Neuropathological and neuromorphometric abnormalities in bipolar disorder: View from the medial prefrontal cortical network

Neuroscience and Biobehavioral Reviews 42 (2014) 132–147

Contents lists available at ScienceDirect

Neuroscience and Biobehavioral Reviews journal homepage: www.elsevier.com/locate/neubiorev

Review

Neuropathological and neuromorphometric abnormalities in bipolar disorder: View from the medial prefrontal cortical network Jonathan B. Savitz a,b,∗ , Joseph L. Price c , Wayne C. Drevets a,d a

Laureate Institute for Brain Research, Tulsa, OK, USA Faculty of Community Medicine, University of Tulsa, Tulsa, OK, USA Department of Anatomy and Neurobiology, Washington University School of Medicine, St. Louis, MO, USA d Janssen Pharmaceuticals of Johnson & Johnson, Inc., Titusville, NJ, USA b c

a r t i c l e

i n f o

Article history: Received 7 October 2013 Received in revised form 20 January 2014 Accepted 19 February 2014 Keywords: Depression Bipolar disorder Neuropathology Magnetic resonance imaging Neuron Oligodendrocyte Postmortem Gene expression Medial prefrontal cortex Developmental

a b s t r a c t The question of whether BD is primarily a developmental disorder or a progressive, neurodegenerative disorder remains unresolved. Here, we review the morphometric postmortem and neuroimaging literature relevant to the neuropathology of bipolar disorder (BD). We focus on the medial prefrontal cortex (mPFC) network, a key system in the regulation of emotional, behavioral, endocrine, and innate immunological responses to stress. We draw four main conclusions: the mPFC is characterized by (1) a decrease in volume, (2) reductions in neuronal size, and/or changes in neuronal density, (3) reductions in glial cell density, and (4) changes in gene expression. These data suggest the presence of dendritic atrophy of neurons and the loss of oligodendroglial cells in BD, although some data additionally suggest a reduction in the cell counts of specific subpopulations of GABAergic interneurons. Based on the weight of the postmortem and neuroimaging literature discussed herein, we favor a complex hypothesis that BD primarily constitutes a developmental disorder, but that additional, progressive, histopathological processes also are associated with recurrent or chronic illness. Conceivably BD may be best conceptualized as a progressive neurodevelopmental disorder. © 2014 Elsevier Ltd. All rights reserved.

Contents 1. 2. 3.

4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. The subgenual ACC (sgACC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. The pregenual ACC (pgACC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. The supragenual ACC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. The orbitofrontal cortex (OFC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. BA 9 of the dorsolateral prefrontal cortex (DLPFC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Abnormalities associated with BD postmortem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Limitations of the postmortem literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Should the abnormalities observed postmortem in BD be described as neuropathology? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4. Evidence for a neurodevelopmental etiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5. Evidence for a neurodegenerative etiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.6. Clinical implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.7. Future directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

∗ Corresponding author at: Laureate Institute for Brain Research, 6655 South Yale Avenue, Tulsa, OK 74136, USA. Tel.: +1 918 502 5104; fax: +1 918 502 5135. E-mail address: [email protected] (J.B. Savitz). http://dx.doi.org/10.1016/j.neubiorev.2014.02.008 0149-7634/© 2014 Elsevier Ltd. All rights reserved.

133 133 133 133 135 135 136 137 139 139 139 140 140 141 141 142

5.

J.B. Savitz et al. / Neuroscience and Biobehavioral Reviews 42 (2014) 132–147

133

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conflict of interest statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

143 143 143 143

1. Introduction The monoamine-deficiency hypothesis of mod disorders continues to receive empirical support (Hasler et al., 2008; Roiser et al., 2009; Savitz et al., 2013b). Nevertheless, bipolar disorder (BD) and other forms of affective illness increasingly are viewed as neuropathological conditions as more sensitive methods of detecting histological abnormalities are employed at postmortem. This conceptual shift has been driven by evidence of abnormalities of neuronal and glial cells at postmortem, as well as evidence of dendritic and neuronal atrophy in preclinical chronic stress models that serve as putative analogs of depression, MRI-defined reductions in gray matter (GM) volume, and cognitive deficits that do not fully resolve with improvement in mood (Savitz and Drevets, 2009a; Savitz et al., 2005a). Congenital abnormalities are one possible explanation for the neurophysiological changes observed in BD, but the neurotrophic effects of mood stabilizing medications like lithium (Moore et al., 2000; Savitz et al., 2010), coupled with longitudinal studies demonstrating volumetric changes over time, also raise the possibility that BD is underpinned by a (presumably) excitotoxicity-mediated histopathological process (Savitz and Drevets, 2009a). Here we review the evidence for, and the nature of the neuropathological changes in the medial prefrontal cortex (mPFC) network in primary BD and interpret these data within the context of the structural MRI, diffusion tensor imaging, and magnetic resonance spectroscopy literature. Based on these data we then discuss the phenomenological but clinically important issue of whether the BD-associated changes observed postmortem should be considered to be a form of neuropathology, and if so whether BD can be considered to be a progressive, neurodegenerative illness. Here we use the term “neuropathology” in the same sense as (Harrison, 2002), i.e. morphometric abnormalities of circuits, neurons, glia and synapses rather than abnormalities solely of receptors and/or neurochemistry. We do not claim that the mPFC is the only region of the brain that shows neuropathological or neurophysiological abnormalities in BD. In contrast, based on our reading of the literature as well as the conclusions of other reviews (Gigante et al., 2011; Harrison, 2002), the abnormalities observed in BD postmortem do not differ qualitatively between subcortical and cortical regions. Further, because of widespread evidence implicating the mPFC in BD and the importance of the mPFC in regulating bodily homeostasis and adaptation to stress along with neurophysiological and neuroendocrine responses to stress, we hypothesize that the mPFC serves as a reasonable vehicle for discussing the broader implications of the nature of the neuropathological changes in BD. 2. Methods Relevant studies published in English were identified through a MEDLINE search, National Library of Medicine, NIH (http://www.pubmed.gov) and cross-referenced papers in the field. The following key words were used: “bipolar disorder”, “postmortem”, “prefrontal cortex”, “anterior cingulate”, “orbitofrontal cortex”. Morphometric postmortem studies are emphasized in this review. That is, we included studies that measured GM volume, and the number and/or density of neurons or glial cells in BD patients and controls. Neuroimaging studies and postmortem studies of gene expression were not systematically reviewed but

were included where relevant to the interpretation or evaluation of morphometric postmortem data. Similarly, the schizophrenia and major depressive disorder (MDD) postmortem literature was beyond the focus of this review and these studies were only discussed in order to contextualize the findings in BD or in cases that samples of both unipolar and bipolar depressives were dominated by the latter subgroup. 3. Results The anterior cingulate cortex (ACC) carries out a diverse array of integrative functions. It is often heuristically divided into dorsal “cognitive”, and ventral “affective” streams. The dorsal ACC (dACC) lies along the superior portion of the cingulate sulcus running dorsal to the corpus callosum (CC), while the regions ventral and/or anterior to the genu of the CC comprise the ventral ACC. A general heuristic is that the dACC forms part of an “executive” attention system that supports response selection, error detection and performance monitoring while the ventral ACC regulates emotional and visceromotor responses (Bush et al., 2000). 3.1. The subgenual ACC (sgACC) Drevets et al. (1997) first demonstrated a reduction of GM volume, cerebral blood flow (CBF), and glucose metabolism in the mPFC ventral to the corpus callosum genu (“subgenual” ACC) in patients with BD and major depressive disorder MDD relative to healthy controls (Fig. 1). The reduction in GM has since been replicated by a number of independent groups and appears to apply to both males and females, individuals scanned early in the course of illness, as well as patients with affective psychosis and bipolar spectrum illness (Drevets et al., 2008); although the abnormality may be specific to, or at least more salient in familial cases (Drevets et al., 1997; Hirayasu et al., 1999; Koo et al., 2008; McDonald et al., 2004). Further, chronic lithium treatment, which exerts robust neurotrophic effects in animal models (Moore et al., 2000), largely normalizes subgenual ACC (sgACC) volume in treatment responders (Moore et al., 2009). These data are supported by magnetic resonance spectroscopy (MRS) studies which find that higher levels of N-acetylaspartate (NAA) in the sgACC, a marker of neuronal integrity, are associated with lithium treatment (Moore and Galloway, 2002; Forester et al., 2008), potentially consistent with the neurotrophic effects associated with chronic lithium administration in preclinical studies. Consistent with the weight of the volumetric imaging data, a 3-D stereological study reported a reduction in the number of nisslstained glia identified morphometrically together with an increase in neuronal density in the sgACC of two independent samples of patients with familial BD as well as familial MDD (Ongur et al., 1998). The groups were similar in age, sex, postmortem interval, and brain pH (an indicator of premortem acidosis that can confound postmortem measurements) although the storage time in fixative was significantly shorter for control brains than for BD brains. Medication effects cannot be ruled out since the majority of subjects in the MDD group were receiving fluoxetine and/or tricyclic antidepressants whereas the BD subjects typically were receiving lithium and/or anticonvulsants. However, both the MDD and the BD groups showed reductions in glial number, suggesting that the reduction in glial cells is not secondary to treatment with specific

134

J.B. Savitz et al. / Neuroscience and Biobehavioral Reviews 42 (2014) 132–147

Fig. 1. Adapted from Ongur et al. (2003) shows the perigenual anterior cingulate cortex region. The yellow and orange shading represent the approximate boundaries of the posterior (BA25) and anterior (BA24) subgenual ACC. The blue shading shows the approximate boundaries of the pregenual ACC, and the green shading, the approximate boundaries of the supragenual ACC. Numbers and associated letters refer to the cytoarchitectonic divisions adapted to Brodmann by Ongur et al. (2003). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

classes of medication. A more recent 2-D immunohistochemistry study did not find a significant decrease in oligodendrocyte number in the sgACC of BD, MDD or schizophrenia samples (Williams et al., 2013). The authors used cresyl hematoxylin and glial fibrillary acidic protein (GFAP) for oligodendrocyte and astrocyte identification, respectively. There was however, a significant decrease in the oligodendrocyte to astrocyte ratio in all 3 patient groups compared with the control group, although this change in oligodendrocyte to astrocyte ratio appeared to be driven by increases in astrocyte number in the patient groups rather than decreases in oligodendrocyte number. The BD group had a longer storage time in fixative than the other 3 subject groups although the storage time did not correlate with oligodendrocyte or astrocyte density (Williams et al., 2013). The medication taken by the patients was not provided and crucially, familial versus non-familial cases of depression were not distinguished from each other. The reported reduction in glial cells which most clearly has implicated the perineuronal and myelinating oligodendroglia (see below), potentially is consistent with the results of diffusion tensor imaging (DTI) studies that have produced evidence of structural abnormalities of white matter (WM) tracts connecting the sgACC (a.k.a., subcallosal gyrus) with limbic nuclei. Wang et al. (2008) reported a decrease in fractional anisotropy (FA) of the anterior cingulum. Furthermore, the strength of the connectivity between the amygdala and perigenual ACC measured with fMRI was found to correlate with the structural integrity of WM fibers as measured by DTI (Wang et al., 2009). FA is a measure of the extent to which the diffusion of water is constrained by fibers. Thus a reduction in FA may be indicative of a loss of structural integrity of WM fibers. The Wang et al. study is potentially congruent with two other studies that reported a greater number of reconstructed white matter

fibers (Houenou et al., 2007) and increased FA (Versace et al., 2008) in the left uncinate fasciculus in BD. The uncinate fasciculus connects the orbitofrontal cortex (OFC) and ventromedial PFC areas that include the sgACC with the amygdala and hippocampus. The latter study also found reduced FA in the right uncinate fasciculus in BD cases suggesting an imbalance in left versus right hemisphere processing of emotion (Versace et al., 2008). McIntosh et al. (2008) and Sussmann et al. (2009) found a bilateral reduction in FA in the uncinate fasciculi of BD patients, while Kafantaris et al. (2009) reported increased apparent diffusion coefficient (ADC), a measure of water mobility, in both the left and right sgACC in adolescents with BD. There are two reports of neuronal abnormalities in the sgACC at postmortem. Chana et al. (2003) conducted a 2-D morphometric analysis to examine the density and size of neurons and glia in the sgACC in patients with MDD, BD, schizophrenia and controls (15 subjects per group). The authors detected a decrease in neuronal somal size in layer V in all the patient groups as well as an increase in neuronal density in layer VI in the BD and schizophrenia groups after statistically controlling for gender, fixation time, and postmortem interval. No difference in glial cell size or density was found in the BD group relative to the control group. Because neuronal size is correlated with dendritic arborization, the decrease in neuronal size is interpreted by the authors to reflect a loss of neuropil, particularly the processes of the larger pyramidal neurons (Chana et al., 2003). Using a stereological counting approach Bouras et al. (2001) reported that autopsied individuals with sporadic BD (N = 21) displayed a significant reduction in cortical thickness and neuronal density (15–20%) in layers III, V and VI of the left sgACC compared with healthy controls (N = 55). There was no significant difference in mean pyramidal neuron size. All individuals were drug naïve or had received psychotropic medication for less than 6 months, and had no history of substance abuse. Methodological limitations prevented the analysis of individual neuron volume or total neuron numbers. Nevertheless, immunocytochemical analysis showed decreased levels of the microtubule-associated proteins, MAP1B and MAP2, which putatively is indicative of dendritic or axonal atrophy (Bouras et al., 2001). The reason for the discrepancy in the results of Chana et al. (2003) and Bouras et al. (2001) with respect to neuronal size in cortical layer V and neuronal density in cortical layer VI remains unclear. Besides methodological differences across the two studies (2-D versus 3-D, see Section 4), the sample of Bouras et al. (2001) was largely medication-free and consisted of cases without a known family history of mood disorders, whereas the majority of the subjects in the Chana et al. (2003) sample had been treated with medication and only 6 subjects of the 15 BD subjects had a positive family history of mood disorders. Table 1 Summary of the postmortem and the neuroimaging findings in BD. Region

sgACC

pgACC

Supragenual ACC

OFC

BA 9

Glutamatergic neuron density Glutamatergic neuron number Glutamatergic neuron size GABAergic neuron density GABAergic neuron number GABAergic neuron size Glial cell density Glial cell number GM volume WM pathology (DTI) Metabolic (MRS)

↓+ − ↓+ ↓+ − − − ↓+ ↓ ++ ++ −

− − − ↓ ++ ↓+ − − − ↓+ − +

− − − ↓+ − − − − − − −

− − − ↑+ − − − − ↓ ++ + −

↓+ − ↓+ ↓ ++ ↓+ − ↓ ++ ↓+ − − +

Notes: sgACC = subgenual anterior cingulate cortex, pgACC = pregenual ACC, OFC = orbital frontal cortex, BA 9 = Brodmann’s area 9, DTI = diffusion tensor imaging, MRS = magnetic resonance spectroscopy. ↑ increased, ↓ decreased, + modest evidence, ++ strong evidence, − no or minimal evidence.

