199Archivio Italiano di Urologia e Andrologia 2018; 90, 3 ORIGINAL PAPER Serenoa repens extracts: In vitro study of the 5α-reductase activity in a co-culture model for Benign Prostatic Hyperplasia Daniela Buonocore 1, Manuela Verri 1, Laura Cattaneo 1, Sara Arnica 1, Michele Ghitti 2, Maurizia Dossena 1 1 University of Pavia, Pavia (Italy), Department of Biology and Biotechnology “L. Spallanzani”; 2 University of Pavia, Pavia (Italy), Department of Earth and Environmental Sciences (DSTA) - Unit of Statistical Analyses (UNISTAT). Objectives. Benign Prostatic Hyperplasia (BPH) is a form of benign tumor that occurs in humans mainly with ageing. It affects more than 50% of over 50 years old males and it is characterized by an increased syn- thesis of dihydrotestosterone (DHT), due to the 5α-reductase activity. The BPH therapeutic approach mainly uses 5α-reduc- tase inhibitors, such as the active compounds present in the extracts deriving from species Serenoa repens. Many lipido- sterolic extracts are available on the market, which are obtained with different solvents, among them ethanol is recog- nized as non-toxic and has less handling risks than hexane. The purpose of the present experimental study was to investi- gate in-vitro the potency of an ethanol extract of S. repens com- paring it with an n-hexane one. Materials and methods. Two different lipido-sterolic extracts of S. repens have been tested: ethanol extract and n-hexane extract, two batches for each one. The inhibitory action of the extract was evaluated estimating in-vitro the activity of enzyme 5α-reductase type I (5α-RI), which was mainly active under the experimental condition of pH 7.5. DHT amount, synthesized from testosterone (1 μM), was evaluated in a co-culture model of epithelial cells and fibroblasts resulting from prostatic biopsy of a patient with BPH. Results. The analysis of the resulting dose-response curves showed that the entire S. repens extracts inhibited the 5α-RI showing no difference between the two kinds of extract or between the batches. The resulting IC50 values were the follow- ing: 8.809 (95% CI = 5.133-15.56) and 9.464 (95% CI = 5.094- 18.27) for ethanol extracts; 11.08 (95% CI = 6.389-19.98) and 12.72 (95% CI = 7.758-21.53) for n-hexane extracts. Conclusions. The potency of ethanol extracts of S. repens was comparable with the one of n-hexane extracts. KEY WORDS: Benign Prostatic Hyperplasia; Dihydrotestosterone; Ethanol extracts; In-vitro study; 5α-reductase; Serenoa repens. Submitted 7 May 2018; Accepted 12 July 2018 Summary No conflict of interest declared. action of enzyme 5α-reductase, which leads to an increase of the prostate size and causes various disorders, especially in the lower urinary tract (1, 2). The approach for the treatment and therapy of BPH is mainly based on the use of inhibitors of 5α-reductase enzyme, in particu- lar natural inhibitors, such as the active compounds present in the alcoholic extracts deriving from species Serenoa repens, a typical palm common in the Subtropical sandy soils and in the southern coasts of the United States. Although the sure mechanism of action of S. repens is not yet fully understood, numerous mecha- nisms have been proposed (3) and the presence of spe- cific fatty acids (saturated and unsaturated fatty acids) and phytosterols in its alcoholic extracts confers upon S. repens an anti-inflammatory and above all antiprolifera- tive action on prostatic tissue (4). Both a systematic Cochrane review of the literature (5) and a meta-analysis (6) assessed the safety profile and the clinical efficacy of S. repens in the treatment of symptoms in patients with BPH. These studies have shown an improvement of uri- nary tract disorders, with a mild to moderate effect and, therefore, a lack of therapeutic equivalence between S. repens extracts of different brands and between extracts of a single brand, but belonging to different manufactur- ing