Context: Cytomegalovirus (CMV) represents the leading congenital viral infection in humans. Although congenital CMV due to vertically transmitted infections is the main cause of CMV-related diseases, adult CMV infections might still be of clinical significance. It is still discussed how far CMV seropositivity, due to horizontal infection in immunocompetent adults, is able to induce significant dysfunction. The present study investigates in how far CMV seropositivity is an additional risk factor for an increasing susceptibility to sensorineural hearing loss induced by acoustic injury during adulthood in a guinea pig CMV (GPCMV) model of noise-induced hearing loss (NIHL). Methods: Two groups (GPCMV seropositive vs. seronegative) of normal hearing adult guinea pigs were exposed to a broadband noise (5–20 kHz) for 2 hours at 115 dB sound pressure level. Frequency-specific auditory brainstem response recordings for determination of auditory threshold shift were carried out and the number of missing outer hair cells was counted 2 weeks after the noise exposure. Results: The data show a slightly increased shift in auditory thresholds in seropositive animals compared to the seronegative control group in response to noise trauma. However, the observed difference was significant at least at high frequencies. The differences in threshold shift are not correlated with outer hair cell loss between the experimental groups. Conclusion: The results point to potential additional pathologies in a guinea pig NIHL model in correlation to GPCMV seropositivity, which should be taken into account when assessing risks of latent/reactivated CMV infection. Due to the relatively slight effect in the present data, the aim of future studies should be a more detailed consideration (e.g., larger sample size) and to localize possible target structures as well as the significance of the infection route.
Keywords: Auditory brainstem response, cytomegalovirus seropositivity, guinea pig, noise-induced hearing loss
How to cite this article:Cytomegalovirus (CMV) represents the leading congenital viral infection in humans. Worldwide, approximately between 0.2% and 1% of infants are born with a CMV infection. Of these total number of infected children, around 10% to 20% show symptoms such as liver, lung or spleen damage, or neurologic (developmental) problems including hearing loss.[1] In congenitally human CMV (cHCMV) infection, in addition to showing central nervous system infections and brain abnormalities, virus-infected cells have been detected throughout the peripheral auditory system, that is, the cochlea, including the stria vascularis (SV), spiral ganglion neurons (SGNs), and the organ of Corti, representing a possible cellular correlate of sensorineural hearing loss (SNHL) in cHCMV-related diseases.[2],[3],[4] Similar results have been demonstrated using different animal models of congenital CMV infection.[5],[6],[7],[8],[9]
Detailed mechanisms of CMV-induced pathologies are still under investigation. Acute CMV infection is able to induce several molecular mechanisms of injury at the cellular level and interact with cellular function including cell cycle perturbation, modification of apoptosis, or modulation of immune responses.[8],[10],[11]
Although congenital CMV due to vertically transmitted infections (primary CMV infection of the mother during pregnancy) is the main cause of CMV-related diseases, CMV infections might still be of clinical significance when occurring during adulthood as well. Percentage of population-wide CMV infections rises from 60% in children up to more than 90% in the elderly, including both vertical (congenital) and horizontal infections.[12] Acute CMV infections as well as viral reactivation are able to induce profound diseases in immunocompromised adults including inflammatory responses and severe diseases of different organs (e.g., lung, liver, gastrointestinal tract, muscles, and sensory systems).[12],[13],[14],[15]
It is still a matter of debate how far CMV seropositivity, due to horizontal infection between individuals in immunocompetent adults, is able to induce significant dysfunction. Although the majority of seropositive adults do not show any clinical symptoms, some cases show severe CMV infection and CMV reactivation followed by profound diseases including gastrointestinal tract disorders, central nervous system injury (e.g., meningitis, encephalitis, myelitis), sensory impairment, or hematologic and vascular diseases.[16]
Severe damage to sensorineural structures in the young adult inner ear can be induced by direct inoculation of CMV in animal models. It has been demonstrated that intracochlear application of guinea pig CMV (GPCMV) in young adult guinea pigs leads to inflammatory responses in the cochlea accompanied by significant reduction of compound action potentials and comparable auditory sequalae as during congenital infection.[17],[18] Although systemic postnatal CMV infection in animal models and humans seems to be asymptomatic in most studies, viremic spread into the inner ear has been demonstrated within a few days postinfection.[19],[20]
The GPCMV is well-established to study the effects of a CMV infection in general. GPCMV is able to cross the placenta and cause fetal infection.[21] Moreover, the hearing range and structure of the guinea pig inner ear are comparable to humans which made them very useful for auditory research for decades.[22] Like in humans, the congenital infection of guinea pigs with species-specific CMV is known to induce diseases ranging from subtle auditory sensorineural and neurologic deficiencies to congenital deafness. Systemic infection with GPCMV in adult animals produces an immune response, but apparently does not affect the excitability of auditory nerve fibers in normal hearing guinea pigs.[23],[24] Thus, hearing thresholds are not changed upon the virus infection. However, due to the known effects of the infection on sensorineural structures of the inner ear, it could be possible that a predamaged ear is more vulnerable to acoustic overstimulation.
