Specificity and Continuity of Schizophrenia and Bipolar Disorder: Relation to 
Biomarkers

Abstract

Schizophrenia and bipolar disorder overlap considerably in terms of symptoms, familial patterns, risk genes, outcome, and treatment response. This article provides an overview of the specificity and continuity of schizophrenia and mood disorders on the basis of biomarkers, such as genes, molecules, cells, circuits, physiology and clinical phenomenology. Overall, the discussions herein provided support for the view that schizophrenia, schizoaffective disorder and bipolar disorder are in the continuum of severity of impairment, with bipolar disorder closer to normality and schizophrenia at the most severe end. This approach is based on the concept that examining biomarkers in several modalities across these diseases from the dimensional perspective would be meaningful. These considerations are expected to help develop new treatments for unmet needs, such as cognitive dysfunction, in psychiatric conditions.

1. Introduction

Schizophrenia and bipolar disorder, operationally defined by clinical features, overlap considerably in terms of symptoms, familial patterns, risk genes, outcome and treatment response [1]. Although Kraepelin differentiated between schizophrenia and bipolar disorder as two forms of psychoses on the basis of the clinical course, he also pointed out that both disorders shared certain symptoms, such as hallucinations, delusions, and mood symptoms [2]. To interpret the overlap of schizophrenia and bipolar disorder, Kasanin [3] introduced a concept of schizoaffective disorder that elicits clinical features of both diseases. Nevertheless, it was sometimes difficult to clearly distinguish between schizophrenia, schizoaffective disorder and bipolar disorder solely by phenomenological features (Fig. 1) [4]. Accordingly, there has been a growing need for a valid diagnostic system based on biological indicators.

Fig. (1).

Typical profile of each disease diagnosed with a combination of psychopathological categories and phenomenological features. Categorical diagnoses of schizophrenia (blue), bipolar disorder (green), and schizoaffective disorder (violet) are accompanied by a patient's quantitative scores (connected by red lines) on five main dimensions of psychopathology [4]. Copyright (2009), with permission from Elsevier. (A higher resolution / colour version of this figure is available in the electronic copy of the article).

Operational diagnostic criteria, such as the Diagnostic and Statistical Manual of Mental Disorders (DSM) and the International Statistical Classification of Diseases and Related Health Problems (ICD), have been developed to improve the reliability of symptom-based diagnosis. However, these diagnostic criteria may not be valid enough because they do not incorporate neuroscience and genetics information, or resolve the issues of coexistence and heterogeneity. Thus, the National Institute of Mental Health (NIMH) has proposed the Research Domain Criteria (RDoC) system, as a new evaluation system to study mental illnesses [5, 6]. RDoC provides a framework that excludes categorical diagnoses, and adopts dimensional evaluation based on genetic, neural and behavioral indicators. The system consists of six research domains and eight analysis units (https://www.nimh.nih.gov/research-priorities/rdoc/index.shtml, 21/2/2019). The research domains include Negative Valence Systems, Positive Valence Systems, Cognitive Systems, Systems for Social Processes, Arousal and Regulatory Systems, and Sensorimotor Systems. For each research domain, genes, molecules, cells, circuits, physiology, behavior, self-reports, and paradigm are provided as analysis units (Fig. 2) [7].

Fig. (2).

Schematic diagram of the RDoC framework.

This review article is intended to provide an up-to-date insight into the specificity and continuity of schizophrenia and bipolar disorder based on the concept of RDoC.

2. Specificity and continuity in genes

Genetic research of schizophrenia and bipolar disorder includes quantitative genetics, such as family, twin and adoption studies, while molecular genetics concerns common risk variants and variants of rare chromosomal structures. Results of quantitative genetics show notable similarities across these disorders [8]. Conventional studies show a substantial familial aggregation with sibling relative risks of around 8-10 for schizophrenia [9-11], bipolar disorder [12, 13], and schizoaffective disorder [14]. Twin studies show concordances of around 40-45% in monozygotic and 0-10% in dizygotic twin pairs for schizophrenia [15], bipolar disorder, and schizoaffective disorder [16]. In a meta-analysis [17], first-degree relatives of schizophrenia patients were shown to possess a higher risk of developing bipolar disorder compared to other relatives. These findings support the genetic link between schizophrenia and bipolar disorder.

