REPORT DOCUMENTATION Of '

A.4P~WW 00U~% Is 14 t ~R .04mq SNW ewam ft o I no-~ "a amvas."~ ww4 inm ft 1144W ahmS 4wp 0Ua W row" "" e"14n4a4&sn44

2243 f ff fMWwt an$ aIk# Y1. AGENCY USE ONLY (LaV@b*j 2 REPORT DATE A~D 'MA2 ,I 778

19904. TITLANDOSUBTITLE r MJNOGN NUMIERS

(see title on reprint) NWED QAXH

s. AUT)4R(S)

Weis .~*~.Work Unit No. C00162, 100159

7 PIfOMtMNG ORGaAMZTKONNA4MS) ANO AOONESS4E5I S. PENFONMING ORGAMZATIONREPORT mumUERp..

Armo'd Fot:ces Radiobiology Research InstituteDefense 4uclear Agency SR90-24 CBethesdo.., M. 20889-5145

9. %a'N /MNTW AGENCY NAME4I) AND A0OflESSIES) 10,SOSRNGmNOINAGENCY REPORT ftMGER

Defouc.e Nuclear AgencyWashington, DC 20305

11, SUP$4.EMENTANY NOTES

12S. DISTRIOU10NAVAN.A8IUIY STATEMFNT12.ITAUONC0

Approved for public release;.distribution unlimited.

13J ASSTRACI (Afeawwo 2W0 w.E ____________

Aeaesuiow poptIT IS ' GR A I 'DYIC TAB

Un~uinouncedDTIC utfotnELECTEBy

CT15199ODistribution/S~E 0 Avvn11nbility Codes'Avail And/or

Disit Special,

14 SuAECT TERMSI u6A f II

14

17 iacuft CLAWAIO 4 Sic'ArWvCLASSCATION It SICUO"r CU*APCATP 30 u"9TA"'ON OPOF RPOR1 OF Twi P0Q 6 OP ASTRACT ASSTRACT

uYECAsSITIIW UNCLASSIFIED 40-wI ftA*to10

INT. 1. RADIAT. 9IOL., 1990, VOL. 57,,-No. 4, 7W9722

SR0-4Advances in radioprotection through the use of combinedagent regimens"'

J. F. WEISSt, K. S. KUMARt, T. L. WALDENt,R. NETAt, A. R. LANDAU ER§ and E. P. CLARKI

tRadistion Biochemistry, *Experimental Hematology, and §BehavioralSciences Departments, Antied Forces Radiobiology Research Institute,Behesda,.MD 20814-5143. USA

'The moat effective radioprotective agents exhibit toxicities that can limit theirusefulness. It may be posible to use coinbinations of agents with differentradioprotective mechanisms of action at less toxic doses, or to reduce the toxicityof the major protective compound by adding another agent. With regard to thelatter possibility, improved radioprotection and reduced lethal toxicity of thephosphor.othioste WR-2721 was observed when at was administered in combin-ation with metals (setlenium. zinc or cotper). The known mechanisms of actionof potential radioprntective agents and varying effects of different dotes andtimes of administration in relation it) radiation exposure must be consideredwhen using combined-agent regimens, A ijumber oif receptor-mediated pro-tectors and other biological compounds, including endlotoxtin, eicosanoids andcytokines. have at least an additive effect when administered with thiol pro-tectorst. Eicosanoids and other binoctive lipids must be administered beforeradiation eAxosure., whereas some imntinomoicdulators have activity when ad-ministered either before or after radiation exposure -For erample. the cytokineinierleukin-? administered simultaneously witb %R-2721 before irrudiation orafter irradiation enhances the radioprotecttie efticscy of WR-2721. The mosteff'ective single agents or combinations of protectorsi result in a decrement inlocomotor activity, an index of behavioral toxicity. Raecent evidence indicatesthat administration of the CNS stimulant caffeine mitigates the behavioraltoxicity of an effrctive radioprotective dose of the phosphoroithioste WR-.1689without altering its radioprotective efficacy These exatmples indicate that theuse of combinations (if agents is a promising approach for maximizing radio-protection with minimal adverse effects..~

1. Protoction by thiolt, antioxidants, anid immunaodulatnnHistorically, the development of radioiprotective agents has been dominated

by the study of sulf hydrvl compounds, particularly the aminothiola and thephosphorot hio tros. Various metnhanismo or combhinations of mechanisms forradiiprotection by thiols have been proposed: it the miolecular level, scavengingof reactive oxvgen tpecieii, hvdrogen transfer reactions, mixed disulfide hypo-thesis, enhancement or protection of repair .-nsvmes. And at the biochemnical -

phsioloRical level, modification of cellular metabolism. ,nductiorn of hvporthtrmia.Anoxia. mediators amnd enzvnmes and biochemical shock (Fove 1981, Weiss andKumar 1998). The most effective thiol protectors developed thus far areS;-2(i -amino~propvlamino)ethv lpho)sphorothioic acid (WR- 2721) and other

Presented at the 22nd Annual Meeting of the European Seciety for Radiation Hioloev,12 I t, Sepiemi-er 1 191), Brusskls

,woi:i Mth go s ix) r~ 19% 1'vi~e Of Im LiA

9010 0.

710 J. P. Wei et al.

phosphorothioates with slightly differing structures, such as S-2[(3-methylaminopropyl)amiro]ethylphosphorothioic acid (WR-3689) (Davidson et al.1980). A dose reduction factor (DRF) as high as 2.7 against 30-day lethality in micehas been achieved with WR-2721 (Yuhas and Storer 1969), but high levels ofprotection are accompanied by side-effects that may be unacceptable in manysituations.

A number of 'natural antioxidants' with low toxicity, such as gluta.hione,,superoxide dismutase, antioxidant yitamins (vitamins E, A, and C), as well assubstances that mimic or induce activity of endogenous antioxidant systems (e.g.selenium), have been studied as radioprotectors. Generally, these natural agentsprovide a low degree of protection compared with phosphorothioates, but they maybe of value in certain situations. Although the post-irradiation administration ofantioxidants or free radic d scavengers would not be expected to have much effect,evidence suggests that this may occur to some extent, and is probably related tomodulation of later reactions, e.g. interaction of radiation-induced radicals ofbiomolecules with reactive oxygen species evolved during normst! cellular processes(reviewed in Weiss and Kumar 1988). There is some evidence of protection againstlethality for mice administered the following antioxidants after irradiation: super-oxide dismutase (Petkau 1987), vitamin A (Seifter et al. 1984) and vitamin E(Malick et al. t978). There is evidence, mostly in vitro, that thiols administeredpost-irradiation can enhance DNA repair (Riklis 1983) and reduce mutageniceffects (Grdina et al. 1985); however, mercaptopropiorylglycine does protectagainst radiation-induced chromosome aberrations in bone marrow when admini-stered to mice after irradiation (Umr Devi and Thomas 1989).

