Taccalonolide Binding to Tubulin Imparts Microtubule ... · laulimalide site on the exterior of the...

Therapeutics, Targets, and Chemical Biology

Taccalonolide Binding to Tubulin Imparts MicrotubuleStability and Potent In Vivo Activity

A.L. Risinger1,3, J. Li1,3, M.J. Bennett4, C.C. Rohena1, J. Peng1,3, D.C. Schriemer4, and S.L. Mooberry1,2,3

AbstractThe taccalonolides are highly acetylated steroids that stabilize cellular microtubules and overcome multiple

mechanisms of taxane resistance. Recently, two potent taccalonolides, AF and AJ, were identified that bind totubulin directly and enhance microtubule polymerization. Extensive studies were conducted to characterizethese new taccalonolides. AF and AJ caused aberrant mitotic spindles and bundling of interphase microtubulesthat differed from the effects of either paclitaxel or laulimalide. AJ also distinctly affected microtubulepolymerization in that it enhanced the rate and extent of polymerization in the absence of any noticeableeffect onmicrotubule nucleation. In addition, the resultingmicrotubules were found to be profoundly cold stable.These data, along with studies showing synergistic antiproliferative effects between AJ and either paclitaxel orlaulimalide, suggest a distinct binding site. Direct binding studies demonstrated that AJ could not be displacedfrom microtubules by paclitaxel, laulimalide, or denaturing conditions, suggesting irreversible binding of AJ tomicrotubules. Mass spectrometry confirmed a covalent interaction of AJ with a peptide of b-tubulin containingthe cyclostreptin-binding sites. Importantly, AJ imparts strong inter-protofilament stability in amanner differentfrom other microtubule stabilizers that covalently bind to tubulin, consistent with the distinct effects of thetaccalonolides as compared with other stabilizers. AFwas found to be a potent and effective antitumor agent thatcaused tumor regression in the MDA-MB-231 breast cancer xenograft model. The antitumor efficacy of sometaccalonolides, which stabilize microtubules in a manner different from other microtubule stabilizers, providesthe impetus to explore the therapeutic potential of this site. Cancer Res; 73(22); 6780–92. �2013 AACR.

IntroductionMicrotubule stabilizers, including paclitaxel, docetaxel, and

ixabepilone, are highly effective in the treatment of manysolid tumors, however, innate and acquired drug resistancelimits their efficacy (1–3). The search for new microtubulestabilizers that can overcome these limitations has beenintense. The taccalonolides are a novel class of microtubulestabilizers isolated from tropical plants of the genus Tacca.Many of the cellular effects of the taccalonolides are similar toall the other microtubule stabilizers in that they increase thedensity of cellular microtubules and interrupt mitotic progres-sion, leading to apoptosis (4). The most abundant naturallyoccurring taccalonolide, A (Fig. 1A), is substantially less potentthan the other microtubule stabilizers in vitro with an IC50

value for inhibition of proliferation of 5 mmol/L (5), however, it

exhibits antitumor effects in animal models (6, 7) and is able toovercome multiple mechanisms of drug resistance includingmutations in the taxane-binding site (4), the expression ofMRP7 (7), bIII-tubulin (7), and P-glycoprotein (Pgp) both invitro and in vivo (4, 7). Other cellular effects of taccalonolide Athat differentiate it from paclitaxel include its high cellularpersistence and the ability to initiate microtubule bundling atthe IC50 for inhibition of proliferation, whereas paclitaxelrequires concentrations 20 times higher than its IC50 (8).Recently, potent new taccalonolides, including taccalonolidesAF and AJ (AF, AJ) that have IC50 values for inhibition ofproliferation of 23 and 4 nmol/L, respectively, in HeLa cells,were identified (5). The potency of these taccalonolides iscomparable with paclitaxel and laulimalide, each of which hasIC50 values of 1 to 3 nmol/L in HeLa cells. AF and AJ causemicrotubule bundling in cells and robustly stimulate thepolymerization of purified tubulin, indicating for the first timea direct interaction of taccalonolides with tubulin (5).

Two major, nonoverlapping microtubule stabilizer-bindingsites on microtubules have been identified: the taxane site inthe interior lumen of the microtubule and the peloruside A/laulimalide site on the exterior of the microtubule (9, 10). Acombination of methods mapped paclitaxel binding to apocket on b-tubulin in the microtubule lumen with contactsat R282, H227, and V23 (10–13). Detailed binding and syner-gism studies demonstrated that overlapping, nonidenticaldrug-binding sites exist within the taxane-binding pocket,which allows chemically diverse agents, including paclitaxel,

Authors' Affiliations: Departments of 1Pharmacology, 2Medicine, and3Cancer Therapy and Research Center, University of TexasHealth ScienceCenter at San Antonio, San Antonio, Texas; and 4Department of Biochem-istry & Molecular Biology, University of Calgary, Alberta, Canada

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Corresponding Author: S.L. Mooberry, University of Texas Health Sci-ence Center at San Antonio, 7703 FloydCurl Drive, San Antonio, TX 78229.Phone: 210-567-4788; Fax: 210-567-4300; E-mail:[email protected]

doi: 10.1158/0008-5472.CAN-13-1346

�2013 American Association for Cancer Research.

CancerResearch

Cancer Res; 73(22) November 15, 20136780

on November 5, 2020. © 2013 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst September 18, 2013; DOI: 10.1158/0008-5472.CAN-13-1346

http://cancerres.aacrjournals.org/

docetaxel, the epothilones, discodermolide, and the dictyos-tatins, to elicit similar effects on microtubule stability (14–18).These drugs enhance microtubule stability by strengtheninglateral protofilament interactions, often by inducing the oth-erwise unstructured M-loop on b-tubulin into an orderedhelical structure (19–21). Increased microtubule stability canalso result from stabilization of lateral protofilament interac-tions on a-tubulin as seen for discodermolide, which does notaffect the M-loop (22). The multiple orientations and interac-tions of diverse compounds within the taxane pocket haverecently been expanded to include the covalent binding ofzampanolide/dactylolide to b-tubulin residues N226 andH227,which also causes M-loop stabilization (21, 23). In addition tothe taxane site, the second major stabilizer-binding site ofpeloruside A and laulimalide wasmapped to the exterior of themicrotubule in a site containing residues F294, R306, N337, andY340 on b-tubulin (9). Although the taxane and laulimalidesites are nonoverlapping, binding to either site initiates nearlyidentical effects on microtubule polymerization and stability(9, 24).

More recent evidence also demonstrates that some taxanesite–binding agents can also bind weakly to a low-affinitysite near the microtubule pore that they may use to gainaccess to the interior of the microtubule (25). The microbialmetabolite cyclostreptin, which covalently binds to T218 ofb-tubulin deep within this pore, was instrumental in thecharacterization of this site, which is referred to as the"gatekeeper" site (26). Cyclostreptin, which causes microtu-bule stabilization in both biochemical and cellular assays,can also bind covalently to N226 within the taxane pocketif it traverses the pore to reach the microtubule lumen(27–29).

In this article, the biochemical, cellular, and in vivo antitu-mor activities of AJ andAFwere evaluated.Wedetermined thatthey bind covalently to microtubules in a similar manner tocyclostreptin. Striking inter-protofilament stability that didnot involve M-loop stabilization was initiated by the covalentbinding of AJ to microtubules. The ability to covalently bind toand stabilize microtubules sheds light on many of the distinctproperties of the taccalonolides, including their cellular

Figure 1. Effects of diverse microtubule stabilizers on microtubule structures. A, chemical structure of AF and AJ, which differ from one another at the C15position. The structures of taccalonolides A and B are also included for reference. HeLa cells were treated with vehicle, (EtOH; B and G), 100 nmol/LAF (C and H), 30 nmol/L AJ (D and I), 15 nmol/L laulimalide (E and J), or 12.5 nmol/L paclitaxel (F and K) and microtubule structures in interphase (B–F), andmitotic (G–K) cells were visualized by indirect immunofluorescence.

