Beta-Sitosterol Alters Collagen Distribution in Prostate Fibroblasts

Abstract

Herbal supplements containing several types of plant sterols, vitamins, and minerals, are marketed for prostate health. In the majority of these supplements, the most abundant plant sterol is saw palmetto extract or its’ principal component, beta-sitosterol. In terms of prostate health, previous work almost exclusively focused on the effects of beta-sitosterol on prostatic epithelium, with little attention paid to the effects on prostatic stroma. This omission is a concern, as the abnormal accumulation of collagen, or fibrosis, of the prostatic stroma has been identified as a factor contributing to lower urinary tract symptoms and dysfunction in aging men. To address whether beta-sitosterol may be promoting prostatic fibrosis, immortalized and primary prostate stromal fibroblasts were subjected to immunoblotting, immunofluorescence, qRT-PCR, ELISA, and image quantitation and analysis techniques to elucidate the effects of beta-sitosterol on cell viability and collagen expression and cellular localization. The results of these studies show that beta-sitosterol is nontoxic to prostatic fibroblasts and does not stimulate collagen production by these cells. However, beta-sitosterol alters collagen distribution and sequesters collagen within prostatic fibroblasts, likely in an age-dependent manner. This is a significant finding as prostate health supplements are used predominantly by middle aged and older men who may, then, be affected disproportionately by these effects.

Keywords:

• Beta-sitosterol
• collagen
• fibroblast
• prostate
• stroma
• supplement

Introduction

In 2020, U.S. consumers spent nearly $25 M on supplements marketed for prostate health, followed by cardiovascular and immune health (Smith et al. 2021). Herbal supplements intended to enhance prostate health may contain several types of plant sterols, vitamins, and minerals. In the majority of these supplements, the most abundant plant sterol is saw palmetto extract or its’ principal component, beta-sitosterol. Analysis of the plant sterol composition of several prostate herbal supplements using data publicly available from independent laboratories, such as Eurofins and Labcorp (formerly Covance), demonstrates a wide range, by weight, for the most common sterols in these supplements –campesterol stigmasterol, beta-sitosterol, brassicasterol, and cholesterol. Among these, beta-sitosterol accounted for 42.4%–78.0% of the total plant sterol content. A multidimensional scaling of the data shows that the herbal supplement Prostagenix is an outlier compared to the other herbal supplements due to its high beta-sitosterol content . Heatmap visualization of the differences between the plant sterol contents of the same herbal supplements suggests that Prostagenix, Prost-T, and Prostvar Ultra share more identity with each other than with the other supplements, due to more similar compositions for beta-sitosterol. It should be noted, however, that these supplements varied widely from each other in terms of composition for campesterol, stigmasterol, cholesterol, and brassicasterol.

Beta-sitosterol demonstrates some anti-tumorigenic activities in vitro in cultured prostate cells and in vivo in mouse prostate tumor models, and has it has been associated with decreased prostate cancer cell invasiveness, motility, and laminin/fibronectin binding (Stone et al. 1978; von Holtz et al. 1998; Awad et al. 2001; Pradhan et al. 2019). It has also been shown to inhibit the 5α-reductase enzyme and thereby prevent the conversion of testosterone to its’ active form, dihydrotestosterone (DHT). DHT is the major growth factor of the prostate, and inhibition of DHT activity has been shown to slow the growth of prostate tumors and benign prostatic hyperplasia (BPH) (Vickman et al. 2020). Hence, the mechanism of action for beta-sitosterol in the prostate is largely attributed to its ability to act as a 5α-reductase enzyme inhibitor, and reduce the activation levels of the androgen receptor, for which DHT is a major ligand (Sultan et al. 1984, Vickman et al. 2020).

