The ANKK1 Gene Associated with Addictions Is Expressed in Astroglial Cells and Upregulated by Apomorphine

ARCHIVAL REPORTS

The ANKK1 Gene Associated with Addictions Is Expressed in Astroglial Cells and Upregulated by Apomorphine Janet Hoenicka, Adolfo Quiñones-Lombraña, Laura España-Serrano, Ximena Alvira-Botero, Leonor Kremer, Rocío Pérez-González, Roberto Rodríguez-Jiménez, Miguel Ángel Jiménez-Arriero, Guillermo Ponce, and Tomás Palomo Background: TaqIA, the most widely analyzed genetic polymorphism in addictions, has traditionally been considered a gene marker for association with D2 dopamine receptor gene (DRD2). TaqIA is located in the coding region of the ANKK1 gene that overlaps DRD2 and encodes a predicted kinase ANKK1. The ANKK1 protein nonetheless had yet to be identified. This study examined the ANKK1 expression pattern as a first step to uncover the biological bases of TaqIA-associated phenotypes. Methods: Northern blot and quantitative reverse-transcriptase polymerase chain reaction analyses were performed to analyze the ANKK1 mRNA. To study ANKK1 protein expression, we developed two polyclonal antibodies to a synthetic peptides contained in the putative Ser/Thr kinase domain. Results: We demonstrate that ANKK1 mRNA and protein were expressed in the adult central nervous system (CNS) in human and rodents, exclusively in astrocytes. Ankk1 mRNA level in mouse astrocyte cultures was upregulated by apomorphine, suggesting a potential relationship with the dopaminergic system. Developmental studies in mice showed that ANKK1 protein was ubiquitously located in radial glia in the CNS, with an mRNA expression pick around embryonic Day 15. This time expression pattern coincided with that of the Drd2 mRNA. On induction of differentiation by retinoic acid, a sequential expression was found in human neuroblastoma, where ANKK1 was expressed first, followed by that of DRD2. An opposite time expression pattern was found in rat glioma. Conclusions: Spatial and temporal regulation of the expression of ANKK1 suggest an involvement of astroglial cells in TaqIA-related neuropsychiatric phenotypes both during development and adult life.

Key Words: Addictions, ANKK1, astrocytes, astroglia, DRD2, polymorphism, radial glia, TaqIA

T

he Taq IA single-nucleotide polymorphism (SNP; rs1800497), located 10.541 bp downstream of the termination codon of the dopamine D2 receptor gene (DRD2; chromosome 11q22– q23), is the most studied genetic variant in a broad range of psychiatric conditions and personality traits (1). The TaqIA polymorphism consists of a single nucleotide C/T change; the two alleles are referred to as A2 (C) and A1 (T). Blum et al. (2) first reported the association between alcoholism and the TaqIA A1 allele and the A1⫹ genotype (hetero- or homozygous for A1). Numerous other studies have related the A1 allele to alcohol addiction (1,3), although others could not replicate this association (1,4). Two meta-analyses of Caucasian alcoholics and control subjects nonetheless support a link between the A1 allele and alcoholism (5,6). The TaqIA SNP has also been related to a variety of addictions and impulsive disorders

From the Department of Psychiatry (JH, LE-S, XA-B, RP-G, RR-J, MAJ-A, GP, TP), Hospital Universitario 12 de Octubre; Centro de Investigación Biomédica en Red de Salud Mental (CIBERSAM) (JH, AQ-L, RR-J, MAJ-A, GP, TP), ISCIII; Protein Tools Unit (LK), Centro Nacional de Biotecnología/ CSIC, Campus de Cantoblanco, Madrid, Spain. Authors LE-S and XA-B contributed equally to this work. Address correspondence to Janet Hoenicka, Ph.D., Laboratory of Neurosciences, Department of Psychiatry, Hospital Universitario 12 de Octubre, Avenida Andalucía SN, Madrid 28041, Spain; E-mail: jhoenicka@ gmail.com. Received Feb 18, 2009; revised Jul 31, 2009; accepted Aug 16, 2009.

0006-3223/10/$36.00 doi:10.1016/j.biopsych.2009.08.012

(1), pharmacogenetics of addictions (7), psychoses (8), and antisocial traits in alcoholics (9,10). A number of dopamine-related endophenotypes are associated with the A1 allele (1,11,12). Of these, a reduction in striatal dopamine D2 receptors for the A1 allele carriers was reported in several studies (1,13), which could explain the association with several neuropsychiatric disorders and learning processes (1,14). It is now known that the TaqIA SNP is located in exon 9 of the ankyrin repeat and kinase domain-containing 1 (ANKK1) gene, where it produces a Glu713-to-Lys (E713K) substitution (15). A positron emission tomography (PET) study of healthy volunteers associated the functional SNP C957T (rs6277) in DRD2 exon seven with low D2 receptor binding potential, a trait previously associated with the nearby TaqIA SNP (16). ANKK1 variants might nonetheless act in the dopaminergic system, which could provide an alternative explanation for some previously reported TaqIA-associated traits. For example, a TaqIA effect on striatal dopamine synthesis, independent of the C957T DRD2 genotype, was reported in a PET study (17). Extensive genotyping of DRD2 and ANKK1 genes suggests that the association between the locus 11q22– q23 and substance dependence is also due to ANKK1 variants (18 –21). Psychopathic traits in alcoholic patients are related to an epistatic interaction between the ANKK1 TaqIA and DRD2 C957T SNP genotypes (22). ANKK1 belongs to the receptor-interacting protein (RIP) serine/threonine kinase family (23). The RIP kinases are important regulators of cell proliferation and differentiation and initiate responses to various environmental factors by activating transcription factors such as NF-␬B or AP-1 (23). cDNA sequence data analysis predicted a 765-amino-acid ANKK1 protein very similar to RIP4, sharing not only the N-terminal kinase domain BIOL PSYCHIATRY 2010;67:3–11 © 2010 Society of Biological Psychiatry

4 BIOL PSYCHIATRY 2010;67:3–11 but also the C-terminal ankyrin repeats (15). A study of ANKK1 gene expression describes low mRNA levels in placenta and spinal cord and its absence in whole brain homogenates (15); nonetheless, the ANKK1 protein has yet to be reported. The characterization of ANKK1 would clarify our understanding of the physiology underlying the TaqIA-associated phenotypes.

