Skip to main page content
U.S. flag

An official website of the United States government

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Jun 21;43(25):4580-4597.
doi: 10.1523/JNEUROSCI.0030-23.2023. Epub 2023 May 5.

Developmental Exposure to Polychlorinated Biphenyls Prevents Recovery from Noise-Induced Hearing Loss and Disrupts the Functional Organization of the Inferior Colliculus

Baher A Ibrahim  1   2 , Jeremy J Louie  1 , Yoshitaka Shinagawa  1   2 , Gang Xiao  1   2   3 , Alexander R Asilador  2   3 , Helen J K Sable  4 , Susan L Schantz  2   3   5 , Daniel A Llano  6   2   3   7
Affiliations

Developmental Exposure to Polychlorinated Biphenyls Prevents Recovery from Noise-Induced Hearing Loss and Disrupts the Functional Organization of the Inferior Colliculus

Baher A Ibrahim et al. J Neurosci. .

Abstract

Exposure to combinations of environmental toxins is growing in prevalence; and therefore, understanding their interactions is of increasing societal importance. Here, we examined the mechanisms by which two environmental toxins, polychlorinated biphenyls (PCBs) and high-amplitude acoustic noise, interact to produce dysfunction in central auditory processing. PCBs are well established to impose negative developmental impacts on hearing. However, it is not known whether developmental exposure to this ototoxin alters the sensitivity to other ototoxic exposures later in life. Here, male mice were exposed to PCBs in utero, and later as adults were exposed to 45 min of high-intensity noise. We then examined the impacts of the two exposures on hearing and the organization of the auditory midbrain using two-photon imaging and analysis of the expression of mediators of oxidative stress. We observed that developmental exposure to PCBs blocked hearing recovery from acoustic trauma. In vivo two-photon imaging of the inferior colliculus (IC) revealed that this lack of recovery was associated with disruption of the tonotopic organization and reduction of inhibition in the auditory midbrain. In addition, expression analysis in the inferior colliculus revealed that reduced GABAergic inhibition was more prominent in animals with a lower capacity to mitigate oxidative stress. These data suggest that combined PCBs and noise exposure act nonlinearly to damage hearing and that this damage is associated with synaptic reorganization, and reduced capacity to limit oxidative stress. In addition, this work provides a new paradigm by which to understand nonlinear interactions between combinations of environmental toxins.SIGNIFICANCE STATEMENT Exposure to common environmental toxins is a large and growing problem in the population. This work provides a new mechanistic understanding of how the prenatal and postnatal developmental changes induced by polychlorinated biphenyls (PCBs) could negatively impact the resilience of the brain to noise-induced hearing loss (NIHL) later in adulthood. The use of state-of-the-art tools, including in vivo multiphoton microscopy of the midbrain helped in identifying the long-term central changes in the auditory system after the peripheral hearing damage induced by such environmental toxins. In addition, the novel combination of methods employed in this study will lead to additional advances in our understanding of mechanisms of central hearing loss in other contexts.