Taccalonolides Covalently Bind to Tubulin

www.aacrjournals.org Cancer Res; 73(22) November 15, 2013 6781




persistence, ability to overcome multiple drug resistancemechanisms, and their potent in vivo effects.

Materials and MethodsCell culture

HeLa cells [American Type Culture Collection (ATCC)]were maintained in Basal Medium Eagle and MDA-MB-231cells (ATCC) were maintained in Improved Modified EagleMedium, both supplementedwith 10%FBS. Cellswerepassagedfor fewer than 6months after resuscitation from liquid nitrogen.

ImmunofluorescenceHeLa cells were treated for 18 hours with vehicle (EtOH),

paclitaxel, laulimalide, AF, or AJ at the minimal concentra-tions that caused maximum G2–M arrest. The cells were fixedand microtubules were visualized by indirect immunofluo-rescence techniques with a b-tubulin antibody as previouslydescribed (4).

Tubulin assaysTubulin polymerization was evaluated turbidimetrically

at 37�C in a SpectraMax plate reader with purified porcinebrain tubulin (Cytoskeleton Inc.). Tubulin (2 mg/mL) inGPEM buffer (80 mmol/L PIPES pH 6.8, 1 mmol/L MgCl2,and 1 mmol/L EGTA) containing 10% glycerol and 1 mmol/LGTP was incubated with vehicle (0.5% v/v) or drug in a finalvolume of 200 mL. Cold-induced tubulin depolymerizationwas evaluated by cooling the plates at 4�C or �20�C in acommercial refrigerator or freezer, respectively. No freezingof the reactions was observed. Effects on steady-state tubu-lin polymerization levels were analyzed by detection oftubulin by SDS-PAGE of the soluble or pellet fractions aftercentrifugation at 25�C.

Electron microscopySamples were collected from the tubulin assays described

above. Aliquots of the samples were taken 4, 11, 30, or 60minutes after warming to 37�C as well as after chilling andrecovery andmixed with equal volumes of a 4% gluteraldehydesolution. Microtubules were then mounted on 200 mesh cop-per grids, washedwith a 10% cytochromeC solution, negativelystained with 8% uranyl acetate, and visualized on a JEOL100CXtransmission electron microscope.

Cellular growth inhibition and synergismAntiproliferative effects were evaluated using the sulforho-

damine B assay as described previously (4). The synergism ofdrug combinations was evaluated with CompuSyn software togenerate isobolograms and combination indices (CI) asdescribed by Chou and Talalay (30).

Drug binding, displacement, and stabilityPurified porcine brain tubulin at a concentration of

20 mmol/L in GPEM buffer containing 10% glycerol and1 mmol/L GTP was incubated with 10 mmol/L drug(s) for1 hour at 37�C in a final volume of 100 mL. After incubation,the reactions were centrifuged at 20,000 � g for 30 minutesat 37�C to separate the bound and unbound drug. The pellet

was resuspended in 6 mol/L urea, 50 mmol/L Tris (pH 8),and 0.03% b-mercaptoethanol and heated to 80�C for 30minutes. Drugs were extracted from the supernatant andpellet fractions with ethyl acetate and detected using anAgilent 6224 Accurate–Mass time-of-flight (TOF) liquidchromatography/mass spectrometry (LC/MS). The quantityof drug(s) in each sample was determined from a standardcurve generated using identical experimental conditions.Each sample was evaluated in triplicate using MassHunterWorkstation software. Drug stability over time in pH 7 PBSwas determined by LC/MS.

Structural mass spectrometryHydrogen/deuterium exchange mass spectrometry (HDX-

MS) methods as described in detail previously (9) were used toexplore the conformational effects of AJ on microtubules.Briefly, 60 mmol/L bovine brain tubulin (Cytoskeleton) wasassembled into stabilized microtubules using 1 mmol/LGMPCPP as a nonhydrolyzable GTP analog and treated witheither 125 mmol/L AJ or docetaxel. Samples were then pro-cessed with a conventional HDX protocol and the resultingpeptides were analyzed by LC/MS on a Qstar Pulsar i using achilled LC apparatus as previously described (31). Deuteriumincorporation for each was quantified with Hydra (32), andsignificant drug-induced alterations of deuteration were deter-mined from quadruplicate datasets for each state. The levels ofaltered deuteration were color coded per peptide on segmentsof microtubule structures (PDB 2XRP; ref. 33). Structures wererendered in all figures using Pymol (http://pymol.sourceforge.net) and all residue numbering is based on bovine sequences inUniProt (a-tubulin: P81948,b-tubulin: Q6B856). The remainingdigests from the HDX analysis were depleted of deuterium atneutral pH, and re-analyzed by nano-liquid chromatography/tandem mass spectrometry (LC/MS-MS) on an Orbitrap Velos(Thermo Scientific). Peptides with significantly reduced inten-sity upon drug treatment triggered a reanalysis of the digest, asthis suggested covalent binding of the drug, and new peakswere sequenced with both collisionally induced dissociationand electron-transfer dissociation, to identify the host peptideand study the nature of the covalent interaction.

In vivo antitumor trialFemale athymic nude mice were maintained in an Asso-

ciation for Assessment and Accreditation of Laboratory Ani-mal Care-approved facility and provided food and waterad libitum. A total of 3� 106 MDA-MB-231 cells supplementedwith Matrigel were bilaterally injected subcutaneously intoeach flank. Mice were randomized into treatment groups(n ¼ 5 mice, 10 tumors) and drug treatments initiated whena median tumor volume of 60 mg was reached. The optimizedformulation of each drugwas comparedwith untreated tumor-bearing mice. AF and AJ were administered intraperitoneallyin 5% EtOH in PBS and paclitaxel in 5% Cremophor in PBSwith a total volume of 0.2 mL per injection. Tumor mass wascalculated using the formula: mass (mg) ¼ [length (mm) �width (mm)2]/2. One-way ANOVA with a Dunnett posttestwas applied to determine statistical significance betweencontrol and taccalonolide treatment groups.

Risinger et al.

Cancer Res; 73(22) November 15, 2013 Cancer Research6782




ResultsEffects of diverse microtubule stabilizers on interphaseand mitotic microtubulesThe effects of the new potent taccalonolides, AF and AJ

(Fig. 1A), on interphase and mitotic microtubules wereevaluated and compared with the effects initiated by pac-litaxel and laulimalide. For direct comparison, the minimumconcentration of each drug that caused maximal G2–Marrest was used. AF and AJ, but not paclitaxel or laulimalide,caused bundling of interphase microtubules at their respec-tive G2–M arrest concentrations (Fig. 1B–F). In addition, themitotic spindle asters formed by AF or AJ were morenumerous and compact as compared with those inducedby paclitaxel or laulimalide (Fig. 1G–K). These effects aresimilar to those previously observed with less potent tacca-lonolides (8).

Effects of diverse microtubule stabilizers on tubulinpolymerizationTo further evaluate differences between the taccalono-

lides and other microtubule stabilizers, the effects of themost potent taccalonolide, AJ, on the kinetics and extent oftubulin polymerization were compared with paclitaxel and

laulimalide over a range of concentrations (Fig. 2). In thevehicle controls, a 9-minute lag period was observed beforeinitiation of tubulin polymerization, after which the rate andextent of polymerization were monitored and normalized to1. At the lowest concentrations that affected tubulin poly-merization, paclitaxel (0.1 mmol/L) or laulimalide (0.25mmol/L) caused only modest increases in the total micro-tubule polymer over vehicle-treated controls, yet theydecreased the lag period from 9 to 4 minutes (Fig. 2A, B,and D). With higher concentrations of paclitaxel or lauli-malide, a dose-dependent decrease in lag period was mea-sured and polymerization occurred immediately after theaddition of 5 mmol/L of either drug with no measurable lagperiod. In contrast, a lag period of at least 5 minutes wasobserved with every concentration of AJ evaluated, includingthose where the rate and extent of total tubulin polymer-ization were greater than that observed with any othertreatment (Fig. 2C and D). The persistence of a lag periodfor polymerization was observed over the entire range ofAJ concentrations and is a property of all taccalonolideswith IC50 values for inhibition of cellular proliferation in thelow nmol/L range, including AF (5). This finding suggeststhat the taccalonolides are not as efficient at initiating

