Mild pentachlorophenol-mediated uncoupling of mitochondria depletes ATP but does not cause an oxidized redox state or dopaminergic neurodegeneration in Caenorhabditis elegans

Abstract

Aims: Mitochondrial dysfunction is implicated in several diseases, including neurological disorders such as Parkinson's disease. However, there is uncertainty about which of the many mechanisms by which mitochondrial function can be disrupted may lead to neurodegeneration. Pentachlorophenol (PCP) is an organic pollutant reported to cause mitochondrial dysfunction including oxidative stress and mitochondrial uncoupling. We investigated the effects of PCP exposure in Caenorhabditis elegans, including effects on mitochondria and dopaminergic neurons. We hypothesized that mild mitochondrial uncoupling by PCP would impair bioenergetics while decreasing oxidative stress, and therefore would not cause dopaminergic neurodegeneration.

Results: A 48-hour developmental exposure to PCP causing mild growth delay (∼10 % decrease in growth during 48 h, covering all larval stages) reduced whole-organism ATP content > 50 %, and spare respiratory capacity ∼ 30 %. Proton leak was also markedly increased. These findings suggest a main toxic mechanism of mitochondrial uncoupling rather than oxidative stress, which was further supported by a concomitant shift toward a more reduced cellular redox state measured at the whole organism level. However, exposure to PCP did not cause dopaminergic neurodegeneration, nor did it sensitize animals to a neurotoxic challenge with 6-hydroxydopamine. Whole-organism uptake and PCP metabolism measurements revealed low overall uptake of PCP in our experimental conditions (50 μM PCP in the liquid exposure medium resulted in organismal concentrations of < 0.25 μM), and no measurable production of the oxidative metabolites tetra-1,4-benzoquinone and tetrachloro-p-hydroquinone.

Innovation: This study provides new insights into the mechanistic interplay between mitochondrial uncoupling, oxidative stress, and neurodegeneration in C. elegans. These findings support the premise of mild uncoupling-mediated neuroprotection, but are inconsistent with proposed broad "mitochondrial dysfunction"-mediated neurodegeneration models, and highlight the utility of the C. elegans model for studying mitochondrial and neurotoxicity.

Conclusions: Developmental exposure to pentachlorophenol causes gross toxicological effects (growth delay and arrest) at high levels. At a lower level of exposure, still causing mild growth delay, we observed mitochondrial dysfunction including uncoupling and decreased ATP levels. However, this was associated with a more-reduced cellular redox tone and did not exacerbate dopaminergic neurotoxicity of 6-hydroxydopamine, instead trending toward protection. These findings may be informative of efforts to define nuanced mitochondrial dysfunction-related adverse outcome pathways that will differ depending on the form of initial mitochondrial toxicity.

Keywords: Adverse outcome pathway; Caenorhabditis elegans; Dopaminergic neurodegeneration; Mitochondrial uncoupling; Redox tone.

Graphical abstract

Fig. 1

PCP causes growth delay at high concentrations, and decreases ATP and cellular oxidation at lower concentrations. Panel A, representative growth curve for one biological replicate with increasing concentrations of PCP. For experiments, L1-stage larvae were incubated with PCP for 48 h in liquid culture in 96-well plates. Following exposure, worms were loaded into a Copas Biosorter and Time of Flight (TOF) was measured as an indicator of worm length. Asterisks represent p < 0.0001 in one-way ANOVA with Bonferroni-corrected post-test compared to control. Panel B, compiled growth curve for three biological replicates in wild-type (N2) and two biological replicates in the other strains used in this study, BY200 and JV2. A 2-way ANOVA analysis showed a significant effect of exposure concentration (p < 0.0001), but not strain or interaction (p = 0.67 and 0.96, respectively). The 50 µM PCP dose used in all subsequent experiments is highlighted in green. Panel C, uptake data for PCP with wild-type (N2) worms. For experiments, incubations were prepared of UV-inactivated E. coli ± 50 µM PCP (first two bars, no worms present), and E. coli + worms ± 50 µM PCP. For incubations with worms, PCP was measured in the supernatant (External [PCP]) and in the lysed worm pellet after 3 brief washes with K-medium (Internal [PCP]). Panel D, ATP content in control and PCP-exposed worms. ATP content was determined using a luminescence-based ATP kit (see Methods) in wild-type (N2) worms after a 48 h exposure to 50 µM PCP. Asterisks represent p < 0.001, student’s unpaired t-test, 9 biological replicates. Panel E, normalized ratio of Grx1-roGFP2 fluorescence (JV2 strain) with excitation wavelengths of 400 nm / 470 nm after a 48 h exposure to 50 µM PCP or a 30 min exposure to 3 % H₂O₂ (positive control). Ratios were normalized to the vehicle control group. Shown are compiled data from > 3 biological replicates. Asterisks represent *, p < 0.05, **, p < 0.01, one-way ANOVA with Bonferroni-corrected post testing for significance compared to the vehicle control.

Fig. 2

Exposure to PCP reduces spare respiratory capacity and increases proton leak in C. elegans. For experiments, L1-stage N2 (wild-type) worms were incubated with 50 µM PCP and UV-inactivated E. coli for 48 h. Following exposure, worms were loaded into a Seahorse XFe24 plate at a density of approximately 75 worms per well. Panel A, depiction of the concept of the mitochondrial stress test, indicating which portion of the measurements are used to calculate the parameters in panels B-E. Note that although injection of DCCD (ATP synthase inhibitor) and FCCP (uncoupler) are shown as sequential injections, for C. elegans experiments we run these inhibitors in separate wells as previously published (Luz et al. 2015). All measurements are normalized to the confirmed number of nematodes in each well (counted after the conclusion of the flux measurements). Panel B, Basal OCR represents the baseline OCR – non-mitochondrial respiration (azide-inhibited OCR). Panel C, ATP-linked OCR is calculated as the difference between baseline OCR and the DCCD-inhibited rate. Panel D, Spare capacity is calculated as the difference between maximal FCCP-induced OCR and the baseline OCR. Panel E, proton leak is calculated as the difference between DCCD-inhibited OCR and the non-mitochondrial azide-inhibited OCR. Data represent n = 5 wells per group per biological replicate and a total of three biological replicates (or six for basal OCR). *, p < 0.05, **, p < 0.01, student’s unpaired t-test.

Fig. 3

PCP exposure does not cause neurodegeneration or exacerbate 6-hydroxydopamine-induced neurodegeneration. For experiments, 48-h exposures were carried out with 50 µM PCP and UV-inactivated E. coli. After exposure, worms were either plated immediately to K-agar plates to recover (K + group), or were immediately exposed to 25 mM 6-hydroxydopamine (6OHDA group), or its vehicle 5 mM ascorbic acid (AA group). The neurons were then visualized in the BY200 strain expressing GFP under the control of the dat-1 (dopamine transporter) promoter. We focused our analysis on the 4 cephalic neurons in the head, with bright cell bodies near the pharynx and dendrites extending to the nose (Panel A, intact neurons). Damage was scored as the presence of bright inclusions or “blebs” (Panel B) or “breaks” in the dendrite (Panel C). Panel D shows the compiled frequency of each score from 2 biological replicates and a total of 50 worms per group (200 neurons per group). P-value shown is based on a Chi-squared analysis.

Fig. 4

TCBQ is taken up into worms, but cannot explain the effects of PCP. Panel A, metabolism data for PCP. For experiments, incubations were prepared with E. coli + N2 worms ± 50 µM TCBQ. TCBQ was measured in the supernatant (External [TCBQ]) and in the lysed worm pellet after 3 brief washes with K-medium (Internal [TCBQ]). Panel B, growth curve for increasing concentrations of TCBQ. For experiments, L1-stage N2 larvae were incubated with TCBQ and UV-inactivated E. coli for 48 h in liquid culture in 6-well plates. Following exposure, worms (n > 40 per group) were placed on food-free K-agar plates and imaged using a stereomicroscope equipped with a camera. The worm volume was then calculated using the ImageJ plugin WormSizer (Moore et al. 2013). Experiments were performed in biological triplicate. **, p < 0.01, one-way ANOVA with Bonferroni-corrected multiple comparisons compared to control.