The Val66Met Polymorphism of the BDNF Gene Influences Trigeminal Pain-Related Evoked Responses

The Journal of Pain, Vol 13, No 9 (September), 2012: pp 866-873 Available online at www.jpain.org and www.sciencedirect.com

The Val66Met Polymorphism of the BDNF Gene Influences Trigeminal Pain-Related Evoked Responses Cherubino Di Lorenzo,* Giorgio Di Lorenzo,y,z Andrea Daverio,y,z Patrizio Pasqualetti,x,{ Gianluca Coppola,# Ioannis Giannoudas,y,z Ylenia Barone,z,** Gaetano S. Grieco,yy Cinzia Niolu,z,** Esterina Pascale,zz Filippo M. Santorelli,xx Ferdinando Nicoletti,{{,## Francesco Pierelli,{{ Alberto Siracusano,y,z,** and Stefano Seri*** *Don Carlo Gnocchi Onlus Foundation, Milan, Italy. y Laboratory of Psychophysiology, zChair of Psychiatry, Department of Systems Medicine, University of Rome ‘‘Tor Vergata,’’ Rome, Italy. x Service of Medical Statistics and Information Technology, Fatebenefratelli Association for Research (AFaR), Rome, Italy. { Epidemiology and Biostatistics, Casa di cura San Raffaele-Cassino, Cassino, Italy. # G.B. Bietti Foundation-Instituto Di Ricovero e Cura a Carattere Scientifico, Department of Neurophysiology of Vision and Neurophthalmology, Rome, Italy. **Psychiatric Clinic, Fondazione Policlinico ‘‘Tor Vergata,’’ Rome, Italy. yy Laboratory of Experimental Neurobiology, C. Mondino National Institute of Neurology Foundation, Instituto Di Ricovero e Cura a Carattere Scientifico, Pavia, Italy. zz Department of Medico-Surgical Sciences and Biotechnologies, Sapienza University of Rome, Rome, Italy. xx Molecular Medicine & Neurodegenerative Diseases, Instituto Di Ricovero e Cura a Carattere Scientifico Fondazione Stella Maris, Pisa, Italy. {{ Instituto Di Ricovero e Cura a Carattere Scientifico—Neuromed, Pozzilli (IS), Italy. ## Department of Pharmacology, Sapienza University of Rome, Rome, Italy. ***School of Life and Health Sciences, Aston Brain Centre, Aston University, Birmingham, United Kingdom.

Abstract: Cortical pain processing is associated with large-scale changes in neuronal connectivity, resulting from neural plasticity phenomena of which brain-derived neurotrophic factor (BDNF) is a central driver. The common single nucleotide polymorphism Val66Met is associated with reduced BDNF activity. Using the trigeminal pain-related evoked potential (tPREP) to repeated electrical painful stimuli, we investigated whether the methionine substitution at codon 66 of the BDNF gene was associated with changes in cortical processing of noxious stimuli. Fifty healthy volunteers were genotyped: 30 were Val/Val and 20 were Met-carriers. tPREPs to 30 stimuli of the right supraorbital nerve using a concentric electrode were recorded. The N2 and P2 component latencies and the N2-P2 amplitude were measured over the 30 stimuli and separately, by dividing the measurements in 3 consecutive blocks of 10 stimuli. The average response to the 30 stimuli did not differ in latency or amplitude between the 2 genotypes. There was a decrease in the N2-P2 amplitude between first and third block in the Val/Val group but not in Met-carriers. BDNF Val66Met is associated with reduced decremental response to repeated electrical stimuli, possibly as a result of ineffective mechanisms of synaptic memory and brain plasticity associated with the polymorphism. Perspective: BDNF Val66Met polymorphism affects the tPREP N2-P2 amplitude decrement and influences cortical pain processing through neurotrophin-induced neural plasticity, or through a direct BDNF neurotransmitter-like effect. Our findings suggest that upcoming BDNF central agonists might in the future play a role in pain management. ª 2012 by the American Pain Society Key words: BDNF, Val66Met, single nucleotide polymorphism, pain-related evoked potential, neural plasticity. Received May 6, 2012; Accepted May 30, 2012. Supported by grant number 08RC02 of Don Gnocchi Onlus Foundation (Ricerca Corrente 2011). Pain research at the Aston Brain Centre is supported by the Dr. Hadwen and the Lord Dowding Trusts for Humane Research and by the Wellcome Trust. Authors disclose no conflict of interest.

