Molecular design of orally active heparin conjugates for the treatment of thrombosis and tumor angiogenesis

Abstract

Oral delivery of therapeutics is extremely challenging. The digestive system is designed in a way that naturally allows the degradation of proteins or peptides into small molecules prior to absorption. For systemic absorption, the intact drug molecules must traverse the impending harsh gastrointestinal environment. Technologies, such as enteric coating, with oral dosage formulation strategies have successfully provided the protection of non-peptide based therapeutics against the harsh, acidic condition of the stomach. However, these technologies showed limited success on the protection of therapeutic proteins and peptides. Importantly, inherent permeability coefficient of the therapeutics is still a major problem that has remained unresolved for decades. Addressing this issue in the context, we utilized the strategies that are developed in enhancing the intestinal permeability of a drug molecule either by modifying the intestinal epithelium or by modifying the drug itself. These modifications have been pursued by using a group of molecules that can be conjugated to the drug molecule to alter the cell permeability of the drug or mixed with the drug molecule to alter the epithelial barrier function, in order to achieve the effective drug permeation. The knowledge of membrane transporters, receptors, and their specific ligands helped scientists to explore the cellular recognition based oral delivery systems by covalent attachment of a drug. Receptor mediated endocytosis proved to be independent with regard to the size of the transported drug molecules. However, the receptor mediated endocytosis pathway potentially involves lysosomal degradation, to which enzyme activity is highly susceptible. On the other side, transporting molecules that are prototypically variants, the escape from lysosomal degradation can be attained by utilizing gradient dependent apical transporters

for example conjugation of a small molecule to bile acid. Bile acid uptake is concerted by the apical sodium-dependent bile acid transporter (ASBT) system. The transport mechanism of intestinal bile acid transporters are followed by a sodium-dependent carrier mediated facilitated diffusion that does not involve active endocytosis. The utilization of membrane-based transporters, as a strategy, is thus mainly confined to the transport of only small molecular substrates because of their size-limited uptake mechanism. The field of active delivery of macromolecules, particularly via transporters, remains largely unexplored. To address this issue, we selected low molecular weight heparin (LMWH) as a model macromolecular drug because it has the potential to treat various diseases such as, deep vein thrombosis (DVT), cancer-associated thrombosis, and arthritis, but is not absorbed in the gastrointestinal (GI) tract. Here, we describe the development of ASBT-targeted high-affinity oligomeric bile acid substrates that mediate the transmembrane transport of LMWH. Several oligomers of deoxycholic acid (oligoDOCA) were synthesized to investigate the substrate specificity of ASBT. To see the binding of oligoDOCA on the substrate-binding pocket of ASBT, molecular docking was used and the dissociation rate constants (KD) were measured using surface plasmon resonance. The KD for tetrameric DOCA (tetraDOCA) was 50-fold lower than that for monomeric DOCA, because tetraDOCA interacted with several hydrophobic grooves in the substrate-binding pocket of ASBT. The synthesized oligoDOCA compounds were subsequently chemically conjugated to macromolecular LMWH. In vitro, tetraDOCA-conjugated LMWH (LHe-tetraD) had highest selectivity for ASBT during its transport. Orally administered LHe-tetraD showed remarkable systemic anticoagulation activity and high oral bioavailability of 33.5 ± 3.2% and 19.9 ± 2.5% in rats and monkeys, respectively. ASBTs are the intestinal transporters that form intermediate complexes with substrates and its conformational change drives the movement of small molecular substrates across the cell membrane. Here, we propose new biological insights on the role of ASBT that uses high-affinity binding macromolecular substrates to functionally transform the membrane transporters so that they behave like receptors, ultimately allowing the apical-basal transport of bound macromolecules. The optimized bile acid-based macromolecular substrate, LHe-tetraD, had high affinity (KD = 0.072 µM) that endowed strong interactions with ASBT. Using co-immunoprecipitation and in situ proximity ligation assay, we observed the physical interactions between LHe-tetraD and ASBT both in the apical and cytoplasmic fractions of ASBT-transfected epithelial cells. ASBT/LHe-tetraD complexes were rapidly internalized in vesicles, localized in early endosomes, dissociated and escaped the vesicular transport. While binding of cytoplasmic ileal bile acid binding proteins cause exocytosis of macromolecules and prevented entry into lysosomes. This newly found transformation process of ASBT suggests a new transport mechanism that could aid in further utilization of ASBT to mediate oral macromolecular drug delivery. The new transport mechanism of ASBT observed here might possess tremendous clinical importance, because the use of heparin or its derivatives as extended therapy, for indications such as thrombosis prophylaxis, demands non-parenteral delivery methods. Thrombogenesis is a major cause of morbidity and mortality in cancer patients. Prophylaxis with LMWH is recommended for cancer patients, but requires non-invasive delivery methods for long-term treatments. When LHe-tetraD was orally administered at a dose of 5 mg/kg in ICR mice, the maximum anti-factor Xa level was increased up to 0.62 ± 0.05 IU/mL without any evidence of liver toxicity, gastrointestinal damage, or thrombocytopenia. LHe-tetraD successfully prevented thrombosis in a rat model of deep vein thrombosis and heat-induced model of cancer-associated thrombosis (CAT) in mice. The CAT was induced in tumor-bearing mice by local heat application, and the fibrin deposition in tumors was evaluated. The oral administration of LHe-tetraD (either a single dose or multiple daily doses for up to 10 days) in mice substantially abolished the coagulation-dependent tropism of fibrinogen in the heated tumors and significantly decreased hemorrhage, compared to the mice treated with saline or subcutaneous injection of LMWH. Thus, the anticoagulation effect of oral LHe-tetraD invokes the benefits of oral delivery and promises to provide an effective and convenient treatment for cancer patients at risk of thrombosis. Overall, these results represent a major advancement in ASBT-mediated LMWH delivery and may facilitate administration of many important therapeutic macromolecules through a non-invasive oral route. In quest of novel drug development opportunities for heparin, a non-invasive oral delivery approach also represents a starting point. Heparin is one of the most abundant biomolecules in nature. Besides their uses as structural materials and anticoagulants, they are to large extents mediating recognition events through their interactions with proteins and other biological entities. Growth factors (GF) interact with extracellular matrix (ECM) biomolecules, such as heparin sulfate (HS) glycosaminoglycan, to enhance their stability and angiogenic signaling in many disease forms, such as cancer. Biomaterials that modulate GF activity by mimicking interactions observed in the native ECM could be designed as an effective treatment strategy to attenuate angiogenic signaling. Basic residues critical for HS binding have been identified on vascular endothelial growth factor (VEGF)

