A Behavioral Economic Analysis of the Effects of Rimcazole on Reinforcing Effects of Cocaine Injection and Food Presentation in Rats

Abstract

RATIONALE AND OBJECTIVES:

Rimcazole, a σ-receptor antagonist with affinity for the dopamine transporter (DAT), decreases rates of cocaine self-administration at doses lower than those that affect food-reinforced responding. As response rates are multiply determined, behavioral-economic analyses were used to provide measures of the reinforcing effectiveness of cocaine and food after rimcazole treatment. Further, effects of combinations of the DAT inhibitor, methylphenidate, and σ-receptor antagonists (BD1008, BD1063) were compared to those of rimcazole to assess mechanism of rimcazole effects.

METHODS:

Male Sprague-Dawley rats were trained to lever press with food reinforcement (one or three 20-mg sucrose pellets) or cocaine injection (0.1 or 0.32 mg/kg) under fixed-ratio (FR) 5-response schedules. Drugs or vehicle were administered (i.p.) 5-min before sessions in which FR value was increased from 5 to 80. Economic demand functions were generated from effects of FR value (price) on intake (consumption), with the parameters of demand, consumption at no cost (Q₀) and sensitivity to price (essential value, EV), derived.

RESULTS:

Rimcazole dose-dependently decreased Q₀ and EV at both cocaine doses/injection. In contrast, rimcazole had no effect on these parameters at either food amount. Combinations of methylphenidate and the σ-receptor antagonists decreased Q₀ at the lower cocaine dose/injection, but had no effect on EV; these treatments were ineffective on both economic parameters at the higher cocaine dose/injection and at either food amount.

CONCLUSIONS:

Though the drug combinations only replicated rimcazole’s effects incompletely, the present results suggest a specific decrease in the reinforcing effects of cocaine due to dual DAT σ-receptor blockade.

INTRODUCTION

Cocaine is thought to exert its behavioral effects through actions at the dopamine transporter (DAT). However, several studies have noted that the pharmacological effects of cocaine are not solely due to actions at the DAT, and that some compounds that act at the DAT do not have actions equivalent to those of cocaine (Brodnik et al. 2017; Reith et al. 2015). Previous studies have indicated that cocaine binds to sigma (σ) receptors at concentrations approximating those at which it has DAT affinity (Lever et al. 2016). Specific molecular interactions between σ receptors and the DAT have been determined with coimmunoprecipitation and bioluminescence resonance energy transfer assays (Hong et al. 2017). That study indicated that cocaine-induced DAT conformation changes were enhanced by preincubation with σ-receptor ligands, and that those changes were blocked by the specific σ-receptor antagonist, CM304. Other studies have documented direct involvement of σ-receptors in the molecular actions of methamphetamine (Sambo et al. 2017).

Several behavioral studies have indicated that σ receptors may be involved in the behavioral effects of cocaine (see review by Katz et al. 2017). In one study, rimcazole (cis-9-[3-(3,5-dimethyl-1-piperazinyl) propyl] carbazole dihydrochloride), a σ-receptor antagonist that also has affinity for the DAT (Izenwasser et al. 1993; Valchar and Hanbauer 1993) decreased the maximum self-administration of cocaine in rats at doses that did not significantly decrease food-maintained responding (Hiranita et al. 2011). Further, combinations of DAT inhibitors and σ-receptor antagonists also decreased maximal cocaine self-administration at dose combinations that did not affect responding maintained by food reinforcement (Hiranita et al. 2011). Those results suggested that combined actions at the DAT and at σ receptors underlie specific decreases in the reinforcing effects of cocaine.

One caveat to that suggestion is that response rates maintained in self-administration procedures are multiply determined and the degree to which the effectiveness of the drug as a reinforcing stimulus impacts those response rates can be unclear (Iglauer and Woods 1974; Johanson and Schuster 1975); see also (Banks et al. 2008). Therefore, the present study looked to another indication of reinforcing strength suggested by adaptations of behavioral economic measures to operant behavior (Hursh and Silberberg 2008). In that analysis, economic demand curves that relate consumption of a commodity to its price are determined. In applications to self-administration, consumption translates to drug intake and price is the behavioral requirement for each injection. Estimates of the sensitivity of intake to price and consumption at a hypothetical zero price are made from the demand curve. These measures have been applied to provide estimates of reinforcing effectiveness for a variety of abused drugs (see review by (Hursh et al. 2005). Further, those measures have been applied to comparisons of the effects of various drug treatments on behaviors maintained by cocaine (Bentzley and Aston-Jones 2015; Bentzley et al. 2014; Cosgrove and Carroll 2002; Oleson et al. 2011; Porter-Stransky et al. 2017; Wade-Galuska et al. 2011; Zanettini et al. 2018). In laboratory studies of the effects of drug pretreatments, it has proved useful to assess a demand curve in a single experimental session (Bentzley and Aston-Jones 2015; Kearns and Silberberg 2016; Oleson and Roberts 2009; Zanettini et al. 2018).

Thus, the present study applied a behavioral economic analysis of demand curves to the effects of rimcazole. Additionally, as the behavioral effects of rimcazole may be mediated by actions at both the DAT and σ receptors, its effects were compared to those of combinations of methylphenidate, as a prototype DAT inhibitor, combined with either of two σ-receptor antagonists, BD1008 and BD1063, as has been done previously (Hiranita et al. 2011).

METHODS

Subjects.

