Alterations in anterior cingulate cortex myoinositol and aggression in veterans with suicidal behavior: A proton magnetic resonance spectroscopy study

Psychiatry Research: Neuroimaging 276 (2018) 24–32

Contents lists available at ScienceDirect

Psychiatry Research: Neuroimaging journal homepage: www.elsevier.com/locate/psychresns

Alterations in anterior cingulate cortex myoinositol and aggression in veterans with suicidal behavior: A proton magnetic resonance spectroscopy study

T

⁎

Chandni Shetha,b, , Andrew Prescotc, Elliott Buelerb,d, Jennifer DiMuziob,d, Margaret Legarretab,d, Perry F. Renshawa,b,d, Deborah Yurgelun-Todda,b,d, Erin McGladea,b,d a

Department of Psychiatry, University of Utah School of Medicine, Salt Lake City, UT, USA Diagnostic Neuroimaging, University of Utah, Salt Lake City, UT, USA Department of Radiology, University of Utah School of Medicine, Salt Lake City, UT, USA d George E. Wahlen Department of Veterans Affairs Medical Center, VA VISN 19 Mental Illness Research, Education and Clinical Center (MIRREC), Salt Lake City, UT, USA b c

A R T I C LE I N FO

A B S T R A C T

Keywords: Suicide Veterans Neurobiology Anterior cingulate cortex Endophenotype

Studies investigating the neurochemical changes that correspond with suicidal behavior (SB) have not yielded conclusive results. Suicide correlates such as aggression have been used to explore risk factors for SB. Yet the neurobiological basis for the association between aggression and SB is unclear. Aggression and SB are both prevalent in veterans relative to civilian populations. The current study evaluated the relationship between brain chemistry in the anterior (ACC) and the posterior cingulate cortex (POC), as well as the relationship between aggression and SB in a veteran population using proton magnetic resonance spectroscopy (1H-MRS). Single-voxel MRS data at 3 Tesla (T) were acquired from the ACC and POC voxels using a 2-dimensional J-resolved point spectroscopy sequence and quantified using the ProFit algorithm. Participants also completed a structured diagnostic interview and a clinical battery. Our results showed that the myoinositol (mI)/H2O ratio in the ACC and POC was significantly higher in veterans who reported SB when compared to veterans who did not. The two groups did not differ significantly with regard to other metabolites. Second, verbal aggression and SB measures positively correlated with mI/H2O in the ACC. Finally, verbal aggression mediated the relationship between mI/ H2O in the ACC and SB.

1. Introduction Suicide is the tenth leading cause of death and is a significant public health issue in the United States (CDC, 2015). Suicide is also prevalent among veterans with an average of 18 individuals dying by suicide every day in the United States, which is approximately 18% of suicides in individuals 18 or older (Department of Veterans Affairs, 2009). The high rate of suicide among veterans coupled with limited number of pharmacological treatments highlight the fact that there is an urgent need to investigate the neurobiological underpinnings of suicide in veterans, with the overall goal of identifying objective biomarkers of suicide. These objective biomarkers may help clinicians to accurately predict who is at greatest risk for suicide so that effective targeted interventions can be developed. Suicide attempts (SA) are at one end of a continuum of behaviors commonly referred to as suicide-related behaviors (SB) that include suicidal ideation (SI) and SA (Nock et al.,

⁎

2008). SI, defined as thinking about, considering or planning suicide and SA, defined as a non-fatal, self-directed, potentially injurious behavior with intent to die, are both significant risk factors for suicide (Baca-Garcia et al., 2011; Isometsa and Lonnqvist, 1998). In the current study, veterans with SI and/or SA were classified as the SB+ group and those without SI and/or SA were classified as the SB- group. The concept of suicide endophenotypes has been used to explore the neurobiological basis of suicide (Mathews et al., 2013). Endophenotypes include neurophysiological, biochemical, and neuropsychological constructs, where heritability and stability (state independence) are important considerations for an ideal endophenotype (Gould and Gottesman, 2006). Aggressive behavior is a construct that meets specific endophenotypic criteria and associations have been widely reported between aggression and suicide risk in case-control studies in clinical populations, cohort studies in epidemiological samples, retrospective studies of individuals who have died by suicide and

Corresponding author at: Department of Psychiatry, University of Utah,383 Colorow Dr., Rm 325, Salt Lake City, UT 84108, USA. E-mail address: [email protected] (C. Sheth).

https://doi.org/10.1016/j.pscychresns.2018.04.004 Received 30 November 2017; Received in revised form 21 April 2018; Accepted 23 April 2018 Available online 24 April 2018 0925-4927/ © 2018 Elsevier B.V. All rights reserved.

Psychiatry Research: Neuroimaging 276 (2018) 24–32

C. Sheth et al.

and without SA as compared to healthy controls (Jollant et al., 2016a), although the difference did not survive correction for multiple comparisons. mI signaling may also play a role in aggressive behavior. For example, a genome-wide association study (GWAS) found that a gene coding for the non-tyrosine receptor kinase, Fyn, was significantly associated with anger (Mick et al., 2014). Fyn interacts with inositol signaling pathways to stimulate intracellular calcium release, suggesting that inositol signaling may underlie expression of anger. Furthermore, in another study of medication free youths with severe mood dysregulation characterized by extreme irritability and anger/aggression, mI/Cre levels in the temporal lobe were significantly lower than controls (Dickstein et al., 2008). The neurochemical basis of SB, especially in the ACC and POC, has heretofore not been explored using MRS imaging. Furthermore, the neurochemical correlates underlying the link between aggression and SB need further investigation. The current study sought to characterize the association between SB, aggression and altered neurochemistry in veterans using an MRS approach. Based on prior evidence (Jollant et al., 2016a), we hypothesized that an increase in mI may be evident in the ACC and POC of veterans with SB, and that changes in mI may be related to aggression, an important suicide endophenotype.

