2.1.

Surgical Technique

The Baylor College of Medicine Institutional Animal Care and Use Committee approved the study protocol. C57BL/6 mice, weighing 20 to $30\mathrm{g}$ , age 4 to 8 weeks were anesthetized with an intraperitoneal (IP) injection of ketamine hydrochloride $(100\mathrm{mg}∕\mathrm{kg})$ and xylazine hydrochloride (5 to $10\mathrm{mg}∕\mathrm{kg}$ ). Body temperature was maintained at $39°\mathrm{C}$ using an electric heating pad (FHC, Bowdoinham, Maine). A retroauricular incision extending ventrally was made and the soft tissues divided to expose the tympanic bulla. The bulla was then opened and the round window membrane exposed. ABR thresholds were then measured (see the following).

The membranous structures within the cochlea are nearly transparent and are bathed in an aqueous environment. To enhance laser energy absorption and achieve optical selectivity for the basilar membrane, an exogenous chromophore was applied. We chose trypan blue (Gibco Laboratories, New York), a water-soluble stain that has maximal optical absorption at $607\mathrm{nm}$ , with an absorption coefficient $0.25{\mathrm{cm}}^{-1}$ . At the laser wavelength of $694\mathrm{nm}$ available to us (see details following), the absorption coefficient of trypan blue is³¹ less than $0.05{\mathrm{cm}}^{-1}$ . Trypan blue was diluted to 0.1% in artificial perilymph (containing, in mM: $130\mathrm{NaCl}$ , $4\mathrm{HCl}$ , $1{\mathrm{MgCl}}_2$ , $2{\mathrm{CaCl}}_2$ , 10 HEPES, 10 glucose). The pH was titrated to 7.3 and the osmolarity was $293\mathrm{mOsm}$ . A flexible micropipette (Microfil, World Precision Instruments, Sarasota, Florida) was inserted through the round window and directed up the duct of scala tympani about $1.5\mathrm{mm}$ . Trypan blue $\left(0.3\mathrm{ml}\right)$ was gently perfused in over $5\min$ . As the fluid drained out the round window, it was aspirated with a cotton wisp. Thus, only the basal region of the cochlea was stained. ABR thresholds were remeasured and the cochlea was then irradiated.

After laser irradiation, the incision was closed with sutures and the mouse was monitored until fully awake. The animal was then returned to its cage and maintained for another 14 to 16 days. At that point, the mouse was reanesthetized and the incision reopened. Any fluid within the middle ear space was aspirated and ABR thresholds were measured for a third time. The mouse was then sacrificed, perfused systemically with fixative, and its temporal bones harvested.

2.2.

Laser Irradiation

A 694-nm-long pulsed $\left(3\mathrm{ms}\right)$ ruby laser (Epilaser, Palomar Medical Technologies, Inc., Burlington, Massachusetts) was used to irradiate the cochlea. The light was coupled from the articulated arm of the laser into a $600\text{-}\mathrm{\mu }\mathrm{m}\text{-}\mathrm{core}\text{-}\mathrm{diam}$ fiber optic probe to facilitate light delivery to the basilar membrane through the exposed cochlea. Light output from the optical fiber was measured with an Ophir (30A, Wilmington, Massachusetts) power meter. It was estimated that the spot size was $1\mathrm{mm}$ in diameter at the cochlea. The radiant exposure applied at the focal spot was selected to be either $15\mathrm{J}∕{\mathrm{cm}}^2$ (one pulse of $15\mathrm{J}∕{\mathrm{cm}}^2$ ) or $180\mathrm{J}∕{\mathrm{cm}}^2$ (six pulses of $30\mathrm{J}∕{\mathrm{cm}}^2$ ) with a repetition rate of $1\mathrm{Hz}$ . We chose the $180\text{-}\mathrm{J}∕{\mathrm{cm}}^2$ dosage because it was the lowest dose used in our previous ex vivo study that caused statistically significant decreases in basilar membrane birefringence.²⁷ We chose the $15\text{-}\mathrm{J}∕{\mathrm{cm}}^2$ dose as the lowest in this set of experiments because it is a dose commonly used in clinical practice to induce collagen changes for skin rejuvenation.^{32, 33}

2.3.

