Abstract
Dynorphin (DYN)/κ-opioid receptor (KOR) activation contributes to aversion, dysphoria, sedation, depression, and enhanced psychostimulant-rewarding effects by inhibiting dopamine (DA) release. The precise neuronal mechanisms underlying these effects remain unclear, limiting the use of KOR agonists in treating mood and substance use disorders. DYN fibers form synapses with DA terminals that express KOR and dopamine transporter (DAT), which is crucial for regulating DA dynamics and related behaviors. Previously, we demonstrated that KOR agonists upregulate DAT activity via ERK1/2 signaling involving phospho-Thr53 DAT (pT53-DAT). However, it remains unclear whether pT53-DAT is involved in KOR-mediated DAT regulation in vivo and whether such a phenomenon contributes to the behavioral effects of KOR agonism. In this study, we utilized male DAT-Ala53 knock-in mice with nonphosphorylatable Ala at position 53 to investigate the role of pT53-DAT in KOR-mediated DAT regulation and its behavioral effects. KOR agonist U69593 increased KOR antagonist-sensitive DAT activity, DAT Vmax, pT53-DAT, and surface expression in WT but not in DAT-Ala53 mice. KOR agonists caused locomotor suppression, conditioned place aversion (CPA), and enhanced cocaine preference [conditioned place preference (CPP)] in WT but not in DAT-Ala53 mice. Conversely, both WT and DAT-Ala53 mice exhibited similar lithium chloride-induced CPA and morphine-induced CPP. These findings provide the first causal evidence that KOR-mediated locomotor suppression, aversive response, and enhancement of cocaine reward manifest through the modulation of DAT activity via DAT-Thr53 phosphorylation. This suggests that targeting specific DAT-regulatory motif(s) may help develop new KOR-directed therapeutic strategies devoid of adverse effects.
- κ-opioid receptor
- cocaine reward
- conditioned place aversion
- dopamine transporter
- knock-in mice
- threonine-53 phosphorylation
Significance Statement
Preclinical and clinical research reveal that cocaine use disorder (CUD) affects mesolimbic dopamine neurotransmission, dopamine transporters (DAT), and DA interactions with the dynorphin (DYN)/κ-opioid receptor (KOR) system. The lack of FDA-approved treatments for CUD highlights a significant gap in our understanding of its neurobiology. While KOR ligands have potential as therapies, their effectiveness is often limited by side effects like aversion, dysphoria, and enhanced cocaine reward. Our study demonstrates that phosphorylation of Thr53 motif in DAT is crucial for KOR-mediated aversion, locomotor suppression, and enhancement of cocaine reward. These findings provide the first neurobiological evidence linking DAT-Thr53 phosphorylation to KOR modulation of DA clearance, highlighting its contribution to adverse behavioral outcomes and opening avenues for effective CUD treatments.
Introduction
Dynorphin (DYN) and its native κ-opioid receptor (KOR) are expressed in the central nervous system and play a role in many physiological functions (Chavkin et al., 1982; Watson et al., 1982; Mansour et al., 1988). DYN/KOR system is implicated in depression, anxiety, drug addiction, and schizophrenia (reviewed in Tejeda et al., 2012). However, the therapeutic potential of targeting this system is limited by adverse effects. KOR activation produces dysphoric and psychotomimetic effects in humans (Pfeiffer et al., 1986) and aversive behaviors in rodents (Shippenberg and Herz, 1987; Land et al., 2008). Systemic administration of KOR agonists, including U69593, U50488, and salvinorin A (SalA), induces conditioned place aversion (CPA) and locomotor suppression in rodents (White et al., 2015). DYN/KOR activation by stress or treatment with KOR agonist U50488 prior to cocaine conditioning potentiates cocaine conditioned place preference (CPP; McLaughlin et al., 2006a; Schindler et al., 2010). Understanding the molecular mechanisms that contribute to these adverse effects associated with KOR activation could pave the way for developing new therapeutics that minimize these negative outcomes.
DYN fibers synapse onto DA terminals, and KOR is expressed on these same DA terminals as a heteroreceptor (Svingos et al., 2001), suggesting that the presynaptic DYN/KOR system modulates DA transmission. Indeed, KOR agonists decrease DA release and extracellular DA concentrations via the activation of KOR located on DA terminals, and the decreased dopaminergic neurotransmission is implicated in KOR agonist-induced CPA and locomotor suppression (Werling et al., 1988; Spanagel et al., 1990). Furthermore, specific elimination of KOR in DA neurons abolishes KOR agonist-induced place aversion and synaptic DA modulation and enhances cocaine sensitization (Chefer et al., 2013; Ehrich et al., 2015). However, the downstream targets on which KOR-triggered signaling cascades act to drive aversion and other behavioral actions are unknown.
The functional properties of DA transporter (DAT) on DA neurons are one of the critical factors in regulating synaptic DA for neurotransmission through clearance of released DA and behavior (Fumagalli et al., 1998). Several cellular protein kinases regulate DAT function via posttranslational modifications such as phosphorylation (reviewed in and see references therein Ramamoorthy et al., 2011; Bermingham and Blakely, 2016; Foster and Vaughan, 2017). Activation or inactivation of intracellular kinase/phosphatase signaling cascades might occur through activation of auto- and heteroreceptors expressed on DA neurons/terminals that consequently modulate DAT-mediated DA clearance and release, DA neurotransmission, and behaviors. However, the role of DAT phosphorylation in a physiologically relevant model system has yet to be determined.
Given the fact that KOR is expressed as a heteroreceptor on DAT-expressing DA terminals, we previously demonstrated that KOR agonists upregulate DAT function via ERK1/2 activation and that this regulation requires phosphorylation of Thr53 in DAT (phospho-T53-DAT or pT53-DAT; Kivell et al., 2014; Mayer et al., 2025). However, a causative relationship of endogenous pT53-DAT and KOR-mediated DAT upregulation has not yet been investigated, and its contribution to KOR agonist-induced aversion, locomotor suppression, and enhancement of cocaine-induced CPP is also unknown.
In the current study, we directly examined the causal role of in vivo pT53-DAT in mediating KOR-linked DAT regulation, locomotor suppression, aversive behavior, and potentiation of cocaine-induced CPP using DAT-Ala53 mice carrying nonphosphorylatable Ala substituted at Thr53 phosphorylation site in DAT and WT counterparts. Our findings indicate that KOR agonists upregulate DAT activity by increasing the total levels of pT53-DAT and DAT Vmax and the surface expression levels of both DAT and pT53-DAT in WT mice but not in DAT-Ala53 mice. Moreover, our results demonstrate that the lack of DAT-Thr53 phosphorylation in DAT-Ala53 mice prevents locomotor suppression, CPA, and potentiation of cocaine-induced CPP produced by KOR agonists. Nonetheless, DAT-Ala53 mice exhibit unaltered LiCl-induced CPA and morphine-induced hyperlocomotion and CPP when compared with WT. These results directly link pT53-DAT to KOR modulation of dopaminergic neurotransmission and consequent behavioral outcomes, including KOR agonist-induced locomotor suppression, aversive behaviors, and potentiation of cocaine reward.
Materials and Methods
Animals and housing
All animal studies and care were performed under the guidelines of the Virginia Commonwealth University Institutional Animal Care and Use Committee, in accordance with the principles and procedures outlined in the National Research Council “Guide for the Care and Use of Laboratory Animals.” Male homozygous DAT-Ala53 and wild-type (WT) littermates of 8–10 weeks were used in this study. Mice were housed in a mouse colony room (3–5 males per cage) in a temperature- and humidity-controlled environment under a 12 h light/dark light schedule (lights on at 7:00 A.M.) and were fed standard mouse chow (irradiated Teklad LM-485 diet) and autoclaved water. All behavioral and biochemical experiments were conducted between 10 A.M. and 5 P.M. Our laboratory has generated DAT-Ala53 knock-in mice on a C57BL/6J background using CRISPR/Cas9 technology. DAT-Ala53 mice exhibit a birth rate consistent with the Mendelian inheritance ratio. Both male and female homozygous mice are viable and show no significant abnormalities in growth, body weight, body temperature, or general health. In DAT-Ala53 mice, there are no significant changes in the distribution of tyrosine hydroxylase (TH) and DAT, the colocalization of TH and DAT, the levels of DA and its metabolites (DOPAC, HVA, 3-MT), striatal DAT transport activity and kinetics, or total and surface DAT protein levels. Importantly, pT53-DAT immunoreactivity is absent in DAT-Ala53 mice. In addition, DAT-Ala53 shows insensitivity to ERK1/2 inhibition and significantly reduced response to cocaine inhibition. Behaviorally, DAT-Ala53 mice exhibit novelty-induced hyperactivity and diminished locomotor activation triggered by cocaine. We recently published the biochemical, neurochemical, and behavioral phenotypes of DAT-A53 mice (Ragu Varman et al., 2021b).
