SEARCH

SEARCH BY CITATION

Keywords:

  • melanocytes;
  • stria vascularis;
  • tyrosinase;
  • tyrosine hydroxylase;
  • transgenic mice;
  • L-DOPA;
  • albinism

Summary

  1. Top of page
  2. Summary
  3. Significance
  4. Introduction
  5. Results
  6. Discussion
  7. Material and methods
  8. Acknowledgements
  9. References
  10. Supporting Information

Strial melanocytes are required for normal development and correct functioning of the cochlea. Hearing deficits have been reported in albino individuals from different species, although melanin appears to be not essential for normal auditory function. We have analyzed the auditory brainstem responses (ABR) of two transgenic mice: YRT2, carrying the entire mouse tyrosinase (Tyr) gene expression-domain and undistinguishable from wild-type pigmented animals; and TyrTH, non-pigmented but ectopically expressing tyrosine hydroxylase (Th) in melanocytes, which generate the precursor metabolite, L-DOPA, but not melanin. We show that young albino mice present a higher prevalence of profound sensorineural deafness and a poorer recovery of auditory thresholds after noise-exposure than transgenic mice. Hearing loss was associated with absence of cochlear melanin or its precursor metabolites and latencies of the central auditory pathway were unaltered. In summary, albino mice show impaired hearing responses during ageing and after noise damage when compared to YRT2 and TyrTH transgenic mice, which do not show the albino-associated ABR alterations. These results demonstrate that melanin precursors, such as L-DOPA, have a protective role in the mammalian cochlea in age-related and noise-induced hearing loss.


Significance

  1. Top of page
  2. Summary
  3. Significance
  4. Introduction
  5. Results
  6. Discussion
  7. Material and methods
  8. Acknowledgements
  9. References
  10. Supporting Information

This manuscript describes how melanin precursors, such as L-DOPA, prevent the profound premature age-related deafness and noise-induced hearing loss associated with albinism in mice. We use two well-defined transgenic mouse models of oculocutaneous albinism type I to study the auditory deficits associated with albinism in mice. We show that melanin and L-DOPA can prevent these hearing alterations. Since L-DOPA alone, a melanin precursor, is enough to rescue the observed deficits we conclude that melanin precursors produced by melanocytes is all what is needed to restore the alterations observed in albino mice by ABR. We do not know the exact mechanism nor we have determined the cause of the hearing loss in these albino mice. However, since melanin (and L-DOPA) can bind calcium we propose that the observed hearing loss phenotype might be caused by alterations in the calcium homeostasis of the endolymph, produced by the stria vascularis of the cochlea, where melanocytes are located within the inner ear. Our work might also be relevant for the corresponding genetic disorder in humans, where, to date, no systematic studies regarding auditory function have been carried out.

Introduction

  1. Top of page
  2. Summary
  3. Significance
  4. Introduction
  5. Results
  6. Discussion
  7. Material and methods
  8. Acknowledgements
  9. References
  10. Supporting Information

Neural crest derived cochlear melanocytes form the intermediate layer of the highly vascularized tissue at the lateral aspect of the cochlear scala media named the stria vascularis. This structure is essential for the maintenance of endolymph homeostasis and cochlear function and hence, for hearing (Meyer zum Gottesberge, 1988; Tachibana, 1999; Takeuchi et al., 2000; Wangemann, 2006). Accordingly, melanocyte deficiencies result in hearing impairment in mammals (Steel and Barkway, 1989).

The functions of the melanin or its precursors produced by strial melanocytes are ill-defined. Based on its complex structure and physicochemical properties, certain activities have been proposed for melanin, including calcium buffering, heavy metal scavenging and anti-oxidant activity (Bush and Simon, 2007; Dräger, 1985; Seagle et al., 2005). Melanin does not seem to be essential for normal hearing, yet despite some paradoxical results (Bartels et al., 2001; Ohlemiller et al., 2006), this pigment is thought to fulfill a protective role against age-related hearing loss (ARHL: Conlee et al., 1988; Hayashi et al., 2007), noise-induced hearing loss (NIHL: Ohlemiller and Gagnon, 2007) and ototoxicity (Conlee et al., 1989; Laurell et al., 2007; Wu et al., 2001). Genetically robust animal models have been essential to study the relationship between pigmentation and hearing (Steel and Kros, 2001). Most of these models analyze the hearing phenotype in melanocyte-deficient animals. Nevertheless, a detailed study of hearing in tyrosinase (Tyr) null mutant and Tyr transgenic mice has not yet been performed, where melanocytes are present and the underlying molecular alteration affects the production of melanin in albino individuals.

