Original Paper International Journal of Cell Cloning 10:28-32 (1992)
In Vitro Megakaryocyte Colony Formation in Patients with Idiopathic Thrombocytopenic Purpura: Differences between Acute and Chronic ITP J . F. Abgrall, N . El-Kassar, C . Berthou, 1. Renard, J . M. Cauvin,A. Le Pailleur, C . Autrand, L. Sensebe, G. Guern, P. Zilliken, G. Le Gall, J . Briere Department of Hematology, Hbpital A. Morvan, Brest University, School of Medicine, Brest, France Key Words. Idiopathic thrombocytopenic purpura Bone-marrow CFU-Meg Abstract. In vitro megakaryocyte colony formation from the bone marrow of patients with acute idiopathic thrombocytopenic purpura (ITP) or chronic ITP was compared using a plasma clot system. The number of megakaryocyte colony-forming units (CFU-Meg) was significantly higher (p c 0.05) in acute ITP compared to chronic ITP (54.3 f 68.4 vs. 12.9 f 15.3/105 nonadherent mononuclear cells, mean k SD), and significantly lower (p c 0.05) in chronic ITP compared to controls (12.9 ? 15.3 vs. 22.8 15.9). A significant correlation was observed between platelet recovery 7 and 30 days after culture, and the number of CFU-Meg ( r = 0.49 and 0.45, respectively, p c 0.05). An inverse correlation was observed between platelet count at the time of culture and the number of Megs per colony (r = -0.48, p c 0.05). These results indicated a difference between acute and chronic ITP in the ability to promote in vitro Meg colony formation and may suppose a different immune mechanism for thrombocytopenia in these two disorders.
+
Introduction Idiopathic thrombocytopenic purpura (1TP) is a disorder characterized by immunologic destruction of platelets of unknown origin, and sometimes associated with viral infection [l]. Two clinical entities are recognized, acute and chronic ITP, with different therapeutic management [2]. In vitro megakaryocyte colony formation has been studied in animal models [3] and in patients [4, 5 , 61. In mice rendered thrombocytopenic by antiplatelet antibodies, Burstein Correspondence: Dr. J.F. Abgrall, HBpital A. Morvan, 29285 Brest Cedex, France. Received June 19, 1991; provisionally accepted July 24, 1991; accepted for publication August 29, 1991. 0731-1454/92~$2.00/00AlphaMed Press
[3] described three events: 1 ) an increase in cell cycle followed by 2) an increase in the number and volume of megakaryocytes (Megs), and 3) 24 - 48 h later an increase in the number of Meg colony units (CFUMeg). Segal and Ballem [4] showed in humans an increase in the percentage of Megs in S phase, inversely correlated to the number of platelets, and a normal number of CFU-Meg. They concluded that antibodies in ITP decreased the number of platelets by acting at the terminal phase of megakaryocytopoiesis. More recently, De Alarcon studied nine children with acute ITP and described an increase in the number of CFU-Meg and an increase in the size of Meg colonies [5]. Sugiyama studied 18 patients with chronic ITP and observed a significant increase in the number of CFU-Meg, which inversely correlated to the number of platelets [6]. In the present study, patients with acute and chronic ITP were evaluated for their ability to promote in vitro Meg colony formation.
Materials and Methods Patients Twenty-one adult patients with acute or chronic ITP were studied. The mean age of the 12 females and 9 males was 49.4 f 24.2 years (range 21 - 80). Mean age was not different between acute and chronic ITP (40.9 k 21.5 vs. 56.1 f 22.4 years). The diagnosis of ITP was done as follows: platelet count under 1 X 10"fl; normal or slightly decreased cellularity on bone marrow aspiration with normal or increased number of Megs; absence of splenomegaly; absence of antinuclear antibody; absence of any etiological factor such as viral infection or drug; and absence of
Abgrall et al.
29
Table I. Clinical and biological data of 21 patients with ITP Platelet count
X
109Amarrow Megs
Patient
Age
Sex
A/C
Day0
Day7
Day30
LES HAM LAR JEA RUE DOH LAO LEV THO LEF PER AND MAR PLE HER SEH MAD GUI LHO LEJ LAN
61 76 70 71 74 23 25 21 78 70 30 71 60 24 49 30 65 80 23 26 30
M F M F F F M F F F F F M F M F F F F M M
C C C C C C C C C C C C C A A A A A A A A
3 11 29 29 30 35 55 57 58 62 78 88 97 2 3 3 5 6 8 16 20
86 87 29 50 200 32 65 70 58 60 80 90 102 113 100 419 90 51 8 840 20
155 60 35 94 43 180 50 I0 60 70 80 70 90 200 350 315 33 1 300 8 350 45
Cellularity
+++
++ +++ ++ +++
+++ +++ +++ ++ +++ +++
++ +++ ++ +++ ++ +++ ++ +++ ++
++
Megs
I I I I I I N 1 N
PAIg
+ ND
+
ND ND
+
ND ND ND
1
+
I N I I I N
ND ND
I N I N N
ND
+ + ND
+
Age is indicated in years. A/C means acute or chronic ITP. Platelet counts were done at days 0, 7 and 30. Marrow cellularity is indicated by +++ as normal and ++ as slightly decreased. The numbers of megakaryocytes on bone marrow smears are normal (N) or increased (I). PAIg represent platelet-associated Ig and are expressed as positive (+), negative (-), or not done (ND).
