An adverse outcome pathway for parkinsonian motor deficits associated with mitochondrial complex I inhibition

Abstract

Epidemiological studies have observed an association between pesticide exposure and the development of Parkinson's disease, but have not established causality. The concept of an adverse outcome pathway (AOP) has been developed as a framework for the organization of available information linking the modulation of a molecular target [molecular initiating event (MIE)], via a sequence of essential biological key events (KEs), with an adverse outcome (AO). Here, we present an AOP covering the toxicological pathways that link the binding of an inhibitor to mitochondrial complex I (i.e., the MIE) with the onset of parkinsonian motor deficits (i.e., the AO). This AOP was developed according to the Organisation for Economic Co-operation and Development guidelines and uploaded to the AOP database. The KEs linking complex I inhibition to parkinsonian motor deficits are mitochondrial dysfunction, impaired proteostasis, neuroinflammation, and the degeneration of dopaminergic neurons of the substantia nigra. These KEs, by convention, were linearly organized. However, there was also evidence of additional feed-forward connections and shortcuts between the KEs, possibly depending on the intensity of the insult and the model system applied. The present AOP demonstrates mechanistic plausibility for epidemiological observations on a relationship between pesticide exposure and an elevated risk for Parkinson's disease development.

Keywords: Adverse outcome pathway; MPTP; Mitochondrial complex I inhibitor; Parkinson’s disease; Pesticide exposure; Regulatory decision-making; Rotenone.

Fig. 1

Schematic overview on the adverse outcome pathway (AOP) for the development of parkinsonian motor deficits by inhibitor binding to mitochondrial complex I. The AOP is initiated by binding of an inhibitor to mitochondrial complex I as the molecular initiating event (MIE), leading to the activation of a series of key events (KEs) that cover various levels of biological organization. Parkinsonian motor deficits were selected as the adverse outcome (AO) of the present AOP, based on its relevance in risk assessment. Key event relationships (KER) (indicated by arrows) represent the available experimental evidence in the literature, illustrating a quantitative relationship between a KE and its corresponding downstream KE. Overlap with other AOPs: overlap of KEs integrated in the AOP “Inhibition of mitochondrial complex I of nigro-striatal neurons leads to parkinsonian motor deficits” with KEs of other AOPs of the AOP-Wiki (https://aopwiki.org) was examined in October 2017. Mitochondrial dysfunction (KE 2) is part of 9, while Neuroinflammation (KE 5) is part of 3 other AOPs in the AOP Wiki in different stages of development

Fig. 2

Key event relationship 1 (KER 1), linking inhibitor binding to complex I (MIE) and the inhibition of complex I (KE 1). The table shows the result of a qualitative assessment of KER 1 on a 3 point scale (weak, moderate, strong). Biological plausibility and experimental support were rated “strong”, according to the available body of experimental support in the literature. However, the molecular mechanisms associated with electron transfer along the respiratory chain, as well as the sites of inhibitor binding and the mechanisms underlying inhibitor-dependent inactivation of complex I, are not fully elucidated yet. NADH nicotinamide adenine dinucleotide, ATP adenosine triphosphate, NDI-1 yeast NADH dehydrogenase. References: [1] Suzuki and King (1983), [2] Kotlyar et al. (1990), [3] van Belzen et al. (1997), [4] Palmer et al. (1968), [5] Degli Esposti et al. (1996), [6] Friedrich et al. (1994), [7] Ohnishi (1998), [8] Lümmen (1998), [9] Brand (2010), [10] Genova et al. (2001), [11] Galkin and Brandt (2005), [12] Lambert and Brand (2004), [13] Schildknecht et al. (2009), [14] Okun et al. (1999), [15] Talpade et al. (2000), [16] Ino et al. (2003), [17] Greenamyre et al. (1992), [18] Higgins and Greenamyre (1996), [19] Grivennikova et al. (1997), [20] Greenamyre et al. (2001), [21] Lambert and Brand (2004), [22] Ichimaru et al. (2008), [23] Okun et al. (1999), [24] Cleeter et al. (1992), [25] Friedrich et al. (1994), [26] Degli Esposti et al. (1993); [27] Degli Esposti and Ghelli (1994), Degli Esposti et al. (1994), [28] Höllerhage et al. (2009), [29] Seo et al. (1998), [30] Sherer et al. (2003), [31] Sharma et al. (2009), [32] Hirst (2013), [33] Vinogradov et al. (1995), [34] Albracht et al. (1997)

