• 沒有找到結果。

Mst3 plays an important role in the hypoxia-induced apoptosis of human trophoblasts

N/A
N/A
Protected

Academic year: 2021

Share "Mst3 plays an important role in the hypoxia-induced apoptosis of human trophoblasts"

Copied!
10
0
0

加載中.... (立即查看全文)

全文

(1)

The FASEB Journal

Research Communication

High regulatability favors genetic selection in SLC18A2,

a vesicular monoamine transporter essential for life

Zhicheng Lin,*,‡,1 Ying Zhao,*,‡ Chee Yeun Chung,†,§ Yanhong Zhou,*,‡ Nian Xiong,*,‡ Charles E. Glatt,and Ole Isacson†,§

*Department of Psychiatry and†Department of Neurology, Harvard Medical School, Boston, Massachusetts, USA;‡Division of Alcohol and Drug Abuse and§Center for Neuroregeneration Research, McLean Hospital, Belmont, Massachusetts, USA; and储Department of Psychiatry, Weill Medical College of Cornell University, New York, New York, USA

ABSTRACT SLC18A2 encodes the vesicular

mono-amine transporter 2 protein that regulates neurotrans-mission and reduces cytosolic toxicity of monoamines. Deletion of this gene causes lethality in mice, and DNA sequence variation in this gene is associated with alco-holism and Parkinson’s disease, among other disorders. The Caucasian SLC18A2 promoter has at least 20 hap-lotypes (A–T), with A representing two-thirds of 1460 chromosomes. It is not known why A is selected in the human lineage. To understand the selection, here we took a functional approach by investigating the regula-tions of 4 representative haplotypes (A, C, G, and T) by 17 agents. We show that 76.5% of the agents were able to regulate A but only 11.8 –23.5% of them regulated the 3 other infrequent ones, observing a positive correlation between haplotype frequency and regu-latability. Pathway and molecular analyses revealed five signaling hubs that regulate the four haplotypes differentially, probably through targeting the poly-morphic core promoter region. These findings sug-gest that greater diversity of transcriptional regula-tions is the driving force for the haplotype selection in SLC18A2.—Lin, Z., Zhao, Y., Chung, C. Y., Zhou, Y., Xiong, N., Glatt, C. E., Isacson, O. High regulat-ability favors genetic selection in SLC18A2, a vesicu-lar monoamine transporter essential for life. FASEB

J. 24, 2191–2200 (2010). www.fasebj.org

Key Words: human䡠 promoter haplotypes 䡠 evolution 䡠 signal-ing cascades䡠 transcription

SLC18A2, the human vesicular monoamine trans-porter 2 gene, is associated with a number of brain disorders, including alcoholism, Parkinson’s disease (PD), and schizophrenia (1– 6). Molecular studies have revealed that the low activity-associated haplotypes are risk factors for these diseases. Consistently, one copy deletion of the gene increases both alcohol consump-tion and dopaminergic vulnerability in ⫹/⫺ mice (7–9), demonstrating the contribution of reduced SLC18A2 expression to alcoholism and PD. Interest-ingly, the SLC18A2 promoter has at least 20 haplotypes (A–T, where T is newly identified in this study), and one of them, designated A, had a frequency of 66.7% in

730 unrelated Caucasians from 2 independent collec-tions. In vitro analyses showed that a medium level of promoter activity was displayed by A, whereas higher activity by C and lower activity by G were observed consistently in different cellular systems. Although highly selected, A was among the disease-associated haplotypes (3, 4).

The vesicular monoamine transporter 2 (VMAT2) is an important molecule for the function of monoamin-ergic neurons that are key participants in locomotion, reward, working memory, and mnemonic brain sys-tems. Acting to remove cytosolic monoamines [dopa-mine (DA), serotonin, norepinephrine, and hista[dopa-mine] by uptake into intracellular vesicles and to discharge the monoamines into extracellular space, VMAT2 pre-vents neurotoxicity of these monoamines in the cytosol and regulates neurotransmission (10). VMAT2 is ex-pressed in central, peripheral, and enteric neurons as well as in platelets (11, 12). The expressional difference between different human brain regions can be ⬎16-fold (13). Data from knockout mice indicate that expression of VMAT2 is essential for survival and that different expression levels have altered behavioral con-sequences. Homozygous (⫺/⫺) knockout mice survive only about 1 postnatal week because of developmental defects, which is in contrast to other monoamine transporters, for which homozygous deletions (⫺/⫺) cause no lethality (14). Nobably, environmental factors can regulate expression of the VMAT2 gene, including stress, clozapine, and environmental contaminants (15–17). Despite the small number of regulation stud-ies available, these data all suggest that the VMAT2 gene is under sensitive physiological regulation.

By investigating haplotype-dependent regulation of SLC18A2, in this study, we took a functional analysis approach to understanding why SLC18A2 promoter haplotype A is highly selected in the human lineage, whereas others are deselected. Elucidation of the bio-logical drive for the selections may help our

under-1Correspondence: McLean Hospital, 115 Mill St., Belmont,

MA 02478, USA. E-mail: zhicheng_lin@hms.harvard.edu doi: 10.1096/fj.09-140368

(2)

standing of the contributions of SLC18A2 to human evolution and diseases.