Methods and Materials Animals and Sample Preparation BALB/c mice (20 –30 g) and Wistar rats (180 –200 g); both from Harlan Interfauna Ibérica, Spain, were used as protein and tissue source. Experimental procedures were conducted in accordance with European Communities Council Directive 86/609/EEC). For immunohistochemistry, eight BALB/c mice and eight Wistar rats were sacrificed under anesthesia (5% halothane), with transcardial perfusion using 40 mL heparinized saline (1000 U/mL in .15 mol/L NaCl), followed by 40 mL 4% paraformaldehyde (PFA) in phosphate buffer (PB) .1 mol/L. Brains were removed, postfixed in 4% PFA (2 hours), and cryoprotected in 10%, 20%, and 30% sucrose PB .1 mol/L (4°C). Tissue sections (35 ␮m) were cut on a cryostat (CM 1900; Leica Microsystems, Wetzler, Germany). For E14.5 mouse embryo analysis, pregnant dams were anesthetized and killed by cervical dislocation before embryos were dissected into cold phosphate-buffered saline (PBS), OCT-embedded, and cut on a cryostat. Slide-mounted human brain tissue sections from five independent control male samples were kindly provided by the Banco de Tejidos para Investigaciones Neurológicas (Madrid, Spain). Northern Blotting A premade blot of total normal human brain RNA (N1234445) was purchased from the BioChain Institute. The probe used was a human ANKK1 gene polymerase chain reaction (PCR) fragment of 92 bp (Table S1 in Supplement 1). Radiolabeled 32P-DNA probe was made using the HexaLabel plus DNA labeling kit (Fermentas, Amherst, New York). Hybridization was performed in PerfectHyb plus hybridization buffer 1X (Sigma, St. Louis, Missouri) at 42°C following conventional protocols. Quantitative mRNA Estimation by Real-Time Reverse Transcriptase PCR Total RNA was isolated from cultured cells with Trizol reagent (Invitrogen, Carlsbad, California) and purified (RNeasy Protect Mini Kit, Qiagen, Valencia, California). We also used RNA from two commercial panels of human and mouse (Human Total RNA Panel IV and Mouse Total RNA master panel, BD Biosciences Clontech, Palo Alto, California). RNA quality and yield were assessed by Nanodrop-1000 spectrophotometer and stored at ⫺80°C. PolyT primers were used for the reverse transcriptase (RT) reaction using Superscript III Reverse Transcriptase Kit (Invitrogen). All quantitative real-time PCR reactions were carried out in triplicate on an ABI Prism 7500 sequence detection system (Applied Biosystems, Foster City, California). Gene expression was analyzed using SyBR Green Master Mix (Roche, Diagnostics, Indianapolis, Indiana). Thermal cycler conditions were 2 min at 50°C, 10 min at 95°C, followed by 30 sec at 95°C (denaturation), 30 sec at 58 – 60°C (annealing), and 30 sec at 72°C (extension) for 40 cycles. The total volume for each reaction was 15 ␮L, comprising 5 ␮L diluted cDNA (10 ng/␮L), 7.5 ␮L Roche SyBR Green Master Mix, 1.6 ␮L double-distilled water, and .45 ␮L of each primer (10-mmol/L concentration). Primers used to test www.sobp.org/journal

J. Hoenicka et al. human, mouse, and rat samples are shown in Table S1 in Supplement 1. Samples were quantified by the standard-curve method (24) and normalized to the GAPDH (glyceraldehyde 3-phosphate dehydrogenase) housekeeping gene in the same cDNA sample. Results were analyzed with SDS 2.1 software (Applied Biosystems). Baseline values of amplification plots were set automatically, and threshold values kept constant to obtain normalized cycle times and linear regression data. Western Blot Total protein extracts were obtained after homogenization in lysis buffer (150 mmol/L NaCl, 20 mmol/L Tris-HCl, 5 mmol/L ethylenediamine tetraacetate, 10% glycerol, 1% NP-40) containing a protease inhibitor cocktail (Complete Mini-Protease Inhibitor Cocktail, Roche). To obtain membrane and cytosolic proteinenriched fractions, brain tissue was homogenized in PBS with protease inhibitors and centrifuged (1000 g, 10 min) to eliminate cell debris, nuclei, and the largest mitochondria. Supernatant was centrifuged (60,000 g, 2 hours, 4°C). The final supernatant, used as the cytosolic fraction, was precipitated with 100% (w/v) trichloroacetic acid (TCA) at 1:4 TCA: protein sample ratio and rinsed twice in ⫺20°C acetone. The pellet was used as the membrane fraction. To obtain both detergent-soluble and insoluble membrane proteins, membranes were extracted with 1% Triton x-100 (Pharmacia Biotech, Freiburg, Germany) and centrifuged at 60,000 g for 2 hours at 4°C. Equal amounts of protein samples (75 ␮g) quantified by bicinchoninic acid protein assay (Pierce, Rockford, IL), were resolved in 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes (Protran, Whatman, Dassel, Germany). ANKK1 proteins were detected using two rabbit polyclonal antibodies (Ab; STk Ab for human, mouse and rat, and mouse-specific STk2 Ab) raised against two peptides corresponding to the N-terminal of the predicted ANKK1 ORF (see Methods in Supplement 1). A 1:600 dilution of STk Ab (2 hours), followed by goat anti-rabbit horseradish peroxidase– conjugated secondary antibody (Jackson ImmunoResearch, West Grove, Pennsylvania; 1 hour). Proteins were processed for chemiluminescence with the ECL Detection System (Amersham, Piscataway, New Jersey). Immunohistochemistry Adjacent sections from one animal of each species were Nissl-stained to delimit brain structures. Sagittal and horizontal mouse embryos were hematoxylin-stained to visualize the central nervous system. Free-floating or slide-mounted tissue sections were washed in PB .1 mol/L, pH 7.4 followed by PBS, and incubated (2 hours) in blocking buffer (PBS, 10% normal goat serum, .5% bovine serum albumin and .1% Triton x-100). All incubations were at room temperature and were followed by several rinses in PBS. Tissue sections were incubated in a solution containing primary antibodies (⬃40 hours, 4°C). For human brain, slide-mounted tissue sections were brought to RT, washed with PBS (5 min), blocked, and incubated with primary antibody (4°C, 12 hours). ANKK1 localization was determined by incubation with polyclonal anti-ANKK1 (STk or STk2) antibodies, followed by goat anti-rabbit Texas Red (1:250, 2 hours; Jackson ImmunoResearch). To identify neurons, astrocytes, oligodendrocytes, and radial glia, we used the following mouse monoclonal antibodies (mAb): antibeta-III tubulin (TUBB3; G712A, Promega), -glial fibrillary acidic protein (GFAP; G3893, Sigma), -myelin/oligodendrocyte-specific protein (MOSP; MAB328, Millipore, Bedford,

