Faculty Bibliography

Studies in rodents indicate that diets deficient in omega-3 polyunsaturated fatty acids (n-3 PUFA) lower dopamine neurotransmission as measured by striatal vesicular monoamine transporter type 2 (VMAT2) density and amphetamine-induced dopamine release. This suggests that dietary supplementation with fish oil might increase VMAT2 availability, enhance dopamine storage and release, and improve dopamine-dependent cognitive functions such as working memory. To investigate this mechanism in humans, positron emission tomography (PET) was used to measure VMAT2 availability pre- and post-supplementation of n-3 PUFA in healthy individuals. Healthy young adult subjects were scanned with PET using [(11)C]-(+)-alpha-dihydrotetrabenzine (DTBZ) before and after six months of n-3 PUFA supplementation (Lovaza, 2 g/day containing docosahexaenonic acid, DHA 750 mg/d and eicosapentaenoic acid, EPA 930 mg/d). In addition, subjects underwent a working memory task (n-back) and red blood cell membrane (RBC) fatty acid composition analysis pre- and post-supplementation. RBC analysis showed a significant increase in both DHA and EPA post-supplementation. In contrast, no significant change in [(11)C]DTBZ binding potential (BP(ND)) in striatum and its subdivisions were observed after supplementation with n-3 PUFA. No correlation was evident between n-3 PUFA induced change in RBC DHA or EPA levels and change in [(11)C]DTBZ BP(ND) in striatal subdivisions. However, pre-supplementation RBC DHA levels was predictive of baseline performance (i.e., adjusted hit rate, AHR on 3-back) on the n-back task (y = 0.19+0.07, r(2) = 0.55, p = 0.009). In addition, subjects AHR performance improved on 3-back post-supplementation (pre 0.65+/-0.27, post 0.80+/-0.15, p = 0.04). The correlation between n-back performance, and DHA levels are consistent with reports in which higher DHA levels is related to improved cognitive performance. However, the lack of change in [(11)C]DBTZ BP(ND) indicates that striatal VMAT2 regulation is not the mechanism of action by which n-3 PUFA improves cognitive performance.

Evidence indicates that synchronization of cortical activity at gamma-band frequencies, mediated through GABA-A receptors, is important for perceptual/cognitive processes. To study GABA signaling in vivo, we recently used a novel positron emission tomography (PET) paradigm measuring the change in binding of the benzodiazepine (BDZ) site radiotracer [(11)C]flumazenil associated with increases in extracellular GABA induced via GABA membrane transporter (GAT1) blockade with tiagabine. GAT1 blockade resulted in significant increases in [(11)C]flumazenil binding potential (BPND) over baseline in the major functional domains of the cortex, consistent with preclinical studies showing that increased GABA levels enhance the affinity of GABA-A receptors for BDZ ligands. In the current study we sought to replicate our previous results and to further validate this approach by demonstrating that the magnitude of increase in [(11)C]flumazenil binding observed with PET is directly correlated with tiagabine dose. [(11)C]flumazenil distribution volume (VT) was measured in 18 healthy volunteers before and after GAT1 blockade with tiagabine. Two dose groups were studied (n = 9 per group; Group I: tiagabine 0.15 mg/kg; Group II: tiagabine 0.25 mg/kg). GAT1 blockade resulted in increases in mean (+/- SD) [(11)C]flumazenil VT in Group II in association cortices (6.8 +/- 0.8 mL g-1 vs. 7.3 +/- 0.4 mL g-1;p = 0.03), sensory cortices (6.7 +/- 0.8 mL g-1 vs. 7.3 +/- 0.5 mL g-1;p = 0.02) and limbic regions (5.2 +/- 0.6 mL g-1 vs. 5.7 +/- 0.3 mL g-1;p = 0.03). No change was observed at the low dose (Group I). Increased orbital frontal cortex binding of [(11)C]flumazenil in Group II correlated with the ability to entrain cortical networks (r = 0.67, p = 0.05) measured via EEG during a cognitive control task. These data provide a replication of our previous study demonstrating the ability to measure in vivo, with PET, acute shifts in extracellular GABA.

