Psychopharmacology (2004) 174: 463–476 DOI 10.1007/s00213-004-1840-8

REVIEW

J. F. Cubells . C. P. Zabetian Human genetics of plasma dopamine β-hydroxylase activity: applications to research in psychiatry and neurology

Received: 5 August 2003 / Accepted: 5 February 2004 / Published online: 16 April 2004 # Springer-Verlag 2004

Abstract Rationale: Norepinephrine (NE) is a key ergic dysfunction to a variety of symptoms in Parkinson’s neurotransmitter in the central and peripheral nervous disease and other degenerative neurological disorders. systems. Dopamine β-hydroxylase (DβH) catalyzes the Conclusions: A model is proposed, in which lower levels synthesis of NE from dopamine (DA) and occurs in the of DβH protein may lead to elevated ratios of DA to NE. plasma as a stable heritable trait. Studies of this trait have This model may explain associations between lower been useful in psychiatric and neurological research. plasma DβH activity and vulnerability to psychotic Objective: To selectively and critically review the litera- symptoms. Genotype-controlled analysis of plasma DβH ture on plasma DβH, and on recent progress under- holds promise for promoting further progress in research standing the molecular genetic basis for its inheritance. on psychiatric and neurological disorders. Based on this review, directions for future research in psychiatry and neurology will be suggested. Methods: Keywords Catecholamines . Norepinephrine . We selectively review the literature on the biochemical and Quantitative trait locus . Sympathetic nervous system . molecular genetics of plasma DβH activity, as well as Psychosis . Depression . Alcoholism . Attention deficit research on plasma and cerebrospinal fluid (CSF) DβHin hyperactivity disorder . Parkinson’s disease . Hypotension psychiatric and neurological disorders. Results: Strong evidence implicates DBH, the structural locus encoding DβH enzyme, as the major quantitative trait locus influencing plasma DβH activity, with one single nucle- Norepinephrine (NE) mediates many important functions otide polymorphism (SNP) accounting for up to 50% of in the central and peripheral nervous systems. It is the the variance. Mutations at DBH appear to be responsible major post-ganglionic neurotransmitter of the sympathetic for the rare syndrome of DβH deficiency. Some nervous system (SNS), where its release regulates vascular biochemical and genetic studies suggest associations tone and cardiac contractility, among other vital functions. between low plasma or CSF DβH and psychotic symp- NE is also a hormone and a precursor for epinephrine in toms in several psychiatric disorders. Studies combining neurosecretory cells of the adrenal medulla. Both genotyping at DBH with biochemical measurement of hormones contribute to “fight–flight” endocrine responses plasma DβH have proven useful in studies of schizophre- such as vascular and cardiac responses, and mobilization nia, cocaine-induced paranoia (CIP), depression, attention of glucose from hepatic glycogen stores. In the central deficit hyperactivity disorder, and alcoholism. Such nervous system, NE is localized within several neuronal studies may also elucidate the contribution of noradren- populations in the hindbrain and midbrain. The locus ceruleus (LC) is the largest of these, and accounts for most J. F. Cubells (*) of the noradrenergic innervation of the forebrain. Norad- Department of Psychiatry, Yale University School of Medicine renergic neurons of the LC project widely throughout the and VA Connecticut Health Care System, brain, innervating virtually every gray-matter region. 950 Campbell Avenue, Numerous sub-cortical brain areas, as well as the neocor- West Haven, CT 06516, USA e-mail: [email protected] tex, are richly innervated by NE-containing fibers (Amaral Tel.: +1-203-937-4943 and Sinnamon 1977; Moore and Bloom 1979; Foote et al. Fax: +1-203-937-3897 1983). NE thus contributes to regulation of many centrally mediated physiological, cognitive, and behavioral func- C. P. Zabetian Department of Neurology, University of Washington School of tions (Grace et al. 1998; Arnsten 2000a,b). Medicine and VA Puget Sound Health Care System, Seattle, WA 98108, USA 464 NE is a target for medication treatment of many human diseases, and may also mediate pathophysiological processes. For example, medications influencing NE- mediated transmission are widely used in treatment of medical illnesses such as hypertension and myocardial ischemia (Esler et al. 2001); neurological conditions such as sleep disorders (Mitler et al. 1993), neurodegenerative disorders (Schirger et al. 1981), and migraine headaches (Holroyd et al. 1991); and psychiatric disorders such as major depressive disorder (Anand and Charney 2000), attention deficit-hyperactivity disorder (ADHD; Arnsten 2000a), and anxiety disorders (Kent et al. 2002). Recent Fig. 1 Diagram of NE synthesis within a hypothetical noradrener- data suggest that functional genetic variation in the β1 and gic neuron, showing localization of DβH within synaptic vesicles. α2c adrenergic receptor proteins modifies risk for poor Dopamine is synthesized in the cytoplasm, and then transported into outcomes in cardiovascular disease (Small et al. 2002). the vesicle by the vesicular monoamine transporter, which derives its energy from the proton gradient across the vesicular membrane. One of the premises underlying the present review is that Abbreviations: tyr, tyrosine; TH, tyrosine hydroxylase; AADC, genetic studies of other NE-related proteins will similarly aromatic L-amino acid decarboxylase elucidate medical, neurological and psychiatric disorders. We focus on DBH, the locus encoding dopamine β- hydroxylase (DβH), the enzyme catalyzing the synthesis vesicular proteins are released from sympathetic nerve of NE from dopamine (DA; Kaufman and Friedman terminals (Smith et al. 1970; Weinshilboum et al. 1971), 1965). and that DβH can be assayed in plasma (Weinshilboum As schematized in Fig. 1,DβH is specifically expressed and Axelrod 1971) led to attempts to use plasma DβH in NE-containing neurons; it is the only catecholamine- levels as an index of sympathetic noradrenergic tone. synthetic enzyme localized within synaptic vesicles, where However, it quickly became clear that plasma DβH levels it occurs in both soluble and membrane-bound forms were quite stable within individuals, even during strenuous (Stewart and Klinman 1988). As a result of its vesicular exercise and cold pressor stimulation, both known to localization, DβH is released together with NE and other produce large increases in circulating catecholamines vesicular contents during synaptic transmitter release (Winer and Carter 1977; Peronnet et al. 1985). The lack (Smith et al. 1970; Weinshilboum et al. 1971). Some of substantial acute changes in plasma DβH levels DβH protein might also be secreted via a constitutive following these stimuli probably reflected the large exocytotic pathway (Oyarce and Fleming 1991). By virtue difference in metabolic kinetics of plasma catecholamines of its secretion into the extracellular space, DβH occurs in versus DβH protein, as the half-life of circulating DβHis the cerebrospinal fluid (CSF), and in plasma or serum (we estimated at 4.2 days in rats (Grzanna and Coyle 1978). henceforth refer to plasma or serum DβH only as plasma Developmental and longitudinal studies show that DβH). Typically, plasma DβH is assayed as enzyme plasma DβH activity is remarkably stable within subjects activity under saturating substrate and co-factor conditions after the age of 5 years (Ogihara et al. 1975; Weinshil- (Nagatsu and Udenfriend 1972), but levels of DβH protein boum 1978; Heiss et al. 1980). In contrast, plasma DβH have also been estimated directly using radioimmune activity varies widely across unrelated individuals (Wein- assay (Dunnette and Weinshilboum 1976;O’Connor et al. shilboum et al. 1973). Robust correlations among levels of 1983, 1994). Since the two sets of measures correlate plasma DβH in first-degree relatives (Weinshilboum et al. strongly (r >0.80: Ebstein et al. 1973; Dunnette and 1973), and nearly perfect correlations between monozy- Weinshilboum 1976;O’Connor et al. 1983), we will use gotic twins (Ross 1973) strongly supported the hypothesis the terms plasma DβH activity and plasma DβH levels that genetic mechanisms contributed to inter-individual interchangeably. The remainder of this paper will review variation in plasma DβH. Population and family genetic recent progress in understanding molecular genetic studies by Weinshilboum et al. (1975) suggested that only mechanisms regulating plasma DβH levels, studies of a few genes accounted for the heritable variation in plasma plasma DβH activity and of DBH genotypes in psychiatric DβH activity. A subgroup of subjects in these studies and neurological disorders, and suggest specific areas of exhibited “very-low” plasma DβH levels, defined as research that may benefit from further genetic and levels below 50 IU/ml in a population distribution ranging biochemical analysis of plasma DβH activity. from approximately 0–2000 nmol/h per ml. Segregation analysis of the very-low DBH trait suggested that individuals exhibiting very low serum DBH activity Overview of past work: plasma DβH is a quantitative were homozygotes carrying an allele at a single hypothe- genetic trait tical locus, which Weinshilboum et al. named “DBHL.” The molecular identity of this locus was unknown at the Weinshilboum (1978) provides a comprehensive review of time. the seminal work on plasma DβH activity, which is very Linkage studies using phenotypic markers subsequently briefly outlined here. The discoveries that DβH and other provided a major step forward towards identifying DBHL. 465 Building on the early work of Elston et al. (1979), Goldin molecular evidence that variation at DBH influences CSF et al. (1982) reported positive evidence for linkage levels of DβH. between a locus influencing plasma DβH activity and Further analysis at DBHshowed that a 19 base-pair the blood-group locus ABO. This observation was insertion/deletion polymorphism (−4784–4803del; Porter confirmed and extended by Elston et al. (Asamoah et al. et al. 1992), located 118 bp upstream of the STR, also 1987; Wilson et al. 1988, 1990), who reported a maximum associated with plasma DβH (Cubells et al. 2000). These lod score >5.0 between ABO and plasma DβH activity. experiments established that sequence variation within the Simultaneously, Lamouroux et al. (1987) cloned the DBH locus strongly associated with phenotypic variation cDNA encoding DβH protein, permitting molecular in DβH activity in the plasma, and importantly, in CSF as mapping by fluorescent in situ hybridization (Craig et al. well. However, individuals representing exceptions to 1988), and molecular linkage studies (Gelernter et al. apparent association rules were observed for each of the 1991; Perry et al. 1991). Those analyses localized the foregoing polymorphisms; i.e. homozygotes for low structural locus DBH to chromosome 9q34, immediately activity-associated alleles exhibited high plasma DβH adjacent to ABO. These findings converge to provide levels, and vice versa. These exceptions suggested that the strong positional and functional evidence that the ABO- observed associations resulted from linkage disequilibrium linked locus controlling plasma DβH activity is the (LD) with an as-yet unidentified molecular variant structural locus DBH. Association studies, reviewed in constituting the true functional polymorphism(s) respon- “Molecular genetic association analysis of plasma DβH”, sible for variation in plasma DβH activity. (LD term refers have substantiated and extended this hypothesis. to polymorphisms in close physical proximity on a chromosome, such that a specific allele at one locus has significant predictive value for specific alleles at the Molecular genetic association