SEARCH

SEARCH BY CITATION

Keywords:

  • HTR1A;
  • HTR2A;
  • HTR2C;
  • polymorphism;
  • SLC6A4;
  • sleep bruxism

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Declarations of Interest
  10. References

Sleep bruxism is a sleep-related movement disorder that can be responsible for various pains and dysfunctions in the orofacial region. The aim of the current case–control association study was to investigate the association of genetic, psychological and behavioral factors with sleep bruxism in a Japanese population. Non-related participants were recruited and divided into either a sleep bruxism group (= 66) or control group (= 48) by clinical diagnoses and 3-night masseter electromyographic recordings by means of a portable miniature device. The Epworth Sleepiness Scale, Temperament and Character Inventory, NEO-Five Factor Inventory and custom-made questionnaires that asked about familial aggregation, alcohol intake, caffeine intake, cigarette smoking, past stressful life events, daytime tooth-contacting habit, temporomandibular disorder, daily headache, snoring, apnea/hypopnea symptoms, leg-restlessness symptoms and nocturnal-myoclonus symptoms were administered. In addition, 13 polymorphisms in four genes related to serotonergic neurotransmission (SLC6A4, HTR1A, HTR2A and HTR2C) were genotyped. These factors were compared between case (sleep bruxism) and control groups in order to select potential predictors of sleep-bruxism status. The statistical procedure selected five predictors: Epworth Sleepiness Scale, leg-restlessness symptoms, rs6313 genotypes, rs2770304 genotypes and rs4941573 genotypes. A multivariate stepwise logistic regression analysis between the selected predictors and sleep-bruxism status was then conducted. This analysis revealed that only the C allele carrier of HTR2A single nucleotide polymorphism rs6313 (102C>T) was associated significantly with an increased risk of sleep bruxism (odds ratio = 4.250, 95% confidence interval: 1.599–11.297, = 0.004).This finding suggests a possible genetic contribution to the etiology of sleep bruxism.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Declarations of Interest
  10. References

Sleep bruxism (SB) is characterized by involuntary masticatory muscle activities during sleep and is classified as somatoform disorders in the International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10 F45.8). SB can be responsible for a variety of pains and dysfunctional conditions in the orofacial region (Manfredini and Lobbezoo, 2010). In the past two decades, it has been shown that SB is part of partial arousal phenomena and regulated centrally rather than peripherally, suggesting a possible involvement of central nervous system (CNS) neurotransmitters in its genesis (Kato et al., 2003; Lavigne et al., 2008). Several lines of pharmacological and therapeutic evidence have implicated serotonin (5-hydroxytryptamine; 5-HT), which is responsible for circadian rhythm, maintaining arousal and regulating muscle tone and breathing regulation (Monti and Jantos, 2008) in the pathogenesis of SB. For example, selective serotonin reuptake inhibitors (SSRIs) prescribed for depression have been reported to cause SB (Ellison and Stanziani, 1993; Gerber and Lynd, 1998), suggesting possible involvement of the 5-HT transporter in SB. Furthermore, SSRI-induced SB was reported to be controlled successfully by 5-HT 1A receptor agonist (Bostwick and Jaffee, 1999). Although a major limitation of these studies was a lack of actual SB measurements by means of laboratory-based or ambulatory electromyogram (EMG), these studies suggest that 5-HT is a candidate neurotransmitter that might be responsible for SB. It has been well documented that the 5-HT transporter is central to the fine tuning of brain serotonergic neurotransmission and of the peripheral actions of 5-HT by regulating the magnitude and duration of serotonergic responses (Murphy et al., 2008). The 5-HT transporter protein is encoded by a single gene, the solute carrier family 6, member 4, SLC6A4, which is located on chromosome 17q11.1–17q12. Previous studies have reported associations of SLC6A4 polymorphisms with sleep disorders, such as disturbed sleep and obstructive sleep apnea syndrome (Fabre et al., 2000; Ylmaz et al., 2005). In addition, the 5-HT receptor subtypes 1A, 2A and 2C have been suggested to regulate sleep and waking in a series of animal experiments that investigated the effects of agonists and antagonists of these receptors on sleep structure (Monti and Jantos, 2008). The gene polymorphisms of these receptors (which are encoded by HTR1A on chromosome 5q11.2–q13, HTR2A on chromosome 13q14–q21 and HTR2C on chromosome Xq24, respectively) have also been investigated for possible association with sleep disorders (Monti and Jantos, 2008).

