Stock solution of carnosic acid and rosemary extract was prepared by dissolving it in DMSO. Working standard solutions ranging from 2.5 µg/ml to 6.5 µg/ml for carnosic acid were diluted with acetonitrile. Rosemary extract was diluted in acetonitrile at a final concentration of 10 µg/ml (i.e. 4 µg/ml of carnosic acid). Each standard reaction contained rosmarinic acid as an internal standard at a ratio of 1∶1 and at a final concentration of 5 µg/ml. The samples were transferred to auto sampler vials and 20 µL of the sample was injected into the LC-MS/MS system for analysis. The standard curve of carnosic acid had a R² value of 0.9778.

The LC-MS/MS system consisted of an Agilent HPLC auto sampler, Agilent quaternary pump, and an API-3200 Qtrap mass spectrometer with a turbo-ion spray source (Applied Biosystem, Foster City, CA). The Q-trap was equipped with an electrospray ionization (ESI) source and operated in the negative-ion mode. Total eluent flow from the LC was directed to the turbo ion spray source without splitting. A valco valve was used to divert the first two minutes of the eluent to waste. Needle voltage was 4.5 kV, turbo ion spray heater temperature 500°C, (GS1) nebulizer gas (nitrogen) 50 psi, and (GS2) turbo heater gas (nitrogen) 50 psi. Curtain gas (nitrogen) was set at 10, and collision gas (CAD, nitrogen) pressure in the collision cell was set at medium. The optimum values for collision energy (CE), declustering potential (DP), and cell entrance potential (CEP) were optimized individually for each compound. Tandem mass spectrometry (MS/MS) optimization was established by directly infusing 100 µg/mL of analyte in acetonitrile.

The instrument was operated in unit resolution mode with the peak width (full width at half-maximum, FWHM) set to 0.7 m/z both at Q1 and Q3. The selected reaction monitoring scheme followed transitions of the precursor to selected product ions with the following values: m/z 331–287 for carnosic acid and m/z 359/161 for rosmarinic acid. Chromatography was performed on a Nova-Pak C18, 4.6 mm×150 mm, 4.0 µm analytical column (Waters, Milford, MA, USA). The mobile phase consisting of 10% ACN containing 0.1% formic acid in water and ACN (0.1%formic acid) mixed with methanol at 1∶1 ratio was delivered at a flow rate of 0.6 mL/min (injection volume 20 µL). The gradient program was as follows: 70–95% B at 0–7 min, hold for 1 min, followed by a 95–70% B from 7–8 min, followed by a re-equilibration at 70% B from 8–15 min. Data was analyzed using the Analyst Software Version 1.4.2 (Framingham, ME, USA).

22Rv1 and LNCaP human epithelial prostate cancer cells were obtained from American Type Culture Collection (ATCC, Manassas, VA). The cells were cultured in RPMI 1640 medium (Cellgro) containing 10% FBS (Atlanta Biologicals), 100 U/ml penicillin and 100 µg/ml streptomycin (Cellgro) at 37° C and 5% CO₂. For cell treatments an oil soluble rosemary extract standardized to >40% carnosic acid was used. Cells were treated with rosemary extract at concentrations ranging from 0–70 µg/ml for 24 hrs and then harvested for different experiments as described previously [12].

Primary prostatic epithelial cells (PrE) were established from radical prostatectomy tissue at the University of Illinois at Chicago Medical Center as described previously [13]. Fresh tissue from the peripheral zone was selected by a pathologist according to an IRB protocol approved by the University of Illinois at Chicago IRB. All samples were obtained following written informed consent from the donors. Briefly, the tissue was digested in collagenase, and plated on collagen-coated dishes in PrEGM media (Lonza, Walkersville, MD) for epithelial cell outgrowth. All cells were used at secondary passage and ∼70 % confluency (cell density). PrECs were treated with 40 µg/ml of standardized rosemary extract for 24 hr followed by downstream experiments.

Cell viability was determined by 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) assay as described previously [14].

BrdU Cell Proliferation Assay Kit was used to assess the proliferation of 22Rv1 and LNCaP cells upon treatment with rosemary extract as per the manufacturer’s instructions. Briefly, cells were seeded at a concentration of 6,000 cells/well in a 96-well plate and allowed to reach 70% confluence. Cells were treated with 0 and 50 µg/ml of rosemary extract and prepared according to the protocol for analysis.

