Skip to content
Frontiers
Search

Review

Frontiers in Chemical Biology ·

Volume 5 - 2026 | https://doi.org/10.3389/fchbi.2026.1830007

Structural insights into cannabinoid receptors CB1 and CB2: determinants of ligand selectivity

  • Ann Varghese1
  • Jie Liu1
  • Wenjing Guo1
  • Tucker A. Patterson1
  • Huixiao Hong1*corresponding author
  1. National Center for Toxicological Research, U.S. Food and Drug Administration, Jefferson, AR, United States
Article metrics0 viewsView details
PDFCiteMetrics

Abstract

Cannabinoid receptors 1 (CB1) and 2 (CB2) are key components of the endocannabinoid system and play central roles in regulating diverse physiological processes, including neural signaling, immune modulation, and inflammation. Both receptors belong to the Class A G protein-coupled receptor (GPCR) family and exhibit distinct tissue distributions and signaling profiles, making them attractive yet challenging targets for receptor-specific drug development. Efforts to design subtype-selective ligands have been hindered by the high degree of sequence and structural similarity between CB1 and CB2, resulting in widespread ligand cross-reactivity. Recent advances in structural biology have yielded high-resolution structures of both CB1 and CB2 in multiple functional states, providing new opportunities to dissect the molecular determinants of receptor selectivity. In this review, we present a comprehensive comparative analysis of CB1 and CB2 structures, with a focus on differences in ligand-binding residues, binding pocket topology, and physicochemical environments that may influence subtype selectivity and signaling. Our analysis suggests that CB1 preferentially accommodates bulkier and more flexible ligands, whereas CB2 favors smaller and more compact chemotypes. In addition, positionally conserved but chemically distinct residues (e.g., Leu193 in CB1 vs. Ile110 in CB2 and Leu359 in CB1 vs. Val261 in CB2) highlight subtle structural differences that can influence ligand binding and receptor activation. Overall, this work aims to inform the rational design of next-generation cannabinoid ligands with enhanced receptor selectivity, reduced side effects, and improved therapeutic potential.

Keywords

Introduction

The endocannabinoid system (ECS) is a vital regulatory network that maintains physiological balance by modulating processes such as pain perception, appetite, mood, immune function, and neuroprotection (Marzo et al., 2004; Mechoulam and Parker, 2013; Lu and Mackie, 2021). Dysregulation of this system is implicated in a wide range of diseases, including chronic pain, epilepsy, multiple sclerosis, Alzheimer’s disease, cancer, and inflammatory disorders, many of which still lack effective therapeutic solutions (Centonze et al., 2007; Maroof et al., 2013; Pisanti et al., 2013; Zieglgänsberger et al., 2022). The ECS is a complex cell-signaling system composed of three major components: endocannabinoids, their biosynthetic and degradative enzymes, and cannabinoid receptors (Lu and Mackie, 2016). It primarily functions through two G protein-coupled receptors (GPCR), cannabinoid receptor 1 (CB1) and cannabinoid receptor 2 (CB2), traditionally referred to as the canonical cannabinoid receptors.

These important ECS components belong to the Class A GPCR family, also known as the rhodopsin family. CB1 and CB2 were discovered more than 30 years ago, with CB1 identified first and initially regarded as the cannabinoid receptor due to its early discovery, until CB2 was later characterized (Devane et al., 1988; Matsuda et al., 1990). CB1 is predominantly expressed in the central nervous system, and is most abundant in brain regions governing cognition, memory (cortex, hippocampus), and motor control (basal ganglia, cerebellum) (Marsicano and Lutz, 1999; Chou et al., 2022). CB1 is also present at lower levels in several peripheral tissues, including sensory neurons, as well as reproductive and digestive systems (Maccarrone et al., 2002; Pinto et al., 2002). CB2 was historically viewed as limited to peripheral tissues and immune cells but recent evidence also show its presence in the brain, particularly robust in microglia and at low, context-dependent levels in some neurons (Sharkey, 2022; Oddi et al., 2023). Both receptors are integral to maintaining physiological homeostasis and serve as promising therapeutic targets for epilepsy, obesity, and pain. In general, CB1 receptors are primarily associated with neurological and metabolic functions due to their predominant expression in the central nervous system, whereas CB2 receptors are more closely linked to immunomodulatory and protective roles, reflecting their enrichment in immune cells and peripheral tissues. CB1 receptors reduce nociceptive transmission primarily by inhibiting neurotransmitter release through Gi/o-mediated suppression of adenylate cyclase and calcium influx, whereas CB2 receptors complement this action by modulating neuroimmune responses and glial activity at spinal, supraspinal and peripheral levels, together contributing to pain regulation (Khan et al., 2025). Moreover, CB1 is involved in regulating food intake and cognitive processes (Bellocchio et al., 2010; Ruehle et al., 2012; Busquets Garcia et al., 2016). CB2 is recognized for its therapeutic potential in neurodegenerative diseases, where its anti-inflammatory and neuroprotective actions can alleviate neuroinflammation and delay progression in disorders like Parkinson’s and Alzheimer’s (Han et al., 2020; Ishiguro et al., 2022; Ruiz De Martín Esteban et al., 2022). The major physiological differences between CB1 and CB2 receptors are listed in Table 1. Due to their diverse functional roles and broad tissue distribution, CB1 and CB2 are considered promising drug targets. In fact, GPCRs remain the most extensively exploited protein family in drug discovery, with roughly one in three FDA-approved therapeutics acting through this receptor class (Lundstrom, 2009).

TABLE 1. Comparison of major physiological roles of CB1 and CB2 receptors.

FeatureCB1 receptorCB2 receptor
Primary locationCentral nervous system (brain, spinal cord); also peripheral tissuesPredominantly immune cells and peripheral tissues; low expression in CNS
Main functionNeuromodulation and Energy BalanceImmunomodulation and Anti-inflammatory responses
PsychoactivityHigh (mediates psychoactive effects)None (non-psychotropic)
Role in painCentral pain processingPeripheral/inflammatory pain relief
MetabolismIncreases appetite/energy storageImproves glucose tolerance

Over the past decades, numerous ligands targeting CB1 and CB2 have been developed, but their clinical outcomes differ significantly. CB1 inverse agonists such as rimonabant and taranabant were initially introduced for obesity treatment, but later withdrawn due to severe psychiatric side effects associated with central CB1 blockade (Moreira and Crippa, 2009; Aronne et al., 2010; Proietto et al., 2010). This has prompted the development of alternative strategies aimed at retaining therapeutic efficacy while minimizing central nervous system (CNS) exposure. In particular, peripherally restricted CB1 ligands that limit blood-brain barrier penetration have emerged as an important strategy to minimize CNS-related side effects.

In contrast, some selective CB2 agonists such as HU-308, JWH-133, and GW-405833 have shown activity in in-vitro assays without eliciting psychoactive responses (Hanuš et al., 1999; Valenzano et al., 2005; Verty et al., 2015; Poleszak et al., 2020). Despite these ongoing efforts, only a handful of phytocannabinoids and their synthetic analogs have reached the market, and no receptor-selective CB1 or CB2 drugs have yet received clinical approval. The major challenge is that most cannabinoid receptor ligands exhibit affinity for both subtypes, reflecting poor selectivity. The lack of selectivity arises from the high degree of homology between the two receptors; CB1 and CB2 share approximately 44% overall sequence identity, which increases to about 68% within their transmembrane regions (Munro et al., 1993). Ongoing research is, therefore, focused on overcoming this challenge, as achieving receptor selectivity remains key to enable the design of cannabinoid-based therapeutics with improved safety and reduced off-target toxicity.

Advances in structural biology have greatly deepened our understanding of cannabinoid receptor architecture and function, providing vital insights to support structure-based drug design. Although substantial progress has been made in CB1 and CB2 receptor-targeted drug discovery, the molecular understanding of ligand selectivity between the two receptors remains incomplete. The availability of three-dimensional structures of CB1 and CB2 provides an opportunity to elucidate molecular features underlying receptor selectivity. This study aims to systematically analyze currently available crystallographic and cryo-EM structures of CB1 and CB2 receptors. By comparing ligand-bound conformations, binding pocket architectures, and key receptor amino acid residue (hereafter referred to as “residue”) interactions, we seek to evaluate the structural determinants that underlie receptor selectivity. Our goal is to provide a comprehensive structural framework to inform the rational design of highly selective cannabinoid ligands for therapeutic applications.

While this review focuses on ligand-binding pocket architecture, receptor activation in CB1 and CB2 involves conserved GPCR molecular switches and ligand-induced conformational rearrangements, including outward movement of transmembrane helix 6 (TM6), reorganization of toggle switch residues (e.g., Phe2003.36/Trp3566.48), and conformational changes in the DRY and NPxxY motifs that facilitate G protein coupling. Structural and biophysical studies have shown that CB1 exhibits pronounced conformational flexibility and ligand-dependent activation states, whereas CB2 adopts comparatively more constrained active conformations (Hua et al., 2016; Shao et al., 2016; Krishna Kumar et al., 2019; Xing et al., 2020). Comprehensive discussions of these activation mechanisms and downstream signaling pathways are available in prior reviews and are beyond the scope of the present work (Wootten et al., 2013).