866

Address reprint requests to Cherubino Di Lorenzo, Via Maresciallo Caviglia, 30 – 00135 – Rome, IT. E-mail: [email protected] 1526-5900/$36.00 ª 2012 by the American Pain Society http://dx.doi.org/10.1016/j.jpain.2012.05.014

Di Lorenzo et al

P

ain is defined as ‘‘an unpleasant sensory and emotional experience associated with actual or potential tissue damage,’’16 suggesting its multidimensional nature and an equal relevance of patients’ ache-dependent sensory perception and of its emotional correlates. Multiple mechanisms contribute to the pathophysiology of pain, each of which is subject to or an expression of neural plasticity,19 ie, the capacity of neurons to change their function, chemical profile, or structure. Repeated noxious stimuli induce plastic modifications of functional connectivity in specific brain networks in the resting state.30 Brain-derived neurotrophic factor (BDNF), a potent modulator of brain plasticity released as a consequence of nociceptors activation, plays both a signaling and a plastic role in pain-related sensory and emotional pathways.29 In recent years, genetic approaches have been proposed to obtain new insights into the pathophysiology of pain,11 and a specific role in modulation of the pain matrix has been proposed for BDNF. A genetic variation within the BDNF gene resulting in valine to methionine substitution at codon 66 (Val66Met) is associated with reduced secretion of the BDNF protein and impaired BDNF signaling. The BDNF Val66Met polymorphism has been implicated in a number of neuropsychiatric conditions due to its role in shaping the anatomy of the central nervous system, as well as its influence on behavior and cognition in humans.2 Time-locked brain responses to noxious electrical stimuli, of which the trigeminal electric pain-related evoked potential (tPREP) is an established paradigm,17 can assess noninvasively the neural response to painful stimuli with high temporal resolution. Noxious electrical stimulation induces plastic modifications in human pain processing31 in areas involved in sensory, affective, cognitive, and attentional processing, interpreted as manifestation of long-term depression.32 Furthermore, when the noxious stimuli are applied repeatedly at constant intensity, a decrement of the amplitude of the brain response over time is observed.25 Recently, the role of BDNF Val66Met in cortical pain processing has been investigated in healthy individuals and in patients with chronic pain; in healthy individuals, polymorphisms of the BDNF gene did not result in differences in the amplitude of the pain-related brain response obtained by subcutaneous electrical stimulation.38 Our aims were to verify whether healthy individuals with methionine substitution at codon 66 of the BDNF gene presented differences in the amplitude of the tPREP and, more importantly, lower amplitude reduction of the response following repeated stimuli compared to those with the Val/Val polymorphism.

Methods Subjects Individuals without personal and familial history of headache or any other chronic pain condition and of neuropsychiatric disorders were considered eligible for this study. None of the participants had used any

The Journal of Pain

867

central nervous system-acting drugs in the last 6 months or analgesic in the 72 hours prior to the experimental procedure. Subjects were asked not to drink coffee, tea, or any other beverage containing stimulants and to refrain from smoking cigarettes in the 2 hours before the beginning of the recording session. Information on the quality of sleep during the night prior to the recording was collected and the experiment was postponed if the subject reported a nonrestoring sleep. Women were recorded during the follicular phase of the menstrual cycle. Written informed consent was obtained from all participants prior to entering the study, which received approval by the local ethics committee.

Genotyping We obtained peripheral blood from all subjects and genomic DNA was extracted using standard procedures. Subjects were genotyped by RFLP analysis, as previously described.13 We used the following primers: SBDNF1 – AAA GAA GCA AAC ATC CGA GGA CAA G; SBDNF2 – ATT CCT CCA GCA GAA AGA GAA GAG G resulting in a 274 base pair (bp) polymerase chain reaction product. In the presence of the G allele, HIN1 II digestion produced 2 products, 57 and 217 bp; whereas the A allele produced 3 products, 57, 77, and 140 bp. Polymerase chain reaction products were electrophoresed on a 2% agarose gel and visualized using ethidium bromide. According to their genotype, patients were classified as Val/Val if homozygous for valine or as Met-carriers (and classified as Val/Met if heterozygous and Met/Met if homozygous for methionine). Met-carriers were grouped together for statistical analysis due to the rarity of homozygotes and in line with methods from independent groups in distinct research areas.3,8,12,35,38 To ensure that researchers were blinded at the time of neurophysiologic data analysis, results of genetic analysis was not made available prior to the analysis of neurophysiological data.

Experimental Setting Stimulation Noxious stimuli were applied on the forehead (approximately 1 cm above the supraorbital foramen) to stimulate the supraorbital nerve—a terminal division of the trigeminal ophthalmic branch—using a custom-built planar concentric electrode (Bionen, Florence, Italy). The physical characteristics of the stimuli were consistent with those reported by Kaube et al18 and were reported in a previous study.34 The duration of each stimulus was 100 ms. Individual thresholds for sensory perception (ST) and pain perception (PT) were defined as the minimum intensity of the electrical stimulus perceived as tactile and painful, respectively, determined by a sequence of 3 descending and ascending current intensities. The stimulation intensity was set at 1.5-fold of the individual PT. The recording session consisted of 31 stimuli of the same intensity with an interstimulus interval ranging between 14 and 16 seconds.

