QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.103.052795v1 308/1/249    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gasior, M. Articles by Bergman, J. Search for Related Content PubMed PubMed Citation Articles by Gasior, M. Articles by Bergman, J.

BEHAVIORAL PHARMACOLOGY

Modification by Dopaminergic Drugs of Choice Behavior under Concurrent Schedules of Intravenous Saline and Food Delivery in Monkeys

Behavioral Pharmacology Laboratory/Alcohol and Drug Abuse Research Center, McLean Hospital, Belmont, Massachusetts

Received for publication April 8, 2003
Accepted September 25, 2003.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
The allocation of "choice" behavior provides a measure that may be useful in developing experimental models of clinical relapse. In the present experiments, indirect monoaminergic agonists [cocaine, 1-(2-[bis(4-fluorophenyl)methoxy]ethyl)-4-(3-phenylpropyl)piperazine (GBR 12909), desipramine, and citalopram], and dopaminergic D₁ family agonists [(±)-6-chloro-7,8-dihydroxy-3-allyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine (SKF 82958), R-(+)-6-bromo-7,8-dihydroxy-3-allyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine (R-(+)-6-BrAPB), and 6-chloro-7,8-dihydroxy-3-methyl-1-(3-methylphenyl)-2,3,4,5-tetrahydro-1H-3-benzazepine (SKF 83959)] and D₂ family agonists [quinelorane, R-(-)-10,11-dihydroxy-N-n-propylnorapomorphine (R-NPA), (+)-N-propyl-hydroxynaphoxazine [(+)-PHNO], and S-(+)-(4aR,10bR)-3,4,4a,10b-tetrahydro-4-propyl-2H,5H-[1]benzopyrano-[4,3-b]-1,4-oxazin-9-ol (PD 128907)] were evaluated for their capacity to alter the distribution of choice behavior in cocaine-experienced monkeys. Rhesus monkeys responded on two levers (injection-lever and food-lever) under concurrent fixed ratio 30; fixed ratio 30 schedules of i.v. cocaine and food delivery. Under training conditions, the distribution of behavior was related to the unit dose of i.v. cocaine: when saline was available, responding occurred predominantly on the food-lever and when reinforcing doses of cocaine were available, responding occurred predominantly on the injection-lever. Drugs were studied by administering i.m. pretreatment doses before components in sessions of i.v. saline availability. Cocaine produced dose-related increases in injection-lever responding in all monkeys, whereas desipramine failed to alter the distribution of behavior in any monkey. The dopamine transport blocker GBR 12909 and each dopamine D₁ family agonist markedly increased injection-lever responding in three of four monkeys; the serotonin transport blocker citalopram and D₂ family agonists were comparably effective in only one or two monkeys. These results agree with previous findings of similarity in the behavioral effects of cocaine and indirect or direct dopamine agonists and suggest, furthermore, that i.v. self-administration behavior engendered by priming doses of cocaine may involve actions mediated through both D₁ and D₂ families of dopamine receptors.

Caine et al., 2000). However, the reinforcing strength of the different consequences is not often measured and normalized, making rigorous comparisons impossible (Mello and Negus, 1996).

Self-administration procedures also are used to identify candidate medications by studying the effects of drugs on self-administration behavior that has undergone extinction. Although addictive behavior among drug users rarely undergoes extinction, such "reinstatement" of self-administration behavior has been proposed as a laboratory model of relapse and a means with which to examine candidate medications for their ability to promote relapse or, alternatively, retard its induction (Shalev et al., 2002). For example, indirect dopamine agonists such as cocaine and direct dopamine D₂ family agonists such as quinpirole or R-NPA may reinstate behavior in rats and monkeys to levels observed during sessions of i.v. cocaine self-administration itself. By contrast, direct dopamine D₁ family or D₃-selective agonists seem not to reinstate behavior but only to decrease the "reinstating" effects of cocaine itself (Self et al., 1996; Khroyan et al., 2000). The extent to which these results reflect treatment-induced changes in drug-seeking behavior or other effects that are unrelated to motivational variables, e.g., rate-altering or discriminative-stimulus effects, is uncertain (Bergman and Katz, 1998). Moreover, the predictive validity of reinstatement procedures for identifying medications that decrease vulnerability to clinical relapse is currently undocumented. This is an important issue because drugs that reinstate self-administration behavior may have effects comparable with those of the abused drug and thereby increase vulnerability to relapse. Alternatively, such pharmacotherapies may serve as highly effective medications, e.g., methadone in the management of heroin addiction. Notwithstanding, findings in reinstatement studies have been used to support the view that dopamine D₁ and D₃ agonists deserve continued consideration as candidate therapies for the management of stimulant-type addiction (Self et al., 1996; Khroyan et al., 2000).

The present studies were conducted to extend the application of i.v. self-administration procedures in the evaluation of candidate medications for stimulant-type addiction to the use of a "choice" procedure using concurrent schedules of i.v. cocaine and food delivery. Under these conditions, the two consequences are concurrently available under identical FR schedules, and the relative reinforcing effects of injections of different doses of i.v. cocaine and food delivery can be empirically described by the distribution of behavior on response levers associated with the consequence (Paronis et al., 2002). The use of such choice procedures to study the relative reinforcing strength of drugs was pioneered early in the development of self-administration research (Iglauer and Woods, 1974; Johanson and Schuster, 1975). Historically, however, they have not been widely applied in the study of candidate medications in laboratory animals (Woolverton and Balster, 1981). In contrast, choice conditions are regularly used in drug self-administration studies in human subjects so as to analyze the effects of drugs in terms of reinforcing strength relative to other known reinforcers and to evaluate candidate medications for drug addiction (Griffiths et al., 1980; Haney et al., 2001; Greenwald et al., 2002).

