myelopoietic stimulating activity of granulocyte/macrophage colony ...

Proc. Nadl. Acad. Sci. USA Vol. 89, pp. 3691-3695, May 1992 Pharmacology

Synergistic action of the benzene metabolite hydroquinone on myelopoietic stimulating activity of granulocyte/macrophage colony-stimulating factor in vitro (acute myelogenous leukemia/hematopoietic progenitor cells/benzene/cytokines)

RICHARD D. IRONS*tt§, WAYNE S. STILLMAN*, DOROTHY B. COLAGIOVANNI*,

AND

VERONICA A. HENRY*

*Molecular Toxicology and Environmental Health Sciences, School of Pharmacy; tDepartment of Pathology, School of Medicine; and tCancer Center, University of Colorado Health Sciences Center, Denver, CO 80262

Communicated by Eugene P. Cronkite, January 27, 1992

benzene have been repeatedly implicated in hematotoxicity (17-20). Benzene and its metabolites produce bone marrow suppression via a cell cycle-specific mechanism, dividing cells being blocked or arrested in G2/M phase (21-23). The quinone metabolites of benzene, hydroquinone (HQ) and p-benzoquinone, are potent inhibitors of microtubule assembly (24-26), and HQ, either by itself or in combination with another benzene metabolite, catechol (CAT), is a potent inducer of aneuploidy in cultured human lymphocytes (27), suggesting a potential role for nondysjunctional events in the evolution of AML secondary to benzene exposure. Various studies have reported that repeated or prolonged exposure to benzene in vivo depresses the number of hematopoietic progenitor cells as defined by the colony-forming unit-spleen (CFU-S) and in vitro colony-forming (CFC) assays (28-33). Cronkite et al. (34) suggested a possible transitory increase in granulocyte/macrophage colony-forming units (CFU-GM) followed by suppression in mice chronically exposed to benzene. Recently, short-term benzene administration in vivo has been reported to result in a relative sparing of murine CFU-GM (35). Leukemogenesis is recognized to be a multifactorial process with the development of AML secondary to drug or chemical exposure involving more than a single direct cytotoxic effect on a cycling progenitor cell (6). Chemotherapeutic agents, such as hydroxyurea or vincristine, also target dividing progenitor cells, but their use in chemotherapy has not been associated with the development of AML. We have hypothesized that a distinguishing characteristic of agents with leukemogenic potential is their ability to produce intrinsic alterations in the regulation of stem cell differentiation in addition to targeting dividing cells. Such compounds might increase the absolute number of dividing cells in the target cell compartment at any given time. Given that malignant transformation of a hematopoietic progenitor cell involves genomic alterations, such as nondysjunctional events in dividing cells, a concomitant increase in the size of the cycling progenitor cell compartment would serve to increase the probability of transforming events in this target cell population. As an initial test of this hypothesis, we examined the influence of in vitro pretreatment with benzene metabolites on intrinsic colony-forming response of murine bone marrow cells stimulated with recombinant granulocyte/ macrophage colony-stimulating factor (rGM-CSF) in the absence of conditioned medium.

The effects of in vitro pretreatment with benABSTRACT zene metabolites on colony-forming response of murine bone marrow cells stimulated with recombinant granulocyte/macrophage colony-stimulating factor (rGM-CSF) were examined. Pretreatment with hydroqulinone (HQ) at concentrations ranging from picomolar to micromolar for 30 min resulted in a 1.5to 4.6-fold enhancement in colonies formed in response to rGM-CSF that was due to an increase in granulocyte/macrophage colonies. The synergism equaled or exceeded that reported for the effects of interleukin 1, interleukin 3, or interleukin 6 with GM-CSF. Optimal enhancement was obtained with 1 ,uM HQ and was largely independent of the concentration of rGM-CSF. Pretreatment with other authentic benzene metabolites, phenol and catechol, and the putative metabolite trans,trans-muconaldehyde did not enhance growth factor response. Coadministration of phenol and HQ did not enhance the maximal rGM-CSF response obtained with HQ alone but shifted the optimal concentration to 100 pM. Synergism between HQ and rGM-CSF was observed with nonadherent bone marrow cells and lineage-depleted bone marrow cells, suggesting an intrinsic effect on recruitment of myeloid progenitor cells not normally responsive to rGM-CSF. Alterations in differentiation in a myeloid progenitor cell population may be of relevance in the pathogenesis of acute myelogenous leukemia secondary to drug or chemical exposure.

