Skip to main content
NCBI home page
Zoological Studies logoLink to Zoological Studies
. 2023 Apr 24;62:e13. doi: 10.6620/ZS.2023.62-13

Genetic Assessment of Fertile F1 Hybrids between Two Hercules Beetles, Dynastes maya Hardy and D. grantii Horn (Scarabaeidae)

Jen-Pan Huang 1,*, Wei-Yun Chen 1, My Hanh Le 1
PMCID: PMC10201344  PMID: 37223431

Abstract

Although hybridization may complicate taxonomic practices, it can be common between animal species. Animal hybridization not only can help with generating phenotypic and species diversity in nature, but also with understanding the genetic and genomic basis of phenotypic evolution in the laboratory. We assessed the genetic composition of captive bred F1 hybrids between two Hercules beetle species using mitochondrial CO1 and nuclear loci from a double-digest restriction-site associated DNA sequencing (ddRADseq) library. We showed that the F1 hybrids were genetically clustered with samples from the maternal species, D. grantii, based on CO1 data. Nuclear genome data, on the other hand, clearly showed that the F1 individuals were genetically intermediate between D. maya, the paternal species, and D. grantii, based on a principal component analysis. Our results also revealed that sampling design may have a major impact on the inferred genetic structure and hybrid individuals using ddRADseq data sets. We discuss the importance and potential from studying the genomics of this hybrid progeny in terms of understanding the origin and maintenance of both intraspecific and interspecific phenotypic divergence and convergence.

Keywords: CO1, Dynastes, Hybridization, Hybrid F1, Genomic, ddRADseq

BACKGROUND

Hybridization and the production of viable and fertile offspring between animal species has been a subject of particular interest and debate among evolutionary biologists. The subject is contentious because it complicates the biological species concept in animal systematics (Mayr 1942; Wilson and Brown 1953; Mallet 2008 2020; Burbrink et al. 2022). Specifically, a successful hybridization between species indicates the incompleteness of hybrid sterility; thus their reproductive isolation may not be viewed as complete to some taxonomic practitioners (Wilson and Brown 1953; Huang 2020; Burbrink et al. 2022). However, beyond the taxonomic semantics (Burma 1949; Mayr 1949; Simpson 1951), hybridization between animal species could be common in nature and can even promote subsequent speciation and adaptation (Wu 2001; Nosil 2008; Stelkens et al. 2009; Hedrick 2013; Meier et al. 2017; Edelman et al. 2019). More importantly, to understand the origin of phenotypic novelty, and thus the emergence and maintenance of the diversity of life forms, hybridization between phenotypically divergent closely related species has been an essential tool to create genetic and phenotypic chimeric individuals to help associate phenotypes with genotypes (True et al. 1997; Tobler et al. 2005; Stelkens et al. 2009; Linnen et al. 2018).

The Hercules beetles (Dynastes MacLeay 1819, Coleoptera; Scarabaeidae) show remarkable phenotypic variation across species, where the variation has been hypothesized to be adaptive (particularly the body coloration and the male horn shape; Hinton and Jarman 1973; Jarman and Hinton 1974; Morón 1987; McCullough et al. 2014; Huang and Knowles 2016; Huang 2016). The adaptive phenotypic variation observed among species can even be found within a species (Morón 1987; Moctezuma and Sánchez-Huerta 2018). For example, a dark body coloration phenotype has been found to be associated with a Neotropical forest habitat in D. maya (Huang 2016), while a light color phenotype has been hypothesized to match the greyish color tree trunks in a pine-oak forest (e.g., Moctezuma and Sánchez-Huerta 2018; García-Morales et al. 2022). The repeated evolution of dark versus greyish body coloration has not only occurred across closely related species (e.g., between D. grantii and D. maya; Fig. 1), but also between populations within the same species (e.g., within D. hyllus; Morón 1987; 2009). The ability to generate hybrid progeny between species of distinct phenotypes will help with understanding the genetic basis associated with both intra and inter specific divergences (Linnen et al. 2018). We thus chose (1) D. maya with dark body coloration and a unique fan shape cephalic horn denticle on adult males (Fig. 1A) and (2) D. grantii with light body coloration as progenitors to create hybrid progeny.