Experimental groups and drugs
Male WT or homozygous DAT-Ala53 mice were assigned to one of four treatment groups: saline + vehicle, saline + U69593 (0.1 mg/kg, s.c.), nor-BNI (10 mg/kg, i.p.) + vehicle, or nor-BNI + U69593 (where nor-BNI was given 24 h prior to vehicle or U69593). U69593 was dissolved in 20% propylene glycol in saline (Thompson et al., 2000), and the vehicle served as 20% propylene glycol alone. Injectable grade isotonic saline (0.9% NaCl) solution was used to dissolve cocaine (3.75 or 7.5 mg/kg, i.p.), nor-BNI, U50488 (10 mg/kg, i.p.), SalA (1 mg/kg, i.p.), morphine (10 mg/kg, s.c., for locomotor assay; 3 mg/kg., s.c., for CPP), and LiCl (160 mg/kg, i.p.). All drugs were given in a volume of 10 µl/g body weight. The concentration of drugs, route of administration, and treatment time are indicated in each figure legend. U69593 (catalog #U103), nor-binaltorphimine (nor-BNI; catalog #5.01087), U50448 (catalog #D8040), morphine (catalog #M8777), LiCl (catalog #L-4408), and propylene glycol (catalog #P4347) were obtained from Sigma-Aldrich. Salvinorin A (SalA) and cocaine were from RTI-11597-93-18, NIDA Drug Inventory Supply and Control System. The drug concentrations were selected according to previous studies that showed effective conditioning behaviors in mice (White et al., 2015; Bagdas et al., 2016; Mannangatti et al., 2017).
Preparation of synaptosomes and determination of DAT-mediated DA uptake and serotonin transporter-mediated 5-HT uptake
WT and DAT-Ala53 male mice received vehicle or KOR agonist U69593 (0.1 mg/kg, s.c.) and were returned to their home cage until experiment initiation. A separate group of animals received nor-BNI (10 mg/kg, i.p.) 24 h before vehicle or KOR agonist administration. After 2 h postinjection of vehicle or U69593, animals were decapitated, and brains were quickly removed and kept on ice. The dose of U69593 and preadministration time were selected based on the published study (Thompson et al., 2000) that showed stimulation of DAT activity by U69593. Dorsal and ventral striata were dissected on a cold dish and collected in 2 ml of ice-cold sucrose buffer (0.32 M sucrose in 5 mM HEPES, pH 7.4). The procedure for synaptosome preparation and DA and 5-HT uptake assay was adapted from our previous study (Ragu Varman et al., 2021a, b). The tissue was homogenized using a Teflon glass homogenizer, and the homogenized samples were centrifuged at 1,000 × g for 10 min at 4°C. After centrifugation, the supernatant was collected and centrifuged at 12,000 × g for 20 min. The resulting pellet was resuspended in ice-cold sucrose buffer. The protein concentration was quantified by a protein assay kit using BSA as a standard. Synaptosomes (30 μg of protein) were incubated in a total volume of 0.3 ml of KRH assay buffer, pH 7.4 (120 mM NaCl, 4.7 mM KCl, 1.2 mM KH2PO4, 1.2 mM MgSO4, 10 mM HEPES, and 10 mM d-glucose) containing 0.1 mM ascorbic acid and 0.1 mM pargyline. For DAT kinetic analysis, [3H] DA (20 nM) was mixed with unlabeled DA so that total DA concentration ranges from 25 to 1,000 nM. Total DAT-mediated DA uptake (DAT-specific + nonspecific DA uptake) was determined by measuring DA uptake in the presence of norepinephrine transporter (NET)-specific inhibitor nisoxetine (50 nM) to block NET-mediated DA uptake. To determine the nonspecific DA uptake, synaptosomes were preincubated with both the DAT- and NET-specific inhibitors (50 nM GBR-12909 and 50 nM nisoxetine) at 37°C for 10 min followed by the addition of 10 nM [3H] DA (catalog #NET673001MC; 18.32 Ci/mmol: PerkinElmer). [3H] DA uptake was terminated at 5 min with the addition of GBR-12909 followed by rapid filtration over GF-B filters using Brandel Cell Harvester (Brandel). In parallel, serotonin transporter (SERT)-mediated 5-HT uptake was determined. To determine the nonspecific 5-HT uptake, synaptosomes were preincubated with SERT-specific inhibitor (0.1 µM fluoxetine) at 37°C for 10 min. 5-HT uptake was initiated by the addition of 10 nM [3H] 5-HT (catalog #NET498001MC; 24.82 Ci/mmol: PerkinElmer), and the uptake was terminated at 5 min with the addition of fluoxetine followed by rapid filtration over GF-B filters. Filters were counted by liquid scintillation counter (MicroBeta2 LumiJET, PerkinElmer) to determine the radioactivity bound to the filters. Nonspecific DA uptake measured in the presence of GBR-12909 and nisoxetine was subtracted from the total [3H] DA uptake measured in the presence of NET inhibitor nisoxetine to determine the DAT-specific DA uptake. Nonspecific 5-HT uptake measured in the presence of fluoxetine was subtracted from the total [3H] 5-HT uptake to determine the SERT-specific 5-HT uptake.
Surface protein biotinylation and determination of total and surface DAT and pT53-DAT levels
Surface protein biotinylation was carried out as described in our previous study (Ragu Varman et al., 2021b). Dorsal or ventral striatal synaptosomes from respective treatment groups and genotypes (under experimental groups and drugs) were used for surface protein biotinylation. Synaptosomes were incubated with the membrane-impermeable EZ link NHS-sulfo-SS-biotin (catalog #21331, Pierce; 1 mg/1 mg protein) for 30 min at 4°C in ice-cold PBS/Ca/Mg buffer (138 mM NaCl, 2.7 mM KCl, 1.5 mM KH2PO4, 9.6 mM Na2HPO4, 1 mM MgCl2, 0.1 mM CaCl2, pH 7.3). At the end of the biotinylation and following centrifugation, the synaptosomes were suspended in the same buffer containing 100 mM glycine to quench excess NHS-sulfo-SS-biotin. The biotinylated synaptosomes were pelleted by centrifugation and suspended in ice-cold radioimmunoprecipitation assay (RIPA) buffer (10 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% SDS, and 1% sodium deoxycholate) supplemented with commercial cocktails of protease and phosphatase inhibitors [1 µg/ml aprotinin, 1 µg/ml leupeptin, 1 µM pepstatin, and 250 µM phenylmethylsulfonyl fluoride (catalog #P8340, Sigma-Aldrich); 10 mM sodium fluoride, 50 mM sodium pyrophosphate, 5 mM sodium orthovanadate, and 1 µM okadaic acid (catalog #P0044, Sigma-Aldrich)]. The suspended synaptosomes were solubilized by passing through a 26-gauge needle six times followed by incubation at 4°C for 60 min, and the lysate was collected by centrifugation at 25,000 × g for 30 min. Biotinylated proteins were isolated by incubating the lysate/supernatant overnight with NeutrAvidin Agarose resin (catalog #29201, Pierce). The NeutrAvidin Agarose resin was washed three times with RIPA buffer by brief centrifugation, and the biotinylated proteins (bound to the NeutrAvidin Agarose resin) were eluted by incubating the resin with 50 µl Laemmli sample buffer (62.5 mM Tris-HCl pH 6.8, 20% glycerol, 2% SDS, and 5% ß-mercaptoethanol). Proteins from total and bound fractions were separated by 7.5% SDS-PAGE and transferred to polyvinylidene difluoride membranes (Bio-Rad). To reduce the background signal, the membrane blots were blocked by incubating with 5% BSA–TBS–Tween 20 buffer for 1 h at room temperature. Total levels of DAT and pT53-DAT were determined using primary antibodies for DAT (mouse anti-dopamine transporter monoclonal, clone mAb16, catalog #MABN669, 1 : 2,000 dilution, MilliporeSigma; pT53-DAT rabbit anti-phospho-Thr53 dopamine transporter polyclonal, RRID:AB-2492078, catalog #p435-53, 1 : 2,000 dilution, PhosphoSolutions, respectively). After washing the membranes, corresponding species-specific horseradish peroxidase-conjugated secondary antibodies (RRID:AB-2313567, RRID:AB-10015289, 1:10,000 dilution, Jackson ImmunoResearch Laboratories) were used to visualize immunoreactive bands by ECL reagents (catalog #GERPN2236, Amersham Biosciences, GE HealthCare). The level of intracellular endoplasmic reticulum protein calnexin was determined by stripping the membranes and reprobing with anti-calnexin antibody (rabbit anti-calnexin polyclonal RRID:AB-312058, catalog #SPA-860, 1 : 2,000, Enzo Life Sciences) to validate that only plasma membrane surface proteins were biotinylated and to ensure that equal protein was loaded in the gel and transferred to membrane. Expression levels of arbitrary protein band densities of pT53-DAT, DAT, and calnexin were quantified using NIH ImageJ software (RRID:SCR-003070, version 1.48j). The protein bands were quantified by ensuring that results were within the linear range of the film exposure (HyBlot CL, Thomas Scientific). The relative arbitrary protein band densities obtained for pT53-DAT and DAT in total fractions were normalized with calnexin.