Here, we have compared hearing in YRT2 and TyrTH transgenic mice with that of their non-transgenic albino littermates. YRT2 mice carry a functional copy of the Tyr locus and are phenotypically undistinguishable from wild-type pigmented mice (Giraldo and Montoliu, 2002; Schedl et al., 1993). By contrast, TyrTH mice are phenotypically albino but ectopically express tyrosine hydroxylase (Th) under the control of Tyr. Th also generates L-DOPA from L-tyrosine, as Tyr (Riley, 1999), but it cannot further oxidize this intermediate to synthesize pigment. The production of L-DOPA in TyrTH mice, in the absence of melanin, was sufficient to rescue the retinal abnormalities present in albino mice (Lavado et al., 2006). We have applied these unique animal models of oculocutaneous albinism type 1 (OCA1: Oetting et al., 2003), already established for the analysis of the visual system (reviewed in: Lavado and Montoliu, 2006), to an analysis of the auditory system.

Hearing in these animals was evaluated during ageing and after noise exposure by means of auditory brainstem responses (ABR) (Murillo-Cuesta et al., 2009). These results show that YRT2 and TyrTH transgenic mice maintain normal auditory thresholds as they age, whereas in the albino non-transgenic mice there is an increase in the prevalence of premature ARHL. Rescue of Tyr expression or ectopic Th activity contributed to the recovery of hearing after a noise insult. These results suggest that melanin precursor metabolites play a key role in hearing maintenance and protection, as assessed by ABR.

Results

  1. Top of page
  2. Summary
  3. Significance
  4. Introduction
  5. Results
  6. Discussion
  7. Material and methods
  8. Acknowledgements
  9. References
  10. Supporting Information

Albino mice show premature age-related hearing loss that is delayed by Tyr and Th expression in melanocytes

To investigate the hearing capacity of albino and pigmented mice, ABRs were measured in pigmented YRT2 transgenic mice and their albino non-transgenic littermates at ages grouped as follows: younger than 2 months old, 2–4 months old and 4–6 months old. In the youngest mice analyzed (1.5 months old) the ABR thresholds were below 45 dB SPL and there was no significant difference between the pigmented and unpigmented mice. By contrast, adult 2- to 4-month-old albino non-transgenic mice displayed a higher average threshold level (∼55 ± 15 dB SPL) than the pigmented YRT2 mice (∼40 ± 7 dB SPL, Figure 1A).

image

Figure 1.  (A) ABR thresholds in response to click stimulus in albino (○, circles), transgenic YRT2 (•, filled circles) and TyrTH (△, triangles) transgenic mice at three different ages (<2 months, 2–4 months and >4 months), the former displaying age related hearing loss. Statistical analysis was performed with anova and post hoc Dunnett test (*P < 0.05, **P < 0.01). Data are presented as mean ± SEM. (B) Percentage of individuals showing normal hearing (≤45 dB SPL, white bards), moderate deafness (>45 to ≤60 dB SPL, grey bars) and profound deafness (≥70 dB SPL, black bars) mice in populations grouped by age (in months) and genotype, as indicated. Each bar represents 100% of the animals analyzed in each of the groups and has been further subdivided in sections depicting the percentage of individuals falling into any of the three considered categories regarding hearing capacity. There were a higher percentage of profoundly deaf albino mice when compared to both transgenic mouse strains. Number of individuals analyzed per group: <2 months, 19 albino and 14 YRT2 mice; 2–4 months, 24 albino, 21 YRT2 and 9 TyrTH mice, and; >4 months, 17 albino, 23 YRT2 and 9 TyrTH mice.

Download figure to PowerPoint

Individual ABR thresholds were grouped according to genotype, age and threshold level (Figure 1B). The percentage of mice with normal hearing capacity was comparable in young individuals (2 months old) regardless of their pigmentation and phenotype. As the mice aged, clear differences arose between YRT2 pigmented and albino non-transgenic littermates. Thus, while 54% of the albino non-transgenic mice displayed an average hearing threshold above 45 dB SPL and 27% had hearing thresholds in the range of profound deafness (over 70 dB SPL), only 5% of YRT2 mice displayed a phenotype of moderate deafness. In addition, most YRT2 mice older than 4 months had normal hearing (65%) and there were no profoundly deaf individuals, whereas 41% of the albino animals of an equivalent age suffered profound deafness.

Like the YRT2 pigmented line, premature age-related hearing loss was not detected in 2- to 4-month-old TyrTH transgenic mice when compared with their albino non-transgenic littermates (Figure 1A), even though TyrTH are phenotypically albino and have no detectable traces of pigment (Lavado et al., 2006). Analyzing the evolution of the hearing thresholds of the three genotypes with age, it was clear that the YRT2 and TyrTH transgenic mouse lines behaved similarly. In conclusion, the expression of either Tyr- or Th-containing transgenes in melanocytes on an albino background delays age-related hearing loss.

Similar differences were evident when the ABR record in response to click stimulus of a 2- to 4-month-old pigmented YRT2 transgenic mouse was compared with that of an albino non-transgenic mouse of an equivalent age with profound deafness (ABR threshold over 70 dB SPL, Figure 2).

image

Figure 2.  ABR patterns from two individuals with an extreme hearing phenotype determined with clicks of different intensities in a fixed scale. (A) Normal hearing YRT2 transgenic mouse. (B) Profoundly deaf albino non-transgenic mouse.

