Original Paper International Journal of Cell Cloning 9:78-88 (1991)

Accelerated Marrow Recovery Following Total-Body Irradiation After Treatment with Vincristine, Lithium or Combined Vincristine-Lithium Roberta M.Johnke, Reed S. Abernattry Department of Radiation Oncology, East Carolina University School of Medicine, Greenville, North Carolina, USA

Key Words. Hematopoiesis

- Radioprotection - Lithium

*

Vincristine

Abstract. Accelerated post-irradiation recovery of hematopoietic marrow has been reported following treatment with lithium (Li) or vincristine (VcR). Because these two agents appear to exert their effects on different, albeit overlapping, hematopoietic populations, it was felt that combining them might lead to a wider spectrum of enhanced post-irradiation marrow regeneration. Results demonstrated that an accelerated recovery, which appeared to be additive in nature, was observed in the marrow following combined VcR-Li/4.5 Gy total-body irradiation. The combined schedule significantly enhanced post-irradiation recovery of white blood cells, E d a y spleen colony-forming units, erythroid burst-forming units, and fibroblastic colony-forming units over radiation alone; and recovery of marrow cellularity, multipotential colony-forming units (CFU-gemm) and granulocytic/monocytic colony-forming units (CFU-gm) over both radiation alone and either drug given singly with the 4.5 Gy. In addition, while data on the ability of regenerating stroma to support CFU-gm and CFU-gemm did not suggest that VcR was acting to enhance post-irradiation marrow recovery by increasing stromal production of hematopoietic growth factors, Li did appear to increase production of one or more of these factors, and this may be part of its mechanism of action.

Introduction For some time, it has been known that radioprotection of laboratory animals can be achieved by prior treatment or “priming” with a diverse array of agents which are themselves cytotoxic [I-51.Among these cytotoxic agents is the antineoplastic drug vincristine (VcR). Priming with VcR one to two days prior to a lethal dose of irradiation has been demonstrated to markedly increase animal Correspondence: Roberta M. Johnke, Ph.D., Department of Radiation Oncology, Division of Radiation Biology and Oncology, East Carolina University School of Medicine, Greenville, NC 27858. Received July 26, 1990; provisionally accepted August 31, 1990; accepted for publication October 9, 1990. 0737-1454191/$2.00/0 oAlphaMed Press

Enhanced Post-Irradiation Marrow Recovery with VcR and Li

79

survival [6]-a response attributed to the ability of VcR to significantly accelerate recovery of critical hematopoietic stem cell populations as well as erythropoietic and granulopoietic progenitors [7-91. However, while radioprotection of the marrow microenvironment is also a key concern, priming with VcR has not been shown to accelerate recovery of the marrow stroma [9]. On the other hand, the monovalent cation lithium (Li) has been reported to markedly accelerate stromal cell recovery following total-body irradiation (TBI) [lo]. Unlike VcR, this psychotropic drug demonstrates negligible erythropoietic ability, but is a well known stimulator of granulopoiesis with marked proliferative effects upon a number of stem cell and progenitor cell populations [II-151. Indeed, Li’s potent granulopoietic stimulation combined with its low toxicity has led to an active interest in the use of this compound to ameliorate the neutropenia that may accompany antineoplastic chemotherapy or radiation [lo, 12, 16, 171. Hence, both Li and VcR have been demonstrated to attenuate the myelosuppression that can occur following radiation. Moreover, Li and VcR appear to exert their favorable hematopoietic effects on different, albeit overlapping, marrow subpopulations. It is conceivable, therefore, that combining these two agents may result in a complementary interaction leading to a wider spectrum of enhanced marrow regeneration following irradiation. This report describes results of experiments designed to test this hypothesis.

Materials and Methods Experimental Animals All investigationswere performed using 8-to12-week-old B6D2F, (C57B1/6 x DBA/2) mice (Jackson Labs, Bar Harbor, ME) allowed autoclaved food and acidified water ad libitum. Animals were maintained in conventional housing on a 12 h light/dark cycle with the dark cycle beginning at 6 p.m. The guidelines of laboratory care as put forth by the National Institute of Laboratory Animal Science were strictly observed.

