- Email: [email protected]
Differentially expressed angiogenic genes in diabetic erectile tissue — Results from a microarray screening
Molecular Genetics and Metabolism 105 (2012) 255–262
Contents lists available at SciVerse ScienceDirect
Molecular Genetics and Metabolism journal homepage: www.elsevier.com/locate/ymgme
Differentially expressed angiogenic genes in diabetic erectile tissue — Results from a microarray screening Ângela Castela a, b, Raquel Soares b, Fátima Rocha c, Rui Medeiros d, Ricardo Ribeiro d, Cátia Monteiro d, Pedro Gomes b, Pedro Vendeira a, e, Ronald Virag f, Carla Costa b, c,⁎ a
Institute for Molecular and Cell Biology of the University of Porto (IBMC-UP), Rua do Campo Alegre, 823, 4150-180 Porto, Portugal Department of Biochemistry (U38-FCT), Faculty of Medicine of the University of Porto, Alameda Prof. Hernani Monteiro, 4200-319 Porto, Portugal Department of Experimental Biology, Faculty of Medicine of the University of Porto, Alameda Prof. Hernani Monteiro, 4200-319 Porto, Portugal d Molecular Oncology Group, Portuguese Institute of Oncology of Porto, Rua Dr. António Bernardino de Almeida, 4200-072 Porto, Portugal e Department of Urology, Hospital de S. João, Alameda Prof. Hernani Monteiro, 4200-319 Porto, Portugal f Centre d'Explorations et Traitements de l'Impuissance, 8, rue de Duras 75008 Paris, France b c
a r t i c l e
i n f o
Article history: Received 4 November 2011 Accepted 4 November 2011 Available online 10 November 2011 Keywords: Angiogenesis Diabetes Erectile dysfunction Immunohistochemistry Microarrays Real-time PCR
a b s t r a c t Diabetes-induced metabolic derangements promote endothelial malfunction, contributing to erectile dysfunction (ED). However, it remains unclear which angiogenic molecular mechanisms are deregulated in diabetic corpus cavernosum (CC). We investigated early and late alterations in cavernosal angiogenic gene expression associated to diabetes. Angiogenic changes were assessed in penile tissue of streptozotocin-induced Wistar rats, in an early (2-week) and established stage (8-week) of diabetes. Differentially expressed genes were identified by microarrays and expression data validated by quantitative real-time PCR (qrt-PCR). At protein level, quantitative immunohistochemistry confirmed the arrays data and dual immunofluorescence for selected alterations and α-smooth muscle actin (α-SMA) identified the cellular location of target proteins. The selected differentially expressed genes were also evaluated in human non-diabetic and diabetic CC by quantitative immunolabeling. At 2-week diabetes there was no differential gene expression between non-diabetic and diabetic CC. At 8-week, 10 genes were found down-regulated in diabetics. The results were validated by qrt-PCR for the insulin-like growth factor-1 (Igf1) and the natriuretic peptide receptor-1 (Npr1) genes. Dual immunofluorescence for IGF-1/ α-SMA showed predominant localization of IGF-1 in SM. NPR-1 expression was diffuse and mostly present in trabecular fibroblasts and SM. Quantitative immunostaining confirmed the decreased expression of both proteins in diabetic tissues. Concordantly, we detected a significant reduction in IGF-1 and NPR-1 protein expressions in human diabetic samples. Microarray analysis identified 10 angiogenic-related molecules deregulated in CC of established diabetes. Among them, IGF-1 and NPR-1 were significantly down-regulated and might result in preventive/therapeutic targets for ED management. © 2011 Elsevier Inc. All rights reserved.
1. Introduction
Abbreviations: Actb, β-actin; AGEs, advanced glycation end products; CC, corpus cavernosum; cGMP, guanosine monophosphate; DAB, diaminobenzidine; DAPI, 4′,6-diamidino2-phenylindole; EC, endothelial cell; ED, erectile dysfunction; EDys, endothelial dysfunction; Gapdh, glyceraldehyde-3-phosphate dehydrogenase; GC, guanilate cyclase; Igf1, insulin-like growth factor-1; NO, nitric oxide; Npr1, natriuretic peptide receptor-1; qrt-PCR, quantitative real-time PCR; SMC, smooth muscle cell; STZ, streptozotocin; T1D, type 1 diabetes; T2D, type 2 diabetes; VEGF, vascular endothelial growth factor; α-SMA, alpha-smooth muscle actin. ⁎ Corresponding author at: Department of Experimental Biology, Faculty of Medicine of the University of Porto, Alameda Prof. Hernani Monteiro, 4200-319 Porto, Portugal. Fax: +351 225513655. E-mail address: [email protected] (C. Costa). 1096-7192/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2011.11.002
Erectile dysfunction (ED) is a common, hard-to-treat complication and important cause of decreased quality of life in men with diabetes, with a prevalence ranging from 15% to 55% of all men with the disease. The basis underlying diabetic-ED is multifactorial involving, among others, changes in central nervous system, peripheral nerve activity and endothelial dysfunction (EDys) [1]. EDys comprises a number of functional alterations in the endothelial monolayer, including changes in vasoregulation, inflammatory activation, and altered barrier function [2,3]. These modifications affect the functionality and integrity of the endothelial bed, impairing cavernosal vasodilation events and blood flow perfusion, and contribute to the development of systemic vascular disease [4,5]. Highlighting the crucial role played by the endothelium, EDys has been considered the central event linking
256
Â. Castela et al. / Molecular Genetics and Metabolism 105 (2012) 255–262
ED and cardiovascular complications in diabetics [6]. Further, EDys is aggravated with diabetes progression, exacerbating vasculopathy [7]. It is established that hyperglycemia-induced formation of advanced glycation end products (AGEs) and increased oxidative stress interfere with the synthesis/bioavailability of nitric oxide (NO), affecting primarily penile vascular endothelium [2,8,9]. However, besides the essential roles of NO in regulating endothelial cell (EC) biological activities, there are numerous endothelial-related molecules controlling relevant pathways required to maintain vascular homeostasis [10]. Still, to date, there is scarce information on how diabetes deleteriously affects the production of essential corporeal vascular mediators, and its effects on endothelial and erectile function. It was suggested that diabetic penile vasculopathy could be related to decreased vascular endothelial growth factor (VEGF) expression/signaling [11,12]. Recently, it was reported that intracorporeal therapy with angiopoietin1 (Ang-1) improved diabetic cavernous endothelial content and ameliorated ED [13]. However, it is reasonable to consider that the aforementioned molecules, and the pathways they mediate, represent only a very small portion of the total changes in vascular-related mediators occurring due to diabetes in erectile tissue. Thus, this study aims to take a more global approach on the range of angiogenic molecular changes occurring with the course of diabetes, in corpus cavernosum (CC) of an experimental model. Further, we intend to verify if the observed angiogenic alterations also occur in human non-diabetic and diabetic erectile tissue. Our overall goal is to advance in the understanding of the vascular pathogenesis of diabetes-associated ED and to provide a rationale for novel neovascular treatments/preventive therapies. 2. Materials and methods 2.1. Animals All experimental procedures and animal handling were conducted according to the ethical guidelines proposed by the Portuguese General Veterinary Directorate (DGV) in the Directive of November 24th 1986 (86/609/EEC), with the recommendations of June 18th 2007 (2007/ 526/EC) proposed by the Council of the European Communities. 40 male Wistar rats (age-matched; 280–350 g; Charles River Laboratories, Barcelona, Spain) were maintained in a 12-hour day/light cycle with free access to food and tap water. Animals were divided in 2 groups: type 1 diabetes (T1D) was induced in 20 rats by streptozotocin (STZ; Sigma-Aldrich, Sintra, Portugal), 50 mg/kg, in citrate buffer (pH 4.6); 20 age-matched controls were injected with vehicle. Diabetes (glucose levels >250 mg/dl) was confirmed by Accu-Check glucose measurement (Roche Diagnostics, Mannheim, Germany). 2 and 8-week diabetics and controls were sacrificed and penises excised. Cohorts of diabetic and control tissues were stored at −80 °C for further molecular analysis. The remaining tissues were fixed in 10% formaldehyde, paraffin embedded and sectioned. 2.2. Human samples All patients gave written informed consent for the removal of cavernous tissue, being this study approved by the local ethics committee and according to the World Medical Association Declaration of Helsinki. 22 diabetic patients with ED [mean age ± standard error (SE); 62.3 ± 1.7] and 10 non-diabetics, non-ED individuals (mean age ± SE; 33.2 ± 4.0) were enrolled in this study. Diabetic individuals, 9 with T1D (mean age ± SE; 59.1 ± 2.6) and 13 with type 2 diabetes (T2D; mean age ± SE; 64.5 ± 2.1) underwent penile implant surgery. Among the 10 non-diabetics, 1 had penile curvature correction and the remaining had procedures for penile enlargement and/or lengthening. During surgical interventions, cavernosal fragments with ≈5– 7.5 mm length/2.5 mm width were harvested and fixed in formaldehyde. Human corporeal tissue was paraffin-embedded and sectioned.