It is the aim of the present study to investigate in how far CMV seropositivity is an additional risk factor for an increasing susceptibility to SNHL induced by acoustic injury during adulthood. As reported in immunodeficient as well as immunocompetent subjects, CMV infection leads to an increased vulnerability to inflammatory diseases.[16] As recent research points to immune responses and inflammatory pathways as major contributors in the development of SNHL after acoustic trauma (e.g., noise exposure),[25],[26],[27] an additional contribution of CMV needs to be explored.
Materials and methodsExperimental groups
Nine young adult normal hearing female guinea pigs (Dunkin Hartley, Cavia porcellus) were used for the experiments. One group consisted of five GPCMV-seronegative (CMV(−)) animals, and the second group consisted of four GPCMV-seropositive (CMV(+)) guinea pigs. Dry blood spot testing was applied to confirm GPCMV serology using enzyme-linked immunosorbent assays and immunofluorescent assays. Testing was performed at the laboratory facilities of the commercial animal breeding company (Envigo, Leicestershire, UK).
The experimental protocol was approved by the governmental Ethics Commission for Animal Welfare (LaGeSo, Berlin, Germany; approval number: G0079/15). This study was carried out in accordance with the recommendations of the EU Directive 2010/63/EU on the protection of animals used for scientific purposes.
Noise exposure
The animals were anesthetized with fentanyl (0.025 mg/kg)/midazolam (1.0 mg/kg)/medetomidine (0.2 mg/kg) and exposed to a broadband white noise (5–20 kHz) at 115 dB sound pressure level (SPL) for 2 hours in a sound-proof chamber (80 × 80 × 80 cm, minimal attenuation 60 dB). Noise was delivered binaurally by a speaker (HTC 11.19; Visaton, Haan, Germany) placed 10 cm above the animal’s head. The speaker was connected to an audio amplifier (Yamaha AS201; Yamaha Corporation, Hamamatsu Shizuoka, Japan) and a DVD player (DK DVD-438, Ratingen, Germany). SPL was calibrated by using a sound level meter (Voltcraft 329, Conrad Electronic SE, Hirschau, Germany) placed close to the animal’s ear.
Animals were observed with a camera placed inside the illuminated chamber during the noise exposure. If necessary, a further dose of anesthetics (50% of the initial dose) was administered. Body temperature was kept at a constant level of 37°C.
Auditory brainstem response
To examine the auditory thresholds, frequency-specific (2, 4, 8, 12, 16, 20, 24, 28, 32, 36, and 40 kHz) auditory brainstem responses (ABRs) were recorded under anesthesia (see above). ABR measurements were carried out 1 day before and 14 days after noise exposure. The tone stimuli were applied binaurally under free-field conditions using a high-frequency loudspeaker (RBT-10, Monacor, Bremen, Germany) placed 10 cm above the animal’s head at different SPLs with a sine-wave generator (Funktionsgenerator FG 250 D; H-Tronic, Hirschau, Germany) and an audio amplifier (Yamaha AS 201; Yamaha Corporation, Hamamatsu Shizuoka, Japan). Subdermal needle electrodes were placed at the vertex (active), mastoid (reference), and at the side of the body (ground). ABR recordings were measured with a multichannel recording system (USB-ME16-FAI-System; Multi Channel Systems, Reutlingen, Germany).
