- Email: [email protected]
Promoter polymorphisms of the NOS3 gene are associated with hypnotizability-dependent vascular response to nociceptive stimulation
Neuroscience Letters 467 (2009) 252–255
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Promoter polymorphisms of the NOS3 gene are associated with hypnotizability-dependent vascular response to nociceptive stimulation Silvano Presciuttini a,∗ , Michele Curcio b , Roberta Chillemi a , Serena Barbuti b , Fabrizio Scatena b , Giancarlo Carli c , Brunello Ghelarducci a , Enrica L. Santarcangelo a a b c
Dept. of Physiological Sciences, University of Pisa, Via San Zeno 31 - 56127 Pisa, Italy Immunohematology Unit, Azienda Ospedaliera-Universitaria Pisana, Italy Dept. of Physiology, University of Siena, Italy
a r t i c l e
i n f o
Article history: Received 18 August 2009 Received in revised form 6 October 2009 Accepted 18 October 2009 Keywords: Hypnotizability Endothelial function NOS3 Haplotype analysis Linkage disequilibrium
a b s t r a c t Hypnotizability is associated with a few physiological characteristics also in the normal awake state. Differences in flow-mediated dilation (FMD) have been observed in subjects with high (Highs) or low (Lows) hypnotizability during nociceptive stimulation. FMD is largely due to the nitric oxide (NO) produced by vascular endothelium through the activity of NO synthase (eNOS). Endothelial NOS is encoded by the NOS3 locus. Aim of this pilot study was to investigate the association between genetic polymorphisms of NOS3 involved in NO blood levels and hypnotizability. Nine single nucleotide polymorphisms (SNPs) of the NOS3 gene were analyzed in the DNA of 24 Highs, 22 Lows, and 61 newborns. Two SNPs, rs1800783 (−1474 T/A) and rs2070744 (−786 T/C), located in the upstream and promoter region of the gene, respectively, showed significant differences between Highs and Lows in allele frequency. Haplotype analysis showed that the newborns were in linkage equilibrium for these SNPs, whereas both Highs and Lows showed linkage disequilibrium. The A-C haplotype (associated with lower NO availability in the general population) was more frequent in Highs, and the T-T haplotype was more frequent in Lows. Thus, the lower FMD reduction observed in Highs during nociceptive stimulation, which is indicative of higher NO availability, should be due to greater efficacy of shear stress-related transcriptional factors and/or to lower effects of NOS inhibitory controls. A consequent theoretical proposal concerns the possible role of NO in the brain vessels where, in stimulation conditions, NO diffusion to the extracellular compartment might be involved in hypnotic responding. © 2009 Elsevier Ireland Ltd. All rights reserved.
Hypnotizability is a multidimensional trait that remains substantially stable during all life, allowing the experience of hypnotic trance [14]. Measured on standard scales, hypnotizability is normally distributed in the general population, with about 10% of all individuals being highly hypnotizable (Highs) and another 10% being scarcely or not at all hypnotizable (Lows) [1,11,8]. Recent evidence indicates that Highs and Lows exhibit different physiological responses to a few cognitive and physical stimuli even in the absence of hypnotic induction and/or specific suggestions [6,31,32]. One of these concerns the “endothelial function” (FMD), consisting of a post-ischemic flow-mediated vasodilation. FMD is due to the flow-mediated shear stress which induces the production of nitric oxide (NO) by vascular endothelium [30]. A reduced vasodilation in basal conditions – “endothelial dysfunction” – is predictive of cardiovascular disease [34], but it can be transiently induced in healthy subjects by cognitive stress [13–15,18,35] and acute pain
∗ Corresponding author. Tel.: +39 050 2213525; fax: +39 050 2213527. E-mail address: [email protected] (S. Presciuttini). 0304-3940/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2009.10.056
[17]. In the absence of physical or cognitive stimulation, not hypnotized Highs and Lows show comparable brachial artery diameters and FMD. Hypnotizability-related differences in FMD appear during cognitive load, when Highs do not exhibit any endothelial dysfunction [15,16], and during experimental pain, when they exhibit a significantly lower endothelial dysfunction than Lows [17]. Theoretically, these differences between Highs and Lows may be due either to sympathetically controlled hemodynamic responses eliciting different post-occlusion shear stress between Highs and Lows and/or to hypnotizability-related thresholds for NO production. The first hypothesis should be rejected, as nociceptive stimulation and cognitive load induce the same reduction of systolic/mean blood pressure in the two groups with respect to basal conditions [28] and, consequently, a similarly reduced shear stress. Thus, a lower threshold for NO production by the Highs vascular endothelium should account for the FMD differences previously observed [15–18]. Endogenous NO is generated from arginine by a family of three distinct calmodulin-dependent NO synthase (NOS) enzymes: NOS from endothelial cells (eNOS) and neurons (nNOS) are constitu-
S. Presciuttini et al. / Neuroscience Letters 467 (2009) 252–255
tively expressed, while expression iNOS (inducible NOS) requires transcriptional activation mediated by citokines. eNOS is mainly responsible for smooth muscle relaxation and nNOS for neuronal signalling; yet, eNOS and nNOS exert a cooperative role in the control of heart contractility [24], and spontaneous myocardial infarction associated with multiple cardiovascular risk factors has been observed in mice only after genetic disruption of the whole NOS system [26]. In addition, NO from eNOS in the nucleus tractus solitarii (NTS) plays a major role in the control of baroreflex gain and arterial pressure through vascular-neuronal signalling [29] and/or the modulation of the interaction between astrocytes and neurons [23]. Endothelial NOS is encoded by the NOS3 locus, which contains 26 exons spanning approximately 21 kilobases. Among the many polymorphisms identified in this gene, two functional single nucleotide polymorphisms (SNPs) have been especially investigated in relation to cardiovascular disease, namely, the promoter −786T>C variant (rs2070744) and the missense mutation 894G>T (Glu298Asp, rs1799983) [10,19,20,24,25]. Aim of the present study was to examine whether individual differences in hypnotizability are associated with single nucleotide polymorphisms (SNPs) of NOS3. As linkage disequilibrium (LD) among SNPs may generate in populations specific haplotypes, which are the actual determinants of individual susceptibility, we investigated a total of nine SNPs of this gene, and determined two-locus haplotype frequencies in the presence of significant LD. We hypothesized that some eNOS haplotypes could be associated with susceptibility to hypnosis. After signing an informed consent approved by the local Ethics committee, 46 healthy