2.3. RNA isolation Total RNA was isolated using TriPure (Roche Applied Science, Indianapolis, USA). Contaminating genomic DNA was removed using the DNA-free™ kit (Ambion, Texas, USA), followed by RNA purification with the RNeasy Mini kit (Qiagen, California, USA). RNA purity and concentration were determined by absorbance reading at 260 nm, 280 nm and 230 nm (NanoDrop™1000 Spectrophotometer; Thermo Scientific, California, USA). RNA integrity was analyzed by denaturing agarose gel electrophoresis. 2.4. Microarray analysis Angiogenic gene expression was evaluated using the Rat Angiogenesis OligoGEArray ® (SuperArray, Maryland, USA). This membrane based-microarray contained a total of 128 oligonucleotide probes, including 113 angiogenic-specific/related genes, 3 blank negative reference spots, 3 artificial sequences, 7 reference genes and 2 prebiotinylated oligonucleotides. ≈3 μg of RNA was used as template to generate/amplify biotin-16-dUTP-labeled-cRNA probes using the TrueLabeling-AMP method (SuperArray). OligoGEArray® membranes were pre-hybridized with GEAhyb solution, followed by hybridization with biotin-labeled-cRNA probes (overnight, 60 °C). Membranes were incubated with alkaline phosphatase-conjugated streptavidin, with the chemiluminescent substrate CDP-Star and exposed to X-ray films. Films were developed, scanned, images acquired and analyzed with the GEArray Analysis Software (SuperArray). Relative expression of each gene was determined after correction for background, subtraction of the global value and normalization, by adjusting to the average intensity signal of the reference gene glyceraldehyde-3-phosphate dehydrogenase (Gapdh). The ratio of gene expression in penile diabetic to control was considered significant when less than 1.0 or greater than 2.0. Microarray evaluation was performed twice for each animal sample. 2.5. Quantitative real-time PCR (qrt-PCR) 1.5 μg of total RNA of 8-week CC samples was reverse transcribed into cDNA using SuperScript™ II Reverse Transcriptase (Invitrogen, California, USA). Each cDNA sample (100 ng/μl) was used in duplicate as template for qrt-PCR using TaqMan® Gene Expression Master Mix and TaqMan® hydrolysis probes (Applied Biosystems, California, USA). 2 target genes were selected from previous microarray experiments: Igf1 (assay_ID Rn99999087_ml) and Npr1 (assay_ID Rn00561678_ml). Amplification in a StepOne Real-Time PCR system (Applied Biosystems), was performed in a final volume of 20 μL, with 1x Gene Expression Master Mix, 250 nM probe and 900 nM primers, using the following cycling conditions: 1 cycle, 50 °C/2 min; 1 cycle, 95 °C/10 s; 45 cycles, 95 °C/15 s and 60 °C/1 min. Raw data was analyzed with Stepone TM Software v2.1 (Applied Biosystems). Quantification cycle (Cq) values were determined for each sample and the comparative Cq method was used to detect relative gene expression ratios normalized to the reference genes, Gapdh (assay_ID Rn01775763_gl) and β-actin (Actb; assay_ID Rn00667869_ml) (Applied Biosystems). Relative mRNA expression was represented as fold change over the calibrator sample (average of gene expression corresponding to the control group) [14]. Relative gene expression changes were calculated using the 2-ΔΔCt method [15]. All samples were prepared and examined in parallel, negative controls were included. 2.6. Immunolabeling and quantification of staining intensity Immunohistochemistry for IGF-1 and NPR-1 was performed in 8week CC and in human cavernosal samples, using the streptavidinbiotin peroxidase method [4]. Slides were incubated overnight with the primary antibodies: cross-reactive rabbit anti-human IGF-1 (1:150;
Â. Castela et al. / Molecular Genetics and Metabolism 105 (2012) 255–262
Abcam, Cambridge, UK); cross-reactive rabbit anti-human NPR-1 (1:900, Abcam), followed by the secondary antibodies (Santa Cruz Biotechnology, California, USA) and with the Avidin–Biotin Complex reagents (DAKO, Glostrup, Denmark). Negative controls were performed by omission of the primary antibody. Reactions were developed using diaminobenzidine (DAB; Sigma-Aldrich) and sections counterstained with hematoxylin. Slides were observed under the Olympus AH3-RFCA microscope and images captured using the Olympus C-35AD-4 camera (Olympus Imaging Europa GmbH, Hamburg, Germany). Signal intensity quantification was performed in 100× digitally captured color images covering entire rat and human CC sections, which were analyzed by ImageJ using color deconvolution (v.1.37a, NIH, Maryland, USA). Color deconvolution was performed by separating the red/green/blue (RGB) image into hematoxylin and brown-colored DAB components [16]. A DAB threshold value of 30 (rat) and 40 (human), separating background DAB from true staining was selected for IGF-1 and NPR-1, by measuring the lowest and the highest mean optical density values in 10 random DAB-positive images. For each image, the sum of the intensity values in the regions of interest (ROI), pixels above the brown threshold, was divided by the ROI total area, producing an average unitless intensity value (“IGF-1 staining” or “NPR-1 staining”). All intensity values within the same group (non-diabetics and diabetics) were averaged to calculate an overall value, a relative measurement of IGF-1 and NPR-1 protein levels.
257
Table 1 Differentially expressed angiogenic genes in 8-week diabetic versus control erectile tissue. Alterations are expressed as the ratio of gene expression in penile diabetic/control rats less than 1.0-fold change. Position GeneBank no 47 12 86
77 11 108
68 69 82 72
NM_178866
Gene description
Insulin like growth factor 1 NM_031530 Chemokine (C-C motif) ligand 2 NM_177927 Serine (or cysteine) peptidase inhibitor, clade F, member 1 NM_013085 Plasminogen activator, urokinase NM_019205 Chemokine (C-C motif) ligand 11 NM_00100151 Tumor necrosis factor ligand superfamily member 12 NM_012613 Natriuretic peptide receptor 1 NM_145098 Neuropilin 1 NM_017043 Prostaglandin-endoperoxide Synthase 1 NM_012801 Platelet-derived growth factor alpha
Gene symbol
Ratio (diabetic/ control)
Igf1
0.043⁎
Ccl2
0.143⁎
Serpinf1 0.159⁎
Plau
0.266⁎
Ccl11
0.423⁎
Tnfsf12
0.423⁎
Npr1
0.424⁎
Nrp1 Ptgs1
0.436 0.640⁎
Pdgfa
0.682
⁎ p b 0.05.
2.7. Dual immunofluorescence Simultaneous expression of IGF-1 and α-smooth muscle actin (α-SMA) and of NPR-1 and α-SMA, was assessed by double immunofluorescence [4]. Briefly, rat and human CC sections were incubated with a mixture of the primary antibodies: cross-reactive rabbit antihuman IGF-1 (1:150; Abcam) and cross-reactive mouse anti-human α-SMA (dilution 1:500; clone ASM, Chemicon, UK); cross-reactive rabbit anti-human NPR-1 (1:900, Abcam) and the aforementioned α-SMA antibody (1:500). Next, slides were incubated with the secondary antibodies: donkey anti-rabbit conjugated with a red fluorochrome (1:1000; Alexa™ 568, Invitrogen, California, USA) and donkey antimouse conjugated with a green fluorochrome (dilution 1:1000; Alexa™ 488, Invitrogen). Negative controls were performed by omission of
the primary antibodies. Nuclei were counterstained with DAPI (4′,6diamidino-2-phenylindole; Invitrogen), slides were observed under the fluorescence microscope (Imager.Z1, Zeiss, Germany) and images captured (Carl Zeiss MicroImaging GmbH, Germany).
2.8. Statistics Data is presented as mean ± standard error (SE). Statistical differences between 8-week non-diabetic and diabetic erectile tissue, as evaluated by microarray analysis, qrt-PCR and quantitative immunohistochemistry, were determined using the Mann–Whitney test (p b 0.05 considered statistically significant).
Fig. 1. Microarray evaluation. (A) Location of the genes included in the Rat Angiogenesis OligoGEArrays (SuperArray, Maryland, USA). This membrane based-microarray contained a total of 128 oligonucleotide probes, including 113 angiogenic-specific/related genes, 3 blank negative reference spots (115–117), 3 artificial sequences (118–120), 7 reference genes (1, 121–126) and 2 pre-biotinylated oligonucleotides (127, 128). (B) Representative membrane array image of the overall expression of the angiogenic-related genes in 8-week control and diabetic animals. Ccl2—chemokine (C-C motif) ligand 2; Ccl11—chemokine (C-C motif) ligand 11; Fgf1,2,6—fibroblast growth factor 1,2,6; Fgfr3—fibroblast growth factor receptor 3; Figf—c-fos induced growth factor; Flt1—vascular endothelial growth factor receptor 1; Igf1—insulin like growth factor 1; Npr1—natriuretic peptide receptor 1; Nrp1—neuropilin 1; Pdgfa—platelet-derived growth factor alpha; Plau—plasminogen activator, urokinase; Ptgs1—prostaglandin-endoperoxide Synthase 1; Serpinf1—serine (or cysteine) peptidase inhibitor, clade F, member 1; Tnfsf12—tumor necrosis factor ligand superfamily member 12; Vegfa,b,c—vascular endothelial growth factor a,b,c.
258
Â. Castela et al. / Molecular Genetics and Metabolism 105 (2012) 255–262
Fig. 2. Confirmation of microarrays results for Igf1 and Npr1 genes by qrt-PCR. (A) Igf1 was significantly down-regulated in 8-week diabetic CC as compared to controls (p = 0.003); (B) Npr1 gene presented a significant reduced expression in 8-week diabetic penile tissue (p= 0.0021). Relative quantification was achieved by normalizing the expression of Igf1 and Npr1 to Gapdh and Actb by the 2-ΔΔCt method [15]. Data presented as Mean ± SE.
3. Results 3.1. Microarrays analysis We analyzed angiogenic gene expression alterations in CC in an early and established stage of diabetes. Control and diabetic tissues, 2 and 8-week diabetes, were evaluated using the Rat Angiogenesis
OligoGEArray (Fig. 1A) and GEArray analyzer. This software allowed us to perform a global qualitative and semi-quantitative expression analysis of 113 angiogenic genes directly and indirectly involved in angiogenesis pathways. As determined by repetitive microarray evaluations, at 2-week diabetes there was no significant differential gene expression between non-diabetic and diabetic CC, as expressed by the ratio of relative gene expression of diabetic/non-diabetic. At 8 weeks of diabetes, there was a decrease in the expression of several genes as observed in the representative membranes in Fig. 1B. Membranes from 8-week animals presented a similar profile of altered genes, however among different samples there was some degree of variability. A number of genes involved directly in the angiogenic process, as members of the fibroblast growth factor (Fgf) and the vascular endothelial growth factor (Vegf) families presented undetectable physiological levels in both 8-week controls and diabetic samples. The Vegf receptor 1, Flt1, had also an extremely low expression, unnoticeable in 8-week tissues. The Vegf receptor 2, Kdr, presented a slight decrease in diabetic CC, which was not overall statistically significant (p = 0.57), since differential expression was considered only when genes presented levels of expression less than 1.0-fold. It was the case of a total of 10 genes which exhibited expression of less than 1.0-fold in STZinduced erectile tissue samples when compared to controls (Table 1). The category of altered genes included: angiogenesis (Nrp1, Serpinf1, Igf1, Pdgfa), guanilate cyclase (GC) activity/cyclic guanosine monophosphate (cGMP) biosynthesis (Npr1), chemokine activity (Ccl2, Ccl11, Tnfsf12), proteolysis (Plau) and inflammation (Ptgs1).