Twenty-seven male Sprague-Dawley rats obtained from Charles River Laboratories (Wilmington, MA) served as subjects. They were housed under temperature- and humidity-controlled conditions and maintained on a 12-hour/12-hour light/dark cycle with lights on at 07:00 AM. The subjects were acclimated to the animal housing facility for at least one week before any experimental procedures. After acclimation, weights of rats were maintained at approximately 350 g by adjusting daily food rations (~12 g of Harlan Rodent Chow) at least one hr after daily sessions. Care of the subjects was in accordance with the guidelines of the National Institutes of Health and the National Institute on Drug Abuse Intramural Research Program Animal Care and Use Program, which is fully accredited by AAALAC International.

Surgery.

The subjects in the cocaine self-administration group were surgically prepared with chronic indwelling external-jugular or femoral vein catheters connected to a subcutaneously placed backmount/external port (313–000BM-10–5UP, PlasticsOne Inc., Roanoke, VA) that exited at the mid-scapular region. Catheter (RJV-1, SAI Infusion Technology, Lake Villa, IL) implantation was performed under sterile conditions using intraperitoneal (i.p.) ketamine (60.0 mg/kg) and xylazine (12.0 mg/kg) or inhaled isoflurane anesthesia. The catheter was closed with a screw cap (C313CAC, PlasticsOne Inc.) on the backmount. Catheters were infused daily with 0.2 ml of a sterile saline solution containing heparin (30.0 IU/ml) and enrofloxacin (5 mg/kg) to minimize the likelihood of infection and the formation of clots or fibroids. All animals were allowed to recover from surgery for approximately one week before cocaine self-administration procedures were initiated.

Apparatus.

Operant-conditioning chambers (modified ENV-008CT; Med Associates, St. Albans, VT) that measured 25.5 cm × 32.0 cm × 25.0 cm were enclosed within sound-attenuating cubicles equipped with a fan for ventilation and white noise to mask extraneous sounds. Within the chamber, on the front wall, were two response levers, 5.0 cm from the midline and 4.0 cm above the grid floor. A downward displacement of either lever with a force approximating 20 g (0.20 N) produced an audible “feedback” click of a relay mounted behind the front wall of the chamber and defined a response. Three light-emitting diodes (LEDs) were located in a row above each lever. A pellet dispenser (ENV-203; Med Associates) delivered 20-mg sucrose food pellets (Bio-Serv, Flemington, NJ) to a receptacle mounted behind accessible from a 5.0 × 5.0 cm opening in the front wall midway between the two levers and 2.0 cm above the floor.

A syringe driver (model 22; Harvard Apparatus, Holliston, MA) placed above each chamber delivered injections of specified volumes and durations from a 10-ml syringe. The syringe was connected by Tygon tubing to a single-channel fluid swivel (375 Series Single Channel Swivels; Instech Laboratories, Inc., Plymouth Meeting, PA) which was mounted on a balance arm above the chamber. Tygon tubing from the swivel connecting the subject’s catheter was protected by a surrounding metal spring.

Procedures.

All subjects were trained during once-daily 1-hour sessions to respond under a fixed-ratio 1-response (FR 1) schedule in which each response produced a 20-mg sucrose food pellet (Bio-Serv, Flemington, NJ). Pellet deliveries were followed by 20-s timeout (TO) periods, during which all lights were off and responses had no scheduled consequences other than the feedback click. When rates of responding did not vary by more than 30% over two consecutive sessions (Zanettini et al. 2018), the response requirement was changed to FR 5. When responding was stable at this FR requirement, the subjects were divided into cocaine self-administration (N=15) and food-reinforcement (N=12) groups. Procedures for the food-reinforcement group were in all important aspects identical to those described above for cocaine self-administration with the exception that completions of the response requirement produced the 20-mg sucrose food pellets.

Cocaine self-administration sessions were conducted in two-hour daily sessions conducted seven days per week. Daily sessions continued until the response rates and patterns of responding showed no substantial session-to-session trends. During these sessions, the LEDs above the right lever were illuminated when cocaine injections were available. Completion of the FR 5 schedule requirement turned off the LEDs and activated the infusion pump, delivering a dose of 0.32 mg/kg. A 20-s TO, during which LEDs were off and responses had no scheduled consequences other than the feedback click, started with the injection. After the TO, the LEDs were re-illuminated, and responding again had scheduled consequences. Once response rates were relatively stable, the session was divided into five 20-min components, each preceded by a 2-min TO. Training proceeded with the FR 5 requirement for all components.

Drug testing began when response rates were again stable within and across sessions. During the test sessions, different doses of drugs or combinations of drugs were injected i.p. 5 min before sessions. Additionally, the number of responses required per infusion was increased sequentially (FR 5, 10, 20, 40 and 80) across the five components, as described previously (Kearns and Silberberg 2016; Zanettini et al. 2018). Training with an FR 5 schedule in all five-components continued for at least two training sessions between successive test sessions, and test sessions were only conducted if there was less than 30% variation in the total number of cocaine injections in the two preceding training sessions.

Drugs.

