Search

Quassinoids often exhibit anti-oxidant and anti-proliferative activity. Emerging evidence suggests that these natural metabolites also display chemopreventive actions. In this study, we investigated the potential for the quassinoid glaucarubulone glucoside (Gg), isolated from the endemic Jamaican plant Castela macrophylla (Simaroubaceae), to display potent cytotoxicity and inhibit human cytochrome P450s (CYPs), particularly CYP1A enzymes, known to convert polyaromatic hydrocarbons (PAHs) into carcinogenic metabolites. Gg reduced the viability of MCF-7 breast adenocarcinoma cells (IC50 = 121 nM) to a greater extent than standard of care anticancer agents 5-fluorouracil, tamoxifen (IC50 > 10 μM) and the tamoxifen metabolite 4-hydroxytamoxifen (IC50 = 2.6 μM), yet was not cytotoxic to non-tumourigenic MCF-10A breast epithelial cells. Additionally, Gg induced MCF-7 breast cancer cell death. Gg blocked increases in reactive oxygen species in MCF-10A cells mediated by PAH benzo[a]pyrene (B[a]P) metabolite benzo[a]pyrene 1,6 quinone, yet down-regulated the expression of genes that promote antioxidant activity in MCF-7 cells. This implies that Gg exhibits anti-oxidant and cytoprotective actions in non-tumourigenic breast epithelial cells and pro-oxidant, cytotoxic actions in breast cancer cells. Furthermore, Gg inhibited the activities of human CYP1A according to non-competitive kinetics and attenuated the ability of B[a]P to induce CYP1A gene expression in MCF-7 cells. These data indicate that Gg selectively suppresses MCF-7 breast cancer cell growth without impacting non-tumourigenic breast epithelial cells and blocks B[a]P-mediated CYP1A induction. Taken together, our data provide a rationale for further investigations of Gg and similar plant isolates as potential agents to treat and prevent breast cancer.

The C-15 senecioyl side chain of brusatol was interchanged with fluorinated acyl groups, and the C-3 hydroxy group of bruceolide was esterified with fluorinated acyl chlorides. These fluorinated quassinoids 11, 12, 13, and 17 showed significant cytotoxic activity against eight human cancer cell lines including small and non-small cell lung, colon, CNS, ovarian and renal cancers, leukemia, and melanoma with 17 being about 100 times more potent than 11, 12, and 13. The activity of 17 was similar to that of bruceantin (1) in this in vitro cell line panel.

Two new quassinoids, bruceanols G [1] and H [4], were isolated from Brucea antidysenterica, and their structures were elucidated by spectral evidence and chemical transformation. Bruceanol G exhibited significant cytotoxicity against the COLO-205 and KB neoplastic cell lines with ED50 values of 0.44 and 0.55 microM, respectively.

Three new quassinoid glucosides, bruceosides D [1], E [2], and F [3], were isolated from Brucea javanica, and their structures were elucidated by spectral evidence and chemical transformation to known compounds. Compounds 1-3 show selective cytotoxicity in the leukemia and non-small cell lung, colon, CNS, melanoma, and ovarian cancer cell lines with log GI50 values in the range of -4.14 to -5.72.

Three new quassinoids, bruceanols D [1], E [2], and F [3], were isolated from Brucea antidysenterica, and their structures were elucidated by spectral evidence and chemical transformation. All of these compounds exhibited cytotoxicity against five human tumor cell lines, malignant melanoma (RPMI-7951), lung carcinoma (A-549), ileocecal adenocarcinoma (HCT-8), epidermoid carcinoma of the nasopharynx (KB), and medulloblastoma (TE-671), and against murine lymphocytic leukemia (P-388).

