Supplementary Legends from The Taccalonolides: Microtubule Stabilizers That Circumvent Clinically Relevant Taxane Resistance Mechanisms

The taccalonolides are a unique class of microtubule stabilizers that do not bind directly to tubulin. Three new taccalonolides, Z, AA, and AB, along with two known compounds, taccalonolides R and T, were isolated from Tacca chantrieri and Tacca integrifolia. Taccalonolide structures were determined by 1D and 2D NMR methods. The biological activities of the new taccalonolides, as well as taccalonolides A, B, E, N, R, and T, were evaluated. All nine taccalonolides display microtubule stabilizing activity, but profound differences in antiproliferative potencies were noted, with IC(50) values ranging from the low nanomolar range for taccalonolide AA (32 nM) to the low micromolar range for taccalonolide R (13 μM). These studies demonstrate that diverse taccalonolides possess microtubule stabilizing properties and that significant structure-activity relationships exist. In vivo antitumor evaluations of taccalonolides A, E, and N show that each of these molecules has in vivo antitumor activity.

A new benzoquinone-type retro-dihydrochalcone, named evelynin, was isolated from the roots and rhizomes of Tacca chantrieri. The structure was elucidated on the basis of the analysis of spectroscopic data and confirmed by a simple one-step total synthesis. Evelynin exhibited cytotoxicity against four human cancer cell lines, MDA-MB-435 melanoma, MDA-MB-231 breast, PC-3 prostate, and HeLa cervical carcinoma cells, with IC50 values of 4.1, 3.9, 4.7, and 6.3 μM, respectively.

Abstract The taccalonolides are a class of structurally and mechanistically distinct microtubule-stabilizing agents isolated from Tacca chantrieri. A crucial feature of the taxane family of microtubule stabilizers is their susceptibility to cellular resistance mechanisms including overexpression of P-glycoprotein (Pgp), multidrug resistance protein 7 (MRP7), and the βIII isotype of tubulin. The ability of four taccalonolides, A, E, B, and N, to circumvent these multidrug resistance mechanisms was studied. Taccalonolides A, E, B, and N were effective in vitro against cell lines that overexpress Pgp and MRP7. In addition, taccalonolides A and E were highly active in vivo against a doxorubicin- and paclitaxel-resistant Pgp-expressing tumor, Mam17/ADR. An isogenic HeLa-derived cell line that expresses the βIII isotype of tubulin was generated to evaluate the effect of βIII-tubulin on drug sensitivity. When compared with parental HeLa cells, the βIII-tubulin–overexpressing cell line was less sensitive to paclitaxel, docetaxel, epothilone B, and vinblastine. In striking contrast, the βIII-tubulin–overexpressing cell line showed greater sensitivity to all four taccalonolides. These data cumulatively suggest that the taccalonolides have advantages over the taxanes in their ability to circumvent multiple drug resistance mechanisms. The ability of the taccalonolides to overcome clinically relevant mechanisms of drug resistance in vitro and in vivo confirms that the taccalonolides represent a valuable addition to the family of microtubule-stabilizing compounds with clinical potential. [Cancer Res 2008;68(21):8881–8]

The interactions of microtubules with most compounds described as stabilizing agents have been studied. Several of them (lonafarnib, dicumarol, lutein, and jatrophane polyesters) did not show any stabilizing effect on microtubules. Taccalonolides A and E show paclitaxel-like effects in cells, but they were not able to modulate in vitro tubulin assembly or to bind microtubules, which suggests that other factors are involved in their cellular effects. The binding constants of epothilones, eleutherobin, discodermolide, sarcodictyins, 3,17 beta-diacetoxy-2-ethoxy-6-oxo-B-homo-estra-1,3,5(10)triene, and dictyostatin to the paclitaxel site; the critical concentrations of ligand-Induced assembly; and their cytotoxicity in carcinoma cells have been measured, and correlations between these parameters have been determined. The inhibition of cell proliferation correlates better with the binding enthalpy change than with the binding constants, suggesting that large, favorable enthalpic contribution to the binding is desired to design paclitaxel site drugs with higher cytotoxicity.

