Background: Macrophage migration inhibitory factor is a proinflammatory cytokine that has been associated with various psychiatric disorders. MicroRNA-451a can directly target macrophage migration inhibitory factor and downregulate its expression in cells. However, the role of macrophage migration inhibitory factor and microRNA-451a in psychiatric patients treated with psychotherapeutic interventions is unknown. In this study, our aim was to investigate levels of macrophage migration inhibitory factor and its regulating microRNA-451a in patients with depression, anxiety, or stress and adjustment disorders who underwent mindfulness-based therapy or treatment as usual. Methods: A total of 168 patients with psychiatric disorders were included from a randomized controlled trial that compared mindfulness-based therapy with treatment as usual. Plasma levels of macrophage migration inhibitory factor and microRNA- 451a were measured at baseline and after the 8-week follow-up using Luminex assay and qPCR. Results: Macrophage migration inhibitory factor levels decreased significantly in patients posttreatment, whereas microRNA- 451a levels showed a nonsignificant change. Macrophage migration inhibitory factor levels were inversely associated with microRNA-451a expression levels at baseline (β = −0.04, P = .008). The change in macrophage migration inhibitory factor levels (follow-up levels minus baseline levels) was associated with the change in microRNA-451a (follow-up levels minus baseline levels) (β = −0.06,P < .0001). The change in either macrophage migration inhibitory factor or microRNA-451a was not associated with improvement in psychiatric symptoms. Conclusion: We demonstrate that the levels of macrophage migration inhibitory factor decreased after psychotherapeutic interventions in patients with psychiatric disorders. However, this reduction was not associated with an improvement in psychiatric symptoms in response to the treatment. We also found an association between macrophage migration inhibitory factor and its regulating microRNA. However, this association needs to be further examined in future studies. Keywords: macrophage migration inhibitory factor; MiR-451a; depression; anxiety; stress and adjustment disorders Received: August 15, 2017; Revised: December 1, 2017; Accepted: January 10, 2018 © The Author(s) 2018. Published by Oxford University Press on behalf of CINP. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any 513 medium, provided the original work is properly cited. For commercial re-use, please contact email@example.com Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/513/4823232 by Ed 'DeepDyve' Gillespie user on 21 June 2018 514 | International Journal of Neuropsychopharmacology, 2018 Significance Statement Psychotherapeutic interventions such as mindfulness-based therapy or TAU may influence inflammatory responses that con- tribute to psychological health. In the present study, we investigated the role of MIF and its regulating miR-451a and found that MIF was significantly decreased after psychotherapeutic intervention. Another important finding is that MIF was inversely related to plasma miR-451a levels in patients with depression, anxiety, or stress and adjustment disorders. For the first time, it is shown that MIF and miR-451a may have a role in patients with psychiatric disorders. its expression and can lead to decreased cell proliferation, col- Introduction ony formation, cell migration and invasion in vitro, suppress Macrophage migration inhibitory factor (MIF) is one of the first xenograft tumor growth in vivo, and modulate epithelial cell cytokine-like proteins that was discovered more than 50 years survival (Bandres et al., 2009; NLiu et . al., 2013; Graham et al., ago (B. R. Bloom and Bennett, 1966; David, 1966). MIF was named 2015; Tang et al., 2015; G. Liu et al., 2016). Furthermore, miR-451 for its ability to recruit macrophages to sites of inflammation is widely expressed in different cells, such as red blood cells, and prevent their random migration. It is synthesized by T- and white blood cells, platelet samples, and serum/plasma (K. Wang B-lymphocytes, monocytes, macrophages, dendritic cells, neu- et al., 2012). However, the relation between circulating miR-451a trophils, eosinophils, mast cells, and basophils. MIF is widely and MIF is largely unknown. distributed in tissues. High levels of MIF expression are noted in Dysregulation of the immune system and inflammatory the endocrine system, especially in the organs that are involved response has been linked to the pathophysiology of certain psy- in stress responses, for example, hypothalamus, pituitary, and chiatric disorders, such as depression (Irwin and Miller, 2007). adrenal glands (Waeber et al., 1997Calandr ; a and Roger, 2003; For example, the role of MIF in the pathobiology of depression Fingerle-Rowson et al., 2003). MIF has pleiotropic effects on has been investigated in different ways (J. Bloom and Al-Abed, inflammation, chemotaxis, cell survival, and proliferation that 2014). MIF is expressed in the brain especially in the areas act as a regulator of innate immune and inflammatory responses concerning behavioral symptoms of depression and anxiety (Mitchell et al., 1999 Calandr ; a and Roger, 2003Bernha ; gen et al., (Conboy et al., 2011). Behaviorally, genetic deletion of MIF has 2007; Savaskan et al., 2012; J. Bloom and Al-Abed, 2014). Inducers resulted in increased anxiety- and depression-like behaviors, for MIF releasing include microbial products such as lipopoly- and the role of MIF in mediating the antidepressant action of saccharides and proinflammatory cytokines. Once released, MIF exercise has been found (Conboy et al., 2011; Moon et al., 2012). acts in an autocrine or paracrine fashion to induce production In blood, the association between MIF and mood disorders may of proinflammatory cytokines (Calandra et al., 1994Calandr ; a, be the opposite. Previous studies have shown that blood MIF 2003; Calandra and Roger, 2003; J. Bloom and Al-Abed, 2014). levels are increased in subjects with mild to moderate depres- MIF also interferes with the antiinflammatory activity of glu- sion, major depression, and other mood disorders (Baugh and cocorticoids at a transcriptional and posttranscriptional level Donnelly, 2003; Hawkley et al., 2007; Musil et al., 2011; J. Bloom (Fingerle-Rowson et al., 2003). Circulating MIF is increased dur - and Al-Abed, 2014). Treatment with antidepressants can reduce ing episodes of inflammation, infection, and stress (Bernhagen the blood levels of MIF (Cattaneo et al., 2013). MIF does not, et al., 1993; Beishuizen et al., 2001; Calandra and Roger, 2003). however, cross the blood brain barrier (Bacher et al., 2002), so MicroRNAs (miRNAs) are a class of small (21-23-nucleotide), the blood level of MIF may be a consequence of different albeit