**Towards Regenerative Audiology: Immune Modulation of Adipose-Derived Mesenchymal Cells Preconditioned with Citric Acid-Coated Antioxidant-Functionalized Magnetic Nanoparticles**

**Adeline Josephine Cumpata 1,†, Dragos Peptanariu 2,†, Ana-Lacramioara Lungoci 2, Luminita Labusca 3,4,\*, Mariana Pinteala <sup>2</sup> and Luminita Radulescu 1,5**


**Abstract:** *Introduction and Background*: Based on stem cells, bioactive molecules and supportive structures, regenerative medicine (RM) is promising for its potential impact on field of hearing loss by offering innovative solutions for hair cell rescue. Nanotechnology has recently been regarded as a powerful tool for accelerating the efficiency of RM therapeutic solutions. Adipose-derived mesenchymal cells (ADSCs) have already been tested in clinical trials for their regenerative and immunomodulatory potential in various medical fields; however, the advancement to bedside treatment has proven to be tedious. Innovative solutions are expected to circumvent regulatory and manufacturing issues related to living cell-based therapies. The objectives of the study were to test if human primary ADSCs preconditioned with magnetic nanoparticles coated with citric acid and functionalized with antioxidant protocatechuic acid (MNP-CA-PCA) retain their phenotypic features and if conditioned media elicit immune responses in vitro. MNP-CA-PCA was synthesized and characterized regarding size, colloidal stability as well as antioxidant release profile. Human primary ADSCs preconditioned with MNP-CA-PCA were tested for viability, surface marker expression and mesenchymal lineage differentiation potential. Conditioned media (CM) from ADSCs treated with MNP-CA-PCA were tested for Il-6 and IL-8 cytokine release using ELISA and inhibition of lectin-stimulated peripheral blood monocyte proliferation. *Results*: MNP-CA-PCA-preconditioned ADSCs display good viability and retain their specific mesenchymal stem cell phenotype. CM from ADSCs conditioned with MNP-CA-PCA do not display increased inflammatory cytokine release and do not induce proliferation of allergen-stimulated allogeneic peripheral blood monocytes in vitro. *Conclusions*: While further in vitro and in vivo tests are needed to validate these findings, the present results indicated that CM from ADSCs preconditioned with MNP-CA-PCA could be developed as possible cell-free therapies for rescuing auditory hair cells.

**Keywords:** hearing loss; adipose-derived mesenchymal cells; conditioned media; magnetic nanoparticles; protocatechuic acid; regenerative medicine

**Citation:** Cumpata, A.J.; Peptanariu, D.; Lungoci, A.-L.; Labusca, L.; Pinteala, M.; Radulescu, L. Towards Regenerative Audiology: Immune Modulation of Adipose-Derived Mesenchymal Cells Preconditioned with Citric Acid-Coated Antioxidant-Functionalized Magnetic Nanoparticles. *Medicina* **2023**, *59*, 587. https://doi.org/10.3390/ medicina59030587

Academic Editors: Adriana Neagos, Daniela Vrinceanu, Codrut Sarafoleanu and Mahmut Tayyar Kalciogu

Received: 17 February 2023 Revised: 12 March 2023 Accepted: 14 March 2023 Published: 16 March 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

#### **1. Introduction**

Regenerative medicine (RM) involves the use of cells, bioactive molecules and supportive structures for replacing or substituting dysfunctional, lost organs or bodily functions [1]. With more than 20 years of basic and translational research and with the rapidly increasing number of clinical trials testing various regenerative strategies, RM is coming to age, already delivering therapies for so-far intractable diseases [2]. Recent developments in stem cell technology and nanomedicine have offered new perspectives introducing the possibility of rescue and regeneration of musculoskeletal tissues [3], sensory organs [4] and auditory epithelia [5] and to prevent or treat deafness of various causes in the adult and pediatric populations [6]. Differentiation methods to obtain hair cell-like cells seem to be effective; however, they proved to be technically challenging, costly and less amenable to large-scale manufacturing [7]. Mesenchymal stem cells (MSCs) are adult tissues which derive from a large variety of tissues of mesenchymal origin (such as bone marrow, adipose tissue, bone and synovium). MSCs assist regeneration mainly by means of the trophic and immunomodulatory effect generated by their paracrine activity. Adipose-derived mesenchymal cells (ADSCs) are MSCs derived from adipose tissue considered to be a convenient source for RM strategies due to their large-scale availability and phenotypic properties [8]. ADSCs were shown to express immunomodulatory and trophic effects in vitro as well as in vivo in an animal model investigating autoimmune and traumatic hearing loss [9]. Many issues in MSC/ADSC-based cell therapies remain unsolved, especially regarding cell survival after transplantation, accumulation within the target tissue as well as fate surveillance. The currently available clinically approved imaging methods are not able to detect a therapeutic cell population after implantation. Cell fate after implantation cannot be resolved non-invasively, which remains a problem that challenges medium- and long-term follow up of stem cell-based therapy and/or tissue-engineered constructs.

The use of nano-scaled materials, particularly of magnetic nanoparticles (MNPs), has evolved as an increasing field of research and application for medicine and life sciences [10]. Iron oxide MNPs internalized by the cells enable cell maneuverability, making them remotely controllable under an applied magnetic field (MF) as well as being traceable in vivo using clinically available magnetic resonance imaging (MRI). MNPs' biocompatibility is reportedly excellent, as they can be degraded by already-existent cellular iron handling molecular pathways. MSCs that have incorporated MNPs were shown to retain their specific phenotype and to become remotely controllable using an applied external magnetic field [11] with large applicability for cell targeting and cellular imaging using MRI or magnetic particle imaging (MPI) [12]. ADSCs loaded with MNPs are shown to retain their main phenotypic features in terms of proliferative and differentiation capability [13]. Little is known, however, about the immunomodulatory potential of ADSCs preconditioned with MNPs. MNPs' interaction with living matter is known to be determined by a their constituent physical parameters such as size, shape and coating materials but also by cell phenotype and functions. As a consequence, every MNP variety and cell type interaction needs to be carefully characterized before considering them as potential therapeutic agents. Several reports indicate that antioxidant drugs are capable of increasing the antiinflammatory and regenerative potential of ADSCs both in vitro and in vivo [14,15]. To date, MNPs' mediation of the delivery of antioxidant compounds to stem cells is underexplored but has important potential in increasing the therapeutic efficiency of ADSCs. PCA is a compound found in some traditional Chinese herbs which was reported to exert good antioxidant properties [16]. Citric acid (CA) is an organic compound found in lots of fruits and vegetables and is also known for its antioxidant and regenerative capabilities [17]. In this study, we tested the interaction between iron oxide MNP coated with citric acid (CA) and functionalized with antioxidant-protocatechuic acid (PCA) and human primary ADSCs. The purpose was to detect if ADSCs retain their basic phenotypic features (viability, differentiation, surface markers) when exposed to MNP-CA-PCA as well as to test if conditioned media from MNP-CA-PCA-preconditioned cells elicit an immune response in vitro. The cell viability and retention of characteristic mesenchymal stem cell phenotype

of ADSCs exposed to MNP-AC-PCA complexes were tested as well as the in vitro effect of MNP-AC-PCA preconditioning on ADSC cytokine release and interaction with the cellular immune system.

#### **2. Materials and Methods**

General information

Ferric chloride (FeCl3 × 6H2O), ferrous chloride (FeCl2 × 4H2O), 25% ammonium solution, citric acid (CA) and protocatechuic acid (PCA) were purchased from Sigma-Aldrich (USA). The Minimum Essential Medium composed of a mixture of Alpha Eagle 1% Penicillin, Streptomycin and Amphotericin B (10K/10K/25 μg in 100 mL), OsteoImage™ Mineralization Assay and AdipoRed™ Assay Reagent were from Lonza (Verviers, Belgium); the fetal bovine serum (FBS), Tryple, StemPro™ Adipogenesis Differentiation Kit and StemPro™ Osteogenesis Differentiation Kit were from Gibco (Langley, VA, USA); the phosphate-buffered saline (PBS) and Live/DeadTM Cell Imaging Kit were from Invitrogen (Eugene, OR, USA); the CellTiter-Glo 2.0 Assay was from Promega (Madison, WI, USA); the Max DiscoveryELISA kits for human IL6 and IL8 were from Bioo Scientific (Austin, TX, USA) and the Pancoll was from Pan-Biotech (Aidenbach, Germany).

Mouse anti-human antibodies for flowcytometry: anti-CD14-PACIFIC BLUE, anti-CD19-ECD, anti-CD34-PC5, anti-CD90 (Thy-1)-APC-AlexaFluor 750 and anti-CD105-PC7 were purchased from Beckman Coulter (Marseille, France) and Alexa Fluor® 488 Anti-CD73 from Abcam.