Fig. 3. Quantification and localization of IGF-1 protein in 8-week control and diabetic penile tissue. (A, B) IGF-1 immunoreactivity was quantified in IGF-1/DAB stained cavernosal samples using ImageJ color deconvolution plugin [16]. (A) Color deconvolution was performed by separating the red/green/blue (RGB) image into hematoxylin and brown-colored DAB components [16]. A DAB threshold value was selected, and the signal above threshold was considered as IGF-1 positive staining; (B) All intensity values within the same group (non-diabetics and diabetics) were averaged to calculate an overall value, a relative measurement of IGF-1protein levels. A significant reduction in IGF-1 production was confirmed in 8-week diabetic penile tissue; (C) IGF-1 expression pattern was evaluated by double immunofluorescence for IGF-1/α-SMA. IGF-1 was mostly identified in the corporeal nondiabetic and diabetic SMC counterpart, although it could also be detected in trabecular fibroblasts. DAB, diaminobenzidine; DAPI, 4′,6-diamidino-2-phenylindole (DNA intercalator). Scale bar = 100 μm. Data presented as Mean ± SE; **P ≤ 0.01.
Â. Castela et al. / Molecular Genetics and Metabolism 105 (2012) 255–262
3.2. qrt-PCR validation of selected genes qrt-PCR was used to confirm decreased expression levels of Igf1 and Npr1 in diabetic CC. These genes were selected based on their biological function and potential role in vascular dysfunction associated to diabetic-ED. qrt-PCR revealed that both Igf1 and Npr1 mRNA expressions were down-regulated by three-fold (p = 0.003 and p = 0.0021, respectively) in 8-week diabetic penile tissue when compared to controls (Fig. 2A and B); consistent with the results obtained by microarrays. These results were further confirmed at protein level. 3.3. Immunohistochemistry and protein quantitative evaluation To confirm at protein level the data obtained by molecular evaluation, we quantified IGF-1 and NPR-1 immunoreactivity staining in DAB-labeled sections at 8-week diabetes, using ImageJ color deconvolution method [18]. Additionally, we identified rat cavernosal components expressing IGF-1 and NPR-1, by performing dual immunofluorescences for IGF-1/α-SMA and NPR-1/α-SMA. Corroborating the array analysis, we observed that IGF-1 immunoreactivity in IGF-1/DAB-stained samples (Fig. 3A) was significantly decreased (pb 0.01) in 8-week diabetics (73.5 ± 2.3) when compared to controls (89.2 ± 4.1) (Fig. 3B). Regarding the expression profile, in non-diabetic penile tissue, IGF-1 was abundantly produced by SMCs, with some expression in trabecular fibroblasts (Fig. 3C, upper panel). Diabetic corporeal tissue presented an analogous IGF-1 pattern, mostly in the
259
SM layer (Fig. 3C, lower panel). In diabetic penis, NPR-1 protein levels were also significantly reduced, as quantified by NPR-1/DAB-labeling immunopositivity (Fig. 4A) with 95.4 ± 3.0 in controls versus 77.4 ± 4.2 in diabetics; p b 0.01 (Fig. 4B). NPR-1 protein was diffusely detected in non-diabetic trabecular fibroblasts and SMCs (Fig. 3D, upper panel). In the diabetic penis, NPR-1 profile was similar to what was observed in control tissues (Fig. 3D, lower panel). 3.4. Quantitative immunohistochemistry in human samples To verify if IGF-1 and NPR-1 decreased expression levels in diabetic CC were not restricted to rat samples, we performed analogous experiments and quantitative evaluation in human cavernosal non-diabetic and diabetic samples. Consistent with the results obtained in experimental erectile tissue, human diabetic CC presented significantly decreased IGF-1 protein levels, as quantified in IGF-1/DAB-stained slides (Fig. 5A) with mean values of 112.0 ± 2.0 in non-diabetics and 94.1 ± 2.1 in diabetic patients (pb 0.001) (Fig. 5B). Similarly, IGF-1 was predominantly detected in the SMC layer of human non-diabetic CC, as evaluated by double immunofluorescence for IGF-1/α-SMA (Fig. 5C, upper panel). In human diabetic CC, IGF-1 had a similar expression profile when compared with controls (Fig. 5C, lower panel). In NPR-1/DAB stained slides (Fig. 6A), we observed that protein levels were diminished in regard to controls (141.9 ± 4.1, non-diabetics; 133.1 ± 1.9, diabetics; p b 0.05) (Fig. 6B). In non-diabetic corporeal tissue, NPR-1 presented a diffuse pattern of expression, mostly identified in trabecular fibroblast-like cells and SMCs (Fig. 6C, upper panel). This staining
Fig. 4. NPR-1 protein quantification and expression pattern in 8-week control and diabetic cavernosal samples. (A, B) NPR-1 immunopositivity was quantified in NPR-1/DAB stained cavernosal samples using ImageJ color deconvolution plugin [16]. (A) Color deconvolution was performed as abovementioned for IGF-1 [16]; (B) Mean values of NPR-1 in non-diabetic and diabetic rat penile tissue, confirming the significant decrease in NPR-1 protein expression in diabetic corporeal tissue; (C) NPR-1 expression profile as assessed by double immunofluorescence for NPR-1/α-SMA, demonstrated a diffused expression in trabecular fibroblasts and SMCs of both control and diabetic CC. DAB, diaminobenzidine; DAPI, 4′,6-diamidino-2-phenylindole (DNA intercalator). Scale bar = 100 μm. Data presented as Mean ± SE; **P ≤ 0.01.
260
Â. Castela et al. / Molecular Genetics and Metabolism 105 (2012) 255–262
Fig. 5. IGF-1 protein evaluation in human non-diabetic and diabetic cavernosal samples. (A, B) IGF-1 protein levels were quantified in IGF-1/DAB stained corporeal tissues using ImageJ color deconvolution. (B) Quantitative immunohistochemistry in IGF-1/DAB stained slides showed a significant reduction in protein expression in human diabetic erectile tissue; (C) Double immunofluorescence for IGF-1/α-SMA demonstrated positive labeling in cavernosal SMCs of both non-diabetic and diabetic samples. In certain vascular spaces IGF-1 was clearly identified lining the sinosoids (upper panel; white arrow). DAB, diaminobenzidine; DAPI, 4′,6-diamidino-2-phenylindole (DNA intercalator). Scale bar = 100 μm. Data presented as Mean ± SE; ***p b 0.001.
pattern was maintained in diabetic CC samples (Fig. 6C, lower panel). Interestingly, when separating the overall group of diabetics in T1D and T2D, no differences were attained for IGF-1 and NPR-1 localization and quantification. 4. Discussion Diabetic men have approximately a three-fold increased risk for the development of ED, compared with healthy individuals [17]. It is known that elevated production of AGEs, increased oxidative stress, and impairment of NO synthesis, affect penile ECs, leading to EDys and ED [2,4,8,9]. However, besides these mechanisms, there is very scarce information on how diabetes detrimentally affects the production of endothelial mediators, contributing to EDys. We evaluated in different stages of diabetes, STZ-induced 2 and 8 weeks T1D, the range of angiogenic-related molecules differentially expressed in erectile tissue, using a membrane-based microarray system. A similar molecular approach using affymetrix chips, was previously used in studies of diabetic-ED [19,20]. However, one report analyzed in established diabetes a wide range of genes involved in multiple biological functions [19], whereas the other focused in the analysis of penile SMCrelated changes in early diabetes [20]. Our study was the first to evaluate alterations in cavernosal vascular-related genes in two different stages of diabetes. At 2-week we did not observe significant changes in genes associated with vascular function. We may assume that at this early time point, diabetic-induced metabolic deregulations have not yet induced significant alterations at gene level. Additionally, considering that diabetic-related hemodynamic changes in erectile function occur
at later stages of the disease [21], it is valid to assume that endothelial major alterations would occur with diabetes progression. At 8-week diabetes we identified 10 angiogenic-related genes down-regulated in diabetic CC. The functional category of altered genes included (Table 1): angiogenesis (Nrp1, Serpinf1, Igf1, Pdgfa), GC activity/cGMP biosynthesis (Npr1), chemokine activity (Ccl2, Ccl11, Tnfsf12), proteolysis (Plau) and inflammation (Ptgs1). Members of the VEGF family of ligands and receptors present in the microarray membranes were barely detected at physiological levels, both in control and diabetic samples and no significant alterations were observed among groups. Our results regarding VEGF expression are divergent to what was previously reported in an obese/T2D rat model [11]. This discrepancy may be due to the experimental model/rat strain and the detection methods. Regarding the down-regulated genes, a reduced expression of Igf1[22], Pdgfa[23], Npr1[24] and Ptgs1[25], has been associated with diabetic complications in other organ systems; suggesting the existence of common molecular pathways underlying diabetes. Among our gene alterations, we validated by qrtPCR Igf1 and Npr1, chosen due to their relevant function in vasorelaxation mechanisms. Reduced Igf1 expression may have deleterious effects in diabetic erectile function, since it has been reported that Igf1 adenoviral gene transfer to STZ-diabetic CC ameliorated ED [26]. In fact, Igf1 displays relevant biological roles as angiogenic mediator and stimulator of myogenesis, crucial events for penile endothelial homeostasis and SMC function [27,28]. The Npr1 gene encodes for a membrane bound GC receptor with a tyrosine kinase domain, involved in cGMP biosynthesis [29]. Npr1 was shown to promote vascular regeneration [29] and due to its role in cGMP formation, plays a relevant role in SMC vasorelaxation events [30]. Next, we
Â. Castela et al. / Molecular Genetics and Metabolism 105 (2012) 255–262
261
Fig. 6. NPR-1 protein analysis in human non-diabetic and diabetic penile tissue. (A, B) Quantitative immunohistochemistry in NPR-1/DAB using ImageJ color deconvolution; (B) NRP-1/DAB protein levels were significantly decreased in human diabetic erectile tissue; (C) NPR-1 expression profile was diffuse in non-diabetic and diabetic trabecular fibroblasts and SMCs. DAB, diaminobenzidine; DAPI, 4′,6-diamidino-2-phenylindole (DNA intercalator). Scale bar = 100 μm. Data presented as Mean ± SE; *p b 0.05.
