Melatonin does not affect disseminated intravascular

Thrombosis Research 139 (2016) 38–43

Contents lists available at ScienceDirect

Thrombosis Research journal homepage: www.elsevier.com/locate/thromres

Full Length Article

Melatonin does not affect disseminated intravascular coagulation but diminishes decreases in platelet count during subacute endotoxaemia in rats Maren Oude Lansink a, Klaus Görlinger b,c, Matthias Hartmann b, Herbert de Groot a, Katharina Effenberger-Neidnicht a,⁎ a b c

Institute for Physiological Chemistry, University Hospital Essen, Essen, Germany Department of Anaesthesiology and Intensive Care, University Hospital Essen, Essen, Germany TEM International GmbH, Munich, Germany

a r t i c l e

i n f o

Article history: Received 27 July 2015 Received in revised form 30 September 2015 Accepted 13 October 2015 Available online 28 October 2015 Keywords: Lipopolysaccharide Sepsis Thromboelastometry Platelet Hemolysis

a b s t r a c t Introduction: Inhibitory effects of exogenous melatonin (MLT) on plasma coagulation and platelet aggregation have already been observed in vivo and in vitro under normal conditions. Here, we studied whether MLT also diminishes the lipopolysaccharide (LPS)-induced disseminated intravascular coagulation (DIC) during subacute endotoxaemia. Materials and methods: Subacute endotoxaemia was induced in male Wistar rats by an intravenous infusion of LPS over a period of 300 min (0.5 mg LPS/kg × h). MLT was administered intravenously 15 min before and 120 min and 240 min after starting of the LPS infusion (3 × 3 mg MLT/kg × 15 min). The kinetic of clot formation was analysed by thromboelastometry. Results: Infusion of LPS led initially to a significant reduction of clotting time (120 min, LPS: 150 ± 21 s vs. SHAM: 292 ± 36 s), and finally a significant increase of clotting time (300 min, LPS: 2768 ± 853 s vs. SHAM: 299 ± 67 s) and a slight increase of clot formation time (300 min, LPS: 1038 ± 657 s vs. SHAM: 98 ± 14 s) as well as a significant decrease of alpha-angle (300 min, LPS: 35 ± 15° vs. SHAM: 72 ± 3°), maximum clot firmness (300 min, LPS: 22 ± 6 mm vs. SHAM: 68 ± 3 mm), and area under the curve (300 min, LPS: 1657 ± 552 mm × 100 vs. SHAM: 6849 ± 307 mm × 100). Simultaneously, a decrease of platelet count (300 min, LPS: 55 ± 8 vs. SHAM: 180 ± 55) and a release of cell-free haemoglobin (240 min, LPS: 46 ± 5 μmol/L vs. SHAM: 16 ± 2 μmol/L) could be observed in the course of subacute endotoxaemia. The additional administration of MLT did not reduce the LPS-induced alterations in parameters of thromboelastometry, but significantly reduced the LPS-induced decrease of platelet count (300 min, LPS + MLT: 130 ± 10) and release of cell-free haemoglobin (240 min, LPS + MLT: 29 ± 3 μmol/L). Conclusion: Melatonin does not affect DIC but diminishes thrombocytopenia and haemolysis during endotoxaemia. © 2015 Published by Elsevier Ltd.

1. Introduction Sepsis is a systemic inflammatory response of the host to an infection and is frequently associated with disseminated intravascular coagulation (DIC), which is a strong predictor of mortality in patients with severe sepsis [1,2]. DIC is characterised by systemic activation of coagulation with simultaneous loss of compensation mechanisms like fibrinolysis and anticoagulant-acting proteins, leading to both, microvascular thrombosis and increased bleeding tendency [3]. The consequence of this can be ⁎ Corresponding author at: Institut für Physiologische Chemie, Universitätsklinikum Essen, Hufelandstr. 55, D-45147 Essen, Germany. E-mail address: [email protected] (K. Effenberger-Neidnicht).

http://dx.doi.org/10.1016/j.thromres.2015.10.025 0049-3848/© 2015 Published by Elsevier Ltd.

the development of ischemia and necrosis of organs up to multi organ dysfunction and failure [3]. In septic patients and animal models of endotoxaemia, the fibrin deposition in blood vessels as a result of DIC maybe also contribute to destruction of erythrocytes membranes and, in turn, a release of haemoglobin [4]. Endotoxaemia induced by injections of lipopolysaccharide (LPS), the endotoxin of Gram-negative bacteria, is a widely used experimental model to study DIC during systemic inflammation [5]. The initiation of coagulation by LPS has been found to result from the induction of the expression of tissue factor on endothelial cells and monocytes [6–8]. Beyond that, a direct stimulation of platelet aggregation and secretion by LPS through the toll-like receptor 4 (TLR4)-dependent pathway has also been observed [9]. Furthermore, anticoagulant proteins and fibrinolysis have been suppressed in LPS-induced DIC as a result of

