Pharmacological characterization and chemical fractionation of aliposterolic extract of saw palmetto (Serenoa repens): Effects on ratprostate contractility

Thiam Chua a, Nicole T. Eise a,b, Jamie S. Simpson b, Sabatino Ventura a,n

a Drug Discovery Biology, Monash Institute of Pharmaceutical Sciences, Monash University, 381 Royal Parade, Parkville, VIC 3052, Australiab Medicinal Chemistry, Monash Institute of Pharmaceutical Sciences, Monash University, Parkville, VIC 3052, Australia

a r t i c l e i n f o

Article history:Received 27 September 2013Received in revised form6 December 2013Accepted 18 December 2013Available online 21 January 2014

Keywords:Chemical fractionationSmooth muscle contractilityBenign prostatic hyperplasiaLower urinary tract symptomsHerbal medicines

a b s t r a c t

Ethnopharmacological relevance: Saw palmetto (Serenoa repens) was first used medicinally by nativeAmerican Indians to treat urological disorders. Nowadays, saw palmetto extracts are widely used inEurope and North America to treat the urinary symptoms associated with benign prostatic hyperplasiaeven though its mechanisms of action are poorly understood. This study aimed to characterize thebioactive constituents of a lipid extract of saw palmetto that are able to affect contractility of the ratprostate gland. The mechanism of action will also be investigated.Materials and methods: A commercially available lipid extract of saw palmetto was subjected tofractionation using normal phase column chromatography. Composition of fractions was assessed byproton nuclear magnetic resonance spectroscopy (1H NMR) and mass spectrometry (MS). Contractileactivities of these fractions were evaluated pharmacologically using isolated preparations of rat prostategland and compared to the activity of the crude extract.Results: Saw palmetto extract inhibited contractions of the rat prostate gland which were consistentwith smooth muscle relaxant activity. Only the ethyl acetate fraction resulting from chromatographyinhibited contractions of isolated rat prostates similarly to the inhibition produced by the crude lipidextract. Comparison with authentic samples and analysis of NMR data revealed that this bioactivity wasdue to the fatty acid components present in the ethyl acetate fraction. Bioassay using variouspharmacological tools identified multiple contractile mechanisms which were affected by the bioactiveconstituents.Conclusion: A fatty acid component of saw palmetto extract causes inhibition of prostatic smooth musclecontractions via a non-specific mechanism.

& 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

The pathogenesis of benign prostatic hyperplasia (BPH) involvesa combination of two elements: the static and dynamic compo-nents. The static component involves prostatic growth, caused bytestosterone diffusing into the prostate cells and being converted todihydrotestosterone by the enzyme 5α-reductase. The dynamiccomponent relates to prostatic smooth muscle tone, which is underthe influence of the sympathetic nervous system and mediated by

α1-adrenoceptors. Conventional pharmacotherapies for BPH consistof 5α-reductase inhibitors (dutasteride and finasteride) and α1-adrenoceptor antagonists (tamsulosin, alfuzosin, prazosin, terazosinand doxazosin). Smooth muscle contraction in the urethra, prostateand bladder neck is considered to be the main contributor to thevoiding symptoms associated with BPH. Hence, α1-adrenoceptorantagonists are regarded as the preferred and more effective drugclass (Miano et al., 2008).

There are a number of phytotherapeutic agents currently usedfor the treatment of BPH (Lowe and Ku, 1996). Most of theseextracts are derived from the roots, seeds, bark or fruits of plantssuch as the African plum, purple cone flower, pumpkin seeds, rye,stinging nettle, red clover and saw palmetto. Of these, extracts ofsaw palmetto berries which were first used for the medicinaltreatment of urological disorders by the native American SeminoleIndians, are by far the most popular treatment for urinarysymptoms caused by BPH (Gerber, 2000; Wilt et al., 2000). Thereare several different saw palmetto preparations on the market but

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/jep

Journal of Ethnopharmacology

0378-8741/$ - see front matter & 2014 Elsevier Ireland Ltd. All rights reserved.http://dx.doi.org/10.1016/j.jep.2013.12.030

Abbreviations: ATP, adenosine-50-triphosphate disodium salt; 2-APB,2-aminoethoxydiphenyl borate; BPH, benign prostatic hyperplasia; ESI/APCI,electrospray ionization/atmospheric pressure chemical ionization; IP3, inositoltriphosphate; MS, mass spectrometry; PDA, phorbol-12,13-diacetate; PKC, proteinkinase C; 1H NMR, proton nuclear magnetic resonance spectroscopy; TLC, thin layerchromatography

n Corresponding author. Tel.: þ61 3 99039566; fax: þ61 3 99039638.E-mail address: Sab.Ventura@monash.edu (S. Ventura).

Journal of Ethnopharmacology 152 (2014) 283–291

the most widely used are the liposterolic extracts of saw palmetto.Liposterolic oily extracts are prepared with either n-hexane, 90%ethanol (w/w) or by supercritical fluid extraction with CO2 and arenormally standardised to contain 70 to 95% of free fatty acids andsterols. Despite the presence of free fatty acids, the liposterolicextracts also contain small amounts of phytosterols, fatty alcohols,flavonoids and polyprenoids (Plosker and Brogden, 1996).

