The Caenorhabditis elegans worm Development and Activity Test (wDAT) can be used to differentiate between reversible and irreversible developmental effects

Abstract

Developmental delay and spontaneous locomotor activity changes, as well as the reversibility of these adverse effects are apical endpoints used in chemical safety evaluations. These endpoints were assessed at sublethal concentrations in C. elegans using 5-fluorouracil (5FU), hydroxyurea (HU), or ribavirin (RV), teratogens that are associated with reduced fetal growth in mammals. C. elegans develop from egg to egg-laying adult in about three days. Synchronized cohorts were exposed either continuously, or for 24 h (early-only) from first-feeding after hatching. Developmental delays were dose-responsive for all three chemicals in both exposure schemes. For 5FU and HU, developmental delays and hypoactivity levels were similar in continuous and early-only exposure groups, consistent with irreversible developmental effects. The observed hypoactivity in developing C. elegans may be related to reported 5FU-induced muscle impairment and HU-induced post-exposure effects on locomotion parameters in mammals. In contrast to 5FU- and HU-induced hypoactivity, RV was associated with a non-significant trend to slight hyperactivity in both exposure schemes. Continuous RV exposures induced delays to sequential developmental milestones that increased with exposure duration. RV-induced delays were significantly reduced but not eliminated in early-only exposure cohorts, consistent with cumulative RV effects on developmental progress. These findings suggest that C. elegans may be a useful model for detecting chemicals with irreversible, reversible, and/or cumulative effects on organismal development.

Graphical Abstract

Highlights

• •As in mammals, 5-FU and HU have irreversible effects on C. elegans development.
• •As in mammals, C. elegans locomotion parameters are altered by 5-FU and HU.
• •As in mammals, some RV development effects are partially reversible in C. elegans.
• •The C. elegans wDAT can detect reversible vs. irreversible developmental effects.
• •C. elegans data can contribute to chemical hazard identification.

1. Introduction

Public health agencies are moving to reduce, refine, and/or replace (3Rs) toxicity testing in mammals, increasing the need for the development of reliable alternative methods [60], [63]. New Approach Methodologies (NAMs) are alternative toxicity tests that use any non-vertebrate animal technology, methodology, or procedure that can provide data to inform human chemical hazard and risk assessment [17], [37]. Improvements in developmental toxicity assessment using human cell based NAM systems are progressing at a rapid rate; however, organismal growth, locomotor activity, and the reversibility of chemical effects on these processes cannot be completely modeled with current in vitro systems [1], [17], [18], [24]. These apical endpoints represent highly complex biological processes for which many aspects are well conserved across animal phyla, providing an opportunity for small model organisms (SMOs) to bridge toxicological information gaps between cell-based in vitro data and human adverse outcomes [5].

C. elegans are non-pathogenic microscopic nematodes with a 3-day life cycle and an approximately 3-week lifespan. C. elegans have a basic body plan, yet conserved aspects of the digestive system, digestive processes, and nutrient uptake, along with a relatively impermeable cuticle, make this SMO useful for oral toxicity testing [31], [83]. The timing of developmental progress from onset of feeding after hatching to egg-laying adult is well studied and highly reproducible among C. elegans control cohorts, and responsive to exposure to mammalian developmental toxicants [3], [33], [7], [73]. The majority of known signal transduction pathways involved in organismal development, from the regulation of early embryonic cell polarity and asymmetric cell division to morphogenesis to fetal- and juvenile-specific processes, have at least some conserved elements in nematodes and humans [32], [36], [42], [47], [67]. C. elegans predict chemically induced developmental toxicity in rats and rabbits nearly as well as those two mammalian species predict developmental effects in each other, supporting the model’s utility as a component of alternative developmental toxicity test batteries and integrated testing strategies [59], [6], [62].

