CHRONOBIOLOGY INTERNATIONAL https://doi.org/10.1080/07420528.2017.1343341

ARNTL, CLOCK and PER3 polymorphisms – links with chronotype and affective dimensions Konrad S. Jankowski a and Monika Dmitrzak-Weglarz a

Faculty of Psychology, University of Warsaw, Warsaw, Poland; Medical Sciences, Poznan, Poland

b

b

Psychiatric Genetics Unit, Department of Psychiatry, Poznan University of

ARTICLE HISTORY

ABSTRACT

Recently, seven single nucleotide polymorphisms (SNPs) of ARNTL, TIM and PER3 genes were found associated with affective temperaments in bipolar disorder patients. This study aimed to test whether a) the same associations appear in a non-clinical sample; b) the SNPs are related to other affective dimensions; c) the SNPs underpin the associations between chronotype and affective temperaments/dimensions. Three hundred thirty-eight university students completed the Temperament Scale of Memphis, Pisa, Paris and San Diego Auto-questionnaire, the Centre for Epidemiological Studies Depression Scale, the Perceived Stress Scale, the General Health Questionnaire, the Seasonal Pattern Assessment Questionnaire and the Composite Scale of Morningness. Seven SNPs of the ARNTL, TIM and PER3 genes were genotyped. According to nominal significance, ARNTL rs7107287 was associated with a cyclothymic temperament, depressive and stress symptoms, general mental health and perceived negative impact of seasonality, while TIM rs10876890 was associated with a hyperthymic temperament, and the TIM rs2291738 was associated with chronotype. Different SNPs were related to chronotype and affective temperaments/dimensions, and therefore, they seem to not underpin relationships between chronotype and affective dysfunction, that is, in the present study, eveningness was related to dysthymic, cyclothymic and irritable temperaments, more symptoms of depression, stress, worse mental health and a negative impact of seasonality, while morningness was related to hyperthymic temperament. The SNPs associations need further replication given that they did not achieve Bonferroni criteria of significance accounting for the number of polymorphisms considered and tests conducted.

1. Introduction Morningness–eveningness, also called chronotype or circadian typology, is a characteristic reflecting preferences to undertake activities at different times of day. In people with evening tendency the timing of sleep, daily activities and acrophases of physiological functions are shifted towards later hours as compared to morning individuals (Adan et al., 2012; Burgess & Fogg, 2008). A recent twin study has indicated that genetic factors explain 52% variability in morningness–eveningness (Barclay et al., 2010). Numerous studies using clinical and non-clinical samples have shown relationships between eveningness and adverse affective functioning. Evening chronotype has been related to unipolar and bipolar affective disorders (Ahn et al., 2008; Dilsaver &

Received 26 February 2017 Revised 2 June 2017 Accepted 13 June 2017 KEYWORDS

Affect; chronotype; circadian clock; clock genes; depression; polymorphism; seasonality; SNP; stress; temperament

Akiskal, 2009; Giglio et al., 2010; Hasler et al., 2010; Robillard et al., 2013; Wood et al., 2009), and seasonal affective disorder (Lee et al., 2011). Amongst clinical and non-clinical samples, eveningness has been associated with more symptoms of depression (Bahk et al., 2014; Chan et al., 2014; Gaspar–Barba et al., 2009; Jankowski, 2016; Konttinen et al., 2014; Merikanto et al., 2015; Müller et al., 2015) and more severe symptoms of seasonality (Jankowski, 2017; Zhang et al., 2015). In non-clinical samples, eveningness has been also associated with dysthymic (depressed-like), cyclothymic (rapidly changing affect), irritable (becoming easily annoyed) and anxious (worrying) temperaments (Chrobak et al., 2017; Park et al., 2015), whereas morningness has been linked to the hyperthymic (hypomanic-like) one (Park et al., 2015). In the conceptual framework, these affective

CONTACT Konrad S. Jankowski [email protected] Faculty of Psychology, University of Warsaw, Stawki 5/7, Warsaw 00-183, Poland. Supplemental data for this article can be accessed at www.tandfonline.com/icbi. © 2017 Taylor & Francis Group, LLC

2

K. S. JANKOWSKI AND M. DMITRZAK-WEGLARZ

temperaments describe normal behaviour which may eventually predispose to bipolar disorder (Akiskal & Akiskal, 2005); thus, the above links seem to be in line with the finding that eveningness is related to greater vulnerability to bipolar disorder in a non-clinical sample (Bullock et al., 2014). The underpinnings of associations between eveningness and adverse affective functions are not clear yet, but one line of research is to look at genetic basis of the relationships. For instance, a recent twin study has indicated an existence of genetic correlation between morningness–eveningness and depressive symptoms (Toomey et al., 2015), but possible molecular basis for these associations is yet not know. Circadian clock genes can be considered as candidates for such a molecular link. Association studies provide, however, often conflicting results regarding links between polymorphisms in circadian clock genes and behavioral outcomes. For instance, some studies have shown that morningness–eveningness is associated with aryl hydrocarbon receptor nuclear translocator-like (ARNTL) gene (Dmitrzak-Węglarz et al., 2016; Parsons et al., 2014), circadian locomotor output cycles kaput (CLOCK) gene (Dmitrzak-Węglarz et al., 2016), period 3 (PER3) gene (Archer et al., 2010; Dorokhov et al., 2017; Jones et al., 2007; Kunorozva et al., 2012; Ojeda et al., 2013; Parsons et al. 2014) or timeless (TIM) gene (Etain et al., 2014). Others, however, failed to show links between morningness–eveningness and these genes: ARNTL (Etain et al., 2014); CLOCK (Barclay et al., 2011; Serretti et al., 2010; Song et al., 2016); PER3 (An et al. 2014; Barclay et al., 2011; Dmitrzak-Węglarz et al., 2016; Etain et al., 2014; Hida et al., 2014; Kripke et al., 2014; Osland et al., 2011); and TIM (Dmitrzak-