underlying disease. Platelet-associated immunoglobulins (PAIg) were measured in 11 patients and were increased in 7. No patient was under steroid therapy or other treatment for thrombocytopenia at the time of culture. Bone marrow megakaryocyte progenitors (CFU-Meg) were studied at the time of diagnosis in all patients except in 4 patients in which bone marrow was plated 2 - 3 weeks after diagnosis. Acute ITP and chronic ITP were defined 3 mos. after culture by persistence of thrombocytopenia. Clinical and biological data of patients are summarized in Table I. Eight patients had acute ITP and 13 patients had chronic ITP. All patients with acute ITP and 7 patients with chronic ITP received steroid therapy after Meg culture was done. Steroid therapy regimen was the same in all patients, i.e., one course of prednisolone (1 m a g body weight), during 3 weeks, followed by a 1 wk decreasing dosage until complete cessation. Platelet counts also were done 7 days and 30 days after culture enabling platelet recovery at day 7 and
at day 30 to be calculated (platelet count at day 7 or 30 minus platelet count at day 0 divided by platelet count at day 0). Bone marrows were obtained by aspiration after receiving informed consent of the patient. One part was used for diagnosis and the other for in vitro culture. Bone marrow specimens for culture were collected in l ml alpha medium (Eurobio, Paris, France) containing 0.1 ml heparin (20 U/ml; Roche, Neuilly sur Seine, France). Megakaryocyte culture Progenitor cells were assayed in a plasma clot culture system as previously described [7]. Briefly, 2 X lo5nonadherent mononuclear cells (NAMC) were plated in duplicate or triplicate in 1.05 ml alpha medium containing 10% citrated bovine plasma (Flow Laboratories, Irvine, Scotland), 1% deionized bovine serum albumin (Sigma Chemical Co., St. Louis, MO), 5% human peripheral blood leukocyte-conditioned medium stimulated by phytohemagglutinin (PHA-
Meg Colony Formation and Idiopathic Thrombocytopenic Purpura
30
Table 11. Platelet count, CFU-Meg number and platelet recovery (PR) in acute and chronic ITP
N
Acute ITP 8
Chronic ITP 13
Platelets CFU-Meg Meg/Colon y PR at day 7 PR at day 30
8.5 f 6.4a 54.3 f 68.4b 9.1 f 2.7 26.9 f 23.S8 39.7 f 3 1.4"
48.6 f 28.7 12.9 f 15.3b 5.9 f 3.4 4.1 f 8.1 4.8 f 13.9
Controls 15
22.8 f 15.9 8 . 2 f 3.5
Platelet count is given 1 X 109/I.CFU-Meg are given for 1 X los NAMC. PR means platelet recovery. Mann-Whitney test shows a significant difference in CFU-Meg between chronic and acute ITP (b= p < 0.05) and between chronic ITP and controls (b = p < 0.05); and also shows a significant difference in platelet recovery at day 7 and day 30 ("= p < 0.01) between acute and chronic ITP. A significant difference was also observed in initial platelet count between acute and chronic ITP p < 0.01). No significant difference was present in the numbers of Megs per colony.
LCM; Wellcome, Paris, France), M 2mercaptoethanol (Sigma), 0.34 mg/ml CaCl,, 10% human AB serum. Cultures were incubated for 12 days in a humidified incubator at 37°C with 5% CO, and air. Meg colonies were identified directly in culture dishes by an indirect immunoperoxidase staining with a monoclonal antibody against the platelet glycoprotein IIb/IIIa complex, HuP1-ml (AMD, Artarmon, Australia). A CFU-Meg-derived colony was defined as a cluster of three or more Megs. The number of Megs per colony was counted in each dish. Granulomonocytic colonies (CFU-GM) were identified in the same dishes as clusters of at least 40 cells and scored under microscopy. Results were compared to 15 control bone marrows obtained after informed consent from hematologically normal subjects undergoing surgery for benign lesions. Results of acute ITP were compared to those of chronic ITP. The Mann-Whitney Rank-sum test and KruskalWallis one-way analysis of variance were used for comparison of means. Kendall and Spearman rank tests were used for analysis of correlation. Chi square, Yates-corrected chi square tests and the Fisher test were used for frequence tables analysis.
