[CANCER RESEARCH 40, 42-46, January 1980] 0008-5472/80/0040-0000$02.00 Effect of DihydroxyanthraquinoneandRadiationon G2Progre@sion1

Bruce F. Kimler

Radiation Biology Laboratory, Department of Radiation Therapy, University of Kansas Medical Center, Kansas City, Kansas 66 10@3

ABSTRACT MATERIALS AND METHODS The effect of dihydroxyanthraquinone (DHAQ), a potential Cell and Culture Conditions. Chinese hamster ovary cells anticancer chemotherapeutic agent, on the progression of were maintained in exponential groMh as monolayer cultures Chinese hamster ovary cells into mitosis and on the division at 37°in6% CO2. McCoy's modified@Medium5Awas supple delay induced by ionizing radiation was studied using the mented with 10% fetal calf serum and penicillin plus strepto mitotic selection procedure for cell cycle analysis. mycin (KC Biological, Lenexa, Kans.) Under these conditions, Following the addition of DHAQ, the number of mitotic cells the cells displayed a doubling time of pproximately 14 hr with @ selected from an asynchronous population remained unaltered G1 5 hr, S 7 hr, G2 i .5 hr, and 0.7 hr as determined for a refractory period and then decreased. This effect was by autoradiographic studies of sync onous cells. Cells were concentration dependent with transition points between the S/ monitored periodically with a fluoresc nt stain technique (i ) to G2 boundary at iO-4 @.tg/mland the G2/M boundary at :i02 assure the absence of Mycop!asma c ntamination. @g/ml.The duration of the transient division delay was de Mitotic Selection Procedure. The itotic selection proce pendent upon the concentration of drug used and the duration dure for cell cycle analysis (6) was utilized to monitor the ofpulseexposure. perturbatory effects of DHAQ upon elI cycle progression. When cells were treated with pulses of DHAQ in addition to Briefly, ten 75-sq cm tissue culture fI ks containing approxi X-irradiation, there was no change in the location of the radia mately I 0@asynchronous cells in e ponential growth were tion transition point. There was an increase in the duration of shaken for 20 sec every i 0 mm on a horizontal shaker. The division delay compared to that produced by X-ray alone that medium, now containing the detached, ounded-up cells in late was dependent upon the concentration and duration of drug mitosis, was collected on ice and repl ed with i 0 ml of fresh treatment. medium. The cell suspension was c unted using a Coulter The effect of DHAQ is similar to that of other cancer chem counter coupled to a frequency size a lyzer to verity that the otherapeutic agents (Adriamycin, bleomycin, and lucanthone), collected cells were predominately mit tic cells with a volume and the same cautions should therefore be considered when twice that of the smallest (G1) cells. @Themitotic index, as combining DHAQ and radiation for clinical use. determined by fixing and staining with acetoorcein, was cus tomarily greater than 90 to 95%. This chnique resulted in a record of the number of cells that mov into a narrow region INTRODUCTION of mitosis, approximately 4 to 22 mm ior to cytokinesis (6), DHAQ2(NSC 279836) is a newly synthesized compound that i.e., the selection window, every i 0 in (the mitotic rate). has been shown to be effective in preliminary antitumor testing Treatment (9 treatment flasks and 1 co rol flask) was accom (8). At present, the drug is entering Phase I clinical investiga pushed by either irradiating the flask or replacing the control tions. Since the compound is structurally related to a family of medium with medium containing the de ired concentration of intercalating antibiotics that includes Adriamycin, actinomycin drug for the required number of shakes. D, and lucanthone, care must be taken to avoid any significant X-lrradiation. A Norelco300 kVp X-ra unitwas operatedat problem with normal tissue damage. Preliminary data suggest 10 ma with i .2 mm aluminum addition I filtration (half-value that this compound has much less cardiotoxicity and nephro layer, i .0 cm aluminum), producing a se rate of i .2 Gy3/ toxicity than do the conventional antibiotics. It seems likely that mm at 45 cm. The irradiation procedure cessitated the flasks DHAQ will one day be utilized with radiation therapy in a being out of the incubator at room tem rature for 4 mm, but combined modality approach. Therefore, it seems prudent to this did not cause any significant pertur ation of mitotic rate. establish some guidelines by which to estimate the potential DrugTreatment. DHAQwas obtained omthe DrugSynthe for synergistic action of the modalities. The mitotic selection sis and Chemistry Branch, Division of C ncer Treatment, Na procedure for cell cycle analysis (6) was used to investigate tional Cancer Institute. The drug was diss Ived In balanced salt the effects of DHAQ on cell cycle progression, the concentra solution, sterilized by filtration, and stor as a concentrated tion-dependent blockade in the G2 phase, and the interaction solution in the freezer until needed. Pr' r to treatment, the of DHAQ and radiation-induced division delay. It is hoped that concentrated solution was diluted in co plete medium. The these results will assist the oncologist in the design of combined drug solutions were protected from hg t so as to prevent modality therapeutic schedules using radiation therapy and photodegradation. Drug-containing medi at the appropriate DHAQ so as to avoid the severe complications that resulted concentrations was added to the flasks fo the desired number from the early use of actinomycin 0 and Adriamycin in combi of shakes before control medium was onc again used. nation with radiation therapy. RESULTS

