www.sciencemag.org/cgi/content/full/329/5998/1537/DC1

Supporting Online Material for

miR-16 Targets the Serotonin Transporter: A New Facet for Adaptive Responses to Antidepressants

Anne Baudry, Sophie Mouillet-Richard, Benoît Schneider, Jean-Marie Launay* Odile Kellermann*

*To whom correspondence should be addressed. E-mail: jean-marie.launay@lrb.aphp.fr (J.-M.L.);

odile.kellermann@parisdescartes.fr (O.K.)

Published 17 September 2010, Science 329, 1537 (2010) DOI: 10.1126/science.1193692

This PDF file includes:

Materials and Methods Figs. S1 to S10 References

1

Supporting Online Material Materials and Methods Materials - Dibutyryl cyclic AMP (Bt2cAMP), cyclohexane carboxylic acid (CCA), SB-216763 and dimethyl sulfoxide (Me2SO) were purchased from Sigma-Aldrich (St Louis, MO, USA). [3H]-paroxetine (0.98-1.01 TBq/mmol), [125I]-RTI-55 (81.4 TBq/mmol) and [125I]-BE 2254 (81.4 TBq/mmol) were from NEN Life Science Products (Boston, MA, USA) whereas [γ-32P]-ATP (220 TBq/mmol) and [3H]-nisoxetine (2.56-3.12 TBq/mmol) were from Amersham Biosciences. [3H]-LY266097 (2.14 TBq/mmol) was synthesized by Dr J. Würch (Roche, Basel, Switzerland). Fluoxetine was kindly provided by Dr M. Bouhassira (Eli Lilly, Indianapolis, Indiana, USA). The S100β siRNA oligo was from Invitrogen (CA, USA). Wnt3a was purchased from R&D Systems (Minneapolis, MN, USA). Cell culture - 1C11 murine cells were grown and induced to differentiate toward the serotonergic pathway in the presence of 1 mM Bt2cAMP and 0.05% CCA, or along the noradrenergic pathway by the addition of 1 mM Bt2cAMP, 0.05% CCA and 2% Me2SO in Dulbecco’s modified Eagle medium (DMEM) supplemented with 5-HT-depleted fetal calf serum (10%), as described previously (S1). For gain/loss-of-function experiments, 1C115-HT and 1C11NE cells were bombarded with tungsten microprojectiles (S2) coated with miR-16 (Ambion) or anti-miR-16 inhibitor (Ambion) at day 2 of the serotonergic program or at day 10 of the noradrenergic program. After bombardment of fluorescent miRNA, 60 to 65% of the cells were fluorescent. Transfection and luciferase assay - Hela cells were cultured in DMEM supplemented with 10% FCS. The oligonucleotides encoding 2 fold the sequence from 2530 to 2580 nt of the 3’UTR sequence of SERT were cloned into the region directly downstream of the stop codon in the luciferase gene in the pCM firefly luciferase vector (kind gift from Dr M. Kress). Constructs were purified using the Plasmid Maxiprep Kit (Qiagen) and verified by sequencing. Hela cells were transfected using Lipofectamine 2000 (Invitrogen) with the pRL-TK renilla luciferase plasmid (kind gift from Dr M. Kress), the reporter pCM/miR-16 sensor plasmid and the indicated amount of miR-16 pre-miR precursor molecules or pre-miR negative

control (scramble) (Ambion). The pRL-TK was used as an internal control. Forty eight hours after transfection, luciferase activities were measured by using the dual-luciferase reporter assay system (Promega). RNA isolation and Northern Blot analysis - Total RNA was isolated using a total RNA isolation kit according to the manufacturer’s instructions (Promega). 20 µg of total RNA were electrophoretically separated on a 12% polyacrylamide denaturating gel and were then transferred to a Hybond membrane (Amersham). Radiolabeled decade RNA markers (Ambion) were loaded as size markers. After UV-crosslinking of the RNA to the membrane, hybridization was performed using the Rapid-hyb buffer as recommended by the manufacturer’s protocol (Amersham). The probe sequence was complementary to the mature form of miR-16 and was labeled with [γ-32P]-ATP (Amersham Biosciences) by T4 polynucleotide kinase (New England Biolabs). After washing, the membranes were imaged using a STORM phosphoimager. U6 was used as an internal control.