Results Platelet count was significantly lower (p < 0.01) in acute ITP compared to chronic ITP (8.5 f 6.4 vs. 48.6 f 28.7 X lo9 plateletsfl) (Table 11). Platelet recovery at day 7 was significantly higher (p < 0.05) in acute ITP compared to chronic ITP (26.9 f 23.8 times the initial platelet count vs. 4.1 f 8.1 times) (Table
11). Platelet count at day 30 was significantly higher (p < 0.01) in acute ITP compared to chronic ITP (217.2 f 123.2 vs. 82.6 f 43 plateletsX109/l), and platelet recovery at day 30 was significantly higher (p < 0.01) in acute ITP compared to chronic ITP (39.7 f 31.4 vs. 4.7 f 13.8). When considering patients with acute or chronic ITP treated or not by steroids, platelet count at the time of culture was significantly lower (p < 0.01) in patients receiving steroids (17.5 f 15.7 vs. 73 f 17.3 X 109fl).Platelet recovery at day 7 was significantly higher (p < 0.01) in treated patients (17.9 f 20.6 vs. 0.86 f 2.1). Platelet recovery at day 30 also was significantly higher (p < 0.01) in treated patients (25.3 f 29.9 vs. 0.86 f 2.1) (data not shown). The number of CFU-Meg was not significantly higher in treated patients (34.1 f 53.8 vs. 15.2 k 20.2), but the number of Megs per colony was significantly higher (p < 0.05) in treated patients (8 f 3.4 vs. 4.9 f 2.8). The number of CFU-Meg was significantly higher (p < 0.05) in acute ITP treated by steroids compared to chronic ITP also treated by steroids (54.3 f 68.4 vs. 1 1 k 10.8 CFU-Meg/lOS NAMC). The number of CFU-Meg was within the normal range in 4 patients, increased in 5 patients and decreased in l l patients (Table III). The numbers of CFU-Meg were significantly different (p < 0.01) in acute ITP, chronic ITP and controls. The mean number of CFU-Meg was significantly higher (p < 0.05) in acute ITP compared to chronic ITP (54.3 k 68.4 vs. 12.9 f 15.3 CFW-Meg/lO' NAMC). The mean number of CFU-Meg was significantly lower (j< 0.05) in chronic ITP compared to controls (12.9 f 15.3 vs. 22.8 f 15.9).The mean number of CFU-Meg in acute ITP was not significantly different from controls (Table 11). These differences did not seem to be
Abgrall et al.
31
Table 111. Numbers of in vitro CFU-Meg and CFU-GM in 2 1 patients with ITP
Patients LES HAM LAR JEA RUE DOH LAO
AIC
CFU-Meg
Meg/colony
C C C
32 13.2 9.6 3
7.1
C C
C C C C C C C C A A A A A A A A
LEV THO
LEF PER AND
MAR PLE HER
SEH MAD GUI LHO LEI
LAN Median
CFU-GM 46 4 ND
9
9.8 11.3 4.3
15 0
19 7.5 28 49 42.5
0
4
1 4
5.8 5.9 5
8 5
52
15
53 21 62
7.4 0
0
25.3 25
7.3 6.6
44
10
13.3 57.5
14.1 6
6
23 218 48
10.8
43 ND 34.6 10.4
8
20
11.2 9.6 1.6
37 30
80 30
14
~~
~
Controls
n = 15
22.8 f 15.9
8.2 f 3.5
40 f 25
~~
~
The numbers of CFU-Meg and CFU-GM are given as the mean of 2 - 3 experiments and are expressed for 1 A/C means acute/chronic ITP. Megs per colony are given as the mean number of Megs in each colony.
caused by any technical problem in the culture, since there was no observed relation between the number of CFU-GM and the number of CFU-Meg. Nine patients had a number of CFU-Meg equal to or below median (median of CFU-Meg = 14), five of which had a number of CFU-GM below median (median of CFU-GM = 30), and four had a number of CFU-GM above median. Nine patients had a number of CFUMeg above median, five of which had a number of CFU-GM equal to or below median, and four had a number above median. The number of Megs per colony were not significantly different in acute ITP, chronic ITP and controls (Table 11). When all data were taken together (acute and chronic ITP) and related to platelet count, no significant correlation was observed between initial platelet count and numbers of CFU-Meg ( r = -0.42).However, a significant inverse correlation was observed between the number of Megs per colony and platelet count at day 0 (Spearman; r = -0.48;p c 0.05) (data not shown). A significant correlation also was observed between the
X
~~
lo5NAMC.
number of CFU-Meg and platelet recovery at day 7 (Spearman; r = 0.49;p < 0.05) and between the number of CFU-Meg and platelet recovery at day 30 (Spearman; r = 0.45; p < 0.05) (Table 11). A significant relation also was observed between the number of Megs per colony and platelet recovery at day 7 (Spearman; r = 0.51; p c 0.05). No difference was observed in the number of CFU-Meg between patients receiving steroids or not receiving steroids.