analysis of plasma DβH others.) In order to map functional variation at DBH more finely, Figure 2 shows the locations of selected polymorphisms at Zabetian et al. (2001) used an extreme-phenotype strategy. DBH, relative to the structure of the gene. Wei et al. They examined DNA sequence in four samples from (1997a) reported the first association between molecular individuals with very-low, two from those with average, alleles at DBH and plasma DβH, observing a significant and two from those with high plasma DβH, by re- relationship between a short tandem repeat polymorphism sequencing all 12 exons of DBH, together with approxi- (STR, Fig. 1) and plasma DβH in a set of unrelated British mately 1.5 kb of 5′ upstream sequence, and at least 50 bp individuals. Of five alleles at this polymorphism, the two at the 3′ and 5′ ends of each intron. This experiment most common (A3 and A4) accounted for more than 70% identified a C–T transition located 1021 bp upstream to the of the total. The A3 allele associated with lower levels of ATG translational start codon (−1021C→T) that showed a DβH, and A4 with higher levels. These observations were clear pattern of association to plasma DβH activity: all confirmed and extended by Cubells et al. (1998), who four very-low DβH individuals were homozygous for the showed the same pattern of allelic association at the STR T allele, while the remainder were homozygous for the C polymorphism to plasma DβH in a sample of European- allele at −1021C→T. To test the hypothesis that American (EA) patients with affective and anxiety −1021C→T was a functional variant, they developed a disorders. They also showed that a synonymous single rapid genotyping method for the variant, and examined its nucleotide polymorphism (SNP), 444A→G, located in association to plasma DβH activity in samples from EAs, exon 2 of DBH (Kobayashi et al. 1989), associated with African–Americans (AA), and ethnic Japanese (Jp). In all plasma DβH levels, and with levels of DβH protein in the three populations, very-low DβH individuals carried two CSF. The latter observation was the first, and to date, only, copies of the T allele at −1021C→T, heterozygotes showed intermediate levels of DβH, and individuals homozygous for the C allele showed higher mean levels (Fig. 3). The strong association of TT genotype with very- low DβH, together with the cross-population consistency of this association, strongly supports the hypothesis that −1021C→T, or another polymorphism in very tight LD with it, is the variant at DBH controlling plasma DβH levels. Furthermore, the T allele appears to be identical to (or in extremely tight LD with) the variant that Weinshilboum had dubbed DBHL some 25 years earlier. Given its location in the 5′ region of the gene, and the observation that very-low DβH individuals have lower circulating levels of DβH protein, the simplest hypothesis Fig. 2 Polymorphisms at the DBH locus. Selected polymorphisms explaining the relationship between −1021C→T and are shown in their relative positions across DBH. The boxed β − polymorphisms represent the two best functional candidates based plasma D H activity is that 1021T lowers expression on currently available evidence. STR short tandem repeat polymor- of DβH relative to −1021C by diminishing DBH gene phism transcription. The functional candidacy of −1021C→T has 466 Fig. 3 Distribution of square- root DβH activity by genotype at −1021C→T in European– Americans. In an additive ge- netic model, genotype at this SNP accounted for approxi- mately 44% of the variance in plasma DβH. Data were square- root transformed to stabilize variances across the genotype groups. Reprinted by permission from Zabetian et al. 2001. © American Society of Human Genetics

recently been bolstered further by the demonstration that TBH. In support of these predictions of functional the magnitude of its association to plasma DβH activity is tolerance, an analysis of variance of plasma DβH activity far greater than that of 11 other SNPs evenly spaced across and −1021C→T, which sequentially added each of these the DBH locus (Zabetian et al. 2003a). non-synonymous SNPs to the model, found that R535C In addition to −1021C→T, three DBH variants have increased the R2 by 0.02, (P=0.0024), while the others did merited consideration as potentially common functional not explain significant additional variance (Zabetian et al. polymorphisms in European-derived populations. All 2001). Interestingly, the only observed homozygote for three are non-synonymous (amino acid-altering) SNPs, 535C (who was heterozygous at −1021C→T) had an located in exons 3 (A197T), 5 (A304S), and 11 (R535C), undetectable level of plasma DβH activity. These data with minor allele frequencies in EAs of approximately support a putative function effect for R535C, but not for 0.05–0.10 (Zabetian et al. 2001). Ishii et al. (1991) A197T or A304S. reported a 13-fold difference in homospecific DBH The following sections will selectively review studies of activity between the 304A and 304S alleles in transient plasma or CSF DβH levels, or molecular variation at transfection assays of COS cells. However, Li et al. (1996) DBH, in human psychiatric or neurological disorders. repeated these experiments in transformed Drosophila Schneider 2 cells, but also performed purification procedures prior to enzymatic assays, and found no DβH deficiency significant difference in homospecific activity (or any other enzyme property) between the two forms. We have Human DβH deficiency is an exceedingly rare disorder of used SIFT (Ng and Henikoff 2001), a program that uses the SNS, first described by Robertson and colleagues in sequence homology to predict whether an amino acid the US (1986), and Man in’t Veld et al. (1987) in the substitution affects protein function, to examine the Netherlands. DβH-deficient patients exhibit profound potential functional consequences of A197T, A304S, and orthostatic hypotension, and a variety of other signs and R535C. SIFT predicted that both A197T and A304S symptoms of SNS failure. Biochemically, they lack should be well tolerated, and inspection of sequence detectable peripheral NE or NE metabolites, and exhibit alignments revealed that the amino acid encoded by the elevated plasma dopamine and dopamine metabolites, minor allele of each SNP occurred at the homologous consistent with a metabolic block at the synthetic step position in proteins closely related to human DBH. For catalyzed by DβH. Plasma DβH is also undetectable in example, at the positions corresponding to 197 and 304 in this syndrome (Robertson et al. 1986; Man in’t Veld et al. human DBH, threonine replaces alanine in bovine DBH, (1987), as are CSF DβH immunoreactive protein and serine replaces alanine in human monooxygenase X (O’Connor et al. 1994) and DβH immunoreactivity in (Chambers et al. 1998), respectively. However, SIFT sympathetic fibers (Mathias et al. 1990). The latter predicted that the third SNP, R535C, would be poorly evidence suggests that DβH protein is absent in this tolerated and thus might affect DBH function. Arg is syndrome, at least in the tissues examined thus far. NE conserved in all available DBH sequences that include the synthesis can be restored in DβH-deficient patients by 3′ end of the gene (five species of mammals), but Lys administration of the β-hydroxylated precursor of NE, occurs in human/mouse MoX and Ile occurs in Drosophila dihydroxy phenylserine (DOPS), which is converted to 467 NE by decarboxylation catalyzed by aromatic amino acid has been reported, are compound heterozygotes for IVS1 decarboxylase. DOPS has proven a remarkably effective +2T→C and one or more of a variety of rare missense or treatment for DβH deficiency (Biaggioni and Robertson deletion mutations in the coding region. 1987), reversing the most disabling symptoms of the An important unsolved problem in molecular analysis syndrome almost completely. Interestingly, no neuropsy- of DβH deficiency is that missense and deletion mutations chiatric phenotype has been reported in patients with DβH account for absence of DβH enzyme activity, but not for deficiency. Whether DβH-deficient patients lack central the apparent absence of immunoreactive DβH protein in DβH and NE is not known. DβH-deficiency patients (Mathias et al. 1990;O’Connor Using the re-sequencing strategy developed by Zabetian et al. 1994). It is possible that the mutations leading to et al. (2001), Kim et al. (2002) showed that rare sequence D100E, V87M and D331N could disrupt splicing, since all variants at DBH associate with DβH deficiency. In two are near intron–exon junctions. Another potential mech- unrelated patients with the syndrome, they found that each anism explaining how these mutations lead to absence of patient carried a single copy of a T→C transition (IVS1 detectable DβH protein is that the mutant peptides do not +2T→C), located within the invariant GT of the donor undergo appropriate post-translational processing, and splice site at the 5′ end of intron 1. Jansen et al. (2001) therefore fail to translocate into nascent dense-core described a case of DβH deficiency who is a homozygote vesicles (or that structure is altered such that they undergo for IVS1+2T→C, strongly suggesting the mutation, or the rapid degradation even if they are targeted to vesicles). haplotype(s) on which it rides, is sufficient to cause DβH These hypotheses require further experimental study. deficiency. However, although Kim et al. (2002) demon- The presence of the splice-disrupting C allele of IVS1 strated that the mutation alters normal splicing of mRNA +2T→C in at least one copy in several unrelated DβH- and results in a transcript containing nonsense sequence deficient patients suggested that this mutation might occur and a premature stop codon, their in vitro analysis detected at a high enough frequency to be detectable in population a small amount of appropriately spliced transcript in the studies. Indeed, in a series of 88 unrelated EA individuals, presence of the C (presumed mutant) allele. This Kim et al. (2002) observed two individuals heterozygous observation raises the possibility that complete absence for this variant (apparent allele frequency=0.011). How- of detectable DβH protein in deficient patients may ever, a recent study of a much larger sample (801 require a “second hit” elsewhere in the gene that further individuals, including 456 EAs and 223 ethnic Germans) suppresses expression of the normal transcript. The DBH found no further examples of the C allele in EAs or haplotype shared by both patients for whom data have Germans, and only one heterozygote in 122 samples from been published contains the T (low-DβH associated) allele AA subjects (Zabetian et al. 2003b). IVS1+2T→C thus of −1021C→T (Zabetian et al. 2003b), suggesting that low appears to be an exceedingly rare variant. expression associated with this allele might contribute to the molecular etiology of the DβH deficiency syndrome. Each of the two DβH deficiency cases examined by Other psychiatric and neurological disorders Kim et al. (2002) also carried rare missense mutations at DBH in trans (i.e. on the opposite chromosome) to the Numerous studies have examined plasma DβH levels, or chromosome bearing IVS1+2T→C. One patient carried a CSF DβH levels, in patients with psychiatric and C→A transversion (D100E) that results in a change in the neurological disorders. Unfortunately, almost all of these encoded amino acid from aspartic acid to glutamic acid at studies were performed before DBH was identified as a position 100 in the primary peptide sequence. Although major locus influencing plasma DβH levels. It is perhaps this change, from one acidic amino acid to another, not surprising, therefore, that many of these studies led to appears at first glance to be conservative, examination of conflicting, inconclusive, or uninterpretable results. More copper-dependent