Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
ACS Chem Biol. Author manuscript; available in PMC 2010 August 21.
Published in final edited form as:
PMCID: PMC2853803

Botryllamides: Natural Product Inhibitors of ABCG2


ABCG2 is a membrane-localized, human transporter protein that has been demonstrated to reduce the intracellular accumulation of substrates through ATP-dependent efflux. Highly expressed in placental syncytiotrophoblasts, brain microvasculature and the gastrointestinal tract, ABCG2 has been shown to mediate normal tissue protection as well as limit oral bioavailability of substrate compounds. Development of ABCG2 inhibitors for clinical use may allow increased penetration of therapeutic agents into sanctuary sites and increased gastrointestinal absorption. Previously identified inhibitors have lacked potency, specificity or were toxic at concentrations needed to inhibit ABCG2; none are in clinical development. A previously developed high throughput assay measuring inhibition of ABCG2-mediated pheophorbide a (PhA) transport was applied to natural product extract libraries. Among the active samples were extracts from the marine ascidian Botryllus tyreus. Bioassay-guided fractionation resulted in purification of a series of botryllamides. Ten botryllamides were obtained, two of which (designated I and J) were novel. Activity against ABCG2 was confirmed by assessing the ability of the compounds to: inhibit ABCG2-mediated BODIPY-prazosin transport in ABCG2-transfected HEK293 cells; compete [125I]-iodoarylazidoprazosin (IAAP) labeling of ABCG2; stimulate ABCG2-associated ATPase activity; and reverse ABCG2-mediated resistance.


The ATP-binding cassette (ABC) superfamily of genes comprise the largest family of transporter genes and encode membrane proteins that transport diverse substrates across biomembranes against a concentration gradient using energy derived from ATP hydrolysis.1 Of the 48 human ABC transporter genes, three are usually associated with multidrug resistance in cancer – ABCB1 (MDR-1) encoding P-glycoprotein (P-gp), ABCC1 encoding the multidrug associated protein 1 (MRP1) and ABCG2 encoding the breast cancer resistance protein (BCRP or ABCG2).2 P-gp was the first ABC transporter described and has been shown to transport a diverse range of substrates including anticancer drugs, antibiotics and steroids.2 MRP1 was the second ABC transporter reported and was found to transport anticancer drugs as well as glucuronide and glutathione conjugates.2 ABCG2 is the most recent ABC transporter linked to multidrug resistance, counting chemotherapeutics, antibiotics, and HMG-CoA inhibitors among its substrates.3 Although its contribution to clinical drug resistance remains under investigation, ABCG2 is involved in modulating the oral availability of drugs and in forming normal protective barriers such as the maternal-fetal barrier and the blood-brain barrier.4,5 ABCG2 has also been reported to be highly expressed in cancer stem cells.6,7 Given these important roles, increased availability of modulators of ABCG2 activity would have significant research and clinical implications.

The search for ABCG2 inhibitors began with the observation that fumitremorgin C (FTC, produced by Aspergillus fumigatus) could reverse non-P-gp, non-MRP multidrug resistance in a drug-resistant cell line (later shown to be due to ABCG2 expression).8 However, clinical development of FTC was never undertaken because of its neurotoxicity in various animal models.9 Compounds reported to be inhibitors of ABCG2 activity include drugs originally identified as P-gp inhibitors, protein tyrosine kinase inhibitors, steroids, flavonoids, FTC derivatives, and other natural products. However, these lack potency or specificity or have toxic effects and none have been developed specifically as ABCG2 targeting agents (see references 2–7 for lists and discussions of specific inhibitors). A potent and specific ABCG2 inhibitor would be of significant value for clarification of the physiological function(s) of ABCG2 for pharmacological applications such as improving drug penetration into sanctuary sites as well as to confirm its purported relevance in multidrug resistance. Therefore, a high-throughput screening (HTS) assay, based on accumulation of pheophorbide a (PhA), was developed10 and used to identify novel inhibitors of ABCG2.11

To identify lead compounds for development as ABCG2 inhibitors, the ABCG2 HTS assay was applied to natural product extracts from the National Cancer Institute’s (NCI) natural products repository. Natural products and derivatives have been and continue to be rich sources of novel therapeutics12,13 and comprise the majority of current cancer chemotherapeutics.12,14 Many of the previously identified substrates and inhibitors of ABCG2 have also been natural products or derivatives, suggesting that this could be a fertile area for discovery of additional, unique inhibitors. After screening and bioassay-guided fractionation of active extracts, a series of botryllamides15,16 with variable activity against ABCG2 have been purified and characterized. Among these are two novel botryllamides. Confirmation and further characterization of the interaction of the botryllamides with ABCG2 included additional cell-based and biochemical assays. Specificity of the botryllamides for ABCG2 was assessed by determining their activity against MRP1 and P-gp.