To date, no genetic marker has been identified for SB; however, clinical studies have reported that 21–50% of SB patients have a direct family member who ground his or her teeth in childhood (Lavigne et al., 2008), that pairwise similarity of SB status was higher in monozygotic twin pairs than in dizygotic twin pairs (Hublin et al., 1998) and that 20 out of 39 children with tooth grinding had either one or two parents who exhibited tooth-grinding during sleep (Abe and Shimakawa, 1966). These studies suggest the possible involvement of genetic factors in the pathogenesis of SB.

For these reasons, the aim of this study was to investigate the association of 5-HT-related genetic factors, as well as psychological and behavioral factors, with SB and to explore the magnitude of their contribution to SB. We performed a case–control association study in a Japanese population.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Declarations of Interest
  10. References

Participants

Non-related Japanese participants (= 120; 59 females and 61 males) aged 22–69 years were recruited from patients and staff of Tokyo Medical and Dental University and Showa University and were enrolled into the study. They were screened to exclude those individuals with age under 20 or more than 70 years, absence of occlusal contact for the posterior teeth, orofacial dysfunction and acute symptoms. Those who had systemic diseases and who took or had a history of taking regular medication that affects the serotonergic system or sleep/wake regulation, or is used to treat movement disorders, were also excluded. The participants were first divided into either the SB group or the control group using clinical diagnostic criteria derived from the literature (Lavigne et al., 1996). Criteria to be included in the SB group were history of tooth-grinding noted by the participant’s sleep partner in the last 6 months plus at least one of the secondary criteria: the presence of tooth wear or shiny spots on restorations (Johansson et al., 1993); report of morning masticatory muscle fatigue or pain; or masseter muscle hypertrophy upon a voluntary clench in maximal intercuspal occlusion. This evaluation process identified 33 participants as sleep bruxers. As these clinical diagnostic criteria identified those who emitted grinding sounds but are not sensitive to SB without grinding sounds (namely clenching type SB), the remaining 87 participants underwent overnight masseter EMG recordings for three nights using a miniature EMG device (BiteStrip; S.L.P. Ltd, Tel Aviv, Israel). This device counts the number of masseter muscle activities above the 30% maximum voluntary contraction level and produces a four-grade score depending upon the number of counted events (L = fewer than 30 events; 1 = 31–59 events; 2 = 60–99 events and 3 = more than 99 events). The manufacturer of the BiteStrip has reported it to have sufficient validity, with average sensitivity of 0.72 and average positive predictive value of 0.75 compared with laboratory-based polysomnographic recordings (Shochat et al., 2007). The BiteStrip recordings revealed that 33 of the 87 participants tested exhibited a score of 3 for two nights or more and were then included into the SB group. Six participants who exhibited a score of 2 were excluded from the analysis because a rational judgement of their status was impossible (Fig. 1). Overall, 66 participants were in the SB group [34 males and 32 females; mean age 30.5 ± 8.3 standard deviation (SD) years] and 48 participants were in the control group (23 males and 25 females; mean age 33.0 ± SD 11.4 years) (Table 1). No significant difference was found in age and gender distribution between the two groups.

image

Figure 1.  Protocol for assignment of recruited participants to study groups.

Download figure to PowerPoint

Table 1.   Participant profile
 SBControlP-value
n = 66n = 48
  1. SD, standard deviation.

Female, n (%)32 (48.5%)25 (52.1%)0.704
Age ± SD (years)30.5 ± 8.333.0 ± 11.40.214

The study was approved by the Ethics Committees of Tokyo Medical and Dental University (Dental Ethics Committee, no. 332, approved on 26 May 2008) and Showa University (Ethics Committee for Genome Research, no. 94, approved on 18 April 2008) and was conducted in accordance with the Declaration of Helsinki. All participants gave written informed consent.

