Ineffective Hemopoiesis in the Myelodysplastic Syndromes (MDS) as Studied by Daily In Situ Observation of Colony-Cluster Formation

Transcription

1 Original Paper International Journal of Cell Cloning (1991) Ineffective Hemopoiesis in the yelodysplastic Syndromes (DS) as Studied by Daily In Situ Observation of Colony-Cluster ormation asami Ohmori, Seiichi Ohmori, Yarunori Ueda, Yataro Yoshida, inoru Okuma irst Division, Department of Internal edicine, aculty of edicine, Kyoto University, Kyoto, Japan Key Words. DS * Ineffective granulopoiesis ing cells In situ * Individual proliferat- Abstract. Daily in situ observation of individual proliferating cells was performed to examine ineffective hemopoiesis in vitro. Bone marrow mononuclear cells (BNC) from 24 myelodysplastic syndrome (DS) patients and 12 controls were cultured for granuloqte-macrophage progenitor (CU-gm) assays using methylcellulose. Individual proliferating cells were mapped at 3 days of culturing and their fates were followed by daily in situ cell counting contained within each cell aggregate until day 8. By retr~spedve analysis of the daily growth of the cells, a significantly greater proportion of noncolony-forming cells in DS were found to proliferate initially, but failed to do so thereafter and degenerated in the culture. Cells showing these abnormal growth characteristics apparently contributed to ineffective granulopoiesis. The present method may be useful for clarifying ineffective granulopoiesis. Introduction The myelodysplastic syndromes (DS) are characterized by varying degrees of peripheral cytopenias despite cellular bone marrow, and by the presence of abnormal clones [I, 21. In vitro marrow cell cultures in DS revealed defective colony formation and increased cluster formation [l-31. These observations are generally assumed to correspond to ineffective hemopoiesis. However, the precise mechanisms leading to these culture abnormalities remain poorly understood. To gain insight into the mechanisms involved in ineffective hemopoiesis in DS, we have developed a new method of serial in situ observation of individual Correspondence: asami Ohmori,.D., irst Division, Department of Internal edicine, aculty of edicine, Kyoto University, Sakyo-ku, Kyoto 66, Japan. Received January 31, 1991; pmisionally accepted ay 3, 1991; accepted for publication June, /91/$2. Alphaed Press

2 Ineffective Hemopoiesis in DS 22 Table I. Patients characteristics Peripheral blood Case ABa Age Sex no. Hb (gldl) PLT (x3/pl) Leukocyte (/pl> b RS RS RS RS EB EB ANLL ANLL ANLL CoL CoL CoL a AB classification according to the rench-american-british criteria [4]. bcases 13 and, as well as Cases 1, 16 and 19 are the same patients studied serially after progression to overt leukemia. proliferating cells. This method involved mapping of individual cell aggregates after 3 days of methylcellulose culture. The subsequent fate of individual proliferating cells was followed up by daily in situ observation until day 8. The results showed that cellular disintegration, eventually leading to a decrease in the number of viable cells within each aggregate, was a conspicuous feature in DS. aterials and ethods Patients Twenty-four consecutive patients with DS were studied. They were diagnosed ac-