however it could only provide a low intrinsic affinity. Conjugation of hydrophobic bile acids on heparin structure might offers substantial increases in avidity for VEGF, the most prominent angiogenic GF, while retaining the similar surface charge distributions for binding. LMWH conjugated with several monomeric and oligomeric deoxycholic acids was developed. For most mono- and oligo-DOCA LMWH conjugates, the antitumor activity was improved proportional to the degree of DOCA conjugation, but non-proportional to their anticoagulant properties. In vivo, dimeric DOCA-conjugated LMWH, namely LHbisD4, had the highest absorption after oral delivery and antitumor activity. The conjugate attenuated VEGF signaling in vitro and inhibited angiogenesis in a spheroid-based model of human vasculature in mice. Regulating growth factor activities in a controlled and site-specific manner is the main hurdle that current antiangiogenic treatment modalities face. Here, we take the advantage of the differential expression patterns of tumoral endothelial cells (TEC) to achieve a selective tumor targeting strategy in biomaterials design. The prion-like protein, doppel, was identified as a TEC-, but not to normal endothelial cells (EC)-, specific surface marker of uncertain function. We found that among the various tumoral endothelial markers identified, doppel expression was most abundant in the vessels and isolated TEC of different tumors types. We present evidence that doppel remained co-localize and formed complexes with the vascular endothelial growth factor receptor 2 (VEGFR2) on the surface of intact TEC. Surprisingly, doppel antibody treatment modulated the cell surface residency of doppel and VEGFR2 and induced their internalization. The internalization of complexes containing doppel and VEGFR2 markedly decreased the binding of VEGF165 to VEGFR2. Thus, doppel antibody effectively inhibited VEGF-induced receptor signaling and the extent of angiogenic sprouting of TEC. The interaction of doppel and VEGFR2 can be viewed as a therapeutic target to modulate VEGF signaling on TECs. This was accomplished by using heparin (an ECM component) or its bile acid-based conjugates, LHbisD4 that selectively bind to doppel, localized to TEC than TEC-/-dpl, and demonstrated high tumor accumulation after oral administration. In addition to doppel-VEGFR2 complex internalization, LHbisD4 also regulated angiogenic signaling via binding with doppel that allowed sequential capturing of VEGF from tumor interstitial to inhibit tumor growth in vivo. This strategy defines the importance of doppel as a therapeutic target and provides a rational for the continued development of doppel-targeted therapies. We also show that biomaterials (LMWH-oligoDOCA) engineered to have tissue and molecular targeting abilities (doppel binding and GF capturing) via bio-inspired interactions could be used as a new class of therapeutics in tumor treatment.