Cancer procoagulant in acute lymphoblastic leukemia Alessio MG, Falanga A, Consonni R, Bassan R, Minetti B, Donati MB, Barbui T. Cancer procoagulant in acute lymphoblastic leukemia. Eur J Haematol 1990: 45: 78-81.
M. G. Alessio’, A. Falanga, R. Consonni, R. Bassan, B. Minetti, M. B. Donati and T. Barbui
In a previous study we characterized cancer procoagulant (CP), a 68 kd cysteine proteinase which directly activates coagulation factor X in various subtypes (from M 1 to M5) of acute non-lymphoblastic leukemia (ANLL). The aim of this study was to determine whether CP is also expressed by acute lymphoblastic leukemia (ALL) cells. Blasts from 25 ALL patients were extracted and tested for their procoagulant properties. 16 samples (64%) shortened the recalcification time of normal human plasma, and 9 (36%) did not. 8 of the 16 active samples showed properties compatible with CP, i.e. independence from factor VII in triggering blood coagulation and sensitivity to cysteine proteinase inhibitors. Selected samples also cross-reacted with a polyclonal antibody raised against purified CP. The specific activity of CP in ALL extracts was significantly lower than in most ANLL types previously studied (all but M4). These finding indicate that CP can be a property of the lymphoid phenotype although its expression may be lower than in the myeloid phenotype.
Divisione di Ematologia, Ospedali Riuniti, Bergamo, and lstituto di Ricerche Farmacologiche Mario Negri, Consorzio Mario Negri Sud, S. Maria Imbaro, Italy
Procoagulant activities of cancer cells are considered relevant to the hypercoagulable state in malignancy (1,2). Besides tissue factor (TF) - the procoagulant most frequently described in leukemic cells (3-6) another procoagulant has recently been identified in acute non-lymphoblastic leukemia (ANLL) blasts (7). It shows properties of “cancer procoagulant” (CP), a factor VII (FVI1)-independent cysteine proteinase with factor X (FX)-activating activity (8, 9). Coagulation abnormalities can occur in acute lymphoblastic leukemia (ALL) (10-12), although severe thrombo-hemorrhagic accidents are reported more rarely than in ANLL. We set out here to determine: 1) whether CP activity is associated with ALL cells; 2) whether there is any relation with their immunophenotype; and 3) how the amount of this activity compares to that of ANLL cells. This study on 25 ALL samples provides evidence that CP can be a property of ALL cells; it was expressed by 1/3 of analyzed samples, but never by malignant T cells. When present, CP specific activity was lower than in most of the ANLL subtypes (7).
* Recipient of a fellowship of the “Associazione Italiana per la Ricerca sul Cancro”. 78
Key words: acute lymphoblastic leukemia cancer procoagulant - tissue procoagulant activity Correspondence:Dr. A. Falanga, M.D., Divisione di Ematologia, Ospedali Riuniti, Largo Barozzi, 1-24100 Bergamo, Italy Accepted for publication 5 March 1990
Material and methods Patients
25 blast cells of patients with ALL (12 males and 13 females, age range 3-57 years) were studied at diagnosis, before chemotherapy. On the basis of immunological markers, the blasts were classified as: 13 “common”, 3 pre-B, 3 “null”, 4 T and 2 B ALL (13). Blast cell extracts
20 to 50 ml of peripheral blood (PB, 18 samples) or 1 to 5 ml bone marrow (BM, 7 samples) were collected in sodium citrate (1 : 10) and diluted 1: 2 with phosphate buffered saline (PBS), pH 7.4, as described (7). Mononuclear cells (> 99% blasts) were separated by Ficoll Hypaque (Lymphoprep, Nyegaard, Oslo) density gradient centrifugation. The lymphoid lineage was identified by phase microscopy after May-Grunwald-Giemsa and cytochemical stainings (PAS, Sudan black, ANAE, peroxidases). ALL cell subtyping was done by indirect immunofluorescence using a panel of monoclonal antibodies directed against surface antigens (13). After separation, mononuclear cells ( 2 5 0 x lo6) were washed three times in PBS, pH 7.4, and resuspended in 1 to 5 ml 20mmol/l Veronal buffer, pH 7.8. Extraction was carried out at 4”C, as described (7). A control group of samples was processed in