Magnetic nanoparticle synthesis

MNP-CA was prepared using the pre-addition method as previously described [18]. Briefly, 20 mL each of FeCl2 × 4H2O and FeCl3 × 6H2O were mixed in a molar ratio of 1:2, 1 mL citric acid (0.5 g/mL) and pre-added to the ferrous and ferric solution followed by 40 mL of distilled water. After heating at 65 ◦C, 14 mL NH4OH was added drop-wise into the mixture under vigorous mechanical stirring (650 rpm); the final product was centrifuged and washed 3 times with distilled water. Then, 1 mL solution PCA (10 mg/mL) was added to 1 mL MNP-CA (50 mg/mL) and submitted to mechanical stirring for 15 min before purification by magnetic decantation and washing 3 times with distilled water. The entrapment efficiency of PCA in a CA shell was calculated by absorbance reading at 287 nm of the supernatant solution using the following formula: Entrapment efficiency (%) = 100 (total drug-free amount of drug)/(total amount of drug), where free amount of drug and total amount of drug were determined using the calibration curves for PCA at 287 Nm vs. concentration of free PCA

MNP characterization

FT-IR

The Fourier transform infrared (FT-IR) spectra were recorded on a Bruker Vertex 70 FTIR instrument in a 400–4000 cm−<sup>1</sup> range, in transmission mode, in KBr pellet.

DLS

The hydrodynamic diameter and zeta potential were recorded using a Delsa Nano C Submicron Particle Size Analyzer (Beckman Coulter, Inc., Fullerton, CA, USA) equipped with a laser diode operating at 658 nm.

STEM images

MNP morphology was analyzed in STEM mode with a Verios G4 UC Scanning electron microscope (Thermo Scientific, Brno, Czech Republic) equipped with an energy-dispersive X-ray spectroscopy analyzer (Octane Elect Super SDD detector, Pleasanton, CA, USA). The STEM studies were performed using the STEM 3+ detector (bright-field mode) at an accelerating voltage of 30 kV. For STEM analysis, the samples were dispersed in water and ultrasonicated, then placed on carbon-coated copper grids with 300 mesh sizes and dried in an oven until the solvent was removed.

Release of antioxidant agent

The release profile of PCA from the MNP-CA-PCA sample was studied in PBS with a pH of 7.4. First, 30 mg loaded magnetic nanoparticles were placed in a 12 kDa dialysis bag and introduced in 100 mL PBS at 37 ◦C, under gentle stirring. Then, 1 mL of supernatant was taken out at fixed intervals (30 min) and replaced with 1 mL fresh buffer. This 1 mL of supernatant was diluted with 1 mL PBS and then assayed by UV-VIS spectrophotometry at 250 nm. The concentration values of the released PCA were determined using the calibration curve of PCA and the following formula:

$$\mathbf{C}\_{-}\mathbf{f}' = \mathbf{C}\mathbf{f} + \mathbf{v}/\mathbf{V}\sum \mathbf{C}\_{-}(\mathbf{f}\_{-}(\mathbf{i}-1));$$

where: v = volume of the release media taken out every time; V = volume measured by UV-VIS; Cf' = concentration of the released drug and Cf = concentration in volume V at specific intervals.

Free radical scavenging activity by DPPH method

The DPPH method was used for measuring the antioxidant activity of functionalized magnetic nanoparticles. First, 3 mL ethanol solution of DPPH (0.1 mg/mL) was added in each 3 mL suspension of MNP-CA-PCA of different concentrations. After 30 min, the absorbance values were measured using 1 cm quartz cuvettes. The absorbance values were read at 517 nm and the radical scavenging activity was determined using the following equation:

$$\text{\textbullet of inhibition} = \text{(Ac} - \text{As)} / \text{Ac} \cdot 100$$

where: As is the absorbance of MNP-CA-PCA samples of different concentrations and Ac is the absorbance of the DPPH solution of 0.05 mg/mL.

Human primary adipose-derived mesenchymal cells (ADSC)

ADSCs were obtained from healthy donors undergoing liposuction procedures for cosmetic reasons after institutional board ethical approval and informed patient consent was obtained; transportation to the laboratory was carried out in sterile conditions. The resulting lipoaspirate was processed within 24 h as previously described [13]. The lipoaspirate was washed three times with PBS, digested with collagenase type I (0.01 mg/mL) for 2 h at 37.5 ◦C and centrifuged twice at 300 g for 5 min at RT. The supernatant consisting of tissue debris was removed and the remaining medium further centrifuged at 300 g for 5 min. Pelleted cells were re-suspended in complete culture media (CCM-αMEM with 10% fetal bovine serum and a 1% Penicillin–Streptomycin–Amphotericin B mixture) and counted. Cells were plated at 1 × 106 cells/cm2 in appropriate tissue culture flasks (CellBIND surface, Corning). Cells in passage 3–4 were used for experiments, counted automatically and incubated at 37 ◦C and at 5% CO2 in an incubator with the media replaced every 3 to 4 days.

#### Flow cytometry

For the flow cytometry experiment, ADSCs were cultured in a T25 flask, harvested by detachment with Tryple, washed with PBS twice and finally resuspended in microcentrifuge tubes in 300 μL PBS each, for the unlabeled and labeled samples. The following markers were tested for presence/absence: negative markers for stem cells CD14, CD 19 and CD 34 on the fluorescence channels PB450, ECD and PC5.5 respectively; positive markers for stem cells CD73, CD90 and CD105 on fluorescence channels FITC, APCA750 and PC7, respectively; 1 μL of each antibody was added. Samples were vortexed briefly and incubated for 15 min at 37 ◦C, centrifuged at 300 g and washed twice with PBS, resuspended in 300 μL and analyzed on a CytoFLEX benchtop flow cytometer (Beckman Coulter Life Sciences, Indiannapolis, IN, USA).

In vitro toxicity of MNPs

In vitro MNP cytotoxicity was tested using the CellTiter-Glo kit. Cells were plated on 96-well white opaque tissue-culture-treated plates at densities of 5 × 103 cells/well in 100 μL/well complete medium and incubated for 24 h. The next day the media were replaced with serial dilutions of magnetite concentrations in complete cell culture medium and the plates were incubated for another 48 h. Before reading the results, the plates were removed from the incubator and kept at RT for 30 min followed by the addition of 100 μL/well of CelltTiter-Glo. Plates were shaken for 2 min and incubated for 15 min at RT. Light emission was assessed by spectrophotometry using the FLUOstar®Omega plate reader (BMG, Offenberg Germany). The relative cell viability is expressed as a percentage of the viability of control (cells treated only with cell culture medium) according to the following formula:

relative cell viability = (RLup − RLub)/(RLuc − RLub) × 100

where RLup, RLub and RLuc have relative light units recorded for samples, blank and control wells, respectively.

Live/dead viability assay

ADSCs were plated on 12-well tissue-culture-treated plates at densities of 40 × 103 cells/well in 1 mL/well complete medium and incubated for 24 h. The next day, the media were replaced with coated and non-coated MNPs at a concentration of 63 and 125 μm/mL, respectively; plates were incubated for another 7 days in CCM. At the end of the experiment, component A and component B from the kit were mixed as per the manufacturer's instructions; plates were incubated at RT for 20 min after which the images were collected using a Leica DMI 3000B inverted microscope (Wetzlar, Germany) using GFP and Texas Red filter cubes.

ADSC differentiation: adipogenesis and osteogenesis

Cells were cultured in 96-well black flat-bottom clear plates for quantitative evaluation and in 12-well plates to be photographed under a microscope. A density of 3200 cells/well was used for the 96-well plates, while 32,000 cells/well were seeded for the 12-well plates in complete αMEM medium. The next day, the medium was replaced with magnetite solutions in αMEM. After another 3 days, the medium was again replaced with commercially available adipocyte differentiation medium, respectively with osteogenic differentiation medium (see general information above). Adipogenesis assay was performed for 11 days while osteogenesis for 19 days as per the manufacturer's instructions.

Assessment of differentiation: adipogenesis

At 11 days, AdipoRed™ reagent was added according to the manufacturer's protocol. Briefly, the differentiation medium was removed and the cells were washed with PBS; AdipoRed™ dissolved in PBS was added and the plates were incubated 10 min at RT. Fluorescence (excitation 485 nm; emission 570) was read with a plate reader as described above. Similarly, the plates were qualitatively investigated with fluorescence microscopy.

Assessment of differentiation: osteogenesis

For the osteogenesis assay, cells were fixed with ethanol for 20 min. For qualitative and quantitative evaluation, Osteoimage™ was used as per the manufacturer's instructions. After staining, the samples were washed 3 times with wash buffer. To quantify the results, the 96-well plates were recorded with the plate reader (excitation 485 nm; emission 505), while the 12-well plates were analyzed with the fluorescence microscope.

Cytokine release

ELISA

ADSCs with or without MNP-AC-PCA were cultured in T25 flasks in αMEM medium with 0.2% FBS and 1% antibiotics for 11 days without changing the medium. After 11 days, the supernatant was removed and kept at −80 ◦C until the day of the ELISA test. For the ELISA test, steps were followed according to the manufacturer's protocol. Briefly, 100 μL of 1× assay diluent was added in the negative control wells, 100 μL of interleukin standards in separate wells, as well as 100 μL of the sample in other separate wells and the assay plates were incubated 2 h at RT. After incubation, the liquid was removed, washed 3 times with 250 μL wash solution, 100 μL detection antibody was added to the plates and incubated 1 h at RT, the liquid was aspirated, washed 3 times with 250 μL wash solution, 100 μL 1× avidin-HRP was added and the plates were incubated for 3 h at RT. The wells were washed 3 times with 250 μL wash solution, 100 μL TMB substrate was added and the plates were incubated for 15 min. Then, 100 μL stop buffer was added and absorbance was detected immediately at 450 nm using a plate reader (same as above).