2

RT-PCR and quantitative real-time analysis - Reverse transcription was performed using SuperScript II and oligo dT primers (Invitrogen). Primers used for the PCR analysis were: SERT forward 5’GGCCTTCCTCCTCCCTTACACC3’ and reverse 5’TGTCACCCACACCACCTTGCC3’; NET forward 5’ACATCGGGAAAGGTTGTCTGGA3’ and reverse 5’AGCCAGGAGCATCAAGAAGAAC3’; 5-HT2B receptor forward 5’AGGAATCGAGACTGATGTGAT3’ and reverse 5’CTTAGGAAAACTGTGGGCACA3’; α1D receptor forward 5’AGTGATCGTGGTCATGTACTGC3’ and reverse 5’CTATTGAAGTAGCCCAGCCAGA3’; TH forward 5’-CCACGGTGTACTGGTTCACT-3’ and reverse 5’-GGCATAGTTCCTGAGCTTGT-3’; TPH2 forward 5’-ACCCCGGAACCAGATACATGCCA-3’ and reverse 5’-CCATACGCCCGCAGTTGACCC-3’; Pre/pri-miR-16 forward 5’-CCTGGTATGCAGTGGTAAGGC-3’ and reverse 5’-CTATTGAGGTGCTAGGAG; GAPDH forward 5’TGAAGGTCGGTGTGAACGGATTTGGC3’ and reverse 5’CATGTAGGCCATGAGGTCCACCAC3’, used as an internal control. Mature miR-16 expression was detected using the TaqMan MicroRNA Assay (Applied Biosystems) according to the manufacturer’s protocol. Data analysis was performed using the comparative ∆∆CT method in Software by Roche. U6 was used as an internal control to normalize the relative abundance of miR-16. Radioligand binding studies - Binding experiments were performed on cell membranes as described in (S3). Briefly, cells were harvested in PBS containing a mixture of protease inhibitors (Roche), pelleted by centrifugation and resuspended in cold buffer (4 mM EDTA, 1 mM EGTA, 0.1 mM PMSF, 10 mM imidazole, pH 7.3). After loading of the supernatant onto a 20% sucrose cushion and centrifugation, the membrane-containing pellet was resuspended in buffer (75 mM KCl, 5 mM MgCl2, 1 mM EGTA, 10 mM imidazole, pH 7.3). Protein contents were measured with the BCA protein assay (Pierce). Crude membranes (20 µg) were incubated for 30 min at 37°C in binding buffer (50 mM Tris, pH 7.4) containing radiolabeled ligand and appropriate competing ligands. After addition of ice-cold 10 mM Tris, pH 7.4, samples were filtered on polyethyleneimine-treated filters and the radioactivity retained on the filters was counted in a scintillation counter (Packard). The specific binding was defined as the binding that was inhibited by 1 µM of homologous unlabeled ligands. Determination of bioamines, metabolites and enzymatic activities - Cellular contents of 5-HT, catecholamines (NE, DA) and their metabolites were measured using HPLC combined with electrochemical detection (S4). Enzymatic activities for bioamine synthesis were determined by radioenzymatic assays as described in (S4). Determination of S100ββββ levels - S100β levels were measured by electrochemiluminescence on a Cobas e601 automate (Roche Diagnostics). Intracerebroventricular injection - Adult 6 to 8 week old male Swiss-Kunming mice (25-30 g) were housed at 22 ± 0.5°C with food and water ad libitum and a reversed 12:12 h light cycle. All animal procedures were performed in accordance with National Institutes of Health guidelines for care and were approved by the Animal Care and Use Committee at Basel University.