1 Supported in part by Mid-America Cancer Center Program, National Cancer The effect of a 60-mm exposure to various concentrations of Institute Grant 5 P30 CAl 5080-04. DHAQ (1 o-@ to i 0°jig/mI) on the selectiofl of mitotic Chinese 2 The abbreviation used Is DHAQ, dihydroxyanthraquinone.

Received April 30, 1979; accepted October 4, 1979. 3 One Gy = 100 red.

42 CANCER @ESEARCHVOL.40

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1980 American Association for Cancer Research. @@@@ @8O I

DHAQ-Radiation Division Delay

hamster ovary cells is shown in Chart i . For comparison, the tion of this experiment. In other experiments, longer treatment effect of X-irradiation (2.0 Gy) on the induction of a progression times produced the same lower plateau region with no recovery blockade (midpoint, 40 mm) and the subsequent period of as long as the drug was present. division delay (206 mm) is also shown. Mitotic cells continued The midpoint of progression blockade as a function of the to enter the selectionwindowat the controlrate for a periodof logarithm of DHAQ concentration is plotted in Chart 2. The 40 to 80 mm after initiation of drug treatment and then declined composite curve from 14 experiments was described by a to a concentration-dependentminimum.The midpointof pro straight line since the midpoint occurred closer to mitosis with gression blockade (the time from initiation of treatment to the increasing drug concentration. Identical midpoints were ob time at which the mitotic rate is 50% of the control; Ref. 5) tamed for pulse exposures (i 0 to 60 mm) as for continuous decreased with increasing concentration of DHAQ. The in treatment, indicating that the response is indeed the result of creasing concentration also resulted in a progressive shorten different points of action in G2 and is not due to different rates ing of the time that cells were selected at the control rate. of drug diffusion, uptake, or intercalation. A plateau was at Concentrations of i 0@ and 10_2 @g/mldid not completely tamed where the midpoint did not decrease further with in reduce the number of cells selected per shake to the constant creasing concentration. This minimum value is coincident with background level of nonmitotic cells and debris over the dura the value obtained for radiation-induced blockade, as has been shown for a number of cancer chemotherapeutic agents (5). @@T_ I I ‘ ‘ The mean value from i 4 experiments for the X-ray transition