Fig. 3

Key event relationship 2 (KER 2), linking the inhibition of complex I (KE 1) and mitochondrial dysfunction (KE 2). The table shows the result of a qualitative assessment of KER 2 on a 3 point scale (weak, moderate, strong). Biological plausibility and empirical evidence were rated “strong”, based on the vast body of experimental evidence available in the literature. A threshold of complex I inhibition, necessary for the induction of mitochondrial dysfunction, has so far not been defined in the literature. Similar limitations apply for the quantitative assessment, respectively, the definition, of mitochondrial dysfunction. ATP adenosine triphosphate, DA dopamine, PD Parkinson’s disease, NDUFS subunits of NADH-ubiquinone oxidoreductase (complex I), ROS reactive oxygen species, NDI-1 yeast NADH dehydrogenase. References: [1] Wirth et al. (2016), [2] Friedrich et al. (1994), [3] Mailloux (2015), [4] Fernandez-Moreira et al. (2007), [5] Berger et al. (2008), [6] Hoefs et al. (2008), [7] Janssen et al. (2006), [8] Lazarou et al. (2009), [9] Dunning et al. (2007), [10] Ogilvie et al. (2005), [11] Saada et al. (2008), [12] Pagliarini et al. (2008), [13] Koopman et al. (2007), [14] Sheehan et al. (1997), [15] Willems et al. (2008), [16] Ye et al. (2015), [17] Han et al. (2016), [18] Dukes et al. (2016), [19] Wang et al. (2011), [20] Li et al. (2014), [21] Giordano et al. (2012), [22] Piao et al. (2012), [23] Wu et al. (2009), [24] Bi et al. (2008), [25] Nakai et al. (2003), [26] Brownell et al. (1998), [27] Koga et al. (2006), [28] Seo et al. (1998), [29] Sherer et al. (2003), [30] Shults et al. (2002), [31] Moon et al. (2005), [32] Wen et al. (2011), [33] Yang et al. (2009), [34] Matthews et al. (1999), [35] Beal (2011); [36] Przedborski et al. (1992), [37] Zhang et al. (2000), [38] Filomeni et al. (2012), [39] Wang et al. (2015), [40] Nataraj et al. (2016), [41] Lee et al. (2011), [42] Tseng et al. (2014), [43] Liu et al. (2015), [44] Thomas et al. (2012), [45] Pöltl et al. (2012), [46] Bose and Beal (2016), [47] Brownell et al. (1998), [48] Choi et al. (2008), [49] Höllerhage et al. (2009)

Fig. 4

Key event relationship 3 (KER 3), linking mitochondrial dysfunction (KE 2) and impaired proteostasis (KE 3). The table shows the result of a qualitative assessment of KER 3 on a 3 point scale (weak, moderate, strong). While a strong experimental basis exists in the literature to justify the rating “strong” for the experimental support linking KE 2 and KE 3, mechanistic understanding on how mitochondrial dysfunction, respectively, its individual features such as a decline in ATP generation, or an elevated formation of free radical species, affect cellular proteostasis, are only incompletely understood. The situation is further complicated by mutual interactions between mitochondrial dysfunction, oxidative stress, and proteasomal stress that lead to self-amplifying futile cycles but allow no definition on an initiating event. PD Parkinson’s disease, UPS ubiquitin proteasomal system, ALP autophagy–lysosomal pathway, ATP adenosine triphosphate, ROS reactive oxygen species, MPTP 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine, NDI-1 single subunit NADH dehydrogenase of S. cerevisiae. References: [1] Betarbet et al. (2005), [2] McNaught et al. (2003), [3] McNaught and Jenner (2001a, b), [4] Ambrosi et al. (2014), [5] Yu et al. (2009), [6] Martini-Stoica et al. (2016), [7] Komatsu et al. (2006), [8] Menzies et al. (2015), [9] Goldberg (2003), [10] Ding et al. (2003), [11] Zheng et al. (2016), [12] Pickart and Cohen (2004), [13] Finley (2009), [14] Voges et al. (1999), [15] Bose and Beal (2016), [16] Wang et al. (2010, b), [17] Farout et al. (2006), [18] Ishii et al. (2005), [19] Demasi et al. (2003), [20] Demasi et al. (2001), [21] Butterfield and Kanski (2001), [22] Sayre et al. (2001), [23] Fornai et al. (2005), [24] Wu et al. (2015), [25] Liu et al. (2013), [26] Yong-Kee et al. (2012), [27] Pan et al. (2009), [28] Seo et al. (2002), [29] Seo et al. (2000), [30] Seo et al. (1998), [31] Sherer et al. (2003), [32] Shamoto-Nagai et al. (2003), [33] Chou et al. (2010), [34] Filomeni et al. (2012)