Leukemic cell procoagulant activity
parallel and consisted of peripheral blood from 4 healthy volunteers and 1 BM from a subject with mild iron-deficiency anemia. Protein content was determined by Bradford’s method (14). Procoagulant activity (PCA) assay
PCA of blast extracts was measured by “one-stage recalcification assay” of normal human plasma (NHP) (7). To assess the presence of a FVII-independent procoagulant, human plasma congenitally deficient in FVII (FVII def. plasma, Behringwerke AG, Marburg, Germany) was used instead of normal plasma in the clotting assay. Standard tissue factor (TF) (rabbit brain thromboplastin, RBT, Sigma Chemical Company, St. Louis, MO) and Russell’s viper venom (RW, Sigma), a standard FX activator, were the coagulation controls for the assay in the absence of FVII. PCA was expressed as seconds or as R W units/mg protein. Units were calculated on a standard curve obtained with different dilutions (from lo-’ to of R W , as described (7). Percent FVII-independent activity corresponded to specific activity in FVII-deficient plasma/specific activity in NHP x 100.
Results
PCA of normal and FVII-deficient plasma of 25 ALL blast extracts is shown in Table 1. 16 samples (64%) shortened the recalcification time of NHP, with activity ranging from 0.43 to 22.6 R W units/mg (mean 7.0 k 1.8 SE). 8 (32%) of them (4 “common”, 1 pre-B, 2 “null”, 1 B ALL), 6 from PB and 2 from BM, were active in both the presence and absence of FVII. The range of specific activity was 0.65-22.6 R W units/mg protein (mean 8.7 2.7 SE) in NHP and 0.2-3 R W units/mg (mean 1.7 f 0.3 SE) in FVII-deficient plasma. FVII-independent activity ranged from 9 to 100% of total PCA. 8 (32%) samples (3 “common”, 2 pre-B, 1 “null”, 1 T, 1 B ALL), 5 from PB and 3 from BM, only showed TF-like procoagulant. Their range of activity was 0.53-17.6 R W units/mg (mean 5.9 2.3 SE) in NHP and 0 in FVII-deficient plasma. 9 (36%) extracts (6 “common”, 3 T ALL), 7 from PB and 2 from BM, did not show any PCA. Control mononuclear cell extracts (4 PB and 1 BM) did not possess PCA in normal or FVII-deficient plasma. Table 1. Specific procoagulant activity of cell extracts from 25 ALL patients Specific activity
Inhibition study
N.
Patient
Type
Sample.
NHP’
FVIIdef.#
% FVll indep. act.
Enzymatic characteristics were defined by testing the sensitivity of selected ALL samples to a series of cysteine proteinase inhibitors including HgC12 (Sigma), iodoacetamide (IA, Sigma) and the peptidyl-diazo-methyl-ketone Z-Ala-Ala CHN2 (ZAA-CHN2, kindly provided by Prof. E. Shaw, Friedrich Miescher Institut, Basel, CH) (15), and to a T F inhibitor, Concanavalin A (ConA, Sigma). R W , a serine protease FX activator, papain, a cysteine protease FX activator, and RBT, a standard TF, were the controls of the inhibition study. Samples and standards were incubated for 30 min at 25 “ C with HgC12 (0.1 mmol/l), IA (2 mmol/l) or ZAA-CHN2 (200 pmol/l) before the plasma recalcification assay. They were incubated with Con A (200 pg/ml) for 50 min at 37°C before testing for PCA.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
O.G. V.C. M.R. C.K. D.A. D.T. D.A. B.D.
preB B Common Common null Common null Common preB preB Common B Common Common null T Common Common Common Common Common Common T T T
BM P P P P BM P P P BM BM P BM P P P P BM P P P
16.3 0.65 5.1 2.1 11.5 3.0 22.6 3.49 1.2 1.5 0.53 17.6 11.6 0.43 12.1 2.39 0 0 0 0 0 0 0 0 0
2.72 0.2 1.3 1.07 1.1 3.0 2.0 2.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
17 31 26 50 10 100 9 63 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Immunological study
ALL extracts were immunologically characterized by double immunodiffusion in 1% agar gel, using a goat polyclonal antibody raised against purified CP (14), kindly provided by Dr S.G. Gordon (University of Colorado, Denver, CO, USA). Samples or controls were allowed to react at 4°C for 48 hours, as previously described (7).