Drugs and Radiation

+

The physical factors of x-irradiation were 275 kVp, 15 mA, H.V.L. = 0.55 mm Cu 0.90 mm Al and a target-to-source distance of 51 cm. During irradiation, animals were placed without anesthesia in acrylic containers, receiving TBI at a rate of 0.4 Gy/min. The experimental treatment dose was 4.5 Gy TBI (no death was observed with this dose during the time frame of these studies). The 12-day spleen colony-forming unit (CFU-s-l2d) recipients received 8.5 Gy TBI. VcR was dissolved in sterile physiological saline at a stock concentration of 0.1 mg/ml and was injected i.p. at a dose of 1 rng/kg body weight 24 h prior to irradiation (this schedule has been demonstrated to result in maximal enhancement of marrow post-irradiation regeneration [9]). Ultrapure lithium carbonate ( 2 99.99% pure; Alpha Products, Danvers, MA) was administered according to a modification of the schedule reported by Gallicchio et al. [Ill. Specifically, Li,CO, was dissolved in phosphate-buffered saline (PBS) and administered i.p. at a dose of 1.5 mglkg body weight immediately following irradiation and again at 24 h post-irradiation (preliminary tests demonstrated that this schedule provided optimal stimulation in our system).

Johnke/ Abernathy

80

Hematological Indices Blood obtained from cardiac puncture was used to assess hemoglobin content and white blood cell (WBC) concentration per ml of blood. Hemoglobin content was determined using a Model HGBR Coulter Hemoglobinometer, and the white cell count was assessed using a Model Z, Coulter Counter (Coulter Electronics, Luton, Bedfordshire, England). At least 5 mice were used per experiment and 3 experiments performed. Individual values were averaged to obtain the means (f SEM). Marruw Cell Suspensions Marrow cell suspensions were obtained by flushing excised femurs with 1.5 ml volumes of modified McCoy's 5A medium (Sigma, St. Louis, MO) containing 10%fetal calf serum (FCS; Hyclone Labs, Logan, Utah). The marrow plugs were aspirated 3-4 times through a 25-gauge needle to obtain a single-cell suspension and cell concentration determined with a hemocytometer. Marrow cellularity was obtained by multiplying cell number per ml by the volume of the cell suspension. Hematopoietic Stem Cell and Progenitor Cell Assays CFU-s-12d For the CFU-s-l2d, appropriate dilutions of the experimental marrow cell suspensions were injected into lethally irradiated (8.5 Gy) B6D2F1 recipients (10 per group) via the tail vein as previously described [18]. Following injection, recipients were maintained under isolation conditions for l2 days with acidified water and autoclaved food available ad libitum. The mice were then sacrificed, the spleens removed, fixed in Bouin's solution (71% saturated picric acid, 24% formalin and 5% glacial acetic acid; all chemicals purchased from Fisher, Pittsburgh, PA) for 24 h and the number of macroscopic foci enumerated using a dissecting microscope. Irradiated, saline-injected mice served as controls for determining endogenous CFU-s. Mean (fSEM) CFU-s-lZd/femur for untreated control mice in these experiments was 2,485 f 656. Granulocyte, Erythrocyte, Monocyte, Megakaryocyte Colony-Forming Units (CFU-gem) Conditions of the C F U - g e m assay have been reported previously [8].As stimuli, each culture contained 1 U of recombinant human erythropoietin (rhEpo; Connaught Labs, Ontario, Canada; specific activity of l0,OOO U/mg protein) and 1.5%(v:v) pokeweed mitogen spleen cell-conditioned medium (PWM-SCM; pokeweed mitogen obtained from GIBCO, Grand Island, NY). Cultures were incubated at 37°C and 7.5% co2/92.5%air humidified to >98%. After 7 days, cultures were stained with benzidine, and those colonies containing both benzidine-positive and bemidine-negative cells were scored as C F U - g e m . Mean (k SEM) CFU-gemmkmur h r untreated control mice in these experimentswas 3,797 f 693. Erythroid Burst-Forming Units (BFU-e) Although a different marrow compartment than the CFU-gemm, the same methods as described above for the C F U - g e m were used to assay for the BFU-e with the exception that 2 U of rhEpo were used. The cultures were incubated for 7 days at 3TC, 7.5% C02/92.5% air and >98% humidity before being stained with benzidine and scored for the number of bemidine-positive colonies ( >50 cells) present. Mean (* SEM) BFU-e/ femur for untreated control mice in these experiments was 2,938 f 854.