We tested the hypothesis that breast-vs. formula-feeding differentially affects the enzymatic activity of three sterol hydroxylases critical in the initial steps of bile acid formation. Thirty baboons were either breast-fed or formula-fed for the first 14 wk of life before weaning to baboon chow. At 14 and 34 wk of age, liver biopsies were assayed for cholesterol 7α-hydroxylase (CYP7A1), 27-hydroxycholesterol-7α-hydroxylase (CYP7B1), and cholesterol 27-hydroxylase (CYP27A1). We also determined the kinetics of 3H-27-hydroxycholesterol (27-OHC) turnover in vivo at both ages. At 14 wk of age, hepatic CYP7A1 activity was low but sevenfold higher among formula-fed vs. breast-fed baboons. By 34 wk, CYP7A1 activity had increased nearly 10-fold in both infant diet groups, and the sevenfold difference in CYP7A1 between previously breast-and formula-fed animals persisted. There were no differences in CYP7B1 activities between infant diet groups at either 14 or 34 wk of age although the activity increased in both groups by about 50% from 14 to 34 wk. CYP27A1 activity also increased between 14 and 34 wk of age, and, compared with CYP7A1, relatively small differences in CYP27A1 activity due to infant diet were observed at each age. Plasma 27-OHC turnover had a half-time of 2–4 min. We had previously reported that after weaning, the total bile acid synthesis rate was higher among baboons that were formula-fed than among breast-fed animals. The present results suggest that this difference is most likely due to significantly higher CYP7A1 activity among formula-fed vs. breast-fed animals.

During the course of a mechanism-based screening program designed to identify new microtubule-disrupting agents from natural products, we identified a crude extract from Tacca chantrieri that initiated Taxol-like microtubule bundling. Bioassay-directed purification of the extract yielded the highly oxygenated steroids taccalonolides E and A. The taccalonolides caused an increased density of cellular microtubules in interphase cells and the formation of thick bundles of microtubules similar to the effects of Taxol. Mitotic cells exhibited abnormal mitotic spindles containing three or more spindle poles. The taccalonolides were evaluated for antiproliferative effects in drug-sensitive and multidrug-resistant cell lines. The data indicate that taccalonolide E is slightly more potent than taccalonolide A in drug-sensitive cell lines and that both taccalonolides are effective inhibitors of cell proliferation. Both taccalonolides are poorer substrates for transport by P-glycoprotein than Taxol. The ability of the taccalonolides to circumvent mutations in the Taxol-binding region of beta-tubulin was examined using the PTX 10, PTX 22, and 1A9/A8 cell lines. The data suggest little cross-resistance of taccalonolide A as compared with Taxol, however, the data from the PTX 22 cell line indicate a 12-fold resistance to taccalonolide E, suggesting a potential overlap of binding sites. Characteristic of agents that disrupt microtubules, the taccalonolides caused G(2)-M accumulation, Bcl-2 phosphorylation, and initiation of apoptosis. The taccalonolides represent a novel class of plant-derived microtubule-stabilizers that differ structurally and biologically from other classes of microtubule-stabilizers.

26-Hydroxycholesterol (26OHC), a major oxysterol in human blood, is believed to play an important role in reverse cholesterol transport, bile acid formation, and regulation of various cellular processes. Using isotope dilution mass spectrometry, we measured plasma 26OHC concentrations in baboons fed either a high cholesterol/saturated fat (HC-SF) or normal chow diet. Plasma 26OHC levels in baboons were comparable to those reported for humans and were positively correlated with plasma cholesterol concentrations. Animals on the HC-SF diet had significantly higher 26OHC levels (0.274±0.058 μM, mean±S.D.) than those on the chow diet (0.156±0.046 μM). In separate experiments, [3H]26OHC was injected into four tethered baboons, and multiple blood samples drawn over a 1-h period were analyzed for [3H]26OHC and 26OHC. Fitting the specific radioactivity data to a two-pool compartmental model indicated a rapidly turning over plasma compartment (t1/2 2.9–6.0 min) and a second compartment with slow turnover (t1/2 76–333 min). The calculated 26OHC production rate was 2.5 μmol/kg body weight/day. Assuming all 26OHC is converted to bile acids, the 26OHC production rate corresponds to about 10% of total bile acid production in adult baboons. These results indicate that rapid turnover of plasma 26OHC at submicromolar concentrations could significantly contribute to bile acid synthesis.

In a series of experiments over the past 20 years, we have demonstrated long-term deferred effects of infant nutrition, particularly breast- as compared with formula-feeding and overfeeding as compared with normal or underfeeding, on serum HDL-cholesterol concentrations, adiposity, and atherosclerosis in the baboon, a large nonhuman primate. Low HDL-cholesterol levels and obesity are associated with accelerated progression of atherosclerosis and with increased risk of coronary heart disease in humans. We have observed other deferred effects of infant nutrition on bile acid metabolism, enzyme activities, and water and electrolyte balance, some of which may be physiologically related to HDL-cholesterol levels or to adiposity. The occurrence of these deferred effects suggests that infant nutrition may program other metabolic systems for life, and that these effects may contribute to other chronic diseases of adults. Although our understanding of the mechanisms by which infant diet regimens affect adult metabolism is meager, it is important to identify these mechanisms because they are likely to provide valuable clues to the causes and ultimately may contribute to the long-range prevention of those diseases.