Leukemias are clonal diseases originating in cells of the hematopoietic stem or progenitor cell compartment. In a majority of cases, the origin of acute myelogenous leukemia (AML) appears to be a cell capable of giving rise to myelo/ monocytic, erythroid, or megakaryocytic cells, monocyte or granulocyte progeny, or limited to the granulocytic pathway (1-6). AML occurs secondary to exposure to a number of cytotoxic agents, such as benzene, and various cancer chemotherapeutic agents (7-11). Secondary AML is distinguished from de novo AML by a high incidence of a "preleukemic phase" (i.e., cytopenias, myelodysplasias, panmyelosis) (6) and frequent cytogenetic abnormalities involving a loss of all or part of chromosomes 5 and 7 (12, 13). The majority of hematotoxic and leukemogenic compounds appears to target cycling cells with cytotoxicity mediated through interference with nucleic acid synthesis, cellular enzyme integrity, or microtubule and/or spindle formation. For such agents, target cell susceptibility is dependent on the rate ofcell proliferation (14). The hematotoxicity of benzene is well established in humans and experimental animals, chronic or repeated exposure resulting in lymphocytopenia, pancytopenia, and aplastic anemia (15, 16). Benzene metabolism is a requirement for toxicity, and the phenolic metabolites of

Abbreviations: AML, acute myelogenous leukemia; rGM-CSF, recombinant granulocyte/macrophage colony-stimulating factor; CFU-GM, granulocyte/macrophage colony-forming unit(s); CFU-M, macrophage colony-forming unit(s); HQ, hydroquinone; CAT, catechol; PH, phenol; BSA, bovine serum albumin; TM, trans,trans-muconaldehyde; IL, interleukin. §To whom reprint requests should be addressed.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. 3691

3692

Pharmacology: Irons et al.