volunteers joined the study which followed the rules of the Declaration of Helsinki. They were selected according to their hypnotizability level through the Italian version [11] of the Stanford Hypnotic Susceptibility Scale (SHSS, form C) among 240 students of the Pisa University and divided in 24 Highs (score ≥ 9/12), age (mean ± SD) 24 ± 3.2 years, 15 females, and 22 Lows (score ≤ 3/12), age 25 ± 2.8 years, 16 females. Sixty-one newborns (umbilical cords from the Immunohemathology Unit Bank at the Azienda Ospedaliera–Universitaria Pisana) were included in the study as representative of the general population. Genomic DNA was extracted through a salting out procedure from Highs’ and Lows’ peripheral blood leukocytes. For privacy requirements, blood samples were coded anonymously. The DNA extracted from 500 l of blood was diluted with 100 l H2 O, quantified by UV measurement at OD 260 nm and stored at −20 ◦ C until further processing. Nine known single nucleotide polymorphisms (SNPs) of NOS3 (Table 1) were analyzed by direct sequencing using published primers (http://genecanvas.idf.inserm.fr/). The “rs” numbers of these SNPs were obtained by cross-checking against the National Center of Biotechnology Information (NCBI) database. Calculations and statistical tests were carried out in Excel (Microsoft Corporation, Redwood, WA, USA) using our own routines. Adherence of genotype frequencies to Hardy–Weinberg equilibrium was assessed by goodness-of-fit tests. Heterogeneity of allele frequencies among population samples was assessed
Table 1 “rs” number and nucleotide change of the typed NOS3 variants. Variant
Base change
Function
rs1800783 rs2070744 rs1800781 rs1549758 rs1799983 rs3918174 rs1800780 rs3918188 rs1799985
−1474 T/A −786 T/C +174 G/A 774 C/T 894 G/T +294 A/G +52 G/T +15 G/A +11 G/T
Upstream Promoter Intron 3 Silent Glu298Asp Intron 9 Intron 12 Intron 14 Intron 23
253
by contingency-table 2 analysis. Difference of allele frequency between Highs and Lows was assessed by the Fisher exact test. Maximum likelihood estimates of two-locus gametic frequencies were obtained by the expectation-maximization (EM) algorithm [36]; linkage disequilibrium (LD) between SNP pairs was measured using the absolute value of Lewontin’s D (|D |); its value is included between 0 (the state of gametic equilibrium) and 1 (complete LD); the significance of D was assessed by 2 . Significance level was set at P < 0.05. The genotype distribution of each of the nine SNPs in each of the three samples was tested for HW equilibrium. Only one out of the 27 tests was statistically significant, but significance disappeared after Bonferroni correction. As shown in Table 2, two SNPs located in the upstream region (−1474T/A, rs1800783) and in the promoter (−786T/C, rs2070744), respectively, showed both significant heterogeneity of allele frequency among the three samples (2 = 7.46, P = 0.024 and 2 = 5.87, P = 0.053, respectively) and were specifically associated with high or low hypnotizability (Fisher exact test). The newborn allele frequencies were intermediate between those of Highs and Lows in both cases. Two-locus linkage disequilibrium (LD) analysis was conducted separately in the three samples. For the above two SNPs, LD was non-significant in the sample of newborns (D = 0.177, P = 0.342), whereas it was highly significant in Highs (D = 0.763, P = 0.002) and borderline significant in Lows (D = 0.544, P = 0.070). Table 3 shows the estimated haplotype frequencies in the three samples; the A-C haplotype was 3.4-fold more frequent in Highs than in newborns (0.472 vs. 0.138) and the T-T haplotype was 1.8-fold more frequent in Lows than in newborns (0.577 vs. 0.318). This study reports the first evidence of a genetic substrate for hypnotizability-related vascular responses. We typed nine SNPs of NOS3 in two highly selected samples (the Highs and the Lows, which belongs to the upper and lower deciles of the population, measured on the scale of hypnotizability) as well as in a sample of the general population (represented by umbilical cords of newborns). With regard to hypnotizability, the general population is an admixture of different phenotypes (most individuals score at intermediate values) and is expected to show allele frequencies intermediate between Highs and Lows for those polymorphisms that are truly associated with the trait, i.e. for which Highs and Lows show significant differences. Indeed, this has been the case of the two SNPs −1474 T/A and −786 T/C, whose allele frequencies were significantly different between Highs and Lows and which showed a trend of allele frequency from Lows to newborns to Highs. This result strongly supports the authenticity of the association, in spite of the small sample sizes of Highs and Lows. In addition, the two SNPs are close to each other (no other SNPs have been typed in between), and the fact that they show similar features is a further indication that the association is not due to random effects. Another result of the present work supports the validity of the association. Haplotype analysis showed that the general population was in linkage equilibrium for the two SNPs, whereas both Highs and Lows showed high or intermediate values of linkage disequilibrium; this is expected if a specific haplotype is selected along with an extreme phenotype. Indeed, the A-C haplotype frequency was increased in Highs, whereas the T-T haplotype frequency was increased in Lows. The overall effect appears to be strong, as it has been detected in small samples. In our analysis, no other SNP or two-locus haplotype showed evidence of association with hypnotizability. The allele C of the −786 T/C SNP has been associated, although not unanimously, with reduced NO activity [10,3] and consequent cardiovascular disease, particularly atherosclerosis, coronary artery disease, myocardial infarction, stroke, cerebral and peripheral artery disease, intracranial aneurysm [see 21,22,27]. Our
254
S. Presciuttini et al. / Neuroscience Letters 467 (2009) 252–255
Table 2 Genotype distribution and allele frequency of 9 SNPs of the NOS3 gene. SNP
Genotype
Highs
Lows
Newborn
P (het)
P (Fisher)
rs1800783
TT AT AA T frequency
3 13 8 0.40
10 9 3 0.66
23 26 12 0.59
0.024
0.013
rs2070744
TT CT CC T frequency
6 11 7 0.479
11 10 1 0.727
21 30 10 0.59
0.053
0.02
rs1800781
GG GA AA G frequency
11 5 8 0.563
6 12 4 0.545
22 25 14 0.566
NS
NS
rs1549758
CC CT TT C frequency
14 7 3 0.729
10 11 1 0.705
30 26 5 0.705
NS
NS
rs1799983
GG GT TT G frequency
18 6 0 0.875
16 6 0 0.864
38 20 3 0.787
NS
NS
AA AG GG A frequency
6 11 7 0.479
8 11 3 0.614
21 30 10 0.59
NS
NS
rs1800780
GG GT TT G frequency
20 4 0 0.917
19 3 0 0.932
45 15 1 0.861
NS
NS
rs3918188
GG GA AA G frequency
16 6 2 0.792
14 8 0 0.818
41 17 3 0.811
NS
NS
rs1799985
GG GT TT G frequency
11 12 1 0.708
7 14 1 0.636
26 31 4 0.68
NS
NS
rs3918174
Note. P (het): P-value for heterogeneity of allele frequency among the three samples; P (Fisher): P-value of the exact test between Highs and Lows (2 × 2). NS: P > 0.05.