J. Hoenicka et al.

BIOL PSYCHIATRY 2010;67:3–11 5

Figure 1. Multiple alignment analysis of predicted amino acid sequences for ANKK1 in Homo sapiens, Mus musculus, and Rattus norvegicus. Shading indicates 100% amino acid sequence identity. The domains are underlined. Polymorphic sites are indicated by 夝. (A) Comparison of the human ANKK1-kinase domain (NP_848605) with a predicted ANKK1kinase from mouse (AceView) and rat (NP_001102469). Degree of identity: H. sapiens-M. musculus: 83.4%; H. sapiens-R. norvegicus: 83%; M. musculus-R. norvegicus: 93.6%. The lysine that binds ATP (
), and the aspartic acid that accepts the proton during phosphotransfer (Œ) are highlighted. Boxes span kinase submotifs. Phylogenetically conserved lysine residues are in bold. (B) Comparison of the predicted human (EAW67217), mouse (AceView), and rat (XP_001073947) ANKK1-ankyrin proteins. Degree of identity: H. sapiens-M. musculus: 82.5%; H. sapiens-R. norvegicus: 80.5%; M. musculus-R. norvegicus: 91.2%.

Massachusetts) and RC2 (Hybridoma Bank, Iowa City, Iowa). After washing in PBS, sections were incubated in the dark (RT, 2 hours) in blocking buffer diluted 1:3 in PBS containing Alexa Fluor goat anti-mouse 488 (1:800; Invitrogen). Sections were rinsed and mounted on gelatin-coated glass slides (G2500; Sigma), air dried, dehydrated in a graded ethanol series, defatted in xylene (30 min), and coverslipped with DePeX mounting medium (VWR International). Sequential scanning images were captured on a Zeiss LSM 510 Meta scanning laser confocal microscope. Cell Culture and Treatment Mouse astrocytes were isolated and maintained as described (25). These cells were treated with the DA receptor D2 antagonist sulpiride [S112 S- (-)] and apomorphine [A4393 (R-)]; both from Sigma). Sulpiride was applied to cultures 2 hours before apomorphine addition (both at 100 ␮mol/L in serum-free medium). Apomorphine treatment was for 6 hours; cells were then trypsinized and RNA was isolated as described earlier.

Human SK-N–SH neuroblastoma and rat C6 glioma cell lines were kindly provided by Drs. R.M. Sainz and J.C. Mayo (Instituto Universitario de Oncología, Oviedo, Spain). To induce cell differentiation, all trans-retinoic acid (ATRA; Sigma) was used. The cell lines were treated for 5 days, with daily addition of fresh complete media with ATRA. In Silico Promoter Analysis To identify transcription factor binding cis elements of ANKK1 and DRD2 genes, we used the Matinspector Web-based search algorithm available from Genomatix Software (26). The search algorithm is described in detail at http://www.genomatix.de. Statistical Analysis Normal distribution of data was tested using the Kolmogorov– Smirnov test. The cell treatment experiments were analyzed by using Student’s t or Mann–Whitney U test. Data are expressed as the mean ⫾ standard deviation (SD) and values of p ⬍ .05 were considered significant. www.sobp.org/journal

6 BIOL PSYCHIATRY 2010;67:3–11

J. Hoenicka et al.

Figure 2. Expression study of the ANKK1 gene. (A) Northern blot analysis of human brain areas: FT, frontal lobe; TL, temporal lobe; PL, parietal lobe; OL, occipital lobe. Arrows indicate bands and molecular weights. A cDNA probe of human 18 S ribosomal RNA was used as control. (B) Reverse transcriptase polymerase chain reaction analysis of mouse Ankk1 mRNA from different tissues. Relative mRNA values (⫾SD) are shown for three independent assays; values were obtained and normalized as described in Methods and Materials. (C) Western blot analysis of mouse tissue extracts. The 40-kDa band was quantified using the ImageJ 1.37v software. The protein level in heart was assigned a value of 100, followed by spinal cord (83.73% ⫾ 5.8%), brain (78.97% ⫾ 7.71%), bone (40.33% ⫾ 3.02%), liver (39.6% ⫾ 3.21%), 5 day mouse brain (27.03% ⫾ 3.69%), lung (26.27% ⫾ 3.31%), and spleen (15.97% ⫾ 3.5%). (D) Western blot analysis of total and subcellular fractions of mouse brain and (E) of cytosol- and membrane-rich brain fractions treated with triton x-100 (4°C) to separate detergent-soluble (DSM) from detergent-resistant membranes (DRM). RE, relative expression.