Enhanced Post-Irradiation Marrow Recovery with VcR and Li

81

Granulocyte/Macrophage Colony-Forming Units (CFU-gm) Conditions of the CFU-gm assay have been described previously [18]. Marrow cells were added to cultures containing McCoy's 5A medium supplemented with bone marrow additives, penicillin (Sigma), streptomycin (Sigma), HEPES buffer (Sigma), 10%FCS and 0.3% agar. Step II purified L cell colony-stimulating factor (L-CSF) [I91was used as a stimulus. The cultures were incubated at 3 7 T , 7.5% C02/92.5%air and >98%humidity for 7 days, stained with 2-iodophenyl, 3-nitrophenyl, 5-phenyl, 2-H tetrazolium chloride (MT) stain and the number of colonies ( >50 cells) determined. Mean (fSEM) CFU-gml femur for untreated control mice in these experiments was 5,634 937.

+

Fibroblastic Colony-Forming Units (CFU-f) CFU-f were assayed by a modification of the method of Piersma et al. [20]. Briefly, appropriate numbers of marrow cells were plated in a-MEM (GIBCO, Grand Island, NY) containing 20% FCS and 0.8% methylcellulose (Fisher, Pittsburgh, PA). Triplicate cultures were plated and placed in large glass Petri dishes containing a central well of sterile water. They were incubated at 3TC, 7.5%C02/92.5%air and >98% humidity for 7 days. Dishes were then gently washed 3-4 times in cold phosphate-buffered saline (PBS) to thoroughly remove the methylcellulose, air dried, fixed in methanol, and stained with a freshly prepared Giemsa solution. Colonies containing >20 fibroblastic cells were scored as CFU-f. Mean (fSEM) CFU-f/femur for untreated control mice in these experiments was 104 f 13. Measurement of the Functional Capacity of the Hematopoietic Microenvironment The functional capacity of regenerating stroma to support CFU-grn colony formation was assessed as described by Gullicchio et al. [lo], and a modification of this technique was used to assess the ability of stroma to support C F U - g e m . Briefly, with the number of plated bone marrow cells adjusted so that each treatment group produced the same number (& 5%)of colonies, the CFU-f for the various groups were cultured in a-MEM containing 20% FCS. Following 7 days of incubation, the medium was decanted and a 0.5% agar layer placed over the stromal colonies. This was then overlayed with normal, nonadherent bone marrow cells (adherent cells were removed by incubation for 90min at 37°C) in either 0.3%agar containing McCoy's 5A medium Supplemented with bone marrow additives, penicillin, streptomycin, HEPES buffer and 10%FCS for CFU-gm assay, or 0.8% methylcellulose containing a-MEM, 2-mercaptoethanol (Fisher, Pittsburgh, PA), bovine serum albumin (Sigma), transferrin (Sigma), sodium selenite (Sigma), 30% FCS and 1 U rhEpo for the CFU-gemm assay. No exogenous stimulus (Step II L-CSF or PWM-SCM) was added. The plates were then returned to the incubator for another 7 days before being scored for their CFU-gm or CFU-gemm colonies ( >50 cells). Morphologic identification of the CFU-grn and C F U - g e m was confirmed by plucking the colonies and staining cytospin preparations with Dif-Quik stain. Statistical Analysis Pooled marrow from a minimum of 3 mice per data point per experiment was used to obtain untreated control values and from a minimum of 5 mice per data point per experiment for radiation alone or VcR and/or Li with radiation, All marrow precursor data were expressed as number of colonies per femur. For convenience, control values were used in some cases to convert radiation alone, VcRlradiation, Lilradiation, and VcR-Li/radiation values into percents of control. Each data point represents the mean (fSEM) from 3 or more experiments. Student's t test was used to find statistical differences among the means of the various schedules. A probability of 0.05 was used as a level of significance.