results show that this allele is part of a haplotype with higher frequency in Highs. Thus, the present results seem to contrast with previous findings on endothelial function suggesting a higher NO activity in Highs during nociceptive stimulation and mental stress on the basis of their scarcely reduced post-occlusion vascular dilation (FMD) [15–17]. As hemodinamic differences induced in Highs and Lows by stimulations have been excluded [28] and, thus, a similarly reduced shear stress occurs in the two groups during stimulations, the evidences concerning FMD and eNOS can be reconciled only by hypothesizing that the higher NO availability accounting for the Highs’ larger FMD is obtained through greater efficacy of transcriptional factors induced by shear stress and/or through lower effects of NOS inhibitory controls affecting eNOS activity and function [9]. Apparently, these modulator factors enable the Highs’ endothelium to provide higher NO availability in spite of their apparently lower NOS3 activity, which is associated with lower
Table 3 Frequency estimates of two SNPs haplotypes at rs1800783 and rs2070744 (−1474 T/A and −786 T/C). Haplotype
Newborns
Highs
Lows
A-C A-T T-C T-T
0.138 0.272 0.272 0.318
0.472 0.132 0.049 0.347
0.191 0.150 0.082 0.577
NO availability in the general population. Thus, the present findings suggest that high hypnotizability might be a protective factor against the NO deficiency possibly occurring in subjects showing this NOS3 variant [21,22,27]. The lack of FMD evaluation in the same subjects undergoing the present NOS investigation might be considered a limitation of the study. However, FMD results assessing a lower FMD reduction in Highs during cognitive and physical stimulation had been previously replicated in four separate groups of subjects [15–18]. Altogether, our results [15–18] suggest that being a High might be important for chronic pain patients for more than the wellassessed control of pain experience [12]. Indeed, a larger availability of NO is a relevant prognostic factor in these patients as they exhibit 13.6% of coronary disease as compared to 6.5% in matched subjects not suffering from chronic pain [2]. A higher endothelial derived NO activity in the NTS of chronic pain patients with high susceptibility to hypnosis might also reduce the sympathetic overactivity or parasympathetic underactivity reported in chronic pain patients [see 7] and contributing to pain chronicization [20]. In addition, the sympathetically mediated control of immune system [19] may be differentially modulated in Highs and Lows and be responsible for a lower frequency of occurrence of diseases due to immune system deficiency in Highs [see 7]. Finally, we would like to emphasize that the endothelial derived NO plays a role not only in the peripheral and brain vascular diameter control, but also in the modulation of neuronal activity due
S. Presciuttini et al. / Neuroscience Letters 467 (2009) 252–255
to its diffusion from brain vessels into the extracellular compartment; thus, it may be involved also in other hypnotizability-related differences. Indeed, it has been shown that nitric oxide inhalation increases imaginative ability [37] that is an important component of hypnotic responding [14,38]. A larger availability of NO might induce a greater general excitability of the cerebral cortex and account also for hypnotizability-related differences observed in the cognitive and sensori-motor domain, i.e. the equivalence of imagery and perception occurring more often and more effectively in Highs than in Lows [4,5,33], the Highs’ ability to imagine through both the visual and tactile/proprioceptive modality [5] and the peculiar postural and locomotor control observed in Highs [25,32]. In conclusion, the NO availability, which depends on individual genotype in stimulation conditions, could have a causal role in hypnotic responding. This “NO hypothesis” might be a new frontier in the physiology of hypnotic susceptibility, but it should be further investigated in its interactions with other factors, i.e. gender, personality traits and social/cultural level, in large samples. Acknowledgments Research funded by Azienda Ospedaliera-Universitaria Pisana. Dr. I. Bianco is sincerely acknowledged for valuable technical collaboration. References [1] G.C. Balthazard, E.Z.E.Z. Woody, Bimodality, dimensionality, and the notion of hypnotic types, Int. J. Clin. Exp. Hypn. 37 (1989) 70–89. [2] A Berger, E.M. Dukes, G. Oster, Clinical characteristics and economic costs of patients with painful neuropathic disorders, J. Pain 5 (2004) 143–149. [3] P.F. Binkley, E. Nunziatta, Y. Liu-Stratton, G.E. Cooke, A polymorphism of the endothelial nitric oxide synthase promoter is associated with an increase in autonomic imbalance in patients with congestive heart failure, Am. Heart J. 149 (2005) 342–348. [4] G. Carli, C. Rendo, L. Sebastiani, E.L. Santarcangelo, Suggestions of altered balance: possible equivalence of imagery and perception, Int. J. Clin. Exp. Hypn. 54 (2006) 206–223. [5] G. Carli, F.I. Cavallaro, C.A. Rendo, E.L. Santarcangelo, Imagery of different sensory modalities: hypnotizability and body sway, Exp. Brain Res. 179 (2007) 147–154. [6] G. Carli, D. Manzoni, E.L. Santarcangelo, Hypnotizability-related integration of perception and action, Cogn. Neuropsychol. 25 (2008) 1065–1076. [7] G. Carli, A. Huber, E.L. Santarcangelo, Hypnotizability and chronic pain: an ambiguous connection, Contemp. Hypn. 25 (2008) 65–77. [8] C. Carvalho, I. Kirsch, G. Mazzoni, I. Leal, Portuguese norms for the Waterloo–Stanford Group C (WSGC) scale of hypnotic susceptibility, Int. J. Clin. Exp. Hypn. 56 (2008) 295–305. [9] A. Chatterjee, J.D. Catravas, Endothelial nitric oxide (NO) and its pathophysiologic regulation, Vasc. Pharmacol. 49 (2008) 134–140. [10] G.E. Cooke, A. Doshi, P.F. Binkley, Endothelial nitric oxide synthase gene: prospects for treatment of heart disease, Pharmcogenomics 8 (2007) 1723–1734. [11] V. De Pascalis, A. Bellusci, P.M. Russo, Italian norms for the Stanford Hypnotic susceptibility scale, form C, Int. J. Clin. Exp. Hypn. 48 (2000) 15–23. [12] S.W.G Derbyshire, M.G. Whalley, D.A. Oakley, Fibromyalgia pain and its modulation by hypnotic and non-hypnotic suggestion: an fMRI analysis, Eur. J. Pain 13 (2009) 542–550. [13] L. Ghiadoni, A.E. Donald, M. Cropley, M.J. Mullen, G. Oakley, M. Taylor, G. O’Connor, J. Betteridge, N. Klein, A. Steptoe, J.E. Deanfield, Mental stress induces transient endothelial dysfunction in humans, Circulation 102 (2000) 2473–2478. [14] J.P. Green, A.F. Barabasz, D. Barrett, G.H. Montgomery, Forging ahead: the 2003 APA Division 30 definition of hypnosis, Int. J. Clin. Exp. Hypn. 53 (2005) 259–264.