Structural overview of CB1 and CB2

CB1 and CB2 adopt the canonical seven-transmembrane helical fold characteristic of the GPCR family (Figure 1) (Baldwin, 1993). CB1 consists of 472 amino acids; CB2 is relatively shorter and consists of 360 amino acids. The amino acid residues of class A GPCRs are commonly annotated using the Ballesteros-Weinstein (B.W.) numbering system which consists of two numbers. The first digit denotes the transmembrane helix and the second indicates the residue’s position with respect to the most conserved residue of that helix, assigned as 50 (Ballesteros and Weinstein, 1995). The orthosteric ligand-binding pockets of CB1 and CB2 are nestled within the transmembrane domain and further shaped by extracellular loops, while the intracellular face of the helical bundle forms the interface for G-protein coupling and downstream signaling. The broad structural similarity in CB1 and CB2 indicates their shared mode of binding or mechanism. The sequential differences in these receptors are mainly observed in the N-terminus, C-terminus, and the extracellular loop-2 (ECL2) which has a major involvement in ligand binding (Montero et al., 2005).

Architecture of cannabinoid receptors CB1 and CB2. The left depicts **(A)** the general topology of G-protein coupled receptors. The right **(B)** shows the overlay of CB1 (magenta) and CB2 (orange) receptor proteins.
Figure 1. Architecture of cannabinoid receptors CB1 and CB2. The left depicts (A) the general topology of G-protein coupled receptors. The right (B) shows the overlay of CB1 (magenta) and CB2 (orange) receptor proteins.

Emerging evidence suggests that extracellular regions are not merely structural connectors but actively contribute to receptor function. In particular, ECL2 can directly participate in ligand binding and receptor activation, while also influencing receptor folding and trafficking. ECL2 has further been implicated in guiding ligand entry into the binding pocket (Woolley and Conner, 2017). In addition, the membrane-proximal region of the CB1 N-terminus modulates ligand binding through an allosteric mechanism, with conserved cysteine residues forming a disulfide bond that influences receptor–ligand interactions and cooperativity (Fay and Farrens, 2013). From an evolutionary perspective, CB1 receptors exhibit a high degree of sequence conservation across species, whereas CB2 receptors are comparatively less conserved (Pertwee et al., 2010).

Cannabinoid receptor ligands are commonly classified either based on their origin—including phytocannabinoids, endocannabinoids, and synthetic cannabinoids—or by their chemical structure, which comprises classical, nonclassical, aminoalkylindole, and eicosanoid types (Pertwee et al., 2010; Hourani and Alexander, 2018). Phytocannabinoids such as Δ9-tetrahydrocannabinol (THC) and cannabidiol (CBD) are plant-derived compounds from Cannabis sativa, whereas endocannabinoids like anandamide and 2-arachidonoylglycerol are produced naturally in the body. Synthetic cannabinoids include a broad range of analogues designed to probe receptor function or develop therapeutics. Structurally, the classical cannabinoids are dibenzopyran derivatives exemplified by Δ9-THC and its synthetic analogue HU-210. Nonclassical cannabinoids lack the characteristic pyran ring of classical cannabinoids and possess bicyclic or tricyclic cores. The aminoalkylindoles like R-(+)-WIN-55,212 contain an indole-based scaffold and eicosanoids (like anandamide and 2-arachidonoylglycerol) are long-chain fatty acid derivatives, specifically from arachidonic acid. Unlike the other groups, these compounds are flexible, lipidic molecules rather than rigid polycyclic systems, yet they are still capable of interacting effectively with CB1 and CB2 receptors. These diverse structural classes indicate the chemical versatility of compounds capable of modulating CB1 and CB2 receptors, contributing to their wide range of pharmacological effects.

The first structure of CB1 was deposited in the protein data bank (PDB) in 2016 (Hua et al., 2016). Following this, in 2019, the first structure of CB2 was deposited (Li et al., 2019). We analyzed a total of 13 CB1 and 8 CB2 three-dimensional structures with bound ligands retrieved from the PDB. To avoid redundancy, only one representative structure per unique ligand-receptor complex was included, with preference given to the highest-resolution structure when multiple entries were available. Among the 13 CB1 structures, five were determined by X-ray crystallography and eight by cryo-EM, whereas among CB2 structures, one was resolved by X-ray crystallography and the remaining using cryo-EM. Several of these structures incorporate stabilizing elements such as nanobodies, antibody fragments (e.g., scFv16), flavodoxin, or glycogen synthase fragments to enhance receptor stability and facilitate structure determination. The CB1 and CB2 receptor-ligand structures that have been reported and analyzed are summarized in Table 2. These include seven CB1 complexes solved with heterotrimeric G proteins; five with Gi (PDB IDs: 6N4B, 9ERX, 8GHV, 9B54, 9EGO) and one with Gs (PDB ID: 8K8J). The remaining CB1 entries represent ligand-bound receptor structures without G-protein coupling, providing insights into inactive or pre-activation states of the receptor. Among the eight CB2 structures, seven were resolved in complex with heterotrimeric Gi proteins, while one structure (PDB ID: 5ZTY) represents a ligand-bound receptor without G-protein coupling.

TABLE 2. Structural, pharmacological, and functional characterization of CB1 and CB2 receptor-ligand complexes.

ReceptorPDB IDResolution (Å)MethodLigandBinding affinityLigand functional typeReceptor stateComplex/Stabilizing elements
CB1 pKiCB2 pKi
CB15TGZ2.80X-ray diffractionAM65388.34-AntagonistInactiveFlavodoxin fusion
5U092.60X-ray diffractionTaranabant9.526.54AntagonistInactiveGlycogen synthase (GlgA) fusion
5XR82.95X-ray diffractionAM8418.948.50AgonistActiveFlavodoxin fusion
5XRA2.80X-ray diffractionAM115429.96-AgonistActiveFlavodoxin fusion
6N4B3.00Electron microscopyMDMB-Fubinaca (FUB)8.57*9.89AgonistActiveHeterotrimeric Gᵢ (Gαᵢ1, Gβ1, Gγ2) + scFv16
8K8J2.88Electron microscopyFenofibrate6.307.7Partial agonistActiveMiniGαs/q (Gαs short, Gβ1, Gγ2) + Nb35nanobody
9B9Z3.30Electron microscopyMRI-18678.64-Peripherally restricted inverse agonist/antagonistInactiveGlgA fusion + CNb36
9BA03.13Electron microscopyMRI-18919.52-Peripherally restricted inverse agonist/antagonistInactiveGlgA fusion + CNb36
9ERX2.90Electron microscopyHU2108.98*9.66*AgonistActiveHeterotrimeric Gᵢ1 (Gαᵢ1, Gβ1, Gγ2) + scFv16
8GHV2.80Electron microscopyAMG3158.116.75AgonistActiveHeterotrimeric Gᵢ (Gαᵢ1, Gβ1, Gγ2) + scFv16
9B542.86Electron microscopyVIP367.66-Gi-biased agonistActiveHeterotrimeric Gᵢ (Gαᵢ1, Gβ1, Gγ2) + scFv16
7FEE2.70X-ray diffractionCP55,9408.92*9.05*AgonistActiveGlgA fusion + allosteric ligand ZCZ011
9EGO3.20Electron microscopy1,350 (YVF)9.02-AgonistActiveHeterotrimeric Gᵢ (Gαᵢ1, Gβ1, Gγ2) + scFv16
CB25ZTY2.80X-ray diffractionAM10257-10.12AntagonistInactiveT4 lysozyme fusion
8GUQ3.08Electron microscopyAPD3715**8.21**AgonistActiveHeterotrimeric Gᵢ (Gαᵢ1, Gβ1, Gγ2) + scFv16
8GUS2.97Electron microscopyHU3085*7.64*AgonistActiveHeterotrimeric Gᵢ (Gαᵢ1, Gβ1, Gγ2) + scFv16
8GUT2.98Electron microscopyLEI-102-8AgonistActiveHeterotrimeric Gᵢ (Gαᵢ1, Gβ1, Gγ) + scFv16
8X3L3.13Electron microscopyYL0254.52**7.46**AgonistActiveHeterotrimeric Gᵢ (Gαᵢ1, Gβ1, Gγ2) + scFv16
6KPF2.90Electron microscopyAM120339.299.43AgonistActiveHeterotrimeric Gᵢ (Gαᵢ1, Gβ1, Gγ2) + scFv16
6PT03.20Electron microscopyWIN 55,212-27.68*8.63*AgonistActiveHeterotrimeric Gᵢ (Gαᵢ1, Gβ1, Gγ2) + scFv16
8GUR2.84Electron microscopyCP55,9408.92*9.05*AgonistActiveHeterotrimeric Gᵢ (Gαᵢ1, Gβ1, Gγ2) + scFv16

Median pKi from several independent determinations.

pEC50 (pKi not available; value shown is functional potency).

Both CB1 and CB2 receptors primarily couple to Gi/o proteins, resulting in inhibition of adenylyl cyclase and activation of MAPK signaling pathways. However, CB1 exhibits broader signaling versatility, with the ability to additionally engage Gs and Gq pathways, whereas CB2 signaling remains largely restricted to Gi/o coupling, leading to a comparatively more limited signaling profile (Ibsen et al., 2017).

Among the 13 CB1-bound ligands reported in the PDB, three are classical cannabinoids (HU210, AM11542, AM841). The remaining ligands include one representative each from the nonclassical (CP55940), aminoalkylindole (MDMB-Fubinaca), diarylpyrazole (AM6538), and ethanolamide (AMG315) classes, as well as five structurally diverse small molecules—Fenofibrate, MRI-1867, MRI-1891, VIP36, YVF—which collectively expand the known chemotype space of CB1 receptor ligands. Across the structurally diverse set of CB1-bound ligands (Table 2), several common features emerge among agonists. These typically include (1) a hydrophobic alkyl chain or bulky lipophilic substituent and (2) a polar headgroup containing hydrogen-bond donors or acceptors—such as phenolic hydroxyl, amide, or carbonyl functionalities—often complemented by aromatic or cyclic ring systems that contribute to receptor binding.

