Anda di halaman 1dari 9


Neuropsychiatric Genetics

A Preliminary Analysis of microRNA as Potential

Clinical Biomarker for Schizophrenia
Xin-yang Sun,1,7 Jin Zhang,2 Wei Niu,3 Wei Guo,4 Hong-tao Song,5 Heng-yu Li,6 Hui-min Fan,7
Lin Zhao,8 Ai-fang Zhong,10 Yun-hua Dai,11 Zhong-min Guo,12 Li-yi Zhang,13* Jim Lu,11,12 and
Qiao-li Zhang9

Prevention and Treatment Center for Psychological Diseases, No.102 Hospital of Chinese Peoples Liberation Army,
Changzhou, Jiangsu, P.R. China

School of medicine, Jiangsu University, Zhenjiang, Jiangsu, P.R. China

Department of Rehabilitation, No.102 Hospital of Chinese Peoples Liberation Army, Changzhou, Jiangsu, P.R. China
Administration office, No.102 Hospital of Chinese Peoples Liberation Army, Changzhou, Jiangsu, P.R. China


Department of Psychiatric Medicine, Bengbu Medical College, Bengbu, Anhui, P.R. China

Department of Burn Surgery, Changhai Hospital of Second Military Medical University, Shanghai, P.R. China
Department of Psychology and Psychiatry, Second Military Medical University, Shanghai, P.R. China


Department of Psychiatry, Suzhou Psychiatric Hospital, Suzhou, Jiangsu, P.R. China

Department of Neurology and Psychiatry, Postgraduate school, Xuzhou Medical College, Xuzhou, Jiangsu, P.R. China
Department of Laboratory, No.102 Hospital of Chinese Peoples Liberation Army, Changzhou, Jiangsu, P.R. China


GoPath Diagnostic Laboratory Co.Ltd, No.801, Changzhou, Jiangsu, P.R. China


GoPath Laboratory LLC, Buffalo Grove, Illinois

Prevention and Treatment Center for Psychological Diseases, No.102 Hospital of Chinese Peoples Liberation Army,
Changzhou, Jiangsu, P.R. China


Manuscript Received: 11 August 2014; Manuscript Accepted: 25 November 2014

MicroRNAs (miRNA, miR) have been implicated as promising

blood-based biomarkers for schizophrenia patients. This study
aimed to clinically validate miRNA as potential schizophrenia
biomarkers. Plasma levels of 10 miRNAs were analyzed using
qPCR in a cohort of 61 schizophrenia patients and 62 normal
controls, as well as 25 patients particularly selected for a six-week
antipsychotic treatment course. Positive And Negative Syndrome Scale (PANSS), Global Assessment Scale (GAS) and
Clinical Global Impression (CGI) were administered to assess
the clinical symptoms. The results demonstrated that a panel of
miRNAs consisting of miR-30e, miR-181b, miR-34a, miR-346
and miR-7 had significantly increased expression levels with
significant combined diagnostic value (AUC:0.713; sensitivity:35.5%; specificity:90.2%). In response to pharmacological

How to Cite this Article:

Sun X-y, Zhang J, Niu W, Guo W, Song Ht, Li H-y, Fan H-m, Zhao L, Zhong A-f,
Dai Y-h, Guo Z-m, Zhang L-y, Lu J, Zhang
Q-l. 2015. A preliminary analysis of
microRNA as potential clinical biomarker
for schizophrenia.
Am J Med Genet Part B 9999:19.

Xin-yang Sun, Jin Zhang, and Wei Niu have contributed equally to this paper and agreed to share the first authorship position together.
Conflict of interest: none.

Correspondence to: Dr. Li-yi Zhang, MM, Prevention and Treatment Center for Psychological Diseases, No.102 Hospital of Chinese Peoples
Liberation Army, North Peace Road 55, Changzhou213003, Jiangsu, Peoples Republic of China.

Correspondence to: Jim Lu, MD, GoPath Laboratories LLC, 1351 Barclay Blvd, Buffalo Grove, IL 60089, United States of America.
Article first published online in Wiley Online Library
( 00 Month 2015
DOI 10.1002/ajmg.b.32292

2015 Wiley Periodicals, Inc.

treatment, expression levels of miR-132, miR-181b, miR-432 and
miR-30e were significantly decreased. In addition, the improvement of clinical symptomatology was significantly correlated
with the changes of miR-132, miR-181b, miR-212 and miR-30e
expression levels. Furthermore, the decreases of plasma levels of
miR-132 and miR-432 were significantly greater in high-effect
subgroup than those in low-effect subgroup after six-week
treatment course. We conclude that miR-30e, miR-181b, miR34a, miR-346 and miR-7 combined as a panel are potentially
useful non-invasive biomarkers for schizophrenia diagnosis.
Markers miR-132, miR-181b, miR-30e and miR-432 are potential
indicators for symptomatology improvements, treatment
responses and prognosis for schizophrenia patients.
2015 Wiley Periodicals, Inc.