Molecular genetic research includes large-scale genome-wide association studies (GWAS), aimed at detecting commonly occurring genetic variants which by themselves have a small effect on the risk for diseases. On the other hand, large chromosomal structural variants, particularly copy number variants (CNV), produce rare but large effects on the risk [8]. GWAS typically deals with more than a million of genetic markers residing in each chromosome, with sample sizes (cases and controls combined) of tens of thousands in recent years [18, 19].

Genetic markers include single-nucleotide polymorphisms (SNPs) that are used to determine whether one of the variants occurs more frequently than expected in affected cases compared with control subjects. An association indicates the presence of a causal genetic variant nearby or, less commonly, i.e., the genetic marker variant itself may have a causal effect [8]. In a review [20] that summarized major GWAS findings for schizophrenia, bipolar disorder, and both disorders combined, the associations indicate a small effect on risk (ORs around 1.1), consistent with a partial overlap of genetic influences from commonly occurring variants on the two disorders (Table 1). These findings show that both diseases share a similar genetic sensitivity.

Table 1.

Genome-wide association study (GWAS) findings for schizophrenia and bipolar disorder from the review by Sullivan et al. [20]. Based on studies with large samples (minimum of around 10000 cases and 10000 controls) and SNP markers showing associations at genome-wide level of statistical significance (p<5×10^-8). Odds Ratio; OR. Copyright (2012), with permission from Springer Nature.

Phenotype	Chromosome where Marker is Located	Nearest Gene	OR
Schizophrenia	1	MIR137	1.12
	2	VRK2	1.09
	2	ZNF804A	1.10
	2	PCGEM1	1.20
	6	MHC	1.22
	8	MMP16	1.10
	8	CSMD1	1.11
	8	LSM1	1.19
	10	CNNM2	1.10
	10	NT5C2	1.15
	11	AMBRA1	1.25
	11	NRGN	1.12
	18	CCDC68	1.09
	18	TCF4	1.20
Bipolar disorder	11	ODZ4	1.14
	12	CACNA1C	1.14
	19	NCAN	1.17
Schizophrenia and Bipolar disorder	2	ZNF804A	1.11
	3	ITIH3-ITIH4	1.12
	10	ANK3	1.22
	12	CACNA1C	1.11

3. Issues of neurometabolites

Proton magnetic resonance spectroscopy (¹H -MRS) is a non-invasive technique that detects magnetic resonance signals produced by atomic nuclei located within molecules in living tissues and measures their chemical composition, energy metabolism, neurotransmitter levels, and neuronal integrity in vivo. ¹H -MRS has increasingly been applied to characterize tissue-based chemical or metabolic abnormalities in psychiatric disorders [21]. This is done by evaluating concentrations of N-acetylaspartate (NAA), creatine (Cr), choline (Cho) and related chemicals [21]. NAA is a metabolite thought to reflect neuronal integrity, and is present exclusively in the brain. Cr is a marker of phosphate metabolism, while Cho indicates the breakdown of cell membranes and cellular turnover [22]. Abnormalities of neurometabolites in various regions of the brain have been implicated in the pathophysiology of schizophrenia and bipolar disorder. For example, both mental disorders show decreased.

NAA levels in the hippocampus and frontal lobes (gray and white matter) [23, 24]. A decrease in NAA concentrations is thought to reflect neuronal or axonal loss, or mitochondrial dysfunction [25], indicating structural abnormalities on a molecular level in schizophrenia and bipolar disorder. Both disorders also show decreased Cr levels in the dorsolateral prefrontal cortex (DLPFC), hippocampus and basal ganglia [26-29], suggesting alterations in the cellular energy metabolism. Conflicting results have been reported for Cho levels in the basal ganglia, hippocampus and DLPFC of schizophrenia patients [21]. In bipolar disorder, increased, decreased, or unaltered Cho levels have been reported in the DLPFC, hippocampus and anterior cingulate cortex [21]. Results of meta-analysis indicate schizophrenia and bipolar disorder share a decline of NAA concentrations and steady-state transition of Cho and Cr.

NAA levels in the thalamus and frontal lobes of schizophrenia patients are significantly decreased, while it is so in the basal ganglia of patients with bipolar disorder [21]. These observations suggest that the changes of several metabolites in the brain may also represent the notion of specificity and continuity pertinent to some psychiatric illnesses (Table 2).

Table 2.

Changes in concentrations of neurometabolites in schizophrenia and bipolar disorder [21]. N-acetylaspartate; NAA, Creatine; Cr, Choline; Cho, and dorsolateral prefrontal cortex; DLPFC.