Measurement of ABR Thresholds

Sine wave stimuli were generated using a digital signal processing system. The stimuli were attenuated to the appropriate intensity, and delivered acoustically to the ear by an electrostatic speaker (RP-2, PA-5, ED-1, and EC-1 Tucker-Davis Technologies). The speaker was calibrated across the frequency range of 1 to $80\mathrm{kHz}$ prior to each experiment using a probe-tip microphone (microphone type 8192, NEXUS conditioning amplifier, Bruel and Kjar, Demark). This was performed after connecting the microphone and speaker to an earbar inserted into the animal’s ear canal.

The ABR was recorded using needle electrodes positioned at the vertex of the skull and along the ventral surface of the tympanic bulla. A ground electrode was placed in the hind leg. The signals were amplified 10,000 times using a biological amplifier (HS4/DB4, Tucker-Davis Technologies, Alachua, Florida), digitized at $100\mathrm{kHz}$ (RP-2, Tucker-Davis Technologies), and digitally filtered to pass frequency components in the 300 to 3000-Hz range.

The stimulus for eliciting the ABR was a 5-ms sine wave tone with ${\cos }^2$ envelope rise and fall times of $0.5\mathrm{ms}$ . The repetition time was $50\mathrm{ms}$ and 250 trials were averaged. The peak-to-peak ABR signal was measured at stimulus intensities ranging from 10 to $80\mathrm{dB}$ sound pressure level (SPL) in 10-dB steps for each frequency. These data were interpolated using a cubic spline algorithm (MATLAB, Release 13, The Mathworks, Natick, Massachusetts) off-line to identify where the signal crossed the average noise floor plus four standard deviations to determine frequency-specific thresholds. We measured ABR thresholds at frequencies between 4 to $80\mathrm{kHz}$ . If no ABR was detected even at our equipment limits of $80\mathrm{dB}$ SPL, we arbitrarily defined the threshold to be $80\mathrm{dB}$ when calculating average thresholds.

Statistical analysis was performed using Excel (Microsoft, Redmond, Washington). ABR thresholds in the same mice before and after trypan blue perfusion were compared using the Student’s two-tailed paired $t$ test. The mean ABR thresholds in each group of irradiated mice were compared to each other and to ABR thresholds of all the mice before trypan blue perfusion using the Student’s two-tailed nonpaired $t$ test. Statistical significance was defined as $p<0.05$ . All presented values are mean $\pm$ SEM (standard error of the mean).

2.4.

Polarized Light Microscopy

After harvesting the cochleas, they were immersed in 4% paraformaldehyde for 2 days, decalcified in 0.2-M EDTA for 2 weeks, embedded in paraffin, and cut perpendicular to the organ of Corti in $7\text{-}\mathrm{\mu }\mathrm{m}$ sections. To enhance the native birefringence of collagen, the sections were stained with picrosirius red (Sigma Aldrich, St. Louis, Missouri). Picrosirius red is a strongly elongated, birefringent molecule that binds parallel to collagen molecules. ^{21, 34, 35, 36, 37} It was originally used in the textile industry to analyze fiber orientation,³⁸ and it has been used to assess collagen organization^{35, 39} since the 1960s. We additionally stained the sections with hemotoxylin (Thermo Shandon, Pennsylvania) to visualize regions with minimal collagen.

Once all of the cochlear cross sections were cut and stained, they were visualized using polarized light microscopy to highlight parallel arrays of collagen. Before starting this procedure, one control specimen was placed on the microscope stage and the two polarizers were rotated to maximize the birefringence of the tissue. Once this was done, the polarizers were locked into place and all of the specimens were studied in the same anatomic orientation. Thus, polarization settings were held constant for all sections. Images were captured digitally using a 6.3-Mpixel camera (Canon, EOS 10D, Tokyo 146-8501, Japan) connected to an upright microscope (Zeiss, Axioskop 2, Germany). The applied light, aperture size, and time exposure were also held constant for all images.

ImageJ [National Institute of Health (NIH), http://rsb.info.nih.gov/ij/] was used to quantify the intensity of the birefringence within different collagen-containing areas of the cochlea. The color images of each turn of each cochlea were converted to 8-bit gray scale. After conversion, areas of high birefringence were white and areas of low birefringence were gray or black. Within each region to be analyzed, the area with the strongest birefringence was chosen and the signal intensity determined by averaging at least 60 pixels. The two-tailed nonpaired Student’s $t$ test was used to compare the birefringence between different regions and laser dosages.

2.5.

Transmission Electron Microscopy

The harvested cochleas were fixed in 2.5% glutaraldehyde in 0.1-M cacodylate buffer containing 2-mM $\mathrm{CaCl}$ at a pH 7.4, and decalcified in 0.2-M EDTA for 2 weeks. They were then washed several times with 0.1-M cacodylate buffer. The samples were postfixed in 1% osmium tetroxide in 0.1-M cacodylate buffer and then washed with cacodylate buffer. Gradual dehydration was performed from 25 to 50% ethanol and the cochleas were then stained with 2% uranyl acetate in 50% ethanol for an hour. The cochleas were further dehydrated in 100% ethanol, followed by the transitional dehydrating agent propylene oxide.

The cochleas were gradually infiltrated with propylene oxide resin (Durucupan ACM Resin). This process was repeated using progressively higher resin concentrations until pure resin was infiltrated. The cochleas were oriented in resin-filled blocks, cured for 2 days at 55°, and then cut into 80-nm thin sections with an ultramicrotome. The sections were stained with 2% uranyl acetate in 50% ethanol for $7\min$ and with Reynold’s lead citrate for $1\min$ . The specimens were imaged using a transmission electron microscope (H-7500, Hitachi).

2.6.

Animals

We studied a total of 18 mice. The first 11 mice were used to develop the technique of in vivo laser irradiation. From these 11 animals, we present only the ABR data collected prior to laser irradiation (i.e., after the surgical exposure of the cochlea and after perfusion of trypan blue into the cochlea). After the laser irradiation procedure was perfected, four mice were irradiated at $15\mathrm{J}∕{\mathrm{cm}}^2$ and 3 mice were irradiated at $180\mathrm{J}∕{\mathrm{cm}}^2$ . One cochlea from each of these groups was prepared for electron microscopy analysis, and the others (three and two cochleas, respectively) underwent ABR threshold testing and polarized light microscopy analysis.

Four contralateral cochleas were used as controls. They were not operated on, and so were not stained with trypan blue or irradiated. One was studied with electron microscopy and three were studied with polarized light microscopy.

3.1.

Polarized Light Microscopy

Polarized light microscopy revealed that the birefringence of the basilar membrane of the basal turn 2 weeks after irradiation was higher than in controls (Fig. 2 ). In a cochlea irradiated at $15\mathrm{J}∕{\mathrm{cm}}^2$ , the organ of Corti architecture appeared normal 2 weeks following laser irradiation. Both the inner and outer hair cells were maintained throughout the cochlea at this laser dosage. At $180\mathrm{J}∕{\mathrm{cm}}^2$ , the basilar membrane demonstrated even more birefringence than that at the lower laser dosage. However, intracochlear fibrosis and inflammatory tissue filled the scalae. The cells that normally make up the organ of Corti (hair cells and supporting cells) were not evident.

Fig. 2

Organ of Corti. Paraffin-embedded cochlear cross sections stained with picrosirius red and hemotoxylin were visualized using polarization microscopy to highlight collagen arrays. (A) A representative control section (cochlea not perfused or irradiated). The normal basilar membrane birefringence can be noted (*). Inner and outer hair cells are also visible (arrows). (B) A representative section of a cochlea 14 days after the laser irradiation with $15\mathrm{J}∕{\mathrm{cm}}^2$ . Note the continued presence of hair cells (arrows) and the normal architecture of the organ of Corti. (C) A representative section of a cochlea 14 days after irradiation with $180\mathrm{J}∕{\mathrm{cm}}^2$ . The basilar membrane birefringence appears brighter (more yellow) than that of the control cochlea. Hair cells are no longer present and scar tissue fills the intracochlear scalae (arrowheads).

We quantified the tissue birefringence from four different regions in each cochlea (Fig. 3 ). The basilar membrane from the basal turn near the round window (BM-base) was the area stained with trypan blue and was directly in the path of the laser beam. The basilar membrane of the apical turn (BM-apex) was likely not stained by the trypan blue. The stria vascularis (SV-base) is the tissue along the lateral wall of the cochlea that produces the electrochemical gradients within the scala media necessary for normal cochlear function. The osseous spiral lamina (OSL-base) is the bone to which the basilar membrane makes its medial attachment. We only analyzed the stria vascularis and osseous spiral lamina within the basal turn of the cochlea.

Fig. 3

Quantification of collagen birefringence. Measurements were taken from the basilar membrane in the basal turn (BM-base), the basilar membrane in the apical turn (BM-apex), the stria vascularis in the basal turn (SV-base), and the osseous spiral lamina in the basal turn (OSL-base). The intensity of birefringence is the average pixel intensity on a scale of 0 to 255. At both 15- and $180\text{-}\mathrm{J}∕{\mathrm{cm}}^2$ laser dosages, there were statistically significant increases in the collagen birefringence in the BM-base compared to controls (*). Additionally, the collagen birefringence in the BM-base after $180\mathrm{J}∕{\mathrm{cm}}^2$ was larger than that after $15\mathrm{J}∕{\mathrm{cm}}^2$ (**). No changes occurred in the other regions of the cochlea.

We found statistically significant increases in the average birefringence of the BM-base in cochleas irradiated with either laser dosage compared to controls ( $173\pm 3$ , $196\pm 0.4$ , and $222\pm 2$ for controls, $15\mathrm{J}∕{\mathrm{cm}}^2$ , and $180\mathrm{J}∕{\mathrm{cm}}^2$ , respectively; $P<0.05$ ). Additionally, the effect was dose dependent in that the mean birefringence of cochleas irradiated at $180\mathrm{J}∕{\mathrm{cm}}^2$ was larger than that of the cochlea irradiated at $15\mathrm{J}∕{\mathrm{cm}}^2$ $(P<0.05)$ . There were no statistically significant changes in tissue birefringence in the BM-apex, SV-base, or OSL-base for cochleae irradiated with either laser dosage compared to controls.

3.2.

Transmission Electron Microscopy

Transmission electron microscopy of the basilar membrane of a control cochlea revealed the expected normal architecture (Fig. 4 ). There were large collagen fibers oriented in a radial fashion with loose connective tissue and ground substance between them. The space between the fibers was filled with white vacuoles. This indicates that the material within these spaces leeched out of the basilar membrane during fixation, and thus did not consist of collagen.

Fig. 4

Transmission electron microscopy of the basilar membrane. A cross section of the basilar membrane from the basal turn is shown. (a) A representative control section demonstrates the normal organization of the basilar membrane. There are collagen fibers (*) separated by loose connective tissue, which appear predominantly as white vacuoles. (b) Fourteen days after irradiation with $15\mathrm{J}∕{\mathrm{cm}}^2$ . The collagen fibers are slightly denser than in the control section. There are also an increased amount of thin collagen fibrils within the loose connective tissue between the fibers (arrow). (c) Fourteen days after irradiation with $180\mathrm{J}∕{\mathrm{cm}}^2$ . At this level of irradiation, the collagen fibers remain denser than in the control section. In many areas, the fibers seem wider with less crisp borders. Thus, there is less space between adjacent collagen fibers and almost no vacuolated space. This is due to the substantial amount of thin collagen fibrils within the loose connective tissue between fibers (arrow).

In the irradiated cochleas, the collagen fibers were denser than in controls, suggesting that remodeling of the existing collagen had occurred. Also, there were substantial numbers of new collagen fibrils within the loose connective tissue regions between the fibers. These fibrils were irregularly organized in between adjacent fibers. However, the new fibrils also appeared to condense around the borders of the preexisting collagen fibers. This may perhaps reflect the early incorporation of these new collagen fibrils into the fibers. Thus, there was a reduced amount of the vacuolated space between the fibers after $15\mathrm{J}∕{\mathrm{cm}}^2$ , and almost no vacuolated space after $180\mathrm{J}∕{\mathrm{cm}}^2$ . Together, these findings are consistent with new collagen deposition within the basilar membrane.^{40, 41}

3.3.

ABR Threshold Shifts

ABR thresholds were measured after opening the tympanic bulla, following perfusion of the cochlea with trypan blue, and just prior to sacrifice 14 to 16 days later (Fig. 5 ). Normal thresholds were found after opening the bulla, as expected. There were no changes in ABR thresholds after trypan blue perfusion, except for 4 to $6\text{-}\mathrm{dB}$ increases in the ABR threshold at 22 and $52\mathrm{kHz}$ that were statistically significant. This indicates that the perfusion procedure and the chromophore application had only minimal effects on cochlear function. This finding is consistent with data from the guinea pig cochlea, in which only a slight high-frequency threshold shift develops when artificial perilymph is perfused through scala tympani in a careful fashion.^{43, 44} Presumably, this hearing loss reflects inadvertent cochlear trauma or cooling of the cochlea because of its exposure to the outside environment.

Fig. 5

ABR thresholds. After initially opening the bulla, the mice had normal thresholds across the frequency spectrum. After perfusing trypan blue through the round window, there were only statistically significant increases in the ABR threshold at 22 and $52\mathrm{kHz}$ . Two weeks after irradiating with $15\mathrm{J}∕{\mathrm{cm}}^2$ , there were 17 to 27-dB increases in ABR thresholds that were statistically significant compared to baseline at every frequency except at $80\mathrm{kHz}$ . After irradiating with $180\mathrm{J}∕{\mathrm{cm}}^2$ , there were 32 to 59-dB increases in ABR thresholds across the frequency spectrum compared to baseline. SPL was defined using a standard reference that is approximately the intensity of a 1000-Hz sinusoid that is just barely audible to a human with normal hearing⁴² $\left(20\mathrm{\mu }\mathrm{Pa}\right)$ . Postirradiation thresholds that were statistically elevated from baseline are marked (*). Thresholds statistically higher after 180 compared to $15\mathrm{J}∕{\mathrm{cm}}^2$ are marked (**).

Two weeks after laser irradiation with $15\mathrm{J}∕{\mathrm{cm}}^2$ , there were increases in ABR thresholds ranging from 17 to $27\mathrm{dB}$ that were statistically significant at every frequency except at $80\mathrm{kHz}$ . After laser irradiation with $180\mathrm{J}∕{\mathrm{cm}}^2$ , there were larger increases in ABR thresholds ranging from 32 to $59\mathrm{dB}$ across the frequency spectrum. Statistically significant increases in ABR thresholds were found in mice irradiated with $180\mathrm{J}∕{\mathrm{cm}}^2$ compared to those irradiated with $15\mathrm{J}∕{\mathrm{cm}}^2$ at 4, 6, 14, and $22\mathrm{kHz}$ that ranged from 22 to $30\mathrm{dB}$ .