904
Schizophrenia Bulletin vol. 40 no. 4 pp. 904–913, 2014 doi:10.1093/schbul/sbt093 Advance Access publication July 16, 2013
© The Author 2013. Published by Oxford University Press on behalf of the Maryland Psychiatric Research Center. All rights reserved. For permissions, please email: journals.permissions@oup.com
Altered Striatal Functional Connectivity in Subjects With an At-Risk Mental State for Psychosis
Orwa Dandash1, Alex Fornito1–4, Jimmy Lee5,6, Richard S. E. Keefe7,8, Michael W. L. Chee8, R. Alison Adcock9, Christos Pantelis1,10, Stephen J. Wood*,1,11, and Ben J. Harrison1
1Melbourne Neuropsychiatry Centre, Department of Psychiatry, The University of Melbourne, Melbourne, Australia; 2Centre for Neural Engineering, The University of Melbourne, Melbourne, Australia; 3The Victorian Research Laboratory National ICT Australia Ltd, Victoria, Australia; 4Monash Clinical and Imaging Neuroscience, School of Psychology and Psychiatry and Monash Biomedical Imaging, Monash University, Clayton, Australia; 5Department of General Psychiatry 1 and Research Division, Institute of Mental Health, Buangkok, Singapore; 6Office of Clinical Sciences, Graduate Medical School, Duke-National University of Singapore, Singapore, Singapore; 7Psychiatry and Behavioral Sciences, Duke University Medical Center, Durham, NC; 8Neuroscience and Behavioral Disorders Program, Graduate Medical School, Duke-National University of Singapore, Singapore, Singapore; 9Department of Psychiatry and Behavioral Sciences and Center for Cognitive Neuroscience, Duke University, Durham, NC; 10Melbourne Health, Melbourne, Australia; 11School of Psychology, University of Birmingham, Edgbaston, Birmingham, UK
*To whom correspondence should be addressed; School of Psychology, University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK; tel: 44 121 414 4917, fax: 44 121 414 4897, e-mail: s.j.wood@bham.ac.uk
Recent functional imaging work in individuals experienc- ing an at-risk mental state (ARMS) for psychosis has implicated dorsal striatal abnormalities in the emergence of psychotic symptoms, contrasting with earlier findings implicating the ventral striatum. Our aims here were to characterize putative dorsal and ventral striatal circuit- level abnormalities in ARMS individuals using resting- state functional magnetic resonance imaging (fMRI) and to investigate their relationship to positive psychotic symptoms. Resting-state fMRI was acquired in 74 ARMS subjects and 35 matched healthy controls. An established method for mapping ventral and dorsal striatal func- tional connectivity was used to examine corticostriatal functional integrity. Positive psychotic symptoms were assessed using the Comprehensive Assessment of At-Risk Mental State and the Positive and Negative Syndrome Scale. Compared with healthy controls, ARMS subjects showed reductions in functional connectivity between the dorsal caudate and right dorsolateral prefrontal cortex, left rostral medial prefrontal cortex, and thalamus, and between the dorsal putamen and left thalamic and lenticu- lar nuclei. ARMS subjects also showed increased func- tional connectivity between the ventral putamen and the insula, frontal operculum, and superior temporal gyrus bilaterally. No differences in ventral striatal (ie, nucleus accumbens) functional connectivity were found. Altered functional connectivity in corticostriatal circuits were significantly correlated with positive psychotic symptoms. Together, these results suggest that risk for psychosis is
mediated by a complex interplay of alterations in both dorsal and ventral corticostriatal systems.
Key words: ARMS/fMRI/resting state/striatum
Introduction
The onset of psychosis is commonly preceded by a pro- dromal period characterized by subthreshold symp- tomatology and functional decline.1 These symptoms have been construed as reflecting an at-risk mental state (ARMS), which is associated with an enhanced risk of converting to frank psychosis within 2 years.2,3 A number of anatomical, functional, and neurochemi- cal brain changes thought to be crucial to the patho- genesis of the disease have been associated with the ARMS.4–6 Chief among these is elevated dopamine transmission in the striatum,7,8 a brain region rich in dopamine D2 receptors targeted by most antipsychotic drugs.9,10
Many pathophysiological models of psychosis ascribe a prominent role to the striatum, comprising the caudate and putamen, given its rich interconnectivity with other regions known to be dysfunctional in psychosis, such as the prefrontal cortex.11–14 In particular, animal studies have emphasized dysfunction of a ventral (limbic) circuit, linking ventral striatum (nucleus accumbens) with medial prefrontal cortex, a system implicated in the mesolimbic dopaminergic pathway.15,16
905
Altered Striatal Connectivity in ARMS Subjects
Recently, however, in vivo positron emission tomo- graphy research has demonstrated elevated dopaminergic activity in the dorsal but not the ventral striatum of patients with established schizophrenia17 and ARMS subjects.8 These elevations correlate with the severity of positive psychotic symptoms8 and prefrontal activation18,19 in the latter group. Thus, in contrast to animal models,15,16 human data implicate dorsal striatal dysfunction in the emergence of psychosis. They also suggest that abnormalities may not be restricted to the striatum but propagate throughout interconnected corticostriatal circuitry, consistent with the idea that psychotic disorders arise as a result of aberrant brain connectivity or “dysconnectivity.”20–22
Despite this possibility, no study to date has directly examined corticostriatal dysfunction in ARMS individu- als from a systems-level perspective in order to character- ize putative, circuit-based risk biomarkers for psychosis. Resting-state functional magnetic resonance imaging (fMRI) provides a powerful probe of the functional integrity of corticostriatal circuitry by mapping networks of brain regions showing coherent fluctuations of spon- taneous neural activity.23–25 These fluctuations are under strong genetic control26,27 and are thought to represent an intrinsic property of brain functional organization.20,28 In this study, we used this technique to investigate the functional connectivity of dorsal and ventral striatal regions (caudate nucleus and putamen) with the rest of the brain in a well-characterized, relatively large sample of ARMS subjects recruited in a community-based set- ting in Singapore.29
The aims of this study were 2-fold. First, we sought to systematically examine functional connectivity within dorsal and ventral corticostriatal circuits to evalu- ate whether the ARMS is associated primarily with alterations in the dorsal circuit, ventral circuit, or both. Second, we sought to characterize circuit-level abnormal- ities related to psychotic symptoms, independently of any covariance with depression and anxiety, which are often present in ARMS individuals. Based on recent neuroim- aging work,8,18 we hypothesized that participants would show altered striatal functional connectivity in the dorsal circuit compared with healthy controls. We also hypoth- esized that altered functional connectivity would be asso- ciated with more severe psychotic symptoms.