3 Ohmori/Ohmori/Ueda/Yoshida/Okuma 23 cording to the rench-american-british (AB) criteria [4] as having refractory anemia (), refractory anemia with ringed sideroblasts (URS), refractory anemia with excess of blasts (EB), EB in transformation (EB-t), and chronic myelomonocytic leukemia (CoL)(Table I). Two cases of EB-t were examined serially after their diseases progressed to acute nonlymphocytic leukemia (ANLL). Control bone marrow was obtained from I2 healthy volunteers. Cell prepamtions Bone marrow was taken after obtaining informed consent. Lightdensity mononuclear cells were separated by icoll-paque (Pharmacia ine Chemicals, Piscataway, NJ) density gradient centrifugation at 4 g for 3 min, washed three times and resuspended in Iscove s modified Dulbecco s medium WD; GIBCO, Grand Island, NY). The cells were further separated into nonadherent and adherent cells after incubation on plastic tissue culture dishes at 37 C for 6 min in an atmosphere of36 C2. Nonadhemt cells were collected by gentle swirling of the dishes and careful aspiration of the floating cells. Granulocyte-acrophage Colony-onning Units (CU-gm) CU-gm cultures were carried out according to the method originally described by Iscove []. Briefly, 1 X lo nonadherent BNC were incubated in 1 ml ID containing.87% methylcellulose, 2% fetal calf serum (CS; kine Scientific, Santa h a, CA) and ng/d recombinant human granulocyte-macrophage colony-stimulating factor (rhg-cs) (Sumitomo Chemicals, Tokyo). This concentration is optimal for forming CU-gm colonies in our laboratory. or mapping of cell proliferation, a 22 mm2 cover glass marked with an 8 mm2 line grid was attached to the bottom of a 3 rnm plastic culture dish (iles Scientific, Naperville, IL) using transparent taping. Individual proliferating cells were located on the grid on day 3 and were examined daily until day 8 by counting the number of viable cells within each cell aggregate under an inverted microscope. Aggregates of more than 4 cells were scored as colonies on day 8, and aggregates of less than 4 cells were scored as clusters on day 8. Statistical analysis was performed using the Wilcoxon test. Cell Viabilify In some experiments, parallel series of cultures were set up with the same conditions as above and were used to examine the viability of cells contained in each cell aggregate. Specifically, cell aggregates at a given day of culture were carefully aspirated using a thin Pasteur pipette and examined for cell viability by the trypan blue dye exclusion method. Results Individual proliferating cells were mapped on day 3 in the whole culture plates from normal and DS bone marrow samples and their subsequent fates were followed by daily in situ cell countings within each cell aggregate. On the basis of daily in situ counting of viable cells, 3 patterns of cell prolikration were arbitrarily defmed: type. A: continuous growth, in which the number of viable cells increased from day 3 through day 8, leading to the formation of a CU-gm colony or a cluster. On day 8, 88% of the proliferating cells of type A formed a colony, while the remaining 12% of the type A cells formed a cluster (aggregates of less than 4 cells). The degenerated cells were <3% of one colony or cluster (ig. L4 and B).

4 Ineffective Hemopoiesis in DS 24 ig. 1. Growth of CU-gm in A) normal marrow and B) marrow from a case of DS (case ). Cells grew continuously and formed a colony. The photomicrographs show the same cell aggregates taken on day 3 through 8.

5 Ohmori/Ohmori/UeddYoshida/Okuma 2 ig. 2. Growth of noncolony-forming cells in DS marrow (Case 12). Cells initially grew but degenerated by day 6 to 8, eventually leading to a reduction in viable cell counts at day 8. The degenerated cells were 23% of per cell aggregates. The photomicrographs show the same cell aggregates taken on day 3 through 8. Type B: the initial growth was followed by disintegration leading to a reduction in viable cell counts. Approximately % of the cells showing type B formed a colony, since aggregates of more than 4 viable cells on day 8 even in the presence of cell degeneration were scored as a colony. The degenerated cells were 23% of one colony (ig. 2). 'Qpe C: the initial growth was followed by cessation of growth without an apparent reduction in viable cell counts. The degenerated cells were c 3% of one colony (ig. 3). The degenerating cells were routinely assessed under an inverted microscope by morphological changes such as cell swelling, irregular disruption of cytoplasmic border and cell lysis (ig. 4A and B). The morphological estimation of cellular degeneration fitted well with the results of trypan blue exclusion. This was also confirmed by the microscopic observation on successive days. Removal of the plates from the incubator for daily observation did not interfere with the growth behavior of cells, since the numbers of day 8 colony and cluster were comparable to those of parallel cultures that had been kept in the incubator. The cells that formed G colonies by day 8 showed type A not only in normal marrow, but also in DS marrow. The cells continued to proliferate in the culture with less than 3% of degeneration. In type B, most of the aggregates contained cells showing signs of disintegration, such as cellular swelling, distortion of the cytoplasmic border and cell

