Neurobiology of schizophrenia

The symptoms of schizophrenia can be divided into two classes: positive and negative. Positive symptoms refer to distortions or exaggerations of normal functions, such as delusions, disordered thought, and hallucinations. In contrast, negative symptoms refer to the absence of certain normal behaviors and emotions: flat affect, apathetic social withdrawal, and poverty of speech.

As noted by Paola, there do seem to be genetic factors as well as environmental factors that contribute to the development of schizophrenia. Based on the neuroanatomical data, as well as epidemiological data, there is evidence that schizophrenia may be related to a disruption in the normal developmental process. It is likely that the disorder, which is heterogeneous in nature, may have a number of etiologies.

So what is the underlying neuropathology associated with schizophrenia?

Paola discussed how many CNS regions, including the frontal and temporal cortex, hippocampus, and cerebellum, appear to be affected anatomically and/or functionally in schizophrenia. Imaging studies suggest that some patients with schizophrenia have one or more of three major anatomical abnormalities.

1- reduced blood flow in frontal lobes

2- thinner medial temporal lobe cortex, frontal cortex, and smaller anterior portion of the hippocampus (especially on the left side, consistent with a defect in memory)

3- Lateral and third ventricles are enlarged, and there is a widening of the sulci, especially in the temporal and frontal lobes

Most patients with schizophrenia exhibit neurological symptoms that suggest the presence of brain damage, including catatonia, unusually rates of blinking, paroxysmal bursts of jerky eye movements, poor visual pursuit of a smoothly moving object, inability to move the eyes without moving the head, poor pupillary light reactions, and speech arrest.

So, alterations in the prefrontal cortex, hippocampus, and globus pallidus, in particular, may be part of a cognitive system that is impaired in schizophrenia. Consistent with this notion, the negative symptoms of schizophrenia include a loss of executive functions such as planning and working memory (which involves prefrontal association areas).

DEVELOPMENTAL INJURY:

Paola spent some time discussing the hypothesis that schizophrenia may be related to alterations in CNS development due to genetic or environmental factors. Several pieces of evidence suggest a developmental disturbance. Altered densities and disorganization of neurons are found in the white matter immediately below layer VI of the cortex. In patients with schizophrenia, the number of neurons in the superficial white matter of both the prefrontal and temporal lobe cortices is significantly reduced, whereas the number in the deeper white matter is significantly greater compared to normal subjects. This suggests that the neurons underwent abnormal migration during development. Such a defect could lead to abnormal patterns of cortical connections in the frontal and temporal lobes. Similarly, pyramidal cells in the hippocampus are disorganized, another indication that the normal developmental process was disrupted.

What causes this? Developmental abnormalities may be related to alterations in genes involved in neurogenesis, migration and differentiation. In addition, epidemiological studies indicate that several environmental factors are associated with schizophrenia. For example, many studies have shown that people born during the late winter and early spring (in the Northern hemisphere) are more likely to develop schizophrenia, a phenomenon known as the seasonality effect. The converse is true in the southern hemisphere. Why? Both may be related to winter flu season occurring during late pregnancy. Others have found that increased rates of influenza or other viral infections, stress, famine, and Rh incompatibility during the 2^nd and 3^rd trimesters is associated with increased risk for schizophrenia, as are birth complications. It might be that the combination of certain genetic factors (polygenic, currently not yet identified) and developmental insults contribute to the neuropathology associated with schizophrenia.

The anatomical changes are associated with neurochemical alterations.

The general view for some time has been that the positive symptoms are related to overactivity of the mesolimbic dopamine pathway, whereas the negative symptoms are related to alterations in the dorsolateral prefrontal cortex and other neurotransmitter systems.

Before going into the dopamine hypothesis, recall the dopaminergic pathways.

There is the:

(1) mesolimbic pathway which projects from the brainstem (midbrain, (ventral tegmental area, or VTA)) to limbic areas, including the nucleus accumbens, amygdala, hippocampus, anterior cingulate cortex, entorhinal cortex and ventral striatum

(2) nigrostriatal pathway, which projects from the substantia nigra to the basal ganglia

(3) mesocortical pathway, which projects from the brainstem (midbrain) areas to the cortex, particularly the prefrontal cortex which is involved in the temporal organization of behavior, in motivation, planning, attention and social behavior.

Much of our understanding of the neurochemical basis for schizophrenia comes from the development of antipsychotic drugs.