255
[15] Z. Jambrik, E.L. Santarcangelo, B. Ghelarducci, E. Picano, L. Sebastiani, Does hypnotisability modulate the stress-related endothelial dysfunction? Brain Res. Bull. 63 (2004) 213–221. [16] Z. Jambrik, L. Chunzeng, E.L. Santarcangelo, L. Sebastiani, B. Ghelarducci, E. Picano, Traditional acupuncture does not modulate the endothelial dysfunction induced by mental stress, Int. J. Cardiovasc. Imaging 20 (2004) 357–362. [17] Z. Jambrik, G.G. Carli, T. Rudish, A. Varga, T. Forster, E.L. Santarcangelo, Modulation of pain-induced endothelial dysfunction by hypnotisability, Pain 116 (2005) 181–186. [18] Z. Jambrik, L. Sebastiani, E. Picano, B. Ghelarducci, E.L. Santarcangelo, Hypnotic modulation of flow-mediated endothelial response to mental stress, Int. J. Psychophysiol. 55 (2005) 221–227. [19] W. Jänig, Pain in the sympathetic nervous system: pathophysiological mechanisms, in: C.J. Mathias, R. Bannister (Eds.), Autonomic Failure, 5th edn, New York Oxford University Press, Oxford, 2008, pp. 175–189. [20] W. Jänig, J.D. Levine, Autonomic-neuroendocrine-immune responses in acute and chronic pain, in: S.B. McMahon, M. Koltzenburg (Eds.), Wall & Mezack’s Textbook of Pain, 5th edn, Elsevier Churchill Livingstone, Edinburgh, 2006, pp. 205–218. [21] P.C. Jaramillo, C. Lanas, F. Lanas, L.A. Salazar, −786T>C polymorphism of the endothelial nitric oxide synthase gene in Chilean subjects with coronary artery disease and controls, Clin. Chim. Acta 387 (2007) 105–108. [22] I.J. Kullo, M. Todd Greene, E. Boerwinkle, J. Chu, S.T. Turner, S.L.R. Kardia, Association of polymorphisms in NOS3 with the ankle-brachial index in hypertensive adults, Aterosclerosis 196 (2008) 905–912. [23] L.H. Lin, O. Taktakishvili, W.T. Talman, Identification and localization of cell types that express endothelial and neuronal nitric oxide synthase in the rat nucleus tractus solitarii, Brain Res. 171 (2007) 42–51. [24] X. Loyer, C. Heymes, J.L. Samuel, Constitutive nitric oxide synthases in the heart from hypertrophy to failure, Clin. Exp. Pharmacol. Physiol. 35 (2008) 483–488. [25] M. Menzocchi, G. Paoletti, A. Huber, G. Carli, F.I. Cavallaro, D. Manzoni, E.L. Santarcangelo, Hypnotizability and sensorimotor integration: an Italian Space Agency Project, Int. J. Clin. Exp. Hypn., in press. [26] S. Nakata, M. Tsutsui, H. Shimokawa, O. Suda, T. Morishita, K. Shibata, Y. Yatera, K. Sabanai, A. Tanimoto, M. Nagasaki, H. Tasaki, Y. Sasaguri, Y. Nakashima, Y. Otsuji, N. Yanagihara, Spontaneous myocardial infarction in mice lacking all nitric oxide synthase isoforms, Circulation 117 (2008) 2211–2223. [27] A. Neyatizadeh, R. Kumar, T. Stobdan, A.K. Goyal, S. Sikdar, M. Gupta, S. Javed, M.A. Qadar Pasha, Endothelial nitric oxide synthase gene haplotypes and circulating nitric oxide levels significantly associate with risk of essential hypertension, Free Radic. Biol. Med. 44 (2008) 1912–1918. [28] G. Paoletti, M. Varanini, G. Carli, R. Balocchi, C. Morizzo, C. Palombo, E.L. Santarcangelo, Does the autonomic system contribute to pain-related endothelial dysfunction? Acta Physiol. 197 (Suppl. 672) (2009) 2. [29] J.F. Paton, H. Waki, A.P. Abdala, J. Dickinson, S. Kasparov, Vascular-brain signaling in hypertension: role of angiotensin II and nitric oxide, Curr. Hypertens. Rep. 9 (2007) 242–247. [30] N. Resnick, H. Yahav, A. Shay-Salit, M. Shushy, S. Schubert, L.C.M. Zilberman, E. Wofovitz, Fluid shear stress and the vascular endothelium: for better and for worse, Prog. Biophys. Mol. Biol. 81 (2003) 177–199. [31] E.L. Santarcangelo, R. Balocchi, E. Scattina, D. Manzoni, L. Bruschini, B. Ghelarducci, M. Varanini, Hypnotizability-dependent modulation of the changes in heart rate control induced by upright stance, Brain Res. Bull. 75 (2008) 692–697. [32] E.L. Santarcangelo, E. Scattina, G. Carli, A. Macerata, D. Manzoni, Hypnotizability-dependent modulation of postural control: effects of alteration of the visual and leg proprioceptive inputs, Exp. Brain Res. 191 (2008) 331–340. [33] E. Scattina, E.L. Santarcangelo, G. Carli, B. Ghelarducci, D. Manzoni, Modulation of vestibular reflex by imagery, Acta Physiol. 197 (Suppl. 672) (2009) 3. [34] M. Shechter, A. Issachar, I. Marai, N. Koren-Morag, D. Freinark, Y. Shahar, A. Shechter, M.S. Feinberg, Long-term association of brachial artery flowmediated vasodilation and cardiovascular events in middle-aged subjects with no apparent heart disease, Int. J. Cardiol. 134 (2009) 52–58. [35] L.E. Spieker, D. Hürlimann, F. Rischitzka, R. Corti, F. Enseleit, S. Shaw, D. Hayoz, J.E. Deanfield, T.F. Lüscher, G. Noll, Mental stress induces prolonged endothelial dysfunction via endothelin-A receptors, Circulation 105 (2002) 2817–2820. [36] J. Terwillinger, J. Ott, Handbook of human genetic linkage, The Johns Hopkins University Press, Baltimore, London, 1994. [37] M.G. Whalley, G.B. Brooks, Enhancement of suggestibility and imaginative ability with nitrous oxide, Psychopharmacology 203 (2009) 745–752. [38] E.Z. Woody, A.J. Barnier, K.M. McConkey, Multiple hypnotizabilities: differentiating the building blocks of hypnotic response, Psychol. Ass. 17 (2005) 200–211.