The direct aim of the present experiments was to determine the effects of differing types of dopaminergic compounds on choice behavior when subjects could lever-press for either i.v. saline injections or food delivery. Results indicate that, like cocaine, the dopamine transport blocker GBR 12909 and direct D₁ family agonists [SKF 82958, R-(+)-6-BrAPB, and SKF 83959] generally produced dose-related increases in the allocation of behavior to the injection-lever. D₂ family agonists [quinelorane, PD 128,907, (+)-PHNO, and R-NPA] also increased responding on the injection-lever; however, drug-induced increases in injection-lever responding plateaued and, on average, comparable levels of behavior were allocated to both levers over a log unit range of doses. These data are consistent with the view that activation of either dopamine D₁- or D₂-receptors can increase drug-seeking behavior in cocaine-experienced subjects.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Subjects. Three female (Mm 211, Mm 258, and Mm 331) and one male (Mm 263) adult rhesus monkeys (Macaca mulatta), weighing 6 to 9.5 kg, were studied in daily experimental sessions 5 to 6 days/week. Two monkeys (Mm 211 and Mm 331) were previously exposed to cocaine in i.v. self-administration experiments, and two monkeys (Mm 258 and Mm 263) had previously participated in experiments with a variety of opioid or dopaminergic drugs. Subjects were individually housed in stainless steel cages between experimental sessions in a temperature- and humidity-controlled room with a 12-h light/dark cycle (7:00 AM lights on). Each monkey had unlimited access to tap water and received a nutritionally balanced diet (5045 High Protein Monkey Diet; PMI Nutrition International, Inc., Brentwood, MO) supplemented regularly with fresh fruit, trail mix, and vitamins. The protocol for animal use in these studies was approved by the Institutional Animal Care and Use Committee. Subjects in this study were maintained in accordance with guidelines provided by the Committee on Care and Use of Laboratory Animals of the Institute of Laboratory Animals Resources, National Institutes of Health. This facility is licensed by the U.S. Department of Agriculture.

Each monkey initially was trained to lever press under a fixed ratio schedule of food presentation. After food-maintained performance was well established, each monkey was prepared with a chronic venous catheter after the general surgical procedure described by Herd et al. (1969). Briefly, under isoflurane anesthesia and in aseptic conditions, one end of a hydrophilically coated polyurethane catheter (inside diameter, 1.0 mm; outside diameter, 1.7 mm) was inserted through a femoral vein and passed to the level of the right atrium. The distal end of the catheter was attached to a titanium vascular access port (model TI200 AC-5H; Access Technologies, Skokie, IL) that was subcutaneously located in the midscapular region.

Apparatus. During experimental sessions, each monkey was seated in a Plexiglas chair within a ventilated, sound-attenuating chamber. Two response levers were mounted on a panel attached to the front of the chair. Each press of either lever with a force of at least 0.25 N produced an audible click of a relay and was recorded as a response. Colored lamps (red or white) behind the levers could be illuminated to serve as visual stimuli. A motor-driven feeder outside the chamber could deliver food pellets to a food receptacle on the chair via flexible tubing. A motor-driven syringe pump outside the chamber could deliver intravenous injections via an external catheter line and noncoring Huber needle through the silicone rubber septum of the monkey's vascular access port. Sterile 0.9% saline (1-2 ml) was used to flush residual drug solution from the port and catheter after experimental sessions. Both the syringe and feeder were operated by automatic programming equipment. Each operation of the syringe pump or feeder lasted 200 ms and delivered a constant volume from the syringe or one food pellet, respectively.

Terminal Schedule and Training Sessions. Each experimental session lasted 110 min and consisted of three 30-min components of schedule-controlled performance separated by 10-min time-out (TO) periods. The terminal schedule comprised a two-member sequence (for details, see Paronis et al., 2002). Briefly, 30 responses (FR30) on the injection-lever in the first member of the sequence produced an i.v. injection of the solution available for self-administration and, subsequently, a 45-s TO period. Each response on the other lever before completion of the schedule requirement produced a 10-s TO period without resetting the response requirement on the injection-lever. During the second member of the sequence, which lasted for the remainder of the 30-min component, responding was reinforced under concurrent FR30 schedules of i.v. injection and food pellet delivery. Specifically, the completion of 30 consecutive responses on the injection-lever was followed by an i.v. injection of the solution available for self-administration, whereas completion of thirty consecutive responses on the second lever (food-lever) was followed by the delivery of a food pellet. A response on one lever reset the response requirement on the other lever. Delivery of each reinforcer (food or i.v. injection) was followed by a 45-s TO period. The left lever was the injection-lever for monkeys Mm 211, Mm 258, and Mm 331 and the food-lever for monkey Mm 263; lever assignments remained constant throughout the present study. Stimulus lights of the same color were illuminated above both levers during both members of the sequence but were not lit during short and long TO periods. All lights were off during TO periods; TO responses were recorded but had no scheduled consequences.

Under training conditions, either saline or cocaine (0.032 or 0.1 mg/kg/inj) was available for i.v. self-administration. Performance for all monkeys was characterized by nearly exclusive responding on the food-lever when saline was available for self-administration, and nearly exclusive responding on the injection-lever when training doses of cocaine were available for self-administration. The availability of saline, 0.032, or 0.1 mg/kg/inj cocaine for self-administration in a particular component within each training session varied with the provisos that 1) saline was never available after cocaine, and 2) 0.032 mg/kg/inj cocaine was never available after 0.1 mg/kg/inj cocaine. The order of presentation of saline, 0.032 and 0.1 mg/kg/inj cocaine varied in a nonsystematic manner from session to session, with the specification that each condition was presented three times out of every 30 training sessions (Paronis et al., 2002). After performance across the different training conditions was stable from day to day, the effects of cocaine and other drugs were studied during test sessions described below.

Experimental Protocol. Test sessions usually were conducted twice a week; training sessions were conducted on other days. All programmed parameters of test and training sessions were identical. The effects of presession i.m. administration of cocaine and other drugs were studied during sessions in which saline was available for self-administration in all three components. In test sessions, the effects of drugs usually were studied by administering cumulative doses 10 min before successive components of the session. Full dose-effect functions for a drug were determined by studying overlapping ranges of cumulative doses up to the highest doses that could be studied safely and/or up to the doses that profoundly altered rates and/or patterns of operant responding in the second member of the terminal schedule. None of the drugs had consistent or dose-dependent effects on performance during the first member of the terminal schedule (data not shown). Drugs also were occasionally studied by administering single doses before the test session. In these latter experiments, data from the first component were used for inclusion with data from cumulative dosing experiments.