mono-oxygenases from several verte- recently, genotype-based studies have been conducted. brate and invertebrate species shows an invariant aspartic The following sections will selectively review these acid residue at this position (Kim et al. 2002), suggesting studies. that the missense mutation could disrupt enzyme function. The second DβH-deficient patient carried two missense mutations within DBH, both in trans to the IVS1+2T→C DβH and psychosis mutation. The first, V87M, results in substitution of a methionine codon for the normal valine, because of a Studies of plasma or CSF DBH levels in schizophrenic G→A transition in exon 1; the second, D331 N, replaces patients, or from subjects exhibiting psychotic symptoms, the normal aspartic acid codon with one encoding have produced a variety of conflicting findings. For asparagine, due to a G→A transition within exon 6. example, Fujita et al. (1978), studied serum DβHin Jansen et al. (2001) have also reported a single nucleotide schizophrenic patients diagnosed by Schneiderian criteria, deletion (of A at position 575) in DBH that results in a and reported that serum DBH activity was significantly frameshift and premature stop codon, presumably result- lower in the schizophrenic group than in a non-schizo- ing in a protein with no catalytic activity. This mutation phrenic control group. These findings were consistent with was observed in trans to IVS1+2T→C. Thus, all but one an autopsy study in which DBH activity from a variety of patient with DβH deficiency, for whom molecular analysis brain regions was lower in chronic schizophrenic brains 468 (diagnostic criteria unspecified) than in control brains ation of haplotypes composed of these variants to cocaine- (Wise and Stein 1973). However, studies employing induced paranoia (CIP) in EA subjects dependent on operational diagnostic criteria for schizophrenia have cocaine. (The term haplotype refers to the presence of two found no difference between either the serum or CSF or more specific alleles in LD on a single chromosome: by DBH activities of patients and controls (Meltzer et al. yielding information about a stretch of chromosome, 1976, 1980; Sternberg et al. 1982, 1983; van Kammen et haplotypes are potentially more informative than single al. 1994). Several genetic association studies, examining a genotypes.) First, they established that significant LD variety of polymorphisms at DBH, have found no occurred between the low-activity-associated deletion evidence for associations to schizophrenia (Meszaros et allele (“del”)at−4784–4803del, and the low-activity al. 1996; Wei et al. 1997b; Williams et al. 1999; Arrufat et associated “A” allele at 444A→G. This observation al. 2000). Thus, it is now clear that neither plasma DβH allowed them to designate the “del-A” haplotype as the levels nor genotypes at DBH distinguish schizophrenic low-activity associated haplotype. They then estimated patients from healthy individuals. haplotype frequencies using an estimation–maximization While neither plasma DβH nor DBH sequence variants (EM) method (Long et al. 1995; EM uses frequencies of associate directly with schizophrenia, there is evidence haplotypes in cases in which haplotypes can be defini- that plasma and CSF DβH levels, or specific alleles at tively identified based on individual genotypes to estimate DBH, associate with clinically important characteristics of the frequencies of haplotypes in individuals in whom it is the disorder. These observations suggest the possibility not possible to specify precisely which haplotypes are that DBH is a modifying gene in schizophrenia. Sternberg present; in the present example, subjects heterozygous at et al. (1982, 1983) measured CSF DβH activities in 30 both −4784–4803del and 444 A→G cannot be definitively patients meeting Research Diagnostic Criteria (Spitzer et assigned haplotypes). Consistent with the hypothesis that al. 1975) for schizophrenia or schizoaffective disorder. low-DβH-associated variants associate with psychosis, DβH activity was significantly lower in the patients who they found that the del-A (i.e., low-activity) haplotype was responded to (typical) antipsychotic medications. In significantly more frequent (P=0.0007) in those who addition, those with lower DβH activities tended to endorsed CIP, as compared to those who did not. Since exhibit better premorbid adjustment, and were more likely 444A→G and −4784–4803del are both in strong LD with to be classified as “reactive” schizophrenics than “pro- -1021C→T, a large prospective study to test the hypoth- cess” schizophrenics. Thus, low CSF DBH activity esis that −1021C→T associates with CIP and other predicted more severe positive symptoms, but better psychotic phenomena in cocaine-dependent individuals overall clinical outcome in this small group of schizo- is currently under way. phrenic patients. Consistent with those findings, a factor Yamamoto et al. (2003) examined the relationship of the analysis of a large number of possible predictive variables same 444G→A/−4784–4803del haplotypes to treatment (Frecska et al. 1990) indicated that low serum DBH outcomes in EA schizophrenic patients. They classified activity possessed a small but significant degree of value patients according to their long-term responses to treat- as a predictor of good response to neuroleptics in ment, and selected two groups: those who had not schizophrenia. van Kammen et al. (1994) reassessed the experienced remission of psychotic symptoms for at results of Sternberg et al. (1983) by measuring DβH least 2 years despite adequate trials of two classes of protein levels in the CSF of 60 male schizophrenic patients antipsychotic medication (non-responders, NR), and those (diagnosed by DSM-IIIR criteria: American Psychiatric who experienced near-complete remission after initiation Association, 1984). Once again, low CSF DBH levels of antipsychotic medication, but who had relapsed when were associated with good premorbid social functioning. medications were reduced or withdrawn (responders, R). However, low CSF DBH was also associated with poor Consistent with the results of Cubells et al. (2000), they scholastic performance and fewer years of formal educa- found significant LD between del and A. As expected tion in these patients. Taken together, the findings suggest from prior observations of equivalent plasma DβH activity that heritable variation in plasma DβH levels may in schizophrenic patients versus controls (reviewed associate with cognitive or psychosocial differences in above), they found no difference in the frequency of del- schizophrenia, or with differences in illness severity and A or other haplotypes in schizophrenic patients as clinical outcome. Sternberg et al. (1983) suggested that compared to healthy control subjects. However, they did lower DβH activity may elevate DA levels, which in turn observe a significantly larger proportion of the low- may contribute to psychotic symptoms such as hallucina- activity-associated del-A haplotype in NR schizophrenic tions and delusions. patients as compared to R patients. They also showed that Since plasma DβH levels are under genetic control by carriers of at least one del-A haplotype had significantly variation at DBH, several studies have attempted to test the higher scores on the brief psychiatric rating scale (BPRS) hypothesis that genetic variants associated with lower than non-carriers. These results are consistent with the plasma DβH levels associate with differences in vulner- associations between more florid psychotic symptoms and ability to psychotic symptoms. After identifying 444A→G lower plasma DβH observed by Sternberg et al. (1983). and −4784–4803del as polymorphisms associated with Taken together, the studies of Cubells et al. (2000) and variation in plasma DβH activity (but prior to discovering Yamamoto et al. (2003) suggest that variants at DBH that −1021C→T), Cubells et al. (2000) examined the associ- associate with low DβH activity also associate with 469 vulnerability to “positive” psychotic symptoms (Crow While acknowledging that larger, better-designed stu- 1985) in idiopathic and psychostimulant-induced psycho- dies were clearly necessary, the authors proposed that the ses. chronic disturbances in hypothalamic-pituitary-adrenal (HPA) axis that occur in UDPF might mediate the diagnosis-related, but −1021C→T-independent, differ- Psychotic depression ences in plasma DβH levels. In this model, DBH gene expression decreases in response to the chronic elevations Also consistent with the hypothesis that lower levels of in circulating cortisol known to exist in UDPF (Nelson and DβH protein associate with vulnerability to psychosis are Davis 1997). Consistent with this model are observations four independent studies reporting significant associations that the proximal 5′ region of the DBH gene contains a between unipolar major depression with psychotic features glucocorticoid response element (GRE; Kobayashi et al. (UDPF) and low plasma DβH activity (Meltzer et al. 1989) and that glucocorticoids exert large effects on DBH 1976; Mód et al. 1986; Sapru et al. 1989; Meyers et al. transcription (McMahon and Sabban 1992). By lowering 1999). Only one study has not replicated this finding expression of the DBH gene, chronic HPA dysfunction (Lykouras et al. 1988). Based on these observations, could thereby lead to psychosis by lowering DβH protein Cubells et al. (2002) postulated that UDPF associates with levels. Lower levels of DβH protein could, in turn, elevate molecular variation at −1021C→T, or at non-synonymous central DA/NE ratios. While data on central DA/NE ratios SNPs at DBH. They compared genotypes and plasma in UDPF do not exist, a study of the peripheral ratio of DβH activity in samples from patients with UDPF to those these catecholamines indeed shows elevated DA/NE with unipolar major depressive disorder without psychosis (Rothschild et al. 1984). (UD). Consistent with previous studies, plasma DβH Further support for the hypothesis that lower levels of activity was significantly lower in the UDPF samples than DβH can increase the risk of psychotic symptoms comes in the UD samples. However, a two-way analysis of from studies showing that DβH inhibitors can predispose variance of plasma DβH activity, using diagnosis and humans to psychosis. In healthy college students co- genotype at −1021C→T as independent variables, showed administration of L-dopa and fusaric acid (an inhibitor of significant main effects of both genotype and diagnosis, DβH) significantly increased psychosis related items of but no significant interaction. The UDPF-associated the BPRS, such as somatic concern, conceptual disorga- decrement in plasma DβH activity was thus independent nization, motor retardation, and unusual thoughts (Hart- of genotype at −1021C→T. Allele frequency analysis mann and Keller-Teschke 1979). Fusaric acid also showed no association of −1021C→T, nor any non- exacerbated psychosis in bipolar patients with psychotic synonymous SNPs at DBH, to UDPF. A trend (P=0.07) features (Sack and Goodwin 1974). Disulfiram, in addition toward association was observed at R535C, the SNP in to its inhibitory effects on aldehyde dehydrogenase also exon 11 (discussed earlier) that associates significantly potently inhibits DβH. Major et al. (1979) found that low with plasma DβH activity independently from levels of CSF DβH predicted disulfiram-induced psycho- −1021C→T (Zabetian et al. 2001). The study had several sis in alcoholics treated with disulfiram. Finally, inhibitors shortcomings, most notably small sample sizes, and of DβH elevate DA/NE ratios (Stanley et al. 1997), and reliance on “samples of convenience” that were not increase potassium-stimulated release of DA from striatal prospectively matched for age, sex, medication treatments slices (Fischer et al. 1980). or other potentially important variables.