The CB2 structures have a smaller but chemically varied ligand set. Of the eight CB2-bound ligands reported, two are classical cannabinoid analogs (HU308, AM12033), one each belongs to the nonclassical (CP55940) and aminoalkylindole scaffolds (WIN 55,212), and the remaining ∼50% comprise novel CB2-selective chemotypes—including AM10257, APD371, LEI-102, and YL025—that extend beyond the traditional cannabinoid framework. The CB1- and CB2-bound ligands identified from the PDB are illustrated in Figures 2, 3, respectively. Collectively, these observations indicate that most experimentally resolved receptor-ligand complexes involve classical or closely related cannabinoid scaffolds. Nevertheless, the presence of novel chemotypes in several structures indicates increasing efforts to explore nontraditional scaffolds with the potential to confer enhanced receptor selectivity and functional diversity.

Two-dimensional structures of the CB1-bound ligands in the PDB.
Figure 2. Two-dimensional structures of the CB1-bound ligands in the PDB.
Two-dimensional structures of ligands bound to CB2 in the PDB.
Figure 3. Two-dimensional structures of ligands bound to CB2 in the PDB.

Ligand binding site architecture

Structural analyses of CB1 and CB2 indicate that ligands predominantly bind to a common site, the phytocannabinoid binding pocket where natural cannabinoids such as THC and CBD bind. Although CB1 and CB2 share a highly conserved overall fold, ligand-specific interactions can introduce subtle differences in the local pocket environment. To examine these variations, residues located within 5.0 Å of the bound ligands (Figure 4) were analyzed across all available CB1 and CB2 structures, and this distance was used to define the binding pocket.

Illustration of binding pockets (residues within 5 Å of bound ligands) of cannabinoid receptors: **(A)** CB1 and **(B)** CB2. Residues are color-coded based on frequency of occurrence across analyzed structures (forest green: high-frequency; hot pink: low-frequency), with carbon atoms following this scheme. Heteroatoms are displayed using standard element-based coloring (N: blue, O: red, S: yellow). Key high-frequency residues are labeled to facilitate interpretation. Residues occupying equivalent positions but differing in chemical properties between CB1 and CB2, which may contribute to subtype selectivity, are enclosed in red boxes. Surface representations of the binding pockets of **(C)** CB1 and **(D)** CB2, colored according to residue frequency of occurence.
Figure 4. Illustration of binding pockets (residues within 5 Å of bound ligands) of cannabinoid receptors: (A) CB1 and (B) CB2. Residues are color-coded based on frequency of occurrence across analyzed structures (forest green: high-frequency; hot pink: low-frequency), with carbon atoms following this scheme. Heteroatoms are displayed using standard element-based coloring (N: blue, O: red, S: yellow). Key high-frequency residues are labeled to facilitate interpretation. Residues occupying equivalent positions but differing in chemical properties between CB1 and CB2, which may contribute to subtype selectivity, are enclosed in red boxes. Surface representations of the binding pockets of (C) CB1 and (D) CB2, colored according to residue frequency of occurence.

In ligand-bound CB1, 14 residues bind to most of the ligands, each appearing in at least 11 out of 13 complex structures (∼85%), and are termed as high-frequency residues, whereas CB2 has 22 such residues, which present in at least 7 out of 8 structures (∼87%) (Figure 5). Residues that bind to less than 11 and 7 ligands for CB1 and CB2, respectively, are called low-frequency residues.

Heatmap representation of ligand-binding residue frequencies across CB1 **(A)** and CB2 **(B)** receptor structures. Residues within 5 Å of bound ligands are shown in green, residues beyond 5 Å are shown in light green, and residues not resolved in the structure are shown in red.
Figure 5. Heatmap representation of ligand-binding residue frequencies across CB1 (A) and CB2 (B) receptor structures. Residues within 5 Å of bound ligands are shown in green, residues beyond 5 Å are shown in light green, and residues not resolved in the structure are shown in red.

To bridge structural observations with experimental validation, available mutagenesis data for high-frequency binding site residues were compiled and are summarized in Table 3. These studies show that alanine substitutions at several high-frequency residues reduce agonist binding affinity or potency, with effects ranging from modest (∼4-fold) to substantial (>200-fold) decreases. In contrast, mutation of the ECL2 residue Phe268 results in more pronounced effects, including loss of ligand binding and impaired receptor signaling. Together, these findings indicate that the identified high-frequency binding site residues play important roles in ligand recognition and receptor activation, providing a basis for comparing binding pocket properties between CB1 and CB2.

TABLE 3. Mutagenesis data for high-frequency ligand-binding residues in CB1 and CB2 receptors.

ReceptorAmino acid residueB.W. numberMutationEffects on binding and/or functionReferences
CB1Ser3837.39Ala↓ agonist potency (∼200 fold, CP55,940)Sitkoff et al. (2011), Hua et al. (2025)
CB1Phe3797.35Ala↓ binding affinity (∼3–4 fold, SR141716A); ↓ agonist potencyShim et al. (2012), Hua et al. (2025)
CB1Met3636.55Ala↓ agonist binding affinity (HU-210, CP55,940, Δ9-THC; ∼4–15 fold)Shim et al. (2011), Díaz et al. (2019)
CB1Leu 3596.51AlaAbolishes THC’s ability to activate the receptorLlinas Del Torrent et al. (2023)
CB1Phe1742.61Ala↓ antagonist functional affinity (rimonabant, AM6538)↓ agonist binding affinity (HU-210, CP55,940, Δ9-THC; ∼22–146 fold)Shim et al. (2011), Hua et al. (2016)
CB1Phe1702.57Ala↓ antagonist functional affinity (rimonabant, AM6538)Hua et al. (2016)
CB1Trp2795.43AlaStrong reduction in antagonist binding (SR141716A; incomplete displacement at 5 µM)McAllister et al. (2003), Shim et al. (2012)
CB1*Phe268ECL2Ala↓ signaling (∼170-fold EC50 shift)Loss of ligand binding (CP55,940/SR141716A)Ahn et al. (2009)
CB1Leu1933.29Ala↓ agonist binding affinity (HU-210, CP55,940, Δ9-THC; ∼10–100 fold)Shim et al. (2011)
CB2Ser2857.39Ala, Leu↓ agonist binding affinity (∼12 fold, HU-243, S→A)Partially impairs CBD’s NAM effect on JWH-133 (S→L)Rhee (2002), Navarro et al. (2021)
CB2Phe2817.35LysAbolishes agonist/antagonist binding (CP55940/SR144528)Feng et al. (2014)
CB2Val2616.51LeuGain of CB1-like activation by THCLlinas Del Torrent et al. (2023)
CB2Trp1945.43Ala, PheAbolishes agonist/antagonist binding (CP55940/WIN55212-2/SR144528)Zhang et al. (2011)
CB2Tyr1905.39Phe, Ile↑ binding affinity (Δ9-THC ∼4-fold, Y→F)↓ binding affinity (anandamide ∼25-fold, Y→F); ↓ binding affinity (loss, Y→I)McAllister et al. (2002)
CB2*Phe183ECL2Gln, ValAbolishes agonist/antagonist binding (CP55940/SR144528)Feng et al. (2014)
CB2Val1133.32GluAbolishes agonist/antagonist binding(CP55,940/SR144528/XIE95-26)Alqarni et al. (2014)
CB2Val1133.32LeuLargely preserves agonist/antagonist binding (minor ↓ SR144528 binding only)Alqarni et al. (2014)
CB2His952.65Ala↓ agonist potency (∼219 fold, CP55,940)Li et al. (2019)
CB2Phe942.64Ala↓ agonist potency (∼78 fold, CP55,940)Li et al. (2019)
CB2Phe912.61Ala↓ agonist potency (∼35 fold, CP55,940)Li et al. (2019)
CB2Phe872.57Ala↓ agonist potency (∼416 fold, CP55,940)Li et al. (2019)

Residues marked with an asterisk are in the ECL2 and are not assigned Ballesteros-Weinstein number.

These binding site residues directly influence ligand binding capacity by mediating key intermolecular interactions that govern ligand affinity, orientation, and stability within the pocket. Hydrophobic residues contribute to van der Waals contacts and shape complementarity, facilitating accommodation of lipophilic ligand moieties, whereas polar and charged residues enable hydrogen bonding and electrostatic interactions that can stabilize specific ligand conformations. In addition, certain residues play structural roles by maintaining pocket geometry and conformational integrity, thereby indirectly modulating ligand accessibility and binding efficiency. Collectively, these interactions define the physicochemical environment of the binding pocket and determine how effectively different ligands can be recognized and retained.

