Piperlongumine

Piperlongumine inhibits the proliferation, migration and invasion of fibroblast-like synoviocytes from patients with rheumatoid arthritis

Abstract
Objectives Recent studies have indicated that piperlon- gumine (PLM) may exert anti-inflammatory effects. In the present study, we determined the effect of PLM on the prolif- eration, apoptosis, migration and invasion of fibroblast-like synoviocytes (FLS) from patients with rheumatoid arthritis (RA) (referred to herein as RA FLS). We further explored the mechanisms by which the studied compound inhibits the functions of RA FLS.Methods RA FLS viability and apoptosis were tested using MTT and Annexin V/PI assays, respectively. We performed an EDU assay to examine the proliferation of RA FLS. The migration and invasion of these cells were measured using a transwell chamber method and wound closure assay. The MMP-1, MMP-3, and MMP-13 levels in the culture super- natants of RA FLS were detected using a Luminex Assay kit. The intracellular ROS levels were detected using DCFH-DA. The expression levels of signal transduction proteins were measured using western blot.Results We found that PLM induced apoptosis in RA FLS at concentrations of 15 and 20 μM. The proliferation of RA FLS was downregulated by PLM at concentrations of 1, 5 and 10 μM. Migration and invasion of RA FLS were reduced by PLM at concentrations of 1, 5 and 10 μM. PLM also inhibited cytoskeletal reorganization in migrating RA FLS and decreased TNF-α-induced intracellular ROS production. Moreover, we demonstrated the inhibitory effect of PLM on activation of the p38, JNK, NF-κB and STAT3 pathways. Conclusions Our findings suggest that PLM can inhibit proliferation, migration and invasion of RA FLS. Moreover, these data suggests that PLM might have therapeutic potential for the treatment of RA.

Introduction
Rheumatoid arthritis (RA) is a chronic autoimmune disease that requires long-term treatment. Certain biological agents bring hope to RA patients, especially those who exhibit inadequate responses to conventional disease-modifying antirheumatic drugs (DMARDs), including methotrexate, leflunomide, and sulfasalazine. However, the insufficient efficacy, unpredictable safety and high cost of promising biological agents prevent some patients from completing treatment [1]. It is, therefore, still necessary to explore alter- native therapeutic agents for RA.Fibroblast-like synoviocytes (FLS) from patients with RA exhibit “tumor-like” properties, activation and other aggres- sive behaviors, such as hyperproliferation, decreased apopto- sis, aberrant migration and invasion. These cells play a key indicates that targeting FLS-mediated inflammation and invasion may be promising for the treatment of RA [4].Piperlongumine is a natural bioactive small molecule isolated from the Piper longum Linn plant. This compound is easily available, inexpensive, and has therapeutic effects against cancer, heart disease, intestinal diseases, diabetes, obesity, joint pain and other conditions in Chinese Herbal and Indian Ayurvedic medicine [6, 7]. As a primary constit- uent of Piper longum Linn, PLM has been reported to exert effects against cancer, platelet aggregation, atherosclerosis, diabetes and inflammation [6, 7]. Mechanistic investigations have found that PLM selectively augments ROS levels in tumor cells and regulates downstream signaling, including NF-κB, MAPK, and STAT3 pathways, in several other cell types [8, 9]. However, the role of PLM in regulating RA FLS functions has not been defined. In the present study, we evaluated the effects of PLM on apoptosis, prolifera- tion, migration and invasion of RA FLS, then subsequently explored the mechanisms how this agent works on RA FLS function.

