Taxoids:New Weaponsagainst Cancer

The chemists who developedthe cancer-fighting agent taxol are creatinga family of similar compounds that may

one day help combat the disease

by K. C. Nicolaou, Rodney K. Guy and Pierre Potier

ust five years ago the chemicalknown as taxol made headlines asa breakthrough treatment for ovar-ian cancer. There was only one

problem—the drug was incredibly hardto come by. Researchers had to extractthe substance from the bark of the Pa-cific yew (Taxus brevifolia) in a processthat inevitably killed the tree. Even morefrustrating, yews grow slowly (a full-grown tree is around 25 feet tall), andeach plant yields little bark. A 100-year-old tree provides only a gram of the com-pound, about half the amount neededfor a single treatment. In addition, yewsthat produce taxol exist within the deli-cate old-growth forest of the PacificNorthwest, and harvesting the endan-gered trees would cause irreparable harmto this ecosystem. As the number ofPacific yews dwindled, environmental-ists argued to protect the few remainingtrees, while cancer patients and theirfamilies pleaded for more of the drug.

J

Today the headlines about taxol arequite different. In 1994 the U.S. Foodand Drug Administration approvedsemisynthetic taxol, made in the labo-ratory and available in unlimited quan-tities, for use in the treatment of variouscancers. Early this year a team of physi-cians based at Emory University an-nounced results from an extensive studyof the drug. Instead of lamenting itsscarcity, the researchers emphasized itsunexpected potency. According to the

94 SCIENTIFIC AMERICAN June 1996

findings, women suffering from ad-vanced ovarian cancer who took taxolin combination with another anticancermedication lived an average of 14months longer than patients who re-ceived other therapies. Taxol is now con-sidered one of the most promising treat-ments for breast and ovarian cancer.Other studies have demonstrated its ef-fectiveness against lung cancer and mela-noma. How did taxol, an agent initiallyknown mainly for its absence, becomerenowned for its powerful presence?

The story of taxol provides an impor-tant lesson about how scientists discov-er and develop new drugs. Chemists firstidentified the compound almost 30 yearsago. Since that time, biologists have de-termined how taxol works, and physi-cians have explored its healing proper-ties. Many researchers, including thethree of us, are pursuing the challengesof developing an entire family of taxol-like compounds—known as taxoids—that may eventually be easier to manu-facture and that may also afford moreand better therapeutic options than theparent molecule taxol.

Discovering Taxol, Again

odern pharmaceutical interest intaxol extends back to the 1960s,

but the medicinal properties of the yewtree have been known for centuries. Inone of his seven books, collectively enti-

M

tled On the Gallic Wars, published in 51B.C., Julius Caesar recorded the deathof the chieftain Catuvolcus, who com-mitted suicide by drinking tea madefrom yew bark. In the northwesternU.S., Native American tribes such asthe Quinault, Multnomah and Nez Per-ce utilized the Pacific yew's bark as adisinfectant, an abortifacient and atreatment for skin cancer. Over the past100 years, however, yew trees attractedlittle attention—at least until very re-cently. In the Pacific Northwest, forinstance, logging companies simplyburned yews after clear-cutting the tow-ering pine and fir trees that surroundedthe much smaller yews.

But in 1962 the botanist Arthur Bar-clay of the U.S. Department of Agricul-ture started the yew on a long and cir-cuitous journey back to being one of themost valuable trees in the Pacific North-western forest. At the time, the NationalCancer Institute (NCI) had requestedthat researchers sample natural sources-such as plants, bacteria and marinelife—in hopes of finding substances thatmight be useful as pharmaceuticals.

Taxoids: New Weapons against Cancer

CANCER THERAPY WITH TAXOL involves repeated intra-venous transfusions, each of which may last up to six hours.Here a woman at the Winship Cancer Center of Emory Univer-

sity receives taxol intravenously for ovarian cancer. Researcheshope derivatives of taxol, known as taxoids, will be easier to ad-minister with simple injections or even tablets.

Barclay collected bark from Pacific yewtrees in the Gifford Pinchot NationalForest, located in Washington State.

