The role of poly(ADP-ribose) polymerases in manganese exposed Caenorhabditis elegans

Catherine Neumann a, Jessica Baesler a b, Gereon Steffen a, Merle Marie Nicolai a e, Tabea Zubel c, Michael Aschner d, Alexander Bürkle c, Aswin Mangerich c, Tanja Schwerdtle a b, Julia Bornhorst a b e

Highlights
•Highly sensitive LC–MS/MS method to quantify poly(ADP-ribosyl)ation in C. elegans.
•Characterizing worms defective in pme-1 or pme-2 (orthologues of PARP1 and PARP2).
•pme-1 (orthologue of PARP1) holding the major worm PARP activity.
•Toxic doses of Mn led to PAR-induction in pme-1-deficient C. elegans.
•Mn-induced oxidative stress in pme-1 or pme-2 deletion mutants.

Abstract
Background and aim
When exceeding the homeostatic range, manganese (Mn) might cause neurotoxicity, characteristic of the pathophysiology of several neurological diseases. Although the underlying mechanism of its neurotoxicity remains unclear, Mn-induced oxidative stress contributes to disease etiology. DNA damage caused by oxidative stress may further trigger dysregulation of DNA-damage-induced poly(ADP-ribosyl)ation (PARylation), which is of central importance especially for neuronal homeostasis. Accordingly, this study was designed to assess in the genetically traceable in vivo model Caenorhabditis elegans the role of PARylation as well as the consequences of loss of pme-1 or pme-2 (orthologues of PARP1 and PARP2) in Mn-induced toxicity.

Methods
A specific and sensitive isotope-dilution liquid chromatography-tandem mass spectrometry (LC–MS/MS) method was developed to quantify PARylation in worms. Next to monitoring the PAR level, pme-1 and pme-2 gene expression as well as Mn-induced oxidative stress was studied in wildtype worms and the pme deletion mutants.

Results and conclusion
While Mn failed to induce PARylation in wildtype worms, toxic doses of Mn led to PAR-induction in pme-1-deficient worms, due to an increased gene expression of pme-2 in the pme-1 deletion mutants. However, this effect could not be observed at sub-toxic Mn doses as well as upon longer incubation times. Regarding Mn-induced oxidative stress, the deletion mutants did not show hypersensitivity. Taken together, this study characterizes worms to model PAR inhibition and addresses the consequences for Mn-induced oxidative stress in genetically manipulated worms.
Graphical abstract

Introduction
Metal ions, such as mercury, lead, manganese, copper, iron, aluminum, bismuth, thallium and zinc play crucial roles in the complex multi-factorial mechanisms of neurodegenerative diseases (summarized in [1]). Excessive and prolonged exposure to the plentiful of the naturally occurring trace element manganese (Mn) has been documented to cause neurological impairment which is termed “manganism”. The motor and cognitive deficits are similar to those observed in idiopathic Parkinson’s disease (PD) [2,3]. Differences from PD include the lack of nigrostriatal dopaminergic neuron damage and the classic response to levodopa [4]. However, Mn exposure is further supposed to be a risk factor for the development of PD [5]. In earlier studies Mn neurotoxicity has been described clinically in workers exposed occupationally to high Mn levels, but the exposure scenarios changed during the last century from acute to chronic low-level environmental and/or occupational exposure [3,6].

Concerns are mounting about adverse neurological effects in children, since Mn overexposure may result in lower IQ scores, changes in cognitive abilities, as well as altered short-term memory and motor control [7,8]. To date the molecular mechanisms behind Mn induced neurotoxic effects remain unclear. It has been attributed to alterations in a variety of cellular functions including disruptive effects on the neurochemistry of neurotransmitters or oxidative stress [2,9]. Taking oxidative stress into account, excessive reactive oxygen and nitrogen species (RONS) formation leads to increase of interactions with macromolecules such as the DNA. Recently, we identified the DNA damage related signaling reaction poly(ADP-ribosyl)ation (PARylation) to be highly sensitive to in vitro Mn exposure, corroborating the sensitization of cells to genotoxic treatment [10,11].
PARylation is a posttranslational modification of proteins, which is associated with numerous cellular processes such as DNA repair, protein turnover, inflammation, aging or metabolic regulation [12,13].