M.O. Lansink et al. / Thrombosis Research 139 (2016) 38–43

decreased expression of thrombomodulin on endothelial cells and increased plasma levels of plasminogen activator inhibitor 1 [6,10,11]. Melatonin (MLT) is a circadian hormone, which is synthesised and secreted primarily by the pineal gland depending on the daylightdarkness cycle with a peak in the night even in nocturnal animals [12]. Due to its lipophilic nature, melatonin has the capability to interact with lipid bilayers. In this respect, an average shallow position in the membrane is suggested [13]. Its receptor-dependent actions are mediated by the G-protein coupled receptors MT1 and MT2 as well as the putative receptor MT3, which has been identified as the enzyme quinone reductase 2 [14–16]. Up to now, the protective effects of MLT during sepsis or endotoxaemia have primarily been attributed to its antiinflammatory and antioxidative action [17]. Only one clinical trial with septic newborns has studied the beneficial effect of MLT treatment in humans, but the emphasis was on survival rate and antioxidative efficacy [18]. Nevertheless, this trial has given the first hint for an effect of MLT on platelet count during sepsis, because the MLT-treated septic newborns had significantly increased platelet counts 24 h and 48 h after MLT administration in comparison to their baseline values and to untreated septic newborns. The coagulation activation and platelet aggregability have a diurnal maximum early in the morning, when melatonin reaches its minimal diurnal values in blood [19–21]. Inhibitory effects of exogenous MLT on plasma coagulation and platelet aggregation have already been observed in vivo and in vitro under normal conditions [22–26]. In healthy young men, the administration of MLT has resulted in diminished blood coagulation activity through the reduction of the mean values of coagulation factor VIII and fibrinogen [22]. In vitro, inhibition of arachidonic acid-, adenosine diphosphate- and collagen-induced platelet aggregation and delta-granule secretion, thromboxane A2 production, and serotonin uptake in platelets by high exogenous doses of MLT have been observed [23–26]. Based on the aforementioned antihaemostatic effects of exogenous MLT, we assume that MLT is a suitable substance to prevent or reduce the development of DIC during (experimental) systemic inflammation or even sepsis. Therefore, we performed the present study with male Wistar rats to examine the effect of exogenous MLT on LPS-induced DIC during subacute endotoxaemia. 2. Materials and methods 2.1. Chemicals and materials Lipopolysaccharide (LPS; from Escherichia coli, serotype 0111:B4) and melatonin (MLT) were purchased from Sigma-Aldrich (St. Louis, USA), buffer tablets pH 6.8, May Gründwald's eosine methylene blue solution modified and Giemsa's azur eosine methylene blue solution from Merck (Darmstadt, Germany), object and cover slides from Engelbrecht Medizin und Labortechnik (Edermünde, Germany). BD Plastipak™ 1 ml was obtained from Becton Dickinson (Madrid, Spain), ethanol ≥99.5% from Carl Roth (Karlsruhe, Germany), isoflurane from AbbVie Deutschland (Ludwigshafen, Germany), ketamine 10% from Ceva (Düsseldorf, Germany), lidocaine (xylocaine 1%) from AstraZeneca (Wedel, Germany), medical oxygen from Air Liquide (Düsseldorf, Germany), Portex® catheters (0.58 mm i.d./0.96 mm o.d.) from Smith Medical International (Grasbrunn, Germany), Ringer's solution from Fresenius Kabi (Bad Homburg, Germany), 2.0 ml self-filling arterial samplers containing 80 IU electrolyte-balanced heparin (PICO50) from Radiometer Medical (Brønshøj, Denmark), sodium citrate solution 3.13% from Eifelfango (Bad Neuenahr-Ahrweiler, Germany) and sodium chloride solution (0.9%) from B. Braun (Melsungen, Germany). 2.2. Animals Male Wistar rats (400–500 g) were obtained from the central animal unit of the University Hospital Essen. Rats were kept under standardised