Questionnaire-based variables

Daytime sleepiness was evaluated by the Japanese version of the Epworth Sleepiness Scale (ESS) and a summary score (0–24) was calculated (Johns, 1991). Presence or absence of sleep-related symptoms experienced by participants was determined using questionnaires regarding snoring, apnea/hypopnea during sleep, leg restlessness before sleep and nocturnal myoclonus (yes/no). Personality traits were assessed by the Japanese version of the Temperament and Character Inventory (TCI) and the NEO Five-Factor Inventory (NEO-FFI) (Cloninger et al., 1993; McCrae and Costa, 1987). The TCI includes 240 items that discriminate four temperament and three character scales. The temperament dimensions include novelty seeking, harm avoidance, reward dependence and persistence, and the character dimensions assess self-directedness, cooperativeness and self-transcendence. The participants responded to each item by answering ‘true’ or ‘false’ and a total score for each temperament and character dimension was calculated (0–40 for novelty seeking, 0–35 for harm avoidance, 0–24 for reward dependence, 0–8 for persistence, 0–44 for self-directedness, 0–42 for cooperativeness and 0–33 for self-transcendence). The NEO-FFI consists of 60 items that discriminate five dimensions, which include neuroticism, extraversion, openness to experience, agreeableness and conscientiousness. The participants responded to each item on a scale between 0 and 4, and a total score for each dimension was calculated (ranging between 0 and 48 for each dimension). Presence of familial aggregation, past stressful life events, daytime tooth-contacting habit, temporomandibular disorder and daily headache (yes/no) and the amount of cigarettes smoked (number/day), alcohol intake (mL/day) and caffeine intake (mg/day) were also determined using custom-made questionnaires (Holmes and Rahe, 1967; Ohayon et al., 2001). Although amphetamines and cocaine have been reported to cause sleep bruxism (Lavigne et al., 2008), their intake was not included in the questionnaires because the number of participants who use them was expected to be extremely low.

Genotyping

Venous blood samples and scraped buccal cells were collected from all the participants for genotyping. Genomic DNA was purified from the blood samples with NucleoSpin Blood Quick Pure (Macherey-Nagel GmbH & Co. KG, Düren, Germany) and from buccal cells with the Gentra Puregene Buccal Cell kit (Qiagen K.K., Tokyo, Japan), according to the manufacturer’s instructions, respectively. Genomic DNA samples were cryopreserved at −20 °C until use. For SLC6A4 gene polymorphisms, two common polymorphisms (5-HTTLPR and STin2 VNTR) were determined with standard polymerase chain reaction (PCR) techniques. 5-HTTLPR is a deletion/insertion (s: short allele, l: long allele, xl: extra-long allele) of 44 base pairs (bp) in the promoter region approximately 1 kb upstream of the transcription site, and STin2 VNTR is a variable number of tandem repeats containing 10/12 copies of a 16–17 bp repeat element located in intron 2. PCR amplification was performed in a 50-μL reaction mix containing 1.5 μL of a 0.3 μm oligonucleotide primer mix, 2 μL of 1 mm MgSO4, 5 μL of 10× KOD plus Buffer (Toyobo Co., Ltd., Osaka, Japan), 5 μL of a 0.2 m MdNTP mix, 1 μL of 1U KOD plus DNA polymerase (Toyobo Co., Ltd), 10 μL of 2× PCRx Enhancer Solution (Life Technologies Japan Ltd, Tokyo, Japan) and 24.5 μL of distilled H2O, with 1 μL of 0.2 μg (5-HTTLPR) or 0.1 μg (STin2 VNTR) genomic DNA. The oligonucleotide primers flanking the polymorphisms were as follows: for 5-HTTLPR, forward 5′-GGC GTT GCC GCT CTG AAT GC-3′ and reverse 5′-GAG GGA CTG AGC TGG ACA ACC AC-3′; and for STin2 VNTR, forward 5′- GTC AGT ATC ACA GGC TGC GAG -3′ and reverse 5′- CAT GTT CCT AGT CTT ACG CCA GTG -3′, respectively.