Although it is unlikely that compounds administered after irradiation will havea greater effect than those administered before irradiation that intercept or immedi-ately repair damage or enhance repair mechanisms, from a practical point of view itis important to develop therapeutic agents, such as immunomodulators or biolog-ical response modifiers, that would enhonce hematopoietic and immunologicalresponses even when administered during the post-irradiation period.

The first and longest-studied of this class is endotoxin. When studying newerimmunomodulators it is useful to recall that although endotoxin is most effectivewhen ndministered 24h before irradiation, it provides slight protection whenadministered shortly before or even after radiation exposure (Ainswcrth 1988). Aneffective category of radioprotectors of the immunomodulator class is polysac-

charides. The extensive studies of Patcien (reviewed in Patchen et al. 1988)indicate that glucan (P-1,3 polygluc0se) acts as a hiolor,.s1 response modifier whenit protects against radiation exposure, e.g. when particulate glucan is given 24 hbefore irradiation. However, the work of Maisin et al. (1986) indicates that glucanand related polyaccharides may also act as free radical scavengers, because highlevels of protection can be obtained by some polysaccharides when they are givenshortly before irradiation (similar to aminothiols). Our studies (Weiss and Kumar1988), using synthetic rsdioprotector/immunomodulators other than glucan, suchas diethyldithiocarbamate (DDC) and levamisole, suggest that many immuno-modulators can modulate oxidative processes and som- exhibit both pro-oxidant

and antioxidant properties. These anomalies point out that it is sometimes difficultto classify radioprotective agents into rigid categories. Also different mechanismsmay predominate, depending on dose and time of delivery. Differences in cell.biochemistry among organs or between normal and tumor tissue can result in

m9

Radoprotectii by combi' ed agenti 711

differential effects, inciuding protection in one tissue and rdiosensitization ortoxicity in another. DDC is an example of a thiol protector with different effects(Kumar of at. 1986). These factors must ultimately be taken into considerationwhen choosing protectors of differing mechanisms for use in combinations.

2. Receptor-medsited radlowpotectionIdentification of specific receptors for many radioprotectors presents a great

advantage as it allows a better understanding of the mechanisms of action ofradioprotective agents at the cellular level. This diverse clas' of radioprotectiveagents, having known receptors, has many subclasses and would include thebioactive lipids, naturally occurring peptides, and some synthetic compounds.Bioactive lipids that are radioprotective include metabolites of arachidonic acid,such as prostacyclin (PG12) and leukotrien C, (LTC4); synthetic analogs ofprostaglandins, such as 16,16 dimethyl prostaglandin E2 'diPGE 2 ); and otherphospholipid moieties (platelet activating factor (PAF)). The radioprotective activ-ity of endotoxin, a lipopolysaccharide, is probably related to its lipid component(reviewed by Ainsworth 1988).' The extracellular and intracellular activitiesthought to be involved in receptor-mediated protection are shown in figure 1.

Of the compounds acting through receptor mediation, the most active appear tobe diPGE2 (Hanson and A insworth 1985, Walden a al. 1987), LTC, (Walden et al.1988), PGi 2 (Hanson 1987b), and PAF (Hughes et al. 1989) DiPGE2 and LTC4effectively protect hematopoietic stem cells and intestinal crypt cells and enhancesurvival of irradiated mice (Hanson and Ainsworth 1985, Hanson 1987a, Waldena al. 1987, 1988). Of the many arachidonic acid mttabolites tested, PGI2 (Hanson1987b) and the prostaglandin analogue misoprostol (Hanson et al. 1988) alsoprovide a high degree of intestinal protection. All of these compouids are onlyeffective when given before irradiation.

The itiodies of Walden and co-workers show that the maximum protectionattainable by treatment with bioactive lipids (30-day survival of irradiated mice) isin the range of protection afforded by the phosphorothioates. However, when acomparison is made based on administration of equitoxic doses (1/4 LD 10), they arePot as effective: WR-2721 or WR-3689>diP(,E 2 > LTC4 = PAF. The major pro-blems with diPGE 2 are extensive diarrhea and behavioral toxicity at radio-protective doses. The eicosanoids mediate many important physiological and patho-logical reactions, ranging from vasoregulation to inflammation, thus complicatingthe elucidation of the mechanisms responsible for their radioprotective properties. I naddition, the eicosanoids also function as mediators of radiation injury (inhibitors ofprostaglandin synthesis, such as indomethacin, can be radioprotective). Protectionmay involve events at the cell membrane level and induction of cell hypoxis, as wellas profound physiological effects (Walden 1987).

Many biological response modifiers, such as endotox'n and glucan, induce thepeptide cytokines interleukin- I (I L- 1) and tumor necrosis factor (TN F), and someof the properties of endotoxin may be due to prostaglandin and leukotrieneinduction as well. A common feature of synthetic immunomodulators that areprotective is their ability to induce cytokines, such as colony-stimulating factors(CSF) and interleukins (Chirigos and Patchen 1988). Although the radioprotectiveeffects of a variety of biological response modifiers may be due to enhancement ofhemstopoietic recovery, or other effects on a variety of immune cells, this may comeabout by the release of cytokines, which in turn induce the release of many

4

712 J. F. Weiss et s.

Figure 1. Cellular activities related to receptor-mediated radioprotetion.

mediators, such as the products of arachidonic acid metabolism, and pathways thatcan be considered inflammatory (Neta 1988). Identification of IL-I and "INF asprotective and therapeutic agents against radiation (Nets 1988, Nets et al. 1988)presents direct evidence that inflammatory pathways participate in prevention ofradiation damage and in repair, because these two cytokines are key inflammator,mediators, produced endogenously in response to multiple exogenous insults(infections, trauma and other physical stresses).

IL-I provides varying degrees of protection, depending on the time of admini-stration in relation to radiation exposure, probably involving a variety of mechin-isms (Nets et al. 1988, 1989). These may include induction of hematopoieticgrowth factors (G-CSF, GM-CSF, IL-3 and IL-6) and induction of scavengingacute-phase proteins and superoxide dismutase, and probably other mechanismsnot yet recognized (Durum et al. 1989). The protective effects of single dosesof AL-1, whether administered before or after irradiation, contrast with G-CSF,which is effective in improving survival of mice and neutrophil recovery onlywhen it is administered after irradiation in multiple doses (Laver et al. 1990).Most studies of biological response modifiers and cytokines indicate a limit ofprotection of 1.3 DRF.