Barclay's yew samples eventually end-ed up at the Research Triangle Institutein North Carolina. There two chemists,Mansukh C. Wani and Monroe E. Wall,discovered that a mixture containing theyew's bark killed artificially preservedleukemia cells. By 1967 Wani and Wallhad isolated the active ingredient fromthis mixture: a previously unknownchemical that they christened taxol be-cause of its similarities to the family ofchemicals known as taxanes and be-cause the substance was found in plantsof the genus Taxus. (Although the name"taxol" is still widely used generically,the pharmaceutical company Bristol-Myers Squibb has registered "Taxol"as a trademark and wants the scientificcommunity to use "paclitaxel" instead.)

Over the next several years, taxol al-most faded back into the forest. The NCIdid not consider the compound particu-

larly promising. In early tests, otherdrugs worked just as well or better thantaxol for the treatment of cancer. Taxolwas also rare and difficult to obtain. ButWall, acting on a strong belief about itspotential, continued to champion thesubstance to the NCI. In 1977 the agen-cy agreed to investigate the matter fur-ther. But even after additional study,taxol still did not stand out from drugsalready in the anticancer pipeline.

Rigid Microtubules

oon after the second round of testsat the NCI, however, a pair of biolo-

gists at the Albert Einstein College ofMedicine in Bronx, N.Y., uncovered anew facet of taxol. In 1978 Susan B.Horwitz and one of her graduate stu-dents, Peter B. Schiff, demonstrated thattaxol killed cancer cells in a manner un-like any other drug known at the time.Over the next 10 years, Horwitz's groupprobed the details of how taxol func-

S

tions in the human body. In particular,the team found that taxol binds tostructures in the cell known as micro-tubules, which serve as part of the cell'sinternal skeleton, or cytoskeleton.

Normally, microtubules are flexibleconstructions that play a crucial role inthe dynamic process of cell division.For example, microtubules are the ma-jor constituents of the cellular appara-tus known as the mitotic spindle, whichhelps to separate the chromosomes dur-ing cell mitosis. When taxol attaches tomicrotubules, they become extremelystable and static, making cell divisionimpossible, thus killing the cells just asthey begin to divide. Cancer cells dividemore frequently than healthy cells, sothe drug primarily attacks tumors, inwhich runaway cell division occurs. Buiother rapidly dividing cells, such aswhite blood cells or hair cells, can alsobe affected; consequently, taxol is notwithout side effects when used to treatcancer. For example, taxol can suppress

Taxoids: New Weapons against Cancer SCIENTIFIC AMERICAN June 1996 95

patients' immune systems, deaden senso-ry nerves or cause nausea and hair loss.

The news of taxol's unusual methodof attacking cancerous cells excited theresearch community. Cancer tends tobecome resistant to treatment over time;because taxol killed tumor cells in anovel fashion, it might offer hope to pa-tients whose disease was not respond-ing to current therapy. By 1984 physi-cians at a number of hospitals, includingthe Dana-Farber Cancer Institute inBoston, the Johns Hopkins OncologyCenter in Baltimore, and MemorialSloan-Kettering Cancer Center in New

SYNTHETIC TAXOL can be producedfrom simple starting materials. Chemistsat Scripps Research Institute combinedfour small molecules and added addition-al fragments in a series of 39 steps to pro-duce taxol in a process known as totalsynthesis. By substituting different molec-ular fragments at any stage in the process,researchers can potentially make hun-dreds of derivatives of taxol, or taxoids.

York City, had begun the first stage ofhuman clinical trials to assess the safetyof taxol. In one of these surveys, Eric K.Rowinsky and his associates at JohnsHopkins reported unprecedented re-sults. In more than 30 percent of pa-tients whose tumors had previouslydefied conventional chemotherapy, tax-ol reduced the size of the growths. Onepatient was even cured. Other studiessoon echoed these findings, and taxolquickly slipped onto the pharmaceuti-cal fast track.