Phenotype	Neurometabolites	Concentration Change	Region
Schizophrenia	NAA	↓↓	Thalamus, Frontal lobe
Bipolar disorder	NAA	↓↓	Basal ganglia
Schizophrenia and Bipolar disorder	NAA	↓	Hippocampus, Frontal lobe
	Cr	↓~±0	DLPFC, Basal ganglia
	Cho	±0	DLPFC, Hippocampus

4. Key proteins in postmortem brain regions

Proteins are major targets for many types of medicine to treat psychiatric disorders [30]. In particular, schizophrenia and bipolar disorder have been associated with aberrant blood cytokine levels. For example, Goldsmith et al. [31] performed meta-analysis of blood cytokines in acutely and chronically ill patients with these disorders, and found increased levels of cytokines (interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α)), a cytokine receptor (sIL-2R), and its antagonist (IL-1RA) (Table 3). Overall, there were cross-diagnostic similarities in the direction of alterations in cytokine levels throughout the course of illness, suggesting common underlying immune dysfunctions. The association between peripheral levels of cytokines and C-reactive protein (CRP) and cognition was also reviewed [32], which indicates worse cognitive performance in schizophrenia patients with higher CRP levels. By contrast, better cognitive functioning was associated with higher concentrations of tumor necrosis factor-α (TNF-α) [32].

Table 3.

Abnormal proteins in schizophrenia and bipolar disorders. Brain-derived neurotrophic factor; BDNF, synaptosomal-associated protein-25; SNAP-25, cyclic-AMP response element binding protein; CREB, amyloid beta A4 precursor protein; APP, striatal-enriched protein tyrosine phosphatase; STEP, growth associated protein 43; GAP43, ras-related C3 botulinum toxin substrate1; RAC1, interleukin; IL, tumor necrosis factor-α; TNF-α, acute phase; AP, chronic phase; CP.

Region	Marker	Schizophrenia	Bipolar Disorder
Hippocampus	Reelin	↓↓	↓
	BDNF	↓↓	↓↓
	Complexin1	↓↓	↓↓
	Complexin2	↓	↓
	SNAP-25	-	↓
	Parvalbumin	↓↓	↓
	Glucocorticoid receptor	↓	↓
	Dopamine 5 receptor	-	↑
	Serotonin 2A receptor	↓↓	↓↓
Prefrontal cortex	Reelin	↓↓	↓
	Kainate receptor KA2 subunit	-	↓
	Glucocorticoid receptor	↓	-
	Dopamine D2 receptor	↓	-
	CREB	-	↓
	Homer1	↓	↓
	APP	↓	↓
	STEP	↓	-
Anterior cingulate cortex	Synaptophysin	-	↓↓
	Neuromodulin (GAP43)	-	↓↓
	Complexin2	-	↓
	Calbindin	↓	↓
	CREB	↓	↓
Lateral cerebellum	RAC1	-	↑
Blood	IL-1β	↑ (AP)↑ (CP)	↑ (CP)
	IL-1RA	↑ (AP)	↑ (AP) ↓ (CP)
	sIL-2R	↑ (AP) ↑ (CP)	↑ (AP) ↑ (CP)
	IL-6	↑ (AP) ↑ (CP)	↑ (AP) ↑ (CP)
	TNF-α	↑ (AP) ↑ (CP)	↑ (AP)

Aberrant regulation of synaptic function is thought to play a role in the etiology of schizophrenia and bipolar disorder. Specifically, normal neurotransmitter release is dependent on a complex group of presynaptic proteins that regulate synaptic vesicle docking, membrane fusion and fission, such as synaptophysin, syntaxin, synaptosomal-associated protein-25 (SNAP-25), vesicle-associated membrane protein (VAMP), α-synuclein and dynamin I. In addition, structural and signaling proteins, such as neural cell adhesion molecule (NCAM), maintain the integrity of the synapse [33]. Postmortem brain studies suggest impaired neuroplasticity. Therefore, much attention has been paid to the imbalance of intracellular signaling systems. For example, Ren et al. [34] reported that cyclic-AMP (cAMP) response element binding protein (CREB), its mRNA expression levels, and CRE-DNA binding activity were decreased in the nuclear fraction of the dorsolateral prefrontal cortex (DLPFC) and cingulate gyrus (CG) in postmortem specimens from subjects with bipolar disorder. On the other hand, these intracellular indicators were decreased in the CG, but not DLPFC of subjects with schizophrenia. These results indicate region-specific abnormalities of expression and function of CREB in both disorders.

