Eur. J. Immunol. 1990.20: 1927-1931 IL6 is a differentiation factor for human megakaryocytes in v i m 1927

Hideo Kimura., Toshiyuki Ishibashi., Tatsumi Uchida., Yukio Maruyama., Paul Friese. and Samuel A. Burstein.

The 1st Department of Internal Medicine., Fukushima Medical College, Fukushima and Saint Francis Medical Research Institute., Department of Medicine, University of Oklahoma Health Sciences Center, Oklahoma City

Interleukin 6 is a differentiation factor for human megakaryocytes in vitro*

The response of cells of the megakaryocytic lineage to interleukin 6 (IL6), a cytokine with multiple biological activities, was studied by adding the factor to human bone marrow (BM) cultures. IL6 alone had no effect on megakaryocytic colony formation in methylcellulose; however, in the presence of maximally stimulating concentrations of IL 3, almost a twofold increment in colony formation was observed. Tritiated thymidine suicide studies of BM incubated for 2 h with growth factors showed that almost one-half of megakaryocytic progenitors (CFU-Mk) preincubated with IL3 or IL3 plus IL6 were in S phase, whereas BM incubated with IL 6 alone was similar to control (- 24% of CFU-Mk in S phase). When P 1 ng/ml of IL6 was added to liquid suspension cultures of BM, the size of individual megakaryocytes was significantly augmented com- pared with that seen in control cultures. Moreover, the DNA content of megakaryocytes grown in the presence of IL6 was increased (modal ploidy 16N) compared with cultures grown with IL3 (modal ploidy 8N). To determine if the effect of IL6 could be direct rather than mediated via accessory BM cells, the factor was added to cultures of isolated single megakaryocytes. Seventy-one percent of cells increased in diameter in the presence of IL6, while only 27% increased in size in the absence of the factor. The data show that IL 6 is a direct maturation factor for human megakaryocytes, promoting increments in size and ploidy of these cells.

1 Introduction

IL6 is a 26-kDa protein defined originally as a B cell stimulatory factor, but shown subsequently to support a broad spectrum of biological activities [ l , 21. IL6 is a potent hybridoma growth factor, a T cell differentiation factor and promoter of the hepatic acute-phase reaction [2-61. The factor has also been shown to act synergistically with IL 1 and IL3 in the proliferation of early hematopoiet- ic progenitor cells, to induce differentiation of murine leukemia cells and to inhibit the growth of human breast carcinoma and leukemia cell lines [7-91. In two recent studies, we have shown that IL 6 influences the platelet- megakaryocytic lineage both in vitro and in vivo. In culture, IL6 has the property of stimulating the maturation of murine megakaryocytes as defined by augmentation of cell diameter, acetylcholinesterase (a marker of the megaka- ryocytic lineage in mice; [lo]) activity and cellular DNA content, yet alone was incapable of promoting megakaryo- cyte colony-forming cell (CFU-Mk) proliferation. Howev- er, in combination with the multipoietin IL3 [ll], IL6 demonstrated significant synergy [12]. In vivo, a 5040% increment in the platelet count was observed, accompanied by a marked increase in megakaryocytic size following i.p.

[I 82611

* This work was supported in part by a Grant-in-Aid for Scientific Research (61570586) from the Ministry of Education, Science and Culture of Japan, grant HL29037 from the National Institutes of Health, the Saint Francis Medical Research Insti- tute, and a grant from the Oklahoma Center for the Advance- ment of Science and Technology.

Correspondence: Samuel A. Burstein, Department of Medicine, Room 271, College of Health Building, University of Oklahoma, Oklahoma City, OK 73190, USA

Abbreviations: CFU-Mk: Megakaryocyte colony-forming cell CFU-GM: Granulocyte-macrophage colony-forming cell

administration of the factor to mice [13]. Recent studies have corroborated these findings in mice, both in vitro and in vivo [14, 151. The recent studies of Lotem et al. have suggested that IL6 may play a critical role in murine megakaryocytopoiesis in vitro, since addition of anti-mouse IL6 to BM cultures almost totally abrogated megakaryo- cytic colony formation induced by IL3 [16]. In this report, we have examined the effect of IL 6 on human megakaryo- cytopoiesis in vitro and show that it (a) synergizes with IL3 in megakaryocytic colony formation, but does not affect directly the cell cycle of CFU-Mk, (b) supports the differentiation of human megakaryocytes as assessed by increments in cell size and ploidy, and (c) interacts directly with the megakaryocyte.

