(CANCER RESEARCH 58, 2817-2824. July I. 1998 ......(CANCER RESEARCH 58, 2817-2824. July I. 1998] Explaining Differences in Sensitivity to Killing by Ionizing Radiation between Human

(CANCER RESEARCH 58, 2817-2824. July I. 1998]

Explaining Differences in Sensitivity to Killing by Ionizing Radiation between

Human Lymphoid Cell LinesDavid R. Aldridge and Ian R. Radford1

Sir Donald and Lady Trescowihick Research Laboratories, Peter MacCallum Cancer Institute, St. Andrews Place, East Melbourne, Victoria 3002, Australia

ABSTRACT

We surveyed five human hematopoietic cell lines (IISlt-2, MOI, I -4.Reh, CEM, and III,-60) to determine whether any simple correlates withsensitivity to killing by y-irradiation might be revealed. The clonogenicsurvival y-ray dose-response curves for these cell lines cover a wide range

of sensitivities. Consistent with previous results for murine hematopoieticcell lines, there was a clear correlation between the rapidity with whichirradiation induced apoptosis and clonogenic radiosensitivity of a cell line,although the relationship between timing of apoptosis and radiosensitivitydiffered between human and murine cell lines. Flow cytometric determination of cell cycle distribution after irradiation showed that differencesbetween human hematopoietic cell lines, in the rate of induction of apoptosis, were generally related to the functioning of cell cycle checkpoints.Whereas the rapidly dying and radiosensitive 1-1SB-2cell line underwent

apoptosis from different points in the cell cycle, the more slowly dying celllines showed a variety of cell cycle arrest profiles and initiated apoptosisafter accumulation of cells in the G2 phase. The lag-phase between arrestin ( ;, and induction of apoptosis was comparable for MOI ,1-4, Reh, andCEM; however. III,-6(1 cells showed a markedly longer G2 arrest that

correlated with their greater radioresistance. The results suggest that thetotal length of time available for DNA damage repair (irrespective ofwhether this time accrues as blockage in G,, S, or G2), prior to potentialactivation of apoptosis, is a critical determinant of radiosensitivity inhuman hematopoietic cell lines. Comparison of the p53 status of these celllines suggested that mutations in the 77*53 gene are contributing to the

delay of induction of apoptosis seen in the more radioresistant cell lines.The sensitivity of MOLT-4 and HL-60 cells to killing by DNA-associatedI25I decays was determined and was found to correlate with the relative

sensitivity of these lines to y-irradiation. The highly localized deposition ofenergy by I25I decays argues that DNA damage is a potent initiator of

apoptosis in these cell lines. The results presented suggest that differencesin the radiosensitivity of the cell lines examined reflect differences in therapidity of induction of apoptosis and that radiation-induced cell death in

hematopoietic cells can be explained as a response to DNA damage.

INTRODUCTION

Tumorigenesis frequently involves genetic changes that impair thecontrol of cell cycling and/or the regulation of apoptosis. Suchchanges can also alter radioresponsiveness, and there is now extensiveliterature on the involvement of genes such as p53, c-myc, and bcl-2

in determining the radiosensitivity of tumor cells. The impact onradiosensitivity of mutation or change in the level of expression ofsuch genes can, however, be variable and cell type dependent (reviewed in Ref. 1), presumably reflecting the complexity of the pathways that lead to cell death and the multiplicity of the regulatorybiochemical interactions that together determine the fate of an irradiated cell. We have examined different aspects of the radiation response of murine lymphoid and myeloid cell lines to determinewhether there are any simple, general correlates with relative radio-

Received 12/9/97; accepted 4/23/98.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1To whom requests for reprints should be addressed, at Research Division, PeterMacCallum Cancer Institute, Locked Bag No. 1, A'Beckett Street. Melbourne. Victoria

8006. Australia. Phone: 61-3-9656-1291; Fax: 61-3-9656-1411; E-mail: i.radford@

pmci.unimclb.cdu.au.

sensitivity. These studies suggested that sensitivity to clonogenickilling by y-rays or DNA-associated 12il decays correlated closely

with how rapidly apoptosis was induced after irradiation (2, 3). Suchdifferences in timing reflected whether induction occurred soon afterirradiation ("rapid interphase" death), following arrest in the G2 phase("delayed interphase" death), or after cell division ("mitotic" death).

Cell lines could show different proportions of death at these timesdependent upon the radiation dose they had received. It was speculated that DNA dsb2 directly initiates the destruction of cell lines

susceptible to rapid interphase death and that chromosomal aberrations can trigger delayed interphase or mitotic apoptosis (4). Theoccurrence of rapid interphase apoptosis appeared to be dependentupon the presence of wild-type p53 in these cells (5).

There have been relatively few comparative radiobiological studiesof human hematopoietic cell lines. The studies performed have soughtto correlate radiosensitivity with differences in capacity for sublethaldamage repair (6, 7). differences in lymphoid lineage maturation orvarious biochemical parameters (8), or differences in the rate of DNAdsb repair (9). However, aside from noting that the most radiosensitive lines were derived from immature lymphoid cells [we came to asimilar conclusion from studies with murine lymphoid lines (3)], thesestudies have been unable to explain the differences in radiosensitivityamong the cell lines examined.

The relationship between p53 status, the kinetics of induction ofapoptosis, and radiosensitivity has been studied in human lymphoidlines, although to a more limited extent than for mouse cell lines.Human studies have compared the TK6 (wild-type p53) and WI-L2-NS (mutant p53) cell lines, which were both derived from theBurkitt's lymphoma WI-L2 cell line. WTK1 (a clonal derivative of

WI-L2-NS, which is also a p53 mutant) showed delayed induction ofapoptosis and was correspondingly more resistant to killing by X-

irradiation than was TK6 (10, 11). Olive et al. (12) also showed acorrelation between slower induction of apoptosis and increased radioresistance in WI-L2-NS as compared with TK6. although, this

correlation did not hold when TK6 cells in different phases of the cellcycle were compared. More recent work has cast doubt on whether thedifferences seen in the WI-L2 system are due solely to the expression

of mutant p53 protein. Yu et al. (13) disrupted p53 function in theTK6 cell line by transfection with the human papillomavirus E6 geneand showed that this had a relatively small effect on the kinetics ofinduction of apoptosis and no effect on the clonogenic survivaldose-response for y-irradiation. The rate of induction of apoptosis inirradiated WI-L2-derived cell lines is, however, significantly slower

than is seen for many of the murine cell lines that we have examined.O'Connor et al. (14) examined a panel of Burkitt's lymphoma lines

and found that cells containing wild-type p53 were generally moresensitive to radiation-induced growth inhibition.