batches (5, 6). Many lipido-sterolic extracts are com- mercially available, which were obtained with different solvents, among them ethanol is recognized as non-toxic and it has less handling risks than hexane (7, 8). The purpose of the present experimental study was to inves- tigate in-vitro the potency of an ethanol extract of S. repens comparing it with an n-hexane one. In particular, the activity of enzyme 5α-reductase was estimated in- vitro concerning the inhibitory action of the extracts, using co-cultures of epithelial cells and fibroblasts result- ing from prostatic biopsy of a patient with BPH. MATERIALS AND METHODS Samples Two extracts, belonging to commercial brands SABA® (Lampugnani Farmaceutici) and PERMIXON® (Pierre Fabre Pharma), supplied with marketing authorization in DOI: 10.4081/aiua.2018.3.199 INTRODUCTION Benign Prostatic Hyperplasia (BPH) is a form of benign tumor that occurs in humans mainly with ageing. In fact, it affects more than 50% of over 50 years old males, with an incidence rate directly proportional to age (1). It is characterized by an increased synthesis of dihydrotestos- terone (DHT) starting from testosterone, due to the Buonocore_Stesura Seveso 03/10/18 09:45 Pagina 199 Archivio Italiano di Urologia e Andrologia 2018; 90, 3 D. Buonocore, M. Verri, L. Cattaneo, S. Arnica, M. Ghitti, M. Dossena 200 several EU Member States (including Italy) and belonging to the class of drugs employed to counter the increase in prostatic volume in males with BPH, have been tested. The two branded drugs contained the lipid-sterolic extract of S. repens (320 mg/soft capsule) as active ingre- dient and particularly, ethanol extract (SABA®) and hexane extract (PERMIXON®). Two different batches of each brand were tested: SABA® (G08363 and G08364); PERMIXON® (F 11917 and G07340), mentioned below as “samples”. The content of one soft capsule was weighed (10 mg) and dissolved in N-hexane; then the solvent was evapo- rated to leave the solid extract that was dissolved in ethanol (70%) to give a concentration of 10 mg/ml. This stock solution was further diluted in appropriate media to provide a working solution of 1 mg/ml (9). Co-culture cell model Two types of primary cells were used: epithelial cells and fibroblasts, previously isolated from prostatic tissue obtained during a biopsy from a patient with established BPH. Informed consent of the patient was obtained. Primary cultures of fibroblast and epithelial cells were obtained as previously described (2). Both the epithelial and fibroblast cells were cultured separately in flasks with culture medium, complete with fetal bovine serum (10%), DMEM and RPMI 1640 respectively, and kept in an incubator in a humidified atmosphere (37°C and 5% CO2). Corresponding to the third generation step, both the epithelial cells and fibroblasts, were transferred into multiwell plates of 12 wells provided with Transwell® (3470 Clear-Corning), or supports consisting of a micro- porous polystyrene membrane. This enabled to keep the two cell populations separate but, at the same time, to allow their interactions. In particular: epithelial cells were sown in the number of 50,000 on the bottom of each well; on the contrary, fibroblasts were sown in the number of 5,000 on the microporous polystyrene membrane. Co-cultures were maintained in DMEM:RPMI-1640 (1:1 v/v) media + 10% fetal calf serum at 37°C in 5% CO2 (10). Furthermore, in order to verify that the fibroblasts were correctly adhering to the membrane, the quantitative tech- nique TEER (Transepithelial Transendothelial Electrical Resistance) was applied; this technique enabled to examine the integrity of the cell junctions (tight junctions) by meas- uring the resistance that the monolayer of fibroblasts opposed to the passage of electric current (11). Measurement of the trans-epithelial resistance value, expressed in Ohms and normalized for the polystyrene membrane area (Ω/cm²), was performed using a voltmeter (Millicell® ERS-2, Merck Millipore) equipped with two elec- trodes of different length (data not reported). 