Abstract
Bleeding resulting from the application of low-molecular-weight heparins (LMWHs) may be treated with protamine sulfate, but this treatment lacks efficiency; its action against antifactor Xa activity is limited to ∼60%. Moreover, protamine sulfate can cause life-threatening hypersensitivity reactions. We developed diblock heparin-binding copolymer (HBC), which can neutralize the anticoagulant activity of parenteral anticoagulants. In the present study, we explored the safety profile of HBC and its potential to reverse enoxaparin, nadroparin, dalteparin, and tinzaparin in human plasma and at in vivo conditions. HBC-LMWH complexes were characterized using zeta potential, isothermal titration calorimetry, and dynamic light scattering. The rat cardiomyocytes and human endothelial cells were used for the assessment of in vitro toxicity. Male Wistar rats were observed for up to 4 days after HBC administration for clinical evaluation, gross necropsy, and biochemistry and histopathological analysis. Rats were treated with LMWHs alone or followed by short-time intravenous infusion of HBC, and bleeding time and antifactor Xa activity were measured. HBC completely reversed antifactor Xa activity prolonged in vitro by all LMWHs with an optimal weight ratio of 2.5:1. The complexes of HBC-LMWHs were below 5 µm. We observed no effects on the viability of cardiovascular cells treated with HBC at concentrations up to 0.05 mg/ml. Single doses up to 20 mg/kg of HBC were well tolerated by rats. HBC completely reversed the effects of LMWHs on bleeding time and antifactor Xa activity in vivo after 20 minutes and retained ∼80% and ∼60% of reversal activity after 1 and 2 hours, respectively. Well-documented efficacy and safety of HBC both in vitro and in vivo make this polymer a promising candidate for LMWHs reversal.
SIGNIFICANCE STATEMENT Over the last decade, there has been significant progress in developing antidotes for the reversal of anticoagulants. Until now, there has been no effective and safe treatment for patients with severe bleeding under low-molecular-weight heparin therapy. Based on our in vitro and in vivo studies, heparin-binding copolymer seems to be a promising candidate for neutralizing all clinically relevant low-molecular-weight heparins.
Introduction
Low-molecular-weight heparins (LMWHs) represent a group of parenteral anticoagulants for prevention and treatment of several thrombotic disorders, that is, unstable angina, myocardial infarction, and deep vein or cancer-associated thrombosis. The global parenteral anticoagulant market is dominated by LMWHs and is still growing. LMWHs offer numerous advantages over standard unfractionated heparin (UFH), including more predictable pharmacodynamics and pharmacokinetics, less restrictive monitoring, safer subcutaneous administration, and more favorable overall safety profile with reduced risk of nonhemorrhagic side effects (Kearon et al., 2012).
LMWHs are not a homogenous group of anticoagulants and may differ in their anticoagulant profiles, pharmacokinetic properties, and recommended dosing regimens (Kearon et al., 2012). LMWHs are obtained from UFH by different methods of depolymerization. The principal mechanism of the anticoagulant action of LMWHs is similar and involves antithrombin. All LMWHs catalyze antithrombin-mediated inactivation of coagulation factors. Because not all molecules of LMWHs contain 18 saccharide units that are necessary for factor IIa (thrombin) inhibition, LMWHs have antifactor Xa to antifactor IIa activity ratios between 1.6:1 and 8:1. Differences in antifactor Xa/antifactor IIa activity ratio between LMWHs depend on their average molecular weight and molecular size distribution (Gray et al., 2008). It was also shown in a cohort study of more than 12,500 patients that LMWHs differ in the frequency of bleeding episodes (van Rein et al., 2017). In the most clinical trials involving the management of thrombotic disorders with LMWHs, the incidence of major bleeding was between 1.5% and 4.7% and depended mainly on the type and duration of anticoagulation used and on the indication claimed (Simonneau et al., 1997; Hull et al., 2000; Turpie et al., 2002; Petersen et al., 2004; Mismetti et al., 2005; Crowther and Warkentin, 2008).
Severe bleeding related to LMWHs may be treated with intravenous administration of protamine sulfate. Protamine sulfate fully neutralizes LMWHs’ antifactor IIa activity, but its action against antifactor Xa activity is limited to ∼60% (Garcia et al., 2012). In the retrospective study, LMWH-induced active bleeding was effectively stopped in 66% of patients who received protamine (van Veen et al., 2011). Protamine sulfate can also cause life-threatening hypersensitivity reactions, such as hypotension, pulmonary hypertension, noncardiogenic pulmonary edema, cardiovascular collapse, and pulmonary vasoconstriction. Additionally, several risk factors for severe hypersensitivity reactions to protamine have been described, including rapid administration, previous administration of protamine or protamine insulin, fish allergy, left ventricular dysfunction, and previous vasectomy (Sokolowska et al., 2016). Therefore, each administration of protamine should be carefully considered, and better alternatives for LMWHs reversal are required.