There are various lines of evidence that support the assumption thatdifferences in hematopoietic cell line radiosensitivity are related insome way to the cell's response to DNA damage. For example.

Warters (15) demonstrated that bromodeoxyuridine labeling enhancedthe sensitivity to X-ray-induced killing of a murine T-cell hybridoma.

2 The abbreviations used are: dsb. double strand break; IL, interleukin; BrdUrd,

bromodeoxyuridine; ECL. enhanced chemiluminescence.

2817

on July 6, 2021. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

RADIOSENSmVITY OF HUMAN LYMPHOID CELLS

and we have shown that murine lymphoid cell lines can be exquisitelysensitive to killing by DNA-associated 125Idecays and that, in gen

eral, their sensitivity to such radioactive decays correlates closely withtheir sensitivity to external beam irradiation (2, 3). The DNA-centric

view has, however, been challenged by investigators who assert acrucial role for ceramide production, following activation of sphin-

gomyelinase at cell membranes, in the induction of apoptosis byirradiation (16). Consistent with the latter hypothesis, it was shownthat acid sphingomyelinase-deficient human or murine lymphoid cellsshowed reduced levels of radiation-induced apoptosis (17). Recently,

a correlation between the level of postirradiation ceramide productionand the relative radiosensitivity of a panel of human lymphoid celllines has been adduced as further support for the importance ofmembrane damage and for the inferred irrelevancy of DNA damage(18).

Accordingly, we wished to determine whether the correlations,found using murine cell lines, between external beam radiosensitivityand the rate of induction of apoptosis and between sensitivity tokilling by external beam radiation and DNA-associated 125I decays

are also observed in human hematopoietic cell lines that show a broadrange of radioresponsiveness. For these purposes, we have used apanel of four human lymphoid (CEM, HSB-2, MOLT-4, and Reh) andone myeloid (HL-60) cell line.

MATERIALS AND METHODS

Cell Lines. All cell lines were obtained from the American Type CultureCollection. Cells were grown in the a modification of Eagle's medium (ICN)

plus 10% PCS (CSL, Melbourne, Australia) at 37Â°C,using sealed flasks that

had been flushed with 5% CO2, 5% O2, and balance N2. All studies wereperformed using asynchronous, log-phase cultures. The cell lines were rou

tinely checked for Mycoplasma contamination using a PCR method (19). Theorigin and some of the characteristic features of each of the cell lines used arepresented in Table 1.

y-Irradiation, 125IIncorporation, Clonogenic Survival Assay, and Sta

tistical Treatment of Survival Data. Cell cultures were irradiated at roomtemperature in a '37Cs -y-ray source at a dose rate of approximately 0.9 Gy/min.

I25I labeling was performed by incubating cells for greater than 1.5 popu

lation doubling times in growth medium containing 50-100 Bq/ml of[125I]iododeoxyuridine (NEN/DuPont) and 2.5 Â¿IMthymidine. Cells were then

washed, and unincorporated label was chased by incubation in growth mediumcontaining 20 /UMthymidine and 20 /J.Mdeoxycytidine for 3 h. After chasing,incorporated '~5I was measured by pelleting the cells and counting the level of

radioactivity in a Compugamma CS (LKB) gamma counter. Counts werecorrected to decays per minute using the method of Eldridge (20). Cell pelletswere then resuspended in a known volume of growth medium, aliquots weretaken, and cell number was determined by Coulter counting. The mean numberof I25I decays occurring per cell, per day was then calculated. After slow

addition of an equal volume of growth medium plus 20% DMSO, the cellsuspension was aliquoted, frozen at -PC/min in a controlled-rate freezer

[Cryologic (Melbourne) model CL3000 with Cryogenesis V3 software] toâ€”¿�60Â°Cand then transferred to liquid nitrogen for storage. Cell aliquots were

removed from liquid nitrogen storage at various times, after known numbers ofI25I decays per cell had occurred, and assayed for clonogenic survival. Accu-

Table 1 General properties of human hematopoielic cell lines

CelllineHSB-2

MOLT-4RehCEMHL-60OriginPre-T

Pre-TEarly BTPre-myelocyleKaryotype"Pseudodiploid

PseudotetraploidPseudodiploidPseudodiploidPseudodiploidDoubling

time(h)29

23212233Cloning

efficiency(% Â±SE)15

Â±363 Â±741 Â±348 Â±850 Â±5TP53

status"Wild

typeWild type/mutant^Mutant

Deleted" Data from American Type Culture Collection catalog.* Data from other studies (see "Results" for references).

mulated decays at day "t" (At) were determined using the equationA, = 86.56/l0(l - i'-0-01155Â»),where Aa represents the number of decays per

cell per day at day 0.Survival assays were performed using single-cell suspensions that were

obtained by gently pipetting the cultures approximately 10 times through a5-ml glass pipette. Dilutions of this cell suspension were then counted using aCoulter counter. For each of the cell lines except HSB-2, clonogenic survival

was determined by plating cells in growth medium plus 0.3% Noble agar(Difco) and 0.4% (packed cell volume) sheep RBCs (Amadeus International,Melbourne, Australia). After 14 days, plates were fixed and stained by theaddition of 0.01% crystal violet in 1.5% acetic acid. Colonies containing >50cells were then scored. The HSB-2 cell line could not be cloned in soft agar.

Consequently, the survival of this cell line was assayed by plating cells ingrowth medium in 96-well, U-bottomed microtiter plates (Greiner) and check

ing for growth after 14 days. Microtiter plate wells were scored as positive ifthey contained greater than 50 cells. Cell numbers added per well wereadjusted to give â€”¿�50%positive wells. In control platings, this corresponded to

around three cells per well. The cloning efficiency of HSB-2 cells was

significantly improved by the addition of 10% pooled human AB serum and5% conditioned medium from the IL-2-secreting gibbon leukaemic cell line

MLA 144 (data not shown). Although Guizani et al. (21) demonstrated thatHSB-2 does not constitutively express the IL-2 receptor, growth factors present

in either the human serum or the conditioned media may be inducing expression of IL-2 receptor. Alternatively, other factors secreted by MLA-144 cellsmay be contributing to the improved growth. Results for HSB-2 cells werecalculated using the equation Ce = [-ln(AO]/Cw, where cloning efficiency (Ce)

is equal to the negative natural log of the fraction of wells without cell growth(N) divided by the number of cells plated per well (Cw). Clonogenic -y-ray

dose-response survival curves for CEM cells plated in either soft agar or

microtiter wells showed no significant difference (data not shown).Survival data were fitted to equations using a computer-generated least-

squares fit. The fit was minimized according to the criterion of Â£Â¡(0â̂€”¿�Â£Â¡)2/VÂ¡,

where O is observed value, E is calculated value, and V is variance. Thestatistical errors associated with the survival data were calculated, after themethod proposed by Boag (22), by assuming a simple Poisson variance on thetotal number of control and test colonies counted.