Results In Silico Evidence for Three ANKK1 Isoforms We first integrated database information for ANKK1 cDNA clones and predicted peptides for man, mouse and rat (Results and Table S2 in Supplement 1). ANKK1 transcription would produce at least three proteins: ANKK1 (with RIP kinase and ankyrin repeat domains), ANKK1-kinase (only the RIP kinase domain), and ANKK1-ankyrin (only the ankyrin repeat domain). We compare predicted ANKK1-kinase and ANKK1-ankyrin proteins for the three species (Figure 1A and 1B). The ANKK1kinases are nearly identical and have conserved modules for the RIP kinase motive (23). Comparison of the polymorphic residues in human ANKK1 with the mouse and rat proteins shows a shared alanine at position Thr239Ala (rs7118900) in the kinase domain. The ANKK1-ankyrin proteins differ in the number of ankyrin domains, with human polymorphisms at Arg318Gly (rs11604671), Arg442Gly (rs4938016), Arg490His (rs2734849), and Glu713Lys (rs18000497, TaqIA). The putative expression of three isoforms suggests distinct biological functions for these proteins. The kinase domain of RIP4, the protein most closely related to ANKK1 (23), activates NF-␬B signaling as efficiently as does the full-length protein (27), whereas the RIP4 ankyrin repeat domain inhibits kinase activity during apoptosis (27). ANKK1 Is Expressed in the Central Nervous System and in Other Tissues To determine whether ANKK1 is expressed in the human central nervous system (CNS), we performed Northern blot hybridization of samples from human brain cortex. We detected www.sobp.org/journal

ANKK1 transcripts of approximately 5.0, 2.6, and .78 kilobases (kb), with a stronger signal in frontal and parietal lobes than in temporal and occipital lobes (Figure 2A). We then used quantitative RT-PCR analysis of total RNA to study the human ANKK1 expression pattern in other tissues, amplifying a 204-bp fragment from exons 4 and 5, which are putatively expressed in all three predicted ANKK1 proteins. The ANKK1 mRNA level in brain was assigned a value of 100, followed by medulla (95.15 ⫾ 9.16), fetal liver (93.92 ⫾ 15.96), fetal brain (91.57 ⫾ 6.37), cerebellum (88.61 ⫾ 5.67), and placenta (85.75 ⫾ 3.21). A similar RT-PCR study of Ankk1 in 11 murine tissues showed constitutive transcripts in all tissues assayed (Figure 2B). To study ANKK1 protein expression, we performed Western blot analysis of different mouse tissues using the polyclonal STk antibody (Ab). We found bands with qualitative and quantitative tissue-specific differences (Figure 2C). Adult whole-brain samples showed a broad 49- to 67-kDa band and a 40-kDa band, whereas newborn mouse brain showed a reduction in the 40-kDa band signal relative to the broad band. The adult whole-brain pattern was found in several mouse CNS areas including prefrontal cortex, hippocampus, corpus callosum, thalamus, bulb, pons, mesencephalon, encephalic trunk, basal ganglia, cerebellum, and spinal cord (not shown). Comparison of mouse, rat, and human brain tissue in Western blot showed no differences among species (not shown). We detected no band compatible with the predicted longest ANKK1 isoform in any of the samples. The 40-kDa band coincided with the expected molecular weight of the predicted rat ANKK1-kinase; the 49- to 67-kDa bands could result from posttranslational modification of the 40 kDa ANKK1-kinase

J. Hoenicka et al.

BIOL PSYCHIATRY 2010;67:3–11 7

Figure 3. ANKK1 is expressed in astrocytes. (A) Top: confocal optical sections showing GFAP (green) and ANKK1kinase (red) colocalization in mouse astrocytes from mouse medulla oblongata (top row; inset in left panel shows sagittal diagram indicating stained section). Boxed area in merged image is amplified (right box). Middle: distribution of TUBB3 (green) in neurons and ANKK1-kinase (red) in astrocytes in mouse; these markers did not colocalize. Bottom: distribution of MOSP (green)-positive oligodendrocytes and ANKK1-kinase (red)-positive astrocytes in mouse brain; colocalization was not detected. Scale bars ⫽ 50 ␮m; inset scale bars ⫽ 25 ␮m. Gi, gigantocellular reticular nucleus; IO, inferior olive; Rob, raphe obscurus nucleus; py, pyramidal tract. (B) Confocal optical section of the rat caudate-putamen (Cpu) and corpus callosum (cc) indicates colocalization of GFAP and ANKK1kinase. Inset shows coronal diagram of the stained section. Scale bar ⫽ 50 ␮m. (C) Colocalization of GFAP and ANKK1-kinase in the molecular layer of human cerebellum as detected by confocal microscopy. Scale bar ⫽ 50 ␮m (left); inset scale bar ⫽ 12.5 ␮m (right). (D) In silico MatInspector analysis of a 5= upstream sequence of ANKK1 conserved in man, mouse, and rat, showing cis elements for specific GFAP expression in astrocytes.

protein. ANKK1-kinase, like other RIP kinases (28), might also be ubiquitinated on conserved lysine residues (Figure 1A). Because some RIP kinases have both cytosolic and membrane-associated forms (29), we performed subcellular fractionation of mouse brain before Western blot analysis with the STk Ab. We found the 40-kDa isoform in the membrane fraction, and the 49- to 67-kDa bands in both cytosolic and membrane-rich fractions (Figure 2D). The absence of the 40-kDa band in the cytosolic fraction indicated ANKK1 association with cell membranes or submembrane cytoskeletal elements. Treatment of tissue membranes with 1% Triton x-100 revealed three bands of 49, 60, and 67 kDa in the detergent-soluble fraction, but not in the insoluble fraction, which showed only the 40-kDa band (Figure 2E). The STk Ab thus identified four bands in mouse brain extracts, three of which were soluble or loosely membraneassociated and one in detergent-insoluble membrane. In the CNS, ANKK1-Kinase Is Expressed Exclusively in Astrocytes We next investigated the cellular location of ANKK1 in the CNS by double immunolabeling with STk Ab and cell-specific marker monoclonal antibodies. The STk Ab signal colocalized exclusively with GFAP-positive astrocytes (Figure 3A), a pattern observed in all mouse brain areas as well as in rat and human brain sections (Figure 3B and 3C). ANKK1 and GFAP were not