Key words: schizophrenia; microRNA; biomarker; antipsychotics; qPCR; PANSS; GAS; CGI

Reliable diagnosis and selective choice of more effective therapeutic
regimens remain challenge for clinical management of schizophrenia (SZ) patients [Kawanishi et al., 2000];[Killackey and Yung,
2007];[Schwarz and Bahn, 2008a];[Schwarz and Bahn, 2008b];
[Lakhan and Kramer, 2009]. Currently, diagnoses of SZ patients
largely rely on self-reported experience and abnormalities in behavior, followed by a clinical assessment of psychiatrists. There is no
objective test available for assisting clinical diagnosis of SZ patients.
Furthermore, there is an urgent need for laboratory tests which
guide the selection of more effective antipsychotic drugs for individual SZ patients. Therefore, biomarkers for differential diagnoses
and prediction of responses to antipsychotic therapy will be essential in developing novel strategies for successful treatment of SZ
Altered expressions of miRNAs have been demonstrated to
contribute to the genetic and biological basis of neuropsychiatric
disorders [Abelson et al., 2005; Kim et al., 2007; Schipper et al., 2007;
Feng et al., 2009; Cox et al., 2010] and specific miRNA expression
profiling has been also observed in postmortem gray matter including the prefrontal cortex, parietal and temporal cortices of SZ
patients [Thomson et al., 2004; Perkins et al., 2007; Beveridge et al.,
2008; Mellios et al., 2009]. As exemplified by the parallel changes in
the levels of certain mRNAs between CNS and peripheral blood
[Vawter et al., 2001; Marazziti et al., 2010], it is plausible to suggest
that disease associated changes in miRNA expression might also be
detectable in peripheral tissues such as plasma and mononuclear
leucocytes. Indeed, blood-based miRNA profiling has become an
increasingly important approach to identify clinically applicable
biomarkers for SZ since brain tissue is not readily accessible.
Hansen et al. studied the association between SZ and genetic
variants of miRNA genes dominantly expressed in brain tissue
using peripheral blood samples, and found that rs17578796 and
rs1700 in miR-206 and miR-198 showed nominal significant allelic
association with SZ[Hansen et al., 2007]. Gardiner et al. explored
miRNA expression profiling in peripheral blood from 112 patients
with SZ and 76 non-psychiatric controls, and revealed 83 signifi-


cantly down-regulated miRNAs in the SZ group[Gardiner et al.,
2011]. Similarly, Lai et al. initially analyzed a panel of seven miRNA
signatures (miR-34a, miR-449a, miR-564, miR-548d, miR-572 and
miR-652 up-regulated; miR-432 down-regulated) in the mononuclear leukocytes from a cohort of SZ patients and controls, and also
found that the expression of a subset of these 7 miRNAs correlated
with negative symptoms, neurocognitive dysfunction and mismatch negativity performance of SZ patients[Lai et al., 2011].
More convincingly, miR-137 was recently shown to be highly
associated (P 1.6  10 11) with the disorder in one of the largest
genome wide association studies conducted to date[Consortium,
2011]. Findings from these studies suggested that the circulating
miRNAs were sufficiently stable and detectable biomarkers in
peripheral blood.
In this study, we investigated the potential significance of 10
miRNAs previously reported to be associated with SZ based upon
adequate literature search as plasma biomarkers for differential
diagnosis, therapeutic responses and prognosis for SZ patients. To
the best of our knowledge, this is the first systematic study investigating the effects of antipsychotic treatment on circulating miRNA
expression in SZ patients.

Sixty-one SZ patients fulfilling the criteria as defined by the
Diagnostic and Statistical Manual 4th edition (DSM-IV) were
prospectively recruited from No.102 Hospital of the Chinese
Peoples Liberation Army from July 2012 to May 2013. Clinical
diagnoses of the patients were made by at least two consultant
psychiatrists, and the diagnoses were further confirmed by an
additional experienced clinical psychiatrist. All patients were first
visitors to the clinics and prior to any clinical treatment for SZ, or in
the absence of psychotropic medication within at least three
months. No patient had history of severe medical diseases, other
psychiatric disorders, structural brain disorders, mental retardation, unstable psychiatric features and movement disorders. Also,
patients who had brain injury causing traumatic amnesia longer
than 24 hr and who received blood transfusion within a month or
electroconvulsive therapy within 6 months, were excluded from the
study. In addition, 62 healthy controls without any family history of
major psychiatric disorders (SZ, bipolar disorder and major depression) within the last three generations were recruited. Similarly,
all healthy controls were without any history of blood transfusion or
severe traumatic event within a month. Patients and healthy controls were matched in gender, age and ethnicity. All individuals
recruited in the study were provided with written informed consent. The study was approved by local Institutional Review Boards.

Medication Intervention
To examine effects of clinical medications on plasma miRNA levels
in SZ patients, 25 out of the 61 SZ patients whose PANSS score were
higher than 60 were particularly selected for receiving six-week
medication intervention. Among 25 patients, seven were treated
with olanzapine (initial dosage 5 mg, average dosage 13 mg, and

dosage range 520mg); eight with quetiapine (initial dosage
100 mg, average dosage 550 mg, and dosage range 100800mg);
six with ziprasidone (initial dosage 40 mg, average dosage 135 mg,
and dosage range 40140mg); additional four with risperidone
(initial dosage 2 mg, average dosage 4.7 mg, and dosage range 2
6mg). All four medications were atypical antipsychotics, treating
both positive and negative symptoms, rarely inducing extrapyramidal symptoms under administration of therapeutic dose.

