The present study, which was designed to explore mechanisms underlying the increase in brain KYNA levels in the PFC of individuals with SZ, revealed distinct abnormalities in KP enzymes in both cortical regions studied, ie, BA 9 and BA 10. These changes in SZ, which were not related to demographics, were restricted to enzymes in the main branch of the KP, whereas the activity of KAT II, in the side arm of the pathway, was in the normal range. Moreover, both the patients’ medication history and complementary studies in chronically risperidone-treated rats indicated that the results of our study were not affected by antipsychotic drug use. Our data therefore provide insights into the pathophysiology of SZ and also suggest new treatment strategies.
The tryptophan metabolite kynurenine occupies a central position in the KP (). In the brain, kynurenine gives rise to two physically segregated branches of the pathway, producing 3-hydroxykynurenine and its downstream metabolites 3-hydroxyanthranilic acid and quinolinic acid in microglial cells and KYNA in astrocytes (cf Introduction).23
Excessive formation of the three microglial compounds, which are neurotoxins and generators of highly reactive free radicals, may play significant roles in brain pathology.33–37
Astrocyte-derived KYNA, in contrast, has neuroprotective properties due to its ability to block neuronal excitation and scavenge free radicals.37–39
Of the enzymes that use kynurenine as a substrate, KMO is the most specific and has the lowest Km
, and is therefore rate limiting. Reduced KMO activity will decrease the flux of the KP toward quinolinic acid and might therefore indirectly provide a degree of neuroprotection. We have previously proposed that this can be exploited for the treatment of Huntington's disease and other neurodegenerative disorders by cautiously targeting KMO with specific enzyme inhibitors.40
The present study revealed a significant decrease in KMO activity in the PFC of individuals with SZ. This reduction, which was tentatively linked to a lower vmax
rather than to a Km
change, was not accompanied by a decrease in the activity of kynureninase, the next enzyme in the metabolic cascade. On the contrary, kynureninase activity in SZ tended to be higher than in controls, though the difference did not attain statistical significance in either of the two prefrontal regions studied. It therefore appears that the observed reduction in KMO activity is not a reflection of a generalized microglial abnormality, which has been invoked to play a significant role in SZ and interpreted as an indication of a compromised immune system in the disease.41–45
In light of recent studies, it is more likely that the impairment of KMO activity in SZ is selective, possibly due to functional sequence variants in the KMO
The activity of 3-HAO, which catalyzes the formation of the NMDA receptor agonist quinolinic acid from 3-hydroxyanthranilic acid, was found to be reduced in BA 9, ie, the dorsolateral subdivision of the PFC that is preferentially involved in sustaining attention and working memory.48
A tendency toward lower 3-HAO activity was also observed in BA 10, though the results were not statistically significant. Decreased 3-HAO activity might account for the elevation in the tissue levels of 3-hydroxyanthranilic acid in SZ, which was recently demonstrated in the anterior cingulate cortex49
and might affect the redox status of neurons and glial cells in the area (see above). In addition, reduced 3-HAO activity will translate into lower quinolinic acid formation and may thus possibly contribute to NMDA receptor hypofunction.
No disease-related changes were seen in the activity of the next enzyme in the cascade, QPRT, in either region of the PFC. This further supports the notion that distinct, rather than generalized, KP impairments exist in the brain of patients with SZ. Studies currently in progress in our laboratory are designed to elucidate the genetic underpinnings and molecular mechanisms of the discrete anomalies in KP metabolism reported here. In particular, we are investigating the possible role of cosubstrates, cofactors, and regulators of the two impaired oxygenases, ie, KMO and 3-HAO, such as molecular oxygen, metal ions, and the endogenous anti-oxidant glutathione,50,51
all of which are established risk factors in SZ (see Brown and Susser52
and Do et al53
for recent reviews).
The question then arises whether and how specific impairments in KP enzymes might account for the significant increases in prefrontal KYNA levels in SZ, which were originally described in 2001.22
The most parsimonious explanation would be that a reduction in KMO activity eventually triggers a shift in cerebral KP metabolism toward enhanced KYNA formation in SZ. As demonstrated in a recent in vivo study in rats, such a redirection of KP metabolism toward increased KYNA synthesis does not occur in the normal brain when KMO activity is acutely reduced by pharmacological means.54
However, KYNA production is indeed enhanced under these conditions when the experiment is performed in injured brain tissue where glial functions are abnormal.54
This mechanism may therefore also operate in SZ, where microglial and astrocytic anomalies in the PFC have been repeatedly described (see above).42,55–57
Moreover, it is quite conceivable that prolonged
downregulation of KMO, as opposed to the effects of acute
enzyme inhibition studied by Amori et al,54
will eventually favor KYNA synthesis over the synthesis of 3-hydroxykynurenine.
The dynamics of the pivotal metabolite kynurenine deserve special consideration in a discussion of possible functional interactions between the two KP branches in the brain of individuals with SZ. Postmortem analysis reveals that kynurenine levels are elevated in the PFC of patients, and this increase is correlated with KYNA levels in the same tissue.22
The explanation for this nexus seems unambiguous because the high Km
of KAT II and all other cerebral kynurenine aminotransferases allows for a proportional increase in KYNA formation when kynurenine levels rise.58
of increased kynurenine levels in the brain of SZ patients is less clear. This elevation, which is also seen in the cerebrospinal fluid59
and must therefore include changes in the extracellular milieu, may be directly related to reduced KMO activity, ie, to an accumulation of the enzyme's substrate. Alternatively or quite possibly in addition, kynurenine levels in the SZ brain might be elevated due to increased activity of the biosynthetic enzymes tryptophan 2,3-dioxygenase60
Notably, these two enzymes, like the entire cerebral KP pathway, are preferentially localized in glial cells,23,60–62
and newly produced kynurenine is readily liberated into the extracellular compartment.63
Irrespective of the underlying enzymatic and cellular mechanism(s), there are reasons to assume that the observed increase in prefrontal KYNA levels plays a role in the pathophysiology of SZ.22,64
Within the PFC, astrocyte-derived KYNA controls the levels of acetylcholine and glutamate14,15,20
by initially targeting and thus reducing the activity of α7nAChRs.12
Thus, increased KYNA levels trigger or exacerbate the nicotinergic and glutamatergic deficits, which have been credibly linked to both cognitive dysfunctions and psychotic manifestations in humans (cf Introduction).1–3,65,66
The demonstration of distinct impairments in cerebral KP metabolism in SZ, which are also observed in the basal ganglia,67
raises the prospect that more than one KP enzyme could be targeted to provide clinical benefits in the disease. This idea, which is an extension of our recently proposed strategy to use selective KAT II inhibitors as cognition enhancers in SZ,68
includes interventions that are aimed specifically at normalizing KMO and 3-HAO activity in the brain of patients. We are currently testing this concept in relevant animal models of SZ.