Serotonin regulates prostate growth through androgen receptor modulation

Aging and testosterone almost inexorably cause benign prostatic hyperplasia (BPH) in Human males. However, etiology of BPH is largely unknown. Serotonin (5-HT) is produced by neuroendocrine prostatic cells and presents in high concentration in normal prostatic transition zone, but its function in prostate physiology is unknown. Previous evidence demonstrated that neuroendocrine cells and 5-HT are decreased in BPH compared to normal prostate. Here, we show that 5-HT is a strong negative regulator of prostate growth. In vitro, 5-HT inhibits rat prostate branching through down-regulation of androgen receptor (AR). This 5-HT’s inhibitory mechanism is also present in human cells of normal prostate and BPH, namely in cell lines expressing AR when treated with testosterone. In both models, 5-HT’s inhibitory mechanism was replicated by specific agonists of 5-Htr1a and 5-Htr1b. Since peripheral 5-HT production is specifically regulated by tryptophan hydroxylase 1(Tph1), we showed that Tph1 knockout mice present higher prostate mass and up-regulation of AR when compared to wild-type, whereas 5-HT treatment restored the prostate weight and AR levels. As 5-HT is decreased in BPH, we present here evidence that links 5-HT depletion to BPH etiology through modulation of AR. Serotoninergic prostate pathway should be explored as a new therapeutic target for BPH.

Benign prostatic hyperplasia (BPH) is one of the main causes of non-neurogenic lower urinary tract symptoms (LUTS) in the aging male 1,2 . The underlying mechanism responsible for BPH is not understood, and only elucidating the etiology of BPH will increase our ability to treat or even prevent its development.
Currently the most accepted hypothesis for the etiology of BPH is, that proposed by McNeal, in which BPH results from the reawakening of inductive potential in adult prostatic stroma in a specific prostatic region defined as transition zone [3][4][5] . This hypothesis claimed that the adult prostatic epithelium retains the ability to respond to inductive stromal signaling with new ductal branching morphogenesis 6,7 . However this hypothesis does not respond to the critical question of why this reawakening of human adult prostatic stroma occurs.
While there is no BPH without testosterone 8 , testosterone levels decrease with age 9,10 and no direct correlation between testosterone concentration and prostate volume has been established yet 11 . Moreover, it is widely accepted that physiologic concentrations of testosterone provide an excess of testosterone for optimal prostatic growth suggesting that testosterone is not the etiologic factor responsible for BPH 12 . On the other hand, several reports have documented an up-regulation of the androgen receptor (AR) in BPH tissue, unveiling a potential role for AR in BPH etiopathogenesis [13][14][15] .
The neuroendocrine prostatic cells secrete various neuroendocrine factors with 5-HT being one of the most abundant. The peculiar morphology of some neuroendocrine cells with dendritic processes extending to lumen and projections surrounding the epithelial-stroma interface justify the hypothesis that neuroendocrine products, namely 5-HT, could regulate prostate growth 16 . Notably, neuroendocrine prostatic cells are mainly located in the transition zone of the normal human prostate 17 , where BPH originates 4 . However, comparing BPH tissue with normal transition zone (without BPH) the number of neuroendocrine cells is extraordinarily decreased [18][19][20] . Also 5-HT was shown to be significantly depleted in BPH tissue 19 . Furthermore, a recent study in a large cohort of Scandinavian men revealed that LUTS are associated with benign prostate enlargement and to decreased plasmatic 5-HT concentration 21 . These findings suggest a potential link between prostatic 5-HT depletion and BPH etiology; however, the function of 5-HT in regulation of benign prostate growth has never been studied.
We hypothesized that 5-HT had an inhibitory function over benign prostate growth and that suppression of prostatic 5-HT production could be responsible for benign prostatic growth. The aim of this study was to define the role of 5-HT in the regulation of benign prostatic growth and to test the pharmacologic modulation of the prostatic serotoninergic system as a new pharmacological target for BPH.

5-HT, 5-Htr1a, and 5-Htr1b specific agonists inhibits rat ventral prostate branching through AR down-regulation.