Clinical Data Collection and Scales Assessment

of the Patients
General information (including name, gender, age, ethnicity, education level, occupation, income level, marital status, substance
abuse history, and family history of psychotic diseases) from all
subjects in the SZ group were collected. One attending and two
resident psychiatric doctors participated in evaluating illness severity
and treatment effects, plus recording dosage information(including
medication, dosage, side effects and treatment course) using positive
and negative syndrome scale(PANSS), global assessment scale(GAS)
and clinical global impression(CGI). Likewise, the general information of healthy controls was recorded. Procedures for data collection
were standardized to ensure the accuracy of the collected data. The
total score of PANSS, the grading score of GAS and global improvement (GI) (from CGI scale) were used for statistical analysis.

miRNA Selection
The targeted 10 miRNAs(miR-132, miR-181b, miR-195, miR-212,
miR-30e, miR-346, miR-34a, miR-432, miR-7 and miR-137) in this
study were selected by thoroughly and extensively surveying current
literatures on Pubmed database, Springer database and the University of British Columbia library database. All published studies
about miRNAs in schizophrenia and antipsychotics from year 2000
to May 2013 were searched and analyzed. The search showed more
than 30 articles, which were further classified in terms of article type.
These 10 miRNA were reported to be associated with SZ in more
than three independent original research articles (Table S1).

miRNA Preparation
Whole blood (5 ml) was collected from all subjects (SZ patients and
healthy controls) using EDTA anticoagulant tube. The plasma was
separated by centrifugation and transferred into a fresh RNase/
DNase-free 2 ml microcentrifuge tube, and stored at 80C until
use. Exact 2 ml of plasma were used for miRNA preparation for all
plasma samples using the miRNeasy Serum/Plasma Kit (Qiagen,
CA) according to the manufacturers protocol. The total miRNA
samples were stored at 80C. C. elegans synthetic miR-39 was used
as a spiked-in and normalization control (Qiagen) as suggested by
the manufacturer[Thomson et al., 2004; Kroh et al., 2010; Wang
et al., 2011; Song et al., 2014].

Quantitative Real-Time PCR

For quantitative detection of miRNA by RT-PCR, purified
plasma miRNA was converted to cDNA by reverse transcription

reactions using TaqMan MicroRNA Reverse Transcription Kit
(Applied Biosystems, Inc., Grand Island, NY) and miRNA-specific stem-loop primers were supplied by the TaqMan MicroRNA
Assays(Applied Biosystems, Inc.). The primer sequences of the
tested miRNAsare shown in Table S2. The RT reactions was
mixed according to the manufacturers protocol and performed
in an Applied Biosystems 9700 PCR instrument using the following conditions: 16C for 30 min, 42C for 30 min, 85C for
5 min, hold at 4C. Real-time PCR reactions were performed in a
5-mL reaction mixture containing 2.5 mL TaqMan Universal PCR
Master Mix II(Applied Biosystems, Inc.), 0.25mL miRNA-specific
primer/probe mix(Applied Biosystems, Inc.), and 2.25 mL diluted
RT cDNA product. PCR reactions in a 384 well reaction plate
were run in a 7900HT fast real-time PCR system (Applied
Biosystems, Inc.) using following cycling parameters: 95C for
10 min, followed by 40 cycles of 95C for 15 sec and 60 C for
1 min. Each reaction was performed in triplicate. Realtime PCR
data were collected by SDS software. For analysis of relative levels
of the tested miRNAs, we used C. elegans synthetic miR-39 as a
normalization control since exactly the same amount of plasma
was applied for miRNA preparation for all plasma specimens.
This synthetic miRNA as a normalization was suggested by the
miRNeasy Serum/Plasma Kit(Qiagen) and has been commonly
used as a normalization for analysis miRNA levels in peripheral
blood in previous publications[Kroh et al., 2010; Song et al.,
2014; Thomson et al., 2004; Wang et al., 2011]. The relative levels
of the tested miRNA in plasma were presented as DDCt which
was calculated using the following formula: 2-(CtmiRNA CtmiR)
NA39 [Thomson et al., 2004; Kroh et al., 2010; Wang et al., 2011;
Song et al., 2014].

Statistical Analysis
Wilcoxon rank sum test was used to test the age difference between
SZ and healthy controls, and expression levels of the individual
tested miRNA between different gender(male or female), different
age(above P75 or below P25), different sibling status(with or
without siblings), different residence(urban or rural), different
educational levels(above college or below high school), different
income levels(above 10,000 for annual income or below 10,000 for
annual income), different marital status(married or unmarried)
and different family history(family patients or sporadic patients) in
SZ group. Wilcoxon rank sum test was also used to compare the
expression levels of miRNA between SZ and healthy controls.
Univariate repeated measures analysis of variance and LSD test
were performed to compare the plasma levels of miRNA before and
after medication. KruscalWallis H test and MannWhitney test
were employed to test the changes of miRNA expression levels
among the groups receiving different medication. Spearman correlation test was carried out for testing the correlation of miRNA
expression levels with scale scores (total PANSS score, GAS score
and CGI score). Receiver-operating characteristics (ROC) curves
were established to evaluate the diagnostic performance of miRNAs
for differentiating between SZ and healthy controls. Conditional
logistic regression analyses were then performed to select the most
significant miRNA with the maximum acceptable limit for adding a
variable being 0.1 and the minimum acceptable limit for removing a


variable being 0.15. Mann-Whitney U test was carried out to test the
expression levels changes between different treating effect subgroups. All statistical analyses were carried out using SPSS version
17.0 software (Chicago, IL). P < 0.05 was considered statistically