2 Materials and methods

2.1 BM preparation

BM was obtained with informed consent from healthy volunteer donors according to the Human Subjects Com- mittee guidelines of the respective institutions.The BM was aspirated directly into 40 U/ml of preservative-free heparin and centrifuged over 1.077 g/cm3 Percoll or Ficoll-Hypaque (Pharmacia Fine Chemicals, Piscataway, NJ) at 400 X g for 20 min at room temperature. The interface cells were washed twice in Iscove’s modification of Dulbecco’s medium (IMDM), and incubated in 10% FCS in IMDM for 2-4 h to partially deplete adherent cells. The nonadherent cells were washed in IMDM and used for subsequent culture.

2.2 Colony assays

CFU-Mk and granulocyte-macrophage colony-forming cell (CFU-GM) were assayed as described previously [ 171. Briefly, 1 x 105 cells were cultured in IMDM supplemented

0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990 OO14-2980/9O/0909- 1927$3.50 + .25/0

1928 H. Kimura, T. Ishibashi, T. Uchida et al. Eur. J. Immunol. 1990. 20: 1927-1931

with 25% normal human plasma, 5 X lop5 M 2-ME, 1% penicillin-streptomycin solution and 0.9% methylcellulose, and stimulated with recombinant IL3 (Genzyme Corp., Boston, MA) in the presence or absence of rIL6 in a 37°C 95% aird% COz tissue culture incubator. Following 13 days in culture, CFU-Mk- and CFU-GM-derived colon- ies were counted in siru. Triplicate plates were enumerated for each data point.

2.3 Assessment of the percentage of CFU-Mk in DNA synthesis

To determine if short-term exposure to IL 6 and IL 3 could induce CFU-Mk into cell cycle, the percentage of CFU-Mk in S phase was measured by [3H]dThd suicide. Nonadher- ent BM cells (1 x 10') were incubated under serum-free conditions in 1 ml alpha medium containing IL3 (100U/ml), IL6 (200ng/ml), both, or medium alone at 37°C for 2 h, followed by washing twice with medium. One-half of the cells were exposed to 20 pCi [3H]dThd (25 Cdmmol; Amersham Int., Amersham, GB) while the other half was exposed to 1.2 pg/ml cold dThd. Following a 20-min incubation at 37 "C, 10 ml of ice-cold alpha medium containing 1% FCS and 100 pg/ml cold dThd was added, and the cells were pelleted and then washed twice in IMDM. Cells (105) were cultured for CFU-Mk and CFU- GM as described above, using 5% PHA-stimulated lym- phocyte conditioned medium to stimulate colony forma- tion [17].

2.4 Megakaryocyte number and size

To determine if IL 6 augments megakaryocytic size, 7.5 x 10" nonadherent BM cells were cultured in suspension in IMDM supplemented with 5% normal human plasma in the presence or absence of IL6.The cultures were set up in a final volume of 200 pl in 96-well flat-bottom culture plates (Cell Wells, Corning Glass Works, Corning, NY). Follow- ing 6 days in culture, the plates were centrifuged at 700 x g for 5 min, and the SN carefully removed. The plates were air dried with a blow drier, followed by fixation in 15% acetone in ethanol for 3 min. Megakaryocytes were stained using a mAb (P4) to the platelet glycoprotein IIb/IIIa complex (provided by Dr. Zaverio Ruggeri, Scripps Clinic and Research Foundation, La Jolla, CA).The specificity of this antibody has been previously characterized and the mAb has been shown to be useful for the detection of megakaryocytes [18]. The plates were incubated with 20 pg/ml of P4 for 2 h, washed extensively and the bound antibody was detected histochemically with an avidin- biotin-glucose oxidase system (Vector Laboratories, Bur- lingame, CA). Cells staining positively for P4 were enumer- ated and their size was measured. This was determined by calculating the geometric mean of two perpendicular diameters measured with an ocular micrometer.