0.6. B.C. C.A. C.P. C.D. M.G. C.P. C.L. C.R. P.D. M.V. F.G. M.M. F.F. Z.M. A.A. R.I.
P BM P P
BM = bone marrow, P = peripheral blood; ‘NHP = normal human plasma; #FVIIdef. = FVIIdeficient plasma. Specific activity is expressed as RW unitslmg protein (see text): 1 unit = activity of 1 mEq/ml of RW in the onestage clotting assay. Results are the means of at least two determinations. Percent FVII-independentactivity corresponds to: (specific activity in FVIIdef. plasma/ specific activity in NHP) X 100.
79
Alessio et al. Table 2. Sensitivity to cysteine proteinase inhibitors and to Con A of ALL samples " A ' (with F VII-independent activity) and "B' (without F VII-independent activity) Clotting time (sec)
"A'
"6'
Inhibitor
Concentration
Untreated samples
Treated samples
lodoacetamide HgC12
149.3f 32.4 136.3f 20.12 131.2f 26.2 147.3f 16.35
197.7 f23.15 192.9 f 25.68 199 f22.75 169.8 f 22.10
0.05 0.025 0.025
Con A
1 mM 0.1 mM 200 pM 200 pg/ml
lodoacetamide HgCI2 ZAA-CHN2 Con A
1 mM 0.1 mM 200 pM 200 pM
159 f 18.3 140.5f 17.7 153.7f 15.2 154.3f14
171.3f 19 155.4f 13 155 f17.4 218.7 f 21.4
N.S. N.S. N.S.
ZAA-CHN2
P
M. G. Alessio’, A. Falanga, R. Consonni, R. Bassan, B. Minetti, M. B. Donati and T. Barbui
In a previous study we characterized cancer procoagulant (CP), a 68 kd cysteine proteinase which directly activates coagulation factor X in various subtypes (from M 1 to M5) of acute non-lymphoblastic leukemia (ANLL). The aim of this study was to determine whether CP is also expressed by acute lymphoblastic leukemia (ALL) cells. Blasts from 25 ALL patients were extracted and tested for their procoagulant properties. 16 samples (64%) shortened the recalcification time of normal human plasma, and 9 (36%) did not. 8 of the 16 active samples showed properties compatible with CP, i.e. independence from factor VII in triggering blood coagulation and sensitivity to cysteine proteinase inhibitors. Selected samples also cross-reacted with a polyclonal antibody raised against purified CP. The specific activity of CP in ALL extracts was significantly lower than in most ANLL types previously studied (all but M4). These finding indicate that CP can be a property of the lymphoid phenotype although its expression may be lower than in the myeloid phenotype.
Divisione di Ematologia, Ospedali Riuniti, Bergamo, and lstituto di Ricerche Farmacologiche Mario Negri, Consorzio Mario Negri Sud, S. Maria Imbaro, Italy
Procoagulant activities of cancer cells are considered relevant to the hypercoagulable state in malignancy (1,2). Besides tissue factor (TF) - the procoagulant most frequently described in leukemic cells (3-6) another procoagulant has recently been identified in acute non-lymphoblastic leukemia (ANLL) blasts (7). It shows properties of “cancer procoagulant” (CP), a factor VII (FVI1)-independent cysteine proteinase with factor X (FX)-activating activity (8, 9). Coagulation abnormalities can occur in acute lymphoblastic leukemia (ALL) (10-12), although severe thrombo-hemorrhagic accidents are reported more rarely than in ANLL. We set out here to determine: 1) whether CP activity is associated with ALL cells; 2) whether there is any relation with their immunophenotype; and 3) how the amount of this activity compares to that of ANLL cells. This study on 25 ALL samples provides evidence that CP can be a property of ALL cells; it was expressed by 1/3 of analyzed samples, but never by malignant T cells. When present, CP specific activity was lower than in most of the ANLL subtypes (7).