Fragile X mental retardation protein (FMRP) is an RNA binding protein with 842 target mRNAs in the mammalian brain. Silencing of the fragile X mental retardation 1 (FMR1) gene leads to the loss of expression of FMRP and upregulated metabotropic glutamate receptor 5 (mGluR5) signaling, resulting in multiple physical and cognitive deficits associated with fragile X syndrome (FXS). Reduced FMRP expression has been reported in subjects with schizophrenia and bipolar disorder who do not carry the mutation for FMR1. Specifically, Folsom et al. [35] investigated the expression of four downstream targets of FMRP-mGluR5 signaling, i.e. homer1, amyloid beta A4 precursor protein (APP), ras-related C3 botulinum toxin substrate1 (RAC1), and striatal-enriched protein tyrosine phosphatase (STEP), in the brains of subjects with either disorder. In the frontal cortex, expressions of APP and homer1 were reduced in both disorders, whereas expressions of STEP were reduced only in subjects with schizophrenia. By contrast, expressions of RAC1 in the lateral cerebellum were, increased only in subjects with bipolar disorder. Overall, proteins involved in the FMRP-mGluR5 signaling pathway are altered in both disorders, consistent with the specificity/continuity concept (Table 3) [35-38].

5. Specificity and continuity in brain morphology

Schizophrenia and bipolar disorder exhibit considerable overlaps in terms of morphological brain changes including ventricular enlargement and global reduction in the brain volume [39]. For example, whole-brain voxel-based morphometry (VBM) studies report the reduction of grey matter volumes, mainly in bilateral insula and anterior cingulate cortex, in both mental disorders [40]. On the other hand, generalized grey matter deficits are greater in schizophrenia compared with bipolar disorder [41]. Indeed, schizophrenia patients show smaller grey matter volumes than bipolar disorder patients in fronto-temporal cortex, thalamus, hippocampus and amygdala [41], suggesting that morphological changes are subtler in bipolar disorder. In addition, atrophy of the hippocampus, amygdala and thalamus, brain areas associated with cognitive function [43], is more prominent in schizophrenia as compared with bipolar disorder [41, 42]. These characteristics are consistent with abnormal expressions of proteins and neurometabolites in both diseases (Tables 2 and 3). Especially, morphological changes in the hippocampus and prefrontal cortex are conspicuous in conjunction with the decrease in NAA concentrations, which is thought to reflect neuronal or axonal loss, or mitochondrial dysfunction in these regions [25]. Moreover, the hippocampus and prefrontal cortex constitute key neural circuits responsible for cognitive impairment [54], a clinical manifestation representing specificity and continuity for schizophrenia and mood disorders.

Advances in neuroimaging also support the hypothesis that schizophrenia and bipolar disorder have common changes in a series of functional connectivity. For example, there are some lines of evidence for white matter alterations shared by the two disorders, in contrast to the case for grey matter deficits [44, 45]. Abnormalities of white matter microstructures, identified from diffusion tensor imaging, have been collated in meta-analyses, supporting altered white matter connectivity as one of the shared features [45].

The dysconnectivity hypothesis [46] suggests that both illnesses arise not from the regionally specific focal pathophysiology in the brain, but rather from impaired integration between neuroanatomical regions. For example, results of meta-analyses indicate pervasive reductions in organization among all major brain regions of white matter of patients with schizophrenia, while patients with bipolar disorder elicit decreased organization, specifically in the left limbic and right temporal-parietal white matter [47-49].

In view of the connectivity in the brain, both psychiatric disorders share white matter alterations incorporating prefrontal, cortico-thalamic, and callosal fibers this potentially contributes to aberrant executive and cognitive function, a common feature across the psychosis spectrum albeit to a lesser degree in bipolar disorder (Table 4) [50-53].

Table 4.

Brain morphology changes in grey matter volume and white matter (WM) connectivity between schizophrenia (SZ) and bipolar disorder (BP).