Węglarz et al., 2016). Inconsistency in the findings points the need for further research in various samples. Polymorphisms in circadian clock genes have been also associated with various affective malfunctions (for a review see: McClung, 2007). Recently, Rybakowski et al. (2014a) investigated multiple single nucleotide polymorphisms (SNPs) in four circadian clock genes – CLOCK, ARNTL, TIM and PER3 – in relation to five affective temperaments (dysthymic, cyclothymic, hyperthymic, irritable and anxious) in bipolar patients. They found that a dysthymic temperament was associated with one SNP of the PER3 gene, a cyclothymic temperament was associated with two SNPs of the TIM gene, a hyperthymic temperament was associated with three SNPs of the ARNTL gene, and an anxious temperament was associated with four SNPs of the ARNTL gene (Table 1). The current research has been inspired by the one by Rybakowski et al. (2014a) and, therefore, aimed to test in a non-clinical sample of young adults the associations between the SNPs detailed in Table 1 and affective temperaments, as well as other affective dimensions, that is, levels of depression, stress, general mental health and seasonality. We also investigated whether the associations between affective dimensions and the circadian clock at the gene level are accompanied by associations at a behavioral level, i.e. chronotype.

2. Methods 2.1. Participants

The participants were 338 university students (51% females) from various universities located mainly in

Table 1. Single nucleotide polymorphisms of clock genes associated with affective temperaments in the study by Rybakowski et al. (2014a). Gene ARNTL

TIM

PER3

SNP ID rs1481892 rs4146388

Chromosomal position 13258497 13263181

MAF .258 .233

Allels G:C C:T

TaqMan assay ID Custom assay C___1870648_10

Function Intron Intron

rs11022780

13347436

.358

T:C

C__31248681_10

Intron

rs7107287

13269545

.233

G:T

C___1870658_10

Intron

rs2291738

55101548

.396

T:C

C___3134217_1_

Intron

rs10876890

55120018

.492

A:T

C___2690213_10

Intron

7770423

.450

C:T

C__11673507_10

Intron

rs228727

Dys dysthymic; Cyc cyclothymic; Hyp hyperthymic; Anx anxious.

Affective dimension Hyp, Anx; CC > CG + GG Hyp, Anx CC > CT + TT Anx CC > CT + TT Hyp, Anx GG > GT + TT Cyc TT < TC + CC Cyc AA < AT + TT Dys CC > CT + TT

CHRONOBIOLOGY INTERNATIONAL

Warsaw. They were Caucasians of Polish origin aged between 18 and 32 years (M = 22.2, SD = 2.2), but the majority (87%) were aged between 19 and 24 years, which is the most common age of university students in Poland. Twenty-one per cent of participants were also gainfully employed, but they were not working during the night. The distribution of study variables and percentage of assayed polymorphisms is shown in Table 2.

2.2. Questionnaires

The Composite Scale of Morningness (CSM; Jankowski, 2015; Smith et al., 1989) was used to measure chronotype. CSM has 13 items scored on a four or five Likert-type response format. Respondents are asked to tick the response that best describes them in terms of the hours they prefer to spend on various activities and sleep, and how easy it is for them to wake up. The total score ranges from 13 (eveningness) to 55 (morningness). Temperament Scale of Memphis, Pisa, Paris and San Diego–Autoquestionnaire in short version (TEMPS-A; Akiskal et al., 2005; Borkowska et al., 2010) was used to assess five affective temperaments: dysthymic, cyclothymic, hyperthymic, irritable and anxious. The shortened version of TEMPS-A is Table 2. Description of the studied variables in the sample. Gene ARNTL

TIM PER3 Variable CSM Dysthymic Cyclothymic Hyperthymic Irritable Anxious CES-D PSS-4 GHQ-28 NIS GSS

SNP ID rs1481892 rs4146388 rs11022780 rs7107287 rs2291738 rs10876890 rs228727 Range 14–51 0–1 0–1 0–1 0–1 0–1 0–51 0–15 4–64 0–5 0–24

Polymorphism frequency CC 11% CG 44% GG 46% CC 50% CT 42% TT 9% CC 23% CT 40% TT 37% GG 39% GT 42% TT 20% CC 22% CT 55% TT 23% AA 17% AT 57% TT 26% CC 38% CT 51% TT 11% Mean (SD) Median Skewness 33 (7.0) 33 .00 .25 (.27) .13 1.1 .44 (.27) .42 .27 .57 (.29) .63 −.28 .19 (.21) .13 1.2 .43 (.36) .33 .32 16 (9.8) 15 .68 6.5 (3.2) 6 .25 24 (11) 22 .95 1.0 (1.2) 1.0 .99 8.9 (4.4) 9.0 .25

Polymorphism frequencies may not add to 100% due to rounding error; SD standard deviation; CSM Composite Scale of Morningness; CES-D Centre for Epidemiological Studies Depression scale; PSS-4 Perceived Stress Scale; GHQ-28 General Health Questionnaire; NIS perceived Negative Impact of Seasonality; GSS Global Seasonality Score.