J.B. Savitz et al. / Neuroscience and Biobehavioral Reviews 42 (2014) 132–147

Summary: Neuroimaging studies are indicative of a reduction in sgACC volume and structural abnormalities of WM tracts connecting the sgACC with limbic nuclei. There is some evidence for a reduction in glial cells at postmortem that may at least partly explain the DTI results of reduced FA in the white matter tracts associated with the sgACC. A decrease in neuronal size and both increases and decreases in neuronal density have been reported that could possibly be secondary to a decrease in neuronal arborization or a decrease in cell numbers (Table 1). The sgACC shares cytoarchitectural and connectional (e.g. with respect to the amygdala) similarities with the ACC situated anterior to the corpus callosum genu (i.e. “pregenual” ACC) suggesting that distinctions of the cortex at the actual sgACC/pgACC interface may be arbitrary. Nevertheless, since the pgACC and sgACC are often discussed separately in the literature, we will follow this framework here. 3.2. The pregenual ACC (pgACC) Fornito et al. (2008) reported that their BD group had reduced volume of the left paracingulate gyrus in the pgACC compared with the healthy control group. In a similar vein, Matsuo et al. (2009a) found an inverse relationship between volume of the left pgACC and a psychometric measure of impulsivity, a personality trait associated with BD (Savitz and Ramesar, 2006). In another study, adolescents with BD who were scanned at baseline and again two years later, manifested greater volume reductions in the left pgACC than healthy controls (Kalmar et al., 2009). Benes and colleagues found that neuronal terminals immunoreactive to glutamic acid decarboxylase (GAD65 ), the rate-limiting enzyme responsible for the conversion of glutamate to GABA, were decreased in the upper cortical layers of the pgACC in a BD sample (Benes et al., 2000). The study was limited by the small sample size (5 BD and 12 controls), however the results were potentially congruent with the subsequent finding of a decrease in neuronal density of non-pyramidal neurons in layer II of the pgACC in BD and schizophrenia (Benes et al., 2001). In this 2-D counting study, 10 BD subjects, 11 schizophrenic subjects, and 12 controls were matched for age although the BD group had a longer postmortem interval and a different sex ratio than the schizophrenic and control groups, and there was a significant correlation between postmortem interval and the density of non-pyramidal neurons in the BD group. Nevertheless, the results remained significant when age, postmortem interval and fixation interval were controlled for statistically. The BD patients with and without neuroleptic exposure also did not differ from each other in non-pyramidal neuron density. Follow-up studies by the same group using in situ hybridization, supported the hypothesis that BD-associated abnormalities in the pgACC appear to be specific to GABAergic cells (Woo et al., 2004). A decrease in neuronal density of GABAergic interneurons expressing the NR2A subunit of the NMDA receptor was reported in layer II of the pgACC in BD subjects and layers II and V in subjects with schizophrenia (Woo et al., 2004). There was no significant difference in neuronal density of GAD67 -containing neurons that did not express the NRA2 receptor. In this study, 17 BD subjects were matched with 17 schizophrenic subjects and 17 controls for age, sex, postmortem interval and pH. The results of the Woo et al. (2004) report were not however, echoed by genetic transcriptomics analyses in a more recent study (Woo et al., 2008b). In this latter study, the authors examined the mRNA expression of GABAergic interneurons that contain the calcium-binding protein, calbindin (CB) since these neurons are preferentially localized to layer II of the pgACC, where Woo et al. (2004) had reported a reduction of neuronal density. Twenty BD subjects, 20 schizophrenic subjects, and 20 controls were matched for age, sex, postmortem interval and pH (Woo et al., 2008b). Although the density of double-labeled CB+NR2A+ neurons

135

were decreased in the schizophrenic samples, BD subjects did not differ from controls in neuronal density or NR2A mRNA expression. However, because only ∼10% of all CB mRNA-containing cells express NR2A mRNA, conceivably small changes in the magnitude of mRNA expression would not be detectable by in situ hybridization (Woo et al., 2008b). In an independent sample, the density of GAD67 -containing GABAergic neurons that express the GluR5 subunit of the kainate receptor was found to be decreased by ∼40% in layer II of the pgACC in BD and schizophrenia (Woo et al., 2007). No change in mRNA expression was observed in GluR6-containing neurons; the GluR6 receptor is expressed predominantly on pyramidal cells. Further, the density of neurons that did not contain GAD67 mRNA (i.e. which predominantly consist of pyramidal neurons) but did express GluR5 mRNA was not altered. Another study found a nominal decrease in mRNA expression of GAD67 in layer II of the pgACC in patients with BD although the decrease in the level of GAD67 expression did not reach statistical significance (Thompson et al., 2009). In this study there were no significant correlations between GAD67 mRNA levels and age, postmortem interval, storage time, or estimated neuroleptic exposure. Although pH correlated significantly with GAD67 mRNA expression, there were no significant group differences in pH. It is unclear if these abnormalities in mRNA expression are confined to GABAergic neurons. In one study 10 “clinically depressed” subjects were matched with 10 controls for age, brain pH, postmortem interval, and RNA integrity. The authors found reduced immediate early gene expression in the depressed cases at postmortem (Covington et al., 2010). In a parallel study, the authors measured mRNA expression in the ventral mPFC of mice following chronic social defeat stress, and detected the same reduction in immediate early gene expression as observed in the depressed subjects at postmortem (Covington et al., 2010). Further, optogenetic stimulation of both GABAergic and glutamatergic neurons in the mouse mPFC increased gene expression and neuronal activity within this region, normalizing depressive behavior (Covington et al., 2010). In a proton MRS study Ongur et al. (2008) detected a higher glutamine–glutamate (GLN/GLU) ratio in the pgACC of manic patients. Since glutamate is converted to glutamine in glial cells, and glutamine is converted to glutamate in neurons, Ongur et al. (2008) interpreted their data as reflecting a breakdown in neuronal–glial cell interactions. Glial cell pathology was also suggested by Mosebach et al. (2013) who reported a non-significant decrease in oligodendrocyte densities in the WM adjacent to the pgACC in BD and MDD but not schizophrenia. The authors also reported an increase in Olig-1 expression in the WM which they interpret to reflect a compensatory response to oligodendrocyte dysfunction. Summary: There is evidence to suggest that at least some populations of patients with BD are characterized by reduced volume of the pgACC measured with MRI, a reduction in GABAergic neuron density, and/or functional abnormalities of GABAergic neurons. These abnormalities appear to involve specifically the non-pyramidal interneurons in layer II of the pgACC. Given evidence from morphometric MRI studies of a BD-associated decrease in pgACC volume in unmedicated patients, the reduction in neuron density and/or the changes in gene expression may reflect a reduction in the absolute number of specific subtypes of GABA-ergic interneurons in BD. Nevertheless, more stereological studies are needed before it can be concluded that neuronal loss occurs in the pgACC in BD. 3.3. The supragenual ACC Decreased volume of the left supragenual ACC (Fig. 1) has been reported in one voxel-based morphometry (VBM) MRI study of BD (Yatham et al., 2007). It is unclear if this putative decrement in

136

J.B. Savitz et al. / Neuroscience and Biobehavioral Reviews 42 (2014) 132–147

volume results from a reduction in neuropil (which occupies most of the GM volume), and whether this decrement is accompanied by loss of oligo-/astro-glial or neuronal cells. In the same series of brains which demonstrated glial cell reduction in the sgACC (Ongur et al., 1998), an absence of significant neuronal cell size, glial cell density or neuronal density changes was found in supragenual BA 24b in BD (Cotter et al., 2001). Glial cell density and neuronal size was however, decreased in layer VI of the supragenual ACC in the MDD sample compared with the control sample. Neurons were identified using cresyl-violet and glial cells were identified on the basis of morphological characteristics (Cotter et al., 2001). The reason for the difference in findings in the BD group between the Cotter et al. (2001) and the Ongur et al. (1998) studies is not clear as both studies used the same set of brains and both studies reported a reduction in glial cell density in the subgenual ACC in MDD (see above). Besides methodological differences, one possibility is that the reduction in glial cells in BD is limited to the sgACC. Another possibility is that Ongur et al. (1998) focused on familial BD patients while Cotter et al. (2001) did not subdivide groups according to family-history. Statistical power also may have contributed to the difference in results as the BD group in the Cotter et al. (2001) study did in fact show a reduction in glial cell density in the left (but not the right) layer VI of the supragenual ACC, whereas the MDD group showed a bilateral reduction in glial cell density in layer VI. Webster et al. (2005) found decreased levels of GFAP, a cytoskeletal marker of astroglia, in the WM of the supracallosal gyrus (BA 24) of subjects with BD. A trend in the same direction was found in the GM, and was most pronounced in layer VI. Nevertheless, the variability in GFAP levels due to differences in glial activation and fixation quality make the significance of these data uncertain. Regarding neuronal abnormalities, a reduction in the number of third-order (but not fourth or fifth-order) branches of pyramidal cell dendrites was found in layer VI of the supragenual ACC in a mixed group of suicide cases with a history of MDD or BD (Hercher et al., 2010). The effect was more pronounced in the cases with BD (n = 5) than with MDD (n = 7). Age, postmortem interval, storage time and pH were not associated with the number of dendritic branches. Connor et al. (2009) reported an increase in neurons (labeled with neuron-specific nuclear protein) situated within the WM ventral to the dACC in ∼25% of their postmortem BD and schizophrenia samples, raising the possibility that aberrant neuronal migration may occur during neural development in BD. No changes in the density of GM neurons were observed. This sample included only four cases who were not receiving antipsychotic medication before death, and three of these four subjects manifested neurons in the deeper white matter, suggesting that the results did not reflect a confounding effect of medication. In addition, no association between the density of neurons within the WM and variables such as age and postmortem interval was evident. Additional evidence for neuronal abnormalities in the supragenual ACC is based largely on immunocytochemical and/or gene expression studies. Cotter et al. (2002a) showed using immunohistochemistry that the density of calbindin-expressing GABAergic neurons in layer II of the supracallosal gyrus was reduced by ∼30% in BD and schizophrenia. There was no group difference in the density of calretinin, and parvalbumin-expressing neurons, although there was a non-significant (20%) reduction in the density of parvalbumin-expressing neurons. Parvalbumin is expressed predominantly in the basket and chandelier subpopulations of GABAergic neurons, calretinin is expressed by bipolar cells and double bouquet neurons, and calbindin is expressed predominantly by double bouquet neurons. Eastwood and Harrison (2001) reported a decrease in the expression of 3 synaptic proteins, namely synaptophysin, growth-associated protein-43 (GAP-43), and complexin 2 in subjects with BD but not schizophrenia. The decrease in these

proteins was correlated with length of illness and was accentuated in subjects with a family history of mood disorders. Complexin 1 is found predominantly in parvalbumin-containing GABAergic interneurons while complexin II is found in predominantly pyramidal neurons, suggesting that the change in protein levels is indicative of reduced synaptic density, particularly in the density of pyramidal cell-based excitatory synapses (Eastwood and Harrison, 2001). Freezer storage time, which was longer in the psychiatric groups was negatively correlated with GAP-43 and synaptophysin levels conceivably confounding the data, although if the data were confounded by storage time one may have expected the schizophrenia group to also differ from the control group, and this was not observed. Summary: There is some evidence for abnormalities in double bouquet and basket and chandelier GABAergic neuron gene expression in the supragenual ACC. The extent and nature of glial cell pathology in this region remains unclear. 3.4. The orbitofrontal cortex (OFC) Pathological reductions in the GM volume of the OFC have been reported in adult (Frangou, 2005; Haznedar et al., 2005; Lyoo et al., 2006; Nugent et al., 2006; Narita et al., 2010) and pediatric BD samples (Wilke et al., 2004; James et al., 2011), with the OFC region implicated most consistently in these studies being the intrasulcal portion of the lateral orbital cortex (BA 47s), an area associated specifically with the visceromotor, or medial prefrontal cortical network (Ongur et al., 2003). Consistent with these data, cortical thinning of the OFC recently was described in a sample of BD patients with a history of psychosis (Foland-Ross et al., 2011). Potentially congruent with this finding, Cecil et al. (2002) recorded a reduction of N-acetyl-aspartate (NAA) and choline concentrations in the OFC of a BD sample hospitalized for manic or mixed episodes. The reduction in NAA may, however, have reflected a confounding treatment effect, since the authors also observed a significant negative correlation with the duration of divalproex sodium treatment. The NAA concentration putatively reflects neuronal function, and is decreased in neuropathological conditions such as stroke and Alzheimer’s disease. The reduction in choline levels is hypothesized to reflect altered membrane phospholipid metabolism, as choline concentrations correlate directly with the loss of cell membrane structures and possibly decreased cell density (Cecil et al., 2002). Regarding DTI studies, Beyer et al. (2005) reported increased ADC values in the WM of the OFC, bilaterally while Frazier et al. (2007) detected reduced microstructural integrity in the WM of the left OFC in a pediatric sample. In contrast, greater WM FA in the region of the left orbitomedial PFC together with reduced WM FA in the right orbitomedial PFC was reported in a sample of BD subjects using tract-based spatial statistical methods which allowed for the differentiation of longitudinal and radial diffusivities (Versace et al., 2008). A partial replication of this finding was obtained in adolescents with BD who showed evidence of reduced WM integrity of the right OFC (Kafantaris et al., 2009). Suicidal behavior may be a confounding variable. Mahon et al. (2012) reported that BD patients with a prior suicide attempt had lower FA values in the left OFC WM than patients without a suicide attempt, and further that FA values correlated inversely with impulsivity. Nevertheless, neither BD group differed significantly from healthy controls in WM integrity of the OFC. The neuropathological correlates of these imaging findings remain unclear as we are aware of only two relevant postmortem studies conducted in BD samples. Cotter et al. (2005) reported decreased neuronal size (21%) in layer I of the caudal OFC and a trend toward decreased neuronal size in layer V of the caudal OFC in patients with BD. The significance of this report remains unclear,

J.B. Savitz et al. / Neuroscience and Biobehavioral Reviews 42 (2014) 132–147

however, because of the relatively low number of neuronal cell bodies in layer I compared to other layers of the cortex. No change in neuronal density or glial cell size or density was detected in this region. In contrast, a greater density of OFC GAD-immunoreactive neurons was reported in both a BD (n = 11) and an MDD (n = 9) sample compared with controls (n = 19) (Bielau et al., 2007). Neuronal density in the right OFC was positively correlated with illness duration. Although the mean dose of psychotropic medication did not correlate significantly with neuronal density in the OFC, positive correlations between medication dose and neuronal density were observed in the hippocampus (Bielau et al., 2007), raising the possibility that the results were confounded by treatment. Summary: Neuroimaging studies have reported reductions in GM volume and changes to the WM tracts in the portion of the lateral OFC functionally linked to the medial prefrontal network, but the neuropathological basis of these abnormalities remain unclear. 3.5. BA 9 of the dorsolateral prefrontal cortex (DLPFC) The neurons in the BA 9 cortex share substantial reciprocal anatomical projections with the mPFC regions that form the visceromotor network, and also share monosynaptic projections with limbic structures such as the amygdala and the periaqueductal gray (PAG) – although the connections between BA 9 and the amygdala are relatively sparse (Price and Drevets, 2010). Brain imaging studies generally have focused on the DLPFC as a whole rather than on BA 9 in isolation. Two studies reported GM volume reductions of a large area of the DLPFC (BA 8, 9, 45, 46) in medicated and remitted BD I patients, and partially medicated, “stable” bipolar-spectrum individuals, respectively (Frangou, 2005; Haznedar et al., 2005). Nevertheless, more circumscribed volume reductions of BA 9 have been reported in a medicated, euthymic pediatric sample (Dickstein et al., 2005). The effect of mood state on these neuromorphometric changes remains unclear. In contrast to Frangou (2005) and Haznedar et al. (2005) who reported volume reductions of the DLPFC in euthymic BD patients, a recent VBM analysis showed that depressed but not euthymic subjects with BD, who were unmedicated for at least 2 weeks prior to scanning, had reduced GM in the dorsomedial (BA 9/10, bilaterally) and right DLPFC (BA 9/46) versus controls (Brooks et al., 2009). NAA has been reported to be decreased in magnetic resonance spectroscopy (MRS) studies of the DLPFC in patients with BD. Molina et al. (2007) found that males with BD had reduced NAA concentration in the right DLPFC (BA 8, 9, 10 and 46) while Kalayci et al. (2012) reported NAA reductions in the DLPFC in subjects with BD, schizoaffective disorder, and schizophrenia. Similar findings have been reported in the pediatric BD literature, with both left and right hemispheres implicated (Chang et al., 2003; Sassi et al., 2005; Olvera et al., 2007). Postmortem studies generally indicate that both the neuronal and glial cell density are decreased in BD samples relative to control samples. Regarding neuronal abnormalities, a 3-D study detected a decrease of 20–30% in the density of neurons in layer III of BA 9 in BD subjects (N = 10) compared with controls (N = 11) matched for age, sex, ethnicity, and postmortem interval (Rajkowska et al., 2001). Additional analyses showed a decrease in pyramidal cell density in layers III and V, but no significant difference between BD subjects and controls in the density of non-pyramidal neurons. A trend toward a reduction in neuron size also was reported by the authors (Rajkowska et al., 2001). The tissue samples were obtained from the magnocellular portion of the BA 9 cortex and thus exclude the medial or parvocellular portions of area 9 (Rajkowska et al., 2001). Medication is a potential confound as 9 out of the 10 patients who participated in this study were receiving lithium, and the duration of lithium use was inversely associated with pyramidal cell density (Rajkowska et al., 2001). Thus it remains