Fig. 5

Key event relationship 4 (KER 4), linking impaired proteostasis (KE 3) and DA neurodegeneration (KE 4). The table shows the result of a qualitative assessment of KER 4 on a 3 point scale (weak, moderate, strong). Literature provides conclusive empirical support for a causal and quantitative relationship between KE 3 and KE 4. Insight into the molecular events responsible for DA neurodegeneration in response to impaired proteostasis, however, can only be classified “moderate” due to essential knowledge gaps. UPS ubiquitin proteasomal system, ALS autophagy–lysosomal system, DA dopamine, UCH-L1 ubiquitin carboxy-terminal hydrolase L1, Ndufs4 NADH:ubiquinone oxidoreductase subunit S4, TFEB transcription factor EB. References: [1] Martini-Stoica et al. (2016), [2] Menzies et al. (2015), [3] McNaught and Jenner (2001a, b), [4] McNaught et al. (2003), [5] Ambrosi et al. (2014), [6] Betarbet et al. (2000), [7] Betarbet et al. (2006), [8] Fornai et al. (2005), [9] Davies (2001), [10] Wang et al. (2010), [11] Schmidt et al. (2005), [12] Kitada et al. (1998), [13] Leroy et al. (1998), [14] Song and Cortopassi (2015), [15] Mader et al. (2012), [16] Dehay et al. (2010), [17] Wu et al. (2015), [18] Giordano et al. (2014), [19] Bentea et al. (2015), [20] Li et al. (2012), [21] Fornai et al. (2003), [22] Decressac et al. (2013), [23] Kilpatrick et al. (2015), [24] Decressac and Björklund (2013), [25] Ebrahimi-Fakhari and Wahlster (2013), [26] Decressac et al. (2012), [27] Shimoji et al. (2005), [28] Zhu et al. (2007b)

Fig. 6

Key event relationship 5 (KER 5), linking DA neurodegeneration (KE 4) and neuroinflammation (KE 5). The table shows the result of a qualitative assessment of KER 5 on a 3 point scale (weak, moderate, strong). Both empirical support and biological plausibility were classified as “moderate”, based on the species-dependent variability of mediators originating from degenerating DA neurons. Experimental support for a causal link of KE 4 and KE 5 is mainly based on in vitro models, whereas in vivo information is rather limited. DAMP damage associated molecular patterns, HMGB1 high mobility group box 1, CX3CR1 fractalkine receptor, MPTP 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine, RAGE receptor for advanced glycation end products, NF-kB nuclear factor kappa B. References: [1] McGeer et al. (2003), [2] Miklossy et al. (2006), [3] Béraud et al. (2013), [4] Thundyil and Lim (2015), [5] Chao et al. (2014), [6] Fossati and Chiarugi (2007), [7] Liu et al. (2012), [8] Fellner et al. (2013), [9] Farina et al. (2007), [10] Efremova et al. (2015), [11] Davalos et al. (2005), [12] Haynes et al. (2006), [13] Koizumi et al. (2007), [14] Shinozaki et al. (2017), [15] Blank and Prinz (2013), [16] Chapman et al. (2000), [17] Streit et al. (2001), [18] Nayak et al. (2011), [19] Lopategui Cabezas et al. (2014), [20] Shan et al. (2011), [21] Zecca et al. (2008), [22] Santoro et al. (2016), [23] Sasaki et al. (2016), [24] Noelker et al. (2013), [25] Abdelsalam and Safar (2015), [26] Schildknecht et al. (2015), [27] Emmanouilidou et al. (2010), [28] Marques and Outeiro (2012)

Fig. 7

Key event relationship 6 (KER 6), linking DA neuroinflammation (KE 5) and DA neurodegeneration (KE 4). The table shows the result of a qualitative assessment of KER 6 on a 3 point scale (weak, moderate, strong). A causal relationship between neuroinflammation and DA neurodegeneration has been demonstrated. Biological plausibility and empirical support were both rated “moderate”, due to the lack of profound knowledge on the mediators that evoke neurodegeneration. Anti-inflammatory and antioxidant agents could not convincingly demonstrate a neuroprotective potential. CNS central nervous system, DA dopamine, IL-1β interleukin-1β, IFN-γ interferon-γ, TNF-α tumor necrosis factor α; TGF: transforming growth factor, NSAID non-steroidal anti-inflammatory drugs. References: [1] McGeer et al. (2003), [2] Miklossy et al. (2006), [3] Liberatore et al. (1999), [4] Norden et al. (2015), [5] Boka et al. (1994), [6] Dong and Benveniste (2001), [7] Lopez-Ramirez et al. (2014), [8] Pan and Kastin (2002), [9] Banks (2005), [10] Heráandez-Romero et al. (2012), [11] Pott Godoy et al. (2008), [12] Villarán et al. (2010), [13] Hirsch and Hunot (2009), [14] Griffin et al. (1998), [15] Blasko et al. (2004), [16] Barbeito et al. (2010), [17] Herrera et al. (2000), [18] Frank-Cannon et al. (2008), [19] He et al. (2013), [20] Ramsey and Tansey (2014), [21] Tanaka et al. (2013), [22] Mount et al. (2007), [23] Ferger et al. (2004), [24] Leng et al. (2005), [25] Sriram et al. (2002), [26] Sriram et al. (2006), [27] Qin et al. (2007), [28] McCoy et al. (2006), [29] Castaño et al. (2002), [30] Brochard et al. (2009), [31] Reynolds et al. (2007), [32] Laurie et al. (2007), [33] Liu et al. (2016), [34] Faust et al. (2009), [35] Du et al. (2001), [36] Tikka et al. (2001), [37] Wu et al. (2002), [38] Shults (2003), [39] NINDS NET-PD Investigators (2006), [40] NINDS-NET-PD Investigators (2008), [41] Chen et al. (2005), [42] Chen et al. (2003), [43] Hernán et al. (2006), [44] Ton et al. (2006), [45] Etminan et al. (2008), [46] Schildknecht et al. (2005), [47] Hoos et al. (2014), [48] Parkinson Study Group (1993), [49] Shoulson (1998)