3

composed of 39 items with a 0 (no) – 1 (yes) response format. Total scores for each temperament dimension are represented by arithmetic means and range from 0 (no signs of a given temperament) to 1 (extreme level of a given temperament). Centre for Epidemiological Studies Depression scale (CES-D; Jankowski, 2016; Radloff, 1977) was used to assess depression symptoms. CES-D contains 20 items asking for the frequency of symptoms during the past week, and responses are given using a Likert-type format, with options ranging from 0 (rarely or none of the time/less than 1 day) to 3 (all of the time/5–7 days). The total score ranges between 0 (no depression symptoms) and 60 (extreme intensity of depression symptoms). General Health Questionnaire-28 (GHQ-28; Goldberg & Hillier, 1979; Makowska & Merecz, 2001) was used to assess general mental health. GHQ-28 contains 28 items regarding symptoms within the previous few weeks in four areas: somatic, anxiety/insomnia, social dysfunction and severe depression. Responses are recorded on a Likert-type format with four options ranging from 0 (not at all) to 3 (much more than usual). The total score ranges from 0 (excellent general mental health) to 84 (poor general mental health). Seasonal Pattern Assessment Questionnaire (SPAQ; Rosenthal et al., 1984) was used to assess seasonality and the perceived negative impact of seasonality. The magnitude of seasonal changes is measured with a global seasonality score (GSS), composed of questions regarding the degree of seasonal changes in sleep length, social activity, mood, weight, appetite and energy. GSS ranges from 0 (no change) to 24 (extremely marked change). The perceived negative impact of seasonality (NIS) is scored using a scale from 0 (no problem) to 5 (disabling problem). Perceived Stress Scale in four items version (PSS4; Cohen et al., 1983) was used to assess stress. PSS-4 consists of four items that ask for the frequency of stress symptoms during the last month. Responses are given on a Likert-type scale with options ranging from 0 (never) to 4 (very often). The total score ranges from 0 (no stress) to 16 (extreme stress). 2.3. Procedure

University students were recruited by mass email advertising possibility to enrol into the study and

4

K. S. JANKOWSKI AND M. DMITRZAK-WEGLARZ

by direct invitations by research assistants at universities and in student dormitories. The research assistants were trained psychology students, and recruitment was conducted between May and July. The exclusion criteria were night working hours (between 22:00 and 6:00) imposed by an employer (in case of working students) and consumption of alcoholic beverages within the past twenty four hours. Participation in the study was anonymous, voluntary and rewarded with a USB flash drive. Participants gave written, informed consent, and the study was conducted according to international ethical standards (Portaluppi et al., 2010) and approved by the institutional review board of the Faculty of Psychology at the University of Warsaw. After providing information about the study procedure and receiving consent, saliva sample was collected and questionnaires were given to fill out. Saliva for DNA extraction was collected using a Saliva DNA Collection, Preservation and Isolation Kit (Norgen Biotek Corp; Ontario, Canada) according to the manufacturer’s instructions. Specifically, participants were asked to rinse their mouth with water and wait for ten minutes before saliva collection. Next, they spit into the collection funnel until the tube reached two mL line. Then, preservative liquid was added, and the tube was shaken for at least ten seconds. After completing all samples, tubes were delivered to the laboratory for genotyping. The polymorphisms of clock genes were genotyped using TaqMan SNP Genotyping Assays (Applied Biosystems, Waltham, MA) and TaqMan Genotyping Master Mix according to the manufacturer’s protocols. The amplification for TaqMan SNP Genotyping Assay plates was done in an ABI PRISM® 7900HT Sequence Detection System. Data acquisition and analysis were performed using the allelic discrimination analysis module in SDS v2.4 software (Applied Biosystems, Waltham, MA). 2.4. Statistical analyses

Statistical analyses were conducted using IBM SPSS Statistics version 23. Preliminary analyses indicated that self-reported variables (TEMPS-A dimensions, CES-D, PSS-4, GHQ-28, GSS, NIS, and CSM scores) were not excessively asymmetric in total sample (Table 2) and in different polymorphisms (Supplement 1) and had similar

variances across different polymorphisms according to Leven’s test (p > .05; Supplement 1). Consequently, one-way analysis of variance (ANOVA) was used as an omnibus test to check for differences between genotypes in the dependent variables, while in case of non-homogenous variances (seven analyses; Supplement 1), Welch ANOVA was used instead. This was followed by LSD (or Games-Howell in case of non-homogenous variances) pairwise comparisons, if the F test was statistically significant (p < .05). Eta-squared (η2) was used as an effect size measure; η2 varies between 0 and 1 and can be interpreted as a percentage of variance in dependent variable explained by a factor (here, a given SNP). The results of genetic comparisons were considered at nominal significance (p < .05) and at significance corrected for 77 comparisons (11 variables * 7 SNPs) using Bonferroni method (p < .00065). Pearson’s correlation was used to assess relationships between affective dimensions and chronotype.