137

conceivable that the reduction in neuronal density may have arisen secondary to a lithium-related increase in volume. In a subsequent immunocytochemistry-based study, no difference in the size, shape or density of three subpopulations of pyramidal neurons was detected in BD subjects relative to controls although there was a reduction in neuronal size in all three pyramidal cell populations in layer V of the BD subjects (Law and Harrison, 2003). There was no significant correlation between the neuronal size and the lifetime exposure to psychiatric medications or age of illnessonset. Postmortem interval and pH had no effects on neuronal density or size (Law and Harrison, 2003). Similarly, Cotter et al. (2002b) reported a reduction in neuronal size, but not neuronal density in layers III (10%), V (14%) and VI (18%) in the BD samples from a 2-D analysis of the Stanley Foundation Neuropathology Consortium dataset. Demographic and histological variables were not significantly correlated with neuronal size or density. On the other hand, another study reported that the density of GABAergic (GAD67 mRNA-containing) neurons in layers II through V of BA 9 was decreased by approximately 25–33% in BD (Woo et al., 2008a). The density of the double-labeled GAD67 /NR2A neurons was decreased in the schizophrenia sample but unchanged in the BD sample compared with controls, raising the possibility that glutamatergic innervation of GABA interneurons via NMDA channels may be abnormal in schizophrenia but not in BD (Woo et al., 2008a). There was no difference between the groups in cortical thickness, postmortem interval, freezer storage time, pH, and age. In a 2-D study, Beasley et al. (2002) quantified the density of GABAergic neurons immunoreactive for the calcium-binding proteins parvalbumin, calbindin, and calretinin in samples from the Stanley Foundation Neuropathology Consortium. A trend toward a significant reduction in the density of parvalbumin and calbindinimmunoreactive neurons in BD was found in all layers. However, with the exception of the schizophrenia group, parvalbumin (large basket and chandelier cells) and calbindin (double bouquet and/or neurogliaform) neuron density did not differ significantly from that of controls (Beasley et al., 2002). Tissue pH and age, which differed between the groups, correlated with calbindin and parvalbumin neuron density but was controlled for statistically. Three BD subjects who were neuroleptic-naive had relative densities similar to those patients who were medicated and there was no correlation between lifetime neuroleptic load and the density of neurons (Beasley et al., 2002). A recent immunohistochemistry study using a monoclonal antibody to both GAD65 and GAD67 reported a decrease in GADimmunoreactive neuropil in layers III and V of the left DLPFC in BD (n = 12) but not MDD (n = 9) subjects relative to controls (n = 18) (Gos et al., 2012). Duration of illness was greater in the BD group compared with the MDD group but did not correlate with GAD immunoreactivity in the DLPFC. Postmortem interval, time of fixation, shrinkage, sex, age, and brain weight did not differ among the groups. Negative correlations between neuroleptic and/or lithium dose were reported in the medial temporal cortex and the OFC but not the DLPFC (Gos et al., 2012). These data were supported by a meta-analysis of Stanley Foundation Neuropathology Consortium samples (schizophrenia, BD, MDD, and controls, 15 per group) matched for age, gender, race, postmortem interval, mRNA quality, brain pH and hemisphere (Kim and Webster, 2010). The size of pyramidal neurons in layer III was significantly reduced in the BD group compared with the controls along with a decrease in the density of calbindin-expressing neurons in layer VI (Kim and Webster, 2010). On the other hand, Bitanihirwe et al. (2010) found no difference in the density or laminar distribution of parvalbumin (PV)-expressing neurons in BA 9, and no changes in the mRNA expression of the NR2A NMDA receptor subunit in these neurons. A 2-D counting study reported an increase in the density of large calretinin-expressing neurons in

138

J.B. Savitz et al. / Neuroscience and Biobehavioral Reviews 42 (2014) 132–147

layer II of BA 9 in 5 BD subjects compared with 5 control subjects (Sakai et al., 2008). However, the significance of this result is unclear given the small sample size and the fact that the increase in the density calretinin-expressing neurons was confined to large neurons. Regarding neuronal gene expression changes, a quantitative polymerase chain reaction (qPCR) analysis of pyramidal cells dissected from the ventral and dorsal banks of the principle sulcus of the DLPFC yielded evidence of increased expression of the AMPA receptor subunit, GRIA1 in layer V of samples from Stanley Foundation Neuropathology Consortium postmortem samples with BD and schizophrenia (O’Connor and Hemby, 2007). GRIA1 mRNA expression was significantly increased in pyramidal cells in both layers II/III as well as layer V in the schizophrenia group. In contrast, there were no group differences in the AMPA receptor subunits, GRIA2-4, in either layer nor were any group differences in the expression of the NMDA receptor subunits (NR1, NR2A2B) observed. Because layer V pyramidal cells project primarily to the striatum, the authors postulate that the increased levels of AMPA receptors result in hyperactivation of the striatum and therefore enhanced subcortical DA responsivity to stress (O’Connor and Hemby, 2007). Beneyto and Meador-Woodruff (2006) analyzed AMPA receptor expression in the DLPFC in schizophrenia, MDD, BD, and controls using in situ hybridization. The authors reported decreased expression of the GRIA2 and GRIA4 AMPA receptor subunits in layers II, V and VI in patients with BD. GRIA3 also was decreased in MDD, and GRIA4 decreased in schizophrenia. Because autoradiography analyses suggested that AMPA subunit transcripts were decreased without a concomitant change in AMPA receptor binding, the authors hypothesized the existence of an abnormality in intracellular trafficking mechanisms of the AMPA receptor (Beneyto and Meador-Woodruff, 2006). Costa and colleagues provided evidence for abnormalities of GABAergic neuron gene expression in BD. Reelin (RELN) is secreted by GABAergic interneurons in layers I and II and binds to integrin receptors located on dendritic spines of pyramidal neurons or integrin receptors located on GABAergic interneurons of layers III through V (Guidotti et al., 2000). Both a decrease in reelin and GAD67 expression in BA 9 and a decrease in RELN-positive cells were found in patients with schizophrenia or BD with psychosis, but not in subjects with MDD (Guidotti et al., 2000). Because other proteins expressed by GABAergic neurons such as GAD65 and DAB1 were not significantly decreased in this BD sample, Guidotti et al. (2000) hypothesized that the decrease in reelin and GAD67 expression was a consequence of changes in gene expression rather than neuronal loss, per se. A follow-up study showed that the decrease in GAD67 mRNA-expressing GABAergic neurons in BA 9 in subjects with a history of psychosis, may be due to hypermethylation of the GAD1 gene promoter by DNA methyltransferase 1 (Veldic et al., 2005). Regarding glial cell abnormalities, a 3-D counting study found a non-significant (6%) reduction in overall nissl-stained glial cell density in BD (Rajkowska et al., 2001). There was however, a significant reduction in the density (19%) of glial cells in layer IIIc in conjunction with glial cell enlargement in layer IIIc, i.e. fewer but larger glial cells were present. The reduction in glial cell density was not dependent on age, postmortem delay, or storage time in formalin (Rajkowska et al., 2001). Similarly, Cotter et al. (2002b) reported no significant difference in glial cell density in BD samples obtained from the Stanley Foundation Neuropathology Consortium although there was a 25% reduction in glial cell density that trended toward significance in layer VI, and furthermore, a significant reduction in glial cell density in the MDD and schizophrenia groups was reported in layer V. Using a potentially more sensitive 3-D methodology (optical dissector), Uranova et al. (2004) did in fact find significantly reduced numbers of nissl-stained glial cells

– specifically morphologically-identified oligodendroglia – in layer VI of BA 9 from the same Stanley Foundation Neuropathology Consortium samples. Group differences in age, postmortem interval and time in formalin were controlled for statistically. There was no change in glial cell density in the WM and cortical thickness also did not differ between healthy and disease groups. In a parallel study the authors showed that the decrease in glial cells extended to layer III, potentially accounting for reports of pyramidal cell dysfunction (Vostrikov et al., 2007). The Stanley Foundation Neuropathology Consortium meta-analysis discussed above, showed a significant reduction in the number of oligodendrocytes per neuron in layer III as well as a significant reduction in the overall density of oligodendrocytes in layer VI (Kim and Webster, 2010). Reduced myelination of the deep (but not gyral) WM in a region encompassing BA 9 and BA 46 was found in postmortem BD, MDD and schizophrenia samples (Regenold et al., 2007). The authors conducted a myelin staining study and quantified the results as the percentage of GM staining for each region. The results did not appear to be confounded by variables such as gender, age, smoking, postmortem interval and antipsychotic medication. Nevertheless, the results may have been driven in part by a healthy control that appeared to be an outlier in terms of WM staining intensity. A significant number of studies also reported abnormal increases or decreases in the mRNA levels or protein products expressed by glial cells. Whether these putative changes are related to the reductions in cell density reported above or whether they result from specific changes in gene expression remains unclear. Certainly, the postmortem evidence for glial cell abnormalities is at least consistent with a quantitative PCR analysis of BA 9 tissue which demonstrated significant reduction in mRNA expression of key protein markers of myelination and oligodendrocyte function (Tkachev et al., 2003). Expression of proteolipid protein 1 (PLP1), myelin associated glycoprotein (MAG), oligodendrocyte specific protein (CLDN11), myelin oligodendrocyte glycoprotein (MOG), and transferrin (TF) was reduced by approximately 2- to 4-fold in BD patients relative to psychiatrically healthy controls (Tkachev et al., 2003). Further, expression of the OLIG2 and SOX10 genes, which code for transcription factors involved in oligodendrocyte differentiation and maturation, was downregulated by 2to 3-fold in BD. Although this decrease in oligodendrocyte-related gene expression conceivably could result from cell loss, not all of the oligodendrocyte-related genes were found by Tkachev et al. (2003) to be downregulated, leading these authors to postulate that their results are more likely to reflect cellular dysfunction than cell death. Another study reported that the level of the creatine kinase B isoform, which is involved in the synthesis of phosphocreatine and preferentially is expressed in oligodendrocytes and astroglia, was reported to reduced in BD subjects relative to controls (MacDonald et al., 2006). Since creatine kinase generates ATP and creatine from high energy phosphates such as phosphocreatine, these data provide evidence for mitochondrial dysfunction and thus may reflect the aforementioned reduction in oligodendroglial function (MacDonald et al., 2006). Notably, Pennington et al. (2008) conducted a proteomic analysis of BA 9 tissue from postmortem samples and found 51 proteins that were abnormally expressed in BD. Of these 51 proteins, 25 were involved in energy metabolism or mitochondrial function, and 15 were cytoskeletal or synapse-associated. Furthermore, MacDonald et al. (2006) reported that other mRNA transcripts of oligodendrocyte-specific proteins such as gelsolin, MAG, and ERBB3 were downregulated in BD. The authors postulated that these changes were not secondary to treatment with the mood stabilizer, lithium, since with the exception of ERBB3, feeding rats for 14 days with lithium chow did not affect the gene expression of the glial cell markers (MacDonald et al., 2006). Consistent with the postmortem data, a decrease in the cerebral creatine and phosphocreatine concentrations also was reported in

J.B. Savitz et al. / Neuroscience and Biobehavioral Reviews 42 (2014) 132–147

an MRS study of a medication-free sample of BD patients using a voxel that encompassed parts of the BA 9 cortex in the left DLPFC (Frey et al., 2007). Summary: A number of morphometric MRI studies have reported reductions of the DLPFC volume although because the resolution of morphometric MRI makes it difficult to distinguish between different cytoarchitectonic regions of the DLPFC, it remains unclear whether these abnormalities are specific to BA 9. There is evidence for reductions in the neuronal (GABAergic and glutamatergic) density and size, as well as the glial cell density in this region. These data appear potentially congruent with reported functional abnormalities of oligodendrocytes and excitatory and inhibitory neurons. The selective nature of the functional abnormalities suggests that the changes in mRNA and protein expression might not be directly related to cell loss, but rather that the reported decreases in density may be secondary to a loss of neuropil or a down-regulation of gene expression.

139

neuronal density and cortical thickness in the sgACC (Bouras et al., 2001), and BA 9 (Rajkowska et al., 2001), suggesting neuronal loss along with a reduction in neuropil. In contrast, Ongur et al. (1998) found an increase in neuronal density without neuronal loss that may be explained by a decrease in cortical thickness attributable to a loss of neuropil in the sgACC in BD subjects relative to controls sample. This finding is consistent with preclinical studies demonstrating the effects of psychological stress on neuronal morphology of the mPFC. Repeated psychological stress putatively causes the loss of dendritic spines as well as the retraction and debranching of the apical dendrites of pyramidal neurons (Cook and Wellman, 2004; Radley et al., 2006). This dendritic atrophy likely has significant functional consequences (Radley et al., 2008) which could also explain the reported reductions in gene expression observed in BD samples. 4.2. Limitations of the postmortem literature

4. Discussion 4.1. Abnormalities associated with BD postmortem Although the abnormalities observed at postmortem vary somewhat across the subregions and cell layers of the mPFC, three common themes emerge from this literature. Firstly, a decrease in volume is found in many morphometric MRI investigations. Secondly, reductions in glial cell density and/or abnormalities of gene expression are relatively widely reported, and WM abnormalities measured using DTI putatively provide support for these findings. Thirdly, there are a number of reports of reductions in neuronal size, neuronal density and/or changes in neuronal gene expression. With the exception of BA 9 where changes in both GABAergic and pyramidal cell density were apparent, most studies reported changes selectively in GABAergic cell density or expression, and particularly in cells staining positively for parvalbumin and/or calbindin, which constitute the basket and chandelier neurons and the double bouquet neurons, respectively. The ontogeny of these changes in GABAergic cells remain unclear. Glutamate functions as a major regulator of inhibitory tone by tonically activating NMDA receptors on GABAergic interneurons (Olney et al., 1999). Thus NMDA receptor hypofunction and/or neurotoxicity of GABAergic cell populations may lead to excitotoxicity-induced dendritic atrophy and/or oligodendrocyte cell loss in the cingulate cortex and other anatomically-related regions. Neurons expressing the NMDA receptor during the synaptogenesis stage of development are highly sensitive to glutamate stimulation such that too much stimulation or too little stimulation may result in excitotoxic neurodegeneration or apoptotic neurodegeneration, respectively (Olney et al., 1999). However, the sequelae of developmental neurotoxicity of the GABAergic interneurons is postulated to manifest itself only in late adolescence when the development of synaptic connections is complete (Farber et al., 1998; Olney et al., 1999). It is unclear how to interpret the reductions in glial cell and neuronal cell density since the weight of the data was derived from studies applying the 2-D counting method which is susceptible to bias (see below). Conceivably reductions in cell density may result from the loss of cells. We are aware of two 3-D studies that also have reported an absolute reduction in glial cell numbers in the sgACC (Ongur et al., 1998), and BA 9 (Uranova et al., 2004). Another 3-D study reported a decrease in glial cell density together with glial cell hypertrophy in BA 9, suggesting the presence of fewer but larger glial cells in BD (Rajkowska et al., 2001). Regarding neuronal abnormalities we are not aware of any 3-D studies that have compared the absolute neuron numbers in the mPFC between BD and control samples. However, two stereological studies reported a decrease in

The technical limitations of various postmortem techniques have been reviewed elsewhere (Dorph-Petersen and Lewis, 2011; Gigante et al., 2011) and are discussed briefly here. The major schism in the literature revolves around the use of 2-D versus 3D (stereological) counting methods. As Dorph-Petersen and Lewis (2011) point out, caution must be exercised in the interpretation of studies of cellular density derived from 2-D studies. Because the precise, absolute volume of the brain region under investigation is unknown in postmortem studies (due to shrinkage during fixation) the true in vivo concentration of cells cannot be accurately derived from the density data. For example, the increase in glial cell and/or neuronal density reported in some studies above, could be attributable either to increased cell numbers or to decreased tissue volume accompanied by no change (or conceivably even a reduction) in cell number (Dorph-Petersen and Lewis, 2011). Conversely, a decrease in cell density could be attributable to cell loss or to increased tissue volume. Regarding transcriptomic studies, the integrity of mRNA is known to be sensitive to pH (Kingsbury et al., 1995) which may differ between diagnostic groups. Transcriptomic studies also cannot always differentiate between changes in gene expression (functional changes) or morphological abnormalities; i.e. a reduction in the number cells immunoreactive to a protein such as GAD65 might result from reduced mRNA expression or protein synthesis of GAD65 rather than a decrease in the number of GABAergic cells. Histological variables such as longer fixation time, which can lead to greater tissue shrinkage and increased cellular density, and increased pH and postmortem interval which can artificially increase the size of neurons, may bias the results of some studies (Gigante et al., 2011). As described in Section 3, these confounds are usually controlled for statistically, although the statistical analyses in question are usually predicated on the assumption that the relationship between the above variables and diagnosisrelated histological change is linear. Nevertheless, demographic and histological variables appear unlikely to solely account for the BD-associated neuropathology described in this review. The impact of psychiatric medications on postmortem tissue appears complex, and generally has not been assessed systematically. We and others have shown that lithium treatment is associated with an increase in GM volume in the context of morphometric MRI-based assessments (Moore et al., 2000; Savitz et al., 2010, 2011). Theoretically, this increase in GM volume could be secondary to an increase in extravascular space thus decreasing cell density, however, preclinical studies favor the hypothesis that lithium impacts GM volume via neuroplasticity effects that result in an increase in neuropil in regions affected by pathological, atrophic processes (Manji et al., 2000). Whether other classes of medications exert neuromorphometric or histological effects remains unclear.