Mixed lymphocyte reaction (MLR)

For MLR, we used the modified protocol by Herzig et al. [19]. First, 5 mL complete blood was collected in a vacutainer containing citrate as anticoagulant. Then, 15 mL Pancoll was placed in a 50 mL tube; 5 mL of blood was mixed with 5 mL PBS and allowed to settle, followed by centrifugation at 900 g for 30 min at 18 ◦C without the brake. PBMCs were extracted and resuspended, diluted with CCM and counted. The number of replicates for each of the 16 conditions was 3 with 6 × 10 × 5 cells per sample (sample set = 18; there were 16 treated and two controls (one positive and one negative) for lectin (Figure S1, Supplementary Materials). The sample set included conditioned media (CM) from ADSCs with or without MNP-CA-PCA at a 50% concentration (ADSC 50% (50 μL/100 μL) 50 μL concentrated CM and 50 μL medium = 100 μL) and at a 25% concentration (ADSC 25% (25 μL/100 μL) 25 uL concentrated CM and 75 μL medium = 100 μL) in the presence and absence of lectin; 96-well plates were used for this assay. Then, 100 μL per sample was left to incubate for 72 h at 37 ◦C. PBMC viability/proliferation was determined using 100 μL celltiter-Glo/well; data were processed in GraphPad.

#### **3. Results**

#### *3.1. MNP Characterization*

#### 3.1.1. FTIR

PCA has its characteristic peaks at 1676 cm and 1299 cm (C=O stretching vibration) and at 1467 cm and 1528 cm (C-C aromatic ring stretching mode). Citric acid has its characteristic peaks at 1753 cm (C=O stretch in the carboxylic groups) and in the 1500–1000 domain (C-O, C-OH, C-C vibrations). MNP has a characteristic peak at 572 cm (Fe-O bond). In the spectrum of MNP-CA, we found the characteristic peaks for the Fe-O bond (611 cm) and citric acid (1612 cm C=O stretch). In the spectrum of MNP-CA-PCA we found the characteristic peaks for Fe-O (611 cm), CA (1622 cm) and PCA (1485 cm) (Figure 1).

**Figure 1.** Fourier transformed infrared (FTIR) spectra of magnetic nanocomposites (**a**) protocatechuic acid (**b**) citric acid (**c**) magnetic nanoparticles (**d**) magnetic nanoparticles covered with citric acid (**e**) magnetic nanoparticles covered with citric acid and loaded with protocatechuic acid.

#### 3.1.2. DLS for the Uncoated MNPs

DLS measurements showed a hydrodynamic diameter of 325.6 nm and a zeta Potential of −2.88 mV. After coating with citric acid, the hydrodynamic diameter increased to 478 nm and zeta potential decreased to −17, 25 mV, which confirms the successful coating with CA. For MNP-CA-PCA, the hydrodynamic diameter was 397.9 nm and the zeta potential was −20.24 mV, which confirmed the adsorption of PCA in the CA shell (Table 1).


**Table 1.** DLS measurements for MNP, MNP-CA and MNP-CA-PCA.

#### 3.1.3. STEM

The STEM measurements of MNPs showed spherical particles of 7–10 nm with a tendency to agglomerate. For MNP-CA, the STEM images showed the CA coating around the bare MNPs. The difference between DLS measurements and the TEM ones consisted of the fact that in TEM, the solvent evaporated slowly, but in DLS the nanoparticles moved in an aqueous medium, resulting in bigger structures (Figure 2a,b).

**Figure 2.** STEM images for MNP (**a**) and MNP-CA (**b**).

#### 3.1.4. EDX

The EDX spectra of MNPs confirmed the presence of magnetite nanoparticles(Table 2) For MNP-CA and MNP-CA-PCA, the EDX spectra showed the presence of carbon because of the coating with citric acid. For MNP-CA-PCA, the oxygen content was the highest because of the presence of PCA (Table 2).


#### 3.1.5. DPPH

For MNP-CA-PCA, the DPPH tests indicated an IC50 value of 600 μg/mL. This confirmed good antioxidant activity. Not only did the magnetic core of MNPs not decrease the antioxidant activity of simple PCA but also had an overall contribution in the final antioxidant activity of the MNP-CA-PCA samples (Figure 3a,b).

**Figure 3.** (**a**) MNP-CA-PCA antioxidant activity; (**b**) PCA antioxidant activity.

#### *3.2. Drug Release*

The drug release profile of MNP-CA loaded with PCA showed a rapid and continuous release of the antioxidant PCA over the course of three days (Figure 4).

**Figure 4.** PCA release profile of MNP-CA functionalized with antioxidant PCA. The blue dots represent the number of measurements eash time the release media was taken out ant the straight portion of lines represent the overnight stopping of stirring and temperature over the course of three days.

#### 3.2.1. In Vitro MNP Cytotoxicity/ADSC Viability

We used two methods to determine cell viability in the presence of MNPs and MNP-CA-PC, respectively. For the quantitative CelltiterGlo test, we firstly tested ADSC viability at 48 h with increasing concentrations of non-coated MNPs in order to determine the particle working concentration. We found that LC 50% was situated between the concentrations of 62.5 and 125 μg/mL (Figure 5a). Next, dextran and citric acid-coated MNPs containing PCA in concentrations of 62.5–125 μg/m were tested for long-term viability against ADSCs. Since MNP-CA-PCA exposure yielded higher viability at 7 days, this coating was chosen to proceed with further experiments (Figure 5b).

**Figure 5.** (**a**) ADSC viability with increasing concentrations of as-prepared MNP; (**b**) ADSC viability with MNP coated with dextran and protocatechuic acid D-PCA and citric acid and protocatechuic acid CA-PCA in 63 mg/mL and 125 mg/mL concentrations added in the culture, respectively.

Qualitatively, ADSC viability was tested using the LIVE/DEAD assay. The method is based on the use of two fluorescent dyes for cell staining: a membrane-permeable dye for living cells and an impermeable dye to mark dead cells. The two dyes can be observed microscopically. The living cells display an intense fluorescence in the green domain (ex/em 488 nm/515 nm), while dead cells emit in the red domain (ex/em 570 nm/602 nm). We found that ADSCs in the presence of MNP-CA-PCA and MNP-D-PCA, 63 and 125 μg/mL, displayed an increased number of dead cells for both types of coatings for 125 μg/mL as compared with 63 μg/mL (Figure 6).

#### 3.2.2. ADSC Differentiation

We tested the differentiation potential of ADSCs cultured in specific, commercially available adipogenic and osteogenic differentiation media in the presence and absence of magnetic nanoparticles. The aim was to assess if the presence of MNP interferes with one of the definitory traits of mesenchymal progenitors: mesenchymal lineage differentiation. For the qualitative assessment of adipogenesis, we used AdipoRed®. The AdipoRed® reagent is a commercially available fluorescent dye used to color intracellular fat droplets because it is associated with triglycerides and in this way provides an image of the stage of adipocytic differentiation. It is a more sensitive assay than other methods such as the Oil Red O assay. ADSCs exposed and not exposed to as-prepared MNPs, MNPs loaded with citric acid or dextran, respectively and functionalized with PCA displayed positive AdipoRed® staining after 10 days in culture, signifying the successful conversion to preadipocytes. This was true for both concentrations of MNPs at 63 and 125 μg, respectively (Figure 7).

**Figure 6.** LIVE/DEAD qualitative evaluation of ADSC viability in the presence of MNP. (**a**–**d**) ADSC from day 0 to day 7 (in that order), bright field (BF) d 10×. (**e**) LIVE/DEAD staining at day 7 after LIVE/DEAD fluorescence, 10×. (**f**–**h**) ADSC-MNP-CA-PCA 63 μg/mL at days 3, 5 and 7, respectively BF 10×; (**i**) ADSC-MNP-CA-PCA LIVE/DEAD 7 day fluorescence. (**j**–**l**) ADSC-MNPs of 125 μg/mL at days 3, 5 and 7 BF 10×; (**m**) ADSC-MNP LIVE/DEAD day 7 fluorescence at BF 10×. (**n**–**p**) ADSC-MNP-D-PCA 63 μg/mL, BF 10×; (**q**) ADSC-MNP-D-PCA 7 day LIVE/DEAD fluorescence BF 10×; (**r**,**s**) ADSC-MNP-D-PCA 125 μg/mL, BF 10×; (**t**) ADSC-MNP-D-PCA 125 μg LIVE/DEAD, fluorescence, BF 10×; ADSC = cells non exposed to magnetic nanoparticles; ADSC-MNP = cells exposed to as-prepared magnetic nanoparticles; ADSC-MNP-CA-PCA = ADSC exposed to magnetic nanoparticles coated with citric acid and functionalized with protocatechuic acid; ADSC-MNP-D-PCA = ADSC exposed to magnetic nanoparticles coated with dextran and functionalized with protocatechuic acid.

