Introduction

Idiopathic orthostatic intolerance (IOI), also termed postural tachycardia syndrome, is characterized by an array of symptoms which occur upon assuming the upright position, with tachycardia and an exaggerated plasma norepinephrine response (Jacob and Biaggioni 1999). IOI is of great relevance for occupational medicine, because the secondary problems cause a wide range of personal, social and occupational difficulties. Firstly, severe IOI implies a substantial impairment of well-being and work performance (Grubb B et al. 2003; Winker et al. 2003). Secondly, it represents a major safety risk for particular professions (i.e. construction workers, roofers, and all employees who operate dangerous machines) (Winker and Rüdiger 2001). Unlike other syndromes with some similar characteristics, such as neurogenic orthostatic hypotension, the basis for IOI remains obscure (Jacob and Biaggioni 1999; Jordan et al. 1999). Several features of the syndrome, including the tachycardia, palpitations and tremulousness, have been considered potentially due to β-adrenergic receptor supersensitivity or other hypo/hyperactive states of autonomic receptors (Farquhar et al. 2000; Jacob and Biaggioni 1999; Jordan et al. 2002). The sympathetic nervous system is highly integrated, with nine different receptors for epinephrine and norepinephrine. Six adrenergic receptors (ARs) are known to have polymorphic variations in their coding regions that result in altered function or regulation (Green et al. 1993, 1994; Mason et al. 1999; Rathz et al. 2002; Small et al. 2001, 2000a, 2000b). Relevant to potential causes of IOI are the Ser/Gly49 and Gly/Arg389 forms of the β1AR; the Arg/Gly16, Gln/Glu27, and Thr/Ile164 forms of the β2AR; the Asn/Lys251 forms of the α2AAR; the “wild-type” and Del301-303 forms of the α2BAR; and the “wild-type” and Del322-325 forms of the α2CAR. Many of these polymorphisms are common in the population, and we have considered that one (or potentially a combination of several) may be associated with, are protective against, or modify IOI. To investigate this issue, two cohorts of young men with or without IOI were identified among military personnel, and genotypes at eight polymorphic loci within the five relevant adrenergic receptor genes were ascertained.

Materials and methods

Study population

Controls (n=60) and patients with IOI (n=20) were recruited from the Austrian Army between January 2003 and December 2003. All 80 subjects were male Caucasians. The participants did not exhibit signs or symptoms of systemic illness and did not take medications, including nonprescription antihistamines or decongestants. All participants had had a traditional medical evaluation when entering the army several months prior to study. Those with a history of metabolic or neurologic diseases that might affect the autonomic nervous system were excluded. Supine blood pressures less than 90/50 or greater than 140/90, neurogenic orthostatic hypotension, mitral valve prolapse and deconditioning were considered exclusion criteria.

All participants underwent a standardised tilt table test. A digital plethysmograph (Task Force Monitor; CNSystems, Graz, Austria) was installed around the second phalanx of the middle finger of the right hand for the continuous noninvasive measurement of arterial blood pressure. The right arm was positioned at heart level. In addition to this beat-by-beat recording, blood pressure and heart rate were measured every 2 min at the left arm. The finger blood pressure signal was calibrated to the oscillometric blood pressure measurement as described (Gratze et al. 1998). Participants were supported by belts at the hips and by a footrest. Subjects equilibrated in the supine position for 30 min and then were subjected to a 75° upright tilt for 30 min. Plasma catecholamines were determined immediately at the conclusion of the supine and upright phases.

Patients were designated as having IOI if they met the following criteria:

  1. 1.

    The occurrence of characteristic orthostatic symptoms for at least 6 months. These included at least three of the following: dizziness, headache, chest discomfort, tremulousness, palpitations, nausea or hyperhydrosis.

  2. 2.

    Orthostatic tachycardia within 5 min after assuming an upright position (heart rate increase ≥30 beats per min without a decrease of ≥20 mmHg in systolic or ≥10 mmHg in diastolic blood pressure).

  3. 3.

    Plasma norepinephrine concentrations >600 pg/ml while standing upright.

Each IOI subject was matched to three controls of similar age (thus a total of 60 controls). The protocol was approved by the ethics committee of the medical faculty of the University of Vienna and the University of Cincinnati College of Medicine, and all participants gave written informed consent before entering the study.