5α-reductase activity assay The co-culture cells were pre-treated for a period of four days with a non-toxic concentration (10 μg/ml) of each sample (SABA®: G08363 and G08364; PERMIXON®: F 11917 and G07340). Afterwards the cells were harvested by trypsinization, centrifuged and the pellet was suspend- ed in RPMI-1640, supplemented with 10% fetal calf serum, and then counted. The pH value of the medium solution was 7.85, an optimum value for the activity of 5 α-reductase type I (12). Cell suspensions were added into tubes containing testosterone (1 μM) as substrate, a NADPH-regenerating system (0.5 mM glucose-6-phos- phate, 0.06 U glucose-6-phosphate dehydrogenase, 50 μM NADPH) (H+ donor) (13) and different concentrations of each sample: 1 mg/ml, 100 μg/ml, 10 μ /ml, 5 μg/ml, 1 μg/ml, 0.1 μg /ml, 0.01 μg/ml; these concentrations were chosen basing on the cytotoxicity results, previously obtained in laboratory applying the MTT test (14)) and in agreement with data present in the literature (9, 10). The tubes were then incubated at 37° C for 30 min in a stirring water bath. The reaction was stopped by dipping the tubes into liquid nitrogen (9, 10, 12). The 5α-reductase type I activity was assessed by measur- ing the conversion of testosterone to dihydrotestosterone (DHT) applying a qualitative/quantitative Enzyme-Linked Immunosorbent Assay (Abnova KA1886), as previously described (15). Enzyme activity is expressed as a percentage of the con- trol. The conversion of 1 μM of testosterone in the absence of inhibitors is defined as 100% activity (2.95- 3.63 nmol of DHT/106 cells/min for type I isoenzyme). Dose-effect response curves were analyzed using a sig- moid maximum-effect model with a variable slope (Graphpad, Prism7). Inhibitory potency was assessed by estimating the IC50 value that represented the concentration (μg / ml) capa- ble of determining the 50% of the maximal effect (enzy- matic inhibition). The highest IC50 value was then divided by the IC50 values obtained for each sample for the relative potency (RP) (the relative potency value = 1 was considered the lowest value). Statistical analysis The nonlinear-regression curves were analyzed applying two different Linear Mixed Models, one to analyze the com- mercial brands (SABA® vs PERMIXON®) and one for the batches. The analyses were performed with software R. RESULTS All the samples of S. repens extracts (F 11917; G07340; G08363; G08364) were found to be capable of inhibit- ing 5α-reductase (5α-RI) in a prostatic co-cultured epithelial and fibroblast cells, as shown by dose-effect curves in Figure 1 (enzyme activity (%) as a function of the logarithm of concentrations). Regarding the potency of each sample, evaluated by IC50 value (Table 1) and highlighted by the relative potency (Table 2), SABA® (IC50 = 8.809 and 9.464) showed a higher efficacy on 5α-RI than PERMIXON® (IC50 = 11.08 and 12.72), but there were non-signifi- cant differences between the two brands and batches, as mentioned below. Regarding the comparison between the two brands (SABA® vs PERMIXON®), the analysis of deviance (Type III tests) for 5α-RI data set showed that the model was non-significant (Response: enzyme; Chisq Df Pr (> Chisq); (Intercept) 28.3600 1 1.007e-07 ***; Brand 2.9228 1 0.08734; Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1). Buonocore_Stesura Seveso 03/10/18 09:45 Pagina 200 Regarding the comparison between the batches (F 11917; G07340; G08363; G08364), the analysis of deviance (Type III tests) showed that the model was non-signifi- cant [Response: enzyme; Chisq Df Pr (> Chisq); (Intercept) 28.0176 1 1.202e-07 ***; Batch 2.9326 3 0.4021; Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1]. Regarding the relative potency (Table 2), both SABA® and PERMIXON® showed comparable