Locomotor activity measurement
Locomotor activity was measured in open-field activity monitoring chambers (RRID:SCR-014296, Med Associates; model ENV-510) in a soundproof box as described previously (Ragu Varman et al., 2021b). The horizontal activity was recorded for 60 min, and data were collected in 4 min bins as centimeters. WT and DAT-Ala53 mice were handled for 5 d before initiation of locomotor activity measurement. Mice received vehicle or U69593 alone or morphine alone or nor-BNI alone or nor-BNI + U69593 in a volume of 1 ml/kg and immediately placed in an activity monitor chamber, and activity was monitored for 60 min. nor-BNI was administered 24 h before vehicle or KOR agonist administration.
Conditioned place aversion and preference study
Drug conditioning was performed using a three-compartment apparatus enclosed in a sound-attenuating cubicle (Med Associates, ENV-3013). The apparatus consisted of white and black chambers (20 cm × 20 cm × 20 cm each), which were distinguished from one another by their different floor textures (white chamber with mesh and black chamber with rod: Med Associates, ENV-3013WM and ENV-3013BR). The smaller gray intermediate compartment with a smooth PVC floor and doors leading to the black and white chambers served as a barrier between the place conditioning chambers. The compartments were equipped with photobeam strips, and the location, locomotion, and time spent by the subject in each compartment were recorded using the program provided by the manufacturer. Mice were handled for 5 d before conditioning experiments. To determine conditioned preference, the time spent in the drug-paired side during the postconditioning test was subtracted from their time during the pretest. A drug is defined as aversive or producing CPA if the animal spends significantly less time in the drug-paired compartment than the vehicle-paired compartment during the test session. In contrast, a drug exhibits a rewarding effect (CPP) when the animal spends significantly more time in the drug-paired compartment than the vehicle-paired compartment during the test session.
(A)#U69593 conditioning
Conditioned place aversion (CPA) to KOR agonists U69593 was adapted as described previously (Chefer et al., 2013; Tejeda et al., 2013; White et al., 2015). On Day 1, mice were introduced into the central gray chamber and were given 15 min to explore both compartments for initial chamber preference (preconditioning test or pretest), and the amount of time spent in each compartment was recorded. Mice were conditioned with U69593 (0.1 mg/kg, s.c.) or vehicle on Day 2 following the pretest. For conditioning, mice were injected once daily for 6 d alternatively either with vehicle or U69593 and confined to a specific compartment for 15 min. Mice injected with U69593 were confined to the compartment where they spent more time during pretest (preferred side during pretest), and mice injected with vehicle were confined to the compartment where they spent less time during pretest (nonpreferred side identified in pretest). nor-BNI was administered 24 h before vehicle or KOR agonist administration. Vehicle control mice received vehicle daily for 6 d. Place conditioning preference was tested 24 h after the last conditioning session in a drug-free state by allowing the animals free access to all three compartments for 15 min (postconditioning test), and time spent in each compartment was recorded.
(B)#U50448, SalA, morphine, and LiCl conditioning
Conditioned place aversion (CPA) to KOR agonists U50448 and SalA, conditioned place preference (CPP) to µ-opioid receptor agonist morphine (morphine conditioning was done in nonpreferred side identified in pretest), and CPA to nonopioid LiCl were determined using the same design as described above with the exception that there were six separate drug-conditioning sessions spread across 3 d, twice daily. After testing for initial chamber preference on Day 1 (pretest), animals received saline in one chamber in the A.M. and U50448 or SalA or morphine or LiCl in the opposite chamber in the P.M. each day for 3 d (conditioning, 30 min duration). Horizontal activity was recorded during each conditioning session to determine whether treatments affect activity. Postconditioning test was conducted 24 h after the last conditioning session without drug treatment by allowing the animals free access to all three compartments for 15 min, and time spent in each compartment was recorded. The drug concentrations (see above, Experimental groups and drugs) were selected according to previous studies that showed effective conditioning behaviors in mice (White et al., 2015; Bagdas et al., 2016; Mannangatti et al., 2017).
(C)#Cocaine conditioning
A schematic of experimental timeline for cocaine conditioning and CPP determination is shown in Figure 6A, adapted as described previously (Medvedev et al., 2005). After 5 d of handling, on Day 6, mice were given 20 min to explore both compartments for initial chamber preference (pretest), and the amount of time spent in each compartment was recorded. Mice were administered designated doses of cocaine on Days 8, 11, 14, and 17 and confined to nonpreferred side identified in pretest for 20 min. Saline was administered on Days 7, 10, 13, and 16 and confined to the opposite compartment for 20 min. CPP tests were performed on Days 9, 12, 15, and 18, allowing animals to freely move and explore the three compartments for 20 min without cocaine or saline administration (Fig. 6A).
(D)#KOR agonist U69593 administration and cocaine conditioning
The experimental timeline for U69593 pretreatment and cocaine conditioning is given in Figure 7A. The impact of U69593 pretreatment on cocaine-induced CPP was determined using the same experimental design as described above in (C) with the exception that U69593 was administered 60 min prior to cocaine or saline administration and conditioning.
Statistical analysis
All values are expressed as mean ± standard deviation (SD). Microsoft Excel (Mac version 16.84) and GraphPad Prism 10 (RRID:SCR 002798 GraphPad) were used for data analysis and statistical evaluations and to generate graphical representations. ImageJ 1.54g (National Institutes of Health) was used to digitally quantify bands from digitized films. Figures are presented as bar graphs showing all points representing each subject or repeats. One-way or two-way or repeated mixed analysis of variance was used, followed by post hoc testing for pairwise comparisons. Student's t tests were used to compare two datasets. A value of p < 0.05 was considered statistically significant. Specific statistical analyses, statistical data, and the significance of each experiment are included in the Results section. In addition, comparisons between groups with p-values and the number of samples used in each experiment are reported in each figure legend.
Results
Endogenous DAT-Thr53 phosphorylation is required for KOR-mediated upregulation of DAT activity, Vmax, and surface expression
Using an in vitro cell culture model and ex vivo rat striatal synaptosomes, we showed that KOR agonist SalA upregulates DAT function, Vmax, and surface expression in an ERK1/2-dependent manner (Kivell et al., 2014). Additionally, mutating Thr53 to nonphosphorylatable Ala in DAT blunted KOR agonist-mediated upregulation DAT activity and surface expression (Mayer et al., 2025). In our subsequent study, we demonstrated that ERK1/2 inhibition failed to inhibit/alter striatal DAT activity in DAT-Ala53 knock-in mice (Ragu Varman et al., 2021b). Therefore, it was essential to investigate how activation of KOR modulates DAT function and whether in vivo KOR-mediated upregulation of endogenous DAT function requires DAT phosphorylation. Given the facts that (1) KOR-triggered ERK1/2 is mediating DAT upregulation, (2) DAT-T53 phosphorylation site is the signature canonical motif for ERK1/2, and (3) the absence of effects of ERK1/2 inhibition on DAT activity in DAT-Ala53 knock-in mice, we sought to investigate the role of DAT-T53 phosphorylation in KOR-mediated DAT upregulation in both ventral and dorsal striata of DAT-Ala53 knock-in mice expressing normal DAT function and expression but lacking Thr53 phosphorylation (DAT-Ala53) by comparing with WT controls. KOR agonist, U69593 (0.1 mg/kg, s.c.) was systemically administered to DAT-Ala53 and WT control mice to activate KOR. DAT-mediated DA uptake in ventral and dorsal striata was determined using synaptosomes 2 h U69593 postadministration. The dose, time, and route of administration were selected from previous publication demonstrating enhanced DAT activity by U69593 (Thompson et al., 2000). When compared with vehicle, U69593 administration significantly enhanced DAT-mediated DA uptake in both ventral and dorsal striata of WT mice (ventral striatum, p = 0.0003; dorsal striatum, p < 0.0001) but not in DAT-Ala53 mice (ventral striatum, p = 0.9969; dorsal striatum, p = 0.8526; Fig. 1A,B). Two-way ANOVA showed significant effect of genotype (ventral striatum, F(1,4) = 59.35, p < 0.0015; dorsal striatum, F(1,4) = 37.86, p < 0.0035) and treatment (ventral striatum, F(3,12) = 5.708, p < 0.0115; dorsal striatum, F(3,12) = 2.276, p < 0.1320) with significant genotype × treatment interaction (ventral striatum, F(3,12) = 7.860, p < 0.0036; dorsal striatum, F(3,12) = 13.84, p < 0.0003). To further confirm that KOR is involved in U69593-mediated upregulation of DAT function, we examined the effect of the long-acting KOR antagonist nor-BNI (Endoh et al., 1992). We injected KOR antagonist nor-BNI (10 mg/kg, i.p.) 24 h before U69593 administration. While nor-BNI alone did not affect the basal DAT function in WT, it prevented U69593-mediated upregulation of DAT function in both ventral (p < 0.0018) and dorsal striata (p < 0.0001). nor-BNI administration alone or with U69593 did not affect DAT function in DAT-Ala53 mice significantly (ventral striatum, p < 0.1960; dorsal striatum, p < 0.9811; Fig. 1A,B). We also examined KOR agonist effect on SERT function in ventral and dorsal striata of DAT-Ala53 to determine if Thr53 mutation in DAT affects SERT function indirectly. Consistent with published observation (Schindler et al., 2012), systemic U69593 administration enhanced SERT-mediated 5-HT uptake in ventral and dorsal striata of WT. Similar to WT, enhanced ventral and dorsal striatal SERT activity was observed in DAT-Ala53 mice following systemic U69593 administration (Fig. 1C,D). The two-way ANOVA revealed a significant effect of U69593 on ventral and dorsal striatal SERT activity in both WT and DAT-Ala53 mice (ventral striatum, F(1,4) = 16.40, p < 0.016; dorsal striatum, F(1,4) = 9.25, p = 0.038) with no significant genotype effect (ventral striatum, F(1,4) = 0.90, p = 0.395; dorsal striatum, F(1,4) = 0.38, p = 0.567) or genotype × treatment interaction (ventral striatum, F(1,4) = 0.000, p = 0.927; dorsal striatum, F(1,4) = 0.01 = 7, p = 0.895). These results indicate the specific effect of DAT-Ala53 mutation and in turn the influence of DAT-Thr53 phosphorylation on KOR-mediated DAT regulation.