Drugs. Cocaine HCl, 1-(2-[bis(4-fluorophenyl)methoxy]ethyl)-4-(3-phenylpropyl)piperazine dihydrochloride (GBR 12909 2HCl), and (±)-6-chloro-7,8-dihydroxy-3-allyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine hydrobromide (SKF 82958 HBr) were obtained from the National Institute on Drug Abuse (Rockville, MD). 6-Chloro-7,8-dihydroxy-3-methyl-1-(3-methylphenyl)-2,3,4,5-tetrahydro-1H-3-benzazepine hydrobromide (SKF 83959 HBr) was supplied by Sigma/RBI (Natick, MA) through the Chemical Synthesis Program of the National Institute of Mental Health, Contract N01MH30003. Desipramine HCl, R-(+)-6-bromo-7,8-dihydroxy-3-allyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine hydrobromide [R-(+)-6-BrAPB HBr], R-(-)-10,11-dihydroxy-N-n-propylnorapomorphine hydrochloride (R-NPA HCl), quinelorane HCl, and S-(+)-(4aR,10bR)-3,4,4a,10b-tetrahydro-4-propyl-2H,5H-[1]benzopyrano-[4,3-b]-1,4-oxazin-9-ol hydrochloride (PD 128,907 HCl) were purchased from Sigma/RBI. Citalopram HBr and (+)-N-propyl-hydroxynaphoxazine [(+)-PHNO] were graciously supplied by Lundbeck A/S (Copenhagen, Denmark) and Merck Sharpe and Dohme (Hoddesdon, UK), respectively. Except for the benzazepines, all drugs were prepared in 0.9% saline and warmed and/or sonicated before administration, as necessary. Stock solutions of the benzazepines were prepared in a solution of 0.1% ascorbic acid in sterile water and further diluted with 0.9% saline. Doses of cocaine refer to its free base form; doses of the remaining drugs refer to their salt forms. Cocaine was studied by making unit doses available for i.v. self-administration during training sessions or, during test sessions, by administering cumulative i.m. doses before components of i.v. saline availability. All other drugs were studied only by administering doses i.m. before components of saline availability.

Data Analysis. Only data collected during the second member of the two-member schedule were used for further analysis. Specifically, four measures of behavior were obtained for all monkeys in each component of the session. These included number of self-administered i.v. injections, number of food pellet deliveries, overall distribution of responding as the percentage of injection-lever and food-lever responding, and overall response rate. The percentage of injection-lever or food-lever responding was calculated by dividing the number of responses emitted on the injection-lever or food-lever by the total number of responses emitted on both levers. All responses (i.e., those in both completed and uncompleted ratios in each component) were included in the analysis of the distribution of responding. Overall response rate is expressed as responses per second and was calculated by dividing the total number of responses on both levers by the total elapsed time, excluding responses and time during TO periods.

Data were collected and analyzed for each monkey separately. Control measures were determined over the course of 15 to 20 sessions and therefore are expressed as mean values with 95% confidence limits (95% CL) for the expression of variability in the individual monkey's data. After testing began, the effects of i.m. drug pretreatments on self-administration behavior during i.v. saline availability were compared with the effects of i.m. saline pretreatment during i.v. saline availability, hereafter referred to as control sessions. Data from at least three temporally proximal control sessions during testing each drug were averaged and used as control values against which to measure the effects of that drug. When the effects of a drug dose were determined more than once in a subject, data were averaged to obtain mean values (± S.D.) for the effects of that dose in that monkey. Dose-response data for the distribution of responding, number of food deliveries, and response rate after drug pretreatments were further analyzed by estimating the dose of a drug that could 1) increase the percentage of injection-lever responding to 50%, 2) decrease food intake by 50%, or 3) decrease overall response rate by 50%. ED₅₀ values first were determined in individual monkeys and were calculated by extrapolation of the linear portion of the dose-response function. In some instances, dose-response data were not amenable to ED₅₀ calculation, and no further analysis was performed. For each measure, ED₅₀ values that were obtained for a drug in at least three monkeys were averaged for expression as a mean ED₅₀ value with the 95% CL shown in parentheses.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Control Performance during Training Sessions. When saline was available for self-administration during training sessions, monkeys allocated their responding almost exclusively to the food-lever (Table 1). Control values averaged for the group of monkeys over the course of the experiments (mean ± S.E.M.) show that 93 ± 3 and 7 ± 3% of responding occurred on the food-lever and the injection-lever, respectively, resulting in the delivery of 26 ± 1 food pellets and 2 ± 1 injections of saline per 30-min session component. Throughout the present experiments, overall response rates remained high in all components during sessions of saline availability and ranged between 2.9 and 3.7 responses/s in monkey Mm 258 and 1.3 and 2.0 responses/s in the remaining three monkeys.

In contrast to saline availability, the availability of training doses of cocaine for i.v. self-administration led to responding predominantly on the injection-lever. Averaged for the group of monkeys, 84 ± 6% of responding occurred on the injection-lever during the availability of 0.032 mg/kg/inj cocaine, resulting in the intake of 10 ± 3 injections and 4 ± 1 food pellet deliveries in the 30-min session component. During availability of the higher training dose of cocaine, 0.1 mg/kg/inj, responding occurred nearly exclusively on the injection-lever (96 ± 1%), resulting in the delivery of an average of 5 ± 2 injections and 1 food pellet per 30-min session component. There also were marked differences in overall rates of responding associated with the availability of saline or cocaine. Averaged for the group of monkeys, overall response rates during saline availability remained relatively high (2.3 ± 0.6 responses/s), reflecting behavior primarily on the food-lever. On the other hand, averaged rates of responding decreased in a dose-related manner during the availability of 0.032 and 0.1 mg/kg/inj cocaine to 0.6 ± 0.2 and 0.1 ± 0.1 responses/s, respectively.