0.16

0.14

0.12

0.1

0.08

0.06

0.04

0.02

0

0.16

0.14

0.12

0.1

0.08

0.06

0.04

0.02

0

PTX

A C

B D

5

1

0.25

10

5

10.25

0 10 20

Time (min)

Lag (min)

Vehicle

0.1 μmol/L PTX

0.25 μmol/L PTX

1 μmol/L PTX

5 μmol/L PTX

10 μmol/L PTX

4

3

1

0

0

9

4

2

0

0

9

7

6

5

5

0.9

2.2

4.6

5.0

5.5

0.8

1.4

2.4

4.6

4.8

1.1

3.1

4.7

6.1

7.8

1.1

1.5

1.9

2.0

2.2

0.9

1.4

1.5

1.8

2.0

1.1

1.7

2.0

2.6

3.0

0.1 μmol/L LAU

0.25 μmol/L LAU

1 μmol/L LAU

5 μmol/L LAU

10 μmol/L LAU

1 μmol/L AJ

5 μmol/L AJ

10 μmol/L AJ

20 μmol/L AJ

30 μmol/L AJ

9 1.0 1.0

Max rate Total polymer

30

0 10 20

Time (min)

30

0 10 20

Time (min)

30

30

20

10

5

0.1

LAU

AJ

Mic

rotu

bu

le p

oly

me

riza

tio

n (

A340)

0.16

0.14

0.12

0.1

0.08

0.06

0.04

0.02

0

Mic

rotu

bu

le p

oly

me

riza

tio

n (

A340)

Mic

rotu

bu

le p

oly

me

riza

tio

n (

A340)

Figure 2. Tubulin polymerization.The kinetics of the polymerizationof 2 mg/mL purified tubulin withmicrotubule stabilizers wasmonitored turbidimetrically.A, paclitaxel (PTX) at 0.1, 0.25, 1,and 5 mmol/L. B, laulimalide at0.25, 1, 5, and10mmol/L.C, AJ at 5,10, 20, and 30 mmol/L. The lowest,dashed line in each graph indicatesmicrotubule polymerization withvehicle alone. D, quantitation oftubulin polymerization parameters.Lag indicates the time required toachieve exponential microtubulepolymerization. The maximumrelative rate and totalpolymerization have beennormalized to vehicle-treatedcontrols.






microtubule polymerization as compared with paclitaxel orlaulimalide.

The effects of each drug on the maximal rate of poly-merization and the total tubulin polymer formed as deter-mined from turbidity measurements were quantified. Dose-dependent increases in both the rate and extent of tubulinpolymerization were observed up to 1 mmol/L paclitaxel or5 mmol/L laulimalide where the total polymer and maxi-mum rate did not increase significantly with higher con-centrations (Fig. 2A and B). For both paclitaxel and lauli-malide, the maximum rate of tubulin polymerization wasapproximately five times greater than vehicle-treated con-trols with a plateau of 2-fold greater total polymer (Fig. 2D).Even though they bind to distinct sites on tubulin, thealmost identical effects of paclitaxel and laulimalide onmicrotubule polymerization are consistent with previousfindings (24).

AJ also caused a dose-dependent increase in the rate andextent of tubulin polymerization, with 10 mmol/L AJ resultingin a 4.7-fold increase in the rate of polymerization over vehicleand a doubling in total polymer formed (Fig. 2C and D).Although these values are almost identical to the effects of10 mmol/L paclitaxel or laulimalide, higher concentrations of20 or 30 mmol/L AJ caused further increases (30%–66%) in boththe rate of polymerization and total polymer formed as mea-sured turbidimetrically.

Because the tubulin polymerization in Fig. 2 was inferredby turbidity measurements, it was important to directlyvisualize the structures formed to determine whether theywere microtubules. The tubulin structures in the presence of10 mmol/L paclitaxel, laulimalide, or AJ, which showedsimilar effects on the rate and total extent of tubulinpolymerization (Fig. 2), were evaluated 4, 11, 30, and 60minutes after polymerization was initiated at 37�C and

compared with vehicle-treated controls. Electron micro-graphs show that paclitaxel and laulimalide caused a sub-stantial increase in microtubule polymer within 4 minutes at37�C, whereas the low number of microtubules formed in thepresence of AJ at this time point was not greater than thoseobserved for vehicle controls (Fig. 3). This is consistentwith the lag period observed during AJ-initiated microtubulepolymerization (Fig. 2C). The amount of microtubule poly-mer increased over time in all conditions in a manner thatwas consistent with turbidity measurements (Fig. 3). Nomajor differences between the microtubule structuresinduced by the three drugs were observed, even at highermagnifications (Supplementary Fig. S1). Microtubule struc-tures were also observed at higher concentrations of AJ,which were required to observe maximum polymerization.Interestingly, the microtubules formed in the presence 30mmol/L AJ did appear somewhat more rigid (SupplementaryFig. S2), but how this contributes to the increased turbidityobserved in the presence of this super-stoichiometric con-centration of AJ is not yet known (Fig. 2). Together, thesedata show that although AJ is not efficient at initiatingmicrotubule polymerization, it can enhance the rate andextent of tubulin polymerization in a manner that is marked-ly different from either paclitaxel or laulimalide.

Differential effects of microtubule stabilizers onmicrotubule stability

Microtubules are depolymerized by cold temperatures, asindicated by the decrease in turbidity following incubationof vehicle-treated microtubules at �20�C for 30 minutes(Fig. 4A, dashed line). Although paclitaxel- and laulimalide-initiated microtubules are somewhat cold stable, this strin-gent drop in temperature caused almost complete depo-lymerization of microtubules formed in the presence of

4 min 37°

Vehicle

PTX

LAU

AJ

11 min 37° 30 min 37° 60 min 37° 30 min 37°–20°

Figure 3. Electron microscopy ofmicrotubules. Electron micrographimages of microtubules formed inthe presence of 2 mg/mL tubulinwith 10 mmol/L paclitaxel,laulimalide, AJ, or vehicle 4,11, 30, or 60 minutes after tubulinreactions were warmed to 37�C,after chilling the reactions at�20�C for 30 minutes, or afterrepolymerization at 37�C for 30minutes after chilling. All imageswere acquired at a magnification of�2,000.

Risinger et al.





either 10 mmol/L paclitaxel or laulimalide (Fig. 4B and C).However, once the vehicle-, paclitaxel- or laulimalide-treatedtubulin reactions were rewarmed to 37�C, microtubulerepolymerization occurred rapidly with kinetics similar tothe initial polymerization. In contrast, microtubules formedin the presence of 10 mmol/L AJ were insensitive to cold-induced depolymerization and once the reaction wasrewarmed, a further increase in turbidity was observed (Fig.4D). The differential cold stability of microtubules in thepresence of paclitaxel, laulimalide, AJ, or vehicle was con-firmed by electron microscopy at low (Fig. 3) and high(Supplementary Fig. S3) magnification. These findings dem-onstrate a dramatic cold stability of AJ-induced microtu-bules and a propensity for increased microtubule polymer-ization upon rewarming. Electron micrographs show thatafter rewarming, the microtubule structures induced by AJappear longer and that the microtubule density is greaterthan that observed after the original polymerization (Fig 3,bottom row and Supplementary Fig. S4). Similar effects wereobserved for taccalonolide AF-induced microtubules (data

not shown), suggesting that resistance to cold-induceddepolymerization is a general property of the taccalonolides.