39

conditions of temperature (22 ± 1 °C), humidity (55 ± 5%), and 12-h/ 12-h light/dark cycles (06:00 a.m. light on/06:00 p.m. light off, standard time) with free access to food (ssniff-Spezialdiäten, Soest, Germany) and water. All rats received human care according to the standards of the Federation of European Laboratory Animal Science Association (FELASA). The experimental procedure has been reviewed and approved by the local Animal Care and Use Committee with the permit number Az.: 84-02.04.2012.A205. 2.3. Anaesthesia, analgesia and surgical procedure Rats were anesthetised with isoflurane (2% in 100% medical O2 at 4 × 103 ml/min for anaesthesia induction; 1.5–2% at 1 × 103 ml/min throughout the experiment) and received ketamine (50 mg/kg, s.c.) and lidocaine (5 mg/kg, s.c.) for analgesia. Afterwards, catheters were inserted in femoral artery and vein. The arterial catheter was perfused with Ringer's solution at a rate of 3 ml/h and the venous catheter with 0.9% sodium chloride solution at a rate of 7 ml/kg × h to maintain the functionality of the catheters. Mean arterial blood pressure (MAP) was recorded via the femoral artery catheter in 10-min intervals. Core temperature was monitored using a rectal sensor and was kept constant above 37 °C with an underlying heated operating table and by covering the animals with aluminium foil. 2.4. Experimental groups LPS was dissolved in 0.9% sodium chloride solution and filtered through a 0.2 μm filter (Minisart®; Sartorius, Göttingen, Germany). Subacute endotoxaemia was induced by intravenous infusion of LPS (0.5 mg LPS/kg × h) using a syringe pump (Fresenius Injectomat 2000; Fresenius Kabi, Bad Homburg, Germany) at a rate of 7 ml/kg × h over a total period of 300 min (Fig. 1). LPS administration started at approximately 09:00 a.m. (standard time). MLT was dissolved in a mixture of ethanol and 0.9% sodium chloride solution (1:90, EtOH:NaCl) filtered. MLT (3 × 3 mg/kg) was infused intravenously using a syringe pump (Medfusion Inc. 2010; Smith Medical International, Grasbrunn, Germany) as three 15-min infusions directly before and 120 min and 240 min after starting LPS administration (Fig. 1) according to our previous study testing the time and frequency of MLT dosing [27]. A similar time, dose, and frequency of MLT treatment was already found to beneficial in experimental sepsis, whereby the emphasis was on hemodynamic effects and anti-oxidative efficacy [28,29]. Experiments were performed with 8 rats per group for SHAM and MLT and 12 rats per group for LPS and LPS + MLT, unless stated otherwise. The following experimental groups were compared: • SHAM (sham control group): 0.9% sodium chloride solution, vehicle, • MLT (melatonin control group): 0.9% sodium chloride solution, 3 × 3 mg MLT/kg × 15 min, vehicle, • LPS (lipopolysaccharide control group): 0.5 mg LPS/kg × h, vehicle, • LPS + MLT (LPS melatonin group): 0.5 mg LPS/kg × h, 3 × 3 mg MLT/kg × 15 min, vehicle. 2.5. Assessment of blood and plasma parameters Blood samples were taken from the femoral artery catheter immediately before starting the LPS infusion (0 min) and at any further hour (60, 120, 180, 240, 300 min) in one tenth volume of citrate (3.13%) or using 2.0 ml self-filling arterial samplers, containing 80 IU electrolytebalanced heparin. Heparinized blood plasma was obtained from blood samples by centrifugation at 3000 ×g for 15 min at 4 °C. Aliquots were stored at −80 °C until their use. The concentration of cell-free haemoglobin was measured in heparinized blood plasma samples using the spectroscopic soret band method [30]. Citrate blood samples were used immediately for blood smear preparation and thromboelastometry

40


Fig. 1. Time scale of the administration of lipopolysaccharide (LPS) and of melatonin (MLT). Catheters were inserted in femoral artery and vein before starting the LPS and MLT administration. Subacute endotoxaemia was induced by an intravenous infusion of LPS (0.5 mg LPS/kg × h over 300 min). MLT was infused 15 min before and 120 min and 240 min after starting the LPS infusion (3 × 3 mg MLT/kg × 15 min). Blood samples were taken from the femoral artery catheter directly before starting the induction of endotoxaemia (0 min) and any further hour (60, 120, 180, 240, 300 min).

(ROTEM®, TEM International, Munich, Germany). Each blood smear was stained by Pappenheim method to determine the amount of platelets per thousand erythrocytes [31]. The kinetic of clot formation was analysed by thromboelastometry using NATEM test over a total time of 2 h and was described in a curve (TEMogram) and in numerical parameters [32]. In NATEM test, blood coagulation was initiated by recalcification with 20 μl start-TEM® (200 mM CaCl2). Clotting time (CT) is the time from test start until an amplitude of 2 mm is reached. Clot formation time (CFT) is the time between 2 mm amplitude and 20 mm amplitude. Alpha-angle is the angle between the baseline and a tangent to the clotting curve through 2 mm point. Maximum clot firmness (MCF) is the maximum amplitude reached during the test run time. Area under the curve (AUC) is the area under the first derivative curve from start of the derivative curve until the maximum clot firmness is reached. In principle, these numerical parameters are similar in rats and humans, but the onset of coagulation in rats is more rapid. This was already demonstrated by Wohlauer et al. [33] using thromboelastography, which is a comparable method to thromboelastometry. The authors concluded that citrated native whole blood is optimal in the assessment of coagulation in rodents and described reference ranges of thromboelastography parameters for Sprague Dawley rats. Moreover, thrombo-elastometry was already used to analyse disseminated intravascular coagulation (DIC) in a model of subacute endotoxaemia with pigs and in septic patients [34–36]. 2.6. Statistics Experiments were performed with 8 rats per group for SHAM and MLT and 12 rats per group for LPS and LPS + MLT, except for platelet count including 5 rats per group for SHAM, 6 rats per group for MLT and 8 rats per group for LPS and LPS + MLT due to missing measurements. The including number of rats was set on empirical basis. The data are presented as mean values ± SEM, analyses were performed using GraphPad Prism 6 (GraphPad Prism Software, Inc., San Diego, USA). All data for non-recurring measures (parameters of thromboelastometry, platelet count and plasma concentration of cell-free haemoglobin) were tested for normality using the Brown–Forsythe test and were determined to have a Gaussian distribution. Homogeneity of variance was tested using Bartlett's test. When Bartlett's test indicated that the group comparison had equal variance, one way analysis of variance (ANOVA) was used followed by a Fisher LSD post-hoc analysis for comparison among multiple groups. An a priori alpha error p of less than 0.05 was considered statistically significant (95% confidence interval). Outliers were identified by creating box-and-whiskers plots. Differences in mean arterial pressure over time from baseline within each group were determined by repeated-measures ANOVA. A Gaussian distribution of the data was assumed. Since sphericity was not assumed, the Geisser-Greenhouse correction was applied. Comparisons among multiple groups were evaluated by one-way ANOVA followed by a