82

Johnke/Abernathy

120

I

A

120

a

2

100 80

0

)I

60

0

K

40 2o

5

0 0

15

Qo

10

t

E

Rad. Alone Li + Rad. VcR + Rad. VcR+Li+Rad.

d

U U

:

5

u

I

0

7

I

14

I

I

I

21

2a

35

Daya Post 4.5 Gy

Fig. 1. The influence of VcR, Li and combined Vcr and Li on blood indices and marrow cellularity: A) the post-irradiation values of blood hemoglobin content; B) the postirradiation values of blood WBC concentration (data points for A and B represent the mean +_ SEM from three separate experiments presented as a percent of control); and C ) the post-irradiation values for nucleated bone marrow cellularity following 4.5 Gy TBI (the mean +_ SEM nucleated bone marrow cells/femur observed for untreated control mice was 13.64 f 0.82 X lo6).

Results Hematological indices and Bone Marrow Cellularity The results of monitoring the hemoglobin content following radiation alone or in combination with VcR andor Li (Fig. lA)indicate that no significant changes

Enhanced Post-Irradiation Marrow Recovery with VcR and Li

83

occurred during the time frame of the experiment for any of the treatment schedules when the data were compared to untreated controls. In the WBC compartment, however, a pronounced response was observed (Fig. 1B). All treatment schedules markedly reduced WBC cellularity during the first week with little variance in cellularity observed among any of the groups. W e recovery did not begin until the third week for radiation alone, however, it began during the second week for those treatments combining VcR andlor Li with the radiation. This resulted in WBC levels significantly higher than radiation alone ( p 5 0.05) for these three schedules during the following three weeks. The combination of VcR and Li did not seem to elevate WBC following radiation any better than did either drug alone. The effect of VcR andor Li on the post-irradiation recovery of femoral marrow cellularity is illustrated in Figure 1C. Initially, all treatments markedly suppressed bone marrow cellularity. By four days following irradiation, however, treatment with VcR and/or Li prior to irradiation resulted in significantly higher ( p 5 0.05) marrow cellularities than did radiation alone, being highest for the VcR-Li/radiation schedule. Cellularity continued to remain significantly elevated over levels with radiation alone until day 14 in the Li-treated group and through the duration of the experimental interval (21 days) in both the VcR- and VcR-Li-treated groups. Recovery Kinetics of the Marrow Precursor Compartments The temporal responses of the stem cell (CFU-s-Ed, CFU-gem), progenitor cell (CFU-gm, BFU-e) and stromal (CFU-f) compartments following 4.5 Gy TBI alone or in conjunction with VcR and/or Li are presented in Table I. CFU-s-Ed and C F U - g e m Results demonstrate that VcR priming 24 h prior to irradiation significantly enhances the rate of CFU-s-Ed recovery when compared to radiation alone, but administration of Li does not appreciably alter the recovery kinetics at any time. Administrationof both VcR and Li also significantlyacceleratesthe post-irradiation recovery of the CFU-s-Ed when compared to radiation alone, but not as much as the VcRhadiation schedule does. Unlike the CFU-s-L?d, recovery from 4.5 Gy in the CFU-gem compartment is significantly accelerated following both Li and VcR. Additionally, combining the VcR and Li resulted in a recovery rate that was even greater than that resulting from either compound alone, returning to above control levels by 7 days post-irradiation. Lihadiation and VcRlradiation schedules did not return to control values until day 14 post-irradiation. CFU-gm and BFU-e The temporal recovery of the CFU-gm progenitor following 4.5 Gy TBI is acceleratedby the addition of either Li or VcR. Furthermore, combination of these two agents acceleratesrecovery significantly better than either agent given singly. VcR priming also acceleratedthe post-irradiation recovery of the erythroid BFU-e, but Li administration did not. The VcR-Lilradiation schedule resulted in no better post-irradiation recovery of the BFU-e than did a VcRhadiation schedule.

Johnkel Abernathy

84

Table I. Effect of VcR and Li on the recovery of marrow precursors following 4.5 Gy Marrow precursor

Days after radiation

Radiation alone

CFU-s-12d

1 7 14 21

CFU-gemm

0 1 3 7 14 0 1 3 7 14 0 1 3 7 14 1 3 7 14

0.4f0.1 9.3 f 1.9 46.7 f 5.4 72.9 f 5.8 0.5f0.1 0.6 f 0.1 18.6 f 1.5 46.6 f 2.1 88.2 f 5.1 1.1f0.2 1.8 f 0.1 10.1 f 1.6 36.5 f 3.0 82.7 f 4.1 0.6f0.1 0.5f0.1 18.0 f 1.3 50.3 f 2.9 94.2 f 8.7 5.6 f 0.3 25.6 f 2.5 88.0 f 3.5 90.5 f 5.5

