Over the last decade, transgenic models of C. elegans have been successfully used to study PD-like pathologies and behaviors (Caldwell and Caldwell, 2008; Harrington et al., 2010). In humans, monogenic forms of PD, caused by a single gene mutation in a dominant or recessive fashion, are well-established, though relatively rare types of the disease. They account for approximately 30% of the familial cases (Klein and Westenberger, 2012).

To model α-synuclein aggregation and accumulation in vivo, researchers have generated transgenic C. elegans strains that express the human α-synuclein gene in body wall muscle cells and in neurons (Table 1). In these models, increased or decreased fluorescence intensity associated with YFP linked to α-synuclein can be quantified to determine the levels of protein expression (Jadiya et al., 2011; Jadiya and Nazir, 2012; Fatima et al., 2014; Fu et al., 2014a; Chen Y. M. et al., 2015; Liu et al., 2015). Loss of fluorescence intensity indicates reduced protein expression, whereas increased fluorescence indicates increased α-synuclein expression. Such changes in protein expression can be visualized using microscopy (Figure 1) and analyzed using freely available programs such as FIJI (Schindelin et al., 2012). Alterations in protein expression can also be assessed using techniques such as fluorescence resonance energy transfer (FRET) or fluorescence recovery after photobleaching (FRAP) (Bodhicharla et al., 2012).

Aberrant neuronal morphologies caused by exposure to neurotoxins or heavy metals can be visualized using fluorescence microscopy and quantified by counting the types and frequencies of aberrations. Such investigations typically focus on dopaminergic neurons of the head, i.e., the four CEPs and two ADEs (Figures 2a,b). Aberrant morphologies include the loss of neuronal cell bodies (Figure 2c), the absence of neuronal processes, broken neurites (Figure 2d), shrinking of dendritic endings, and the appearance of vacuoles (Nass et al., 2002; Pu and Le, 2008; Yao et al., 2010b; Masoudi et al., 2014). In addition, neurons exposed to toxins may appear dark, rounded and/or small, exhibit neuritic blebbing (Figure 2e), and lose GFP expression (Nass et al., 2002; Berkowitz et al., 2008; Pu and Le, 2008; VanDuyn et al., 2010; Ali and Rajini, 2012; Fu et al., 2014a; Masoudi et al., 2014). Selective degeneration can be scored based on any of these morphological changes or the absence of the neurons. In addition, C. elegans dopaminergic neurons that express human α-synuclein degenerate by mid-life (Hamamichi et al., 2008). In contrast, most genetic mouse models of α-synuclein fail to show degeneration of dopamine neurons (Blesa et al., 2012a; Blesa and Przedborski, 2014).

Dopamine reuptake transporters (DAT-1 in C. elegans) play a crucial role in the uptake of environmental toxins such as 6-OHDA and MPTP, which enter neurons, cause cell degeneration, and decrease the levels of endogenous dopamine (Gainetdinov et al., 1997; Nass et al., 2002; Pu and Le, 2008; Ali and Rajini, 2012; Masoudi et al., 2014). The dopamine content in C. elegans treated with a neurotoxin can be measured by reverse phase high performance liquid chromatography (RP-HPLC) with electrochemical detection (Pehek et al., 2005; Yao et al., 2010b; Satapathy et al., 2016). Reduced levels of dopamine and the resulting behavioral deficits are found in C. elegans overexpressing LRRK2 (both wild-type and mutated forms; Yao et al., 2010b). LRRK2 animals have a 50–72% reduction in dopamine levels compared to the wild-type control strain N2 (Yao et al., 2010b).