Determination of Apoptotic Morphology. Cell samples, taken at intervalsafter irradiation, were pelleted by centrifugation at 1000 rpm, gently resuspended in a small volume of residual growth medium, and then mixed with anequal volume of growth medium containing 10 fig/ml of ethidium bromide and3 p.g/ml of acridine orange. At least 500 cells were then scored for apoptoticnuclear morphology under a fluorescence microscope.

Cell Cycle Distribution. Cell cultures were labeled with 10 JXMBrdUrd for30 min and then analyzed using the anti-BrdUrd antibody technique (23).

Cellular DNA content was determined by staining samples with propidiumiodide. Samples were processed using a Becton Dickinson FACStar Plus flowcytometer, and the data were analyzed using the WinMDI version 2.5 program.Cells were counted as apoptotic if they had a sub-G, or sub-G2 DNA content

and lacked BrdUrd incorporation. This assignment was checked by cell sortingand microscopic examination. Levels of apoptosis determined by flow cytom-

etry were generally lower than values obtained by scoring morphology (presumably due to overlap between the DNA content of apoptotic and G, or G2cells); however, the two methods gave qualitatively similar results (compareFigs. 2 and 4).

Western Blotting. Protein was extracted from each cell line by lysing IO7

cells in 0.5 ml of 50 mm Tris (pH 7.5) plus 2% SDS. To reduce sampleviscosity, DNA was sheared by 20 passages through an 18-gauge needle.

Samples were then boiled for 3 min, cooled on ice, and spun for 5 min at10,000 rpm in a microcentrifuge to remove insoluble material. Supematantswere assayed for protein concentration using bicinchoninic acid and thenstored at â€”¿�70Â°C.Prior to electrophoresis, buffer containing bromphenol blue

was added to give a final concentration of 10% glycerol, 10 HIMDTT, and 50mM Tris (pH 7.5). Samples were boiled for 3 min and allowed to coolimmediately prior to loading.

Equal amounts of total protein (100 /j,g unless stated otherwise) wereseparated on a 12.5% SDS-PAGE gel and then transferred to a polyvinylidene

difluoride membrane (Millipore) for 1 h at 50 V in transfer buffer containing25 mM Tris base, 192 mM glycine, and 20% methanol at pH 8.3 (24). Themembrane was then allowed to dry overnight at room temperature, stained for

2818




(0>3CO

.1

I .010)Â§>o .001o

.0001

Dose (Gy)

Fig. I. Clonogenic survival of different human hemalopoietic cell lines as a functionof -y-ray dose. Data were fitted to the equation S = exp - [aD + ÃŸD2].where S is

survival, D is dose, and a and ÃŸare constants. Calculated values of a and ÃŸ:HSB-2(1.75 Â±0.06 and 0.087 Â±0.029); MOLT-4 (1.49 Â±0.05 and 0.064 Â±0.018); Reh(1.06 Â±0.01 and 0.005 Â±0.005); CEM (0.61 Â±0.03 and 0.081 Â±0.006); and HL-60(0.62 Â±0.02 and 0.029 Â±0.004). Each data set is from at least two separate experiments.

total protein with Ponceau-S (Sigma), and blocked with 3% skim milk powder,

2% gelatin, and 0.1% Tween 20 in PBS (pH 7.4). After incubation for 2 h atroom temperature, fresh blocking solution containing the primary antibody(AC 21 (Santa Cruz Biotechnology) for Bcl-2 or PAb 421 for p53) was added,and the membrane was incubated overnight at 4Â°C.The membrane was then

washed three times in PBS plus 0.2% Tween 20, followed by incubation withperoxidase-labeled secondary antibody for l h at room temperature. Afterfour X 15-min washes in PBS plus 0.2% Tween 20 at room temperature, the

membrane was incubated for 1 min in ECL buffer [0.4 mg/ml luminol, 0.1mg/ml 4-iodophenol, and 1 /Â¿I/mlH2O2 (36%) in 60 mM Tris buffer, pH 7.5],and proteins were then visualized on Fuji Rx medical X-ray film using

exposure times varying from 30 s to 5 min. ECL provides qualitative, but notquantitative, information about protein levels.

tion are presented in Fig. 1. The a and ÃŸvalues derived from fittingthe data to a linear-quadratic function of dose are given in the figurelegend. The results show clearly that HSB-2 and HL-60 are. respec

tively, the most radiosensitive and the most radioresistant of the celllines surveyed, with approximately a 200-fold difference between thelines in the level of survival after a dose of 4 Gy. MOLT-4 cells show

a comparable, although somewhat more radioresistant, response tothat of HSB-2. Reh and CEM are considerably more radioresistantthan HSB-2 and MOLT-4. Both show similar clonogenic dose re

sponses, although CEM is significantly more radioresistant than Rehat lower radiation doses. Our dose responses for MOLT-4 and HL-60

are comparable with those of previous studies (7, 27).Rate of Induction of Cell Death after y-Irradiation Correlates

with Radiosensitivity. Based on the y-ray survival dose-responses

presented in Fig. 1. cultures of the different cell lines were givenequitoxic radiation doses sufficient to reduce clonogenic survival to0.5% of the control value and were then incubated at 37Â°C.The timing

of occurrence of cell death was then determined in culture samplestaken at various times after irradiation. All five cell lines showedinduction of apoptotic nuclear morphology after irradiation. HL-60

cultures showed some necrotic death, although apoptosis was thepredominant form of death over the time interval examined (data notshown). As shown in Fig. 2, the rate of induction of apoptosis afterequitoxic doses of y-irradiation varied greatly between the five cell

lines examined, and the relative ordering of these responses reflectedthe clonogenic survival dose responses. For example, HSB-2 is the

most radiosensitive of the cell lines examined and also shows the mostrapid induction of apoptosis. The more radioresistant MOLT-4 cellline shows a pronounced delay of â€”¿�4-6h before commencement ofinduction of apoptosis at a similar rate to that of HSB-2. Reh culturesare more radioresistant than MOLT-4 and show a somewhat longer

delay before initiating apoptosis. The subsequent entry of Reh cellsinto apoptosis occurs at a much slower rate than is seen for MOLT-4or HSB-2. CEM cells are more resistant to y-irradiation than Reh and

show a significantly longer delay prior to the induction of apoptosis.Once apoptosis has begun in irradiated CEM cultures, the rate ofaccumulation of apoptotic cells appears to be similar to that of Reh.