**Figure 7.** Adipogenic differentiation of ADSCs exposed and not exposed to magnetic nanoparticles: (**a**) ADSC-MNP-CA-PCA 63 μg/mL adipogenic differentiation after 10 days, BF 20×.; (**b**) ADSC-MNP-CA-PCA 125 μg/mL; (**c**) ADSCs not exposed to MNP ((**a**–**c**) bright field); (**d**) ADSC-MNP-CA-PCA 63 μg/mL fluorescence (Adipored). The lipids are reddish-orange and the nuclei are blue (Hoechst); (**e**) ADSC-MNP-CA-PCA 125 μg/mL fluorescence (AdipoRed); (**f**) ADSCs not exposed to MNP (**d**–**f**) fluorescence (Adipored). The scale bar is 100 μm. Representative results are from one donor.

For assessing osteogenic conversion, we used fluorescent OsteoImage® dye (Lonza) that specifically binds to mineralized extracellular matrices deposited by pre-osteoblasts and osteoblasts if osteogenic conversion is successful. We found that ADSCs exposed and not exposed to as-prepared MNPs, MNPs loaded with citric acid or dextran, respectively and those functionalized with PCA displayed positive OsteoImage® staining after 19 days in culture, signaling osteogenic conversion. This was true for both concentrations of MNPs, i.e., 63 and 125 μg/mL, respectively (Figure 8).

**Figure 8.** Osteogenic differentiation of ADSCs exposed and not exposed to MNP after 19 days in culture: (**a**) ADSC-MNP-CA-PCA 63 μg/mL osteogenic differentiation after 19 days, BF 20×; (**b**) ADSC-MNP-CA-PCA 125 μg/mL; (**c**) ADSCs not exposed to MNP ((**a**–**c**) bright field); (**d**) ADSC-MNP-CA-PCA 63 μg/mL fluorescence (OsteoImage®). The deposited mineralized extracellular matrix is green and the nuclei are blue (Hoechst); (**e**) ADSC-MNP-CA-PCA 125 μg/mL fluorescence OsteoImage® (**f**) ADSCs not exposed to MNP (**d**–**f**) fluorescence OsteoImage® <sup>20</sup>×. The scale bar is 100 μm. Representative results are from one donor; ADSC-MNP = cells exposed to non-coated magnetic nanoparticles; ADSC-MNP-CA-PCA = ADSC exposed to magnetic nanoparticles coated with citric acid and functionalized with protocatechuic acid; ADSC-MNP-D-PCA = ADSC exposed to magnetic nanoparticles coated with dextran and functionalized with protocatechuic acid.

Quantitative evaluation of adipogenesis, respectively, osteogenesis using a spectofotometric assesment of fluorescent AdipoRed® and OsteoImage® revealed that MNP-CA-PCAloaded cells in both concentrations displayed increased differentiation potential compared with non-MNP-loaded controls (Figure 9).

**Figure 9.** Quantitative differentiation assay. (**a**) Adipogenic differentiation of ADSCs exposed and not exposed to magnetic nanoparticles MNP-CA-PCA; (**b**) osteogenic differentiation of ADSCs exposed and not exposed to MNP-CA-PCA.

#### 3.2.3. Mesenchymal Stem-Cell-Specific Surface Markers

Flow cytometry was used to identify positive and negative cell markers that define the phenotype of mesenchymal stem cells and identify the cells tested as being ADSCs. The positive markers for these cells are CD 73, CD 90 and CD 105. The negative markers or the markers that should not be present on ADSCs are CD 14, CD 19 and CD 34. Figure 9 depicts the presence of the positive markers and the absence of the negative markers for ADSCs representative of one donor (Figure 10).

**Figure 10.** Flow cytometry evaluation of surface markers. (**a**) gating around all significant events, (**b**) gating around the events in the negative control, (**c**) CD 14 and CD 19 where a cluster can be observed in the negative quadrants signifying the absence of CD 14 and CD 19, (**d**) negative control for CD 14 and CD 19, (**e**) Cell cluster CD 19 and CD 34 found in the negative quadrants, (**f**) negative control for CD 19 and CD 34, (**g**) CD 90 and CD 105 cell clusters found in the positive quadrants signifying the presence of these markers in the ADSCs, (**h**) negative control CD 90 and CD 105 in the negative quadrants, (**i**) Cell clusters CD 90 and CD 73 in the positive quadrants and (**j**) negative control CD 73 and CD 105 in the negative quadrants.

#### 3.2.4. Cytokine Release

The alteration of the anti-inflammatory profile of ADSCs in the presence of MNPs was investigated by measuring the level of interleukins IL-6 and IL-8 released in serum-free culture media using the ELISA technique. We found that after 11 days in culture, ADSCs and ADSC-MNP-AC-PCA release had strikingly comparable levels of IL-6, averaging 590/609 pg/m, while non-significant increases of IL-8 could be detected in ADSC-MNP-AC-PCA (averaging 171/2 compared with 213/6 pg/m (Figure 11a,b).

**Figure 11.** Cytokine release in CM from ADSCs preconditioned with MNP-CA-PCA: (**a**) IL-6; (**b**) IL-8. In green adipose derived stem cells (ADSC) cytokine release in brown adipose derived stem cells treated with MNP-CA-Pca cytokine release.

#### 3.2.5. MLR

We tested the ability of conditioned media from ADSCs and ADSC-MNP to influence peripheral blood monocyte reactivity to induce stimulation with a common allergen (lectin). As expected, fresh isolated human PBMCs in the presence of lectin displayed higher viability than non-stimulated PBMCs, showing their reactivity and induced cellular immune response. We found that conditioned media from ADSCs inhibited the proliferation of non-stimulated PBMCs by 10%, while conditioned media from ADSC-MNP resulted in inhibition of proliferation by 3%. A similar profile was recorded in the case of lectinstimulated PBMC-conditioned media from ADSCs, which decreased lectin-stimulated PBMC proliferation by 7.7%, while conditioned media from ADSC-MNP decreased stimulated PBMC only by 0.3% (Figure 3). While ADSCs have the ability to decrease proliferation of allergen-stimulated PBMCs, albeit non-significantly, MNP treatment decreases this ability. Lectin-stimulated as well as naïve PBMCs, however, did not increase proliferations in the presence of ADSC-MNP-conditioned media in this model in vitro (Figure 12).

**Figure 12.** The effect of CM from ADSC-CA-MNP on the proliferative reaction of lectin-stimulated allogeneic peripheral blood monocytes (PBMC) tested using CellTiter-Glo viability assay.

#### **4. Discussion**

Recently, the use of nanoparticles and especially MNPs has elicited constant interest for the development of potentially novel breakthrough therapies. MNPs' relevance as drug delivery systems constitutes a special field of research and their relevance for regenerative medicine is increasingly recognized. Given the significance of immune reactivity in regenerative processes, the design of therapeutic agents that have the ability to modulate soluble and cellular immune responses is of the utmost importance [20]. The use of stem-cell-based therapies as regenerative and immune-modulatory agents has already been tested in clinical trials for a large variety of therapeutic applications [21]. Stem-cell-based regenerative strategies for congenital or acquired hearing loss are expected to offer improved therapies [22]. However, the complexity of producing and marketing products that are based on living cells in terms of manufacturing, costs and ethical and regulatory approval has consistently impaired their advancement to clinical settings. Serum-free conditioned media (CM) from ADSCs has been sought out as a modality to avoid the hurdles involved in the therapeutic delivery of living cells [23]. CM from ADSCs preconditioned with various molecules has proven efficient in rescuing hair cell loss due to paracrine regenerative and immunomodulatory activity [24].

In this study, CM from human primary ADSCs preconditioned with MNPs coated with citric acid and antioxidant protocatechuic acid were tested for their effects on immunomodulatory cytokine release and in limiting cellular responses initiated in allogeneic human PBMC by a known allergen (lectin) using a modified protocol [19].

In-house-synthetized MNP-CA-PCA proved to have regular geometry, excellent dispersibility and good colloidal stability (zeta potential −24, 2 compared with as-prepared MNPs (−2), certifying the method used for coating and functionalization results in MNPs with superior characteristics related to their interaction with living matter.

Herein, we reported the synthesis and characterization of a novel formulation of core-shell iron oxide MNPs coated with CA and functionalized with antioxidant molecule PCA as well as their interaction with ADSCs.

The adsorption of an organic shell on the surface of MNPs allows the stabilization of MNPs, decreasing their tendency to agglomerate. Additionally, the zeta potential of coated MNPs tends to increase (positive or negative), making them more suitable for biomedical applications [25].

Furthermore, the core-shell nanoparticles can be used for loading therapeutic agents (drugs, redox substances) which can release the molecules at the site of interest [26].