. 2.0 Gy X-ray EXP K 98 point (36 mm) was used to define T0(see below). It was often DHAQ difficult to obtain data for concentrations above i 0' @ig/mldue to the increasedselectionof nonmitoticcells. However, in no experiment was there a drug transition point less than the X i05 ray transition point, confirming the concept of a minimum transition point. The linear portion of the curve was calculated w by least-squares regression analysis from i O@to i 02 @sg/mlto I fit the equation U, w 0. T= T0+ m log(C0/C) ‘I, .J -J where T is the midpoint value in mm prior to the selection Midpoint window, T0is the minimum transition point attained (36 mm), m 40mm is the slope constant to be derived, C0 is the concentration at 64 mm which T0is obtained (i 02 ,@g/ml;2.2 x i 02 SM), and C is the 75 mm independent variable concentration. The value of i 0 for the 85 mln slope constant is significantly lower than are the values for actinomycin D, Adriamycin, lucanthone, mitomycin C, and bleo V lO@}tg/ml 105 mln -a mycin (a range of 25 to 68) that have been previously reported @ ‘U- iôo 200 300 400 500 (5). This difference cannot be due to the choice of a value of TIMEAFTERINITIATIONOFTREATMENT,MINUTES 102 @sg/mlfor C0 since Co would have to change by several Chart 1. Effect of various concentrations of DHAQ on the selection of mitotic orders of magnitude to appreciably alter the value of the slope cells. At time 0, the control medium (0) was replaced with medium containing DHAQ at the indicated concentration for a 60-mm drug exposure or a flask constant. The low value for m indicates that the effectiveness received 2.0 Gy x-ray. of DHAQ for inducing a concentration-dependent blockade of

Concentration, pM @ . 0.2 o io@ lo@ io2

40 . @ @i2O 5

oIOO @ Chart 2. Effect of DHAQ concentration on the midpoint of pro l@j PG gression blockade, i.e., the time prior to selection after which cells are refractory to drug-Induced delay and proceed through mitosis. The data from 14 experiments were averaged; numbers above bars @ (S.D.), number of individual determinations that contributed to the @6O t-.,...-...'.J average. The location of the X-ray midpoint of progression blockade @ is indicated for the average value obtained from the 14 experiments. @i_ @ a40 °- .-@ @-@------X@ray36@'n'fl——-----@--- 0 @ ;!.2O@,IIII ,d@ )Q3 102 161 100 101 02 I0@ Concentration, pg/mi

JANUARY1980 43

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1980 American Association for Cancer Research. B. F. Kimler G2progression is less than that of the other agents mentioned. of the mitotic rate to about 30% of the control value at the Moreover, the concentration required to produce the minimum nadir. The midpoint of progression bk$ckade is 78 mm prior to midpoint of blockage, i.e., C0, was greater for DHAQ than for selection, in agreement with Charts 1 @nd2.After a period of the others (5), indicating that the absolute molar effectiveness time, the mitotic rate increased as tells recovered from a of DHAQ for G2 blockade was also less. transient division delay. The combina on of DHAQ and X-ray Chart 3 shows the effect of a i 0-mm pulse of i 0_2,@gDHAQ produced a mitotic rate curve that w identical to the curve per ml on the division delay produced by 2.0 Gy X-ray. These produced by X-ray alone (dashed un ) over the descending conditions (i 0 mm and i 0_2 @g/ml)werechosen (see below) portion of the curve. The midpoint of p ogression blockade for since they produced quantifiable effects with a minimum of both treatments was 37 mm prior to lection. However, the complications, e.g., midpoints close to the X-ray midpoint, long recovery and eventual progression of cells into the selection delay times, etc. DHAQ alone produced a moderate depression window occurred later after the combi treatment than after X-ray alone. The midpoint division dela (the time between the midpoints of the descending and a nding portions of the 0 10mm m@2 @sg)ml DHAQ ‘EXPKl03@ mitotic rate curve) was 237 mm for t e combined treatment a 10mm 102 jig/mI DHAQ, 2.OGy X-ray' compared to 207 mm for X-ray only. The plateau region of mitotic rate following recovery from the division delay induced by DHAQ plus X-ray was higher than t mitotic rate following recovery from the delay induced by the i 0-mm pulse of DHAQ alone. This is similar to the ‘‘overshoot‘thatoccurs following w radiation-induced division delay, where the mitotic rate from 4 I an irradiated flask is greater than that fr m a control flask (see U) Chart 1). w a- The effect of various concentrations ol DHAQ on the midpoint U) division delay produced by radiation is shown in Chart 4. To -J -J compensate for variation between exr triments (not needed @ i04 within experiments) in the absolute valt i of the midpoint divi sion delay induced by 2.0 Gy X-ray, the values are expressed as relative to the delay produced by ‘rayonly within each experiment (average midpoint division d ay induced by 2.0 Gy X-ray, 204 mm). There was still consid able variation by this method as evidenced by the large error rs; however, in each experiment there was a consistent trend of increasing division I I I I delay with increasing concentration. Th re was no change in 0 100 200 300 400 500 the midpoint division delay induced by radiation if given in TIME AFTER INITIATIONOFTREATMENT,MINUTES combination with concentrations of 1O@ r i O@ @sgDHAOper Chart 3. Effect of DHAQ and radiation on the selection of mitotic cells. DHAQ (102 @g/ml)waspresent for the first 10 mm of treatment; the irradiated flasks ml. Above these concentrations, the div ion delay increased received 2.0 Gy X-ray at time 0. - - —-, response to X-ray alone. Midpoint of as a linear function of the logarithm of th concentration. Thus, progression blockade for DHAQ only, 78 mm; midpoint of progression blockade for DHAQ plus X-ray, 37 mm; midpoint division delay for DHAO plus X-ray, 237 irradiated cells that experienced a progr ssion blockade were mm. delayed for a longer period before they e$ntually appeared in