Other peptide hormones, such as luteinizing hormone-releasing hormone(LHRH), may also protect through receptor mediation. Pre-treatment of rats withanalogs of LHRH produces specific protection from radiation-induced injury to thetestes (Schally et al. 1987). This may have important clinical applications. Methyl-xanthines, isoproterenol and norepinephrine act through receptors to increasecellular cyclic AMP levels, which might contribute to protective activity.

3. The problem of toxicity and combinations of radioprotective agentsA major question in chemical radioprotection remains: can protective mechan-

isms be separated from mechanisms of toxicity? As Maisin and Bacq (1975) haveemphasized, the goal of non-toxic radioprotection appears difficult to attain becauseradioprotection and toxicity seem to be intimately linked in all organisms, particu-larly in mammals. The problem of toxicity of single protectors and combinations

Radio~otctim by combined agents 713

is more acute when the intended use is at radiation accident sites or in space, whereperformance is an important factor. Behavioral and other tolicities are lessproblematic in clinical applications, where side-effects can be controlled. Clinicalstudies with WR-2721 have shown that it produces a variety of side-effects,including nausea, vomiting and hypotension (Blumberg et al. 1982). Animal studies

" indicate that WR-2721 also produces a decrement in performance (Bogo et al. 1985,Landauer et al. 1987). The use of WR-2721 as an adjunct to radiotherapy andchemotherapy should prove beneficial (Glover et al. 1988), especially if side-effectscould be minimized. Similarly, because of exploitable differences in eicosanoidmetabolism between normal and tumor 'tissue, the use of these bioactive lipids asprotectors (or inhibition of their synthesis) in cancer treatment may some day be areality (Walden 1987, Hanson et al. 1988). Protective cytokines, such as IL-I andG-CSF, may also have considerable side-effects, but they are in clinical use.

Performance decrement must be evaluated when developing radioprotectors foruses where behavioral toxicity would be an unacceptable sioe-effect. In a series ofstudies, Landauer and co-workers demonstrated that the most effective protectorshave the greatest behavioral side-effects, as measured by alterations in locomotoractivity (Lcndauer et al 1989b). In general, biological compounds produce decre-ments at least as large as chemical radioprotectors. Studies demonstrate that whenradioprotective efficacy is compared on the basis of doses with equal behavioraltoxicity, WR-2721 and WR-3689 (which have undesirable behavioral effects) arestill superior to other radioprotectors tested.

There is now considerable evidence suggesting that the use of combinations ofagents is a valid concept (Maisin et al.' 1968, Patcher et al. 1989, Szanyik andSantha 1976, Urs Devi and Thomas 1988, Weiss et al. 1987). A review of data oncombined radioprotection indicates that the maximum achievable DRF, usingcombinations of protectors, will be approximately 3. However, it is extremelyunlikely that this will prove practical because of concomitant toxicity. A morereasonable approach is to use combinations providing a lower DRF but acceptabletoxicity. In addition, DRFs obtained with combined lower doses of protectors canbe further extended by bone marrow transplants and other supportive careadministered after irradiation. When the intended use of a combination of pro-tectors is as an adjunct to radiotherapy, appropriate preclinical testing must be doneto determine how the combination alters tumor protection versus normal tissueprotection.

Throughout the remainder of this paper we discuss protection aginst -- orX-irradiation by combinations of agents, illustrated also by previously unpublisheddata. In all radioprotection studies reported in the tables, CD2F1 male mice wereirradiated bilaterally with cobalt-60 at I Gy/min. The LD50,30 for saline-treatedmice was 8.0-8.5 Gy. Other experimental details are given in Weiss et al. (1987) andLandauer et al. (1987).

4. Combinations of metals and phosphorothioatesTreatment of mice with salts of various metals (copper, zinc or selenium)

provided a small radioprotective effect. Therefore, it was of interest to determinewhether combinations of metals and thiol compounds would be beneficial. We firstconcentrated or. selenium (Se) because of its known inter-relationship with endo-genous antioxidant defense systems, such as vitamin E and glutathione peroxidase(Jacobs et a!. 1983). The effect of Se, as sodium selenite, on the acute toxicity and

714 J. F. Weiss et al.

radioprotective effect of WR-2721, was studied in male CD2F1 mice (Weiss et al.1987). Injection of 1.6mg/kg Se 24h before WR-2721 (800-1200mg/kg, i.p.)decreased the lethal toxicity of WR-2721 significantly. Se injection alone(1.6mg/kg) 24 h before cobalt-60 irradiation increased survival (DR F= -1.1). Anenhancement of the protective effect of WR-2721 also occurred hen Se was Iinjected 24h before WR-2721 (200-600mg/kg, i.p., 0-5 h before irra iation). Forexample, after exposure to 22 Gy (1 Gy/min), 30-day survival was 00 per centwhen mice were treated with both Se and 600mg/kg WR-2721, an 13 per centwhen they were treated with WR-2721 alone. The DRFs for 30-day survival after400mg/kg WR-2721 were 2.6 with Se and 2.2 without Se pre-treatme nt.

Pre-administration of zinc aspartate 2 min before cysteami: e or 2-fl-aminoethylisothiouronium-Br-HBr (AET) increased the radioprotec ive effect ofthe thiols tFloersheim and Floersheim 1986). Brown etal. (1988) demo strated thatzinc chloride administration 5 min before WR-2721, resulted in an inc -ease in boththe hematopoietic and gastrointestinal DRFs.

In subsequent studies we compared the effects of Se, zinc (Zn) as inc chlorideand copper (Cu) as copper sulfate. We used lower doses than those used previouslyin combination with thiolh, and WR-2721 was also administered at a r ,latively lowdose. Mice were either pre-treated with the metals, or the metals were 2 dministeredsimultaneously with WR-2721. Table I compares the effects of the n etals- on thelethal toxicity of WR-2721 (1000mg/kg). Cu pre-administration ( 1h or 24h)resulted in an increase in survivors, but simultaneous' adminis1ratien withWR-2721 was not effective. Zn was effective when it was giver 3b beforeWR-2721. Although this, study was done with a small number of nimals, theresults suggest that Se was more effective than the other metals when it %as givensimultaneously with WR-2721.

Table 2 shows that pre-treatment with each of the three metals at - 3 h enhancesradioprotection by WR-2721 (200 mg) at 14 Gy exposure when all trea tments werei.p. However, when WR-2721 was administered orally and the metals i.p., Se wasnot effective, whereas Cu and Zn had a small effect. Table 3 shows the comparativeradioprotective effects of solutions of metals and WR-2721 combi ed. In thissituation, Se was the most effective, in a simiiar way to its effect on the lethaltoxicity of WR-2721 (Table I).

