Proc. Natl. Acad. Sci. USA
`Vol. 77, No. 3, pp. 1561-1565, March 1980
`Cell Biology
`
`Taxol stabilizes microtubules in mouse fibroblast cells
`(cell cycle/cytoskeleton/cell migration/antimitotic agents)
`PETER B. SCHIFF AND SUSAN BAND HORWITZ
`Departments of Cell Biology and Molecular Pharmacology, Albert Einstein College of Medicine, Bronx, New York 10461
`Communicated by Harry Eagle, December 18, 1979
`
`ABSTRACT
`Taxol, a potent inhibitor of human HeLa and
`mouse fibroblast cell replication, blocked cells in the G2 and M
`phase of the cell cycle and stabilized cytoplasmic microtubules.
`The c oplasmic microtubules of taxol-treated cells were visu-
`alized by transmission electron microscopy and indirect im-
`munofluorescence microscopy. More than 90% of the cells
`treated with 10 1AM taxol for 22 hr at 370C displayed bundles of
`microtubules that appeared to radiate from a common site (or
`sites), in addition to their cytoplasmic microtubules. Untreated
`cells that were kept in the cold (40C) for 16 hr lost their mi-
`crotubules, whereas cells that were pretreated with taxol for 22
`hr at 370C continued to display their microtubules and bundles
`of microtubules in the cold. Taxol inhibited the migration be-
`havior of fibroblast cells, but these cells did not lose their ability
`to produce mobile surface projections such as lamellipodia and
`filopodia.
`Taxol was isolated from the plant Taxus brevifolia and char-
`acterized as an experimental antitumor drug by Wani et al. (1).
`Our work has shown that taxol enhances in vitro the rate, extent,
`and nucleation phase of microtubule polymerization and sta-
`bilizes microtubules. Microtubules assembled in vitro in the
`presence of taxol are resistant to depolymerization by cold (40C)
`or 4 mM CaC12. The optimal effects of the drug on in vitro
`polymerization and stabilization of microtubules are observed
`near stoichiometric equivalence with tubulin dimers (2).
`We now report that taxol is a potent inhibitor of the repli-
`cation of HeLa and BALB/c fibroblast cells. HeLa cells treated
`with taxol accumulate in the G2 and M phase of the cell cycle.
`These cells contain microtubules plus bundles of microtubules
`all of which appear to have the structure of normal microtub-
`ules by transmission electron microscopy. Cytoplasmic mi-
`crotubules in BALB/c fibroblasts treated with a concentration
`of taxol that completely inhibits the replication of these cells
`are resistant to depolymerization by cold (40C) or by 10,uM
`steganacin. This is consistent with our observation that mi-
`crotubules treated with taxol in vitro become resistant to de-
`polymerization.
`It has been reported that colchicine and other drugs that
`inhibit the polymerization of microtubules also inhibit fibroblast
`and macrophage cell migration but do not alter the ability of
`these cells to produce mobile surface projections (3-5). We find
`that taxol, a promoter and stabilizer of microtubules, completely
`inhibits fibroblast migration. However, the taxol-treated fi-
`broblast cells are able to produce mobile lamellipodia and fi-
`lopodia.
`
`MATERIALS AND METHODS
`Materials. Taxol was obtained from the National Cancer
`Institute. Steganacin was kindly provided by the late S. Morris
`Kupehan. Both drugs were dissolved in dimethyl sulfoxide at
`a concentration of 10 mM and stored at -20°C. The final
`concentration of dimethyl sulfoxide in each experiment was
`
`The publication costs of this article were defrayed in part by page
`charge payment. This article must therefore be hereby marked "ad-
`vertisement" in accordance with 18 U. S. C. §1734 solely to indicate
`this fact.
`
`0.5% or less, a concentration that had no effect on control re-
`actions.
`Cells. HeLa (human) cells, strain S3, were grown in suspen-
`sion culture in Joklik's modified Eagle's minimal essential
`medium supplemented with 5% fetal calf serum and 1% glu-
`tamine. A primary cell line of male BALB/c mouse fibroblasts
`was provided by Susie Chen. These fibroblasts and Swiss 3T3
`mouse fibroblasts were grown as monolayers in Dulbecco's
`modified Eagle's medium supplemented with 10% fetal calf
`serum. Fibroblast cells used in experiments were no older than
`20 passages.