homogeneously distributed within astrocyte cell bodies and processes (Figure 3A, merged images), and their relative distribution varied in different brain areas (not shown). These differences may be due to the diversity of physiologic roles of cells in the astrocyte lineage. In view of the ANKK1 cellular location, we used the MatInspector program (26) to compare sequence fragments 5 kb upstream of human ANKK1 and GFAP. We found similar genomic structure with an overall identity of approximately 50% (not shown). Human, mouse, and rat ANKK1 5= sequences were also compared. They shared putative cis elements for GATA, IK2, Chop-C/EBP, and NF-␬B binding (Figure 3D), all of which are necessary for astrocyte-specific expression of the GFAP gene (30). Homologous region in rat and mouse promoters, covering ⫺218 to ⫺136 from ATG was also identified. The rat promoter showed a distal C/EBP site at ⫺670. In mouse, we identified a high density of NF-␬B and C/EBP binding sites between ⫺649 and ⫺619 from ATG. Potential Functional Relationship Between ANKK1 and DRD2 Genes The proximity of ANKK1 and DRD2 genes (whose 3= ends overlap; Figure 4A) is compatible with functional interaction. There is also evidence that ANKK1 and DRD2 act in a coordinated manner (22); it nonetheless remains unclear whether one or both of these genes are responsible for the psychiatric www.sobp.org/journal

8 BIOL PSYCHIATRY 2010;67:3–11

J. Hoenicka et al.

Figure 4. Relationship between ANKK1 and DRD2 genes. (A) The seven haplotype blocks identified in Caucasians in a 108-kb fragment of the locus in which DRD2 and ANKK1 are located. The 18 SNP captured by multimarker tagging are shown. (B) Comparison of ANKK1 and DRD2 5= upstream sequences, showing the Kruppel-like cis elements KLF4, CPBP/KLF6, MZF, and ZBP-89, as well as PPAR/RXR, NF–KB, NFAT, MYC/MAX, MAZ, POZ, SP1, AP-2, TCF1, and TCF/LEF1. Shared cis elements are underlined (C) Ankk1 and Drd2 expression analysis by reverse transcriptase polymerase chain reaction after apomorphine treatment of mouse astrocytes, showing strong upregulation (Ankk1: p ⬍ .001; Drd2: p ⫽ .006). **p ⬍ .01; ***p ⬍ .001. RE, relative expression.

phenotypes and neuropsychologic traits observed in association with the TaqIA SNP. Using Haploview 2.0, we analyzed SNP genotype Hapmap data (http://www.hapmap.org/) within the DRD2 and ANKK1 genes for Caucasian populations. We identified one shared haplotype block (Block 2) that extends from ANKK1 intron three to DRD2 intron 7 (Figure 4A). Multimarker tagging analysis identified a three SNP marker corresponding to the combination of the minor alleles of rs11604671 (allele T), rs1800497 (TaqIA, allele A1: T) SNP in ANKK1, and rs493807 (allele G) SNP in DRD2. This TTG marker captures 17 DRD2 SNP, including the functional rs2283265 and rs1076560 SNP that affect DRD2 splicing (31), indicating that TaqIA is a marker of DRD2 expression variants (Figure 4A). In Block 2, we also observed strong linkage disequilibrium (r2 ⫽ .902) between the ANKK1 rs7118900 SNP (Thr239Ala) and TaqIA, suggesting that this SNP is a marker for variations in the kinase activity of ANKK1 protein. The threonine at position 239 creates an additional phosphorylation consensus site, whose phosphorylation could modify ANKK1 function in astrocytes. The TaqIA SNP genotype might thus tag functional differences in ANKK1 in astrocytes and D2 receptor in astrocytes and neurons. To determine whether ANKK1 and DRD2 share a transcriptional regulatory pathway, we used the MatInspector program (26) to search for cis elements in a 1-kb region 5= upstream. The promoters shared a number of putative consensus binding sites for transcriptional factors inwww.sobp.org/journal

volved in cell growth, proliferation, differentiation, embryogenesis, and survival (Figure 4B). To examine further the potential relationship between ANKK1 and the D2 dopamine receptor in vitro, we studied the Ankk1 response to apomorphine, a dopamine receptor agonist. We treated mouse astrocytes and found significantly increased Ankk1 expression (3.58 ⫾ .56-fold, t ⫽ 6.05, df ⫽10, p ⬍ .001). To clarify whether the effect of apomorphine on Ankk1 expression is due specifically to D2 receptor apomorphine induction, we treated astrocytes with sulpiride, a selective D2 receptor antagonist, followed by apomorphine. This treatment also increased Ankk1 expression (2.05 ⫾ .21-fold; Figure 4C). This Ankk1 upregulation was lower than that observed using apomorphine alone but remains significant compared with untreated astrocytes (t ⫽ 7.9, df ⫽ 9, p ⬍ .001). Treatment with sulpiride alone had no effect on Ankk1 (1.1 ⫾ .09-fold; t ⫽ .487, df ⫽ 9, p ⫽ .636; Figure 4C). Although these results are preliminary, they suggest that Ankk1 gene expression, and the dopaminergic system may be functionally linked in mouse astrocytes. ANKK1-Kinase Is Developmentally Regulated and Expressed in Radial Glia Given that both RIP4 and RIP2 are regulators of differentiation processes (32,33), we next studied Ankk1 expression in mouse samples from embryonic Days (E) 7, 11, 15, and 17 using

