Maturation of limbic regions in Asperger syndrome: A preliminary study using proton magnetic resonance spectroscopy and structural magnetic resonance imaging

Psychiatry Research: Neuroimaging 184 (2010) 77–85

Contents lists available at ScienceDirect

Psychiatry Research: Neuroimaging j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p s yc h r e s n s

Maturation of limbic regions in Asperger syndrome: A preliminary study using proton magnetic resonance spectroscopy and structural magnetic resonance imaging Finian M. O'Brien a,f,g, Lisa Page a, Ruth L. O'Gorman b, Patrick Bolton c, Ajay Sharma d, Gillian Baird e, Eileen Daly a, Brian Hallahan a,f, Ronán M. Conroy g, Catherine Foy a, Sarah Curran a, Dene Robertson a, Kieran C. Murphy f,g, Declan G.M. Murphy a,⁎ a

Section of Brain Maturation (Department of Psychological Medicine), Institute of Psychiatry, Kings College London, UK The Neuroimaging Research Group, Institute of Psychiatry, Kings College London, UK c Social, Genetic, Developmental & Psychiatry Research Centre, Institute of Psychiatry, Kings College London, UK d Southwark Health Centre, London, UK e St. Georges Hospital Medical School, London, UK f Department of Psychiatry, Beaumont Hospital, Dublin, Ireland g The Royal College of Surgeons in Ireland, Dublin, Ireland b

a r t i c l e

i n f o

Article history: Received 8 April 2009 Received in revised form 5 July 2010 Accepted 11 August 2010 Keywords: Autistic disorder Asperger syndrome Magnetic resonance imaging Magnetic resonance spectroscopy Amygdala Hippocampus Aging

a b s t r a c t People with autistic spectrum disorders (ASD, including Asperger syndrome) may have developmental abnormalities in the amygdala–hippocampal complex (AHC). However, in vivo, age-related comparisons of both volume and neuronal integrity of the AHC have not yet been carried out in people with Asperger syndrome (AS) versus controls. We compared structure and metabolic activity of the right AHC of 22 individuals with AS and 22 healthy controls aged 10–50 years and examined the effects of age between groups. We used structural magnetic resonace imaging (sMRI) to measure the volume of the AHC, and magnetic resonance spectroscopy (1H-MRS) to measure concentrations of N-acetyl aspartate (NAA), creatine + phosphocreatine (Cr + PCr), myo-inositol (mI) and choline (Cho). The bulk volume of the amygdala and the hippocampus did not differ significantly between groups, but there was a significant difference in the effect of age on the hippocampus in controls. Compared with controls, young (but not older) people with AS had a significantly higher AHC concentration of NAA and a significantly higher NAA/Cr ratio. People with AS, but not controls, had a significant age-related reduction in NAA and the NAA/Cr ratio. Also, in people with AS, but not controls, there was a significant relationship between concentrations of choline and age so that choline concentrations reduced with age. We therefore suggest that people with AS have significant differences in neuronal and lipid membrane integrity and maturation of the AHC. © 2010 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Asperger syndrome (AS) is a subtype of the autistic spectrum disorders (ASD). A core feature of ASD is abnormal social-communicative behaviour, and it has been proposed that people with ASD have aberrant development of the limbic system (Bauman and Kemper, 1988; Pierce et al., 2001). In the healthy human population, the amygdaloid region (comprising the amygdala and hippocampus) of the limbic system plays a crucial role in mediating social behaviours. The amygdala is involved in emotional arousal, processing of facial expressions (Baron-Cohen et al., 1999; Critchley et al., 2000; Pierce et al., 2001), assigning significance to environmental stimuli, and ⁎ Corresponding author. P.O. Box 50, Section of Brain Maturation, Division of Psychological Medicine, Institute of Psychiatry, London SE5 8AF, UK. Tel.: + 44 207 8480984; fax: + 44 207 8480650. E-mail address: [email protected] (D.G.M. Murphy). 0925-4927/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.pscychresns.2010.08.007

emotional learning (Brothers et al., 1990; Anderson and Phelps, 2001; Grelotti et al., 2002; Helmuth, 2003). In contrast, the hippocampus is implicated in memory function and stress regulation (De Long, 1992; Strange and Dolan, 1999; Hoschl and Hajek, 2001; Burgess et al., 2002). Both structures have a close, reciprocal relationship during memory and social tasks (Adolphs, 2002; Critchley, 2003; Richardson et al., 2004; Richter-Levin, 2004). People with ASD may have differences in structural integrity and function of the amygdaloid region compared with controls. Lesions to temporal lobe structures, including the amygdaloid region, of both animals and humans are associated with impaired socio-affective behaviours (De Long et al., 1981; Gillberg, 1986; Hoon and Reiss, 1992; Bachevalier, 1994, 1996; Bolton, 2004). Also, neuro-pathological studies of individuals with autism found abnormalities in the limbic region, including the amygdala–hippocampal complex (AHC). However, results have been inconsistent. Some reported increased cell density, reduced neuron size and abnormalities in dendritic

78

F.M. O'Brien et al. / Psychiatry Research: Neuroimaging 184 (2010) 77–85

arborization (Palmen et al., 2004), whereas a recent study found reduced neuronal density and no difference in neuronal size (Schumann and Amaral, 2006). Similarly, in vivo volumetric studies have yielded conflicting results. Hippocampal volume has been reported to be either decreased (Aylward et al., 1999; Saitoh et al., 2001) or not different (Saitoh et al., 1995; Piven et al., 1998; Haznedar et al., 2000; Howard et al., 2000); and amygdala volume has been reported as decreased (Pierce et al., 2001; Schultz, 2005), increased (Abell et al., 1999; Howard et al., 2000; Sparks et al., 2002) and not different (Haznedar et al., 2000). Also, functional magnetic resonance imaging (fMRI) studies found that the amygdaloid region (and particularly the right side) is hypoactive when individuals with ASD process emotional faces, suggesting that neuro-circuitry in this region may develop abnormally (Pierce et al., 2001). Moreover, it has been suggested that there may be a genetic basis to limbic abnormalities in ASD, and one study reported increased hippocampal volume (Rojas et al., 2004) in parents of people with ASD. In contrast another found no difference in parental temporal lobe volume (including the AHC) (Palmen et al., 2005). The reasons for the variable findings from prior studies are unknown, but may reflect differences in data analysis and/ or heterogeneity in the samples studied (some included children, some adults, some with mental retardation and some without, and some mixed samples of people with autism and AS). Nevertheless there is evidence from both autopsy and in vivo imaging studies that limbic regions may be developmentally abnormal in people with ASD. Recent evidence suggests that people with ASD have age-related differences from controls in brain size and anatomy. For example, some studies report that young, but not older, children with ASD have increased volume of the whole brain and some frontal regions (Aylward et al., 2002; Courchesne, 2004). Also, one study found that boys with high-functioning autism have no significant difference in brain volume, including the AHC (Herbert et al., 2003). In contrast, others reported that young children with autism have an increased volume of both the hippocampus and the amygdala, but that adolescents only have increased hippocampal (and not amygdala) volume (Schumann et al., 2004). In addition, a recent study found increased right amygdala (but not left amygdala or hippocampal) volume in young children with ASD that was associated with more severe social and communication impairments (Munson et al., 2006). Thus, it has been proposed that people with ASD have age-related differences from controls in limbic regions. Furthermore, we previously reported significant differences in aging of the whole brain in adults of normal intelligence with ASD (McAlonan et al., 2002) and in programmed cell death of medial frontal regions in adults with ASD (Murphy et al., 2002). To date, however, no previous studies have compared in vivo the maturation of limbic regions in physically healthy people with and without ASD as measured by vivo proton magnetic resonance spectroscopy (1H-MRS) — a neuro-imaging technique that can be used to assess brain metabolism and neuronal integrity through quantification of N-acetyl aspartate (NAA), creatine (Cr + PCr), myo-inositol (mI), glutamate + glutamine + GABA (Glx) and choline (Cho) containing substances (Koller et al., 1984; Danielson and Ross, 1999). NAA synthesis is catalyzed by a mitochondrial enzyme and is present in high concentration in both grey and white matter, and so NAA is often treated as a marker of axonal and neuronal integrity and viability (Koller et al., 1984; Clark, 1998; Danielson and Ross, 1999; Bhakoo et al., 2001; Brandao and Domingues, 2004; Soares and Law, 2009). In contrast, Cr + PCr and Cho have been used as indicators of phosphate and lipid membrane metabolism, respectively (Koller et al., 1984; Davie et al., 1994; Danielson and Ross, 1999; Brandao and Domingues, 2004; Soares and Law, 2009), while mI and Glx are considered markers of glial and excitatory neurotransmitter regulatory activity in the brain, respectively (Danielson and Ross, 1999; Soares and Law, 2009). Seven 1H-MRS studies have examined the AHC in people with ASD (Otsuka et al., 1999; Mori et al., 2001; Page et al., 2006; Endo et al.,

