The OCD brain strikes again

Who would have ever thought that Cameron Diaz, the cute voice actress behind Fiona from Shrek, actually suffers from obsessive-compulsive disorder (OCD)? In fact, she opens doors with her elbows, refusing to touch the germ-ridden doorknobs with her bare hands!

Although OCD is not a disease that “kills”, it can be a distressing experience for individuals. A classic example is someone who has an obsession with cleanliness. Driven by a fear of germs, an OCD person may wash his or her hands over and over again. Or, like Cameron Diaz, he or she may polish the doorknobs so many times that the paint wears off. These frequent upsetting thoughts (obsessions) compel a person to repeat certain rituals. While washing hands provides temporary relief from the anxiety caused by germaphobia, it does not completely erase the unreasonable fear of germs¹. As a result, the germaphobe will continue to wash his or her hands to put an end to the endless obsessions, leading up to a vicious cycle that controls the lives of OCD individuals.

Now this doesn’t mean that if you are afraid of germs you automatically have OCD. Anyone who engages in good hygienic practice will dislike germs. It does not hurt to wash our hands every once in a while to stay clean, since at the end of the day, we are still in control of our own lives. However, the usual hygienic practice turns into OCD when washing hands disrupts and takes control of the individual. If the OCD person does not wash his or her hands, the anxiety of not being able to satisfy the urge causes extreme stress.

OCD does not have to manifest itself only through hand washing. Besides a fear of germs, OCD can also show up through other means such as obsessions with order and symmetry.

A major problem in mental health issues such as OCD is the difficulty in quantifying the disease. Psychiatry, the specialty that deals with mental disorders, has no objective measurement (such as a blood test) that can be used to diagnose and treat patients. Instead, psychiatrists rely on subjective personal accounts given by their patients, which may be unreliable.

The lack of a biomarker reflects a more fundamental issue: no one knows what goes wrong inside the brain. While it is safe to say that OCD and other mental health disorders result from neurons not working correctly, the tall hurdle lies in figuring out which out of the millions of neurons or neurotransmitters are implicated in disease conditions. Without establishing the molecular reasoning behind behavior, it is impossible to develop diagnostic assays and effective cures.

Rigorous research in the past three decades has begun to shed light on the pathophysiology of OCD, linking clinical observations with specific neural pathways. Early anatomical studies demonstrated the existence of specialized neural loops that connect the basal ganglia to the cerebral cortex, resulting in functional behavior such as initiation and control of movement². The importance of basal ganglia circuitry in OCD was first identified almost a hundred years ago by Constantin Von Economo, a Romanian neurologist³.  Sometimes during a viral infection, the virus causes inflammation of the brain. Severe cases can result in permanent brain damage. Von Economo described patients who suffered from basal ganglia dysfunction due to influenza infections. Interestingly, the patients displayed compulsive behavior, “having to” act, while “not wanting to”. This description fits with current OCD individuals, who realize that their behavior is unreasonable, yet cannot put a stop to their ritual-like habits.

As more evidence piled on, scientists have narrowed it down to the cortico-striato-thalamo-cortical (CSTC) circuit. This loop consists of the prefrontal cortex, the striatum (part of the basal ganglia), and the thalamus⁴. In simple terms, the thalamus receives sensory information from the outside world and relays it to the prefrontal cortex. The prefrontal cortex then processes this information and makes complex decisions on how the human body should respond. While sensory information from the thalamus is crucial, too much thalamic traffic can overwhelm the prefrontal cortex. Thus, the prefrontal cortex can inhibit the thalamus via the striatum, preventing overflow of sensory information. Essentially, the CSTC is the engine for numerous functions, depending on which area of the prefrontal cortex is involved and where on the striatum the cortical neurons project to. For instance, in cognitive tasks such as paying attention and maintaining goals, neurons in the dorsolateral prefrontal cortex project to the caudate region of the striatum, whereas regulation of motor activity involves the prefrontal motor cortex⁴.

Figure 1: The CSTC circuit. It has been postulated that there are many different CSTC pathways depending on which region of the cerebral cortex is utilized. Figure adapted from Stahl, 2008.

The proposed CSTC loop that is implicated in OCD involves the orbitofrontal cortex. Evidence from human and animal studies suggests that the orbitofrontal cortex plays a role in emotional and motivational aspects of behavior⁵. Subsequent works show that the orbitofrontal cortex is also involved in inhibitory responses⁵. Thus, disruption of the orbifrontal cortex may lead to a deficit in repressing one’s own behavior, resulting in the inability of OCD individuals to control their compulsive cycles. However, given that the orbitofrontal cortex functions in a pathway with other brain regions, a dysregulation anywhere in the CSTC loop can result in OCD. The CSTC loop is a combination of excitatory (glutamatergic) and inhibitory (GABA-ergic) neurons. As the names suggest, excitatory neurons release glutamate to activate their targets, while inhibitory neurons release gamma-Aminobutyric acid (GABA) to suppress downstream targets. The normal function of the CSTC rests upon the balance of activity between these neurons. When this balance is disrupted, OCD occurs.