The new epithelial gland formation observed in BPH is normally seen only during prostate branching morphogenesis 22 . For this reason, we first tested the hypothesis that 5-HT could regulate prostate growth using in vitro cultures of rat ventral prostate explants (VPs) from P1 newborns. During 4 days in culture, 5-HT supplementation induced a significant dose-dependent inhibition of rat VPs growth (Fig. 1a), as expressed by decreased area (Fig. 1b), as well the number of peripheral explant buds (Fig. 1c). In medium conditions without additional testosterone supplementation, inhibitory effect of 5-HT over VPs growth was maximal at 100 µM where a reduction of 40% in prostate area D 4 /D 0 (p < 0.001) and a reduction of 42% in the number of peripheral buds D4/D0 (p < 0.001) was observed in comparison to the control group (0 µM 5-HT). As expected, testosterone supplementation of VPs exerted a strong stimulatory effect on prostate branching morphogenesis, mainly in the number of peripheral buds (Fig. 1c), but again, 5-HT at 100 µM reduced 33% the prostate area D 4 / D 0 (p < 0.001) and 36% the number of peripheral buds D4/D0 (p < 0.001) in comparison to control group (0 µM 5-HT + testosterone).
From all 5-HT receptors, 5-Htr1a and 5-Htr1b were the most extensively studied in the regulation of malignant prostate growth 23,24 , so we tested if these receptors could contribute to the 5-HT inhibitory function in normal prostate growth. By immunofluorescence, we found that both receptors are strongly expressed in rat prostate but with a slightly different distribution pattern with 5-Htr1a predominantly expressed in prostate epithelium (Fig. 1d), while 5-Htr1b being expressed both in epithelium and stroma (Fig. 1e). To determine the contribution of both receptors in 5-HT inhibition of prostate branching morphogenesis, VPs were treated with drugs that specifically activate 5-Htr1a or 5-Htr1b. The selective 5-Htr1a agonist 8-OH-DPAT, (Fig. 1f,g and h) and the selective 5-Htr1b agonist, anpirtoline, (Fig. 1i, j and k) induced a significant dose-dependent inhibition of VPs growth. The inhibitory effect was maximal in VPs supplemented with testosterone and treated with 100 µM of anpirtoline, where a reduction of 39% in prostate area D 4 /D 0 (p < 0.001) and a reduction of 66% in the number of peripheral buds D4/D0 (p < 0.001) was observed in comparison to the control group (0 µM anpirtoline + testosterone).
Since, androgens are a major prostatic stimulatory factor, we asked if the 5-HT inhibitory effect was related to the AR stimulatory pathway. By western blot analysis we showed that testosterone supplementation induced AR up-regulation, but 5-HT treatment significantly decreased AR expression either with or without testosterone supplementation ( Fig. 2a and b) suggesting that the inhibitory function of 5-HT could be related to inhibition of the AR pathway. Similarly, both the selective 5-Htr1a agonist 8-OH-DPAT, (Fig. 2a and c) and the selective 5-Htr1b agonist, anpirtoline, ( Fig. 2a and d) induced a significant AR down-regulation, more evident in anpirtoline treated VPs. Taken together these results indicate that in vitro 5-HT inhibits rat prostate growth through 5-Htr1a and 5-Htr1b, by down-regulating AR. 5-HT, 5-Htr1a or 5-Htr1b specific agonists inhibit growth of androgen sensitive human benign prostate cells through AR down-regulation. Next, we asked if this mechanism is also present in human prostate. With this purpose we performed in vitro 5-HT treatment of different human cell lines from epithelium of BPH (BPH-1), normal prostate epithelium (PNT1A) and normal prostate stroma (WPMY-1). We found that 5-HT significantly reduced cell viability of BPH-1 and WPMY-1 namely in the presence of testosterone but without changing PNT1A cell viability (Fig. 3a,b,c and Supplementary Fig. 1). The inhibitory effect of 5-HT was maximal in BPH-1 cells supplemented with testosterone. Under these conditions, 100 µM of 5-HT decreased cell viability by 35% compared to control (0 µM 5-HT + testosterone) (p < 0.001).
Next, we tested if the growth inhibitory function of 5-HT in androgen human prostate cells BPH-1 and WPMY-1 was mediated by 5-Htr1a and 5-Htr1b. First, we showed that both receptors are expressed in the three human prostate cell lines (Fig. 3d) and that their specific agonists significantly inhibited cell viability (  Additionally, Ki-67 staining confirmed that proliferation of BPH-1 and WPMY-1 cells supplemented with testosterone was significantly reduced by 5-HT treatment ( Fig. 4a and c), while PNT1A cells proliferation was not affected (Fig. 4b). Similarly, both specific agonists of 5-Htr1a and 5-Htr1b strongly inhibited cell proliferation ( Fig. 4d-i) but again only in BPH-1 and WPMY-1 cells.