potencies between the two batches and PERMIXON® showed the lowest values. DISCUSSION All the extracts of S. repens that we tested, SABA® (G08363 and G08364) and PERMIXON® (F 11917 and G07340), have shown to inhibit enzyme 5α-reductase in a co-culture model of human prostatic cells (epithelial and fibroblasts), as reported also in the literature (9, 10). Even though the potency of SABA samples of inhibiting the activity of enzyme 5α-reductase type I showed high- er values than PERMIXON’s values, there were no sig- nificant statistical differences between the two brands or the batches. These in-vitro results pointed out that the S. repens ethanol extract (SABA®) was equivalent to the n- hexane one (PERMIXON®). This conclusion is important because basing on the equivalence of efficacy for both tested extracts, obtained with different extraction methods, it would be better to use organic solvents (ethanol) recognized as environ- mentally safer and alternative to hexane, which is a sol- vent obtained from petrochemical sources that can remain in potential traces in edible oils after refining and can be emitted during extraction and recovery and that has been identified as an air pollutant since it can react with other pollutants to produce ozone and photochem- ical oxidants (7, 8, 16). Moreover, data that we obtained allow us to speculate on the possible effects of the ethanol extract from S. repens on the pathology of BPH: it is known that clinical benefits are gen- erally associated with the free fatty acid content, along with a small contribution from unsaponifiable components (17, 18). Particularly, the relative inhibitory efficacy of the various free fatty acids seems to depend on the length of the carbon chain and its sat- uration state (17). For example, lauric acid (short saturated, C12 chain) inhibits both 5 α-reductase type I and II, while oleic acid with a C18 unsaturated chain (C18 ∆9) and linoleic acid (C18 ∆9, 12) have a good activ- ity on type I but not on type II (15). SABA® and PERMIXON® (19) were analyzed in 201Archivio Italiano di Urologia e Andrologia 2018; 90, 3 S. repens inhibits 5α-reductase Table 1. Comparative potency of the two extracts on 5α-reductase type I, given by IC50 value and related 95% CI. PERMIXON SABA F11917 G07340 G08363 G08364 IC50 (5α-RI)1 11.08 12.72 8.809 9.464 95% CI2 6.39 to 19.98 7.76 to 21.53 5.13 to 15.56 5.09 to 18.27 15α-RI: 5α-reductase type I. 2CI: Confidence Interval. Table 2. Relative potency of the various extracts on 5α-reductase type I (considering RP = 1 as the lowest potency). 5α-RI1 SABA SABA PERMIXON PERMIXON Batch G08363 G08364 F11917 G07340 RP2 1.444 1.344 1.148 1 15α-RI: 5α-reductase type I. 2RP: relative potency. Figure 1. Inhibition of 5α-reductase type I. Enzyme activity, depending on the logarithm of concentrations, is expressed in percentage as compared with the positive control (100% of enzyme activity converting testosterone 1μM, in the absence of inhibitors). Buonocore_Stesura Seveso 03/10/18 09:45 Pagina 201 Archivio Italiano di Urologia e Andrologia 2018; 90, 3 D. Buonocore, M. Verri, L. Cattaneo, S. Arnica, M. Ghitti, M. Dossena 202 terms of concentration in free fatty acids, methyl and ethyl esters, long-chain esters and glycerides. These analyses revealed that each of the individual FFAs ana- lyzed was found in similar proportions in the tested products, with lauric and oleic acids present at the high- est concentrations in every tested sample and similar concentrations of methyl and ethyl esters were found. Instead, glyceride content was particularly high (420%) in SABA® and not in PERMIXON®. Generally, a similar content composition was observed between SABA® and PERMIXON® (19). So, our results are explained by data reported above: similar content composition is associated with the same potency of ethanol and hexane extracts. CONCLUSIONS Ethanol extract of S. repens has shown in vitro a potency of inhibiting the activity of enzyme 5α-reductase compa- rable with potency of n-hexane extract of S. repens. REFERENCES 1. Rył A, Rotter I, Grzywacz A, et al. Molecular analysis of the SRD5A1 and SRD5A2 genes in patients with benign prostatic hyper- plasia with regard to metabolic parameters and selected hormone levels. Int J Environ Res Public Health. 