(Unfortunately, taxol did have poten-tially serious drawbacks: many peopleexperienced severe allergiclike reactionsto treatment, and one person died fromthis response. The cause of such compli-cations remains unclear, but doctorshave adjusted the dosage of the drugand how it is administered to minimizethe risk of adverse reactions. Neverthe-less, as with all chemotherapies, theside effects of taxol continue to troublephysicians and patients.)

When the promising stories of taxol'sbenefits surfaced, the NCI found itselffaced with two challenges. First, al-though taxol appeared to be excitinglyeffective, it was far from perfect. Butthis problem was typical for new drugs.Second, and more unusual, the supply

of taxol was running short. Conse-quently, between 1984 and 1989 physi-cians could conduct only a limited num-ber of extensive clinical trials. In 1989the NCI and Bristol-Myers Squibb estab-lished a contract that arranged for thecompany to produce the compound forthe NCI in exchange for gaining accessto the results of the NCI's clinical trials.Soon after, Bristol-Myers Squibb beganlarge-scale harvesting of the Pacific yewbut predicted that supplies would lastonly five years. Faced with this impend-ing shortage, scientists in many fields,including horticulture, forestry, cellularbiology and chemistry, scrambled tofind new ways to produce taxol.

Conquering a Molecular Mount Everest

hemists in particular exhibited aserious interest in taxol. To them,

molecules as large and complex astaxol, which contains 112 atoms, areaesthetically as well as scientifically ap-pealing. Its intricate architecture pre-sented a unique challenge to research-ers—such as the three of us—who spe-cialize in synthesizing natural products.We knew that the task of makingartificial taxol would be a lengthy one,requiring years of work. In the courseof the project, we would most likelycome to understand the compound's id-iosyncrasies—what parts of the struc-ture were particularly stable or fragileand how the molecule interacted withother chemicals. Such informationwould help us address broader ques-tions about the precise molecular func-tion of taxol in the body of a cancer pa-tient. Eventually, we hope scientists willunderstand the architecture of taxoland how the compound attaches to mi-crotubules so thoroughly that they willbe able to custom-design new drugs withthe benefits of taxol but with fewerharmful side effects.

C

Between 1983 and 1993, more than30 research groups struggled to synthe-size taxol or simpler, related compounds.But taxol proved to be an exceedinglydifficult molecule to construct; at times,it seemed unconquerable. Initially, manygroups explored the technique knownas semisynthesis in their attempts. Inthis process, chemists essentially start atabout the halfway point in the synthe-sis; rather than joining many smallfragments together to produce the finalproduct, they begin with a substance thatis very similar to the desired structure(and that is, ideally, cheap and available

96 SCIENTIFIC AMERICAN June 1996 Taxoids: New Weapons against Cancer

in large quantities). Then, by slightly al-tering this molecule, they obtain thecompound of interest in only a few steps.

In the early 1980s one of us (Potier)at the National Center of Scientific Re-search in France, along with Andrew E.Greene and his colleagues at the JosephFourier University in Grenoble, carriedout the first successful semisynthesis oftaxol. The investigators observed thattaxol could be dissected into two parts:the complex center of the molecule,known as the taxane core, and a sim-pler structure known as the side chain,which is connected to the core. WhilePotier and his group were screening theEuropean yew (T. baccata) for taxollikesubstances, they realized that the tax-ane core could be isolated from the nee-dles of this plant. They then figured outa straightforward way to attach theside chain. Because the team obtainedthe taxane core from the needles, whichgrow back after harvesting, the proce-dure offered hope that supplies of taxolmight not always be limited.

Such hope proved justified when, in1993, Bristol-Myers Squibb announcedthat it would no longer harvest Pacificyews. The company had adopted a pro-cess for the commercial production oftaxol that was initially developed inde-pendently by Iwao Ojima of the StateUniversity of New York at Stony Brookand by Robert A. Holton of FloridaState University. These two researchersalso employed a semisynthesis, but theirside chain and the method they used toattach it to the core differed from theFrench version.

