Rapamycin Delays Age-Related Hearing Loss

Rapamycin Added to Diet in Late Mid-Life Delays Age-Related Hearing Loss in UMHET4 Mice

Richard A. Altschuler 1,2,3 *, Lisa Kabara1 , Catherine Martin1 , Ariane Kanicki 1 , Courtney E. Stewart 1,2, David C. Kohrman1 and David F. Dolan1 1 Kresge Hearing Research Institute, Department of Otolaryngology, Head and Neck Surgery, University of Michigan, Ann Arbor, MI, United States, 2 VA Ann Arbor Health Care System, Ann Arbor, MI, United States, 3 Department of Cell and Developmental Biology, University of Michigan, Ann Arbor, MI, United States

Published: 07 April 2021.

This is a continuation of a first study which demonstrated rapamycin added to diet at 4 months of age had significantly less age-related outer hair cell loss in the basal half of the cochlea at 22 months of age compared to mice without rapamycin. The present study tested adding rapamycin to diet later in life, at 14 months of age, and added a longitudinal assessment of auditory brainstem response (ABR).

The present study used UMHET4 mice, a 4 way cross in which all grandparental strains lack the Cdh23753A allele that predisposes to early onset, progressive hearing loss. UMHET4 mice typically have normal hearing until 16–17 months, then exhibit threshold shifts at low frequencies/apical cochlea and later in more basal high frequency regions.

The mice ABR thresholds at 4, 12, 24, and 48 kHz were assessed and tested at 12, 18, and 24 months of age and compared to baseline ABR thresholds acquired at 5 months of age to determine threshold shifts (TS).

Threshold shifts at 12 months of age — No threshold shift from baseline at any frequency

Threshold shifts at 18 months of age — Mice with rapamycin added to diet had a significantly lower mean TS at 4 and 12 kHz compared to mice on the control diet. There was no significant difference in TS at 24 and 48 kHz.

Threshold shifts at 24 months of age — The mean 4 kHz TS in the rapamycin diet group was no longer significantly lower than the control diet group, while the 12 kHz mean remained significantly lower. Mean TS at 24 and 48 kHz in the rapamycin diet group became significantly lower than in the control diet group at 24 months.

The results show that a later life addition of rapamycin can decrease age-related hearing loss in the mouse model, however, it also suggests that this decrease is a delay/deceleration rather than a complete prevention.

Introduction:

Age-related hearing loss (ARHL) occurs in approximately one-third of people in the United States over the age of 65 increasing to approximately half of those over the age of 75 (e.g., Gates, 2006; Gates et al., 2010). ARHL can reduce ability to communicate, quality of life and social integration and has been identified as a major risk factor for depression and dementia (e.g., Gates et al., 2010; Lin et al., 2011; Davis and Smith, 2013).

One major cause of ARHL is an age-related loss of sensory hair cells, predominantly outer hair cells, and an accompanying decrease in auditory sensitivity as measured by threshold shifts (TS) in auditory brain stem response (ABR) in animal models and audiometric thresholds in people. The underlying mechanisms responsible for age-related hair cell loss remain unknown and there are no treatments currently being clinically applied to prevent or reduce this pathology.

The National Institute on Aging Intervention Testing Program (NIA-ITP) tests for treatments that can be added to diet to increase lifespan, using UMHET3 mice, a four-way cross, as a model. Four-way cross mice (from four different grandparent strains) provide for genetic heterogeneity and reduce strain specific effects. Among several effective treatments identified through NIA-ITP, addition of rapamycin to diet at 9 months of age was found to increase life span by 26% in male mice and 23% in female mice (Miller et al., 2014). We hypothesized that ARHL might share underlying mechanisms, such that treatments that enhance life span could also reduce and/or delay ARHL. This is consistent with studies that demonstrate the positive effects of rapamycin on age-related disorders in animal models, including decreases in cardiac pathology (Dai et al., 2014); muscle weakness (Bitto et al., 2016), cancer incidence (Anisimov et al., 2011), and cognitive decline (Halloran Hussong et al., 2012; Majumder et al., 2012).

