Original Paper. Introduction

Transcription

1 Original Paper International Journal of Cell Cloning 1: (1992) Retinyl Acetate and All-Trans-Retinoic Acid Enhance Erythroid Colony Formation In Vitro by Circulating Human Progenitors in an Improved Serum-Free Medium Paul N. Correa, Arthur A. Axelrad Department of Anatomy and Cell Biology, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada Key Words. Retinoids IGF-I Epo Erythropoiesis Serum-free medium Abstract. Retinyl acetate (RA) dramatically increased the production of early (d16) erythroid colonies in vitro by circulating human progenitor cells growing in an improved serum-free (SF) medium. In the absence of either erythropoietin (Epo) or insulin-like growth factor I (IGF-I), RA alone was able to induce the hemoglobinization of cells in these erythroid colonies. RA synergized with Epo or with IGF-I to yield increased numbers of well-hemoglobinized early colonies. In the presence of defined burst promoting activity (BPA) provided by recombinant human interleukin 3 (rhul-3) and hemin, RA and all-trans-retinoic acid (ATRA) were identical with respect to their ditferentiation-inducing function for early erythroid colonies. ATRA increased the number of these colonies in a concentration-dependent manner, with maximal stimulation (3.5-fold) occurring at 3 nm in the presence of 5.5 ng/d L-3,.1 mm hemin, 3. U/ml Epo and 3 nm IGF-I. This appears to be the first demonstration of erythropoietic activity of two metabolic derivatives of vitamin A in SF medium. Introduction Vitamin A, its metabolites and its analogues have long been known to play a fundamental role in the maintenance of healthy hemopoietic tissue growth and differentiation [ Early morphological studies showed that hemopoiesis was substantially depressed in vitamin A-deprived rats [l, 21. Retinoids have also been shown to block the proliferation of malignant myeloid and melanoma cells in vivo and of transformed cell lines in vitro [ More recently, there have been several re- Correspondence: Dr. A. A. Axelrad, Department of Anatomy, Medical Sciences Building, University of Toronto, Toronto, Ontario, Canada M5S 1A8. Received March ; provisionally accepted May 1, 1992; accepted for publication June 29, /92/$2./ OAlphaMed Press ports of successful antileukemic therapy with alltrans-retinoic acid (ATRA) [ However, very little information exists on the effects of retinoids on the clonal growth of hemopoietic progenitor cells in culture. In the early 198s, Douer and Koefler published two studies in vitro on the effects of retinoids using normal human peripheral blood mononuclear cells (PBMNC), one on erythroid and the other on myeloid progenitors [19, 21. These studies were performed in the presence of a high concentration of fetal calf serum (FCS), which contains both retinol and retinyl esters [21]. Moreover, their erythroid culture system employed a T lymphocyte conditioned medium that contained an undefined erythroid potentiating activity (EPA) active on late erythroid colony forming unit (CFU-E) progenitor cells [22], as well as an undefined burst-promoting activity (BPA) which exerts its effect on early erythroid burst-forming unit (BFU-E) progenitors. A BPA [23, 241 has also been shown to be present as a contaminant in semi-purified erythropoietin (Epo) preparations [25] such as the one used by Douer and Koefler [ 191, and in serum albumin [26]. It is also known that serum albumin binds retinoids nonspecifically with high affhity and that it is the major vehicle for their plasma transport [27]. We have recently developed an improved serumfree (SF) medium for the growth of erythroid bursts and colonies from PBMNC, which contains a defined BPA-like activity, BPA-free albumin, and optimal concentrations of the recombinant hemopoietic growth factors recombinant human insulin-like growth factor I (rhuigf-i) and rhuepo [28]. In this study, we have investigated the effects of retinyl acetate (RA) and of ATRA on the production in vitro of d16 (early) erythroid colonies on a retinoid-depleted background. We have found that both retinoids potentiate the effects of Epo and IGF-I,

