Thursday, October 16, 2008

Endogenous Reward System: the Etiology and Treatment of Depression

by Tara Deliberto


The possibility exists that dysfunction at various levels of a biological reward system, such as the dopaminergic mesolimbic pathway, could be involved in etiology and maintenance of depression. Current descriptions of the role that specific factors, such as dopamine, play in the endogenous reward system have been incomplete; therefore, other factors are likely involved as well.

Keywords: Depression, endorphin, beta-endorphin, etiology, treatment


It is well known that environmental reinforcement has an effect on one’s behavior; therefore, a biological system through which the organism is able to process and experience reinforcement must exist, yet is not extremely well understood. It is possible that through dysfunction at various levels of a biological reward system, psychological disorders such as depression can result or be maintained. Furthermore, if acquired or genetic dysfunctions in the biological reward system result in psychopathology, it can be argued that ameliorating these problems can result in further success in treating depression.

Several possible brain structures have emerged as being involved in the biological reward system. In recent years, the nucleus accumbens (NAS) has been supported as the key structure in the reward center of the brain. The involvement of the NAS in reinforcement has been demonstrated for both natural (logically related to the task) and artificial rewards (not logically related to the task; Rada, Colasante, Skirzewski, Hernandez & Hoebel, 2006). Reinforcement is hypothesized to be mediated through regulation of dopamine in the NAS; however, several pieces of evidence point to additional contributing factors. For example, while depletion of dopamine in the NAS has been shown to affect operant responding, it does not necessarily affect unconditioned reinforcement (Salmone & Correa, 2002). This means that DA does not affect conditioning to reinforcers that are naturally reinforcing. Because natural reinforcers such as food can be quite strong, DA’s role in conditioning may not be as important as previously thought. Therefore, other biological factors are likely involved in learning through reinforcement.

Also present in the NAS is beta-endorphin, an endogenous opioid peptide that has been implicated in the reward process. For example, it has been shown that beta-endorphin levels have increased in the NAS in rats during extinction learning. The results from this study indicate that beta-endorphin can be involved in both learning and adaptive responding to distress (Rada et al., 2006). Because beta-endorphin may be involved with operant conditioning and responses to stress, it could be involved with the onset or maintenance of depression (and impulsive behavior as well).

Role of beta-Endorphin in Onset

Due to a possible dysfunction in the reward system from environmental or biological etiology, the organism may not experience the environment as rewarding. It has been well-observed that people may actively seek out maladaptive ways to be rewarded. Various addictive disorders such as alcoholism, opiate addiction, cigarette smoking, excessive exercise, and starvation have all been linked to beta-endorphin dysfunction, and therefore, can possible be viewed as reward seeking behaviors. For example, people with a high genetic risk for developing alcoholism have lower levels of beta-endorphin. Perhaps not surprisingly, drinking alcohol increases levels of beta-endorphin (Dai, Thavundayil & Glanoulakis, 2005; Zalewska-Kaszubska & Czarnecka , 2005). In addition, it has been indicated that single-nucleotide polymorphism in the mu opiod receptor gene alters beta-endorphin binding and activity exists and could be related to opiate addiction (Bond, LaForge, Tian, Melia, Zhang, Borg, et al., 1998). Furthermore, people who smoke cigarettes have lower levels of basal beta-endorphin than those who do not smoke (Lee, Joe, Sohn, Na, Kee & Chae, 2005). Those who smoke more than 20 cigarettes per day have lower levels of beta-endorphin than those who smoke less than 10 cigarettes per day. In addition, those who smoke less than 10 cigarettes per day have lower beta-endorphin levels than controls (de Arbol, Munoz, Ojeda, Cascales, Irles, Miranda, et al., 2000). In addition, people with autism who engage in non-suicidal self-injury, the direct and deliberate destruction of one’s own body tissue without intent to die, have been shown to have lower baseline levels of beta-endorphin compared to controls (Sandman, Barron, Chicz-DeMet, & DeMet, 1990). It is also possible that engaging in self-injury increases beta-endorphin (Favazza & Conteiro, 1988). Increasingly, high frequency of engaging in this behavior is coming to be viewed as addiction. Along with self-injury beginning to be viewed as an addiction, food/sugar addiction is also being considered. It has recently been shown that endogenous opiods play a role in what is viewed as sugar addiction (Avena, Rada, Hoebel, 2008). Excessive exercise as well as starvation have also been shown to increase beta-endorphin (Davis & Claridge, 1998). In rats, activity anorexia- the phenomenon in which after rats are exposed to an intermittent schedule of food reinforcement for lever pressing, they run in the wheel in their cage for the period between reinforcers- has been observed. In activity anorexia, within a week of being reinforced on a variable ratio schedule, rats will run up to 12 miles in one day and voluntarily deny food when it is presented (Collier, G. & Levitsky, 1968). If allowed to continue, rats will starve themselves to death. It seems that a positive feedback loop (simply stated, which continues in the same direction as the problem) develops in which increased running and decreased eating are somehow reinforced. This phenomenon suggests that a strong, endogenous, biological reward system maintains these behaviors, which may have developed for evolutionary purposes but become maladaptive under certain conditions, such as when food is presented and denied. It is probable that due to an environmental or a directly biological (such as a genetic predisposition) factor causing dysfunction in the reward system, people actively seek out stimuli that raise their beta-endorphin levels such as smoking, drinking, exercising, and restricting food intake. People with reward system dysfunctions could seek out stimuli that are both maladaptive and potent in an effort to change their experiences.