140

J.B. Savitz et al. / Neuroscience and Biobehavioral Reviews 42 (2014) 132–147

For example, primate studies have reported discrepant effects of antipsychotic treatment on the density of glial cells (Selemon et al., 1999; Konopaske et al., 2008). Nevertheless, as is the case with the histological variables discussed above, researchers usually attempt to statistically control for medication effects. Further, in some cases the same pathological changes are reported across MDD, BD and schizophrenia groups which tend to be treated with different classes of medication. As discussed by Gigante et al. (2011) and Harrison (2002), the effects of non-prescription drugs and alcohol may be an equally serious source of bias because of the elevated prevalence of substance abuse in BD. Alcohol is neurotoxic and alcoholism has been associated with neuronal and glial cell abnormalities, notably glial cell loss (Korbo, 1999). Finally, BD is a clinically and pathophysiologically heterogeneous syndrome which in combination with the small sample sizes typical of postmortem studies decreases the probability of detecting subtype-specific pathology. The cytoarchitecturally heterogeneous nature of the mPFC also poses challenges. Even within a particular layer, neurons are heterogeneous in terms of their connectivity, morphology and functional properties (Law and Harrison, 2003). 4.3. Should the abnormalities observed postmortem in BD be described as neuropathology? The question of whether BD is a genuine neuropathological condition also remains at least partly unanswered. The extant data indicate that BD is associated with morphometric abnormalities, including reductions in cortical thickness and changes in glial and neuronal cell density that likely results from a combination of dendritic atrophy and/or cell loss. In this sense the answer to whether such findings reflect histopathology is affirmative. However, BD is not a classic neurodegenerative disorder. For example, people with BD do not exhibit the classic morphometric signatures of Huntington’s disease (gliosis), Alzheimer’s disease (␤-amyloid plaques and neurofibrillary tangles) or Parkinson’s disease (Lewy body inclusions). The possibility of cell loss and/or dendritic atrophy in BD should also be viewed in the context of healthy aging which is putatively characterized by a progressive loss of GM volume that may be secondary to neuronal shrinkage and dendritic atrophy (Fjell and Walhovd, 2010). The key question is not whether BD is characterized by morphometric abnormalities but rather whether these abnormalities are progressive, fixed, or reversible. This question is embedded in the larger debate about whether the pathogenesis of BD involves a neurodevelopmental or a neurodegenerative process. 4.4. Evidence for a neurodevelopmental etiology As Harrison points out, the paucity of postmortem evidence for the proliferation and hypertrophy of glia (i.e. gliosis), the hallmark of neuronal degeneration (Oppenheimer, 1984), argues against the existence of a neurodegenerative or inflammatory process in BD (Harrison, 2002); although see Torres-Platas et al. (2011) who reported astrocytic hypertrophy suggestive of mild astrogliosis in the WM adjacent to the dorsal part of the genu of the corpus callosum (BA24) in depressed suicides, and Dean et al. (2013) who reported increased levels of the transmembrane form of tumor necrosis factor alpha (TNF) in BA 24 of BD subjects at postmortem. In contrast, RELN, which plays a key role in neuronal migration during development is putatively under-expressed in BA 9 inter-neurons in BD patients with psychosis (Guidotti et al., 2000). This finding appears consistent with the above-mentioned evidence that neuronal counts within the white matter are elevated in the supragenual ACC in BD subjects versus controls (Connor et al., 2009). Affective disorders may have neurological antecedents. Although the evidence for clearly-defined development problems

in BD is modest, some individuals who go on to develop BD show premorbid abnormalities in motor, language and cognitive function (Savitz et al., 2005a). Certainly impairments in a wide spectrum of cognitive domains are already present at first-episode (Lewandowski et al., 2011; Zabala et al., 2010; Zanelli et al., 2010) as well as in children with BD (Bearden et al., 2007a; Pavuluri et al., 2009). Further, a number of reviews of the extant literature indicate the existence of neurocognitive abnormalities in a subset of healthy individuals at-risk for BD (Olvet et al., 2013; Savitz et al., 2005b). There is also modest evidence for an association between perinatal complications and BD. One study reported that premature infants were 2.7-fold more likely to develop BD (Nosarti et al., 2012). Similarly, a linear relationship between birth weight and the risk for BD as well as other psychiatric disorders was found (Abel et al., 2010), and children delivered by cesarean section had a 2.5-fold greater risk for BD in a large population-based register (Chudal et al., 2013). A recent study that reported an association between perinatal hypoxia and reduced left amygdala and right hippocampal volumes in psychotic and non-psychotic BD subjects, respectively, raises the possibility that perinatal complications increase the risk for developing BD by impacting key circuits involved in the regulation of emotion (Haukvik et al., 2013). Regarding cross-sectional morphometric MRI and DTI studies, volume reductions of the perigenual ACC have been reported in first-episode adult patients with BD. (Hirayasu et al., 1999) found GM deficits of the left sgACC in familial BD patients during their first-episode of mania. Similarly, GM volume reductions in the left supragenual ACC were observed in mood stabilizer and neuroleptic-naïve subjects with a first episode of mania (Yatham et al., 2007). A DTI study of patients who had remitted from a first episode of mania reported an increase in radial diffusivity of several WM tracts including the left anterior frontal cortex and the left cingulum (Chan et al., 2010). Similarly, untreated BD patients with a first psychotic episode had increased radial diffusivity compared with healthy controls in several WM tracts including the cingulum, internal capsule, and longitudinal fasciculus, suggesting a neurodevelopmental disorganization of WM tracts and/or aberrant myelination of fibers (Lu et al., 2011). Potentially congruent with the DTI data, Strakowski et al. (1993) reported an increased frequency of white matter hyperintensities (WMH) in first-episode adult patients with BD. Volume reductions of the mPFC and WM abnormalities have also been found in pediatric subjects with BD. Specifically, a GM volume deficit in the left perigenual ACC has been reported in both medicated and unmedicated children and/or adolescents with BD (Chiu et al., 2008; Kaur et al., 2005; Singh et al., 2012; Wilke et al., 2004). The fact that in some cases these morphometric abnormalities were observed in pediatric cases with a single recent episode of mania raises the possibility that the reduction in GM volume of the perigenual ACC predates illness onset or occurs very early in the course of the disorder. With respect to WM abnormalities, Adler et al. (2006) found lower FA in the superior frontal WM tracts in mediation-naïve adolescents with BD compared with controls. In a similar vein, WMH, which are usually associated with vascular disease, are also putatively over-represented in pediatric BD samples (Botteron et al., 1995; Pillai et al., 2002). Morphometric changes in the ACC also appear to be present in people at risk for BD. Volume reduction of the right perigenual ACC region including both the sgACC and the pgACC has been shown to be associated with increasing genetic risk for BD (based on the number of BD relatives) (McDonald et al., 2004). Similarly, this region has been reported to be reduced in volume in a sample of boys with subclinical depression (Boes et al., 2008). Evidence for a genetic contribution to BD-associated morphometric abnormalities is emerging in the literature. The met allele of

J.B. Savitz et al. / Neuroscience and Biobehavioral Reviews 42 (2014) 132–147

the val66met SNP in the brain-derived neurotrophic factor (BDNF) gene, which may influence susceptibility to BD (Sklar et al., 2002), has been associated with a reduction in gray matter volume in the dorsal ACC in healthy controls and patients with BD (Matsuo et al., 2009b). In addition, Ho et al. (2007) found that met allele carriers with schizophrenia showed greater gross frontal lobe volume reductions over time than val/val homozygotes. More recently, a putative risk allele for BD in the ankyrin G (ANK3) gene was not only found to be correlated with cognitive deficits in first-episode psychosis, but was associated with widespread reductions in cortical thickness including the medial frontal gyri and orbital gyri (Cassidy et al., 2013). 4.5. Evidence for a neurodegenerative etiology Chronic psychological stress leading to elevations in circulating cortisol and glutamate-induced excitotoxicity may be one cause of progressive neuronal degeneration in BD. As discussed above, chronic or repeated exposure to various types of stressors have been shown to cause reductions in dendritic spine density and alterations in gene expression in the mPFC of rodents (Covington et al., 2010; Liston et al., 2006). Nevertheless, the loss of spines and synapses is not accompanied by a decrease in cell numbers or volume (Cerqueira et al., 2007), and furthermore, this stress-associated dendritic remodeling has been shown to be reversible (Radley et al., 2005), arguing against progressive neurodegeneration. Glial cells may also be adversely effected by stress. Rats exposed to chronic social defeat display reduced gliogenesis and reduced oligodendroglial cell counts in the mPFC; an effect that was reversed by treatment with fluoxetine (Czeh et al., 2007). Evaluation of the evidence for progressive neurodegeneration in BD is confounded by treatment with medication which alters the natural course of the disorder. For example, a higher prevalence of Alzheimer’s disease has been reported in BD but chronic lithium treatment reportedly reduces the risk to that of the general population (Nunes et al., 2007; Kessing et al., 2008). In other words, a BD-associated neurodegenerative process may be partially masked by treatment with medication so that the histopathological changes observed at postmortem are more subtle than would have been the case prior to the development of effective treatments for BD. The hypothesis that medication may exert neuroprotective effects receives some support from the morphometric MRI literature. No change in the degree of gyrification of the prefrontal cortex was observed in medicated subjects with BD versus healthy controls at 4-year follow-up – although BD subjects carrying the BDNF val66met met allele showed greater loss in prefrontal gyrification over time than val/val homozygotes (Mirakhur et al., 2009). Medication effects may also explain counter-intuitive reports of progressive increases in GM volume in BD subjects over time. For instance, Bearden et al. (2007b) found an increase in GM volume of the sgACC in BD patients with a mean duration of illness of 15 years. However, these morphometric abnormalities were apparent only in the subset of the BD patients who were treated with lithium. Lisy et al. (2011) reported that adolescents and adults with BD, the majority of whom were receiving treatment with atypical antipsychotics and/or anticonvulsants, displayed an increase in the GM volume of the ventrolateral PFC, caudate, and medial temporal lobe between the baseline assessment and the 3–34-month follow-up assessment. Nevertheless, a number of longitudinal studies have reported decreases in GM volume of the brain, including the perigenual ACC, over the time-period between subjects’ first presentation with BD and later follow-up scans – even in cases where subjects were medicated (Lim et al., 2013; Schneider et al., 2012). For instance, a progressive reduction in GM volume of the pregenual ACC was observed in a small sample of BD adolescents followed for 2.5 years

141

after a first manic episode (Farrow et al., 2005). This result has been partially replicated in other pediatric BD samples which also show progressive reductions in GM volume of the mPFC, including the pregenual ACC and the sgACC (Gogtay et al., 2007b; Kalmar et al., 2009). Similarly, young adults scanned after a first episode of mania showed a progressive loss of GM in the sgACC between the baseline scan and an 18-month follow-up scan (Koo et al., 2008). Congruent with these data, several cross-sectional studies have reported negative correlations between length of illness and GM volume in BD. Lyoo et al. (2006) found reductions in the cortical thickness of multiple cortical regions including the pregenual ACC and the OFC as well as a significant negative correlation between illness duration and cortical thickness of the middle frontal cortex (BA 46) of medicated subjects with BD. In a similar vein, an inverse association between illness duration and total GM but not WM volume was reported in a predominantly medicated BD sample (Frey et al., 2008). A series of comprehensive reviews of the literature indicate that BD is characterized by neurocognitive impairments that extend into periods of remission (Arts et al., 2007; Robinson et al., 2006; Savitz et al., 2005a). Consistent with the morphometric MRI literature, several studies have reported correlations between the clinical course/history of the illness and the degree of neuropsychological impairment of subjects with BD (Kessing, 1998). Specifically, greater chronicity, duration of illness, and a greater number of manic and/or depressive episodes has been associated with deficits in traditional neuropsychological measures of executive function (Cavanagh et al., 2002; Frangou et al., 2005; Lopez-Jaramillo et al., 2010; Martinez-Aran et al., 2004; van Gorp et al., 1998; Zubieta et al., 2001). Congruent with these data a recent 10-year follow-up study of BD outpatients reported a temporal decline in executive function that was likely related to illness duration (Torrent et al., 2012). Moreover, a longitudinal follow-up study of elderly individuals reported more cognitive dysfunction at baseline and a more rapid cognitive decline in the BD subjects than their healthy counterparts (Gildengers et al., 2009). Nevertheless, most longitudinal studies published to date report that the BD-associated cognitive deficits are largely stable or decline modestly over time (Arts et al., 2011; Balanza-Martinez et al., 2005; Chaves et al., 2011; Delaloye et al., 2011; Gildengers et al., 2013). 4.6. Clinical implications Neuropathological changes in the mPFC network potentially undermine cognition, regulation of the immune system, affective and hedonic processing, as well as the balance between sympathetic and parasympathetic activity leading to aberrant cognitive function, emotional behavior, and comorbid somatic illnesses. As discussed above, patients with BD perform worse than healthy controls on a wide-range of neuropsychological tasks, in particular on measures of verbal memory and executive function. These cognitive deficits may be partly related to dysfunction of the mPFC network. Not only is there a rich neuropsychological literature pertaining to the impact of lesions to the OFC and the mPFC, but reductions in the GM volume and the WM volume of the perigenual ACC have been shown to be associated with impaired executive function in patients with BD (Zimmerman et al., 2006). The OFC and the pregenual ACC to which it projects, creates a representation of the affective value of both primary and more abstract rewards and thus damage to this network impairs emotional behavior and affect, potentially leading to the impulsivity and euphoria characteristic of mania (Rolls, 2000; Rolls and Grabenhorst, 2008). The ventromedial PFC modulates the electrophysiological response of the ventral tegmental area (VTA) dopamine neurons suggesting that this cortex also may participate in evaluating the salience of rewards (Drevets et al., 1998). Thus

142

J.B. Savitz et al. / Neuroscience and Biobehavioral Reviews 42 (2014) 132–147

conceivably the morphometric abnormalities of neurons, glia and synapses observed in the mPFC may result in dysregulated mesolimbic dopamine release in BD subjects: i.e. diminished dopamine signaling during depressive episodes and elevated dopamine signaling during manic episodes, resulting in anhedonia and exaggerated hedonic responses, respectively. As reviewed elsewhere (Drevets et al., 2008), the decrements in GM volume of the perigenual ACC appear, at least in familial cases, to be restricted to the left hemisphere. Rodents with lesions of the left infralimbic cortex and perigenual ACC show heightened sympathetic arousal and corticosterone release in response to restraint stress (Sullivan and Gratton, 1999). In contrast, animals with right-sided lesions show the opposite pattern of diminished corticosterone release and stress-associated pathology. Sullivan and Gratton (1999) hypothesized that left-sided lesions of the ventromedial PFC disinhibit the right ventromedial PFC resulting in the physiological analogs of depression, i.e. heightened sympathetic autonomic and HPA axis function. Thus neuropathological changes in the mPFC may result in the reduction in parasympatheticto-sympathetic tone observed in a proportion of mood disorder patients, potentially explaining the elevated risk for ventricular tachycardia and myocardial infarction in depressed patients with cardiovascular disease (Carney et al., 2005). The reduction in parasympathetic and vagal tone observed in depressed and manic subjects (Brunoni et al., 2013; Henry et al., 2010; Kemp et al., 2012; Udupa et al., 2007) may additionally result in abnormal autonomic responses, and by implication abnormal affective responses to both internal or external stimuli. For instance, Lane et al. (2009) reported that the vagally-mediated component of heart rate variability correlated with changes in regional cerebral blood flow (rCBF) in the mPFC during the induction of happiness, sadness, and disgust. More recently our group found that the correlation between rCBF in the medial OFC and the balance of sympathetic to parasympathetic tone was in opposite directions in depressed and healthy subjects during the performance of a mildly stressful motor task (Nugent et al., 2011). The lateral OFC receives afferent projections from the nucleus tractus solitarius, the principal terminus of interoceptive inputs from the vagus nerve, and potentially acts through the perigenual ACC to regulate visceromotor activity (Conway et al., 2012; Thayer et al., 2012). The reciprocal relationship between the vagus nerve and the mPFC may have implications for the regulation of the immune system, which is putatively disturbed in mood disorders (Dantzer et al., 2008; Irwin and Cole, 2011; Miller et al., 2009; Padmos et al., 2008; Savitz et al., 2013a). The vagus nerve modulates immune function via its effect on the hypothalamic–pituitary–adrenal (HPA) axis and peripheral leukocytes (Rosas-Ballina and Tracey, 2009). Specifically, activation of the cholinergic anti-inflammatory pathway by electrical stimulation of the vagus nerve reduces the release of proinflammatory cytokines such as tumor necrosis factor alpha (TNF) and interleukin 6 (IL6) therefore restoring homeostasis and preventing tissue damage after activation of the immune system (Rosas-Ballina and Tracey, 2009). Interestingly a recent study reported that in healthy subjects with high heart rate variability (indicating high tonic vagal nerve activity) exposed to an acute stressor, rCBF in the left pregenual ACC was positively correlated with the peripheral concentrations of natural killer (NK) cells (Ohira et al., 2013). In contrast, there was no correlation between NK cell levels and rCBF in subjects with low vagal tone. Ohira et al. (2013) hypothesize that high tonic vagal activity is a prerequisite for the rapid regulation of immune, autonomic, and endocrine responses to acute stress. The direct clinical implications of the pathophysiological changes in the mPFC network remain to be determined. Nevertheless, if progressive damage to the mPFC (and other brain regions) is indeed occurring over time then this putative pathological process

