Relationship between Arginase 1 and Arginase 2 levels and genetic polymorphisms with erectile dysfunction

Nitric Oxide 51 (2015) 36e42

Contents lists available at ScienceDirect

Nitric Oxide journal homepage: www.elsevier.com/locate/yniox

Relationship between Arginase 1 and Arginase 2 levels and genetic polymorphisms with erectile dysfunction Riccardo Lacchini a, Jaqueline J. Muniz a, Yuri T.D.A. Nobre b, Adauto J. Cologna b, Antonio C.P. Martins b, Jose E. Tanus-Santos c, * a Department of Psychiatric Nursing and Human Sciences, Ribeirao Preto College of Nursing, University of Sao Paulo, Av. Bandeirantes, 3900, 14049-900 Ribeirao Preto, SP, Brazil b Department of Surgery and Anatomy, Faculty of Medicine of Ribeirao Preto, University of Sao Paulo, Av. Bandeirantes, 3900, 14049-900 Ribeirao Preto, SP, Brazil c Department of Pharmacology, Faculty of Medicine of Ribeirao Preto, University of Sao Paulo, Av. Bandeirantes, 3900, 14049-900 Ribeirao Preto, SP, Brazil

a r t i c l e i n f o

a b s t r a c t

Article history: Received 24 July 2015 Received in revised form 23 October 2015 Accepted 26 October 2015 Available online xxx

Arginase 1 and Arginase 2 are homologous enzymes that convert L-Arginine to Urea and L-ornithine and compete with nitric oxide synthases for L-Arginine. Increased Arginase 1 and 2 activity may reduce nitric oxide production by the endothelium in disease states, including erectile dysfunction (ED). Here we aimed at assessing whether Arginase 1 and 2 plasma levels, plasma arginase activity, or genetic factors are associated with ED risk and severity. Blood samples were collected from healthy controls (n ¼ 106) and from patients with ED (n ¼ 110) after completion of the IIEF questionnaire (international index of erectile function). Plasma Arginase 1 and 2 concentrations were assessed by ELISA, while plasma arginase activity was measured by spectrophotometry. Genotypes of ARG1 (rs2781659, rs2781667, rs2246012 and rs17599586) and ARG2 (rs3742879 and rs10483801) were determined by Taqman genotyping assays by real-time polymerase chain reaction. Increased Arginase 2 concentrations were found in clinical ED and are associated with increased risk for ED. ARG1 rs2781659 AA and rs2781667 TT genotypes are associated with lower IIEF scores (higher severity) only in clinical ED. Similarly, the ARG1 GTCC haplotype is associated with higher IIEF scores in clinical ED. This study shows that plasma Arginase 2 concentrations may serve as risk factor for ED. Besides, Arginase 1 genetic variations affect ED severity. © 2015 Elsevier Inc. All rights reserved.

Keywords: Arginase Nitric oxide Erectile dysfunction ARG1, ARG2 Genetic polymorphisms

1. Introduction Erectile dysfunction (ED) is a common disease that impacts health and social aspects of affected patients [1]. This multifactorial disease [2] has different etiologic origins including neurogenic, vasculogenic, and endocrine mechanisms [3]. Vasculogenic ED is commonly associated with cardiovascular diseases (CVD) [4,5] and patients with CVD are exposed to increased risk of developing ED [6]. Indeed, ED is now considered an early marker for future CVD [4] because it reflects endothelial dysfunction and impaired relaxation of smooth muscle cells [7] in patients with ED and CVD. Nitric oxide (NO) is one of the most important molecules involved in the physiology of erection and its bioactivity usually

* Corresponding author. E-mail addresses: [email protected], [email protected] (J.E. TanusSantos). http://dx.doi.org/10.1016/j.niox.2015.10.003 1089-8603/© 2015 Elsevier Inc. All rights reserved.

decreases in CVD [2,8]. While this molecule is produced by three different isoforms of NO synthases (NOS), two isoforms (endothelial and neuronal NOS, eNOS and nNOS, respectively) are considered of paramount importance for normal erectile function [2,9e13]. All NOS enzymes utilize L-arginine as substrate for NO synthesis [2,11]. However, this substrate is also used by other enzymes ubiquitously expressed in the body, Arginase 1 and Arginase 2 [11], which consume L-arginine producing L-ornithine and urea [14,15]. They are expressed in the corpus cavernosum [16e18] and co-localize with eNOS in the endothelium [19]. Importantly, Arginases 1 and 2 compete with NOS for L-Arginine, and therefore the higher the arginase activity, the lower the NO production will be as a result of local microenvironmental substrate exhaustion [14,15,20,21]. This competition has already been shown in several different conditions [15,20e23] including experimental erectile dysfunction [24,25]. Furthermore, inhibition of Arginase activity improved erectile function in animal models of ED [18,24,26e29]. Since NO is the main molecule responsible for starting and