noncoding, single-stranded RNAs that inhibit gene expres- related processes of depression (J. Bloom and Al-Abed, 2014). sion by promoting messenger-RNA (mRNA) degradation or Recently, the role of miRNAs in the development of depres- inhibiting translation (Ambros, 2004). They influence a variety sion and in antidepressant treatment has gained significant of physiological cell processes during development and tissue attention. Studies conducted on postmortem brains from sub- homeostasis by regulating the expression of around 90% of all jects who suffered from depression and subsequently commit- human genes (Miranda et al., 2006). Numerous miRNAs have ted suicide showed alterations in 29 miRNAs compared with been recently detected in several body fluids, including serum, nonpsychiatric control subjects (Smalheiser et al., 2012). Several plasma, and cerebrospinal fluid (Weber et al., 2010). Blood cells studies have reported on the role of miR-124-3p in depression. extensively contact plasma; thus they can be the major con- MiR-124-3p may be used as a biomarker for diagnosis and anti- tributors to extracellular miRNA content in plasma (Pritchard depressant treatment response, because miR-124-3p is a tar - et al., 2012; Turchinovich et al., 2012; Makarova et al., 2016), get of antidepressants and shows similar changes in the blood which can be derived under different conditions (Turchinovich and brain of patients suffering with depression (Dwivedi, 2017; and Burwinkel, 2012). The biological significance of secreted Roy et al., 2017). MiR-135a has also been found to be lower in miRNAs remains to be determined (Turchinovich et al., 2012). the blood of patients with depression and was increased after One explanation is that cells secrete miRNAs after reducing the treatment with antidepressants (Issler et al., 2014). To our know- target mRNA expression, while the unused miRNA is redun- ledge, the role of miR-451a in psychiatric disorders has not been dant (Squadrito et al., 2014). On the other hand, the presence fully investigated (Camkurt et al., 2015 W ; an et al., 2015). One of extracellular miRNAs involves cell-to-cell communication research group found that treatment with antidepressant could via selective miRNA export system (Villarroya-Beltri et al., 2013; reverse the stress-induced change of miR-451 expression in rat Turchinovich et al., 2016; Thomou et al., 2017). Several miRNAs hippocampus (O’Connor et al., 2013). However, it is not known targeting MIF were identified in silico using bioinformatic pro- whether psychotherapeutic interventions can affect plasma lev- grams such as TargetScan and miRSystem (version 20120229; els of MIF or miR-451a in patients with psychiatric disorders. and miRDB, version 4.0) (Lu et al., 2012). MIF was identified as a We therefore aimed to examine whether psychotherapeutic target of miR-451a and this was confirmed by multiple studies. interventions (mindfulness therapy or treatment as usual [TAU], For example, miR-451a can directly target MIF and downregulate mainly cognitive-behavioral therapy [CBT]) are associated with Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/513/4823232 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Wang et al. | 515 MIF or miR-451a levels in patients with psychiatric disorders patient filled in 3 self-rated questionnaires (above-mentioned (depression, anxiety, and stress and adjustment disorders) who PHQ-9, HADS-A/HADS-D, and MADRS-S) at baseline and after 8 have been treated in primary health care. Mindfulness-based weeks of follow-up. The patients received antidepressants and therapies and CBT are effective ways of treating depressive dis- tranquilizers (pharmacotherapy) if deemed necessary. Blood orders and are associated with reductions in proinflammatory samples were collected at the same time as the assessment of processes according to previous research (Steptoe et al., 2007; self-rated symptoms before and after treatment. Patients with Irwin et al., 2015; Sundquist et al., 2015; Black and Slavich, 2016; missing clinical information or poor-quality plasma samples, Walsh et al., 2016; Memon et al., 2017). for example, hemolysis samples, were excluded from our study. Our study population was based on a previously published randomized controlled trial (RCT) conducted by our research Plasma Collection group. The study population included patients with depression, Whole blood (6 mL) was collected from each participant in EDTA anxiety, or stress and adjustment disorders from 16 primary tubes. Blood samples were centrifuged at 2000 g for 10 minutes health care centers in Sweden who were treated with mindful- at 4°C, and the plasma was then aliquoted and stored at -80°C ness-based therapy or TAU for 8 weeks (Sundquist et al., 2015). In before further processing. Blood samples were processed and the present analysis, we analyzed MIF and miR-451a collected in the plasma was frozen within 8 hours of collection (Friebe and the RCT at baseline and after the 8-week follow-up. Our overall Volk, 2008). aim was to investigate levels of MIF and its regulating miR-451a in patients with depression, anxiety, or stress and adjustment disorders who underwent mindfulness-based therapy or TAU. Detection of MIF and miRNA in Plasma Samples We first investigated the potential changes in plasma levels of MIF and miR-451a after 8 weeks of mindfulness-based therapies Plasma MIF levels were determined using the bead-based multi- or TAU in the patients with depression, anxiety, or stress and ad- plex assay for the Luminex platform (R&D Systems Inc) based on justment disorders. Later we explored the potential association the manufacturer’s instructions. To summarize, 80 µL of plasma between plasma levels of MIF and miR-451a at baseline and be- was diluted 1:2 in the dilution buffer and then incubated with tween changes in levels of MIF and miRNA-451a posttreatment. antibody-coated magnetic bead. Protein levels were measured Finally, we investigated whether the possible changes in MIF using the Bio-Plex suspension array system and data were ana- and miR-451a were associated with the improvement in psychi- lyzed with Bio-Plex Manager software (Version 4). Absolute con- atric symptoms (assessed using MADRS-S) posttreatment. centrations were calculated from a standard curve generated from 8 serially diluted standards provided in the kit. The intra- and inter assay CV were 3.8% and 4.4%, respectively. Duplicate Methods samples were assayed and all results were reported as means. Plasma miRNA measurement was performed as previously Study Subjects and Sample Collection described (X. Wang et al., 2014, 2015). In brief, 50 µL of total RNA The study population included 168 patients (age 21–65 years) was isolated from 200 µL of plasma using the miRNeasy Mini Kit with depression, anxiety, or stress and adjustment disorders. (Qiagen GmbH) according to the