Effect of U69593 on DAT and SERT activities in ventral and dorsal striatal synaptosomes. nor-BNI–sensitive DAT activity increases in the (A) ventral and (B) dorsal striatal synaptosomes of WT, but not DAT-Ala3 mice following systemic U69593 administration. Drug administrations, synaptosome preparations, and DAT-specific 3H-labeled DA uptake assays were performed as described in Materials and Methods. DAT-specific [3H] DA uptake is given in pmol/mg protein/min and presented as mean ± SD. Each data point represents an individual mouse. When compared with vehicle control, U69593 increased DAT activity in the WT mice (ventral striatum, ***p = 0.0003; dorsal striatum, ****p = 0.0001) but not in the DAT-Ala53 mice (ventral striatum, p = 0.997; dorsal striatum, p = 0.853). nor-BNI blocked U69593 effect on DAT activity in the WT mice (ventral striatum, ^^p = 0.002; dorsal striatum, ^^^^p = 0.0001). Significant difference between WT-U69593 and DAT-Ala53-U69593 (ventral striatum, ##p = 0.002; dorsal striatum, ####p = 0.0001). ns, nonsignificant when compared between specific pairs as indicated in figures. WT (n = 5) and DAT-Ala53 (n = 5) mice. Data were analyzed by two-way RM ANOVA with Tukey’s multiple comparisons. U69593 increased SERT activity in ventral (C) and dorsal (D) striatal synaptosomes of WT (ventral striatum, *p = 0.026; dorsal striatum, *p = 0.042) and DAT-Ala53 (ventral striatum, *p = 0.022; dorsal striatum, *p = 0.037) mice. Data were analyzed by two-way ANOVA with Sidak's multiple comparisons.
To examine the kinetic properties of DAT-mediated DA uptake, saturation analysis was performed in synaptosomes obtained from ventral and dorsal striata of WT and DAT-Ala53 mice administered with U69593 or vehicle. Consistent with our previous study (Ragu Varman et al., 2021b), there was no effect of genotype on the Vmax and Km of DAT-mediated DA uptake in ventral and dorsal striata of vehicle administrated mice (Fig. 2). Compared with vehicle administration, U69593 administration in WT mice increased DAT Vmax in both ventral (Fig. 2A,E) and dorsal striata (Fig. 2B,F; ventral striatum ± SD, vehicle: 16.08 ± 0.3625 pmol/mg/min, U69593: 23.85 ± 1.117 pmol/mg/min; dorsal striatum ± SD, vehicle: 18.62 ± 0.870 pmol/mg/min, U69593: 24.40 ± 1.651 pmol/mg/min) without altering Km (ventral striatum ± SD, vehicle: 168.93 ± 18.451 nM, U69593: 157.16 ± 6.018 nM; dorsal striatum ± SD, vehicle: 122.78 ± 43.928 nM, U69593: 109.69 ± 16.919 nM). However, in DAT-Ala53 mice, compared with vehicle control, systemic U69593 did not produce significant change in DAT Vmax in both ventral (Fig. 2C,E) and dorsal striata (Fig. 2D,F; ventral striatum ± SD, vehicle: 15.22 ± 0.931 pmol/mg/min, U69593: 16.20 ± 1.491 pmol/mg/min; dorsal striatum ± SD, vehicle: 19.74 ± 0.900 pmol/mg/min, U69593: 19.98 ± 0.886 pmol/mg/min). Km was unaltered in ventral and dorsal striata (ventral striatum ± SD, vehicle: 145.36 ± 21.553 nM, U69593: 165.76 ± 14.165 nM; dorsal striatum ± SD, vehicle: 128.10 ± 21.287 nM, U69593: 128.80 ± 30.153 nM). Two-way ANOVA of Vmax showed significant effect of genotype (ventral striatum, F(1,2) = 93.10, p < 0.011; dorsal striatum, F(1,4) = 12.2, p < 0.025) and treatment (ventral striatum, F(1,2) = 90.0, p < 0.011; dorsal striatum, F(1,4) = 25.9, p < 0.007) with significant genotype × treatment interaction (ventral striatum, F(1,2) = 21.2, p < 0.011; dorsal striatum, F(1,4) = 72.97, p < 0.001). These results suggest that the phosphorylation of Thr53 is required to enhance endogenous DAT activity in ventral and dorsal striata following KOR activation.
U69593-induced increase in DAT Vmax is blunted in the ventral and dorsal striata of DAT-Ala53 mice. Drug administrations, synaptosome preparations, and DAT-specific DA transport kinetic analysis were performed as described in Materials and Methods. Maximal velocity (Vmax) and substrate (DA) apparent affinity (Km) of DAT were determined by using Michaelis–Menten equation (Prism). DA uptake values (pmol/mg protein/min) are presented as mean ± SD. WT (n = 3) and DAT-Ala53 (n = 3). When compared with vehicle, U69593 increased DAT Vmax significantly in (A, E) ventral striatum (**p = 0.003) and in (B, F) dorsal striatum (*p = 0.012) of WT mice. U69593 failed to increase DAT Vmax in (C, E) ventral (p = 0.765) and (D, F) dorsal striata (p = 0.999) of DAT-Ala53 mice. U69593 did not alter Km in ventral and dorsal striata of both WT and DAT-Ala53 mice. ^^^^p < 0.0001 (ventral striatum); ^p = 0.0206 (dorsal striatum) WT-U69593 versus DAT-Ala53-U69593. ns, nonsignificant. Comparisons between specific pairs are indicated in the figures. Data were analyzed using two-way ANOVA with Bonferroni’s multiple comparisons.
To study the mechanism underlying the elevated DAT function following U69593 administration, we sought to determine whether systemic U69593-mediated upregulation of DAT activity and Vmax is accompanied by increased pT53-DAT and cell surface DAT. Membrane surface protein biotinylation and immunoblotting with specific antibodies to pT53-DAT, total DAT (DAT), and calnexin were used to quantify total and surface resident of DAT, pT53-DAT, and calnexin in ventral and dorsal striata of WT and DAT-Ala53 mice (Fig. 3). Systemic administration of U69593 in WT mice significantly increased pT53-DAT without altering total DAT and intracellular marker calnexin in ventral and dorsal striata (Fig. 3A,C,E). Consistent with increased DA uptake and Vmax, systemic U69593 administration significantly increased surface (biotinylated) levels of DAT and pT53-DAT in ventral and dorsal striata of WT mice (Fig. 3B,D,F). In contrast to WT, systemic U69593 administration failed to alter surface DAT level in the ventral and dorsal striata of DAT-Ala53 mice (Fig. 3B,F) without changes in total DAT (Fig. 3A,E). Two-way ANOVA revealed significant genotype effect (ventral striatum: pT53-DAT, F(1,4) = 176.2, p = 0.0002; DAT, F(1,4) = 12.70, p = 0.023; dorsal striatum: pT53-DAT, F(1,4) = 66.7, p < 0.0001; DAT, F(1,4) = 20.08, p = 0.011) and treatment effect (ventral striatum: pT53-DAT, F(1,4) = 13.43, p = 0.2153; DAT, F(1,4) = 33.24, p < 0.005; dorsal striatum: pT53-DAT, F(1,4) = 46.69, p = 0.0024; DAT, F(1,4) = 29.53, p < 0.006) with significant genotype × treatment interaction in both ventral and dorsal striata (ventral striatum: pT53-DAT, F(1,4) = 13.18, p = 0.0221; DAT, F(1,4) = 25.95, p = 0.007; dorsal striatum: pT53-DAT, F(1,4) = 46.61, p = 0.0024; DAT, F(1,4) = 17.66, p = 0.014). As reported in our previous publication (Ragu Varman et al., 2021b), the pT53-DAT antibody did not detect pT53-DAT apart from some nonspecific bands in ventral and dorsal striata of DAT-Ala53 mice. Notably, the intracellular endoplasmic reticulum marker calnexin is undetectable in biotinylated fractions, indicating that synaptosomes were intact and intracellular DAT protein was not biotinylated and recovered along with surface biotinylated DAT or pT53-DAT (Fig. 3B). Thus, the observed biotinylated DAT represents resident DAT on the surface plasma membrane. Collectively, these results demonstrate that in vivo DAT-Thr53 phosphorylation in the ventral and dorsal striata enhances DAT surface density and confirms a direct relationship between KOR-mediated DAT upregulation and DAT-Thr53 phosphorylation.