Table i. Effect of metal, (08 mg/kg, i.p.) on lethal toxicity of WR-2721 (10(0mg/kg, i.p.,CD2FI male mice).

Treatment 30-day survivors

WR-2721 1/30

Cu at -3h 9/10Cu at -24h 7/10WR-2721 and Cu simultaneously 1/10

Zn at -3h 8/10Zn at -24h 0/10WR-2721 and Zn simultaneously 1/10

Seat -3 h 4/10Se at -24h 3/10WR-2721 and Se simultaneously 3/10

- -.- -.- ~- -

Radioprotection by combined agen:s 715

Table 2. Effect of pre-treatment with metals (0"9 mg/kg, i.p., -3 h) on radioprotection byWR-2721 in CD2!1 m.ie mice.

Treatrment 30-day survivors

WR-2721 (700mg/kg, p.o.),I h before irradiation (14Gy cobalt-60) WR-2721 7148 (15°%)

+Cu 11/32 (34%)+Zn 18/32(56%)+Se 3/32 (9%)

WR-2721 (200mg/kg, i.p.),30min before irradiation (14Gy cobalt-60) WR-2721 15/40(38%)

+Cu 34/40(85%)+ Zn 27/40 (68%)+Se 34140 (85%)

The results of metal effects on toxicity and radioprotection by WR-2721 suggestdifferent mechanisms for the potentiation by metals of thiol radioprotection.Floersheim and Floersheim (1986) suggested that Zn stabilizes thiol protectors, but

there is little experim,"tal evidence for this mechanism. Inhibition of alkalinephosphatase by metals would alter the kinetics of conversion of WR-2721 to itsactive free thiol WR-1065. This might occur with higher concentrations of Zn(Brown et al. 1988) or Se (Weiss et al. 1987). The improvement in radioprotectiveeffect of WR-2721 by Se and/or suppression of toxic metabolites formed duringmetabolism of WR-2721 may be due to induction of glutathione peroxidase activityby Se administration (Kumar et al. 1988). I: is possible that other endogenousprotective organometallic compounds are formed when the metals are injected, orthe metals are forming new compounds by reacting with WR-1065. This appears tobe most likely in the case of Se (Kumar and Weiss 1989).

Because oxygen plays an important role in the modulation of radiation sensitiv-ity, modulation by metals of oxygen uptake by WR- 1065 was investigated using anin vitro model system (Kumar and Weiss 1989). The highest rate of oxygenco.asumption by WR-1065 occurred in the presence of Cu, followed by Se, and Znhad very little effect. Purdie et al. (1983) suggested that WR-1065 is oxidized to thedisulfide, and the consequent anoxia may contribute to protection by the drug. Ourstudies indicate that formation of the disulfide of WR-1065 is accelerated in thepresence of increased lev.s of Cu. Earlier in vivo work by Yuhas et al. (1973)

Table 3. Radioprotection in CD2F1 male mice by combined treatment with WR-2721 andmetals. Solutions of WR-2721 (200mg/kg) and metals (0.8 mg/kg) administered i.p.simultaneously 30 min before irradiation.

30-day survivors

Treatment 14 Gy 15 Gy

W R-2721 8/16 (50 0) 1/16 (6%)WR-2721 and Cu 11/16 (690%) 9/16 (561)WR-2721 and Zn 13/161(810o) 5/16 (31 o)WR-2721 and Se 16/16 (1000,,) 13/16 (81 %)

716 J.F. Weiss et al.

showed that the protective effect of WR-2721 was probably influenced by oxygentension. Denekamp et al. (1982) reported that an optimum level of oxygen is neededfor maximum protection with WR-2721, and protection is lower below and abovethat level of oxygen. It is difficult to correlate our in vitro results with the observedradioprotective effects in mice of the metal and WR-2721 combinations, but thestudies established that metal ions are important factors in the interaction betweensulfhydryl compounds and oxygen with respect to radioprotection.

An important adjunct to studies on radioprotection by combinations of agentsare determinations of behavioral toxicity of single agents compared with combin-ations of agents. Automated quantitation of spontaneous locomo 'or activity hasbeen found to be a sensitive measurement of the behavioral to\ city of radio-protectors (Landauer et al. 1987). Table 4 summarizes the effects of combinationsof metals and WR-2721 on mouse locomotor activity. Mice (n= 11 /group) weretested during the nocturnal phase of their light/dark cycle. All treatments resultedin locomotor decrements. Administration of metals alone resulted in an earlier onsetof locomotor decrement than the WR-2721 treatment, but it took longer to recoverfrom the decrement produced by WR-2721. When WR-2721 was administeredwith the metals, the Cu combination resulted in the most severe locomotordecrement, due mainly to the longer recovery time. The Zn and WR-2721 mixtureappears to be the least toxic because WR-2721 alone produced a greater perfor-mance, decrement than the combi.iation. Information on the behavioral toxicity,lethal toxicity and radioprotective effects of combinations can be useful forcomparative assessments of combined radioprotective regimens.

5. Combinations of receptor-mediated and other biological compoundswith phosphorothioates

The first indication that biological response modifiers or immunomodulatorsmight be effecti.e in combination with thiols resulted from studies of endotoxinand AET. Administration of endotoxin at 24 h and AET 15 min before radiationexposure of mice resulted in greater than additive protection (Ainsworth et al.1970). Unpublished work by Walden, using'detoxified endotoxin (monophosphoryllipid A) injected at -24 h, produced an additive effect with WR-2721 administeredat -30 min.

The combined use of WR-2721 and diPGE 2 has been investigated (Hanson1987a, Steel et al. 1988, Landauer et al. 1989b). The studies have shown a favorableprotective response with y-irradiation. When WR-2721 was administered 15 minbefore irradiation with 0.4mg/kg diPGE 2 given 5 min before irradiation, 30-daysurvival increased as compared with that of WR-2721 alone. In this case, protectionby' the combined agents was slightly less than additive. The DRF for WR-2721(200mg/kg) was 1.9; for diPGE 2, 1-45; and for the combination, 2.15. A protectiveresponse that was greater than additive was obtained for survival at 6 days whendiPGE2 was administered 1 h before irradiation and before WR-2721 (Hanson1987a). Greater protection'of murine intestinal crypt cells was observed after thecombined treatment, but protection by high doses of WR-2721 (approximately300-400mg/kg) could not be improved by the addition of diPGE2 . The greaterprotection produced by using the combination of agents in this study could haveresulted from the order of administration, which produced different physiologicalresponses, or from modified catabolism .of the radioprotectors. Misoprostol, asynthetic analog of prostaglandin E l , was effective in protecting intestinal clono-

' .. ... ... ... .. .. .... ... .. .. ... ... ...