Download figure to PowerPoint

The frequency audiograms and I-V peak latencies obtained at different ages for the three genotypes were also studied (Figure 3). In younger animals, there were no detectable differences between the audiograms from albino and YRT2 pigmented mice (Figure 3A). However, older albino non-transgenic individuals displayed an increase in the ABR threshold at the frequencies analyzed, especially at 16 and 20 kHz, when compared to the transgenic YRT2 and TyrTH animals (Figure 3C, E). No significant differences were observed in the auditory thresholds between YRT2 and TyrTH transgenic mice at the frequencies studied.

image

Figure 3.  Frequency audiograms (A, C, E) and I–V peak latencies (B, D, F) obtained at different ages (<2 months, A–B; 2–4 months, C–D; >4 months old, E–F) for the three genotypes studied (albino, ○, open circles; YRT2, •, filled circles; and TyrTH, △, triangles). No significant differences were found between young individuals (A, B). Older albino mice showed a significant increase in tone-burst thresholds (C) and peak latencies (D). manova was applied to compare ABR thresholds means and peak latencies among the three genotypes in the different age groups. No statistically significant differences were observed between YRT2 and TyrTH transgenic mice for the parameters analyzed (*P < 0.05, **P < 0.01, ***P < 0.001). Data are presented as mean 6 SEM. Number of individuals analyzed per group: <2 months, 19 albino and 14 YRT2 mice; 2–4 months, 20 albino, 17 YRT2 and 9 TyrTH mice, and; >4 months, 11 albino, 16 YRT2 and 7 TyrTH mice.

Download figure to PowerPoint

The peak and interpeak latencies did not differ with age among the three genotypes compared, indicating that the hypopigmented phenotypes were not associated with alterations in the transduction of the stimuli along the central auditory pathway (Figure 3B, D, F; and Table S1).

Melanin precursor metabolites protect albino mice from noise-induced hearing loss

The differences between albino, YRT2 and TyrTH transgenic mice were further explored by exposing them to an excessive noise insult. This noise insult caused a threshold shift greater than 40 dB SPL, producing post-exposure hearing thresholds above 90 dB SPL in all the animals tested. One week after exposure to the insult, the threshold levels decreased about 20 dB SPL in TyrTH and YRT2 mice, but only 10 dB SPL in albino non-transgenic mice. The differences in the hearing thresholds between albino and transgenic mice became significant 2 and 3 weeks after exposure, although there were no significant differences between TyrTH and YRT2 mice during the recovery period. By 3 weeks after exposure to the noise insult, albino mice failed to recover basal ABR thresholds and they remained in a state of profound deafness. Conversely, mice able to synthesize melanin (YRT2) or its precursors (TyrTH) in melanocytes showed significant hearing recovery of ∼30 dB SPL after 3 weeks (Figure 4).

image

Figure 4.  Evolution of ABR thresholds after noise insult. A general linear model with repeated measures was applied to compare the evolution of ABR thresholds in the three genotypes during the experiment. No significant differences were found between the different genotypes before and immediately after noise insult. The recovery of ABR thresholds was impaired in albino animals after 3 weeks when compared to YRT2 and TyrTH transgenic littermates. Legends: albino, ○, open circles; YRT2, •, filled circles; and TyrTH, △, triangles; *P < 0.05, **P < 0.01, ***P < 0.001. Data are presented as mean 6 SEM. Number of individuals analyzed per group: 10 albino, 4 YRT2 and 6 TyrTH mice. All animals used were <2 months.

Download figure to PowerPoint

Analysis of cochlear morphology and cytoarchitecture

The different hearing responses observed in the mice studied suggested that the presence of melanin (as in YRT2 transgenic mice), or just melanin precursors, such as L-DOPA (as in TyrTH transgenic mice), may protect cochlear cells from ageing. A comparative histological study was carried out to rule out the possibility that morphological abnormalities in the cochlea might be responsible for the observed abnormal hearing phenotype in 2–4 and >4-month-old albino non-transgenic mice. Cochlear cytoarchitectonic and cell-type specific markers were studied as previously reported (Camarero et al., 2001, 2002; Murillo-Cuesta et al., 2009). As expected, there was no pigment at the stria vascularis of the albino non-transgenic and TyrTH transgenic mice since they are both unable to produce melanin. In contrast, pigmented intermediate cells were evident at the stria vascularis of YRT2 transgenic mice (Figure 5) that, as other pigment cells throughout the body, are known to bind calcium (Dräger, 1985). The presence of Th, and hence, the production of L-DOPA, was confirmed in cochlear melanocytes of TyrTH mice by immunofluorescence confocal microscopy (Figure 6). In accordance with the melanin and intermediate metabolite content of the different pigment cell populations, there was an increase in the L-DOPA concentration in the cochlea of TyrTH mice, as compared to albino non-transgenic mice (Wakamatsu K, Cantero M., Ito S. and Montoliu L., manuscript in preparation).

