Androstenediol Stimulates Myelopoiesis and Enhances Resistance to Infection in Gamma-Irradiated Mice Mark H

International Journal of Immunopharmacology 22 (2000) 1±14 www.elsevier.com/locate/ijimmpharm

Androstenediol stimulates myelopoiesis and enhances resistance to infection in gamma-irradiated mice Mark H. Whitnall a,*, Thomas B. Elliott a, Rita A. Harding a, Cynthia E. Inal a, Michael R. Landauer a, Catherine L. Wilhelmsen a, LuAnn McKinney a, Venita L. Miner a, William E. Jackson III a, Roger M. Loria b, G. David Ledney a, Thomas M. Seed a

aRadiation Casualty Management Team, Armed Forces Radiobiology Research Institute, Bethesda, MD 20889, USA bVirginia Commonwealth University's Medical College of Virginia, Richmond, VA 23298, USA

Abstract

The ionizing radiation-induced hemopoietic syndrome is characterized by defects in immune function and increased mortality due to infections and hemorrhage. Since the steroid 5-androstene-3b,17b-diol (5-androstenediol, AED) modulates cytokine expression and increases resistance to bacterial and viral infections in rodents, we tested its ability to promote survival after whole-body ionizing radiation in mice. In unirradiated female B6D2F1 mice, sc AED elevated numbers of circulating neutrophils and platelets and induced proliferation of neutrophil progenitors in bone marrow. In mice exposed to whole-body 60Co gamma-radiation (3 Gy), AED injected 1 h later ameliorated radiation-induced decreases in circulating neutrophils and platelets and marrow granulocyte-macrophage colony- forming cells, but had no eect on total numbers of circulating lymphocytes or erythrocytes. In mice irradiated (0, 1 or 3 Gy) and inoculated four days later with Klebsiella pneumoniae, AED injected 2 h after irradiation enhanced 30- d survival. Injecting AED 24 h before irradiation or 2 h after irradiation increased survival to approximately the same extent. In K. pneumoniae-inoculated mice (irradiated at 3±7 Gy) and uninoculated mice (irradiated at 8±12 Gy), AED (160 mg/kg) injected 24 h before irradiation signi®cantly promoted survival with dose reduction factors (DRFs) of 1.18 and 1.26, respectively. 5-Androstene-3b-ol-17-one (dehydroepiandrosterone, DHEA) was markedly less ecacious than AED in augmenting survival, indicating speci®city. These results demonstrate for the ®rst time that a DHEA-related steroid stimulates myelopoiesis, and ameliorates neutropenia and thrombocytopenia and enhances resistance to infection after exposure of animals to ionizing radiation. # 2000 International Society for Immunopharmacology. Published by Elsevier Science Ltd. All rights reserved.

* Corresponding author. Tel.: +1-301-295-9262; fax: +1- 707-922-0011. E-mail address: [email protected] (M.H. Whit- nall).

0192-0561/00/$20.00 # 2000 International Society for Immunopharmacology. Published by Elsevier Science Ltd. All rights reserved. PII: S0192-0561(99)00059-4 2 M.H. Whitnall et al. / International Journal of Immunopharmacology 22 (2000) 1±14

Keywords: Androst-5-ene-3 beta, 17 beta-diol; Prasterone; Hematopoiesis; Neutropenia; Experimental radiation injuries; Ionizing radiation; Gram-negative bacterial infections; Klebsiella pneumoniae

1. Introduction countermeasure to promote survival during the early phase of the hemopoietic syndrome are to Acute exposure to elevated doses of ionizing increase production of neutrophils and platelets radiation causes defects in hemopoiesis, resulting [32]. Very little is known about the eects of this in low numbers of circulating blood cells and pla- family of steroids on myelopoiesis and thrombo- telets and increased susceptibility to infection and cytopoiesis. One report indicated a slight inhibi- hemorrhage [1±3]. Past eorts to stimulate hemo- tory eect of 7±10 days of dietary DHEA poiesis in myelosuppressed animals have involved administration on granulopoiesis in irradiated administration of components of microbial cell mice [38]. AED increased formation of granulo- walls or their synthetic analogs [4±8], and natural mas during infection with Mycobacterium tuber- factors such as cytokines, prostaglandins, and culosis in mice [27]. In view of the ecacy of peptides or their synthetic analogs [9±12]. It AED and DHEA in promoting resistance to would be advantageous to develop a small-mol- infection in unirradiated animals, the limited in- ecule, nontoxic pharmacological agent that formation available on the eects of these ster- would ameliorate hemopoietic radiation injury. oids on myelopoiesis and thrombocytopoiesis, From the point of view of low toxicity, ease of and their modulation of expression of cytokines storage and administration, steroids would make known to be radioprotective, we tested their abil- attractive candidates. ity to enhance survival and reduce hemopoietic Administration of the adrenocortical steroid 5- injury after whole-body irradiation in mice. androstene-3b-ol-17-one (dehydroepiandrosterone, DHEA) or its metabolite 5-androstene- 3b,17b-diol (5-androstenediol, AED) to rodents 2. Experimental procedures results in increased immunocompetence and greater survival during infection [13±23]. AED is 2.1. Mice more potent than DHEA in promoting survival during bacterial or viral infection in rodents All studies were carried out in accordance with [14,15]. The DHEA-related steroids preserve the principles and procedures of the National immunocompetence after thermal injury [17] and Research Council Guide for the Care and Use of during aging [24]. They also modulate expression Laboratory Animals [71]. B6D2F1/J female mice of cytokines such as interleukin-1 (IL-1), IL-3, (Jackson Laboratory, Bar Harbor, ME), 18±24 IL-6, tumor necrosis factor (TNF) and inter- weeks of age, 22±30 g body weight, were held in feron-gamma (IFNg) [17,22,25±34]. These cyto- quarantine for two weeks. Mice were chosen for kines are known to aect recovery from these studies because of the similar responses of radiation-induced hemopoietic injury [32±34]. murine and human hemopoiesis to drugs and One report indicated small radioprotective or toxic insults [39]. Up to 10 mice were housed in radiosensitizing eects of ip DHEA pretreat- sanitized 46 24 15 cm polycarbonate boxes ments, depending on the interval between treat- with ®lter coversÂ Â (MicroIsolator; Lab Products, ment and irradiation [35]. Inc, Maywood, NJ) on autoclaved hardwood Although the DHEA-related steroids can chip bedding in a facility accredited by the As- stimulate proliferation or activation of lympho- sociation for Assessment and Accreditation of cytes, natural killer cells and macrophages Laboratory Animal Care International. Mice [16,20,22,24,31,36,37], the primary goals of a were given feed and acidi®ed (pH 2.5) water M.H. Whitnall et al. / International Journal of Immunopharmacology 22 (2000) 1±14 3 freely. The animal holding room was maintained 548 EtOH) and DHEA (melting point 149± À with conditioned fresh air that was changed at 1518C, rotation +138 EtOH) were purchased least 10 times per h at approximately 218C and from Steraloids (Wilton, NH). PEG-400 was 508 (210%) relative humidity and with a 12-h obtained from Sigma (St Louis, MO). PEG-400 light/dark full spectrum lighting cycle. All has been recommended as a nontoxic vehicle for research was approved by the Institutional Ani- administration of steroids, based on histopatho- mal Care and Use Committee of the Armed logical analysis of sc injection sites [42]. Forces Radiobiology Research Institute (AFRRI). 2.4. Peripheral blood element counts

