29
Immunologic Disorders of the Inner Ear

C. Arturo Solares, Gordon B. Hughes, and Vincent K. Tuohy

The concept that the inner ear is able to initiate an immune response that may lead to otologic dysfunction is relatively recent. There is also reasonable evidence that supports the role of organ-specific autoimmunity in a subset of patients with rapidly progressive sensorineural hearing loss (SNHL). Autoimmune sensorineural hearing loss (ASNHL) has been defined largely on clinical parameters. It typically produces a bilateral, sometimes fluctuating, hearing loss that progresses rapidly over weeks or months.¹ The diagnosis is made by excluding ototoxicity, systemic disease, and other factors that may induce rapidly progressive hearing loss, and by showing a therapeutic response to corticosteroid treatment.

Although autoantibodies and autoreactive T cells have been implicated in the etiopathogenesis of ASNHL, several central issues remain unresolved, including the relative prominence of B-cell or T-cell autoimmunity in the initiation and progression of ASNHL, and the identity of the putative inner ear self-antigen(s) that target ASNHL. The importance of these concepts is that if detected early enough, hearing loss, once thought to be progressive and irreversible, can be restored through appropriate and judicious use of immunosuppressive drugs. This chapter discusses seminal evidence that has led to our current understanding of inner ear immune responses and their relationship to clinical disease. In addition, a summary of the clinical aspects and therapeutic options for ASNHL is presented.

Immunology of the Inner Ear

The blood-brain and blood-labyrinthine barriers are important determinants of the immune responses in the brain and the inner ear, respectively.² Due to the blood-brain barrier, the brain is largely isolated from the systemic immune responses under normal conditions. This relative isolation led to the suggestion that the brain is an immunoprivileged site.³ However, it is now known that isolation of the brain from systemic immune system is far from complete because immune responses can be elicited in the brain under the right circumstances.⁴ Due to similarities between the blood-brain and blood-labyrinthine barriers, there was a traditional belief that the inner ear also was immunoprivileged, but this concept has been widely disproved by many authors.^5–7

Under normal conditions the inner ear contains immunoglobulins and immunocompetent cells.^5,8–10 Immunoglobulins can cross the blood-labyrinthine barrier and are present in inner ear fluids at higher concentrations relative to levels in the central nervous system (CNS).^5,8 Discolo et al,¹⁰ using CD45 immunolabeling, identified a small population of resident monocytes in mouse cochleas in the inferior spiral ligament. Other authors have identified the endolymphatic sac as a source of immunocompetent cells in the inner ear.¹¹ Thus, the inner ear not only derives its protection through diffusion of immunoglobulin from the systemic circulation and the CNS, but also contains a full array of immunocompetent cells.

Harris⁶ performed intracochlear keyhole limpet hemocyanin (KLH) immunization in guinea pigs previously primed to bovine serum albumen (BSA). He subsequently observed increased anti-KLH antibody levels in the inner ear perilymph without a corresponding increase in anti-BSA levels and without an increase in anti-KLH levels in the cerebrospinal fluid (CSF). Harris concluded that the inner ear was fully capable of initiating a local immune response to an antigen by a resident population of immunocompetent T cells.⁶ Several subsequent studies involving obliteration of the endolymphatic sac indicated that this inner ear structure known to be involved in endolymph drainage and absorption was also critical for maintaining local memory responses to antigens and initiating the efferent limb of the inner ear immune response.^11–15 Memory responses to inner ear KLH inoculation produced a hearing deficit and cochlear histopathology only when guinea pigs were immunized to KLH.¹⁶ This series of studies clearly showed that the inner ear was immunocompetent.

