- Email: [email protected]
Association analysis of the functional Ala111Glu polymorphism of the glyoxalase I gene in panic disorder
Neuroscience Letters 396 (2006) 163–166
Association analysis of the functional Ala111Glu polymorphism of the glyoxalase I gene in panic disorder Pierluigi Politi a,∗ , Piercarlo Minoretti b , Colomba Falcone b , Valentina Martinelli a , Enzo Emanuele b a
Department of Applied Health and Behavioural Sciences, Section of Psychiatry, University of Pavia, Via Bassi 21, I-27100, Pavia, Italy b Interdepartmental Center for Research in Molecular Medicine (CIRMC), University of Pavia, Pavia, Italy Received 20 October 2005; received in revised form 11 November 2005; accepted 14 November 2005
Abstract The zinc metalloenzyme glyoxalase I (GLO1) is thought to play a role in anxiety disorders because a reduced brain expression of GLO1 has been associated with increased anxiety-behaviours in mice. Recently, a functional Ala111Glu polymorphism in GLO1 has been shown to result in a reduced enzyme activity. The present study tested the hypothesis that this common genetic variant could confer susceptibility to panic disorder using an Italian population sample of 162 panic disorder patients and 288 matched controls. Statistical analysis failed to show association with the overall diagnosis of the disease. However, a weak but significant association was demonstrated between this polymorphism and panic disorder without agoraphobia. While our data suggest that this polymorphism is unlikely to have a major function in the pathogenesis of panic disorder, it could play a role in the subgroup of patients without agoraphobic avoidance. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Glyoxalase I; Panic disorder; Agoraphobia; Single nucleotide polymorphism; Association study
Panic disorder (PD) is a common anxiety disorder characterized by unexpected and recurrent episodes of intense fear or anxiety that are often accompanied by physical symptoms such as dizziness, sweating, palpitations, and chest pain. The prevalence of this disabling psychiatric condition in the general population is of 1–3% [20], with about 50% of the cases displaying an association with agoraphobia [12]. Although the causes of PD are believed to be multifactorial, several lines of evidence point to a role for genetic factors in PD pathogenesis [18]. Accordingly, a recent meta-analysis of family and twin studies reported an estimated heritability of 48% for the overall diagnosis of PD [6]. In addition, sensitivity to anxiety has been proposed as an important risk factor for the development of PD [16]. Of interest, a very recent proteomic study in mice has identified the zinc metalloenzyme glyoxalase I (lactoylglutathione lyase, E.C. 4.4.1.5, GLO1) as a putative biomarker of trait anxiety [10]. Indeed, GLO1 has been found to be consistently expressed to a higher extent in several brain areas of low-anxiety-
∗
Corresponding author. Tel.: +39 0382 987878; fax: +39 0382 526723. E-mail address: [email protected] (P. Politi).
0304-3940/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2005.11.028
related behaviours animals as compared to both wild-type and high-anxiety-related behaviours littermates. Working from these observations, researchers suggested that it would be important to establish whether GLO1 may have an impact beyond that of a biomarker in anxiety-like behaviours [10]. Specifically, it remains to be established whether GLO1 could be involved in the genetic predisposition to anxiety disorders. Glyoxalase I is an antioxidant enzyme that, together with the cofactor glutathione, is involved in the detoxification of ␣ketoaldehydes, thereby preventing the formation of prooxidant compounds such as methylglyoxal [17]. The gene encoding for GLO1 has a common polymorphism (reported in dbSNP as rs2736654) that causes an Ala111Glu substitution in the protein sequence. The presence of the additional acidic charge from the 111Glu residue has been shown to result in a conformational change of the protein associated with a decreased enzyme activity [8]. To test the hypothesis that the GLO1 gene might play a role in anxiety-related behaviours, we examined the relationship between the functional GLO1 Ala111Glu polymorphism and PD in an Italian population sample. Since PD patients are often co-morbid with agoraphobia, we also wanted to investigate whether a specific association with agoraphobia could exist in the patient group.
