Characterization of molecular interactions between cannabidiol and human plasma proteins (serum albumin and γ-globulin) by surface plasmon resonance, microcalorimetry, and molecular docking

Highlights

• Direct binding between CBD and HSA and γ-globulin was measured by SPR.

• Thermodynamic binding parameters were determined by ITC.

• Thermal stability of CBD and HSA and γ-globulin binding was analyzed by DSC.

• CBD and warfarin bind to allosteric sites on HSA in a competitive SPR assay.

Abstract

A cannabidiol (CBD) oral solution (Epidiolex®) has been approved by the United States Food and Drug Administration to treat seizure conditions. However, the biomedical and pharmaceutical applications of CBD are hindered partially due to a limited understanding of CBD’s pharmacokinetic behaviors, such as its interactions with plasma proteins. Herein, we investigated the molecular interactions between CBD and two plasma proteins, namely, human serum albumin (HSA) and γ-globulin, using biophysical techniques including surface plasmon resonance (SPR), isothermal titration calorimetry, and differential scanning calorimetry, as well as molecular docking. CBD bound to HSA and γ-globulin in an exothermic manner (enthalpy: −9.3 ×10⁴ and −3.7 ×10⁴ kcal/mol, respectively) with a binding affinity of 1.8 × 10⁻⁵ and 1.3 × 10⁻⁵ M, respectively. The binding ratio between CBD and HSA or γ-globulin was approximately 1:1 and 3:1, respectively. Furthermore, computational modeling suggested that CBD and warfarin may bind to HSA independently, supported by data from a competitive SPR binding assay. Findings from the current study elucidate CBD’s plasma protein binding characteristics and shed light on their impact on CBD’s pharmacokinetic property.

Introduction

Cannabidiol (CBD) is a Cannabis derived pharmaceutical ingredient with promising pharmacological effects. CBD’s therapeutic effects on various diseases, including cancers, neurodegenerative diseases, epilepsy and anxiety conditions, and autoimmune diseases are supported by reported preclinical studies and human clinical trials [1], [2], [3]. Several clinical trials are currently underway to assess CBD’s safety and efficacy for the treatment of conditions such as schizophrenia, anxiety disorder, Crohn's disease, and Parkinson's disease. However, by far, only one CBD-based drug (an oral solution of CBD; Epidiolex® by GW Pharmaceuticals) has successfully received approval from the United States Food and Drug Administration (FDA) to treat patients with seizures conditions. Limited pharmaceutical and biomedical applications of CBD are, at least partially, due to the lack of comprehensive understanding of CBD’s pharmacokinetic (PK) and pharmacodynamics (PD) behaviors, including its plasma protein binding (PPB) capacity [4], [5].

PPB is a critical determinant for the development of active pharmaceutical ingredients. The portion of a drug that binds to plasma proteins is not bioavailable, and only the drug in its free state (fraction of unbound drug) can pass across biological membranes and become pharmacologically active. Therefore, a drug’s PK parameters, such as the apparent volume of distribution, clearance rate, and half-life, can be affected mainly by its propensity to plasma proteins. The most abundant plasma protein is human serum albumin (HSA), which constitutes up to 4.5% of the weight of human blood [6]. HSA has a set of eight hydrophobic binding sites, which enable it to bind to a wide range of endogenous substances (e.g. hormones and fatty acids) and xenobiotics (e.g. drugs) with varying affinities [7]. Notably, two binding pockets on HSA (i.e. Sudlow site I and II) tend to bind neutral and negatively charged hydrophobic drugs, such as warfarin (3-(α-acetonylbenzyl)− 4-hydroxycoumarin) [8]. These two binding sites are also referred to as ‘warfarin-azapropazone site’ and ‘indole-benzodiazepine site’, respectively.

Given that warfarin primarily binds to HSA site I with a high affinity, the majority fraction of administered warfarin (c.a. 95%) is sequestered by HSA, resulting in a low degree of free drug. Numerous studies have revealed that warfarin’s PPB capacity accounts for its PK characteristics, including small distribution volume and low clearance, which are pivotal factors to be considered for its medical uses [5]. On the contrary, only a few human clinical trials have reported data on CBD’s PK/PD properties. A study with human subjects (healthy volunteers) suggests that CBD’s bioavailability is low (0.9–6.5%) and varying by multiple factors including administration routes (i.e. oral solution, oral capsule, and oromucosal spray), dose range (5–6000 mg), and feeding status (fasted- and fed-state) [9]. Moreover, the PPB capacity of a drug can affect the concentration of its bioavailable form (i.e. free drug status) at the site of action, as well as its elimination rate in the systemic circulation, which can collectively affect the overall drug bioavailability [10]. However, it was not clear whether CBD’s low bioavailability is attributed to its PPB capacity. Although a study analyzed CBD’s PPB capacity using an atmospheric pressure chemical ionization MS/MS method [11], the direct binding profiles between CBD and plasma proteins are not well elucidated.

Herein, we aimed to characterize the protein interactions between CBD and two major plasma proteins, namely, HSA and γ-globulins, using a combination of biosensor and microcalorimetric tools. CBD’s protein binding characteristics, including affinity, stoichiometry and thermodynamics, and thermal transition changes with HSA were obtained by surface plasmon resonance (SPR), isothermal titration calorimetry (ITC), and differential scanning calorimetry (DSC). In addition, the binding profile of CBD on HSA is depicted by a computational modeling method and further supported by competitive SPR binding experiments with CBD and warfarin.