Schedule Dependence of Vincristine Lethality in Sarcoma 180 Cells following Partial Synchronization with Hydroxyurea

Hamza Mujagic, Bruce M. Conger, Charles A. Smith, et al. Cancer Res 1983;43:3598-3603.

Updated Version

Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/43/8/3598

E-mail alerts Reprints and Subscriptions Permissions

Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at pubs@aacr.org. To request permission to re-use all or part of this article, contact the AACR Publications Department at permissions@aacr.org.

Downloaded from cancerres.aacrjournals.org on December 13, 2012 Copyright 1983 American Association for Cancer Research

[CANCER RESEARCH 43. 3598-3603.

August 1983]

Schedule Dependence of Vincristine Lethality in Sarcoma 180 Cells following Partial Synchronization with Hydroxyurea
Hamza Mujagic,1 Bruce M. Conger, Charles A. Smith, Sandra J. Occhi pinti, William H. Schuette, and Stanley E. Shackney
Section of Cell Kinetics, Clinical Pharmacology Branch, Division of Cancer Treatment, National Cancer Institute [H. M., B. M. C., C. A. S., S. J. O., S. E. S.], and Applied Clinical Engineering Section, BiomdicalEngineering and Instrumentation Branch, Division of Research Services [W. H. S.J, NIH, Bethesda, Maryland 20205

ABSTRACT The effects of exposure to 0.1, 0.5, or 2 pM vincristine for 4 hr were studied in Sarcoma 180 cells at various times after synchronization with 5 mw hydroxyurea for 1 hr. Maximum sensitivity to the lethal effects of vincristine was observed at 10 to 14 hr after hydroxyurea exposure at the higher vincristine concentrations, compared to a period of a maximum sensitivity to a second dose of hydroxyurea at 8 to 12 hr. Serial flow cytometry studies indicated that the apparent decrease in sen sitivity to vincristine at 14 to 18 hr was due to the division of cells in the leading segment of the synchronized wave and their entry into the relatively resistant G phase prior to vincristine exposure. Synchronized cells that had not divided at the time of vincristine exposure were blocked transiently in G2. Serial metaphase index studies suggested that the G2 cells closest to the end of the cell cycle at the time of vincristine exposure were likely to exhibit the greatest degree of mitotic disorganization when they overcame the G2 block and entered metaphase. The present studies suggest that sensitivity to vincristine increases progressively as cells approach mitosis. The molecular mecha nisms underlying this phenomenon are considered in relation to the increase in cell tubulin content during the course of cell cycle progression. INTRODUCTION The Vinca alkaloids, VCR2 and vinblastine, are cytotoxic tubulin-binding agents that are effective in the treatment of leukemias, lymphomas, and a variety of solid tumors in animals and in humans. Early studies of the Vinca alkaloids suggested that their lethal effects were cell-cycle phase-specific (8, 11, 14, 24). These agents were shown to disrupt the mitotic spindle (8,11,14), and their lethality was attributed largely to their effects on cells in mitosis (3, 8, 24). However, in other studies, the Vinca alkaloids were found to exert their lethal effects on interphase cells (5, 9, 10,12,13, 25). When synchronized cells were exposed to VCR, a nadir in clonogenic cell survival was observed during S phase (9, 13) or during late S and G2 (25). In autoradiographic studies of cells doubly labeled with [3H]dThd and [14C]dThd, it was shown that cells exposed to VCR during S and G2 were later arrested in mitosis and subsequently underwent necrosis (5,10). In our own studies on the effects of VCR in asynchronously
1To whom requests lor reprints should be addressed, at Building 10, Room 12C216, National Cancer Institute, Bethesda, Md. 20205. 2 The abbreviations used are: VCR, vincristine; dThd, thymidine; HU, hydroxy urea; HBSS. Hanks' balanced salt solution. Received October 22, 1982; accepted May 5, 1983.

growing Sarcoma 180 cells (16), VCR produced a transient block of interphase cells in G2. We also found that not all cells that were lethally damaged by VCR accumulated in mitosis, and that many of the cells that did accumulate in mitosis did not become necrotic but went on to become polyploid cells. In order to explore these observations in greater detail, we have studied the effects of VCR on cell survival, cell cycle progression, DMA synthesis, and metaphase accumulation in Sarcoma 180 cells that were synchronized by prior exposure to hydroxyurea for 1 hr. The results of these studies are described in this paper. We have found that the lethal effects of VCR became more pronounced as cells progressed through the cell cycle and approached mitosis. The mechanism underlying this observation may be related to the reduplication of cell tubulin content late in the cell cycle. MATERIALS AND METHODS