Phenotype	Morphology	Volume/Connectivity Change	Region
Schizophrenia	Grey matter	↓↓	Insula
	Grey matter	↓↓	Anterior cingulate cortex
Schizophrenia and Bipolar disorder	Grey matter	↓↓ (SZ), ↓ (BP)	Fronto-temporal cortex
	Grey matter	↓↓ (SZ), ↓ (BP)	Thalamus
	Grey matter	↓↓ (SZ), ↓ (BP)	Hippocampus
	Grey matter	↓↓ (SZ), ↓ (BP)	Amygdala
Schizophrenia	White matter	↓↓	Fronto-temporal WM
Bipolar disorder	White matter	↓	Left limbic WM
	White matter	↓	Right temporal-parietal WM
Schizophrenia and Bipolar disorder	White matter	↓↓ (SZ), ↓ (BP)	Prefrontal WM
	White matter	↓↓ (SZ), ↓ (BP)	Cortico-thalamic WM
	White matter	↓↓ (SZ), ↓ (BP)	Callosal fiber

In schizophrenia, the hippocampus is supposed to be hyperactive, leading to overdrive in the responsivity of midbrain dopamine (DA) neurons that project to the associative striatum, which is proposed to underlie positive symptoms [54]. Additionally, hyperactivity of the hippocampus may interfere with the function of other circuits. Thus, the hippocampal projection to the prefrontal cortex (PFC) may lead to disruption of activity and rhythmicity of the PFC, leading to cognitive impairment [54]. Moreover, the hippocampal projection to basolateral amygdala (BLA) may interfere with the BLA-limbic cortical control of emotional responses, possibly leading to negative symptoms [54]. In this way, altered hippocampal function potentially disrupts multiple interconnected circuits, and causes major symptoms of schizophrenia and bipolar disorders (Figs. 1 and 3) [54].

Fig. (3).

Circuitry of dopamine system regulation and its disruption in schizophrenia. Hyperactive, dysrhythmic limbic hippocampus potentially disrupts multiple interconnected circuits, and could contribute to all 3 symptom classes of schizophrenia [54]. Hipp: Hippocampus, PFC: Prefrontal cortex, BLA: Basolateral amygdala, VP: Ventral pallidum, DA: Dopamine. Copyright (2019), with permission from Oxford University Press.

6. Specificity and continuity in brain connectivity

The medial prefrontal cortex (MPFC) plays a crucial role in the psychophysiology of schizophrenia and bipolar disorders [55]. The ventral and orbital parts of the MPFC are extensively and reciprocally connected to the limbic circuit and surrounding prefrontal cortical regions [56]. Abnormalities of these neural systems may be responsible for the emotional dysregulation of bipolar disorder [57]. For example, patients with bipolar disorders exhibit increased functional connectivity between MPFC and the amygdala compared with healthy controls [58]. MPFC is also associated with internal, self-referential processing [59], and has been suggested to underlie the impairments in reality monitoring of schizophrenia [60].

The MPFC is a major hub of the default mode network, which is typically more active during the resting state than during the performance on tasks that demand external attention, and thought to mediate internal mental activity [61]. Chai et al. [55] examined functional connectivity between MPFC and other brain regions in schizophrenia and bipolar disorder using resting-state functional magnetic resonance imaging (fMRI). The schizophrenia group did not exhibit any resting-state correlations between the MPFC and the ventral lateral prefrontal cortex (VLPFC) or insula. In contrast, the bipolar disorder group exhibited positive correlations between the MPFC vs. insula and VLPFC. Under the same conditions, the control group exhibited negative correlations between these regions. Moreover, the decoupling of dorsal lateral prefrontal cortex (DLPFC) with MPFC in both disorders was observed, consistent with the impaired executive functioning. In sum, functional connectivity between MPFC and insula/VLPFC may distinguish between bipolar disorder and schizophrenia, whereas both diseases share the decoupling DLPFC from MPFC, which may provide a common pathogenesis.

Recent fMRI studies have shown altered brain dynamic functional connectivity (DFC) in mental disorders. Thus, Du et al. [62] examined DFC across a spectrum of symptomatically-related disorders, including schizophrenia, schizoaffective disorder and bipolar disorder with psychosis. They conducted group information guided independent component analysis to estimate both group-level and subject-specific connectivity states from DFC, using fMRI data from patients and healthy control subjects. Regarding the dominant state, widespread group differences were found in 166 functional connectivity, which mainly involved the thalamus and cerebellum, as well as frontal, temporal, occipital, fusiform, postcentral, cuneus, supramarginal and calcarine cortices. Specifically, 22 functional connectivity associated with the postcentral, frontal, and cerebellar cortices were weakened across health control, bipolar disorder with psychosis, schizoaffective disorder, and schizophrenia groups, while 34 functional connectivity associated with the insular, temporal, frontal, fusiform, lingual, occipital, supramarginal cortices, as well as thalamus and cerebellum, were strengthened across those groups (Fig. 4). The degree of these abnormalities, i.e., hypo-connectivity and hyper-connectivity, was in the ascending order of bipolar disorder with psychosis, schizoaffective disorder, and schizophrenia relative to healthy controls. These results are consistent with those in previous studies that observed more severe gray matter deficits [63] and functional impairments [64] in these disorders. These findings support the view that schizophrenia, schizoaffective disorder and bipolar disorder with psychosis are in a continuum of severity, with bipolar disorder with psychosis closer to normality and schizophrenia at the most severe end.