3. Results 3.1. Associations with circadian clock genes

Considering nominal p values before Bonferroni correction, ANOVA indicated that polymorphism in one SNP of the ARNTL gene (rs7107287) was associated with cyclothymic temperament, current depression symptoms, stress, general mental health and perceived negative impact of seasonality. Specifically, participants with TT variant in rs7107287 exhibited greater levels of cyclothymic temperament, depressive symptoms, stress, general mental health and more negative impact of seasonality as compared to those with GG variant (Table 3). Participants with GT variant in rs7107287 exhibited lower levels of depressive symptoms compared to those with TT variant and higher levels of general mental health as compared to those with GG variant (Table 3). Polymorphism in the rs10876890 SNP of the TIM gene was related to hyperthymic temperament in a way that participants with AT variant showed higher levels of hyperthymic affect than those with TT variant (Table 3). Polymorphism in

CT > CC**

η .003 .005 .017 .003 .001 .017 .011 .007 .004 .007 .029** TT > GG*,GT** TT > GG** GG < GT*,TT** TT > GG**

Dysthymic Cyclothymic Hyperthymic Irritable Anxious CES-D PSS-4 GHQ-28 NIS GSS CSM

GG .21 ± .24 .40 ± .26 .57 ± .28 .20 ± .21 .40 ± .36 14 ± 10 6.0 ± 3.2 22 ± 9.1 .73 ± 1.1 8.5 ± 4.3 33 ± 6.5

GT .26 ± .28 .45 ± .27 .57 ± .29 .17 ± .20 .45 ± .38 17 ± 9.0 6.6 ± 3.2 25 ± 11 .95 ± 1.1 8.9 ± 4.7 32 ± 6.9

TT .29 ± .29 .52 ± .25 .53 ± .28 .22 ± .23 .46 ± .36 19 ± 10 7.1 ± 2.9 28 ± 12 1.4 ± 1.4 9.6 ± 4.2 33 ± 7.7

2

η .008 .026* .001 .006 .004 .033** .023* .045** .046** .010 .001

TT > GG**

CC .28 ± .28 .48 ± .27 .53 ± .27 .21 ± .21 .44 ± .36 18 ± 11 7.0 ± 3.2 25 ± 9.1 1.1 ± 1.2 9.4 ± 4.4 31 ± 6.2

CT .24 ± .26 .42 ± .26 .59 ± .29 .18 ± .21 .43 ± .37 15 ± 9.5 6.4 ± 3.2 24 ± 11 .93 ± 1.2 8.5 ± 4.6 33 ± 7.2

TT .24 ± .26 .46 ± .26 .53 ± .27 .19 ± .21 .42 ± .38 16 ± 9.2 6.2 ± 2.9 24 ± 10 .80 ± 1.2 9.1 ± 4.1 32 ± 6.8

2

Pairwise comparison

TIM

rs2291738

ARNTL

rs7107287

Table 3. Polymorphisms in ARNTL and TIM genes associated with affective dimensions and morningness.

*p < .05; **p < .01; η2 effect size derived from analysis of variance; CES-D Centre for Epidemiological Studies Depression scale; PSS-4 Perceived Stress Scale; GHQ-28 General Health Questionnaire; NIS perceived Negative Impact of Seasonality; GSS Global Seasonality Score; CSM Composite Scale of Morningness.

AA .27 ± .28 .46 ± .27 .53 ± .27 .20 ± .22 .46 ± .41 17 ± 9.5 6.3 ± 2.7 25 ± 12 .91 ± 1.2 8.9 ± 4.3 33 ± 6.3

Pairwise comparison

rs10876890

AT .22 ± .24 .42 ± .26 .60 ± .29 .19 ± .21 .41 ± .35 15 ± 9.1 6.3 ± 3.2 24 ± 11 .92 ± 1.2 8.6 ± 4.5 33 ± 7.0

TT .29 ± .30 .48 ± .27 .49 ± .26 .19 ± .21 .45 ± .38 18 ± 11. 6.9 ± 3.4 24 ± 9.3 1.0 ± 1.2 9.5 ± 4.5 32 ± 6.5

η2 .010 .003 .030** .001 .003 .014 .007 .001 .000 .002 .005

AT > TT**

Pairwise comparison

CHRONOBIOLOGY INTERNATIONAL

5

the rs2291738 SNP of the TIM gene was associated with morningness–eveningness. Specifically, participants with CT variant in rs2291738 showed greater morningness than those with CC variant (Table 3). The remaining SNPs of the ARNTL (rs1481892, rs4146388, rs11022780) and PER3 (rs228727) genes were unrelated to self-reported markers of circadian and affective functioning. None of the associations achieved Bonferroni criteria of significance; the lowest p value was that for association between variants of rs7107287 ARNTL gene and general mental health for which nominal p = .001. 3.2. Genetic basis of association between affective functioning and chronotype