An alternative method is to measure dopamine content using HPLC followed by the detection of chemiluminiscence (Kuwahara et al., 2006; Tsunoda, 2006; Fu et al., 2014a). Post separation, colorimetric oxidation, fluorescence derivatization with ethylenediamine, and peroxyoxalate chemiluminiscence reaction detection are then performed on the extracts containing dopamine and its metabolites (Tsunoda, 2006). This method is highly sensitive, with detection limits in the fentomolar range, and makes it possible to measure dopamine in small-volume samples. HPLC with chemiluminiscence was used to show reduced dopamine levels and a reduced locomotory phenotype in transgenic C. elegans strains expressing A30P or A53T mutant α-synuclein in dopamine neurons (Kuwahara et al., 2006). In another study, the same technique revealed that the dopamine content of 6-OHDA-treated animals was 64% less than untreated controls (Fu et al., 2014a). Interestingly, the levels of dopamine in 6-OHDA-treated animals are elevated after treatment with the natural compound n-butylidenephthalide (Fu et al., 2014a). Additionally, using HPLC with UV detection, Ali and Rajini showed that the dopamine levels of MPTP and organophosphorous insecticide-exposed C. elegans are lower than in untreated controls (Ali and Rajini, 2012).

Prior research suggests that the aggregation of α-synuclein oligomers is associated with lipid peroxidation due to ROS overload, which can alter cellular membrane composition (Binukumar et al., 2010; Angelova et al., 2015). In C. elegans, intracellular fat droplets can be stained with the fluorescent dye Nile red (a lipophilic stain that fluoresces in a lipid environment), visualized with fluorescence microscopy (Figure 3), and quantified by analyzing the fluorescence staining. Several studies have shown that the NL5901 and OW13 strains have lower Nile red fluorescence than wild-type control N2 animals of the same age, indicating a reduced lipid content in PD strains (Jadiya et al., 2011; Jadiya and Nazir, 2012; Fu et al., 2014a). However, further investigation is warranted to describe the role of lipid content in C. elegans models of PD. For example, the lipid content of PD strains should be directly compared to their corresponding genetic controls, not just to wild-type N2 control strains. In addition, another widely used stain for measuring lipid content in C. elegans, Oil Red O, is a fat-soluble diazo dye that has been widely used to stain lipid droplets in mammalian cells and tissues, and has recently been applied to observing lipid stores in C. elegans (O'Rourke et al., 2009; Elle et al., 2010; Wahlby et al., 2014). This stain measures fat stores contained only in lipid droplets, which correlate well with biochemically-measured lipids (total fatty-acid methyl esters). Whereas Nile red primarily stains acidic lysosome-related gut granules in live or fixed animals (Elle et al., 2010; Wahlby et al., 2014), Oil Red O shows a better correlation with triglyceride levels. Depending on the solvent, Oil Red O can also stain cellular structures in non-adipogenic cell lineages (Elle et al., 2010). Yen and colleagues showed that fixed, but not Nile red-fed (live) wild-type N2 animals reveal fat stores that match label-free coherent anti-Stokes Raman scattering (CARS) imaging or Oil Red O and Nile Red fixed imaging (Yen et al., 2010). Another study by Barros and coworkers compared different methods (Nile red, BODIPY, Sudan Black and Oil Red O) to study the effect of dopamine signaling on fat content in C. elegans. Results showed similarity between fixative based dyes (Sudan Black and Oil Red O) and vital dyes (BODIPY and Nile Red) with smaller measurable decreases for the vital dyes (Barros et al., 2014). It will be interesting to further elucidate the role of lipid content in cellular hallmarks of PD by using these additional tools.

The locomotion rates of C. elegans change in the presence or absence of food, and feeding status. For example, well-fed, wild-type animals move more slowly in the presence of bacteria versus when there is no bacterial food source on the petri dish. This foraging behavior is dependent on dopaminergic neurons, which mechanically sense the presence and availability of bacterial food by its texture, and when food is present, decrease the animals' locomotion (Sawin et al., 2000). This slowing in response to abundant food in well-fed animals is called basal slowing. Deficits in dopaminergic function are associated with higher locomotion in the presence of food in well-fed animals, as evidenced by their lower basal slowing response (Yao et al., 2010b; Chen et al., 2013). Starved C. elegans slow down more dramatically in the presence of food, a phenomenon referred to as the enhanced slowing response, which ensures that the animals do not leave their newly found food source (Sawin et al., 2000; Rivard et al., 2010). Unlike basal slowing, which is controlled by dopaminergic neurons, the enhanced slowing response is regulated by serotonin (Sawin et al., 2000).