The APO-DIRECT Kit was used for measuring cell cycle by flow cytometry. Briefly, 22Rv1 and LNCaP cells were cultured in medium containing 1% FBS for 24 hr, and then treated with DMSO or 40 µg/ml rosemary extract for 24 hr. Cell fixation and staining were performed according to manufacturer’s manual. Cells were then subjected to flow cytometry for cell cycle analysis.

Lysates were prepared with the same lysis buffer used for Western blotting. 10–40 micrograms of lysates were analyzed using PathScan cleaved caspase-3 sandwich ELISA following the manufacturer's instructions. Lysates were mixed with 50 µL of sample diluent at a ratio of 1∶1 and incubated in antibody-coated microwell strips. One hundred microliters of cleaved caspase-3 detection antibodies were added to each well. Binding was detected with 100 µL of horseradish peroxidase–linked streptavidin antibody and 100 µL of 3,3′,5,5′-tetramethylbenzidine substrate solution. The colored reaction product was measured in a standard ELISA plate reader at 450 nm.

22Rv1 and LNCaP cells were cultured in chamber slides at a concentration of 0.1×10⁶ cells/chamber and allowed to grow to 70% confluence. Cells were then treated with 0 and 50 µg/ml rosemary extract for 24hrs. Apoptosis was detected by TUNEL assay using the DeadEnd Fluorometric TUNEL System according to the manufacturer’s instructions followed by counterstaining with DAPI.

The human PSA ELISA kit was used for the quantitative determination of PSA levels in cell culture media per the manufacturer's protocol. Following collection of media, the samples were stored at −20°C until assayed for secreted PSA. A standard curve was plotted using available PSA standards (0, 2, 10, 20, 40, and 80 ng/mL) to quantify individual samples. The equation of the standard PSA plot was y = 0.0383x + 0.0718 with a R² value of 0.9953.

Control as well as rosemary extract treated cells were collected and whole cell lysates prepared. Protein concentrations in supernatants were measured using a BCA protein assay kit (Thermo Scientific). 10–40 µg protein was separated by SDS gel electrophoresis and transferred on nitrocellulose membrane (Millipore). The blots were blocked with 5% milk in TBS/Tween 20 for 1 hr at room temperature. Membranes were then incubated with the following primary antibodies, androgen receptor (AR), Bax, PERK, CHOP, IRE1α, BiP and caspase-4 overnight at 4°C and then with the corresponding HRP-labeled secondary antibodies for 1 hr at room temperature. The blots were developed with Super Signal West Femto chemiluminesent substrate (Thermo Scientific) and visualized using the FlourChemE Imager. All blots were probed with β-actin antibody to confirm equal loading of proteins.

22Rv1 and LNCaP cells were treated with 40 µg/ml rosemary extract or DMSO for 24 hrs. Total RNA was purified from control and treated cells. One step RT-PCR was performed according to the manufacturer’s instruction. One microgram of total RNA was used as template for each reaction. House-keeping gene GAPDH was used as an internal control. Primer sequences were: GAPDH forward 5′-ACC ACA GTC CAT GCC ATC AC-3′, GAPDH reverse TCC ACC ACC CTG TTG CTG TA-3′, XBP-1-A forward 5′-TTA CGA GAG AAA ACT CAT GGC C-3′, XBP-1-A reverse 5′-GGG TCC AAG TTG TCC AGA ATG C-3′.

Cells were cultured in 4-chamber slides (BD Biosciences) and treated with 0 and 50 µg/ml rosemary extract for 24hrs. After treatment, cells were washed once in PBS, fixed in 4% PFA for 15 mins and permeabilized in 0.2% Triton X-100 for 30 mins on ice. After blocking with 5% BSA in PBS for 1 hr, cells were incubated with CHOP antibody for 2 hrs followed by 1 hr treatment with corresponding fluorescent labeled secondary antibody. The slides were mounted and counterstained using Vectashield mounting medium containing DAPI (Vector labs). Cell images were taken at 40X magnification using Nikon microscope.

13nM control and CHOP siRNA and 50nM control and BiP siRNA were transfected in 22Rv1 and LNCaP cells using the TransIT-siQuest reagent for 24 hrs. After siRNA transfection, cells were treated for another 24 hrs with 0 and 50 µg/ml rosemary extract and lysates prepared thereafter.

