Blood coagulation factors and platelet response to drug‐induced hepatitis and hepatosis in rats

Abstract Background Knowing the variability of blood coagulation responses to liver damage of different origins can provide a key to curing liver tissues or to mitigating treatment side effects. The aim of the present work was to compare the changes in the main components of hemostasis under experimental drug‐induced hepatosis and hepatitis in rats. Methods We modeled diclofenac‐induced hepatitis and tetracycline‐induced hepatosis. Hemostasis response was gauged by measuring fibrinogen, factor X, protein C (PC), and prothrombin in plasma. The decarboxylated form of prothrombin was detected by measuring prothrombin index and ecamulin index. Platelet reactivity was studied using aggregometry. Results Both hepatitis and hepatosis decreased the synthesis of fibrinogen, factor X, and prothrombin. However, protein carboxylation was not disrupted in hepatosis but was much impaired in hepatitis. PC decreased in both models as a consequence of its consumption possibly during inflammatory response. Platelet aggregation rate was lower in hepatosis but higher in hepatitis. Conclusions Our findings imply the need for a thorough monitoring of the hemostasis system in liver diseases to avoid possible thrombotic complications. Its state indicates the disorder's rate and character.

Common agents that damage the liver are medical preparations.
Antibiotics, sulfanilamides, hormones, and so on can cause pathological states that occasionally worsen into chronic diseases even more aggravating than the causes for their initial prescription. [4][5][6] Liver injury leads to the disruption of protein synthesis. 7,8 Hepatocytes are the main source of clotting factors. The disruption of liver functions leads to changes in the γ-carboxylation of vitamin K-dependent factors (mainly factors VII, IX, and X; prothrombin; and proteins С, S, and Z). When decarboxylated, these factors are functionally inactive. They can neither form enzymatic complexes on the surface of the membrane lipid bilayer nor be activated in the physiological clotting cascade. 9 Their accumulation lowers the procoagulant potential of blood plasma. [10][11][12] Toxic damage of liver can lead to pathological conditions such as necrosis, cholestasis, fat or toxic hepatitis, cirrhosis, and fibrosis.
All these processes imbalance the hemostasis system affecting this very sensitive mechanism on so many levels of regulation that it cannot be easily predicted or healed. Thromboses and hemorrhages are common for liver diseases. [13][14][15] Hemostasis dysfunctions during liver injuries are broadly studied. However, there is no direct head-to-head comparison of how the blood coagulation system (BCS) reacts to modeled liver injuries of different origins. Therefore, the main aim of the present work was to compare the system's response to hepatitis and hepatosis induced in the same rat line. We speculate that the specifics of BCS response to liver injuries of different origins can shed some light on how to eliminate side effects of drugs and cure the liver.
Ecamulin was purified from Echis multisquamatis venom according to the method of Solovjov et al. 16

| Animal models
Two different forms of drug-induced hepatopathologies were modeled: toxic hepatitis and fatty liver disease.
The experiment was performed using male Wistar rats (weight: 200-220 g). For 14 days until the experiment began, the animals were quarantined and clinically examined daily. They were fed a standard diet. Food and drinking water were provided ad libitum.
During the quarantine period, we monitored changes in their body weight and feed intake.

| Model of fatty hepatosis
The rats were divided into two groups: control and experimental (n = 10 in each group). For experimental modeling of acute fatty hepatosis according to Gryshchenko et al,17 the rats of the experimental group were intragastrically administered 4% aqueous solution of tetracycline hydrochloride once a day for 7 days at 250 mg/kg body weight.
Animals of the control group were intragastrically administered an equivalent volume of double-distilled water. The experiment lasted 1 week. The biological material was collected on the eighth day under ether anesthesia.

| Model of toxic hepatitis
This study also used two groups of rats: control and experimental (n = 10 in each group). Toxic hepatitis was induced by intragastric administration of a 4% aqueous solution of sodium diclofenac at 12.5 mg/kg body weight, once a day for 14 days, as described in Serdyukov et al. 18 The animals of the control group were given an equivalent amount of double-distilled water intragastrically. The biological material was collected on the 15th day.