Over the years, various mechanisms of action have beenproposed for the beneficial effects of extracts of saw palmetto. Itis believed that saw palmetto works by a variety of mechanisms,befitting its polypharmaceutic nature. The pharmacological activ-ity of saw palmetto includes; inhibition of 5α-reductase activity(Di Silverio et al., 1992; Iehle et al., 1995), anti-inflammatoryactivity (Paubert-Braquet et al., 1997), anti-androgen properties(Ravenna et al., 1996) and anti-edema effects (Tarayre et al., 1983).However, saw palmetto extract also appears to have spasmolyticand smooth muscle relaxant activity as demonstrated in theurinary bladder of guinea pigs and rats (Gutierrez et al., 1996).This indicates that saw palmetto extracts may alleviate symptomsby inhibiting both the dynamic and static components of BPH.

Previously, we have demonstrated that commercially availableethanolic extracts of saw palmetto contain the indirectly actingsympathomimetic tyramine (Cao et al., 2006; Chua et al., 2011).This effect causes an indirect α1-adrenoceptor mediated contrac-tion via the release of noradrenaline from sympathetic neurons,which would cause urethral obstruction and be detrimental toBPH sufferers. The present study was undertaken to examine theacute pharmacological properties of a widely used liposterolicextract of saw palmetto on the isolated rat prostate gland in vitro.In addition, chemical fractionation was performed to elucidate thebioactive compounds contained in saw palmetto that mediate itspharmacological activity.

2. Materials and methods

2.1. Animals and tissues

Male Sprague-Dawley rats (200–300 g) were housed at theMonash Clayton and Parkville animal houses in a controlledenvironment of 22 1C and exposed to a photoperiod of 12 h light/12 h dark. Animals were provided with food and water ad libitum.The animals were killed by cervical dislocation. An abdominalincision was made, exposing the male urogenital tract and the leftand right lobes of the prostate were removed providing twoprostate preparations from each rat. The excess fat and connectivetissues were carefully removed and prostates were then placed in aPetri dish containing Krebs–Henseleit solution (mM: NaCl 118.1, KCl4.69, KH2PO4 1.2, NaHCO3 25.0, glucose 11.7, MgSO4 1.1 and CaCl22.5, carbogenated to pH 7.4). Prior approval for animal experimen-tation was obtained from the Monash University Standing Commit-tee on Animal Ethics in Animal Experimentation (Ethics numberVCPA.2009.15).

2.2. Isolated organ bath studies

Isolated prostates were mounted in separate 10 ml glass, water-jacketed organ baths, containing Krebs–Henseleit solutionbubbled with 5% CO2 in O2, and maintained at 37 1C. One end ofthe prostate was attached to a perspex tissue holder and the otherto an isometric Grass FT03 force-displacement transducer(Telefactor, West Warwick, RI, USA), which measured the forceproduced by the tissue. A PowerLab data acquisition system (Chartv5) (ADInstruments, Bella Vista, NSW, Australia) received informa-tion from the force transducer for recording on a personalcomputer. Preparations were equilibrated for 60 min, under a

resting force of 0.5–0.8 g. To ensure the viability of the tissue,nerve terminals within the prostatic smooth muscle were elec-trically stimulated via two vertical parallel platinum electrodesincorporated in the tissue holder, which was connected to a GrassS88 stimulator delivering pulses of 0.5 ms, 60 V, at 0.01 Hz.

2.2.1. Effects of saw palmetto on nerve-mediated contractionTo assess whether the liposterolic extract of saw palmetto

affected contractile responses produced by nerve-stimulation,frequency response curves (0.5 ms, 60 V, 0.1–20 Hz) to electricalfield stimulation were conducted using a frequency progressionratio of approximately one third of a log unit. The pulses weredelivered for 10 pulses at low frequencies (r1 Hz) and for 10 s athigher frequencies (Z1 Hz). Trains of electrical field stimulationwere delivered at intervals of 10 min. An initial frequency–response curve was constructed to determine the contractileresponse of each tissue at each frequency. A subsequent fre-quency–response curve was constructed 1 h later in the presenceof liposterolic extract of saw palmetto or vehicle after the tissuehad been exposed to the extract or vehicle for 15 min. Oneprostate from each rat was used to construct frequency–responsecurves to saw palmetto extract while the contralateral prostatewas used as a vehicle control.

In a subset of experiments, fractions produced by chromato-graphic fractionation and commercially available fatty acids wereused instead of the liposterolic extract of saw palmetto, using thesame discrete frequency–response curve protocol as above. Theeffect of triglycerides was also examined using the commerciallyavailable glyceryl trioleate. The above protocol was adapted usinga single frequency of 10 Hz.

2.2.2. Effects of saw palmetto extract on contractile responsesinduced by noradrenaline, ATP, acetylcholine and KCl

Following the initial 60 min equilibration period discrete con-centration–response curves to saw palmetto were constructedagainst various agonists. The agonists used were noradrenaline(1 μM), acetylcholine (10 μM), ATP (10 mM) and high potassiumKrebs–Henseleit solution (mM: NaCl 38.0, KCl 80.0, KH2PO4 1.2,NaHCO3 25.0, glucose 11.7, MgSO4 1.1 and CaCl2 2.5, carbogenatedto pH 7.4) (KCl (80 mM)). In these experiments, one prostate fromeach rat was used to construct concentration–response curves tosaw palmetto extract while the contralateral prostate was used asa vehicle control. Agonists were added to the organ bath 15 minafter addition of each concentration of saw palmetto extract. Oncethe contractile responses of the agonist had reached a plateau, thetissue was washed out with four to five times the bath volume,followed by the addition of the next saw palmetto extractconcentration.