R. Lacchini et al. / Nitric Oxide 51 (2015) 36e42

maintaining erection, and its production is reduced in ED [2], it is possible that increased arginase expression or activity contribute to the pathogenesis of ED. While increased arginase levels has been shown in animal models and in humans with cardiovascular dysfunction [16,22,23], it is unknown whether plasma levels or activity of Arginase may predict ED risk. In addition, recent evidence suggests that plasma Larginine hydrolysis by arginases limits the amounts of L-arginine available to promote endothelial NO synthesis [30], even though it is not clear to which extent endothelial cells rely on plasma Larginine to produce NO. While several studies have shown that genetic variations in genes encoding the components of the NOecGMP pathway may affect the pathogenesis of ED and therapeutic responses to drugs [2,8,12,13,31,32], no previous study has examined whether polymorphisms in the genes encoding Arginase 1 and Arginase 2 (ARG1 and ARG2, respectively) are associated with ED. This study aims at assessing whether plasma levels of Arginase 1 and 2, or Arginase activity, are associated with ED and predict the degree of disability. In addition, this study aims at examining whether those possible associations are influenced by genotypes or haplotypes of ARG1 and ARG2 genes. 2. Materials and methods 2.1. Study population Patients included in this study (n ¼ 110) were enrolled from the ~o Preto Urology outpatient Clinic of the University Hospital, Ribeira Medical School. Healthy controls (n ¼ 106) were enrolled from the ~o Preto of the University of Sa ~o population of the campus of Ribeira Paulo, and not related to patients. This study was approved by the Institutional Review Board at the Ribeir~ ao Preto Medical School and informed consent was obtained from all subjects. All patients underwent physical and history examination. Inclusion criteria for the Clinical ED group were age between 40 and 80 years, complaints regarding sexual activity, and medical diagnosis of erectile dysfunction. Exclusion criteria were hormonal disorders (testosterone, hypophysis, thyroid hormones), psychiatric disorders, neurogenic bladder dysfunction, hypogonadism, penile implants, cerebrovascular accident, central nervous system trauma, and anatomical abnormalities such as Peyronie's disease. We have not excluded diabetic patients. The exclusion criteria enriched our Clinical ED group with vasculogenic ED patients because other hormonal, neurogenic, and anatomical causes for ED were excluded [8,13,33,34]. Inclusion criteria for the healthy subjects were absence of any diseases and exclusion criteria was International Index for Erectile Function (erectile function domain) score below 25. Erectile function was evaluated in both groups using the International Index for Erectile Function (IIEF) questionnaire, the erectile function (EF) domain [35]. Venous blood samples were drawn from all included participants after overnight fasting and plasma was separated by centrifugation at 2000 g for 15 min, immediately frozen, and stored at 70  C. Genomic DNA was extracted from whole blood samples by a salting-out method and stored at 20  C until analyzed. 2.2. Laboratorial analyses Serum lipid profile (total cholesterol, triglycerides, high-density lipoprotein cholesterol), glucose, testosterone, urea, and creatinine concentrations were measured with commercially available kits using standard techniques. Low-density lipoprotein concentration was calculated according to Friedewald's formula.

37

2.3. Enzyme linked immunoassays (ELISAs) of Arginase 1 and Arginase 2 and Arginase activity Human Arginase 1 and Arginase 2 concentrations were measured in plasma samples using commercially available ELISA kits (MBS912500 and MBS2021960 respectively, purchased from MyBioSource; San Diago, CA, USA), following the manufacturer's instructions. To assess plasma Arginase activity, plasma samples were depleted of urea using 10 KDa membrane filter columns (Vivaspin 500 model VS0102, Sartorius, Goettingen, Germany), which were centrifuged at 14,000 g for 30 min at 4  C. The samples were then reconstituted using Mili-Q water and analyzed by the QuantiChrom Arginase Assay kit (BioAssay Systems, Hayward, CA, USA), following the manufacturer's instructions. 2.4. Genotype determinations Genotypes for the four polymorphisms in ARG1 and two polymorphisms in ARG2 were determined by Taqman Allele Discrimination Assays (Applied Biosystems, Foster City, CA, USA). Assay IDs were: C___3063957_10 (rs2781659), C__15933286_10 (rs2781667), C__15933284_10 (rs2246012), C__25596209_10 (rs17599586), C__25960528_10 (rs3742879) and C___2778311_10 (rs10483801). All experiments were performed with JumpStart Taq ReadyMix for Quantitative PCR 1x (SigmaeAldrich, St Louis, MO, USA), Taqman assays 1x and 5 ng of template in 10 uL reaction volume. Thermal cycling was performed under standard conditions and fluorescence was recorded by the StepOne Plus Real Time PCR equipment (Applied Biosystems, Foster City, CA, USA). The results were analyzed with manufacturer's software. 2.5. Haplotype inference Haplotypes were estimated using the program PHASE version 2.1 (http://stephenslab.uchicago.edu/software.html#phase). The possible ARG1 gene haplotypes including the four polymorphisms studied here in (rs2781659A>G, rs2781667C>T, rs2246012T>C and rs17599586C>T) were ACTC, ACTT, ACCC, ACCT, ATTC, ATTT, ATCC, ATCT, GCTC, GCTT, GCCC, GCCT, GTTC, GTTT, GTCC, and GTCT. The possible ARG2 gene haplotypes including the two polymorphisms studied here (rs3742879A>G and rs10483801C>A) were AC, AA, GC and GA. Haplotypes with frequencies lower than 5% in any group were not taken into consideration for the statistical analysis. 2.6. Statistical analysis The clinical features and the biochemical parameters of the studied groups were compared by unpaired t-test, by ManneWhitney test, or by c2-test, when appropriate. Deviations from HardyeWeinberg equilibrium were assessed by c2-test. A P < 0.05 was considered as statistically significant in all analyses. The effects of genotypes on disease risk, IIEF score and on arginase markers (arginase 1 and 2 levels and arginase activity) were assessed by multivariate logistic regression analysis or multivariate linear regression analysis. Univariate and bivariate analyses were used to assess for the potential confounding influence of each covariate on the relationship between ARG1 and ARG2 genetic markers, disease risk, Arginase 1 and Arginase 2 levels, Arginase activity, or IIEF score. Variables of clinical importance identified by the bivariate approach were included in the final multivariate regression models. Calculations were made with JMP software, version 5.1.2 (SAS institute, Cary, NC, USA). The main dependent variables taken into consideration in this study were disease status and IIEF score. Age, smoking status and ethanol consumption were used as