Piperlongumine was obtained from Cayman Chemical (#20069-09-4, Michigan, USA). Recombinant human TNF-α was purchased from R&D Systems (#210-TA-CF, Minneapolis, MN, USA). Methyl thiazolyl tetrazolium (MTT) (#M2128), collagenase I (#1148089), crystal vio- let solution (#V5265), phalloidin (#P1951), 2 × loading buffer (#S3401), anti-β-actin antibody (#A1978), annexin V-fluorescein isothiocyanate (Annexin V-FITC), a pro- pidium iodide (PI) staining assay kit (Annexin V-FITC/ PI kit) (#APOAF), and gelatin (#G7041) were purchased from Sigma (Saint Louis, MO, USA). DAPI was purchased from Molecular Probes (#D1306, Oregon, USA). The Cell- Light™ EdU Apollo®567 In Vitro Imaging Kit was pur- chased from RiboBio (#C10310-1, Guangzhou, China). 2′,7′-dichlorodihydro-fluorescein diacetate (DCFH-DA) was purchased from Beyotime (#S0033, Haimen, China). Anti- rabbit IgG (#7074S) and anti-mouse IgG (#7076), p-ERK (#4370S)/ERK (#4695S), p-JNK (#4668S)/JNK (#9252S), p-p38 (#9215S)/p38 (#9212S), p-STAT3 (#9134S)/STAT3(#4904S), and p-p65 (#3033S)/p65 (#8242S) antibodies were purchased from Cell signaling Technology (Beverly, MA, USA). MMP-1/MMP-3/MMP-13 Luminex Assays kits were purchased from Affymetrix (EPX050-10015-901, EPX030-10829-90, USA). DMEM/F12 (#C11330500CP), FBS (#10099141), Trypsin EDTA (#25200-072), PBS
(#70011069), and other standard cell culture products were purchased from Invitrogen (Carlsbad, CA, USA).

Synovial tissues were obtained from patients with active RA (6 females and 1 male, aged 45–67 years) who were undergoing synovectomy or joint replacement. RA was diag- nosed according to the 1987 revised criteria of the American College of Rheumatology [10]. The study was performed according to the recommendations of the Declaration of Hel- sinki and approved by the Medical Ethical Committee of the First Affiliated Hospital, Sun Yat-sen University, China. All of the patients gave informed consent to take part in the study.
Freshly isolated synovial tissue was cut into small pieces and digested with collagenase I for 2 h at 37 °C to isolate FLS. The FLS were grown in DMEM/F12 medium sup- plemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 100 μg/ml streptomycin in a 5% CO2 incuba- tor. When the cells reached sub-confluence (70–80%), they were routinely trypsinized and passaged. Passage 4–6 FLS were used in our experiments, during which period the cell population is believed to be homogeneous (< 1% CD11b positive, < 1% phagocytic, and < 1% FcgRII- and FcgRIII- receptor positive) [11].RA FLS were cultured in serum-free DMEM/F12 for 24 h at a density of 1 × 104 cells/well in 96-well plates in tripli- cate. After starvation, the cells were untreated (as negative control) or treated with 0.1% dimethyl sulfoxide (DMSO) (#D2650, Sigma, USA) or various concentrations of PLM (5, 10, 15, 20 μM) for 48 h in DMEM/F12 with 10% FBS. After this period, the cells were cultured in FBS-free DMEM/F12 for 12 h. Ten microliters 5% MTT (dissolved in PBS, PH 7.4, Sigma, USA) was added to the cells, which were incubated for 4–6 h. The cells were washed twice with PBS, followed by the addition of 100 μl DMSO. The cells were shaken gently for 10 min at 37 °C to achieve complete dissolution. The absorbances were read on a spectrophotom- eter at wavelengths of 570 and 620 nm. RA FLS apoptosis was assayed using an Annexin V-FITC/ PI kit. RA FLS were seeded into 6-well culture plates and grown to confluence. After serum starvation in FBS-free DMEM/F12 for 24 h, the cells were untreated (as negative control) or treated with 0.1% DMSO or various concentra- tions of PLM (5, 10, 15, 20 μM) for 48 h in DMEM/F12 with 10% FBS. The apoptosis assay was then performed using an Annexin V-FITC/PI Apoptosis Detection Kit, according to the manufacturer’s instructions. The cells were analyzed using flow cytometry (Gallios, Beckman Coulter, USA), and data were analyzed using FlowJo software.Proliferation assay5-Ethynyl-2′-deoxyuridine (EDU) is a thymidine analog that can incorporate into replicating DNA instead of thymine (T), allowing for the labeling of proliferating cells. RA FLS were cultured in serum-free DMEM/F12 for 24 h at a den- sity of 1 × 104 cells/well in 96-well plates in triplicate. After starvation, the cells were untreated (as negative controls) or treated with 0.1% DMSO or various concentrations of PLM (1, 5, 10 μM) for 24 h in DMEM/F12 with 10% FBS. The cell