2.2.3. Effects of various pharmacological tools on actions of sawpalmetto extract

Following the 60 min equilibration period, frequency–responsecurves to electrical field stimulation were constructed to deter-mine the contractile response of the tissue at each frequency.A subsequent frequency–response curve was constructed in thepresence of a test drug after the tissue had been exposed to thedrug for 15 min. The test drugs used were the β-adrenoceptorantagonist, propranolol (1 mM); the Rho-kinase inhibitor, Y27632(10 mM); the protein kinase C (PKC) activator, phorbol-12,13-diacetate (PDA) (1.0 mM); the calmodulin antagonist, w-7(100 mM); the L-type Ca2þ channel blocker, nifedipine (1.0 mM);the inositol triphosphate (IP3) receptor antagonist, 2-amino-ethoxydiphenyl borate (2-APB) (60 mM) and ryanodine (10 mM);.A final frequency–response curve was established in the presence

T. Chua et al. / Journal of Ethnopharmacology 152 (2014) 283–291284

of the test drug and saw palmetto extract following pre-incubationfor 15 min.

In a subset of experiments investigating the role of PKC, tissueswere not electrically field stimulated after the 60 min equilibrationperiod but instead, were exposed to a priming dose of noradrena-line (0.1 mM) or carbachol (0.1 mM). Once the contractile responseof the tissue had reached a maximum, the tissues were washedand allowed 30 min to recover. Following this, discrete concentra-tion–response curves to noradrenaline (1.0 nM–30 mM) or carba-chol (0.1 mM–1.0 mM) were constructed using a dose progressionratio of half a log unit. Once the contractile response to eachconcentration of noradrenaline or carbachol had reached aplateau, tissues were washed and allowed 15 min to recover beforethe next concentration was applied. If no response was observedafter 30 s, the tissues were washed and allowed 15 min to recover.Following the initial concentration–response curve, tissues wereallowed to rest for 45 min before being exposed to PDA (1.0 mM)for 15 min before a second concentration–response curve wasconstructed. During the second curve, PDA was replaced after eachbath wash. To test whether saw palmetto extract was acting via amechanism involving PKC, a final concentration–response curvewas constructed in the presence of PDA and saw palmetto extractfollowing pre-incubation for 15 min.

2.3. Drugs and vehicle solutions

A sample of liposterolic extract of saw palmetto was generouslydonated by Indena SpA (Sabalselects, Milan, Italy). This extract isexported widely throughout the world to be used in variouscommercially available saw palmetto preparations. The extractshows the following analytical specifications at 20 1C: density,0.89; refractive index, 1.45; acid value, 212.9, iodine value, 45.3,saponification value, 226.8; peroxide value, 0.5. Table 1 shows thechemical composition of Sabalselects. The extract was dissolved at500 mg/ml in absolute ethanol and 10% v/v Tween-80 and wasstored at room temperature. The required dilutions were preparedon the morning of experimentation. Lauric acid, oleic acid, linoleicacid and myristic acid were purchased from Sigma Aldrich(St Louis, MI, USA). A concentration of 50 mg/ml glyceryl trioleate(Sigma Aldrich, St Louis, MI, USA), was made fresh on the morningof experimentation using absolute ethanol and 4% v/v Tween-80.

The following drugs were used: (�) arterenol (noradrenaline)bitartrate salt, adenosine-50-triphosphate disodium salt (ATP),acetylcholine chloride, trans-4-[(1R)-1-aminoethyl]-N-4-pyridinyl-cyclohexanecarboxamide dihydrochloride (Y27632), phorbol-12,13-diacetate (PDA), carbamylcholine chloride (carbachol), nife-dipine, 2-aminoethoxydiphenyl borate (2-APB) and DL-propra-nolol hydrochloride (all purchased from Sigma Aldrich; St Louis,MI, USA). N-(6-Aminohexyl)-5-chloro-1-naphthalene)sulfonamidehydrochloride (w-7) and ryanodine were purchased from Tocris(Ellisville, MI, USA).

Stock solutions were prepared as follows: noradrenaline andisoprenaline in catecholamine diluent (mM: NaCl 154, NaH2PO4

1.2 and ascorbic acid 0.2), PDA and 2-APB in DMSO and nifedipine

and ryanodine in absolute ethanol. All other drugs were dissolvedand diluted to required concentrations in distilled water.

2.4. Measurement and analysis of data

All data analyses were carried out using GraphPad Prism(version 4.00). Results are expressed as the mean7standard errorof the mean (SEM). The value of n represents the number ofexperimental animals used. Curves were analysed by repeatedmeasures ANOVA and P-values indicate the probabilities of asignificant interaction between frequency or concentration andtreatment. Values of Po0.05 were considered significant.

2.5. Chemistry

1H NMR spectra were recorded at 300 MHz, respectively, using aBruker Avance DPX 300 spectrometer equipped with a SiliconGraphics workstation. Chemical shifts for all 1H spectra are reportedin ppm, in either d-chloroform, d4-methanol, or d6-DMSO, andreferenced to residual proteo-solvent. Low-resolution mass spectro-metry analyses were performed using a Micromass Platform II singlequadrupole mass spectrometer equipped with an electrosprayionization/atmospheric pressure chemical ionization (ESI/APCI) ionsource. Sample management was facilitated by an Agilent 1100series HPLC system. In reporting spectroscopic data, the followingabbreviations have been used: s, singlet; d, doublet; t, triplet;q, quartet; p, pentet; m, multiplet; br, broad; app, apparent. Thinlayer chromatography (TLC) was carried out on silica gel 60 F254 pre-coated plates (0.25 mm, Merck, ART. 5554). Flash column chromato-graphy was carried out using Merck Silica gel 60, 230–400 meshASTM, in a glass column of diameter 3 cm and silica length of 15 cm.All solvents were reagent grade and used as received.