2007; Zeegers et al., 2007; Gabis et al., 2008; Kleinhans et al., 2009) and these reported conflicting results. Those of children reported either no difference in NAA concentration (Zeegers et al., 2007) or a significantly lower NAA (Otsuka et al., 1999; Mori et al., 2001; Endo et al., 2007; Gabis et al., 2008), whereas we and one other group found no difference in NAA of adults (Page et al., 2006; Kleinhans et al., 2009). However, some of these prior studies included those with mental retardation and epilepsy and none examined the effects of age. Thus, the discrepant findings between prior 1H-MRS studies of the AHC may be explained by differences in intelligence quotient (IQ) and health (because both factors are independently associated with differences in NAA) (Vermathen et al., 1997; Jung et al., 1999), and/ or differences in brain maturation. In summary, it has been proposed that people with ASD have abnormalities in the development of the AHC, and some studies have suggested that the right side, in particular, may be most affected. However, relatively few studies have investigated neuronal integrity or aging of this region in people with ASD, and no previous studies have examined the effect of age on both volume and neuronal integrity. Also, 1H-MRS can provide evidence of neuronal abnormalities even when brain anatomy appears to be grossly normal as measured using structural magnetic resonace imaging (sMRI) (Connelly et al., 1998; Filippi et al., 2002). Therefore, we used both sMRI and 1H-MRS to compare volume, neuronal integrity and age-related differences in the amygdala–hippocampal complex of non-mentally retarded, physically healthy, adolescents and adults with AS and controls. We hypothesised that those participants with AS would have developmental abnormalities in the amygdala–hippocampal complex. 2. Methods 2.1. Participants People with AS were recruited through our clinical research program in autism (part of the MRC UK A.I.M.S. network). We studied 22 high-functioning individuals with AS, of whom four were females; their mean (S.D.) age was 23 (13) years and Full Scale IQ (FSIQ) was 101 (22). The age range for AS subjects was 10–46 years. Twenty-one were right-handed and one left-handed. Twelve were less than 16 years of age and ten were older than 20. People with AS were clinically diagnosed using the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10) research criteria (World Health Organisation, 1993), and the diagnosis was confirmed where parental informants were available with the Autism Diagnostic Interview (ADI) (17 cases) (Lord et al., 1994), or the Autism Diagnostic Observation Schedule (ADOS) (Lord et al., 2000) (5 cases) when parents were not available or unwilling. Clinical interviews and ADI or ADOS assessment to confirm diagnosis and level of autistic dysfunction were carried out at entry to the study. We included subjects with no reported language delay and who met autistic criteria for abnormal social and obsessional behaviour [i.e. we excluded people with autism or pervasive disorder not otherwise specified (PDD-NOS)]. Controls were recruited locally through advertisement and comprised 22 individuals (2 female) with a mean age of 23 (11) years and FSIQ of 110 (12). The age range was 10–50 years. Twentyone were right-handed and one left-handed. Ten were aged less than 16 years, and twelve were aged over 20 years. All participants in the study underwent structured physical and psychiatric examination to assess the presence of a co-morbid Axis I or Axis II disorder as determined by the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV; American Psychiatric Association, 1994). Exclusion criteria were a history of physical or psychiatric disorders affecting brain function (e.g., epilepsy, affective disorder or psychosis), a genetic disorder putatively associated with autistic spectrum disorders (e.g., Fragile X syndrome),

F.M. O'Brien et al. / Psychiatry Research: Neuroimaging 184 (2010) 77–85

79

or a clinically abnormal finding on magnetic resonance imaging. No participants had ever taken psychotropic medication at the time of the study. We measured overall intelligence using the Wechsler Intelligence Scale for Children-Third Edition (Wechsler, 1991) or the Wechsler Adult Intelligence Scale-Revised (WAIS-R) (Wechsler, 1981), as appropriate. There were no significant differences between subjects with AS and controls in age, FSIQ, handedness, sex or education. Research was approved by the respective local research ethics committees, and all participants (and/or their carer) provided written consent/assent as appropriate.

2.2. MRI scanning protocol and procedures All participants underwent scanning using a 1.5-T system with a transmit–receive quadrature head coil (GE Signa Nvi; General Electric, Milwaukee, Wis) running LX software. Structural MR imaging was carried out to exclude those participants with clinically detectable brain abnormalities and to measure the volume of the amygdala and hippocampus. Differences in the proportion of grey and white matter in the volume of interest (VOI) of the 1H-MRS acquisition were also investigated. Structural imaging sequences included a 3D fast inversion recovery prepared spoiled gradient acquisition in the steady state (IR-SPGR), with inversion time = 450 ms, echo time = 2.8 ms, and repetition time = 13.8 ms. The sequence involved the acquisition of 124 contiguous coronal slices with a slice thickness of 1.5 mm, a field of view of 220 mm, and a matrix of 256 × 256, resulting in an inplane resolution of 0.859 × 0.859 mm2. These data were used to manually trace regional brain matter using the Measure software (MEASURE version 0.8.7) and previously described methods (Rojas et al., 2004; Page et al., 2006). There were two raters for which the intra-class correlation co-efficients (ICCs) and the intra-rater ICCs were greater than 0.9 for volumetric measurements of total brain, hippocampus and amygdala. Single-voxel 1H MR (1H-MRS) spectroscopy was performed in the same scanning session using a point-resolved spectroscopy (PRESS) sequence with repetition time = 3 s, echo time = 35 ms, 160 averages, number of complex data points = 2048, and spectral width = 2500 Hz. MR spectra were acquired from a 6-ml (20 × 20 × 15 mm3) volume prescribed over the right amygdala–hippocampal complex using coordinates derived from the coronal IR-SPGR images (Figs. 1 and 2).

Fig. 2. An example of typical MR spectra from the amygdala–hippocampal complex (in AS). The in vivo data are shown in black, and the LCModel fit is overlaid in red.