CFU-gm

BFU-e

CFU-f

+

+

Li radiation

VcR radiation

0.4f0.1 0.5f0.1 6.3 f 1.0 20.7 f 1.6" 42.0 f 4.9 79.7 f 7.3" 77.0 f 6.9 98.5 f 7.0" 0.5k0.2 0.3f0.1 1.2 f 0.1" 1.5 f 0.1" 37.6 f 2.8" 35.6 f 2.1" 67.4 f 2.7" 75.6 f 3.3" 106.7 f 6.5 114.7 _+ 8.0" 1.2f0.2 0.8f0.2 2.2 f 0.1 2.0 f 0.1 16.6 f 1.8" 15.8 f 0.9" 59.0 f 3.4" 50.6 f 2.7" 101.6 f 4.5" 90.5 f 5.6 0.6f0.2 0.8f0.2 0.4f0.1 0.8f0.3 22.3 f 2.3 35.2 f 1.7" 46.8 f 3.8 70.6 f 3.7" 90.1 f 3.9 111.2 f 5.9 5.2 f 0.5 5.8 f 0.6 47.7 f 2.2" 20.2 f 1.7 122.5 f 8.0" 74.1 f 4.1 113.5 f 5.3" 82.9 f 4.9

VcR & Li radiation

+

0.5f0.1 23.8 f 2.3ab 67.4 f 4.9ab 87.3 f 7.3 0.4f0.1 2.4 f 0.2abc 54.1 f 3.1abc 118.6 f 5.8abc 110.0 f 6.9 1.2f0.2 3.8 f 0.3"bc 32.9 f 2.2"k 70.4 f 2.9abc 107.7 f 7.5" 1.3f0.4 0.750.2 40.3 f 2Sab 69.6 f 2.9ab 97.5 f 8.5 3.8 f 0.9 46.0 f 2.2" 97.5 f 6.3 119.3 f 7.9"

"Significantly different from radiation alone (p I0.05) bVcR & Li radiation significantly different from Li radiation treatment ( p 5 0.05) 'VcR & Li radiation significantly different from VcR + radiation treatment ( p 5 0.05) Values are expressed as % of control; each data point is the mean f SEM from at least 3 experiments. Student's r test was used to determine the statistical differences among the mean values of the various treatments.

+ +

+

CFU-f The post-irradiation response of the CFU-f was monitored as an index of stromal integrity. The Likadiation schedule produced a strong proliferative response in the CFU-f compartment with numbers exceeding control levels by day 7, but the VcRIradiation treatment failed to enhance recovery over that from radiation alone. VcR-Lihadiation also enhanced CFU-f regeneration when compared to radiationalone, but no more so than the Lihadiation schedule did. Ability of the CFU-fto Support CFU-gm and CFU-gemm Growth The data on CFU-gem colony formation and growth (Fig. 2A) demonstrated

85

Enhanced Post-Irradiation Marrow Recovery with VcR and Li

I

I

0 Rad. Alone 0 Li + Rad.

+ Rad. VcR+Li+Rad.

w VcR

I

0

7

14

Daya Post 4.6 Gy

Fig. 2. The influence of VcR, Li and combined VcR and Li on the ability of irradiated marrow CFU-f to support the growth and formation of CFU-gm and CPU-gem: A) CFUg e m from untreated, non-adherent bone marrow cells (the mean f SEM C F U - g e m supported by untreated control CFU-f underlayers was 33.2 f 3.1), and B) CFU-gm from untreated, non-adherent bone marrow cells (the mean f SEM CFU-gm supported by untreated control CFU-f underlayers was 108.1 f 11.4). The X-axis represents the days after 4.5 Gy TBI that CFU-f underlayers were initiated.

that CFU-f from the radiation alone group supported a somewhat higher number of CFU-gemm than did C F U f from the other three treatment groups, but none of the differences among the groups were statistically significant at the 5% level. Conversely, support of CFU-gm (Fig. 2B) was significantly greater than radiation alone for both Lihadiation and VcR-Likadiation schedules ( p s 0.01) on day 1 post-irradiation and remained so through day 7 post-irradiation. However, the combined VcR/Li regimen appeared to be somewhat less effective than treatment with Li alone, and no improvement over radiation alone was observed for the VcWradiation schedule. By day 14 post-irradiation, no treatment group’s stroma was significantly better at supporting CFU-gm than any other groups.