To conclude that circadian clock genes underlie associations between affective functioning and chronotype two preconditions should be met: a) there exist associations between affective functioning and chronotype; b) the same polymorphism is related to both affective functioning and chronotype. Pearson’s correlation indicated that eveningness (as measured by the CSM) was related to higher levels of dysthymic, cyclothymic, irritable and anxious temperaments and to lower levels of hyperthymic temperament (Table 4). Furthermore, eveningness was associated with more intense symptoms of depression, stress, worse mental health and higher perceived negative impact of seasonality, but was unrelated to the magnitude of seasonality (Table 4). It should be noted that the above relationships were small according to common guidelines (Cohen, 1992). Despite the above associations, chronotype and affective functions were related to different SNPs (TIM rs2291738 versus ARNTL rs7107287 and TIM rs10876890, respectively) and therefore do not seem to have the same underlying SNP within the studied SNPs. 4. Discussion This study demonstrated that in a sample of university students: 1) SNPs in the ARNTL and TIM genes were associated with cyclothymic and hyperthymic temperaments, symptoms of depression, stress, general mental health and perceived

6

K. S. JANKOWSKI AND M. DMITRZAK-WEGLARZ

Table 4. Pearson correlations between morningness and measures of affective functioning. CSM −.24*** −.13* .16** −.13* −.11* −.17** −.17** −.18*** −.20*** −.09

Dysthymic Cyclothymic Hyperthymic Irritable Anxious CES-D PSS-4 GHQ-28 NIS GSS

*p< .05; **p<.01; ***p<.001; CSM Composite Scale of Morningness; CES-D Centre for Epidemiological Studies Depression scale; PSS-4 Perceived Stress Scale; GHQ-28 General Health Questionnaire; NIS perceived Negative Impact of Seasonality; GSS Global Seasonality Score.

negative impact of seasonality; 2) the above affective dimensions were also associated with chronotype, but these associations do not seem to have a common genetic underpinning in the studied polymorphisms because chronotype and affective dimensions were related to different SNPs (TIM rs2291738 versus ARNTL rs7107287 and TIM rs10876890, respectively). In the following sections, the results are discussed in more details in reference to previous research (see also Table 5) and study limitations are outlined.

In the studied sample, only one SNP of the ARNTL gene (rs7107287) and one SNP of the TIM gene (rs10876890) were related to affective temperaments (cyclothymic versus hyperthymic, respectively). In contrast, the precursor study (Rybakowski et al., 2014a) found that a total of seven SNPs of the ARNTL, TIM and PER3 genes showed associations with temperamental dimensions (Table 1). This suggests that the genetic underpinnings of symptom intensity in a non-clinical sample may differ from that observed within a sample of patients with bipolar disorder. Nevertheless, the association of the ARNTL SNP rs7107287 with a cyclothymic temperament observed in our study is in line with a previous report indicating that this SNP is associated with bipolar disorder when patients are compared with the general population (Mansour et al., 2006). Furthermore, ARNTL rs7107287 (in interaction with PER3 rs2172563) has been also indicative of bipolar disorder in comparison with unipolar disorder (Dmitrzak-Weglarz et al., 2015), whereas within a sample of bipolar patients, it was related to the degree of lithium response (Rybakowski et al., 2014b). Our results strengthen the conclusion that the SNP in rs7107287 of the ARNTL gene is related to bipolar

Table 5. Single nucleotide polymorphisms of ARNTL rs7107287, TIM rs2291738 and rs10876890 showing associations in the current and previous research. Gene SNP ID ARNTL rs7107287

TIM

Study Current study

Sample Non-clinical

Dmitrzak-Weglarz et al., 2015 MD patients versus non-clinical Mansour et al., 2006 BD patients versus non-clinical Rybakowski et al., 2014a BD patients Rybakowski et al., 2014b BD patients rs2291738 Current study Non-clinical Mansour et al., 2006 BD patients versus non-clinical Rybakowski et al., 2014a BD patients Shi et al., 2008 BD patients with families Utge et al., 2010 Depression patients versus non-clinical Utge et al., 2011 Non-clinical rs10876890 Current study Non-clinical Dmitrzak-Weglarz et al., MD patients versus 2015 non-clinical Rybakowski et al., 2014a BD patients Rybakowski et al., 2014b BD patients Shi et al., 2008 BD patients with families

MD mood disorder; BD bipolar disorder.

Result Association with cyclothymic temperament, depressive symptoms, stress, general mental health, perceived negative impact of seasonality Epistatic interaction with PER3 rs2172563 associated with BD Association with BD Association Association Association Association

with with with with

hyperthymic and anxious temperaments the degree of lithium prophylactic response morningness–eveningness BD

Association with cyclothymic temperament Association with insomnia during a manic episode in BD patients Association with depression amongst individuals with fatigue Association with short sleep Association with hyperthymic temperament Association with MD Association with cyclothymic temperament Association with the degree of lithium prophylactic response Association with insomnia during a manic episode in BD patients

