Exploring rat corpus cavernosum alterations induced by finasteride treatment and withdrawal

Despite its efficacy for treating androgenetic alopecia, finasteride, an inhibitor of 5α‐reductase (i.e., the enzyme converting testosterone, T, into dihydrotestosterone, DHT), is associated with several side effects including sexual dysfunction (e.g., erectile dysfunction). These side effects may persist after drug suspension, inducing the so‐called post‐finasteride syndrome (PFS). The effects of subchronic treatment with finasteride (i.e., 20 days) and its withdrawal (i.e., 1 month) in rat corpus cavernosum have been explored here. Data obtained show that the treatment was able to decrease the levels of the enzyme 5α‐reductase type II in the rat corpus cavernosum with increased T and decreased DHT levels. This local change in T metabolism was linked to mechanisms associated with erectile dysfunction. Indeed, by targeted metabolomics, we reported a decrease in the nitric oxide synthase (NOS) activity, measured by the citrulline/arginine ratio and confirmed by the decrease in NO2 levels, and a decrease in ornithine transcarbamylase (OTC) activity, measured by citrulline/ornithine ratio. Interestingly, the T levels are negatively correlated with NOS activity, while those of DHT are positively correlated with OTC activity. Finasteride treatment also induced alterations in the levels of other molecules involved in the control of penile erection, such as norepinephrine and its metabolite, epinephrine. Indeed, plasma levels of norepinephrine and epinephrine were significantly increased and decreased, respectively, suggesting an impairment of these mediators. Interestingly, these modifications were restored by suspension of the drug. Altogether, the results reported here indicate that finasteride treatment, but not its withdrawal, affects T metabolism in the rat corpus cavernosum, and this alteration was linked to mechanisms associated with erectile dysfunction. Data here reported could also suggest that the PFS sexual side effects are more related to dysfunction in a sexual central control rather than peripheral compromised condition.


INTRODUCTION
Penile erection is the result of a complex neurovascular process regulated by central and local mediators such as neurotransmitters, neuropeptides, peptides, and sexual hormones that ultimately influence the corpus cavernosum. 1,2Erectile dysfunction (ED) and overall male sexual impairment have a multifactorial etiology and may be related with psychogenic (i.e., stress, depression, anxiety), not endocrine (i.e., neurogenic, vasculogenic, and iatrogenic), or endocrine causes. 1,2Several drugs, 3,4 including 5α-reductase inhibitors (5α-RIs), 5,6 may impair male sexual function.A flagship example of 5α-RIs is the finasteride, which was approved in 1997 for treating androgenetic alopecia (AGA) to block the conversion of testosterone (T) into dihydrotestosterone (DHT).4][15][16][17] Indeed, young men treated for AGA with finasteride showed ED, decreased libido, orgasm failure, and ejaculatory disorders. 18,19[22][23][24][25][26][27][28][29][30][31][32][33][34] The effects of finasteride treatment were also reported in animal models.For instance, finasteride treatment in castrated rats decreased erectile response, although receiving T replacement 35,36 and in control male rats, it decreased intracavernosal pressure during electrical stimulation of cavernous nerves. 37Treatment with finasteride or other 5α-RIs, such as dutasteride, significantly suppressed plasma DHT levels. 38,39Additionally, drug treatment in adult male rats resulted in a loss of the weight of the corpus cavernosum, 38 reduced cross-sectional penile area, 40 and increased collagen density in the cavernosal tissue, 37 while in aged male rats resulted in morphological alterations of mitochondria and increased apoptosis in cavernous smooth muscle cells. 41Altogether, these experimental observations suggested the corpus cavernosum as a target for finasteride-induced ED.However, the molecular mechanisms have not been explored so far in detail, and in particular, if these are also involved in the persistent ED induced by this drug.
On this basis, we have explored the effect of subchronic treatment with finasteride and its withdrawal in rat corpus cavernosum, assessing: (i) the levels of steroids, such as pregnenolone (PREG), proges- Interestingly, finasteride also shows off-targets, such as phenylethanolamine N-methyltransferase (PNMT), the limiting enzyme in the formation of epinephrine from norepinephrine.Indeed, as recently demonstrated, finasteride treatment was able to inhibit this enzyme in the rat adrenal gland. 42This is relevant since the balance of epinephrine and norepinephrine levels is crucial for achieving erection. 1,43Therefore, the plasma levels of norepinephrine and epinephrine were assessed by LC-MS/MS after finasteride treatment and withdrawal.After sacrifice, the plasma and corpora cavernosa were immediately stored at −80 • C until the analysis.