The search for new anti-cancer drugs is one of the most prominent research areas of natural products. Numerous active compounds isolated from Brazilian Cerrado plant species have been studied with promising results. Aim of the study To investigate the cytotoxic potential of 412 extracts from Brazilian Cerrado plants used in traditional medicine belonging to 21 families against tumor cell lines in culture. Material and method Maceration of 50 plant species resulted in 412 hexane, dichloromethane, ethanol and hydroalcohol extracts. The cytotoxicity of the extracts was tested against human colon carcinoma (HCT-8), melanoma (MDA-MB-435), and brain (SF-295) tumor cell lines, using the thiazolyl blue test (MTT) assay. Bioassay-guided fractionation was performed for one active extract. Results and Conclusions Twenty-eight of the 412 tested extracts demonstrated a substantial antiproliferative effect, at least 85% inhibition of cell proliferation at 50 μg/mL against one or more cell lines. Those extracts are obtained from different parts of Anacardiaceae, Annonaceae, Apocynaceae, Clusiaceae, Flacourtiaceae, Sapindaceae, Sapotaceae, Simaroubaceae and Zingiberaceae. Complete dose-response curves were generated and IC 50 values were calculated for these active extracts against four cell lines HCT-8, MDA-MB-435, SF-295 and HL-60 (leukemia), and their direct cytotoxic effects were determined. In summary, 14 extracts of 13 species showed toxicity in all tested tumor cell lines, with IC 50 values ranging from 0.1 to 19.1 μg/mL. The strongest cytotoxic activity was found for the hexane extract of Casearia sylvestris var. lingua stem bark, with an IC 50 of 0.1 μg/mL for HCT-8, 0.9 μg/mL for SF-295, 1.2 μg/mL for MDA-MB-435, and 1.3 μg/mL for HL-60, and Simarouba versicolor root bark, with an IC 50 of 0.5 μg/mL for HCT-8, 0.7 μg/mL for SF-295, 1.5 μg/mL for MDA-MB-435, 1.1 μg/mL for HL-60. Bioassay-guided fractionation of the last extract led to the isolation of glaucarubinone, which showed pronounced activity against the four cell lines studied. Further studies of the active extracts are necessary for chemical characterization of the active compounds and more extensive biological evaluations.

After removing lipophilic material, the ground root bark of Quassia africana Baill. (Simaroubaceae) was extracted with ethanol 95 %. Partitioning between chloroform, ethyl acetate and water yielded three crude extracts. Pronounced activities were shown by the chloroform and ethyl acetate crude extracts against Herpes simplex, Semliki forest, Coxsackie and Vesicular stomatitis viruses. By repeated column chromatography and preparative thin layer chromatography on silica gel, two quassinoids, i. e., quassin and simalikalactone D were isolated. Structures of the pure compounds were established primarily using NMR spectroscopy. Mass spectral information confirmed the assigned structures. Simalikalactone D was responsible, at least in part, for the high antiviral activity observed for the chloroform crude extract. Quassin showed no activity. For quassinoids the ester group at C-15 and the epoxymethano bridge between C-8 and C-13 appeared to be important structural features in order to exhibit a pronounced antiviral activity.

In an effort to discover new chemotherapeutic/chemopreventive agents from natural sources, brusatol (1) was found to induce HL-60 cellular differentiation, accompanied by strong antiproliferative and cytotoxic effects. A series of natural and semisynthetic quassinoids (1-48) was designed to effect both antiproliferative and differentiation-inducing properties. Compounds were assessed in vitro using the HL-60 promyelocytic cell model. Changes in activity due to structural modification of the core structure glaucarubolone (24) were consistent with activities reported in other cell systems. However, the following were novel SAR findings: (1) semisynthetic analogues with a hydroxylated ring at the beta-position of the ester side chain at C-15 were able to induce cellular differentiation at concentrations lower than those inducing cell growth arrest, and (2) quassinoids inhibiting DNA synthesis with greater efficacy than reducing cellular viability possessed alkyl substitutions at the alpha-position of the C-15 ester side chain. Analogues from this latter group and brusatol (1) and bruceantin (2) inhibited dimethylbenz(a)anthracene-induced preneoplastic lesion formation in a mouse mammary organ culture. The novel finding of 1 and glaucarubolone analogues as potent inducers of differentiation leads to potential novel applications in the field of cancer.

A plasma membrane-associated NADH oxidase of transformed cells was shown to be inhibited by nanomolar and subnanomolar concentrations of the antitumor quassinoid, glaucarubolone. The inhibition was seen with plasma membrane vesicles of HeLa cells at two log orders less glaucarubolone than with plasma membrane vesicles of rat liver. Assignment of a drug-binding site to the external surface of the HeLa cell plasma membrane was supported by findings where full activity of the glaucarubolone in the inhibition of NADH oxidase activity of isolated plasma membrane vesicles and of growth of HeLa cells was given on a molar glaucarubolone basis by an impermeant conjugate of glaucarubolone in which the glaucarubolone moiety was linked via the C-15 hydroxyl to amino polyethyleneglycol (ave Mr 5,000). The activity of the conjugate, and to a lesser extent, of free glaucarubolone was modulated by the redox environment of the cells and of the plasma membrane vesicles. Activity, both in the inhibition of NADH oxidase activity and in the inhibition of growth, was enhanced by oxidizing conditions in the presence of oxidized glutathione compared to reducing conditions in the presence of reduced glutathione.