0 t%j 0 ,@

0

@ Chart 4. Effect of DHAQ concentration on midpoint dIvision de- .@ @ lay of nonirradiated and irradiated (2.0 Gy X-ray) cells. DHAQ was @ present for the first 10 mm of treatment. The midpoint divIsion delay is expressed as relative to the delay produced by 2.0 Gy X-ray @, alone (204 ±20 mm for 6 experiments). 0, treatment with DHAO @“ only; •,combinedtreatment with DHAQ plus X-ray. Numbers above ‘@ @ the bars (S.D.), number of determinations that contributed to the 0'- average. .@ .@ 0

44

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1980 American Association for Cancer Research. DHAQ-Radiation Division Delay

the selection window. Also shown in Chart 4 is the midpoint division delay for cells treated with DHAQ only. The slope of this line is similar to the slope of the line calculated for the X 400 ray plus DHAQ division delays. Chart 5 shows the effect of pulse treatments of various durations on the values of midpoint of progression blockade and midpoint division delay observed for either nonirradiated or irradiated cells. The location of the midpoint of progression 300 blockade for either DHAQ alone or DHAQ plus X-ray was independent of the duration of the DHAQ pulse, being 72 and 37 mm prior to selection, respectively. The plot of midpoint .@ @ division delay (expressed as relative to the delay produced by 200 2.0 Gy X-ray alone) for cells that received either DHAQ only or DHAQ plus X-ray shows a linear response as a function of the duration of DHAQ treatment. The data are consistent with the 2 lines possessing equal slopes and being merely displaced one from the other by a constant factor. 100 Chart 6 shows the effect of various doses of X-ray on the response of cells either nontreated or treated with DHAQ. The midpoint of progression blockade for either X-ray alone or X ray plus DHAQ was independent of the dose of X-ray received, 0 the value being 37 mm. Radiation-induced division delay as 0 1.0 2.0 3.0 4.0 measured by midpoint division delay is seen as a linear function X-ray dose, Gy of dose for both X-ray and X-ray plus DHAQ. Extrapolation Chart 6. Effect of X-ray dose on the midpoint of progression blockade and yields a displacement between the 2 responses of approxi midpoint division delay for nontreated and DHAO-treated cells. DHAO (102 @jg/ mately 40 mm at zero dose. For both conditions, the slope of ml) was presentfor the first 10 mm of treatment;Irradiationwas at time 0. 0. treatment with X-ray only; •,combinedtreatment with DHAQ plus X-ray. Numbers the line indicates a dose-dependent, radiation-induced division above bars (S.D.). number of determinations that contributed to the average. delay of approximately i hr/Gy.

