Sesquiterpenes as Immunosuppressants Alba Munoz-Suano,1 David Chinchilla,2 Susana Aguilar,1 Jose M.G. Molinillo,2 Francisco A. Macias,2 Manuel A. Rodríguez-Iglesias,1 and Francisco Garcia-Cozar1,3 Background. The search for new immunosuppressive drugs is a high priority. Sesquiterpenes constitute a family of compounds with a great variety of biological activities due to their reactive moieties. Methods. Human tumor cell lines and murine primary cells and human peripheral blood mononuclear cells or primary CD4⫹ cells from healthy individuals were stimulated in the presence of sesquiterpenes. Cell division was analyzed by 5-(and-6)-carboxyfluorescein diacetate succinimidyl ester, cell cycle progression by Hoecht, and cell death by Anexin-V and propidium iodine staining. Cytokine secretion was analyzed by means of a bioplex assay. Results. Two sesquiterpene derivatives of the 18 previously shown to inhibit vegetal cell growth are shown to block cell division and cell cycle progression in human and murine cell lines and primary cells. Cytokine secretion is also impaired on stimulation in the presence of sesquiterpenes. Conclusions. Here, we show that sesquiterpenes heliannuols constitute a novel family of molecules with potential use as immunosuppressants. Moreover, we show that an assay based on the allelopathic effect of plant leads can be used as a cost-effective screening previous to studies in mammalian cells. Keywords: Immunosuppressants, Lead Compounds, Sesquiterpenes, Heliannuols. (Transplantation 2009;88: S24–S30)
mmunosuppressants have allowed widespread organ transplantation, but their utility remains limited by toxicity. Thus, the search for new drugs with immunosuppressive activity is still ongoing. Target-based drug design methods, frequently do not give the desired effect, so classical drug discovery procedures based on testing the pharmacological effect of natural product is still the approach that yields most pharmaceuticals. Chemical studies of Helianthus annuus L. have shown that this species is a rich source of terpenoids, particularly sesquiterpenoids (1, 2). Along with these compounds, the results obtained with heliannuols in biological assays make them interesting leads in bioactivity studies (3). This biomolecules own a wide spectrum of biological activities including the inhibition of growth of other plants termed allelopathy (4 – 6). In this work, we have tested the attractive possibility of using molecules positive in an initial test based on the alleopathic effect, using the etiolated wheat coleoptiles bioassay (7) that test the effect of biomolecules on plant cell growth, to narrow the number of biomolecules to be tested in immunological assays down to eighteen. From those 18 biomolecules, two showed to be active in inhibiting immune cells.
I
MATERIALS AND METHODS Plant Material Extraction and Isolation Tested compounds belong to two different skeletons. The fist one is heliannuols, bioactive sesquiterpenes isolated from Helianthus annuus. They were obtained using the procedures described by Macias et al. (IH1 [8]; IH3, IH4, IH14, IH15, IH16 [9]; and IH2, IH13, IH17, and IH18 [10]). The second family is benzoxazinoids, alkaloids and their degradation products isolated from the family Poaceae. Tested compounds were prepared following procedures previously reported (IH5, IH6, IH7, IH8, IH9, IH10, IH11 [11] and IH12 [12]). All extracts were analyzed using the Coleoptile Wheat Bioassay (7).
This study was supported by the Ministerio de Educacion y Ciencia (Spain) grants AGL2005-05190/AGR and SAF2005-00458; Ministerio de Salud (Spain) grant FIS PI050715; Consejeria de Salud de Andalucia (Spain) grant 0080/2005; the Ramon y Cajal Program (F.J.G.C.); and Junta de Andalucia, Andalucia, Spain (A.M.S. was a graduate fellow). The authors declare no conflict of interest. 1 Departamento de Bioquímica y Biología Molecular (Inmuologia), Universidad de Cadiz, Hospital Universitario de Puerto Real, Unidad de Investigacio´n, Carretera NIV Km665, 11510 Puerto Real, Cadiz, Spain. 2 Grupo de Alelopatia, Departamento de Quimica Organica, Facultad de Ciencias, Universidad de Cadiz, Ada. Republica Saharaui, Apdo. 40, 11510 Puerto Real, Spain. 3 Address correspondence to: Francisco Garcia-Cozar, Hospital Universitario de Puerto Real, Unidad de Investigacio´n, Carretera NIV Km665, 11510 Puerto Real, Cadiz, Spain. E-mail:
[email protected] Copyright © 2009 by Lippincott Williams & Wilkins ISSN 0041-1337/09/8803S-24 DOI: 10.1097/TP.0b013e3181af66fa
Cell Culture and Stimulation The Jurkat (American Type Culture Collection, Manassas, VA) cell line was maintained routinely in RPMI-1640 medium supplemented with 2 mM L-glutamine, 10 mM HEPES, 10% FBS, 1% nonessential amino acids, 1% sodium pyruvate, and 1% penicillin/streptomycin at 37°C. For primary mouse cultures, cells were obtained by mechanical dissociation of the spleens and the lymphatic nodes of DO11.10 transgenic mice. Red blood cells were lysed by treatment with a buffer containing 0.17 M Tris-0.16 M ammonium chloride. Cells were stimulated with the cognate ovalbumin (OVA) peptide and cultured in the presence of splenocytes from BALB/C treated with mytomicin C used as feeders. Cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 2 mM L-glutamine, 100 mg/mL streptomycin, 100 U/mL penicillin, 20 mg/mL gentamicin, 1 g/mL pyruvate, and 1 g/mL nonessential amino acids, HEPES, vitamin and L-arginin, L-asparragin, folic acid, 50 M 2-ME (Invitrogen Life Technologies, Milan, Italy), and 10% heat-inactivated FBS (Euroclone, Milan, Italy) at 37°C in a 5% CO2 atmosphere by periodic addition of IL-2 (donated by the AIDS reagent Program) at 40 IU/mL every 48 hr for 7 days. After that period, CD4⫹ T cells were immunomagnetically purified by means of an anti-mouse CD4 biotinilated
S24 |
Transplantation • Volume 88, Number 3S, August 15, 2009
www.transplantjournal.com
© 2009 Lippincott Williams & Wilkins
antibody (clon L3T4, BD Pharmingen) and streptavidinIMag particles BD (Becton Dickinson). CD4⫹ cells were stimulated with hamster anti-mouse CD3 (clon 145–2C11, BD Pharmingen) and hamster anti-mouse CD28 (clon 37.51, BD Pharmingen) at 1 g/mL. Human peripheral blood samples were obtained from healthy donors, and peripheral blood mononuclear cells (PBMC) isolated by density gradient centrifugation using Lymphocyte separation Medium (D⫽1.077⫾0.001) (Eurobio, Montpellier, France). PBMC were subsequently stimulated with 1 g/mL phytohemagglutinin-P (PHA) (Sigma) and cultured in Dulbecco’s modified Eagle’s medium supplemented with 2 mM L-glutamine, 100 mg/mL streptomycin, 100 U/mL penicillin, 20 mg/mL gentamicin, 1 g/mL pyruvate, and 1 g/mL nonessential amino acids, HEPES, vitamin and L-arginin, L-asparragin, folic acid, 50 M 2-ME (Invitrogen Life Technologies, Milan, Italy), and 10% heat-inactivated FBS (Euroclone, Milan, Italy) at 37°C in a 5% CO2 atmosphere by periodic addition of IL-2 (AIDS reagents) at 40 IU/mL every 48 hr for 7 days. After that period, CD4⫹ T cells were immunomagnetically purified by means of an anti-human CD4 biotinilated antibody (clon RPA-T4, BD Pharmingen) and streptavidin-IMag
Munoz-Suano et al.