case registries (Mann et al., 1999; Rusch et al., 2008). Aggression is often reported as a concern among veterans (Hellmuth et al., 2012; McFall et al., 1999). Furthermore, it has been shown that veterans with higher aggression scores are more likely to demonstrate SI and SA (Flanagan et al., 2014; Goldstein et al., 2012). However, the neurobiological basis underlying the association between aggression and SB is not clear. Neuroimaging tools have been increasingly used to characterize the biological factors underlying vulnerability to suicide. Structural and functional alterations have been reported in several brain regions in individuals at risk for suicide as well as in postmortem studies of people who died by suicide (Desmyter et al., 2013). The anterior cingulate cortex (ACC), which has been implicated in cognitive and emotion processing (Carter et al., 1998), has demonstrated differences in individuals with SB. For example, reduced gray matter density in the rostral ACC measured by voxel-based morphometry was observed in MDD patients at high risk for suicide when compared to non high-risk MDD patients (Wagner et al., 2011). In addition, a meta-analysis showed increased dorsal and rostral ACC activation during emotional tasks and reduced activation in these regions during cognitive tasks to be associated with a history of SB (van Heeringen et al., 2014). In a study of combat-exposed veterans, those with SI showed more engagement of the dorsal ACC during error processing as compared to veterans without SI (Matthews et al., 2012). Furthermore, ACC topdown regulation of aggressive impulses has been extensively studied (Siever, 2008; Sterzer and Stadler, 2009) with high trait aggression associated with decreased activation of the dorsal ACC in response to frustration in a study of healthy males (Pawliczek et al., 2013). There are relatively fewer studies that have investigated changes in POC in association with SB and aggression. Using fMRI, hyper-connectivity from the POC to the medial prefrontal cortex (mPFC) and hypo-connectivity in the opposite direction were demonstrated in schizophrenic patients with high risk of suicide compared to healthy controls (Zhang et al., 2013). With regard to aggression, activation of the POC was associated with increasing revenge stimulus intensity in a study of violent criminal psychopaths when compared to healthy controls (Veit et al., 2010). Collectively these studies suggest that the ACC and POC may be neuroanatomical correlates associated with suicide endophenotypes and an increased risk for suicide. In addition to structural and functional imaging techniques, proton1 magnetic resonance spectroscopy (1H-MRS) provides a non-invasive means to measure in vivo levels of biochemical compounds in brain tissue. 1H-MRS can quantify levels of neurotransmitters such as glutamate, glutamine, gamma amino butyric acid (GABA) as well as molecules involved in membrane and intracellular processes such as total creatine (tCre) that reflects a combined signal of Cre + phosphocreatine (PCr), choline that represents phosphorylcholine + glycerophosphocholine (PC+GPC), N-acetylaspartate (NAA), lactate, and myoinositol (mI) (Bittsansky et al., 2012). MRS studies investigating the neurochemistry of suicide are scarce (Jollant et al., 2016a; Li et al., 2009) and none have been reported on a veteran population. To our knowledge, there have been 2 studies using MRS imaging to investigate the neurochemistry in individuals with SB. A potential candidate in the ACC and POC that may be linked to both aggression and an increased risk of suicide is mI, an osmolyte that is primarily synthesized in glial cells in the brain (Brand et al., 1993). Although normally considered a glial marker, a few recent studies have suggested that mI may also be localized in neuronal cells (Fisher et al., 2002). Previous studies have linked alterations in mI to suicide. For example, alterations in cortical mI levels have been reported in postmortem analysis of the brains of individuals who died by suicide (Shimon et al., 1997). In addition, postmortem analysis of brains of individuals who died by suicide revealed hypertrophy of astrocytes (Torres-Platas et al., 2011), which may be a relevant finding since mI is found in high concentrations in astrocytes. An MRS study found increased mI/Cre in the dorsal prefrontal cortex in depressed adults with

2. Methods 2.1. Participants Eighty-one veterans (16 females) were enrolled in the study. Participants were recruited from a local VA hospital as well as from the community via flyers and word of mouth. The Institutional Review Boards at the University of Utah and the George E. Wahlen Department of Veterans Affairs (VA) Medical Center approved this study. All subjects provided written informed consent as per the IRB and Declaration of Helsinki. Participants were compensated financially for their time. Exclusion criteria included major sensorimotor handicaps, estimated full scale IQ < 80, history of autism, claustrophobia, electroconvulsive therapy, active neurological disease, and any MRI contraindications. 2.2. Procedures Participants completed the Structured Clinical Interview for DSMIV-TR (SCID-IV-TR), a clinician administered, semi-structured interview to determine general adaptive functioning (GAF) as well as current and past mental health symptoms. Participants also completed a clinical battery including the Columbia Suicide Severity Rating Scale (C-SSRS), Hamilton Rating Scale for Anxiety and Depression (HAM-A, HAM-D) (Hamilton, 1959, 1960) and the Buss–Perry Aggression Questionnaire (BPAQ). The CSSRS assesses lifetime presence of SI, plans, intensity of ideation and SA. SA includes an actual attempt, an interrupted attempt, or an aborted/self-interrupted attempt. Veterans who reported SI and/or SA were classified as the SB group. Constructs on the C-SSRS have been found to be acceptable internal consistency as well as convergent, divergent, and predictive validity and predict SA in a 24-week follow-up period (Posner et al., 2011). The BPAQ is a 29-item, four-factor instrument that measures physical aggression, verbal aggression, anger, and hostility, with the scales showing good internal consistency and stability over time (Buss and Perry, 1992). The HAM-A and HAM-D are widely used rating scales to measure the severity of anxiety and depressive symptoms respectively. The HAM-D has shown utility in determining the level of depression before, during, and after treatment (Hamilton, 1960). It is based on the clinician's interview with the patient and probes depressive symptoms such as depressed mood, guilty feelings, SB, sleep disturbances, anxiety, and weight loss. Research has demonstrated a validity coefficient of 0.85 (Reynolds and Mazza, 1998). The HAM-A is a rating scale developed to quantify the severity of anxiety symptomatology, and it is often used in psychotropic 25

Psychiatry Research: Neuroimaging 276 (2018) 24–32

C. Sheth et al.

Fig. 1. Top left: myoinositol structure, top right: axial and mid-sagittal slices extracted from a tissue-segmented 3D MP-RAGE MRI dataset recorded from a single subject. Red rectangle depicts the positioning of the MRS voxel in the ACC and the blue rectangle depicts the positioning of the MRS voxel in the POC. Fitted (top), raw (middle) and residuals (bottom) 2D-J 1H-MRS spectra analyzed using Prior Knowledge Fitting (ProFit). Dashed boxes indicate the 2D spectral regions where the mI protons resonate. The color bars to the right show contouring amplitudes and signal phase. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

FOV = 256 × 256 × 224 mm; 1 mm isotropic resolution) were obtained to facilitate the positioning of an obliqued MRS voxel (25 × 25 × 30 mm3) within predominantly gray matter of the ACC. The voxel was placed midline to primarily cover the dorsal anterior cingulate cortex based on midsagittal T1-weighted images, with the ventral anterior edge of the voxel aligned with the centroid of the genu of the corpus callosum (Fig. 1). The MRS voxel was obliqued along the sagittal plane with its smallest dimension spanning the anterior–posterior axis and the largest dimension in the superior–inferior

drug evaluations (Hamilton, 1959). Finally, participants completed a 1 H-MRS scan at the visit. 2.3. Magnetic resonance spectroscopic imaging: acquisition and analysis 1

H-MRS measurements were performed using a 3.0 Tesla Siemens (Erlangen, Germany) Verio™ whole-body MRI scanner. Three-dimensional, high-resolution, magnetization-prepared, rapid gradient echo (MP-RAGE) MR images (TR/TE/TI = 2000/3.53/1100 ms; 26