Despite having fewer available structures (8 CB2 vs. 13 CB1), CB2 has more pocket residues (22 CB2 vs. 14 CB1), suggesting that CB2 has a larger ligand binding pocket and ligands can adopt more binding orientations and engage more flexible binding residues, whereas CB1 has a smaller ligand binding pocket, and thus ligands may show greater similarity in their poses and interaction patterns. Among the high-frequency residues, the proportion of polar (21.4% in CB1 vs. 18% in CB2) and charged (7.1% in CB1 vs. 4.5% in CB2) residues was slightly higher in CB1, whereas nonpolar residues were more abundant in CB2 (77.2% in CB2 vs. 71.4% in CB1). These observations imply that the CB2 binding pocket is relatively more hydrophobic, which may favor interactions with lipophilic ligands, while CB1 features a slightly higher contribution from polar and charged residues that could support additional hydrogen-bonding or electrostatic interactions. Our analysis agrees with earlier computational and mutagenesis studies showing that CB2 appears to rely more heavily on aromatic stacking interactions within its hydrophobic subdomain than CB1 (Ashton et al., 2008). This distinction is also reflected in previous analyses of C1 phenol analogs of Δ8/Δ9-THC, where the phenolic hydroxyl group was found to be important for CB1 binding by acting as a hydrogen-bond donor, but not essential for CB2, where hydrogen bonding at this position is unnecessary (Huffman et al., 1996; Bow and Rimoldi, 2016). Moreover, experimental and modeling studies by Huffman and co-workers revealed that removal of the phenolic hydroxyl group did not diminish receptor affinity as previously assumed. Instead, CB1 can adapt by reorienting the ligand such that the pyran oxygen forms a compensatory hydrogen bond with Lys192. This flexibility demonstrates that while the phenolic hydroxyl is preferred, it is not essential for CB1 binding, indicating the adaptable nature of the CB1 binding pocket (Huffman et al., 1996). Prior studies highlight Lys192 as a crucial hydrogen bonding partner for CB1 ligand affinity (Song and Bonner, 1996; Ashton et al., 2008; Reggio, 2010). The functional importance of Lys192 is further supported by site-directed mutagenesis studies. Substitution of Lys192 with alanine, glutamine, or glutamate (Lys192Ala, Lys192Gln, or Lys192Glu) markedly reduces the binding affinity of several CB1 agonists, including anandamide, HU-210 and CP55,940 (Chin et al., 1998; Pan et al., 1998; Hurst et al., 2002).

However, our structural analysis indicates that its involvement varies among available receptor-ligand three dimensional structures of CB1. Lys192 interacts with seven of the 13 CB1 crystal structures analyzed in this study, indicating this residue is important for ligand binding, but not necessary. Furthermore, static crystallographic data provides only a partial view of receptor-ligand interactions, as proteins are inherently dynamic; future molecular dynamics and ensemble-based analyses could, therefore, offer deeper insight into the transient role of Lys192 during ligand recognition and activation.

To better understand the structural basis of the similarities between cannabinoid receptors, we selected one CB1 and one CB2 receptor structures with the highest resolution and superimposed them with particular attention to the binding pockets. The comparison revealed that most residues forming the ligand-binding pocket in CB1 and CB2 occupy equivalent positions, indicating a high degree of conservation between the two subtypes. This observation suggests that the overall topology of the phytocannabinoid binding pocket is maintained across both receptors. Such conservation, however, poses a major challenge in designing subtype-selective ligands. A closer inspection revealed positionally equivalent, but chemically distinct residues at a few sites (Table 4), such as Leu193 in CB1 corresponding to Ile110 in CB2, and Leu359 in CB1 corresponding to Val261 in CB2. These nonpolar substitutions preserve hydrophobicity, but may introduce subtle variations in side-chain length that may influence pocket geometry. Even such conservative changes can slightly alter side-chain packing, shifting pocket shape or volume just enough to affect how bulky or elongated ligands are accommodated.

TABLE 4. Positional Equivalence of binding pocket residues in CB1 and CB2.

CB1 amino acid residueB.W. numberCB2 amino acid residueB.W. number
*Phe268ECL2*Phe183ECL2
Val1963.32Val1133.32
Phe1702.57Phe872.57
Ser3837.39Ser2857.39
Thr1973.33Thr1143.33
Leu3596.51Val2616.51
Phe3797.35Phe2817.35
Met3636.55Met2656.55
Leu1933.29Ile1103.29
Phe1742.61Phe912.61
Trp2795.43Trp1945.43
Ser1732.60Ser902.60
Cys3867.42Cys2887.42
His1782.65His952.65
Phe1772.64Phe942.64
Phe2003.36Phe1173.36
*Pro269ECL2*Pro184ECL2
*Ile271ECL2*Ile186ECL2
Phe1893.25Phe1063.25
Tyr2755.39Tyr1905.39
Trp3566.48Trp2586.48
Leu2765.40Leu1915.40

Residues marked with an asterisk are in the ECL2 and are not assigned Ballesteros-Weinstein number. Amino acid residues in green and pink represent high and low frequency residues respectively.

Importantly, experimental evidence supports the functional relevance of these differences. Alanine scanning mutagenesis studies have shown that substitution of Leu193 and nearby residues within the transmembrane helix 2-helix 3 (H2/H3) region (including Phe174, Phe177, Phe189 and Lys192) to alanine significantly reduces the binding affinity of the agonist CP55,940, highlighting the importance of this region in ligand recognition (Murphy and Kendall, 2003; McAllister et al., 2004; Shim et al., 2011). In addition, residue at 6.51, corresponding to Leu359 in CB1 and Val261 in CB2, plays a critical role in ligand-dependent receptor activation. Substitution of Leu3596.51 with Val in CB1 abolishes THC-mediated signaling, whereas the reciprocal mutation in CB2 (Val6.51Leu) restores CB1-like activation by THC. These findings demonstrate that even a single conservative substitution at this position can directly influence ligand efficacy and subtype selectivity, identifying 6.51 as a key determinant of receptor activation in cannabinoid receptors (Llinas Del Torrent et al., 2023).

In addition to residue-level variations, differences in the size and volume of the binding pocket are important considerations in understanding receptor selectivity, as they dictate the range of ligand sizes each receptor can accommodate. CB1 and CB2 receptors differ, to a modest extent, in the sizes of ligands they bind to. Based on experimental structural data from the PDB, ligands bound to CB1 often reach molecular weights of about 590 Da, whereas those in CB2 complexes are smaller, typically up to 450 Da (Table 5). Such differences suggest that pocket dimensions and flexibility may serve as contributing factors to subtype selectivity. In this context, ligand shape and conformational rigidity become equally important factors. Prior reports have shown that variations in ligand rigidity and hydrophobicity can influence receptor preference, with more compact and hydrophobic ligands tending to favor CB2 (Thakur G. A. et al., 2005; Lu et al., 2008; Bow and Rimoldi, 2016). Further experimental and computational analyses of long-chain cannabinoid agonists showed higher binding affinities for CB1 than CB2, supporting the notion that CB1 accommodates larger and more flexible ligands (Lin et al., 1998; Hillard et al., 1999; Di Marzo et al., 2001; Thakur G. et al., 2005; Yang et al., 2020).

TABLE 5. Ligand molecular weight, binding pocket surface area (SASA and SES), and selectivity data for CB1 and CB2 receptor–ligand complexes.

ReceptorPDB IDLigandLigandchemotypeLigand M.W. (Da)Pocket SASA (Å2)Pocket SES (Å2)ΔpKi (pKi CB1 – pKi CB2)
CB15TGZAM6538Diarylpyrazole542.413,4222,654>0*
CB15U09TaranabantAcyclic amide515.953,3302,6232.98
CB15XR8AM841Classical cannabinoid445.662,62620510.44
CB15XRAAM11542Classical cannabinoid449.462,9782,480>0*
CB16N4BMDMB-Fubinaca (FUB)Aminoalkylindazole397.443,0042,458−1.32
CB18K8JFenofibrateStructurally distinct/Novel360.832,9822,413−1.40
CB19B9ZMRI-1867Structurally distinct/Novel547.983,2682,332>0*
CB19BA0MRI-1891Structurally distinct/Novel591.003,2592,414>0*
CB19ERXHU210Classical cannabinoid386.572,9072,490−0.68
CB18GHVAMG315Ethanolamide375.593,1942,5751.36
CB19B54VIP36Aminoalkylindazole526.603,2682,604>0*
CB17FEECP55,940Non-classical cannabinoid376.573,0442,509−0.13
CB19EGO1,350 (YVF)Structurally distinct/Novel437.433,1912,643>0*
CB1 (Mean ± SD)​​​458 ± 78.13,113.3 ± 215.32,480.5 ± 162.8​
CB25ZTYAM10257Structurally distinct/Novel421.573,0392,517<0*
CB28GUQAPD371Tricyclic pyrazole-3-carboxamide357.412,9862,429−3.21**
CB28GUSHU308Classical Cannabinoid414.623,0332,674−2.64
CB28GUTLEI-102Structurally distinct/Novel454.542,9642,571<0*
CB28X3LYL025Structurally distinct/Novel409.393,0042,449−2.94**
CB26KPFAM12033Classical Cannabinoid399.573,0872,598−0.14
CB26PT0WIN 55,212-2Aminoalkylindole426.513,0622,511−0.95
CB28GURCP55,940Non-classical cannabinoid376.573,0662,429−0.13
CB2 (Mean ± SD)​​​407.52 ± 30.133,030.1 ± 42.52,522.2 ± 87.9​

These values represent ΔpEC50 values (pEC50(CB1) − pEC50(CB2)), used where Ki data were unavailable.

For ligands lacking reported affinity at one receptor, selectivity is indicated qualitatively as >0 (CB1-selective) or <0 (CB2-selective).

Ligand flexibility is an important determinant of how different chemotypes adopt the structural motifs required for receptor docking. Endogenous cannabinoids such as anandamide (AEA) are highly flexible due to their long polyunsaturated acyl chains and can adopt folded or “hairpin-like” conformations within the binding pocket. This conformational adaptability enables alignment of key pharmacophoric elements, including the polar headgroup and hydrophobic tail, with complementary regions of the receptor. In contrast, more rigid synthetic agonists, such as HU-210 and related polycyclic compounds, are conformationally constrained and effectively pre-organized to present similar spatial arrangements for receptor interaction. Despite these differences, both flexible and rigid ligands converge on comparable binding geometries, engaging conserved interaction hotspots within the pocket. These observations highlight that ligand flexibility and conformational pre-organization represent alternative strategies for achieving efficient receptor docking and stabilization.