| Plasma collection
The blood was collected by puncturing the heart. Sodium citrate (3.8%) was added to the whole blood in the ratio of 1:9 immediately after blood collection. For the aggregometry study, platelet-rich plasma (PRP) was obtained from whole blood by centrifugation at 160g for 30 min at 25°C. PRP was centrifuged at 1300g for 15 min at 25°C, and the platelet-poor plasma (PPP) was collected above the platelet pellet and frozen at −35°C. PPP was thawed at 37°C prior to the measurements.

| Biochemical and hematological testing
An automatic hematological analyzer BC 2800 Vet (Mindray, Shenzhen, China) was used to measure blood composition quickly.
In particular, it allowed the estimation of the number of white blood cells (WBC); number of lymphocytes, granulocytes, platelets, and red blood cells; percentage of lymphocytes, mid-sized cells, granulocytes, and platelets; and WBC histogram, hemoglobin concentration, and mean platelet volume.
Biochemical parameters of blood plasma samples were determined using a fully automatic analyzer BioSystem A15 (BioSystems, Barcelona, Spain). We focused on the measuring of alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase, γ-glutamyltranspeptidases, total bilirubin, and total protein levels.
The measurements were performed according to the manufacturer's recommendations.

| Carracci hematoxylin and eosin staining
Histological samples of the aorta for light-microscopic examination were prepared as previously described. 17,18 The samples were stained with hematoxylin and eosin solution according to Carracci and studied using microscopy. Stained histological samples were studied using an Olympus BX51 microscope (Olympus, Tokyo, Japan).

| Fibrinogen concentration
Fibrinogen concentration in the blood plasma was determined using the modified spectrophotometric method. The blood plasma (0.2 ml) and phosphate buffered saline (1.7 ml) were mixed in a glass tube.
Coagulation was initiated by adding 0.1 ml of thrombin-like enzyme from the venom of Aghistrodon halys halys (1 NIH/ml) to avoid fibrin cross-linking. The mixture was incubated for 30 min at 37°C. The fibrin clot was removed and dissolved in 5 ml of 1.5% аcetic acid. The concentration of protein was measured using the POP spectrophotometer (Optizen, Daejeon, Korea) at 280 nm (ε = 1.5). 19

| Factor X level
Total factor Х level was determined using RVV (reagent that specifically activates factor X, purchased from Sigma-Aldrich, Saint Louis, MO, USA) and factor Xa-specific chromogenic substrate S22765

| PC level
The total PC level in blood plasma was determined using PC activator and PC-specific chromogenic substrate S2236 (p-Glu-Pro-Arg-pNa). 21 In a 96-well plate, 0.02 ml of plasma sample, 0.03 ml of S2236 solution (0.25 mM), and 0.03 ml of PC activator solution were admixed in TBS with 0.001 М CaCl 2 to obtain a final volume of 0.25 ml.
The synthesis of colorful pNa was monitored at 405 nm using a ThermoMultiscan (ThermoFisher). Results are presented as percentage from control values.

| Prothrombin index
The prothrombin test is based on the application of thromboplastin that acts through the tissue factor pathway of coagulation and activates only the functionally active carboxylated and uncleaved forms of prothrombin.
The results of the prothrombin test were presented as prothrombin index (PI) calculated using the formula PI = An/Ap, where An is the blood plasma clotting time of healthy controls and Ap is the experimental blood plasma clotting time.

| Ecamulin index
The test is based on the application of prothrombin activator from the venom of E. multisquamatis-ecamulin. Ecamulin activates prothrombin, descarboxy-prothrombin, and prethrombin 1, thus permitting the determination of total prothrombin level. 22 The results of the ecamulin test were presented as ecamulin index (EI) calculated using the formula EI = An/Ap, where An is the blood plasma clotting time of healthy controls and Ap is the experimental blood plasma clotting time.

| Platelet aggregation
Platelet aggregation was measured based on changes in the turbidity of PRP. 23 In a typical experiment, 250 μl of PRP was incubated with 25 μl of 0.025 M CaCl 2 and 25 μl of 12.5 μM adenosine diphosphate at 37°C. Aggregation was monitored for 10 min using the aggregometer Solar 2110 (Solar, Minsk, Belarus).