Psychiatry Research: Neuroimaging 276 (2018) 24–32

C. Sheth et al.

the False Discovery Rate (FDR) originally described by Benjamini and Hochberg (Benjamini, 1995) to determine significance at the FDR p < 0.05 level. We explored whether verbal aggression mediates the relationship between mI/H2O in the ACC and C-SSRS scores, using the PROCESS procedure in SPSS, which uses bootstrapping, a nonparametric resampling procedure, to estimate the bias-corrected 95% confidence intervals of the indirect effect of each of the 5000 resampled datasets (Hayes, 2013; Preacher and Hayes, 2008). Our analysis used bootstrap percentile confidence intervals to infer the observed significance level of the effects. If the confidence intervals did not include 0, the effect was deemed to be significant (Gardner and Altman, 1986). All analyses were performed in SPSS 20 (IBM, Chicago, IL).

orientation. Within-voxel B0 shimming was achieved using a manufacturer-supplied phase map procedure in combination with interactive manual shimming until a full-width at half-maximum (FWHM) of ≤11 Hz was observed for the real component of the ACC unsuppressed water signal. A PRESS sequence was used to acquire two-dimensional (2D) J-resolved 1H MRS spectra measurements, modified to enable TE stepping: TR/TE range = 2400/31-229 ms; signal averages per TE = 4; deltaTE = 2 ms; 3-pulse WET water suppression. The spectral data were obtained using a maximum-echo sampling scheme whereby the analogue-to-digital converter (ADC) on-time was fixed for all 100 TE steps (Schulte and Boesiger, 2006). Outer-volume suppression (OVS) was achieved using six saturation bands positioned at least 1.5-cm away from the MRS voxel faces and band saturation was achieved using hyperbolic secant adiabatic full passage RF pulses. A three-pulse water elimination through T1-effects (WET (Ogg et al., 1994)) scheme was interleaved with the OVS module for global water suppression. An additional water unsuppressed 2D J-resolved 1H MRS dataset was recorded from each voxel with 2 signal averages recorded for each TE step.

2.5. A dimensional approach: correlations with mI/H2O Beyond group comparisons and the inherent limitations of this approach such as high variability, we analyzed the relationship between SB, aggression and mI/H2O in the combined sample of veterans with and without SB, based on the approach employed by Jollant and colleagues (Jollant et al., 2016a,2016b). Evaluation of the combined sample allowed us to take a dimensional rather than categorical approach, in line with the Research Domain Criteria (RDoC) initiative by the NIMH (Cuthbert and Insel, 2013). The RDoC framework may be particularly useful in understanding suicide risk since SB is not specific to one disorder but is seen across several psychiatric illnesses, including depression, bipolar disorder, anxiety, substance use disorder, schizophrenia, and impulse control disorders (Borges et al., 2010; Nock et al., 2010).

2.3.1. Tissue segmentation To control for within-voxel tissue variability, skull-stripping and brain tissue-type segmentation was applied to all MP-RAGE images using the Brain Extraction Tools (Smith, 2002)and FAST (Zhang et al., 2001) tools provided with the freely‐available FMRIB software library (Smith et al., 2004). MATLAB (TheMathWorks, Natick, MA) was used to extract the 3D volume corresponding to the positioned MRS voxel and calculate within-voxel gray matter (GM), white matter (WM) and cerebrospinal fluid (CSF) tissue content for each subject. The within-voxel GM % was calculated as the ratio to the total brain matter within the voxel, i.e., 100 × GM/(GM+WM).

3. Results 3.1. Clinical and demographic variables

2.3.2. MRS data processing All 2D J-resolved 1H MRS data were quantified using the prior knowledge fitting (ProFit) algorithm without additional line broadening applied to prior spectral fitting (Prescot et al., 2012; Schulte and Boesiger, 2006). Before the 2D fast Fourier transformation (FFT), the raw 2D matrix was zero-filled to 200 points along the indirectly detected J dimension. The ProFit algorithm fits basis spectra from a total of nineteen metabolites to the raw 2D spectral surface without considering the effects of spatial localization (Prescot et al., 2012; Schulte and Boesiger, 2006). The basis set comprised of N-acetylaspartate (NAA), glycerophosphocholine (GPC), phosphorylcholine (PC), alanine (Ala), aspartate (Asp), glucose (Glc), glycine (Gly), lactate (lac), Nacetylaspartylglutamate (NAAG), ascorbic acid (Asc), phosphoethanolamine (PE), taurine (Tau), scyllo-inositol (sI), Cre + PCr, glutamate, glutamine, GABA, mI, and glutathione (GSH). Using identical 2D 1H MRS methodology in 10 healthy volunteers, we previously have reported an inter-subject CV of 11–13% for both ACC and POC mI levels normalized to water (Prescot and Renshaw, 2013). All metabolite levels were expressed as metabolite/water ratios and corrected for the withinvoxel CSF fraction using segmented MRI data. Metabolite/water ratios thus are expressed as institutional units (Iu) and presented as the mean ± standard error mean (SEM).

Clinical and demographic characteristics for each group are included in Table 1. There were no significant differences between veterans with and without SB with regard to age, education and IQ. Males with SB ranged in age from 20 to 54; females with SB: 25–49; males with no SB: 25–53, and females with no SB: 28–39. As expected, the SB + group had higher scores of depression and anxiety on the HAM-D and HAM-A, as compared to SB−. Furthermore, the BPAQ total score as well as scores on each of the subscales (physical aggression, verbal aggression, anger, hostility) were significantly higher in the SB+ group. When age and sex were adjusted for, the difference between the SB+ and SB− groups with regard to the clinical measures continued to be significant, except for the physical aggression subscale on the BPAQ. 3.2. Tissue segmentation Fig. 1 displays tissue-segmented axial and sagittal images extracted Table 1 Demographic and clinical variables.

2.4. Statistical analysis

Sex Age (mean ± S.D.) Education (mean ± S.D.) WASI-IQ HAM-D HAM-A BPAQ-total score BPAQ-physical aggression BPAQ-verbal aggression BPAQ-anger BPAQ-hostility

Group differences in demographic and clinical measures between SB + and SB− groups were evaluated using Student's t-test. Logarithmic transformations were used for MRS measures to test group differences if the measures failed the Shapiro-Wilk's test for normality. We adjusted for the effects of age, sex by including them as covariates in a general linear model. We also adjusted for the presence of depressive symptoms by including current and/or past major depressive disorder (MDD) diagnosis as a covariate. Pearson's and Spearman's tests were used for correlation analyses and we corrected for multiple comparisons using 27

Veterans with SB

Veterans without SB

46 males, 11 females 37.2 ± 9.1 15 ± 2.2 110.8 ± 9.6 9.8 ± 7.5 10.1 ± 8.3 80.8 ± 19.2 26.6 ± 12.3 14.5 ± 4.1 17.7 ± 6.2 22.0 ± 6.8

