Orbital and Medial Prefrontal Cortex Functional Connectivity of Major Depression Vulnerability and Disease

Accepted Manuscript Orbital and medial prefrontal cortex functional connectivity of major depression vulnerability and disease Zoe Samara, Elisabeth A.T. Evers, Frenk Peeters, Harry B.M. Uylings, Grazyna Rajkowska, Johannes G. Ramaekers, Peter Stiers PII:

S2451-9022(18)30018-1

DOI:

10.1016/j.bpsc.2018.01.004

Reference:

BPSC 236

To appear in:

Biological Psychiatry: Cognitive Neuroscience and Neuroimaging

Received Date: 5 July 2017 Revised Date:

12 January 2018

Accepted Date: 16 January 2018

Please cite this article as: Samara Z., Evers E.A.T., Peeters F., Uylings H.B.M., Rajkowska G., Ramaekers J.G. & Stiers P., Orbital and medial prefrontal cortex functional connectivity of major depression vulnerability and disease, Biological Psychiatry: Cognitive Neuroscience and Neuroimaging (2018), doi: 10.1016/j.bpsc.2018.01.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

RI PT

Short Title: OMPFC connectivity in major depression

M AN U

SC

Title: Orbital and medial prefrontal cortex functional connectivity of major depression vulnerability and disease

Affiliations: 1

TE D

Authors: Zoe Samara1, Elisabeth A.T. Evers1, Frenk Peeters1, Harry B.M. Uylings2, Grazyna Rajkowska3, Johannes G. Ramaekers1 and Peter Stiers1

2

EP

Department of Neuropsychology and Psychopharmacology, Maastricht University, 6229 ER Maastricht, The Netherlands Department of Anatomy and Neuroscience, VU University Medical Center, Amsterdam, The Netherlands

3

AC C

Department of Psychiatry and Human Behavior, University of Mississippi Medical Center, Jackson, MS 39216-4505, USA

Correspondence should be addressed to Zoe Samara, Department of Neuropsychology and Psychopharmacology, Maastricht University, Universiteitssingel 40 (East), 6229 ER Maastricht, The Netherlands. E-mail:[email protected]

Keywords: prefrontal, functional connectivity, major depression, vulnerability, MRI, family history Abstract: 249 Main Text: 3997 Tables: 2 Figures: 3 Supplemental Information: 1

1

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

ABSTRACT Background: Pathophysiology models of major depression (MD) center on the dysfunction of various cortical areas within the orbital and medial prefrontal cortex (OMPFC). While

RI PT

independent structural and functional abnormalities in these areas are consistent findings in MD, the complex interactions among them and the rest of the cortex remain largely unexplored. Methods: We used resting state functional magnetic resonance imaging (fMRI) connectivity to

SC

systematically map alterations in the communication between OMPFC fields and the rest of the brain in MD. Functional connectivity (FC) maps from participants with current MD (N=35),

M AN U

unaffected first-degree relatives (N=36) and healthy controls (N=38) were subjected to conjunction analyses to distinguish FC markers of MD vulnerability and FC markers of MD disease).

Results: FC abnormalities in MD vulnerability were found for dorsal medial wall regions and the anterior insula and concerned altered communication of these areas with the inferior parietal

TE D

cortex and dorsal posterior cingulate, occipital areas and the brainstem. FC aberrations in current MD included the anterior insula, rostral and dorsal anterior cingulate and lateral orbitofrontal areas and concerned altered communication with the dorsal striatum, the

EP

cerebellum, the precuneus, the anterior PFC, somato-motor cortex, dorsolateral PFC, and visual areas in the occipital and inferior temporal lobes.

AC C

Conclusions: Functionally delineated parcellation maps can be used to identify putative connectivity markers in extended cortical regions such as the OMPFC. The anterior insula and the rostral anterior cingulate play a central role in the pathophysiology of MD, being consistently implicated both in the MD vulnerability and MD disease states.

2

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

INTRODUCTION Major Depression (MD) is a debilitating syndrome presenting with various symptoms in mood, reward and decision making (1). The efficacy of MD neuro-modulation therapies such as

RI PT

transcranial magnetic stimulation (TMS; 2) and deep brain stimulation (DBS; 3) with various targets demonstrates in a “proof of concept” manner that MD is a systems-level disorder

involving a complex pattern of altered interactions between several brain areas. Convergent

SC

neuroimaging findings suggest that, together with limbic and autonomic centers, subregions within the orbital and medial prefrontal cortex (OMPFC) (4-6) play a central role in MD’s

M AN U

aberrant networks. Understanding of the neuro-circuitry pathways underlying MD symptoms and treatment response remains elusive, despite the large body of evidence documenting abnormalities in various OMPFC subregions and the efficacy of connectivity-based treatments with OMPFC targets.

Elucidation of MD related changes, if any, in the functional wiring between the OMPFC

TE D

subregions and key brain areas for autonomic, cognitive and affective control is critical for generating new hypotheses on MD disease mechanisms and for identifying connectivity biomarkers to guide diagnosis and treatment. On the theoretical side, neuro-circuit models of

EP

MD propose altered pathways based only on tracer or lesion findings in animals and rare post mortem human reports (13). Thus far, MRI studies of MD have examined functional wiring of the

AC C

OMPFC either only for select coordinates derived from activation studies or coarsely, as either part of the anterior default mode network (DMN) or with predefined general anatomical labels (7-10; see reviews in 11,12). Recent neuroimaging advances on MRI connectivity analysis tools enable us to study

the cortical organization systematically, for a specified area or the cerebral cortex as a whole (see review in 13). These tools use homogeneity in functional connectivity in an attempt to delineate cortical fields, which are anatomically defined as architectonic and connectional homogeneous patches of cortex (14-15). While it is still open how these fields correspond to 3

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

anatomically defined cytoarchitectonic areas, at the group level these data-driven methods yield a systematical parcellation of cortex in subregions that are neuro-anatomically plausible in general terms of location, spatial extend and whole brain (functional) connectivity (13; 17-23).

RI PT

In the current study we make use of such a parcellation of the orbital and medial prefrontal cortex in the left hemisphere of 34 healthy adults (16). That study delineated 19

functional subregions in the left hemisphere cortex covering the anterior medial PFC, orbital

SC

surface, and the ventral anterior insula (see white outline in Figures 1, 2 and 3). Their spatial organization and intrinsic as well as cortical and subcortical functional connectivity

M AN U

characteristics were generally in accordance with current knowledge of the anatomical organization of this part of the cerebral cortex from cytoarchitectonic and tracer studies in the macaque monkey and cytoarchitectonic and diffusion MRI studies in humans (for a detailed discussion see ref. 18, including their supplementary Information). Here, we used this parcellation map as a lay-out to identify FC changes associated with OMPFC in MD.

TE D

We focused on the left OMPFC since only this hemisphere was extensively studied in (16). Pathophysiology studies of MD and rTMS interventions (i.e. excitatory left DLPFC vs inhibitory right) in patients with MD suggest that left and right PFC might be differentially

EP

involved, while TMS treatment centered on the left dorsolateral PFC is the most widely used (25). We sought specifically to test the hypotheses that OMPFC subregions show altered

AC C

connectivity patterns which are: 1) common to MD patients and at-risk individuals compared to healthy controls (connectivity markers of MD vulnerability) and, 2) unique to the MD group compared to at-risk individuals and controls (connectivity markers of acute MD or MD disease). With this design, we thus aimed to uncover functional connectivity abnormalities that might confer susceptibility to MD by giving rise to emotional and cognitive symptoms observed in unaffected at-familial MD risk individuals as well as in MD patients in remission. At the same time, we specifically aimed to dissociate these abnormalities from the functional connectivity

4

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

aberrations which characterize the acute MD state and which may underlie the cognitive,

AC C

EP

TE D

M AN U

SC

RI PT

affective and somatic symptoms observed during depressive episodes.

5

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

METHODS & MATERIALS Clinical Assessments Thirty-five participants experiencing a major depressive episode (cutoff BDI-II > 20;

RI PT

mean age 38.9; SD 11.6; 71% female) (MD group), thirty-six unaffected first-degree relatives (mean age 34; SD 14.6; 72% female) (FH group) and thirty-eight non-psychiatric controls (mean age 36; SD 16.4; 68% female) (HC group) participated after providing informed consent.

SC

Patients were included if they met DSM-IV-TR criteria for MD and excluded if they were diagnosed with Bipolar I or II, substance dependence or were taking benzodiazepines.

M AN U

Participants in the other two groups were excluded in the presence of any axis I diagnosis. Screening procedures and diagnostic and medication histories are reported in the Supplement. The study was approved by the Medical Ethical Committee of Maastricht University. Participants completed the Beck Depression Inventory-II (BDI-II), Brief Symptom Inventory (BSI), and the Quick Inventory of Depressive Symptomatology Self-Report (QIDS-SR) (26-28). They also

TE D

completed Raven’s Standard Progressive Matrices, a 60-item test of non-verbal IQ (29).

MRI Acquisition & Preprocessing

EP

MRI data included a resting state scan of minimally 6.4 minutes (TR=2.5 sec, 2x2x3mm voxels, 153 or 203 volumes, equally spread across the three groups), a high-resolution T1-

AC C

weighted anatomical scan and a field map. The EPI sequence was optimized to reduce susceptibility artefacts in the orbital cortex: low echo time (25 ms), thinner slices (3mm) oriented with the orbital surface (±30°), and field-map dist ortion correction (see Samara et al. (2017, p. 2951 for quantifications and discussion of the orbital signal quality with this sequence and scanner). Data preprocessing with SPM 5 (Welcome Trust Center for Neuroimaging) included correction for slice timing, head motion, and spatial distortion, normalization to the Montreal Neurological Institute (MNI) 152 template, 6 mm FWHM smoothing, removal of linear trends, of white matter and CSF signals, and of affine head motion parameters, and finally 0.1-0.01 Hz 6

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

band pass filtering. To further eliminate detrimental effects of head motion (30, 31) volumes with a frame-wise displacement exceeding 10% of the slice thickness were removed and groups were equated on the mean frame-wise displacement (p=.57). Acquisition parameters and

RI PT

preprocessing are detailed in the Supplement.