2 281 RA and ATRA Enhance Erythroid Differentiation of PBMNC each of which can also function as a normal regulator of erythroid differentiation on its own. Materials and Methods Cell Preparation After informed consent, human peripheral blood was removed by venipuncture and immediately layered over alpha-minimal essential medium (a-mem) containing 2% FCS (#2-614; Gibco, Grand Island, NY) and 1 U/ml of preservative-free sodium heparin (# MF; Gibco). Mononuclear cells were separated by Ficoll-Paque (Pharmacia, Montreal, Quebec) density-gradient centrifugation at 4 g for 4 min. Cell suspensions from the interface were washed three times, only the first wash containing FCS (2%), and cell counts were performed with.5% Trypan Blue (#63-525; Gibco) in phosphate buffered saline (PBS). Retinoids ATRA preparations (Sigma Chemical Co., St. Louis, MO) were stored at -2 C. ATRA was prepared following Douer and Koeffler s method [19]. All dilutions were made directly in a-mem. Vitamin A acetate (Nutritional Biochemicals Corp., Cleveland, OH) at 3 x lo- M was dissolved together with the SF-lipid preparation and stored at 4 C. All vessels containing retinoids were covered with aluminum foil, and all plating in medium containing retinoids was performed in subdued light. Serum-free Medium We have recently developed an improved SF medium for the growth in vitro of erythroid progenitor cells from normal human peripheral blood [28]. Briefly, this medium employs a defatted, BPAfree, Epo-free and globulin-free bovine serum albumin (BSA) (#A-281; Sigma) plus all the components previously found necessary for murine BFU-E culture [29], and these are supplemented with 1 pg/ ml of each deoxy- and ribonucleoside (Sigma) and 7 x lo- M d-a-tocopherol(18 pure; Clinic Products, Windsor, Ontario). To provide a defined BPA, we have added 5.5 ng/ml rhuil-3 (Amgen, Thousand Oaks, CA) and.1 mm hemin (Sigma). The complete SF medium also contained 3 U/ml rhuepo, 3 x M rhuigf-i (generous gifts from Amgen, Thousand Oaks, CA, courtesy of Dr. J. Egrie) and 3 x M RA (Nutritional Biochemicals Cop, Cleveland, OH). All media were mixed with.8% methylcellulose (Methocel A4M, premium grade; Dow Chemical Co., Midland, MI) in multi-wells (#76--4; Flow Laboratories, MacLean, VA), and the cultures were incubated at 37 C in a humidified atmosphere and 5% CO, for indicated time periods. Scoring of Erythroid Colonies Early erythroid colonies were scored in situ at d14-16 as clusters of 25 hemoglobinized cells. We have previously [28] defined an erythroid burst as either a single colony or a cluster of burst-component or day 14 colonies or subcolonies, each having 15 hemoglobinized (or benzidine-positive) cells, scored at days 14 to 16 of growth. Thus the colonies we have counted are not bursts (except those bursts that comprise only a single colony of 25 cells). We have also studied the effect of these retinoids upon burst formation and d8 CFU-E-derived colony formation, but we have not included the data; they show that in the absence of retinoids no d8 colonies mature. Statistical Analysis Student s t-test (two-tailed) was employed to determine the significance of differences between any two sets of data. Results Addition of RA at physiological serum concentration (3 nm) [21] to a minimal SF medium, which contained exogenous BPA-like activity (5.5 ng/ml rhuil-3 and.1 mm hemin) and 1% BPAfree BSA, partially replaced Epo in the production of d16 erythroid colonies from PBMNC (Fig. 1, compare bars 1, 2 and 3). Colonies obtained with RA displayed a light hemoglobinization which was yellow-orange in color (Fig. 1, bar 2). Addition of RA to rhuepo dramatically increased the number of d 16 colonies -fourfold (Fig. 1, bar 3 versus bar 4). A similar relationship was observed for RA and IGF-I; addition of the retinoid to 3 x 1 M rhuigf-i increased the number of d16 colonies -fivefold (Fig. 1, bar 5 versus bar 6). When rhuigf-i was added at 3 x M to the retinoid, the number of d16 colonies increased almost tenfold over what was obtained with RA alone (Fig. 1, bar 2 versus bar 6) (compare ref. 28). Lastly, when RA was added to both Epo and IGF-I, d16 colonies increased fivefold (Fig. 1, bar 7 versus bar 8). It was also in the groups containing all the growth factors that the most intense hemoglobinization was observed (Fig. 1, bar 8). All these data demonstrate that the actions of RA and either Epo or IGF-I are synergistic for d16 erythroid colony production.