19 males, 5 females 36.2 ± 9.7 14.7 ± 1.7 114.3 ± 11.1 1.9 ± 2.5 3.0 ± 4.1 63.6 ± 22.3 25.5 ± 17.4 11.6 ± 3.5 11.3 ± 3.8 15.1 ± 6.2

p-value

0.62 0.57 0.16 <0.001** <0.001** <0.001** 0.047* 0.01* <0.001** <0.001**

Psychiatry Research: Neuroimaging 276 (2018) 24–32

C. Sheth et al.

3.4. Correlation between mI/H2O and C-SSRS scores

Table 2 Tissue Fractions (GM and WM) calculated for both groups and brain regions under investigation. Values are expressed as group mean % fraction ± SD. Brain region

Group

ACC

Veterans Veterans Veterans Veterans

POC

GM with SB without SB with SB without SB

69.44 70.22 61.87 61.39

WM ± ± ± ±

5 3 3.9 4.3

30.55 29.78 38.12 38.61

When all study participants were examined as a whole, there was a significant correlation between C-SSRS scores and mI/H2O in the ACC (Spearman's rho = 0.27, p = 0.02). Partial correlations run with sex as a covariate showed that the relationship between mI/H2O in the ACC and C-SSRS scores continued to be significant after controlling for sex (Spearman's rho = 0.25, p = 0.03). However, mI/H2O ratio in the POC did not demonstrate any significant correlation with C-SSRS scores (Spearman's rho = 0.14, p = 0.24). When corrected for multiple comparisons using the FDR method, the relationship between mI/H2O in the ACC and C-SSRS scores continued to be significant (qFDR < 0.05).

p-value ± 5 ± 3 ± 3.9 ± 4.3

0.52 0.64

Table 3 Metabolite concentrations normalized to water in the ACC (mean ± standard error mean) (institutional units). NAA: N-Acetylaspartate, Cre + PCr: creatine + phosphocreatine, GABA: gamma amino butyric acid, Gln: glutamine, Glu: glutamate.

NAA/H2O Cre + PCr/H2O Gpc/H2O GABA/H2O Gln/H2O Glu/H2O mI/H2O

Veterans with SB × 10−4

Veterans without SB × 10−4

p-value

2.86 2.16 0.58 0.37 0.52 2.34 1.54

2.60 2.12 0.61 0.37 0.48 2.34 1.38

0.72 0.76 0.60 0.83 0.86 1.00 0.03

± ± ± ± ± ± ±

0.11 0.08 0.02 0.03 0.03 0.10 0.06

± ± ± ± ± ± ±

0.17 0.13 0.03 0.03 0.05 0.12 0.10

3.5. Aggression and mI/H2O in the ACC The scores on the verbal aggression subscale and anger subscale of the BPAQ showed a significant correlation with mI/H2O in the ACC (Spearman's rho = 0.4, p < 0.01 for verbal aggression and Spearman's rho = 0.23, p = 0.05), when the study participants were analyzed as a whole. Further, the total BPAQ scores and hostility subscale scores showed a trend level significant correlation with mI/H2O in the ACC (Spearman's rho = 0.21, p = 0.08 for total scores and Spearman's rho = 0.22, p = 0.06 for the hostility subscale). In contrast, the physical aggression subscale scores did not show a significant correlation with mI/H2O (Spearman's rho = 0.1, p = 0.4). When sex was included as a covariate, the correlation between verbal aggression scores and mI in the ACC continued to be significant (Spearman's rho = 0.3, p = 0.01). The correlations between the hostility subscale, total score and mI concentration exhibited trend-level significance after controlling for sex (Spearman's rho = 0.21, p = 0.08 for total scores and Spearman's rho = 0.21, p = 0.08 for hostility subscale). The correlation between the anger subscale and mI/H2O ratio in the ACC lost significance after controlling for sex (Spearman's rho = 0.19, p = 0.11), suggesting that sex may explain the relationship between anger and mI in the ACC. The correlation between verbal aggression on the BPAQ and mI/H2O in the ACC survived correction for multiple comparisons (qFDR < 0.05).

from a 3D MP-RAGE dataset recorded from a single HC subject. Table 2 displays the within-voxel GM and WM content for both subject cohorts. We did not find significant differences between the two groups. Further, the CSF content within the ACC and POC also did not differ between the two groups (p=0.42 for ACC and p=0.62 for POC).

3.3. SB and metabolite concentrations Metabolite concentrations normalized to water for each group in the ACC and POC are presented in Tables 3 and 4 respectively. There were no significant group differences in any of the metabolite concentrations; however, as hypothesized, the mI/H2O ratio in the ACC was significantly higher in the SB+ group as compared to the SB– group (p = 0.03). When adjusted for the effects of age and gender, the difference in mI/H2O ratio in the ACC continued to be significant (age or gender as covariate, p = 0.03). The difference in mI/H2O ratio continued to be significant in the ACC after adjusting for MDD diagnosis (p = 0.01). In the POC, mI/H2O was significantly higher in the SB+ group as compared to the SB− group (p = 0.05). The difference continued to be significant when adjusted for gender (p = 0.05), however; when age was added as a covariate, the difference between the two groups only showed a trend towards significance (p = 0.07), suggesting that age may mediate the relationship between mI/H2O in the POC and SB. With MDD diagnosis as covariate, there was a strong trend towards significance (p = 0.06).

3.6. Aggression and mI concentrations in the POC In the POC, verbal aggression scores on the BPAQ were significantly correlated with mI/H2O ratio (Spearman's rho = 0.26, p = 0.03). When sex was included as a covariate, the relationship continued to be significant (Spearman's rho = 0.25, p = 0.03). However, the other subscales and the total BPAQ were not significantly correlated with mI/ H2O ratio in the POC (Spearman's rho = 0.08, p = 0.5 for total scores, Spearman's rho = 0.06, p = 0.62 for physical aggression, Spearman's rho = 0.15, p = 0.2 for anger, and Spearman's rho = 0.14, p = 0.22 for hostility). None of the correlations survived correction for multiple comparisons (qFDR > 0.05).

Table 4 Metabolite concentrations normalized to water in the POC (mean ± standard error mean)(institutional units). NAA: N-Acetylaspartate, Cre + PCr: Creatine + phosphocreatine, GABA: Gamma amino butyric acid, Gln: Glutamine, Glu: Glutamate.