To further explore the cold sensitivity of drug-treatedmicro-tubules, less stringent conditions were used that caused onlypartial depolymerization of paclitaxel- or laulimalide-inducedmicrotubules. In addition, the extent of microtubule polymer-ization was monitored by separation of soluble tubulin fromthe microtubule pellet by centrifugation to more directlydetermine the distribution of polymerized and soluble tubulin.Consistent with the turbidimetric assays described above, at 10mmol/L of each drug initiated a similar degree (75%–81%) oftubulin polymerization after 30 minutes at 37�C (Fig. 4E, blackbars). After an additional 10minutes of incubation at 4�C, littlemicrotubule depolymerizationwas observed in the presence ofanymicrotubule stabilizer (Fig. 4E, gray bars), but, as expected,vehicle-treated microtubules were completely depolymerized(data not shown). After a longer 30 minutes of incubation at4�C, there was a 21% to 23% loss of paclitaxel- and laulimalide-induced polymer but no decrease in AJ-induced polymer (Fig.4E, white bars). Extended cold exposure did not cause

0.16

0.14

0.12

0.1

0.08

0.06

0.04

0.02

0

PTX

0 20 70

Time (min)

Vehicle

90 0 20 70

Time (min)

90 0 20 70

Time (min)

90 0 20 70

Time (min)

90

81 82

60

55

75

69

5252

8078

7376

LAU AJ

37°C

10’ 4°C

30’ 4°C

12 h 4°C

PTX

A B

E

C D

LAU AJ

Mic

rotu

bu

le p

oly

me

riza

tio

n (

A340) 0.16

0.14

0.12

0.1

0.08

0.06

0.04

0.02

0

Mic

rotu

bu

le p

oly

me

riza

tio

n (

A340)

% P

oly

me

rize

d t

ubu

lin

0.16

0.14

0.12

0.1

0.08

0.06

0.04

0.02

0

Mic

rotu

bu

le p

oly

me

riza

tio

n (

A340) 0.16

0.14

0.12

0.1

0.08

0.06

0.04

0.02

0

Mic

rotu

bu

le p

oly

me

riza

tio

n (

A340)

85

80

75

70

65

60

55

50

45

40

Figure 4. Cold stability ofmicrotubule polymers in the presence ofmicrotubule stabilizers. Polymerization of 2mg/mLpurified tubulin in the presence of vehicle(A), 10 mmol/L paclitaxel (PTX; B), 10 mmol/L laulimalide (LAU; C), or 10 mmol/L AJ (D) was monitored turbidimetrically for 30 minutes at 37�C. Sampleswere then cooled at �20�C for 30 minutes to depolymerize cold-sensitive microtubules (dashed line) and then the plate was returned to 37�C andturbiditymonitored for an additional 30minutes. E, the percentage of tubulin in the polymerized form in the presence of paclitaxel, laulimalide, or AJ after initialpolymerization (black bars) or the indicated times after transfer to 4�C was determined by centrifugation.






substantial further loss of polymer for any condition (Fig. 4E,hatched bars). Although these results confirm reports thatpaclitaxel and laulimalide impart some cold stability to micro-tubules, they further demonstrate that the cold stability initi-ated by AJ is more robust.

Because AJ-induced microtubules were markedly cold sta-ble, their ability to resist mechanical disruption was alsoevaluated. Under conditions of rigorous pipetting where vehi-cle-, paclitaxel-, and laulimalide-induced microtubules weresheared into numerous short microtubule polymers asobserved by electron microscopy, AJ-induced microtubuleswere not affected (data not shown). These studies demonstratethat the taccalonolides impart robust stability tomicrotubulesthat differentiate them from paclitaxel- and laulimalide-sta-bilized microtubules.

Synergism and displacement studiesSynergism studies have been useful in identifying com-

pounds with nonoverlapping binding sites (34–36). Theability of AF to cause synergistic antiproliferative effects incombination with paclitaxel or laulimalide was evaluated.Low antiproliferative concentrations of AF in combinationwith paclitaxel or laulimalide caused synergistic effects asdetermined by isobologram analysis and calculation of CIs(30). Combination indices of 0.65 to 0.84 were found with AFand laulimalide, indicating the two drugs act synergistically(Fig. 5A). A lesser degree of synergism was detected with thecombination of AF and paclitaxel, with CI values rangingfrom 0.84 to 0.95 (Fig. 5B). These finding suggested thepossibility that the taccalonolides might bind to a site thatis pharmacologically distinct from the two major stabilizer-

1

0.5

0

1

0.5

0

100

80

60

40

20

0

100

80

60

40

20

0

100

80

60

40

20

0

0 0.5

CI

0.650.84

0.77

AF

1 0 0.5

AF

PTX

A

C

D

B E

F

G

– Tubulin + Tubulin

Supernatant Pellet

% P

TX

in p

elle

t

% A

J in p

elle

t

(estim

ate

d)

% L

AU

in p

elle

t

– Tubulin + Tubulin

Supernatant Pellet

AJ

1

LA

U

PT

X

CI

93

99 99

50

62

41

LAU

alone

LAU +

PTX

AJ +

PTX

AJ +

LAU

LAU +

AJ

LAU 30’

after AJ

PTX

alone

AJ

alone

PTX +

DTX

PTX +

LAU

PTX +

AJ

PTX 30’

after AJ

94 93

93 90 93

93

0.840.85

0.95

Figure 5. Synergism and displacement studies of the taccalonolides and other stabilizers. Isobolograms depicting synergy of AF with laulimalide(LAU; *, 0.5 nmol/L laulimalide þ 10 nmol/L AF; &, 0.5 nmol/L laulimalide þ 20 nmol/L AF; D, 0.25 laulimalide þ 10 nmol/L AF; A) or AF with paclitaxel(PTX; *, 0.5 nmol/L paclitaxel þ 10 nmol/L AF; &, 0.5 nmol/L paclitaxel þ 20 nmol/L AF; D, 0.5 paclitaxel þ 30 nmol/L AF; B). The CIs calculated foreach point on the isobologram are listed. LC/MS traces of paclitaxel (C) or AJ (D) in the absence of microtubules or in the supernatant or microtubulepellet after incubation with purified tubulin for 30 minutes. Quantitation of the percentage of laulimalide (E), paclitaxel (F), or AJ (G) present in the microtubulepellet after incubation alone or in combination with other stabilizers. The percentage of AJ in the pellet was estimated on the basis of the amount of drug thatwas depleted from the supernatant fraction.

Risinger et al.





binding sites on tubulin and biochemical studies wereinitiated to test this possibility.Displacement studieswere conducted to determinewhether

the taccalonolides could compete with paclitaxel or laulima-lide binding to microtubules. High-resolution mass spectrom-etry was used to detect drugs in the microtubule-containingpellet or residual supernatant after incubation with purifiedtubulin and compared with the total amount of drug detectedin the absence of tubulin. As expected, when 10 mmol/L ofpaclitaxel was incubated with 10 mmol/L of a/b-tubulin,essentially no drug was detected in the supernatant and99% was extracted from the microtubule pellet (Fig. 5C andF). In contrast, AJ was undetectable in the microtubule pelletunder the same conditions, even though only 7% of the drugremained in the supernatant (Fig. 5D). We hypothesized thatthe inability to detect significant levels of AJ in either fractionwas a result of a tight interaction between AJ andmicrotubulesthat was not amenable to extraction with organic solvents.More stringent extractions of the microtubule pellet with avariety of detergents, proteases, and heat were unable toliberate AJ from the microtubule pellet, indicating the possi-bility of a covalent interaction of AJwithmicrotubules. Becauseof its inability to be extracted from the microtubule pellet,the percentage of bound AJ to microtubules was estimatedon the basis of depletion of AJ from the supernatant using astandard curve (Fig. 5G).To measure drug displacement, equimolar combinations of

laulimalide, paclitaxel, and AJ were incubated with tubulin andthe percentage of each drug present in the microtubule pelletwas calculated. When laulimalide was incubated with tubulinalone, 93% was bound to microtubules (Fig. 5E). As expected,the percentage of laulimalide bound to the pellet was notdecreased with paclitaxel coincubation as they bind to non-overlapping sites. AJ was unable to displace laulimalide wheth-er it was added simultaneously or before laulimalide (Fig. 5E)and conversely laulimalide was unable to displace AJ (Fig. 5G),indicating that these two drugs do not compete for binding tomicrotubules.Similarly, paclitaxel was unable to be displaced frommicro-