MATERIALS AND METHODS Mice. Male C57BL/6 mice, 4-6 weeks old, were purchased from The Jackson Laboratory and held in quarantine for 1 week. The animals were housed with up to 10 mice per cage and were given food (3000, Agway, Syracuse, NY) and water ad libitum. Growth Factors. Murine rGM-CSF (50 units/ng) was a generous gift from Immunex (Seattle, WA). Bone Marrow Cells. Mice were killed by cervical dislocation and bone marrow was flushed from femora with phosphate-buffered saline (PBS) containing 1% bovine serum albumin (BSA) using a 5-ml syringe with a 22-gauge needle. A single cell suspension was obtained using a Pasteur pipet, which was then purified over a discontinuous gradient (Lympholyte-M, Accurate Scientific, Westbury, NY). The recovered buffy layer was removed and washed twice in PBS/ BSA. Nonadherent cells were obtained by incubating the cells at 2 x 106 per ml in culture flasks for 1 hr at 370C in PBS/BSA. The number of nucleated cells was determined with a hemocytometer using Turk's solution and cell viabilities were determined by trypan blue exclusion. Cell suspensions were kept on ice during all procedures unless otherwise noted. Enrichment of Hematopoietic Progenitor Cells. Bone marrow cells were obtained as described above, incubated with an antibody cocktail of lineage-specific antibodies (either rat IgG or IgM) for 45 min at 40C, and washed twice with PBS/BSA to remove unbound antibody. Lineage-specific antibodies consisted of CD4 (GK1.5), CD8 (53-6.72), B220 (RA3-3A1/6.1), and Mac-1 (M1/70.15.11.5) [American Type Culture Collection (ATCC)] and Gr-1 (RB6-8C5), which was obtained from Irving Weissman (Stanford University). Antibodies were purified using protein G and titered for optimal binding. Labeled cells (1.4-1.5 x 108) in 1 ml of PBS/BSA were applied to a chromatography column of 1.6-cm internal diameter containing Sepharose 4B linked with anti-rat K monoclonal antibody (MAR 18.5, ATCC). The bed height of the affinity column was 4.7 cm and the flow rate was 9.8 ml/cm2 per hr. The purity of lineage negative cells was determined by flow cytometry, using fluorescein-conjugated anti-rat K as the secondary reagent. Flow Cytometry. Cells were analyzed on a model 753 EPICS V flow cytometer (Coulter). Logarithmically integrated green fluorescence was gated on the forward angle, and 900 light scatter of live cells was measured on 10,000 cells per sample. Chemical Exposure. Nonadherent bone marrow cells or enriched progenitor cells were aliquoted into control and treatment groups, centrifuged at 200 x g for 5 min, and resuspended in PBS without BSA. Concentrations of HQ (Sigma), phenol (PH) (Sigma), CAT (Sigma), or trans,transmuconaldehyde (TM) (supplied by James Ruth, University of Colorado) or PBS was added to each suspension (2 million cells per ml). Cells were incubated at 37°C for 30 min, complete Iscove's medium was added, and cells were centrifuged for 10 min at 200 x g. Cells were resuspended in complete medium and cultured. Except where noted, cell viabilities determined after 30 min of pretreatment for different concentrations of compounds were >90%. CFU-C Assay. Cells were plated in 35-mm culture dishes at a concentration of 3 to 5 x 104 cells per ml in 1 ml of modified Iscove's medium containing 10% heat-inactivated fetal bovine serum (Life Technology, Grand Island, NY), 50 mM 2-mercaptoethanol, 1.2% (wt/vol) methyl cellulose, and rGM-CSF (either 5 or 10 ng/ml unless otherwise stated). Colonies consisting of 50 or more cells were scored on days 8 and 14 using a dissecting microscope. Correlation of colony morphology with individual cell types present was confirmed by phase-contrast microscopy of methyl cellulose colonies in