The molecular basis of the immune response has been extensively studied. Several mediators such as interleukins, interferons, and tumor necrosis factors have been shown to be present in the inner ear. Studies by Gloddek and Harris have helped elucidate the roles of interleukin-2 (IL-2) and transforming growth factor-β (TGF-β) in the cascade of events that takes place. IL-2 cannot be detected in the perilymph under unstimulated conditions, but is a component of the inner-ear immune response. Following inner-ear challenge of the scala tympani with KLH, IL-2 was measurable at 6 hours, peaked at 18 hours, and declined over 5 days. These same investigators also reported an early egress of polymorphonuclear leukocytes and the appearance of fewer monocytes and lymphocytes, and hypothesized that it could be secondary to the actions of IL-2.¹⁵ Yeo and Ryan¹⁷ studied the role of TGF-β in the inner ear. Leukocytes within the scala tympani and scala vestibuli were labeled with mRNA probes to TGF-β1. Following scala tympani challenge with KLH, this label was detected at 1 day, peaked at 3 days, and decreased by 1 week.¹⁷ TGF-β is a chemoattractant for monocytes, T cells, and neutrophils. It also increases the levels of IL-1, IL-6, and platelet-derived growth factor. TGF-β also interferes with the IL-2 response, deactivates macrophages, and inhibits production of interferon-γ (IFN-γ) and tumor necrosis factor-α (TNF-α). More recently, Hirose and Keasler¹⁸ reported that after acoustic injury, there was a substantial increase in inner ear expression of monocyte chemoattractant protein (MCP-1), TNF-α, and IL-1, suggesting that inflammation may have a role in repair after acoustic trauma. Thus, it has become increasingly clear that immune responses may occur in the inner ear and that these responses involve a complex cascade of events potentially activated by any of several mechanisms.

In addition to the damage incurred during responses to invading pathogens, the inner ear also may incur damage from autoimmune-mediated inflammation. There is evidence that the inner ear can be affected in a variety of non-organ-specific autoimmune diseases. Hearing loss is a rare complication of polyarteritis nodosa.19 Wegener’s granulomatosis has been associated with both middle and inner ear pathology.²⁰ Systemic lupus erythematosus can produce necrotizing vasculitis and progressive SNHL or disequilibrium.¹ Some literature reports have implicated rheumatoid arthritis in otologic pathology.²¹ Cogan’s syndrome, characterized by interstitial keratitis and vestibuloauditory dysfunction,²² may result from hypersensitivity response to one or more inflammatory agents associated with vasculitis.²³ However, lymphocyte proliferation on exposure to corneal antigen^24,25 and inner ear antigen,²⁶ suggests organ-specific autoimmunity.

ASNHL is a clinical entity characterized by bilateral, rapidly progressive hearing loss occurring over weeks to months in the absence of systemic immunologic disease. Although its name implies an autoimmune etiopathogenesis, data implicating self-recognition events in the development and progression of ASNHL until recently have been limited to its therapeutic response to immunosuppressive treatment, to detection of serum antibody to heat shock protein 70 kD (HSP70), and to recall responses of peripheral blood mono-nuclear cells (PBMCs) to crude inner ear homogenate. Its contemporary clinical definition is derived from a seminal study by McCabe,¹ whosesuspicions began when immunosuppressive therapy with corticosteroids and cyclophosphamide not only cleared up a patient’s nonhealing mastoid infection but coincidentally improved her rapidly progressive SNHL. Although some of McCabe’s earliest reported patients may have had Wegener’s granulomatosis, others clearly had isolated progressive SNHL and benefited from treatment with corticosteroids or cyclophosphamide. He proposed that inner ear-specific autoreactive T cells may mediate ASNHL, and several recent studies have provided support for T-cell-mediated ASNHL.^27,28 However, others have proposed that ASNHL may be the result of autoantibody-mediated injury to the inner ear.^29,30 As described below, our laboratory has recently developed a T-cell-mediated inner ear-specific autoimmune mouse model that mimics the clinical features of ASNHL.³¹ We hope that this murine model will facilitate identifying the stages leading to ASNHL and provide a venue for discovering new diagnostic markers and novel treatments for this disorder.