Growth of Crandall feline kidney cells permanently infected with feline immunodeficiency virus was inhibited by the anti-cancer quassinoid glaucarubolone whereas growth of uninfected cells was not inhibited. Similar results were obtained for human MOLT-4 cells infected with HIV-1. The results suggest that cell lines permanently infected with either the feline or the human lentivirus exhibit growth response characteristics to the quassinoids in common with other cell lines malignantly transformed. In addition the quassinoids may delay viral infection suggesting some commonality between the mechanism responsible for inhibition of the growth of the transformed phenotype and viral infection.

Short-term in vitro assays for tumor promoters and antitumor promoters (Epstein-Barr virus activation test) were carried out for semisynthetic quassinoids (3-7), which were obtained by esterification of the C-15 OH group of deacetylated isobrucein-B (2). All the ester derivatives showed higher antitumor promoting activity than that of the potent compound 2. A compound containing a fluorinated aliphatic ester showed the highest potency.

Short-term in vitro assays for tumor promoters and antitumor promoters (Epstein-Barr virus activation test) were carried out for 14 quassinoids isolated from Ailanthus altissima. Some quassinoids, including ailantinol B, ailantinol C, ailanthone, and shinjulactone A, showed moderate activity at a molar ratio of 1:100 (TPA/quassinoids), and the results led to the elucidation of structure-activity relationships.

Bruceoside C [5], a new quassinoid glucoside, and related compounds were isolated from Brucea javanica, and their structures were elucidated by spectral data. Bruceoside C showed potent cytotoxicities against KB, A-549, RPMI, and TE-671 tumor cells.

The known bruceanic acid A [1] and its methyl ester 2, as well as the new bruceanic acids B [3], C [4], and D [5], have been isolated from Brucea antidysenterica. The structures of 1-5 were elucidated by spectral data. Compound 1 demonstrated cytotoxicity against KB and TE671 tumor cells. Compound 5 was cytotoxic against P-388 lymphocytic leukemia cells.

A series of brusatol, bisbrusatol, and bruceantin esters were examined for their ability to inhibit protein synthesis in P-388 lymphocytic leukemia cells. Compounds which produced high T/C % values (170-272) resulted in ID50 of 5.4-15.5 microM for inhibition of whole cell protein synthesis, ID50 of 1.3-13 microM for inhibition of endogenous protein synthesis in cell homogenates, and ID50 of 1.9-6 microM for inhibition of polyuridine directed polyphenylalanine synthesis using "runoff" ribosomes and a "pH 5" enzyme preparation. The polyuridine directed polyphenylalanine synthesis requires neither initiation nor termination factors, suggesting that quassinoids are exclusively elongation inhibitors. Bruceantin, brusatol, and bisbrusatolyl malonate allowed a runoff of the polyribosomes to 80S free ribosomes. However, formation of the ternary complex and 80S initiation complex were not inhibited by the quassinoids. Thus, these agents do not affect the individual steps leading to the formation of a stable 80S initiation complex in P-388 cells. Brusatol, bruceantin, and bisbrusatolyl malonate inhibited the formation of the first peptide bond between puromycin and [3H]methionyl-transfer RNA bound to the initiation complex, indicating peptidyl transferase activity is inhibited by the quassinoids in P-388 cells. These studies also suggest that the free 80S ribosome is the site of binding by the quassinoid. Ribosomes actively conducting protein synthesis will continue protein synthesis and terminate before the quassinoids bind. This proves quassinoids are elongation inhibitors of tumor cells. A strong correlation was observed between potent antileukemic activity and the ability to inhibit protein synthesis in P-388 lymphocytic leukemia cells.