RESULTS

Optimizing Clonogenic Assays. Our initial studies suggested that,when plated in growth medium plus semi-solid agar, the human cell

lines chosen generally had low cloning efficiencies that showed a highlevel of inter-experimental variability. This result was of concern

given claims that, at least for some cell lines, there is a correlationbetween cloning efficiency and radiosensitivity (25). The controlcloning efficiencies of CEM, HL-60, MOLT-4, and Reh were, how

ever, found to be greatly increased by the addition of sheep RBCs tothe medium. For example, the cloning efficiency of HL-60 increasedfrom 1% to ~50% in the presence of sheep RBCs. The ability of

sheep RBCs to increase the cloning efficiency of hematopoietic cellswas first demonstrated by Bradley et al. (26). Even with the additionof conditioned media, sheep RBCs, or lethally irradiated feeder cells,we were unable to clone HSB-2 cells in soft agar. This cell line was,

however, found to be clonable in microtiter plates as described in"Materials and Methods." Our inability to clone HSB-2 cells in soft

agar may indicate a requirement for cell-to-cell contact or that some

aspect of the cloning methodology is lethal to this cell line. Controlcloning efficiencies, obtained for each of the cell lines using the abovemethods, are shown in the Table. The values obtained are, in general,significantly greater than those reported in previous studies.

Clonogenic Survival y-Irradiation Dose Responses. Dose-responses for the five human hematopoietic cell lines under investiga-

oOu

oo.

o"'S

SSIL

1.0

0.8

0.6

0.2

10 20

Time (h)

30 40

Fig. 2. Rate of induction of apoptosis. as determined by nuclear morphology, inâ€¢¿�y-irradiatedhuman lymphoid cell lines. Cell cultures were given equitoxic doses (HSB-2.3 Gy; MOLT-4. 3.5 Gy; Reh, 5 Gy; CEM. 5 Gy; and HL-60. 6.5 Gy) and then incubatedat 37Â°C.The response of the murine ST4 lymphoid cell line after 3 Gy of -y-irradialion is

included for comparison [data from Radford et al. (36)|. Each data set was pooled fromat least two separate experiments.

2819




.1

.01

HL-60(31.211.9)

MOLT-4(22.110.6)

20 40

125

60 80

I decays/cell

100

Fig. 3. Clonogenic survival responses of MOLT-4 and HL-60 cells exposed toDNA-associated 125Idecays. Data were fitted to the equation S = exp - [DIDâ€ž\,where

S is survival, D is dose, and Ã¶â€žis a constant. Calculated values for D0 and its associatederror are shown on the figure. Each data set is from two separate experiments.

The apparent decline in the numbers of apoptotic CEM cells at â€”¿�20

h after irradiation is an artifact caused by the fragility of apoptoticCEM cells (i.e., the true kinetics of appearance of apoptotic CEM cellsare probably more comparable with those of Reh than to HL-60).HL-60 is clearly the most radioresistant cell line examined and also

shows the most pronounced delay prior to the induction of apoptosis.Accumulation of apoptotic cells was not observed in HL-60 cultures

until at least 24 h after irradiation. It is of interest to note that even themost rapidly dying of the human cell lines (HSB-2) shows markedlyslower kinetics of induction of apoptosis than does a murine lym-

phoma such as ST4 (Fig. 2).Sensitivity to Killing by DNA-associated 125I Decays. Because

of the continuing debate over the importance of DNA damage ininitiating radiation-induced apoptotic cell death and because the levelof DNA damage induced per unit dose of y-irradiation can vary

between cell lines (28), we were interested in comparing the sensitivity of our panel of cell lines to killing by DNA-associated I25I

decays.The technique used for I25I decay experiments involves freezing

cells and then storing them over liquid nitrogen for periods up to 2-3

months to allow the accumulation of DNA damage without interference from DNA repair activities. The cells must then show reasonablecloning efficiencies after thawing. Our attempts to determine thesensitivity of these cell lines to DNA-associated 125I decays were,

however, generally unsuccessful due to low and variable cloningefficiencies after freezing and thawing. MOLT-4 and HL-60 were the

only lines to show reasonable cloning efficiencies after freezing andthawing (30 Â±9% for MOLT-4 and 6.7 Â±0.4% for HL-60). The

clonogenic survival curves for these cell lines, after accumulation of125I-induced DNA damage, are shown in Fig. 3. As for the y-ray

survival results, HL-60 cells (D0 = 38.2 Â±1.9 decays) were found tobe far more resistant to 125I decay-induced DNA damage than are

MOLT-4 cells (D0 = 22.1 Â±0.6 decays).Effect of y-Irradiation on Cell Cycling. Studying the effect of

y-irradiation on subsequent cell cycling can indicate whether induc

tion of apoptosis is related to cell cycle events (e.g., cell cycleblockage or mitosis) and how differences in radiosensitivity might berelated to these events (4). Cell cultures were pulse labeled with

BrdUrd at intervals after exposure to equitoxic y-ray doses (sufficient

to reduce clonogenic survival to 0.5% of the control value) and werethen analyzed by flow cytometry. Cell cycle distributions, as a function of time after irradiation, for each of the five cell lines are shownin Fig. 4.

Flow cytometric analysis of irradiated HSB-2 cultures showed no

marked accumulation of cells in any phase of the cell cycle prior to anincrease in the levels of apoptosis. Irradiation has rapid and profoundeffects on these cells as evidenced by an almost total suppression ofmitotic activity after 3 Gy of y-irradiation and a marked decrease inBrdUrd incorporation into S-phase cells within 2 h after irradiation

(data not shown). Flow cytometric profiles of cultures that had beenincubated for 4 or more hours after irradiation, showed a significantincrease in the level of cells with sub-G, or intermediate to G, and G2

(but lacking BrdUrd incorporation) DNA content (Fig. 5), suggestingthat the induction of apoptosis in HSB-2 cells can occur at different

points in the cell cycle and is not linked to one checkpoint. In contrastto the results obtained for HSB-2, the other more rapidly dying cellline, MOLT-4, shows an accumulation of cells in G2 that parallels andprecedes the induction of apoptosis. Consistent with MOLT-4 cells

undergoing apoptotic death after blockage in G2, flow cytometricprofiles showed a progressive increase, with time after irradiation, inthe proportion of cells showing a DNA content intermediate to G, andG2 (but lacking BrdUrd incorporation; Fig. 5). Similar to MOLT-4,

the increase in the fraction of apoptotic Reh cells follows and occursin parallel with the accumulation of cells in the G2 phase. Thisaccumulation occurs more slowly than is seen for MOLT-4, appar

ently due to a more prolonged blockage of Reh cells in Gl and/or S.Consistent with induction of apoptosis in G2, flow cytometric profilesfrom irradiated Reh cultures undergoing apoptosis showed increasedlevels of cells with a DNA content intermediate to G, and G2 (butlacking BrdUrd incorporation; see Fig. 5). Although Reh and CEMhave comparable sensitivities to killing by y-irradiation, their cell