Clinical Characteristics of the Patients
The demographic information of all the 123 subjects (both the SZ
patients and healthy controls) is shown in Table I. No significant
differences (P > 0.05) were found in age and genders between 61
SZ patients and 62 healthy controls. Among patients from SZ
group, no significant differences (P > 0.05) in miRNA expression
levels were found between genders, sibling status, residential
locations, educational levels, marital statuses and family history.
The 61 SZ patients were further classified into higher and lower
age subgroups according to age percentile. Patients with ages
above 37.5 were classified into the higher age subgroup (P75,
n 16) and those with ages below 19 into the lower age subgroup
(P25, n 13). Statistical analysis revealed that only miR-30e
expression level was significantly elevated in the lower age
subgroup as compared to the higher age subgroup(P < 0.01,
Table S3). In addition, 25 out of all the 61 patients who had
PANSS score more than 60 were selected to undergo six-week
antipsychotics treatment. These 25 patients who underwent
three-week treatment and 18 patients who underwent six-week
treatment with complete experimental and clinical data were
included for the final analysis.

TABLE I. Demographic Variables of the two Groups

Sibling status
Educational level
Junior high and below
Senior high and above
Marital status
Family history

SZ(n 61)

Control(n 62)

27.84 (10.64)





Plasma miRNA Levels in SZ Patients and

Healthy Controls Before Medication
Of the ten candidate miRNAs, miR-137 was excluded for the final
analysis due to its exceedingly high CT value making it impossible to
be detected. Average expression levels of individual tested miRNAs
in plasma samples from SZ and healthy control are shown in
Table S4. Five miRNAs (miR-181b, miR-30e, miR-346, miR-34a
and miR-7) in SZ patients were significantly up-regulated as
compared to healthy controls (P < 0.05P < 0.001). Relative levels
of the five significantly altered miRNAs are demonstrated in
Figure 1. To further evaluate the diagnostic value of these five
miRNAs, ROC (Receiver Operating Characteristic) curves of the
five miRNAs were constructed, and AUC (Area under ROC curve)
was established to assess the predictive power of each miRNA(Figure S1), and the combined predictive power of all 5 miRNAs
was also computed [Pepe et al., 2006;[Pepe and Thompson, 2000]
(AUC:0.713; sensitivity:35.5%; specificity:90.2%, Fig. 2). To determine the independent predicting value of the five miRNAs for SZ
patients, we conducted multivariate logistic regression analysis
using each of the five miRNAs as independent variables, and SZ
and healthy controls as dependent variables. Among the tested
miRNAs, plasma level of miR-181b most significantly associated
with SZ patients(Odd Ratio:2.483; 95%CI:1.4674.201), implicating miR-181b as a strong independent predictor for SZ patients.

Plasma miRNA Levels in Responses to Clinical

To identify blood-based miRNA biomarkers for therapeutic
responses in SZ patients, we analyzed the plasma levels of all
nine miRNAs before medication, and 3 and 6 weeks after medication. As shown in Table II, significant decreases in plasma levels
were observed in 3 and 6 weeks after medication for miR-132
(P < 0.05) and 6 weeks for miR-181b, miR-30e and miR-432
(P < 0.05P < 0.01). Dynamic changes of the 4 significant miRNAs
in responses to medications are shown in Figure S2. To determine
the plasma miRNA changes associating with different medications,
we compared the plasma miRNA changes among the SZ patients
who received different psychotropic medication. Significantly decreased plasma levels of 4 miRNAs (miR-132, miR-195, miR-30e
and miR-432) were observed in responses to all four medications
(P < 0.05)(Table S5). Among them, drug olanzapine had the
strongest effect on plasma miRNA level changes (Table S6).


Correlations of Plasma miRNA Levels with

Clinical Effects of Medication in Sz Patients


To further explore plasma miRNAs as non-invasive biomarkers

guiding clinical treatment of SZ patients, we analyzed the correlations of plasma miRNA changes with therapeutic effects of
psychotropic medications and clinical symptoms of the patients
after medication. As shown in Table S7, the total PANSS score was
decreased, and the GAS and SI scores were increased after medication, indicating the clinical improvements and positive therapeutic
effects of the applied medication on the patients. We found that
plasma levels changes of four miRNAs (miR-132, miR-181b, miR-



Fig. 1. Comparision of aberrant miRNAs expression levels between SZ and healthy controls. Plasma levels of selected MiRNAs were analyzed
in 61 SZ patients and 62 matched healthy controls using quantitative real-time PCR. The figure shows DDCT of 5 miRNAs significantly
different between SZ patients and healthy controls. The plots were constructed using GraphPad Prism 5 software and statistical difference
was analyzed by Wilcoxon sum test.


after medication after six-week medication. Score reduction rate
less than 0.5 was taken as low treating effect, and score reduction
rate more than 0.5 as high treating effect. The significant improved and improved assessed by CGI scale were reclassified as
high-effect subgroup, while the slightly improved and not
improved assessed by CGI scale as low-effect subgroup. As
indicated in Table IV, the decreases of plasma levels of miR-132
and miR-432 were significantly greater in high-effect subgroup
than those in the low-effect subgroup after six-week treatment
course(P < 0.05).