Several new reversal agents for LMWHs (mainly enoxaparin and tinzaparin) are being developed. Ciraparantag is a cationic, synthetic small molecule designed as an antidote for direct oral anticoagulants, UFH, and enoxaparin (Laulicht et al., 2013). Andexanet alfa, a recombinant derivative of factor Xa, has been designed to reverse the anticoagulant effects of direct and indirect factor Xa inhibitors (Lu et al., 2013). However, andexanet alfa has received approval only for uncontrolled and life-threatening bleeding in patients treated with oral anticoagulants: apixaban and rivaroxaban (Heo, 2018). Another antidote currently in the preclinical phase of development is Universal Heparin Reversal Agent (UHRA). UHRA, a synthetic dendrimeric polymer, has been designed to reverse the anticoagulant activity of all available heparins (Shenoi et al., 2014; Kalathottukaren et al., 2017a) . In a recently published in vitro comparative study, UHRA directly bound both UFH and enoxaparin. Andexanet alfa and ciraparantag bound the above heparins relatively weakly. For andexanet alfa and UHRA, binding correlated positively with the neutralization activity. The authors did not observe the reversal activity of ciraparantag in assays typically used to assess anticoagulant effects (Kalathottukaren et al., 2018). None of the three above antidotes have been approved by the Food and Drug Administration for the management of LMWH-induced bleeding.
Therefore, to address the significant and still-unmet clinical need, we developed a heparin-binding polymer (HBC) [poly(ethylene glycol) (PEG) 41–poly(3-(methacryloylamino)propyl trimethylammonium chloride) (PMAPTAC) 53], a reversal agent for parenteral anticoagulants. HBC is a diblock synthetic polymer that directly binds anticoagulant molecules and effectively neutralizes their effects in the living organism (Kalaska et al., 2016, 2018, 2019). In this work, we further explore the potential of HBC to safely reverse LMWHs widely used in the clinical practice (i.e., enoxaparin, nadroparin, dalteparin, and tinzaparin) at in vivo conditions and in human plasma.
Materials and Methods
In Vitro Neutralization Assay.
Enoxaparin, nadroparin, dalteparin, and tinzaparin were purchased from Sanofi-Aventis (France), Aspen Notre Dame de Bondueville (France), Pfizer Europe MA EEIG (UK), and Sigma-Aldrich (Germany), respectively. The neutralization of the above LMWHs was analyzed by measuring antifactor Xa activity in a 96-well plate reader (Synergy HTX, BioTek) according to the modified kit manufacturer instructions (BioMedica Diagnostics, Canada), as described previously (Kalaska et al., 2016). In brief, sodium citrate–anticoagulated pooled plasma was obtained from healthy rats or human volunteers by centrifugation of whole blood at 3500 g for 20 minutes at 4°C. Two hundred microliters of pooled plasma were mixed with 10 µl of LMWHs (20 µg/ml of plasma for enoxaparin and tinzaparin; 2 U/ml for nadroparin and dalteparin). After 150 seconds of incubation (37°C), 10 µl of a solution containing increasing concentrations of HBC (0.01–0.1 mg/ml of plasma) was added. After another 150 seconds of incubation (37°C), antifactor Xa activity was measured at 405 nm. All the procedures involving healthy human volunteers were approved by the Local Ethics Committee of the Medical University of Bialystok (Permit No. R-I-002/193/2019).
Zeta Potential Measurements.
Zeta potential of all HBC-LMWH complexes in ratios based on the in vitro neutralization assay was measured using a Zetasizer Nano-ZS (Malvern Instruments, UK). The samples were prepared in PBS (pH 7.4), filtered through a Chromafil filter (0.45 μm), and measured at 25°C. The data were analyzed using the Malvern software.
Isothermal Titration Calorimetry Measurements.
All isothermal titration calorimetry (ITC) measurements were made using Malvern MICRO-CAL PEAQ-ITC. HBC was dissolved in PBS; its concentration was 12 g/l (0.001078 mol/l). LMWHs’ commercial solutions were also diluted in PBS to concentrations of about 125 U/ml or 1.25 mg/ml. The measurement temperature was 25°C, and the mixing speed was 750 rpm. The measuring chamber contained 270 µl of LMWH solution to which 19 portions of HBC solution, 2 µl each, were added.
The binding constant (Ka) between a molecule of HBC and a free binding site of a respective LMWH was calculated using the formula:
Dynamic Light Scattering Measurements.