2.2. Irradiation Blood (0.6±1.0 ml) was obtained from halothane-anesthetized mice by cardiac puncture Mice were placed in ventilated Plexiglas con- using a heparinized syringe attached to a 21- tainers and exposed bilaterally to gamma-radi- gauge needle. Blood was collected in EDTA-con- ation from the AFRRI 60Co source as described taining sample tubes. Mice were euthanized by previously [40]. Exposure time was adjusted so cervical dislocation after blood collection. White that each animal received a midline tissue- blood cell (WBC), red blood cell (RBC) and pla- absorbed dose of 1±12 Gy at a nominal dose rate telet (PLT) counts were performed using a of 0.4 Gy/min at ambient temperature. Using a Hematology System 9000 (Biochem Immunosys- standardized technique, the midline dose rate was tems). Wright-stained blood smears from the measured by placing a 0.5 cc tissue-equivalent same samples were made for dierential counts ionization chamber at the center of a 2.5-cm di- of neutrophils and lymphocytes by light mi- ameter cylindrical acrylic mouse phantom. The croscopy. calibration factor for the ionization chamber was calculated by means of a standard obtained from 2.5. Granulocyte-macrophage colony-forming cell an Accredited Dosimetry Calibration Laboratory (GM-CFC) assay (M.D. Anderson Cancer Center, Houston, TX) certi®ed by the National Institute of Standards Hemopoietic progenitor cells committed to and Technology. The tissue±air ratio, de®ned as granulocyte-macrophage dierentiation (GM- the ratio of the dose rate measured in the phan- CFC) were assayed by a single-layer modi®cation tom to the dose rate in free air, for this array of a double-layer semisolid agar technique was 0.96. Variation within the exposure ®eld was described by Patchen et al. [6]. Brie¯y, femoral less than 4% (unpublished results). Dosimetric marrow was extracted and cell suspensions were measurements were made in accordance with the prepared by ¯ushing with 3 ml of McCoy's 5A American Association of Physicists in Medicine medium containing 10% heat-inactivated fetal protocol for the determination of absorbed dose bovine serum (HIFBS; Hyclone, Logan, UT). from high-energy photon and electron beams Each cell suspension represented a pool of mar- [41]. Sham-irradiated mice were treated in exactly row from four femurs, i.e., both femurs from the same way as the irradiated animals, except each of two mice. The total number of nucleated that the 60Co rods were not raised from the cells in each suspension was determined with a shielding water pool. Coulter counter. The agar-medium mixture con- sisted of equal volumes of 0.66% agar and 2.3. Steroids double-strength supplemented CMRL 1066 medium (Gibco, Grand Island, NY). The medium AED, DHEA, vehicle (PEG-400) or sterile was supplemented with ®nal concentrations of 0.9% NaCl (saline) was injected sc in a volume 10% HIFBS, 5% tryptic soy broth, 5% heat- of 0.1 ml. AED doses were 40, 80, 160 or 320 mg/ inactivated horse serum, antibiotics, and L-serine. kg. AED (melting point 179±1818C, rotation One millilitre of the agar-medium mixture was 4 M.H. Whitnall et al. / International Journal of Immunopharmacology 22 (2000) 1±14 added to each 35-mm plastic Petri dish (two injected sc with K. pneumoniae four days after dishes per suspension) and mixed with 50 ml of sham-irradiation or irradiation when circulating 0.1 ng/ml recombinant mouse GM-CSF (Gen- leukocytes are depressed [1]. Mice were inocu- zyme, Cambridge, MA). Cell suspensions were lated sc rather than iv or ip, because the aim was then mixed into the agar-medium mixture to a to establish infection leading to sepsis, not rapid ®nal concentration of 0.5 105 cells/ml for uni- septic shock. After sc inoculations of K. pneumo- rradiated animals, and 1.0Â 105 or 1.5 105 niae in these mice, the infection remains largely cells/ml for irradiated animalsÂ to ensure sucientÂ localized to the injection site. We have found pre- colonies per plate for quantitation. Control ex- viously that K. pneumoniae are not detectable in periments were done to con®rm linearity of colo- blood of inoculated mice until a few hours before nies at cell concentrations of 0.5±1.5 105 cells/ death (unpublished data). ml. Colonies (>50 cells) were countedÂ after Dierent doses of the bacteria were inoculated seven days incubation in a 378C humidi®ed en- for each of three radiation dose levels (0, 1 or 3 vironment containing 5% CO2. The average of Gy) to approximate the LD95/30, because the the counts for the two dishes was taken as the eects of radiation on hemopoiesis and suscepti- value for each pool. Six animals were used per bility to infection are dependent on the dose of group in each of two experiments. radiation [1,2]. The LD95/30 for bacteria at each radiation dose was calculated from probit analy- 2.6. Histology of bone marrow sis of previous experiments in our laboratory (unpublished). The target LD95/30 doses in col- For histological examination of myeloid hyper- ony-forming units of bacteria (CFU) and their plasia in bone marrow after injection of AED, 95% con®dence intervals were: 0 Gy, 2.3 107 mice were euthanized with halothane, tissues (1.4 107±4.4 107); 1 Gy, 1.1 107 (0.61Â were immersed in formalin, bones were decalci- 107±2.5Â 107);Â 3 Gy, 6.6 106 (2.4Â 106±7.8 Â 6 Â Â Â Â ®ed and routine H&E-stained 6-mm paran sec- 10 ). The actual doses were estimated by dilution tions were prepared. plating of inocula onto Trypticase Soy Agar and incubating overnight at 358C. The actual doses 2.7. Infection with K. pneumoniae (mean of triplicate spreads) were as follows: 0 Gy, 1.87 107; 1 Gy, 1.13 107; 3 Gy, 4.02 For induced-infection studies, a clinical isolate 106. SinceÂ dierent bacterialÂ doses were given forÂ of K. pneumoniae, capsule type 5 (strain AFRRI dierent radiation doses, and it was impossible 7), that was kept frozen at 708C in skim milk, to deliver a precise LD95/30 to each group, dier- was grown overnight at 358C on Trypticase Soy ent mortality rates were observed in the three ve- Agar (Becton±Dickinson, Sparks, MD). Five hicle-injected control groups (see Section 3). typical colonies were inoculated into 8 ml of Bacterial doses for the other induced-infection brain heart infusion broth (Becton±Dickinson) experiments were prepared and calculated in the and incubated overnight at 358C. Two milliliters same manner and are indicated in the text and of this overnight suspension was inoculated into ®gure legends. 95 ml of prewarmed brain heart infusion broth. The culture was incubated at 358C with shaking 2.8. Survival measurement for approximately 2.5 h. The optical density of bacterial growth was monitored with a spectro- Animals were checked twice daily, seven days photometer at a wavelength of 550 nm. Late log- per week, to monitor survival and to euthanize phase cells were washed and suspended in cold mice that were in a moribund state. To verify saline to yield 109 viable bacteria per ml. Appro- that mortality in the induced-infection exper- priate dilutions for inoculations were made in iments was associated with K. pneumoniae injec- cold saline. tion, heart blood from recently deceased animals To induce a bacterial infection, all mice were (or moribund animals euthanized by cervical dis- M.H. Whitnall et al. / International Journal of Immunopharmacology 22 (2000) 1±14 5 location) was cultured overnight at 358C on with a randomization t-test to obtain exact P- Columbia sheep blood agar plates (Becton±Dick- values. inson, Sparks, MD). Colonies were identi®ed as Thirty-day survival values were compared K. pneumoniae by Biolog analysis. using the generalized Savage (Mantel±Cox) pro- cedure (BMDP Statistical Software, Inc, Los 2.9. Statistical evaluation Angeles, CA). To calculate dose reduction factors (DRFs), probit analysis was performed on mor- Signi®cant eects of radiation and drug treat- tality data. ment on blood element and GM-CFC colony counts were evaluated using Analysis of Variance (ANOVA). For comparison of individual groups, 3. Results one-way ANOVA was followed by the Newman± Keuls test. Dierences were considered signi®cant 3.1. Eects of AED on non-irradiated mice if P < 0.05. For the dose-response study, regression analysis was used to assess whether Injection of the PEG-400 vehicle sc had no there was a signi®cant relationship between dose eect on any of the measured hematologic par- and eect. For some blood cell counts in irra- ameters (peripheral blood element counts or diated and sham-irradiated mice, log transform- GM-CFC) in comparison to saline injection four, ations of the data were made before ANOVA to seven, 10, 14 or 18 days after injection in unirra- attain approximate normality and equality of diated or irradiated mice (not shown). variance. In these cases, data are reported as geo- When AED was injected into naive mice (not metric means and standard errors on semilog subjected to irradiation or sham-irradiation), a plots. All other hematological data are reported dose-dependent increase (r = 0.977, P = 0.002) in as arithmetic means and standard errors on lin- circulating neutrophils was observed four days ear plots. after injection (Fig. 1). The maximum eect For histological analysis of bone marrow, (112% increase compared to vehicle) was coded slides were scored blind using a ®ve-level observed with a dose of 320 mg/kg. Levels of cir- semiquantitative scale and the results analyzed culating neutrophils had returned to control