CHRONOBIOLOGY INTERNATIONAL

disorder, as this condition can be seen as an extreme of the cyclothymic temperament. The finding that the SNP in TIM rs10876890 is associated with hyperthymic temperament seems to be in line with a finding that this polymorphism differentiates patients with mood disorders from individuals without the condition (DmitrzakWeglarz et al., 2015). A previous study indicated that the above SNP was associated with insomnia during a manic episode (Shi et al., 2008). This association has now been strengthened by our results, as the TIM rs10876890 showed association with hyperthymic temperament, which can be viewed as a manic-like tendency. Overall, the accumulating evidence suggests that TIM rs10876890 is involved in bipolar disorder as it also showed associations with other features of the disorder, like the cyclothymic temperament in bipolar patients (Rybakowski et al., 2014a) and the degree of lithium prophylactic response (Rybakowski et al., 2014b). The present study indicated that ARNTL rs7107287 could be also relevant for other affective variables than those typically considered in the field of bipolar disorder research (Mansour et al., 2006; Rybakowski et al., 2014a; Rybakowski et al., 2014b), namely to symptoms of depression, stress, general mental health and perceived negative impact of seasonality. Future studies could confirm the importance of this SNP for such affective symptoms in other non-clinical samples. Amongst considered SNPs, only one (TIM rs2291738) was associated with chronotype – the most marked behavioural characteristic of individual differences in the circadian system (Adan et al., 2012). However, given that the rs2291738 was unrelated to affective dimensions, it can be concluded that the observed associations between affective variables and chronotype (Table 4) do not have a common genetic underpinning within the studied SNPs (Table 1). The associations between affective variables and chronotype showed that eveningness is related to rather less desirable characteristics, that is, more pronounced dysthymic, cyclothymic, irritable and anxious temperaments, more symptoms of depression, stress, perceived negative impact of seasonality and worse general mental health. Morningness, on the other hand, was associated to hyperthymic temperament. These results replicate and strengthen

7

recent findings. Specifically, the direction of recently observed associations of chronotype with affective temperaments (Chrobak et al., 2017; Park et al., 2015), depressive symptoms (Jankowski, 2016; Merikanto et al., 2015) and seasonality (Jankowski, 2017) found in community and student samples are similar to those reported here. Furthermore, the observation of poorer general mental health and emotional distress, as measured with GHQ-28 and PSS-4 scales, in evening individuals found here also replicates previous studies using the same assessment methods (Prat & Adan, 2013; Simor et al., 2015). There still remains, however, an unanswered question of whether or not circadian clock genes underlie these associations – the idea suggested by a twin study showing the genetic association between chronotype and depressive symptoms (Toomey et al., 2015). There are two important limitations of this study which have to be kept in mind when considering the results. Firstly, the study had limited capacity to detect very weak effects which often occur in gene association studies. Namely, the sample size did not allow to spot effects below 2% of explained variance; to detect SNP effect explaining 1% of variance with .80 power at nominal p < .05, a sample composed of 969 participants would be required (Faul et al., 2007). In other words, we cannot exclude the possibility that amongst the studied SNPs some underpin associations between chronotype and affective functions if the associations between SNPs and behavioural outcomes are very weak. This limitation points a need for further studies in larger samples. Another limitation, which is partly related to the one mentioned above, refers to multiple comparisons and resultant increase in type I error. The SNPs associations found in this study did not achieve Bonferroni criteria of significance accounting for the number of polymorphisms considered and tests conducted. It does not have to imply that the results obtained here are accidental (Perneger, 1998; Rothman, 1990), especially that the observed associations are rather in line with previous findings, but also indicates that further replication studies are required. Declaration of interest The authors report no conflicts of interests.

8

K. S. JANKOWSKI AND M. DMITRZAK-WEGLARZ

Funding This work was supported by the National Science Centre [DEC-2011/03/D/HS6/05760].

References Adan A, Archer SN, Hidalgo MP, et al. (2012). Circadian typology: A comprehensive review. Chronobiol Int. 29:1153–75. Ahn YM, Chang J, Joo YH, et al. (2008). Chronotype distribution in bipolar I disorder and schizophrenia in Korean sample. Bipolar Disord. 10:271–5. Akiskal KK, Akiskal HS. (2005). The theoretical underpinnings of affective temperaments: Implications for evolutionary foundations of bipolar disorder and human nature. J Affect Disord. 85:231–9. Akiskal HS, Mendlowicz MV, Jean-Louis G, et al. (2005). TEMPS-A: validation of a short version of a self-rated instrument designed to measure variations in temperament. J Affect Disord. 85:45–52. An H, Zhu Z, Zhou C, et al. (2014). Chronotype and a PERIOD3 variable number tandem repeat polymorphism in Han Chinese pilots. Int J Clin Exp Med. 7:3770–6. Archer SN, Carpen JD, Gibson M, et al. (2010). Polymorphism in the PER3 promoter associates with diurnal preference and delayed sleep phase disorder. Sleep. 33:695–701. Bahk YC, Han E, Lee SH. (2014). Biological rhythm differences and suicidal ideation in patients with major depressive disorder. J Affect Disord. 168:294–7. Barclay NL, Eley TC, Buysse DJ, et al. (2010). Diurnal preference and sleep quality: Same genes? A study of young adult twins. Chronobiol Int. 27:278–96. Barclay NL, Eley TC, Mill J, et al. (2011). Sleep quality and diurnal preference in a sample of young adults: Associations with 5HTTLPR, PER3, and CLOCK 3111. Am J Med Genet B Neuropsychiatr Genet. 156B: 681–90. Borkowska A, Rybakowski JK, Drozdz W, et al. (2010). Polish validation of the TEMPS-A: The profile of affective temperaments in a college student population. J Affect Disord. 123:36–41. Bullock B, Corlass-Brown J, Murray G. (2014). Eveningness and seasonality are associated with the bipolar disorder vulnerability trait. J Psychopathol Behav. 36:443–51. Burgess HJ, Fogg LF. (2008). Individual differences in the amount and timing of salivary melatonin secretion. PLoS One 3(8):e3055. Chan JWY, Lam SP, Li SX, et al. (2014). Eveningness and insomnia: Independent risk factors of nonremission in major depressive disorder. Sleep. 37:911–7. Chrobak AA, Tereszko A, Dembinska-Krajewska D, et al. (2017). Morningness–eveningness and affective temperaments assessed by the Temperament Evaluation of Memphis, Pisa and San Diego-Autoquestionnaire (TEMPS-A). Chronobiol Int. 34:57–65.