BIOL PSYCHIATRY 2010;67:3–11 9

J. Hoenicka et al.

Figure 5. ANKK1 study during embryonic development. (A) Reverse transcriptase polymerase chain reaction analysis of Ankk1 and Drd2 at four mouse embryonic stages showed significant differences; Ankk1 (p ⬍ .001) and Drd2 (p ⬍ .001). (B and C) In vitro differentiation studies upon treatment with all-trans retinoic acid for 5 days. (B) Human neuroblastoma SK-N-SH cell line: ANKK1 showed early upregulation at day 1 (p ⬍ .001) and a moderate decrease by Day 5 (p ⫽ .036). DRD2 expression was upregulated at Day 3 of treatment (p ⬍ .001), followed by a decrease by Day 5 (p ⫽ .048). (C) The rat C6 glioma cell line showed an upregulation of Drd2 at Day 1 postinduction (p ⬍ .02) and of Ankk1 at Day 3 (p ⬍ .01), followed by a slight decline in the expression of both genes at Day 5 (Drd2: p ⫽ .84; Ankk1: p ⫽ .09). Values are shown ⫾ SD (n ⫽ 3). *p ⬍ .05; ***p ⬍ .001. RE, relative expression. (D) Confocal optical sections showing DAPI (blue), RC2 (green), and ANKK1-kinase (red) colocalization in embryoc radial glia from developing medulla (inset in left panels shows horizontal diagram indicating stained section) and olfactory epithelium (bottom). Scale bars ⫽ 50 ␮m.

RT-PCR. Ankk1 expression increased from E7 and peaked at E15, followed by almost complete loss of expression by E17 (KruskallWallis ␹2: 34.76, df ⫽ 3, p ⬍ .001) (Figure 5A). A parallel study revealed that the Drd2 gene also peaked at E15 (Figure 5A). Thus, the expression of both Ankk1 and Drd2 may be regulated during neurogenesis, a process which has been shown to peak around E14 in the mice telencephalon (34). We then analyzed ANKK1 and DRD2 gene expression in human SK-N-SH neuroblastoma treated with retinoic acid, an in vitro model for neuron differentiation. We observed upregulation of ANKK1 at Day 1 (t ⫽ 6.012, df ⫽ 7.193; p ⬍ .001) and DRD2 at Day 3 postinduction (t ⫽ 5.991, df ⫽ 9.551; p ⬍ .001), followed by a slight decline in expression of both genes at day 5 (ANKK1: t ⫽ 2.418, df ⫽ 10.218; p ⫽ .036; DRD2: t ⫽ 2.14, df ⫽ 16.292; p ⫽ .048; Figure 5B). Given that ANKK1 is expressed in astrocytes we also studied the expression of these two genes in the rat C6 glioma cell line on treatment with retinoic acid. In this case, we observed upregulation of Drd2 at Day 1 postinduction (p ⬍ .02) and of Ankk1 at Day 3 (p ⬍ .01), followed by a slight decline in the expression of both genes at Day 5 (Drd2 p ⫽ .84; Ankk1: p ⫽ .09; Figure 5C). We found an opposite time pattern for the induction of these two genes between the two cell lines. These results are consistent with a temporally coordinated regulation of ANKK1 and DRD2 expression in development. To study the prenatal histological location of ANKK1, we analyzed the developing CNS in horizontal and sagittal sections obtained from mouse embryos at E14.5, a developmental stage at which the Ankk1 reached high-level expression. We found with the mouse-specific STk2 Ab that ANKK1 was expressed in the developing spinal cord, hindbrain, olfactory bulb, midbrain, and forebrain (not shown). ANKK1-positive cells commonly displayed long and fine processes that pass through the neuroepithelial layer and terminate in small conical end feet at the pial surface, matching the morphology described for radial glial cells. To confirm ANKK1 expression in radial glial cells, double immunolabeling to colocalize STk2 Ab and RC2 (a nestin marker specific for radial glia cells) was performed. We observed a complete overlap of the STk2 Ab and RC2 signals (Figure 5D).

Discussion Although the TaqIA ANKK1 SNP has been studied in a wide variety of psychiatric conditions, especially addictive and antisocial disorders, the ANKK1 protein remained to be identified. Here we describe, for the first time, a biological plausibility for the TaqIA-associated phenotypes. On the basis of sequence and database information, the putative ANKK1 protein was initially proposed as a 85-kDa protein (15); analyses of expressed sequence tags collections also predict shorter protein isoforms (35–37). Our RNA and protein studies showed that ANKK1 is constitutively expressed in CNS and its 40-kDa product localized to various subcellular compartments. In human, rat, and mouse CNS, we observed that the ANKK1kinase is expressed exclusively in astrocytes. The evidence about ANKK1-kinase adult brain location gives birth to a new perspective on our understanding of the pathophysiology underlying the TaqIA-associated phenotypes, because it challenges the view that only neurons are involved and assigns a new and active role to the astroglia contributing to the regulation of this genetic link. Bidirectional communication between astrocytes and neurons has been demonstrated, and astrocytes are assumed to be an integral part of the synapses (38). These glial cells are implicated in the pathophysiology of learning processes and the rewarding effects of substance of abuse (39). Therefore, the regulation of information processing by astrocytes might contribute to the TaqIA-associated phenotypes. In addition, in mice we found that ANKK1 is located in the radial glia of the developing CNS. Radial glial cells are the major source of cortical neurons in rodents (40), and they constitute the radial system across the whole developing CNS that guides migrating neurons (41). Moreover, most radial glial cells transdifferentiate to mature astrocytes after completion of the neuronal migration (42). In humans, there is also direct evidence that radial glia serve as neuronal progenitors in the cerebral cortex (43). Hence, it is tempting to consider that polymorphisms of the genes involved on the function of the radial glia may be related to subtle changes in the cytoarchitecture of the brain, which are responsible for individual differences www.sobp.org/journal

10 BIOL PSYCHIATRY 2010;67:3–11 in the brain function. Given the localization of ANKK1 in both adult and developing CNS, it is plausible that multiple mechanisms may contribute to the association of the well-studied TaqIA ANKK1 SNP with different psychiatric conditions and neuropsychological phenotypes. Our study suggests a potential relationship between ANKK1 and the dopaminergic system. ANKK1 and DRD2 genes overlap and share haplotypic blocks, and their promoters have identical cis elements for transcriptional regulation. In addition, our in vitro studies showed that ANKK1 is activated by apomorphine, a dopaminergic agonist. Finally, ANKK1 and DRD2 expression patterns in mouse embryonic samples, human neuroblastoma, and rat glioma, suggest that the interaction of these proteins may be relevant for CNS development, including neurogenesis and gliogenesis. For instance, in mice, both the neurogenesis pick and the beginning of the differentiation of radial glia into GFAP-positive cells occur at E15 (44). In conclusion, we are providing the first direct evidence of ANKK1 location in adult and developing CNS, which assigns a potential role for the astroglial cells in TaqIA-associated phenotypes. The evidence that TaqIA is tagging both ANKK1 and DRD2 functional gene variants represents a broader scenario to help understanding neuropsychiatric disorders associated with this SNP. The repercussion of such variants on the role of astrocytes and the radial glia is compelling and shall be addressed in future works.