Methods
Participants
Eighty help-seeking individuals were recruited from the Longitudinal Youth At Risk study,29 a research study based at the Institute of Mental Health (IMH), Singapore. Individuals were recruited from psychiat- ric clinics at IMH, the Singapore Armed Forces, and from community mental health services run by trained
counselors. Comprehensive Assessment of At-Risk Mental State (CAARMS)30 was used to establish at- risk status for the development of psychosis. Included individuals had no history of psychotic disorder, neu- rological disorder, serious medical disorders, or men- tal retardation and no current substance use. Baseline symptoms were assessed with the Positive and Negative Syndrome Scale (PANSS) of schizophrenia,31 the Calgary Depression Scale for Schizophrenia (CDSS),32 and the Beck Anxiety Inventory (BAI).33 Overall func- tioning was assessed using the Global Assessment of Functioning34 (see online supplementary table 1 for fur- ther information). All assessments were carried out by trained psychologists (N = 14, CAARMS interrater reli- ability κ = 0.99).
At intake, 25 ARMS subjects were receiving antide- pressants (tricyclic or serotonin reuptake inhibitors); 2 were receiving antianxiety treatment (benzodiazepines); 12 were receiving both antidepressants and benzodiaz- epines; and 2 received low-dose antipsychotic treatment (chlorpromazine 50 mg and seroquel 25 mg). Repeating analyses after excluding the last 2 subjects did not alter the findings, so we report on the full sample here. Subsequent to MRI scan, 6 ARMS subjects made a transition to psy- chosis as outlined by the CAARMS (see online supple- mentary table 2 for further information).
Forty subjects with a similar sociodemographic background to ARMS subjects matched for age, gender, handedness, and educational attainment were recruited as healthy controls. Sample characteristics were compared between the groups (table  1) using univariate ANOVA in the Statistical Package for the Social Sciences (v.20). Educational attainment was assessed by the Primary School Leaving Examination, a standardized multidisciplinary test of scholastic achievement. Exclusion criteria were history of severe head injury, personal history of psychotic or neurologic disorders, and substance or alcohol dependence assessed with the Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, fourth edition.35 All participants gave their written and informed consent approved by the ethics review board of the Singaporean National Healthcare Group.
Image Acquisition
T2*-weighted echo-planar images were acquired for each subject in the resting condition with a 3.0 T MRI scan- ner (Siemens Magnetom, Trio Tim) that is equipped with a 12-channel array coil. Subjects were instructed to lie still with their eyes closed and not to fall asleep. Functional volumes were acquired in 3 mm3 voxel size, with a time to repetition (TR) of 2000 ms, time to echo (TE) of 30 ms, flip angle of 90° in 64 × 64 matrix size, and 192 mm field of view (FOV). For 7 control subjects, 320 volumes were acquired with 28 slices each. For the remaining participants, 240 volumes comprising 36 slices
906
O. Dandash et al
each were obtained. Repeating analyses after excluding the 7 subjects did not significantly alter the findings, so we report on the full sample here. For registration of functional images, high-resolution anatomical T1 images were acquired using a 3-dimensional magnetic-prepared rapid gradient echo (MPRAGE) sequence and multiecho MPRAGE. One hundred and ninety-two contiguous sag- ittal slices of 1.0 mm thickness using a TR of 2300 ms, TE of 2980 ms, flip angle of 9°, and a FOV of 256 mm in 256 × 256 matrix were acquired with a voxel size of 1.0 mm3.
Preprocessing
A validated resting-state functional connectivity mapping procedure was used to characterize ventral and dorsal corticostriatal systems via primary seed regions of interest located in ventral and dorsal areas of the caudate nucleus and putamen (see Di Martino et  al36 for anatomical delineation of these regions and Harrison et  al25 and online supplementary text for a full description of the methodology implemented herein).
First-Level, Within-Subjects Analysis
Functional connectivity maps were estimated for each participant by including the striatal region time series and nuisance signals as predictors of interest/no-interest
in whole-brain, linear regression analyses in SPM8 for each hemisphere. Prior to model estimation, a high-pass filter set at 128 seconds was used to remove low-frequency drifts. Each of the 3 nuisance covariates were orthogo- nalized (using an iterative Gram-Schmidt method) and then removed from each seed’s time series along with 6 head-motion parameters (3 translation and 3 rotation) by linear regression, resulting in a general linear model that comprised “noise-cleaned” regions and 9 nuisance vari- ables. Contrast images were generated for each partici- pant by estimating the regression coefficient between all brain voxels and each region’s time series.
