Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Mar 14.
Published in final edited form as: Eur J Pharmacol. 2024 Dec 10;987:177192. doi: 10.1016/j.ejphar.2024.177192

Insurmountable antagonism of human mu opioid receptors by buprenorphine is due to hemi-equilibrium

Michael J Wedemeyer 1, Teresa S Chavera 1, Kelly A Berg 1, William P Clarke 1,*
PMCID: PMC12984539  NIHMSID: NIHMS2149387  PMID: 39667424

Abstract

Buprenorphine is an FDA approved drug for the treatment of opioid use disorder and is a long-lasting, low efficacy (partial) agonist of the μ opioid receptor. As a partial agonist, buprenorphine can act as either an agonist or an antagonist depending on the efficiency of the cellular signaling system. Here we investigated the antagonist properties of buprenorphine using a genetically-encoded biosensor to monitor cAMP levels in real time in HEK293 cells expressing a relatively low density of the human μ receptor.

Pre-treatment of cells with buprenorphine decreased the maximal response (Emax) of the μ receptor agonists DAMGO and fentanyl, consistent with expectations of a long-lasting partial agonist in these cells. However, buprenorphine pretreatment also reduced the potency of these agonists. In a Schild analysis, the reduction in Emax saturated with increasing concentrations of buprenorphine, while the reduction in potency did not. This unexpected behavior of buprenorphine on potency was mediated through the orthosteric site, as pretreatment with the opioid receptor antagonist naloxone prevented buprenorphine antagonism. Computation simulations using a model of hemi-equilibrium indicated that buprenorphine’s slow receptor dissociation could account for both effects of antagonism. In support of this kinetic explanation, buprenorphine antagonism became surmountable with additional agonist incubation time.

In this system, buprenorphine appeared to behave both as a competitive and non-competitive (insurmountable) antagonist due to its slow dissociation resulting in partial re-equilibration (hemi-equilibrium). As fast-acting biosensors become more prevalent, it will be important to consider kinetically-mediated effects such as hemi-equilibrium when characterizing the pharmacological properties of antagonists.

Keywords: Buprenorphine, Mu opioid receptor, Pain, Substance use disorder, Opioids, Hemi-equilibrium

1. Introduction

Buprenorphine is an FDA (Food and Drug Administration) approved drug for the treatment of opioid use disorder (OUD) through its use as an opioid substitute (Gowing et al., 2017; Larochelle et al., 2018; Shulman et al., 2019). Buprenorphine binds the μ opioid receptor with high affinity, yet has relatively low intrinsic efficacy (Butler, 2013; Virk et al., 2009). Acting as a low-efficacy agonist, buprenorphine provides relief from withdrawal symptoms and cravings for patients suffering from OUD and promotes reduced opioid use. Additionally, buprenorphine is a desirable analgesic drug with a reduced potential for toxicity and overdose due to weak activation of the μ receptor (Davis et al., 2018; Gress et al., 2020; Hayes et al., 2008; Pergolizzi et al., 2008).

As a result of its weak agonist efficacy (partial agonist), buprenorphine can also behave as an antagonist in cells where μ receptor expression or receptor-effector coupling efficiency is low (Cowan et al., 1977; Kenakin, 2019a; Virk et al., 2009). When receptor density at the cell surface is high, buprenorphine measurably activates μ receptor dependent Gi signaling, which produces analgesia and also reduces cellular cAMP (cyclic adenosine monophosphate) levels (Ehrlich et al., 2019; Khanna and Pillarisetti, 2015; Kuo et al., 2020). However, when receptor density is low, buprenorphine-mediated Gi activation is insufficient to produce these effects. Thus, buprenorphine can act as an antagonist for μ receptor agonists with higher intrinsic efficacy, such as enkephalins, morphine, or DAMGO ([D-Ala2, N-MePhe4, Gly-ol]-enkephalin) (Cowan et al., 1977; Lewis, 1985; Virk et al., 2009). This phenomena of partial agonism underlies the seemingly conflicting nature of buprenorphine in the literature, where it has separately been reported to provide analgesia equivalent to full agonists or to be a μ receptor antagonist capable of precipitating withdrawal symptoms (Greenwald et al., 2003; Lutfy and Cowan, 2005; Spadaro et al., 2022). Many studies such as these have characterized the pharmacology of buprenorphine in vivo. However the presence of the δ-, κ-, and nociceptin opioid receptors, to which buprenorphine also binds, can confound interpretability (Lutfy et al., 2003; Wang et al., 2015). Relatively little is known about buprenorphine’s pharmacological profile when acting as an antagonist at μ receptors.

Another remarkable property of buprenorphine is its slow dissociation from the μ receptor, which has previously been established via radioligand binding in rodent brain (Boas and Villiger, 1985; Villiger and Taylor, 1981). This trait reportedly contributes to buprenorphine’s lower abuse liability and reduced withdrawal symptoms after cessation (Tzschentke, 2002). A recent study (Bidlack et al., 2018) quantified the binding kinetics of buprenorphine at the μ receptor in vitro and measured the dissociation T1/2 value to be between 200 min or over 23 h depending on buffer conditions. This slow dissociation from the μ receptor was not observed with binding to the δ or κ opioid receptors.

In tandem with its low intrinsic efficacy, buprenorphine’s slow dissociation may potentially result in long-term insurmountable (i.e. pseudo-irreversible) antagonism (Walker et al., 1995). Such antagonism would reduce the maximal response of μ receptor agonists. This situation has been postulated to occur in patients who are treated with buprenorphine for opioid maintenance therapy and require subsequent acute pain management (Buresh et al., 2020; Huang et al., 2014; Mccormick et al., 2013). In this study we characterize buprenorphine antagonism at the human μ receptor expressed in HEK293 cells at a relatively low density (i.e. without receptor reserve).

2. Methods

2.1. Chemicals and reagents

Buprenorphine (#B9275), Forskolin (#F6886), DAMGO (#E7384), and naloxone (#N7758) were purchased from Sigma-Aldrich. HEK-293 cells were from American Type Culture Collection (ATCC, # CRC-1573). Thermo Fisher Scientific supplied the minimum essential medium (αMEM, #12561056), lipofectamine (#L3000001), CO2 independent media (#18045088), and Geneticin (G418, #11811023). The mammalian expression vector encoding the human μ opioid receptor was obtained from the cDNA Resource Center (cDNA.org). The heat-inactivated horse serum was from GeminiBio (# 100–508). The D-Luciferin was purchased from GoldBio (#eLUCK), GloSensor-22F was from Promega (#E2301), and Hygromycin B was from Roache (#843555).

2.2. Cell culture

HEK-293 cells stably expressing the cAMP biosensor, GloSensor-22F, (Promega) and the human μ receptor (Zamora et al., 2021) were maintained with αMEM containing 10% heat inactivated horse serum, 50 μg/ml Hygromycin B, and 300 μg/ml G418 at 37°C and 5% CO2. These cells express a low level of the μ receptor such that receptor reserve for the high efficacy agonists, DAMGO and fentanyl, is not present. The lack of receptor reserve for DAMGO and fentanyl was confirmed by the lack of a dextral shift in the agonist concentration-response curves following reduction of μ receptor density by treatment with the irreversible antagonist, β-funaltrexamine (β-FNA) (Zamora et al., 2021).

2.3. GloSensor Gi/o-mediated cAMP inhibition assay

Cellular levels of cAMP were measured using the genetically-encoded biosensor, GloSensor, following manufacturer (Promega) instructions. In brief, 40,000 cells per well were seeded 48 h before the assay into poly-D-lysine coated white-walled 96-well plates with clear bottoms. On the day of the assay, the media was replaced with CO2 independent media (CIM) containing 450 μg/ml of D-Luciferin and 10% heat-inactivated horse serum. After 2 h of incubation in the dark at 30°C, bioluminescence was measured using FluoStar Omega microplate reader (BMG Labtech) with internal temperature set to 30 °C.