Results and Discussion

To identify novel modulators of ABCG2, 89,229 natural product extracts from the NCI natural products repository were screened for ABCG2 inhibitory activity using a previously-described high-throughput assay.10 Extracts were evaluated for their ability to increase intracellular accumulation of the ABCG2-specific substrate pheophorbide a in ABCG2-overexpressing NCI-H460 MX20 cells. Thirty-five confirmed hits (i.e., reproducible activity of ≥ 50% of the activity of the positive control, FTC - see reference 10 for additional details) were obtained. Among these was a CH2Cl2-MeOH extract of the marine ascidian Botryllus tyreus that was collected along the coast of Papua New Guinea. Assay-guided fractionation of this extract by solvent partitioning and repeated chromatography on C18 stationary phase yielded known compounds, botryllamide A–H (1–8). Figure 1 shows the structures of these compounds and the related botryllamides described below. Botryllamides A–H were previously isolated and characterized as a result of chemical studies of several Botryllus species.15,16 The botryllamides have been reported to exhibit weak cytotoxicity to several tumor cell lines and their biosynthesis appears to involve the conjugation of two tyrosine subunits. In the current investigation they were identified by comparison of their spectral data with published values.15,16 In addition to the known botryllamides, two new compounds, designated botryllamide I (9) and J (10), were identified from the B. tyreus extract. See “Supporting Information” for complete NMR spectroscopic and physical data for compounds 9 and 10. In the course of assigning the structure of botryllamide J (10), it became apparent that the previously assigned structure of botryllamide H had to be revised to 11.

Figure 1
Structures of botryllamides

Botryllamide I (9) was obtained as a glassy solid after final C18 HPLC purification. Its molecular formula was established as C19H19NO4 by HRESIMS measurements (obsd [M-H] m/z 324.1236, calcd for C19H18NO4 324.1241). Compound 9 was clearly related to the other botryllamides as its 1H NMR spectrum showed characteristic resonances for two methoxy groups (δH 3.74 and 3.76) and two pairs of (2H) aromatic doublets that were indicative of two p-substituted phenol rings. 2D NMR established the presence of two olefins in 9 and HMBC correlations from H-3 (δH 6.05) and H-10 (δH 7.31) to C-1 (δC 163.9) confirmed they were connected through a carbonyl group. The 1H NMR spectrum acquired in CD3OH revealed the presence of an amide NH proton (δH 7.26) that was coupled (8.4 Hz) to H-10. HMBC correlations (δH 3.74 / δC 149.3 and δH 3.76 / δC 160.0) required that the methoxy groups were located at C-2 and C-15, respectively. The geometry of the C-10 / C-11 double bond was E based on the 14.6 Hz coupling between H-10 and H-11. The geometry of the C-2 / C-3 double bond in 9 could be inferred as E from the characteristic 13C NMR chemical shift of C-3 (δC 108.6). It was previously established with botryllamides A–D (1–4) that when Δ2,3 is Z C-3 resonates downfield (δC > 120), and when Δ2,3 is E C-3 is shifted upfield (δC < 110).15 Therefore, the structure of botryllamide I (9) could be assigned as the Δ2,3 geometrical isomer of botryllamide E (5). This was confirmed by the observation that botryllamide I (9) could be irreversibly converted to botryllamide E (5) by exposure to sunlight.

Given this observation, in order to avoid the possibility of light-induced isomerization of botryllamides, dry compounds and stock solutions were light-protected during storage. Similarly, incubations were performed in the dark or under subdued light conditions.

