Facial nerve palsy after reactivation of herpes simplex virus type 1 in diabetic mice.

OBJECTIVES/HYPOTHESIS: Bell's palsy is highly associated with diabetes mellitus (DM). Either the reactivation of herpes simplex virus type 1 (HSV-1) or diabetic mononeuropathy has been proposed to cause the facial paralysis observed in DM patients. However, distinguishing whether the facial palsy is caused by herpetic neuritis or diabetic mononeuropathy is difficult. We previously reported that facial paralysis was aggravated in DM mice after HSV-1 inoculation of the murine auricle. In the current study, we induced HSV-1 reactivation by an auricular scratch following DM induction with streptozotocin (STZ). STUDY DESIGN: Controlled animal study. METHODS: Diabetes mellitus was induced with streptozotocin injection in only mice that developed transient facial nerve paralysis with HSV-1. Recurrent facial palsy was induced after HSV-1 reactivation by auricular scratch. RESULTS: After DM induction, the number of cluster of differentiation 3 (CD3)(+) T cells decreased by 70% in the DM mice, and facial nerve palsy recurred in 13% of the DM mice. Herpes simplex virus type 1 deoxyribonucleic acid (DNA) was detected in the facial nerve of all of the DM mice with palsy, and HSV-1 capsids were found in the geniculate ganglion using electron microscopy. Herpes simplex virus type 1 DNA was also found in some of the DM mice without palsy, which suggested the subclinical reactivation of HSV-1. CONCLUSIONS: These results suggested that HSV-1 reactivation in the geniculate ganglion may be the main causative factor of the increased incidence of facial paralysis in DM patients.

A cell culture model of facial palsy resulting from reactivation of latent herpes simplex type 1.

HYPOTHESIS: Reactivation of herpes simplex virus type 1 (HSV-1) in geniculate ganglion neurons (GGNs) is an etiologic mechanism of Bell's palsy (BP) and delayed facial palsy (DFP) after otologic surgery. BACKGROUND: Several clinical studies, including temporal bone studies, antibody, titers, and intraoperative studies, suggest that reactivation of HSV-1 from latently infected GGNs may lead to both BP and DFP. However, it is difficult to study these processes in humans or live animals. METHODS: Primary cultures of GGNs were latently infected with Patton strain HSV-1 expressing a green fluorescent protein-late lytic gene chimera. Four days later, these cultures were treated with trichostatin A (TSA), a known chemical reactivator of HSV-1 in other neurons. Cultures were monitored daily by fluorescent microscopy. Titers of media from lytic, latent, and latent/TSA treated GGN cultures were obtained using plaque assays on Vero cells. RNA was harvested from latently infected GGN cultures and examined for the presence of viral transcripts using reverse transcription-polymerase chain reaction. RESULTS: Latently infected GGN cultures displayed latency-associated transcripts only, whereas lytically infected and reactivated latent cultures produced other viral transcripts, as well. The GGN cultures displayed a reactivation rate of 65% after treatment with TSA. Media from latently infected cultures contained no detectable infectious HSV-1, whereas infectious virus was observed in both lytically and latently infected/TSA-treated culture media. CONCLUSION: We have shown that cultured GGNs can be latently infected with HSV-1, and HSV-1 in these latently infected neurons can be reactivated using TSA, yielding infectious virus. These results have implications for the cause of both BP and DFP.

Latency of herpes simplex virus type-1 in human geniculate and vestibular ganglia is associated with infiltration of CD8+ T cells.

Herpes simplex virus type-1 latency and CD8+ T-cell occurrence were investigated in the trigeminal, geniculate, and vestibular ganglia from seven deceased humans. The HSV-1 "latency-associated transcript" was assessed by in situ hybridization and quantitative RT-PCR. Infiltration of CD8+ T cell was detected by immunohistochemistry and quantitative RT-PCR. The data show that HSV-1 latency and CD8+ T-cell infiltration are not solely confined to the trigeminal ganglia but can also occur in other cranial ganglia along the neuroaxis. However, the HSV-1 latency transcripts in the geniculate and vestibular ganglia were expressed at a very low level. The difference in CD8 transcript levels among HSV-1 latently infected trigeminal ganglia, geniculate, and vestibular ganglia was less conspicuous. Colocalization of latent HSV-1 and CD8+ T cells in geniculate and vestibular ganglia supports further the hypothesis that HSV-1 reactivation is possible in these ganglia and is the cause of Bell's palsy and vestibular neuritis.