All studies were carried out in Sarcoma 180 (Foley strain CCRF11; supplied by American Tissue Type Culture, Rockville, Md.) grown in vitro in Earle's Medium 199 (Flow Laboratories, Rockville, Md.), which was supplemented with 10% fetal bovine serum, 2 mw t-glutamine, 100 units of penicillin per ml, and 100 HQ of streptomycin per ml. Cultures were grown in monolayer in 250-ml plastic tissue culture flasks (growth surface area, 75 sq cm) (Costar, Cambridge, Mass.) containing 10 ml of medium. Cells were plated at an initial concentration of 1 x 105 cells/ml. Medium was changed on Days 2 and 4, and cells were subcultured on Day 5. Cell Survival Studies. Two-day-old log-phase cell cultures were in cubated at 37with 5 rriM HU for 1 hr. At the end of the drug exposure period, the medium containing drug was removed, and the cells were rinsed 3 times with 5 ml of HBSS. Cells were then resuspended in Medium 199. A drug-free control flask was included and was treated identically in each experiment. At various intervals after HU pretreatment, VCR was added at final concentrations of 0.1, 0.5, or 2 UM for 4 hr. For reference purposes, time intervals between drugs are reported from the onset of exposure to each drug. At the end of the VCR exposure period, the medium was removed, and the cells were washed 4 times with HBSS. The cells were then harvested by incubation (37) with 0.25% trypsin for 8 min. Controls treated with HU alone were also obtained at each time point. Total cell counts in each flask were determined using a Coulter Counter (Coulter Electronics, Hialeah, Fla.). Cell suspensions were diluted with Medium 199, and known numbers of cells were cloned in soft agar. Nine-day-old colonies were fixed, stained with Giemsa stain, and counted visually. Clonogenic cell yield for each flask, defined as the number of colonies enumerated divided by the number of cells plated, was averaged for 5 replicate flasks at each drug concentration. The viable cell count per flask was calculated as the total cell number per flask multiplied by the clonogenic cell yield at a given time point. The surviving viable cell fraction following VCR exposure was calculated as the number of viable HU-treated cells per flask divided by the number of viable HU-treated control cells per flask, harvested at the same time, Thus, the reported

3598

CANCER

RESEARCH

VOL. 43

Downloaded from cancerres.aacrjournals.org on December 13, 2012 Copyright 1983 American Association for Cancer Research

Vincristine Schedule Dependence

surviving cell fractions after VCR exposure reflected the lethal effects produced by VCR alone. Each experiment was performed at least 3 times, and the reported results represented the log means of replicate studies. For comparison, flasks of cells exposed initially to 5 rriM HU for 1 hr were exposed to a second dose of HU at the same concentration for 1 hr at the same time intervals at which VCR exposures were initiated. Metaphase Index Studies. Serial metaphase indices were obtained at intervals during and after HU and HU plus VCR exposure. Aliquots of cells were cytocentrifuged on glass slides and stained with acetic orcein. Only those cells containing condensed, discrete chromosomes in the absence of a nuclear membrane were included in the numerator of the metaphase index. Metaphases were further classified on the basis of the degree of disruption of normal metaphase chromosome patterns. Meta phases in which the chromosomes were organized in a single ring (or band-like structure, when viewed on edge) were considered to have intact kinetochore microtubules (2) and were classified as organized for our purposes. Metaphases in which chromosomes were spread ran domly throughout the cell, and/or in which there were multiple-ring structures, each involving 2 to 3 chromosomes, were classified as disorganized metaphases. Two thousand cells were counted at each time point. Each experiment was performed 3 times, and reported results represent the means of replicate studies. Flow Cytometry Studies. Serial DNA histograms were obtained be fore, during, and after exposure to HU and HU plus VCR. Cells were washed with HBSS, harvested by trypsinization, fixed in cold 70% ethanol, and stained with mithramycin (100 ^g/ml) in 15 mM MgCl.. and 0.15 M NaCI at a final cell concentration of 1 x 10s cells/ml. Nuclear fluorescence was measured with a Los Alamos cell sorter. Data were recorded, stored, and analyzed on a DEC 11/40 computer system using software developed at the Los Alamos Scientific Laboratory, Los Alamos, N. M. At least 20,000 cells were measured in each sample. All DNA histograms were normalized with respect to total cells analyzed per sample for display purposes. Prior to the measurement of each drug-treated sample, a sample of untreated control cells was run, and amplifier gain settings were adjusted so that the G, peak of the control cells appeared in the same reference channel (Channel 60 of 256). Immediately after the measurement of each drug-treated sample, the position of the G peak of control cells was