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Promoter polymorphisms of the NOS3 gene are associated with hypnotizability-dependent vascular response to nociceptive stimulation Silvano Presciuttini a,∗ , Michele Curcio b , Roberta Chillemi a , Serena Barbuti b , Fabrizio Scatena b , Giancarlo Carli c , Brunello Ghelarducci a , Enrica L. Santarcangelo a a b c
Dept. of Physiological Sciences, University of Pisa, Via San Zeno 31 - 56127 Pisa, Italy Immunohematology Unit, Azienda Ospedaliera-Universitaria Pisana, Italy Dept. of Physiology, University of Siena, Italy
a r t i c l e
i n f o
Article history: Received 18 August 2009 Received in revised form 6 October 2009 Accepted 18 October 2009 Keywords: Hypnotizability Endothelial function NOS3 Haplotype analysis Linkage disequilibrium
a b s t r a c t Hypnotizability is associated with a few physiological characteristics also in the normal awake state. Differences in flow-mediated dilation (FMD) have been observed in subjects with high (Highs) or low (Lows) hypnotizability during nociceptive stimulation. FMD is largely due to the nitric oxide (NO) produced by vascular endothelium through the activity of NO synthase (eNOS). Endothelial NOS is encoded by the NOS3 locus. Aim of this pilot study was to investigate the association between genetic polymorphisms of NOS3 involved in NO blood levels and hypnotizability. Nine single nucleotide polymorphisms (SNPs) of the NOS3 gene were analyzed in the DNA of 24 Highs, 22 Lows, and 61 newborns. Two SNPs, rs1800783 (−1474 T/A) and rs2070744 (−786 T/C), located in the upstream and promoter region of the gene, respectively, showed significant differences between Highs and Lows in allele frequency. Haplotype analysis showed that the newborns were in linkage equilibrium for these SNPs, whereas both Highs and Lows showed linkage disequilibrium. The A-C haplotype (associated with lower NO availability in the general population) was more frequent in Highs, and the T-T haplotype was more frequent in Lows. Thus, the lower FMD reduction observed in Highs during nociceptive stimulation, which is indicative of higher NO availability, should be due to greater efficacy of shear stress-related transcriptional factors and/or to lower effects of NOS inhibitory controls. A consequent theoretical proposal concerns the possible role of NO in the brain vessels where, in stimulation conditions, NO diffusion to the extracellular compartment might be involved in hypnotic responding. © 2009 Elsevier Ireland Ltd. All rights reserved.
Hypnotizability is a multidimensional trait that remains substantially stable during all life, allowing the experience of hypnotic trance [14]. Measured on standard scales, hypnotizability is normally distributed in the general population, with about 10% of all individuals being highly hypnotizable (Highs) and another 10% being scarcely or not at all hypnotizable (Lows) [1,11,8]. Recent evidence indicates that Highs and Lows exhibit different physiological responses to a few cognitive and physical stimuli even in the absence of hypnotic induction and/or specific suggestions [6,31,32]. One of these concerns the “endothelial function” (FMD), consisting of a post-ischemic flow-mediated vasodilation. FMD is due to the flow-mediated shear stress which induces the production of nitric oxide (NO) by vascular endothelium [30]. A reduced vasodilation in basal conditions – “endothelial dysfunction” – is predictive of cardiovascular disease [34], but it can be transiently induced in healthy subjects by cognitive stress [13–15,18,35] and acute pain
∗ Corresponding author. Tel.: +39 050 2213525; fax: +39 050 2213527. E-mail address: [email protected] (S. Presciuttini). 0304-3940/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2009.10.056
[17]. In the absence of physical or cognitive stimulation, not hypnotized Highs and Lows show comparable brachial artery diameters and FMD. Hypnotizability-related differences in FMD appear during cognitive load, when Highs do not exhibit any endothelial dysfunction [15,16], and during experimental pain, when they exhibit a significantly lower endothelial dysfunction than Lows [17]. Theoretically, these differences between Highs and Lows may be due either to sympathetically controlled hemodynamic responses eliciting different post-occlusion shear stress between Highs and Lows and/or to hypnotizability-related thresholds for NO production. The first hypothesis should be rejected, as nociceptive stimulation and cognitive load induce the same reduction of systolic/mean blood pressure in the two groups with respect to basal conditions [28] and, consequently, a similarly reduced shear stress. Thus, a lower threshold for NO production by the Highs vascular endothelium should account for the FMD differences previously observed [15–18]. Endogenous NO is generated from arginine by a family of three distinct calmodulin-dependent NO synthase (NOS) enzymes: NOS from endothelial cells (eNOS) and neurons (nNOS) are constitu-
S. Presciuttini et al. / Neuroscience Letters 467 (2009) 252–255
tively expressed, while expression iNOS (inducible NOS) requires transcriptional activation mediated by citokines. eNOS is mainly responsible for smooth muscle relaxation and nNOS for neuronal signalling; yet, eNOS and nNOS exert a cooperative role in the control of heart contractility [24], and spontaneous myocardial infarction associated with multiple cardiovascular risk factors has been observed in mice only after genetic disruption of the whole NOS system [26]. In addition, NO from eNOS in the nucleus tractus solitarii (NTS) plays a major role in the control of baroreflex gain and arterial pressure through vascular-neuronal signalling [29] and/or the modulation of the interaction between astrocytes and neurons [23]. Endothelial NOS is encoded by the NOS3 locus, which contains 26 exons spanning approximately 21 kilobases. Among the many polymorphisms identified in this gene, two functional single nucleotide polymorphisms (SNPs) have been especially investigated in relation to cardiovascular disease, namely, the promoter −786T>C variant (rs2070744) and the missense mutation 894G>T (Glu298Asp, rs1799983) [10,19,20,24,25]. Aim of the present study was to examine whether individual differences in hypnotizability are associated with single nucleotide polymorphisms (SNPs) of NOS3. As linkage disequilibrium (LD) among SNPs may generate in populations specific haplotypes, which are the actual determinants of individual susceptibility, we investigated a total of nine SNPs of this gene, and determined two-locus haplotype frequencies in the presence of significant LD. We hypothesized that some eNOS haplotypes could be associated with susceptibility to hypnosis. After signing an informed consent approved by the local Ethics committee, 46 healthy volunteers joined the study which followed the rules of the Declaration of Helsinki. They were selected