MATERIALS AND METHODS

Haplotype-reporter hybrid constructions

SLC18A2 promoter region-luc (fruit fly luciferase gene) hy-brids were constructed by shuttling 8.5-kb NheI/BamHI frag-ments of plasmids pGL3e-hVMAT2– 6.1A and pGL3e-hVMAT2– 6.1C (3) into pGL4.14[luc2/Hygro] (Promega, Madison, WI, USA), resulting in 12.1-kb plasmids pSLC18A2– 6.3A-luc and pSLC18A2– 6.3C-luc. A 5.5-kb haplotype G or T fragment was cloned by PCR from a G or T carrier and after confirmation of PCR fidelity by DNA sequencing (Supplemental Fig. S1B; Supplemental Table S1A, B) was shuttled into pSLC18A2– 6.3A-luc, resulting in pSLC18A2– 6.3G-luc or pSLC18A2– 6.3T-luc. The control vector was generated by replacing the 2.0-kb NheI/HindIII fragment in pGL4.14[luc2/Hygro] with the 2.2-kb NheI/HindIII fragment of pGL3-Enhancer. To

linearize the plasmids, NheI located immediately upstream of SLC18A2 DNA was used for restriction digestion (Supplemen-tal Fig. S1A), followed by purification with a QIAEX II Gel Extraction Kit (Qiagen, Valencia, CA, USA). Haplotype con-firmation of SY-SY5Y genomic DNA for SLC18A2 used PCR and DNA resequencing primers listed in Supplemental Table S1E.

Transfection, agent treatments, and luciferase activity measurements

The transfection procedure has been described before (3). Eighteen-hour treatments of transfected cells with agents, as listed in Table 1, started 27 h after transfection. For luciferase (Luc) activity measurements, a Luciferase Assay System Kit (Promega) in Bio-Tek Synergy HT/KC4 and a 96-well format were used, according to the manufacturer’s instructions. Cell numbers in each well were estimated by protein amount based on Protein Assay Reagent (Bio-Rad, Hercules, CA, USA). An arbitrary unit of SLC18A2 promoter activity is calculated as Luc activity: (readout/protein)(SLC18A2⫺MOCK)/

TABLE 1. Agents used in regulation studies

Symbol Name Major activity Solvent Stock conc. Working conc. Source

AMPH d-Amphetamine Treatment of ADHD;

psychostimulant

H2O 50 mM 10␮M Sigma-Aldrich Corp., St.

Louis, MO, USA

DA Dopamine Neurotransmitter 50␮M

ascorbic acid

50 mMa 10␮M Sigma-Aldrich Corp.

Dynb Dynorphin A Endogenous

␬-agonist

H2O 0.5 mM 0.25␮M Phoenix Pharmaceuticals,

Belmont, CA, USA

Forskolin Adenylyl cyclase

activator

DMSO 50 mM 10␮M EMD Biosciences, San Diego, CA, USA IGF-1 Insulin-like growth

factor AKT signaling activator H2O/0.1% BSA 0.1 mg/ml 50 ng/ml Sigma-Aldrich Corp.

KN-62 CaM kinase inhibitor DMSO 15 mM 3␮M Tocris Cookson, Inc.,

Ellisville, MO, USA LY294002 PI3 kinase inhibitor DMSO 15 mM 3␮M Tocris Cookson, Inc. MK801 (⫹)-MK-801 or dizocilpine NMDA receptor inhibitor H2O 5 mM 1␮M Sigma-Aldrich Corp. MPP⫹ Methylpiperidinopy-razole

Neurotoxin H2O 0.5 mMa 5␮M Sigma-Aldrich Corp.

OA Okadaic acid Protein phosphatase inhibitor

DMSO 0.1 mM 20 nM Tocris Cookson, Inc. 6-OHDA 6-Hydroxydopamine Neurotoxin 50␮M

ascorbic acid

50 mM 10␮M Sigma-Aldrich Corp.

PACAP38 Pituitary adenylyl cyclase activating polypeptide 38 amino acids Adenylate cyclase stimulator H2O 50␮M 10 nM EMD Biosciences. . . . 4␣-PDD 4␣-Phorbol 12,13-didecanoate Negative control of PMA, PDD

DMSO 5 mM 1␮M EMD Biosciences

PDD Phorbol 12,13-didecanoate

PKC activator DMSO 5 mM 1␮M EMD Biosciences

PMA Phorbol 12-myristate 13-acetate

PKC activator DMSO 5 mM 1␮M EMD Biosciences

SB202190 p38 kinase inhibitor DMSO 50 mM 10␮M Tocris Cookson, Inc. TNF-␣ Tumor necrosis

factor-␣

Apoptosis inducer H2O 0.1 mg/ml 100 ng/ml Sigma-Aldrich Corp.

U0126 MEK1/2 inhibitor DMSO 50 mM 10␮M Tocris Cookson, Inc.

ADHD, attention deficit/hyperactivity disorder; CaM kinases, Ca2⫹/calmodulin-dependent protein kinase II; NMDA, N-methyl-d-aspartate.

(3)

(readout/protein)(Vector⫺MOCK), where MOCK means

trans-fection without DNA, a background control.

Stable cell line establishment and immunocytochemistry

Establishment of stable cell lines followed a previous protocol (18), using BstEII-linearized pSLC18A2– 6.3A-luc (Supplemental Fig. S1A) and cloning medium containing 100␮g/ml hygromy-cin B (Invitrogen, Carlsbad, CA, USA) until colonies appeared, followed by Luc activity and immunocytochemical analyses (19). Luc activity uses (readout/protein)(SLC18A2⫺MOCK).

Electrophoretic mobility shift assay (EMSA)

For 35S-labeling of oligonucleotides (oligos) (listed in

Sup-plemental Table S1C), double-stranded (ds) oligos were prepared in annealing buffer (10 mM Tris-HCl, pH 8.0, and 100 mM MgCl2), and 20␮M ds oligo was labeled in a 30-␮l

solution containing 0.5␮l of 20 ␮M annealed ds oligos, 3 ␮l of 10⫻ Reaction Buffer 2 (New England Biolabs, Ipswich, MA, USA), 6 ␮l of [35S]dATP (60 pmol; PerkinElmer Life and

Analytical Sciences, Waltham, MA, USA), and 1␮l of Large Fragment DNA Polymerase I (Invitrogen), incubated at room temperature for 28 min followed by purification in a Chroma Spin-10 column (Clontech, Mountain View, CA, USA). DNA binding reactions were conducted with a Novagen EMSA Accessory Kit (EMD Biosciences, Gibbstown, NJ, USA), and electrophoresis was performed with 0.6% nondenaturing DNA retardation gels according to the EMSA kit instructions. Gels were dried on filter papers and exposed to X-ray film for 3–5 d for autoradiography.