U69593 enhances total and surface DAT-Thr53 phosphorylation and surface DAT expression in ventral and dorsal striata of WT but not the DAT-Ala53 mice. Drug administrations and synaptosome preparations are given in Materials and Methods. Using indicated specific antibodies, total and surface expression of pT53-DAT, DAT and calnexin were visualized and quantified following surface protein biotinylation. Representative immunoblots from three independent experiments show changes in (A) total and (B) surface (biotinylated) pT53-DAT (∼60–55 kDa), DAT (∼65 kDa) and calnexin (∼90 kDa) in the ventral and dorsal striata of WT and DAT-Ala53 mice. The specificities of DAT and pT53-DAT immunoreactive bands by the antibodies used were established in our previous publication (Ragu Varman et al., 2021b). NS, nonspecific band. Bar graph shows quantified arbitrary band densities of (C) total pT53-DAT and (E) DAT and (D) biotinylated surface pT53-DAT and (F) DAT. Each data point represents an individual mouse, and the data are presented as mean ± SEM. WT (n = 3) and DAT-Ala53 (n = 3) mice. Compared with vehicle control, U69593 increased total and surface pT53-DAT levels and surface DAT expression without altering the total expression of DAT and calnexin in the ventral and dorsal striata of WT but not DAT-Ala53 mice. *p = 0.05, **p < 0.01, ^^p < 0.01, ^^^p < 0.001. ns, nonsignificant. Comparisons between specific pairs are indicated in the figures. Data were analyzed by two-way ANOVA with Bonferroni’s multiple comparisons.
DAT-Thr53 phosphorylation is required for KOR agonist-induced locomotor suppression
It is known that systemic administration of KOR agonists in rodents induces reduced locomotor activity (White et al., 2015; Brust et al., 2016). Since KOR activation reduces DA release (Werling et al., 1988; Spanagel et al., 1990, 1992; Devine et al., 1993; Brust et al., 2016), and our results showed that KOR activation also increases DA clearance through DAT-Thr53 phosphorylation-dependent trafficking mechanism (Figs. 1–3), we hypothesized that DAT-Thr53 phosphorylation may have a direct causal role in the behavioral effects of KOR agonists. To test our hypothesis, we compared the effect of systemic U69593 on the modulation of locomotor activity in WT and DAT-Ala53 mice. We selected U69593 dose (1.0 mg/kg) that is shown to affect locomotor activity (White et al., 2015). Consistent with published studies (White et al., 2015), when compared with vehicle administration, systemic U69593 administration significantly decreased locomotor activity (total distance traveled in 60 min) of WT mice (p < 0.0001; Fig. 4B). However, U69593 administration did not have any significant effect on the locomotor activity of DAT-Ala53 mice (p > 0.1357; Fig. 4B). Locomotor activity measured in 4 min bins is shown in Figure 4, C and D. Two-way ANOVA analysis of the total activity for the duration of 60 min showed a significant effect of genotype (F(1,14) = 13.94, p < 0.0022) and U69593 administration (F(1,14) = 12.70, p < 0.0031) with significant genotype × U69593 interaction (F(1,14) = 8.482, p < 0.0114). Pretreatment with long-acting KOR antagonist nor-BNI attenuated U69593-induced locomotor suppression in WT mice (p < 0.0009) suggesting the involvement of KOR activation. Moreover, the significant difference between WT and DAT-Ala53 mice with respect to the effect of U69593 on locomotor activity (p < 0.0006) demonstrates an essential role of DAT-Thr53 phosphorylation in KOR-mediated locomotor suppression.
U69593-induced suppression of locomotor activity is blunted in DAT-Ala53 mice. A, Representative activity traces of 60 min locomotor activity, (B) total distance traveled for 60 min, and (C, D) time course of locomotor activity measured in 4 min bins over 60 min period postadministration of vehicle or U69593 or nor-BNI or pre–nor-BNI following U6953 in WT and DAT-Ala53 mice are shown. Systemic administration of vehicle or U69593 and locomotor monitoring were performed as described in Materials and Methods. Each data point represents an individual mouse, and the data are presented as mean ± SEM. WT (n = 6) and DAT-Ala53 (n = 6) mice. Comparison of U69593 effects on locomotor activity between genotypes revealed that U69593 decreased locomotor activity in WT (***p = 0.001) but not in DAT-Ala53 (p = 0.999) mice. nor-BNI blocked U69593-induced locomotor suppressing effect in WT (^^^p = 0.001). ##p = 0.0001 WT-U69593 versus DAT-Ala53-U69593. ns, nonsignificant (p = 0.999). Comparisons between specific pairs are indicated in the figures. Data were analyzed by two-way ANOVA with Bonferroni’s multiple comparisons.
Phosphorylation of DAT-Thr53 is required for KOR agonist-induced CPA
It is well known that KOR agonists produce CPA in rodents (Shippenberg and Herz, 1986; Zhang et al., 2005; McLaughlin et al., 2006a; Chefer et al., 2013; Tejeda et al., 2013; White et al., 2015; Brust et al., 2016). To investigate whether DAT-T53 phosphorylation-dependent DAT regulation plays a role in KOR agonist-triggered CPA, we tested the effect of systemic administration of KOR agonists (U69593, U50488, and SalA) on WT and DAT-Ala53 mice (Fig. 5). To keep continuity and comparisons with published literature, the dose of KOR agonists and conditioning procedures were adapted from published CPA protocols (Shippenberg and Herz, 1986; Zhang et al., 2005; McLaughlin et al., 2006a; Chefer et al., 2013; Tejeda et al., 2013; White et al., 2015; Brust et al., 2016) and described in Materials and Methods. As expected, all three agonists tested produced CPA in WT mice. Multiple comparisons indicate that WT mice administered with U69593 or U50488 or SalA showed significant CPA (U69593, p < 0.0001; U50488, p = 0.0268; SalA, p = 0.017) compared with vehicle administration (Fig. 5A,C). Pretreatment with long-acting KOR antagonist, nor-BNI blocked U69593 triggered CPA (p = 0.0001) suggesting the involvement of KOR activation (Fig. 5A). Interestingly, U69593 and U50488 did not induce significant CPA in DAT-Ala53 mice (U69593,p = 0.404; U50488, p = 0.268). On the other hand, SalA elicited significant CPA in DAT-Ala53 mice (p = 0.001; Fig. 5C). However, the SalA-induced CPA was significantly lower in DAT-Ala53 mice compared with WT mice (p = 0.01). There was significant genotype difference between WT and DAT-Ala53 mice in response to KOR agonists (U69593, p = 0.0013; U50488, p = 0. 0032; SalA, p = 0.01). Two-way mixed ANOVA showed significant effect of treatments (U69593, F(1,31) = 25.67, p < 0.0001; U50488, F(1,7) = 18.2, p = 0.004; SalA, F(1,7) = 7.363, p = 0.030) and significant differences in the CPA produced by all three KOR agonists between genotypes (U69593, F(1,31) = 8.389, p < 0.01; U50488, F(1,7) = 18.2 p = 0.004; SalA, F(1,7) = 20.60, p = 0.003) with significant genotype × treatment interaction except for SalA (U69593, F(1,31) = 9.044, p < 0.001; U50488, F(1,7) = 9.36, p < 0.018; SalA, F(1,7) = 2.793, p = 0.139). Furthermore, there were no significant differences in posttest movement counts between genotypes following treatment with all three KOR agonists (Fig. 5B,D). These observations directly link DAT-Thr53 phosphorylation and consequent DAT functional regulation to KOR agonist-induced aversive behavior.
U69593, U50488, and SalA-induced conditioned place aversion are blunted in DAT-Ala53 mice. Mice were conditioned with indicated KOR agonists as described in Materials and Methods. Each data point represents an individual mouse, and bars are plotted as average values ± S.D. A, The postconditioning effects of vehicle and U69593 and (B) the posttest movement counts are shown. Number of mice used for each genotype: vehicle (n = 10) and U69593 (n = 8). U69593 produced significant CPA in WT (****p < 0.0001) but not in DAT-A53 (ns = 0.453) mice in comparison with corresponding vehicle controls. A significant difference in the CPA effect of U69593 between WT and DAT-Ala53 mice (##p = 0.0013). nor-BNI blocked U69593-induced CPA in WT mice (^p = 0.025). C, The postconditioning effects of U50488 and SalA and (D) the posttest movement counts are presented. U50488 and SalA produced significant CPA in WT (***p < 0.0003 and ^p = 0.045, respectively). DAT-A53 mice displayed no CPA to U50488 (ns = 0.999) and attenuated CPA to SalA (§p = 0.015). ####p < 0.0001 WT-U50488 versus DAT-Ala53-U50488. ^^p = 0.006 DAT-Ala53-saline versus DAT-Ala53-SalA. ns, nonsignificant compared between specific pairs as indicated in figures. Number of mice used: WT-saline (n = 8), WT-U50488 (n = 9), WT-SalA (n = 9), DAT-Ala53-saline (n = 8), DAT-Ala53-U50488 (n = 10), and DAT-Ala53-SalA (n = 9). Data were analyzed by two-way ANOVA mixed-effects analysis with Bonferroni’s multiple comparisons.