Proc. Natl. Acad. Sci. USA 89 (1992) situ and light microscopic examination of Cytospin prepara-

tions of aspirated cells fixed and stained with modified Wright's stain. Statistical Analysis. Five plates were scored for each treatment group and the results are expressed as the mean ± 1 SEM. Significant differences between pretreated and control groups were determined using the Student's t test and were calculated using absolute values.

RESULTS Influence of Phenolic Metabolites of Benzene on ColonyForming Cell Response to rGM-CSF in Murine Bone Marrow. The colony-forming response of nonadherent Ficoll-purified murine bone marrow cells cultured in the presence of different concentrations of rGM-CSF is shown in Fig. 1. Granulocyte/macrophage colonies (CFU-GM) predominate over a wide range of growth factor concentration, with CFU-GM response plateauing at =2.5 ng of rGM-CSF per ml. Given these culture conditions, the number of granulocyte colonies (CFU-G) formed is negligible; however, macrophage colonies (CFU-M) are observed in small numbers. No significant differences in colony number or type are noted between day 8 and day 14 cultures for any treatment group. Therefore, only day 8 culture data are presented. HQ pretreatment of bone marrow cells for 30 min results in a significant enhancement of GM-CSF-induced CFU-GM over a range of concentrations (10-8-10-5 M), with maximal enhancement consistently observed at 10-6 M. Higher concentrations (10-3_10-2 M) result in suppression of CFU-GM. Pretreatment of bone marrow cells with HQ in the absence of GM-CSF results in no colony formation. The HQ-induced increase in colony-forming response is almost entirely due to an increase in CFU-GM colonies. The results of two representative experiments are shown in Fig. 2. The results of three additional separate experiments are presented in Table 1. The increase in the number of responding cells relative to controls ranges between 1.5- and 4.6-fold. Although the magnitude of the HQ-induced enhancement appears to be influenced by the concentration of GM-CSF (Table 1), optimal stimulation invariably occurs at 10-6 M HQ following pretreatment with HQ alone. Accordingly, the concentration-response of HQ-induced enhancement appears to be relatively independent of either the amount of GM-CSF present or the magnitude of baseline variation in rGM-CSFinduced CFU-GM in individual experiments. In contrast to HQ, pretreatment with CAT produces no significant effect on CFU-GM over an extended range of concentrations (10-1o10-2 M) (Fig. 3A), although cytotoxicity is observed at pretreatment concentrations of 10-2 M or greater. Similarly, pretreatment of cells with PH alone does not enhance colony formation at any concentration but suppresses CFU-GM at co