image

Figure 5.  Cochlear morphology in albino and YRT2 pigmented mice. (A) Dissection of the auditory inner ear showing the cochlea in toto in a pigmented mouse, the round window (RW) and the oval window (OW). Note the pigmented stria vascularis at the lateral wall. (B) Cross section of the dissected cochlea shown in (A) showing the pigmented stria (arrowheads). (C) Low magnification light micrograph of a mid-modiolar section from an albino mouse cochlea. Note the scala vestibule (SV), the scala tympani (ST), the scala media (SM) where the auditory receptor is located, and the spiral ganglion (SG, in the osseus Rosenthal`s canal). (D) A detail of a single turn of the cochlea showing the different structures: Reissner`s membrane (RM), spiral limbus (SL), basilar membrane (BM), stria vascularis (StV), spiral ligament (SpL), the spiral ganglion and the organ of Corti (OC). (E and F) Detail of the stria vascularis in albino (E) and YRT2 (F) mice, with the marginal cells (MC) close to the scala media, the intermediate cells (IC) and the basal cells (BC). The melanosomes containing melanin granules are only found in YRT2 mice (arrows in F). (G and H) The organ of Corti in albino (G) and in YRT2 (H) mice showing the neurosensorial cells (inner hair cell, IHC; outer hair cells, OHC), the tectorial membrane (TM), the spiral limbus (SL), the basilar membrane (BM), the tunnel of Corti (arrowhead) and the myelinated cochlear nerves fibers (CNF). There are no differences between genotypes in the neurosensorial organ. (I and J) Detail of the spiral ganglion in albino (I) and in YRT2 (J) mice where no alterations are evident. Insets: magnification of the spiral ganglion neurons. Scale bar 0.5 mm (A, B and C); 100 μm (D, G and H); 50 μm (I and J); 25 μm (E, F and insets in I and J).

Download figure to PowerPoint

image

Figure 6.  Detection of tyrosine hydroxylase (Th) in cochlear melanocytes of TyrTH transgenic mice. Cryostat sections of mouse retinas (A–D) and cohleas (E–L) from albino mice (E–H) and TyrTH transgenic mice (A–D; I–L) stained with and antibody against Th (Lavado et al., 2006) and observed with confocal microscopy. (A, E, I) DIC optics images depicting and adult mouse retina (A) and two stria vascularis (E and I); (B, F, J) TO-PRO3, labeling cell nuclei (shown in blue); (C, G, K) antibody against Th, labeling Th-expressing cells (shown in red); and (D, H, L) merge. Th-positive amacrine cells (shown in red) can be seen in a control retina section from a non-transgenic albino littermate, most apparent stripe of nuclei corresponds to the outer nuclear layer (panel C and D). Th-positive melanocytes present in the stria vascularis (shown in red) can be seen in TyrTH transgenic mice (panel K and L) as compared to albino non-transgenic littermates (panel G and H). Panels (M, N, O and P) shown below, correspond to higher magnifications of panels (G, H, K and L), respectively, demonstrating the presence of Th-positive melanocytes in the stria vascularis of TyrTH transgenic mice and their absence in non-transgenic albino mice. All images were captured using the same setting parameters at the confocal microscope. Scale bars 25 μm.

Download figure to PowerPoint

To further explore the basis of albino-associated hearing loss, the different cellular types and organization of the organ of Corti were studied with different cochlear specific markers. Actin in the hair bundles was labeled with phalloidin (Lin et al., 2005), the afferent fibers of the organ of Corti were labeled with neurofilament 200 kDa (Camarero et al., 2002), and synaptophysin was used to label presynapses in the inner hair cells and efferent fibers arriving at the organ of Corti (Knipper et al., 1995; Camarero et al., 2001). No additional cellular alterations were observed in the cochlea of any of the genotypes studied (Figure S1).

Similarly, morphological analyses after noise insult showed that the three mouse genotypes presented a similar pattern of cell degeneration, cell death and cytoarchitectonic alterations (Figure 7). Indeed, the cell damage provoked by the insult primarily affected fibrocytes of the spiral limbus and type IV fibrocytes of the spiral ligament, confirming previous reports (Ohlemiller and Gagnon, 2007; Wang et al., 2002). No structural differences were evident at the stria vascularis nor were there any alterations in the markers of strial activity like Na+K+-ATPase (Figure S2) or spiral ligament cell proliferation, like Ki67 (data not shown).

image

Figure 7.  Morphology of the cochlear from albino (A–C) and YRT2 pigmented mice (D–F) after noise insult. Characteristic morphological alterations include the degeneration of fibrocytes in the spiral limbus (SL, B and E) and of type IV fibrocytes in the spiral ligament (C and F) that were evident in several cochlear turns, although some remained unaffected (A and D). The Organ of Corti showed no significant lesions (C and F) and the pattern of lesions was similar for both genotypes. OHC, outer hair cells; SpL, spiral ligament. Scale bar 100 μm.