Fig. 1.

Fig. 1.

Open in a new tab

The parental beetles and exemplar hybrid F1 offspring. (A) The paternal sample, a male D. maya, of the hybrid progeny. (B) Copulation between the paternal D. maya and a female D. grantii. (C) and (D) Lateral and dorsal views of a newly emerged male F1 hybrid. (E) and (F) Lateral and dorsal view of the same male F1 hybrid after feeding. The body coloration changed according to the environmental humidity and after feeding. (G) A pure-bred female D. grantii (lower-right; red arrow) and two hybrid F1 females.

Animal hybridization and the maintenance of viable and fertile progeny can be challenging because incompatibility may have, at least partially, developed during the divergence process. For example, prezygotic isolation has been shown to develop more rapidly between sympatric species than between allopatric species (reinforcement; Coyne and Orr 1989; Stelkens et al. 2009; Turelli et al. 2014). To avoid pre-and postzygotic reproductive isolation when creating hybrid progeny, the choice of allopatric parental species, or populations, may lead to a higher success rate of producing viable and fertile progeny (Jennings et al. 2014; Castillo 2017). Hence, in our hybrid progeny experiment, we selected allopatric parental samples. Specifically, we selected a male of D. maya from Chiapas, Mexico, and two wild-caught females of D. grantii from Payson, Arizona, USA. These geographic populations of Dynastes beetles do not have any other Dynastes species living in their close proximity, and thus we would not expect reinforcement to have evolved. Indeed, the beetle paired and mated immediately, and viable F1 were successfully created.

In this study, we used a mitochondrial locus and genome-wide sequence variation data to study the genetic affinity of seven adult F1 beetles that emerged in 2021 (the F1 beetle eggs were produced in 2019). The mitochondrial genome is maternally inherited in insects and the use of a mitochondrial locus can help us verify the maternal origin of the putative hybrids (Dong et al. 2021). Nuclear genetic data, especially genomic data, have been demonstrated to be effective in identifying hybrid individuals in recent studies (e.g., Malinsky et al. 2018). We did not apply the conventional approach of using microsatellites to detect hybrids because a set of microsatellite markers is currently not available for the Dynastes beetle system; restriction site-associated DNA sequencing represents a better alternative and cost-effective approach (Andrews et al. 2016). Furthermore, there are more advanced analyses to detect genetic nuances showing the evolutionary relationships among individuals (e.g., RADpainter analysis; Malinsky et al. 2018) that cannot be applied to microsatellite data. We specifically tested (1) whether the maternal genealogy of the F1 adults belongs to D. grantii, (2) whether the genomic composition of the F1 samples reveals a mixed origin between D. maya and D. grantii, and (3) how robust the analytical tools are to detect hybrids and their parental species when different sampling strategies are employed.

MATERIALS AND METHODS

Beetle Samples and DNA extractions

A captive-bred adult male of D. maya (Fig. 1A), originally from Chiapas, Mexico, and two wild-caught females of D. grantii (Fig. 1B), originally from Payson, Arizona, were used to create hybrid progeny in October, 2019. The hybrid offspring were fed with commercially available larval growing medium (Beetle Breeding Mat PRO, Fujikon). Seven adult beetles from this hybrid experiment, three males and four female emerged in June, 2021 and became active feeding and mating in July, 2021 (Fig. 1C–G). The F1 hybrid adult beetles also mated randomly with each other and successfully produced viable F2 offspring. The F2 larvae are currently at the third instar stage (kept in an Insect rearing room with a constant temperature [22°C] at Academia Sinica).

The paternal sample of D. maya was dissected with tissue samples preserved in a cryo-freezer (-80°C) in the BEECHEN lab at the Biodiversity Research Center, Academia Sinica. We failed to preserve the maternal samples because the beetles disintegrated soon after laying eggs. The seven adult F1 hybrid samples were dissected with the tissues stored in absolute ethanol in a cryo-freezer. Three voucher specimens, one male and two females, were curated in the BEECHEN lab at Academia Sinica. Four additional specimens, two males and two females, were vouchered in a personal collection (Mr. William H. Reynolds). Genomic DNAs for the paternal D. maya sample and the seven adult F1 hybrids were extracted using the DNeasy Blood & Tissue Kits (Qiagen) following the manufacturer’s instructions.

Amplifying and Sequencing Mitochondrial CO1

We amplified the DNA barcoding region of the mitochondrial cytochrome c oxidase subunit 1 (CO1) for the seven F1 hybrid individuals with the primer pair LCO1490 and HCO2198 (Folmer et al. 1994). The polymerase chain reaction (PCR) was performed in a total volume of 25 μL using PuReTaq Ready-To-Go PCR Beads (GE Healthcare), with the following PCR profile: initial denaturation at 94°C for 2 min, followed by 35 cycles of denaturation at 94°C for 20 seconds, annealing at 50°C for 40 seconds, and extension at 72°C for 45 seconds. The final extension step was set to 72°C for 10 min. We used NautiaZ Gel/PCR DNA Purification Mini Kit (Nautia Gene, Taipei, Taiwan) to clean-up PCR products. The purified PCR products were sequenced using the above-mentioned primer set by the Institute of Biomedical Sciences, Academia Sinica.

Phylogenetic Analyses Using Mitochondrial CO1

In addition to our newly obtained CO1 sequences from the seven hybrid individuals, we retrieved additional CO1 sequences from the paternal D. maya sample (Morgan et al. 2022) and samples from other closely related species (Huang and Knowles 2016). The information about the samples is listed in table 1. The CO1 sequences were aligned using MUSCLE (Edgar 2004) implemented in SEAVIEW (version 5; Gouy et al. 2021). The CO1 sequence alignment saved in phylip format was uploaded into the online PhyML web server (http://www.atgc-montpellier.fr/phyml/) for a phylogenetic analysis (Guindon et al. 2010). We selected the Smart Model Selection (SMS) in the Automatic model selection panel in PhyML without a priori estimation and specification of molecular evolutionary models before reconstructing a maximum likelihood phylogeny (Lefort et al. 2017). We further specified to use 1000 bootstrap replications for estimating branching supports.

Table 1.

Samples with CO1 used for molecular phylogenetic analysis

graphic file with name zoolstud-62-013-t001.jpg

Open in a new tab

Genomic Library Preparation and Processing

The genomic DNA from the seven F1 hybrid individuals and the paternal D. maya sample was processed into one double-digest restriction-site associated DNA sequencing (ddRADseq) genomic library (Peterson et al. 2012). We firstly digested the DNA with restriction enzymes MseI and EcoRI (New England Biolabs) and then ligated the DNA fragments of each individual with Illumina adaptors including unique 7-bp barcodes. We subsequently amplified the fragments individually using PCR with 12 cycles using the iProof High-Fidelity DNA polymerase (Bio-Rad). The amplified products were quantified individually using Qubit (ThermoFisher Scientific) and then pooled according to their DNA concentration. We used a Pippin Prep machine (Sage Science) to select the fragment size ranging from 350 to 450 bp from the pooled DNA library (cf. Huang 2016), and then the size-selected DNA library was sequenced on an Illumina MiSeq platform (150-bp single-end) at the Biodiversity Research Center Sequencing Core, Academia Sinica (Taipei, Taiwan).

We used ipyrad (version 0.9.62; Eaton and Overcast 2020) to demultiplex the raw reads (step 1) generated from the Illumina MiSeq run. Before further processing and assembling the ddRADseq loci, we retrieved 12 additional data sets from four White Hercules beetle species (table 2; Huang 2016; 2019). Specifically, six D. grantii (the maternal species of the F1 hybrids), two D. hyllus (sister species to D. grantii; Huang and Knowles 2016; Huang 2017), one D. maya (an additional sample of the paternal species), and three D. moroni (sister species to D. maya; Huang and Knowles 2016; Huang 2017) were included in this study. The retrieved individuals from the same species were from different geographic origins, except D. moroni. The ipyrad settings for filtering and assembling the loci de novo are listed here: minimum read depth for statistical base calling and majority rule base calling was set to 6, the maximum of low quality base calls allowed in a read was 5, the clustering threshold of sequencing similarity was set to 0.85, and only those loci that were present in at least 10 individuals (allowing a maximum of 50% missing data) were retained. Importantly, because most (15 out of 20) of the individuals included in the study were F1 hybrids or the parental species, we specified a high value of allowed maximum shared heterozygous sites per loci (maxSH) among individuals to 0.75. Note that we tried different settings (e.g., set the clustering thresholds to 88% and 90% or set the maxSH to 0.25 or 0.50), but the resulting inferences remained the same (see also Eaton 2014). As a result, the following analyses were based on data sets generated by our specified settings listed above.

Table 2.

Samples with ddRADseq data used in population genomic analyses

graphic file with name zoolstud-62-013-t002.jpg

Open in a new tab

Population Genomic Analyses

The genetic structure among White Hercules beetle species and the F1 hybrids was first accessed using principal component analysis (PCA). Specifically, we manually edited the .str file output from ipyrad to include grouping information (taxon assignment) and then imported the edited .str file in R for PCA using the package adegenet (Jombart 2008). Because the results from PCA may be affected by missing data, we also edited the original .str file (which allowed 50% missing data) to generate a new input file that only allowed 20% missing data via a customized R script (Huang 2018). PCAs were performed for both data sets that allowed 50% and 20% missing data. We estimated the mean frequency of the corresponding allele among samples with data and used it to substitute missing data before performing PCAs (Jombart 2008).

We further used RADpainter (Malinsky et al. 2018) to assess the genetic relationships among samples. The program first calculates SNP differences between alleles within a sample and then the SNP differences between alleles between samples. For each locus, the program then identifies the likely allele transmitted from the donor individual to the recipient individual by identifying the allele that has the least number of SNP differences. As a result, this approach uses the full information from sequence variation, instead of haplotype information (such as the input data format for PCA), available from the genomic data set. A coancestry matrix between individuals for the full data set can be generated by the program after summing the coancestry values across all loci. We used a customized python script (https://github.com/edgardomortiz/fineRADstructure-tools) to convert the .allele file generated from ipyrad processing into RADpainter input format. Specifically, we specified the python script to generate two input data sets that represented 50% and 20% missing data (minimum individuals included in the data sets = 10 and 16, respectively). Note that RADpainter has been shown to work well with 50% missing data, but for a fair comparison between PCA and RADpainter results we chose to also show results generated from similar filtering criteria. We further performed the RADpainter analyses using data sets that either excluded the immediate paternal sample of the F1 hybrids (DmaRef) or excluded the F1 hybrid individuals (allowing 50% missing data for both analyses). The rationales were to investigate the impact of including and excluding these samples on the resulting inferences of coancestry plots. The RADpainter and fineRADstructure programs were then used to produce a coancestry matrix and generate a cluster based on estimated coancestry between pairs of samples. We followed the example setting from the manual to specify the number of MCMC iterations to assign individuals to populations (http://cichlid.gurdon. cam.ac.uk/fineRADstructure.html). Results were plotted using a customized R script distributed with the RADpainter package.

RESULTS

The F1 Beetle Samples

The adult F1 individuals exhibited intermediate phenotypes between D. maya and D. grantii. However, the adult male body color could change to closely resemble D. maya after feeding (Fig. 1E and F). One concern of our study design is that the maternal samples were of wild origin, and thus the F1 individuals may not actually be the products of interspecific hybridization. Specifically, although successful mating was observed between the paternal D. maya and the maternal samples (Fig. 1B), we could not rule out the possibility that the offspring were the result of prior successful mating of the maternal samples in the wild (Simmons 2002). That is, the F1 samples can be resulted from using previously stored sperms from wild males of D. grantii by the maternal samples; the observed morphological peculiarity of the F1 individuals could still be intraspecific variation. Hence, the following genetic assessments are necessary to validate the identities of the F1 beetles.