2.5 DNA content

To compare the effect of IL6 with that of the known megakaryocytic growth-promoter IL 3, the DNA content of these cells was measured by FCM. Nonadherent BM cells (1.5 x 105) were cultured as described above in a final volume of 1.5 ml in 24-well culture plates (Falcon 3047; Becton Dickinson, Lincoln Park, NJ). Following 8 days in

culture, the top 1 ml of medium was carefully removed to concentrate the cells, and the underlying cells (in - 0.5 ml) were then aspirated into a 15-ml culture tube.The cells were incubated at 4°C for 30 min with 10 pg/ml of FITC-P4 [19,20]. Subsequently, an equal volume of 0.1% sodium citrate solution containing 40 pg/ml propidium iodide and 0.1% Triton X-100 was added and the cells were incubated at 4°C for 20 min. FCM was performed on a Coulter Epics Vcytometer (Hialeah, FL) using the optical and flow settings described previously for the cytometric evaluation of normal human BM megakaryocytes [19,20]. Small megakaryocytes exhibiting low fluorescence were distin- guished from nonspecifically stained cells and autofluores- cence by setting a gate at a level of fluorescence greater than that of cells stained with a control antibody (fluores- ceinated murine IgGI). The ploidy distribution was deter- mined by setting markers at the nadirs between peaks using the 2N and 4N peaks of the cells as internal standards.

2.6 Single-cell cultures

To determine if the influence of IL6 was direct or mediated via BM accessory cells, the factor was added to isolated single megakaryocytes derived from megakaryocytic colon- ies. BM was enriched for progenitor cells on a 1.060/1.072 g/cm3 discontinuous Percoll gradient, followed by incubation in 10% FCS for 2 h at 37°C to partially deplete adherent cells. Nonadherent cells (7.5 x 104) were cultured with 100 U/ml IL3 in methylcellulose as described above. Following 12 days in culture, a maximum of five individual megakaryocytic colonies were plucked in 2 4 volumes, and dispersed in 2 ml of IMDM [21]. Individual cells were removed in 1-p1 volumes and set up in 35 pl of IMDM supplemented with 2% normal human plasma, 1% penicillin-streptomycin solution and 5 X M 2-ME in the presence or absence of 100 ng/ml IL6. The geometric mean diameters of the cells were determined under an inverted microscope equipped with an ocular micrometer before and after 24 h in culture.

2.7 Growth factors

Human rIL6 produced in E. coli was kindly provided by Ajinomoto Co. (Kawasaki, Japan), or was purchased from R & D Systems (Minneapolis, MN). Human rIL3 was purchased from Genzyme.

3 Results 3.1 IL 6 augments megakaryocytic colony formation

Fig. 1A shows that at concentrations 2 20 ng/ml, IL 6 acts synergistically with IL 3 in the promotion of CFU-Mk- derived colonies. This effect was noted at the both the optimal (maximally stimulating) concentration of IL 3 (100 U/ml), where almost a 2-fold increase in detectable colonies was observed, and at a suboptimal concentration (30 U/ml) of IL3, where almost a 4-fold enhancement of colonies was noted. However, IL6 alone promoted a maximum of two-three megakaryocytic colonies, and usually none (data not shown). Avery slight but statistically significant enhancement of GM colony formation was observed at concentrations of IL6 > 50 n g / d (Fig. 1B).

Eur. J. Immunol. 1990. 20: 1927-1931 IL6 is a differentiation factor for human megakaryocytes in vitro 1929

3.2 IL3, but not IL6 affects the cell cycle status of CFU-Mk and CFU-GM

Pre-incubation of BM cells with IL6 did not alter the percentage of CFU-Mk in Sphase compared to control (24 k 6% vs. 23 k 6%, respectively). Similarly, no differ- ence in the percentage of CFU-GM in S phase was observed (29 k 4% vs. 28 k 5%). Conversely, pre-incubation with IL3 induced a significant increment in the percentage of both CFU-Mk (48 f 13%) and CFU-GM (51 f 13%) in S phase (p < 0.001). The effect of pre-incubation of BM with both IL3 and IL6 was not significantly different than the effect of IL3 alone on either CFU-Mk or CFU-GM (Table 1).

