Research Paper

Supplementation with Pfaffia glomerata (Sprengel) Pedersendoes not affect androgenic–anabolic parameters in male rats

Ney Felipe Fernandes a, Anderson Joel Martino-Andrade a,Ana Carolina dos Santos Lourenço b, Juliane Centeno Muller b,Katherinne Maria Spercoski a, Fabiola Nihi b, Marilis Dallarmi Miguel c,Vinícius Bednarczuk de Oliveira c, Paulo Roberto Dalsenter b, Rosana Nogueira Morais a,n

a Departamento de Fisiologia, Universidade Federal do Paraná, Centro Politécnico, 81531-980 Curitiba, PR, Brazilb Departamento de Farmacologia, Universidade Federal do Paraná, Centro Politécnico, 81531-980 Curitiba, PR, Brazilc Departamento de Farmácia, Universidade Federal do Paraná, 632, Avenida Prof. Lothário Meissner, 80210-170, Curitiba, PR, Brazil

a r t i c l e i n f o

Article history:Received 23 June 2014Received in revised form7 November 2014Accepted 30 November 2014Available online 9 December 2014

Keywords:Pfaffia glomerataTestosteroneRatsAndrogenic effects

a b s t r a c t

Ethnopharmacological relevance: Paffia spp (Amaranthacea) has a widespread use of in Brazil as a possiblehormonal supplement and a substitute of Panax ginseng, although information on its reproductive effectsis missing.Aim of the study: To evaluated possible anabolic–androgenic or anti-androgenic effects of Pfaffiaglomerata (PG) extract using intact eight-months-old male rats and pre-pubertal castrated rats.Materials and methods: Three different dose levels of PG (8.5, 30 and 85 mg/kg/day) were administeredto eight-months-old rats for 28 days or to castrated males for 7 days (Hershberger assay). In theexperiment with intact animals, 24 h fecal samples were collected for quantification of fecal metabolitesof androgens throughout treatment. At the end of the treatment period, animals were euthanized forevaluation of serum testosterone, reproductive organ weights, number of spermatids per testis, diameter ofseminiferous tubules and cross-sectional area of soleus muscle fibers. In the Hershberber assay, androgenic oranti-androgenic effects were evaluated by the weights of androgen-dependent tissues: ventral prostate,seminal vesicle, glans penis and levator ani muscle/bulbocavernosus muscle.Results: No effects were observed in the concentrations of fecal metabolites of androgens monitored duringthe treatment of intact eight-months-old rats. Moreover, at the end of treatment, no changes were seen inany of the investigated parameters. In the Hershberger assay, the PG extract did not induce androgenic oranti-androgenic effects at the dose levels tested. Significant effects were only observed in animals treatedwith testosterone and testosterone plus flutamide, which were used as positive controls for androgenicityand anti-androgenicity, respectively.Conclusions: At the dose levels tested, PG extract does not induce anabolic–androgenic or anti-androgeniceffects in rats.

& 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

The aging process in men encompasses slow progressive modifica-tions in sexual steroid levels, with a decrease in serum testosterone ofapproximately 1% per year after the age of 30. This reduction on and-rogens is part of the “andropause” or late onset hypogonadism (LOH)syndrome in ageing men, which includes alterations in sexual, phy-sical, and mental functions. Sexually related manifestations includereduced libido, erectile dysfunction, and decreased ejaculatory forceand volume (for review see Singh (2013) and Huhtaniemi (2014)).

Given the association between LOH and hazards of osteoporosis,cognitive decline, and metabolic dysfunction, testosterone therapyreplacement may be promising. However, there is still some con-troversy either in terms of the benefits or risks associated, since resultsfrom clinical trials are not conclusive (Singh, 2013; Huhtaniemi, 2014).One alternative to hormonal replacement is the use of phytoandro-gens, which could havemultiple clinical applications, such as being thedirect supply of androgens, stimulating androgen production or inter-fering with androgen metabolism. The use of vegetal drugs could bean affordable and safer alternative to elders (Edouard et al., 2014), alt-hough some products easily obtained from internet as “natural”herbal/hormonal dietary supplements, depending on the componentsof the mixture, might be associated with prostate cancer progression(Shariat et al., 2008).

Contents lists available at ScienceDirect

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

Journal of Ethnopharmacology

http://dx.doi.org/10.1016/j.jep.2014.11.0490378-8741/& 2014 Elsevier Ireland Ltd. All rights reserved.

n Corresponding author. Tel.: þ55 41 33611719; fax: þ55 41 33611714.E-mail address: moraisrn@ufpr.br (R.N. Morais).

Journal of Ethnopharmacology 161 (2015) 46–52

There are several medicinal herbs historically used worldwide astonic and aphrodisiacs, with improvement of sexual behavior andperformance, including at least 12 species also proved capable ofincreasing serum testosterone levels in experimental animals (forreview see Nantia et al. (2009), Mendes (2011), Kotta et al. (2013),Chauhan et al. (2014) and Edouard et al. (2014)). However, possiblemechanisms of action as well as therapeutic applications are stillnot clear, even for the most studied species, as for example, Tribulusterrestris (TT) L. (Zygophyllaceae), a well-patronized medicinal herbby Chinese and Indian medicine (Arsyad, 1996; Gauthaman et al.,2002, 2003; Neychev and Mitev, 2005; Gauthaman and Ganesan,2008; Saudan et al., 2008; Martino-Andrade et al., 2010).