38

R. Lacchini et al. / Nitric Oxide 51 (2015) 36e42

independent variables. Other variables including race and body mass index were tested but not included in the final regression models. Plasma levels of Arginase 1 and 2 and Arginase activity were used either as dependent or independent variables, depending on the scope of analysis. 3. Results Clinical features of included subjects are shown in Table 1. Significant between groups differences were found in age, smoking status, systolic blood pressure, and biochemical parameters such as cholesterol and fasting glucose (all P < 0.05). However, the groups were similar with respect to the other parameters shown in Table 1 (P > 0.05). Fig. 1 shows plasma Arginase 1 and Arginase 2 concentrations and Arginase activity in the studied groups. While Arginase 1 and Arginase Activity were not different between groups, higher Arginase 2 levels were found in patients with ED (P ¼ 0.009). Accordingly, multivariate logistic regression analysis (Table 2) showed that Arginase 2 levels are positively associated with ED risk (P ¼ 0.010, OR ¼ 7.9, 95% confidence interval: 1.7 to 39.1), whereas Arginase 1 concentrations and Arginase Activity showed no significant effect. No associations were found between plasma Arginase 1 and 2 levels or plasma Arginase activity and changes in IIEF scores (Supplementary Table 1). Supplementary Table 2 shows the genotypes distributions for ARG1 and ARG2 polymorphisms in the Control and Clinical ED groups. All genotypes frequencies were in HardyeWeinberg equilibrium. No associations were found between genotypes and disease risk, even after correction for confounding factors. Supplementary Table 3 shows the haplotypes distributions for ARG1 and ARG2 polymorphisms in the Control and Clinical ED groups. Again, no significant association was found between haplotypes and ED after correction for covariates. Interestingly, however, both genotypes and haplotypes were associated with disease severity reflected by alterations in the IIEF score (Tables 3 and 4).

Table 1 Clinical features of the Control and Clinical Erectile Dysfunction (ED) groups.

n Age (years) Ethnicity (White/non-White) Waist circumference (cm) BMI (kg m2) Smoking Never smoker (n) Ex-smoker (n) Current smoker (n) Ethanol consumption Higher than 30 g/day Lower than 30 g/day SBP (mm Hg) DBP (mm Hg) HDL (mg dl1) LDL (mg dl1) Total cholesterol (mg dl1) Triglycerides (mg dl1) Fasting glucose (mg dl1) Urea (mg dl1) Creatinine (mg dl1) IIEF score

Control

Clinical ED

P

106 47 ± 9 58/48 97 ± 11 28 ± 4

110 56 ± 11 52/58 100 ± 12 28 ± 5

e <0.001* 0.274 0.098 0.784

62 26 18

37 56 17

<0.001*

14 92 130 ± 19 88 ± 13 36 ± 11 127 ± 38 206 ± 46 175 ± 106 100 ± 40 35 ± 9 0.99 ± 0.30 28 ± 2

12 98 138 ± 20 91 ± 12 41 ± 9 112 ± 35 181 ± 38 161 ± 111 127 ± 54 33 ± 12 1.05 ± 0.32 10 ± 7

0.604 <0.001* 0.083 <0.001* 0.003* <0.001* 0.342 <0.001* 0.221 0.149 <0.001*

SBP: systolic blood pressure; DBP: diastolic blood pressure; HDL: high density lipoprotein cholesterol; LDL: low density lipoprotein cholesterol; IIEF: international index of erectile function score. Results are expressed as means ± SD or as absolute numbers. Groups were compared using chi-squared tests, Student T test or ManneWhitney tests, as appropriate. * Statistically significant.