NAA/H2O Cre + PCr/H2O Gpc/H2O GABA/H2O Gln/H2O Glu/H2O mI/H2O

Veterans with SB × 10−4

Veterans without SB × 10−4

p-value

2.93 1.95 0.46 0.35 0.54 2.07 1.37

2.83 ± 0.06 1.88 ± 0.06 0.43 ± 0.01 0.34 ± 0.01 0.55 ± 0.05 2.06 ± 0.06 1.28 ± 0.03

0.38 0.45 0.93 0.95 0.84 0.93 0.05

± ± ± ± ± ± ±

0.06 0.04 0.01 0.01 0.02 0.04 0.03

3.7. Does verbal aggression mediate the relationship between mI/H2O in the ACC and suicide? We next examined whether verbal aggression scores on the BPAQ mediated the relationship between mI/H2O in the ACC and C-SSRS scores since the relationship between mI/H2O in the ACC and verbal aggression, and mI/H2O in the ACC and C-SSRS scores survived correction for multiple comparisons. As shown in Fig. 2, there was a significant indirect effect of mI/H2O in the ACC on C-SSRS scores through verbal aggression, indirect effect (ab)=0.09, 95% CI (0.02,0.22). Verbal aggression accounted for less than half of the effect PM (ratio of indirect to total effect of mI/H2O on C-SSRS scores)=0.36, thus supporting a partial mediation model. 28

Psychiatry Research: Neuroimaging 276 (2018) 24–32

C. Sheth et al.

Fig. 2. Path a is the effect of mI/H2O in the ACC on BP-verbal aggression scores and is significant. Path b is the effect of BP-verbal aggression scores on C-SSRS scores and is significant. Path c is the total effect of mI/H2O in the ACC on C-SSRS scores with BP-verbal aggression not in the model; it is statistically significant. Path c is the direct effect of mI/H2O on C-SSRS scores with verbal aggression in the model. Path c is no longer significant with verbal aggression in the model, indicating that verbal aggression statistically mediates the relationship between mI/H2O in the ACC and C-SSRS.

4. Discussion

increase in mI. Though speculative, the above results suggest that increased mI may be observed in the SB group in brain regions other than the ACC and POC. mI is found at high concentrations in different brain regions and is thought to increase as a result of glial proliferation due to reactive astrocytosis and microgliosis (Ashwal et al., 2004; Brand et al., 1993; Kierans et al., 2014). Relevant to this observation, two recent reports have shown an association between microgliosis and SB (Holmes et al., 2018; Steiner et al., 2008). Specifically, Steiner et al. (2008) showed significant microgliosis in the DLPFC, ACC and mediodorsal thalamus of individuals who died by suicide, using immunostaining for the HLA-DR antigen, which is up-regulated in activated microglia as compared to constitutive levels of expression found in human microglia (Gehrmann et al., 1993; Mattiace et al., 1990; Ulvestad et al., 1994). Moreover, this observation was independent of diagnosis of a psychiatric illness (Steiner et al., 2008). In a study of drug-free adults with major depression in a moderate to severe major depressive episode, Holmes and colleagues measured translocator protein (TSPO), which is up-regulated in activated microglia, using positron emission tomography (PET) and TSPO-specific radioligands. The results showed elevated TSPO availability in the ACC, suggestive of microglial activation, in depressed individuals with SI when compared to depressed patients without SI and healthy controls (Holmes et al., 2018). However, there were no significant differences in TSPO availability between depressed patients without SI and healthy controls, suggesting that microglial activation may be more dependent on SB rather than depression status (Holmes et al., 2018). Further, astrocytic activation in the ACC of

In our veteran-focused study, we show that mI/H2O in the ACC and POC are significantly higher in veterans in the SB+ group when compared to the SB− group. However, the results in the POC did not survive covariation with age. Using a dimensional approach consistent with the goals of the RDoC initiative (Insel et al., 2010), we demonstrate a positive correlation between mI/H2O ratio in the ACC with CSSRS scores and verbal aggression subscale on the BPAQ. Finally, the results imply that verbal aggression may partially mediate the relationship between mI/H2O in the ACC and C-SSRS scores. Together these findings shed light on neurochemical correlates of suicide endophenotypes that may underlie a higher risk of suicide. There are relatively few published MRS studies that have investigated neurochemical underpinnings of SB. Li and colleagues reported lower NAA/tCre ratio in the left hippocampus of suicide attempters as compared to healthy controls (Li et al., 2009). In another recent study of 15 suicide attempters, Jollant and colleagues showed an increase in mI/tCre and choline/tCre in suicide attempters and patient controls (depressed without SA) compared to healthy controls in the dorsal prefrontal cortex. Further, the study also reported a decrease in NAA/tCre in suicide attempters as compared to healthy controls (Jollant et al., 2016a). Collectively, these studies suggest that SB may be associated with impairments in neuronal and membrane integrity as well as increased inflammation. In contrast to the above studies, we did not observe group differences in NAA and choline in the ACC and POC. However, similar to the study by Jollant and colleagues, we observed an

Fig. 3. Correlation of mI/H2O in the ACC and (A) C-SSRS scores (B) verbal aggression scores on the BPAQ in the SB+ and SB− groups. 29

Psychiatry Research: Neuroimaging 276 (2018) 24–32

C. Sheth et al.

in many ways, rather than just SB including the presence of psychiatric diagnoses, which may contribute to the observed between group difference and hence is potentially confounding. Finally, the coefficient of variance (CV) for metabolite ratios to water in the ACC were higher than previously reported (Prescot and Renshaw, 2013), which may reflect the heterogeneity of the population and hence the negative findings for other metabolites should be interpreted with caution. In summary, the current study reports increased mI/H2O in the ACC in veterans in the SB+ group as compared to the SB− group. Further, the results indicate that in this study population verbal aggression mediates the relationship between mI/H2O in the ACC and C-SSRS scores. To the best of our knowledge, this is the first study to demonstrate a role for mI as it relates to aggression and SB.