tubules by laulimalide, but docetaxel was able to inhibitpaclitaxel binding by 50%, consistent with their ability tocompetitively bind to the taxane site (Fig. 5F). When AJ wasadded simultaneously with paclitaxel in equimolar concentra-tions, the percentage of paclitaxel associated with the micro-tubule pellet decreased by 38%, indicating that AJ inhibitedpaclitaxel binding, although to a lesser extent than docetaxel(Fig. 5F). Given the inhibition of paclitaxel binding by AJ, it wassurprising to find that the converse was not true; the distri-bution of AJ was not affected by paclitaxel (Fig. 5G).On the basis of the assumption that the binding of AJ to

microtubules was irreversible and that this property maycontribute to its inability to be displaced by paclitaxel, theeffect of the order of drug addition on binding was tested.WhenAJwas added to tubulin 30minutes before paclitaxel, thepercentage of paclitaxel associated with themicrotubule pelletdecreased by 59% (Fig. 5F). Together, these results demon-strate that AJ inhibits paclitaxel binding, but not laulimalidebinding and that neither paclitaxel nor laulimalide affected AJ

binding. Although these data are consistent with some com-petition for microtubule binding between paclitaxel and AJ,the fact that the timing of AJ addition impacts the extent towhich it is able to displace paclitaxel combined with theinability of paclitaxel to displace AJ suggests that the twodrugs interact with tubulin in different ways.

Characterization of taccalonolide binding by massspectrometry

Mass spectrometry was used to aid in characterizing theinteraction between the taccalonolides with tubulin/microtu-bules and determine their impact on microtubule stability.Peptic fragments of tubulin were generated in the presence ofAJ or docetaxel and detected by LC/MS using ion extractionchromatography to identify peptides by their predicted mass.Strikingly, the peptide containing residues 212–230 of b-tubu-lin was observed in the presence of docetaxel (Fig. 6A, bluetrace) but not with AJ (Fig. 6A, red trace). Because our bindingexperiments suggested that AJ may covalently interact withtubulin, the generation of peptide(s) conjugated to AJ wasexplored. Indeed, a peptide with the predicted mass of b212–230 linked to AJ was detected in the presence of AJ (Fig. 6B, redtrace) but not docetaxel (Fig. 6B, blue trace), confirming that AJcovalently binds to this peptide on b-tubulin. In addition tob212–230, AJwas also found to bind directly to another, slightlyshorter peptic fragment, b213–230 (Supplementary Fig. S5).The peptides towhichAJ binds contain both theT218 andN226cyclostreptin-binding residues and are distinct but partiallyoverlapping with the taxane site (Fig. 6C and D), consistentwith our displacement data (Fig. 5F). Therefore, AJmay bind toone or both of the residues that interact with cyclostreptin oranother nearby residue. Unfortunately, the labeled residue(s)could not be localized in this study because tandem massspectrometry using collisionally induced dissociation onlyinduced the removal of AJ as a neutral loss fragment, whereaselectron-transfer dissociation did not generate an informativepeptide sequence, only charge reduction. More complex fol-low-up studies to map the residue(s) of b-tubulin that directlybind to AJ and model drug binding are ongoing.

In the absence of the ability to precisely map AJ binding totubulin, we usedHDX-MS to further probe the allosteric effectselicited by this covalent interaction as has been done for manyother microtubule-binding drugs (9, 19, 20, 22). A decreasedrate of hydrogen exchange demonstrates a lower accessibilityof residues to solvent, which is used to identify less flexible andmore stable regions of a protein. This method was used todemonstrate the effect of AJ or docetaxel on the stability ofmicrotubule peptides as compared with GMPCPP-stabilizedmicrotubules, which allows for differentiation of drug-specificeffects from those associated with general microtubule stabi-lization. Decreases in accessibility to hydrogen deuteriumexchange on tubulin residues in the presence of AJ are depictedin red on a b-tubulin monomer (Fig. 6D, right) and in thecontext of a larger microtubule fragment to show widerallosteric effects (Fig. 6C, right). Because we could not deter-mine whether the covalent binding of AJ to b213–230 directlyor indirectly altered the exchange properties of this peptide, itwas excluded from our analysis. The changes observed with AJ






binding were compared with those elicited by the binding ofdocetaxel to microtubules (Fig. 6C and D, left). Although someof the regions on tubulin that showed decreased exchangeupon drug binding are similar for the two drugs, there arenotable differences. The binding of docetaxel results in stabi-lization and therefore decreased hydrogen deuteriumexchange in the M-loop of b-tubulin as depicted in by itsrepresentation in red (Fig. 6C andD, left), which is also the casefor most other microtubule stabilizers (19–21). In contrast, theM-loop is not stabilized upon AJ binding, as shown by the lackof red in this area (Fig. 6C and D, right). The stabilization of theintradimer interface, involving helix H7 on b-tubulin, is alsoobserved with docetaxel and other stabilizers (37) but not AJ(Fig. 6D). Rather, AJ induced a markedly higher stabilization of

the lateral inter-protofilament contacts centering ona-tubulinin a manner that does not seem to involve the M-loop (Fig. 6C,right).

In vivo antitumor activity of taccalonolides AF and AJThe distinct binding characteristics of the taccalonolides led

us to evaluate the in vivo antitumor effects of AF and AJ in theMDA-MB-231 breast cancer xenograft model. Initial dosetolerance tests indicated that doses of 2 to 2.5 mg/kg AFadministered twice a week caused a dose-dependent, butgenerally recoverable weight loss. For these antitumor trials,the positive control, paclitaxel, was administered at 10 mg/kgon days 1, 3, 5, and 8 for a cumulative dose of 40 mg/kg (Fig. 7,filled squares). AF was found to have potent dose-dependent

12

10

8

6

4

2

0

6

5

4

3

2

1

0

β212-230

A C

B

D

M-

loop

M-

loop

H7-

loop

T218

N226

AJ-

binding

site

β212-230

+ AJ

β

α

Inte

nsity (

×10

3)

Inte

nsity (

×10

3)

Figure 6. Mapping the interaction of AJwithmicrotubules by structural mass spectrometric analysis. A, chromatogramof the b212–230 peptide [m/z 544.7995(þ4)] showing the peak intensity for docetaxel- (blue) or AJ (red)–treated microtubules. B, chromatogram of the AJ-bound form of the b212–230 peptide[m/z 713.87 (4þ)] showing the peak intensity for docetaxel- (blue) or AJ (red)–treated microtubules. C and D, lumen-to-exterior view of two parallelprotofilaments in surface rendering (C) and a zoomed-in view of b-tubulin in cartoon (same orientation as C; D). Structures on left represent docetaxel-treatedGMPCPP-microtubules and structures on right represent taccalonolide-treated GMPCPP-microtubules. Red highlights depict reductions in deuteriumlabeling induced by drug binding. Docetaxel is shown as purple spheres, and the peptide covalently bound to AJ (b213–230) is shown in yellow. An estimationof the binding region for AJ is marked with a black box in C. The M-loop involved in taxane-binding and interprotofilament stabilization is highlighted(black box in C, left blue box in D). Notice the M-loop is unaffected in AJ-treatedmicrotubules as indicated by green coloring. The H7 helix and loop, involvedin taxane-binding and intraprotofilament stabilization, is also highlighted (right blue box in D) and unaffected by AJ. The residues to which cyclostreptincovalently binds, T218 and N226, are marked for reference.