Reactions were run on a TaKaRa Thermal Cycler MP (Takara Bio Inc., Shiga, Japan) using the following cycling parameters: an initial denaturation at 94 °C for 2 min, followed by 35 cycles at 94 °C denaturation for 15 s, 61 °C (5-HTTLPR) and 59 °C (STin2 VNTR) annealing for 30 s, 68 °C extension for 1 min, followed by a final extension step of 68 °C for 7 min. The fragments were separated and directly sized by electrophoresis in 3% agarose gels. After the gels were stained with ethidium bromide, the reaction products were visualized under ultraviolet (UV) illumination to distinguish the 484-bp ‘s’ fragment, 528-bp ‘l’ fragment, and 572-bp ‘xl’ fragment in 5-HTTLPR and the 267-bp ‘10’ fragment and 301-bp ‘12’ fragment in STin2 VNTR.

Further genetic analysis was performed to detect single nucleotide polymorphisms (SNPs) related to HTR1A, HTR2A and HTR2C. A SNP of HTR1A, rs6295, was derived from the Japanese SNP (JSNP) database (Hirakawa et al., 2002). Tag SNPs of HTR2A and HTR2C were derived from the Haplotype Map (HapMap) database in a Japanese population in the National Center for Biotechnology Information (NCBI) build 36 (International HapMap Consortium., 2003). The SNPs of HTR2A investigated were rs6313, rs1923884, rs2770304 and rs4941573, and those of HTR2C were rs518147, rs17260565, rs498177, rs12838742, rs6579495 and rs2192371. SNP genotyping was performed with a TaqMan Genotyping Assay (Life Technologies Japan Ltd, Tokyo, Japan) consisting of two allele-specific TaqMan probes labeled with fluorescein dye and two gene-specific primers. Amplification reactions were run in a volume of 10 μL containing 5 μL of 2× TaqMan Genotyping Master Mix buffer, 0.25 μL of 1× mix of unlabeled PCR primers and probes and 5–10 ng of template genomic DNA. All amplification and detection was conducted in 96-well PCR plates using the Applied Biosystems 7500 sequence detection system (Life Technologies Japan Ltd). Thermal cycling was initiated with a denaturation step of 10 min at 95 °C, followed by 40 cycles of 15 s at 92 °C and 1 min at 60 °C. After PCR was completed, allelic discrimination was analyzed using the Applied Biosystems Sequence Detection Software version 1.3 (Life Technologies Japan Ltd).

Statistical analyses

In order to identify the factors that influenced SB status, we applied univariate analysis using chi-square tests or Fisher’s exact tests for the categorical variables. For the continuous variables, Shapiro–Wilk tests (< 0.05) were first applied to determine whether the data sets were approximated by normal distributions. After that, we applied Student’s t-tests or Mann–Whitney U-tests for the variables. The tests were two-tailed, and the significance level was set at < 0.10 (Kimura et al., 2009).

We also performed a multivariate stepwise logistic regression analysis (< 0.05 to enter, > 0.10 to remove) between the predictors that were found by univariate analyses and SB status as an outcome, applying a forward selection method. Odds ratios and their 95% confidence intervals were calculated to evaluate the effects of factors. The significance level was set at < 0.05 in the multivariate model. The analysis was performed using spss version 11.5J (IBM Japan Ltd, Tokyo, Japan).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Declarations of Interest
  10. References

The univariate analyses selected five factors as predictors: ESS scores, leg-restlessness symptoms, rs6313 genotypes, rs2770304 genotypes and rs4941573 genotypes. The ESS score was slightly higher in the SB group (mean score of 10.6 ± 4.6) than in the control group (9.2 ± 3.9) (= 0.094, Table 2). No significant difference between the SB and control groups was found for each dimension of the personality traits as evaluated by TCI and NEO-FFI (Table 2). Similarly, no significant difference was observed between the two groups in the amount of cigarettes smoked, alcohol and caffeine intake, presence of possible familial aggregation, past stressful life events, daytime tooth-contacting habit, temporomandibular disorder, daily headache, daily snoring, apnea/hypopnea and nocturnal myoclonus (Tables 2 and 3). Interestingly, for the sleep-related symptoms, the presence of leg restlessness before sleep was associated significantly with SB status (14 of 59 participants in the SB group versus three of 40 participants in the control group, = 0.036; Table 3). For the genotype of 5-HT-related polymorphisms, significant differences were found in three of the four HTR2A SNPs examined: rs6313 (T/T genotype, eight of 66 SB participants versus 18 of 48 control participants, = 0.001); rs2770304 (A/A genotype, 16 of 66 SB participants versus 24 of 48 control participants, = 0.004); and rs4941573 (C/C genotype, 11 of 66 SB participants versus 19 of 48 control participants, = 0.006). No significant differences between the SB group and the control group were found in the polymorphisms of SLC6A4, HTR1A and HTR2C that were tested (Table 4).