868

The Journal of Pain

BDNF Gene Polymorphism Influences Trigeminal Cortical Pain Modulation

Electrophysiological Recordings

Sample Size Calculation

Subjects were seated in a reclining chair with a mounted headrest throughout the experiment. The recordings were performed using Ag/AgCl disk electrodes (Bionen, Florence, Italy). According to published methodology the active electrode was attached to the scalp on the Cz location of the International 10–20 System and referred to linked earlobes (A1 and A2).17 A cephalic ground electrode was placed on Fpz. One polygraphic channel was used to monitor vertical and horizontal eye movements with the active electrode placed 1 cm lateral to the outer canthus of the right eye and the reference electrode placed on lower eyelid. Electrode impedances were maintained below 5 KU. The signal was amplified by a digital EEG system (Galileo MIZAR-Sirius, EBNeuro, Florence, Italy) with a sampling rate of 2048 Hz and 24-bit resolution and stored on a hard disk with a digital trigger from the built-in electrical stimulator.

An expert not involved in the neurophysiological or genetic data collection (P.P.) performed sample size calculation and statistical analysis (see next section). Because our primary endpoint was to detect differences in the slope of N2-P2 amplitude between Val/Val and Met-carriers, a sample size calculation was based on pilot data from 20 subjects enrolled independently from the current study. Twelve were Val/Val and 8 were Val/Met, with a slope index respectively 3.13 6 3.27 and .93 6 5.47. Assuming that values in each subject group were normally distributed with within-group standard deviation of 4.29, to fulfill a desired power of 90% with significance level at 5%, the required sample size resulted in 50 subjects: 30 Val/Val and 20 Met-carriers.

Data Analysis One of the investigators (A.D.) blinded to the results of the genetic study performed the analysis off-line. Digital data were exported to the European Data Format using NPX Lab 2010 (available at www. brainterface.com) and analyzed in the EEGLAB environment (http://www.sccn.ucsd.edu/eeglab/index.html) under Matlab 7.7.0 R2010a (Mathworks Inc, Natick, MA). EEG signal was digitally band-pass filtered between 1 and 100 Hz and segmented into 700-ms epochs (100 ms pre- to 600 ms post-stimulus). The first epoch of each subject’s recording was rejected to avoid possible contamination by a startle response. The remaining 30 epochs were averaged in 3 sequential blocks of 10 responses and analyzed separately. We performed artifact rejection using the EEGLAB semiautomatic procedure.7 The EOG channel was used to further verify the efficacy of the rejection method in the identification of ocular artifacts. The N2 and P2 components of the averaged evoked response were identified as the most negative and the most positive peak, respectively, occurring between 90 and 260 ms after the stimulus. We measured the latency of the N2 and P2 components, as well as the N2-P2 interpeak amplitude for each block. The N2-P2 brain response encodes cognitive and evaluative aspects of stimulus processing, represents an indirect readout of central nociceptive networks,1,9 and can be therefore considered a reliable neurophysiological correlate of the synaptic plasticity attributed to BDNF activity. We then computed the slope of the linear regression line (as expressed by the formula: y = mx 1 b) of the N2-P2 amplitudes over the 3 blocks in each subject to investigate changes of the evoked response amplitude over time. This derived measure of the evoked potential amplitude has the advantage of being relatively insensitive to interindividual variability. A negative slope of the regression line was indicative of a decremental response, whereas a positive slope was an index of an incremental response.

Statistical Analysis Because the indices showed a nonnormal distribution, we logarithmically transformed all the variables to obtain a better approximation to a Gaussian curve, achieving an appropriate equivalence to a normal distribution (Kolmogorov-Smirnov test, P > .2). Analysis of variance (ANOVA) models were used to measure difference among groups and between blocks and groups. The differences in age, ST, PT, and grand averages of the N2 and P2 latencies and of the N2-P2 amplitude of the 3 blocks as well as in the slope index were analyzed using a 2-way ANOVA (between-subject factors: ‘‘gender,’’ women versus men; ‘‘genotype,’’ Val/Val versus Met-carriers). Gender was used as a between-subject factor due to evidence of gender differences in pain perception, sensitivity, and tolerance.26 Repeated measures ANOVA (rm-ANOVA) followed by univariate ANOVAs were performed on N2-P2 measure to investigate the group effects (between-subject factors: ‘‘gender’’ and ‘‘genotype’’) on the 3 blocks (within-subjects factor: block 1 to block 3). Univariate results were analyzed only if Wilks’ Lambda multivariate significance criterion was achieved. The sphericity of the covariance matrix was tested with Mauchly sphericity test; in the case of violation of the sphericity assumption, Greenhouse-Geisser epsilon (ε) adjustment was used. To analyze changes between blocks, polynomial contrasts were constructed. Because data were divided in 3 blocks, only linear (c1 = .707 *block1.707 * block3, hereafter called ‘‘slope’’) and quadratic (c2 = .408 * block1.816 * block2 1.408 * block3) contrasts could be defined and assessed for the main term (block effect in the whole sample) and for the ‘‘block’’  ‘‘genotype’’ interaction (differences of block effects between Val/Val and Met-carriers). Statistical significance was set at P < .05.