BPH, a proliferative but nonmalignant enlargement of the prostate, is a pathobiology that contributes to Lower Urinary Tract Symptoms (LUTS), a costly and critical medical problem for millions of aging men. This spectrum disorder encompasses symptoms such as weak stream, nocturia, incomplete emptying and intermittent urination, all of which are indicative of lower urinary tract dysfunction (LUTD). If left untreated or treated ineffectively, LUTD can progress to bladder dysfunction, which can lead to urinary retention, recurrent urinary tract infections, bladder calculi, and, eventually, renal impairment. Both the epithelial and stromal components of the prostate may exhibit enlargement associated with LUTS. Surgical ablation of prostate tissue and medical approaches including 5α-reductase inhibitors (finasteride, dutasteride), α-adrenergic receptor antagonists (doxazosin, tamsulosin), and PDE5 (sildenafil, tadalafil) inhibitors have been used to improve urinary flow (Laborde and McVary 2009, Strope et al. 2012).

More recently, expansion of the stromal component of the prostate concurrent with fibroblast activation and increased collagenization has been recognized as another pathobiology contributing to LUTS (Rodriguez-Nieves and Macoska 2013; Macoska 2019). However, the effects of beta-sitosterol/saw palmetto on the prostatic stroma in general or on fibroblast activation and collagen deposition has not been well elucidated. Therefore, the current study explored the effects of beta-sitosterol on prostate fibroblast viability, collagen production, and collagen cellular localization.

Methods

Cell culture

N1 cells were derived from prostate transition zone tissue explanted, grown as monolayer cells, and transduced with a recombinant LXSNE6E7 retrovirus (Begley et al. 2006). The cells were grown in 5% HIE culture media (Ham’s F-12, 5% Fetal Bovine Serum (FBS), Insulin [5 μg/mL], EGF [10 ng/mL], Hydrocortisone [1 μg/mL], Fungizone [0.5 μg/mL], Gentamicin [0.05 mg/mL]), Plasmocin [0.5 ug/mL]). Prior to treatment, cells were serum starved for 48 hr using Serum Free (SF) Ham’s F12, EGF [50 ng/mL], 0.1% BSA, Insulin [5 μg/mL], Transferrin [5 μg/mL], 50 μM sodium selenite, 10 uM 3,3′, 5-triiodo-L-thyronine, HEPES [10 mM], Hydrocortisone [1 μg/mL], Fungizone [0.5 μg/mL], Gentamicin [0.05 mg/mL]) Ethanolamine [5 mM] (HIE) media. pHPF primary human fibroblast cells were purchased at passage 3 from Lifeline Cell Technology and were maintained in FibroLife S2 Fibroblast Medium (Lifeline LL-0011). Prior to treatment, cells were starved for 48 h using SF HIE as described above. N1 cells were used between passages 46–54, and pHPF cells between passages 4–9. All treatments were performed in SF HIE media.

Western blot

Cells were cultured as above on 6 cm petri dishes and treated with vehicle + ethanol, 10 µM, 20 µM, 30 µM, 40 µM, or 50 µM of beta-sitosterol (Sigma S1270-10MG, dissolved in 100% ethanol), vehicle + sodium citrate, or 4 ng/mL TGFβ (Sigma T7039, dissolved in 20 mM 3.5pH Sodium Citrate) for 48 h. Intracellular protein was extracted, quantified, and prepared for electrophoresis as previously described (D’Arcy et al. 2022). Extracellular protein was collected from conditioned cell media and concentrated using Amicon Ultra 15 centrifugal filters (100kD NMWL, 4000 × g at 20 C for 20 min). Conditioned cell media concentrate was diluted 1:10 with ddH₂O then quantified and prepared for electrophoresis as previously described (Patalano et al., 2018). Membranes were blocked for one hour using a 5% milk in TBS-T solution followed by primary antibody incubation a 5% milk TBS-T solution with anti-Collagen 1 (#EPR7785) from Abcam, anti-Collagen 3 (#30565), and GAPDH (#2118) from Cell Signaling Technologies. All primary antibodies were incubated overnight at a 1:1000 concentration, except for GAPDH which was used at 1:5000, followed by 3 times washing with TBS-T. Secondary antibody incubations with IRDye 800CW Goat anti-rabbit IgG (LiCor 926-32211) and IRDye 680RD Goat anti-Rabbit IgG (LiCor 926-68071) were used at a 1:5000 concentration for 1 h at room temperature. Membranes were washed three times with TBS-T, imaged on LiCoR Odyssey CLx, then quantified and analyzed using ImageJ. Immunoblots shown are representative of triplicate experiments.