S25
particles BD (Becton Dickinson). CD4⫹ cells were stimulated with mouse antibodies anti-human CD3 (clon UCHT1, BD Pharmingen) at 2 g/mL and mouse anti-human CD28 (clon CD28.2, BD Pharmingen,) at 2 g/mL. 5-(and-6)-Carboxyfluorescein diacetate succinimidyl ester Proliferation, Cell Death, and Flow Cytometry Cells were left unstimulated or stimulated with anti-CD3 plus anti-CD28 in the presence or absence (DMSO vehicle was added at an equivalent volume) of sesquiterpenes and incubated for the indicated period (see figure legends). For 5-(and-6)carboxyfluorescein diacetate succinimidyl ester (CFSE) assay, cells were washed and resuspended in 0.5 mL PBS containing 5 M CFSE (Molecular Probes, Eugene, OR) and incubated at 37°C for 15 min. Labeling was quenched by addition of an equal volume of PBS containing 0.5% BSA, cells were washed twice in PBS, resuspended in complete medium and cell number determined by trypan blue exclusion. Cell Death Assay Cell death was evaluated with APOPTEST FITC (Nexins research, Netherlands) according to the manufacturer’s
FIGURE 1. Effect of the sesquiterpenes in D5 cell division upon anti-CD3 plus anti-CD28 stimulation. (A) Sesquiterpenes that have been shown to be active in the etiolated wheat coleoptiles bioassay. (B) Upper two: CFSE staining of cells unstimulated (gray solid curve) or stimulated in the absence of sesquiterpenes (empty curve). Lower three: overlay of the previous curves with those corresponding to cells stimulated in the presence of sesquiterpenes or DMSO vehicle (striped curves).
S26 |
www.transplantjournal.com
protocol. At least 10,000 live events were collected on a Cyan-ADP-MLE II flow cytometer (DakoCytomation). Acquisition and analysis was performed using summit software (DakoCytomation). Cell Cycle Assay For cell cycle analysis, cells were stimulated in the presence or absence of sesquiterpenes. Four days after, 17 M Hoechst 33342 (SIGMA) was added, and cells were incubated for two additional hours at 37°C. Before the analysis, propidium iodide was added at 250 g/mL to exclude dead cells with less than diploid DNA content. Relative DNA content of life cells was analyzed on a Cyan-ADP-MLE (DakoCytomation) using a UV enterprise laser set at 30 mW. The percentage of cells in various stages of the cell cycle was determined using the Summit 4.3 software (DakoCytomation). Cytokine Assay CD4 cells were stimulated with anti-CD3 plus antiCD28 in the presence or absence of sesquiterpenes. After 16 hr, supernatants were collected and stored at – 80°C until the analysis was performed. IL-2, IL-4, IL-10, TNF-␣, and IFN-␥ were determined using BioSource Multiplex Assays for Luminex (Invitrogen) as described by de Jager et al. (13) according
Transplantation • Volume 88, Number 3S, August 15, 2009
to the manufacturer’s protocol. Each was performed in triplicate, with the experiment repeated three times. Results were obtained in a Luminex platform and data analysis were performed using a Luminex 100 IS software version 2.3 (BioSource, Invitrogen). A four-parameter logistic regression model was used when more than six standards were within 70% to 130% of the expected values. Data were compared using the Wilcoxon rank sums test and differences were considered significant if the P value was less than 0.05.
RESULTS Effect of IH3 and IH4 on the Murine Antigen Specific CD4ⴙ T-Cell Line D5 From the 18 sesquiterpenes that had been shown to be active on the etiolated wheat coleoptiles bioassay (shown in Fig. 1A), only two (IH3 and IH4) were able to induce inhibition ofcelldivisioninthehumantumorcelllineJurkat(datanotshown). Once these sesquiterpenes were tested in tumor cell lines, we wanted to analyze whether these compounds had activity on cell division of the nontumor cell line D5 (14), so we analyzed the effect on D5 cell division on antigenic stimulation in the presence or absence of IH3 or IH4. Cells were labeled with CFSE and subsequently treated during half an
FIGURE 2. Effect of sesquiterpenes in primary murine CD4⫹ T-cell division on anti-CD3 plus anti-CD28 stimulation. 5-(and-6)-Carboxyfluorescein diacetate succinimidyl ester staining of cells left unstimulated (gray solid curve), stimulated in the absence (empty curve) or stimulated in the presence of sesquiterpenes (striped curves) in the absence (A) or presence (B) of exogenous IL-2. (C) Cell cycle distribution of cells unstimulated and stimulated in the absence or presence of sesquiterpenes.