Discussion Twenty-one adult patients with ITP were studied for their ability to promote in vitro bone marrow Meg colony formation in order to characterize differences between acute and chronic ITP. Patients with acute ITP had a significantly higher number of CFU-Meg compared to chronic ITP; patients with chronic ITP had a significantly lower number of CFU-Meg compared to controls. These findings are in accordance
32
Meg Colony Formation and Idiopathic Thrombocytopenic Purpura
with De Alarcon [ 5 ] who also observed an increased number of CFU-Meg in children with acute ITP, but are different from the results of Sugiyama [6] showing an increased number of CFU-Meg in chronic ITP. Segal and Ballem [4] did not observe any difference from controls in CFU-Meg numbers in their patients. No significant difference was present in the number of Megs per colony between chronic ITP, acute ITP, and controls. Sugiyama observed a significant inverse correlation between platelet count and CFU-Meg numbers. This finding was not observed in the present study. When examining platelet recovery at day 7 and day 30 after culture in acute and chronic ITP, a significant relation was observed between the number of CFU-Meg and platelet recovery at day 7 and day 30. A significant inverse correlation was observed between platelet count at day 0 and the number of Megs per colony. When platelet count was low, the number of Megs per colony was high. The present study shows a difference between acute and chronic ITP in the ability to promote in vitro bone marrow Meg colony formation, Meg colony formation being normal in acute ITP and impaired in chronic ITP. The explanation for this impaired colony formation is unknown, but we can suggest some hypotheses: 1) the immune mechanism of thrombocytopenia could be different in acute ITP compared to chronic ITP, as suggested by McMillan who demonstrated that in chronic ITP, antiplatelet auto-antibody also binds to megakaryocytes [8]. In the present study, we have not tested the binding of patients’ auto-antibodies either on platelets or on megakaryocytes, but we could hypothesize that in acute ITP the antibody could be directed mainly against platelet antigens, and in chronic ITP the antibody could act at the level of Megs. 2 ) Another immune mechanism is suggested by Trenf [9] in chronic ITP who observed an abnormal T suppressor function. In the present work, T suppressor function was not tested, and T cell depletion was not done on bone marrow before plating. Therefore, we are unable to give any suggestion for a possible role of T suppressor cells on Meg colony formation in our patients. 3) Another explanation could be that Meg colony formation could be the result of different feedback mechanisms in response to thrombocytopenia in acute ITP and chronic ITP. Additionally, an important finding in this study was the correlation between culture data and platelet
recovery, i.e., when the numbers of CFU-Meg were high, platelet recovery at day 7 and day 30 was also high. The use of in vitro Meg colony culture in ITP thus could be valuable in clinical practice as predictive of platelet recovery. From this study we were unable to determine the value of Meg culture in predicting the efficiency of steroid therapy. Determination of the specificity of the antibodies present in these patients is under investigation currently in order to know if Meg culture patterns are related to any peculiar platelet antibody specificity.
Acknowledgment We thank f.McLaughlin for technical assistance in translation.
References Karpatkin S. Autoimmune thrombocytopenic purpura. Blood 198056~329-343. McMillan R. Chronic idiopathic thrombocytopenic purpura. N Eng J Med 1981;304:1135-1147. Burstein S, Adamson JW, Erb SK, Harker L. Megakaryocytopoiesis in the mouse: response to varying platelet demand. J Cell Physiol 1981;109:333-341. Segal GM, Ballern P, Gemsheimer T, Slichter SJ, Adamson JW.Autoimmune thrombocytopenia: progenitor cell response to platelet demand. In: Levine, RF, et al., eds. Megakaryocyte Development and Function. New York: Alan R. Liss, Inc., 1986;341-345. De Alarcon PA, Mazur EM, Schmieder JA. In vitro megakaryocytopoiesis in children with acute idiopathic thrombocytopenic purpura. Am J Ped Hematol Oncol 1987;9:212-218. Sugiyama H, Yagita M, Takahashi T, Nakamura K, Iho S, Hoshino T, Irnura H. Megakaryocytopoiesis in idiopathic thrombocytopenic purpura. Acta Haernatol Jap 198730:119-128. Han ZC, Briere J, Abgrall JF, Sensebe L, Parent D, Guem G. Characteristics of megakaryocyte colony formation i n normal individuals and in primary thrombocythemia: studies using an optimal cloning system. Exp Hematol 1989;17:46-52. McMillan R . Immune thrombocytopenia. Clin Haematol 1983; 12:69-88. Trent RJ, Clancy RL, Dank V, Basten A. Disordered immune homeostasis in chronic idiopathic thrornbocytopenic purpura. Clin Exp Irnmunol 1981;45:9- 17.