The size of complexes formed by HBCs and LMWHs in ratios based on the in vitro neutralization was studied with the dynamic light scattering (DLS) technique. LMWHs were dissolved in PBS and bovine serum. To that mixture, the HBC solutions in PBS were added, and the formation of complexes was monitored by DLS using a Zetasizer Nano-ZS (Malvern Instruments).
Cell Lines Used and Cell Viability Assay.
Primary human umbilical vein endothelial cells (HUVECs) (American Type Culture Collection–1730) and nontransformed embryonic rat heart–derived cells [H9c2(2-1), American Type Culture Collection CRL-1446] were used as healthy control cell lines for assessment of the cardiovascular system toxicity of HBC. HUVECs were cultured in Medium 199 containing 20% fetal bovine serum with cell growth supplements at 37°C in the atmosphere containing 5% CO2 and at 95% humidity. All experiments were performed with mycoplasma-free cells. The ATPlite luminescence test (Perkin Elmer Life Sciences) was used to assess the proliferation and viability of HUVECs in vitro. Cells were seeded in 96-well plates, and after overnight incubation, the cells were treated with increasing concentrations of HBC for 72 hours. Cells treated with formulation buffer were used as a control. H9c(2-1) cells were cultured in Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum at 37°C in the atmosphere containing 5% CO2 and at 90% humidity. The experiment was carried out for 24 hours after adding HBC to cell media. To assess cell viability, crystal violet (CrV) assay was used (Popiołek et al., 2019). Briefly, cells were seeded into 24-well plates at the density of about 1.1 × 104 cells/cm2. After incubation, the medium was removed, and the cells were washed and fixed using 1 ml of 4% v/v formaldehyde/PBS and treated for 2 minutes with a CrV solution. Then, unbound CrV was removed by rinsing with water. After drying, the destaining solution was added to each well and left for 30 minutes. Finally, absorbance of the obtained solution at 540 nm was measured, which was proportional to the number of living cells.
Animals.
Male Wistar rats were obtained from the Centre of Experimental Medicine in the Medical University of Bialystok, grouped into cages as appropriate, and allowed to have ad libitum access to sterilized tap water and standard chow in specific pathogen-free conditions. All the procedures involving animals were approved by the Local Ethical Committee on Animal Testing (Permit Nos. for the safety study: 38/2017 and 85/2019; Permit Nos. for the efficacy study: 15/2013 and 60/2018) and conducted in accordance to the Directive 2010/63/EU of the European Parliament and the Council on the protection of animals, Animal Research: Reporting of In Vivo Experiments guidelines, and national laws.
Safety Study in Rats.
Thirty-five male Wistar rats were randomly assigned to five groups. HBC was administered into the tail vein of six rats at the initial dose of 5 mg/kg. The rats were observed continuously for the first 30 minutes after HBC administration for clinical signs of toxicity. Then they were kept under observation for up to 4 days after HBC administration (intermittently for 4 hours and then at 6, 24, 48, 72, and 96 hours after HBC administration). Body weights were recorded before HBC administration and before the gross necropsy on day 4. Following a minimum of 1 day without severe clinical findings, HBC was administered to a group of six rats at a dose of 10 mg/kg. These rats were observed for clinical signs of toxicity. This procedure was repeated in six rats treated with HBC at a dose of 20 mg/kg and 10 rats treated with HBC at a dose of 40 mg/kg. PBS was also administered to seven animals as a control group. On the 4th day, four rats treated with PBS and four rats treated with HBC at the dose of 40 mg/kg were anesthetized by an intraperitoneal injection of pentobarbital (45 mg/kg), blood was collected from the heart and centrifuged, and serum was used for the measurement of alanine aminotransferase and aspartate aminotransferase activities and total bilirubin, urea, and creatinine concentrations. Immediately after the blood collection, the whole liver, lungs, kidney, and thymus were removed for routine histological analysis using hematoxylin and eosin staining.
Efficacy Study in Rats.
In the first procedure, 56 male Wistar rats were randomly assigned to seven groups, anesthetized by an intraperitoneal injection of pentobarbital (45 mg/kg), and placed in a supine position on a heated operating table. The body weights ranged from 200 to 250 g just before anesthesia induction. Nadroparin (800 U/kg), dalteparin (800 U/kg), or tinzaparin (8 mg/kg) was administered into the right femoral vein 10 minutes before the measurement of bleeding time. HBC (20 mg/kg) was intravenously infused (5 minutes) after LMWH administration. Bleeding time was measured as described previously (Kalaska et al., 2015). If bleeding was still present at the end of the 480-second observation period, a value of 480 seconds was ascribed for the sake of statistical analysis. Twenty minutes after anticoagulant administration, antifactor Xa activity was analyzed with ELISA technique using a microplate reader (Synergy HTX, BioTek) according to the kit manufacturer instructions (BioMedica Diagnostics). If antifactor Xa activity was more than or equal to 1 U/mg, a value of 1 U/mg was ascribed for the sake of statistical analysis.