To determine the basal slowing response, animals are washed in buffer (typically M9) and then transferred to NGM plates with or without OP50 bacterial lawns. Basal slowing, which is measured as the frequency of body bends, is then recorded for 20–60 s and analyzed using data acquisition software as follows: (basal slowing = [rate of movement in the absence of food − rate of movement in the presence of food]/rate of movement in the presence of food) (Cooper et al., 2015). Basal slowing or food sensing behavior can also be measured as: (basal slowing = [movement rate of the animals in the presence of bacteria/movement rate in absence of bacteria] × 100) (Kuwahara et al., 2006). The study by Cooper et al. used food-sensing behavior to assess the functional loss of dopamine neurons in C. elegans expressing the familial Parkinson mutant human α-synuclein in dopamine neurons. The results showed that C. elegans expressing either human mutant α-synuclein (A53T) or human mutant LRRK2 (G2019S) exhibited deficits in this dopamine-related behavior (Cooper et al., 2015). This deficiency can be rescued by a mutation in the insulin-IGF1 receptor C. elegans ortholog, daf-2, a key modulator of aging pathways (Kenyon et al., 1993). Interestingly, the overexpression of LRRK2 (both wild-type and G2019S mutated forms) and cat-2 deletion disrupt the age-dependent basal slowing response. This diminished behavior can be rescued by treatment with exogenous dopamine (Yao et al., 2010b; Johnson et al., 2015).

Area-restricted searching (ARS) is a foraging behavior in which wild-type animals minimize searching in areas that have abundant food and extend the search to larger areas when food is scarce. As the time since removal from food increases, animals turn less frequently towards the food (Hills et al., 2004; Gray et al., 2005; Chen et al., 2013). This is a goal-directed behavior that involves dopamine signaling. Removing or damaging dopaminergic neurons can lead to abnormal or abolished ARS behavior. For example, ARS behavior was rescued by administering exogenous dopamine to animals with defective dopamine signaling (Hills et al., 2004). ARS can be measured by transferring well-fed animals to NGM plates and videotaping them for 60 s after 5 and 30 min. The number of turns that exceed 90 degrees are counted from the tracks of each animal at each time-point (Cooper et al., 2015). ARS is impaired in both α-synuclein and LRRK2 PD mutants (Cooper et al., 2015). Daf-2 mutations increase searching in both PD strains, suggesting a role for aging in modulating dopamine-dependent behaviors in nematode models of PD (Cooper et al., 2015).

Caenorhabditis elegans can sense and respond to a multitude of environmental cues. These responses can be both aversive and attractive (Bargmann, 2006). For example, under standard laboratory culturing conditions, untreated wild-type (N2) animals avoid ethanol. However, when these animals are continuously exposed to ethanol, they develop a tolerance to and preference for ethanol, a response which is controlled by the dopamine system (Davies et al., 2004; Lee et al., 2009). Unlike wild-type animals, cat-2 and tph-1 mutants lacking a functional dopamine system do not develop an ethanol preference to chronic ethanol exposure (Lee et al., 2009). Ethanol avoidance is significantly decreased in non-ethanol-pretreated animals that express human mutant α-synuclein and mutant LRRK2 compared to those expressing wild-type α-synuclein and wild-type LRRK2 (Cooper et al., 2015). Interestingly, ethanol avoidance is restored in an α-synuclein mutant with a deletion of the daf-2 gene, indicating that slowing aging also slows PD symptoms. To assay ethanol preference as a surrogate measure of the dopamine system, animals are incubated on an ethanol plate and transferred to assay plates that are divided into equal quadrants. Ethanol is provided in two quadrants, and animals are allowed to move freely for 30 min; the time preference for the quadrants is scored during this time. A preference index (PI) is calculated as ([number of animals in the ethanol quadrants]–[number of animals in control quadrants])/the total number of animals tested (Lee et al., 2009). This assay could also be used to assess the PI of PD animals with an impaired dopaminergic system caused by chemical exposure.

In C. elegans, the response to the aversive odorant nonanol is regulated by dopamine signaling (Bargmann, 2006; Kimura et al., 2010; Fatima et al., 2014; Sashidhara et al., 2014; Satapathy et al., 2016). When a drop of nonanol is placed near the head of a wild-type worm, the worm senses it and moves away as a chemotactic “aversive” response. However, when its dopamine content is diminished, the animals take longer to respond to the chemical stimulus. The response time to nonanol is increased 2-fold in the α-synuclein overexpressing strain NL5901 after treatment with ida-1 (ortholog of mammalian diabetes autoantigen IA-2) RNAi (Fatima et al., 2014). In contrast, certain botanical compounds have shown to reduce the time required by both wild-type (N2 exposed to 6-OHDA/pesticide) and α-synuclein overexpressing strains (NL5901) to respond to nonanol (Sashidhara et al., 2014; Satapathy et al., 2016). This suggests that the “nonanol repulsion assay” can be used as an indirect measure of dopamine content in nematodes with impaired dopamine signaling.

Gait can be defined as alterations in the patterns of movement based on the environment currently occupied by an animal. In humans, the basal ganglia regulate motor movement during gait, which activates dopaminergic neurons (Marsden, 1982; Mink and Thach, 1991; Fukuyama et al., 1997; Koepp et al., 1998). In PD, dysfunction in the basal ganglia region contributes to impaired gait functions and rhythms (Morris et al., 1996; Hausdorff et al., 1998; Sofuwa et al., 2005). Gaits in C. elegans are mainly characterized as crawling (on solid “agar” media) and swimming (in liquid media) (White et al., 1986; Pierce-Shimomura et al., 2008). On agar, nematodes move or crawl in a classical sinusoidal fashion. This changes to “thrashing” or swimming when the animals are moved to liquid media. The mechanisms behind this gait transition are unknown; however, roles for bioamine neurotransmitters such as dopamine and serotonin have been implicated (Mesce and Pierce-Shimomura, 2010).

In C. elegans, dopamine is responsible for a wide array of behaviors including the gait transition from swim to crawl (Vidal-Gadea et al., 2011). The activation of dopamine neurons by optogenetics confirmed that the switch from crawling to swimming involves signaling through D1-like dopamine receptors, which is similar to the pattern the animals exhibit when they crawl off the bacterial food source (Sawin et al., 2000; Vidal-Gadea et al., 2011). Under both conditions, dopamine functions by decreasing the speed of the animal's movement. Animals with impaired dopaminergic signaling can exhibit opposing behavioral phenotypes, and the genetic ablation of all dopaminergic neurons can impair the transitions between swimming and crawling and lead to paralysis in animals due to incessant swimming (the swimming-induced paralysis or SWIP phenotype; Vidal-Gadea et al., 2011). The dopamine transporter DAT-1 plays an important role in dopamine reuptake and clearance. In animals that exert maximal physical activity during swimming, mutations in DAT-1 lead to SWIP (McDonald et al., 2007). In humans, PD is characterized by impaired gait and the failure to transition between locomotory patterns (Jankovic, 2008). The failure of C. elegans to transition between swimming and crawling when the dopamine system is impaired reinforces the validity of C. elegans PD models.

A swim-to-crawl assay involves growing animals on NGM plates seeded with OP50 bacteria and then changing the environmental conditions to affect movement. Such changes could include increasing or decreasing the viscosity of the medium or providing mechanical stimulation with magnetic particles, as described by Vidal-Gadea et al. (2011). Gait transitions are evaluated by video recording movement before and after altering the conditions. For swim-to-crawl transition, C. elegans lacking dopaminergic neurons will exhibit truncated movement upon transitioning from a liquid to an agar medium. Similarly, animals lacking the DAT-1 receptor accumulate high amounts of endogenous dopamine, which induces a switch from the swim to crawl phenotype before causing the SWIP phenotype (McDonald et al., 2007; Vidal-Gadea et al., 2011). In C. elegans, the swim-to-crawl assay has been used to demonstrate that the membrane protein tetraspanin (TSP-17) protects dopaminergic neurons against 6-OHDA-mediated neurodegeneration and the toxicity caused by increased concentrations of endogenous intracellular dopamine (Masoudi et al., 2014).

Previous studies in C. elegans indicate that dopaminergic neurons are mechanosensory (Loer and Kenyon, 1993; Liu and Sternberg, 1995; Duerr et al., 1999; Sawin et al., 2000; Bettinger and McIntire, 2004; Hills et al., 2004; Sanyal et al., 2004; Abdelhack, 2016). Dopaminergic neurons respond to anterior touch stimulation (Sanders et al., 2013). Caenorhabditis elegans lacking tyrosine hydroxylase (cat-2 mutants) display defective food-sensing behavior because they fail to slow down when they encounter a bacterial food source. This basal slowing response is mediated by dopamine signaling and depends on physically touching the bacterial food source (Sawin et al., 2000). Such interactions between dopamine and mechanosensory touch responses are not well understood. Nevertheless, these interactions appear to be necessary for regulating foraging in nematodes (Sawin et al., 2000). This confirms a role for dopamine in modulating the response to a non-localized mechanical stimulus (such as taps) administered to the NGM plate (Sanyal et al., 2004). Animals respond to external tapping by escalating their forward or backward motion. Repeated tapping attenuates the reversal frequencies and leads to habituation (Rose and Rankin, 2001). The time required to respond to the tap can be used as a measure of dopaminergic function as the loss of dopaminergic function can alter this behavior (Sanyal et al., 2004). Although this behavior has not yet been studied in animals with Parkinson's-like symptoms, mechanosensory touch responses have been studied in C. elegans neurodegenerative models of Huntington's disease, Alzheimer's disease, and tauopathies (Parker et al., 2001; Miyasaka et al., 2005; Gordon et al., 2008). Therefore, this behavior can be used to assess healthy/impaired dopaminergic function in wild-type and PD animals using cat-2 mutants as a negative control, since these mutants habituate to tapping faster than wild-type strains (Chen et al., 2013).

Under favorable conditions, the life cycle of C. elegans includes the egg stage, four larval stages (L1-L4), and an adult stage, which is reproductive in hermaphrodites and lasts for 3–5 days. When exposed to overcrowded conditions, limited food, or chemical or physical stressors, animals enter an alternative stage after L2 known as dauer diapause (Cassada and Russell, 1975; Fielenbach and Antebi, 2008). The entry to dauer is regulated by daf-16 (forkhead box O or FOXO) and its upstream regulator daf-2 (insulin receptor), which are important modulators of aging and lifespan (Kenyon et al., 1993; Lee et al., 2001). Although this behavior occurs independent of dopamine signaling, once the animals enter the arrest phase they respond to any changes in dopamine signaling by increasing their body movement (Gaglia and Kenyon, 2009). Therefore, dauer movement assays can be used to assess this behavioral change. For example, dauer formation can be induced by exposing djr-1.2 mutants to the heavy metal Mn, transferring them to NGM plates without bacterial food, and storing for 72 h (Chen P. et al., 2015). The dauer diapause can then be determined using body movements which is defined as one complete body bend in forward or backwards direction in a 1 min duration (Gaglia and Kenyon, 2009; Chen P. et al., 2015). The cat-2 deletion mutants that have diminished DA signaling are used as a positive control. Both djr-1.2 and cat-2 mutants exhibit increased dauer movement compared with controls. When exposed to Mn, the djr-1.2 mutants show a further increase in movement compared with untreated controls, indicating reduced dopamine signaling (Chen P. et al., 2015). However, this behavior can be rescued by the overexpression of DAF-16. This behavioral assay was also important for assessing the interactions between aging, a PD environmental risk factor (i.e., Mn), and the PD-associated homolog DJ-1 (Chen P. et al., 2015).

Fecundity is an important assay for determining the egg-laying behavior of C. elegans and is controlled by dopamine (Schafer and Kenyon, 1995; Weinshenker et al., 1995). The exposure to environmental toxins can lead to changes in dopamine signaling, which in turn can alter fecundity or brood size in C. elegans (VanDuyn et al., 2010). Fecundity can be measured by performing progeny count assays. Age-synchronous adults are placed on individual plates each day until they cease reproducing. The number of eggs or viable progeny is then counted. When the assays are performed by counting progeny, the plates are incubated at a specific temperature and the eggs are allowed to develop for 48 h before the brood size is determined (Hodgkin and Barnes, 1991; Scerbak et al., 2016). The assessment of brood size or total progeny has been performed in neurotoxin-treated (6-OHDA and insecticide) models (Satapathy et al., 2016) and in different PD mutants (Cooper et al., 2015) of C. elegans. LRRK2 mutants have decreased fecundity due to decreased levels of DA, and this decrease cannot not be rescued by the daf-2 mutation (Yao et al., 2010b; Cooper et al., 2015). Also, a significant decrease in brood size (25–31%) occurs in animals exposed to 6-OHDA, which can be slightly increased by curcumin treatment (Satapathy et al., 2016). Overall, fecundity can be used to measure healthspan in wild-type and PD animals to assess the effects of experimental treatments on the overall pathology and behavioral phenotypes of C. elegans.

In C. elegans, defecation is a behavior controlled by a series of muscle contractions, i.e., a motor program that occurs in the intestinal “enteric” muscles of the animals. On average, it occurs every 50 s (Dal Santo et al., 1999) and this cycle remains constant at 20°C. Dopamine has been implicated in controlling the defecation cycle (Weinshenker et al., 1995; McDonald et al., 2006; Vidal-Gadea and Pierce-Shimomura, 2012). Previous studies have demonstrated that excess dopamine reduces the defecation rate by decreasing expulsion muscle contractions (Weinshenker et al., 1995). Defecation is carried out in three steps: posterior body muscle contraction (pBoc), anterior body muscle contraction (aBoc), and expulsion muscle contraction (Branicky et al., 2001; Kwan et al., 2008). The length of the defecation cycle can be determined by viewing animals with a dissecting microscope and measuring the duration between two consecutive pBoc contractions in adult animals at 20°C or as specified (Branicky et al., 2001; Cooper et al., 2015). A recent study showed slower rates of defecation in α-synuclein and LRRK2 mutants compared to normal rates in cat-2 mutants, suggesting that defecation behavior occurs independent of dopamine in these PD models (Cooper et al., 2015). However, cat-2 mutants may not completely lack dopamine (Sanyal et al., 2004). The rate of defecation should be further investigated as an indicator of physiological outcome in PD animals.

In C. elegans, locomotion or motility is a useful marker to assess healthspan (Bansal et al., 2015). The dorsal and ventral muscles coordinate to control the classical sinusoidal locomotion patterns in nematodes (Croll, 1975; Donnelly et al., 2013). Motility can be assessed in aged individuals using an A-B-C class-based system (Herndon et al., 2002). Class A represents a normal sinusoidal pattern, class B represents spontaneous reversals or induced motion with gentle prodding, and class C represents no movement or only movement of the head in response to gentle prodding. These patterns are also influenced by the presence or absence of food and exposure to mechanical or chemical stimuli (Omura et al., 2012). Studies have shown that disrupting DA signaling using genetic mutations or exposure to environmental toxins (6-OHDA or MPTP) can change the locomotory behavior of C. elegans (Ali and Rajini, 2012; Cooper et al., 2015; Liu et al., 2015). Such altered behavior can be assessed by observing changes in the typical sinusoidal pattern, including irregular body bends or thrashing behavior. Body bends are counted as one muscle contraction that leads to a complete bend of the dorsal or ventral side of the animal (Ghosh and Emmons, 2008). The term “thrashing” is used to define motility when nematodes are placed in a drop of liquid (e.g. M9 buffer), and it is determined by measuring the frequency of lateral movements or the direction of mid-body bending (Buckingham and Sattelle, 2009). Locomotory behavior can be quantified by viewing or recording worm movements through a stereomicroscope. Numerous automated programs facilitate the analysis of digitally recorded data, including Worm Tracker 2.0, OptoTracker, The Parallel Worm Tracker, Nemo, Multimodal illumination and tracking system, the Multi Worm Tracker, and CoLBeRT (Husson et al., 2013). Recently, a microfluidic device was also used to measure the locomotion of C. elegans using an electric signal (Jung et al., 2016).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.