Table 2.   Univariate analyses for continuous variables
 SBControlP-value
n = 66n = 48
Mean ± SDMean ± SD
  1. ESS, Epworth Sleepiness Scale; TCI, Temperament and Character Inventory; NEO-FFI, NEO-Five-Factor Inventory.

  2. *< 0.10.

  3. These six variables were examined using Student’s t-test, and the others were examined using the Mann–Whitney U-test.

ESS scores10.6 ± 4.69.2 ± 3.90.094*†
TCI scores
 Novelty seeking21.5 ± 4.521.7 ± 4.70.986
 Harm avoidance17.3 ± 6.216.6 ± 6.00.551
 Reward dependence16.5 ± 3.515.5 ± 3.60.128
 Persistence4.7 ± 1.94.7 ± 1.70.590
 Self-directedness28.8 ± 6.429.0 ± 6.70.780
 Cooperativeness29.3 ± 5.029.1 ± 5.40.852
 Self-transcendence11.3 ± 5.012.3 ± 6.30.587
NEO-FFI scores
 Neuroticism24.6 ± 7.624.5 ± 8.60.947
 Extraversion26.2 ± 6.925.1 ± 8.50.465
 Openness to experience30.1 ± 4.829.6 ± 5.50.244
 Agreeableness27.7 ± 5.227.5 ± 6.40.834
 Conscientiousness26.9 ± 6.826.7 ± 6.30.895
Cigarettes smoked (per day)3.0 ± 7.21.8 ± 4.90.258
Alcohol intake (mL day−1)13.0 ± 16.012.3 ± 18.30.403
Caffeine intake (mg day−1)85.9 ± 81.881.2 ± 77.90.767
Table 3.   Univariate analyses for categorical variables
 SBControlP-value
n = 66 (%)n = 48 (%)
  1. *P < 0.05.

  2. †Number of sleep bruxism group participants is 59 and number of control group participants is 40 in analyses for these variables based on the number of valid responses obtained.

  3. ‡This variable was examined using Fisher's exact test.

Possible familial aggregation
 Yes24 (36.4)13 (27.1)0.296
 No42 (63.6)35 (72.9)
Past stressful life events
 Yes43 (65.2)28 (58.3)0.458
 No23 (34.8)20 (41.7)
Daytime tooth-contacting habit
 Yes41 (62.1)27 (56.3)0.528
 No25 (37.9)21 (43.7)
Temporomandibular disorder
 Yes9 (13.6)4 (8.3)0.379
 No57 (86.4)44 (91.7)
Daily headache
 Yes16 (24.2)17 (35.4)0.194
 No50 (75.8)31 (64.6)
Daily snoring
 Yes26 (39.4)16 (33.3)0.508
 No40 (60.6)32 (66.7)
Apnea/hypopnea symptoms
 Yes4 (6.8)2 (5.0)1.000
 No55 (93.2)38 (95.0)
Leg restlessness symptoms
 Yes14 (23.7)3 (7.5)0.036*
 No45 (76.3)37 (92.5)
Nocturnal myoclonus symptoms
 Yes9 (15.3)10 (25.0)0.379
 No50 (84.7)30 (75.0) 
Table 4.   Univariate analyses for genotypes of polymorphisms
GenotypeSBControlP-valueGenotypeSBControlP-value
n = 66 (%)n = 48 (%)n = 66 (%)n = 48 (%)
  1. *< 0.01.

  2. This variable was examined using Fisher’s exact test.

  3. Number of control group participants is 47 due to a failure of the amplification.

  4. As for HTR2C, in general, females possess two alleles and males possess one allele in each SNP, because HTR2A is located on chromosome X.

