THROMBOSIS RESEARCH 63; 39-45,199i 0049-3848/91 $3.00 + .OO Printed in the USA. Copyright (c) 1991 Pergamon Press pk. All rights reserved.
EFFECTS OF WARFARIN ON VITAMIN K-DEPENDENT COAGULATION FACTORS IN ATLANTIC SALMON AND RAINBOW TROUT WITH SPECIAL REFERENCE TO FACTOR X Ragnar Salte and Kari Norberg AKVAFORSK
(Institute of Aquaculture Research), N 1432 AS-NLH, Norway.
(Received
11.21991;
accepted
in revised form 15.4.1991
P.O.Box
10,
by Editor U. Abildgaard)
ABSTRACT Plasma factor X (FX) was severely depleted in Atlantic salmon and rainbow trout after 18 days of oral warfarin administration; this response was further reflected in prolonged prothrombin time. Thus, both species require vitamin K for the maintenance of normal haemostasis. FX activity in controls of rainbow trout was 3to 5-fold that of salmon. The FX of both species show several of the features of human FX: it is inhibited by SBTI, inactivated in activated by RW, the presence of heparin, and adsorbs to insoluble barium precipitates. Rainbow trout sustained warfarin poisoning better than did salmon, with mortalities occurring 8 days later in trout. After 18 days all remaining fish had developed a severe haemorrhagic anaemia with poikilocytosis, leucopenia, and anisocytosis, thrombocytopenia.
A primary function of vitamin K in vertebrates is in the synthesis of cogulation factors. Reports as to its essentiality in fish are, however, scarce: diets deficient in vitamin K have lead to increased whole blood coagulation time and anaemia in brook trout Salvelinus fontinalis (l),rainbow trout Oncorhynchus mykiss (2), and lake trout Salvelinus namaycush (3), while Murai and Andrews (4), working also with the vitamin K antagonist Key words: warfarin,
vitamin K, haemostasis,
trout.
39
factor X, salmon,
40
WARFARIN & COAGULATION IN SALMON
Vol. 63, No. 1
dicoumarolconcludedthat channel catfish Ictalurus punctatus had marginal, if any, requirement for this vitamin. Vitamin K is essential in the carboxylation of glutamic acid to y-carboxyglutamic acid, which is present in residues coagulation factors II, VII, IX and X, and in Protein C (5) and Protein S (6). y-carboxyglutamic acid is required in the calcium phospholipid binding of the binding and calcium dependent coagulation proteins, and adsorb to insoluble barium precipitates acid to be (5,7). The latter authors showed y-carboxyglutamic present in rainbow trout plasma, and barium adsorbed factors have also been identified in the common carp Cyprinus carpio (8). The present study was conducted to obtain information on the essentiality of vitamin K for the maintenance of normal haemostasis in Atlantic salmon Salmo salar L. compared to rainbow trout, and to study the effect of a vitamin K antagonist on the vitamin K-dependent coagulation factor X. In the following the term factor X refers to a factor X-like activity.
MATERIALS
AND METHODS
Experimental fish Atlantic salmon and rainbow trout were held in one sea cage of 27 m3 capacity at ambient water temperature which averaged 8.3 OC. Fish nlasma Plasma from healthy Atlantic salmon and rainbow trout for the characterization of factor X were obtained from NIVA, Marine Research Station, Solbergstrand. Human nlasma Normal plasma was obtained from KabiVitrum Diagnostica, Melndal, Sweden. Abnormal control plasma was obtained from Nycomed a.s, Oslo, Norway. Reaoents a-Cbo-D-arg-gly-arg-pNA MW 714.6 (S 2765) was obtained from KabiVitrum. Normotest (NT), Thrombotest (TT) and Cephotest reagents were obtained from Nycomed. Bovine thrombin (TopostasineR) was obtained from Hoffmann-La Roche & Co AG, Basel, Switzerland. Russell's Viper Venom (RW) was obtained from KabiVitrum. Soybean trypsin inhibitor (SBTI) and BaSO, were obtained from Sigma Chemical Co, St. Louis, USA. Heparin sodium was obtained from Novo Nordisk AS, Copenhagen, Denmark. Warfarin sodium was obtained from the Norwegian Medicinal Depot, Oslo, and gelatine capsules were filled to 50mg or 500 ug by Ulleval Hospital Pharmacy, Oslo. Methods. Atlantic salmon (n=15, mean weight approximately 350g) and rainbow