Connection between the inner ear and the lymphatic system

OBJECTIVE

The aim of this study was to identify the lymphatic drainage of the inner ear in guinea pigs.

STUDY DESIGN

Prospective study.

METHODS

The prospective study was performed in guinea pigs by injection of keyhole limpet hemocyanin (KLH) into either the right-side scala tympani or the middle ear cavity. The left side was not injected and served as a control. Fifteen minutes after injection, the animals were killed by intracardiac perfusion with paraformaldehyde and tissue specimens (right and left temporal bones, cervical lymph nodes, and the spleen) were collected. The presence of KLH in each specimen was determined by immunohistochemical assay of frozen sections using polyclonal mouse anti-KLH antibodies.

RESULTS

After injection into the middle ear, labeled cells were identified in the parotid, superficial ventral, mandibular, and deep cranial cervical lymph nodes. However, after inner ear injections KLH was present in only the parotid and superficial ventral cervical nodes. The spleen contained KLH-positive cells following injection into either the middle or inner ear, but not all animals contained labeled spleen cells.

CONCLUSIONS

The inner ear has a connection to the lymphatic drainage system. Because fewer lymph nodes contained labeled cells after inner ear injection than after middle ear injection, it is concluded that the inner ear does not simply drain to the middle ear and subsequently to the lymph nodes but seems likely to have its own connections.

Mitochondrial cytochrome oxidase immunolabeling in aged human temporal bones

Presbycusis, an age-related hearing loss, is accompanied by histopathological cochlear changes including variable amounts of degeneration of the auditory receptors, neurons and the stria vascularis. The causes of degeneration are unknown, although acoustic trauma and exposure to ototoxic agents are certainly contributors to the cellular degeneration. Acquired mitochondrial DNA defects are postulated as important determinants of aging in neuromuscular tissues. The cochlear neurons are highly metabolic and are, therefore, likely to be affected by mitochondrial DNA defects. Sequence analysis has demonstrated a significant number of acquired mutations in the cytochrome oxidase gene in the neurons from aged human cochleas. The current study used immunohistochemical labeling of cytochrome oxidase in the neuronal cell bodies in archival celloidin sections to evaluate relationships among label density, hearing loss, number of neurons and mitochondrial DNA changes within individual cochleas. Label density was less in many aged temporal bones, but not all. There was no relationship among any other variables. It is concluded that while there may be a decrease in the amount of cytochrome oxidase expression in aged spiral ganglion cell bodies, there are many other factors that contribute to hearing loss and cellular degeneration.

Autoimmune inner ear disorders

There is considerable evidence to suggest that hearing and vestibular function can be influenced by autoimmune processes. A number of systemic autoimmune disorders include hearing loss and vertigo as part of their constellation of symptoms. Although classic evidence for a specific autoimmune etiology mediated by immune response directed solely at the inner ear is elusive, it appears that autoimmune damage can also exist as an entity confined to the labyrinth. Antigenic targets of autoimmunity within the labyrinth seem to be diverse. Partly because of this, the condition is difficult to diagnose. However, autoimmune disorders of the inner ear are of special interest since they are among the few forms of hearing loss that are amenable to medical treatment. Recent progress in understanding the etiology, diagnosis and treatment of autoimmune damage to the inner ear is reviewed.

Immunohistochemistry and microwave decalcification of human temporal bones

Processing of human temporal bones is a long, expensive process and the resulting celloidin sections are difficult to use for immunohistochemistry. We tested the ability of immunohistochemical assays to work in human temporal bones that were decalcified using a microwave oven. Tissue was trimmed to an approximate cube (1.5-2 cm/side) containing only the cochlea and immersed in fresh EDTA with paraformaldehyde every 6 h. This sized block required 190-400 h to decalcify. The decalcified tissue was embedded in paraffin and sectioned. Sections were immunoassayed with anti-cytochrome c oxidase, anti-neurofilament or anti-peripherin. All three antibodies labeled the appropriate structures. This procedure may stimulate advancement in the understanding of human inner ear pathology.

Intratympanic immunosuppressives for prevention of immune-mediated sensorineural hearing loss

HYPOTHESIS

Round window membrane application of immunosuppressives reduces cochlear inflammation and hearing loss in a guinea pig model of sterile labyrinthitis.

BACKGROUND

Systemic immunosuppressives are used to treat sensorineural hearing loss due to inflammation (autoimmune, bacterial, viral), which in animal models causes hearing loss. Transtympanic application of drugs is an attractive and possibly efficacious method of treatment that avoids systemic toxicities.

METHODS

Sterile labyrinthitis was created using keyhole limpet hemocyanin (KLH). Dexamethasone (0.048 mg/day and 0.288 mg/day), cyclosporine (0.5%), prednisolone acetate (1%), fluorouracil, (5%), and FK506 (0.01%) were delivered to the round window membrane with one injection (filling middle ear space) or osmotic minipumps. Efficacy was evaluated by auditory brainstem response and histology.

RESULTS

No drug was effective at reducing hearing loss or inflammation.

CONCLUSION

Local application of immunosuppressives did not suppress inner ear inflammatory infiltrates and hearing loss in KLH-induced labyrinthitis in a guinea pig model.

Characterization of an experimentally induced inner ear immune response

OBJECTIVE

To determine the effects of a sterile immune response on the structure and function of the cochlea.

METHODS

An immune response was created in guinea pigs by systemically sensitizing the animals to keyhole limpet hemocyanin and subsequently challenging the inner ear with the protein. Animals were allowed to survive for 1 to 5 weeks, after which the cochlea was evaluated histologically. Hearing was measured by auditory brainstem response before the inner ear challenge, during the survival period, and prior to sacrifice.

RESULTS

Inflammatory cells infiltrated the cochlea from the circulation. Surface preparations and plastic sections of the organ of Corti 1 and 2 weeks after the initiation of the inflammation demonstrated degeneration of the sensory and supporting cells in cochlear turns containing inflammatory cells. Good preservation of structures was seen in the more apical cochlear turns with little or no inflammatory cells. In cochleas from animals that survived 5 weeks, most of the infiltrated cells were cleared after undergoing apoptosis and the inflammatory matrix in the scala tympani began to calcify. Hearing loss was moderate to severe depending on the amount of inflammation.