Radioprotection by combined agents 717

Table 4. Behavioral toxicity (locomotor decrement) in CD2FI male mice treated i.p-simultaneously with metals (0-8 mg/kg) and WR-2721 (200mg/kg).

Onset of Maximumlocomotor locomotor decrement Time to recoverdecrement from decrement

Treatment (min) (min) percentage (h)

WR-2721 25 25 70 4-0Se 15 15 70 08WR-2721 and Se 10 20 95. 4-5Cu 10 10 66 0-6WR-2721 and Cu 10 30 80 7.0Zn 10 10 35 0-3WR-2721 and Zn 20 30 70 3-5

genic cells when combined with WR-2721 (Hanson et al. 1988). The combinationof misoprostol (25 pg/mouse) followed by a high dose of WR-2721 (10 mg/mouse)extended the longevity of mice irradiated with 20 Gy and 23.5 Gy.

When WR-2721 followea by diPGE2 was administered before fission neutronirradiation, there was no improvement in the survival of mice compared with thosegiven WR-2721 alone (Steel et al. 1988). Behavioral toxicity studies conducted onmice receiving the combination of WR-2721 and diPGE 2 demonstrated a greaterbehavioral decrement measured by lucomotor activity than those produced byeither agent alone (Landauer et al. 1989b).

Although Maisin et al. (1986) failed to find improved radioprotection withpolysaccharides and AET, Patchen it al. (1990) reported enhanced protection byWR-2721 in combination with glucan. The protective effect (DKF 1-2) of particu-late glucan administered at -20 h was additive with the protective effect of WR-2721 (200mg/kg, - 30min). Furthermore, an even greater protection was obtained,when Se was also given at - 20 h. Treatment with each of the three agents, whichact by different mechanisms, resulted in an incease in hematopoietic stem cells(endogenous spleen colony-forming unit assay). Treatment with the combination ofthe three agents was most effective in this regard, as well as in accelerating be-emarrow and splenic granulocyte-macrophage colony-forming cell regeneration(Patchen et at. 1990). A greater than additive effect was obtained when WR-2721was given before irradiation and soluble glucan (Glucan-F) was administered afterirradiation (Patchen. et al. 1989). This study established a potential role for thepost-irradiation use of immunomodulators in combination with traditional thiolradioprotectc 's. Such combinations appear to depend on the sequential thiol-mediated cell protection and immunomodulator-mediated hematopoieticstimulation.

We demonstrated that IL-I was effective in improving the survival of irradiatedCD2Fl mice when it was adminis*ered at times ranging from 20h before to 2 h afterradiation exposure. When administered after irradiation i-a combination with WR-2721 (200mg/kg, -30min), IL-I enhanced survival at radiation doses (15-16Gy)causing hematopoietic and gastrointestinal injury (Neta et al. 1989). Table 5 showsthe effect of simultaneous administration of I L-I (human recombinant interleukin-la, Hoffmann-LaRoche) and WR-2721 at 30 min before irradiation. IL-1 alone hasno protective effect at the high radiation doses tested. Greater than additive

718 J. F. Weiss et al.

Table S. Radioprotective effects of combinations of WR-2721 and IL-! in CD2F1 malemice. Simultaneous i.p. administration 30min before irradiation.

30-day survivors

Treatment 14 Gy ISGy

WR-2721 (200mg/kg) 14/32 (44%) 0/24 (0%)+IL-1 (4pg/kg) 23/24 (96%) 6/16(38%)+ IL- (400pg/kg) 16/16 (100%) 13/16(81%)

l8Gy 20Gy

WR-2721 (400mg/kg) 14/32 (44%) 1/32 (3%)+ 1L- 1 (4 pg/kg) 30/32 (94%) 15/32 (47%)+ IL-I (400pg/kg) 25/32(78%) 6/32(19%)

protection was obtained with combinations of IL-i and 200mg/kg WR-2721 (1/4the LD1 0 dose) or 400mg/kg WR-2721. The highcr dose of IL-I (400mg/kg) didnot provide much benefit over the lower dose (4 pg/kg). WR-2721 may beprotecting hematopoietic stem cells, which in turn may be amplified because of

IL-I administration. Whether similar explanations can be used to account for im-proved protection with WR-2721 and IL-I in the gastrointestinal dose range isbeing investigated. Although the biochemical mechanisms of protection by IL-I

are unclear, it is known that IL-I treatment of mice induces ceruloplasmin andmetaliothionein (reviewed by Neta 1988), both of which have antioxidant andpossible radioprotective effects. The induction of superoxide dismutase by IL-Ialso was observed recently in vitro (Masuda et al. 1988). Vaishnav et al. (1989)showed that IL-I can induce manganese-superoxide dismutase in vivo, but only at6h after administration with the higher dose (400pg/kg). Therefore, superoxidedismutase may contribute to the radioprotective effect. of I L-I, but may not be theonly radioprotective mediator.

It would be useful to combine more than one cytokine witha phosphorothioate,because additive protection has been observed with combinations of cytokines orbiological response modifiers: IL-1 and TNF (Neta et al. 1988), IL-I and G-CSF(Laver et al. 1990) and glucan and BM41.332 (Patchen et al. 1988). Presently, theonly recommended biological factors for the treatment of radiation injuries inhumans are recombinant G-CSF and GM-CSF (Browne et al. 1990) because moreclinical data arz available for these agents. Because WR-2721 is also in clinical use,animal studies on combinations of pre-irradiation administration of phosphoro.thioates combined with post-irradiation administration of G-CSF or GM-CSFshow promise for early acceptance for human use (Patchen and MacVittie,

unpublished work). The further addition of post-irradiation bone marrow tran-plants his been shown previous!y to be of value with a number of protectors, forexample, in combination with IL-I (Oppenheim et al. 1989).

6. Combinations of:awieine and phosphorothioatesResearch on chemical radioprotectors needs to be expanded to include studies

on drugs that will prevent radiation-induced behavioral disruption and per-

Radiopratection by comb d agents 719

Table 6. Effect of caffeine on radioproteetion by WR.3689 in CD2FI male mice. Simulta-neous oral administration I h before irradiation.

