SUPPLEMENTARY INFORMATION
Convergentlines of evidence support CAMKK2 as a schizophrenia susceptibility gene
Xiong-jian Luo, Ming Li et al.
Correspondence to Xiong-jian Luo, Ph.D., Flaum Eye Institute and Department of Ophthalmology, University of Rochester, Rochester 14642, New York, USA. E-mail: ;
Correspondence to Lin Gan, Ph.D., Flaum Eye Institute and Department of Ophthalmology, University of Rochester, Rochester 14642, New York, USA. E-mail:.
Members of the MooDS SCZ Consortium
Jana Strohmaier1, Sandra Meier1, Josef Frank1, Stephanie H. Witt1, Franziska A. Degenhardt2, Andreas J. Forstner2, Buket F. Basmanav2, Peter Propping2, Manuel Mattheisen3,Christoph Lange4,Markus Leber5, Thomas G. Schulze6, C. Christoph Schultz7, Ralf G.M. Schlösser7, Igor Nenadic7, Heinrich Sauer7, Rainald Mössner8, Wolfgang Maier8,Ina Giegling9, Dan Rujescu9, Bertram Mueller-Myhsok10, Sven Cichon2, Markus M. Noethen2, Marcella Rietschel1
1Department of Genetic Epidemiology in Psychiatry, Central Institute of Mental Health, University of Mannheim, D-68159 Mannheim, Germany
2Department of Genomics, Life & Brain Center, and Institute of Human Genetics, University of Bonn, D-53127 Bonn, Germany
3Department of Biomedicine, Aarhus University, 8000 Aarhus C, Denmark
4Department of Genomic Mathematics, University of Bonn, D-53127 Bonn, Germany
5Institute for Medical Biometry, Informatics, and Epidemiology, University of Bonn, 50127 Bonn, Germany
6Department of Psychiatry and Psychotherapy, University Medical Center Georg-August-Universität, 37075 Goettingen, Germany
7Department of Psychiatry and Psychotherapy, Friedrich-Schiller-University Jena, Jena, Germany
8Department of Psychiatry, University of Bonn, Germany
9Department of Psychiatry, University of Halle-Wittenberg, 06112 Halle/Saale, Germany
10Department of Statistical Genetics, Max-Planck-Institute of Psychiatry, 80804 Munich, Germany
Replication sample information
(1)The SGENE-plus sample, subset not included in the PGC genome-wide analysis
This sample consisted of 1,932 schizophrenia cases and 92,074 controls from 10 European study groups. There were 93 cases and 88 controlsfrom England;59 cases and 147 controls from Finland; 123 cases and 50 controls from a population isolate in Finland; 197 cases and 203 controls from Georgia; 163 cases and 185 controls from Germany (Munich);593 cases and 90,670 controls from Iceland; 84 cases and 89 controls from Italy; 152 cases and 137 controls from Macedonia; 419 cases and 405 controls from Russia; and49 cases and 100 controls from Serbia. Cases were diagnosed according to ICD-10 or DSM-IV criteria, and genotyping was carried out at deCODE genetics using Illumina arrays. Samples that were low-yield, duplicates of a higher-yield sample already included, not consistent in reported and X-chromosome-heterozygosity-determined sex, or likely to have non-European ancestry were removed. Markers were excluded if they had yield < 0.95 or a Hardy-Weinberg equilibrium p-value < 1 x 10-5 in controls. Association analysis was carried out using logistic regression assuming an additive genetic model. In each study group, genomic control was used to correct for relatedness and potential population stratification. Results from thevarious study groups were combined using inverse-variance weighted, fixed-effects meta-analysis.