Fig. 4 Hypothetical model contrasting NE neurons with high or low inheritance or pharmacologic inhibition), allowing DβH to become DβH activity. In a, higher vesicular DβH activity efficiently the rate-limiting step in NE biosynthesis. In this situation, the ratio converts DA to NE, and the neuron releases little or no DA during of DA to NE in the synapse would be higher, with greater relative synaptic activity. This low DA/NE ratio in the synapse thus favors stimulation of DA receptors, and relatively inefficient α2 receptor- stimulation of presynaptic and post-synaptic NE receptors, with little mediated negative feedback on NE synthesis and release. Conditions to no activity at DA receptors. In addition, NE synthesis and release favoring a rate-limiting role for DβH might include high-demand are under efficient negative feedback through stimulation of situations such as acute stress or psychostimulant intoxication presynaptic α2 receptors. In b,DβH activity is low (due to genetic 470 In summary, findings from a variety of studies suggest inter-individual variance in plasma DβH. The residual associations between lower plasma DβH and vulnerability variance presumably reflects a combination of other loci to psychotic symptoms in several idiopathic and drug- influencing plasma DβH (Wilson et al. 1988, 1990; induced psychoses. A model that unifies these findings is Province 2000), and environmental or physiological summarized in Fig. 4. In the model, we postulate that influences on the measure. The plethora of inconclusive under certain circumstances, DβH can be rate-limiting for studies of plasma DβH, published largely in the 1970’s NE synthesis, leading to less efficient conversion of DA to and 1980’s, thus bears re-examination in genotype- NE. This metabolic “bottleneck” would therefore elevate controlled studies. vesicular and synaptic DA/NE ratios, possibly leading to emergence of psychotic symptoms. If such a mechanism operates in the human brain, it is most likely to be Plasma DβH and childhood psychiatric disorders: ADHD functionally important in regions that receive substantial and conduct disorder noradrenergic innervation and also contain DA receptors. The prefrontal cortex (PFC) is an example of such a brain Biochemical studies have not shown strong relationships region. Interactions between NE and DA receptor- between ADHD and plasma DβH (Rogeness et al. 1982; mediated neurotransmission are known to alter PFC Bowden et al. 1988). Both of the latter studies, however, function in ways thought to be important to schizophrenic found interesting results in subsets of children with ADHD psychopathology (Arnsten 2000b), and NE fibers in this and conduct disorder. A study examining DSM-III region appear capable of taking up and releasing DA conduct disorder and ADHD in boys (mean age 11 (DeVoto et al. 2001). Experiments examining the impact years) found that regardless of the presence or absence of of DβH inhibitors on PFC function might elucidate DA– ADHD, the DSM-III diagnosis of conduct disorder, NE interactions, and the role of DβH, in schizophrenia undersocialized type (CDU) associated with significantly and other psychotic disorders. lower plasma DβH activity (Rogeness et al. 1982) than did conduct disorder, socialized type (CDS). Interestingly, many of the boys with CDU had very low DβH values. Genotype-controlled analysis of plasma DβH in alcohol This observation was replicated in several later studies dependence (Rogeness et al. 1984, 1986, 1989; Bowden et al. 1988), but not by all (Pliszka et al. 1988). Given this intriguing Alcohol withdrawal induces profound, sometimes life- set of findings, together with a recent linkage study threatening, changes in SNS function. Köhnke et al. suggesting possible linkage between adolescent-offender (2002) therefore postulated that −1021C→T, by altering substance abuse and markers on 9q34 (Stallings et al. DβH activity and changing NE synthesis, would associate 2003), molecular genetic analysis of the DBH locus, and with delirium tremens (DT). To test the hypothesis, they genotype-controlled analysis of plasma DβH, in children measured plasma DβH activity and determined genotypes with CD seem overdue. at −1021C→T in healthy subjects, and in alcoholic A growing number of studies have tested for associa- subjects with and without histories of DT. They found tions between ADHD and genotypes at IVS5+192C→T, a no association between DT and −1021C→T alleles or DBH SNP of no known function located in intron 5, which plasma DβH levels (measured at least 3 weeks after is easily genotyped by differential cleavage with the recovery from DT). However, similar to the results of restriction endonuclease TaqI. Daly et al. (1999) con- Cubells et al. (2002) in UDPF, plasma DβH activity was ducted haplotype-based haplotype relative risk (HHRR: significantly lower in alcoholic subjects (regardless of DT Terwilliger and Ott 1992) and transmission-disequilibrium status) compared with healthy subjects, independent of analysis (TDT: Spielman et al. 1993) on Irish parent- genotype at −1021C→T. The simplest explanation of offspring trios ascertained by the presence of ADHD in the these findings, that chronic ethanol is toxic to NE neurons, children. They found significant evidence for differential is supported by autopsy results showing de-pigmentation transmission of the “A2” (cleaved, or “T”) allele of IVS5 of the LC in brains from chronic alcoholics (Arango et al. +192C→T, an effect that appeared to be more robust in 1994). families containing at least one parent retrospectively In addition to providing evidence for ethanol-related diagnosed with ADHD. These findings supported linkage changes in NE function, the results of Köhnke et al. (2002) and association between ADHD and DBH. Roman et al. are important because (1) they replicate the association of (2002) also found HHRR evidence for association −1021C→T to plasma DβH activity in an independent between ADHD and the A2 allele of IVS5+192C→T, in sample of Europeans; and (2) they confirm the utility of a familial sample gathered in Brazil. However, they noted simultaneous measurement of plasma DβH and genotypes a stronger finding from families that did not contain at −1021C→T. Since the variance of plasma DβH activity parents with ADHD, a pattern opposite to the trend is smaller within each genotypic group, comparisons of observed by Daly et al. (1999). Another attempt to plasma DβH accounting for genotype at −1021C→T are replicate these findings in ADHD parent-offspring trios more accurate than comparisons made without taking found no significant differential transmission of the TaqI genotype into account. In other words, determination of allele (one-sided p=0.07), nor of a 3-locus haplotype genotypes at −1021C→T can factor out up to 50% of the composed of the DBH STR, −4784–4803del, and the TaqI 471 allele (Wigg et al. 2002). Although a full discussion of responsible for the profound sympathetic failure charac- family-based association methods is beyond the scope of teristic of the disease (Wenning et al. 1997). In Parkinson’s this review, it is important to note that unlike case— disease (PD), marked cell loss, degenerative changes control association studies, which are based only on including the presence of Lewy bodies, and diminished comparison of allele or haplotype frequencies in two activity of catecholaminergic enzymes have consistently groups of unrelated individuals, HHRR and TDT studies been reported within both the LC and SNS (Nagatsu et al. focus on alleles transmitted from parents to offspring. 1979; Mann and Yates 1983; Wakabayashi and Takahashi These family-based methods are therefore robust to 1997). Several groups have suggested a link between LC potential problems associated with coincidental differ- degeneration and the increased incidence of dementia in ences in allele frequency between comparison groups patients with PD (Mann and Yates 1983; Gaspar and Gray related to between-group differences in ancestral geo- 1984; Cash et al. 1987). More recently, widespread loss of graphic population origins (a phenomenon known as postganglionic sympathetic neurons in PD has been “population stratification”). demonstrated by 123I-MIBG SPECT imaging (Yoshita Müller Smith et al. (2003) recently reported a case- 1998; Braune et al. 1999; Satoh et al. 1999; Druschky et control association between ADHD and the TaqIallele, in a al. 2000) and likely plays a role in the SNS dysfunction prospectively gathered sample of subjects who are that affects a subgroup of PD patients (Singer et al. 1992; currently adults, but were ascertained during childhood. Magalhaes et al. 1995). In contrast to the two positive family-based studies, this In contrast to the lack of support for DβH as a marker of case-control study found an association between ADHD acute changes in SNS activity, there is evidence that and the A1 (uncleaved, or “C”) allele. Why Müller Smith plasma and CSF DβH levels might reflect long-term et al. (2003) observed a direction of association between trends in noradrenergic function. Treatment of rats for 5 ADHD and DBH opposite to that of the other groups is not weeks with guanethidine at doses causing partial destruc- clear. However, as pointed out by Müller Smith et al. tion of sympathetic nerves resulted in a decrease in plasma (2003), within-study stratification cannot be ruled out. In DBH activity of 53% (Grobecker et al. 1977). A similar another family-controlled study of ADHD, Payton et al. study using 6-hydroxydopamine found a 40% reduction in (2001) did not find evidence supporting differential rat serum DBH (Coyle et al. 1974). Studies of CSF in PD transmission of alleles of A304S (see above for discussion patients have consistently demonstrated diminished levels of this non-synonymous SNP). However, because A304S of DβH activity and/or protein by immunoassay. Nagatsu has a minor allele frequency of approximately 5% in et al. (1982), Mogi et al. (1988), Hurst et al. (1985), and Europeans (Cubells et al. 1997), that study was quite O’Connor et al. (1994) found DBH levels decreased to 6, underpowered. 