Fig. 1. Circulating neutrophils measured four days and platelets 14 days after sc injection of 40, 80, 160 or 320 mg/kg AED or vehicle (PEG-400) into unirradiated female B6D2F1/J mice. Regression analysis indicated dose-dependent increases in neutrophils (r = 0.977, P = 0.002) and platelets (r = 0.935, P = 0.01). Each point represents the mean2SEM of eight animals. 6 M.H. Whitnall et al. / International Journal of Immunopharmacology 22 (2000) 1±14 values 10 and 14 days after treatment (cf. sham- into naive mice (Fig. 2). A blind semiquantitative irradiated mice in Fig. 3). analysis of myeloid hyperplasia (four animals per A dose-dependent increase in platelets (r = group) revealed a signi®cant shift in the myeloid- 0.935, P = 0.01) was observed 14 days after to-erythroid ratio between vehicle-treated and injection of AED into naive mice (Fig. 1). The AED-injected mice at four and 10 days (P = maximum eect (20% increase compared to ve- 0.014), but not at 14 days (P = 0.14). hicle) was again observed at the 320 mg/kg dose. No signi®cant dierences in circulating platelets 3.2. Eects of AED on irradiated mice between AED-treated and control mice were observed four and 10 days after treatment (cf. To test the ability of AED to ameliorate radi- sham-irradiated mice in Fig. 5). No eect on ation-induced defects in hemopoiesis, mice were lymphocytes or red blood cells was observed at exposed to bilateral whole-body gamma-radiation any time after injection of AED into naive mice and received a dose of 3 Gy (or were sham-irra- (not shown). diated). One hour after irradiation or sham-ir- Histological examination of bone marrow radiation, mice were injected with 320 mg/kg showed marked proliferation of immature my- AED or PEG-400 vehicle. All between-group eloid elements after injection of 320 mg/kg AED dierences in blood cell elements (i.e., neutro-

Fig. 2. Myeloid hyperplasia in femur four days after sc AED (320 mg/kg, right) compared with vehicle (left). Note the relative increase in the proportion of myeloid to erythroid precursors after AED injection. H&E-stained paran sections; original magni®- cation 700 . Â M.H. Whitnall et al. / International Journal of Immunopharmacology 22 (2000) 1±14 7

Fig. 4. Femoral marrow GM-CFC at various times after ir- Fig. 3. Circulating neutrophils at various times after irradiation (3 Gy) or sham-irradiation. AED (320 mg/kg) or ve- radiation (3 Gy) or sham-irradiation. AED (320 mg/kg) or vehicle (PEG-400) was injected sc 1 h after irradiation or sham- hicle (PEG-400) was injected sc 1 h after irradiation or sham- irradiation. ANOVA and Newman±Keuls analysis indicated irradiation. ANOVA and Newman±Keuls analysis indicated signi®cant eects of AED compared to vehicle seven, 10 and signi®cant eects of AED compared to vehicle four days after 14 days after irradiation (P < 0.05). Each point represents the sham-irradiation and four and seven days after irradiation (P mean2SEM of three pools of marrow. Each pool comprised < 0.05). Each point represents the mean 2 SEM of 6±8 ani- marrow from four femurs (two mice). mals.