While the CC genotype for the rs2781659 polymorphism was associated with reduced IIEF scores in the clinical ED group (B ¼ 7.2, P ¼ 0.006), the variant TT genotype was associated with increased IIEF score (B ¼ þ6.9, P ¼ 0.008). Besides, the CC genotype for rs2781667 polymorphism was associated with increased IIEF scores (B ¼ þ7.2, P ¼ 0.009), and the variant TT genotype was associated with reduced IIEF scores (B ¼ 6.0, P ¼ 0.021). With respect to haplotypes, the common GTCC haplotype of ARG1 was associated with increased IIEF scores only in the clinical ED group (B ¼ þ2.52, P ¼ 0.006). No associations were observed on Control group. We further analyzed which factors could predict plasma arginase activity (Table 5). Interestingly, Arginase 1 and Arginase 2 levels correlated positively with Arginase Activity only in the clinical ED group (B ¼ þ0.11, P ¼ 0.043 for Arginase 1 levels, and B ¼ þ0.28, P ¼ 0.044 for Arginase 2 levels). Furthermore, we analyzed whether the studied genotypes and haplotypes could predict Arginase 1 and Arginase 2 levels (Supplementary Tables 4 and 5, respectively) and Arginase activity (Supplementary Tables 6 and 7). We found that the rs17599586 polymorphism affects arginase activity only in the clinical ED group (P ¼ 0.031), although specific genotypes effects were not significant (Supplementary Table 6). Other genotypes and haplotypes were not associated with changes in Arginase 1 and Arginase 2 levels, nor with Arginase Activity. 4. Discussion Arginase 1 and Arginase 2 are homologous enzymes that consume L-arginine to produce L-ornithine and urea [14,15]. These enzymes compete with NO synthases for the same substrate [14,15,30], and therefore may reduce NO production [21,24,30]. The colocalization of arginases with NO synthases [19] and the fact that arginase inhibitors increase NO production [36] strengthen this possibility. Consistent with abnormal upregulation of arginases in the corpus cavernosum in ED [16,24,25,29], arginase inhibitors were shown to improve erectile function in animal models of ED [24,26]. In humans, upregulation of Arginase 2 in the corpus cavernosum from patients with ED [16] may exert similar effects to those found in human endothelial cells submitted to hypoxic conditions, when reduced production of NO by eNOS is accompanied Arginase 2 upregulation [19]. However, there are no clinical studies that have evaluated whether the expression or activity of Arginases are increased in plasma from ED patients, neither whether genetic factors involving variations in the genes encoding Arginase 1 and 2 are associated with ED. Here, we were first to assess whether plasma Arginase 1 and 2 levels, and Arginase activity and genetic polymorphisms affect the risk for ED, and the degree of ED as reflected by the IIEF score. The main findings reported here are: 1) Plasma Arginase 2 levels are increased in patients with ED patients and may be a useful biomarker for ED; 2) Arginase 1 genotypes and haplotypes are associated with variations in the IIEF score in patients with ED, thus suggesting a role for Arginase 1 in the degree of ED. Arginase 1 and Arginase 2 are enzymes produced ubiquitously in human body [15]. While Arginase 1 is cytosolic and more abundantly expressed in the liver, Arginase 2 is mitochondrial and is more widely expressed in the kidney, prostate, gastrointestinal tract, and vasculature, although both forms were found in vascular endothelial and smooth muscle cells [15,37]. However, our results suggest that Arginase 2 may be more relevant as a clinical biomarker for ED than Arginase 1. Indeed, Arginase 2 was increased in plasma of patients with ED (Fig. 1), which is consistent with previous reports on corpus cavernosum from ED patients [16].

R. Lacchini et al. / Nitric Oxide 51 (2015) 36e42

39

Fig. 1. Plasma levels of Arginase 1, Arginase 2 and Arginase Activity in the Control and clinical erectile dysfunction (ED) groups. Data of Arginase1, Arginase 2 and Arginase activity (inserts A, B and C, respectively) are presented as scatter plot. Lines indicate medians (Arginase 1 and Arginase Activity) or mean (Arginase 2), defined according to normality tests. Data were compared using ManneWhitney or Student's T test, as appropriate. *Statistically significant.

Table 2 Multivariate logistic regression analysis showing the influence of plasma levels of Arginase 1, Arginase 2 and Arginase Activity on Erectile Dysfunction risk. Source

P

OR (95% CI)

Arginase 1 levels Arginase 2 levels Arginase activity

0.928 0.010* 0.676

0.94 (0.27e3.35) 7.92 (1.68e39.51) 0.66 (0.09e4.62)

OR: Odds Ratio, CI: Confidence Interval. Results are corrected for age, smoking status and ethanol consumption. R2:0.16. * Statistically significant.

Interestingly, we found lower plasma Arginase 1 than Arginase 2 concentrations, both in ED patients and in controls. Nevertheless, Arginase 1 may be relevant to disease conditions because it may be locally upregulated, as shown in animal models [24], and thus impair regional NO production without being detected at abnormal circulating concentrations. In agreement with this idea, we found that Arginase 1 and Arginase 2 correlated with plasma enzyme activity only under disease conditions (ED), which presumably leads to upregulation of both enzymes [16,24]. To date there are no

Table 3 Multivariate linear regression analysis showing the influence of genotypes of ARG1 and ARG2 on IIEF scores in the Control and Clinical erectile dysfunction (ED) groups. Source

Control 2

Age (years) Smoking status Never smoker Ex smoker Current smoker Ethanol consumption (>30 g/day) ARG1 rs2781659 AA AG GG rs2781667 CC CT TT rs2246012 TT TC CC rs17599586 CC CT TT ARG2 rs3742879 AA AG GG rs10483801 CC CA AA

Clinical ED

R ¼ 0.20

RMSE ¼ 1.87

R2 ¼ 0.37

RMSE ¼ 5.52

B

P

B

P

0.06

0.004*

0.28

<0.001*

þ0.15 þ0.15 0.30 0.17

0.603 0.624 0.401 0.565

þ0.07 þ2.21 2.28 0.99

0.941 0.008* 0.037* 0.278

P ¼ 0.512 B þ1.01 0.35 0.67 P ¼ 0.251 B 1.40 þ1.03 þ0.37 P ¼ 0.189 B þ1.64 1.11 0.54 P ¼ 0.607 B 0.34 0.65 þ0.99 P ¼ 0.719 B þ0.25 þ0.15 0.40 P ¼ 0.920 B 0.07 0.12 þ0.19

P 0.261 0.718 0.466 P 0.100 0.266 0.654 P 0.145 0.075 0.384 P 0.638 0.356 0.460

P 0.474 0.664 0.434 P 0.832 0.719 0.696

R2: portion of variability explained by the model. RMSE: root mean square error. * Statistically significant.