Fisher LSD post-hoc analysis. An a priori alpha error p of less than 0.05 was considered statistically significant (95% confidence interval). 3. Results 3.1. Effect of melatonin (MLT) on thromboelastometry during subacute endotoxaemia The parameters of thromboelastometry did not alter in melatonin control group (MLT) and sham control group (SHAM) during the whole experimental time (Fig. 2). The induction of subacute endotoxaemia by a continuous infusion of lipopolysaccharide (LPS) led to a significant reduction of clotting time (CT) at time point 120 min (LPS: 150 ± 21 s vs. SHAM: 292 ± 36 s, p = 0.0032) and 180 min (LPS: 145 ± 27 s vs. SHAM: 249 ± 29 s, p = 0.0316; Fig. 2A). Following longer time of observation, the LPS control group (LPS) showed a significant increase of CT (300 min, LPS: 2768 ± 853 s vs. SHAM: 299 ± 67 s, p = 0.0318) and a slight increase of clot formation time (CFT, 300 min, LPS: 1038 ± 657 s vs. SHAM: 98 ± 14 s, p = 0.0580) as well as a significant decrease of alpha-angle (300 min, LPS: 35 ± 15 ° vs. SHAM: 72 ± 3 s, p = 0.0006), maximum clot firmness (MCF, 300 min, LPS: 22 ± 6 mm vs. SHAM: 68 ± 3 mm, p b 0.0001), and area under the curve (AUC, 300 min, LPS: 1657 ± 552 mm × 100 mm vs. SHAM: 6849 ± 307 mm × 100, p b 0.0001) in comparison to SHAM (Fig. 2). The administration of MLT during subacute endotoxaemia did not reduce the LPS-induced alterations in the parameters of thromboelastometry (LPS + MLT; Fig. 2). 3.2. Effect of melatonin (MLT) on platelet count and plasma concentration of cell-free haemoglobin during subacute endotoxaemia In sham control group (SHAM), the platelet count was not a subject to significant variation (Fig. 3A). In melatonin control group (MLT), however, the platelet count was significantly increased starting at time point 180 min in comparison to SHAM (300 min, MLT: 306 ± 19 vs. SHAM: 180 ± 55, p = 0.0011, Fig. 3A). Subacute endotoxaemia resulted in a significant decrease of platelet count 240 min after starting of LPS infusion till the end of experiment (300 min, LPS: 55 ± 8 vs. SHAM: 180 ± 55, p = 0.0009, Fig. 3A). The administration of MLT during subacute endotoxaemia significantly reduced the LPS-induced decrease of platelet count (300 min, LPS + MLT: 130 ± 10 vs. LPS: 55 ± 8, p = 0.0100, Fig. 3A). Cell-free haemoglobin was low in sham control group rats (SHAM) and was not influenced by MLT (Fig. 3B). Subacute endotoxaemia caused a significant increase in the plasma concentration of cell-free haemoglobin 180 min after starting of LPS infusion till the end of experiment (Fig. 3B). The administration of MLT diminished the LPS-induced release of cell-free haemoglobin, which was significant at time point 240 min (240 min, LPS: 46 ± 5 μmol/L vs. SHAM: 16 ± 2 μmol/L, p b 0.0001; LPS + MLT: 29 ± 3 μmol/L vs. LPS: 46 ± 5 μmol/L, p = 0.0021, Fig. 3B).


41

Fig. 2. Effect of melatonin (MLT) on parameters of thromboelastometry during subacute endotoxaemia. The clotting time (CT; A), the clot formation time (CFT; B), the alpha-angle (C), the maximum clot firmness (MCF; D), and the area under the curve (AUC; B) of the groups SHAM (n = 8), MLT (n = 8), LPS (n = 12) and LPS + MLT (n = 12) are shown (mean values ± SEM). Subacute endotoxaemia was induced by an intravenous infusion of LPS (0.5 mg LPS/kg × h over 300 min). MLT was infused 15 min before and 120 min and 240 min after starting the LPS infusion (3 × 3 mg MLT/kg × 15 min). #p b 0.05 vs. SHAM.

3.3. Effect of melatonin (MLT) on mean arterial pressure during subacute endotoxaemia Mean arterial blood pressure (MAP) was physiologic for the whole experimental time in the sham control group rats (SHAM; Fig. 4). Treatment with MLT resulted in an increase of MAP from 70 min to 300 min in comparison to SHAM (p b 0.0001 for MLT vs. SHAM, Fig. 4). Subacute endotoxaemia induced by LPS infusion led to a significant increase of MAP starting at time point 130 min (p = 0.0243 for LPS vs. SHAM, Fig. 4). The administration of MLT during subacute endotoxaemia significantly enhanced MAP 70 min after starting of LPS-infusion till the end of experiment in comparison to LPS control group (p b 0.0001 for LPS + MLT vs. LPS, Fig. 4).