As with disorders like substance abuse and anorexia nervosa, it is possible that people with major depression have dysregulated reward or pleasure systems. Indeed, a major symptom of depression, anhedonia, is described as the lack of ability to feel pleasure. It has been shown that people with major depression release a smaller amount of beta-endorphin compared to controls when presented with a social stressor (Young, Lopez, Murphy-Weinberg, Watson & Akil, 2000). Similar to learned helplessness, it is possible that depression results from learning that physiological rewards cannot be experienced either because of a direct inability to experience pleasure from the reward, the inability to make the association between the award and the experience of pleasure, an actual lack of rewards in the environment (or excess of punishments), or any combination of biological and environmental factors within a given person. The possibility also exists that due to dysfunction involved with beta-endorphin, specific learned helplessness towards gaining weak adaptive rewards and specific addictions to potent maladaptive reinforcers can co-occur in one organism, for example, comorbidity between major depression and alcohol dependence.

Although the studies mentioned above may indicate that beta-endorphin may be involved in the experience of pleasure and aid in the formation of associations between stimuli and responses, the results of several studies appear to be contrary to this notion. For example, lateral hypothalamus self-stimulation (LHSS) has been shown to be an effective reinforcer in rats; however, increases in extracellular levels of beta-endorphin in rats were not observed in the NAS as a result of LHSS in one study (Zangen & Shalev, 2003). These results indicate that stimulation of the lateral hypothalamus may not related to beta-endorphin in the reward system; however, the results of the study indicate that beta-endorphin may play a role adaptively responding to stress because extracellular levels increased in the NAS during extinction and presentation to aversive stimuli. In addition, Maes, et al (1994) found that administering corticotrophin-releasing hormone (CRH) 9.5 hours after dexamethasone resulted in enhancement of beta-endorphin. If a lack of beta-endorphin is involved in depression, one might expect that beta-endorphin would be decreased rather than increased in people with depression compared to controls; however, since no control group was used and all participants had depression, it remains unclear if people with depression have less beta-endorphin secretion compared to controls. Also in this study, levels of beta-endorphin were not different between people with major or minor depression. This indicates that the level of beta-endorphin may not be related to clinical severity of depression. It is important to remember that the role of beta-endorphin in the onset, maintenance, and treatment of depression remains speculative.

Despite this contradictory evidence, several different pieces of information regarding dopamine and serotonin may provide a link for beta-endorphin with depression. For example, beta-endorphin has been shown to decrease dopamine turnover in the hypothalamus and striatum (George & Van Loon, 1982). Antidepressant drugs such as bupropion (Wellbutrin) that act by selectively inhibiting the uptake of dopamine may indirectly work by acting on endorphins as well. Serotonin, a neurotransmitter known to be involved with major depression, when given to rats either exogenously or endogenously, facilitates the release of beta-endorphin in the arcuate nucleus in the hypothalamus and the NAS (Zangen, Nakash, & Yadid, 1999). Therefore, it is possible the interaction between serotonin and beta-endorphin is responsible for part of the effectiveness of selective serotonin reuptake inhibitors (SSRIs) used to treat depression.