CO1 Phylogeny

The CO1 alignment contained a total of 20 sequences with a length of 631 bp. A maximum likelihood phylogeny based on the alignment with a sample from D. tityus specified as outgroup is shown in figure 2A (lnL = -1321.48122). The phylogeny was estimated by the best-fit molecular evolution model: GTR + I + G (lnL = -1318.28186). Three highly supported (> 90% bootstrapping support) lineages were revealed from the reconstructed phylogeny: (1) D. hyllus, (2) D. grantii and the F1 hybrids, and (3) D. maya and D. moroni. Individuals from D. moroni also formed a monophyletic group with high bootstrapping support, but this D. mononi lineage was found nested within D. maya. All but one (hyb_1) hybrid individuals shared an identical CO1 sequence. All the CO1 sequences derived from the F1 hybrids represented D. grantii.

Fig. 2.

Fig. 2.

Open in a new tab

Genetic assessment of the hybrid F1 individuals and samples of their closely related taxa. (A) a maximum likelihood tree based on mitochondrial CO1 sequences. Numbers above branches indicate bootstrapping support values from 1000 replicates, where only values > 700 were shown. (B) A principal component analysis (PCA) based on 9844 unlinked SNP data (50% missing data allowed). Note that, taxa MA and GR are the parental species of the hybrids (hyb). An image showing a breeding pair of adult F1 hybrids was inserted in the PCA plot. [HY: D. hyllus. GR: D. grantii. hyb: F1 hybrid individuals. MA: D. maya. MO: D. moroni.]

Genomic Diversity and Differentiation

A total of 33,545,117 raw reads passed the initial quality control, with an average of 1,677,256 good reads per sample (SE = 210,069). Clustering analyses within samples identified an average of 154,672 clusters (SE = 17,888) with at least 5-folds of sequencing coverage (mean sequencing depth = 12 per cluster within a sample) among the samples. The total number of loci assembled across all individuals was 62,012; however, a final data set of 9,844 loci (about 15% of the total loci) was retained based on our filtering criteria (see materials and methods section for details). The average number of loci per sample after filtering was 6,203 (SE = 460).

The result from a PCA using a data set comprising 9,844 loci (50% missing data allowed) revealed well-separated clusters among species, showing different species are genetically distinct (Fig. 2B). Furthermore, the cluster of F1 hybrid individuals were located spatially between the D. grantii and D. maya clusters, supporting their hybrid origin. A PCA result using 868 loci (20% missing data allowed) yielded the same inference (result not shown). We chose to show results using the 50% missing data set across analyses for the statistical power (more loci) to unravel genetic nuances. Stringent filtering criteria (e.g., allowing 20% or no missing data) can result in some genetically identical individuals.

A total of 8,112 loci were kept after further processing and filtering using the fineRADstructure tools for a data set containing all the sequenced samples while allowing 50% missing data per locus for a RADpainter analysis. The fineRADstructure tools python script further excluded loci with ambiguous allele information and indels. The resulting coancestry plot also revealed genetic clusters according to species (Fig. 3A). However, the immediate paternal sample of the F1 hybrid (DmaRef) was clustered with the F1 hybrid offspring with an extremely high number of loci indicating coancestry, instead of with the other D. maya sample (Dma7Br). Nevertheless, DmaRef had a high number of loci showing coancestry with Dma7Br as well as with samples from the sister species D. moroni, Dhym2, Dhym3, and Dhym5. The F1 hybrid individuals also shared a high number of coancestry loci with the other D. maya sample, Dma7Br, and some samples from their maternal species, D. grantii, DGRR1, DGRR2, and DGRP1. When a smaller data set with a lower missing data percentage (allowing 20% missing data; a total of 602 loci were kept) was used for RADpainter analysis, the same pattern was revealed (Fig. 3B). When the immediate paternal sample, DmaRef, was excluded from the RADpainter analysis (allowing 50% missing data; a total of 7636 loci were retained; Fig. 3C), the F1 hybrid individuals were genetically close to samples from the maternal species D. grantii; furthermore, the F1 hybrid individuals also shared a high level of coancestry with Dma7Br, a sample from the paternal species D. maya. When the F1 hybrid individuals were excluded from the RADpainter analysis (allowing 50% missing data, a total of 6742 loci were kept), the two D. maya samples, Dma7Br and DmaRef, formed a well-supported genetic cluster. In summary, the F1 hybrid individuals shared an excessive high number of coancestry loci with their immediate father to a point that the model inferred them belonging to one genetic cluster; additionally, there were high levels of coancestry in general between the F1 hybrids and samples from D. grantii and that between the F1 hybrids and Dma7Br, the other sample from D. maya.