3.3 IL6 promotes an increase in megakaryocytic size, but not number in short-term liquid BM culture

One possible explanation for the observed synergy is that IL 6 increases the number of colonies detectable by render-

250 - 225 -

200 - P 175 - -

c

150 - rr) 125 - 2

100 -

ing individual cells larger. Therefore, size was determined on GPIIb/IIIa+ cells grown for 6 days in liquid culture with different concentrations of IL 6.Table 2 shows the results of two separate experiments using concentrations of IL 6 ranging from 0 to 100ng/ml in cultures grown with 5% human plasma. A significant increment in megakaryocytic diameter was observed in all experiments, and at all tested concentrations of IL6 in comparison with controls ( p < O . O 5 ) . Conversely, the factor at any of the tested concentrations did not augment significantly the number of GPIIb/IIIa+ cells. Similar results were observed in two additional experiments using 2% human plasma (data not shown). At concentrations of plasma below 2%, or in serum-free medium, cell growth was poor in these liquid cultures.

3.4 IL6 increases the ploidy of cultured megakaryocytes

To determine if IL6 increases the DNA content of cultured megakaryocytes, the ploidy of these cells was determined

20 c 1

" 0 20 50 100

IL6 (ng/ml)

200

Figure 1 . IL6 is synergistic with IL3 in the promotion of megakaryocytic colony growth.The 100% level on the ordinate represents the number of colonies promoted by 100 U/ml (optimal) IL3. The error bars represent 1 SD. (A) At concentrations of IL6 h 20 ng/ml, a significant increase in megakaryocytic colony formation was noted in the presence of either 30 (open circles) or 100 Ulml (closed circles) IL3 (five experiments; p<O.001). In the absence of IL3, a negligible number of colonies was observed. The absolute number of CFU-Mk-derived colonies at 100 Ulml of IL3 alone was 14,28,22, 17 and 43, respectively, for each of the five experiments. (B) IL6 (> 50 ng/ml) in the absence of IL3 (open triangles) slightly enhanced the numbers of CFU-GM-derived colonies (five experiments: p < 0.01). No enhancement of colony formation was observed in the presence of either 30 (open circles) or 100 U/ml (closed circles) IL 3. The absolute number of CFU-GM-derived colonies at 100 U/ml of IL3 alone was 75, 62, 65, 72 and 140, respectively, for each of five experiments.

Table 1. Effect of IL6 and IL3 on the percentage of progenitors in S phasea)

a) % Reduction in colonies after exposure to PHIdThd CFU-Mk CF'U-GM

Exp. control IL6 IL3 IL3+ contrOl IL6 IL3 IL3+ IL6 IL 6

1 29 30 60 64 28 30 58 56 2 27 26 51 54 35 29 63 60 3 19 16 50 50 30 31 51 43 4 26 29 51 48 30 33 52 51 5 15 20 26 30 m 22 31 36 Mean 23 24 48 49 29 29 51 50 SD 6 6 13 12 5 4 12 9

BM cells were preincubated for 2 h with 100 Ulml IL3, 200ng/ml IL6, both, or medium alone,washed, split into two equal aliquots and incubated with either [3H]dThd or cold dThd for 20 min. The cells were then cultured in methylcellulose for 13days and CFU-Mk and CFU-GM-derived co- lonies were enumerated. The percent reduction is cal- culated by: l -No. of colon- ies enumerated from

[3H]dThd-exposed BM/No. of colonies enumerated from cold dThd-exposed BM, and represents the fraction of progenitors killed by exposure to [3H]dThd.

1930 H. Kimura, T. Ishibashi, T. Uchida et al. Eur. J. Immunol. 1990. 20: 1927-1931

- 4 0 x v

.