To determine opioid agonist responses, baseline luminescence measurements were collected once per minute for 5 min, followed by addition of the opioid agonist and another 5 min of data collection. Then, forskolin (FSK, 10 μM) was added and luminescence was recorded for 30 min. An example trace of the raw data is included in Supplemental Fig. 1. When used, antagonists were added with D-Luciferin and incubated in the dark at 30°C for 2 h, or as otherwise indicated.

The data shown in Fig. 1 were generated by addition of vehicle (CIM) or 10 nM buprenorphine to the cells 2 h prior to the assay. After a baseline reading at the start of the assay, fentanyl was added to corresponding wells. After another 5 min read, FSK was added and the plate was read for 30 min. The peak values of the timecourse data were normalized to FSK stimulation (100%) and plotted in Fig. 1 to construct the concentration response curve.

Fig. 1. Buprenorphine reduces both efficacy and potency of fentanyl.

Fig. 1.

Open in a new tab

HEK293 cells stably expressing the human μ receptor were pretreated with vehicle or buprenorphine (10 nM) for 2 h. The amount of cAMP in these cells was measured after the addition of fentanyl and stimulation with forskolin to activate adenylyl cyclase. With vehicle pretreatment (black circles), fentanyl reduces cAMP in a concentration dependent manner. Buprenorphine pretreatment (blue squares) results in both a rightward shift (reduced potency) and a reduced Emax (reduced efficacy). Data points represent mean ± SEM from at least six independent experiments (n = 6). Note: in this and subsequent figures, the SEM may be within the size of the symbol.

The data shown in Fig. 2 were generated following the same protocol, except with the μ receptor agonist, DAMGO, and pretreatment with multiple concentrations of buprenorphine. In addition to the plotted concentration response curve, the maximal DAMGO response and the log of the (dose ratio −1) were plotted for Schild analysis.

Fig. 2. Schild analysis reveals both competitive and non-competitive properties of buprenorphine.

Fig. 2.

Open in a new tab

HEK293 cells stably expressing the human μ receptor were pretreated with vehicle or a set concentration of buprenorphine (0.1–1000 nM) for 2 h. (A) The amount of cAMP was measured after addition of DAMGO and forskolin. The rightward shift of the DAMGO concentration response curve was buprenorphine concentration-dependent. (B) The Schild regression analysis of DAMGO potency shifts caused by buprenorphine is linear with a slope of 1.07. (C) The maximum reduction of cAMP by DAMGO is shown with increasing concentrations of buprenorphine. Buprenorphine reduces the maximal effect of DAMGO in a concentration dependent manner, until a plateau is reached at 10 nM. Data points represent mean ± SEM from at least four independent experiments (n = 4).

The data shown in Fig. 6 (and Supplemental Fig. 3) were generated with a modified protocol. Various concentrations of DAMGO were added to cells simultaneously with vehicle or 10 nM buprenorphine. After 2 or 4 h of incubation in the dark at 30°C, bioluminescence was measured using the standard protocol without washout. The peak vales of the timecourse data were normalized to both FSK (100%) and the maximal response of DAMGO (0%) as extended DAMGO exposure reduced the maximal response.

Fig. 6. Buprenorphine antagonism becomes surmountable with increased time of incubation with agonist.

Fig. 6.

Open in a new tab

Buprenorphine (10 nM) and DAMGO were simultaneously added to cells 2 or 4 h prior to a 5-min baseline and 30-min read of FSK-stimulated luminescence. Data shown has been normalized to DAMGO’s maximal effect at a time-matched vehicle control to account for receptor desensitization. After 2 h of co-incubation, buprenorphine antagonism was not surmountable (* Emax difference to vehicle, p = 0.0315). With 4 h of co-incubation, buprenorphine no longer reduced the maximal effect of DAMGO (Emax difference, p = 0.9076). Significant difference in Emax was determined by one-way ANOVA with Dunnett multiple comparisons test (F (2, 16) = 4.018, p = 0.039). Data points represent mean ± SEM from at least six independent experiments (n = 6).

2.4. Wash and naloxone protection

The data shown in Fig. 3 were generated by pretreatment of vehicle, 10 nM buprenorphine, or 100 nM naloxone for 2 h with or without subsequent washout. To washout opioid antagonists, media was removed after antagonist preincubation by gently flicking the plate. Fresh CO2 independent media was gently added along the well edge on a 37°C hot plate. This process was repeated for a total of at least three times over a period of 15 min. After the final wash, media was replaced with CO2 independent media containing 450 μg/ml of D-Luciferin and 10% heat-inactivated horse serum. The plate was placed in the plate reader at 30°C for 10 min before opioid agonist and FSK-mediated bioluminescence measurements were recorded as described above.

Fig. 3. Buprenorphine antagonism is partially reversed via washout.

Fig. 3.

Open in a new tab

HEK293 cells stably expressing the human μ receptor were treated with buprenorphine (10 nM) (A) or naloxone (100 nM) (B) for 2 h then rigorously washed prior to addition of forskolin and DAMGO. Buprenorphine antagonism was resistant to washout, with partial recovery of DAMGO potency but not efficacy. Naloxone pretreatment reduces the potency of DAMGO, but does not affect the maximal effect of DAMGO. This wash paradigm was effective at reversing naloxone antagonism. Data points represent mean ± SEM from at least four independent experiments (n = 4).

To protect the orthosteric site from occupancy by buprenorphine (Fig. 4), cells were incubated with vehicle or naloxone (10 μM; >10,000 × Ki) for 15 min in the dark at 30°C before the addition of vehicle or buprenorphine (10 nM). After 2 h, bioluminescence was recorded either in the presence of naloxone, or after washout as described above.

Fig. 4. Protection of the orthosteric pocket with naloxone prevents buprenorphine antagonism.

Fig. 4.

Open in a new tab

HEK293 cells stably expressing the human μ receptor were treated with vehicle or naloxone (10 μM) for 15 min before treatment with vehicle or buprenorphine (10 nM) for 2 h. (A) After addition of forskolin and DAMGO, cAMP levels were measured. This concentration of naloxone significantly shifted the DAMGO response curve to the right, and the addition of buprenorphine had no further effect on potency and no effect on efficacy. (B) Cells were washed after incubation with naloxone or naloxone plus buprenorphine. Following the wash, forskolin and DAMGO were added and cAMP levels measured. Naloxone pretreatment prevented the buprenorphine-mediated shift in DAMGO potency and the reduction in maximal response after washout. Data points represent mean ± SEM from at least four independent experiments (n = 4).

2.5. Molecular modeling

To identify potential binding hotspots on the μ receptor (Fig. 5A–C), four structures were downloaded from the Protein Data Bank (PDB) (PDB IDs: 4DKL, 5C1M, 6DDR, 6DDF). These structures were pre-processed by removing all non-receptor atoms before they were submitted to online servers for computational solvent mapping and binding site prediction (FTMap and FTSite (Kozakov et al., 2015; Wakefield et al., 2019),).

Fig. 5. Computational analysis supports a hemi-equilibrium mechanism for buprenorphine-mediated antagonism.

Fig. 5.