Risinger et al.





antitumor effects with inhibition of tumor growth almostidentical to that of paclitaxel when the lower, 2 mg/mL, dosewas administered on days 1, 4, and 8 for a cumulative dose of 6mg/kg (Fig. 7, open triangles). When a higher, 2.5 mg/kg dose,was administered on days 1 and 5 for a cumulative dose of 5mg/kg, complete inhibition of tumor growth and even mea-sureable reductions in tumor size were observed (Fig. 7, opencircles). This is significant because tumor shrinkage is rarelyobserved in this model. However, it is important to note thatthis dosing of AF also caused significant weight loss andrepresented the LD20 with one mouse succumbing to toxicityon day 15, which was 10 days after the last dose was admin-istered. These data indicate that AF has antitumor activity invivo with the potential to reduce tumor mass, albeit with anarrow therapeutic window, consistent with the effects of lesspotent taccalonolides (6, 7).Following initial dose tolerance testing, antitumor studies

were also performed with 0.5 mg/kg AJ dosed on days 1, 3, 5,and 8 for a cumulative dose of 2mg/kg. This intense schedule ofAJ showed no indication of antitumor effects (Fig. 7, opensquares) despite the fact that an average weight loss of greaterthan 10% was observed and 2 mice succumbed to toxicity on

days 11 and 12. The fact that no measureable antitumoractivities were observed at the LD40 demonstrates that AJ doesnot have a therapeutic window for antitumor activity. Follow-up studies to optimize the dose and schedule of AJ confirmedthis as lower dosing regimens that minimized toxicity alsoproduced no antitumor effects. Modest antitumor effects wereobserved with a dosing regimen of 0.85 mg/kg on days 1, 4, and8 albeit with unacceptable toxicity that led to the LD80.Although AJ was highly potent, causing notable weight lossat doses 20-fold lower than paclitaxel, the lack of antitumorefficacy at doses lower than the LD80 demonstrated theabsence of any therapeutic window.

The fact that a small but notable therapeutic window wasidentified for AF but not AJ prompted an investigation intodifferences in chemical stability that might relate to in vivometabolism. Analysis of the chemical breakdown of both AFand AJ showed very different chemical stability in aqueousconditions; AJ showed marked stability in pH 7 PBS of over 20hours, whereas AF had a t1/2 of 9 hours. Analysis of thedegradation products of AF indicated that AJ was the majorbreakdown product. Given the high toxicity observed for AJ invivo and the fact that AF is rapidly broken down into AJ inaqueous solution at physiologic pH, it is possible that thetoxicity associated with AF administration in vivo could be dueat least in part to its conversion into AJ. We hypothesize thatthe identification of taccalonolides that do not have thisliability of hydrolysis at C15 might facilitate the discovery ofa taccalonolide with a better therapeutic window.

DiscussionAll microtubule-stabilizing agents in clinical use bindwithin

the classical taxane pocket on microtubules. Laulimalide siteagents, which show synergism with taxane-binding drugs(24, 34, 36), are able to circumvent some clinically relevantforms of taxane resistance in vitro (38–40), but have so far failedto advance to clinical trials due in part to the lack of clearantitumor effects in murine models (38, 41). We show thatpaclitaxel and laulimalide have almost identical effects onmicrotubule polymerization and stability despite the fact thatthey bind to distinct sites onmicrotubules, which is consistentwith previous reports (9, 24). In contrast, AF and AJ have verydifferent properties, including a slow rate of microtubulenucleation as evidenced by a persistent lag period for poly-merization even at concentrations that eventually enhance therate, extent, and cold stability of microtubule polymerizationas compared with other stabilizers. These results suggest thatthe taccalonolides do not stimulate microtubule nucleation;but that once they are formed they are remarkably stable. Thisfinding is reminiscent of those obtained with cyclostreptin,which causes weak tubulin assembly reactions with a signif-icant lag period but enhanced cold stability (27, 42). The abilityof cyclostreptin to bind to b-tubulin residue T218 deep withinthe pore of the microtubule has prompted references tocyclostreptin as a "gatekeeper" for agents that require trans-port through this pore to gain access to the taxane site on theinterior of the microtubule (43). Similar to zampanolide,cyclostreptin can also covalently modify the N226 residue inthe taxane pocket on the inside of themicrotubule if it is able to

200

180

160

140

120

100

80

60

40

20

0

Control

1 3 5 7 9 11

Time (d)

13 15

10 mg/kg PTX

2 mg/kg AF

2.5 mg/kg AF

0.5 mg/kg AJ

Media

n tum

or

mass (

mg)

Figure 7. Antitumor effects of AF and AJ. Mice bilaterally implanted withMDA-MB-231 breast cancer cells were treated with 2.0 mg/kg of AF ondays 1, 4, and 8, 2.5 mg/kg AF on days 1 and 5, or 0.5 mg/k AJ on days 1,3, 5, and 8. Paclitaxel (PTX) was administered at a dose of 10 mg/kg ondays 1, 3, 5, and 8 as a positive control. Median tumor volumeswith SEM(n ¼ 10) are graphically represented. ��, P < 0.01.






traverse the pore without becoming covalently attached (26).Despite the robust literature detailing the biochemical effectson tubulin (26–28, 42), the antitumor effects of cyclostreptinhave not been well characterized (44).

The ability of AJ to bind covalently to residue(s) in the 213–230 peptide of b-tubulin is reminiscent of both cyclostreptinand zampanolide binding because this peptide encompassesboth the T218 pore site and the N226 residue in the taxanepocket. Although the identification of the exact residue(s) ofb-tubulin that are covalently modified by AJ was not deter-mined in these studies due to technical limitations, the allo-steric effects on microtubule stabilization afforded by AJbinding are striking (Fig. 6). It has been demonstrated thatthe covalent binding of zampanolide tomicrotubules increasesthe lateral interaction between protofilaments by promotinghelical structuring of the M-loop on b-tubulin (21). In contrast,the microtubule stabilization promoted by covalent binding ofAJ seems to cause profound interprotofilament stabilitythrough a distinct mechanism that does not involve changesto the M-loop. Interestingly, it has also been proposed that thebinding of cyclostreptin to T218 is not anticipated to affect M-loop stability or tubulin polymerization and that its ability tobind within the taxane pocket is what enhances microtubulestabilization (45). The distinct interprotofilament interactionsinitiated by AJ indicate, for the first time, that a drug cancovalently bind and stabilize microtubules independently ofM-loop stabilization, demonstrating that it interacts withtubulin in a manner distinct from all other stabilizers thathave been previously characterized. Preliminary data indicatethat all taccalonolides that are sufficiently potent to enhancethe polymerization of tubulin in biochemical extracts bindirreversibly to the resulting microtubule polymer, suggestingthat covalent binding of the taccalonolides to microtubules islikely a general property of this class ofmicrotubule stabilizers.Further biochemical analysis is warranted to fully characterizethe interaction of the taccalonolides with tubulin.

The distinct allosteric effects imparted on microtubule sta-bility by AJ shed insight into many unique properties of thiscompound class as compared with other microtubule stabili-zers. For example, it may facilitate the ability of the taccalono-lides to cause bundling of interphase microtubules at antipro-liferative concentrations, which is important in light of recentevidence that suggests that interphase effects play an importantrole in the antitumor properties of microtubule-targeted agents(46). The covalent binding of the taccalonolides tomicrotubulesmay also help to explain their potency in vivo. As observed forother taccalonolides (6, 7), AF and AJ weremuchmore potent invivo thanwould have been expected from their in vitro IC50s. Forexample, paclitaxel has an in vitro IC50 10-fold lower than AF, yetwas administered at 5-fold higher concentrations than AF toobserve similar antitumor effects. This represents an effective50-fold increase in in vivo potency for AF as compared withpaclitaxel, which is likely at least partially attributable to theability of taccalonolides to bind covalently tomicrotubules. Thiswould not be immediately reflected in an in vitro IC50 where thecells are constantly bathed in the drug, but becomes evident invivo where nonbound drug is actively cleared. The high cellularpersistence of the taccalonolides in clonogenic assays after drug

washout was the first indication that their cellular effects wereirreversible (8).