In Brazil, many plants are used as tonic, fortifier, aphrodisiac,and anti-aging medicines (Mendes, 2011). Pfaffia paniculata (Mar-tius) Kuntze and Pfaffia glomerata (Sprengel) Pedersen, both fromthe Amaranthaceae family, are among the most commercializedmedicinal herbs of this gender and both are popularly called“Brazilian ginseng” (Pinello et al., 2006; Gosmann et al., 2003).The roots of these species are largely used in folk medicine, butthere is incomplete data about their pharmacological effectivenessand safety. They have been popularly indicated for several usesincluding as anti-carcinogenic, tonic, invigorating, aphrodisiac,anti-inflammatory, and anti-anemic agents (De Oliveira, 1986;Silva et al., 2005; Pinello et al., 2006; Carneiro et al., 2007). Oneparticular aspect related to the commercialization of the “Brazilianginseng” is that it may be sold as a less expensive substitute ofKorean ginseng (Panax ginseng, C. A. Mayer, Araliaceae), wellrecognized as a tonic and aphrodisiac (Edouard et al., 2014), oreven being part of adulterations of products sold as Panax ginseng(Mendes, 2011). Chemical constituents from roots of Pfaffia speciesinclude allantoin, ecdysteroids, pfaffic acid and their glycosides(nortriterpene saponins), stigmasterol and sitosterol (Gosmannet al., 2003; Zimmer et al., 2006; Nakamura et al., 2010). Ecdyster-one is only found in Pfaffia glomerata (Shiobara et al., 1993) and isused as criteria of identification for the species. Probably due totheir content of saponins and sitosterol, Pfaffia spp are expected tobe as effective on stimulating sexual function as Panax ginseng(Malini and Vanithakumari, 1992; Furukawa et al., 2006; Yue et al.,2007). However, there is very limited data on the reproductiveeffects of Pfaffia spp. In rats, Pfaffia paniculata did not increase theactivity of sexually potent animals, but normalized sexual motiva-tion and performance of sexually sluggish/impotent individuals(Arletti et al., 1999). Oshima and Gu (2003) investigated the effectsof Pfaffia paniculata root powder on the concentration of sexualhormones in mice. After 30 days of supplementation, highertestosterone levels were detected in males treated with Pfaffiapaniculata when compared to controls. For Pfaffia glomerata thereare only data on its positive effects on both learning and memoryof old mice supplemented for 150 days (Marques et al., 2004).

In addition to saponins, Pfaffia glomerata presents approximately1% of ecdysterone, a substance that is supposed to have ergogenicproperties (Tóth et al., 2008). In vivo studies with ecdysteroids inmammals are difficult to be made, because of their fast catabolismand elimination (Kizelsztein et al., 2009) and the few availableresults remain quite controversial. Ecdysteroids are a large family ofpolyhydroxylated steroids produced by invertebrates, but also bysome plant species (phytoecdysteroids). In rats, 20-hydroxyecdy-sone, an ecdysteroid extracted from Silene viridiflora, promoted anincrease in the size of muscular fibers, after local subcutaneousapplication of 5 mg/kg for 7 days (Tóth et al., 2008). Based on theseresults the authors suggested that 20-hydroxyecdysone could beused as an alternative to synthetic anabolic steroids in therapiesagainst muscle atrophy—like the ones that occur naturally throughthe aging process (Lynch et al., 2007). On the other hand, in thestudy made by Wilborn et al. (2006) with men treated with a dailydose of 200 mg ecdisterone, no effects were observed on maximum

power, fat levels or serum testosterone. Marques et al. (2002), in adouble-blind clinical study controlled with placebo, evaluated thetonic effects in physical activity attributed to the roots of Pfaffiaglomerata and found no benefits to the volunteers in overall apt-itude and physical activity.

Due to the widespread use of Paffia glomerata as a substitute ofPanax ginseng in Brazil and the scarce information on its possiblereproductive effects, in the present study we investigated the effectsof Pfaffia glomerata dry extract on reproductive and androgen-dependent tissues of adult 8-months-old male rats, as well as pos-sible androgenic and anti-androgenic activities of this plant in theHershberger assay. Data generated here could help to understandthe possible benefits and risks of using this herbal medicine.

2. Materials and methods

2.1. Pfaffia glomerata dry extract

The dry ethanolic extract of roots of Pfaffia glomerata (Sprengel)Pedersen, Amaranthaceae (Brazilian Ginseng) was kindly donated bya local pharmaceutical company (Herbarium Laboratório Botânico,Colombo, Brazil) and contained 0.96% of ecdisterone, according to thetechnical authentication chromatography provided by the company.We also did run a phytochemical analysis of the dry extract of Pfaffiaglomerata by various methods for identification of compounds, suchas phenolic compounds tannins (Riffault et al., 2014), and flavonoids(Wagner et al., 1983). Other secondary metabolites were testedincluding saponins (Mousli and Tazerouti, 2007), steroids and alka-loids (Reich and Schibli, 2006). For the polyphenols, TLC was usedand eluted with a mixture of ethyl acetate/HCOOH/CH3COOH/H2O(100:11:11:27) acetate. Neu reagent (1 g of boric acid diphenyl etherethylamino in 100 mL of MeOH) was sprayed, and after the TLC wasdried, visualization of the spots was done at 366 nm. A fingerprintHPLC analysis of the extract was also performed using a highperformance liquid chromatography (HPLC Merck-Hitachi) equippedwith a pump (L-2130), UV–vis detector (DAD L- 2450), rheodynemanual injector (loop 20 μL). The extract was dissolved in the elutingsolvent and the solution was filtered with Millipore membrane(0.45 mM pore diameter). The samples (10 mg/mL) were eluted usingcolumn and pre column Waters Xterra reverse phase column C185 mm (4.6�250mm2). Mobile phase A: acetonitrile, B: water, gradi-ent elution: 0–30 min: 20–80% A, 30–35 min: 95% A, 35–38 min:20–80% A, 38–40min, 20% A. It was employed a flow rate of1 ml/min, an injection volume of 20 mL and the peak was detectedat 242 nm (Zimmer et al., 2006).

2.2. Animals

Two sets of experiments were conducted, one experiment withintact and the other with castrated male rats (Hershberger assay).For the experiment with intact animals, 8-months-old male Wistarrats were obtained from the breeding stock of the animal facility ofthe Universidade Federal do Paraná. At this age, animals are nolonger used for breeding as a result of lower reproductive perfo-rmance, i.e., reduced number of impregnated females and/or littersizes. Animals were housed individually in cages (33 cm of width,40 cm of depth and 16,5 cm of height) divided by a metal partition(one animal in each cage compartment), under 12 h dark/lightcycle and controlled temperature (2272 1C).