Starting from Scratch

s Potier, Greene and others focused. their efforts on producing taxol by

semisynthesis, researchers elsewhere, in-cluding the two of us—Nicolaou andGuy—at Scripps Research Institute, con-tinued to work on a total synthesis. Byconstructing taxol with simple buildingblocks, we would be able to modify thecompound's structure at any position,thereby creating a variety of taxol deriv-atives, or taxoids, some of which mightprove less expensive and more potentthan taxol itself. In early 1994 twogroups almost simultaneously reporteda total synthesis of taxol. Nicolaou,Guy and their colleagues first publishedthe results of their work in the journalNature; Holton's group recounted itssuccess in the journal of the AmericanChemical Society.

A

Any synthesis of taxol must take intoaccount the inherent symmetry, or"handedness," of natural products.Structures with this property—like ourtwo hands—exist as mirror images ofeach other; we refer to each mirror im-age as an enantiomer. But often only oneenantiomer can produce an effect in thehuman body. Not surprisingly then, sci-entists believe only one form of taxolcan combat cancer. The proper enan-tiomer can be selected early on, by start-ing with chemicals of the appropriateconfiguration and then maintaining thisorientation at every step of the synthe-sis. This approach restricts the choice ofstarting material, however, and therebylimits the versatility of the synthesis. Toavoid such constraints and to reservethe option of constructing both enan-tiomers, our group at Scripps employedthe technique known as resolution,which allowed us to distinguish be-tween enantiomers. We were then freeto work with a mixture of enantiomersand to select the relevant configurationnear the end of the synthesis.

To further streamline the efficiency ofour method, we assembled taxol using

MAKING TAXOTERE from a complexsubstance similar to both taxol and Tax-otere, which is found in the needles of theEuropean yew (inset), offered a methodof producing a taxollike compound in afew steps. Scientists at the National Cen-ter of Scientific Research in France com-bined the taxane core with a small sidechain in a technique known as semisyn-thesis. Although semisynthesis provides aquick way to produce taxol and Tax-otere, it does not lend itself as easily tomaking a variety of taxoids.

what is called convergent synthesis. Us-ing this approach, one begins with sev-eral small pieces and joins them togeth-er to obtain the desired product; in con-trast, linear synthesis involves modifyinga single starting compound sequential-ly. The final structure can be altered in

Taxoids: New Weapons against Cancer SCIENTIFIC AMERICAN June 1996 97

a convergent synthesis fairly easily byintroducing different building blocks atany stage of the process; in a linear syn-thesis, the choice of building blocks ismuch more restricted. In this way, wecould make small, systematic changesin taxol's central core or side chain.

Chemists routinely make these kindsof changes in a compound's structure toevaluate how the molecular frameworkof the drug influences its potency. Forexample, suppose that for some hypo-thetical drug, replacing a hydroxylgroup (-OH) with a hydrogen atommakes the substance much less effec-tive. One would then assume that thehydroxyl group is directly involved inthe chemical's interaction with thebody. Drawing on this information, re-searchers can make new molecules byaltering or eliminating segments that ei-ther do not influence potency or causeharmful side effects. Or parts of thestructure that reduce potency can be al-tered or eliminated to improve the drug.

For example, Potier and his col-leagues produced the first notable tax-oid, which they named Taxotere. The

structure of Taxotere differs from taxolat two sites, but fortunately the taxoidalso combats the growth of tumors.Physicians in Japan and Europe com-monly use Taxotere as a therapy forbreast and ovarian cancers; in late 1995the FDA approved Taxotere for womenwith drug-resistant or metastatic breastcancer. Taxotere and taxol appear tohave subtle differences in their ability totreat certain cancers. Extensive use ofboth drugs in clinical trials should al-low scientists to define any advantagesthat one may have over the other.

Nicolaou, Guy and their colleagues atScripps have produced two importantclasses of taxol derivatives that mightone day yield functional pharmaceuti-cals. First, they simplified taxol's struc-ture and produced a taxoid that is some-what easier to make than taxol but, inpreliminary tests, can still kill certaintypes of cancer cells. Second, the grouphas developed a class of taxoids thatdiffers slightly at what seems to be theregion of taxol that attaches to micro-tubules. Scientists are continuing workto improve taxol's potency by tinkeringwith this binding site and thus makingtaxol more efficient at connecting to mi-crotubules and preventing cell division.