Botryllamide J (10) was isolated as a pale yellow solid that was soluble in DMSO, but not in MeOH. The molecular formula of 10 was established as C18H14N2O4 by HRESIMS ([M-H] m/z 321.0879) and this formula was isomeric with botryllamide H (8). The 1H NMR spectrum in DMSO-d6 was very similar to that obtained with 8 as it showed two doublet resonances (δH 6.79 and 7.35, each 2H) characteristic of a phenol, and three resonances at δH 7.08 (d, 8.5 Hz), 7.38 (t, 8.5 Hz), and 7.58 (d, 8.5 Hz) consistent with a 1, 2, 3-trisubstituted aromatic ring. A three proton spin system consisting of H-11 (δH 5.85), H-10 (δH 6.74), and an amide NH (δH 7.26) was apparent and a 9.8 Hz coupling between H-10 and H-11 revealed that the geometry of the C-10 / C-11 double bond in 10 was Z. HMBC correlations from H-3 (δH 7.45) to C-1 (δC 163.5) and C-4 (δC 124.1), from H-11 to C-13 (δC 130.1) and from the NH to C-2 (δC 156.9) helped establish the enamide amide and enol functionalities that connected the two aromatic rings. Similarity between the 1H and 13C resonances for the trisubstituted ring in 10 with those in botryllamide H (8) indicated that the substituents and substitution patterns were identical. All of the observed HMBC correlations about the ring were consistent with this assignment and the presence of an isonitrile functionality was supported by an IR absorption at 2120 cm−1 and a broadened 13C resonance at δC 162.3. The upfield shift of C-3 (δC 102.8) in 10 indicated that the Δ2,3 geometry was E. Since the chemical shift reported for C-3 in 8 was also upfield at 102.3, it appears that botryllamide H also has a Δ2,3 olefin with E geometry and that its structure should be revised to 11. Thus, botryllamide J (10) was assigned to be the Δ10,11 geometric isomer of the revised structure of botryllamide H (11).

Figure 2 shows activities in the screening assay for each botryllamide obtained from fractionation of the original extract. Results are expressed as percent of activity of 1 µM of the known ABCG2 inhibitor FTC which was set to 100%. Each purified botryllamide was resuspended in DMSO and serial dilutions prepared. Maximal activities and IC50 values are shown in Table 1. As observed in figure 2, all of the botryllamides except C and H had maximal activity at least 60% of that obtained with the positive control, FTC.

Figure 2
Activity of botryllamides in screening assay
Table 1
Effects of botryllamides on PhA accumulationa

To confirm the activity of these compounds as ABCG2 inhibitors, and to demonstrate that the activity was not linked to use of pheophorbide a as a substrate, an alternative flow-cytometry based substrate efflux assay, two biochemical assays, and a cell-based assay for ABCG2 inhibition were applied. The flow cytometry-based assay examined the ability of the botryllamides to inhibit efflux of the fluorescent ABCG2 substrate BODIPY-prazosin from ABCG2-transfected HEK293 cells.18 Cells were incubated with BODIPY-prazosin alone or in the presence of 10 µM botryllamide or FTC. As seen in Figure 3 (FTC panel), ABCG2-transfected cells incubated with BODIPY-prazosin alone demonstrate lower intracellular levels of the fluorescent compound (solid line) compared to cells incubated with BODIPY-prazosin in the presence of 10 µM of the known ABCG2 inhibitor FTC (dashed line). The higher intracellular fluorescence of BODIPY-prazosin found in cells incubated with prazosin in the presence of inhibitor compared to cells incubated with prazosin alone, denoted by difference between the solid histogram and the dashed histogram is indicative of inhibition of ABCG2-mediated BODIPY-prazosin transport. Intracellular fluorescence of BODIPY-prazosin was 3.4-fold higher in the presence of FTC versus in the absence of inhibitor. All of the botryllamides except H and J showed substantial activity (2.3- to 3.3-fold increased fluorescence). Based on their ability to cause accumulation of ABCG2 substrates, botryllamides A-G, I and J appear to be ABCG2 inhibitors. Botryllamide H was inactive in both the screening assay and the flow cytometry assay. Although C was inactive in the screening assay, it was active in the flow cytometry assay. Botryllamide J was found to be active in the screening assay but was marginally active in the BODIPY-prazosin assay. The combination of two different drug efflux assays thus identified 7 ABCG2-active botryllamides.

Figure 3
Accumulation of BODIPY-prazosin

Interaction with the ABCG2 protein by the botryllamides was also confirmed by two biochemical assays. First, the botryllamides were tested for their ability to inhibit the photolabeling of ABCG2 with [125I]IAAP (a prazosin analog) in membranes isolated from ABCG2-overexpressing MCF-7 FLV1000 cells. Previously, IAAP has been used as a photolabel to characterize the drug binding sites of ABC transporters such as ABCG2.23 For this assay, membrane protein isolated from MCF-7 FLV1000 cells was incubated (under low light conditions) with [125I]IAAP in the presence or absence of test or control compounds. Exposure to a UV light source followed, to crosslink the IAAP with the ABCG2 protein. The labeled ABCG2 was then immunoprecipitated and separated by electrophoresis, yielding the band in the control lane in Figure 4. When membrane protein is incubated with a compound known to interact with ABCG2 at the IAAP binding site, such as FTC, it prevents IAAP labeling of the ABCG2 protein, as seen in the FTC lane (Figure 4). As seen in Figure 4, at 20 µM, all of the botryllamides, including those found to be inactive in one or both of the substrate accumulation assays, significantly reduced [125I]IAAP incorporation into ABCG2. [125I]IAAP binding was reduced to 19–37% of control, comparable to that obtained with FTC (14% of control). Although reduced IAAP binding could be due to conformational changes induced by the inhibitor, this assay is generally understood to indicate that decreased binding is due to competition of the inhibitor for the drug-binding site. Thus, all of the botryllamides were found to interact with ABCG2 at the IAAP binding site, regardless of whether or not they were active in the screening or flow cytometry assays.