The quassinoids bruceantin, brucein D, brucein E, bruceoside A, and brusatol significantly inhibited P-388 lymphocytic leukemic cell RNA and protein synthesis in tissue culture. However, DNA synthesis inhibition seemed to correlate more directly with the anti-neoplastic activity of these compounds in the in vivo P-338 survival system. In vitro, brusatol and bruceoside A marginally inhibited 10-day P-388 lymphocytic leukemia DNA polymerase, RNA polymerase, thymidylate synthetase, dihydrofolate reductase, phosphoribosyl pyrophosphate aminotransferase, and cathepsin protease activities. In vivo studies demonstrated similar inhibition and elevated cyclic AMP levels, correlating positively with the antineoplastic activity of individual compounds. Purine synthesis was inhibited drastically by brusatol in vivo, and one key inhibition site in purine synthesis was at phosphoribosyl pyrophosphate aminotransferase, the regulatory enzyme. Histone phosphorylation and ribonucleotide reductase activity also were inhibited marginally by brusatol.

A phase II evaluation of bruceantin was carried out in 15 patients with refractory metastatic breast cancer. All patients had received extensive prior therapy including adriamycin, cytoxan, 5-FU, methotrexate, and a vinca alkaloid. Except for two patients with stable disease, no complete or partial response was observed. Drug toxicity, mainly nonhematologic, was severe, with nausea, vomiting, mild hypotension, and fever being the most frequently encountered.

The Eastern Cooperative Oncology Group (ECOG) conducted a Phase II trial of Bruceantin in malignant melanoma. Twenty-two patients, thirteen without prior cytotoxic chemotherapy, were entered. All patients were evaluable for response and toxicity. Dose limiting toxicity was found to be hypotension during Bruceantin infusion. Other prominent side effects were nausea, vomiting, anorexia, fever, chills, and weakness. Only minor hematologic toxicity was encountered. Two partial responses, both in previously treated patients were observed (response rate — 9%). Bruceantin has only limited activity against malignant melanoma and is unlikely to contribute to systemic therapy of this disease, either as a single agent or in combinations of cytotoxic drugs.

Deux nouveaux quassinoïdes ont été isolés des fruits de #Simaba multiflora$ récoltés en Guyane Française : la 13, 18-dehydro- 6 alpha-senecioyloxy-chaparinne et la 12-dehydro 6 alpha-senecioyloxy-chaparinne. L'élucidation de la structure de ce dernier quassinoïde, actif sur la leucémie murine P388, faut appel aux techniques de la RMN bidimensionnelle, 2D delta delta, dite de corrélation hétéro-nucléaire. (Résumé d'auteur)

During the phase I clinical trial of a new antitumor agent, bruceantin, the pharmacology was studied in 18 cancer patients. The drug was infused intravenously (IV) for 3 h at doses ranging from 1 to 3.6 mg/m2 per day for 5 days. The plasma drug disappearance curves were biphasic, with a fast initial half-life of less than 15 min. The second half-life (t1/2 beta) varied from 0.7 to 38 h among different patients and was not dose-related. The difference between the t1/2 beta on day 1 and that on day 5 was not significant. In patients with normal liver function, the mean plasma concentration at the end of infusion was 22 ng/ml, and the value of the area under the concentration X time curve (AUC) was 111 (ng/ml)h. In contrast, in patients with abnormal liver function the corresponding values were 115 ng/ml and 830 (ng/ml)h, respectively. In addition, these patients had a slower elimination half-life of 10.9 h and a decreased total clearance of 157 ml/min/m2, as compared with 2.6 h and 671 ml/min/m2, respectively, for the normal group. All these differences were statistically significant. Patients with abnormal liver function developed more severe toxicity, including fever, severe nausea, vomiting, and hypotension. Two patients with severe hepatic dysfunction received a reduced dose and developed no toxicity. These results demonstrated the importance of the effects of liver dysfunction on drug disposition and showed that the dosage should be reduced in patients with hepatic dysfunction.

Based on the fact that some known antineoplastic agents possess an ester moiety within their structure, the esters of helenalin, a sesquiterpene lactone, and of brusatol and bruceantin, quassinoids, were synthesized and tested for antileukemic activity in the P-388 screen. These agents gave different T/C % values dependent on the P-388 lymphocytic leukemia strain and the host strain of mice used. Later studies demonstrated that the agents caused different degrees of inhibition of nucleic acid and protein synthesis in the various P-388 strains. The higher the degree of inhibition of precursor incorporation into the nucleic acid or protein, the higher was the T/C % value obtained in a given P-388 strain. The study demonstrates the lack of consistency of P-388 lymphocytic leukemia cell lines used in various laboratories and indicates that the inbred strain of mice is a critical factor in the tolerance of drug toxicity and, thus, T/C % obtained.