In Vitro Megakaryocyte Colony Formation in Patients with Idiopathic Thrombocytopenic Purpura: Differences between Acute and Chronic ITP J . F. Abgrall, N . El-Kassar, C . Berthou, 1. Renard, J . M. Cauvin,A. Le Pailleur, C . Autrand, L. Sensebe, G. Guern, P. Zilliken, G. Le Gall, J . Briere Department of Hematology, Hbpital A. Morvan, Brest University, School of Medicine, Brest, France Key Words. Idiopathic thrombocytopenic purpura Bone-marrow CFU-Meg Abstract. In vitro megakaryocyte colony formation from the bone marrow of patients with acute idiopathic thrombocytopenic purpura (ITP) or chronic ITP was compared using a plasma clot system. The number of megakaryocyte colony-forming units (CFU-Meg) was significantly higher (p c 0.05) in acute ITP compared to chronic ITP (54.3 f 68.4 vs. 12.9 f 15.3/105 nonadherent mononuclear cells, mean k SD), and significantly lower (p c 0.05) in chronic ITP compared to controls (12.9 ? 15.3 vs. 22.8 15.9). A significant correlation was observed between platelet recovery 7 and 30 days after culture, and the number of CFU-Meg ( r = 0.49 and 0.45, respectively, p c 0.05). An inverse correlation was observed between platelet count at the time of culture and the number of Megs per colony (r = -0.48, p c 0.05). These results indicated a difference between acute and chronic ITP in the ability to promote in vitro Meg colony formation and may suppose a different immune mechanism for thrombocytopenia in these two disorders.
+
Introduction Idiopathic thrombocytopenic purpura (1TP) is a disorder characterized by immunologic destruction of platelets of unknown origin, and sometimes associated with viral infection [l]. Two clinical entities are recognized, acute and chronic ITP, with different therapeutic management [2]. In vitro megakaryocyte colony formation has been studied in animal models [3] and in patients [4, 5 , 61. In mice rendered thrombocytopenic by antiplatelet antibodies, Burstein Correspondence: Dr. J.F. Abgrall, HBpital A. Morvan, 29285 Brest Cedex, France. Received June 19, 1991; provisionally accepted July 24, 1991; accepted for publication August 29, 1991. 0731-1454/92~$2.00/00AlphaMed Press
[3] described three events: 1 ) an increase in cell cycle followed by 2) an increase in the number and volume of megakaryocytes (Megs), and 3) 24 - 48 h later an increase in the number of Meg colony units (CFUMeg). Segal and Ballem [4] showed in humans an increase in the percentage of Megs in S phase, inversely correlated to the number of platelets, and a normal number of CFU-Meg. They concluded that antibodies in ITP decreased the number of platelets by acting at the terminal phase of megakaryocytopoiesis. More recently, De Alarcon studied nine children with acute ITP and described an increase in the number of CFU-Meg and an increase in the size of Meg colonies [5]. Sugiyama studied 18 patients with chronic ITP and observed a significant increase in the number of CFU-Meg, which inversely correlated to the number of platelets [6]. In the present study, patients with acute and chronic ITP were evaluated for their ability to promote in vitro Meg colony formation.
Materials and Methods Patients Twenty-one adult patients with acute or chronic ITP were studied. The mean age of the 12 females and 9 males was 49.4 f 24.2 years (range 21 - 80). Mean age was not different between acute and chronic ITP (40.9 k 21.5 vs. 56.1 f 22.4 years). The diagnosis of ITP was done as follows: platelet count under 1 X 10"fl; normal or slightly decreased cellularity on bone marrow aspiration with normal or increased number of Megs; absence of splenomegaly; absence of antinuclear antibody; absence of any etiological factor such as viral infection or drug; and absence of
Abgrall et al.
29
Table I. Clinical and biological data of 21 patients with ITP Platelet count
X
109Amarrow Megs
Patient
Age
Sex
A/C
Day0
Day7
Day30
LES HAM LAR JEA RUE DOH LAO LEV THO LEF PER AND MAR PLE HER SEH MAD GUI LHO LEJ LAN
61 76 70 71 74 23 25 21 78 70 30 71 60 24 49 30 65 80 23 26 30
M F M F F F M F F F F F M F M F F F F M M
C C C C C C C C C C C C C A A A A A A A A
3 11 29 29 30 35 55 57 58 62 78 88 97 2 3 3 5 6 8 16 20
86 87 29 50 200 32 65 70 58 60 80 90 102 113 100 419 90 51 8 840 20
155 60 35 94 43 180 50 I0 60 70 80 70 90 200 350 315 33 1 300 8 350 45
Cellularity
+++
++ +++ ++ +++
+++ +++ +++ ++ +++ +++
++ +++ ++ +++ ++ +++ ++ +++ ++
++
Megs
I I I I I I N 1 N
PAIg
+ ND
+
ND ND
+
ND ND ND
1
+
I N I I I N
ND ND
I N I N N
ND
+ + ND
+
Age is indicated in years. A/C means acute or chronic ITP. Platelet counts were done at days 0, 7 and 30. Marrow cellularity is indicated by +++ as normal and ++ as slightly decreased. The numbers of megakaryocytes on bone marrow smears are normal (N) or increased (I). PAIg represent platelet-associated Ig and are expressed as positive (+), negative (-), or not done (ND).
