SEARCH

SEARCH BY CITATION

Keywords:

  • Cocaine;
  • hippocampus;
  • neurogenesis;
  • self-administration;
  • water T-maze;
  • working memory

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Authors Contribution
  9. References

Drug addiction is a chronic brain disorder, characterized by the loss of the ability to control drug consumption. The neurobiology of addiction is traditionally thought to involve the mesocorticolimbic system of the brain. However, the hippocampus has received renewed interest for its potential role in addiction. Part of this attention is because of the fact that drugs of abuse are potent negative regulators of neurogenesis in the adult hippocampus and may as a result impair learning and memory. We investigated the effects of different dosages of contingent cocaine on cell proliferation and neurogenesis in the dentate gyrus of the hippocampus and on working memory during abstinence, using the water T-maze test, in adult rats. We found that cocaine, in addition to the changes it produces in the reward system, if taken in high doses, can attenuate the production and development of new neurons in the hippocampus, and reduce working memory.


INTRODUCTION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Authors Contribution
  9. References

It has been demonstrated that repeated exposure to substances of abuse results in brain region-specific neuroadaptations that correlate with the loss of the ability to control drug use (Everitt, Dickinson & Robbins 2001). This evidence indicates a prominent role for limbic brain regions, such as the prefrontal cortex (PFC; Everitt & Robbins 2005) and the hippocampus in drug addiction (Berke & Eichenbaum 2001). Drug-induced neuroadaptations in the hippocampus are of particular interest, because of the fact that this structure is anatomically well positioned to influence brain reward circuitry. The hippocampus receives input from both the nucleus accumbens (NAc) and ventral tegmental area (VTA) and sends output to the NAc (Kelley & Domesick 1982; Totterdell & Smith 1989; Blood & Zatorre 2001). Furthermore, neuroadaptations in the PFC were recently shown to impair working memory in rats trained to self-administer cocaine (George et al. 2008). Given that the PFC densely and reciprocally innervates the hippocampal formation (Swanson & Kohler 1986; Jay, Glowinski & Thierry 1989; Jay & Witter 1991), it is essential to study the hippocampus's contribution to this task.

The sub-granular zone (SGZ) of the hippocampal dentate gyrus (DG) and the subventricular zone of the lateral ventricle are the two known neurogenic sites of the adult brain. In the hippocampus, contingents of newly generated cells born in the SGZ of the DG travel short distances to become incorporated as granular neurons in a dynamic process that continues throughout life (Alvarez-Buylla & Lim 2004; Kempermann, Wiskott & Gage 2004). The role of adult neurogenesis in the hippocampus is of present unclear. In light of the critical role of the hippocampus in the process of forming and recovering certain types of memory (Squire, Stark & Clark 2004), the integrity of the neurogenesis process may be necessary for the acquisition and consolidation of memories (Shors et al. 2001; Cao et al. 2004; Crandall et al. 2004). Only a limited number of studies have examined the role of working memory that involves both the hippocampus and the PFC (Wall & Messier 2001; Jones 2002). Winocur et al. (2006) suggest that the suppression of hippocampal neurogenesis interferes with working memory at long intra-trial delays. Others argue that it might not fulfill a unitary function in memory and may have opposite roles in distinct types of memory (Saxe et al. 2007).

Accrued experimental evidence shows that addictive substances, such as alcohol (Herrera et al. 2003), opiates (Eisch et al. 2000; Eisch & Harburg 2006) and amphetamines (Teuchert-Noodt, Dawirs & Hildebrandt 2000) can negatively affect the self-renewal capacity of the hippocampus by diminishing the rate of proliferation of neural progenitors or by impairing the long-term survival of neural precursors, or both.