`Flow Microfluorometry. Flow microfluorometric analysis
`of the DNA content per cell by propidium iodide (50,ug ml-')
`staining in 0.1% sodium citrate has been described (6). Sus-
`pensions of stained cells were analyzed in a Cytofluorograph
`(model 4802A, Ortho Diagnostics Instruments, Westwood, MA),
`using an argon ion laser at 488 nm.
`Transmission Electron Microscopy. HeLa cells were sedi-
`mented (200 X g for 5 min at room temperature) after incu-
`bation with taxol at 37°C for 20 hr, resuspended in Joklik's
`modified Eagle's medium without serum, and sedimented
`again. The pellet of cells was fixed with 2% (wt/vol) glutaral-
`dehyde buffered with Joklik's modified Eagle's medium for
`1 hr at room temperature. Cells were postfixed in 1% osmium
`tetroxide, dehydrated, and embedded in Epon 812. Thin sec-
`tions of cells were stained with 4% uranyl acetate in 40% (vol/
`vol) ethanol, then with 0.1% lead citrate, and viewed with a
`Siemens Elmiskop 1A electron microscope.
`Immunofluorescence. Cytoplasmic microtubules were
`visualized by indirect immunofluorescence microscopy (7).
`Fibroblast cells were grown on glass coverslips in tissue culture
`dishes (Falcon) and were allowed to attach for 24 hr prior to the
`addition of drug. After the cells were incubated with drug for
`the desired time, the coverslips were washed once in phos-
`phate-buffered saline and fixed in 3.7% (wt/vol) formaldehyde
`for 8 min at room temperature. The coverslips were washed
`with phosphate-buffered saline, immersed in cold methanol
`(-20°C) for 4 min, immersed in cold acetone (-20°C) for 3
`min, and then allowed to air dry. The dry coverslips were
`covered with 10 Mil of rabbit antiserum to tubulin, provided by
`Marc Kirschner (8), diluted 1:30 in phosphate-buffered saline.
`After 50 min at 37°C the coverslips were washed extensively
`and covered with 10,ul of rhodamine-conjugated goat antise-
`rum to rabbit IgG (Cappel Laboratories, Cochranville, PA),
`diluted 1:10 in phosphate-buffered saline. After 50 min at 37°C
`they were again washed and mounted with Aqua-mount
`(Lerner Laboratories, Stamford, CT) on microscope slides. A
`Zeiss Photomicroscope III equipped with epifluorescent optics
`and a X63 oil immersion objective lens was used to view the
`cells. Photographs were taken on Kodak Tri-X 35-mm film and
`developed in Diafine.
`Migration Assay. A method for the visualization of
`phagokinetic paths of individual cultured cells moving on a gold
`particle-coated substrate has been described by Albrecht-
`Buehler (9).
`
`1561
`
`IMMUNOGEN 2190, pg. 1
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Proc. Natl. Acad. Sci. USA 77 (1980)
`sults were observed when HeLa cells were incubated with either
`0.25 or 10MM taxol for 20 hr. The microtubules had an average
`diameter of 250 A, the same value obtained for control cells.
`Indirect immunofluorescence microscopy, with antibodies
`against tubulin, was used to examine the effect of taxol on
`cytoplasmic microtubules in BALB/c fibroblast cells. Interphase
`fibroblast cells presented a characteristic display of cytoplasmic
`microtubules (Fig.3A). The microtubules in these cells are seen
`radiating outward from the perinuclear region to the plasma
`membrane. Cells that had been treated for 22 hr with either 1
`or 10 MtM taxol, concentrations of drug that completely inhibit
`fibroblast cell replication, displayed, in addition to their cyto-
`plasmic microtubules, bundles of microtubules that appear to
`radiate from a common site (or sites) (Fig. 3B). Small bundles
`of microtubules could be seen as early as 20 min
`after the ad-
`MM taxol to the cells. The taxol-treated cells also
`dition of10
`differed from normal cells in having a microtubule-free zone
`between the distal ends of their microtubules and the plasma
`membrane (Fig. 3B). More than 90% (132 out of 145) of the
`drug-treated cells viewed at 22 hr in a typical experiment had
`a characteristic morphology that included microtubule bundles
`and microtubule-free zones. The structure of the microtubules
`was verified by transmission electron microscopy and the
`morphology was observed to be the same as that described
`above for HeLa cells.