Pretreatment with Cocaine during Saline Availability. Acute pretreatment with i.m. cocaine produced dose-related changes in the distribution of behavior, in the number of food deliveries and i.v. saline injections, and in the overall rate of responding in all subjects (Fig. 1; Table 2). Changes in the distribution of responding were reflected in dose-dependent increases in the number of i.v. saline injections and decreases in food pellet deliveries (Fig. 1, middle and bottom). All monkeys responded predominantly on the injection-lever after the highest i.m. doses of cocaine. Thus, the average number of self-administered saline injections during the 30-min session component increased from a control value of 3 ± 1to10 ± 2 and 9 ± 3 injections, respectively, after pretreatment with 0.32 and 1.0 mg/kg cocaine. Correspondingly, the number of food pellet deliveries decreased from control values of 24 ± 2 to 5 ± 4 and 1 ± 1 pellets, respectively. Overall response rates also decreased in a dose-dependent manner after pretreatment with i.m. cocaine, consistent with the decrease in the number of overall reinforcers (Table 2). The potency with which i.m. cocaine altered behavior did not differ significantly across measures of response distribution, reinforcer density, and response rate. Thus, averaged ED₅₀ values ranged only from 0.17 ± 0.05 to 0.26 ± 0.09 mg/kg across the different behavioral measures (Table 3).

View larger version (17K): [in this window] [in a new window]   Fig. 1. Effects of cocaine on responding under the concurrent schedules of i.v. saline and food presentation. Top plot shows data in individual monkeys and the group mean (solid line) for the percentage of responses produced on the injection-lever (responses on the injection lever/total responses on the injection-lever and food-lever) per 30-min session component. Middle and bottom plots show group means ± S.E.M. for the number of self-administered saline injections and food pellets deliveries, respectively, per 30-min session component. Points above "SAL" represent the distribution of responses and number of reinforcers after saline pretreatment during at least three sessions of i.v. saline availability (control performance). These values are averaged across components of the session for clarity of presentation. Abscissae give cumulative doses of i.m. cocaine. Response rates and ED50 values for cocaine are given in Tables 2 and 3.

View larger version (20K): [in this window] [in a new window]   Fig. 2. Effects of selective dopamine (GBR 12909), serotonin (citalopram), and norepinephrine (desipramine) transport inhibitors on responding under the concurrent schedules of i.v. saline and food presentation. See Fig. 1 for legend and other details.

Pretreatment with Selective Monoamine Transport Inhibitors. Pretreatment with inhibitors of the transport of dopamine (GBR 12909), norepinephrine (desipramine), and serotonin (citalopram) had differing behavioral effects during i.v. saline availability (Fig. 2). In three of four monkeys, the dopamine transport inhibitor GBR 12909 produced dose-related decreases in overall response rates and food deliveries but increased responding on the injection-lever (Fig. 2, middle and bottom left; Table 3). In the fourth monkey, Mm 263, GBR 12909 also decreased response rates and food pellet delivery but, up to a dose that completely eliminated performance (10 mg/kg), did not shift the distribution of behavior to the injection-lever.

The potency of GBR 12909 differed <10-fold among the three monkeys for which responding increased on the injection-lever. Cumulative i.m. doses that resulted in the full allocation of responding to the injection-lever and, as well, in delivery of maximum number of saline injections, ranged from 3.2 to 18 mg/kg GBR 12909 (see individual plots in Fig. 2, top left). Averaged for these three monkeys, and irrespective of dose, the maximum number of injections was 15 ± 6 injections. For these monkeys, the potency of GBR 12909 was similar across the different behavioral measures of the present experiments. Thus, the i.m. pretreatment dose of GBR 12909 calculated to produce equal responding on the injection- and food-levers was 6.20 ± 3.0 mg/kg, and the estimated ED₅₀ dose for decreasing response rates was 7.3 ± 3.5 mg/kg (Table 3).

The serotonin transport inhibitor citalopram consistently produced dose-dependent decreases in overall response rates and food-maintained behavior and had varying effects on the distribution of responding across subjects. Averaged for the group of monkeys, i.m. pretreatment with the highest cumulative doses of citalopram, 3.2 or 10 mg/kg, led to similar levels of responding on both the injection-lever and the food-lever, and decreased response rates and food deliveries to approximately 25% of control values (Fig. 2; Table 3). However, the change in the distribution of behavior was primarily attributable to the effects of citalopram in one monkey, Mm 258. After the cumulative dose of 10 mg/kg citalopram, >80% of responding by this monkey occurred on the injection-lever, increasing the number of i.v. saline injections from a control value of two injections to 14 injections. Citalopram otherwise only moderately altered the distribution of behavior in the other three monkeys, with maxima of 18 to 45% of responding on the injection-lever after pretreatment with the highest cumulative doses (Fig. 2, top). These limited changes in the distribution of behavior did not result in a significant increase in the number of saline injections in any subject. Regardless of changes in the distribution of responding, citalopram exerted its behavioral effect with comparable potency across measures. Thus, ED₅₀ values averaged across monkeys for the effects of citalopram on response rate and food delivery were nearly identical and, for the two monkeys in which comparisons could be made, individual ED₅₀ values were highly comparable across the measures of response distribution, food deliveries, and response rate (Table 3).

In contrast to GBR 12909 and citalopram, the norepinephrine transport inhibitor desipramine did not markedly change the distribution of responding during saline availability in any monkey (Fig. 2, top). Responding was allocated exclusively to the food-lever after cumulative i.m. doses of desipramine up to 3.2 mg/kg in all monkeys and up to 10 mg/kg in three of four monkeys. The fourth monkey, Mm 258, also responded primarily (>70%) on the food-lever after the cumulative i.m. dose of 10 mg/kg desipramine. In this monkey, however, some responding also occurred on the injection-lever (<30%), resulting in eight injections of saline. In addition to differences in effects on the distribution of behavior, desipramine also could be distinguished from GBR 12909 and citalopram by its effects on response rates and food delivery. Thus, as shown in Fig. 2 (middle and bottom right) and Table 2, the highest dose of desipramine, 10 mg/kg, reduced response rates by >50% of control values in all subjects (ED₅₀ = 5.3 ± 1.1 mg/kg) but decreased the number of food deliveries by >50% in only one monkey (Mm 211). Pretreatment with i.m. doses of desipramine >10 mg/kg previously were noted to produce tremor and other adverse effects and therefore were not studied in the present experiments.