Immunofluorescence

Cells were cultured as above on 6-well plates and treated with vehicle + ethanol, 20uM beta-sitosterol, vehicle + sodium citrate, or 4 ng/mL TGFβ for 48 h. Cells were washed in warm PBS then fixed in 4% PFA (1 mL/well) for 10 min followed by permeabilization in .1% Triton x-100 (Fisher) for 15 min. Wells were then washed three times in warm PBS, blocked in 10% goat serum for 1 h, then incubated with biotin-conjugated anti-Collagen 1 (Rockland 600-406-103) overnight at 1:500 concentration. Secondary antibody incubation with rabbit IgG fluorescein-conjugated (R&D F0112) was used at 1:500 concentration in 0.1% BSA for 1 h. Cells were then counterstained with DAPI (Invitrogen D1306) at 1ug/mL concentration for 1 min and washed in PBS. Images were captured in 300 µL PBS per well on EVOS FL Auto Imaging System at 20× magnification. Captured images were overlayed and contrast was adjusted using ImageJ. For puncta/cell quantification, cells were identified by intact DAPI-stained nuclei and were counted within a 300uM radius circle from 4 areas per well. Collagen puncta were visually identified as isolated, high intensity green fluorescent signals within a cell border.

RNA purification and quantitative real time PCR (qRT-PCR)

Cells were cultured as described above on 6 cm petri dishes and treated with vehicle + ethanol, 10 µM, 20 µM, 30 µM, 40 µM, or 50 µM of beta-sitosterol, vehicle + sodium citrate, or 4 ng/mL TGFβ for 48 h. RNA was extracted using Trizol Reagent (Invitrogen), assessed for purity by A260/A280 ratio and quantified using a Nanodrop spectrophotometer. 1 µg of RNA was reversed transcribed using a High-Capacity cDNA Reverse Transcription Kit with RNase Inhibitor (Applied Biosystems 4374966). qRT-PCR was preformed using a QuantStudio 12K Flex Real-Time PCR system and software (Applied Biosystems) using TaqMan Universal PCR Master Mix (Applied Biosystems 4304437) using molecular probes Hs00164004_m1 COL1A1 Human, Hs1028956_m1 COL1A2, Hs00943809_m1 COL3A1 Human, and Hs02786624_g1 GAPDH Human from Thermo Scientific. Reactions were performed with technical duplicates including no template controls and amplification of GAPDH as an endogenous control transcript. Cycle threshold values were calculated for each target from each average experimental value and normalized to GAPDH. Fold change for beta-sitosterol treatments were calculated between the beta-sitosterol treatment value and the vehicle + ethanol vehicle value. Fold change for TGFβ treatment was calculated between the TGFβ treatment and vehicle + sodium citrate value. Data represents triplicate experiments.

ELISA assay

Cells were cultured as described above and treated with vehicle + ethanol, 20um or 40um beta-sitosterol, vehicle + sodium citrate, or 4 ng/mL TGFβ for 48 h. Cells were then counted and 40k cells/treatment were subjected to Cell Death Detection ELISA Plus (Millipore Sigma 11774425001) in technical triplicates. Data represents triplicate experiments.

Data processing

Data was processed, normalized, and visualized using the R programming language (version 4.2.2). Principal Component Analysis (PCA) was performed using the R function prcomp on the normalized data. Hierarchical clustering analysis was performed using the heatmap and hclust R function. Figures were generated using ggplot2.

Results

Supplement composition

An analysis of the sterol composition of several prostate health supplements examined showed that they contain the phytosterols campesterol, stigmasterol, brassicasterol, and beta-sitosterol, and that the herbal supplement Prostagenix clustered away from the other 5 supplements examined (Figure 1(A)). Heatmap visualization of the differences between the plant sterol contents of the same herbal supplements suggests that Prostagenix, Prost-T, and Prostvar Ultra share more identity with each other than with the other supplements, due to more similar compositions for beta-sitosterol (Figure 1(B)). The composition of Prostagenix showed that it contained the highest relative level of beta-sitosterol level (Figure 1(B)). Although these supplements varied widely from each other in terms of composition, beta-sitosterol is clearly the most prevalent sterol in all of those examined.