Quantitation of mRNA levels by quantitative real-time PCR (qRT-PCR)

Reverse transcription was performed with Superscript II re-verse transcriptase (Qiagen) oligo(dT) as the primer. Quan-titative PCR reactions was conducted with SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA) and

primers listed in Supplemental Table S1D. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and␤-actin were used as internal controls. Relative expression levels were calculated by using the 2⫺⌬⌬CTmethod (20). SLC18A2 primers showed amplification efficiency approximately equal (within 5% dif-ference) to that of the internal controls.

Southern blotting

Biotin labeling of a probe (1.1-kb AvaI/HindIII fragment within the luc gene) was performed with the BrightStar Psoralen-Biotin Nonisotopic Labeling Kit and associated pro-tocols (Ambion; Applied Biosystems Inc.). Genomic DNA, isolated from cultured cells by using the Blood & Cell Culture DNA Midi Kit (Qiagen), was digested with KpnI/BamHI, followed by 0.7% agarose electrophoresis. After capillary transfer, the DNA on the positively charged nylon membrane (Ambion) was hybridized with the biotin-labeled probe in ULTRAhyb hybridization buffer (Ambion) at 42°C for 24 h. After several washes with SSC/0.1% SDS and 1⫻ wash buffer, membranes were treated with strep-alkaline phosphatase in blocking buffer for 15 min, washed with 1⫻ wash buffer, and treated with detection solution CDP-Star (BrightStar Bio-Detect Kit; Ambion), followed by exposure to a film.

RESULTS

Expression systems for regulation study

To study regulation of the SLC18A2 promoter, hu-man cell lines that express endogenous SLC18A2 were searched among 5 human cell lines, including 4 DA cell lines [SH-SY5Y, IMR-32, SK-N-AS, and BE(2)-M17] and a non-neuronal cell line (HEK293T) by using qRT-PCR. As a result, SH-SY5Y expressed the highest levels of endogenous SLC18A2 mRNA and IMR-32 the second highest (Fig. 1A). The data on BE(2)-M17 were

Figure 1.Search for expression systems. A) Endogenous SLC18A2 expression in human cell lines. The DA cell line SH-SY5Y expressed endogenous SLC18A2 the most, compared with two other DA cell lines, IMR-32 and SK-N-AS, based on qRT-PCR. Non-neuronal cell line HEK293T did not express detectable mRNA levels (n⫽3 each in duplicate). P ⫽ 0.0006 overall, P ⬍ 0.001 for SH-SY5Y vs. others; ANOVA. B) PACAP38 regulation of haplotype A in circular vs. linearized pSLC18A2– 6.3A-luc in SH-SY5Y (n⫽6–12). *P ⬍ 0.05; Student’s t test.

(4)

unreliable and are not shown here. We therefore used the human DA cell lines SH-SY5Y and IMR-32 as two systems for SLC18A2 regulation analysis that mimic different brain regions to some extent.

In this study we also chose linearized plasmid DNAs for expressional analyses. In an animal study, Miller and colleagues (21) had reported that PACAP38 up-regulated VMAT2 levels in mouse striatum. However, PACAP38 failed to up-regulate SLC18A2 promoter ac-tivity when pSLC18A2– 6.3A-luc (6.3-kb haplotype A in a Luc reporter vector; see Materials and Methods) was transiently expressed in SH-SY5Y. With the plasmid DNA linearized, PACAP38 was able to up-regulate the promoter activity (Fig. 1B). These data suggested that the supercoiled structure of pSLC18A2– 6.3A-luc af-fected regulation of the promoter activity. Therefore, regulation data from transient expression systems in this study were obtained with NheI-linearized DNAs.

Regulation of transient expression in SH-SY5Y and IMR-32

Seventeen agents (substances) were used in this regu-lation study, including 5 endogenous ones (listed in Table 1). All of the resultant regulation data are summarized in Table 2 (see Supplemental Material for details), and only those showing statistical significance based on t tests (in pairwise treatments) or by ANOVA posttest (Tukey’s test) are described here.

Endogenous molecules

DA is an endogenous substrate of the VMAT2 protein and binds to DA receptors, inducing intracellular sig-naling that could regulate SLC18A2. In SH-SY5Y, treat-ment with 10 ␮M DA for 18 h tended to up-regulate promoter activity of haplotype A by 1.55-fold (P⫽0.058 by t tests), compared with the negative control (H2O Vc). DA had no effect on three other haplotypes in SH-SY5Y or IMR-32 (data not shown). Another four endogenous agents studied were PACAP38, Dyn, IGF-1, and TNF-␣, among which PACAP38 was the only one that up-regulated the promoter activity of A (2.03-fold) and C (2.02-fold) in SH-SY5Y and of A (1.46-fold), C (1.75-fold), and T (2.34-fold) in IMR-32, compared with the negative control.

Exogenous molecules, including AMPH, MK801, and forskolin

In SH-SY5Y/A (pSLC18A2– 6.3A-luc) cells, MK801 up-regulated A (2.06-fold), C (2.37-fold), and G (2.48-fold) but not T. Forskolin up-regulated the promoter activity of A by 2.02-fold but not the promoter activity of any other haplotypes in SH-SY5Y.

