Low-dose hypersensitivity after fractionated low-dose irradiation in vitro

  • Published on
    06-Mar-2017

  • View
    212

  • Download
    0

Transcript

  • int. j. radiat. biol 2001, vol. 77, no. 6, 655 664

    Low-dose hypersensitivity after fractionated low-dose irradiationin vitro

    S. C. SHORT, J. KELLY, C. R. MAYES, M. WOODCOCK and M. C. JOINER*

    (Received 8 September 2000; accepted 30 January 2001)

    Abstract. fully enable repair and therefore exhibit maximalPurpose : It was demonstrated previously that some radioresistant radioresistance. If this is the case, the radioresistancetumour cell lines respond to decreasing single, low radiation seen after doses > 1Gy may mainly re ect this extradoses by becoming increasingly radiosensitive. This paper reports repair. This is termed increased radioresistancethe response of four radioresistant human glioma cell lines to

    (IRR). A modi cation of the linear-quadratic (LQ)multiple low-dose radiation exposures given at various intervals.Three of the cell lines (T98G, U87, A7) were proven already to equation has been used to model the response ofshow low-dose hyper-radiosensitivity (HRS) after single low cell lines that behave in this way. In this induced-doses; the fourth, U373, does not show HRS after acute doses. repair (IR) model, the sensitivity parameter of theMaterials and methods : Clonogenic cell-survival measurements were cell population, a, is not constant but is typicallymade in vitro using the Dynamic Microscopic Image Processing

    > 1Gy 1 at very low doses and decreases as the doseScanner (DMIPS) or Cell Sorter (CS) following exposure to240 kVp X-rays one or more times. increases. Statistical tting of this model to the dataResults: A consistent, time-dependent hypersensitive response to from cell lines showing HRS has con rmed thata second, or subsequent, dose was observed in the cell lines that this concept of dose-dependent radioresistance is ademonstrated HRS. This time-dependent change in radiosensit-

    tenable explanation of low-dose radiation e ectsivity did not occur in the radioresistant cell line that did not(Lambin et al. 1993, Short et al. 1999b). In the IRshow HRS (U373). In one cell line that demonstrated strong

    HRS, T98G, a similar time-dependent hypersensitive response model, low-dose hypersensitivity is explained as thewas also seen when the cells were irradiated whilst held in the constitutive response of cells without the enabling ofG1-phase of the cell cycle. In this same cell line, signi cantly repair mechanisms triggered by higher doses.increased cell kill was demonstrated when three very low doses

    The single-dose IRR phenomenon has similarities(0.4Gy) were given per day, 4 h apart, for 5 days, comparedto the adaptive response seen in some cell lines inwith the same total dose given as once-daily 1.2Gy fractions.

    Conclusions : These data demonstrate the possibility that a multiple- which a small initial priming radiation dose (usuallydose per day, low-dose per fraction regimen, termed ultra- < 0.5 Gy) can induce resistance to a second, higherfractionation, could produce increased tumour cell kill in challenge dose given some time later (Wol et al.radioresistant tumours compared with the same total dose given

    1988, Fan et al. 1990, Shadley 1994, Seong et al.as conventional-sized 2Gy fractions.1995) . This adaptive e ect has been most thoroughlystudied in human lymphocyte populations and, in

    1. Introduction these cell lines, is time-dependent; an interval of atleast 4 h is necessary between the priming dose andThere is now overwhelming evidence that somethe challenge dose for the maximum adaptive radio-radioresistant cell lines respond to very low radiationresistance to be seen. Like IRR, adaption is assumeddoses (< 1Gy) with an increasing cell kill per gray.to be related to an enabling or induction of repairThis phenomenon is termed low-dose hyper-mechanisms, in this situation, as a response to theradiosensitivity or HRS (Lambin et al. 1993, 1994, rst dose. However, Wouters and Skarsgard (1997)1996, Marples and Joiner 1993, Joiner 1994, Singhargued that IRR and the adaptive response must beet al. 1994, Skarsgard et al. 1996, Wouters et al. 1996,separate phenomena. They investigated the e ect ofMarples et al. 1997, Joiner et al. 1999, Short et al.a small priming dose on the survival of HT29 cells,1999a, b). HRS has now been demonstrated in morewhich demonstrate HRS, and found that a primingthan 26 human cell lines in vitro. It has been suggesteddose abolished the hypersensitive response to dosesthat it re ects a di erential triggering or induction< 1Gy but did not increase radioresistance to higherof cellular repair mechanisms, so that only after thedoses given after the same priming dose. This indi-levels of damage that occur at high doses do cellscates that IRR at low doses can be triggered withoutinducing a classical adaptive response at high doses*Author for correspondence; e-mail: joiner@graylab.ac.ukand that the two mechanisms of increased radio-Gray Laboratory Cancer Research Trust, PO Box 100, Mount

    Vernon Hospital, Northwood, Middlesex HA6 2JR, UK. resistance must therefore be di erent. The precise

    International Journal of Radiation Biology ISSN 0955-3002 print/ISSN 1362-3095 online 2001 Taylor & Francis Ltdhttp://www.tandf.co.uk/journals

    DOI: 10.1080/0955300011004132 6

    Int J

    Rad

    iat B

    iol D

    ownl

    oade

    d fr

    om in

    form

    ahea

    lthca

    re.c

    om b

    y M

    ichi

    gan

    Uni

    vers

    ity o

    n 10

    /30/

    14Fo

    r pe

    rson

    al u

    se o

    nly.