Pretreatment with Dopamine D₁ Family Agonists. Dopamine D₁ family agonists that differ in efficacy had comparable effects on behavior during i.v. saline availability (Fig. 3). Thus, one or more cumulative i.m. doses of both the high-efficacy agonists SKF 82958 and R-(+)-6-BrAPB and the partial agonist SKF 83959 markedly increased injection-lever responding in three monkeys; each drug was ineffective in one of four monkeys. Optimum doses of the D₁ agonists produced between 65 and 100% responding on the injection-lever and corresponding increases in the number of saline injections (Fig. 3, middle). Of note, changes in the allocation of behavior were observed in different sets of three monkeys for the different drugs (Fig. 3, top). Thus, in the different sets of three monkeys the D₁ agonists SKF 82958, R-(+)-6-BrAPB, and SKF 83959 increased the number of i.v. saline injections from control values of one to two injections to, respectively, 11 ± 5, 17 ± 7, and 14 ± 7 injections per 30-min session component. In all four monkeys, each of the three drugs also produced dose-related decreases in food-pellet deliveries and in overall response rates (Fig. 3, bottom; Table 2). Based on averaged ED₅₀ values, the potency of each D₁ agonist was similar across all behavioral measures (Table 4).

View larger version (20K): [in this window] [in a new window]   Fig. 3. Effects of dopamine D1 agonists [SKF 82958, R-(+)-6-BrAPB, and SKF 83959] on self-administration behavior under the concurrent schedule of i.v. self-administration and food delivery during sessions of i.v. saline availability. See Fig. 1 for legend and other details.

Pretreatment with Dopamine D₂ Family Agonists. Dopamine D₂ [quinelorane, R-NPA, and (+)-PHNO] or D₃/D₂ agonists (PD 128907) had qualitatively comparable effects on behavior in the present experiments. Averaged for the group of monkeys, each of the D₂ family agonists produced dose-related increases in injection-lever responding and the number of i.v. saline injections (Fig. 4). On average, the most effective doses of the drugs produced a similar distribution of behavior, with 42 to 59% of responding allocated to the injection-lever. Each of the D₂ family agonists also produced dose-related decreases in food pellet deliveries and response rate (Fig. 4, bottom; Table 2). The greatest effects were observed with (+)-PHNO and R-NPA, which, averaged across monkeys, reduced response rates and the number of food-pellet deliveries to <20% of control values. Pronounced effects also were observed with quinelorane and PD 128,907, which, on average, decreased response rates and the number of food-pellet deliveries to approximately 30 to 40% of control values. The highest doses of the D₂ family agonists either eliminated responding or produced agitation and self-scratching in individual monkeys; higher doses of drugs were not studied to avoid adverse behavioral effects. Notwithstanding some differences among subjects, averaged ED₅₀ values indicate that each of the D₂ family agonists was comparable in potency across behavioral measures (Table 4).

View larger version (20K): [in this window] [in a new window]   Fig. 4. Effects of dopamine D2 [quinelorane, R-NPA, and (+)-PHNO] and D3/D2 (PD 128,907) agonists on self-administration behavior under the concurrent schedule of i.v. self-administration and food delivery during sessions of i.v. saline availability. See Fig. 1 for legend and other details.

Although the different D₂ family agonists comparably altered the distribution of behavior, the magnitude of effects differed among monkeys, with the greatest effects observed in monkeys Mm 211 and Mm 258. For example, optimum doses of the four drugs produced 64 to 100% injection-lever responding in monkey Mm 211 and 78 to 100% injection-lever responding in monkey Mm 258 (Fig. 4, top). D₂ family agonists also produced the greatest increase in the number of saline injections in these two monkeys. PD 128,907 was most effective in this regard, with 17 and 30 injections of i.v. saline after optimum doses in monkeys Mm 211 and Mm 258, respectively.

Alterations in the distribution of responding were less pronounced in the remaining two monkeys, Mm 263 and Mm 331. Optimum doses of each of the four D₂ family agonists produced <50% injection-lever responding and five or fewer i.v. saline injections. Mm 331 allocated <30% of responses to the injection-lever after pretreatment with R-NPA and (+)-PHNO and responded almost exclusively on the food-lever after pretreatment with quinelorane or PD 128907. Injection-lever responding by Mm 263 increased to 33 and 48% after optimum doses of quinelorane and (+)-PHNO but was less than 5% after pretreatment with R-NPA or PD 128907.

Doses of each of the D₂ agonists that increased injection-lever responding also induced responding during periods of nonreinforcement (i.e., during post-reinforcement and intercomponent time-out periods) and, in some cases, occasioned switching between injection-lever and food-lever responding throughout the component. These effects were not quantified; however, it is noteworthy that such effects were not observed with other drugs in the present study.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
Recent self-administration studies using concurrent schedules of i.v. injections and food delivery indicate that the distribution of behavior (i.e., choice) may provide an index of relative reinforcing strength that is functionally independent of other performance-altering effects of self-administered cocaine (Paronis et al., 2002). The main goal of the present experiments was to use similar methods to study how priming doses of dopaminergic drugs modify choice behavior by comparing their effects on the distribution of injection- and food-lever responding during i.v. saline availability. The effects of priming doses on i.v. self-administration behavior in the absence of reinforcing drug injections previously have been assessed in reinstatement procedures in rats and monkeys. However, these methods differ in important ways. For example, reinstatement methods employ a single schedule of reinforcement, and responding is formally extinguished for test conditions. In the present choice procedure, behavior is maintained under concurrent schedules and is not extinguished but, reliably, shifts to the food-lever when response-contingent i.v. injections no longer are reinforcing. As discussed below, such procedural differences may yield dissimilar results with some dopaminergic drugs.

Monoamine Transport Inhibitors. Both cocaine and GBR 12909 induced dose-related changes in choice behavior during i.v. saline availability, resulting in nearly exclusive allocation of responding to the injection-lever after the highest doses of both drugs. These results support previous findings that cocaine and GBR 12909 can reinstate self-administration behavior (de Wit and Stewart, 1981; deVries et al., 1999; Schenk, 2002), and extend those observations to increases in injection-lever responding under concurrently available schedules of i.v. saline and food delivery. To the extent that the priming effects of GBR 12909 may be attributable to its selective inhibition of dopamine transport, the present results further support the view that its effects and comparable effects of cocaine involve prominent dopaminergic actions.