30-day survivors

Treatment IOGy I IGy 12 Gy

WR-3689 (400 hig/kg) 12/32 (38),) - -+caffeine (40mg/kg) 14/32 (4400) -

WR-3689 (500mg/kg) 4/8 (500%) 2(8 (25%)+caffeine (40mg/kg) 10/16 (630/) 4(16 (23%)

formance decrement, as well as studies on agents that will modify the behavioraltoxicity of radioprotectors (Bogo 1988). Landauer et al. (1989a) recently deter-mined that the CNS stimulant caffeine zan mitigate the locomotor decrementproduced by WR-3689. These data and results on radioprotection by combinationsof caffeine and WR-3689 will be published in detail elsewhere. The timing ofcaffeine administration in relation to rdministration of the phosphorothioateappears to be important in mitigating the behavioral effect, but not the radioprotec-tive efficacy, oi WR-3689. Caffeine administration does not have an adverse effecton the radioprotective efficacy of WR-3689 wheti the drugs are administered byvarious routes. Table 6 shows the survival of irradiated mice (10-12Gy) aftersimultaneous oral administration of WR-3689 and caffeine I h before irradiation.No significant difference in survival was observed. Although caffeine is generallyconsidered to be a sensitizer, results on radiosensitization effects are, in general,obtained from in vitro cell irradiation studies. There is evidence that, in mice,caffeine provides some protection of jejunal cr'pt cells, as do other inhibitors ofcyclic AMP phosphodiesterase (Lehnert 197'9). The results on combinations ofcaffeine and WR-3689 provide enc.ouragement that the toxicities of major radio-protective compounds can be ameliorated.

AcknowledgemnentsThis research was supported by the Armed Forces Radiobiology Research

Institute,. Defense Nuclear Agency, under work units 00162 and 00159. Viewspresented in this paper are "'iose of the authors; no endorsement by the DefenseNuclear Agency has been given or should be inferred Research was conductedaccording to the principles enunciated in the 'Guide for the Care and Use ofLaboratory Animals' prepared by the Institute of Laboratory Animal Resources,National Research Council. WR-2721 and WR-3689 were obtained from the DrugSynthesis and Chemistry Branch, Division of Cancer Treatment, National CancerInstitute, Bethesda, MD. The assistance of the technical staff if the RadiationBiochemistry and Behavioral Sciences Departments, AFRRI, is gratefullyacknowledged.

ReferencesAmNswowrm, E. J., 1988, From endot-)xins to newer immunomodulators: Survival-

promoting effects of microbial polysaccharide complexes in irradiated animals.Pharmacology Gnd Therapeutics, 39, 223-241.

720 J. F. Wesset A.

AtNswoml4m, E. J., LARSEN, R. M., MITCHELL. F. A., and TAYs. 1.st F., 1970, Survival-promoting effects of endlotoxin in mice, dogst, and sheep. RAdksitiow Protection endSestuittratow, Proceedings of thme Seemed Internsational Sympoium on Raimouselstitv4i,and Radio, Protrie Dugs, edited by H. L. Mormson and M. Quintiiant (TaylorFmiscis, Uendons). pp. 381-388.

BLLmaaRG. A. L., NaiLSON, D. F., GRAMICow31t, M., GiAWUR, D., Gim-x, J. H.. YtHAu,j. M., aid KIAGPIMur~AN, M. M., 1982. Clinical trials of WR-.2721 with radiationtherapy. Iitte~eatsi'nal Journald of Raditon Onecolocy. Bioilosly, Physics. 8, 561- 563.

Bo~uo, V., 1988, Bethavioral radioprotection. Pharmacrology and Therapeutics, 390, 73-78.lOGo, V., )ACONFS, A. J., and Wtiss, J. r., 1985, Behavioral toxicity and efficacy of WR.2721

as a radiopr-itectant. Radiation Restart h, 104. 192- 190.Bitow-4, [D. Q.. (;RAIVM, W. J., NIAcd(Nzir, L )., Prrrouc J. W.. and Shaw, L. W.. 1948,

Can WR-2721 he improved upon? Pharmerolory and Thevateutica. 39,157 -164WMitim %v. D)., VW,'is. j F., '.ts(ViT7Ie, 'I". J.. and P'it iii, M. V. (eai, I~AI. Trlelvt of

Radiation Injturies (Plenum, New York) (in press).CHIaloos., M, A.. and PATCMEN. M. L., 1988, Survey of newer biological resprinse modifiers

for possible use in raedioprotectin. Pharmaiolotry and Theropetus. 39, 243-246.DAVIDS00N. [). E. GRIVAN NI. M., and swwavsyT. R., 19go. Biltogical characteristics of

some improved radiorotttit Radiation Vilartisert. Their V" to e the ('lwit.AMaaqnefof Cancer, edited bit L. W. Bradv (Naaaon., New York). pp. 109 320.

DRNIAMP, J., NImIu eAFL VS. IX, ROls, A.. and !STVWAxT, F. A., 1982. Hadioprotertton ofmouse skin 'ov WR-2721: the critical int'huence of toxvvef tenstion. littvrnwuatlesjJowrnal of Radiation Om'ologv. R(,wg. Phviuocs. , 531 534,

.()rm S, pperiteim. j. ;., an14 \PTA, W, 19119. IMMUnophVIOloI0Cal role of ier.leukin-1. Immuonphisiolog: Role of Creli tand ('yti~tmit f,mumuiv and lntlaen.stat ion, edited by J. J. ()ppenheime and F. She-.ach (C tfisrd t'nivereittv P'ressO.xfoird), pp 210~ 225

Fiwss.Pits, G. L~. and Fi oe4Pi-. PF, 198M, Protection altaitivt ioni,.ing radiation andtsisertior~ %%ith thila bv zinc apartate. Bfritishi Journal of Hadwhilfv. 59. 597 602,

Fovit, W. 0.. 14041, lRadinprotective, drumls. Hiurget'i Mediereal ('Aemeirv. edited by NO. F,Wolff (Wiley, New York 1. pp, It 45.

(;i~na o1 sl, Fox, K. ' ., WiViitw , C., Ki,11411MAN, M. NO., Ti-aist, A , and (;I ivi.J 4.1998, Cloiil trial* of WR-2721 prior to alkvlating agent 4-lhemotherspv and radio.theraspv. Ihaorenat(oo tiled Tliraprute, 39, 1 7.,

(',nNDsA, 1). J,, NAGV, It., Il I V, C.. K., Wtiis, R. I_. and PreAtiio, C , 1905. l'hiradiopr~itector WRIA)(iS reducesl raedtatin-intlaived muntat, is aot the hyproxarithine.itudnine phosphoribosvl tranasferrse locus in V79 cells. (Ci& %teeeaa. 4.,929 9311

WlN~~, R., 1Q97,, Haditipoteciion of murine intestine by W14.2721, 16,lh-dimethviprostaiglandin F2 and 'ie combination of both agents. pdseleo Rresteeih, Ill,16!? 173.

l-IANSO, W, R.. 1997h. Radiation protftion hv exogenous airechidinsir acid and sevrrsilmetabolites, I'rostaillondto ansd Li~ptd .efrtooise In Rad.Jeon /npjurs- edited byT 1X .Walden and It. N. lugts (Plenumr, New York), op, 2313 241.

