Auditory Brainstem Responses in the C57BL/6J Fragile X Syndrome Knockout Mouse Model

Sensory hypersensitivity, especially in the auditory system, is a common symptom in Fragile X Syndrome (FXS), the most common monogenic form of intellectual disability. However, linking phenotypes across genetic background strains of mouse models has been a challenge and could underly some of the issues with translatability of drug studies to the human condition. This study is the first to characterize the auditory brainstem response (ABR), a minimally invasive physiological readout of early auditory processing that is also used in humans, in a commonly used mouse background strain model of FXS, C57BL/6J. We measured morphological features of pinna and head and used ABR to measure hearing range, monaural and binaural auditory responses in hemizygous males, homozygous females and heterozygous females compared to wildtype mice. Consistent with previous work we showed no difference in morphological parameters across genotypes or sexes. Male FXS mice had increased threshold for high frequency hearing at 64 kHz compared to wildtype males, while females had no difference in hearing range between genotypes. In contrast, female homozygous FXS mice had decreased amplitude of wave IV of the monaural ABR, while there was no difference in males for amplitudes and no change in latency of ABR waveforms across sexes and genotypes. Lastly, FXS males had increased latency of the binaural interaction component (BIC) at 0 ITD compared to wildtype males. These findings further clarify auditory brainstem processing in FXS by adding more information across genetic background strains allowing for a better understanding of shared phenotypes.


Introduction 27
Fragile X Syndrome (FXS) is the most common monogenic form of autism spectrum disorder (ASD) 28 and shares many attributes of ASDs including auditory hypersensitivity and other sensory disruptions 29 (Abbeduto and Hagerman, 1997 Tian et al., 2017). Despite the common use of these models to study FXS, phenotypes are not always 33 shared between species and background strains, particularly for sensory processing. As a result, drug 34 therapies have struggled to rescue the human disorder (Dahlhaus, 2018). One of the most common 35 symptoms described in people with FXS and ASD is auditory hypersensitivity (Ethridge et al., 2017;36 Stefanelli et al., 2020). The mechanisms that underly auditory alterations are unknown, but likely 37 involve the entirety of the ascending pathway from the periphery to the cortex (reviewed in 38 McCullagh et al., 2020b). A complete characterization of auditory processing from the periphery to 39 cortex across sexes, background strains, and models is needed to fully understand shared phenotypes 40 and circuitry involved in this common symptom. 41 The auditory brainstem is one brain region in the ascending auditory pathway that has shown to have 42 anatomical, physiological, and behavioral alterations in FXS mouse models (Brown et al., 2010;43 Beebe 2019; Lu, 2019). The auditory brainstem is the first site of binaural processing of sound location in 46 the brain using interaural timing and level differences (ITD and ILD respectively) to compute sound 47 source locations (Grothe et al., 2010). This brain area is also involved in separating spatial channels 48 allowing for complex listening environments. Disruptions in this spatial separation and binaural 49 processing could lead to auditory hypersensitivity due to inability to separate sound sources 50 (Bronkhorst, 2015). One measure of auditory brainstem physiology, and binaural hearing, that can be 51 directly translated between animal models and humans is the auditory brainstem response (ABR) 52 (Laumen et al., 2016). 53 The ABR is a minimally invasive physiological measure that allows for simultaneous assessment of 54 sound processing across multiple brainstem nuclei, as each wave of the ABR directly corresponds to 55 distinct areas of the ascending auditory brainstem pathway. These features make the ABR an 56 attractive translational tool. Indeed, recent evidence suggests that ABR measurements are an early 57 indicator for auditory dysfunction in ASD (Santos et al., 2017). ABRs can also be used to assess 58 binaural hearing, which is essential for sound localization and hearing in noisy environments and 59 often impaired in ASD (Visser et al., 2013 CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted October 28, 2021. ; https://doi.org/10.1101/2021.10.27.466191 doi: bioRxiv preprint (Brittan-Powell and Dooling, 2004), or the lowest level (dB SPL) a response could be detected 123 (independent of wave) or 2.5 dB SPL below the lowest level that elicited a response. Audiogram 124 stimuli consisted of tone bursts (2 ms ± 1 ms on/off ramps) of varying frequency and intensity. 125