Improving Taxol

ll three of us have also been. attempting to solve one of

taxol's major pharmacologicaldrawbacks: its inability to dis-solve in water. This propertymakes administering taxol topatients complex and difficult.Currently doctors administer

A

PACIFIC YEW TREE provided the origi-nal source of the anticancer agent taxol.

taxol intravenously over several hours;the liquid medium used in this process,Cremophor El, has caused complicationsin some patients. A water-soluble com-pound would be much easier to handle.One new taxoid developed at Scrippsdissolves in water and could possibly beadministered with fewer side effects.

Other water-soluble taxoids producedin the laboratory allow us to examinein greater detail how taxol actually at-taches to microtubules. Because taxolitself is so resistant to solubility, investi-gators have typically analyzed its crys-talline, or solid, structure. Unfortunate-ly, the solid form of a molecule does notalways accurately reflect the way thecompound exists in the aqueous en-vironment of the cell. By observing howdissolved taxoids attach to microtubules,we can get a sense of which segments ofthe taxoid molecules are most likely tointeract with cells. Obviously, if wewant to manipulate taxol's structure tobetter its effectiveness, we need to knowwhere and how this binding occurs. Wemay be able to enhance taxol's ability tolatch onto microtubules and thus killcells. At the very least, we would notwant to alter the binding site in such away as to diminish taxol's potency.

Clearly, the story of taxol is not com-plete. But in the discovery of new drugs,one rarely cries "Eureka!" Rather theprocess takes years of detailed researchto determine how a drug works and howto improve its potency. In the case oftaxol, scientists have made significantprogress, not only figuring out how toproduce large quantities of the original-ly scarce drug but also finding new ap-plications for its use in cancer therapy.Now we have turned to yet anotherchallenge—tinkering with taxol's struc-ture until we find a less expensive, moreeffective medication.

The Authors

K. C. NICOLAOU, RODNEY K. GUYand PIERRE POTIER share an interest inthe molecular design and chemistry of natu-ral products. Nicolaou, who holds joint ap-pointments at Scripps Research Instituteand the University of California, San Diego,began investigating taxoids in 1992. Guystarted his doctoral studies at Scripps in1991; after graduation, he will move toSouthwestern Medical Center in Dallas.Potier began his research on taxol in 1980at the National Center of Scientific Researchin Gif-sur-Yvette, France, where he is cur-rently the director.

Further Reading

THE YEW TREE: A THOUSAND WHISPERS. Hal Hartzell, Jr. Hulogosi Communications, Eu-gene, Ore., 1991.

TAXOL AND TAXOTERE: DISCOVERY, CHEMISTRY, AND STRUCTURE-ACTIVITY RELATION-SHIPS. D. Guenard, F. Gueritte-Voegelein and P. Potier in Accounts of Chemical Research,Vol. 26, No. 4, pages 160-167; April 1993.

AN OVERVIEW OF EXPERIENCE WITH TAXOL (PACLITAXEL) IN THE U.S.A. R. C. Donehowerand E. K. Rowinsky in Cancer Treatment Reviews, Vol. 19, Supplement C, pages 63-78;October 1993.

CHEMISTRY AND BIOLOGY OF TAXOL. K. C. Nicolaou, W.-M. Dai and R. K. Guy in Ange-wandte Chemie, International Edition in English, Vol. 33, No. 1, pages 45-66; January17, 1994.

TOTAL SYNTHESIS OF TAXOL. K. C. Nicolaou, Z. Yang, J. J. Liu, H. Ueno, P. G. Nantermet,R. K. Guy, C. F. Claiborne, J. Renaud, E. A. Couladouros, K. Paulvannan and F. J.Sorensen in Nature, Vol. 367, pages 630-634; February 17, 1994.

98 SCIENTIFIC AMERICAN June 1996 Taxoids: New Weapons against Cancer