Results Fifty participants (24 women, 26 men; age, mean 6 SD: 29.67 6 3.28 and 28.96 6 3.50, respectively) were enrolled in the study. Thirty were genotyped as Val/Val, 18 as Val/Met, and 2 as Met/Met, in line with the Hardy-Weinberg equilibrium (P = .73). The 2 subjects

Di Lorenzo et al

The Journal of Pain

Descriptive Statistics of Gender, Age, ST, and PT in Val/Val and Met-Carriers

Table 1.

Gender* Women Men Agey STz PTx

VAL/VAL

MET-CARRIERS

14 16 29.73 6 2.94 1.60 6 .62 8.87 6 2.29

10 10 28.65 6 3.95 1.65 6 .75 10.05 6 2.89

NOTE. Data are frequencies and mean 6 SD. *Yates chi-square test: c2 = .003, df = 1, P = .95. yGenotype, F1,46 = 1.24, P = .27; gender, F1,46 = .66, P = .42; genotype  gender, F1,46 = .09, P = .76. zGenotype, F1,46 = .06, P = .81; gender, F1,46 = .21, P = .65; genotype  gender, F1,46 = .003, P = .96. xGenotype, F1,46 = 2.71, P = .11; gender, F1,46 = 2.38, P = .13; genotype  gender, F1,46 = .62, P = .44.

genotyped as Met/Met did not differ from Val/Met subjects in terms of behavior of response to repeated painful stimulation. On this basis, we collated the 2 genotypes in a single group. Val/Val and Met-carriers did not significantly differ in gender, age, ST, and PT (Table 1). The number of remaining epochs (mean 6 SD) after rejection of those contaminated by artifacts for the 3 blocks was 8.98 6 .65, 9.04 6 .70, and 9.1 6 .71, respectively. An rm-ANOVA model showed no effect of ‘‘block’’ (Wilks’ Lambda = .98, F2,47 = .61, P = .55) and of ‘‘block’’  ‘‘genotype’’ interactions (Wilks’ Lambda = .99, F2,47 = .11, P = .90) on the number of epochs left for the analysis. Two-way ANOVA showed no significant effect of ‘‘genotype’’ (F1,46 = .92, P = .34), ‘‘gender’’ (F1,46 = 3.24, P = .08), and ‘‘genotype’’  ‘‘gender’’ interaction (F1,46 = .88, P = .35) on the N2-P2 amplitude of the grand average of the 3 blocks. No significant differences were seen for the N2 (‘‘genotype’’: F1,46 = 2.52, P = .12; ‘‘gender’’: F1,46 = 1.25, P = .27; ‘‘genotype’’  ‘‘gender’’: F1,46 = .67, P = .42) and P2 latency (‘‘genotype’’: F1,46 = 2.36, P = .13; ‘‘gender’’: F1,46 = 2.99, P = .09; ‘‘genotype’’  ‘‘gender’’: F1,46 = 2.01, P = .16). Rm-ANOVA models for N2-P2 amplitude revealed statistically significant ‘‘block’’ (Wilks’ Lambda = .81, F2,45 = 5.13, P < .01) and ‘‘block’’  ‘‘genotype’’ effects (Wilks’ Lambda = .83, F2,45 = 4.66, P = .01) (Fig 1); ‘‘block’’  ‘‘gender’’ and ‘‘block’’  ‘‘genotype’’  ‘‘gender’’ effects were not significant. After checking that the sphericity assumption was not violated (Mauchly’s W = .94, c2 = 2.67, df = 2, P = .26), univariate tests were performed and showed significant ‘‘block’’ (F2,92 = 6.4, P < .002) and ‘‘block’’  ‘‘genotype’’ effects (F2,92 = 5.91, P < .004) (Fig 2 and Table 2). The effect size, as measured by f statistic, was .34 for ‘‘block’’ main effect and .33 for ‘‘block’’  ‘‘genotype’’ interaction term, with values within Cohen’s conventional values for medium (.25) and large (.40) effects. Polynomial contrasts of the 3 blocks allowed assessment of linear and quadratic components of ‘‘block’’ main effect and of ‘‘block’’  ‘‘genotype’’ interaction term. In both cases, the P values for the quadratic component were not significant (F1,46 = .19, P = .67 and F1,46 = .07, P = .8, respectively), while the linear value (slope) re-

869

sult was significant for the main term (F1,46 = 10.76, P = .002) and for the interaction (F1,46 = 6.18, P = .017). These findings indicate that the linear decrease observed in the whole sample was significantly different in Val/Val and Met-carriers. Furthermore, the slope was significantly different from 0 only in Val/Val (3.79, 95% confidence interval [5.41, 2.17]) and not in Met-carriers (.54, 95% confidence interval [2.52, 1.45]) (Fig 3). The difference between slope index means, indicated by the above significant interaction (P = .02), was confirmed when tested by means of Welch’s robust test for equality of means (P = .03). This procedure was applied to take into account the variance heterogeneity of the 2 groups. As shown in Fig 3, the larger heterogeneity in Metcarriers was related to the coexistence of subjects with clear N2-P2 decrease (similar to Val/Val subjects) and of subjects with null or even positive slope. Finally, the significance of the ‘‘block’’  ‘‘genotype’’ interaction and of its linear component was reliably confirmed even after entering as covariate the N2-P2 amplitude of the first block. The N2 and P2 latency did not change significantly over the 3 blocks (rm-ANOVA model results: ‘‘block,’’ P = .87; ‘‘block’’  ‘‘genotype,’’ P = .93; ‘‘block’’  ‘‘gender,’’ P = .33; ‘‘block’’  ‘‘genotype’’  ‘‘gender,’’ P = .54; ‘‘block,’’ P = .88; ‘‘block’’  ‘‘genotype,’’ P = .85; ‘‘block’’  ‘‘gender,’’ P = .06; ‘‘block’’  ‘‘genotype’’  ‘‘gender,’’ P = .46).

Discussion The amplitude of evoked potentials is a reliable quantitative measure of the extent of neural activation in response to sensory stimuli. In particular, a relationship between the amplitude of the N2-P2 responses and subjective rating of pain intensity has been documented.15 Consistent with data on healthy individuals from a previous study,38 we did not find significant differences between the 2 genotypes in the latency of the N2 and P2 components and in the amplitude of the N2-P2 interpeak amplitude of the grand average of the 3 blocks. However, the most interesting aspect of our study is the time course of brain response to repeated painful stimuli. When analyzing the sample as a whole, we found a reduction of the N2-P2 amplitude to repeated stimuli. This phenomenon, which has been well characterized in response to CO2 laser stimulation,37 is confirmed in our study using electrical stimulation. Owing to the complexity of levels and modes of action of BDNF in the nervous system, a single and unifying explanation of our findings is too ambitious, and only plausible hypotheses are possible at this stage. In agreement with recent evidence,31 the response decrement could be explained as a manifestation of neural plasticity mediated by pre- or post-synaptic modulation and interpreted as a form of synaptic learning and memory aimed at protecting the brain from painful overstimulation and synaptic exhaustion.33 When we analyze brain responses from the 2 groups separately, Met-carriers present lesser amplitude decrement of the evoked

870

The Journal of Pain

BDNF Gene Polymorphism Influences Trigeminal Cortical Pain Modulation

Figure 1. Grand average tPREP of the whole sample (gray line), Val/Val (black solid line), and Met-carriers (black dashed line) in the 3 blocks.

response over time than Val-Val individuals (Fig 2). Owing to the lack of data in humans, we can only speculate on the biological significance of our findings by inferring from animal studies, with the inherent limitations of such extrapolation. Although the presence of the Metallele does not alter BDNF protein function per se, it has been associated with abnormalities in intracellular trafficking and regulated secretion of BDNF,5,10 as well as with impaired synaptic transmission and spike timing-dependent plasticity.27 Evidence of the effect of the Met-allele on BDNF function in humans is only

indirect; MR spectroscopy studies in Met-carriers have reported reduced N-acetyl-aspartate levels in the hippocampus,36 strongly correlated with glutamate concentration.28 BDNF is known to regulate glutamate release and receptor function through TrkB receptors, and its compromised function results in impaired synaptic plasticity of glutamatergic synapses.4 Our findings support the hypothesis that BDNF Val66Met is associated with an atypical cortical response to pain in response to repeated painful stimuli. We hypothesize that this is a result of a cascade of plastic events

Di Lorenzo et al

The Journal of Pain

871

Figure 2. Line plot of N2-P2 amplitudes of each participant for the 3 blocks in Val-Val (left panel) and Met-carriers (right panel). Line in bold represents the mean amplitude values for each block.

resulting in the abnormal activation of anti-nociceptive systems. Following repetitive trigeminal painful stimuli, BDNF production associated with the wild-type form favors cortical plasticity of the anti-nociceptive system and results in a progressive reduction in N2-P2 amplitude across the blocks. Because BDNF is considered a central driver of neural plasticity and is involved in different mechanisms of learning and memory,6 its impaired function (due to the polymorphic presence of Met allele) could have a negative influence on plastic modifications of the nervous system; this might be one of the biological underpinnings of the difference in the decremental response to painful repeated stimulation in the 2 genetic groups. Table 2.