The water suppression and shimming were optimised using a standard automated pre-scan, incorporating a first-order automatic shimming method based on a fast 3-planar acquisition of B0 maps, which are used to calculate the appropriate shim values. The average water line width was 6 Hz for both AS subjects and controls. In vivo metabolite levels for NAA, Cr + PCr, mI, Glx and Cho were measured using LCModel software version 6.1-0 (LCModel; Provencher, 1993) with a standard 1.5 T GE basis of concentration-calibrated model spectra from 15 metabolites, correcting for eddy current effects and coil loading (http://s-provencher.com/pages/lcmodel.shtml). The metabolite concentrations reported by LCModel were divided by the fractional content of brain tissue (p[GM] + p[WM], (where p [GM] and p[WM] represent the percentage of grey matter and white matter in the voxel)) to correct for any cerebrospinal fluid (CSF) in the MRS voxel. The tissue and CSF percentages were calculated from the IR-SPGR volume using the segmentation function of SPM99 software (Statistical Parametric Mapping available at: http://www.fil.ion.bpmf. ac.uk/spm). These corrected concentrations were then calibrated to

Fig. 1. Inversion recovery FSPGR magnetic resonance image from a healthy control subject illustrating the location of the 1H-MRS voxel in the right amygdala–hippocampal complex.

80

F.M. O'Brien et al. / Psychiatry Research: Neuroimaging 184 (2010) 77–85

institutional molar units with respect to a phantom of known metabolite concentrations scanned with the same MRS protocol as the in vivo spectra, (TR = 3000, TE = 35 ms). (This calibration factor is given by the ratio of the known concentration of NAA in the phantom and the concentration output from LCModel.) Relaxation time corrections were not performed, but potential errors from relaxation effects would be expected to be small given that the MR spectra were acquired with a relatively short echo time and a long repetition time, and that the errors associated with relaxation time effects are proportional to the difference in T2 relaxation rates in vivo and in vitro (LCModel; Provencher, 1993). As expected, some of these metabolite peaks that were included in the LCModel did not reach statistical significance when fitted (including glutamate or glutamate/glutamine (i.e. the combined signal Glx) in the young population. However, NAA, Cr + PCr, mI and Cho did reach statistical significance for all spectra derived from the amygdala–hippocampal VOI, and concentrations were therefore derived from these metabolite peaks. 2.3. Statistical analysis Data were analyzed blind to subject status using the Statistical Package for the Social Sciences (SPSS) (SPSS Inc, Chicago, IL) and Stata SE Release 11 (StataCorp., 2009). We carried out both between-group and within-group analysis of the data. Firstly we compared (as two groups) all people with AS versus all controls. The data were normally distributed (using P–P plots); therefore, parametric tests were employed. The volume of a brain region may be affected by differences in whole brain volume. Hence we compared both the ‘raw’ volume of the amygdala and hippocampus, and the volume when ‘normalized', with the whole brain (by expressing it as% of total brain volume). We compared mean group differences in metabolite concentrations and ratios, and hand-traced% regional brain volume using student t-tests when comparing people with AS and controls. We used analysis of co-variance (ANCOVA), with sex as a co-variate when comparing the Older Groups as females were more prevalent (not significantly) in the AS cohort. To further investigate age-related differences, we used two methods. Firstly, we split each sample into ‘young’ or ‘adult’ depending on age. “Youngsters” were defined as less than 16 years, and adults as older than 20 years. Group differences (young versus adult AS and controls) in metabolite concentrations were tested with a one-way analysis of variance (ANOVA), with Group as the betweensubject factor. Also hippocampal and amygdala raw volumes were separately analyzed using one-way ANCOVA, with Group as the single factor and Whole Brain Volume as the co-variate. All analyses of group differences underwent subsequent post-hoc comparisons using appropriate statistical tests (i.e. parametric or non-parametric depending on the distribution of the data). Secondly, we used fractional polynomial regression to model the relationship between the parameter of interest and age separately, in cases and controls. Fractional polynomial regression allows the detection of nonlinear relationships, deriving the best-fitting curve from a family of candidate curves. In each case, the fitted curve was compared with a simple linear regression model in order to detect significant departures from linearity. Evidence of differential relationships between age and each parameter were examined using a Chow test, which simultaneously tests the hypothesis that each group has a different slope and intercept. Results are reported as significant when P b 0.05 (two-tailed). In order to determine if our data were “driven” by people in whom we were not able to carry out parental interviews, or by sampling heterogeneity, we carried out a further between-group analysis after restricting the inclusion criteria to include only those people with AS for whom ADI scores were available (n = 17).

We carried out correlation analyses to ascertain if metabolite concentrations were associated with hippocampal or amygdala volume. Lastly, we carried out an exploratory post-hoc analysis to relate those metabolite concentrations that differed significantly between groups to increased severity of symptoms as measured by the Autism Diagnostic Interview (ADI-R). 3. Results There were no significant differences between the whole group of people with AS and controls in age, sex, raw or corrected volume of the amygdala and the hippocampus; volume of white matter, gray matter and CSF content of the MRS VOI; or any metabolite concentration or metabolite ratio (Table 1). However, this whole group analysis obscured age-dependent differences. There were no significant differences observed between people with AS and controls at any age in the normalized whole volume of the amygdala and the hippocampus. However, in controls but not those with AS, there was a significant decline with age in raw volumes of whole (F = 3.9, d.f. = 2, 19, P = 0.037) and right (t = −2.1, P = 0.042) hippocampus. Also, normalized right hippocampus showed a sharp, curvilinear decline in volume at earlier ages (Chow test P = 0.012) in controls only, but this relationship appeared to disappear after age 20 years (Fig. 3). No significant relationship with age or case status was observed with raw or normalized volumes of whole and right amygdala. The younger, but not older, group of people with AS had a significantly higher NAA concentration (t = 2.221, d.f. = 2, 20; P = 0.038) and NAA/Cr ratio (t = 2.644; d.f. = 2, 20; P = 0.016) than controls in the AHC (Table 1). Significant interactions between age and case–control status were found for AHC concentrations of NAA (Chow test P = 0.017) and NAA/ Cr ratio (Chow test P = 0.003). In people with AS, but not controls, concentrations of NAA declined with increasing age (t = − 2.9, P = 0.009) (Fig. 4) and examination of fractional polynomial fits showed no significant departure from linearity in either group. Age was related to NAA/Cr ratio in both AS cases and controls, but the shape of this relationship differed (Fig. 5) so that in those with AS, levels declined with age (t = − 2.4, P = 0.028), while in controls there was a significant increase with age (t = 2.8, P = 0.011). Again, there was no evidence from fractional polynomial curve fitting of a significant nonlinear relationship. Also, in people with AS, but not controls, there was a significant relationship between concentrations of choline and age (Chow test P = 0.002) so that choline concentrations reduced with age (Fig. 6). We carried out a power analysis using nQuery Advisor, version 4 (Statistical Solutions, MA, USA) and were able to detect a difference in NAA concentration of 15% with a power of 80%. These results were unchanged when we only included those people for whom ADI scores were obtainable. Also, we found no significant relationship between decreased NAA and NAA/Cr and more severe clinical symptoms. We found no correlations between any of the metabolites and amygdala or hippocampal volume when we examined individuals with AS and controls separately or when we examined the Older Group alone. We did, however, note negative correlations between hippocampal volume and NAA/Cr (r = −0.497, P = 0.02) and Cho/Cr (r = −0.528, P = 0.01) in the Younger Group. 4. Discussion We found that otherwise healthy people with AS, over the age of 10 years, have no significant differences from controls in bulk volume of the amygdala and hippocampus. However, we report preliminary evidence that people with AS have significant age-related differences from controls in structure and neuronal integrity of this region.