3

CMA/11 microdialysis guide cannula (CMA Microdialysis, Stockholm, Sweden) were stereotaxically implanted into the raphe or the locus coeruleus of avertin anaesthetized mice according to the mouse brain atlas of Paxinos and Franklin (2001) (S5). At the end of the experiment, examination by light microscopy of hematoxylin-eosin stained tissue sections was performed to confirm the placement of the probe and to ensure that excessive damage had not occurred. The CMA/11 microdialysis probe (0.25 mm diameter and 1 mm membrane length; CMA Microdialysis, Stockholm, Sweden) was connected to a syringe pump by means of a two channel swivel. It was perfused at a rate of 2 µL/min with a solution of artificial cerebrospinal fluid (125 mM NaCl, 3 mM KCl, 1.3 mM CaCl2, 1 mM MgCl2, 23 mM NaHCO3, and 1.5 mM KH2PO4, pH 7.4) containing either fluoxetine at 1 µM, Wnt3a at 50 ng/ml, LiCl at 1 mM, SB-216763 at 100 nM, S100β at 1 nM or antibodies against S100β at 1 µg/ml. For manipulation of miR-16 levels in vivo, 1 µl of mi-R16 or anti-miR16 at 2 µM were directly injected at 36 h intervals. SiRNA against S100β (2 µg, 1 µg/µl) was also directly infused at 36 h intervals. The mouse was allowed to move freely around the cage, although care was taken to ensure that the tubing connecting the probe was kept out of reach. At the indicated times, samples were collected in microtubes containing 1 µL of 10% HCl to prevent oxidation of the recovered compounds. Then, mice were deeply anesthetized with isoflurane and decapitated, and the raphe and the locus coeruleus were collected for RNA extraction and analysis of miR-16 and pre/pri-miR16 expression by real-time PCR as well as radioligand binding experiments. Immunostaining - Mice were anesthetized with avertin and perfused transcardially with 4% paraformaldehyde in PBS. After postfixation (2-4 h) and cryoprotection in 10% sucrose for at least 4 h and 20% in PBS overnight, free-floating cryostat sections (40 µm) were processed by standard procedures and collected onto slides (Superfrost slides; Fisher Scientific, Fair Lawn, NJ). Slides were washed in PBS and incubated for 1 h at room temperature in blocking solution (1% BSA, 5% normal goat serum, 0.3% Triton X-100 in PBS). Sections were then incubated overnight at 4°C with TH antibody (Chemicon; 1:1000) and SERT antibody (Chemicon; 1:1000). After three washes with PBS, sections were incubated for 1-2 h at room temperature with the appropriate fluorescently labelled secondary antibodies. Following three washes in PBS, sections were mounted under coverslips with Fluoromount G (Fisher Scientific). Images were obtained with a Zeiss Meta 510 scanning microscope. For double-label immunohistochemical imaging, the two channels were collected separately with single wavelength excitation and then merged to produce the composite image. GSK3ββββ activity assay - Mice were decapitated and brain regions were rapidly dissected and homogenized in ice-cold lysis buffer containing 10 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 0.5% Nonidet P-40, 10 µg/ml leupeptin, 10 µg/ml aprotinin, 5 µg/ml pepstatin, 1 mM phenylmethylsulfonyl fluoride, 1 mM sodium vanadate, 50 mM sodium fluoride, and 100 nM okadaic acid. The lysates were centrifuged at 20,000 × g for 10 min and the protein concentrations in the supernatants were measured with the BCA protein assay (Pierce). The Duo Set IC ELISA GSK3β kit was used to measure GSK3β phosphorylated at Ser9 in raphe extracts, according to the manufacturer’s protocol (R&D Systems). The relative GSK3β activity was deduced from the ratio of phosphorylated-Ser9 GSK3β (i.e. inactive GSK3β) to total GSK3β.

4

The Forced Swimming Test (FST) - The FST employed was similar to that described elsewhere (S6). Briefly, male mice were dropped individually into glass cylinders (height 25 cm, diameter 10 cm) containing 10-13.5 cm of water, maintained at 23-25°C, and left there for 6 min. In such a situation, from which there is no escape, the animals rapidly became immobile, that is, they float in an upright position and make minimal movements to keep their heads above water. The duration of immobility was recorded during the last 4 min of the 6-min testing period. Unpredictable chronic mild stress (UCMS) - The stress regimen was previously described (S7). Male mice were repeatedly subjected to various socio-environmental stressors according to a “random” schedule for a total period of 6 weeks. Treatment was administered for the last 5 weeks. UCMS-exposed mice were maintained under standard laboratory conditions but were isolated in small individual cages while non-stressed control mice were reared (4 per cage) in larger cages. The stressors were: altered bedding (sawdust change, removal of sawdust, damp sawdust, substitution of sawdust with 21°C water, rat feces); cage tilting (45°); cage exchange (mice were positioned in the empty cage of another male); altered length and time of light/dark cycle, forced swimming. Body weight and fur coat state were assessed weekly. The total score of the coat state resulted from the sum of scores obtained from 5 different body parts: head, neck, dorsal coat, ventral coat, and hindpaws. For each body area, a score of 0 was given for a well-groomed coat and 1 for an unkempt coat. Thus the more elevated the total score the more unkempt is the fur. A dirty state is characterized by a fluffy, greasy, less dense coat or piloerection. Both of these physical parameters have been pharmacologically validated (S8). Locomotor activity - All animals were tested on each of two successive days in an automated locomotor activity monitoring system. The system consisted of 12 clear perspex chambers (35x35x20 cm) with metal grid floors (1 cm mesh), separated by opaque dividers. Each chamber contained two photocells, which monitored locomotor activity by detecting breaks in 2 parallel light beams situated 18 cm apart and 5.5 cm above the floor. The 45 min testing took place between 10:00 A.M. and 15:00 P.M. Sucrose preference test - The consumption of sucrose solution was assessed during the last week of UCMS with a two-bottle-choice drinking paradigm (S9). During 4 days, mice were housed 2 per cage (in order to exclude the confounding effect due to the social isolation) and accustomed to the presence of bottles of water. One bottle contained tap water and the other a daily fresh 4% sucrose (Sigma) solution; both were freely available 24/24 h. Their relative position (left vs. right) was random and changed every day. The day preceding the sucrose solution intake test, mice were deprived of food and water for 24 h. Mice were trained to consume sucrose solution with choice of water in 3 sessions with a 2 day-interval. Fluid consumption was monitored on a daily basis by weighing the bottles. Sucrose preference was determined as follows: (sucrose solution consumed/ sucrose solution consumed + water consumed) x 100. Statistics - The results are reported as the means ± standard errors of the means (S.E.M.). Non parametric (Wilcoxon and Kruskal-Wallis) tests were used for comparisons except for fig. S2 (one-way ANOVA, Bonferroni post hoc test). A P-value < 0.05 was considered significant.