© 2009 Lippincott Williams & Wilkins
hour with the sesquiterpenes, previous to the addition of the cognate antigen arsOVA. CFSE dilution was determined 4 days after the stimulation by flow cytometry (Fig. 1B). As shown in Figure 1(B), D5 cells stimulated in the presence of IH3 or IH4 were blocked in cell division. Effect of IH3 and IH4 on Murine Primary CD4 T Cells From TcR Transgenic DO11.10 Mice To assess the effect of these compounds on division of murine primary cells, CD4 T cells were purified from 1 week mature murine OVA-stimulated T-cell blasts. Cells were labeled with CFSE, IH3 or IH4 was added to the cultures (at 1 M final concentration), and cells were incubated 30 min at 37°C before stimulation with anti-CD3 plus anti-CD28. Results were analyzed by flow cytometry 4 days after the stimulation. As shown in Figure 2, addition of sesquiterpenes caused an inhibition of cell division. Unstimulated control cells, without sesquiterpenes showed higher cell division than stimulated cells treated with sesquiterpenes, which could be caused by residual stimulation in response to the first stimulation with OVA. It is interesting to note that residual stimulation is also blocked by IH3 and IH4. To assess if the effect on cell division could be due to an impaired production of IL-2; IL-2 was added to the culture at the moment of stimulation. Cells pretreated with IH4 partially responded to IL-2 and showed a decreased inhibition, showing limited cell division, whereas cells pretreated with
Munoz-Suano et al.
S27
IH3 showed no change in the inhibition even in the presence of IL-2 (Fig. 2B). We next wanted to analyze whether the effect on cell division is related with cell cycle progression. Thus, murine CD4 T cells were labeled “in vivo” with Hoechst 33342, 4 days after the stimulation in presence or absence of sesquiterpenes. In cells pretreated with sesquiterpenes IH3 or IH4, the percentage of cells in phase S-M changed from 33.47% in the absence of sesquiterpenes to 9.21% or 4.21% in cells stimulated in the presence of sesquiterpenes IH3 or IH4, respectively (Fig. 2C). Effect of IH3 and IH4 on Human Primary Naïve PBMC To assess if sesquiterpenes IH3 and IH4 have a similar effect on human primary lymphoid cells, human PBMC were isolated from healthy donors, pretreated with 1 M final concentration of the corresponding sesquiterpene, and then stimulated with PHA. After 48 hr, life cells were labeled with Hoechst 33342 to analyze the effect of sesquiterpenes on cell cycle distribution. At this time point, 21.57% of cells stimulated with PHA were in S-M phase while only 1.13% and 2.01% of cells stimulated in the presence of IH3 and IH4. At 72 hr, differences were maintained (Fig. 3) as well as at 96 hr, although the percentage of cells in phase S-M is slightly increasing in the sesquiterpene-treated cells, indicating that the effect is reversible (Fig. 3).
FIGURE 3. Effect of sesquiterpenes in the cell cycle distribution of primary human naïve peripheral blood mononuclear cells on phytohemagglutinin-P stimulation. Peripheral blood mononuclear cells from healthy donors were pretreated with 1 M final concentration of the corresponding sesquiterpene 1 hr before phytohemagglutinin-P addition, labeled with Hoechst 33342 48 hr after, and analyze in a Cyan-ADP-MLE II flow cytometer (DakoCytomation).
S28 |
www.transplantjournal.com
Transplantation • Volume 88, Number 3S, August 15, 2009
FIGURE 4. Effect of sesquiterpenes in primary human CD4⫹ T cell division upon anti-CD3 plus anti-CD28 stimulation. 5-(and6)-Carboxyfluorescein diacetate succinimidyl ester staining of cells left unstimulated (gray solid curve), stimulated in the absence (empty curve) or stimulated in the presence of Sesquiterpenes (striped curves) in the absence (A) or presence (B) of exogenous IL-2. (C) Cell cycle distribution of cells unstimulated and stimulated in the absence or presence of sesquiterpenes.
FIGURE 5. Effect of sesquiterpenes in cell death of primary human CD4⫹ T cell on anti-CD3 plus anti-CD28 stimulation. CD4⫹ cells were stimulated in the presence or absence of sesquiterpenes and aliquots were stained at the indicated time periods with Anexin-V and propidium iodine.