Adriamycin (3, 5) and several other cancer chemotherapeutic DISCUSSION agents (2, 5). The location of the transition point in G2 is The action of DHAQ to induce a progression blockade of G2 concentration dependent, near the S/G2 boundary (located cells towards mitosis is qualitatively similar to that observed for approximately 120 mm prior to selection) at low concentra tions, and at the G2/M boundary (approximately 40 mm prior to selection) at high concentrations. A minimum transition point @@ ,II ExpKIO3 I—' is also produced that is located at or near the X-ray transition 1.6 . 0-2@tg/mlDHAQ±2.0GyX-ray point. This implies the completion of some process or structure at this time, 36 mm prior to selection, such that cells beyond ‘.4 S 0.20 Relative units that point are refractory to delay produced by either X-ray or ‘H 44 mm DHAQ and proceed into mitosis at the control rate. As sug @ l.a Slope 3.5 mln delay mm treatment gested previously (5), the linear nature of concentration-de pendent transition points past the last point in the cell cycle at which metabolic synthesis is required for progression and the existence of a common minimum transition point argue for a physical mechanism of action. This could occur by intercalation of the drug into DNA causing perturbation of the chromatin structure or the division apparatus. It is not surprising that _f'@ T2t9min(9) DHAQ would be a functional analog of Adriamycin since by at .@ least 2 criteria it is a structural analog. First, both Adriamycin t 4. 35mm and DHAQ are substituted anthraquinones. The planar 3-ring structure has been postulated to be the active structure that I V 3Tt4min(ll) confers cytotoxic properties to the molecule (8). Adriamycin possesses a glycosidic linkage whereas DHAQ is bisubstituted 0 10 20 30 60 with aminoalkylamino chains. Secondly, there is a particular Pulse duration, mInutes triangular arrangement of 2 oxygen atoms and one nitrogen Chart 5. Effect of DHAQ pulse duration on the midpoint of progression block atom that has also been implicated as the functional moiety of ade and midpoint division delay of nonirradiated and irradiated (2.0 Gy X-ray) cells. DHAQ (1o@ pg/mI) was presentfor the time indicatedat the initiationof various cytotoxic compounds (9). The only difference is that treatment; Irradiation was at time 0. The midpoint division delay is expressed as DHAQ possesses 2 such triangular arrangements whereas relative to the delay produced by 2.0 Gy X-ray alone (205 ±20 mm for 4 Adriamycin has only one. This has been used to explain the experiments). 0. treatment with DHAQ only; •,combinedtreatment with DHAQ plus X-ray. Numbers above bars (S.D.), number of determinations that contributed higher level of cytotoxic activity observed when DHAQ is corn totheaverage. pared to Adnarnycin.