Download figure to PowerPoint

In summary, albino, TyrTH and pigmented YRT2 mouse cochlea had different pigment content in the stria vascularis melanocytes but no other obvious morphological differences that could account for the observed hearing loss associated with the albino condition.

Discussion

  1. Top of page
  2. Summary
  3. Significance
  4. Introduction
  5. Results
  6. Discussion
  7. Material and methods
  8. Acknowledgements
  9. References
  10. Supporting Information

In this study, we have shown that melanin precursor metabolites, such as L-DOPA, prevent premature age-related and noise-induced hearing loss associated to albinism in mice. YRT2 and TyrTH transgenic mouse models maintain normal auditory thresholds as they age and recover hearing after a noise insult, assessed by ABR, whereas the albino mice, in which the transgenic strains were created and maintained, show premature profound deafness and poor recovery of auditory thresholds after noise damage. Non-transgenic albino mice and TyrTH transgenic mice do not have pigment in the stria vascularis, whereas the presence of melanin is clearly evident in the pigmented YRT2 transgenic mouse. In our hands, the cochlea of the albino mouse did not show any other anatomical, architectonic or morphological differences, when compared with the cochleas of TyrTH and pigmented YRT2 mice. These results strongly suggest that rather than structural differences, the presence of melanin and, specifically, of its precursor metabolites, such as L-DOPA, is responsible for the auditory responses observed in the YRT2 pigmented or TyrTH transgenic mice, as compared to albino mice.

Melanocytes are required for the normal development of the cochlea and for the mature cochlea to function normally, as evident in mutant mice with congenital or late-onset melanocyte anomalies (Cable and Steel, 1998;Cable et al., 1992, 1994; Carlisle et al., 1990; Motohashi et al., 1994; Steel and Barkway, 1989; Tachibana et al., 2003). The stria vascularis is a highly vascularised epithelium of the lateral cochlear wall formed by marginal, intermediate melanocytes and basal cells. This structure acts as a blood-barrier to maintain the status of the endolymph, the sensory cells ionic environment. Electrophysiological and genetic studies have shown that strial cells and their cellular networks are essential for the maintenance of endolymph homeostasis and the generation of endocochlear potentials (Cohen-Salmon et al., 2007; Forge et al., 2003; Meyer zum Gottesberge, 1988; Tachibana, 1999, 2001; Wangemann, 2006). Hence, these cells are essential for hearing and accordingly, genetic deficiencies in melanocytes produce hearing impairment in mice and humans. For example, mutations in the human MITF locus cause the Waardenburg syndrome type IIA (Beighton et al., 1991;Saito et al., 2003; Tassabehji et al., 1994), whereas mutations in the homologous mouse Mitf locus also cause sensorineural deafness (Steingrímsson et al., 1994).

The function of strial melanin is not fully understood and different possibilities have been proposed on the basis of its structure and chemical characteristics (Laurell et al., 2007; Ohlemiller et al., 2006; Tachibana, 1999; Wu et al., 2001). Melanin binds Ca2+ (Dräger, 1985) and it could act as both an intracellular calcium buffer and reservoir that contributes to the calcium homeostasis of the endolymph (Bush and Simon, 2007; Gill and Salt, 1997). Calcium is a key element during sound-induced mechanical activation of the sensory hair cells in the Organ of Corti (Ferrary et al., 1988; Salt et al., 1989). During stimulation, calcium enters hair cells through specific channels and it is then recycled and transported back to the stria vascularis to be released again into the endolymph (Mammano et al., 2007). Interestingly, the Ca2+ concentration in the endolymph is 40 and 100 times lower than that of the intrastrial fluid and the plasma, respectively (Wangemann, 2006). Therefore, the calcium-binding capacity of melanin could be central to the calcium homeostasis of the endolymph.

Age-related hearing loss in mammals has been widely studied using mice as experimental models (Ohlemiller, 2006). Several factors appear to modulate hearing loss associated with ageing, including the mouse strain, specific genetic modifications and radical oxygen species accumulation (Erway et al., 1996; Ohlemiller et al., 2006; Riva et al., 2007; Staecker et al., 2001). Changes in melanin composition have also been proposed to participate in age-induced hearing loss (Hayashi et al., 2007). The absence of strial melanin has been recently shown to coincide with age-associated marginal cell loss and endocochlear potential decline (Ohlemiller et al., 2009). However, there are few studies on the hearing abilities of NMRI mice, the albino strain used in this work, during ageing. Young NMRI mice present normal hearing (Müller et al., 1997; Sterbing and Schrott-Fischer, 2002) and a progressive hearing loss with ageing, where the high frequencies appear to be the most affected ones (Drayton and Noben-Trauth, 2006), and which is less dramatic than that observed in other albino mouse strains (Drayton and Noben-Trauth, 2006; Ohlemiller et al., 2009).