Fig. 2. Combined diagnostic performance of miR-181b,miR-30e,

miR-346, miR-34a and miR-7. ROC curve was constructed to
examine the diagnostic value of the 5 miRNAs combined as a
panel. The ROC curve is based on the relative levels(DDCT) of
the 5 miRNAs in plasma from all 123 SZ patients and healthy
controls. SPSS 17.0 software was used to construct the ROc
curve. Area under curve (AUC): 0.713.

212, and miR-30e) highly correlated with the clinical scores changes
of the patients after medication (Table III). To verify the role of
plasma miRNAs as biomarkers predicting therapeutic effects in SZ
patients, we reclassified all patients into high-effect and low-effect
subgroup based on the PANSS score reduction rate((Spre-Spost)/
Spre, Spre represents pre-medication score, Spost represents postmedication score) and the degree of improvement of the symptoms

In this study, we attempted to explore the effects of antipsychotics

upon abnormally expressed miRNAs in SZ patients. The results
revealed that miRNA-181b, miRNA-30e, miRNA-346, miRNA34a, and miRNA-7 expression were significantly up-regulated in SZ
patients before psychotropic medication, suggesting that these five
miRNAs significantly contribute to the molecular mechanism of
SZ. Using high throughput microchip analysis combined with
Northern blotting and quantitative RT-PCR, a study analyzed 76
differentially expressed miRNAs from postmortem tissues and
found that significantly increased expression of miR-181b was
observed in superior temple gyrus(STG) and dorsolateral prefrontal cortex(DL-PFC) of SZ patients[Beveridge et al., 2008]. This
observation supports the finding in our current study. However,
analysis of blood miRNA profiles in SZ patients using a genomewide microarray approach in another study showed that miR-181b
expression in peripheral mononuclear leucocytes was significantly
down-regulated[Gardiner et al., 2011], which is inconsistent with
our findings. Although it remains unclear, one of the reasons
underlying this discrepancy might be due to different specimens
in different studies. The up-regulation of miR-181b expression
both in brain tissues and in peripheral tissues (plasma) in SZ
patients implies that miR-181b potentially plays a very important
role in pathogenesis of SZ. Functional importance of miRNA-181b
in SZ is further supported by the reports that over-expression of
miR-181b led to dysregulation of cortical gene expression in SZ
patients[Hunsberger et al., 2009], and altered expression of miR181b induced abnormal expressions of Vsnl1 and Gria2 in SZ and

TABLE II. Comparison of miRNA Expression in SZ Patients Between Pre-Medication, Three-Week and Six-Week Post-Medication(DDCt)
(n 18)

5.14  0.33
2.06  0.26
3.64  0.43
5.66  0.27
1.95  0.42
4.95  0.43
9.21  0.43
4.79  0.39
9.78  0.32

Three-week post-medication
5.18  0.33*
1.95  0.27
3.67  0.28
5.85  0.32
2.40  0.40
5.42  0.22
9.31  0.28
4.88  0.38
11.13  0.56

Note: Compared with pre-medication,* represents P < 0.05, ** represents P < 0.01; compared with three-week post-medication, D represents P < 0.05.

Six-week post-medication
5.95  0.28*D
2.80  0.24**
4.27  0.36
6.13  0.24
3.07  0.34*
5.30  0.21
9.97  0.59
5.91  0.44D
10.76  0.42


TABLE III. Correlation Between PANSS and GAS Scores Changes and miRNA DDCt Value Changes in SZ Patients Before, Three-Week and
Six-Week Post-Medication(r)
PANSS total score changes



GAS score changes





Note: D1represents the differences of three-week post-medication and pre-medication;D2 represents the differences of six-week post-medication and three-week post-medication;D3represents the
differences of six-week post-medication and pre-medication;* represents P < 0.05.

autism patients[Beveridge et al., 2008; Beveridge et al., 2009;

Mellios et al., 2009; Gardiner et al., 2011; Shi et al., 2012].
Consistent with our study, alterations of miR-30e has also been
reported in SZ patients. The result that miR-30e expression level
was significantly elevated in the lower age subgroup as compared to
the higher age subgroup suggests that miR-30e might be the one
miRNA exhibiting differential expression in SZ patients at the
earliest stage. Perkins first reported aberrant miR-30e expression
in prefrontal cortex of postmortem schizophrenic brain[Perkins
et al., 2007]. Xu et al. revealed a strong association between the
polymorphism ss178077483 in the miRNA-30e precursor (premiR-30e) and the risk of SZ[Xu et al., 2010]. In addition, miR-34a
was also reported to significantly up-regulate in the blood mononuclear cells of patients with Alzheimers disease[Schipper et al.,
2007]. In cultured cell lines, mood stabilizers reduced the expression of miR-34a and consequentially led to increased expression of
GRM7, a confirmed target gene of miR-34a and a candidate gene for
SZ identified in recent association studies[Zhu et al., 2009]. Furthermore, miR-34a was reported to be one of the seven signature
miRNAs in the prefrontal cortex of patients with SZ[Kim et al.,
2010]. Expression of miR-34a also causes dramatic reprogramming
of gene expression [Chang et al., 2007].For miR-346, the majority of
studies investigated alteration of miR-346 in brain tissues. The gene
encoding miR-346, is located in intron 2 of the glutamate receptor