Figure 1. Sterol content of prostate health supplements. (A) Multidimensional scaling of the campesterol stigmasterol, beta-sitosterol, brassicasterol, and cholesterol content of several herbal supplements intended to enhance prostate health. Prostagenix is an outlier compared to the other herbal supplements due to its high beta-sitosterol content. (B) Heatmap visualization of the differences between the sterol contents of the same herbal supplements shown in Figure 1A. Prostagenix, Prost-T, and Prostvar Ultra share more identity with each other than with the other supplements, due to more similar compositions for beta-sitosterol.

Beta-sitosterol does not induce prostate fibroblast apoptosis

Though not uniformly reported, several publications showed or implied that beta-sitosterol induced apoptosis in epithelial prostate cells in vitro or tissues in vivo (Awad et al. 2001, Awad et al. 2005, Petrangeli et al. 2009, Cole et al. 2015, Pradhan et al. 2019, Sudeep et al. 2019). However, only one study examined the prostatic stroma in this regard. This study reported that patients treated with a saw palmetto extract demonstrated increased apoptosis in the epithelial and, to a lesser extent, in the stromal compartments of the prostate (Vacherot et al. 2000). Therefore, we performed studies to test whether beta-sitosterol would induce apoptosis in primary human prostate fibroblasts, pHPF cells. As shown in Figure 2, the percent apoptosis (compared to positive control) of cells treated with vehicle, 20 µM or 40 µM beta-sitosterol, or 4 ng/ml TGFβ, was low, and was not significantly different between treatments. This in vitro study suggests that that beta-sitosterol may be less efficient in inducing apoptosis in prostate stromal fibroblasts than prostate epithelial cells, perhaps consistent with the observations of Vacherot et al. (2000).

Figure 2. Beta-sitosterol does not induce prostate fibroblast apoptosis. pHPF cells were serum-starved for 48 h in SF HIE media as described above then treated with 20 μM or 40 μM beta-sitosterol or vehicle + ethanol, or with 4 ng/ml TGFβ or vehicle + sodium citrate for 48 h, then subjected to the Roche cell Death ELISA assay. The positive control is set at 100%. None of the treatments promoted apoptosis above 3%.

Beta-sitosterol does not induce the transcription of collagen-encoding genes

To determine whether beta-sitosterol affected collagen induction and accumulation, RNA was purified from both N1 immortalized and pHPF primary prostate stromal fibroblasts grown in in SFHIE media and treated with increasing doses of beta-sitosterol or 4 ng/ml TGFβ and examined by qRT-PCR for COL1A1, COL1A2, and COL3A1 transcript levels. Treatment with beta-sitosterol did not significantly alter collagen transcript expression levels in N1 cells (Figure 3(A)) or pHPF cells (Figure 3(B)) compared to vehicle. One or more of the 3 collagen genes were transcriptionally upregulated in response to treatment with TGFβ, a powerful pro-fibrotic, in N1 or pHPF cells. These results suggested that beta-sitosterol did not induce or repress collagen gene transcript expression in prostate stromal fibroblasts.

Figure 3. Beta-sitosterol does not induce the transcription of collagen-encoding genes. N1 cells (A) or pHPF cells (B) were grown in in serum-free media and treated with increasing doses of beta-sitosterol or TGFβ and their respective vehicles, then assessed by qRT-PCR. Data from treated cells was normalized to that of respective vehicles, which was set at 1-fold. Treatment with beta-sitosterol failed to elicit transcription of the COL1A1, COL1A2, or COL3A1 genes, whereas treatment with TGFβ significantly (p < .05) increased transcription of these same genes in N1 cells and of the COL1A2 gene in pHPF cells.