ANTIPSYCHOTIC (Neuroleptic) DRUGS

The first useful treatment for schizophrenia was chlorpromazine. It was identified by the French neurosurgeon Henri Laborit, who thought that anxiety experienced by patients before surgery led to the release of histamine from mast cells (which are found in the brain) and that the histamines might contribute to the undesirable side effects of anesthesia. He looked for a drug to block the release of histamine and through trial and error found chlorpromazine, which successfully calmed the patients. In 1951, John Delay and Pierre Deniker found that a high dosage of chlorpromazine calmed highly agitated and aggressive patients who had schizophrenia or manic depressive symptoms.

Chlorpromazine and other drugs of the phenothiazene class, the butryophenones (Haldol), and the thioxanthenes are all considered typical antipsychotics. There is a newer class of drugs, the atypical antipsychotics (i.e. clozapine), which are also effective and produce fewer side effects.

How do they produce their actions? As with many therapeutic drugs, the efficacy of the drugs was established before its mechanism of action was understood. Arvid Carlsson found that the effective antipsychotics block dopamine receptors. The typical antipsychotic drugs have a high affinity for D2 receptors, which are therefore thought to be one of the major sites of the therapeutic action of these drugs. In fact, the clinical potency of the typical antipsychotic agents is closely correlated with their affinity for the D2 receptor. (This does not mean, however, that they do not affect other receptors, including cholinergic, adrenergic, and histaminic. Typical of most therapeutic drugs, their sites of action may be many.)

The therapeutic effects of typical antispychotics are believed to be related to reductions in mesolimbic DA functioning, but because D2 receptors are found in all of these dopaminergic systems, side effects may occur. Recall that reductions in dopaminergic functioning are the cause of Parkinson’s disease. Typical antipsychotic drugs often produce a syndrome resembling parkinsonism, due to a reduction in dopaminergic activity in the nigrostriatal pathway. In fact, the term “neuroleptic” was coined by the first clinicians who observed the psychomotor slowing and emotional quieting associated with antipsychotic drugs. What is more, if the individual uses the typical antipsychotic for a prolonged period, it can lead to tardive dyskinesia, characterized by uncontrollable, or hyperkinetic, movements, especially of the face, neck and extremities. How does a DA antagonist lead to symptoms characteristic of a chorea? If the DA receptors are blocked for prolonged periods, it can lead to receptor supersensitivity, or up-regulation. The receptors try to compensate for the reduction in DA activation and overcompensate. This is the current, long-held hypothesis. However, there are cases in which the tardive dyskinesia may be permanent, so this process may be more complicated.

It appears that the antagonism of dopaminergic activity in the mesolimbic system contributes to the therapeutic effects of antipsychotic drugs on the positive symptoms, but not really the negative symptoms. There is a lot of debate, but some argue that it is possible that neuroleptics may exacerbate negative symptoms. In fact, the first clinicians that coined the term neuroleptic described the neuroleptic-induced deficit syndrome. So, typical antipsychotics may improve positive symptoms but may have no effect or may even exacerbate negative symptoms and may produce undesirable motor dysfunction.

Attempts were made to develop effective drugs that did not produce the adverse side effects—thus, the development of atypical antipsychotic drugs. The atypical antipsychotic drugs do not produce the same motor symptoms as typical antipsychotic drugs, because they have a preference for the mesolimbic system over the nigrostriatal system. How? There are many subtypes of DA receptors and the concentration of these subtypes differs in various neuronal regions. For example, D2 receptors are expressed at particularly higher levels in neurons of the caudate and the nucleus accumbens, but are also present in the amygdala, hippocampus, and parts of the cerebral cortex. The amygdala, hippocampus, and neocortex are possible sites of therapeutic action. The D2 receptors in the caudate and putamen, however, presumably contribute to the psychomotor side effects of the antipsychotic drugs.

In contrast, the D3 and D4 receptors are found primarily in the limbic system and cortex and only weakly expressed in the basal ganglia, so they do not give rise to the side effects associated with the extrapyramidal (basal ganglia) systems.

An example of an atypical antipsychotic is clozapine. Clozapine is an effective therapeutic agent with fewer side effects compared to typical antipsychotic medications. It is a very complicated drug--it interacts with at least nine neurotransmitter receptors. Clozapine blocks both D2 and D4 receptors, but has a higher affinity for D4 receptors, which may contribute to its therapeutic effects (as mentioned above), but minimize the adverse motor effects. In addition to antagonizing DA receptors, clozapine is a 5HT2 receptor antagonist. The 5HT2 receptor is implicated in the hallucinogenic effects of LSD. It may be the combination of DA and 5HT antagonism that leads to its beneficial effects. Blockade of 5HT2 receptors may be directly therapeutic, but may also impact dopaminergic activity. For example, in the basal ganglia, serotonin input to the nigrostriatal DA neurons causes the DA neurons to be inhibited. So, blocking 5HT receptors may lead to an increase in DA. Serotonergic blockade may increase DA release in the striatum, which counteracts the antagonistic effects on DA receptors. Thus, the net effect of this drug on nigrostriatal DA is not great. BUT, the relationship of 5HT and DA is not the same in the mesolimbic DA pathway; thus, the therapeutic effect remains. Unfortunately, however, clozapine also comes with an increased risk of potentially fatal bone marrow toxicity, so other novel pharmacologic agents are being developed.