Fig. (4).

The mean static functional connectivity matrix across subjects and its visualized pattern for health control (HC), bipolar disorder with psychosis (BPP), schizoaffective disorder (SAD) and schizophrenia (SZ) group, respectively [62]. The red and blue lines represent positive and negative connectivity strengths, respectively. Copyright (2017), with permission from John Wiley and Sons. (A higher resolution / colour version of this figure is available in the electronic copy of the article).

7. Specificity and continuity in neurocognition

Neurocognitive impairment has long been recognized as a core feature of schizophrenia [65]. In contrast, the importance of cognitive problems in bipolar disorder has been recognized more recently [68]. Individuals with bipolar disorder also demonstrate persistent and trait-like cognitive deficits during remission, while there are some effects of mood state on cognition with acute manic or depressed patients demonstrating profound cognitive deficits [66]. The impairment is most notable in attention, verbal learning and executive function [67], with performance falling 0.5-1 standard deviation (SD) below average. Moreover, these cognitive deficits significantly contribute to functional disability in both schizophrenia and bipolar disorder [65, 68].

Schizophrenia shows cognitive heterogeneity [69], and generally has four subgroups; one with almost normal and one with profoundly impaired cognitive performance, and two intermediate subgroups [70]. Similarly, bipolar disorder has cognitive heterogeneity, with some subgroups whose cognitive deficits are less severe than those reported in schizophrenia [71]. Accordingly, Burdick et al. conducted a cluster analysis of data from the Measurement and Treatment Research to Improve Cognition in Schizophrenia (MATRICS) cognitive battery test in 136 bipolar disorder patients, and found three distinct subgroups, as follows [72] (Fig. 5):

Fig. (5).

Neurocognitive profiles of bipolar disorder clusters and the schizophrenia sample. The X-axis indicates the MATRICS Consensus Cognitive Battery (MCCB) domains. The Y-axis depicts a Z-scale score with a mean of 0 and a standard deviation of 1. Z scores were computed based upon the healthy control sample. Patients are divided into lines based on scoring for each cognitive domain [72]. Copyright (2014), with permission from Cambridge University Press. (A higher resolution / colour version of this figure is available in the electronic copy of the article).

Cluster 1, global impairment group, presenting diffuse and severe cognitive dysfunction, with performance falling between 1 and 2 SD below the mean of healthy controls (Global Group);

Cluster 2, selective impairment group, presenting modest deficits on specific domains, with performance ranging between normal and -1 SD below average (Selective Group);

Cluster 3, intact group, performing comparably to healthy controls on all domains, with superior performance vs. healthy controls on social cognition (Intact Group).

In this way, these subtypes of bipolar disorder were based on degree and pattern of cognitive decline, with the Global Group demonstrating cognitive deficits comparable to those of schizophrenia [72]. This supports the concept of continuity between bipolar disorder and schizophrenia on the basis of behavioral paradigms.

Conclusion

Information from multiple modalities of measures, herein reviewed, support the notion that schizophrenia and bipolar disorder share several biological substrates responsible for the clinical manifestations. These considerations suggest the utility of dimensional perspectives to develop new therapeutics for unmet needs, such as cognitive dysfunction, which may compensate the limitations of categorical diagnostic classifications for psychiatric disorders.

CONSENT FOR PUBLICATION

Not applicable.

FUNDING

This work was partly supported by Japan Society for the Promotion of Science (JSPS) KAKENHI No. 17K10321, Intramural Research Grant (29-1, 30-1, 30-8) and Young Investigator Encouragement Grant for Neurological and Psychiatric Disorders of National Center of Neurology and Psychiatry (NCNP), and AMED under Grant Numbers 18dk0307069 and 18dk0307081, Japan.

CONFLICT OF INTEREST

The authors declare no conflict of interest, financial or otherwise.