The drugs used in the present study were as follows: (−)-cocaine hydrochloride (Sigma-Aldrich), methylphenidate hydrochloride (NIDA), rimcazole dihydrochloride (synthesized in the Medicinal Chemistry Section, NIDA), BD1008 (Tocris Bioscience, Ellisville, MO), and BD1063 (Tocris Bioscience). Drug pretreatments were administered i.p. 5 min before sessions. Effects of rimcazole (3.2, 10 and 32 mg/kg) or its vehicle, and combinations of methylphenidate (1 mg/kg) and BD1008 or BD1063 (3.2 and 10 mg/kg) or saline injections were examined. All drug solutions were prepared fresh daily in 0.9% NaCl. Rimcazole was dissolved in DMSO and Tween-80 and diluted to final volumes with sterile water. Testing of drug pretreatments was conducted first with the greater unit dose of cocaine or food amount followed by the lower values. Pretreatment times and doses of drugs used in the present study were chosen based on published data demonstrating selective effects of rimcazole and the drug combinations (Hiranita et al. 2010; Katz et al. 2003).

Data Analysis.

Response rates were determined for each subject by dividing the number of responses delivered in each component by the duration of the component, excluding the TO periods. Demand curves were plotted with number of reinforcers earned (consumption) as a function of FR requirement (price) and the data were fitted using the exponential model of Hursh and Silberberg (2008):

$$logQ=log{Q}_0+k\left(e^{-a\left(Q_{0}C\right)}-1\right),$$ [1],

where Q represents consumption of the reinforcer, C represents cost (presently the number of required responses), Q₀ represents consumption at no cost, α represents a fitted parameter related to the decline in consumption with increased cost, and k is a scaling constant reflecting the consumption range. The value of α is inversely related to reinforcer effectiveness and can be converted to “essential value” (EV), which is directly related to reinforcer effectiveness, by the following equation:

$$EV=\frac{1}{\left(100\times a\times k^{1.5}\right)}$$ [2],

where α and k are as described for equation [1]. The percent change in Q₀ and EV, derived from the fits, for the different drug pretreatments relative to vehicle or saline pretreatments were calculated and normalized to control values to facilitate comparisons across reinforcer magnitude and type. Percent changes in Q₀ and EV from control (vehicle, saline) values were analyzed using a one-way repeated measures ANOVA, with Tukey’s post-hoc test where applicable. Changes in response rates were analyzed using two-way repeated measures ANOVA with FR value and treatments as factors.

RESULTS

Performances maintained by both cocaine injection or food presentation were characteristic of temporal patterns under FR schedules with responding following a short pause and a rapid response rate until the completion of the required number of responses (data not shown). Response rates varied little from the first to last components when the FR value was unchanged at five responses. With 0.1 mg/kg/injection of cocaine those rates in the first and last components averaged 0.168 (±0.027) and 0.153 (±0.026) responses/sec, respectively. With 20 mg food pellets those response rates in the first and last components averaged 1.62 (±0.065) and 1.59 (±0.069) responses/sec, respectively. None of the changes in response rates across the five components were statistically significant. At the higher cocaine dose/injection (F_4,40=8.70; p<0.001), but not the higher food amount, there was a significant effect of component with post-hoc tests indicating no differences in response rates in all components following the first.

Effects of Rimcazole on Cocaine Self-Administration

Increasing FR value increased response rates maintained by 0.1 mg/kg/injection of cocaine from 0.272 (±0.079) at FR 5 to 0.411 (±0.116) responses/sec at FR 20, with a decrease at the highest FR value to 0.164 (±0.095) responses/sec (Figure 1A, filled symbols). The response rates at FR 5 and FR 80 were approximately 66% and 40%, respectively, of that obtained at FR 20. Rimcazole produced dose-related decreases in response rates maintained by 0.1 mg/kg/injection of cocaine across the range of FR values assessed (Figure 1A, open symbols). As with controls, response rate as a function of FR value had an inverted U-shape after rimcazole treatment, with maxima less than that obtained after vehicle injection, typically at FR 20. Statistical analysis indicated that there was a significant effect of FR value (F_4,20=3.28, p=0.032) an effect of rimcazole dose that approached significance (F_3,15=2.62, p=0.089), and no interaction of the two factors.

Figure 1:

Effects of rimcazole on various aspects of performances with changes in FR schedule value at different reinforcing doses of cocaine (0.1 or 0.32 mg/kg per injection). Error bars indicate standard errors of the mean. Top row: effects of FR value on response rates (responses/sec) either after injection of vehicle or different doses of rimcazole. Abscissae: FR value; Ordinates: Responses/sec. Second row: demand curves of intake (consumption) as a function of FR value (price). Abscissae: log FR value; Ordinates: log intake (mg/kg). Symbols depict actual values; lines are fits to the points using Equation [1]. Third row: Effects of rimcazole dose on the Q₀ parameter of the fitted demand equation expressed as percentage change from vehicle treatment. Fourth row: Effects of rimcazole dose on the EV parameter derived from α of the fitted demand equation as derived from Equation [2], expressed as percentage change from vehicle treatment.

The demand curves for cocaine (0.1 mg/kg/injection) derived from the response-rate data showed monotonic decreases in drug intake (consumption) with increases in FR value (price), giving Q₀ and EV values of 45.3 (±6.58) and 9.76 (±2.33), respectively (Figure 1B filled circles; Table 1). A similar demand curve was obtained after each dose of rimcazole with differences in the fitted Q₀ and EV values (Figure 1B; open symbols). The fitted model accounted for ≥97% of the variance in the data (Table 1).

Table 1:

Fitted Consumption (Q₀) and Essential Value (EV) at different cocaine doses per injection or food amounts per presentation derived from the exponential equations for demand and the variance accounted for by the model after vehicle or various treatments.