We tested cell viability using two distinct methods, i.e., a qualitative (LIVE/DEAD) and a quantitative one (CellTiter-Glo) as a modality to investigate a larger panel of enzymatic equipment that is mandatory for cell viability and/or proliferation. The CellTiter-Glo test is based on the reaction between the luciferase enzyme and its substrate (luciferin), a reaction catalyzed by cellular ATP that results in the formation of oxyluciferin and the emission of photons. Since the efficiency of the reaction depends on the amount of ATP, and the latter depends on the number of living cells, the CellTiter-Glo test indirectly measures cell viability as well as cell proliferation. The CellTiter-Glo test is much more reliable than the colorimetric tests in which the absorbance of the chemicals to be tested can be difficult to subtract from the calculations and where the chemicals can produce reactions with the reagents of the test kits, leading to interference [27]. LIVE/DEAD detects the presence of intact cellular membranes as well as the presence of enzyme esterase activity characteristic of viable cells. Both tests performed indicated the excellent viability of ADSCs in the presence of MNP- CA-PCA added in the culture media and allowed to establishment of a working concentration for further experiments (i.e., 63 and 125 μg/mL, respectively). Previous studies reported good to excellent biocompatibility of MNPs with different coatings interacting with ADSCs in vitro as well as in vivo [28,29]. Good viability was reported in different mesenchymal stem cell types upon internalization of commercially available MNPs with different coatings [30]. Herein, we reported that human primary ADSCs interacting with as-prepared custom-made magnetite (Fe3O4) MNP as well as MNPs

coated with citric acid and antioxidant PCA display excellent short- and medium-term viability up to 7 days in culture. Compared with MNP-D-PCA, we found cell viability to be increased when exposed to MNP-CA-PCA, albeit non-significantly, at every concentration, therefore used we used the latter for further experiments. This effect was found to be, however, dose dependent since cell viability dropped below 50% at 50 μg/mL added in culture media (Figure 3). A particularity of MNPs that governs their interaction with cells and is distinctive from other nanoparticles is the existence of an intrinsic mechanism for iron storage and metabolism. Ferritin-dependent iron metabolism is an endogenous mechanism ubiquitous and evolutionarily conserved in all mammalian cells. Ferritin protein cages were shown to store degradation products of MNP cores, concomitantly slowing down the process due to their colloidal behavior in the acidic medium of the lysosomes [31]. A possible biosynthesis of bio magnetite by human mesenchymal stem cells after MNP degradation has been proposed as well, a process that might be cell-state-dependent [32].

We next tested if human primary cell interaction with MNP-AC-PCA interferes with their basic phenotypic features. The defining characteristics of ADSCs consist of their ability to adhere to a plastic surface of a culture dish, expression of CD73, CD90 and CD105, and lack of CD11b or CD34, CD19 or CD79, CD45 and HLA-DR as well as trilineage mesenchymal differentiation [33,34]. We found that after being exposed to different concentrations of as-prepared MNPs and CA-PCA-coated MNPs, ADSCs retained their adherence to a plastic culture dish for up to 7 days in culture. Cells retained their characteristic surface marker profile (Figure 10) and were capable of undergoing adipogenesis as well as osteogenesis in the presence of MNP-CA-PCA at 63 μg/mL as well as 125 μg/mL (Figures 7–9). Moreover both adipogenesis as well as osteogenesis were increased in the presence of MNP-CA-PCA, a fact that suggests cells not only preserve but increase their differentiation potential. Similar reports using commercially available [35] or proprietary MNPs with various coatings [27] indicate that ADSCs retain their differentiation potential to mesenchymal lineages in the presence of MNPs with various coatings. Enhanced cell metabolism produced by the presence or iron as well as changes in cell shape and cytoskeletal rearrangement were proposed to explain this finding. Several groups reported the increased osteogenic potential of bone marrow mesenchymal stem cells (BMSCs) loaded with MNPs [36,37]. Adipogenesis and increased osteogenesis were previously reported in ADSCs incorporating MNPs [27,38], a process that appears to, however, decrease with increasing concentrations of MNPs [39,40]. Herein, we found that both 63.5 as well as 125 μg/mL MNP-CA-PCA support ADSC adipogenesis and osteogenesis in vitro in the experimental conditions created.

We then tested the effect of MNP-CA-PCA presence on inflammatory cytokine expression by ADSCs. Serum-free CM from ADSCs treated with 125 μg/mL MNP-CA-PCA were found to release comparable levels of IL-6 and slightly non-significant increased levels of IL-8. IL-6 is a common denominator of acute inflammation and member of the pro-inflammatory cytokine family [41]. Since the presence of MNPs does not increase its release, it is possible ADSCs do not acquire an inflammatory phenotype in their presence. A slight increase in IL-8, another pro-inflammatory cytokine which can, however, act as an anti-inflammatory myokine [42], can possibly indicate ADSC reactivity. This process is very likely to be donor-dependent since very different values were obtained in CM media from different donors.

MSCs and especially ADSC cell suspension, CM or CM-derived extracellular vesicles are known to inhibit proliferation of allergen-stimulated allogeneic monocytes [43,44]. Herein, we found that CM from ADSCs but also from ADSC-CA-PCA reduce allogeneic blood monocyte proliferation under non-allergen stimulation, which indicates their ability to modulate cellular immune responses in vitro. This is a very important finding that has not been reported before which promotes the use of cell-free regenerative solutions. Hearing loss of various degenerative, traumatic congenital abnormalities [45] could benefit from such therapeutic approaches. Previous reports indicate that free iron ions are potentially capable of promoting immune modulation by stem cells [46] in a process that could

potentially be similar to macrophage polarization. This process is very likely fine-tuned with many factors involved (including ferric/ferrous iron balance, transferrin and hepcidin protein activity) [47]; therefore, our findings need to be further tested in vivo. Further tests are warranted to detect the potential of MNP-CA-PCA-preconditioned allogeneic ADSCs both in vitro and in vivo and to discriminate donor-related differences regarding ADSC immune profile. The present findings lay the foundation for and justify further investigation of the use of CM from preconditioned ADSCs as regenerative therapies for rescuing auditory hair cells and for treating hearing loss, with the important potential to translate the aforementioned to clinically available therapies.

#### **5. Conclusions**

This is, to our knowledge, the first report on the effect of MNP-CA-PCA preconditioning on ADSC phenotype retention and immune profile. We found that ADSC pretreatment with 63.5 and 125 μg/mL did not interfere with stem cell phenotype and supported cells' immune modulatory activity. Further tests are needed to validate the use of CM from MNP-CA-PCA-preconditioned ADSCs to rescue auditory hair cells and potentially in other sensory epithelia. These findings are important for the design of regenerative approaches to address hearing loss with important potential to translate them to clinically available therapies.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/medicina59030587/s1, Figure S1: MLR experiment set up details.

**Author Contributions:** Conceptualization, L.L. and L.R.; data curation, D.P. and L.L.; formal analysis, A.J.C., D.P. and A.-L.L.; funding acquisition, M.P. and L.R.; investigation, A.J.C., D.P. and A.-L.L.; project administration, L.R.; supervision, M.P. and L.R.; writing—original draft, A.J.C. and D.P.; writing—review and editing, L.L., M.P. and L.R. All authors have read and agreed to the published version of the manuscript.

**Funding:** This scientific research is financed by the University of Medicine and Pharmacy "Grigore T. Popa" Iasi-grant no.4711/25:02:2021.

**Institutional Review Board Statement:** The study was conducted in accordance with the Declaration of Helsinki, and donor sample collection was approved by the Institutional Review Board (or Ethics Committee) of County Emergency Hospital Saint Spiridon Iasi Romania Nr 31567/1507/2015 and Ethical Committee of County Emergency Hospital Saint Spiridon Iasi Nr 23/01/08/201.

**Informed Consent Statement:** Informed consent was obtained from all adipose tissue donors.

**Data Availability Statement:** Data are available from authors at reasonable request.

**Acknowledgments:** Authors wish to thank Florin Zugun-Eloae for critical referral of the manuscript.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

**Violeta Necula, Alma Aurelia Maniu, László-Péter Ujváry \*, Maximilian-George Dindelegan \*, Mara Tănase, Mihai Tănase and Cristina Maria Blebea**

> Otorhinolaryngology Department, "Iuliu Ha¸tieganu" University of Medicine and Pharmacy, 400347 Cluj-Napoca, Romania

**\*** Correspondence: ujvarypeter@outlook.com (L.-P.U.); maximilian.dindelegan@gmail.com (M.-G.D.)

**Abstract:** Otosclerosis is a pathological condition affecting the temporal bone, and is characterized by remodelling of the labyrinthine bone tissue through a dynamic process of osteolysis and osteogenesis. This condition progressively leads to hearing loss, tinnitus, and vertigo. Stapedotomy, a surgical procedure involving the removal of the stapes superstructure and its replacement with a prosthesis, is the treatment of choice to improve hearing in individuals with otosclerosis. However, vestibular dysfunction is a significant complication associated with this procedure, which can occur intraoperatively or postoperatively, ranging from the immediate postoperative period to weeks, months, or even years after surgery. This paper aims to provide a comprehensive review of the most important causes of vertigo associated with otosclerosis and stapes surgery with the goal of minimizing the incidence of this complication. Understanding the underlying factors contributing to vertigo in this context is crucial for the prevention and effective management of vertigo in patients undergoing stapedotomy.