For the Hershberger assay, 7-weeks-old male Wistar rats weobtained from the animal facility of the Universidade Federal doParaná and castrated via the scrotum (midline incision) under fullanesthesia (ketamine 75 mg/kg and xylazine 1.5 mg/kg, both viaintraperitoneal injection). After surgery, animals were subcutaneously

N.F. Fernandes et al. / Journal of Ethnopharmacology 161 (2015) 46–52 47

injected (0.1 mL/100 g of body weight) with Agropen L.A. (Virbac,France: penicilin G procain 10.000.000 UI; penicilin G benzathine10.000.000 UI; dihydroestreptomycin sulfate 20 g; vehicle 100 mL)and Dipyrone (66 mg/kg). The treatments were started 7 days aftersurgery in order to allow the complete recovery of the animals.

In both experiments, animals received tap water and standardpellet food Nuvital CR1 (Nuvilab, Colombo, Brazil) ad libitum. Allexperimental procedures were approved by the Committee onAnimal Research and Ethics of the Universidade Federal do Paraná(Consent Number 513), which is in accordance with national andinternational Guidelines of animal welfare.

2.3. Evaluation of Pfaffia glomerata extract in intact male ratstreated during 28 days

2.3.1. TreatmentThe 8-months-old animals were treated during 28 days by oral

route. The Pfaffia glomerata (PG) dry extract was dissolved in distilledwater (vehicle) and administered by gavage at a dosing volume of5 mL/kg body weight. Animals were randomly distributed into fourtreatment groups (n¼12/group): control, which received distilledwater (vehicle), and three doses of Pfaffia glomerata dry extract,8.5 mg/kg/day (PG 8.5), 30 mg/kg/day (PG 30) and 85 mg/kg/day(PG 85). The lowest dose (8.5 mg/kg/dia) corresponds, approxi-mately, to the usual and prescribed dose to humans (approximately600 mg/day for a 70 kg person). The 30 mg/kg dose corresponds tothe extrapolated “therapeutic dose” for rats, calculated by allometry(Nevill, 1994), and the highest dose (85 mg/kg/day) corresponds toten times the human prescribed dose.

All animals were treated in the morning (8–10 h) for 28 days.24 h after the last administration, animals were euthanized bydecapitation for collection of blood and organs.

2.3.2. Organ weightsThe absolute and relative (organ weight/body weight�100) wei-

ghts of reproductive and other androgen-sensitive organs and tissueswere recorded: testes, epididymides, ventral prostate, seminal vesicle(with coagulating glands), and levator ani/bulbocavernosus muscle(LABC). The ventral prostate was weighed without the prostaticcapsule and seminal vesicle without its contents. The absolute andrelative weights of liver and kidneys were also recorded.

2.3.3. Extraction and quantification of fecal androgens24 h fecal samples were collected from cage bedding on days 1,

7, 14 and 28 of treatment and frozen at �20 1C until processing.Feces were dried out at 60 1C for 24 h.

Fecal steroids were extracted according to the procedurepreviously described by Touma and Palme (2005). Briefly, an ali-quot of approximately 0.5 g of the dried fecal sample was placedinto a glass tube containing 5.0 mL of 80% ethanol and vigorouslyshaken using Multi-Pulse vortexer (Glas-Col, Terre Haute, IN) for30 min. Each sample was centrifuged at 1000� g for 15 min, andthe supernatant recovered and diluted at 1:1 ratio with a phos-phate buffered solution.

Fecal extracts were analyzed for metabolites of androgens byenzyme immunoassay (Munro et al., 1991). The antibody againsttestosterone (R1556/7) and the respective HRP-conjugate wereobtained from Coralie Munro at the University of California (Davis,CA, USA). Serial dilutions of pooled fecal extracts produced displa-cement curves parallel to those of standards. The intra- and inter-assay coefficients of variation were less than 10%. Data were corr-ected for dilution and are expressed as ng/g of feces.

2.3.4. Serum testosteroneOne day after the final treatment, blood was collected by

decapitation and clotted at room temperature. After centrifugation,serum samples were transferred to clean tubes and kept at �20 1C.Serum testosterone was quantified by enzyme immunoassay asdescribed in section 2.3.3, but without extraction.

2.3.5. Spermatids countingAt the end of the treatment period, the left testes were used for

quantification of homogenization-resistant spermatids per testis,which correspond to spermatids in stages 17 to 19 of development.After removal of the tunica albuginea, each testis was minced andhomogenized for one minute in 10 mL 0.9% NaCl containing 0.5%Triton X-100 at medium speed in a Potter S Tissuemizer (Sartorius,Goettingen, Germany). The homogenate was diluted 10 times in 0.9%NaCl for microscopic counting of the number of homogenization-resistant spermatids in a Bürker haemocytometer chamber (BrandGmbH, Wertheim, Germany).

2.3.6. Testicular histologyThe right testes were immersed-fixed in ALFAC fixative (80%

ethanol, 10% formaldehyde and 5% glacial acetic acid) for approxi-mately 2 h. The capsule of each testis was cut at the poles to allowpenetration of the fixative. After this preliminary fixation, testeswere cut transversally to the long axis in 2 or 3 slices, which wereplaced again into the fixative for additional 24 h (Russell et al.,1990). Following fixation, testes were placed in 70% ethanol untilroutinely processed and embedded in paraffin. Sections of 5 mmthickness were obtained, placed onto glass slides and stained withhematoxilin and eosin.

The diameter of seminiferous tubules was measured in 20 roundtubular sections per animal at 100� magnification using the 3.0 ver-sion of UTHSCSA ImageTool software (UTHSCSA, San Antonio, TX).Images were acquired on a Zweiss Axiophot microscope (Oberkochen,Germany) coupled with an imaging system (ASI, Vista,CA) at 100�magnification.

2.3.7. Cross-sectional area of soleus muscle fibersTo evaluate the possible anabolic effects of the Pfaffia glomerata

treatment, the soleus muscle from the right hind limb was coll-ected and fixed in ALFAC solution for 24 h. After routine histologicalprocessing, muscles were transversally cut at 5 mm and sta-ined with hematoxilin and eosin. The total cross-sectional area ofhundred muscle fibers was measured in randomly selected fields ineach animal at a magnification of 100� . Analyses were performedwith the 3.0 version of the UTHSCSA Image Tool software(UTHSCSA, San Antonio, TX). Images were acquired on a ZweissAxiophot microscope (Oberkochen, Germany) coupled with animaging system (ASI, Vista, CA).