underlying disease. Platelet-associated immunoglobulins (PAIg) were measured in 11 patients and were increased in 7. No patient was under steroid therapy or other treatment for thrombocytopenia at the time of culture. Bone marrow megakaryocyte progenitors (CFU-Meg) were studied at the time of diagnosis in all patients except in 4 patients in which bone marrow was plated 2 - 3 weeks after diagnosis. Acute ITP and chronic ITP were defined 3 mos. after culture by persistence of thrombocytopenia. Clinical and biological data of patients are summarized in Table I. Eight patients had acute ITP and 13 patients had chronic ITP. All patients with acute ITP and 7 patients with chronic ITP received steroid therapy after Meg culture was done. Steroid therapy regimen was the same in all patients, i.e., one course of prednisolone (1 m a g body weight), during 3 weeks, followed by a 1 wk decreasing dosage until complete cessation. Platelet counts also were done 7 days and 30 days after culture enabling platelet recovery at day 7 and
at day 30 to be calculated (platelet count at day 7 or 30 minus platelet count at day 0 divided by platelet count at day 0). Bone marrows were obtained by aspiration after receiving informed consent of the patient. One part was used for diagnosis and the other for in vitro culture. Bone marrow specimens for culture were collected in l ml alpha medium (Eurobio, Paris, France) containing 0.1 ml heparin (20 U/ml; Roche, Neuilly sur Seine, France). Megakaryocyte culture Progenitor cells were assayed in a plasma clot culture system as previously described [7]. Briefly, 2 X lo5nonadherent mononuclear cells (NAMC) were plated in duplicate or triplicate in 1.05 ml alpha medium containing 10% citrated bovine plasma (Flow Laboratories, Irvine, Scotland), 1% deionized bovine serum albumin (Sigma Chemical Co., St. Louis, MO), 5% human peripheral blood leukocyte-conditioned medium stimulated by phytohemagglutinin (PHA-
Meg Colony Formation and Idiopathic Thrombocytopenic Purpura
30
Table 11. Platelet count, CFU-Meg number and platelet recovery (PR) in acute and chronic ITP
N
Acute ITP 8
Chronic ITP 13
Platelets CFU-Meg Meg/Colon y PR at day 7 PR at day 30
8.5 f 6.4a 54.3 f 68.4b 9.1 f 2.7 26.9 f 23.S8 39.7 f 3 1.4"
48.6 f 28.7 12.9 f 15.3b 5.9 f 3.4 4.1 f 8.1 4.8 f 13.9
Controls 15
22.8 f 15.9 8 . 2 f 3.5
Platelet count is given 1 X 109/I.CFU-Meg are given for 1 X los NAMC. PR means platelet recovery. Mann-Whitney test shows a significant difference in CFU-Meg between chronic and acute ITP (b= p < 0.05) and between chronic ITP and controls (b = p < 0.05); and also shows a significant difference in platelet recovery at day 7 and day 30 ("= p < 0.01) between acute and chronic ITP. A significant difference was also observed in initial platelet count between acute and chronic ITP p < 0.01). No significant difference was present in the numbers of Megs per colony.
LCM; Wellcome, Paris, France), M 2mercaptoethanol (Sigma), 0.34 mg/ml CaCl,, 10% human AB serum. Cultures were incubated for 12 days in a humidified incubator at 37°C with 5% CO, and air. Meg colonies were identified directly in culture dishes by an indirect immunoperoxidase staining with a monoclonal antibody against the platelet glycoprotein IIb/IIIa complex, HuP1-ml (AMD, Artarmon, Australia). A CFU-Meg-derived colony was defined as a cluster of three or more Megs. The number of Megs per colony was counted in each dish. Granulomonocytic colonies (CFU-GM) were identified in the same dishes as clusters of at least 40 cells and scored under microscopy. Results were compared to 15 control bone marrows obtained after informed consent from hematologically normal subjects undergoing surgery for benign lesions. Results of acute ITP were compared to those of chronic ITP. The Mann-Whitney Rank-sum test and KruskalWallis one-way analysis of variance were used for comparison of means. Kendall and Spearman rank tests were used for analysis of correlation. Chi square, Yates-corrected chi square tests and the Fisher test were used for frequence tables analysis.