10

12

14

16

18

TIME, HR Chart 1. Surviving clonogenic cell fraction following exposure to VCR for 4 hr, initiated at various intervals after exposure to 5 RIM HU for 1 hr. Data are normalized with respect to time-matched controls treated initially with HU alone and reflect the effects of VCR administration only. ,0.1 uVCR; *, MMVCR; 0.5 . M 2 VCR. Effects on cell survival of a second dose of 5 mM HU for 1 hr at various intervals after the first are shown for comparison ( , haded s region). For discussion, see text. Data represent the means of 3 experiments. Bars, S.E.

intervals after an exposure for 1 hr to 5 mw HU. The effects of exposure to a second dose of HU are shown for comparison. Since the data are normalized with respect to time-matched controls treated with HU alone (see "Materials and Methods"),

they reflect the effects of administration of the second drug only on cell survival. When cells were exposed to 0.1 MMVCR for 4 hr, starting at 2 to 10 hr after exposure to HU, the surviving clonogenic cell fraction ranged from 0.70 to 0.75 (Chart 1). When cells were exposed to 0.1 ^M VCR at 12 to 14 hr after exposure to HU, the surviving cell fraction fell to 0.5. Exposure to 0.1 /UMVCR at 16 and 18 hr produced surviving cell fractions of 0.6 and 0.75, respectively. When cells were exposed to 0.5 ^M VCR at various intervals after exposure to HU, the schedule-dependent effects on cell <50; G, Channels 50 to 70; S, Channels 71 to 99; G2 M, Channels 100 to 140; and post-G2-M, Channels >140. The fraction of cells in each survival were more pronounced. The nadir in the surviving cell fraction (0.25 to 0.31) was observed at 12 to 14 hr (Chart 1). region was calculated for each histogram. Because the gain settings were adjusted to an external reference standard within one channel of Cell killing was also greater at 6 to 10 hr with 0.5 //M VCR than with 0.1 UM VCR. The effects of 2 >IMVCR were similar to those 60 for each sample, valid comparisons could be made among sequential histograms, and corresponding data from multiple experiments could be observed at 0.5 mw VCR. The nadir in the surviving cell fraction grouped for analysis. The reported results represent the means of 3 (0.22 to 0.27) also occurred at 12 to 14 hr. The decrease in experiments. Although by this method the cell fraction in each region surviving cell fraction at 10 hr was greater with exposure to 2 was contaminated to some extent by cells from adjacent regions, this M than with 0.5 M VCR VCR. proved to be of little practical consequence. Major shifts of cells from By comparison, the nadir in the surviving cell fraction following one region of the histogram to another that were apparent on visual exposure to a second dose of HU was observed between 6 and inspection were faithfully reflected in the processed data. 12 hr and ranged from 0.45 to 0.55 (see Chart 1, shaded region). While there was extensive overlap in the intervals of maximum RESULTS sensitivity to a second dose of HU and to VCR, respectively, the Cell Survival Studies. Chart 1 shows the effects on cell interval of maximum sensitivity to VCR extended 2 to 4 hr later survival of VCR exposure for 4 hr, commencing at various time than that for a second dose of HU, and the nadir was lower at
rechecked to ensure that it had not drifted during the course of the measurement. DNA Histogram Analysis. Because of the presence of large numbers of nuclear fragments and polyploid cells at various times after VCR exposure (see Chart 4, and Ref. 15), conventional methods of DNA histogram analysis could not be used. In order to quantitate population shifts among the cell cycle phases, and in order to quantitate the development of nuclear fragmentation and polyploidy during and at various times after drug exposure, the DNA histogram was analyzed by a modification of the method of Alabaster and Cassidy (1). The DNA histogram was divided into bounded regions as follows: pre-G,. Channels AUGUST 1983

3599

Downloaded from cancerres.aacrjournals.org on December 13, 2012 Copyright 1983 American Association for Cancer Research

H Mu/agicet

al region in the presence of drug through the 14-hr time point (Chart 2D, shaded region), ana the fraction of cells in the Gz-M region remained elevated or continued to rise for at least 2 hr after drug removal. Peak fractions of VCR-treated cells in the Gz-M region

VCR concentrations of 0.5 (iu or greater. In earlier radioautographic studies of the effects of exposure of 5 m** HU for 1 hr, peak rates of DMA synthesis during the pos ttreat men t period were observed at 6 to 10 hr. Concorri itantly, a wave of recruited eels was shown by flow cytometry to move through early S at 6 hr, mid-S at 8 to 10 hr, and late Searly G2 at 12 hr (6). Preliminary studies of changes in the metaphase index following exposure to 5 IM HU for 1 hr indicated a decrease in the metaphase index during the first 10 to 12 hr, and a rise at 14 to 18 hr. That is, exposure to HU for 1 hr was followed by the appearance of a cohort of rapidly cycling, syn chronized ceOsthat moved through mid-S at 8 to 10 hr, traversed late S and 62 at 12 to 14 hr, and underwent mitosis between 14 and 18 hr. Thus, one interpretation of the cell survival studies shown in Chart 1 might be that susceptibility to VCR peaked in late S and earty Gz and then decreased late in G2. Alternatively, it may be that cells in late Gz were no less susceptible to VCR effects than were eels in late S and early G.. and that the apparent decrease in drug effectiveness at 14 to 18 hr was due to the vanguard of the partially synchronized wave of cells dividing and entering a relatively resistant G, phase at this time. In order to distinguish between these 2 possibilities, we chose the 10-and 14-hr time points after HU exposure for more detailed