SLC6A4, 5-HTTLPRHTR2C, rs518147 (−697G>C)
 s/s40 (60.7)26 (54.1)  G/G25 (78.1)21 (84.0) 
 s/l22 (33.3)19 (39.6)  G/C6 (18.8)3 (12.0) 
 l/l3 (4.5)2 (4.2)  C/C1 (3.1)1 (4.0) 
 s/xl1 (1.5)1 (2.1)  G31 (91.2)21 (91.3) 
 s/s40 (60.6)26 (54.2)0.327 C3 (8.8)2 (8.7) 
 s/l, l/l, s/xl26 (39.4)22 (45.8)  G/G, G56 (84.8)42 (87.5)0.687
SLC6A4, STin2 VNTR G/C, C/C, C10 (15.2)6 (12.5) 
 10/1057 (86.4)38 (79.2) HTR2C, rs17260565 (−147+4941A>G)
 10/128 (12.1)10 (20.8)  G/G1 (3.1)0 (0.0) 
 12/121 (1.5)0 (0.0)  G/A6 (18.8)4 (16.0) 
 10/1057 (86.4)38 (79.2)0.309 A/A25 (78.1)21 (84.0) 
 10/12, 12/129 (13.6)10 (20.8)  G2 (5.9)1 (4.3) 
HTR1A, rs6295 (−1019C>G) A32 (94.1)22 (95.7) 
 G/G37 (56.0)26 (54.1)  G/G, G3 (4.5)1 (2.1)0.637
 G/C25 (37.9)19 (39.6)  G/A, A/A, A63 (95.5)47 (97.9) 
 C/C4 (6.1)3 (6.3) HTR2C, rs498177 (−147+5558A>G)
 G/G37 (56.1)26 (54.2)0.841 G/G1 (3.1)1 (4.2) 
 G/C, C/C29 (43.9)22 (45.8)  G/A8 (25.0)7 (29.2) 
HTR2A, rs6313 (102C>T) A/A23 (71.9)16 (66.6) 
 T/T8 (12.1)18 (37.5)  G6 (17.6)3 (13.0) 
 T/C45 (68.2)23 (47.9)  A28 (82.4)20 (87.0) 
 C/C13 (19.7)7 (14.6)  G/G, G7 (10.6)4 (8.5)0.760
 T/T8 (12.1)18 (37.5)0.001* G/A, A/A, A59 (89.4)43 (91.5) 
 T/C, C/C58 (87.9)30 (62.5) HTR2C, rs12838742 (285543T>C)
HTR2A, rs1923884 (614-12062G>A) T/T25 (78.1)21 (84.0) 
 A/A19 (28.8)18 (37.5)  T/C6 (18.8)3 (12.0) 
 A/G34 (51.5)21 (43.7)  C/C1 (3.1)1 (4.0) 
 G/G13 (19.7)9 (18.8)  T31 (91.2)21 (91.3) 
 A/A19 (28.8)18 (37.5)0.327 C3 (8.8)2 (8.7) 
 A/G, G/G47 (71.2)30 (62.5)  T/T, T56 (84.8)42 (87.5)0.687
HTR2A, rs2770304 (613+11160G>A) T/C, C/C, C10 (15.2)6 (12.5) 
 A/A16 (24.2)24 (50.0) HTR2C, rs6579495 (345486T>G)
 A/G39 (59.1)18 (37.5)  G/G1 (3.1)1 (4.0) 
 G/G11 (16.7)6 (12.5)  G/T6 (18.8)3 (12.0) 
 A/A16 (24.2)24 (50.0)0.004* T/T25 (78.1)21 (84.0) 
 A/G, G/G50 (75.8)24 (50.0)  G2 (5.9)2 (8.7) 
HTR2A, rs4941573 (613+1668T>C) T32 (94.1)21 (91.3) 
 C/C11 (16.7)19 (39.6)  G/G, G3 (4.5)3 (6.3)0.695
 C/T43 (65.1)22 (45.8)  G/T, T/T, T63 (95.5)45 (93.7) 
 T/T12 (18.2)7 (14.6) HTR2C, rs2192371 (322171A>G)
 C/C11 (16.7)19 (39.6)0.006* G/G4 (12.5)3 (12.0) 
 C/T, T/T55 (83.3)29 (60.4)  G/A15 (46.9)11 (44.0) 
     A/A13 (40.6)11 (44.0) 
     G11 (32.4)7 (30.4) 
     A23 (67.6)16 (69.6) 
     G/G, G15 (22.7)10 (20.8)0.809
     G/A, A/A, A51 (77.3)38 (79.2) 

Finally, the multivariate stepwise logistic regression analysis was performed with the following predictors: ESS scores, leg-restlessness symptoms, rs6313 genotypes, rs2770304 genotypes and rs4941573 genotypes. The analysis revealed that only the C allele carrier of HTR2A SNP rs6313 was associated with SB status (odds ratio = 4.250, 95% confidence interval: 1.599–11.297, = 0.004; Table 5). ESS scores, leg-restlessness symptoms, rs2770304 genotypes and rs4941573 genotypes were not associated significantly with SB status.