Results Platelet count was significantly lower (p < 0.01) in acute ITP compared to chronic ITP (8.5 f 6.4 vs. 48.6 f 28.7 X lo9 plateletsfl) (Table 11). Platelet recovery at day 7 was significantly higher (p < 0.05) in acute ITP compared to chronic ITP (26.9 f 23.8 times the initial platelet count vs. 4.1 f 8.1 times) (Table
11). Platelet count at day 30 was significantly higher (p < 0.01) in acute ITP compared to chronic ITP (217.2 f 123.2 vs. 82.6 f 43 plateletsX109/l), and platelet recovery at day 30 was significantly higher (p < 0.01) in acute ITP compared to chronic ITP (39.7 f 31.4 vs. 4.7 f 13.8). When considering patients with acute or chronic ITP treated or not by steroids, platelet count at the time of culture was significantly lower (p < 0.01) in patients receiving steroids (17.5 f 15.7 vs. 73 f 17.3 X 109fl).Platelet recovery at day 7 was significantly higher (p < 0.01) in treated patients (17.9 f 20.6 vs. 0.86 f 2.1). Platelet recovery at day 30 also was significantly higher (p < 0.01) in treated patients (25.3 f 29.9 vs. 0.86 f 2.1) (data not shown). The number of CFU-Meg was not significantly higher in treated patients (34.1 f 53.8 vs. 15.2 k 20.2), but the number of Megs per colony was significantly higher (p < 0.05) in treated patients (8 f 3.4 vs. 4.9 f 2.8). The number of CFU-Meg was significantly higher (p < 0.05) in acute ITP treated by steroids compared to chronic ITP also treated by steroids (54.3 f 68.4 vs. 1 1 k 10.8 CFU-Meg/lOS NAMC). The number of CFU-Meg was within the normal range in 4 patients, increased in 5 patients and decreased in l l patients (Table III). The numbers of CFU-Meg were significantly different (p < 0.01) in acute ITP, chronic ITP and controls. The mean number of CFU-Meg was significantly higher (p < 0.05) in acute ITP compared to chronic ITP (54.3 k 68.4 vs. 12.9 f 15.3 CFW-Meg/lO' NAMC). The mean number of CFU-Meg was significantly lower (j< 0.05) in chronic ITP compared to controls (12.9 f 15.3 vs. 22.8 f 15.9).The mean number of CFU-Meg in acute ITP was not significantly different from controls (Table 11). These differences did not seem to be
Abgrall et al.
31
Table 111. Numbers of in vitro CFU-Meg and CFU-GM in 2 1 patients with ITP
Patients LES HAM LAR JEA RUE DOH LAO
AIC
CFU-Meg
Meg/colony
C C C
32 13.2 9.6 3
7.1
C C
C C C C C C C C A A A A A A A A
LEV THO
LEF PER AND
MAR PLE HER
SEH MAD GUI LHO LEI
LAN Median
CFU-GM 46 4 ND
9
9.8 11.3 4.3
15 0
19 7.5 28 49 42.5
0
4
1 4
5.8 5.9 5
8 5
52
15
53 21 62
7.4 0
0
25.3 25
7.3 6.6
44
10
13.3 57.5
14.1 6
6
23 218 48
10.8
43 ND 34.6 10.4
8
20
11.2 9.6 1.6
37 30
80 30
14
~~
~
Controls
n = 15
22.8 f 15.9
8.2 f 3.5
40 f 25
~~
~
The numbers of CFU-Meg and CFU-GM are given as the mean of 2 - 3 experiments and are expressed for 1 A/C means acute/chronic ITP. Megs per colony are given as the mean number of Megs in each colony.
caused by any technical problem in the culture, since there was no observed relation between the number of CFU-GM and the number of CFU-Meg. Nine patients had a number of CFU-Meg equal to or below median (median of CFU-Meg = 14), five of which had a number of CFU-GM below median (median of CFU-GM = 30), and four had a number of CFU-GM above median. Nine patients had a number of CFUMeg above median, five of which had a number of CFU-GM equal to or below median, and four had a number above median. The number of Megs per colony were not significantly different in acute ITP, chronic ITP and controls (Table 11). When all data were taken together (acute and chronic ITP) and related to platelet count, no significant correlation was observed between initial platelet count and numbers of CFU-Meg ( r = -0.42).However, a significant inverse correlation was observed between the number of Megs per colony and platelet count at day 0 (Spearman; r = -0.48;p c 0.05) (data not shown). A significant correlation also was observed between the
X
~~
lo5NAMC.
number of CFU-Meg and platelet recovery at day 7 (Spearman; r = 0.49;p < 0.05) and between the number of CFU-Meg and platelet recovery at day 30 (Spearman; r = 0.45; p < 0.05) (Table 11). A significant relation also was observed between the number of Megs per colony and platelet recovery at day 7 (Spearman; r = 0.51; p c 0.05). No difference was observed in the number of CFU-Meg between patients receiving steroids or not receiving steroids.