Figure 4
Photoaffinity labeling of ABCG2 with [125I]-IAAP

The botryllamides were then tested for their ability to stimulate or inhibit ATP hydrolysis of ABCG2 in isolated membranes. All of the botryllamides stimulated ABCG2-mediated ATPase activity in this assay, indicating that all the compounds directly interact with the protein (Figure 5). Many ABCG2 inhibitors, including FTC, inhibit ABCG2-associated ATPase activity while substrates often stimulate the ATPase activity.23 However, there is precedent for stimulation of ATP hydrolysis by ABCG2 inhibitors (for example, curcumin – see reference 24 for discussion). As a result, the stimulation of ATPase activity observed with the botryllamides does not preclude their tentative identification as ABCG2 inhibitors. The data obtained from the biochemical assays clearly indicate that the botryllamides directly interact with this transporter.

Figure 5
ATPase assay

In a further strategy, the botryllamides were assessed for their ability to sensitize drug-resistant ABCG2-overexpressing cells to killing by the cytotoxic ABCG2 substrate mitoxantrone. To overcome drug resistance mediated by expression of ABC transporters such as P-gp, MRP1 or ABCG2, nontoxic inhibitors of the proteins are often used to inhibit the activity of the transporters. NCI-H460 MX20 cells are resistant to the cytotoxic effects of mitoxantrone compared to ABCG2-negative parental NCI-H460 cells due to overexpression of ABCG2. As observed in Figure 6, when NCI-H460 MX20 cells were incubated with 10 µM mitoxantrone for 3 days, approximately 88% of the cells are alive compared to cells incubated with vehicle (DMSO). At this concentration of mitoxantrone, approximately 50% of the NCI-H460 parental cells survived compared to cells incubated with vehicle (data not shown). When the NCI-H460 MX20 cells were exposed to mitoxantrone in the presence of a nontoxic concentration of FTC (10 µM), sensitivity to mitoxantrone was restored, resulting in the survival of only 57% of the cells. Figure 6 shows the results of treating NCI-H460/MX20 (i.e. ABCG2-overexpressing) cells with mitoxantrone in the presence or absence of 10 µM botryllamide. Each of the botryllamides, except C, H, and J, had an effect comparable to that of FTC and reduced the number of cells suviving in the presence of mitoxantrone to roughly half (45–55%) of control. As seen in the open bars, none of the botryllamides alone caused appreciable killing of these cells. Although this was not designed to be a stability study, the results also suggest that the botryllamides remain stable in cell culture medium for at least the 3 day period of the assay. The mitoxantrone sensitization experiments provide results with the botryllamides and a third ABCG2 substrate. Given the wide range of chemical classes known to be ABCG2 substrates, it is important to validate new inhibitors using multiple diverse substrates, since variation in the drug binding site has been previously observed for ABCG2.17 PhA (a naturally-occurring chlorophyll metabolite), prazosin (a quinazoline derivative), and mitoxantrone (a synthetic anthracenedione) provide this breadth.

Figure 6
Mitoxantrone sensitization assay

Since some botryllamides have been previously reported to have cytotoxic activity and to evaluate their potential as ABCG2 substrates, parental NCI-H460 (ABCG2-) and NCI-H460/MX20 (ABCG2+) cells were exposed to up to 100 µM of each botryllamide for 7 d. At 10 µM, none of the botryllamides reduced cell survival by more than 27% (see part A of the figure in “Supporting Information”). At very high concentrations (100 µM – see part B of the figure in “Supporting Information”), botryllamides A, B, D, F, and G showed significant cytotoxicity. However, unlike with other cytotoxics that are substrates of transporters, overexpression of ABCG2 had no protective effect suggesting that, in addition to having relatively low cytotoxicity, the botryllamides are unlikely to be ABCG2 substrates.