* Recipient of a fellowship of the “Associazione Italiana per la Ricerca sul Cancro”. 78
Key words: acute lymphoblastic leukemia cancer procoagulant - tissue procoagulant activity Correspondence:Dr. A. Falanga, M.D., Divisione di Ematologia, Ospedali Riuniti, Largo Barozzi, 1-24100 Bergamo, Italy Accepted for publication 5 March 1990
Material and methods Patients
25 blast cells of patients with ALL (12 males and 13 females, age range 3-57 years) were studied at diagnosis, before chemotherapy. On the basis of immunological markers, the blasts were classified as: 13 “common”, 3 pre-B, 3 “null”, 4 T and 2 B ALL (13). Blast cell extracts
20 to 50 ml of peripheral blood (PB, 18 samples) or 1 to 5 ml bone marrow (BM, 7 samples) were collected in sodium citrate (1 : 10) and diluted 1: 2 with phosphate buffered saline (PBS), pH 7.4, as described (7). Mononuclear cells (> 99% blasts) were separated by Ficoll Hypaque (Lymphoprep, Nyegaard, Oslo) density gradient centrifugation. The lymphoid lineage was identified by phase microscopy after May-Grunwald-Giemsa and cytochemical stainings (PAS, Sudan black, ANAE, peroxidases). ALL cell subtyping was done by indirect immunofluorescence using a panel of monoclonal antibodies directed against surface antigens (13). After separation, mononuclear cells ( 2 5 0 x lo6) were washed three times in PBS, pH 7.4, and resuspended in 1 to 5 ml 20mmol/l Veronal buffer, pH 7.8. Extraction was carried out at 4”C, as described (7). A control group of samples was processed in
Leukemic cell procoagulant activity
parallel and consisted of peripheral blood from 4 healthy volunteers and 1 BM from a subject with mild iron-deficiency anemia. Protein content was determined by Bradford’s method (14). Procoagulant activity (PCA) assay
PCA of blast extracts was measured by “one-stage recalcification assay” of normal human plasma (NHP) (7). To assess the presence of a FVII-independent procoagulant, human plasma congenitally deficient in FVII (FVII def. plasma, Behringwerke AG, Marburg, Germany) was used instead of normal plasma in the clotting assay. Standard tissue factor (TF) (rabbit brain thromboplastin, RBT, Sigma Chemical Company, St. Louis, MO) and Russell’s viper venom (RW, Sigma), a standard FX activator, were the coagulation controls for the assay in the absence of FVII. PCA was expressed as seconds or as R W units/mg protein. Units were calculated on a standard curve obtained with different dilutions (from lo-’ to of R W , as described (7). Percent FVII-independent activity corresponded to specific activity in FVII-deficient plasma/specific activity in NHP x 100.
Results
PCA of normal and FVII-deficient plasma of 25 ALL blast extracts is shown in Table 1. 16 samples (64%) shortened the recalcification time of NHP, with activity ranging from 0.43 to 22.6 R W units/mg (mean 7.0 k 1.8 SE). 8 (32%) of them (4 “common”, 1 pre-B, 2 “null”, 1 B ALL), 6 from PB and 2 from BM, were active in both the presence and absence of FVII. The range of specific activity was 0.65-22.6 R W units/mg protein (mean 8.7 2.7 SE) in NHP and 0.2-3 R W units/mg (mean 1.7 f 0.3 SE) in FVII-deficient plasma. FVII-independent activity ranged from 9 to 100% of total PCA. 8 (32%) samples (3 “common”, 2 pre-B, 1 “null”, 1 T, 1 B ALL), 5 from PB and 3 from BM, only showed TF-like procoagulant. Their range of activity was 0.53-17.6 R W units/mg (mean 5.9 2.3 SE) in NHP and 0 in FVII-deficient plasma. 9 (36%) extracts (6 “common”, 3 T ALL), 7 from PB and 2 from BM, did not show any PCA. Control mononuclear cell extracts (4 PB and 1 BM) did not possess PCA in normal or FVII-deficient plasma. Table 1. Specific procoagulant activity of cell extracts from 25 ALL patients Specific activity
Inhibition study
N.