A number of studies have demonstrated that chronic cocaine exposure (non-contingent or contingent drug administration) decreases the rate of proliferation (Yamaguchi et al. 2004, 2005; Dominguez-Escriba et al. 2006; Noonan et al. 2008) of neural progenitor cells in the DG without altering their rate of survival (Dominguez-Escriba et al. 2006; Noonan et al. 2008). There remains, however, considerable uncertainty over the functional significance of this rate of proliferation in relation to cocaine abuse. One study (Del Olmo et al. 2006) demonstrated that cocaine self-administration affects long-term potentiating (LTP) but does not notably affect performance in the Morris water maze test. In contrast to these findings, in a different study, results showed an improvement in the rats' performance in a difficult Morris water maze task following cocaine self-administration (Del Olmo et al. 2007).

We used the self-administration method combined with the water T-maze test to examine the effect of withdrawal of dose-dependent chronic cocaine intake on neurogenesis in the DG of the hippocampus, and on working memory during abstinence in adult rats.

MATERIALS AND METHODS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Authors Contribution
  9. References

Subjects

Male Sprague–Dawley rats (250–280 g) were maintained on a 12–12 hour dark-light cycle with free access to food and water. All experimental procedures were approved by the Animal Care and Use Committee of the University and were performed in accordance with the guidelines of the National Institutes of Health.

Behavioral procedures

Cocaine self-administration

The rats were implanted with intravenous silastic catheters under anesthesia (ketamine 100 mg/kg and xylazine 10 mg/kg) into the right jugular vein (Roth-Deri et al. 2003). The catheter was secured to the vein with silk sutures and was passed subcutaneously to the top of the skull where it exited into a connector (a modified 22-gauge cannula; Plastics One, Roanoke, VA) mounted to the skull with MX-80 screws (Small Parts, Inc., Miami Lakes, FL) and dental cement (Yates & Bird, Chicago, IL). Catheters were flushed every 24 hours with sterile saline and the antibiotic gentamicine (0.08 mg/ml) to prevent infections.

Rats were trained to self-administer cocaine or saline as previously described (Roth-Deri et al. 2003). Briefly, 5 days after catheterization, rats were transferred to operant conditioning chambers (Medical-Associates, Inc.; St. Albans, VT; hardware and software of interface were designed by Dr M. Ben Tzion using LabVIEW, National Instruments) for 1 hour daily until stable maintenance levels were attained (at least 3 days of 20% deviation from the mean) during their dark cycle, and were allowed to self-administer saline (0.13 ml/infusion) or cocaine (0.13 ml/infusion; 0.5 mg/kg/20 seconds; or 1.5 mg/kg/20 seconds; obtained from the National Institute of Drug Abuse, Research Technology Branch, Rockville, MD) via a lever press under an FR-1 schedule of reinforcement. The group defined as 0.5 mg/kg cocaine, received 1 mg/kg cocaine per infusion during the first 5 days of training and then switched to 0.5 mg/kg cocaine for the rest of the experiment. During the infusion, a light located above the active lever was lit for 20 seconds. During the 20-second intervals of cocaine infusion, active lever presses were recorded, but no additional cocaine reinforcement was provided. Presses on the inactive lever were recorded, but did not activate the infusion pump.

Water T-maze

General procedure.  The water escape T-maze consisted of grey Plexiglas T-maze pool (1-cm thick) filled with water (23 ± 1°C). The main alley (100 cm, 20 cm, 40 cm) was connected to two side arms (right and left) (45 cm, 20 cm, 40 cm) by two sliding doors manually operated to close off either arm. At the end of each arm there was a platform (Plexiglas, 15 cm, 18 cm) submerged 2 cm below the surface of the water. The apparatus was set on a 75-cm high table, in a room without visual cues that could be used by the animals to guide their response. In addition, the behavioral task was performed under red light, during the dark period of the circadian light/dark cycle. The delayed alternation task in the water escape T-maze consisted of a pseudorandom sequence of 10 discrete trial pairs of forced-choice runs. Briefly, each trial pair consisted of a forced run in which animals were given access to only one arm (right or left), where a submerged platform to escape from the water was located, followed by a choice run in which animals have access to both arms, but the platform was found in the arm opposite the one they had just entered on the previous forced run. Therefore, rats had to learn to alternate in order to find the submerged platform during the choice run. If the animal chose the arm where the platform is not present (i.e. the same arm where it was forced to enter in the forced run), the sliding door of that arm was closed and the animal was maintained 10 seconds in the water (failed choice). Afterwards the sliding door was opened to allow the animal to find the submerged platform located in the opposite arm. Once the animal reached the platform after either forced or choice runs, the sliding door was closed and it was maintained on the platform for 10 seconds. The retention intra-trial interval (delay) between the forced run and the choice run was set at 10 seconds and the intertrial interval between trial pairs was set at 30 seconds. During intratrial and intertrial intervals, animals were placed in a plastic holding cage (27 cm, 27 cm, 23 cm) that was placed adjacent to the maze.