confirmed the microarray data at protein level by quantitative immunohistochemistry [16] and dual immunofluorescence for IGF-1/ α-SMA and NPR-1/ α-SMA. In agreement with its reported roles in SM proliferation/myogenesis and angiogenesis, IGF-1 was mostly detected in SMCs and in some ECs-lining sinosoids [27,28]. NPR-1 was particularly identified in trabecular fibroblasts and SMCs. SMC expression of NPR-1 is consistent with its role in cGMP formation and cavernosal vasorelaxation [30]. In corporeal trabecular fibroblasts, NPR-1 protein expression was never described. It is established that decreased SMCs and expansion of collagen producing trabecular fibroblasts is one of diabetes-induced pathophysiological changes [12]. As a recent publication suggested that NPR-1 has inhibitory effects in renal fibrosis in mice [31], we hypothesized the same role in penile trabeculae. We may speculate that lower NPR-1 expression in diabetic trabeculae may potentially contribute to cavernosal fibrosis. Finally, we verified that IGF-1 and NPR-1 protein levels were decreased in human diabetic CC and that the expression pattern of these proteins was similar to rat penile tissue. Taken together, our study was the first to evaluate angiogenic-related gene alterations in STZ-induced diabetic erectile tissue, using microarrays. We reported a downregulation in Igf1 and Npr1 gene expression in diabetic CC, which was later confirmed at protein level both in rat and human diabetic erectile tissue. In conclusion, we demonstrated a down-regulation in 10 angiogenicrelated genes in diabetic CC of an established STZ-model. Among them, Igf1 and Npr1 were selected due to their relevant roles in cavernosal endothelial and SMC components. A reduced expression in IGF-1 and NPR-1 was also detected in human diabetic erectile tissue, as compared to non-diabetic CC. These data indicate that IGF-1 and NPR-1
in penile tissue may play important roles in the maintenance of corporeal homeostasis/erectile process, and should be taken into account as potential targets in diabetic-ED. Conflict of interest The authors disclose any financial and personal relationships with other people or organizations that could inappropriately influence this work. Acknowledgments This study was supported by ESSM Grant for Medical Research 2007, by the Portuguese Foundation for Science and Technology (PTDC/SAUOSM/65599/2006; SFRH/BPD/40554/2007) and by the Associação Portuguesa de Urologia (Prémio Abbott Urologia 2008). References [1] I. Sáenz de Tejada, J. Angulo, S. Cellek, N. González-Cadavid, J. Heaton, R. Pickard, U. Simonsen, Pathophysiology of erectile dysfunction, J. Sex. Med. 2 (2005) 26–39. [2] J. Zúrová-Nedelcevová, J. Navarová, K. Drábiková, V. Jancinová, M. Petríková, I. Bernátová, V. Kristová, V. Snirc, V. Nosál'ová, R. Sotníková, Participation of reactive oxygen species in diabetes-induced endothelial dysfunction, Neuro Endocrinol. Lett. 27 (2006) 168–171. [3] W. Bakker, E.C. Eringa, P. Sipkema, V.W. van Hinsbergh, Endothelial dysfunction and diabetes: roles of hyperglycemia, impaired insulin signaling and obesity, Cell Tissue Res. 335 (2009) 165–189. [4] C. Costa, R. Soares, A. Castela, S. Adães, V. Hastert, P. Vendeira, R. Virag, Increased endothelial apoptotic cell density in human diabetic erectile tissue — comparison with clinical data, J. Sex. Med. 6 (2009) 826–835.
262
Â. Castela et al. / Molecular Genetics and Metabolism 105 (2012) 255–262
[5] M.M. Hartge, U. Kintscher, T. Unger, Endothelial dysfunction and its role in diabetic vascular disease, Endocrinol. Metab. Clin. North Am. 35 (2006) 551-5ix. [6] M. Kirby, G. Jackson, U. Simonsen, Endothelial dysfunction links erectile dysfunction to heart disease, Int. J. Clin. Pract. 59 (2005) 225–229. [7] J. Xu, M.H. Zou, Molecular insights and therapeutic targets for diabetic endothelial dysfunction, Circulation 120 (2009) 1266–1286. [8] A. Nessar, Advanced glycation end products—role in pathology of diabetic complications, Diabetes Res. Clin. Pract. 67 (2005) 3–21. [9] B. Musicki, M.F. Kramer, R.E. Becker, A.L. Burnett, Inactivation of phosphorylated endothelial nitric oxide synthase (Ser-1177) by O-GlcNAc in diabetes-associated erectile dysfunction, Proc. Natl. Acad. Sci. USA 102 (2005) 11870–11875. [10] J. Folkman, Fundamental concepts of the angiogenic process, Curr. Mol. Med. 3 (2003) 643–651. [11] S. Jesmin, I. Sakuma, A. Salah-Eldin, K. Nonomura, Y. Hattori, A. Kitabatake, Diminished penile expression of vascular endothelial growth factor and its receptors at the insulin-resistant stage of a type II diabetic rat model: a possible cause for erectile dysfunction in diabetes, J. Mol. Endocrinol. 31 (2003) 401–418. [12] M. Yamanaka, M. Shirai, H. Shiina, Y. Tanaka, H. Enokida, A. Tsujimura, K. Matsumiya, A. Okuyama, R. Dahiya, Vascular endothelial growth factor restores erectile function through inhibition of apoptosis in diabetic rat penile crura, J. Urol. 173 (2005) 318–323. [13] H.R. Jin, W.J. Kim, J.S. Song, S. Piao, M. Tumurbaatar, S.H. Shin, M.J. Choi, B. Tuvshintur, K.M. Song, M.H. Kwon, G.N. Yin, G.Y. Koh, J.K. Ryu, J.K. Suh, Intracavernous delivery of synthetic angiopoietin-1 protein as a novel therapeutic strategy for erectile dysfunction in the type II diabetic db/db mouse, J. Sex. Med. 7 (2010) 3635–3646. [14] V. Catalán, J. Gómez-Ambrosi, A. Rodríguez, C. Silva, F. Rotellar, M.J. Gil, J.A. Cienfuegos, J. Salvador, G. Frühbeck, Expression of caveolin-1 in human adipose tissue is upregulated in obesity and obesity-associated type 2 diabetes mellitus and related to inflammation, Clin. Endocrinol. (Oxf) 68 (2008) 213–219. [15] K.J. Livak, T.D. Schmittgen, Analysis of relative gene expression data using real-time quantitative PCR and the 2(ΔΔCT) method, Methods 25 (2001) 402–408. [16] A.C. Ruifrok, D.A. Johnston, Quantification of histochemical staining by color deconvolution, Anal. Quant. Cytol. Histol. 23 (2001) 291–299. [17] J. Dey, M.D. Shepherd, Evaluation and treatment of erectile dysfunction in men with diabetes mellitus, Mayo Clin. Proc. 77 (2002) 276–282. [18] P.V. Babu, K.E. Sabitha, C.S. Shyamaladevi, Therapeutic effect of green tea extract on advanced glycation and cross-linking of collagen in the aorta of streptozotocin diabetic rats, Clin. Exp. Pharmacol. Physiol. 33 (2006) 351–357. [19] C.J. Sullivan, T.H. Teal, I.P. Luttrell, K.B. Tran, M.A. Peters, H. Wessells, Microarray analysis reveals novel gene expression changes associated with erectile dysfunction in diabetic rats, Physiol. Genomics 23 (2005) 192–205.
[20] J.D. Hipp, K.P. Davies, M. Tar, M. Valcic, A. Knoll, A. Melman, G.J. Christ, Using gene chips to identify organ-specific, smooth muscle responses to experimental diabetes: potential applications to urological diseases, BJU Int. 99 (2007) 418–430. [21] T.J. Bivalacqua, M.F. Usta, H.C. Champion, S. Leungwattanakij, P.A. Dabisch, D.B. McNamara, P.J. Kadowitz, W.J. Hellstrom, Effect of combination endothelial nitric oxide synthase gene therapy and sildenafil on erectile function in diabetic rats, Int. J. Impot. Res. 16 (2004) 21–29. [22] H. Yang, A.J. Scheff, D.S. Schalch, Effects of streptozotocin-induced diabetes mellitus on growth and hepatic insulin-like growth factor I gene expression in the rat, Metabolism 39 (1990) 295–301. [23] Q.H. Wu, W.S. Chen, Q.X. Chen, J.H. Wang, X.M. Zhang, Changes in the expression of platelet-derived growth factor in astrocytes in diabetic rats with spinal cord injury, Chin. Med. J. (Engl) 123 (2010) 1577–1581. [24] V.G. Marrachelli, F.J. Miranda, J.M. Centeno, I. Miranda, M. Castelló-Ruiz, M.C. Burguete, T. Jover-Mengual, J.B. Salom, G. Torregrosa, E. Alborch, Mechanisms underlying the diabetes-induced hyporeactivity of the rabbit carotid artery to atrial natriuretic peptide, Pharmacol. Res. 63 (2011) 190–198. [25] C. Fang, Z. Jiang, D.R. Tomlinson, Expression of constitutive cyclo-oxygenase (COX-1) in rats with streptozotocin-induced diabetes; effects of treatment with evening primrose oil or an aldose reductase inhibitor on COX-1 mRNA levels, Prostaglandins Leukot. Essent. Fatty Acids 56 (1997) 157–163. [26] W.Y. Pu, L.Q. Hu, H.P. Wang, Y.X. Luo, X.H. Wang, Improvement in erectile dysfunction after insulin-like growth factor-1 gene therapy in diabetic rats, Asian J. Androl. 9 (2007) 83–91. [27] E.D. Rabinovsky, R. Draghia-Akli, Insulin-like growth factor I plasmid therapy promotes in vivo angiogenesis, Mol. Ther. 9 (2004) 46–55. [28] M. Kim, E.C. Hwang, I.K. Park, K. Park, Insulin-like growth factor-1 gene delivery may enhance the proliferation of human corpus cavernosal smooth muscle cells, Urology 76 (2010) 511.e5-9. [29] M. Kuhn, K. Völker, K. Schwarz, J. Carbajo-Lozoya, U. Flögel, C. Jacoby, J. Stypmann, M. van Eickels, S. Gambaryan, M. Hartmann, M. Werner, T. Wieland, J. Schrader, H. A. Baba, The natriuretic peptide/guanylyl cyclase-a system functions as a stressresponsive regulator of angiogenesis in mice, J. Clin. Invest. 119 (2009) 2019–2030. [30] E. Andrade, P.M. Andrade, R.C. Borra, J. Claro, M. Srougi, cDNA microarray analysis of differentially expressed genes in penile tissue after treatment with tadalafil, BJU Int. 101 (2008) 508–512. [31] T. Nishikimi, C. Inaba-Iemura, K. Ishimura, K. Tadokoro, S. Koshikawa, K. Ishikawa, K. Akimoto, Y. Hattori, K. Kasai, N. Minamino, N. Maeda, H. Matsuoka, Natriuretic peptide/natriuretic peptide receptor-A (NPR-A) system has inhibitory effects in renal fibrosis in mice, Regul. Pept. 154 (2009) 44–53.