3 CorrealAxelrad 288 f m 8r p 5rp$);L Hemin 1 mm rhu Epo 3 Ulml rhu IGF4 3 x 1% Retinyl acetate d16 e w~ony wbur Fig. 1. Effect of retinyl acetate on the production of d16 (early) erythroid colonies by circulating erythroid progenitors in an improved SF medium. All assays were set up with a modification of the method of Stewart et al. [29] and employed a defatted, BPA-free, Epo-free and globulin-free BSA. Colonies had 25 hemoglobinized (orange to red) cells B 1- P) r; z All-trans retinoic acuj 3 nm rhu Epo 3. U/ml rhu-il-3 Hemin.1 mm rhu IGF-I 311% 4 OD R o Rnk orang. Yellow oraw Orange red T Fig. 2. Effect of ATRA on early erythroid colony fonnation by circulating progenitors in an improved SF medium. Colonies had 15 hemoglobinized cells (orange to red in color). To find out how ATRA behaves under these conditions, we substituted ATRA for RA (at 3 nm) in the same SF medium. We found that in the absence of any added BPA-like activity (IL-3 or hemin) and of any Epo-like activity (Epo or IGF-I), ATRA was not capable by itself of supporting early colony growth and differentiation (Fig. 2, bar 1). In the presence of Epo, the normal regulator of erythropoietic differentiation, ATRA did not provide a BPAlike activity in that no d16 colony formation was detected (Fig. 2, bar 2). In the presence of rhuil-3, erythroid colonies could be morphologically recognized, but no hemoglobinization could be detected (Fig. 2, bar 3). However, a moderate number of hemoglobinized colonies could be detected when hemin was present at.1 mm (in the absence of IL-3), showing that hemin by itself has BPA-like properties (Fig. 2, bar 4). In the presence of IL-3 (at optimal concentration) and.1 mm hemin, and in the absence of Epo, similarly to RA, ATRA induced some lightly hemoglobinized colonies, though many more colonies could be observed to be erythroid-like in morphology and not hemoglobinized (Fig. 2, bar 5). Substituting optima1 concentrations of Epo and IGF-I for ATRA (on an IL-3 + hemin background) yielded a comparable number of d 16 colonies (not significantly different from bar 5, Fig. 2; t = 2.46 for 2 degrees of freedom and p c.5) and a much stronger degree of hemoglobinization (Fig. 2, bar 6); still, other erythroid-like colonies could be observed which had not matured. However, colony maturation was dramatically increased when ATRA was added to complete the SF medium (Fig. 2, bar 7), as it induced a fourfold increase in the number of hemoglobinized d16 colonies. We believe that this is not so much a result of an increase in the actual number of colonies associated to form each burst as it is a result of the retinoids promoting the hemoglobinization of those colonies that are already there, hence permitting their scoring as recognizable erythroid colonies. Thus ATRA appears to function in a manner analogous to RA, both retinoids acting as essential cofactors of erythroid differentiation. Lastly, titration of ATRA was performed in the improved SF medium, i.e., in the presence of the full complement of erythropoietic growth factors. Without ATRA, this medium was capable of supporting the growth of 9 well-hemoglobinized d16 colonies having -5-1 cells per colony. In the presence of ATRA at concentrations between 3 x lo- and 3 x 1% M, we found that the number of d16 colonies increased proportionately with the log of concentration, to a maximum of 4.5-fold (Fig. 3). However, no plateau of activity was reached at 3 x lo-* M;