F.M. O'Brien et al. / Psychiatry Research: Neuroimaging 184 (2010) 77–85

81

Table 1 Demographic data on participants [data are given as means (± S.D.); metabolite concentrations are given in institutional units (IU)]. Independent t-tests were used to compare differences between individuals with Asperger syndrome and healthy controls in children, and analysis of co-variance with sex as a co-variate (where appropriate) was used to compare adults with Asperger syndrome and healthy controls (due to the greater proportion of males in the healthy control group). Children

Asperger's N = 12

Controls N = 10

Significance (two-tailed)

Age Sex FSIQ

13 (2) m = 11; f = 1 98 (24)

13 (2) m = 9; f = 1 109 (6)

0.97 0.89 0.23

Hippocampus (cm3) Total Right Left

5.9 (0.7) 3.0 (0.4) 2.9 (0.3)

6.2 (0.6) 3.2 (0.4) 2.9 (0.2)

0.43 0.46 0.48

Amygdala (cm3) Total Right Left

5.2 (1.0) 2.5 (0.5) 2.6 (0.5)

5.3 (0.5) 2.7 (0.3) 2.7 (0.3)

0.69 0.52 0.90

0.71 (0.1) 0.22 (0.1) 0.15 (0.3) 1.5 (0.1) 0.3 (0.0) 4.9 (0.3) 4.9 (0.6) 1.0 (0.1) 5.2 (0.4)

0.50 0.41 0.26 0.27 0.13 0.82 0.038* 0.016* 0.54

Amygdala–hippocampal complex VOI Gray matter vol.% 0.68 (0.1) White matter vol.% 0.26 (0.1) CSF vol.% 0.05 (0.0) Cho concentration (IU) 1.6 (0.1) Cho/Cr ratio 0.3 (0.0) Cr + PCr concentration (IU) 4.8 (0.4) NAA concentration (IU) 5.6 (0.8) NAA/Cr ratio 1.1 (0.1) Myo-inositol 5.0 (0.8) concentration (IU) Adults

Aspergers N = 10

Controls N = 12

Significance (two-tailed)

Age Sex FSIQ

35 (10) m = 7; f = 3 105 (15)

32 (8) m = 11; f = 1 112 (15)

0.35 0.20 0.21

Hippocampus (cm3) Total Right Left

5.9 (0.8) 3.0 (0.4) 2.8 (0.3)

5.8 (0.8) 3.0 (0.4) 2.8 (0.4)

0.72 0.75 0.87

Amygdala (cm3) Total Right Left

5.8 (0.8) 2.8 (0.3) 2.9 (0.4)

5.3 (0.6) 2.6 (0.3) 2.7 (0.3)

0.12 0.08 0.22

0.49 (0.2) 0.45 (0.2) 0.04 (0.0) 1.4 (0.3) 0.3 (0.0) 4.6 (1.0) 4.8 (1.0) 1.1 (0.2) 4.1 (1.4)

0.60 0.73 0.87 0.76 0.14 0.67 0.57 0.14 0.49

Amygdala–hippocampal complex VOI Gray matter vol.% 0.49 (0.1) White matter vol.% 0.42 (0.2) CSF vol.% 0.05 (0.0) Cho concentration (IU) 1.4 (0.2) Cho/Cr ratio 0.3 (0.0) Cr + PCr concentration (IU) 4.9 (1.4) NAA concentration (IU) 4.6 (1.0) NAA/Cr ratio 1.0 (0.1) Myo-inositol 4.5 (1.1) concentration (IU)

Fig. 3. Relationship between normalized right hippocampal volume and age in people with Asperger syndrome and controls (significant change with age in controls P = 0.012).

these structures. However, it was not practically possible to use a smaller VOI due to the technical constraints of MRS in this region (e.g. the need for a sufficient signal-to-noise ratio), and if we had placed the VOI more anteriorly or posteriorly, we would have included nontarget tissue. Further, this was a cross-sectional study, and so we are only able to describe age-related differences — and not individual changes over time. Hence there may have been undetectable agerelated confounders (e.g. health differences and/or secular effects) that affected our findings. Moreover, some of our results may be explained by type 1 error as we carried out multiple comparisons. Nevertheless, this pragmatic study design allows analysis across a wide age range (approximately 35 years), which would not be achievable in a longitudinal brain imaging study. Also, our results stayed the same when we restricted our analysis to only include those with an ADI. Moreover, the significant between-group differences in NAA we found were internally consistent — i.e. we found differences both when we compared means in the young and old subgroups, and when we compared correlations with age across both whole groups. In addition we only included physically healthy, medication free, AS individuals, who were not mentally retarded. We also employed two neuro-imaging techniques – sMRI and 1H-MRS – to help explore both brain structure and metabolism in the same study population. Hence our finding of differences in metabolism/neuronal integrity, while

The two figures highlighted in bold and with an asterisk are the significant between group findings for that analysis.

Nevertheless, we did not include people with ‘classical autism’ (i.e. with abnormalities in language development and/or learning disability) or very young children. Hence, we do not know if our findings will generalize to other groups within the autistic spectrum. Also, we did not obtain spectra from other brain regions (due to time constraints involved in scanning children). Thus we were not able to examine the neuronal integrity of other limbic regions, or include ‘control’ regions not classically implicated in ASD. Moreover, our VOI included contributions from both the amygdala and hippocampus; and so we do not know if neuronal integrity is abnormal in only one, or both, of

Fig. 4. Relationship between NAA and age in people with Asperger syndrome and controls (significant change with age of NAA in cases P = 0.009).

82

F.M. O'Brien et al. / Psychiatry Research: Neuroimaging 184 (2010) 77–85

Fig. 5. Relationship between NAA:Cr and Age in people with Asperger syndrome and controls (significant change with age of NAA:Cr in cases P = 0.028 and controls P = 0.011).

preliminary, cannot be explained by potential confounding factors such as intellectual disability, epilepsy, or gross (detectable) differences in brain anatomy. Our finding that the bulk volumes of the amygdala and the hippocampus are not significantly different from those of controls is consistent with prior work in younger and older people with AS. For example, the only previous study of children and adolescents found no difference in gross volume of the amygdala and the hippocampus (Schumann et al., 2004) in AS. Similarly, normal amygdala (Haznedar et al., 2000; Dziobek et al., 2006) and hippocampal size (Haznedar et al., 2000) has been reported in adults with AS. Moreover, our finding of no significant age-related difference in amygdala volume is consistent with the only previous study that examined aging of this region in AS (Schumann et al., 2004) and extends those findings through to middle adulthood. In contrast, our finding that raw volumes of the whole hippocampus and both raw and normalized volumes of the right hippocampus differ significantly with age in healthy controls, but not in AS, is consistent with previous studies of normal aging (Pruessner et al., 2001; Rajah et al., 2009). However, it is not consistent with the only previous aging study of this region in AS, which reported no significant differences from controls (Schumann et al., 2004). Nevertheless, unlike our study, Schumann et al. focused on children aged between 7 and 19 years (i.e. a narrower age range than

Fig. 6. Relationship between Cho and Age in people with Asperger syndrome and controls (significant change with age of Cho in cases P = 0.002, but not controls).