Contents lists available at SciVerse ScienceDirect
Molecular Genetics and Metabolism journal homepage: www.elsevier.com/locate/ymgme
Differentially expressed angiogenic genes in diabetic erectile tissue — Results from a microarray screening Ângela Castela a, b, Raquel Soares b, Fátima Rocha c, Rui Medeiros d, Ricardo Ribeiro d, Cátia Monteiro d, Pedro Gomes b, Pedro Vendeira a, e, Ronald Virag f, Carla Costa b, c,⁎ a
Institute for Molecular and Cell Biology of the University of Porto (IBMC-UP), Rua do Campo Alegre, 823, 4150-180 Porto, Portugal Department of Biochemistry (U38-FCT), Faculty of Medicine of the University of Porto, Alameda Prof. Hernani Monteiro, 4200-319 Porto, Portugal Department of Experimental Biology, Faculty of Medicine of the University of Porto, Alameda Prof. Hernani Monteiro, 4200-319 Porto, Portugal d Molecular Oncology Group, Portuguese Institute of Oncology of Porto, Rua Dr. António Bernardino de Almeida, 4200-072 Porto, Portugal e Department of Urology, Hospital de S. João, Alameda Prof. Hernani Monteiro, 4200-319 Porto, Portugal f Centre d'Explorations et Traitements de l'Impuissance, 8, rue de Duras 75008 Paris, France b c
a r t i c l e
i n f o
Article history: Received 4 November 2011 Accepted 4 November 2011 Available online 10 November 2011 Keywords: Angiogenesis Diabetes Erectile dysfunction Immunohistochemistry Microarrays Real-time PCR
a b s t r a c t Diabetes-induced metabolic derangements promote endothelial malfunction, contributing to erectile dysfunction (ED). However, it remains unclear which angiogenic molecular mechanisms are deregulated in diabetic corpus cavernosum (CC). We investigated early and late alterations in cavernosal angiogenic gene expression associated to diabetes. Angiogenic changes were assessed in penile tissue of streptozotocin-induced Wistar rats, in an early (2-week) and established stage (8-week) of diabetes. Differentially expressed genes were identified by microarrays and expression data validated by quantitative real-time PCR (qrt-PCR). At protein level, quantitative immunohistochemistry confirmed the arrays data and dual immunofluorescence for selected alterations and α-smooth muscle actin (α-SMA) identified the cellular location of target proteins. The selected differentially expressed genes were also evaluated in human non-diabetic and diabetic CC by quantitative immunolabeling. At 2-week diabetes there was no differential gene expression between non-diabetic and diabetic CC. At 8-week, 10 genes were found down-regulated in diabetics. The results were validated by qrt-PCR for the insulin-like growth factor-1 (Igf1) and the natriuretic peptide receptor-1 (Npr1) genes. Dual immunofluorescence for IGF-1/ α-SMA showed predominant localization of IGF-1 in SM. NPR-1 expression was diffuse and mostly present in trabecular fibroblasts and SM. Quantitative immunostaining confirmed the decreased expression of both proteins in diabetic tissues. Concordantly, we detected a significant reduction in IGF-1 and NPR-1 protein expressions in human diabetic samples. Microarray analysis identified 10 angiogenic-related molecules deregulated in CC of established diabetes. Among them, IGF-1 and NPR-1 were significantly down-regulated and might result in preventive/therapeutic targets for ED management. © 2011 Elsevier Inc. All rights reserved.
1. Introduction
Abbreviations: Actb, β-actin; AGEs, advanced glycation end products; CC, corpus cavernosum; cGMP, guanosine monophosphate; DAB, diaminobenzidine; DAPI, 4′,6-diamidino2-phenylindole; EC, endothelial cell; ED, erectile dysfunction; EDys, endothelial dysfunction; Gapdh, glyceraldehyde-3-phosphate dehydrogenase; GC, guanilate cyclase; Igf1, insulin-like growth factor-1; NO, nitric oxide; Npr1, natriuretic peptide receptor-1; qrt-PCR, quantitative real-time PCR; SMC, smooth muscle cell; STZ, streptozotocin; T1D, type 1 diabetes; T2D, type 2 diabetes; VEGF, vascular endothelial growth factor; α-SMA, alpha-smooth muscle actin. ⁎ Corresponding author at: Department of Experimental Biology, Faculty of Medicine of the University of Porto, Alameda Prof. Hernani Monteiro, 4200-319 Porto, Portugal. Fax: +351 225513655. E-mail address: [email protected] (C. Costa). 1096-7192/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2011.11.002
Erectile dysfunction (ED) is a common, hard-to-treat complication and important cause of decreased quality of life in men with diabetes, with a prevalence ranging from 15% to 55% of all men with the disease. The basis underlying diabetic-ED is multifactorial involving, among others, changes in central nervous system, peripheral nerve activity and endothelial dysfunction (EDys) [1]. EDys comprises a number of functional alterations in the endothelial monolayer, including changes in vasoregulation, inflammatory activation, and altered barrier function [2,3]. These modifications affect the functionality and integrity of the endothelial bed, impairing cavernosal vasodilation events and blood flow perfusion, and contribute to the development of systemic vascular disease [4,5]. Highlighting the crucial role played by the endothelium, EDys has been considered the central event linking
256
Â. Castela et al. / Molecular Genetics and Metabolism 105 (2012) 255–262
ED and cardiovascular complications in diabetics [6]. Further, EDys is aggravated with diabetes progression, exacerbating vasculopathy [7]. It is established that hyperglycemia-induced formation of advanced glycation end products (AGEs) and increased oxidative stress interfere with the synthesis/bioavailability of nitric oxide (NO), affecting primarily penile vascular endothelium [2,8,9]. However, besides the essential roles of NO in regulating endothelial cell (EC) biological activities, there are numerous endothelial-related molecules controlling relevant pathways required to maintain vascular homeostasis [10]. Still, to date, there is scarce information on how diabetes deleteriously affects the production of essential corporeal vascular mediators, and its effects on endothelial and erectile function. It was suggested that diabetic penile vasculopathy could be related to decreased vascular endothelial growth factor (VEGF) expression/signaling [11,12]. Recently, it was reported that intracorporeal therapy with angiopoietin1 (Ang-1) improved diabetic cavernous endothelial content and ameliorated ED [13]. However, it is reasonable to consider that the aforementioned molecules, and the pathways they mediate, represent only a very small portion of the total changes in vascular-related mediators occurring due to diabetes in erectile tissue. Thus, this study aims to take a more global approach on the range of angiogenic molecular changes occurring with the course of diabetes, in corpus cavernosum (CC) of an experimental model. Further, we intend to verify if the observed angiogenic alterations also occur in human non-diabetic and diabetic erectile tissue. Our overall goal is to advance in the understanding of the vascular pathogenesis of diabetes-associated ED and to provide a rationale for novel neovascular treatments/preventive therapies. 2. Materials and methods 2.1. Animals All experimental procedures and animal handling were conducted according to the ethical guidelines proposed by the Portuguese General Veterinary Directorate (DGV) in the Directive of November 24th 1986 (86/609/EEC), with the recommendations of June 18th 2007 (2007/ 526/EC) proposed by the Council of the European Communities. 40 male Wistar rats (age-matched; 280–350 g; Charles River Laboratories, Barcelona, Spain) were maintained in a 12-hour day/light cycle with free access to food and tap water. Animals were divided in 2 groups: type 1 diabetes (T1D) was induced in 20 rats by streptozotocin (STZ; Sigma-Aldrich, Sintra, Portugal), 50 mg/kg, in citrate buffer (pH 4.6); 20 age-matched controls were injected with vehicle. Diabetes (glucose levels >250 mg/dl) was confirmed by Accu-Check glucose measurement (Roche Diagnostics, Mannheim, Germany). 2 and 8-week diabetics and controls were sacrificed and penises excised. Cohorts of diabetic and control tissues were stored at −80 °C for further molecular analysis. The remaining tissues were fixed in 10% formaldehyde, paraffin embedded and sectioned. 2.2. Human samples All patients gave written informed consent for the removal of cavernous tissue, being this study approved by the local ethics committee and according to the World Medical Association Declaration of Helsinki. 22 diabetic patients with ED [mean age ± standard error (SE); 62.3 ± 1.7] and 10 non-diabetics, non-ED individuals (mean age ± SE; 33.2 ± 4.0) were enrolled in this study. Diabetic individuals, 9 with T1D (mean age ± SE; 59.1 ± 2.6) and 13 with type 2 diabetes (T2D; mean age ± SE; 64.5 ± 2.1) underwent penile implant surgery. Among the 10 non-diabetics, 1 had penile curvature correction and the remaining had procedures for penile enlargement and/or lengthening. During surgical interventions, cavernosal fragments with ≈5– 7.5 mm length/2.5 mm width were harvested and fixed in formaldehyde. Human corporeal tissue was paraffin-embedded and sectioned.