may be responsible for the progressively deteriorating clinical course, worsening of cognitive function, and comorbidity observed in some BD patients. For instance, treatment resistance and shortening of the inter-episode interval have been reported over time in parallel with an increase in the number of discrete episodes of depression and/or mania (Berk et al., 2011; Kessing et al., 2000) leading to attempts to define several clinical stages of BD based on neuroimaging, cognitive, and molecular biomarkers (Berk et al., 2007; Kapczinski et al., 2009; Lin et al., 2013; McNamara et al., 2010). These data raise the possibility that early diagnosis and treatment of both BD and co-occurring medical illnesses may help to reduce the morbidity associated with BD. 4.7. Future directions Although challenging, further MRI studies of the earliest possible manifestations of BD illness in pediatric or first-episode subjects may lend more clarity to the relative contributions of developmental versus degenerative effects on brain structure. Another informative approach is to evaluate brain structure in the unaffected, first-degree relatives of subjects with BD. Theoretically, a proportion of these genetically at-risk individuals may share neurodevelopmental abnormalities with their affected relatives (Savitz and Drevets, 2009b; Savitz et al., 2013c), and as discussed above, GM abnormalities of the mPFC in at-risk relatives of subjects with BD have been reported in the literature. One challenge with studying at-risk relatives is that by definition these individuals are healthy and thus may be resilient rather than vulnerable to developing mood disorders. A possible solution is to focus on younger high-risk individuals who have not yet passed through the typical age of illness onset. Indeed cortical GM loss is present in the siblings of subjects with childhood-onset schizophrenia, but these deficits normalize by the age of 20 (Gogtay et al., 2007a). Ultimately, however, long-term longitudinal studies of individuals at-risk for BD will be required to rigorously differentiate between developmental and degenerative mechanisms of disease. Given that BD has been hypothesized to be a progressive illness, surprising few neuroimaging studies have addressed the relative contributions of psychotic episodes and/or predominant polarity of mood to this pathological process. One method of addressing these factors is to compare subjects with BD type I, which is characterized by both manic and depressive episodes, and BD type II, which has primarily a depressive polarity with intermittent hypomanic episodes (Sanchez-Moreno et al., 2009). Several studies have reported that subjects with BD type I display greater impairment than their counterparts with BD type II on tests of executive function and verbal memory (Aminoff et al., 2013; Bora et al., 2010; Harkavy-Friedman et al., 2006; Hsiao et al., 2009) although other studies have found no differences or only subtle differences between type I BD and type II BD (Bora et al., 2011). Nevertheless, cognitive differences between BD type I and BD type II cannot necessarily be attributed to the greater neurotoxicity of manic episodes as BD type I and BD type II also differ in respect to the presence of psychosis, treatment regimen, and possibly even developmental etiology. In fact, a history of psychosis may be the primary driver of cognitive impairment in BD (Aminoff et al., 2013; Bora et al., 2007; Glahn et al., 2007; Martinez-Aran et al., 2008; Savitz et al., 2009; Simonsen et al., 2011). Another confound is the potential neuropathological effects of factors such as obesity, sedentary lifestyle, smoking and comorbid illnesses such as cardiovascular disease and type 2 diabetes on the neuromorphometric and histopathological changes that are reported in BD. In other words, are the putative neurodegenerative changes attributable to BD per se, or are they a consequence of commonly comorbid illnesses or lifestyle sequelae? A recent study reported that BD subjects with type 2 diabetes and/or insulin

J.B. Savitz et al. / Neuroscience and Biobehavioral Reviews 42 (2014) 132–147

resistance had lower NAA and creatine levels in the PFC than euglycemic BD subjects and healthy controls (Hajek et al., 2013). Potentially consistent with these data, Bond et al. (2013) found a correlation between elevated body mass index (BMI) and GM/WM reductions in the multiple brain regions including the mPFC, in recovered first-episode mania patients but not healthy controls. Such a relationship conceivably may reflect the interrelationships between obesity, inflammation, and the role of neuroactive cytokines on neuroplasticity. Future studies that stratify according to these variables may provide better insight into the relative pathophysiological contributions of BD versus its co-existing conditions. One factor that has been convincingly demonstrated to confound morphometric (and functional) neuroimaging studies is treatment with medication, in particular lithium (Bearden et al., 2007b; Hafeman et al., 2012; Moore et al., 2000, 2009; Savitz et al., 2010, 2011). While it is challenging to study untreated individuals with BD, this approach is crucially important in order to obtain a more accurate picture of the neuropathological characteristics of BD. Regarding postmortem studies, we are aware of only one study of BD individuals who were either drug-naïve or had received psychotropic medication for less than 6 months prior to death (Bouras et al., 2001). Another gap in the literature is the paucity of studies evaluating microglia and/or infiltrating macrophages in mood disorders. Given emerging evidence for the role of immune dysregulation in affective illness (Dantzer et al., 2008; Miller et al., 2009; Padmos et al., 2008; Savitz et al., 2013a), postmortem studies of microglia would be an important addition to the literature. One immunohistological study has reported evidence for increased levels of microglia in depressed suicide victims (Steiner et al., 2008). Nevertheless, it remains unclear whether increased numbers and/or priming of microglia cells found in such individuals subserve an adaptive or a pathological role in neuroplasticity. Given MRI-derived evidence that GM loss is more pronounced in left perigenual ACC of patients with mood disorders it may be interesting to attempt to replicate these results in the postmortem brain. Fourthly, the perigenual ACC shares extensive anatomical connections with multiple regions such as the hippocampal subiculum, orbitomedial PFC, and hypothalamus implicated in the modulation of emotional behavior. Thus the dendritic atrophy, oligodendrocyte loss, and GM loss observed in the mPFC in some subjects with BD raises the possibility that abnormal synaptic interactions between the mPFC and the hippocampal subiculum, amygdala, and OFC among other regions, may contribute to disturbances in emotional regulation characteristic of BD. While resource intensive, further studies using identical methodologies to examine the relationship between different brain regions in the same cohort of subjects may improve our understanding of the distributed nature of neuropathology in BD. 5. Conclusion BD is characterized by neuropathological abnormalities of glial and neuronal cells which likely affect the excitatory and inhibitory circuits involved in the regulation of emotional processing and associated endocrine and behavioral responses. The magnitude of these abnormalities is modest compared with that of the neuropathological changes observed in the classical neurodegenerative diseases. The weight of the evidence from stereological studies is suggestive of glial cell loss, predominantly oligodendrocytes. It remains unclear whether neuronal cell loss occurs in BD. In contrast, the putative elevations in neuronal density appear secondary to dendritic atrophy (because the vast majority of the GM volume is attributable to the neuropil). Nevertheless, the possibility that a loss of specific populations of GABAergic interneurons occurs in BD is supported by the extant literature. It is likely that functional gene

143

expression abnormalities occur concomitantly with the structural glial and neuronal cell abnormalities. Interpretation of the literature is complicated by the fact that many of the neuropathological changes observed in patients with BD may be confounded by limitations in the different methodological techniques applied in identifying these changes, as well as by treatment-related effects. Moreover such changes may reflect compensatory responses to distal processes which, therefore, may be epiphenomenal to the fundamental pathology of the illness. The causative agents of the histopathological changes observed in BD remain unclear and likely vary across individual patients in a complex interplay of genetic vulnerabilities, psychological stressors, comorbid medical conditions (e.g. metabolic syndrome), substance abuse, and environmental exposures (e.g. microbial infections, perinatal complications). The question of whether the pathogenesis of BD involves a developmental or a degenerative illness remains unanswered, and may ultimately turn out to be an overly simplistic question because it is predicated on the implicit assumption that BD is a single type of disease. Given the modest nature of the abnormalities observed at postmortem, together with the absence of gliosis, and the possibility that comorbid factors such as substance abuse and metabolic syndrome may account for some of the neuropathological changes, we favor the hypothesis that BD is primarily a disorder of neurodevelopmental etiology. Nevertheless, the developmental and degenerative hypotheses are not necessarily mutually exclusive. Conceivably, neurodevelopmental abnormalities lead to an increased susceptibility to subsequent environmental insults such as psychological stress, autoimmune disease, infectious disease, oxidative stress, and episodes of depression and mania that further undermine brain structure and function. As has been previously suggested in the case of schizophrenia (Woods, 1998), we hypothesize that BD is best described as a “progressive neurodevelopmental disorder”; that is an abnormal ontogeny results in suboptimal neural function potentially leading to progressive neuropathological changes over time. Conflict of interest statement Wayne Drevets, M.D. is an employee of Janssen Pharmaceuticals of Johnson & Johnson, Inc., Titusville, NJ, USA. Jonathan Savitz, Ph.D. and Joseph Price, Ph.D. have no conflicts of interest to declare. Acknowledgements JS and WCD received support from The William K. Warren Foundation. The foundation played no role in the writing of the manuscript. References Abel, K.M., Wicks, S., Susser, E.S., Dalman, C., Pedersen, M.G., Mortensen, P.B., Webb, R.T., 2010. Birth weight, schizophrenia, and adult mental disorder: is risk confined to the smallest babies? Arch. Gen. Psychiatry 67, 923–930. Adler, C.M., Adams, J., DelBello, M.P., Holland, S.K., Schmithorst, V., Levine, A., Jarvis, K., Strakowski, S.M., 2006. Evidence of white matter pathology in bipolar disorder adolescents experiencing their first episode of mania: a diffusion tensor imaging study. Am. J. Psychiatry 163, 322–324. Aminoff, S.R., Hellvin, T., Lagerberg, T.V., Berg, A.O., Andreassen, O.A., Melle, I., 2013. Neurocognitive features in subgroups of bipolar disorder. Bipolar Disord. 15, 272–283. Arts, B., Jabben, N., Krabbendam, L., van Os, J., 2007. Meta-analyses of cognitive functioning in euthymic bipolar patients and their first-degree relatives. Psychol. Med., 1–15. Arts, B., Jabben, N., Krabbendam, L., van Os, J., 2011. A 2-year naturalistic study on cognitive functioning in bipolar disorder. Acta Psychiatr. Scand. 123, 190–205. Balanza-Martinez, V., Tabares-Seisdedos, R., Selva-Vera, G., Martinez-Aran, A., Torrent, C., Salazar-Fraile, J., Leal-Cercos, C., Vieta, E., Gomez-Beneyto, M., 2005. Persistent cognitive dysfunctions in bipolar I disorder and schizophrenic patients: a 3-year follow-up study. Psychother. Psychosom. 74, 113–119.

144

J.B. Savitz et al. / Neuroscience and Biobehavioral Reviews 42 (2014) 132–147

Bearden, C.E., Glahn, D.C., Caetano, S., Olvera, R.L., Fonseca, M., Najt, P., Hunter, K., Pliszka, S.R., Soares, J.C., 2007a. Evidence for disruption in prefrontal cortical functions in juvenile bipolar disorder. Bipolar Disord. 9 (Suppl. 1), 145–159. Bearden, C.E., Thompson, P.M., Dalwani, M., Hayashi, K.M., Lee, A.D., Nicoletti, M., Trakhtenbroit, M., Glahn, D.C., Brambilla, P., Sassi, R.B., Mallinger, A.G., Frank, E., Kupfer, D.J., Soares, J.C., 2007b. Greater cortical gray matter density in lithiumtreated patients with bipolar disorder. Biol. Psychiatry 62, 7–16. Beasley, C.L., Zhang, Z.J., Patten, I., Reynolds, G.P., 2002. Selective deficits in prefrontal cortical GABAergic neurons in schizophrenia defined by the presence of calciumbinding proteins. Biol. Psychiatry 52, 708–715. Benes, F.M., Todtenkopf, M.S., Logiotatos, P., Williams, M., 2000. Glutamate decarboxylase(65)-immunoreactive terminals in cingulate and prefrontal cortices of schizophrenic and bipolar brain. J. Chem. Neuroanat. 20, 259–269. Benes, F.M., Vincent, S.L., Todtenkopf, M., 2001. The density of pyramidal and nonpyramidal neurons in anterior cingulate cortex of schizophrenic and bipolar subjects. Biol. Psychiatry 50, 395–406. Beneyto, M., Meador-Woodruff, J.H., 2006. Lamina-specific abnormalities of AMPA receptor trafficking and signaling molecule transcripts in the prefrontal cortex in schizophrenia. Synapse 60, 585–598. Berk, M., Brnabic, A., Dodd, S., Kelin, K., Tohen, M., Malhi, G.S., Berk, L., Conus, P., McGorry, P.D., 2011. Does stage of illness impact treatment response in bipolar disorder? Empirical treatment data and their implication for the staging model and early intervention. Bipolar Disord. 13, 87–98. Berk, M., Hallam, K.T., McGorry, P.D., 2007. The potential utility of a staging model as a course specifier: a bipolar disorder perspective. J. Affect. Disord. 100, 279–281. Beyer, J.L., Taylor, W.D., MacFall, J.R., Kuchibhatla, M., Payne, M.E., Provenzale, J.M., Cassidy, F., Krishnan, K.R., 2005. Cortical white matter microstructural abnormalities in bipolar disorder. Neuropsychopharmacology 30, 2225–2229. Bielau, H., Steiner, J., Mawrin, C., Trubner, K., Brisch, R., Meyer-Lotz, G., Brodhun, M., Dobrowolny, H., Baumann, B., Gos, T., Bernstein, H.G., Bogerts, B., 2007. Dysregulation of GABAergic neurotransmission in mood disorders: a postmortem study. Ann. N. Y. Acad. Sci. 1096, 157–169. Bitanihirwe, B.K., Lim, M.P., Woo, T.U., 2010. N-methyl-d-aspartate receptor expression in parvalbumin-containing inhibitory neurons in the prefrontal cortex in bipolar disorder. Bipolar Disord. 12, 95–101. Boes, A.D., McCormick, L.M., Coryell, W.H., Nopoulos, P., 2008. Rostral anterior cingulate cortex volume correlates with depressed mood in normal healthy children. Biol. Psychiatry 63, 391–397. Bond, D.J., Ha, T.H., Lang, D.J., Su, W., Torres, I.J., Honer, W.G., Lam, R.W., Yatham, L.N., 2013. Body mass index-related regional gray and white matter volume reductions in first-episode mania patients. Biol. Psychiatry. Bora, E., Vahip, S., Akdeniz, F., Gonul, A.S., Eryavuz, A., Ogut, M., Alkan, M., 2007. The effect of previous psychotic mood episodes on cognitive impairment in euthymic bipolar patients. Bipolar Disord. 9, 468–477. Bora, E., Yucel, M., Pantelis, C., 2010. Neurocognitive markers of psychosis in bipolar disorder: a meta-analytic study. J. Affect. Disord. 127, 1–9. Bora, E., Yucel, M., Pantelis, C., Berk, M., 2011. Meta-analytic review of neurocognition in bipolar II disorder. Acta Psychiatr. Scand. 123, 165–174. Botteron, K.N., Vannier, M.W., Geller, B., Todd, R.D., Lee, B.C., 1995. Preliminary study of magnetic resonance imaging characteristics in 8- to 16-year-olds with mania. J. Am. Acad. Child Adolesc. Psychiatry 34, 742–749. Bouras, C., Kovari, E., Hof, P.R., Riederer, B.M., Giannakopoulos, P., 2001. Anterior cingulate cortex pathology in schizophrenia and bipolar disorder. Acta Neuropathol. 102, 373–379. Brooks 3rd, J.O., Bonner, J.C., Rosen, A.C., Wang, P.W., Hoblyn, J.C., Hill, S.J., Ketter, T.A., 2009. Dorsolateral and dorsomedial prefrontal gray matter density changes associated with bipolar depression. Psychiatry Res. 172, 200–204. Brunoni, A.R., Kemp, A.H., Dantas, E.M., Goulart, A.C., Nunes, M.A., Boggio, P.S., Mill, J.G., Lotufo, P.A., Fregni, F., Bensenor, I.M., 2013. Heart rate variability is a trait marker of major depressive disorder: evidence from the sertraline vs. electric current therapy to treat depression clinical study. Int. J. Neuropsychopharmacol. 16, 1937–1949. Bush, G., Luu, P., Posner, M.I., 2000. Cognitive and emotional influences in anterior cingulate cortex. Trends Cogn. Sci. 4, 215–222. Carney, R.M., Freedland, K.E., Veith, R.C., 2005. Depression, the autonomic nervous system, and coronary heart disease. Psychosom. Med. 67, S29–S33. Cassidy, C., Buchy, L., Bodnar, M., Dell’elce, J., Choudhry, Z., Fathalli, F., Sengupta, S., Fox, R., Malla, A., Lepage, M., Iyer, S., Joober, R., 2013. Association of a risk allele of ANK3 with cognitive performance and cortical thickness in patients with first-episode psychosis. J. Psychiatry Neurosci. 38, 120242. Cavanagh, J.T., Van Beck, M., Muir, W., Blackwood, D.H., 2002. Case–control study of neurocognitive function in euthymic patients with bipolar disorder: an association with mania. Br. J. Psychiatry 180, 320–326. Cecil, K.M., DelBello, M.P., Morey, R., Strakowski, S.M., 2002. Frontal lobe differences in bipolar disorder as determined by proton MR spectroscopy. Bipolar Disord. 4, 357–365. Cerqueira, J.J., Mailliet, F., Almeida, O.F., Jay, T.M., Sousa, N., 2007. The prefrontal cortex as a key target of the maladaptive response to stress. J. Neurosci. 27, 2781–2787. Chan, W.Y., Yang, G.L., Chia, M.Y., Woon, P.S., Lee, J., Keefe, R., Sitoh, Y.Y., Nowinski, W.L., Sim, K., 2010. Cortical and subcortical white matter abnormalities in adults with remitted first-episode mania revealed by Tract-Based Spatial Statistics. Bipolar Disord. 12, 383–389. Chana, G., Landau, S., Beasley, C., Everall, I.P., Cotter, D., 2003. Two-dimensional assessment of cytoarchitecture in the anterior cingulate cortex in major depressive disorder, bipolar disorder, and schizophrenia: evidence for decreased

neuronal somal size and increased neuronal density. Biol. Psychiatry 53, 1086–1098. Chang, K., Adleman, N., Dienes, K., Barnea-Goraly, N., Reiss, A., Ketter, T., 2003. Decreased N-acetylaspartate in children with familial bipolar disorder. Biol. Psychiatry 53, 1059–1065. Chaves, O.C., Lombardo, L.E., Bearden, C.E., Woolsey, M.D., Martinez, D.M., Barrett, J.A., Miller, A.L., Velligan, D.I., Glahn, D.C., 2011. Association of clinical symptoms and neurocognitive performance in bipolar disorder: a longitudinal study. Bipolar Disord. 13, 118–123. Chiu, S., Widjaja, F., Bates, M.E., Voelbel, G.T., Pandina, G., Marble, J., Blank, J.A., Day, J., Brule, N., Hendren, R.L., 2008. Anterior cingulate volume in pediatric bipolar disorder and autism. J. Affect. Disord. 105, 93–99. Chudal, R., Sourander, A., Polo-Kantola, P., Hinkka-Yli-Salomaki, S., Lehti, V., Sucksdorff, D., Gissler, M., Brown, A.S., 2013. Perinatal factors and the risk of bipolar disorder in Finland. J. Affect. Disord. Connor, C.M., Guo, Y., Akbarian, S., 2009. Cingulate white matter neurons in schizophrenia and bipolar disorder. Biol. Psychiatry 66, 486–493. Conway, C.R., Sheline, Y.I., Chibnall, J.T., Bucholz, R.D., Price, J.L., Gangwani, S., Mintun, M.A., 2012. Brain blood-flow change with acute vagus nerve stimulation in treatment-refractory major depressive disorder. Brain Stimul. 5, 163–171. Cook, S.C., Wellman, C.L., 2004. Chronic stress alters dendritic morphology in rat medial prefrontal cortex. J. Neurobiol. 60, 236–248. Cotter, D., Hudson, L., Landau, S., 2005. Evidence for orbitofrontal pathology in bipolar disorder and major depression, but not in schizophrenia. Bipolar Disord. 7, 358–369. Cotter, D., Landau, S., Beasley, C., Stevenson, R., Chana, G., MacMillan, L., Everall, I., 2002a. The density and spatial distribution of GABAergic neurons, labelled using calcium binding proteins, in the anterior cingulate cortex in major depressive disorder, bipolar disorder, and schizophrenia. Biol. Psychiatry 51, 377–386. Cotter, D., Mackay, D., Chana, G., Beasley, C., Landau, S., Everall, I.P., 2002b. Reduced neuronal size and glial cell density in area 9 of the dorsolateral prefrontal cortex in subjects with major depressive disorder. Cereb. Cortex 12, 386–394. Cotter, D., Mackay, D., Landau, S., Kerwin, R., Everall, I., 2001. Reduced glial cell density and neuronal size in the anterior cingulate cortex in major depressive disorder. Arch. Gen. Psychiatry 58, 545–553. Covington 3rd, H.E., Lobo, M.K., Maze, I., Vialou, V., Hyman, J.M., Zaman, S., LaPlant, Q., Mouzon, E., Ghose, S., Tamminga, C.A., Neve, R.L., Deisseroth, K., Nestler, E.J., 2010. Antidepressant effect of optogenetic stimulation of the medial prefrontal cortex. J. Neurosci. 30, 16082–16090. Czeh, B., Muller-Keuker, J.I., Rygula, R., Abumaria, N., Hiemke, C., Domenici, E., Fuchs, E., 2007. Chronic social stress inhibits cell proliferation in the adult medial prefrontal cortex: hemispheric asymmetry and reversal by fluoxetine treatment. Neuropsychopharmacology 32, 1490–1503. Dantzer, R., O’Connor, J.C., Freund, G.G., Johnson, R.W., Kelley, K.W., 2008. From inflammation to sickness and depression: when the immune system subjugates the brain. Nat. Rev. Neurosci. 9, 46–56. Dean, B., Gibbons, A.S., Tawadros, N., Brooks, L., Everall, I.P., Scarr, E., 2013. Different changes in cortical tumor necrosis factor-alpha-related pathways in schizophrenia and mood disorders. Mol. Psychiatry 18, 767–773. Delaloye, C., Moy, G., de Bilbao, F., Weber, K., Baudois, S., Haller, S., Xekardaki, A., Canuto, A., Giardini, U., Lovblad, K.O., Gold, G., Giannakopoulos, P., 2011. Longitudinal analysis of cognitive performances and structural brain changes in late-life bipolar disorder. Int. J. Geriatr. Psychiatry 26, 1309–1318. Dickstein, D.P., Milham, M.P., Nugent, A.C., Drevets, W.C., Charney, D.S., Pine, D.S., Leibenluft, E., 2005. Frontotemporal alterations in pediatric bipolar disorder: results of a voxel-based morphometry study. Arch. Gen. Psychiatry 62, 734–741. Dorph-Petersen, K.A., Lewis, D.A., 2011. Stereological approaches to identifying neuropathology in psychosis. Biol. Psychiatry 69, 113–126. Drevets, W.C., Ongur, D., Price, J.L., 1998. Neuroimaging abnormalities in the subgenual prefrontal cortex: implications for the pathophysiology of familial mood disorders. Mol. Psychiatry 3, 220–226 (190–221). Drevets, W.C., Price, J.L., Simpson, J.R., Todd, R.D., Reich, T., Vannier, M., Raichle, M.E., 1997. Subgenual prefrontal cortex abnormalities in mood disorders. Nature 386, 824–827. Drevets, W.C., Savitz, J., Trimble, M., 2008. The subgenual anterior cingulate cortex in mood disorders. CNS Spectr. 13, 663–681. Eastwood, S.L., Harrison, P.J., 2001. Synaptic pathology in the anterior cingulate cortex in schizophrenia and mood disorders. A review and a Western blot study of synaptophysin, GAP-43 and the complexins. Brain Res. Bull. 55, 569–578. Farber, N.B., Newcomer, J.W., Olney, J.W., 1998. The glutamate synapse in neuropsychiatric disorders. Focus on schizophrenia and Alzheimer’s disease. Prog. Brain Res. 116, 421–437. Farrow, T.F., Whitford, T.J., Williams, L.M., Gomes, L., Harris, A.W., 2005. Diagnosisrelated regional gray matter loss over two years in first episode schizophrenia and bipolar disorder. Biol. Psychiatry 58, 713–723. Fjell, A.M., Walhovd, K.B., 2010. Structural brain changes in aging: courses, causes and cognitive consequences. Rev. Neurosci. 21, 187–221. Foland-Ross, L.C., Thompson, P.M., Sugar, C.A., Madsen, S.K., Shen, J.K., Penfold, C., Ahlf, K., Rasser, P.E., Fischer, J., Yang, Y., Townsend, J., Bookheimer, S.Y., Altshuler, L.L., 2011. Investigation of cortical thickness abnormalities in lithium-free adults with bipolar I disorder using cortical pattern matching. Am. J. Psychiatry. Forester, B.P., Finn, C.T., Berlow, Y.A., Wardrop, M., Renshaw, P.F., Moore, C.M., 2008. Brain lithium, N-acetyl aspartate and myo-inositol levels in older adults with bipolar disorder treated with lithium: a lithium-7 and proton magnetic resonance spectroscopy study. Bipolar Disord. 10, 691–700.

J.B. Savitz et al. / Neuroscience and Biobehavioral Reviews 42 (2014) 132–147 Fornito, A., Malhi, G.S., Lagopoulos, J., Ivanovski, B., Wood, S.J., Saling, M.M., Pantelis, C., Yucel, M., 2008. Anatomical abnormalities of the anterior cingulate and paracingulate cortex in patients with bipolar I disorder. Psychiatry Res. 162, 123–132. Frangou, S., 2005. The Maudsley bipolar disorder project. Epilepsia 46 (Suppl. 4), 19–25. Frangou, S., Donaldson, S., Hadjulis, M., Landau, S., Goldstein, L.H., 2005. The Maudsley Bipolar Disorder Project: executive dysfunction in bipolar disorder I and its clinical correlates. Biol. Psychiatry 58, 859–864. Frazier, J.A., Breeze, J.L., Papadimitriou, G., Kennedy, D.N., Hodge, S.M., Moore, C.M., Howard, J.D., Rohan, M.P., Caviness, V.S., Makris, N., 2007. White matter abnormalities in children with and at risk for bipolar disorder. Bipolar Disord. 9, 799–809. Frey, B.N., Stanley, J.A., Nery, F.G., Monkul, E.S., Nicoletti, M.A., Chen, H.H., Hatch, J.P., Caetano, S.C., Ortiz, O., Kapczinski, F., Soares, J.C., 2007. Abnormal cellular energy and phospholipid metabolism in the left dorsolateral prefrontal cortex of medication-free individuals with bipolar disorder: an in vivo 1H MRS study. Bipolar Disord. 9 (Suppl. 1), 119–127. Frey, B.N., Zunta-Soares, G.B., Caetano, S.C., Nicoletti, M.A., Hatch, J.P., Brambilla, P., Mallinger, A.G., Soares, J.C., 2008. Illness duration and total brain gray matter in bipolar disorder: evidence for neurodegeneration? Eur. Neuropsychopharmacol. 18, 717–722. Gigante, A.D., Young, L.T., Yatham, L.N., Andreazza, A.C., Nery, F.G., Grinberg, L.T., Heinsen, H., Lafer, B., 2011. Morphometric post-mortem studies in bipolar disorder: possible association with oxidative stress and apoptosis. Int. J. Neuropsychopharmacol. 14, 1075–1089. Gildengers, A.G., Chisholm, D., Butters, M.A., Anderson, S.J., Begley, A., Holm, M., Rogers, J.C., Reynolds 3rd, C.F., Mulsant, B.H., 2013. Two-year course of cognitive function and instrumental activities of daily living in older adults with bipolar disorder: evidence for neuroprogression? Psychol. Med. 43, 801–811. Gildengers, A.G., Mulsant, B.H., Begley, A., Mazumdar, S., Hyams, A.V., Reynolds Iii, C.F., Kupfer, D.J., Butters, M.A., 2009. The longitudinal course of cognition in older adults with bipolar disorder. Bipolar Disord. 11, 744–752. Glahn, D.C., Bearden, C.E., Barguil, M., Barrett, J., Reichenberg, A., Bowden, C.L., Soares, J.C., Velligan, D.I., 2007. The neurocognitive signature of psychotic bipolar disorder. Biol. Psychiatry. Gogtay, N., Greenstein, D., Lenane, M., Clasen, L., Sharp, W., Gochman, P., Butler, P., Evans, A., Rapoport, J., 2007a. Cortical brain development in nonpsychotic siblings of patients with childhood-onset schizophrenia. Arch. Gen. Psychiatry 64, 772–780. Gogtay, N., Ordonez, A., Herman, D.H., Hayashi, K.M., Greenstein, D., Vaituzis, C., Lenane, M., Clasen, L., Sharp, W., Giedd, J.N., Jung, D., Nugent 3rd, T.F., Toga, A.W., Leibenluft, E., Thompson, P.M., Rapoport, J.L., 2007b. Dynamic mapping of cortical development before and after the onset of pediatric bipolar illness. J. Child Psychol. Psychiatry 48, 852–862. Gos, T., Steiner, J., Bielau, H., Dobrowolny, H., Gunther, K., Mawrin, C., Krzyzanowski, M., Hauser, R., Brisch, R., Bernstein, H.G., Jankowski, Z., Braun, K., Bogerts, B., 2012. Differences between unipolar and bipolar I depression in the quantitative analysis of glutamic acid decarboxylase-immunoreactive neuropil. Eur. Arch. Psychiatry Clin. Neurosci. 262, 647–655. Guidotti, A., Auta, J., Davis, J.M., Di-Giorgi-Gerevini, V., Dwivedi, Y., Grayson, D.R., Impagnatiello, F., Pandey, G., Pesold, C., Sharma, R., Uzunov, D., Costa, E., 2000. Decrease in reelin and glutamic acid decarboxylase67 (GAD67) expression in schizophrenia and bipolar disorder: a postmortem brain study. Arch. Gen. Psychiatry 57, 1061–1069. Hafeman, D.M., Chang, K.D., Garrett, A.S., Sanders, E.M., Phillips, M.L., 2012. Effects of medication on neuroimaging findings in bipolar disorder: an updated review. Bipolar Disord. 14, 375–410. Hajek, T., Calkin, C., Blagdon, R., Slaney, C., Alda, M., 2013. Type 2 diabetes mellitus: a potentially modifiable risk factor for neurochemical brain changes in bipolar disorders. Biol. Psychiatry. Harkavy-Friedman, J.M., Keilp, J.G., Grunebaum, M.F., Sher, L., Printz, D., Burke, A.K., Mann, J.J., Oquendo, M., 2006. Are BPI and BPII suicide attempters distinct neuropsychologically? J. Affect. Disord. 94, 255–259. Harrison, P.J., 2002. The neuropathology of primary mood disorder. Brain 125, 1428–1449. Hasler, G., Fromm, S., Carlson, P.J., Luckenbaugh, D.A., Waldeck, T., Geraci, M., Roiser, J.P., Neumeister, A., Meyers, N., Charney, D.S., Drevets, W.C., 2008. Neural response to catecholamine depletion in unmedicated subjects with major depressive disorder in remission and healthy subjects. Arch. Gen. Psychiatry 65, 521–531. Haukvik, U.K., McNeil, T., Lange, E.H., Melle, I., Dale, A.M., Andreassen, O.A., Agartz, I., 2013. Pre- and perinatal hypoxia associated with hippocampus/amygdala volume in bipolar disorder. Psychol. Med., 1–11. Haznedar, M.M., Roversi, F., Pallanti, S., Baldini-Rossi, N., Schnur, D.B., Licalzi, E.M., Tang, C., Hof, P.R., Hollander, E., Buchsbaum, M.S., 2005. Fronto-thalamo-striatal gray and white matter volumes and anisotropy of their connections in bipolar spectrum illnesses. Biol. Psychiatry 57, 733–742. Henry, B.L., Minassian, A., Paulus, M.P., Geyer, M.A., Perry, W., 2010. Heart rate variability in bipolar mania and schizophrenia. J. Psychiatr. Res. 44, 168–176. Hercher, C., Canetti, L., Turecki, G., Mechawar, N., 2010. Anterior cingulate pyramidal neurons display altered dendritic branching in depressed suicides. J. Psychiatr. Res. 44, 286–293. Hirayasu, Y., Shenton, M.E., Salisbury, D.F., Kwon, J.S., Wible, C.G., Fischer, I.A., Yurgelun-Todd, D., Zarate, C., Kikinis, R., Jolesz, F.A., McCarley, R.W., 1999.