All animal experiments were performed in accordance with the guidelines approved by the Animal Care and Use Committee of the University of Illinois at Chicago. The protocol was approved by the animal care committee at the University of Illinois at Chicago (Protocol Number: ACC-11-019). All animals were maintained under pathogen free conditions on a 12-hr light/dark cycle and fed with food and water ad libitum. 5 week old, 16 athymic J: NU outbred mice (Jackson Labs) were subcutaneously injected with 22Rv1 cells in the right and left flanks at a density of 1×10⁶cells/100 µl of RPMI + matrigel media. Mice were then divided into two groups of 8 mice each and the control group animals received 100 µl olive oil daily by oral gavage. The other 8 mice in the treated group received rosemary extract (100 mg/kg) dissolved in olive oil orally for 22 days. Body weights and tumor volumes were measured two times a week as described before [15]. Animal tissues were collected at the end of the study and stored at –80°C. Tissue lysates were then prepared in 1X RIPA buffer (Cell Signaling Technology) and its protein content measured using BCA assay kit (Thermo Scientific).Lysates were subjected to immunoblot analysis and probed for AR, PSA and CHOP.

All statistical analysis was done using GraphPad QuickCals software. Statistical comparisons between control and treated groups were performed by Student’s t test. All statistical tests were two- sided and P<0.05 was considered statistically significant.

For this study, we used an oil-soluble rosemary extract preparation containing carnosic acid and other phenolic diterpenes. Since carnosic acid accounts for majority of the antioxidant activity of rosemary extract we standardized the extract to carnosic acid using a LC/MS/MS method. The representative LC/MS chromatograms of carnosic acid and rosmarinic acid as internal standard are shown in Figure 1A. Figure 1B represents carnosic acid peak fragmented as a single ion at mass 331. LC/MS data showed 43% carnosic acid present in rosemary extract. The chemical structures of carnosic acid and rosmarinic acid are depicted in Figure 1C.

To determine the antiproliferative activity of rosemary extract we treated prostate cancer cells, 22Rv1 and LNCaP, with 0 and 50 µg/ml of rosemary extract for 24 hrs. As shown in Figure 2A, rosemary extract decreased proliferation of both 22Rv1 and LNCaP cells by 76.5 and 94.6% respectively. Likewise, we also observed a dose-dependent decrease in cell viability of 22Rv1 and LNCaP cells upon treatment with increasing concentrations of rosemary extract after 48hrs (Figure 2B). The IC₅₀ values observed for 22Rv1 and LNCaP cells are 13.3 µg/ml and 27 µg/ml respectively.

Treatment of 22Rv1 cells with 40 µg/ml rosemary extract for 24 hrs resulted in accumulation of cells in the G2 phase of the cell cycle whereas treatment of LNCaP cells resulted in both G1 and G2 cell cycle arrest as shown in Figure 2C.

To examine if rosemary extract promotes apoptosis we looked at the expression levels of pro-apoptotic protein, Bax, by Western blotting. With increasing concentrations of rosemary extract we saw an increase in expression of Bax in 22Rv1 and LNCaP cells (Figure 3A). To further confirm if rosemary extract induces apoptosis we performed a TUNEL assay on 22Rv1 and LNCaP cells. As shown in Figure 3B, 22Rv1 and LNCaP cells displayed green fluorescence for TUNEL staining upon treatment with rosemary extract at 50 µg/ml for 24 hrs compared to untreated control cells. Etoposide was used as a positive control for visualizing apoptosis. Additionally, we also detected increased production of cleaved caspase-3 in rosemary extract treated cells compared to untreated controls (Figure 3C). We observed a 20 fold and 6.6 fold increase in cleaved caspase 3 production in 50 µg/ml rosemary extract treated 22Rv1 and LNCaP cells respectively, compared to untreated control cells.

The androgen receptor has long been a target for prostate cancer. Hence we evaluated the expression of AR following treatment with rosemary extract in 22Rv1 and LNCaP cells. We observed a dose dependent decrease in expression of AR (Figure 4A) in prostate cancer cells with concentrations as low as 40 µg/ml and 20 µg/ml in 22Rv1 cells and LNCaP cells, respectively. Along with decreased AR expression we also observed 6 fold decrease in PSA protein secreted in culture media of LNCaP cells after treatment with 50 µg/ml of rosemary extract for 24 hrs as measured by PSA ELISA (Figure 4B)

Next, we evaluated the expression levels of ER stress proteins PERK, IRE1α, XBP-1, BiP, CHOP and caspase-4 upon treatment with 0–50 µg/ml of rosemary extract in both 22Rv1 and LNCaP cells. As shown in Figure 5A, treatment with rosemary extract, resulted in an increase in PERK, often the first ER stress protein to be modulated. An increase in protein expression of IRE1α was observed at 20 µM. A significant increase in BiP, an ER stress chaperone protein, and CHOP was also observed upon treatment with 20 µM of rosemary extract (Figure 5A). In response to ER stress, activated IRE1α cleaves XBP1 mRNA to remove a 26-nucleotide intron and generate a translational frameshift mutation. Thus, using RT-PCR we tested for XBP1 splicing using primers specific for the 26-nucleotide intron. As shown in Figure 5B, rosemary extract treatment induced splicing of XBP-1 mRNA in both prostate cancer cell lines compared to untreated controls. GAPDH was used as an internal control. All upstream signals of UPR ultimately lead to activation of caspase-4 during ER stress. Our results show an increase in caspase-4 expression in both 22Rv1 and LNCaP cells upon treatment with rosemary extract (Figure 5C).