2.3. RNA isolation Total RNA was isolated using TriPure (Roche Applied Science, Indianapolis, USA). Contaminating genomic DNA was removed using the DNA-free™ kit (Ambion, Texas, USA), followed by RNA purification with the RNeasy Mini kit (Qiagen, California, USA). RNA purity and concentration were determined by absorbance reading at 260 nm, 280 nm and 230 nm (NanoDrop™1000 Spectrophotometer; Thermo Scientific, California, USA). RNA integrity was analyzed by denaturing agarose gel electrophoresis. 2.4. Microarray analysis Angiogenic gene expression was evaluated using the Rat Angiogenesis OligoGEArray ® (SuperArray, Maryland, USA). This membrane based-microarray contained a total of 128 oligonucleotide probes, including 113 angiogenic-specific/related genes, 3 blank negative reference spots, 3 artificial sequences, 7 reference genes and 2 prebiotinylated oligonucleotides. ≈3 μg of RNA was used as template to generate/amplify biotin-16-dUTP-labeled-cRNA probes using the TrueLabeling-AMP method (SuperArray). OligoGEArray® membranes were pre-hybridized with GEAhyb solution, followed by hybridization with biotin-labeled-cRNA probes (overnight, 60 °C). Membranes were incubated with alkaline phosphatase-conjugated streptavidin, with the chemiluminescent substrate CDP-Star and exposed to X-ray films. Films were developed, scanned, images acquired and analyzed with the GEArray Analysis Software (SuperArray). Relative expression of each gene was determined after correction for background, subtraction of the global value and normalization, by adjusting to the average intensity signal of the reference gene glyceraldehyde-3-phosphate dehydrogenase (Gapdh). The ratio of gene expression in penile diabetic to control was considered significant when less than 1.0 or greater than 2.0. Microarray evaluation was performed twice for each animal sample. 2.5. Quantitative real-time PCR (qrt-PCR) 1.5 μg of total RNA of 8-week CC samples was reverse transcribed into cDNA using SuperScript™ II Reverse Transcriptase (Invitrogen, California, USA). Each cDNA sample (100 ng/μl) was used in duplicate as template for qrt-PCR using TaqMan® Gene Expression Master Mix and TaqMan® hydrolysis probes (Applied Biosystems, California, USA). 2 target genes were selected from previous microarray experiments: Igf1 (assay_ID Rn99999087_ml) and Npr1 (assay_ID Rn00561678_ml). Amplification in a StepOne Real-Time PCR system (Applied Biosystems), was performed in a final volume of 20 μL, with 1x Gene Expression Master Mix, 250 nM probe and 900 nM primers, using the following cycling conditions: 1 cycle, 50 °C/2 min; 1 cycle, 95 °C/10 s; 45 cycles, 95 °C/15 s and 60 °C/1 min. Raw data was analyzed with Stepone TM Software v2.1 (Applied Biosystems). Quantification cycle (Cq) values were determined for each sample and the comparative Cq method was used to detect relative gene expression ratios normalized to the reference genes, Gapdh (assay_ID Rn01775763_gl) and β-actin (Actb; assay_ID Rn00667869_ml) (Applied Biosystems). Relative mRNA expression was represented as fold change over the calibrator sample (average of gene expression corresponding to the control group) [14]. Relative gene expression changes were calculated using the 2-ΔΔCt method [15]. All samples were prepared and examined in parallel, negative controls were included. 2.6. Immunolabeling and quantification of staining intensity Immunohistochemistry for IGF-1 and NPR-1 was performed in 8week CC and in human cavernosal samples, using the streptavidinbiotin peroxidase method [4]. Slides were incubated overnight with the primary antibodies: cross-reactive rabbit anti-human IGF-1 (1:150;
Â. Castela et al. / Molecular Genetics and Metabolism 105 (2012) 255–262
Abcam, Cambridge, UK); cross-reactive rabbit anti-human NPR-1 (1:900, Abcam), followed by the secondary antibodies (Santa Cruz Biotechnology, California, USA) and with the Avidin–Biotin Complex reagents (DAKO, Glostrup, Denmark). Negative controls were performed by omission of the primary antibody. Reactions were developed using diaminobenzidine (DAB; Sigma-Aldrich) and sections counterstained with hematoxylin. Slides were observed under the Olympus AH3-RFCA microscope and images captured using the Olympus C-35AD-4 camera (Olympus Imaging Europa GmbH, Hamburg, Germany). Signal intensity quantification was performed in 100× digitally captured color images covering entire rat and human CC sections, which were analyzed by ImageJ using color deconvolution (v.1.37a, NIH, Maryland, USA). Color deconvolution was performed by separating the red/green/blue (RGB) image into hematoxylin and brown-colored DAB components [16]. A DAB threshold value of 30 (rat) and 40 (human), separating background DAB from true staining was selected for IGF-1 and NPR-1, by measuring the lowest and the highest mean optical density values in 10 random DAB-positive images. For each image, the sum of the intensity values in the regions of interest (ROI), pixels above the brown threshold, was divided by the ROI total area, producing an average unitless intensity value (“IGF-1 staining” or “NPR-1 staining”). All intensity values within the same group (non-diabetics and diabetics) were averaged to calculate an overall value, a relative measurement of IGF-1 and NPR-1 protein levels.
257
Table 1 Differentially expressed angiogenic genes in 8-week diabetic versus control erectile tissue. Alterations are expressed as the ratio of gene expression in penile diabetic/control rats less than 1.0-fold change. Position GeneBank no 47 12 86
77 11 108
68 69 82 72
NM_178866
Gene description
Insulin like growth factor 1 NM_031530 Chemokine (C-C motif) ligand 2 NM_177927 Serine (or cysteine) peptidase inhibitor, clade F, member 1 NM_013085 Plasminogen activator, urokinase NM_019205 Chemokine (C-C motif) ligand 11 NM_00100151 Tumor necrosis factor ligand superfamily member 12 NM_012613 Natriuretic peptide receptor 1 NM_145098 Neuropilin 1 NM_017043 Prostaglandin-endoperoxide Synthase 1 NM_012801 Platelet-derived growth factor alpha
Gene symbol
Ratio (diabetic/ control)
Igf1
0.043⁎
Ccl2
0.143⁎
Serpinf1 0.159⁎
Plau
0.266⁎
Ccl11
0.423⁎
Tnfsf12
0.423⁎
Npr1
0.424⁎
Nrp1 Ptgs1
0.436 0.640⁎
Pdgfa
0.682
⁎ p b 0.05.
2.7. Dual immunofluorescence Simultaneous expression of IGF-1 and α-smooth muscle actin (α-SMA) and of NPR-1 and α-SMA, was assessed by double immunofluorescence [4]. Briefly, rat and human CC sections were incubated with a mixture of the primary antibodies: cross-reactive rabbit antihuman IGF-1 (1:150; Abcam) and cross-reactive mouse anti-human α-SMA (dilution 1:500; clone ASM, Chemicon, UK); cross-reactive rabbit anti-human NPR-1 (1:900, Abcam) and the aforementioned α-SMA antibody (1:500). Next, slides were incubated with the secondary antibodies: donkey anti-rabbit conjugated with a red fluorochrome (1:1000; Alexa™ 568, Invitrogen, California, USA) and donkey antimouse conjugated with a green fluorochrome (dilution 1:1000; Alexa™ 488, Invitrogen). Negative controls were performed by omission of
the primary antibodies. Nuclei were counterstained with DAPI (4′,6diamidino-2-phenylindole; Invitrogen), slides were observed under the fluorescence microscope (Imager.Z1, Zeiss, Germany) and images captured (Carl Zeiss MicroImaging GmbH, Germany).
2.8. Statistics Data is presented as mean ± standard error (SE). Statistical differences between 8-week non-diabetic and diabetic erectile tissue, as evaluated by microarray analysis, qrt-PCR and quantitative immunohistochemistry, were determined using the Mann–Whitney test (p b 0.05 considered statistically significant).
Fig. 1. Microarray evaluation. (A) Location of the genes included in the Rat Angiogenesis OligoGEArrays (SuperArray, Maryland, USA). This membrane based-microarray contained a total of 128 oligonucleotide probes, including 113 angiogenic-specific/related genes, 3 blank negative reference spots (115–117), 3 artificial sequences (118–120), 7 reference genes (1, 121–126) and 2 pre-biotinylated oligonucleotides (127, 128). (B) Representative membrane array image of the overall expression of the angiogenic-related genes in 8-week control and diabetic animals. Ccl2—chemokine (C-C motif) ligand 2; Ccl11—chemokine (C-C motif) ligand 11; Fgf1,2,6—fibroblast growth factor 1,2,6; Fgfr3—fibroblast growth factor receptor 3; Figf—c-fos induced growth factor; Flt1—vascular endothelial growth factor receptor 1; Igf1—insulin like growth factor 1; Npr1—natriuretic peptide receptor 1; Nrp1—neuropilin 1; Pdgfa—platelet-derived growth factor alpha; Plau—plasminogen activator, urokinase; Ptgs1—prostaglandin-endoperoxide Synthase 1; Serpinf1—serine (or cysteine) peptidase inhibitor, clade F, member 1; Tnfsf12—tumor necrosis factor ligand superfamily member 12; Vegfa,b,c—vascular endothelial growth factor a,b,c.
258
Â. Castela et al. / Molecular Genetics and Metabolism 105 (2012) 255–262
Fig. 2. Confirmation of microarrays results for Igf1 and Npr1 genes by qrt-PCR. (A) Igf1 was significantly down-regulated in 8-week diabetic CC as compared to controls (p = 0.003); (B) Npr1 gene presented a significant reduced expression in 8-week diabetic penile tissue (p= 0.0021). Relative quantification was achieved by normalizing the expression of Igf1 and Npr1 to Gapdh and Actb by the 2-ΔΔCt method [15]. Data presented as Mean ± SE.
3. Results 3.1. Microarrays analysis We analyzed angiogenic gene expression alterations in CC in an early and established stage of diabetes. Control and diabetic tissues, 2 and 8-week diabetes, were evaluated using the Rat Angiogenesis
OligoGEArray (Fig. 1A) and GEArray analyzer. This software allowed us to perform a global qualitative and semi-quantitative expression analysis of 113 angiogenic genes directly and indirectly involved in angiogenesis pathways. As determined by repetitive microarray evaluations, at 2-week diabetes there was no significant differential gene expression between non-diabetic and diabetic CC, as expressed by the ratio of relative gene expression of diabetic/non-diabetic. At 8 weeks of diabetes, there was a decrease in the expression of several genes as observed in the representative membranes in Fig. 1B. Membranes from 8-week animals presented a similar profile of altered genes, however among different samples there was some degree of variability. A number of genes involved directly in the angiogenic process, as members of the fibroblast growth factor (Fgf) and the vascular endothelial growth factor (Vegf) families presented undetectable physiological levels in both 8-week controls and diabetic samples. The Vegf receptor 1, Flt1, had also an extremely low expression, unnoticeable in 8-week tissues. The Vegf receptor 2, Kdr, presented a slight decrease in diabetic CC, which was not overall statistically significant (p = 0.57), since differential expression was considered only when genes presented levels of expression less than 1.0-fold. It was the case of a total of 10 genes which exhibited expression of less than 1.0-fold in STZinduced erectile tissue samples when compared to controls (Table 1). The category of altered genes included: angiogenesis (Nrp1, Serpinf1, Igf1, Pdgfa), guanilate cyclase (GC) activity/cyclic guanosine monophosphate (cGMP) biosynthesis (Npr1), chemokine activity (Ccl2, Ccl11, Tnfsf12), proteolysis (Plau) and inflammation (Ptgs1).