We thank Francesc Palau for his critical reading of the manuscript, Catherine Mark for editorial assistance, Raquel Cuadros for her technical assistance in the production of antibodies, and Begoña Santiago for helping us with confocal microscopy. This work was supported by the Fondo de Investigacíon Sanitaria, Instituto Salud Carlos III, Grant No. 05/0731 and 08/0529. CIBERSAM is an initiative of the Instituto de Salud Carlos III. Competing Interests: All authors declared no biomedical financial interests or potential conflicts of interest. Supplementary material cited in this article is available online. 1. Noble EP (2003): D2 dopamine receptor gene in psychiatric and neurologic disorders and its phenotypes. Am J Med Genet B Neuropsychiatr Genet 116:103–125. 2. Blum K, Noble EP, Sheridan PJ, Ritchie T, Montgomery A, Wood RC, et al. (1990): Allelic association of human dopamine D2 receptor gene in alcoholism. J Am Med Assoc 263:2055–2060. 3. Comings DE, Comings BG, Muhleman D, Dietz G, Shahbahrami B, Tast D, et al. (1991): The dopamine D2 receptor locus as a modifying gene in neuropsychiatric disorders. J Am Med Assoc 266:1793–1800. 4. Gelernter J, Southwick S, Goodson S, Morgan A, Nagy L, Charney DS (1999): No association between D2 dopamine receptor (DRD2) “A” system alleles, or DRD2 haplotypes, and posttraumatic stress disorder. Biol Psychiatry 45:620 – 625. 5. Munafo MR, Matheson IJ, Flint J (2007): Association of the DRD2 gene Taq1A polymorphism and alcoholism: A meta-analysis of case-control studies and evidence of publication bias. Mol Psychiatry 12:454 – 461. 6. Smith L, Watson M, Gates S, Ball D, Foxcroft D (2008): Meta-analysis of the association of the Taq1A polymorphism with the risk of alcohol dependency: A HuGE gene-disease association review. Am J Epidemiol 167:125–138. 7. David SP, Strong DR, Munafò MR, Brown RA, Lloyd-Richardson EE, Wileyto PE, et al. (2007): Bupropion efficacy for smoking cessation is influenced by the DRD2 Taq1A polymorphism: Analysis of pooled data from two clinical trials. Nicotine Tob Res 9:1251–1257.

www.sobp.org/journal

J. Hoenicka et al. 8. Dubertret C, Gouya L, Hanoun N, Deybach JC, Adès J, Hamon M, Gorwood P, et al. (2004): The 3= region of the DRD2 gene is involved in genetic susceptibility to schizophrenia. Schizophr Res 67:75– 85. 9. Ponce G, Jimenez-Arriero MA, Rubio G, Hoenicka J, Ampuero I, Ramos JA, et al. (2003): The A1 allele of the DRD2 gene (TaqI A polymorphisms) is associated with antisocial personality in a sample of alcohol-dependent patients. Eur Psychiatry 18:356 –360. 10. Hoenicka J, Ponce G, Jimenez-Arriero MA, Ampuero I, RodríguezJiménez R, Rubio G, et al. (2007): Association in alcoholic patients between psychopathic traits and the additive effect of allelic forms of the CNR1 and FAAH endocannabinoid genes, and the 3= region of the DRD2 gene. Neurotox Res 11:51– 60. 11. Jimenez-Arriero MA, Ponce G, Rodriguez-Jimenez R, et al. (2006): Taql-A polymorphism linked to the DRD2 gene and P300 in alcoholic patients. Eur J Psychiatry 20:45–53. 12. Rodriguez-Jimenez R, Avila C, Ponce G, Ibáñez MI, Rubio G, JiménezArriero MA, et al. (2006): The TaqIA polymorphism linked to the DRD2 gene is related to lower attention and less inhibitory control in alcoholic patients. Eur Psychiatry 21:66 – 69. 13. Jönsson EG, Nöthen MM, Grünhage F, Farde L, Nakashima Y, Propping P, Sedvall GC, et al. (1999): Polymorphisms in the dopamine D2 receptor gene and their relationships to striatal dopamine receptor density of healthy volunteers. Mol Psychiatry 4:290 –296. 14. Klein TA, Neumann J, Reuter M, Hennig J, von Cramon DY, Ullsperger M (2007): Genetically determined differences in learning from errors. Science 318:1642–1645. 15. Neville MJ, Johnstone EC, Walton RT (2004): Identification and characterization of ANKK1: A novel kinase gene closely linked to DRD2 on chromosome band 11q23.1. Hum Mutat 23:540 –545. 16. Hirvonen M, Laakso A, Nagren K, Rinne JO, Pohjalainen T, Hietala J (2004): C957T polymorphism of the dopamine D2 receptor (DRD2) gene affects striatal DRD2 availability in vivo. Mol Psychiatry 9:1060 –1061. 17. Laakso A, Pohjalainen T, Bergman J, Kajander J, Haaparanta M, Solin O, et al. (2005): The A1 allele of the human D2 dopamine receptor gene is associated with increased activity of striatal L-amino acid decarboxylase in healthy subjects. Pharmacogenet Genom 15:387–391. 18. Gelernter J, Yu Y, Weiss R, Brady K, Panhuysen C, Yang BZ, et al. (2006): Haplotype spanning TTC12 and ANKK1, flanked by the DRD2 and NCAM1 loci, is strongly associated to nicotine dependence in two distinct American populations. Hum Mol Genet 15:3498 –3507. 19. Dick DM, Wang JC, Plunkett J, Aliev F, Hinrichs A, Bertelsen S, et al. (2007): Family-based association analyses of alcohol dependence phenotypes across DRD2 and neighboring gene ANKK1. Alcohol Clin Exp Res 31: 1645–1653. 20. Yang BZ, Kranzler HR, Zhao H, Gruen JR, Luo X, Gelernter J (2007): Association of haplotypic variants in DRD2, ANKK1, TTC12 and NCAM1 to alcohol dependence in independent case control and family samples. Hum Mol Genet 16:2844 –2853. 21. Yang BZ, Kranzler HR, Zhao H, Gruen JR, Luo X, Gelernter J (2008): Haplotypic variants in DRD2, ANKK1, TTC12, and NCAM1 are associated with comorbid alcohol and drug dependence. Alcohol Clin Exp Res 32: 2117–2127. 22. Ponce G, Hoenicka J, Jiménez-Arriero MA, Rodríguez-Jiménez R, Aragüés M, Martín-Suñé N, et al. (2008): DRD2 and ANKK1 genotype in alcoholic patients with psychopathic traits: Association and interaction study. Br J Psychiatry 193:121–125. 23. Meylan E, Tschopp J (2005): The RIP kinases: Crucial integrators of cellular stress. Trends Biochem Sci 30:151–159. 24. Larionov A, Krause A, Miller W (2005): A standard curve based method for relative real time PCR data processing. BMC Bioinform 6:62. 25. Giulian D, Baker TJ (1986): Characterization of ameboid microglia isolated from developing mammalian brain. J Neurosci 6:2163–2178. 26. Quandt K, Frech K, Karas H, Wingender E, Werner T (1995): MatInd and MatInspector: New fast and versatile tools for detection of consensus matches in nucleotide sequence data. Nucleic Acids Res 23:4878 – 4884. 27. Meylan E, Martinon F, Thome M, Gschwendt M, Tschopp J (2002): RIP4 (Dik/PKK), a novel member of the RIP kinase family, activates NF-kappa B and is processed during apoptosis. EMBO Rep 3:1201–1208. 28. Hasegawa M, Fujimoto Y, Lucas PC, Nakano H, Fukase K, Núñez G, Inohara N (2008): A critical role of Rick/RIP2 polyubiquitination in nodinduced NF-kappaB activation. EMBO J 27:373–383. 29. Micheau O, Tschopp J (2003): Induction of TNF receptor I-mediated apoptosis via two sequential signaling complexes. Cell 114:181–190.