Role of beta-Endorphin in Treatment

Not only may beta-endorphin be involved in the onset of depression, it may be involved in various forms of treatments as well. Recently, a N-methyl-D-aspartate (NDMA) receptor antagonist, also known as ketamine hydrochloride, was administered in a randomized, placebo controlled, double-blind study to a group of people with depression. After only two hours after administration, the low dose of ketamine had an antidepressant effect that lasted about a week (Zarante, Singh, Carlson, Butsche, Ameli, Luckenbaugh, et al., 2006). Of those who received ketamine, 71% met criteria for response and 29% met criteria for remission the day after administration. Therefore, 100% of patients at least had an immediate response to ketamine. Furthermore, 35% of those treated with ketamine maintained a response for at least 1 week. Perhaps not surprisingly, ketamine has been shown to directly act on the reward system by stimulating both immediate and delayed secretion of beta-endorphin in mice. In this study, ketamine increased secretion of beta-endorphin in some participants that was up to three times greater than their level at baseline (YaDaeu, Morelli & Billingsley, 2003). Although it may be unanticipated to see ketamine, also a street drug, be administered as a prescription, perhaps it would be effective to have controlled low dosages at a hospital where a doctor or nurse would administer the drug to a patient starting SSRI treatment during the first few weeks SSRIs are not effective. Theoretically, this could reduce the suicide risk for people with depression, providing that ketamine interacts with SSRIs in a safe manner. As suicide rates are highest in the immediate post-hospitalization period (Qin & Nordentoft, 2005), a fast acting drug, that is perhaps directly acting on the biological dysfunction seen in major depression, could have a large impact on suicide and depression.

For example, electro-convulsive shock therapy (ECT) has been shown to increase the level of beta-endorphin. In particular, ECT has been shown to activate the hypothalamic-pituitary-adrenal (HPA) axis, known to be involved in depression, resulting in a release of beta-endorphin (Young, Gruhnhaus, Haskett, Pande, Murphy-Weinberg, Akil, et al, 1991). The results from this study indicate that the effect of ECT on length of seizure has a relationship to the amount of beta-endorphin released during treatment, with decreasing duration of seizures leading to decreases in beta-endorphin. Perhaps with the advancement of research on opioids, ECT will no longer be needed to treat people with depression because the mechanisms through which ECT are effective may be uncovered and able to be administered in a more favorable manner: pharmaceutically. If ketamine and ECT do work through similar mechanisms, it would be interesting to conduct an experiment comparing the effects of ECT and ketamine as treatments for depression. Although evidence exists for the release of beta-endorphin during ECT, one study found that levels of endorphins were returned to baseline after administration of ECT [19]. In addition, there was an increase in endorphin level in all ten participants, one of whom was a control. This study provides evidence that endorphin level increases in response to stress, perhaps similarly to cortisol, rather than acting as a mechanism through which ECT is effective; however it is conceivable that perhaps even a brief period of experiencing pleasure as a result of beta-endorphin release after a serious episode of depression is enough for the person to learn that they can experience happiness. Perhaps a brief period of happiness is enough to instill a feeling of hope that perhaps the person will be able to experience happiness again in the future. Perhaps ECT could act as a jump start into a negative feedback loop (which reverses in direction away from depression and toward equilibrium).