We estimated the effects of breast- and formula-feeding on cholesterol and bile acid metabolism for 1.5 years after weaning in 35 newborn baboons that were breast-fed (n = 12) or fed one of two formulas with high (n = 11) or low (n = 12) polyunsaturated/saturated (P/S) fatty acid composition. Infants were weaned at 15 weeks to a high cholesterol, saturated fat diet. Because formula P/S ratio did not affect any variable for 1.5 years after weaning, the data were averaged for the two formula groups. After weaning, serum cholesterol and lipoprotein cholesterol concentrations among the infant diet groups were not different until after 52 weeks of age. From 70 to 97 weeks of age, serum cholesterol and high density lipoprotein-2 (HDL2)-cholesterol (HDL2-C) concentrations were lower (P < 0.04) among baboons that were breast-fed as infants compared with those fed formulas. We observed no significant postweaning differences in low density lipoprotein (LDL)-C, HDL3-C, or serum apolipoprotein A-I, B, or E concentrations. At 97 weeks of age baboons that were breast-fed until 15 weeks compared with those formula-fed had a 25% lower total bile acid synthetic rate (36.6 vs. 48.6 mumol/day per kg body weight, P < 0.02) due principally to a 29% lower cholic acid synthetic rate (23.2 vs 32.5 mumol/day per kg body weight, P < 0.004). Baboons breast-fed as infants had a 44% higher hepatic LDL-receptor mRNA concentration than those formula-fed (1.45 vs. 1.01 pg mRNA/micrograms total RNA, P < 0.003).(ABSTRACT TRUNCATED AT 250 WORDS)

To determine whether the amino acid pattern of human milk is unique, we compared the amino acid pattern of human milk with the amino acid patterns of the milks of great apes (chimpanzee and gorilla), lower primates (baboon and rhesus monkey) and nonprimates (cow, goat, sheep, llama, pig, horse, elephant, cat and rat). Amino acid pattern was defined as the relative proportion of each amino acid (protein-bound plus free) (in mg) to the total amino acids (in g). Total amino acid concentration was lower in primate milk than in nonprimate milk. There were commonalities in the overall amino acid pattern of the milks of all species sampled; the most abundant amino acids were glutamate (plus glutamine, 20%), proline (10%) and leucine (10%). Essential amino acids were 40%, branched-chain amino acids 20%, and sulfur amino acids 4% of the total amino acids. The amino acid pattern of human milk was more similar to those of great apes than to those of lower primates. For example, cystine was higher and methionine was lower in primate milks than in nonprimate milks, and in great ape and human milks than in lower primate milks. Because the milk amino acid patterns of the human and elephant, both slow-growing species, were dissimilar, the amino acid pattern of human milk seems unrelated to growth rate.

To determine whether the amino acid composition of milk changes during lactation, we compared the amino acid pattern (concentration of each individual amino acid relative to the total amino acid concentration) of colostrum with that of mature milk in six mammalian species. In the human, horse, pig and cow, the pattern of amino acids changed between colostrum and mature milk: glutamate, proline, methionine, isoleucine and lysine increased; cystine, glycine, serine, threonine and alanine decreased. In these four species, the total amino acid concentration also decreased 75% between colostrum and mature milk. In the baboon ( Papio cynocephalus anubis and Papio cynocephalus anubis/Papio cynocephalus cynocephalus ) and rhesus monkey ( Macaca mulatta ), however, there was little change in the pattern of amino acids between colostrum and mature milk, and total amino acid concentration decreased only about 25% between colostrum and mature milk. Mature milk rather than colostrum was the most similar among the three primates in both amino acid pattern and total amino acid concentration. We conclude, in those species in which total amino acid concentrations decline substantially between colostrum and mature milk, amino acid patterns also change. The presence of a change in amino acid pattern and total amino acid concentration during lactation appears to be unrelated to phylogenetic order.