In IMR-32, haplotype dependence was less evident. Forskolin up-regulated promoter activity by the largest amount (2.80-fold) on G, second (1.84-fold) on A or TABLE 2. Summary of haplotype-dependent regulations

Agent Haplotype Reference figure SH-SY5Y IMR-32 A C G T A C G T Endogenous DA 1a,** 2A, 3B PACAP38 1## 1# 1# 1# 1# 2B, S2 Dyn 1b,### 3B, S2 IGF-1 3B, S2 TNF-␣ 1a,# 2B, S2 Exogenous AMPH S3 MK801 1### 1## 1# S3 Forskolin 1** 1** 1** 1*** 1* S3 PMA/ PDD 2###,### 2##,# 2#,# 2**, NA S4 OA 1# S4 LY294002c 2a,### 2D U0126c 2a,# 2D SB202190c 2a,###;b,# 2D, 3B KN-62 S4 MPP⫹ 1a,*,*** 1** NA 2** NA 2E, S5 6-OHDA 1a,*,** 1** 1* 1* 2E, S5 Up 9 3 2 1 3 2 1 2 Down 4 1 1 1 1 0 0 0 Total 13 4 3 2 4 2 1 2

All data from the transient expression experiments SH-SY5Y and IMR-32 as indicated, unless noted. No arrow indicates no significant effect based on either ANOVA posttests or t tests. NA, not analyzed.aData from endogenous expression of SH-SY5Y.bData from stable expression of

A7.cUnreliable data in transient expression systems. *P⬍ 0.05, **P ⬍0.01, ***P ⬍ 0.001; Student’s t tests of pairwise treatments.#P⬍ 0.05,##P

(5)

(1.74-fold) on C and least (1.66-fold) on T. None of the other treatments including those with MK801 had any effect on the promoter activities. That is, AMPH did not affect any of these haplotypes in either of the two cell lines (Supplemental Fig. S3).

PKC, CaMKII, and phosphatase effectors: two PKC activators, PMA and PDD, the CaMK inhibitor KN-62, and the protein phosphatase inhibitor OA

In SH-SY5Y, PMA and PDD down-regulated the promoter activities of all 4 haplotypes, by⬃0.6-fold for C, 0.5-fold for

A, 0.4-fold for G, and 0.14-fold for T. OA up-regulated the A promoter activity (by 2.02-fold) but had no significant

effect on 3 others; KN-62 had no effect on any of the haplotypes. In IMR-32, no significant effects were ob-served for any of the haplotypes (Supplemental Fig. S4).

Neurotoxins: MPP(a substrate of VMAT2) and 6-OHDA, the two most common toxins used in animal models for PD

In SH-SY5Y/A, MPP⫹ and 6-OHDA up-regulated the promoter activity by 1.62- and 2.24-fold. In SH-SY5Y/C, 6-OHDA displayed an up-regulation of 1.56-fold, and MPP⫹ up-regulation was not statistically significant. In SH-SY5Y/G, only MPP⫹displayed a significant up-regula-tion (1.95-fold). In SH-SY5Y/T, 6-OHDA displayed a significant up-regulation (2.92-fold). In IMR-32/A, MPP⫹ down-regulated but 6-OHDA up-regulated the promoter

activity by 0.75- and 1.39-fold. Neither of the toxins regulated C or G, and 6-OHDA also did not affect T in IMR-32 (Supplemental Fig. S5).

Regulation of endogenous expression in SH-SY5Y

Selected agents were examined for their effects on endog-enous mRNA levels of SLC18A2 in SH-SY5Y by using qRT-PCR. DA was able to up-regulate SLC18A2 by 1.57-fold (Fig. 2A), a finding consistent with what was observed in the transient expression. TNF-␣ up-regulated the en-dogenous activity by 1.71-fold (Fig. 2B). PMA had no effect (Fig. 2C). The MAP kinase inhibitors LY294002 (PI3 kinase), U0126 (MEK1/2), and SB202190 (p38 kinase) down-regulated the endogenous activity by 0.61-, 0.80-, and 0.64-fold (Fig. 2D). In the transient expression systems, the regulation data were generally unreliable because of vector effects, but haplotype dependence was observed for PI3 kinase inhibitor-mediated down-regula-tion in A (0.65-fold) vs. G (1.10-fold) in SH-SY5Y. No haplotype dependence was observed in IMR-32. Consis-tent with the transient expression, both MPP⫹ and 6-OHDA up-regulated the endogenous activity by 2.02-and 1.71-fold, compared with the negative controls (Fig. 2E). These regulation data, except for PACAP38 and PMA, were all consistent with the findings from transient expression of haplotype A in SH-SY5Y. More consistently,

SLC18A2 haplotyping of the SH-SY5Y genomic DNA

con-firmed that this cell line carried homozygous A.

Figure 2.Regulation of endogenous SLC18A2 activity in SH-SY5Y by various agents, based on qRT-PCR analysis of mRNA levels. A) DA; n⫽ 4. B) Other endogenous agents; n ⫽ 3. P ⫽ 0.021 overall,#P⬍ 0.05 vs. H2O; ANOVA. C) PKC activator PMA; n

5. D) MAP kinase inhibitors; n⫽ 3–4. P ⬍ 0.0001 overall;#

P⬍ 0.05,###P⬍ 0.001 vs. DMSO; ANOVA. E) Neurotoxins; n ⫽ 3–5. *P⬍ 0.05, **P ⬍ 0.01 vs. corresponding controls; Student’s t test. Shaded bars indicate solvent controls.

(6)

Regulation of stable expression

Chromatin structure may influence promoter regula-tion (22), suggesting that expression from plasmid DNAs may or may not be able to reflect the chromo-somal expression. To understand the potential discrep-ancy in the regulation of SLC18A2 between plasmids and chromosomes, stable cell lines were generated for haplotype A-luc hybrid to integrate into host chromo-somes and to see whether the same regulations could be observed in the stable cell lines. Among 12 colonies obtained, one stable cell line, termed A7, expressed

normal Luc activity (within 1- to 2-fold of the activity observed in transient expression) and was a pure cell line, based on immunocytochemical staining of Luc (Fig. 3A). Regulation analyses showed that Dyn up-regulated the promoter activity expressed in the host chromosome, whereas SB202190 again down-regulated the activity in this stable cell line (Fig. 3B). Generally speaking, A7 displayed a regulation pattern similar to those observed in the transient or endogenous expres-sion system. Southern blotting analyses of the genomic DNA digested with KpnI/BamHI showed that A7 car-ried a single copy of the hybrid and that this hybrid had