In the second procedure, 43 male Wistar rats were randomly divided into six groups and anesthetized in the same manner as described in the first procedure. Rats were weighted just before anesthesia induction and the body weights ranged from 190 to 220 g. Nadroparin and HBC were also administered in the same doses and manner as described in the first procedure. Bleeding time was measured 10 minutes before blood collection. Antifactor Xa activity was analyzed 1 or 2 hours after anticoagulant administration. Bleeding time and antifactor Xa activity were measured as described above.
Statistical Analysis.
In the in vivo study, n refers to the number of animals in each experimental group. For each test, the experimental unit was an individual animal. The data are shown as mean ± S.D. or median with lower and upper limits and analyzed with GraphPad Prism 6 software (GraphPad Prism 6 software) using the unpaired Student’s t test or Mann-Whitney test. P values less than 0.05 were considered significant.
Results
The Synthesis of HBC (PEG41-PMAPTAC53) by Controlled Radical Polymerization.
The synthesis of HBC by reversible-addition fragmentation chain transfer polymerization technique and selection strategy of the leading compound was previously described (Kalaska et al., 2016). Reversible-addition fragmentation chain transfer is one of the controlled/living radical polymerization processes, allowing synthesis of very well-defined polymers in terms of molecular weight and molecular weight dispersity. Briefly, a diblock copolymer with the degrees of polymerization of PEG and PMAPTAC blocks equal to 41 and 53, respectively, was synthesized (Fig. 1). The cationic PMAPTAC198 homopolymer was used as a control to assess the toxicity-reducing activity of the PEG block.
Chemical structure of HBC (PEG41-PMAPTAC53) (Kalaska et al., 2016).
Cell Viability Reduction.
HUVECs and cardiomyocytes were treated with the diblock HBC and a cationic PMAPTAC198 homopolymer. HUVEC viability was not negatively affected by HBC and PMAPTAC198 at concentrations up to 0.05 mg/ml. Both studied polymers significantly impaired HUVEC viability at concentrations of 0.15 mg/ml (Fig. 2A). The conjugation of PMAPTAC with PEG significantly reduced the negative effect of the polycation PMAPTAC block on cardiomyocyte viability (Fig. 2B). At 0.15 mg/ml, the maximum concentration studied, the viability of the cardiomyocytes treated with HBC was still about 100%, whereas the viability of those treated with PMAPTAC198 decreased to about 50%.
Effect of HBC and PMAPTAC198 on HUVECs (A) and cardiomyocyte viability (B). Results are shown as mean ± S.D.
Efficacy in the In Vitro Study.
HBC concentration-dependently and completely reversed antifactor Xa activity in vitro of enoxaparin, nadroparin, dalteparin, and tinzaparin in rat and human sodium citrate–anticoagulated plasma with an optimal ratio of 0.05 mg of the polymer for 0.02 mg of both enoxaparin and tinzaparin and 2 U of both dalteparin and nadroparin (Fig. 3). We chose the concentration and dose of HBC for further in vitro and in vivo studies based on the estimated ratios: 2.5 mg of HBC per 1 mg of enoxaparin and tinzaparin and 2.5 mg of HBC per 100 U of dalteparin and nadroparin.
Neutralization of enoxaparin (A), nadroparin (B), dalteparin (C), and tinzaparin (D) by HBC in sodium citrate–anticoagulated plasma with the antifactor Xa assay. Results are shown as mean ± S.D.
Zeta Potential of HBC-LMWH Complexes.
Based on the effective neutralization ratios found, we next investigated the zeta potential of the HBC-LMWH complexes. In all cases, the zeta potential of complexes had a value close to zero (Table 1). The polycation completely neutralized the negative charge of LMWH molecules. Such a result was to be expected when the proportion of polymers in the mixture corresponded to the total binding of LMWH by polycation. DLS measurement showed a relatively large diameter of complexes due to potential values close to zero and the lack of electrostatic stabilization of the particles. In the case of tinzaparin, the complexes were the smallest.
Zeta potential of LMWH-HBC complexes and thermodynamic parameters for the interaction of LMWHs with HBC
Calorimetric Measurements and Thermodynamic Parameters of the Complexation Reaction.
Thermodynamic parameters of the interactions between all studied LMWHs and HBC are given in Table 1, and the integral enthalpies of interaction are shown in Fig. 4. For enoxaparin and nadroparin, the drop inflection points in the respective curves are less clearly defined than for dalteparin and tinzaparin (Fig. 4, A and B vs. Fig. 4, C and D), which may be due to higher dispersity of the molecular weight of the former LMWHs.