proliferation assays were then performed using a Cell- Light™ EdU Apollo®567 In Vitro Imaging Kit, according to the manufacturer’s instructions.Cell migration and invasion assaysThe RA FLS chemotaxis assay was performed using the Boyden chamber method using a 6.5-mm diameter filter with a 8.0 µm pore size (Transwell #3422, BD Biosciences, USA). The bottoms of the upper chambers were coated with 0.1% gelatin for the cell migration assay or with Matrigel basement membrane matrix (#BD356234, BD Biosciences, USA) for the cell invasion assay. After 24 h of serum star- vation, the RA FLS were untreated (as negative controls) or treated with 0.1% DMSO or various concentrations of PLM (1, 5, 10 µM) for 12 h in serum-free DMEM/F12. The cells were then suspended in serum-free DMEM/F12 and plated on the upper chambers (2 × 104 cells/chamber) in trip- licate. Six hundred microliters DMEM/F12 with 10% FBS as the chemoattractant was added to the lower chambers. The chambers were incubated in a humidified CO2 incubator (5% CO2/95% air) at 37 °C for 6 h (cell migration assay) or 24 h (cell invasion assay). After incubation, the cells that had migrated to the lower surface of the filter membrane were fixed in methanol for 10 min at room temperature and stained with 0.1% crystal violet solution. Non-migrating cells on the upper surface of the filter membrane were wiped out gently with cotton swabs. The stained cells on the lower surface were observed, photographed, and counted under a microscope (Zeiss, Germany). Migration and invasion were quantified using the mean number of migrated cells per 5 random fields per chamber.Wounding closure assayRA FLS were seeded into 6-well culture plates and grown to confluence. The cells were starved for 24 h and then untreated (as negative controls) or treated with 0.1% DMSO or PLM (5 µM) for 12 h in serum-free DMEM/F12. The plates were then scratched with 200 μl pipette tips and washed twice with PBS to remove detached cells. The remaining cells were incubated in DMEM/F12 with 10% FBS for 36 h, after which they were photographed under a microscope (Zeiss, Germany) at the indicated time. Migra- tion was quantified by counting the cells that moved beyond a reference line.RA FLS were plated on 6-well culture plates and grown to confluence. After 24 h of serum starvation, the cells were untreated (as negative controls) or treated with TNF-α (10 ng/ml) in the absence or presence of PLM (5 µM) for 24 h in serum-free DMEM/F12. Cell culture supernatants were collected to detect the MMP-1, MMP-3, and MMP-13 levels using a Luminex Assays kit, according to the manu- facturer’s instruction.ROS levels were determined using 2′,7′-dichlorodihydro- fluorescein diacetate (DCFH-DA). After 24 h of serum starvation, the RA FLS were untreated (as negative con- trols) or treated with 0.1% DMSO or various concentrations of PLM (1, 5, 10 µM) or TNF-α (10 ng/ml) in serum-free DMEM/F12 for 1 or 6 h. Then these cells were suspended in serum-free DMEM/F12 containing 10 µM DCFH-DA and incubated at 37 °C for 20 min. Fluorescence was then measured using flow cytometry (Gallios, Beckman Coulter, USA), with an excitation wavelength of 488 nm and an emision wavelength of 525 nm. The level of intracellular ROS was calculated using the mean fluorescence intensity (MFI).Protein extracts were quantified using the bicinchoninic acid (BCA) protein assay (#23225, Thermo scientific Pierce, USA), mixed with 2 × loading buffer (4% SDS, 20% glyc- erol, 10% 2-mercaptoethanol, 0.004% bromophenol blue and0.125 M Tris HCl, pH 6.8), and heated at 95 °C for 10 min. The prepared samples were subjected to 10–12% SDS–poly- acrylamide gel electrophoresis and transferred onto nitrocel- lulose membranes (#10600002, GE healthcare, UK) using a wet transfer system. The membranes were blocked with 5% nonfat dried milk (#M203, Amersco, USA) in TBST (15 mM Tris-HCl pH 8.0, 150 mM NaCl, and 0.1% Tween 20) and probed with primary antibodies directed against p-p38/p38, p-JNK/JNK, p-ERK/ERK, p-p65/p65 and p-STAT3/STAT3 at 4 °C overnight. The bound antibodies were detected using anti-rabbit or mouse IgG conjugated with HRP, followed by use of the ECL Plus Western blotting detection system (#RPN2108, GE Healthcare, UK), according to the manu- facturer’s instructions.F-actin stainingRA FLS were cultured on confocal dishes (#NNU#150680, Nunc, Danmark) or glass coverslips (#174969, Nunc, Dan- mark) in DMEM/F12 with 10% FBS. When the cells were approximately 80% confluent, they were serum-starved for 24 h and were then untreated (as negative controls) or treated with 5%FBS in the absence or presence of PLM (5 µM) for 6 h. The cells were then fixed with 4% paraformaldehyde for 10 min and permeabilized with 0.2% Triton X-100 in PBS for 10 min at room temperature. For the detection of F-actin and pseudopodia organization, the cells were stained with rhodamine-labeled phalloidin for 10 min. The cell nuclei were visualized using DAPI, and coverslips were mounted on glass slides using antifade mounting media. The samples were observed and photographed using a confocal fluores- cence microscope (Zeiss LSM710, Germany).The data are expressed as the mean ± SEM from at least 3 independent experiments and analyzed using Student’s t test. For comparisons of three or more groups, the analyses were performed by one-way ANOVA. If F reached significance, two post-hoc tests were applied: Dunnett’s post-hoc test when comparing each group with the control or the Sidak post-hoc test if a multiple group comparison was necessary. The post-hoc tests were performed for F values correspond- ing to P < 0.05 and when there was no significant variance in homogeneity. A non-parametric analysis was performed to analyze the normalization of the generated data. P values less than or equal to 0.05 were considered significant. Results We first determined the effect of PLM on RA FLS viabil- ity and apoptosis. Our results showed that PLM did not affect cell viability at concentrations of 5 or 10 μM, while higher concentrations (15 and 20 μM) reduced cell viability (Fig. 1a). We also found that higher concentrations of PLM (15 and 20 μM) could promote apparent apoptosis, while a lower concentration of PLM (5 μM) did not induce apparent apoptosis (Fig. 1b, c). In addition, we determined the effect of PLM on RA FLS proliferation using an EDU assay. We choose a concentration of PLM below 10 μM to prevent cytotoxic effects in our subsequent experiments. As shown in Fig. 1d, e, we demonstrated that RA FLS proliferation was significantly inhibited by PLM at concentrations of 1, 5 and 10 μM.The effect of PLM treatment on the migration and invasion of RA FLSTo determine whether PLM modulates cell migration, RA FLS migration was evaluated using the transwell Boyden chamber and wound closure assays. To prevent cytotoxic- ity, we choose the concentrations of 1, 5 and 10 μM. Com- pared to untreated and DMSO-treated cells, we found that PLM (1, 5 and 10 μM) reduced cell migration in a dose- dependent manner (Fig. 2a). In the wound closure assay, we found that untreated or DMSO-treated cells migrated into the wounded area in 36 h, while PLM-treated (5 μM) cells failed to advance into the wounded area during the same period (Fig. 2c). The ability to invade cartilage is a critical pathogenic behavior of RA FLS. We, therefore, investigated the role of PLM (1, 5 and 10 μM) on RA FLS invasion using a transwell chamber coated with Matrigel matrix. Consistent with its effect on the migration of RA FLS, PLM treatment also inhibited the invasive ability of these cells in a dose- dependent manner (Fig. 2b).To determine the role of PLM in regulating actin organi- zation in RA FLS, fluorescent phalloidin staining was used to visualize polymerized actin (F-actin). As shown in Fig. 3a, compared to unstimulated FLS, 5% FBS pro- moted F-actin stress fiber formation and assembly, while this effect was prevented by PLM treatment at a concentra- tion of 5 μM. Furthermore, we observed the effect of 5 μM PLM on the formation of pseudopodia in migrating cells. As shown in Fig. 3b, FLS treated with 5% FBS exhibited protrusive pseudopodial structures at their leading edge, whereas cells treated with 5% FBS plus 5 μM PLM rarely produced pseudopodia, suggesting that PLM inhibited cell migration by regulating the formation of pseudopodia.The effect of PLM treatment on the secretion of MMP-1, MMP-3, and MMP-13To assess the role of PLM in regulating aggressive pheno- type of RA FLS, we evaluated its effect on TNF-α-induced secretion of MMP-1, MMP-3 and MMP-13. As shown in Fig. 4a–c, stimulation with TNF-α increased the secretion with PLM (1, 5, 10 μM) for 24 h for the proliferation assay. The val- ues are the mean ± SEM from at least 3 independent experiments.