Poly(ADP-ribose) polymerase-1 (PARP1) and poly(ADP-ribose) polymerase-2 (PARP2) are localized in the nucleus and both of them participate in the early DNA damage response. Thereby the catalytic activity of PARP1 is stimulated 500-fold by DNA with single-strand or double-strand DNA breaks. Although, the basal level of ADP-ribosylation is relatively low, PARPs can consume up to 90% of cellular NAD+ upon DNA damage attaching ADP-ribose moieties onto various acceptor proteins or PARP1 itself [14]. Consequently, over-activation as well as inhibition of PARP1 or PARP2 does have severe consequences [12,13,15]. Although PARP1 inhibitors have excelled in targeting cancers, its beneficial application in neurodegenerative settings has been controversial [16,17]. On the one hand, it is used as therapeutic option for stroke in clinical trials [16]. On the other hand, PARP1 inhibition diminishes mitochondrial capacity and rate of DNA repair with severe consequences for neuronal cells as cell death [16]. Additionally, since PARP1 activation has been associated with neurite outgrowth and long-term memory [18,19], it is conceivable that chronic PARP1 inhibition may attenuate neurogenesis and learning. Considering the importance of PAR homeostasis as well as findings showing that elevated dietary Mn exposure may cause neurobehavioral and neurocognitive deficits in children [[20], [21], [22]], the role of the DNA damage response in Mn-induced toxicity merits further investigation.

The simplicity and several key features of the nematode Caenorhabditis elegans (C. elegans) turned it into an appealing model organism to study the role of PAR in Mn-induced toxicity in vivo. Characteristics that have been contributed to its success include among others the genetic manipulability, the well-characterized genome and the ease of maintenance. The nematode is less complex than a mammalian system, while still sharing considerable genetic homology (60–80%) [23].

Section snippets
C. elegans strains, Mn treatment and Mn-induced lethality assay
The C. elegans strains were handled and maintained at 20 °C as previously described [24]. The following strains were used in this study: WT N2 Bristol strain, OH7193 (otIs181 [Pdat-1::mCherry + Pttx-3::mCherry] III.; him-8(e1489) IV.) and the deletion mutants RB1042 (parp-1(ok988) I.) and VC1171 (parp-2(ok344) II.). All strains were provided by the Caenorhabditis Genetic Center (CGC; University of Minnesota).

Effect of 1 h and 4 h Mn exposure on the survival of C. elegans
To assess the effect of Mn toxicity and determine optimal dosing, wildtype (WT) worms as well as the deletion mutants of pme-1 and pme-2 were treated with increasing Mn doses. The dose–response survival curves (Fig. 1) show that the genetic deletion of either gene did not increase mortality in L4 worms exposed 1 h or 4 h to Mn, with an LD50 indistinguishable from wildtype (WT) worms for the respective exposure time. 1 h Mn exposure leads to an LD50 of 250 mM (Fig. 1A), whereas 4 h exposure.

Discussion
A balanced regulation of the DNA damage response reaction PARylation is of central importance and a dysregulation (inhibition as well as over-activation) is associated with detrimental consequences for several aspects of brain physiology and physiopathology [16]. This implicates the importance of DNA damage response in neural homeostasis. We have recently identified the DNA damage related signaling reaction PARylation to be highly sensitive to in vitro Mn exposure [10,11].

Declaration of Competing Interest
The authors declare no conflict of interest.

Acknowledgements
We thank the German Research Foundation (DFG) for the financial support of BO 4103/2-1, INST 38/537-1, as well as the DFG Research Unit TraceAge (FOR 2558) and the Konstanz Research School Chemical Biology (KoRS-CB, GSC 218). We thank for the European UPF 1069 Regional Development Fund (EFRE). We would also like to thank the Caenorhabditis Genetics Center (CGC), which is funded by the NIH Office of Research Infrastructure Programs (P40 OD010440), for providing the C. elegans strains used in this work.