5

Supplementary Figures

Fig. S1. (A) The 1C11 murine neuroectodermal precursor and its serotonergic (1C115-HT) and noradrenergic (1C11NE) neuronal derivatives. (B) RT-PCR analysis of TPH2, TH, 5-HT2B receptor and α1DR transcripts during serotonergic and noradrenergic differentiation of 1C11 cells. Normalization was performed using the level of expression of GAPDH.

6

Fig. S2. miR-16 is enriched in noradrenergic vs serotonergic neuronal cells. (A) Various algorithms (miRanda, Targetscan, Pictar, miRbase) predicted complementarity between miR-16 and the 3’UTR of SERT mRNA. (B) SERT 3’UTR luciferase and Renilla luciferase constructs were transfected into Hela cells with increasing amounts (1 to 20 pmol) of miR-16 or scrambled oligonucleotides. Relative sensor luciferase activity is shown as a percentage of the control. The data represent means ± SEM of four independent experiments, *P<0.0001 vs untransfected cells. (C) Left panel. Detection of miR-16 expression by Northern blot analysis in mouse brain and during 1C11 serotonergic (4 days) and noradrenergic (12 days) differentiation. U6 mRNA was used as a loading control. Right panel. The histogram shows the quantification of miR-16 levels normalized to U6 mRNA. The results are expressed as the means ± SEM, *P<0.0001 vs 1C11 precursor cells. (D) Quantification of miR-16 expression by real-time PCR in the mouse raphe and the locus coeruleus, vs 1C11 precursor cells. The values are the means ± SEM of n=7 animals, *P<0.001 vs raphe.

7

Fig. S3. MiR-16 reduction in 1C11NE cells does not alter the noradrenergic phenotype. MiR-16 sense (miR16) or antisense (anti-miR16) oligonucleotides were introduced into 1C11NE cells at day 10 after induction of differentiation along the noradrenergic pathway. At day 12 of the program, cell extracts were collected to assess (A) the intracellular contents of dopamine (DA), noradrenaline (NE) and 3-methoxy-4-hydroxyphenylglycol (MHPG), (B) the amount of α1D receptors ([125I]-BE 2254 binding). Each bar represents the means ± SEM of seven independent experiments.

8

Fig. S4. The Pre/Pri-miR-16 expression level in mouse raphe and the locus coeruleus was determined by real-time PCR. Values are the means ± SEM (n=4 animals), *P<0.01.

9

Fig. S5. Injection of fluoxetine into raphe leads to an increase in GSK3β activity in raphe and to a decrease in GSK3β activity in the locus coeruleus. Stereotaxic injection (2 µL/min) of fluoxetine (1 µM) into mouse raphe was carried out for 3 days. GSK3β activity was measured in raphe (A) and locus coeruleus (B) extracts. Each bar represents the means ± SEM of n=7 animals, *P<0.01 vs control.

10

Fig. S6. Injection of fluoxetine into the locus coeruleus does not modify miR-16 expression in the locus coeruleus. Fluoxetine (1 µM) was injected by stereotaxis (2 µL/min) into the mouse locus coeruleus for 3 days. The level of expression of miR-16 in the locus coeruleus was determined by real-time PCR. The data represent the means ± SEM (n=7 animals).

11

Fig. S7. De novo SERT expression in the locus coeruleus occurs in TH-positive neurons. There is superposition of SERT (red) and TH (green) immunofluorescence in the locus coeruleus after stereotaxic infusion of fluoxetine (1 µM, 2 µL/min) in raphe for 3 days (Top panel: bar=25 µm). The bottom panel is an enlarged field of the merged images (bar=50 µm).