Figure 2: The CSTC model of OCD. The CSTC model of OCD. The red lines indicate inhibition of targets and the green arrows show excitation of targets. This loop involving the orbitofrontal cortex is thought to regulate compulsivity, and its dysregulation leads to ritual-like behaviors in OCD. Excitatory cortical neurons from the orbitofrontal cortex first project onto inhibitory medium spiny neurons (MSN) of the striatum. The remaining steps of the pathway mainly involve neurons deactivating each other, leading to modulation of thalamic activity (I’m going to spare you from the technical terms so you don’t get bored). The direct pathway increases thalamic activity, while the indirect pathway decreases thalamic output⁶. Both pathways involve the globus pallidus as well, an important component of the basal ganglia. The external and internal parts of the globus pallidus are called GPe and GPi, respectively. Together with the pars reticula of the substantia nigra (SNr), the GPi neurons directly project onto thalamic neurons. Lastly, other neurons the subthalamic nucleus (STN) and the GPe also have functions in the indirect pathway. A major hypothesis holds that when this delicate balance of inhibition vs excitation is disrupted, OCD will occur. The cause of the imbalance is yet to be discovered. Figure modified from Kalra and Swedo, 2009.

Indeed, neuroimaging studies have provided evidence for the CSTC model. In 1987, Baxter and others used positron emission tomography (PET) to study glucose metabolism in the brains of 14 OCD patients⁷ The amount of glucose metabolized reflected the function of different regions in the brain. Compared to normal people, OCD patients displayed elevated glucose activity in the orbital gyri and caudate nucleus, indicating hyperactivity of the orbitofrontal cortex and striatum in conditions of OCD. Furthermore, successful treatments using selective serotonin reuptake inhibitors have shown to down-regulate such hyperactivity, demonstrating the relevance of orbitofrontal cortex involvement in OCD⁸ Subsequent reports utilizing functional magnetic resonance imaging (fMRI) techniques have also demonstrated orbitofrontal dysfunction. Instead of relying on glucose metabolism, fMRI measures brain activity via cerebral blood flow. Five years ago, Chamberlain and others from Cambridge University showed that OCD individuals exhibited impaired orbitofrontal function in a task called reversal learning⁹.

Reversal learning is a critical biological process that allows us to adapt our behavior to changes in circumstances. Playing video games is a perfect example of reversal learning. When a young dude plays the new Call of Duty game, he gets quite a kick from shooting enemies and unlocking new levels. Thus, the brain associates playing Call of Duty with the reward of entertainment. As his mind is lost in the 6-hour video game sessions, the teenager forgets to study for an upcoming test. As a result, he receives a failing grade and subsequent yellings from his mother. The brain then updates itself: video games are now associated with a lackluster performance in school. The teenager learns to repress his own urge to play video games, an activity previously linked with a reward, but is now connected with a punishment. The orbitofrontal cortex is necessary in order to suppress these previously-important, but now inappropriate responses¹⁰. Perhaps, in the cases of OCD, a dysfunctional orbitofrontal cortex results in the inability to repress an urge (such as washing hands), leading to a never-ending cycle of repetitive behavior.

Given the body’s ability to maintain homeostasis, scientists have wondered: does hyperactivity of the orbitofrontal-striatal loop directly lead to OCD? Whenever there is an imbalance, the body can respond in numerous ways via compensatory mechanisms. Since OCD has been proposed to result from an imbalance between inhibitory and excitatory signals, elevated activity of the orbitofrontal-striatal circuit may simply be the brain’s way of counteracting against abnormal activities in other unknown areas. A really cool experiment to do would be to somehow activate only the neurons in the orbitofrontal cortex and striatum. If selective activation of those neurons leads to repetitive behavior, then we can say with confidence that the hyperactivity of the orbitofrontal-striatal loop does indeed contribute to OCD pathogenesis.