Treatments
We have used the dose of 1 mg/rat/day, on the basis of our previous published observations. 44,45In particular, this low dose of finasteride was effective in decreasing the plasma DHT levels in a persistent way. 44nasteride (1 mg/rat/day; Sigma-Aldrich, Italy) was dissolved in a vehicle solution of sesame oil and ethanol (5% v/v).Solutions were injected subcutaneously, at a volume of 100 μL/day for 20 days.Finasterideand vehicle-treated rats were sacrificed at 24 h (n = 12) after the last injection and 1 month (n = 12) after drug suspension to investigate the effects of the subchronic treatment and its withdrawal, respectively.

2.3.1
Reagents and chemicals

Quantitative analysis of steroid molecules in corpus cavernosum
The levels of steroids were evaluated in corpora cavernosa (25-35 mg/sample). 13C 3 -17β-E (2 ng/sample) 13 C 5 -PROG (0.4 ng/sample), and 13 C 2 D 2 -PREG (10 ng/sample) were used as internal standards and added to all samples.For quantitative analysis, corpora cavernosa tissues were homogenized using the TissueLyser, in methanol/acetic acid 1%.Corpus cavernosum was purified by organic phase extraction and the analysis was performed by liquid chromatography (LC) supplied of Surveyor LC Pump Plus and Surveyor Autosampler Plus (Thermo Fisher Scientific MA, USA) with a linear ion trap-mass spectrometer LTQ (Thermo Fisher Scientific MA, USA), operated in positive atmospheric pressure chemical ionization as previously described. 32The chromatographic separation was achieved with a Hypersil Gold column C18 (100 The mobile phases consisted of 0.1% formic acid in water and 0.1% formic acid in methanol.25 μL sample was injected at a flow rate of 0.250 mL/min.The injector needle was washed with methanol-water (1:1 v/v).Quantitative analysis was performed on the basis of freshly prepared and extracted calibration curves.LC-MS/MS data were evaluated using the software Excalibur release 2.0 SR2 (Thermo Fisher Scientific MA, USA).

Quantitative analysis of norepinephrine and epinephrine in plasma
The quantitative analysis of norepinephrine and epinephrine in plasma samples was achieved as previously described. 46Briefly, 700 μL of plasma samples were mixed with 10 μL of internal standard solution (500 ng/sample of D 6 -NE) and 300 μL of buffer solution (1 mol/L Tris and 0.05 mol/L EDTA in water).Samples were purified with Sep-Pak 1cc (100 mg) Alumina B cartridges (Waters, Milano, Italy), eluted with 550 μL of elution solution (2.5% formic acid and 4% water in acetonitrile) and evaporated to 50 μL under a stream of N 2 .250 μL of 35 mmol/L HSA in aqueous 0.5% formic acid were added to samples and the solution was filtered with a 0.2 μm filter (SRC grade, regenerated cellulose membrane filter, CHMLAB Group).For the quantitative analysis 5 μL/sample were injected in API 3500 (AB Sciex, USA) mass spectrometer, equipped with an electrospray source (ESI+) and triple quadrupole analyzer, interfaced with a pump for the HPLC model EXION SL (Sciex, USA).The chromatographic separation was achieved using Luna Omega 5 μm PS C18 100 Å column (Phenomenex, USA).The mobile phases consisted of 0.1% formic acid in water and 0.1% formic acid in acetonitrile at flow rate of 0.300 μL/min.The quantitative analysis was performed on the basis of calibration curves freshly prepared in water-acetonitrile (98:2 v/v).LC-MS/MS data were evaluated using the software Analyst software (Sciex, USA).Amino acids, energy metabolites, and acyl-carnitines quantification and data processing were achieved as previously described. 47Data processing and analysis were performed by MetaboAnalyst 5.0 web tool.