Second-Level, Between-Group Analysis
For each striatal region, participants’ contrast images were included in a random effects 2 × 2 factorial design (Group [control, ARMS] by Hemisphere [left, right]) in SPM8. Within-group statistical maps were thresholded at a false discovery rate of P < .05 for the whole-brain volume (figure 1). Between-group effects (main effects of group and group × hemisphere interactions) were mapped by implicitly masking t contrasts (1 tailed) with a global conjunction of the within-subjects effect calculated for both groups for the combined left and right hemispheres. Nuisance covariates included age, gender, and 4 addi- tional summary measures of head motion to further
Table 1. Demographics and Clinical Characteristics of the At-Risk Mental State (ARMS) Group and the Healthy Comparison Group
ARMS Subjects (N = 74)
Characteristics Mean SD Mean SD t/χ2 p
Age (y) 21.4 3.57 22.8 3.94 −1.839 .068 Educational attainment 196.5 47.3 206.7 27.1 −1.40 .166
Gender N % N % χ2 p
Male 47 67 21 60 0.599 .439 Female 27 33 14 40
Handednessa N % N % χ2 p
Left 4 5.7 0 0 4.921 .178 Mixed 6 8.6 1 2.9 Right 60 85.7 34 97.1
Clinical scoresa Mean SD — — — —
CAARMS positive symptoms scaleb
Unusual thought content 4.0 3.4 — — — — Nonbizarre ideation 5.5 3.2 — — — — Perceptual abnormality 4.3 3.3 — — — — Disorganized speech 1.7 2.8 — — — — PANSS PANSS total 49.6 11.5 — — — — PANSS positive 11.0 2.8 — — — —
Note: CAARMS, Comprehensive Assessment of At-Risk Mental State; PANSS, Positive and Negative Syndrome Scale. aThe data of 4 (5.4%) ARMS subjects were not available at the time of the analysis. bSummed across intensity and frequency measures.
907
remove residual head-motion effects37: (1) the number of significant micromovements (instances of >0.10 mm rela- tive displacement between adjacent volumes); (2) mean head displacement; (3) maximum head displacement; and (4) mean head rotation. Between-group statistical maps were thresholded using a P < .05 ([family-wise error] FWE) cluster-wise corrected threshold determined using a permutation-based framework38 (1000 permutations with a cluster-forming threshold of P < .01 uncorrected), as implemented in the REST toolbox.39
Symptom Correlates of Striatal Functional Connectivity
We investigated associations between striatal functional connectivity and measures of positive symptom severity taken from the CAARMS and PANSS positive symptom scale in ARMS subjects (N = 70). Secondary associations
with CDSS and BAI scales were also performed to test the specificity of findings to positive symptoms. These scores were entered as covariates into separate general linear models, which also included age, gender, and the 4 head motion derivatives as nuisance covariates. Correlations between clinical ratings and connectivity measures were masked by the between-group difference observed between ARMS and control subjects (figure  2) to identify the symptom correlates of striatal functional dysconnectivity observed in ARMS subjects. All results were displayed at P < .05 (FWE) cluster-wise corrected, with appropriate correction for the search volume employed.
Results
Across groups, functional connectivity of the dorsal and ventral striatum (figure  1) recapitulated previous
Fig. 1. Significant within-group (seed effect) functional connectivity maps of the dorsal caudate (DC), ventral striatum/nucleus accumbens (VS), dorsal putamen (DP), and ventral putamen (VP) seeds (in blue). Green indicates healthy comparison subjects, while red indicates at-risk mental state subjects and yellow indicates areas of overlap; R, right hemisphere; L, left hemisphere. Sagittal slices are displayed at x = ±6 (DC and VS); x = −3 and x = 4 (DP); x = ±8 (VP). Results are displayed at P < .05 (false discovery rate) corrected.
908
O. Dandash et al
results25,36 and was consistent with known anatomical con- nectivity.11,12 When functional connectivity was compared between groups, significant differences were observed for the dorsal caudate, dorsal putamen, and ventral putamen seed regions (figure 2; table 2). With regard to the dor- sal caudate, ARMS subjects demonstrated significantly reduced functional connectivity with the right dorsolat- eral prefrontal cortex, left rostral medial prefrontal cor- tex, and thalamus bilaterally (figure 2A). For the dorsal putamen seed region, ARMS subjects demonstrated reduced functional connectivity with the left lenticular nucleus and the medial dorsal, ventral lateral, and ven- tral anterior thalamic nuclei (figure 2B). ARMS subjects demonstrated increased functional connectivity between ventral putamen seeds and lateral superior temporal gyrus, frontal operculum and the insula bilaterally (fig- ure 2C). There were no significant group differences for the ventral striatal (nucleus accumbens) seed, nor were there any significant group × hemisphere interactions for any seed region.
Brain-Behavior Associations
In ARMS participants, higher scores on the CAARMS nonbizarre ideation subscale predicted lower functional connectivity between the dorsal caudate and the left ros- tromedial prefrontal cortex (figure 3A). Higher scores on the perceptual abnormality subscale predicted lower func- tional connectivity between the dorsal caudate and right dorsolateral prefrontal cortex (figure 3A). Additionally, higher unusual thought content (CAARMS) and PANSS positive scores predicted lower functional connectiv- ity between the dorsal putamen and the medial dorsal, ventral lateral, and ventral anterior thalamic nuclei (fig- ure 3B). Finally, higher unusual thought content scores (CAARMS) predicted higher functional connectivity between ventral putamen and language areas, superior temporal gyrus and the insula bilaterally, whereas higher PANSS positive scores predicted higher functional con- nectivity with the left inferior frontal gyrus (figure 3C).
No further significant associations with CAARMS or PANSS positive scores were identified.