The results presented here complement these studies by showing that Tyr activity fulfils a fundamental role in maintaining hearing through the production of melanin precursors, such as L-DOPA and its derivatives, including melanin. Profound age-related hearing loss primarily affected albino mice and it was delayed by Tyr expression. Interestingly, when a novel source of L-DOPA was provided through ectopic Th activity in albino cochlear melanocytes, the hearing phenotype was comparable to that in the pigmented animals. These data suggest that L-DOPA production, a melanin precursor, is sufficient to delay ARHL in albino non-transgenic mice. Our results cannot exclude that the role of L-DOPA in the process could be indirect, through its metabolite derivatives. Also, we cannot exclude a collaborative role of melanin in the process, in addition to the fundamental role that our results demonstrate for L-DOPA and/or its subsequent metabolite derivatives.

Noise insult is a common threat to hearing during animal life span and genetic factors may contribute to noise susceptibility in mice (Davis et al., 2001; Ohlemiller and Gagnon, 2007). Several functional studies have reported that mutations in the Ahl gene encoding otocadherin CDH23 (Di Palma et al., 2001) renders mice more susceptible to NIHL (Harding et al., 2005). However, other genes as well as systemic and local factors must also be involved. Studies on the potential protective effect of melanin against NIHL suggest that different types of melanin may fulfill distinct roles, although strain differences may make it difficult to compare these data (Barrenäs and Holgers, 2000; Bartels et al., 2001).

We have not determined the precise cause of the hearing loss phenotype observed in these albino mice, assessed by ABR analyses. The protective role of melanin may be due to its physical-chemical properties. As discussed above, melanin has free radical scavenging properties and therefore, it could act as an antioxidant during cochlear injury induced by oxidative stress in the presbycusis (Jiang et al., 2007) and NIHL (Samson et al., 2008). In addition, it is also a low-affinity high-capacity calcium-binding polymer that may contribute to the maintenance of calcium homeostasis (Bush and Simon, 2007; Dräger, 1985; Hong and Simon, 2007). Similarly, L-DOPA or the derived catecholamine compounds could participate in a reversible mechanism that would release calcium upon oxidation (Patrick Riley, personal communication). Lower calcium concentrations and altered endocochlear potentials have been recorded in albino guinea pigs (Gill and Salt, 1997), which is in agreement and would be predicted by our results. Deregulation of cochlear calcium homeostasis contributes to cochlear injury (Lahne and Gale, 2008; Shen et al., 2007), yet the correct cochlear calcium homeostasis and redox balance would be preserved in pigmented YRT2 transgenic mice and phenotypically albino TyrTH transgenic mice through L-DOPA production. The characteristic absence of pigment in albino individuals, along with their associated lack of intermediate pigment metabolites, such as L-DOPA, could therefore lead to abnormal calcium homeostasis in the endolymph. Such alterations would eventually affect normal auditory function, causing earlier age-related hearing loss and higher sensitivity to noise-induced hearing damage.

The data presented here demonstrate that impaired hearing responses during ageing and in response to insult can be detected in a mouse model of human OCA1. In the light of our results, it would be of interest to assess the relevance of these findings in the homologous human genetic condition.

Material and methods

  1. Top of page
  2. Summary
  3. Significance
  4. Introduction
  5. Results
  6. Discussion
  7. Material and methods
  8. Acknowledgements
  9. References
  10. Supporting Information

Mouse strains and handling

Two mouse strains genetically modified for the study of OCA1 were used here: the YRT2 (line #1999, Schedl et al., 1993) and TyrTH transgenic mice (line #6775, Lavado et al., 2006). Both these strains were generated and maintained on an albino mutant (Tyrc) outbred NMRI mouse background (NMRI/Hsd; Harlan Interfauna Iberica SL, Barcelona, Spain), and all the transgenic mice analyzed were hemizygous for the corresponding transgene. YRT2 transgenic mice carry the entire genomic region responsible for mouse Tyr expression within a yeast artificial chromosome and they are undistinguishable from wild-type agouti mice (Giraldo and Montoliu, 2002; Schedl et al., 1993). YRT2 mice were readily identified at birth through their obvious pigmentation pattern over the albino phenotype of the NMRI genetic background. TyrTH transgenic mice are externally albino but they ectopically express neuronal Th in melanocytes and retinal pigment epithelium cells and thus, they were identified at weaning by PCR. The Taq DNA polymerase (Roche) was used for PCR reactions as recommended by the manufacturer with the TyrTH1 5′-GCTTAGTGTAAAACAGGCTGAGAGTATTTG-3′ and TyrTH2 5′-AAGGGAGCAGATGAGTAGGGAGGGC-3′ oligonucleotides that specifically amplified a product of 0.6 kb. The reactions were initiated by a 4 min incubation at 94°C, followed by 40 cycles of 30 s at 94°C, 1 min at 66°C and 1 min at 72°C, and they were terminated with a 10-min extension at 72°C. Non-transgenic albino NMRI littermates from both transgenic lines (YRT2 and TyrTH) were used as controls for the experimental procedures. The number of mice used for this work was: 34 YRT2 transgenic mice, 23 TyrTH transgenic mice and 49 non-transgenic albino NMRI littermates, of ages ranging from 1.5 to 6 months old.