Statistical Analyses

SC

We sought to identify FC abnormalities in 19 OMPFC fields of the left-hemisphere

parcellation map (18). For ease of comprehension, we refer to these fields by their presumed

M AN U

anatomical interpretations, as discussed in (18). The locations of the fields are indicated as outlines in Figures 1, 2 and 3 (for details see 18). A set of 19 seed time courses was created for each participant by selecting voxels within a sphere of 4mm radius around the center of mass (COM) of each field. The focus on the field centers ensured that only voxels were included that were representative of each field’s functional dynamic, despite individual variation in the exact

TE D

location of the fields in each individual data set. To further increase the field specificity of the connectivity profiles we computed for each field the first Eigen vector of the low frequency (0.01 - 0.1 Hz) signal in the time courses of the selected voxels. Per seed, whole brain FC profiles

EP

were calculated as the Fisher-Z transformed Pearson’s correlation between each seed’s timecourse and the time-course of each voxel in that participant’s brain. These field and participant

AC C

specific FC maps were subjected to a random effects general linear model (GLM) analysis to estimate differences between the three groups, with age, gender and IQ Raven score as covariates.

To test for connectivity abnormalities of MD vulnerability, we created conjunction maps

of the F contrasts between 1) the HC and the MD group and 2) the HC and the FH groups. These maps highlighted voxels whose FC with the seed differed both between controls and patients, AND between controls and at-risk participants. Further, to ensure in our vulnerability measure that FH participants were not different from the MD patients, we removed from the 7

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

conjunction maps all voxels significant at a lenient threshold (.001 uncorrected) in the F contrast between MD versus FH groups. This eliminated as markers of vulnerability all voxels in which the FC change in the at-risk participants was intermediate between the healthy controls and the

RI PT

MD patients. To test for connectivity abnormalities associated with acute MD, we created new

conjunction maps, now of the F contrasts between 1) the MD group and HC and 2) the MD and

SC

FH groups. These conjunction maps now highlighted voxels whose FC with the seed differed both between MD patients and controls, AND between patients and at-risk participants. Further,

M AN U

to ensure in our disease manifestation measure that at-risk participants were not different from the healthy controls, we removed from the conjunction maps all voxels significant at a lenient threshold (.001 uncorrected) in the F contrast between FH versus HC. This excluded all voxels with a FC change in the at-risk participants that was intermediate between the healthy controls and the MD patients so that the final maps contained voxels whose FC was exclusively altered

TE D

in the disease state.

All conjunction maps were tested for significance with Monte Carlo simulations (10.000 iterations; voxel-level p =.01, cluster-level p=.0025 – Bonferroni-adjusted for 19 independent

EP

comparisons). We ensured that the connectivity differences reported here were not driven by antidepressant medication use or comorbidity (MD group), or self-reported symptoms (both

AC C

depression and other psychiatric complaints in the FH group) by regressing the influence of these variables from the MD and FH individual maps prior to the final across-groups analysis and using the residual, cleaned maps for these groups in the next step. Gray matter differences between the three groups were controlled by removing from our results maps all voxels which their probability of being gray matter was significantly different (.001 uncorrected) among the groups. We determined the direction of FC differences in MD vulnerability and in disease manifestation with ROI-based pairwise t- tests using the r-Z-transformed values in the 8

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

conjunction maps. Correlation analyses for the MD group between the FC changes in the acute phase and our clinical measures were also performed (for details of all analyses see

AC C

EP

TE D

M AN U

SC

RI PT

Supplement).

9

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

RESULTS Demographic and Clinical Characteristics MD patients had their first episode on average at the age of 29.3 (SD= 11.4) and a mean

RI PT

of 2.9 episodes (SD=3). Comorbidity (having at least one additional axis I diagnosis) and use of antidepressants at the time of the scan were reported by 51% and 48% of the MD group respectively. Most prevalent comorbid diagnosis was social phobia and the majority of

SC

participants were using SSRI’s (Supplementary Table 1). Average BDI-II and QIDS-SR scores at the time of the scan in the MD group were 32.74 and 16.57 respectively. At-risk participants

M AN U

had clinically and statistically significantly lower scores of depressive and general psychiatric symptomatology compared to the MD group (Table 1, one put last column). Their symptoms scores were somewhat elevated compared to the healthy control participants (Table 1, last column), but none scored near the cut-off for clinical symptomatology. Groups were matched in age, gender and IQ (all p’s >.2).

Neuroimaging Analysis

TE D

INSERT TABLE 1 HERE

FC abnormalities of MD vulnerability

EP

In our analysis for markers of MD vulnerability we found that MD patients and at-risk individuals share various FC abnormalities in the OMPFC compared to controls. The GLM

AC C

analysis revealed changes in the FC of four OMPFC regions: anterior insula, para-cingulate regions 32 pregenual (p32) and 32 dorsal (d32) (see 18 for area homology, nomenclature and relevant references) and dorsomedial frontal area BA9 (Figure 1). ROI-based post-hoc analysis indicated that MD vulnerability is characterized by decreased coupling between area BA9 and an area of the ventral intraparietal sulcus and between area p32 and dorsal posterior cingulate and contralateral inferior parietal cortex (see Table 2 for p values of post-hoc tests). In contrast, increased FC was observed between the anterior insula and the brainstem and between medial region d32 and the contralateral dorsal occipital cortex. It should be noted that in MD 10

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

vulnerability no significant changes in the functional coupling among any of the 19 OMPFC fields with each other were observed. INSERT FIGURE 1 HERE

RI PT

FC abnormalities of MD Disease In our analysis for aberrations in MD disease, we found that MD patients, in addition to the FC changes that they share with the at-risk group, also show unique impairments in the

SC

communication of the OMPFC with several other brain areas in the rest of the cortex. The

random effects GLM analyses revealed additional changes for six OMPFC fields, summarized in

M AN U

Figure 2 and Table 2. Two of the OMPFC fields implicated in vulnerability show also abnormalities associated with MD disease: the anterior insula and the dorsomedial prefrontal cortex (p32). The anterior insula had changed coupling ipsilaterally with the lateral occipital cortex, fusiform gyrus, cerebellum, and putamen, and contralaterally with the caudate nucleus. Field p32 had altered coupling with the ipsilateral precuneus. Four additional fields had changed

TE D

functional coupling in the MD patients compared to the other two groups. On the medial side, p24 had changed coupling with the anterior superior frontal gyrus, while mid-cingulate field a24’ had changed coupling with the dorsolateral PFC and the left para-central lobule. At the orbital

EP

side, changed coupling was found, first, between lateral orbital area 47/12 and the dorsal occipital cortex bilaterally, the left inferior temporal lobe and the parietal para-central lobule and

AC C

subcortically with the contralateral body of the caudate nucleus. Secondly, central orbital area 11 had changed coupling with the dorsal precuneus. ROI-based post-hoc analysis indicated that MD patients had decreased FC between area p32 and the precuneus compared to both at-risk individuals and controls. In contrasts, the MD group had increased FC compared to both groups between all the other OMPFC cortical fields and the cortical and subcortical regions reported. Similarly to the MD vulnerability results, no functional coupling changes in between the 19 OMPFC fields were observed in this analysis. INSERT FIGURE 2 HERE 11

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

Correlation Analyses MD patients and FH participants share a set of FC indicators and show increased selfreport measures of depressive symptomatology compared to the HC participants. Within the MD

range of symptoms and FC indicators in the patient group alone.

AC C

EP

TE D

M AN U

SC

INSERT TABLE 2 HERE

RI PT

patients this correlation did not reach significance, however, reflecting the reduced variability

12

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

DISCUSSION We examined changes in the functional connectivity of the OMPFC associated with MD. By starting from a previously established parcellation scheme (16) we delineated the specific

RI PT

subregions of OMPFC involved in the disease. By including a group of participants at familial risk for MD, alongside MD patients and healthy controls, we distinguished between the FC

alterations that confer vulnerability to MD (i.e., MD = FH ≠ HC) and the FC correlates of the

SC

current MD episode (i.e., MD ≠ FH = control). Our results suggest that individuals with familial risk for MD, in the absence of clinically significant symptoms, share with patients aberrations in

M AN U

functional connectivity of OMPFC fields that center on two foci, one in the dorsomedial prefrontal cortex and one in the anterior insula. Acute MD patients, in addition to these vulnerability changes, exhibit unique FC alterations originating from these same two foci, and from one additional focus in the orbital frontal cortex. This suggests that neurobiologically MD starts off with confined predisposing FC abnormalities and progresses to a disease state with

TE D

more extensive OMPFC alterations (Figure 3).