Next, we investigated if this inhibitory function of 5-HT, 5-Htr1a and 5-Htr1b specific agonists was related to changes in the AR pathway. We observed that testosterone induced an up-regulation of AR in both BPH-1 and WPMY-1 cells. 5-HT, 5-Htr1a and 5-Htr1b specific agonists inhibited the AR up-regulation induced by testosterone in both BPH-1 and WPMY-1 cells (Fig. 5b). Regarding 5-HT effect in AR down-regulation this was very significant in WPMY-1 cells after testosterone supplementation (p < 0,001) while in BPH-1 cells a non-significant down-regulation of AR was observed ( Fig. 5c and d). Also for both 5-Htr1a and 5-Htr1b specific agonists only Anpirtoline induced a significant down-regulation of AR in both BPH-1 and WPMY-1 cells (Fig. 5d,e,g,h). Additionally, by immunofluorescence analysis we observed that expression of AR after testosterone treatment was decreased in BPH-1 cells treated with 5-HT, 5-Htr1a and 5-Htr1b specific agonists ( Supplementary Fig. 2).
Remarkably, the absence of inhibitory action of 5-HT, 5-Htr1a and 5-Htr1b specific agonists on PNT1A cells viability and proliferation, even in presence of testosterone, co-existed with a complete absence of AR expression in these cells (Fig. 5a). These data strongly argue that 5-HT's inhibitory function on growth of benign human prostate cells is related with the suppression of the AR pathway.

In vivo ablation of peripheral 5-HT synthesis in mice induces benign prostatic growth. 5-HT
synthesis is initiated by tryptophan hydroxylase (Tph). Tph type 1 (Tph1) and 2 (Tph2) regulate 5-HT production in non-neuronal and neuronal tissues, respectively 25,26 . The majority of 5-HT in the body is produced by Tph1. In fact, Tph1 −/− mice exhibit very low levels of circulating 5-HT, while brain serotonin is not affected 25 . Based on our in vitro findings which suggest that 5-HT has a strong inhibitory action on prostate growth through down-regulation of AR, we used Tph1 −/− mice to evaluate the effect of peripheral 5-HT depletion on mouse prostate gland growth. Remarkably, Tph1 −/− mice exhibited a significantly 37% higher prostate-to-body weight ratio compared to wild-type at 20 weeks (p < 0.001) ( Fig. 6a and b) without changes in body weight ( Supplementary  Fig. 3a). Interestingly, histology of the prostate gland revealed that Tph1 −/− mice exhibit areas of hyperplasia in epithelium and stroma (Fig. 6a, lower panel). To determine if 5-HT treatment could revert higher prostate mass in Tph1 −/− mice, we performed intraperitoneal injections of 5-HT at 10 consecutive days. 5-HT treatment resulted in significant mass reduction in prostate gland compared to levels similar to the wild-type ( Fig. 6c and d) again without affecting animal weight ( Supplementary Fig. 3b). Next, we asked if the higher prostate mass in Tph1 −/− mice was associated with different expression of AR. We could not demonstrate significant differences of AR expression by western blot analysis in total prostate (data not shown), however by immunofluorescence dorsolateral prostate of Tph1 −/− mice appeared to express more AR (Fig. 6e, lower panel). So, we investigated and demonstrated by qRT-PCR that the dorsolateral prostate of Tph1 −/− mice has increased levels of AR mRNA expression, while 5-HT treatment partially restores it to levels to wild-type mice (Fig. 6e, upper panel), reinforcing our hypothesis.
It is well established that castration induces a strong reduction in size of the prostate gland as well as in seminal vesicles, while testosterone supplementation makes both organs return to normal size 27,28 . Interestingly, also seminal vesicles of Tph1 −/− mice were significantly larger than the ones of wild-type (Fig. 6f) suggesting that 5-HT could regulate androgen sensitivity not only in prostate gland but also in seminal vesicles. Again, we observed that the mass of seminal vesicles was partially restored by 5-HT treatment in Tph1 −/− mice (Fig. 6g). Lastly, we tested if wild-type mice challenged with 5-HT treatment continue responding to 5-HT's inhibitory action, and we demonstrated that both prostate gland (Fig. 6h) as well as seminal vesicle mass (Fig. 6i) were reduced while animal weight was not affected (Supplementary Fig. 3c).