The effects of cocaine and GBR 12909 contrasted with those of the norepinephrine transport inhibitor desipramine, which did not generally alter the distribution of behavior during i.v. saline availability. The present data with desipramine are consistent with previous findings that norepinephrine transport inhibitors have limited cocaine-like reinforcing actions and, as well, that desipramine is not effective in modulating the reinforcing effects of cocaine (Woolverton, 1987; Mello et al., 1990; Foltin and Fischman, 1994). Drug-discrimination studies previously have provided some evidence for the involvement of noradrenergic actions in discriminative-stimulus effects of cocaine (Spealman, 1995). Nevertheless, the lack of desipramine-induced increases in injection-lever responding suggests that shared actions of cocaine and desipramine did not contribute greatly to the priming effects of cocaine in the present experiments.

Citalopram reliably decreased overall response rates and food intake, yet produced inconsistent changes in the distribution of behavior. Serotonin uptake inhibitors have been reported to be ineffective in either maintaining i.v. self-administration behavior or inducing reinstatement behavior in cocaine-trained subjects (Griffiths et al., 1978; Baker et al., 2001). Indeed, serotonergic drugs seem to reduce, rather than augment, the effects of cocaine in self-administration procedures (Kleven and Woolverton, 1993; Grottick et al., 2000; Czoty et al., 2002). The present results are consistent with such findings and do not support a prominent serotonergic role in the priming effects of cocaine on choice behavior.

Direct Dopamine Agonists. The behavioral effects of D₁ family and D₂ family receptor agonists overlapped those of cocaine and GBR 12909. On average, high levels of injection-lever responding followed treatment with each agonist. However, considerable variability was observed among individual monkeys, especially after treatment with D₂ family agonists. Although the basis for differences in the response of individual subjects to dopamine agonists is not well understood, such findings also have been reported in drug discrimination studies in monkeys. For example, D₁ and D₂ agonists have been shown to mimic the discriminative-stimulus effects of cocaine or methamphetamine in most, but not all, subjects (Spealman et al., 1991, Tidey and Bergman, 1998). Incomplete overlap in the stimulus effects of indirect and direct dopamine agonists supports the view that nondopaminergic mechanisms also may contribute to the behavioral effects of cocaine and related drugs.

Dopamine D₁ agonists, including the partial agonist SKF 83959, which produced considerable increases in injection-lever responding in the present studies, previously were reported to be ineffective in reinstating self-administration behavior in both rats and monkeys (Self et al., 1996, Khroyan et al., 2000). For example, priming doses of 0.32 or 1.0 mg/kg of the D₁ agonist R-(+)-6-BrAPB, which produced >75% responding on the injection-lever in three of four monkeys in the present experiments, failed to restore extinguished responding in monkeys trained under a second-order schedule of i.v. cocaine self-administration (Khroyan et al., 2000). The reasons for such dissimilar results in choice and reinstatement experiments are uncertain but may indicate that different features of cocaine's pharmacology contribute to injection-lever responding in the two procedures. For example, doses of cocaine that reinstate self-administration behavior under second-order schedules in monkeys (0.3 and 1.0 mg/kg; Khroyan et al., 2000) also increase response rates under simple and second-order fixed interval schedules of food presentation (Gonzalez and Goldberg, 1977; Spealman et al., 1989). However, D₁ agonists, although sharing discriminative-stimulus and reinforcing effects with cocaine, do not typically increase response rates (Bergman et al., 1995; Katz et al., 1995; Khroyan et al., 2000). In conjunction, these observations raise the possibility that reinstatement procedures in cocaine-trained subjects are less sensitive to drugs that do not increase responding than to drugs that, like cocaine, have rate-stimulant effects. From this perspective, the present choice procedures seem to be sensitive to D₁-mediated actions involved in priming effects of cocaine that are independent of its rate-stimulant actions.

The effects of D₂ agonists in the present experiments also incompletely overlapped those of cocaine. Several behavioral actions of D₂ agonists may have contributed to these results. Like cocaine, D₂ agonists exhibit both reinforcing and rate-increasing effects in monkeys that can contribute to increases in injection-lever responding (Woolverton et al., 1984; Bergman et al., 1995; Grech et al., 1996). D₂ agonists also may induce response switching in a manner that is independent of the probability of reinforcement (Evenden and Doggett, 1989). In the present studies, D₂ agonists were the only drugs that caused responding during periods of nonreinforcement (i.e., during postreinforcement and intercomponent time-out periods) and that occasioned switching back and forth between the injection- and food-levers both between and within fixed ratio 30 response units. Possibly, such effects served to limit the extent to which priming injections of D₂ agonists mimicked cocaine in some monkeys, resulting overall in intermediate levels of injection-lever responding. The occurrence of elevations in noncontingent responding and response switching with D₂ agonists, but not cocaine, suggests that the role of D₂ mechanisms in the priming effects of cocaine is modulated by other neurochemical actions.

Reinstatement studies suggest that the priming effects of D₂ family agonists also may depend on subtype selectivity. For example, the reportedly D₃-selective ligand PD 128,907 seems neither to reinstate self-administration behavior nor to augment cocaine-induced reinstatement in monkeys, whereas, like cocaine, D₂ family agonists, including quinelorane and R-NPA, fully restore extinguished self-administration behavior in both rats and monkeys (Self et al., 1996; Khroyan et al., 2000). In view of such previous data, it is noteworthy that the effects of PD 128,907 were indistinguishable from those of R-NPA and other D₂ family agonists in the present studies. The reasons for such dissimilar effects of D₂ family agonists in the two types of self-administration studies are unclear. One possibility is that squirrel monkeys (reinstatement studies) and rhesus monkeys (choice studies) are differently sensitive to the effects of D₂ agonists. For example, doses >0.1 mg/kg PD 128,907 are reported to produce adverse behavioral effects in squirrel monkeys and were not studied as priming doses in reinstatement studies, whereas doses as high as 0.32 mg/kg were studied in rhesus monkeys in the present experiments and did not decrease response rate below 1.0 responses/s. Another possibility is that procedures based on extinguished behavior (reinstatement) and ongoing behavior (choice) provide assay conditions that are dissimilarly sensitive to D₂ family receptor-mediated actions or to distinctions between D₂-receptor subtype. In this regard, the functional consequences of D₃-receptor activation are not well understood, and the role of distinctive D₃-mediated actions in PD 128,907's behavioral effects remains unclear (Spealman, 1996; Caine et al., 1997; but see Katz and Alling, 2000; Dekeyne et al., 2001; Rinken et al., 1999).