An alternative explanation of the progressive reduction of evoked potentials amplitude to repeated painful stimuli might involve brainstem monoaminergic nuclei. These structures are central effectors of the endogenous anti-nociceptive system and play a crucial role in the central processing of sensory stimuli by controlling the signal-to-noise ratio in cortical and thalamo-cortical neurons.24 BDNF has shown to promote serotoninergic axons22 and modulate serotoninergic activity in the brainstem, inducing analgesia and pain control.23 Alternatively to or jointly with the induction of neural plasticity, this serotoninergic modulation might account for the role of BDNF in processing repeated painful stimuli. BDNF acts as pain modulator

Indices of tPREP in Whole Sample, Val/Val, and Met-Carriers

N2 latency (ms) 1st block 2nd block 3rd block Grand average P2 latency (ms) 1st block 2nd block 3rd block Grand average N2-P2 amplitude (V) 1st block 2nd block 3rd block Grand average Slope of N2-P2 blocks

WHOLE SAMPLE

VAL/VAL

MET-CARRIERS

129.85 6 10.77 130.18 6 11.07 130.82 6 13.45 130.28 6 10.42

128.29 6 11.82 128.24 6 11.36 129.18 6 14.10 128.57 6 10.87

132.20 6 8.72 133.08 6 10.19 133.28 6 12.34 132.85 6 9.37

208.61 6 25.95 208.26 6 29.09 208.22 6 30.55 208.37 6 27.10

213.80 6 26.88 213.12 6 32.36 213.90 6 34.68 213.61 6 29.90

200.83 6 22.97 200.98 6 22.15 199.71 6 21.06 200.50 6 20.53

32.92 6 9.38* 28.57 6 8.99* 25.34 6 8.99* 28.94 6 8.43 3.79 6 2.91

33.66 6 15.83 32.97 6 13.46 32.59 6 12.55 33.07 6 12.84 .54 6 6.01

33.22 6 12.23* 30.33 6 11.09* 28.24 6 11.04* 30.60 6 10.49 2.49 6 4.65

NOTE. Data, shown as mean 6 SD, are raw values. *Indicates significant difference among blocks (see Results section for details).

872

The Journal of Pain

BDNF Gene Polymorphism Influences Trigeminal Cortical Pain Modulation

Figure 3. Slope values of the N2-P2 amplitude of the tPREP in Val/Val (left panel) and Met-carriers (right panel). Each circle represents 1 participant. In the right panel, black circles identify Met/Met subjects. Mean and dispersion around the mean is rendered with the middle line indicating the mean value and the vertical bar the 95% Confidence Interval.

at multiple levels. The activation of central terminals of nociceptors induces BDNF release, further upregulated by peripheral nerve inflammation, generating hyperalgesia and allodynia. Moreover, BDNF action on postsynaptic receptors is known to induce intracellular signaling cascades leading to central sensitization, enhancing the central neurons responsiveness. This activity suggests that BDNF may be considered as a positive neuromodulator in pain transduction as well.29 The infusion of BDNF close to periaqueductal gray and dorsal raphe nuclei has shown to induce a decrease of the response to thermal and chemical stimuli, modulated by increased opiatergic, neuropeptidergic, and serotoninergic activity, and an analgesic effect of BDNF was reported following its intra-cerebroventricular

References 1. Apkarian AV, Bushnell MC, Treede RD, Zubieta JK: Human brain mechanisms of pain perception and regulation in health and disease. Eur J Pain 9:463-484, 2005 2. Bath KG, Lee FS: Variant BDNF (Val66Met) impact on brain structure and function. Cogn Affect Behav Neurosci 6:79-85, 2006 3. Beste C, Kolev V, Yordanova J, Domschke K, Falkenstein M, Baune BT, Konrad C: The role of the BDNF Val66Met polymorphism for the synchronization of error-specific neural networks. J Neurosci 30:10727-10733, 2010 4. Carvalho AL, Caldeira MV, Santos SD, Duarte CB: Role of the brain-derived neurotrophic factor at glutamatergic synapses. Br J Pharmacol 153(Suppl 1):S310-S324, 2008 5. Chen ZY, Patel PD, Sant G, Meng CX, Teng KK, Hempstead BL, Lee FS: Variant brain-derived neurotrophic factor (BDNF) (Met66) alters the intracellular trafficking and activity-dependent secretion of wild-type BDNF in neurosecretory cells and cortical neurons. J Neurosci 24: 4401-4411, 2004