Fig. 3. Quantification and localization of IGF-1 protein in 8-week control and diabetic penile tissue. (A, B) IGF-1 immunoreactivity was quantified in IGF-1/DAB stained cavernosal samples using ImageJ color deconvolution plugin [16]. (A) Color deconvolution was performed by separating the red/green/blue (RGB) image into hematoxylin and brown-colored DAB components [16]. A DAB threshold value was selected, and the signal above threshold was considered as IGF-1 positive staining; (B) All intensity values within the same group (non-diabetics and diabetics) were averaged to calculate an overall value, a relative measurement of IGF-1protein levels. A significant reduction in IGF-1 production was confirmed in 8-week diabetic penile tissue; (C) IGF-1 expression pattern was evaluated by double immunofluorescence for IGF-1/α-SMA. IGF-1 was mostly identified in the corporeal nondiabetic and diabetic SMC counterpart, although it could also be detected in trabecular fibroblasts. DAB, diaminobenzidine; DAPI, 4′,6-diamidino-2-phenylindole (DNA intercalator). Scale bar = 100 μm. Data presented as Mean ± SE; **P ≤ 0.01.
Â. Castela et al. / Molecular Genetics and Metabolism 105 (2012) 255–262
3.2. qrt-PCR validation of selected genes qrt-PCR was used to confirm decreased expression levels of Igf1 and Npr1 in diabetic CC. These genes were selected based on their biological function and potential role in vascular dysfunction associated to diabetic-ED. qrt-PCR revealed that both Igf1 and Npr1 mRNA expressions were down-regulated by three-fold (p = 0.003 and p = 0.0021, respectively) in 8-week diabetic penile tissue when compared to controls (Fig. 2A and B); consistent with the results obtained by microarrays. These results were further confirmed at protein level. 3.3. Immunohistochemistry and protein quantitative evaluation To confirm at protein level the data obtained by molecular evaluation, we quantified IGF-1 and NPR-1 immunoreactivity staining in DAB-labeled sections at 8-week diabetes, using ImageJ color deconvolution method [18]. Additionally, we identified rat cavernosal components expressing IGF-1 and NPR-1, by performing dual immunofluorescences for IGF-1/α-SMA and NPR-1/α-SMA. Corroborating the array analysis, we observed that IGF-1 immunoreactivity in IGF-1/DAB-stained samples (Fig. 3A) was significantly decreased (pb 0.01) in 8-week diabetics (73.5 ± 2.3) when compared to controls (89.2 ± 4.1) (Fig. 3B). Regarding the expression profile, in non-diabetic penile tissue, IGF-1 was abundantly produced by SMCs, with some expression in trabecular fibroblasts (Fig. 3C, upper panel). Diabetic corporeal tissue presented an analogous IGF-1 pattern, mostly in the
259
SM layer (Fig. 3C, lower panel). In diabetic penis, NPR-1 protein levels were also significantly reduced, as quantified by NPR-1/DAB-labeling immunopositivity (Fig. 4A) with 95.4 ± 3.0 in controls versus 77.4 ± 4.2 in diabetics; p b 0.01 (Fig. 4B). NPR-1 protein was diffusely detected in non-diabetic trabecular fibroblasts and SMCs (Fig. 3D, upper panel). In the diabetic penis, NPR-1 profile was similar to what was observed in control tissues (Fig. 3D, lower panel). 3.4. Quantitative immunohistochemistry in human samples To verify if IGF-1 and NPR-1 decreased expression levels in diabetic CC were not restricted to rat samples, we performed analogous experiments and quantitative evaluation in human cavernosal non-diabetic and diabetic samples. Consistent with the results obtained in experimental erectile tissue, human diabetic CC presented significantly decreased IGF-1 protein levels, as quantified in IGF-1/DAB-stained slides (Fig. 5A) with mean values of 112.0 ± 2.0 in non-diabetics and 94.1 ± 2.1 in diabetic patients (pb 0.001) (Fig. 5B). Similarly, IGF-1 was predominantly detected in the SMC layer of human non-diabetic CC, as evaluated by double immunofluorescence for IGF-1/α-SMA (Fig. 5C, upper panel). In human diabetic CC, IGF-1 had a similar expression profile when compared with controls (Fig. 5C, lower panel). In NPR-1/DAB stained slides (Fig. 6A), we observed that protein levels were diminished in regard to controls (141.9 ± 4.1, non-diabetics; 133.1 ± 1.9, diabetics; p b 0.05) (Fig. 6B). In non-diabetic corporeal tissue, NPR-1 presented a diffuse pattern of expression, mostly identified in trabecular fibroblast-like cells and SMCs (Fig. 6C, upper panel). This staining
Fig. 4. NPR-1 protein quantification and expression pattern in 8-week control and diabetic cavernosal samples. (A, B) NPR-1 immunopositivity was quantified in NPR-1/DAB stained cavernosal samples using ImageJ color deconvolution plugin [16]. (A) Color deconvolution was performed as abovementioned for IGF-1 [16]; (B) Mean values of NPR-1 in non-diabetic and diabetic rat penile tissue, confirming the significant decrease in NPR-1 protein expression in diabetic corporeal tissue; (C) NPR-1 expression profile as assessed by double immunofluorescence for NPR-1/α-SMA, demonstrated a diffused expression in trabecular fibroblasts and SMCs of both control and diabetic CC. DAB, diaminobenzidine; DAPI, 4′,6-diamidino-2-phenylindole (DNA intercalator). Scale bar = 100 μm. Data presented as Mean ± SE; **P ≤ 0.01.
260
Â. Castela et al. / Molecular Genetics and Metabolism 105 (2012) 255–262
Fig. 5. IGF-1 protein evaluation in human non-diabetic and diabetic cavernosal samples. (A, B) IGF-1 protein levels were quantified in IGF-1/DAB stained corporeal tissues using ImageJ color deconvolution. (B) Quantitative immunohistochemistry in IGF-1/DAB stained slides showed a significant reduction in protein expression in human diabetic erectile tissue; (C) Double immunofluorescence for IGF-1/α-SMA demonstrated positive labeling in cavernosal SMCs of both non-diabetic and diabetic samples. In certain vascular spaces IGF-1 was clearly identified lining the sinosoids (upper panel; white arrow). DAB, diaminobenzidine; DAPI, 4′,6-diamidino-2-phenylindole (DNA intercalator). Scale bar = 100 μm. Data presented as Mean ± SE; ***p b 0.001.
pattern was maintained in diabetic CC samples (Fig. 6C, lower panel). Interestingly, when separating the overall group of diabetics in T1D and T2D, no differences were attained for IGF-1 and NPR-1 localization and quantification. 4. Discussion Diabetic men have approximately a three-fold increased risk for the development of ED, compared with healthy individuals [17]. It is known that elevated production of AGEs, increased oxidative stress, and impairment of NO synthesis, affect penile ECs, leading to EDys and ED [2,4,8,9]. However, besides these mechanisms, there is very scarce information on how diabetes detrimentally affects the production of endothelial mediators, contributing to EDys. We evaluated in different stages of diabetes, STZ-induced 2 and 8 weeks T1D, the range of angiogenic-related molecules differentially expressed in erectile tissue, using a membrane-based microarray system. A similar molecular approach using affymetrix chips, was previously used in studies of diabetic-ED [19,20]. However, one report analyzed in established diabetes a wide range of genes involved in multiple biological functions [19], whereas the other focused in the analysis of penile SMCrelated changes in early diabetes [20]. Our study was the first to evaluate alterations in cavernosal vascular-related genes in two different stages of diabetes. At 2-week we did not observe significant changes in genes associated with vascular function. We may assume that at this early time point, diabetic-induced metabolic deregulations have not yet induced significant alterations at gene level. Additionally, considering that diabetic-related hemodynamic changes in erectile function occur
at later stages of the disease [21], it is valid to assume that endothelial major alterations would occur with diabetes progression. At 8-week diabetes we identified 10 angiogenic-related genes down-regulated in diabetic CC. The functional category of altered genes included (Table 1): angiogenesis (Nrp1, Serpinf1, Igf1, Pdgfa), GC activity/cGMP biosynthesis (Npr1), chemokine activity (Ccl2, Ccl11, Tnfsf12), proteolysis (Plau) and inflammation (Ptgs1). Members of the VEGF family of ligands and receptors present in the microarray membranes were barely detected at physiological levels, both in control and diabetic samples and no significant alterations were observed among groups. Our results regarding VEGF expression are divergent to what was previously reported in an obese/T2D rat model [11]. This discrepancy may be due to the experimental model/rat strain and the detection methods. Regarding the down-regulated genes, a reduced expression of Igf1[22], Pdgfa[23], Npr1[24] and Ptgs1[25], has been associated with diabetic complications in other organ systems; suggesting the existence of common molecular pathways underlying diabetes. Among our gene alterations, we validated by qrtPCR Igf1 and Npr1, chosen due to their relevant function in vasorelaxation mechanisms. Reduced Igf1 expression may have deleterious effects in diabetic erectile function, since it has been reported that Igf1 adenoviral gene transfer to STZ-diabetic CC ameliorated ED [26]. In fact, Igf1 displays relevant biological roles as angiogenic mediator and stimulator of myogenesis, crucial events for penile endothelial homeostasis and SMC function [27,28]. The Npr1 gene encodes for a membrane bound GC receptor with a tyrosine kinase domain, involved in cGMP biosynthesis [29]. Npr1 was shown to promote vascular regeneration [29] and due to its role in cGMP formation, plays a relevant role in SMC vasorelaxation events [30]. Next, we
Â. Castela et al. / Molecular Genetics and Metabolism 105 (2012) 255–262
261
Fig. 6. NPR-1 protein analysis in human non-diabetic and diabetic penile tissue. (A, B) Quantitative immunohistochemistry in NPR-1/DAB using ImageJ color deconvolution; (B) NRP-1/DAB protein levels were significantly decreased in human diabetic erectile tissue; (C) NPR-1 expression profile was diffuse in non-diabetic and diabetic trabecular fibroblasts and SMCs. DAB, diaminobenzidine; DAPI, 4′,6-diamidino-2-phenylindole (DNA intercalator). Scale bar = 100 μm. Data presented as Mean ± SE; *p b 0.05.