4 289 RA and ATRA Enhance Erythroid Differentiation of PBMNC Concentration of ATFiA (M) Fig. 3. Early colony formation by circulating erythroid progenitors in an improved SF medium as a function of ATRA concentration. Final concentrations: 5.5 ng/ml rhuil-3,.1 mm hemin. 3. U/ml Epo and 3 x lo- M IGF-I, and 1% defatted BPA-free and Epo-free BSA. Colonies had 25 hemoglobinized cells. instead an optimal concentration of ATRA was found, and at 3 x lo- M the number of colonies significantly decreased. However, the highest cellularity of colonies (-2 cellskolony) was observed at this concentration. Further, greater ATRA concentrations dramatically reduced the cellularity of the hemoglobinized colonies (to -7 cellskolony) as well as their overall numbers, and toxic effects became extreme at 3 x M, at which level no colonies, erythroid or otherwise, were detectable (only single, colorless cells and monocytic doublets could be seen). This suggests that the effect of ATRA at high concentration may be to block proliferation of erythroid progenitor cells. Discussion We have shown that d16 erythroid colonies can be obtained in the absence of RA, the addition of Epo or IGF-I alone being sufficient for their maturation. Nevertheless we observed a clear synergism of these two growth factors with RA, as the combined effects of the retinoid with either Epo or IGF-I were greater than the sum of their separate effects. By itself, RA could promote the maturation of a low number of d 16 colonies, but their hemoglobinization remained light. The most intense hemoglobinization, as well as the highest cellularity of these colonies, was observed when all three factors were added together. This then suggests that RA acts both as an erythroid differentiation factor on its own and as a cofactor that potentiates other erythroid growth and differentiation factors. These findings are supported by the results obtained with ATRA, which suggest that the stimulatory activity of these retinoids is Epo-like (i.e., promotes erythroid differentiation) and not BPA-like (i-e., does not promote proliferation). Because of the dramatic potentiation observed, it appears that RA or ATRA must be considered indispensable for the optimization of erythroid growth in SF medium. These retinoids may thus also be essential for normal erythropoiesis. In our system, ATRA appeared to reach maximal erythroid stimulatory activity at 3 nm, three times the physiological concentration in normal human plasma [3, 311, and it increased the colony number by -45%. In their assay system, Douer and Koeffler also found that maximal stimulation of BFU-E-derived colony growth by ATRA occurred at 3 nm and was on the order of 225% over controls without ATRA [ 191. However, in their study, erythroid colonies were grown in the presence of 3% FCS. This is important because the overall efficiency of early erythroid colony formation in our SF medium appeared to be more than four times higher than that reported by Douer and Koeffler (4 colonies of >5 cells per 15 PBMNC versus 2 colonies of >4 cells per 2 x lo5 PBMNC) in their serum-containing medium. This may be the combined result of the presence, in our medium, of other critical growth factors (specifically, hemin and IGF-I) and the absence of inhibitory factors normally present in serum.

5 CorredAxelrad 29 Acknowledgments promyelocytic leukemia. Br J Haematol 1984;57: This work was SupPofled by a grant from the 15 Daenen S, Vellenga E, van Dobbenbugh OA, Halie Medical Research Council of Canada. MR. Retinoic acid as antileukemic therapy in a patient with acute promyelocytic leukemia and Aspergillus pneumonia. Blood 1986;67: References Findlay GM, McKenzie RD. The bone marrow in deficiency diseases. J Pathol Bacteriol 1922;25: Wolbach SB, Howe PR. Tissue changes following deprivation of fat soluble A vitamin. J Exp Med 1925;42: Hodges RE, Sauberlich HE, Canham JE, et al. Hematopoietic studies in vitamin A deficiency. Am J Clin Nutr 1978;31: Felix EL, Lloyd B, Cohen MH. Inhibition of the growth and development of a transplantable murine melanoma by vitamin A. Science 1975;189: Harisiadis L, Miller RC, Hall EJ, Borek C. A vitamin A analogue inhibits radiation-induced oncogenic transformation. Nature 1978;274: Flynn P, Miller W, Weisdorf D, Arthur D. Banning R, Branda R. Retinoic acid treatment of