suicide victims has also been reported (Torres-Platas et al., 2011). Thus, it may be possible that increased mI in the ACC is representative of astrogliosis or microglial activation, both of which have been linked to suicide. To the best of our knowledge, this is the first study to show a positive correlation between mI/H2O ratio in the ACC and C-SSRS scores. Aggression and anger-related behavioral traits have been implicated in the diathesis of SB (Carballo et al., 2008). Neurobiological studies have shown impaired serotonergic function to be linked with both aggression and SB, suggesting that it may be a mechanism underlying the association between aggression and SB (Mann et al., 1999). However, whether there exist alternate neurochemical correlates associated with SB that could also be linked to aggression has thus far not been investigated. In the current study, we observed a positive relationship between verbal aggression and mI/H2O in the ACC, which survived correction for multiple comparisons. Further, verbal aggression partially mediated the relationship between mI/H2O in the ACC and CSSRS scores. To date, most of the evidence for a role of mI in aggression is indirect with a GWAS study reporting that Fyn, a gene that interacts with inositol signaling pathways, was significantly associated with anger (Mick et al., 2014). Aberrant temporal lobe mI/tCre ratios were observed in youth with extreme aggression in an MRS study (Dickstein et al., 2008), suggesting that aberrant mI signaling in the brain may contribute to an aggressive phenotype. There are several candidate mechanisms through which mI may play a role in SB and SB related endophenotypes such as aggression. First, mI is incorporated in neuronal cell membranes as inositol phospholipids where it serves as a key metabolic precursor for Gq-coupled receptors and also plays a crucial role in phosphoinositide (PI) signaling (Harvey et al., 2002). Interestingly, Gq-coupled receptors such as 5HT2A receptors have been implicated in the neurobiology of suicide and aggression (Arora and Meltzer, 1989; Mann et al., 1986; Oquendo et al., 2006; Rosell et al., 2010; Stanley and Mann, 1983). Further, PI signaling deficits have been observed in suicide (Dwivedi et al., 2008; Lo Vasco et al., 2015; Pacheco et al., 1996). Second, elevated mI levels in the frontal cortex have been linked to poor decision-making (Jollant et al., 2016b) and cognitive abilities assessed by the MiniMental State Examination (MMSE) (Lim et al., 2012), which are neuropsychological endophenotypes of suicide (Jollant et al., 2013; Richard-Devantoy et al., 2014). Finally, it is noteworthy that lithium, a drug that has shown promise in reducing suicide risk (Cipriani et al., 2013; Kovacsics et al., 2009) reduces inositol levels in the brain (Moore et al., 1999; O'Donnell et al., 2000; Williams et al., 2002). Together, these data suggest that dysregulated mI signaling may result in disruptions in the inositol second messenger signaling which may be linked to frontally-mediated cognitive ability. There are a number of factors that must be considered when interpreting the study findings. First, it is a cross-sectional study that relied on self-report, which makes the causal role of mI in aggression and SB difficult to assess. Longitudinal studies are needed to determine causality and the direction of a relationship between mI, aggression and SB. Second, it is important to consider the temporal relationship between SB and the timing of the scan. It is possible that neurochemical differences exist among subjects depending on time since they last experienced SB. Third, we combined subjects with SI and SA into one group (SB+). We acknowledge that these two phenotypes may only partly share underlying etiology and neurobiology. Thus, future studies should consider investigating the neurobiological mechanisms underlying SI and SA separately. Fourth, we measured mI in ACC and POC voxels, preventing us from making conclusions about mI or other neurochemical alterations elsewhere in the brain that may be associated with SB. In addition, we acknowledge that there may be some signal contamination from nearby brain regions. Fifth, veterans in the current study were not asked to stop taking prescription medications for ethical reasons. Thus, we cannot rule out the confounding effects of medications on the observed differences. Sixth, the two groups differed

Funding This manuscript is based upon work supported by the Department of Veterans Affairs, but does not necessarily represent the views of the Department of Veterans Affairs or the United States Government. This presentation is based upon work supported with resources and the use of facilities at the VISN 19 MIRECC. This project is funded by the Military Suicide Research Consortium (MSRC), an effort supported by the Office of the Assistant Secretary of Defense for Health Affairs under award no. W81XWH-10-2-0178; and through merit review 5I01CX000253-02. Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.pscychresns.2018.04.004. References Arora, R.C., Meltzer, H.Y., 1989. Serotonergic measures in the brains of suicide victims: 5HT2 binding sites in the frontal cortex of suicide victims and control subjects. Am. J. Psychiatry 146, 730–736. Ashwal, S., Holshouser, B., Tong, K., Serna, T., Osterdock, R., Gross, M., et al., 2004. Proton spectroscopy detected myoinositol in children with traumatic brain injury. Pediatr. Res. 56, 630–638. Baca-Garcia, E., Perez-Rodriguez, M.M., Oquendo, M.A., Keyes, K.M., Hasin, D.S., Grant, B.F., et al., 2011. Estimating risk for suicide attempt: are we asking the right questions? Passive suicidal ideation as a marker for suicidal behavior. J. Affective Disord. 134, 327–332. Benjamini, Y., Hochberg, Y., 1995. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. Royal Stat. Soc 57 (1), 288–300. Bittsansky, M., Vybohova, D., Dobrota, D., 2012. Proton magnetic resonance spectroscopy and its diagnostically important metabolites in the brain. Gen. Physiol. Biophys. 31, 101–112. Borges, G., Nock, M.K., Haro Abad, J.M., Hwang, I., Sampson, N.A., Alonso, J., et al., 2010. Twelve-month prevalence of and risk factors for suicide attempts in the World Health Organization World Mental Health Surveys. J. Clin. Psychiatry 71, 1617–1628. Brand, A., Richter-Landsberg, C., Leibfritz, D., 1993. Multinuclear NMR studies on the energy metabolism of glial and neuronal cells. Dev. Neurosci. 15, 289–298. Buss, A.H., Perry, M., 1992. The aggression questionnaire. J. Personal. Soc. Psychol. 63, 452–459. Carballo, J.J., Akamnonu, C.P., Oquendo, M.A., 2008. Neurobiology of suicidal behavior. An integration of biological and clinical findings. Arch. Suicide Res. 12, 93–110. Carter, C.S., Braver, T.S., Barch, D.M., Botvinick, M.M., Noll, D., Cohen, J.D., 1998. Anterior cingulate cortex, error detection, and the online monitoring of performance. Science ( N.Y.) 280, 747–749. CDC, 2015. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Web-based Injury Statistics Query and Reporting System (WISQARS), Atlanta, Georgia. Cipriani, A., Hawton, K., Stockton, S., Geddes, J.R., 2013. Lithium in the prevention of suicide in mood disorders: updated systematic review and meta-analysis. BMJ (Clin. Res. edn.) 346, f3646. Cuthbert, B.N., Insel, T.R., 2013. Toward the future of psychiatric diagnosis: the seven pillars of RDoC. BMC Med. 11, 126. Department of Veterans Affairs, W.D., 2009. Health Services Research and Development Services. Strategies for Suicide Prevention in Veterans. Desmyter, S., Bijttebier, S., van Heeringen, K., 2013. The role of neuroimaging in our understanding of the suicidal brain. CNS Neurol. Disord. Drug Targets 12, 921–929. Dickstein, D.P., van der Veen, J.W., Knopf, L., Towbin, K.E., Pine, D.S., Leibenluft, E., 2008. Proton magnetic resonance spectroscopy in youth with severe mood