6 Ineffective Hemopoiesis in DS 26 'ig. 3. Growth of noncolony-forming cells in normal maficrw. Cells initially grew but subsequently ceased to grow without an apparent reduction in viable cell counts. The degenerated cells were <3% of per cell aggregates. The photomicrographs show the same cell aggregates taken on day 3 through 8. lysis. Such signs of degeneration were not observed at day 3, when most cells appeared viable, but became evident with time in culture. The degenerative changes yielded a decrease in the number of viable cells contained within single aggregates in DS, but a similar reduction in the number of viable cells was seen in a smaller fraction of noncolony-forming cells in normal marrow. In type C, most of the aggregates contained cells proliferating until days or 6, but failed to do so thereafter. During culture, increasingly more cells ceased to proliferate, retaining the same number of cell aggregates. The majority of these noncolony-forming cells corresponded to cluster-forming cells. Table I1 shows the distribution of the three growth patterns in the DS and control marrow. Unlike the control marrow, most proliferating cells from DS marrow were of pattern B, and the difference in the mean proportion of proliferating cells showing pattern B between the control (median [9% interval], [O-171) and DS marrow (6 [4-9.81) was statistically significant (p <.1). Although most of the noncolony-forming cells exhibited pattern B or C in both the control and DS marrow, there were significantly fewer proliferating cells showing pattern C in DS ( [O-31) than in the control marrow (3 [6.-1, p <.1). Three cases of untreated aplastic anemia were also examined. CU-gm and non-

7

8 Ineffective Hemopoiesis in DS 28 Wle II. Distribution of colony cluster-forming cells in DS and control marrow according to growth patterns DS CU-p % % % (Case no.) Colony IclusteP Ab Bb Cb f f 7 OfOl Of 7f31 16k7 Of1 Of f 31 2 f 1 OfO/ 2f2 2f21 9f9 f 1 7 f 3 1 * f 1 f f 4 Of1 2f2 4f1124f3 1 f f 28 Of1 Of Of1 2f2 OfOl Of 7f21 16f2 Of1 7f Of1 Of Of1 Of 3f2138f1 1f12f3 OfOl 6f2 24 f f 2 ean f SD 3.8 f f edian (W% interval) Controls (n = 12) 69.2 f f 31.2 Number of colonieslclusters per 1 X los BNC on day 8. (ean f SD of triplicate cultures.) These numbers of colonylcluster were counted on day 8 by parallel standard culture used with same patients BNCs. b attern A = continuous growth; pattern B = initial growth followed by disintegration; pattern C = initial growth followed by cessation of growth. ccontrols included 12 healthy volunteers. cultures suggested either a reduction in the number of progenitors or defects in their capacity to proliferate and differentiate to give rise to colonies. The former possibility is unlikely since cellular marrow can hardly be explained solely by a lack of progenitors.

9 Ohmori/hmori/UedalYoshida/Okuma 29 The present study demonstrated that colony-forming cells, albeit small in number in DS, grew in cultures into CU-gm colonies, as did the control CU-gm (ig. IA and B). However, 6 - % of noncolony-forming cells exhibited initial proliferation, followed by degeneration (ig. 2). These degenerative changes resulted in the reduction of the number of viable cells within single cell aggregates. This growth pattern was also found in a small proportion of the control marrow ( [O-171). However, the difference in the proportion of cells showing pattern B was significantly high in DS (p <.1) (Table IT). This indicates that they are the cells largely responsible for ineffective granulopoiesis in DS. Cells showing similarly impaired growth were also seen in the late erythroid progenitor culture in DS (data not shown). It should be added that all aggregates consisting of more than four cells were mapped and counted after three days of culture. Some might well be immature cells located close together that had divided at least once. However, the majority were clones that arose from single progenitors. Although our primary objective was to study the progenitor cell growth in DS versus control marrows, the present method enabled us to examine the growth patterns of the immature proliferating myeloid cells as well. Our observation revealed that type B growth with degeneration and cell death was also characteristically seen for immature myeloid cells in DS, whereas such cells in control marrows showed exclusively C pattern. It thus appears that abortive growth with degeneration is a striking feature not only of progenitor cells but also of immature dividing myeloid cells in DS. There are several possibilities concerning the mechanisms resulting in growth pattern B. irst, DS progenitors have been reported to respond to higher concentrations of G-CS [9]. The use of increased concentrations of G-CS or other sources of CS, such as human placental-conditioned medium, yielded essentially the same results in growth behavior in our hands (data not shown). Thus, the present results could not be attributed to theconcentrations or sources of CS. Second, DS progenitors may have inherent defects in their potential to proliferate. A third possibility is related to the growth inhibition by leukocytes [-l2] or by marrow macrophages in DS [13]. In a separate study, we have shown the presence of a group of marrow macrophages in DS that have inhibitory activity on CU-gm growth [B]. The inhibition seems to be mediated by soluble factors released by macrophages. The identity of these soluble hctors and their relationship to similar substances [-l2] is currently being examined. The precise relationship between the inhibitory macrophages and ineffective hemopoiesis is under investigation. It also remains to be seen whether growth patterns A and B are the property of the clones of the same origin. In conclusion, daily in situ observation of mapped methylcellulose cultures may provide a useful model for the study of ineffective granulopoiesis.