Aside from pharmaceutical treatments and ECT, it is possible that increased beta-endorphin is also a mechanism behind psychological treatments. Although studies have shown that cognitive-behavioral therapy (CBT) is effective, studies aimed at determining mechanisms of change have pointed to behavioral activation (BA) as the component of treatment that accounts for the vast majority of symptom decreases (YaDaeu, Morelli, & Billingsley, 2003). An intervention strategy used in BA is increasing the client’s engagement in pleasurable activity and increasing mastery over situations. The activities are presented to clients in gradations, so that clients are gradually and reliably reinforced. In BA, clinicians often remark that “activity breeds activity” (Hayes, S., Follette, V. & Linehan, 2004), possibly indicating some evidence for a negative feedback loop. While no studies have measured beta-endorphin levels throughout the course of BA treatment, as activity increases, it is possible that beta-endorphin is involved in active engagement in behavior and learning positive associations with being active. If it is shown that beta-endorphin levels are increased through BA (either in frequency or strength of release), evidence would be provided for a potential biological mechanisms through which activity would breed more activity. BA may act through showing people from their own experience that they can engage in pleasurable activity. Perhaps similarly to ECT, engaging in behavioral activity after a depressive episode could have the propensity to instill a feeling of hope at the initiation of treatment from beta-endorphin release. In short, while BA is a psychological mechanism of change, perhaps a negative feedback loop can be a biological mechanism of change underlying the psychological experience of symptom improvement. Because BA has been a successful treatment in major depression, researchers are beginning to contemplate the role of behavioral inactivation as a cause of depression. Future studies examining how behavioral inactivity could result in major depression are needed. Like the positive feedback loop previously mentioned in regards to activity anorexia in rats, a positive feedback loop could be the cause for onset of major depression in humans while BA could result in a negative feedback loop that reverses the effects of depression.

Similar to BA, an evolving therapy called Positive Psychotherapy (PPT) has been developed by positive psychologists for people with depression (Seligmnan, Steen, Park, & Peterson, 2005). Amongst other techniques, PPT therapists employ discussion about self-identified positive qualities, having clients document why they are grateful, and having clients engage in pleasurable activities. While PPT is a new therapy, evidence exists that people treated with PPT have decreased depressive symptoms compared to a control condition receiving treatment as usual; however, PPT has not yet been compared to CBT or BA. Like BA, it can be argued that release of beta-endorphin can be involved in the formation and maintenance of associations between engaging in pleasurable activity and being rewarded, and therefore, be one potential mechanism through with the therapy is effective.

While BA and PPT include activity for treating depression, many studies have been conducted that directly examine the role of exercise on depression. Through various studies, perhaps the most supported link between psychological interventions and beta-endorphin release has been in the use of exercise to treat psychological disorders. For example, one recent study examined 38 inpatients with major depression who were randomly assigned to an exercise or non-exercise group. After 10 days, the reduction of depressive symptoms and the number of patients who experienced clinical response during inpatient treatment were significantly larger than the non-exercise group (Knubben, Reischies, Adli, Schlattmann, Bauer, & Dimeo, 2007) and the increase in beta-endorphin as a result of exercise is a well-documented phenomenon (Bender, Nagy, Barna, Tefner, Kadas & Geher, 2007). Like BA and PPT, the mechanism through which exercise is effective could possibly be beta-endorphin release or reward system activation. Because exercise most likely increases more beta-endorphin than BA or PPT, it would be interesting to see if exercise is a more potent form of treatment than BA or PPT.

While many studies were conducted to measure beta-endorphin and other endogenous opioids in people with depression during the 1980’s and 1990’s, research has waned in recent years. Perhaps studying neurotransmitters such as serotonin, norepinephrine, and dopamine was seen as having more promising clinical utility because administering opiates directly was not viewed as a viable option. Although opiates are dangerous because of the potential for abuse, recent findings on the strong and immediate effects of administering ketamine to people with depression may spark a renewed interest in the study of beta-endorphin and may prompt researchers to find a way to safely administer agents that act on beta-endorphin amongst other neurotransmitters. Although this paper attempts to link beta-endorphin to a wide variety of biological functions (the experience of pleasure and the formation of associations through conditioning), possible onset and maintenance of psychological disorders and unhealthy behaviors (alcoholism, opiate addition, smoking, anorexia nervosa, bulimia nervosa, and major depression), and treatments for major depression (SSRIs, ECT, BA, PPT, and exercise), many other biological mechanisms also contribute. Although examining only beta-endorphin as the sole mechanism in psychological disorders and treatment is vastly overly simplistic, the intention of this paper is to point to beta-endorphin as one possible contributing factor that is perhaps underutilized as tool in the treatment of major depression and understudied in relation to the onset of psychological disorders. Although, as evidence suggests that beta-endorphin may be involved in forming associations between behaviors and the feeling of pleasure as well as the experience of pleasure itself, future studies should focus on determining if beta-endorphin is directly related to the onset, maintenance, and treatment of major depression and other psychologically maladaptive behaviors.