2017; 14:1318. 2. Bayne CW, Donnely F, Chapman K, et al. A Novel coculture model for benign prostatic hyperplasia expressing both isoforms of 5α-reductase. JCEM. 1998; 83:206. 3. Buck AC. Is there a scientific basis for the therapeutic effects of Serenoa repens in benign prostatic hyperplasia? Mechanisms of action. J Urol. 2004; 172:1792. 4. Capasso F, De Pasquale R, Grandolini G. Droghe contenenti lipi- di. In: Farmacognosia. Botanica, chimica e farmacologia delle piante medicinali. Italia: Ed. Springer Verlag, 2014. 5. Wilt T, Ishani A, MacDonald R. Serenoa repens for benign pro- static hyperplasia. Cochrane Database Syst Rev. 2002; CD001423. 6. Boyle P, Robertson C, Lowe F, Roehrborn C. Updated meta- analysis of clinical trials of Serenoa repens extract in the treatment of symptomatic benign prostatic hyperplasia. BJUI. 2004; 39:751. 7. Johnson LA & Lusas EW. Comparison of alternative solvents for oils extraction. J Am Oil Chem Soc. 1983; 60:229. 8. Ferreira-Dias S, Valente DG, Abreu JMF. Comparison between ethanol and hexane for oil extraction from Quercus suber L. fruits. Grasas y Aceites. 2003; 54:378. 9. Scaglione F, Lucini V, Pannacci M, et al. Comparison of the Potency of Different Brands of Serenoa repens Extract on 5α- Reductase Types I and II in Prostatic Co-Cultured Epithelial and Fibroblast Cells. Pharmacology. 2008; 82:270. 10. Bayne CW, Donnely F, Ross M, Habib FK. Serenoa repens (Permixon®): A 5α-reductase Types I and II Inhibitor-New evidence in a Coculture Model of BPH. Prostate. 1999; 40:232. 11. Srinivasan B, Kolli Reddy A, Esch MB, et al. TEER measurement techniques for in vitro barrier model systems. J Lab Autom. 2015; 20:107. 12. Smith CM, Ballard SA, Worman N, et al. 5 alpha-reductase expression by prostate cancer cell lines and benign prostatic hyper- plasia in vitro. J Clin Endocrinol Metab. 1996; 81:1361. 13. Weisser H, Tunn S, Behnke B, Krieg M. Effects of the sabal serru- lata extract IDS 89 and its subfractions on 5 alpha-reductase activity in human benign prostatic hyperplasia. Prostate. 1996; 28:300. 14. Mosmann T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J Immunol Methods. 1983; 65:55. 15. Di Silverio F, D'Eramo G, Lubrano C, et al. Evidence that Serenoa repens extract displays an antiestrogenic activity in prostatic tissue of benign prostatic hypertrophy patients. Eur Urol. 1992; 21:309. 16. Wan PJ, Pakarinen DR, Hron RJSr, et al. Alternative hydrocar- bon solvents for cottonseed extraction. J Am Oil Chem Soc. 1995; 72:653. 17. Paubert-Braquet M, Cousse H, Raynaud JP, et al. Effects of the lipido-sterolic extract of Serenoa repens (Permixon ®) and its major components on basic fibroblast growth factor-induced proliferation of cultures of human prostate biopsies. Eur Urol. 1998; 33:340. 18. Raynaud JP, Cousse H, Martin PM. Inhibition of type 1 and type 2 5alpha-reductase activity by free fatty acids, active ingredients of Permixon®. J Steroid Biochem Mol Biol. 2002; 82:233. 19. Habib FK & Wyllie MG. Not all brands are created equal: a comparison of selected components of different brands of Serenoa repens extract. Prostate Cancer Prostatic Dis. 2004; 7:195. Correspondence Daniela Buonocore, MD (Corresponding Author) daniela.buonocore@unipv.it Manuela Verri, MD manuela.verri@unipv.it Laura Cattaneo, MD laura.cattaneo02@universitadipavia.it Sara Arnica, MD sara.nica90@gmail.com Maurizia Dossena, MD maurizia.dossena@unipv.it Via Ferrata, 9-27100 Pavia, Italy Michele Ghitti, MD ghitti.michele@gmail.com Via Ferrata, 1-27100 Pavia, Italy Buonocore_Stesura Seveso 03/10/18 09:45 Pagina 202