Western blot
For Western blotting, 15 mg of corpus cavernosum were homogenized using the TissueLyser in Phosphate buffer saline (PBS) without

Statistical analysis
LC-MS/MS, Western blot data and NO 2 data were analyzed by unpaired two-tailed Student's t-test, after checking for the normal distribution with the Kolmogorov-Smirnov test.p < 0.05 was considered significant.Linear regression analysis and Pearson's correlation coefficient were computed to assess the potential relationship between two different variables.Analyses were performed using Prism, version 7.0a (GraphPad Software Inc., San Diego, CA, USA).

Assessment of steroid levels and 5α-R type II after finasteride treatment and its withdrawal in rat corpus cavernosum
The levels of 10 different steroids (i.e., PREG, PROG, DHP, ALLO, ISOALLO, DHEA, T, DHT, 3α-diol and 17β-E) were evaluated by LC-MS/MS in the corpus cavernosum after subchronic finasteride treatment (i.e., 24 h after the last injection, n = 12) and withdrawal (i.e., 1 month after drug suspension, n = 12).Data obtained indicate that T and DHT levels were affected by finasteride treatment (Figure 1A), while the other steroids were not significantly modified (data not shown).Indeed, a statistically significant increase in T levels F I G U R E 2 Assessment of enzyme protein expression.After subchronic finasteride treatment the relative protein expression of 5α-R type II was assessed by Western blotting.Data are expressed as % of vehicle ± SD with n = 6 for each group.Unpaired Student's t-test analysis: **p < 0.01 vs. vehicle group.
was associated with a significant decrease in the levels of its metabolite, DHT (Figure 1A).The levels of T and DHT (Figure 1B), as well as the other steroids analyzed (data not shown), were restored by suspension of the drug.
Interestingly, the local alteration of T conversion induced by the finasteride treatment was supported by a significant decrease in the protein expression of 5α-R type II (Figure 2).

Assessment of catecholamine plasma levels after finasteride treatment and its withdrawal
The levels of norepinephrine and its metabolite (i.e., epinephrine) were quantified in plasma by LC-MS/MS analysis, after subchronic finasteride treatment (n = 12) and withdrawal (n = 12).As reported in Figure 3A, subchronic finasteride treatment resulted in a significant increase in the levels of norepinephrine, associated with a decrease in those of epinephrine.In contrast, the assessment performed at finasteride withdrawal did not show significant variations in their levels (Figure 3B).

Targeted metabolomics by LC-MS/MS coupled with nitrate/nitrite colorimetric assay after finasteride treatment and its withdrawal in corpus cavernosum
To test if and to what extent finasteride-treated rats are characterized by specific metabolic patterns, we performed targeted metabolomics to investigate nucleotide, amino acid, biogenic amine, acylcarnitine, glycolysis, and tricarboxylic acid cycle intermediate levels in the corpus cavernosum, after subchronic finasteride treatment (n = 12) and withdrawal (n = 12).As reported in the representative scheme in F I G U R E 3 Evaluation of catecholamine levels in plasma.After subchronic finasteride treatment (A) and withdrawal (B), norepinephrine and epinephrine levels were analyzed by LC-MS/MS.Data are expressed as pg/μL ± SD with n = 6 for each group.Unpaired Student's t-test analysis: *p < 0.05 vs. vehicle group.
Figure 4A, specific enzymatic activities were modulated by finasteride treatment.In particular, as shown in Figure 4B, the activity of ornithine transcarbamylase (OTC), catalyzing the reaction between carbamoyl phosphate and ornithine resulting in citrulline synthesis, was significantly decreased.Similarly, a decrease in the activity of the nitric oxide synthase (NOS), an enzyme that converts arginine to citrulline, was observed (Figure 4B).As a confirmation of suppressed NOS activity, the concentration of nitrogen dioxide (NO 2 ) was assessed in the corpus cavernosum.As shown in Figure 4C, treatment with finasteride significantly decreased the NO 2 content.No significant changes in other specific metabolic patterns after finasteride treatment and withdrawal were observed (data not shown).