Based on ABCG2-specific assays, it may be concluded that 5 of the active botryllamides are likely to be ABCG2 inhibitors. These compounds have the following approximate order of potency based on the PhA assay as confirmed by the BODIPY-prazosin assay and the mitoxantrone resistance reversal assay: G > B > D, F > A. Botryllamide H was found to be inactive in the PhA screen, BODIPY-prazosin and mitoxantrone resistance reversal assays. Botryllamide C was not particularly active in the PhA screen or the mitoxantrone resistance reversal assay, but was active in the BODIPY-prazosin assay. Botryllamide J was active in the PhA assay, but was only marginally active in the BODIPY-prazosin and mitoxantrone resistance reversal assays. The flow cytometry-based BODIPY-prazosin assay includes relatively short (≤ 2 h total) incubation times in contrast to the initial screening assay and mitoxantrone sensitization assay which include much longer incubations (18 h to 3 d). It is not unreasonable to assume that the differential activity of some of the botryllamides observed in these assays may be due to relative stability of the compounds in culture medium or inside the cells. The longer incubation times as well as the more directly relevant mitoxantrone sensitization results suggest that only those compounds with activity in these assays may have practical applications in a cellular or in vivo environment.

Each of the botryllamides was also tested for its ability to inhibit P-gp and MRP1, other ABC transporters known to confer drug resistance. Many inhibitors of P-gp are also able to inhibit ABCG2.21 Results are shown in Figure 7. P-gp or MRP1 expressing HEK293 cells were incubated with the fluorescent substrates rhodamine or calcein AM alone (solid line), respectively, or in the presence of 50 µM botryllamide (dashed line). Only botryllamides A and C significantly inhibited P-gp-mediated rhodamine transport (17.6- and 46.4-fold change in fluorescence, compared to 29.7-fold for the positive control, valspodar). By contrast, all of the botryllamides had detectable activity against MRP1-mediated calcein transport (2.0- to 4.5-fold increase in cellular fluorescence). However, only botryllamides A (4.5-fold) and C (3.7-fold) approached the activity of the positive control (5.1-fold). Thus, two of the botryllamides (A and C) showed significant activity against both P-gp and MRP1. Since it had no effect on ABCG2-mediated transport, botryllamide H may be relatively MRP1-selective, although it appears to have low potency. However, it must be noted that high concentrations (50 µM) of botryllamide were required for their inhibitory effects on P-gp or MRP1 and that when the experiment was attempted at lower concentrations (5 µM or less – data not shown) the effect was almost always lost, suggesting that the botryllamides are generally more specific for ABCG2 than for P-gp or MRP1. None of the compounds showed significant specificity for P-gp. Botryllamides E and G appear to be ABCG2-selective since they had minimal effects on P-gp and MRP1.

Figure 7
Assays of botryllamide effects on P-gp and MRP1

The botryllamides have several notable structural features, including variations in the bromination patterns, double bond geometry, and methylation. However, none of these features, except possibly the isonitrile group on botryllamides H and J, appear to consistently correlate with activity against ABCG2 and none are predictive of their ability to interact with the ABCG2 protein. In contrast to the lack of structural features associated with ABCG2 activity, an inhibitory effect on P-gp (only observed with botryllamides A and C) appears to require bromination and O-methylation at C-15 as well as trans configurations around both double bonds. For example, botryllamide G differs from A and C only in lacking the O-methylation while D differs from C only in configuration around the Δ2,3 double bond, and E lacks bromination. Botryllamides D, E, and G were inactive against P-gp. Although these results are only suggestive, it may be possible to isolate or synthesize additional botryllamide analogs with variable relative selectivity and potency to test these hypotheses.

The structures of botryllamides I and J have not been described previously, and thus they represent new members of this class of ascidian metabolites. Botryllamide I is a geometrical isomer of botryllamide E and the only structural difference is the Δ2,3 E double bond configuration in I versus a Δ2,3 Z geometry in E. Botryllamide J is structurally related to botryllamide H. Unlike the other botryllamides, they both lack methylation of the C-2 enol group and they contain a relatively rare isonitrile substituent. The Δ9,10 enamine double bond in botryllamide J was assigned as Z based on 1H-1H coupling constant and NOE evidence. This is the first compound in the botryllamide family with cis geometry at this position. The Δ2,3 double bond in botryllamide J was assigned as E based on well established 13C chemical shift evidence for C-3 (δ 102.8). Careful analysis of the published spectral data for botryllamide H16 revealed that the chemical shift of C-3 in H (δ 102.3) was characteristic of a Δ2,3 double bond with E geometry. The original botryllamide H publication did not provide a rationale for assigning the Δ2,3 geometry and it now appears that the initial structure of botryllamide H (8) needs to be revised to 11. With regard to biological activity, however, there is nothing noteworthy about either of these compounds in the assays described here.