Figure 3.Characterization of stable SH-SY5Y/pSLC18A2– 6.3A-luc cell line A7. A) Luc immunoreactivity in naive and A7 cells. B) Regulation of A by DA, other endogenous agents, and kinase effectors. Shaded bars indicate solvent controls; bent arrow indicates a negative control and its comparison. ANOVA: P⬍ 0.0001 overall,###

P⬍ 0.001vs. H2O for endogenous agents; P

0.0005 overall,#

P⬍ 0.05 vs. DMSO for kinase effectors; P value is indicated by t tests, compared with corresponding control H2O

Vc (n⫽12–14). C) Southern blotting analysis of A7. plas, plasmid DNA as a positive control for the 6.5-kb target. D) EMSA analysis of SNPs ⫺106A/G, ⫺103C/A, and ⫺62G/A (Supplemental Figure S1C). SNPs at ⫺106 and ⫺103 were linked and analyzed together as AC in G, GA in A, and GC in C or T. Three nuclear proteins with differential binding activities are termed ␣, ␤, and ␦ here. Oligo specificity of these binding activities has been confirmed by competition experiments; the persistent band below the arrows was poly(A/T)-related activity.

(7)

been truncated by⬃3 kb, probably at the 5⬘ end (Fig. 3C). The exact location of truncation has not been mapped yet, but A7 appeared to carry a 1- to 1.5-kb core promoter region upstream of the luc gene, and the observed regulations were executed by this core pro-moter.

Transcription factor (TF) binding to polymorphic sites in the core promoter

Because EMSA may faithfully reveal TF-promoter inter-actions that occur in vivo (23–25), this method was used to investigate whether polymorphic sites in the core promoter region were bound by nuclear proteins. This core promoter region harbored 3 single-nucleotide polymorphisms (SNPs) at ⫺106, ⫺103, and ⫺62 (3). To understand whether TFs could recognize these polymorphic sites, EMSA analyses were performed by using 35S-labeled ds oligos (Supplemental Table S1C) and nuclear proteins isolated from SH-SY5Y. As a result, there appeared to be 3 nuclear proteins that bound to these sites:␣ bound to ⫺106 to ⫺103 G–A in haplotype

A but not to other haplotypes; ␤ bound most to

haplotype G, less to A, and not at all to C/T; and ␦ bound to the⫺62G allele that was carried by A, G, or T (Fig. 3D; Supplemental Fig. S1C). These differential binding activities were consistent with the haplotype-dependent regulations observed in our transient ex-pression systems.

DISCUSSION

Functional basis for haplotype selection

The critical finding in our analyses is that haplotype A, the highly selected one, is more regulatable than the infrequent haplotypes C, G, and T. A was regulated by

13 (76.5%) of the 17 agents tested, displaying the most responses relative to C (4 agents or 23.5%), G (3 agents or 17.6%), and T (2 agents or 11.8%) in SH-SY5Y. In particular, A was regulated by 4 of the 5 endogenous agents, compared with only 1 agent for C and none for

G or T. These agents were relatively less active in

IMR-32, probably because SLC18A2 is less active in this cell line, as IMR-32 expresses only 12% of the mRNA levels in SH-SY5Y (Fig. 1A). The lesser activity suggests that fewer signaling pathways activate SLC18A2 in IMR-32. A was regulated by 4 (23.5%) agents, compared with

C or T by 2 (11.8%) agents and G by 1 (5.9%) agent in

IMR-32, a high A-regulatability pattern similar to that in SH-SY5Y (Table 2). These data on more regulations of

A are consistent with the fact that A is recognized by

more nuclear proteins than the other haplotypes (Fig. 3D).

Interestingly, haplotype frequency is positively and almost linearly correlated with the number of signifi-cant regulations (Fig. 4). Cladistic analysis of the four haplotypes shows a large distance between the two most frequent haplotypes, A and C. Evidently, A has been heavily favored during human biological history; T is relatively close to A but is deselected (Fig. 4A). These data imply that high SLC18A2 regulatability is critical for survival during human evolution, which is consis-tent with variations in VMAT2 expression influencing brain integrity, function, and behaviors in knockout mice. In particular, the high regulatability may underlie forms of synaptic plasticity in which regulation of neurotransmission relies on variation in VMAT2-facili-tated quantal size of transmitter release (26). An in-triguing question is the following: because low-activity haplotypes are associated with brain diseases, why did high-activity haplotypes such as C not have a greater survival advantage and were not highly selected (the frequency of C was 14.4% in the Caucasian popula-tions)? Among other speculations (14, 27), we may

Figure 4. A) Cladogram of SLC18A2 promoter haplotypes A, C, G, and T (see ref. 3 for method). Scale bar ⫽ 0.1

substitution/site. B, C) Positive correlation between haplotype frequency and regulatability for 5 endogenous agents (B) or all studied agents (C) in 2 DA cell lines, SH-SY5Y and IMR-32. IMR-32 had too few endogenous agent-based regulations to show a correlation in B. Haplotype frequency: 66.7% for A, 14.4% for C, 3.68% for G, and⬃1% for T, based on 1460 Caucasian chromosomes combined from 2 independent studies (3, 4). Haplotype T appeared to be a variant of haplotype H (3), due to a different simple sequence length polymorphism allele at ⫺5199 and 2 different intron 1 alleles at ⫹321 and ⫹517 (Supplemental Figure S1C; Table 3 in ref. 3), possibly representing a subgroup of the haplotype 1 that had a 2% frequency in ⬎300 Caucasians (4). Because haplotypes G, H, K, and M together counted for 6% of the total haplotypes in 333 Caucasians (3), we estimate that this new haplotype, termed haplotype T, has a frequency of⬃1%. The r2value indicates degree of linear fitness

(8)

postulate that C is much less regulatable, reducing the capacity for synaptic plasticity. It would be interesting to see whether SLC18A2 is associated with other monoam-ine-related features of human beings.