Monaural ABRs 126
Click stimuli (0.1 ms transient) were presented to each ear independently to generate monaural 127 evoked potentials. Peak amplitude (voltage from peak to trough) and latency (time to peak amplitude) 128 were measured across the four peaks of the ABR waveform at 90 dB SPL ( Figure 1A). The trough 129 was considered the lowest point for that wave. Monaural data from the two ears were averaged to 130 determine monaural amplitude and latency for each animal. Like hearing thresholds across 131 frequency, click threshold was determined for each genotype and sex. Click threshold is determined 132 by decreasing intensity of sound in 5 -10 dB SPL steps until ABR waveforms disappear. 133

Binaural ABRs 134
Click stimuli at 90 dB SPL were also presented to both ears simultaneously to generate a binaural 135 evoked potential.

Analysis of ABR waveforms 165
Custom python software was used to analyze evoked potentials for monaural and binaural stimuli 166 (New et al., 2021). To account for fluctuation in the baseline signal of the ABR, raw traces were 167 . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made     Results 210 We used both morphological and physiological features to examine hearing differences in a 211 commonly used mouse model of FXS, C57BL/6J across genotypes and sexes. Hearing measurements 212 . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made Fmr1, or Fmr1 het animals for pinna attributes ( Figure 2C pinna width, 2D pinna length, 2B effective 218 diameter). In addition, pinna characteristics were the same between the sexes (p = 0.175 pinna width, 219 p = 0.96 pinna length, p = 0.267 effective diameter Figure 2B-D). When genotypes were compared 220 within the same sex, there were no differences in weight, but sexes were significantly different 221 independent of genotype (p = 0.0023) with females weighing significantly less than males. Similar to 222 pinna morphology, there was no significant difference in either distance between pinna or distance 223 from the nose to pinna between the genotypes or sexes ( Figure 2E and F). These data suggest that 224 mice do not share the same craniofacial changes, at 225 least in the measurements described here, as people 226 with FXS. 227      Figure 3B). 256 There were no significant differences in hearing 257 range between the sexes, but Fmr1 males did have significantly higher threshold at 64 kHz than 258 . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted October 28, 2021. ; https://doi.org/10.1101/2021.10.27.466191 doi: bioRxiv preprint female Fmr1 mice suggesting that the phenotype is mostly driven by males (p = 0.0317) and indeed 259 both female Fmr1 and female Fmr1 het mice were not different than B6 females ( Figure 3A). Best 260 frequencies for both genotypes, as indicated by lower threshold, of mice were between 8 -64 kHz 261 consistent with specialized high frequency hearing. 262   Fmr1 mutants to monaural click stimuli compared to B6 mutant mice to determine if they have a 295 similar ABR phenotype to the FVB strain. We saw no differences in overall click threshold for either 296 genotype or sex (p = 0.102 genotype and p = 0.47 for sex). Amplitude of monaural responses was 297 significantly lower for wave IV of the ABR in Fmr1 females compared to Fmr1 het females ( Figure  298 4A upper). Indeed, Fmr1 het female amplitudes were closer to B6 than Fmr1 females, though Fmr1 299 females were not significantly different from B6. In contrast, Fmr1 male amplitudes for waves I-IV 300 were not different from B6 ( Figure 4A lower). When sexes were combined, Fmr1 het females had 301 significantly higher amplitudes than B6, and were close to being significantly higher than Fmr1 mice 302 (p = 0.0593). Consistent with sex driving the differences in genotype, peak amplitudes varied 303 between the sexes. Female B6 mice had significantly higher amplitude peaks I and IV compared to 304 B6 males (p = 0.0295 peak I and p = 0.0289 peak IV). In contrast, there were no sex differences 305 between male and female Fmr1 mice suggesting a more male-like phenotype (independent of 306 genotype) in homozygous Fmr1 females. There were no differences between the sexes or genotypes 307 in latency of monaural peaks ( Figure 4C and D). 308 . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made