2.5.1. ChromatographySaw palmetto extract (1.14 g) was diluted with hexane (10 ml).

The diluted extract was then applied to the top of a normal-phasechromatography column, and eluted with 200 ml of the followingsolvents in order: hexane, chloroform, ethyl acetate and methanol.Solvents were evaporated to dryness.

3. Results

3.1. Effects of saw palmetto on rat prostate smoothmuscle contractility

Trains of electrical field stimulation (0.5 ms, 60 V, 0.1–20 Hz)evoked frequency-dependent contractions, which were reprodu-cible over the time course of the experimental protocol and notaffected by vehicle. Incubation of isolated prostates with sawpalmetto extract (1.0 mg/ml) attenuated responses to electricalfield stimulation (Po0.05, n¼4) but incubation of prostatewith extract at a lower concentration (0.1 mg/ml) had no effect(P¼0.23, n¼4) (Fig. 1).

Table 1Chemical composition of Sabalselects. Source: Indena SpA. Composition in fatty acids, alcohols and sterols elucidated by a validated GLC method.

Saturated fatty acids (59.8%) Unsaturated fatty acids (33.7%) Fatty alcohols (0.20%) Sterols (0.32%)

Caproic acid (1.5%) Oleic acid (28.5%) Hexacosanol (0.017%) Campesterol (0.07%)Caprylic acid (2.3%) Linoleic acid (4.6%) Octacosanol (0.146%) Stigmasterol (0.03%)Capric acid (2.5%) Linolenic acid (0.6%) Tetracosanol (0.004%) β-Sitosterol (0.22%)Lauric acid (30.2%) Triacontanol (0.033%)Myristic acid (12.0%)Palmitic acid (9.5%)Stearic acid (1.8%)

T. Chua et al. / Journal of Ethnopharmacology 152 (2014) 283–291 285

3.2. Chromatographic fractions

Saw palmetto extract (1.14 g) was fractionated by normal-phasechromatography on silica gel to give four fractions: hexane(160 mg), chloroform (10 mg), ethyl acetate (1.07 g) and methanol(20 mg). Apart from solvent peaks, the NMR spectrum of thehexane fraction showed no significant signals and was thereforenot bioassayed. Both chloroform (P¼0.41, n¼6, Fig. 2A) andmethanol (P¼0.95, n¼6, Fig. 2C) fractions did not have anypharmacological effect on the isolated prostate tissues. In contrast,the ethyl acetate fraction (Po0.05, n¼6, Fig. 2B) attenuatedresponses to electrical field stimulation to an extent similar tothat produced by the crude extract. The NMR spectrum of theethyl acetate fraction showed peaks corresponding to saturatedand unsaturated fatty acids and triglycerides as the major compo-nents (Fig. 3), consistent with the composition given in Table 1.

3.3. Pharmacological assay of commercially available fatty acids

Commercially available fatty acids (lauric acid, oleic acid,linoleic acid and myristic acid) were shown to attenuate electricalfield stimulation induced contractions of isolated preparationsof rat prostate (Fig. 4). In addition, these inhibitory actions of thecommercially available fatty acids (0.01–10.0 mg/ml) and sawpalmetto extract (0.01–10.0 mg/ml) on electrical field stimulationinduced rat prostate contractions were concentration dependent(Fig. 5). The triglyceride, glyceryl trioleate did not have an effect onelectrical field stimulation induced contraction of the isolated ratprostate (results not shown).

3.4. Effects of saw palmetto extract on contractile responses inducedby noradrenaline, ATP, acetylcholine and KCl

Saw palmetto extract produced a concentration-dependentinhibition of isolated preparations of rat prostate tonic contrac-tions induced by noradrenaline (1 μM), ATP (1 mM), acetylcholine(10 μM) or KCl (80 mM) (Fig. 6). Contractions to these agonistswere not affected by vehicle over the same time course.

3.5. Mechanisms of action

3.5.1. β-Adrenoceptor antagonismPropranolol (1 μM) did not affect the response elicited by

electrical field stimulation of isolated rat prostate tissues. Further-more, saw palmetto extract (1.0 mg/ml), still reduced contractileresponses to electrical field stimulation compared to controltissues as well as in the presence of propranolol (Fig. 7).

3.5.2. Rho kinase inhibitionIncubation of isolated rat prostates with the Rho kinase inhibitor

Y27632 (10 mM) attenuated responses to electrical field stimulation.

Fig. 1. Mean contractile response to electrical field stimulation of isolated preparations of rat prostate to trains of electrical field stimulation (0.5 ms, 60 V, 0.1 to 20 Hz for 10pulses or 10 s) following the incubation of (A) saw palmetto extract (0.1 mg/mL) or (B) saw palmetto extract (1.0 mg/mL). Columns represent the mean force from fourprostate preparations. Error bars represent S.E.M. P-values represent the probability of a significant interaction between treatment and frequency (two-way repeatedmeasures ANOVA).

Fig. 2. Mean concentration response curves to electrical field stimulation ofisolated preparations of rat prostate to trains of electrical field stimulation(0.5 ms, 60 V, 0.1 to 20 Hz for 10 pulses or 10 s) following the incubation of(A) chloroform fraction (0.01 mg/mL), (B) ethyl acetate fraction (1.3 mg/mL) and(C) methanol fraction (0.02 mg/mL). Each column represents the mean force fromsix prostate preparations. Error bars represent S.E.M. P-values represent theprobability of a significant interaction between treatment and concentration (twoway repeated measures ANOVA).

T. Chua et al. / Journal of Ethnopharmacology 152 (2014) 283–291286

Contractions in response to electrical field stimulation in thepresence of Y27632 (10 mM) were further reduced in the presenceof saw palmetto extract (1.0 mg/ml) compared to those observed incontrol tissues and in the presence of Y27632 alone (Fig. 7).