**Keywords:** otosclerosis; vertigo; dizziness; stapes surgery

#### **1. Introduction**

Otosclerosis, also known as otospongiosis, is a progressive primary bone disorder of the otic capsule. It is characterized by abnormal focal resorption and recalcification of the endochondral layer of the temporal bone. This results in progressive conductive hearing loss, evolving in severe cases to a combination of conductive and sensorineural hearing loss. The otosclerotic process starts in the anterior part of the oval window, near the fissula ante fenestram resulting in fenestral otosclerosis. The disease can progress beyond the fissula ante fenestram and extend to the pericochlear otic capsule in cochlear otosclerosis or retrofenestral otosclerosis. The advanced stages can affect structures such as round window, semicircular canals, labyrinth, or vestibular nerve endings [1].

While progressive hearing loss is the main complaint in otosclerosis, other symptoms such as tinnitus and balance disorders are frequently associated with the condition. Up to 30% of patients with otosclerosis may experience vestibular symptoms, including instability, dizziness, and vertigo [2]. These symptoms can occur before or after treatment, and may arise as immediate or delayed complications of stapes surgery.

Surgical intervention is the preferred treatment for otosclerosis, with the aim of restoring the mechanism of sound transmission from the ossicular chain to the inner ear receptors, resulting in an air–bone gap closure of less than 10 dB in over 80% of cases [3,4]. Stapedotomy has become the most commonly performed surgical technique, replacing stapedectomy. For patients who are not suitable for or decline surgery, hearing aids can be recommended to improve hearing [5].

Considering the significant impact of vertigo on quality of life, this study aims to evaluate the presence of vertigo in otosclerosis and assess the risk of developing vestibular disorders following stapes surgery. By understanding the relationship between otosclerosis,

**Citation:** Necula, V.; Maniu, A.A.; Ujváry, L.-P.; Dindelegan, M.-G.; T ˘anase, M.; T ˘anase, M.; Blebea, C.M. Vertigo Associated with Otosclerosis and Stapes Surgery—A Narrative Review. *Medicina* **2023**, *59*, 1485. https://doi.org/10.3390/ medicina59081485

Academic Editor: Silviu Albu

Received: 16 July 2023 Revised: 13 August 2023 Accepted: 15 August 2023 Published: 18 August 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

209

vertigo, and stapes surgery, clinicians can better manage and minimize the incidence of this distressing complication.

#### **2. The Otosclerotic Process**

The otosclerotic process consists of abnormal replacement of enchondral bone with cancellous bone and subsequently with sclerotic bone. The process occurs in waves of osteolysis followed by osteogenesis. During the active phase of otospongiosis, the resorption process consists of replacing the normal bone around the blood vessels, which has cellular fibrous connective tissue, with mononuclear histiocytes, osteocytes, and osteoclasts [6]. A number of studies have shown that the enzymes secreted by the cells from the otospongiotic foci play a role in bone decalcification [7,8], while other studies have investigated the level of alkaline phosphates in the decalcification process of the otic capsule [9]. The final stage of the process is the otosclerosis stage, when the bone becomes mineralized and presents a mosaic appearance [10].

Lim et al. [11] described three types of otosclerotic lesions: cellular or spongiotic, characterized by the activation of monocytes, macrophages, osteoblasts, and osteoclasts; fibrotic, in which extensive bone fibrosis occurs; and sclerotic, characterized by a marked reduction of bone cells. Chevance et al. [12] reported the presence of osteolytic enzymes in the perilymph of patients who underwent surgery for otosclerosis, and suggested that these enzymes may have an important role in the development of otosclerotic lesions in the inner ear.

Stapes fixation occurs due to calcification of the annular ligament and the invasion of otosclerotic lesions at the oval window [13]. Otosclerotic lesions in the cochlear endosteum can lead to atrophic and hyalinization changes of the spiral ligament [14]. Damage to the spiral ligament can disrupt the chemical balance of ion-fluid recycling [15] and obstruct the endolymphatic duct and sac, resulting in biochemical changes [16,17]. Gros et al. [18] observed that vestibular disorders are frequently associated with sclerotic lesions. Saim and Nadol [19] reported that patients with vestibular symptoms have elevated bone-conduction thresholds and suggested that the degeneration of the vestibular nerve and Scarpa ganglion cells could be responsible for these symptoms, regardless of otosclerotic damage to the vestibular end organs.

Otosclerosis is associated with inflammation, disturbed collagen expression, and the presence of viral receptors and antigens in the otosclerotic foci [20]. At the molecular level, the bone remodelling process is regulated by a series of cytokines, signalling molecules that play a crucial role in regulating various cellular processes including bone remodelling. In otosclerosis, cytokines such as osteoprotegerin (OPG), receptor activator of nuclear factor kB (RANK), and RANK liand (RANKL), as well as transforming growth factor ß1 (TGF-ß1), are involved in controlling the balance between bone resorption and bone formation [20,21]. The presence of the measles virus and concurrent inflammation may trigger the abnormal bone remodeling that is a characteristic of otosclerosis. In the active phase, there is an increase in inflammation, detectable measles virus particle, local expression of tumor necrosis factor alpha (TNF-α), and negativity for OPG expression. During this phase, the balance between bone resorption and formation may be disrupted. In the inactive phase of otosclerosis, OPG positivity and TNF-α negativity are observed, along with absence of inflammation [22,23]. The increased level of TNF-α can stimulate osteoclast activation, induce RANKL expression, and reduce osteoclast apoptosis. This sequence of events ultimately leads to osteolysis and contributes to the process of otospongiosis [22,24]. TNF-α overexpression stimulates osteoclast formation both by inhibiting OPG secretion and by stimulating RANKL formation [23,24].

#### **3. Anatomy of the Membranous Labyrinth**

The anatomy of the membranous labyrinth is essential to understanding certain pathological processes; in surgical procedures involving the inner ear, these particularly relate to otosclerosis and stapes surgery. The middle ear communicates with the inner ear through the oval and round windows. The footplate articulates with the oval window through the annular ligament. Beyond the oval window is the vestibule, which is filled with perilymph. The membranous labyrinth is supported by periotic connective tissue within the perilymphatic space, which is medial and superior to the utricle and saccule and absent lateral to them [25].

The otolith organs of the vestibule are the macula of the utricle and saccule, located medially in the vestibule. Their role is to detect the position and direction of the head as well as the linear and gravitational acceleration [26].

The utricle has an elongated shape and communicates with the semicircular canals. On the inferior wall, the more lateral is the macula of the utricle, oriented horizontally. It is localized next to the upper edge of the oval window at 0.5–1 mm distance [27]. In surgical procedures, the macula of the utricle can be observed as a white plaque within the vestibule [28].

The saccule is situated in the spherical recess, and its macula has a vertical orientation perpendicular to the macula of the utricle. The anterior wall of the saccule is adjacent to the footplate. Between these two structures is found a connective tissue, named the reinforced area of the saccular membrane [29]. The macula of the saccule projects below the horizontal line passing through the arm of the stapes [30]. The distance between the saccule and the anterior edge of the oval window is between 1 and 1.5 mm [27]. The saccule communicates with the utricle through the utriculo-saccular duct, with the cochlea through the reuniens duct, and with the endolymphatic duct through the sinus of the duct.

The membrana limitans is a membranous structure, similar to a network, which delimits the superior vestibular labyrinth from the inferior part, laying below the utricle and supporting its macula [31]. In certain cases, the membrana limitans can present thin fibrillary attachments to the footplate, especially in the posterior third [25]. Its role is more of a support than a barrier, as it has a discontinuous structure that allows the passage of perilymph. Its insertion is in the superior part of the vestibule, superior to the oval window, immediately above the stapes footplate. In certain cases, the membrana limitans can be directly inserted in the footplate [25]. Pauw et al. [32] reported that the distance between the footplate and the utricle is smaller in patients with otosclerosis than in normal subjects. These rapports and adhesions of the labyrinthine structures to the footplate may partially explain the vertigo experienced during stapes surgery.

Vestibular symptoms can be part of the clinical manifestations of otosclerosis or can occur during or after stapes surgery, either immediately or with delayed onset.

#### **4. Preoperative Vertigo**

The incidence of preoperative vestibular symptoms in otosclerosis patients varies greatly from one study to another. Different studies have reported incidence ranging from 8.6 to 30% [33–35]. While the exact cause of vertigo in patients with otosclerosis is not fully understood, several factors have been proposed (Table 1).

One factor is the presence of otosclerotic foci, which can affect the endolymphatic duct and sac, leading to hydrops [15]. This abnormal accumulation of fluid in the inner ear can contribute to vertigo. Temporal bone studies have shown the presence of endolymphatic hydrops (EH) in specimens with extensive otosclerotic lesions in the cochlear endosteum or in the vestibular aqueduct, obstructing the flow of endolymph and disrupting labyrinthine fluid homeostasis [15,16]. The presence of endolimphatic hydrops can be visualized on delayed three-dimensional (3D) fluid-attenuated inversion recovery (FLAIR) MRI images obtained after intravenous administration of gadolinium. Sone et al. [36] suggested that the presence of preoperative asymptomatic vestibular EH could serve as a predictive factor for postoperative complications following stapes surgery. EH located adjacent to the oval window could be a contraindication for stapes surgery. The proximity of EH to the surgical site may increase the risk of vestibular complications after the procedure, including vertigo.