in our study) and did not examine volumetric differences with age separately within case and control groups. Hence differences in methodology and sample age could partially explain this variability in observations. Therefore, we suggest that our preliminary findings indicate that people with AS, when adults are included, have agerelated differences in the hippocampus. Lastly, our findings provide tentative support for the suggestion that differences in bulk volume and aging of the amygdala and hippocampus may only be detected in those with more clinically severe subtypes of autism, or at a much younger age (Sparks et al., 2002; Lotspeich et al., 2004; Schumann et al., 2004; Nacewicz et al., 2006). Our finding of an elevation in NAA, but not other metabolites, in the right amygdala–hippocampal complex may result from differences in brain tissue composition of the VOI we studied (because the relative concentrations of NAA, Cho, and Cr differ in grey and white matter, (Pfefferbaum et al., 1999). However, the grey:white matter ratio in the people we studied was similar to those in previous reports (Miller et al., 1980). Also we found no significant between-group difference in the proportion of grey matter/white matter/CSF contained in the MRS VOIs so that our measures of VOI tissue proportion are most likely reliable, and our results cannot be fully explained by these potential confounds. Our finding of no difference in NAA concentration in the right AHC of our adult (older) subgroup is consistent with both previous studies of this region in adults with ASD, one of which examined people with AS (Page et al., 2006) and the other, a mixed group of people with AS and high-functioning autism (Kleinhans et al., 2009). However, the increased concentration of NAA we found in “youngsters” with AS differs from the other previous MRS studies of this region in ASD. NAA concentration in the right amygdala–hippocampal complex has been reported as lower (Otsuka et al., 1999; Endo et al., 2007; Gabis et al., 2008) or not different (Mori et al., 2001); and that in the left AHC was found to be significantly reduced (Mori et al., 2001; Gabis et al., 2008) or not different (Zeegers et al., 2007). The reason for this discrepancy in findings is unknown. However, some of these previous studies of young people included a variety of different subtypes of ASD (including those with a history of language delay and mental retardation). Also, although three (Endo et al., 2007; Zeegers et al., 2007; Gabis et al., 2008) controlled for the effects of age and gender, two of these studies (Zeegers et al., 2007; Gabis et al., 2008) included groups containing younger children (age range 2-6 years and 7-16 years, respectively) than in studied a group containing mostly females with ASD (Endo et al., 2007). In contrast we studied a predominantly male group that included only non-retarded people with AS from (relatively late) childhood upwards. Hence, it is likely that (similar to our anatomical results) clinical differences in the populations studied partially account for the differences in our findings. NAA is present in high concentration in the grey matter, white matter and neuronal cell bodies, and a mitochondrial enzyme catalyzes its synthesis — and NAA is often treated as a measure of neuronal density and/or viability (Danielson and Ross, 1999; Soares and Law, 2009). Thus, the increased NAA we found in the ‘youngsters’ with AS, and the significant differences we found between cases and controls in the relationship between age and NAA, may be explained by differences in neuronal density and/or integrity; for example, an increased density of neurons and/or a potentially abnormal mitochondrial metabolism of NAA in the AHC (Koller et al., 1984; Clark, 1998; Danielson and Ross, 1999; Bhakoo et al., 2001; Driscoll et al., 2003; Soares and Law, 2009). Cho is generally accepted as a marker of lipid membrane density and integrity; and is a proxy measure for the synthesis and degradation of phospholipids (Soares and Law, 2009). To our best knowledge, no other study has previously examined Cho concentration in the AHC of people with AS only, but our finding of no significant group differences in Cho concentration is consistent with many other studies of children, adolescents and adults with ASD

F.M. O'Brien et al. / Psychiatry Research: Neuroimaging 184 (2010) 77–85

(Otsuka et al., 1999; Mori et al., 2001; Page et al., 2006; Gabis et al., 2008). Nevertheless, one 1H-MRS study of this region in adults did report an increased concentration of Cho in ASD (Suzuki et al., 2010). The reason for this variability in results is unknown, but the latter study examined hippocampus only and involved relatively small numbers of adults with high-functioning autism so that clinical differences in methodology and sample characteristics may account for the differences observed. However, our finding that Cho concentration changed with age in the AS group but not controls, suggests significant differences in the maturation of neuronal membranes. We found no significant group differences in Cr + PCr and mI concentration of the right AHC and no changes with age of these metabolites in people with AS as compared with healthy controls. This suggests that there are no differences (detectable using 1H-MRS at 1.5 tesla) in phosphate energy metabolism or neuro-glial concentration of this region in AS. To our knowledge, no prior studies have examined Cr + PCr and mI concentrations in this region of people with AS exclusively. However, our finding is consistent with two recent studies of this region in children (Endo et al., 2007; Gabis et al., 2008), but not with a study of adults with ASD (Page et al., 2006). Rather, in that latter study, we reported a significantly higher concentration of Cr + PCr, but not of NAA and mI, in the right AHC of 25 people with high-functioning autism or Asperger syndrome, compared with healthy controls. It is likely that clinical differences in populations studied are partly responsible for differences in results between these studies. One possible explanation for the age-associated differences we found in NAA and Cho is that people with AS have a developmental abnormality in the right AHC which (when they are young) lead to an increased density of neurons rather than abnormal neuronal metabolism per se and that, with age, is associated with reduced neuronal and lipid membrane integrity. This suggestion is consistent with findings from some neuro-pathological studies of autistic subjects (Palmen et al., 2004) that reported an increased cell packing density in limbic regions. On the other hand, this hypothesis is not supported by a recent study (and the first to use quantitative stereological methods) which reported reduced neuronal density in people with autism (Schumann and Amaral, 2006) and consistent with other MRS studies of ASD reporting reduced NAA concentrations in the AHC (Otsuka et al., 1999; Endo et al., 2007; Gabis et al., 2008). However, the discrepancies in these findings may be explained by differences in subregional development within the limbic system in ASD. For example, one MRS study reported reduced NAA/Cr concentration ratios in both the right and left AHC and increased Cho/Cr concentrations in the left (but not the right) AHC in children with ASD. Moreover, language impairment within that ASD group was associated with higher Cho/Cr concentration ratios in the left (but not the right) AHC (Gabis et al., 2008). These findings suggest that neuronal and lipid cell membrane integrity in the left and right AHC is differentially abnormal in ASD. Furthermore, another recent study found NAA and Cr concentrations in amygdala to correlate with clinical severity of autism (Kleinhans et al., 2009). Thus, similar to volumetric findings, the degree of abnormality of the AHC detected in MRS may be associated with clinical subtype of autism. Overall, our findings add to growing evidence that the limbic region develops and matures abnormally in ASD. However, replication of this study and further studies are required to examine subregional neuronal and lipid membrane integrity in ASD and to determine if age-dependent differences in NAA and Cho are explained by abnormalities in neuronal, axonal and lipid membrane density, metabolism, or both. To our best knowledge, there are no prior 1H-MRS studies of normal age-related differences in the amygdala–hippocampal complex from childhood to middle age in either healthy controls or people with ASD. However, one cross-sectional study reported that NAA/Cr concentrations are significantly higher in the right, compared with the left AHC in a group of normally developing children (Gabis et al., 2008), suggesting that the right and the left AHC develop differently. Also, a

