|Home | About | Journals | Submit | Contact Us | Français|
Variation in the monoamine-oxidase-A gene has been associated with volumetric changes in corticolimbic regions with differences in their response to relevant emotional tasks. Here we show no changes in baseline regional brain metabolism as a function of genotype indicating that, unchallenged, corticolimbic activity is not modulated by the MAOA genotype.
Monoamine oxidase A (MAO-A) degrades neurotransmitters, including serotonin, dopamine and norepinephrine implicated in regulating mood and behavior (Shih & Thompson, 1999). A functional polymorphism in the MAO-A gene promoter is described as 4-tandem-repeats (high MAO-A activity) in 60% and 3-repeats (low MAO-A activity) in 40% in healthy men (Sabol et al., 1998). Studies in humans (Foley et al., 2004) and primates (Newman et al., 2005) support a link between the low MAO-A genotype and susceptibility to antisocial behavior in the face of childhood maltreatment. Functional MRI studies found that healthy subjects with the low MAO-A genotype have mostly reduced volume (Meyer-Lindenberg et al., 2006) and changed cortical activity to cognitive and emotional tasks as compared to the high genotype (Fan et al., 2003; Passamonti et al., 2006; Meyer-Lindenberg et al., 2006).
It is not known, however, whether brain function without a task, at resting baseline would differ as a function of the MAO-A genotype. As compared to fMRI, glucose utilization measured by PET is a quantifiable and absolute measure of brain activity. Using the aggregation of brain activity over time without a specific challenge yet at an alert state, may provide an informative gestalt of baseline neuronal activity. We hypothesized that brain function without any task, therefore without a specific challenge to expose susceptibility, will not show genotype modulated differences in brain metabolism.
Thirty-eight male subjects (32±6 years of age) participated in this PET study following a thorough physical examination and interview by a neurologist to verify healthy status. All 38 control subjects were fasting four hours before this PET study. Non-smoking status was ascertained by self-report and verified by breath CO test. Two PET scans were obtained 2 hours apart at rest: a [11C]clorgyline [reported elsewhere (Fowler et al., 2007)] followed by a 18FDG scan to measure glucose metabolism both done at rest.
All subjects provided cheek swab samples containing their DNA which was analyzed for MAO-A (Freeman et al., 2003). Polymerase chain reactions (PCRs) were performed as described previously (Sabol et al., 1998). The PCR products were analyzed on an Applied Biosystems 3100 Genetic analyzer resulting in 12 subjects (32%) having the 3-repeat allele (low) and 26 (68%) having the 4-repeat allele (high), distributions that parallel previous studies (Meyer-Lindenberg et al, 2006). Note that we did not have subjects with 3.5 or 5 repeat alleles in this sample possibly due to the relative rarity of these alleles in the population (less than 2%). There were no significant differences between the two genotype groups on age, education, right hand dominance (Oldfield, 1971), socioeconomic status (Hollingshead, 1975), verbal (Wilkinson, 1993) and non-verbal (Wechsler, 1999) measures of intelligence, and on self-reported depression (Beck et al., 1996) (all P > 0.62). Subjects were fully informed and provided written consent in accordance with the local Institutional Review Board.
The 18FDG scans were acquired on a whole body PET scanner (Siemen’s HR+ 4.5×4.5×4.8 mm at center of field-of-view) in 3D mode providing 63 contiguous planes of 2.4mm each. To stabilize the head, an individually molded head-holder was made for each subject. Subjects were kept supine with their eyes open in a quiet room with a nurse monitoring to prevent sleep and maintain awake state. Catheters were placed in the antecubital vein for radiotracer injection and the radial artery for blood sampling. A transmission scan was obtained with a 68Ge rotating rod source before the emission scan to correct for attenuation before the radiotracer injection.18FDG (Hamacher et al., 1986) (3–5 mCi) was injected. Serial blood samples were taken from time of 18FDG injection through 55 minutes. Emission data were attenuation-corrected and reconstructed using filtered back projection and metabolic images were computed (Reivich et al., 1985).