Cohen J. (1992). A power primer. Psychol Bull. 112:155–9. Cohen S, Kamarck T, Mermelstein R. (1983). A global measure of perceived stress. J Health and Social Behav. 24:385–96. Dilsaver SC, Akiskal, H.S. (2009). “Mixed hypomania” in children and adolescents: Is it a pediatric bipolar phenotype with extreme diurnal variation between depression and hypomania? J Affect Disord. 116:12–7. Dmitrzak-Weglarz MP, Pawlak JM, Maciukiewicz M, et al. (2015). Clock gene variants differentiate mood disorders. Mol Biol Rep. 42:277–88. Dmitrzak-Węglarz M, Pawlak J, Wiłkość M, et al. (2016). Chronotype and sleep quality as a subphenotype in association studies of clock genes in mood disorders. Acta Neurobiol Exp (Wars). 76:32–42. Dorokhov VB, Puchkova AN, Taranov AO, et al. (2017). A pilot replication study of two PER3 single nucleotide polymorphisms as potential genetic markers for morning and evening earliness-lateness. Biol Rhythm Res. DOI: 10.1080/ 09291016.2016.1275400. Etain B, Jamain S, Milhiet V, et al. (2014). Association between circadian genes, bipolar disorders and chronotypes. Chronobiol Int. 31:807–14. Faul F, Erdfelder E, Lang AG, et al. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Meth. 39:175–91. Gaspar–Barba E, Calati R, Cruz–Fuentes CS, et al. (2009). Depressive symptomatology is influenced by chronotypes. J Affect Disord. 119:100–6. Giglio LM, Magalhães PV, Andersen ML, et al. (2010). Circadian preference in bipolar disorder. Sleep Breath. 14:153–5. Goldberg DP, Hillier VF. (1979). A scaled version of the General Health Questionnaire. Psychol Med. 9:139–45. Hasler BP, Allen JJB, Sbarra DA, et al. (2010). Morningnesseveningness and depression: Preliminary evidence for the role of BAS and positive affect. Psychiatry Res. 176:166–73. Hida A, Kitamura S, Katayose Y, et al. (2014). Screening of clock gene polymorphisms demonstrates association of a PER3 polymorphism with morningness-eveningness preference and circadian rhythm sleep disorder. Sci Rep. 4:6309. Jankowski KS. (2015). Composite Scale of Morningness: psychometric properties, validity with Munich ChronoType Questionnaire and age/sex differences in Poland. Eur Psychiatry. 30:166–71. Jankowski KS. (2016). Morningness-eveningness and depressive symptoms: Test on the components level with CES-D in Polish students. J Affect Disord. 196:47–53. Jankowski KS. (2017). Morningness-eveningness and seasonality. Biol Rhythm Res. 48:331–42. Jones KH, Ellis J, von Schantz M, Skene DJ, et al. (2007). Age-related change in the association between a polymorphism in the PER3 gene and preferred timing of sleep and waking activities. J Sleep Res. 16:12–6. Konttinen H, Kronholm E, Partonen T, et al. (2014). Morningness–eveningness, depressive symptoms, and