numbers were reduced by 33% and 43%, re- phils, GM-CFC and platelets) reported below spectively, and had recovered to normal after (shown in Figs. 3±5) were signi®cant (P < 0.05, 18 days, compared to sham-irradiated mice ANOVA and Newman±Keuls test). (Fig. 5). In AED-treated irradiated mice, plate- Irradiation resulted in a 78% decrease in neutrophils at four days (compared to sham- irradiated animals at the same time point), with recovery to normal by 10 days (Fig. 3). AED signi®cantly attenuated the radiation- induced neutropenia: in irradiated mice, AED induced a 150% increase in neutrophils (compared to vehicle) at four days (Fig. 3). In sham-irradiated mice, AED caused a 70% increase in neutrophils above vehicle control at four days and had little or no eect at other times (Fig. 3). GM-CFC were reduced by approximately 50% four days after irradiation compared to sham-irradiated mice, 75% at seven and 10 days and 45% at 14 days, with full recovery by 18 days (Fig. 4). AED induced an approxi- Fig. 5. Circulating platelets at various times after irradiation mate doubling of GM-CFC in irradiated mice (3 Gy) or sham-irradiation. AED (320 mg/kg) or vehicle compared to vehicle control at 7±14 days, (PEG-400) was injected sc 1 h after irradiation or sham-ir- with full recovery by 14 days (Fig. 4). No radiation. ANOVA and Newman±Keuls analysis indicated signi®cant eects of AED compared to vehicle 10, 14 and 18 eect of AED on GM-CFC was observed in days after irradiation and 14 and 18 days after sham-ir- sham-irradiated animals (Fig. 4). radiation (P < 0.05). Each point represents the mean2SEM Seven and 10 days after irradiation, platelet of 6±8 mice. 8 M.H. Whitnall et al. / International Journal of Immunopharmacology 22 (2000) 1±14 let numbers fully recovered after 10 days (Fig. Gy. One hour later, all mice were injected sc with 5). either PEG-400 vehicle or 320 mg/kg AED. Four After irradiation in vehicle-treated mice, circu- days later, all mice were injected sc with an ap- lating lymphocytes were reduced by 80% at four proximate LD95/30 of K. pneumoniae. days, and were still 62% below the sham-irra- At 30 days after K. pneumoniae inoculation (34 diated baseline at 18 days (not shown). In con- days after irradiation), the percentages of animals trast to the eects observed on neutrophils and surviving in each group were as follows: 0 Gy± GM-CFC, AED had no eect on numbers of Vehicle, 0%; 0 Gy±AED, 65%; 1 Gy±Vehicle, lymphocytes in sham-irradiated or irradiated 5%; 1 Gy±AED, 95%; 3 Gy±Vehicle, 15%; 3 mice (not shown). Irradiation reduced erythro- Gy±AED, 90%. The dierences in mortality cyte numbers in vehicle-treated animals by 12% between the vehicle-injected groups were due to at 4 days and 6% at 14 days (not shown). AED the fact that the dierent doses of bacteria deliv- had no eect on this parameter (not shown). ered in each radiation condition could only ap- To determine the eect of time of AED admin- proximate the LD95/30 (see Section 2). The istration on neutropenia after irradiation, AED increases in survival in the AED-treated groups or vehicle was administered 24 h before, 2 h compared to the vehicle-treated groups were sig- before, 2 h after, or 24 h after whole-body ni®cant (P < 0.001). The time course of mortality gamma-irradiation and circulating neutrophils for the 3-Gy groups is shown in Fig. 7. were counted four days after irradiation. The amelioration of radiation-induced neutropenia 3.3.2. Irradiation at 3 Gy followed by LD100/30 K. was similar in all four groups (Fig. 6). pneumoniae: comparison of AED and DHEA The speci®city of AED's survival-enhancing 3.3. Eects of AED on resistance to infection eects was assessed by comparing it to its parent steroid, DHEA. AED diers from DHEA only 3.3.1. Irradiation at 0, 1 or 3 Gy followed by by the presence of a 17-hydroxyl group in place LD95/30 K. pneumoniae of the 17-ketone group. A dose of 80, 160 or To directly test the eect of AED on resistance 320 mg/kg of either steroid (or vehicle) was to infection, mice were exposed to whole-body injected 2 h after 3-Gy whole-body gamma ir- gamma-radiation at doses of 0 (sham), 1 or 3 radiation. Four days after irradiation, all mice

Fig. 6. Circulating neutrophils four days after irradiation (3 Fig. 7. Survival of mice exposed to whole-body gamma-radi- Gy). AED (320 mg/kg) or vehicle (PEG-400) was injected sc ation (3 Gy), injected sc with AED (320 mg/kg) 1 h later, and at various times before or after sham-irradiation. Mean 2 inoculated with K. pneumoniae (4.0 106 CFU) four days Â SEM of eight mice per group indicated. later. N = 20 mice per group. M.H. Whitnall et al. / International Journal of Immunopharmacology 22 (2000) 1±14 9 were injected with an approximate LD100/30 of K. pneumoniae (5.6 106 CFU). At 30 days after K. pneumoniae inoculationÂ (34 days after irradiation), the percentages of animals surviving in each group were as follows: AED 320 mg/kg, 30%; DHEA 320 mg/kg, 0%; AED 160 mg/kg, 50%; DHEA 160 mg/kg, 5%; AED 80 mg/kg, 20%; DHEA 80 mg/kg, 10%; Vehicle, 0%. The time course of mortality for the 160 mg/kg groups is shown in Fig. 8.

3.3.3. Irradiation at 3 Gy followed by LD100/30 K. pneumoniae: comparison of AED administration times To determine the eect of time of AED admin- Fig. 9. Survival of mice exposed to whole-body gamma-radi- istration relative to irradiation on its ability to ation (3 Gy), injected sc with AED (240 mg/kg) at various enhance resistance to gram-negative bacterial times relative to irradiation, and inoculated with K. pneumoniae (5.9 106 CFU) four days after irradiation. N = 16 mice infection, AED (240 mg/kg) or vehicle was admi- Â per group. nistered 14 d, 7 d, 3 d, or 24 h, respectively before or 2 h after 3-Gy whole-body gamma ir- 3.3.4. Radiation dose reduction factor (DRF) in radiation. Four days after irradiation, all mice mice inoculated with bacteria were injected with an approximate LD of K. 100/30 In order to calculate a radiation DRF for pneumoniae (5.9 106 CFU). Only the groups AED in mice inoculated with bacteria, mice were treated 24 h beforeÂ or 2 h after irradiation exposed to whole-body gamma irradiation and showed signi®cant improvement in survival com- received doses ranging between 3 and 7 Gy. All pared to vehicle (Fig. 9). mice were inoculated with K. pneumoniae four