JANUARY1980 45

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1980 American Association for Cancer Research. B. F. Kimler As with Adriamycin (3) and bleomycin (2), DHAQ does not is suggested that a prudent approach to c mbining DHAQ and alter the location of the radiation transition point. This is con radiation therapy for clinical use would tak@I into account the at sistent with the minimum transition point for radiation and least additive nature of the interaction of these 2 modalities. DHAQ (at high concentrations) being the same. Thus, any cells While effects on progression blockade do ‘iotprovethat there refractory to radiation-induced blockade are also refractory to will be a clinical problem as was produced )y radiation therapy DHAQ-induced blockade. The effect of DHAQ and radiation on and Adriamycin, the similarity between DH @tQand Adriamycin division delay appears to be additive. That is, the presence of and between their effects in vitro do sugg@st caution. There is DHAQ does not prevent the repair of that damage associated still an obvious need for studies to corn are the effects of with radiation-induced division delay. Consequently, once a DHAQ and radiation on the clonogenic pot rntial of mammalian cell recovers from DHAQ-induced division delay it will appear cells to the effects of Adriamycin and r @diation(4). These in the selection window, having already recovered from radia investigations are currently under way in this laboratory and tion-induced division delay. As seen in Charts 5 and 6, the will be reported at a later date. effect of 10_2 ,@gDHAQ per ml is to displace the value of midpoint division delay by a factor of approximately 40 mm. it ACKNOWLEDGMENTS is also observed that there is a rather constant difference The author would like to thank the staff of the Departr lent of Radiation Therapy between the location of the radiation transition point and the for thelr support. The technical assistance of Iris L. At rahams, Mary L. Barnes, @ transition point for that concentration of DHAQ. The additional and Patrick W. Gunn is acknowledged with gratitude Special thanks are ex 40 mm of delay could be explained by the reversion of those tended to Dr. C. C. Cheng for his introduction to DH @IQ and for many helpful conversations. cells located between the DHAQ and radiation transition points, i.e., those cells between 78 and 36 mm prior to selection, so REFERENCES that when they once again resume progression they do so from the DHAQ transition point. Thus, the transition point following 1. Chen, T. R. In situ detection of Mycoplasma cont mination in cell cultures combined treatment would be identical with the transition point by fluorescent Hoechst 33258 stain. Exp. Cell Re .,104:255—262,1977. @ 2. Kimler, B. F. Effect of bleomycin and radiation c G2 progression.Int.J. following radiation only, but the time of appearance of cells in Radiat. Oncol. Blol. Phys., in press, 1979. the selection window would be delayed by that additional 40 3. Kimler, B. F., and Leeper, D. B. The effect of Adu smycin and radiation on mm. A similar phenomenon has been observed with radiation G2progression. Cancer Res., 36: 3212—3216,19 ‘6. 4. Kimler, B. F.. and Leaper, D. B. Effect of adriamyc I and radiation on G3 cell and Adriamycin (3) but not with radiation and either actinorny survival. Int. J. Radiat. Oncol. Blol. Phys., in pres 1979. cm 0 or lucanthone (7). Overall, it appears that the action of 5. Kimler, B. F., Schnelderman, M. H., and Leaper, ( B.Inductionofconcen tration-dependent blockade in the G2 phase of @ecellcycle by cancer DHAQ and radiation is at best additive for this particular 40- chemotherapeutic agents. Cancer Res., 38: 809—114,1978. mm-wide region of cells. For long delays, i.e., those produced 6. Schneiderman, M. H., Dewey, W. C., Leaper, D. B@. and Nagasawa, H. Use by large X-ray doses, high concentrations of DHAQ, and long of the mitotic selection procedure for cell cycle i nalysis: comparison be tween the X-ray and cycloheximideG2markers. I xp. Call Res., 74: 430— pulse treatments, the effect may be subadditive. Overall, there 438, 1972. is no synergistic action to increase the duration of radiation 7. Tomasovic, S. P., and Dewey, W. C. Comparative studies of the effects of drugson X-ray-inducedG2delay. Radlat.Res., 74 . 112—128, 1978. induced division delay beyond this factor of 40 mm. 8. Zee-Cheng, R. K. V., and Cheng, C. C. Anti-neo@lastic agents. Structure This report has demonstrated the close similarity of Adria activity relationship study of bis(substituted amiuoalkylamlno) anthraqui mycin and DHAQ in regard to their effects on cell cycle pro nones. J. Med. Chem., 21: 291—294,1978. 9. Zee-Cheng, R. K. V., and Cheng, C. C. Comm4n receptor-complement gression blockades in the G2 phase and on the duration of feature among some antileukemic compounds. J. Pharm. Scm.,59: 1630— radiation-induced division delay. Based on this information, it 1634,1970.

46 CANCERFjESEARCHVOL.40

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1980 American Association for Cancer Research. Effect of Dihydroxyanthraquinone and Radiation on G2 Progression

Bruce F. Kimler

Cancer Res 1980;40:42-46.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/40/1/42

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/40/1/42. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1980 American Association for Cancer Research.