ionotropic delta 1 (GRID1) gene, which has been previously

implicated in SZ susceptibility [Fallin et al., 2005; Guo et al.,
2007]. Expressions of both miR-346 and GRID1 were reduced in
SZ patients as compared to normal controls. A previous comprehensive study which analyzed expression changes of all known
human miRNAs and their potential target genes in SZ patients
identified 10 miRNAs associated with SZ, including miR-346[Guo
et al., 2007]. However, circulating miR-346 was reported undetectable in peripheral blood in a previous study[Shi et al., 2012].
Accumulated data suggested the critical role of miR-346 in SZ
and further investigation of its function and clinical importance in
SZ patients should be warranted. Involvement of miR-7 in SZ was
first reported by Perkins, who found that miR-7 expression was
dysregulated in the prefrontal cortex of individuals with SZ and
schizoaffective disorder[Perkins et al., 2007]. The above finding was
further supported by the observations that miR-7 was one of the 26
aberrantly expressed miRNAs in DL-PFC and three members of
miR-7(miR-71, miR-72, miR-73) were among the 7 differentially expressed miRNAs in prefrontal cortex of SZ patients[Emamian et al., 2004; Bajestan et al., 2006; Kim et al., 2010].
Importance of miR-7 in SZ warrants further investigation.
In this study, we analyzed the five differentially expressed
miRNAs (miR-181b, miR-30e, miR-34a, miR-346, and miR-7)
as a panel for SZ diagnosis, and achieved an AUC of 0.713 and

TABLE IV. Comparison of miRNA Expression Changes in SZ Patients Between Different Symptomatology Improvement Subgroups After
Six-Week Medicaiton(x  s)

*represents P < 0.05

High-effect subgroup (n 8)
1.59  1.05*
1.35  1.52
0.79  1.86
1.01  1.82
1.71  1.92
0.33  2.80
1.18  3.23
2.04  2.17*
2.06  3.12

Low-effect subgroup (n 10)

-0.04  1.53
-0.10  1.12
0.20  1.74
-0.24  1.09
0.28  1.90
0.65  1.19
-0.15  1.44
-0.39  2.39
0.53  0.49

sensitivity of 35.5% sensitivity and specificity of 90.2% respectively.
Among these five miRNA, miR-181b was the most significant
predictor for SZ with OR of 2.438 as demonstrated by multivariate
logistic regression analysis. Taken together, findings from our
current and other previous studies suggest that miR-181b, miR30e, miR-34a, miR-346 and miR-7 can be developed to blood-based
biomarkers for SZ diagnosis.
Currently, dopamine receptors system, 5-HT receptors, and
GABA system, are the major pharmacological treatments for SZ
patients. However, the pharmacological actions of these drugs
remain unclear, making it difficult to select effective treatment
and predict treatment responses. Up to date, very limited numbers
of studies have investigated pharmaceutical impact on miRNA
expression. Our current study observed altered expression levels of
miR-30e, miR-181b, miR-132, and miR-432 upon medication,
suggesting that these four miRNAs might be functionally implicated in clinical responses to psychotropic drugs in SZ patients. We
further explored the effects of different psychotropic medications
on plasma miRNA levels. Among the four medication subgroups,
the expression of miR-30e, miR-195, miR-132, and miR-432
showed significantly different patterns of changes, with the olanzapine showing the strongest effect. The findings imply that these
four miRNA are the most sensitive to psychotropic medication, and
olanzapine has the most effective therapeutic effect on SZ patients.
Association of plasma miRNA changes with therapeutic effects of
drugs was further supported by the observations that symptomatic
improvements of SZ after treatment tightly correlated with plasma
levels changes of four miRNAs, including miR-132, miR-181b,
miR-212 and miR-30e. In addition, we found that the decreases of
plasma levels of miR-132 and miR-432 were significantly greater in
high-effect subgroup than those in the low-effect subgroup at
different periods of treatment course. Recent findings of Miller
et al. suggest that miR-132 dysregulation and subsequent abnormal
expression of miR-132 target genes contribute to the neurodevelopmental and neuromorphological pathologies present in schizophrenia [Miller et al., 2012]. These findings taken together suggest
that miR-30e, miR-181b, miR-132, and miR-432 can be potential
indicators for therapeutic responses of SZ patients.
In conclusion, miR-181b, miR-30e, miR-34a, miR-346 and miR7 combined as a panel might serve as useful biomarkers for
diagnosis of SZ patients, and markers miR-30e, miR-181b, miR132 and miR-432 combined are potential indications for symptomatology improvements, treatment responses and prognosis for
schizophrenia patients.

It would be ideal to have a microarray analysis before the PCR
verification. Another limitation of our study was the sample size,
which may not be powerful enough to detect variants with minor
effect. Besides, observational study results need to be confirmed by
prospective studies. Thus, further epidemiological and functional
studies in a larger cohort are warranted to validate these results.
Since this study did not involve other psychiatric diseases, such as
bipolar disorder and major depressive disorder and so on, the
conclusion of this study needs further differential verification
against other psychiatric diseases.