CONCLUSION

Although in general the immune response serves to protect an organism from infection, these results demonstrate that bystander injury associated with local immune responses in the cochlea, an organ incapable of regeneration, causes permanent cochlear destruction and hearing loss.

Effects of a hair cell transcription factor, Brn-3.1, gene deletion on homozygous and heterozygous mouse cochleas in adulthood and aging

The transcription factor Brn-3.1, is expressed in the inner ear hair cells throughout life and is necessary for the development of these cells. Mutant mice in which the Brn-3.1 encoding region has been deleted have no identifiable hair cells, greatly reduced numbers of spiral ganglion cells and are deaf. A mutation in the human homologue of this gene has been shown to be related to adult onset, sensorineural hearing loss (Vahava et al., 1998). The question whether haploinsufficiency in the mutant Brn-3.1 mouse with a mixed C57BL6/129Sv genetic background could affect the adult or aged cochlea was tested, therefore, by measuring the auditory brainstem responses and examining the cochlea's histologically at 2, 18 and 24 months of age. The heterozygotes had a comparable hearing to the wild-type animals and similar patterns of cochlear degeneration. Both groups showed an about 30 dB hearing loss beginning at 18 months of age, outer hair cell degeneration and loss of spiral ganglion neurons in the basal turn. There appeared to be no effect of Brn-3.1 haploinsufficiency on the mouse cochlea, implying that one intact copy of the gene is sufficient to maintain a normal cochlea.

Ultrastructural pathology in the stria vascularis of the MRL-Fasl(lpr) mouse

The MRL-Fas(lpr) mouse, a model of multisystemic, organ nonspecific autoimmune disease, has been proposed as a model of immune-mediated inner ear disease. A preliminary study employing light microscopy indicated that it develops cochlear pathology that appeared most striking in the stria vascularis, where cells underwent edema and degeneration. However, other structures, including the inner and outer hair cells and the supporting cells, also appeared to display pathology. The current study analyzed cochlear ultrastructure using transmission electron microscopy to better delineate the cochlear lesions found in these animals. MRL-Fas(lpr) animals were allowed to develop systemic disease (20 weeks old) and then had auditory brainstem response (ABR) thresholds determined. Animals were then killed and their cochleas prepared for electron microscopy. Age-matched MRL-+/+ and BALB/c mice served as controls. Results indicated that MRL-Fas(lpr) mice demonstrated elevated ABR thresholds. In contrast to a preliminary report, the cochlear pathology was observed exclusively in the stria vascularis, where cells demonstrated hydropic degeneration. Strial capillary structure was normal as were the rest of the cellular cochlear constituents. No inflammatory infiltrate was noted. These studies confirm that the MRL-Fas(lpr) mouse develops cochlear abnormalities focused in the stria vascularis. Whether the mechanism of the cellular degeneration involves autoimmune, genetic, or uremic processes has yet to be determined.

Application of labeling techniques to archival temporal bone sections

Pathology of the human inner ear has traditionally been studied in celloidin-embedded, hematoxylin and eosin-stained sections of the temporal bone. Although the traditional histologic approach has yielded valuable information, it is now possible to extend these studies to include analysis of molecules using immunohistochemical and histochemical staining techniques. Fourteen antibodies and 6 lectins have been applied to 420 archival, celloidin-embedded human temporal bone sections. Tissues had been fixed in 10% formalin, embedded in celloidin, sectioned, and stored for as many as 40 years. The staining intensities varied among sections, so they were ranked from 'no label" to "dense label." To investigate the relationships between the extent of postmortem changes (PMCs), storage time, and staining intensity for each antibody, the sections were graded according to their PMCs, which ranged from good preservation of the temporal bone histologic structure to severe postmortem autolysis. Although statistical analysis indicated that both extent of PMCs and storage time in general decrease the staining intensity, both poorly fixed tissue and sections stored for a long time can yield good immunostaining results with some antibodies.

GDNF protects the cochlea against noise damage

Glial-derived neurotrophic factor (GDNF) was tested for its ability to prevent hearing and sensory cell loss in guinea pigs exposed to acoustic trauma. Hearing was measured prior to any treatment. Animals were exposed to damaging levels of noise either before or after local application of GDNF to one ear. Four weeks later, hearing and sensory cell loss was greater in the control ear than in the ear receiving GDNF before acoustic trauma or 2 h after trauma, but not 4 or 6 h after trauma. The results indicate that GDNF treatment in vivo can prevent cochlear sensory cell damage and hearing loss if present during or shortly after acoustic trauma.

Immunohistochemical analysis of proliferating cells in a sterile labyrinthitis animal model

Inflammatory reactions within the cochlea lead to the formation of fibrotic tissue and bone. To determine which cells are involved in the proliferation of the inflammatory response within the cochlea, sterile labyrinthitis was created by inoculating keyhole limpet hemocyanin (KLH) into the scala tympani of systemically sensitized animals. Cellular proliferation was assessed immunohistochemically using the monoclonal antibody Ki-67. Proliferating cells were identified among inflammatory cells and fibroblasts within the matrix, as well as in endosteal cells lining the scala tympani. Inflammatory and potential osteoprogenitor cells were labeled as late as 6 weeks after inoculation, suggesting the absence of a strong immunosuppressive mechanism. Endosteal cells may proliferate and secrete the extracellular matrix used by the inflammatory cells to move within the cochlear scalae. They may also participate in the ossification of the inflammatory matrix.

Clinical diagnoses associated with histologic findings of fibrotic tissue and new bone in the inner ear

Fibrotic tissue or new bone occurs following inner ear inflammation, fracture, or surgery. The prevalence is unknown and was investigated using the National Temporal Bone, Hearing and Balance Pathology Resource Registry database. A search yielded 264 temporal bones with diagnoses of otosclerosis, tumor, Meniere's disease, meningitis, labyrinthitis, chronic otitis media, autoimmune disease, temporal bone fracture, or sensorineural hearing loss. All autoimmune cases contained some new bone, whereas only 20% to 30% of the labyrinthitis/meningitis cases were reported to contain new bone. Otosclerosis, Meniere's disease, and otitis media had relatively few cases containing new bone. Although new bone may derive from surgical trauma, it is also likely to be a result of the disease process. It seems that all these disease processes may contain a common feature that acts as a stimulus to induce fibrosis or bone growth in the inner ear.