Evidence from biochemical, imaging, and treatment studies suggest abnormalities of the serotonin system in autism spectrum disorders, in particular in frontolimbic areas of the brain. We used the radiotracers [(11)C]MDL 100907 and [(11)C]DASB to characterize the 5-HT(2A) receptor and serotonin transporter in Asperger's Disorder. Seventeen individuals with Asperger's Disorder (age=34.3 +/- 11.1 years) and 17 healthy controls (age=33.0 +/- 9.6 years) were scanned with [(11)C]MDL 100907. Of the 17 patients, eight (age=29.7 +/- 7.0 years) were also scanned with [(1)(1)C]DASB, as were eight healthy controls (age=28.7 +/- 7.0 years). Patients with Asperger's Disorder and healthy control subjects were matched for age, gender, and ethnicity, and all had normal intelligence. Metabolite-corrected arterial plasma inputs were collected and data analyzed by two-tissue compartment modeling. The primary outcome measure was regional binding potential BP(ND). Neither regional [(1)(1)C]MDL 100907 BP(ND) nor [(1)(1)C]DASB BP(ND) was statistically different between the Asperger's and healthy subjects. This study failed to find significant alterations in binding parameters of 5-HT(2A) receptors and serotonin transporters in adult subjects with Asperger's disorder.

OBJECTIVE: Positron emission tomography (PET) studies performed with [(11) C]raclopride have consistently reported lower binding to D(2/3) receptors and lower amphetamine-induced dopamine (DA) release in cocaine abusers relative to healthy controls. A limitation of these studies that were performed with D(2/3) antagonist radiotracers such as [(11) C]raclopride is the failure to provide information that is specific to D(2/3) receptors configured in a state of high affinity for the agonists (i.e., D(2/3) receptors coupled to G-proteins, D(2/3 HIGH) ). As the endogenous agonist DA binds with preference to D(2/3 HIGH) relative to D(2/3 LOW) receptors (i.e., D(2/3) receptors uncoupled to G-proteins) it is critical to understand the in vivo status of D(2/3 HIGH) receptors in cocaine dependence. Thus, we measured the available fraction of D(2/3) (HIGH) receptors in 10 recently abstinent cocaine abusers (CD) and matched healthy controls (HC) with the D(2/3) antagonist and agonist PET radiotracers [(11) C]raclopride and [(11) C]NPA. METHODS: [(11) C]raclopride and [(11) C]NPA binding potential (BP) (BP(ND) ) in the striatum were measured with kinetic analysis using the arterial input function. The available fraction of D(2/3 HIGH) receptors, i.e., % R(HIGH) available = D(2/3 HIGH) /(D(2/3 HIGH) + D(2/3 LOW) ) was then computed as the ratio of [(11) C]NPA BP(ND) /[(11) C]raclopride BP(ND) . RESULTS: No differences in striatal [(11) C]NPA BP(ND) (HC = 1.00 +/- 0.17; CD = 0.97 +/- 0.17, P = 0.67) or available % R(HIGH) (HC = 39% +/- 5%; CD = 41% +/- 5%, P = 0.50) was observed between cocaine abusers and matched controls. CONCLUSIONS: The results of this [(11) C]NPA PET study do not support alterations in D(2/3 HIGH) binding in the striatum in cocaine dependence.