*P < 0.05 versus CTR, **P < 0.01 versus CTR, ***P < 0.001 versus CTRwere incubated with PLM (1, 5, 10 μM) for 12 h. Then wells with RA FLS were scratched using 200 μl tips when a confluent mon- olayer had formed (0 h). The cells were then continually cultured in DMEM/F12 with 10% FBS for 36 h. The values are the mean ± SEM from at least 3 independent experiments. **P < 0.01 versus CTR,***P < 0.001 versus CTR of MMP-1, MMP-3 and MMP-13, whereas this increase was inhibited by PLM treatment.The effect of PLM treatment on intracellular ROS in RA FLSWe next investigated whether PLM modulates ROS produc- tion by RA FLS. We first determined the effect of PLM on ROS production by unstimulated RA FLS. These cells were exposed to PLM (5 μM) at different time points (0, 1 and 6 h). As shown in Fig. 5a, treatment with PLM did not affect intracellular ROS production at different time points. We also found that different doses of PLM (1, 5, 10 μM) did not affect ROS production for 6 h (Fig. 5b). We next evalu- ated the effect of PLM on TNF-α-induced ROS production. TNF-α (10 ng/ml) stimulation increased intracellular ROS levels for 6 h (Fig. 5c). Therefore, we co-treated with TNF-α and PLM (5 μM) for 6 h, finding that PLM significantly decreased TNF-α-induced intracellular ROS production (Fig. 5d).The effect of PLM treatment on signaling pathways in RA FLSWe further examined the effect of PLM on the activation of NF-κB, STAT3 and MAPK signaling pathways, which play important roles in the pathogenesis of RA. As shown in Fig. 6a, b, we found that stimulation with TNF-α signifi- cantly increased the phosphorylation of p65 and STAT3, indicating the activation of the NF-κB and JAK/STAT3 pathways, respectively. However, PLM treatment signifi- cantly suppressed TNF-α-induced activation of p65 and STAT3. We also observed that PLM treatment decreased TNF-α-induced phosphorylation of p38 and JNK, but not ERK, significantly. These results indicate the inhibitory effect of PLM on the activation of the p38 and JNK MAPK pathways (Fig. 6c–e). Discussion In the present study, we found that PLM induced apoptosis when used at higher concentrations (15 and 20 μM) and inhibited RA FLS proliferation at concentrations of 1, 5 and 10 μM. In addition, the migration and invasion of RA FLS were inhibited by PLM in a dose-dependent manner at con- centrations of 1, 5 and 10 μM. PLM treatment regulated F-actin formation and decreased TNF-α-induced secretion of MMP-1, MMP-3 and MMP-13. We furthermore observed an inhibitory effect of PLM on activation of the NF-κB, p38, JNK and STAT3 pathways. These data suggest that PLM can regulate RA FLS function and this that compound is a potential therapeutic agent for the treatment of RA.RA FLS exhibit tumor-like characteristics, including anchorage-independent growth and resistance to apopto- sis [12]. Moreover, these cells migrate towards and invade the cartilage and bone and are thereby required for pannus development. Increasing evidence indicates that regulation of activated FLS migration and invasion may be a new thera- peutic strategy to prevent joint destruction in RA [3].Recent studies indicated that PLM inhibits the migration of other cell lines, such as vascular smooth muscle cells [13]. Increasing evidence suggests the potential importance of FLS-mediated joint destruction in RA [3]; however, no effective treatments have been found to directly target FLS to improve joint destruction. Consistent with previous reports [13, 14], we demonstrated that PLM treatment significantly reduced migration as determined using wound closure and transwell assays. Similar results were obtained regarding the invasive behavior of RA FLS through the use of Matrigel- coated transwell membranes. We also found that PLM inhib- ited the reorganization of F-actin and pseudopodia forma- tion in migrating RA FLS. As the formation of stress fibers and pseudopodia are critical steps that control cell motility, our findings further support the observations obtained for in vitro migration of RA FLS. The proteases that regulate Luminex assay. The values are the mean ± SEM from at least 3 independent experiments. *P < 0.05 versus CTR, **P < 0.01 versus CTR,***P < 0.001 versus CTR, #P < 0.05 versus TNF-α the remodeling of the extracellular matrix play an impor- tant role in the progressive destruction of joints observed in RA. Some studies have shown that FLS, which produce MMP-1, MMP-3, MMP-9 