Open in a new tab

(A) Four allosteric hot spots on the μ receptor were computationally identified that may contribute significant binding energy. Three allosteric hot spots are extrahelical, located between either trans-membrane (TM) domains II, III, and IV, or TM IV and V. TM domains I, VI, and VII are not shown for clarity. (B) One allosteric hot spot was identified on the intracellular side of the receptor, which overlaps with the G protein binding site. (C) Buprenorphine was computationally docked into these hot spots, however the binding scores of the allosteric sites were significantly less than the binding score to the orthosteric site. Data represents mean ± SEM of the normalized binding scores from each of the four μ receptor structures used. (D) Using an equation that incorporates partial dissociation of a pre-equilibrated antagonist (i.e. hemi-equilibrium), the effect of increasing concentrations of buprenorphine on DAMGO concentration-response curves was simulated. This simulation predicts a non-saturable rightward shift of the DAMGO concentration-response curve and a saturable decrease in DAMGO efficacy, similar to the experimental data shown in Fig. 2A. (E) Schild regression of the calculated potency shifts demonstrate competition with a linear fit and a slope of 0.91. (F) Similar to the experimental data (Fig. 2C), the reduction of DAMGO efficacy by buprenorphine saturates near 10 nM. Simulated curves generated using GraphPad Prism.

The Rosetta ligand docking protocol on the ROSIE server (Lyskov et al., 2013a) was used to determine if buprenorphine could bind to each identified allosteric hot spot or to the orthosteric site for each of the four structures. A small ensemble of at least 200 models was constructed for each docking run. To validate this docking protocol, we redocked the μ receptor ligand, BU72, into the orthosteric pocket of the 5C1M structure, which produced models with root-mean-square deviations of ligand position of less than 0.8 Å. As a negative control, an unrelated molecule, the chemokine receptor antagonist IT1t, was docked into orthosteric site of 5C1M. The top six models (lowest interface score) in each ensemble were visually inspected and the interface scores were averaged. For comparison of the allosteric sites, these scores were normalized to interface energy of orthosteric buprenorphine (1) and the negative control (0), and averaged across all four structures. The averaged interface scores were; orthosteric buprenorphine: (−10.82), purple site: (−9.11), blue site: (−9.00), orange site: (−8.93), yellow site: (−8.73), and negative control: (−8.33).

Buprenorphine docked into the orthosteric pocket of the μ receptor with a low Rosetta interface energy score (i.e. high docking score), and matched the orientation (root-mean-square deviation: 1.3 Å) of previous buprenorphine docking performed by Ellis et al. (2018). Orthosteric docking of the chemokine receptor antagonist, IT1t, was used as a negative control, and interface energy scores of buprenorphine binding to putative allosteric sites were normalized to produce a docking score between 0 (IT1t) and 1 (orthosteric buprenorphine).

2.6. Schild analysis and hemi-equilibrium equations

The magnitude of the buprenorphine-mediated rightward shifts in DAMGO potency was quantified using the following equation (Arunlakshana and Schild, 1959):

log(DR-1)=log[B]+pKB Equation 1

where DR is the dose ratio of EC50 concentrations of the agonist in the presence or absence of the antagonist, [B] is the concentration of the antagonist, and pKB is the −log of the equilibrium dissociation constant of the antagonist. A regression analysis on a series of dose ratios with antagonist concentration was constructed. This will result in a straight line with a slope of one if the antagonism is competitive and there is not reduction in maximal effect. When a reduction in maximal effect is observed, the regression of log(DR-1) versus log[B] can still provide insight into the mechanism of action of insurmountable antagonists (Kenakin et al., 2006). In the case of hemi-equilibrium, little to no correction factor is needed (Kenakin et al., 2006).

To simulate buprenorphine antagonism (Fig. 5D–F), the following equation was used (Kenakin, 2019a; Kenakin et al., 2006), which describes the effect of an antagonist on agonist concentration-response curves under conditions of hemi-equilibrium:

Response=[A]KA1-ϑ1-e-k2ϕt+ρBe-k2ϕtτEmax[A]KA1-ϑ1-e-k2ϕt+ρBe-k2ϕtτ+1+1 Equation 2

where

ϑ=[B]KB/[B]KB+[A]KA+1
ρB=[B]KB/[B]KB+1
ϕ=[B]KB+[A]KA+1/[A]KA+1

The concentration of the agonist and antagonist are [A] and [B], and the equilibrium dissociation constants for the agonist and antagonist are KA and KB, respectively, τ is a measure of the efficacy of the agonist in the system (Black and Leff, 1983), and t is the time when the agonist response is measured. The KA for DAMGO was estimated to be 50 nM (based on the average EC50 for DAMGO in this system without receptor reserve) and a value of 2.3 for τ was chosen that produces a curve that matched the experimental DAMGO concentration-response curve. The k2 (0.0029 min−1) and estimated KB for buprenorphine (0.4 nM) was obtained from Bidlack et al. (2018). The time for the measurement of the agonist response was 35 min. With these values, the equilibration time multiplied by the rate of dissociation (i.e. t × K2) is 0.01, which is within hemi-equilibrium considerations (Kenakin, 2019a).

2.7. Data and statistical analysis

All statistical analyses were done using Prism software (GraphPad Software, Inc., version 9.5). Results are expressed as mean ± standard error of the mean (SEM). Figures depict the mean data points, and the curves were plotted using the mean of fitted parameters from each experiment. The peak value for each luminescence trace following agonist application was determined and fit to a logistic equation using non-linear regression to provide estimates of maximal response (Rmax) and potency (EC50).

R=R0R0-Ri1+10A-LogEC50 Equation 3

Where R is the measured response at a given agonist concentration (A), Ro is the response in the absence of agonist, Ri is the response after maximal inhibition by the agonist, and EC50 is the concentration of agonist that produces half-maximal response. Experimental repetitions are provided in each figure, and each experimental condition was performed at least in triplicate. All data analysis and statistical evaluations between treatment groups were done using the individual curve fit parameters and these statistics (including the geometric mean ± SEM) are reported in the text and/or figure legends. Significant differences in Emax were determined by one-way ANOVA with Dunnett multiple comparisons test. p < 0.05 was considered statistically significant.

3. Results

3.1. Buprenorphine reduces both the potency and efficacy of fentanyl

To assess the properties of buprenorphine acting as an antagonist, a luminescent biosensor (GloSensor, Promega) was used to monitor agonist-mediated inhibition of FSK-stimulated cAMP levels in HEK293 cells expressing a relatively low density of human μ receptor. As shown in Supplemental Figs. 1 and 2, in this cell system, buprenorphine agonism was not detectable and buprenorphine pretreatment did not alter baseline or FSK-stimulated cAMP levels. To measure the antagonist properties of buprenorphine, cells were pretreated with vehicle or buprenorphine (10 nM) for 2 h before the addition of fentanyl, a μ receptor agonist used clinically as an analgesic.

With vehicle pretreatment, fentanyl reduced FSK-stimulated cAMP levels by 70%, and was a potent agonist with an EC50 of ~30 nM (Fig. 1, Supplemental Table 1). Buprenorphine pretreatment antagonized fentanyl, and resulted in an insurmountable reduction of maximal fentanyl efficacy of 29% (Emax 50%). This reduction in maximal effect, a characteristic of noncompetitive antagonism, was originally attributed to pseudo-irreversible binding of buprenorphine preventing receptor occupancy by fentanyl during the course of the experiment.

However, buprenorphine pretreatment also reduced the potency of fentanyl, resulting in a rightward shift of the concentration response curve and a ~60-fold increase in EC50. A reduction in agonist potency is a characteristic of competitive antagonists. While this can also be observed for non-competitive antagonists in a system with high receptor-effector coupling efficiency (e.g. receptor reserve or “spare receptors”), this cell line expresses a low density of the μ receptor and lacks receptor reserve for DAMGO or fentanyl (Zamora et al., 2021). Therefore, an irreversible or pseudo-irreversible antagonist will occupy receptors necessary for the maximal response (i.e. reduced Emax) without affecting agonist potency (i.e. no change in EC50). Thus, the reduction in fentanyl potency could not be explained solely through a pseudo-irreversible mechanism.