Overall, it is interesting to consider that a deficiency of a substance like beta-endorphin could naturally lead to people engaging in behaviors that increase beta-endorphin such as drug use or result in the experience of depression. It is also worth noting that various forms of treatments for depression such as ECT, medication, psychological therapy, and exercise therapy all may increase beta-endorphin as well, implicating the treatment work through correcting a deficiency.


Avena, N.M., Rada, P., & Hoebel, B.G. (2008). Evidence for sugar addiction: behavioral and neurochemical effects of intermittent, excessive sugar intake. Nueroscience & Biobehavioral Reviews, 32(1), 20-39.

Bender, T., Nagy, G., Barna, I., Tefner, I., Kadas, E. & Geher, P. (2007). The effect of physical therapy on beta-endorphin levels. Eur J Appl Physiol, 100, 371-382.

Bond, C., LaForge, S., Tian, M., Melia, D., Zhang, S., Borg, L., et al. (1998). Single-nucleotide polymorphism in the human mu opiod receptor gene alters beta-endorphin binding and activity: Possible implications for opiate addiction. Proc Natl Acad Sc, 95, 9608-9613.

Collier, G. & Levitsky, D. (1968). Operant running as a function of deprivation and effort. J Comp Physiol Psychol, 66, 522-523.

Davis, C & Claridge, G. (1998). The eating disorders as addiction: psychobiological perspective. Addictive Behaviors, 23, 463-475.

Dai, X., Thavundayil, J., & Glanoulakis, C. (2005). Difference in the peripheral levels of beta-endorphin in response to alcohol and stress as a function of alcohol dependence and family history of alcoholism. Alcoholism: Clinical and Experimental Research, 29, 1965-1975.

del Arbol, J. L., Raya, M., Munoz, J., Ojeda, L., Lopez-Cascales, A., Irles, J., Miranda, M., et al. (2007). Plasma concentrations of beta-endorphin in smokers who consume different numbers of cigarettes per day. Pharmacol Biochem Behav, 67, 25-28.

Dimidjian, S., Hollon, S., Dobson, K., Schmaling, K., & Kohlenberg, R., et al. (2006). Randomized control trial of behavioral activation, cognitive therapy, and antidepressant medication in the acute treatment of adults with major depression. Journal of Consulting and Clinical Psychology, 74, 658-670.

Favazza, A. & Conteiro, K. (1988). The plight of chronic self-mutilators. Community and Mental Health Journals. 24(1), 22-30.

George, S. R., & Van Loon, G. R. (1982). Beta-Endorphin alters dopamine uptake by the dopamine neurons of the hypothalamus and striatum. Brain Res, 248, 293-303.

Ghadirian, A., Gianoulakis, C., & Nair, A. (1988). The effect of electroconvulsive therapy on endorphins in depression. Biol Psychiatry, 23, 459-464.

Hayes, S., Follette, V. & Linehan, M. (2004). Mindfulness and Acceptance: Expanding the cognitive-behavioral tradition. New York: Guilford Press, 160.

Knubben, K., Reischies, F., Adli, M., Schlattmann, P., Bauer, M., & Dimeo, F. (2007). A randomized, controlled study on the effects of a short-term endurance training programme in patients with major depression. Br J Sports Med, 41, 29-33.

Lee, Y., Joe, K., Sohn, I., Na., C., Kee, B., & Chae, S. C. (2005). hanges of smoking behavior and serum adrenocorticotropic hormone, cortisol, prolactin, and andogenous opiod levels in nicotine dependence after naltrexone treatment. Prog Neuropsychopharmacol Biol Psychiatry, 29, 639-647.

Maes, M., Meltzer, H., Cosyns, P., Calabrese, J., D’hondt, P., & Blockx, P. (1994). Adrenocorticotropic hormone, beta-endorphin and cortisol responses to CRH in unipolar depressed patients pretreated with dexamethasone. Prog Neuro-psychopharmacol & boil psychiat, 18, 1273-1292.

Qin, P & Nordentoft, M. Suicide risk in relation to psychiatric hospitalization: evidence based on longitudinal registers. Arch Gen Psychiatry, 62, 427-432.

Rada, P., Colasante, C., Skirzewski, M., Hernandez, L., & Hoebel, B. (2006). Behavioral depression in the swim test causes biphasic, long-lasting change in accumbens acetylcholine release, with partial compensation by acetylcholinesterase and muscarinic-1 receptors. Neuroscience, 141, 67-76.