Training procedure.  On the first day, animals were gently immersed in the water escape T-maze for 1 minute, without platforms. On the second day, they were subjected to 10 trial pairs of forced-forced alternation runs in which during the second run of the pair, the animals had access only to the opposite arm they had visited previously. From the third day on, animals were subjected to 10 trial pairs of forced-choice alternation runs. A different, pseudorandom sequence of forced runs was used every day (e.g. L-R-L-L-R-L-R-R-L-R), for all animals tested. Animals were trained until they performed correctly 7 or more out of 10 trials (> 70% correctness) for 3 consecutive days (Del Arco et al. 2007).

Testing brain cell- proliferation and neurogenesis

Administration of BrdU and tissue preparations

Rats housed under standard conditions were injected i.p. with BrdU (Sigma-Aldrich; 50 mg per kg body weight) three times at 4-hour intervals, after reaching stable maintenance levels. Twenty-four hours (proliferation test) or 28 days (neurogenesis test) after the first BrdU injection, rats were euthanized and perfused transcardially, first with phosphate buffered saline (PBS) and then with 4% paraformaldehyde. Their brains were removed, postfixed overnight and equilibrated in phosphate-buffered 30% sucrose. Coronal hippocampal sections (40-µm thick) were collected on a freezing cryostat. Every ninth section (total of nine sections, 360 µm apart) was taken and stored free-floating in PBS containing sodium azide (1%) at 4°C for immunohistochemical analysis.

Immunohistochemistry

For BrdU staining of the rats' specimens' tissue, sections were washed with PBS, incubated in 2N HCl at 37°C for 30 minutes and then blocked for 1 hour with blocking solution (PBS containing 20% normal horse serum and 0.5% Triton X-100). The tissue sections were stained overnight with specified combinations of the following primary antibodies: rat anti-BrdU (1:200; Oxford Biotechnology, Kidlington, Oxfordshire, UK) and mouse anti-NeuN (1:200; Chemicon, Temecula, CA). Secondary antibodies used for both rat tissues were Cy-3-conjugated donkey anti-rat (1:200; Jackson ImmunoResearch, West Grove, PA) and Cy-2 donkey anti-mouse (1:200; Jackson ImmunoResearch).

Quantification

Proliferation in the DG was measured by counting manually the cells that were labeled for BrdU. Neurogenesis was evaluated by counting the cells that were double labeled with BrdU and NeuN (using fluorescent microscope Nikon Eclipse E400, Yokohama, Japan; magnification ×400). We counted the number of labeled cells in nine coronal sections per rat brain that were stained and mounted on coded slides (blind to the observer). To obtain an estimate of the total number of labeled cells per DG, the total number of cells counted in the selected coronal sections from each brain was multiplied by the volume index (the ratio between the volume of the DG and the total combined volume of the selected sections). Cellular co-labeling of BrdU and NeuN was confirmed by confocal microscopy (Zeiss LSM 510 laser scanning microscope, Jena, Germany).