Cocaine 0.1 mg/kg/injection
Treatment	Q0 (SEM)	EV (SEM)	R2
Vehicle	45.3 (±6.58)	9.76 (±2.33)	1.00
Rimcazole 3.2	45.9 (±6.65)	9.38 (±2.24)	1.00
Rimcazole 10	32.0 (±4.64)	9.70 (±2.31)	0.97
Rimcazole 32	16.2 (±2.35)	3.40 (±0.811)	0.98
Cocaine 0.3 mg/kg/injection
Treatment	Q0 (SEM)	EV (SEM)	R2
Vehicle	12.8 (±1.03)	4.75 (±0.941)	0.98
Rimcazole 3.2	12.7 (±0.952)	3.26 (±0.534)	0.98
Rimcazole 10	11.7 (±0.871)	3.34 (±0.548)	0.98
Rimcazole 32	8.67 (±3.31)	2.12 (±0.348)	1.00
Food 20 mg/delivery
Treatment	Q0 (SEM)	EV (SEM)	R2
Vehicle	89.3 (±11.2)	12.9 (±1.09)	0.97
Rimcazole 3.2	85.4 (±10.7)	13.3 (±1.12)	0.96
Rimcazole 10	114 ±(14.2)	13.1 (±1.11)	0.98
Rimcazole 32	92.6 ±(11.6)	11.8 (±0.998)	0.99
Food 60 mg/delivery
Treatment	Q0 (SEM)	EV (SEM)	R2
Vehicle	81.6 (±9.31)	10.9 (±1.53)	0.99
Rimcazole 3.2	113 (±12.8)	11.3 (±1.58)	0.98
Rimcazole 10	113 (±12.9)	17.4 (±2.45)	0.99
Rimcazole 32	111 (±12.7)	10.2 (±1.43)	0.99
Cocaine 0.1 mg/kg/injection
Treatment	Q0 (SEM)	EV (SEM)	R2
Saline – Saline	47.0 (±2.56)	10.7 (±1.69)	1.00
MPD 1.0 – Saline	42.0 (±2.28)	11.4 (±1.79)	1.00
MPD 1.0 – BD1008 3.2	42.5 (±2.31)	10.7 (±1.68)	0.99
MPD 1.0 – BD1008 10	36.2 (±1.97)	9.84 (±1.55)	1.00
Cocaine 0.3 mg/kg/injection
Treatment	Q0 (SEM)	EV (SEM)	R2
Saline – Saline	13.9 (±1.67)	6.55 (±1.58)	0.95
MPD 1.0 – Saline	12.8 (±1.55)	5.60 (±1.35)	0.91
MPD 1.0 – BD1008 3.2	11.4 (±1.38)	6.81 (±1.64)	0.93
MPD 1.0 – BD1008 10	11.7 (±1.41)	4.98 (±1.20)	0.86
Food 20 mg/delivery
Treatment	Q0 (SEM)	EV (SEM)	R2
Saline – Saline	102.5 (±15.5)	12.7 (±0.488)	0.97
MPD 1.0 – Saline	122 (±18.4)	13.2 (±0.508)	0.94
MPD 1.0 – BD1008 3.2	98.6 (±7.96)	14.6 (±0.559)	0.99
MPD 1.0 – BD1008 10	112 (±16.9)	13.1 (±0.502)	0.96
Food 60 mg/delivery
Treatment	Q0 (SEM)	EV (SEM)	R2
Saline – Saline	92.9 (±12.2)	14.1 (±1.33)	0.99
MPD 1.0 – Saline	84.1 (±11.1)	16.3 (±1.54)	0.99
MPD 1.0 – BD1008 3.2	82.1 (±10.8)	12.9 (±1.22)	1.00
MPD 1.0 – BD1008 10	817.1 (±10.7)	15.6 (±1.47)	0.99
Cocaine 0.1 mg/kg/injection
Treatment	Q0 (SEM)	EV (SEM)	R2
Saline – Saline	47.2 (±5.56)	8.32 (±1.37)	1.00
MPD 1.0 – Saline	38.0 (±4.47)	11.2 (±1.85)	1.00
MPD 1.0 – BD1063 3.2	37.3 (±4.39)	9.04 (±1.49)	0.99
MPD 1.0 – BD1063 10	34.4 (±4.05)	11.5 (±1.89)	0.99
Cocaine 0.3 mg/kg/injection
Treatment	Q0 (SEM)	EV (SEM)	R2
Saline – Saline	14.9 (±1.03)	11.5 (±5.29)	0.89
MPD 1.0 – Saline	15.2 (±1.05)	11.8 (±5.41)	0.95
MPD 1.0 – BD1063 3.2	15.5 (±1.07)	12.0 (±5.53)	0.89
MPD 1.0 – BD1063 10	17.2 (±1.18)	13.3 (±6.11)	0.84
Food 20 mg/delivery
Treatment	Q0 (SEM)	EV (SEM)	R2
Saline – Saline	103 (±15.5)	12.7 (±0.488)	0.97
MPD 1.0 – Saline	122 (±18.4)	13.2 (±0.508)	0.94
MPD 1.0 – BD1063 3.2	135 (±20.5)	13.4 (±0.516)	0.96
MPD 1.0 – BD1063 10	120 (±18.1)	13.2 (±0.506)	0.97
Food 60 mg/delivery
Treatment	Q0 (SEM)	EV (SEM)	R2
Saline – Saline	92.9 (±12.2)	14.1 (±1.33)	0.99
MPD 1.0 – Saline	84.1 (±11.1)	16.3 (±1.54)	0.99
MPD 1.0 – BD1063 3.2	77.3 (±10.2)	16.5 (±1.56)	1.00
MPD 1.0 – BD1063 10	85.2 (±11.2)	16.2 (±1.53)	0.99