Herpes virus and Ménière's disease.

OBJECTIVE: The main goal of this study was to examine the vestibular ganglia from patients with intractable classic Ménière's disease (MD) for the presence or absence of DNA from three neurotropic viruses herpes simplex virus 1 and 2 (HSV1, HSV2) and varicella zoster virus (VZV) and to investigate the hypothesis that MD is associated with virus reactivation within Scarpa's ganglion. STUDY DESIGN: Polymerase chain reaction (PCR) was performed with nested primer sets specific for viral genomic DNA of HSV1, HSV2 and VZV in biopsies of the ganglion scarpae of patients with MD who underwent vestibular neurectomy. Included were patients with MD classified as definite MD according to American Academy of Otolaryngology/Head and Neck Surgery criteria. The ganglion scarpae and ganglion geniculi harvested at autopsy from patients without history of MD or facial palsy served as control specimens. RESULTS: No viral DNA was detected in the vestibular ganglion of 7 patients with definite MD. In 34% of the vestibular ganglia of the control group we detected either HSV1 or VZV. Only one Scarpa's ganglion had both viruses present at the same time. Thirty-two out of 34 ganglia from the geniculate segment of the facial nerve contained either HSV1 and/or VZV genomic DNA. Eight specimens contained both viruses simultaneously. Altogether viral DNA was found in 94% of ganglia. Viral genomic DNA of HSV2 was not detected. CONCLUSION: Although HSV and VZV appear to be present in many ganglion cells throughout the human body, we were unable to find genomic DNA of these viruses in patients with definite MD and disabling vertigo, who underwent vestibular neurectomy. Based on these results, reactivation of HSV1 and VZV in the vestibular ganglion does not seem to play a role in the pathogenesis of MD.

Do viruses cause inner ear disturbances?

The association of viral infection to inner ear disease is controversial. Experiments on animals show that several viruses are capable of causing hearing loss, if applied into the perilymph. Some of these have specific affinity to the cellular type of the inner ear, as sensory epithelia and cochlear nerve. Some viruses as adenoviruses and Coxsackie virus B have specific CAR receptors that are identified in different cell types, whereas other act by attaching onto nonspecific cellular surface receptors. Some viruses such as varicella zoster virus (VZV) do not cause disease in rodents. We assessed 273 patients with clinical, serological, neuro-otologic and endoscopic evaluations. Of the 273 patients, 43 served as control subjects. The patients either had Ménière's disease (n = 158), recurrent vertigo of unknown etiology (n = 56), or hearing loss (n = 17). Antibodies against neurotropic and common viruses were evaluated. VZV, influenza B, CBV5 and RSV titers were significantly elevated in patients with inner ear disease when compared with the control group. In analyzing the internal relationship, VZV and influenza B were intercorrelated. We did not find a correlation between hearing loss and viral titers. In conclusion, VZV, Coxsackie virus B5 and influenza B virus may be the main causes of inner ear disorder. The spiral and Scarpa's ganglion are potential sites harboring viral DNA for possible latent infection.

Role of T-lymphocyte subsets in facial nerve paralysis owing to the reactivation of herpes simplex virus type 1.

CONCLUSION: Although both T-cell subsets are essential for inhibiting HSV-1 reactivation in the GG, CD4 + T cells play a more important role in host defense against virus replication. OBJECTIVE: To elucidate the host immunological factors that participate in herpes simplex virus type 1 (HSV-1) reactivation in the geniculate ganglia (GG) and lead to facial paralysis, we developed a mouse model of facial paralysis that involved the reactivation of HSV-1 following general immune suppression. MATERIAL AND METHODS: Eight weeks after recovery from primary facial paralysis caused by inoculating the auricle with HSV-1 the auricle was scratched and mice (n = 69) were given an i.p. injection of either anti-CD4 (n = 46) or anti-CD8 (n = 23) monoclonal antibody to deplete specific T-lymphocyte subsets. Following this reactivation procedure, the rate of recurrent facial paralysis was compared between the two models. The GG were examined histopathologically and using polymerase chain reaction to detect HSV-1 DNA. RESULTS: Facial paralysis developed in 42% of mice in the anti-CD4 model and in 13% in the anti-CD8 model. HSV-1 DNA was detected in 50% of the mice in both models. Histopathologically, neurons were destroyed in parts of the GG and numerous virus particles were seen in the surviving neurons.

Bell's palsy with ipsilateral numbness.

Bell's palsy is an idiopathic facial palsy of the peripheral type. A herpes virus is the most likely mechanism. We report a patient with the often encountered combination of a facial palsy with ipsilateral sensory changes. Magnetic resonance imaging showed had contrast enhancement in the greater petrosal nerve. Viral spread through anatomical connections could be an explanation for the association of facial palsy with numbness.

Bell's palsy and Herpes simplex virus: fact or mystery?

HYPOTHESIS AND BACKGROUND: In recent years, progress has been made in the understanding of Bell's palsy, the most common form of acute facial weakness. Herpes simplex virus (HSV) reactivation within the geniculate ganglion with subsequent inflammation and entrapment of the nerve at the meatal foramen has been proposed to be the pathogenetic mechanism. We challenged its accuracy by analyzing our own data on the presence of viral genomic DNA of HSV-1 and 2, human herpes virus (HHV)-6A/B, as well as varizella zoster virus (VZV) in patients with Bell's palsy and in control patients without the disease. METHODS: Polymerase chain reaction was performed with primer sets specific for viral genomic DNA of HSV-1, HSV-2, and VZV in facial muscle biopsy specimens from patients with Bell's palsy. As control specimens, the Scarpa's ganglion of patients with Meniere's disease and the geniculate ganglion harvested at autopsy from patients without history of facial palsy. In a second study, we used polymerase chain reaction with primers specific for HSV-1, -2, and HHV-6A, -6B to analyze for the presence of these viruses in tear fluid samples from control patients and patients with acute Bell's palsy. RESULTS: HSV-1 and VZV genomic DNA were detected in 86 and 43%, respectively, of geniculate ganglion preparations from control specimen. We were not able to detect the presence of HSV-1, HSV-2, or VZV genomic DNA in ganglion scarpae or muscle biopsy results in control and Bell's palsy patients. HHV-6A could be detected in tear fluid samples in 40% of control patients and 30% of Bell's palsy patients. CONCLUSIONS: The sole presence of HSV genomic DNA within the sensory ganglion along the facial nerve does not explain the direct association with Bell's palsy. The missing link would be the identification of an active replicating virus, an investigation that has not yet been carried out.

Quantitative analysis of herpes simplex virus in cranial nerve ganglia.

A susceptible individual exposed to herpes simplex virus (HSV) will develop latent infection in multiple cranial nerve ganglia. There are a few quantitative studies of the viral load within the trigeminal ganglion, but none that investigate other cranial nerve ganglia. In this study, human trigeminal, geniculate, vestibular (Scarpa's) and cochlear (spiral) ganglia were obtained from willed body donors. Real time quantitative polymerase chain reaction (PCR) analysis of the HSV DNA polymerase gene was performed on ipsilateral ganglion sets from the same individual. Viral load, expressed as HSV genomes per 105 cells, was significantly greater in the vestibular ganglion (mean +/- SD, 176705 +/- 255916) than in the geniculate (9948 +/- 22066), cochlear (3527 +/- 9360), or trigeminal (2017 +/- 5578) ganglia. There was not a significant correlation among ganglia from the same individual. The results support the hypothesis that neuronal subpopulations have variable susceptibility to HSV infection.

Demyelination associated with HSV-1-induced facial paralysis.

In 1995, we developed an animal model of transient homolateral facial nerve paralysis by inoculating Herpes simplex virus type 1 (HSV-1) into the auricle of mice. This study examined the mechanism of facial nerve paralysis in this model histopathologically. Using the immunofluorescence technique with anti-HSV-1 antibody, the time course of viral spread and the site of viral replication were investigated over the entire course of the facial nerve. Furthermore, viral replication and nerve degeneration at the site of viral replication were observed by electron microscopy. On the 7th day after inoculation, facial paralysis was observed in more than 60% of mice. Immunofluorescence study revealed HSV-1 in the geniculate ganglion, the descending root, and the facial nucleus at this stage. On the 9th day, the descending root in the sections stained with osmium looked pale, because prominent demyelination had occurred in this region; electron micrographs showed many degenerated oligodendrocytes and large naked axons. In contrast, the facial nucleus neurons showed no remarkable degeneration, despite HSV-1 particles in their cytoplasm. From these findings, we concluded that facial nerve paralysis in this model is caused mainly by facial nerve demyelination in the descending root.

Chickenpox and the geniculate ganglion: facial nerve palsy, Ramsay Hunt syndrome and acyclovir treatment.

Facial nerve palsy has long been considered to have an infectious etiology. Recent diagnostic analyses in children and adults have provided convincing evidence that reactivation of varicella-zoster virus (VZV), sometimes during infectious mononucleosis, can lead to cranial nerve VII palsy. The site of reactivation from latency is the geniculate ganglion. Virus most likely enters the ganglion during chickenpox, via the sensory branches of the facial nerve located on the ear and tongue. Retrospective reviews suggest that patients with VZV-related facial nerve palsy have poorer outcomes than other cases of Bell's palsy. Therefore treatment with acyclovir is suggested when VZV reactivation i slikely.

Prevalence of HSV-1 LAT in human trigeminal, geniculate, and vestibular ganglia and its implication for cranial nerve syndromes.

Herpes simplex virus type 1 (HSV-1) enters sensory neurons and can remain latent there until reactivation. During latency restricted HSV-1 gene expression takes place in the form of latency-associated transcripts (LAT). LAT has been demonstrated to be important not only for latency but also for reactivation, which may cause cranial nerve disorders. Tissue sections of the trigeminal ganglia (TG), geniculate ganglia (GG), and the vestibular ganglia (VG) from seven subjects were examined for the presence of LAT using the in situ hybridization technique. LAT was found on both sides in allTG (100%), on both sides of five subjects (70%) in the GG, and in none of the VG. Using a second more sensitive detection method (RT-PCR), we found LAT in the VG of seven of ten other persons (70%). This is the first study to demonstrate viral latency in the VG, a finding that supports the hypothesis that vestibular neuritis is caused by HSV-1 reactivation. The distribution of LAT in the cranial nerve ganglia indicates that primary infection occurs in the TG and GG and subsequently spreads along the faciovestibular anastomosis to the VG.

Delayed facial palsy after stapedectomy.

OBJECTIVE: To study the incidence, pathogenesis, and prevention of delayed facial palsy after stapedectomy. STUDY DESIGN: Retrospective case review. SETTING: Otology/neurotology referral center. PATIENTS: A series of 2152 stapedectomy procedures in 2106 patients over 12 years. INTERVENTION: Delayed facial palsy after stapedectomy was studied. MAIN OUTCOME MEASURE: House-Brackmann facial nerve grading system and serum antibody titer tests for herpes simplex virus type I and type II, and varicella zoster virus. RESULTS: Delayed facial palsy occurred in 11 of 2152 procedures. Delayed facial palsy occurred from 5 to 16 days (mean 8) after stapedectomy. Predisposing factors were bony facial canal dehiscence with bare or bulging facial nerve herniation in 5 patients; chorda tympani nerve stretched, manipulated, or cut in 2 patients; granulomatous reaction to Gelfoam in 1 patient; fever blisters on the upper lip in 1 patient; and sinusitis in 2 patients. Elevated anti-varicella antibody titers were found in all 6 patients studied. Anti-herpes simplex type I and II antibody titers were elevated in 5 of 6 patients. Acyclovir was effective in preventing delayed facial palsy in 1 patient who had undergone revision stapedectomy and experienced delayed facial palsy after previous stapedectomy in the same ear with elevated anti-herpes antibody titer. CONCLUSIONS: Delayed facial palsy occurred in 0.51% of patients after stapedectomy. Serologic investigation suggests activation of latent herpesvirus. Mechanical irritation of the facial or chorda nerve during operation may trigger the activation. The anti-herpesvirus agent acyclovir may prevent delayed facial palsy after stapedectomy in patients suspected of having this complication.

Herpes virus reactivation and gadolinium-enhanced magnetic resonance imaging in patients with facial palsy.

OBJECTIVE: This study investigated whether magnetic resonance imaging (MRI) patterns were different between patients with Bell's palsy and those with herpetic facial palsy in whom varicella-zoster virus (VZV) or herpes simplex virus type 1 (HSV-1) reactivation had been confirmed by polymerase chain reaction (PCR) or serologic assay. STUDY DESIGN: A retrospective study of 15 patients with acute peripheral facial palsy was performed to compare virologic tests and gadolinium (Gd)-enhanced MRI findings. RESULTS: Ramsay Hunt syndrome was diagnosed in one patient. By use of virologic tests, zoster sine herpete (VZV reactivation without zoster) was diagnosed in four patients and HSV-1 reactivation in three. Bell's palsy was diagnosed in the remaining seven patients. No significant difference in the frequency of Gd-enhanced MRI was observed between herpetic facial palsy and Bell's palsy. However, in those patients who underwent MRI on the day viral reactivation was confirmed by PCR, Gd enhancement of the meatal fundus was observed infrequently. In addition, when MRI was performed within 10 days of the onset of palsy, Gd enhancement was not detected at the geniculate ganglion in any patients with herpetic facial palsy. By contrast, both the meatal fundus and the geniculate ganglion were enhanced in all patients with Bell's palsy, regardless of when MRI was performed with respect to the onset of palsy. CONCLUSION: This study shows a difference in the pattern of Gd enhancement at the meatal fundus and the geniculate ganglion between patients with Bell's palsy and those with herpetic facial palsy. The results suggest that the meatal fundus or the geniculate ganglion may be affected first by virus reactivation in patients with herpetic facial palsy.

Mouse model of Bell's palsy induced by reactivation of herpes simplex virus type 1.

In order to investigate the mechanism of Bell's palsy, we developed an animal model of facial nerve paralysis induced by the reactivation of herpes simplex virus type 1 (HSV-1). Eight weeks after recovery from facial nerve paralysis caused by inoculation with HSV-1, the mice were treated with auricular skin scratch at the site of the previous inoculation, or with intraperitoneal injection of anti-CD3 monoclonal antibody (mAb), or combination of both procedures. No mice developed facial nerve paralysis when they were treated with either auricular scratch or mAb injection alone. In contrast, 20% of mice developed facial nerve paralysis with the combined treatment. With one exception, no mouse treated with either auricular scratch or mAb injection showed HSV-I DNA in their facial nerve tissue, whereas 4 out of 6 mice receiving both treatments showed HSV-1 DNA on day 10 after treatment. Histopathological findings showed neuronal degeneration in the geniculate ganglion and demyelination of the facial motor nerve in paralyzed mice. These findings suggest that a combination of stimuli, local skin irritation, and general immunosuppression is essential for successfully inducing facial nerve paralysis in mice with latent HSV-1 infection.

Herpes simplex virus in the vestibular ganglion and the geniculate ganglion-role of loose myelin.