DOPAMINE HYPOTHESIS

Given that the antipsychotics all block dopamine transmission, it was hypothesized that schizophrenia is related to excessive dopamine neurotransmission. Consistent with this hypothesis, drugs that increase DA can induce psychotic episodes resembling paranoid schizophrenia (remember the MPTP patients who were treated with l-dopa?). Some of these drugs can also cause repetitive stereotyped behaviors. However, there is still no direct evidence of excessive activity in DA neurons in schizophrenia.

What evidence is there of dopaminergic dysfunction?

Well, alterations in dopaminergic functioning could be caused by dysfunction at a number of sites: it could be related to an increase in DA release, in alterations in autoreceptors or postsynaptic receptors, or alterations in reuptake or breakdown. A couple of recent studies suggest there may be increased DA release. For example, Laruelle (1996, 1997) found that amphetamine causes a great release of DA in the striatum of schizophrenic patients.

What about receptors? Data can be collected via postmortem studies or via PET scans with radioactive ligands for DA receptors. The results are equivocal, likely due to the difficulty of discerning changes associated with the pathology, changes associated with adaptive responses to the pathology, and changes associated with long-term medication. Also, many studies may have focused on the wrong anatomical area or receptor subtype. Perhaps it is not the D2 receptor that is altered!

Two recent studies have found evidence for increased D3 and D4 DA receptors in the brains of deceased schizophrenics. Murray (1995) found a twofold increase in the concentration of D4 receptors in the nucleus accumbens and Gurevich (1997) found a twofold increase in D3 receptors in both the neostriatum and nucleus accumbens. The patients had been drug-free for at least one month before their deaths.

Given the inconsistency in findings, perhaps the site of pathology is not the dopaminergic systems themselves. Perhaps DA transmission alterations are the consequence of abnormalities in their regulation by prefrontal and limbic cortical regions.

NEGATIVE SYMPTOMS MAY BE RELATED TO HYPOFRONTALITY AND REDUCTIONS IN DA NEUROTRANSMISSION

The plot thickens.

What about the negative symptoms?

Daniel Weinberger (1988) has postulated that the mesolimbic and mesocortical DA systems are disturbed in different ways in schizophrenia. First, an increased activity in the mesolimbic pathway (and D2 and D3 receptors) would account for the positive symptoms. Second, decreased DA activity of the mesocortical connections in the prefrontal cortex would account for the negative symptoms. Perhaps the imbalance between cortical and subcortical DA transmission underlies the development of schizophrenia. Weinberger proposes that activity in the mesocortical pathway to the prefrontal cortex normally inhibits the mesolimbic pathway by feedback inhibition. He suggests that alterations in the mesocortical system, like a reduction in the activity of the frontal lobes (hypofrontality), may lead to the alterations in the mesolimbic system.

This scheme is still untested, however, there is experimental evidence for an interaction between these pathways. If you chemically lesion the mesocortical pathway in experimental animals, it enhances synaptic responsiveness in the mesolimbic pathway, specifically in its termination in the nucleus accumbens. For example, King et al. (1997) found that destruction of the DA axons and terminals in the prefrontal cortex causes increased DA secretion in the nucleus accumbens. Moreover, Okubo (1997) found a reduction the number of D1 DA receptors in the prefrontal cortex of schizophrenics using PET, and the number correlated with the severity of negative symptoms. Similarly, Akil et al (1999) found a reduction in the length of DA axons in the prefrontal cortex of deceased schizophrenic patients.

There is also direct evidence of both structural and functional damage in the frontal lobes, associated with deficits in executive functioning. As mentioned above, there is disorganization of neurons and reduced prefrontal activity. For example, there are reductions in dorsolateral prefrontal cortex activity in schizophrenics, particularly evident when the patient is challenged with a task that requires the functional integrity of this area (Taylor, 1996). Similar manipulations of executive functioning can be achieved with DA agonists and antagonists (remember PKU?)