Plasma catecholamines determination

Plasma catecholamine levels were determined by high-performance liquid chromatography (HPLC) (Chromsystems; Munich, Germany) as previously described (Tschernko et al. 1996). Briefly, plasma was collected and frozen to −70°. Samples were mixed with dihydroxybenzylamine for recovery estimations and were afterward extracted with A1203. Eluted catecholamines were automatically injected and separated on a C18 silica column by HPLC. Catecholamines were measured by an electrochemical detector (Pharmacia, Sweden).

Genotyping

Genotyping was performed at the University of Cincinnati by S. Liggett. Genomic DNA was isolated from samples of peripheral blood, and the adrenergic receptor polymorphisms were determined exactly as previously described (Small et al. 2002a).

Statistical analysis

Data were analyzed with the Statistical Package for Social Sciences (SPSS, Chicago, IL, USA). The results are expressed as mean ± SD. Two-tailed t tests and Mann–Whitney tests were used for comparison between groups concerning the following variables: age, weight, height, heart rate, blood pressure and plasma norepinephrine concentration. Haemodynamic variables were compared in one-way analysis of variance with genotype as factor. All group differences were evaluated in post hoc analysis with the Scheffe test. Odds ratios as a measure of relative risk for IOI and 95% confidence intervals (95% CI) were calculated by standard methods. Chi-square tests were used to test for associations between IOI and genotype or allele. Differences were considered statistically significant at P<0.05.

Results

Characteristics of the subjects

The characteristics of the patients with IOI and the controls are shown in Table 1. The mean age of the patients with IOI was 19.8 ± 2.4 years, and 20.4 ± 2.6 years for the controls. There were no differences in weight, size or age. Heart rate and blood pressure in the supine position were similar in both groups. Based on the predefined criteria, the IOI subjects had ~30 bpm higher heart rates, and upright norepinephrine levels approximately twofold greater, compared with the normal subjects. In addition, we noted that the IOI subjects had diastolic blood pressures in the supine position that were slightly higher than controls, as well as higher supine norepinephrine levels.

Table 1 Age, weight, size, and autonomic responses of study population. ± Values are mean + SD. To convert the values for norepinephrine to nanomoles per litre, multiply by 0.0059
Full size table

Potential relationships between adrenergic receptor polymorphisms and IOI

The distribution of genotypes of the adrenergic receptor polymorphisms is shown in Table 2. A polymorphism of the β1AR (Gly49) was significantly less common in IOI (allele frequency of β1Gly49 was 0.26 in controls and 0.10 in patients with IOI; P=0.04). The odds ratio based on the allele frequencies for the Gly49 allele was 0.319 (95% CI, 0.105–0.969). The odds ratio for IOI based on the Gly49 genotype distribution determined by two-by-three chi-square test was 0.88 (95% CI, 0.81–0.97, P=0.16). There were no homozygous Gly49 IOI subjects, so an analysis using the two homozygous and the heterozygous genotypes could not be considered. There was no association (Table 2) with the other functional β1AR polymorphism at position 389 and IOI. β2AR polymorphisms have also been shown to affect cardiac and vascular function (Brodde et al. 2001; Dishy et al. 2001; Hoit B et al. 2000), but in our study there was no difference in the frequencies of the two major β2AR polymorphisms between IOI and controls.

Table 2 Distribution of adrenergic receptor variants among controls and patients with IOI. P Values for the comparison of allele frequencies between controls and patients with IOI were determined by two-by-two chi-square tests; P values for the comparisons of genotype distributions between controls and patients with IOI were determined by two-by-three chi-square tests; CI confidence interval, NA No calculation possible, since there were no subjects who were homozygous for the β2Ile164 polymorphism or the α2aaN251K polymorphismphism
Full size table

The α2AAR and α2CAR control presynaptic norepinephrine release; however, coding polymorphisms of these receptors’ genes are uncommon in Caucasians (Table 2; Small and Liggett 2001; Small et al. 2003). For the α2AAR, the Lys251 allele was not found in any of the 80 individuals. The α2CDel322-325 was uncommon in both groups, and there was no statistical difference between allele frequencies between IOI patients (0.025) and controls (0.075, P=0.26). The α2BDel301-303 polymorphism is common in Caucasians (Small et al. 2001), and has been associated with vascular responsiveness to catecholamines (Heinonen et al. 2002). However, we found no difference in allele frequency between our IOI and control subjects (0.28 in both groups).