Next, we evaluated the ability of rosemary extract to modulate the ER stress protein CHOP/GADD153 in vitro. 22Rv1 and LNCaP cells were grown in chamber slides for 24 hours and treated with rosemary extract for 24 hours and prepared as described in Materials and Methods. We observed an increase in overall protein expression of CHOP (Figure 5D). This can be seen by the increased red fluorescence in CHOP panels along with the merged panel containing the counterstain with 4′,6-diamidino-2-phenylindole (DAPI). Being that CHOP is a transcription factor localized to the nucleus is consistent with its role in the unfolded protein response pathway. Observations were consistent in both 22Rv1 and LNCaP cells.

Our next objective was to determine the potential of rosemary extract to promote ER stress in non-tumorigenic cancer cells. Prostate epithelial cells were isolated from benign peripheral zone in patients undergoing radical prostatectomy. Interestingly, we observed rosemary extract was unable to increase the expression of the critical ER stress proteins, PERK, in benign human prostate epithelial cells (Figure 5E). Interestingly, a decrease in PERK expression in prostate epithelial cells was observed following treatment with rosemary extract.

To understand the role of the ER stress chaperone protein BiP in AR degradation we treated 22Rv1 and LNCaP cells with rosemary extract with or without BiP-siRNA. As seen in Figure 6A, expression of BiP protein is increased in rosemary treated cells along with a decrease in AR expression. However, when cells were treated with a combination of BiP-siRNA and rosemary extract, androgen receptor degradation was reversed. These results suggest that BiP may have a critical role in regulating AR degradation (Figure 6A).

Next, we evaluated the role of BiP in regulating caspase-3 activation in 22Rv1 and LNCaP cells. As expected rosemary extract treatment induced activation of caspase-3 in both 22Rv1 and LNCaP cells. However, when cells were treated with combination of BiP siRNA and rosemary extract the activation of caspase-3 was decreased in both 22Rv1 and LNCaP cells (Figure. 6B and 6C respectively). These results suggest that rosemary extract induced BiP expression is essential for apoptosis.

22Rv1 and LNCaP cells were further evaluated to understand if CHOP is a key regulator in AR degradation. 22Rv1 and LNCaP cells were treated with rosemary extract with or without CHOP-siRNA. As expected a significant increase in CHOP expression and decrease in AR expression was observed following treatment with rosemary extract. However, when cells were treated with a combination of CHOP-siRNA and rosemary extract androgen receptor degradation was reversed. These results suggest that CHOP may have a critical role in regulating AR degradation (Figure 6D).

Next, we evaluated the role of CHOP in regulating caspase-3 activation in 22Rv1 and LNCaP cells. We observed rosemary extract to increase activation of caspase-3 as observed earlier. Interestingly, when cells were treated with CHOP siRNA and rosemary extract the activation of caspase-3 was decreased in 22Rv1 cells (Figure 6E) and LNCaP cells (Figure 6F). These results suggest that rosemary extract induced CHOP expression is essential for apoptosis.

Athymic nude mice were administered either olive oil alone or rosemary extract dissolved in olive oil to determine the efficacy of rosemary extract on progression of prostate cancer in vivo. As evident from their body weight measurements, mice tolerated rosemary extract well for 22 days (Figure 7A). Development of tumors was observed at day 14 in both groups of mice, with a smaller volume observed in mice treated with rosemary extract as shown in Figure. 7B. At day 21, tumors in the treated group measured 695 mm³ whereas tumors from control mice measured 1295 mm³ depicting a 46% reduction in tumor size in rosemary extract treated mice compared to control animals (Figure. 7B). A representative image of control and rosemary extract treated mice along with pictures showing different dimension of tumors from each group is shown in Figure 7C. We also analyzed mouse tissue lysates using Western blotting for expression of AR, PSA and CHOP. As shown in Figure 7D, expression of AR and PSA is decreased and that of CHOP is increased in rosemary extract treated tissue lysates compared to lysates from control group animals.