acute promyelocytic leukemia: in vitro and in vivo observations. Blood 1983;62: Lotan R, Giotta G, Nork E, Nicolson GL. Characterization of the inhibitory effects of retinoids on the in vitro growth of two malignant murine melanomas. J Natl Cancer Inst 1978;6: Lotan R. Different susceptibilities of human melanoma and breast carcinoma cell lines to retinoic acid-induced growth inhibition. Cancer Res 1979;39: Lotan R, Nicolson GL. Heterogenicity in growth inhibition by P-trans-retinoic acid of metastatic B16 melanoma clones and in vivo-selected cell variant lines. Cancer Res 1979;39: Meyskens F, Salmon S. Inhibition of human melanoma colony formation by retinoids. Cancer Res 1979;39: Meyskens F, Fuller BB. Characterization of the effects of different retinoids on the growth and differentiation of a human melanoma cell line and selected subclones. Cancer Res 198; Breitman TR, Collins SJ, Keene BR. Terminal differentiation of human promyelocytic leukemic cells in primary culture in response to retinoic acid. Blood : 1OOO- 14. Koeffler HP. Induction of differentiation of human acute myelogenous leukemia cells: therapeutic implications. Blood 1983;62: Nilsson B. Probable in vivo induction of differentiation by retinoic acid of promyelocytes in acute 16 Fontana JA, Roger JS, Durham JP. The role of 13-cis retinoic acid in the remission induction of a patient with acute promyelocytic leukemia. Cancer 1986;57: Meng-er H, Yu-chen Y, Shu-rong C, et al. Use of all-trans retinoic acid in the treatment of acute promyelocytic leukemia. Blood 1988;72: Castaigne S, Chomienne C, Ballerini P. et al. Alltrans retinoic acid a novel differentiation therapy for acute promyelocytic leukemia. Blood 1989;74 (suppl 1): Douer D, Koeffler Hp. Retinoic acid enhances growth of human early erythroid progenitor cells in vitro. J Clin Invest 1982;69: Douer D. Koeffler Hp. Retinoic acid enhances colonystimulating factor-induced clonal growth of normal human myeloid progenitor cells in vitro. Exp Cell Res 1982; 138: HyClone Laboratories. Fetal bovine serum: variation in components that influence cell growth and function. Art to Science in Tissue Culture Golde DW, Bersch N, Quan SG, Lusis AJ. Production of erythroid-potentiating activity by a human T-lymphoblast cell line. Proc Natl Acad Sci USA 198;77: Porter PN, Ogawa M. Erythroid burst-promoting activity (BPA). In Dunn CDR ed. Current Concepts in Erythropoiesis. New York, Wiley & Sons, 1983: Linch DC, Lipton JM, Nathan DG. Identification of three accessory cell populations in human bone marrow with erythroid burst-promoting properties. J Clin Invest 1985;75: Konwalinka G. Geissler D, Peschel C, et al. Human erythropoiesis in vitro and the source of burst-promoting activity in serum-free system. Exp Hematol 1986;14: Akahane K, Tojo A, Urabe A, Takaku F. Pure erythropoietic colony and burst formations in serum-free culture and their enhancement by insulin-like growth factor I. Exp Hematol 1987;15: Sani BP, Titus BC, Banejee CK. Determination of binding affinities of retinoids to retinoic acid-binding protein and serum albumin. Biochem J 1978;171: Correa PN, Axelrad AA. Production of erythropoietic bursts by progenitor cells from adult human peripheral blood in an improved serum-free medium: role of insulin-like growth factor I. Blood 1991;78:

6 29 1 RA and ATRA Enhance Erythroid Differentiation of PBMNC 29 3 Stewart S, Zhu BD, Axelrad AA. A serum-free medium for the production of erythropoietic bursts by murine bone marrow cells. Exp Hematol 1984;12: De Leenheer AP, Lambert WE, Claeys I. All-transretinoic acid: measurement of reference values in 31 human serum by high performance liquid chromatography. J Lipid Res 1982;23: De Ruyter MG, Lambert WE, De Leenheer AP. Retinoic acid: an endogeneous compound of human blood. Unequivocal demonstration of retinoic acid in normal physiological conditions. Anal Biochern 1979;98:42-49.