Potential relationships between adrenergic receptor polymorphisms and phenotypic variables

The IOI cohort was defined by symptoms, blood pressure, and norepinephrine criteria. We considered that, while for some polymorphisms there were no associations with IOI as so defined, there may be associations with one physiologic aspect of the syndrome. To approach this possibility, potential relationships between blood pressure, heart rate, and norepinephrine levels, without regard to diagnosis of IOI, were explored.

In this model, we again found a relationship between heart rate and the position 49 polymorphism of the β1AR. Gly49 homozygotes (only found in controls) had upright heart rates of 69 ± 11 bpm, compared with 86 ± 17 bpm for Ser49 homozygotes, and an intermediate rate for heterozygotes of 82 ± 17 bpm (P=0.04 by ANOVA). In the group of IOI subjects, no relationship was found between heart rate and the β149 polymorphism. Also no association was found between the other β1AR polymorphic locus, at position 389, and baseline haemodynamic variables (Table 3). The two common β2AR polymorphisms studied showed no relationship between blood pressure and heart rate as well. Concerning the α2AR polymorphisms, the low frequency of the α2CDel322-325 allele limited analysis, but no trends were noted in heart rate or blood pressure parameters and presence of the allele. The common α2BAR polymorphism did not show any trend in association of heart rate or blood pressure parameters.

Table 3 β1AR genotype and haemodynamic variables of all subjects (n=80) . ± Values are means ± SD
Full size table

None of the receptor variants were associated with supine, upright, or the change in plasma norepinephrine. As noted earlier, the genes for the two receptors (α2AAR and α2CAR) that control presynaptic release of norepinephrine had either no or rare variability at the loci examined in these Caucasian cohorts.

Discussion

Genetic variability within the coding regions of ARs in the human population was first demonstrated in 1992 with the β2AR (Heinonen et al. 2002). Subsequently, nonsynonymous coding polymorphisms of the α1A, α2A, α2B, α2C, β1 and β3ARs have been identified (Small et al. 2003). (To date, the α1B and α1D coding regions have not been closely examined for variation.) Using heterologous expression systems in cells (Green et al. 1993, 1994; Mason et al. 1999; Rathz et al. 2002; Small et al. 2001, 2000a, 2000b) or mice (Mialet Perez et al. 2003; Turki et al. 1996), the relevance of these variants has been ascertained within the constraints imposed by such models. The α1A polymorphism was found to have no functional effect in recombinantly expressed CHO cells (Shibata et al. 1996). However, each of the other nonsynonymous polymorphisms of the α2A, α2B, α2C, β1, β2 and β3ARs have been reported to have a signaling phenotype (reviewed in Small and Liggett [2001]; Small et al. [2003]). These include alterations in expression, ligand binding affinities, functional coupling to G proteins, and regulation.

In case control as well as a limited number of family studies, the nonsynonymous coding polymorphisms of the α2B, α2C, β1 and β2ARs have been associated with a number of cardiovascular phenotypes (Small et al. 2003). These include associations for risk such as hypertension (Bray et al. 2000) and heart failure (Small et al. 2002b). In addition, certain of these polymorphisms have been associated with disease modification or disease subsets, such as survival (Liggett et al. 1998) and exercise tolerance in heart failure (Wagoner et al. 2000, 2002), and the response to β-blockers (Johnson et al. 2003). Other studies have revealed effects of certain adrenergic receptor polymorphisms on cardiac or vascular function in response to agonists or sympathetic nervous system activation in normal individuals (Brodde et al. 2001; Dishy et al. 2001). Results from these in vitro and human studies prompted us to examine the potential for associations of the functional coding polymorphisms of ARs and IOI. The features of IOI, either in terms of clinical symptoms and signs, or the physiologic findings that define the syndrome, are highly suggestive of altered cardiac, vascular, or norepinephrine kinetic responses in response to orthostatic challenge. Indeed, in recent studies of the rare entity familial orthostatic intolerance, a mutation of the coding region of the norepinephrine transporter gene (which is not polymorphic) has been identified as causative (Shannon et al. 2000).