2.4. Test of androgenicity and anti-androgenicity with castratedmale rats—Hershberger assay

The Hershberger assay is based on the assessment of androgen-dependent tissues growth in castrated male rats (androgenicity) aswell as on blockage of testosterone trophic effects (anti-andro-genicity) in these organs (Owens et al., 2006).

Castrated male rats were treated daily for 7 consecutive dayswith different substances (n¼9 animals/group). For evaluation of apossible androgenic action of Pfaffia glomerata, castrated rats weretreated with three dose levels of the dry extract (8.5, 30 and85 mg/kg/day) by gavage and additional subcutaneous injectionsof oily vehicle (canola oil) (Owens et al., 2006). The positivecontrol group was subcutaneously treated with testosteronepropionate (CAS no. 57-85-2; Sigma-Aldrich, Steinheim, Germany)

N.F. Fernandes et al. / Journal of Ethnopharmacology 161 (2015) 46–5248

at 0.25 mg/kg/day and distilled water by gavage, while negativecontrols received distilled water (gavage) and canola oil (subcuta-neous injections). In order to test a possible anti-androgenic activity ofPfaffia glomerata, a group of castrated was treated subcutaneouslywith 0.25 mg/kg/day of testosterone propionate and the highest doseof the dry extract (85 mg/kg/day) by gavage. The positive controlgroup for anti-androgenicity was treated by gavage with flutamide(Galena Laboratory, Curitiba, Brazil), which was dissolved in distilledwater and administered at 5.0 mg/kg/day, and subcutaneous injec-tions of testosterone propionate (0.25 mg/kg/day). The dosing volumefor all solutions was 5.0 mL/kg body weight when using the oral routeand 1.0 mL/kg when using subcutaneous injections. 24 h after thefinal administration, animals were weighed and decapitated. Theventral prostate, seminal vesicle (with coagulating glands), glanspenis, levator ani/bulbocavernosus muscle (LABC), liver and kidneyswere removed and dissected carefully to remove the surrounding fatand connective tissue. Organ weights were measured and expressedas absolute (g) and relatives weights (organ weight�100/bodyweight). The ventral prostate was weighed without the prostaticcapsule and seminal vesicle without its contents.

2.5. Statistical analysis

Data were analyzed using GraphPad Prism 5.0 (GraphPad Soft-ware Inc, La Jolla, CA). Normality and homogeneity of varianceswere evaluated prior to statistical analysis. Data were analyzed byanalysis of variance (ANOVA) followed by Tukey multiple compar-ison test. For the analysis of fecal androgenic metabolites, repeated-measures-ANOVA was performed. Differences were considered tobe statistically significant at a probability level of 5% (po0.05).

3. Results

3.1. Phytochemical analysis of the dry extract of Pfaffia glomerata

Phytochemical screening confirmed the extract was positive forsaponins, flavonoids and steroids. These results confirm other studiesthat identified these compounds in the extract of Pfaffia glomerata(Felipe et al., 2014). The HPLC chromatographic fingerprint of Pfaffiaglomerata monitored at 242 nm (Fig. 1) showed the presence of threemajor peaks. Monitoring at 242 nm is characteristic for this species,and the peak 1, showing maximum absorbance at this wavelengthcould be ecdysterone, as identified by Zimmer et al. (2006), andpreviously confirmed through chromatography by the company thatkindly provided us the dry extract. Peaks 2 and 3 have yet to beidentified.

3.2. Evaluation of Pfaffia glomerata extract in intact male ratstreated during 28 days

3.2.1. Organ weightsAbsolute and relative organ weights are shown in Table 1.

Treatment of male rats with different doses of Pfaffia glomerata dryextract did not induce any significant change in the weight of rep-roductive or other androgen-sensitive organs, when compared tocontrol group. Likewise, no differences were detected for liver andkidney weights.

3.2.2. Fecal and serum androgensThe results of the fecal and serum androgen levels are sum-

marized in Figs. 2 and 3, respectively. The concentrations of fecalmetabolites of androgens in treated groups were not significantlydifferent from that of control at any time point investigated. Like-wise, no significant differences were observed between treated

and control groups for the concentration of serum testosterone atthe end of treatment.

3.2.3. Number of spermatidsThe number of homogeneization-resistant spermatids per testis

is shown in Fig. 4. Treatment with dry extract of Pfaffia glomerataproduced no significant effects on the number of spermatids, whencompared to control group.

Fig. 1. HPLC fingerprint of extract of Pfaffia glomerata, detected at 242 nm. Peak 1:(time 05.86 min.) 2: (time 09.78 min.) 3: (time 11.89 min).

Table 1Absolute and relative (%) organ weights of intact male rats treated during 28 dayswith distilled water (control), Pfaffia glomerata 8.5 mg/kg (PG 8.5 mg/kg), Pfaffiaglomerata 30 mg/kg (PG 30 mg/kg) or Pfaffia glomerata 85 mg/kg (PG 85 mg/kg).

Parameters Groups

Control PG 8.5 mg/kg PG 30 mg/kg PG 85 mg/kg

Prostate (g) 0.5170.16 0.5670.14 0.4770.18 0.5070.08Prostate (%) 0.1270.07 0.1270.03 0.1070.04 0.1170.02Testis (g) 1.7770.17 1.7870.09 1.8070.16 1.8770.15Testis (%) 0.3770.05 0.3970.05 0.3870.05 0.4270.05Epididymis (g) 0.6570.06 0.6470.05 0.6170.06 0.6470.04Epididymis (%) 0.1470.02 0.1470.01 0.1470.01 0.1470.01Kidney (g) 1.4870.13 1.4070.13 1.4170.14 1.3670.14Kidney (%) 0.3170.03 0.3070.03 0.3070.03 0.3070.03Liver (g) 15.472.54 15.172.68 14.972.38 14.071.46Liver (%) 3.1970.51 3.2470.35 3.1470.40 3.0970.20Seminal vesicle (g) 0.8370.12 0.7970.19 0.7570.13 0.7770.14Seminal vesicle (%) 0.1770.03 0.1770.05 0.1670.03 0.1770.04LABC (g) 1.3770.24 1.3070.24 1.2570.22 1.3270.22LABC (%) 0.2970.06 0.2870.05 0.2770.05 0.2970.05

n¼12 animals/group; LABC¼Levator ani/bulbocavernosus muscle.