Keywords: auditory midbrain; inferior colliculus; noise-induced hearing loss; oxidative stress; polychlorinated biphenyls; two-photon imaging.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
Developmental exposure to PCBs impairs hearing in male mice. A, The timeline of the experimental design of the PCB dosing of dams and hearing assessment of the pups. B, A line plot of the hearing threshold across different frequencies for all pups came from dams under different exposures (one-way ANOVA: F(2,33) = 6.1, p = 0.005, Fisher post hoc test: *p = 0.002 and 0.007 for Oil vs 6 and 12 mg/kg PCB, respectively, and p = 0.58 for 6 vs 12 mg/kg PCB at 4 kHz and F(2,33) = 10.8, *p = 2.4 × 10−4 and 7.1 × 10–5 for Oil vs 6 and 12 mg/kg PCB, respectively, and p = 0.28 for 6 vs 12 mg/kg PCB at 8 kHz and F(2,32) = 3.1, p = 0.06 at 16 kHz and F(2,26) = 1.2, p = 0.31 at 24 kHz). C, A line plot of the hearing threshold across different frequencies for male pups came from dams under different exposures (one-way ANOVA: F(2,16) = 8.3, p = 0.003, Fisher post hoc test: *p = 9.1 × 10−4 and 0.01 for Oil vs 6 and 12 mg/kg PCB, respectively, and p = 0.19 for 6 vs 12 mg/kg PCB at 4 kHz and F(2,16) = 5.5, p = 0.02, Fisher post hoc test: *p = 0.02 and 0.005 for Oil vs 6 and 12 mg/kg PCB, respectively, and p = 0.43 for 6 vs 12 mg/kg PCB at 8 kHz and F(2,15) = 2.0, p = 0.17 at 16 kHz and F(2,11) = 1.1, p = 0.36 at 24 kHz). D, A line plot of the hearing threshold across different frequencies or female pups came from dams under different exposures (one-way ANOVA: F(2,14) = 1.1, p = 0.36 at 4 kHz and F(2,13) = 3.2, p = 0.08 at 8 kHz and F(2,14) = 1.1, p = 0.35 at 16 kHz and F(2,14) = 0.4, p = 0.67 at 24 kHz). * The significance against Oil-treated group.
Figure 2.
Figure 2.
Developmental exposure to PCB inhibits the recovery from NIHL. A, The timeline of the experimental design of the PCBs dosing of dams, the overexposure to the noise of the pups, and hearing assessment of the pups before or after noise overexposure. B, A bar graph showing the threshold to flat noise under different exposures across two-time points (days 0 and 120; one-way ANOVA: F(7,76) = 14.3, p = 9.6 × 10−12, Fisher post hoc test: *p = 0.009, 2.8 × 10−7, or 2.1 × 10−8 for Oil/NU vs PCB/NU, Oil/NE, or PCB/NE, respectively, at day 0 and *p = 0.03, 9.3 × 10−6, or 3.9 × 10−7 for Oil/NU vs PCB/NU, Oil/NE, or PCB/NE, respectively, at day 120 and #p = 4.8 × 10−4 or 0.7.1 × 10−5 for PCB/NU vs Oil/NE or PCB/NE, respectively, at day 0 and #p = 0.03 or 0.005 for PCB/NU vs Oil/NE or PCB/NE, respectively, at day 120 and p = 0.74 and 0.48 for Oil/NE vs PCB/NE at days 0 and 120, respectively, and p = 0.45, 0.24, 0.57, 0.30 for day 0 vs 120 at Oil/NU, PCB/NU, Oil/NE, and PCB/NE, respectively). C, A line graph of the ABR threshold to pure tones from different exposure groups (at 5 kHz, one-way ANOVA: F(3,36) = 8.7, p = 1.8 × 10−4, Fisher post hoc test: *p = 0.004, 1.6 × 10−4, or 5.2 × 10−5 for Oil/NU vs PCB/NU, Oil/NE, or PCB/NE, respectively, and p = 0.19 for PCB/NU vs Oil/NE, 0.11 for PCB/NU vs PCB/NE, and 0.79 for Oil/NE vs PCB/NE; at 10 kHz, one-way ANOVA: F(3,36) = 11.4, p = 2.0 × 10−5, Fisher post hoc test: *p = 0.005, 5.0 × 10−5, or 4.6 × 10−6 for Oil/NU vs PCB/NU, Oil/NE, or PCB/NE, respectively, and #p = 0.02 PCB/NU vs PCB/NE and p = 0.08 for PCB/NU vs Oil/NE and 0.63 for