In summary, the experiments reported here were conducted to pharmacologically characterize the dopaminergic contribution to the priming effects of cocaine when i.v. injections of saline and a nondrug reinforcer, food, are concurrently available. In contrast to reinstatement procedures, the effects of priming injections of drugs were studied on ongoing, rather than extinguished, behavior and measured in terms of allocation, rather than rates, of behavior. Results indicate that dopamine D₁ family agonists, like cocaine and GBR 12909, produced marked increases in i.v. self-administration behavior during saline availability, whereas lesser effects on the distribution of behavior were observed with nondopaminergic transport inhibitors or with dopamine D₂ family agonists. The value of the present approach may lie in providing a new metric with which to experimentally study relapse. Moreover, the present findings may lead to an improved understanding of the role of dopaminergic mechanisms in relapse to cocaine addiction.

	Footnotes

 
This work was supported by National Institute of Health Grants DA10566, DA11453, and DA03774.

DOI: 10.1124/jpet.103.052795.

ABBREVIATIONS: FR, fixed ratio; TO, time-out; inj, injection; CL, confidence limits.

Address correspondence to: Dr. Jack Bergman, Behaviroal Pharmacology Laboratory/Alcohol and Drug Abuse Research Center, McLean Hospital, 115 Mill St., Belmont, MA 02478. E-mail: jbergman{at}hms.harvard.edu

	References

Top Abstract Materials and Methods Results Discussion References

 