(2)The German sample from MooDS SCZ consortium
This sample consisted of 1,332 schizophrenia cases and 866 controls. It represents the non-overlapping part (PGC, “SGENE-Bonn”) of the MooDS SCZ Consortium sample that has been described in detail in Priebe et al.1 In brief, the study was approved by the ethics committees of all study centers and each participant provided written informed consent prior to inclusion.All patients were recruited from consecutive admissions to psychiatric inpatient units and of German descent according to self-reported ancestry. A lifetime “best estimate” diagnosis2 of SCZ or SCZA according to DSM-IV criteria3was assigned on the basis of the Structured Clinical Interview4or the OPCRIT5, medical records, and family history.Venous blood samples were obtained from all participants. These were genotyped separately using the following Illumina's BeadChips: HumanHap550v3, Human610-Quadv1, and Human660W-Quad. Community based controls were taken from a sample that has previously been used to study the genetics of major depression6. More details on this sample can be found in Muglia et al.6Only those markers common to all utilized genotyping chips were subjected to a stringent quality control protocol (for a detailed description see Rietschel et al.7) and subsequently imputed.Imputation was carried out using IMPUTE8 (version 2) and the February 2012 release of the 1,000 Genomes Project data9. Imputed SNPs with an info score < 0.4 or a MAF < 0.03 were removed during the post imputation QC and the sample-specific association analyses using SNPTEST10 and an additive model.
(3)The Jewish-Israeli sample
The Jewish family sample consisted of 107 nuclear families [331 individuals with DNA of whom 155 are affected, 43 Ashkenazi and 64 non-Ashkenazi families (both parents being of Ashkenazi or non-Ashkenazi origin), mixed families (one parent Ashkenazi and one non-Ashkenazi) were not included]; 70 were ‘triad’ families (parents and one affected offspring), while the rest had two or more offsprings, with at least one affected with schizophrenia. Clinical evaluation included a semi-structured interview, the Schedule for Affective Disorders and Schizophrenia-Lifetime Version (SADS-L), a Family History Diagnostic Interview (FHRDC) and assessment of medical records. Primary diagnoses were established by a best-estimate procedure11 according to Research Diagnostic Criteria. This sample has been used previously and no population stratification was detected.12, 13
(4)The Chinese sample from Shanghai
This sample consisted of 3,750 schizophrenia cases and 6,468 controls. All schizophrenic patients were diagnosed according to Diagnostic and Statistical Manual of Mental Disorders (DSM)-IV criteria by two independent psychiatrists. Healthy controls were randomly selected from Han Chinese volunteers and individuals with mental illness were excluded. Genomic DNA was extracted from peripheral blood lymphocytes using FlexiGene DNA kits (Fuji). SNPs were genotyped by the Affymetrix Genome-Wide Human SNP Array 6.0. SNPs with call rates < 95% in either cases or controls were removed. And SNPs with minor allele frequency (MAF) <1% or significant deviation from Hardy-Weinberg equilibrium (HWE; P<1×10−3) in controls were also excluded.The association of single SNPs with schizophrenia was conducted by logistic regression using PLINK. More detailed information about sample description (recruitment of schizophrenic patient and healthy controls, diagnosis), genotyping, quality control, population stratification analysis, and statistical analysis can be found in the original paper.14
(5)The Chinese sample from Beijing
This sample contained 746 schizophrenia cases and 1,599 healthy controls.Individuals with schizophrenia were interviewed by at least two experienced psychiatrists according to the Diagnostic and Statistical Manual of Mental Disorders IV (DSM-IV). Subjects with severe medical complications or other psychiatric disorders were excluded from this study. All healthy controls were clinically diagnosed to be free of autoimmune or psychiatric disorders or family history of such disorders (including first-, second- and third-degree relatives). Controls were matched to caseson age and gender. This study was approved by the Institutional Review Board of each hospital and written informed consent was obtained from all participants.Genomic DNA was extracted using the QIAamp DNA Mini Kit (QIAGEN) and genotyping was performed using Illumina HumanHap610- Quad BeadChips. For quality control, SNPs with genotype call rates <90% were removed. In addition, SNPs with minor allele frequencies <1% or significant deviation from Hardy-Weinberg equilibrium (P < 1×10−5) in the controls were also excluded. After stringent quality control filtering, the total genotyping rate was 0.999291 in the analyzed individuals (746 cases and 1,599 controls).Single SNP association was carried out by the Cochran-Armitage trend test using PLINK [v1.07] software. Quantile-Quantile plot was used to evaluate the overall significance of the genome-wide association results. For more details, please refer to the original paper.15