Beta-sitosterol inhibits the export of collagen 1 and collagen 3 proteins in N1, but not pHPF, cells

This study next examined collagen protein accumulation in vitro. For these studies, N1 or pHPF cells were grown for 24 h in 5% HIE media, starved in SF HIE media for 48 h, treated with increasing doses of beta-sitosterol or 4 ng/ml TGFβ in SF HIE media for 48 h, and lysed to collect protein (Begley et al. 2008). Previous studies have shown that N1 cells may be grown for 4 or more days in SF HIE with no changes in cell morphology, phenotype, or viability (D’Arcy et al., 2022). The studies conducted here demonstrated the same for the pHPF cells. The recovered protein was subjected to immunoblotting against antibodies for collagen 1, collagen 3, or GAPDH (as a loading control). The cell culture media was also collected, subjected to protein purification, and similarly immunoblotted against antibodies for the same proteins. As seen in Figure 4(A), the vehicle treated N1 cells produce a low basal level of collagen 1 protein, which they secrete into the media. This level actually diminished with increasing concentrations of beta-sitosterol in the media but increases intracellularly in the whole cell lysate. High levels of collagen 1 are evident both intra- and extracellularly in cells treated with TGFβ. Very little collagen 3 protein was evident in either the beta-sitosterol- or TGFβ- treated cells. Quantification of collagen 1 and 3 protein expression showed that the ratio of intracellular: extracellular collagen was significantly (p < .001) higher for N1 cells treated with 30, 40, or 50 µM beta-sitosterol (Figure 4(C)). pHPF cells also demonstrated diminishing levels of collagen 1 secretion into the media concomitant with increased beta-sitosterol dosage (Figure 4(B)). However, unlike N1 cells, pHPF cells did not demonstrate a parallel up-regulation of collagen 1 in the whole cell lysate. pHPF cells also demonstrated appreciable production of collage 3 protein in vehicle-, beta-sitosterol-, and TGFβ-treated cells (Figures 4(B,D)). The levels of intracellular and extracellular collagen 3 appeared to diminish with increasing dosage of beta-sitosterol (Figure 4(B)), but this did not achieve statistical significance (Figure 4(D)). These results suggest that, consistent with the qRT-PCR data, beta-sitosterol was not affecting the overall levels of collagen 1 or 3 produced by the cells. However, beta-sitosterol did appear to interfere with the secretion of collagen 1 by the N1 cells into the media such that collagen1 protein was sequestering inside the cells and less was being secreted into the media at higher beta-sitosterol dosages.

Figure 4. Beta-sitosterol inhibits the export of collagen 1 and collagen 3 proteins in N1, but not pHPF, cells. N1 or pHPF cells were grown for 24 h in 5% HIE media, starved in SF HIE media for 48 h, then treated with increasing doses of beta-sitosterol or 4 ng/ml TGFβ for 48 h in SF HIE media. Protein recovered from the cells and media was subjected to immunoblotting against antibodies for collagen 1 (COL1), collagen 3 (COL3), or GAPDH (as a loading control). (A,C) N1 cells produce a low basal level of collagen 1 protein, which is decreasingly secreted into the media at higher beta-sitosterol concentration (A) resulting in a significantly (p < .001) higher intracellular:Extracellular ratio of collagen 1 with increasing beta-sitosterol dosage compared to vehicle (C). (B,D) pHPF cells produce high levels of both collagen 1 and collagen 3 (B), and secretion of both proteins into the media is unaffected by beta-sitosterol treatment (D).

Immunofluorescence was next used to further examine the possibility that beta-sitosterol mediated sequestration of collagen 1 protein in N1 cells. Although collagen distribution appeared similar between ethanol vehicle-, sodium citrate vehicle-, and TGFβ-treated cells (Figure 5(A,C,D)), beta-sitosterol treated cells demonstrated a high number of internal fluorescently stained collagen protein in a ‘punctate’ type pattern (Figure 5(B)). To evaluate this quantitatively, puncta were counted in cells treated with ethanol vehicle (N = 169), 20 µM beta-sitosterol (N = 146), sodium citrate vehicle (N = 175) or TGFβ (N = 191). 14, 17, 16, and 22 puncta were identified and counted in ethanol vehicle-, 20 µM beta-sitosterol-, sodium citrate vehicle-, and TGFβ-treated cells, respectively. The ratio of puncta/cell for beta-sitosterol treated cells was significantly higher than that for ethanol vehicle treated cells (p < .001) (Figure 5(C)). Cells treated with TGFβ or sodium citrate exhibited a ratio of puncta/cell like that of ethanol vehicle. This data suggested that beta-sitosterol treatment was associated with collagen sequestration within the cells, consistent with observations from the immunoblot studies.