Estimated heat of binding for the titration of HBC into enoxaparin (A), nadroparin (B), dalteparin (C), and tinzaparin (D).
The Size of LMWH-HBC Complexes.
The size of complexes formed by LMWHs and HBCs in PBS and bovine serum was studied using DLS technique. The particle size of complexes in PBS and serum have diameters increasing in the following order: tinzaparin (120 and 220 nm), dalteparin (0.5 and 1 µm), enoxaparin (0.8 and 5 µm), and nadroparin (in both cases, two populations 0.4/>5 and 0.2/>5 µm). The presence of the polycation itself in the serum only slightly affected the size of the objects present in the mixture, and this effect can be considered negligible. The particle size of the complex formed in the serum was, in all cases, smaller than the diameter of the capillary blood vessels. A relatively large dimension equal to a fraction of the size of blood cells will promote the rapid elimination of complexes from the blood system (Fig. 5).
The size profile of enoxaparin-HBC (A), nadroparin-HBC (B), dalteparin-HBC (C), and tinzaparin-HBC (D) complexes in the aqueous solutions (PBS, pH 7.4) and bovine serum, and PBS and HBC (E) in bovine serum obtained by the dynamic light scattering analysis.
Safety in Animals.
There were no significant HBC-related effects on the body weight gain in rats treated for up to 4 days at the doses of 5, 10, and 20 mg/kg. A 5% reduction in body weight occurred in the group treated with HBC at the dose of 40 mg/kg. Administration of HBC at the dose of 20 mg/kg resulted in slightly decreased locomotive activity and apathy at approximately 3 hours postdose in 50% of rats. We observed transient mild nasal swelling and edema of the limbs in five rats. All rats treated with HBC at the dose of 40 mg/kg had decreased locomotive activity and apathy (both 10–15 minutes following HBC administration); 40% of rats had an increased thirst (14–20 minutes following HBC administration); 50% of rats had an increased secretion of porphyrins (2 to 3 hours following HBC administration). All these findings were resolved within approximately 24 hours postdose. Furthermore, mild nasal swelling and edema of the limbs were observed in 40% of rats treated with HBC at the dose of 40 mg/kg. These findings appeared immediately postdose and were resolved within approximately 6 hours. HBC at the dose of 40 mg/kg resulted in kidney congestion and color heterogeneity in 40% of rats. Gross necropsy findings in 30% of rats included the presence of urinary white sediment. Severe gross necropsy findings occurred in one rat and included lung congestion, kidney congestion, color heterogeneity, thymus congestion, and liver abnormalities. All rats survived the experiment (Table 2). We found higher serum concentrations of urea and creatinine in two out of four rats treated with HBC at the dose of 40 mg/kg (Table 3). The same two rats presented significant histopathological changes of acute tubular necrosis in kidney tissue, whereas two others presented mild tubular toxicity, which should not impair renal function. There were small areas of slight congestion and inflammatory cell infiltration in the lungs. HBC did not cause significant histopathological changes in other studied tissues (Fig. 6).
Selected clinical and gross necropsy observations during safety study in rats
The values indicate a total number of observations/number of animals concerned.
Biochemical parameters in rats treated with HBC at the dose of 40 mg/kg
Results are shown as a median with lower and upper limits.
Representative micrographs of tissue sections of liver (A and B), lung (C and D), renal cortex (E and F), renal medulla (G and H), and thymus (I and J) 4 days after intravenous administration of PBS (left column; A, C, E, G, and I) and HBC at the dose of 40 mg/kg (right column; B, D, F, H, and J) in rats. Original magnification, 100×; hematoxylin and eosin staining.
Efficacy in Animals.
All currently in vivo–studied LMWHs (tinzaparin, dalteparin, and nadroparin) prolonged bleeding time and increased antifactor Xa activity at all studied time points. HBC completely reversed the effects of LMWHs on bleeding time and antifactor Xa activity to the approximate control values in a 20-minute window (Fig. 7). Our previous study demonstrated that HBC also reversed an increased bleeding time and antifactor Xa activity by enoxaparin in the same animal model (Kalaska et al., 2016). We observed an almost-complete reversal of bleeding time, and antifactor Xa activity increased by nadroparin 1 hour from the administration, whereas more than 60% of reversal activity was maintained after 2 hours (Fig. 8).
Neutralization of nadroparin, dalteparin, and tinzaparin by HBC in rats as measured by bleeding time (A) and antifactor Xa activity (B). ***P < 0.001 vs. vehicle; ###P < 0.001 vs. appropriate low-molecular-weight heparin, unpaired Student’s t test. Results are shown as mean ± S.D. and analyzed with GraphPad Prism software 6.