Dependence in SLC18A2 activity

The most consistent findings from this study are the haplotype dependence and the cell type-dependence in regulation of SLC18A2 promoter activity. Such haplo-type dependence could range from 0.91-fold for G to 2.24-fold for A (a 2.5-fold difference) with 6-OHDA or from 0.14-fold for T to 0.48-fold for A (a 3.4-fold difference) with PMA in SH-SY5Y. Among the 17 agents investigated, 76.5% (13 agents) of them showed haplo-type dependence in either of the DA cell lines (Table 2). Some agents regulated one haplotype but not another, including DA, Dyn, TNF-␣, forskolin, OA, and 6-OHDA. Others affected different haplotypes to vari-ous extents, including PMA and 6-OHDA in SH-SY5Y,

forskolin in IMR-32, and PACAP38 in both cell lines. These haplotype-dependent regulations suggest that the 6.3-kb haplotypes are regulated by multiple signal-ing cascades, among which some may recognize TFs that bind to conserved elements or polymorphic sites in the core promoter.

The SLC18A2 regulatome: A-dominant

Several signaling cascades appear to contribute to the selection of A, based on the regulation data. These cascades are conjoined by five hubs including cAMP response element-binding protein (CREB) 1,␤-arrestin 1, G-protein ␣s, c-Raf-1, and ERK-1 (Fig. 5) and may make few points. Foremost, upstream of CREB1 is CaMKII that had been expected to activate CREB1, an activator of the VMAT2 gene (28). However, this acti-vation was not present in our expression systems be-cause the CaMKII inhibitor KN-62 did not affect the promoter activity. Second, Ca2⫹ appeared to play an

Figure 5.Signaling cascades that regulate A. Green asterisks, agents that regulated A (see Table 2); red boxes, major hubs each withⱖ4 interactions; thick aquamarine lines, canonical pathways; thin green arrows, activation; thin red arrows, inhibition; thin gray arrows, unspecified; X, pathway that was not supported by the data from this study. Oxidative stress-related (MPP⫹and 6-OHDA) pathways are not included. Bottom rectangle: schematic diagram of SLC18A2 regulated by TFs, including CREB1 and others (indicated by question mark) such as those indicated in Fig. 3D; CRE (green), cAMP-responsive element; black arrows, unknown links between TFs and pathways. Pathways analysis was performed with MetaCore (36).

(9)

inhibitory role here because its reduced concentration seems to correlate with increased promoter activity, as suggested by data with PMA, PDD, MK801, and SB202190. This Ca2⫹-reverse correlation is further sup-ported by the nicotine/nicotinic acetylcholine receptor (nAChR)-mediated down-regulation of A by 22% (P⫽0.010, n⫽4; compared with the negative control), consistent with the finding that expression of a hyper-sensitive nAChR caused neurotoxicity and DA neuron death in mice (29). Third, a single agent such as Dyn may regulate A via multiple pathways (bradykinin re-ceptor B2, G-protein ␣s, ␬-type opioid receptor, the L-type Ca2⫹ channel ␣1C subunit, and ␤-arrestin 1). Fourth, expression of an endogenous agent such as TNF-␣ can regulate as well as be regulated by related pathways (␤-arrestin 1, c-Jun pathway, and CREB1), implying the regulation of A by balancing of different related pathways. Finally, infrequent haplotypes are reg-ulated by a limited number of pathways that involve CREB1 and Ca2⫹. Because agents that regulated the rare haplotypes all regulated the dominant A (Table 2), these rare haplotypes may miss regulatory elements due to DNA sequence variation, which merits further investigation.

Technical caveats associated with reporter systems

Transient expression of the Luc-based reporter is used for transcriptional analyses by many biological fields (30 –34), and data obtained by using such a reporter system could be spurious according to our observations. Foremost, regulation of the promoter on linearized plasmid DNA expressed transiently could result in data more similar to those obtained with endogenous, sta-ble, or even in vivo expression than data obtained with circular plasmid DNA, based on the PACAP38 and DA data (35) (Figs. 1B; 2A, B; and 3B), suggesting that the regulations required an open DNA conformation (21). The second observation was that kinase effectors could regulate mysterious promoters of the vector and such regulations of vector depended on the expression systems. These technical caveats associated with transient expression of reporter assays mandate data confirmation by a means different from transient expression.

CONCLUSIONS

Haplotype A is regulated by more pathways than the infrequent haplotypes C, G, and T, suggesting a positive correlation between haplotype frequency and number of regulators. TF-based further dissection of the haplo-type-specific regulatory cascades can reveal specific mechanisms for haplotype selections. Haplotype-de-pendence can also be cell type-dependent and trans-genic technology will help to clarify specific contribu-tions of distinct SLC18A2 haplotypes to human evolution and diseases.

This work was funded by grants from the U.S. National Institutes of Health (P50-NS039793 to O.I., P01-ES016732 to

C.E.G., and R01-DA021409 to Z.L.) and by a Michael J. Fox Foundation Award (to Z.L.).

REFERENCES

1. Zubieta, J. K., Taylor, S. F., Huguelet, P., Koeppe, R. A., Kilbourn, M. R., and Frey, K. A. (2001) Vesicular monoamine transporter concentrations in bipolar disorder type I, schizo-phrenia, and healthy subjects. Biol. Psychiatry 49, 110 –116 2. Zucker, M., Valevski, A., Weizman, A., and Rehavi, M. (2002)

Increased platelet vesicular monoamine transporter density in adult schizophrenia patients. Eur. Neuropsychopharmacol. 12, 343– 347

3. Lin, Z., Walther, D., Yu, X. Y., Li, S., Drgon, T., and Uhl. G. R. (2005) SLC18A2 promoter haplotypes and identification of a novel protective factor against alcoholism. Hum. Mol. Genet. 14, 1393–1404

4. Glatt, C. E., Wahner, A. D., White, D. J., Ruiz-Linares, A., and Ritz, B. (2006) Gain-of-function haplotypes in the vesicular monoamine transporter promoter are protective for Parkinson disease in women. Hum. Mol. Genet. 15, 299 –305