3.2. Schild analysis reveals the reduction of efficacy, but not potency, saturates with increasing concentrations of buprenorphine

Buprenorphine antagonism appeared to have characteristics of both competitive (i.e. reduced agonist potency) and non-competitive (i.e. reduced agonist efficacy) antagonism. As competitive and non-competitive antagonism are mutually exclusive properties for a single ligand binding a single site, a Schild analysis (Arunlakshana and Schild, 1959) was performed to characterize the decrease in both agonist potency and efficacy by buprenorphine (Fig. 2A–Supplemental Table 1). Concentration-response curves for cAMP were obtained with the μ receptor agonist DAMGO after a 2-h pre-incubation with vehicle or buprenorphine (0.1–1000 nM). Similar to fentanyl, both DAMGO potency and efficacy were reduced by 10 nM buprenorphine pretreatment.

Schild analysis of the dose ratio (ratio of EC50 concentrations of the agonist in the presence of the antagonist divided by that in the absence of the antagonist) reveal that buprenorphine pretreatment results in parallel rightward shifts in the concentration-response curve (slope of 1.07) of DAMGO throughout the concentration range tested (Fig. 2B). The buprenorphine KB was calculated to be 0.1 nM, which is consistent with the values reported by Bidlack et al. (2018). While competitive antagonists, such as naloxone, will similarly produce a parallel rightward shift in the concentration-response curve, no diminution of the maximal response is expected. Thus, while one expectation of competitive antagonism is met by buprenorphine (parallel rightward shifts), the expectation of an unchanged maximal response is violated. Nevertheless, the Schild analysis can be useful to analyze the mechanism of action of insurmountable antagonists (Kenakin et al., 2006).

Analysis of the reduction in maximal DAMGO response by buprenorphine pretreatment shows that increasing concentrations of buprenorphine did not completely eliminate DAMGO signaling and the effect saturates near 10 nM buprenorphine (Fig. 2C). This is inconsistent with either simple competitive antagonists, which do not reduce the maximal effect, or irreversible antagonists which will completely eliminate agonist signaling at saturating concentrations (Kenakin, 2019a).

3.3. Buprenorphine’s reduction of agonist potency, but not efficacy, is reduced by washout

Two hypotheses could explain an antagonist that saturably reduces an agonist’s maximal effect while also reducing the potency in a parallel manner. First, buprenorphine could interact with the orthosteric site and partially re-equilibrate with the agonist, a condition known as hemi-equilibrium. Alternatively, buprenorphine could bind to an allosteric site, similar to the structurally-related molecule methocinnamox, which has recently been shown to act as both an orthosteric pseudo-irreversible antagonist and an allosteric modulator of the μ receptor (Zamora et al., 2021).

To begin to exclude possible mechanisms for buprenorphine’s antagonism of the μ receptor, we rigorously washed cells with drug-free media after a 2-h incubation with 10 nM buprenorphine (Fig. 3A–Supplemental Table 2). This wash procedure will remove un-bound buprenorphine before the addition of DAMGO but it will not significantly reduce receptor-bound buprenorphine due to its slow receptor dissociation rate. The reduction of DAMGO’s Emax was not significantly reduced by buprenorphine washout, supporting that this effect on efficacy was due to pseudo-irreversible binding.

This wash procedure partially reversed the reduction of DAMGO potency, indicating that this effect of buprenorphine was not due to irreversible binding to the μ receptor. Comparatively, the effect of the fast acting, competitive antagonist naloxone was completely abolished by washout (Fig. 3B). Thus, a subset of buprenorphine’s effect is wash-sensitive and is responsible for the reduction in DAMGO’s potency while a wash-resistant component reduces DAMGO efficacy.

3.4. Protection of the orthosteric binding pocket with naloxone prevents buprenorphine antagonism

We performed an orthosteric protection experiment to determine if possible allosteric sites on the μ receptor mediated the peculiar antagonism by buprenorphine. The orthosteric pocket was fully occupied by a 15-min preincubation with a high concentration of the orthosteric antagonist, naloxone (10 μM, >10,000 × Ki), before incubation with vehicle or buprenorphine (10 nM) for 2 h (Fig. 4A). If buprenorphine maintained its antagonism of DAMGO through orthosteric blockade, this would support an allosteric mechanism.

As expected for a competitive antagonist, 10 μM naloxone reduced DAMGO potency (from 54 nM to 141 μM, a 2630-fold increase in EC50, Supplemental Table 2), without a decrease in Emax. The pretreatment and presence of naloxone prevented buprenorphine-mediated reduction of DAMGO’s Emax, indicating that the buprenorphine effect on DAMGO Emax requires occupancy of the orthosteric site. Additionally, buprenorphine did not reduce the potency of DAMGO beyond that produced by naloxone pretreatment, suggesting that this was also mediated through a naloxone-sensitive site. Naloxone pretreatment prevented both effects of buprenorphine (reduced potency and efficacy of DAMGO) after washout (Fig. 4B, compare with Fig. 3A). Thus, buprenorphine is not able to bind to the μ receptor once naloxone occupies the orthosteric site. The naloxone sensitivity of buprenorphine antagonism suggests an allosteric component for buprenorphine is unlikely, as an allosteric component would likely be naloxone-insensitive, as occurs with methocinnamox, (Zamora et al., 2021).

3.5. Computational analysis supports a hemi-equilibrium mechanism for buprenorphine antagonism

While buprenorphine antagonism was sensitive to naloxone pre-treatment, there remains a possibility that naloxone binding caused a conformational change in the μ receptor that masked an allosteric binding site for buprenorphine. To address this possibility, an unbiased computational protocol (Kozakov et al., 2015) identified potential allosteric sites on four structures of the μ receptor, representing both active and inactive states. Four allosteric hot spots on the μ receptor were identified (Fig. 5A and B), which are similar to known allosteric regions in other class A G protein-coupled receptors (GPCRs) (Hedderich et al., 2022). We next docked buprenorphine into these sites using Rosetta (Combs et al., 2013; DeLuca et al., 2015; Kothiwale et al., 2015; Lyskov et al., 2013b). Additionally, we docked buprenorphine into the orthosteric site, as well as an unrelated compound as a negative control, to normalize binding scores (see methods). Comparison of the interface energy scores of the putative allosteric sites were 0.3 or lower (normalized to a buprenorphine orthosteric binding score of 1) (Fig. 5C). Thus, buprenorphine is unlikely to bind the μ receptor at these identified allosteric hot spots.

If buprenorphine acts solely through the orthosteric site, then a kinetic mechanism is likely to explain this antagonist behavior. Competitive antagonists, such as naloxone, dissociate relatively quickly from receptors to rapidly reach equilibration between the agonist, antagonist, and receptor. This results in surmountable, concentration-dependent rightward shifts in agonist potency as shown for naloxone in Fig. 3B and 4A. If there is no equilibration due to very slow antagonist dissociation from the receptor (i.e. irreversible or pseudo-irreversible binding) then antagonism will be insurmountable but will not change agonist potency in a system without receptor reserve for that agonist. Between these two kinetic extremes exists a unique condition, hemi-equilibrium, where receptor occupancy may partially reach equilibrium as described by Paton and Rang (Paton and Rang, 1965). This phenomenon is caused by slow, but not very slow, antagonist dissociation and occurs when the response to agonist is measured before equilibration can occur between the agonist, receptor, and antagonist (Gress et al., 2020; Kenakin, 2019a; Riddy et al., 2015). In such cases, a competitive antagonist can reduce both the maximal response and potency of an agonist.