Time course of reversal of nadroparin by HBC in rats as measured by bleeding time (A and B) and antifactor Xa activity (C and D) 1 or 2 hours after nadroparin administration. ***P < 0.001 vs. vehicle; #P < 0.05; ###P < 0.001 vs. nadroparin, unpaired Student’s t test. Results are shown as mean ± S.D. and analyzed with GraphPad Prism software 6.
Discussion
We have previously shown the ability of HBC to neutralize UFH, enoxaparin, and fondaparinux. The complete reversal of enoxaparin demonstrated the potential advantage of HBC over protamine sulfate (Kalaska et al., 2016). In this work, we evaluate the safety profile of HBC and its reversal activity against dalteparin, nadroparin, and tinzaparin. We found that HBC bound all tested LMWHs forming the smallest complexes with dalteparin and tinzaparin. The efficacy and safety studies in rats showed that HBC could be a universal antidote for all LMWHs. We also confirmed the neutralization of all LMWHs by HBC in human plasma.
Our in vitro studies investigated the cardiomyocyte viability after exposure to HBC and PMAPTAC198 as a control to check the cytotoxicity reduction of the block copolymer compared with homopolymer. Reversal of heparins by cationic protamine sulfate can cause right ventricular failure and bradycardia (Ocal et al., 2005; Sokolowska et al., 2016) and may induce left ventricular dysfunction because of its direct effect on cardiomyocyte contractile processes (Hird et al., 1995). Because of the negative mitochondrial membrane potential in cardiomyocytes, cationic compounds may be taken up by mitochondria and may, therefore, lead to cardiomyocyte damage (Hasinoff et al., 2003, 2016). Indeed, cationic PMAPTAC198 homopolymer decreased cardiomyocyte viability, whereas HBC had no significant effect on cell viability, confirming a well-known strategy of copolymerization or functionalization with PEG to reduce polycation toxicity (Pasut and Veronese, 2012). Though HBC and PMAPTAC198 did not affect human endothelial cells viability in vitro at concentrations up to 0.05 mg/ml, endothelial cells seem to be more sensitive than cardiomyocytes to the effects of both polycationic compounds applied in the higher concentration. The reduced viability may result from their interaction with the anionic surface of endothelial cells, as was shown in the case of protamine sulfate (Chang and Voelkel, 1989).
HBC sufficiently reversed antifactor Xa activity increased in vitro by studied LMWHs in the concentration of 0.025 mg/ml. In contrast, protamine sulfate was partially effective and showed an anticoagulant effect at concentrations higher than 0.05 mg/ml (Shenoi et al., 2014). The efficacious concentrations of HBC were three to six times lower than those maximums studied in the cardiomyocyte viability experiment. UHRAs demonstrated complete tinzaparin neutralization at similar ratios (Shenoi et al., 2014). Andexanet alfa in the concentration of ∼3 µM also completely reversed enoxaparin (Lu et al., 2013). However, the authors focused on one LMWH (tinzaparin or enoxaparin) and have not investigated the cardiotoxicity.
We applied DLS and ITC techniques to understand the mechanism of interaction between HBC and LMWHs. The enthalpy of binding was positive, indicating an endothermic interaction. It was a bit unexpected for electrostatic interaction. However, endothermic interactions between oppositely charged polyelectrolytes have been reported (Ma et al., 2009). On the other hand, entropy values were positive, indicating that the neutralization process is entropy driven. The values of 
vary from 0.38 to 0.69, indicating that one chain of HBC may bind more than one molecule of an LMWH (from 1.4 to 2.6 on the average). This is not surprising, taking into account that one HBC chain bears on the average 53 cationic groups, whereas the chains of the LMWHs bear from about 22 to 32 negative sulfate groups (assuming they all have the same charge density equal to 2.7 sulfate groups per disaccharide repeat unit, i.e., a charge of −5/kDa) (Kalathottukaren et al., 2017a). It means that the number of HBC cationic groups per chain is about 1.6–2.5 times greater than that of LMWHs anionic groups, which is close to 1/
, that is, the number of LMWH chains complexed by one HBC chain. Moreover, the value of 
for HBC-enoxaparin can be compared with the respective value for dendrimeric UHRA, for which the 
value varies from 0.17 to 0.28 (Shenoi et al., 2014), and indicates that one UHRA dendrimer may bind from 3.6 to 5.9 chains of enoxaparin. It is about two times more than the number of enoxaparin chains complexed by one HBC molecule. Considering that the molecular weight of UHRA is >23 kDa (i.e., almost twice that of HBC), the ability to bind enoxaparin of UHRA and HBC is comparable in terms of enoxaparin chains bound per kiloDalton of a polycation. The binding constants for HBC-LMWH complexes are of the order of 107–108, so the binding is rather strong also when compared with UHRA-enoxaparin binding for which Ka was of the order of 104–105 (Shenoi et al., 2014).