Cognitive function analyses
The Irish sample
This sample consisted of282schizophrenicpatients with DSM-IV–diagnosed and 85healthy participants.All schizophrenic cases were recruited from 5 sites across Ireland and the participants were aged 18 to 65 years. All cases had no history of substance abuse in the preceding 6 months,and had no previous head injury with loss of consciousness ora history of seizures.Diagnosis was verified by trained psychiatristsaccording to the Structured Clinical Interview for DSM-IV Axis1 Disorders (SCID).The control sample was also recruited from Ireland based on their responsesto local media advertisements. The ages of controls were 18 to 65 years. All controls have no history of majormental health problems, intellectual disability, or acquired braininjury and no history of substance misuse in the preceding 6months based on self-report. Control participants were excludedif their first-degree relative had a history of psychosis. The assessments of patients and controls were conducted in accordance with the relevant ethicscommittee approval from each participating site. All subjects (patientsand controls) were of Irish ancestry (i.e., 4 grandparents born inIreland), and all provided written informed consent.
Cognitive functions known to be impaired in schizophrenia were assessed, including IQ, episodic memory, working memory, and attention.IQ was measured using selected subtests(vocabulary, similarities, block design, and matrix reasoning)from the WAIS, third edition,16 yielding a full-scale verbaland performance IQ.Episodic memory was assessed using the logical memory immediate and delayed fromthe Wechsler Memory Scale, third edition (WMS-III).17Verbal and spatial working memory were evaluated using the Wechsler Letter Number Sequencing task(WMS-III) and the spatial working memory task from the CambridgeNeuropsychological Test Automated Battery.17 The association between rs1063843 and cognitive performance was performed by 1-way analysis of variance(ANOVA) with age and sex as covariates. All of the tests were conducted by SPSS 14.18More detailed information about sample description, genotyping, quality control, cognitive assessments, and statistical analysis can be found in the original paper.19
The Chinese sample
The Chinese sample consisted of 342 healthy (i.e., no history of neurological and psychiatric disorders based on self-reports) Chinese college students from Beijing Normal University. Among them, 197 subjects were females and 145 individuals were males. Their age ranged from 18 to 23 years old. This experiment was approved by the IRB of the State Key Laboratory of Cognitive Neuroscience and Learning at Beijing Normal University, China. Written consent form was obtained from all participants after a full explanation of the study procedure.
Cognitive and behavioral measures included working memory, executive functions (as assessed with the Attention Network Test, the Wisconsin Card SortingTask, and a reversal learning test), IQ, personality and motivation traits, etc. Cognitive functionsincluded in this study are listed in Supplementary Table S1.DNA was extracted and purified from blood sample using the standard procedure. The genotyping was performed by Affymetrix Genome-Wide Human SNP Array 6.0with standard protocol. The association between rs1063843 and cognitive performance was performed by ANOVA. All statistical analyses were carried out in SPSS15.0 for Windows.Detailed information of cognitive and behavioral data acquisition, genotyping, and statistical analysis can be found in the previous studies.20, 21
Table S1. Cognitive performance assessment in Chinese sample (N=342).Cognitive domain / Task / Brief description / Index / Reference
Memory / Wechsler Memory Scale –3rd Edition (WMS-III) / Two subscales: Picture recall (Subjects were showed pictures of 20 simple objects for 30 seconds and then asked to recall as many as possible) and picture recognition (Subjects were showed pictures of 8 simple objects for 30 seconds and then asked to pick them out from 28 pictures). / Number of items correctly recalled or recognized / Wechsler, 194517, 22