Baker DA, Tran-Nguyen TL, Fuchs RA, and Neisewander JL (2001) Influence of individual differences and chronic fluoxetine treatment on cocaine-seeking behavior in rats. Psychopharmacology 155: 18-26.[CrossRef][Medline]
Bergman J and Katz JL (1998) Behavioral pharmacology of cocaine and the determinants of abuse liability, in Cocaine Abuse: Behavior, Pharmacology and Clinical Applications (Higgins ST and Katz JL eds) pp 51-79, Academic Press, San Diego.
Bergman J, Rosenzweig-Lipson S, and Spealman RD (1995) Differential effects of dopamine D1 and D2 receptor agonists on schedule-controlled behavior of squirrel monkeys. J Pharmacol Exp Ther 273: 40-48.[Abstract/Free Full Text]
Caine SB, Koob GF, Parsons LH, Everitt BJ, Schwartz JC, and Sokoloff P (1997) D3 receptor test in vitro predicts decreased cocaine self-administration in rats. Neuroreport 8: 2373-2377.[Medline]
Caine SB, Negus SS, and Mello NK (2000) Effects of dopamine D1-like and D2-like agonists on cocaine self-administration in rhesus monkeys: rapid assessment of cocaine dose-effect functions. Psychopharmacology 148: 41-51.[CrossRef][Medline]
Czoty PW, Ginsburg BC, and Howell LL (2002) Serotonergic attenuation of the reinforcing and neurochemical effects of cocaine in squirrel monkeys. J Pharmacol Exp Ther 300: 831-837.[Abstract/Free Full Text]
de Wit H and Stewart J (1981) Reinstatement of cocaine-reinforced responding in the rat. Psychopharmacology 75: 134-143.[CrossRef][Medline]
Dekeyne A, Rivet J-M, Gobert A, and Millan MJ (2001) Generalization of serotonin (5-HT)_1A agonists and the antipsychotics, clozapine, ziprasidone and S16924 [GenBank] , but not haloperidol, to the discriminative stimuli elicited by PD128, 907 and 7-OH-DPAT. Neuropharmacology 40: 899-910.[Medline]
deVries TJ, Schoffelmeer AN, Binnekade R, and Vanderschuren LJ (1999) Dopaminergic mechanisms mediating the incentive to seek cocaine and heroin following long-term withdrawal of IV drug self-administration. Psychopharmacology 143: 254-260.[CrossRef][Medline]
Evenden JL and Doggett SJ (1989) A comparison of the effects of amphetamine, apomorphine and white noise on response switching in the rat. Psychopharmacology 97: 238-242.[Medline]
Foltin RW and Fischman MW (1994) Cocaine self-administration research: treatment implications. NIDA Res Monogr 145: 139-162.[Medline]
Gonzalez FA and Goldberg SR (1977) Effects of cocaine and d-amphetamine on behavior maintained under various schedules of food presentation in squirrel monkeys. J Pharmacol Exp Ther 201: 33-43.[Abstract/Free Full Text]
Grech DM, Spealman RD, and Bergman J (1996) Self-administration of D1 receptor agonists by squirrel monkeys. Psychopharmacology 125: 97-104.[CrossRef][Medline]
Greenwald MK, Schuh KJ, Hopper JA, Schuster CR, and Johanson CE (2002) Effects of buprenorphine sublingual tablet maintenance on opioid drug-seeking behavior by humans. Psychopharmacology 160: 344-352.[CrossRef][Medline]
Griffiths RR, Bigelow GE, Liebson I, and Kaliszak JE (1980) Drug preference in humans: double-blind choice comparison of pentobarbital, diazepam and placebo. J Pharmacol Exp Ther 215: 649-661.[Abstract/Free Full Text]
Griffiths RR, Brady JV, and Snell JD (1978) Progressive-ratio performance maintained by drug infusions: comparison of cocaine, diethylpropion, chlorphentermine and fenfluramine. Psychopharmacology 56: 5-13.[CrossRef][Medline]
Grottick AJ, Fletcher PJ, and Higgins GA (2000) Studies to investigate the role of 5-HT(2C) receptors on cocaine- and food-maintained behavior. J Pharmacol Exp Ther 295: 1183-1191.[Abstract/Free Full Text]
Haney M, Ward AS, Foltin RW, and Fischman MW (2001) Effects of ecopipam, a selective dopamine D1 antagonist, on smoked cocaine self-administration by humans. Psychopharmacology 155: 330-337.[CrossRef][Medline]
Herd JA, Morse WH, Kelleher RT, and Jones LG (1969) Arterial hypertension in the squirrel monkey during behavioral experiments. Am J Physiol 217: 24-29.[Free Full Text]
Iglauer C and Woods JH (1974) Concurrent performances: reinforcement by different doses of intravenous cocaine in rhesus monkeys. J Exp Anal Behav 22: 179-196.[CrossRef][Medline]
Johanson CE and Schuster CR (1975) A choice procedure for drug reinforcers: cocaine and methylphenidate in the rhesus monkey. J Pharmacol Exp Ther 193: 676-688.[Abstract/Free Full Text]
Katz JL and Alling KL (2000) Discriminative stimulus effects of putative D3 dopamine receptor agonists in rats. Behav Pharmacol 11: 483-493.[Medline]
Katz JL, Alling K, Shores E, and Witkin JM (1995) Effects of D1 dopamine agonists on schedule-controlled behavior in the squirrel monkey. Behav Pharmacol 6: 143-148.[Medline]
Kleven MS and Woolverton WL (1993) Effects of three monoamine uptake inhibitors on behavior maintained by cocaine or food presentation in rhesus monkeys. Drug Alcohol Depend 31: 149-158.[CrossRef][Medline]
Khroyan TV, Barrett-Larimore RL, Rowlett JK, and Spealman RD (2000) Dopamine D1- and D2-like receptor mechanisms in relapse to cocaine-seeking behavior: effects of selective antagonists and agonists. J Pharmacol Exp Ther 294: 680-687.[Abstract/Free Full Text]
Mello NK, Lukas SE, Bree MP, and Mendelson JH (1990) Desipramine effects on cocaine self-administration by rhesus monkeys. Drug Alcohol Depend 26: 103-116.[CrossRef][Medline]
Mello NK and Negus SS (1996) Preclinical evaluation of pharmacotherapies for treatment of cocaine and opioid abuse using drug self-administration procedures. Neuropsychopharmacology 14: 375-424.[CrossRef][Medline]
Negus SS, Brandt MR, and Mello NK (1999) Effects of the long-acting monoamine reuptake inhibitor indatraline on cocaine self-administration in rhesus monkeys. J Pharmacol Exp Ther 291: 60-69.[Abstract/Free Full Text]
Paronis CA, Gasior M, and Bergman J (2002) Effects of cocaine under concurrent fixed ratio schedules of food and IV drug availability: a novel choice procedure in monkeys. Psychopharmacology 163: 283-291.[CrossRef][Medline]
Rinken A, Finnman UB, and Fuxe K (1999) Pharmacological characterization of dopamine-stimulated [35S]-guanosine 5'(gamma-thiotriphosphate) ([35S]GTP-gammaS) binding in rat striatal membranes. Biochem Pharmacol 57: 155-162.[CrossRef][Medline]
Schenk S (2002) Effects of GBR 12909, WIN 35,428, and indatraline on cocaine self-administration and cocaine-seeking in rats. Psychopharmacology 160: 263-270.[Medline]
Self DW, Barnhart WJ, Lehman DA, and Nestler EJ (1996) Opposite modulation of cocaine-seeking behavior by D1- and D2-like dopamine receptor agonists. Science (Wash DC) 271: 1586-1589.[Abstract]
Shalev U, Grimm JW, and Shaham Y (2002) Neurobiology of relapse to heroin and cocaine seeking: a review. Pharmacol Rev 54: 1-42.[Abstract/Free Full Text]
Spealman RD (1995) Noradrenergic involvement in the discriminative stimulus effects of cocaine in squirrel monkeys. J Pharmacol Exp Ther 275: 53-62.[Abstract/Free Full Text]
Spealman RD (1996) Dopamine D3 receptor agonists partially reproduce the discriminative stimulus effects of cocaine in squirrel monkeys. J Pharmacol Exp Ther 278: 1128-1137.[Abstract/Free Full Text]
Spealman RD, Bergman J, Madras BK, and Melia KF (1991) Discriminative stimulus effects of cocaine in squirrel monkeys: involvement of dopamine receptor subtypes. J Pharmacol Exp Ther 258: 945-953.[Abstract/Free Full Text]
Spealman RD, Madras BK, and Bergman J (1989) Effects of cocaine and related drugs in nonhuman primates. II. Stimulant effects on schedule-controlled behavior. J Pharmacol Exp Ther 251: 142-149.[Abstract/Free Full Text]
Tidey JW and Bergman J (1998) Drug discrimination in methamphetamine-trained monkeys: agonist and antagonist effects of dopaminergic drugs. J Pharmacol Exp Ther 285: 1163-1174.[Abstract/Free Full Text]
Woolverton WL (1987) Evaluation of the role of norepinephrine in the reinforcing effects of psychomotor stimulants in rhesus monkeys. Pharmacol Biochem Behav 26: 835-839.[CrossRef][Medline]
Woolverton WL (1996) Intravenous self-administration of cocaine under concurrent VI schedules of reinforcement. Psychopharmacology 127: 195-203.[Medline]
Woolverton WL and Balster RL (1981) Effects of antipsychotic compounds in rhesus monkeys given a choice between cocaine and food. Drug Alcohol Depend 8: 69-78.[CrossRef][Medline]
Woolverton WL, Goldberg LI, and Ginos JZ (1984) Intravenous self-administration of dopamine receptor agonists by rhesus monkeys. J Pharmacol Exp Ther 230: 678-683.[Abstract/Free Full Text]

This article has been cited by other articles:

				G. T. Collins and J. H. Woods Drug and Reinforcement History as Determinants of the Response-Maintaining Effects of Quinpirole in the Rat J. Pharmacol. Exp. Ther., November 1, 2007; 323(2): 599 - 605. [Abstract] [Full Text] [PDF]

				J. Bergman and C. A. Paronis Measuring the reinforcing strength of abused drugs. Mol. Interv., October 1, 2006; 6(5): 273 - 283. [Abstract] [Full Text] [PDF]

				M. A. Nader and P. W. Czoty PET Imaging of Dopamine D2 Receptors in Monkey Models of Cocaine Abuse: Genetic Predisposition Versus Environmental Modulation Am J Psychiatry, August 1, 2005; 162(8): 1473 - 1482. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.103.052795v1 308/1/249    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gasior, M. Articles by Bergman, J. Search for Related Content PubMed PubMed Citation Articles by Gasior, M. Articles by Bergman, J.