manufacturer’s protocol, with All the patients were recruited from the 16 primary health care minor modifications. miRNAs were reverse transcribed using a centers that had participated in a randomized controlled trial Universal cDNA Synthesis kit (Exiqon). Quantitative real-time (RCT) of mindfulness therapy compared with TAU, mainly CBT. PCR (qPCR) was carried out in 384-well plates using the CFX38 A detailed description of the study design is provided in a pre- real-time PCR detection system (Bio-Rad Laboratories). At pre- vious article (Sundquist et al., 2015). Patients were recruited sent, there are different methods to detect hemolysis, includ- between January 4, 2012 and March 22, 2012 at the 16 primary ing low levels, in plasma /serum (Blondal et al., 2013 Kirsc ). ( hner health care centers in the county of Scania (Skåne in Swedish) et al., 2011; Shah et al., 2016). Blondal et al. suggested that the in southern Sweden. ratio (miR-451a/miR-23a-3p) of red blood cell-enriched miR-451a Patients who fulfilled the inclusion criteria were included to miR-23a-3p, the latter not affected by hemolysis, could be in the study. All clinical diagnoses were made by doctors at the used as an indicator of hemolysis. The ratio of miR-451a to miR- 16 primary health care centers, including one or more of the 23a-3p was found to be the most sensitive method that could following ICD-10 psychiatric diagnoses: F32.0, mild depressive detect as little as 0.001% hemolysis in serum. Briefly, miR-451a/ episode; F32.1, moderate depressive episode; F32.9, depressive miR-23a-3p ratios of <5, 5–7, and >7 indicate low, mild, or high episode, unspecified; F33.0, recurrent depressive disorder, cur- risk of hemolysis. Both miR-451a and miR-23a-3p were meas- rent episode mild; F33.1, recurrent depressive disorder, current ured in all the plasma samples. Samples with a ratio of ≥7 were episode moderate; F41.0, panic disorder; F41.1, generalized anx- excluded in the analysis. Our measurements showed that the Ct iety disorder; F41.2, mixed anxiety and depressive disorder; value for miR-451a was 23.1± 0.8 (mean ± SD) and the Ct value for F41.3, other mixed anxiety disorders; F41.8, other specified anx- miR-23a–3p was 26.3 ± 1.4 in all the patient samples. This indi- iety disorders; F41.9, anxiety disorder, unspecified; F43.2, adjust- cates that miRNA-451a levels displayed little variation between ment disorders; F43.8, other reactions to severe stress; and F43.9, individuals. There were 6% of samples with a ratio of 5 to 7. For reaction to severe stress, unspecified; age 20 to 64 years; ability these 6% of samples, the Ct value for miR-451a was 22.7 ± 0.7 to speak and read Swedish; and a score of ≥10 on the Patient (mean ± SD) and the Ct value for miR-23a–3p was 28.2 ± 0.9. There Health Questionnaire (PHQ)-9 or ≥7 on the Hospital Anxiety and were no significant differences in miRNA-451a levels between Depression Scale (HADS) or a total score on the Montgomery- these 6% of samples and the whole study population. Therefore, Åsberg Depression Rating Scale (MADRS-S) between 13 and 34 we included these 6% in our analysis. At present, no generally (mild to moderate depression). The exclusion criteria were as accepted standards for normalization of miRNA PCR data have follows: severe personality disorder, risk of suicide, pregnancy, been established. Different methods have been suggested. Some thyroid disease, current psychotherapy of any kind, and par - studies have used spike-in oligonucleotides U6 as normaliza- ticipation in any other psychiatric intervention study. Each tion controls in their study (Zampetaki et al., 2012 Gr; aham et al., Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/513/4823232 by Ed 'DeepDyve' Gillespie user on 21 June 2018 516 | International Journal of Neuropsychopharmacology, 2018 2015). Nonetheless, these synthetic miRNAs are not incorporated Table 1. Characteristics of the Study Population at Baseline (n = 168) in microvesicles or protein lipid complexes. Therefore, extraction Variables Patients (n = 168) efficiency is not accounted for. Alternatively, endogenous miR- NAs that are detectable in all samples show lower variation of Age, y expression levels and have been used as internal controls previ- Mean (SD) 42.2 (11.0) ously (Bye et al., 2013; X. Wang et al., 2016). In the present study, Sex, n (%) the Ct values were normalized according to the ∆Ct method Male 22 (13) with the internal controls miR-425-5p and miR-186-5p. These Female 146 (87) 2 reference miRNAs were selected based on the screening of BMI miRNA expression using a Serum/Plasma Focus microRNA PCR Mean (SD) 26.9 (5.6) Panel (Exiqon) comprising 179 LNA microRNA primer sets in the Smoking status, n (%) selected 11 samples (X.W ang et al., 2015). The geometric mean Yes 22 (13.1) No 143 (85.1) of 2 or more selected reference genes is more accurate than a Alcohol status, n (%) single gene for normalization (Vandesompele et al., 2002). The Standard size drinks per week normalization stability of these two miRNAs was confirmed in ≤1 144 (85.7) our study with geNorm software (Song et al., 2012) and the result >1 19 (11.3) is shown in supplementary Figure 1. These 2 miRNAs have also Antidepressants, n (%) been used as reference miRNAs in previous studies (Bye et al., Yes 55 (33) 2013; X. Wang et al., 2014; Chen et al., 2017). Our measurements No 97 (58) showed that the Ct value for miR-186-5p was 29.4 ± 0.9 (mean ± SD) Tranquilizers, n (%) and the Ct value for miR-425–5p was 28.3 ± 1.0 in all the pa- Yes 24 (14) tient samples. This indicates that those two reference miRNAs No 121 (72) are robust. The relative expression of miR-451a was calculated Baseline MADRS-S with miR-425-5p and miR-186-5p using the following equation: Median score (IQR) 20 (10) ∆Ct = Ct – Ct . The miR-425-5p&miR-186-5p (geometric mean of two miRNAs) miR-451a Baseline MIF (pg/mL) changes of miRNA level after follow-up, the expression of miRNA Median score (IQR) 5398 (4339) at 8 weeks follow-up is compared to expression of the miRNA at Baseline miR-451a (∆Ct) baseline. It was calculated with the formula CT= ∆Ct - Mean (SD) 5.75 (1.30) follow-up ∆Ct . baseline Abbreviations: BMI, body mass index; IQR, interquartile range. 9 patients had missing on BMI. Ethical Considerations 3 (1.8%) had missing on smoking status. 5 (3%) had missing on alcohol status. The study was performed according to the principles of the 16 (9%) had missing on antidepressants. Declaration of Helsinki. It was reviewed and approved by the 23 (14%) had missing on tranquilizers. Ethics Committee of Lund University, prior to its commence- ment, on 5 October 2011 (application no. 2011/491). Written informed consent was obtained from all participants. Table 2. MIF and miR-451a (∆Ct) for Patients at Baseline and 8-Week Follow-up (Mindfulness and TAU) Statistical Analysis Variables Baseline Follow-up Difference P-value Data are presented as mean and SD for age, body mass index MIF (BMI), and baseline miR-451a levels. Median and interquartile Median (IQR) 5398 (4339) 4561 (3131) -718 (2797) <.0001 range (IQR) were used for baseline MADRS-S and MIF, whereas miR-451a sex, smoking, and alcohol status and antidepressant and tran- Mean (SD) 5.75 (1.30) 5.80 (1.32) 0.05 (1.05) .54 quilizer use are presented as numbers and percentages (Table ). 