This study presents the first direct evidence for herpes simplex virus type 1 (HSV-1) infection in the neurons of the vestibular ganglion. Although many investigators have reported electron microscopic evidence of HSV-1 infection in sensory ganglia, HSV-1 infection in the vestibular ganglion has not been described. Vestibular ganglion neurons have a unique structure, with a loose myelin sheath instead of the satellite cell sheath that is seen in other ganglia. This loose myelin is slightly different from compact myelin which is known as too tight for HSV-1 to penetrate. The role of loose myelin in terms of HSV-1 infection is completely unknown. Therefore, in an attempt to evaluate the role of loose myelin in HSV-1 infection, we looked for HSV-1 particles, or any effects mediated by HSV-1, in the vestibular ganglion as compared with the geniculate ganglion. At the light microscopic level, some neurons with vacuolar changes were observed, mainly in the distal portion of the vestibular ganglion where the communicating branch from the geniculate ganglion enters. At the electron microscopic level, vacuoles, dilated rough endoplasmic reticulum and Golgi vesicles occupied by virus were observed in both ganglia neurons. In contrast, viral infections in Schwann and satellite cells were observed only in the geniculate ganglion, but not in the vestibular ganglion. These results suggest that loose myelin is an important barrier to HSV-1 infection, and it must play an important role in the prevention of viral spread from infected neurons to other cells.

Amplification of herpes simplex virus type 1 DNA in human geniculate ganglia from formalin-fixed, nonembedded temporal bones.

Polymerase chain reaction (PCR) has provided new insights in molecular biology. Recently, some studies have been focused on temporal bone pathology, with amplification of DNA from fixed sections of celloidin-embedded bones. The purpose of our study was to elucidate the utility of PCR in detection of minor concentrations of DNA from nonoptimal stored samples. We obtained geniculate ganglia from 30 temporal bones preserved in formalin for a long time, without any process of embedding. By performing a nested PCR assay, we detected herpes simplex virus type 1 DNA in 13 of 30 ganglia (43%). We conclude therefore that study of temporal bones stored under poor conditions by PCR is possible, although there are some limitations when compared with fresh or optimally archived samples.

Detection of herpes simplex virus-1 by nested PCR. An experimental model.

Nested polymerase chain reaction (nested PCR) was performed using a reaction mix batch-prepared and kept frozen in single reaction tubes at -20 degrees C until use. Twenty-one New Zealand white rabbits were infected with herpes simplex virus type 1 (HSV-1). Eleven animals were killed on day seven and the other ten were sacrificed on day 21. Viral culture and nested PCR was used to determine the presence of HSV-1 in samples from the tongue, HSV-1 was detected in 90.47% of the animals; in 84.21% by nested PCR and in 52.63% by culture. Nested PCR assay had greater sensitivity than culture in animals sacrificed on day seven with significative difference (p < 0.05). Higher sensitivity and faster results were obtained with this method, so we found it reliable and useful in the setting of a clinical laboratory dealing with diagnosis of herpes virus infections.

Bell's palsy and herpes simplex virus.

Bell's palsy, which is defined as idiopathic peripheral facial paralysis of sudden onset, accounts for > 50% of all cases of facial paralysis. Different theories on the etiology of Bell's palsy have been proposed and investigated. Various clinical studies have suggested an etiological link between Bell's palsy and herpes simplex virus (HSV). In addition, animal experiments have shown the ability of HSV to induce facial paralysis. In our opinion, the possible link between Bell's palsy and HSV can only be explored properly by studying the human facial nerve, and especially the geniculate ganglion itself. Different groups have tried to detect hypothetically reactivated and hypothetically latent HSV in the facial nerves of Bell's palsy patients and control patients, respectively. The isolation of infectious HSV from facial nerve tissue by conventional cell culture methods appeared to be very difficult, also when Bell's palsy patients were tested. Instead, modern molecular methods, such as in situ hybridization and the polymerase chain reaction (PCR) could easily detect HSV DNA in geniculate ganglia. The detection of HSV-specific latency-associated transcripts in the ganglia of control patients provided further evidence for the hypothetically latent state of HSV in the geniculate ganglia in these patients. Recent PCR experiments performed by a Japanese group strongly suggest that the area adjacent to the geniculate ganglia does not usually contain any HSV at all, except in patients with Bell's palsy. This well-controlled study provides conclusive evidence that reactivation of HSV genomes from the geniculate ganglia is the most important cause of Bell's palsy. Consequently, it has been suggested that "Bell's palsy" be renamed as "herpetic facial paralysis".