cycle responses are significantly different (Fig. 4). Unlike Reh, CEMcells show no postirradiation blockage in the G, phase, but they doshow a pronounced blockage in mid- to late-S phase that is followed

by a rapid exit from S and accumulation in the G2 phase. As withMOLT-4 and Reh, the latter event precedes the appearance of apop

totic cells in culture. Consistent with the death of CEM cells in G2phase, flow cytometric profiles of apoptosing cultures showed increased levels of cells with a DNA content intermediate to G( and G2(but lacking BrdUrd incorporation; data not shown). HL-60, the most

radioresistant and the slowest dying of the cell lines in our panel, lacksa marked G, arrest but shows prolonged blockage of cells in G2 priorto the appearance of apoptosis in the culture (Fig. 4). The timing ofoccurrence of apoptosis in irradiated HL-60 cells also coincides with

the resumption of mitotic activity in these cultures (data not shown).This result and the flow cytometric data suggest that death of thesecells occurs both in G2 and during or around the time of mitosis.

p53 and Bcl-2 Status. The p53 protein is an important modulator

of the cell cycle and a determinant of radiosensitivity in lymphoidcells. Accordingly, we sought to correlate the status and expression ofthe TP53 gene with the radiosensitivity and cell cycle response of thecell lines examined. The TP53 gene of the HSB-2, MOLT-4, CEM,and HL-60 cell lines has been sequenced by other investigators, and

the results are summarized in Table 1. Similar sequence data for Rehcells could not be found in the literature. We tested for the presenceof p53 protein in each of the cell lines by Western blotting. The blotswere probed with the PAb 421 monoclonal antibody, which is specificfor an epitope in the COOH-terminal region of the protein (29), and

reactive proteins were detected using ECL. Expression of p53 proteinwas found in all of the cell lines, except for HL-60 (Fig. 6). The latterresult is consistent with the finding that HL-60 has a large deletion

820




1.0

Fig. 4. Effect of -y-irradiation on cell cycle progression

of human lymphoid lines. The distribution of cells between the different phases of the cycle (G,, S. and G2)and the level of apoptosis (Api were determined by flowcytometry using cultures given equitoxic doses and thenincubated at 37Â°C.Each data set was pooled from two

separate experiments.

o -

0 2 4 602468 10 0

Time (h) Time (h)

8 12 16 20 24

Time (h)

1.0

.2 .8

CL

.4

8.2

CEM (5 Gy) HL-60 (6.5 Gy)

4 8 12 16 20 240

Time (h)

10 15 20 25 30 35

Time (h)

within the TP53 gene that prevents protein expression (30). The twomost rapidly dying and radiosensitive cell lines, HSB-2 and MOLT-4,have both been shown to express wild-type p53 (31, 32). In additionto a wild-type gene, the MOLT-4 cell line also contains a TP53

mutation that leads to aberrant splicing of the mRNA, causing a frameshift and truncating the protein before translation of exons 10 and 11(32). Of the more slowly dying cell lines, none are known to expresswild-type p53. Both copies of the TP53 gene in CEM cells contain

single base substitutions that produce missense mutations (31), and asmentioned above, the HL-60 cell line does not express p53 protein due

to gene deletions. Reh does express p53 protein (Fig. 6): however, wedo not know whether this protein is mutant or wild type.

Members of the Bcl-2 family of proteins are important modulators

of apoptosis and have been shown to influence radiosensitivity (33).Western blotting, using a polyclonal antibody to Bcl-2 protein,

showed that all of the cell lines used in this study expressed a reactiveprotein of Mr -26,000 (Fig. 6). Reh and HL-60 differed from the

other lines in showing an additional reactive band of slightly highermolecular weight. The additional band might represent expression ofa Bcl-2-related protein, such as Bcl-x,. which has a molecular weight

of 29.000 (34), or a phosphorylated form of the protein. A variant ofHL-60 has been shown to express Bcl-xL (35).

DISCUSSION

We have examined whether the correlation between the rate ofinduction of apoptosis and the clonogenic survival y-ray dose-

response, observed for murine hematopoietic cell lines, is also seen inhuman hematopoietic cell lines. The results presented in this reportclearly demonstrate, for the cell lines examined, a correlation betweenthe rapidity with which apoptosis is induced after irradiation and theclonogenic survival dose-response. For example, the HSB-2 cell line,which shows the most rapid induction of apoptosis after y-irradiation,

was also found to be the cell line most sensitive to clonogenic killing,whereas MOLT-4, Reh, CEM, and HL-60, which show progressively

longer delays prior to the induction of apoptosis after an equitoxicradiation dose, are correspondingly more resistant to killing byy-irradiation.

Although a general correlation between the relative rate of induction of apoptosis and the clonogenic survival y-ray dose-response is

seen for both human and murine hematopoietic cell lines, one notabledifference is that although the survival dose-response curves for the

more radiosensitive human and murine cell lines are similar, equitoxicradiation doses induce apoptosis far more rapidly in the murine lines.For example, we found that at least 90% of ST4 cells (a murine pre-T

2821



RADIOSENSÃ•TMTY OF HUMAN LYMPHOID CELLS

HSB-2 (Control) HSB-2 (3 Gy + 5 hrs)

uO

O3E'

2mO

MOLT-4 (Control) 2

Reh (Control)

O

mO

MOLT-4 (3.5 Gy + 10 hrs)

O

Reh (5 Gy + 24 hrs)

DNA ContentFig. 5. Dot-pio! distributions for BrdUrd incorporation (ordinale) versus DNA content

(abscissa) of cells from control and irradiated MOLT-4 and Reh cultures. Regionscorresponding to cells in G, and G2 phases of the cell cycle are indicated.