30

Psychiatry Research: Neuroimaging 276 (2018) 24–32

C. Sheth et al.

Mick, E., McGough, J., Deutsch, C.K., Frazier, J.A., Kennedy, D., Goldberg, R.J., 2014. Genome-wide association study of proneness to anger. PLoS One 9, e87257. Moore, G.J., Bebchuk, J.M., Parrish, J.K., Faulk, M.W., Arfken, C.L., Strahl-Bevacqua, J., et al., 1999. Temporal dissociation between lithium-induced changes in frontal lobe myo-inositol and clinical response in manic-depressive illness. Am. J. Psychiatry 156, 1902–1908. Nock, M.K., Borges, G., Bromet, E.J., Cha, C.B., Kessler, R.C., Lee, S., 2008. Suicide and suicidal behavior. Epidemiol. Rev. 30, 133–154. Nock, M.K., Hwang, I., Sampson, N.A., Kessler, R.C., 2010. Mental disorders, comorbidity and suicidal behavior: results from the National Comorbidity Survey Replication. Mol. Psychiatry 15, 868–876. O'Donnell, T., Rotzinger, S., Nakashima, T.T., Hanstock, C.C., Ulrich, M., Silverstone, P.H., 2000. Chronic lithium and sodium valproate both decrease the concentration of myo-inositol and increase the concentration of inositol monophosphates in rat brain. Brain Res. 880, 84–91. Ogg, R.J., Kingsley, P.B., Taylor, J.S., 1994. WET, a T1- and B1-insensitive water-suppression method for in vivo localized 1H NMR spectroscopy. J. Magn. Reson. Ser. B 104, 1–10. Oquendo, M.A., Russo, S.A., Underwood, M.D., Kassir, S.A., Ellis, S.P., Mann, J.J., et al., 2006. Higher postmortem prefrontal 5-HT2A receptor binding correlates with lifetime aggression in suicide. Biol. Psychiatry 59, 235–243. Pacheco, M.A., Stockmeier, C., Meltzer, H.Y., Overholser, J.C., Dilley, G.E., Jope, R.S., 1996. Alterations in phosphoinositide signaling and G-protein levels in depressed suicide brain. Brain Res. 723, 37–45. Pawliczek, C.M., Derntl, B., Kellermann, T., Gur, R.C., Schneider, F., Habel, U., 2013. Anger under control: neural correlates of frustration as a function of trait aggression. PLoS One 8, e78503. Posner, K., Brown, G.K., Stanley, B., Brent, D.A., Yershova, K.V., Oquendo, M.A., et al., 2011. The Columbia-Suicide Severity Rating Scale: initial validity and internal consistency findings from three multisite studies with adolescents and adults. Am. J. Psychiatry 168, 1266–1277. Preacher, K.J., Hayes, A.F., 2008. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav. Res. Methods 40, 879–891. Prescot, A.P., Renshaw, P.F., 2013. Two-dimensional J-resolved proton MR spectroscopy and prior knowledge fitting (ProFit) in the frontal and parietal lobes of healthy volunteers: assessment of metabolite discrimination and general reproducibility. J. Magn. Reson. Imaging. 37, 642–651. Prescot, A.P., Richards, T., Dager, S.R., Choi, C., Renshaw, P.F., 2012. Phase-adjusted echo time (PATE)-averaging 1 H MRS: application for improved glutamine quantification at 2.89 T. NMR Biomed. 25, 1245–1252. Reynolds, W.M., Mazza, J.J., 1998. Reliability and validity of the Reynolds Adolescent Depression Scale with young adolescents. J. Sch. Psychol. 36, 295–312. Richard-Devantoy, S., Berlim, M.T., Jollant, F., 2014. A meta-analysis of neuropsychological markers of vulnerability to suicidal behavior in mood disorders. Psychol. Med. 44, 1663–1673. Rosell, D.R., Thompson, J.L., Slifstein, M., Xu, X., Frankle, W.G., New, A.S., et al., 2010. Increased serotonin 2A receptor availability in the orbitofrontal cortex of physically aggressive personality disordered patients. Biol. Psychiatry 67, 1154–1162. Rusch, N., Spoletini, I., Wilke, M., Martinotti, G., Bria, P., Trequattrini, A., et al., 2008. Inferior frontal white matter volume and suicidality in schizophrenia. Psychiatry Res. 164, 206–214. Schulte, R.F., Boesiger, P., 2006. ProFit: two-dimensional prior-knowledge fitting of Jresolved spectra. NMR Biomed. 19, 255–263. Shimon, H., Agam, G., Belmaker, R.H., Hyde, T.M., Kleinman, J.E., 1997. Reduced frontal cortex inositol levels in postmortem brain of suicide victims and patients with bipolar disorder. Am. J. Psychiatry 154, 1148–1150. Siever, L.J., 2008. Neurobiology of aggression and violence. Am. J. Psychiatry 165, 429–442. Smith, S.M., 2002. Fast robust automated brain extraction. Hum. Brain Mapp. 17, 143–155. Smith, S.M., Jenkinson, M., Woolrich, M.W., Beckmann, C.F., Behrens, T.E., JohansenBerg, H., et al., 2004. Advances in functional and structural MR image analysis and implementation as FSL. NeuroImage 23 (Suppl 1), S208–S219. Stanley, M., Mann, J.J., 1983. Increased serotonin-2 binding sites in frontal cortex of suicide victims. Lancet (London, England) 1, 214–216. Steiner, J., Bielau, H., Brisch, R., Danos, P., Ullrich, O., Mawrin, C., et al., 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. Sterzer, P., Stadler, C., 2009. Neuroimaging of aggressive and violent behaviour in children and adolescents. Front. Behav. Neurosci. 3, 35. Torres-Platas, S.G., Hercher, C., Davoli, M.A., Maussion, G., Labonte, B., Turecki, G., et al., 2011. Astrocytic hypertrophy in anterior cingulate white matter of depressed suicides. Neuropsychopharmacology 36, 2650–2658. Ulvestad, E., Williams, K., Bo, L., Trapp, B., Antel, J., Mork, S., 1994. HLA class II molecules (HLA-DR, -DP, -DQ) on cells in the human CNS studied in situ and in vitro. Immunology 82, 535–541. van Heeringen, K., Bijttebier, S., Desmyter, S., Vervaet, M., Baeken, C., 2014. Is there a neuroanatomical basis of the vulnerability to suicidal behavior? A coordinate-based meta-analysis of structural and functional MRI studies. Front. Hum. Neurosci. 8, 824. Veit, R., Lotze, M., Sewing, S., Missenhardt, H., Gaber, T., Birbaumer, N., 2010. Aberrant social and cerebral responding in a competitive reaction time paradigm in criminal psychopaths. NeuroImage 49, 3365–3372. Wagner, G., Koch, K., Schachtzabel, C., Schultz, C.C., Sauer, H., Schlosser, R.G., 2011. Structural brain alterations in patients with major depressive disorder and high risk

