Boswellic acids and malignant glioma: induction of apoptosis but no modulation of drug sensitivity.

Steroids are essential for the control of oedema in human malignant glioma patients but may interfere with the efficacy of chemotherapy. Boswellic acids are phytotherapeutic anti-inflammatory agents that may be alternative drugs to corticosteroids in the treatment of cerebral oedema. Here, we report that boswellic acids are cytotoxic to malignant glioma cells at low micromolar concentrations. In-situ DNA end labelling and electron microscopy reveal that boswellic acids induce apoptosis. Boswellic acid-induced apoptosis requires protein, but not RNA synthesis, and is neither associated with free radical formation nor blocked by free radical scavengers. The levels of BAX and BCL-2 proteins remain unaltered during boswellic acid-induced apoptosis. p21 expression is induced by boswellic acids via a p53-independent pathway. Ectopic expression of wild-type p53 also induces p21, and facilitates boswellic acid-induced apoptosis. However, targeted disruption of the p21 genes in colon carcinoma cells enhances rather than decreases boswellic acid toxicity. Ectopic expression of neither BCL-2 nor the caspase inhibitor, CRM-A, is protective. In contrast to steroids, subtoxic concentrations of boswellic acids do not interfere with cancer drug toxicity of glioma cells in acute cytotoxicity or clonogenic cell death assays. Also, in contrast to steroids, boswellic acids synergize with the cytotoxic cytokine, CD95 ligand, in inducing glioma cell apoptosis. This effect is probably mediated by inhibition of RNA synthesis and is not associated with changes of CD95 expression at the cell surface. Further studies in laboratory animals and in human patients are required to determine whether boswellic acids may be a useful adjunct to the medical management of human malignant glioma.

Acetyl-11-keto-beta-boswellic acid induces apoptosis in HL-60 and CCRF-CEM cells and inhibits topoisomerase I.

Antiproliferative action of different pentacyclic triterpenes has repeatedly been reported, and some lipoxygenase inhibitors have been shown to induce cell death in various cell systems. Acetyl-11-keto-beta-boswellic acid (AKBA) is a pentacyclic triterpene that inhibits 5-lipoxygenase in a selective, enzymedirected, nonredox, and noncompetitive manner. To investigate a possible effect of AKBA on leukemic cell growth, proliferation of HL-60 and CCRF-CEM cells was assayed in the presence of AKBA and a structural analog without effect on 5-lipoxygenase, amyrin. Cell counts and [3H]thymidine incorporation were significantly reduced in a dose-dependent manner in the presence of AKBA (IC50 = 30 microM) but not amyrin. An additive effect of AKBA with the crosslinking of the CD95 receptor was also observed. Flow cytometric analysis of propidium iodide-stained cells indicated that the cells underwent apoptosis. This was confirmed by flow cytometric detection of sub-G1 peaks in AKBA-treated cells and by DNA laddering. However, because HL-60 and CCRF-CEM do not express 5-lipoxygenase mRNA constitutively, a mechanism distinct from inhibition of 5-lipoxygenase must account for the effect of AKBA. In a DNA relaxation assay with phiX174RF DNA, AKBA inhibited topoisomerase I from calf thymus at concentrations of >/=10 microM. A semiquantitative cDNA polymerase chain reaction approach was used to estimate the relative level of expression of topoisomerases in both cell lines. The data suggest that induction of apoptosis in HL-60 and CCRF-CEM by AKBA may be due to inhibition of topoisomerase I in these cells.

Characterization of an acetyl-11-keto-beta-boswellic acid and arachidonate-binding regulatory site of 5-lipoxygenase using photoaffinity labeling.

AKBA (acetyl-11-keto-beta-boswellic acid), a natural pentacyclic triterpene, is an orally active leukotriene-synthesis inhibitor, which acts by a 5-lipoxygenase-directed, non-redox, non-competitive mechanism. It is the only leukotriene-synthesis inhibitor so far identified that inhibits 5-lipoxygenase activity as an allosteric regulator and not by a reducing or competitive mechanism. To characterize AKBA's effector site we prepared azido125I-KBA (4-azido-5-125iodo-salicyloyl-beta-alanyl-11-keto-beta-bo swellic acid) as a photoaffinity analogue, which inhibited 5-lipoxygenase activity as efficiently as the lead compound and specifically labeled human 5-lipoxygenase protein. The labeling of 5-lipoxygenase by azido-125I-KBA strictly depended on the presence of calcium ([Ca2+]free > 500 nM) and was abolished by heat denaturation or by prior incubation with a series of pentacyclic triterpenes (e.g., amyrin, beta-boswellic acid, AKBA and 18a-glycyrrhetinic acid). In contrast, 18-beta-glycyrrhetinic acid and competitive 5-lipoxygenase inhibitors (e.g., ZM-230,487 and L-739,010) did not affect labeling. Arachidonic acid, in enzyme-activity-inhibiting concentrations, reduced photoincorporation (IC50 about 10 microM), whereas a variety of other long-chain fatty acids and their derivatives (e.g., arachidinic acid, arachidonic acid methyl ester, lipoxins A4 and B4) had no effect. The inhibitory arachidonate action on labeling was not affected by blocking the substrate-binding site by micromolar amounts of the competitive inhibitor L-739,010. Therefore, we suggest that AKBA binds in presence of calcium to a site which is distinct from the substrate binding site of 5-lipoxygenase. The AKBA-binding site is likely to be identical with a regulatory, second arachidonate binding site of the enzyme.