Discussion Twenty-one adult patients with ITP were studied for their ability to promote in vitro bone marrow Meg colony formation in order to characterize differences between acute and chronic ITP. Patients with acute ITP had a significantly higher number of CFU-Meg compared to chronic ITP; patients with chronic ITP had a significantly lower number of CFU-Meg compared to controls. These findings are in accordance
32
Meg Colony Formation and Idiopathic Thrombocytopenic Purpura
with De Alarcon [ 5 ] who also observed an increased number of CFU-Meg in children with acute ITP, but are different from the results of Sugiyama [6] showing an increased number of CFU-Meg in chronic ITP. Segal and Ballem [4] did not observe any difference from controls in CFU-Meg numbers in their patients. No significant difference was present in the number of Megs per colony between chronic ITP, acute ITP, and controls. Sugiyama observed a significant inverse correlation between platelet count and CFU-Meg numbers. This finding was not observed in the present study. When examining platelet recovery at day 7 and day 30 after culture in acute and chronic ITP, a significant relation was observed between the number of CFU-Meg and platelet recovery at day 7 and day 30. A significant inverse correlation was observed between platelet count at day 0 and the number of Megs per colony. When platelet count was low, the number of Megs per colony was high. The present study shows a difference between acute and chronic ITP in the ability to promote in vitro bone marrow Meg colony formation, Meg colony formation being normal in acute ITP and impaired in chronic ITP. The explanation for this impaired colony formation is unknown, but we can suggest some hypotheses: 1) the immune mechanism of thrombocytopenia could be different in acute ITP compared to chronic ITP, as suggested by McMillan who demonstrated that in chronic ITP, antiplatelet auto-antibody also binds to megakaryocytes [8]. In the present study, we have not tested the binding of patients’ auto-antibodies either on platelets or on megakaryocytes, but we could hypothesize that in acute ITP the antibody could be directed mainly against platelet antigens, and in chronic ITP the antibody could act at the level of Megs. 2 ) Another immune mechanism is suggested by Trenf [9] in chronic ITP who observed an abnormal T suppressor function. In the present work, T suppressor function was not tested, and T cell depletion was not done on bone marrow before plating. Therefore, we are unable to give any suggestion for a possible role of T suppressor cells on Meg colony formation in our patients. 3) Another explanation could be that Meg colony formation could be the result of different feedback mechanisms in response to thrombocytopenia in acute ITP and chronic ITP. Additionally, an important finding in this study was the correlation between culture data and platelet
recovery, i.e., when the numbers of CFU-Meg were high, platelet recovery at day 7 and day 30 was also high. The use of in vitro Meg colony culture in ITP thus could be valuable in clinical practice as predictive of platelet recovery. From this study we were unable to determine the value of Meg culture in predicting the efficiency of steroid therapy. Determination of the specificity of the antibodies present in these patients is under investigation currently in order to know if Meg culture patterns are related to any peculiar platelet antibody specificity.
Acknowledgment We thank f.McLaughlin for technical assistance in translation.
References Karpatkin S. Autoimmune thrombocytopenic purpura. Blood 198056~329-343. McMillan R. Chronic idiopathic thrombocytopenic purpura. N Eng J Med 1981;304:1135-1147. Burstein S, Adamson JW, Erb SK, Harker L. Megakaryocytopoiesis in the mouse: response to varying platelet demand. J Cell Physiol 1981;109:333-341. Segal GM, Ballern P, Gemsheimer T, Slichter SJ, Adamson JW.Autoimmune thrombocytopenia: progenitor cell response to platelet demand. In: Levine, RF, et al., eds. Megakaryocyte Development and Function. New York: Alan R. Liss, Inc., 1986;341-345. De Alarcon PA, Mazur EM, Schmieder JA. In vitro megakaryocytopoiesis in children with acute idiopathic thrombocytopenic purpura. Am J Ped Hematol Oncol 1987;9:212-218. Sugiyama H, Yagita M, Takahashi T, Nakamura K, Iho S, Hoshino T, Irnura H. Megakaryocytopoiesis in idiopathic thrombocytopenic purpura. Acta Haernatol Jap 198730:119-128. Han ZC, Briere J, Abgrall JF, Sensebe L, Parent D, Guem G. Characteristics of megakaryocyte colony formation i n normal individuals and in primary thrombocythemia: studies using an optimal cloning system. Exp Hematol 1989;17:46-52. McMillan R . Immune thrombocytopenia. Clin Haematol 1983; 12:69-88. Trent RJ, Clancy RL, Dank V, Basten A. Disordered immune homeostasis in chronic idiopathic thrornbocytopenic purpura. Clin Exp Irnmunol 1981;45:9- 17.