E Z IL3 = IL6

L a, a v)

t a, a,

10 c

%

0 5 - 9

0 -

2N 4N 8N 16N 32N 64N Ploidy distribution

Figure 2. IL6 promotes an increment in megakaryocytic ploidy in comparison with IL3. Following 8 days in liquid culture, cells were labeled with fluoresceinated anti-GPIIb/IIIa and the DNA stained with propidium iodide. The solid bars represent 2300 GPIIb/IIIa+ cells grown with 100 ng/ml IL 6; the hatched bars represent 2426 GPIIb/IIIa+ cells grown with 100 U/ml IL3.The data represent the mean t 1 SEM of nine experiments, with an average of - 250 megakaryocytes analyzed per experiment.The differences between IL6 and IL3 addition is significantly different for cells 2 16N in ploidy (p < 0.05 for 16N, p < 0.02 for 32N and p < 0.01 for 64N; independent t-test).

-

20

I

a, Q

u) l o c C 9)

9)

L 0 5 +t

0

20 I ( B ) - P 1 5 1 L U I

Log red fluorescence 2N 4N 8N 16N 32N 64N

I L 3

Log red fluorescence

2N 4N 8N 16N 32N 64N

Figure 3. FCM of a representative experiment. Megakaryocyte ploidy was measured after 8 days in culture in the presence of IL6 (421 megakaryocytes; panel A) or IL3 (156 megakaryocytes; panel B). The abscissa represents propidium iodide fluorescence. The ploidy distribution is also indicated on the abscissa.The modal ploidy of the IL6-treated culture is 16N, while the mode of the IL3-treated culture is 8N.

by FCM 8 days after initiation of culture. This time was chosen since too few megakaryocytes were analyzable at earlier time points, while adherent cell proliferation inhib- ited megakaryocytic growth at later time points in these liquid cultures. Adherent cell overgrowth appeared to be related to the percentage and the source of plasma used to supplement the cultures, with S 5% plasma providing the most consistent results. Fig. 2 shows that megakaryocytes stimulated with 100 ng/ml IL 6 exhibited a significant increment in DNA content (modal ploidy 16N) compared to cells grown with 100 U/ml IL3 (modal ploidy 8N). Fig. 3 is a representative experiment showing the differences observed between cultures containing IL 6 or IL 3.

Table 2. Influence of IL6 on the number and size of GPIIb/IIIa+ cells in liquid cultures

Exp. 1 Exp. 2 IL 6 Megakaryo- DiameteP) Megakaryo- Diameter (ng/ml) cytedwell (pm * 1 SD) cytedwell (pm f 1 SD)

(+ 1 SD) (k 1 SD)

0 33 (4) 23.4 (6.8) 43 (6) 24.5 (6.5) 1 31 (5 ) 26.8(6.7) 42 (7) 27.9(8.5) 2.5 34 (2) 27.0(8.0) 39 (5 ) 29.1 (7.7) 5 37 (5 ) 27.5 (7.6) 45 (3) 29.4 (8.4)

10 39 (5 ) 27.0(7.4) 45 (3) 30.2(8.1) 20 35 (4) 27.1 (7.2) 48 (7) 30.6 (5.8) 50 40 (6) 28.1 (9.4) 49 (4) 31.4 (8.5)

100 38 (4) 28.2 (8.5) 47 ( 5 ) 31.3 (8.0)

a) The dimaeter represents the geometric mean of two perpendi- cular diameters. At least 50 cells were measured at each point. At all concentrations of IL6 tested, there was no significant difference in the number of megakaryocytes, in both Expts. 1 and 2. Conversely, at all concentrations of added IL6, there was a significant difference in megakaryocyte diameter compared with controls (p < 0.05).

