Supplementary Information
Secreted Frizzled-related Protein 3 (sFRP3) regulates antidepressant responses in mice and humans
M-H. Jang1,2,3#, Y. Kitabatake1,2,4#, E. Kang1,5, H. Jun1,2, M. V. Pletnikov2,6,7, K. M. Christian1,2, R. Hen8, S. Lucae9, E. B. Binder9,10*, H. Song1,2,5,7*and G-l. Ming1,2,7*
1Institute for Cell Engineering,Johns Hopkins University School of Medicine, Baltimore, MD 21205.
2Department of Neurology,Johns Hopkins University School of Medicine, Baltimore, MD 21205.
3Department of Neurologic Surgery, Department of Biochemistry and Molecular Biology, Mayo College of Medicine, Rochester, Minnesota 55901.
4Department of Pediatrics, Osaka University School of Medicine, 2-2 Yamadaoka, Suita,Osaka, 565-0871 Japan.
5Pre-doctoral Program in Human Genetics, Johns Hopkins University School of Medicine, Baltimore, MD 21205.
6Department of Psychiatry,Johns Hopkins University School of Medicine, Baltimore, MD 21205.
7The Solomon H. Snyder Department of Neuroscience,Johns Hopkins University School of Medicine, Baltimore, MD 21205.
8Department of Psychiatry, Columbia University, New York, NY 10032, USA; The New York State Psychiatric Institute, New York, NY 10032.
9Max Planck Institute of Psychiatry, Kraepelinstr, Munich 80804, Germany.
10Department of Psychiatry and Behavioral Sciences, Emory University School of Medicine, Atlanta, Georgia 30322.
#: contributed equally to this work.
Supplementary Information includes Supplementary Methods and Materials, Acknowledgements and References.
Supplementary Methods and Materials
In situ Hybridization. Animals were perfused with 4% paraformaldehyde and brains were removed and post-fixed overnight at 4°C in the same fixative. After overnight cryoprotection in PBS with 30% sucrose at 4°C, tissue samples were embedded in O.C.T. mounting medium and stored at -80°C until use. Sections (14-µm) were cut onto Superfrost-Plus slides (Fisher Scientific, Pittsburgh, PA) and subjected to in situ hybridization analysis as described previously1. Briefly, coronal sections of frozen brain were cut on a cryostat and mounted onto slides. The digoxygenin-labelled antisense RNA corresponding to the full-length coding sequence of sfrp3 was hybridized with sections for 18 hours at 65oC in the hybridization buffer. The sections were washed in 5x SSC and 1% SDS and then twice in 2x SSC buffer at 65oC for 30 minutes per wash. Hybridized RNA was visualized by overnight treatment with alkaline phosphatase-conjugated anti-digoxygenin antibody at 4°C followed by incubation in 35 µg/ml nitroblue tetrazolium (NBT) and 18 µg/ml 5-bromo-4-chloro-3-indolyl phosphate (BCIP) at room temperature for two days. All experiments were processed in parallel for direct comparison of labelling. All animal procedures used in this study were approved by the Animal Care and Use Committee of the Johns Hopkins University School of Medicine.
Quantitative Analysis of sfrp3 Expression. Hippocampi were rapidly dissected from antidepressants-treated and sham-control treated animals, frozen on dry ice, and stored at -80°C. Total RNA was isolated and subjected to quantitative real-time PCR as previously described1. Quantitative real-time PCR experiments were performed on an ABI PRISM 7900HT sequence detection system using the SYBR-green PCR master mix (Applied Biosystems, Foster City, CA, USA). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mRNA was used as an internal control for mRNA quantification. The following primers were used: GAPDH: 5′- GTATTGGGCGCCTGGTCACC-3′ (forward), 5′- CGCTCCTGGAAGATGGTGATGG-3′ (reverse); sFRP3: 5′- CAAGGGACACCGTCAATCTT-3′ (forward), 5′- CATATCCCAGCGCTTGACTT-3′ (reverse). All reactions were performed in triplicates for each animal and mean levels of sfrp3 mRNA were normalized to the mean of GAPDH levels. Changes in sfrp3 mRNA levels in antidepressants-treated animals were then calculated relative to levels in vehicle-treated controls.