In summary, novel ABCG2 inhibitors have been identified by screening natural product extracts from the NCI natural products repository. Among the ABCG2 inhibitors that have been previously identified (including tyrosine kinase inhibitors, estrogen agonists and antagonists, curcumins, and other classes of synthetic and natural compounds), the majority lack potency, selectivity or exhibit toxicity. None of the ABCG2 inhibitors identified to date have proven effective in clinical applications. Botryllamides form a unique structural class of diverse inhibitors of drug transport proteins. Synthesis and/or purification of additional members of this class of compounds may lead to a better understanding of structure-activity relationships and the potential for designing botryllamide derivatives with defined specificity patterns and potency. Given the aforementioned problems with ABCG2 inhibitors, including FTC,9 the availability of a new compound class should prove to be very useful for increasing understanding of ABCG2 function and potential as a therapeutic target. Botryllamides could also prove useful in normalizing oral bioavailability of ABCG2 substrate drugs which, in turn, could reduce intra- and inter-patient variability, thus improving clinical efficacy of ABCG2 substrate drugs. It should be noted that all of the active botryllamides were able to inhibit BODIPY-prazosin efflux and to sensitize ABCG2 overexpressing cells to mitoxantrone to the same extent and at the same concentrations as observed with FTC. Botryllamide G is the most potent and specific of these compounds as an ABCG2 inhibitor and will serve as a starting point for further analog synthesis. Until such analogs are developed, botryllamide G itself may prove useful in further studies of ABCG2 function and for probing the potential uses of botryllamides as modulators of ABCG2. Although the active botryllamides had relatively low cytotoxicity at concentrations at which they were active, analog development may allow for further reduction in toxicity. Similarly, the results suggest that it may be possible to discover or develop botryllamides or analogs with MRP1 and/or P-gp selectivity. Given that these compounds showed relatively low cytotoxicity, they may also provide a starting point for development of clinically useful derivatives.

Materials and Methods


PhA was obtained from Frontier Scientific (Logan, UT). FTC was from the Drug Synthesis and Chemistry Branch, Developmental Therapeutics Program (DTP), Division of Cancer Treatment and Diagnostics, NCI (Bethesda, MD). Natural product extracts were obtained from the Natural Products Branch of the DTP. Cell culture media were from Invitrogen (Carlsbad, CA) and fetal bovine serum (FBS) from Hyclone (Kansas City, MO).

Cell culture

NCI-H460 human lung non-small-cell carcinoma cells (NCI, Frederick, MD) were selected for overexpression of ABCG2 by maintenance in RPMI1640/10% FBS supplemented with 20 nM mitoxantrone (MX) and designated NCI-H460/MX20.18 ABCG2-transfected, ABCB1-transfected (i.e. P-gp-expressing), and ABCC1-transfected (i.e., MRP1-expressing) HEK293 cells were maintained as previously described.19

Screening assay for ABCG2 inhibitors

Accumulation of pheophorbide a (PhA), a fluorescent ABCG2 substrate,20,21 formed the basis of the assay for inhibitors of ABCG2 activity.10 Briefly, PhA (1 µM final) was added to NCI-H460/MX20 cells immediately followed by extracts (10 µg/ml), compound, or vehicle (DMSO/PBS) control and incubated (18 h). After removal of medium and washing with PBS containing Ca2+ and Mg2+, fluorescence intensity was read on a fluorescence plate reader in bottom read mode (395 nm excitation, 670 nm emission). Each plate had control wells containing vehicle (DMSO) or 10 µM (final concentration) of FTC. Data were normalized to FTC-treated cells on the same plate and reported as % of FTC fluorescence. Final DMSO concentrations were below 0.2% (v:v).

Flow cytometry

Purified compounds were confirmed for their ability to inhibit ABCG2-mediated transport in a flow cytometry assay using BODIPY-prazosin as a substrate.19 The ability of the botryllamides to inhibit P-gp-mediated rhodamine 123 efflux and MRP1-mediated calcein efflux was assessed as previously described.19,22 Briefly, transfected HEK293 cells expressing ABCG2, P-gp, or MRP1 were trypsinized and incubated in complete medium (phenol red–free Richter’s medium with 10% FCS and penicillin/streptomycin) containing 200 nM of BODIPY-prazosin, 0.5 µg/mL of rhodamine 123, or 200 nM of calcein AM, respectively, in the presence or absence of the desired concentration of inhibitor for 30 min at 37° C. Cells were then washed and incubated in substrate-free medium continuing with or without inhibitor for 1 h. Positive controls for inhibition of ABC transporters were 10 µM FTC for ABCG2, 3 µg/ml valspodar for P-gp and 50 µM MK- 571 for MRP1.

