This study used immunohistochemical analysis to investigate dexamethasone uptake by auditory hair cells after transtympanic versus systemic administration. The utility of this modality to study hair cell steroid uptake is demonstrated in Figure 1, with minimal background labelling in negative controls (Figure 1A-B) and very strong labelling throughout the cochlea for positive control animals receiving extreme high dose dexamethasone (Figure 1C-D).
The preferential uptake of dexamethasone by inner hair cells compared to outer hair cells is striking, and to our knowledge not previously reported. Clear examples are demonstrated in Figures 1F and 3A (blue arrow marking inner hair cell and yellow arrow marking outer hair cells). This phenomenon was consistently observed for both transtympanic and systemic administration. Previous studies evaluating the oto-protective effects of glucocorticoids in acoustic trauma and aminoglycoside ototoxicity have clearly demonstrated greater sensitivity of outer hair cells to these cochlear insults, while inner hair cells are much more resistant [9, 10, 21]. With glucocorticoid administration inner hair cells were entirely preserved, while significantly more outer hair cells were preserved in a dose-dependent fashion. A greater affinity for glucocorticoids combined with patterns of endogenous steroid production may help explain the innate resistance of inner hair cells to ototoxic insults previously described.
In the present study both transtympanic and high dose systemic dexamethasone achieved strong uptake in basal inner hair cells at 1 hour, as demonstrated in Figure 2A and C. In contrast, Parnes et al. showed 1 hour dexamethasone peak levels in perilymph of 0.22 mg/L after high-dose intravenous (IV) injection, versus 1.553 mg/L in perilymph and 9.062 mg/L in endolymph after transtympanic injection . Yang et al. found even higher perilymph concentrations at 1 hour of 2.17 mg/L after transtympanic injection . Based on these previous studies of perilymph distribution it may be expected that hair cell uptake should be much lower with systemic versus transtympanic delivery. It was surprising to discover equally strong uptake of dexamethasone by auditory hair cells between the two delivery modalities.
After systemic and transtympanic injection, a decreasing basal-apical gradient of steroid uptake was present within hair cells which has not been previously characterized (Figure 2A-D). Salt et al. have demonstrated a perilymphatic gradient by sampling from the cochlear apex after transtympanic dexamethasone injection, generated by diffusion of steroid across the RWM . Knowing this, it is intriguing to see that vascular delivery of dexamethasone to the cochlea after systemic administration also leads to a basal-apical gradient within auditory hair cells. Such a gradient may have significant implications for steroid efficacy in different regions of the cochlea. Although the only previous study that used immunohistochemistry to study cochlear dexamethasone uptake failed to demonstrate this phenomenon, findings from the present study appear to support the current understanding of cochlear dexamethasone pharmacokinetics [18, 20].
Understanding the time course of steroid uptake and elimination from the cochlea is critical to the treatment of inner ear diseases. Previous studies have demonstrated peak perilymphatic concentrations at 30 min to 1 hour for systemic and transtympanic glucocorticoids [14, 16, 17, 19, 20]. At 6 hours, concentrations are markedly decreased, and at 24 hours glucocorticoids are largely eliminated from the inner ear. Building upon these data, this study evaluated hair cell dexamethasone uptake at 1, 6, and 12 hours. Temporal trends for high dose systemic administration demonstrated strong labelling at 1 hour, markedly decreased labelling at 6 hours and essentially no signal at 12 hours (Figure 3D-F), which parallels the aforementioned perilymph pharmacokinetics studies. However for transtympanic dexamethasone a very interesting effect of maximal 1 hour fluorescence (Figure 3A) and prolonged labelling of moderate strength up to 12 hours was observed (Figure 3B, 3C), suggesting retention of dexamethasone solution in the middle ear, continued permeation through the RWM into the perilymph, and continued auditory hair cell uptake. Eustachian tube losses and variable RWM permeability are challenges with transtympanic injection, but the present study demonstrates the potential for such strategies to markedly prolong glucocorticoid uptake by auditory hair cells. Preliminary work on sustained-release drug delivery vehicles such as dexamethasone hydrogels appears to demonstrate promising results [22, 23].
Regarding study limitations, immunohistochemical analysis has not been traditionally utilized for precise quantification. However, more powerful and sensitive confocal microscopes along with better secondary antibodies have allowed the detection of even trace amounts of a molecule of interest, and as such the dynamic range of the fluorescent label as it relates to concentration has been more accepted. In this study, outcomes were measured by subjective assessment of labeling by two independent reviewers, limiting the results to a descriptive analysis and thereby avoiding any errors that may be introduced by attempting to develop a new objective assessment method. In future experiments, as steroid dosing is decreased closer to clinical levels, more subtle differences in labeling may require increased precision and necessitate developing a form of objective measurement. However the large doses and subsequent obvious differences in labeling between groups allowed us to easily identify patterns through observation alone. Importantly, to limit variability a rigorous methodology was adhered to for all experiments, control mice were utilized, and importantly the confocal and laser settings were kept identical throughout the study. All sections were processed using the same batch of reagents and antibodies, and using the same time and temperature parameters.
Systemic dexamethasone concentrations used in this study are greater than those used in clinical practice. Parnes et al. demonstrated that with IV dosing of 0.2 mg/kg of dexamethasone, based on the dosing used clinically in humans, there was no demonstrable uptake in the plasma, CSF or perilymph in guinea pigs. High dose IV dexamethasone at 8 mg/kg yielded 1 hour drug peaks in the perilymph of 0.22 mg/L and the plasma of 2.123 mg/L . The dosing used in this study is in keeping with previously published literature. An extreme high dose of 100 mg/kg was also used as a maximal positive control (Figure 1), which would be toxic for clinical applications. This achieved very high intensity of labelling within hair cells, and interestingly this abolished the basal-apical gradient observed in the other experimental groups. Establishing that the organ of Corti can be maximally ‘saturated’ by such high dose dexamethasone, and that the basal-apical gradient can be overcome, serves as an intriguing metric with which to compare novel drug delivery strategies.