confirmed the microarray data at protein level by quantitative immunohistochemistry [16] and dual immunofluorescence for IGF-1/ α-SMA and NPR-1/ α-SMA. In agreement with its reported roles in SM proliferation/myogenesis and angiogenesis, IGF-1 was mostly detected in SMCs and in some ECs-lining sinosoids [27,28]. NPR-1 was particularly identified in trabecular fibroblasts and SMCs. SMC expression of NPR-1 is consistent with its role in cGMP formation and cavernosal vasorelaxation [30]. In corporeal trabecular fibroblasts, NPR-1 protein expression was never described. It is established that decreased SMCs and expansion of collagen producing trabecular fibroblasts is one of diabetes-induced pathophysiological changes [12]. As a recent publication suggested that NPR-1 has inhibitory effects in renal fibrosis in mice [31], we hypothesized the same role in penile trabeculae. We may speculate that lower NPR-1 expression in diabetic trabeculae may potentially contribute to cavernosal fibrosis. Finally, we verified that IGF-1 and NPR-1 protein levels were decreased in human diabetic CC and that the expression pattern of these proteins was similar to rat penile tissue. Taken together, our study was the first to evaluate angiogenic-related gene alterations in STZ-induced diabetic erectile tissue, using microarrays. We reported a downregulation in Igf1 and Npr1 gene expression in diabetic CC, which was later confirmed at protein level both in rat and human diabetic erectile tissue. In conclusion, we demonstrated a down-regulation in 10 angiogenicrelated genes in diabetic CC of an established STZ-model. Among them, Igf1 and Npr1 were selected due to their relevant roles in cavernosal endothelial and SMC components. A reduced expression in IGF-1 and NPR-1 was also detected in human diabetic erectile tissue, as compared to non-diabetic CC. These data indicate that IGF-1 and NPR-1
in penile tissue may play important roles in the maintenance of corporeal homeostasis/erectile process, and should be taken into account as potential targets in diabetic-ED. Conflict of interest The authors disclose any financial and personal relationships with other people or organizations that could inappropriately influence this work. Acknowledgments This study was supported by ESSM Grant for Medical Research 2007, by the Portuguese Foundation for Science and Technology (PTDC/SAUOSM/65599/2006; SFRH/BPD/40554/2007) and by the Associação Portuguesa de Urologia (Prémio Abbott Urologia 2008). References [1] I. Sáenz de Tejada, J. Angulo, S. Cellek, N. González-Cadavid, J. Heaton, R. Pickard, U. Simonsen, Pathophysiology of erectile dysfunction, J. Sex. Med. 2 (2005) 26–39. [2] J. Zúrová-Nedelcevová, J. Navarová, K. Drábiková, V. Jancinová, M. Petríková, I. Bernátová, V. Kristová, V. Snirc, V. Nosál'ová, R. Sotníková, Participation of reactive oxygen species in diabetes-induced endothelial dysfunction, Neuro Endocrinol. Lett. 27 (2006) 168–171. [3] W. Bakker, E.C. Eringa, P. Sipkema, V.W. van Hinsbergh, Endothelial dysfunction and diabetes: roles of hyperglycemia, impaired insulin signaling and obesity, Cell Tissue Res. 335 (2009) 165–189. [4] C. Costa, R. Soares, A. Castela, S. Adães, V. Hastert, P. Vendeira, R. Virag, Increased endothelial apoptotic cell density in human diabetic erectile tissue — comparison with clinical data, J. Sex. Med. 6 (2009) 826–835.
262
Â. Castela et al. / Molecular Genetics and Metabolism 105 (2012) 255–262
[5] M.M. Hartge, U. Kintscher, T. Unger, Endothelial dysfunction and its role in diabetic vascular disease, Endocrinol. Metab. Clin. North Am. 35 (2006) 551-5ix. [6] M. Kirby, G. Jackson, U. Simonsen, Endothelial dysfunction links erectile dysfunction to heart disease, Int. J. Clin. Pract. 59 (2005) 225–229. [7] J. Xu, M.H. Zou, Molecular insights and therapeutic targets for diabetic endothelial dysfunction, Circulation 120 (2009) 1266–1286. [8] A. Nessar, Advanced glycation end products—role in pathology of diabetic complications, Diabetes Res. Clin. Pract. 67 (2005) 3–21. [9] B. Musicki, M.F. Kramer, R.E. Becker, A.L. Burnett, Inactivation of phosphorylated endothelial nitric oxide synthase (Ser-1177) by O-GlcNAc in diabetes-associated erectile dysfunction, Proc. Natl. Acad. Sci. USA 102 (2005) 11870–11875. [10] J. Folkman, Fundamental concepts of the angiogenic process, Curr. Mol. Med. 3 (2003) 643–651. [11] S. Jesmin, I. Sakuma, A. Salah-Eldin, K. Nonomura, Y. Hattori, A. Kitabatake, Diminished penile expression of vascular endothelial growth factor and its receptors at the insulin-resistant stage of a type II diabetic rat model: a possible cause for erectile dysfunction in diabetes, J. Mol. Endocrinol. 31 (2003) 401–418. [12] M. Yamanaka, M. Shirai, H. Shiina, Y. Tanaka, H. Enokida, A. Tsujimura, K. Matsumiya, A. Okuyama, R. Dahiya, Vascular endothelial growth factor restores erectile function through inhibition of apoptosis in diabetic rat penile crura, J. Urol. 173 (2005) 318–323. [13] H.R. Jin, W.J. Kim, J.S. Song, S. Piao, M. Tumurbaatar, S.H. Shin, M.J. Choi, B. Tuvshintur, K.M. Song, M.H. Kwon, G.N. Yin, G.Y. Koh, J.K. Ryu, J.K. Suh, Intracavernous delivery of synthetic angiopoietin-1 protein as a novel therapeutic strategy for erectile dysfunction in the type II diabetic db/db mouse, J. Sex. Med. 7 (2010) 3635–3646. [14] V. Catalán, J. Gómez-Ambrosi, A. Rodríguez, C. Silva, F. Rotellar, M.J. Gil, J.A. Cienfuegos, J. Salvador, G. Frühbeck, Expression of caveolin-1 in human adipose tissue is upregulated in obesity and obesity-associated type 2 diabetes mellitus and related to inflammation, Clin. Endocrinol. (Oxf) 68 (2008) 213–219. [15] K.J. Livak, T.D. Schmittgen, Analysis of relative gene expression data using real-time quantitative PCR and the 2(ΔΔCT) method, Methods 25 (2001) 402–408. [16] A.C. Ruifrok, D.A. Johnston, Quantification of histochemical staining by color deconvolution, Anal. Quant. Cytol. Histol. 23 (2001) 291–299. [17] J. Dey, M.D. Shepherd, Evaluation and treatment of erectile dysfunction in men with diabetes mellitus, Mayo Clin. Proc. 77 (2002) 276–282. [18] P.V. Babu, K.E. Sabitha, C.S. Shyamaladevi, Therapeutic effect of green tea extract on advanced glycation and cross-linking of collagen in the aorta of streptozotocin diabetic rats, Clin. Exp. Pharmacol. Physiol. 33 (2006) 351–357. [19] C.J. Sullivan, T.H. Teal, I.P. Luttrell, K.B. Tran, M.A. Peters, H. Wessells, Microarray analysis reveals novel gene expression changes associated with erectile dysfunction in diabetic rats, Physiol. Genomics 23 (2005) 192–205.
[20] J.D. Hipp, K.P. Davies, M. Tar, M. Valcic, A. Knoll, A. Melman, G.J. Christ, Using gene chips to identify organ-specific, smooth muscle responses to experimental diabetes: potential applications to urological diseases, BJU Int. 99 (2007) 418–430. [21] T.J. Bivalacqua, M.F. Usta, H.C. Champion, S. Leungwattanakij, P.A. Dabisch, D.B. McNamara, P.J. Kadowitz, W.J. Hellstrom, Effect of combination endothelial nitric oxide synthase gene therapy and sildenafil on erectile function in diabetic rats, Int. J. Impot. Res. 16 (2004) 21–29. [22] H. Yang, A.J. Scheff, D.S. Schalch, Effects of streptozotocin-induced diabetes mellitus on growth and hepatic insulin-like growth factor I gene expression in the rat, Metabolism 39 (1990) 295–301. [23] Q.H. Wu, W.S. Chen, Q.X. Chen, J.H. Wang, X.M. Zhang, Changes in the expression of platelet-derived growth factor in astrocytes in diabetic rats with spinal cord injury, Chin. Med. J. (Engl) 123 (2010) 1577–1581. [24] V.G. Marrachelli, F.J. Miranda, J.M. Centeno, I. Miranda, M. Castelló-Ruiz, M.C. Burguete, T. Jover-Mengual, J.B. Salom, G. Torregrosa, E. Alborch, Mechanisms underlying the diabetes-induced hyporeactivity of the rabbit carotid artery to atrial natriuretic peptide, Pharmacol. Res. 63 (2011) 190–198. [25] C. Fang, Z. Jiang, D.R. Tomlinson, Expression of constitutive cyclo-oxygenase (COX-1) in rats with streptozotocin-induced diabetes; effects of treatment with evening primrose oil or an aldose reductase inhibitor on COX-1 mRNA levels, Prostaglandins Leukot. Essent. Fatty Acids 56 (1997) 157–163. [26] W.Y. Pu, L.Q. Hu, H.P. Wang, Y.X. Luo, X.H. Wang, Improvement in erectile dysfunction after insulin-like growth factor-1 gene therapy in diabetic rats, Asian J. Androl. 9 (2007) 83–91. [27] E.D. Rabinovsky, R. Draghia-Akli, Insulin-like growth factor I plasmid therapy promotes in vivo angiogenesis, Mol. Ther. 9 (2004) 46–55. [28] M. Kim, E.C. Hwang, I.K. Park, K. Park, Insulin-like growth factor-1 gene delivery may enhance the proliferation of human corpus cavernosal smooth muscle cells, Urology 76 (2010) 511.e5-9. [29] M. Kuhn, K. Völker, K. Schwarz, J. Carbajo-Lozoya, U. Flögel, C. Jacoby, J. Stypmann, M. van Eickels, S. Gambaryan, M. Hartmann, M. Werner, T. Wieland, J. Schrader, H. A. Baba, The natriuretic peptide/guanylyl cyclase-a system functions as a stressresponsive regulator of angiogenesis in mice, J. Clin. Invest. 119 (2009) 2019–2030. [30] E. Andrade, P.M. Andrade, R.C. Borra, J. Claro, M. Srougi, cDNA microarray analysis of differentially expressed genes in penile tissue after treatment with tadalafil, BJU Int. 101 (2008) 508–512. [31] T. Nishikimi, C. Inaba-Iemura, K. Ishimura, K. Tadokoro, S. Koshikawa, K. Ishikawa, K. Akimoto, Y. Hattori, K. Kasai, N. Minamino, N. Maeda, H. Matsuoka, Natriuretic peptide/natriuretic peptide receptor-A (NPR-A) system has inhibitory effects in renal fibrosis in mice, Regul. Pept. 154 (2009) 44–53.