Folate: a possible role in erectile dysfunction?

Abstract

A growing body of evidence suggests a role for homocysteine (Hcys) and folate (FA) in erectile function (EF): Hcys appears to impair EF affecting endothelium via several mechanism whereas the role of FA remains to be elucidated, besides decreasing Hcys. To assess correlation between erectile dysfunction (ED) and serum levels of FA, Hcys, and B12, we enrolled 31 patients affected by ED (Group A; age 52.83 ± 11.89 years) and 31 healthy adults (Group B; age 49.14 ± 13.63 years). Fasting blood samples were taken for each subject. ED was assessed by the International Index of Erectile Function-5 (IIEF-5). IIEF-5 mean score was significantly lower in Group A than in Group B (10.71 ± 4.24 versus 23.32 ± 1.33, p < .001). Compared to Group B, Group A also showed significantly lower serum FA levels (5.11 ± 1.79 versus 7.9 ± 3.55 ng/ml, p < .001) and significantly higher serum Hcys levels (13.61 ± 3.55 versus 9.17 ± 2.32 µmol/L, p < .001). No significant correlation was observed between Hcys and FA both groups. Our results showed a significant association among ED, FA deficiency and hyperomocisteinemia. Lack of correlation between FA and Hcys suggests that FA deficit may directly impair EF.

Keywords:

• Folic acid
• erectile dysfunction
• homocysteine
• nitric oxide synthase
• tetrahydrobiopterin

Introduction

Erectile dysfunction (ED) is defined as the consistent inability to obtain and/or maintain an erection for satisfactory sexual intercourse. According to the DSM-V, erectile disorder is characterized by the marked difficulty in obtaining erection during sexual activity, marked difficulty in maintaining an erection until completion of sex and marked decrease in erectile rigidity. One out of these three aforementioned symptoms must have been present for at least 6 months and to be experienced in almost all or all (∼75–100%) of sexual activities. Symptoms must be deemed to have caused significant distress in the individual. Sexual dysfunction is not better explained by nonsexual mental disorder, a consequence of severe relationship distress, or other stressors, and it is not attributable to substance or medical condition [1,2]. Its incidence and prevalence are growing more and more and they’re strongly linked to aging. ED has a multifactorial etiology and according to the causes it can be defined as arteriogenic, neurogenic, cavernosal, endocrinological or iatrogenic. In most cases, ED is a consequence of impaired penile blood flow: vascular function is commonly assessed by means of duplex penile ultrasonography, by measuring peak systolic and diastolic velocity in the cavernosal arteries within the corpora cavernosa before and after the intracavernosal injection of a test dose of a standard vasodilator (e.g. 20 µg of PGE1). Many different pathological conditions may affect endothelium: diabetes, hypertension, obesity, arteriosclerosis (i.e. including atherosclerosis), hypercholesterolemia, hypertriglyceridemia, and chronic inflammatory status. In turn, these conditions may be worsened by endothelial dysfunction. Common causes of these morbidities listed above are represented by unhealthy lifestyles—such as smoking, lack of physical exercise, bad nutritional habits—and genetics. These factors may also contribute to the development and maintenance of hyperomocisteinemia, which is able to impair significantly endothelial function and consequently erectile function by several mechanism as reported by Liu et al. [3,4]. In this setting, folate (FA) may play a crucial role in improving both endothelial and erectile function decreasing hyperomocisteinemia [5]; however, in the last years a growing body of evidence suggests that FA might improve endothelial function by means of other mechanism [6,7]. Indeed, FA seems to act directly on nitric oxide (NO) metabolism increasing its bioavailability, in particular improving efficiency of NO synthase. Besides its role as the main biological mediator of the erection progress, NO is of paramount importance in maintaining a proper endothelial function [8]. Therefore, in this study we aim to observe serum FA levels and their correlations with homocysteinemia (Hcys), serum testosterone, and erectile function.