Working memory / In the 2-back working memory task, subjects judged whether the current item was the same (or related) to the one presented two trials earlier. Three sessions involved morphological, phonological, and semantic judgment. / Overall accuracy / Xue, 200423
Cambridge Face Memory Test / Look at 6 faces for 20 seconds as learning, then in testing stage subjects have to pick out the ones present from 3 faces, with the other 2 never presented before / Accuracy / Duchaine, 200624
False memory / Subjects saw 50 picture slides depicting a story. After 30 minutes of filler tasks, subjects were asked to read narrations of the story, with 38 accurate and 12 inaccurate descriptions. After 10 minutes of filler tasks, subjects took the recognition test and then the source monitoring test. / Overall false memory and robust false memory test / Loftus, 200325
Executive function / Attention network test / Subjects saw several small arrows on the computer screen and had to judge the direction of the arrow in the middle (left or right). The 6 peripheral arrows can either in the same or inverse direction to the middle one. There were also cues to alert subjects or point to the position where arrows will be presented / Alert, orientation, conflict / Fan, 200226
Stroop test / Four Chinese color words were used: red, green, blue, and yellow. Each character was presented in the color either congruent or incongruent with its meaning. Subjects were asked to respond to the print color and ignore the meaning of the word. / Stroop effect, e.g., RT Difference between
incongruent and congruent conditions / Adopted from Stroop, 193527
Wisconsin card sorting task / Subjects had to select one from four cards that fits a rule. Rules included color, form, and amount of items on the cards, and rules changed during the experiment / Preserved error / Berg, 194828
Stop-signal test / Subjects had to press a key when a letter appears on the screen as soon as possible, except when a tone followed the letter. / SSRT, the average response time minus the time for inhibition signal / Aron, 200629
Reversal Learning / Two figures were presented simultaneously side-by-side, and one of them was designated as the target. Subjects had to learn which one was the “target” according to feedback, but the feedback was wrong some of the times and the rules changed after a few items. / Persistent error / Cools, 200230
IQ / Raven's Advanced Progressive Matrices / Subjects were given 30 minutes to complete as many items as possible. In each test item, subjects were asked to select from several alternatives the missing segment that would complete a larger pattern. / Number of items correctly finished / Raven, 195631
Wechsler Adult Intelligence Test / Six subscales: Verbal section included general knowledge, similarities, digit span; Performance section included picture completion, symbol digit coding, block design / Verbal IQ, Performance IQ / Wechsler, 195532
Gong, 1992 for Chinese version33
Personality / Temperament and Character Inventory-Revised / 7 aspects of personality: Novelty Seeking, Harm Avoidance, Reward Dependence, Persistence, Self-Directedness, Cooperativeness, Self-Transcendence / 7 subscales scores / Cloninger, 199334
Highly Sensitive Person Scale / Sensitivity to stimuli in daily life / Total score / Aron, 199735
Internal Self Awareness (ISA) / Self awareness of one’s emotions, feelings, and thoughts / Total score / Scheier, 198536
EAS temperament survey (Tempo) / Activity items of EAS / Total score / Buss, 198437
The BIS/BAS Scales / Behavior inhibition/approach systems of personality / BIS and BAS score / Carver,199438
Work Preference Inventory / Intrinsic/Extrinsic Motivation / Intrinsic/
Extrinsic Motivation score / Amabile, 199439
Expression analysis of CAMKK2 in human tissues
Tissue specific expression profiling analysis
To investigate the tissue specific expression pattern of CAMKK2, we obtained expression data from the Gene Enrichment Profiler,40 a database that contained gene expression data for 126 different cell types/tissues. To explorewhether the expression of CAMKK2is tissue specific, an enrichment scores is used to benchmark expression levels in one tissue compared to all other tissues. The expression intensities were converted to enrichment scores toreflect the enrichment of a gene in a tissue given its expression inall tissues.Genes that are widely expressed score low and around 0. However, genes that showed tissue specific expression score closer to 1. The enrichment score allows comparison between genes and therefore can be used to rank genes in each tissue.