The antitumor activities of two ailanthone derivatives with 15 beta-acyloxy side chains were investigated. The cytotoxic activity of 11 beta, 20-epoxy-1 beta, 11 alpha, 12 alpha-trihydroxy-15-beta-[E)-3-methyl-2-octenoyl) oxypicras-3,13(21)-diene-2,16-dione (SUN2071) and 11 beta, 20-epoxy-1 beta, 11 alpha, 12 alpha-trihydroxy-15 beta-[E)-2-undecenoyl) oxypicras-3,13(21)-diene-2,16-dione (SUN0237) was close to that of bruceantin and vincristine. SUN2071 was shown to be a potent inhibitor of protein synthesis in L1210 cultured cells. When administered i.p. to i.p. inoculated P388 leukemia mice, daily treatment with SUN2071 and SUN0237 significantly increased the lifespan (increases in lifespan in excess of 100% were achieved). These increases were comparable to those achieved with vincristine. The therapeutic ratio of SUN2071 was also close to that of vincristine. However, the compounds were ineffective when administered as a single injection. Daily i.p. treatment with SUN2071 demonstrated significant tumor growth inhibition in mice inoculated s.c. with colon-38 and moderate activity against i.p. L1210 leukemia and i.p. B16 melanoma. The compounds were ineffective when tested against the Lewis lung carcinoma and colon-26. In a preliminary toxicological study, SUN2071 at a therapeutic dose in daily consecutive i.p. injection produced leucopenia.

beta-Sitosterol, epilupeo, amarolide-11-acetate, amarolide-2,11-diacetate, ailanthinone and glaucarubinone have been isolated from S. versicolor. The cytotoxic and antileukemic activities of extracts of this plant are due chiefly to glaucarubinone.

Chaparrinone (1) and 6alpha-tigloyoxychaparrinone (2) were shown to be responsible for the antitumor and cytotoxic activities of the root bark of Ailanthus integrifolia ssp. calycina. The structure of the latter compound was established by analysis of spectral data. Compound 2 exhibited a more pronounced biological activity than chaparrinone (1) and demonstrates, for the first time, anticancer activity of a simaroubolide ester substituted at only the 6-position.

A phase I clinical study of bruceantin was conducted in 66 patients with various types of advanced solid tumors to evaluate its toxicity and efficacy. The initial dose of 0.2 mg/m2/day x 5 days repeated at 2-week intervals was progressively increased to a maximum dose of 4.5 mg/m2/day. Hypotension was the dose-limiting toxic effect; it was delayed, cumulative, and occurred more often in patients with abnormal pretreatment liver function. Nausea, vomiting, and fever were common at higher doses, and diarrhea, stomatitis, alopecia, paresthesia, and rash were observed in some patients. The hematologic toxicity of bruceantin was moderate at high doses and was manifested mainly as thrombocytopenia; it was more severe in patients with abnormal hepatic and renal functions. No objective tumor regressions were observed. The recommended dose of bruceantin is 3.5 mg/m2/day x 5 days for phase II studies.

A microbiologic assay of the agar diffusion type, employing a strain of yeast tentatively identified as Candida macedoniensis, was developed to study the disposition, excretion, and metabolism of the antitumor agent bruceantin (BN; NSC 165563) in the mouse. Bioautographic studies showed the assay to be specific for BN. Normal and tumor-bearing male BDFl mice were studied. Tumor-bearing mice were implanted sc with 10 6 L1210 ascites cells and held for 7 days prior to dosing. Average tumor weight was 307 mg and advanced generalized disease was evident. Three groups of 5 or 10 mice were injected iv with BN (1.5 mg/kg; approx. LD10). At various times after injection, blood, tissues, urine, and feces were obtained and extracted with chloroform to recover BN. Recovery of BN was in the range of 91 to 121%. Disposition and excretion of BN were similar for normal and tumor-bearing mice. Decay of BN in blood was biphasic with α-phase half-lives of 5 to 6 min. Estimated half-lives for the β-phase were possibly >0.5 day. Average zero time intercepts for α-and β-phases were 550 and 51 ng/ml respectively. Higher levels of BN were found in lung, pancreas, intestine, and spleen (1–6 μg/g) than in liver, kidney, and tumor (0.3–0.5 μg/g) after 15 min. Concentrations of BN in brain and peritoneal fat were below detectable limits (0.1 μg/g tissue). Urine and fecal excretion of BN accounted for 2% of the dose after 24 h. In vitro metabolism studies using a postmitochondrial microsome fraction of liver, lung, and kidney suggest that bruceantin is inactivated by an NADPH-dependent enzyme present in liver but not in lung or kidney.