Discussion
Currently, the etiology of BPH is unknown. However, it is accepted today that BPH is a consequence of aging and the simultaneous presence of testosterone 3 . BPH almost universally affects human males and a significant part of men will develop bothersome LUTS because of benign prostate enlargement. Although some of these men respond to current medical treatment (mainly α 1 -adrenoreceptors antagonists and 5 α-reductase inhibitors) a large portion continues to need a surgical procedure to treat resistant LUTS or have even more serious complications of BPH 29 , creating the emerging necessity for novel therapies.
In this study, we investigated the function of 5-HT in the regulation of non-malignant prostatic growth. Here, we demonstrated for the first time in several in vitro and in vivo models that 5-HT is a powerful negative regulator of prostatic growth through down-regulation of AR. We found that 5-Htr1a and 5-Htr1b are strongly expressed in the rodent prostate gland as well in human benign prostate cells, and that both receptors could mediate the inhibitory action of 5-HT on prostate growth.
Our in vitro and animal findings lead us to propose a new mechanism to explain the development of BPH in humans (Fig. 7). Our proposed model explains how the depletion of neuroendocrine cells and serotonin observed in prostatic transition zone with aging [18][19][20] , could be the etiologic factor responsible for the initiation and progression of BPH. In our model, the depletion of serotonin induces an up-regulation of androgen receptor in the prostatic transition zone leading to the stimulation of benign prostatic growth in this specific prostatic region.
Our neuroendocrine model for the etiopathogenesis of BPH would resolve an intriguing question about the crucial participation of androgens in the development of BPH. Previous studies have showed that testosterone does not increase with aging and some studies have even reported that plasmatic testosterone is decreased in the aging human male [9][10][11] . In a similar way, intraprostatic testosterone, in particular DHT, are not increased in BPH comparatively to normal prostate 30 and even the administration of testosterone in supra-physiologic concentrations to eugonadal men does not induce the development of BPH 31,32 . This data were the basis for the saturation model of prostate growth proposed by Morgentaler et al. suggesting that prostate gland growth is extraordinarily sensitive to low androgen concentrations (near the castrate range) but insensitive to androgen concentrations above a certain saturation point 12 . The critical point, in saturation model is that the rate-limiting step for prostatic growth, which is the concentration of AR. Recent studies demonstrated that AR is in fact up-regulated in stroma and epithelium of BPH tissue comparatively to normal prostate, implicating AR in etiophatogenesis of BPH [13][14][15] . Our findings seem to provide an explanation for this current view about the participation of androgens in the development of BPH. In this way, the loss of neuroendocrine cells and serotonin in prostatic transition zone up-regulates the AR and then permits the development of BPH, even with a decreased plasmatic concentration of androgens observed in the aging male.
Our first in vitro experimental approach focused on the function of 5-HT in the regulation of rat prostate branching morphogenesis. Because BPH is the result of new branching morphogenesis, this model permits the study the influence of 5-HT exactly in the mechanism by which BPH develops and progresses. Here, we demonstrate that 5-HT strongly inhibit the branching morphogenesis of the prostate gland through down-regulation of AR. In fact, other organs like the prostate which have a development process of branching morphogenesis, such as the mammary gland, 5-HT also have demonstrated a development inhibitory action 33 .
Although new in the prostate gland, the serotoninergic inhibitory mechanism through down-regulation of AR, is well known in the brain. In fact, the complete masculinization of the brain is dependent of a perinatal surge of testosterone and a simultaneous decrease in hypothalamic 5-HT concentrations 34 . In the brain, it has been demonstrated that regulation of 5-HT concentration is crucial for normal sexual differentiation, where 5-HT down-regulates AR 35,36 . In agreement with our findings, Sayed et al. demonstrated that dapoxetine decreased AR expression and prevents testosterone-induced BPH in rats 37 . However, both in brain and prostate the full mechanism responsible for this down-regulation remains to be elucidated.