Optogenetics is a new handy tool in neuroscience that can shine a light on this mystery regarding whether hyperactivity of the orbitofrontal-striatal circuit is the primary cause of OCD. Essentially, optogenetics allows scientists to control specific neurons by turning on a blue light. You may find it surprising, but this fancy-sounding technique originated from photosynthetic green algae. As green algae make their food from photosynthesis, they will tend to move towards areas of brighter light. Light-sensitive ion channels known as rhodopsin mediate this migration from low- to high-intensity light¹¹. When light shines onto the rhodopsins, they open up themselves and allow ions to pass through, leading to a cascade of downstream chemical reactions that control the algae’s movement.

Neural activation operates in very similar ways. When a neuron gets turned ON, ions flood inside the cell through opened channels. Thus, controlling neural activity could be done by genetically engineering the desired neurons to express light-sensitive algal rhopdosins. In 2005, the Deisseroth lab at Stanford University utilized channelrhodopsin 2 (ChR2) in order to manipulate cultured rat neurons¹². By exposing the ChR2-expressing neurons to blue light, control of neural activity was narrowed down to milliseconds. Because of the temporal resolution offered by optogenetics, hundreds of labs have gotten their hands onto this technology in order to elucidate intricate neural circuits driving human behavior.

Since OCD is a disease of defective neural circuitry, researchers have taken advantage of optogenetics to study the orbitofrontal-striatal circuit implicated in compulsive behavior. In order to target this specific neural pathway, researchers at Columbia University injected an adeno-associated virus (AAV) into the orbitofrontal cortex of the mouse. Since the AAV carried a gene that encodes ChR2, cortical neurons infected with the virus will also carry this gene and express ChR2. The researchers then implanted opto-electrodes that can shine blue light inside the brain, allowing them to exert control over neural activities. The cortical neurons directly connect to striatal neurons. Thus, when the ChR2-expressing cortical neurons are turned ON by a blue light, striatal neurons will also become activated. Repeated activation of the oribitofrontal-striatal loop over several days by optogenetics means led to an increase in OCD-like grooming behavior in mice, indicating that this specific circuitry is indeed a contributor to OCD behavior¹³

Now that we are more confident that hyperactivity of the orbitofrontal-striatal circuit leads to OCD, we can ask the next question: if we decrease the activity, will we able to suppress compulsive behavior? In 2007, the Feng lab generated a mouse model of OCD that exhibited defects in the orbitofrontal-striatal circuit¹⁴. These mice did not have Sapap3, a protein that is involved in regulating neural signaling between cortical and striatal neurons. As a result, they groom themselves so much that they develop lesions on the head, neck, and snout regions!

Using the Sapap3 mutant mice as a mouse model of OCD, researchers at MIT utilized optogenetics to see whether compulsive behavior can be rescued by targeting specific neurons¹⁵. When the scientists measured the firing rates of the MSN in the striatum, they discovered that the baseline activity levels of these neurons were significantly elevated in the Sapap3 mutants. A source of MSN inhibition came from GABAergic fast-spiking interneurons (FSIs), which are also located in the striatum. Similar to the MSN, the FSI receives controlling inputs from cortical neurons. When the FSI is activated, it will inhibit the activity of the MSN (wow, an inhibitory neuron inhibiting another inhibitory neuron. NEURO-ception!).

By optogenetically activating the FSI, the normally-hyperactive MSN was suppressed. Strikingly, obsessive grooming exhibited by the Sapap3 mutants was rescued following MSN inhibition, indicating that decreasing activity levels in the orbitofrontal-striatal loop can alleviate OCD behavior. While previous studies have correlated successful treatments of OCD with down-regulation of implicated neural activities, this report is the first to show a direct relationship between suppression of oribitofrontal-striatal activity and amelioration of compulsive behavior¹⁶.

Figure 3: Hyperactivity of the pathway between the orbitofrontal cortex and striatum led to obsessive grooming in normal mice. Repression of this pathway by down-regulating MSN activity led to alleviation of compulsive behavior in an OCD mouse obsessed with self-grooming. Figure made by Duy Phan, 2013 with cartoon drawings taken from Chen et al., 2013 and images taken from Fudan-Lux team.