ranged from 0.8 to 0.9. However, concomitant metaphase index studies showed that the fraction of cells in metaphase at 10 to 16 hr was less than 0.05 at all drug concentrations (see below and Chart 4 and associated text). Thus, the cells that accumu lated in the Gz-M region of the ONA histogram between 12 and 18 hr were almost all premi tot ic cells in Gz. A similar VCRinduced G, block was observed in asynchronously growing sarcoma 180 cells as well (16). The Gz block induced by VCR was transient, and its duration was dependent on VCR concentration. Following exposure to 0.1 MVCR, the fraction of cells in 62 fed to control values by 22 hr (Chart 2D). Following exposure to 0.5 and 2 /IM VCR, a significant decrease in the G2 fraction was observed at 22 hr, but return to control values did not occur until 26 hr. The progressive and sustained increase in the G, cell fraction of 0.2 to 0.3 above the HU-treated controls between 12 and 18 hr at VCR concentrations of 0.5 and 2 < (Chart 2D) would M suggest that VCR induced a 62 block, not only in all cells that were in G2 at the time of drug exposure, but also in S-phase ceils that were within 6 to 8 hr of mitosis at the time of initial study by means of flow cytometry. Row Cytometry Studies. Chart 2 shows the effects on the drug exposure. This is supported by the observations that VCRfractions of cete in Afferent regions of the DMA histogram of a treated cells continue to empty out of the S region at up to 18 4-hr exposure to VCR that was initiated at 10 hr after HU hr (Chart 2C) but do not proceed from the Gz-M region into the exposure. The effects of VCR concentrations of 0.1, 0.5, and 2 G region until after 18 hr (Chart 2B. 0.5 and 2 ^M VCR). This ftu are shown in comparison with the effects of initial exposure would suggest that the wave of cells recruited by HU that was passing through G. at the time of VCR exposure had a trailing to HU alone. The presence of VCR did not prevent the continued progression of the wave of synchronized cells through the S segment that extended well back into S and included approxi region and into the Gz-M region of the DNA histogram at any of mately 20 to 30% of the population, i.e., that, initially, HU induced the VCR concentrations studied (Chart 2C, shaded region). only partial synchronization of these cells. At ail 3 VCR concentrations, cells accumulated in the G2-M Cells arrested in Gz by VCR underwent one of several subse quent fates. A small fraction of the cells was arrested transiently in metaphase at 18 and 22 hr (see below); presumably, these metaphases were included in the descending limb of the wave of cells in the G2-M region of the histogram (Chart 2D). Some cells resumed cycling and reappeared in the G, region of the histogram. A larger proportion of cells exposed to 0.1 MM VCR recovered than after exposure to 0.5 or 2 UM VCR, and recovery occurred more rapidly at the lowest drug concentration (Chart 2).Some G2-arrested cells subsequently underwent fragmen tation and appeared in the pre-Gi region of the histogram (Chart 2A), white others went on to become polyploid cells that ap peared in the post-Gz-M region of the histogram (Chart 2). Chart 3 shows the effects on the fractions of cells in different regions of the DNA histogram of a 4-hr exposure to VCR initiated at 14 hr after HU exposure. The patterns are somewhat more complex, due to the fact that cells in the leading segment of the HU-recruited wave had divided and reappeared in G, between 12 and 14 hr, just prior to VCR exposure (Chart 3B), while cells in the broader trailing segment of the HU-recruited wave had not divided yet when VCR was introduced. While VCR promptly 02 6 10 14 18 22 26 30 blocked the entry of new cells into d, it did not prevent the 20% TIME, HR of the total population that had already reentered G from pro Chart 2. Changes in the tractions of cete as a function of time in diffrent gressing out of the G region between 14 and 22 hr (Chart 3, regions of the DNA histogram in cete exposed to 5 nu* HU for 1 hr and to various concentrations of VCR at 10 hr. A. PRE-G, (eel fragments) region. 8. G, region. C. shaded region and beyond). This subgroup of recruited cells that S region. O. G^M region. E. POST GJA region ipredormnanUy pofyploid cete). . had escaped the effects of VCR could be seen passing through HU controls; T, 0.1 *u VCR; A. 0.5 IM VCR; , jriwVCR. Shaded region, time 2 of exposure to VCR. Al data points represent the means of 3 experiments. Bars, and leaving the S region between 16 and 24 hr (Chart 3C). S.E. For discussion, see text Presumably, it was this subgroup of recruited, partially synchro3600
CANCER RESEARCH VOL. 43