The quassinoids (brusatol, bruceantin, bisbrusatolyl esters, and bisbruceantinyl esters of succinic and malonic acids) were observed not to be universal protein synthesis inhibitors. Rather, they were selective for both the types of cancers, e.g., P-388 lymphocytic leukemia, Ehrlich and hepatoma carcinoma and L-1210 lymphoid leukemia, as well as types of normal tissues (e.g., lymphocytes), in which they demonstrated protein synthesis inhibition. The data suggest that the observed difference in the magnitude of protein synthesis inhibition of two P-388 lymphocytic leukemia cell lines by the quassinoids was at the ribosomal levels, whereas the observed difference in normal livers from various strains of mice involve differences in cell membrane transport of the quassinoids into the various tissues. Detailed studies indicated that the mode of action of the quassinoids as protein synthesis inhibitors was identical in all of the cells where inhibition was observed; i.e., the elongation step of protein synthesis was blocked by the quassinoids. The data derived from assays for polyuridine-directed polyphenylalanine synthesis of isolated ribosomes demonstrated that the ID50 values obtained were consistent with the observed inhibition of whole cell protein synthesis inhibition for P-388 cells, as well as for BDF1 and DBA/2 liver cells. The ID50 values obtained from various cells, e.g., P-388 cells and normal liver, were all in the microM range and are consistent with previously published values for the quassinoids in the rabbit reticulocyte and yeast systems.

The South American Simaba cuspidata Spruce and North Indian Ailanthus grandis Prain were investigated as sources of potentially useful antineoplastic agents. Both of these Simaroubaceae plant species were found to produce 6 alpha-tigloyloxychaparrinone (4a) and the new quassinoid 6 alpha-tigloyloxychaparrin (3b). The latter structure was determined by interpretation of spectral data and oxidation to 6 alpha-tigloyloxychaparrinone (4a). While both glycol 3b and alpha-ketol 4a were found to significantly inhibit growth of the murine P388 lymphocytic leukemia cell line, only the alpha-ketol (4a) inhibited growth of the corresponding in vivo system.

Information on the metabolism of antitumor drugs of plant origin is not extensive for most compound types. This can be attributed to the fact that really only two groups of plant-derived drugs are broadly used clinically, the alkaloids of the vincristine-vinblastine family and the epipodophyllotoxin derivatives etoposide and teniposide. An important factor in fewer plant-derived drugs than microbial or synthetic ones being developed is the relatively small investment in plant research in the U.S. pharmaceutical industry. We have tried to cover, in addition to clinically useful drugs, those compounds which have been in trials recently (even if inactive) and those which are currently in development. Some compounds which might otherwise have been included were omitted because no metabolic studies have been reported. These include maytansine, camptothecin, tetrandrine, thalicarpine and acronycine. We have in some cases included data on metabolism of related compounds which do not have antitumor activity or which have not been developed to clinical study where we felt that this aided the perspective of metabolism of compounds as a group or class. The drug development program with which we are affiliated at the National Cancer Institute is heavily oriented towards seeking target-specific or at least highly target-selective compounds using a screen composed of batteries of human tumor cell lines organized by tissue type (Boyd et al., 1986). The screen looks for selective cytotoxicity towards target cells which could result from such possibilities as specific receptors being present, greater sensitivity of enzyme systems or isozymes, permeability differences or transport specificity, or tissue specific metabolic activation. This last situation, tissue-specific metabolic activation, is discussed in Section 6 and may become of increasing significance as the emphasis in cancer drug discovery turns away from the search for broad spectrum agents, which often have broad toxicity as well, towards more selective ones. We therefore anticipate that metabolic studies of antitumor agents will increasingly become a key component of decisions on development of new drugs towards clinical trials and decisions on clinical protocols.