The authors wish to thank all the subjects who participated in the
study. Authors also acknowledge invaluable assistance by numerous mental health professionals in different clinical departments
within and without No.102 Hospital of PLA. In particular, the
authors wish to convey sincere thanks to Gopath Global LLC.,
Chicago in USA for their professional laboratory services and
No.102 Hospital clinical laboratory of PLA for professional laboratory assistance.

Abelson JF, Kwan KY, ORoak BJ, Baek DY, Stillman AA, Morgan TM,
Mathews CA, Pauls DL, Rasin M-R, Gunel M. 2005. Sequence variants in
SLITRK1 are associated with Tourettes syndrome. Science 310(5746):317320.
Bajestan SN, Sabouri AH, Nakamura M, Takashima H, Keikhaee MR,
Behdani F, Fayyazi MR, Sargolzaee MR, Bajestan MN, Sabouri Z. 2006.
Association of AKT1 haplotype with the risk of schizophrenia in Iranian
population. Am J Med Genet Part B Neuropsychiatr Genet 141(4):
Beveridge N, Gardiner E, Carroll A, Tooney P, Cairns M. 2009. Schizophrenia is associated with an increase in cortical microRNA biogenesis.
Mol Psychiatr 15(12):11761189.
Beveridge NJ, Tooney PA, Carroll AP, Gardiner E, Bowden N, Scott RJ,
Tran N, Dedova I, Cairns MJ. 2008. Dysregulation of miRNA 181b in the
temporal cortex in schizophrenia. Hum Mol Genet 17(8):11561168.
Chang TC, Wentzel Ea Fau - Kent OA, Kent Oa Fau - Ramachandran K,
Ramachandran K Fau - Mullendore M, Mullendore M Fau - Lee KH, Lee
Kh Fau - Feldmann G, Feldmann G Fau - Yamakuchi M, Yamakuchi M
Fau - Ferlito M, Ferlito M Fau - Lowenstein CJ, Lowenstein Cj Fau - Arking
DE, Arking De Fau - Beer MA, Beer Ma Fau - Maitra A, Maitra A Fau Mendell JT, Mendell JT. 2007. Transactivation of miR-34a by p53 broadly
influences gene expression and promotes apoptosis 26(5):745752.
Consortium SPG-WAS. 2011. Genome-wide association study identifies
five new schizophrenia loci. Nat Genet 43(10):969976.
Cox MB, Cairns MJ, Gandhi KS, Carroll AP, Moscovis S, Stewart GJ,
Broadley S, Scott RJ, Booth DR, Lechner-Scott J. 2010. MicroRNAs miR17 and miR-20a inhibit T cell activation genes and are under-expressed in
MS whole blood. PLoS One 5(8):e12132.
Emamian ES, Hall D, Birnbaum MJ, Karayiorgou M, Gogos JA. 2004.
Convergent evidence for impaired AKT1-GSK3b signaling in schizophrenia. Nat Genet 36(2):131137.
Fallin MD, Lasseter VK, Avramopoulos D, Nicodemus KK, Wolyniec PS,
McGrath JA, Steel G, Nestadt G, Liang K-Y, Huganir RL. 2005. Bipolar I
disorder and schizophrenia: A 440single-nucleotide polymorphism
screen of 64 candidate genes among Ashkenazi Jewish case-parent trios.
Am J Hum Genet 77(6):918936.
Feng J, Sun G, Yan J, Noltner K, Li W, Buzin CH, Longmate J, Heston LL,
Rossi J, Sommer SS. 2009. Evidence for X-chromosomal schizophrenia
associated with microRNA alterations. PLoS One 4(7):e6121.
Gardiner E, Beveridge N, Wu J, Carr V, Scott R, Tooney P, Cairns M. 2011.
Imprinted DLK1-DIO3 region of 14q32 defines a schizophrenia-associated miRNA signature in peripheral blood mononuclear cells. Mol
Psychiatry 17(8):827840.
Guo S-Z, Huang K, Shi Y-Y, Tang W, Zhou J, Feng G-Y, Zhu S-M, Liu H-J,
Chen Y, Sun X-D. 2007. A Case-control association study between the
GRID1 gene and schizophrenia in the Chinese Northern Han population.
Schizophr Res 93(1):385390.

Hansen T, Olsen L, Lindow M, Jakobsen KD, Ullum H, Jonsson E,
Andreassen OA, Djurovic S, Melle I, Agartz I. 2007. Brain expressed
microRNAs implicated in schizophrenia etiology. PloS one 2(9):e873.
Hunsberger JG, Austin DR, Chen G, Manji HK. 2009. MicroRNAs in
mental health: from biological underpinnings to potential therapies.
Neuromolecular Med 11(3):173182.
Kawanishi Y, Tachikawa H, Suzuki T. 2000. Pharmacogenomics and
schizophrenia. Eur J Pharmacol 410(2):227241.
Killackey E, Yung AR. 2007. Effectiveness of early intervention in psychosis.
Curr Opin Psychiatry 20(2):121125.
Kim AH, Reimers M, Maher B, Williamson V, McMichael O, McClay JL,
van den Oord EJ, Riley BP, Kendler KS, Vladimirov VI. 2010. MicroRNA
expression profiling in the prefrontal cortex of individuals affected with
schizophrenia and bipolar disorders. Schizophr Res 124(1):183191.