Downloaded from cancerres.aacrjournals.org on December 13, 2012 Copyright 1983 American Association for Cancer Research

Vincristine Schedule Dependence observed at 4 to 8 hr after the removal of VCR. When Chart 4 is considered together with Charts 2D and 3D, it becomes apparent that premitotic cells were blocked tran siently in G2 during VCR exposure and for 2 to 4 hr after drug removal. Thus, the delayed appearance of the peaks in the metaphase index until 4 to 8 hr after drug removal, regardless of the time of VCR exposure (Chart 4), is directly attributable to prior VCR-induced G2 block. The differences in the ranges of peak values of the metaphase index in Chart 4, A and B, are attributable to differences in the proportions of cells in the population that were susceptible to VCR-induced metaphase arrest in the 2 drug schedules. When VCR was introduced at 10 hr after HU exposure, the HU-recruited wave was just entering G2 (Chart 2D); when VCR was introduced at 14 hr, many of the HU-recruited cells had already divided and entered d (Chart 3B). It is clear that, of all the cells that were blocked transiently in G2 by VCR, only a small proportion were subsequently arrested in metaphase after 4 hr of exposure to the drug. To explore the possibility that the cells that were closest to mitosis at the time of VCR addition were those that were more likely to undergo metaphase arrest once the G2 block was overcome, metaphases were subclassified with respect to the degree of disruption of spatial organization of the chromosomes in the metaphase figure (see "Materials and Methods"). Organized and disorganized metaphases were then scored separately. The total metaphase index and organized metaphase index are shown in Chart 5 for each VCR concentration and drug schedule. At concentrations of 0.1 UM VCR. nearly all metaphases were organized, whether the drug was administered at 10 or 14 hr after HU (Chart 5, A and B, respectively). At the higher drug concentrations, disor ganized metaphases first appeared at the end of the 4-hr drug exposure period and peaked at 4 hr after the termination of drug exposure, regardless of drug schedule (Chart 5, /42,/43, B2, and B3). In contrast, at 8 hr after the termination of drug exposure, the metaphase peaks consisted almost entirely of organized metaphases. These data are consistent with the premise that the cells closest to mitosis were the first to overcome VCRinduced G2 block and, subsequently, exhibited the greatest degree of disruption of metaphase organization. Presumably, cells that were 2 to 4 hr away from mitosis at the time of VCR exposure overcame VCR-induced G2 block at 4 to 8 hr after

~0 2

10 14 18 22 26 30 34 38 TIME. HR

02

10 14 18 22 28 30 34 38

Chart 3. Changes in the fractions of cells as a function of time in cells exposed to 5 mu HU for 1 hr, and to 0.1, 0.5, and 2 IM VCR at 14 hr. Legend as in Chart 2. Shaded region, time of exposure to VCR. All data represent the means of 3 experiments. Bars, S.E. For discussion, see text.

nized cells that contributed to the late rise and fall in the G2-M region between 22 and 30 hr (compare Charts 2D and 3D). If this is so, then the wave of cells in the G2-M region between 14 and 30 hr (Chart 3D) actually consisted of 2 components, namely, the static VCR-blocked trailing segment of the initial HU-recruited wave, upon which was superimposed the moving wave of cells that had already escaped into d by the time of onset of VCR exposure and had traversed the cell cycle once more over the succeeding 10 to 12 hr. It is of some interest that, in this series of experiments, one can identify clearly a cohort of recruited cells treated with HU alone in which synchrony was sustained through a second cell cycle. An initial wave of HU-treated controls could be seen entering G 14 to 16 hr (Chart 30), traversing S between 16 at and 24 hr (Chart 3C), and traversing the G2-M region between 22 and 30 hr (Chart 3D). In retrospect, sustained synchrony may also have been present among the HU controls in Chart 2, to D, but the synchronized waves are not as well defined. In summary, Charts 2 and 3 show that the leading segment of the synchronized wave began to divide between 10 and 14 hr after HU exposure. Cells that had not divided by the time of initiation of VCR exposure were promptly blocked in G2; in contrast, cells that had divided and entered G the time of at introduction of VCR were not prevented from traversing the cell cycle. Metaphase Index Studies. During the course of the flow cytometry studies, aliquots of cells were also taken for parallel determinations of the metaphase index at each time point. The results are shown in Chart 4. Whether VCR was added at 10 hr after HU (Chart 4/4) or at 14 hr after HU (Chart 4),there were only minimal changes in the metaphase index after 4 hr of drug exposure, and for 2 hr thereafter. Increases in the metaphase index above 0.1 were observed only in cells exposed to 0.5 and 2 UMVCR at 10 hr after HU exposure (Chart 4/4); peak metaphase indices ranging from 0.10 to 0.15 were seen at 4 to 8 hr after VCR removal. When 0.5 and 2 UM VCR were added at 14 hr after HU, there were modest rises in the metaphase index above control values; peak values ranging from 0.05 to 0.075 were