The development of a reversal agent for LMWHs is a multistep process involving an evaluation of its safety. The nonclinical safety evaluation mainly includes a characterization of toxic effects, gross necropsy observations, dose dependence, and relationship to exposure. According to International Conference on Harmonisation guidelines, it should support first-in-human clinical trials . Nonclinical toxicity studies of the most promising LMWH-reversal agents, ciraparantag and andexanet alfa, have been conducted in different animal models. Ciraparantag was evaluated to perform the core safety pharmacology studies in rat and dog models. Based on the lack of adverse systemic toxicity and mortality, doses of ciraparantag up to 35 mg/kg in dogs and 40 mg/kg in rats were chosen as maximum tolerated doses. Ciraparantag in the single dose of 40 mg/kg was well tolerated in dogs. There were mild signs of hypersalivation and loss of balance. In rats, ciraparantag in a single dose of 40 mg/kg caused decreased activity and red-stained nares (Sullivan et al., 2015). Similar treatment-emergent adverse events were observed in healthy volunteers later (Ansell et al., 2016). The authors have not investigated LMWH-reversal effects of ciraparantag in humans. Andexanet alfa was tested in a single and repeat-dose toxicity study. It was administered to rats or monkeys alone and in combination with apixaban, rivaroxaban, or enoxaparin. There were no systemic or tissue pathologies in studied animals after single intravenous injections of up to 60 mg/kg. In registration data available at Food and Drug Administration, five mortalities of rats treated with andexanet alfa twice daily for 14 days were observed. In monkeys, they found neither significant safety concerns nor unexpected outcomes. All observations were transient or reversible and included transient hypoactivity, labored respiration, vomiting, and anaphylactic reaction (Lu et al., 2017). Similarly to other antidotes (Kalathottukaren et al., 2017b), we tested the toxicity of HBC administered without heparin. Besides mild and occasional clinical and gross necropsy findings, single doses up to 20 mg/kg of HBC were well tolerated. We observed changes in rats’ behavior and signs of organ damage in the group of rats treated with a dose of 40 mg/kg of HBC. In some of the rats, we performed additional biochemistry and histopathological analysis, which revealed a sign of acute tubular necrosis and increased biochemical markers of kidney damage in two cases. Nephrotoxicity may result from the binding of cationic HBC with an anionic surface of renal epithelial cells, similarly to the postulated chemical interactions between crystals and epithelial cell surfaces in kidney stone formation (Lieske et al., 1997; Wesson and Ward, 2006). Similarly to protamine, the efficacious dose of HBC for LMWHs reversal is closely related to the dose of studied LMWH and is based on the estimated optimal mass ratios. In the previous study (Kalaska et al., 2016), we used HBC at the dose of 6 mg/kg that corresponded to 3 mg/kg of enoxaparin. If we assume 3 mg/kg as a therapeutic dose of enoxaparin in rats, the multiple to the no-observed-adverse-effect-level and adverse level for HBC would be ∼4 and ∼8, respectively. LMWHs were administered in supratherapeutic doses to induce clinically relevant major bleeding. Indeed, anticoagulation by LMWHs increased bleeding time and antifactor Xa activity. The strong anticoagulant effect was observed in 100% of the animals. HBC administered in dose corresponding to 2.5:1 mass ratio, but not exceeding safety dose of 20 mg/kg, completely reversed the effects of LMWHs on bleeding time and antifactor Xa activity to the approximate control values in a 20-minute window. A similar study has been conducted to determine the reversal effect of andexanet alfa on enoxaparin. However, administration of andexanet alfa only partially reversed antifactor Xa activity and decreased bleeding in ∼80% of the animals in a rat tail–transection model (Hollenbach et al., 2012). We further evaluated time-related changes of the reversal activity of HBC against nadroparin as an example, since HBC effect was comparable for all studied LMWHs. The almost-complete neutralization of bleeding time and antifactor Xa activity was observed for 1 hour after a 5-minute infusion of HBC, whereas more than 60% of its reversal activity was maintained after 2 hours. Based on the literature data (Ibrahim et al., 2017), the inhibition of factor Xa activity by LMWH expires after ∼6 hours from intravenous administration into rats. Thus, if a reversal of anticoagulation longer than 2 hours is needed for surgical intervention or during uncontrolled bleeding, HBC should probably be administered second time or, similarly to andexanet alfa, should be kept infused till the end of surgery or bleeding. Andexanet alfa due to the short half-life (∼1 hour) had to be infused additionally to nonbleeding healthy volunteers to reverse activity of enoxaparin (Crowther et al., 2014, Siegal et al., 2015). In the ANNEXA-4 trial, andexanet alfa was administered as an intravenous bolus followed by a 2-hour infusion to 352 patients who had major acute bleeding after administration of rivaroxaban, apixaban, or enoxaparin. Of the 15 patients treated with enoxaparin who could be evaluated for anticoagulation efficacy, 13 (87%) were adjudicated as having at least good hemostatic efficacy 12 hours after andexanet alfa infusion (Connolly et al., 2019). Thus, a 2-hour interval seems to be a convenient therapeutic window for the reversing of LMWHs in patients, especially those with a high risk of thrombosis, in which restoring of anticoagulant protection is needed. In the case of LMWHs reversal, more clinical trials are required to assess the effects of andexanet alfa on this drug-related bleeding (Sartori and Cosmi, 2018). Ciraparantag, another candidate for LMWHs reversal, was evaluated in rats and healthy volunteers. It stopped tail bleeding in rats treated with enoxaparin, and the effect was stronger than that of protamine sulfate (Laulicht et al., 2013). In healthy volunteers, ciraparantag entirely and rapidly reversed the clotting time induced by enoxaparin but also caused adverse events in more than half of patients (Ansell et al., 2016).