| Statistics
Statistical data analysis was performed using the Kruskal-Wallis test (https://www.socsc istat istics.com/tests/ krusk al/defau lt.aspx). All blood coagulation assays were replicated thrice. The results are presented as boxplot diagrams with median, maximal and minimum values, and interquartile range. The results were considered significant at p < 0.05.
The Wilcoxon-Mann-Whitney test was used to estimate and compare the differences between two independent groups. This is widely used for comparisons of small groups by the level of any qualitatively measured attribute and allows to identify differences in the value of parameter. 24,1 1 If it will not be a problem, could You please delete this paragraph? Thank You in advance.

| Modeling of hepatosis and hepatitis
Administration of tetracycline for 7 days led to the complexity of characteristic symptoms of the acute stage of fatty hepatosis. 17 Depression, loss of appetite, thirst, weight loss by 10-15 g, dimming of the wool, and so on, were observed starting from the third day of the experiment. Analysis of blood cells demonstrated an increase in leukocytes by 27% (6.1 ± 0.5 × 10 9 /L vs. 4.8 ± 0.2 × 10 9 /L in the control group), a decrease in red blood cells by 27% (4.8 ± 0.3 × 10 12 /L vs. 6.6 ± 0.5 × 10 12 /L in the control group), and a decrease in hemoglobin level by 23% (141.0 ± 3.6 g/L vs. 182.7 ± 12.1 g/L in the control group). Hematocrit decreased from 35.2 ± 2.3% to 27.2 ± 1.7%, whereas plateletcrit increased (0.322 ± 0.016% vs. 0.249 ± 0.018% in controls).
We also observed an increase in the segmentoid neutrophils (36.0 ± 0.4% vs. 33.5 ± 0.8% in the control group) and a decrease in rod-shaped neutrophils (5.5 ± 0.5% vs. 8.0 ± 0.2% in the control group). The 2.3-fold increase in eosinophils (3.5 ± 0.5% vs. Hematological analysis detected leucocytosis by the increasing number of leucocytes from 5.7 ± 0.2 × 10 9 /L in the control group to 17.10 ± 0.8 × 10 9 /L with consequent shift of the neutrophil nucleus to the right side and a 2.6-fold decrease in the number of monocytes (3.6 ± 0.1% vs. 9.3 ± 0.3% in the control group). We also detected a decrease in the number of red blood cells (6.9 ± 0.2 × 10 12 /L vs. 3.9 ± 0.1 × 10 12 /L in controls), yet hematocrit and hemoglobin levels were unchanged. All this suggested an inflammation process and anemia. 26 The total protein concentration decreased from 41. 11.3 ± 0.4 IU in the control group).
These changes indicate pathological changes in liver function, including the disruption of pigment metabolism and cholestasis. 27 Morphological changes in liver during hepatosis or hepatitis are shown in Figure 1.

| BCS parameters
To evaluate the condition of the BCS we measured the main coagulation factors (fibrinogen, prothrombin, and factor X) and the crucial anticoagulant factor PC.
We found a tendency to reduce 15%-20% in fibrinogen level in blood plasma of both experimental groups compared to controls ( Figure 2). This was in spite of inflammatory processes accompanying liver injuries and can be associated with the suppression of protein synthesis.
Decrease in the plasma content of factor X and PC, which are also produced by the liver, confirmed this suggestion. Moreover, total PC decreased from 100 ± 10% in controls to 30 ± 6% and 51 ± 15% in the hepatosis and hepatitis groups, respectively (Figure 3). Decrease in the level of total factor X was less obvious. However, it was reduced up to 40% in both models and reached 84 16% in hepatitis and 77 ± 12% in hepatosis groups versus 100 ± 10% in controls (Figure 4). F I G U R E 2 Concentration of fibrinogen in blood plasma of rats with experimental drug-induced hepatosis (n = 10) and hepatitis (n = 10). Here and later only one control is presented, as the control parameters were similar in both. The p-value is 0.16525. The result is not significant at p < 0.05. The liver is also responsible for carboxylation of PC, fibrinogen, prothrombin, and factor X. In the case of prothrombin we were able to evaluate its total content as well as the content of carboxylated forms using a combination of two tests: EI and PI.
By applying EI we demonstrated a decrease in the total content of prothrombin-to 75 ± 20% in hepatitis and 80 ± 14% in hepatosis, similar to factor X and PC (20%-30%) (Figures 5 and 6). The use of PI also confirmed that prothrombin was not only less synthesized but also much less carboxylated in the case of hepatitis compared to controls ( Figure 4). Simultaneously, hepatosis affected the content of prothrombin but not its carboxylation ( Figure 5).
Pathological changes in protein synthesis and carboxylation during drug-induced hepatitis were also accompanied by a statistically significant increase in the rate of platelet aggregation (up to 60 ± 4.5% compared to 43 ± 11% in control). In contrast, hepatosis provoked the decrease in platelet aggregation rate to 39 ± 6.5% ( Figure 7).