26X02
TIME,

61014182226303438
HR

Chart 4. Changes in the metaphase to 5 rriM HU for 1 hr, and to 0.1, 0.5, Legend as in Chart 2. Shaded region, represent the means of 3 experiments.

index as a function of time in cells exposed and 2 ^M VCR at 10 hr (A) or at 14 hr (fl). time of exposure to VCR. All data points Bars, S.E. For discussion, see text.

AUGUST

1983

3601

Downloaded from cancerres.aacrjournals.org on December 13, 2012 Copyright 1983 American Association for Cancer Research

H. Mujagic et al.
dure itself can lead to erroneous conclusions regarding the phase specificity of drug lethality. For example, the mitotic shake tech nique might select for the most rapidly proliferating component 0.10 of a population, and this selected cohort might traverse the cell cycle more rapidly than expected. On the other hand, synchro nizing agents such as HU are cytotoxic and might introduce unwanted retardant effects on cell cycle progression. Regardless of which method is chosen, the degree of synchronization can be expected to diminish progressively as cells traverse the cell cycle (22), and this, too, can affect the interpretation of cell survival studies. Since the technique of mitotic selection does not provide sufficient numbers of cells for analysis by flow cytometry, we used HU synchronization in order to examine directly the behav ior of synchronized sarcoma 180 cells before, during, and after exposure to VCR for 4 hr. Although the cell survival patterns superficially resembled those that might be expected for a cycle phase-specific drug (Chart 1), a more detailed analysis of the data showed that the sensitivity of cells to the lethal effects of VCR increased progressively as they traversed S and G2. It was apparent from our serial flow cytometry studies that the HUsynchronized cohort of cells actually traversed the late portion of cell cycle in a broad wave with a relatively long trailing segment (Charts 2 and 3). The apparent decrease in VCR lethality between 12 and 18 hr after exposure to HU (Chart 1) could be attributed to the division of an increasingly larger proportion of cells in the HU-recruited cohort and their entry into a relatively resistant G phase during this period (Charts 2B and 38). There were striking 6 10 14 18 22 26 30 02 6 10 14 18 22 26 30 34 38 differences in the subsequent behavior of cells that had just TIME,HR divided prior to VCR exposure and cells that were approaching Charts. Changes in th total metaphase index ( ) and in the organized mitosis but had not yet divided at the time of VCR addition. VCR metaphase index ( ) a function of time after exposure to 5 mw HU for as 1 hr and 0.1, 0.5, or 2 UM VCR at 10 hr (A,, A,, and A3, respectively) and 0.1, 0.5, had no effect on the movement out of d of cells that had already or 2 put VCR at 14 hr (,,B,and B3, respectively). Difference between total and divided (Chart 3) nd had no effect on their subsequent move a organized metaphase index at each time point represents the fraction of disorga nized metaphases. Shaded region, time of exposure to VCR. Each data point ment through S (Chart 2C). However, the entry of new cells into represents the mean of 3 experiments. Bars, S.E. For discussion, see text. d was halted rapidly (Charts 2B and 3B), and cells began to accumulate in the G2-M region of the histogram (Charts 2D and 3D) soon after the introduction of VCR. drug removal, and their metaphases exhibited a greater degree In the present study, there were several other indications that of organization. cells approaching mitosis were especially sensitive to the effects of VCR. The transient blockade of cells in G2 (Charts 2D and 3D) is itself a manifestation of this phenomenon. Our metaphase DISCUSSION index data are also consistent with the premise that the G2 cells Effects of VCR in Relation to the Cell Cycle. Published closest to the end of the cell cycle at the time of initial exposure studies on the cell cycle-dependent lethality of VCR have yielded to VCR were the cells that were likely to exhibit the greatest seemingly conflicting results. Madoc-Jones and Mauro (13) and degree of mitotic disorganization when they overcame the G2 Hill and Whelan (9) have claimed that VCR is S-phase-specific block and entered metaphase (Chart 5). on the basis of their studies in HU-synchronized and mitotically Mechanisms of VCR Lethality. The finding that sensitivity to VCR lethality increases progressively as cells traverse the late selected cells, respectively. That is, VCR sensitivity was thought to peak during S and decline thereafter as cells approached portion of the cell cycle may have important implications with mitosis. In contrast, the studies of Wibe (25) in mitotically se regard to understanding the relation between known molecular lected cells suggested that sensitivity to the lethal effects of VCR mechanisms of VCR action and drug lethality. VCR is known to increased progressively as cells traversed the cell cycle, i.e., that bind specifically to tubulin (17-19), a major component of intracells in G2 were more sensitive to VCR than were cells in S. cellular microtubular structures that are essential for a variety of Studies in asynchronously growing cells double-labeled with normal cell functions, including cell division. The lethal effects of [3H]dThd and [14C]dThd have suggested that cells in both S and the Vinca alkaloids are commonly attributed to their tubulinbinding properties (23). Historically, the primary target of tubulinG2, at the time of VCR exposure, were later arrested in meta phase and subsequently went on to become necrotic (5, 10). binding agents was thought to be the mitotic spindle, the disrup These findings were subsequently confirmed in time-lapse cine tion of which prevents centriole migration (2, 8, 21). However, other microtubule structures, such as the kinetochore microtumatography studies (12). Cell survival studies in synchronized cells must be interpreted bules, are also involved in the mitotic process. The transient VCR-induced G2 block observed in these and other published with care. Artifacts associated with the synchronization proce
0.15

