Sonoporation of the Round Window Membrane on a Sheep Model: A Safety Study

Abstract

Sonoporation using microbubble-assisted ultrasound increases the permeability of a biological barrier to therapeutic molecules. Application of this method to the round window membrane could improve the delivery of therapeutics to the inner ear. The aim of this study was to assess the safety of sonoporation of the round window membrane in a sheep model. To achieve this objective, we assessed auditory function and cochlear heating, and analysed the metabolomics profiles of perilymph collected after sonoporation, comparing them with those of the control ear in the same animal. Six normal-hearing ewes were studied, with one sonoporation ear and one control ear for each. A mastoidectomy was performed on both ears. On the sonoporation side, Vevo MicroMarker^® microbubbles (MBs; VisualSonics-Fujifilm, Amsterdam, The Netherlands) at a concentration of 2 × 10⁸ MB/mL were locally injected into the middle ear and exposed to 1.1 MHz sinusoidal ultrasonic waves at 0.3 MPa negative peak pressure with 40% duty cycle and 100 μs interpulse period for 1 min; this was repeated three times with 1 min between applications. The sonoporation protocol did not induce any hearing impairment or toxic overheating compared with the control condition. The metabolomic analysis did not reveal any significant metabolic difference between perilymph samples from the sonoporation and control ears. The results suggest that sonoporation of the round window membrane does not cause damage to the inner ear in a sheep model.

Keywords: inner ear; metabolomics; perilymph; round window membrane; sonoporation.

Figure 1

Study protocol. (1) 6 sheep under general anesthesia. (2) Plasma sample after anesthesia (before sonoporation). (3) Verification of NH status using ABR. (4) Sonoporation ear: a. Mastoidectomy, b. Measure of promontory temperature before sonoporation, c. Sonoporation of ultrasound activated MBs injected into the middle ear, and d. Verification of promontory temperature after sonoporation. (5) Verification of hearing thresholds. (6) Perilymph and plasma sampling. (7) Control ear: mastoidectomy. (8) Perilymph sampling before sacrifice of the animal. (9) Metabolomic analysis of the samples.

Figure 2

Visualization of the round window membrane (RWM) and anatomical structures of the middle ear after mastoidectomy on a left ewe ear. Spatial orientation is shown with anterior (ANT) and superior (SUP) localizations. mm: millimeters.

Figure 3

Diagram of the sonoporation of RWM. After surgical mastoidectomy, MBs solution (in green) was injected in the middle ear and ultrasound were applied using an ultrasound probe connected to a degassed water-filled cone. Promontory temperature was measured before and after sonoporation using a thermal probe (in orange). M: Malleus, U: Uncus, S: Stapes, RWM: Round window membrane.

Figure 4

Representative measurement of bone conduction ABR for sonoporation side (right ear) before (A) and after sonoporation (B) and control side (left ear) (C) for sheep #5. Wave IV can be clearly observed and was not modified by sonoporation.

Figure 5

Unsupervised multivariate analysis of perilymph comparing data from sonoporation and control groups. Score-plot of the PCA is shown with adjustment of S1 and C6 data constructed from the metabolites found according to the first two components p1 and p2. The control samples are in purple (C1 to C6) and the sonoporation samples are in green (S1 to S6). Each group is associated with an ellipse regrouping all its samples. The thicker dots correspond to the center of each ellipse. Two samples with the same metabolomic profile appear grouped on this graph (i.e., samples S4 and C4).

Figure 6

Volcano plot representing metabolite dysregulations between control and sonoporation groups, with thresholds set at log2FC < −1 or >1, and −log10 (p-value) > 1.30 (p < 0.05). The vertical dotted bars represent the log2FC thresholds at −1 and 1. As all metabolites were under the horizontal dotted line, none of the expression variations were significant. The graph was similar using the FDR adjusted p-value.

Figure 7

Unsupervised multivariate metabolomic analysis of plasma collected before and after sonoporation. The plot score of the PCA was constructed from the metabolites found according to components p1 and p2. Pre-sonoporation samples are shown in purple (C1 to C6) and post-sonoporation samples are shown in green (D1 to D6). Each group is associated with an ellipse grouping all its samples. The thicker dots correspond to the center of each ellipse.