4. Discussion In the present study, the lipopolysaccharide control group (LPS) showed initially a reduction of clotting time (CT) in the analysis of thromboelastometry and subsequently an increase of CT and clot formation time (CFT) as well as a decrease of area under the curve

(AUC), maximum clot firmness (MCF), and alpha-angle (Fig. 2). Simultaneously, a decrease of platelet count could be observed during manifestation of subacute endotoxaemia (Fig. 3A). The initial reduction of CT points out a hypercoagulable state (presumably resulting from a release of tissue factor from monocytes and endothelial cells [6–8]); the increase of CT and CFT as well as the decrease of AUC, MCF, alpha-angle, and platelet count afterwards are an indication of a hypocoagulable state resulting from a massive consumption of coagulation factors and platelets [34]. Both states of coagulation are typical for disseminated intravascular coagulation (DIC) and were already detected in a model of subacute endotoxaemia with pigs and in septic patients by thromboelastometry [34–36]. In our experiments, the administration of melatonin (MLT) during subacute endotoxaemia did not reduce the LPS-induced alterations in the parameters of thromboelastometry (Fig. 2). In consequence, the administration of MLT does not prevent the LPS-induced activation of coagulation and resulting consumption coagulopathy (DIC). In some contrast to our results, the oral administration of 3 mg MLT has shown an inhibitory effect on plasmatic coagulation in healthy young men [22]. A possible explanation for this discrepancy could be the strong activation of coagulation through LPS, which can apparently not be

42


Fig. 3. Effect of melatonin (MLT) on platelet count (A) and plasma concentration of cell-free haemoglobin (B) during subacute endotoxaemia. Platelet count per 1000 erythrocytes (A) and plasma concentration of cell-free haemoglobin (B) of the groups SHAM (A: n = 5; B: n = 8), MLT (A: n = 6; B: n = 8), LPS (A: n = 8; B: n = 12) and LPS + MLT (A: n = 8; B: n = 12) are shown (mean values ± SEM). Subacute endotoxaemia was induced by an intravenous infusion of LPS (0.5 mg LPS/kg × h over 300 min). MLT was infused 15 min before and 120 min and 240 min after starting the LPS infusion (3 × 3 mg MLT/kg × 15 min). # p b 0.05 vs. SHAM,*p b 0.05 vs. LPS.

prevented by the intravenous administration of even higher cumulative MLT dose (3 × 3 mg/kg). In addition, the daytime of MLT administration could also be crucial to achieve an adequate antihaemostatic effect, given the fact that MLT is a circadian hormone, which is synthesised

Fig. 4. Effect of melatonin (MLT) on mean arterial pressure (MAP) during subacute endotoxaemia. Mean arterial pressure (MAP) of the groups SHAM (n = 8), MLT (n = 8), LPS (n = 12) and LPS + MLT (n = 12) are shown (mean values ± SEM, n ≥ 8). Subacute endotoxaemia was induced by an intravenous infusion of LPS (0.5 mg LPS/kg × h over 300 min). MLT was infused 15 min before and 120 min and 240 min after starting the LPS infusion (3 × 3 mg MLT/kg × 15 min). #p b 0.05 MLT vs. SHAM, ##p b 0.05 LPS vs. SHAM, *p b 0.05 LPS + MLT vs. LPS.