`Indirect immunofluorescence has been used to demonstrate
`that the display of cytoplasmic microtubules in mammalian
`cultured cells is sensitive to cold and antimitotic drugs such as
`colchicine(10, 11) that inhibit microtubule polymerization in
`vitro. Normally, cells that have been in the cold (40C) for 16
`hr lose their microtubules (Fig. 3C); however, cells incubated
`with 10 MLM taxol for 22 hr at37°C and then shifted to 40C for
`16 hr still displayed their microtubules and bundles of mi-
`crotubules (Fig. 3D). The same results were observed after a
`1-hr incubation with taxol. Steganacin, an antimitotic agent that
`is a competitive inhibitor of the binding of [3H]colchicine to
`purified tubulin, is a potent inhibitor of microtubule poly-
`merization invitro (12, 13). BALB/c fibroblasts that had been
`incubated with 10MM steganacin for 2 hr at 37°C did not dis-
`play their microtubules. However, fibroblasts that were pre-
`treated with 10 MM taxol for 22 hr at 370C and then treated
`with 10MM steganacin for 2 hr continued to display their mi-
`crotubules.
`The effect of taxol and steganacin on 3T3 fibroblast cell
`migration behavior has been examined by using the phagoki-
`netic track assay (9). When cells were viewed at low magnifi-
`cation with dark-field microscopy at 24 hr, both taxol and ste-
`ganacin were found to completely inhibit cell migration at
`concentrations of 1 or 10,MM, whereas control cultures pro-
`duced numerous phagokinetic tracks of individual cells re-
`moving and ingesting gold particles (Fig. 4 A, C, and E). Ad-
`dition of taxol to a 24-hr culture of migrating 3T3 fibroblast cells
`completely inhibited further cell migration. When the drug-
`treated cells were viewed at higher magnification by phase-
`contrast microscopy, it was evident that the cells were able to
`clear and ingest the gold particles from the area of attachment
`during the 24-hr experiment (Fig. 4 B, D, and F). The stega-
`nacin-treated cells generally appeared to extend fewer and
`smaller lamellipodia than the taxol-treated cells.
`In a separate experiment, photographic sequences were made
`of single 3T3 fibroblast cells incubated with 10MuM taxol at 370C
`(Fig. 5). After 15 min in the presence of taxol, ruffling lamel-
`lipodia appeared in numerous places around the cell perimeter.
`At 25 min the cell extended a large lamellipodium. At 58 and
`88 min the cell had retracted and, again, extended lamellipodlia,
`respectively. At 130 min the cell had retracted its lamellipodia
`and seemed to have lost all its polarity; ruffling of the mem-
`
`1562
`
`Cell Biology: Schiff and Horwitz
`
`Microscopy. The
`Dark-Field and Phase-Contrast
`phagokinetic tracks (9) of the 3T3 cells were observed in
`dark-field illumination with a Zeiss PhotomicroscopeII, using
`a X2.5 objective lens. A X40 objective was used for the phase-
`contrast micrographs. Photographs were exposed and developed
`as described above.
`
`RESULTS
`Flow microfluorometry was used to examine the effect of taxol
`on the distribution of DNA in HeLa cells as a function of time.
`Cells in exponential growth (3.2 X105 cells ml-') were incu-
`bated with 0.25 AM taxol. Untreated cell cultures had a dou-
`bling time of 19 hr. DNA content of individual cells in taxol-
`treated cultures was observed 3, 6,9,18, and 27 hr after addition
`of the drug (Fig. 1). At 18 and 27 hr, essentially all of the
`drug-treated cells had a tetraploid DNA content. Approxi-
`mately 70% of the taxol-treated cells were in mitosis at 20 hr,
`as determined by transmission electron microscopy; such cells
`contained condensed chromosomes and had lost their nuclear
`membranes.