Methods

Participants

We enrolled 31 patients affected by ED (Group A; age 52.83 ± 11.89 years) and 31 healthy adults as control group (Group B; age 49.14 ± 13.63 years); patients were recruited from our andrologic clinic of Experimental Medicine Department and control men were volunteered healthy hospital members. We included subjects according with the following criteria: (i) two or more sexual intercourse weekly; (ii) absence of either chronic or on demand use of phosphodiesterase-5 inhibitors (PDE5i) in the last 6 months; (iii) absence of psychoactive drugs and/or alcohol abuse; (iv) absence of on-treatment with drugs affecting endocrine status; (v) absence of end-stage renal disease, hepatic and heart failure; (vi) absence of neurological disease, pelvic trauma, psychiatric disorder, diabetes, coronary artery disease, vitamin deficiency, and strenuous physical activity.

Main outcome measures

All subjects underwent full physical examination, including measurement of blood pressure.

Both healthy men and patients complaining of ED completed the International Index of Erectile Dysfunction (IIEF-5) questionnaire in order to establish these men as either a control or as a patient. The IIEF-5 is a validated questionnaire routinely used in clinical assessment of ED [9]. Every participant understood and completed the questionnaire. All men were sexually active and they complete questionnaire based on their sexual activities in the last six months. IIEF-5 score ≤21 identified men to be included in ED group.

We assessed penile vascular function by means of duplex ultrasonography following the intracavernosal injection of 20 µg of PGE1. Doppler examinations were performed by a single operator in order to reduce variability; according to standard operating procedures [10], normal response was defined by a peak-systolic velocity (PSV) > 30 cm/s end-diastolic velocity (EDV)  <  3 cm/s and resistive index [RI = (PSV − EDV)/PSV] > 0.8.

Laboratory tests

Fasting blood samples were collected from every subject, and they were prepared as following: (i) Whole blood was drawn into vacutainer tube(s) containing no anticoagulant. We drew ∼10 ml blood. (ii) Blood was incubated in an upright position at room temperature for 45 min to allow clotting. We used a clot-activator tube, so we inverted carefully five times to mix clot activator and blood before incubation. (iii) We centrifuged for 15 min at manufacturer’s recommended speed (1200 RCF). (iv) We aspirated the supernatant (serum) at room temperature using a Pasteur pipette and pool into a polypropylene tube. (v) We inspected serum for turbidity. Turbid samples have been centrifuged and aspirated again to remove remaining insoluble matter. (vi) Serum was apportioned into 0.5 ml aliquots and were stored at −25 °C until assayed. The serum levels of total testosterone, vitamin B₁₂, folate were measured by electrochemiluminiscence (ECL) using the Architect i1000 analyzer (Abbott Laboratories, Abbott Park, Chicago, IL) according to the protocol provided by the manufacturer. Regarding total testosterone, based on kit manufacturer (Second-Generation Testosterone Kit, Abbott Laboratories) measuring interval ranges from 0.15 nmol/L (4 ng/dl) to 65.47 nmol/L (1154 ng/dl) accuracy is within ±5.4%, within-laboratory imprecision is ≤10% CV for samples with testosterone concentrations ≥0.5 (14 ng/dl) to 35 nmol/L (10.10 ng/dl), limit of detection (LoD) is 0.5 nmol/L (1.44 ng/dl), limit of blank (LoB) is 0.03 nmol/L (0.87 ng/dl) and limit of quantification (LoQ) is 0.08 nmol/L (2.30 ng/dl).

Regarding vitamin B₁₂, based on kit manufacturer (ARCHITECT B12 assay, Abbott Laboratories) measuring interval ranges from 125 to 1218 pg/mL (92–899 pmol/L), accuracy is within ±3.6%, within-laboratory imprecision is ≤11% CV for low, medium and high concentrations, LoD is 125 pg/mL (92 pmol/L), LoB is 83 pg/mL (61 pmol/L), and LoQ is 125 pg/mL (92 pmol/L).