To obtain enrichment score, each tissue is compared pairwise to each of the other tissues. The LIMMA module (Bioconductor) is used to calculate a linear model coefficient for each pairwise comparison. The enrichment score is defined for each probe as the sum of all linear model coefficients with P<0.05. For more detail, please refer to the work of Benita et al.40
Analyses of temporal-spatial expression pattern of CAMKK2 in human brain
To determine the temporal-spatial expression of CAMKK2 in human brains, we extracted expression data of CAMKK2 from Human Brain Transcriptome (HBT)41 and Brain Cloud42 databases. The HTBdatabase includes transcriptome of 16 regions comprisingthe cerebellar cortex, mediodorsal nucleus of the thalamus, striatum,amygdala, hippocampus, and 11 areas of the neocortex. Totally, 1,340 tissue samples were collected from 57 developingand adult post-mortem brains. The BrainCloud database ( contains genome-wide gene expression data from the human postmortem dorsolateral prefrontal cortex (DLPFC) of normal subjects (N = 261) across the lifespan. More detailed information can be found in the original papers.41, 42
Table S2.Genes that were significantly associated with schizophrenia in recent GWAS of schizophrenia.
Gene / SNP / P value in original studya / ReferencesMIR137 / rs1625579 / 1.59 × 10−11 / Ripke et al.43
PCGEM1 / rs17662626 / 4.65 × 10−8
TRIM26 / rs2021722 / 2.18 × 10−12
CSMD1 / rs10503253 / 4.14 × 10−8
MMP16 / rs7004633 / 2.75 × 10−8
CNNM2 / rs7914558 / 1.82 × 10−9
NT5C2 / rs11191580 / 1.11 × 10−8
STT3A / rs548181 / 2.91 × 10−8
CCDC68 / rs12966547 / 2.60 × 10−10
TCF4 / rs17512836 / 2.35 × 10−8
AMBRA1 / rs11819869 / 3.89×10-9 / Rietschel et al.44
VRK2 / rs2312147 / 1.9 × 10-9 / Steinberget al.45
TCF4 / rs4309482 / 7.8 × 10-9
SDCCAG8 / rs6703335 / 4.22×10-8 / Hamshere et al.46
ITIH3/4 / rs2239547 / 3.62×10-10
CACNA1C / rs4765905 / 1.23×10-8
NOTCH4 / rs2071287 / 3.4×10-8 / Ikeda et al.47
BRP44 / rs10489202 / 9.5×10-9 / Shi et al.14
LSM1 / rs16887244 / 1.27×10-10
WHSC1L1 / rs1488935 / 5.06×10-9
ZKSCAN4 / rs1233710 / 4.76 × 10−11 / Yue et al.15
PGBD1 / rs2142731 / 5.14 × 10−10
NKAPL / rs1635 / 6.91 × 10−12
TSPAN18 / rs11038172 / 7.21 × 10−10
TSPAN18 / rs11038167 / 1.09 × 10−11
TSPAN18 / rs835784 / 2.73 × 10−11
BCL9 / rs672607 / 1.23×10−11 / Li et al.48
MYO18B,ADRBK2 / rs5761163 / 3.44×10-7 / Purcell et al.49
MHC region / rs3130375 / 3.66×10-7
ZNF804A / rs1344706 / 1.95×10-7 / O’Donovan et al.50
IFT74 / rs10812518 / 3.0 × 10-7 / Stefansson et al.51
IFT74 / rs12337896 / 3.7 × 10-7
HIST1H2BJ / rs6913660 / 4.6×10-8
PRSS16 / rs13219354 / 1.4×10-7
HIST1H2BL / rs7776351 / 2.3×10-7
PLAA / rs7863476 / 6.3×10-7
PGBD1 / rs13211507 / 8.3×10-11
NOTCH4 / rs3131296 / 2.3×10-10
NRGN / rs12807809 / 2.4×10-9
TCF4 / rs9960767 / 4.1×10-9
aThe smallest p value in original GWAS of schizophrenia was listed when there were several stages (e.g., stage 1, stage 2, and combined analysis).