Fig. 10. Probit mortality curves of mice exposed to whole- Fig. 8. Survival of mice exposed to whole-body gamma-radi- body gamma-radiation and inoculated with K. pneumoniae ation (3 Gy), injected sc with 160 mg/kg AED, 160 mg/kg (5.5 104 CFU) four days later. Mice were injected sc with Â DHEA or vehicle 2 h later, and inoculated with K. pneumo- AED (160 mg/kg) or vehicle 24 h before irradiation. DRF niae (5.6 106 CFU) four days after irradiation. N = 20 mice calculated from ratio of LD of AED-treated to vehicle- Â 50/30 per group. treated mice. N = 10 mice per group. 10 M.H. Whitnall et al. / International Journal of Immunopharmacology 22 (2000) 1±14 days later (5.52 104 CFU). AED (160 mg/kg) radiation quality [43±45]. The primary cause of or vehicle was injectedÂ 24 h before irradiation. mortality during the early phase of the radiation- Mortality for the two treatments is plotted in induced hemopoietic syndrome is sepsis resulting Fig. 10. The probit line for AED-treated mice from opportunistic infection due to low numbers was shifted to the right with a DRF of 1.18 of neutrophils and increased translocation of bac- (95% con®dence limits 1.10±1.28, P = 0.0008). teria across the gastrointestinal mucosa [1±5]. This is complicated by thrombocytopenia and 3.3.5. DRF with no bacterial inoculation concomitant hemorrhage and defects in the adap- The survival-enhancing ecacy of AED was tive immune system resulting from apoptosis of then explored with a more stringent hemopoietic lymphocytes and de®cient lymphopoiesis [1,2]. injury. Mice were exposed to whole-body The main focus of eorts to enhance survival gamma-radiation and received doses ranging during this period has been to ®nd ways of sti- between 8 and 12 Gy in the absence of bacterial mulating granulopoiesis and thrombocytopoiesis challenge. AED (160 mg/kg), vehicle or saline [32±33]. was injected 24 h before irradiation. Mortality The neutropenia and thrombocytopenia for the AED and vehicle treatments is plotted in observed after ionizing radiation are due to the Fig. 11. The probit line for AED-treated mice loss of progenitor cells in hemopoietic tissue was shifted to the right compared with vehicle, capable of proliferating and developing into with a DRF of 1.26 (95% con®dence limits 1.20± mature neutrophils and platelets [2]. Maintenance 1.35, P < 0.001). The probit lines for vehicle and proliferation of these progenitor cells are (Fig. 11) and saline (not shown) were not signi®- regulated by a variety of soluble factors present cantly dierent (DRF=1.02). in the microenvironment of hemopoietic tissue as well as contact with stromal cells [46]. Cytokines that stimulate granulopoiesis and thrombocyto- 4. Discussion poiesis have therefore been used as treatments for whole-body ionizing radiation [3,32±34]. Ionizing radiation causes defective hemopoiesis Other immunomodulators that protect against as a function of radiation dose, dose-rate and hemopoietic injury do so in large part by stimulating expression of these cytokines [7]. Among the important hemopoietic cytokines are granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage (GM)-CSF, IL-1, IL-3, IL-6, IL-11, IL-12, thrombopoietin (TPO), TNF and stem cell factor (SCF) [3,32±34]. Some of these cytokines have proven to be ineective when administered systemically, or are too toxic for this application [32,34,47]. We reasoned that stimulation of local expression of cytokines in tissues by a natural circulating hormone such as AED might enhance hemopoiesis with low toxicity. DHEA-related steroids are known to initiate a cascade of cytokines, including IL-1 [17,27,31], IL-3 [36], and IL-6 [22,26,48±50]. These cytokines Fig. 11. Probit mortality curves of mice exposed to whole- synergistically stimulate myelopoiesis [32,34], body gamma-radiation with no bacterial challenge. Mice were injected sc with AED (160 mg/kg) or vehicle 24 h before ir- which supports our hypothesis that the observed radiation. DRF calculated from ratio of LD50/30 of AED-trea- improvement in survival after AED treatment ted to vehicle-treated mice. N = 10 mice per group. was partly due to increased numbers of circulat- M.H. Whitnall et al. / International Journal of Immunopharmacology 22 (2000) 1±14 11 ing neutrophils resulting from heightened myelo- rule out additional mechanisms. DHEA and poiesis. This is the ®rst report of a stimulatory DHEA metabolites have direct eects on mature eect of DHEA-related steroids on myelopoiesis. cells involved in both innate and adaptive An earlier study demonstrated strong inhibition immune responses [17,18,21,36,37,56±62]. It of lymphopoiesis and slight inhibition of granulo- should also be noted that administering AED poiesis in DHEA-fed mice exposed to sublethal three days before irradiation had no survival- radiation [38]. The discrepancy between those enhancing eect in spite of the fact that myeloid results and our ®ndings may be due to the fact hyperplasia and neutrophils were elevated four that DHEA and AED can have distinct eects days after AED injection in normal animals. on target cells [36], or to the method of adminis- Further research is necessary to determine the re- tration (long-term dietary vs single sc injection). lationships between the kinetics of AED radio- The mechanism by which circulating platelets protection, cell cycling and radioresistance of were increased by AED administration is not progenitor cells [34], and the precise timing and known. However, in view of the stimulatory mechanisms of the radiation injury that is ame- eect of AED on neutrophil progenitors, there liorated by AED. may have been a similar eect on megakaryocyte A possible mediator of in vivo eects of progenitors. Stimulation of IL-3 [36] and IL-6 DHEA-related steroids is the peroxisome prolif- [22] expression by the DHEA-related steroids erator-activated receptor (PPAR)-alpha, a mem- would be expected to enhance megakaryocytopoi- ber of the steroid/thyroid nuclear receptor esis and platelet production [32]. superfamily of ligand-activated transcription fac- Increases in circulating neutrophils can be tors [63]. Although DHEA does not activate the caused by increased movement of neutrophils receptor in transient transfection/trans-activation into blood vessels from hemopoietic reservoirs or assays [63], some of the eects of DHEA-related other tissues [51]. Although we cannot rule this steroids in vivo are dependent on the presence of out as a contributing factor to the increases in PPAR-alpha [64]. DHEA and AED may exert neutrophil counts observed after AED adminis- their eects partly via metabolites; 5-androstene- tration, we believe that AED stimulates myelo- 3b,7b,17b-triol [65] and 7a-hydroxy-DHEA [66] poiesis for the following reasons: (1) a marked