and secreted by pineal gland in darkness, and its effectors are subjected to this diurnal variation [25,37,38]. Beyond that, the absence of a significant diminishment of LPS-induced DIC by exogenous MLT could be caused by a significant enhancement of the LPS-induced release of the cytokine IL-6 [39]. IL-6 is known to influence coagulation by increasing expression of tissue factor, fibrinogen, factor VIII and VWF and by reducing levels of inhibitory proteins such as antithrombin and protein S [40]. Although the administration of MLT did not reduce the LPS-induced DIC during subacute endotoxaemia, interestingly, a significant diminishment of the LPS-induced decrease of platelet count by MLT was detected (Fig. 3A). Since the platelet count was also elevated in the MLT control group, it presumably directly stimulates the formation of platelets and/or their release into blood (Fig. 3A). Protective effects on platelet recovery by administration of MLT have already been shown in an irradiated mouse model and in patients with thrombocytopenia [41–43]. Due to the fact, that megakaryopoiesis takes over several days, and in our in vivo model the effect of MLT was already visible after three hours, an increase of platelet count by MLT through megakaryopoiesis is rather unlikely. One possible mechanism could be a rapid release of platelets by the spleen, where 30% of the platelets are stored [44]. A general involvement of alpha-adreno receptor in the release of platelets from spleen is controversially discussed [45,46]. Beyond that, MLT membrane receptors MT1 and MT2, whose expression have already been detected in spleen, was maybe involved in a MLT-induced release of platelets from spleen [47]. However, the elevated platelet count could also be attributed to the observed MLT-induced increase in mean arterial pressure (MAP) as well (Fig. 4), because shear stress stimulates the release of platelets from its progenitor cells [48]. According to this, a simultaneous shear stress stimulated-release of cell-free haemoglobin would be expectable. On the contrary to that, we observed a diminishment of cell-free haemoglobin during subacute endotoxaemia by an administration of MLT (Fig. 3B). Due to the missing reductive effect of MLT on LPS-induced DIC, a protection of erythrocytes from fibrin deposition-caused lysis could be excluded [4]. The reduced haemolysis by MLT was maybe conditioned by its known capability to interact with lipid bilayers [13]. The possible interaction of MLT with the membrane of erythrocytes may diminish the LPS-induced influence of membrane stability [49]. In septic rats, an increase of the LPS-induced reduced deformability of erythrocytes by MLT has already been shown, albeit it has been attributed to its antioxidative action [50,51]. In the present study, some limitations exist: Our model of subacute endotoxaemia reflects a mild form of systemic inflammation [39] and has to be distinguished from peracute endotoxaemia or endotoxaemic shock. Thus, we only can make a statement of the MLT effect during mild systemic inflammation accompanied by minor changes in all system organs. However, we cannot exclude that MLT is protective when functional impairment and tissue injury is more severe (multiple organ failure), e.g. following LPS administration in higher doses or as a bolus. In any case, we expect a beneficial effect of MLT on platelet count also during severe systemic inflammation with major changes in all system organs. Moreover, we only can make a statement of the effect of a mid-morning application of MLT (3 × 3 mg/kg). Albeit MLT is a circadian hormone with a diurnal peak in the night [12], the major effect of putative anti-coagulant MLT in pharmaceutical doses would be expectable in the morning, because coagulation activation and platelet aggregability have a diurnal maximum early in the morning [19,20]. Furthermore, we only can make a statement of the effect of a MLT prophylaxis plus treatment, since MLT was applied as three 15-min infusion directly before and every two hours after starting of the LPS administration. Accordingly, we cannot infer to a mere therapeutic application of MLT (at a time when sepsis/endotoxaemia has become manifest). On the other hand, however, one clinical trial with septic newborns has already indicated an effect of a therapeutic administered MLT on platelet count [18]. To sum up, despite these limitations, the present results indicate that exogenous MLT has no protective effect on LPS-induced DIC during


subacute endotoxaemia. The beneficial effects of MLT on platelet count and hemolysis may result from a shear-stress or spleen-mediated release of platelet and its capability to stabilise erythrocytes, respectively. Abbreviations AUC CFT CT DIC EtOH IL-6 LPS MAP MCF MLT MT1 MT2 MT3 SEM TLR4

area under the curve clot formation time clotting time disseminated intravascular coagulation ethanol interleukin 6 lipopolysaccharide mean arterial pressure maximum clot firmness melatonin melatonin receptor 1 melatonin receptor 2 melatonin receptor 3 standard error of the mean toll-like receptor 4