Repeating the analysis while covarying for CDSS and BAI ratings did not significantly alter the correlation between corticostriatal functional connectivity and positive symptom severity. To control for medication effects, we ran a separate analysis in which we investigated differences in functional connectivity of striatal seeds (that showed significant between-group differences) between ARMS subjects receiving antidepressants (n = 34) and those who did not (n = 36), masked by the observed pattern of group differences. No significant differences were found. To further exclude the effect of diagnosis and/or treatment, we repeated the between- group analysis after excluding subjects with Axis-I diagnosis of depressive and/or anxiety disorder in addition to subjects receiving antidepressants and/or benzodiazepines. Despite lower power, the same effects were observed (P < .01 uncorrected), indicating our results are not attributable to diagnosis or treatment effects (see online supplementary figure 1).
Discussion
It remains controversial whether psychotic symptoms emanate from primary alterations of dorsal or ventral corticostriatal circuitry. To address this question, we systematically characterized the functional integrity of dorsal and ventral corticostriatal circuits at a systems level with resting-state fMRI in a large sample of ARMS subjects experiencing attenuated psychotic symptoms. ARMS subjects showed a selective reduction of functional connectivity in dorsal corticostriatal circuits and increased functional connectivity between ventral putamen and brain regions largely involved in language, including the superior temporal and transverse gyri, insula, and frontal operculum bilaterally. No functional connectivity changes were found for the ventral striatum/ nucleus accumbens seeds. The magnitude of both dorsal and ventral circuit alterations correlated significantly with
Fig. 2. z score statistical maps of significant between-group differences in functional connectivity of the (A) dorsal caudate (DC); (B) dorsal putamen (DP); and (C) ventral putamen (VP) seeds. Right hemisphere to the right of the image. At-risk mental state (N =7 0), healthy controls (N = 35). Results are displayed at P < .05 (FWE) cluster-wise corrected. See table 2 for more information.
909
Altered Striatal Connectivity in ARMS Subjects
the severity of positive psychotic symptoms as measured by the positive symptoms scales of the CAARMS and the PANSS.
The alterations of corticostriatal functional connectivity in ARMS subjects are in strong agreement with accumulat- ing evidence of functional and neurochemical changes in these structures prior to the onset of frank psychosis8,40–42
(reviewed in Fusar-Poli, McGuire, Borgwardt5) and their general role in the pathophysiology of schizophrenia.13,43 In addition, our findings confirm and extend the find- ings of resting-state fMRI studies in first episode44 and established schizophrenia patients45 (reviewed in Fornito and Bullmore20 and Pettersson-Yeo et  al46) by show- ing that functional dysconnectivity is evident in subjects
Table 2. Brain Regions Demonstrating Significant Between-Group Differences in Functional Connectivity and Association With Positive Psychotic Symptoms in ARMS Subjects
Main Effect Anatomical Region Hemisphere MNI Coordinates (x, y, z) z (df = 1,208) Voxels
Dorsal caudate Rostral medial prefrontal cortex
Left −8, 74, 8 3.34 215
Dorsal lateral prefrontal cortex
Right 26, 18, 62 3.60 298
Thalamus Left −6, −12, 14 4.12 828 Thalamus Right 8, −22, 12 4.30
Dorsal putamen Lentiform nucleus Left −26, −8, 0 4.19 580 Thalamus Left −12, −10, 16 3.07
Ventral putamen Operculum Right 58, −14, 10 4.59 629 Inferior postcentral gyrus Left −58, −24, 24 3.83 238 Inferior frontal gyrus Left −48, 18, −4 3.77 316 Operculum Left −56, −6, 6 3.64 239 Insula Left −44, −6, −4 3.23
Positive Symptoms Association Anatomical Region Hemisphere
MNI Coordinates (x, y, z) z (df = 1,131) Voxels
Dorsal caudate Dorsal lateral prefrontal cortex
Right 32, 38, 44 3.10 35
Rostral medial prefrontal cortex
Left −10, 72, 8 3.04 17
Dorsal putamen Thalamus Left −12, −10, 16 3.40 27 Thalamus Left −12, −12, 20 3.0 13
Ventral putamen Insula/operculum Right 48, −4, 4 3.32 8 Insula/operculum Left −44, −4, 2 2.72 9 Inferior frontal gyrus Left −56, 10, 6 3.15 27
Note: ARMS, at-risk mental state; FWE, family-wise error. Results are viewed at P < .05 (FWE) cluster-wise corrected.
Fig. 3. z scores statistical maps of significant brain-behavior association between estimates of functional connectivity of striatal seeds and positive psychotic symptoms on the Comprehensive Assessment of At-Risk Mental States (unusual thought contents, nonbizarre ideation, and perceptual abnormality) and the PANSS positive scales in at-risk mental state subjects (N = 70). Results are masked with the between-group difference (main effect) mask and displayed at a P < .05 (FWE) cluster-wise corrected. See table 2 for more information and online supplementary figure 2 for scatter plots.
910
experiencing positive psychotic symptoms characteristic of the psychosis prodrome and that it is not a secondary effect of antipsychotic treatment, comorbidity, or illness chronicity.
The involvement of dorsal corticostriatal circuitry is consistent with recent reports indicating that the ARMS is associated with elevation of dopamine transmission specifically in the dorsal, so called “associative,” regions of the caudate nucleus.8,18,19 In past work, this elevation was associated with severity of psychotic symptoms8 and altered prefrontal function.18,19 Our findings extend these results by demonstrating a functional dysconnection in the form of reduced connectivity between dorsal striatal and prefrontal regions while also highlighting an impor- tant role for the ventral putamen in this circuit-level abnormality.