But does this evidence really mean that the underlying pathology lies within the dopaminergic systems? Does the effectiveness of a pharmacologic agent necessarily indicate that the pathology lies at the drug’s site of action? No. For example, the primary defect in Parkinson’s is a decrease in DA levels, but some symptoms can be alleviated by drugs that block cholinergic transmission. So, the neuropathology may only indirectly affect DA neurotransmission. Therefore, one can not necessarily extrapolate the causal mechanisms of a disease by the mechanism of action of a therapeutic agent.

As I mentioned in class, this can be further illustrated by the hypothetical case of three presynaptic neurons converging on a postsynaptic neuron, with each releasing a different transmitter. Transmitter A and DA reduce the excitability, whereas transmitter B directly excites it. If schizophrenia resulted from a defect in neuron A or its transmitter, causing an excess of postsynaptic activity, one could improve the symptoms by simply blocking the action of DA.

There are other issues to consider. Just like our antidepressants, antipsychotic drugs occupy DA receptors very quickly, but the maximal therapeutic effects are often delayed by several weeks. Thus, the acute blockade of DA transmission is not likely to be sufficient for the therapeutic effects. Activity in certain neural circuits may need to adjust to a new level of modulation. In addition, the drugs or the resulting alterations in neuronal activity might produce changes in gene expression, which may not become manifest for one or more weeks. So, is it a direct change in DA receptors, or changes in other neurotransmitter systems (like 5HT)? Incidentally, long-term administration of atypical antipsychotic agents leads to a downregulation of 5HT2 receptors.

GLUTAMATE HYPOTHESIS

It has been hypothesized that schizophrenia is associated with a reduction in glutamate activity.

PCP, an NMDA receptor antagonist, produces a psychosis that resembles that in schizophrenia and exacerbates psychosis in patients with schizophrenia. PCP leads to impairments in working memory, attention deficits, decreased drive, thought disorders, hallucinations, and delusions. In essence, it produces both the positive and negative symptoms associated with schizophrenia. Other NMDA receptor antagonists, like ketamine, also produce psychosis. I mentioned that NMDA receptor-mediated excitotoxicity may produce cell death in stroke patients. One treatment for stroke is the administration of certain NMDA receptor antagonists, the side effects of which include psychotic behaviors.

PCP binds to NMDA receptors, a subtype of glutamatergic receptor. NMDA receptors interact with DA systems. For example, NMDA receptors are present on the DA axon terminals in the prefrontal cortex and enhance DA release from the terminals. Thus, PCP inhibits release of dopamine in the prefrontal cortex. In contrast, at the DA terminal in the nucleus accumbens, PCP increases DA release and inhibits reuptake, much like amphetamine.

Thus, the behavioral effects of PCP may be due in part to its ability to enhance DA release in the mesolimbic pathway while blocking DA release in the mesocortical pathway. With such an action, both positive and negative symptoms should appear.

What evidence is there of alterations in the glutamatergic systems? As Paola pointed out and as noted in Keshavan’s paper, reduced CSF glutamate levels, altered glutamate metabolism, and altered NMDA receptor subunit gene expression have been reported

.

So, evidence suggests that schizophrenia may begin with the loss and/or disorganization of neurons somewhere in the brain that causes hypofrontality, leading to a reduction in the release of DA in the prefrontal cortex. The hypofrontality produces the negative symptoms of schizophrenia. It also causes an increase in the activity of the DA neurons in the mesolimbic system, which produces positive symptoms.

If glutamate is the culprit, then NMDA agonists should reduce schizophrenic symptoms. Unfortunately, NMDA agonists may lead to excitotoxic cell death, so modulators of NMDA receptors are better candidates for therapeutic effects. You may recall that the NMDA receptor has a number of modulatory sites, one of which is a glycine site. Glycine agonists, like d-cylcoserine, improve negative symptoms (Heresco-Levy et al., 1999). Stay tuned for the development of novel antipsychotics.

So, if schizophrenia is related to developmental insults, why do the symptoms not appear until adolescence? Various hypotheses have been proposed. It may be that the dysfunction is not uncovered until adolescence, when the synaptic pruning causes the reduction in synaptic connections. Or that excessive pruning occurs during adolescence, either of which might reduce the connections below a threshold. It is also possible that the younger brain is simply not sensitive to the effects of NMDA receptor antagonism like the adult. For example, ketamine produces psychosis in adults, but not children.