Fig. 8

Key event relationship 7 (KER 7), linking mitochondrial dysfunction (KE 2) and DA neurodegeneration (KE 4). The table shows the result of a qualitative assessment of KER 7 on a 3 point scale (low, moderate, strong). The literature is currently lacking a generally accepted definition of mitochondrial dysfunction. There is currently no consensus on the contribution of individual processes (e.g. mitochondrial membrane potential loss, ROS formation, drop in ATP formation, release of pro-apoptotic factors, etc.) to overall mitochondrial dysfunction nor a quantitative assessment of these processes for threshold definition. However, for some endpoints, semi-quantitative information is available. Notably, the support that KER 7 prevails over KER 3 and KER 4 is limited to few experimental situations, and human evidence has not been established. ATP adenosine triphosphate, ROS reactive oxygen species, MPP ⁺ 1-methyl-4-phenylpyridinium, NDI-1 single subunit NADH dehydrogenase of S. cerevisiae. References: [1] Bose and Beal (2016), [2] Banerjee et al. (2009), [3] Subramaniam and Chesselet (2013), [4] Herrero-Mendez et al. (2009), [5] Almeida et al. (2001), [6] Almeida et al. (2004), [7] Nedergaard et al. (1993), [8] Guzman et al. (2009), [9] Chan et al. (2007), [10] Surmeier et al. (2011), [11] Surmeier and Schumacker (2013), [12] Bolam and Pissadaki (2012), [13] Matsuda et al. (2009), [14] Pissadaki and Bolam (2013), [15] Pacelli et al. (2015), [16] Schildknecht et al. (2017), [17] Chan et al. (1991), [18] Fabre et al. (1999), [19] Hasegawa et al. (1990), [20] Nicklas et al. (1985), [21] Przedborski et al. (1996), [22] Sherer et al. (2003), [23] Sherer et al. (2007), [24] Marella et al. (2008), [25] Ekstrand et al. (2007), [26] Du et al. (2001), [27] Choi et al. (2014), [28] Hajieva et al. (2009), [29] Chen et al. (2015), [30] Marella et al. (2008), [31] Wen et al. (2011), [32] Beal et al. (1998), [33] Adhihetty and Beal (2008), [34] Cunha et al. (2014), [35] Seo et al. (1998, 2000, 2002), [36] Shults et al. (2002), [37] Moon et al. (2005), [38] Wen et al. (2011), [39] Wang et al. (2012), [40] Leist et al. (1998), [41] Leist et al. (1997)

Fig. 9

Key event relationship 8 (KER 8), linking DA neurodegeneration (KE 4) and parkinsonian motor deficits (AO). The table shows the result of a qualitative assessment of KER 8 on a 3 point scale (weak, moderate, strong). Literature provides strong evidence for a causal correlation between the levels of striatal dopamine and the onset of parkinsonian motor deficits. These correlations can be observed in MPTP exposed rodents, primates, including humans, and in human PD. A potential contribution of other brain areas, respectively, their demise, to parkinsonian motor deficits, was only inadequately investigated so far. DA dopamine, PD Parkinson’s disease, l -DOPA l-3,4-dihydroxyphenylalanine, DAT dopamine transporter, VMAT-2 vesicular monoamine transporter 2, TH tyrosine hydroxylase. References: [1] Lynd-Balta and Haber (1994a), [2] Lynd-Balta and Haber (1994b), [3] Joel and Weiner (2000), [4] Alexander et al. (1986), [5] Obeso et al. (2008a), [6] Blandini et al. (2000), [7] Ehringer et al. (1960), [8] Bernheimer et al. (1973), [9] Koller (1992), [10] Kirik et al. (1998), [11] Earle (1968), [12] Lloyd et al. (1975), [13] Benamer et al. (2000), [14] Rakshi et al. (1999), [15] Lin et al. (2014), [16] Pirker (2003), [17] Rinne et al. (1995), [18] Tissingh et al. (1998), [19] Lloyd et al. (1975), [20] Yam et al. (1998), [21] Gilmour et al. (2011), [22] Heimer et al. (2002), [23] Papa et al. (1999), [24] Hutchinson et al. (1997), [25] Levy et al. (2001), [26] Parkinson Study Group (1993), [27] Pålhagen et al. (1998), [28] Pålhagen et al. (2006); [29] Parkinson Study Group (1996), [30] Olanow et al. (2008), [31] Widner et al. (1992), [32] Kordower et al. (1998), [33] Kordower et al. (1995), [34] Mendez et al. (2008), [35] Schumacher et al. (2000), [36] Ben-Hur et al. (2004), [37] Bezard et al. (2001), [38] Blesa et al. (2012), [39] Mitchell et al. (1989), [40] Filion and Tremblay (1991), [41] Bergman et al. (1990), [42] Aziz et al. (1991), [43] Porras et al. (2012), [44] Jenner (2008), [45] Bédard et al. (1986), [46] Clarke et al. (1987), [47] Langston et al. (2000), [48] Smith et al. (2003), [49] Kuoppamäki et al. (2007), [50] Seniuk et al. (1990), [51] Muthane et al. (1994), [52] Moratalla et al. (1992), [53] Snow et al. (2000), [54] Forno et al. (1986), [55] Petzinger et al. (2006), [56] Jakowec et al. (2004), [57] Rothblat et al. (2001), [58] Meredith and Kang (2006)