Based on the previous studies on enoxaparin, one could assume that other LMWHs would also be neutralized by HBC. However, it would be impossible to conclude that HBC is an efficient and safe reversal agent for all main LMWHs without clear direct experimental evidence, especially because other antidotes were less effective against heparins with shorter chains. We also expanded the panel of safety studies in rodents and the toxicity in cardiomyocytes and characterized in more detail the mechanism of binding. We confirmed the efficacy of HBC against LMWHs in human plasma and its safety in HUVECs. It seems that the reversal activity of HBC after a single injection could last longer than the effect of andexanet alfa, which needs to be infused to maintain activity. Nevertheless, the second injection or prolonged infusion of HBC may be required in some clinical scenarios depending on how much time has passed since LMWHs injection and how much time is needed for surgical intervention. Thus, the evaluation of repeated-dose toxicity together with the estimation of the optimal route of HBC administration in nonhuman primates is required before the administration of HBC to patients.
To sum up, we present here a new pharmacological tool for binding all heparins. The studies with protamine, andexanet alfa, and ciraparantag show that there is still space for a more efficient, safer, and cost-effective reversal agent indicated for LMWHs, and yet no such antidote has been currently registered. The results of safety experiments point to a therapeutic potential of HBC administered in doses up to 20 mg/kg. We estimated the neutralization weight ratio of HBC to LMWH for 2.5:1 in rats. Then, we confirmed it in human plasma. Importantly, HBC stopped bleeding from tail of heparinized rats and brought to control level typical coagulation parameters used in the clinical practice. HBC retained ∼80% and ∼60% of reversal activity after 1 and 2 hours, respectively. Based on this study, we may expect similar effect of HBC in patients treated with LMWHs.
Acknowledgments
We would like to thank Marcin Szczeciński, Piotr Rózga, and Marcin Feder from Adamed Pharma S.A. (Poland) for the critical review of the results and performing experiment in HUVECs.
Authorship Contributions
Participated in research design: Kalaska, Szczubiałka, Mogielnicki.
Conducted experiments: Kalaska, Miklosz, Kamiński, Swieton, Jakimczuk, Yusa, Mogielnicki.
Contributed new reagents or analytic tools: Kalaska, Miklosz, Kamiński, Swieton, Yusa, Szczubiałka, Mogielnicki.
Performed data analysis: Kalaska, Kamiński, Swieton, Mogielnicki.
Wrote or contributed to the writing of the manuscript: Kalaska, Miklosz, Kamiński, Swieton, Jakimczuk, Yusa, Pawlak, Nowakowska, Szczubiałka, Mogielnicki.
Footnotes
- Received September 30, 2019.
- Accepted January 9, 2020.
This work was funded by National Science Centre, Poland Grant 2016/21/B/ST5/00837. The efficacy studies in human plasma were supported by Medical University of Bialystok project no. SUB/2/DN/19/003/2211.
Abbreviations
- CrV
- crystal violet
- DLS
- dynamic light scattering
- HBC
- heparin-binding copolymer
- HUVEC
- human umbilical vein endothelial cell
- ITC
- isothermal titration calorimetry
- LMWH
- low-molecular-weight heparin
- PEG
- poly(ethylene glycol)
- PMAPTAC
- poly(3-(methacryloylamino)propyl trimethylammonium chloride)
- UFH
- unfractionated heparin
- UHRA
- Universal Heparin Reversal Agent
- Copyright © 2020 by The Author(s)
This is an open access article distributed under the CC BY-NC Attribution 4.0 International license.