Figure 5. Beta-sitosterol treatment is associated with collagen sequestration in N1 cells. N1 cells treated with vehicle + ethanol, 20 μM beta-sitosterol, vehicle + sodium citrate or 4 ng/ml TGFβ were subjected to immunofluorescence for collagen 1. Collagen 1 distribution appeared similar between ethanol vehicle-, sodium citrate vehicle-, and TGFβ-treated cells but beta-sitosterol treated cells demonstrated a high number of internal fluorescently stained collagen protein in a ‘punctate’ type pattern (A). the puncta/cell ratio was significantly (p < .001) higher in beta-sitosterol treated than other treated cells (B). puncta were counted in cells treated with ethanol vehicle (N = 169), 20 µM beta-sitosterol (N = 146), sodium citrate vehicle (N = 175) or TGFβ (N = 191). 14, 17, 16, and 22 puncta were identified and counted in ethanol vehicle-, 20 µM beta-sitosterol-, sodium citrate vehicle-, and TGFβ-treated cells, respectively.

Discussion

The first primary finding of this study is that beta-sitosterol is nontoxic and does not induce apoptosis in primary human prostate stromal fibroblasts. Previous studies in prostate cancer epithelial cells showed that beta-sitosterol was largely both toxic and pro-apoptotic in those cells using either lower or higher levels of phytosterol than those used in this study (von Holtz et al. 1998, Awad et al. 2001, Awad et al. 2005, Petrangeli et al. 2009, Pradhan et al. 2019). In vivo studies of prostate cancer epithelial cells implanted subcutaneously into nude mice suggested that the resulting tumors grew less well and up-regulated caspase 6, suggestive of apoptosis (Cole et al. 2015). Beta-sitosterol mediated cellular apoptosis has also been studied in a testosterone-induced Wistar rat model of BPH. Rats treated with beta-sitosterol exhibited significant decreases in prostate weight to body weight ratio and increased epithelial compartment apoptosis (Sudeep et al. 2019). Taken together, these in vitro and in vivo studies are consistent in their reports of the pro-apoptotic effects of beta-sitosterol on prostatic epithelium.

Studies of human prostate tissues also provide data regarding the pro-apoptotic properties of beta-sitosterol/saw palmetto on both the epithelial and stromal compartments. Vacherot et al. reported that both epithelial and stromal prostatic tissues from men treated with saw palmetto extract exhibited significantly increased apoptosis and reduced proliferation (Vacherot et al. 2000). However, a contemporaneous study by Marks et al. reported reduced epithelial, but not stromal, volume in men with BPH treated with finasteride, a pharmaceutical aza-steroid commonly used to treat BPH (Marks et al. 2000). Comparison of these two studies suggest that either aza-steroids in general exhibit variability in pro-apoptotic effects on prostatic stroma, or that pharmaceutical-grade and supplement-grade aza-steroids exert different effects on prostatic stroma.

Taken together, the in vitro, in vivo, and human tissue studies cited above suggest that beta-sitosterol induces apoptosis in the epithelial, but to a lesser extent or not at all, in the stromal compartment of the prostate. This latter finding is consistent with the data presented here, that beta-sitosterol did not elicit apoptosis of primary human prostate stromal cells.