3.5.3. PKC activationThe PKC activator (PDA 1.0 μM) had no effect on prostatic

contractions elicited by electrical field stimulation. Furthermore,saw palmetto extract (1.0 mg/ml) inhibited contractions of the ratprostate produced by electrical field stimulation (Fig. 7) in thepresence of PDA (1.0 mM). Similarly, PDA (1.0 μM) did not affectcontractions elicited by noradrenaline (1.0 nM–30 mM) or carba-chol (0.1 mM–1.0 mM) and did not affect the inhibition of nora-drenaline or carbachol induced contractions caused by sawpalmetto extract (1.0 mg/ml).

3.5.4. Calmodulin antagonismThe calmodulin antagonist w-7 (100 mM) attenuated responses to

electrical field stimulation in isolated preparations of rat prostate.

Saw palmetto extract (1.0 mg/ml) further reduced contractionscompared to those observed in control tissues and in the presenceof w-7 alone (Fig. 7).

3.5.5. Calcium channel antagonistsResponses to electrical field stimulation in the presence of the

Ca2þ channel blockers nifedipine (1.0 mM), 2-APB (60 mM) andryanodine (10 mM) were significantly reduced compared to thoseobserved in control tissues. Administration of saw palmettoextract (1.0 mg/ml) further inhibited the contractile response toelectrical field stimulation in the presence of each of these Ca2þ

channel blockers (Fig. 8).

4. Discussion

Saw palmetto extract is the most popular medicinal herb usedin the symptomatic treatment of BPH. The use of saw palmetto islargely based on its reputed anti-androgenic, anti-inflammatoryand anti-proliferative effects (Di Silverio et al., 1992; Iehle et al.,1995; Paubert-Braquet et al., 1997; Ravenna et al., 1996), which

Fig. 3. 1H NMR spectrum: Ethyl acetate fraction in d-chloroform.

Fig. 4. Mean contractile response to electrical field stimulation of isolated preparations of rat prostate to trains of electrical field stimulation (0.5 ms, 60 V, 0.1 to 20 Hz for 10pulses or 10 s) in the presence or absence of (A) lauric acid (1.0 mg/mL), (B) oleic acid (1.0 mg/mL), (C) linoleic acid (1.0 mg/mL) and (D) myristic acid (1.0 mg/mL). Barsrepresent the mean force from four prostate preparations. Error bars represent S.E.M. P-values represent the probability of a significant interaction between treatment andfrequency (two-way repeated measures ANOVA).

Fig. 5. Mean contractile response to electrical field stimulation (0.5 ms, 60 V, 10 Hzfor 10 s) of isolated preparations of rat prostate following the incubation of (●)vehicle, (■) lauric acid, (◆) oleic acid, ( ) myristic acid, ( ) linoleic acid and ( )saw palmetto extract. Each point represents the mean force from six prostatepreparations. Error bars represent S.E.M. P-values represent the probability of asignificant interaction between treatment and frequency (two-way repeatedmeasures ANOVA).

T. Chua et al. / Journal of Ethnopharmacology 152 (2014) 283–291 287

produce a decrease in prostatic size. This study investigated theeffects of a liposterolic saw palmetto extract on the contractility ofisolated rat prostate glands in vitro.

To deduce the bioactive chemical components of the lipostero-lic extract of saw palmetto, normal-phase chromatography wasconducted. The fractions were pharmacologically evaluated but

Fig. 6. Mean contractile response to (A) noradrenaline, (B) ATP, (C) acetylcholine and (D) KCl-induced contraction of isolated preparations of rat prostate in the presence ofvarious concentration of saw palmetto extract. Each bar represents the mean force from six prostate preparations. Error bars represent S.E.M. P-values represent theprobability of a significant interaction between treatment and frequency (two-way repeated measures ANOVA).

Fig. 7. Mean contractile response to electrical field stimulation of isolated preparations of rat prostate to trains of electrical field stimulation (0.5 ms, 60 V, 0.1 to 20 Hz for 10pulses or 10 s) in the presence and absence of the β-adrenoceptor antagonist propranolol (1 μM, upper left panel), the Rho kinase inhibitor Y27632 (10 μM, upper rightpanel), the PKC activator PDA (1 mM, lower left panel) or the calmodulin antagonist w-7 (100 mM, lower right panel) either alone or in combination with saw palmetto extract(SPE, 1.0 mg/ml). Columns represent the mean force from six prostate preparations. Error bars represent S.E.M. P-values represent the probability of a significant interactionbetween treatment and frequency (two-way repeated measures ANOVA).

T. Chua et al. / Journal of Ethnopharmacology 152 (2014) 283–291288

only the ethyl acetate fraction was found to attenuate electricalfield stimulation induced contractions of the rat prostate. The 1HNMR spectrum of the active fraction showed signals for saturatedand unsaturated fatty acid components consistent with paststudies and the manufacturer's certificate of analysis that sawpalmetto liposterolic extracts are composed mainly of fatty acidsof varying chain lengths (Breu et al., 1992). Studies have shownthat fatty acids, aliphatic alcohols and sterols present in theliposterolic extract can interact with cellular membranes, ionicchannel conductance or intracellular mediators (Ordway et al.,1991; Stasek et al., 1993; Zwijsen et al., 1992). These findingsprompted us to assess the inhibitory effect of fatty acids onelectrical field stimulation induced contractions.