Table 5.   Multivariate stepwise logistic regression analysis
 Odds ratio95% CIP-value
  1. −2 Log-likelihood = 124.6; Cox & Snell R2 = 0.087; Nagelkerke R2 = 0.117. CI, confidence interval.

  2. *P < 0.01.

  3. 1: T/C genotype and C/C genotype; 0: T/T genotype.

rs63134.2501.599–11.2970.004*
Constant0.500 0.109

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Declarations of Interest
  10. References

This study explored multi-dimensional factors—genetic, psychological and behavioral—in terms of their possible relationship to SB. To our knowledge, this is the first case–control study that included specific genetic polymorphisms as possible risk factors for SB. The results of this study suggest that the C allele carrier of HTR2A SNP rs6313 is associated with a 4.25-fold increased risk of SB.

Polymorphism of the 5-HT 2A receptor gene

The serotonergic system in the CNS shows reduced activity during sleep, most completely in rapid eye movement (REM) sleep, and has a role in maintaining arousal as well as regulating muscle tone and some of the phasic events during sleep (Monti and Jantos, 2008). While ingestion of l-tryptophan, a 5-HT precursor, did not reduce SB levels as measured by EMG (Etzel et al., 1991), SSRI use has been reported to be associated with SB, and SSRI-induced SB has been reported to be managed successfully by 5-HT 1A receptor agonist (Bostwick and Jaffee, 1999). For this reason, we investigated 5-HT-related polymorphisms for their possible association with SB, and found that a HTR2A polymorphism, specifically the C allele of rs6313, was associated significantly with SB.

Regarding the effect of HTR2A SNP rs6313 on gene expression, a previous post-mortem study has demonstrated that the presence of the C allele was related to reduced 5-HT binding to 5-HT 2A receptor in the superior frontal cortex compared with the T allele (Turecki et al., 1999). An in vitro study reported that the G allele of rs6311 SNP, which is in absolute linkage disequilibrium with the C allele of rs6313, has the potential to modulate HTR2A promoter activity negatively (Parsons et al., 2004). These findings suggest that the presence of the C allele, which is the risk indicator for SB, may predispose an individual to low overall expression of 5-HT2A receptor which, in turn, may predispose him or her to SB in combination with other factors. However, it should be noted that low overall expression of 5-HT2A receptor may be caused by multiple mechanisms (Monti and Jantos, 2008). In addition, SB itself is a multi-factorial disorder. Therefore, the mechanism underlying the association found in the current study remains to be established. Future investigations are warranted on functional differences between the C and T alleles of the5-HT2A receptor gene, which may possibly underlie the increased prevalence of the C allele among SB and contribute to this disorder. If confirmed, these findings may help to unravel one of the molecular mechanisms contributing to SB.

Other variables (stress, temperament and character, behavior, sleep-related disorders)

Although the HTR2A SNP was associated significantly with SB by the regression analysis, we understand that this association does not fully explain SB. SB is a multi-factorial disorder, and several studies have reported that behavioral factors such as alcohol and caffeine intake and cigarette smoking were related to SB (Ohayon et al., 2001). We found similar trends, but none of these associations was statistically significant. Moreover, other reported risk factors for SB, such as stressful life events, anxiety and familial aggregation (Hublin et al., 1998; Ohayon et al., 2001), were also found to be not significant in this study. These inconsistencies might be explained partially by the difference in the methodology to establish the case and control groups. These previous population-based studies did not perform objective measurement of SB, but largely used an SB diagnosis based on self-report of the participants (Lavigne et al., 2008; Ohayon et al., 2001). However, reliability of self-awareness of behavior during sleep has a certain limitation.