have higher potency than DHEA or AED in sti- increase in myelopoiesis was observed in histo- mulating certain immune responses. Induction of logical sections of bone marrow from mice liver peroxisomes is not a factor in the present injected with AED; and (2) increases in marrow experiments, because that morphological re- GM-CFC were observed in irradiated mice when sponse has only been reported after repeated dos- mice were treated with AED after irradiation, ing of DHEA over a period of a week or longer making it unlikely that the eects of AED were [67,68]. solely due to preventing radiation injury. This Although DHEA is an adrenal androgen, the conclusion is supported by the fact that ameliora- immunostimulatory properties of its metabolite tion of radiation-induced neutropenia by AED AED probably are not related to anabolic andro- was similar whether the AED was injected 24 h gen-like properties. (1) The androgenicity of before, 2 h before, 2 h after or 24 h after ir- AED is trivial compared to testosterone or even radiation. AED could stimulate myelopoiesis in DHEA [69]. (2) The ecacy of AED in improv- part by acting on stromal cells: DHEA-related ing survival in irradiated and infected mice was steroids interact with cell lineages making up the markedly superior to that of DHEA in the pre- hemopoietic microenvironment: osteoblasts [52], sent study. (3) Testosterone produces immune de- adipocytes [53], ®broblasts [54], endothelial cells pression in vivo [70]. [55] and macrophages [21,37,56]. The present ®ndings demonstrate changes in The present study supports the hypothesis that myelopoiesis induced by AED, and suggest that AED boosts resistance to infection in part by sti- the DHEA-related steroids should be explored as mulating myelopoiesis. However, our data do not a novel type of radioprotectant and radiothera- 12 M.H. Whitnall et al. / International Journal of Immunopharmacology 22 (2000) 1±14 peutic agent. Our results indicate that AED or its and in combination with aminothiols. Adv Space Res in vivo metabolites promote survival after radi- 1992;12:(2)233±48. [7] Peterson VM, Adamovicz JJ, Elliott TB, Moore MM, ation primarily by enhancing restoration of the Madonna GS, Jackson III WE, Ledney GD, Gause WC. innate immune system, and possibly also via Gene expression of hematoregulatory cytokines is elev- eects on platelet production. The identi®cation ated endogenously after sublethal gamma irradiation and of active metabolites of DHEA and AED, their is dierentially enhanced by therapeutic administration target molecules and cells, and characterization of biologic response modi®ers. J Immunol of the changes in gene expression and cell func- 1994;153:2321±30. [8] Fedorocko P, MackovaÂ O. Combined modality radiopro- tion caused by irradiation and these steroids, will tection: enhancement of survival and hematopoietic be necessary to develop more ecacious treat- recovery in gamma-irradiated mice by the joint use of ments for radiation-induced hemopoietic injury. liposomal muramyl tripeptide phosphatidylethanolamine (MTP-PE) and indomethacin. Int J Immunopharmacol 1996;18:329±37. [9] Neta R, Oppenheim JJ, Douches SD. Interdependence of Acknowledgments the radioprotective eects of human recombinant interleukin 1a, tumor necrosis factor a, granulocyte colony- The authors thank Jason Deen, Shameek stimulating factor, and murine recombinant granulocyte- macrophage colony-stimulating factor. J Immunol Ghose, Samy Bouhaouala, Karen Quinones, 1988;140:108±11. Brian Hively, Peter Wang and Robin-Anne Fer- [10] Waddick KG, Song CW, Souza L, Uckun FM. ris for excellent technical assistance. We are Comparative analysis of the in vivo radioprotective grateful to Dr. Janis Hammer for providing a eects of recombinant granulocyte colony-stimulating preprint of a manuscript on comparison of toxic factor (G-CSF), recombinant granulocyte-macrophage CSF, and their combination. Blood 1991;77:2364±71. eects of vehicles for sc injection of steroids. [11] MacVittie TJ, Weiss JF, Browne D. Advances in the Supported in part by 5R25CA22032 and a gift treatment of radiation injuries. Tarrytown, NY: from the SCDR Cancer Research Fund. Pergamon/Elsevier Science Inc, 1996. [12] Kumar KS, Srinivasan V, Palazzolo D, Kendrick JM, Clark EP. Synergistic protection of irradiated mice by a combination of iloprost and misoprostol. In: Honn KV, References Nigam S, Marnett LJ, editors. Eicosanoids and other bioactive lipids in cancer, in¯ammation, and radiation [1] Fliedner TM, Nothdurft W, Heit H. Biological factors injury 2. New York: Plenum Press, 1997. p. 831±9. aecting the occurrence of radiation syndromes. In: [13] Ben-Nathan D, Lachmi B, Lustig S, Feuerstein G. Broerse JJ, MacVittie TJ, editors. Response of dierent Protection by dehydroepiandrosterone in mice infected species to total body irradiation. Boston: Martinus with viral encephalitis. Arch Virol 1991;120:263±71. Nijho Publishers, 1984. p. 209±19. [14] Loria RM, Padgett DA. Androstenediol regulates sys- [2] Hendry JH, Feng-Tong Y. Response of bone marrow to temic resistance against lethal infections in mice. Arch low LET irradiation. In: Hendry JH, Lord BI, editors. Virol 1992;127:103±15. Radiation toxicology: bone marrow and leukaemia. [15] Loria RM, Padgett DA. Mobilization of cutaneous London: Taylor and Francis, 1995. p. 91±116. immunity for systemic protection against infections. Ann [3] Mauch P, Constine L, Greenberger J, Knospe W, N Y Acad Sci 1992;650:363±6. Sullivan J, Liesveld JL, Deeg HJ. Hematopoietic stem [16] Rasmussen KR, Martin EG, Healey MC. Eects of cell compartment: acute and late eects of radiation dehydroepiandrosterone in immunosuppressed rats therapy and chemotherapy. Int J Radiat Oncol Biol Phys infected with Cryptosporidium parvum. J Parasitol 1995;31:1319±39. 1993;79:364±70. [4] Elliott TB, Madonna GS, Ledney GD, Brook I. [17] Araneo B, Daynes R. Dehydroepiandrosterone functions Combined therapy for postirradiation infection. as more than an antiglucocorticoid in preserving immu- Microecol Ther 1989;19:105±8. nocompetence after thermal injury. Endocrinology [5] Madonna GS, Ledney GD, Elliott TB, Brook I, Ulrich 1995;136:393±401. JT, Myers KR, Patchen ML, Walker RI. Trehalose [18] Padgett DA, Sheridan JA, Loria RM. Steroid hormone dimycolate enhances resistance to infection in neutrope- regulation of a polyclonal TH2 immune response. Ann N nic animals. Infect Immun 1989;57:2495±501. Y Acad Sci 1995;774:323±5. [6] Patchen ML, MacVittie TJ, Solberg BD, D'Alesandro [19] Gianotti L, Alexander JW, Fukushima R, Pyles T. MM, Brook I. Radioprotection by polysaccharides alone Steroid therapy can modulate gut barrier function, host M.H. Whitnall et al. / International Journal of Immunopharmacology 22 (2000) 1±14 13