cell line), exposed to a y-ray dose sufficient to reduce clonogenic

survival to 0.5%, showed apoptotic morphology 3 h after irradiation(36). In contrast, the most rapidly dying human cell line used in thisstudy (HSB-2) does not reach this level of apoptosis until at least 24 hafter receiving an equitoxic y-ray dose, despite having a similarclonogenic survival y-ray dose-response to ST4. Unfortunately, because we were unable to determine the sensitivity of HSB-2 cells tokilling by DNA-associated 125Idecays, we cannot rigorously compare

the response to DNA damage of ST4 and HSB-2. If we assume thatthere are similar y-ray dose-responses for DNA lesion induction in the

two cell lines, then perhaps the simplest explanation for the differencein timing is that, after the detection of lethal cellular injury, thebiochemical events that lead to the triggering of apoptosis occur muchmore slowly in HSB-2 than in a murine line such as ST4. Two factors

that may be relevant to this human/murine difference are: (a) there isa marked difference in the population doubling times of cell lines suchas ST4 and HSB-2 (9 h versus 29 h). The data of Olive et al. (12),

however, suggest that human TK6 cells have a relatively rapid doubling time (around 12 h) and yet show induction of apoptosis afterirradiation at a comparable rate to HSB-2 cells; and (b) the human cell

lines that were used in this study all derived from tumors that relapsedafter intensive radiotherapy and/or chemotherapy of the primary tumor, whereas the derivation of murine lines like ST4 did not involveexposure to cytotoxic agents. Such treatment may well have selectedfor cells with down-regulated apoptotic responses, and it is perhaps

relevant that Bcl-2 protein is readily detectable in the human cell lines

that we have used.Irradiation of cells can induce blockage at checkpoints in different

phases of the cell cycle, including p53-dependent arrest in G, (37) andp53-independent blocks in S (38) and G2 (39). These blocks areultimately caused by inhibition of cyclin-dependent kinases that are

necessary for cycle progression. In the cell lines that we have examined, the rate of induction of apoptosis, after y-irradiation, generally

appears to be related to the functioning of these cell cycle checkpoints.HSB-2 may be an exception in this regard, because it shows no clear

evidence for accumulation at checkpoints and appears to undergoinduction of apoptosis in different phases of the cell cycle. Consistentwith previous observations that the more rapidly dying and radiosensitive murine lymphoid cell lines contain wild-type p53 (5), HSB-2encodes wild-type TP53 (31), and we have shown that it expressesp53 protein. It may also be relevant to the radiosensitivity of HSB-2

that, due to homozygous deletion of the Rb gene (40), this cell linedoes not express pRb. Expression of pRb was associated with p53-independent protection against radiation-induced apoptosis in both a

human osteosarcoma and a lymphocytic cell line (41, 42). WhenMOLT-4 or Reh cells were exposed to a y-ray dose that reducessurvival to 0.5% (unlike HSB-2), both showed induction of apoptosis

after accumulation of cells in G-,. Due apparently to blockage earlierin the cycle, Reh cells accumulated in G2 much more slowly than didMOLT-4 cells and showed a corresponding increase in the time

between irradiation and the appearance of apoptotic cells that correlated with their greater radioresistance. The molecular basis for thisdifference is unclear. Both MOLT-4 and Reh express pRb (40, 43)and, as shown in this study, p53. Although MOLT-4 cells express bothwild-type and a truncated p53 (32), the status of the p53 expressed by

Reh cells is not known to us. The relative slowness of the inductionof apoptosis in Reh cells suggests that they contain mutant p53. LikeMOLT-4 and Reh, irradiated CEM cells appeared to die after block

age in G2. CEM cells differed, however, in showing a prominentblockage in mid- to late-S phase that preceded accumulation in G2.

Consistent with their slow kinetics of induction of apoptosis and theirradioresistance, CEM cells express pRb (40) and mutant p53 (31).Finally, the response of HL-60 to a y-ray dose that reduces survival to

0.5% is characterized by transient arrest in S and a marked andprolonged blockage in G2. Other authors have also described radiation-induced G-, arrest and apoptosis of HL-60 cells (39, 44). Although MOLT-4, Reh, and CEM cultures showed an increase in the

level of apoptosis that lagged about 4 h behind the accumulation ofcells in G2, our results suggest that HL-60 cells can remain arrested in

G2 for much longer times before undergoing apoptosis. The basis forthis difference has not been examined, although in other cell typesexpression of a ras oncogene was associated with a decrease in cyclinBl mRNA levels and a lengthening of G2 arrest (45).

HSB-2 MOLT-4 Reh CEM HL-60

p53

bcl-2

Fig. 6. Western blotting analysis of p53 and Bcl-2 protein expression in humanhematopoietic cell lines.

2822



RADIOSENSITIVE OF HUMAN LYMPHOID CELLS

The importance of the duration of postirradiation arrest at the G,and G2 checkpoints in determining radiosensitivity has been tested invarious nonlymphoid cell types that undergo apoptosis. There isevidence to suggest that although the p53-dependent G, checkpoint

may not be critical in this regard (46), the length of the G2 delay maybe directly related to radiosensitivity (45). Consistent with our previous results for murine hematopoietic cell lines (4), the data presentedin this paper suggest that the total length of time available for DNAdamage repair (irrespective of whether this time accrues as blockagein G|, S, or G-,), prior to potential activation of apoptosis, is a critical

determinant of radiosensitivity in human hematopoietic cell lines.We have assumed that the explanation for the differences in radio-

sensitivity between hematopoietic cell lines lies in understanding theresponse of the cell to DNA damage. Consistent with this view, wehave shown in this report a correlation between the sensitivity ofMOLT-4 and HL-60 to y-irradiation and to DNA-associated 125Idecays. It is interesting to note that the D0 values for I25I decay-

induced cell killing of MOLT-4 and HL-60 (22.1 Â± 0.6 and

38.2 Â±1.9 decays, respectively) are very similar to the values for themurine lines WEHI-7 and 18-81 (22.4 Â±1.3 and 40.8 Â±2.8 decays,respectively; Ref. 3). Both MOLT-4 and WEHI-7 cells show postir-

radiation accumulation in and subsequent apoptosis in G2 phase,whereas HL-60 and 18-81 both exhibit a very delayed induction ofapoptosis (4). I25I decays have been shown to deposit the majority of

their energy in close proximity to the site of decay (47) and, whenoccurring in DNA, induce dsb with close to 100% efficiency (48).Accordingly, the 125Iresults (presented in this and previous studies)

suggest that radiation-induced DNA dsb is a potent initiator of apop

tosis, and they are a strong indicator that these lesions are alsoresponsible for y-ray-induced apoptosis. An opposing view, that radiation-induced apoptosis is triggered by membrane damage-initiated

production of ceramide, has been supported recently by data showinglower postirradiation production of ceramide in HL-60 than in a moreradiosensitive Burkitt's lymphoma line (18). The latter study did not,

however, compare the ceramide responses of the cell lines at equitoxicradiation doses; in addition, it has been suggested that the ceramideassay method used by Michael et al. (18) gives artifactual results (49).

We have shown that differences in the radiosensitivity of a panel ofhuman hematopoietic cell lines can be related to differences in therapidity of induction of apoptosis and that radiation-induced apoptosis

can be explained as a response to DNA damage.

REFERENCES

10.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

1. Olive. P. L.. and Durand, R.E. Apoptosis: an indicator of radiosensitivity in vitro? Int. 3 jJ. Radial. Biol.. 71: 695-707, 1997.

2. Radford. I. R. Mouse lymphoma cells that undergo interphase death show markedly 32increased sensitivity to radiation-induced DNA double-strand breakage as comparedwith cells that undergo mitotic death. Int. J. RadiÃ¢t.Biol., 59: 1353-1369, 1991.