administration.23 These apparently contrasting activities can be explained by the different effect of BDNF depending on the neurons and structures in which it is produced, released, and left to act. BDNF can therefore be associated with both an enhancement of the warning effect associated with afferent stimuli, and a decreased response to reduce the noise and protect the brain from the overstimulation. The latter effect seems relevant to explain the results of our study, because the analgesia induced by the neurotrophins could account for the decremental response to repeated painful stimulation that we have observed in Val/Val subjects and not in Met-carriers who are known to produce reduced levels of BDNF. Pain is not only a perceptual event but also an emotional experience. Studies on nociceptive blink reflex elicited with concentric electrodes have shown that emotional control modulates trigeminal nociception and pain perception, indicating supraspinal level processing.39 Emotional aspects are difficult to partial out in pain research; nonetheless, they are known to recognize well-characterized biological underpinnings,14 in which BDNF could play an important role.9,20,21 In conclusion, tPREP N2-P2 amplitude has shown a decremental in response to repeated painful stimuli. BDNF could mediate this phenomenon as a result of multiple mechanisms, including modulation on anti-nociceptive systems (due to processes of neural plasticity or to a direct neurotransmitter-like effect), induction of a serotoninrelated analgesia, and a plastic effect at the supraspinal level.

Acknowledgments We are in debt to the 2 anonymous reviewers for their invaluable advice and comments to the first version of this manuscript.

6. Cowansage KK, LeDoux JE, Monfils MH: Brain-derived neurotrophic factor: A dynamic gatekeeper of neural plasticity. Curr Mol Pharmacol 3:12-29, 2010 7. Delorme A, Sejnowski T, Makeig S: Enhanced detection of artifacts in EEG data using higher-order statistics and independent component analysis. NeuroImage 34:1443-1449, 2007 8. Di Lorenzo C, Di Lorenzo G, Sances G, Ghiotto N, Guaschino E, Grieco GS, Santorelli FM, Casali C, Troisi A, Siracusano A, Pierelli F: Drug consumption in medication overuse headache is influenced by brain-derived neurotrophic factor Val66Met polymorphism. J Headache Pain 10: 349-355, 2009 9. Duric V, McCarson KE: Effects of analgesic or antidepressant drugs on pain- or stress-evoked hippocampal and spinal neurokinin-1 receptor and brain-derived neurotrophic factor gene expression in the rat. J Pharmacol Exp Ther 319: 1235-1243, 2006 10. Egan MF, Kojima M, Callicott JH, Goldberg TE, Kolachana BS, Bertolino A, Zaitsev E, Gold B, Goldman D, Dean M, Lu B, Weinberger DR: The BDNF val66met polymorphism affects activity-dependent secretion of BDNF and

Di Lorenzo et al

The Journal of Pain

873

human memory and hippocampal function. Cell 112: 257-269, 2003

26. Mogil JS, Bailey AL: Sex and gender differences in pain and analgesia. Prog Brain Res 186:141-157, 2010

11. Foulkes T, Wood JN: Pain genes. PLoS Genet 4:e1000086, 2008

27. Pattwell SS, Bath KG, Perez-Castro R, Lee FS, Chao MV, Ninan I: The BDNF Val66Met polymorphism impairs synaptic transmission and plasticity in the infralimbic medial prefrontal cortex. J Neurosci 32:2410-2421, 2012