Oil/NE vs PCB/NE; at 14 kHz, one-way ANOVA: F(3,36) = 15.1, p = 1.6 × 10−6, Fisher post hoc test: *p = 0.02, 4.0 × 10−6, or 1.2 × 10−6 for Oil/NU vs PCB/NU, Oil/NE, or PCB/NE, respectively, and #p = 0.04 or 0.002 for PCB/NU vs Oil/NE or PCB/NE, respectively, and p = 0.81 for Oil/NE vs PCB/NE; at 16 kHz, one-way ANOVA: F(3,34) = 12.6, p = 1.1 × 10−5, Fisher post hoc test: *p = 0.003, 1.7 × 10−5, or 3.5 × 10−6 for Oil/NU vs PCB/NU, Oil/NE, or PCB/NE, respectively, and #p = 0.04 or 0.02 for PCB/NU vs Oil/NE or PCB/NE, respectively, and p = 0.81 for Oil/NE vs PCB/NE; at 20 kHz, one-way ANOVA: F(3,36) = 11.8, p = 1.6 × 10−5, Fisher post hoc test: *p = 0.002, 1.4 × 10−5, or 7.6 × 10−6 for Oil/NU vs PCB/NU, Oil/NE, or PCB/NE, respectively, and p = 0.08 for PCB/NU vs Oil/NE, 0.06 for PCB/NU vs PCB/NE, and 0.94 for Oil/NE vs PCB/NE; and at 24 kHz, Kruskal–Wallis ANOVA, χ2 = 9.04, p = 0.03, post hoc Dunn's test: p = 1.0, 0.22, and 0.41 for Oil/NU vs PCB/NU, Oil/NE, and PCB/NE, respectively, and p = 1.0 for all remaining comparisons). D, A bar graph of the hearing threshold shift to the flat noise immediately, a week, or four months after acoustic trauma for the pups from dams exposed to PCBs or Oil (PCB/NE vs Oil/NE; one-way ANOVA: F(5,38) = 4.4, p = 0.002, Fisher post hoc test: ap = 0.003, p = 0.42, and bp = 8.3 × 10−4 for day 0 vs 7, day 0 vs 120, and day 7 vs 120, respectively, at Oil/NE and p = 0.58, 0.15, and 0.06 for day 0 vs 7, day 0 vs 120, and day 7 vs 120, respectively, at PCB/NE and p = 0.85, $p = 0.03, and p = 0.63 for Oil/NE vs PCB/NE at days 0, 7, and 120, respectively). E, A line graph of the hearing threshold shift to different pure tone frequencies immediately or a week after acoustic trauma for the pups from dams exposed to PCBs or Oil (PCB/NE vs Oil/NE; at 5 kHz, one-way ANOVA: F(3,32) = 7.3, p = 7.0 × 10−4, Fisher post hoc test: *p = 1.5 × 10−4 and p = 0.32 for day 0 vs 7 at Oil/NE and PCB/NE, respectively, and #p = 0.007 for PCB/NE vs Oil/NE at day 7; at 10 kHz, one-way ANOVA: F(3,32) = 4.9, p = 0.007, Fisher post hoc test: *p = 0.002 and p = 0.39 for day 0 vs 7 at Oil/NE and PCB/NE, respectively, and #p = 0.02 for PCB/NE vs Oil/NE at day 7; and one-way ANOVA: F(3,32) = 1.7, p = 0.19 at 14 kHz, F(3,32) = 0.94, p = 0.42 at 16 kHz, F(3,32) = 0.17, p = 0.9 at 20 kHz, and F(3,32) = 0.51, p = 0.67 at 24 kHz). The exposure groups are plotted as Oil/NU: black, PCB/NU: red, Oil/NE: blue, and PCB/NE: purple. * The significance against (Oil/NU) group. # The significance against (PCB/NU) group. $ The significance against (Oil/NE) group. a The significance against day 0. b The significance against day 120.
Figure 3.
Figure 3.
The effect of PCBs and/or noise exposure on the DCIC cells. A, A brightfield image of the surface of the DCIC. B, The time traces of the evoked calcium signals by different stimuli of pure tones obtained from non-GABAergic (top) and GABAergic (bottom) cells of the DCIC imaged from the surface; The length and the color of the gradient green bars indicate the intensity and the frequency of the stimulus, respectively. C, Pseudocolor images show the activity of the cells on the DCIC across different animals from (Oil/NU: first row, PCB/NU: second row, Oil/NE: third row, and PCB/NE: fourth row) in response to the pure tone based on their best frequency as graded green circles for the responsive cells and solid red circles for the nonresponsive cells [Pearson correlation 0.69 (p = 2.4 × 10−37), 0.56 (p = 2.6 × 10−37), 0.7 (p = 6.5 × 10−69), and −0.007 (p = 0.92) for non-GABAergic cells of Oil/NU, PCB/NU, Oil/NE, and PCB/NE, respectively, and Pearson correlation 0.63 (p = 1.9 × 10−6), 0.56 (p = 3.7 × 10−13), 0.62 (p = 5.6 × 10−17), and −0.03 (p = 0.85) for GABAergic cells of Oil/NU, PCB/NU, Oil/NE, and PCB/NE, respectively]. D, A heat map showing the fraction of the responsive non-GABAergic [left; χ2 test: χ2 = 227 (*p < 0.001), 111 (*p < 0.001), and 9.9 (*p = 0.001) for Oil/NE vs PCB/NU, Oil/NE, and PCB/NE, respectively, and χ2 = 36.3 (#p < 0.001) and 311 (#p < 0.001) for PCB/NU vs Oil/NE and PCB/NE, respectively, and χ2 = 181 ($p < 0.001) for Oil/NE vs PCB/NE) and GABAergic (right; χ2 test: χ2 = 79.8 (*p < 0.001), 54.6 (*p < 0.001), and 0.09 (p = 0.67) for Oil/NE vs PCB/NU, Oil/NE, and PCB/NE, respectively, and χ2 = 4.2 (#p = 0.04) and 65.2 (#p < 0.001) for PCB/NU vs Oil/NE and PCB/NE, respectively, and χ2 = 42 ($p < 0.001) for Oil/NE vs PCB/NE] cells to sounds across all exposure groups. E, A cumulative distribution function for non-GABAergic (left; Kruskal–Wallis ANOVA: χ2 = 156, *p < 0.001 for Oil/NE vs PCB/NU, Oil/NE, and PCB/NE, respectively, and #p = 0.003 and #p = 7.1 × 10−7 for PCB/NU vs Oil/NE and PCB/NE, respectively, and $p < 0.001 for Oil/NE vs PCB/NE) and GABAergic (right; Kruskal–Wallis ANOVA: χ2 = 59.9, *p = 5.0 × 10−6, *p = 4.2 × 10−4, *p < 0.001 for Oil/NE vs PCB/NU, Oil/NE, and PCB/NE, respectively, and p = 1.0 and #p = 5.5 × 10−5 for PCB/NU vs Oil/NE and PCB/NE, respectively, and $p = 7.0 × 10−7 for Oil/NE vs PCB/NE) cells of the DCIC based on their best frequency. F, A line graph showing the fraction of evoked non-GABAergic [left; χ2 test: χ2 = 53.3 (*p < 0.001), χ2 = 0.0043 (p = 0.99), and 43.6 (#p < 0.001) for Oil/NU vs PCB/NU, Oil/NE, and PCB/NE, respectively, at 40 dB SPL and χ2 = 0.06 (p = 0.99), χ2 = 0.34 (p = 0.95), and 0.52 (p = 0.91) for Oil/NU vs PCB/NU, Oil/NE, and PCB/NE, respectively, at 60 dB SPL] and GABAergic [right; χ2 test: χ2 = 3.7 (p = 0.29), χ2 = 0.65 (p = 0.88), and χ2 = 9.9 (*p = 0.02) for Oil/NU vs PCB/NU, Oil/NE, and PCB/NE at 40 dB SPL and χ2 = 0.42 (p = 0.92), χ2 = 5.1 (p = 0.16), and 0.001 (p = 0.99) for Oil/NU vs PCB/NU, Oil/NE, and PCB/NE, respectively, at 60 dB SPL] cells of the DCIC at each sound level within the responsive cell population across all exposure groups. The exposure groups are plotted as Oil/NU: black line, PCB/NU: red line, Oil/NE: blue line, and PCB/NE: purple line. * The significance against (Oil/NU) group. # The significance against (PCB/NU) group. $ The significance against (Oil/NE) group. Cereb: cerebellum; CTx: cortex, DCIC: dorsal cortex of the inferior colliculus, TS: the transverse sinus. A rainbow color code was selected for different frequencies, N.R in gray: nonresponsive cells.
Figure 4.
Figure 4.
The effect of PCBs and/or noise exposure on the inhibition and excitation balance of the DCIC. A, Heat maps showing the excitatory (left) and inhibitory (right) average response across the animals of each exposure group (Oil/NU: first row, PCB/NU: second row, Oil/NE: third row, and PCB/NE: fourth row). B, C, Line graphs showing the sound level required to evoke 20% of the sound response in non-GABAergic (one-way ANOVA: At 10 kHz, F(3,20) = 9.0, p = 5.7 × 10−4, Fisher post hoc test: *p = 0.0002, 0.002, and 0.0002 for Oil/NU vs PCB/NU, Oil/NE, and PCB/NE, respectively, p = 0.2 for PCB/NU vs Oil/NE, 0.79 for PCB/NU vs PCB/NE, and 0.28 for Oil/NE vs PCB/NE and at 14.1 kHz, F(3,20) = 8.0, p = 0.001, Fisher post hoc test: *p = 0.004, 0.008, and 0.0001 for Oil/NU vs PCB/NU, Oil/NE, and PCB/NE, respectively, p = 0.62 for PCB/NU vs Oil/NE, 0.18 for PCB/NU vs PCB/NE, and 0.06 for Oil/NE vs PCB/NE and F(3,20) = 0.24, p = 0.86 at 5 kHz, F(3,20) = 0.53, p = 0.67 at 7.1 kHz, F(3,20) = 1.7, p = 0.19 at 20 kHz, F(3,20) = 2.3, p = 0.1 at 28.3 kHz, and F(3,20) = 0.75, p = 0.35 at 40 kHz) or GABAergic (one-way ANOVA: at 10 kHz, F(3,20) = 6.9, p = 0.002, Fisher post hoc test: *p = 0.006, 0.02, and 0.0003 for Oil/NU vs PCB/NU, Oil/NE, and PCB/NE, respectively, p = 0.49 for PCB/NU vs Oil/NE, 0.29 for PCB/NU vs PCB/NE, and 0.07 for Oil/NE vs PCB/NE and at 14.1 kHz, F(3,20) = 6.9, p = 0.002, Fisher post hoc test: *p = 0.01, 0.01, and 0.0002 for Oil/NU vs PCB/NU, Oil/NE, and PCB/NE, respectively, p = 0.78 for PCB/NU vs Oil/NE, 0.16 for PCB/NU vs PCB/NE, and 0.08 for Oil/NE vs PCB/NE and at 28.3 kHz, F(3,20) = 3.3, p = 0.04, Fisher post hoc test: *p = 0.02, p = 0.06 and *p = 0.01 for Oil/NU vs PCB/NU, Oil/NE, and PCB/NE, respectively, p = 0.46 for PCB/NU vs Oil/NE, 0.92 for PCB/NU vs PCB/NE, and 0.37 for Oil/NE vs PCB/NE and F(3,20) = 1.1, p = 0.38 at 5 kHz, F(3,20) = 1.1, p = 0.36 at 7.1 kHz, F(3,20) = 1.6, p = 0.23 at 20 kHz, F(3,20) = 0.75, p = 0.35 at 40 kHz) cells, respectively, across different pure tone frequencies for each group. D, E, Line graphs showing the inhibition/excitation ratio across different amplitudes or frequencies (one-way ANOVA: at 5 kHz, F(3,18) = 4.9, p = 0.01, Fisher post hoc test: *p = 0.037, 0.039, and 0.001 for Oil/NU vs PCB/NU, Oil/NE, and PCB/NE, respectively, p = 0.73 for PCB/NU vs Oil/NE, 0.23 for PCB/NU vs PCB/NE, and 0.09 for Oil/NE vs PCB/NE and at 7.1 kHz, F(3,16) = 5.4, p = 0.009, Fisher post hoc test: *p = 0.01 for Oil/NU vs PCB/NU, #p = 0.02 and 0.003 for PCB/NU vs Oil/NE and PCB/NE, respectively, p = 0.47 for Oil/NU vs PCB/NU, 0.06 for Oil/NU vs Oil/NE, and 0.24 for Oil/NE vs PCB/NE and F(3,19) = 1.8, p = 0.18 at 10 kHz, F(3,19) = 1.4, p = 0.27 at 14.1 kHz, F(3,18) = 0.36, p = 0.78 at 20 kHz, F(3,17) = 1.4, p = 0.29 at 28.3 kHz, and F(3,17) = 1.3, p = 0.30 at 40 kHz), respectively, of pure tone stimulus for each exposure group. The exposure groups are plotted as Oil/NU: black line, PCB/NU: red line, Oil/NE: blue line, and PCB/NE: purple line. * The significance against (Oil/NU) group. # The significance against (PCB/NU) group.
Figure 5.
Figure 5.
The association between the downregulation of inhibition and oxidative stress. A, A cartoon image showing the process of tissue collection and gel running for Western blotting (some features were taken from https://biorender.com). B, An image showing the Western blottings of GAD67 (first row), actin (second row), and SOD2 (third row). C–F, Line graphs showing the correlation between the normalized expression levels of the SOD2 enzyme and either age, body weight, or the normalized expression level of the GAD67 enzyme for the (PCB/NE) group as well as ABR for the (Oil/NE) group. *p-values for each correlation.
See this image and copyright information in PMC