Experiment 1a: effect of cocaine self-administration on hippocampal proliferation

A catheter was surgically implanted i.v. as described previously. Rats were randomly divided into three groups, trained to self-administer either cocaine (0.5 mg/kg, n = 5 or 1.5 mg/kg, n = 11) or saline (n = 7). On day 15 of the experiment (1 day after cocaine-trained rats reached maintenance), all three groups were injected with BrdU as described previously. Twenty-four hours after the first BrdU injection, rats were euthanized and brains were labeled for BrdU (Fig. 1).

image

Figure 1. Flow chart of experimental procedures. Rats were trained to self-administer cocaine (0.5 or 1.5 mg/kg) or saline for 14 days. When the rats achieved stable maintenance, they were injected i.p. with BrdU (50 mg/kg) three times, at 4-hour intervals for 1 day. Twenty-four hours or 28 days after BrdU injection, rats were euthanized and their brains were excised. Cells that were double positive for BrdU, and the mature neuronal marker, NeuN, were counted in the hippocampi (Experiment 1a, 1b). In a following experiment, rats were trained to self-administer cocaine (1.5 mg/kg) or saline for 14 days. After reaching stable maintenance levels, they started a water escape T-maze training procedure. On the first day of training, animals were immersed in the maze for 1 minute, without platforms. On the second day, they were subjected to 10 trial pairs of forced-forced alternation runs. From the third day on, rats were subjected to 10 trial pairs of forced-choice alternation runs. At the cessation of the last trial, rats were euthanized and their brains were excised. Cells that were double positive for BrdU, and the mature neuronal marker, NeuN, were counted in the hippocampi (Experiment 2)

Download figure to PowerPoint

Experiment 1b: effect of cocaine self-administration on hippocampal neurogenesis

The same procedure as in experiment 1a was used, albeit that rats were injected with BrdU and 28 days later (day 43 of the experiment), rats were euthanized and brains were double labeled for BrdU and NeuN. Rats were trained to self-administer cocaine (0.5 mg/kg, n = 5 or 1.5 mg/kg, n = 15) or saline (n = 12). On day 15 of the experiment (1 day after cocaine-trained rats reached maintenance), all three groups were injected with BrdU as described previously. Twenty-eight days later (day 43 of the experiment), the rats were euthanized and brains were double labeled for BrdU and NeuN (Fig. 1).

Experiment 2: association of the performance of rats in a water T-maze and neurogenesis after cocaine self-administration

An i.v. catheter was surgically implanted as described previously. Rats were randomly divided into two groups, trained to self-administer either cocaine (1.5 mg/kg, n = 15) or saline (n = 15). On day 15 of the experiment (1 day after the cocaine-trained rats reached maintenance), both groups were injected with BrdU (as described previously) and on the next day rats were trained in the water T-maze. At the secession of the water T-maze training (day 36 of the experiment), rats were euthanized and brains were double labeled for BrdU and NeuN (Fig. 1).

Statistical analysis

For the behavioral data (cocaine self-administration and water T-maze test), we applied one way analysis of variance (ANOVA) with repeated measures followed by a post hoc test Student Newman–Keuls. The between groups variable was the factor of treatment (1.5 mg/kg cocaine, 0.5 mg/kg cocaine or saline) and the within subjects' variable was factor of days.

The histological data (cell proliferation and survival) were analyzed by ANOVA followed by Student Newman–Keuls post hoc test. For comparison between cell proliferation and cell survival, we used two-way ANOVA followed by Student Newman–Keuls post hoc test.

To analyze how cell proliferation and neurogenesis in the hippocampus is affected by cocaine consumption, we used linear regression analysis.

Data are presented as mean ± SEM. Results were considered significantly different if P < 0.05.

RESULTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Authors Contribution
  9. References

Cocaine self-administration

Rats were trained to self-administer cocaine (0.5 or 1.5 mg/kg, respectively) or saline over 14 days. A one-way ANOVA with repeated measures for the active lever presses revealed a main effect of treatment [F(2, 840) = 76.14, P < 0.001], time [F(13, 840) = 14.32, P < 0.001] and treatment × time interaction [F(26, 840) = 12.6, P < 0.001]. For number of inactive lever presses, there was a main effect of treatment [F(2, 840) = 192.5, P < 0.001], no main effect of time [F(13, 840) = 1.18, P > 0.05] and treatment × time interaction [F(26, 840) = 1.48, P = 0.05]. The difference between inactive lever presses between treatment groups stems from relative high inactive lever responses of the saline treated rats compared with cocaine self-administered rats (0.5 and 1.5 mg/kg cocaine, P < 0.05, the Student Newman–Keuls post hoc test). However, the lack of distinction, displayed by saline-treated rats, between active and inactive levers indicates that rats were not rewarded by saline. Also, in the groups of cocaine self-administration (0.5 or 1.5 mg/kg), the number of inactive lever responses was low and did not differ significantly throughout the days of the experiment (P > 0.05, the Student Newman–Keuls post hoc test) (Fig. 2).