Rimcazole dose-dependently decreased fitted Q₀ values compared to control for 0.1 mg/kg/injection of cocaine (Figure 1C; Table 1). These effects of rimcazole were statistically significant (F_3,15=6.50, p=0.005). Rimcazole also decreased the calculated EV values compared to control for 0.1 mg/kg/injection cocaine (Figure 1D; Table 1), with ANOVA indicating that these effects were statistically significant (F_3,15=4.19, p=0.024).

At 0.32 mg/kg/injection of cocaine response rates were uniformly lower than those obtained at the 0.1 mg/kg dose of cocaine. Increasing FR value increased response rates maintained by the higher dose from 0.069 (±0.011) at FR 5 to 0.157 (±0.035) responses/sec at FR 40, with a decrease at the highest FR value to 0.103 (±0.049) responses/sec (Figure 1E, filled circles). The response rates at FR 5 and FR 80 were approximately 44% and 66%, respectively, of that obtained at FR 40. Rimcazole decreased the response rates maintained, particularly at FR 40 (Figure 1E, open circles). Statistical analysis indicated a significant effect of FR value (F_4,24=4.42, p=0.008) and rimcazole dose (F_3,18=9.88, p<0.001), but no interaction of the two factors.

The cocaine demand curves fitted to the data for 0.32 mg/kg/injection of cocaine also showed monotonic decreases in cocaine intake with increases in FR value (Figure 1F), and gave Q₀ and EV values of 12.8 (±1.03) and 4.75 (±0.941), respectively (Table 1). The demand curves obtained after doses of rimcazole were shaped similarly to control, with differences in fitted Q₀ and EV values, and accounted for >95% of the variance. Rimcazole dose-dependently decreased fitted Q₀ values compared to control (Figure 1G; Table 1). These effects of rimcazole were statistically significant (F_3,18=3.46, p=0.038). Rimcazole also produced dose-related decreases in the calculated EV values compared to control (Figure 1H; Table 1) that were statistically significant (F_3,18=4.19, p=0.021).

Effects of Rimcazole on Food-Reinforced Responding

Increasing FR value increased response rates maintained by 20 mg/delivery of food pellets from 1.20 (±0.254) at FR 5 to 1.73 (±0.326) responses/sec at FR 10; of higher FR values response rates progressively decreased, both without (Figure 2A, filled symbols) and with rimcazole pretreatment (Figure 2A, open symbols). Statistical analysis indicated that there was a significant effect of FR value (F_4,20=12.1, p<0.001), and no effects of rimcazole dose or the interaction of the two factors.

Figure 2:

Effects of rimcazole on various aspects of performances with changes in FR schedule value at different reinforcing amounts of food (20 or 60 mg/presentation). Error bars indicate standard errors of the mean. Top row: effects of FR value on response rates (responses/sec) either after injection of vehicle or different doses of rimcazole. Abscissae: FR value; Ordinates: Responses/sec. Second row: demand curves of intake (consumption) as a function of FR value (price). Abscissae: log FR value; Ordinates: log intake (mg/kg). Symbols depict actual values; lines are fits to the points using Equation [1]. Third row: Effects of rimcazole dose on the Q₀ parameter of the fitted demand equation expressed as percentage change from vehicle treatment. Fourth row: Effects of rimcazole dose on the EV parameter derived from α of the fitted demand equation as derived from Equation [2], expressed as percentage change from vehicle treatment.

The food demand curves fitted to the data for 20 mg/delivery of food pellets showed monotonic decreases in food intake with increasing FR values and gave Q₀ and EV values of 89.3 (±11.2) and 12.9 (±1.09), respectively (Figure 2B; Table 1). Similar demand curves were obtained at each dose of rimcazole, and the model accounted for ≥96% of the variance (Table 1). Rimcazole did not significantly alter fitted Q₀ (Figure 2C) or EV values (Figure 2D) compared to control.

As with the smaller food amount, increasing FR value progressively decreased response rates maintained by 60 mg/delivery at FR values greater than 10, both without and with rimcazole pretreatment (Figure 2E, compare open and filled symbols). Response rates at FR 10 of 1.24 (±0.33) responses/sec progressively decreased to 0.094 (±0.047) at FR 80 (Figure 2E). Rimcazole had little effect on response rates across the range of FR values. Statistical analysis indicated that there was a significant effect of FR value (F_4,24=26.1, p<0.001), no effect of rimcazole dose, or the interaction of dose and FR value.

The demand curves fitted to the data for 60 mg/delivery of food pellets showed monotonic decreases in food intake with increases in FR value and gave Q₀ and EV values of 81.6 (±9.31) and 10.9 (±1.53), respectively (Figure 2F; Table 1). Similar demand curves were obtained after each dose of rimcazole, and the model accounted for ≥98% of the variance. Rimcazole did not significantly alter fitted Q₀ values (Figure 2G) or EV values (Figure 2H) compared to control.