Functional connectivity in MD vulnerability

EP

The first of the two foci of FC abnormality associated with MD vulnerability concerns an increased functional coupling between anterior insula and the periaqueductal grey (PAG). The

AC C

PAG is a brainstem control center for survival-related responses during stress and threat (32). Insula-PAG coupling strength has been shown to increase with perceived painfulness of a stimulus (33), experienced versus observed pain (34, 35), stressfulness of cognitive tasks (36) and proximity of a virtual predator (37). Against this background, the increased insula-PAG coupling observed here may suggest a proclivity in at-risk individuals to interpret anticipated events as more emotionally threatening or generally negative. In line with this behavioral studies of vulnerability for MD show anticipatory affect biases, including increased threat reactivity (38, 39). 13

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

An affective bias hypothesis could also make sense of the second focus of vulnerabilityrelated alterations in functional coupling, between regions of the dorsal medial PFC and the posterior, “receptive” brain. The involved dorsal medial PFC region extends from the cingulate

RI PT

sulcus until the dorsal convexity of the superior frontal gyrus at around the intersection of the cognitive and affective part of the medial PFC (40). This region has dense anatomical

connections to autonomic centers (41, 42), and has been implicated in assessment salience

SC

and affective value of emotional information (43), conflict monitoring in the presence of

emotional distractors (40, 44-49) as well as in awareness and regulation of emotional responses

M AN U

(40, 50-52). This affective bias hypothesis is furthermore in agreement with behavioral characteristics observed in people with genetic MD susceptibility and remitted depressed (53, 54) such as difficulty directing attention away from negative stimuli (54), reappraising (55) and suppressing negative distractors (56).

The altered FC of medial regions associated with this second focus agrees with this

TE D

pattern. On the one hand, BA9 and d32 showed changed coupling with the posterior intraparietal sulcus, which is part of the dorsal visual attention network (57). On the other hand, area p32 was uncoupled from the dorsal inferior parietal cortex implicated in the computation of

EP

priorities for attentional allocation (58, 59), the re-evaluation of conflicting choices between options (60, 61), and the reappraisal/suppression of emotion (55). Interestingly, the inferior

AC C

parietal lobule has been found to respond to cognitive behavioral therapy (62, 63), a treatment aimed at modifying attentional biases and reinforcing emotion-regulation strategies. Area p32 also lost functional coupling with the posterior cingulate cortex (PCC), which may relate to areas p32’s role in memory consolidation (64). Dorsal PCC has a mixed pattern of anatomical connectivity linking frontal default mode regions with parietal attention areas (65, 66) and is thought to balance internal and external attentional foci in order to retrieve and update behavioral strategies (67). In line with this, PCC lesions result in deficits in implementing new strategies and changing cognitive set (67). 14

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

Taken together, the observed pattern of FC dysregulation between these OMPFC areas and the brainstem, dorsal attention areas, the inferior parietal and the posterior cingulate cortex might be the manifestation of a predisposition in MD vulnerable individuals for anticipating or

RI PT

perceiving environmental events as threatening, painful and overall negative. Increased occurrence of (perceived) negative experiences, combined with a diminished ability to

cognitively regulate the resulting emotional arousal might make at risk individuals more prone to

SC

stress, chronic negative affect and a dysregulated HPA axis.

M AN U

Functional connectivity in MD disease

In addition to the changes associated with MD vulnerability, patients with acute MD exhibited an extension of changes in the coupling of the two foci marking MD vulnerability (i.e the anterior insula and the rostral cingulate), and an emergence of a new focus in the orbital frontal cortex. The cerebral targets of FC aberrations across these foci tend to aggregate in two

TE D

functional zones. The first of these zones comprises areas of higher-order vision and attention in the posterior brain and was already prominent in the vulnerability markers, while the second, in the superior frontal gyrus, seems specific to disease manifestation.

EP

All three medial OMPFC foci show changed coupling with higher visual cortex, both in the ventral (loci d, e and j in Figure 2) and the dorsal (loci f to i in Figure 2) stream. The insular

AC C

(aI) and orbital (47/12) foci have changed coupling with the anterior ventral stream, responsible for the analysis and representation of visual objects and their identity (68). The orbital area 47/12 had changed coupling with the dorsal visual stream, namely the ventral intraparietal sulcus (“VIPS” in 69), already highlighted as MD vulnerability indicator. The additional loss of coupling with the contralateral oculomotor/attentional part of the caudate nucleus further confirms the functional involvement of visual attentional processes. Finally, medial (area p32) and orbital foci (area 11 and 47/12) had changed coupling with parts of the non-limbic precuneus also implicated in visual processing (66, 69). This part of the precuneus has 15

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

previously been linked to MD, with patients showing increased node centrality (70) and abnormal activation to attentional targets and emotional distractors (71). This pattern of FC disturbances originating from all three OMFC foci of MD-related FC change strengthens the

RI PT

earlier suggestion that changed affective modulation of information processing is a core aspect of MD. In MD patients, the dorsal medial focus of FC change indicative of MD vulnerability gets an important extension into the cingulate gyrus (Figure 3), with two additional areas (p24 and

SC

a24’) showing changed functional coupling. The cingulate extension spans the transition from the anterior to the mid-cingulate section (MCC) of the cingulate cortex (72), with a24’ being the

M AN U

rostral part of MCC, including the rostral cingulate motor area in the cingulate sulcus. This, together with the action selection/monitoring function attributed to the mid-cingulate cortex (73) might explain the changed coupling of a24’ with the motor-related paracentral gyrus. Both these areas show changed functional coupling with the superior frontal gyrus. The function of this part of SFG is not well studied, but it shows consistent functional coupling with nodes of the default

TE D

mode network (74, 17, 75, 76). Moreover, introspective tasks increase activity specifically in this part of SFG, with activity extending medially into BA32 (77), while damage to SFG and BA32 correlates with impairment in re-appraising the personal relevance of negative events (78).

EP

Interestingly, this part of SFG has also been implicated in expressions of positive affect such as

AC C

laughter (79) and humor (80) important for self-centered re-appraisal.

Limitations

While our results provide the first insightful dissociation of functional connectivity

abnormalities in MD vulnerability and disease, replication is needed, particularly given the heterogeneity of clinical populations. Although we corrected rigorously for medication and axis I comorbidity effects, replication should ideally be sought in a large independent sample of unmedicated MD participants. Secondly, while the FC abnormalities described here reflect significant group differences related to MD vulnerability and disease, we could not examine 16

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

whether they are specific to MD. Future studies should examine whether these FC changes could differentiate between unipolar depression and other psychiatric syndromes. Lastly, our focus on OMPFC prevents drawing conclusions regarding other brain areas. For instance,

RI PT

executive dysfunctions often manifest in MD, which would predict FC abnormalities associated with more cognition-centered networks.

SC

General conclusions

The current study systematically delineated functional connectivity alterations of cortical

M AN U

units in the OMPFC in MD using a published parcellation map. This enabled a more fine-grained localization of the pathogenetically relevant subregions within the affective cortex than previously. We also employed conjunction analyses on groups of healthy controls, MD susceptible individuals and patients during MD episodes and dissociated the patterns of connectivity aberrations in MD vulnerability versus those of in MD manifestation. Our results

TE D

complement previous evidence that the OMPFC, anatomically and functionally linked to autonomic and affective centers, shows connectivity aberrations in both MD states. Moreover, we provide evidence that these functional aberrations concern primarily the relationship of

EP

specific OMPFC subparts and the posterior receptive cortex and various regions within specifically the dorsal and ventral visual processing streams. This in line with the increasingly

AC C

recognized relationship between emotion and visual perception: visual stimuli and experiences trigger and shape emotions and affect as well as being shaped by mood states (81, 82). Furthermore, our results highlight two OMPFC regions, both implicated in MD

vulnerability and disease manifestation as being central to depression’s pathophysiology, the anterior insula and the rostral anterior cingulate area p32. Both of these areas have been previously shown to have altered activity and connectivity in MD (83-86). More importantly, they are both central nodes within two extended neural networks, the salience network and the default mode network that have also been implicated in MD (87, 88). Here we provide evidence 17

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

that the connectivity of these central nodes or hubs is altered not only in the MD disease state but long before any clinical symptoms appear in individuals with familial MD vulnerability.

RI PT

Acknowledgements This work was supported by the Netherlands Initiative Brain and Cognition (NIHC), a part of the Netherlands Organization for Scientific Research (NWO) (grant number 056-13-012). This work

SC

was also supported by the National Institutes of Health, grant P30 GM103328 (author GR).

M AN U

Financial Disclosures

AC C

EP

TE D

The authors report no biomedical financial interests or potential conflicts of interest.

18

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

References 1. American Psychiatric Association (1994): Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington DC: American Psychiatric Press.

RI PT

2. Downar J, Daskalakis ZJ (2013): New targets for rTMS in depression: a review of convergent evidence. Brain Stimul 6:231–40. 3. Riva-Posse P, Holtzheimer PE, Garlow SJ, Mayberg HS (2013): Practical considerations in the development and refinement of subcallosal cingulate white matter deep brain stimulation for treatment-resistant depression. World Neurosurg 80:25–34. 4. Price JL, Drevets WC (2010): Neurocircuitry of mood disorders. Neuropsychoph 35:192–216.

SC

5. Pizzagalli DA (2010): Frontocingulate Dysfunction in Depression: Toward Biomarkers of Treatment Response. Neuropsychoph:1–24.

M AN U

6. Hamani C, Mayberg H, Stone S, Laxton A, Haber S, Lozano AM (2010): The Subcallosal Cingulate Gyrus in the Context of Major Depression. Biological Psychiatry 69:301-308. 7. Greicius MD, Flores BH, Menon V, Glover GH, Solvason HB, Kenna H, et al. (2007): Resting-state functional connectivity in major depression: abnormally increased contributions from subgenual cingulate cortex and thalamus. Biol Psychiatry 62:429–37. 8. Guo H, Cheng C, Cao X, Xiang J, Chen J, Zhang K (2014): Resting-state functional connectivity abnormalities in first-onset unmedicated depression. Neural Regen Res 9:153–63.

TE D

9. Sheline YI, Price JL, Yan Z, Mintun MA (2010): Resting-state functional MRI in depression unmasks increased connectivity between networks via the dorsal nexus. Proc Natl Acad Sci 107:11020– 11025. 10. Sundermann B, Beverborg MO, Pfleiderer B (2014): Toward literature-based feature selection for diagnostic classification: a meta-analysis of resting-state fMRI in depression. Front Hum Neurosci 8:692.