We tested, with 14-wk-old baboons before weaning, the hypothesis that bile acid metabolism is differentially affected by breast feeding or by feeding formulas with a high polyunsaturated:saturated fatty acid ratio or with a low ratio, similar to that of breast milk. Bile lipid content, bile acid pool size, fractional turnover rate, synthetic rate and conjugate composition were measured in a single bile sample 9 d after an injection on d 1 of a mixture of [14C]cholic and [14C]chenodeoxycholic acids and an injection of a mixture of [3H]cholic acid and [3H]chenodeoxycholic acid on d 8. The principal biliary bile acid was chenodeoxycholic acid. The only difference in chenodeoxycholic acid metabolism among the infant diet groups was a lower chenodeoxycholic acid synthetic rate in baboons fed the low polyunsaturated:saturated formula compared with those fed the high polyunsaturated:saturated formula or breast-fed. Cholic acid metabolism was significantly affected by infant diet: breast-fed infants had a smaller cholic acid pool size, lower cholic acid percentage of total bile acids, higher cholic acid glycine:taurine conjugate ratio and larger cholic acid fractional turnover rate than formula-fed animals. The polyunsaturated:saturated fatty acid ratio in the formulas did not significantly affect these variables. These results show that differences in bile acid metabolism between breast- and formula-fed infant baboons are limited principally to cholic acid. These differences likely are due to factors other than fatty acid saturation.

We previously reported that female baboons overfed during infancy were not fatter at weaning, but developed hypertrophic obesity after puberty. To clarify the mechanisms of this dietary effect on adipocyte hypertrophy, we determined the effects of infant overfeeding on preweaning plasma hormone and triglyceride levels and their relationship with fat cell volume at weaning (19 weeks of age). Newborn female baboons from 3 sires and 24 dams were fed either 280 kilojoules (normally fed; n = 12) or 395 kilojoules (overfed; n = 10) per 100 g Similac formula for 18 weeks. Both formulas contained 9.2%, 43.1%, and 48.5% of calories as protein, carbohydrate, and fat, respectively. During the first 9 weeks, overfed infants had significantly higher fasting and postprandial insulin, total T3, and free T3 concentrations; lower cortisol levels; and lower excretion of urinary 17-hydroxycorticosteroids (17-OHCS) than normally fed infants. These effects were no longer significant at 17-18 weeks. Infant diet did not influence fasting and postprandial plasma triglyceride levels, and fat cell volume was not influenced by energy intake. However, fat cell volume was positively associated with postprandial triglyceride concentrations and inversely associated with postmeal nadir cortisol levels. These results demonstrate that infant overfeeding initiates early alterations in insulin, T3, free T3, and cortisol, but these effects persist only as long as there is a significant increase in energy intake.

5α-Cholest-8(14)-en-3β-ol-15-one is a potent inhibitor of cholesterol biosynthesis which has significant hypocholesterolemic activity upon oral administration to rodents and nonhuman primates. In the present study the metabolism of the 15-ketosterol has been investigated after the oral administration of a mixture of [2,4-3H]5α-cholest-8(14)-en-3β-ol-15-one and [4-14C]cholesterol to 8 baboons. Blood samples were obtained at 4, 8, 12, 16, and 24 h after administration of the labeled sterols. Clear differences in the time courses of the levels of 3H and 14C in plasma were observed, 3H in plasma showed maximum values at 4 to 8 h, whereas maximum values for the levels of 14C were observed much later. 3H in plasma was shown to be primarily in the form of its metabolites, i.e. esters of the 15-ketosterol, cholesterol, and cholesteryl esters. The levels of the 15-ketosterol and of each of these metabolites showed different changes with time. The labeled cholesterol (and the cholesterol moiety of the cholesteryl esters), formed from the [2,4-3H]-15-ketosterol, was characterized by chromatography and by purification by way of its dibromide derivative. At 24 h after the administration of the labeled sterols, the distribution of 3H in plasma lipoproteins fractions paralleled that of 14C, with most of the 3H and 14C in high density lipoproteins (HDL) and low density lipoproteins (LDL). Almost all of the 3H in HDL and in LDL was found as cholesterol, cholesteryl esters and esters of the 15-ketosterol. The distribution of 3H in HDL in LDL of the free 15-ketosterol, esters of the 15-ketosterol, cholesterol, and cholesteryl esters was similar to that of plasma, thereby indicating no unusual concentration of any of the 3H labeled components in HDL or LDL.