References [1] M. Levi, The coagulant response in sepsis, Clin. Chest Med. 29 (2008) 627–642. [2] J.F. Dhainaut, S.B. Yan, D.E. Joyce, V. Pettila, B. Basson, J.T. Brandt, et al., Treatment effects of drotrecogin alfa (activated) in patients with severe sepsis with or without overt disseminated intravascular coagulation, J. Thromb. Haemost. 2 (2004) 1924–1933. [3] M. Levi, Disseminated intravascular coagulation, Crit. Care Med. 35 (2007) 2191–2195. [4] H. Heyes, W. Kohle, B. Slijepcevic, The appearance of schistocytes in the peripheral blood in correlation to the degree of disseminated intravascular coagulation. An experimental study in rats, Haemostasis 5 (1976) 66–73. [5] L.O. Berthelsen, A.T. Kristensen, M. Tranholm, Animal models of DIC and their relevance to human DIC: a systematic review, Thromb. Res. 128 (2011) 103–116. [6] K.L. Moore, S.P. Andreoli, N.L. Esmon, C.T. Esmon, N.U. Bang, Endotoxin enhances tissue factor and suppresses thrombomodulin expression of human vascular endothelium in vitro, J. Clin. Invest. 79 (1987) 124–130. [7] M. Guha, M.A. O'Connell, R. Pawlinski, A. Hollis, P. McGovern, S.F. Yan, et al., Lipopolysaccharide activation of the MEK-ERK1/2 pathway in human monocytic cells mediates tissue factor and tumor necrosis factor alpha expression by inducing Elk-1 phosphorylation and Egr-1 expression, Blood 98 (2001) 1429–1439. [8] K. Zacharowski, C. Sucker, P. Zacharowski, M. Hartmann, Thrombelastography for the monitoring of lipopolysaccharide induced activation of coagulation, Thromb. Haemost. 95 (2006) 557–561. [9] G. Zhang, J. Han, E.J. Welch, R.D. Ye, T.A. Voyno-Yasenetskaya, A.B. Malik, et al., Lipopolysaccharide stimulates platelet secretion and potentiates platelet aggregation via TLR4/MyD88 and the cGMP-dependent protein kinase pathway, J. Immunol. 182 (2009) 7997–8004. [10] M. Colucci, J.A. Paramo, D. Collen, Generation in plasma of a fast-acting inhibitor of plasminogen activator in response to endotoxin stimulation, J. Clin. Invest. 75 (1985) 818–824. [11] H. Asakura, Y. Suga, T. Yoshida, Y. Ontachi, T. Mizutani, M. Kato, et al., Pathophysiology of disseminated intravascular coagulation (DIC) progresses at a different rate in tissue factor-induced and lipopolysaccharide-induced DIC models in rats, Blood Coagul. Fibrinolysis 14 (2003) 221–228. [12] E. Challet, Minireview: entrainment of the suprachiasmatic clockwork in diurnal and nocturnal mammals, Endocrinology 148 (2007) 5648–5655. [13] E.J. Costa, C.S. Shida, M.H. Biaggi, A.S. Ito, M.T. Lamy-Freund, How melatonin interacts with lipid bilayers: a study by fluorescence and ESR spectroscopies, FEBS Lett. 416 (1997) 103–106. [14] S.M. Reppert, D.R. Weaver, T. Ebisawa, Cloning and characterization of a mammalian melatonin receptor that mediates reproductive and circadian responses, Neuron 13 (1994) 1177–1185. [15] S.M. Reppert, C. Godson, C.D. Mahle, D.R. Weaver, S.A. Slaugenhaupt, J.F. Gusella, Molecular characterization of a second melatonin receptor expressed in human retina and brain: the Mel1b melatonin receptor, Proc. Natl. Acad. Sci. U. S. A. 92 (1995) 8734–8738. [16] O. Nosjean, M. Ferro, F. Coge, P. Beauverger, J.M. Henlin, F. Lefoulon, et al., Identification of the melatonin-binding site MT3 as the quinone reductase 2, J. Biol. Chem. 275 (2000) 31311–31317. [17] V. Srinivasan, S.R. Pandi-Perumal, D.W. Spence, H. Kato, D.P. Cardinali, Melatonin in septic shock: some recent concepts, J. Crit. Care 25 (2010) 656 (e1-.e6). [18] E. Gitto, M. Karbownik, R.J. Reiter, D.X. Tan, S. Cuzzocrea, P. Chiurazzi, et al., Effects of melatonin treatment in septic newborns, Pediatr. Res. 50 (2001) 756–760. [19] P. Chrusciel, A. Goch, M. Banach, D.P. Mikhailidis, J. Rysz, J.H. Goch, Circadian changes in the hemostatic system in healthy men and patients with cardiovascular diseases, Med. Sci. Monit. 15 (2009) 203–208.