In a recent positron emission tomography (PET) study, we demonstrated the ability to measure amphetamine-induced dopamine (DA) release in the human cortex with the DA D(2)/(3) radioligand [(1)(1)C]FLB 457. As previous studies in animals have shown that a relatively high fraction of the [(1)(1)C]FLB 457 signal in the cerebellum represents specific binding to D(2)/(3) receptors, there was concern that the use of the cerebellum as a measure of nonspecific binding (i.e., reference region) to derive [(1)(1)C]FLB 457 binding potential (BP) (BP(ND) ) would bias cortical DA release measurements. Thus, we evaluated the fractional contribution of specific binding to D(2)/(3) receptors in the human cerebellum for [(1)(1)C]FLB 457. Six healthy human subjects (5M/1F) were studied twice with [(1)(1)C]FLB 457, once at baseline and again following a single oral dose of 15 mg of aripiprazole, a D(2)/(3) partial agonist. [(1)(1)C]FLB 457 distribution volume (V(T) ) was estimated using kinetic analysis in the cortical regions of interest and potential reference regions. The change in [(1)(1)C]FLB 457 V(T) following aripiprazole ranged from -33 to -42% in the cortical regions of interest (ROIs). The aripiprazole-induced change in [(1)(1)C]FLB 457 V(T) in three potential reference regions suggests significant specific binding the cerebellum (CER, -17 +/- 12%), but not pons (PON, -10 +/- 10%) and centrum semiovale (CESVL, -3 +/- 12%). Nevertheless, a reanalysis of the published [(1)(1)C]FLB 457 test-retest and amphetamine studies suggests that the use of the PON V(T) and CESVL V(T) as an estimate of nonspecific binding to derive [(1)(1)C]FLB 457 BP(ND) in DA release studies is unlikely to be successful because it leads to less reproducible outcome measures, which in turn diminishes the ability to measure DA release in the cortex. D(2)/(3) blocking studies with aripiprazole and [(1)(1)C]FLB 457 suggest specific binding to D(2)/(3) receptors in the cerebellum. These data also suggest that the contribution of specific binding to D(2)/(3) receptors in the cerebellum is lower than that in the cortical ROIs and that CER V(T) is mostly representative of nonspecific binding. Nevertheless, caution is advised when using reference tissue methods that rely solely on the cerebellum signal as an input function to quantify [(1)(1)C]FLB 457 BP(ND).

The aim of this study was to compare 5-HT(1A) availability in vivo in individuals with schizophrenia before and during treatment with the atypical antipsychotic ziprasidone. Six individuals with schizophrenia underwent two PET scans with [(11)C]WAY 100635; the first while medication-free (baseline) and the second while taking the atypical antipsychotic ziprasidone (on-medication). Regional volumes of distribution (V(T), mL g(-1)) were derived using a two-tissue compartment kinetic model. Outcome measures included binding potential relative to the plasma (BP(P), mL g(-1)) and the binding potential relative to the nonspecific distribution volume (BP(ND), unitless). No significant differences were observed in regional BP(P) or BP(ND) with ziprasidone treatment. A significant correlation was noted between BP(P) measured in the orbitofrontal cortex during the on-medication condition and degree of improvement in negative symptoms with treatment (r = 0.96, p = 0.004). Consistent with the published literature of changes in 5-HT(1A) binding during treatment with 5-HT(1A) receptor agonists, this study did not detect a significant reduction in 5-HT(1A) binding with ziprasidone. The finding of a relationship between 5-HT(1A) binding and the degree of improvement in negative symptoms provides further support for the role of the 5-HT(1A) receptor in the pathophysiology and treatment of this symptom domain.

In a recent PET study, we demonstrated the ability to measure amphetamine-induced DA release in the human cortex with the relatively high affinity dopamine D(2)/(3) radioligand [(1)(1)C]FLB 457 (Narendran et al., [2009] Synapse 63:447-461). The aim of this study was to evaluate the reproducibility and reliability of [(1)(1)C]FLB 457 in the same imaging paradigm we used to measure amphetamine-induced DA transmission. Six healthy human subjects (three males/three females)were studied twice with [(1)(1)C]FLB 457, once at baseline and again 3 h following the end of the baseline scan. D(2)/(3) receptor binding parameters were estimated using a two-tissue compartment kinetic analysis in the cortical regions of interest and cerebellum (reference region). The test-retest variability and intraclass correlation coefficient were assessed for distribution volume (VT), binding potential relative to plasma concentration (BP(P)), and binding potential relative to non-displaceable uptake (BP(ND)) of [(1)(1)C]FLB 457. The test-retest variability of [(1)(1)C]FLB 457 VT, BPP, and BP(ND) were