3. Radford. I. R. Radiation response of mouse lymphoid and myeloid cell lines. Part I. 33Sensitivity to killing by ionizing radiation, rate of loss of viability, and cell type oforigin. Int. J. Radial. Biol.. 65: 203-215. 1994.

4. Radford, I. R., and Murphy. T. K. Radiation response of mouse lymphoid and 34myeloid cell lines. Part III. Different signals can lead to apoptosis and may influencesensitivity to killing by DNA double-strand breakage. Int. J. RadiÃ¢t. Biol.. 65:229-239, 1994. 35.

5. Radford. I. R. p53 status, DNA double-strand break repair proficiency, and radiationresponse of mouse lymphoid and myeloid cell lines. Ini. J. Radial. Biol.. 66: 557-560.

1994.6. FitzGerald. T. J., McKenna. M., Kase. K.. Daugherty, C.. Rothstein. L.. and 36.

Greenberger. J. S. Effect of X-irradiation dose rate on the clonagenic survival ofhuman and experimental animal hematopoietic tumor cell lines: evidence for heterogeneity. Int. J. Radial. Oncol. Biol. Phys., 12: 69-73, 1986. 37.

7. Lehnen. S.. Rybka. W. B.. Suissa. S., and Giambaltisto. D. Radialion response ofhaematopoietic cell lines of human origin. Ini. J. RadiÃ¢t.Biol.. 49: 423-431. 1986.

8. Uckun. F. M.. Mitchell. J. B., Obuz. V.. Park. C. H., Waddick, K.. Friedman, N.. 38.Oubaha. L.. Min, W. S., and Song. C. W. Radiation sensitivity of human B-lineagelymphoid precursor cells. Int. J. Radial. Oncol. Biol. Phys., 21: 1553-1560, 1991.

2823

Evans. H. H.. Ricanati. M.. Horng. M. F.. Jiang. Q.. Menci. J.. and Olive. P. DNAdouble-strand break rejoining deficiency in TK6 and other human B-lymphoblast celllines. RadiÃ¢t.Res.. 134: 307-315. 1993.

Amundson, S. A., Xia. F.. Wolfson. K.. and Liber. H. L. Different cytotoxic andtnutagenic responses induced by X-rays in two human lymphoblastoid cell linesderived from a single donor. MutÃ¢t.Res.. 286: 233-241. 1993.Xia. F., Wang, X.. Wang. Y. H.. Tsang. N. M.. Yandell. D. W., Kelsey. K. T.. andLiber, H. L. Altered p53 status correlates with differences in sensitivity to radiation-

induced mutation and apoptosis in two closely related human lymphohlasl lines.Cancer Res., 55: 12-15, 1995.

Olive, P. L., Banath. J. P., and Durand. R. E. Development of apoptosis andpolyploidy in human lymphohlast cells as a function of position in the cell cycle at thetime of irradiation. Radial. Res.. 146: 595-602. 1996.

Yu. Y.. Li. C. Y.. and Little. J. B. Abrogation of p53 function by HPV16 Â£6genedelays apoptosis and enhances mutagenesis hut does not alter radiosensitivity in TK6human lymphoblast cells. Oncogene. 14: 1661-1667. 1997.O'Connor, P. M., Jackman. J.. Jondle. D.. Bhatia. K.. Magralh. !.. and Kohn. K. W.

Role of the p53 tumor suppressor gene in cell cycle arrest and radiosensitivity ofBurkitt's lymphoma cell lines. Cancer Res.. 5.?: 4776-4780. 1993.

Warters. R. L. Radiation-induced apoptosis in a murine T-cell hybridoma. CancerRes., 52: 883-890, 1992.Haimovitz-Friedman. A.. Kan. C. C.. Ehleiter. D.. Persaud. R. S., McLoughlin, M .Fuks. / and Kolesnick. R. N. loni/ing radiation acts on cellular membranes togenerate ceramide and initiate apoptosis. J. Exp. Med.. 1X0: 525-535, 1994.Santana. P.. Pena. L. A.. Haimovitz-Fricdman. A.. Martin, S.. Green. D., McLoughlin,M.. Cordon-Cardo. C.. Schuchman, E. H.. Fuks, Z., and Kolesnick. R. Acid sphin-gotnyelinase-deficient human lymphoblasts and mice are defective in radiation-induced apoptosis. Cell. 86: 189-199. 19%.Michael. J. M., Lavin, M. F.. and Watters. D. J. Resistance to radiation-inducedapoptosis in Burkitt's lymphoma cells is associated with defective ceramide signaling.

Cancer Res.. 57: 3600-3605, 1997.

Pruckler, J. M.. Pruckler. J. M.. and Ades. E. W. Detection by polymerase chainreaction of all common Mvcoplasma in a cell culture facility. Pathobiology, 63: 9-11.

1995.Eldridge. J. S.. and Crowther, P. Absolute determination of 1125 in clinical applications. Nucleonics. 22: 56-59. 1964.Guizani, L.. Lang. P., Stancou, R.. and Bertoglio. J. Regulation of interleukin-2 andinterleukin-4 receptor expression in human and ape lymphoid cell lines. LymphokineCytokine Res.. 12: 25-32. 1993.Boag. J. W. The statistical treatment of cell survival dala. In: Cell survival after lowdoses of radiation, pp. 40-53. London: John Wiley and Sons. 1975.Dean. P. N., Dolbeare. F.. Gratzner. H.. Rice. G. C., and Gray, J. W. Cell-cycleanalysis using a monoclonal amibody lo BrdUrd. Cell Tissue Kinel.. 17: 427-436.

1984.Towbin. H.. Staehelin, T., and Gordon. J. Electrophorelic transfer of proteins frompolyacrylamide gels lo nitrocellulose sheets: procedure and some applications. Proc.Nati. Acad. Sci. USA. 76: 4350-4354. 1979.

Brenner. D. J., Zaider. M.. Geard, C. R., and Georgsson. M. A. Cell survival andplating efficiency. Radial. Res.. ///: 572-576. 1987.

Bradley. T. R., Telfer, P. A., and Fry, P. The effect of erythrocyles on mouse bonemarrow colony developmenl in vitro. Blood. 38: 353-359, 1971.Szekely. J. G.. and Lobreau. A. U. High radiosensitivity of the MOLT-4 leukaemiccell line. Int. J. Radial. Biol., 48: 277-284. 1985.Radford. I. R. Evidence for a general relationship between the induced level of DNAdouble-strand breakage and cell-killing after X-irradiation of mammalian cells. Int. J.Radial. Biol.. 49: 611-620. 1986.Gannon. J. V., Greaves, R., Iggo, R., and Lane, D. P. Aclivating mutalions in p53produce a common conformational effect. A monoclonal antibody specific for themutant form. EMBO J.. 9: 1595-1602. 1990.