The four major fatty acids present in the Sabaselects lipos-terolic extract of saw palmetto are lauric acid (30.2%), oleic acid(28.5%), myristic acid (12.0%) and linoleic acid (4.6%), present asthe free fatty acids and mixed triglycerides (�55:45). Commer-cially available samples of these fatty acids were pharmacologi-cally assayed and all were found to inhibit contraction in responseto electrical field stimulation. The commercially available fattyacids were found to be equipotent, producing IC50 values from 0.67

to 0.85 mg/ml which is similar to the IC50 value of the sawpalmetto extract (Fig. 5). A triglyceride (glyceryl trioleate) wasalso assayed and showed no effects. The similar IC50 values for thewhole extract and free fatty acids which are the major compo-nents, shows that the activity is principally due to the free fattyacids in the mixture, rather than very minor components, such assterols. Fatty acids have been shown to inhibit 5α-reductaseactivity and have receptor-binding activity at the α1-adrenoceptor,1,4-dihydropyridine Ca2þ channel and muscarinic receptors (Abeet al., 2009; Liang and Liao, 1992). These reports indicate that fattyacids present in the liposterolic extract of saw palmetto wouldindeed be efficacious in the symptomatic relief of symptomscaused by the dynamic component of BPH.

In assessing the mechanism of action, the blockade of smoothmuscle contraction caused by the saw palmetto extract does notappear to be dependent on a specific receptor mechanism sincesaw palmetto attenuated the response to noradrenaline, acetyl-choline, ATP and KCl. This lack of specificity suggests that sawpalmetto acts on some process which is common to the effects ofall of these agonists, most likely upon a separate inhibitoryreceptor mechanism or intracellular mediators which are down-stream from the receptor site.

It is generally accepted that the coupling of β-adrenoceptors toadenylyl cyclase elevates intracellular cAMP levels, which is believedto be the key trigger for eliciting β-adrenoceptor mediated smoothmuscle relaxation. Such β-adrenoceptor mechanisms have beenshown to be present in the rat prostate (Kalodimos and Ventura,2001). In our study, the liposterolic extract of saw palmetto con-tinued to have a relaxant effect in the presence of propranolol, whichindicates that the extract does not act through β-adrenoceptors.

The downstream target Rho kinase plays an important role inthe regulation of myosin light chain phosphatase activity (Rientoand Ridley, 2003). Rho kinase acts by phosphorylating the myosin-binding subunit of myosin light chain phosphatase hence inhibit-ing its activity. Due to this effect, myosin light chain cannot bedephosphorylated, promoting the contractile state of the smoothmuscle cell. In order to clarify whether or not the relaxant effectcould be due to the inhibition of this enzyme, the extract wasassayed in the presence of an inhibitor of Rho kinase. Y27632 is ahighly potent, selective Rho kinase inhibitor, which inducesrelaxation of smooth muscle by competing with the ATP-bindingsite on the enzyme. Y27632 has previously been shown to lowerblood pressure through relaxation of vascular smooth muscle(Chitaley et al., 2001; Uehata et al., 1997). In our study, Y27632also had prostate smooth muscle relaxant activity as demonstratedby its inhibition of contractions in response to electrical fieldstimulation. However, the liposterolic saw palmetto extract wasstill able to inhibit contractions in the presence of Y27632indicating that Rho kinase inhibition does not play an importantrole in the relaxant effect of the saw palmetto extract in theisolated rat prostate.

To investigate the possibility that saw palmetto acts throughPKC, we investigated the effect of PKC activation using thesynthetic phorbol ester PDA. PDA is known to enhance contractionof smooth muscle through mimicking diacylglycerides whichactivate PKC (Baraban et al., 1985; Wagner et al., 1987). ActivatedPKC has contraction-promoting effects such as phosphorylation ofL-type Ca2þ channels, receptor-operated Ca2þ channels, sarco-plasmic reticulum IP3 Ca2þ release channels, extrusion pumps inthe sarcolemma or sarcoplasmic reticular membrane as well asactivation of Naþ/Kþ exchange and other proteins that regulatecross-bridge cycling (Besterman and Cuatrecasas, 1984; Naka et al.,1983; Nishizuka, 1984; Rembold and Murphy, 1988). Surprisinglyin our hands, the addition of PDA had little effect on prostaticcontractions in response to electrical field stimulation or exogen-ous administration of noradrenaline or carbachol. This observed

Fig. 8. Mean contractile response of isolated preparations of rat prostate to trainsof electrical field stimulation (0.5 ms, 60 V, 0.1 to 20 Hz for 10 pulses or 10 s) in thepresence or absence of (A) the L-type Ca2þ channel blocker nifedipine (1.0 mM),(B) IP3 receptor antagonist 2-APB (60 mM) or (C) ryanodine (10 mM) either alone orin combination with saw palmetto extract (SPE, 1.0 mg/mL). Columns represent themean force from four prostate preparations. Error bars represent S.E.M. P-valuesrepresent the probability of a significant interaction between treatment andfrequency (two-way repeated measures ANOVA).

T. Chua et al. / Journal of Ethnopharmacology 152 (2014) 283–291 289

lack of effect of PDA alone on prostatic smooth muscle contractilitysuggests that PKC may not be an important intracellular messen-ger in prostatic smooth muscle cells. In addition, saw palmetto wasable to inhibit prostatic contractions in the presence of PDA, whichsuggests that inhibition of PKC is not involved in the observedrelaxant effect of liposterolic saw palmetto extract.