3602

CANCER

RESEARCH

VOL. 43

Downloaded from cancerres.aacrjournals.org on December 13, 2012 Copyright 1983 American Association for Cancer Research

Vincristine Schedule Dependence studies (25) may reflect the disruption or prevention of formation of microtubular structures associated with very early stages of mitosis. Microtubules are also involved in the maintenance of cell shape and locomotion; disruption of the cell cytoskeleton may account for reported observations of interphase cell lysis by tubulin-binding agents (12, 13, 20). Total cell tubulin content normally doubles between cell divi sions, increasing in sigmoidal fashion as cells progress through the cell cycle (4, 7). Peak rates of tubulin synthesis occur during a phase of the cell cycle that overlaps S and G2 (4). Using fluorescent antitubulin antibody to determine cell tubulin content, we have shown by flow cytometry that the cell tubulin content distribution in Sarcoma 180 cells spans a 2-fold range and exhibits a prominent postmitotic peak and a saddle shape, much like the DMA histogram (15). This would imply that cell tubulin content increases in parallel with cell DMA content as cells progress through the cell cycle; like cell DMA content, cell tubulin content doubles between cell divisions and halves at mitosis. It is conceivable that the increase in sensitivity to the lethal effects of VCR during the course of cell cycle progression in Sarcoma 180 cells might be related to the increase in cell tubulin content during the course of cell cycle progression. That is, cells exposed to VCR before they have synthesized a full complement of tubulin might, after VCR removal, go on to synthesize a sufficient amount of new, VCR-free tubulin to form the microtu bular structures that are required for mitosis. On the other hand, cells in which tubulin synthesis had nearly been completed at the time of VCR exposure would not synthesize enough new tubulin for normal cell division, even if the drug were removed prior to mitosis. Additional studies will be required to explore this possi bility.
arrested in metaphase by vincristine. Cell Tissue Kinet., 73: 239-250,1980. 6. Ford, S. S., and Shackney, S. E. Lethal and sublethal effects of hydroxyurea in relation to drug concentration and duration of drug exposure in sarcoma 180 in vitro. Cancer Res., 37: 2628-2637, 1977. 7. Forrest, G. L, and Klevecz, R. R. Synthesis and degradation of microtubule protein in synchronized Chinese hamster cells. J. Biol. Chem., 277: 31473151, 1972. 8. George, P., Journey, L. J., and Goldstein, M. H. Effect of vincristine on the fine structure of HeLa cells during mitosis. J. Nati. Cancer Inst., 35: 355-375, 1965. 9. Hill, B. T., and Whelan, R. D. H. Comparative cell killing and kinetic effects of vincristine or vindesine in mammalian cell lines. J. Nati. Cancer Inst., 67: 437443,1981. 10. Jellinghaus, W., Schultze, B., and Maurer, W. The effect of vincristine on mouse jejunal crypt cells of differing cell age: double labeling autoradiographic studies using 3H- and 14C-TdR. Cell Tissue Kinet., 70: 147-156,1977. 11. Krishan, A. Time-lapse and ultrastructure studies on the reversal of mitotic arrest induced by vinblastine sulfate in Earie's L cells. J. Nati. Cancer Inst., 47:581-595,1968. 12. Lengsfeld, A. M., Schultze, B., and Maurer, W. Time-lapse studies on the effect of vincristine on HeLa cells. Eur. J. Cancer, 77: 307-319, 1981. 13. Madoc-Jones, H., and Maura, F. Interphase action of vinblastine and vincristine: differences in their lethal action through the mitotic cycle of cultured mammalian cells. J. Cell Physiol., 72: 185-196, 1968. 