  • 656 S. C. Short et al.

    mechanisms involved in these responses are currently Hospital, from his own cell collection in December1996. U373 is a human-grade III astrocytoma cellunknown, but there is evidence that protein synthesis

    and nucleic acid turnover are involved in both line supplied by Dr G. Perlman in 1996 from a cellline originally established by J. Ponten and held atadaption and IRR (Youngblom et al. 1989, Marples

    and Joiner 1995) . Any of the many examples of gene the American Type Culture Collection (ATCC). Allcells were maintained in exponential growth in mono-regulation that have been demonstrated following

    radiation exposure to higher doses could also be layer culture in vitro in Eagles minimum essentialmedium (EMEM) supplemented with 10% foetal calfinvolved in the induction of radioresistance after low

    doses. Some groups have demonstrated induction of serum and 0.01% sodium pyruvate. The cells weremaintained in a 90% N2 , 5% CO2 , 5% O2 atmo-early response genes, stress genes and increased heat

    shock protein-like molecules following very low- sphere at 37 C. They were routinely passaged twicea week using a calcium-free salt solution with 0.05%dose radiation (Yamaoka et al. 1994, Prasad et al.

    1995, Sadekova et al. 1997, Amundson et al. 1999) . trypsin and 0.02% EDTA. Single-dose cell survivalcurves for these cell lines have been published (ShortHowever, whilst DNA repair mechanisms seem likely

    to be involved in both responses ( Joiner et al. 1999), et al. 1999a, b), and it has been shown that T98Gcells that have been treated with serum deprivationthis remains to be de nitively proven.

    There is speculation about the possible utilization to produce a reversible G1 arrest also demonstrateHRS after single low doses (Short et al. 1999a).of the HRS response clinically. If HRS is prevalent

    in radioresistant cell lines, then it is possible thatradiotherapy using multiple very low doses per frac-tion might be more e ective in radioresistant tumours 2.1. Experiments using asynchronously growing cells andthan conventional radiotherapy regimens using 2 Gy two consecutive low dosesper fraction. This prediction will only hold true ifthere is a hypersensitive response to each dose in a The Dynamic Microscopic Image Processing

    Scanner (DMIPS) and the Cell Sorter (CS) werefractionated regimen. This might not necessarily bethe case if, for example, the e ect of a subsequent used to measure clonogenic cell survival after expo-

    sure to very low doses of 240 kVp X-rays. Thedose is in uenced by an adaptive response to thepreceding one, as discussed above, thus producing a DMIPS is an automatic scanning microscope that

    locates and counts cells on the surface of a cultureresistant response rather than a sensitive one. Smithet al. (1999) investigated the response of two cell lines, ask and logs their positions in two dimensions so

    that each cell position can be revisited after anC3H 10T1/2 and V79, to consecutive low doses totest the above prediction. They concluded that when appropriate interval to establish whether colony

    formation has occurred. This allows accurate estima-multiple 0.3Gy doses were administered at 3-h inter-vals in vitro, the resulting cell survival did not provide tion of the surviving fraction (SF) even when the

    values are close to unity (Marples and Joiner 1993) . rm evidence to support the presence of HRS.Although their results were consistent with the pres- To measure clonogenic survival after consecutive low

    doses, T98G tumour cells were harvested from theence of some hypersensitivity to low doses, overallthey were adequately modelled using the LQ formula. growth surface using EDTA/trypsin whilst in expo-

    nential growth, washed and diluted to 3 103The data presented here represent a further, moredetailed study, in which the authors set out to cells ml 1 then plated to Nunclon high optical clarity

    asks at 3 103 cells per ask with 5ml medium.investigate the e ect of consecutive and multiple verylow doses on a series of human cell lines already These cells were allowed to attach to the growth

    surface (30min) then the medium was decanted o known to demonstrate low-dose hypersensitivity. Thisstudy provided a test of the principle of ultrafrac- and the asks re lled completely with fresh medium

    and sealed. Flasks were then irradiated with consecut-tionation, i.e. the use of very low doses per fractionto overcome radioresistance. It also investigated ive doses of 0.4 or 0.5 Gy at 37 C using a Pantak