We analyzed the effects of dietary cholesterol, type of dietary fat, sex and sire progeny family on lecithin-cholesterol acyltransferase activity in 80 adult baboons. The animals were the progeny of 80 dams and 6 sires and were randomly assigned at birth to breast feeding or to one of three formulas containing 0.02, 0.30 or 0.60 mg cholesterol/ml. After weaning at 4 months of age the animals were fed one of four diets that were either high or low in cholesterol with 40% of the calories from either saturated or unsaturated fat. The fractional and molar rates of lecithin-cholesterol acyltransferase activity were measured at 7-8 years of age by an HPLC method. Infant diet (breast vs. formula feeding or level of cholesterol in formula had no effect on enzyme activity later in life. The adult diets that were high in cholesterol decreased the fractional lecithin-cholesterol acyltransferase rate by 20% / compared to diets low in cholesterol (7.89 vs. 9.84%/h, P less than 0.002), but dietary cholesterol did not affect the molar activity. Animals fed the high cholesterol diets had higher unesterified cholesterol concentrations compared to those fed the low cholesterol diets (38.1 mg/dl vs. 31.6 mg/dl, P less than 0.0001). The molar lecithin-cholesterol acyltransferase rate was increased 13% by saturated compared to unsaturated fat (83.3 vs. 73.6 nmol/h per ml plasma, P less than 0.07), but no effect of dietary fat was observed on the fractional enzyme activity. Females compared to males had significantly higher fractional (10.9 vs. 7.14%/h, P less than 0.0001) and molar lecithin-cholesterol acyltransferase activities (99.3 vs. 61.7 nmol/h per ml plasma, P less than 0.0001). After adjustment for the effects of diet and sex we observed differences in the fractional activity (range, 7.2-10.8%/h, P less than 0.04) and in the molar rate (range, 63.6-99.8 nmol/h per ml plasma, P less than 0.07) among the six sire progeny groups. The differences among sire progeny groups are evidence for genetic differences in lecithin-cholesterol acyltransferase activities among the baboon families.

Cholesterol turnover was studied in four baboons by injecting [14C]cholesterol 186 days and [3H]cholesterol 4 days before necropsy, and fitting a two- or three-pool model to the resulting specific activity-time data. At necropsy, cholesterol mass and specific activity were determined for the total body (minus the central nervous system) and for many tissues. A pool model permits the estimation, from the plasma specific activity-time curve alone, of total body cholesterol within a limited range, depending upon the extent of side pool synthesis. The principal aim of this study was to estimate the extent of cholesterol synthesis in the side pools of the model, by computing the amount of side pool synthesis needed to equal the measured total body cholesterol. Central pool synthesis varied from 61 to 89% of the total cholesterol production rate. Thus, approximately 25% (11 to 39%) of the production rate arose from peripheral (pool 3 for the three-pool, and pool 2 for the two-pool model) cholesterol synthesis. Moreover, the finding that the measured total body cholesterol fell within the range obtained from the kinetic analysis by using reasonable assumptions (namely, that zero or that half the production rate occurred in the side pools), provides evidence for the physiological validity of the model. A second aim of this study was to explore cholesterol turnover in various tissues. A pool model predicts that rapidly turning over tissues will have higher specific activities at early times and lower specific activities at later times after injection of tracer relative to slowly turning over tissues, except where significant synthesis occurs. Tissues were ranked 1 to 17 for 3H and 17 to 1 for 14C cholesterol specific activity values. Except for the GI tract and testis, the tissues had similar ranks for both 3H and 14C, further validating model predictions. Results in all four baboons were similar. Turnover rates for the different tissues loosely fell into three groups which were turning over at fast, intermediate, and slow rates. Finally, the magnitude of variation of cholesterol specific activity was moderate for several distributed tissues (fat, muscle, arteries, and the alimentary tract), but was small for liver. Cholesterol turnover in serial biopsies of skin, muscle, and fat could, however, be fitted with a single pool to estimate tissue turnover rates.

Fecal neutral steroids were fractionated by high performance liquid chromatography (HPLC) into three major fractions: 5 beta-H, 3-keto steroids; 5 beta-H, 3 beta-hydroxy steroids; and 5 alpha-H and delta 5-3 beta-hydroxy steroids. This separation was achieved in about 10 minutes, with greater than 97% recovery of standards in each fraction. Gas-liquid chromatographic quantitation of fecal steroids fractionated by either HPLC or thin-layer chromatography gave nearly identical results. A method using both C18 reverse phase and silica HPLC to purify radiolabeled sterols is also described.

The molecular heterogeneity of platelet-activating factor (PAF) produced by stimulated human neutrophilic polymorphonuclear leukocytes (PMN) was assessed by both normal and reverse phase high performance liquid chromatography (HPLC). As detected by rabbit platelet stimulation, at least 5 PAF molecules were separated by HPLC. Fast atom bombardment (FAB) mass spectrometry revealed one of these PAFs was acetyl glyceryl ether phosphorylcholine (AGEPC) with a C 16:0 alkyl chain in the sn -1 position. Although the structures of the remaining PAFs are unknown, two of the peaks of PAF activity had the same retention times on reverse phase HPLC as the C 15 - and C 18 -saturated alkyl chain AGEPC homologues. These studies indicate that the human PMN produces multiple molecular species of PAF.