EP

11. Wang L, Hermens DF, Hickie IB, Lagopoulos J (2012): A systematic review of resting-state functionalMRI studies in major depression. J Affect Disord 142:6–12.

AC C

12. Mulders PC, van Eijndhoven PF, Schene AH, Beckmann CF, Tendolkar I (2015): Resting state functional connectivity in major depressive disorder: A review. Neuroscience & Biobehavioral Reviews 56:330–344. 13. Eickhoff SB, Thirion B, Varoquaux G, Bzdok D (2015): Connectivity-based parcellation: Critique and implications. Hum Brain Mapp 36:4771-92. 14. Krubitzer L (1995): The organization of neocortex in mammals: are species differences so different? Trends Neurosci 18(9): 408-17. 15. Passingham RE, Stephan KE, Kötter R (2002): The anatomical basis of functional the cortex. Nat Rev Neurosci 3(8):606-16.

really

localization in

19

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

16. Samara Z, Evers EAT, Goulas A, Uylings BM, Rajkowska G, Ramaekers JG, et al. (2017): Human orbital and anterior medial prefrontal cortex: Intrinsic connectivity parcellation and functional organization. Brain Struct Funct 17:1378-2. 17. Girvan M, Newman MEJ (2002): Community structure in social and biological networks. Proc Natl Acad Sci U S A 99:7821–6.

RI PT

18. Rubinov M, Sporns O (2010): Complex network measures of brain connectivity: uses and interpretations. Neuroimage 52:1059–69.

19. Barnes KA, Cohen AL, Power JD, Nelson SM, Dosenbach YBL, Miezin FM, et al. (2010): Identifying Basal Ganglia divisions in individuals using resting-state functional connectivity MRI. Front Syst Neurosci 4:18.

SC

20. Goulas A, Uylings HBM, Stiers P (2012): Unravelling the intrinsic functional organization of the human lateral frontal cortex: a parcellation scheme based on resting state fMRI. J Neurosci 32:10238– 52.

M AN U

21. Barbas H, Pandya DN (1989): Architecture and intrinsic connections of the prefrontal cortex in the rhesus monkey. J Comp Neurol 286:353-375. 22. Ongür D, Price JL. (2000): The organization of networks within the orbital and medial prefrontal cortex of rats, monkeys and humans. Cereb Cortex 10:206–219. 23. Yeterian EH, Pandya DN, Tomaiuolo F, Petrides M (2012): The cortical connectivity of the prefrontal cortex in the monkey brain. Cortex 48:58–81.

TE D

24. Price JL, Drevets WC (2012): Neural circuits underlying the pathophysiology of mood disorders. Trends Cogn Sci 16:61–71. 25. Chen J, Zhou C, Wu B, Wang Y, Li Q, Wei Y, et al. (2013): Left versus right repetitive transcranial magnetic stimulation in treating major depression: a meta-analysis of randomised controlled trials. Psych Res 30;210(3):1260-4.

EP

26. Beck AT, Steer RA, Ball R, Ranieri W (1996): Comparison of Beck Depression Inventories-IA and -II in psychiatric outpatients. J Pers Assess 67:588–597. 27. Derogatis L, Melisaratos N (1983): The Brief Symptom Inventory: An introductory report. Psychological Medicine 13 :595-605.

AC C

28. Rush AJ, Trivedi MH, Ibrahim HM, Carmody TJ, Arnow B, Klein DN (2003): The 16-item Quick Inventory of Depressive Symptomatology (QIDS) Clinician Rating (QIDS-C) and Self-Report (QIDS-SR): A psychometric evaluation in patients with chronic major depression. Biol Psych 54:573-583. 29. Raven J, Raven JC, Court JH (2003): Manual for Raven's Progressive Matrices and Vocabulary Scales. Section 1: General Overview. San Antonio, TX: Harcourt Assessment. 30. Van Dijk KRA, Sabuncu MR, Buckner RL (2012): The influence of head motion on intrinsic functional connectivity MRI. Neuroimage 59:431–438. 31. Power JD, Barnes KA, Snyder AZ, Schlaggar BL, Petersen SE (2012): Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. Neuroimage 59:2142–2154.

20

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

32. Bandler R, Keay KA, Floyd N, Price J (2000): Central circuits mediating patterned autonomic activity during active vs. passive emotional coping. Brain Res Bull 53:95–104. 33. Ploner M, Lee MC, Wiech K, Bingel U, Tracey I (2010): Prestimulus functional connectivity determines pain perception in humans. Proc Natl Acad Sci U S A 107:355–360.

RI PT

34. Lamm C, Meltzoff AN, Decety J (2010): How do we empathize with someone who is not like us? A functional magnetic resonance imaging study. J Cogn Neurosci 22:362–376. 35. Zaki J, Ochsner KN, Hanelin J, Wager TD, Mackey SC (2007): Different circuits for different pain: patterns of functional connectivity reveal distinct networks for processing pain in self and others. Social neuroscience 2:276–291.

SC

36. Linnman C, Moulton EA, Barmettler G, Becerra L, Borsook D (2013): Neuroimaging of the Periaqueductal Gray : State of the Field. Neuroimage 60(1): 505–522.

M AN U

37. Mobbs D, Petrovic P, Marchant JL, Hassabis D, Weiskopf N, Seymour B, et al. (2007): When fear is near: threat imminence elicits prefrontal-periaqueductal gray shifts in humans. Science 317:1079–1083. 38. Dearing KF, Gotlib IH (2009): Interpretation of ambiguous information in girls at risk for depression. J Abnorm Child Psychol 37:79–91. 39. Grupe DW, Nitschke JB (2013): Uncertainty and anticipation in anxiety: an integrated neurobiological and psychological perspective. Nat Rev Neurosci 14:488–501. 40. Bush G, Luu P, Posner MI (2000): Cognitive and emotional influences in anterior cingulate cortex. Trends Cogn Sci 4:215–222.

TE D

41. Porro CA, Cettolo V, Francescato MP, Baraldi P (2003): Functional activity mapping of the mesial hemispheric wall during anticipation of pain. Neuroimage 19:1738–1747. 42. Critchley HD (2005): Neural mechanisms of autonomic, affective, and cognitive integration. J Comp Neurol 493:154–66.

EP

43. Grabenhorst F, Rolls ET, Parris BA (2008): From affective value to decision-making in the prefrontal cortex. Eur J Neurosci 28:9:1930-9. 44. Etkin A, Egner T, Peraza DM, Kandel ER, Hirsch J (2006): Resolving emotional conflict: a role for the rostral anterior cingulate cortex in modulating activity in the amygdala. Neuron 51:871–82.

AC C

45. Foland-Ross LC, Gotlib IH (2012): Cognitive and neural aspects of information processing in major depressive disorder: an integrative perspective. Front Psychol 3:489. 46. Roy M, Shohamy D, Wager TD (2012): Ventromedial prefrontal-subcortical systems and the generation of affective meaning. Trends Cogn Sci 16:147–56. 47. Egner T, Etkin A, Gale S, Hirsch J (2008): Dissociable neural systems resolve conflict from emotional versus nonemotional distracters. Cereb Cortex 18:1475–84. 48. Salomons TV, Johnstone T, Backonja M-M, Davidson RJ (2004): Perceived controllability modulates the neural response to pain. J Neurosci 24:7199–203. 49. Rey G, Desseilles M, Favre S, Dayer A, Piguet C, Aubry J-M, Vuilleumier P (2014): Modulation of brain response to emotional conflict as a function of current mood in bipolar disorder: preliminary findings from a follow-up state-based fMRI study. Psychiatry Res 223:84–93. 21

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

50. Herwig U, Kaffenberger T, Baumgartner T, Jäncke L (2007): Neural correlates of a “pessimistic” attitude when anticipating events of unknown emotional valence. Neuroimage 34:848–58. 51. Mayberg H (2003): Modulating dysfunctional limbic-cortical circuits in depression: towards development of brain-based algorithms for diagnosis and. Br Med Bull193–207.

RI PT

52. Piefke M, Weiss PH, Zilles K, Markowitsch HJ, Fink GR (2003): Differential remoteness and emotional tone modulate the neural correlates of autobiographical memory. Brain 126:650-68. 53. Foland-Ross LC, Hardin MG, Gotlib IH (2013): Neurobiological markers of familial risk for depression. Curr Top Behav Neurosci 14:181-206.

SC

54. Browning M, Holmes EA, Charles M, Cowen PJ, Harmer CJ (2012): Using attentional bias modification as a cognitive vaccine against depression. Biol Psych 72(7):572-9.

55. Viviani R (2013): Emotion regulation, attention to emotion, and the ventral attentional network. Front Hum Neurosci 7:746.

M AN U

56. Kerr N, Scott J, Phillips ML (2005): Patterns of attentional deficits and emotional bias in bipolar and major depressive disorder. British J Clinical Psychol 44:343–356. 57. Corbetta M, Shulman GL (2002): Control of goal-directed and stimulus-driven attention in the brain. Nat Rev Neurosci 3:201–215. 58. Caspers S, Eickhoff SB, Rick T, von Kapri A, Kuhlen T, Huang R, et al. (2011): Probabilistic fibre tract analysis of cytoarchitectonically defined human inferior parietal lobule areas reveals similarities to macaques. Neuroimage 58:362–80.

TE D

59. Caspers S, Schleicher A, Bacha-Trams M, Palomero-Gallagher N, Amunts K, Zilles K (2013): Organization of the human inferior parietal lobule based on receptor architectonics. Cereb Cortex 23:615–28. 60. Kim C, Chung C, Kim J (2010): Multiple cognitive control mechanisms associated with the nature of conflict. Neurosci Lett 476:156–160.