Wolf, D.. and Roller. V. Major dÃ©telionsin Ihe gene encoding Ihe p53 lumor antigencause lack of p53 expression in HL-60 cells. Proc. Nati. Acad. Sci. USA, 82:790-794. 1985.

Cheng. J.. and Haas. M. Frequenl mutations in the p5J tumor suppressor gene inhuman leukemia T-cell lines. Mol. Cell. Biol.. 10: 5502-5509, 1990.

Chow. V. T.. Quek, H. H.. and Tock. E. P. Alternative splicing of ihe p5J tumorsuppressor gene in the Moll-4 T-lymphohlaslic leukemia cell line. Cancer Lett., 73:141-148, 1993.

Strasser. A., Harris, A. W.. Jacks. T.. and Cory, S. DNA damage can induce apoptosisin proliferating lymphoid cells via p53-independenl mechanisms inhibitahle by Bcl-2.Cell. 79: 329-339. 1994.Boise, L. H.. Gonzalez-Garcia. M.. Poslema, C. E., Ding, L.. Lindslen. T.. Turka,L. A.. Mao. X., Nunez. G., and Thompson. C. B. bcl-x, a bcl-2-relaled gene thatfunctions as a dominant regulator of apoptotic cell death. Cell, 74: 597-608. 1993.Han, Z., Chatterjee. D.. Early. J.. Pantazis. P.. Hendrickson. E. A., and Wyche. J. H.Isolalion and characlerizalion of an apoptosis-resistanl varianl of human leukemiaHL-60 cells lhat has switched expression from Bcl-2 to Bcl-xL. Cancer Res., 56:1621-1628. 1996.

Radford. I. R., Murphy. T. K.. Radley. J. M.. and Ellis. S. L. Radiation response ofmouse lymphoid and myeloid cell lines. Part II. Apoptolic dcalh is shown by all linesexamined. Int. J. Radial. Biol., 65: 217-227. 1994.Kuerbilz. S. J.. Plunkell. B. S., Walsh. W. V.. and Kaslan. M. B. Wild-type p53 is a

cell cycle checkpoint determinant following irradiation. Proc. Nati. Acad. Sci. USA,Â«9..7491-7495. 1992.D'Anna. J. A.. Valdez. J. G.. Habbersell. R. C.. and Crissman, H. A. Associalion of

G,-S-phase and lale S-phase checkpoints wilh regulation of cyclin-dependent kinasesin Chinese hamsler ovary cells. Radial. Res.. 148: 260-271. 1997.



RADIOSHNSmVITY OF HUMAN LYMPHOID CELLS

39. Han. Z.. Chatterjee. D.. He, D. M.. Early, J.. Pantazis, P.. Wyche, J. H., andHendrickson, E. A. Evidence for a G2 checkpoint in p53-independent apoptosisinduction by X-irradiation. Mol. Cell. Biol., 15: 5849-5857, 1995.

40. Cheng. J.. Scully, P.. Shew. J. Y.. Lee. W. H.. Vila, V., and Haas, M. Homozygousdeletion of the retinoblastoma gene in an acute lymphoblaslic leukemia (T) cell line.Blood, 75: 730-735, 1990.

41. Haas-Kogan, D. A.. Kogan. S. C.. Levi. D.. Dazin, P.. A. T. A.. Fung, Y. K.. andIsrael, M. A. Inhibition of apoplosis by the retinoblastoma gene produci. EMBO J.,14: 461-472. 1995.

42. Ishii. H., Igarashi, T., Saito, T., Nakano, T., Mori. M., Ohyama, H., Miyamoto, T.,Saito. Y., and Oh. H. Retinoblastoma protein expressed in human non-Hodgkin's

lymphoma cells generates resistance against radiation-induced apoptosis. Am. J.Hematol., 55: 46-48, 1997.

43. Christoffersen. J.. Smeland, E. B.. Stokke. T.. Tasken. K.. Andersson. K. B.. andBlomhoff. H. K. Retinoblastoma protein is rapidly dephosphorylated by elevatedcyclic adenosine monophosphate levels in human B-lymphoid cells. Cancer Res., 54:2245-2250, 1994.

44. Abend. M., Gilbert/, K. P.. Rhein. A., and van Beuningen. D. Early and late G2 arrestof cells undergoing radiation-induced apoptosis or micronucleation. Cell Proliferation, 29: 101-113, 1996.

45. McKenna, W. G.. Bernhard. E. J.. Markiewicz. D. A., Rudoltz. M. S.. Maity. A., andMuschel, R. J. Regulation of radiation-induced apoptosis in oncogene-transfectedfibroblasts: influence of H-ras on the G2 delay. Oncogene, 12: 237-245, 1996.

46. Gupta, N., Vij, R.. Haas-Kogan. D. A.. Israel. M. A.. Deen, D. F.. and Morgan, W. F.Cytogenetic damage and the radiation-induced G1-phase checkpoint. Radial. Res..Â¡45:289-298, 1996.

47. Charlton. D. E. The range of high LET effects from I25I decays. RadiÃ¢t.Res., 107:

163-171. 1986.48. Kandaiya, I. S.. Lobachevsky. P. N., D'Cunha. G.. and Martin. R. F. DNA strand

breakage by I2il-decay in a synthetic oligodeoxynucleotide I. Fragment distribution

and evaluation of DMSO protection effect. Acta Oncol., 35: 803-808. 1996.49. Walts, J. D.. Gu. M.. Polverino. A. J.. Patterson. S. D., and Aebersold, R. Fas-induced

apoptosis of T cells occurs independently of ceramide generation. Proc. Nati. Acad.Sci. USA. 94: 7292-7296, 1997.

2824



1998;58:2817-2824. Cancer Res David R. Aldridge and Ian R. Radford Radiation between Human Lymphoid Cell LinesExplaining Differences in Sensitivity to Killing by Ionizing

Updated version

http://cancerres.aacrjournals.org/content/58/13/2817

Access the most recent version of this article at:

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://cancerres.aacrjournals.org/content/58/13/2817To request permission to re-use all or part of this article, use this link

http://cancerres.aacrjournals.org/content/58/13/2817http://cancerres.aacrjournals.org/cgi/alertsmailto:[email protected]://cancerres.aacrjournals.org/content/58/13/2817http://cancerres.aacrjournals.org/

Documents

(CANCER RESEARCH 58, 2817-2824. July I. 1998 ......(CANCER RESEARCH 58, 2817-2824. July I. 1998] Explaining Differences in Sensitivity to Killing by Ionizing Radiation between Human