43

[20] G.H. Tofler, D. Brezinski, A.I. Schafer, C.A. Czeisler, J.D. Rutherford, S.N. Willich, et al., Concurrent morning increase in platelet aggregability and the risk of myocardial infarction and sudden cardiac death, N. Engl. J. Med. 316 (1987) 1514–1518. [21] A. Brzezinski, Melatonin in humans, N. Engl. J. Med. 336 (1997) 186–195. [22] P.H. Wirtz, M. Spillmann, C. Bartschi, U. Ehlert, R. von Kanel, Oral melatonin reduces blood coagulation activity: a placebo-controlled study in healthy young men, J. Pineal Res. 44 (2008) 127–133. [23] L.I. Kornblihtt, L. Finocchiaro, F.C. Molinas, Inhibitory effect of melatonin on platelet activation induced by collagen and arachidonic acid, J. Pineal Res. 14 (1993) 184–191. [24] C.M. Leach, G.D. Thorburn, A comparison of the inhibitory effects of melatonin and indomethacin on platelet aggregation and thromboxane release, Prostaglandins 20 (1980) 51–56. [25] F.J. Martin, G. Atienza, M. Aldegunde, J.M. Miguez, Melatonin effect on serotonin uptake and release in rat platelets: diurnal variation in responsiveness, Life Sci. 53 (1993) 1079–1087. [26] A. Valevski, I. Modai, Z. Jerushalmy, L. Kikinzon, A. Weizman, Effect of melatonin on active transport of serotonin into blood platelets, Psychiatry Res. 57 (1995) 193–196. [27] K. Effenberger-Neidnicht, M. Al Nator, L. Brencher, M. Oude Lansink, H. de Groot, Prevention of endotoxin-induced hypotension in rats by the circadian hormone melatonin, Infection 41 (2013) 11–12. [28] C.C. Wu, C.W. Chiao, G. Hsiao, A. Chen, M.H. Yen, Melatonin prevents endotoxininduced circulatory failure in rats, J. Pineal Res. 30 (2001) 147–156. [29] J.Y. Wu, M.Y. Tsou, T.H. Chen, S.J. Chen, C.M. Tsao, C.C. Wu, Therapeutic effects of melatonin on peritonitis-induced septic shock with multiple organ dysfunction syndrome in rats, J. Pineal Res. 45 (2008) 106–116. [30] J.-L. Soret, Analyse spectrale: Sur le spectre d'absorption du sang dans la partie violette et ultra-violette, C. R. 97 (1883) 1269–1273. [31] A. Pappenheim, Zur Blutzellfärbung im klinischen Bluttrockenpräparat und zur histologischen Schnittpräparatfärbung der hämatopoetischen Gewebe nach meinen Methoden, Folia Haematol. 13 (1912) 337–344. [32] D. Whiting, J.A. DiNardo, TEG and ROTEM: technology and clinical applications, Am. J. Hematol. 89 (2014) 228–232. [33] M.V. Wohlauer, E.E. Moore, J. Harr, E. Gonzalez, M. Fragoso, C.C. Silliman, A standardized technique for performing thromboelastography in rodents, Shock 36 (2011) 524–526. [34] H. Schochl, C. Solomon, A. Schulz, W. Voelckel, A. Hanke, M. Van Griensven, et al., Thromboelastometry (TEM) findings in disseminated intravascular coagulation in a pig model of endotoxinemia, Mol. Med. 17 (2011) 266–272. [35] M.C. Muller, J.C. Meijers, M.B. Vroom, N.P. Juffermans, Utility of thromboelastography and/or thromboelastometry in adults with sepsis: a systematic review, Crit. Care 18 (2014) 30. [36] S. Zeerleder, C.E. Hack, W.A. Wuillemin, Disseminated intravascular coagulation in sepsis, Chest 128 (2005) 2864–2875. [37] M.M. Del Zar, M. Martinuzzo, D.P. Cardinali, L.O. Carreras, M.I. Vacas, Diurnal variation in melatonin effect on adenosine triphosphate and serotonin release by human platelets, Acta Endocrinol. (Copenh) 123 (1990) 453–458. [38] M. Martinuzzo, M.M. Del Zar, D.P. Cardinali, L.O. Carreras, M.I. Vacas, Melatonin effect on arachidonic acid metabolism to cyclooxygenase derivatives in human platelets, J. Pineal Res. 11 (1991) 111–115. [39] K. Effenberger-Neidnicht, L. Brencher, M. Broecker-Preuss, T. Hamburger, F. Petrat, H. de Groot, Immune stimulation by exogenous melatonin during experimental endotoxemia, Inflammation 37 (2014) 738–744. [40] R. Kerr, D. Stirling, C.A. Ludlam, Interleukin 6 and haemostasis, Br. J. Haematol. 115 (2001) 3–12. [41] J.Y. Ye, E.Y. Liang, Y.S. Cheng, G.C. Chan, Y. Ding, F. Meng, et al., Serotonin enhances megakaryopoiesis and proplatelet formation via p-Erk1/2 and F-actin reorganization, Stem Cells 32 (2014) 2973–2982. [42] P. Lissoni, G. Tancini, S. Barni, F. Paolorossi, F. Rossini, P. Maffe, et al., The pineal hormone melatonin in hematology and its potential efficacy in the treatment of thrombocytopenia, Recenti Prog. Med. 87 (1996) 582–585. [43] P. Lissoni, M. Mandala, F. Rossini, L. Fumagalli, S. Barni, Growth factors: thrombopoietic property of the pineal hormone melatonin, Hematology 4 (1999) 335–343. [44] H. Wadenvik, J. Kutti, The spleen and pooling of blood cells, Eur. J. Haematol. 41 (1988) 1–5. [45] K. Freden, P. Lundborg, L. Vilen, J. Kutti, The peripheral platelet count in response to adrenergic alpha-and beta-1-receptor stimulation, Scand. J. Haematol. 21 (1978) 427–432. [46] Y. Ojiri, K. Noguchi, N. Shiroma, T. Matsuzaki, M. Sakanashi, M. Sakanashi, Uneven changes in circulating blood cell counts with adrenergic stimulation to the canine spleen, Clin. Exp. Pharmacol. Physiol. 29 (2002) 53–59. [47] H. Ishii, N. Tanaka, M. Kobayashi, M. Kato, Y. Sakuma, Gene structures, biochemical characterization and distribution of rat melatonin receptors, J. Physiol. Sci. 59 (2009) 37–47. [48] J.N. Thon, A. Montalvo, S. Patel-Hett, M.T. Devine, J.L. Richardson, A. Ehrlicher, et al., Cytoskeletal mechanics of proplatelet maturation and platelet release, J. Cell Biol. 191 (2010) 861–874. [49] M. Nagel, S. Brauckmann, F. Moegle-Hofacker, K. Effenberger-Neidnicht, M. Hartmann, H. de Groot, et al., Impact of bacterial endotoxin on the structure of DMPC membranes, Biochim. Biophys. Acta 2015 (1848) 2271–2276. [50] M.B. Yerer, S. Aydogan, H. Yapislar, O. Yalcin, O. Kuru, O.K. Baskurt, Melatonin increases glutathione peroxidase activity and deformability of erythrocytes in septic rats, J. Pineal Res. 35 (2003) 138–139. [51] M.B. Yerer, H. Yapislar, S. Aydogan, O. Yalcin, O. Baskurt, Lipid peroxidation and deformability of red blood cells in experimental sepsis in rats: the protective effects of melatonin, Clin. Hemorheol. Microcirc. 30 (2004) 77–82.