To delve deeper into the structural characteristics of the cannabinoid binding pockets, we computed the solvent-accessible surface area (SASA) and solvent-excluded surface (SES) for all the ligand-bound structures. SASA represents the total molecular surface accessible to solvent molecules, providing an estimate of surface openness, whereas SES defines the molecular boundary excluding solvent penetration, reflecting pocket compactness and atomic packing. Together, these parameters serve as complementary indicators of pocket size and flexibility. Consistent with the wider molecular weight range of CB1-bound ligands, the CB1 ligand binding pocket exhibits a larger SASA range (2,626–3,421 Å2) and mean ± standard deviation (SD) (3,113 ± 215 Å2) (Table 5), indicating that it is more adaptable for binding compounds. Classical cannabinoid ligands bound to CB1 receptors typically exhibit lower molecular weights and occupy more compact binding pockets, whereas diarylpyrazole chemotypes show higher SASA values, consistent with their bulkier scaffolds and the associated expansion of binding pocket expansion. In contrast, the CB2 ligand binding pocket displays a narrower SASA range (2,964–3,087 Å2) and mean ± SD (3,030 ± 43 Å2), reflecting a more compact and spatially constrained cavity. Accordingly, the relationship between ligand molecular weight and SASA is less pronounced for CB2, where the tighter pocket likely favors enhanced shape complementarity and more efficient hydrophobic burial for well-fitting ligands. This observation aligns with previous structure-activity analyses, where CB1 tolerates bulkier or branched substituents more readily than CB2, which exhibits a preference for smaller or more compact molecules. A similar trend was observed for the SES, where CB1 exhibited a larger SES range (2051–2,654 Å2) and SD (163 Å2) compared to CB2 (range: 2,429–2,674 Å2; SD: 88 Å2), suggesting greater conformational variability and a more flexible surface topology in CB1.

However, these interpretations should be cautiously considered, as they are based on a relatively small and imbalanced set of available structures. In particular, more ligand-bound CB1 complexes have been deposited in the PDB compared to CB2, which may contribute to the observed variability and limit the generalizability of these trends.

Although CB1 and CB2 binding pockets do not exhibit strikingly structural contrasts, subtle variations in pocket geometry, residue composition, and physicochemical environment can be strategically exploited to enhance subtype selectivity. In particular, ligand size, shape, and flexibility can be tuned to match binding pocket constraints, with CB1 generally accommodating bulkier and more flexible ligands, whereas CB2 favors smaller, more compact scaffolds. Targeted modifications at chemically distinct but positionally equivalent residues in CB1 and CB2, such as Leu1933.29 and Ile1103.29 and Leu3596.51/Val2616.51, can further promote differential binding and improve subtype selectivity. Achieving receptor selectivity typically involves not only optimizing interactions with the intended target, but also minimizing affinity for the alternate subtype, which may be influenced by differences in receptor conformational dynamics. Computational studies by Dutta et al. suggest that receptor activation mechanisms further contribute to selectivity. During activation, the CB1 binding pocket undergoes substantial conformational changes, partly driven by N-terminal fluctuations, whereas the CB2 pocket remains relatively stable. Consequently, CB2-selective ligands can engage multiple metastable states of CB2 while showing reduced affinity for corresponding CB1 states, thereby decreasing overall CB1 binding (Dutta and Shukla, 2023). These activation-dependent differences in binding pocket volume provide useful insights for improving draggability, and a better understanding of binding pocket dynamics may have broader implications for cannabinoid drug discovery. A careful balance of polar and hydrophobic ligand functional groups, informed by both structural and computational analyses, may therefore, provide a practical route toward developing highly selective CB1 or CB2 modulators.

The hydrophobic nature of these binding pockets may also contribute to favorable ligand binding through solvent-mediated thermodynamic effects. Water molecules confined within narrow, hydrophobic pockets are often poorly hydrogen-bonded and energetically unfavorable (Carey et al., 2000; Young et al., 2007). Displacement of these structured waters upon ligand binding is therefore thermodynamically advantageous, as it relieves unfavorable ligand–water and water–pocket interactions while allowing the released water molecules to form stronger hydrogen-bond networks in the bulk solvent (Schaeffer, 2008). This process contributes both enthalpically and entropically to binding affinity and helps explain the substantial gains often observed upon introduction of lipophilic substituents in confined binding environments.

Another promising strategy for achieving subtype selectivity is allosteric modulation. Unlike orthosteric ligands, allosteric modulators bind to topologically distinct and less conserved sites, enabling greater receptor subtype selectivity and offering opportunities to fine-tune signaling bias. Several putative allosteric binding sites have been identified in both receptors using computational/experimental techniques. In CB1, multiple allosteric regions have been identified across transmembrane and intracellular domains including the TM3–6–7 interface, a pocket between TM1/TM4, and sites within the N-terminal loop. In CB2, proposed allosteric sites are located adjacent to the orthosteric binding pocket and within transmembrane regions. Based on their effects on orthosteric ligand-driven signaling, allosteric modulators are broadly classified as positive or negative allosteric modulators, which enhance or attenuate receptor activity (Yang et al., 2022; Pandey et al., 2025; 2026). Representative chemical classes of allosteric modulators include indole-2-carboxamide derivatives, diaryl ureas, and 2-phenylindoles for CB1, whereas comparatively fewer classes, such as 2-oxo-pyridine-3-cycloheptanecarboxamide derivatives, have been identified for CB2 (Gado et al., 2019).

The CB1-bound ligand fenofibrate (PDB ID: 8K8J), although resolved within the orthosteric binding pocket, has been reported to exhibit allosteric modulatory effects at higher concentrations. This behavior has been attributed to potential bitopic binding, whereby the ligand engages in both orthosteric and allosteric regions, resulting in negative allosteric modulator activity. Notably, these modulatory effects are observed at concentrations exceeding those required for its primary orthosteric activity.

Future perspectives

CB1 and CB2 play vital roles in many physiological processes important for human health, survival, and recovery. Because their overall structures are highly similar, it is often difficult to design ligands that bind selectively to one receptor without affecting the other. Most cannabinoid ligands reported so far show only moderate selectivity, differing by a few-fold in potency between CB1 and CB2. Resolving these subtle selectivity differences requires comprehensive structural insight into both receptors. However, the current structurally available data poses limitations. Since CB1 was discovered earlier, most available three-dimensional structures represent CB1-bound complexes, while CB2 structures remain relatively few. This imbalance may bias comparative analyses and restrict conclusions about CB2-specific interactions. Expanding the CB2 structural repertoire and investigating receptor behavior within membrane environments may reveal conformations that favor selective ligand binding. Further research integrating evolutionary analysis, structural modeling, and dynamic simulations will be crucial to fully understand the mechanisms underlying receptor selectivity.

Moreover, CB1 is highly conserved across species; however CB2 has undergone more rapid evolutionary divergence, sharing only ∼81% sequence identity between human and rat receptors (McPartland et al., 2007). This divergence, particularly within the C-terminal and extracellular regions, may significantly influence ligand binding, signaling efficacy, and receptor regulation (Brown et al., 2002; Ashton et al., 2008). Accounting for these interspecies differences is, therefore, essential when translating ligand activity and therapeutic outcomes from preclinical models to humans. Future mutagenesis and binding studies focused on residues near the phytocannabinoid pocket could elucidate how subtle microenvironmental changes fine-tune ligand recognition, guiding the development of more selective cannabinoid receptor modulators.

While structural differences between CB1 and CB2 can be leveraged to improve subtype selectivity, limiting CNS exposure remains critical for minimizing adverse effects associated with CB1 activation. Recent advances in medicinal chemistry and structure-based design have demonstrated that tuning physicochemical properties such as polarity or charge can effectively restrict blood–brain barrier penetration and promote peripheral selectivity. Integrating these pharmacokinetic considerations with structural insights provides a complementary strategy for developing safer cannabinoid therapeutics with improved efficacy and reduced CNS liabilities.

Recently, deep learning and geometric modeling frameworks such as DiffDock, MaSIF, and PIGNet have emerged as powerful tools for predicting ligand affinity and selectivity (Krishnan et al., 2022; Moon et al., 2022; Brown, 2025). These models learn 3D geometric and physicochemical complementarity directly from structural data, enabling more accurate prediction of binding poses and affinity. Importantly, by evaluating surface complementarity or low-confidence interaction scores, such models may help identify ligands unlikely to bind effectively, an aspect particularly relevant for CB1 and CB2, whose active sites are highly similar and differ only in subtle physicochemical features that may influence selectivity.

CB1 and CB2 receptors remain promising targets for future drug discovery. Their closely related architecture, coupled with minor yet functionally significant differences in binding-pockets provide valuable opportunities to design ligands that precisely exploit these microvariations. With continued progress in structural biology, molecular modeling, and pharmacological studies, these receptors are poised to remain central to next-generation efforts aimed at achieving tissue-specific and functionally selective cannabinoid modulation.

In addition to receptor selectivity within the cannabinoid system, it is important to consider potential off-target interactions of these ligands with other receptor families. For example, certain cannabinoid-related compounds, such as CBD, have been reported to interact with the serotonergic system, including agonist activity at the 5-HT1A receptor. Such polypharmacological behavior may contribute to their observed biological effects and represents an important consideration for future studies aimed at fully characterizing ligand activity beyond CB1 and CB2.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. Check English errors.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Author disclaimer

This manuscript reflects the views of its authors and does not necessarily reflect those of the U.S. Food and Drug Administration. Any mention of commercial products is for clarification only and is not intended as approval, endorsement, or recommendation.

Footnotes

Edited by: Mohammad Khursheed Siddiqi, Virginia Commonwealth University, United States

Reviewed by: Pankaj Pandey, University of Mississippi, United States

Thomas Piccariello, Synthonics, Inc., United States

Statements

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The author HH declared that they were an editorial board member of Frontiers at the time of submission. This had no impact on the peer review process and the final decision.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This study was funded by the US Food and Drug Administration (FDA). This research was supported in part by an appointment to the Research Participation Program at the National Center for Toxicological Research (AV), administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and the U.S. Food and Drug Administration.

Author contributions

AV: Data curation, Formal Analysis, Visualization, Writing – original draft. JL: Data curation, Writing – original draft. WG: Visualization, Writing – original draft. TP: Writing – review and editing. HH: Conceptualization, Supervision, Writing – review and editing.

References

  1. Ahn K. H., Bertalovitz A. C., Mierke D. F., Kendall D. A. (2009). Dual role of the second extracellular loop of the cannabinoid receptor 1: ligand binding and receptor localization. Mol. Pharmacol. 76, 833 – 842. doi:10.1124/mol.109.057356
  2. Alqarni M., Myint K. Z., Tong Q., Yang P., Bartlow P., Wang L. (2014). Examining the critical roles of human CB2 receptor residues Valine 3.32 (113) and Leucine 5.41 (192) in ligand recognition and downstream signaling activities. Biochem. Biophys. Res. Commun. 452, 334 – 339. doi:10.1016/j.bbrc.2014.08.048
  3. Aronne L. J., Tonstad S., Moreno M., Gantz I., Erondu N., Suryawanshi S. (2010). A clinical trial assessing the safety and efficacy of taranabant, a CB1R inverse agonist, in obese and overweight patients: a high-dose study. Int. J. Obes. 34, 919 – 935. doi:10.1038/ijo.2010.21
  4. Ashton J., Wright J., McPartland J., Tyndall J. (2008). Cannabinoid CB1 and CB2 receptor ligand specificity and the development of CB2-Selective agonists. Curr. Med. Chem. 15, 1428 – 1443. doi:10.2174/092986708784567716
  5. Baldwin J. M. (1993). The probable arrangement of the helices in G protein-coupled receptors. EMBO J. 12, 1693 – 1703. doi:10.1002/j.1460-2075.1993.tb05814.x
  6. Ballesteros J. A., Weinstein H. (1995). “ [19] integrated methods for the construction of three-dimensional models and computational probing of structure-function relations in G protein-coupled receptors,” in Methods in Neurosciences (Elsevier), 366 – 428. doi:10.1016/S1043-9471(05)80049-7
  7. Bellocchio L., Lafenêtre P., Cannich A., Cota D., Puente N., Grandes P. (2010). Bimodal control of stimulated food intake by the endocannabinoid system. Nat. Neurosci. 13, 281 – 283. doi:10.1038/nn.2494
  8. Bow E. W., Rimoldi J. M. (2016). The structure–function relationships of classical cannabinoids: CB1/CB2 modulation. Perspect. Med. Chem. 8, 17. doi:10.4137/PMC.S32171
  9. Brown B. P. (2025). A generalizable deep learning framework for structure-based protein–ligand affinity ranking. Proc. Natl. Acad. Sci. 122, e2508998122. doi:10.1073/pnas.2508998122
  10. Brown S. M., Wager-Miller J., Mackie K. (2002). Cloning and molecular characterization of the rat CB2 cannabinoid receptor. Biochim. Biophys. Acta BBA - Gene Struct. Expr. 1576, 255 – 264. doi:10.1016/S0167-4781(02)00341-X
  11. Busquets Garcia A., Soria-Gomez E., Bellocchio L., Marsicano G. (2016). Cannabinoid receptor type-1: breaking the dogmas. F1000Research 5, 990. doi:10.12688/f1000research.8245.1
  12. Carey C., Cheng Y.-K., Rossky P. J. (2000). Hydration structure of the α-chymotrypsin substrate binding pocket: the impact of constrained geometry. Chem. Phys. 258, 415 – 425. doi:10.1016/S0301-0104(00)00178-6
  13. Centonze D., Bari M., Rossi S., Prosperetti C., Furlan R., Fezza F. (2007). The endocannabinoid system is dysregulated in multiple sclerosis and in experimental autoimmune encephalomyelitis. Brain 130, 2543 – 2553. doi:10.1093/brain/awm160
  14. Chin C., Lucas‐Lenard J., Abadji V., Kendall D. A. (1998). Ligand binding and modulation of cyclic AMP levels depend on the chemical nature of residue 192 of the human cannabinoid receptor 1. J. Neurochem. 70, 366 – 373. doi:10.1046/j.1471-4159.1998.70010366.x
  15. Chou S., Ranganath T., Fish K. N., Lewis D. A., Sweet R. A. (2022). Cell type specific cannabinoid CB1 receptor distribution across the human and non-human primate cortex. Sci. Rep. 12, 9605. doi:10.1038/s41598-022-13724-x
  16. Devane W. A., Dysarz F. A., Johnson M. R., Melvin L. S., Howlett A. C. (1988). Determination and characterization of a cannabinoid receptor in rat brain. Mol. Pharmacol. 34, 605 – 613. doi:10.1016/S0026-895X(25)09876-1
  17. Di Marzo V., Bisogno T., De Petrocellis L., Brandi I., Jefferson R. G., Winckler R. L. (2001). Highly selective CB1 cannabinoid receptor ligands and novel CB1/VR1 vanilloid receptor “hybrid” ligands. Biochem. Biophys. Res. Commun. 281, 444 – 451. doi:10.1006/bbrc.2001.4354
  18. Díaz Ó., Dalton J. A. R., Giraldo J. (2019). Revealing the mechanism of agonist-mediated cannabinoid receptor 1 (CB1) activation and phospholipid-mediated allosteric modulation. J. Med. Chem. 62, 5638 – 5654. doi:10.1021/acs.jmedchem.9b00612
  19. Dutta S., Shukla D. (2023). Distinct activation mechanisms regulate subtype selectivity of cannabinoid receptors. Commun. Biol. 6, 485. doi:10.1038/s42003-023-04868-1
  20. Fay J. F., Farrens D. L. (2013). The membrane proximal region of the cannabinoid receptor CB 1 N-Terminus can allosterically modulate ligand affinity. Biochemistry 52, 8286 – 8294. doi:10.1021/bi400842k
  21. Feng Z., Alqarni M. H., Yang P., Tong Q., Chowdhury A., Wang L. (2014). Modeling, molecular dynamics simulation, and mutation validation for structure of cannabinoid receptor 2 based on known crystal structures of GPCRs. J. Chem. Inf. Model. 54, 2483 – 2499. doi:10.1021/ci5002718
  22. Gado F., Meini S., Bertini S., Digiacomo M., Macchia M., Manera C. (2019). Allosteric modulators targeting cannabinoid cb1 and cb2 receptors: implications for drug discovery. Future Med. Chem. 11, 2019 – 2037. doi:10.4155/fmc-2019-0005
  23. Han Q.-W., Yuan Y.-H., Chen N.-H. (2020). The therapeutic role of cannabinoid receptors and its agonists or antagonists in Parkinson’s disease. Prog. Neuropsychopharmacol. Biol. Psychiatry 96, 109745. doi:10.1016/j.pnpbp.2019.109745
  24. Hanuš L., Breuer A., Tchilibon S., Shiloah S., Goldenberg D., Horowitz M. (1999). HU-308: a specific agonist for CB 2, a peripheral cannabinoid receptor. Proc. Natl. Acad. Sci. 96, 14228 – 14233. doi:10.1073/pnas.96.25.14228
  25. Hillard C. J., Manna S., Greenberg M. J., DiCamelli R., Ross R. A., Stevenson L. A. (1999). Synthesis and characterization of potent and selective agonists of the neuronal cannabinoid receptor (CB1). J. Pharmacol. Exp. Ther. 289, 1427 – 1433. doi:10.1016/S0022-3565(24)38289-8
  26. Hourani W., Alexander S. P. H. (2018). Cannabinoid ligands, receptors and enzymes: pharmacological tools and therapeutic potential. Brain Neurosci. Adv. 2, 2398212818783908. doi:10.1177/2398212818783908
  27. Hua T., Vemuri K., Pu M., Qu L., Han G. W., Wu Y. (2016). Crystal structure of the human cannabinoid receptor CB1. Cell. 167, 750 – 762.e14. doi:10.1016/j.cell.2016.10.004
  28. Hua T., Vemuri K., Nikas S. P., Wu Y., Qu L., Pu M. (2025). Crystal structures of agonist-bound human cannabinoid receptor CB1. Nature 646, 754 – 758. doi:10.1038/s41586-025-09454-5
  29. Huffman J. W., Yu S., Showalter V., Abood M. E., Wiley J. L., Compton D. R. (1996). Synthesis and pharmacology of a very potent cannabinoid lacking a phenolic hydroxyl with high affinity for the CB2 receptor. J. Med. Chem. 39, 3875 – 3877. doi:10.1021/jm960394y
  30. Hurst D. P., Lynch D. L., Barnett-Norris J., Hyatt S. M., Seltzman H. H., Zhong M. (2002). N-(Piperidin-1-yl)-5-(4-chlorophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide (SR141716A) interaction with LYS 3.28(192) is crucial for its inverse agonism at the cannabinoid CB1 receptor. Mol. Pharmacol. 62, 1274 – 1287. doi:10.1124/mol.62.6.1274
  31. Ibsen M. S., Connor M., Glass M. (2017). Cannabinoid CB 1 and CB 2 receptor signaling and bias. Cannabis Cannabinoid Res. 2, 48 – 60. doi:10.1089/can.2016.0037
  32. Ishiguro H., Kibret B. G., Horiuchi Y., Onaivi E. S. (2022). Potential role of cannabinoid type 2 receptors in neuropsychiatric and neurodegenerative disorders. Front. Psychiatry 13, 828895. doi:10.3389/fpsyt.2022.828895
  33. Khan M. G., Hussain S. H. A., Alkhayl F. F. A., Ahsan M., Ridha-Salman H. (2025). Cannabinoids in neuropathic pain treatment: pharmacological insights and clinical outcomes from recent trials. Naunyn. Schmiedeb. Arch. Pharmacol. 398, 13087 – 13107. doi:10.1007/s00210-025-04134-7
  34. Krishna Kumar K., Shalev-Benami M., Robertson M. J., Hu H., Banister S. D., Hollingsworth S. A. (2019). Structure of a signaling cannabinoid receptor 1-G protein complex. Cell. 176, 448 – 458.e12. doi:10.1016/j.cell.2018.11.040
  35. Krishnan S. R., Bung N., Vangala S. R., Srinivasan R., Bulusu G., Roy A. (2022). De novo Structure-Based Drug Design Using Deep Learning. J. Chem. Inf. Model. 62, 5100 – 5109. doi:10.1021/acs.jcim.1c01319
  36. Li X., Hua T., Vemuri K., Ho J.-H., Wu Y., Wu L. (2019). Crystal structure of the human cannabinoid receptor CB2. Cell. 176, 459 – 467.e13. doi:10.1016/j.cell.2018.12.011
  37. Lin S., Khanolkar A. D., Fan P., Goutopoulos A., Qin C., Papahadjis D. (1998). Novel analogues of arachidonylethanolamide (anandamide): affinities for the CB1 and CB2 cannabinoid receptors and metabolic stability. J. Med. Chem. 41, 5353 – 5361. doi:10.1021/jm970257g
  38. Llinas Del Torrent C., Raïch I., Gonzalez A., Casajuana-Martin N., Lillo J., Rebassa J. B. (2023). The Leu/Val 6.51 side chain of cannabinoid receptors regulates the binding mode of the alkyl chain of Δ 9 -Tetrahydrocannabinol. J. Chem. Inf. Model. 63, 5927 – 5935. doi:10.1021/acs.jcim.3c01054
  39. Lu H.-C., Mackie K. (2016). An introduction to the endogenous cannabinoid system. Biol. Psychiatry 79, 516 – 525. doi:10.1016/j.biopsych.2015.07.028
  40. Lu H.-C., Mackie K. (2021). Review of the endocannabinoid system. Biol. Psychiatry Cogn. Neurosci. Neuroimaging 6, 607 – 615. doi:10.1016/j.bpsc.2020.07.016
  41. Lu D., Guo J., Duclos R. I., Bowman A. L., Makriyannis A. (2008). Bornyl- and Isobornyl-Δ 8 -tetrahydrocannabinols: a novel class of cannabinergic ligands. J. Med. Chem. 51, 6393 – 6399. doi:10.1021/jm8005299
  42. Lundstrom K. (2009). “ An overview on GPCRs and drug discovery: structure-based drug design and structural biology on GPCRs,” in G Protein-Coupled Receptors in Drug Discovery. Editor Leifert W. R. (Totowa, NJ: Humana Press), 51 – 66. doi:10.1007/978-1-60327-317-6_4
  43. Maccarrone M., Falciglia K., Di Rienzo M., Finazzi-Agrò A. (2002). Endocannabinoids, hormone-cytokine networks and human fertility. Prostagl. Leukot. Essent. Fat. Acids PLEFA 66, 309 – 317. doi:10.1054/plef.2001.0354
  44. Maroof N., Pardon M. C., Kendall D. A. (2013). Endocannabinoid signalling in Alzheimer’s disease. Biochem. Soc. Trans. 41, 1583 – 1587. doi:10.1042/BST20130140
  45. Marsicano G., Lutz B. (1999). Expression of the cannabinoid receptor CB1 in distinct neuronal subpopulations in the adult mouse forebrain. Eur. J. Neurosci. 11, 4213 – 4225. doi:10.1046/j.1460-9568.1999.00847.x
  46. Marzo V. D., Bifulco M., Petrocellis L. D. (2004). The endocannabinoid system and its therapeutic exploitation. Nat. Rev. Drug Discov. 3, 771 – 784. doi:10.1038/nrd1495
  47. Matsuda L. A., Lolait S. J., Brownstein M. J., Young A. C., Bonner T. I. (1990). Structure of a cannabinoid receptor and functional expression of the cloned cDNA. Nature 346, 561 – 564. doi:10.1038/346561a0
  48. McAllister S. D., Tao Q., Barnett-Norris J., Buehner K., Hurst D. P., Guarnieri F. (2002). A critical role for a tyrosine residue in the cannabinoid receptors for ligand recognition. Biochem. Pharmacol. 63, 2121 – 2136. doi:10.1016/S0006-2952(02)01031-6
  49. McAllister S. D., Rizvi G., Anavi-Goffer S., Hurst D. P., Barnett-Norris J., Lynch D. L. (2003). An aromatic microdomain at the cannabinoid CB 1 receptor constitutes an agonist/inverse agonist binding region. J. Med. Chem. 46, 5139 – 5152. doi:10.1021/jm0302647
  50. McAllister S. D., Hurst D. P., Barnett-Norris J., Lynch D., Reggio P. H., Abood M. E. (2004). Structural mimicry in class A G protein-coupled receptor rotamer toggle switches: the importance of the F3.36(201)/W6.48(357) interaction in cannabinoid CB1 receptor activation. J. Biol. Chem. 279, 48024 – 48037. doi:10.1074/jbc.M406648200
  51. McPartland J. M., Norris R. W., Kilpatrick C. W. (2007). Coevolution between cannabinoid receptors and endocannabinoid ligands. Gene 397, 126 – 135. doi:10.1016/j.gene.2007.04.017
  52. Mechoulam R., Parker L. A. (2013). The endocannabinoid system and the brain. Annu. Rev. Psychol. 64, 21 – 47. doi:10.1146/annurev-psych-113011-143739
  53. Montero C., Campillo N. E., Goya P., Páez J. A. (2005). Homology models of the cannabinoid CB1 and CB2 receptors. A docking analysis study. Eur. J. Med. Chem. 40, 75 – 83. doi:10.1016/j.ejmech.2004.10.002
  54. Moon S., Zhung W., Yang S., Lim J., Kim W. Y. (2022). PIGNet: a physics-informed deep learning model toward generalized drug–target interaction predictions. Chem. Sci. 13, 3661 – 3673. doi:10.1039/D1SC06946B
  55. Moreira F. A., Crippa J. A. S. (2009). The psychiatric side-effects of rimonabant. Rev. Bras. Psiquiatr. 31, 145 – 153. doi:10.1590/S1516-44462009000200012
  56. Munro S., Thomas K. L., Abu-Shaar M. (1993). Molecular characterization of a peripheral receptor for cannabinoids. Nature 365, 61 – 65. doi:10.1038/365061a0
  57. Murphy J. W., Kendall D. A. (2003). Integrity of extracellular loop 1 of the human cannabinoid receptor 1 is critical for high-affinity binding of the ligand CP 55,940 but not SR 141716A. Biochem. Pharmacol. 65, 1623 – 1631. doi:10.1016/S0006-2952(03)00155-2
  58. Navarro G., Gonzalez A., Sánchez-Morales A., Casajuana-Martin N., Gómez-Ventura M., Cordomí A. (2021). Design of negative and positive allosteric modulators of the cannabinoid CB 2 receptor derived from the natural product cannabidiol. J. Med. Chem. 64, 9354 – 9364. doi:10.1021/acs.jmedchem.1c00561
  59. Oddi S., Fiorenza M. T., Maccarrone M. (2023). Endocannabinoid signaling in adult hippocampal neurogenesis: a mechanistic and integrated perspective. Prog. Lipid Res. 91, 101239. doi:10.1016/j.plipres.2023.101239
  60. Pan X., Ikeda S. R., Lewis D. L. (1998). SR 141716A acts as an inverse agonist to increase neuronal voltage-dependent Ca2+ currents by reversal of Tonic CB1 cannabinoid receptor activity. Mol. Pharmacol. 54, 1064 – 1072. doi:10.1124/mol.54.6.1064
  61. Pandey P., Zagzoog A., Laprairie R. B., Neal W. M., Doerksen R. J., Chittiboyina A. G. (2025). Determination of the negative allosteric binding site of cannabidiol at the CB1 receptor: a combined computational and site-directed mutagenesis study. ACS Chem. Neurosci. 16, 311 – 328. doi:10.1021/acschemneuro.4c00343
  62. Pandey P., Zagzoog A. H., Zaharia T., Chatterjee S., Khan I. A., Paris J. J. (2026). Cannabidiphorol (CBDP) acts as a negative allosteric modulator at two distinct sites of cannabinoid receptor 1. Commun. Chem. [in press]. doi:10.1038/s42004-026-01990-z
  63. Pertwee R. G., Howlett A. C., Abood M. E., Alexander S. P. H., Di Marzo V., Elphick M. R. (2010). International union of basic and clinical pharmacology. LXXIX. Cannabinoid receptors and their ligands: beyond CB1 and CB2. Pharmacol. Rev. 62, 588 – 631. doi:10.1124/pr.110.003004
  64. Pinto L., Capasso R., Di Carlo G., Izzo A. A. (2002). Endocannabinoids and the gut. Prostagl. Leukot. Essent. Fat. Acids PLEFA 66, 333 – 341. doi:10.1054/plef.2001.0345
  65. Pisanti S., Picardi P., D’Alessandro A., Laezza C., Bifulco M. (2013). The endocannabinoid signaling system in cancer. Trends Pharmacol. Sci. 34, 273 – 282. doi:10.1016/j.tips.2013.03.003
  66. Poleszak E., Wośko S., Sławińska K., Wyska E., Szopa A., Sobczyński J. (2020). Ligands of the CB2 cannabinoid receptors augment activity of the conventional antidepressant drugs in the behavioural tests in mice. Behav. Brain Res. 378, 112297. doi:10.1016/j.bbr.2019.112297
  67. Proietto J., Rissanen A., Harp J. B., Erondu N., Yu Q., Suryawanshi S. (2010). A clinical trial assessing the safety and efficacy of the CB1R inverse agonist taranabant in obese and overweight patients: low-dose study. Int. J. Obes. 34, 1243 – 1254. doi:10.1038/ijo.2010.38
  68. Reggio P. (2010). Endocannabinoid binding to the cannabinoid receptors: what is known and what remains unknown. Curr. Med. Chem. 17, 1468 – 1486. doi:10.2174/092986710790980005
  69. Rhee M. H. (2002). Functional role of serine residues of transmembrane dopamin VII in signal transduction of CB2 cannabinoid receptor. J. Vet. Sci. 3, 185 – 191. doi:10.4142/jvs.2002.3.3.185
  70. Ruehle S., Rey A. A., Remmers F., Lutz B. (2012). The endocannabinoid system in anxiety, fear memory and habituation. J. Psychopharmacol. (Oxf.) 26, 23 – 39. doi:10.1177/0269881111408958
  71. Ruiz De Martín Esteban S., Benito-Cuesta I., Terradillos I., Martínez-Relimpio A. M., Arnanz M. A., Ruiz-Pérez G. (2022). Cannabinoid CB2 receptors modulate microglia function and amyloid dynamics in a mouse model of Alzheimer’s disease. Front. Pharmacol. 13, 841766. doi:10.3389/fphar.2022.841766
  72. Schaeffer L. (2008). “ The role of functional groups in drug–receptor interactions,” in The Practice of Medicinal Chemistry (Elsevier), 359 – 378. doi:10.1016/B978-0-12-417205-0.00014-6
  73. Shao Z., Yin J., Chapman K., Grzemska M., Clark L., Wang J. (2016). High-resolution crystal structure of the human CB1 cannabinoid receptor. Nature 540, 602 – 606. doi:10.1038/nature20613
  74. Sharkey K. A. (2022). “ Proposed mechanisms of cannabinoid hyperemesis syndrome—how can cannabinoid pathways both relieve and cause vomiting?,” in Cyclic Vomiting Syndrome and Cannabinoid Hyperemesis (Elsevier), 175 – 200. doi:10.1016/B978-0-12-821597-5.00003-4
  75. Shim J.-Y., Bertalovitz A. C., Kendall D. A. (2011). Identification of essential cannabinoid-binding domains. J. Biol. Chem. 286, 33422 – 33435. doi:10.1074/jbc.M111.261651
  76. Shim J.-Y., Bertalovitz A. C., Kendall D. A. (2012). Probing the interaction of SR141716A with the CB1 receptor. J. Biol. Chem. 287, 38741 – 38754. doi:10.1074/jbc.M112.390955
  77. Sitkoff D. F., Lee N., Ellsworth B. A., Huang Q., Kang L., Baska R. (2011). Cannabinoid CB1 receptor ligand binding and function examined through mutagenesis studies of F200 and S383. Eur. J. Pharmacol. 651, 9 – 17. doi:10.1016/j.ejphar.2010.10.056
  78. Song Z. H., Bonner T. I. (1996). A lysine residue of the cannabinoid receptor is critical for receptor recognition by several agonists but not WIN55212-2. Mol. Pharmacol. 49, 891 – 896. doi:10.1016/S0026-895X(25)09082-0
  79. Thakur G. A., Duclos R. I., Makriyannis A. (2005). Natural cannabinoids: templates for drug discovery. Life Sci. 78, 454 – 466. doi:10.1016/j.lfs.2005.09.014
  80. Thakur G., Nikas S., Makriyannis A. (2005). CB1 cannabinoid receptor ligands. Mini-Rev. Med. Chem. 5, 631 – 640. doi:10.2174/1389557054368772
  81. Valenzano K. J., Tafesse L., Lee G., Harrison J. E., Boulet J. M., Gottshall S. L. (2005). Pharmacological and pharmacokinetic characterization of the cannabinoid receptor 2 agonist, GW405833, utilizing rodent models of acute and chronic pain, anxiety, ataxia and catalepsy. Neuropharmacology 48, 658 – 672. doi:10.1016/j.neuropharm.2004.12.008
  82. Verty A. N. A., Stefanidis A., McAinch A. J., Hryciw D. H., Oldfield B. (2015). Anti-obesity effect of the CB2 receptor agonist JWH-015 in diet-induced Obese mice. PLOS ONE 10, e0140592. doi:10.1371/journal.pone.0140592
  83. Woolley M. J., Conner A. C. (2017). Understanding the common themes and diverse roles of the second extracellular loop (ECL2) of the GPCR super-family. Mol. Cell. Endocrinol. 449, 3 – 11. doi:10.1016/j.mce.2016.11.023
  84. Wootten D., Christopoulos A., Sexton P. M. (2013). Emerging paradigms in GPCR allostery: implications for drug discovery. Nat. Rev. Drug Discov. 12, 630 – 644. doi:10.1038/nrd4052
  85. Xing C., Zhuang Y., Xu T.-H., Feng Z., Zhou X. E., Chen M. (2020). Cryo-EM structure of the human cannabinoid receptor CB2-Gi signaling complex. Cell. 180, 645 – 654.e13. doi:10.1016/j.cell.2020.01.007
  86. Yang J.-F., Williams A. H., Penthala N. R., Prather P. L., Crooks P. A., Zhan C.-G. (2020). Binding Modes and Selectivity of Cannabinoid 1 (CB1) and Cannabinoid 2 (CB2) Receptor Ligands. ACS Chem. Neurosci. 11, 3455 – 3463. doi:10.1021/acschemneuro.0c00551
  87. Yang X., Wang X., Xu Z., Wu C., Zhou Y., Wang Y. (2022). Molecular mechanism of allosteric modulation for the cannabinoid receptor CB1. Nat. Chem. Biol. 18, 831 – 840. doi:10.1038/s41589-022-01038-y
  88. Young T., Abel R., Kim B., Berne B. J., Friesner R. A. (2007). Motifs for molecular recognition exploiting hydrophobic enclosure in protein–ligand binding. Proc. Natl. Acad. Sci. 104, 808 – 813. doi:10.1073/pnas.0610202104
  89. Zhang Y., Xie Z., Wang L., Schreiter B., Lazo J. S., Gertsch J. (2011). Mutagenesis and computer modeling studies of a GPCR conserved residue W5.43(194) in ligand recognition and signal transduction for CB2 receptor. Int. Immunopharmacol. 11, 1303 – 1310. doi:10.1016/j.intimp.2011.04.013
  90. Zieglgänsberger W., Brenneisen R., Berthele A., Wotjak C. T., Bandelow B., Tölle T. R. (2022). Chronic pain and the endocannabinoid system: smart lipids – a novel therapeutic option? Med. Cannabis Cannabinoids 5, 61 – 75. doi:10.1159/000522432