Inhibition by boswellic acids of human leukocyte elastase.

Frankincense extracts and boswellic acids, biologically active pentacyclic triterpenes of frankincense, block leukotriene biosynthesis and exert potent anti-inflammatory effects. Screening for additional effects of boswellic acids on further proinflammatory pathways, we observed that acetyl-11-keto-beta-boswellic acid, an established direct, nonredox and noncompetitive 5-lipoxygenase inhibitor, decreased the activity of human leukocyte elastase (HLE) in vitro with an IC50 value of about 15 microM. Among the pentacyclic triterpenes tested in concentrations up to 20 microM, we also observed substantial inhibtion by beta-boswellic acid, amyrin and ursolic acid, but not by 18beta-glycyrrhetinic acid. The data show that the dual inhibition of 5-lipoxygenase and HLE is unique to boswellic acids: other pentacyclic triterpenes with HLE inhibitory activities (e.g., ursolic acid and amyrin) do not inhibit 5-lipoxygenase, and leukotriene biosynthesis inhibitors from different chemical classes (e.g., NDGA, MK-886 and ZM-230,487) do not impair HLE activity. Because leukotriene formation and HLE release are increased simultaneously by neutrophil stimulation in a variety of inflammation- and hypersensitivity-based human diseases, the reported blockade of two proinflammatory enzymes by boswellic acids might be the rationale for the putative antiphlogistic activity of acetyl-11-keto-beta-boswellic acid and derivatives.

A test system for leukotriene synthesis inhibitors based on the in-vitro differentiation of the human leukemic cell lines HL-60 and Mono Mac 6.

Differentiation of HL-60 cells along the granulocytic lineage by DMSO in the presence of transforming growth factor-beta and low concentrations of 1,25-dihydroxyvitamin D3 leads to the upregulation of 5-lipoxygenase activity in 100,000 g supernatants and intact cells to levels which are comparable to normal granulocytes. Similarly, differentiation of the human monocytic cell line Mono Mac 6 by 1,25-dihydroxyvitamin D3 and transforming growth factor-beta strongly upregulates the 5-lipoxygenase pathway. Here, we describe an assay system for leukotriene biosynthesis inhibitors which is based on the in-vitro differentiation of HL-60 and Mono Mac 6 cells. Different leukotriene biosynthesis inhibitors like the nonredox type inhibitor ZM 230487, the redox type inhibitor BW A4C and the FLAP inhibitor MK886 were tested and the results were compared with an assay system based on normal human granulocytes. ZM 230487, BWA4C and MK886 showed similar potencies in these cell lines as compared to normal leukocytes. Thus, the in-vitro differentiation of HL-60 and Mono Mac 6 cells provides an excellent model for the screening of drugs affecting the 5-lipoxygenase pathway.

Anti-inflammatory actions of pentacyclic triterpenes.

Pentacyclic triterpenes (PTs) as aglycones of saponins have a wide distribution in plants, and many of them have been used as anti-inflammatory remedies in folk medicine. This survey critically reviews the effects of PTs on proinflammatory mediator signalling pathways and data from experimental animal models and clinical trials. Because the knowledge of their actions is far from being satisfactory a critical summary of the partly promising but mostly scattered and preliminary data might promote productive research on chances and risks of PTs. Antiproliferative and anti-infectious actions and effects on intracellular cell signalling and hormone metabolism are beyond the scope of this short review, although such effects might also contribute to the understanding of the systemic anti-inflammatory actions of aglycones.

Acetyl-11-keto-beta-boswellic acid (AKBA): structure requirements for binding and 5-lipoxygenase inhibitory activity.

1. 5-Lipoxygenase (5-LOX) products from endogenous arachidonic acid in ionophore-stimulated peritoneal polymorphonuclear leukocytes (PMNL) and from exogenous substrate (20 microM) in 105,000 g supernatants were measured. 2. The effects of natural pentacyclic triterpenes and their derivatives on 5-LOX activity were compared with the inhibitory action of acetyl-11-keto-beta-boswellic acid (AKBA), which has been previously shown to inhibit the 5-LOX by a selective, enzyme-directed, non-redox and non-competitive mechanism. 3. The 5-LOX inhibitory potency of AKBA was only slightly diminished by deacetylation of the acetoxy group or reduction of the carboxyl function to alcohol in intact cells (IC50 = 1.5 vs. 3 and 4.5 microM, respectively) and in the cell-free system (8 vs. 20 and 45 microM). 4. beta-Boswellic acid (beta-BA), lacking the 11-keto function, inhibited 5-LOX only partially and incompletely, whereas the corresponding alcohol from beta-BA, as well as amyrin, acetyl-11-keto-amyrin, 11-keto-beta-boswellic acid methyl ester had no 5-LOX inhibitory activity up to 50 microM in either system. 5. beta-BA only partially prevented the AKBA-induced 5-LOX inhibition, whereas the non-inhibitory compounds, amyrin and acetyl-11-keto-amyrin, almost totally antagonized the AKBA effect and shifted the concentration-inhibition curve for the incomplete inhibitor beta-BA to the right. In contrast, the non-inhibitory 11-keto-beta-BA methyl ester exerted no antagonizing effect. 6. The results demonstrate that the pentacyclic triterpene ring system is crucial for binding to the highly selective effector site, whereas functional groups (especially the 11-keto function in addition to a hydrophilic group on C4 of ring A) are essential for 5-LOX inhibitory activity.