All the procedures that required the use of animals complied with Spanish and European legislation concerning vivisection and the use of genetically modified organisms, and the protocols were approved by the local Ethics Committees on Animal Experimentation.

Hearing evaluation

Hearing was assessed in transgenic YRT2, TyrTH and non-transgenic albino NMRI mice by ABR in the Non-invasive Neurofunctional evaluation Service (http://www.iib.uam.es/servicios/nine/en/intro.en.html) as described previously (Cediel et al., 2006; Murillo-Cuesta et al., 2009). In brief, the mice were anesthetized by intraperitoneal injection of ketamine (100 mg/kg) and xylazine (10 mg/kg) and then placed on a heated pad in a sound-attenuating chamber. Two different sound stimuli were designed with SigGenRPTM TDT software (Tucker-DavisTM Technologies, Alachua, FL, USA) and they were calibrated with a PS9200 ¼-inch microphone (ACO Pacifics Inc., Belmont, CA, USA) using SigCalRPTM TDT software. Click stimuli lasted 0.1 ms and tone-burst (4–28 kHz) stimuli lasted 5 ms (2.5 ms for the rise and decay with no plateau). These sound stimuli were generated by a TDT System III RP2.1 processor and a PA5 attenuator, and they were delivered with an electrostatic TDT free-field ES1 speaker located 5 cm from the pinnae. Electrical responses were recorded with three stainless steel needle electrodes subcutaneously placed at the vertex (active), ventrolateral to the right ear (reference) and in the lumbar region (ground). The responses were analyzed with BioSigRPTM TDT software and ABR waveforms were recorded in 5–10 dB steps down from the maximum intensity of 90 dB SPL to 25 dB SPL. The ABR threshold was defined as the stimulus level that evoked a peak-to-peak voltage 2 SDs above the mean background activity, and they were obtained for each click and tone burst stimulus as described previously (Cediel et al., 2006; Ngan and May, 2001; Zheng et al., 1999). Audiograms were obtained from the ABR thresholds observed in response to tone-burst stimuli. The click-ABR peak latencies (PL, I–V) and interpeak latencies (IPL, I-II and I-IV) were also determined. The sequence of peaks in the ABR recording reflects the synaptic activity of consecutive nuclei along the afferent auditory pathway in the brainstem. In the mouse, five positive (I-V) peaks have been identified, corresponding to the cochlea and auditory nerve, cochlear nucleus, superior olivary complex, lateral lemniscus and inferior collicullus, respectively (Galbraith et al., 2006; Parham et al., 2001).

Noise-induced hearing loss

To study the response of transgenic YRT2 and TyrTH mice, and of non-transgenic albino mice to noise, unrestrained and awake individuals were exposed to 12 h swept sine noise (0.5–20 kHz) at 95 dB SPL in an home-made and designed sound-proof reverberant chamber. The noise stimulus was designed with Wavelab Lite software (Wavelab Lite, Steinberg Media Technologies GMBH, Hamburg, Germany) and delivered through standard loudspeakers with an adequate frequency response range. ABR was performed before the noise insult to confirm normal hearing function, and at 24 h, 1, 2 and 3 weeks after the animals were subjected to the noise insult, to determine the magnitude of threshold shift.

Cochlear morphology and histological procedures

Adult mice were deeply anaesthetized and transcardially perfused with 4% paraformaldehyde in phosphate buffered saline (PBS; pH 7.4). Temporal bones containing the inner ear were removed and the cochlea were isolated, post-fixed in fresh fixative solution and decalcified for 2 weeks with 10% EDTA as described previously (Camarero et al., 2001). For each mouse, one cochlea was embedded in paraffin wax and the other was prepared for cryostat sectioning following standard procedures. Paraffin sections were obtained at 5 μm and used for general cytoarchitectonic studies (Nissl-staining using 1% cresyl violet: Camarero et al., 2002) and for immunohistochemistry with antibodies against the cell-cycle nuclear marker Ki67 (rabbit anti-Ki-67, Novocastra Laboratories Ltd., Newcastle upon Tyne, UK, dilution 1:500, Giménez et al., 2005) and the Na+K+-ATPase isoform β2 (rabbit anti-NaK-ATPase, Upstate Biotechnology Inc., Lake Placid, NY, USA, dilution 1:200, as described by Meyer zum Gottesberge, 1988). Cryostat sections were obtained at 15 μm and used for histochemistry with Alexa-488 conjugated phalloidin (Molecular Probes, Invitrogen, Life Technologies, Carlsbad, CA, USA, dilution 1:100) after permeabilization of the sections with 0.2% Triton X-100. Additional immunohistochemical studies on cryostat sections were performed to detect neurofilament (NF) 200 kDa (mouse anti-NF, Sigma-Aldrich Quimica SA, Madrid, Spain, dilution 1:100), synaptophysin (rabbit anti-synaptophysin, Dako, Denmark A/S, Glostrup, Denmark, dilution 1:100) as described previously (Camarero et al., 2001, 2002) and tyrosine hydroxylase (goat anti-Th, Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA, dilution 1:250, Lavado et al., 2006). Appropriate Alexa-488 or Alexa-546 conjugated secondary antibodies were used to detect primary antibody binding (Molecular Probes, dilution 1:400, for NF and synaptophysin, or 1:500, for Th). Finally, NF and synaptophysin stained sections were mounted in Vectashield containing DAPI (Vector, Burlingame, CA, USA) while the Th sections were mounted in Prolong containing TO-PRO3 (Molecular Probes) to visualize nuclei. Histological sections were prepared by the CNB Histology facility (http://www.cnb.csic.es/~histocnb).

Statistical analysis

Statistical analysis was carried out with SPSS (v.15; SPSS Inc, Chicago, Illinois, USA) software. The data obtained from the transgenic and non-transgenic mice were compared by a general linear model with repeated measures. Univariant (anova) and multivariate analysis of variance (manova) of ABR thresholds and latencies were carried out to compare the hearing parameters between the experimental groups at different time points. An analysis of the prevalence of normal hearing, and of moderate-severely deaf animals, among the whole population was also conducted. Post hoc analyses were performed with the Dunnett test when comparing both transgenic groups to the control non-transgenic group. In addition, independent sample t tests with a Bonferroni correction were applied to determine at which age there were significant differences in the group thresholds. Alternatively the Student–Newman–Keuls test was used when there were no a priori differences. An analysis of the prevalence of normal hearing, and of moderate-severely deaf animals, among the whole population was also conducted. Significance was set at P ≤ 0.05 and all values are presented as mean ± SEM.

Acknowledgements

  1. Top of page
  2. Summary
  3. Significance
  4. Introduction
  5. Results
  6. Discussion
  7. Material and methods
  8. Acknowledgements
  9. References
  10. Supporting Information

This work was supported by funds from: the Spanish Ministry of Education and Science (BFU2006-12185) and the ‘Comunidad de Madrid’ (GR/SAL/0654/2004) to LM; the ‘Comunidad de Madrid’ (IV PRICYT IC0530), DIGNA Biotech (Ref. 065510), MICINN (SAF2008–00470) and the ‘Fundación Investigación Médica Mutua Madrileña’ (Ref. 20070504) to IVN; and the CIBERER (INTRA/08/761,1) to IVN and LM. The authors are most grateful to Richard A. King, Glen Jeffery, Angela M. Meyer zum Gottesberge, Patrick Riley, Francisco Solano and Hortensia Sánchez-Calderón for their useful comments, to Laura Barrios for support with the statistical analysis, and to Lourdes Rodríguez de la Rosa, Antonio Rodríguez, Juan José Lazcano, Soledad Montalbán and Óscar Sánchez for technical assistance.

References

  1. Top of page
  2. Summary
  3. Significance
  4. Introduction
  5. Results
  6. Discussion
  7. Material and methods
  8. Acknowledgements
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Significance
  4. Introduction
  5. Results
  6. Discussion
  7. Material and methods
  8. Acknowledgements
  9. References
  10. Supporting Information

Figure S1. Morphology of the Organ of Corti in albino NMRI (A, C, E) and pigmented YRT2 (B, D, F) mice. (A and B) Actin-labeling of the hair bundles with phalloidin. (C and D) Neurofilament 200 kDa labeling of the afferent fibers (arrows) of the organ of Corti and the synapse region (arrowhead). (E and F) Synaptophysin labeling of presynapses in the inner hair cells and efferent fibers arriving at the inner hair cells (IHC, arrowhead) and outer hair cells (OHC, arrowhead) of the organ of Corti. No significant differences were found between genotypes. Scale bar 100 μm.

Figure S2. Na-K-ATPase expression in the stria vascularis of non-transgenic albino NMRI and transgenic (YRT2 and TyrTH) mice at different times after acoustic insult. Albino NMRI mice at (A) 24 h, (B) 3 days and (C) 1 month and transgenic mice (D–E: TyrTH; F: YRT2) at (D) 24 h; (E) 3 days; (F) 1 month. There were no morphological differences between the genotypes studied. Note the mild Na-K-ATPase expression in sections 24 h after acoustic insult.

Table S1. Interpeak latencies I–II and I–IV in response to click evoked stimuli at 90 dB SPL at two different ages. No statistically significant differences were found in the IPL I–IV (central conduction time) value between the genotypes, suggesting a peripheral hearing loss in the albino NMRI mice.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
PCMR_646_sm_TableS1_FigS1_S2.doc1479KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.