Another finding of the current study is that beta-sitosterol can inhibit procollagen secretion by prostate stromal fibroblasts. Due to their large size, procollagen proteins undergo a specific mechanism for secretion into the extracellular space. Once translated and properly folded, pro-collagen proteins are sequestered into COPII vesicles that form from the endoplasmic reticulum (ER) membrane. These vesicles transport large cargo, such as procollagens, to the Golgi apparatus. There, the procollagens are enclosed by endosomes that bud off from the Golgi and travel to the plasma membrane, where they fuse and undergo an exocytic process to release them into the extracellular space, where they undergo further post-translational modification to mature collagen (Sato and Nakano 2007). Our group has previously shown that two pro-fibrotic proteins, CXCL12 and TGFβ, exert opposite effects on COPII vesicle formation and procollagen secretion. Activation of the TGFβ/TGFβR axis increases procollagen production but concurrently inhibits COPII vesicle formation by prostate fibroblasts, thereby inhibiting procollagen secretion. In contrast, activation of the CXCL12/CXCR4 axis increases both procollagen production and secretion by prostate fibroblasts, the latter by upregulating the expression of COPII vesicle components (Patalano et al. 2018). Although beta-sitosterol did not increase or decrease the amount of collagen produced by either the immortalized or primary prostate fibroblasts, it did alter the distribution of collagen in N1 cells, as evidenced by the significant increase in the ratio of intracellular: extracellular collagen observed in the western blot studies. The immunofluorescence studies also suggest that collagen protein may be sequestered within beta-sitosterol treated N1 cells, as these cells demonstrated discrete intracellular collagen-staining ‘puncta’ in the beta-sitosterol treated N1 cells. However, the low resolution of the images produced by the immunofluorescence studies do not definitively support this interpretation of the images. Intracellular sequestration of collagen in beta-sitosterol treated pHPF primary human fibroblast cells was not observed either by western blot or immunofluorescence studies. Taken together, these findings suggest that, under some conditions or circumstances, beta-sitosterol may repress collagen export into the extracellular space.

What circumstances or conditions, then, may differentially affect the beta-sitosterol export of collagen into the extracellular space in N1 and pHPF cells? The N1 cells used in this study were grown from stromal nodular tissue from the prostate of a 71 year old man with LUTS and immortalized with a with a recombinant LXSNE6E7 retrovirus (Begley et al. 2006). They closely resemble primary fibroblast cells histologically and phenotypically (Gharaee-Kermani et al. 2012; Gharaee-Kermani et al. 2016). Unlike prostate epithelial cells immortalized by transduction using the same methods, N1 cells exhibit no chromosomal alterations. However, like aging primary human prostate stromal fibroblasts, N1 cells secrete many inflammatory mediators, e.g. cytokines, chemokines, and interleukins, which can act as both autocrine and paracrine growth factors that promote proliferation, gene transcription, and other key cellular activities (Begley et al. 2008). This distinguishes N1 cells from primary prostate stromal fibroblasts, which don’t exhibit this secretory phenotype unless induced to undergo senescence (Coppe et al. 2008). The pHPF cells used in this study were grown from the prostate of a young adult, are not immortalized, and stop proliferating after passage 16. pHPF cells, like other primary cells in culture, will senesce as they reach the end of their replicative capacity. However, for the studies reported here, pHPF cells were used at passages 4–9, well before they would undergo senescence and stop proliferating. Therefore, it is highly feasible that the differences in age-related phenotypes exhibited by the N1 and pHPF cells, respectively, may have influenced the cellular response of these cells to beta-sitosterol.

Limitations of this study includes the lack of available human prostate fibroblast cell lines, which restricted the number of such cell lines that could be tested. Another limitation is the variability of beta-sitosterol content between prostate health supplements, which likely limits the generalization of the findings presented here to the extent of effects individuals taking these supplements may experience. A third limitation is the paucity of existing data on the effects of beta-sitosterol on human prostate fibroblasts in vitro or prostatic stromal tissues in rodent models or human specimens. This prevented a comprehensive evaluation of the results of these studies in comparison to existing data. However, it also raised the possibility of resolving this paucity by perhaps examining biopsy tissues taken from men for the purposes of prostate cancer diagnosis. The majority of such biopsies do not harbor tumor foci, and these tumor-free tissues could be examined from men with a history of taking/not taking prostate health supplements containing beta-sitosterol to help resolve whether beta-sitosterol exerts similar or dissimilar effects on human prostatic stroma. It may also help resolve the proposal that beta-sitosterol may affect collagen distribution in the prostatic stroma in an age-dependent manner, which is important in the context that the use of prostate health supplements is predominantly by middle aged and older men similar in age to the patient from which the N1 cells were isolated and cultured.