150

o0

100-

4a

50-

Co

0 0~

o0

n JI0

5

10

15

20

[GM-CSF] (ng/ml)

FIG. 1. Concentration dependence of colony formation on rGMCSF in murine bone marrow cells. Nonadherent bone marrow cells were plated at a density of 1 x 104 cells per dish and cultured in the presence of various concentrations of the growth factor for 8 days. *, Mixed granulocyte/macrophage colonies; *, macrophage colonies; *, granulocyte colonies. Error bars indicate 1 SEM for five

cultures and are omitted when they are smaller than the symbol.

Pharmacology: Irons et al. .9

120' 90-

0

60

075

30 -

0

a)

i

25o 25 0Q Cof

0

0

0

-

Control 12 a

200-

0 o

150-

gin

100-

0

f96% depleted of cells expressing lineage-specific markers (Fig. 7). Pretreatment of this enriched progenitor cell population with HQ results in a 2-fold enhancement of GM-CSF-induced CFU-GM comparable to that obtained in routine bone marrow cell preparations (Fig. 8). The marginal increases in CFU-M observed in bone marrow preparations pretreated with HQ are completely abrogated in lineage-depleted cells (data not shown), suggesting that any effect of HQ on CFU-M either is mediated via a more differentiated responding cell population not present in lineage-depleted cells or requires accessory cell involvement.

DISCUSSION GM-CSF is an important growth factor that supports the growth and differentiation of myeloid progenitor cells. Cells responding to GM-CSF are resting cells that are rapidly recruited into cycle, the earliest cell giving rise to CFU-GM being the multipotential myeloid progenitor cell (41-43), 120

en

80

41

0 0

200

40

-a

0

0

300

0

0 0 a)

250

implicated as a target cell in the development of AML. Pretreatment of murine bone marrow cells with HQ at concentrations between 10-1o and 10-6 M results in an enhancement of CFU-GM in response to GM-CSF. The magnitude of HQ-enhanced colony formation equals or exceeds that described for the synergistic action of IL-3, IL-1, and IL-6 in combination with GM-CSF (41, 44). Pretreatment of bone marrow cells with HQ in the absence of GM-CSF results in no colony formation, suggesting that production of GM-CSF, IL-3, M-CSF, G-CSF, IL-1, or IL-6 cannot explain these results. In addition, the magnitude of HQ-induced enhancement of CFU-GM far exceeds that observed by varying the concentration of either GM-CSF or IL-3 in this culture system (IL-3 data not shown). A number of additional factors suggest that HQ-induced enhancement of CFU-GM represents an intrinsic effect on GM-CSF response of progenitor cells rather than an effect mediated by accessory cells. ILs thought to play an accessory role in myeloid differentiation are IL-1 and IL-6, both of which have been reported to exert synergistic effects indirectly via the production of GM-CSF or IL-3 (41). In contrast to interactions reported for these growth factors and c-kit ligand (45), no difference in colony size or type is observed between control and HQ-pretreated cultures, suggesting that increased recruitment of CFC is not accompanied by an increase in cell proliferation rate. The enhancement in colony number observed in HQ-pretreated cultures results in a uniform distribution of colonies on the plate rather than a cluster or "burst" pattern characteristic of accessory cell participation and is observed in preparations depleted of stromal and macrophage cells. Finally, the potency of HQ-induced ena)