`When these cells were examined at higher magnification,
`it was noted that, in addition to normal microtubules, they
`contained bundles of microtubules (Fig. 2 A and B). These re-
`
`Control
`
`I
`
`L~
`
`I I
`
`o I
`ITaxol
`1~I3 hr
`
`|
`
`"
`
`z
`
`7I
`.8I
`
`1
`
`Taxol
`6hr
`
`t t
`
`2C
`
`4C
`
`f t
`4C
`2C
`Cellular DNA content
`Flow microfluorometry of the DNA content of HeLa cells
`FIG. 1.
`in the absence and presence of taxol. Cells growing exponentially were
`diluted to 3.2 X 105 ml1' at the start of the experiment. Approximately
`6 X 104 cells were analyzed for DNA content at each time point. The
`histograms depict 100-channel analyses of cellular DNA content. The
`arrows indicate the modal positions of cells having diploid (2C) and
`tetraploid (4C) DNA contents. The full ordinate scale indicates a cell
`count of approximately 4 X 103 cells per channel. In control cultures,
`the proportion of cells with various DNA contents did not vary sig-
`nificantly during the time course of the experiment. The DNA dis-
`tributions shown were determined in cultures exposed to 0.25 uM
`taxol for the indicated times.
`
`IMMUNOGEN 2190, pg. 2
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Proc. Natl. Acad. Sci. USA 77 (1980)
`
`1563
`
`- vS P; >w as asA;,s K K f
`-
`
`d
`
`|s
`
`l
`4?,
`;A
`
`t Ditiz E r ' ; *- * id t ' ' ;, -Aim<'~~~~~~~~~~*
`
`*~4
`
`cells appear to be associated with the endoplasmic reticulum.
`It is not clear whether the association is a direct or indirect effect
`of taxol, but this observation does suggest that the endoplasmic
`reticulum may be involved in the organization of microtubules
`in cells.
`Indirect immunofluorescence microscopy, using antibodies
`against tubulin, was used to examine the effect of taxol on
`cytoplasmic microtubules and to determine if taxol could
`convert "labile" cytoplasmic microtubules into "stable" mi-
`crotubules like those found in cilia and flagella (16, 17). The
`"stable" microtubules of cilia and flagella are resistant to de-
`polymerization by low temperatures and drugs that normally
`block polymerization of microtubules in vitro. More than 90%
`of the drug-treated cells viewed at 22 hr had a characteristic
`morphology consisting of microtubules, microtubule bundles,
`and microtubule-free zones between the distal ends of their
`microtubules and their plasma membranes. These microtubules
`were resistant to depolymerization by cold (40C) and stega-
`nacin. The bundles appear to radiate from a common site (or
`
`Fic. 2.
`Electron micrographs of thin sections of HeLa cells treated with 10 /IM taxol for 20 hr. (A) Mitotic cell. MTB, microtubule bundle;
`Cr, chromosome; arrows point to individual microtubules in longitudinal profile (B) Interphase cell. Nu, nucleus; arrows point to microtubules'
`in cross section. Scale bars: 0.5 Mm.
`brane was observed all around the perimeter of this fibroblast
`cell. The cell extended a lamellipodium again 223 min after the
`addition of taxol. Mobile filopodia were also observed during
`the photographic sequence. Untreated cells generally had
`ruffling only at the leading edge of the cell, as has been reported
`by other investigators (14, 15).
`
`DISCUSSION
`Flow microfluorometry was used to demonstrate that taxol does
`not inhibit the first round of DNA synthesis but blocks cells in
`the G2 and M phases of the cell cycle. In addition to their in-
`dividual cytoplasmic microtubules, the taxol-treated cells
`contain bundles of microtubules. Although approximately 70%
`of the taxol-treated cells are in mitosis after exposure to the drug
`for 20 hr, they do not make a normal mitotic apparatus even
`though they lose their nuclear membranes and condense their
`DNA. Using transmission electron microscopy, we have shown
`that the structure of the microtubules in these cells seems to be
`normal. Some of the microtubule bundles in the taxol-treated
`
`Cell Biology: Schiff and Horwitz
`
`W
`
`et
`
`~~~~~~~~~~~~~~~I_
`
`Indirect immunofluorescence of BALB/c fibroblast cells, using antibodies against tubulin. (A) Control cell, (B) cell exposed to 10
`FIG. 3.