Regarding folate, based on kit manufacturer (ARCHITECT Folate Assay, Abbott Laboratories) measuring interval ranges from 1.5 (3.4 nmol/L) to 20.0 ng/mL (45.32 nmol/L), accuracy is within ±10%, within-laboratory imprecision is ≤12% total CV for serum samples from 3.5 (7.93 nmol/L) to 20 ng/mL (45.32 nmol/L), LoD is 0.5 ng/mL (1.1 nmol/L), LoB is 0.3 ng/mL (0.7 nmol/L), and LoQ is 1.5 ng/mL (3.4 nmol/L).

Homocysteine was measured using high-performance liquid chromatography (HPLC) using Hewlett-Packard (model 1100) HPLC system by Agilent Technologies (Santa Clara, CA). Measuring interval ranges from 1.0 to 85.0 μmol/L accuracy is within ±9%, intra-assay imprecision is ≤5.5%, inter-assay imprecision is ≤8.0%, LoD is 1.0 μmol/L, LoB 0.7 μmol/L, and LoQ is 3.4 μmol/L.

Statistical analysis

Statistical analyses were performed using the Statistical Package for Social Sciences (SPSS) version 21.0 (SPSS Inc., Chicago, IL) for Windows 10. Data are shown as mean ± standard deviation for normally distributed data and as medians (IQR) for non-normally distributed data. Shapiro–Wilk test of normality was used in order to assess normality of distribution. Correlation coefficients were assessed by Pearson method. All statistical analyzes were two-sided, and p < .05 was considered as significant.

Results

Clinical data of both groups are listed in Table 1. No significant differences between the groups have been observed in regards to age, total testosterone and vitamin B₁₂ whereas Group A showed a significant higher serum Hcys level and lower serum FA level compared to Group B (Hcys 13.61 ± 3.55 versus 9.17 ± 2.32, p < .001; FA 5.11 ± 1.79 versus 7.9 ± 3.55, p < .001, respectively). Of course, IIEF-5 score was significant lower in Group A than in Group B (10.71 ± 4.24 versus 23.32 ± 1.33, p < .001). IIEF-5 median score was 11 in Group A and 23 in Group B whereas the interquantile range was 5.5 in Group A and 2 in Group B. Surprisingly, in both groups serum FA levels were not significantly correlated to serum Hcys levels; only slight negative correlation coefficients were observed, far from being significant (Group A: r = −0.247, p = .197; Group B: r = −0.163, p = .372) (Tables 2 and 3). Interestingly, serum FA level resulted strongly correlated with IIEF-5 score in group A (r = 0.538, p = .002) (Table 2).

Table 1. Clinical data and hematochemical values of the subjects.

	ED (Group A)	Controls (Group B)	p
Age (years)	52.83 ± 11.89	49.14 ± 13.63	.297
IIEF-5 score	10.71 ± 4.24	23.32 ± 1.33	<.001
Testosterone (ng/ml)	4.46 ± 1.46	4.1 ± 1.84	.45
Vitamin B12 (pg/ml)	393.49 ± 145.07	375.18 ± 120.0.22	.59
Homocysteine (µmol/ml)	13.61 ± 3,55	9.17 ± 2.32	<.001
Folate (ng/ml)	5.11 ± 1.79	7.9 ± 3.55	<.001

Data are expressed as mean ± standard deviation.

Table 2. Correlation coefficients of FA concentrations with clinical parameters in ED group (Group A).

	Folate (r)	p
Age	0.225	.290
IIEF-5	0.538	.002
Testosterone	−0.198	.342
Vitamin B12	−0.192	.310
Homocysteine	−0.247	.197

Table 3. Correlation coefficients of FA concentrations with clinical parameters in control group (Group B).