trout (n=15, mean weight approximately 500 g& were individually marked with Carlin tags and put into one 27 m sea cage. After two weeks adaptation to handfeeding with a standard moist pelleted feed, fish were netted one at a time, and anaesthetized in well aerated seawater containing 300 mg/l chlorbutanol. All fish were given a gelatine capsule containing 50 mg warfarin sodium into the stomach with a pair of tweezers. Every third salmon and rainbow trout sampled were placed on a plastic mat ventral side up prior to warfarin administration. Blood was drawn from the caudal vein into siliconized, evacuated blood collecting tubes (VenojectR) containing 0.13M Na, citrate as anticoagulant (9 volumes of blood into 1 volume of citrate), and centrifuged at 2,000 g for 10 minutes. Plasma was pipetted off into plastic tubes within 15 minutes and analyzed within 3
Vol. 63, No. 1
WARFARIN & COAGULATION IN SALMON
hours. Similarely, additional blood samples were taken with heparin as anticoagulant (45 U.S.P Units heparin sodium per 3 ml of blood). The fish were allowed to recover in a fiber glass tank to control whether they kept the capsule down, after which they were returned to the cage. The fish were handfed ad libitum twice a day with freshly made moist pellets, each containing one gelatine capsule with 500 1.19of warfarin throughout the experiment. The number of pellets consumed was counted each day. After 8 and 18 days 5, respectively 3, of each species were individually netted and anaesthetized as above. Citrated and heparinized blood, and blood with I$-EDTA 0.06ml O.l9mol/l per 3ml as anticoagulant (after 18 days only) was drawn as above. All fish sampled were subsequently killed by spinal transsection and submitted to post mortem examination. Mortalities were recorded daily, and dead fish were all submitted to post mortem examination. The experiment was terminated after 5 weeks, and the remaining fish were killed. Prothrombin time was measured with bovine (TT) and rabbit (NT) thromboplastin. Both tests as well as the activated partial thromboplastintime (APTT) test Cephotestwere performed manually according to manufacturer's recommendations; we have previously reported that coagulation tests developed for human purposes effectively disclose haemostatic disorders also in salmonids(9). Fibrinogen values were determined by the DADE method based on the procedure described by Clauss (lO).For the accuracy and precision control of the analyses an abnormal control plasma was included in each series of analyses. Factor X activity was determined on a Titertec Multiscan spectrophotometer (Labsystems, Helsinki, Finland) at 20°C: Citrated plasma was diluted I.:20 in 50 mm01 Tris-HCL with 20 mg/l polybrene pH 7.8. Equal volumes of RW 0.067 mg/ml and CaCl, 0.1 mol were mixed. 60~1 diluted plasma was mixed with 60 ~1 RW/CaCl,. After 60 s 60 ~1 of S 2765 1.2 mm01 in distilled water was added. The amidolytic activity was read after 90 min at 405 nm. To further classify the factor X activity salmon and rainbow trout plasmas were either preincubated with BaSO, 0.25g in 0.5ml plasma under continuous tilting for 90 min or diluted 1:lO and mixed with equal volumes of SBTI 8mg/ml in 50 mm01 Tris-HCl pH 7.8, alternatively both, prior to assay as above. In addition the amidolytic activity was determined in plasma diluted 1:20 in Tris-HCl pH 7.8 with 3U/ml heparin. Packed cell volume was determined from heparinized blood by a microhaematocrit method. Erythrocyte, leukocyte, and thrombocyte counts were determined from EDTA blood diluted with Dacie's fluid in a Biirker haemacytometer. Blood cell morphology was determined from smears (EDTA-blood) stained with May-Griinwald and Giemsa. RESULTS Both Atlantic salmon and rainbow trout required vitamin K to maintain normal haemostasis; both species responded to warfarintreatment with salmon being the fastest responder (Table 1 and 2). Mortalities occurred 8 days earlier in salmon than in rainbow trout. From the onset of mortalities and onwards the dying proceeded at the same rate in both species.