Fig. 2. Androgenic metabolites analyzed in 24-h fecal samples at different timepoints throughout treatment. Animals were treated during 28 days with vehicle(control) or different doses of Pfaffia glomerata (PG) extract. There were nosignificant differences between groups; n¼12 animals/group.

N.F. Fernandes et al. / Journal of Ethnopharmacology 161 (2015) 46–52 49

3.2.4. Testicular and skeletal muscle histologyHistological evaluation of testicular samples revealed no sig-

nificant differences in the mean diameter of seminiferous tubules(Fig. 5). In the same way, analysis of cross-sectional area of soleusmuscular fibers from treated and control animals indicated no sig-nificant differences between groups (Fig. 6).

3.3. Test of androgenicity and anti-androgenicity with castratedmale rats—Hershberger assay

Results of the Hershberger assay are shown in Table 2. Treatmentof castrated rats with testosterone propionate significantly increasedthe absolute and relative weight of ventral prostate, seminal vesicle,LABC and glans penis, when compared to control (vehicle only)group. However, treatment with dry extract of Pfaffia glomeratainduced no changes in organ weights when compared to vehiclecontrol. Animals treated with testosterone and flutamide, used as apositive control for anti-androgenicity, exhibited a significant reduc-tion in the weight of these organs and tissues when compared totestosterone-treated rats (po0.05). This anti-androgenic effect wasnot observed in the group treated with testosterone and the highestdose of the plant extract, which displayed similar organ weightsthan the group treated with testosterone.

4. Discussion

In addition to the use of medicinal herbs to treat several diseases,there has been increasing interest in the possible endocrine mod-ulating effects of plants and their derivatives for use in geriatric andsports medicine. In the present study, 8 months-old-rats were usedas an experimental model to evaluate the effects of Pfaffia glomerata(PG) supplementation in the concentrations of androgens, skeletalmuscle histology and endocrine sensitive organs and tissues. Accord-ing to data from the animal facility of the Universidade Federal doParaná, rats aged 8 months are already in a stage of reduced repr-oductive performance. To further evaluate possible endocrine mod-ulating effects of Pfaffia glomerata, the dry extract was also tested inpre-pubertal castrated male rats (Hershberger assay).

The concentration of fecal metabolites of androgens in intact ratswas evaluated on days 1, 7, 14 and 28 of treatment in 24-h fecalsamples collected from cage beddings. Overall, treatment with PGextract did not change the concentration of fecal androgen metabo-lites throughout treatment, when compared to controls. This non-invasive method has been previously used (Martino-Andrade et al.,2010) and presents several advantages over traditional hormonal

analysis in serum, such as possibility of measurement at differenttime points without the need of stressful procedures of bloodcollection, usually leading to small amounts of blood, or euthanasiaof large number of animals. In addition, this noninvasive approachhas the advantage of using 24-h fecal samples, therefore, reducingthe interference of hormonal pulsatility and circadian rhythms. Theabsence of effects in the concentrations of fecal androgens is further

Fig. 3. Serum testosterone concentration in male rats treated with three doses ofPfaffia glomerata (PG) or vehicle (control group) for 28 days. There were nosignificant differences between groups; n¼12 animals/group, except for PG 8.5(n¼11).

Fig. 4. Number of homogenization resistant spermatids in male rats treated during28 days with Pfaffia Glomerata (PG) or vehicle (control). There was no significantdifference between groups; n¼12 animals/group, except for PG 8.5 (n¼11).

Fig. 5. Seminiferous tubule diameter in rats after 28 days of treatment with PfaffiaGlomerata (PG) or vehicle (control). There were no significant differences betweengroups.

Fig. 6. Cross-sectional area of soleus muscular fibers from animals treated withPfaffia glomerata (PG) or vehicle (control) during 28 days. There were no significantdifferences between groups.

N.F. Fernandes et al. / Journal of Ethnopharmacology 161 (2015) 46–5250

corroborated by unchanged levels of serum testosterone at the end oftreatment, as well as the lack of alterations in the weight of androgensensitive tissues such as prostate, seminal vesicle, epididymis andlevator ani/bulbocavernosus muscle (LABC). These data does notcorroborate the findings for Pfaffia paniculata in mice (OSHIMA andGU, 2003). Considering that the different classes of phytoandrogenscould be classified as cognate or true phytoandrogens, phytoandro-gen-ogenic, and functional mimetics of androgens (Edouard et al.,2014), we could say that under our experimental conditions, PG wasnot either capable of eliciting an androgen response through bindingto androgen receptor (as a true androgen) or to stimulate androgenproduction (as a phytoandroen-ogenic). It is also interesting to rem-ark that even a proven aphrodisiac such as Panax ginseng does notstimulate androgen production, having several of its effects attrib-uted to the activation of endothelial nitric oxide synthase to releasenitric oxide through its main phytosterol (ginsenoside). This activa-tion would occur through a non-genomic pathway of androgen, est-rogen and progesterone receptors (Furukawa et al., 2006; Edouardet al., 2014).

Some medicinal herbs containing steroidal saponins, such asTribulus terrestris and Panax ginseng have been indicated to enhancespermatogenesis and fertility (Arsyad, 1996; Gauthaman et al., 2003;Park, 2007). In the present study, we tested if PG would also havesome effect on spermatogenesis, by measuring testicular weight andthe number of homogenization-resistant spermatids, which are themost advanced (mature) spermatids present in the testis, whichwere both unaffected by PG treatment. Mean diameter of seminifer-ous tubules, which is a parameter used to further evaluate thespermatogenic activity (Russell et al., 1990), was also unchanged.Taken together, these results indicate that, despite the presence ofsteroidal saponins in its composition, the selected doses of Pfaffiaglomerata extract are not able to induce changes in testicularfunction of aging rats supplemented during 28 days.