Citation

Varghese, A., et al. (2026). Structural insights into cannabinoid receptors CB1 and CB2: determinants of ligand selectivity. Frontiers in Chemical Biology, Volume 5. https://doi.org/
Received
Accepted
Published
Volume
5 - 2026

Copyright

Copyright © 2026 Varghese, Liu, Guo, Patterson and Hong.

License: https://creativecommons.org/licenses/by/4.0/

*Correspondence: Huixiao Hong

Article metrics

Views
0
Downloads
0
Citations
0
Shares
0

Related articles

Original ResearchNanoprobe XRF analysis reveals the intracellular distribution of organometallic rhenium complexes with different directing ligandC. Rumancev, S. K. Vellas, A. Gräfenstein, D. Obitz, J. Garrevoet, G. Falkenberg, N. Metzler-Nolte, A. RosenhahnPerspectiveAdvances in enhanced sampling simulations of G-protein-coupled receptorsKeya Joshi, Victor Ayo Adediwura, Yinglong MiaoOriginal ResearchA bifunctional carborane-quinazoline combines tumor growth inhibition and boron delivery for boron neutron capture therapy in melanoma and non-small cell lung cancer modelsMaría Alejandra Dagrosa, Marina Carpano, Antonella Pastini, Estanislao Pavlik, Andrés Bechis, María Silvina Olivera, Pablo Vallecorsa, Belén Dávila, Luciano Parodi, Esther Yardena Salem, María Sol Espain, Agustina Mariana Portu, Laura Todaro, Alejandro Urtreger, Pablo Menendez, Hugo Cerecetto, Marcos Couto