IfANVw-, W R.. and AeNA-WOrNri4 E. J., 19S.5, 16,16.lDimethvi proet etaind in FZ induvesradsoprotection in murine -niettinal and bemstopoictic stem cells, Radiation RifvfettA.100, 2100 247

IIA'.NUN. W. R , IPI-0iMAN. K A a nd CoilI i, P W. 19M8, Radiatione priotititn en 9eisa

hy prolstaulaeidioai anJ We~tel tomporwinds of the atrachidonic aicid cstradir Mearin,.oloeq and Iheidrspiewtiz. 39, 14 7 156.

III a;NPM. II. N_, W~AlIMON, T, 1. , And Sirto, L.. K , 1989. Itsdioprtetitire rfitas o (f plsicletActiP. inrg factor in mice , ,Ibsfraitsti of aper for the 37th Apeseial 1Ireeis *I theHois aoon Heirar. h ,sv Sealft., If A1. r I t

J.Artis, A j , RINKi-4, W. A , i~vs~V , and Wri-A, J. F., 14111. Effftua o 4 iamin Ftrnd owlenium on alutatbionr perixidaser attivily and survival (if irraiaited rnscr

Plrorpotuevgi of the 7th Ip,1criatlonal "oirets of Radiatlion Rriprrh edited 6v jIllroerse. G, W Farendsers. 14 H. Kai, and A. J. van der Kiwil ('sarinus Niltboff.Amsterdam),.D 1)5 I

Radioliorouectson by co wn~d aeuts 721

Kvuha. K. A., and Wates, J. F., 1989, Effeict o( meotls an olypaf conossmpnion bradiopeorectiwe thoola e ts v.. Fvarieoo of Raiiiieaag Boia. PwscritAq, of the 21Asieeul MeraoWq/ ofith Farooot -w Seatu for Radeatsco Rmilacy edited by E. Ralua(V(H. WeinhtimtDoetfield Resoch, FL. and Balaban, RehavocPhilsodelpuats (in

Kt HAm. K. S., SANcwO, A. M., and Wuites. ). F.. 1986j A novel intevrsciuon of Jweihvl-dithicocairbscate with the glutathione glutathione p"'roxsaie system. liwdrremeuiJournal 4 Radevswe(on oorwv. Reolnqv, P~vtoors, 12. 1463 1467.

Ki %A*, K. S., VAIiHnwv Y. N., and Wiiiii. J F., 19&M' adioprowtxe by mntioiumlantenzymes wid ensy.me, mimetics. &~r.ret Wd teeatm~rs, 39. 301 309lm

LANDAUII, M. R , I)Avis. 11. 1). DipuwiaNre, ) A., and Wpietam. V F. 19017. Door and time,relationships of the raditwprotectir WR-2721 o n kwiototnsm c-ty in mice. Phev,....eoohiy. Bwevwet aned Mokoer". 27, 373 576

I.Aswx.~I3 M. R. J)~vua H. D)., Kt'0An. K 14 . arid Wif* 3 F, 19)9s, Cafftne mmasale.the beha'. oral tnwicity 4f the radionwitectcur WM.-)M9. AAirevmw, olf Papport fev ths,

22id luied Ifuuueof the Kaiwnfuv sNw,,ro few Rmbgatoi oraitmvn Hrossife, p. 146I..ioAepU, M. 14. WAI..oeN T L. 0-d DAVIS. H. I . 19119b, NehaVoCWal efekt C-4

radwopenitective Write in mi*c uimliminalion al' 1R-2721 end 16, IS domethvlproawglanint. Froa tier, of Radiatteuu ReoofWy. Praorft-,Aos of thu, 21st Awasoal

Mveq~lo vI thr ew uopt - Sirwiot for PauAwtat Rmooqv. edited by F. Rilthe (VOI.Weinheim" leerfloold Rearh. Fl. and Ueabson. Rcrbu'tA PhuladelphialI(in press)

LIMPU I., (.,s I if, A.. Aa.ut). MI ,Aem r ( . "' ho"0". I)- . Okra:1 1', R. J., andloliy-,v. XI k -". t'N* N'rlhutsiriit% r rtlne 1,-'ntrvlruhin- I tilloing cotoeuiretwo cheiwiditotherspirv rivectiml 4 Ro,*tu.. Ior.ti. edited by 1) Hriwne, j3Weiss.. T j3la ~i .::tiend MI V Pillo I Plentim, Nroo YVo) (In preoisj

irH"Nit . S . 19J79. Radt~ormeuuc..urn tif nmier iflte~tifl hv inhihittrs (If O CIk ANMPphoefdese e , Iervaatm.o Jnairooomo of Naduetioa. Oarclogy. Rwdt~v. Pht'ure 5.925 9.13,

NIAC..'4 j R , and 444'Q.Z %fI. 19'74. .1uct. IanamieArkv~oParosesrileovanud Iwrtheeepeaiw iffire-Cam( ot Radwolvrtrolr .4g"t. edited by . %I llstqIlseritio"Vr. \ew Yourk), M 1% 19

NIosi'. J N4 . Kl%11iui.JAiI*A, Ai ,n NI',tW M Wll1. G 19M, Piyset.hartdirs induuie?aduipri'tttuou ofuriw horMuupwiori tem crita and oncrease the 1.1 SOP .1) dia

Rad.v,'. Rwa~h. 44,276 2111MlssMgue j IR . NIArTVI w'.. , 1!; tnooeww- NIAijtls riui A !and V.A..t two Pkitso,' 1968,

Redutiuon of .oilot- end Itos'.g-term radoetuim 10+1111111' bV miflUts (If 4hemkialjPrvetrs. Roadsea. 11"irw~k 31, A 44

%I Niii ,.u. NI J% 1. 14i RNI , and Stv.anow;J 1974. FKftt (of %,itsonior F foo post.irredisevoon death in mete Kirorentia, M. 121h, 1217

NtAt f)A. A . 1~'., ) 1, . K, umni. V . ArP~iI A, F- Orppe',eitic J and %14r%4AIIA, K , I 965. Inktmtsuun of motu onral onaopenses superns.de dimnutater by

* ierleukin. I P'.SEH JuwaIf 2. 1107 1091N IT A. R4. I 96N, Role a!4 tvtnhone in radocipereutorn Pteirnosrwiv deed Teipviewo,

251 2M6\@eT. 4 Ki m'..K t ,and Wsi.) F . 1990 F han.cmuntuof *unuIIa.rradiated micetwv

treatment '..uth interculoin. If [L. I 1Ahwinran iniownt.natwurowith W14.2721 .As~rooritq h 17th ?4 fun.! viflusof th 'swiMalm"..uHvuw Smwuvt% Snuvrle 14'.4. p I xq\P~t, 4 ,OPP~u4CI4,JJ.oand I)"' rs.i D I. PAN5. I ntorendem.v uf ther redimnsnottrr-