We predicted how buprenorphine would antagonize DAMGO under hemi-equilibrium conditions (Equation (1), Fig. 5D). Values from buprenorphine radioligand binding from Bidlack et al. (2018), were used as estimates for parameters in the hemi-equilibrium equation (see methods). Notably, we used the Koff value of 0.0029 min−1, as this was measured in the presence of guanosine diphosphate which better mimics the cellular environment. When compared with our experimental data (compare Fig. 5 with Fig. 2) qualitatively similar results could be generated from simulation of a hemi-equilibrium mechanism. Specifically, the simulation reveals a reduction of potency that did not saturate and a linear Schild regression with a slope near 1 (Fig. 5E, compare with Fig. 2B). Additionally, the simulation shows that the predicted reduction in efficacy saturated at buprenorphine concentrations near 10 nM, equivalent to in vitro data (compare Fig. 5F with Fig. 2C). Thus, these results suggest that the competitive (shift in DAMGO potency) and apparent non-competitive (reduced DAMGO maximal response) effects of buprenorphine are consistent with hemi-equilibrium effects.

3.6. Reduction of agonist efficacy by buprenorphine is time dependent

If hemi-equilibrium were responsible for buprenorphine’s apparent insurmountable antagonism, then extending the assay incubation time should result in simple competitive antagonism. To test this, we incubated vehicle or buprenorphine (10 nM) with various DAMGO concentrations (0.1 nM-300 μM) for 0, 2, or 4 h before forskolin addition and measurement of luminescence (Fig. 6, Supplemental Table 3). After accounting for loss of efficacy due to receptor desensitization (the result of prolonged DAMGO exposure) (Supplemental Fig. 3), buprenorphine reduced both the potency and efficacy of DAMGO with 2 h of co-incubation. However, buprenorphine antagonism became surmountable (no reduction in DAMGO Emax) after 4 h of co-incubation, indicating that this was mediated through a kinetic effect (i.e. hemi-equilibrium).

4. Discussion

Buprenorphine is an effective medication for the treatment of opioid use disorder (Han et al., 2021). Many studies have investigated agonist effects of buprenorphine, but due to its relatively weak intrinsic efficacy (i.e. partial agonism), buprenorphine can also act as an antagonist at the μ receptor in systems with low receptor-effector coupling efficiency. Buprenorphine has a relatively slow dissociation rate from the μ receptor (Bidlack et al., 2018) and many studies have shown that buprenorphine’s effects in vivo are long-lasting (Cowan et al., 1977; Virk et al., 2009). As a result, it has been suggested that buprenorphine may act as a pseudo-irreversible antagonist at the μ receptor to reduce the maximal response of μ receptor agonists (Walker et al., 1995). The current study was performed to characterize the antagonist properties of buprenorphine at the human μ receptor.

Our data demonstrate that, in cells without receptor reserve for the high efficacy agonists, DAMGO or fentanyl, buprenorphine behaved as an insurmountable μ receptor antagonist. However, buprenorphine also reduced the potency of these agonists. This reduction in both potency and maximal response was not consistent with competitive antagonism (solely expected potency reduction) or pseudo-irreversible antagonism (solely expected efficacy reduction in a system without receptor reserve for the agonist). A shift in potency could not have been mediated through buprenorphine-induced receptor internalization, as this would only depress the maximal response in this system without receptor reserve. Further, buprenorphine-mediated agonism was undetectable and buprenorphine does not readily induce receptor internalization (Virk et al., 2009).

As the pseudo-irreversible antagonist, methocinnamox, has been shown to orthosterically reduce agonist maximal response and allosterically shift agonist potency (Zamora et al., 2021), we considered the possibility that buprenorphine may similarly act at multiple sites on the μ receptor. Saturation of drug effect, such as the saturation of buprenorphine’s Emax reduction (Fig. 2), could be due to saturation of the drug binding to an allosteric site. Further, one explanation for buprenorphine’s reduction in agonist potency or efficacy having different sensitivities to washout could be explained by buprenorphine binding to two sites with different affinities. However, unlike the fast-acting antagonist naloxone, buprenorphine’s reduction of DAMGO potency was not completely reversed by a rigorous washout procedure. The cause of this remains unknown, but we speculate that it may be due to buprenorphine’s higher lipophilicity (PubChem XLogp3 of 5) compared to naloxone’s (XLogP3 of 2.1) resulting in incomplete washout of un-bound buprenorphine.

Although allosteric sites have been identified on the μ receptor (Kandasamy et al., 2021; Livingston et al., 2018; Stahl et al., 2021), an allosteric site for buprenorphine has not been described in the literature. An in vitro binding study reported that buprenorphine had a slightly larger specific maximal binding (Bmax) than expected, but otherwise the data fit a single site model (Bidlack et al., 2018). We used an unbiased modeling approach to search the allosteric landscape for buprenorphine binding sites, but this computational search did not yield any strong candidates. While this data did not support an allosteric mechanism, it did not rule out an undiscovered allosteric site for buprenorphine.

More definitively, both the reductions in agonist maximal response and potency by buprenorphine were completely prevented by naloxone protection of the orthosteric site (Fig. 4). This naloxone pretreatment also prevented buprenorphine antagonism after washout, suggesting that buprenorphine cannot bind the μ receptor when naloxone is bound. As naloxone is an orthosteric antagonist, the simplest explanation is that buprenorphine’s reduction of both agonist potency and efficacy are mediated by the orthosteric site. However, as orthosteric ligands change the conformation of the receptor, it remains possible that the naloxone-bound μ receptor occludes an undiscovered allosteric site.

In the absence of support for an allosteric site of action, we considered that the slow receptor dissociation rate of buprenorphine may be responsible for the reduction in agonist efficacy and potency through a kinetic mechanism. Considering the recently determined buprenorphine dissociation rate reported by Bidlack et al. (2018), we speculated that hemi-equilibrium conditions may be responsible. Hemi-equilibrium can occur when a competitive antagonist does not have sufficient time to reach equilibrium with an agonist during the measurement of a response that requires only a short amount of time of agonist exposure (Cullum et al., 2022; Kenakin, 2019a; Kenakin et al., 2006).

To demonstrate if hemi-equilibrium conditions can allow a competitive, surmountable antagonist to reduce the maximal effect of an agonist, we performed a simulation using an equation that describes the effect of an antagonist under hemi-equilibrium conditions (Equation (2)). Using reasonable estimates for parameters in this equation, we found that the simulated effect of buprenorphine was qualitatively similar to our experimental results and could reduce both the potency and maximal response of DAMGO.

If hemi-equilibrium conditions were responsible for buprenorphine’s apparent non-competitive antagonism (insurmountable reduction in DAMGO Emax), then allowing sufficient time for buprenorphine to dissociate would return buprenorphine’s apparent behavior to that of a simple competitive antagonist. To test this, we co-incubated buprenorphine and DAMGO, as well as DAMGO alone, to observe any time-dependent kinetic effects. With 2 h of co-incubation, buprenorphine reduced DAMGO potency and maximal response, similar to the experiments with 2-h buprenorphine pretreatment. After 4 h of co-incubation, buprenorphine reduced DAMGO potency but not efficacy.

This time dependent difference in antagonism can be explained by re-equilibration kinetics (hemi-equilibrium), where a proportion of receptors remain bound to buprenorphine and are unable to bind DAMGO to contribute to the DAMGO maximum effect. As more time elapses, buprenorphine will dissociate from these receptors and allow DAMGO to exert its full effect. Thus, the low-efficacy agonist buprenorphine can act as an insurmountable antagonist due to the phenomenon of hemi-equilibrium.