Phosphorylation of DAT-Thr53 modulates cocaine’s efficacy to elicit CPP
Recently, we reported that in the absence of DAT-Thr53 phosphorylation, cocaine's ability to inhibit DAT-mediated DA uptake and to trigger hyperlocomotion is significantly blunted in DAT-Ala53 knock-in mice (Ragu Varman et al., 2021b). We aimed to investigate whether DAT-T53 phosphorylation plays a role in the rewarding properties of cocaine using the CPP test. The CPP test measures an animal's ability to learn and recall a specific contextual environment associated with the rewarding effects of cocaine experience in a drug-free condition. To evaluate the sensitivity to cocaine and the acquisition of context related to rewarding effects of cocaine, we adapted a CPP procedure where we conducted postconditioned tests (CPP tests) sequentially at specific intervals for the duration of conditioning with different doses of cocaine (Fig. 6A; Medvedev et al., 2005). After the pretest, cocaine (3.75 or 7.5 mg/kg cocaine, i.p.) or saline (i.p.) was administered to both WT and DAT-Ala53 mice and confined to a compartment for 20 min The postconditioned test (CPP test) was conducted on the day immediately following first pairing, which consisted of one saline and one cocaine conditioning session, as shown in Figure 6A, by allowing animals to move freely in all three compartments for 20 min without the administration of cocaine or saline. Significant genotype differences were evident in cocaine-induced CPP depending on the dose used (Fig. 6B). When the sequential CPP data were analyzed for the entire procedure, we observed WT mice exhibited dose and CPP test-dependent progressive increases in cocaine-induced CPP. In contrast, DAT-Ala53 mice failed to display CPP test-dependent progressive increase in cocaine-induced CPP when conditioned with low cocaine dose (3.75 mg/kg) but did display CPP test-dependent progressive increase in cocaine-induced CPP when conditioned with high cocaine dose (7.5 mg/kg; Fig. 6B). Two-way RM ANOVA showed significant effects of cocaine dose on CPP tests (F(3,5) = 48.03, p < 0.0001) and genotypes (F(1,7) = 69.59, p < 0.0001), with significant genotype × cocaine dose–CPP test interaction (F(2.9,20.64) = 4.53, p = 0.014). We conducted separate two-way ANOVA analyses of CPP tests to determine the effect of cocaine dose on conditioning, genotype, and the interaction between treatment and genotype in each CPP test. In the first CPP test, neither WT mice nor DAT-Ala53 mice displayed any conditioned place preference at both doses of cocaine. However, significant effects of cocaine dose and genotype evolved during CPP tests 2, 3, and 4. Notably, WT mice developed significant CPP at a low dose of 3.75 mg/kg cocaine, while DAT-Ala53 did not show CPP at any CPP tests at this dose (Fig. 6B). Additionally, compared with WT mice, DAT-Ala53 mice exhibited significantly reduced CPP at higher dose of 7.5 mg/kg cocaine, as illustrated in Figure 6B. Two-way ANOVA mixed model revealed significant effect of treatment (CPP test 3, F(1.2,8.41) = 104.2, p < 0.0001; CPP test 4, F(1.4,9.7) = 60.04, p < 0.0001) and genotype (CPP test 3, F(1,7) = 14.22, p = 0.007; CPP test 4, F(1,7) = 32.26, p = 0.0008), with significant genotype × treatment–CPP test interaction (CPP test 3, F(1.9,13.4) = 7.27, p = 0.0078; CPP test 4, F(1.6,11.39) = 6.04, p ≤ 0.0202). The CPP scores observed in CPP test 4 did not differ significantly from CPP test 3 in both WT and DAT-Ala53 mice. These findings suggest that the reduced ability of cocaine to inhibit DAT activity in DAT-Ala53 mice (Ragu Varman et al., 2021b) directly links pT53-DAT to the lack of cocaine-induced CPP (3.75 mg/kg) or diminished cocaine-induced CPP (7.5 mg/kg). This indicates that lower cocaine affinity for DAT correlates with reduced CPP in DAT-Ala53 mice.
Cocaine-induced place preference is suppressed in DAT-Ala53 mice. A, Experimental timeline of place preference following cocaine conditioning. After pretest, mice were conditioned with 3.75 or 7.5 mg/kg cocaine (i.p.) for 20 min. The postconditioned test (CPP test) was conducted the day following one saline and one cocaine conditioning day, as shown in A and as described in Materials and Methods. Number of mice used for each genotype: 3.75 mg/kg (n = 8) and 7.5 mg/kg (n = 8) cocaine. B, The results of four CPP tests were shown after each subsequent set of saline and cocaine conditioning. Each data point represents an individual mouse, and bars are plotted as average values ± SD. Compared with WT mice conditioned with cocaine, DAT-A53 mice display attenuated cocaine-induced CPP at the following CPP tests: 3.75 mg/kg cocaine produced no CPP at CPP tests 2 (****p < 0.0001), 3 (****p ≤ 0.0004), and 4 (***p = 0.0007), and 7.5 mg/kg cocaine-induced CPP was significantly reduced at CPP tests 2 (^^p < 0.0071), 3 (^^p < 0.0074), and 4 (^^p < 0.0069). ns, nonsignificant. Comparisons between specific pairs are indicated in the figure. Data were analyzed using two-way ANOVA with Tukey’s multiple comparisons.
DAT-Thr53 phosphorylation drives KOR-induced potentiation of cocaine-induced CPP
Activation of DYN/KOR pathways by stress or treatment with KOR agonist U50488 prior to conditioning with cocaine potentiates cocaine-induced CPP (McLaughlin et al., 2006a; McLaughlin et al., 2006b; Bruchas et al., 2010; Schindler et al., 2012). Given that the DAT-Thr53 phosphorylation is required for KOR-mediated regulation of DAT, locomotor suppression, and CPA (Figs. 1–5), we hypothesize that DAT-Thr53 phosphorylation may play a direct role in KOR-mediated potentiation of cocaine-induced CPP. To test this hypothesis, we adapted cocaine-induced CPP protocol as shown in Figure 7A, using a cocaine dose of 7.5 mg/kg which produced CPP in both WT and DAT-Ala53 mice (Fig. 6B). To investigate the effects of KOR activation, animals were preadministered with U69593 (0.1 mg/kg, s.c.) or vehicle and were kept in their home cages for 60 min before conditioning, as illustrated in Figure 7A. This 60 min pretreatment duration was chosen because previous studies have demonstrated that this timing potentiates cocaine-induced CPP without causing CPA (McLaughlin et al., 2006a). Consistent with published findings (McLaughlin et al., 2006a), preadministration of U69593 (0.1 mg/kg, s.c.) 60 min prior to conditioning did not result in CPP or CPA in either WT or DAT-Ala53 mice across all four CPP tests conducted examined (Fig. 7B). Overall data analyzed using two-way ANOVA mixed model demonstrated a significant effect of treatment CPP test (F(5,79) = 24.74, p < 0.0001) and genotype (F(1,14) = 49.21, p < 0.0001). There was also a significant interaction between genotype and treatment in the CPP tests (F(6,77.28) = 5.34, p = 0.0002). We performed separate two-way ANOVA analyses of CPP tests to compare the effect of treatment, genotype, and treatment × genotype interaction in each CPP test. No significant differences in CPP scores between genotypes were observed after U69593 treatment and cocaine conditioning during CPP test 1 (treatment, F(1,49) = 0.41, p = 0.6372; genotype, F(1,28) = 2.98, p = 0.0953; genotype × treatment, F(2,56) = 0.61, p = 0.549). During CPP test 2, differences in treatment (F(1,27) = 8.58, p = 0.0013) and genotype (F(1,14) = 6.46, p = 0.0235) were evident with no significant genotype × treatment interaction (F(1,28) = 1.11, p = 0.3447). Consistent with results presented in Figure 6B, 7.5 mg cocaine conditioning produced significant CPP in CPP tests 3 and 4 in WT and reduced cocaine-induced CPP in DAT-Ala53 mice (Fig. 7B). Notably, pretreatment of U69593 significantly potentiated cocaine-induced CPP in WT during CPP tests 3 and 4, but this effect was not observed in DAT-Ala53 mice (Fig. 7B). Two-way ANOVA mixed model revealed significant effect of treatment (CPP test 3, F(1.7, 24) = 34.89, p < 0.0001; CPP test 4, F(1.8, 25) = 85.03, p < 0.0001) and genotype (CPP test 3, F(1,14) = 58.60, p < 0.0001; CPP test 4, F(1,14) = 41.19, p < 0.0001) with significant genotype × treatment CPP test interaction (CPP test 3, F(1.8,25) = 8.52, p = 0.0019; CPP test 4, F(1.6,20) = 16.86, p = 0.0001). These results combined with the data presented in Figures 1–5 further reiterate our conclusion that DAT-Thr53 phosphorylation drives KOR-mediated upregulation of DAT activity and surface expression and thus plays a direct role in KOR agonist-induced locomotor suppression, CPA, and potentiation of cocaine-induced CPP.