Pepe MS, Thompson ML. 2000. Combining diagnostic test results to
increase accuracy. Biostatistics 1(2):123140.
Perkins DO, Jeffries CD, Jarskog LF, Thomson JM, Woods K, Newman MA,
Parker JS, Jin J, Hammond SM. 2007. MicroRNA expression in the
prefrontal cortex of individuals with schizophrenia and schizoaffective
disorder. Genome Biol 8(2):R27.
Schipper HM, Maes OC, Chertkow HM, Wang E. 2007. MicroRNA
expression in Alzheimer blood mononuclear cells. Gene Regul Sys
Biol 1:263.
Schwarz E, Bahn S. 2008a. Biomarker discovery in psychiatric disorders.
Electrophoresis 29(13):28842890.
Schwarz E, Bahn S. 2008b. The utility of biomarker discovery approaches
for the detection of disease mechanisms in psychiatric disorders. Br J
Pharmacol 153(S1) S133S136.

Kim J, Inoue K, Ishii J, Vanti WB, Voronov SV, Murchison E, Hannon G,

Abeliovich A. 2007. A MicroRNA feedback circuit in midbrain dopamine
neurons. Science 317(5842):12201224.

Shi W, Du J, Qi Y, Liang G, Wang T, Li S, Xie S, Zeshan B, Xiao Z. 2012.

Aberrant expression of serum miRNAs in schizophrenia. J Psychiatr Res

Kroh EM, Parkin RK, Mitchell PS, Tewari M. 2010. Analysis of circulating
microRNA biomarkers in plasma and serum using quantitative reverse
transcription-PCR (qRT-PCR). Methods 50(4):298301.

Song H-t, Sun X-y, Zhang L, Zhao L, Guo Z-m, Fan H-m, Zhong A-f, Niu
W. 2014. A preliminary analysis of association between the downregulation of microRNA-181b expression and symptomatology improvement in schizophrenia patients before and after antipsychotic
treatment. J Psychiatr Res 54:134140.

Lai C-Y, Yu S-L, Hsieh MH, Chen C-H, Chen H-Y, Wen C-C, Huang Y-H,
Hsiao P-C, Hsiao CK, Liu C-M. 2011. MicroRNA expression aberration
as potential peripheral blood biomarkers for schizophrenia. PloS one
Lakhan SE, Kramer A. 2009. Schizophrenia genomics and proteomics: are
we any closer to biomarker discovery. Behav Brain Funct 5(2):19.
Marazziti D, Catena DellOsso, Baroni M, Masala S, DellOsso I, Consoli B,
Giannaccini G, Betti G, Lucacchini L. 2010. Alterations of the dopamine
transporter in resting lymphocytes of patients with different psychotic
disorders. Psychiatr Res 175(1):5457.
Mellios N, Huang H-S, Baker SP, Galdzicka M, Ginns E, Akbarian S. 2009.
Molecular determinants of dysregulated GABAergic gene expression in
the prefrontal cortex of subjects with schizophrenia. Biol Psychiatry
Miller BH, Zeier Z Fau - Xi L, Xi L Fau - Lanz TA, Lanz Ta Fau - Deng S,
Deng S Fau - Strathmann J, Strathmann J Fau - Willoughby D, Willoughby D Fau - Kenny PJ, Kenny Pj Fau - Elsworth JD, Elsworth Jd Fau Lawrence MS, Lawrence Ms Fau - Roth RH, Roth Rh Fau - Edbauer D,
Edbauer D Fau - Kleiman RJ, Kleiman Rj Fau - Wahlestedt C, Wahlestedt
C. 2012. MicroRNA-132 dysregulation in schizophrenia has implications
for both neurodevelopment and adult brain function. (1091 -6490
(Electronic)) 109(8):31253130.
Pepe MS, Cai T, Longton G. 2006. Combining predictors for classification
using the area under the receiver operating characteristic curve. Biometrics 62(1):221229.

Thomson JM, Parker J, Perou CM, Hammond SM. 2004. A custom

microarray platform for analysis of microRNA gene expression. Nat
Methods 1(1):4753.
Vawter MP, Barrett T, Cheadle C, Sokolov BP, Wood III, Donovan WH,
Webster DM, Freed M, Becker WJ. 2001. Application of cDNA microarrays to examine gene expression differences in schizophrenia. Brain Res
Bull 55(5):641650.
Wang Y, Zheng D, Tan Q, Wang MX, Gu L-Q. 2011. Nanopore-based
detection of circulating microRNAs in lung cancer patients. Nature
Nanotechnol 6(10):668674.
Xu Y, Li F, Zhang B, Zhang K, Zhang F, Huang X, Sun N, Ren Y, Sui M, Liu
P. 2010. MicroRNAs and target site screening reveals a pre-microRNA30e variant associated with schizophrenia. Schizophr Res 119(1):
Zhu Y, Kalbfleisch T, Brennan MD, Li Y. 2009. A MicroRNA gene is hosted
in an intron of a schizophrenia-susceptibility gene. Schizophr Res 109(1):8689.

Additional supporting information may be found in the online
version of this article at the publishers web-site.