The phenomenon of hemi-equilibrium is usually associated with very rapid signal collection, such as the measurement of calcium transients (evaluated at a timescale of seconds) where insurmountability can occur with simple competitive antagonists (Berg et al., 2006; Christopoulos et al., 1999; Cullum et al., 2022; Vauquelin and Charlton, 2010). This state of hemi-equilibrium depends on both the time of the assay and antagonist dissociation rate. As demonstrated here, an antagonist with sufficiently slow receptor dissociation can result in insurmountable antagonism on the time scale of minutes to hours. While extended agonist exposure time will allow the system to reach equilibrium, this can also lead to changes in the system due to prolonged agonist action (e. g. receptor desensitization) (Williams et al., 2013).

Notably, this desensitization was observed in the co-incubation experiment (Supplemental Fig. 3), where extended DAMGO incubation resulted in a reduced maximal response after addition of forskolin. To compensate for this desensitization, the maximal response was normalized to the DAMGO-only condition at the corresponding incubation time. This desensitization means that while DAMGO could surmount buprenorphine after 4 h of co-incubation, the maximum inhibition of the cAMP signal will be less than that of DAMGO without extended incubation. This effect would be more pronounced with an antagonist that itself induces desensitization, and is likely to impact the magnitude of biological processes even after agonist-antagonist-receptor equilibrium is reached. Additionally, transient agonist signals, such as those that occur during synaptic transmission, may occur well before equilibrium with an antagonist is reached (Ogawa and Komatsu, 2010; Runyan et al., 2017), and so these neurons would experience a competitive antagonist with a sufficiently slow receptor dissociation as an insurmountable antagonist.

It has been postulated that perioperative opioid analgesia may be disrupted in patients who are on buprenorphine maintenance therapy (Roberts and Meyer-Witting, 2005). This remains a disputed topic, as clinical reports differ on the effectiveness of opioid analgesics in these patients (Chern et al., 2012; Webster et al., 2020). The level of μ receptor occupancy by buprenorphine may vary between patients and through time, which may explain the variability in opioid analgesic effectiveness in clinical reports. Our data suggest that if sufficient levels of the μ receptor are occupied by buprenorphine, opioid analgesics may be unable to surmount buprenorphine due to its slow receptor dissociation, thereby interfering with analgesic efficacy. Although additional studies are needed to translate these observations into clinical practice, these data imply that pain should be closely monitored and alternative non-opioid analgesics should be considered for patients on buprenorphine maintenance therapy.

With the growing interest in discovering new allosteric modulators of GPCRs (Wold et al., 2019), it is important to consider that hemi-equilibrium conditions may influence drug response. For example, a reduction in maximal effect is characteristic of an allosteric or non-competitive antagonist. However, as demonstrated here it can also occur with a competitive antagonist under conditions of hemi-equilibrium. Thus, there is a potential danger of misinterpreting a drug’s pharmacological characteristics if additional experiments, such as orthosteric protection or Schild analysis, are not performed (Kenakin, 2019c; Kenakin et al., 2006). As more emphasis is now being placed on real-time data collection, particularly with the development of fast acting biosensors, hemi-equilibrium conditions may become more prevalent (Kenakin, 2019b; Riddy et al., 2015).

In conclusion, the current study has demonstrated that buprenorphine can reduce both the efficacy and potency of μ receptor agonists. This is consistent with buprenorphine’s slow receptor dissociation rate, relative to the time for measurement of the agonist response, resulting in hemi-equilibrium. Thus, buprenorphine may appear as both a competitive and non-competitive antagonist, depending on the duration of agonist treatment. While conditions of hemi-equilibrium are more commonly observed at shorter time scales, it is likely to be more prevalent in the future as more data is collected with fast-acting biosensors. While an allosteric mechanism was not supported by these data, future studies such as radioligand binding will be required to eliminate the possibility of an undiscovered allosteric site. Non-equilibrium steady-states occur frequently in vivo, which warrant further study on the dynamics between receptors, agonists, and antagonists under hemi-equilibrium conditions.

Supplementary Material

Supplemental Data

Acknowledgements

This work was supported by the National Institutes of Health grants [DA048214] and [DA057787] to WPC and KAB; MJW was supported by a T32 Training Program [DA031115] and an F32 fellowship [DA056212].

Abbreviations

μ receptor

Mu opioid receptor

cAMP

Cyclic adenosine monophosphate

DAMGO

[D-Ala2, N-MePhe4, Gly-ol]-enkephalin (μ receptor agonist)

FDA

Food and Drug Administration

OUD

Opioid use disorder

FSK

Forskolin

PDB

Protein Data Bank

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.ejphar.2024.177192.

Footnotes

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

CRediT authorship contribution statement

Michael J. Wedemeyer: Writing – review & editing, Writing – original draft, Visualization, Validation, Investigation, Formal analysis, Data curation, Conceptualization. Teresa S. Chavera: Writing – review & editing, Methodology, Investigation, Data curation. Kelly A. Berg: Writing – review & editing, Supervision, Project administration, Investigation, Formal analysis. William P. Clarke: Writing – review & editing, Supervision, Project administration, Funding acquisition, Formal analysis, Conceptualization.

Data availability

Data will be made available on request.