The C. elegans genome encodes conserved genes for the biosynthesis and function of neurotransmitters acetylcholine, dopamine, and serotonin; in both nematodes and mammals, excitatory motor neurons release acetylcholine and inhibitory motor neurons release γ-aminobutyric acid (GABA) [2]. Molecules involved in neurotransmitter packaging, release, re-uptake, and postsynaptic signal cascades are also highly conserved, making the machinery of the neuromuscular junction similar between C. elegans and humans [2], [40]. These concordant neuronal features are the basis for C. elegans’ high positive detection rate for human neurotoxicants [51].

Changes in developmental measures such as growth and behavior, at any stage of development, reversible or permanent, are important components of chemical safety evaluations [16]. Growth inhibition and delayed milestone acquisition are apical effects that can reflect a variety of underlying functional deficits, and are therefore used as general endpoints that require further testing to identify causative toxic modes of action (MOA) [1], [46]. Similarly, within a toxicology context, behavior can be considered the final output of the nervous system, yet the measures of behavior or locomotion used in developmental neurotoxicity testing can reflect neurotoxicity, structural anomalies, general malaise, or other etiology, each with their own underlying causes [24]. Thus, while adverse outcome data from any whole animal model will not provide specific MOA information, the benefit of apical endpoint testing is that toxicity data is generated without prior identification of a chemical target [10].

Developmental toxicity in vitro test batteries are useful for hazard identification, and efficient at detecting human-relevant chemical bioactivity; however, cellular bioactivity does not necessarily predict apical adverse effects, and in vitro systems are limited in their ability to model chronic effects [15], [61], [74]. In contrast, SMOs with short lifespans can rapidly model adverse outcomes from acute vs. chronic exposures, suggesting the utility of supplementing in vitro data with information from a model like C. elegans with demonstrated mammalian-concordant effects for apical endpoints such as growth inhibition and altered locomotion [3], [51], [6]. What is less well studied is the suitability of C. elegans to differentiate between reversible and irreversible chemical effects on development.

The worm Development and Activity Test (wDAT) measures the time to C. elegans developmental milestone acquisition and stage-specific locomotor activity [33]. Previously, we found that three drugs known to reduce fetal weight in rodents, 5-fluorouracil, hydroxyurea, and ribavirin, cause developmental delays and altered locomotor activity in C. elegans continuously exposed for three days from first feeding [35]. The present work is a case study conducted to compare the effects of shorter-term early exposures for these three drugs to continuous exposures during the entire C. elegans post-hatching developmental period, and thereby determine if chemicals with reversible vs. irreversible effects can be distinguished in this model.

2. Materials and methods

2.1. Chemicals and reagents

Test articles and nutrient chemicals for C. elegans Habitation Reagent (CeHR) were purchased from Sigma Aldrich (Saint Louis, MO, USA). C. elegans Habitation Medium (CeHM) was a freshly prepared mixture of 20 % non-fat cows’ milk purchased at local grocery stores and 80 % CeHR. The CeHM formulation using CeHR containing cytochrome c, which allows for C. elegans to develop in nutrient medium at the same rate as C. elegans grown on E. coli, was utilized to avoid potential added xenobiotic effects of E. coli metabolism [34], [71]. Sublethal concentrations that allowed for at least semi-synchronous developmental timing within continuously exposed C. elegans cohorts were determined previously [35].

2.2. C. elegans culture maintenance and chemical exposures

The C. elegans N2 wild-type strain was obtained from the Caenorhabditis Genetics Center, a program funded by the NIH Office of Research Infrastructure Programs (P40 OD010440). C. elegans CeHM-fed cultures were maintained in vented, canted neck flasks on orbital shakers set to 60 rpm and maintained at 20 ± 1 °C in hot/cold incubators. C. elegans dauer larvae are an alternate, food-scarcity induced life-stage that does not eat or reproduce [82]. Dauers release pheromones into the medium that increase the stress resistance gene expression of other C. elegans in the same medium [22], [49], therefore only healthy, well-fed cultures without dauers were utilized for hypochlorite egg isolation (bleach prep, [34]). Age-synchronized newly hatched cohorts were generated by egg isolation followed by an 18.5 h ± 30 min incubation in non-nutrient M9 phosphate buffer [85]. C. elegans XO males arise spontaneously by chromosome non-disjunction and make up about 0.1 % of healthy control populations [27]. C. elegans males do not propagate in CeHM cultures maintained on shakers [32], therefore exposed populations were presumed to be 99.9 % XX hermaphrodites.