defense, and survival in thermally injured mice. J Surg androstenediol, and androstenetriol. J Neuroimmunol Res 1996;62:53±8. 1998;84:61±8. [20] Padgett DA, Loria RM, Sheridan JF. Endocrine regu- [32] MacVittie TJ, Farese AM. Experimental approaches to lation of the immune response to in¯uenza virus infec- treatment of radiation-induced hemopoietic injury. In: tion with a metabolite of DHEA-androstenediol. J Hendry JH, Lord BI, editors. Radiation toxicology: bone Neuroimmunol 1997;78:203±11. marrow and leukaemia. London: Taylor & Francis, [21] Suitters AJ, Shaw S, Wales MR, Porter JP, Leonard J, 1995. p. 141±94. Woodger R, Brand H, Bodmer M, Foulkes R. Immune [33] Patchen ML. Single and combination cytokine therapies enhancing eects of dehydroepiandrosterone and dehy- for the treatment of radiation-induced hemopoietic droepiandrosterone sulphate and the role of steroid sul- injury. Adv Biosci 1996;94:21±36. phatase. Immunology 1997;91:314±21. [34] Neta R. Modulation of radiation damage by cytokines. [22] Carr DJ. Increased levels of IFN-gamma in the trigem- Stem Cells Dayt 1997;15(Suppl. 2):87±94. inal ganglion correlate with protection against HSV-1- [35] Sonka J, Strakova M. Survival of irradiated mice pre- induced encephalitis following subcutaneous adminis- treated by dehydroepiandosterone. Agressologie tration with androstenediol. J Neuroimmunol 1970;11:421±6. 1998;89:160±7. [36] Padgett DA, Loria RM. In vitro potentiation of lympho- [23] Daigle J, Carr DJ. Androstenediol antagonizes herpes cyte activation by dehydroepiandrosterone, androstene- simplex virus type 1-induced encephalitis through the diol, and androstenetriol. J Immunol 1994;153:1544±52. augmentation of type 1 IFN production. J Immunol [37] McLachlan JA, Serkin CD, Bakouche O. 1998;160:3060±6. Dehydroepiandrosterone modulation of lipopolysacchar- [24] Yen SSC, Morales AJ, Khorram O. Replacement of ide-stimulated monocyte cytotoxicity. J Immunol DHEA in aging men and women. Potential remedial 1996;156:328±35. eects. Ann N Y Acad Sci 1995;774:128±42. [38] Risdon G, Kumar V, Bennett M. Dierential eects of [25] Danenberg HD, Alpert G, Lustig S, Ben-Nathan D. dehydroepiandrosterone on murine lymphopoiesis and Dehydroepiandrosterone protects mice from endotoxin myelopoiesis. Exp Hematol 1991;19:128±31. toxicity and reduces tumor necrosis factor production. [39] Hendry JH, Lord BI, editors. Radiation toxicology: bone Antimicrob Agents Chemother 1992;36:2275±9. marrow and leukaemia. London: Taylor & Francis Inc, [26] Araghi-Niknam M, Zhang Z, Jiang S, Call O, Eskelson 1995. CD, Watson RR. Cytokine dysregulation and increased [40] Myska JC, Adams TL, Bhatt RC, Broom JG, Pitcher oxidation is prevented by dehydroepiandrosterone in CD, Sharpnack FM II, St. John TJ, Torres BA, Vavrina mice infected with murine leukemia retrovirus. Proc Soc G, Gerstenberg HM. Arrays for use at the cobalt ir- Exp Biol Med 1997;216:386±91. radiation facility. AFRRI Technical Report 97-2. Armed [27] Hernandez-Pando R, De La Luz Streber M, Orozco H, Forces Radiobiology Research Institute, September, Arriaga K, Pavon L, Al-Nakhli SA, Rook GA. The 1997. eects of androstenediol and dehydroepiandrosterone on [41] American Association of Physicists in Medicine (AAPM the course and cytokine pro®le of tuberculosis in BALB/ Task Group 21). A protocol for the determination of c mice. Immunology 1998;95:234±41. absorbed dose from high-energy photon and electron [28] Inserra P, Zhang Z, Ardestani SK, Araghi-Niknam M, beams. Med Phys 1983;10:741±71. Liang B, Jiang S, Shaw D, Molitor M, Elliott K, Watson [42] Hammer JG, Harris HA, Komm BS, Goodwin T, RR. Modulation of cytokine production by dehydroe- Goldberg K, Van Winkle T. Comparison of gross and piandrosterone (DHEA) plus melatonin (MLT) sup- histologic eects of six vehicles for subcutaneous injec- plementation of old mice. Proc Soc Exp Biol Med tion of hydrophobic steriod mimetic compounds in rats. 1998;218:76±82. Contemp Top Lab Anim Sci 1996;35:72 [abstract]. [29] Kimura M, Tanaka S-I, Yamada Y, Kiuchi Y, [43] Seed TM, Fritz TE, Tolle DV, Poole CM, Lombard LS, Yamakawa T, Sekihara H. Dehydroepiandrosterone Doyle DE, Kaspar LV, Cullen SM, Carnes BA. Survival decreases serum tumor necrosis factor-a and restores patterns and hemopathological responses of dogs under insulin sensitivity: independent eect from secondary continuous gamma irradiation. In: Broerse JJ, MacVittie weight reduction in genetically obese Zucker fatty rats. TJ, editors. Response of dierent species to total body ir- Endocrinology 1998;139:3249±53. radiation. Boston: Martinus Nijho Publishers, 1984. p. [30] Moynihan JA, Callahan TA, Kelley SP, Campbell LM. 137±59. Adrenal hormone modulation of type 1 and type 2 cyto- [44] Fritz TE, Seed TM, Tolle DV, Lombard LS. Late eects kine production by spleen cells: dexamethasone and of protracted whole-body irradiation of beagles by dehydroepiandrosterone suppress interleukin-2, interleu- cobalt-60 gamma rays. In: Thompson RC, McGaey JA, kin-4, and interferon-gamma production in vitro. Cell editors. Life-span radiation eects studies in animals: Immunol 1998;184:58±64. what can they tell us? Washington, DC: Oce of [31] Padgett DA, Loria RM. Endocrine regulation of murine Scienti®c and Technical Information, United States macrophage function: eects of dehydroepiandrosterone, Department of Energy, 1986. p. 116±41. 14 M.H. Whitnall et al. / International Journal of Immunopharmacology 22 (2000) 1±14

[45] Seed TM, Kaspar LV. Probing altered hematopoietic variant of glucose-6-phosphate dehydrogenase. progenitor cells of preleukemic dogs with JANUS ®ssion Carcinogenesis 1987;8:1567±70. neutrons. Radiat Res 1991;128:S81±S86. [59] Suzuki T, Suzuki N, Daynes RA, Engleman EG. [46] Lord BI, Hendry JH. Radiation toxicology: bone mar- Dehydroepiandrosterone enhances IL2 production and row and leukaemia. In: Hendry JH, Lord BI, editors. cytotoxic eector function of human T cells. Clin Radiation toxicology: bone marrow and leukaemia. Immunol Immunopathol 1991;61:202±11. London: Taylor & Francis, 1995. p. 1±21. [60] Meikle AW, Dorchuck RW, Araneo BA, Stringham JD, [47] Vial T, Descotes J. Clinical toxicity of cytokines used as Evans TG, Spruance SL, Daynes RA. The presence of a haemopoietic growth factors. Drug Safety 1995;13:371± dehydroepiandrosterone-speci®c receptor binding com- 406. plex in murine T cells. J Steroid Biochem Mol Biol [48] Araneo BA, Shelby J, Li GZ, Ku W, Daynes RA. 1992;42:293±304. Administration of dehydroepiandrosterone to burned [61] Okabe T, Haji M, Takayanagi R, Adachi M, Imasaki K, mice preserves normal immunologic competence. Arch Kurimoto F, Watanabe T, Nawata H. Up-regulation of Surg 1993;128:318±25. high-anity dehydroepiandrosterone binding activity by [49] Daynes RA, Araneo BA, Ershler WB, Maloney C, Li dehydroepiandrosterone in activated human T lympho- GZ, Ryu SY. Altered regulation of IL-6 production with cytes. J Clin Endocrinol Metab 1995;80:2993±6. normal aging. Possible linkage to the age-associated [62] Maradny E, Kanayama N, Maehara K, Kobayashi T, decline in dehydroepiandrosterone and its sulfated de- Terao T. Dehydroepiandrosterone sulfate potentiates the rivative. J Immunol 1993;150:5219±30. eect of interleukin-8 on the cervix. Gynecol Obstet [50] Kim HR, Ryu SY, Kim HS, Choi BM, Lee EJ, Kim Invest 1996;42:191±5. HM, Chung HT. Administration of dehydroepiandroster- [63] Waxman DJ. Role of metabolism in the activation of one reverses the immune suppression induced by high dehydroepiandrosterone as a peroxisome proliferator. J dose antigen in mice. Immunol Invest 1995;24:583±93. Endocrinol 1996;150:S129±S147. [51] Seed TM, Cullen SM, Kaspar LV, Tolle DV, Fritz TE. [64] Peters JM, Zhou YC, Ram PA, Lee SS, Gonzalez FJ, Hemopathologic consequences of protracted gamma ir- Waxman DJ. Peroxisome proliferator-activated receptor radiation: alterations in granulocyte reserves and gra- alpha required for gene induction by dehydroepiandros- nulocyte mobilization. Blood 1980;56:42±51. terone-3 beta-sulfate. Molec Pharmacol 1996;50:67±74. [52] Fujikawa H, Okura F, Kuwano Y, Sekizawa A, Chiba [65] Loria RM, Padgett DA, Huynh PN. Regulation of the H, Shimodaira K, Saito H, Yanaihara T. Steroid sulfa- immune response by dehydroepiandrosterone and its tase activity in osteblast cells. Biochem Biophys Res metabolites. J Endocrinol 1996;150:S209±S220. Commun 1997;231:42±7. [66] Mor®n R, Courchay G. Pregnenolone and dehydroepian- [53] Lea-Currie YR, Wu SM, McIntosh MK. Eects of acute drosterone as precursors of native 7-hydroxylated metab- administration of dehydroepiandrosteron-sulfate on adi- olites which increase the immune response in mice. J pose tissue mass and cellularity in male rats. Int J Obes Steroid Biochem Mol Biol 1994;50:91±100. Relat Metab Disord 1997;21:147±54. [67] Frenkel RA, Slaughter CA, Orth K, Moomaw CR, [54] Nakashima N, Haji M, Sakai Y, Ono Y, Umeda F, Hicks SH, Snyder JM, Bennett M, Prough RA, Putnam Nawata H. Eect of dehydroepiandrosterone on glucose RS, Milewich L. Peroxisome proliferation and induction uptake in cultured human ®broblasts. Metabolism of peroxisomal enzymes in mouse and rat liver by dehy- 1995;44:543±8. droepiandrosterone feeding. J Steroid Biochem [55] Meikle AW, Daynes RA, Araneo BA. Adrenal androgen 1990;35:333±42. secretion and biologic eects. Endocrinol Metab Clin [68] Yamada J, Sakuma M, Ikeda T, Fukuda K, Suga T. North Am 1991;20:381±400. Characteristics of dehydroepiandrosterone as a peroxi- [56] Mohan PF, Jacobson MS. Inhibition of macrophage some proliferator. Biochim Biophys Acta 1991;1092:233± superoxide generation by dehydroepiandrosterone. Am J 43. Med Sci 1993;306:10±5. [69] Coey DS. Androgen action and the sex accessory tis- [57] Whitcomb JM, Schwartz AG. Dehydroepiandrosterone sues. In: Knobil E, Neill J, editors. The physiology of and 16 alpha-Br-epiandrosterone inhibit 12-O-tetradeca- reproduction. New York: Raven Press, 1988. p. 1081± noylphorbol-13-acetate stimulation of superoxide radical 119 [Chapter 24]. production by human polymorphonuclear leukocytes. [70] Angele MK, Ayala A, Cio WG, Bland KI, Chaudry Carcinogenesis 1985;6:333±5. IH. Testosterone: the culprit for producing splenocyte [58] Pascale R, Garcea R, Ruggiu ME, Daino L, Frassetto S, immune depression after trauma hemorrhage. Am J Vannini MG, Cozzolino P, Lenzerini L, Feo F, Schwartz Physiol 1998;274:C1530±C1536. AG. Decreased stimulation by 12-O-tetradecanoylphor- [71] National Research Council. Guide for the care and use bol-13-acetate of superoxide radical production by poly- of laboratory animals. Washington, DC: National morphonuclear leukocytes carrying the Mediterranean Academy Press, 1996.