https://orcid.org/0009-0008-4862-1881
Abstract
Lassa fever (LF), caused by Lassa virus (LASV) infection, typically leads to mild symptoms in humans, but some survivors experience audiovestibular problems. Here we present vestibular histopathological insights in our LF model mice. We observed (1) hemorrhage within the vestibular ganglion and stroma beneath the sensory epithelium, (2) preserved hair cells and supporting cells, (3) LASV antigen presence in the vestibular ganglion cells and the stroma beneath the sensory epithelium, and (4) CD3-positive T-lymphocyte infiltration in the vestibular ganglion and the stroma underlying the sensory epithelium. LASV and/or its immune response likely contributes to the pathogenesis of vestibular dysfunction.
Lassa fever (LF), a viral hemorrhagic fever caused by infection with Lassa virus (LASV), is endemic to West African countries such as Mali, Sierra Leone, and Nigeria [1]. LF has an approximately 1%–10% fatality rate with the majority infected with LASV unknowingly overcoming the disease with only mild symptoms [1]. One of the important LF sequelae is audiovestibular dysfunction. Approximately 25% of LF survivors develop sudden-onset sensorineural hearing loss [1, 2], and up to 50% of LF patients develop vestibular symptoms such as dizziness, vertigo, and ataxia [3] after surviving the acute phase of the disease or later in the convalescence phase. The mechanism underlying audiovestibular dysfunctions caused by LASV infection remains poorly understood, with limited detailed reports on pathology and histopathology available.
Previously, we documented histopathological changes in the inner ear cochlea associated with hearing loss in our LF model mice [4]. Our findings revealed severe destruction of spiral ganglion cells and the cochlear nerve, accompanied by substantial infiltration of CD3-positive T lymphocytes in the perineural area. Remarkably, cochlear hair cells remained intact despite severe hearing loss following LASV infection [4, 5]. In some of these mice, we noted behaviors suggestive of vestibular dysfunction, including head bobbing and head tilting. Thus, our LF model mice represent a unique tool for investigating the pathology behind the audiovestibular dysfunction phenotype, which closely mirrors that observed in human LASV infection. This study focuses on the histopathological examination of the vestibular apparatus in our LF model mice, offering crucial insights into the mechanisms underlying vestibular symptoms in LF patients.
METHODS
Viruses and Biosafety
The details were described previously [4]. LASV LF2350 (GeneBank accession numbers PP826286 and PP826287) and LF2384 (GeneBank accession numbers PP826288 and PP826289) were directly isolated from nonfatal and fatal cases during a 2012 outbreak in Sierra Leone. All work with infectious LASV was performed in the biosafety level 4 (BSL-4) facility at University of Texas Medical Branch in accordance with institutional and safety guidelines.
Mice
We examined the inner ear vestibular apparatus after LASV infection in the same mice used in the previous study [4]. The mice deficient in signal transducer and activator of transcription 1 (Stat1) showed symptoms similar to LF in the acute infectious phase and developed hearing loss in the convalescent phase [4]. On the other hand, interferon-α/β and -γ receptor (IFN-α/βγR) deficient mice developed a transient mild disease in the acute phase and no hearing loss [4]. The details of the experiments were described previously [4], and the main objective was to evaluate the auditory function and cochlear pathology of the LF model mice. Briefly, female mice at 6–8 weeks of age were intraperitoneally infected with 105 plaque-forming units of LASV. Phosphate-buffered saline was injected into mice as a mock control. The inner ears were collected from mice at 60 days postinfection. The systemic symptoms and hearing test results were presented in our previous study [4]. All animal studies were reviewed and approved by the Institutional Animal Care and Use Committee at University of Texas Medical Branch (No. 1010054A) and were carried out according to the National Institutes of Health guidelines.
Histology
After the mice were euthanized, the dissected temporal bones were fixed in 10% formalin for 72 hours before removal from the BSL-4 facility. The temporal bones were embedded in paraffin, thin sectioned at 4- to 7-μm slices, and subjected to either hematoxylin and eosin staining or immunolabeling with anti-Lassa glycoprotein antibody (IBT Bioservices) or anti-CD3 antibody (Dako). The immunolabeled slides were processed with secondary antibody and then with 3, 3’-diaminobenzidine, and then counterstained with hematoxylin. The slides were observed with an Olympus IX 71 microscope.