Patient
Type
Sample.
NHP’
FVIIdef.#
% FVll indep. act.
Enzymatic characteristics were defined by testing the sensitivity of selected ALL samples to a series of cysteine proteinase inhibitors including HgC12 (Sigma), iodoacetamide (IA, Sigma) and the peptidyl-diazo-methyl-ketone Z-Ala-Ala CHN2 (ZAA-CHN2, kindly provided by Prof. E. Shaw, Friedrich Miescher Institut, Basel, CH) (15), and to a T F inhibitor, Concanavalin A (ConA, Sigma). R W , a serine protease FX activator, papain, a cysteine protease FX activator, and RBT, a standard TF, were the controls of the inhibition study. Samples and standards were incubated for 30 min at 25 “ C with HgC12 (0.1 mmol/l), IA (2 mmol/l) or ZAA-CHN2 (200 pmol/l) before the plasma recalcification assay. They were incubated with Con A (200 pg/ml) for 50 min at 37°C before testing for PCA.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
O.G. V.C. M.R. C.K. D.A. D.T. D.A. B.D.
preB B Common Common null Common null Common preB preB Common B Common Common null T Common Common Common Common Common Common T T T
BM P P P P BM P P P BM BM P BM P P P P BM P P P
16.3 0.65 5.1 2.1 11.5 3.0 22.6 3.49 1.2 1.5 0.53 17.6 11.6 0.43 12.1 2.39 0 0 0 0 0 0 0 0 0
2.72 0.2 1.3 1.07 1.1 3.0 2.0 2.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
17 31 26 50 10 100 9 63 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Immunological study
ALL extracts were immunologically characterized by double immunodiffusion in 1% agar gel, using a goat polyclonal antibody raised against purified CP (14), kindly provided by Dr S.G. Gordon (University of Colorado, Denver, CO, USA). Samples or controls were allowed to react at 4°C for 48 hours, as previously described (7).
0.6. B.C. C.A. C.P. C.D. M.G. C.P. C.L. C.R. P.D. M.V. F.G. M.M. F.F. Z.M. A.A. R.I.
P BM P P
BM = bone marrow, P = peripheral blood; ‘NHP = normal human plasma; #FVIIdef. = FVIIdeficient plasma. Specific activity is expressed as RW unitslmg protein (see text): 1 unit = activity of 1 mEq/ml of RW in the onestage clotting assay. Results are the means of at least two determinations. Percent FVII-independentactivity corresponds to: (specific activity in FVIIdef. plasma/ specific activity in NHP) X 100.
79
Alessio et al. Table 2. Sensitivity to cysteine proteinase inhibitors and to Con A of ALL samples " A ' (with F VII-independent activity) and "B' (without F VII-independent activity) Clotting time (sec)
"A'
"6'
Inhibitor
Concentration
Untreated samples
Treated samples
lodoacetamide HgC12
149.3f 32.4 136.3f 20.12 131.2f 26.2 147.3f 16.35
197.7 f23.15 192.9 f 25.68 199 f22.75 169.8 f 22.10
0.05 0.025 0.025
Con A
1 mM 0.1 mM 200 pM 200 pg/ml
lodoacetamide HgCI2 ZAA-CHN2 Con A
1 mM 0.1 mM 200 pM 200 pM
159 f 18.3 140.5f 17.7 153.7f 15.2 154.3f14
171.3f 19 155.4f 13 155 f17.4 218.7 f 21.4
N.S. N.S. N.S.
ZAA-CHN2
P