`MuM taxol for 22 hr, (C) control cell kept at 40C for 16 hr, (D) cell incubated with 10MM taxol for 22 hr, then shifted to 40C for 16 hr. Scale bars:
`20 kiln. Arrows indicate the edge of the plasma membrane in the plane of focus.
`
`IMMUNOGEN 2190, pg. 3
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`1564
`
`Cell Biology: Schiff and Horwitz
`
`Proc. Natl. Acad. Sci. USA 77 (1980)
`
`_1
`..AA
`
`'I~~~~~l
`
`Sasil;.
`
`....
`
`-- 4as A'
`
`..
`
`Migration behavior of control (A and B), taxol-treated (C and D), and steganacin-treated (E and F) Swiss 3T3 fibroblast cells. Cells
`FIG. 4.
`(1000-2000 per dish) were seeded into drug-containing medium in 35 X 10 mm tissue culture dishes. A, C, and E show the phagokinetic track
`patterns left by individual control, taxol-treated (10 uM), and steganacin-treated (10 /AM) 3T3 cells, respectively. Particle-free dark areas result
`from the cells ingesting or removing the gold particles. Illumination is dark field. B, D, and F are phase-contrast micrographs of control, taxol-
`treated, and steganacin-treated cells, respectively, from the same experiment. Both drug-treated and control cells were able to make particle-free
`areas and ingest the gold particles. However, the drug-treated cells did not make tracks. Scale bars: 200 ,m.
`sites) within the cell, which may represent the microtubule-
`organizing centers. These may be the same sites from which
`microtubules have been reported to originate in cells (8, 11).
`Although it is difficult to rule out a redistribution of microtu-
`bules in the cells, an intriguing alternative is that these bundles
`represent new initiations that have occurred at microtubule-
`organizing centers. Mouse fibroblasts are known to contain a
`primary cilium (18); therefore, some of the bundles observed
`in taxol-treated cells may be elongated or newly formed cilium.
`One explanation for the microtubule-free zone is that the cell
`preferentially initiates formation of new microtubules at the
`microtubule-organizing centers with the available tubulin di-
`mers, instead of elongating existing microtubules as the cell
`extends lamellipodia on the glass coverslip. In addition, the cell
`may not be able to depolymerize existing microtubules to
`provide additional dimers f6r elongation.
`Little is known about the biological machinery involved in
`cell migration. The microtubule cytoskeleton may play a role
`in cell migration by determining the polarity of a migrating cell
`(19). A model based on intrinsic microtubule behavior has been
`proposed to explain the orderly separation of chromosomes
`during mitosis (20). Colcemid, colchicine, and vinblastine, drugs
`that inhibit mitosis, also inhibit fibroblast and macrophage cell
`migration (3-5). However, these drug-treated cells are not ut-
`
`terly immobile. They are still able to produce mobile lamelli-
`podia and filopodia. Ruffling of the membrane becomes more
`evenly distributed around the cell perimeter of these drug-
`treated cells, instead of occurring mainly at the leading edge
`as is true of untreated cells (14, 15). The results of the time-lapse
`photographic sequences and the migration experiments dem-
`onstrate that taxol inhibits cell migration but does not affect cell
`motility. These experiments further suggest that migrating cells
`must be able both to polymerize and depolymerize their cyto-
`plasmic microtubules in order to differentiate between their
`front and back ends.
`The inhibition of HeLa and BALB/c fibroblast cell repli-
`cation and the inability of 3T3 fibroblasts to migrate in the
`presence of low concentrations of taxol could derive from the
`cell being unable to depolymerize its microtubule cytoskeleton.
`This may represent a mechanism of action for a chemothera-
`peutic agent and explain the observed antitumor activity of
`taxol. The drug should be a useful tool in studying the regulation
`of microtubule assembly and cellular functions, such as cell
`migration, that may be mediated by microtubules. The drug
`could also provide a method for the isolation of intact cyto-
`plasmic and spindle microtubules.
`We are indebted to Danuta Hornig and Dr. Carl Schildkraut for
`advice concerning flow microfluorometry, Jane Fant for assistance with
`
`IMMUNOGEN 2190, pg. 4
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Cell Biology: Schiff and Horwitz
`
`Proc. Natl. Acad. Sci. USA 77 (1980)
`
`1565
`
`FIG. 5.