References

  1. Arunlakshana O, Schild HO, 1959. Some Quantitative uses of drug antagonists. Br. J. Pharmacol. Chemother 14 (1), 48–58. 10.1111/j.1476-5381.1959.tb00928.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Berg KA, Navailles S, Sanchez TA, Silva YM, Wood MD, Spampinato U, Clarke WP, 2006. Differential effects of 5-methyl-1-[[2-[(2-methyl-3-pyridyl)oxyl]-5- pyridyl]carbamoyl]-6-trifluoromethylindone (SB 243213) on 5-hydroxytryptamine2C receptor-mediated responses. J. Pharmacol. Exp. Therapeut 319 (1), 260–268. 10.1124/jpet.106.104448. [DOI] [PubMed] [Google Scholar]
  3. Bidlack JM, Knapp BI, Deaver DR, Plotnikava M, Arnelle D, Wonsey AM, Toh MF, Pin SS, Namchuk MN, 2018. In vitro pharmacological characterization of buprenorphine, samidorphan, and combinations being developed as an adjunctive treatment of major depressive disorder. J. Pharmacol. Exp. Therapeut 367 (2), 267–281. 10.1124/jpet.118.249839. [DOI] [PubMed] [Google Scholar]
  4. Black JW, Leff P, 1983. Operational models of pharmacological agonism. Proceedings of the Royal Society of London - Biological Sciences 220 (1219), 141–162. 10.1098/rspb.1983.0093. [DOI] [PubMed] [Google Scholar]
  5. Boas RA, Villiger JW, 1985. Clinical actions of fentanyl and buprenorphine: the significance of receptor binding. Br. J. Anaesth 57 (2), 192–196. 10.1093/bja/57.2.192. [DOI] [PubMed] [Google Scholar]
  6. Buresh M, Ratner J, Zgierska A, Gordin V, Alvanzo A, 2020. Treating perioperative and acute pain in patients on buprenorphine: narrative literature review and practice recommendations. J. Gen. Intern. Med 35 (12), 3635–3643. 10.1007/s11606-020-06115-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Butler S, 2013. Buprenorphine-Clinically useful but often misunderstood. Scandinavian Journal of Pain 4 (3), 148–152. 10.1016/j.sjpain.2013.05.004. De Gruyter. [DOI] [PubMed] [Google Scholar]
  8. Chern SYS, Isserman R, Chen L, Ashburn M, Liu R, 2012. Perioperative pain management for patients on chronic buprenorphine: a case report. J. Anesth. Clin. Res 3 (10). 10.4172/2155-6148.1000250. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Christopoulos A, Parsons AM, Lew MJ, El-Fakahany EE, 1999. The assessment of antagonist potency under conditions of transient response kinetics. Eur. J. Pharmacol 382 (3), 217–227. 10.1016/S0014-2999(99)00550-6. [DOI] [PubMed] [Google Scholar]
  10. Combs SA, Deluca SL, Deluca SH, Lemmon GH, Nannemann DP, Nguyen ED, Willis JR, Sheehan JH, Meiler J, 2013. Small-molecule ligand docking into comparative models with Rosetta. Nat. Protoc 8 (7), 1277–1298. 10.1038/nprot.2013.074. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cowan A, Lewis JW, Macfarlane IR, 1977. Agonist and antagonist properties of buprenorphine, a new antinociceptive agent. Br. J. Pharmacol 60 (4), 537–545. 10.1111/j.1476-5381.1977.tb07532.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cullum SA, Veprintsev DB, Hill SJ, 2022. Kinetic analysis of endogenous β2-adrenoceptor-mediated cAMP GloSensor™ responses in HEK293 cells. Br. J. Pharmacol 10.1111/bph.16008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Davis MP, Pasternak G, Behm B, 2018. Treating chronic pain: an overview of clinical studies centered on the buprenorphine option. Drugs 78 (12), 1211–1228. 10.1007/s40265-018-0953-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. DeLuca S, Khar K, Meiler J, 2015. Fully flexible docking of medium sized ligand libraries with rosettaligand. PLoS One 10 (7), e0132508. 10.1371/journal.pone.0132508. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Ehrlich AT, Semache M, Gross F, Da Fonte DF, Runtz L, Colley C, Mezni A, Le Gouill C, Lukasheva V, Hogue M, Darcq E, Bouvier M, Kieffer BL, 2019. Biased signaling of the mu opioid receptor revealed in native neurons. iScience 14, 47–57. 10.1016/j.isci.2019.03.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Ellis CR, Kruhlak NL, Kim MT, Hawkins EG, Stavitskaya L, 2018. Predicting opioid receptor binding affinity of pharmacologically unclassified designer substances using molecular docking. PLoS One 13 (5), 1–18. 10.1371/journal.pone.0197734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gowing L, Ali R, White JM, Mbewe D, 2017. Buprenorphine for managing opioid withdrawal. Cochrane Database Syst. Rev 2017 (2). 10.1002/14651858.CD002025.pub5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Greenwald MK, Johanson CE, Moody DE, Woods JH, Kilbourn MR, Koeppe RA, Schuster CR, Zubieta JK, 2003. Effects of buprenorphine maintenance dose on μ-opioid receptor availability, plasma concentrations, and antagonist blockade in heroin-dependent volunteers. Neuropsychopharmacology 28 (11), 2000–2009. 10.1038/sj.npp.1300251. [DOI] [PubMed] [Google Scholar]
  19. Gress K, Charipova K, Jung JW, Kaye AD, Paladini A, Varrassi G, Viswanath O, Urits I, 2020. A comprehensive review of partial opioid agonists for the treatment of chronic pain. Best Pract. Res. Clin. Anaesthesiol 34 (Issue 3), 449–461. 10.1016/j.bpa.2020.06.003. Best Pract Res Clin Anaesthesiol. [DOI] [PubMed] [Google Scholar]
  20. Han B, Jones CM, Einstein EB, Compton WM, 2021. Trends in and characteristics of buprenorphine misuse among adults in the US. JAMA Netw. Open 4 (10), e2129409. 10.1001/jamanetworkopen.2021.29409. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hayes BD, Klein-Schwartz W, Doyon S, 2008. Toxicity of buprenorphine overdoses in children. Pediatrics 121 (4), e782–e786. 10.1542/peds.2007-1774. [DOI] [PubMed] [Google Scholar]
  22. Hedderich JB, Persechino M, Becker K, Heydenreich FM, Gutermuth T, Bouvier M, Bünemann M, Kolb P, 2022. The pocketome of G-protein-coupled receptors reveals previously untargeted allosteric sites. Nat. Commun 13 (1), 1–12. 10.1038/s41467-022-29609-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Huang A, Katznelson R, de Perrot M, Clarke H, 2014. Perioperative management of a patient undergoing Clagett window closure stabilized on Suboxone® for chronic pain: a case report. Can. J. Anesth 61 (9), 826–831. 10.1007/s12630-014-0193-y. [DOI] [PubMed] [Google Scholar]
  24. Kandasamy R, Hillhouse TM, Livingston KE, Kochan KE, Meurice C, Eans SO, Li MH, White AD, Roques BP, McLaughlin JP, Ingram SL, Burford NT, Alt A, Traynor JR, 2021. Positive allosteric modulation of the mu-opioid receptor produces analgesia with reduced side effects. Proceedings of the National Academy of Sciences of the United States of America 118 (16), e2000017118. 10.1073/pnas.2000017118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kenakin T, 2019a. Orthosteric drug antagonism. In: A Pharmacology Primer. 10.1016/b978-0-12-813957-8.00006-0. [DOI] [Google Scholar]
  26. Kenakin T, 2019b. Analytical pharmacology: how numbers can guide drug discovery. ACS Pharmacol. Transl. Sci 2 (1), 9–17. 10.1021/acsptsci.8b00057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kenakin T, 2019c. Biased receptor signaling in drug discovery. Pharmacol. Rev 71 (2), 267–315. 10.1124/pr.118.016790. [DOI] [PubMed] [Google Scholar]
  28. Kenakin T, Jenkinson S, Watson C, 2006. Determining the potency and molecular mechanism of action of insurmountable antagonists. J. Pharmacol. Exp. Therapeut 319 (2), 710–723. 10.1124/jpet.106.107375. [DOI] [PubMed] [Google Scholar]
  29. Khanna IK, Pillarisetti S, 2015. Buprenorphine – an attractive opioid with underutilized potential in treatment of chronic pain. J. Pain Res 8, 859–870. 10.2147/JPR.S85951. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kothiwale S, Mendenhall JL, Meiler J, 2015. BCL::Conf: small molecule conformational sampling using a knowledge based rotamer library. J. Cheminf 7 (1), 1–15. 10.1186/s13321-015-0095-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Kozakov D, Grove LE, Hall DR, Bohnuud T, Mottarella SE, Luo L, Xia B, Beglov D, Vajda S, 2015. The FTMap family of web servers for determining and characterizing ligand-binding hot spots of proteins. Nat. Protoc 10 (5), 733–755. 10.1038/nprot.2015.043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kuo A, Magiera J, Rethwan N, Andersson Å, Leen Lam A, Wyse B, Meutermans W, Lewis R, Smith M, 2020. In vitro profiling of opioid ligands using the cAMP formation inhibition assay and the β-arrestin2 recruitment assay: No two ligands have the same profile. Eur. J. Pharmacol 872, 172947. 10.1016/j.ejphar.2020.172947. [DOI] [PubMed] [Google Scholar]