Table 3. Responsiveness of isolated single megakaryocytes to IL 6a)

No. of cells increasing in diametedtotal cells measured (%)

Initial cell diameter - I L 6 + IL6 (vm)

10-20 20-30 Total

201 65 (31) 491 63 (78) 19/ 77(25) 49/ 76 (64) 39/142 (27) 98/139 (71)

a) The size of single megakaryocytes was determined prior to and following 24 h in culture with or without 100 nglml IL6. An increase in size was defined as an increment of 2 0.5 pm. The range of size increments was 0.65-10.82 pm for cells cultured withIL6(mean4.16f2.44), and0.73-5.94 pmforcontrolcells (mean 2.90f 1.91). At each of the initial cell diameter classifications listed, the percentage of cells increasing in size was significantly greater in the IL6-treated group than in controls (p < 0.001; X-square).

Eur. J. Immunol. 1990. 20: 1927-1931 IL6 is a differentiation factor for human megakaryocytes in vitro 1931

3.5 IL6 promotes an increase in the size of single megakaryocytes

Although these data show that IL 6 augments megakaryo- cytic size and DNA content, it is conceivable that these changes are mediated indirectly by other BM cells; this is of particular concern in densely seeded liquid cultures. To determine whether the effect of IL6 was directly on the megakaryocyte, the factor was added to single megakaryo- cytes isolated from CFU-Mk-derived colonies. Table 3 shows that 78% of cells initially 10-20 pm in diameter, and 64% of cells initially 20-30 pm in diameter increased in size when cultured with IL6 compared with 31% and 25% of control cells, respectively.

4 Discussion

The present data show that IL6 has a significant influence on the maturation of human megakaryocytes, and define an additional activity for this multifunctional cytokine. Although IL 6 does not by itself promote megakaryocytic colony formation, the factor synergizes with IL 3, perhaps by increasing the capacity to detect the colonies by augmenting cell size and cytoplasmic maturity (as assessed by increments in GPIIbAIIa). IL 6 does not, by itself trigger CFU-Mk into cell cycle, as assessed by [3H]dThd suicide: experiments employing a 2-h pre-incubation with the growth factor. This is in contrast to the effects observed on primitive progenitor cells from 5-fluorouracil-treated mice , where only the combination of IL 3 and IL 6 triggered these cells into cycle (although the conditions of the experiment were quite different from ours; [22]). However, IL6 may also have additional effects on progenitors that are inde- pendent of augmentation of megakaryocytic maturation, perhaps by influencing the commitment to the megakaryo- cytic lineage from multilineage progenitors, or enhancing the maturational activity of IL 3 [21]. Nonetheless, all of the biological effects of IL 6 on megakaryocytes do not appear to be dependent on the presence of added IL3 or other growth factors, since the addition of IL6 alone promotes the growth of isolated single megakaryocytes. This conclu- sion must be tempered by the fact that the cultures contain small amounts of plasma; therefore, the possibility remains that the effects of added IL 6 are modulated by the presence of growth factors that may be present therein.

Our data corroborate the observations of Bruno and Hoffman that IL6 enhances the growth of human mega- karyocytic colonies when suboptimal concentrations of IL 3 are used [23]. In contrast, those investigators found that IL 6 alone increased the number of CFU-Mk-derived colonies, and that optimal concentrations of both IL6 and IL 3 were no more effective than optimal amounts of IL 3 alone. The reasons for these discrepancies are unclear, but are likely to be related to differences in culture methodol- ogy.

The establishment of IL6 as a maturational factor for human megakaryocytopoiesis in vitro, in conjunction with previous in vitro and in vivo murine studies, suggest that this factor subserves several of the characteristics of a thrombocytopoiesis stimulatory factor or a thrombopoie- tin. However, these studies do not indicate that IL6 is the only such factor. Further in vitro studies will be required to

determine if any of the unpurified conditioned media used as sources of megakaryocytic potentiator or synergistic activity contain IL6. The influence of IL6 on human platelet production is unknown, although there is intriguing suggestive evidence that elevations in IL6 are observed in some inflammatory disorders, conditions that are frequent- ly accompanied by thrombocytosis [24,25].

Received January 23, 1990; in final revised form May 14, 1990.