The most described 5-HT receptors in prostatic cells, are the 5-Htr1a and 5-Htr1b. Therefore, we characterize its expression in rat prostate for the first time. As we demonstrated both 5-Htr1a and 5-Htr1b are strongly expressed in rat prostate and the activation of these receptors resulted in a significant inhibition of prostate branching morphogenesis, which also occurred through down-regulation of AR. In human prostatic cells the function of 5-HT, 5-Htr1a, and 5-Htr1b have been previously studied but only in malignant cells 23,24 . These previous reports showed that 5-HT has a proliferative effect in several malignant cell lines through 5-Htr1a and 5-Htr1b. Curiously, this stimulatory effect was mainly evident in androgen-insensitive cells 23 . Here we studied, for Finally, to test in vivo our mechanistic approach to the etiopathogenesis of BPH, we genetically ablated the peripheral production of 5-HT. Using Tph1 −/− mice we demonstrated that prostatic 5-HT depletion induces benign prostatic growth. The inhibition of peripheral 5-HT synthesis in mice, through genetic deletion of Tph1, simulates a decrease in the prostate transition zone 5-HT observed in the aging male. This led us to propose that the decrease in neuroendocrine cells and 5-HT in the human transition zone could contribute to the development of BPH. The increased mass of Tph1 −/− prostates was associated to an up-regulation of AR in dorso-lateral prostate samples suggesting that, at least in part, the excessive prostatic growth in Tph1 −/− could be attributed to AR up-regulation.
In conclusion, our findings suggest that 5-HT is a strong negative regulator of prostate growth through AR down-regulation. As 5-HT is decreased in BPH, we present here evidence that links 5-HT-producing neuroendocrine cell depletion to BPH etiology. Therefore, this new described serotoninergic inhibitory pathway over benign prostatic growth should be explored as a new target for BPH treatment.

Methods
Ethics and animal work. Mice and rats were maintained in accordance with the guidelines of "Guide to the Care and Use of Experimental Animals" National Academy of Science, and the EU Directive 2010/63/EU. This study was approved by the Animal Ethics Committee of the Institution were the study was performed (SECVS 003/2016) and by the National Competent Authority for Animal Protection (DGAV 0421/000/000/2016).
Rat ventral prostate cultures. Newborn male Sprague-Dawley rats were sacrificed 24-hours after birth.
Ventral prostate lobes were microdissected using a stereomicroscope (Leica MZ6, Switzerland) and processed for organ culture. Organ culture was performed as previously described 38 . Briefly, rat ventral prostates (VPs) from P1 newborns were cultured for 4 days at 37 °C in a humidified atmosphere of 5% CO 2 . Medium and VPs were transferred to porous membranes (Millicell CM filters, Millipore Corp., Bedford, Massachusetts) in 12 well plate for floating explant cultures. Each VPs were dipped into 500 μl of 1:1 mixture of DMEM and Ham's F-12 nutrient supplemented with 100 μg/mL streptomycin, 100 units/mL penicillin, 10 µg/mL transferrin and 10 µg/mL insulin. Media were replenished at 48 hours of culture. Branching morphogenesis in all groups was monitored daily by a stereomicroscope and photographs were taken at day 0 and day 4. The number of peripheral buds was manually counted and the prostate tissue area was measured in ImageJ using the beProstate plugin (Version 1.0) (developed by Biomedical Engineering Solutions Research Group, Life and Health Sciences Research Institute, University of Minho; available at http://www.besurg.com/sites/default/files/beProstateApp.zip). The differences between day 0 (D0: 0 hours) and day 4 (D4: 96 hours) of culture, were expressed as D4/D0 ratio. A total of 427 VPs were cultured divided in three experimental groups: in specific pathogen-free conditions in a room maintained at a constant temperature of 23 °C on a 12-h light-dark cycle. Food and water were provided ad libitum. All treatment groups were age matched and randomized to treatment at the initiation of an experiment. The researchers performing the experiments were blinded to experimental groups during all testing. Animals were excluded from analysis if signs of fight with skin lesions were present. Tph1 −/− mice on a C57BL/6 background were provided by M. Bader (Max Delbrück Center for Molecular Medicine, Berlin, Germany). For assessment of the morphological evolution with age both male wild-type and Tph1 −/− mice were sacrificed at different time points (at least 6 animals for both groups at each time-point): 7, 12, 16 and 20 weeks-old. For pharmacological studies, wild-type and Tph1 −/− mice with 19 week-old were treated daily with intraperitoneal injections of 0.9% saline or 5-HT (100 mg/Kg) during 10 consecutive days (at least 6 animals for each group). Mice were sacrificed and prostate tissue (all lobes combined) and seminal vesicles were micro dissected away from other urogenital and fat tissues. Total prostate was weighted immediately after dissection. The right lobes were separated from the left and processed for histology or western blotting. For histologic analysis hemi-right prostate was fixed in 10% PFA, processed and embedded in paraffin and stained with hematoxylin and eosin (H&E). Hemi-left prostate was separated in three lobes (ventral, dorsolateral and anterior) and processed to qRT-PCR.