0 0

II

50 100 150 200 Log Fluorescence Intensity

FIG. 7. Flow cytometric histogram of bone marrow cells incubated with lineage-specific antibodies (-) or mouse anti-rat K secondary reagent (-). (A) Nonadherent bone marrow cells. (B) Affinity-purified bone marrow cells. The vertical line at channel 25 is an artifact created by the discriminator of the flow cytometer.

4 2 6 8 Control 10 - Log [Hydroquinone] (M)

12

0

2

100-

0

a

0

4 6 8 Control 10 - Log [Muconaldehyde] (M)

FIG. 6. Effects of pretreatment with various concentrations of TM on rGM-CSF-stimulated colony formation. m, Mixed granulocyte/macrophage colonies; *, macrophage colonies. Error bars indicate the SEM for five cultures and are omitted when they are smaller than the symbol. rGM-CSF concentration = 10 ng/ml. *, Difference significant from controls stimulated with rGM-CSF only (P 0.05). -

100

50

0

2

10C

af)

00-

a)

Control 10 15C )~B

-)

24

z

50 - I

A

0

r

7 6 5 9 Control 12 - Log [Hydroquinone] (M)

3

FIG. 8. Effects of HQ pretreatment on rGM-CSF-induced CFU-GM in lineage-depleted hematopoietic progenitor cells. Error bars indicate the SEM for five cultures and are omitted when they are smaller than the symbol. rGM-CSF concentration = 5 ng/ml. *, Difference significant from controls stimulated with rGM-CSF only (P 0.05). -

Pharmacology:

Irons

et

al.

hancement is not diminished in hematopoietic progenitor cells essentially depleted of lymphocytic, macrophage, and granulocytic cells. The most likely explanation for HQ-induced enhancement of CFU-GM is the recruitment of an early hematopoietic precursor population that would normally not respond to GM-CSF under the conditions employed in these assays. Previous studies have revealed that repeated benzene administration in vivo is accompanied by an initial increase in bone marrow cell turnover that is maximal at 4 hr but declines to control rates within 8 hr (46). Together with the transitory increase in GM colonies reported by Cronkite et al. (34), these findings suggest that in vivo exposure to benzene results in a shift in a population of resting cells into replicative cycle consistent with the increased recruitment of a hematopoietic progenitor cell population. The potential of HQ to alter intrinsic growth factor response and induce differentiation in a myeloid progenitor cell population may be important in the pathogenesis of AML secondary to benzene exposure. Benzene leukemogenesis may result from the dual ability of its metabolites to promote progenitor cell differentiation and induce cytogenetic changes in replicating cells. If other leukemogenic agents act similarly, alterations in myeloid progenitor cell differentiation may be important in the pathogenesis of secondary AML in general. HQ has been shown to be a potent disrupter of microtubule assembly via covalent interaction with SH groups required for GTP binding to the tubulin molecule (24-26). In addition, HQ, by itself or in synergy with CAT, has been demonstrated to produce aneuploidy in cultured human lymphocytes (27). Mechanisms to explain HQ-induced enhanced response to GM-CSF may include alterations in receptor-ligand affinity via direct binding to SH groups associated with the receptor complex, modulation of membrane-associated cytoskeletal elements, or direct effects on signal transduction or gene expression. Understanding the significance of altered progenitor cell growth factor response and its role in secondary leukemogenesis will require a critical examination of these alternatives. We thank Immunex Corporation, Seattle, WA, for their generous gift of rGM-CSF and acknowledge the support of the flow cytometry core facility of the University of Colorado Cancer Center. These studies were supported in part by grants from the American Petroleum Institute, the Chemical Manufacturers Association, and The Center for Space Environmental Health/National Aeronautics and Space Administration (NAGW-2356). 1. Fialkow, P. J., Denman, A. M., Singer, J. W., Jacobson, R. J. & Lowenthal, M. N. (1978) in Differentitation of Normal and Neoplastic Hematopoietic Cells, eds. Clarkson, B., Marks, P. A. & Till, J. E. (Cold Spring Harbor Lab., Cold Spring Harbor, NY), pp. 131-144. 2. Fialkow, P. J., Singer, J. W., Adamson, J. W., Berkow, R. L., Friedman, J. M., Jacobson, R. J. & Moohr, J. W. (1979) N. Engl. J. Med. 301, 1-5. 3. Fialkow, P. J., Singer, J. W., Adamson, J. W., Vaidya, K., Dow, L. W., Ochs, J. & Moohr, J. W. (1981) Blood 57, 1068-1073. 4. Najfeld, V., Zucker-Franklin, D., Adamson, J. W., Singer, J. W., Troy, K. & Fialkow, P. J. (1988) Leukemia 2, 351-357. 5. Fialkow, P. J., Singer, J. W., Raskind, W. H., Adamson, J. W., Jacobson, R. J., Bernstein, I. D., Dow, L. W., Najfeld, V. & Veith, R. (1987) N. Engl. J. Med. 317, 468-473. 6. Foucar, K., McKenna, R. W., Bloomfield, C. D., Bowers, T. K. & Brunning, R. D. (1979) Cancer 43, 1285-12%. 7. McBride, G. (1977) J. Am. Med. Assoc. 237, 2697-2698. 8. Allan, W. S. A. (1970) Lancet ii, 775. 9. Kyle, R. A., Pierre, R. V. & Bayrd, E. D. (1975) Arch. Intern. Med. 135, 185-192.