Binaural hearing 310
While the monaural ABR provides information about binaural areas of the brainstem (potentially 311 peaks III and IV), since they are elicited by either sound played directly to one ear (closed field) or 312 equally to both ears (open field), little information can be gained about binaural integration of sound 313 information. We used the BIC of the ABR to measure binaural processing ability of the brainstem as 314 the BIC varies with ITDs played to both ears. We saw no differences in amplitude of the BIC at any 315 ITD between the two genotypes (p = 0.809) or with sex (p = 0.6904, Figure 5A, B), though there was 316 a significant difference between Fmr1 male and female mouse BIC amplitudes at 1.5 ms ITD. 317 Latency of the BIC was significantly slower in male Fmr1 compared to B6 ( Figure 5C, lower panel) 318 only at 0 ITD, with no difference in genotype for female mice ( Figure 5C, upper panel). When data 319 were combined for sexes across genotype, there was no significant difference in latency of the BIC at 320 any ITD ( Figure 5D). There were differences in latency of the BIC between B6 (-1.5 ms) and Fmr1 321 (1 ms) males and females though there was no overall main effect of sex (p = 0.3367). 322 This is the first study to characterize the ABR in the C57BL/6J Fmr1 mutant mouse, and in particular 354 highlights morphological characteristics, hearing range, monaural ABRs, and binaural integration 355 across sexes and in heterozygote females. Consistent with previous work, we see an increase in 356 . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made latency of the BIC at 0 ITD, but not other ITDs or changes in amplitude of the BIC across ITD 362 compared to B6 animals suggesting changes in timing of the processing of binaural information that 363 does not change overall ITD following ability. 364 Pinnae size and shape are the first feature available to determine sound localization ability in animals 365 with external ears (Butler, 1975;Musicant and Butler, 1984). conflicting results may be in part due to the earlier onset age-related hearing loss, which can be seen 400 as decreases in early waves of the ABR, that occurs in the B6 background (Hunter and Willott, 401 1987). Changes in wave I amplitude specific to FXS may be masked by overall decreases in wave I 402 . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made 2019), though they also see a decrease in amplitude of the BIC. Our findings that the BIC latency is 419 only significant in males at 0 ITD potentially suggests that there is overall slowing of binaural 420 processing in the brainstem, but that it is not dependent on ITD, which would be consistent with mice 421 that do not rely as predominantly on ITD cues compared to other species. 422 The subject of sex differences in animal models is important for fully understanding the complexities 423 of disorders such as autism spectrum disorder or FXS which seem to impact females differently than 424 males (Werling and Geschwind, 2013;Nolan et al., 2017). In FXS, due to it being an X-linked 425 disorder, there is a higher prevalence in males than females, which can undergo X-inactivation on the 426 effected X chromosome (genetic mosaicism) (Kirchgessner et al., 1995). However, mice offer a 427 unique opportunity to measure both heterozygote and homozygous females giving insight into 428 potential sex differences related to loss of Fmr1 on one or both X chromosomes. Our data suggest 429 that there are indeed differences in auditory phenotypes between heterozygous and homozygous 430 females (wave IV amplitude) in addition to differences between males and females. These and future 431 data comparing female Fmr1 subtypes may give insight into the role of X-inactivation in auditory 432 brainstem processing phenotypes. 433 In conclusion, this study offers important insight into auditory phenotypes that may be shared or 434 differ between background strains of FXS mice.  (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted October 28, 2021. ; https://doi.org/10.1101/2021.10.27.466191 doi: bioRxiv preprint . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted October 28, 2021. ; https://doi.org/10.1101/2021.10.27.466191 doi: bioRxiv preprint