12

Fig. S8. The down-regulation of the level of miR-16 and the de novo expression of SERT in the locus coeruleus after a 20-day intraperitoneal injection of fluoxetine in mice were abolished by siRNA-mediated knockdown of S100β in raphe. Mice (n=5 to 9 per group) were chronically exposed to fluoxetine for 20 days (daily intraperitoneal injections, 5 mg/kg). During the treatment, S100β-siRNA (2 µg, 1 µg/µl) or scrambled oligonucleotides (2 µg, 1 µg/µl) were injected every 36 h by stereotaxis into the raphe. (A) The miR-16 level as determined by real-time PCR and (B) SERT expression as quantified by [3H]-paroxetine binding were measured in locus coeruleus extracts. The values are the means ± SEM, *P < 0.01 vs control. **P < 0.01 vs scramble.

13

Fig. S9. FST behavioral responses induced by fluoxetine are mimicked by miR-16 elevations in raphe or miR-16 decreases in the locus coeruleus. (A, B) Stereotaxic infusion of fluoxetine (1 µM, 2 µL/min) or miR-16 (1 µl, 2 µM, every 36 h) into mouse raphe or anti-miR16 (1 µl, 2 µM, every 36 h) into the locus coeruleus 3 days before testing, similarly reduced the time of immobility in the FST (A), whereas they had no impact on locomotor activity (B). The data are the means ± SEM (n=6 to 10 mice per group). *P<0.05 vs control. (C, D) Mice (n=5 to 10 per group) were chronically exposed to fluoxetine for 10 days (daily intraperitoneal injection, 5 mg/kg). During the last 3 days of treatment, the indicated oligonucleotides (1 µl, 2 µM) were injected every 36 h by stereotaxis into the raphe or the locus coeruleus. A decrease of miR-16 in the raphe or an increase of miR-16 in the locus coeruleus both alleviated the effect of fluoxetine on immobility time in the FST (C). Locomotor activity was unaffected whatever the treatment (D). The data are the means ± SEM (n=6 to 10 mice per group). *P<0.05 and ** P<0.01 vs control; § P<0.05 vs scramble.

14

Fig. S10. MiR-16 targeting SERT is a new actor mediating the adaptive response of serotonergic and noradrenergic neurons to fluoxetine treatment. MiR-16 is differentially expressed in serotonergic (green) vs noradrenergic (blue) neurons. In raphe, fluoxetine augments miR-16 level by increasing GSK3β activity. This up-regulation of miR-16 expression leads to a reduction of SERT expression. Serotonergic neurons additionaly respond to fluoxetine treatment by releasing S100β. This neurotrophic factor subsequently acts on the locus coeruleus: it decreases GSK3β activity, down-regulates the level of miR-16 and thus unlocks serotonergic functions in noradrenergic cells.

15

Supplementary References S1. S. Mouillet-Richard et al., Regulation by neurotransmitter receptors of serotonergic

or catecholaminergic neuronal cell differentiation. J Biol Chem 275, 9186-9192 (2000).

S2. S. Mouillet-Richard et al., Signal transduction through prion protein. Science 289, 1925-1928 (2000).

S3. J. M. Launay, B. Schneider, S. Loric, M. Da Prada, O. Kellermann, Serotonin transport and serotonin transporter-mediated antidepressant recognition are controlled by 5-HT2B receptor signaling in serotonergic neuronal cells. Faseb J 20, 1843-1854 (2006).

S4. S. Mouillet-Richard et al., Prions impair bioaminergic functions through serotonin- or catecholamine-derived neurotoxins in neuronal cells. J Biol Chem 283, 23782-23790 (2008).

S5. G. Paxinos, K. B. J. Franklin, The mouse brain in stereotaxic coordinates (Academic Press, San Diego, ed. 2, 2001).

S6. R. D. Porsolt, A. Bertin, M. Jalfre, Behavioral despair in mice: a primary screening test for antidepressants. Arch Int Pharmacodyn Ther 229, 327-336 (1977).

S7. A. Surget et al., Corticolimbic transcriptome changes are state-dependent and region-specific in a rodent model of depression and of antidepressant reversal. Neuropsychopharmacology 34, 1363-1380 (2009).

S8. L. Santarelli et al., Requirement of hippocampal neurogenesis for the behavioral effects of antidepressants. Science 301, 805-809 (2003).

S9. P. Willner, A. Towell, D. Sampson, S. Sophokleous, R. Muscat, Reduction of sucrose preference by chronic unpredictable mild stress, and its restoration by a tricyclic antidepressant. Psychopharmacology (Berl) 93, 358-364 (1987).

Supporting Online Material www.sciencemag.org Materials and Methods Figs. S1 to S10 References