Effects of Combinations of Methylphenidate and BD1008 on Cocaine Self-Administration

As obtained previously, increasing FR values at 0.1 mg/kg/injection of cocaine for comparison with pretreatments of methylphenidate and BD1008 combinations produced a bitonic change in response rates with absolute values similar to those obtained previously (Figure 3A, filled symbols). Response rates were little affected by the combinations of methylphenidate and BD1008 across the range of FR values assessed (Figure 3A, open symbols). Statistical analysis indicated that there was a significant effect of FR value (F_4,20=3.55, p=0.024), but no effects of the methylphenidate/BD1008 treatments, or the interaction of treatment and FR value.

Figure 3:

Effects of combinations of methylphenidate and BD1008 on various aspects of performances with changes in FR schedule value at different reinforcing doses of cocaine (0.1 or 0.32 mg/kg per injection). All details of the graphs are as in Figure 1.

The demand curves for cocaine (0.1 mg/kg/injection) again showed monotonic decreases in cocaine intake with increases in FR value (Figure 3B; filled symbols). The values of Q₀ and EV derived from the fits were 47.0 (±2.56) and 10.7 (±1.69), respectively, and the model accounted for ≥99% of the variance (Table 1). Similar demand curves were obtained with each combination of treatments (Figure 3B), with decreases in fitted Q₀ values compared to vehicle control (Figure 3C). These effects were statistically significant (F_3,15=4.73, p=0.016), with post-hoc comparisons indicating that Q₀ was significantly decreased at the highest dose of BD1008 in combination with methylphenidate. In contrast, the calculated EV values were not changed compared to control (Figure 3D).

As shown above, response rates maintained at 0.32 mg/kg/injection of cocaine were uniformly lower than those obtained at the lower dose, and increasing FR value progressively increased response rates. The combination of methylphenidate and BD1008 did not alter average response rates maintained by this higher cocaine dose across the range of FR values assessed (Figure 3E, open symbols). Statistical analysis indicated that there were significant effects of FR value (F_4,28=2.91, p=0.039), but not of the combination treatments, or the interaction of the two factors.

The demand curves for cocaine at 0.32 mg/kg/injection showed decreased cocaine intake with increases in FR value, though the decreases were less than those obtained at the lower cocaine dose (Figure 3F). Values of Q₀ and EV derived from the fits were 13.9 (±1.67) and 6.55 (±1.58), respectively with >95% of the variance accounted for by the exponential model (Table 1). Similar demand curves were obtained after each combination of methylphenidate and increasing doses of BD1008, with no effect on the fitted Q₀ (Figure 3G) or EV (Figure 3H) values.

Effects of Combinations of Methylphenidate and BD1008 on Food-Reinforced Responding

Increases in FR value from 5 to 10 increased response rates maintained by 20 mg/delivery of food pellets, with progressive decreases in response rates at higher FR values (Figure 4A, filled symbols). The combination of methylphenidate and BD1008 did not alter average response rates maintained by 20 mg/delivery of food pellets across the range of FR values (Figure 4A, open symbols). Statistical analysis indicated that there was a significant effect on response rates of FR value (F_4,20=26.0, p<0.001), no effect of the combination treatments, but an interaction of the two factors (F_12,60=2.56, p=0.008).

Figure 4:

Effects of combinations of methylphenidate and BD1008 on various aspects of performances with changes in FR schedule value at different reinforcing amounts of food (20 or 60 mg/presentation). All details of the graphs are as in Figure 2.

The food demand curves fitted to the data for 20 mg/delivery of food pellets showed decreased intake with increases in FR value and gave Q₀ and EV values of 102 (±15.5) and 12.7 (±0.488), respectively (Figure 4B; Table 1). Similar demand curves were obtained at each dose of rimcazole, with the model accounting for >94% of the variance. Rimcazole did not alter fitted Q₀ (Figure 4C) or EV values (Figure 4D) compared to control.

Increases in FR value decreased response rates maintained by 60 mg/delivery of food pellets at values greater than FR 10 with little effect of the combination of methylphenidate and BD1008 (Figure 4E, compare open and filled symbols). Statistical analysis indicated that there was a significant effect of FR value on response rates (F_4,24=38.5, p<0.001), but no significant effects of the treatment combinations or the interaction of the two factors.

The demand curves fitted to the response-rate data for 60 mg/delivery of food pellets showed the characteristic monotonic decreases in intake with increases in FR value (Figure 4F), and yielded Q₀ and EV values of 92.9 (±12.2) and 14.1 (±1.33), respectively (Table 1). Similar demand curves were obtained at each drug combination, with fits to the model accounting for >99% of the variance. Combinations of methylphenidate and BD1008 had no effect on fitted Q₀ values (Figure 4G). A significant effect on EV values (Figure 4H; F_3,18=3.25, p=0.046) was not due to differences between dose effects and control values but instead reflected differences between effects of the different doses.

Effects of Combinations of Methylphenidate and BD1063 on Cocaine Self-Administration

Increasing FR values at 0.1 mg/kg/injection of cocaine again produced a bitonic change in response rates with absolute values similar to those obtained previously, with little effect of the combinations of methylphenidate and BD1063 across the range of FR values assessed (Figure 5A, compare open and filled symbols). However, at two dose combinations there were trends towards increased response rates at the highest FR values (Figure 5A, open symbols). Statistical analysis indicated that there was a significant effect on response rates of FR value (F_4,20=3.80, p=0.019), no effect of the drug treatments nor an interaction of the two factors.