In conclusion, this study reports that beta-sitosterol, a large component of many prostate health supplements, is nontoxic to prostatic stroma and does not stimulate collagen production by immortalized or primary prostate stromal fibroblasts. However, beta-sitosterol may affect collagen distribution and secretion by prostate stromal fibroblasts that sequesters collagen within the cell, possibly in an age-dependent manner. This is a significant finding as prostate health supplements are used predominantly middle aged and older men.

Acknowledgements

The authors would like to acknowledge the Alton J. Brann Foundation Endowment (J.A.M.) for supporting this work and for services provided by the CPCT Genomics Core supported by NIH/NCI U54 CA156734 (J.Am).

Disclosure statement

The authors declare no conflicts of interest. The authors alone are responsible for the content and writing of the article.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.

Notes on contributors

Quentin D’Arcy

Quentin D’Arcy, B.S., is a Biology pre-doctoral student at the University of Massachusetts Boston. He earned a B.S. in Biology with Honors from the University of Massachusetts Amherst and spent several years at the Massachusetts Department of Health prior to joining the Macoska laboratory in 2020. His dissertation research focuses on mechanisms promoting fibrosis in the lower urinary tract including those involving pro-fibrotic cytokines, health supplements, and regulatory RNAs. He plans to pursue a position in academia or industry in the Boston/Cambridge area upon completion of his graduate studies.

Marissa Sarna-McCarthy

Marissa Sarna-McCarthy, B.S., is a Biology pre-doctoral student at the University of Massachusetts Boston. She earned a B.S. in Biology from Bay Path University while working part time as a chemistry lab preparatory technician at the Baystate Medical Center prior to joining the Macoska laboratory in 2020. Her dissertation research focuses on cytokine-induced anti-apoptotic mechanisms that increase the survival of activated fibroblasts and myofibroblasts, hence, promote collagen production and fibrosis in the lower urinary tract. She plans to remain in Massachusetts and pursue a position in academia or industry upon completion of his graduate studies.

Delaney Bowen

Delaney Bowen, B.S., is a Biology Master’s student and the President of the Graduate Student Association at the University of Massachusetts Boston. She earned a B.S. in Biology from the University of Massachusetts Boston in 2021 and joined the Macoska Laboratory the following year. Her thesis research focuses on the role(s) of regulatory RNAs, particularly miRNAs, in promoting fibrosis in the lower urinary tract. Upon graduation, she plans to attend medical school to become a physician scientist.

Fidias O. Soto

Fidias Soto, B.S., graduated from the University of Massachusetts Boston with a B.S. in Biology with Honors and Distinction in 2023. As an undergraduate student, he received the College of Science and Mathematics Undergraduate Research Fellowship, the McCone Award, and Biology Department Research Award. He is now pursuing a patient-centered position in the Harvard health and hospital system and plans to attend medical school to become a physician scientist.

Kourosh Zarringhalam

Kourosh Zarringhalam, Ph.D., is an Associate Professor in the Department of Mathematics at the University of Massachusetts Boston. His research utilizes approaches from applied mathematics, data sciences, and computational molecular biology to develop mathematical models and algorithms for parasitology and cancer applications.

Jill A. Macoska

Jill Macoska, Ph.D., is the Distinguished University Professor of Science and Mathematics, Alton J. Brann Endowed Chair, and Director of the Center for Personalized Cancer Therapy and Genomics Core at the University of Massachusetts Boston. She has led peer-reviewed and NIH-funded research for the past 30 years focused on elucidating the molecular genetic alterations and dysfunctional inter- and intra- cellular signaling mechanisms that promote urinary tract (kidney, bladder, prostate) pathobiology. This work has resulted in >100 peer-reviewed publications, many of which were co-authored by the >50 post-doctoral, pre-doctoral, or undergraduate trainees she has mentored in her laboratory. Notably, the Macoska laboratory established the concept of peri-urethral fibrosis as a pathobiology promoting male lower urinary dysfunction (LUTD), a finding that has since become a major driving force towards understanding and treating LUTD.