JohnkelAbernathy

86

Discussion The effect of both VcR and Li on the post-irradiation recovery kinetics of peripheral blood, bone marrow cellularity and the hematopoietic stem cell, progenitor cell and stromal cell compartments has been well studied [6-121. The present studies are in general agreement with these previous reports, although the stimulatory effect of Li administered with 4.5Gy appears to be smaller than that reported previously for 2.0 Gy (perhaps due to such factors as differences in experimentaldesign between the two studies, increased radiation-induceddeath following 4.5Gy of those cells responsible for secreting hematopoietic stimulatory factors, or because regeneration following 4.5 Gy cannot be enhanced as greatly as that following 2.0 Gy since the marrow’s recovery systems are already functioning near their maximum after this higher radiation dose). Additionally, the present data on the ability of regenerating stroma to support CFU-gm and CFU-gemm add insight into the role of stroma in VcR and Li hematopoietic stimulation. Results indicate that increased stromal production of hematopoietic growth factors is not the reason for VcR priming effects on the marrow, but may mediate, at least in part, the action of Li on this organ. Some investigators have postulated that Li may function to stimulate hematopoiesis by increasing production and/or release of one or more of the CSFs from accessory cells in the stroma [21-231, and the results from this study would tend to support this idea. However, failure to observe increased support of CFU-gemm colonies by CFU-f from animals treated with Lihadiation instead of radiation alone would also suggest that the nature of the CSFs produced by the stroma after Li treatment is restricted. As postulated, because the various marrow compartmentsrespond differently to VcR priming and Li treatment, combining these two agents with radiation did, indeed, result in acceleratedpost-irradiation recovery rates over a wider spectrum of marrow subpopulationsthan did radiation alone or either agent given singly with the radiation. Specifically, results demonstratedthat combining the VcR and Li stimulatory effects led to a post-irradiationrecovery of marrow cellularity, CFU-gemm and CFU-gm that was significantly accelerated when compared to the other three treatments monitored. Unexpectedly, however, the accelerated post-irradiation recovery of the marrow following combined VcR-Li was not translated into significantly better peripheral blood indices than were observed following either agent alone (Figs. lA and B), at least not during the time frame of this experiment. In general, the enhanced recovery seen in the hematopoietic compartments monitored following the VcR-Likadiation treatment appeared to be additive in nature, suggesting that Li and VcR are most likely acting on the bone marrow through independent mechanisms. Nevertheless, some data did indicate that the combination of VcR and Li may be exerting a moderately detrimental effect in some compartments. For example, the VcR-Lihadiation schedule, while significantly better than radiation alone, did not accelerate CFU-s-12d recovery during the second and third weeks following radiation as markedly as did just VcR alone with radiation. However, these results may reflect the strong granulopoieticpull that

Enhanced Post-Irradiation Marrow Recovery with VcR and Li

87

Li has been reported to exert on the CFU-s stem cell compartment [I21 of the marrow. In addition, the post-irradiation benefits observed in the C F U f compartment appeared to be somewhat better for the Lihadiation group than for the VcR-Li/ radiation group, suggesting that VcR priming may be exerting a slightly negative influence on the stroma. Indeed, VcR priming prior to irradiation, although not affecting the number of cells initially killed, appears to slow, somewhat, the postirradiation recovery of the stromal CFU-f. The reasons for this are unclear. Perhaps VcR added to the radiation results in increased toxicity to a population(s) of cells responsible for secreting factors that promote C F U f growth and recovery. Studies are currently underway to better define the action of both VcR and Li on the marrow stroma in an effort to better understand the regulatory role (both humoral and cellular) that the hematopoietic microenvironment plays in the recovery of hematopoiesis. In conclusion, both Li and VcR are clinically available drugs known to exert marked stimulatory effects on the post-irradiation recovery of hematopoietic marrow. The data from this study suggest that combining these two moderately stimulatory regimes with radiation can result in an additive interaction to achieve a wider spectrum of enhanced marrow regeneration.

Acknowledgment This study was supported, in part, by Grant #42036 awarded by the National Cancer Institute, National Institutes of Health, and Department of Health and Human Services.