Figure 5:

Effects of combinations of methylphenidate and BD1063 on various aspects of performances with changes in FR schedule value at different reinforcing doses of cocaine (0.1 or 0.32 mg/kg per injection). All details of the graphs are as in Figure 1.

The demand curves for cocaine (0.1 mg/kg/injection) showed monotonic decreases in cocaine intake with increases in FR value (Figure 5B; filled symbols). Values of Q₀ and EV derived from the fits were 47.2 (±5.56) and 8.32 (±1.37), respectively (Table 1). Similar demand curves were obtained with each combination of methylphenidate and BD1063 (Figure 5B; open symbols), with greater than 99% of the variance accounted for by the model. The drug combination decreased fitted Q₀ values compared to vehicle control (Figure 5C; F_3,15=3.35, p=0.047). The calculated EV was not changed compared to control for the 0.1 mg/kg/injection dose of cocaine (Figure 5D).

Increasing FR value progressively increased response rates maintained by 0.32 mg/kg/injection of cocaine, with little effect of the combination of methylphenidate and BD1063 (Figure 5E, compare open and filled symbols). Statistical analysis indicated that there were effects on response rates of FR value (F_4,28=3.24, p=0.026), but not of the drug combination (F_3,21=2.61, p=0.078) or the interaction of the two factors (F_12,84=1.26, p=0.259).

The demand curves for cocaine at the 0.32 mg/kg/injection dose also showed decreased cocaine intake with increases in FR value, though the decreases were less than those obtained at the lower cocaine dose (Figure 5F). Values of Q₀ and EV derived from the fits were 14.9 (±1.03) and 11.5 (±5.29), respectively (Table 1). Similar demand curves were obtained after each combination of methylphenidate and increasing doses of BD1063, with the model accounting for ≥84% of the variance. There were no effects of the drug treatments on the fitted Q₀ (Figure 5G) nor EV (Figure 5H) values.

Effects of Combinations of Methylphenidate and BD1063 on Food-Reinforced Responding

Increases in FR value again increased response rates maintained by 20 mg/delivery of food pellets from FR 5 to FR 10, with progressive decreases in response rates at higher FR values, with no effects of the combination of methylphenidate and BD1063 (Figure 6A, compare open and filled symbols). Statistical analysis indicated that there were significant effects of FR value on response rates (F_4,20=32.5, p<0.001), but no significant effects of the drug treatments or their interaction with FR value.

Figure 6:

Effects of combinations of methylphenidate and BD1063 on various aspects of performances with changes in FR schedule value at different reinforcing amounts of food (20 or 60 mg/presentation). All details of the graphs are as in Figure 2.

The food demand curves fitted to the data for 20 mg/delivery of food pellets showed the characteristic monotonic decreases in intake with increases in FR value and gave Q₀ and EV values of 103 (±15.5) and 12.7 (±0.488), respectively (Figure 6B; Table 1). Similar demand curves were obtained at each drug treatment, with the model accounting for ≥94% of the variance. Combinations of methylphenidate and BD1063 affected neither fitted Q₀ (Figure 6C) nor EV (Figure 6D) values compared to control (Table 1).

Increases in FR value decreased response rates maintained by 60 mg/delivery of food pellets at values greater than FR 10, with little effect of the combination of methylphenidate and BD1063 (Figure 6E, compare open and filled symbols). Statistical analysis indicated that there was a significant effect on response rates of FR value (F_4,24=48.4, p<0.001), no significant effect of the combination treatments or the interaction of the two factors.

The demand curves fitted to the response-rate data for 60 mg/delivery of food pellets showed the characteristic decreases in consumption with increases in price (Figure 6F), and yielded Q₀ and EV values of 92.9 (±12.2) and 14.1 (±1.33), respectively (Table 1). Similar demand curves were obtained at each drug combination, with the model accounting for >99% of the variance. The combinations of methylphenidate and BD1063 altered neither fitted Q₀ (Figure 6G) nor EV (Figure 6H) values compared to control.

DISCUSSION

In the present study, the effects of rimcazole on the self-administration of cocaine were compared with its effects on responding comparably maintained by food reinforcement. A previous study had shown that rimcazole decreased cocaine self-administration at doses that did not appreciably affect food-reinforced responding (Hiranita et al. 2011), suggesting a specific effect on the reinforcing effects of cocaine. As response rates maintained by drug injections can be influenced by many factors and therefore do not by necessity reflect changes in the reinforcing effectiveness of the self-administered drug, the present study looked to an indication of reinforcing strength suggested by adapting behavioral economic measures to operant behavior (Hursh and Silberberg, 2008). The demand curves obtained indicated that rimcazole at appropriate doses decreased the reinforcing effectiveness of cocaine, but not food reinforcement. A similar result was reported previously for the effects of analogs of benztropine on cocaine self administration (Zanettini et al. 2018).

Similar to previous economic comparisons of behaviors maintained by food or cocaine (Kearns and Silberberg 2016; Zanettini et al. 2018), the derived Q₀ and EV values for food reinforcement were greater than those for cocaine self-administration. Further, response rates maintained by food reinforcement were greater than those maintained by cocaine injection. Previous studies have identified several factors which can affect the parameters of demand curves, such as the level of food deprivation (Hursh and Roma 2013; Hursh and Silberberg 2008; Macenski and Meisch 1999), history and experience with the reinforcer (Christensen et al. 2008a; Christensen et al. 2008b; Oleson and Roberts 2009; Panlilio et al. 2013), expression of DA-receptor subtypes (Soto et al. 2011), as well as environmental rearing conditions (Hofford et al. 2016; Yates et al. 2019). As many of these factors can be difficult to equate across types of reinforcers, the present study focused on relative change in the exponential demand curve parameter values as a function of drug treatments.