EP

61. Mars RB, Jbabdi S, Sallet J, O’Reilly JX, Croxson PL, Olivier E, et al. (2011): Diffusion-weighted imaging tractography-based parcellation of the human parietal cortex and comparison with human and macaque resting-state functional connectivity. J Neurosci 31:4087–100.

AC C

62. Goldapple K, Segal Z, Garson C, Lau M, Bieling P, Kennedy S, et al., (2004): Modulation of corticallimbic pathways in major depression: treatment-specific effects of cognitive behavior therapy. Arch Gen Psychiatry 61(1):34-41. 63. Fu CHY, Williams SCR, Cleare AJ, Scott J, Mitterschiffthaler MT, Walsh ND, et al. (2008): Neural responses to sad facial expressions in major depression following cognitive behavioral therapy. Biol Psychiatry 64:505–12. 64. Nieuwenhuis IL, Takashima A (2011): The role of the ventromedial prefrontal cortex in memory consolidation. Behav Brain Res 15;218(2):325-34. 65. Vincent JL, Snyder AZ, Fox MD, Shannon BJ, Andrews JR, Raichle ME, et al. (2006): Coherent spontaneous activity identifies a hippocampal-parietal memory network. J Neurophysiol 96(6):3517-31. 66. Margulies DS, Vincent JL, Kelly C, Lohmann G, Uddin LQ, Biswal BB, et al. (2009): Precuneus shares 22

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

intrinsic functional architecture in humans and monkeys. Proc Natl Acad Sci U S A 106:20069– 74. 67. Pearson JM, Heilbronner SR, Barack DL, Hayden BY, Platt ML (2012): Posterior Cingulate Cortex: Adapting Behavior to a Changing Environment. Trends Cogn Sci 15(4):143–151.

RI PT

68. Kravitz DJ, Saleem KS, Baker CI, Ungerleider LG, Mishkin M (2013): The ventral visual pathway: An expanded neural framework for the processing of object quality. Trends Cogn Sci 17:26–49. 69. Stiers P, Peeters R, Lagae L, Van Hecke P, Sunaert S (2006): Mapping multiple visual areas in the human brain with a short fMRI sequence. Neuroimage 29:74-89. 70. Zhang J, Wang J, Wu Q, Kuang W, Huang X, He Y, et al. (2011): Disrupted brain connectivity networks in drug-naive, first-episode major depressive disorder. Biol Psychiatry 70:334–42.

SC

71. Wang L, LaBar KS, Smoski M, Rosenthal MZ, Dolcos F, Lynch TR, et al. (2008): Prefrontal mechanisms for executive control over emotional distraction are altered in major depression. Psychiatry Res 163:143–55.

M AN U

72. Palomero-Gallagher N, Vogt BA, Schleicher A, Mayberg HS, Zilles K (2009): Receptor architecture of human cingulate cortex: evaluation of the four-region neurobiological model. Hum Brain Mapp 30(8):2336-55. 73. Rushworth MFS, Walton ME, Kennerley SW, Bannerman DM (2004). Action sets and decisions in the medial frontal cortex. Trends Cogn Sci 8:410-7. 74. Fox MD, Snyder AZ, Vincent JL, Corbetta M, Van Essen DC, Raichle ME (2005): The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc Natl Acad Sci U S A 102:9673–9678.

TE D

75. Li B, Liu L, Friston KJ, Shen H, Wang L, Zeng LL, et al. (2013): A treatment-resistant default mode subnetwork in major depression. Biol Psychiatry 74:48–54. 76. Yeo BT, Krienen FM, Sepulcre J, Sabuncu MR, Lashkari D, Hollinshead M, et al. (2011): The organization of the human cerebral cortex estimated by functional connectivity. J Neurophysiol 106:1125–1165.

EP

77. Goldberg II, Harel M, Malach R (2006): When the Brain Loses Its Self: Prefrontal Inactivation during Sensorimotor Processing. Neuron 50:329–339.

AC C

78. Falquez R, Couto B, Ibanez A, Freitag MT, Berger M, Arens EA, et al. (2014): Detaching from the negative by reappraisal: the role of right superior frontal gyrus (BA9/32). Front Behav Neurosci 9;8:165. 79. Fried I, Wilson C, MacDonald K, Behnke E (1998): Electric current stimulates laughter. Nature 391:650. 80. Campbell DW, Wallace MG, Modirrousta M, Polimeni JO, McKeen NA, Reiss JP (2015): The neural basis of humour comprehension and humour appreciation: The roles of the temporoparietal junction and superior frontal gyrus. Neuropsychologia 79:10-20. 81. Zadra JR, Clore GL (2011): Emotion and Perception: The Role of Affective Information. Wiley Interdiscip Rev Cogn Sci 2(6):676-685. 82. Zhang X, Zuo B, Erskine K, Hu T (2016): Feeling light or dark? Emotions affect perception of brightness. J Environ Psychol 47:107-11. 23

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

83. Horn DI, Yu C, Steiner J, Buchmann J, Kaufmann J, Osoba A, et al.(2010): Glutamatergic and resting- state functional connectivity correlates of severity in major depression - the role of pregenual anterior cingulate cortex and anterior insula. Front Syst Neurosci 15:4:33.

RI PT

84. Pezawas L, Meyer-Lindenberg A, Drabant EM, Verchinski BA, Munoz KE, Kolachana BS, et al. (2005): 5-HTTLPR polymorphism impacts human cingulate-amygdala interactions: a genetic susceptibility mechanism for depression. Nat Neurosci 8:828–34. 85. Sliz D, Hayley S (2012): Major Depressive Disorder and Alterations in Insular Cortical Activity: A Review of Current Functional Magnetic Imaging Research. Front Hum Neurosci 6:323.

SC

86. McGrath CL, Kelley ME, Holtzheimer PE, Dunlop BW, Craighead WE, Franco AR, et al. (2013): Toward a Neuroimaging Treatment Selection Biomarker for Major Depressive Disorder. JAMA Psychiatry 70(8): 821–829.

M AN U

87. Kaiser RH, Andrews-Hanna JR, Wager TD, Pizzagalli DA (2015): Large-scale network dysfunction in Major Depressive Disorder: Meta-analysis of resting-state functional connectivity. JAMA Psychiatry 72(6):603-611.

AC C

EP

TE D

88. Manoliu A, Meng C, Brandl F, Doll A, Tahmasian M, Scherr M, et al. (2013): Insular dysfunction within the salience network is associated with severity of symptoms and aberrant inter-network connectivity in major depressive disorder. Front Hum Neurosci 7:930.

24

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

Tables Table 1. Sample characteristics and clinical measures. FH

HC

RI PT

MD SD

Mean

SD

Mean

SD

F (2,106)

BDI-II

32.74

7.66

5.78

5.04

2.24

2.97

328.68

.000

a

.020

BSI GSI

1.67

0.55

0.34

0.38

0.14

0.17

159.24

.000

a

.098

BSI PST

38.66

7.67

12.42

10.16

6.5

7.27

147.18

.000

a

.010

BSI PSDI

2.26

0.46

1.18

0.39

0.89

QIDS

16.57

3.38

3.91

3.46

2.1

Age

38.88

11.8

34.03

% Female

71%

---

72%

Raven IQ

45.2

8.99

48.28

SC

Mean

p Values

108.07

.000a

1.84

252.58

.000

M AN U

0.41

b

.009b

a

14.77

36.21

16.65

0.98

.377

---

68%

---

χ (2) = .14

.930

6.5

46.03

7.46

1.53

.221

2

b

b

.030

TE D

MD, patient group; FH, at-familial risk group; HC, healthy controls; BDI, Beck Depression Inventory; BSI, Brief Symptom Inventory; GSI, global severity index; PST, positive symptom total; PSDI, positive a

symptom distress index; QIDS, Quick Inventory of Depressive Symptomatology. The MD group differs

EP

significantly from relatives and controls in the F test; bThe FH group differs significantly from controls in post-hoc comparisons. All post-hoc p’s in the post-hoc tests between MD vs HC and MD vs FH are

AC C

smaller than .000.

25

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

Table 2. Summary of MRI results. Anatomy labels are given according to Automated Anatomical Labeling (Tzourio-Mazoyer et al., 2002) and p values concern F tests (see Supplement for all p’s). Voxel Map

OMPFC area

Anatomy

Average MNI

Extent

p Value

Ant. Insula

Brainstem

3, -26, -15

73

.000005

MD Vulnerability

p32

Posterior cingulate

1, -30, 30

61

.000048

MD Vulnerability

p32

BA 40/39

44, -50, 45

61

.000027

MD Vulnerability

BA9

Intraparietal sulcus

-30, -72, 32

69

.000001

MD Vulnerability

d32

Occipital

23, -91, 20

82

.000001

MD Disease

47/12 (OFC)

Inferior temporal

48, -46, -18

53

.000001

MD Disease

47/12 (OFC)

Occipital (RH)

26, -92, 20

61

.000003

MD Disease

47/12 (OFC)

Occipital (LH)

-25, -87, 21

90

.000001

MD Disease

47/12 (OFC)

Paracentral Lobule

2, -47, 73

62

.000003

MD Disease

47/12 (OFC)

Caudate body (RH)

12, 3, 15

55

.000014

MD Disease

Ant. Insula

Cerebellum

29, -61, -36

76

.0000001

MD Disease

Ant. Insula

Fusiform (medial)

34, -72, -18

93

.000006

MD Disease

Ant. Insula

Putamen

-23, 5, 5

68

.000002

MD Disease

Ant. Insula

Caudate

13, 25, 2

61

.000538

MD Disease

Ant. Insula

Occipital (LH)

-34, -85, 3

54

.000003

MD Disease

p32

Precuneus (LH)

-8, -72, 33

60

.000000

MD Disease

p32

Precuneus (RH)

11, -71, 41

69

.000130

MD Disease

p24

Super. frontal gyrus

-27, 58, 15

72

.000269

MD Disease

a24'

DLPFC

-10, 42, 47

68

.000037

SC

M AN U

TE D

EP

AC C

RI PT

MD Vulnerability

MD Disease

a24'

Paracentral Lobule

-7, -25, 63

58

.000008

MD Disease

11 (OFC)

Precuneus

4, -66, 52

58

.000929

26

ACCEPTED MANUSCRIPT OMPFC connectivity in major depression – Samara et al.