In addition to the androgenic evaluations, we also investigatedpossible anabolic effects of the PG extract by measuring the cross-sectional area of fibers from the soleus muscle in the intact8-months-old rats. The results obtained show that the oral admin-istration of PG extract does not change the size of soleus fibers at thedose levels tested, indicating that ecdysteroids and other possibleactive anabolic–androgenic components are either ineffective orpresent at concentrations that are unable to induce anabolic effectsunder the described experimental conditions. Previously, Tóth et al.(2008) reported that 5 mg/kg of 20-hydroxyecdysone (20E) admi-nistered as subcutaneous injection in the left thigh of rats for 7 dayswas able to increase cross-sectional area of soleus and extensordigitorum longus muscles. The effects of 20E were observed innormal and regenerating muscles, but more pronounced effects

were seen in treated (left) hindlimbs, supporting a major local effect.In our study, rats receiving the highest PG extract dose (85 mg/kg),were supplemented with �0.85 mg/kg of ecdysterone, based onthe �1% content of ecdysterone in the dry extract, which is muchlower than the dose of 5 mg/kg reported by Tóth et al. (2008). Also,the supplementation with PG extract for 6 months did not improveperformance on ergoespirometric tests or reached maximum load in38 young men (Marques et al., 2002).

In the present study, possible anabolic–androgenic effects of PGextract were also investigated in castrated male rats treated for7 days (Hershberger assay). The results obtained indicate that thethree dose levels tested were not able to stimulate the selectedandrogen-sensitive tissues: glans penis, ventral prostate, seminalvesicle and levator ani/bulbocavernosus muscle (LABC). The responseto PG administration did not differ from that of control rats receivingdistilled water (vehicle). In contrast, administration of testosteronepropionate significantly increased the weight of these tissues. Using asimilar Hershberger assay protocol, Wilson et al. (2002) demon-strated that the anabolic steroid trenbolone acetate was able toincrease the weight of LABC at doses lower than those needed toproduce changes in glans penis, seminal vesicle and prostate. Thisdifferentially anabolic effect, with lower androgenic potency on sexaccessory organs, may be related to the ability of 5-alpha-reductase,absent in LABC, to inactivate trenbolone in these accessory tissues. Incontrast, Wilson et al. (2002) also demonstrated that testosteronepropionate can produce similar anabolic–androgenic effects in alltissues evaluated in the Hershberger assay, as it can be converted intothe more potent androgen dihydrotestosterone (DHT) and stimulateboth testosterone and DHT-dependent tissues. The lack of anabolic–androgenic effects of PG in the Hershberger assay is in accordancewith the results obtained in the experiment with intact 8-months-old rats that also showed no effects in androgen sensitive tissues,body weight or skeletal muscle. In addition, the highest dose of PGextract was also investigated for a possible anti-androgenic effect.However, in contrast to flutamide, used as positive control, there wasno evidence of anti-androgenic effects of PG.

Overall, the obtained data indicate that PG has no direct andro-genic–anabolic effects under the experimental conditions tested. Theextract was also unable to produce changes in the concentration offecal metabolites of androgens throughout a 28-day period, as well asin the serum concentration of testosterone at the end of treatment,indicating that PG does not change the concentration of endogenousandrogens in intact aging rats. Moreover, testis weight, number ofspermatids and tubular diameter were also unchanged by PG tre-atment. It is important to mention that inclusion of a positive controlgroup in the experiment with intact male rats would strengthenour results, even though the parameters investigated have been

Table 2Absolute and relative (%) organ weights of castrated male rats treated during 7 days (Hershberger assay).

Parameter Treatment

Control (n¼9) PG 8.5 mg/kg(n¼10)

PG 30 mg/kg(n¼8)

PG 85 mg/kg(n¼9)

Testosterone(n¼9)

TestosteroneþFlutamide(n¼9)

TestosteroneþPG 85(n¼9)

Glans penis (g) 0.03970.006a 0.03870.013 a 0.03670.007 a 0.03670.009 a 0.06670.011b 0.03770.009a 0.06670.009b

Glans penis (%) 0.01770.005 a 0.01670.005 a 0.01670.003 a 0.01570.003 a 0.02670.006b 0.01670.004a 0.02770.005b

LABC (g) 0.17770.052 a 0.17970.037 a 0.15970.037 a 0.15770.028 a 0.47770.078b 0.16270.024a 0.45170.070b

LABC (%) 0.07570.016 a 0.07670.011 a 0.06670.010 a 0.06670.008 a 0.18770.028b 0.06870.010a 0.18570.027b

Prostate (g) 0.01170.004 a 0.01370.005 a 0.01070.001 a 0.01170.003 a 0.10970.030b 0.05270.053c 0.10270.022b

Prostate (%) 0.00470.002 a 0.00570.002 a 0.00470.001 a 0.00570.001 a 0.04370.011b 0.02170.021c 0.04270.012b

Seminal vesicle(g)

0.04370.018 a 0.04370.020 a 0.04670.014 a 0.03970.011 a 0.32070.058b 0.04370.014a 0.29470.056b

Seminal vesicle(%)

0.01870.007 a 0.01870.008 a 0.01970.005 a 0.01670.004 a 0.12670.029b 0.01870.006a 0.12270.030b

Different letter superscripts indicate significant differences between groups (po0.05; ANOVA/Tukey). LABC¼Levator ani/bulbocavernosus muscle. Control¼vehicle onlygroup (oral and subcutaneous route); PG¼Pfaffia glomerata (mg/kg/day) by gavage and vehicle (canola oil) by subcutaneous injection; Testosterone¼testosterone propionate0.25 mg/kg/day by subcutaneous injection and vehicle (distilled water) by gavage; flutamide¼5.0 mg/kg/day by gavage.

N.F. Fernandes et al. / Journal of Ethnopharmacology 161 (2015) 46–52 51

previously shown to sensitively respond to androgens in the intactrat (Korenchevsky et al., 1937). Overall, the combined results of theshort- (Hershberger) and long-term (28-days) protocols used in thepresent study provide some evidence for a lack of androgenic andanti-androgenic activity of Pfaffia glomerata at the dose levels tested.Taken together these results indicate that this plant may not beeffective for some of the traditional uses in folk medicine such as an“aphrodisiac” or “invigorating”, “tonic” agent. On the other hand, thepresent data show evidence for the absence of overall male repro-ductive and endocrine toxicity, although additional experimentalstudies using extended treatment periods would be necessary to fullycharacterize the efficacy and toxicity profile of this plant.

Acknowledgments

The authors are indebted to Herbarium Laboratório Botânico(Colombo, Brazil) for kindly providing the Pfaffia glomerata dryextract. Ney Felipe Fernandes was a recipient of a scholarship fromCAPES, Brazil.