The in vivo potency of AF and AJ, presumably initiated bytheir covalent binding, also has the advantage of allowingthem to be administered in aqueous solvents without theneed for Cremophor or polysorbate-80, which can causehypersensitivity reactions. The covalent attachment of thetaccalonolides to microtubules also likely explains theirability to circumvent drug resistance mediated by expres-sion of the Pgp drug efflux pump, which limits the efficacy ofmany anticancer agents (4, 7). Indeed, in addition to thetaccalonolides, other microtubule stabilizers that covalentlybind microtubules, including cyclostreptin and zampano-lide, have shown the ability to circumvent Pgp-mediateddrug resistance (23, 26, 28). Despite the potency of AF and AJ,it is possible that the lack of antitumor efficacy for AJ and thenarrow therapeutic window of AF and other taccalonolidesmay also be due in part to irreversible binding. In addition toidentifying new taccalonolides with a larger therapeuticwindow, an alternative strategy is to conjugate a taccalo-nolide to a tumor-targeting antibody. This has been suc-cessful with the recent U.S. Food and Drug Administration(FDA) approval of T-DM1 (47). The prospects of using thisstrategy to specifically load a drug that irreversibly binds toits target into cancer cells may be highly effective.

Disclosure of Potential Conflicts of InterestA.L. Risinger, J. Li, and J. Peng have ownership interest (including patents) in

patent on the taccalonolides (inventor) issued to the UT system. S.L. Mooberry hascommercial research grant from Remeditex Ventures, honoraria from SpeakersBureau of CMEducation, ownership interest (including patents) in investorpatents, and is a consultant/advisory board member of Eisai Preclinical AdvisoryBoard. No potential conflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: A.L. Risinger, C.C. Rohena, J. Peng, S.L. MooberryDevelopment of methodology: A.L. Risinger, J. Li, J. Peng, D.C. SchriemerAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): A.L. Risinger, J. Li, M.J. Bennett, C.C. Rohena, D.C.SchriemerAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): A.L. Risinger, J. Li, M.J. Bennett, C.C. Rohena, J. Peng,D.C. Schriemer, S.L. MooberryWriting, review, and/or revision of the manuscript: A.L. Risinger, J. Li, M.J.Bennett, C.C. Rohena, J. Peng, D.C. Schriemer, S.L. MooberryStudy supervision: J. Peng, S.L. Mooberry

AcknowledgmentsThe authors thank Dr. Phil Crews for providing laulimalide (fijianolide B).

Support of the CTRCMass Spectrometry Shared and Macromolecular Structure[nuclear magnetic resonance (NMR)] Resources is gratefully acknowledged.

Grant SupportThis work was supported by National Cancer Institute (NCI) CA121138 (S.L.

Mooberry), Department of Defense, Congressionally Directed Medical ResearchProgram Postdoctoral Award BC087466 (A.L. Risinger), COSTAR ProgramNational Institute of Dental and Craniofacial Research DE 14318 (J. Li), NCIP30 CA054174 (S.L. Mooberry), the Clinical and Translational Science Awardsupport grant (S.L. Mooberry), Remeditex Ventures (S.L. Mooberry), AlbertaIngenuity-Health Solutions 201000620 (D.C. Schriemer), and Natural Sciencesand Engineering Research Council, Canadian Institutes of Health ResearchCollaborative Health Research Projects 385880 (D.C. Schriemer).

The costs of publication of this article were defrayed in part by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received May 10, 2013; revised July 30, 2013; accepted August 22, 2013;published OnlineFirst September 18, 2013.

Risinger et al.





References1. Fojo T, Menefee M. Mechanisms of multidrug resistance: the potential

role of microtubule-stabilizing agents. Ann Oncol 2007;18(Suppl 5):v3–8.

2. Kavallaris M. Microtubules and resistance to tubulin-binding agents.Nat Rev Cancer 2010;10:194–204.

3. Rowinsky EK, Donehower RC. Paclitaxel (Taxol). N Engl J Med1995;332:1004–14.

4. Tinley TL, Randall-Hlubek DA, Leal RM, Jackson EM, Cessac JW,Quada JC Jr, et al. Taccalonolides E and A: plant-derived steroids withmicrotubule-stabilizing activity. Cancer Res. 2003;63:3211–20.

5. Li J, Risinger AL, Peng J, Chen Z, Hu L, Mooberry SL. Potent tacca-lonolides, AF and AJ, inform significant structure–activity relationshipsand tubulin as the binding site of these microtubule stabilizers. J AmChem Soc 2011;133:19064–7.

6. Peng J, Risinger AL, Fest GA, Jackson EM, Helms G, Polin LA, et al.Identification and biological activities of new taccalonolide microtu-bule stabilizers. J Med Chem 2011;54:6117–24.

7. Risinger AL, Jackson EM, Polin LA, Helms GL, LeBoeuf DA, Joe PA,et al. The taccalonolides: microtubule stabilizers that circumventclinically relevant taxane resistance mechanisms. Cancer Res 2008;68:8881–8.

8. Risinger AL, Mooberry SL. Cellular studies reveal mechanistic differ-ences between taccalonolide A and paclitaxel. Cell Cycle 2011;10:2162–71.

9. BennettMJ,Barakat K,Huzil JT, Tuszynski J, SchriemerDC.Discoveryand characterization of the laulimalide-microtubule binding mode bymass shift perturbation mapping. Chem Biol 2010;17:725–34.

10. Nogales E, Wolf SG, Khan IA, Ludue~na RF, Downing KH. Structure oftubulin at 6.5 A and location of the taxol-binding site. Nature1995;375:424–7.

11. Rao S, He L, Chakravarty S, Ojima I, Orr GA, Horwitz SB. Character-ization of the Taxol binding site on themicrotubule. Identification of Arg(282) in beta-tubulin as the site of photoincorporation of a 7-benzo-phenone analogue of Taxol. J Biol Chem 1999;274:37990–4.

12. Rao S, KraussNE, Heerding JM, Swindell CS, Ringel I, Orr GA, et al. 30-(p-azidobenzamido)taxol photolabels theN-terminal 31 amino acids ofbeta-tubulin. J Biol Chem 1994;269:3132–4.

13. Rao S, Orr GA, Chaudhary AG, Kingston DG, Horwitz SB. Character-ization of the taxol binding site on themicrotubule. 2-(m-Azidobenzoyl)taxol photolabels a peptide (amino acids 217–231) of beta-tubulin. JBiol Chem 1995;270:20235–8.

14. Honore S, Kamath K, Braguer D, Horwitz SB, Wilson L, Briand C, et al.Synergistic suppression of microtubule dynamics by discodermolideand paclitaxel in non–small cell lung carcinoma cells. Cancer Res2004;64:4957–64.

15. Martello LA, McDaid HM, Regl DL, Yang CP, Meng D, Pettus TR, et al.Taxol and discodermolide represent a synergistic drug combination inhuman carcinoma cell lines. Clin Cancer Res 2000;6:1978–87.

16. Kowalski RJ, Giannakakou P, Gunasekera SP, Longley RE, Day BW,Hamel E, et al. The microtubule-stabilizing agent discodermolidecompetitively inhibits the binding of paclitaxel (Taxol) to tubulin poly-mers, enhances tubulin nucleation reactions more potently than pac-litaxel, and inhibits the growth of paclitaxel-resistant cells. Mol Phar-macol 1997;52:613–22.

17. Madiraju C, Edler MC, Hamel E, Raccor BS, Balachandran R, Zhu G,et al. Tubulin assembly, taxoid site binding, and cellular effects of themicrotubule-stabilizing agent dictyostatin. Biochemistry 2005;44:15053–63.

18. Kowalski RJ, Giannakakou P, Hamel E. Activities of the microtubule-stabilizing agents epothilones A and B with purified tubulin and in cellsresistant to paclitaxel (Taxol(R)). J Biol Chem 1997;272:2534–41.

19. Khrapunovich-Baine M, Menon V, Yang CP, Northcote PT, Miller JH,Angeletti RH, et al. Hallmarks of molecular action of microtubulestabilizing agents: effects of epothilone B, ixabepilone, pelorusideA, and laulimalide on microtubule conformation. J Biol Chem 2011;286:11765–78.