CHRONOBIOLOGY INTERNATIONAL

emotional eating: a population–based study. Chronobiol Int. 31:554–63. Kripke DF, Klimecki WT, Nievergelt CM, et al. (2014). Circadian polymorphisms in night owls, in bipolars, and in non-24-hour sleep cycles. Psychiatry Investig. 11:345–62. Kunorozva L, Stephenson KJ, Rae DE, et al. (2012). Chronotype and PERIOD3 variable number tandem repeat polymorphism in individual sports athletes. Chronobiol Int. 29:1004–10. Lee HJ, Rex KM, Nievergelt CM, et al. (2011). Delayed sleep phase syndrome is related to seasonal affective disorder. J Affect Disord. 133:573–9. Makowska Z, Merecz D. (2001). Polska adaptacja Kwestionariuszy Ogólnego Stanu Zdrowia Davida Goldberga: GHQ-12 i GHQ-28. In Dudek B., ed. Ocena zdrowia psychicznego na podstawie badań kwestionariuszami Davida Goldberga. Łódź: Oficyna Wydawnicza Instytutu Medycyny Pracy, pp. 191–264. Mansour HA, Wood J, Logue T, et al. (2006). Association study of eight circadian genes with bipolar I disorder, schizoaffective disorder and schizophrenia. Genes Brain Behav. 5:150–7. McClung CA. (2007). Circadian genes, rhythms and the biology of mood disorders. Pharmacol Therap. 114:222–32. Merikanto I, Kronholm E, Peltonen M, et al. (2015). Circadian preference links to depression in general adult population. J Affect Disord. 188:143–8. Müller MJ, Cabanel N, Olschinski C, et al. (2015). Chronotypes in patients with nonseasonal depressive disorder: Distribution, stability and association with clinical variables. Chronobiol Int. 32:1343–51. Ojeda DA, Perea CS, Niño CL, et al. (2013). A novel association of two non-synonymous polymorphisms in PER2 and PER3 genes with specific diurnal preference subscales. Neurosci Lett. 553:52–6. Osland TM, Bjorvatn B, Steen VM, et al. (2011). Association study of a variable-number tandem repeat polymorphism in the clock gene PERIOD3 and chronotype in Norwegian university students. Chronobiol Int. 28:764–70. Park CI, An, SK, Kim HW, et al. (2015). Relationships between chronotypes and affective temperaments in healthy young adults. J Affect Disord. 175:256–9. Parsons MJ, Lester KJ, Barclay NL, et al. (2014). Polymorphisms in the circadian expressed genes PER3 and ARNTL2 are associated with diurnal preference and GNβ3 with sleep measures. J Sleep Res. 23:595–604. Perneger TV. (1998). What’s wrong with Bonferroni adjustments. Br Med J. 316:1236–8. Portaluppi F, Smolensky M, Touitou Y. (2010). Ethics and methods for biological rhythm research on animals and human beings. Chronobiol Int. 27:1911–29. Prat G, Adan A. (2013). Relationships among circadian typology, psychological symptoms, and sensation seeking. Chronobiol Int. 30:942–9.

9

Radloff LS. (1977). The CES–D scale: a self–report depression scale for research in the general population. Applied Psychol Measur. 1:385–401. Robillard R, Naismith SL, Rogers NL, et al. (2013). Delayed sleep phase in young people with unipolar or bipolar affective disorders. J Affect Disord. 145:260–3. Rothman K. (1990). No adjustments are needed for multiple comparisons. Epidemiology. 1:43–6. Rosenthal NE, Bradt GH, Wehr TA. (1984). Seasonal pattern assessment questionnaire (SPAQ). Bethesda, MD: National Institute of Mental Health. Rybakowski JK, Dmitrzak-Węglarz M, Dembińska-Krajewska D, et al. (2014a). Polymorphism of circadian clock genes and temperamental dimensions of the TEMPS-A in bipolar disorder. J Affect Disord. 159:80–4. Rybakowski JK, Dmitrzak-Weglarz M, Kliwicki S, et al. (2014b). Polymorphism of circadian clock genes and prophylactic lithium response. Bipolar Disord. 16:151–8. Serretti A, Gaspar-Barba E, Calati R, et al. (2010). 3111T/C clock gene polymorphism is not associated with sleep disturbances in untreated depressed patients. Chronobiol Int. 27:265–77. Shi J, Wittke-Thompson JK, Badner JA, et al. (2008). Clock genes may influence bipolar disorder susceptibility and dysfunctional circadian rhythm. Am J Med Genet B Neuropsychiatr Genet. 147B:1047–55. Simor P, Zavecz Z, Pálosi V, et al. (2015). The influence of sleep complaints on the association between chronotype and negative emotionality in young adults. Chronobiol Int. 32:1–10. Smith CS, Reilly C, Midkiff K. (1989). Evaluation of three circadian rhythm questionnaires with suggestion for an improved measure of morningness. J Applied Psychol. 75:728–38. Song HM, Cho CH, Lee HJ, et al. (2016). Associations of CLOCK, ARNTL, PER2, and GNB3 polymorphisms with diurnal preference in a Korean population. Chronobiol Int. 33:1455–63. Toomey R, Panizzon MS, Kremen WS, et al. (2015). A twin– study of genetic contributions to morningness–eveningness and depression. Chronobiol Int. 32:303–9. Utge SJ, Soronen P, Loukola A, et al. (2010). Systematic analysis of circadian genes in a population based sample reveals association of TIMELESS with depression and sleep disturbance. PLoS One. 5:e9259. Utge S, Kronholm E, Partonen T, et al. (2011). Shared genetic background for regulation of mood and sleep: Association of GRIA3 with sleep duration in healthy Finnish women. Sleep. 34:1309–16. Wood J, Birmaher B, Axelson D, et al. (2009). Replicable differences in preferred circadian phase between bipolar disorder patients and control individuals. Psychiatry Res. 166:201–9. Zhang L, Evans DS, Raheja UK, et al. (2015). Chronotype and seasonality: morningness is associated with lower seasonal mood and behavior changes in the old order Amish. J Affect Disord. 174:209–14.