Modification of the relaxing effect of the liposterolic sawpalmetto extract by the Ca2þ channel inhibitors nifedipine, 2-APB and ryanodine was also investigated. Smooth muscle contrac-tion depends upon the concentration of Ca2þ ions free within thecytoplasm of the cell (Bolton, 1979; Filo et al., 1965; van Breemenand Saida, 1989). Ca2þ mediates smooth muscle contraction,through the formation of a Ca2þ-calmodulin complex which inturn stimulates phosphorylation of the myosin light chain. It iscommonly accepted that the source of smooth muscle cytosolicCa2þ comes from the extracellular space through voltage-operatedand receptor-operated Ca2þ channels, as well as from intracellularstores (sarcoplasmic reticulum) (Hurwitz, 1986). The mechanismfor Ca2þ extrusion from the sarcoplasmic reticulum into thecytosol involves primarily IP3 and ryanodine sensitive Ca2þ releasechannels (Chadwick et al., 1990). Incubation of prostates withnifedipine, 2-APB or ryanodine resulted in a depression of smoothmuscle contraction in response to electrical field stimulation. Thisindicates that all three of these channels are involved in the influxof Ca2þ into the cytosol from extracellular and intracellularsources. However, saw palmetto was still able to relax prostatecontractions in the presence of nifedipine, 2-APB and ryanodine,indicating that modulation of other mechanisms is more likely tobe responsible for saw palmetto's inhibitory effect.

As mentioned earlier, the contraction of smooth muscle isinitiated by the formation of Ca2þ-calmodulin. This complexactivates myosin light chain kinase (MLCK) to phosphorylate thelight chain of myosin, subsequently allowing cross-bridge cyclingto occur between actin and myosin. We explored whether therelaxing effect could be a consequence of inhibition of the serine/threonine-specific protein kinase MLCK using the calmodulininhibitor, w-7. As expected, w-7 inhibited contractile responsesto electrical field stimulation; however, the liposterolic sawpalmetto extract still inhibited electrical field stimulation-induced contractions in its presence. These results suggest thatthe liposterolic extract of saw palmetto does not act through MLCKinhibition.

Taken together our results suggest that the liposterolic extractof saw palmetto may provide therapeutic relief against urinarysymptoms associated with BPH by relaxing prostatic smoothmuscle through a non-specific mechanism. Indeed, the active fattyacids present in the liposterolic extract may have subtle effects onmultiple mechanisms which cannot be isolated but add up to havea substantial smooth muscle relaxant effect. However, it seemsunlikely that the recommended daily dose of this extract (320 mg)would translate to a concentration in the prostate of 1 mg/mlwhich would be needed to produce adequate smooth musclerelaxation. Nevertheless, such a mechanism is likely to alsoproduce relaxation of other smooth muscles as has been describedpreviously (Gutierrez et al., 1996). This lack of specificity is likely toproduce a multitude of side effects unless the extract is selectivelydistributed to the prostate gland, as has been shown for the α1-adrenoceptor antagonist tamsulosin (Korstanje et al., 2011) whichis used to treat urinary symptoms associated with BPH.

Acknowledgements

This work was supported in part by funding from Indena S.p.A.manufacturers of Sabalselects, an original CO2 Serenoa repensfruit extract and Pathway International, Australian suppliers ofSabalselects.

References

Abe, M., Ito, Y., Suzuki, A., Onoue, S., Noguchi, H., Yamada, S., 2009. Isolation andpharmacological characterization of fatty acids from saw palmetto extract.Anal. Sci. 25, 553–557.

Baraban, J.M., Gould, R.J., Peroutka, S.J., Snyder, S.H., 1985. Phorbol ester effects onneurotransmission: interaction with neurotransmitters and calcium in smoothmuscle. Proc. Nat. Acad. Sci. U.S.A. 82, 604–607.

Besterman, J.M., Cuatrecasas, P., 1984. Phorbol esters rapidly stimulate amiloride-sensitive Naþ/Hþ exchange in a human leukemic cell line. J. Cell Biol. 99,340–343.

Bolton, T.B., 1979. Mechanisms of action of transmitters and other substances onsmooth muscle. Physiol. Rev. 59, 606–718.

Breu, W., Hagenlocher, M., Redl, K., Tittel, G., Stadler, F., Wagner, H., 1992. Anti-inflammatory activity of sabal fruit extracts prepared with supercritical carbondioxide. In vitro antagonists of cyclooxygenase and 5-lipoxygenase metabolism.Arzneimittelforschung 42, 547–551.

Cao, N., Haynes, J.M., Ventura, S., 2006. Saw palmetto is an indirectly actingsympathomimetic in the rat-isolated prostate gland. Prostate 66, 115–123.

Chadwick, C.C., Saito, A., Fleischer, S., 1990. Isolation and characterization of theinositol trisphosphate receptor from smooth muscle. Proc. Nat. Acad. Sci. U.S.A.87, 2132–2136.

Chitaley, K., Weber, D., Webb, R.C., 2001. RhoA/Rho-kinase, vascular changes, andhypertension. Curr. Hypertens. Rep. 3, 139–144.

Chua, T., Simpson, J.S., Ventura, S., 2011. Ethanol extracts of saw palmetto containthe indirectly acting sympathomimetic: tyramine. Prostate 71, 71–80.

Di Silverio, F., D'Eramo, G., Lubrano, C., Flammia, G.P., Sciarra, A., Palma, E.,Caponera, M., Sciarra, F., 1992. Evidence that Serenoa repens extract displaysan antiestrogenic activity in prostatic tissue of benign prostatic hypertrophypatients. Eur. Urol. 21, 309–314.

Filo, R.S., Bohr, D.F., Ruegg, J.C., 1965. Glycerinated skeletal and smooth muscle:calcium and magnesium dependence. Science 147, 1581–1583.