20. Xiao H, Verdier-Pinard P, Fernandez-Fuentes N, Burd B, Angeletti R,Fiser A, et al. Insights into the mechanism of microtubule stabilizationby Taxol. Proc Natl Acad Sci U S A 2006;103:10166–73.

21. Prota AE, Bargsten K, Zurwerra D, Field JJ, Díaz JF, Altmann KH, et al.Molecular mechanism of action of microtubule-stabilizing anticanceragents. Science 2013;339:587–90.

22. Khrapunovich-Baine M, Menon V, Verdier-Pinard P, Smith AB III,Angeletti RH, Fiser A, et al. Distinct pose of discodermolide in taxolbinding pocket drives a complementary mode of microtubule stabili-zation. Biochemistry 2009;48:11664–77.

23. Field JJ, Pera B, Calvo E, Canales A, Zurwerra D, Trigili C, et al.Zampanolide, a potent new microtubule-stabilizing agent, covalentlyreacts with the taxane luminal site in tubulin alpha,beta-heterodimersand microtubules. Chem Biol 2012;19:686–98.

24. Gapud EJ, Bai R, Ghosh AK, Hamel E. Laulimalide and paclitaxel: acomparison of their effects on tubulin assembly and their synergisticaction when present simultaneously. Mol Pharmacol 2004;66:113–21.

25. Barasoain I, García-Carril AM,MatesanzR,MaccariG, Trigili C,MoriM,et al. Probing the pore drug binding site of microtubules with fluores-cent taxanes: evidence of two binding poses. Chem Biol 2010;17:243–53.

26. Buey RM, Calvo E, Barasoain I, Pineda O, Edler MC, Matesanz R, et al.Cyclostreptin binds covalently to microtubule pores and lumenaltaxoid binding sites. Nat Chem Biol 2007;3:117–25.

27. Bai R, Vanderwal CD, Díaz JF, Hamel E. Interaction of a cyclostreptinanaloguewith themicrotubule taxoid site: the covalent reaction rapidlyfollows binding. J Nat Prod 2008;71:370–4.

28. Calvo E, Barasoain I, Matesanz R, Pera B, Camafeita E, PinedaO, et al.Cyclostreptin derivatives specifically target cellular tubulin and furthermap the paclitaxel site. Biochemistry 2012;51:329–41.

29. Sato B, Nakajima H, Hori Y, Hino M, Hashimoto S, Terano H. A newantimitotic substance, FR182877. II. The mechanism of action. JAntibiot 2000;53:204–6.

30. Chou TC, Talalay P. Quantitative analysis of dose-effect relationships:the combined effects of multiple drugs or enzyme inhibitors. AdvEnzyme Regul 1984;22:27–55.

31. Bennett MJ, Chik JK, Slysz GW, Luchko T, Tuszynski J, Sackett DL,et al. Structural mass spectrometry of the alpha beta-tubulin dimersupports a revised model of microtubule assembly. Biochemistry2009;48:4858–70.

32. Slysz GW, Baker CA, Bozsa BM, Dang A, Percy AJ, Bennett M, et al.Hydra: software for tailored processing of H/D exchange data fromMSor tandem MS analyses. BMC Bioinformatics 2009;10:162.

33. Fourniol FJ, Sindelar CV, Amigues B, Clare DK, Thomas G, PerderisetM, et al. Template-free 13-protofilament microtubule-MAP assemblyvisualized at 8 A resolution. J Cell Biol 2010;191:463–70.

34. Clark EA, Hills PM, Davidson BS, Wender PA, Mooberry SL. Laulima-lide and synthetic laulimalide analogues are synergistic with paclitaxeland 2-methoxyestradiol. Mol Pharm 2006;3:457–67.

35. Hamel E, Day BW, Miller JH, Jung MK, Northcote PT, Ghosh AK, et al.Synergistic effects of peloruside A and laulimalide with taxoid sitedrugs, but not with each other, on tubulin assembly. Mol Pharmacol2006;70:1555–64.

36. Wilmes A, Bargh K, Kelly C, Northcote PT, Miller JH. Peloruside Asynergizes with other microtubule stabilizing agents in cultured cancercell lines. Mol Pharm 2007;4:269–80.

37. Huzil JT, Chik JK, SlyszGW, FreedmanH, Tuszynski J, Taylor RE, et al.A unique mode of microtubule stabilization induced by peloruside A. JMol Biol 2008;378:1016–30.

38. Liu J, TowleMJ, ChengH, Saxton P, ReardonC,Wu J, et al. In vitro andin vivo anticancer activities of synthetic (�)-laulimalide, a marinenatural product microtubule stabilizing agent. Anticancer Res 2007;27:1509–18.

39. Mooberry SL, Tien G, Hernandez AH, Plubrukarn A, Davidson BS.Laulimalide and isolaulimalide, new paclitaxel-like microtubule-stabi-lizing agents. Cancer Res 1999;59:653–60.

40. Pryor DE, O'Brate A, Bilcer G, Díaz JF, Wang Y, Wang Y, et al. Themicrotubule stabilizing agent laulimalide does not bind in the taxoidsite, kills cells resistant to paclitaxel and epothilones, and may notrequire its epoxide moiety for activity. Biochemistry 2002;41:9109–15.

41. Johnson TA, Tenney K, Cichewicz RH, Morinaka BI, White KN, Ama-gata T, et al. Sponge-derived fijianolide polyketide class: further






evaluation of their structural and cytotoxicity properties. J Med Chem2007;50:3795–803.

42. Edler MC, Buey RM, Gussio R, Marcus AI, Vanderwal CD,Sorensen EJ, et al. Cyclostreptin (FR182877), an antitumor tubu-lin-polymerizing agent deficient in enhancing tubulin assemblydespite its high affinity for the taxoid site. Biochemistry 2005;44:11525–38.

43. Snyder JP. The microtubule-pore gatekeeper. Nat Chem Biol 2007;3:81–2.

44. Sato B, Muramatsu H, Miyauchi M, Hori Y, Takase S, Hino M, et al. Anew antimitotic substance, FR182877. I. Taxonomy, fermentation,

isolation, physico-chemical properties and biological activities. J Anti-biot 2000;53:123–30.

45. Prussia AJ, Yang Y, Geballe MT, Snyder JP. Cyclostreptin and micro-tubules: is a low-affinity binding site required? Chem Biochem2010;11:101–9.

46. Komlodi-Pasztor E, Sackett DL, Fojo AT. Inhibitors targeting mitosis:tales of howgreat drugs against a promising target were brought downby a flawed rationale. Clin Cancer Res 2012;18:51–63.

47. Boyraz B, Sendur MA, Aksoy S, Babacan T, Roach EC, KizilarslanogluMC, et al. Trastuzumab emtansine (T-DM1) for HER2-positive breastcancer. Curr Med Res Opin 2013;29:405–14.

Risinger et al.





2013;73:6780-6792. Published OnlineFirst September 18, 2013.Cancer Res A.L. Risinger, J. Li, M.J. Bennett, et al.

ActivityIn VivoPotent Taccalonolide Binding to Tubulin Imparts Microtubule Stability and

Updated version

10.1158/0008-5472.CAN-13-1346doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2013/12/23/0008-5472.CAN-13-1346.DC2 http://cancerres.aacrjournals.org/content/suppl/2013/12/23/0008-5472.CAN-13-1346.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/73/22/6780.full#ref-list-1

This article cites 47 articles, 18 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/73/22/6780.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/73/22/6780To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-13-1346

http://cancerres.aacrjournals.org/content/suppl/2013/12/23/0008-5472.CAN-13-1346.DC1

http://cancerres.aacrjournals.org/content/suppl/2013/12/23/0008-5472.CAN-13-1346.DC2

http://cancerres.aacrjournals.org/content/73/22/6780.full#ref-list-1

http://cancerres.aacrjournals.org/content/73/22/6780.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/73/22/6780


Taccalonolide Binding to Tubulin Imparts Microtubule ... · laulimalide site on the exterior of the...

Documents

Transcript of Taccalonolide Binding to Tubulin Imparts Microtubule ... · laulimalide site on the exterior of the...