Data were analyzed using Statistical Parametric Mapping (SPM) (Friston et al., 1995) on the “absolute” and the “relative” (images normalized to the mean metabolic activity of all voxels within the brain) metabolic images. The relative scaling corrects for these individual differences by accounting for differences in global metabolism. For this purpose the metabolic measures were spatially normalized using a 2×2×2 mm3 voxel size and the template provided in the SPM 99 package and subsequently smoothed with a 16 mm isotropic Gaussian kernel. Two separate independent-samples t-tests were performed to compare the absolute and the relative images obtained from the participants during rest.
Manual regions of interest (ROI) were also drawn on the metabolic images of each subject. For ROI placement we re-sliced the metabolic images along the AC-PC line and summed the 63 planes in groups of 2 obtaining 23 planes of 4.76mm thickness with 11 planes above and 11 planes below the thalamus. We applied Talairach and Tournoux (1988) template while manually adjusting the position of the ROI for each individual. Values were computed using the weighted average from the different planes for the regions in Figure 1. Each of these ROI was identified in at least two contiguous slices. To obtain a global metabolic value for each individual we chose 10–12 of these planes (depending on the size of the individual brain) typically choosing 6–7 planes above, and 4–5 planes below the thalamus using a program that thresholded at 20% of the maximum metabolic rate for the plane (Volkow et al., 2006).
Whole-brain SPM significance was set at P<0.05, cluster-level corrected, though uncorrected thresholds were also inspected. The ROI from manual drawings are reported at the threshold level of P<0.05, Bonferroni corrected for regions implicated in the MAO A genotype (Buckholtz et al., 2007).
Using whole brain analyses in SPM, there were no significant differences in absolute or relative baseline metabolism between the genotyped groups. Reducing the significance threshold to P<0.05, uncorrected, there were still no differences in absolute or relative metabolism. The ROI results confirmed the non significant SPM findings (all P>0.30) (Figure 1).
Baseline metabolism, a marker of resting brain function (Volkow et al., 2006) did not differ as a function of the MAO-A genotype. Posthoc analysis revealed that the sample size in this study would yield a power of 80% for the t-test to detect differences at effect size 1 (P<0.05, two-tailed). In our study the pooled SDs was about 20% of the population mean. This indicates that the effect of MAO-A genotype on brain metabolism at rest is smaller than the variability in brain metabolism at rest in adult subjects. Failure to see a difference could reflect the need to challenge with a relevant task to detect the differences in activation patterns as a function of genotype. Given volumetric differences in amygdala and cingulate that were observed as a function of genotype (Meyer-Lindenberg et al., 2006) one would expect differences in metabolism unless the lack of challenge created a ceiling effect in these healthy subjects.
This finding mirrors our negative results in the same sample with [11C]clorgyline, a radiotracer with specificity for brain MAO-A (Fowler et al., 2007). It appears that the MAO-A genotype has a modulatory effect on neuronal maturation in utero, and is relevant in childhood only through sensitivity to environmental insult (Caspi et al., 2002) and in adulthood through reactivity to emotional stimuli in fMRI studies (Meyer-Lindenberg, et al., 2006). However, fMRI experiments cannot measure brain function at absolute baseline making it difficult to distinguish differences due to the task challenge from those pertaining to general brain function at baseline (Canli et al., 2005). Although Independent Component Analysis (ICA) can detect baseline fluctuations and map regions that have similar time-varying responses ICA does not provide absolute measures in fMRI. Since we measure a substantial period of time when the brain is not harnessed task demands we call this absolute baseline. Here we show that at resting baseline, the MAO-A genotype does not impart a significant effect on glucose metabolism in brain.
This finding corroborates our hypothesis that in the context of externalizing behavior phenotypes (such as antisocial behavior), significant associations between genotype and behavior emerge primarily in response to a challenge serving to perturb the individual beyond their baseline.
This work was carried out at Brookhaven National Laboratory under contract DE-AC-298CH10886 with the U.S. Department of Energy and supported by its Office of Biological and Environmental Research and by NIH-NIDA (K05DA020001), NIH CGRC (MO1RR10710) and by the National Association for Research on Schizophrenia and Depression (NARSAD). We thank the PET team for advice and assistance in different aspects of these studies. We are also grateful to the subjects who volunteered for these studies.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.