(((Iduny -otumultcone fact'w. and mnurtrue tr'mbiriant 'agoanulmy- tor-mak riupfoset mohuny.v

O~e'Nin~te 3 . it t 14, Iuwm~wei%.P . (.uimA, 14 , KI'.'.v. J . and LiunAu'(. I) .119694ntitulukin I enhatmes surki'.aI ,tholkl' irradiated mote treated otith alligerus..uitwmearrow tell Mand 4, 22 17 221

Ptrus'. NIM %11 . 'I) YoNPA'Eif-u1111 , CH(NurnuXI, %I A , an- Wsis. I F , I "A,Hadliu'vrrse-tmnn byv huidulg of re "Ia.oiers slorice and in (NIsbinafion with WR.2 2 1 IAa'mod'iqv end 7hevaperaftr 39.,247 2 %4

722

PAVNUN, NO. L.. MAcVinTir, T. .,and Jm-Kwr4, W. E.. 1919, Pastirradistion glucanadimnitration enhan~ces the radsoprotective effects, of WR.2721. Rawkwataio Rwairch,117. 5"-9.

PAWWV4. M. L.. %4A'%'srrw.,, T. J., and Waiss, J. F,. 19111, Combined modality radio.protection: The use of glucnn and selenium with WR-272t.Iuwao. oraloRadetao Ouo/4 ologhy, Physi (In press).

Pnmti. A.. 1 987, Role of superoxide dismutase in modificatuion of radiation injury. BritshJournul ofCancer, 55 (Suppl. VII11), 87--95.

PvUMME, J. W.. INHAW.11, F. R., *-it'oirwm, H., and LAWN. ii, J. L.. 1M1, Interaction ofcultured mammalvan cells withWR-2721 and its thWo WR-1065: implications formechanisms 4f radioprotectin, International Journal of Rd,*at.. Bialoy, 43w517-527.

Rtm.,., E., 1913. DNA repair as a probe of rsdaosensitivity and radioproection. Radio-p..Ierhwi sd,4starugm edited by' 0, F. Nygaard and NO. G. Simic (AcademicPres. New Yorkt), pp. 363 3W.

Wimui.y, A. V., PAZ-Hot-CA,) L . l iaHuars J. V., KAaAAHimAA, T., DMeijex, L. (;AMW4l.a.. and SA"Pon,,I. NO., 19117, Pirrtective effects of analogs of luteinizing hormone.releasing hormone againat X-ruodition..nduced damage in rts". Ihorweio of lA,Noionalj Ariiiu of .eairn. USA. K4,85 1- 47.

.411P-rem, F_. RVerriRA, Gi., PAtJAWVN, J.. TRaAMMIatt F.. WlrINXWVIG, J., I)am#rTitivs A. A.,end 1 4aoa . M., M94, Morbidity and mortality reduction by supplementalvitamin A or P-carotcne in ('HA mice given total body radiation. Jourinal q/ (heNational arwve Imieiiui,. 73, 1107 1177.

10i4(.a' A,, and Sim;, K. 19962, T1ime-scale and nature of radiaeronn-bologicsl damageapproache to radiaion protetionw and "oirradiation therapy. Progress to BiopAvuct-and Maiforuae Reilocy, 39, 64 199M.

t4-rwui., L, K.. WAiviue%, 1. L.., IIiienr, if. %., and JwuCsoi W. V. 1963, Protction (If (1i141against mted isson neutron--1 in:-,, 1:1) irradiation hby W14-2721, I1i,1A-dimeihyIP(*I 1, and the combinatiov of boith agents, Radiation Revewrkh I IS, OVIS 64.

S4rA%vmx I., fH, and tANWA. A.. 1976, Sivnee;tstic effect of radooproaective substancesha% ing different mechanism* of action. M*fotj3ta of Rfadroviasttity o/ 11toltogivolSvsenste (International Atomic Energy Agency, Viennol. pp. 47 59,

Lits I)Vy i, P, and ITwiiaAs, Ht., 1 99X9, Hone marrow* cell protection and sodiitkation of drug. oxctyt combination of protector. faraolg and rhiepreft. 39, 213 214.

V'ttsit4tAt' V N._ Ki ttAm, K, '4. J'se L F.. and \sPiA. R.. 19019, Induction of superoxidedismoiaaer: A me, hanism for radutoprotectami by interleukin.? (IL-I)? .Aardarts ofJ'()triv /,,r 1Ar 7t .UInu lao A'lrriing of /IV Radeaow Rotearrh Soroev. Spoeatle, W.-#,p 110. I

Wuk inse, T,' L., 191,7 A paradoxtical role for ricoastnids: Radtopotertan'ta and radiowseit-iae". Prostaglan da d i~ .i.f'sanaa n RdAdte,, Inpary. edited hby 1' L. Walden'and H. N. Hiughes (Penum. Ney% York,. pp. 203 27 1.

Wkviewa. T L., P,%rcwc NU 31, . and mwiu~rr, *r ., 191411, I.,ukotriene-inducedradiop-Pection of hemnatenpoitic stem cells in mice. RadwseAuo Reore*. 113,119 195

WAeIMN, T1 L., PATCyrwpN. M., and S'vio, 14, I_, 1"I1, 16.16-lDieetbyl prostaglandin Ftincreases survi-sl in mice follo-Aina irradiation. Rodiwtow Roworte h., 1", 4404k44.

Ieviss J. F . and Ki ttu. 'K. 1% . 19149,. Anioeidani mechanism in radiation injur-, andradioprewectiont. (,ilmlow -Aniftvda,gIens.,franss Vol. 11. edited by C.'K,(hopw W(RCPress, (9lantn), p., 163 11M9

Wrimia, J F, l4txwua, It. I_ and ki-twN K. *. 19117, Selenium pretreatment enhances theradsaiprofttive effect and reduces the lethal towiity of WR-2721- Free Ri4i&elResearch Coaeamaaont. 3. 3 3 Mtf.

yvit~s, J, M,, Pmui-ia, J. (9 and 'oairw, L. 'if . 1971. tlome Pharivaolosi: efectsof WkI272?: T~he'ir role in toiity' and rsdioproetw Radation Raswarch, 4, 222 233.

YttHAP, J' ML1, and *sioaena, J, H. 1969, Chemv~e.ottinn against three modes of radiationdeath in the mouse. ,f ateroesoia JoseneWlof Raetwo Roology, 15, 233 237.