Behaviour Tests. All behavioural tests were performed in the behaviour Core of the Johns Hopkins University School of Medicine. Sfrp3 knockout (KO) mouse were generated by replacing the first coding exon with LacZ2. Littermates of male sfrp3 wild-type (WT) and KO mice, all on C57BL/6 background, at the age of 10-12 weeks were given fluoxetine (20 mg/kg body weight, i.p.; Sigma) once daily for 4 weeks, and testedat least 72 hours after the last change of cage bedding. Mice were transferred from the housing room to the Behaviour Core holding area 1 hour before testing. All tests were carried out by an investigator blind to genotype. For the tail-suspension test, mice were suspended by their tails (with adhesive tape) and immobility time was recorded during the last 4 minutes of the 6-minute testing period3. For forced swimming test, animals were individually forced to swim in an open cylindrical container (diameter 10 cm, height 25 cm), with a depth of 19 cm of water at 23°C-25°C. The duration of immobility was recorded during the last 4 minutes of the 6-minute testing period4. For the locomotor activity, animals were placed in the center of the activity chamber (50 X 50 X 50 cm) with infrared beams (San Diego Instruments Inc., San Diego, CA, USA) and were allowed to freely explore the area for 30 minutes5. The photo-beam based system automatically recorded using the software by the same manufacturer. For the elevated plus maze, the elevated plus maze consists of a central platform and four arms placed 50 cm above the floor as previously described6. Two arms are enclosed within walls and the other two (open) have low rims. Animals were placed on the central platform and their behaviour was recorded for 5 minutes. Time spent in different compartments (closed and open arms, central platform) wasmeasured. For the light-dark test, the apparatus was a rectangular box made of Plexiglas divided by a partition into two compartments. One compartment was dark and the other compartment was brightly illuminated. The time spent in the dark compartment was recorded for the 5-minute test session7.
Human Genetic Association Analysis
Subjects and Treatment. A total of 541 patients recruited within the Munich Antidepressant Response Signature (MARS) project that had been genotyped on the Illumina 610k SNP array were used in the analysis. Details of recruitment and phenotyping procedures were described in Hennings et al., 2009 (ref8). All patients presented with a moderate or severe depressive episode within a unipolar (95.0 % of which 71.2% had a recurrent episode) or bipolar affective disorder (5.0 %, of which 70.0% were bipolar 2). All patients were in-patients and treated with different antidepressants and antidepressant combinations according to doctors’ choice. Patients were rated weekly for depression severity using the 21 item Hamilton Depression Rating Scale (HAM-D)9. Response to antidepressant treatment was quantified using the latency in weeks until patients experienced a 25% (duration to partial response) or 50% decrease (duration to response) in symptom severity from admission. In analyzed samples, 53.8 % were women with anaverage age of 48.8 years (SD: 14.0). Patients had moderate to severe depressive symptoms at admission with a mean HAM-D of 24.9 (SD: 6.0). The average in-patient stay was 11.0 (SD: 17.6) weeks.For the case/control analyses of selected FRZB SNPs, patients with unipolar depression from two cohorts, including MARS (total cases N = 1562) and the respective matched controls (total N = 1405) were used. The recruitment, genotyping and phenotyping of these cohorts as well as previous genetic case-control analyses have been previously described10, 11.
SNP Selection and Genotyping. 19 SNPs within the FRZB locus on chromosome two were selected from the Illumina 610k array to span the gene as well as 10 kb 3’ and 5’ of it, covering over 46kb. SNP genotyping was performed using 200 ng of DNA extracted from whole blood according to the standard Illumina protocol. Raw data were transformed into genotype calls using the Illumina Beadstudio software. All SNPs were in Hardy-Weinberg equilibrium and had a mean call rate of 99.2% (see Supplementary Figures2A and Bfor linkage disequilibrium in this sample).