16, 41, and 52%, respectively, of control values. CSF As is often the case in single-locus association studies, DβH is probably derived from both central (LC) and different groups have reported differing, and difficult-to- peripheral (sympathetic nerves, adrenal medulla) sources resolve, outcomes in association studies of DBH and (O’Connor et al. 1994). The literature is less clear in ADHD. Clearly, more studies and larger sample sizes will regard to PD-associated changes in plasma DβH activity. be necessary before a possible association between ADHD Lieberman et al. (1972) found DβH levels reduced to less and DBHcan be evaluated with confidence. No studies of than half that of controls in untreated PD patients, but no −1021C→T in ADHD, nor of genotype-controlled difference in L-dopa treated PD patients versus controls. analyses of plasma DβH activity in ADHD, have been Markianos et al. (1981) reported no difference in DβH reported yet. Whether this lack of results at the presumably activity between both treated and untreated PD patients functional marker reflects lack of study, or a publication versus controls, but did find diminished activity in the PD bias against negative results, remains unclear. subgroup with dementia. Nagatsu et al. (1982) described a significant reduction in plasma DβH activity levels in PD patients in comparison to age-matched controls. Thus, in Plasma DβH and DBH genotypes in neurodegenerative general, conditions which result in the loss of noradren- diseases ergic neurons are associated with diminished levels of DβH in CSF and plasma. Substantial neuronal loss within the LC and/or SNS has Compensatory changes in surviving dopaminergic been well documented in several neurodegenerative neurons have been well documented in PD and in MPTP diseases. In Alzheimer’s disease (AD), significant reduc- lesioned-rodents. These include increased levels (per cell) tion of neurons within the LC has been frequently reported of tyrosine hydroxylase (TH) mRNA in the substantia (Tomlinson et al. 1981; Burke et al. 1999), and some nigra and ventral tegmental area (Joyce et al. 1997; Tong authors have suggested that dysfunction of the central et al. 2000), and TH protein in the striatum (Bezard et al. noradrenergic system might be responsible for the 2000), possibly by increased transcriptional activation of increased incidence of depression and impaired stress the TH gene. Similar changes in noradrenergic neurons responses observed in AD (Forstl et al. 1994; Hoogendijk were reported by Burke et al. (1999) in a series of AD et al. 1999). In multiple system atrophy (MSA), degen- cases in which substantial (79%) cell loss within the LC eration of neurons within the hypothalamus and interme- was observed together with an 11-fold increase in DβH diolateral cell columns of the spinal cord is thought to be protein in the remaining neuronal cell bodies. Such 472 mechanisms, particularly transcriptional activation of sequence conserved across species (and therefore likely to catecholamine-synthetic genes, are probably critical to have functional significance) have yet to be conducted. In the ability of surviving noradrenergic neurons to function- addition, resequencing of intronic sequence and larger ally compensate in neurodegenerative diseases. Any portions of the 5′ and 3′ regions will be necessary to factors, either environmental or genetic, that interfere examine the possibility that functional sequence variation with these adaptations might result in an earlier onset or in LD with −1021C→T accounts for the strong association more severe symptoms of noradrenergic dysfunction. of this SNP to variation in plasma DβH activity. Second, Zabetian et al. (2001) postulated that −1021C→T could more studies examining the relationship of DBH geno- represent one such genetic factor, and serve as a marker for types to CSF DβH levels (Cubells et al. 1998), are the maximum potential production of DβH (i.e., NE- necessary. Another fundamentally critical issue is whether synthetic “functional reserve”) in noradrenergic neurons. plasma and CSF DβH activities, and DBH genotypes, are For example, in PD patients homozygous for the T allele correlated with cellular and tissue levels of DβH protein. of −1021C→T, the surviving noradrenergic neurons in the In studies led by C.P.Z., we are approaching this question SNS might have a reduced capacity to upregulate DBH by examining levels of DβH mRNA and protein in NE- expression, but appropriately increase TH expression. producing human tissues that have been genotyped at Thus, as the disease progressed and the population of −1021C→T. A second set of approaches to this issue will noradrenergic neurons continued to decline, a shift from involve rigorous testing of the hypothesis that −1021C→T TH to DβH as the rate-limiting enzyme in catecholamine alters transcription of DBH. K.S. Kim and colleagues synthesis would occur, eventually resulting in insufficient (personal communication) conducted a transient transfec- stores of NE. This hypothesis predicts that PD patients tion analysis, in DβH-expressing cultured human neuro- homozygous for −1021T would exhibit a substantial blastoma cells, of −1021C→T alleles on expression of deficit in the functional reserve of the SNS, causing an reporter constructs driven by a 1.14 kb fragment of the earlier onset and more severe symptoms of sympathetic human DBH proximal 5′ region. They did not observe a dysfunction than in heterozygotes or those homozygous transcriptional effect of the −1021C→T polymorphism in for −1021C. These predictions would hold true regardless this system. However, as shown in previous studies of whether −1021C→T was itself functional, or simply (Ishiguro et al. 1993), the reporter system they employed acting as an LD marker for an unrecognized functional is sensitive only to sequence variation within approxi- polymorphism elsewhere in the DBH gene. The same mately 600 bp of the transcriptional start site. The negative concepts would apply to central noradrenergic dysfunction results of the transient transfection analysis of −1021C→T in AD, MSA, and PD. are therefore inconclusive. Reporter-gene experiments on A study to examine whether −1021C→T is predictive of DBH promoter function in transgenic mice have shown the onset and severity of sympathetic dysfunction in PD is that an element necessary for tissue-specific expression of currently underway in one of our labs (C.P.Z.). Patients DβH resides between −600 and −1100 bp (Hoyle et al. with PD who are homozygous for either the C or T allele 1994); −1021C→T also resides in this region. Since the will be evaluated longitudinally for symptoms of SNS transgenic mouse system is sensitive to sequence changes dysfunction using a previously validated questionnaire in the region surrounding −1021C→T, it would presum- (Suarez et al. 1999). Signs of sympathetic failure will be ably afford more interpretable data on the variant’s assessed by noninvasive cardiovascular monitoring and by influence on transcription. Another important task is to direct measurements of plasma catecholamines in response identify other loci influencing plasma DβH. Phenotype- to postural changes. The results of this study could directly based linkage data suggest that at least one locus other impact future pharmacological strategies for treating than DBH influences plasma DβH (Wilson et al. 1988, orthostatic hypotension in selected cases of PD. For an 1990; Province 2000). This putative second locus, which example, DOPS has been used quite successfully to treat accounts for approximately 10–15% of the total variance orthostatic hypotension in the syndrome of DβH defi- in plasma DβH, appears to be linked to C3, the locus ciency (see above), and with modest success in a small encoding C3 complement-component protein, located on clinical trial of 10 patients with MSA and pure autonomic chromosome 19p. If the molecular identity of this second failure (Freeman et al. 1999). If homozygosity for the T locus can be ascertained, then it will be possible to allele at −1021C→T proves to render patients relatively examine the locus for associations to UDPF and alcohol- DBH-deficient in PD, DOPS therapy might be similarly ism, as discussed above. successful in these individuals. Basic studies on the influence of non-synonymous SNPs at DBH, especially of R535C, on the structure and function of DβH protein are necessary as well. Since DβH Future directions is a homotetrameric protein (O’Connor et al. 1983), functional variation in amino acid sequence raises the The data and hypotheses just reviewed raise