So, put this together with Keshavan’s paper. Keshavan hypothesizes that because glutamate plays a critical role in development, including cell migration, differentiation, and synaptic sculpturing, glutamatergic dysfunction may lead to developmental injury. Related to the NMDA receptor’s role in development, there is a transient period during late gestation and early postnatal development during which the brain overexpresses NMDA receptors. During this period, the brain is hypersensitive to NMDA receptor dysfunction, as well as NMDA receptor-mediated excitotoxicity. So, glutamate may certainly play a role in the developmental abnormalities. Glutamate and NMDA receptors are also important for later plasticity and neuronal connectivity (recall our discussions of the neurobiology of learning and memory). During adolescence, when there is pruning and structuring of connections, loss of NMDA receptors may exaggerate the loss of synaptic connections and lead to aberrant connectivity.

How does this lead to schizophrenic symptoms?

The question of tonic vs. phasic DA release came up in class. Phasic depends on DA cell activity and is enhanced when DA neurons fire in bursts. It is transient and selectively affects receptors within or near the synapse. Tonic release is a steady-state level of DA in the extracellular space. So DA transmission occurs in two different temporal modes, tonic and phasic. We never got into this, but many of these neurotransmitters are released from vesicle-containing varicosities along the axon, some of which make synapses and others that do not. Each of these types of DA release are regulated by different mechanisms, although they may influence one another. Glutamate enhances tonic DA release in the prefrontal cortex. The cortical areas (both frontal and temporal) modulate both tonic and phasic DA release in the striatum. If there is a reduction in glutamate, it may lead to a reduction in tonic DA release in the prefrontal cortex (associated with negative symptoms). This, in turn, can modulate the tonic and phasic DA release in subcortical areas, leading to an increase in phasic DA release. This can help explain how reductions in glutamate may lead to both reductions of prefrontal DA and increases in mesolimbic DA.

Finally, Keshavan proposes that increases in mesolimbic DA might lead to NMDA receptor-mediated excitotoxicity and further cell loss. To date, the evidence for this is still limited. Nevertheless, it does provide a parsimonious theory of the pathophysiology of schizophrenia.

Other neurochemical systems, including serotonergic, GABergic, and cholinergic may also play a role. In addition, as Paola mentioned, alterations in connectivity among cortical regions may be more critical than neurochemical alterations, although they likely go hand-in-hand.

It should be evident that there is much to be learned about the underlying neuropathology of schizophrenia. There is a lot of research focusing on the genetic factors that contribute to the etiology. You should also be getting a sense of the difficulty in determining the mechanisms of damage, given that systems do not operate independently of one another. This is also evident in other disorders, like depression. As an experimenter, it can be challenging to pick apart abnormalities that are directly related to the disease state, rather than indirectly altered in response to the disease state or to medications.

Akil, et al. 1999. Lamina-specific alterations in the dopamine innervation of the prefrontal cortex in schizophrenic subjects. American Journal of Psychiatry, 156, 1580-1589.

Brier, A., et al. 1997. Schizophrenia is associated with elevated amphetamine-induced synaptic dopamine concentrations: Evidence from a novel positron emission tomography method. Proceedings of the National Academy of Sciences, 94, 2569-2574.

Egan, M.F., Weinberger, D.R. 1997. Neurobiology of schizophrenia. Current Opinion in Neurobiology, 7, 701-707.

Gurevich, E.V., Bordelon, Y., Shapir, R.M., Arnold, S.E., Gur, R.E., and Joyce, J.N. 1997. Mesolimbic dopamine D3 receptors and use of antipsychotics in patients with schizophrenia: A postmortem study. Archives of General Psychiatry, 54, 225-232.

King, D., Zigmond, M.J., and Finlay, J.M. 1997. Effects of dopamine depletion in the medial prefrontal cortex on the stress-induced increase in extracellular dopamine in the nucleus accumbens core and shell. Neuroscience, 77, 141-153.

Laruelle, M., et al. 1996. Single photon emission computerized tomography imaging of amphetamine-induced dopamine release in drug-free schizophrenic subjects. Proceedings of the National Academy of Sciences, 93, 9235-9240.

Murray, A.M., Hyde, T.M, Knable, M.B., Herman, M.M., Bigelow, L.B., Carter, J.M., Weinberger, D.R., and Kleinman, J.E. 1995. Distribution of putative D4 dopamine receptors in postmortem striatum from patients with schizophrenia, Journal of Neuroscience, 15, 2186-2191.

Okubo, Y., et al. 1997. Decreased prefrontal dopamine D1 receptors in schizophrenia revealed by PET. Nature, 385, 634-636.

Pearlson, G.D. 2000. Neurobiology of schizophrenia. Annals of Neurology, 48, 445-466.

Weinberger, D.R. 1988. Schizophrenia and the frontal lobe. Trends in Neuroscience, 11, 367-370.