Update of

Similar articles

See all similar articles

Cited by

References

    1. Abbott SD, Hughes LF, Bauer CA, Salvi R, Caspary DM (1999) Detection of glutamate decarboxylase isoforms in rat inferior colliculus following acoustic exposure. Neuroscience 93:1375–1381. 10.1016/s0306-4522(99)00300-0 - DOI - PubMed
    1. Akerboom J, et al. . (2012) Optimization of a GCaMP calcium indicator for neural activity imaging. J Neurosci 32:13819–13840. 10.1523/JNEUROSCI.2601-12.2012 - DOI - PMC - PubMed
    1. Alenazi FSH, Ibrahim BA, Briski KP (2015) Estradiol regulates dorsal vagal complex signal transduction pathway transcriptional reactivity to the AMPK activator 5-aminoimidazole-4-carboxamide-riboside (AICAR). J Mol Neurosci 56:907–916. 10.1007/s12031-015-0541-6 - DOI - PubMed
    1. Amanipour RM, Zhu X, Duvey G, Celanire S, Walton JP, Frisina RD (2018) Noise-induced hearing loss in mice: effects of high and low levels of noise trauma in CBA mice. Conf Proc IEEE Eng Med Biol Soc 2018:1210–1213. - PubMed
    1. Asilador A, Llano DA (2020) Top-down inference in the auditory system: potential roles for corticofugal projections. Front Neural Circuits 14:615259. 10.3389/fncir.2020.615259 - DOI - PMC - PubMed

Publication types

Substances

LinkOut - more resources