according to their hypnotizability level through the Italian version [11] of the Stanford Hypnotic Susceptibility Scale (SHSS, form C) among 240 students of the Pisa University and divided in 24 Highs (score ≥ 9/12), age (mean ± SD) 24 ± 3.2 years, 15 females, and 22 Lows (score ≤ 3/12), age 25 ± 2.8 years, 16 females. Sixty-one newborns (umbilical cords from the Immunohemathology Unit Bank at the Azienda Ospedaliera–Universitaria Pisana) were included in the study as representative of the general population. Genomic DNA was extracted through a salting out procedure from Highs’ and Lows’ peripheral blood leukocytes. For privacy requirements, blood samples were coded anonymously. The DNA extracted from 500 l of blood was diluted with 100 l H2 O, quantified by UV measurement at OD 260 nm and stored at −20 ◦ C until further processing. Nine known single nucleotide polymorphisms (SNPs) of NOS3 (Table 1) were analyzed by direct sequencing using published primers (http://genecanvas.idf.inserm.fr/). The “rs” numbers of these SNPs were obtained by cross-checking against the National Center of Biotechnology Information (NCBI) database. Calculations and statistical tests were carried out in Excel (Microsoft Corporation, Redwood, WA, USA) using our own routines. Adherence of genotype frequencies to Hardy–Weinberg equilibrium was assessed by goodness-of-fit tests. Heterogeneity of allele frequencies among population samples was assessed
Table 1 “rs” number and nucleotide change of the typed NOS3 variants. Variant
Base change
Function
rs1800783 rs2070744 rs1800781 rs1549758 rs1799983 rs3918174 rs1800780 rs3918188 rs1799985
−1474 T/A −786 T/C +174 G/A 774 C/T 894 G/T +294 A/G +52 G/T +15 G/A +11 G/T
Upstream Promoter Intron 3 Silent Glu298Asp Intron 9 Intron 12 Intron 14 Intron 23
253
by contingency-table 2 analysis. Difference of allele frequency between Highs and Lows was assessed by the Fisher exact test. Maximum likelihood estimates of two-locus gametic frequencies were obtained by the expectation-maximization (EM) algorithm [36]; linkage disequilibrium (LD) between SNP pairs was measured using the absolute value of Lewontin’s D (|D |); its value is included between 0 (the state of gametic equilibrium) and 1 (complete LD); the significance of D was assessed by 2 . Significance level was set at P < 0.05. The genotype distribution of each of the nine SNPs in each of the three samples was tested for HW equilibrium. Only one out of the 27 tests was statistically significant, but significance disappeared after Bonferroni correction. As shown in Table 2, two SNPs located in the upstream region (−1474T/A, rs1800783) and in the promoter (−786T/C, rs2070744), respectively, showed both significant heterogeneity of allele frequency among the three samples (2 = 7.46, P = 0.024 and 2 = 5.87, P = 0.053, respectively) and were specifically associated with high or low hypnotizability (Fisher exact test). The newborn allele frequencies were intermediate between those of Highs and Lows in both cases. Two-locus linkage disequilibrium (LD) analysis was conducted separately in the three samples. For the above two SNPs, LD was non-significant in the sample of newborns (D = 0.177, P = 0.342), whereas it was highly significant in Highs (D = 0.763, P = 0.002) and borderline significant in Lows (D = 0.544, P = 0.070). Table 3 shows the estimated haplotype frequencies in the three samples; the A-C haplotype was 3.4-fold more frequent in Highs than in newborns (0.472 vs. 0.138) and the T-T haplotype was 1.8-fold more frequent in Lows than in newborns (0.577 vs. 0.318). This study reports the first evidence of a genetic substrate for hypnotizability-related vascular responses. We typed nine SNPs of NOS3 in two highly selected samples (the Highs and the Lows, which belongs to the upper and lower deciles of the population, measured on the scale of hypnotizability) as well as in a sample of the general population (represented by umbilical cords of newborns). With regard to hypnotizability, the general population is an admixture of different phenotypes (most individuals score at intermediate values) and is expected to show allele frequencies intermediate between Highs and Lows for those polymorphisms that are truly associated with the trait, i.e. for which Highs and Lows show significant differences. Indeed, this has been the case of the two SNPs −1474 T/A and −786 T/C, whose allele frequencies were significantly different between Highs and Lows and which showed a trend of allele frequency from Lows to newborns to Highs. This result strongly supports the authenticity of the association, in spite of the small sample sizes of Highs and Lows. In addition, the two SNPs are close to each other (no other SNPs have been typed in between), and the fact that they show similar features is a further indication that the association is not due to random effects. Another result of the present work supports the validity of the association. Haplotype analysis showed that the general population was in linkage equilibrium for the two SNPs, whereas both Highs and Lows showed high or intermediate values of linkage disequilibrium; this is expected if a specific haplotype is selected along with an extreme phenotype. Indeed, the A-C haplotype frequency was increased in Highs, whereas the T-T haplotype frequency was increased in Lows. The overall effect appears to be strong, as it has been detected in small samples. In our analysis, no other SNP or two-locus haplotype showed evidence of association with hypnotizability. The allele C of the −786 T/C SNP has been associated, although not unanimously, with reduced NO activity [10,3] and consequent cardiovascular disease, particularly atherosclerosis, coronary artery disease, myocardial infarction, stroke, cerebral and peripheral artery disease, intracranial aneurysm [see 21,22,27]. Our
254
S. Presciuttini et al. / Neuroscience Letters 467 (2009) 252–255
Table 2 Genotype distribution and allele frequency of 9 SNPs of the NOS3 gene. SNP
Genotype
Highs
Lows
Newborn
P (het)
P (Fisher)
rs1800783
TT AT AA T frequency
3 13 8 0.40
10 9 3 0.66
23 26 12 0.59
0.024
0.013
rs2070744
TT CT CC T frequency
6 11 7 0.479
11 10 1 0.727
21 30 10 0.59
0.053
0.02
rs1800781
GG GA AA G frequency
11 5 8 0.563
6 12 4 0.545
22 25 14 0.566
NS
NS
rs1549758
CC CT TT C frequency
14 7 3 0.729
10 11 1 0.705
30 26 5 0.705
NS
NS
rs1799983
GG GT TT G frequency
18 6 0 0.875
16 6 0 0.864
38 20 3 0.787
NS
NS
AA AG GG A frequency
6 11 7 0.479
8 11 3 0.614
21 30 10 0.59
NS
NS
rs1800780
GG GT TT G frequency
20 4 0 0.917
19 3 0 0.932
45 15 1 0.861
NS
NS
rs3918188
GG GA AA G frequency
16 6 2 0.792
14 8 0 0.818
41 17 3 0.811
NS
NS
rs1799985
GG GT TT G frequency
11 12 1 0.708
7 14 1 0.636
26 31 4 0.68
NS
NS
rs3918174
Note. P (het): P-value for heterogeneity of allele frequency among the three samples; P (Fisher): P-value of the exact test between Highs and Lows (2 × 2). NS: P > 0.05.