5-Lipoxygenase inhibition by acetyl-11-keto-ß-boswellic acid (AKBA) by a novel mechanism.

Acetyl-11-keto-ß-boswellic acid (AKAB) from Boswellia serrata and B. carterii acts directly on purified 5-lipoxygenase of human blood leukocytes at a selective site for pentacyclic triterpenes that is different from the arachidonate substrate binding site. The pentacyclic triterpene ring is crucial for binding to the enzyme, whereas functional groups (11-keto function in addition to a hydrophilic group on C 4 of ring A) are essential for the 5-lipoxygenase activity.

Structure-activity relationships of the nonredox-type non-competitive leukotriene biosynthesis inhibitor acetyl-11-keto-ß-boswellic acid.

Acetyl-11-keto-ß-boswellic acid (AKBA) from Boswellia serrata Roxb. and italics Boswellia carterii Birdw. is the first selective, direct, non-competitive and non-redox-type inhibitor of 5-lipoxygenase, the key enzyme for leukotriene biosynthesis (Safayhi et al., 1992). Previously, we showed that AKBA interacts with the 5-lipoxygenase via a pentacyclic triterpene selective effector site (Safayhi et al., 1995). In order to study the impact of AKBA's functional groups on enzyme inhibition, natural and synthetic analogues of this boswellic acid were tested for 5-lipoxygenase inhibition in intact rat neutrophils (Sailer et al., 1996 a). The results reveal that the carboxylic group of AKBA combined with the 11-keto-group is essential for enzyme inhibition, whereas the acetoxy-group on position C-3 α increases the affinity of AKBA to its effector site. Furthermore, other experiments demonstrated that minor structural modifications could cause a total loss of binding affinity and/or inhibitory activity of these compounds.

Mechanism of 5-lipoxygenase inhibition by acetyl-11-keto-beta-boswellic acid.

The formation of 5-lipoxygenase (EC 1.13.11.34) products from endogenous substrate by intact rat neutrophilic granulocytes and from exogenous arachidonic acid by rat granulocyte 105,000 x g supernatants and affinity chromatography-purified human leukocyte 5-lipoxygenase was inhibited by acetyl-11-keto-beta-boswellic acid (IC50 values of 1.5 microM, 8 microM, and 16 microM, respectively). With other pentacyclic triterpenes lacking the 11-keto function and/or the carboxyl function on ring A (e.g., amyrin and ursolic acid), no 5-lipoxygenase inhibition was observed. The presence of the noninhibitory pentacyclic triterpenes both in intact cells and in the cell-free system caused a concentration-dependent reversal of the 5-lipoxygenase inhibition by acetyl-11-keto-beta-boswellic acid, whereas the inhibitory actions of 5-lipoxygenase inhibitors from different chemical classes (MK-886, L-739,010, ZM-230,487, and nordihydroguaiaretic acid) were not modified. The inhibition by acetyl-11-keto-beta-boswellic acid and the antagonism by noninhibitory pentacyclic triterpenes were not due to nonspecific lipophilic interactions, because lipophilic four-ring compounds (cholesterol, cortisone, and testosterone) neither inhibited the activity of the 5-lipoxygenase nor antagonized the inhibitory action of acetyl-11-keto-beta-boswellic acid. Therefore, we conclude that acetyl-11-keto-beta-boswellic acid acts directly on the 5-lipoxygenase enzyme at a selective site for pentacyclic triterpenes that is different from the arachidonate substrate binding site.

Chamazulene: an antioxidant-type inhibitor of leukotriene B4 formation.

Matricine and its transformation product chamazulene are constituents of chamomile extracts. Both have been demonstrated to exert anti-inflammatory activity in vivo. Since preparations from chamomile are used for the treatment of inflammatory skin and bowel diseases, we studied the effects of these compounds on the leukotriene production in neutrophilic granulocytes. Chamazulene inhibited the formation of leukotriene B4 in intact cells and in the 105,000 x g supernatant fraction in a concentration-dependent manner. The IC50 values were 15 and 10 microM, respectively. Matricine showed no effect up to 200 microM. Chamazulene (IC50: 2 microM), but not matricine, blocked the chemical peroxidation of arachidonic acid. Additionally, matricine (up to 200 microM) had no effects on the cyclooxygenase and 12-lipoxygenase activities in human platelets. Therefore, it is concluded that chamazulene, but not matricine, may contribute to the anti-inflammatory activity of chamomile extracts by inhibiting the leukotriene synthesis and additional antioxidative effects.