It is as yet unclear whether altered striatal dopamine is a cause or consequence of the apparent functional dyscon- nectivity reported herein. Certainly, there is evidence that pharmacological manipulation of dopamine transmission can alter brain functional connectivity. In one study, acute depletion of dopamine precursors (tyrosine and phenyl- alanine) in healthy controls led to a reduction in fronto- striatal functional connectivity,47 whereas administration of l-Dopa, another dopamine precursor, enhanced it.48 Neither of these agents represent pharmacological mod- els of psychosis, thus the effects of these drugs need not necessarily mimic our findings, though they do suggest that general increase in dopamine leads to increased fron- tostriatal functional connectivity. This observation seems to be at odds with our report of reduced functional con- nectivity, coupled with other findings of increased striatal dopamine in ARMS subjects.7,8 However, it is well known that there is an inverted U-shaped relationship between dopamine levels and optimal cortical function and behav- ior,49,50 and well-established models posit that deviations from optimal dopamine levels in either direction (ie, too much or too little) will increase the noise of informa- tion processing in neural systems.51,52 Adding noise to 2 variables (eg, regional fMRI time series) will generally decrease the correlation between them. Thus, both abnor- mal increases and decreases in dopamine could plausibly lead to reduced functional connectivity.
Accumulating neuroimaging evidence in subjects at clinical high risk for psychosis shows significant neuroanatomical alterations in cortical regions before the onset of psychosis.6,53,54 Similar changes have generally not been observed in striatal55 or thalamic regions (reviewed in Fusar-Poli et al4 and Fusar-Poli, McGuire, Borgwardt5). The strong connectivity of cortical afferents with the striatum11,12 and their role in controlling subcortical dopaminergic pathways56,57 imply that cortical alterations may precede and influence striatal dopamine transmission as a result of disrupted functional connectivity between these regions. Importantly, disinhibition of subcortical dopamine and the concomitant emergence of psychotic
symptoms have historically been viewed as the final step in a common pathway arising from aberrant cortical afferents.58–60 It is therefore plausible that disruption of corticostriatal functional connectivity is driven by cortical abnormality. Our observed association between corticostriatal dysconnectivity and positive psychotic symptoms (figure 3), as well as other work61–64 implicating the prefrontal cortex in mediating aspects of psychotic symptoms, is in support of this notion, though further work is required to establish causal influences of this perturbation.
Alterations in striatothalamic connectivity are con- sistent with models implicating cortical disinhibition of subcortical dopamine in the emergence of psychotic symptoms.60,65 These models posit that disinhibition of subcortical dopamine increases the flow of sensory information to the thalamus, which results in a failure of the thalamus to “filter out” irrelevant stimuli before they reach the cortex, thus predisposing an individual to psychotic symptoms. Given that the thalamus receives the bulk of striatal output via GABAergic projections to the medial dorsal, ventral anterior, and ventral lateral nuclei,12 it was perhaps not surprising that ARMS sub- jects experiencing more severe positive psychotic symp- toms showed disrupted functional connectivity between the dorsal putamen and these nuclei (figure 3B).
The selective localization of striatothalamic dyscon- nectivity to the left hemisphere is also in agreement with previous reports showing neurochemical abnormalities in the left thalamus in ARMS subjects.41,66 Importantly, reduced thalamic glutamate correlated with altered activation and electrophysiological response in the pre- frontal cortex and superior temporal gyrus, as well as reduced gray matter volume in the lateral temporal, inferior frontal, and insular cortices,67 brain regions that showed increased functional connectivity with the ven- tral putamen (figure  2C). These findings suggest that the abnormal increase in functional connectivity of the ventral putamen may reflect an abnormal increase in cortical glutamatergic tone that could potentially be caused by thalamocortical disinhibition of pyramidal cells.65,68 Alternatively, the ventral putamen could be in overdrive to compensate for a primary dorsal corticos- triatal abnormality. Recent evidence in ARMS subjects points to alteration in ventral striatal and medial tempo- ral areas function that is not present in control subjects,69 and our more recent observation in unaffected siblings of first-episode psychosis patients shows reduced dorsal and increased ventral corticostriatal connectivity.70 Thus, dorsal frontostriatal hypoconnectivity may be a generic risk biomarker for psychosis, while ventral system hyper- connectivity may emerge as a compensatory response or secondary consequence.
Despite the relatively large sample of ARMS sub- jects, between-group differences in functional connec- tivity may be an underestimate of the actual underlying
911
Altered Striatal Connectivity in ARMS Subjects
disturbance given the smaller sample size of healthy controls. A  major strength, however, is the absence of substance abuse problems due to the generally low prevalence of substance use in Singapore71–73 (see online supplementary table  1 for more details). In particular, cannabis use, which is often more common in ARMS individuals,74–76 can influence dopamine levels77,78 and corticostriatal function79 and thus represents a major confound in imaging studies. Though nearly half of our ARMS sample were taking antidepressants, the asso- ciation we observed between corticostriatal functional connectivity and psychotic symptoms was evident inde- pendent of depressive symptom, and there were no dif- ferences between participants on/off these medications. These findings support a specific involvement of corti- costriatal dysconnectivity in the emergence of psychotic symptoms. The lack of outcome data in our analyses prevents any inferences concerning the specific risk for psychosis in our sample. Nonetheless, the robust associa- tion observed with positive symptom severity in individ- uals deemed to be at risk for psychosis, as defined using extensively validated operational criteria,3,30,80 suggests that this circuit-level disruption is intimately related to the expression of psychotic symptoms. This assertion is further supported by our recent finding of similar dorsal circuit changes in patients with first-episode psychosis and their unaffected relatives.70 Six individuals in our ARMS group have since made the transition to psycho- sis, and we are presently following up the remainder of participants to clarify clinical outcomes and more accu- rately characterize risk levels.