In a previous study (Altschuler et al., 2018) we evaluated cochleae from 22 months old UMHET3 mice that had rapamycin added to their diet at 4 months of age as well as from control littermates with normal diet. The 22 months old rapamycin-treated mice had significantly less loss of outer hair cells in the basal half of the cochlea compared to the untreated controls. This sparing of hair cell loss was restricted to the basal half of the cochleae, while the apical half of the cochleae exhibited equivalently large losses of outer hair cells in both rapamycin-fed and normal diet controls (Altschuler et al., 2018).

These results showing rapamycin could reduce or delay age-related hair cell loss in the basal half of the cochlea at 22 months of age raised two pertinent issues. First, the apparent limitation of the treatment effect of rapamycin to the basal half of the cochlea could reflect differences in mechanisms underlying hair cell loss along the cochlear spiral. Alternatively, the effect of rapamycin treatment could be due to delaying rather than preventing age-related hair cell loss. Hair cell loss occurs earlier in apical vs. basal cochlea in most mouse strains (for reviews; Gratton and Vazquez, 2003; Ohlemiller, 2006). If the effect of rapamycin is to “delay” rather than prevent age-related hair cell loss, a treatment induced difference in apical cochleae might also have been present at an earlier time, but by 22 months of age the delay was over and the hair cell loss had equilibrated. The current study was designed to address these points by generating a longitudinal assessment of auditory brain stem response (ABR) thresholds at 4, 12, 24, and 48 kHz in individual mice at 5, 12, 18, and 24 months of age. The current study also addressed a second question of whether beginning rapamycin treatment at a more clinically relevant later age would still be effective in reducing or delaying ARHL. Recent studies have shown that rapamycin can extend life span in mice even when added to diet at 19–20 months of age (Harrison et al., 2009; Zhang et al., 2014) and can also reduce age-related pathologies such as cancer incidence and decreased muscle (Zhang et al., 2014) and cardiac function (Quarles and Rabinovitch, 2020) with late life application. We therefore chose to add rapamycin at a later time but prior to the first appearance of ARHL. Three of the four “grandparent strains” of the UMHET3 mice carry homozygous ahl alleles (Cdh23753A) that predispose to early onset, hair cell loss and progressive deafness in mice, thus restricting the progeny that can be used in auditory aging studies and decreasing the utility of this four way cross for auditory aging studies (Noben-Trauth et al., 2003; Mianné et al., 2016). For this reason, earlier studies from our group (Schacht et al., 2012; Altschuler et al., 2015) developed a different four-way cross, UMHET4, in which the grandparent strains lack the ahl predisposing Cdh23753A alleles. We returned to use of UMHET4 mice in the current study so that all of the progeny could be used. Our previous studies using UMHET4 mice (Schacht et al., 2012; Altschuler et al., 2015) and pilot animals in the current study showed that UMHET4 mice generally have a later appearing ARHL than UMHET3 mice and that ABR TS do not commonly initiate until around 18 months of age. The current study therefore tested the influence of adding rapamycin to mouse diet at 14 months of age.

image (12)

FIGURE 1. Comparison of mean auditory brain stem response threshold shifts (compared to 5 months of age) in the group with rapamycin added to diet at 14 months of age (rapa) vs. the group on control diet without rapamycin (control) at 12, 18, and 24 months of age at 4 kHz (A), 12 kHz (B), 24 kHz (C) and 48 kHz (D). Asterisks indicate significant differences. Please note differences between A, B, C, and D in scale bars for dB SPL on the “y” axis.

Discussion:

The results of the present study extend results of our previous study (Altschuler et al., 2018) that found rapamycin added to diet at 4 months of age reduced outer hair cell loss in the basal half but not apical half of the cochleae of 22 months old UMHET3 mice. The present study used a longitudinal measure of ABR to show rapamycin added to diet reduced mean threshold shifts at 4 and 12 kHz in 18 months old UMHET4 mice, largely by reducing the percent of mice showing TS. It is not well understood why hair cell loss occurs first in apical regions, then in basal regions in most mouse models and it has been suggested that different mechanisms could be influencing basal vs. apical age-related hearing loss (e.g., Schulte and Schmiedt, 1992; Dubno et al., 2013; Wu et al., 2020).