DISCUSSION
Data here reported indicate that the subchronic treatment with finasteride is able to directly affect the enzyme 5α-R type II in the rat corpus cavernosum and consequently the conversion of T into DHT.
Indeed, we here reported a decrease in its protein expression, associated with rise in the levels of its substrate and a decrease in the levels of its first product, DHT.Therefore, our observations indicated that a subchronic treatment with this drug, in addition to what we previously reported in plasma, 44 also locally decreases the levels of DHT.As already demonstrated, androgens, particularly DHT, play a role in penile erection. 35,36,38,48,49These effects involved changes in the NOS activity, which, as confirmed in the present study by metabolomic analysis, decreased after subchronic treatment with finasteride in the corpus cavernosum.As we reported, the decrease in the NOS activity, measured by the citrulline/arginine ratio and confirmed by the decrease in NO 2 levels, was associated with a decrease in the OTC activity, measured by citrulline/ornithine ratio.In this context, it is interesting to note that levels of T in the corpus cavernosum, which we reported to be increased after subchronic drug treatment, are negatively correlated with NOS activity (p = 0.01; r = −0.69),while tissue levels of DHT, which we demonstrated to be decreased, are positively correlated with OTC activity (p = 0.01; r = + 0.71).This is particularly interesting because, even if previous observations already indicated the presence of the enzyme 5α-R type II in the corpus cavernosum, 50,51 to our knowledge this is the first observation suggesting that a local change in T metabolism is linked to mechanisms associated with ED.
Indeed, to our knowledge, clinical and experimental studies evaluating ED have analyzed only steroid changes in plasma. 1 In this context, it is interesting to note that the finasteride treatment did not increase T levels in plasma, 44 suggesting that the T accumulation in corpus cavernosum could be involved in the dysfunction of the NOS activity.The finasteride subchronic treatment in male rats also confirmed alterations in the levels of other molecules involved in the control of penile erection, such as norepinephrine and its metabolite, epinephrine. 1 Indeed, we reported here that the plasma levels of norepinephrine and epinephrine were significantly increased and decreased, respectively, suggesting ED. 1,43 The inhibition of this catecholamine conversion seems to be due to a direct inhibition of the PNMT enzyme in the adrenal gland, which we have previously reported to be an off-target of the finasteride. 42Altogether, these observations suggest that the subchronic finasteride treatment in male rats induces molecular alterations associated with ED, not only in plasma, but also directly in the corpus cavernosum.A possible limitation of this study could be the only comparison of the results obtained with finasteride treatment versus those obtained in vehicle treated animals.Thus, the lack of a baseline control group that was not exposed to finasteride.The situation seems to be very different 1 month after drug suspension, suggesting full reversibility of the effects.Indeed, in this case we did not observe alterations in the corpus cavernosum levels of T and DHT, as well as of the other steroids we have assessed (i.e., PREG, PROG, DHP, ALLO, ISOALLO, DHEA, 3α-diol, and 17β-E).In agreement, our metabolomic analysis in the corpus cavernosum and evaluation of catecholamine levels in plasma showed no changes after finasteride withdrawal.Notably, a wider distribution of catecholamine levels following drug withdrawal than during treatment was highlighted, suggesting that animals may have a variable response to finasteride.
As previously mentioned, PFS patients (i.e., at finasteride withdrawal) show ED. 1,28,43However, the results obtained here seem to indicate that finasteride withdrawal does not alter the peripheral machinery controlling penile erection.On the other hand, it is important to highlight that observations obtained in PFS patients and in male rats after finasteride suspension show the nervous system as an important target, with alterations that could be associated with sexual dysfunction, such as depressive symptomatology and peripheral neuropathy. 7,20,27,28,32,45In addition, it is important to remember that male sexual behavior (e.g., sexual motivation and reward to sexual performance) is under the control of complex neural circuits. 1,52erefore, future experiments will be important to explore whether finasteride withdrawal affects these circuits and molecules involved, such as neurotransmitters and neuropeptides. 1,52 terone (PROG), dihydroprogesterone (DHP), allopregnanolone (ALLO), isoallopregnanolone (ISOALLO), dehydroepiandrosterone (DHEA), T, DHT, 5α-androstane-3α, 17β-diol (3α-diol) and 17β-estradiol (17β-E) by liquid chromatography-tandem mass spectrometry (LC-MS/MS); (ii) the expression of 5α-R type II by Western blot analysis; (iii) nitrogen dioxide concentration by nitrate/nitrite colorimetric assay; (iv) nucleotide, amino acid, biogenic amine, acylcarnitine, glycolysis, and tricarboxylic cycle intermediate levels by targeted metabolomics performed with LC-MS/MS analysis.
data were obtained by LC-MS/MS.We used an API-3500 triple quadrupole mass spectrometer (AB Sciex, MA, USA) coupled with an ExionLC AC System (AB Sciex).10 mg of tissue were smashed in a TissueLyser for 5 min at maximum speed in 250 μL of icecold methanol-water-acetonitrile (55:25:20 v/v).Lysates were spun at 15,000×g for 15 min at 4 • C. Samples were then dried under N 2 flow at 40 • C and resuspended in 125 μL of ice-cold methanol-wateracetonitrile (55:25:20 v/v) for subsequent analyses.