**Table 1.** Summarised data on vertigo associated with otosclerosis.

Detachment of the otoconia from the macula of the utricle is another factor that could contribute to positional vertigo, and may explain anomalies observed in the ocular and cervical vestibular-evoked myogenic potentials (oVEMP and cVEMP), which are used to assess the status of the utricle and saccule, respectively [37–39].

The third factor is vestibular end organ and/or neural degeneration due to otosclerotic foci involving the utricular or ampullary nerve. These changes could be related to the utricular deficit and oVEMP anomalies in patients with vertigo and otosclerosis [40].

Additionally, hydrolytic enzymes originating in the otosclerotic foci have been identified in the perilymph of otosclerosis patients; these can produce vascular and neuroepithelial lesions. Moreover, the cytokines produced in these foci can cause changes in labyrinthine fluids' chemical composition and homeostasis [41].

Degeneration of receptor cells in the vestibule and changes in the nonsensory epithelium have been observed in temporal bone studies of otosclerosis patients. Kaya et al. [42] studied temporal bones harvested from patients diagnosed with otosclerosis and found a decrease in the population of vestibular dark cells and vestibular transitional cells in temporal bone specimens with endosteal involvement. The role of these cells is to maintain the homeostasis of the labyrinthine fluid by controlling the transport of ions and water in order to prevent vestibular dysfunction. Another study by Hizli et al. [43] found that the mean density of type I hair cells in the saccule was significantly reduced in cases with endosteal involvement, suggesting that the extension of the otosclerotic foci towards the endosteum may be an important factor in the occurrence of vestibular symptoms in patients with otosclerosis. They suggested that this might explain the abnormal oVEMP and cVEMP response in patients with otosclerosis and vestibular symptoms.

Saka et al. [44] studied the vestibular-evoked myogenic potential in response to boneconducted sound (BC-VEMP) in a group of 25 patients and showed that 9 of 10 patients with vestibular dysfunction presented abnormal BC-VEMPs. This suggests that saccular dysfunction may be involved in these patients, possibly due to saccular hydrops or the extension of otosclerotic foci to the saccular macula or saccular afferent, considering the anatomical proximity of the saccule to the oval window.

On the other hand, another study involving 27 patients with otosclerosis and vertigo reported abnormal oVEMPs in response to impulsive stimulation, suggesting pathological abnormalities related to the utricle [40]. Hayasi et al. [37] studied 35 temporal bones with otosclerosis, reporting a higher incidence of cupular deposits compared to temporal bone without otosclerosis. They suggested that the origin of these deposits was probably the otoconia from the utricle, from where they detached and migrated to the cupula of semicircular canals.

The specific manifestation of vestibular symptoms can vary from person to person, and may be influenced by factors such as the extent of otosclerotic involvement and individual differences in anatomy and physiology. A study conducted by Eza-Nuñez et al. [45] highlighted the diversity of vestibular symptoms experienced by patients with otosclerosis. Patients with otosclerosis mention vertigo or imbalance, which can manifest in different ways, including a single episode or recurrent attacks either triggered by positional changes or occurring spontaneously. In their study, positional vertigo was associated with otosclerosis in 32.5% of patients and Ménière syndrome was reported in 30% of patients. Around 27.5% of patients experienced spontaneous recurrent vertigo, approximately 7.5% of patients presented with chronic unrelapsing imbalance, and a small percentage of patients (2.5%) had acute unilateral vestibulopathy [45].

These findings suggest that multiple factors, including endolymphatic hydrops, detachment of otoconia, degeneration of receptor cells, and cupular deposits, may lead to vertigo in otosclerosis patients. Further research is needed in order to fully understand the underlying mechanisms and develop targeted interventions for vestibular symptoms in this population.

#### **5. Immediate Postoperative Vertigo**

Treatment of otosclerosis is mainly surgical, generally with good results. The most common technique is stapedotomy, a minimally invasive technique that has largely replaced stapedectomy due to having fewer complications, including vestibular disorder [33]. The surgical approach can be either classical, using a microscope, or endoscopic, and the stapedotomy can be carried out by a conventional or laser-assisted technique.

Vertigo is reported to occur intraoperatively in 2.1% of patients, mainly due to the manipulation of the footplate. It may occur due to frequent suctioning in the middle ear, and less often to a floating footplate. This is treated by reassurance of the patient and vertiginous drugs [46] (Table 1).

After surgical treatment, immediate postoperative vertigo can occur in a significant percentage of patients. The reported incidence of vestibular symptoms varies among studies, ranging from 3.4% to 70% [47–49] (Table 1).

Early postoperative vertigo is usually temporary, and authors report remission of symptoms in most cases after 5 to 7 days with conservative management, including medication and bed rest [50,51].

The use of a CO2 Laser in footplate perforation has been suggested to reduce the prevalence of postoperative vertigo due to minimal mechanical trauma to the inner ear through lesser footplate manipulation [47,52].

Several factors contribute to the occurrence of immediate postoperative vertigo. A possible cause of premature vertigo could be serous and chemical labyrinthitis, which involves irritation of the membranous labyrinth, particularly the macula of the utricle located near the oval window [53]. Suction of the perilymph from the vestibule or contact of the instrument with the membranous labyrinth can trigger vertigo [54]. According to measurements taken Pauw et al., penetration of instruments or the prosthesis into the vestibule is considered less risky in the centre and lower third of the oval window [32].

Nystagmus is observed postoperatively in approximately 65.7% of patients, and may persist for over one month, as shown by Fukuda et al. [55] in a study conducted in 2021.

Singh et al. [52] used posturography to evaluate patients, and found that patients experienced vestibular deficits and increased subjective symptom scores at one week after surgery, with remission occurring after four weeks.

Assessment of cVEMP with air conducted stimuli before surgery and three months after stapedotomy showed a significant reduction in the amplitude of P1/N2 waves in patients who complained of dizziness and vertigo, suggesting a saccular lesion in these patients [56]. The reduction of air conduction (AC) and bone conduction (BC) VEMPs in patients with otosclerosis was reported by Trivelli et al. [57], with the observation that although the air conduction thresholds improved after surgery in all patients, AC-VEMP and BC-VEMP did not significantly improve in operated patients. Akazawa et al. [58] evaluated the cervical and ocular VEMPs through bone-conducted vibration before and after surgery, finding no significant changes in VEMPs in the operated ear after stapes surgery.

Postoperative vertigo following stapedotomy may be attributed to traumatization of the utricle, release of proteolytic enzymes, antigen–antibody reactions, pressure changes in labyrinthine fluids, and reduction of blood supply to the labyrinth caused by a floating footplate during the operation [53,59].

Among the three semicircular canals (SCC), the lateral SCC appears to be the most affected in both otosclerosis and after stapes surgery. SCC function can be evaluated by vHIT. Postoperative vHIT results have indicated subclinical damage to the lateral and posterior SCC. This is further supported by studies on temporal bones which revealed degeneration of the sensory epithelium in the cristae of the SCC [60,61]. Kujala et al. [62] evaluated patients after stapes surgery and found latent spontaneus horizontal-torsional nystagmus in 33% of patients on the day of surgery. The presence of this nystagmus suggests minimal impairment of the SCC.

Overall, immediate postoperative vertigo is a common occurrence following surgery for otosclerosis, though it is usually temporary and resolves with conservative management. Monitoring vestibular function through VEMPs and other tests can provide insights into the underlaying mechanisms and help to evaluate the impact of surgery on vestibular function.

#### **6. Late Postoperative Vertigo**

Late postoperative vertigo can occur following stapes surgery for otosclerosis, with a reported incidence ranging from 0.5% to 17% [54]. The persistence of vertigo beyond four weeks is observed in approximately 4% of patients, as shown by Birch et al. [63] in a study of 722 patients, while in Albera's study 17% (58/347) [54] showed changes in the caloric test even up to 15 years after surgery. A small percentage of patients (2.6%) may experience vertigo lasting over 12 months, indicating permanent postoperative vestibular hypofunction [64] (Table 1).

One potential cause of late postoperative vertigo is the perilymphatic fistula, which occurs due to inadequate sealing around the prosthesis at the oval window. Its incidence is variable from one study to another, ranging from 1.3% to 10% [65–68]. Pedersen et al. [65] suggested that the cause may be inadequate sealing around the prosthesis in the oval window. A systematic review of the results and complications of stapes surgery confirms that perilymphatic fistula is a rare complication of stapes surgery and represents approximately one third of surgical revision cases [69]. Although its existence has been highlighted intraoperatively in only a few cases, the correlation between filling with tissue or fibrin glue and remission of symptoms suggests that the perilymphatic fistula is often underestimated [70].