Gerber, G.S., 2000. Saw palmetto for the treatment of men with lower urinary tractsymptoms. J. Urol. 163, 1408–1412.

Gutierrez, M., Garcia de Boto, M.J., Cantabrana, B., Hidalgo, A., 1996. Mechanismsinvolved in the spasmolytic effect of extracts from Sabal serrulata fruit onsmooth muscle. Gen. Pharmacol. 27, 171–176.

Hurwitz, L., 1986. Pharmacology of calcium channels and smooth muscle. Annu.Rev. Pharmacol. Toxicol. 26, 225–258.

Iehle, C., Delos, S., Guirou, O., Tate, R., Raynaud, J.P., Martin, P.M., 1995. Humanprostatic steroid 5 alpha-reductase isoforms—a comparative study of selectiveinhibitors. J. Steroid Biochem. Mol. Biol. 54, 273–279.

Kalodimos, P.J., Ventura, S., 2001. Beta2-adrenoceptor-mediated inhibition of fieldstimulation induced contractile responses of the smooth muscle of the ratprostate gland. Eur. J. Pharmacol 431, 81–89.

Korstanje, C., Krauwinkel, W., van Doesum-Wolters, F.L., 2011. Tamsulosin shows ahigher unbound drug fraction in human prostate than in plasma: a basis foruroselectivity? Br. J. Clin. Pharmacol. 72, 218–225.

Liang, T., Liao, S., 1992. Inhibition of steroid 5 alpha-reductase by specific aliphaticunsaturated fatty acids. Biochem. J. 285 (2), 557–562.

Lowe, F.C., Ku, J.C., 1996. Phytotherapy in treatment of benign prostatic hyperplasia:a critical review. Urology 48, 12–20.

Miano, R., De Nunzio, C., Asimakopoulos, A.D., Germani, S., Tubaro, A., 2008.Treatment options for benign prostatic hyperplasia in older men. Med. Sci.Monit. 14, RA94–RA102.

Naka, M., Nishikawa, M., Adelstein, R.S., Hidaka, H., 1983. Phorbol ester-inducedactivation of human platelets is associated with protein kinase C phosphoryla-tion of myosin light chains. Nature 306, 490–492.

Nishizuka, Y., 1984. The role of protein kinase C in cell surface signal transductionand tumour promotion. Nature 308, 693–698.

Ordway, R.W., Singer, J.J., Walsh , J.V., 1991. Direct regulation of ion channels byfatty acids. Trends Neurosci. 14, 96–100.

Paubert-Braquet, M., Mencia Huerta, J.M., Cousse, H., Braquet, P., 1997. Effect of thelipidic lipidosterolic extract of Serenoa repens (Permixon) on the ionophoreA23187-stimulated production of leukotriene B4 (LTB4) from human polymor-phonuclear neutrophils. Prostaglandins Leukot. Essent. Fatty Acids 57, 299–304.

Plosker, G.L., Brogden, R.N., 1996. Serenoa repens (Permixon). A review of itspharmacology and therapeutic efficacy in benign prostatic hyperplasia. DrugsAging 9, 379–395.

Ravenna, L., Di Silverio, F., Russo, M.A., Salvatori, L., Morgante, E., Morrone, S.,Cardillo, M.R., Russo, A., Frati, L., Gulino, A., Petrangeli, E., 1996. Effects of thelipidosterolic extract of Serenoa repens (Permixon) on human prostatic celllines. Prostate 29, 219–230.

Rembold, C.M., Murphy, R.A., 1988. [Ca2þ]-dependent myosin phosphorylation inphorbol diester stimulated smooth muscle contraction. Am. J. Physiol 255,C719–723.

Riento, K., Ridley, A.J., 2003. Rocks: multifunctional kinases in cell behaviour. Nat.Rev. Mol. Cell Biol. 4, 446–456.

Stasek , J.E., Natarajan, V., Garcia, J.G., 1993. Phosphatidic acid directly activatesendothelial cell protein kinase C. Biochem. Biophys. Res. Commun. 191, 134–141.

Tarayre, J.P., Delhon, A., Lauressergues, H., Stenger, A., Barbara, M., Bru, M.,Villanova, G., Caillol, V., Aliaga, M., 1983. Anti-edematous action of a hexaneextract of the stone fruit of Serenoa repens Bartr. Ann. Pharm. Fr. 41, 559–570.

Uehata, M., Ishizaki, T., Satoh, H., Ono, T., Kawahara, T., Morishita, T., Tamakawa, H.,Yamagami, K., Inui, J., Maekawa, M., Narumiya, S., 1997. Calcium sensitization of

T. Chua et al. / Journal of Ethnopharmacology 152 (2014) 283–291290

smooth muscle mediated by a Rho-associated protein kinase in hypertension.Nature 389, 990–994.

van Breemen, C., Saida, K., 1989. Cellular mechanisms regulating [Ca2þ]i smoothmuscle. Annu. Rev. Physiol. 51, 315–329.

Wagner, B., Schachtele, C., Marme, D., 1987. Phorbol 12,13-dibutyrate-inducedcontraction of isolated rabbit vascular smooth muscle. Eur. J. Pharmacol. 140,227–232.

Wilt, T.J., Ishani, A., Rutks, I., MacDonald, R., 2000. Phytotherapy for benign prostatichyperplasia. Public Health Nutr. 3, 459–472.

Zwijsen, R.M., Oudenhoven, I.M., de Haan, L.H., 1992. Effects of choleste1ssrol andoxysterols on gap junctional communication between human smooth musclecells. Eur. J. Pharmacol. 228, 115–120.

T. Chua et al. / Journal of Ethnopharmacology 152 (2014) 283–291 291