Introduction

Parkinson’s disease (PD) is a common neurodegenerative syndrome with complex clinical features1characterized by a loss of dopaminergic neurons2. PD is also a progressive neurodegenerative disorder that is characterized by the dopamine deficiency3. To date, several species including the mouse and Drosophila have been studied as animal models of PD4.

In insects, a Drosophila model with essential features of human PD has been established; the mutant forms produce adult-onset degeneration of dopaminergic neurons and locomotor dysfunction5. In the Drosophila model, crawling speed is also known to be increased by the action of L-3,4-dihydroxyphenylalanine (L-dopa)6. In model flies of PD, the flies exhibit dopamine deficiency in the brain accompanied by abnormalities in locomotor activity7, like that of humans. In recent years, the relationship between human disease genes, including PD, has been studied in the Coleopteran beetle species Tribolium castaneum8,9,10 whose entire genome has been sequenced11. We believe that the search for insect groups that can serve as models for PD outside of Drosophila species is important for the future treatment of this disease.

However, to our knowledge, there is no study that establishes a relationship between PD and behavioral abnormalities of T. castaneum in the human genetic disease database from the phenotype to the molecular level. Since PD is a disorder involving abnormal gait mediated by abnormal dopamine expression, a study comparing locomotor activity in dopamine-deficient individuals among three species, mice, nematodes, and T. castaneum was recently published12,13. In T. castaneum, the strains artificially selected for longer (L-strain) and shorter (S-strain) durations of tonic immobility, or death feigning, were established14,15. Maekawa et al.12 found that experiments in which strains were selected for the duration of tonic immobility, strains selected for longer duration of tonic immobility (L strain) showed more abnormal walking behavior. Also, the L-strain had reduced dopamine expression in the brain compared to strains that did not show immobility16, including in the tyrosine metabolic system, decreased expression of dopamine-related genes17, and amplification of mutations in these genes18. This set of features suggests a similarity in the L-strain of T. castaneum to Parkinson’s syndrome in humans. It has also been reported that injecting dopamine or feeding caffeine to adults of the L-strain shortens the immobility duration19. Ecologically, tonic immobility has been elucidated to be adaptive as a predation-avoidance behavior14,15, and its physiological and ecological significance has recently received attention20,21,22.

Adults of the L-strains in T. castaneum express lower levels of dopamine in the brain compared to the S strains16 and thus have lower walking activity than S strains16,23. Adults of the L-strain in T. castaneum can avoid predation by pretending to be dead when attacked by their natural enemy Adanson’s house jumper (jumping spider) Hasarius adansoni14,15. Against another natural enemy the Assassin bug Amphibolus venator, adults of the L-strain with slow movement are harder to find and less likely to be eaten than the well-moving S strain23. Therefore, the L-strain of T. castaneum with dopamine deficiency, which frequently exhibits death-feigning behavior, is less susceptible to attacks by natural enemies due to its slow movement. However, it should be noted that while this behavior is defensive against predators, the underlying dopamine deficiency and associated movement abnormalities in the L-strain parallel key features of human PD, supporting its potential as a model for studying dopamine-related motor control disorders.

A previous molecular study identified 518 differentially expressed genes (DEGs) between the L and S strains by transcriptome analysis, such as RNA sequencing (RNA-seq), which is rapidly gaining momentum in an effort to reveal the molecular mechanisms underlying the physiological mechanisms17. The study revealed tyrosine metabolic pathways including dopamine-related genes, stress-responsive genes, and insulin signaling pathways that differed between L and S strains The study revealed tyrosine metabolic pathways including dopamine-related genes, stress-rimmobility, or death feigning, .esponsive genes, and insulin signaling pathways that differed between L and S strains16. Subsequent analyses of gene variants in the strains using DNA resequencing have shown that the duration of tonic immobility involves many gene pathways, including caffeine metabolism, tyrosine metabolism, tryptophan metabolism, metabolism of xenobiotics by cytochrome P450, longevity-regulating pathways, and circadian rhythm18. These previous studies suggest that many metabolic pathways and related genes may be involved in the decision-making process of anti-predator animal behavior by forming a network in addition to the tyrosine metabolic system, including dopamine17. The association of Tribolium castaneum with PD was established from studies in which the herbicide paraquat was added to the diet of adult T. castaneum, elicited PD symptoms10 but no studies have explored the relationship between tonic immobility and genes affecting dopaminergic pathways on a genomic level. We therefore investigated the relationship between tonic immobility and genes affecting dopaminergic pathways in Tribolium castaneum.

In the present study, we hypothesized that the L-strain selected for duration of tonic immobility would show more variation in the homologous human PD genes than the S-strains. To substantiate our hypothesis, we performed comprehensive and comparative gene expression analyses of T. castaneum L-and S-strains using DNA resequencing.

Materials and methods

1. Insects

The red flour beetle, T. castaneum(Herbst 1797), is a stored-product insect found worldwide and a model genome species. The protocol of artificial selection for the duration of tonic immobility was described in Miyatake et al14.Briefly, the duration of death feigning was measured in 100 male and 100 female adult beetles that were randomly selected. From the population, 10 males and 10 females with the shortest and longest durations of tonic immobility were allowed to reproduce for the next generation. The selection regime was continued for more than 20 generations15,16.

2. DNA extraction

Female individuals of L or S strains were frozen in liquid nitrogen. Head and thoracic tissues without legs were removed from the frozen bodies using a pair of fine spring scissors. Each tissue was minced with the scissors and then treated with ISOGEN (Nippongene, Tokyo, Japan) according to the manufacturer’s instructions. Therefore, the DNA contains not only the genes in the brain, but also in other tissues such as the muscles, thoracic ganglion and salivary glands in the thorax. The quality and quantity of the extracted DNA were determined at 230, 260, and 280 nm using a spectrophotometer (Nanodrop™ 2000, Thermo Fisher Scientific, Waltham, MA, USA).

3. Library construction and sequencing

A total of 1–10 ng genomic DNA was fragmented by shearing to an average size of 300 bp using an Adaptive Focused Acoustics sonicator (Covaris, Woburn, MA, USA). After purification, the paired-end DNA library was constructed using a KAPA Hyper Prep kit (KAPA Biosystems, Wilmington, MA, USA). The fragmented DNA was end-repaired, dA-tailed, and ligated with the paired-end adapter according to the manufacturer’s instructions. The adapter-ligated DNA was amplified by 14 cycles of high-fidelity polymerase chain reaction (PCR). Library quality and concentration were assessed using an Agilent Bioanalyzer 2100 (Agilent Technologies, Waldbronn, Germany) and an Agilent DNA 1000 kit. In addition, the library concentration was precisely determined using the KAPA Library Quantification Kit (KAPA Biosystems). The paired-end libraries were sequenced for 200 cycles (2 × 100 bp) using the HiSeq 2500 sequencer (Illumina, San Diego, CA, USA). Reads in FASTQ format were generated using the conversion software bcl2fastq2 (Illumina, version 2.18, https://support.illumina.com/sequencing/sequencing_software/bcl2fastq-conversion-software/downloads.html). We submitted the read data to the Read Archive of DDBJ (accession number DRA011837).

4. Read mapping to the reference whole genome

A series of data analyses were performed using CLC Genomics Workbench version 12 (Qiagen, Hilden, Germany, https://digitalinsights.qiagen.com/downloads/product-downloads/). After adapter trimming and quality filtering, the clean read data were mapped to the reference genome of T. castaneum (Tcas5.2) that was obtained from the NCBI genome database (https://www.ncbi.nlm.nih.gov/). The mapping parameters were as follows: mismatch cost = 2, insertion cost = 3, deletion cost = 3, length fraction = 0.9, and similarity fraction = 0.9. After local realignment of the mapped reads, duplicate PCR reads were discarded.

5. Small variant detection

Variant calling based on single nucleotide variation (SNV), insertion and deletion (InDel), multi-nucleotide variants (MNV), and replacement was performed using the CLC Genomics Workbench built-in “Fixed Ploidy Variant Detection” tool. The calling parameters were as follows: ploidy = 2, required variant probability = 90.0, minimum coverage = 10, minimum frequency = 20, minimum central quality = 40, and minimum neighborhood quality = 30. Furthermore, high-quality variants were selected using QUAL = 80. Genes with non-synonymous substitutions in the selected variants were enriched as Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) ontology (KO) terms using the web-based tool “DAVID 6.8” (https://david.ncifcrf.gov)24. The enriched terms were statistically analyzed using the modified Fisher’s exact test (p < 0.05) contained in the tool.

6. Structural variant detection

Variant calling based on copy number variation (CNV) was performed using CNVnator version 0.3.3 (https://github.com/abyzovlab/CNVnator)25. The calling parameters were as follows: size ≥ 4000, normalized RD ≥ 2 or ≤ 0.5, and E-value by t-test statistics < 0.05. In comparison, variant calling based on large-scale InDel and presence/absence variation (PAV) was run using the CLC Genomics Workbench built-in tool “InDel and Structural Variants”. The genes contained in the identified regions were enriched as GO and KO terms using the web-based tool “DAVID 6.8”. The enriched terms were statistically analyzed using the modified Fisher’s exact test (p < 0.05) contained in the tool.

7. Mutation analysis for Parkinson’s disease-related genes

Since we could not find a database specialized for PD-related genes across different genera, we searched for PD-related genes in the OMIM human genetic disease database (https://www.omim.org/) and the KEGG database for human PD (hsa05012, https://www.kegg.jp/entry/hsa05012). The acquired information must be associated with T. castaneum because it is of human origin. To map Parkinson-related genes from humans and T. castaneum, the corresponding amino acid sequence were obtained from the acquired information. These were homologized to the entire amino acid sequence of T. castaneum by PSI-BLAST (protein-protein homology search for distantly related proteins, https://ftp.ncbi.nlm.nih.gov/blast/executables/blast+/LATEST/). We made them correspond to the top hits with E-value < 1e-4 and % identity ≥ 70%. Finally, the Gene IDs encoding the corresponding T. castaneum proteins were mapped and matched against the mutation list.

8. Pathway analysis

PD-related metabolic systems were analyzed using previous RNA-seq analysis data (accession number DRA007602)17. Similar to the method described by Uchiyama et al. (2019), gene IDs of DEGs between the L- and S-strains were replaced with human homolog IDs. An Ingenuity Pathway Analysis version January 2024 (IPA, Qiagen, https://analysis.ingenuity.com/pa/installer/select) based on the human dataset of the knowledge base was used to investigate the metabolic systems involved in these genes.

Results

1. Small variant genes in dopaminergic pathways associated with parkinsonism

Based on the reference genome sequence of T. castaneum, small variations were detected in 6,734 regions in the L-strain and 1,476 regions in the S-strain (Supplementary Data). Here, a ‘region’ refers to a genomic locus containing SNVs or InDels within or adjacent to annotated genes. Each region represents a gene or genomic segment with at least one variant meeting our quality criteria. Of the genes containing these variations, 12 genes showed a strong homology with human genes associated with PD and dopaminergic pathways (Table 1). One of these (cytochrome c oxidase subunit I: COX1) was detected in both L- and S-strains. Ten genes were found only in the L-strain, and some of these encode heat shock proteins and NADH-related enzymes, whereas one gene, heat shock protein 70 A1 was found in the S-strain. Thus, the results appear to be biased toward the L-strain: the L-strain had more hits for PD-related genes compared to the S-strain.

Table 1 Small variant genes involved in Parkinson’s disease in Tribolium castaneum strains with homologous human genes.
Full size table

2. Structural variant genes in dopaminergic pathways associated with PD

Large-scale insertions and deletions were detected in 763 regions in the L-strain and 242 regions in the S-strain (Supplementary Data). Eleven genes showed a strong homology with human genes associated with PD and dopaminergic pathways (Table 2). Ten genes, including calcium/calmodulin-dependent protein kinase type II alpha chain genes were found in the L-strain, whereas one gene, the NADH-ubiquinone oxidoreductase 49 kDa subunit was found in the S-strain. The gene encoding the NADH-ubiquinone oxidoreductase 49 kDa subunit protein had copy number variations in both L- and S-strains (Table 2).

Table 2 Structural variant genes involved in Parkinson’s disease in Tribolium castaneum strains with homologous human genes.
Full size table

3. DEGs involved in Parkinson’s disease by RNA-seq

A total of 518 DEGs (L < S: 286 genes, L > S: 232 genes) were detected between L- and S-strains by RNA-seq (Uchiyama et al., 2019). Nine DEGs for L < S expression differences and seven DEGs for L > S expression differences were matched to human genes associated with PD and dopaminergic pathways (more than 40% identity) (Table 3). In L < S expression differences, four isoforms of insulin-like receptor genes were listed as PD-related genes in humans and had synonymous SNVs in both L- and S-strains. In L > S expression differences, the tyrosine hydroxylase (TH) gene and four isoforms of protein henna genes (TH genes in human) were matched to genes associated with dopaminergic pathways in humans (Table 3). These genes contained synonymous SNVs in L- and/or S-strains. In comparative analyses with the genome of Drosophila melanogaster, four isoforms of insulin-like receptor genes of T. castaneum were matched to Drosophila Lrrk PD-related genes (Table 4). Homolog comparisons of the T. castaneum beetle with Drosophila and humans show that the strains artificially selected for longer durations of tonic immobility have both genetic mutations and variations of gene expression in PD-related genes.

Table 3 Differentially expressed genes (DEGs) involved in Parkinson’s disease in Tribolium castaneum strains with homologous human genes.
Full size table
Table 4 Differentially expressed genes (DEGs) involved in Parkinson’s disease in Tribolium castaneum strains with homologous fly genes.
Full size table

In canonical pathway analysis, the DEGs by RNA-seq were associated with monoamine signaling, including serotonin receptor signaling, adrenergic receptor signaling and cAMP-mediated signaling (Fig. 1, Tables S1 and S2). Among the top 30 canonical pathways, a total of 20 genes were associated with monoamine signaling. Of these, 17 genes appeared in multiple pathways, indicating considerable overlap. Notably, CALM1, CAMK2B, and PLCG1 were involved in more than half of the pathways, suggesting their central roles in the signaling networks. In contrast, DVL1, GSTP1, and ZNF746 were each uniquely associated with a single pathway. Insulin processing is also associated with the DEGs identified by RNA-seq. In ingenuity pathway analysis (IPA), the DEGs by RNA-seq were closely related with PD (Fig. 2 and Table S3; equivalent to PD 16 (PARK16), also see https://www.omim.org/entry/613164). Several genes were also related with PD 9 (PARK9, https://www.omim.org/entry/606693) and type 14 (equivalent to PD 14 (PARK14, https://www.omim.org/entry/612953).

Fig. 1
Fig. 1
Full size image

A diagram of the canonical pathway analysis. The top 30 pathways are shown from bottom to top on the vertical axis. The number of genes that overlap with the pathways is indicated by the size of the circles in the figure.

Fig. 2
Fig. 2
Full size image

Network of molecules associated with Parkinson’s disease. This network diagram depicts molecular interactions and expression changes related to Parkinson’s disease. The central node represents Parkinson’s disease (equivalent to Parkinson disease 16 (PARK16)), surrounded by nodes corresponding to associated genes. Node colors indicate expression levels: red for upregulation, green for downregulation, with color intensity reflecting the magnitude of change. Node shapes distinguish molecule types. Edges represent predicted molecular relationships; dashed lines indicate indirect interactions. Blue edges denote predicted inhibition, while gray edges indicate relationships with unspecified effects. Numbers in the upper left corner of each node show the number of detected isoforms or transcript variants for that molecule in the dataset. An asterisk following a gene name indicates that multiple identifiers in the dataset map to a single gene in the Global Molecular Network. Additionally, the nodes for ATP13A2 and PLA2G6 are linked to PARK9 and PARK14, respectively.

Discussion

The present study revealed that the strains artificially selected for longer durations of tonic immobility in T. castaneum have both genetic mutations and variations of gene expression in homologous human PD-associated genes. PD is a neurodegenerative disorder that is characterized by the degeneration of dopaminergic neurons. This disorder causes unintentional or uncontrollable movements, difficulty with balance and coordination, and ultimately difficulty walking26. Artificial selection for tonic immobility duration in T. castaneum may enrich for genetic variants that affect dopaminergic systems in ways analogous to those implicated in PD. Results of genomic analysis showed that the L-strain had more hits for human PD-related genes than the S-strain. Insertions, deletions, and SNVs　in heat shock protein genes and NADH-related enzyme genes were found in the L-strain (see Table 1). Heat shock proteins increase in cells in response to stimuli such as heat and oxidative stress, helping to repair and protect denatured proteins and maintain their structure. These proteins are involved in Ras-related small GTPase-mediated signal transduction pathways, and their mutations and dysfunction have been suggested to be a cause of PD27. Further, NADH is involved in mitochondrial ATP production via aerobic respiration, and mutations in this enzyme can lead to mitochondrial dysfunction. Mitochondrial dysfunction is known to cause oxidative stress and neurodegeneration in the brain, and its association with PD has been discussed28.

Large-scale insertions and deletions in calcium/calmodulin-dependent protein kinase type II alpha chain genes were also found in the L-strain (see Table 2). Calcium/calmodulin-dependent protein kinase phosphorylates α-synuclein to activate the mitochondrial calcium uniporter and thereby causes mitochondrial calcium overload and mitochondrial damage in PD29. Thus, the mutations detected in this study are involved in mitochondrial dysfunction, aerobic respiration, and ATP production, which may affect the motor control system and the duration of tonic immobility. Our previous study17 has shown that the increased expression of genes involved in dopamine synthesis and tyrosine metabolism, including the tyrosine hydroxylase (TH) gene, DOPA decarboxylase (Ddc), N-acetyltransferase (Nat), tyrosine aminotransferase (Tat), 4-hydroxyphenylpyruvate dioxygenase (Hpd) in the L-strain is strain-specific, resulting from artificial selection for longer immobility duration. Moreover, since both the L- and S-strains used in our analyses consisted of individuals within one month after eclosion, it is likely that the observed differences were not influenced by factors other than selection, such as aging.

Although PD–related genes such as LRRK2, PRKN, SNCA, PINK1, PARK7, VPS35, and DJ-1, are also present in insects, we did not detect mutations in these genes in this study. It is therefore important to note that the genes with mutations identified in this study are not the causal genes for PD itself, rather genes involved in dopaminergic pathways and motor control that are associated with dopaminergic signaling with and show a potential link to parkinsonism9,30,31,32,33.

Tyrosine hydroxylase

Tyrosine hydroxylase (TH), which is highly expressed in the L-strain, is one of the dopamine synthesis enzymes that synthesizes L-DOPA and is implicated in PD. Mutations in the TH gene detected were common between T. castaneum and humans (Table 3: T. castaneum Gene ID 654918). Four isoforms of the protein henna in T. castaneum (Table 3: Gene ID 655392) were also homologous genes of TH in humans. Henna variants were differentially expressed between L- and S- strains, and SNVs were detected. These may cause different dopamine synthesis between strains and PD-like symptoms in the L-strain. It is interesting to note that both the L- and S- strains have SNVs in TH and Henna genes. The key distinction lies not solely in the presence of these mutations, but rather in their combined effects together with differential gene expression patterns and the accumulation of additional variants in dopaminergic pathways observed specifically in the L-strain. The S-strain, despite possessing these SNVs, does not exhibit the characteristic phenotypes of prolonged tonic immobility, reduced locomotor activity, or decreased brain dopamine levels that are observed in the L-strain and are analogous to PD symptoms. This suggests that the PD-like phenotype in the L-strain results from a complex interplay of multiple genetic factors rather than single gene mutations, making the L-strain, but not the S-strain, a suitable model for studying dopamine-related motor control disorders similar to parkinsonism. IPA analyses of DEGs in RNA-seq showed a relationship between TH and PARK16. Recent reports indicate that T. castaneum shares similarities with human PARK17 and PARK5 and may serve as a good model for neurodegenerative diseases34. Regulation of TH has been linked to PD35. In humans, some patients with hereditary L-DOPA-responsive dystonia, a neurological disorder with clinical similarities to PD, have mutations in the TH gene36,37. PD is an aging-related movement disorder mainly caused by a deficiency of neurotransmitter dopamine in the striatum of the brain38. In T. castaneum, not only the expression of TH but also the expression of other enzymes involved in tyrosine and dopamine metabolism was high in the L-strain, which may lead to the active conversion of tyrosine into several metabolic pathways and the active conversion of dopamine. It is conceivable that such active conversion of tyrosine and dopamine may result in decreased dopamine levels in the L-strain. Beetles of the L-strain with longer duration of death feigning walk less than those of the S-strain with shorter duration of death feigning16. In addition, anomalies in the pattern of gait angles have been observed in L-strain beetles12,13. This would be related to the phenotypic symptoms seen in the L-strain. Therefore, an association of TH with parkinsonism is suggested in the L-strain of T. castaneum based on the characteristics described above.

Dopamine is a neuroactive substance and functions as a neurotransmitter, neuromodulator, and neurohormone in both vertebrates and invertebrates39,40. Motor control by dopamine has been reported not only in humans in connection with PD and other vertebrates41,42, but also in insects including the fruit fly D. melanogaster43,44 and the honey bee Apis mellifera45,46. In the L-strain of T. castaneum, mRNA expression was stronger for many of the genes in the tyrosine metabolic system than in the S-strain17. Canonical pathway analyses in this study showed that DEGs involved in monoamine signaling including adrenergic receptor signaling and cAMP-mediated signaling were frequently hit. Also, dopamine shortens the duration of tonic immobility19, suggesting the enhancement of motor activity by dopamine in T. castaneum. Our results showed that artificial selection for duration of tonic immobility resulted in variations of gene sequences involved in tyrosine metabolism and differences in the expression of genes encoding TH and involved in monoamine signaling. Since tyrosine is a precursor of dopamine, mutations in tyrosine metabolism and changes in expression of genes for dopamine synthesis may cause PD-like behavior as a disorder of motor activity in T. castaneum. Thus, dopamine is a common molecule linking PD in humans and T. castaneum.

Insulin-like receptor

Molecular studies47,48,49 have shown a possible functional link between insulin and dopamine, with strong evidence demonstrating the action of dopamine in pancreatic islets. In addition, insulin affects feeding and cognition through central nervous system mechanisms that are largely independent of glucose utilization47. Therapies used for the treatment of type 2 diabetes mellitus appear to be promising candidates for symptomatic and/or disease-modifying effects in neurodegenerative diseases, including PD. In contrast, the old dopamine agonist bromocriptine has been repositioned for the treatment of type 2 diabetes mellitus. Therefore, we focused on the contribution of insulin receptor signaling to PD. The common gene mutated in the T. castaneum L-strain, humans, and flies is the insulin-like receptor. Our results showed that artificial selection based on the duration of tonic immobility resulted in variations in gene sequences encoding insulin-like receptors and differences in the expression of these genes. It was also revealed that several trehalose transporter genes have variations of sequence and different expressions between strains. Since the trehalose transporter is involved in sugar transport, it is thought to be related to insulin-like receptors. Insulin-like growth factors (IGF147, and IGF248) and autophagy-related genes49 have been proposed as biomolecules of interest related to idiopathic PD. Thus, mutations in genes involved in insulin signaling and sugar transportation may cause differences in motor control in T. castaneum. This mechanism may be shared with vertebrates including humans.

Insulin signaling is a nutrition-responsive pathway in both vertebrates and invertebrates48. Molecular studies have shown a possible functional link between insulin and dopamine, as there is strong evidence demonstrating the action of dopamine in pancreatic islets, as well as the effects of insulin on feeding and cognition through central nervous system mechanisms49. In the brain, insulin is associated with the specific neurodegenerative process of PD49. The brain insulin receptors (IRs) have been identified in neurons and astrocytes from the human cerebral cortex, the rat olfactory bulb, hypothalamus, hippocampus, and cerebellum50,51,52,53, and in dopaminergic neurons of the ventral tegmental area (VTA) and substantia nigra (SN)54. Insulin can trigger two main downstream pathways: the phosphoinositide-3-kinase (PI3K/Akt) cascade, involved in neuronal survival, and the Ras-Raf-mitogen activated protein kinase (MAPK/ERK) signaling, involved in cell death55. In fact, differential expression of genes involved in the MAPK/ERK signaling cascade (RAF/MAP kinase cascade) between strains were observed (Table S1). This result may be a phenomenon occurring in conjunction with the activation of insulin signaling. Diabetes is known to increase the risk of developing PD in humans51. In insects, insulin signaling is the signaling pathway that responds to blood glucose. In T. castaneum, insulin receptor expression differs between strains, and SNVs have been detected in both strains. This suggests that insulin signaling genes may be under selection for the tonic immobility trait, which may be related to diabetes and PD. Insulin appeared to decrease dopamine concentration in the VTA via increased reuptake of dopamine through dopamine transporter (DAT), reducing conditioned place preference for high-fat food and modulating the reward-related feeding behavior54. Moreover, an insulin-mediated decrease in dopamine in the VTA suppressed the salience of food once satiety had been reached56. In D. melanogaster, the activity levels of insulin signaling in the brain affect the dopamine levels in the brains57. The relationships between insulin signaling and dopamine synthesis in the brains of T. castaneum need to be further investigated.

The present study showed that there were common genes involved in motor control between tonic immobility in T. castaneum and PD in humans. Comparisons between PD-related genes of D. melanogaster and humans have been reported5,7, but species-specific behaviors such as tonic immobility could be observed in other species58. Variations in the duration of tonic immobility may be based on different motor activities or thresholds of motor control59. These characteristics overlap with PD.

The present study also shows that the beetle strains feigning death for longer durations face dopamine deficiency in their brains, which is similar to that in the human PD. This is the first paper to show that predator avoidance strategies in animals are closely related to the parkinsonism-like phenotype via altered levels of dopamine involved in movement disorders.

The series of studies17,18 together with the present results suggest that many metabolic pathways and related genes may be involved in the decision-making process of anti-predator animal behavior by forming a network in addition to the tyrosine metabolic system, including dopamine. Considering these findings, it could also be noted that the data presented for the L- strain of T. castaneum in the present study reflect pathways affected by tonic immobility and dopaminergic signaling, which show associations with parkinsonism, but are not necessarily, specific only to PD or PD-related genes.