Figures Figure 1. Functional connectivity changes associated with MD Vulnerability. Region nomenclature corresponds to (18). Voxel clusters of MD vulnerability related FC changes (MD, FH ≠ HC) are colored in

RI PT

the same color as the region that provided the seed for the FC map. A) Lateral view of the left hemisphere. B) Medial view of the left hemisphere. C) Lateral view of the right hemisphere. D) Posterior view of both hemispheres. E) Bar plots showing connectivity strength (Z(r)) in the three groups for the anterior insula (aI) and the rostral perigenual cingulate (p32) (see Supplement for all plots). Error bars

SC

represent 95% confidence intervals. MD, patient group; FH, at-risk group; HC, healthy controls.

M AN U

Figure 2. Functional connectivity abnormalities of MD Disease. Region nomenclature corresponds to (18). Voxel clusters of MD disease related FC changes (MD ≠ FH, HC) are color-coded according to the orbital or medial region that provided the seed for the FC map. A) Lateral view of the left hemisphere. B) Posterior-dorsal view of the left hemisphere. C) Medial view of the left hemisphere. D) Slices through the cerebellum and basal ganglia to show cluster of altered connectivity with the anterior insula region and the lateral orbital 47/12 region. D) Ventral view of both hemispheres. E) Ventral view of both

TE D

hemispheres. F) Bar plots showing connectivity strength (Z(r)) in the three groups for the anterior insula (aI) and the rostral cingulate (p32) (see Supplement for all plots). Error bars represent 95% confidence

EP

intervals. MD, patient group; FH, at-risk group; HC, healthy controls.

Figure 3. Summary representation of all connectivity findings. Regions in orange were found to have

AC C

altered FC in MD vulnerability, while regions in red were found to have altered FC in the acute MD phase. The anterior insula and the rostral cingulate (marked with asterisk) were found to have changed connectivity with various areas in the rest of the cortex both in MD vulnerability and during depressive episodes.

27

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT Samara et al.

Supplement

Orbital and Medial Prefrontal Cortex Functional Connectivity of Major Depression Vulnerability and Disease

RI PT

Supplementary Information

Design & Participants

Thirty-five MD patients meeting DSM-IV-Text Revision criteria for a nonpsychotic major depressive episode (cutoff BDI-II>20; mean age 38.89; SD 11.63; 71% female), thirty-six

SC

unaffected first-degree relatives of MD sufferers (mean age 34.03; SD 14.57; 72% female) and thirty-eight healthy, non-psychiatric controls (mean age 36.21; SD 16.43; 68% female)

M AN U

participated in the study after providing informed consent. The three groups did not differ significantly in terms of age (F (2, 106) = .98; p=.38), gender (χ2 (2) = 0.14; p=.93), or IQ (F (2, 106) = 1.53; p=.22). Nonetheless, we included age and gender as covariates in our fMRI analyses to further remove variability associated with these parameters (see below). Participants in all three

TE D

groups were recruited via advertisements in the local press; MD patients were also recruited from the regional institute for outpatient mental healthcare (RIAGG, Maastricht NL). MD patients recruited from the community were screened for axis I disorders with SCID-I (DSM-IV structured clinical interview) (1) and BDI-II by the experimenter and trained research assistants. Outpatients

EP

of the regional institute for mental healthcare were interviewed for axis I disorders with SCID-I at

AC C

their intake at the center. All aspects of our screening and experimental procedure were approved by the Medical Ethical Committee of the Academic hospital of Maastricht University and conducted in accordance with the University’s and Committee’s guidelines. MD patients were included if they met DSM-IV-Text Revision criteria for major depressive

disorder as the primary diagnosis and had a BDI-II score of >20 (moderate to severe depression) and excluded if they met criteria for Bipolar I or II, substance dependence or were taking benzodiazepines. MD patients with additional axis I diagnoses (except substance dependence and psychotic disorders) or taking other classes of antidepressants were allowed to participate.

1

ACCEPTED MANUSCRIPT Samara et al.

Supplement

Table S1 details the diagnostic and medication histories of MD participants. To exclude the possibility that our main results were driven by comorbidity and medication use in the MD sample, we filtered out from our results maps the effects of these confounds (for details see section on

RI PT

MRI analyses). Participants with familial MD were included if they had a first-degree relative with MD, as detailed in a screening questionnaire administered to them. The questionnaire included DSM-IV

SC

items for the diagnosis of MD and the differential diagnosis of Bipolar I, II and psychotic features. Participants in this group were excluded if they had ever been personally diagnosed with any axis

M AN U

I disorder and were assessed with the Symptoms-Checklist 90 (SCL-90) (2) for current psychiatric symptomatology (exclusion cutoff for males SCL-90 > 116, females SCL-90 > 130; above average population norms).

Healthy controls were excluded if they had ever been diagnosed with any axis-I disorder and if they had a first or second degree relative with psychiatric history. They were assessed with

TE D

the SCL-90 for current psychiatric symptomatology and were excluded based on the same cutoffs as the familial MD group. All participants were screened for the following somatic conditions and MRI contra-indications: neurological disorders, epilepsy, severe head injury, claustrophobia,

EP

pregnancy or lactation, metal implants, pacemakers and intrauterine contraceptive devices.

AC C

Magnetic Resonance Imaging Acquisition Scanning was conducted on a Siemens MAGNETOM Allegra 3Tesla MRI head-only

scanner. Head motion was constrained by the use of foam padding. Magnetic resonance imaging data, acquired in one scanning session for all participants, included a resting state EPI scan (repetition time=2.5 sec, 153 or 203 volumes) optimized for the reduction of susceptibility artefacts, a high-resolution T1-weighted anatomical scan and a field map. During the resting state scan, participants were instructed to fixate on a cross at the center of the screen, keep their eyes open and refrain from intentionally engaging in specific mental tasks or falling asleep. For each

2

ACCEPTED MANUSCRIPT Samara et al.

Supplement

subject, 153 or 203 T2*- weighted gradient echo planar images (EPI) with 41 slices were acquired. EPI can suffer substantial loss of BOLD sensitivity and geometric distortions due to magnetic field inhomogeneities near air tissue interfaces. In order to minimize MRI signal loss and recover the

RI PT

true spatial signal positions in the OFC we: a) used an optimized echo time, b) tilted the slices (~30o angle), and c) generated a field map to correct susceptibility-related signal displacements offline. Imaging parameters for the resting state sequence were as follows: TR, 2500ms; TE,

SC

25ms; flip angle, 90o; matrix size, 128 X 96; and FOV, 256mm; distance factor, 20%; resulting in a voxel size of 2X2X3mm. The gradient echo image used to generate the field map had the same

M AN U

grid and slice orientation as the functional images (TR, 704ms; TE, 5.11, 7.57 ms; flip angle, 60o). In order to enable the localization of functional data, a high-resolution T1-weighted image was acquired with the following parameters: TR, 2250ms; TE, 2.6ms; flip angle, 9o; FOV, 256mm; slice thickness, 1mm; matrix size, 256X256; number of slices, 192; voxel size, 1X1X1mm.

TE D

Preprocessing

Preprocessing of fMRI data was performed using SPM 5 software and MATLAB scripts (Mathworks, Natick, MA, USA). The functional data were subjected to the following preprocessing:

EP

slice time correction, spatial correction using the field map, realignment, co-registration with the anatomical scan, normalization to the Montreal Neurological Institute (MNI) 152 template,

AC C

reslicing to 2 mm isotropic voxels and smoothing with a 6 mm full width half maximum Gaussian kernel. The results of the co-registration and normalization steps were visually inspected to ensure accuracy of spatial matching. T1-weighted images were segmented into grey matter, white matter and CSF tissue maps and these maps were used later in the analyses. Further, we removed non-neuronal contributions from the BOLD signal by regressing the following nuisance variables: the six realignment parameters obtained by rigid body head motion correction, the time series extracted from cerebrospinal fluid and white matter, the session specific mean and the

3

ACCEPTED MANUSCRIPT Samara et al.

Supplement

intrinsic autocorrelations. The residual volumes of the multiple regression were Fourier band pass filtered (0.01 – 0.1 Hz).

RI PT

Frame-wise Displacement Correction Head motion has been shown to significantly distort measures of FC (underestimating long-range and overestimating short-range connectivity) while commonly used preprocessing

SC

steps (e.g. realignment of volumes to the first scan, regression of the realignment parameters) do not adequately counter its effect (3,4). In order to reduce the bias of head motion in our data we

M AN U

took the following approach: 1) we estimated the scan-to-scan head motion and identified scans during which the frame-wise displacement exceeded a particular threshold (absolute motion difference in the z direction > 0.4mm (1/10 of voxel size); rotation in the x direction > 0.26o (angle corresponding to 0.4mm z-displacement of frontopolar voxels, assuming the rotation point in the middle of the brain is 88mm from the anterior end of the brain’s frontal pole) (5), 2) we marked

TE D

and excluded the identified volumes together with the 1-back and the 2-forward frames (to avoid spin history assumptions’ violations caused by movement) (4), and 3) we included in the analyses only participants for whom at least 120 volumes (i.e. 5 minutes) (6) of resting data were available

EP

after the correction (mean duration, 6.4 min; SD, 0.8 min). To ensure that differences in our analyses were not better accounted for by the amount of data available after correcting for frame-

AC C

wise displacements, we quantified the amount of volumes available for FC analyses in each group and performed analysis of variance (p>.63). To ensure that differences in our main analyses could not be explained by differences in the remaining frame-wise displacement, we compared the postcleaning mean relative translation distance between the three groups using analysis of variance (p>.57).