83

number of 1H-MRS studies reported that hippocampal NAA concentration diminishes in healthy older adulthood (Schuff et al., 1999; Horska et al., 2002). Moreover, 1H-MRS studies which examined the effect of age on the neuronal integrity/metabolism of other brain structures in the general population reported that, overall, the concentration of NAA and Cho increases in grey and white matter until approximately the first and third decades (respectively) — before declining gradually thereafter (Toft et al., 1994; Kato et al., 1997; Kadota et al., 2001). Thus normal cerebral maturation is generally associated with an initial elevation in NAA and Cho, with subsequent down-regulation (Kreis et al., 1993). Morphological studies of neural development are generally consistent with these spectroscopic findings, reporting an initial rapid overproduction of neural synapses in early to late childhood (Kostovic et al., 1995), with subsequent synapse elimination late in childhood and adolescence (Huttenlocher and Dabholkar, 1997), followed by slow reduction in synaptic density thereafter (Huttenlocher, 1979). Thus, we suggest that our findings of significantly increased NAA concentration in young (but not older) people and the significant changes with age of NAA and Cho in AS are consistent with an abnormal early neuronal integrity and maturation of limbic regions into middle adulthood. Moreover, clinical and behavioural studies have consistently reported that adults with AS continue to have abnormal socio-emotional behaviours (Wing, 1993; Gillberg, 1995; Wing, 1996), and that both children and adults with AS have abnormalities in limbic response to facial emotions (Bachevalier, 1994; Baron-Cohen et al., 2000; Critchley et al., 2000; Pierce et al., 2001; Schultz, 2005). Therefore, we suggest that these differences in neuronal, axonal and lipid membrane development in the amygdala– hippocampal complex as measured by 1H-MRS may relate to findings of functional abnormality in the AHC from childhood to adulthood, so that developmental differences may continue to affect brain function (and behaviour) in adults with AS. We have previously reported that adults with AS have a significant increase in NAA, Cho, and Cr in medial frontal (but not parietal) regions (Murphy et al., 2002). Thus we do not suggest that the abnormalities in neuronal integrity of people with AS are restricted to the amygdala–hippocampal complex. Rather, our findings, and the work of others (Aylward et al., 2002; Courchesne, 2004; Page et al., 2006; Kleinhans et al., 2007), indicate that people with AS have abnormal post-natal brain growth, and differences in the maturation (and perhaps programmed cell death) of limbic regions. These are detectable, using MRS, in the amygdala–hippocampal complex of young adolescents, and some frontal regions in adults. Further developmental studies involving sMRI and 1H-MRS are required to determine if our findings in the AHC can be replicated, and whether differences in cerebral maturation/aging also occur in other brain regions. Future studies should also use voxel-based morphometry in order to better characterize the potential confound of gray and white matter differences in the MRS VOI (in addition to simply measuring bulk volume). 5. Conclusion Our findings add to increasing evidence that individuals with autistic spectrum disorders have neuro-developmental abnormalities in the limbic system (possibly reflecting differences in cell packing and lipid membrane density and/or metabolism). Disclosures The authors have no competing interests. Acknowledgements This work was made possible by support from the MRC UK A.I.M.S. network, and the South London and Maudsley NHS Foundation Trust.

84

F.M. O'Brien et al. / Psychiatry Research: Neuroimaging 184 (2010) 77–85

References Abell, F., Krams, M., Ashburner, J., Passingham, R., Friston, K., Frackowiak, R., Happe, F., Frith, C., 1999. The neuroanatomy of autism: a voxel-based whole brain analysis of structural scans. NeuroReport 10, 1647–1651. Adolphs, R., 2002. Neural systems for recognizing emotion. Current Opinion in Neurobiology 12, 169–177. American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders, Fourth edition. APA, Washington, DC. Anderson, A.K., Phelps, E.A., 2001. Lesions of the human amygdala impair enhanced perception of emotionally salient events. Nature 411, 305–309. Aylward, E.H., Minshew, N.J., Goldstein, G., Honeycutt, N.A., Augustine, A.M., Yates, K.O., Barta, P.E., Pearlson, G.D., 1999. MRI volumes of amygdala and hippocampus in non-mentally retarded autistic adolescents and adults. Neurology 53, 2145–2150. Aylward, E.H., Minshew, N.J., Field, K., Sparks, B.G., Singh, N., 2002. Effects of age on brain volume and head circumference in autism. Neurology 59, 175–183. Bachevalier, J., 1994. Medial temporal lobe structures and autism: a review of clinical and experimental findings. Neuropsychologia 32, 627–648. Bachevalier, J., 1996. Brief Report: medial temporal lobe and autism: a putative animal model in primates. Journal of Autism and Developmental Disorders 26, 217–220. Baron-Cohen, S., Ring, H.A., Wheelwright, S., Bullmore, E.T., Brammer, M.J., Simmons, A., Williams, S.C., 1999. Social intelligence in the normal and autistic brain: an fMRI study. European Journal of Neuroscience 11, 1891–1898. Baron-Cohen, S., Ring, H.A., Bullmore, E.T., Wheelwright, S., Ashwin, C., Williams, S.C., 2000. The amygdala theory of autism. Neuroscience and Biobehavioral Reviews 24, 355–364. Bauman, M.L., Kemper, T.L., 1988. Limbic and cerebellar abnormalities—consistent findings in early infantile autism [abstract]. Journal of Neuropathology & Experimental Neurology 47, 369. Bhakoo, K., Craig, T., Styles, P., 2001. Cellular and developmental tissue distribution of NAA catabolic enzyme, aspartoacylase. Insights into NAA Function. International Society for Magnetic Resonance in Medicine 9, 490. Bolton, P.F., 2004. Neuroepileptic correlates of autistic symptomatology in tuberous sclerosis. Mental Retardation and Developmental Disabilities Research Review 10, 126–131. Brandao, L., Domingues, R.C., 2004. MR Spectroscopy of the Brain. Lippincott Williams & Wilkins, Philadelphia. Brothers, L., Ring, B., Kling, A., 1990. Response of neurons in the macaque amygdala to complex social stimuli. Behavioral Brain Research 41, 199–213. Burgess, N., Maguire, E.A., O'Keefe, J., 2002. The human hippocampus and spatial and episodic memory. Neuron 35, 625–641. Clark, J.B., 1998. N-acetyl aspartate: a marker for neuronal loss or mitochondrial dysfunction. Developmental Neuroscience 20, 271–276. Connelly, A., Van Paesschen, W., Porter, D.A., Johnson, C.L., Duncan, J.S., Gadian, D.G., 1998. Proton magnetic resonance spectroscopy in MRI-negative temporal lobe epilepsy. Neurology 51, 61–66. Courchesne, E., 2004. Brain development in autism: early overgrowth followed by premature arrest of growth. Mental Retardation and Developmental Disabilities Research Reviews 10, 106–111. Critchley, H., 2003. Emotion and its disorders. British Medical Bulletin 65, 35–47. Critchley, H.D., Daly, E.M., Bullmore, E.T., Williams, S.C., Van Amelsvoort, T., Robertson, D.M., Rowe, A., Phillips, M., McAlonan, G., Howlin, P., Murphy, D.G., 2000. The functional neuroanatomy of social behavior: changes in cerebral blood flow when people with autistic disorder process facial expressions. Brain 123, 2203–2212. Danielson, E.R., Ross, B., 1999. Magnetic Resonance Spectroscopy Diagnosis of Neurological Diseases. Marcel Dekker, Inc., New York. Davie, C.A., Hawkins, C.P., Barker, G.J., Brennan, A., Tofts, P.S., Miller, D.H., McDonald, W.I., 1994. Serial proton magnetic resonance spectroscopy in acute multiple sclerosis lesions. Brain 117, 49–58. De Long, G.B., 1992. Autism, amnesia, hippocampus, and learning. Neuroscience and Biobehavioral Reviews 16, 63–70. De Long, G.R., Bean, S.C., Brown, F.R., 1981. Acquired reversible autistic disorder in acute encephalopathic illness in children. Archives of Neurology 38, 191–194. Driscoll, I., Hamilton, D.A., Petropoulos, H., Yeo, R.A., Brooks, W.A., Baumgartner, R.N., Sutherland, R.J., 2003. The aging hippocampus: cognitive, biochemical and structural findings. Cerebral Cortex 13, 1344–1351. Dziobek, I., Fleck, S., Rogers, K., Wolf, O.T., Convit, A., 2006. The ‘amygdala theory of autism’ revisited: linking structure to behavior. Neuropsychologia 44, 1891–1899. Endo, T., Shioiri, T., Kitamura, H., Kimura, T., Endo, S., Masuzawa, N., Someya, T., 2007. Altered chemical metabolites in the amygdala–hippocampus region contribute to autistic symptoms of autism spectrum disorders. Biological Psychiatry 62, 1030–1037. Filippi, C.G., Ulu, A.M., Deck, M.D.F., Zimmerman, R.D., Heier, L.A., 2002. Developmental delay in children: assessment with proton MR spectroscopy. American Journal of Neuroradiology 23, 882–888. Gabis, L., Wei, H., Azizian, A., DeVincent, C., Tudorica, A., Kesner-Baruch, Y., Roche, P., Pomeroy, J., 2008. 1H-magnetic resonance spectroscopy markers of cognitive and language ability in clinical subtypes of autism spectrum disorders. Journal of Child Neurology 23, 766–774. Gillberg, C., 1986. Brief Report: onset at age 14 of a typical autistic syndrome: a case report of a girl with herpes simplex encephalitis. Journal of Autism and Developmental Disorders 16, 369–375. Gillberg, C., 1995. The prevalence of autism and autism spectrum disorders. In: Koot, F.C.V.H.M. (Ed.), The Epidemiology of Child and Adolescent Psychopathology. Oxford University Press, Oxford, pp. 227–257.