| DISCUSS ION
Liver diseases are dangerous and spread widely, leading to acute liver failures and thus impairing the quality of human life. [28][29][30] Drug hepatotoxicity can also be a reason for acute liver failure. The symptoms can manifest immediately or several months into therapy.
Accurate diagnostics can be a basis for the selection of appropriate hepatoprotective therapy. 4,31,32 The pathophysiological state of the liver leads to disorders in many systems of the human body, including the very sensitive and important system of hemostasis. In our study we performed complex hemostasis analysis and confirmed significant changes in hemostasis in rat models of two drug-induced liver pathologies: hepatosis and hepatitis. Some of these changes (e.g., the decrease in PC or hyperaggregation of platelets) are alarming signs of hypercoagulation and the risk of thrombosis or bleeding. [33][34][35] These findings result in two main assumptions: (1) hemostasis system parameters should be monitored thoroughly during liver complications to avoid possible thrombotic complications, and (2) F I G U R E 4 Clotting factor X in blood plasma of rats with experimental drug-induced hepatosis (n = 10) and hepatitis (n = 10).
The p-value is 0.00593. The result is significant at p < 0.05. According to our data, liver diseases result in a decrease in main blood coagulation factors (factor X, prothrombin, and fibrinogen). The most obvious decrease was observed in the case of the main anticoagulant proenzyme PC. Its level decreased also because of the consumption during intravascular thrombin generation and inflammation. 36,37 We observed that both hepatitis and hepatosis decreased the synthesis of fibrinogen, factor X, and prothrombin. However, the results of the prothrombin and ecamulin tests suggest that carboxylation of proteins was not disrupted in hepatosis, so the vitamin K-dependent proteins in this pathology are functionally active. In the case of hepatitis, the functional properties of proteins were dramatically disordered due to the production of their decarboxylated forms that cannot be involved in the coagulation cascade in physiological pathways.
We also detected a tendency toward a lower platelet aggregation rate in the case of hepatosis, which can be assumed as a predisposition to bleeding. However, this fact must be evaluated in complex with other parameters of blood coagulation.
On the contrary, the increase in platelet aggregation rate during hepatitis confirms the imbalance in hemostasis, which must be studied more precisely.
We conclude that the determination of the rate of total synthesis and carboxylation of clotting factors can provide vital information on liver condition and the risk of bleeding/thrombosis. Therefore, we recommend detecting the decarboxylated forms of prothrombin during liver diseases using all validated methods available. 9,38,39 PC level should also be estimated during liver diseases as the most sensitive factor that can indicate the severity of liver disease and also the risk of intravascular coagulation. Finally, platelet aggregation should be considered as an additional functional test for analyzing the risk of intravascular coagulation during liver diseases. 40,41 Newly established differences between BCS responses to hepatosis and hepatitis indicate the necessity to study the pathological mechanisms of these diseases to find a method to avoid complications.

ACK N OWLED G M ENT
The authors are grateful to Dr Nadiia Druzhyna for her advice and help during the preparation of the manuscript.

FU N D I N G I N FO R M ATI O N
National Academy of Sciences of Ukraine research grant number 0119U002512.

CO N FLI C T O F I NTE R E S T
The authors declare no competing interests.

E TH I C S A PPROVA L
All animal studies were carried out in accordance with the standards of the "European Convention for the Protection of Vertebrate Animals