P. Politi et al. / Neuroscience Letters 396 (2006) 163–166
164
The study sample comprised 162 psychiatric outpatients who met the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR) criteria for PD. Each patient was assessed by a board-certified psychiatrist using the Mini-International Neuropsychiatric Interview, a short, structured diagnostic interview with documented reliability [15]. According to previous methodology [3], only patients with predominant PD which occurred primarily in the course of the disorder were included. PD subjects comorbid with major or bipolar depression (n = 79) were not included in the present study. A total of 288 age- and gender-matched healthy volunteers were selected as a control group. All of them were interviewed using the Mini-International Neuropsychiatric Interview and none had a personal or family history of psychiatric disorders among first-degree relatives. All subjects described here were Caucasian whites of Italian descent. In our study, the criterion “of Italian descent” was met when an individual’s parents and four grandparents originated from Italy. The study protocol followed the guidelines of our local ethics committee and the investigation was conducted with the ethical requirements defined in the Helsinki Declaration. Blood samples were drawn by venipuncture after the completion of informed consent forms. Genomic DNA was purified from peripheral blood by using the QiaAmp DNA Mini kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. The procedure for detecting the Ala111Glu polymorphism was based on restriction isotyping as described by Junaid et al. [8], with minor modifications. Briefly, DNA was amplified in a GeneAmp PCR System 9700 thermalcycler (Applera, Monza, Italy) with the oligonucleotide primers: sense 5 -TCAGAGTGTGTGATTTCGTG-3 , and antisense 5 -CATGGTGAGATGGTAAGTGT-3 . Standard PCR was carried out in a 50 l reaction volume containing Taq PCR Master Mix (Eppendorf, Milan, Italy) at the following conditions: 95 ◦ C for 90 s, 35 cycles of a 94 ◦ C denaturing step for 30 s, a 56 ◦ C annealing step for 45 s, and a 72 ◦ C extension for 60 s, followed by a final extension at 72 ◦ C for 5 min, resulting in a 713 bp fragment. An aliquot (8 l) of the PCR product was digested using 2.0 U of the endonuclease SfaNI (New England Biolabs, Beverly, MA, USA) for 3 h at 37 ◦ C. The digested products were subjected to a 1% agarose gel electrophoresis and were visualized under ultraviolet light after ethidium bromide staining. The GLO1 111Ala allele yielded two distinctive fragments of 453 and 260 bp, while an unrestricted fragment of 713 bp identified the GLO1 111Glu allele. All genotyping was done with laboratory personnel blinded to case–control status of the
samples, which included quality control samples for validation. Concordance for quality control samples was 100%. The allele frequencies were estimated by the allele counting method. The categorical data were analyzed by χ2 tests. For continuous variables, the differences were evaluated using the Student’s t-test. Multiple logistic regression analysis was performed to test whether the Ala111Glu polymorphism was associated with the risk of PD after adjustment for age and gender. Results were expressed as odds ratios (ORs) with their 95% confidence intervals (CIs). Data were analyzed with the MedCalc statistical package (MedCalc, Mariakerke, Belgium). Statistical testing was two-tailed, with α set at 0.05. Because two comparisons were performed for PD patients, i.e., with and without agoraphobia, a Bonferroni-corrected threshold of α < 0.025 was considered. The study power was calculated using the StatMate software, version 2.0 (GraphPad, San Diego, CA, USA). Comparing the PD patients and healthy control subjects, the mean age (39.9 ± 10.2 years and 39.1 ± 11.6 years, respectively) and gender distribution (male/female 64/98 and 119/169, respectively) were similar (P = 0.39 and P = 0.78, respectively). A total of 101 patients (62.3%) were diagnosed as PD with agoraphobia and 61 (37.7%) as PD without agoraphobia. The prevalence of agoraphobia in our sample of PD patients was similar to that observed in previous investigations [12,13]. The frequency of GLO1 Ala111Glu polymorphism by diagnostic group is presented in Table 1. No deviation from Hardy–Weinberg equilibrium was seen in either the entire cohort of PD subjects or in controls. Similarly, PD patients with and without agoraphobia were in Hardy–Weinberg equilibrium. Based on the observed prevalence of the minor 111Glu allele, our sample size had a 95% power to detect a relative risk of 1.48 for PD between carriers and noncarriers with a significance level (alpha) of 0.05 (two-tailed). Allele and genotype frequencies did not differ in the entire group of PD patients and controls. In the subgroup of patients with PD without agoraphobia (n = 61), however, an association of the investigated polymorphism was observed (genotypes: χ2 = 7.72; d.f. = 2; P = 0.021; alleles: χ2 = 5.97; d.f. = 1; P = 0.015; thus, P < 0.025 [Bonferroni adjusted]), with more 111Glu alleles occurring in this PD subgroup. After allowance for age and gender, the multivariate-adjusted odds ratio for PD without agoraphobia associated with the homozygous Glu/Glu genotype was 1.64 (95% CI: 1.10–2.46, P = 0.014). In the entire PD patient group, no significant differences were demonstrated comparing the three GLO1 genotypic subgroups for age at onset
Table 1 Genotype and allele distribution of the Ala111Glu polymorphism of the GLO1 gene in the entire cohort of PD patients, in the subgroups with and without agoraphobia, and healthy comparison subjects Subject group
n
Genotype (%) Ala/Ala
Panic disorder With agoraphobia Without agoraphobia Control subjects a
162 101 61 288
51 (31.5) 38 (37.6) 13 (21.3) 89 (30.9)
Allele frequency (%) Ala/Glu 75 (46.3) 48 (47.5) 27 (44.3) 145 (50.3)
Compared with the control group. P values are calculated by means of χ2 tests.
Glu/Glu
P
Ala
Glu
P
36 (22.2) 15 (19.1) 21 (34.4) 54 (18.8)
0.611a
177 (56.4) 124 (61.4) 53 (52.0) 323 (56.1)
147 (45.4) 78 (38.6) 69 (48.0) 253 (44.9)
0.727a 0.218a 0.015a
0.407a 0.021a
P. Politi et al. / Neuroscience Letters 396 (2006) 163–166
(36.5 ± 10.9 years, 36.2 ± 10.8 years and 35.9 ± 11.1 years for Ala/Ala, Ala/Glu and Glu/Glu, respectively; P = 0.62). A growing body of evidence has recently implicated a role for chronic or moderate oxidative stress in the pathogenesis of anxious states. Indeed, previous clinical investigations have reported an imbalance of antioxidant enzyme activities in both patients with social phobia [1] and obsessive-compulsive disorder [11]. Additionally, experiments in genetically engineered mice have suggested that a reduced protection against oxidative brain injury is associated with a significant increase in anxiety as shown by fewer entries and less time spent in the open arm of an elevated plus-maze [4]. The recent identification of the antioxidant enzyme glyoxalase I as a biomarker of anxiety-like behaviours in mice [10] prompted us to investigate the putative role of GLO1 in the genetic predisposition to PD. To address this issue, we performed association analysis using a non-synonymous polymorphism in the coding sequence of GLO1 (Ala111Glu), which was not investigated previously in relation to PD. We selected this well-defined single nucleotide polymorphism as it has been previously shown to have functional consequences by decreasing the enzyme activity of GLO1 [8]. Of note, the same polymorphism has been previously associated with several diseases such as prostate cancer [14] and diabetes [9], as well as Alzheimer’s disease [2] and autism [8]. The latter observations are particularly intriguing inasmuch as they seem to suggest an involvement of GLO1 in the pathogenesis of some neurological and psychiatric conditions. The results of our study, however, did not point toward a major role of GLO1 Ala111Glu polymorphism in the pathogenesis of PD. Indeed, no association of this common genetic with the overall diagnosis of PD was discerned. Nevertheless, this polymorphism was found to be associated with the diagnosis of “pure” PD without agoraphobia. Although the mechanisms behind this specific association cannot be directly inferred from the present study, it is feasible that the 111Glu residue can exert this effect through the modulation of oxidative stress metabolism [8], which has been recently related to the level of anxiety-like behaviours in mice [7]. In humans, an imbalance of oxidative stress metabolism may be associated with a higher arousal state [1] that could in turn lead to a higher susceptibility to panic attacks in stressful life circumstances. The fact that the GLO1 Ala111Glu polymorphism is associated with the diagnosis of “pure” PD without agoraphobia only, could be explained by this biological model and is consistent with a differential molecular pathogenesis of panic attacks and agoraphobia [12,19]. In the interpretation of our findings, several important caveats should be considered. Firstly, the number of subjects in the subgroup of PD without agoraphobia was relatively small. We thus believe that our positive findings should be considered explorative, and need to be confirmed in larger sample numbers. Second, as our study cohort consisted only of Italian subjects without ethnical diversity, these results require replication in different ancestry populations. As a rationale for further research, we suggest that future studies should focus on different syndromic and non-syndromic PD subtypes, as well physiologic
165
endophenotypes of panic and phobic anxiety, inasmuch as each of which might differ with regard of susceptibility genes [19]. Third, it should be noted that our study shares the limitations of case–control investigations with regard of population stratification risk. Finally, we are aware that the use of only one single nucleotide polymorphism in this report negates the exploration of significant interactions among distinct polymorphisms in a given haplotype. As several polymorphisms have been identified bioinformatically in the GLO1 gene [5], a larger number of genetic markers should be examined to shed more light on the putative association of this gene with PD subtypes, as well as with other anxiety disorders. In summary, our data do not provide evidence for a major role of the GLO1 Ala111Glu polymorphism in the pathogenesis of PD. In the subset of patients without concurrent agoraphobia, however, a minor role of this genetic variant to the pathogenesis of the disease could be possible and deserves further investigation. References [1] M. Atmaca, E. Tezcan, M. Kuloglu, B. Ustundag, H. Tunckol, Antioxidant enzyme and malondialdehyde values in social phobia before and after citalopram treatment, Eur. Arch. Psychiatry Clin. Neurosci. 254 (2004) 231–235. [2] F. Chen, M.A. Wollmer, F. Hoerndli, G. Munch, B. Kuhla, E.I. Rogaev, M. Tsolaki, A. Papassotiropoulos, J. Gotz, Role for glyoxalase I in Alzheimer’s disease, Proc. Natl. Acad. Sci. U.S.A. 101 (2004) 7687–7692. [3] J. Deckert, M. Catalano, Y.V. Syagailo, M. Bosi, O. Okladnova, D. Di Bella, M.M. Nothen, P. Maffei, P. Franke, J. Fritze, W. Maier, P. Propping, H. Beckmann, L. Bellodi, K.P. Lesch, Excess of high activity monoamine oxidase A gene promoter alleles in female patients with panic disorder, Hum. Mol. Genet. 8 (1999) 621–624. [4] C. Desrumaux, P.Y. Risold, H. Schroeder, V. Deckert, D. Masson, A. Athias, H. Laplanche, N. Le Guern, D. Blache, X.C. Jiang, A.R. Tall, D. Desor, L. Lagrost, Phospholipid transfer protein (PLTP) deficiency reduces brain vitamin E content and increases anxiety in mice, FASEB J. 19 (2005) 296–297. [5] C.P. Gale, P.J. Grant, The characterisation and functional analysis of the human glyoxalase-1 gene using methods of bioinformatics, Gene 340 (2004) 251–260. [6] J.M. Hattema, M.C. Neale, K.S. Kendler, A review and meta-analysis of genetic epidemiology of anxiety disorders, Am. J. Psychiatry 158 (2001) 1568–1578. [7] I. Hovatta, R.S. Tennant, R. Helton, R.A. Marr, O. Singer, J.M. Redwine, J.A. Ellison, E.E. Schadt, I.M. Verma, D.J. Lockhart, C. Barlow, Glyoxalase 1 and glutathione reductase 1 regulate anxiety in mice, Nature (2005) (Epub ahead of print). [8] M.A. Junaid, D. Kowal, M. Barua, P.S. Pullarkat, S. Sklower Brooks, R.K. Pullarkat, Proteomic studies identified a single nucleotide polymorphism in glyoxalase I as autism susceptibility factor, Am. J. Med. Genet. A 131 (2004) 11–17. [9] R.L. Kirk, J. Theophilus, S. Whitehouse, J. Court, P. Zimmet, Genetic susceptibility to diabetes mellitus: the distribution of properdin factor B (Bf) and glyoxalase (GLO) phenotypes, Diabetes 28 (1979) 949–951. [10] S.A. Kromer, M.S. Kessler, D. Milfay, I.N. Birg, M. Bunck, L. Czibere, M. Panhuysen, B. Putz, J.M. Deussing, F. Holsboer, R. Landgraf, C.W. Turck, Identification of glyoxalase-I as a protein marker in a mouse model of extremes in trait anxiety, J. Neurosci. 25 (2005) 4375–4384. [11] M. Kuloglu, M. Atmaca, E. Tezcan, O. Gecici, H. Tunckol, B. Ustundag, Antioxidant enzyme activities and malondialdehyde levels in patients with obsessive-compulsive disorder, Neuropsychobiology 46 (2002) 27–32.
166
P. Politi et al. / Neuroscience Letters 396 (2006) 163–166
[12] Y.J. Lee, C. Hohoff, K. Domschke, P. Sand, G. Kuhlenbaumer, A. Schirmacher, C.M. Freitag, J. Meyer, G. Stober, P. Franke, M.M. Nothen, J. Fritze, R. Fimmers, H.S. Garritsen, F. Stogbauer, J. Deckert, Norepinephrine transporter (NET) promoter and 5 -UTR polymorphisms: association analysis in panic disorder, Neurosci. Lett. 377 (2005) 40– 43. [13] C. Rothe, D. Koszycki, J. Bradwejn, N. King, V. De Luca, S. Shaikh, P. Franke, H. Garritsen, J. Fritze, J. Deckert, J.L. Kennedy, Association study of serotonin-2A receptor gene polymorphism and panic disorder in patients from Canada and Germany, Neurosci. Lett. 363 (2004) 276– 279. [14] A.A. Samadi, S.A. Fullerton, D.G. Tortorelis, G.B. Johnson, S.D. Davidson, M.S. Choudhury, C. Mallouh, H. Tazaki, S. Konno, Glyoxalase, I phenotype as a potential risk factor for prostate carcinoma, Urology 57 (2001) 183–187. [15] D.V. Sheehan, Y. Lecrubier, K.H. Sheehan, P. Amorim, J. Janavs, E. Weiller, T. Hergueta, R. Baker, G.C. Dunbar, The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation
[16]
[17]
[18]
[19] [20]
of a structured diagnostic interview for DSM-IV and ICD-10, J. Clin. Psychiatry 59 (1998) 22–33. N.M. Simon, M.W. Otto, J.A. Smits, D.C. Nicolaou, H.E. Reese, M.H. Pollack, Changes in anxiety sensitivity with pharmacotherapy for panic disorder, J. Psychiatr. Res. 38 (2004) 491–495. P.J. Thornalley, Glyoxalase I-structure, function and a critical role in the enzymatic defence against glycation, Biochem. Soc. Trans. 31 (2003) 1343–1348. O.A. van den Heuvel, B.J. van de Wetering, D.J. Veltman, D.L. Pauls, Genetic studies of panic disorder: a review, J. Clin. Psychiatry 61 (2000) 756–766. S. Villafuerte, M. Burmeister, Untangling genetic networks of panic, phobia, fear and anxiety, Genome Biol. 4 (2003) 224. M.M. Weissman, R.C. Bland, G.J. Canino, C. Faravelli, S. Greenwald, H.G. Hwu, P.R. Joyce, E.G. Karam, C.K. Lee, J. Lellouch, J.P. Lepine, S.C. Newman, M.A. Oakley-Browne, M. Rubio-Stipec, J.E. Wells, P.J. Wickramaratne, H.U. Wittchen, E.K. Yeh, The cross-national epidemiology of panic disorder, Arch. Gen. Psychiatry 54 (1997) 305–309.
Association analysis of the functional Ala111Glu polymorphism of the glyoxalase I gene in panic disorder Pierluigi Politi a,∗ , Piercarlo Minoretti b , Colomba Falcone b , Valentina Martinelli a , Enzo Emanuele b a
Department of Applied Health and Behavioural Sciences, Section of Psychiatry, University of Pavia, Via Bassi 21, I-27100, Pavia, Italy b Interdepartmental Center for Research in Molecular Medicine (CIRMC), University of Pavia, Pavia, Italy Received 20 October 2005; received in revised form 11 November 2005; accepted 14 November 2005
Abstract The zinc metalloenzyme glyoxalase I (GLO1) is thought to play a role in anxiety disorders because a reduced brain expression of GLO1 has been associated with increased anxiety-behaviours in mice. Recently, a functional Ala111Glu polymorphism in GLO1 has been shown to result in a reduced enzyme activity. The present study tested the hypothesis that this common genetic variant could confer susceptibility to panic disorder using an Italian population sample of 162 panic disorder patients and 288 matched controls. Statistical analysis failed to show association with the overall diagnosis of the disease. However, a weak but significant association was demonstrated between this polymorphism and panic disorder without agoraphobia. While our data suggest that this polymorphism is unlikely to have a major function in the pathogenesis of panic disorder, it could play a role in the subgroup of patients without agoraphobic avoidance. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Glyoxalase I; Panic disorder; Agoraphobia; Single nucleotide polymorphism; Association study
Panic disorder (PD) is a common anxiety disorder characterized by unexpected and recurrent episodes of intense fear or anxiety that are often accompanied by physical symptoms such as dizziness, sweating, palpitations, and chest pain. The prevalence of this disabling psychiatric condition in the general population is of 1–3% [20], with about 50% of the cases displaying an association with agoraphobia [12]. Although the causes of PD are believed to be multifactorial, several lines of evidence point to a role for genetic factors in PD pathogenesis [18]. Accordingly, a recent meta-analysis of family and twin studies reported an estimated heritability of 48% for the overall diagnosis of PD [6]. In addition, sensitivity to anxiety has been proposed as an important risk factor for the development of PD [16]. Of interest, a very recent proteomic study in mice has identified the zinc metalloenzyme glyoxalase I (lactoylglutathione lyase, E.C. 4.4.1.5, GLO1) as a putative biomarker of trait anxiety [10]. Indeed, GLO1 has been found to be consistently expressed to a higher extent in several brain areas of low-anxiety-
∗
Corresponding author. Tel.: +39 0382 987878; fax: +39 0382 526723. E-mail address: [email protected] (P. Politi).
0304-3940/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2005.11.028
related behaviours animals as compared to both wild-type and high-anxiety-related behaviours littermates. Working from these observations, researchers suggested that it would be important to establish whether GLO1 may have an impact beyond that of a biomarker in anxiety-like behaviours [10]. Specifically, it remains to be established whether GLO1 could be involved in the genetic predisposition to anxiety disorders. Glyoxalase I is an antioxidant enzyme that, together with the cofactor glutathione, is involved in the detoxification of ␣ketoaldehydes, thereby preventing the formation of prooxidant compounds such as methylglyoxal [17]. The gene encoding for GLO1 has a common polymorphism (reported in dbSNP as rs2736654) that causes an Ala111Glu substitution in the protein sequence. The presence of the additional acidic charge from the 111Glu residue has been shown to result in a conformational change of the protein associated with a decreased enzyme activity [8]. To test the hypothesis that the GLO1 gene might play a role in anxiety-related behaviours, we examined the relationship between the functional GLO1 Ala111Glu polymorphism and PD in an Italian population sample. Since PD patients are often co-morbid with agoraphobia, we also wanted to investigate whether a specific association with agoraphobia could exist in the patient group.
P. Politi et al. / Neuroscience Letters 396 (2006) 163–166
164
The study sample comprised 162 psychiatric outpatients who met the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR) criteria for PD. Each patient was assessed by a board-certified psychiatrist using the Mini-International Neuropsychiatric Interview, a short, structured diagnostic interview with documented reliability [15]. According to previous methodology [3], only patients with predominant PD which occurred primarily in the course of the disorder were included. PD subjects comorbid with major or bipolar depression (n = 79) were not included in the present study. A total of 288 age- and gender-matched healthy volunteers were selected as a control group. All of them were interviewed using the Mini-International Neuropsychiatric Interview and none had a personal or family history of psychiatric disorders among first-degree relatives. All subjects described here were Caucasian whites of Italian descent. In our study, the criterion “of Italian descent” was met when an individual’s parents and four grandparents originated from Italy. The study protocol followed the guidelines of our local ethics committee and the investigation was conducted with the ethical requirements defined in the Helsinki Declaration. Blood samples were drawn by venipuncture after the completion of informed consent forms. Genomic DNA was purified from peripheral blood by using the QiaAmp DNA Mini kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. The procedure for detecting the Ala111Glu polymorphism was based on restriction isotyping as described by Junaid et al. [8], with minor modifications. Briefly, DNA was amplified in a GeneAmp PCR System 9700 thermalcycler (Applera, Monza, Italy) with the oligonucleotide primers: sense 5 -TCAGAGTGTGTGATTTCGTG-3 , and antisense 5 -CATGGTGAGATGGTAAGTGT-3 . Standard PCR was carried out in a 50 l reaction volume containing Taq PCR Master Mix (Eppendorf, Milan, Italy) at the following conditions: 95 ◦ C for 90 s, 35 cycles of a 94 ◦ C denaturing step for 30 s, a 56 ◦ C annealing step for 45 s, and a 72 ◦ C extension for 60 s, followed by a final extension at 72 ◦ C for 5 min, resulting in a 713 bp fragment. An aliquot (8 l) of the PCR product was digested using 2.0 U of the endonuclease SfaNI (New England Biolabs, Beverly, MA, USA) for 3 h at 37 ◦ C. The digested products were subjected to a 1% agarose gel electrophoresis and were visualized under ultraviolet light after ethidium bromide staining. The GLO1 111Ala allele yielded two distinctive fragments of 453 and 260 bp, while an unrestricted fragment of 713 bp identified the GLO1 111Glu allele. All genotyping was done with laboratory personnel blinded to case–control status of the
samples, which included quality control samples for validation. Concordance for quality control samples was 100%. The allele frequencies were estimated by the allele counting method. The categorical data were analyzed by χ2 tests. For continuous variables, the differences were evaluated using the Student’s t-test. Multiple logistic regression analysis was performed to test whether the Ala111Glu polymorphism was associated with the risk of PD after adjustment for age and gender. Results were expressed as odds ratios (ORs) with their 95% confidence intervals (CIs). Data were analyzed with the MedCalc statistical package (MedCalc, Mariakerke, Belgium). Statistical testing was two-tailed, with α set at 0.05. Because two comparisons were performed for PD patients, i.e., with and without agoraphobia, a Bonferroni-corrected threshold of α < 0.025 was considered. The study power was calculated using the StatMate software, version 2.0 (GraphPad, San Diego, CA, USA). Comparing the PD patients and healthy control subjects, the mean age (39.9 ± 10.2 years and 39.1 ± 11.6 years, respectively) and gender distribution (male/female 64/98 and 119/169, respectively) were similar (P = 0.39 and P = 0.78, respectively). A total of 101 patients (62.3%) were diagnosed as PD with agoraphobia and 61 (37.7%) as PD without agoraphobia. The prevalence of agoraphobia in our sample of PD patients was similar to that observed in previous investigations [12,13]. The frequency of GLO1 Ala111Glu polymorphism by diagnostic group is presented in Table 1. No deviation from Hardy–Weinberg equilibrium was seen in either the entire cohort of PD subjects or in controls. Similarly, PD patients with and without agoraphobia were in Hardy–Weinberg equilibrium. Based on the observed prevalence of the minor 111Glu allele, our sample size had a 95% power to detect a relative risk of 1.48 for PD between carriers and noncarriers with a significance level (alpha) of 0.05 (two-tailed). Allele and genotype frequencies did not differ in the entire group of PD patients and controls. In the subgroup of patients with PD without agoraphobia (n = 61), however, an association of the investigated polymorphism was observed (genotypes: χ2 = 7.72; d.f. = 2; P = 0.021; alleles: χ2 = 5.97; d.f. = 1; P = 0.015; thus, P < 0.025 [Bonferroni adjusted]), with more 111Glu alleles occurring in this PD subgroup. After allowance for age and gender, the multivariate-adjusted odds ratio for PD without agoraphobia associated with the homozygous Glu/Glu genotype was 1.64 (95% CI: 1.10–2.46, P = 0.014). In the entire PD patient group, no significant differences were demonstrated comparing the three GLO1 genotypic subgroups for age at onset
Table 1 Genotype and allele distribution of the Ala111Glu polymorphism of the GLO1 gene in the entire cohort of PD patients, in the subgroups with and without agoraphobia, and healthy comparison subjects Subject group
n
Genotype (%) Ala/Ala
Panic disorder With agoraphobia Without agoraphobia Control subjects a
162 101 61 288
51 (31.5) 38 (37.6) 13 (21.3) 89 (30.9)
Allele frequency (%) Ala/Glu 75 (46.3) 48 (47.5) 27 (44.3) 145 (50.3)
Compared with the control group. P values are calculated by means of χ2 tests.
Glu/Glu
P
Ala
Glu
P
36 (22.2) 15 (19.1) 21 (34.4) 54 (18.8)
0.611a
177 (56.4) 124 (61.4) 53 (52.0) 323 (56.1)
147 (45.4) 78 (38.6) 69 (48.0) 253 (44.9)
0.727a 0.218a 0.015a
0.407a 0.021a
P. Politi et al. / Neuroscience Letters 396 (2006) 163–166
(36.5 ± 10.9 years, 36.2 ± 10.8 years and 35.9 ± 11.1 years for Ala/Ala, Ala/Glu and Glu/Glu, respectively; P = 0.62). A growing body of evidence has recently implicated a role for chronic or moderate oxidative stress in the pathogenesis of anxious states. Indeed, previous clinical investigations have reported an imbalance of antioxidant enzyme activities in both patients with social phobia [1] and obsessive-compulsive disorder [11]. Additionally, experiments in genetically engineered mice have suggested that a reduced protection against oxidative brain injury is associated with a significant increase in anxiety as shown by fewer entries and less time spent in the open arm of an elevated plus-maze [4]. The recent identification of the antioxidant enzyme glyoxalase I as a biomarker of anxiety-like behaviours in mice [10] prompted us to investigate the putative role of GLO1 in the genetic predisposition to PD. To address this issue, we performed association analysis using a non-synonymous polymorphism in the coding sequence of GLO1 (Ala111Glu), which was not investigated previously in relation to PD. We selected this well-defined single nucleotide polymorphism as it has been previously shown to have functional consequences by decreasing the enzyme activity of GLO1 [8]. Of note, the same polymorphism has been previously associated with several diseases such as prostate cancer [14] and diabetes [9], as well as Alzheimer’s disease [2] and autism [8]. The latter observations are particularly intriguing inasmuch as they seem to suggest an involvement of GLO1 in the pathogenesis of some neurological and psychiatric conditions. The results of our study, however, did not point toward a major role of GLO1 Ala111Glu polymorphism in the pathogenesis of PD. Indeed, no association of this common genetic with the overall diagnosis of PD was discerned. Nevertheless, this polymorphism was found to be associated with the diagnosis of “pure” PD without agoraphobia. Although the mechanisms behind this specific association cannot be directly inferred from the present study, it is feasible that the 111Glu residue can exert this effect through the modulation of oxidative stress metabolism [8], which has been recently related to the level of anxiety-like behaviours in mice [7]. In humans, an imbalance of oxidative stress metabolism may be associated with a higher arousal state [1] that could in turn lead to a higher susceptibility to panic attacks in stressful life circumstances. The fact that the GLO1 Ala111Glu polymorphism is associated with the diagnosis of “pure” PD without agoraphobia only, could be explained by this biological model and is consistent with a differential molecular pathogenesis of panic attacks and agoraphobia [12,19]. In the interpretation of our findings, several important caveats should be considered. Firstly, the number of subjects in the subgroup of PD without agoraphobia was relatively small. We thus believe that our positive findings should be considered explorative, and need to be confirmed in larger sample numbers. Second, as our study cohort consisted only of Italian subjects without ethnical diversity, these results require replication in different ancestry populations. As a rationale for further research, we suggest that future studies should focus on different syndromic and non-syndromic PD subtypes, as well physiologic
165
endophenotypes of panic and phobic anxiety, inasmuch as each of which might differ with regard of susceptibility genes [19]. Third, it should be noted that our study shares the limitations of case–control investigations with regard of population stratification risk. Finally, we are aware that the use of only one single nucleotide polymorphism in this report negates the exploration of significant interactions among distinct polymorphisms in a given haplotype. As several polymorphisms have been identified bioinformatically in the GLO1 gene [5], a larger number of genetic markers should be examined to shed more light on the putative association of this gene with PD subtypes, as well as with other anxiety disorders. In summary, our data do not provide evidence for a major role of the GLO1 Ala111Glu polymorphism in the pathogenesis of PD. In the subset of patients without concurrent agoraphobia, however, a minor role of this genetic variant to the pathogenesis of the disease could be possible and deserves further investigation. References [1] M. Atmaca, E. Tezcan, M. Kuloglu, B. Ustundag, H. Tunckol, Antioxidant enzyme and malondialdehyde values in social phobia before and after citalopram treatment, Eur. Arch. Psychiatry Clin. Neurosci. 