results show that this allele is part of a haplotype with higher frequency in Highs. Thus, the present results seem to contrast with previous findings on endothelial function suggesting a higher NO activity in Highs during nociceptive stimulation and mental stress on the basis of their scarcely reduced post-occlusion vascular dilation (FMD) [15–17]. As hemodinamic differences induced in Highs and Lows by stimulations have been excluded [28] and, thus, a similarly reduced shear stress occurs in the two groups during stimulations, the evidences concerning FMD and eNOS can be reconciled only by hypothesizing that the higher NO availability accounting for the Highs’ larger FMD is obtained through greater efficacy of transcriptional factors induced by shear stress and/or through lower effects of NOS inhibitory controls affecting eNOS activity and function [9]. Apparently, these modulator factors enable the Highs’ endothelium to provide higher NO availability in spite of their apparently lower NOS3 activity, which is associated with lower
Table 3 Frequency estimates of two SNPs haplotypes at rs1800783 and rs2070744 (−1474 T/A and −786 T/C). Haplotype
Newborns
Highs
Lows
A-C A-T T-C T-T
0.138 0.272 0.272 0.318
0.472 0.132 0.049 0.347
0.191 0.150 0.082 0.577
NO availability in the general population. Thus, the present findings suggest that high hypnotizability might be a protective factor against the NO deficiency possibly occurring in subjects showing this NOS3 variant [21,22,27]. The lack of FMD evaluation in the same subjects undergoing the present NOS investigation might be considered a limitation of the study. However, FMD results assessing a lower FMD reduction in Highs during cognitive and physical stimulation had been previously replicated in four separate groups of subjects [15–18]. Altogether, our results [15–18] suggest that being a High might be important for chronic pain patients for more than the wellassessed control of pain experience [12]. Indeed, a larger availability of NO is a relevant prognostic factor in these patients as they exhibit 13.6% of coronary disease as compared to 6.5% in matched subjects not suffering from chronic pain [2]. A higher endothelial derived NO activity in the NTS of chronic pain patients with high susceptibility to hypnosis might also reduce the sympathetic overactivity or parasympathetic underactivity reported in chronic pain patients [see 7] and contributing to pain chronicization [20]. In addition, the sympathetically mediated control of immune system [19] may be differentially modulated in Highs and Lows and be responsible for a lower frequency of occurrence of diseases due to immune system deficiency in Highs [see 7]. Finally, we would like to emphasize that the endothelial derived NO plays a role not only in the peripheral and brain vascular diameter control, but also in the modulation of neuronal activity due
S. Presciuttini et al. / Neuroscience Letters 467 (2009) 252–255
to its diffusion from brain vessels into the extracellular compartment; thus, it may be involved also in other hypnotizability-related differences. Indeed, it has been shown that nitric oxide inhalation increases imaginative ability [37] that is an important component of hypnotic responding [14,38]. A larger availability of NO might induce a greater general excitability of the cerebral cortex and account also for hypnotizability-related differences observed in the cognitive and sensori-motor domain, i.e. the equivalence of imagery and perception occurring more often and more effectively in Highs than in Lows [4,5,33], the Highs’ ability to imagine through both the visual and tactile/proprioceptive modality [5] and the peculiar postural and locomotor control observed in Highs [25,32]. In conclusion, the NO availability, which depends on individual genotype in stimulation conditions, could have a causal role in hypnotic responding. This “NO hypothesis” might be a new frontier in the physiology of hypnotic susceptibility, but it should be further investigated in its interactions with other factors, i.e. gender, personality traits and social/cultural level, in large samples. Acknowledgments Research funded by Azienda Ospedaliera-Universitaria Pisana. Dr. I. Bianco is sincerely acknowledged for valuable technical collaboration. References [1] G.C. Balthazard, E.Z.E.Z. Woody, Bimodality, dimensionality, and the notion of hypnotic types, Int. J. Clin. Exp. Hypn. 37 (1989) 70–89. [2] A Berger, E.M. Dukes, G. Oster, Clinical characteristics and economic costs of patients with painful neuropathic disorders, J. Pain 5 (2004) 143–149. [3] P.F. Binkley, E. Nunziatta, Y. Liu-Stratton, G.E. Cooke, A polymorphism of the endothelial nitric oxide synthase promoter is associated with an increase in autonomic imbalance in patients with congestive heart failure, Am. Heart J. 149 (2005) 342–348. [4] G. Carli, C. Rendo, L. Sebastiani, E.L. Santarcangelo, Suggestions of altered balance: possible equivalence of imagery and perception, Int. J. Clin. Exp. Hypn. 54 (2006) 206–223. [5] G. Carli, F.I. Cavallaro, C.A. Rendo, E.L. Santarcangelo, Imagery of different sensory modalities: hypnotizability and body sway, Exp. Brain Res. 179 (2007) 147–154. [6] G. Carli, D. Manzoni, E.L. Santarcangelo, Hypnotizability-related integration of perception and action, Cogn. Neuropsychol. 25 (2008) 1065–1076. [7] G. Carli, A. Huber, E.L. Santarcangelo, Hypnotizability and chronic pain: an ambiguous connection, Contemp. Hypn. 25 (2008) 65–77. [8] C. Carvalho, I. Kirsch, G. Mazzoni, I. Leal, Portuguese norms for the Waterloo–Stanford Group C (WSGC) scale of hypnotic susceptibility, Int. J. Clin. Exp. Hypn. 56 (2008) 295–305. [9] A. Chatterjee, J.D. Catravas, Endothelial nitric oxide (NO) and its pathophysiologic regulation, Vasc. Pharmacol. 49 (2008) 134–140. [10] G.E. Cooke, A. Doshi, P.F. Binkley, Endothelial nitric oxide synthase gene: prospects for treatment of heart disease, Pharmcogenomics 8 (2007) 1723–1734. [11] V. De Pascalis, A. Bellusci, P.M. Russo, Italian norms for the Stanford Hypnotic susceptibility scale, form C, Int. J. Clin. Exp. Hypn. 48 (2000) 15–23. [12] S.W.G Derbyshire, M.G. Whalley, D.A. Oakley, Fibromyalgia pain and its modulation by hypnotic and non-hypnotic suggestion: an fMRI analysis, Eur. J. Pain 13 (2009) 542–550. [13] L. Ghiadoni, A.E. Donald, M. Cropley, M.J. Mullen, G. Oakley, M. Taylor, G. O’Connor, J. Betteridge, N. Klein, A. Steptoe, J.E. Deanfield, Mental stress induces transient endothelial dysfunction in humans, Circulation 102 (2000) 2473–2478. [14] J.P. Green, A.F. Barabasz, D. Barrett, G.H. Montgomery, Forging ahead: the 2003 APA Division 30 definition of hypnosis, Int. J. Clin. Exp. Hypn. 53 (2005) 259–264.