Immunofluorescence analysis. Immunofluorescence for AR, Ki-67, 5-Htr1a and 5-Htr1b was performed on formalin-fixed and paraffin-embedded rat ventral prostates or in human prostate cell lines. Briefly, deparaffinized and rehydrated slides were submitted to adequate heat-induced antigen retrieval for 20 min at 98 °C with 10 mM citrate buffer (pH 6.0). Regarding cell lines, all of them were plated in glass coverslips placed into 12-well plates at a density of 5 × 10 5 cells per well, and allowed to adhere overnight. Then, the cells were fixed in cold methanol by 5 minutes at −20 °C. In both paraffin and cell, for block unspecific ligations the cells/tissues were incubated with a solution of PBS containing 10% FBS for 30 minutes at room temperature followed by incubation with a primary antibody against AR (1:1000 dilution; sc-816: Santa Cruz Biotechnology, Santa Cruz, California), Ki-67 (1:100 dilution, AP10244C; Gennova, Sevilla, Spain), 5-Htr1a (1:100 dilution; sc-10801: Santa Cruz Biotechnology, Santa Cruz, California) and 5-Htr1b (1:100 dilution; sc-28937: Santa Cruz Biotechnology, Santa Cruz, California). The cells/tissues were then washed in a PBS solution with 0.5% FBS and incubated with a goat anti-rabbit antibody conjugated with FITC for cells and with TRITC for tissues (dilution 1:500, Life Technology, Carlsbad, California) for 1 hour at room temperature in the dark. Finally, the cells were counterstained with 40,6-diamidino-2-phenylindole (DAPI). The images were obtained using a fluorescence microscopy (BX16; Olympus).
Western blot analysis. Western blot analysis for androgen receptor (AR) was done in both VPs, mouse prostate tissue and in human cell lines. All the samples were properly processed for western blot analysis and lysed in a buffer containing 50 mM Tris pH 7.6-8, 150 mM NaCl, 5 mM EDTA, 1 mM Na3VO 4 , 10 mM NaF, 10 mM NaPyrophosphate, 1% NP-40 and 1/7 of Protease cocktail inhibitors (Roche). Western blotting was done Scientific REpoRtS | 7: 15428 | DOI:10.1038/s41598-017-15832-5 using standard 10% SDS-PAGE gels, loading 20 µg of protein per lane. For AR detection a specific antibody was used (1:1000 dilution; sc-816: Santa Cruz Biotechnology, Santa Cruz, California). β-Actin was used for loading control (1:500 dilution; sc-1616; Santa Cruz Biotechnology, Santa Cruz, California). After incubation with appropriate secondary antibodies, they were detected by chemiluminescence (Thermo Scientific Pierce ECL Western Blotting) in ChemiDoc ™ XRS + System (Bio-Rad). Quantification of western blot results was done using the band densitometry analysis, performed with ImageJ software.
RNA extraction and qRT-PCR. Total RNA was isolated from the dorsolateral prostate of different groups with Trizol (Invitrogen, Carlsbad, California). Then, after quantification using the NanoDrop ® , 500 ng of total RNA was reverse transcribed into first strand cDNA using the iScript ™ cDNA Synthesis Kit (Bio-Rad Laboratories, Hercules, California). Primers used to measure the expression levels of AR was designed using the Primer3 software, on the basis of the respective GenBank sequence. All accession numbers and primer sequences are available on request. The reference gene for hypoxanthine guanine phosphoribosyl transferase (Hprt) (accession number from GenBank: NM_013556) was used as an internal standard for the normalization of the expression of selected transcripts. qRT-PCR was performed on a CFX 96TM real time system instrument (Bio-Rad Laboratories, Hercules, California), with the QuantiTect SYBR Green RT-PCR reagent kit (Qiagen, Hamburg, Germany), using equal amounts of RNA from each one of the samples. Product fluorescence was detected at the end of the elongation cycle. All melting curves exhibited a single sharp peak at the expected temperature.
Statistics. Data are presented as mean ± SEM. Statistical analysis was performed using GraphPad Prism by Student's t test or ANOVA where appropriate. A Bonferroni post hoc test was used to test for significant differences revealed by ANOVA. Statistical significance was confirmed at p < 0.05.