image

Figure 2. Cocaine or saline self-administration. Rats were trained to self-administer saline (0.13 ml/infusion; panel a) or cocaine (0.13 ml/infusion; 0.5 mg/kg/20 seconds, panel b; or 1.5 mg/kg/20 seconds, panel c), on a fixed ratio 1 schedule of reinforcement in a daily 1-hour session, until stable maintenance levels were attained

Download figure to PowerPoint

Experiment 1a: effect of cocaine self-administration on cell proliferation

When the rats achieved stable maintenance of cocaine administration, they were injected i.p. with BrdU three times, at 4-hour intervals. Twenty-four hours later, rats were euthanized and brains were sliced and labeled for BrdU. Slices were counter-stained with NeuN. A one-way ANOVA showed a significant decrease in newly formed cells (BrdU+; F(2, 19) = 4.1, Student Newman–Keuls post hoc analysis P < 0.05) in the dentate gyri of rats trained to self-administer cocaine at dose of 1.5 mg/kg compared with control (saline treated), whereas cocaine at dose of 0.5 mg/kg did not affect the number of newly formed cells in comparison with control (Fig. 3a).

image

Figure 3. Cocaine self-administration decreases cell proliferation and neurogenesis in the hippocampus. Adult hippocampi slices were prepared as described in Methods. Panel a: Quantification of BrdU+ cells in the dentate gyrus (DG) of rats showed significantly less newly formed cells (proliferation) in the dentate gyri of rats trained to self-administer cocaine 1.5 mg/kg compared with control (saline treated), whereas cocaine 0.5 mg/kg did not affect the number of newly formed cells compared with control (*P < 0.05 versus 0.5 mg/kg cocaine and saline groups). Panel b: Quantification of BrdU+ and NeuN+ cells in the DG of rats showed significantly less newly formed cells (BrdU+) and neurons (BrdU+ NeuN+; differentiation) in the dentate gyri of rats trained to self-administer cocaine 1.5 mg/kg compared with control (saline treated), whereas cocaine 0.5 mg/kg did not affect the number of newly formed cells and neurons compared with control (panel b; *P < 0.001 versus the correspondence measure in the 0.5 mg/kg cocaine and saline groups). Panel c: Comparison of total BrdU+ cells in the DG of rats showed a significant decrease in cell proliferation and in cell survival in the dentate gyri of rats trained to self-administer cocaine 1.5 mg/kg compared with saline self-administeed group and to 0.5 mg/kg cocaine group (P < 0.05, Student Newman–Keuls). Panels d–f: Representative micrographs of newly generated cells, BrdU+ (red) summed in panel a are depicted; in saline self-administration (d), 0.5 mg/kg cocaine self-administration (e), 1.5 mg/kg cocaine self-administration (f). Slices were counter stained with NeuN+ (green) for anatomical demonstration. Panel g–i: A representative micrograph (double labeling) of newly generated neurons NeuN+ (green) and BrdU+ (red) summed in panel b are depicted; in saline self-administration (g); 0.5 mg/kg cocaine self-administration (h); 1.5 mg/kg cocaine self-administration (i)

Download figure to PowerPoint

Experiment 1b: Effect of cocaine self-administration on hippocampal neurogenesis

Some of the rats that achieved stable maintenance of cocaine self-administration were injected i.p. with BrdU three times at 4-hour intervals, and their brains were removed after 28 days. These brains were analyzed for BrdU and NeuN. A one-way ANOVA of BrdU+ cells in the DG showed a main effect of group [F(2,29) = 21.57; P < 0.001]. Student Newman–Keuls post hoc test showed a significant decrease in newly formed cells in the dentate gyri of rats trained to self-administer cocaine at 1.5 mg/kg compared with control (saline treated), whereas 0.5 mg/kg cocaine did not affect the number of newly formed cells in comparison with control (P < 0.001).