RESULTS
In the inner ear, hemorrhagic changes were observed in the perilymphatic space, vestibular ganglion, and the stroma underlying the sensory epithelium, the cristae of the semicircular canals, and the maculae of the saccule and the utricle in both Stat1−/− and IFN-α/βγR−/− mice infected with LASV (Figure 1A and 1B). Systemic hemorrhagic changes were also observed in the central nervous system (data not shown). On the other hand, the hair cells and supporting cells constituting the sensory epithelium in the cristae of the semicircular canals and in the maculae of the saccule and utricle were not damaged (Figure 1B).
The vestibular histopathology of Lassa virus (LASV)-infected Stat1−/− and IFN-α/βγR−/− mice. A, Low-magnification view of the vestibular histopathology. The cristae of the anterior (Ant), lateral (Lat), and posterior (Post) semicircular canals, maculae of the saccule (Sac) and the utricle (Utr), the superior vestibular nerve (SVN), ganglion cells (G), perilymphatic space (PLS), cochlea (C), and cochlear nerve (CN) are shown. Arrowheads indicate hemorrhage. Black scale bars = 400 μm. B, High-magnification view of the histopathology of the cristae of the anterior semicircular canals, the maculae of the saccule and the utricle, and the superior vestibular ganglion cells. The hemorrhagic changes are significant, while the sensory epithelium is intact in LASV-infected mice compared to mock-infected mice. The stroma contains native brown pigments. Arrows indicate hair cells. Arrowheads indicate hemorrhage. White scale bars =100 μm.
LASV antigen was detected in the stroma of the hair cells and supporting cells in the cristae of the semicircular canals, the maculae of the saccule and the utricle, and the vestibular ganglion, both in Stat1−/− and IFN-α/βγR−/− mice infected with LASV (Figure 2A). The vestibular apparatus of the mock-infected mice did not show any presence of the LASV antigen.
Immunohistochemistry of LASV antigen and CD3-positive T lymphocytes in the vestibular apparatus. A, The LASV antigen was detected in the stroma underlying the hair cells and supporting cells in the cristae of the semicircular canals and the maculae of the saccule and the utricle, and the vestibular ganglion cells in mice infected with LASV. Anti-Lassa glycoprotein antibody was used. Arrows indicate LASV antigen-positive cells. Black scale bars = 100 μm. B, There was significant infiltration of CD3-positive T lymphocyte in the stroma of the cristae of the semicircular canals and the maculae of the saccule and the utricle, and around the vestibular ganglion cells in mice infected with LASV, compared to those in the mock-infected mice. Arrows indicate CD3-positive cells. Black scale bars = 100 μm. C, The LASV antigen and CD3-positive T-lymphocyte infiltration were detected in the PLS (surrounded by dashed lines) in LASV-infected Stat1−/− mice. Arrows indicate LASV antigen-positive cells. White scale bars = 400 μm. Abbreviations: IFN, interferon-α/β and -γ receptor; IHC, immunohistochemistry; LASV, Lassa virus; PLS, perilymphatic space; Post, posterior semicircular canal; Sac, saccule; Utr, utricle.
CD3-positive T lymphocytes infiltrated the stroma of the cristae of the semicircular canals and the maculae of the saccule and the utricle, and around the vestibular ganglion cells both in Stat1−/− and IFN-α/βγR−/− mice infected with LASV (Figure 2B). The CD3-positive T-lymphocyte infiltration appeared to be more prominent in LASV-infected Stat1−/− mice than in IFN-α/βγR−/− mice, presumably because IFN-α/βγR−/− mice are more severely deficient of the interferon pathway [6]. In addition, LASV antigen and CD3-positive T-lymphocyte infiltration were present in the perilymphatic space within the vestibular apparatus in LASV-infected Stat1−/− mice (Figure 2C).
DISCUSSION
Our observations of the vestibular sensory epithelium of our LF model mice parallel those found in the auditory apparatus, where cochlear hair cells remained intact despite severe hearing loss following LASV infection [4]. These findings suggest that the development of vestibular dysfunction due to LASV infection does not involve significant alterations in the epithelial mechanoreceptors. However, further ultrastructural investigations of vestibular hair cells and supporting cells would be essential to substantiate this hypothesis.
The LASV antigen presence in the vestibular apparatus indicates that the stroma of the vestibular hair cells and vestibular ganglion cells may be the target of the LASV infection. This finding was similar to vestibular dysfunction caused by infection of other viruses, such as herpes simplex virus, human immunodeficiency virus, or mumps virus, where the viral antigen was detected in the vestibular hair cells or/and vestibular ganglion cells [7–9]. The question of whether the vestibular pathophysiology results from direct viral infection, activation of the host immune system by infection, or a combination of both remains a subject of ongoing investigation.
Inflammation around the vestibular nerve is known to cause sudden vertigo and gait imbalance, which is speculated to be associated with viral infections such as a latent herpes simplex virus or influenza virus [10]. However, the specific virus responsible remains unidentified. In contrast to the significant degeneration in the spiral ganglion cells in the cochlea of LASV-infected Stat1−/− mice [4], LASV infection did not result in degeneration of the vestibular ganglion cells. This implies that while the cochlear ganglion cells are more susceptible to viral-induced degeneration, the vestibular ganglion cells may be resistant or affected differently. The presence of T lymphocytes in the severely damaged spiral ganglion in LASV-infected Stat1−/− mice related to the development of hearing loss [4]. Considering that T-lymphocyte depletion impacts the onset of hearing loss [5], we suggest that T-lymphocyte infiltration and host immune response are closely related to the onset of LASV-induced hearing loss in our model mice. Our LF model mice highlight a novel finding where the primary pathogenesis of audiovestibular dysfunction involves mechanical damage and/or dysfunction of the eighth cranial nerve. This model underscores its utility in studying such mechanisms, offering valuable insights for broader applications in understanding audiovestibular disorders caused by viral infections.
The intraperitoneal injection of other viruses has been reported to result in viremic spread to the spiral ganglion and vestibular ganglion cells, leading to structural disruption of the inner ear in hamster infected with reovirus and guinea pig infected with cytomegalovirus [11, 12]. Cytomegalovirus, well known as a virus associated with deafness, infects the inner ear constructing cells both directly and via hematogenous routes, triggering an immune response that affects vestibular function [12, 13]. In our study, the LASV distribution and the hemorrhagic changes in the vestibular ganglion cells raise questions about the exact pathways of infection and how these processes contribute to the observed vestibular dysfunction. Further research is needed to elucidate the mechanisms underlying LASV-induced imbalance.
The edema/hydrops of the vestibular perilymphatic space cause notable distortion of the spatial and planar orientation of the vestibular organs. The fluid imbalance between perilymph and endolymph underlies the audiovestibular pathogenesis of Ménière disease and hydrocephalus [14]. As the orientation of the vestibular organs to the head is important for the maintenance of healthy vestibular function [15], we speculate that the edema and cell infiltrations under the vestibular sensory epithelium can cause altered afferent vestibular signaling leading to vestibular dysfunction even without any cell damage.
Several viral infections, such as measles and mumps, are known to cause sensorineural hearing loss and balance dysfunction in the acute phase of the disease [9]. However, the LASV-induced audiovestibular dysfunction mostly occurs during convalescent phase, following the acute stage of illness [3]. Thus, we focused on the histology at 60 days postinfection, corresponding to the postacute phase. Our LF model closely mirrors the time course of audiovestibular dysfunction onset observed in human patients, providing a representative model that simulates human inner ear pathology caused by LASV infection.
In summary, our study revealed significant histopathological features in the inner ear vestibular apparatus of LF model mice, including hemorrhage, widespread distribution of LASV antigen, CD3-positive T-lymphocyte infiltration in the neuroepithelial stroma and perineural area, and edema/hydrops in the perilymphatic space. This study represents a pioneering report from a translational mouse model of viral infection-induced vestibular dysfunction, highlighting its potential as a crucial tool for understanding and addressing similar conditions in humans. This study highlights the critical need for an accurate model to study balance problems in human LF patients and emphasizes the future importance of vaccine safety and efficacy in preventing such complications. Understanding the mechanisms of LASV-induced vestibular dysfunction will aid in the development of safer and more effective vaccines to mitigate the impact of LF on survivors’ quality of life.
Note
Financial support. This work is supported by the National Institute on Deafness and Other Communication Disorders (grant number R01 DC021542); the National Institute of Allergy and Infectious Diseases (grant numbers U01AI151801, R01 AI129198, K99 AI156012); and the University of Texas Medical Branch, Institute for Human Infections and Immunity. M. H. S. was supported by the Uehara Memorial Foundation and the Astellas Foundation for Research on Metabolic Disorders. J. M. was supported by an overseas research fellowship from the Japan Society for the Promotion of Science. S. P. was supported by the John S. Dunn Foundation.
References
World Health Organization. Lassa fever. Fact sheet. https://www.who.int/en/news-room/fact-sheets/detail/lassa-fever. Accessed 6 November 2023.
Author notes
T. S. and M. H. S. contributed equally.
Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