and MMP-10, have more invasive capability than cells that do not produce these MMPs [15]. Therefore, we assessed the effects of PLM on the secretion of MMP-1, MMP-3 and MMP-13. Our results demonstrated the inhibitory effect of PLM on TNF-α-induced secretion of MMP-1, MMP-3 and MMP-13. We found that PLM signifi- cantly inhibited proliferation of RA FLS, but no correlation was observed between the inhibition of proliferation and the decreased migration and invasion of these cells. FLS are slowly growing cells, with an average doubling time of 7 days [16], whereas the incubation times for the migration and invasion experiments were 6 and 24 h, respectively. In addition, to exclude the cytotoxic effects of PLM, we choose to use the concentrations of 1, 5 and 10 μM. On the basis of treated with TNF-α (10 ng/ml) alone for 0, 1 and 6 h. d The effect of PLM on TNF-α-induced intracellular ROS production. RA FLS were treated with TNF-α (10 ng/ml) and PLM (5 μM) for 6 h. The values are the mean ± SEM from at least 3 independent experiments.**P < 0.01 versus CTR, #P < 0.05 versus TNF-α these data, our findings indicate that (1) PLM inhibits RA FLS migration and invasion and (2) that these effects are not associated with its impacts on proliferation and apopto- sis. Collectively, our findings suggested that PLM treatment might contribute to inhibit the aberrant aggressive behavior of RA FLS.It has been reported that PLM can selectively kill tumor cell by upregulating intracellular ROS levels at concen- trations of 10 μM or lower and does not increase ROS levels in normal or immortalized non-transformed cells at concentrations of 10 μM or higher [8]. In our study, we observed that PLM treatment did not affect intracel- lular ROS production in unstimulated-RA FLS at various concentration (1, 5, 10 μM); however, PLM could reduce TNF-α-induced ROS production. These results suggest that PLM might modulate ROS production under inflam- matory conditions. Indeed, in a previous study by our group, we found that PLM treatment decreased intracel- lular ROS production in LPS-induced dendritic cells [17]. These findings suggest that the role of PLM in regulating intracellular ROS production may be associated with cell type.Several signaling pathways are activated in the synovial tissue and cells of patients with RA and are involved in RA FLS activation. The MAPK pathway, including p38, ERK, JNK, can regulate the proliferation, apoptosis, migration and cytokine secretion of RA FLS. It is reported that p38 inhibition can decrease MMP and cytokine secretion, reduce cartilage degradation, inhibit osteoclast forma- tion and attenuate disease severity in mice with collagen- induced arthritis [18, 19]. JNK is also required for MMP and collagenase production, joint inflammation and bone destruction [20, 21]. ERK appears to participate in the secretion of certain cytokines as well as cell proliferation [22]. In the present study, we found that PLM suppressed TNFα-induced activation of p38 and JNK, but not ERK, in RA FLS. This result suggests that p38 and JNK might in part mediate the observed PLM-induced inhibition of RA FLS proliferation, migration and invasion. As a key signal transcription factor that regulates the expression of inflammatory genes and cell proliferation, NF-κB is involved in synovial inflammation, hyperplasia, and bone destruction in RA [23–25]. STAT3 also modulates the function of RA FLS, including cell growth, survival, and apoptosis, and is considered to be an important factor in the maintenance of synovial inflammation in RA [26]. In this work, we found that PLM inhibited TNF-α-induced phosphorylation of p65 and STAT3. Taken together, our findings indicate that PLM can regulate RA FLS function via multiple signaling pathways that are involved in the activation and aggressiveness of this cell type. ROS are critical for the activation of MAPK, STAT3 and NF-κB pathways [27–29]; therefore, we speculate that PLM acts on RA FLS by inhibiting ROS-mediated activation of these pathways. In summary, our findings suggest that PLM is capa- ble of modulating FLS activity and function and that this occurs through inhibition of the p38, JNK, NF-κB and STAT3 pathways. Further studies are needed to explore the therapeutic potential of PLM for RA. Acknowledgements The authors would like to thank Jinjin Fan for her technical Piperlongumine assistance.