41
42
WARFARIN & COAGULATION IN SALMON
Vol. 63, No. 1
TABLE 1 Response in Coagulation and Haematological parameters in Atlantic salmon during Oral Administration of warfarin. Dosage regimen see text. Days
Thrombotest, s Normotest, s Cephotest, s Fibrinogen, mg% FX, %= HCT, % Erythrocytes, mill mme3 MCV, /Jo Leucocytes, 1000 mme3 Thrombocytea, 1000 mmS3
O(n=S)
8(n=5)
18(n=3)
88(75-93)a
>300b >300 ,300 43(37-50) 12(10-15) 34.8(32-6) 1.1(0.9-1.3) 318(265-337) 37.1(31.6-45.4) 2.6(0.8-4.4)
>900b
88.4(76-97)
30.4(29-32)
154(130-190) 20.8(17-25) 36(32-38) 1.2(0.9-1.4) 313(263-404) 23.9(19.2-30.8) 2.2(1.6-3.6)
>900 >900
Cl5 3.7(3-4) 14.3(10-18) 0.5(0.4-0.6) 322(270-405) 2.6(0.8-4.4) 0.9(0.8-1.2)
a
mean values, range in brackets bobserved for 300, respectively 900 s 'relative to a human reference (100%) plasma The normal plasma factor X activity in controls of rainbow trout (Time 0, Table 2) was 3- to 5-fold that of salmon (Time 0, Table 1). After 18 days on warfarin factor X was severely depleted in both species. At the same time all fish had developed a severe haemorrhagic anaemia with anisocytosis, poikilocytosis, leucopenia and thrombocytopenia.
TABLE 2 Response in Coagulation and Haematological parameters Trout during Oral Administration of Warfarin.
in Rainbow
Days
Thrombotest, s Normotest, s Cephotest, s Fibrinogen, mg% FX, % HCT, % Erythrocytes, mill mmo3 MCV, /Jo Leucocytes, 1000 mmS3 Thrombocytes, 1000 mmm3
O(n=5)
8(n=5)
74(61-103) 73.8(57-117) 28(26-31) 202(195-210) 112(95-124) 30.2(28-35) 1.2(1.1-1.4) 246(215-318) 20.6(16.4-25.2) 4.2(3.6-5.2)
>300 142.5(n=3>300) 32,3(28-40) 173(145-195) 23.4(16-35) 28.4(25-32) 1.2(1.0-1.4) 252(180-308) 21.8(16.4-35.2) 5.4(2.8-6.4)
18(n=3) >900 >900 .900
EFFECTS OF WARFARIN ON VITAMIN K-DEPENDENT COAGULATION FACTORS IN ATLANTIC SALMON AND RAINBOW TROUT WITH SPECIAL REFERENCE TO FACTOR X Ragnar Salte and Kari Norberg AKVAFORSK
(Institute of Aquaculture Research), N 1432 AS-NLH, Norway.
(Received
11.21991;
accepted
in revised form 15.4.1991
P.O.Box
10,
by Editor U. Abildgaard)
ABSTRACT Plasma factor X (FX) was severely depleted in Atlantic salmon and rainbow trout after 18 days of oral warfarin administration; this response was further reflected in prolonged prothrombin time. Thus, both species require vitamin K for the maintenance of normal haemostasis. FX activity in controls of rainbow trout was 3to 5-fold that of salmon. The FX of both species show several of the features of human FX: it is inhibited by SBTI, inactivated in activated by RW, the presence of heparin, and adsorbs to insoluble barium precipitates. Rainbow trout sustained warfarin poisoning better than did salmon, with mortalities occurring 8 days later in trout. After 18 days all remaining fish had developed a severe haemorrhagic anaemia with poikilocytosis, leucopenia, and anisocytosis, thrombocytopenia.
A primary function of vitamin K in vertebrates is in the synthesis of cogulation factors. Reports as to its essentiality in fish are, however, scarce: diets deficient in vitamin K have lead to increased whole blood coagulation time and anaemia in brook trout Salvelinus fontinalis (l),rainbow trout Oncorhynchus mykiss (2), and lake trout Salvelinus namaycush (3), while Murai and Andrews (4), working also with the vitamin K antagonist Key words: warfarin,
vitamin K, haemostasis,
trout.