In summary, our results indicate that the ARMS is associated with a disruption of dorsal and ventral cor- ticostriatal functional connectivity and that this disrup- tion correlates with the severity of positive psychotic symptoms. These findings converge with recent evidence that dopamine elevations are present specifically in dorsal striatal regions of ARMS subjects8 and patients with schizophrenia17 while also highlighting a role for ventral corticostriatal circuitry. Further work examin- ing the utility of this phenotype in predicting transition to psychosis will be critical in determining its clinical relevance.
Supplementary Material
Funding
National Research Foundation Singapore under the National Medical Research Council Translational and Clinical Research Flagship Program (NMRC/ TCR/003/2008). The Australian Postgraduate Award (2010); the Melbourne Neuropsychiatry Centre Research
Higher Degree Award (2010) to O.D.; CR Roper Fellowship and National Health and Medical Research Council Project Grant (1050504 to A.F.); Department of Veteran’s Affair; Feinstein Institute for Medical Research; GlaxoSmithKline; National Institute of Mental Health; Novartis; Psychogenics; Research Foundation for Mental Hygiene, Inc; the Singapore National Medical Research Council (to R.S.E.K.); National Health and Medical Research Counsel of Australia Senior Principal Research Fellowship (628386 to C.P.); National Health and Medical Research Counsel of Australia Program Grant (566529 to C.P.); National Health and Medical Research Council of Australia Clinical Career Development Award (628509 to B.J.H.).
Acknowledgments
The authors have declared that there are no conflicts of interest in relation to the subject of this study.
References
1. Yung AR, McGorry PD, McFarlane CA, Jackson HJ, Patton GC, Rakkar A. Monitoring and care of young people at incipient risk of psychosis. Schizophr Bull. 1996;22:283–303.
2. Cannon TD, Cadenhead K, Cornblatt B, et  al. Prediction of psychosis in youth at high clinical risk: a multisite lon- gitudinal study in North America. Arch Gen Psychiatry. 2008;65:28–37.
3. Yung AR, Nelson B, Stanford C, et al. Validation of “prodro- mal” criteria to detect individuals at ultra high risk of psycho- sis: 2 year follow-up. Schizophr Res. 2008;105:10–17.
4. Fusar-Poli P, Borgwardt S, Crescini A, et al. Neuroanatomy of vulnerability to psychosis: a voxel-based meta-analysis. Neurosci Biobehav Rev. 2011;35:1175–1185.
5. Fusar-Poli P, McGuire P, Borgwardt S. Mapping prodro- mal psychosis: a critical review of neuroimaging studies. Eur Psychiatry. 2012;27:181–191.
6. Pantelis C, Velakoulis D, McGorry PD, et  al. Neuro- anatomical abnormalities before and after onset of psychosis: a cross-sectional and longitudinal MRI comparison. Lancet. 2003;361:281–288.
7. Howes OD, Bose SK, Turkheimer F, et  al. Dopamine synthesis capacity before onset of psychosis: a prospec- tive [18F]-DOPA PET imaging study. Am J Psychiatry. 2011;168:1311–1317.
8. Howes OD, Montgomery AJ, Asselin MC, et  al. Elevated striatal dopamine function linked to prodromal signs of schizophrenia. Arch Gen Psychiatry. 2009;66:13–20.
9. Ginovart N, Kapur S. Role of dopamine d(2) recep- tors for antipsychotic activity. Handb Exp Pharmacol. 2012;212:27–52.
10. Laruelle M. Imaging dopamine transmission in schizophrenia. A review and meta-analysis. Q J Nucl Med. 1998;42:211–221.
11. Alexander GE, DeLong MR, Strick PL. Parallel organiza- tion of functionally segregated circuits linking basal ganglia and cortex. Annu Rev Neurosci. 1986;9:357–381.
12. Haber SN. The primate basal ganglia: parallel and integrative networks. J Chem Neuroanat. 2003;26:317–330.
912
O. Dandash et al
13. Pantelis C, Barnes TR, Nelson HE. Is the concept of fron- tal-subcortical dementia relevant to schizophrenia? Br J Psychiatry. 1992;160:442–460.
14. Robbins TW. The case of frontostriatal dysfunction in schizo- phrenia. Schizophr Bull. 1990;16:391–402.
15. Murase S, Grenhoff J, Chouvet G, Gonon FG, Svensson TH. Prefrontal cortex regulates burst firing and transmitter release in rat mesolimbic dopamine neurons studied in vivo. Neurosci Lett. 1993;157:53–56.
16. Stevens JR. An anatomy of schizophrenia? Arch Gen Psychiatry. 1973;29:177–189.
17. Kegeles LS, Abi-Dargham A, Frankle WG, et al. Increased synaptic dopamine function in associative regions of the striatum in schizophrenia. Arch Gen Psychiatry. 2010;67:231–239.
18. Fusar-Poli P, Howes OD, Allen P, et  al. Abnormal fronto- striatal interactions in people with prodromal signs of psy- chosis: a multimodal imaging study. Arch Gen Psychiatry. 2010;67:683–691.
19. Fusar-Poli P, Howes OD, Allen P, et al. Abnormal prefrontal activation directly related to pre-synaptic striatal dopamine dysfunction in people at clinical high risk for psychosis. Mol Psychiatry. 2011;16:67–75.
20. Fornito A, Bullmore ET. What can spontaneous fluctua- tions of the blood oxygenation-level-dependent signal tell us about psychiatric disorders? Curr Opin Psychiatry. 2010;23:239–249.
21. Fornito A, Zalesky A, Pantelis C, Bullmore ET. Schizophrenia, neuroimaging and connectomics. Neuroimage. 2012;62: 2296–2314.
22. Stephan KE, Friston KJ, Frith CD. Dysconnection in schizo- phrenia: from abnormal synaptic plasticity to failures of self- monitoring. Schizophr Bull. 2009;35:509–527.