Proc. Natl. Acad. Sci. USA 89 (1992)

3695

10. Kyle, R. A., Pierre, R. V. & Bayrd, E. D. (1970) N. Engl. J. Med. 283, 1121-1125. 11. Kantarjian, H. M. & Keating, M. J. (1987) Semin. Oncol. 14, 435-443. 12. Le Beau, M. M., Albain, K. S., Larson, R. A., Vardiman, J. W., Davis, E. M., Blough, R. R., Golomb, H. M. & Rowley, J. D. (1986) J. Clin. Oncol. 4, 325-345. 13. Koeffler, H. P. & Rowley, J. D. (1985) in Neoplastic Diseases of the Blood, eds. Wiernik, P. H., Canellos, G. P., Kyle, R. A. & Schiffer, C. A. (Churchill Livingstone, New York), pp. 357-381. 14. Van Putten, L. M. & Lelieveld, P. (1971) Eur. J. Cancer 7, 11-16. 15. Laskin, S. & Goldstein, B. D. (1977) J. Toxicol. Environ. Health 2, 1-148. 16. Greenburg, L., Ma~yers, M. R., Goldwater, L. J. & Smith, A. R. (1939) J. Ind. Hyg. Toxicol. 21, 395-420. 17. Eastmond, D. A., Smith, M. T. & Irons, R. D. (1987) Toxicol. Appl. Pharmacol. 91, 85-95. 18. Rickert, D. E., Baker, T. S., Bus, J. S., Barrow, C. S. & Irons, R. D. (1979) Toxicol. Appl. Pharmacol. 49, 417-423. 19. Wierda, D. & Irons, R. D. (1982) Immunopharmacology 4, 41-54. 20. Irons, R. D. (1985) J. Toxicol. Environ. Health 16, 673-678. 21. Muirhead, K. A., Irons, R. D., Bruns, R. & Horan, P. K. (1980) J. Histochem. Cytochem. 28, 526-532. 22. Horan, P. K., Muirhead, K. A., Gorton, S. & Irons, R. D. (1980) Lab. Anim. Sci. 30, 76-79. 23. Irons, R. D. (1981) Environ. Health Perspect. 39, 39-49. 24. Irons, R. D., Neptun, D. A. & Pfeifer, R. W. (1981) J. Reticuloendothel. Soc. 30, 359-372. 25. Irons, R. D. & Neptun, D. A. (1980) Arch. Toxicol. 45, 297305. 26. Irons, R. D., Pfeifer, R. W., Aune, T. M. & Pierce, C. W. (1984) J. Immunol. 133, 2032-2036. 27. Robertson, M. L., Eastmond, D. A. & Smith, M. T. (1991) Mutat. Res. 249, 201-209. 28. Green, J. D., Snyder, C. A., LoBue, J., Goldstein, B. D. & Albert, R. E. (1981) Toxicol. Appl. Pharmacol. 58, 492-503. 29. Seidel, H. J., Barthel, E. & Zinser, D. (1989) Exp. Hematol. 17, 300-303. 30. Cronkite, E. P., Inoue, T., Carsten, A. L., Miller, M. E., Bullis, J. E. & Drew, R. T. (1982) J. Toxicol. Environ. Health 9, 411-421. 31. Frash, V. N., Yushkov, B. G., Karaulov, A. V. & Skuratov, V. L. (1976) Bull. Exp. Biol. Med. 83, 985-987. 32. Gill, D. P., Jenkins, V. K., Kempen, R. R. & Ellis, S. (1980) Toxicology 16, 163-171. 33. Uyeki, E. M., El Ashkar, A., Shoeman, D. W. & Bisel, T. U. (1977) Toxicol. Appl. Pharmacol. 40, 49-57. 34. Cronkite, E. P., Drew, R. T., Inoue, T., Hirabayashi, Y. & Bullis, J. E. (1989) Environ. Health Perspect. 82, 97-108. 35. Dempster, A. M. & Snyder, C. A. (1990) Arch. Toxicol. 64, 539-544. 36. Goldstein, B. D., Witz, G., Javid, J., Amoruso, M. A., Rossman, T. & Wolder, B. (1982) Adv. Exp. Med. Biol. 136, 331-339. 37. Witz, G., Rao, G. S. & Goldstein, B. D. (1985) Toxicol. Appl. Pharmacol. 80, 511-516. 38. Witz, G., Maniara, W. M., Mylavarapu, V. J. & Goldstein, B. D. (1990) Biochem. Pharmacol. 40, 1275-1280. 39. Witz, G., Kirley, T. A., Maniara, W. M., Mylavarapu, V. J. & Goldstein, B. D. (1990) in Biological Reactive Intermediates IV, ed. Witmer, C. M. (Plenum, New York), pp. 613-618. 40. Moore, M. A. S. (1991) Blood 78, 1-19. 41. Migliaccio, G., Migliaccio, A. R. & Adamson, J. W. (1991) Exp. Hematol. 19, 3-10. 42. Griffin, J. D., Cannistra, S. A., Sullivan, R., Demetri, G. D., Ernst, T. J. & Kanakura, Y. (1990) Int. J. Cell. Cloning 8, 35-45. 43. DiPersio, J. F., Golde, D. W. & Gasson, J. D. (1990) Int. J. Cell. Cloning 8, 63-75. 44. Schaafsma, M. R., Falkenburg, J. H. F., Duinkerken, N., van Damme, J., Altrock, B. W., Willemze, R. & Fibbe, W. E. (1989) Blood 74, 2398-2404. 45. McNiece, I. K., Stewart, F. M., Deacon, D. M. & Quesenberry, P. J. (1988) Exp. Hematol. 16, 383-388. 46. Irons, R. D., Heck, H. d'A., Moore, B. J. & Muirhead, K. A. (1979) Toxicol. Appl. Pharmacol. 51, 399-409.