Autoimmune Sensorineural Hearing Loss

Autoreactive T Cells in Autoimmune Sensorineural Hearing Loss

The first compelling evidence that autoreactive T cells may be implicated in ASNHL was provided by McCabe and McCormick who observed leukocyte migration inhibition (LMI) in response to homogenized inner ear membranes in activated PBMC from all 54 of their ASNHL study subjects.³² The study did not incorporate normal control subjects or a non-inner-ear control antigen in the experimental design. Hughes and colleagues²⁷ further implicated T-cell autoreactivity in ASNHL by showing that PBMC from ASNHL patients elicited recall proliferative responses to human inner ear homogenate. PBMC from 13/58 (22%) ASNHL subjects with unilateral or bilateral asymmetric SNHL responded to human inner ear antigens. In comparison, only 1/15 (7%) of normal control subjects showed PBMC proliferation to inner ear homogenate. Although the positive predictive value of these proliferation studies was suggested to be 79%, proliferation may best be viewed as a functional IL-2 assay and as such provides limited ability to detect much of the autoreactive T-cell repertoire notoriously known to have cryptic features with low antigen affinity and low precursor frequencies.³³

Lorenz et al²⁸ used the enzyme-linked immunosorbent spot (ELISPOT) assay to measure secretion of specific cytokines by individual cells, thereby providing 10- to 200-fold increased sensitivity over conventional proliferation and enzyme-linked immunosorbent assay (ELISA) for detecting T-cell immunore-activity.^34,35 They found that the PBMC frequencies of IFN-γ-producing T cells specifically responsive to human inner ear homogenate were elevated significantly in 25% (3/12) of ASNHL patients but in none (0/12) of the age- and sex-matched normal control study subjects.²⁸ These findings indicated that proinflammatory effector T cells specific for inner ear antigens may play a pivotal role in the development and progression of ASNHL.

Although ELISPOT analysis showed autoreactivity in only 25% of ASNHL patients, the true incidence of self-recognition is likely much higher for several reasons: (1) Assessment of self-recognition at a single time point reduces the likelihood of detection, as autoreactivity is better detected when the analysis is performed serially.³⁶ (2) The use of IFN-γ or any single cytokine for evaluating precursor frequencies of antigen-specific T cells prevents identification of autoreactive T cells that produce other cytokines.³⁷ (3) Antigen-specific autoreactive T cells may not produce IFN-γ because they may instead make regulatory cytokines such as IL-10 and TGF-β. (4) In humans and mice, the autoreactive repertoire may be compartmentalized in such a way that T cells making IFN-β are underrepresented in the PBMC.³⁸ (5) Despite the high sensitivity of the ELISPOT assay, there is still enough “noise” in the assay to prevent detection of low-affinity autoreactive T-cell clones. Thus, ELISPOT detection of self-recognition in 25% of ASNHL patients may be viewed as a minimum rather than actual incidence of inner ear autoreactivity.

Although the ELISPOT provides enhanced sensitivity over LMI and proliferation for detecting T-cell autoreactivity, inner ear homogenate, although useful in determining autoreactivity, provides little help in identifying candidate self-antigens. An epitope-mapping peptide series derived from inner ear-specific proteins may represent a more ideal experimental design. Such overlapping peptides have been most useful in identifying T-cell epitopes that target human autoimmune diseases such as multiple sclerosis36,39 and insulin-dependent diabetes mellitus.^40,41 Proteins such as cochlin⁴² represent promising candidates for generating epitope-mapping peptides because their expression appears to be confined exclusively to the inner ear. Ideally, the ELISPOT assay should detect changes over time in the precursor frequencies of T cells producing a variety of proinflammatory (IFN-γ, IL-2, TNF-α, TNF-β) and antiinflammatory (IL-4, IL-5, IL-10) cytokines in response to overlapping peptides derived from candidate inner ear-specific proteins. Such studies are currently ongoing in our laboratory and may help clarify the development of self-recognition and identify candidate peptides for developing antigen-specific immunotherapies for ASNHL.⁴³

Autoantibodies in Autoimmune Sensorineural Hearing Loss

In addition to evidence supporting the involvement of autoreactive T cells in ASNHL, several studies have implicated autoantibodies as potential mediators of inner ear injury. Arnold and colleagues⁴⁴ showed that sera from 15/21 (71%) patients with bilateral SNHL of unknown etiology contained antibodies capable of immunostaining human inner ear tissues. Although this study did not include normal control subjects, a later report from the same group confirmed the initial findings by showing immunostaining of human inner ear tissues with serum from 64/119 (54%) subjects with hearing loss compared with 1/25 (4%) normal control subjects.⁴⁵ The single normal control subject who showed positive serum immunostaining was subsequently diagnosed with rheumatoid arthritis (RA). These studies, however, did not show a strong correlation between the presence of inner ear-specific serum antibodies and a therapeutic response to corticosteroid treatment, a clinical hallmark of ASNHL.