dysregulation. Psychiatry Res. 163, 30–39. Dwivedi, Y., Rizavi, H.S., Teppen, T., Zhang, H., Mondal, A., Roberts, R.C., et al., 2008. Lower phosphoinositide 3-kinase (PI 3-kinase) activity and differential expression levels of selective catalytic and regulatory PI 3-kinase subunit isoforms in prefrontal cortex and hippocampus of suicide subjects. Neuropsychopharmacology 33, 2324–2340. Fisher, S.K., Novak, J.E., Agranoff, B.W., 2002. Inositol and higher inositol phosphates in neural tissues: homeostasis, metabolism and functional significance. J. Neurochem. 82, 736–754. Flanagan, J.C., Teer, A., Beylotte, F.M., Killeen, T.K., Back, S.E., 2014. Correlates of recent and lifetime aggression among veterans with co-occurring PTSD and substance use disorders. Ment. Health Subst. Use Dual Diagn. 7, 315–328. Gardner, M.J., Altman, D.G., 1986. Confidence intervals rather than P values: estimation rather than hypothesis testing. Br. Med. J. (Clin. Res. edn.) 292, 746–750. Gehrmann, J., Banati, R.B., Kreutzberg, G.W., 1993. Microglia in the immune surveillance of the brain: human microglia constitutively express HLA-DR molecules. J. Neuroimmunol. 48, 189–198. Goldstein, G., Luther, J.F., Haas, G.L., 2012. Medical, psychiatric and demographic factors associated with suicidal behavior in homeless veterans. Psychiatry Res. 199, 37–43. Gould, T.D., Gottesman II, , 2006. Psychiatric endophenotypes and the development of valid animal models. Genes Brain Behav. 5, 113–119. Hamilton, M., 1959. The assessment of anxiety states by rating. Br. J. Med. Psychol. 32, 50–55. Hamilton, M., 1960. A rating scale for depression. J. Neurol. Neurosurg. Psychiatry 23, 56–62. Harvey, B.H., Brink, C.B., Seedat, S., Stein, D.J., 2002. Defining the neuromolecular action of myo-inositol: application to obsessive-compulsive disorder. Prog. Neuro-psychopharmacol. Biol. Psychiatry 26, 21–32. Hayes, A.F., 2013. Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach. Guildford Press, New York. Hellmuth, J.C., Stappenbeck, C.A., Hoerster, K.D., Jakupcak, M., 2012. Modeling PTSD symptom clusters, alcohol misuse, anger, and depression as they relate to aggression and suicidality in returning U.S. veterans. J. Trauma. Stress 25, 527–534. Holmes, S.E., Hinz, R., Conen, S., Gregory, C.J., Matthews, J.C., Anton-Rodriguez, J.M., et al., 2018. Elevated translocator protein in anterior cingulate in major depression and a role for inflammation in suicidal thinking: a positron emission tomography study. Biol. Psychiatry 83 (1), 61–69. Insel, T., Cuthbert, B., Garvey, M., Heinssen, R., Pine, D.S., Quinn, K., et al., 2010. Research domain criteria (RDoC): toward a new classification framework for research on mental disorders. Am. J. Psychiatry 167, 748–751. Isometsa, E.T., Lonnqvist, J.K., 1998. Suicide attempts preceding completed suicide. Br. J. Psychiatry 173, 531–535. Jollant, F., Guillaume, S., Jaussent, I., Bechara, A., Courtet, P., 2013. When knowing what to do is not sufficient to make good decisions: deficient use of explicit understanding in remitted patients with histories of suicidal acts. Psychiatry Res. 210, 485–490. Jollant, F., Near, J., Turecki, G., Richard-Devantoy, S., 2016a. Spectroscopy markers of suicidal risk and mental pain in depressed patients. Prog. Neuro-psychopharmacol. Biol. Psychiatry. Jollant, F., Richard-Devantoy, S., Ding, Y., Turecki, G., Bechara, A., Near, J., 2016b. Prefrontal inositol levels and implicit decision-making in healthy individuals and depressed patients. Eur. Neuropsychopharmacol. 26, 1255–1263. Kierans, A.S., Kirov II, , Gonen, O., Haemer, G., Nisenbaum, E., Babb, J.S., et al., 2014. Myoinositol and glutamate complex neurometabolite abnormality after mild traumatic brain injury. Neurology 82, 521–528. Kovacsics, C.E., Gottesman II, , Gould, T.D., 2009. Lithium's antisuicidal efficacy: elucidation of neurobiological targets using endophenotype strategies. Ann. Rev. Pharmacol. Toxicol. 49, 175–198. Li, J., Kuang, W.H., Zou, K., Deng, W., Li, T., Gong, Q.Y., et al., 2009. [A proton magnetic spectroscopy research on hippocampus metabolisms in people with suicide-attempted depressions]. Sichuan da xue xue bao. Yi xue ban = Journal of Sichuan University. Medical science edition 40, 59–62. Lim, T.S., Hong, Y.H., Lee, H.Y., Choi, J.Y., Kim, H.S., Moon, S.Y., 2012. Metabolite investigation in both anterior and posterior cingulate gyri in Alzheimer's disease spectrum using 3-tesla MR spectroscopy. Dementia Geriatr. Cognit. Disord. 33, 149–155. Lo Vasco, V.R., Leopizzi, M., Della Rocca, C., Fais, P., Montisci, M., Cecchetto, G., 2015. Impairment and reorganization of the phosphoinositide-specific phospholipase C enzymes in suicide brains. J. Affective Disord. 174, 324–328. Mann, J.J., Stanley, M., McBride, P.A., McEwen, B.S., 1986. Increased serotonin2 and beta-adrenergic receptor binding in the frontal cortices of suicide victims. Arch. Gen. Psychiatry 43, 954–959. Mann, J.J., Waternaux, C., Haas, G.L., Malone, K.M., 1999. Toward a clinical model of suicidal behavior in psychiatric patients. Am. J. Psychiatry 156, 181–189. Mathews, D.C., Richards, E.M., Niciu, M.J., Ionescu, D.F., Rasimas, J.J., Zarate Jr., C.A., 2013. Neurobiological aspects of suicide and suicide attempts in bipolar disorder. Transl. Neurosci. 4, 203–216. Matthews, S., Spadoni, A., Knox, K., Strigo, I., Simmons, A., 2012. Combat-exposed war veterans at risk for suicide show hyperactivation of prefrontal cortex and anterior cingulate during error processing. Psychosom. Med. 74, 471–475. Mattiace, L.A., Davies, P., Dickson, D.W., 1990. Detection of HLA-DR on microglia in the human brain is a function of both clinical and technical factors. The Am. J. Pathol. 136, 1101–1114. McFall, M., Fontana, A., Raskind, M., Rosenheck, R., 1999. Analysis of violent behavior in Vietnam combat veteran psychiatric inpatients with posttraumatic stress disorder. J. Trauma. Stress 12, 501–517.

31

Psychiatry Research: Neuroimaging 276 (2018) 24–32

C. Sheth et al.

effective connectivity in the posterior cingulate and medial prefrontal cortex between first-episode schizophrenic patients with suicide risk and healthy controls. PLoS One 8, e63477. Zhang, X., Ugurbil, K., Chen, W., 2001. Microstrip RF surface coil design for extremely high-field MRI and spectroscopy. Magn. Reson. Med. 46, 443–450.

for suicide: evidence for a distinct neurobiological entity? NeuroImage 54, 1607–1614. Williams, R.S., Cheng, L., Mudge, A.W., Harwood, A.J., 2002. A common mechanism of action for three mood-stabilizing drugs. Nature 417, 292–295. Zhang, H., Wei, X., Tao, H., Mwansisya, T.E., Pu, W., He, Z., et al., 2013. Opposite

32