4

ACCEPTED MANUSCRIPT Samara et al.

Supplement

fMRI FC Analysis To systematically examine connectivity changes in each OMPFC cortical field associated with MD, we generated whole-brain FC maps seeding from the center of mass of each of the 19

RI PT

subregions. The time-course of each seed was de-noised (Eigen decomposition) before calculating Pearson’s r between each seed’s time-course and the time-course of all voxels in the brain. Values in the whole-brain FC maps were r- to-Z transformed using Fisher’s formula.

SC

Subsequently, the FC maps of OMPFC field were subjected to an analysis of covariance (ANCOVA) with group as a factor and age, gender and IQ Raven score as covariates.

M AN U

To test for connectivity patterns of MD vulnerability, we created conjunction maps of the F contrasts between 1) the healthy controls and the MD patients and 2) the healthy controls and the familial MD participants. The F contrasts entered in SPM for the conjunction maps in this analysis were: 1) Healthy controls ~ MD patients (1, 0, -1) and 2) Healthy controls ~ familial MD (0, 1, -1). A third F contrast was created for: MD patients ~ familial MD (1, -1, 0). Thus, MD

TE D

vulnerability FC markers were defined as changes: 1) significantly different between the healthy controls and the MD patients and vulnerable individuals (conjunction map) and logical 2) not significantly different between MD patients and MD vulnerable individuals (third F map).

EP

Conjunction maps were tested for significance with Monte Carlo simulations (10,000 iterations; voxel-level p =.01, cluster-level p=.0025 – Bonferroni-adjusted for 19 independent comparisons).

AC C

To retain in the analysis only brain areas with altered FC in both the disease and the vulnerability state, we masked out of the conjunction maps all voxels significant at a lenient threshold (.001 uncorrected) in the F contrast between MD patients and MD vulnerable participants. To test for connectivity markers of MD disease, we created conjunction maps of the F

contrasts between 1) the MD patients and the healthy controls and 2) the MD patients and the MD vulnerable individuals. The F contrasts entered in SPM for the conjunction map in this analysis were: 1) MD patients ~ healthy controls (1, 0, -1) and 2) MD patients ~ familial MD (1, -1, 0). A third F contrast was created for: Healthy controls ~ familial MDD (0, 1, -1). Thus, MD disease FC

5

ACCEPTED MANUSCRIPT Samara et al.

Supplement

markers were defined as changes: 1) significantly different between the MD patients and the familial MD participants and healthy controls (conjunction map) and logical 2) not significantly different between familial MD and healthy controls (third F map). Conjunction maps were tested

RI PT

for significance with Monte Carlo simulations (10,000 iterations; voxel-level p =.01, cluster-level p=.0025 – Bonferroni-adjusted for 19 independent comparisons). To retain in the analysis only brain areas with altered FC in exclusively the disease state, we masked out from the conjunction

SC

maps all voxels significant at a lenient threshold (.001 uncorrected) in the F contrast between familial MD and controls.

M AN U

Since ANOVA is a general test for differences between groups, we determined the direction of FC differences in MD vulnerability and disease with ROI-based post-hoc tests in SPSS using the r-Z-transformed connectivity values of the statistically significant voxels in the conjunction maps. After being cluster-based corrected (Monte Carlo) the 19 FC conjunction maps (one for each OMPFC subregion) were used as binary masks to extract the Z values of each

TE D

individual in the three groups. Subsequently, analyses of variance and post-tests with Bonferroni adjusted p values were run in SPSS to establish the direction of the FC changes. Finally, to relate our FC markers to disease variables, correlation analyses for the MD group between the disease

EP

markers and the clinical measures were run in SPSS.

AC C

Control Analyses

To ensure that the reported connectivity differences were not driven by antidepressant

medication use and comorbidity in our MD sample, we regressed out of the individual maps variance associated with medication and comorbidity effects (see Supplementary Table S1 for details on axis I comorbid diagnoses and antidepressant medication profile use). We then used the regression’s residual, cleaned maps of the MD group in the across groups GLM. To ensure that the reported connectivity differences between the FH and the other 2 groups were due to trait characteristics and not state effects we regressed out variance

6

ACCEPTED MANUSCRIPT Samara et al.

Supplement

associated with depression symptoms (BDI-II total score) and general psychiatric complaints (BSI Positive Symptom Distress Index score). We then used the residuals as input in the across groups GLM for the FH group.

RI PT

Finally, to establish that our FC differences were not accounted for by probability of grey matter differences in the tested voxels we removed from our F maps all voxels significant in an ANOVA between all groups at a lenient threshold (.001 uncorrected). For this comparison we

SC

used the probability of the significant FC voxels/clusters being grey matter of the modulated and

M AN U

warped grey matter tissue masks created during the segmentation step of the preprocessing.

Functional Connectivity Changes in the Subgenual/Ventromedial Cortex One surprising outcome of the current study is that we did not find evidence for the involvement of the subgenual or ventromedial PFC in MD vulnerability or disease, while seminal studies report it and DBS is based on altering the connectivity around this OMFPC region.

TE D

Published evidence suggests that functional connectivity of the subgenual/ventromedial PFC is a state characteristic of MDD, and resolves with the resolution of the episode. Failure to replicate this finding across studies might be explained by methodological differences, e.g. seed

EP

coordinates used in seed-based studies, adequate signal intensity in the ventromedial PFC, normalization of brains to templates and/or smoothing might contaminate subgenual signal with

AC C

that from the ventral striatum, grey matter differences between patients and controls, etc. In our study we seeded from 2 clusters that cover for the most part the ventromedial PFC, cluster 5 and cluster 19. Neither of these showed differences in our analyses. We report here evidence for the involvement of the ACC at the level of the rostral/perigenual ACC and not more ventrally. To ensure this null finding was not related to our seed coordinates (which resulted from the previous parcellation scheme), we run control analysis using seeds from studies that have reported subgenual cortex activity or connectivity aberrations (7-9; dorsal and ventral seeds).

7

ACCEPTED MANUSCRIPT Samara et al.

Supplement

None of these analyses yielded significant differences between depressed patients and healthy controls. Aside from methodological differences, failure to replicate subgenual/ventromedial

RI PT

cortex abnormalities might be explained by MDD sample characteristics such as severity, i.e. DBS patients are by definition patients with chronic, severe and refractory depression (10) or clinical heterogeneity, i.e. subgenual/ventromedial PFC abnormalities might occur more frequently in

AC C

EP

TE D

M AN U

SC

MDD patients with significant symptoms of rumination (11) or comorbid anxiety (12,13).

8

ACCEPTED MANUSCRIPT Samara et al.

Supplement

Supplementary Figures & Tables Supplementary Table S1. Demographic and clinical data of the MD patient group F

31

F

59

F

49

M

34

F

Current Medications

Med Type

PTSD

Venlafaxine

SNRI

RI PT

22

Axis I Comorbidity

Social Phobia

18

F

25

F

49

F

20

M

52

M

31

F

23

F

Specific Phobia

52

F

Specific Phobia

47

M

44

M

25

F

36

F

39

F

52

M

44

M

GAD

Social Phobia PTSD

TE D

Social Phobia, GAD

Citalopram

SSRI SNRI, Antipsychotic SSRI, TetraCA Melatonergic

PTSD

Citalopram

SSRI

Escitalopram

SSRI

37

M

Specific Phobia, Social Phobia

38

F

Social Phobia

44

F

AC C

EP

F

54

NDRI

SSRI

49

43

Bupropion

Citalopram

F

34

TCA

Panic Disorder with agoraphobia

55

19

Nortriptyline

SSRI

F

38

SSRI

Paroxetine

F

22

Paroxetine

Citalopram, Mirtazapine Agomelatine

42

F

SSRI

SSRI

Venlafaxine, Seroquel

54

36

Citalopram Sertraline

SC

Gender

M AN U

Age

Social Phobia

F

F

F

GAD, Panic with agoraphobia, ADHD/ADD

Citalopram

SSRI

M

Panic Disorder with agoraphobia, Social Phobia

Escitalopram

SSRI

M

Social Phobia

F

Agoraphobia

44 F Panic disorder, OCD Fluvoxamine SSRI Abbreviations: PTSD (Post Traumatic Stress Disorder); GAD (Generalized Anxiety Disorder); ADHD/ADD (Attention Deficit Hyperactivity Disorder); OCD (Obsessive Compulsive Disorder); SNRI (Serotonin Norepinephrine Reuptake Inhibitor); SSRI (Selective Serotonin Reuptake Inhibitor); TCA (Tricyclic Antidepressant); NDRI (Norepinephrine Dopamine Reuptake Inhibitor); TetraCA (Tetracyclic Antidepressant).

9

ACCEPTED MANUSCRIPT Samara et al.

Supplement

Supplementary Table S2. Summary table of MRI results Map

OMPFC area

Anatomy

Average MNI

Post-Hoc p Value

p(Bonf.)

Post-Hoc 0.000237b

p(Bonf.)