5. Schwab, S. G., Franke, P. E., Hoefgen, B., Guttenthaler, V., Lichtermann, D., Trixler, M., Knapp, M., Maier, W., and Wilde-nauer, D. B. (2005) Association of DNA polymorphisms in the synaptic vesicular amine transporter gene (SLC18A2) with alco-hol and nicotine dependence. Neuropsychopharmacology 30, 2263–2268

6. Talkowski, M. E., Kirov, G., Bamne, M., Georgieva, L., Torres, G., Mansour, H., Chowdari, K. V., Milanova, V., Wood, J., McClain, L., Prasad, K., Shirts, B., Zhang, J., O’Donovan, M. C., Owen, M. J., Devlin, B., and Nimgaonkar, V. L. (2008) A network of dopaminergic gene variations implicated as risk factors for schizophrenia. Hum. Mol. Genet. 17, 747–758 7. Hall, F. S., Sora, I., and Uhl, G. R. (2003) Sex-dependent

modulation of ethanol consumption in vesicular monoamine transporter 2 (VMAT2) and dopamine transporter (DAT) knockout mice. Neuropsychopharmacology 28, 620 – 628

8. Takahashi, N., Miner, L. L., Sora, I., Ujike, H., Revay, R. S., Kostic, V., Jackson-Lewis, V., Przedborski, S., and Uhl, G. R. (1997) VMAT2 knockout mice: heterozygotes display reduced amphetamine-conditioned reward, enhanced amphetamine lo-comotion, and enhanced MPTP toxicity. Proc. Natl. Acad. Sci. U. S. A. 94, 9938 –9943

9. Gainetdinov, R. R., Fumagalli, F., Wang, Y. M., Jones, S. R., Levey, A. I., Miller, G. W., and Caron, M. G. (1998) Increased MPTP neurotoxicity in vesicular monoamine transporter 2 heterozygote knockout mice. J. Neurochem. 70, 1973–1978 10. Mosharov, E. V., Larsen, K. E., Kanter, E., Phillips, K. A., Wilson,

K., Schmitz, Y., Krantz, D. E., Kobayashi, K., Edwards, R. H., and Sulzer, D. (2009) Interplay between cytosolic dopamine, cal-cium, and␣-synuclein causes selective death of substantia nigra neurons. Neuron 62, 218 –229

11. Peter, D., Liu, Y., Sternini, C., de Giorgio, R., Brecha, N., and Edwards, R. H. (1995) Differential expression of two vesicular monoamine transporters. J. Neurosci. 15, 6179 – 6188

12. Lebrand, C., Cases, O., Wehrle´, R., Blakely, R. D., Edwards, R. H., and Gaspar, P. (1998) Transient developmental expres-sion of monoamine transporters in the rodent forebrain. J. Comp. Neurol. 401, 506 –524

13. Taylor, S. F., Koeppe, R. A., Tandon, R., Zubieta, J. K., and Frey, K. A. (2000) In vivo measurement of the vesicular monoamine transporter in schizophrenia. Neuropsychopharmacology 23, 667– 675

14. Kalueff, A. V., Ren-Patterson, R. F., and Murphy, D. L. (2007) The developing use of heterozygous mutant mouse models in brain monoamine transporter research. Trends Pharmacol. Sci.

28,122–127

15. Rusnak, M., Kvetnansky, R., Jelokova, J., and Palkovits, M. (2001) Effect of novel stressors on gene expression of tyrosine hydroxylase and monoamine transporters in brainstem norad-renergic neurons of long-term repeatedly immobilized rats. Brain Res. 899, 20 –35

16. Rehavi, M., Roz, N., and Weizman, A. (2002) Chronic clozapine, but not haloperidol, treatment affects rat brain vesicular mono-amine transporter 2. Eur. Neuropsychopharmacol. 12, 261–268

(10)

17. Richardson, J. R., and Miller, G. W. (2004) Acute exposure to Aroclor 1016 or 1260 differentially affects dopamine transporter and vesicular monoamine transporter 2 levels. Toxicol. Lett. 148, 29 – 40

18. Lin, Z., Zhang, P. W., Zhu, X., Melgari, J. M., Huff, R., Spieldoch, R. L., and Uhl, G. R. (2003) Phosphatidylinositol 3-kinase, protein kinase C, and MEK1/2 kinase regulation of dopamine transporters (DAT) require N-terminal DAT phos-phoacceptor sites. J. Biol. Chem. 278, 20162–20170

19. Lin, Z., Wang, W., Kopajtic, T., Revay, R. S., and Uhl, G. R. (1999) Dopamine transporter: transmembrane phenylalanine mutations can selectively influence dopamine uptake and co-caine analog recognition. Mol. Pharmacol. 56, 434 – 447 20. Livak, K. J., and Schmittgen, T. D. (2001) Analysis of relative

gene expression data using real-time quantitative PCR and the 2(⫺⌬⌬CT) method. Methods 25, 402– 408

21. Guillot, T., III, Mercer, A., Zachrisson, O., and Miller, G. W. (2006) PACAP38 regulates striatal expression of VMAT2. Pro-gram No. 824.16/BB27, Neuroscience Meeting Planner [on-line], Society for Neuroscience, Atlanta, GA, USA

22. Chakrabarti, S. K., James, J. C., and Mirmira, R. G. (2002) Quantitative assessment of gene targeting in vitro and in vivo by the pancreatic transcription factor, Pdx1. Importance of chromatin structure in directing promoter binding. J. Biol. Chem. 277, 13286–13293 23. Wang, G., Qi, X., Wei, W., Englander, E. W., and Greeley, G.H.,

Jr. (2006) Characterization of the 5⬘-regulatory regions of the rat and human apelin genes and regulation of breast apelin by USF. FASEB J. 20, 2639 –2641