`Selected frames from a time-lapse phase-contrast study of a taxol-treated Swiss 3T3 fibroblast cell. Cells were allowed to attach
`to the glass coverslip 24 hr before the addition of 10 jM taxol at 370C. After addition of the drug the coverslip was mounted on a slide and sealed
`with wax to form a chamber containing culture medium. The stage of the microscope and slide were kept at :370C with an airstream incubator.
`Cells were exposed to light only during film exposure. Untreated cells behaved normally in these conditions. The photographic sequence begins
`15 min after the addition of taxol to the cells. Scale bars: 40 jm.
`
`the transmission electron microscopy, Dr. Robert Pollack for teaching
`us the methodology for immunofluorescence, and Dr. Richard Burger
`for his criticism of the manuscript. We are very grateful to Dr. Guenter
`Albrecht-Buehler for expert assistance with the cell migration exper-
`iments and valuable discussions. This work was supported in part by
`American Cancer Society Grant CH-86 and U.S. Public Health Service
`Grants CA 23187 and CA 15714. P.B.S. was supported by U.S. Public
`Health Service Training Grant in Pharmacological Sciences 6T 32 GM
`07260. S.B.H. is a recipient of an Irma T. Hirschl Career Scientist
`Award.
`Wani, M. C., Taylor, H. L., Wall, M. E., Coggon, P. & McPhail,
`1.
`A. T. (1971) J. Am. Chem. Soc. 93,2325-2327.
`Schiff, P. B., Fant, J. & Horwitz, S. B. (1979) Nature (London)
`277,665-667.
`Vasiliev, J. M., Gelfand, I. M., Domnina, L. V., Ivanova, 0. Y.,
`Komm, S. G. & Olshevskaja, L. V. (1970) J. Embryol. Exp.
`Morphol. 24, 625-640.
`Bhisey, A. N. & Freed, J. J. (1971) Exp. Cell Res. 64, 419-429.
`Gail, M. H. & Boone, C. W. (1971) Exp. Cell Res. 65, 221-
`227.
`Crissman, H. A. & Steinkamp, J. A. (1973) J. Cell Biol. 59,
`766-771.
`Weber, K., Bibring, T. & Osborn, M. (1975) Exp. Cell Res. 95,
`11-120.
`
`2.
`
`3.
`
`4.
`5.
`
`6.
`
`7.
`
`8.
`
`9.
`10.
`
`11.
`
`12.
`
`13.
`
`14.
`
`15.
`
`16.
`17.
`
`18.
`19.
`
`20.
`
`Spiegelman, B. M., Lopata, M. A. & Kirschner, M. W. (1979) Cell
`16, 239-252.
`Albrecht-Buehler, G. (1977) Cell 11, 395-404.
`Weber, K., Pollack, R. & Bibring, T. (1975) Proc. Natl. Acad. Sci.
`USA 72, 459-463.
`Osborn, M. & Weber, K. (1976) Proc. Natl. Acad. Sci. USA 73,
`867-871.
`Wang, R. W., Rebhun, L. I. & Kupchan, S. M. (1977) Cancer Res.
`37,3071-3079.
`Schiff, P. B., Kende, A. S. & Horwitz, S. B. (1978) Biochem.
`Biophys. Res. Commun. 85,737-746.
`Abercrombie, M., Heaysman, J. E. M. & Pegrum, S. M. (1970)
`Exp. Cell Res. 59, 393-398.
`Abercrombie, M., Heaysman, J. E. M. & Pegrum, S. M. (1970)
`Exp. Cell Res. 60, 437-444.
`Inoue, S. & Sato, H. (1967) J. Gen. Physiol. 50, 259-292.
`Wilson, L. (1974) Fed. Proc. Fed. Am. Soc. Exp. Biol. 33, 151
`(abstr.).
`Wheatley, D. N. (1971) J. Anat. 110, 367-382.
`Vasiliev, J. M. & Gelfand, I. M. (1977) Int. Rev. Cytol. 50,
`159-274.
`Margolis, R. L., Wilson, L. & Kiefer, B. I. (1978) Nature (London)
`272, 450-452.
`
`IMMUNOGEN 2190, pg. 5
`Phigenix v. Immunogen
`IPR2014-00676
`
`