References Millar JL, McElwain TJ. The concept of priming. Eur J Cancer Clin Oncol 1985; 21: l303-UO5. Millar JL, Blackett NM, Hudspeth BN. Enhanced post-irradiation recovery of the haemopoietic system in animals pretreated with a variety of cytotoxic agents. Cell Tissue Kinet 1978;11:543-55 3. Millar JL, McElwain TJ. Combinations of cytotoxic agents that have less than expected toxicity on normal tissues in mice. Antibiot Chemother 1978;23:271-282. Millar JL, Smith IE, McElwain TJ. Pretreatment with certain cytotoxic drugs reduces the normal tissue toxicity of anti-cancer agents in mice and man. Br J Cancer 1981; 43~705-706. Phelps TA, Blackett NM. Protection of intestinal damage by pretreatment with cytarabine (cytosine arabinoside). Int J Radiat Oncol Biol Phys 1978;5:1617-1620. Smith WW, Wfison SM. Effects of vinblastine and vincristine on survival and hemopoiesis in irradiated mice. JNCI l967;39:1055-1066. Smith WW, Wilson SM, Fred S S . Kinetics of stem cell depletion and proliferation: effects of vinblastine and vincristine in normal and irradiated mice. JNCI 1968;40: 847-854. Johnke RM, Kovacs CJ, Loven DP, Abernathy RS, Hooker JL. Altered radiosensitivity of hematopoietic stem cells by vincristine pretreatment: superoxide dismutase as a possible mechanism. NCI Monographs 1988;6:193-197.

Johnke/Abernathy

88

9 Johnke RM, Abernathy RS. Accelerated post-irradiation recovery of hematopoietic marrow following priming with low doses of vincristine. Rad Res 1990;122:234-240. 10 Gallicchio VS, Chen MG, Watts TD. Ability of lithium to accelerate the recovery of granulopoiesis after subacute radiation injury. Acta Radio1 Oncol 1984;23:361-366. 1 1 Gallicchio VS, Chen MG. Modulation of murine pluripotential stem cell proliferation in vivo by lithium carbonate. Blood 1980;56:1150-1152. 12 Gallicchio VS, Chen MG, Watts TD, Gamba-Vitala C. Lithium stimulates the recovery of granulopoiesis following acute radiation injury. Exp Hematol 1983;11:553-563. 13 Boggs DR, Joyce RA. The hematopoietic effects of lithium. Semin Hematol 1983; 20:129-138. 14 Barr D,Galbraith PR. Lithium and hematopoiesis. Can Med Assoc J 1983;128:123-126. 15 Chatelain C, Burstein SA, Harker LA. Lithium enhancement of megakaryocytopoiesis in culture: mediation via accessory marrow cells. Blood 1983;62:172-176. 16 Lyman GH, Williams CC, Preston D. The use of lithium carbonate to reduce infection and leukopenia during systemic chemotherapy. N Engl J Med 1980;302:257-260. 17 Barren AJ. Haematological effects of lithium and its use in treatment of neutropenia. Blut 1980;40:1-6. 18 Kovacs CJ, Johnke RM, Evans MJ, Emma DA, Hooker JL. Development of latent residual drug damage to hematopoietic marrow during the subsequentgrowth of tumors. Exp Hematol 1986;14:165-172. 19 Waheed A, Shadduck RK. Purification and properties of L cell-derived colonystimulating factor. J Lab Clin Med 1979;94:180-194. 20 Piersma AH, Brockbank KGM, Ploemacher RE, vanvliet E, Brakel-van Peer KMJ, Visser PJ. Characterization of fibroblastic stromal cells from murine bone marrow. Exp Hematol 1985;13:237-243. 2 1 Ramsey R, Hays EF. Factors promoting colony-stimulating activity (CSA) in macrophages and epithelial cells. Exp Hematol 1979;5:245-254. 22 Quesenberry PJ, Coppola MA, Gualtieri RJ, et al. Lithium stimulation of murine hematopoiesis in liquid culture: an effect mediated by marrow stromal cells. Blood 1984;63:121-127. 23 McGrath HE, Liang C-M, Alberico TA, Quesenbeny PJ. The effect of lithium on growth factor production in long-term bone marrow cultures. Blood 1987;701136-ll42.