KOR activation-mediated potentiation of cocaine-induced conditioned place preference is blunted in DAT-Ala53 mice. A, Schematic of experimental timeline for the preadministration of U69593, cocaine conditioning, and place preference tests. B, Four CPP tests were shown after each subsequent set of saline and cocaine conditioning from WT and DAT-A53 mice, as shown in A. After the pretest, vehicle (10 ml/kg, s.c.) or U69593 (0.32 mg/kg, s.c.) was administered and kept in a home cage for 60 min each day before conditioning (20 min) saline (i.p.) or cocaine (7.5 mg/kg, i.p.). The postconditioned test (CPP test) was conducted on the day following one saline and one cocaine conditioning (24 h later) as described in Materials and Methods. Each data point represents an individual mouse, and bars are plotted as average values ± SD. The number of mice used was 15 for each treatment group for each genotype. In comparison with WT, attenuated cocaine-induced CPP was evident in DAT-Ala53 in CPP tests 3 (*p < 0.0032) and 4 (**p < 0.0038). U69593 pretreatment potentiates cocaine-induced CPP in WT (CPP test 3, #p < 0.0258; CPP test 4, ##p < 0.0055) but not in DAT-Ala53. U69593 effect on cocaine-induced CPP is different between WT and DAT-Ala53 in CPP tests 2 (^p < 0.0174), 3 (^^^^p < 0.0001), and 4 (^^^^p < 0.0001). ns, nonsignificant. Comparisons between specific pairs are indicated in the figure. Data were analyzed using two-way ANOVA with Tukey’s multiple comparisons.
DAT-Thr53 phosphorylation does not play a role in morphine-induced hyperlocomotion
We examined the effect of morphine, a μ-opioid receptor agonist, on the modulation of locomotor activity to ensure that the lack of locomotor suppression in DAT-Ala53 mice following U69593 administration (Fig. 4) is not a generalized phenomenon. As shown in Figure 8, systemic morphine administration (10 mg/kg, s.c.) triggered locomotor activity in both WT and DAT-Ala53 mice. There were significant morphine treatment effects (F(1,9) = 37.62, p = 0.0002) in both genotypes without genotype effects (F(1,9) = 5.084, p = 0.0506). We observed that DAT-Ala53 mice display higher morphine-induced locomotor activity (p = 0.024) compared with WT mice. These results indicate that the absence of U69593-mediated locomotor suppression in DAT-Ala53 mice is not a generalized phenomenon and is specific to U69593.
Morphine-triggered locomotor activity is similar in WT and DAT-Ala53 mice. Administration of morphine and its effect on locomotor activity were described in Materials and Methods, and the data are presented as mean ± SD. WT (n = 10) and DAT-Ala53 (n = 9). A, Representative activity traces of 60 min locomotor activity and (B) total distance traveled for 60 min after administering vehicle or morphine in WT and DAT-Ala53 mice are shown. Bonferroni’s multiple comparative analysis of the impact of morphine on locomotor activity showed that morphine evoked significant locomotor activation in WT (**p = 0.003) and DAT-Ala53 (^^^^p = 0.0001) mice. Data were analyzed using two-way ANOVA mixed-effects analysis with Tukey’s multiple comparisons.
Morphine-induced CPP and LiCl-induced CPA are independent of DAT-Thr53 phosphorylation
Attenuated aversive effects of KOR agonists (Fig. 5), blunted cocaine-induced CPP (Fig. 6B), and lack of KOR potentiation of cocaine-induced CPP (Fig. 7B) in DAT-A53 mice might occur from altered learning and performance of the conditioned response or generalized malaise. To rule out these possibilities, we examined the conditioned response of morphine and nonopioid lithium chloride (LiCl). As presented in Figure 9A, and consistent with published literature (Leite-Morris et al., 2014; Mannangatti et al., 2017), morphine (3 mg/kg, s.c.) conditioning produced place preference in WT mice. Similar to WT mice, DAT-Ala53 mice also exhibited morphine-induced CPP. Morphine-induced CPP did not differ between genotypes. Two-way ANOVA indicated significant morphine conditioning effect in both genotypes (F(1,26) = 35.60, p = 0.0001) and no significant genotype effect (F(1,26) = 5.28, p = 0.474) with no significant genotype × treatment interaction (F(1,26) = 0.757, p = 0.392). The posttest movement counts were not different between WT and DAT-Ala53 (p = 0.089, t = 1.28, df = 14; Fig. 9B). After a 2-week washout period, we utilized the cohorts of WT and DAT-Ala53 mice used for measuring KOR agonist-triggered CPA (Fig. 5) to assess the conditioned aversive effects of another nonopioid drug, lithium chloride. As predicted (Shippenberg et al., 1988; Bagdas et al., 2016), LiCl (160 mg/kg, i.p.) produced conditioned aversion in WT (Fig. 9C). Surprisingly, the conditioned aversive effect of LiCl was intact in DAT-Ala53 mice (Fig. 9C). LiCl-induced CPA did not differ between genotypes (Fig. 9C). Two-way ANOVA indicated significant LiCl conditioning effect (F(1,28) = 28.70, p = 0.0001) in both genotypes without significant genotype × treatment interaction (F(1,28) = 0.0101, p = 0.9206). Moreover, both WT and DAT-Ala53 exhibited similar posttest movement counts (p = 0.648, t = 0.466, df = 14; Fig. 9D). Overall, intact morphine-induced CPP and LiCl-induced CPA observed in DAT-Ala53 mice indicate that attenuation of KOR-mediated aversive behavior, blunted cocaine-induced CPP, and lack of KOR potentiation of cocaine-induced CPP observed in DAT-Ala53 mice are not due to altered learning and/or performance of conditioned response or generalized malaise.
DAT-Ala53 mice display morphine-induced CPP and LiCl-induced CPA similar to WT. The same cohorts of WT and DAT-Ala53 mice used above in Figure 5 were used to condition with morphine or LiCl or saline, and their effects on place preference or aversion were determined as described in Materials and Methods. A, The posttest of conditioned effects of morphine, (C) LiCl, and total locomotor distance traveled during posttest (B) morphine and (D) LiCl in comparison with WT and DAT-Ala53 are presented. Each data point represents an individual mouse, and bars are plotted as averages ± SD. Conditioning with morphine produced significant CPP in both WT (****p = 0.0001) and DAT-Ala53 (^p = 0.017) mice and showed no differences in morphine-induced CPP between WT and DAT-Ala53 mice (ns = 0.943). WT (n = 8) and DAT-Ala53 (n = 8). LiCl conditioning produced significant CPA in both WT (**p = 0.0047) and DAT-Ala53 (^^p = 0.0032) mice and showed no difference between WT and DAT-Ala53 (ns = 0.309). Number of mice used: WT-saline (n = 8), WT-LiCl (n = 8), and DAT-Ala53-saline (n = 8), and DAT-LiCl (n = 8). Comparison of specific pairs is indicated in the figures. Data were analyzed by two-way ANOVA with Tukey’s multiple comparisons.
Discussion
The primary findings from the current study reveal that KOR-mediated DAT-Thr53 phosphorylation is a primary causal posttranslational mechanism underlying the effects of KOR agonists. Specifically, in vivo Thr53 phosphorylation of DAT occurring in DA terminals dictates (1) KOR-mediated upregulation of DAT functional kinetics and surface expression, (2) locomotor suppression induced by KOR agonists, (3) aversive behaviors triggered by KOR agonists, and (4) enhanced cocaine-induced CPP caused by KOR agonists.
Functional DAT activity governs extracellular DA dynamics (Giros et al., 1996) and KOR activation has been shown to enhance DAT function through Thr53 phosphorylation in cell culture (Kivell et al., 2014; Mayer et al., 2025) revealing a potential role of in vivo DAT-Thr53 phosphorylation in KOR-mediated DA clearance and animal behavior. However, an investigation of this hypothesis is lacking to date. To test this hypothesis, in the current study, we used DAT-Ala53 knock-in mouse model, which lacks DAT-Thr53 phosphorylation (Ragu Varman et al., 2021b) to determine the role of DAT-Thr53 phosphorylation in KOR-mediated DAT regulation and behaviors. Systemic KOR agonist U69593 administration elevated nor-BNI–sensitive DAT-mediated DA uptake and transport velocity (Vmax) without altering substrate affinity (Km) in WT mice, but not in DAT-Ala53 mice. U69593 also enhanced total pT53-DAT levels and surface DAT and pT53-DAT levels in wild-type mice but had no effect in DAT-Ala53 mice, suggesting that KOR-mediated in vivo DAT upregulation is dependent on phosphorylation of DAT-Thr53, which is also the ERK1/2 site (Ragu Varman et al., 2021b). DA is also cleared via norepinephrine transporter (NET; Moron et al., 2002). We performed DA uptake assays in the presence of NET-specific inhibitor nisoxetine; thus, the results represent the impact of KOR modulation on DAT-specific DA uptake.