We found that the bradycardia-promoting β1Gly49 polymorphism was underrepresented in the IOI group. See Fig. 1 for the basis of our hypothesis that the β1Gly49 variant may decrease the risk for IOI. In transfected cells, Gly49 β1AR undergoes enhanced agonist-promoted down-regulation (Rathz et al. 2002) and thus in the heart would represent a genotype less prone to tachycardia. This is consistent with the current findings. The Gly49 receptor has an atypical glycosylation pattern, which appears to be related to an enhanced steady-state degradation of internalized receptors. The Gly49 receptor has previously been reported to be associated with an improved outcome in heart failure (January et al. 1998) and resting heart rate (Ranade et al. 2002). The Gly49 allele frequencies were found to differ significantly between controls and patients (see Table 2). The observed allele frequency for controls was found to be in good accordance with a recently reported allele frequency for Caucasians (Small et al. 2002a). In summary, we have shown a significant association between Gly49 genotype and heart rate, as well as a significantly different allele frequency between IOI subjects and controls. Comparison of genotype distributions between controls and patients did not differ significantly. Nevertheless a trend supporting our hypothesis, that the β1Gly49 might be protective for IOI, was observed (none of the patients was homozygous for glycine, compared with 11.7% of the controls). To confirm these findings, further studies should include a larger number of patients.

Fig. 1
figure 1

Basis of the hypothesis that the α2CDel322-325 and the β1Arg389 variants may increase and the β1Gly49 variant may decrease the risk for idiopathic orthostatic intolerance. Under normal conditions, norepinephrine is released from vesicles in the neuron into the synaptic space, where it can interact with presynaptic and postsynaptic α- and β-adrenergic receptors. β1-Adrenergic receptors couple to the stimulatory G protein Gs, activating adenylyl cyclase and increasing intracellular cyclic AMP (cAMP). Eighty percent of the norepinephrine in the synaptic space is taken up by the norepinephrine transporter into the neuron, which releases it, and approximately 20% spills over into circulation. Norepinephrine is then converted to dihydroxyphenylglycol (DHPG) by monoamine oxidase. The α2-adrenergic receptor inhibits norepinephrine release at presynaptic nerve endings through negative feedback. The presence of the dysfunctionally running α2CDel322-325 receptor would therefore be expected to result in an in vivo increased norepinephrine release at the synapse. The hyperfunctional β1Arg389 receptor would be expected to increase chronotropy at the myocyte. Therefore, the presence of these two receptor polymorphisms were hypothesised to be risk factors for IOI. The β149Gly receptor variant, which should lead to decreased function via down-regulation was hypothesised to be a protecting factor for IOI.

Full size image

Our current results now indicate that the underrepresentation of Gly49 (or increased frequency of Ser49) represents one of likely several genetic, and gene–environment interactions, associated with IOI. While it is attractive to consider that this polymorphism in combination with another adrenergic receptor polymorphism might contribute an additive effect, we found no evidence for interactions between any of the common adrenergic receptor polymorphisms and IOI risk. However, the low allele frequencies of β2Ile164, α2ALys251 and α2CDel322-325 have limited our ability to ascertain effects due to sample size. It should also be noted that promoter, 5′ UTR, or 3′ UTR polymorphisms of ARs may influence receptor expression (Drysdale et al. 2000), and these polymorphisms may not be in linkage disequilibrium with the coding polymorphismic sites examined in the current study (Drysdale et al. 2000). Thus, we cannot exclude some non-coding polymorphisms/haplotypes of ARs as contributing to IOI pathophysiology. Studies to address this will require identification and characterisation of these non-coding polymorphisms for each adrenergic receptor gene. We note that the odds ratio for protection against IOI is modest (0.88), and we are underpowered to address certain adrenergic receptor polymorphisms due to their low frequency. Further, as noted, non-coding polymorphisms were not considered. So, we cannot unequivocally exclude an adrenergic receptor polymorphism “load-score” (comprising multiple polymorphisms of multiple receptors) that associates with IOI. Such a study will require larger cohorts and genotyping of additional polymorphisms within all the genes. And finally, our study was restricted to males, but the syndrome is more common in females. Although a difference in allele frequencies has never been found by the Department of Molecular Genetics, University of Cincinnati (S. Liggett, personal communication), a sex–genotype–disease interaction cannot be excluded for IOI.