© 2009 Lippincott Williams & Wilkins
Effect of IH3 and IH4 on Human Primary CD4ⴙ T Cells We next wanted to analyze the effect of active sesquiterpenes in human CD4⫹ T cells. Thus, CD4 cells were immunomagnetically purified, from PBMC blast that had been stimulated with PHA and cultured during a week. Cells were subsequently labeled with CFSE and treated with sesquiterpenes (1 M) half an hour before stimulation with soluble antiCD3 plus anti-CD28 antibodies. Four days after stimulation, proliferation was determined by means of CFSE dilution determined by flow cytometry. CD4⫹ cells stimulated in the presence of sesquiterpenes showed an intensity of CFSE staining similar to those unstimulated controls (Fig. 4A). To assess if this effect is dependent exclusively on IL-2 inhibition, IL-2 was added to the cultures and the effect of sesquiterpenes was evaluated (Fig. 4B). In the presence of IL-2, only those cells pretreated with IH4 showed a slight increase in cell division, while cells stimulated in the presence of IH3, did not divide even in the presence of IL-2. We next wanted to analyze like in primary murine CD4 T cells, the effect of sesquiterpenes in cell cycle progression by labeling life cells with Hoechst 33342. As shown in Figure 4(C), only 6.18% and 7.88% of cells stimulated in the pres-
FIGURE 6. Effect of sesquiterpenes in cytokines secretion by primary human CD4⫹ T cells. CD4⫹ cells were stimulated in the presence or absence of sesquiterpenes or either left unstimulated. Culture supernatants were collected 24 hr after the second stimulation and assayed for the indicated cytokines using the BioSource Multiplex Assays for Luminex (Invitrogen).
Munoz-Suano et al.
S29
ence of sesquiterpenes were in phase S-M while 22.14% of cells stimulated in the absence of sesquiterpenes have reached the S or M phases. As cell cycle block could also be due to mitotic cell death (15), we wanted to analyze cell death induced in the presence of sesquiterpenes to evaluate its involvement in the observed effect. Thus, human CD4 T cells were pretreated during 30minutes with 1 M final concentration of IH3 and IH4 before stimulation with anti-CD3 and anti CD28, and an annexinV assay was performed 6 and 12 hr later. As shown in Figure 5, there is an increase in cell death that is maximal during the first 6 hr after stimulation in the presence of sesquiterpenes. In the absence of stimulation, sesquiterpenes did not show any increase in cell death (data not shown). Effect of IH3 and IH4 on Cytokine Production by Human Primary CD4ⴙ T Cells Cytokine production is one characteristic feature of effector CD4 T cells, thus we quantitated IL-2, IL-4, IL-10, IFN-␥, and TNF-␣ in culture supernatants, 24 hr after stimulation in the presence or absence of IH3 or IH4 by means of a multiplex assay. As shown in Figure 6, cells pretreated with
S30 |
www.transplantjournal.com
sesquiterpenes showed a diminished production of IL-2, IL-4, IL-10, IFN-␥, and TNF-␣ in response to stimulation.
DISCUSSION Molecules of the sesquiterpene family display several relevant biological activities. Thus, sesquiterpenes have been shown to inhibit NO and NFB activation (16) and to have anti-inflammatory effects “in vivo” (17). Here, we show for the first time an effect of heliannuols, a sesquiterpene subfamily that has not previously been tested on cells from the immune system. We show that heliannuols are able to inhibit cell division in human and murine cell lines as well as primary cell blasts and naïve primary cells. Even though there is an increase in cell death at early time points, it does not account for the observed block in cell division. This data, together with our results on cell cycle distribution; point to a block in cell cycle progression as the most likely cause for the effect of IH3 and IH4. The effect on cell cycle progression seems to be reversible, as cells start entering S phase 96 hr after heliannuol addition (Fig. 3), indicating that after the molecules are degraded, cells are able to overcome the inhibition and initiate DNA replication. Secretion of all cytokines tested (IL2, IL4, IL10, IFN-␥ and TNF-␣) is also affected by IH3 and IH4, in agreement with results observed with related molecules in immune and non immune cells, specifically sesquiterpenes lactones, have been shown to inhibit IL-8 production by epithelial tumor cell line HeLa (18). Other sesquiterpenes isolated from Inula Britannica inhibit TNF-␣ secretion by the murine macrophage cell line RAW264.7 (16) and sesquiterpenes isolated from Thai Zedoary inhibit IL-4 secretion by the rat basophilic leukemia cell line RBL-2H3 (19). We are the first to show an effect of molecules from the sesquiterpene family, specifically heliannuols, on primary human T cells. Heliannuols are active at inhibiting T-cell activation at doses that are adequate to make them potential targets for the development of new lead compounds with therapeutic activity. As an additional conclusion of our work, the etiolated wheat coleoptiles bioassay that measures the activity of molecules on plant cell activation emerges as a useful screening previous to specific immune assays on mammalian cells, as a high proportion of molecules that are positive for the allelopathic effect are also active in inhibiting T cell activation. ACKNOWLEDGMENTS IL-2 was obtained through the AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, NIH.