In more recent studies, Western blot analysis has been used to study reactivity from patients with ASNLH against bovine inner ear material, and it appears that a 68-kD protein is the likely putative antigen.⁴⁶ Moscicki and colleagues⁴⁷ used Western blot analysis to show that sera from 89% of patients with actively progressing bilateral hearing loss reacted with a 68-kD protein constituent of inner ear extract, whereas patients with inactive disease showed no immunoreactivity. Moreover, patients who were antibody-positive showed a significantly increased incidence of responsiveness to corticosteroid treatment compared with antibody-negative patients. Similar results were subsequently obtained by Gottschlich and colleagues,⁴⁸ whose compiled data showed that 90/279 (32%) patients with bilateral rapidly progressive SNHL had elevated anti—68-kD titers, whereas only 5% of control subjects were seropositive. Thus, these studies correlated immunoreactivity directed against a 68-kD antigen with actively progressing bilateral hearing loss.

Considerable evidence indicates that the 68-kD antigen targeted by autoantibody in ASNHL is HSP70, a heat shock protein whose synthesis is greatly enhanced in a variety of tissues following exposure to various stressors, including autoimmunity.^49–53 Monoclonal antibody specific for HSP70 binds the 68-kD antigen, and anti—68-kD sera from ASNHL subjects predominantly target the C-terminal p427–461 region of HSP70.^54,55 Moreover, evidence indicates that a positive Western blot for HSP70 may correlate with corticosteroid responsiveness in ASNHL subjects in a manner similar to seropositive 68-kD immunoreactivity.^{47,48,52,56,57} Indeed, it has been postulated that autoantibodies against HSP70 may be implicated directly in the pathogenesis of ASNHL^52,53; however, immunization of BALB/c or CBA/J mice with bovine HSP70 induced high titer antibody responses to HSP70 without any subsequent changes in auditory brainstem responses (ABRs).⁵⁸ Thus, there is no direct proof that HSP70 antibodies are immunopathogenic or cochleopathic in ASNHL. Although the pathogenicity of anti-HSP70 antibodies remains in question, Western blot detection of anti-HSP70 antibodies still provides a useful laboratory marker for supporting the diagnosis of ASNHL with a specificity of 90% and a positive predictive value of 91%.⁵⁶ However, with a sensitivity of only 42%, the need to develop a more sensitive assay is obvious.

Boulassel and colleagues⁵⁹ showed that ASNHL patients often have elevated immunoglobulin G (IgG) antibody titers to cochlin, an inner ear-specific protein.⁵⁹ Cochlin (formerly known as coch-5B2) is a product of the COCH gene mapped in humans to chromosome 14q12-q13, and its mutations are associated with an autosomal dominant, nonsyndromic, progressive SNHL with vestibular pathology (DFNA9).^42,60,61 Cochlin is an integral part of the inner ear extracellular matrix and is expressed in the regions of the fibrocytes of the spiral limbus and of the spiral ligament—inner ear regions showing histologic abnormalities in humans with COCH mutations.^42,62 Cochlin appears to be the most abundant protein expressed in inner ear tissues,⁶³ and as such is likely to have a high level of constitutive presentation by local professional antigen-presenting cells, thereby making it a probable target in autoimmune disease of the inner ear and a prime candidate for developing diagnostic markers.

Animal Models for Autoimmune Sensorineural Hearing Loss

The inability to examine temporal bone histopathology in humans during the active phase of disease has impaired identification of inner ear-specific antigens that target ASNHL, elucidation of the sequence of inflammatory and immune events leading to the development of ASNHL, and development of novel therapeutic strategies for preventing progressive hearing loss. Several animal models have been described for the study of ASNHL. In this section we review some of the most relevant model systems used in ASNHL studies not included in other sections of this chapter.

Despite its lack of inner ear-specific expression and its ubiquitous presence in a variety of organs, collagen type II (CII) has been proposed as a potential target antigen in ASNHL. Yoo and colleagues64 observed significantly decreased amplitudes and delayed latencies in ABR recordings following immunization of female Lewis rats with either bovine or ovine CII. Because hearing loss did not occur in rats immunized with type I collagen, the authors proposed that CII autoreactivity played a key role in the pathogenesis of ASNHL. In subsequent studies, this same group showed that CII-immunized rats, guinea pigs, and chinchillas underwent otospongiotic changes in the osseous labyrinth of the inner ear as well as degeneration of spiral ganglion cells and atrophy of the cochlear nerve, organ of Corti, and stria vascularis. In addition, endolymphatic hydrops, hearing loss, and vestibular dysfunction were observed in all CII-immunized animals:^64,65 Although the observed CII-induced inner ear abnormalities were quite striking, the results have not been reproduced. Using the same CII immunization protocol to prime Wistar-Furth rats, Harris and colleagues³⁰ were unable to induce hearing loss within 11 months after immunization; CII-associated inner ear pathology was not observed despite high titers of serum and perilymph anti-CII antibodies.

MRL/MpJ-lpr/lpr (MRL/lpr) and C3H/lpr mice have been used recently in ASNHL studies because they develop spontaneous hearing loss secondary to a systemic lymphoproliferative disorder. The Ipr gene is an autosomal-recessive Fas deletion mutant responsible for failure of Fas-mediated apoptosis, which produces a nonspecific lymphoproliferative disorder associated with spontaneous development of various autoimmune disorders, including systemic lupus erythematosus, glomerulonephritis, polyarteritis, RA, and sialoadenitis.^66–68 Several authors have reported hearing disorders associated with expression of the Ipr gene.^69–72 Affected animals have circulating antibodies directed against blood vessels in the stria vascularis,⁷³ breakdown of endothelial tight junctions that make up the stria blood-labyrinth barrier,⁶⁹ and corticosteroid-responsive auditory dysfunction.⁷⁴

Recent studies have indicated that the NZB/kl substrain of the autoimmune-prone NZB mouse strain spontaneously develops high-frequency hearing loss with age as determined by elevated ABR thresholds. Hearing loss appears to be due to thickening of the capillary basement membrane of the stria vascularis as a result of IgM and IgG immune complex deposition.^75,76 The similarly derived NZB/san substrain failed to develop spontaneous hearing loss.⁷⁶ The hearing loss observed in NZB/kl mice is the result of a systemic disorder that involves accompanying renal pathology.⁷⁷ Thus, despite its usefulness as an immune inflammatory model for inner ear pathology, there is no evidence of organ-specific autoimmunity in the hearing loss that occurs in NZB/kl mice.

Autoimmune SNHL is likely mediated by autoreactivity targeted against antigens that are inner ear-specific rather than systemically expressed. Harris⁷⁸ induced cochlear lesions in guinea pigs following immunization with either autologous or bovine inner ear homogenates. Although hearing loss occurred in 12/38 (32%) total ears, there was no correlation between degree of hearing loss, histologic changes, and serum antibody titers to inner ear homogenate. Soliman^79,80 also immunized guinea pigs with bovine inner ear homogenate and showed that 36% of the animals developed endolymphatic hydrops, whereas 20% showed ABR documented hearing loss. In addition, immunoglobulin deposition was evident in inner ear structures including the basilar membrane, endolymphatic sac, and mid-modiolar blood vessels. Similarly, Yamanobe and Harris⁸¹ induced labyrinthitis in guinea pigs after immunization with bovine inner ear homogenate. Animals developed hearing loss by day 7 and were shown to have cellular infiltration of the inner ear that regressed by 4 weeks postimmunization.

Guinea pigs and mice immunized with chicken and guinea pig inner ear tissues developed hearing loss associated with production of serum antibodies to hair cell stereocilia. However, hearing loss was transient.⁸² A pathogenic role for T-cell-dependent immunoglobulins was clearly apparent in studies showing hearing loss in guinea pigs injected with mouse monoclonal antibody (MAb) KHRI-3, an IgG1 generated against guinea pig cochlear hair cells.⁸³

Stay updated, free articles. Join our Telegram channel

Join