Polygenic risk scores implicate genetic pathways involved in neurodevelopmental disorders in hearing thresholds and hearing asymmetry in children

An efficient auditory system contributes to cognitive and psychosocial development. A right ear advantage in hearing thresholds (HT) has been described in adults and atypical patterns of left/right hearing threshold asymmetry (HTA) have been described for psychiatric and neurodevelopmental conditions. Previous genome-wide association studies (GWAS) on HT have mainly been conducted in elderly participants whose hearing is more likely to be affected by environmental effects. We analysed HT and HTA in a children population cohort (ALSPAC, n = 6,743, 7.6 years). Better hearing was associated with more advanced cognitive skills and higher socioeconomic status (SES). Mean HTA was negative (−0.28 dB), suggesting a left ear advantage in children but mainly driven by females (−0.48 dB in females v -0.09 dB in males). We performed the first GWAS on HT in children and the very first GWAS on HTA (n = 5,344). Single marker trait association analysis did not yield significant hits. Polygenic risk score (PRS) analysis revealed associations of PRS for schizophrenia with HT, which remained significant after controlling for SES and cognitive skills, and of PRS for autism spectrum disorders (ASD) with HTA. Gene-based analysis for HTA reached genome-wide significance for MCM5, which is implicated in axon morphogenesis. This analysis also highlighted other genes associated with contralateral axon crossing. Some of these genes have previously been reported for ASD. These results further support the hypothesis that pathways distinguishing the left/right axis of the brain (i.e. commissural crossing) contribute to both different types of asymmetries (i.e. HTA) and neurodevelopmental disorders.


Introduction
The auditory system involves a series of complex distributed cerebral networks and 2 its impairment affects psychosocial, emotional and cognitive development 1 . Hearing-3 impaired children are at increased risk for learning disabilities 2 and even within the 4 normal range, better hearing has been associated with better reading skills, working 5 memory and nonverbal IQ in a sample of 1,638 UK school children 3 . 6 Hearing ability is usually defined as the threshold in decibel (dB) at which a tone is 7 perceived, so that lower values indicate better hearing. An age-related hearing decline 8 is well documented. In a Korean general population sample (n > 15,000) the hearing 9 threshold (HT) for medium frequencies declined from 3 dB in adolescents to 38 dB in 10 elderly participants 4 . A sex difference in favour of women was found in adults (n = 11 10,145; 30-69 years old), but not in children and young adults (n = 3, 458). 12 Successful hearing requires transformation of changes in air pressure into vibrations 13 in the basilar membrane that is transferred onto sensory hair cells of the inner ear, 14 whose depolarization is initiated by deflection of mechano-sensitive hair bundles 5 . The values indicate an advantage of the left ear. In a study of more than 50,000 adults 9 , a 23 right ear advantage (HTA between 1 dB and 4 dB) has been reported with more 24 pronounced HTA in males than in females. In a children sample of n = 1,191, a right 25 ear advantage has been reported, albeit to a smaller extent than in adults 10 . Other 26 authors found a general right ear advantage in males (n = ~400, HTA between 0.1 dB 27 and 0.5 dB) and a left ear advantage in females for specific frequencies (n = ~400, HTA 28 between -0.1 and -0.4 dB) 11 . Smaller studies reported a general left ear advantage in 29 children 12,13 . 30 An absence of HTA has been reported in schizophrenia 14,15 and ADHD 16 . Moreover, 31 symmetrical contralateral suppression in the olivary complex in the left and the right 32 ear in schizophrenia 17 is in contrast with the right ear advantage typically found in 33 controls. In children and adolescents, a right ear advantage has been reported in a 34 sample of n = 22 with autism spectrum disorder (ASD) while no asymmetry was found 35 in the control group 18 . A developmental effect towards stronger HTA has been 36 reported in controls that was absent in ASD children (n = 24) 19 . Reduced laterality in 1 processing auditory stimuli was reported in ASD and bipolar disorder (BIP) 2 suggesting that HTA is linked to neurodevelopmental disorders 20,21 . factors involved in hearing function in children. 22 We analysed HT and HTA in children from the Avon Longitudinal Study of Parents 23 and Children (ALSPAC) (n = 6,743). Consistent with previous studies we found that 24 better hearing is associated with enhanced cognitive skills and higher SES. We report 25 the first GWAS for HT in children and the very first GWAS for HTA (n = 5,344). In  of delivery from 1st April 1991 to 31st December 1992 were invited to take part in the 5 study, resulting in 14,062 live births and 13,988 children who were alive at 1 year of 6 age 47,48 . From age seven, all children were invited annually for assessments on a wide 7 range of physical, behavioural and neuropsychological traits. Informed written 8 consent was obtained from the parents after receiving a complete description of the 9 study at the time of enrolment into ALSPAC, with the option to withdraw at any time.