1 To estimate the change between baseline and the 8-week fol- a Difference tested by Wilcoxon sign-rank test. low-up, we used the median and IQR for MIF and the mean and Difference tested by paired t test. SD for miR-451a and tested it with a nonparametric test (Wilcoxon to analyze the changes in MIF/miR-451a and the changes in sign-rank test) for MIF and a paired test for miR-451a ( t Table 2). Linear regression models were used to test the association MADRS-S adjusted for MADRS-S at baseline, both unadjusted and adjusted for the potential confounders (Table). This 5 was between MIF (transformed with the common logarithm because of a highly skewed distribution) and miR-451a at baseline, both also examined by testing the potential difference in changes in MIF (using a T-test) between responders (defined as ≥50 % unadjusted and adjusted for potential confounders (Table) . 3 The following potential confounders were considered, age, sex, decrease in MADRS-S after follow-up) (Trivedi et al., 2009) and nonresponders (supplementary Table 1). The results from lin- BMI, smoking, and alcohol status and pharmacotherapy (anti- depressants and/or tranquilizers). In addition, the associations ear regression models for the associations between MADRS-S and MIF and miR-451a levels at both baseline and follow-up are between changes in MIF (follow-up levels minus baseline lev- els) and changes in miR-451a (follow-up levels minus baseline shown in a scatterplot together with a Pearson correlation coef- ficient (ρ) and a coefficient of determination (R2) (supplemen- levels) were tested using linear regression and adjusted for the potential confounders (Table ). 4 These results are also shown in tary Figure 3). A sensitivity analysis was performed where the period a figure format together with a Pearson correlation coefficient (ρ) and a coefficient of determination (R2) (supplementary Figure 2). between blood collection and isolation of the samples was also considered in the analysis, but it did not change our results To assess the potential association between MIF/miR-451a and treatment response, we used linear regression analysis (data not shown). Hence, it was not included in the final models. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/513/4823232 by Ed 'DeepDyve' Gillespie user on 21 June 2018 ∆∆ Wang et al. | 517 Table 4. Associations between the Change in MIF (Follow-Up Level - Baseline Level) and Change in miR-451a (∆∆Ct = ∆Ct -∆Ct ) follow-up baseline Unadjusted Adjusted b c Variables β P value 95% CI β P value 95% CI miR-451a -0.08 <.0001 -0.10; -0.05 -0.06 <.0001 -0.09; -0.03 MIF was transformed with the common logarithm (log10). Association tested by a linear regression model. Adjusted for age, sex, BMI, smoking and alcohol status, and pharmacotherapy treatment (antidepressants and/or tranquilizers). Table 5. Associations between the Change in MIF (Follow-Up Level - Baseline Level)/miR-451a (∆∆Ct = ∆Ct -∆Ct ) and the follow-up baseline Change in MADRS-S Adjusted for MADRS-S at Baseline Unadjusted Adjusted b c Variables β P value 95% CI β P value 95% CI MIF MADRS-S 2.52 .36 -2.88; 7.93 1.37 .65 -4.56; 7.30 miR-451a MADRS-S 0.64 .21 -0.37; 1.65 0.49 .37 -0.58; 1.57 MIF was transformed with the common logarithm (log10). Association tested by a linear regression model. Adjusted for age, sex, BMI, smoking and alcohol status, and pharmacotherapy treatment (antidepressants and/or tranquilizers). MiR-451a is enriched in erythrocytes and hemolysis may affect the results. We used a well-known method, the ratio of miR-451a to miR-23a-3p, for detecting hemolysis in plasma/serum. As mentioned in the methods section, samples with a ratio of 7 or more were excluded from the analysis. In total, 6% of the sam- ples with a ratio between 5 and 7 were included in the present study. We also performed an additional analysis where these samples were excluded, but this did not change the results (data not shown). Statistical analyses were performed by using IBM SPSS Statistics 23 (IBM) and STATA version 14 (StataCorp LP). Results Patient Characteristics The clinical characteristics of the patients are shown in Table 1. The mean age in the whole group (n= 168) was 42.2 years (SD = 11) and most of these participants were women. The mean BMI was 26.9. Most participants were nonsmokers (85.1%) or low consumers of alcohol. The median scores at baseline indicated mild to moderate symptoms of depression and/or anxiety. After treatment, the median scores decreased, indicating none to mild symptoms (similar results for both HADS-A/HADS-D and PHQ-9, data not shown) (Sundquist et al., 2015). The Effects of Treatment (Mindfulness-Based Therapy or TAU) on Plasma MIF and miR-451a Table 2 shows that the MIF levels decreased significantly after the 8 weeks of treatment (median: 4561, IQR: 3131, pg/mL) com- pared with the baseline levels (median: 5398, IQR: 4339, pg/mL) (P < .0001). By contrast, miR-451a levels showed a nonsignificant change. A random-intercept linear regression model was used to examine the potential difference in effect between mindfulness Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/513/4823232 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Table 3. Associations Between MIF and miR-451a (Ct) and Other Clinical Variables at Baseline (n = 168) Unadjusted /Univariate analysis Adjusted b c Variables β P-value 95% CI β P-value 95% CI miR-451a -0.03 .02 -0.06; -0.005 -0.04 .008 -0.07; -0.01 Age 0.006 .001 0.002; 0.009 Sex (female vs male) -0.04 .46 -0.16; 0.07 BMI 0.003 .35 -0.004; 0.01 Smoking status (yes vs no) 0.12 .03 0.01; 0.24 Alcohol status (>1 vs ≤1) 0.04 .50 -0.08; 0.17 Pharmacotherapy treatment (yes vs no) -0.05 .25 -0.13; 0.03 MIF was transformed with the common logarithm (log10). Association tested by a linear regression model. Adjusted for age, sex, BMI, smoking and alcohol status, and pharmacotherapy treatment (antidepressants and/or tranquilizers). Antidepressants and/or tranquilizers. 