P ¼ 0.017* B 7.21 þ0.37 þ6.85 P ¼ 0.029* B þ7.20 1.23 5.96 P ¼ 0.299 B 2.28 0.58 þ2.86 P ¼ 0.228 B þ0.34 2.13 þ1.79 P ¼ 0.544 B 0.25 1.37 þ1.62 P ¼ 0.519 B 0.58 0.82 þ1.40

P 0.006* 0.832 0.008* P 0.009* 0.466 0.021* P 0.125 0.713 0.257 P 0.806 0.135 0.441

P 0.838 0.283 0.439 P 0.509 0.347 0.261

40

R. Lacchini et al. / Nitric Oxide 51 (2015) 36e42

Table 4 Multivariate linear regression analysis showing the influence of haplotypes of ARG1 and ARG2 on IIEF scores in the Control and clinical erectile dysfunction (ED) groups. Source

Control R2 ¼ 0.12

Age (years) Smoking status Never smoker Ex smoker Current smoker Ethanol consumption (>30 g/day) ARG1 ACTC GTTC GTTT GTCC ARG2 AC AA GC

Clinical ED RMSE ¼ 1.86

R2 ¼ 0.29

RMSE ¼ 5.51

B

P

B

P

0.07

<0.001*

0.31

<0.001*

þ0.02 þ0.16 0.17 0.05 P ¼ 0.860 B 0.11 þ0.14 þ0.14 0.17 P ¼ 0.379 B þ0.14 þ0.16 0.30

0.933 0.480 0.484 0.776

þ0.03 þ1.94 1.97 1.46 P ¼ 0.049* B 0.57 0.50 1.45 þ2.52 P ¼ 0.872 B þ0.00 þ0.31 0.32

0.967 0.001* 0.010* 0.011*

P 0.605 0.616 0.655 0.602 P 0.434 0.467 0.169

P 0.363 0.565 0.156 0.006* P 0.995 0.613 0.665

R2: portion of variability explained by the model. RMSE: root mean square error. * Statistically significant.

Table 5 Multivariate linear regression analysis showing the influence of Arginase 1 and Arginase 2 plasma levels on plasma Arginase Activity in the Control and clinical erectile dysfunction (ED) groups. Source

Control R2 ¼ 0.01

Age (years) Smoking status Never smoker Ex smoker Current smoker Ethanol consumption (>30 g/day) Arginase 1 levels (ng/mL) Arginase 2 levels (ng/mL)

Clinical ED RMSE ¼ 0.26

R2 ¼ 0.10

RMSE ¼ 0.33

B

P

B

P

0.00

0.742

þ0.00

0.848

0.01 þ0.02 0.01 þ0.01 0.03 þ0.02

0.774 0.636 0.834 0.835 0.568 0.812

0.06 þ0.04 þ0.02 þ0.05 þ0.11 þ0.28

0.230 0.353 0.798 0.367 0.043* 0.044*

R2: portion of variability explained by the model. RMSE: root mean square error. * Statistically significant.

reports of arginase 1 and 2 levels in plasma as biomarkers for erectile dysfunction. Although there is evidence showing the expression of these enzymes in several tissues, there is a lack of evidence showing how these enzymes are present in plasma. We believe that these enzymes may be released from endothelial cells (which express both isoforms [19]) or from red blood cells (which express arginase 1 [36]) to plasma when they are destroyed naturally. In diseases that involve endothelial dysfunction, arginases were shown to be upregulated [16,22,23]. If this is true, it is very likely that the increase in Arginase 2 levels that we show here actually reflects a more general endothelial dysfunction and may not be specifically related to endothelial dysfunction occurring only in the corpus cavernosum. Nevertheless, this is consistent with the idea that ED would be an early manifestation of a more generic endothelial dysfunction, which over the years would end up presenting itself as coronary disease, or other cardiovascular diseases [4]. The effects of genetic polymorphisms of ARG1 and ARG2 genes on ED risk and on IIEF scores were also examined. We selected four polymorphisms of ARG1 gene (Table 3) with basis on previous functional or clinical implications. The rs2781659 G allele reduces the expression of ARG1 promoter [38], and the rs2791665 polymorphism (which is at high linkage disequilibrium with the rs2781667, evaluated here) also reduces the expression of ARG1 in vitro [38]. Moreover, the variant alleles of the four selected

polymorphisms were associated conditions affected by NO signaling including asthma [38e41], blood pressure [42], and myocardial infarction [43]. The selection of these four polymorphisms allowed the assessment of all most common ARG1 haplotype blocks in Caucasians, according to Seattle SNPS website (http://gvs.gs.washington.edu/GVS138/; using 10% as minor allele frequency cutoff). Regarding ARG2, while there is no clear functional evidence implicating the studied polymorphisms, it has been shown that both the rs3742879 and the rs10483801 polymorphisms affect exhaled NO concentrations significantly [44], and affect the susceptibility to asthma [41,45], or the therapeutic responses of sickle cell disease patients to hydroxyurea, which depends on NO formation [46]. Although we found negative results with respect to the association of ARG1 and ARG2 polymorphisms with ED phenotype (Supplementary Tables 1 and 2), the rs2781659 AA and rs2781667 TT genotypes were found in association with worse (lower) IIEF scores among ED patients, whereas the rs2781659 GG and the rs2781667 CC genotypes were associated with better (higher) IIEF scores (Tables 3 and 4). Interestingly this was only observed in the Clinical ED group, which is consistent with our hypothesis that although Arginase 1 may not be upregulated in plasma, it may be locally upregulated under disease condition (as shown in animal models [24]) and may affect the severity of ED. We found a protective effect of the GTCC haplotype against ED,

R. Lacchini et al. / Nitric Oxide 51 (2015) 36e42

which is reflected by its association with increased IIEF in patients with clinical ED. Further studies are required to determine the functional implications of these genetic markers combined in specific haplotypes blocks. Indeed, we found no significant effects of the studied polymorphisms on Arginase1 and Arginase 2 plasma concentrations (Supplementary Tables 4 and 5). However, the ARG1 rs17599586 polymorphism seems to affect arginase activity (P ¼ 0.031, Supplementary Table 6). While this last result shows limited effects for this polymorphism, molecular mechanisms remain to be determined. As strengths of the present study, we should take into consideration the evaluation of healthy controls (without comorbidities; those with IIEF score <21 were excluded from our study) and patients, the vasculogenic-enriched clinical ED group, and the fact that we have covered the most common haplotypes blocks for ARG1. As limitations, we should mention the relatively low number of patients to assess disease susceptibility, which is robust for the biochemical analysis we performed. Nevertheless, power analysis calculation using PGA software [47] showed statistical power above 80% to detect a 2.28 odds ratio in the association with ED. Another limitation of this study is that we have evaluated polymorphisms that do not have a clear functional effect described yet. This makes it more difficult to establish a causative role for the studied polymorphisms, and may explain why there are some inconsistencies between associations we show here with what would be expected from previously reported clinical data. Further studies on this gene are needed to identify the individual functional effect of each of the studied polymorphisms individually and the effect of them when combined in haplotypes. 5. Conclusions Circulating Arginase 2 concentrations increase in clinical ED and are associated with increased risk for ED, and therefore may be a useful biomarker for ED. Moreover, ARG1 rs2781659 AA and rs2781667 TT genotypes were associated with lower IIEF scores (increased severity) in clinical ED, whereas ARG1 GTCC haplotype is associated with higher IIEF scores in clinical ED, thus suggesting a genetic contribution of ARG1 variations to ED. Conflict of interest All authors declare no conflicts of interest. Acknowledgments  Pesquisa do This study was supported by Fundaç~ ao de Amparo a Estado de Sao Paulo (FAPESP e Grant 2013/13346-2) and Conselho gico (CNPq). Nacional de Desenvolvimento Científico e Tecnolo Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.niox.2015.10.003. References [1] K.T. McVary, Clinical practice. Erectile dysfunction, N. Engl. J. Med. 357 (2007) 2472e2481. [2] R. Lacchini, J.E. Tanus-Santos, Pharmacogenetics of erectile dysfunction: navigating into uncharted waters, Pharmacogenomics 15 (2014) 1519e1538. [3] G. Jackson, Treatment of erectile dysfunction in patients with cardiovascular disease : guide to drug selection, Drugs 64 (2004) 1533e1545. [4] B.A. Inman, J.L. Sauver, D.J. Jacobson, M.E. McGree, A. Nehra, M.M. Lieber, V.L. Roger, S.J. Jacobsen, A population-based, longitudinal study of erectile dysfunction and future coronary artery disease, Mayo Clin. Proc. 84 (2009) 108e113.

41

[5] M. Bocchio, P. Scarpelli, S. Necozione, F. Pelliccione, R. Mhialca, C. Spartera, F. Francavilla, S. Francavilla, Intima-media thickening of common carotid arteries is a risk factor for severe erectile dysfunction in men with vascular risk factors but no clinical evidence of atherosclerosis, J. Urol. 173 (2005) 526e529. [6] R. Kloner, Erectile dysfunction and hypertension, Int. J. Impot. Res. 19 (2007) 296e302. [7] H. Solomon, J.W. Man, G. Jackson, Erectile dysfunction and the cardiovascular patient: endothelial dysfunction is the common denominator, Heart 89 (2003) 251e253. [8] J.J. Muniz, R. Lacchini, J.T. Sertorio, A.A. Jordao Jr., Y.T. Nobre, S. Tucci Jr., A.C. Martins, J.E. Tanus-Santos, Low nitric oxide bioavailability is associated with better responses to sildenafil in patients with erectile dysfunction, Naunyn Schmiedeb. Arch. Pharmacol. 386 (2013) 805e811. [9] K.J. Hurt, B. Musicki, M.A. Palese, J.K. Crone, R.E. Becker, J.L. Moriarity, S.H. Snyder, A.L. Burnett, Akt-dependent phosphorylation of endothelial nitric-oxide synthase mediates penile erection, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 4061e4066. [10] K.J. Hurt, S.F. Sezen, G.F. Lagoda, B. Musicki, G.A. Rameau, S.H. Snyder, A.L. Burnett, Cyclic AMP-dependent phosphorylation of neuronal nitric oxide synthase mediates penile erection, Proc. Natl. Acad. Sci. U. S. A. 109 (2012) 16624e16629. [11] B. Musicki, A.L. Burnett, eNOS function and dysfunction in the penis, Exp. Biol. Med. (Maywood) 231 (2006) 154e165. [12] R. Lacchini, J.J. Muniz, Y.T. Nobre, A.J. Cologna, A.C. Martins, J.E. Tanus-Santos, nNOS polymorphisms are associated with responsiveness to sildenafil in clinical and postoperative erectile dysfunction, Pharmacogenomics 15 (2014) 775e784. [13] J.J. Muniz, R. Lacchini, T.O. Rinaldi, Y.T. Nobre, A.J. Cologna, A.C. Martins, J.E. Tanus-Santos, Endothelial nitric oxide synthase genotypes and haplotypes modify the responses to sildenafil in patients with erectile dysfunction, Pharmacogenomics J. 13 (2013) 189e196. [14] W. Durante, F.K. Johnson, R.A. Johnson, Arginase: a critical regulator of nitric oxide synthesis and vascular function, Clin. Exp. Pharmacol. Physiol. 34 (2007) 906e911. [15] J. Pernow, C. Jung, Arginase as a potential target in the treatment of cardiovascular disease: reversal of arginine steal? Cardiovasc Res. 98 (2013) 334e343. [16] T.J. Bivalacqua, W.J. Hellstrom, P.J. Kadowitz, H.C. Champion, Increased expression of arginase II in human diabetic corpus cavernosum: in diabeticassociated erectile dysfunction, Biochem. Biophys. Res. Commun. 283 (2001) 923e927. [17] J.D. Cox, N.N. Kim, A.M. Traish, D.W. Christianson, Arginase-boronic acid complex highlights a physiological role in erectile function, Nat. Struct. Biol. 6 (1999) 1043e1047. [18] E. Cama, D.M. Colleluori, F.A. Emig, H. Shin, S.W. Kim, N.N. Kim, A.M. Traish, D.E. Ash, D.W. Christianson, Human arginase II: crystal structure and physiological role in male and female sexual arousal, Biochemistry 42 (2003) 8445e8451. [19] C.P. Prieto, B.J. Krause, C. Quezada, R. San Martin, L. Sobrevia, P. Casanello, Hypoxia-reduced nitric oxide synthase activity is partially explained by higher arginase-2 activity and cellular redistribution in human umbilical vein endothelium, Placenta 32 (2011) 932e940. [20] F. Quitter, H.R. Figulla, M. Ferrari, J. Pernow, C. Jung, Increased arginase levels in heart failure represent a therapeutic target to rescue microvascular perfusion, Clin. Hemorheol. Microcirc. 54 (2013) 75e85. [21] D.E. Berkowitz, R. White, D. Li, K.M. Minhas, A. Cernetich, S. Kim, S. Burke, A.A. Shoukas, D. Nyhan, H.C. Champion, J.M. Hare, Arginase reciprocally regulates nitric oxide synthase activity and contributes to endothelial dysfunction in aging blood vessels, Circulation 108 (2003) 2000e2006. [22] S.R. Kashyap, A. Lara, R. Zhang, Y.M. Park, R.A. DeFronzo, Insulin reduces plasma arginase activity in type 2 diabetic patients, Diabetes Care 31 (2008) 134e139. [23] F.K. Johnson, K.J. Peyton, X.M. Liu, M.A. Azam, A.R. Shebib, R.A. Johnson, W. Durante, Arginase promotes endothelial dysfunction and hypertension in obese rats, Obes. (Silver Spring) 23 (2) (2014) 383e390. [24] T.J. Bivalacqua, A.L. Burnett, W.J. Hellstrom, H.C. Champion, Overexpression of arginase in the aged mouse penis impairs erectile function and decreases eNOS activity: influence of in vivo gene therapy of anti-arginase, Am. J. Physiol. Heart Circ. Physiol. 292 (2007) H1340eH1351. [25] H.A. Toque, M.J. Romero, R.C. Tostes, A. Shatanawi, S. Chandra, Z.N. Carneiro, E.W. Inscho, R.C. Webb, R.B. Caldwell, R.W. Caldwell, p38 Mitogen-activated protein kinase (MAPK) increases arginase activity and contributes to endothelial dysfunction in corpora cavernosa from angiotensin-II-treated mice, J. Sex. Med. 7 (2010) 3857e3867. [26] R. Segal, J.L. Hannan, X. Liu, O. Kutlu, A.L. Burnett, H.C. Champion, J.H. Kim, J. Steppan, D.E. Berkowitz, T.J. Bivalacqua, Chronic oral administration of the arginase inhibitor 2(S)-amino-6-boronohexanoic acid (ABH) improves erectile function in aged rats, J. Androl. 33 (2012) 1169e1175. [27] K.P. Nunes, H.A. Toque, R.B. Caldwell, R.W. Caldwell, R.C. Webb, Extracellular signal-regulated kinase (ERK) inhibition decreases arginase activity and improves corpora cavernosal relaxation in streptozotocin (STZ)-induced diabetic mice, J. Sex. Med. 8 (2011) 3335e3344. [28] H. Masuda, Significance of nitric oxide and its modulation mechanisms by endogenous nitric oxide synthase inhibitors and arginase in the micturition disorders and erectile dysfunction, Int. J. Urol. 15 (2008) 128e134.