Statistical Analysis. Association of the 19 SNPs with latency to partial response and latency to response was tested with PLINK version 1.07 using an allelic model, including only the 18 SNPs with a minor allele frequency greater than 5% and correcting for age and gender. The PLINK haplotype QTL module was used to determine haplotype-based associations of top three SNPs. The Benjamini & Hochberg step-up false discovery rate control was used to correct for multiple testing.
Gene Expression Analysis. To gain functional insight into these SNP variants, we interrogated SNP genotype statistical associations with brain transcript expression using a public database Braincloud12. For the BrainCloud database, 269 post-mortem human brains without neuropathological or neuropsychiatric diagnosis were obtained at autopsy. From each subject in the brain collection, RNA from prefrontal cortex grey matter tissue was analyzed using spotted oligonucleotide microarrays yielding data and log2 intensity ratioswere further adjusted to reduce the impact of known and unknown sources ofsystematic noise on gene expression measures using surrogate variable analysis after normalization. For genotyping, DNA was obtained from cerebellar tissue in the collection and applied to IlluminaBeasArrays and genotypes were called using BeadExpress software.We examined the genetic regulation of transcript expression in the human brain for rs4666865 and rs1530056 in FRZB. The statistical results of association of SNP genotype and transcript expression in the human brain are taken from the Braincloud12.
Acknowledgements: We thankmembers of Ming and Song Laboratories for help and critical comments, L. Liu, Y. Cai and S. Hussaini for technical support,A. Rattner and J. Nathans for sfrp3 knockout mice, and D. Weinberger for association analysis between SNPs and FRZB expression in human brains using BrainCloud.This work was supported by NIH (MH090115) and NARSAD to M-H. J.,by Postdoctoral Fellowship Award from the Sankyo Foundation of Life Science in Japan to Y.K.,by NIH (NS048271, HD069184), NARSAD, and MSCRF to G-l.M., by NARSAD, NIH (NS047344, MH087874, ES021957), and IMHRO to H.S., and by a grant of the Exzellenz-Stiftung of the Max Planck Society to S.L and E.B.B.
References
1.Ma DK, Jang MH, Guo JU, Kitabatake Y, Chang ML, Pow-Anpongkul N, et al. Science 2009;323:1074-7.
2.Jang MH. et al. Secreted frizzled-related protein 3 regulates activity-dependent adult hippocampal neurogenesis. Cell Stem Cell2013; (in press).
3.Cryan JF, Mombereau C, Vassout A. Neurosci Biobehav Rev 2005;29:571-625.
4.Porsolt RD, Bertin A, Jalfre M. Arch Int Pharmacodyn Ther 1977;229:327-36.
5.Kitabatake Y, Hikida T, Watanabe D, Pastan I, Nakanishi S. Proc Natl Acad Sci U S A 2003;100:7965-70.
6.Shumyatsky GP, Tsvetkov E, Malleret G, Vronskaya S, Hatton M, Hampton L, et al. Cell 2002;111:905-18.
7.Hascoet M, Bourin M. Pharmacol Biochem Behav 1998;60:645-53.
8.Hennings JM, Owashi T, Binder EB, Horstmann S, Menke A, Kloiber S, et al. J Psychiatr Res 2009;43:215-29.
9.Benjamini Y, Hochberg Y. Journal of the Royal Statistical Society, Series B (Methodological) 1995;57:289-300.
10.Muglia P, Tozzi F, Galwey NW, Francks C, Upmanyu R, Kong XQ, et al. Mol Psychiatry 2010;15:589-601.
11.Kohli MA, Lucae S, Saemann PG, Schmidt MV, Demirkan A, Hek K, et al. Neuron 2011;70:252-65.
12.Colantuoni C, Lipska BK, Ye T, Hyde TM, Tao R, Leek JT, et al. Nature2011;478:519-23.
1