Transplantation • Volume 88, Number 3S, August 15, 2009
REFERENCES 1.
2.
3. 4.
5.
6. 7. 8.
9.
10. 11.
12.
13.
14.
15.
16.
17.
18.
19.
Spring O, Rodon U, Macias FA. Sesquiterpenes from noncapitate glandular trichomes of Helianthus annuus. Phytochemistry 1992; 31: 1541. Macias FA, Molinillo JMG, Varela RM, et al. Structural elucidation and chemistry of a novel family of bioactive sesquiterpenes—Heliannuols. J Org Chem 1994; 59: 8261. Macias FA, Torres A, Galindo JLG, et al. Bioactive terpenoids from sunflower leaves cv. Peredovick. Phytochemistry 2002; 61: 687. Macias FA, Varela RM, Torres A, et al. Potential allelopathic activity of natural plant heliannanes: A proposal of absolute configuration and nomenclature. J Chem Ecol 2000; 26: 2173. Macias FA, Varela RM, Torres A, et al. Allelopathic studies in cultivar, Part 10: Bioactive norsesquiterpenes from Helianthus annuus potential allelopathic activity. Phytochemistry 1998; 48(4): 631. Macias FA, Molinillo JMG, Varela RM, et al. Allelopathy—A natural alternative for weed control. Pest Manage Sci 2007; 63: 327. Hancock CR, Lacey HJ, Barlow HWB. East malling coleoptile straight growth test method. J Exp Botany 1964; 15: 166. Macias FA, Varela RM, Torres A, et al. Novel sesquiterpene from bioactive fractions of cultivar sunflowers. Tetrahedron Lett 1993; 34: 1999. Macias FA, Chinchilla D, Molinillo JMG, et al. Synthesis of heliannane skeletons. Facile preparation of (⫹/⫺)-heliannuol D. Tetrahedron 2003; 59: 1679. Macias FA, Chinchilla D, Molinillo JMG, et al. A stereoselective route towards heliannuol A. Tetrahedron 2008; 64: 5502. Macias FA, Marin D, Oliveros-Bastidas A, et al. Isolation and synthesis of allelochemicals from gramineae: Benzoxazinones and related compounds. J Agric Food Chem 2006; 54: 991. Macias FA, De Siqueira JM, Chinchilla N, et al. New herbicide models from benzoxazinones: Aromatic ring functionalization effects. J Agric Food Chem 2006; 54: 9843. de Jager W, te Velthuis H, Prakken BJ, et al. Simultaneous detection of 15 human cytokines in a single sample of stimulated peripheral blood mononuclear cells. Clin Diagn Lab Immunol 2003; 10: 133. Nalefski EA, Wong JG, Rao A. Amino acid substitutions in the first complementarity-determining region of a murine T-cell receptor alpha chain affect antigen-major histocompatibility complex recognition. J Biol Chem 1990; 265: 8842. Schneider L, Essman F, Kletke A, et al. The transforming acidic coiled coil 3 protein is essential for spindle-dependent chromosome alignment and mitotic survival. J Biol Chem 2007; 282: 29273. Jin HZ, Lee D, Lee JH, et al. New sesquiterpene dimers from Inula britannica inhibit NF-kappaB activation and NO and TNF-alpha production in LPS-stimulated RAW264.7 cells. Planta Med 2006; 72: 40. Jimenez-Estrada M, Chilpa RR, Apan TR, et al. Anti-inflammatory activity of cacalol and cacalone sesquiterpenes isolated from Psacalium decompositum. J Ethnopharmacol 2006; 105: 34. Lindenmeyer MT, Hrenn A, Kern C, et al. Sesquiterpene lactones as inhibitors of IL-8 expression in HeLa cells. Bioorg Med Chem 2006; 14: 2487. Matsuda H, Tewtrakul S, Morikawa T, et al. Anti-allergic principles from Thai zedoary: Structural requirements of curcuminoids for inhibition of degranulation and effect on the release of TNF-alpha and IL-4 in RBL-2H3 cells. Bioorg Med Chem 2004; 12: 5891.