42

R. Lacchini et al. / Nitric Oxide 51 (2015) 36e42

[29] N. Numao, H. Masuda, Y. Sakai, Y. Okada, K. Kihara, H. Azuma, Roles of attenuated neuronal nitric-oxide synthase protein expression and accelerated arginase activity in impairing neurogenic relaxation of corpus cavernosum in aged rabbits, BJU Int. 99 (2007) 1495e1499. [30] F. Mariotti, K.J. Petzke, D. Bonnet, I. Szezepanski, C. Bos, J.F. Huneau, H. Fouillet, Kinetics of the utilization of dietary arginine for nitric oxide and urea synthesis: insight into the arginine-nitric oxide metabolic system in humans, Am. J. Clin. Nutr. 97 (2013) 972e979. [31] R. Lacchini, J.J. Muniz, Y.T. Nobre, A.J. Cologna, A.C. Martins, J.E. Tanus-Santos, VEGF genetic polymorphisms affect the responsiveness to sildenafil in clinical and postoperative erectile dysfunction, Pharmacogenomics J. 13 (5) (2012) 437e442. [32] R. Lacchini, P.S. Silva, J.E. Tanus-Santos, A pharmacogenetics-based approach to reduce cardiovascular mortality with the prophylactic use of statins, Basic Clin. Pharmacol. Toxicol. 106 (2010) 357e361. [33] R. Lacchini, J.J. Muniz, Y.T. Nobre, A.J. Cologna, A.C. Martins, J.E. Tanus-Santos, VEGF genetic polymorphisms affect the responsiveness to sildenafil in clinical and postoperative erectile dysfunction, Pharmacogenomics J. 13 (2013) 437e442. [34] J.J. Muniz, R. Lacchini, V.A. Belo, Y.T. Nobre, S. Tucci Jr., A.C. Martins, J.E. TanusSantos, Circulating matrix metalloproteinases and their endogenous inhibitors in patients with erectile dysfunction, Int. J. Impot. Res. 24 (2012) 38e43. [35] R.C. Rosen, A. Riley, G. Wagner, I.H. Osterloh, J. Kirkpatrick, A. Mishra, The international index of erectile function (IIEF): a multidimensional scale for assessment of erectile dysfunction, Urology 49 (1997) 822e830. [36] J. Yang, A.T. Gonon, P.O. Sjoquist, J.O. Lundberg, J. Pernow, Arginase regulates red blood cell nitric oxide synthase and export of cardioprotective nitric oxide bioactivity, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 15049e15054. [37] S. Choi, C. Park, M. Ahn, J.H. Lee, T. Shin, Immunohistochemical study of arginase 1 and 2 in various tissues of rats, Acta histochem. 114 (2012) 487e494. [38] Q.L. Duan, B.R. Gaume, G.A. Hawkins, B.E. Himes, E.R. Bleecker, B. Klanderman, C.G. Irvin, S.P. Peters, D.A. Meyers, J.P. Hanrahan, J.J. Lima, A.A. Litonjua, K.G. Tantisira, S.B. Liggett, Regulatory haplotypes in ARG1 are associated with

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

altered bronchodilator response, Am. J. Respir. Crit. Care Med. 183 (2011) 449e454. H. Li, I. Romieu, J.J. Sienra-Monge, M. Ramirez-Aguilar, B. Estela Del RioNavarro, E.O. Kistner, H.K. Gjessing, C. Lara-Sanchez Idel, G.Y. Chiu, S.J. London, Genetic polymorphisms in arginase I and II and childhood asthma and atopy, J. Allergy Clin. Immunol. 117 (2006) 119e126. A.A. Litonjua, J. Lasky-Su, K. Schneiter, K.G. Tantisira, R. Lazarus, B. Klanderman, J.J. Lima, C.G. Irvin, S.P. Peters, J.P. Hanrahan, S.B. Liggett, G.A. Hawkins, D.A. Meyers, E.R. Bleecker, C. Lange, S.T. Weiss, ARG1 is a novel bronchodilator response gene: screening and replication in four asthma cohorts, Am. J. Respir. Crit. Care Med. 178 (2008) 688e694. J.M. Vonk, D.S. Postma, H. Maarsingh, M. Bruinenberg, G.H. Koppelman, H. Meurs, Arginase 1 and arginase 2 variations associate with asthma, asthma severity and beta2 agonist and steroid response, Pharmacogenet Genomics 20 (2010) 179e186. D. Meroufel, J. Dumont, S. Mediene-Benchekor, S. Benhammamouch, P. Ducimetiere, D. Cottel, M. Montaye, P. Amouyel, T. Brousseau, Characterization of arginase 1 gene polymorphisms in the Algerian population and association with blood pressure, Clin. Biochem. 42 (2009) 1178e1182. J. Dumont, M. Zureik, D. Cottel, M. Montaye, P. Ducimetiere, P. Amouyel, T. Brousseau, Association of arginase 1 gene polymorphisms with the risk of myocardial infarction and common carotid intima media thickness, J. Med. Genet. 44 (2007) 526e531. M.T. Salam, T.M. Bastain, E.B. Rappaport, T. Islam, K. Berhane, W.J. Gauderman, F.D. Gilliland, Genetic variations in nitric oxide synthase and arginase influence exhaled nitric oxide levels in children, Allergy 66 (2011) 412e419. M.T. Salam, T. Islam, W.J. Gauderman, F.D. Gilliland, Roles of arginase variants, atopy, and ozone in childhood asthma, J. Allergy Clin. Immunol. 123 (2009) 596e602, 602 e1e8. Q. Ma, D.F. Wyszynski, J.J. Farrell, A. Kutlar, L.A. Farrer, C.T. Baldwin, M.H. Steinberg, Fetal hemoglobin in sickle cell anemia: genetic determinants of response to hydroxyurea, Pharmacogenomics J. 7 (2007) 386e394. I. Menashe, P.S. Rosenberg, B.E. Chen, PGA: power calculator for case-control genetic association analyses, BMC Genet. 9 (2008) 36.