	Folate (r)	p
Age	0.237	.208
IIEF-5	0.131	.475
Testosterone	0.229	.270
Vitamin B12	−0.144	.432
Homocysteine	−0.163	.372

Discussion

In the last years, a growing number of evidence has shown the role of FA in endothelial function and consequently in etiology and in treatment of ED [3,11–13]. Our results seems to support this role; indeed, serum FA levels are significantly lower in ED group compared to control group (Table 1) and serum FA level appears to be significantly associated with the severity of ED (Table 2) as observed also by Karabakan et al. [12]. Furthermore, our data suggest that the role of FA goes beyond treatment of hyperomocisteinemia, directly affecting endothelium; indeed, we observed significant lower serum FA and higher serum Hcys in ED group compared to control group, although no correlation was observed between FA and Hcys (Table 2). We suggest that FA may affect ED increasing NO availability in several ways.

NO acts as a master regulator of endothelium: indeed, among its multiple properties, NO activates soluble guanylyl cyclase to produce cGMP. Inhibition of cGMP catabolism is the mechanism of action of PDE5i, vasodilator agents used as a first-line treatment for ED [14–16]. NO production requires structural integrity of endothelial NO synthase (eNOS); tetrahydrobiopterine (BH₄) represents the cofactor necessary to keep oxidase and reductase subunits linked together promoting morphological integrity of eNOS. BH₄ also increases the affinity of cofactors, such as NADPH oxidase, for eNOS, and plays a significant role in the reaction of molecular oxygen with l-arginine. In the absence of BH₄, eNOS becomes uncoupled with its cofactor, therefore losing its structural integrity, and electron flow to molecular oxygen becomes uncoupled from l-arginine oxidation, resulting in production of superoxide (O₂⁻) instead of NO [17]. Moreover, O₂⁻ rapidly reacts with NO producing peroxynitrite (ONOO⁻) further decreasing NO bioavailability; consequently, even BH₄ is rapidly oxidized to dihydrobiopterine (BH₂) by ONOO⁻. BH₂ displaces BH₄ from its binding sites not showing the same ability of BH₄ to couple electron flow to l-arginine oxidation and acting as a competitive inhibitor of BH₄ to NOS [18,19]. These events lead to a propagation of the process [20]. Therefore, BH₄ seems to play a crucial role in endothelial function, and FA and its metabolites affect several biological pathways of BH₄. 5-Methyltetrahydrofolate (5-MTHF) seems to stimulate and/or restore endogenous BH₄ regeneration from quinonoid BH₂ while at the same time also upregulating activity of dihydrofolate reductase (DHFR) in reducing BH₂, thus regenerating BH₄, as part of the salvage pathway for biopterin synthesis [8,18,21,22]. Such an upregulation of BH₄ recycling pathway may increase NO production both increasing BH₄ bioavailability and decreasing BH₂ which acts as competitive inhibitor of BH₄ [8]. 5-MTHF may also increase the effectiveness of BH₄ in eNOS coupling improving redox state and/or enhancing binding affinity of BH₄ to eNOS. Indeed, 5-MTHF may act as a direct scavenger of reactive oxygen species (ROS): in particular, 5-MTHF seems to prevent BH₄ intracellular oxidation by ONOO⁻ to BH₂ [19,23]_. Last but not least, chemical structure of 5-MTHF seems to be very similar to that of BH₄ and it may be able to bind directly the pterin site in eNOS stimulating NO production [19,24].

Finally, we observed a significant lower serum FA levels in ED patients compared to healthy men and significant correlation between serum FA levels and grade of severity of ED confirming what has been observed in previous studies [3,12]; however, our most intriguing contribution is represented by the absence of correlation between FA and Hcys. This result suggests that FA may directly affect erectile function with different mechanism from decreasing Hcys. In particular, understanding relations across FA and BH₄ biological pathways seems to be promising. Notwithstanding, correlation does not imply causation and further studies are warranted to verify in larger clinical trials the relationship between FA status and endothelial/erectile function and to assess if a treatment with FA may be effective in improving endothelial/erectile function. Comforting results regarding FA administration in patients affected by ED, in combination with either tadalafil or myoinositol, have already been published [11,13]. Further steps may be represented by the evaluation of variables that may affect simultaneously both serum FA levels and erectile function and the assessment of the treatment with FA that may guarantee best clinical results.

Disclosure statement

No potential conflict of interest was reported by the authors.