12. Gerritsen L, Tendolkar I, Franke B, Vasquez AA, Kooijman S, Buitelaar J, Fernandez G, Rijpkema M: BDNF Val66Met genotype modulates the effect of childhood adversity on subgenual anterior cingulate cortex volume in healthy subject. Mol Psychiatry 17:597-603, 2012 13. Guerini FR, Beghi E, Riboldazzi G, Zangaglia R, Pianezzola C, Bono G, Casali C, Di Lorenzo C, Agliardi C, Nappi G, Clerici M, Martignoni E: BDNF Val66Met polymorphism is associated with cognitive impairment in Italian patients with Parkinson’s disease. Eur J Neurol 16:1240-1245, 2009 14. Iannetti GD, Mouraux A: From the neuromatrix to the pain matrix (and back). Exp Brain Res 205:1-12, 2010 15. Iannetti GD, Zambreanu L, Cruccu G, Tracey I: Operculoinsular cortex encodes pain intensity at the earliest stages of cortical processing as indicated by amplitude of laserevoked potentials in humans. Neuroscience 131:199-208, 2005 16. IASP subcommittee on taxonomy: The need of a taxonomy. Pain 6:249-252, 1979 17. Katsarava Z, Ayzenberg I, Sack F, Limmroth V, Diener HC, Kaube H: A novel method of eliciting pain-related potentials by transcutaneous electrical stimulation. Headache 46: 1511-1517, 2006 18. Kaube H, Katsarava Z, Kaufer T, Diener H, Ellrich J: A new method to increase nociception specificity of the human blink reflex. Clin Neurophysiol 111:413-416, 2000 19. Klein T, Magerl W, Hopf HC, Sandkuhler J, Treede RD: Perceptual correlates of nociceptive long-term potentiation and long-term depression in humans. J Neurosci 24:964-971, 2004 20. Liu MG, Chen J: Roles of the hippocampal formation in pain information processing. Neurosci Bull 25:237-266, 2009 21. Magarinos AM, Li CJ, Gal Toth J, Bath KG, Jing D, Lee FS, McEwen BS: Effect of brain-derived neurotrophic factor haploinsufficiency on stress-induced remodeling of hippocampal neurons. Hippocampus 21:253-264, 2011 22. Mamounas LA, Blue ME, Siuciak JA, Altar CA: Brain-derived neurotrophic factor promotes the survival and sprouting of serotonergic axons in rat brain. J Neurosci 15: 7929-7939, 1995 23. Merighi A, Salio C, Ghirri A, Lossi L, Ferrini F, Betelli C, Bardoni R: BDNF as a pain modulator. Prog Neurobiol 85: 297-317, 2008 24. Mesulam MM: Large-scale neurocognitive networks and distributed processing for attention, language, and memory. Ann Neurol 28:597-613, 1990 25. Miltner W, Larbig W, Braun C: Habituation of subjective ratings and event-related potentials to painful intracutaneous electrical stimulation of the skin. J Psychophysiol 1: 221-228, 1987

28. Petroff OA, Errante LD, Rothman DL, Kim JH, Spencer DD: Neuronal and glial metabolite content of the epileptogenic human hippocampus. Ann Neurol 52: 635-642, 2002 29. Pezet S, McMahon SB: Neurotrophins: Mediators and modulators of pain. Annu Rev Neurosci 29:507-538, 2006 30. Riedl V, Valet M, Woller A, Sorg C, Vogel D, Sprenger T, Boecker H, Wohlschlager AM, Tolle TR: Repeated pain induces adaptations of intrinsic brain activity to reflect past and predict future pain. Neuroimage 57:206-213, 2011 31. Rottmann S, Jung K, Ellrich J: Electrical low-frequency stimulation induces homotopic long-term depression of nociception and pain from hand in man. Clin Neurophysiol 119: 1895-1904, 2008 32. Rottmann S, Jung K, Vohn R, Ellrich J: Long-term depression of pain-related cerebral activation in healthy man: An fMRI study. Eur J Pain 14:615-624, 2010 33. Sappey-Marinier D, Calabrese G, Fein G, Hugg JW, Biggins C, Weiner MW: Effect of photic stimulation on human visual cortex lactate and phosphates using 1H and 31P magnetic resonance spectroscopy. J Cereb Blood Flow Metab 12:584-592, 1992 34. Serrao M, Coppola G, Di Lorenzo C, Di Fabio R, Padua L, Sandrini G, Pierelli F: Nociceptive trigeminocervical reflexes in healthy subjects. Clin Neurophysiol 121: 1563-1568, 2010 35. Soliman F, Glatt CE, Bath KG, Levita L, Jones RM, Pattwell SS, Jing D, Tottenham N, Amso D, Somerville LH, Voss HU, Glover G, Ballon DJ, Liston C, Teslovich T, Van Kempen T, Lee FS, Casey BJ: A genetic variant BDNF polymorphism alters extinction learning in both mouse and human. Science 327:863-866, 2010 36. Stern AJ, Savostyanova AA, Goldman A, Barnett AS, van der Veen JW, Callicott JH, Mattay VS, Weinberger DR, Marenco S: Impact of the brain-derived neurotrophic factor Val66Met polymorphism on levels of hippocampal N-acetylaspartate assessed by magnetic resonance spectroscopic imaging at 3 Tesla. Biol Psychiatry 64:856-862, 2008 37. Valeriani M, de Tommaso M, Restuccia D, Le Pera D, Guido M, Iannetti GD, Libro G, Truini A, Di Trapani G, Puca F, Tonali P, Cruccu G: Reduced habituation to experimental pain in migraine patients: A CO(2) laser evoked potential study. Pain 105:57-64, 2003 38. Vossen H, Kenis G, Rutten B, van Os J, Hermens H, Lousberg R: The genetic influence on the cortical processing of experimental pain and the moderating effect of pain status. PLoS One 5:e13641, 2010 39. Williams AE, Rhudy JL: Supraspinal modulation of trigeminal nociception and pain. Headache 49:704-720, 2009