several sets of interesting possibility that mixed homotetramers, com- basic questions on plasma DβH and DBHgenotype. First, posed of allelomorphic monomers, are expressed by further research on the relationship between sequence individuals heterozygous for non-synonymous SNPs. variation at DBH and plasma DβH activity is necessary. Thus, in addition to functional effects on monomer Comprehensive resequencing studies of noncoding DBH structure, which may alter the catalytic properties of the 473 holoenzyme, it is also possible that non-synonymous Burke WJ, Li SW, Schmitt CA, Xia P, Chung HD, Gillespie KN SNPs at DBH may act at the level of assembly of (1999) Accumulation of 3,4-dihydroxyphenylglycolaldehyde, the neurotoxic monoamine oxidase A metabolite of norepi- monomers into the dimers and tetramers thought to nephrine, in locus ceruleus cell bodies in Alzheimer’s disease: constitute the active enzyme (O’Connor et al. 1983). mechanism of neuron death. Brain Res 816:633–637 Additional studies of plasma DβH and DBH genotypes Cash R, Dennis T, L’Heureux R, Raisman R, Javoy-Agid F, Scatton are also necessary in studies of human disease. An B (1987) Parkinson’s disease and dementia: norepinephrine and dopamine in locus ceruleus. Neurology 37:42–46 important principle that has emerged in research thus far is Chambers KJ, Tonkin LA, Chang E, Shelton DN, Linskens MH, that more information can be gained from combined study Funk WD (1998) Identification and cloning of a sequence of plasma DβH and DBH genotypes, than from studies of homologue of dopamine beta-hydroxylase. Gene 218:111–120 either alone (Cubells et al. 2002; Köhnke et al. 2002). Coyle JT, Wooten GF, Axelrod J (1974) Evidence for extranora- β drenergic dopamine-beta-hydroxylase activity in rat salivary Genotype-controlled analysis of plasma D H holds real gland. J Neurochem 22:923–929 promise for analysis of disorders in many medical Craig SP, Buckle VJ, Lamouroux A, Mallet J, Craig IW (1988) specialties, including psychotic, affective, attentional, Localization of dopamine beta hydroxylase (DBH) gene to and substance-use disorders in psychiatry; neurodegener- chromosome 9q34. Cytogenet Cell Genet 48:48–50 ative and SNS disorders in neurology; and possibly for Crow TJ (1985) The two-syndrome concept: origins and current status. Schizophr Bull 11:471–486 cardiovascular diseases such as hypertension and myocar- Cubells JF, Kobayashi K, Nagatsu T, Kidd KK, Kidd JR, Calafell F, dial ischemia in internal medicine. Kranzler HR, Ichinose H, Gelernter J (1997) Population genetics of a functional variant of the dopamine beta-hydrox- ylase gene (DBH). Am J Med Genet 74:374–379 Acknowledgements This study was supported by the Yale Cubells JF, van Kammen DP, Kelley ME, Anderson GM, O’Connor University Physician Scientist Training Program (K12-DA-00089), DT, Price LH, Malison R, Rao PA, Kobayashi K, Nagatsu T, R01-DA 12422, K02-DA015766, the VA Research Enhancement Gelernter J (1998) Dopamine beta-hydroxylase: two poly- Award Program, and the VA New England Mental Illness Research morphisms in linkage disequilibrium at the structural gene Education and Clinical Center (J.F.C.), and K08-NS44138 (C.P.Z.). DBH associate with biochemical phenotypic variation. Hum Genet 102:533–540 Cubells JF, Kranzler HR, McCance-Katz E, Anderson GM, Malison RT, Price LH, Gelernter J (2000) A haplotype at the DBH References locus, associated with low plasma dopamine β-hydroxylase activity, also associates with cocaine-induced paranoia. Mol Amaral DG, Sinnamon HM (1977) The locus coeruleus: neurobi- Psychiatry 5:56–63 ology of a central noradrenergic nucleus. Prog Neurobiol Cubells JF, Price LH, Meyers BS, Anderson GM, Zabetian CP, 9:147–196 Alexopoulos, Nelson JC, Sanacora G, Kirwin P, Carpenter L, Anand A, Charney DS (2000) Norepinephrine dysfunction in Malison RT, Gelernter J (2002) Genotype-controlled analysis of depression. J Clin Psychiatry 61:16–24 plasma dopamine β-hydroxylase activity in psychotic unipolar Arango V, Underwood MD, Mann JJ (1994) Fewer pigmented major depression. Biol Psychiatry 51:358–364 neurons in the locus coeruleus of uncomplicated alcoholics. Daly G, Hawi Z, Fitzgerald M, Gill M (1999) Mapping suscepti- Brain Res 650:1–8 bility loci in attention deficit hyperactivity disorder: preferential Arnsten AF (2000a) Genetics of childhood disorders: XVIII. transmission of parental alleles at DAT1, DBH and DRD5 to ADHD. Part 2. Norepinephrine has a critical modulatory affected children. Mol Psychiatry 4:192–196 influence on prefrontal cortical function. J Am Acad Child DeVoto P, Flore G, Pani L, Gessa GL (2001) Evidence for co-release Adolesc Psychiatry 39:1201–1203 of noradrenaline and dopamine from noradrenergic neurons in Arnsten AF (2000b) Stress impairs prefrontal cortical function in the cerebral cortex. Mol Psychiatry 6:657–664 rats and monkeys: role of dopamine D1 and norepinephrine Druschky A, Hilz MJ, Platsch G, Radespiel-Troger M, Druschky K, alpha-1 receptor mechanisms. Prog Brain Res 126:183–192 Kuwert T, Neundorfer B (2000) Differentiation of Parkinson’s Arrufat FJ, Diaz R, Queralt R, Navarro V, Marcos T, Massana G, disease and multiple system atrophy in early disease stages by Massana J, Ballesta F, Oliva R (2000) Analysis of the means of I-123-MIBG-SPECT. J Neurol Sci 175:3–12 polymorphic (GT)(n) repeat at the dopamine beta-hydroxylase Dunnette J, Weinshilboum RM (1976) Human serum dopamine-β- gene in Spanish patients affected by schizophrenia. Am J Med hydroxylase: correlation of enzyme activity with immunoreac- Genet 96:88–92 tive protein in genetically defined samples. Am J Hum Genet Asamoah A, Wilson AF, Elston RC, Dalferes E Jr, Bereson GS 28:155–166 (1987) Segregation and linkage analyses of dopamine β- Ebstein RP, Park DH, Freedman LS, Levitz SM, Ouchi T, Goldstein hydroxylase activity in a large six-generation pedigree. Am J M (1973) A radioimmunoassay of human circulatory dopa- Med Genet 27:613–621 mine-beta-hydroxylase. Life Sci 13:769–784 Bezard E, Jaber M, Gonon F, Boireau A, Bloch B, Gross CE (2000) Elston RC, Namboodiri KK, Hames CG (1979) Segregation and Adaptive changes in the nigrostriatal pathway in response to linkage analyses of dopamine β-hydroxylase activity. Hum increased 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-in- Hered 29:284–292 duced neurodegeneration in the mouse. Eur J Neurosci Esler M, Rumantir M, Kaye D, Jennings G, Hastings J, Socratous F, 12:2892–2900 Lambert G (2001) Sympathetic nerve biology in essential Biagionni I, Robertson D (1987) Endogenous restoration of hypertension. Clin Exp Pharmacol Physiol 28:986–989 noradrenaline by precursor therapy in dopamine-beta-hydrox- Fischer HD, Rudolph E, Schmidt J, Hilse H, Oehme P, Piesche L ylase deficiency. Lancet 2:1170–1172 (1980) Interaction at the synaptic level of fusaric acid with Bowden CL, Deutsch CK, Swanson JM (1988) Plasma dopamine- neurotransmitters. Acta Biol Med Germ 39:935–940 beta-hydroxylase and platelet monoamine oxidase in attention Foote SL, Bloom FE, Aston-Jones G (1983) Nucleus locus ceruleus: deficit disorder and conduct disorder. J Am Acad Child new evidence of anatomical and physiological specificity. Adolesc Psychiatry 27:171–174 Physiol Rev 63:844–914 Braune S, Reinhardt M, Schnitzer R, Riedel A, Lucking CH (1999) Cardiac uptake of [123I]MIBG separates Parkinson’s disease from multiple system atrophy. Neurology 53:1020–1025 474 Forstl H, Levy R, Burns A, Luthert P, Cairns N (1994) Joyce JN, Smutzer G, Whitty CJ, Myers A, Bannon MJ (1997) Disproportionate loss of noradrenergic and cholinergic neurons Differential modification of dopamine transporter and tyrosine as cause of depression in Alzheimer’s disease—a hypothesis. hydroxylase mRNAs in midbrain of subjects with Parkinson’s, Pharmacopsychiatry 27:11–15 Alzheimer’s with parkinsonism, and Alzheimer’s disease. Mov Frecska BG, Horváth S, Zádor G, Arató M (1990) Predicting Disord 12:885–897 neuroleptic response from a combination of multilevel variables Kaufman S, Friedman S (1965) Dopamine-beta-hydroxylase. Phar- in acute schizophrenic patients. Acta Psychiatr Scand 82:408– macol Rev 17:71–100 412 Kent JM, Mathew SJ, Gorman JM (2002) Molecular targets in the Freeman R, Landsberg L, Young J (1999) The treatment of treatment of anxiety. Biol Psychiatry 52:1008–1030 neurogenic orthostatic hypotension with 3,4-DL-threo-dihydrox- Kim C-Y, Zabetian CP, Cubells JF, Cho S, Biaggioni I, Cohen BM, yphenylserine: a randomized, placebo-controlled, crossover Robertson D, Kim K-S (2002) Mutations in the dopamine β- trial. Neurology 53:2151–2157 hydroxylase gene are associated with human dopamine β- Fujita K, Ito T, Maruta K, Teradaira R, Beppu H, Nakagami Y, Kato hydroxylase deficiency. Am J Med Genet 108:140–147 Y, Nagatsu T, Kato T (1978) Serum dopamine-β-hydroxylase Kobayashi K, Kiuchi K, Ishii A, Kaneda N, Kurosawa Y, Fujita K, in schizophrenic patients. J Neurochem 30:1569–1572 Nagatsu T (1989) Human dopamine β-hydoxylase gene: two Gaspar P, Gray F (1984) Dementia in idiopathic Parkinson’s disease, mRNA types having different 3-terminal regions are produced a neuropathological study of 32 cases. Acta Neuropathol through alternative polyadenylation. Nucl Acids Res 17:1089– 64:43–52 1102 Gelernter J, Gejman PV, Bisghini S, Kidd KK (1991) Sequence Köhnke M, Zabetian CP, Anderson GM, Kolb W, Gaertner I, tagged site (STS) Taq1 