References

Arletti, R., Benelli, A., Cavazzuti, E., Scarpetta, G., Bertolini, A., 1999. Stimulatingproperty of Turnera diffusa and Pfaffia paniculata extracts on the sexualbehaviour of male rats. Psychopharmacology 143, 15–19.

Arsyad, K.M., 1996. Effect of protodioscin on the quantity and quality of spermsfrom males with moderate idiopathic oligozoospermia. Medika 22, 614–618.

Carneiro, C.S., Costa-Pinto, F.A., da Silva, A.P., Pinello, K.C., da Silva, T.C., Matsuzaki, P.,Nagamine, M.K., Górniak, S.L., Haraguchi, M., Akisue, G., Dagli, M.L., 2007. Pfaffiapaniculata (Brazilian ginseng) methanolic extract reduces angiogenesis in mice.Experimental and Toxicologic Pathology 58, 427–431.

Chauhan, N.S., Sharma, V., Dixit, V.K., Thakur, M., 2014. A review on plants used forimprovement of sexual performance and virility. BioMed Research Interna-tional, 19. http://dx.doi.org/10.1155/2014/868062 (Article ID 868062).

De Oliveira, F., 1986. Pfaffia paniculata(Martius) Kuntze: the Brazilian ginseng.Revista Brasileira de Farmacognosia 1, 86–92.

Edouard, M.J., Miao, L., Fan, G.W., Ojong, B.B., Zhen, H., Zhang, J., Gao, X.M., Zhu, Y.,2014. Yang-tonifying traditional Chinese medicinal plants and their potentialphytoandrogenic activity. Chinese Journal of Natural Medicines 12, 321–334.

Felipe, D.F., Brambilla, L.Z.S., Porto, C., Pilau, E.J., Cortez, D.A.G., 2014. Phytochemicalanalysis of Pfaffia glomerata inflorescences by LC-ESI-MS/MS. Molecules 19,15720–15734.

Furukawa, T., Bai, C.X., Kaihara, A., Kaihara, A., Ozaki, E., Kawano, T., Nakaya, Y., Awais, M.,Sato, M., Umezawa, Y., Kurokawa, J., 2006. Ginsenoside Re, a main phytosterol ofPanax ginseng, activates cardiac potassium channels via a nongenomic pathway ofsex hormones. Molecular Pharmacology 70, 1916–1924.

Gauthaman, K., Ganesan, A.P., 2008. The hormonal effects of Tribulus terrestris andits role in the management of male erectile dysfunction: an evaluation usingprimates, rabbit and rat. Phytomedicine 15, 44–54.

Gauthaman, K., Adaikan, G.P., Prasad, V.N.R., 2002. Aphrodisiac properties ofTribulus Terrestris extract (Protodioscin) in normal and castrated rats. LifeScience 71, 1385–1396.

Gauthaman, K., Ganesan, A.P., Prasad, R.N., 2003. Sexual effects of puncturevine(Tribulus terrestris) extract (protodioscin): an evaluation using a rat model.Journal of Alternative and Complementary Medicine 9, 257–265.

Gosmann, G., Gattuso, S., Gattuso, M., Fenner, R., Pacheco, E.F., Ferraz, A., Savi, L.A.,Barardi, C.R.M., Simões, C.M.O., Sortino, M., Zacchino, S., Gnerre, C., Testa, B.,Rates, S.M.K., 2003. Botanical (morphological, micrographic), chemical andpharmacological characteristics of Pfaffia species (amaranthaceae) native tosouth Brazil. Revista Brasileira de Ciências Farmacêuticas 39, 141–147.

Huhtaniemi, I., 2014. Late-onset hypogonadism: current concepts and controversies ofpathogenesis, diagnosis and treatment. Asian Journal of Andrology 16, 192–202.

Kizelsztein, P., Govorko, D., Komarnystsky, S., Evans, A., Wang, Z., Cefalu, W.T.,Raskin, I., 2009. 20-Hydroxyecdysone decreases weight and hyperglycemia in adiet-inducedobesity mice model. The American Journal of Physiology, Endocri-nology and Metabolism 296, 433–439.

Korenchevsky, V., Dennison, M., Hall, K., 1937. The effects of testosterone andtestosterone propionate on adult male rats (compared with those on femalerats). Biochemical Journal 31, 1434–1437.

Lynch, G.S., Schertzer, J.D., Ryall, J.G., 2007. Therapeutic approaches for musclewasting disorders. Pharmacology & Therapeutics 113, 461–487.

Kotta, S., Ansari, S.H., Ali, J., 2013. Exploring scientifically proven herbal aphrodisiacs.Pharmacognosy Reviews 7, 1–10. http://dx.doi.org/10.4103/0973-7847.112832.

Malini, T., Vanithakumari, G., 1992. Comparative study of the effects of beta-sitosterol, estradiol and progesterone on selected biochemical parameters ofthe uterus of ovariectomised rats. Journal of Ethnopharmacology 36, 51–55.

Marques, L.C., Danucalov, M.A., Torres, F., Galduróz, J.C.F., Carlini, E.L.A., Silva, A.C.,2002. Double-blind clinical study of standardized extract (BNT-08) of the rootsof Pfaffia glomerata (Spreng.) Pedersen: evaluation of the tonic effect onphysical activity. Revista Brasileira de Farmacognosia 12 (Suppl.), S44–S47.

Marques, L.C., Galvão, S.M.P., Espínola, E., Dias, R.F., Mattei, R., Oliveira, M.G.M.,Carlini, E.L.A., 2004. Psychopharmacological assessment of Pfaffia glomerataroots (Extract BNT-08) in rodents. Phytotherapy Research 18, 566–572.

Martino-Andrade, A.J., Morais, R.N., Spercoski, K.M., Rossi, S.C., Vechi, M.F., Golin, M.,Lombardi, N.F., Greca, C.S., Dalsenter, P.R., 2010. Effects of tribulus terrestris onendocrine sensitive organs in male and female wistar rats. Journal of Ethno-pharmacology 127, 165–170.

Mendes, F.R., 2011. Tonic, fortifier and aphrodisiac: adaptogens in the Brazilian folkmedicine. Brazilian Journal of Pharmacognosy 21, 754–763.

Mousli, R., Tazerouti, A., 2007. Direct method of preparation of dodecanesulfona-mide derivatives and some surface properties. Journal of Surfactants andDetergents 10, 279–285.