  33. Larochelle MR, Bernson D, Land T, Stopka TJ, Wang N, Xuan Z, Bagley SM, Liebschutz JM, Walley AY, 2018. Medication for opioid use disorder after nonfatal opioid overdose and association with mortality: a cohort study. Ann. Intern. Med 169 (3), 137–145. 10.7326/M17-3107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Lewis JW, 1985. Buprenorphine. Drug Alcohol Depend. 14 (3–4), 363–372. 10.1016/0376-8716(85)90067-5. [DOI] [PubMed] [Google Scholar]
  35. Livingston KE, Stanczyk MA, Burford NT, Alt A, Canals M, Traynor JR, 2018. Pharmacologic evidence for a putative conserved allosteric site on opioid receptors. Mol. Pharmacol 93 (2), 157–167. 10.1124/mol.117.109561. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Lutfy K, Cowan A, 2005. Buprenorphine: a unique drug with complex pharmacology. Curr. Neuropharmacol 2 (4), 395–402. 10.2174/1570159043359477. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Lutfy K, Eitan S, Bryant CD, Yang YC, Saliminejad N, Walwyn W, Kieffer BL, Takeshima H, Carroll FI, Maidment NT, Evans CJ, 2003. Buprenorphine-induced antinociception is mediated by μ-opioid receptors and compromised by concomitant activation of opioid receptor-like receptors. J. Neurosci 23 (32), 10331–10337. 10.1523/jneurosci.23-32-10331.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Lyskov S, Chou FC, Conchúir SÓ, Der BS, Drew K, Kuroda D, Xu J, Weitzner BD, Renfrew PD, Sripakdeevong P, Borgo B, Havranek JJ, Kuhlman B, Kortemme T, Bonneau R, Gray JJ, Das R, 2013a. Serverification of molecular modeling applications: the Rosetta online server that includes everyone (ROSIE). PLoS One 8 (5). 10.1371/journal.pone.0063906. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Lyskov S, Chou FC, Conchúir SÓ, Der BS, Drew K, Kuroda D, Xu J, Weitzner BD, Renfrew PD, Sripakdeevong P, Borgo B, Havranek JJ, Kuhlman B, Kortemme T, Bonneau R, Gray JJ, Das R, 2013b. Serverification of molecular modeling applications: the Rosetta online server that includes everyone (ROSIE). PLoS One 8 (5). 10.1371/journal.pone.0063906. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Mccormick Z, Chu SK, Chang-Chien GC, Joseph P, 2013. Acute pain control challenges with buprenorphine/naloxone therapy in a patient with compartment syndrome secondary to McArdle’s disease: a case report and review. Pain Med. 14 (8), 1187–1191. 10.1111/pme.12135. [DOI] [PubMed] [Google Scholar]
  41. Ogawa T, Komatsu H, 2010. Differential temporal storage capacity in the baseline activity of neurons in macaque frontal eye field and area V4. J. Neurophysiol 103 (5), 2433–2445. 10.1152/JN.01066.2009. [DOI] [PubMed] [Google Scholar]
  42. Paton W, Rang H, 1965. The uptake of atropine and related drugs by intestinal smooth muscle of the Guinea-pig in relation to acetylcholine receptors. Proceedings of the Royal Society of London. Series B. Biological Sciences 163 (August). [DOI] [PubMed] [Google Scholar]
  43. Pergolizzi J, Böger RH, Budd K, Dahan A, Erdine S, Hans G, Kress HG, Langford R, Likar R, Raffa RB, Sacerdote P, 2008. Opioids and the management of chronic severe pain in the elderly: consensus statement of an international expert panel with focus on the six clinically most often used world health organization step III opioids (Buprenorphine, Fentanyl, Hydromorphone, Met. Pain Pract 8 (4), 287–313. 10.1111/j.1533-2500.2008.00204.x. [DOI] [PubMed] [Google Scholar]
  44. Riddy DM, Valant C, Rueda P, Charman WN, Sexton PM, Summers RJ, Christopoulos A, Langmead CJ, 2015. Label-free kinetics: exploiting functional hemi-equilibrium to derive rate constants for muscarinic receptor antagonists. Mol. Pharmacol 88 (4), 779–790. 10.1124/mol.115.100545. [DOI] [PubMed] [Google Scholar]
  45. Roberts DM, Meyer-Witting M, 2005. High-dose buprenorphine: perioperative precautions and management strategies. Anaesth. Intensive Care 33 (1), 17–25. 10.1177/0310057x0503300104. [DOI] [PubMed] [Google Scholar]
  46. Runyan CA, Piasini E, Panzeri S, Harvey CD, 2017. Distinct timescales of population coding across cortex. Nature 548 (7665), 92–96. 10.1038/NATURE23020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Shulman M, Wai JM, Nunes EV, 2019. Buprenorphine treatment for opioid use disorder: an overview. CNS Drugs 33 (6), 567–580. 10.1007/s40263-019-00637-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Spadaro A, Long B, Koyfman A, Perrone J, 2022. Buprenorphine precipitated opioid withdrawal: prevention and management in the ED setting. AJEM (Am. J. Emerg. Med.) 58, 22–26. 10.1016/j.ajem.2022.05.013. [DOI] [PubMed] [Google Scholar]
  49. Stahl EL, Schmid CL, Acevedo-Canabal A, Read C, Grim TW, Kennedy NM, Bannister TD, Bohn LM, 2021. G protein signaling-biased mu opioid receptor agonists that produce sustained G protein activation are noncompetitive agonists. Proceedings of the National Academy of Sciences of the United States of America 118 (48). 10.1073/pnas.2102178118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Tzschentke TM, 2002. Behavioral pharmacology of buprenorphine, with a focus on preclinical models of reward and addiction. Psychopharmacology 161 (1), 1–16. 10.1007/s00213-002-1003-8. Psychopharmacology (Berl). [DOI] [PubMed] [Google Scholar]
  51. Vauquelin G, Charlton SJ, 2010. Long-lasting target binding and rebinding as mechanisms to prolong in vivo drug action. Br. J. Pharmacol 161 (3), 488–508. 10.1111/j.1476-5381.2010.00936.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Villiger JW, Taylor KM, 1981. Buprenorphine: characteristics of binding sites in the rat central nervous system. Life Sci. 29 (26), 2699–2708. 10.1016/0024-3205(81)90529-4. [DOI] [PubMed] [Google Scholar]
  53. Virk MS, Arttamangkul S, Birdsong WT, Williams JT, 2009. Buprenorphine is a weak partial agonist that inhibits opioid receptor desensitization. J. Neurosci 29 (22), 7341–7348. 10.1523/JNEUROSCI.3723-08.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Wakefield AE, Mason JS, Vajda S, Keserű GM, 2019. Analysis of tractable allosteric sites in G protein-coupled receptors. Sci. Rep 9 (1). 10.1038/s41598-019-42618-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Walker EA, Zernig G, Woods JH, 1995. Buprenorphine antagonism of mu opioids in the rhesus monkey tail-withdrawal procedure. J. Pharmacol. Exp. Therapeut 273 (3), 1345–1352. https://pubmed.ncbi.nlm.nih.gov/7791107/. [PubMed] [Google Scholar]
  56. Wang PC, Ho IK, Lee CWS, 2015. Buprenorphine-elicited alteration of adenylate cyclase activity in human embryonic kidney 293 cells coexpressing κ-, μ-opioid and nociceptin receptors. J. Cell Mol. Med 19 (11), 2587–2596. 10.1111/jcmm.12644. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Webster L, Gudin J, Raffa RB, Kuchera J, Rauck R, Fudin J, Adler J, Mallick-Searle T, 2020. Understanding buprenorphine for use in chronic pain: expert opinion. Pain Med. 21 (4), 714–723. 10.1093/PM/PNZ356. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Williams JT, Ingram SL, Henderson G, Chavkin C, von Zastrow M, Schulz S, Koch T, Evans CJ, Christie MJ, 2013. Regulation of μ-opioid receptors: desensitization, phosphorylation, internalization, and tolerance. Pharmacol. Rev 65 (1), 223–254. 10.1124/pr.112.005942. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Wold EA, Chen J, Cunningham KA, Zhou J, 2019. Allosteric modulation of class A GPCRs: targets, agents, and emerging concepts. J. Med. Chem 62 (1), 88–127. 10.1021/acs.jmedchem.8b00875. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Zamora JC, Smith HR, Jennings EM, Chavera TS, Kotipalli V, Jay A, Husbands SM, Disney A, Berg KA, Clarke WP, 2021. Long-term antagonism and allosteric regulation of mu opioid receptors by the novel ligand, methocinnamox. Pharmacology Research and Perspectives 9 (6). 10.1002/prp2.887. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Data
NIHMS2149387-supplement-Supplemental_Data.pdf (323.6KB, pdf)

Data Availability Statement

Data will be made available on request.

RESOURCES