23. Di Martino A, Kelly C, Grzadzinski R, et al. Aberrant stri- atal functional connectivity in children with autism. Biol Psychiatry. 2011;69:847–856.
24. Furman DJ, Hamilton JP, Gotlib IH. Frontostriatal func- tional connectivity in major depressive disorder. Biol Mood Anxiety Disord. 2011;1:11.
25. Harrison BJ, Soriano-Mas C, Pujol J, et al. Altered corticos- triatal functional connectivity in obsessive-compulsive disor- der. Arch Gen Psychiatry. 2009;66:1189–1200.
26. Fornito A, Bullmore ET. Connectomic intermediate phe- notypes for psychiatric disorders. Front Psychiatry. 2012;3:32.
27. Glahn DC, Winkler AM, Kochunov P, et  al. Genetic con- trol over the resting brain. Proc Natl Acad Sci U S A. 2010;107:1223–1228.
28. Fox MD, Raichle ME. Spontaneous fluctuations in brain activity observed with functional magnetic resonance imag- ing. Nat Rev Neurosci. 2007;8:700–711.
29. Chong SA, Campbell A, Chee M, et  al. The Singapore flagship programme in translational and clinical research in psychosis. Early Interv Psychiatry. 2011;5:290–300.
30. Yung AR, Yuen HP, McGorry PD, et al. Mapping the onset of psychosis: the Comprehensive Assessment of At-Risk Mental States. Aust N Z J Psychiatry. 2005;39:964–971.
31. Kay SR, Fiszbein A, Opler LA. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull. 1987;13:261–276.
32. Addington D, Addington J, Maticka-Tyndale E. Assessing depression in schizophrenia: the Calgary Depression Scale. Br J Psychiatry. 1993;22:39–44.
33. Beck AT, Epstein N, Brown G, Steer RA. An inventory for measuring clinical anxiety: psychometric properties. J Consult Clin Psychol. 1988;56:893–897.
34. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Press; 1994.
35. First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview for DSM-IV-TR Axis I Disorders, Research Version, Patient Edition. (SCID-I/P). New York: Biometrics Research, New York State Psychiatric Institute; 2002.
36. Di Martino A, Scheres A, Margulies DS, et  al. Functional connectivity of human striatum: a resting state FMRI study. Cereb Cortex. 2008;18:2735–2747.
37. Van Dijk KR, Sabuncu MR, Buckner RL. The influence of head motion on intrinsic functional connectivity MRI. Neuroimage. 2012;59:431–438.
38. Ward BD. Simultaneous Inference for fMRI Data. Milwaukee, WI: Biophysics Research Institute, Medical College of Wisconsin; 2000.
39. Song XW, Dong ZY, Long XY, et  al. REST: a toolkit for resting-state functional magnetic resonance imaging data processing. PLoS One. 2011;6:e25031.
40. de la Fuente-Sandoval C, Leon-Ortiz P, Favila R, et al. Higher levels of glutamate in the associative-striatum of subjects with prodromal symptoms of schizophrenia and patients with first-episode psychosis. Neuropsychopharmacology. 2011;36:1781–1791.
41. Fusar-Poli P, Stone JM, Broome MR, et al. Thalamic gluta- mate levels as a predictor of cortical response during execu- tive functioning in subjects at high risk for psychosis. Arch Gen Psychiatry. 2011;68:881–890.
42. Morey RA, Inan S, Mitchell TV, Perkins DO, Lieberman JA, Belger A. Imaging frontostriatal function in ultra-high-risk, early, and chronic schizophrenia during executive processing. Arch Gen Psychiatry. 2005;62:254–262.
43. Pantelis C, Barnes TR, Nelson HE, et al. Frontal-striatal cog- nitive deficits in patients with chronic schizophrenia. Brain. 1997;120 (Pt 10):1823–1843.
44. Zhou Y, Liang M, Jiang T, et  al. Functional dysconnectiv- ity of the dorsolateral prefrontal cortex in first-episode schizophrenia using resting-state fMRI. Neurosci Lett. 2007;417:297–302.
45. Tu PC, Hsieh JC, Li CT, Bai YM, Su TP. Cortico-striatal dis- connection within the cingulo-opercular network in schizo- phrenia revealed by intrinsic functional connectivity analysis: a resting fMRI study. Neuroimage. 2012;59:238–247.
46. Pettersson-Yeo W, Allen P, Benetti S, McGuire P, Mechelli A. Dysconnectivity in schizophrenia: where are we now? Neurosci Biobehav Rev. 2011;35:1110–1124.
47. Nagano-Saito A, Leyton M, Monchi O, Goldberg YK, He Y, Dagher A. Dopamine depletion impairs frontostriatal functional connectivity during a set-shifting task. J Neurosci. 2008;28:3697–3706.
48. Kelly C, de Zubicaray G, Di Martino A, et  al. L-dopa modulates functional connectivity in striatal cognitive and motor networks: a double-blind placebo-controlled study. J Neurosci. 2009;29:7364–7378.
49. Egan MF, Goldberg TE, Kolachana BS, et  al. Effect of COMT Val108/158 Met genotype on frontal lobe func- tion and risk for schizophrenia. Proc Natl Acad Sci U S A. 2001;98:6917–6922.
50. Meyer-Lindenberg A, Miletich RS, Kohn PD, et  al. Reduced prefrontal activity predicts exaggerated striatal
913
dopaminergic function in schizophrenia. Nat Neurosci. 2002;5: 267–271.
51. Goldman-Rakic PS, Muly EC, Williams GV. D(1) recep- tors in prefrontal cells and circuits. Brain Res Rev. 2000;31:295–301.
52. Johnson SW, Palmer MR, Freedman R. Effects of dopamine on spontaneous and evoked activity of caudate neurons. Neuropharmacology. 1983;22:843–851.