Recorded evoked field potentials (sampling rate 20 kHz) were filtered (bandpass 0.15–3 kHz) and averaged (100×) by the MC Rack software (version 4.6.2.0; Multi Channel Systems). The response threshold was determined by a 5-dB stepwise decrease of stimulus intensity until the neural response disappeared. Mean threshold shift for each frequency was calculated from the thresholds prior to and 14 days after the noise exposure. Results are represented as frequency-specific mean relative auditory threshold shift [±standard error (SE)] in dB of the GPCMV-negative group compared to the GPCMV-positive group.
Hair cell counts
After the final ABR measurement, the animals were sacrificed and tissue was fixed by an intracardial perfusion with 0.9% sodium chloride solution followed by 4% paraformaldehyde (PFA; Sigma, Darmstadt, Germany) in phosphate-buffered saline (PBS; Gibco, Carlsbad, CA, USA). The bullae were removed and stored in 4% PFA solution. Before decalcification with 0.4 M ethylenediaminetetraacetic acid (EDTA, pH 8.0; Roth, Karlsruhe, Germany) for 3 days, the cochleae were removed from the bullae and the surrounding bony structures carefully abraded. To stain outer hair cells (OHCs), cochleae were washed in PBS, incubated for 30 minutes with 0.2% Triton-X (Sigma) in PBS, followed by an incubation with Alexa Fluor 488 phalloidin solution (Molecular Probes, Eugene, OR, USA; 1:40 with PBS) for 60 minutes on a shaker in the dark. After washing in PBS, the organ of Corti (OC) was dissected from the modiolus with microscissors. The OC fragments were embedded with Roti Mount Fluorocare including 4’,6-diamidino-2-phenylindole (DAPI; Roth) on microscope slides with cover slips. The stained specimens were digitized under a microscope using a 20× objective (AxioLab1; Zeiss, Oberkochen, Germany) and fluorescence channels for phalloidin (470 nm) and DAPI (365 nm). Pictures were taken using an AxioCam ICc1 microscope camera (Zeiss) and ZEN 2.3 software (Zeiss). The number of missing OHCs was counted based on the fluorescence staining. For further analysis, the length of the basilar membrane was measured with the segmented line tool in ImageJ (version 1.51t; National Institutes of Health, Bethesda, MD, USA). Results are represented as mean (±SE) relative loss of OHCs (normalized to mean analyzed OC length for all animals/ears). Two OC areas were analyzed separately depending on their tonotopic cochlear location (5–13 and 13–19 mm from apex, respectively) corresponding to the estimated frequency range between 1 and 10 kHz or 10 and 40 kHz.[28],[29]
Statistical analysis
ABR threshold shift and OHC counts were statistically compared between the CMV(+) and CMV(−) group. Data distribution was tested applying the Shapiro–Wilk test. Not normally distributed data were compared using the Mann–Whitney U test, and normally distributed data were compared using the t test for independents samples. All statistical analysis was implemented with SPSS (version 24; IBM, Chicago, IL, USA). The level of significance for the tests was P < 0.05.
ResultsAuditory threshold shift
The data of the present study demonstrated an increase in the measured ABR thresholds in both groups, CMV− and CMV+, 14 days after noise exposure. Moreover, the CMV+ animals showed a higher auditory threshold shift compared to the CMV− group. In the CMV+ group, threshold shift ranged frequency dependent between 13.75 and 23.75 dB (mean shift of 18.52 ± 7.51 dB over the entire tested frequency range), with the highest shift at 32 kHz, whereas in CMV− animals, threshold shift ranged between 8.00 and 16.00 dB (mean shift of 10.73 ± 8.33 dB over the entire tested frequency range). Differences between the groups showed statistical significance at 32 and 36 kHz (P = 0.021 and 0.032, respectively). No statistically significant differences were shown for the other tested frequencies (P-values: 2 kHz: 0.139; 4 kHz: 0.197; 8 kHz: 0.556; 12 kHz: 0.431; 16 kHz: 0.286; 20 kHz: 0.652; 24 kHz: 0.905; 28 kHz: 0.589; 40 kHz: 0.087) [Figure 1].
Figure 1 Threshold shift (mean ± standard error) of the auditory brainstem response at different frequencies of noise-exposed guinea pigs [CMV+ group (dashed line): n = 4; CMV− group (solid line): n = 5). Asterisks indicate significant differences between the groups (P < 0.05). CMV, cytomegalovirus.Outer hair cell loss
Noise exposure induced only a slight OHC loss in both CMV+ and CMV− groups. Mean total OHC loss was 101.3 ± 11.6 cells for CMV− animals and 80.9 ± 10.2 in the CMV+ group. Additionally, OHC loss was analyzed separately for the basal and medial apical cochlear regions. For the basal area (above 10 kHz), mean OHC loss was 30.9 ± 5.0 in the CMV− group and 36.1 ± 3.5 in the CMV+ group. Accordingly, mean cell loss for the medial/apical region of the organ of Corti (below 10 kHz) was 70.4 ± 10.5 for CMV− animals and 44.8 ± 10.4 for CMV+ animals [Figure 2]. No statistically significant differences have been detected between the two groups for the total loss of OHC over the entire length of the organ of Corti (P = 0.237) as well as for the separate analysis of the basal OC region (P = 0.431) or the medial/apical OC region (P = 0.107), respectively.
Figure 2 Outer hair cell loss (mean ± standard error; normalized to mean analyzed length of the organ of Corti) of noise-exposed guinea pigs [CMV+ group (gray columns): n = 4; CMV− group (black columns): n = 5]. Data are represented as the total number of OHC loss (right) and further separated for the basal (left) and medial/apical (middle) cochlear region. No significant differences between the experimental groups have been detected for cell loss in the basal cochlear region, the medial/apical region as well as for the entire organ of Corti, respectively. CMV, cytomegalovirus; ns, not significant; OHC, outer hair cell. DiscussionThe present study investigated in how far CMV seropositivity represents a risk factor for an increasing susceptibility to SNHL induced by acoustic injury in adult guinea pigs without apparent auditory anomalies. The data show an increased shift in auditory thresholds in seropositive animals compared to a seronegative control group, which was significant at high frequencies. Interestingly, differences in threshold shift are not correlated with OHC loss between the experimental groups.
Generally, there is a discrepancy between OHC loss and the amount of hearing loss of both groups in the present experiments and OHC loss alone does not sufficiently explain the shifts in auditory thresholds, which is in line with earlier studies.[30] Therefore, other targets which are possibly responsible for the increased noise-induced hearing loss (NIHL) in seropositive animals have to be taken into account. Studies in the periphery of the auditory system, that is, the cochlea, have shown morphologic changes of SGNs, the SV (including loss of endocochlear potential), supporting cells (SC), fibrocytes of the spiral ligament and spiral limbus, as well as of the OHC stereocilia after noise exposure, which could additionally contribute to auditory threshold increase.[31],[32] These structures could also be identified as target structures of pathologic processes after congenital or adult CMV infection.[6],[33] Degeneration of neural structures has been described as well, including loss of synaptic connectivity between inner hair cells and SGNs, followed by subsequent cell death within the auditory nerve.[34]
The described noise-induced cochlear injury is often accompanied or followed by profound immune responses and inflammatory reactions. NIHL induces calcium upregulation due to neuronal hyperexcitation in sensory structures which is followed by cellular stress (formation of reactive oxygen species and free radicals, excitotoxicity) in several cochlear tissue including OHCs, SV, and SGNs.[35],[36] Moreover, an upregulation of stress signaling molecules has been demonstrated in the cochlea, presumably representing a positive feedback loop for additional and long-lasting calcium uptake leading to sustained toxic calcium levels in the cells. Proinflammatory mediators such as tumor necrosis factor alpha (TNF-α) and transient receptor potential vanilloid 1 (TRPV1) seem to be responsible for sustained calcium influx and were shown to be upregulated after NIHL[37],[38] and could lead to inflammation and apoptosis in target tissue.[39] Immune responses and activation of inflammatory cytokines are proposed to play a significant role in the auditory system and have been detected throughout the entire cochlea.[30],[40],[41] Their activation during cochlear pathologies, including NIHL and subsequent energy deficiency due to overexcitation and reduced blood flow, has been demonstrated in several studies, whereby interleukins (e.g., IL-1β, IL-6) and TNF-α upregulation were predominantly involved. The effects were particularly present within the SV and neural cochlear tissue (SGN and supporting structures).[41],[42],[43],[44]
Inflammatory reactions and pathophysiologic responses including expression of apoptosis-related genes and proteins have been detected in central auditory structures including the nucleus cochlearis of the auditory brainstem and might additionally contribute to the observed auditory disorders.[42],[45],[46],[47]
The CMV infection in juvenile or adult animal models is causing immune responses and inflammation in target structures as well as profound auditory injury, particularly after peripheral auditory system or cerebral inoculation.[33],[48],[49],[50] However, acute inflammatory reactions were shown to be present after systemic adult CMV infection as well.[51] Cytokines including TNF-α play a major role in tissue inflammation during CMV-related immunopathologies and are particularly responsible for sustained inflammatory response during active and reactivated CMV infections. Reactivation of CMV infection could occur in seropositive but asymptomatic (immunocompetent) subjects long time after initial CMV infection. It is induced by TNF-α release and binding to TNF receptors in latently CMV-infected cells, initiating viral replication and inflammation in virus-infected cells.[11],[13],[14],[15] In animal models, it seems that CMV reactivation may act as a possible contributor enhancing the inflammation-related cellular pathologies involving the above-mentioned pathways and could occur in seropositive animals as well.[48],[52]
In summary, several studies suggest that cellular injury could probably occur after asymptomatic congenital as well as late-onset horizontal CMV infection between animals. Therefore, it could be possible that asymptomatic CMV infection, particularly if accompanied by additional impact (e.g., noise trauma in the present study), gaining subsequent pathologies in the target structures. Human studies in hearing impaired patients suggest intracochlear CMV reactivation as one possible contributor to the development of auditory system deficits.[53] The described mechanisms could explain the results of the present study. It seems possible that CMV-infected cells can be damaged to a larger extent in case of activation of inflammatory and cell death pathways involving certain cytokines, which play a critical role in CMV pathologies during (re)activated infection. Although we could not determine the exact time of infection as well as the CMV activation status in our animals, asymptomatic CMV-induced acute tissue inflammation, as described above, possibly represents a significant contributor gaining acoustic injury.[51]
In the auditory system, the activation of the mentioned pathways including, for example, different interleukins and TNF-α has been extensively investigated and could possibly contribute to CMV reactivation increasing cellular vulnerability in auditory injury.[38],[52] As mentioned earlier, even in the seronegative animals, OHC loss does not sufficiently explain our results on auditory threshold shift. Therefore, other target structures seem to be responsible for the observed hearing loss in both seropositive and seronegative animals. The most promising candidates are the SV as well as neuronal structures in the cochlea and possibly in the central auditory system as well, which turned out to be largely involved in auditory system pathologies but are also highly affected during CMV infection-related auditory injury.[5],[6],[9],[23],[27],[38],[42] Moreover, these structures contain large amount of virus during CMV infection.[9],[20],[54]
Taken together, the results of the present study demonstrate significant additional pathologies in a guinea pig NIHL model in correlation to GPCMV seropositivity (and possibly to the CMV activation status), which must be taken into account when assessing risks of latent/reactivated CMV infection. Further, the impact on auditory research has to be considered and implemented into the experimental design. Future studies should clarify which target structures are responsible and further investigate how far the infection route (vertical or horizontal) contributes to the observed effects.
Acknowledgment
The authors thank Dr Tanja Schmidt, Dr Konstanze Grote, and Dr André Dülsner from the Research Institutes of Experimental Medicine at Charité - University Medicine Berlin, Germany for their support of the present study.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References
Correspondence Address:
Dr. Moritz Groschel
Department of Otolaryngology, Unfallkrankenhaus Berlin, Warener Str 7, 12683 Berlin
Germany
Source of Support: None, Conflict of Interest: None
CheckDOI: 10.4103/nah.nah_4_21
留言 (0)