145

Subgenual cingulate cortex volume in first-episode psychosis. Am. J. Psychiatry 156, 1091–1093. Ho, B.C., Andreasen, N.C., Dawson, J.D., Wassink, T.H., 2007. Association between brain-derived neurotrophic factor Val66Met gene polymorphism and progressive brain volume changes in schizophrenia. Am. J. Psychiatry 164, 1890–1899. Houenou, J., Wessa, M., Douaud, G., Leboyer, M., Chanraud, S., Perrin, M., Poupon, C., Martinot, J.L., Paillere-Martinot, M.L., 2007. Increased white matter connectivity in euthymic bipolar patients: diffusion tensor tractography between the subgenual cingulate and the amygdalo-hippocampal complex. Mol. Psychiatry 12, 1001–1010. Hsiao, Y.L., Wu, Y.S., Wu, J.Y., Hsu, M.H., Chen, H.C., Lee, S.Y., Lee, I.H., Yeh, T.L., Yang, Y.K., Ko, H.C., Lu, R.B., 2009. Neuropsychological functions in patients with bipolar I and bipolar II disorder. Bipolar Disord. 11, 547–554. Irwin, M.R., Cole, S.W., 2011. Reciprocal regulation of the neural and innate immune systems. Nat. Rev. Immunol. 11, 625–632. James, A., Hough, M., James, S., Burge, L., Winmill, L., Nijhawan, S., Matthews, P.M., Zarei, M., 2011. Structural brain and neuropsychometric changes associated with pediatric bipolar disorder with psychosis. Bipolar Disord. 13, 16–27. Kafantaris, V., Kingsley, P., Ardekani, B., Saito, E., Lencz, T., Lim, K., Szeszko, P., 2009. Lower orbital frontal white matter integrity in adolescents with bipolar I disorder. J. Am. Acad. Child Adolesc. Psychiatry 48, 79–86. Kalayci, D., Ozdel, O., Sozeri-Varma, G., Kiroglu, Y., Tumkaya, S., 2012. A proton magnetic resonance spectroscopy study in schizoaffective disorder: comparison of bipolar disorder and schizophrenia. Prog. Neuropsychopharmacol. Biol. Psychiatry 37, 176–181. Kalmar, J.H., Wang, F., Spencer, L., Edmiston, E., Lacadie, C.M., Martin, A., Constable, R.T., Duncan, J.S., Staib, L.H., Papademetris, X., Blumberg, H.P., 2009. Preliminary evidence for progressive prefrontal abnormalities in adolescents and young adults with bipolar disorder. J. Int. Neuropsychol. Soc. 15, 476–481. Kapczinski, F., Dias, V.V., Kauer-Sant’Anna, M., Frey, B.N., Grassi-Oliveira, R., Colom, F., Berk, M., 2009. Clinical implications of a staging model for bipolar disorders. Expert Rev. Neurother. 9, 957–966. Kaur, S., Sassi, R.B., Axelson, D., Nicoletti, M., Brambilla, P., Monkul, E.S., Hatch, J.P., Keshavan, M.S., Ryan, N., Birmaher, B., Soares, J.C., 2005. Cingulate cortex anatomical abnormalities in children and adolescents with bipolar disorder. Am. J. Psychiatry 162, 1637–1643. Kemp, A.H., Quintana, D.S., Felmingham, K.L., Matthews, S., Jelinek, H.F., 2012. Depression, comorbid anxiety disorders, and heart rate variability in physically healthy, unmedicated patients: implications for cardiovascular risk. PLoS One 7, e30777. Kessing, L.V., 1998. Cognitive impairment in the euthymic phase of affective disorder. Psychol. Med. 28, 1027–1038. Kessing, L.V., Andersen, E.W., Andersen, P.K., 2000. Predictors of recurrence in affective disorder – analyses accounting for individual heterogeneity. J. Affect. Disord. 57, 139–145. Kessing, L.V., Sondergard, L., Forman, J.L., Andersen, P.K., 2008. Lithium treatment and risk of dementia. Arch. Gen. Psychiatry 65, 1331–1335. Kim, S., Webster, M.J., 2010. Correlation analysis between genome-wide expression profiles and cytoarchitectural abnormalities in the prefrontal cortex of psychiatric disorders. Mol. Psychiatry 15, 326–336. Kingsbury, A.E., Foster, O.J., Nisbet, A.P., Cairns, N., Bray, L., Eve, D.J., Lees, A.J., Marsden, C.D., 1995. Tissue pH as an indicator of mRNA preservation in human post-mortem brain. Brain Res. Mol. Brain Res. 28, 311–318. Konopaske, G.T., Dorph-Petersen, K.A., Sweet, R.A., Pierri, J.N., Zhang, W., Sampson, A.R., Lewis, D.A., 2008. Effect of chronic antipsychotic exposure on astrocyte and oligodendrocyte numbers in macaque monkeys. Biol. Psychiatry 63, 759–765. Koo, M.S., Levitt, J.J., Salisbury, D.F., Nakamura, M., Shenton, M.E., McCarley, R.W., 2008. A cross-sectional and longitudinal magnetic resonance imaging study of cingulate gyrus gray matter volume abnormalities in first-episode schizophrenia and first-episode affective psychosis. Arch. Gen. Psychiatry 65, 746–760. Korbo, L., 1999. Glial cell loss in the hippocampus of alcoholics. Alcohol. Clin. Exp. Res. 23, 164–168. Lane, R.D., McRae, K., Reiman, E.M., Chen, K., Ahern, G.L., Thayer, J.F., 2009. Neural correlates of heart rate variability during emotion. Neuroimage 44, 213–222. Law, A.J., Harrison, P.J., 2003. The distribution and morphology of prefrontal cortex pyramidal neurons identified using anti-neurofilament antibodies SMI32, N200 and FNP7. Normative data and a comparison in subjects with schizophrenia, bipolar disorder or major depression. J. Psychiatr. Res. 37, 487–499. Lewandowski, K.E., Cohen, B.M., Ongur, D., 2011. Evolution of neuropsychological dysfunction during the course of schizophrenia and bipolar disorder. Psychol. Med. 41, 225–241. Lim, C.S., Baldessarini, R.J., Vieta, E., Yucel, M., Bora, E., Sim, K., 2013. Longitudinal neuroimaging and neuropsychological changes in bipolar disorder patients: review of the evidence. Neurosci. Biobehav. Rev. 37, 418–435. Lin, A., Reniers, R.L., Wood, S.J., 2013. Clinical staging in severe mental disorder: evidence from neurocognition and neuroimaging. Br. J. Psychiatry Suppl. 54, s11–s17. Liston, C., Miller, M.M., Goldwater, D.S., Radley, J.J., Rocher, A.B., Hof, P.R., Morrison, J.H., McEwen, B.S., 2006. Stress-induced alterations in prefrontal cortical dendritic morphology predict selective impairments in perceptual attentional set-shifting. J. Neurosci. 26, 7870–7874. Lisy, M.E., Jarvis, K.B., DelBello, M.P., Mills, N.P., Weber, W.A., Fleck, D., Strakowski, S.M., Adler, C.M., 2011. Progressive neurostructural changes in adolescent and adult patients with bipolar disorder. Bipolar Disord. 13, 396–405.

146

J.B. Savitz et al. / Neuroscience and Biobehavioral Reviews 42 (2014) 132–147

Lopez-Jaramillo, C., Lopera-Vasquez, J., Gallo, A., Ospina-Duque, J., Bell, V., Torrent, C., Martinez-Aran, A., Vieta, E., 2010. Effects of recurrence on the cognitive performance of patients with bipolar I disorder: implications for relapse prevention and treatment adherence. Bipolar Disord. 12, 557–567. Lu, L.H., Zhou, X.J., Keedy, S.K., Reilly, J.L., Sweeney, J.A., 2011. White matter microstructure in untreated first episode bipolar disorder with psychosis: comparison with schizophrenia. Bipolar Disord. 13, 604–613. Lyoo, I.K., Sung, Y.H., Dager, S.R., Friedman, S.D., Lee, J.Y., Kim, S.J., Kim, N., Dunner, D.L., Renshaw, P.F., 2006. Regional cerebral cortical thinning in bipolar disorder. Bipolar Disord. 8, 65–74. MacDonald, M.L., Naydenov, A., Chu, M., Matzilevich, D., Konradi, C., 2006. Decrease in creatine kinase messenger RNA expression in the hippocampus and dorsolateral prefrontal cortex in bipolar disorder. Bipolar Disord. 8, 255–264. Mahon, K., Burdick, K.E., Wu, J., Ardekani, B.A., Szeszko, P.R., 2012. Relationship between suicidality and impulsivity in bipolar I disorder: a diffusion tensor imaging study. Bipolar Disord. 14, 80–89. Manji, H.K., Moore, G.J., Chen, G., 2000. Clinical and preclinical evidence for the neurotrophic effects of mood stabilizers: implications for the pathophysiology and treatment of manic-depressive illness. Biol. Psychiatry 48, 740–754. Martinez-Aran, A., Torrent, C., Tabares-Seisdedos, R., Salamero, M., Daban, C., Balanza-Martinez, V., Sanchez-Moreno, J., Manuel Goikolea, J., Benabarre, A., Colom, F., Vieta, E., 2008. Neurocognitive impairment in bipolar patients with and without history of psychosis. J. Clin. Psychiatry, e1–e7. Martinez-Aran, A., Vieta, E., Colom, F., Torrent, C., Sanchez-Moreno, J., Reinares, M., Benabarre, A., Goikolea, J.M., Brugue, E., Daban, C., Salamero, M., 2004. Cognitive impairment in euthymic bipolar patients: implications for clinical and functional outcome. Bipolar Disord. 6, 224–232. Matsuo, K., Nicoletti, M.A., Peluso, M.A., Hatch, J.P., Nemoto, K., Watanabe, Y., Nery, F.G., Monkul, E.S., Zunta-Soares, G.B., Bowden, C.L., Soares, J.C., 2009a. Anterior cingulate volumes associated with trait impulsivity in individuals with bipolar disorder. Bipolar Disord. 11, 628–636. Matsuo, K., Walss-Bass, C., Nery, F.G., Nicoletti, M.A., Hatch, J.P., Frey, B.N., Monkul, E.S., Zunta-Soares, G.B., Bowden, C.L., Escamilla, M.A., Soares, J.C., 2009b. Neuronal correlates of brain-derived neurotrophic factor Val66Met polymorphism and morphometric abnormalities in bipolar disorder. Neuropsychopharmacology 34, 1904–1913. McDonald, C., Bullmore, E.T., Sham, P.C., Chitnis, X., Wickham, H., Bramon, E., Murray, R.M., 2004. Association of genetic risks for schizophrenia and bipolar disorder with specific and generic brain structural endophenotypes. Arch. Gen. Psychiatry 61, 974–984. McIntosh, A.M., Maniega, S.M., Lymer, G.K., McKirdy, J., Hall, J., Sussmann, J.E., Bastin, M.E., Clayden, J.D., Johnstone, E.C., Lawrie, S.M., 2008. White matter tractography in bipolar disorder and schizophrenia. Biol. Psychiatry 64, 1088–1092. McNamara, R.K., Nandagopal, J.J., Strakowski, S.M., DelBello, M.P., 2010. Preventative strategies for early-onset bipolar disorder: towards a clinical staging model. CNS Drugs 24, 983–996. Miller, A.H., Maletic, V., Raison, C.L., 2009. Inflammation and its discontents: the role of cytokines in the pathophysiology of major depression. Biol. Psychiatry 65, 732–741. Mirakhur, A., Moorhead, T.W., Stanfield, A.C., McKirdy, J., Sussmann, J.E., Hall, J., Lawrie, S.M., Johnstone, E.C., McIntosh, A.M., 2009. Changes in gyrification over 4 years in bipolar disorder and their association with the brain-derived neurotrophic factor valine(66) methionine variant. Biol. Psychiatry 66, 293–297. Molina, V., Sanchez, J., Sanz, J., Reig, S., Benito, C., Leal, I., Sarramea, F., Rebolledo, R., Palomo, T., Desco, M., 2007. Dorsolateral prefrontal N-acetyl-aspartate concentration in male patients with chronic schizophrenia and with chronic bipolar disorder. Eur. Psychiatry 22, 505–512. Moore, G.J., Bebchuk, J.M., Wilds, I.B., Chen, G., Manji, H.K., 2000. Lithium-induced increase in human brain grey matter. Lancet 356, 1241–1242. Moore, G.J., Cortese, B.M., Glitz, D.A., Zajac-Benitez, C., Quiroz, J.A., Uhde, T.W., Drevets, W.C., Manji, H.K., 2009. A longitudinal study of the effects of lithium treatment on prefrontal and subgenual prefrontal gray matter volume in treatment-responsive bipolar disorder patients. J. Clin. Psychiatry 70, 699–705. Moore, G.J., Galloway, M.P., 2002. Magnetic resonance spectroscopy: neurochemistry and treatment effects in affective disorders. Psychopharmacol. Bull. 36, 5–23. Mosebach, J., Keilhoff, G., Gos, T., Schiltz, K., Schoeneck, L., Dobrowolny, H., Mawrin, C., Muller, S., Schroeter, M.L., Bernstein, H.G., Bogerts, B., Steiner, J., 2013. Increased nuclear Olig1-expression in the pregenual anterior cingulate white matter of patients with major depression: a regenerative attempt to compensate oligodendrocyte loss? J. Psychiatr. Res. Narita, K., Suda, M., Takei, Y., Aoyama, Y., Majima, T., Kameyama, M., Kosaka, H., Amanuma, M., Fukuda, M., Mikuni, M., 2010. Volume reduction of ventromedial prefrontal cortex in bipolar II patients with rapid cycling: a voxel-based morphometric study. Prog. Neuropsychopharmacol. Biol. Psychiatry. Nosarti, C., Reichenberg, A., Murray, R.M., Cnattingius, S., Lambe, M.P., Yin, L., MacCabe, J., Rifkin, L., Hultman, C.M., 2012. Preterm birth and psychiatric disorders in young adult life. Arch. Gen. Psychiatry 69, E1–E8. Nugent, A.C., Bain, E.E., Thayer, J.F., Sollers 3rd, J.J., Drevets, W.C., 2011. Heart rate variability during motor and cognitive tasks in females with major depressive disorder. Psychiatry Res. 191, 1–8. Nugent, A.C., Milham, M.P., Bain, E.E., Mah, L., Cannon, D.M., Marrett, S., Zarate, C.A., Pine, D.S., Price, J.L., Drevets, W.C., 2006. Cortical abnormalities in bipolar disorder investigated with MRI and voxel-based morphometry. Neuroimage 30, 485–497.