39
factor X, salmon,
40
WARFARIN & COAGULATION IN SALMON
Vol. 63, No. 1
dicoumarolconcludedthat channel catfish Ictalurus punctatus had marginal, if any, requirement for this vitamin. Vitamin K is essential in the carboxylation of glutamic acid to y-carboxyglutamic acid, which is present in residues coagulation factors II, VII, IX and X, and in Protein C (5) and Protein S (6). y-carboxyglutamic acid is required in the calcium phospholipid binding of the binding and calcium dependent coagulation proteins, and adsorb to insoluble barium precipitates acid to be (5,7). The latter authors showed y-carboxyglutamic present in rainbow trout plasma, and barium adsorbed factors have also been identified in the common carp Cyprinus carpio (8). The present study was conducted to obtain information on the essentiality of vitamin K for the maintenance of normal haemostasis in Atlantic salmon Salmo salar L. compared to rainbow trout, and to study the effect of a vitamin K antagonist on the vitamin K-dependent coagulation factor X. In the following the term factor X refers to a factor X-like activity.
MATERIALS
AND METHODS
Experimental fish Atlantic salmon and rainbow trout were held in one sea cage of 27 m3 capacity at ambient water temperature which averaged 8.3 OC. Fish nlasma Plasma from healthy Atlantic salmon and rainbow trout for the characterization of factor X were obtained from NIVA, Marine Research Station, Solbergstrand. Human nlasma Normal plasma was obtained from KabiVitrum Diagnostica, Melndal, Sweden. Abnormal control plasma was obtained from Nycomed a.s, Oslo, Norway. Reaoents a-Cbo-D-arg-gly-arg-pNA MW 714.6 (S 2765) was obtained from KabiVitrum. Normotest (NT), Thrombotest (TT) and Cephotest reagents were obtained from Nycomed. Bovine thrombin (TopostasineR) was obtained from Hoffmann-La Roche & Co AG, Basel, Switzerland. Russell's Viper Venom (RW) was obtained from KabiVitrum. Soybean trypsin inhibitor (SBTI) and BaSO, were obtained from Sigma Chemical Co, St. Louis, USA. Heparin sodium was obtained from Novo Nordisk AS, Copenhagen, Denmark. Warfarin sodium was obtained from the Norwegian Medicinal Depot, Oslo, and gelatine capsules were filled to 50mg or 500 ug by Ulleval Hospital Pharmacy, Oslo. Methods. Atlantic salmon (n=15, mean weight approximately 350g) and rainbow trout (n=15, mean weight approximately 500 g& were individually marked with Carlin tags and put into one 27 m sea cage. After two weeks adaptation to handfeeding with a standard moist pelleted feed, fish were netted one at a time, and anaesthetized in well aerated seawater containing 300 mg/l chlorbutanol. All fish were given a gelatine capsule containing 50 mg warfarin sodium into the stomach with a pair of tweezers. Every third salmon and rainbow trout sampled were placed on a plastic mat ventral side up prior to warfarin administration. Blood was drawn from the caudal vein into siliconized, evacuated blood collecting tubes (VenojectR) containing 0.13M Na, citrate as anticoagulant (9 volumes of blood into 1 volume of citrate), and centrifuged at 2,000 g for 10 minutes. Plasma was pipetted off into plastic tubes within 15 minutes and analyzed within 3
Vol. 63, No. 1
WARFARIN & COAGULATION IN SALMON
hours. Similarely, additional blood samples were taken with heparin as anticoagulant (45 U.S.P Units heparin sodium per 3 ml of blood). The fish were allowed to recover in a fiber glass tank to control whether they kept the capsule down, after which they were returned to the cage. The fish were handfed ad libitum twice a day with freshly made moist pellets, each containing one gelatine capsule with 500 1.19of warfarin throughout the experiment. The number of pellets consumed was counted each day. After 8 and 18 days 5, respectively 3, of each species were individually netted and anaesthetized as above. Citrated and heparinized blood, and blood with I$-EDTA 0.06ml O.l9mol/l per 3ml as anticoagulant (after 18 days only) was drawn as above. All fish sampled were subsequently killed by spinal transsection and submitted to