Ant. Insula

Brainstem, PAG

3, -26, -15

0.000005

MD Vulnerability

p32

Dorsal PCC

1, -30, 30

0.000048

0.000091a

0.001494b

0.000116a

0.000363b

MD Vulnerability

p32

BA 40/39

44, -50, 45

0.000027

MD Vulnerability

BA9

IPS

-30, -72, 32

MD Vulnerability

d32

Occipital

23, -91, 20

MD Disease

47/12 (OFC)

IT

48, -46, -18

RI PT

MD Vulnerability

0.000014a

47/12 (OFC)

Occipital (RH)

26, -92, 20

MD Disease

47/12 (OFC)

Occipital (LH)

-25, -87, 21

MD Disease

47/12 (OFC)

Paracentral Lobule

2, -47, 73

MD Disease

47/12 (OFC)

Caudate body (RH)

12, 3, 15

0.000026a

0.000030b

0.000001

0.000004a

0.000030b

0.000001

0.000025c

0.000004d

0.000003

0.000061c

0.000016d

0.000001

0.000011c

0.000007d

0.000003

0.000046c

0.000013d

0.000014

0.000157c

0.000064d 0.000014d

SC

MD Disease

0.000001

Ant. Insula

Cerebellum

29, -61, -36

0.000001

MD Disease

Ant. Insula

Fusiform

34, -72, -18

0.000006

0.000381c

0.000010d

MD Disease

Ant. Insula

Putamen

-23, 5, 5

0.000002

0.000040c

0.000008d

MD Disease

Ant. Insula

Caudate

13, 25, 2

0.000538

0.000984c

0.000514d

-34, -85, 3

0.000003

0.000085c

0.000011d

M AN U

MD Disease

0.000007c

MD Disease

Ant. Insula

Occipital (LH)

MD Disease

p32

Precuneus (LH)

-8, -72, 33

0.000000

0.000000c

0.000036d

MD Disease

p32

Precuneus (RH)

11, -71, 41

0.000130

0.000352c

0.001185d

MD Disease

p24

aPFC

-27, 58, 15

0.000269

0.000082c

0.000097d

-10, 42, 47

0.000037

0.000168c

0.000290d

a24'

DLPFC

MD Disease

a24'

Paracentral Lobule

-7, -25, 63

0.000008

0.000124c

0.000034d

MD Disease

11 (OFC)

Precuneus

4, -66, 52

0.000929

0.004654c

0.002322d

TE D

MD Disease

AC C

EP

Post-hoc Bonferroni corrected p values a and b concern the comparisons MD patient group versus HC and familial MD group versus HC respectively for the MD vulnerability markers. Post-hoc Bonferroni corrected p values c and d concern the comparisons MD patient group versus the familial MD group and MD patient group versus HC respectively for the MD disease markers.

10

ACCEPTED MANUSCRIPT Samara et al.

Supplement

Supplementary Table S3. Mean and standard deviations of functional connectivity per group. MD Map

OMPFC area

Anatomy

FH

HC

Mean

SD

Mean

SD

Mean

SD

0.08

0.11

0.06

0.09

-0.04

0.11

Ant. Insula

Brainstem, PAG

MD Vulnerability

p32

dPCC

0.02

0.13

0.04

0.09

0.14

0.12

MD Vulnerability

p32

BA 40/39

-0.10

0.13

-0.09

0.10

0.02

0.11

MD Vulnerability

BA9

IPS

-0.06

0.12

MD Vulnerability

d32

Occipital

0.02

0.07

MD Disease

47/12 (OFC)

IT

0.11

0.10

MD Disease

47/12 (OFC)

Occipital (RH)

0.06

0.10

MD Disease

47/12 (OFC)

Occipital (LH)

0.05

0.07

MD Disease

47/12 (OFC)

Paracentral Lobule

0.09

0.09

MD Disease

47/12 (OFC)

Caudate body (RH)

-0.03

MD Disease

Ant. Insula

Cerebellum

0.08

MD Disease

Ant. Insula

Fusiform

0.04

MD Disease

Ant. Insula

Putamen

MD Disease

Ant. Insula

Caudate

MD Disease

Ant. Insula

Occipital (LH)

MD Disease

p32

MD Disease

RI PT

MD Vulnerability

0.08

0.04

0.08

0.01

0.10

-0.08

0.08

-0.01

0.10

-0.01

0.11

-0.04

0.09

-0.05

0.11

-0.05

0.10

-0.06

0.08

-0.02

0.09

-0.03

0.12

0.12

0.07

0.07

0.08

0.11

0.10

-0.01

0.06

-0.01

0.08

0.12

-0.06

0.10

-0.09

0.09

M AN U

SC

-0.06

0.11

0.05

0.09

0.04

0.09

0.07

0.16

-0.05

0.10

-0.06

0.15

0.07

0.08

-0.02

0.09

-0.03

0.10

Precuneus (LH)

-0.02

0.09

0.11

0.08

0.08

0.10

p32

Precuneus (RH)

-0.09

0.13

0.02

0.11

0.00

0.10

MD Disease

p24

aPFC

0.18

0.14

0.07

0.08

0.07

0.10

MD Disease

a24'

DLPFC

0.07

0.13

-0.05

0.10

-0.04

0.12

MD Disease

a24'

Paracentral Lobule

0.04

0.12

-0.07

0.11

-0.08

0.12

MD Disease

11 (OFC)

0.10

0.10

0.01

0.10

0.01

0.13

TE D

0.15

AC C

EP

Precuneus

11

ACCEPTED MANUSCRIPT Supplement

SC

RI PT

Samara et al.

AC C

EP

TE D

M AN U

Supplementary Figure S1. Bar plots for FC differences in MD vulnerability. Region nomenclature and letters correspond to main Figure 1.

Supplementary Figure S2. Bar plots for FC differences in MD disease. Region nomenclature and letters correspond to main Figure 2.

12

ACCEPTED MANUSCRIPT Samara et al.

Supplement

SC

RI PT

C03

C11

AC C

EP

C10

TE D

M AN U

C08

Supplementary Figure S3, continued next page.

13

ACCEPTED MANUSCRIPT Samara et al.

Supplement

C13

C14

C15

EP

TE D

M AN U

SC

RI PT

C12

AC C

Supplementary Figure S3. F maps of effects of medication and comorbidity in the MD sample using multiple linear regression at p< .05 (uncorrected). Nuisance effects are depicted in all maps in blue while regions with statistically significant altered functional connectivity (corresponding to manuscript results in Figure 2- markers of MD disease state) are shown as either red or yellow.

14

ACCEPTED MANUSCRIPT Samara et al.

Supplement

C10

M AN U

SC

RI PT

C08

C14

EP

TE D

C11

AC C

Supplementary Figure S4. F maps of effects of depression (BDI-II) and general psychopathology (BSI PSDI) in the FH sample using multiple linear regression at p< .05 (uncorrected). Nuisance effects are depicted in all maps in blue while regions with statistically significant altered functional connectivity (corresponding to manuscript results in Figure 1- markers of MD vulnerability) are shown in red.

15

ACCEPTED MANUSCRIPT Samara et al.

Supplement

Supplemental References 1. First MB, Spitzer RL, Gibbon M, Williams JBW (2002): Structured Clinical Interview for DSMIV-TR Axis I Disorders, Research Version, Patient Edition. (SCID-I/P) New York: Biometrics Research, New York State Psychiatric Institute.

RI PT

2. Derogatis LR, Savitz KL (1999): The SCL-90-R, Brief Symptom Inventory and Matching Clinical Rating Scales. In Maruish ME, The use of psychological testing for treatment planning and outcomes assessment. Philadelphia: Lawrence Erlbaum: 679-724. 3. Van Dijk KRA, Sabuncu MR, Buckner RL (2012): The influence of head motion on intrinsic functional connectivity MRI. Neuroimage. 59:431–438.

SC

4. Power JD, Barnes KA, Snyder AZ, Schlaggar BL, Petersen SE (2012): Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. Neuroimage. 59:2142–2154.

M AN U

5. Talairach T, Tournoux P (1988): Co-planar stereotaxic atlas of the human brain. Thieme Medical Publishers Inc: New York. 6. Van Dijk KRA, Hedden T, Venkataraman A, Evans KC, Lazar SW, Buckner RL (2010): Intrinsic functional connectivity as a tool for human connectomics: theory, properties, and optimization. J Neurophysiol. 103:297–321.

TE D

7. Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwald JM, Kennedy SH (2005): Deep brain stimulation for treatment-resistant depression. Neuron 45(5):651-660. 8. Fox MD, Buckner RL, White MP, Greicius MD, Pascual-Leone A (2012): Efficacy of transcranial magnetic stimulation targets for depression is related to intrinsic functional connectivity with the subgenual cingulate. Biol Psych 72(7):595-603.

AC C

EP

9. Connolly CG, Wu J, Ho TC, Hoeft F, Wolkowitz O, Eisendrath S, Paulus MP (2013): RestingState Functional Connectivity of Subgenual Anterior Cingulate Cortex in Depressed Adolescents. Biol Psych 74(12):898–907. 10. Berlim MT, McGirr A, Van den Eynde F, Fleck MPA, Giacobbe P (2014): Effectiveness and acceptability of deep brain stimulation (DBS) of the subgenual cingulate cortex for treatmentresistant depression: A systematic review and exploratory meta-analysis. J Affect D 159:31-38. 11. Hamilton P, Farmer M, Fogelman P, Gotlib IH (2015): Depressive rumination, the defaultmode network, and the dark matter of clinical neuroscience. Biol Psych 78(4):224-230. 12. Gold PW (2015): The organization of the stress system and its dysregulation in depressive illness. Mol Psych 20:32-47.

16

ACCEPTED MANUSCRIPT Samara et al.

Supplement

AC C

EP

TE D

M AN U

SC

RI PT

13. Jaworska N, Yücelc K, Courtrightb A, MacMasterb FP, Sembob M, MacQueen G (2016): Subgenual anterior cingulate cortex and hippocampal volumes in depressed youth: the role of comorbidity and age. J Affect D 190:726-732.

17