5 References

1

2

3

4

5

6

7

8

9

10 11

12

13

14

15

Hirano, T. ,Yasukawa, K., Harada, H., Taga, T., Watanabe, Y., Matsuda, T., Kashiwamura, S., Nakajima, K., Koyama, K., Iwamatsu, A., Tsunasawa, S., Sakiyama, F., Matsui, H., Takahara,Y.,Taniguchi,T. and Kishimoto,T., Nature 1986.324: 73. Chiu, C. P., Moulds, C., Coffman, R. L., Rennick, D. and Lee, E, Proc. Natl. Acad. Sci. USA 1988. 85: 7099. VanDamme, J., Opdenakker, G., Simpson, R. J., Rubira, M. R., Cayphas, S.,Vink, A., Billiau, A. and Van Snick, J., J. Exp. Med. 1987. 165: 914. Gauldie, J., Richards, C., Hamish, D., Lansdorp, P. and Baumann, H., Proc. Natl. Acad. Sci. USA 1987. 84: 7251. Takai, Y., Wong, G. G., Clark, S. C. and Raulet, D. H., J. Immunol. 1988. 140: 508. Kishimoto, T. and Hirano, T., Annu. Rev. Immunol. 1988. 6: 485. Ikebuchi, K., Wong, G. G., Clark, S. C., Ihle, J. N., Hirai,Y. and Ogawa, M., Proc. Natl. Acad. Sci. USA 1987. 84: 9035. Chen, L., Mory,Y., Zilberstein, A. and Revel, M., Proc. Natl. Acad. Sci. USA 1988. 85: 8037. Chen, L., Novick, D., Rubinstein, M. and Revel, M., FEBS Lett. 1988. 239: 299. - Jackson, C. W., Blood 1973. 42: 413. Ihle, J. N., Keller, J., Orozlan, S., Henderson, L. E., Copeland, T. D., Fitch, F., Prystowsky, M. B., Goldwasser, E., Schrader, J. W., Palaczymski, E., Dy, M. and Lebel, B., J. Immunol. 1983. 131: 282. Ishibashi, T., Kimura, H., Uchida, T., Kariyone, S., Friese, P. and Burstein, S. A., Proc. Natl. Acad. Sci. USA 1989. 86: 5953. Ishibashi, T., Kimura, H., Shikama, Y., Uchida, T., Kariyone, S., Hirano, T., Kishimoto, T., Takatsuki, F. and Akiyama, Y., Blood 1989. 74: 1241. Warren, M. K., Conroy, L. B. and Rose, J. S., Exp. Hematol. 1989. 17: 1095. Hill, R., Warren, K., Stenberg, P., Levin, J., Corash, L. and Drummond, R., Blood 1989. 74 suppl. I : 207a.

16 Lotem, J., Shabo,Y. and Sachs, L., Blood 1989. 74: 1545. 17 Kimura, H., Burstein, S. A., Thoming, D., Powell, J. S.,

Harker, L. A., Fialkow, P. J. and Adamson, J. W., J. Cell. Physiol. 1984. 118: 87.

18 Ishibashi,T., Ruggeri, Z. M., Harker, L. A. and Burstein, S.A., Blood 1986. 67: 1286.

19 Tomer, A., Harker, L. A. and Burstein, S. A., Blood 1987. 70: 1735.

20 Tomer, A., Harker, L. A. and Burstein, S. A., Blood 1988. 71: 1244.

21 Ishibashi, T. and Burstein, S. A., Blood 1986. 67: 1512. 22 Ikebuchi, K., Ihle, J. N., Hirai,Y., Wong, G. G., Clark, S. C.

and Ogawa, M., Blood 1988. 72: 3007. 23 Bruno, E. and Hoffman, R., Exp. Hematol. 1989. 17: 1038. 24 Hirano,T., Matsuda,T.,nmer, M., Miyasaka, N., Buchan, G.,

Tang, B., Sato, K., Shimizu, M., Maini, R., Feldmann, M. and Kishimoto,T., Eur. J. Immunol. 1989. 18: 1797.

25 Marchasin, S., Wallerstein, R. 0. and Aggeler, P. M., Calif. Med. 1964. 101: 95.