10
Ethical approval for the present study was obtained from the ALSPAC Law and Ethics 11 Committee and the Local Research Ethics Committees. The ALSPAC study website 12 contains details of all the data that is available through a fully searchable data 13 dictionary (http://www.bris.ac.uk/alspac/researchers/data-access/data-dictionary/).     For HT, we specifically tested for replication of 28 markers associated with quantitative hearing phenotypes in previous studies (p < 10 -5 29 28,31 , p < 10 -6 29,30 ). We also tested for replication of markers showing genome-wide 30 significance in case-control GWAS on age-related hearing loss (p < 5 × 10 -8 38,40 ). This      Table   2 S2) or HTA (Supplementary Table S3). However, there was a consistent pattern across 3 neurodevelopmental subgroups with more negative HTA compared to the control 4 group, indicating more leftward asymmetry.  In gene set enrichment analysis no GO term reached the Bonferroni-corrected level of 6 significance (p = 6.8 × 10 -6 ) for either HT or HTA. "Translational termination (GO:  Figure 4). The negative 6 association suggests that a genetic predisposition towards higher EA is associated with 7 better hearing.  There was no effect of PRS for IQ or BIP on either 6 phenotype.  1 We report a comprehensive study of HT and HTA in children to dissect their 2 relationship with cognitive abilities and neurodevelopmental disorders both at 3 phenotypic and genetic level. We confirm that better hearing is associated with better 4 performance on a range of cognitive abilities (Figure 2). We also report the results of 5 the first GWAS on HT in children. Single marker, gene-based and gene set enrichment 6 analyses did not lead to any statistically significant results. However, PRS for both 7 ADHD and schizophrenia were significantly associated with HT (Figure 4), with 8 genetic risk for ADHD and schizophrenia increasing and decreasing HT, respectively. 9 We also report the very first GWAS for HTA. Although we did not detect any single 10 marker associations, we found that PRS for ASD were statistically associated with 11 HTA (Figure 4), suggesting that higher genetic risk for ASD is associated with a shift 12 towards the left ear, indicating more asymmetry with better left ear performance.

13
Gene-based analysis for HTA highlighted several genes involved in axon guidance 14 and commissural axon crossing in sensory cerebral networks. Some of these genes 15 have previously been implicated in ASD. 16 At the phenotypic level, previous studies reported lower HT in females 65 . However, 17 this sex difference emerged around the age of 30, while there was no sex difference in 18 children and young adults 4 . Our data (n = 6,743) showed slightly higher HT in females 19 (6.30 dB) compared to males (5.98 dB) (Figure 1). Therefore, our data are in agreement 20 with a sex-specific developmental trajectory resulting in better hearing in female 21 adults that is not detectable in children. Consistent with previous studies 3 , we found 22 that in the normal range of variation, HT is negatively associated with several 23 cognitive skills (Figure 2). We did not detect any association between PRS for IQ and 24 HT suggesting that this association is not mediated by shared biological pathways. A 25 cause-effect relationship would be possible, but cannot be easily explained by our data.

26
The analysis of PRS for ADHD and schizophrenia instead support a role of genes 27 implicated in neurodevelopmental disorders contributing to HT. There was no 28 association between HT and ADHD at the behavioural level, however this analysis 29 was based on a very small sample of children meeting the criteria for ADHD (n = 21) 30 in our dataset and therefore the results might not be conclusive (Supplementary Table   31 S2). Hearing deficits in ADHD have been reported in terms of speech perception 66 , but 32 not in air conduction thresholds. Moreover, the effect of ADHD PRS on HT was 33 reduced after adjusting for cognitive skills, PRS and EA PRS (Supplementary Table   34 S7), suggesting that the effect was mediated by cognitive factors. In contrast, we found 35 that higher genetic risk for schizophrenia is associated with better hearing. This effect 36 was still found after adjusting for cognitive skills (Supplementary Table S7). PRS for 37 schizophrenia have recently been associated with better language skills, but not overall 1 school performance 67 . Thus, the association between better hearing and language 2 development ( Figure 2) could be based on shared biological pathways which also 3 increase the risk for schizophrenia. On the phenotypic level, there are no HT 4 differences between individuals affected by schizophrenia and controls 68 . Although no 5 single marker trait associations reached significance in the GWAS for HT 6 (Supplementary Figure S4), the top marker on chromosome 15 (rs1039444) is located 7 in an intron of RAB8B, which encodes for a GTPase that is expressed in inner and outer 8 hair cells and is involved in autosomal recessive deafness 69 . Targeted analysis for 9 markers reported in previous GWAS for HT in adults replicated association with only 10 one marker, rs12955474, which is located in an intron of the CCBE1 gene 28 . Other with the behavioural data, where we find a more leftward asymmetry in cases than 24 controls across different subgroups for disorders. 25 We found more negative HTA in neurodevelopmental conditions including ASD 26 compared to controls on the behavioural level that is congruent with the PRS results.

27
Different types of asymmetry such as structural brain asymmetry 73 , frontal alpha 28 asymmetry 74 , language processing 75 and handedness 76 have been implicated in ASD.

29
Gene-based analysis on HTA revealed genome-wide significance for the MCM5 gene

22
A possible limitation of the current study is that stimulus presentation was not 23 randomised, but the right ear was always tested first. Testing the right ear first has 24 been more common in previous studies than vice versa, which could either result in a 25 learning effect (favouring the left ear) or in a fatigue effect (favouring the right ear) 97 . 26 However, in adults, a right ear advantage is more common even in studies in which 27 the order of stimulus presentation has been randomised 97 , so the effect of stimulus 28 presentation should be minimal. The main limitation of this study is the rather small 29 sample size. It is of note that air conduction thresholds are not routinely collected in In summary, our results highlight the importance of HT for cognitive development. 3 We find that PRS for schizophrenia are implicated in HT, while PRS for ASD are 4 implicated in HTA. This is in line with the increasing evidence supporting a role of 5 asymmeties in ASD. Gene-based analysis highlighted genes involved in axon guidance 6 and ASD for HTA suggesting a role of genes involved in contralateral axon crossing 7 at the midline. The results support the hypothesis of shared pathways contributing to 8 different types of asymmetries, neurodevelopment and disorders. 9 Acknowledgements 1 We are extremely grateful to all the families who took part in this study, the midwives 2 for their help in recruiting them, and the whole ALSPAC team, which includes 3 interviewers, computer and laboratory technicians, clerical workers, research 4 scientists, volunteers, managers, receptionists and nurses. We are grateful to Veera M. 5 Rajagopal for useful comment to the manuscript.   17 The authors declare no competing financial interests in relation to the work described. retarded, and normal children. J Autism Dev Disord 13, 43-56 (1983