53. Fornito A, Yung AR, Wood SJ, et al. Anatomic abnormali- ties of the anterior cingulate cortex before psychosis onset: an MRI study of ultra-high-risk individuals. Biol Psychiatry. 2008;64:758–765.
54. Sun D, Phillips L, Velakoulis D, et al. Progressive brain struc- tural changes mapped as psychosis develops in ‘at risk’ indi- viduals. Schizophr Res. 2009;108:85–92.
55. Hannan KL, Wood SJ, Yung AR, et al. Caudate nucleus vol- ume in individuals at ultra-high risk of psychosis: a cross- sectional magnetic resonance imaging study. Psychiatry Res. 2010;182:223–230.
56. Karreman M, Moghaddam B. The prefrontal cortex regu- lates the basal release of dopamine in the limbic striatum: an effect mediated by ventral tegmental area. J Neurochem. 1996;66:589–598.
57. Svensson TH, Tung CS. Local cooling of pre-frontal cor- tex induces pacemaker-like firing of dopamine neurons in rat ventral tegmental area in vivo. Acta Physiol Scand. 1989;136:135–136.
58. Carlsson M, Carlsson A. Interactions between glutamatergic and monoaminergic systems within the basal ganglia–impli- cations for schizophrenia and Parkinson’s disease. Trends Neurosci. 1990;13:272–276.
59. Laruelle M, Abi-Dargham A. Dopamine as the wind of the psychotic fire: new evidence from brain imaging studies. J Psychopharmacol. 1999;13:358–371.
60. Carlsson A. The current status of the dopamine hypothesis of schizophrenia. Neuropsychopharmacology. 1988;1:179–186.
61. Fusar-Poli P, Broome MR, Matthiasson P, et al. Prefrontal function at presentation directly related to clinical outcome in people at ultrahigh risk of psychosis. Schizophr Bull. 2011;37:189–198.
62. Sabb FW, van Erp TG, Hardt ME, et al. Language network dysfunction as a predictor of outcome in youth at clinical high risk for psychosis. Schizophr Res. 2010;116:173–183.
63. Corlett PR, Murray GK, Honey GD, et al. Disrupted pre- diction-error signal in psychosis: evidence for an associative account of delusions. Brain. 2007;130:2387–2400.
64. Frith C. The neural basis of hallucinations and delusions. C R Biol. 2005;328:169–175.
65. Carlsson A, Waters N, Carlsson ML. Neurotransmitter interactions in schizophrenia–therapeutic implications. Biol Psychiatry. 1999;46:1388–1395.
66. Stone JM, Bramon E, Pauls A, Sumich A, McGuire PK. Thalamic neurochemical abnormalities in individuals with prodromal symptoms of schizophrenia - relationship
to auditory event-related potentials. Psychiatry Res. 2010;183:174–176.
67. Stone JM, Day F, Tsagaraki H, et  al.; OASIS. Glutamate dysfunction in people with prodromal symptoms of psy- chosis: relationship to gray matter volume. Biol Psychiatry. 2009;66:533–539.
68. Deutsch SI, Rosse RB, Schwartz BL, Mastropaolo J. A revised excitotoxic hypothesis of schizophrenia: therapeutic implications. Clin Neuropharmacol. 2001;24:43–49.
69. Allen P, Chaddock CA, Howes OD, et al. Abnormal relation- ship between medial temporal lobe and subcortical dopa- mine function in people with an ultra high risk for psychosis. Schizophr Bull. 2012;38:1040–1049.
70. Fornito A, Harrison BJ, Goodby E, et al. Functional dyscon- nectivity of corticostriatal circuitry as a risk phenotype for psychosis. JAMA Psychiatry. 2013. In press.
71. Sim K, Swapna V, Mythily S, et al. Psychiatric comorbidity in first episode psychosis: the Early Psychosis Intervention Program (EPIP) experience. Acta Psychiatr Scand. 2004;109:23–29.
72. Subramaniam M, Abdin E, Vaingankar J, Phua AM, Tee J, Chong SA. Prevalence and correlates of alcohol use dis- orders in the Singapore Mental Health Survey. Addiction. 2012;107:1443–1452.
73. Verma SK, Subramaniam M, Chong SA, Kua EH. Substance abuse in schizophrenia. A  Singapore perspective. Soc Psychiatry Psychiatr Epidemiol. 2002;37:326–328.
74. Arseneault L, Cannon M, Poulton R, Murray R, Caspi A, Moffitt TE. Cannabis use in adolescence and risk for adult psychosis: longitudinal prospective study. BMJ. 2002;325:1212–1213.
75. Kristensen K, Cadenhead KS. Cannabis abuse and risk for psychosis in a prodromal sample. Psychiatry Res. 2007;151:151–154.
76. Machielsen M, van der Sluis S, de Haan L. Cannabis use in patients with a first psychotic episode and subjects at ultra high risk of psychosis: impact on psychotic- and pre-psychotic symptoms. Aust N Z J Psychiatry. 2010;44:721–728.
77. Cheer JF, Wassum KM, Heien ML, Phillips PE, Wightman RM. Cannabinoids enhance subsecond dopamine release in the nucleus accumbens of awake rats. J Neurosci. 2004;24:4393–4400.
78. Chen J, Paredes W, Lowinson JH, Gardner EL. Delta 9-tetrahydrocannabinol enhances presynaptic dopa- mine efflux in medial prefrontal cortex. Eur J Pharmacol. 1990;190:259–262.
79. Bhattacharyya S, Crippa JA, Allen P, et al. Induction of psy- chosis by Δ9-tetrahydrocannabinol reflects modulation of prefrontal and striatal function during attentional salience processing. Arch Gen Psychiatry. 2012;69:27–36.