    240 kVp X-ray set (HVL 1.3mm Cu) with varyingwhether an adaptive response could be elicited aftervery low doses and how this relates to the induction intervals between doses (012 h). After irradiation,

    the asks were scanned using the DMIPS to recordof resistance at high doses.the positions of 200300 cells. The asks were thenreturned to the incubator at 37 C. After 7 days, the2. Materials and methods asks were returned to the DMIPS and rescannedusing a revisiting programme to check manually forT98G and U87 glioblastoma cells were obtained

    from ECACC in December 1996. A7 cells were a colony formation at each previously logged cell posi-tion. SF was calculated for each irradiated ask bygift from Professor H. Suit, Massachusetts General

    Int J

    Rad

    iat B

    iol D

    ownl

    oade

    d fr

    om in

    form

    ahea

    lthca

    re.c

    om b

    y M

    ichi

    gan

    Uni

    vers

    ity o

    n 10

    /30/

    14Fo

    r pe

    rson

    al u

    se o

    nly.

  • 657Low-dose hypersensitivity after fractionated low-dose irradiation

    Table 1. Fraction of the T98G cell population in each phasedividing its plating e ciency by the mean platingof the cell cycle following serum deprivation.e ciency of three unirradiated control asks.

    U87, A7 and U373 cells were unsuitable for ana- Cell-cycle phase Fraction (%)*lysis using the DMIPS because of their non-discrete

    G1 84.4 0.83colony morphologies. The DMIPS protocol was alsoS 8.55 0.63di cult to use when more than two consecutive dosesG2 7.62 1.29were given, because of the time required to scan

    each ask, which had to be delayed until after the * Values are fraction SD. nal irradiation. Therefore alternative protocols,using the CS (U87 and U373) and conventional population, con rming that > 80% of cells had G1survival assays (A7), were used in experiments with DNA content at the time of irradiation. These condi-these cells and in which more than two consecutive tions were used to reduce the potential confoundingdoses were given. The survival of A7 cells was e ect of cell-cycle progression during experiments inassessed using conventional assays because these cells which cells were irradiated with consecutive very lowhad a low plating e ciency following cell sorting. doses. T98G cells were plated and irradiated asIn the cell sort protocol, cells were harvested as above but serum-de cient medium (0.5% FCS) wasdescribed above and plated using a modi cation of used in place of standard EMEM (10% FCS). Whenthe technique rst described by Durand (1986) . In the irradiations were completed, 3.5ml mediumthis protocol, a FACSVantage equipped with a sort was removed from each sealed 40-ml ask and rep-enhancement module (Becton Dickinson) was used laced with 3.5 ml FCS, thereby restoring the serumto dispense droplets containing single cells directly to concentration to 10%.25 cm2 culture asks containing 5ml medium. Cellswere identi ed by their forward and side scatter 2.3. Fractionated very low doseswithout the use of uorescent stains. Four hundredto 1000 cells were counted into each dish, left to In a further set of experiments using T98G cells,attach for 1 h in an incubator at 37 C in a 90% N2 , the SF after 15 fractions of 0.4 Gy, given three times5% CO2 , 5% O2 atmosphere, then irradiated using a day, was compared with the same total dose givena Pantak X-ray set, as described above, to give 0.4Gy as once-daily 1.2Gy fractions. Six asks, each withor 0.5Gy 3 with intervals of 06 h. The asks were 3 103 cells were prepared as described above andreturned to the incubator for 1214 days, then irradiated with three fractions per day at 4-h intervalswashed and stained with crystal violet and colonies (09:00, 13:00 and 17:00 hours each day) each day> 50100 cells were manually scored as survivors. for 5 days. A parallel set of asks received 1.2Gy asSF was calculated relative to the plating e ciencies a single dose once per day for 5 days. On day 6, thefor three non-irradiated controls, as in the DMIPS cells in each ask were removed from the growthexperiments. These two methods have been com- surface using trypsin/EDTA and the total cellpared by Short et al. (1999a) . In conventional assays, number in each ask was measured using a haemo-cells were harvested as described then the cell sus- cytometer to count cell numbers in a known dilution.pension was diluted and pipetted to 25 cm2 asks Five hundred cells from each ask were then replatedcontaining 3 103 cells per ask. These asks were to 6 cm Petri dishes and incubated at 37 C in a 90%then irradiated as described above, and SF was calcu- N2 , 5% CO2 , 5% O2 atmosphere for 14 days tolated in the same way as in DMIPS and cell-sort allow them to grow to 50100 cell colonies. Theexperiments. medium was then removed from the dishes and they

    were stained with crystal violet. Colonies with > 50cells were manually scored as survivors and SF was2.2. Consecutive doses in growth-arrested cells calculated as described above. Relative clonogensper ask was calculated asT98G cells enter a reversible G1 arrest in condi-

    tions of serum deprivation or high cell density (con- uence arrest) (Stein 1979) . This has been con rmed

    Final cells per flask (treated)

    Final cells per flask...

Recommended

View more >