255
[15] Z. Jambrik, E.L. Santarcangelo, B. Ghelarducci, E. Picano, L. Sebastiani, Does hypnotisability modulate the stress-related endothelial dysfunction? Brain Res. Bull. 63 (2004) 213–221. [16] Z. Jambrik, L. Chunzeng, E.L. Santarcangelo, L. Sebastiani, B. Ghelarducci, E. Picano, Traditional acupuncture does not modulate the endothelial dysfunction induced by mental stress, Int. J. Cardiovasc. Imaging 20 (2004) 357–362. [17] Z. Jambrik, G.G. Carli, T. Rudish, A. Varga, T. Forster, E.L. Santarcangelo, Modulation of pain-induced endothelial dysfunction by hypnotisability, Pain 116 (2005) 181–186. [18] Z. Jambrik, L. Sebastiani, E. Picano, B. Ghelarducci, E.L. Santarcangelo, Hypnotic modulation of flow-mediated endothelial response to mental stress, Int. J. Psychophysiol. 55 (2005) 221–227. [19] W. Jänig, Pain in the sympathetic nervous system: pathophysiological mechanisms, in: C.J. Mathias, R. Bannister (Eds.), Autonomic Failure, 5th edn, New York Oxford University Press, Oxford, 2008, pp. 175–189. [20] W. Jänig, J.D. Levine, Autonomic-neuroendocrine-immune responses in acute and chronic pain, in: S.B. McMahon, M. Koltzenburg (Eds.), Wall & Mezack’s Textbook of Pain, 5th edn, Elsevier Churchill Livingstone, Edinburgh, 2006, pp. 205–218. [21] P.C. Jaramillo, C. Lanas, F. Lanas, L.A. Salazar, −786T>C polymorphism of the endothelial nitric oxide synthase gene in Chilean subjects with coronary artery disease and controls, Clin. Chim. Acta 387 (2007) 105–108. [22] I.J. Kullo, M. Todd Greene, E. Boerwinkle, J. Chu, S.T. Turner, S.L.R. Kardia, Association of polymorphisms in NOS3 with the ankle-brachial index in hypertensive adults, Aterosclerosis 196 (2008) 905–912. [23] L.H. Lin, O. Taktakishvili, W.T. Talman, Identification and localization of cell types that express endothelial and neuronal nitric oxide synthase in the rat nucleus tractus solitarii, Brain Res. 171 (2007) 42–51. [24] X. Loyer, C. Heymes, J.L. Samuel, Constitutive nitric oxide synthases in the heart from hypertrophy to failure, Clin. Exp. Pharmacol. Physiol. 35 (2008) 483–488. [25] M. Menzocchi, G. Paoletti, A. Huber, G. Carli, F.I. Cavallaro, D. Manzoni, E.L. Santarcangelo, Hypnotizability and sensorimotor integration: an Italian Space Agency Project, Int. J. Clin. Exp. Hypn., in press. [26] S. Nakata, M. Tsutsui, H. Shimokawa, O. Suda, T. Morishita, K. Shibata, Y. Yatera, K. Sabanai, A. Tanimoto, M. Nagasaki, H. Tasaki, Y. Sasaguri, Y. Nakashima, Y. Otsuji, N. Yanagihara, Spontaneous myocardial infarction in mice lacking all nitric oxide synthase isoforms, Circulation 117 (2008) 2211–2223. [27] A. Neyatizadeh, R. Kumar, T. Stobdan, A.K. Goyal, S. Sikdar, M. Gupta, S. Javed, M.A. Qadar Pasha, Endothelial nitric oxide synthase gene haplotypes and circulating nitric oxide levels significantly associate with risk of essential hypertension, Free Radic. Biol. Med. 44 (2008) 1912–1918. [28] G. Paoletti, M. Varanini, G. Carli, R. Balocchi, C. Morizzo, C. Palombo, E.L. Santarcangelo, Does the autonomic system contribute to pain-related endothelial dysfunction? Acta Physiol. 197 (Suppl. 672) (2009) 2. [29] J.F. Paton, H. Waki, A.P. Abdala, J. Dickinson, S. Kasparov, Vascular-brain signaling in hypertension: role of angiotensin II and nitric oxide, Curr. Hypertens. Rep. 9 (2007) 242–247. [30] N. Resnick, H. Yahav, A. Shay-Salit, M. Shushy, S. Schubert, L.C.M. Zilberman, E. Wofovitz, Fluid shear stress and the vascular endothelium: for better and for worse, Prog. Biophys. Mol. Biol. 81 (2003) 177–199. [31] E.L. Santarcangelo, R. Balocchi, E. Scattina, D. Manzoni, L. Bruschini, B. Ghelarducci, M. Varanini, Hypnotizability-dependent modulation of the changes in heart rate control induced by upright stance, Brain Res. Bull. 75 (2008) 692–697. [32] E.L. Santarcangelo, E. Scattina, G. Carli, A. Macerata, D. Manzoni, Hypnotizability-dependent modulation of postural control: effects of alteration of the visual and leg proprioceptive inputs, Exp. Brain Res. 191 (2008) 331–340. [33] E. Scattina, E.L. Santarcangelo, G. Carli, B. Ghelarducci, D. Manzoni, Modulation of vestibular reflex by imagery, Acta Physiol. 197 (Suppl. 672) (2009) 3. [34] M. Shechter, A. Issachar, I. Marai, N. Koren-Morag, D. Freinark, Y. Shahar, A. Shechter, M.S. Feinberg, Long-term association of brachial artery flowmediated vasodilation and cardiovascular events in middle-aged subjects with no apparent heart disease, Int. J. Cardiol. 134 (2009) 52–58. [35] L.E. Spieker, D. Hürlimann, F. Rischitzka, R. Corti, F. Enseleit, S. Shaw, D. Hayoz, J.E. Deanfield, T.F. Lüscher, G. Noll, Mental stress induces prolonged endothelial dysfunction via endothelin-A receptors, Circulation 105 (2002) 2817–2820. [36] J. Terwillinger, J. Ott, Handbook of human genetic linkage, The Johns Hopkins University Press, Baltimore, London, 1994. [37] M.G. Whalley, G.B. Brooks, Enhancement of suggestibility and imaginative ability with nitrous oxide, Psychopharmacology 203 (2009) 745–752. [38] E.Z. Woody, A.J. Barnier, K.M. McConkey, Multiple hypnotizabilities: differentiating the building blocks of hypnotic response, Psychol. Ass. 17 (2005) 200–211.