The perilymphatic fistula can persist postoperatively if the hole around the prosthesis has not been closed, or may appear later if the graft or prosthesis moves as a consequence of increased pressure due to coughing or sneezing. Usually, complaints involve fluctuating hearing loss and vertigo, and the audiogram indicates deterioration of cochlear function [71]. Incidence can be reduced in the case of stapedectomy by placing a graft over the oval window [72].

Due to the risk of meningitis and hearing loss, the presence of perilymphatic fistula represents a serious complication. If a perilymphatic fistula is suspected and the symptoms do not improve with treatment, surgical exploration of the ear is necessary in order to close the fistula with a soft tissue graft. Persistence of the fistula can lead to irreversible hearing loss and persistence of vertigo [2].

According to Nakashima et al., the incidence of perilymphatic fistula can reach 22% in patients in whom the obliteration was performed with gelfoam and 4% in those in whom the obliteration was performed with tissue [73]. An older study comparing gelfoam, fat tissue, and fascia showed an incidence of perilymphatic fistula of 3.5% in case of gelfoam, 1.9% in case of fat tissue, and 0.6% in the case of fascia [74]. According to Lim et al. [75], from an auditory point of view fatty tissue is to be preferred in stapedotomy and fascia in stapedectomies.

Other causes of postoperative vertigo include irritation produced by a protracted prosthesis or a displaced one. Symptoms intensify when moving the head or during the Valsalva manoeuvre. The patient may experience dizziness related to hiccupping, burping, yawning, popping of the ears, and specific acoustic stimuli [76].

Reparative granuloma, which is a pyogenic inflammatory reaction, autoimmune or allergic reaction, or exuberant healing response, can occur in approximately 0.1% to 18% of cases after stapedectomy or stapedotomy [77–80]. Reparative granulomas occur more frequently after stapedectomy, and are characterized by sensorineural hearing loss along with vertigo. Typically, reparative granuloma manifests 7 to 15 days after surgery [81].

Persistent late vertigo can be due to bone fragments entering the vestibule during surgery; additional causes include direct compression of the saccule due to adhesion between the prosthesis and the tympanic membrane, Eustachian tube dysfunction, and Tullio phenomenon [82–84]. Stapes surgery can damage the inner ear and eventually leads to endolymphatic hydrops without decreasing the hearing threshold at low frequencies [85]. Endolymphatic hydrops (EH) can be associated with otosclerosis as a secondary condition following stapes surgery, when EH can occur immediately after the surgery or at a later time. Clinical manifestations include low-frequency fluctuating sensorineural hearing loss, episodic vertigo, tinnitus, and aural fullness. However, according to Halpin et al. [85] these symptoms are much more rare compared to the presence of histopathological findings on TB specimens of patients who underwent otosclerosis surgery.

Rarely, late-onset vertigo can be associated with pneumolabyrinth or barotrauma, a condition presented by Mandala et al. in a patient who started to have vertigo years after surgery [86]. Additionally, Gomes et al. [87] published a case report about a patient who came back 4 weeks after stapedectomy for displacement of the prosthesis and the graft. In most cases, this complication occurred a few weeks or months after the surgery. The diagnosis is based on HRCT scan showing the presence of air bubbles in the vestibule in patients with vertigo and a positive fistula test.

Several factors can contribute to prolonged vertigo, including age, sex, stapes surgery in the opposite ear, the seal around the prosthesis in the footplate, and postoperative hearing outcomes. A history of stapes surgery in the opposite ear has been identified as a significant predictive factor for prolonged nystagmus and subjective vestibular symptoms [55].

Intractable vertigo may be an indicator for revision surgery in otosclerosis cases [54]. Prompt diagnosis and appropriate management are crucial in addressing the underlying causes of late postoperative vertigo and improving patient outcomes.

#### **7. Treatment of Vertigo**

The treatment of vertigo in patients with otosclerosis depends on the underlying cause and the severity of symptoms.

The immediate postoperative vertigo during the surgical procedure and in the first few days resolves mostly with bed rest and symptomatic treatment within approximately 5 to 7 days, rarely lasting more than 4 weeks.

If vertigo symptoms persist or are caused by specific complications such as a perilymphatic fistula or malposition of the prosthesis, surgical intervention can be considered to obliterate a possible perilymphatic fistula, reposition the prosthesis, or take other corrective measures to address the underlying cause of vertigo.

Postoperative or late vertigo can be improved by treatment with the latest generation of bisphosphonates or by physical therapy, including vestibular rehabilitation exercises. These exercises aim to improve balance, reduce dizziness, and promote central compensation of the vestibular system.

Medical treatment of otosclerosis includes drugs that can directly influence bone metabolism, anti-inflammatory agents that address the inflammatory etiology, targeted (biological) therapies, and, last but not least, anti-measles vaccination [88].

Bone metabolism inhibitors aim to preserve hearing thresholds and alleviate symptoms such as tinnitus and vertigo associated with otosclerosis. Sodium fluoride, often combined with calcium carbonate and vitamin D, is used to slow down the progression of otosclerosis. It has been shown to reduce the deterioration of hearing loss and help to control tinnitus and vestibular symptoms. However, there are differing opinions as to its overall effectiveness in treating otosclerosis. Studies have stated that sodium fluoride can slow down the evolution of the disease in more than 50% of cases [89], while others have shown reduced efficiency in the treatment of otosclerosis [88].

Studies have shown that bisphosphonates can influence vestibular symptoms in patients with otosclerosis before or after surgery. Brookler and Tanyeri have reported that 54% of patients presented an improvement in vestibular symptoms after treatment with bisphosphonates, while 39% reported disappearance of dizziness and 35% presented improvement in the results of tests performed with a rotatory chair [90]. The newer generations of bisphosphonates (e.g., risedronate, zoledronate) have more favorable tolerability and are more powerful bone resorption inhibitors [91].

Bioflavonoids can reduce bone resorption by inhibiting the phosphodiesterase enzyme. While they might not significantly affect hearing loss, they have been found to significantly reduce tinnitus [92].

Vitamin D's anti-inflammatory effects and vitamin A's ability to inhibit osteoclast differentiation could potentially have a beneficial impact on otosclerosis [88,93].

Regarding anti-inflammatory agents, corticosteroids are often used in inner ear diseases, including otosclerosis. Transtympanic administration can be a solution to minimize systemic side effects. Nonsteroidal anti-inflammatory drugs have been considered for their inhibitory effect on bone resorption; however, there is a lack of long-term clinical data in otosclerosis [88,92].

From the class of immunosuppressive agents, only cyclosporine A has been studied, and there is limited data on the use of other immunosuppressive drugs in treating otosclerosis [94].

Emerging treatments such as anti-TNF-α agents (e.g., infliximab), recombinant human OPG (rhOPG), and other anti-osteoporotic targeted therapies (e.g., denosumab, odanacatib) hold potential for otosclerosis treatment, however, more long-term studies are needed [92].

Fluoride-based medications and bisphosphonates are among the treatments considered for medical management of otosclerosis-related hearing loss and associated symptoms such as tinnitus and vertigo.

#### **8. Limitations of the Study**

The reported percentages of balance disorders in otosclerosis patients both pre- and postoperatively varies significantly in the literature due to differences in study design and patient assessments. The main complaint in otosclerosis is typically hearing loss, with balance disorders often being secondary, which can lead to variations in the assessment and reporting of vestibular symptoms. Postoperative vertigo can vary depending on factors such as the individual patient, surgeon's experience, surgical technique, and type of prosthesis used.

Advancements in technology have provided new tools for assessing vestibular deficits, allowing for more objective evaluation of vertigo. Objective measures can provide valuable information about vestibular function before and after surgery.

Multicentre studies with larger patient populations and standardized evaluation protocols would be beneficial in providing more comprehensive and reliable data on balance disorders in otosclerosis patients before and after surgery.

#### **9. Conclusions**

Vertigo is a common manifestation in otosclerosis both before and after surgical treatment. It can present as benign paroxysmal positional vertigo, vertigo attacks, or hydrops, as well as dizziness or light-headedness. While vertigo quite frequently appears immediately after surgery, the symptoms typically subside quickly with medical treatment, and only persist for longer periods of time in a very few cases.

When vertigo occurs months or years after the surgery, this can be an alarming signal of a complication that may require surgical reintervention. In such cases, careful evaluation and appropriate management are crucial in order to address the underlying cause and alleviate the vertigo symptoms.

Overall, understanding the occurrence and characteristics of vertigo in otosclerosis both pre- and postoperatively is essential for effective diagnosis, treatment, and patient management. Further research and standardized assessment protocols are needed in order to provide more comprehensive data and improve the management of balance disorders in otosclerosis patients.

**Author Contributions:** Conceptualization, V.N., A.A.M., L.-P.U., M.-G.D., M.T. (Mara Tănase), M.T. (Mihai Tănase) and C.M.B.; writing—original draft preparation, V.N., A.A.M., L.-P.U., M.-G.D., M.T. (Mara Tănase), M.T. (Mihai Tănase) and C.M.B.; writing—review and editing, V.N., A.A.M., L.-P.U., M.-G.D., M.T. (Mara Tănase), M.T. (Mihai Tănase) and C.M.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data sharing not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