Grelotti, D.J., Gauthier, I., Schultz, R.T., 2002. Social interest and the development of cortical face specialization: what autism teaches us about face processing. Developmental Psychobiology 40, 213–225. Haznedar, M.M., Buchsbaum, M.S., Wei, T.-C., Hof, P.R., Cartwright, C., Bienstock, C.A., Hollander, E., 2000. Limbic circuitry in patients with autism spectrum disorders studied with positron emission tomography and magnetic resonance imaging. The American Journal of Psychiatry 157, 1994–2001. Helmuth, L., 2003. Fear and trembling in the amygdale. Cognitive Neuroscience 300, 568–569. Herbert, M.R., Ziegler, D.A., Deutsch, C.K., O'Brien, L.M., Lange, N., Bakardjiev, A., Hodgson, J., Adrien, K.T., Steele, S., Makris, N., Kennedy, D., Harris, G.J., Caviness Jr., V.S., 2003. Dissociations of cerebral cortex, subcortical and cerebral white matter volumes in autistic boys. Brain 126, 1182–1192. Hoon, A.H., Reiss, A.L., 1992. The medial temporal lobe and autism — a case report and review. Developmental Medicine and Child Neurology 34, 252–265. Horska, A., Kaufmann, W.E., Brant, L.J., Naidu, S., Harris, J.C., Barker, P.B., 2002. In vivo quantitative proton MRSI study of brain development from childhood to adolescence. Journal of Magnetic Resonance Imaging 15, 137–143. Hoschl, C., Hajek, T., 2001. Hippocampal damage mediated by corticosteroids—a neuropsychiatric research challenge. European Archives of Psychiatry and Clinical Neuroscience 251, II81–II88. Howard, M.A., Cowell, P.E., Boucher, J., Broks, P., Mayes, A., Farrant, A., Roberts, N., 2000. Convergent neuroanatomical and behavioural evidence of an amygdala hypothesis of autism. NeuroReport 11, 2931–2935. Huttenlocher, P.R., 1979. Synaptic density in human frontal cortex: developmental changes and effects of aging. Brain Research 163, 195–205. Huttenlocher, P.R., Dabholkar, A.S., 1997. Regional differences in synaptogenesis in human cerebral cortex. The Journal of Comparative Neurology 387, 167–178. Jung, R.E., Brooks, W.M., Yeo, R.A., Weers, D., Hart, B., Sibbitt, W.L., 1999. Biochemical markers of intelligence: a proton MR spectroscopy study of the normal human brain. Proceedings of the Royal Society B 266, 1375–1379. Kadota, T., Horinouchi, T., Kuroda, C., 2001. Development and aging of the cerebrum: assessment with proton MR spectroscopy. American Journal of Neuroradiology 22, 128–135. Kato, T., Nishina, M., Matsushita, K., Hori, E., Mito, T., Takashima, S., 1997. Neuronal maturation and N-acetyl-L-aspartic acid development in human fetal and child brains. Brain & Development 19, 131–133. Kleinhans, N.M., Schweinsburg, B.C., Cohen, D.N., Muller, R.A., Courchesne, E., 2007. Nacetyl aspartate in autism spectrum disorders: Regional effects and relationship to fMRI activation. Brain Research 1162, 85–97. Kleinhans, N.M., Richards, T., Weaver, K.E., Liang, O., Dawson, G., Aylward, E., 2009. Brief Report: biochemical correlates of clinical impairment in high functioning autism and Asperger's disorder. Journal of Autism and Developmental Disorders 39 (7), 1079–1086. Koller, K.J., Zacsek, R., Coyle, J.T., 1984. N-acetyl-aspartyl-glutamate: regional levels in rat brain and the effects of brain lesions as determined by a new HPLC method. Journal of Neurochemistry 43, 1136–1142. Kostovic, I., Judas, M., Petanjek, Z., Simic, G., 1995. Ontogenesis of goal directed behaviour: anatomo-functional considerations. International Journal of Psychophysiology 19, 85–102. Kreis, R., Ernst, T., Ross, B.D., 1993. Development of the human brain: in vivo quantification of metabolite and water content with proton magnetic resonance spectroscopy. Magnetic Resonance in Medicine 30, 424–437. Lord, C., Rutter, M., Le Couteur, A., 1994. Autism Diagnostic Interview-Revised: a revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders. Journal of Autism and Developmental Disorders 24, 659–685. Lord, C., Risi, S., Lambrecht, L., Cook Jr., E.H., Leventhal, B.L., DiLavore, P.C., Pickles, A., Rutter, M., 2000. The autism diagnostic observation schedule-generic: a standard measure of social and communication deficits associated with the spectrum of autism. Journal of Autism and Developmental Disorders 30, 205–223. Lotspeich, L.J., Kwon, H., Schumann, C.M., Fryer, S.L., Goodlin-Jones, B.L., Buonocore, M.H., Lammers, C.R., Amaral, D.G., Reiss, A.L., 2004. Investigation of neuroanatomical differences between autism and Asperger syndrome. Archives of General Psychiatry 61, 291–298. McAlonan, G.M., Daly, E., Kumari, V., Critchley, H.D., Van Amelsvoort, T., Suckling, J., Simmons, A., Sigmundsson, T., Greenwood, K., Russell, A., Happe, F., Howlin, P., Murphy, D.G., 2002. Brain anatomy and sensorimotor gating in Asperger's syndrome. Brain 125, 1594–1606. Miller, A.K.H., Alston, R.L., Corsellis, J.A., 1980. Variation with age in the volumes of grey and white matter in the cerebral hemispheres of man: measurements with an image analyzer. Neuropathology and Applied Neurobiology 6, 119–132. Mori, K., Hashimoto, T., Harada, M., Yoneda, Y., Yneda, Y., Shimakawa, S., Fujii, E., Yamaue, T., Miyazaki, M., Saijo, T., Kuroda, Y., 2001. Proton magnetic resonance spectroscopy of the autistic brain. No To Hattatsu 33, 329–335. Munson, J., Dawson, G., Abbott, R., Faja, S., Webb, S.J., Friedman, S.D., Shaw, D., Artru, A., Dager, S.R., 2006. Amygdala volume and behavioral development in autism. Archives of General Psychiatry 63, 686–693. Murphy, D.G.M., Critchley, H.D., Schmitz, N., McAlonan, G., van Amelsvoort, T., Robertson, D., Daly, E., Rowe, A., Russell, A., Simmons, A., Murphy, K.C., Howlin, P., 2002. Asperger syndrome: a proton magnetic resonance spectroscopy study of brain. Archives of General Psychiatry 59, 885–891. Nacewicz, B.M., Dalton, K.M., Johnstone, T., Long, M.T., McAuliff, E.M., Oakes, T.R., Alexander, A.L., Davidson, R.J., 2006. Amygdala volume and nonverbal social impairment in adolescents and adult males with autism. Archives of General Psychiatry 63, 1417–1428.