F I G U R E 1
Ca2+  and Mg 2+ ; EDTA 0.5 M pH 8; Igepal) supplemented with a protease and phosphatase cocktail inhibitor (Cell Signaling, Italy).Then, to remove particulate matter, the tissue homogenates were centrifuged at 2000 rpm for 5 min at 4 • C. The protein content was quantified using a Bradford Assay (Bio-Rad, Italy) and samples containing equal amounts of protein (20 μg) were heated to 100 • C for 5 min.Samples were run on Criterion TGX 4−15% gradient gels (Bio-Rad, Italy) and transferred to nitrocellulose membranes.Stain-free technology was used to determine equal loading.Nitrocellulose membranes were blocked in 5% bovine serum albumin (BSA)-PBS-Tween for 1 h at room temperature and exposed to primary antibody (rabbit 5α-R type II; abcam 124877, 1:2000) overnight.Then, a conjugated secondary antibody, according to the primary antibody was chosen.The protein bands were detected on membranes using the ECL method.ECL signals were acquired with a ChemiDocTM XRS+ system (Bio-Rad, Italy) and analyzed with Image LabTM software version 5.2.1.The detected protein targets were normalized with the total lane values obtained with the stain-free technology (Bio-Rad, Italy).The mean control value within a single experiment was set to 100 and all other values were expressed as a percentage.Assessment of androgen levels in the corpus cavernosum.After subchronic finasteride treatment (A) and withdrawal (B), T and DHT levels were analyzed by LC-MS/MS.Data are expressed as pg/mg ± SD with n = 6 for each group.Unpaired Student's t-test analysis: ***p < 0.001 vs. vehicle group.

F I G U R E 4
Evaluation of the metabolomic status and NO 2 production in corpus cavernosum after finasteride treatment.(A) As summarized in the representative picture, ornithine, which comes from the urea cycle goes into the mitochondria to be converted in citrulline by ornithine transcarbamylase (OTC) and carbamoyl phosphate.Citrulline can dissolve in the cytoplasm to be used in the urea cycle to produce also the substrate of nitric oxide synthase (NOS) (i.e., arginine).The last step of this pathway is the production by NOS of citrulline and nitrogen dioxide (NO 2 ).These pathways are altered by finasteride treatment; (B) OTC and NOS activities studied by targeted metabolomics; (C) NO 2 content analyzed by nitrate/nitrite colorimetric assay.Data are expressed as % of vehicle ± SD (B) or μM/mg ± SD (C); n = 6 for each group.Unpaired Student's t-test analysis: *p < 0.05 vs. vehicle group; **p < 0.01 vs. vehicle group.