A one-way ANOVA of BrdU+ NeuN+ cells in the DG showed a main effect of group [F(2,32) = 16.95; P < 0.001]. The post hoc test showed a significant decrease in newly formed neurons in the dentate gyri of rats trained to self-administer cocaine at 1.5 mg/kg compared with control (saline treated), whereas cocaine 0.5 mg/kg did not affect the number of newly formed neurons in comparison with control (P < 0.001; Figure 3b).

A two-way ANOVA of total BrdU+ cells in the DG of rats showed a main effect of group [F(2,36) = 23.4; P < 0.001], a main effect of proliferation and survival [F(5,36) = 2.4; P < 0.01] and no main effect in interaction [F(10,36) = 0.8; P > 0.05]. A post hoc test showed a significant decrease in cell proliferation and cell survival in the dentate gyri of rats trained to self-administer cocaine 1.5 mg/kg compared with the saline self-administered group and to the 0.5 mg/kg cocaine group (P < 0.05, P < 0.01, respectively; Student Newman–Keuls). Moreover, the extent of decrease in the number of BrdU+ cells was significantly greater when tested 28 days after BrdU injection than the decrease in number of BrdU+ cells when tested 24 hours after BrdU injection, only when rats were allowed to self administer 1.5 mg/kg cocaine (P < 0.001, Student Newman–Keuls; Figure 3c).

Experiment 2: association of the performance of rats in a water T-maze and neurogenesis after cocaine self-administration

Because Experiment 1b indicated a decrease in neurogenesis only when rats were allowed to self-administer 1.5 mg/kg, we conducted another experiment to test learning and memory abilities, after reaching stable maintenance, using the water T-maze in parallel to neurogenesis in the same subjects. The results indicate that from day 10 of the test and onward, cocaine-administered rats showed significantly lower levels of performance compared with saline-administered rats and did not reach the criterion for learning the task [performing correctly in at least 7 out of 10 trials (70%) for three consecutive days throughout all tested days (Del Arco et al. 2007)]. ANOVA with repeated measures of the correct choices revealed a significant effect of treatment F[(1, 448) = 63.08, P < 0.0001], time [F(15,448) = 33.58, P < 0.001] and treatment × time interaction [F(15,448) = 2.51, P < 0.005] (Fig. 4a).

image

Figure 4. Association of the performance of rats in a water T-maze and neurogenesis after cocaine self-administration. Panel a: Rats were trained to self-administer cocaine (1.5 mg/kg/infusion) or saline. After reaching stable maintenance, they were tested for learning and memory abilities in the water T-maze. From day 10 of the test and onward, cocaine-administrated rats showed significantly lower levels of performance compared with saline administrated rats and did not reach the learning criterion throughout all tested days (P < 0.005, Student Newman–Keuls). Panel b: Adult hippocampi slices were prepared as described in Methods. Quantification of BrdU+ and NeuN+ cells in the dentate gyrus of rats showed significantly less newly formed cells (BrdU+) and neurons (BrdU+ NeuN+; differentiation) in the dentate gyri of rats trained to self-administer cocaine 1.5 mg/kg compared with saline treated rats (*P < 0.001, Student Newman–Keuls)

Download figure to PowerPoint

Results of behavioral performance were associated with the number of newly formed matured neurons. A one-way ANOVA of BrdU+ cells in the DG of rats showed a main effect of group [F(1,28) = 45.15; P < 0.001]. A post hoc test showed a significant decrease in newly formed cells in the dentate gyri of rats trained to self-administer cocaine compared with saline-treated rats (P < 0.001).