Rimcazole decreased both EV and Q₀ parameters fitted to the exponential demand function of cocaine. The value Q₀ is influenced by the combined effects of parameters of the reinforcer that contribute to its magnitude, such as duration, size/potency, and concentration (Hursh and Winger 1995), though at least for pharmacological reinforcers duration may be of minimal influence (Ko et al. 2002). The effects of rimcazole on the Q₀ parameter under the present cocaine self-administration procedure are similar to the decreases in number of injections obtained when the concentration of the drug is increased, and Q₀ values decreased accordingly in the present study with increases in cocaine dose or the amount of food per delivery. This outcome is not inconsistent with reports of reinforcer-selective downward shifts in the cocaine dose-effect curve produced by rimcazole as well as rimcazole analogs (Hiranita et al. 2011).

Variables shown to affect EV include onset of self-administered drug action (Winger et al. 2002), type of reinforcer (Hursh and Winger 1995), and whether there is a source of the reinforcer outside of the experimental session, that is whether the economy is open or closed (Hursh 1991; Hursh and Silberberg 2008). The present effects on EV produced by rimcazole resemble those produced with changing to an open economy by providing supplemental food when examining food-reinforced behaviors (Hursh 1991; Hursh and Silberberg 2008) or by administration of morphine in subjects self-administering the short acting mu-opioid agonist remifentanil (Wade-Galuska et al. 2011). In a previous study (Hiranita et al. 2011), the qualitative effects of rimcazole administration on responding maintained by cocaine were similar to those of providing supplemental food to subjects whose responding was maintained by food reinforcement. It was suggested that those findings with rimcazole resemble a “satiating” effect of the treatment. The present results on selective effects of rimcazole on the parameters of cocaine self-administration demand curves are consistent with that suggestion (Zanettini et al. 2018).

It is of interest that the effects of rimcazole on EV and Q₀ in subjects self-administering cocaine in the present experiment did not extend to food reinforced responding. Decreases in the values of these parameters with cocaine self-administration relative to food-reinforced responding imply a specific reduction in the reinforcing effectiveness of cocaine. However, additional studies should explore that specificity further by examining behaviors maintained by other reinforcers in addition to cocaine and food. Currently, it remains possible that the differential effects of the pretreatments obtained on cocaine self-administration and food-reinforced responding represent a specific insensitivity to treatments of the latter rather than a specific sensitivity of the former.

Because rimcazole has affinity for the DAT as well as σ receptors (Izenwasser et al. 1993; Valchar and Hanbauer 1993), the present study, following Hiranita et al. (2011), compared the effects of rimcazole with those of combinations of DAT inhibitors and σ antagonists. In that previous study, rimcazole and its analogs produced selective decreases in response rates similar to those produced by the drug combinations. Rimcazole decreased Q₀ and EV at both cocaine doses but at neither food amount. However, the drug combinations had less pronounced effects. Both combinations decreased Q₀, but had no significant effects on EV, and only had those effects at the lower of the two cocaine doses/injection. Nonetheless, neither drug combination had effects on either Q₀ or EV derived from responding maintained by food reinforcement.

Hiranita et al. (2011) showed that combinations of 0.1 mg/kg of WIN 35,428 with several σ-receptor antagonists had effects like those of rimcazole, though a higher dose of WIN 35,428 with those same antagonists did not. Possibly elements of the present procedure, in particular the changing values of the FR schedule across components, necessary for the determination of the demand curves but imposed incidentally with drug treatment, contributed to a diminished sensitivity to the effects of the combinations that may have been captured at alternate dose combinations. It is also possible that other (off target) effects of the combinations may be evident with the present baseline that were less influential with the baseline used by Hiranita et al. (2011).

Whatever factor(s) contributed to the diminished sensitivity to the combinations of drugs, a previous study (Zanettini et al. 2018) found effects similar to those of rimcazole produced by benztropine analogs which also have both DAT and σ-receptor affinity (Hiranita et al. 2017). The implication of our findings is that under certain conditions dual blockade of the σ-receptors and DAT decreases the reinforcing effectiveness of cocaine specifically. The ideal conditions under which the reinforcing effectiveness of cocaine is blunted with price manipulation may depend on a more finely tuned selection of combined σ-receptor and DAT blockade than that presently achieved with drug combinations.

Rimcazole was originally developed as an antipsychotic medication but clinical trials were terminated due to limited efficacy and an incident of grand mal seizure in one of sixteen patients (Gilmore et al. 2004). As a consequence, rimcazole is not likely to find its way into clinical use. That stated, the effectiveness and specificity demonstrated by rimcazole in this and the previous study, the behavioral economic results of Zanettini et al. (2018) with compounds having similar effects, together with molecular modulation of stimulant effects by compounds with σ-receptor affinity (e.g. Hong et al. 2017; Sambo et al. 2017) suggest combined DAT and σ-receptor targeting for the discovery of drugs that may be useful in the clinical management of stimulant abuse.