A series of quassinoid compounds was investigated to determine the structure-activity relationships for the binding and inactivation steps associated with the inhibition of protein synthesis by these compounds in a rabbit reticulocyte system. The binding of these compounds to rabbit reticulocyte run-off ribosomes was measured in a competition-binding assay using [14C]trichodermin as the competing ligand. Previous studies have shown that trichodermin and quassionoids compete for a single binding site on yeast ribosomes. Our experiments confirmed this behavior with rabbit reticulocyte ribosomes and ind9icated that binding equlibrium could be reached in 2.5 min at 30°C. The comparative binding data indicated the following: (a) The nature of the side-chain at position 15 has very little effect on the association of these compounds with rabbit reticulocyte ribosomes. There is one exception to this general rule, however. saturation of the double bond in brusaatol (compound II) to give dihydrobrusatol (compound IV) increases the d form 1.6μM to 8.7 μM. (b) Easy access to the A ring is required for optimal binding. The addition of any bulky hydrophobic group to the A ring increased the Kd significantly. However, the double bond in the A ring does not appear to be required for binding since the Kd values for dihydrobrusatol (compound IV) and tetrahydrobrusatol (compound VI) to the corresponding lactol (compound X) has only a minor effect on Kd. In order to determine whether inactivation of rabbit reticulocyte ribosomes had any structural requirements which differed from those required for simple binding, we also compared the concentrations of these compounds required to give 50% inhibition of protein synthesis (ID50) in rabbit reticulocye lysates. Previous studies had indicated that binding of quassinoids to the peptidyl transferase center and inhibition of peptide bond formation was probably the only mode of action of these compounds. Our experiments indicated that the ID50 values for several of thse compounds were the same in whole reticulocytes and reticulocyte lysates, suggesting that transport and/or metabolism of these compounds was probably not limiting for their inhibitory effects. The inhibition studies indicated that (a) The nature of the side-chain at position 15 had no detectable effect on the inhibition: (b) the double bonds in the A ring and lactone ring are probably involved in the inactivation of rabbit reticulocyte ribosomes.

Yadanziosides M [3] and P [1] were isolated from Brucea antidysenterica. Their structures were elucidated by spectral data. Bruceantinoside B reported previously was reinvestigated and revealed to be a mixture of yadanzioside P and bruceantinoside C [2]. The structure of yadanzioside P was found to be identical to that of bruceantinoside B.

Inhibition of the uptake of 3 H-thymidine into TLX-5 mouse lymphoma cells has been used to assess the cytotoxicity of extracts and fractions of BRUCEA JAVANICA. Six bruceolides, two dehydrobruceolides and a dihydrobruceolide have been isolated from roots, fruits and stems. Bruceine A was the major bruceolide of the roots and of the fruits. The potent antineoplastic bruceolides, bruceantin and bruceantinol, previously isolated from B. ANTIDYSENTERICA and B. GUINEENSIS have been isolated together with the novel compounds dehydrobruceine A and dihydrobruceine A. Bruceantarin and dehydrobruceine B, not previously reported from this species, and bruceines B and C were also isolated. The identification of these nine compounds by means of their UV, EIMS, CIMS, FDMS and 1 H NMR spectra is discussed.

An investigation of the Guyana plant Simarouba amara Aubl. (Simaroubaceae) for antineoplastic quassinoids led to isolation and structural determination of the new quassinoids 2'-acetylglaucarubine (1a) and 13,18-dehydroglaucarubinone (2). The previously known 2'-acetylglaucarubinone (3a) and glaucarubinone (3b) were also obtained. The new quassinoid 2 was found significantly to inhibit growth of the murine lymphocytic leukemia P388.

Brusatol, a quassinoid with potent antineoplastic activity against P-388 lymphocytic leukemia cell proliferation, significantly inhibited P-388 cell hexokinase, phosphofructokinase, malic dehydrogenase, and succinic dehydrogenase. Mitochondrial oxidative phosphorylation, basal, and adenosine diphosphate-stimulated respiration, utilizing succinate and alpha-ketoglutarate as the substrate, was suppressed significantly by in vivo treatment with brusatol. However, brusatol treatment had no effect on liver oxidative phosphorylation. Brusatol greatly increased P-388 cyclic AMP levels but had no effect on liver cyclic nucleotides. Similar inhibitory effects on P-388 cell oxidative phosphorylation were found in vitro with brusatol, bruceoside A, and bruceantin. Brusatol had no effect on adenosine triphosphatase activity or on uncoupling of oxidative phosphorylation. Rather, brusatol appeared to increase the concentration of reduced mitochondrial electron-transport cofactors, thereby blocking aerobic respiration. A proposed mechanism of action is discussed.