The worm Development and Activity Test (wDAT) uses wMicroTracker (wMT) infrared activity tracking devices from PhylumTech (Santa Fe, Argentina) to monitor locomotion [33], [35]. For continuously exposed (wDAT-c) C. elegans, age-synchronized cohorts were exposed throughout the developmental period and into adulthood (Fig. 1A). First larval stage (L1) worms were counted and diluted to approximately 1000 L1s per mL CeHM, then 900 µL/well of L1s in CeHM were transferred to 12-well plates. 100 µL of water or 10x aqueous dosing solution was added per well within 30 min post-first-feeding after hatching. Each plate contained three replicate wells per condition. A randomized dosing pattern was utilized such that, over four independent experiments, each well position was utilized once and only once per condition. Plates were then transferred to wMT instruments maintained without shaking in 20 °C incubators for four days of monitoring.

Fig. 1.

The worm Development and Activity Test (wDAT). (A) Feeding after hatching marked time zero (feed) of post-hatching development. The midpoints for the four larval stages are marked L1-L4. Early-only exposures (wDAT-e) were for the first 24 h of post-hatching development, followed by brief washing, re-feeding, and a return to trackers for post-exposure evaluation. Continuous (wDAT-c) exposures began within 30 min of first feeding after hatching and continued for the duration of the assessment. (B) An example of activity curves from an individual 5-fluorouracil (5FU) wDAT-c experiment, with time in hours on the x-axis and population locomotor activity as detected by infrared beam interruptions on the y-axis. Activity in control wells is indicated by the dashed blue line. The inactive lethargus period between the L1 and L2 stages in controls is indicated with an up arrow at 19 h post initiation of feeding after hatching. At the highest tested 5FU concentration in this experiment, synchronous milestone acquisition was lost after the second larval stage (bracket). (C) With early-only wDAT-e exposures, L2 activity curves were interrupted by the washing procedure (wash & re-feed). The difference between the height of the control L4 activity peak (stripped blue star) and the 3 µg/mL 5FU L4 activity peak (solid gray star) indicates the change (δ) in activity at that stage. The difference in hours between the peaks indicates the L4 delay for 3 µg/mL 5FU exposure group.

For early-only exposures (wDAT-e), the following alterations to the wDAT-c were made: after 24.5 h + /- 30 min exposures at approximately 2700 worms/mL, C. elegans cohorts were quickly washed, for a total of no more than 10 min without food. The washing procedure was done in 15 mL conical tubes and consisted of diluting each 1 mL of growing worms in CeHM + /- test article in 14 mL water followed by centrifugation at 800 x g for 1 min in an Eppendorf (Hamburg, Germany) 5810 R centrifuge. Wash supernatants were then removed by pipetting rather than aspiration to facilitate containment and proper disposal of hazardous chemicals, leaving approximately 200 µL of worms in a 15-fold water dilution of the original exposure. 14.8 mL M9 buffer was then added for a further 75-fold dilution, and the centrifugation and removal of supernatant was repeated, again leaving approximately 200 µL of washed worms in the tube. This was resuspended in 2.8 mL of fresh CeHM for a further 15-fold dilution, dispensed in 1 mL aliquots to wells of a fresh 12-well plate, and returned to wMTs for monitoring for an additional 3 days. For a 20 µg/mL exposure, this washing procedure would reduce the residual test article exposure from the medium to 1.2 ng/mL. A wDAT worksheet and checklist for the continuous and early-only exposure schemes, and the raw wMT data output files for all experiments, are available in Supplementary Materials.