A one-way ANOVA of BrdU+ NeuN+ cells in the DG of rats showed a main effect of group [F(1,28) = 32.54; P < 0.001]. A post hoc test showed a significant decrease in newly formed neurons in the dentate gyri of rats trained to self-administer cocaine compared with saline-treated rats (P < 0.001; Fig. 4b).

DISCUSSION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Authors Contribution
  9. References

Our results indicate a significant decrease in cell proliferation and in newly generated neurons in the DG of the hippocampus in rats trained to self-administer cocaine. This effect was concomitant with impairment in the performance of working memory during withdrawal from cocaine.

This decrease was only identified when a 1.5 mg/kg/injection cocaine was allowed, but not at a lower dose. The extent of decrease in the number of BrdU+ cells was significantly greater when tested 28 days after BrdU injection than the decrease in the number of BrdU+ cells when tested 24 hours after BrdU injection. Therefore, we postulate that not only cell proliferation was affected by this specific dose, but also cell survival.

In contrast to our results, others have reported that chronic cocaine administration does not alter the survival of neural progenitor cells in the DG (Dominguez-Escriba et al. 2006; Noonan et al. 2008). However, in one study (Dominguez-Escriba et al. 2006), BrdU was injected prior to non-contingent cocaine administration and assayed 24 days later, while in another study, BrdU was injected at maintenance (Noonan et al. 2008) and the self-administration protocol was used only with a rather low dose of cocaine (0.5 mg/kg/per injection).

Cocaine at a dose of 0.5 mg/kg/injection did not affect cell proliferation or survival of newly formed neurons, but results of a threefold higher dose (1.5 mg/kg/injection) showed a decrease in proliferation and to higher extent in the survival of newly generated neurons in the DG. The total amount of cocaine injected, during the 5 days of maintenance in rats trained on 1.5 mg/kg/injection, did not significantly differ from the total amount of cocaine injected in rats trained on 0.5 mg/kg/injection (51 ± 3 mg/kg and 42 ± 4 mg/kg, respectively). Because of cocaine's rapid half-life, and the fact that BrdU labels divide progenitors, it is likely that any effect of cocaine on the progenitor cells at the time of labeling is a result of the amount of cocaine present that day. In the final cocaine session before BrdU injection, rats in both the 1.5 and 0.5 mg/kg/injection groups consumed about 7.5 mg/kg/session. Thus, the progenitor cells should have been exposed to equal amounts of cocaine in each group. However, neurogenesis is a process rather than a single time event and therefore, although the amount of cocaine was equal at the time of BrdU injection, the total longitudinal cocaine consumption during the entire experiment was higher in 1.5 than in 0.5 mg/kg-trained rats (122 ± 6 mg/kg and 90 ± 5 mg/kg, respectively; P < 0.05). The amount of cocaine intake negatively correlated with the number of cell proliferation and neurogenesis (r2 = 0.75, P < 0.001 and r2 = 0.76, P < 0.0005, respectively). It has been previously suggested that high versus low doses and prolonged versus short access to cocaine result in a different pattern of cocaine intake that reflects changes in the brain receptors (Ben Shahar et al. 2007; Wee, Specio & Koob 2007). Based on these reports and on our results, it seems that the intensity and duration of exposure to the substance relate to the decrease in neurogenesis.

Integrated theories regarding the role of the hippocampus in learning and memory have been suggested (Rolls et al. 1998). However, the literature on the function of adult hippocampal neurogenesis leaves room for debate (Kempermann 2002; Kempermann et al. 2004; Lledo, Alonso & Grubb 2006; Jessberger et al. 2009). Experiments aimed at knocking out neurogenesis have proven to be inconclusive regarding the effect on memory. A significant reduction in the number of new neurons in the adult hippocampus was associated with impaired memory performance in some tasks, but not with others. Specifically, treatment with an anti-mitotic agent reduced the amount of fear acquired after exposure to trace fear conditioning paradigm, but did not affect contextual fear conditioning or spatial navigation learning in the Morris water maze (Shors et al. 2002). These results suggest that neurogenesis may be associated with the formation of some types of hippocampal-dependent memories. Similarly, the effects of cocaine self-administration on memory are unclear. On the one hand, cocaine self-administration has been shown not to affect spatial memory (Del Olmo et al. 2006) while on the other hand, a different study showed an improvement in performance in a difficult Morris water maze task following cocaine self-administration (Del Olmo et al. 2007).