Munro, C.J., Stabenfeldt, G.H., Cragun, J.R., Addiego, L.A., Overstreet, J.W., Lasley, B.L.,1991. Relationship of serum estradiol and progesterone concentrations to theexcretion profiles of their major urinary metabolites as measured by enzymeimmunoassay and radioimmunoassay. Clinical Chemistry 37, 838–844.

Nakamura, S., Chen, G., Nakashima, S., Matsuda, H., Pei, Y., Yoshikawa, M., 2010.Brazilian natural medicines. IV. New noroleanane-type triterpene andecdysterone-type sterol glycosides and melanogenesis inhibitors from the rootsof Pfaffia glomerata. Chemical and Pharmaceutical Bulletin 58, 690–695.

Nantia, E.A., Moundipa, P.F., Monsees, T.K., Carreau, S., 2009. Medicinal plants aspotential male anti-infertility agents: a review. Andrologie 19, 148–158.

Neychev, V.K., Mitev, V.I., 2005. The aphrodisiac herb Tribulus terrestris does notinfluence the androgen production in young men. Journal of Ethnopharmacol-ogy 101, 319–323.

Oshima, M., Gu, Y., 2003. Pfaffia paniculata-induced changes in plasma estradiol-17beta, progesterone and testosterone levels in mice. Journal of Reproductionand Development 49, 175–180.

Owens, W., Zeiger, E., Walker, M., Ashby, J., Onyon, L., Gray Jr., L.E., 2006. The OECDprogram to validate the rat Hershberger bioassay to screen compounds for in vivoandrogen and antiandrogen responses. Phase 1: use of a potent agonist and a potentantagonist to test the standardized protocol. Environmental Health Perspectives 114,1259–1265.

Park, K.S., 2007. Korean ginseng induces spermatogenesis in rats through theactivation of cAMP-responsive element modulator (CREM). Fertility and Steri-lity 88, 1000–1002.

Pinello, K.C., Fonseca Ede, S., Akisue, G., Silva, A.P., Salgado Oloris, S.C., Sakai, M.,Matsuzaki, P., Nagamine, M.K., Palermo Neto, J., Dagli, M.L., 2006. Effects of Pfaffiapaniculata (Brazilian ginseng) extract on macrophage activity. Life Sciencies 78,1287–1292.

Reich, E., Schibli, A., 2006. High Performance for the Analysis of Medicinal Plants.Thieme, New York, Stuttgart, USA.

Riffault, L., Destandau, E., Pasquier, L., André, P., Elfakir, C., 2014. Phytochemical analysisof Rosa hybrida cv. ‘Jardin de Granville’ by HPTLC, HPLC-DAD and HPLC-ESI-HRMS:polyphenolic fingerprints of six plant organs. Phytochemistry 99, 127–134.

Russell, L.D., Ettlin, R.A., Hikim Sinha, A.P., Clegg, E.D., 1990. Histological andHistopathological Evaluation of the Testis. Cache River Press, Clearwater.

Saudan, C., Baume, N., Emery, C., Strahm, E., Saugy, M., 2008. Short term impact ofTribulus terrestris intake on doping control analysis of endogenous steroids.Forensic Science International 178, e7–e10.

Shariat, S.F., Lamb, D.J., Iyengar, R.G., Roehrborn, C.G., Slawin, K.M., 2008. Herbal/hormonal dietary supplement possibly associated with prostate cancer pro-gression. Cinical Cancer Research 14, 607–611.

Shiobara, Y., Inoue, S.S., Kato, K., Nishiguchi, Y., Oishi, Y., Nishimoto, N., Oliveira, F.,Akisue, G., Akisue, M.K., Hashimoto, G., 1993. A nortriterpenoid, triterpenoidsand ecdysteroids from Pfaffia glomerata. Phytochemistry 32, 1527–1530.

Silva, T.C., Paula da Silva, A., Akisue, G., Luis Avanzo, J., Kazumi Nagamine, M.,Fukumasu, H., Matsuzaki, P., César Raspantini, P., Haraguchi, M., Lima Górniak, S.,Dagli, M.L., 2005. Inhibitory effects of Pfaffia paniculata (brazilian ginseng) onpreneoplastic and neoplastic lesions in a mouse hepatocarcinogenesis model.Cancer Letters 226, 107–113.

Singh, P., 2013. Andropause: current concepts. Indian Journal of Endocrinology andMetabolism 17, S621–S629.

Tóth, N., Szabó, A., Kacsala, P., Héger, J., Zádor, E., 2008. 20-Hydroxyecdysone increasesfiber size in a muscle-specific fashion in rat. Phytomedicine 15, 691–698.

Wagner, H., Bladt, S., Zgainski, E.M., 1983. Plant Drug Analysis: A Thin LayerChromatography Atlas. Springer-Verlag, Berlin, Heidelberg, New York.

Wilborn, C.D., Taylor, L.W., Campbell, B.I., Kerksick, C., Rasmussen, C.J., Greenwood, M.,Kreider, R.B., 2006. Effects of methoxyisoflavone, ecdysterone, and sulfo-polysaccharide supplementation on training adaptations in resistance-trainedmales. Journal of the International Society of Sports Nutrition 3, 19–27.

Wilson, V.S., Lambright, C., Ostby, J., Gray Jr., L.E., 2002. In vitro and in vivo effects of 17β-Trenbolone: a feedlot effluent contaminant. Toxicological Sciences 70, 202–211.

Yue, P., Mak, N., Cheng, Y., Leung, K.W., Ng, T.B., Fan, D.T., Yeung, H.W., Wong, R.N.,2007. Pharmacogenomics and the Yin/Yang actions of ginseng: anti-tumor,angiomodulating and steroid-like activities of ginsenosides. Chinese Medicine2007 (2), 6. http://dx.doi.org/10.1186/1749-8546-2-6.

Zimmer, A.R., Bruxel, F., Bassani, V.L., Gosmann, G., 2006. HPLC method for thedetermination of ecdysterone in extractive solution from Pfaffia glomerata.Journal of Pharmaceutical and Biomedical Analysis 40, 450–453.

N.F. Fernandes et al. / Journal of Ethnopharmacology 161 (2015) 46–5252