Although we have identified neural circuits that may be responsible for OCD thanks to optogenetics, we are still far from a complete understanding. First, we still do not understand how OCD arises in individuals. One hypothesis is that the microcircuits necessary for fine-tuning the inhibitory responses becomes dysfunctional in OCD conditions. As stated before, the FSI is an interneuron that can function to down-regulate MSN activity. Recent evidence shows that the FSI expresses a calcium-binding protein called parvalbumin (PV) that is required for neurotransmission¹⁷. Due to the importance of PV, it is likely that FSIs lacking this protein will not be able be able to communicate properly within their neural circuits. Strikingly, the OCD mice displayed significantly fewer PV-positive striatal neurons than wild types. There are two ways that we can interpret the reduction of PV-positive neurons: either the FSIs have undergone cellular death or they could have changed identity to no longer produce PV. Obviously, a cell that is dead will no longer express PV, thus explaining why the researchers found fewer PV-positive cells. The second explanation is equally interesting. The FSI itself might still exist, but they are functionally defective without PV. Whether it is cellular death or cellular dysfunction, the consequence is the same: an FSI that is not working will not be able to inhibit its target, the MSN. Perhaps, without functional FSI, individuals lose the critical inhibitory mechanisms to repress certain urges, resulting in compulsive behavior. Do humans with OCD have changes in PV-expressing cells? While no human study has looked at PV in the context of OCD, other reports have found decreased number of PV-positive neurons in patients with Tourette syndrome¹⁸. Like OCD individuals, people suffering from Tourette syndrome also display repetitive behavior that they cannot control. It is possible that there is a common mechanism underlying a variety of repetitive actions, but more research is needed to backup this claim.

The second question we face in the next step of OCD research is: which type of MSN is hyperactive in OCD?  If we design new OCD treatments to target the MSN, we must delineate the precise population of MSN that is affected. There are two distinct different populations of MSNs: striatonigral (SN) neurons that participate in the direct pathway and striatopallidal (SP) neurons that participate in the indirect pathway. Besides their targets, SN and SP neurons also differ in their chemical and physiological properties¹⁹. In particular, SN and SP neurons express different dopamine receptors. Although recent evidence has shown that MSNs have elevated activities in OCD mice, are both types of MSN affected or just one of the two types? We may be able to answer this question by investigating the thalamus, as the two types of MSN have opposite effects on thalamic activity. An analysis of 16 Brazilian OCD patients showed elevated cerebral blood flow in the left and right thalamus, suggesting that the thalamus is hyperactive²⁰. Hyperactivity of the thalamus indicates that MSNs participating in the direct pathway (SN) are more likely to be the ones that are affected. We know from the OCD mice that MSNs exhibit abnormally high activity levels. For an activated MSN to increase thalamic activity seen from human neuroimaging studies, the MSN must have participated in the direct pathway. Therefore, we may hypothesize that the SN neurons are the culprit of OCD, or at least contribute to manifestation of the disease. Future optogenetic studies that manipulate the SN neurons will provide further evidence for this hypothesis. Additionally, it will be critical to determine whether other circuits and neurons are involved as well.

Figure 4: The direct pathway dysfunction hypothesis of OCD (update from previous orbitofrontal-striatal models). Strong evidence showing that OCD has a genetic basis indicates the possible existence of an OCD-related gene¹⁶. This gene may have normal functions in constitutive maintenance of the FSI. When a mutation occurs, rendering the gene defective, FSI can no longer perform its inhibitory roles. Because of a reduction in inhibitory mechanisms, SN neurons become hyperactive and increase thalamic activity via the direct pathway. An excess of glutamatergic signals from the thalamus causes further activation of the SN neurons, resulting in an endless positive feedback loop. This never-ending pathway manifests itself in compulsive behavior of OCD individuals. Figure made by Duy Phan, 2013, with an image taken from Kemper, 2012.

The third and greatest challenge that we face in the future of OCD is translating the discoveries made on the laboratory bench top to treatments in the clinic. Now that we have significant progress in delineating the neural circuitry behind OCD, how will we target specific neurons in this pathway? At this stage, optogenetics is still too far-fetched to be safely applied in humans (walking around with a funky opto-electrode sounds rather dangerous). Potentially, we can exploit the chemical and physiological differences between neurons in order manufacture drugs that target only the neurons that we want. However, many neurons in the brain have similar properties, such as the expression of identical receptors. Given that the neurons are so similar, how will we target the correct cells without disrupting the function of surrounding healthy neurons? More groundwork on the physiology of neurons is needed to look for differences between subtypes of neurons that we can take advantage of.

Despite these enormous challenges, we must applaud the neuroscientists for advancing our understanding of the organic basis behind behavior. Medical fields dealing with mental health diseases have evoked a sense of the “unknown” and “incomprehensible”, since we have always struggled with a greater understanding how the brain works. Now that scientists are un-tangling the neuronal mess in the brain, we are gaining more and more understanding of the neural circuitry behind activities that make us humans, such as higher-order decision making. If we can uncover the mystery behind the brain, we will no longer fear the “incomprehensible” mental disorders.