KOR agonists cause locomotor suppression in mice, which is absent in KOR knock-out mice (White et al., 2015; Brust et al., 2016). Indeed, systemic KOR agonist U69593 suppressed the spontaneous open-field activity in WT mice, an effect blocked by KOR antagonist nor-BNI pretreatment, confirming the role of KOR activation in locomotor suppression. Remarkably, systemic U69593 failed to alter locomotor activity in DAT-Ala53 mice. However, µ-opioid receptor agonist morphine administration stimulated locomotor activity in DAT-Ala53 mice similar to WT mice, suggesting DAT-Thr53 phosphorylation is specifically involved in the effects of KOR agonist. DAT plays a critical role in regulating extracellular DA and locomotor activity (Giros et al., 1996). Our findings that the absence of KOR-mediated DAT functional regulation and KOR agonist-induced locomotor suppression in DAT-Ala53 mice suggest that the DAT-Thr53 phosphorylation mediated by KOR signaling increases DA clearance and, in conjunction with the evidence that KOR signaling/activation also decreases DA release (Heidbreder et al., 1993), may consequently cause locomotor suppression.
KOR agonists produce CPA in rodents (Shippenberg and Herz, 1986; McLaughlin et al., 2006a; Tejeda et al., 2013; White et al., 2015), which requires KOR expression in DA neurons (Chefer et al., 2013). In agreement, systemic administration of three KOR agonists, U69593, U50488, and SalA, produced CPA in WT mice. KOR antagonist, nor-BNI, prevented U69593-induced CPA, suggesting the specific involvement of KOR in KOR agonist-induced CPA. Surprisingly, U69593 and U50488 completely failed to elicit CPA, while SalA produced significantly attenuated CPA in DAT-Ala53 mice, suggesting DAT-Thr53 phosphorylation is required for KOR agonist-induced CPA.
Our studies provide insights into the role of DAT-Thr53 phosphorylation in regulating cocaine’s affinity for DAT and consequent cocaine CPP, a behavioral assay used to examine the rewarding effects of cocaine. Our previous study demonstrated that DAT-Ala53 renders DAT to lower cocaine affinity and cause suppressed cocaine-induced locomotor activity (Ragu Varman et al., 2021b). In our current findings, we observed that DAT-Ala53 mice exhibited blunted CPP at lower cocaine doses and significantly reduced CPP at higher cocaine doses compared with WT mice, correlating with our previous finding that the reduced affinity of cocaine for DAT in DAT-Ala53 mice is directly associated with reduced reward response. Previous studies have demonstrated that administering KOR agonists before cocaine conditioning potentiates cocaine-induced CPP (McLaughlin et al., 2006a; Schindler et al., 2010). KOR agonist pretreatment significantly potentiated cocaine CPP in wild-type mice, but not in DAT-Ala53 mice suggesting a mechanistic role for KOR-mediated DAT-Thr53 phosphorylation in enhancing cocaine reward.
Deficiencies in learning and performance of the conditioned response or preference for natural reward in DAT-Ala53 mice might contribute to observed alterations in cocaine CPP and/or KOR-mediated CPA, independent of Thr53 phosphorylation or DAT regulation. We compared the conditioned response to µ-opioid receptor agonist morphine and nonopioid LiCl in WT and DAT-Ala53 mice. Morphine is known to induce CPP (Mannangatti et al., 2017) while LiCl produces CPA (Shippenberg et al., 1988). Our data show that morphine-induced CPP and LiCl-induced CPA in DAT-Ala53 mice are similar to that of WT mice, suggesting that DAT-Ala53 mice are capable of learning and responding to conditioned drug effects. Additionally, we previously reported that DAT-Ala53 mice show normal natural reward preference in two-bottle sucrose choice preference test (Ragu Varman et al., 2021b). Thus, the absence of KOR agonist-mediated CPA and potentiation of cocaine CPP in DAT-Ala53 mice cannot be attributed to defects in their learning and performance of the conditioned response or natural reward preference.
It has been reported that KOR-triggered p38 MAPK activation via GRK3/arrestin in ventral tegmental area of DA neurons is required for CPA in mice (Ehrich et al., 2015). However, mice lacking ß-arrestin-2 still show aversion with KOR agonists, similar to normal mice (White et al., 2015). Additionally, KOR activation in dorsal raphe nucleus by stress stimuli and downstream GRK3-dependent activation of p38 MAPK, as well as serotonin transporter (SERT) stimulation in the ventral striatal serotonergic terminals, are required to elicit CPA (Land et al., 2009; Schindler et al., 2012). Although these studies demonstrate the possible involvement of both DA and 5-HT systems in KOR-mediated behaviors, the specific relationship to the complex KOR-induced CPA requires further investigation.
KOR-mediated activation of p38 MAPK may regulate functional aspects of several substrates within DA or 5-HT neurons via phosphorylation, and the nature of these substrates is unknown. In DAT-Ala53 mice, KOR activation increases serotonin transporter (SERT) activity in the ventral and dorsal striata. The lack of CPA responses to KOR agonists in these mice indicates that KOR-mediated SERT upregulation does not independently cause aversive effects. Instead, it likely interacts with dopamine transmission through DAT phosphorylation at Thr53. Since (1) KOR agonist-mediated DAT upregulation is ERK1/2-dependent (Kivell et al., 2014), (2) ERK1/2 inhibition failed to modulate DAT function in DAT-Ala53 mice (Ragu Varman et al., 2021b), and (3) DAT-Thr53 is a phosphorylation site for ERK1/2 (Gorentla et al., 2009), we interpret that KOR agonists upregulate DAT function at DA terminals through ERK1/2-dependent DAT-Thr53 phosphorylation, eliciting CPA, locomotor suppression, and modulation of cocaine CPP.
The possibility of compensatory changes arising from constitutive expression of Thr53Ala mutant DAT contributing to observed changes in KOR agonist-induced CPA, locomotor suppression, and cocaine CPP cannot be ruled out. Identifying such compensatory mechanisms and their neuronal substrates will provide further insights into the downstream effectors of KOR/ERK1/2-mediated DAT-Thr53 phosphorylation. Stress triggers DYN release, activates KOR signaling, and promotes aversive and proaddictive behaviors (Bruchas et al., 2007; Land et al., 2008; Bruchas et al., 2010). Investigating whether stress influences these behaviors through DYN/KOR-DAT-Thr53 phosphorylation is important.
A report by Calipari et al. (2017) showed enhanced DAT activity and DAT-Thr53 phosphorylation in association with increased ERK1/2 phosphorylation, enhanced cocaine potency to inhibit DA clearance, and a concomitant potentiation of cocaine-induced CPP in estrous females. It is suggested that the increased DA neuron activity in the ventral tegmental area during estrous cycle may reflect neuronal mechanisms underlying elevated DAT-Thr53 phosphorylation, cocaine affinity, and CPP, thus emphasizing the existence of sexual dimorphism in DAT-Thr53 phosphorylation and its role in cocaine-reward learning processes (Calipari et al., 2017). Although these studies demonstrate a correlative association between changes in DAT-Thr53 phosphorylation and behavior, the causal link between endogenous DAT-Thr53 phosphorylation and the behavioral effects of cocaine in females remains unclear. Furthermore, studies have shown that KOR activation produces CPA and potentiates cocaine-induced CPP in both males and females (Abraham et al., 2018), indicating the complex interactions among sex, DAT-Thr53 phosphorylation, and the behavioral effects of KOR agonists and cocaine which could be explored in future studies using DAT-Ala53 female mice to understand the dimorphic role of sex in DYN/KOR-mediated DAT-Thr53 phosphorylation signaling that translates to behavioral outcomes produced by KOR agonists and cocaine.
KOR agonists are known to produce several adverse effects, including sedation, aversion, dysphoria, depression, and enhanced rewarding effects of psychostimulants, which limit their therapeutic application (reviewed in Bruijnzeel, 2009; Wee and Koob, 2010; Tejeda et al., 2012; Margolis and Karkhanis, 2019). Our studies demonstrated that DAT-Thr53 phosphorylation mediated by KOR activation is a primary causal posttranslational mechanism driving the effects of KOR agonist on DAT regulation, locomotor suppression, aversion, and cocaine CPP modulation. These findings suggest that phosphorylation of DAT-Thr53 and other DAT-regulatory motifs are potential therapeutic targets to alter DA neurotransmission and form a basis for developing effective pharmacotherapeutics that minimize the adverse effects associated with KOR agonism.
Footnotes
This work was supported by the National Institutes of Health RO1 DA054694 (S.R. and L.D.J.).
The authors declare no competing financial interests.
D.R.J.’s present address: Centre for Integrative Omics Data Science, Yenepoya (deemed to be university), Mangalore 575018, India.
- Correspondence should be addressed to Lankupalle D. Jayanthi at ljayanthi{at}vcu.edu and lankupalle.jayanthi{at}vcuhealth.org or Sammanda Ramamoorthy at sramamoorthy{at}vcu.edu and sammanda.ramamoorthy{at}vcuhealth.org.