Immunopathology of the inner ear: an update

We have reviewed the events of an inner-ear immune response. The perilymph contains antibody, presumably derived from the systemic circulation and CSF, which would allow for neutralization and help with opsonization and complement fixation. The endolymphatic sac contains immunocompetent cells capable of processing and presenting viral or bacterial antigen, potentiating the immune response, attacking the invaders directly or attacking infected cells, and developing immunoglobulin responses in situ. The early release of mediators such as IL-2 likely emanate from the endolymphatic sac and result in potentiation and regulation of the response and may assist in changes in the SMV, including expression of ICAM-1, which aid in the egress of immune cells from the systemic circulation. PMNs arrive first, followed by T cells and B cells, with secretion of specific antibody a relatively late event. Concomitant with the increase in cellular constituents is the formation of a dense extracellular matrix. The inner ear appears to have remarkable difficulty in clearing this matrix, ultimately resulting in ossification. The immune response is unfortunately deleterious to the inner ear, resulting in degeneration of the organ of Corti, stria vascularis, and spiral ganglion. Hearing loss is consistently seen following sterile and virally induced labyrinthitis. The inner ear also appears to be a target for autoimmune disease. While inner-ear damage has been described as part of non-organ-specific autoimmune disease, specific disease against the hearing apparatus is also likely. Experimental paradigms have allowed alterations of both the afferent and efferent limbs of this response; ultimately, with the hope that we can alter the course of the response and the subsequent damage in patients.

Comparison of anti-heat shock protein 70 (anti-hsp70) and anti-68-kDa inner ear protein in the sera of patients with Meniere's disease

The 68-kDa antigen detected in the sera of patients with autoimmune inner ear disease is known to represent the highly inducible heat shock protein 70 (hsp70). To evaluate the existence of anti-hsp70 in the sera of patients with Meniere's disease and to develop a more reliable method to detect this antibody, the sera of patients and controls were examined. Bovine kidney (MDBK) cells were cultured and some of them were heat shocked. Proteins in the cells were separated using sodium dodecyl sulfate polyacrylamide gel electrophoresis. Sera were reacted simultaneously with the blots of non-heat-shocked cells and heat-shocked cells. The serum was considered positive if the band in the 70-kDa location was denser in the lane with heat-shocked cells relative to non-heat-shocked cells. Presence of the antibody against the 68-kDa protein was compared with the result of immunoblotting with MDBK cells. In immunoblotting with MDBK cells, 33.3% of patients with Meniere's disease had anti-hsp70, while in the control group, only 5% had this antibody. Of the 60 cases, 13 were positive against both hsp70 and the 68-kDa protein, whereas 7 were positive only against hsp70 and 6 only against the 68-kDa protein. These differences appeared to result from the greater sensitivity of the differential anti-hsp assay and from difficulties in interpreting the results in blots with bovine inner ear extracts because of faint, broad, or overlapping multiple bands. Quite a number of patients with Meniere's disease have anti-hsp70, and it may be indicative of an immune etiology of the disease. The Western blot using heat-shocked and non-heat-shocked cells could be a reliable method to detect this antibody.

Cochlear spiral ganglion cell degeneration in wild-caught mice as a function of age

Presbyacusis in humans is an age-related bilateral sensorineural hearing impairment generally associated with degeneration of cochlear hair cells and spiral ganglion cells (SGC) predominantly in the basal turn but present in the apical turn. Investigations of cochleas of aged rats and gerbils reveal a large loss of SGCs in the apical as well as the basal turns. Genetically inbred aged mice, on the other hand, seem to have variable amounts of SGC loss beginning in some strains very early in the life span of the animals and greatest in the basal turn. Three age groups of wild-caught, then laboratory-bred, mice were investigated to determine the pattern of SGC degeneration. In 18-19-month-old animals the main loss of SGCs occurred in the basal turn (49% loss compared to 2-3 months) followed by the apical turn (31%). The greatest SGC losses in the 28-31-month-old animals were in both the apical (76%) and basal turns (74%). Thus, this strain of mice is similar to other rodents in that both ends of the ganglion are affected by SGC degeneration associated with aging.

Reduction of endotoxin-induced inflammation of the middle ear by polymyxin B

Endotoxin (ET) is an aggregate of lipo-oligosaccharide and protein found in the cell wall of gram-negative bacteria. A potent mediator of inflammatory responses, ET has been detected in middle ear effusions from patients with otitis media with effusion and chronic suppurative otitis media and used to induce inflammation of the middle ear mucosa and disruption of mucociliary transport in experimental animals. Polymyxin B, a polypeptide antibiotic, has been shown to bond to and inactivate the ET molecule. This study investigated the efficacy of polymyxin B as a modulator of the inflammatory response to endotoxin in the middle ear. In a guinea pig model, cellular infiltrate, effusion volume, and mucosal edema in response to ET were reduced in the presence of polymyxin B. These results suggest a potential role for the use of polymyxin B in the management of middle ear effusion.

Role of transcription factors Brn-3.1 and Brn-3.2 in auditory and visual system development

The neurally expressed genes Brn-3.1 and Brn-3.2 (refs 1-6) are mammalian orthologues of the Caenorhabditis elegans unc-86 gene that constitute, with Brn-3.0 (refs 1-3,8,9), the class IV POU-domain transcription factors. Brn-3.1 and Brn-3.2 provide a means of exploring the potentially distinct biological functions of expanded gene families in neural development. The highly related members of the Brn-3 family have similar DNA-binding preferences and overlapping expression patterns in the sensory nervous system, midbrain and hindbrain, suggesting functional redundancy. Here we report that Brn-3.1 and Brn-3.2 critically modulate the terminal differentiation of distinct sensorineural cells in which they exhibit selective spatial and temporal expression patterns. Deletion of the Brn-3.2 gene causes the loss of most retinal ganglion cells, defining distinct ganglion cell populations. Mutation of Brn-3.1 results in complete deafness, owing to a failure of hair cells to appear in the inner ear, with subsequent loss of cochlear and vestibular ganglia.