Regarding associations with other sleep disorders, a previous study has reported that the symptom of restless leg syndrome (RLS) was related to SB (Lavigne and Montplaisir, 1994). Another study has reported that obstructive sleep apnea syndrome, loud snorers, participants with moderate daytime sleepiness and RLS were related to SB (Ohayon et al., 2001). Although we did not make a diagnosis of RLS in the current study, the presence of leg restlessness that was reported by the participants before sleep was associated significantly with SB, suggesting a possible association between RLS and SB. This association warrants further investigation. Regarding sleep apnea, none of the variables that were suggestive of sleep apnea was associated significantly with SB. However, we found that daytime sleepiness as measured by ESS tended to be slightly higher in the SB group, which aligned with previous study results (Ohayon et al., 2001). With regard to other parasomnias, a population-based study reported co-occurrence of SB with sleep-talking (= 0.39), nightmares (= 0.30), and sleep-walking (= 0.25) in adults (Hublin et al., 2001). Because the number of study participants with these disorders was small and they were identified by questionnaire-based evaluation using dichotomous alternatives, future studies with more participants with sleep disorders who are identified by valid diagnostic procedures are highly recommended.

Limitations of the study

One of the limitations of this study was that we did not conduct laboratory-based polysomnographic recordings, which are the gold standard of SB measurements. However, obtaining such recordings from 120 participants takes a significant amount of time and expense and requires technical complexity (Lavigne et al., 2008). A previous population-based survey of 13 057 participants that reported several psychological and behavioral risk factors for SB used a clinical questionnaire on sleep bruxism using the ICSD minimal set of criteria in order to identify SB participants (Ohayon et al., 2001). As mentioned above, a diagnosis of SB status without actual recordings might be not very accurate and has the potential to lead to a false-negative diagnosis, especially in cases in which a patient makes no grinding sound with clenched teeth. Therefore, in the current study we first conducted clinical interviews and examinations for the clinical diagnosis of SB, and then obtained nocturnal EMG recordings by use of a miniature EMG device (Shochat et al., 2007) in the participants who had a negative report of grinding sound in order to make a final diagnosis. However, this single-channel recording device, which is regarded as a Type 4 recording system according to the American Academy of Sleep Medicine (AASM) classification of sleep apnea evaluation systems (Collop et al., 2007), has limited diagnostic validity. Therefore we excluded the six participants who exhibited marginal EMG level measured by this device because we could not make a rational judgement of their status.

Another limitation was that, although polymorphisms of other neurotransmitters (such as orexin, histamine, acetylcholine, noradrenaline, or even dopamine) might be candidates to be investigated, methodological considerations limited the number of polymorphisms relevant to 5-HT that were genotyped in this study. Several lines of sleep research have reported possible associations of orexin, histamine, acetylcholine, noradrenaline and dopamine with sleep/wake regulation (Szabadi, 2006). Because the majority of SB events are reported to occur associated with micro-arousal phenomena (Kato et al., 2003; Lavigne et al., 2008), the genes for these neurochemicals might be responsible for SB development. Therefore, they should be regarded as candidate genes to be investigated in future studies. It should also be noted that our study sample was a convenience sampling of dental patients and university staff, which makes the study results less generalizable to the Japanese population. Population-based genome-wide association studies that include genes which have been identified to be associated with movement disorders or sleep regulation should be encouraged in the future.

Conclusions

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Declarations of Interest
  10. References

The current study revealed the genomic association of the 5-HT 2A receptor single nucleotide polymorphism with sleep bruxism through an exploration of genetic, psychological and behavioral factors. As this is an exploratory design study, confirmatory and larger studies that investigate the exclusively found association are highly recommended in the future. If the association is confirmed by such studies, the result may help to elucidate the mechanism of sleep bruxism.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Declarations of Interest
  10. References

The authors are grateful to S. Yoshizawa, N. Watanabe, and S. Ando for their kind assistance with data collection and analyses. This study was supported in part by a Strategic Research Foundation Grant-aided Project for Private Universities grant (S0801016) from the Ministry of Education, Culture, Sport, Science and Technology, Japan, 2008.

Declarations of Interest

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Declarations of Interest
  10. References

The authors have indicated no financial conflicts of interest.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Declarations of Interest
  10. References