2.3. Data analysis

Test articles were water soluble, and each test plate contained water controls that were condition-matched for wDAT-c or wDAT-e handling. The difference between controls in plates run side-by-side, labeled ‘0 L-R’ in figures, was used to determine the minimum meaningful change, 4 % for delay, and 10 % for spontaneous locomotor activity [35]. Four independent experiments were conducted per chemical, each consisting of a unique preparation of dosing solutions and an independently prepared C. elegans L1 cohort. For each wDAT experiment, data from the three replicate wells were averaged within the PhylumTech wMT™ software [68]. Activity peak midpoint detection was determined from wMT motion detection curves smoothed using the mean of seven consecutive readings for each timepoint. Data was normalized to in-plate control values and presented in figures as the relative time to peak for each stage (delay) with the formula [(time to exposed peak – time to control peak) ÷ (time to control peak)], and the relative peak height (population locomotor activity) with the formula [(height of exposed peak – height of control peak) ÷ (height of control peak)]. Statistical significance was determined using the Student’s t-test.

3. Results

3.1. Control developmental stage acquisition timing

An outline of control developmental stage timing and chemical exposures with the worm Development and Activity Test (wDAT) is provided in Fig. 1A. Continuous exposures (wDAT-c) began within 30 min of first feeding after hatching and continued throughout the assessment period. Early-only exposures (wDAT-e) were for the first 24 h after first feeding, followed by washing, re-feeding, and return to trackers for assessment.

C. elegans develop through four larval stages prior to the adult stage, with a quiescent sleep-like period called lethargus between each stage [69]. C. elegans do not eat and are mostly still during lethargus, and these cyclical changes in behavior allow for the sensitive detection of developmental stage timing and duration [33], [52]. At 20 °C in C. elegans Habitation Medium (CeHM), the control lethargus period between the first (L1) and the second larval stages occurred at about 19 h (arrow, Fig. 1B). Combining the water control data from all wDAT-c and wDAT-e experiments in this study, the control activity peak midpoints for the second (L2), third (L3), and fourth (L4) larval stages occurred at 26.5 ± 1 h, 38 ± 1.5 h, and 51 ± 1 h post initiation of L1 feeding, respectively.

Chemical concentration ranges that did not reduce viability and produced delays and locomotor activity changes without altering synchronous development within cohorts to the point that wDAT data become uninformative were determined previously [35]. Nevertheless, in some experiments, synchronous developmental timing within a population was lost at the highest tested concentration, as indicated by the loss of detectable peaks and valleys (bracket, Fig. 1B). For early-only (wDAT-e) exposures, which involved a 10 min washing and re-feeding procedure at 24.5 h + /-30 min, the normal L2 activity curve was disrupted (wash & re-feed, Fig. 1C), and therefore only L3 and L4 data was evaluated from wDAT-e experiments. After midpoint determination of stage-specific activity peaks (stars, Fig. 1C), the change in spontaneous locomotor activity was the difference between the control and exposed peak heights for each stage, and the delay was the difference in mid-peak time (dotted straight lines, Fig. 1C), normalized to plate-matched control values.

3.2. Effects of 5-fluorouracil

For continuous exposures of 5-fluorouracil (5FU), dose-responsive developmental delays ranged from 6 % to 45 % at 1–5 µg/mL (Fig. 2 A). Note that in two of four independent experiments, activity curves for the 4 µg/mL 5FU cohorts at L2 and the 5 µg/mL 5FU cohorts for L4 were not clear enough to determine a peak time or height (example, Fig. 1B) and were therefore marked with an ‘2/4’ in Fig. 2. Mean delays with early-only 5FU exposures were not significantly different from continuous exposures (Fig. 2B), indicating that delays to later juvenile developmental milestones were not reversed or significantly reduced after removal of 5FU. Stage-specific spontaneous locomotor activity levels decreased with increasing 5FU exposures (Fig. 2 C). At lower tested 5FU continuous exposures, there was some locomotor adaptation with increasing maturity, as indicated by significant L2 hypoactivity but closer-to-control L4 activity for the 1 and 2 µg/mL 5FU exposure cohorts (brackets, Fig. 2 C). However, at 5 µg/mL 5FU, activity levels were about 50 % of controls at L2, L3, and L4, indicating no adaptation over time at the highest tested exposure. At 5 µg/mL 5FU, and at L3 only, there was a modest but statistically significant difference in hypoactivity levels between the continuous and early-only exposure groups, with a 54 % reduction in spontaneous locomotor activity relative to controls with continuous exposure, and a 33 % reduction with early-only exposures (# symbol, Fig. 2D). As this change was seen at only one exposure concentration and at one developmental stage, its significance is unclear.

Fig. 2.

5-fluorouracil data from four independent wDAT experiments. Solid bars indicate continuous exposures (wDAT-c), and stripped bars indicate early-only (wDAT-e) 24 h exposures from first feeding after hatching. ‘0 L-R’ is the difference between the water controls in simultaneously run plates. Error bars show the standard error of the mean, and Student’s t-test p-values of ≤ 0.0500 are indicated by * for difference from control, a bracket for difference from an earlier stage, and # for a significant difference between continuous and early-only exposure groups. The 2/4 symbol indicates that developmental timing was not synchronous enough within the cohort to detect activity peaks in two of four experiments for the indicated stages and concentrations.

3.3. Effects of hydroxyurea

Across a range of continuous exposure concentrations, hydroxyurea (HU) had a greater effect on delays to L2 than to L3 or L4 (brackets, Fig. 3A), suggesting greater harm to early-stage C. elegans, and/or adaptation with time or maturity. With removal of HU, the early-only exposure cohorts’ developmental delays to L3 and L4 were dose-responsive (Fig. 3B), and not significantly different from continuous exposures at the same concentrations, consistent with an irreversible effect on developmental timing to later, post-exposure juvenile stages. Note that a power failure resulted in loss of L2 tracking data in a single continuous HU exposure experiment, therefore HU L2 results are based on only three out of four independent experiments (‘3/4’ symbols, Fig. 3), resulting in Student’s t-test p-values not quite reaching the ≤ 0.0500 threshold despite sizeable differences in continuous-exposure hypoactivity between L2 and L4 at 60–80 µg/mL HU (Fig. 3C). At 80 µg/mL HU, and for L4 only, there was a statistically significant change in hypoactivity levels with the wDAT-e relative to the wDAT-c (# symbol, Fig. 3D), with a continuous exposure mean locomotor activity reduction of 14 % below controls, and an early-only exposure reduction of 26 %. As this difference between continuous and early-only exposure groups occurred only at a single stage and concentration, it may not be biologically meaningful.

Fig. 3.

Hydroxyurea data from four independent wDAT experiments. Solid bars indicate continuous exposures (wDAT-c), and stripped bars indicate early-only (wDAT-e) 24 h exposures from first feeding after hatching. ‘0 L-R’ is the difference between the water controls in simultaneously run plates. Error bars show the standard error of the mean, and Student’s t-test p-values of ≤ 0.0500 are indicated by * for difference from control, a bracket for difference from an earlier stage, and # for a significant difference between continuous and early-only exposure groups. The 3/4 symbol indicates that L2 data was lost due to a power failure in a single experiment, and therefore the L2 data shown is the mean for three out of four total experiments.

3.4. Effects of ribavirin

With continuous ribavirin (RV) exposures, inhibition of developmental progress increased over time, as indicated by increasing delays as C. elegans progressed through the developmental stages (brackets, Fig. 4A). Consistent with cumulative RV effects on developmental timing, removal of RV after 24 h of exposure resulted in significant reductions, but not elimination, of delays to L3 and L4 (# symbols, Fig. 4B).

Fig. 4.

Ribavirin data from four independent wDAT experiments. Solid bars indicate continuous exposures (wDAT-c), and stripped bars indicate early-only (wDAT-e) 24 h exposures from first feeding after hatching. ‘0 L-R’ is the difference between the water controls in simultaneously run plates. Error bars show the standard error of the mean, and Student’s t-test p-values of ≤ 0.0500 are indicated by * for difference from control, brackets for difference from an earlier stage, and # for a significant difference between continuous and early-only exposure groups.

In contrast to the dose-responsive hypoactivity induced by 5FU and HU, there was a non-significant trend towards mild hyperactivity at the L3 stage with both continuous and early-only RV exposures at 2.5–10 µg/mL (Fig. 4C&D).

4. Discussion

Developmental toxicity refers to an environmental insult that results in any reversible or irreversible structural or functional alteration that interferes with normal growth, differentiation, or behavior [43]. Screening with human cell based in vitro models has demonstrated potential to detect chemical effects on specific modes of action (MOA) and pathways related to specific human adverse outcomes [65], [9]. However, the majority of non-drug chemicals interact with biological systems in a non-selective manner, and many individual pathways critical to organismal development are activated by a wide variety of chemical classes [70], [76], [77]. The benefit of apical endpoint toxicity testing in living animals is that chemical effects on a multitude of complex processes involved in organismal development and function are covered, while the drawback is that MOA are not identified [10], [24]. Assessments of growth and behavior, including tests that evaluate non-cognitive behavioral endpoints such as spontaneous locomotor activity, are critical components of developmental toxicity testing and chemical hazard assessment [1], [54]. The addition of data from small model organisms such as C. elegans to in vitro battery testing can provide MOA-agnostic coverage of adverse effects on different life stages and complex biological processes, providing a guide to reduction and refinement of follow-up testing in laboratory mammals [1], [44], [51], [64].

5-fluorouracil (5FU) is a chemotherapy drug with cytotoxic, reproductive, developmental, and teratogenic effects [57]. Maternal 5FU exposures reduce fetal weight in mice and rats, and in C. elegans continuous exposure from first feeding after hatching induces delays to developmental milestones at 2–5 µg/mL 5FU in nutrient medium [30], [35], [45]. Here we demonstrate that the dose-responsive delays seen with continuous 5FU exposures over the 3-day C. elegans post-hatching developmental period were not significantly reduced with 24 h exposures followed by washing and refeeding (Fig. 2 A&B). This irreversible and dose-responsive 5FU effect on C. elegans developmental progress is consistent with rodent data showing that a single maternal 5FU exposure during gestation reduces later fetal and birth weights in a dose response manner [30], [45], [66].

Continuous 5FU exposures were also associated with significant dose-response reductions in spontaneous locomotor activity at L3 and L4, and these later juvenile hypoactivity effects were only slightly reduced in the early-only exposure groups (Fig. 2 C&D). At 5 µg/mL continuous 5FU, the highest tested concentration, spontaneous locomotor activity was reduced by about 50 % at all evaluated timepoints. At lower, but still toxic continuous 5FU exposures, there were greater reductions in locomotor activity at L2 than at L4, suggesting either adaptation over time with chronic exposures or compensatory mechanisms increasing with developmental maturity (Fig. 2 C).

Fatigue and reduced physical activity are frequent side effects associated chemotherapeutics including 5FU [38], [58]. While C. elegans does not have an equivalent to the blood brain barrier (BBB), 5FU can cross the BBB in humans and cause injury to cells of the central nervous system [81]. In mice, multiple studies have shown that 5FU reduces voluntary physical activity [14], [50], [75], indicating that hypoactivity may be a concordant effect of 5FU in C. elegans and mammals. Additionally, 5FU impairs skeletal muscle repair and remodeling in mice and decreases muscle fiber size in rats [11], [78], suggesting that adverse effects on muscle structure and function could also be responsible for the observed 5FU-induced hypoactivity in developing C. elegans.

Hydroxyurea (HU) is an antiproliferative drug that inhibits DNA replication and induces cell cycle arrest. Prenatal HU exposure in rats is associated with postnatal reductions in weight, activity index, and rearing frequency, as well as delays in the development of forward locomotion [21], [79], [80]. Thus, in mammals, fetal HU exposures induce both persistent growth retardation and reduced values for postnatal locomotion parameters [48], [53]. This study demonstrates that HU-induced developmental delays and hypoactivity were similar in the continuous and early-only exposure groups (Fig. 3A&B), indicating that, as in mammals, these adverse developmental effects of early HU exposure in C. elegans are irreversible.

Ribavirin (RV) is an antiviral drug and known mammalian teratogen. Maternal oral RV exposures result in fetal malformations and reduced fetal weights in rodents [20], [39], [41]. RV reversibly inhibits DNA synthesis in mouse embryos, and the teratogenic effects of RV depend on cell types that are most actively dividing at the time of exposure [41]. In adult mice, RV-induced anemia is dose dependent, cumulative, and reversible, while RV effects on immune cells are at least partly reversible, though recovery is lost for some cell types at higher exposures [26]. In adult rats, RV has reversible effects on male reproductive parameters, and partly reversible effects on the integrity of bone marrow DNA [55], [56]. A developmental study in ferrets demonstrated that inhaled RV induced cumulative reductions on body weight in suckling kits, and limited recovery was observed post-exposure [28]. Together these data indicate that some RV-induced effects in mammals are cumulative and partially reversible, though recovery is dependent on the exposure timing, duration, and concentration, as well as the endpoint evaluated.

The effects of RV on C. elegans developmental delays increased with continuous exposure duration, with significant cumulative effects on delay to successive developmental stages (Fig. 4A). In early-only RV exposures, delays to post-exposure milestones were significantly reduced but still present and dose-responsive (Fig. 4B), consistent with a partially reversible RV developmental effect. This finding is in stark contrast to an earlier wDAT study with cannabidiol (CBD), that found increased delays to the third larval stage with early-only CBD relative to continuous exposures, consistent with a withdrawal effect after removal of a molecule that can activate signaling cascades involved in development [8]. Together these data suggest that a 4-day assay in C. elegans can model chronic and cumulative developmental exposure effects and differentiate them from acute, irreversible, and withdrawal-like effects.

In mammals, reduced fetal body weight and locomotor changes are general toxicity endpoints that can reflect other, more specific developmental adverse outcomes such as teratogenesis, neurotoxicity, organ targeted effects, and/or neonatal complications [1], [46]. The irreversible 5FU- and HU-induced effects on developmental timing and locomotion detected with the modified wDAT method presented here are consistent with irreversible effects on C. elegans morphology induced by 5FU- and HU [32]. Similarly, the significant reductions in the incidence of adult dysmorphic phenotypes seen in C. elegans exposed for only the first 24 h of post-hatching development relative to those continuously exposed to RV [32] are consistent with the partly reversible RV-induced developmental delays shown here. A key difference between these two types of assessments is that the wDAT is a rapid and semi-automated assay, while morphology analyses are complex and time-consuming, and not yet compatible with currently available AI solutions. Together, these data indicate that, as in mammals, C. elegans larval growth and developmental locomotor activity levels can act as screening endpoints that may reflect other conserved developmental adverse effects. Additionally, the large number of C. elegans wild-type, mutant, and transgenic strains available at low cost from the Caenorhabditis Genetics Center [13], [23] suggest that the wDAT could be used to evaluate chemical effects on development in a wide variety of different C. elegans disease and sensitive-population models.

Data from rapid alternative assays with high positive predictivity for adverse effects are useful for flagging chemicals for further study but are not sufficient on their own for determining safety in humans. The Ames bacterial reverse mutation test is an example of a non-vertebrate technology that has been used for decades within regulatory contexts for hazard identification and to direct targeted follow-up testing. The Ames test utilizes bacteria to identify chemicals that induce mutations in DNA and is used as a screen for genotoxic activity [19]. In the United States and around the world, the Ames test is a requirement for regulatory approval of many types of chemicals [84]. While the Ames test has a high positive predictivity for genotoxicity, some genotoxic chemicals are false negatives, likely due to bacterial differences from humans in factors such as uptake, metabolic activation, chromosome structure and DNA repair processes [19], [84]. A negative Ames test result does not rule out genotoxicity in humans, and carcinogens that act by nongenotoxic mechanisms are not detected by the test, yet “it is still important to perform the bacterial reverse mutation test” due to its high positive predictivity within a limited context-of-use [19]. The Ames test example suggests how alternative toxicity test models with high positive detection rates for specific types of adverse outcomes can be used within regulatory frameworks as indicators of chemical hazard rather than as definitive evaluations of safety. Bacteria are phylogenetically much more distant from mammals than C. elegans. Unlike bacteria, C. elegans utilize conserved eukaryotic machinery for DNA packaging, epigenetic modification, replication, repair, and transcription, as well as mRNA modification and translation [29], [4], [72]. C. elegans assays have high positive predictivity for mammalian developmental toxicants and neurotoxicants [12], [25], [3], [33], [51], [6], indicating that, like the Ames test, C. elegans data could be used to help direct further, more targeted testing in laboratory mammals, but for a wider variety of endpoints.

5. Conclusion

5-fluorouracil (5FU) irreversibly inhibits juvenile development in C. elegans and fetal growth in rodents. 5FU-induced hypoactivity in C. elegans may be related to reported 5FU-induced fatigue in humans, reduced voluntary physical activity in mice, and/or impairment of skeletal muscle in mice and rats. In both mammals and C. elegans, developmental hydroxyurea exposure has irreversible effects on growth, and early exposures have post-exposure effects on locomotion parameters. In both mammals and C. elegans, developmental ribavirin exposures induce growth retardation. The cumulative and partly reversible RV effects on C. elegans developmental delays are consistent with reported partly reversible RV effects on perinatal weight gain in ferrets. These concordant findings indicate that C. elegans may be a useful model for the rapid detection and differentiation of chemicals with reversible and irreversible adverse effects on organismal development. Further testing with larger and more diverse chemical panels will indicate how this model can contribute to chemical hazard assessment within regulatory frameworks.

Abbreviations

5FU

5-fluorouracil

CeHM

C. elegans Habitation Medium, a mixture of 80 % CeHR and 20 % non-fat cows’ milk

CeHR

C. elegans Habitation Reagent, a chemically defined nutrient medium

HU

hydroxyurea

L1

first larval stage

L2

second larval stage

L3

third larval stage

L4

fourth larval stage

MOA

mode of action

RV

ribavirin

SMO

small model organism

wDAT

worm Development and Activity Test

wDAT-c

wDAT continuous exposures from first feeding after hatching

wDAT-e

wDAT 24 h exposures from first feeding after hatching (early-only)

Funding

The work represented in this manuscript was conducted as part of the authors’ employment. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author statement

• •All authors consent to the publication of this updated version of the manuscript.
• •All authors declare that they have no conflicts of interest, actual or perceived.
• •The funding for this study and the authors’ contributions to this manuscript were entirely internal to the U.S. FDA.
• •C. elegans studies are not covered by IACUC regulations.
• •CRediT authorship contribution statements are accurate.
• •Suzanne Fitzpatrick agreed to be acknowledged in this publication for her comments that influenced the study design.
• •No generative AI or AI-assisted technologies were utilized beyond internet searches for supporting data and appropriate references.
• •All raw data used in this study is provided as supplemental material.

CRediT authorship contribution statement

Nicholas Olejnik: Resources. Jeffrey Yourick: Writing – review & editing. Piper Reid Hunt: Writing – review & editing, Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Conceptualization. Robert L. Sprando: Writing – review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary material

Supplementary material

Supplementary material

Supplementary material

Supplementary material

Data availability

all raw data is in the supplemental files