Since 4 kHz is processed in the apical half of the mouse cochlea, ~1.25 mm from apex (Viberg and Canlon, 2004), the results of the present study show that the influence of rapamycin is not restricted to the basal half of the cochlea. This provides indirect evidence that at least some components underlying age-related hearing loss (those that can be influenced by rapamycin) are present in both basal and apical cochleae.

Rapamycin acts on mammalian-target-of-rapamycin (mTOR) pathways (both mTORC1 and mTORC2). These pathways are multifaceted and in turn act on other functional signaling pathways including those associated with metabolism, proliferation, immune response and cell survival (Inoki et al., 2005a,b; Perl, 2015 for reviews; Wataya-Kaneda, 2015). The mTORC1 pathway can influence endoplasmic reticulum (ER) stress and the unfolded protein response (UPR) (e.g., Ye et al., 2015). ER stress-related factors have been shown to increase in the cochleae of aged mice (Wang et al., 2015) and ER stress-mediated apoptosis has been associated with noise-induced, ototoxic drug-induced and age-related hearing losses (Oishi et al., 2015; Wang et al., 2015; Hu et al., 2016; Mahdi et al., 2016). ER stress pathways could therefore be a target of the rapamycin effect on ARHL. Rapamycin could also stimulate the survival pathway of p-Akt (S473) via mTOR2 signaling, including reducing mitochondrial stress. Rapamycin could also act through its influence on the inflammatory response or through inhibition of oxidative stress pathways previously implicated in hair cell pathology (Yamasoba et al., 2013, for review). Future studies will be necessary to identify the specific target or targets and pathways underlying the treatment effect seen in the current study.

Another important result is that the rapamycin treatment-related sparing of 4 kHz hearing loss at 18 months is no longer present at 24 months of age and there is an associated large hair cell loss in the apical cochlea of most rapamycin diet and control diet mice at 24 months of age, with no difference between the groups. This suggests that rapamycin treatment delays but does not prevent hearing loss. It would be valuable to identify the specific mechanisms by which rapamycin delays hair cell loss, not only to increase understanding of general mechanisms underlying ARHL but to determine if the delay could be extended and even turned into prevention. The timing of the last ABR and terminal euthanasia in the present study was before large TS generally occurs at higher frequencies in the UMHET4 mouse model. A greater rapamycin induced sparing of hair cell loss and TS than observed in the current study might therefore be found at a later age when greater hair cell loss is occurring, as observed in the previous study in UMHET3 mice where ARHL occurs more rapidly. The lack of correlation between ABR TS and OHC loss in the more basal cochlea at 24 months of age is consistent with reports of age-related reduction or loss in OHC function with reduced distortion product otoacoustic emissions (DPOAE) appearing before age-related OHC loss (e.g., Syka, 2010, for review). One explanation is an age-related disruption of prestin in morphologically intact OHC (Chen et al., 2009; Syka, 2010). It would be valuable to examine the influence of rapamycin treatment on age-related decrements in DPOAE. The variability in the progression and extent of ARHL seen in the control diet UMHET4 mice may reflect their genetic diversity and we have previously shown this variability can be correlated with polymorphisms in specific genetic loci (Schacht et al., 2012). The variability seen in the treatment effect of rapamycin in the rapamycin diet group might also reflect UMHET4 genetic diversity and it would be interesting to identify such differences in future studies.

The present study also addressed the question of whether beginning rapamycin treatment later in life than the 4 months of age used in Altschuler et al. (2018) would be effective. The results show beginning treatment later in life, at 14 months of age is still effective. This is consistent with studies showing late life rapamycin delivery also enhances life span and delays/reduces age-related declines in cardiac, muscle and cognitive functions (Quarles and Rabinovitch, 2020, for review). The literature also suggests that late life intermittent administration of rapamycin and rapamycin-like compounds (“rapalogs”) with less side-effects in people, can also increase life span and decrease age-related declines (Anisimov et al., 2011; Arriola Apelo et al., 2016; Shavlakadze et al., 2018; Quarles and Rabinovitch, 2020). It would therefore be valuable to test late-life intermittent treatment effects of rapamycin and rapalogs on ARHL. Thus, with this data we can conclude that rapamycin delays age-related hearing loss within mice with further lateral considerations for the same effect in humans.