post mortem examination. Mortalities were recorded daily, and dead fish were all submitted to post mortem examination. The experiment was terminated after 5 weeks, and the remaining fish were killed. Prothrombin time was measured with bovine (TT) and rabbit (NT) thromboplastin. Both tests as well as the activated partial thromboplastintime (APTT) test Cephotestwere performed manually according to manufacturer's recommendations; we have previously reported that coagulation tests developed for human purposes effectively disclose haemostatic disorders also in salmonids(9). Fibrinogen values were determined by the DADE method based on the procedure described by Clauss (lO).For the accuracy and precision control of the analyses an abnormal control plasma was included in each series of analyses. Factor X activity was determined on a Titertec Multiscan spectrophotometer (Labsystems, Helsinki, Finland) at 20°C: Citrated plasma was diluted I.:20 in 50 mm01 Tris-HCL with 20 mg/l polybrene pH 7.8. Equal volumes of RW 0.067 mg/ml and CaCl, 0.1 mol were mixed. 60~1 diluted plasma was mixed with 60 ~1 RW/CaCl,. After 60 s 60 ~1 of S 2765 1.2 mm01 in distilled water was added. The amidolytic activity was read after 90 min at 405 nm. To further classify the factor X activity salmon and rainbow trout plasmas were either preincubated with BaSO, 0.25g in 0.5ml plasma under continuous tilting for 90 min or diluted 1:lO and mixed with equal volumes of SBTI 8mg/ml in 50 mm01 Tris-HCl pH 7.8, alternatively both, prior to assay as above. In addition the amidolytic activity was determined in plasma diluted 1:20 in Tris-HCl pH 7.8 with 3U/ml heparin. Packed cell volume was determined from heparinized blood by a microhaematocrit method. Erythrocyte, leukocyte, and thrombocyte counts were determined from EDTA blood diluted with Dacie's fluid in a Biirker haemacytometer. Blood cell morphology was determined from smears (EDTA-blood) stained with May-Griinwald and Giemsa. RESULTS Both Atlantic salmon and rainbow trout required vitamin K to maintain normal haemostasis; both species responded to warfarintreatment with salmon being the fastest responder (Table 1 and 2). Mortalities occurred 8 days earlier in salmon than in rainbow trout. From the onset of mortalities and onwards the dying proceeded at the same rate in both species.
41
42
WARFARIN & COAGULATION IN SALMON
Vol. 63, No. 1
TABLE 1 Response in Coagulation and Haematological parameters in Atlantic salmon during Oral Administration of warfarin. Dosage regimen see text. Days
Thrombotest, s Normotest, s Cephotest, s Fibrinogen, mg% FX, %= HCT, % Erythrocytes, mill mme3 MCV, /Jo Leucocytes, 1000 mme3 Thrombocytea, 1000 mmS3
O(n=S)
8(n=5)
18(n=3)
88(75-93)a
>300b >300 ,300 43(37-50) 12(10-15) 34.8(32-6) 1.1(0.9-1.3) 318(265-337) 37.1(31.6-45.4) 2.6(0.8-4.4)
>900b
88.4(76-97)
30.4(29-32)
154(130-190) 20.8(17-25) 36(32-38) 1.2(0.9-1.4) 313(263-404) 23.9(19.2-30.8) 2.2(1.6-3.6)
>900 >900
Cl5 3.7(3-4) 14.3(10-18) 0.5(0.4-0.6) 322(270-405) 2.6(0.8-4.4) 0.9(0.8-1.2)
a
mean values, range in brackets bobserved for 300, respectively 900 s 'relative to a human reference (100%) plasma The normal plasma factor X activity in controls of rainbow trout (Time 0, Table 2) was 3- to 5-fold that of salmon (Time 0, Table 1). After 18 days on warfarin factor X was severely depleted in both species. At the same time all fish had developed a severe haemorrhagic anaemia with anisocytosis, poikilocytosis, leucopenia and thrombocytopenia.
TABLE 2 Response in Coagulation and Haematological parameters Trout during Oral Administration of Warfarin.
in Rainbow
Days
Thrombotest, s Normotest, s Cephotest, s Fibrinogen, mg% FX, % HCT, % Erythrocytes, mill mmo3 MCV, /Jo Leucocytes, 1000 mmS3 Thrombocytes, 1000 mmm3
O(n=5)
8(n=5)
74(61-103) 73.8(57-117) 28(26-31) 202(195-210) 112(95-124) 30.2(28-35) 1.2(1.1-1.4) 246(215-318) 20.6(16.4-25.2) 4.2(3.6-5.2)
>300 142.5(n=3>300) 32,3(28-40) 173(145-195) 23.4(16-35) 28.4(25-32) 1.2(1.0-1.4) 252(180-308) 21.8(16.4-35.2) 5.4(2.8-6.4)
18(n=3) >900 >900 .900