It has been suggested that cocaine seeking as a goal-directed behavior involves long-term adaptations in the PFC that can serve to explain the impairment of working memory seen only in rats exposed to prolonged excess of cocaine (George et al. 2008). It has been postulated that these alterations serve as the mechanism underlying the compulsivity aspect of drug seeking. Nonetheless, it is important to note that working memory involves both the hippocampus and PFC; the hippocampus–orbitomedial PFC circuit activates, maintains, monitors and modifies current and recent past cognitive and emotional representations and enables the integration of cognition, emotion and behavior (Wall & Messier 2001).

Our aim in this study was to test working memory performance of previously addicted rats during abstinence in relation to the formation of new hippocampal neurons (BrdU+/NueN+). A study by George et al. (2008) that examined working memory following cocaine self-administration did not discover changes when using 0.5 mg/kg/injection. Accordingly, previously reported data (Noonan et al. 2008) as well as the present data showed that 0.5 mg/kg/injection cocaine self-administration did not affect neurogenesis. In light of these findings, we assumed that if a connection between working memory and neurogenesis exists, it may be found in rats trained to inject 1.5 mg/kg cocaine, the dose we found to have an effect on newly formed neurons. Indeed, a decrease in neurogenesis was associated with a deficit in working memory performance at this dose. Research has shown (Wang, Scott & Wojtowicz 2000) that young neurons are completely unaffected by GABA-A inhibition and produce LTP more readily than mature neurons. As LTP is considered to be a neural basis of learning, even a minor decrease in the number of young neurons may exert a significant impact on hippocampal dependent learning. Several computational theories have recently been suggested to explain the functional role of neurogenesis, including the idea that memory capacity increases with the number of dentate granule cells, while neuronal turnover with a fixed dentate layer size improves recall by minimizing interference between highly similar items (Becker 2005). Another hypothesis is that newly formed neurons assist adaptation to new situations. The assumption behind this hypothesis is that old neurons are rather stable and preserve an optimal encoding of known environments, while the plasticity of new neurons enables them to adapt and encode new features present in new environments. A simple network simulation has demonstrated that the addition of new plastic neurons proved to be a successful strategy for adaptation, with limited interference between different memories (Wiskott, Rasch & Kempermann 2006).

Based on some of the existing literature (Simon et al. 2002) and on our own study's results, it is tempting to speculate that cocaine attenuates the development and survival of new neurons, which in turn leads to a reduction in cognitive performance. Furthermore, we believe that the difficulty in creating new memories could eventually evoke memories of past drug-induced behavior and as a result increase the propensity for drug relapse.

Acknowledgements

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Authors Contribution
  9. References

This study was supported by a grant from The Israel Ministry of Health to GY and with a fellowship by the Israel Anti-Drug Authority to ES. The study is part of Einav Sudai's PhD dissertation, in the Mina & Everard Goodman Faculty of Life Sciences, Bar-Ilan University, Ramat-Gan 52900, Israel. The authors wish to thank Dr D. Har-Even for his critical help with the statistics analyses and Dr Y. Ziv for consultation and helpful advice with neurogenesis data.

Authors Contribution

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Authors Contribution
  9. References

ES performed the experiments, conducted data acquisition and analyses, and wrote the manuscript; OC performed the experiments that constitute the main body of this work; AS performed the water T maze experiments and data analysis; LA and IRD performed the in-vivo experiments; IG technical support and surgery; YF paper preparation and technical support; NK carried out the histological and immunofluorescence studies; SA constructed the figures; MBZ instrumental and program design and support; GY supervised the project, designed the experiments and helped write the manuscript.

References

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Authors Contribution
  9. References