Intracellular fluorescence of BODIPY-prazosin, rhodamine 123 or calcein was detected with a FACSort flow cytometer equipped with a 488 nm argon laser and 530 nm bandpass filter. At least 10000 events were collected. Dead cells were eliminated based on propidium iodide staining.

Cytotoxicity assays

Botryllamides were assessed for their ability to affect NCI-H460 and NCI-H460/MX20 cell growth in the presence or absence of 10 µM mitoxantrone as described.10 After 2 days treatment, cell numbers were assessed using an XTT assay as described.10

Photoaffinity labeling of ABCG2 with [125I]-IAAP

ABCG2 expressed in MCF-7 FLV1000 cells, was photo-labeled with [125I]-IAAP as described.23 Briefly, crude membranes (1 mg protein/ml) of MCF-7 FLV1000 cells were incubated with 20 µM of the indicated compound for 10 min at room temperature in 50 mM Tris-HCl, pH 7.5. After addition of 3–6 nM [125I]-IAAP (2200 Ci/mmole) (PerkinElmer Life Sciences, Wellesley, MA) samples were incubated for 5 minutes under subdued light, then exposed to ultraviolet (UV, 365 nm) light for 10 min. Labeled ABCG2 was immunoprecipitated using BXP-21 antibody. The radioactivity incorporated into the ABCG2 band was quantified using the STORM 860 PhosphorImager system (Molecular Dynamics, Sunnyvale, CA) and ImageQuaNT software (Molecular Dynamics).

ABCG2-associated ATPase

Crude membrane protein (100 µg protein/ml) from High-five cells expressing ABCG2 was incubated at 37°C with compound (5 µM) in the presence and absence of BeFx (0.2 mM beryllium sulfate and 2.5 mM sodium fluoride) in ATPase assay buffer (50 mM KCl, 5 mM NaN3, 2 mM EGTA, 10 mM MgCl2, 1 mM DTT pH 6.8) for 10 min. The reaction was started by the addition of 5 mM ATP, incubated for 20 min at 37°C, and terminated by addition of SDS solution (0.1 ml of 5% SDS). Release of inorganic phosphate was quantified with a colorimetric reaction.23 The specific activity was recorded as BeFx-sensitive ATPase activity.

Data analysis/statistical significance

In each case quantitative data were compared to controls present in the same experiment. P < 0.05 was considered to be significant.

Supplementary Material



This project has been funded in whole or in part with federal funds from the National Cancer Institute, National Institutes of Health, under contract N01-CO-12400. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.

This Research was supported in part by the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research.


Supporting information available:

Collection and extraction of Botryllus tyreus, purification of botryllamides A–J (1–10), NMR and other spectroscopic data for botryllamides I (9) and J (10), and cytotoxicity/growth inhibitory activity of all 10 botryllamides are available as supporting information. This material is free via the internet.