10 Ineffective Hemopoiesis in DS 3 Acknowledgments This work was supported in part by Grants in Aid for Scientific Research (62742, ) from the inistry of Educaton, Grantskr Intractable Diseases from the inistry of Health and Welfare of Japan. We thank Dr. S. Oguma for statistical analysis, s. C. Yoshida and H. Kitagaw for their valuable assistance References Yoshida Y. Biology of myelodysplastic syndrome. Int J Cell Cloning 1987,: Jacobs A. yelodysplastic syndromes: pathogenesis, functional abnormalities, and clinical implications. J Clin Path1 l98;38: Senn JS, Pinkerton PH. Defective in vitro colony formation by human bone marrow preceding overt leukaemia. Br J Haematoi 1972;23: Bennett J, Catovsky D, Daniel T, et al. The rench-american-british PAB) co-operative group. Br J Haematol l982;1: Iscove NN, Senn JS, Till JE, cculloch EA. Colony formation by normal and leukemic human marrow cells in culture: effect of conditioned medium from human leukocyte. B ld 1971;371-. Cazzola, Barosi G, Berzuini C, Dacco, Ester Orlandi, Stefanelli, Ascari E. Quantitative evaluation of erythrocyte activity in dysmyelopoietic syndromes. Br J Haematol 1982;: -62. Ricketts C, Jacobs A, Cavil1 I. errokinetics and erythropoiesis in man: the measurement of effective erythropoiesis, ineffective erythropoiesis and red cell lifespan using 9e. Br J Haematol l97;31:6-7. Yoshida Y, Oguma S, Uchino H, aekawa T. Clinical features and prognosis of re- fractory myelodysplastic anemias: a Japanese cooperative study. Acta Haematol Jpn 198% : Carlo-Stella C, Cazzola, Bergamaschi G, Bernasconi P, Dezza L, Invernizzi R, Pedrazzoli P. Growth of human hematopoietic colonies from patients with myelodysplastic syndromes in response recombinant human granulocyte-macrophage colonystimulating factor. Leukemia l989;3: Bromeyer HE, Bognacki J, Dorner H, aria de Sousa. Identification of leukemiaassociated inhibitory activity as acidic isoferritins. J Exp ed 1981;33: Cukrova V, Neuwirtova R, Cermak J, alaskowva V, Neuwirt J. Leukocytederived inhibitory activity in patients with myelodysplastic syndrome. Blut 1987, Cukrova V, Neuwirtova R, Cermak J, Neuwirt J. Inhibitor of normal granulopoiesis produced by cells of DS patients. Neoplasm 1989;36: Ohmori, Ohmori S, Ueda Y, 'Ibhyama K, Yoshida Y, Uchino H. yelodysplastic syndrome (DS)-associated inhibitory activity on hemopoietic progenitor cells. Br J Haematol 199;74: