2.1.

Animals

85 female CD1 albino mice, weighing 20 to 30 g at the age of 5 to 8 weeks were used for this study. Out of these, the temporal bones of 74 animals were processed for the analyses with transmitted light and polarized light microscopy, and 11 animals for transmission electron microscopy (TEM). Auditory brainstem responses (ABRs) were recorded to monitor hearing.

2.2.

Surgical Technique

The “Saarländische Landesamt für Verbraucherschutz” approved the study protocol in accordance with the “Deutschen Tierschutzgesetz und der Richtlinie des Rates der europäischen Gemeinschaft zum Schutz der für experimentelle Zwecke verwendeten Tiere.”

The mice were anesthetized with an intraperitoneal injection ($0.1\mathrm{ml}/\mathrm{g}$ BW) of 2.125 ml 0.9% isotone solution of sodium chloride, 0.25 ml ketamine, and 0.125 ml rompun. During the experiment, the body temperature was maintained at 38°C using an electric heating pad.

After trimming the hair around the ear, we performed a 2-mm long incision beginning at the incisura intertragica and extending it along the cartilaginous outer ear canal exposing the TM on both sides. After recording ABR, the external ear was anchored with sutures to expose the entire TM.

To enhance the laser energy absorption and achieve optical selectivity for the proposed defined part of the TM, we stained the target area of the posterior (post) superior quadrant (diameter ca. $400\mu \mathrm{m}$) with the pigment Fuchsin (0.5% solution with alcohol, pharmacy, Saarland University), a water-soluble solution that has its maximal optical absorption at 455 to 600 nm.^27,28 We stained the contralateral ear as well as a control. The anterior (ant.) part of the TM was, on both sides, not touched. After staining, we irradiated the marked area on the left TM (Fig. 1).

The ABR thresholds were then recorded again and the incisions on both ears sutured. Until fully awake, the mouse was monitored and put under an infrared lamp to protect it from hypothermia.

After 14, 28, or 56 days, according to the respective test group, the mice were anesthetized again, the incisions on both ears reopened and final ABR thresholds were measured. The animals were then sacrificed through cervical dislocation under deep anesthesia. The petrous part of the temporal bone was extracted ex vivo and the tympanic cavity opened and perfused with fixative.

2.3.

Laser Irradiation

We used a 532-nm pulsed neodynium-doped yttrium orthovanadate ($\mathrm{Nd}:{\mathrm{YVO}}_4$) (Xiton Photonics GmbH, Kaiserslautern) as the laser source. The pulsed laser light was applied through a laser fiber with a diameter of $365\mu \mathrm{m}$ that we directed to the stained area of the TM with a micromanipulator. We estimated the distance as close as visible possible to the stained area, without touching the TM, giving an irradiation diameter of $\sim 400\mu \mathrm{m}$ on the TM. The final position adjustment has been performed using a continuous waved pilot laser, with a power of 0.1 mW, for a maximum of 30 s (Fig. 1). We then irradiated the stained area of the TM for 30 s with a power of either 10 mW (0.3 J), 25 mW (0.75 J), or 50 mW (1.5 J) at a laser repetition rate of 2 kHz (Table 1).

Table 1

Summary of the analyzed animals with focus onto the irradiation duration, intensity, and total dose of laser energy applied as well as the day of sacrifice.

2.4.

Electrophysiological Monitoring: ABR

We used a digital signal processing system that generated sine wave stimuli (Agilent 33500 B Series Trueform Waveform Generator, Keysight Technologies GmbH, Germany) and delivered them through a free-field speaker (Beyerdynamic GmbH & Co. KG) placed in front of the mouse at 5 cm distance to each ear. With this system, we recorded click- and frequency-specific ABRs using four needle electrodes: one on each mastoid, one at the vertex (reference), and one at the base of the tail (ground). The recorded signals were then amplified through the biosignal amplifier (g.USBamp, g.tec medical engineering GmbH, Austria), digitized at 19.2 kHz and digitally filtered to obtain the frequencies in the 300 to 2500 Hz range. The stimulus intensities ranged from 0 to 80 dB SPL increased in 10 dB steps at 2, 4, 8, 12, 16, 20, 32, 48, and 80 kHz.

We estimated the hearing thresholds for each recording visually, offline and defined it as the lowest intensity, where the Jewett complex was identifiable. Additionally, within the click ABR data, we measured the amplitude and latency of the wave I peak to peak.²⁹

2.5.

Transmitted Light Microscopy

We explanted the temporal bones, opened the tympanic cavity, and fixated the samples with a 4% formaldehyde for 2 days. We performed the decalcification in 20% EDTA with 100% citric acid (pH 7.2) for 4 days, dehydrated them in ascending ethanol series of 70% and 100% (360 min), placed them in 100% Xylol (180 min), and finally embedded them in paraffin. Afterward, we cut the blocks in $8\mu \mathrm{m}$ sections and stained with hematoxylin–eosin to get a first impression of the anatomy. We then used the picrosirius red staining (Sigma-Aldrich Chemie GmbH, München) (MORPHISTO^® Evolutionsforschung und Anwendung GmbH, Frankfurt) to analyze the morphological changes within the TM.

For transmission microscopy, we used three to four slides of each ear. The interval between each utilized slide-series was $\sim 210\mu \mathrm{m}$. We then measured the thickness of the posterior part of the TM from the left (irradiated) and the right (nonirradiated) ear as a control (Fig. 1 dashed squares).

2.6.

Polarized Light Microscopy

We used picrosirius red and added hemotoxylin to enhance the natural birefringence of collagen and visualize regions with just a minimal quantity of collagen.^30,31 Picrosirius red is an elongated, birefringent molecule, which binds parallel to collagen molecules being used to identify and characterize collagen structures within biological tissues.^{17,30,32–34}

The sections were then visualized and imaged under polarized light microscopy. To ensure that the settings of the microscope were always the same, we started with a control ear and rotated the polarizers until the birefringence of the tissue was maximized. Furthermore, the remaining microscope settings were optimized. Once this was done, the polarizers were locked and the light exposure, the aperture size, the position were also held constant for all the other slides and images captured.

We used ImageJ (National Institutes of Health) to quantify the highest intensity of birefringence analyzing areas of 3000 to 4000 pixels within the irradiated and the corresponding control TM.

2.7.

Transmission Electron Microscopy

For a more detailed analysis of the structural changes within the TM, 4 weeks after irradiation, we used TEM. The explanted petrous bones were fixed for 1 day in 1% glutaraldehyd in 0.1 M cacodylate buffer containing 1% paraformaldehyde and ${\mathrm{H}}_2\mathrm{O}$ at a pH 7.2 to 7.4 and afterward decalcified in 20% EDTA for 6 days, washed with 0.1 M cacodylat buffer, and fixed with 0.2% osmium tetroxide. As a next step, the samples were washed in distillated ${\mathrm{H}}_2\mathrm{O}$ and gradually dehydrated from 80% to 100% ethanol and three times stained with acetone, two times with epon (3:1, 1:1, 1:3), and subsequently embedded in pure epon followed by polymerization for 2 days at 60°C. After cutting them in about 70-nm thin sections with an ultramicrotome (LEICA EM UC7, © 2015 Leica Biosystems, Nussloch), they were positioned on a copper grid, contrasted with 2% uranyl acetate, postcontrasted with lead citrate, and analyzed with a transmission electron microscope (FEI TECNAI 12, The Netherlands).

2.8.

Statistical Methods

We used SPSS and performed all tests at an alpha level of 0.05. The hearing thresholds for the click stimulation as well as for the frequency-specific stimulation, as described in Sec. 2.2, were analyzed with a nonparametric test (Friedman). Additionally, we conducted a repeated measure analysis of variance (rmANOVA) for the factor “power” and an rmANOVA for the factor “weeks” using Huynh–Feldt correction for the degrees of freedom. For the analysis of wave I of the ABR recordings, the TM thickness, and the TM polarization, we conducted a one factorial ANOVA, followed by a Bonferroni posthoc test to compare the individual groups. We also performed a Levene test of variance homogeneity. If this turned to be significant, we performed a more robust test, the Welch ANOVA, followed by a Games-Howell procedure as posthoc test. Additionally, in the cases when the data were not distributed normally, we performed a nonparametric test, the Friedman test, followed by a Bonferroni posthoc test.

3.1.

Transmission Light Microscopy

The thickness of the TM increased with increasing intensity of irradiation (Fig. 2). This demonstrated to be valid and statistically significant for the animals exposed to 25 mW for 30 s (0.75 J) and 50 mW for 30 s (1.5 J) independent of the time postexposure. The animals exposed to 10 mW for 30 s (0.3 J) did not have any increase in TM thickness in comparison to the contralateral nonexposed ear. To quantify these changes, we subtracted from the thickness of the stained and irradiated area of the (left) TM, the thickness of the same area of the stained nonirradiated (right) ear in each animal, and plotted the averaged results of these differences (Fig. 3). The maximal increase in TM thickness could be observed for the group of mice irradiated with 50 mW for 30 s (1.5 J) and sacrificed after 2 weeks. Even though the trend was toward an increase in thickness with the increase in optical power applied, this change was, however, statistically significant just between 10 mW for 30 s (0.3 J) and 25 mW for 30 s (0.75 J) or 10 mW for 30 s (0.3 J) and 50 mW for 30 s (1.5 J) but not significant between 25 mW for 30 s (0.75 J) and 50 mW for 30 s (1.5 J). This effect could be observed at all time points of sacrifice 2, 4, or 8 weeks.

Fig. 2

Transmission light microscopy. Four weeks after irradiation with (a) $5\mu \mathrm{J}/\text{pulse}$ (10 mW for $30\mathrm{s}\sim 0.3\mathrm{J}$); (c) $12.5\mu \mathrm{J}/\text{pulse}$ (25 mW for $30\mathrm{s}\sim 0.75\mathrm{J}$); (e) $25\mu \mathrm{J}/\text{pulse}$ (50 mW for $30\mathrm{s}\sim 1.5\mathrm{J}$); (a), (c), (e) (upper posterior quadrant of the irradiated left TM) in comparison with the contralateral ear; (b), (d), (f) (upper posterior quadrant of the nonirradiated right TM).

Fig. 3

Comparison between the analyzed mice groups presenting their TM thickness of section (2) (for reference, see Fig. 1) plotted on the $y$ axis relative to the irradiation levels of 10 mW for $30\mathrm{s}\sim 0.3\mathrm{J}$, 25 mW for $30\mathrm{s}\sim 0.75\mathrm{J}$ or 50 mW for $30\mathrm{s}\sim 1.5\mathrm{J}$ ($x$ axis) for the different incubation times of (a) 2, (b) 4, and (c) 8 weeks.

The mice who were sacrificed after 2 weeks demonstrated a significant increase in the TM-thickness between the groups irradiated with 10 mW for 30 s (0.3 J) compared to the once irradiated with 25 mW for 30 s (0.75 J) ($p=0.007$) and 50 mW for 30 s (1.5 J) ($p=0.000$), but no significant change between the mice irradiated with a power of 25 mW for 30 s (0.75 J) to the group irradiated with a power of 50 mW for 30 s (1.5 J) ($p=0.090$).

The mice sacrificed after 4 weeks demonstrated a significant increase of the TM thickness between the group irradiated with 10 mW for 30 s (0.3 J) to the groups irradiated with 25 mW for 30 s (0.75 J) ($p=0.001$) and 50 mW for 30 s (1.5 J) ($p=0.002$). Again, no significant TM thickness increase could be observed between the mice exposed to 25 mW for 30 s (0.75 J) and the ones exposed to 50 mW for 30 s (1.5 J) ($p=0.165$).

Similar to the mice sacrificed after 4 weeks, the mice sacrificed after 8 weeks demonstrated a significant change in TM thickness between the 10 mW for 30 s (0.3 J) irradiated mice to the 25 mW for 30 s (0.75 J) ($p=0.000$), and 50 mW for 30 s (1.5 J) ($p=0.002$) animals, however, no significant change in the TM thickness of mice exposed to 25 mW for 30 s (0.75 J) compared the ones exposed to 50 mW for 30 s (1.5 J) ($p=0.305$) (Fig. 3).

Additionally, we compared the sections cranial (1) and caudal (3) to the stained and irradiated area (2) of the TM (Fig. 1). The analysis demonstrated that a tissue proliferation was induced just in the area of the stained and irradiated part of the TM ($p<0.005$). The parts above and underneath this area were not significant different to their stained and nonirradiated controls (contralateral ear) ($p=1.000$).

3.2.

Polarized Light Microscopy

We analyzed the TMs birefringence as an estimation of the build/remodeled collagen using polarized microscopy, as described in Sec. 2.6 (Fig. 4), and compared the birefringence of the stained and irradiated area of the left TM with the same area of the control-right TM (stained and not irradiated).

Fig. 4

Polarized light microscopy. Four weeks after irradiation of the upper posterior quadrant of the stained and irradiated left TM with (a) $5\mu \mathrm{J}/\text{pulse}$ (10 mW for $30\mathrm{s}\sim 0.3\mathrm{J}$); (c) $12.5\mu \mathrm{J}/\text{pulse}$ (25 mW for $30\mathrm{s}\sim 0.75\mathrm{J}$); (e) $25\mu \mathrm{J}/\text{pulse}$ (50 mW for $30\mathrm{s}\sim 1.5\mathrm{J}$) (b), (d), (f) in comparison with the upper posterior quadrant of the stained nonirradiated right TM (contralateral ear).

The mice irradiated with 25 mW for 30 s (0.75 J) or 50 mW for 30 s (1.5 J) demonstrated after 4 and 8 weeks an increase of the TM birefringence. The statistical analysis demonstrated that this increase was statistical significant 4 weeks postirradiation: ($p=0.045$) for 25 mW for 30 s (0.75 J) and ($p=0.019$) for 50 mW for 30 s (1.5 J), however, not significant between the animals irradiated with 25 mW for 30 s (0.75 J) to the 50 mW for 30 s (1.5 J) ($p=1.000$).

The incubation time of 2 weeks did not induce a statistical significant change in the TM-polarization either ($p=0.568$) (Fig. 5).

Fig. 5

Comparison between the analyzed mice groups presenting their TM birefringence (difference) of section (2) Legend: difference in birefringence plotted on the $y$ axis relative to the irradiation value of 10 mW for $30\mathrm{s}\sim 0.3\mathrm{J}$, 25 mW for $30\mathrm{s}\sim 0.3\mathrm{J}$ or 50 mW for $30\mathrm{s}\sim 0.3\mathrm{J}$ ($x$ axis) for the different incubation times of (a) 2, (b) 4, and (c) 8 weeks.

Within the mice groups, which were sacrificed after 8 weeks, a significant increase in birefringence could be demonstrated between the mice group irradiated with 10 mW for 30 s (0.3 J) and the ones irradiated with 25 mW for 30 s (0.75 J) ($p=0.024$) as well, however, no significant change between the 25 mW for 30 s (0.75 J) irradiation and the ones irradiated with 50 mW for 30 s (1.5 J) ($p=1.000$). Interestingly, despite the clear trend toward increase in birefringence with increasing energy, the statistical analysis revealed no significant increase in birefringence between the mice irradiated with 10 mW for 30 s (0.3 J) and 50 mW for 30 s (1.5 J) ($p=0.135$). After exposing the stained TMs to 10 mW for 30 s (0.3 J), the increase of birefringence was not statistically significant either (Fig. 5).

Over time the increase in birefringence was significant within the first 4 weeks for the animals irradiated with 25 mW for 30 s (0.75 J) ($p=0.001$) and 50 mW for 30 s (1.5 J) ($p=0.000$) followed by a stagnation within the next 4 weeks (nonsignificant increase in birefringence in the mice irradiated with a power of 25 mW for 30 s (0.75 J) ($p=0.612$) and 50 mW for 30 s (1.5 J) ($p=1.000$)). The animals irradiated with 10 mW for 30 s (0.3 J) had no significant increase in birefringence ($p=0.773$).

As expected, a significant increase in birefringence was observed in the main stained and irradiated Sec. 2 of the TM ($p=0.003$ when comparing the irradiated to the nonirradiated control and $p=0.029$ for the difference between Secs. 1 and 2 and $p=0.002$ for the comparison of Secs. 2 and 3). The birefringence within the area of the outer Secs. 1 and 3 did not differ significantly ($p=0.831$).

3.3.

Transmission Electron Microscopy

The nonirradiated TMs [Figs. 6(b), 6(d), and 6(f)] revealed the typical three layered structure.^7,8 Within the TM irradiated with 10 mW for 30 s (0.3 J) [Fig. 6(a)], we observed a slight change in the radial organized collagen layer, which seemed to be arranged more loosely and was slightly interspersed with circular collagen fibers. Within the TM irradiated with 25 mW for 30 s (0.75 J) [Fig. 6(c)], the new circular structured collagen fibers appeared to be at least doubled compared to the contralateral, nonirradiated TMs. In addition, the ratio between the different collagen layers appeared modified with a larger increase in the circular structured collagen. Within the TM irradiated with 50 mW for 30 s (1.5 J) [Fig. 6(e)], the reorganized collagen layers demonstrated a further increase in thickness and the delimitation between the different organized collagen layers was loose.

Fig. 6

Transmission electron microscopy. Four weeks after irradiation with (a) $5\mu \mathrm{J}/\text{pulse}$ (10 mW for $30\mathrm{s}\sim 0.3\mathrm{J}$); (c) $12.5\mu \mathrm{J}/\text{pulse}$ (25 mW for $30\mathrm{s}\sim 0.75\mathrm{J}$); (e) $25\mu \mathrm{J}/\text{pulse}$ (50 mW for $30\mathrm{s}\sim 1.5\mathrm{J}$); (a), (c), (e) (upper posterior quadrant of the irradiated left TM) in comparison with the contralateral ear; (b), (d), (f) (upper posterior quadrant of the nonirradiated right TM).

3.4.

Measurement of ABR

The click-ABR of each mouse demonstrated no significant threshold shift directly after irradiation. With our recording setup, no significant difference between the irradiated and the contralateral nonirradiated ear ($p=0.687$) could be recorded. The final click-ABR recordings demonstrated a threshold shift of 10 to 40 dB in 56 of 75 mice that was as well observed for both ears (the irradiated and the nonirradiated ears).

The frequency-specific ABRs demonstrated similar results as the click-ABRs (Table 2). Twenty-five mice had in 1 to 2 frequencies a threshold shift at the second and third measurement compared to the first, both ears were affected. Nine mice were deaf from the start. The amplitude and latency of wave I did not present a major difference between the irradiated and the contralateral, not irradiated ear ($p=0.686$).

Table 2

Frequency-specific ABR: significance analysis for the individual frequencies.