F.M. O'Brien et al. / Psychiatry Research: Neuroimaging 184 (2010) 77–85 Otsuka, H., Harada, M., Mori, K., Hisaoka, S., Nishitani, H., 1999. Brain metabolites in the hippocampus–amygdala region and cerebellum in autism: an H-MRS spectroscopy study. Neuroradiology 41, 517–519. Page, L.A., Daly, E., Schimtz, N., Simmonds, A., Toal, F., Deeley, P.Q., Ambery, F., McAlonan, G.M., 2006. Brief Report: an in-vivo H1 magnetic resonance spectroscopy study of amygdala–hippocampal and parietal regions in autism. The American Journal of Psychiatry 163, 2189–2192. Palmen, S.J.M.C., Van Engeland, H., Hof, P.R., Schmitz, C., 2004. Neuropathological findings in autism. Brain 127, 2572–2583. Palmen, S.J.M.C., Hulshoff Pol, H.E., Kemner, C., Schnack, H.G., Sitskoorn, M.M., Appels, M.C.M., Kahn, R.S., Van Engeland, H., 2005. Brain anatomy in non-affected parents of Autistic probands: a MRI study. Psychological Medicine 35, 1411–1420. Pfefferbaum, A., Adalsteinsson, E., Spielman, D., Sullivan, E.V., Lim, K.O., 1999. In vivo spectroscopic quantification of the N-acetyl moiety, creatine, and choline from large volumes of brain gray and white matter: effects of normal aging. Magnetic Resonance in Medicine 41, 276–284. Pierce, K., Muller, R.A., Ambrose, J., Allen, G., Courchesne, E., 2001. Face processing occurs outside the fusiform “face area” in autism: evidence from functional MRI. Brain 124, 2059–2073. Piven, J., Bailey, J., Ranson, B.J., Arndt, S., 1998. No difference in hippocampus volume detected on magnetic resonance imaging in autistic individuals. Journal of Autism and Developmental Disorders 28, 105–110 correction, 28, 271. Provencher, S.W., 1993. Estimation of metabolite concentrations from localized in vivo proton NMR spectra. Magn Reson Med 30, 672. Pruessner, J.C., Collins, D.L., Pruessner, M., Evans, A.C., 2001. Age and gender predict volume decline in the anterior and posterior hippocampus in early adulthood. The Journal of Neuroscience 21, 194–200. Rajah, M.N., Bastianetto, S., Bromley-Brits, K., Cools, R., D'Esposito, M., Grady, C.L., Poirier, J., Quirion, R., Raz, N., Rogaeva, E., Song, W., Pruessner, J., 2009. Biological changes associated with healthy versus pathological aging: a symposium review. Ageing Research Reviews 8, 140–146. Richardson, M.P., Strange, B.A., Dolan, R.T., 2004. Encoding of emotional memories depends on the amygdala and hippocampus and their interactions. Nature Neuroscience 7, 278–285. Richter-Levin, G., 2004. The amygdala, the hippocampus, and emotional modulation of memory. The Neuroscientist 10, 31–39. Rojas, D.C., Smith, J.A., Benkers, T.L., Camou, S.L., Reite, M.L., Rogers, S.J., 2004. Hippocampus and amygdala volumes in parents of children with autistic disorder. The American Journal of Psychiatry 161, 2038–2044. Saitoh, O., Courchesne, E., Egaas, B., Lincoln, A.J., Schreibman, L., 1995. Cross-sectional area of the posterior hippocampus in autistic patients with cerebellar and corpus callosum abnormalities. Neurology 45, 317–324. Saitoh, O., Karns, C.M., Courchesne, E., 2001. Development of the hippocampal formation from 2 to 42 years: MRI evidence of smaller area dentata in autism. Brain 124, 1317–1324.

85

Schuff, N., Amend, D.L., Knowlton, R., Norman, D., Fein, G., Weiner, M.W., 1999. Agerelated metabolite changes and volume loss in the hippocampus by magnetic resonance spectroscopy and imaging. Neurobiology of Aging 20, 279–285. Schultz, R.T., 2005. Developmental deficits in social perception in autism: the role of the amygdale and fusiform face area. International Journal of Developmental Neuroscience 23, 125–141. Schumann, C.M., Amaral, D.G., 2006. Stereological analysis of amygdala neuron number in Autism. The Journal of Neuroscience 26, 7674–7679. Schumann, C.M., Hamstra, J., Goodlin-Jones, B.L., Lotspeich, L.J., Kwon, H., Buonocore, M.H., Lammers, C.R., Reiss, A.L., Amaral, D.G., 2004. The amygdala is enlarged in children but not adolescents with autism; the hippocampus is enlarged at all ages. The Journal of Neuroscience 24, 6392–6401. Soares, D.P., Law, M., 2009. Magnetic resonance spectroscopy of the brain: review of metabolites and clinical applications. Clinical Radiology 64, 12–21. Sparks, B.F., Friedman, S.D., Shaw, D.W., Aylward, E.H., Echelard, D., Artru, A.A., Maravilla, K.R., Giedd, J.N., Munson, J., Dawson, G., Dager, S.R., 2002. Brain structural abnormalities in young children with autism spectrum disorders. Neurology 59, 184–192. StataCorp., 2009. Stata Statistical Software: Release 11. College Station, TX: StataCorp LP. Strange, B., Dolan, R., 1999. Functional segregation within the human hippocampus. Molecular Psychiatry 4, 508–511. Suzuki, K., Nishimura, K., Sugihara, G., Nakamura, K., Tsuchiya, K.J., Matsumoto, K., Takebayashi, K., Isoda, H., Sakahara, H., Sugiyama, T., Tsujii, M., Takei, N., Mori, N., 2010. Metabolite alterations in the hippocampus of high-functioning adult subjects with autism. International Journal of Neuropsychopharmacology 13, 529–534. Toft, P.B., Lethm, H., Lou, H.C., Pryds, O., Henriksen, O., 1994. Metabolite concentrations in the developing brain estimated with proton MR spectroscopy. Journal of Magnetic Resonance Imaging 4, 674–680. Vermathen, P., Ende, G., Laxer, K.D., Knowlton, R.C., Matson, G.B., Weiner, M.W., 1997. Hippocampal N-acetylaspartate in neocortical epilepsy and mesial temporal lobe epilepsy. Annals of Neurology 42, 194–199. Wechsler, D., 1981. Wechsler Adult Intelligence Scale-Revised (WAIS-R). The Psychological Corporation, New York. Wechsler, D., 1991. Weschler Intelligence Scale for Children, 3rd edition. The Psychological Corporation, New York. (WISC-III). Wing, L., 1993. The definition and prevalence of autism: a review. European Child & Adolescent Psychiatry 2, 61–74. Wing, L., 1996. Autistic spectrum disorders. BMJ 312, 327–328. World Health Organisation, 1993. The ICD-10 Classification for Mental and Behavioural Disorders: Clinical Description and Diagnostic Guidelines. WHO, Geneva, Switzerland. Zeegers, M., van der Grond, J., van Daalen, E., Buitelaar, J., van Engeland, H., 2007. Proton magnetic resonance spectroscopy in developmentally delayed young boys with or without autism. Journal of Neural Transmission 114, 289–295.