Late sequelae of cochlear infection

Inflammatory reactions within the inner ear are deleterious to cochlear function and can result in server hearing loss that does not recover. This study investigated a guinea pig model of long-term cytomegalovirus infection. At high doses active inflammation was still present after 35 days. At lower doses some ears showed partial resolution while others were still inflamed. Hearing was totally lost in all cases of persistent inflammation. There was some residual hearing in the cases that had resolved. Cochlear structures including the organ of Corti, stria vascularis, and spiral ganglion were partially degenerated. Fibrotic matrix within scala tympani was ossified in many cases. These changes are consistent with those described for human cochleas following putative viral infections.

Assessment of serum antibodies in patients with rapidly progressive sensorineural hearing loss and Menière's disease

The immunoreactivity of sera from patients with rapidly progressive sensorineural hearing loss (SNHL) or Menière's disease with bovine inner ear material was determined using the Western blot technique. Patients with other otologic conditions, autoimmune disorders, or arthritic disorders and age-matched randomly chosen patients with no hearing complaints served as controls. Twenty-two percent of the patients with bilateral rapidly progressive SNHL and 30% of the patients with Menière's disease had antibodies that reacted with a 68 kd antigen in the inner ear material. In the control groups, the incidence of reactivity was 5.0% (P < .001). When the results of this study were compiled with data collected previously, it was found that of 279 patients with bilateral rapidly progressive SNHL, 90 (32%) were positive with the 68 kd protein. Thus, the anti-68 kd antibody may provide a good marker for an immune etiology of these patients' hearing loss.

Na,K-ATPase in the cochlear lateral wall of human temporal bones with endolymphatic hydrops

Meniere's disease has traditionally been thought to arise from a disruption in longitudinal endolymphatic flow. This view has been brought into question by recent experimental studies that have focused attention on derangements of cochlear fluid and electrolyte homeostatic mechanisms in Meniere's disease, including abnormalities in Na,K-ATPase enzymes found in the cochlear lateral wall. The current study examined the immunohistochemical labeling pattern of the major ion-transporting enzyme of the stria vascularis, Na,K-ATPase, in archival sections of hydropic and nonhydropic human temporal bones for increased density of label that could indicate overproduction of fluid. The results showed good labeling of the stria vascularis in the celloidin sections. The hydropic ears tended to have darker label, but the difference was not statistically significant. The findings are consistent with normal functioning of the stria vascularis in cases of Meniere's disease.

Antigen diffusion from the perilymphatic space of the cochlea

The diffusion pattern of horseradish peroxidase (HRP) injected into the scala tympani of the cochlear basal turn of guinea pigs was studied to test whether antigen presented in this manner can gain access to the endolymphatic sac. By two hours, HRP reaction product was found throughout the cochlea, with the greatest amounts in the spiral ligament, spiral limbus, basilar membrane, and organ of Corti. In several cochleas, very weak labeling was seen in the stria vascularis. HRP reaction product was maximal in the basal turn. By two hours, HRP reaction product was also observed in the endolymphatic sac lumen, epithelial cells, subepithelial tissue, and perisaccular connective tissue. It was more common in the proximal portion. At this time, macrophages within the lumen already appeared to have phagocytosed the HRP. By 72 hours after injection, the inner ear was cleared of HRP. The results of this study support the hypothesis that antigen in the scala tympani gains access to the endolymphatic sac lumen, where it may be presented by macrophages to the systemic immune system. Antigen most likely does not gain access to the endolymphatic space in the cochlea, but it gets to the endolymphatic sac through the perilymph and the perisaccular tissue.

Evidence linking the 68 kilodalton antigen identified in progressive sensorineural hearing loss patient sera with heat shock protein 70

Immunoblotting against bovine inner ear extracts has previously identified a 68 kd antigen reactive with 22% to 58% of sera of patients with rapidly progressive sensorineural hearing loss (PSNHL) of suspected autoimmune causation. Efforts to purify and characterize this diagnostic antigen suggest that it is ubiquitous rather than inner ear-specific, and may represent the highly inducible heat shock protein (hsp) 70. The antigens identified by PSNHL sera and anti-hsp 70 monoclonal antibodies copurify on ion exchange and adenosine triphosphate affinity chromatography, and comigrate on one- and two-dimensional gel electrophoresis. Additionally, immunoblotting with positive patient sera shows dramatically increased expression of the 68 kd antigen by bovine kidney cells following heat shock in culture. Reactivity with stress proteins of various classes has been reported in a number of autoimmune diseases; however, anti-hsp 70 appears uniquely associated with ulcerative colitis and PSNHL.

Fibronectin-like immunoreactivity of the basilar membrane of celloidin-embedded human temporal bone sections

Dysfunction of the mechanical properties of the basilar membrane is a potential cause of presbycusis. In cases of minimal sensorineural or strial degeneration it is believed to play a major role. The membrane has been shown to be partly composed of fibronectin. Fibronectin immunoreactivity is diminished in aged rats. Mesothelial cell line the perilymphatic surface of the membrane and are reduced in number in the aged rat cochlea. Fibronectin immunoreactivity was examined in human temporal bone sections (6 months to 92 years old). Hematoxylin and eosin stained section (17 to 97 years) were immunoreactivity was demonstrable in the human cochlea, but was not reduced, even in the eldest cases examined The number of mesothelial cells was reduced, however, and was related to the age of the individual, but not to the clinical diagnosis or audiogram shape. These two factors do not, therefore, appear to give rise to hearing losses associated with presbycusis.

2-Deoxyglucose uptake patterns in response to pure tone stimuli in the aged rat inferior colliculus

The tonotopic map of the inferior colliculus (IC) of aged rats (25 months old) was examined to determine whether age-related changes known to occur in the cochlea are reflected in the 2-deoxyglucose (2-DG) uptake pattern of the IC. Because aged animals have hearing losses, auditory brainstem response thresholds were measured. Animals with threshold shifts of no greater than 30 dB relative to young animals were used. Animals were injected with radiolabeled 2-DG and stimulated with continuous pure tones presented at 70 dB above the behavioral thresholds for young animals at either 1, 4 or 32 kHz for one hour in a sound attenuated booth. The stimulus sound pressure levels were chosen to achieve comparable sensation levels between the young and aged animals. The tonotopic map of the IC in aged rats was different from that reported previously for young animals (Ryan et al., 1988), in that, the regions stimulated by 1 and 4 kHz were shifted towards the higher frequencies and the uptake areas were twice as broad for the aged animals as for the young animals. The observed 2-DG uptake patterns are consistent with an activation pattern of a high intensity stimulus and a loss of responsive elements in the cochlear apex. Similar broad and shifted bands of activated tissue may contribute to difficulties in auditory perception in aged humans with increased thresholds and sound amplification.

Development of high endothelial venule-like characteristics in the spiral modiolar vein induced by viral labyrinthitis

The postcapillary venule is known to be the preferred site of lymphocyte migration into the extravascular compartment. In this study, induction of these high endothelial venule (HEV)-like vessels has been investigated by inoculation of live cytomegalovirus or inactivated virus into scala tympani of guinea pig cochleas. Endothelial cells of the spiral modiolar vein (SMV) were evaluated histologically 2 to 6 days following inoculation. The experimental group endothelial cells had larger nuclei and increased amounts of cytoplasm. In addition, many lymphocytes were adherent to and present within the vascular wall. Control SMV endothelia showed normal structure and morphology throughout the study. These findings reiterate that nonlymphoid tissues may develop HEV-like characteristics and suggest that HEV-like morphology may be acquired during an acute inflammatory condition in the cochlea.

Evidence of direct communication of bone marrow cells with the endolymphatic sac in experimental autoimmune labyrinthitis

In experimental autoimmune labyrinthitis, we found that the guinea pigs sensitized with bovine inner ear antigen (IEAg) developed cellular infiltration within the endolymphatic sac (E. sac). In this study, we investigated the distribution of immune mediated cells within bony vascular channels. The channels, which are normally found around the E. sac, are tiny and contain few immune-mediated cells. In contrast, in guinea pigs sensitized with IEAg bony vascular channels were large and contained immune-mediated cells. We hypothesize that immune-mediated cells in these channels are the result of the immune response directed against inner ear antigen, and that some immune-mediated cells seen in the E. sac during inner ear inflammatory events have migrated from the adjacent bone marrow.

Inner ear inflammation and round window otosclerosis

Recently, it has been suggested that otosclerosis represents the host's ongoing immunologic response to measles or other viral antigens. Documentation of past inflammation within the inner ear would serve as further evidence that this mechanism may be at play in the pathogenesis of the disease. Among the characteristic signs of prior inflammation in the inner ear is the presence of lamellar bone at the site of inflammation. This has been described in the temporal bone of a patient with immune-mediated deafness and with the temporal bones of experimental models of immune-mediated inner ear disease. Review of temporal bones with round window otosclerosis from the Eastern Temporal Bone Bank at the Massachusetts Eye and Ear Infirmary show that in four of ten cases there are characteristic signs of a prior severe inflammatory event centered in the scala tympani adjacent to the otosclerotic lesion. Otosclerosis, therefore, may have an inflammatory stage that is the consequence of a host response to an inciting event.

Fibronectin-like immunoreactivity of the basilar membrane of young and aged rats

Dysfunction of cochlear mechanics has been hypothesized to be a source of age-related hearing loss and the basilar membrane mass and stiffness contribute to normal cochlear mechanics. Fibronectin, a large, extracellular matrix protein and a major component of the basilar membrane, may contribute to both the mass and stiffness of the membrane. Mesothelial cells underlying the basilar membrane may produce the fibronectin and also contribute to the mass of the membrane. Changes in either the fibronectin or the mesothelial cells might, therefore, have an effect on cochlear mechanics. In order to assess basilar membrane changes in aged animals, young adult (2-4 months) and aged (24-26 months) Sprague-Dawley rats were evaluated for the presence of fibronectin-like protein and mesothelial cells. The basilar membrane in the young animals had strong fibronectin-like immunoreactivity throughout its length. The old animals, on the other hand, showed normal fibronectin immunoreactivity in the basilar membrane of the basal turn, but little or no reactivity in the apical cochlear turn. The number of mesothelial cells was reduced throughout the length of the membrane in aged animals, with the greatest loss in the basal turn (60% fewer cells). These two degenerative changes, which appear to be independent of each other, may contribute to the observed threshold shifts in aged cochleas.

Selective retrograde transport of nipecotic acid, a GABA analog, labels a subpopulation of gerbil olivocochlear neurons

Perfusion of the gerbil cochlea with micromolar quantities of 3H-gamma-aminobutyric acid (GABA) results in rapid, selective labeling of 50-60% of the olivocochlear (OC) efferent terminals on afferent dendrites beneath the inner hair cells, and all of the efferent terminals beneath the outer hair cells. In order to identify the neurons from which these GABA-accumulating terminals originate, the cell bodies were localized by using retrograde transport of 3H-nipecotic acid, a metabolically inert GABA analog. With survival times of 6-30 hours after cochlear injection, myelinated OC efferent fibers and cell bodies were well labeled, with the greatest number being labeled at 12-18 hours. All of the labeled neurons belonged to the medial OC system, and no lateral OC neurons were labeled. It is concluded that the GABA-accumulating endings in the gerbil cochlea arise from medial OC neurons, and therefore that medial OC efferent neurons in this species project to both inner and outer hair cell regions.

Spiral modiolar vein: its importance in viral load of the inner ear

Guinea pig-specific cytomegalovirus and Sendai virus were inoculated into the cochleas of seronegative guinea pigs to study the route of entry of cells participating in inner ear inflammation. Inflammatory cells accumulated around the spiral modiolar vein and appeared to be streaming from this vein into the scala tympani via a collecting venule. Inactivated virus inoculated into the cochlea and normal control cochlea failed to show inflammatory cell infiltrates. The spiral modiolar vein appears to play an important role in the movement of cells from the systemic circulation into the inner ear as part of the host's normal defense against invading pathogens such as viruses.

Effects of bacterial endotoxin applied to the guinea pig cochlea

The ototoxic potential of bacterial endotoxin present in middle ear effusions was evaluated by inoculating endotoxin solutions of 10 micrograms/mL or 1000 micrograms/mL directly into the scala tympani of the guinea pig cochlea. Median auditory brainstem response threshold shifts of 12.5 and 20 dB were observed in the experimental ears exposed to the higher dose at days 2 and 4, respectively, with gradual improvement to near baseline by day 28. Histologic examination revealed inflammatory infiltrates in the scala tympani and scala vestibuli of several of these cochleas. The lower concentration, closer to that found in human effusions, elicited no significant loss of hearing and a milder inflammatory response. It is concluded that brief exposure of the cochlea to endotoxin from middle ear effusions is insufficient to cause hearing loss. Under chronic conditions, however, persistent inflammation within the cochlea may result in a sensorineural deficit.

Cochlear degeneration in aged rats of four strains

Animals with various degrees of inbreeding, some of which are albino, are frequently used in biological research. Albinos do not produce melanin and it is therefore absent from the cochlea. While the function of melanin is unknown, it has been hypothesized that it is involved in cochlear homeostasis. It is possible then, that age-related degeneration may be affected by the presence or absence of melanin. We therefore evaluated young (2-6 months old) and aged (24-36 months old) cochleas in 4 different rat strains: albino Fischer 344 and Lewis rats and pigmented Lewis-Brown Norway F1 rats and Brown Norway rats. Cochlear morphology was the same across all strains of young adult animals with the exception that the pigmented animals had small, darkly stained granules in the stria vascularis. The aged pigmented animals all had large granules as well as small ones. Degeneration of spiral ganglion cells in the apical region of the ganglion had occurred in the old animals of all strains. Strial degeneration at the apex was also present in aged animals. There was no correlation between the presence or absence of melanin and the magnitude of cochlear degenerative changes in the aged animals. The presence or absence of melanin therefore, appears to have no effect on cochlear degeneration in the aged rat cochlea.

Type B Haemophilus influenzae-induced otitis media in the mouse

To characterize the middle and inner ear cellular inflammatory responses to otitis media using immunohistochemical methods, we inoculated type B Haemophilus influenzae into the middle ears of healthy adult BALB/c mice. Mac-1+ neutrophils and macrophages appeared in the middle ear at 3 days. Lyt-1+ T cells and Lyt-2+ T suppressor/cytotoxic cells entered the middle ear mucosa on days 7 and 14. IgG+ and IgM+ T cells were present at all time points, with IgA+ lymphocytes forming the majority of mucosal immunoglobulin-bearing cells at 2 weeks. The cochlear scala tympani contained Lyt-1+ and Mac-1+ cells and two endolymphatic sacs stained diffusely with anti-IgA and -IgG antibodies. Lyt-1/L3T4+ T lymphocytes greatly outnumbered B lymphocytes, suggesting that helper/inducer T cells play a more important role in acute otitis media than has been recognized. Inner ear changes occurred after a single episode of otitis media.

Spiral modiolar vein: its importance in inner ear inflammation

The inner ear responds immunologically to foreign proteins and pathogens introduced into the cochlea. The route of entry of the cells participating in the inner ear inflammatory process is the spiral modiolar vein with its collecting venules. Since this vein is located adjacent to the scala tympani, the end result of cellular infiltration is fibro-osseous obliteration of this scala preferentially. This observation has implications for the placement of cochlear implant electrodes of patients with labyrinthitis as the cause of hearing loss.

Effects of cyclophosphamide on the pathogenesis of cytomegalovirus-induced labyrinthitis

Cyclophosphamide was used in this study to define the contribution of the inflammatory response relative to direct cytopathic effects of guinea pig cytomegalovirus (GPCMV) in inducing sensorineural hearing loss in the guinea pig. The eighth nerve compound action potential (CAP) threshold on day 7 after inoculation of GPCMV into the scala tympani was an average of 35 dB greater for control animals than for those that were immunosuppressed with daily intraperitoneal injections of cyclophosphamide (20 mg/kg body weight). The amount of GPCMV antigen in the cochlea, detected immunohistochemically did not correlate with the CAP threshold. However, the greater the inflammatory response to GPCMV in the cochlea, the higher the CAP threshold and thus the greater the hearing loss. This study demonstrates that the inflammatory response to GPCMV may be more important than direct cytopathic effects of the virus in producing sensorineural hearing loss in GPCMV-induced labyrinthitis.

Spiral ganglion cell endings in the cochlear nucleus of young and old rats

The spiral ganglion cells (SGCs) forming the auditory nerve have been shown to degenerate with age in both human and animal models, presumably resulting in post-synaptic sites in the cochlear nucleus (CN) that have lost their inputs. The present study examined the morphological changes in the surviving SGC endings in the CN of aged animals. The auditory nerves of 2-3 MO and 25-26 MO male, Fisher 344 rats were anterogradely labeled with horseradish peroxidase and the CN prepared histochemically. This resulted in Golgi-like labeling of afferent fibers and their terminals. All endings within each section were drawn and the area and number of components per ending were measured. Young and old animals both had about the same proportion of "simple", "string" and "complex" endings within the ventral CN, with complex endings being predominant in both age groups. The area of many complex endings was greater in the old animals with some endings being twice as large as any seen in young animals. There was no evidence of smaller endings in the old animals, suggesting that endings are not shrinking with age. A comparison of the number of components per complex ending revealed significantly more complexity in the endings of aged animals. Following the degeneration of SGCs it seems likely that the remaining cells, by increasing the area and altering the shape of their central terminals, may cover some of the post-synaptic sites made available by degenerated endings in aged animals.

Spiral modiolar vein: its importance in inner ear inflammation

The inner ear responds immunologically to foreign proteins and pathogens introduced into the cochlea. The route of entry of the cells participating in the inner ear inflammatory process is the spiral modiolar vein with its collecting venules. Since this vein is located adjacent to the scala tympani, the end result of cellular infiltration is fibro-osseous obliteration of this scala preferentially. This observation has implications for the placement of cochlear implant electrodes of patients with labyrinthitis as the cause of hearing loss.

Collaterals from lateral and medial olivocochlear efferent neurons innervate different regions of the cochlear nucleus and adjacent brainstem

Two populations of superior olivary neurons which project to different sensory cell regions in the cochlea also give off collateral projections to the ventral cochlear nucleus (VCN) and adjacent brainstem. To determine whether these VCN projections also have different targets they were characterized by selective retrograde amino acid transport. Retrograde transport of 3H-d-aspartate (D-ASP) selectively labeled the unmyelinated fibers and neurons of the lateral olivocochlear (OC) system including a dense collateral projection to the central VCN. Retrograde transport of 3H-nipecotic acid (NIP) labeled the myelinated fibers and neurons of the medial OC system, including collateral projections to the peripheral VCN, subpeduncular granule cells, and nucleus Y. Medial and lateral OC efferent collaterals thus innervate different regions of the CN. Lateral system collaterals overlap extensively with Type I spiral ganglion cell afferent input. They are well positioned to play a role in modulating afferent input to the central auditory system, as is the primary projection of these efferents to the cochlea. The medial system collaterals project near the recently described afferent projections of Type II spiral ganglion cells. The medial system collaterals may therefore be related to the function of outer hair cells, as the medial system primary axons appear to be in the cochlea.

Immunologic responses in experimental cytomegalovirus labyrinthitis

To better understand the pathogenesis of cytomegalovirus labyrinthitis, a guinea pig model was created. Following inoculation at several sites (cardiac, perilymph, and endolymphatic sac) in both seronegative and seropositive animals, the immunologic, histologic, and electrophysiologic responses were measured. Seronegative animals uniformly showed progressive hearing loss with marked inflammation and degeneration of neural elements. In animals inoculated into the endolymphatic sac, an associated endolymphatic hydrops developed in addition to deafness. Seropositivity protected the hearing, but endolymphatic sac inoculations resulted in mild hydrops due to local inflammation that was devoid of evidence of viral replication. The question of whether hearing loss was attributable to local inflammatory responses rather than the cytopathic effects of the virus was then examined. To test this hypothesis, animals were immunosuppressed with cyclophosphamide prior to intracochlear inoculation of cytomegalovirus. The immunosuppressed animals showed significantly better hearing than the controls, and this correlated directly with the degree of cellular infiltration of the scala tympani. These studies confirm the importance of host immune responses in the pathogenesis of hearing loss due to cytomegalovirus.

Comparison of immune-mediated models of acute and chronic otitis media

The distribution of immunoglobulin-bearing cells and the pattern of histopathological changes in the middle ear (ME) mucosa, round window membrane (RWM), and inner ear were compared during acute and chronic immune-mediated otitis media with effusion (OME) in the guinea pig as an animal model. In both acute and chronic immune responses (IRs), mucosal hyperplasia, edema, neovascularization, and cellular infiltration were observed. IgG+ cells were predominant in both the acute and chronic IRs. The number of IgA+ cells, however, increased in the mucosa and RWM during chronic IRs. Only the chronic IR resulted in gland formation within the ME and inflammation within the cochlea. These results indicate that the chronic IR was more similar to reports of clinical OME than the acute IR. The cochlear inflammation associated with chronic OME can lead to sensorineural hearing loss, as reported in clinical studies.

Endotoxin-induced otitis media with effusion in the mouse. Immunohistochemical analysis

A model of endotoxin-induced otitis media with effusion was developed in healthy BALB/c mice. Endotoxin extracted from Salmonella typhimurium (10 micrograms/ml) was injected into the middle ear bulla transtympanically and the resultant inflammatory response was analyzed using immunohistochemical methods. Serous effusions were observed otoscopically and mucosal edema and infiltrate were studied histologically, both peaking at 3 days and essentially resolving by 7 days. However, persistence of a small area of inflammation in the round window niche was consistently present at 2 weeks. The specific cellular responses to endotoxin were analysed by histologic and immunohistologic examination of the murine temporal bone. We used antibodies to identify T-lymphocytes (anti-Lyt-1, -Lyt-2), macrophages and neutrophils (anti-Mac-1), and immunoglobulins (anti-IgA, -IgG, -IgM). In the mucosal/submucosal infiltrate Mac-1+ and Lyt-1+ cells peaked from days 1 to 3. Within the luminal effusion Mac-1+ and Lyt-1+ cells as well as diffuse (not cell-associated) IgG, IgM, and IgA staining were most prevalent from days 1 to 3, while IgG-bearing plasma cells formed the majority of the luminal cells observed at 1 to 2 weeks. No significant influx of T-suppressor cells was observed at any time. The round window niche and anterior portion of the tympanic bulla mucosa were found to be the most reactive areas. These results suggest that endotoxin alone is capable of producing an inflammatory infiltrate in the mouse consistent with otitis media with effusion, and that the interaction of endotoxin with resident immunocytes of the middle ear, as well as serum derived immune cells, is consistent with the known phlogistic properties of endotoxin.

Antigen-specific immune response in the inner ear

The specificity of inner ear immune responses was investigated by challenging each inner ear of presensitized animals with different antigens. Animals presensitized systemically with keyhole limpet hemocyanin (KLH) and bovine serum albumin (BSA) were challenged with KLH in the right and BSA in the left inner ears. Two weeks later perilymph anti-KLH levels were increased significantly in the right inner ears compared to the levels in the left inner ears. In contrast, perilymph anti-BSA levels were increased significantly in the left inner ears compared to the levels in the right inner ears. These results suggested that the rise in perilymph antibody following inner ear antigen challenge was predominantly the result of an antigen-specific immune response in the inner ear and not simply the result of an increase in vascular permeability of serum contamination from the experimental procedure itself.

Development of morphological and physiological changes in the cochlea induced by cytomegalovirus

The effect of viral infection in the cochlea was investigated by inoculation of live cytomegalovirus or inactivated virus. Auditory thresholds were measured on the day of inoculation and on the terminal day. Two to 8 days following inoculation, the animals were killed and the cochleas were evaluated histologically. The compound nerve potential showed an increase in threshold prior to the cochlear microphonic, indicating the nerve was affected prior to the outer hair cells. All experimental cochleas contained inflammatory and cytomegalic inclusion cells and showed degenerative changes. The number of infected cells was small relative to the histopathology. Control cochleas had normal structure and function. The degeneration, therefore, might be mediated by inflammation as well as by the cytopathic effect of the virus. Viral infections, therefore, might be better managed with anti-inflammatory therapy in addition to antiviral agents.

Spiral ganglion cell density in young and old gerbils

The Mongolian gerbil, like other mammalian species, has a decreased number of spiral ganglion cells as a function of age. This loss of cells was first seen in 24- to 30-month old animals in the basal end of the ganglion. In the oldest individuals the apical end of the ganglion was also affected. There were approximately 15-25% fewer cells in the affected areas in the 36- to 42-month old animals. In the oldest animals degeneration of the stria vascularis was seen in the apical turn and some degenerative changes in the organ of Corti were seen throughout the length of the cochlear duct. The aging pattern in the gerbil cochlea, is similar to that described for other species. Vacuoles, previously described in the gerbil cochlear nucleus, were also seen in the auditory nerve within the modiolus, but central to the Schwann-glial border in all animals. Vacuoles were not present within the spiral ganglion or the peripheral processes of the ganglion cells. Because the ganglion cell axons should be similar on either side of the Schwann-glial border, but the vacuoles were confined to the central nervous system, it is concluded that the degenerative process affects glial cells as opposed to neurons.

The spatial representation of frequency in the rat dorsal cochlear nucleus and inferior colliculus

The spatial distribution of neural activity produced by tones was assessed in the rat dorsal cochlear nucleus (DCN) and inferior colliculus (IC), using the 2-deoxyglucose (2-DG) technique. Eight pure tones, spanning the range of reported single unit characteristic frequencies in the rat, were presented at 40 dB above behavioral threshold. The relationship between frequency of stimulation and location of neural activity within each nucleus was evaluated quantitatively. Based on the 2-DG uptake pattern across animals, a tonotopic axis in the transverse plane was defined for each nucleus. This axis transected the centers of regions of evoked 2-DG uptake for each frequency. There was an orderly relationship between stimulus frequency and the location of evoked neural activity along the axis. Each pure tone stimulus activated an approximately equal proportion of this axis, for all frequencies tested, in both the DCN and IC. This suggests the existence of equal 'spatial bandwidths, in rat central auditory structures, across its entire frequency range. Equal spatial bandwidths could facilitate signal analysis strategies which require interaction between neurons with closely-related CFs. In the horizontal plane, however, the proportion of stimulated tissue was not equal across frequency. High-frequency (greater than 8 kHz) tones produced increased neural activity along a much greater extent of the anterior-to-posterior axis of the IC than did low-frequency tones.(ABSTRACT TRUNCATED AT 250 WORDS)

Temporal sequence of viral antigen expression in the cochlea induced by cytomegalovirus

Injection of guinea-pig cytomegalovirus into seronegative guinea pig cochleas results in viral labyrinthitis. The location of infected cells 1-8 days following inoculation was examined using an immunohistochemical assay for viral antigens. Mesothelial cells below the basilar membrane and those lining the scala tympani, especially below the osseous spiral lamina, were the first cells to express antigen. This occurred 2 days following inoculation. Infection then spread rapidly toward the brain and through the perilymphatic duct to involve inflammatory cells, mesothelial cells of Reissner's membrane, spiral ganglion cells, and the vestibular ganglion cells. In any one cochlea, more mesothelial cells and inflammatory cells expressed viral antigen than any other cell type. It seems then, that there is a differential susceptibility to viral infection and antigen expression in the various cell types of the cochlea. Typical cytomegalic inclusions were common in mesothelial cells. However, many cells labelled for viral antigens did not contain inclusions. The identification of viral antigens then, is a more sensitive technique for demonstrating infection than is the location of cytomegalic inclusion cells.

Experimental cytomegalovirus infection: viremic spread to the inner ear

The entry route of viruses into the inner ear is still controversial. Guinea pig cytomegalovirus (CMV) has been used to create an animal model of systemic infection as well as labyrinthitis. In this study, seronegative guinea pigs were administered intracardiac inoculations of varying doses of guinea pig CMV. After three or eight days, the animals were killed and necropsy specimens examined for viral antigen. In the specimens examined three days after inoculation, no label was seen in the cochlea. However, label was found in the spleen with doses of 100 microL, and also in the liver in animals inoculated with 300 microL of virus. After eight days, an animal that received 200 microL of the virus showed labeling in the modiolar blood vessels and perivascular infiltrates. Two of the four animals that received 300 microL of the virus showed labeling in spiral ganglion cells. None of the animals showed viral antigen within the stria vascularis, nor were there signs of acute labyrinthitis. At eight days, animals inoculated with greater than 100 microL of virus showed labeled cells in multiple organs. These data suggest that the entry route of guinea pig cytomegalovirus into the inner ear might involve viremic spread to modiolar blood vessels and subsequent spread to spiral ganglion cells.

The development of endolymphatic hydrops following CMV inoculation of the endolymphatic sac

The effect of inoculation of cytomegalovirus (CMV) into the endolymphatic sac was examined in CMV-seronegative and seropositive animals. Seronegative animals developed hearing loss, infection in the epithelial cells of the endolymphatic sac and perisaccular connective tissue, and endolymphatic hydrops. Control animals inoculated with inactivated CMV showed no hearing loss, viral infection, or endolymphatic hydrops. Seropositive animals showed complete protection from hearing loss and viral infection, yet a monocytic infiltrate was seen surrounding the endolymphatic sac. Associated with this localized immune response was mild evidence of endolymphatic hydrops in 40% (2/5) of the animals. Control seropositive animals inoculated with inactivated CMV showed no hearing loss or morphological changes. CMV then, can infect cells of the endolymphatic sac resulting in hearing loss and endolymphatic hydrops. The immune response to CMV in seropositive animals is protective, but is associated with endolymphatic sac inflammation.

The spiral ganglion and hair cells of Bronx waltzer mice

Inner hair cells and spiral ganglion cells were counted in a mutant mouse strain (Bronx waltzer) reported to have very few of these cells (6) in order to determine if the remaining ganglion cells would be predominantly type II cells. These cell counts indicate a 50% reduction of spiral ganglion cells in Bronx waltzer cochleas compared to normal mice. Averaged throughout the cochlea about 11% of the remaining cells are type II cells while in normal mice this percentage is 5%. In some regions however, as many as 20% of the remaining cells are type II cells. Counts of IHC in surface preparations reveal 37 normal looking IHC (about 5% of the normal population) in each of two Bronx waltzer mouse cochleas. There were also about 120 shrunken IHC in each cochlea, representing 17% of the normal cell population. While there appears to be an increased proportion of type II cells in the Bronx waltzer spiral ganglion there are also many more type I cells than might be expected from the small number of IHC.