RFLP at dopamine β-hydroxylase Buchkremer G, Vonthein R, Schick S, Lutz U, Köhnke AM, (DBH). Nucl Acids Res 19:1957 Cubells JF (2002) A genotype-controlled analysis of plasma Goldin LR, Gershon ES, Lake CR, Murphy DL, McGinniss M, dopamine β-hydroxylase in healthy subjects and alcoholics: Sparkes RS (1982) Segregation and linkage studies of plasma evidence for alcohol-related differences in noradrenergic func- dopamine-beta-hydroxylase (DBH), erythrocyte catechol-O- tion. Biol Psychiatry 52:1151–1158 methyl transferase (COMT) and platelet monoamine oxidase Lamouroux A, Vigny A, Faucon Biguet N, Darmon MC, Franck R, (MAO): possible linkage between the ABO locus and a gene Henry JP, Mallet J (1987) The primary structure of human controlling DBH activity. Am J Hum Genet 34:250–262 dopamine-beta-hydroxylase: insights into the relationship Grace AA, Gerfen CR, Aston-Jones G (1998) Catecholamines in the between the soluble and the membrane-bound forms of the central nervous system. Overview Adv Pharmacol 42:655–670 enzyme. EMBO J 6:3931–3937 Grobecker H, Roizen MF, Jacobowitz DM, Kopin IJ (1977) Effect Li B, Tsing S, Kosaka AH, Nguyen B, Osen EG, Bach C, Chan H, of prolonged treatment with adrenergic neuron blocking drugs Barnett J (1996) Expression of human dopamine beta-hydrox- on sympathoadrenal reactivity in rats. Eur J Pharmacol 46:125– ylase in Drosophila Schneider 2 cells. Biochem J 313:57–64 133 Lieberman AN, Freedman LS, Goldstein M (1972) Serum dopa- Grzanna R, Coyle JT (1978) Absence of a relationship between mine-beta-hydroxylase activity in patients with Huntington’s sympathetic neuronal activity and tulnover of serum dopami- chorea and Parkinson’s disease. Lancet 1:153–154 ne-β-hydroxylase. Naunyn-Schmiedeberg’s Arch Pharmakol Long JC, Williams RC, Urbanek M (1995) An E–M algorithm and Exp Pathol 304:231–236 testing strategy for multiple-locus haplotypes. Am J Hum Genet Hartmann E, Keller-Teschke M (1979) The psychological effects of 56:799–180 dopamine-β-hydroxylase inhibition in normal subjects. Biol Lykouras E, Markianos M, Malliaras D, Stefanis C (1988) Psychiatry 14:455–462 Neurochemical variables in delusional depression. Am J Heiss G, Tyroler HA, Gunnells JC, McGuffin WL, Hames CG Psychiatry 145:214–217 (1980) Dopamine-β-hydroxylase in a bi-racial community: Magalhaes M, Wenning GK, Daniel SE, Quinn NP (1995) demographic, cardiovascular and familial factors. J Chron Dis Autonomic dysfunction in pathologically confirmed multiple 33:301–310 system atrophy and idiopathic Parkinson’s disease—a retro- Holroyd KA, Penzien DB, Cordingley GE (1991) Propranolol in the spective comparison. Acta Neurol Scand 91:98–102 management of recurrent migraine: a meta-analytic review. Major LF, Lerner P, Ballenger JC, Brown GL, Goodwin FK, Headache 31:333–340 Lovenberg W (1979) Dopamine-beta-hydroxylase in the cere- Hoogendijk WJ, Feenstra MG, Botterblom MH, Gilhuis J, Sommer brospinal fluid: relationship to disulfiram-induced psychosis. IE, Kamphorst W, Eikelenboom P, Swaab DF (1999) Increased Biol Psychiatry 14:337–44 activity of surviving locus ceruleus neurons in Alzheimer’s Mann DM, Yates PO (1983) Pathological basis for neurotransmitter disease. Ann Neurol 45:82–91 changes in Parkinson’s disease. Neuropathol Appl Neurobiol Hoyle GW, Mercer EH, Palmiter RD, Brinster RL (1994) Cell- 9:3–19 specific expression from the human dopamine beta-hydroxylase Man in’t Veld AJ, Boomsma F, Moleman P, Schalekamp MADH promoter in transgenic mice is controlled via a combination of (1987) Congenital dopamine-β-hydroxylase deficiency. A positive and negative regulatory elements. J Neurosci 14:2455– novel orthostatic syndrome. Lancet 1:183–188 2463 Markianos M, Hadjikonstantinou M, Sfagos C (1981) Normal Hurst JH, LeWitt PA, Burns RS, Foster NL, Lovenberg W (1985) plasma dopamine-beta-hydroxylase in non-treated and treated CSF dopamine-beta-hydroxylase activity in Parkinson’s dis- Parkinson patients. Acta Neurol Scand 63:267–272 ease. Neurology 35:565–568 Mathias CJ, Bannister RB, Cortelli P, Heslop K, Polak JM, Ishiguro H, Kim KT, Joh TH, Kim KS (1993) Neuron-specific Raimbach S, Springall DR, Watson L (1990) Clinical, expression of the human dopamine beta-hydroxylase gene autonomic and therapeutic observations in two siblings with requires both the cAMP-response element and a silencer region. postural hypotension and sympathetic failure due to an inability J Biol Chem 268:17987–17994 to synthesize noradrenaline from dopamine because of a Ishii A, Kobayashi K, Kiuchi K, Nagatsu T (1991) Expression of deficiency of dopamine beta hydroxylase. Q J Med 75:617–633 two forms of human dopamine-beta-hydroxylase in COS cells. McMahon A, Sabban EL (1992) Regulation of expression of Neurosci Lett 125:25–28 dopamine beta-hydroxylase in PC12 cells by glucocorticoids Jansen M, Boomsma F, van den Heuvel B, Deinum J, Steenbergen- and cyclic AMP analogues. J Neurochem 59:2040–2047 Spanjers G, Wevers R (2001) Mutation analysis in two patients Meltzer HY, Hyong HW, Carroll BJ, Russo P (1976) Serum with congenital dopamine beta-hydroxylase deficiency, abstract dopamine-β-hydroxylase in the affective psychoses and schizo- III-ICS-10. Presented at the Ninth International Catecholamine phrenia: Decreased activity in unipolar psychotically depressed Symposium, Kyoto, Japan patients. Arch Gen Psychiatry 33:585–591 475 Meltzer HY, Nasr SJ, Tong C (1980) Serum dopamine-β-hydrox- Peronnet F, Cleroux J, Perrault H, Thibault G, Cousineau D, de ylase in schizophrenia. Biol Psychiatry 15:781–788 Champlain J, Guilland JC, Klepping J (1985) Plasma norepi- Meszaros K, Lenzinger E, Fureder T, Hornik K, Willinger U, nephrine, epinephrine, and dopamine beta-hydroxylase activity Stompe T, Heiden AM, Resinger E, Fathi N, Gerhard E, Fuchs during exercise in man. Med Sci Sports Exerc 17:683–688 K, Miller-Reiter E, Pfersmann V, Sieghart W, Aschauer HN, Perry SE, Summar L, Phillips JA, Robertson D (1991) Linkage Kasper S (1996) Schizophrenia and the dopamine-beta-hydrox- analysis of the human dopamine β-hydroxylase gene. Geno- ylase gene: results of a linkage and association study. Psychiatr mics 10:493–495 Genet 6:17–22 Pliszka SR, Rogeness GA, Medrano MA (1988) DBH, MHPG, and Meyers BS, Alexopoulos GS, Kakuma T, Tirumalasetti F, Gabriele MAO in children with depressive, anxiety, and conduct M, Alpert S, Bowden C, Meltzer HY (1999) Decreased disorders: relationship to diagnosis and symptom ratings. dopamine beta-hydroxylase activity in unipolar geriatric delu- Psychiatry Res 24:35–44 sional depression. Biol Psychiatry 45:448–452 Porter CJ, Nahmias J, Wolfe J, Craig IW (1992) Dinucleotide repeat Mitler MM, Hajdukovic R, Erman MK (1993) Treatment of polymorphism at the human dopamine β-hydroxylase (DBH) narcolepsy with methamphetamine. Sleep 16:306–317 locus. Nucleic Acids Res 20:1429 Mogi M, Harada M, Kojima K, Inagaki H, Kondo T, Narabayashi H, Province MA (2000) A single, sequential, genome-wide test to Arai T, Teradaira R, Fujita K, Kiuchi K, Nagatsu T (1988) identify simultaneously all promising areas in a linkage scan. Sandwich enzyme immunoassay of dopamine beta-hydroxylase Genet Epidemiol 19:301–322 in cerebrospinal fluid from control and parkinsonian patients. Robertson D, Goldberg MR, Hollister AS, Onrot J, Wiley R, Neurochem Int 12:187–191 Thompson JG, Robertson RM (1986) Isolated failure of Moore RY, Bloom FE (1979) Central catecholamine neuron autonomic noradrenergic neurotransmission: evidence for systems: anatomy and physiology of the norepinephrine and impaired beta-hydroxylation of dopamine. N Engl J Med epinephrine systems. Annu Rev Neurosci 2:113–168 314:1494–1497 Mód L, Rihmer Z, Magyar I, Arató M, Alföldi A, Bagdy G (1986) Rogeness GA, Hernandez JM, Macedo CA, Mitchell EL (1982) Serum DBH activity in psychotic vs nonpsychotic unipolar and Biochemical differences in children with conduct disorder bipolar depression. Psychiatry Res 19:331–333 socialized and undersocialized. Am J Psychiatry 139:307–311 Müller Smith K, Daly M, Fischer M, Yiannoutsos CT, Bauer L, Rogeness GA, Hernandez JM, Macedo CA, Mitchell EL, Amrung Barkley R, Navia BA (2003) Association of the dopamine beta SA, Harris WR (1984) Clinical characteristics of emotionally hydroxylase gene with attention deficit hyperactivity disorder: disturbed boys with very low activities of dopamine-beta- genetic analysis of the Milwaukee longitudinal study. Am J hydroxylase. J Am Acad Child Psychiatry 23:203–208 Med Genet Part B (Neuropsychiatr Genet) 119B:77–85 Rogeness GA, Hernandez JM, Macedo CA, Amrung SA, Hoppe SK Nagatsu T, Udenfriend S (1972) Photometric assay of dopamine-β- (1986) Near-zero plasma dopamine-beta-hydroxylase and hydroxylase activity in rat serum. Clin Chem 18:980–983 conduct disorder in emotionally disturbed boys. J Am Acad Nagatsu T, Kato T, Nagatsu I, Kondo Y, Inagaki S, Iizuka R, Child Psychiatry 25:521–527 Narabayashi H (1979) Catecholamine-related enzymes in the Rogeness GA, Crawford L, McNamara A (1989) Plasma dopamine- brain of patients with parkinsonism and Wilson’s disease. Adv beta-hydroxylase and preschool behavior in children with Neurol 24:283–292 conduct disorder. Child Psychiatry Hum Dev 20:149–56 Nagatsu T, Wakui Y, Kato T, Fujita K, Kondo T, Yokochi F, Roman T, Schmitz M, Polanczyk GV, Eizirik M, Rohde LA, Hutz Narabayashi H (1982) Dopamine-beta-hydroxylase activity in MH (2002) Further evidence for the association between cerebrospinal fluid of Parkinsonian patients. Biomed Res 3:95– attention-deficit/hyperactivity disorder and the dopamine-beta- 98 hydroxylase gene. Am J Med Genet (Neuropsychiatric Genet- Nelson JC, Davis JM (1997) DST studies in psychotic depression: a ics) 114:154–158 meta-analysis. Am J Psychiatry 154:1497–1503 Ross SB, Wetterberg L, Myrhead M (1973) Genetic control of Ng PC, Henikoff S (2001) Predicting deleterious amino acid plasma dopamine-β-hydroxylase. Life Sci 12:529–532 substitutions. Genome Res 11:863–874 Rothschild AJ, Langlais PJ, Schatzberg AF, Walsh FX, Cole JO, O’Connor DT, Levine GL, Frigone RP (1983) Homologous radio- Bird ED (1984) Dexamethasone increases plasma free dopa- immunoassay of human plasma dopamine β-hydroxylase: mine in man. J Psychiatr Res 18:217–23 analysis of homospecific activity, circulating plasma pool and Sack RL, Goodwin FK (1974) Inhibition of dopamine-β-hydro- intergroup differences based upon race, blood pressure and xyalse in manic patients: a clinical trial with fusaric acid. Arch cardiac function. J Hypertens 1:227–233 Gen Psychiatry 31:649–654 O’Connor DT, Cervenka JH, Stone RA, Levine GL, Parmer RJ, Sapru MK, Rao BSSR, Channabasavanna SM (1989) Serum Franco-Bourland RE, Madrazo I, Langlais PJ, Robertson D, dopamine-β-hydroxylase activity in clinical subtypes of de- Biaggioni I (1994) Dopamine β-hydroxylase immunoreactivity pression. Acta Psychiatr Scand 80:474–478 in human cerebrospinal fluid: properties, relationship to central Satoh A, Serita T, Seto M, Tomita I, Satoh H, Iwanaga K, noradrenergic activity and variation in Parkinson’s disease and Takashima H, Tsujihata M (1999) Loss of123I-MIBG uptake by congenital dopamine β-hydroxylase deficiency. Clin Sci the heart in Parkinson’s disease: assessment of cardiac sympa- 86:149–158 thetic denervation and diagnostic value. J Nucl Med 40:371– Ogihara T, Nugent CA, Shen SW, Goldfein S (1975) Serum 375 dopamine β-hydroxylase activity in parents and children. J Lab Schirger A, Sheps SG, Thomas JE, Fealey RD (1981) Midodrine. A Clin Med 85:566–570 new agent in the management of idiopathic orthostatic hypo- Oyarce AM, Fleming PJ (1991) Multiple forms of human dopamine tension and Shy-Drager syndrome. Mayo Clin Proc 56:429– β-hydroxylase in SH-SY5Y neuroblastoma cells. Arch Bio- 433 chem Biophys 290:503–510 Singer C, Weiner WJ, Sanchez-Ramos JR (1992) Autonomic Payton A, Holmes J, Barrett JH, Hever T, Fitzpatrick H, Trumper dysfunction in men with Parkinson’s disease. Eur Neurol AL, Harrington R, McGuffin P, O’Donovan M, Owen M, Ollier 32:134–140 W, Worthington J, Tapar A (2001) Examining for association Small KM, Wagoner LE, Levin AM, Kardia SLR, Liggett SB (2002) between candidate gene polymorphisms in the dopamine Synergistic polymorphisms of (beta) 1- and (alpha)(2C)- pathway and attention deficit hyperactivity disorder: a family- adrenergic receptors and the risk of congestive heart failure. based study. Am J Med Genet (Neuropsychiatr Genet) N Engl J Med 347:1135–1142 105:464–470 Smith AD, de Potter WD, Moerman EJ, de Schaepdryver AF (1970) Release of dopamine-β-hydroxylase and chromogranin A upon stimulation of the splenic nerve. Tissue Cell 2:547–568 476 Spielman RS, McGinnis RE, Ewens WJ (1993) Transmission test Weinshilboum RM, Thoa NB, Johnson DG, Kopin IJ, Axelrod J for linkage disequilibrium: the insulin gene region and insulin- (1971) Proportional release of norepinephrine and dopamine β- dependent diabetes mellitus (IDDM). Am J Hum Genet hydroxylase from sympathetic nerves. Science 174:1349–1351 52:506–516 Weinshilboum RM, Raymond FA, Elvenack LRl, Weidman WH Spitzer RL, Endicott J, Robbins E (1975) Research diagnostic (1973) Serum dopamine-β-hydroxylase: sibling-sibling corre- criteria. New York State Psychiatric Institute, Biometrics lation. Science 181:943–945 Research, New York Weinshilboum RM, Schrott HG, Raymond FA, Weidman WH, Stallings MC, Corley RP, Hewitt JK, Krauter KS, Lessem JM, Elvevack LR (1975) Inheritance of very low serum DBH Mikulich SK, Rhee SH, Smolen A, Young SE, Crowley TJ activity. Am J Hum Genet 27:573–585 (2003) A genome-wide search for quantitative trait loci Wenning GK, Tison F, Ben Shlomo Y, Daniel SE, Quinn NP (1997) influencing substance dependence vulnerability in adolescence. Multiple system atrophy: a review of 203 pathologically proven Drug Alcohol Depend 70:295–307 cases. Mov Disord 12:133–147 Stanley WC, Li B, Bonhaus DW, Johnson LG, Lee K, Porter S, Wigg K, Zai G, Schachar R, Tannock R, Roberts W, Malone M, Walker K, Martinez G, Eglen RM, Whiting RL, Hegde SS Kennedy JL, Barr CL (2002) Attention deficit hyperactivity (1997) Catecholamine modulatory effects of nepicastat (RS- disorder and the gene for dopamine beta-hydroxylase. Am J 25560-197), a novel, potent and selective inhibitor of dopamine Psychiatry 159:1046–1048 β-hydroxylase. Br J Pharmacol 121:1803–1809 Williams HJ, Bray N, Murphy KC, Cardno AG, Jones LA, Owen Sternberg DE, van Kammen DP, Lerner P, Bunney WE (1982) MJ (1999) No evidence for association between schizophrenia Schizophrenia: dopamine-β-hydroxylase activity and treatment and a functional variant of the human dopamine beta-hydrox- response. Science 216:1423–1425 ylase gene (DBH). Am J Med Genet 88:557–559 Sternberg DE, van Kammen DP, Lerner P, Ballenger JC, Marder SR, Wilson AF, Elston RC, Siervogel RM, Tran LD (1988) Linkage of a Post RM, Bunney WE (1983) CSF dopamine-β-hydroxylase in gene regulating dopamine-β-hydroxylase activity and the ABO schizophrenia: low activity associated with good prognosis and blood group locus. Am J Hum Genet 42:160–166 good response to neuroleptic treatment. Arch Gen Psychiatry Wilson AF, Elston RC, Sellers TA, Bailey-Wilson JE, Gerstin JM, 40:743–747 Deen DK, Sorant AJM, Tran LD, Amos CI, Siervogel RM Stewart LC, Klinman JP (1988) Dopamine beta-hydroxylase of (1990) Stepwise oligogenic segregation and linkage analysis adrenal chromaffin granules: structure and function. Annu Rev illustrated with dopamine-β-hydroxylase activity. Am J Med Biochem 57:551–592 Genet 35:425–432 Suarez GA, Opfer-Gehrking TL, Offord KP, Atkinson EJ, O’Brien Winer N, Carter C (1977) Effect of cold pressor stimulation on PC, Low PA (1999) The Autonomic Symptom Profile: a new plasma norepinephrine, dopamine-β-hydroxylase, and renin instrument to assess autonomic symptoms. Neurology 52:523– activity. Life Sci 20:887–894 528 Wise CD, Stein L (1973) Dopamine-β-hydroxylase deficits in the Terwilliger JD, Ott J (1992) A haplotype-based “haplotype relative brains of schizophrenic patients. Science 181:344–347 risk” approach to detecting allelic associations. Hum Hered Yamamoto K, Cubells JF, Gelernter J, Benkelfat C, Lalonde P, 42:337–346 Bloom D, Lal S, Labelle A, Turecki F, Rouleau GA, Joober R Tomlinson BE, Irving D, Blessed G (1981) Cell loss in the locus (2003) Dopamine beta-hydroxylase (DBH) gene and schizo- coeruleus in senile dementia of Alzheimer type. J Neurol Sci phrenia phenotypic variability: a genetic association study. Am 49:419–428 J Med Genet Part B (Neuropsychiatr Genet) 117B:33–38 Tong ZY, Kingsbury AE, Foster OJ (2000) Up-regulation of tyrosine Yoshita M (1998) Differentiation of idiopathic Parkinson’s disease hydroxylase mRNA in a sub-population of A10 dopamine from striatonigral degeneration and progressive supranuclear neurons in Parkinson’s disease. Brain Res Mol Brain Res palsy using iodine-123 meta-iodobenzylguanidine myocardial 79:45–54 scintigraphy. J Neurol Sci 155:60–67 van Kammen DP, Kelley ME, Gilbertson MW, Gurklis J, O’Connor Zabetian CP, Anderson GM, Buxbaum SG, Elston RC, Ichinose H, DT (1994) CSF dopamine β-hydroxylase in schizophrenia: Nagatsu T, Kim KS, Kim C-H, Malison RT, Gelernter J, associations with premorbid functioning and brain computer- Cubells JF (2001) A quantitative trait locus analysis of human ized tomography scan measures. Am J Psychiatry 151:372–378 dopamine β-hydroxylase activity: evidence for a major func- Wakabayashi K, Takahashi H (1997) Neuropathology of autonomic tional polymorphism at the DBH locus. Am J Hum Genet nervous system in Parkinson’s disease. Eur Neurol 38:2–7 68:515–522 Wei J, Ramchand CN, Hemmings GP (1997a) Possible control of Zabetian CP, Buxbaum SG, Elston RC, Köhnke M, Anderson GM, dopamine beta-hydroxylase via a codominant mechanism Gelernter J, Cubells JF (2003a) The Structure of linkage associated with the polymorphic (GT)n repeat at its gene disequilibrium at the DBH locus strongly influences the locus in healthy individuals. Hum Genet 99:52–55 magnitude of association between diallelic markers and plasma Wei J, Hai-Min X, Ramchand CN, Hemmings GP (1997b) Is the dopamine beta-hydroxylase activity. Am J Hum Genet polymorphic microsatellite repeat of the dopamine β-hydrox- 72:1389–1400 ylase gene associated with biochemical variability of the Zabetian CP, Romero R, Robertson D, Sharma S, Padbury JF, catecholamine pathway in schizophrenia? Biol Psychiatry Kuivaniemi H, Kim K-S, Kim C-H, Kohnke MD, Kranzler HR, 41:762–767 Gelernter J, Cubells JF (2003b) A revised allele frequency Weinshilboum RM (1978) Serum dopamine-β-hydroxylase. Phar- estimate and haplotype analysis of the DBH deficiency macol Rev 30:133–166 mutation IVS1+2T>C in African- and European-Americans. Weinshilboum RM, Axelrod J (1971) Serum dopamine β-hydrox- Am J Med Genet 123A:190–192 ylase. Circ Res 28:307–315