Nunes, P.V., Forlenza, O.V., Gattaz, W.F., 2007. Lithium and risk for Alzheimer’s disease in elderly patients with bipolar disorder. Br. J. Psychiatry 190, 359–360. O’Connor, J.A., Hemby, S.E., 2007. Elevated GRIA1 mRNA expression in Layer II/III and V pyramidal cells of the DLPFC in schizophrenia. Schizophr. Res. 97, 277–288. Ohira, H., Matsunaga, M., Osumi, T., Fukuyama, S., Shinoda, J., Yamada, J., Gidron, Y., 2013. Vagal nerve activity as a moderator of brain–immune relationships. J. Neuroimmunol. 260, 28–36. Olney, J.W., Newcomer, J.W., Farber, N.B., 1999. NMDA receptor hypofunction model of schizophrenia. J. Psychiatr. Res. 33, 523–533. Olvera, R.L., Caetano, S.C., Fonseca, M., Nicoletti, M., Stanley, J.A., Chen, H.H., Hatch, J.P., Hunter, K., Pliszka, S.R., Soares, J.C., 2007. Low levels of N-acetyl aspartate in the left dorsolateral prefrontal cortex of pediatric bipolar patients. J. Child Adolesc. Psychopharmacol. 17, 461–473. Olvet, D.M., Burdick, K.E., Cornblatt, B.A., 2013. Assessing the potential to use neurocognition to predict who is at risk for developing bipolar disorder: a review of the literature. Cogn. Neuropsychiatry 18, 129–145. Ongur, D., Drevets, W.C., Price, J.L., 1998. Glial reduction in the subgenual prefrontal cortex in mood disorders. Proc. Natl. Acad. Sci. U.S.A. 95, 13290–13295. Ongur, D., Ferry, A.T., Price, J.L., 2003. Architectonic subdivision of the human orbital and medial prefrontal cortex. J. Comp. Neurol. 460, 425–449. Ongur, D., Jensen, J.E., Prescot, A.P., Stork, C., Lundy, M., Cohen, B.M., Renshaw, P.F., 2008. Abnormal glutamatergic neurotransmission and neuronal–glial interactions in acute mania. Biol. Psychiatry 64, 718–726. Oppenheimer, D.R., 1984. Diseases of the Basal Ganglia, Cerebellum, and Motor Neurons. Greenfield’s Neuropathology, John Wiley, New York, pp. 699–747. Padmos, R.C., Hillegers, M.H., Knijff, E.M., Vonk, R., Bouvy, A., Staal, F.J., de Ridder, D., Kupka, R.W., Nolen, W.A., Drexhage, H.A., 2008. A discriminating messenger RNA signature for bipolar disorder formed by an aberrant expression of inflammatory genes in monocytes. Arch. Gen. Psychiatry 65, 395–407. Pavuluri, M.N., West, A., Hill, S.K., Jindal, K., Sweeney, J.A., 2009. Neurocognitive function in pediatric bipolar disorder: 3-year follow-up shows cognitive development lagging behind healthy youths. J. Am. Acad. Child Adolesc. Psychiatry 48, 299–307. Pennington, K., Beasley, C.L., Dicker, P., Fagan, A., English, J., Pariante, C.M., Wait, R., Dunn, M.J., Cotter, D.R., 2008. Prominent synaptic and metabolic abnormalities revealed by proteomic analysis of the dorsolateral prefrontal cortex in schizophrenia and bipolar disorder. Mol. Psychiatry 13, 1102–1117. Pillai, J.J., Friedman, L., Stuve, T.A., Trinidad, S., Jesberger, J.A., Lewin, J.S., Findling, R.L., Swales, T.P., Schulz, S.C., 2002. Increased presence of white matter hyperintensities in adolescent patients with bipolar disorder. Psychiatry Res. 114, 51–56. Price, J.L., Drevets, W.C., 2010. Neurocircuitry of mood disorders. Neuropsychopharmacology 35, 192–216. Radley, J.J., Rocher, A.B., Janssen, W.G., Hof, P.R., McEwen, B.S., Morrison, J.H., 2005. Reversibility of apical dendritic retraction in the rat medial prefrontal cortex following repeated stress. Exp. Neurol. 196, 199–203. Radley, J.J., Rocher, A.B., Miller, M., Janssen, W.G., Liston, C., Hof, P.R., McEwen, B.S., Morrison, J.H., 2006. Repeated stress induces dendritic spine loss in the rat medial prefrontal cortex. Cereb. Cortex 16, 313–320. Radley, J.J., Rocher, A.B., Rodriguez, A., Ehlenberger, D.B., Dammann, M., McEwen, B.S., Morrison, J.H., Wearne, S.L., Hof, P.R., 2008. Repeated stress alters dendritic spine morphology in the rat medial prefrontal cortex. J. Comp. Neurol. 507, 1141–1150. Rajkowska, G., Halaris, A., Selemon, L.D., 2001. Reductions in neuronal and glial density characterize the dorsolateral prefrontal cortex in bipolar disorder. Biol. Psychiatry 49, 741–752. Regenold, W.T., Phatak, P., Marano, C.M., Gearhart, L., Viens, C.H., Hisley, K.C., 2007. Myelin staining of deep white matter in the dorsolateral prefrontal cortex in schizophrenia, bipolar disorder, and unipolar major depression. Psychiatry Res. 151, 179–188. Robinson, L.J., Thompson, J.M., Gallagher, P., Goswami, U., Young, A.H., Ferrier, I.N., Moore, P.B., 2006. A meta-analysis of cognitive deficits in euthymic patients with bipolar disorder. J. Affect. Disord. 93, 105–115. Roiser, J.P., Levy, J., Fromm, S.J., Nugent, A.C., Talagala, S.L., Hasler, G., Henn, F.A., Sahakian, B.J., Drevets, W.C., 2009. The effects of tryptophan depletion on neural responses to emotional words in remitted depression. Biol. Psychiatry 66, 441–450. Rolls, E.T., 2000. The orbitofrontal cortex and reward. Cereb. Cortex 10, 284–294. Rolls, E.T., Grabenhorst, F., 2008. The orbitofrontal cortex and beyond: from affect to decision-making. Prog. Neurobiol. 86, 216–244. Rosas-Ballina, M., Tracey, K.J., 2009. Cholinergic control of inflammation. J. Intern. Med. 265, 663–679. Sakai, T., Oshima, A., Nozaki, Y., Ida, I., Haga, C., Akiyama, H., Nakazato, Y., Mikuni, M., 2008. Changes in density of calcium-binding-protein-immunoreactive GABAergic neurons in prefrontal cortex in schizophrenia and bipolar disorder. Neuropathology 28, 143–150. Sanchez-Moreno, J., Martinez-Aran, A., Tabares-Seisdedos, R., Torrent, C., Vieta, E., Ayuso-Mateos, J.L., 2009. Functioning and disability in bipolar disorder: an extensive review. Psychother. Psychosom. 78, 285–297. Sassi, R.B., Stanley, J.A., Axelson, D., Brambilla, P., Nicoletti, M.A., Keshavan, M.S., Ramos, R.T., Ryan, N., Birmaher, B., Soares, J.C., 2005. Reduced NAA levels in the dorsolateral prefrontal cortex of young bipolar patients. Am. J. Psychiatry 162, 2109–2115. Savitz, J., Drevets, W.C., 2009a. Bipolar and major depressive disorder: neuroimaging the developmental-degenerative divide. Neurosci. Biobehav. Rev. 33, 699–771.

J.B. Savitz et al. / Neuroscience and Biobehavioral Reviews 42 (2014) 132–147 Savitz, J., Frank, M.B., Victor, T., Bebak, M., Marino, J.H., Bellgowan, P.S., McKinney, B.A., Bodurka, J., Kent Teague, T., Drevets, W.C., 2013a. Inflammation and neurological disease-related genes are differentially expressed in depressed patients with mood disorders and correlate with morphometric and functional imaging abnormalities. Brain Behav. Immun. 31, 161–171. Savitz, J., Nugent, A.C., Bellgowan, P.S.F., Wright, N., Tinsley, R., Zarate, J.C.A., Herscovitch, P., Drevets, W.C., 2013b. Catecholamine depletion in first-degree relatives of individuals with mood disorders: an [18 F]fluorodeoxyglucose positron emission tomography study. Neuroimage Clin. 2, 341–355. Savitz, J., Nugent, A.C., Bogers, W., Liu, A., Sills, R., Luckenbaugh, D.A., Bain, E.E., Price, J.L., Zarate, C., Manji, H.K., Cannon, D.M., Marrett, S., Charney, D.S., Drevets, W.C., 2010. Amygdala volume in depressed patients with bipolar disorder assessed using high resolution 3 T MRI: the impact of medication. Neuroimage 49, 2966–2976. Savitz, J., Solms, M., Ramesar, R., 2005a. Neuropsychological dysfunction in bipolar affective disorder: a critical opinion. Bipolar Disord. 7, 216–235. Savitz, J., van der Merwe, L., Stein, D.J., Solms, M., Ramesar, R., 2009. Neuropsychological status of bipolar I disorder: impact of psychosis. Br. J. Psychiatry 194, 243–251. Savitz, J.B., Drevets, W.C., 2009b. Imaging phenotypes of major depressive disorder: genetic correlates. Neuroscience 164, 300–330. Savitz, J.B., Nugent, A.C., Bogers, W., Roiser, J.P., Bain, E.E., Neumeister, A., Zarate Jr., C.A., Manji, H.K., Cannon, D.M., Marrett, S., Henn, F., Charney, D.S., Drevets, W.C., 2011. Habenula volume in bipolar disorder and major depressive disorder: a high-resolution magnetic resonance imaging study. Biol. Psychiatry 69, 336–343. Savitz, J.B., Ramesar, R.S., 2006. Personality: is it a viable endophenotype for genetic studies of bipolar affective disorder? Bipolar Disord. 8, 322–337. Savitz, J.B., Rauch, S.L., Drevets, W.C., 2013c. Clinical application of brain imaging for the diagnosis of mood disorders: the current state of play. Mol. Psychiatry 18, 528–539. Savitz, J.B., Solms, M., Ramesar, R.S., 2005b. Neurocognitive function as an endophenotype for genetic studies of bipolar affective disorder. Neuromolecular Med. 7, 275–286. Schneider, M.R., DelBello, M.P., McNamara, R.K., Strakowski, S.M., Adler, C.M., 2012. Neuroprogression in bipolar disorder. Bipolar Disord. 14, 356–374. Selemon, L.D., Lidow, M.S., Goldman-Rakic, P.S., 1999. Increased volume and glial density in primate prefrontal cortex associated with chronic antipsychotic drug exposure. Biol. Psychiatry 46, 161–172. Simonsen, C., Sundet, K., Vaskinn, A., Birkenaes, A.B., Engh, J.A., Faerden, A., Jonsdottir, H., Ringen, P.A., Opjordsmoen, S., Melle, I., Friis, S., Andreassen, O.A., 2011. Neurocognitive dysfunction in bipolar and schizophrenia spectrum disorders depends on history of psychosis rather than diagnostic group. Schizophr. Bull. 37, 73–83. Singh, M.K., Chang, K.D., Chen, M.C., Kelley, R.G., Garrett, A., Mitsunaga, M.M., Bararpour, L., Howe, M., Reiss, A.L., Gotlib, I.H., 2012. Volumetric reductions in the subgenual anterior cingulate cortex in adolescents with bipolar I disorder. Bipolar Disord. 14, 585–596. Sklar, P., Gabriel, S.B., McInnis, M.G., Bennett, P., Lim, Y.M., Tsan, G., Schaffner, S., Kirov, G., Jones, I., Owen, M., Craddock, N., DePaulo, J.R., Lander, E.S., 2002. Family-based association study of 76 candidate genes in bipolar disorder: BDNF is a potential risk locus. Brain-derived neutrophic factor. Mol. Psychiatry 7, 579–593. Steiner, J., Bielau, H., Brisch, R., Danos, P., Ullrich, O., Mawrin, C., Bernstein, H.G., Bogerts, B., 2008. Immunological aspects in the neurobiology of suicide: elevated microglial density in schizophrenia and depression is associated with suicide. J. Psychiatr. Res. 42, 151–157. Strakowski, S.M., Woods, B.T., Tohen, M., Wilson, D.R., Douglass, A.W., Stoll, A.L., 1993. MRI subcortical signal hyperintensities in mania at first hospitalization. Biol. Psychiatry 33, 204–206. Sullivan, R.M., Gratton, A., 1999. Lateralized effects of medial prefrontal cortex lesions on neuroendocrine and autonomic stress responses in rats. J. Neurosci. 19, 2834–2840. Sussmann, J.E., Lymer, G.K., McKirdy, J., Moorhead, T.W., Maniega, S.M., Job, D., Hall, J., Bastin, M.E., Johnstone, E.C., Lawrie, S.M., McIntosh, A.M., 2009. White matter abnormalities in bipolar disorder and schizophrenia detected using diffusion tensor magnetic resonance imaging. Bipolar Disord. 11, 11–18. Thayer, J.F., Ahs, F., Fredrikson, M., Sollers 3rd, J.J., Wager, T.D., 2012. A meta-analysis of heart rate variability and neuroimaging studies: implications for heart rate variability as a marker of stress and health. Neurosci. Biobehav. Rev. 36, 747–756. Thompson, M., Weickert, C.S., Wyatt, E., Webster, M.J., 2009. Decreased glutamic acid decarboxylase(67) mRNA expression in multiple brain areas of patients with schizophrenia and mood disorders. J. Psychiatr. Res. 43, 970–977. Tkachev, D., Mimmack, M.L., Ryan, M.M., Wayland, M., Freeman, T., Jones, P.B., Starkey, M., Webster, M.J., Yolken, R.H., Bahn, S., 2003. Oligodendrocyte dysfunction in schizophrenia and bipolar disorder. Lancet 362, 798–805.

147

Torrent, C., Martinez-Aran, A., del Mar Bonnin, C., Reinares, M., Daban, C., Sole, B., Rosa, A.R., Tabares-Seisdedos, R., Popovic, D., Salamero, M., Vieta, E., 2012. Longterm outcome of cognitive impairment in bipolar disorder. J. Clin. Psychiatry 73, e899–e905. Torres-Platas, S.G., Hercher, C., Davoli, M.A., Maussion, G., Labonte, B., Turecki, G., Mechawar, N., 2011. Astrocytic hypertrophy in anterior cingulate white matter of depressed suicides. Neuropsychopharmacology 36, 2650–2658. Udupa, K., Sathyaprabha, T.N., Thirthalli, J., Kishore, K.R., Lavekar, G.S., Raju, T.R., Gangadhar, B.N., 2007. Alteration of cardiac autonomic functions in patients with major depression: a study using heart rate variability measures. J. Affect. Disord. 100, 137–141. Uranova, N.A., Vostrikov, V.M., Orlovskaya, D.D., Rachmanova, V.I., 2004. Oligodendroglial density in the prefrontal cortex in schizophrenia and mood disorders: a study from the Stanley Neuropathology Consortium. Schizophr. Res. 67, 269–275. van Gorp, W.G., Altshuler, L., Theberge, D.C., Wilkins, J., Dixon, W., 1998. Cognitive impairment in euthymic bipolar patients with and without prior alcohol dependence. A preliminary study. Arch. Gen. Psychiatry 55, 41–46. Veldic, M., Guidotti, A., Maloku, E., Davis, J.M., Costa, E., 2005. In psychosis, cortical interneurons overexpress DNA-methyltransferase 1. Proc. Natl. Acad. Sci. U.S.A. 102, 2152–2157. Versace, A., Almeida, J.R., Hassel, S., Walsh, N.D., Novelli, M., Klein, C.R., Kupfer, D.J., Phillips, M.L., 2008. Elevated left and reduced right orbitomedial prefrontal fractional anisotropy in adults with bipolar disorder revealed by tract-based spatial statistics. Arch. Gen. Psychiatry 65, 1041–1052. Vostrikov, V.M., Uranova, N.A., Orlovskaya, D.D., 2007. Deficit of perineuronal oligodendrocytes in the prefrontal cortex in schizophrenia and mood disorders. Schizophr. Res. 94, 273–280. Wang, F., Jackowski, M., Kalmar, J.H., Chepenik, L.G., Tie, K., Qiu, M., Gong, G., Pittman, B.P., Jones, M.M., Shah, M.P., Spencer, L., Papademetris, X., Constable, R.T., Blumberg, H.P., 2008. Abnormal anterior cingulum integrity in bipolar disorder determined through diffusion tensor imaging. Br. J. Psychiatry 193, 126–129. Wang, F., Kalmar, J.H., He, Y., Jackowski, M., Chepenik, L.G., Edmiston, E.E., Tie, K., Gong, G., Shah, M.P., Jones, M., Uderman, J., Constable, R.T., Blumberg, H.P., 2009. Functional and structural connectivity between the perigenual anterior cingulate and amygdala in bipolar disorder. Biol. Psychiatry 66, 516–521. Webster, M.J., O’Grady, J., Kleinman, J.E., Weickert, C.S., 2005. Glial fibrillary acidic protein mRNA levels in the cingulate cortex of individuals with depression, bipolar disorder and schizophrenia. Neuroscience 133, 453–461. Wilke, M., Kowatch, R.A., DelBello, M.P., Mills, N.P., Holland, S.K., 2004. Voxel-based morphometry in adolescents with bipolar disorder: first results. Psychiatry Res. 131, 57–69. Williams, M.R., Hampton, T., Pearce, R.K., Hirsch, S.R., Ansorge, O., Thom, M., Maier, M., 2013. Astrocyte decrease in the subgenual cingulate and callosal genu in schizophrenia. Eur. Arch. Psychiatry Clin. Neurosci. 263, 41–52. Woo, T.U., Kim, A.M., Viscidi, E., 2008a. Disease-specific alterations in glutamatergic neurotransmission on inhibitory interneurons in the prefrontal cortex in schizophrenia. Brain Res. 1218, 267–277. Woo, T.U., Shrestha, K., Amstrong, C., Minns, M.M., Walsh, J.P., Benes, F.M., 2007. Differential alterations of kainate receptor subunits in inhibitory interneurons in the anterior cingulate cortex in schizophrenia and bipolar disorder. Schizophr. Res. 96, 46–61. Woo, T.U., Shrestha, K., Lamb, D., Minns, M.M., Benes, F.M., 2008b. N-methyl-daspartate receptor and calbindin-containing neurons in the anterior cingulate cortex in schizophrenia and bipolar disorder. Biol. Psychiatry 64, 803–809. Woo, T.U., Walsh, J.P., Benes, F.M., 2004. Density of glutamic acid decarboxylase 67 messenger RNA-containing neurons that express the N-methyl-d-aspartate receptor subunit NR2A in the anterior cingulate cortex in schizophrenia and bipolar disorder. Arch. Gen. Psychiatry 61, 649–657. Woods, B.T., 1998. Is schizophrenia a progressive neurodevelopmental disorder? Toward a unitary pathogenetic mechanism. Am. J. Psychiatry 155, 1661–1670. Yatham, L.N., Lyoo, I.K., Liddle, P., Renshaw, P.F., Wan, D., Lam, R.W., Hwang, J., 2007. A magnetic resonance imaging study of mood stabilizer- and neuroleptic-naive first-episode mania. Bipolar Disord. 9, 693–697. Zabala, A., Rapado, M., Arango, C., Robles, O., de la Serna, E., Gonzalez, C., RodriguezSanchez, J.M., Andres, P., Mayoral, M., Bombin, I., 2010. Neuropsychological functioning in early-onset first-episode psychosis: comparison of diagnostic subgroups. Eur. Arch. Psychiatry Clin. Neurosci. 260, 225–233. Zanelli, J., Reichenberg, A., Morgan, K., Fearon, P., Kravariti, E., Dazzan, P., Morgan, C., Zanelli, C., Demjaha, A., Jones, P.B., Doody, G.A., Kapur, S., Murray, R.M., 2010. Specific and generalized neuropsychological deficits: a comparison of patients with various first-episode psychosis presentations. Am. J. Psychiatry 167, 78–85. Zimmerman, M.E., DelBello, M.P., Getz, G.E., Shear, P.K., Strakowski, S.M., 2006. Anterior cingulate subregion volumes and executive function in bipolar disorder. Bipolar Disord. 8, 281–288. Zubieta, J.K., Huguelet, P., O’Neil, R.L., Giordani, B.J., 2001. Cognitive function in euthymic bipolar I disorder. Psychiatry Res. 102, 9–20.