1. What is Obsessive-Compulsive Disorder (OCD)? (n.d.). Retrieved from National Institute of Mental Health website: http://www.nimh.nih.gov [↩]
2. Alexander, G. E., DeLong, M. R., & Strick, P. L. (1986). Parallel Organization of Functionally Segregated Circuits Linking Basal Ganglia and Cortex. Annual Reviews Neuroscience. [↩]
3. Kalra, S., & Swedo, S. (2009). Children with obsessive-compulsive disorder: are they just “little adults”? Journal of Clinical Investigation. [↩]
4. Stahl, S. (2008). Stahl’s Essential Psychopharmacology: Neuroscientific Basis and Practical Applications (3rd ed.). Cambridge University Press. [↩] [↩]
5. Menzies, L., Chamberlain, S., Laird, A., Thelen, S., Sahakian, B., & Bullmore, E. (2008). Integrating evidence from neuroimaging and neurophychological studies of obsessive-compulsive disorder: The orbitofronto-striatal model revisited. Neuroscience and Biobehavioral Reviews. [↩] [↩]
6. Stocco, A., Lebiere, C., & Anderson, J. (2011). Conditional Routing of Information to the Cortex: A Model of the Basal Ganglia’s Role in Cognitive Coordination. Psychology Review. [↩]
7. Baxter, Phelphs, Mazziotta, Guze, Schwartz, & Selin. (1987). Local cerebral glucose metabolic rates in obsessive-compulsive disorder. A comparison with rates in unipolar depression and in normal controls. Archives of General Psychiatry. [↩]
8. Menzies, Chamberlain, Laird, Thelen, Sahakian, & Bullmore. (2008). Integrating evidence from neuroimaging and neuropsychological studies of obsessive-compulsive disorder: The orbitofrontal-striatal model revisited. Neuroscience and Biobehavioral Reviews. [↩]
9. Chamberlain, Menzies, Hampshire, Suckling, Fineberg, & Aitken. (2008). Orbitofrontal Dysfunction in Patients with Obsessive-Compulsive Disorders and Their Unaffected Relatives. Science. [↩]
10. (Menzies, Chamberlain, Laird, Thelen, Sahakian, & Bullmore. (2008). Integrating evidence from neuroimaging and neuropsychological studies of obsessive-compulsive disorder: The orbitofrontal-striatal model revisited. Neuroscience and Biobehavioral Reviews. [↩]
11. Sineshchekov, Jung, & Spudich. (2002). Two rhodopsins mediate phototaxis to low- and high-intensity light in Chlamydomonas reinhardtii. Proceedings of the National Academy of Sciences. [↩]
12. Boyden, Zhang, Bamberg, Nagel, & Deisseroth. (2005). Millisecond-timescale, genetically targeted optical control of neural activity. Nature Neuroscience. [↩]
13. Ahmari, Spellman, Douglass, Kheirbek, Simpson, Deisseroth, . . . Hen. (2013). Repeated Cortico-Striatal Stimulation Generates Persistent OCD-Like Behavior. Science. [↩]
14. Welch, Lu, Rodriguiz, Trotta, Peca, & Ding. (2007). Cortico-striatal synaptic defects and OCD-like behaviours in Sapap3-mutant mice. Nature. [↩]
15. Burguiere, Monteiro, Feng, & Graybiel. (2013). Optogenetic Stimulation of Lateral Orbitofrontal-Striatal Pathway Suppresses Compulsive Behaviors. Science. [↩]
16. Menzies, L., Chamberlain, S., Laird, A., Thelen, S., Sahakian, B., & Bullmore, E. (2008). Integrating evidence from neuroimaging and neurophychological studies of obsessive-compulsive disorder: The orbitofronto-striatal model revisited. Neuroscience and Biobehavioral Reviews. [↩] [↩]
17. Orduz, Bischop, Schwaller, Schiffmann, & Gall. (2013). Parvalbumin tunes spike-timing and efferent short-term plasticity in striatal fast spiking interneurons. Journal of Physiology. [↩]
18. Kataoka, Kalanithi, Grantz, Schwartz, Saper, & Leckman. (2010). Decreased Number of Parvalbumin and Cholinergic Interneurons in the Striatum of Individuals with Tourette Syndrome. Journal of Comparative Neurology. [↩]
19. Stocco, A., Lebiere, C., & Anderson, J. (2011). Conditional Routing of Information to the Cortex: A Model of the Basal Ganglia’s Role in Cognitive Coordination. Psychology Review. [↩]
20. Lacerda, Dalgalarrondo, Caetano, Carmargo, Etchebehere, & Soares. (2003). Elevated thalamic and prefrontal regional cerebral blood flow in obsessive–compulsive disorder: a SPECT study. Psychiatry Research: Neuroimaging. [↩]