Salmone, J. & Correa, M. (2002). Motivational views of reinforcement: implications for understanding the behavioral functions of nucleus accumbens dopamine. Behavioral Brain Research, 137, 3-25.

Sandman, C.A. Barron, J.K., Chicz-DeMat, A. & DeMat, E.M. (1990). Plasma-B-endorphin levels in patients with self-injurious behavior and stereotypy. American Journal of Mental Retardation, 95(1), 84-92.

Seligmnan, M., Steen, T., Park, N., & Peterson, C. (2005). Positive Psychology Progress: empirical validation of interventions. American Psychologist, 60, 410-421.

Taylor, A.H., Ussher, M.H., Faulkner, G. (2007). The acute effects of exercise on cigarette cravings, withdrawal symptoms, affect, and smoking behaviour: a systematic review. Addiction, 102(4), 534-43.

Rothman, R.B. Blough, B.E., & Baumann, M.H. (2008). Dopamine/serotonin releasers as medications for stimulant addictions. Progress in Brain Research, 172, 385;406.

Wallenstein, M.B. & Nock, M.K. (2007). Physical exercise as a treatment for non-suicidal self-injury: evidence from a single-case study. American Journal of Psychiatry, 164(2),350-351.

YaDaeu, J., Morelli, C., & Billingsley, J. (2003). Ketamine stimulates secretion of beta-endorphin from a mouse pituitary cell line. Regional Anesthesia and Pain Medicine, 28, 12-16.

Young, E., Lopez, J., Murphy-Weinberg, V., Watson, S., & Akil, H. (2000). Hormonal evidence for altered responsiveness to social stress in major depression. Neuropsychopharmacology, 23, 411-418.

Young, E.A., Grunhaus, L., Haskett, R.F., Pande, A., Murphy-Weinberg, V., Akil, H., et al. (1991). Heterogeneity in the beta-endorphin immunoreactivity response to electroconvulsive therapy. Arch Gen Psychiatry, 48, 534-539.

Zalewska-Kaszubska, J. & Czarnecka, E. (2005). Deficit in beta-endorphin peptide and tendency to alcohol abuse. Peptides, 26, 701-705.

Zangen, A. & Shalev, U. (2003). Nuclues accumbens beta-endorphin levels are not elevated by brain stimulation reward but do increase with extinction. European Journal of Neuroscience, 17, 1067-1072.

Zangen, A., Nakash, R., & Yadid, G. (1999). Serotonin-mediated increases in the extracellular levels of beta-endorphin in the arcuate nucleus and the nucleus accumbens: a microdialysis study. J Neurochem, 73, 2569-2574.

Zarante, C., Singh, J., Carlson, P., Butsche, N., Ameli, R., Luckenbaugh, et al. (2006). A randomized trial of an N-Methyl-D-aspartate Antagonist in treatment-resistant major depression. Arch Gen Psychiatry, 6, 856- 864.


Tara said...

I have noticed that this blog post, while the oldest, consistently gets the most hits out of any on the entire blog each day. I have also noticed that it is the second link that comes up in Google when searching "endogenous reward system." If anyone would like to discuss this particular article with me, please feel free to leave a comment right below this one! If not, I would be more than happy to discuss this over email.

Craig Marker said...

Of the many points that struck me in this article, I like to point out two that were really fascinating.
First, I liked your discussion of how people with reward system dysfunctions could seek out stimuli that are both maladaptive, but exerting an effect on the reward system to feel better. The second point was on the mechanism of behavioral activation, in which you said: "as activity increases, it is possible that beta-endorphin is involved in active engagement in behavior and learning positive associations with being active." That sounds like a great future study.
Overall, a great article.

Tara Deliberto said...

Thanks, Dr. Marker! Your comment is really appreciated.

First - Yes, it makes sense to me that people engage in specific types of behaviors for specific physiological reasons.

Second - it also makes sense to me that endorphin increase could be the biological mechanism of change in behavioral activation.

One of my favorite pieces of this blog post relates to the fact that dopamine does not mediate behaviors that are naturalistically reinforcing.

I also find it interesting that endorphin may play a role in mediating serotonin.

Thanks again reading!