J. Hoenicka et al. 30. Lee Y, Messing A, Su M, Brenner M (2008): GFAP promoter elements required for region-specific and astrocyte-specific expression. Glia 56: 481– 493. 31. Zhang Y, Bertolino A, Fazio L, Blasi G, Rampino A, Romano R, et al. (2007): Polymorphisms in human dopamine D2 receptor gene affect gene expression, splicing, and neuronal activity during working memory. Proc Natl Acad Sci U S A 104:20552–20557. 32. Holland P, Willis C, Kanaly S, Glaccum M, Warren A, Charrier K, et al. (2002): RIP4 is an ankyrin repeat-containing kinase essential for keratinocyte differentiation. Curr Biol 12:1424 –1428. 33. Ehlers S, Mueck T, Adams S, Landuzzi L, Lollini PL, Munz B (2008): RIP2 regulates growth and differentiation of normal myoblasts and of rhabdomyosarcoma cells. Eur J Cell Biol 87:163–172. 34. Gotz M, Huttner WB (2005): The cell biology of neurogenesis. Nat Rev Mol Cell Biol 6:777–788. 35. Venter JC, Adams MD, Myers EW, Li PW, Mural RJ, Sutton GG, et al. (2001): The sequence of the human genome. Science 291:1304 –1351. 36. Florea L, Di FV, Miller J, Turner R, Yao A, Harris M, et al. (2005): Gene and alternative splicing annotation with AIR. Genome Res 15:54 – 66. 37. Thierry-Mieg D, Thierry-Mieg J (2006): AceView: A comprehensive cDNA-supported gene and transcripts annotation. Genome Biol 7(suppl 1):S12:1–14.

BIOL PSYCHIATRY 2010;67:3–11 11 38. Araque A, Parpura V, Sanzgiri RP, Haydon PG (1999): Tripartite synapses: Glia, the unacknowledged partner. Trends Neurosci 22:208 –215. 39. Miyatake M, Narita M, Shibasaki M, Nakamura A, Suzuki T (2005): Glutamatergic neurotransmission and protein kinase C play a role in neuronglia communication during the development of methamphetamineinduced psychological dependence. Eur J Neurosci 22:1476 –1488. 40. Gotz M, Hartfuss E, Malatesta P (2002): Radial glial cells as neuronal precursors: A new perspective on the correlation of morphology and lineage restriction in the developing cerebral cortex of mice. Brain Res Bull 57:777–788. 41. Misson JP, Austin CP, Takahashi T, Cepko CL, Caviness VS Jr (1991): The alignment of migrating neural cells in relation to the murine neopallial radial glial fiber system. Cereb Cortex 1:221–229. 42. Chanas-Sacre G, Rogister B, Moonen G, Leprince P (2000): Radial glia phenotype: Origin, regulation, and transdifferentiation. J Neurosci Res 61:357–363. 43. Mo Z, Moore AR, Filipovic R, Ogawa Y, Kazuhiro I, Antic SD, Zecevic N (2007): Human cortical neurons originate from radial glia and neuronrestricted progenitors. J Neurosci 27:4132– 4145. 44. Fox IJ, Paucar AA, Nakano I, Mottahedeh J, Dougherty JD, Kornblum HI (2004): Developmental expression of glial fibrillary acidic protein mRNA in mouse forebrain germinal zones—Implications for stem cell biology. Brain Res Dev Brain Res 153:121–125.

www.sobp.org/journal