518 | International Journal of Neuropsychopharmacology, 2018 and TAU on MIF and miR-451a levels in the patients. This ana- miR-451a post-treatment; (3) no significant associations were lysis showed no significant difference in patients treated with found between the changes in either MIF or miR-451a and mindfulness or TAU (data not shown). improvement of psychiatric symptoms. It is of note that mindfulness-based therapy or TAU in our study reduced MIF levels. MIF is important in the homeosta- The Association Between Plasma MIF and miR-451a sis of the host immune response. MIF has been reported to be in Patients involved in the pathobiology of depression/anxiety (Musil et al., We further investigated the potential role of miR-451a in reg- 2011; J. Bloom and Al-Abed, 2014). Edwards et al. reported that ulating MIF levels. T able 3 shows that MIF levels were signifi- elevated MIF related to depressive symptoms, a smaller corti- cantly associated with age and smoking status, with higher age sol reactivity to acute stress, and lowered morning cortisol val- and smoking (yes vs no) associated with higher levels of MIF ues, indicating that MIF may act as a neuro-immune mediator (β = 0.006, P = .001 and β = 0.12, P = .03). Higher BMI and an alcohol linking depressive symptoms with inflammation and hypothal- consumption of more than one standard size drink were also amic–pituitary–adrenal axis dysregulation (Edwards et al., 2010). positively associated to MIF, although not significant. We per - A previous study also showed that MIF levels were reduced after formed linear regression analysis to assess the potential associ- treatment with antidepressants in patients with depression ation between MIF and miR-451a levels at baseline. Unadjusted (Cattaneo et al., 2013). However, it is not known whether mind- linear regression revealed that MIF levels at baseline were fulness-based therapy or CBT have effect on MIF levels. In the inversely related to miR-451a levels (β = −0.03, P = .02). Results present study, we found that MIF was significantly decreased remained significant after adjusting for age, sex, BMI, smoking, after psychotherapeutic interventions. In line with our finding, and alcohol status and pharmacotherapy (β = −0.04, P = .008). Walsh et al. reported that a 4-week mindfulness-based interven- In addition, we investigated the association between tion can reduce salivary IL-6 and TNF-α in women with depres- changes (follow-up level minus baseline level) in MIF and miR- sive symptomatology (Walsh et al., 2016). Furthermore, Moreira 451a posttreatment. With the linear regression analysis (as et al. found that CBT significantly decreases the serum levels of shown in Table 4), we found that the change in levels of MIF IL-6 and TNF-α in patients with depressive symptoms (Moreira was significantly associated with the change in levels of miR- et al., 2015). Based on the evidence above, mindfulness treat- 451a (β = −0.08, P < .0001), although miR-451a showed a non- ment/CBT may also seem to have an effect on inflammatory significant change before and after treatment (Table ). 2 These immune response in depression, but the underlying mechanism results are also presented in supplementary Figure ρ 2 = ( −0.40 is not well established. and r = 0.16). Adjustment for age, sex, BMI, smoking and alco- There is not sufficient evidence to show that pharmaco- hol status, and pharmacotherapy did not change the results logical antidepression treatment affects the miR-451a expres- (β = −0.06, P < .0001). sion. We found only one study showing that treatment with antidepressants could reverse the stress-induced change of miR- 451a expression in rat hippocampus. However, it is not known The Potential Role of MIF/miRNA-451a in the whether psychotherapeutic interventions can affect plasma Improvement in Psychiatric Symptoms in Response levels of MIF or miR-451a in patients with psychiatric disorders. to the Treatment In the present study, we found a nonsignificant change in miR- Moreover, we performed a linear regression analysis to evaluate 451a after the treatment. We were unable to identify the exact the potential association between the changes in MIF/miR-451a mechanism behind this however, as most miRNAs regulate their and the improvement in psychiatric symptoms in response to targets at cellular levels. We have quantified the circulating the treatment (MADRS-S score). However, there were no signifi- miRNA that may not represent the whole effect of regulation by cant associations between the change in either MIF or miR-451a this miRNA at the cellular level; small changes in the circulat- and the change in MADRS-S (Table 5 ). Similar results were found ing miR-451a may represent large changes in cellular miR-451a for both HADS-A/HADS-D and PHQ-9 (data not shown in tables). levels. This was also investigated by examining the difference in MIF It is well known that miR-451a can directly target MIF and changes between responders (defined as a percentage reduction downregulate its expression in the cells (Bandres et al., 2009; of ≥50 % in the MADRS score after follow-up) and nonrespond- N. Liu et al., 2013). The circulating miR-451a is usually secreted ers, with the same conclusion of no significant association (sup- from their cells of origin (Arroyo et al., 2011 Valadi et ; al., 2007). plementary Table 1). Scatterplots (supplementary Figure 3) were However, the relation between circulating MIF and miR-451a has then used to examine potential associations between MIF/miR- never been reported. In the present study, we found that there 451a and MADRS-S (symptoms) at baseline and follow-up separ - is a significant inverse association between plasma levels of ately. A statistically significant but weak association was found MIF and miR-451 levels in patients with psychiatric disorders. only for MIF and MADRS-S at follow-up (ρ = −0.17; R = 0.03). Our study is the first to evaluate a possible association of MIF with circulating miRNA in patients with psychiatric disorders. Similarly, a recently published study reported another circulat- Discussion ing miRNA (miR-939) regulates multiple proinflammatory genes This is, to our knowledge, the first study to explore the poten- in other disease also (McDonald et al., 2016). We also in this study tial role of plasma MIF and miR-451a in patients with depres- show a significant association between the change in MIF levels sion, anxiety, or stress and adjustment disorders treated with and the change in miR-451a levels in patients with psychiatric mindfulness-based therapy or TAU. Our main findings were disorders after psychotherapeutic interventions. However, the the following: (1) the levels of MIF were significantly decreased biological mechanism behind this relationship still needs to be after 8 weeks of treatment in patients, whereas miR-451a lev- examined. Therefore, in any future studies, it will be of interest to els showed a nonsignificant change; (2) a significant association examine the relationship between MIF and miR-451a at cellular was found between levels of MIF and miR-451a in patients at level in patients with depression, anxiety, and stress and adjust- baseline as well as between changes in MIF and changes in ment disorders (Cattaneo et al., 2016) (McDonald et al., 2016). Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/513/4823232 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Wang et al. | 519 We have previously shown that mindfulness therapy and associated with an improvement in psychiatric symptoms in TAU reduced psychiatric symptoms in patients with depres- response to the treatment. The association between plasma sion, anxiety, and stress and adjustment disorders (Sundquist levels of MIF and miR-451 may relate to the regulatory process et al., 2015). Therefore, we investigated whether the changes in occurring at the cellular level, even though the biological mech- MIF and miR-451a were associated with the improvement in anisms need to be examined in future studies. psychiatric symptoms in response to psychotherapeutic inter - ventions. However, we were unable to detect any significant Supplementary Material associations between response and MIF or miR-451a. In agree- ment with our findings, a recent study reported that a mindful- Supplementary data are available at International Journal of ness-based intervention can reduce salivary cytokines although Neuropsychopharmacology online. the reduction was not related to the improvement in depres- sive symptoms (Walsh et al., 2016). Cattaneo et al also reported Acknowledgments that treatment with antidepressants reduced the levels of MIF, but the reduction was not associated with treatment response This project was supported by the Swedish Research Council (2012– (Cattaneo et al., 2013). Recent evidence shows that the expres- 2378 and 2014–10134) and FORTE to Jan Sundquist, the Swedish sion levels of proinflammatory and inflammatory cytokines in Research Council (2014–2517) and FORTE to Kristina Sundquist, as postmortem brain have a positive relationship with the levels well as ALF funding from Region Skåne awarded to Jan Sundquist in plasma in patients with depression (Pandey et al., 2012 Miller ; and Kristina Sundquist. We would like to thank science editor and Raison, 2016). However, it is important to note that MIF does Patrick Reilly for critical reading of the manuscript as well as to bio- not cross the blood brain barrier (Bacher et al., 2002; J Bloom . medical analyst Hamideh Rastkhani for support in the laboratory. and Al-Abed, 2014). Therefore, the changes in MIF in the plasma may differ from the changes in the brain. Furthermore, multiple Statement of Interest preinflammatory cytokines are involved in the development of depression. It is possible that elevated levels of some of these None. inflammatory cytokines may not be the consequence of the depression but could be a result of brain dysfunction associ- References ated with depression and therefore may not be associated with response to treatment. Therefore, a simple reduction in MIF lev- Ambros V (2004) The functions of animal microRNAs. Nature els may not be enough to improve the depressive symptoms in 431:350–355. response to mindfulness therapy or TAU. Taken together, MIF Arroyo JD, Chevillet JR, Kroh EM, Ruf IK, Pritchard CC, Gibson may not act as an independent predictive biomarker for mind- DF, Mitchell PS, Bennett CF, Pogosova-Agadjanyan EL, fulness therapy or TAU in the circulation. Stirewalt DL, Tait JF, Tewari M (2011) Argonaute2 complexes Our study has several strengths. We provide evidence that carry a population of circulating microRNAs independent psychotherapeutic interventions may reduce MIF levels in of vesicles in human plasma. Proc Natl Acad Sci U S A 108: patients with psychiatric disorders. MIF levels and miR-451a lev- 5003–5008. els at baseline as well as changes in levels posttreatment were Bacher M, Weihe E, Dietzschold B, Meinhardt A, Vedder H, Gemsa significantly associated with each other. These findings suggest D, Bette M (2002) Borna disease virus-induced accumulation that miRNA may interact with the inflammatory process linked of macrophage migration inhibitory factor in rat brain astro- to the psychotherapeutic interventions in patients with psychi- cytes is associated with inhibition of macrophage infiltration. atric disorders. As such, this novel contribution has never been Glia 37:291–306. shown in previous studies. Although our findings are promis- Bandres E, Bitarte N, Arias F, Agorreta J, Fortes P, Agirre X, Zarate ing, several general limitations should be considered. First, we R, Diaz-Gonzalez JA, Ramirez N, Sola JJ, Jimenez P, Rodriguez do not have postmortem samples to further confirm our find- J, Garcia-Foncillas J (2009) MicroRNA-451 regulates macro- ings. Second, the blood samples used for the assays were fro- phage migration inhibitory factor production and prolif- zen between 4 to 8 hours after collection, which may affect our eration of gastrointestinal cancer cells. Clin Cancer Res results. However, according to Friebe et al., the cytokine levels in 15:2281–2290. EDTA tubes are not affected if the samples are prepared within Baugh JA, Donnelly SC (2003) Macrophage migration inhibitory 8 hours at room temperature (Friebe and Volk, 2008). The time factor: a neuroendocrine modulator of chronic inflammation. points for both blood collection and plasma isolation were, how- J Endocrinol 179:15–23. ever, available in our patient samples; the results were adjusted Beishuizen A, Thijs LG, Haanen C, Vermes I (2001) Macrophage for this time point and the results were not affected. Third, we migration inhibitory factor and hypothalamo-pituitary-adre- did not perform separate analyses according to the subtype of nal function during critical illness. J Clin Endocrinol Metab psychiatric disorders. However, overlapping symptoms are rela- 86:2811–2816. tively common among these conditions and the patients were Bernhagen J, Calandra T, Mitchell RA, Martin SB, Tracey KJ, therefore analyzed together. Fourth, our findings cannot be used Voelter W, Manogue KR, Cerami A, Bucala R (1993) MIF is a to guide clinical practice methods at present; it needs to be pituitary-derived cytokine that potentiates lethal endotoxae- confirmed in independent cohorts. Moreover, our study did not mia. Nature 365:756–759. include patients that did not receive any therapy as a control. It Bernhagen J, Krohn R, Lue H, Gregory JL, Zernecke A, Koenen RR, might be possible that the changes in MIF and miR-451a were Dewor M, Georgiev I, Schober A, Leng L, Kooistra T, Fingerle- simply a reflection of variability over time. Rowson G, Ghezzi P, Kleemann R, McColl SR, Bucala R, In conclusion, our results suggest that 8 weeks of psycho- Hickey MJ, Weber C (2007) MIF is a noncognate ligand of CXC therapeutic interventions can reduce the MIF levels in patients chemokine receptors in inflammatory and atherogenic cell with psychiatric disorders. However, this reduction was not recruitment. Nat Med 13:587–596. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/513/4823232 by Ed 'DeepDyve' Gillespie user on 21 June 2018 520 | International Journal of Neuropsychopharmacology, 2018 Black DS, Slavich GM (2016) Mindfulness meditation and the Friebe A, Volk HD (2008) Stability of tumor necrosis factor alpha, immune system: a systematic review of randomized con- interleukin 6, and interleukin 8 in blood samples of patients trolled trials. Ann N Y Acad Sci 1373:13–24. with systemic immune activation. Arch Pathol Lab Med Blondal T, Jensby Nielsen S, Baker A, Andreasen D, Mouritzen P, 132:1802–1806. Wrang Teilum M, Dahlsveen IK (2013) Assessing sample and Graham A, Falcone T, Nothnick WB (2015) The expression of mirna profile quality in serum and plasma or other biofluids. microRNA-451 in human endometriotic lesions is inversely Methods 59:S1–S6. related to that of macrophage migration inhibitory factor Bloom BR, Bennett B (1966) Mechanism of a reaction in vitro asso- (MIF) and regulates MIF expression and modulation of epi- ciated with delayed-type hypersensitivity. Science 153:80–82. thelial cell survival. Hum Reprod 30:642–652. Bloom J, Al-Abed Y (2014) MIF: mood improving/inhibiting fac- Hawkley LC, Bosch JA, Engeland CG, Marucha PT, Cacioppo JT tor? J Neuroinflammation 11:11. (2007) Loneliness, dysphoria, stress and immunity: a role for Bye A, Røsjø H, Aspenes ST, Condorelli G, Omland T, Wisløff U cytokines. Cytokine Stress Immun 67–85. (2013) Circulating microRNAs and aerobic fitness–the HUNT - Irwin MR, Miller AH (2007) Depressive disorders and immu- study. Plos One 8:e57496. nity: 20 years of progress and discovery. Brain Behav Immun Calandra T (2003) Macrophage migration inhibitory factor and 21:374–383. host innate immune responses to microbes. Scand J Infect Irwin MR, Olmstead R, Breen EC, Witarama T, Carrillo C, Sadeghi Dis 35:573–576. N, Arevalo JM, Ma J, Nicassio P, Bootzin R, Cole S (2015) Calandra T, Roger T (2003) Macrophage migration inhibitory Cognitive behavioral therapy and tai chi reverse cellular and factor: a regulator of innate immunity. Nat Rev Immunol genomic markers of inflammation in late-life insomnia: a 3:791–800. randomized controlled trial. Biol Psychiatry 78:721–729. Calandra T, Bernhagen J, Mitchell RA, Bucala R (1994) The Issler O, Haramati S, Paul ED, Maeno H, Navon I, Zwang R, Gil S, macrophage is an important and previously unrecognized Mayberg HS, Dunlop BW, Menke A, Awatramani R, Binder EB, source of macrophage migration inhibitory factor. J Exp Med Deneris ES, Lowry CA, Chen A (2014) MicroRNA 135 is essen- 179:1895–1902. tial for chronic stress resiliency, antidepressant efficacy, and Camkurt MA, Acar Ş, Coşkun S, Güneş M, Güneş S, Yılmaz MF, intact serotonergic activity. Neuron 83:344–360. Görür A, Tamer L (2015) Comparison of plasma microRNA Kirschner MB, Kao SC, Edelman JJ, Armstrong NJ, Vallely MP, van levels in drug naive, first episode depressed patients and Zandwijk N, Reid G (2011) Haemolysis during sample prepar - healthy controls. J Psychiatr Res 69:67–71. ation alters microRNA content of plasma. Plos One 6:e24145. Cattaneo A, Gennarelli M, Uher R, Breen G, Farmer A, Aitchison Liu G, Xu Z, Hao D (2016) MicroRNA-451 inhibits neuroblastoma KJ, Craig IW, Anacker C, Zunsztain PA, McGuffin P, Pariante proliferation, invasion and migration by targeting macro- CM (2013) Candidate genes expression profile associated with phage migration inhibitory factor. Mol Med Rep 13:2253–2260. antidepressants response in the GENDEP study: differentiat- Liu N, Jiang N, Guo R, Jiang W, He QM, Xu YF, Li YQ, Tang LL, Mao ing between baseline ‘predictors’ and longitudinal ‘targets’. YP, Sun Y, Ma J (2013) Mir-451 inhibits cell growth and inva- Neuropsychopharmacology 38:377–385. sion by targeting MIF and is associated with survival in naso- Cattaneo A, Ferrari C, Uher R, Bocchio-Chiavetto L, Riva MA, pharyngeal carcinoma. Mol Cancer 12:123. Pariante CM (2016) Absolute measurements of macrophage Lu TP, Lee CY, Tsai MH, Chiu YC, Hsiao CK, Lai LC, Chuang EY (2012) migration inhibitory factor and interleukin-1-beta mRNA MiRSystem: an integrated system for characterizing enriched levels accurately predict treatment response in depressed functions and pathways of microRNA targets. Plos One 7:e42390. patients. Int J Neuropsychopharmacol 19. Makarova JA, Shkurnikov MU, Wicklein D, Lange T, Samatov Chen L, Jin Y, Wang L, Sun F, Yang X, Shi M, Zhan C, Shi Y, Wang TR, Turchinovich AA, Tonevitsky AG (2016) Intracellular and Q (2017) Identification of reference genes and mirnas for qrt- extracellular microRNA: an update on localization and bio- PCR in human esophageal squamous cell carcinoma. Med logical role. Prog Histochem Cytochem 51:33–49. Oncol 34:2. McDonald MK, Ramanathan S, Touati A, Zhou Y, Thanawala RU, Conboy L, Varea E, Castro JE, Sakouhi-Ouertatani H, Calandra Alexander GM, Sacan A, Ajit SK (2016) Regulation of proin- T, Lashuel HA, Sandi C (2011) Macrophage migration inhibi- flammatory genes by the circulating microRNA hsa-mir-939. tory factor is critically involved in basal and fluoxetine-stim- Sci Rep 6:30976. ulated adult hippocampal cell proliferation and in anxiety, Memon AA, Sundquist K, Ahmad A, Wang X, Hedelius A, depression, and memory-related behaviors. Mol Psychiatry Sundquist J (2017) Role of IL-8, CRP and epidermal growth 16:533–547. factor in depression and anxiety patients treated with mind- David JR (1966) Delayed hypersensitivity in vitro: its mediation fulness-based therapy or cognitive behavioral therapy in pri- by cell-free substances formed by lymphoid cell-antigen mary health care. Psychiatry Res 254:311–316. interaction. Proc Natl Acad Sci U S A 56:72–77. Miller AH, Raison CL (2016) The role of inflammation in depres- Dwivedi Y (2017) MicroRNA-124: a putative therapeutic tar - sion: from evolutionary imperative to modern treatment tar - get and biomarker for major depression. Expert Opin Ther get. Nat Rev Immunol 16:22–34. Targets 21:653–656. Miranda KC, Huynh T, Tay Y, Ang YS, Tam WL, Thomson AM, Lim Edwards KM, Bosch JA, Engeland CG, Cacioppo JT, Marucha PT B, Rigoutsos I (2006) A pattern-based method for the identi- (2010) Elevated macrophage migration inhibitory factor (MIF) fication of microRNA binding sites and their corresponding is associated with depressive symptoms, blunted cortisol heteroduplexes. Cell 126:1203–1217. reactivity to acute stress, and lowered morning cortisol. Brain Mitchell RA, Metz CN, Peng T, Bucala R (1999) Sustained mitogen- Behav Immun 24:1202–1208. activated protein kinase (MAPK) and cytoplasmic phospho- Fingerle-Rowson G, Koch P, Bikoff R, Lin X, Metz CN, Dhabhar lipase A2 activation by macrophage migration inhibitory FS, Meinhardt A, Bucala R (2003) Regulation of macrophage factor (MIF). Regulatory role in cell proliferation and gluco- migration inhibitory factor expression by glucocorticoids in corticoid action. J Biol Chem 274:18100–18106. vivo. Am J Pathol 162:47–56. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/513/4823232 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Wang et al. | 521 Moon HY, Kim SH, Yang YR, Song P, Yu HS, Park HG, Hwang O, nonresponse in major depressive disorder: impact on func- Lee-Kwon W, Seo JK, Hwang D, Choi JH, Bucala R, Ryu SH, Kim tioning. Int Clin Psychopharmacol 24:133–138. YS, Suh PG (2012) Macrophage migration inhibitory factor Turchinovich A, Burwinkel B (2012) Distinct AGO1 and AGO2 mediates the antidepressant actions of voluntary exercise. associated mirna profiles in human cells and blood plasma. Proc Natl Acad Sci U S A 109:13094–13099. RNA Biol 9:1066–1075. Moreira FP, Cardoso Tde A, Mondin TC, Souza LD, Silva R, Jansen Turchinovich A, Weiz L, Burwinkel B (2012) Extracellular mirnas: K, Oses JP, Wiener CD (2015) The effect of proinflammatory the mystery of their origin and function. Trends Biochem Sci cytokines in cognitive behavioral therapy. J Neuroimmunol 37:460–465. 285:143–146. Turchinovich A, Tonevitsky AG, Burwinkel B (2016) Extracellular Musil R, Schwarz MJ, Riedel M, Dehning S, Cerovecki A, Spellmann mirna: a collision of two paradigms. Trends Biochem Sci I, Arolt V, Müller N (2011) Elevated macrophage migration 41:883–892. inhibitory factor and decreased transforming growth factor- Valadi H, Ekstrom K, Bossios A, Sjostrand M, Lee JJ, Lotvall JO beta levels in major depression–no influence of celecoxib (2007) Exosome-mediated transfer of mRNAs and microRNAs treatment. J Affect Disord 134:217–225. is a novel mechanism of genetic exchange between cells. Nat O’Connor RM, Grenham S, Dinan TG, Cryan JF (2013) MicroRNAs Cell Biol 9: 654–659. as novel antidepressant targets: converging effects of keta- Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, mine and electroconvulsive shock therapy in the rat hippo- De Paepe A, Speleman F (2002) Accurate normalization of campus. Int J Neuropsychopharmacol 16:1885–1892. real-time quantitative RT-PCR data by geometric a-ver Pandey GN, Rizavi HS, Ren X, Fareed J, Hoppensteadt DA, Roberts aging of multiple internal control genes. Genome Biol RC, Conley RR, Dwivedi Y (2012) Proinflammatory cytokines 3:RESEARCH0034. in the prefrontal cortex of teenage suicide victims. J Psychiatr Villarroya-Beltri C, Gutiérrez-Vázquez C, Sánchez-Cabo F, Pérez- Res 46:57–63. Hernández D, Vázquez J, Martin-Cofreces N, Martinez-Herrera Pritchard CC, Kroh E, Wood B, Arroyo JD, Dougherty KJ, Miyaji DJ, Pascual-Montano A, Mittelbrunn M, Sánchez-Madrid F MM, Tait JF, Tewari M (2012) Blood cell origin of circulating (2013) Sumoylated hnrnpa2b1 controls the sorting of mir - microRNAs: a cautionary note for cancer biomarker studies. nas into exosomes through binding to specific motifs. Nat Cancer Prev Res (Phila) 5:492–497. Commun 4:2980. Roy B, Dunbar M, Shelton RC, Dwivedi Y (2017) Identification of Waeber G, Calandra T, Roduit R, Haefliger JA, Bonny C, Thompson microRNA-124-3p as a putative epigenetic signature of major N, Thorens B, Temler E, Meinhardt A, Bacher M, Metz CN, depressive disorder. Neuropsychopharmacology 42:864–875. Nicod P, Bucala R (1997) Insulin secretion is regulated by Savaskan NE, Fingerle-Rowson G, Buchfelder M, Eyüpoglu IY the glucose-dependent production of islet beta cell macro- (2012) Brain miffed by macrophage migration inhibitory fac- phage migration inhibitory factor. Proc Natl Acad Sci U S A tor. Int J Cell Biol 2012:139573. 94:4782–4787. Shah JS, Soon PS, Marsh DJ (2016) Comparison of methodolo- Walsh E, Eisenlohr-Moul T, Baer R (2016) Brief mindfulness gies to detect low levels of hemolysis in serum for accurate training reduces salivary IL-6 and TNF-α in young women assessment of serum microRNAs. Plos One 11:e0153200. with depressive symptomatology. J Consult Clin Psychol Smalheiser NR, Lugli G, Rizavi HS, Torvik VI, Turecki G, Dwivedi Y 84:887–897. (2012) MicroRNA expression is down-regulated and reorgan- Wan Y, Liu Y, Wang X, Wu J, Liu K, Zhou J, Liu L, Zhang C (2015) ized in prefrontal cortex of depressed suicide subjects. Plos Identification of differential microRNAs in cerebrospinal fluid One 7:e33201. and serum of patients with major depressive disorder. Plos Song J, Bai Z, Han W, Zhang J, Meng H, Bi J, Ma X, Han S, Zhang Z One 10:e0121975. (2012) Identification of suitable reference genes for qpcr ana- Wang K, Yuan Y, Cho JH, McClarty S, Baxter D, Galas DJ (2012) lysis of serum microRNA in gastric cancer patients. Dig Dis Comparing the microRNA spectrum between serum and Sci 57:897–904. plasma. Plos One 7:e41561. Squadrito ML, Baer C, Burdet F, Maderna C, Gilfillan GD, Lyle R, Wang X, Sundquist K, Hedelius A, Palmér K, Memon AA, Ibberson M, De Palma M (2014) Endogenous rnas modulate Sundquist J (2015) Circulating microRNA-144-5p is associated microRNA sorting to exosomes and transfer to acceptor cells. with depressive disorders. Clin Epigenetics 7:69. Cell Rep 8:1432–1446. Wang X, Sundquist J, Zöller B, Memon AA, Palmér K, Sundquist Steptoe A, Hamer M, Chida Y (2007) The effects of acute psycho- K, Bennet L (2014) Determination of 14 circulating microRNAs logical stress on circulating inflammatory factors in humans: in swedes and iraqis with and without diabetes mellitus type a review and meta-analysis. Brain Behav Immun 21:901–912. 2. Plos One 9:e86792. Sundquist J, Lilja Å, Palmér K, Memon AA, Wang X, Johansson LM, Wang X, Sundquist K, Elf JL, Strandberg K, Svensson PJ, Sundquist K (2015) Mindfulness group therapy in primary care Hedelius A, Palmer K, Memon AA, Sundquist J, Zöller B patients with depression, anxiety and stress and adjustment dis- (2016) Diagnostic potential of plasma microRNA signatures orders: randomised controlled trial. Br J Psychiatry 206:128–135. in patients with deep-vein thrombosis. Thromb Haemost Tang Y, Wan W, Wang L, Ji S, Zhang J (2015) MicroRNA-451 inhibited 116:328–336. cell proliferation, migration and invasion through regulation of Weber JA, Baxter DH, Zhang S, Huang DY, Huang KH, Lee MJ, MIF in renal cell carcinoma. Int J Clin Exp Pathol 8:15611–15621. Galas DJ, Wang K (2010) The microRNA spectrum in 12 body Thomou T, Mori MA, Dreyfuss JM, Konishi M, Sakaguchi M, fluids. Clin Chem 56:1733–1741. Wolfrum C, Rao TN, Winnay JN, Garcia-Martin R, Grinspoon SK, Zampetaki A, Willeit P, Tilling L, Drozdov I, Prokopi M, Renard JM, Gorden P, Kahn CR (2017) Adipose-derived circulating mirnas Mayr A, Weger S, Schett G, Shah A, Boulanger CM, Willeit J, regulate gene expression in other tissues. Nature 542:450–455. Chowienczyk PJ, Kiechl S, Mayr M (2012) Prospective study on Trivedi MH, Corey-Lisle PK, Guo Z, Lennox RD, Pikalov A, Kim circulating microRNAs and risk of myocardial infarction. J Am E (2009) Remission, response without remission, and Coll Cardiol 60:290–299. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/513/4823232 by Ed 'DeepDyve' Gillespie user on 21 June 2018
International Journal of Neuropsychopharmacology – Oxford University Press
Published: Jan 24, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera