High-Dose Ionizing Radiation Impairs Healthy Dendrite Growth in C. elegans

Abstract

Purpose

The nervous system is vulnerable to radiation damage, and further optimization is required to increase the efficacy of radiation therapy while reducing harm to neurons. Given recent developments in heavy ion therapy, experimental models would be valuable to improve these therapies. We used the nematode Caenorhabditis elegans (C. elegans) to evaluate the effects of high-dose radiation on neuron development.

Methods and Materials

In this study, we used confocal microscopy to assess dendritic growth of the PVD nociceptor after high-dose gamma-irradiation from a Cs-137 source.

Results

Irradiation during an early larval stage (L2) delayed overall development but also independently impaired dendrite outgrowth in the PVD nociceptive neuron. Irradiation at L4 larval stage did not result in significant alterations in dendrite morphology.

Conclusions

The nematode C. elegans can serve as a high-throughput model to study the effects of high-dose radiation on dendrite growth. We propose that C. elegans can be useful for studies of experimental radiation therapy modalities and dose rates for translational research.

Introduction

The reported effects of radiation therapy on the brain are extensive. These include cerebral edema, headache, weakness, anorexia, and aggravation of preexisting deficits.^1,2 The late effects are associated with white matter damage caused by vascular injuries, demyelination, or radiation-induced necrosis. Pediatric patients are appreciably more predisposed to cognitive deficits after radiation therapy,1, 2, 3 which disrupts ongoing neurodevelopment resulting in debilitating cognitive deficits that may present years later.^4,5 It is important to investigate the effect of radiation on developing neurons to optimize radiotherapies that reduce acute healthy tissue damage and late cognitive effects.

The contribution of a neuron to overall neural circuitry is governed in part by the structure of dendritic arbors, which receive input. The geometry of dendritic arbors, which range in size and complexity, directly determines synaptic density and the extent of neuron-receptive fields, influencing the firing patterns of neural pathways throughout the nervous system. Mammalian neurons typically display complex dendritic arbors that connect higher-order functioning regions of the central nervous system. Rodent models have been used previously to analyze the effects of high-dose radiation on dendritic branching.⁶ Given significant findings that point to evolutionarily conserved genetic mechanisms across phylogeny, model organisms that offer shorter reproductive life cycles and more robust sample sizes could be especially useful for studies of radiation therapy. For example, dendrite branch geometry has been shown to be tightly regulated by adhesion and guidance receptors driven by transcription factors expressed across invertebrate and vertebrate species.7, 8, 9, 10, 11

Studies in the model organism C. elegans have identified a bilateral nociceptor, the PVD neuron, that envelops the body region and exhibits a highly branched dendritic arbor defined by a repeating menorah-like structure for mediating responses to aversive stimuli.12, 13, 14 Each menorah is rooted in a primary (1⁰) dendrite aligned with the body axis and is comprised of a single secondary (2⁰) dendrite, with terminal tertiary (3⁰), and quaternary (4⁰) branches (Fig. 1A). After embryonic development and hatching, C. elegans develops through 4 larval stages (L1, L2, L3, L4) during a 2-day period to reach the adult stage.¹⁵ PVD neurons are generated at the L2 larval stage when the single axon and 1⁰ dendrite are generated. Higher order PVD dendrites are generated by a stereotypical series of orthogonal branching events with 2⁰ and 3⁰ dendrites arising in L3 larvae and 4⁰ dendrites emerging in the L4 larval stage.16, 17, 18 Because C. elegans is transparent and the PVD neuron arises during larval development, live cell imaging of GFP labeled PVD has been used extensively to track dendrite morphogenesis.^11,17,18 Parallels between mammalian and C. elegans dendritic arbor development support the notion that C. elegans can be exploited as a paradigm to study the effects of extrinsic factors, such as high-dose radiation, on dendrite morphogenesis.¹⁹ The study of radiation on C. elegans has been carried out for several decades with ionizing radiation (IR) as the most prominent modality; however, these studies have primarily investigated radiation effects on germ line, aging, and motor behavior.²⁰ Although the C. elegans nervous system has been extensively investigated on physiological, cellular, and molecular levels, the effects of high-dose radiation on neuron development, specifically dendritic arbor development, are unknown.

Figure 1.

PVD neuron in C. elegans irradiated at larval stage 2 and imaged at larval stage 4. (A) Fluorescent image of larval stage 4 PVD neuron expressing GFP (left) and schematic (right) with labeled PVD soma, axon, and dendritic branches in 0 Gy control. (B-D) Fluorescent images showing PVD at larval stage 4 after irradiation in the larval stage 2 stage at (B) 0 Gy (sham), (C) 10 Gy, (D) 20 Gy, and (E) 40 Gy. (F) PVD dendrite branch count by dose in larval stage 4 larvae (mean ± SEM). Individual data points correspond to branch number measured in a single animal. PVD 4⁰ branches show a dose-dependent decrease. 2-way analysis of variance corrected for multiple comparisons using Sidak test. Alpha 0.05 (95% CI) P < .0001 (****). N = 6-9 per dose group.

The present study investigates the relationship between gamma-radiation dose and PVD dendrite morphology. We used a Cs-137 irradiator and confocal microscopy to show that C. elegans can be used as an efficient and low-cost model for studies of the effects of high-dose radiation on developing neurons. Thus, our findings provide an experimental model for research investigating how high-dose radiation affects neuronal development and the mechanisms that protect neurons from IR damage. Ultimately, our findings may enable future work to discriminate between the most effective and least harmful radiation therapy modalities and doses.

Methods and Materials

Nematode strains and genetics

The C. elegans strain wyIs592(pser2prom3::myrGFP; odr-1::dsRed)²¹ was used to visualize PVD for all experiments in this study. All studies in this work used the C. elegans hermaphrodite.

C. elegans age synchronization

Age-synchronized populations of L1 larvae were generated with a standard protocol. C. elegans were grown on 100 mm nematode growth medium (NGM) OP50-1 nutrient agar plates. Embryos were collected using bleach treatment of gravid adult hermaphrodites and allowed to hatch overnight into sterile M9 medium at 20 °C. This synchronized population of larvae was then plated on NGM plates at 20 °C for experiments.²²

¹³⁷Cs irradiator specifications

The Shephard Mark I Cesium-137 Irradiator Model 68 used in this study was calibrated on July 29, 2020, at 10 Gy/7.37 minutes 15 cm away from the rod source in the center of the turntable and decay corrected for the date of irradiation.

Confocal microscopy

Nematodes were immobilized with 15mM levamisole and 0.05% tricaine mix on a 2% agarose pad in M9 as previously described.²³ Images were obtained using a Nikon A1R laser-scanning confocal microscope. Z stacks were collected with 40X oil objective with NA = 1.49 and 0.5µm/step spanning the focal depth of the PVD neuron and its dendritic branches. Single-plane projections were generated with NIS-Elements Advanced Research software. C. elegans larval stages were identified by independent morphologic markers: L2 (postdeirid)¹⁵ and L4 (vulval development).²⁴

Widefield microscopy to determine larval age

Nematodes were immobilized with 15mM levamisole and 0.05% tricaine mix on a 2% agarose pad in M9 as previously described.²³ Images were obtained using a Zeiss Axioplan upright widefield microscope with 63x oil objective (NA = 1.4) with differential interference contrast (DIC) optics. Animals were left to grow at 20 °C for approximately 49 hours from hatching before scoring at L4. To account for slowed development due to radiation, irradiated animals were left to grow at 20 °C for an additional 2 hours before scoring (Fig. 2).

Figure 2.

High dose radiation (40 Gy) delays C. elegans larval development. (A) Representative images of vulval substages 4.7 to 4.9. (B) Individual data points represent vulval-substages (4.0-4.9) during larval stage 4 development. Sham (nonirradiated) controls and larvae exposed to 40 Gy at the larval stage 2 larval stage were scored for vulval morphology after 49 hours of growth at 20°C. Vulval morphogenesis was significantly retarded in 40 Gy-treated versus control samples. Unpaired t test (parametric, 2-tailed). Alpha 0.05 (95% CI), P < .0001 (****). N = 20 per dose group.

Image analysis

Fiji ImageJ (v. 2.0.0) was used to extract data from each image. The frequency of 2°, 3°, and 4° branches was counted manually, and 1° dendrite length was measured using NeuronJ (v. 1.4.3).

Statistical analyses

Student's t tests were used for all comparisons between 2 groups. Unpaired t tests for both parametric and nonparametric distributions were used when analyzing vulval morphology to distinguish between 2 groups for developmental age. Analysis of variance was used for all comparisons between 3 or more groups with a Tukey posthoc correction for multiple comparisons. P values < .05 were considered significant. Analysis was conducted using GraphPad Prism software, v8.4.2.

Results

High-dose radiation slows morphologic development of C. elegans

To detect potential effects of high-dose radiation on PVD dendrite morphogenesis, age-synchronized populations of C. elegans were split into 3 treatment groups and one nonirradiated control group. Upon reaching the mid-L2 larval stage, treatment groups were exposed to ascending doses of radiation of 10 Gy, 20 Gy, or 40 Gy (10 Gy/7.37 min), and the sham control group was not irradiated (0 Gy). After radiation exposure, all groups were maintained at 20 °C until reaching the late L4 larval stage and were subsequently imaged in the same session on the confocal microscope. Development to the L4 stage was confirmed by identifying the morphologic feature of the vulva, the egg-laying portal, that undergoes morphogenesis during this developmental period.²⁴

Our results demonstrate a significant negative correlation between radiation exposure and PVD dendritic branch count resulting in less complex PVD dendritic architecture (Fig. 1). Specifically, the extent of quaternary branching shows a dose-dependent reduction with the largest difference between sham and 40 Gy (Fig. 1F, P < .0001). The sham group exhibited an average of 99 quaternary branches, whereas the 10 Gy, 20 Gy, and 40 Gy groups presented an average of 45, 36, and 29 quaternary branches, respectively. No statistically significant differences were observed between dose groups for the secondary or tertiary branches. Summary statistics are represented in Table 1.

Table 1.

Mean (SE) PVD branch count by dose: L2rad L4image imaged in same session (not same morphologic age)

	Sham			10 Gy			20 Gy			40 Gy
	Mean	SE	N	Mean	SE	N	Mean	SE	N	Mean	SE	N
2° count	32.22	1.57	9	37.00	1.86	9	39.17	2.76	6	37.00	1.95	8
3° count	31.78	1.23	9	33.33	1.49	9	29.17	2.12	6	26.88	2.50	8
4° count	98.56	3.93	9	64.22	3.55	9	38.17	9.63	6	24.00	7.34	8

We conducted an additional test to determine whether high-dose radiation delays overall C. elegans development. Because quarternary branches emerge during the L4 larval stage,16, 17, 18 we used vulval morphogenesis which occurs during the L4 period for precise comparisons of developmental age. The vulva undergoes stereotypical morphologic changes that define a series of 10 successive substages (4.0-4.9) of L4 development.²⁴ For this experiment, one sham group was not irradiated, and another was exposed to 40 Gy at the L2 stage after ∼29 hours of growth at 20 °C. Both groups were imaged on a widefield microscope in the same session to score vulval morphogenesis after an additional 20 hours at 20 °C (ie, for a total 49 hours of growth). Our results indicate that the sham group was an average of 2.5 vulval substages older than the 40 Gy group (Fig. 2, Table 2, P < .0001). This result suggests that high-dose radiation delays overall L4 development by ∼2 hours.

Table 2.

Mean (SE) vulva sub-stage by dose: L2rad L4 imaged in same session

	Sham	40 Gy
Vulva sub-stage (mean)	4.63	4.35
Difference ± SE		–0.26+ –0.06
P value		<0.0001

High-dose radiation during development disrupts dendrite morphogenesis of the PVD nociceptive neuron

Because our results indicated that high-dose radiation impairs overall larval development, we conducted an additional radiation experiment and used vulval substages to age-match sham and irradiated groups for scoring PVD quarternary branches. As noted previously, C. elegans larvae were irradiated at the L2 larval stage with 40 Gy for comparison to a sham control group. Both 40 Gy and control worms were imaged at the vulval substage of 4.8 ± 0.1.

A significant reduction in quaternary branches was observed in the 40 Gy group compared with the sham (Fig. 3, P < .0001). The sham group had a mean of 122 quaternary branches and the worms irradiated at 40 Gy had a mean of 80 quaternary branches. No significant differences were observed among secondary or tertiary branches between the sham and 40 Gy group. Summary statistics are represented in Table 3. Taken together, our experiments indicate that irradiation impairs both overall larval development (Fig. 2) that delays PVD morphogenesis (Fig. 1) and also specifically retards outgrowth of terminal (eg, quarternary) PVD dendritic branches (Fig. 3).

Figure 3.

Irradiation during larval development impairs PVD dendritic branching. (A-B) Fluorescent images showing representative PVD neurons in larval stage 4 larvae after irradiation at the larval stage 2 stage with either (B) 0 Gy (sham) or (B) 40 Gy. (C) PVD dendrite branch count by dose (mean ± SEM). Individual data points represent the number of each type of branch in a single animal. Vulval morphology was used to age-match sham and 40 Gy treatment groups. PVD quaternary (4^o) branching is reduced at 40 Gy in comparison to control. Multiple unpaired t tests (parametric, 2-tailed). Alpha 0.05 (95% CI), ns (not significant), P < .0001 (****). N = 10 per dose group.

Table 3.

Mean (SE) PVD branch count by Dose: L2rad L4image imaged in staggered sessions (same morphologic age)

	Sham			40 Gy
	Mean	SE	N	Mean	SE	N
Vulva sub-stage	4.85	0.02	10	4.76	0.03	10
2° count	46.10	1.99	10	49.10	3.93	10
3° count	40.80	0.83	10	37.50	3.40	10
4° count	121.60	4.99	10	80.60	10.08	10

Irradiation at the L4 larval stage does not disrupt higher-order PVD dendritic branching

We investigated the sensitivity of mature PVD dendrite morphology by exposing late-stage L4 larvae to high-dose IR. C. elegans were age-synchronized and grown at 20 °C. At the L4 stage, larvae were divided into 2 groups that were either irradiated at 20 Gy or maintained as a sham (ie, nonirradiated) control. PVD neurons in each group were imaged at the adult stage.

Our results indicate that exposure of L4 larvae to 20 Gy does not result in detectable dendritic branching defects (Fig. 4). These results indicate that the most radiation therapy sensitive periods for PVD dendrite growth are early stages of development (eg, L2 larval stage) before significant higher order branching has occurred. Summary statistics are represented in Table 4.

Figure 4.

Irradiation of the mature PVD neuron at the larval stage 4 does not result in dendritic branching defects. (A-B) Fluorescent images showing representative PVD neurons in adults after irradiation with (A) 0 Gy (sham) or (B) 20 Gy at the larval stage 4. (C) PVD dendrite branch count by dose (mean ± SEM). Individual data points represent the number of each type of PVD branch (2^o, 3^o, 4^o) in a single animal. Multiple unpaired t tests (parametric, 2-tailed). Alpha 0.05 (95% CI), ns (not significant). N = 9 (sham) and N = 15 (40 Gy).

Table 4.

Mean (SE) PVD Branch Count by Dose: L4rad Adult image imaged in same sessions

	Sham			40 Gy
	Mean	SE	N	Mean	SE	N
2° count	32.22	1.57	9	35.13	1.05	15
3° count	31.78	1.23	9	34.00	0.90	15
4° count	98.56	3.93	9	105.53	1.95	15

Discussion

Here we have exploited the nervous system of C. elegans to detect deleterious effects of high-dose radiation on dendrite morphogenesis. Our findings suggest that high-dose radiation slows the overall development of C. elegans as well as specifically retarding dendritic branching of the PVD nociceptive neuron. Thus, the PVD nociceptor can serve as a useful model neuron to study the relationship between dendritic growth and high-dose radiation. The reproducibility and speed of C. elegans development provide advantages that mammalian models lack while also yielding quantitative results at much lower cost.

Radiation-induced reductions in dendritic complexity have been previously detected in mouse models. One study used GFP labeling to visualize the dendritic morphology of hippocampal neurons subjected to doses of 1 to 10 Gy gamma-irradiation during late adolescent development.⁶ Measurements at 10- and 30-days after 10 Gy irradiation detected significant reductions in dendritic branching within the granule cell layer of the hippocampus.⁶ Reduced dendritic complexity in mouse hippocampal neurons was also reported after ⁵⁶Fe-irradiation.²⁵ We suggest that the parallel effects of high-intensity IR on dendritic branching in nematodes and mammals may be indicative of shared biologic processes that are impaired by irradiation. Thus, studies investigating the mechanisms that drive PVD dendrite morphogenesis in C elegans^26,27 can provide insight into the molecular causes of IR-induced effects on dendritic branching in more complex nervous systems. Furthermore, studies of C. elegans could also explore functional endpoints, such as sensory²⁶ or motor behavior,²⁷ in conjunction with assessment of neurite formation, to gain a more comprehensive understanding of the effect of high-intensity IR.

Our findings that PVD dendritic branching is disrupted by exposure to gamma-irradiation suggest that this model could be used to study relative toxicity of experimental radiotherapies. For example, particle radiation therapy, specifically proton, carbon ion and ultrahigh-dose-rate (FLASH) therapies, are now under active investigation for their healthy tissue-sparing properties during treatment for malignancies in vital organs of the body.²⁸ Future work could use the PVD neuron to test these novel radiotherapies to identify optimal irradiation modalities that preserve dendritic morphogenesis.

Conclusion

This study provides a proof-of-principle that C. elegans can be used as a model to detect radiation impairment of neuron dendritic formation, at dose levels relevant to human central nervous system radiosurgery. Measurable branch reduction after early irradiation represents a crucial comparator for investigations of experimental radiotherapies with potentially neuron-sparing properties. Future studies can use C. elegans as a high-throughput model to optimize radiation therapy modalities and attain quantitative comparisons of neural toxicity on much faster timescales than those of mammalian studies.

Disclosures

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.

Ionizing Radiation Impairs Dendrite Growth

Appendix. Supplementary materials

Supplementary Tables

Supplementary Table 1. PVD branch count of adult animals irradiated at L4 stage (raw data, columns represent individual animals)