254 (2004) 231–235. [2] F. Chen, M.A. Wollmer, F. Hoerndli, G. Munch, B. Kuhla, E.I. Rogaev, M. Tsolaki, A. Papassotiropoulos, J. Gotz, Role for glyoxalase I in Alzheimer’s disease, Proc. Natl. Acad. Sci. U.S.A. 101 (2004) 7687–7692. [3] J. Deckert, M. Catalano, Y.V. Syagailo, M. Bosi, O. Okladnova, D. Di Bella, M.M. Nothen, P. Maffei, P. Franke, J. Fritze, W. Maier, P. Propping, H. Beckmann, L. Bellodi, K.P. Lesch, Excess of high activity monoamine oxidase A gene promoter alleles in female patients with panic disorder, Hum. Mol. Genet. 8 (1999) 621–624. [4] C. Desrumaux, P.Y. Risold, H. Schroeder, V. Deckert, D. Masson, A. Athias, H. Laplanche, N. Le Guern, D. Blache, X.C. Jiang, A.R. Tall, D. Desor, L. Lagrost, Phospholipid transfer protein (PLTP) deficiency reduces brain vitamin E content and increases anxiety in mice, FASEB J. 19 (2005) 296–297. [5] C.P. Gale, P.J. Grant, The characterisation and functional analysis of the human glyoxalase-1 gene using methods of bioinformatics, Gene 340 (2004) 251–260. [6] J.M. Hattema, M.C. Neale, K.S. Kendler, A review and meta-analysis of genetic epidemiology of anxiety disorders, Am. J. Psychiatry 158 (2001) 1568–1578. [7] I. Hovatta, R.S. Tennant, R. Helton, R.A. Marr, O. Singer, J.M. Redwine, J.A. Ellison, E.E. Schadt, I.M. Verma, D.J. Lockhart, C. Barlow, Glyoxalase 1 and glutathione reductase 1 regulate anxiety in mice, Nature (2005) (Epub ahead of print). [8] M.A. Junaid, D. Kowal, M. Barua, P.S. Pullarkat, S. Sklower Brooks, R.K. Pullarkat, Proteomic studies identified a single nucleotide polymorphism in glyoxalase I as autism susceptibility factor, Am. J. Med. Genet. A 131 (2004) 11–17. [9] R.L. Kirk, J. Theophilus, S. Whitehouse, J. Court, P. Zimmet, Genetic susceptibility to diabetes mellitus: the distribution of properdin factor B (Bf) and glyoxalase (GLO) phenotypes, Diabetes 28 (1979) 949–951. [10] S.A. Kromer, M.S. Kessler, D. Milfay, I.N. Birg, M. Bunck, L. Czibere, M. Panhuysen, B. Putz, J.M. Deussing, F. Holsboer, R. Landgraf, C.W. Turck, Identification of glyoxalase-I as a protein marker in a mouse model of extremes in trait anxiety, J. Neurosci. 25 (2005) 4375–4384. [11] M. Kuloglu, M. Atmaca, E. Tezcan, O. Gecici, H. Tunckol, B. Ustundag, Antioxidant enzyme activities and malondialdehyde levels in patients with obsessive-compulsive disorder, Neuropsychobiology 46 (2002) 27–32.
166
P. Politi et al. / Neuroscience Letters 396 (2006) 163–166
[12] Y.J. Lee, C. Hohoff, K. Domschke, P. Sand, G. Kuhlenbaumer, A. Schirmacher, C.M. Freitag, J. Meyer, G. Stober, P. Franke, M.M. Nothen, J. Fritze, R. Fimmers, H.S. Garritsen, F. Stogbauer, J. Deckert, Norepinephrine transporter (NET) promoter and 5 -UTR polymorphisms: association analysis in panic disorder, Neurosci. Lett. 377 (2005) 40– 43. [13] C. Rothe, D. Koszycki, J. Bradwejn, N. King, V. De Luca, S. Shaikh, P. Franke, H. Garritsen, J. Fritze, J. Deckert, J.L. Kennedy, Association study of serotonin-2A receptor gene polymorphism and panic disorder in patients from Canada and Germany, Neurosci. Lett. 363 (2004) 276– 279. [14] A.A. Samadi, S.A. Fullerton, D.G. Tortorelis, G.B. Johnson, S.D. Davidson, M.S. Choudhury, C. Mallouh, H. Tazaki, S. Konno, Glyoxalase, I phenotype as a potential risk factor for prostate carcinoma, Urology 57 (2001) 183–187. [15] D.V. Sheehan, Y. Lecrubier, K.H. Sheehan, P. Amorim, J. Janavs, E. Weiller, T. Hergueta, R. Baker, G.C. Dunbar, The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation
[16]
[17]
[18]
[19] [20]
of a structured diagnostic interview for DSM-IV and ICD-10, J. Clin. Psychiatry 59 (1998) 22–33. N.M. Simon, M.W. Otto, J.A. Smits, D.C. Nicolaou, H.E. Reese, M.H. Pollack, Changes in anxiety sensitivity with pharmacotherapy for panic disorder, J. Psychiatr. Res. 38 (2004) 491–495. P.J. Thornalley, Glyoxalase I-structure, function and a critical role in the enzymatic defence against glycation, Biochem. Soc. Trans. 31 (2003) 1343–1348. O.A. van den Heuvel, B.J. van de Wetering, D.J. Veltman, D.L. Pauls, Genetic studies of panic disorder: a review, J. Clin. Psychiatry 61 (2000) 756–766. S. Villafuerte, M. Burmeister, Untangling genetic networks of panic, phobia, fear and anxiety, Genome Biol. 4 (2003) 224. M.M. Weissman, R.C. Bland, G.J. Canino, C. Faravelli, S. Greenwald, H.G. Hwu, P.R. Joyce, E.G. Karam, C.K. Lee, J. Lellouch, J.P. Lepine, S.C. Newman, M.A. Oakley-Browne, M. Rubio-Stipec, J.E. Wells, P.J. Wickramaratne, H.U. Wittchen, E.K. Yeh, The cross-national epidemiology of panic disorder, Arch. Gen. Psychiatry 54 (1997) 305–309.