Fractionation of Simaba multiflora A. Juss. guided by bioassay has resulted in the isolation of a new antileukemic quassinoid 6alpha-senecioyloxychaparrinone (2) and the previously reported quassinoid chaparrinone (3). The structure of the former has been established by spectral and chemical methods. Compound 2 has high anti-neoplastic activity against several mouse leukemia systems (P-388, L-1210 and in solid-tumor B-16 melanoma). This demonstrates for the first time that the presence of a C-15 ester function is not required for antineoplastic activity in the quassinoids.

Three cytotoxic, quassinoid glycosides, the new bruceantinoside C [1] and the known yadanziosides G [2] and N [3] were isolated from the stem of Brucea antidysenterica. The structures of 1-3 were determined from their spectral data.

Several quassinoids, obtained by isolation and derivatization from Simaba multiflora and Soulamea soulameoides, were evaluated for growth inhibitory and insecticidal effects against the tobacco budworm (Heliothis virescens) and for antifeedant effects against H. virescens and the fall armyworm (Spodoptera frugiperda). The relative activity of the quassinoids as insect growth inhibitors generally paralleled their known relative potency as antileukemic and cytotoxic agents.

Patients with objectively measurable soft tissue sarcoma, bone sarcoma, or mesothelioma who had failed at least one prior chemotherapy regimen received either bleomycin (20 U/M22 i.v. day 1 each week), chlorozotocin (150 mg/M2 i.v. q6 weeks), MGBG (500 mg/M2 i.v. each week, escalated in 50 mg/M2 weekly increments to a maximum dose of 700 mg/M2), or bruceantin (5.5 mg/M2 days 1, 8, 15, and 22, with cycles repeated every 6 weeks). One hundred eighty patients were evaluable: 53 on bleomycin, 51 on chlorozotocin, 38 on MGBG, and 38 on bruceantin. Two partial responses resulted from bleomycin, and one each from chlorozotocin and MGBG. Both responders on bleomycin had mesothelioma. Seventy-four percent of the patients were of ECOG performance status 0 or 1, and over half on each arm had moderate or worse toxicity. At these doses and schedules, none of the four drugs tested was active against previously treated sarcomas. Bleomycin, however, should be considered for further evaluation in mesothelioma patients.

The structure of soularubinone 3, a new antileukemic quassinoid isolated from the leaves of Soulamea tomentosa (Brongn, and Gris), has been established by spectral and chemical methods. It has been shown to be the C-15 beta-hydroxy-isovaleric ester of glaucarubolone. Soularubinone shows significant antineoplastic activity against mouse leukemia P-388 and inhibits cell transformation induced by Rous sarcoma virus. The known quassinoid chaparrinone I has also been isolated.

The antitumoral agents PR toxin, bouvardin and grandilactones A and B inhibit the growth of HeLa cells, and abolish polyphenylaline synthesis by polyuridylic acid in acellular eukaryotic systemns. However, only bouvardin and the PR toxin inhibit polypeptide synthesis in polysomes using endogenous m-RNA. In addition, grandilactones A and B inhibit the "fragment" reaction, which indicates that they affect specifically peptide bond formation, and the lack of effect on polysomes of these compounds would be due to the fact that, in polysomes, the substrate is bound to either the donor on the acceptor site. Apparently, the PR toxin acts by blocking the enzymatic binding to 80 S ribosomes of (3H)phenylalanyl-tRNA. The inhibitory action of bouvardin seems to be result of the selective blocking of peptidyl-tRNA translocation dependent on elongation factor EF-2 and GTP.

A radioimmunoassay for a new anticancer drug, bruceantin, has been developed using [ 3 H]acetylbruceantin and antibody induced by immunizing rabbits with succinylbruceantin-bovine serum albumin conjugates. [ 3 H]Acetylbruceantin was synthesized by reacting bruceantin with [ 3 H]acetyl anhydride. The assay is simple and reproducible. The standard curve was linear on a logit-log plot, and the lower limit of sensitivity of the assay was 1 ng/ml. Using this assay, drug levels were easily determined in tissues of experimental animals following bruceantin administration. The assay procedure does not require sample extraction for plasma, urine, and bile. Bruceantin in other tissues can be extracted quantitatively with ethanol before being measured by the radioimmunoassay.