14. Malawista, S. E., Sato, H., and Bensch, K. G. Vinblastine and griseofulvin reversibly disrupt the living mitotic spindle. Science (Wash. D. C.), 760: 770772, 1968. 15. Mujagic, H., Chen, S-S., Geist, R., Occhipinti, S. J., Conger, B., Smith, C. A., Schuette, W. A., and Shackney, S. E. Effects of vincristine on cell survival, cell cycle progression, and mitotic accumulation in asynchronously growing sar coma 180 cells. Cancer Res., 43: 3591-3597, 1983. 16. Mujagic, H., Smith, C. E., Schuette, W. H., and Shackney, S. E. The cell tubulin content frequency histogram obtained by flow cytometry in sarcoma 180 cells in vitro. Proc. Am. Assoc. Cancer Res., 23: 28, 1982. 17. Owellen, R. J., Donigian, D. W., Hartke, C. A., Dickerson, R. M., and Kuhar, M. J. The binding of vinblastine to tubulin and to paniculate fractions of mammalian brain. Cancer Res., 34: 3180-3186,1974. 18. Owellen, R. J., Hartke, C. A., Dickerson, R. M., and Hains, F. O. Inhibition of tubulin-microtubule polymerization by drugs of the Vinca alkaloid class. Cancer Res., 36: 1499-1502,1976. 19. Owellen, R. J., Owens, A. H., Jr., and Donigian, D. W. The binding of vincristine, vinblastine and colchicine to tubulin. Biochem. Biophys. Res. Commun., 47: 685-691,1972. 20. Shackney, S. E., Bunn, P. A., and Ford, S. S. The effects of colcemid on mouse bone marrow. Cell Tissue Kinet., 9: 363-369, 1976. 21. Stubblefield, E. Ceninole replication in mammalian cells. In: The Proliferation and Spread of Neoplastic Cells, M. D. Anderson Hospital and Tumor Institute at Houston, pp. 175-193. Baltimore: The Williams & Wilkins Co., 1968. 22. Terasima, T., and Tolmach, L. J. Growth and nucleic acid synthesis in syn chronously dividing populations of HeLa cells. Exp. Cell Res., 30: 344-362, 1963. 23. Tucker, R. W., Owellen, R. J., and Harris, S. B. Correlation of cytotoxicity and mitotic spindle dissolution by vinblastine in mammalian cells. Cancer Res., 37: 4346-4351,1977. 24. Valeriote, F. A., Bruce, W. R., and Meeker, B. E. Comparison of the sensitivity of normal hematopoietic and transplanted lymphoma colony-forming cells of mice to vinblastine administered in vivo. J. Nati. Cancer Inst., 36: 21-27,1966. 25. Wibe, E. Age-dependent cell inactivation by vincristine alone or in combination with 1-propargyl-5-chloropyrimidin-2-one. Cancer Res., 40: 2069-2073,1980. 26. Wibe, E., Oftebro, R., Christensen, T., Laland, S. G., Pettersen, E. 0., and Lindmo, T. Inhibitory effects of the new mitotic inhibitor 5-chloropyrimidin-2one and of vincristine on human cells in vitro. Cancer Res., 38: 560-565, 1978.

REFERENCES
1. Alabaster, O., and Cassidy, M. Flow microfluorimetric analysis of P388 murine leukemia after administration of vincristine and maytansine in vivo. J. Nati. Cancer Inst., 60: 649-652,1978. 2. Brinkley, B. R., Stubblefield, E., and Tsu, T. C. The effects of colcemid inhibition and reversal on the fine structure of the mitotic apparatus of Chinese hamster cells in vitro. J. Ultrastruct. Res., 79:1-18,1967. 3. Bruchovsky, N., Owen, A. A., Becker, A. J., and Till, J. E. Effects of vinblastine on the proliferative capacity of L cells and their progress through the division cycle. Cancer Res., 25:1232-1237,1965. 4. Bucher, N. L. R., and Berkley, P. Microtubule protein in relation to the cell cycle in regenerating rat liver. In: C. V. F. Simpson (ed.), The Cell Cycle in Malignancy and Immunity, pp. 67-75. Springfield, Va.: National Technical Information Service, United States Department of Commerce, 1975. 5. Camplejohn, R. S., Schultze, B., and Maurer, W. An in vivo double-labeling study with the JF-1 mouse ascites tumour of the subsequent fate of cells

AUGUST 1983

3603

Downloaded from cancerres.aacrjournals.org on December 13, 2012 Copyright 1983 American Association for Cancer Research