1. Dean M, Annilo T. Evolution of the ATP-binding cassette (ABC) transporter superfamily in vertebrates. Annu. Rev. Genomics Hum. Genet. 2005;6:123–142. [PubMed]
2. Gottesman MM, Fojo T, Bates SE. Multidrug resistance in cancer: role of ATP-dependent transporters. Nat Rev Cancer. 2002;2:48–58. [PubMed]
3. Robey RW, Polgar O, Deeken J, To KW, Bates SE. ABCG2: determining its relevance in clinical drug resistance. Cancer Metastasis Rev. 2007;26:39–57. [PubMed]
4. Breedveld P, Beijnen JH, Schellens JHM. Use of P-glycoprotein and BCRP inhibitors to improve oral bioavailability and CNS penetration of anticancer drugs. Trends Pharmacol. Sci. 2006;27:17–24. [PubMed]
5. Hardwick LJA, Velamankanni S, van Veen HW. The emerging pharmacotherapeutic significance of the breast cancer resistance protein (ABCG2) Br. J. Pharmacol. 2007;151:163–174. [PMC free article] [PubMed]
6. Dean M, Fojo T, Bates S. Tumour stem cells and drug resistance. Nat. Rev. Cancer. 2005;5:275–284. [PubMed]
7. Lobo NA, Shimono Y, Qian D, Clarke MF. The biology of cancer stem cells. Annu. Rev. Cell Dev. Biol. 2007;23:675–699. [PubMed]
8. Rabindran SK, He H, Singh M, Brown E, Collins KI, Annable T, Greenberger LM. Reversal of a novel multidrug resistance mechanism in human colon carcinoma cells by fumitremorgin C. Cancer Res. 1998;58:5850–5858. [PubMed]
9. Allen JD, van Loevezijn A, Lakhai JM, van der Valk M, van Teilingen O, Reid G, Schellens JH, Koomen GJ, Schinkel AH. Potent and apecific inhibition of the breast cancer resistance protein multidrug transporter in vitro and in mouse intestine by a novel analogue of fumitremorgin C. Mol. Cancer Ther. 2002;1:417–425. [PubMed]
10. Henrich CJ, Bokesch HR, Dean M, Bates SE, Robey RW, Goncharova EI, Wilson JA, McMahon JB. A high-throughput cell-based assay for inhibitors of ABCG2 activity. J. Biomol. Screen. 2006;11:176–183. [PubMed]
11. Henrich CJ, Robey RW, Bokesch HR, Bates SE, Shukla S, Ambudkar SV, Dean M, McMahon JB. New inhibitors of ABCG2 identified by high-throughput screening. Mol. Cancer Ther. 2007;6:3271–3278. [PMC free article] [PubMed]
12. Newman DJ, Cragg GM. Natural products as sources of new drugs over the last 25 years. J. Nat. Prod. 2007;70:461–477. [PubMed]
13. Lam KS. New aspects of natural products in drug discovery. Trends Microbiol. 2007;15:279–289. [PubMed]
14. Tan G, Gyllenhaal C, Soejarto DD. Biodiversity as a source of anticancer drugs. Curr. Drug Targets. 2006;7:265–277. [PubMed]
15. McDonald LA, Swersey JC, Ireland CM, Carroll AR, Coll JC, Bowden BF, Fairchild CR, Cornell L. Botryllamides A-D, new brominated tyrosine derivatives from styelid ascidians of the genus Botryllus. Tetrahedron. 1995;51:5237–5244.
16. Rao MR, Faulkner DJ. Botryllamides E-H, four new tyrosine derivatives from the ascidian Botrylloides tyreum. J. Nat. Prod. 2004;67:1064–1066. [PubMed]
17. Ejendal KF, Hrycyna CA. Differential sensitivities of the human ATP-binding cassette transporters ABCG2 and P-glycoprotein to cyclosporine A. Mol. Pharmacol. 2005;67:902–911. [PubMed]
18. Robey RW, Honjo Y, van de Laar A, Miyake K, Regis JT, Litman T, Bates SE. A functional assay for detection of the mitoxantrone resistance protein MXR (ABCG2) Biochim. Biophys. Acta. 2001;1512:171–182. [PubMed]
19. Robey RW, Honjo Y, Morisaki K, Nadjem TA, Runge S, Risbood M, Poruchynsky MS, Bates SE. Mutations at amino-acid 482 in the ABCG2 gene affect substrate and antagonist specificity. Br. J. Cancer. 2003;89:1971–1978. [PMC free article] [PubMed]
20. Jonker JW, Buitelaar M, Wagenaar E, van der Valk MA, Scheffer GL, Scheper RJ, Plosch T, Kuipers F, Elferink RP, Rosing H, Beijnen JH, Schinkel AH. The breast cancer resistance protein protects against a major chlorophyll-derived dietary phototoxin and protoporphyria. Proc. Natl. Acad. Sci. USA. 2002;99:15649–15654. [PubMed]
21. Robey RW, Steadman K, Polgar O, Morisake K, Blayney M, Mistry P, Bates SE. Pheophorbide a is a specific probe for ABCG2 function and inhibition. Cancer Res. 2004;64:1242–1246. [PubMed]
22. Alvarez M, Robey R, Sandor V, Nishiyama K, Matsumoto Y, Paull K, Bates S, Fojo T. Using the National Cancer Institute anticancer drug screen to assess the effect of MRP expression on drug sensitivity profiles. Mol. Pharmacol. 1998;54:802–814. [PubMed]
23. Shukla S, Robey RW, Bates SE, Ambudkar SV. The calcium channel blockers, 1,4-dihydropyridines, are substrates of the multidrug resistance-linked ABC drug transporter, ABCG2. Biochemistry. 2006;45:8940–8951. [PubMed]
24. Chearwae W, Shukla S, Limtrakul P, Ambudkar SV. Modulation of the function of the multidrug resistance-linked ATP-binding cassette transporter ABCG2 by the cancer chemopreventive agent curcumin. Mol. Cancer. Ther. 2006;5:1995–2006. [PubMed]