24. Zanotto, E., Shah, Z. H., and Jacobs, H. T. (2007) The bidirec-tional promoter of two genes for the mitochondrial transla-tional apparatus in mouse is regulated by an array of CCAAT boxes interacting with the transcription factor NF-Y. Nucleic Acids Res. 35, 664 – 677

25. Tseng, P. C., Hsu, H. C., Janmanchi, D., Lin, C. H., Kuo, Y. H., Chou, C. K., and Yeh, S. F. (2008) Helioxanthin inhibits interleukin-1␤-induced MIP-1 ␤ production by reduction of c-jun expression and binding of the c-jun/CREB1 complex to the AP-1/CRE site of the MIP-1 ␤ promoter in Huh7 cells. Biochem. Pharmacol. 76, 1121–1133

26. Pothos, E. N., Larsen, K. E., Krantz, D. E., Liu, Y., Haycock, J. W., Setlik, W., Gershon, M. D., Edwards, R. H., and Sulzer, D. (2000) Synaptic vesicle transporter expression regulates vesicle pheno-type and quantal size. J. Neurosci. 20, 7297–7306

27. Nikoshkov, A., Drakenberg, K., Wang, X., Horvath, M. C., Keller, E., and Hurd, Y. L. (2008) Opioid neuropeptide geno-types in relation to heroin abuse: dopamine tone contributes to

reversed mesolimbic proenkephalin expression. Proc. Natl. Acad. Sci. U. S. A. 105, 786 –791

28. Watson, F., Kiernan, R. S., Deavall, D. G., Varro, A., and Dimaline, R. (2001) Transcriptional activation of the rat vesic-ular monoamine transporter 2 promoter in gastric epithelial cells: regulation by gastrin. J. Biol. Chem. 276, 7661–7671 29. Schwarz, J., Schwarz, S. C., Dorigo, O., Stu¨ tzer, A., Wegner, F.,

Labarca, C., Deshpande, P., Gil, J. S., Berk, A. J., and Lester, H. A. (2006) Enhanced expression of hypersensitive␣4* nAChR in adult mice increases the loss of midbrain dopaminergic neurons. FASEB J. 20, 935–946

30. Yu, M. J., Miller, R. L., Uawithya, P., Rinschen, M. M., Khosits-eth, S., Braucht, D. W., Chou, C. L., Pisitkun, T., Nelson, R. D., and Knepper, M. A. (2009) Systems-level analysis of cell-specific AQP2 gene expression in renal collecting duct. Proc. Natl. Acad. Sci. U. S. A. 106, 2441–2446

31. Pfeffer, M., Mu¨ ller, C. M., Mordel, J., Meissl, H., Ansari, N., Deller, T., Korf, H. W., and von Gall, C. (2009) The mammalian molecular clockwork controls rhythmic expression of its own input pathway components. J. Neurosci. 29, 6114 – 6123 32. Ren, X. P., Wu, J., Wang, X., Sartor, M. A., Qian, J., Jones, K.,

Nicolaou, P., Pritchard, T. J., and Fan, G. C. (2009) MicroRNA-320 is involved in the regulation of cardiac ischemia/reperfu-sion injury by targeting heat-shock protein 20. Circulation 119, 2357–2366

33. Towers, E., Gilley, J., Randall, R., Hughes, R., Kristiansen, M., and Ham, J. (2009) The proapoptotic dp5 gene is a direct target of the MLK-JNK-c-Jun pathway in sympathetic neurons. Nucleic Acids Res. 37, 3044 –3060

34. Ramagopalan, S. V., Maugeri, N. J., Handunnetthi, L., Lincoln, M. R., Orton, S. M., Dyment, D. A., Deluca, G. C., Herrera, B. M., Chao, M. J., Sadovnick, A. D., Ebers, G. C., and Knight, J. C. (2009) Expression of the multiple sclerosis-associated MHC class II Allele HLA-DRB1*1501 is regulated by vitamin D. PLoS Genet. 5, e1000369

35. Guillot, T. S., Richardson, J. R., Wang, M. Z., Li, Y. J., Taylor, T. N., Ciliax, B. J., Zachrisson, O., Mercer, A., and Miller, G. W. (2008) PACAP38 increases vesicular monoamine transporter 2 (VMAT2) expression and attenuates methamphetamine toxic-ity. Neuropeptides 42, 423– 434

36. Ekins, S., Nikolsky, Y., Bugrim, A., Kirillov, E., and Nikolskaya, T. (2007) Pathway mapping tools for analysis of high content data. Methods. Mol. Biol. 356, 319 –350

Received for publication July 1, 2009. Accepted for publication January 21, 2010.

數據

TABLE 1. Agents used in regulation studies
Figure 1. Search for expression systems. A) Endogenous SLC18A2 expression in human cell lines
Figure 2. Regulation of endogenous SLC18A2 activity in SH-SY5Y by various agents, based on qRT-PCR analysis of mRNA levels
Figure 3. Characterization of stable SH-SY5Y/pSLC18A2– 6.3A-luc cell line A7. A) Luc immunoreactivity in naive and A7 cells
+3

參考文獻

相關文件

The following should be included in the histopathological di fferential diagnosis: hyalinizing clear cell carcinoma of the salivary gland, clear cell variant of central

Before along with the evolution of time, the development of science and technology, the progress of cell phone, the cell phone that we used, from the beginning has the basic

The prepared nanostructured titania were applied for the photoanodes of dye-sensitized solar cell.. The photoanodes were prepared by the doctor blade technique and the area

That, if a straight line falling on two straight lines makes the interior angles on the same side less than two right angles, the two straight lines, if produced indefinitely, meet

To enable the research team to gain a more in- depth understanding of the operation of the Scheme, 40 interviews were conducted, including 32 in eight case study

Two examples of the randomly generated EoSs (dashed lines) and the machine learning outputs (solid lines) reconstructed from 15 data points.. size 100) and 1 (with the batch size 10)

magnetic field lines that pass through the loop is

The case where all the ρ s are equal to identity shows that this is not true in general (in this case the irreducible representations are lines, and we have an infinity of ways