Fig. 3.

Fig. 3.

Open in a new tab

Coancestry plots generated by RADpainter analyses. A higher value of coancestry indicates a higher number of shared alleles found between individuals. Results from a data set of all the sequenced individuals that allowed (A) 50% missing data (8112 loci) or (B) 20% missing data (602 loci). (C) Results from a data set that excluded the immediate paternal sample of the F1 hybrids and allowed 50% (minimum individual = 10 per locus) missing data (a total of 19 individuals; 7,636 loci). (D) Results from a data set that excluded the F1 hybrids and allowed 50% (minimum individual = 7 per locus) missing data (a total of 13 individuals; 6,742 loci). F1 hybrid samples are highlighted in orange boxes; D. grantii samples, the maternal species of the F1 hybrids, are highlighted in grey boxes; D. maya samples, the paternal species of the F1 hybrids, are highlighted with brown underlines; the immediate paternal sample of the F1 hybrids, DmaRef, is indicated with an asterisk. Numbers above branches of the distance tree are bootstrapping support values.

DISCUSSION

In this study, we validated viable and fertile hybrid progeny between D. maya and D. grantii using molecular data. Specifically, we showed that the F1 hybrids (1) were clustered with those individuals from the maternal species D. grantii based on mitochondrial CO1 phylogeny, (2) were genetically intermediate between samples from D. maya and D. grantii based on genome-wide SNP data sets, and (3) shared an excessive amount of alleles with samples from both D. maya and D. grantii based on nuclear genomic sequence data. As a result, we reported that the F1 hybrids were not only phenotypically intermediate (Fig. 1C–G), but also genetically intermediate between the two parent species (Figs. 2 and 3). Our findings here indicate that the Hercules beetle system could become a suitable model system to study rapid and repeated convergent and divergent phenotypic evolution both between and within species, where we might be able to identify the genomic basis of phenotypic adaptation in the near future via association-mapping approaches (see Linnen et al. 2018).

Our study also revealed that the identification and detection of hybrid individuals may not be as straightforward as expected using ddRADseq data, especially when the studied individuals included the immediate parental sample that created the hybrid progeny (Fig. 3). Note that, with high quality reference genomes, one can apply additional methods to better test for hybrid hypotheses using heterozygosity information (e.g., Slon et al. 2018). However, currently a good quality reference genome for Dynastes beetles is still not available, and thus we could only apply the best available approach specifically designed for ddRADseq data to test for hybrids (Malinsky et al. 2018). Based on RADpainter results, sample DmaRef, the immediate paternal sample of the F1 progeny in our study, was genetically clustered with the F1 hybrids while showing excessive allele sharing with its conspecific sample, Dma7Br and samples from its sister species, D. moroni. This genetic clustering pattern remained unchanged when analyzing data sets that allowed different percentages of missing data (Fig. 3A and B). As a result, DmaRef, instead of the hybrid F1 individuals, could have been inferred as a hybrid (cf. Fig. 3 in Malinsky et al. 2018). The F1 individuals nevertheless clearly showed excessive allele sharing with Dma7Br and many samples from the maternal species, D. grantii, when sample DmaRef was excluded from the analysis (Fig. 3C), indicating their hybrid origin. Sampling design has been a major factor determining the resulting genetic clustering using molecular data (Puechmaille 2016; Lawson et al. 2018). Our results further caution the interpretation of hybrids and their parental species using molecular data, where specific attention has to be paid regarding sampling strategy, especially when immediate parental samples might have been included in the study. We showed that by sequentially excluding individuals from different clusters, we could identify inconsistencies in inferred genetic clusters and better understand the genetic composition of hybrids. Note that there is an alternative hypothesis: that the paternal sample could have been a natural hybrid and that our fineRAD structure result thus reflects that possibility. However, we believe such an scenario is unlikely because, as mentioned in the introduction section, the paternal and maternal species/populations were distributed allopatrically with respect to other Dynastes beetles. Future reference genome resources will help us test for this hypothesis by evaluating the genome-wide heterozygosity information of the paternal and F1 samples.

Our successful validation of the genetic composition of F1 hybrids clearly indicated that taxonomically well-accepted biological species can indeed interbreed and produce viable and fertile offspring and added to the growing evidence for hybridization in animals (Mallet 2008; Hendrick 2013). Although distinct animal species do hybridize in nature and can often be successfully created under artificial condition (Tobler et al. 2005; Lai and Ko 2008; Stelkens et al. 2009; Hedrick 2013; Meier et al. 2017; Linnen et al. 2018), hybridization between animal species sparks controversies (Stelkens et al. 2009), especially in communities focusing on animal taxonomy (Wilson and Brown 1953; Andújar et al. 2014; Maxwell et al. 2019; Huang 2020; Burbrink et al. 2022). We agree with Wilson and Brown (1953) that hybrid sterility was never the defining criterion for biological species, and our results revealed the fact that pre-and post-zygotic reproductive isolation between allopatric taxa may require a long evolutionary time to develop (Huang 2020). While hybridization does have implications in animal taxonomy, we believe the opportunity created by having this hybrid progeny extends beyond systematics.

Different genomic compositions originated from different parental species, and the phenotypic variation can be expected in F2 adults. An associated mapping approach can be applied to identify the loci underlying phenotypic diversity given sufficient sampling size (Tobler et al. 2005; Linnen et al. 2018). Specifically, genomic data have been routinely applied in the Dynastes beetle system for systematic, evolution, and conservation oriented studies (Huang and Knowles 2016; Huang 2016 2019 2021). Furthermore, genomic resources for Dynastes beetles are becoming available (e.g., Morgan et al. 2022). We argue that the Hercules beetles can become a model system to understand repeated and parallel phenotypic evolutions. The phenotypic diversity in beetles and the diverse life-history characteristics have attracted the attention of generations of evolutionary biologists (Beebe 1947; Emlen 2001; McCullough et al. 2014). The Hercules beetle was among the first series of scarab beetles named by Linné (Krell et al. 2012). With the hybrid progeny, we may finally move on to understand the genomic basis of phenotypic and species diversity while testing the theories explaining the origin of such diversity generated during the long history of study interest in scarab beetles.

Supplementary materials

SM. 1.

co1.nex.

SM. 2.

hyb.alleles.

SM. 3.

hyb.str.

SM. 4.

hyb_stats.txt.

Acknowledgments

We thank Yi-Chih Hong and Leigh Hsu for providing parental beetle samples of D. maya and D. grantii for our hybridization experiments. Yi-Hsiu Kuan helped with permits to import and breed live exotic insects. Yi-Chih Hong helped with providing breeding and larval rearing materials. The study was supported by a research grant from the Ministry of Science and Technology to J.-P. Huang (MOST 108-2621-B-001-001-MY3). One reviewer and the associate editor provided valuable comments and suggestions that improved the clarity of the manuscript.

Footnotes

Authors’ contributions: J-PH conceived the study, processed and analyzed the data, and wrote the manuscript. W-YC extracted the DNAs and W-YC and MHL prepared the ddRADseq library. All authors read and revised the manuscript.

Compting interests: The authors declare there is no conflict of interest.

Availability of data and materials: The CO1 sequences have been submitted to GenBank (accession numbers: ON661062 to ON661068); the newly generated ddRADseq reads have been archived in the Sequence Read Archive under bioproject PRJNA845960. The CO1 alignment, the .str files for PCA, and the .allele file for fineRAD structure analyses were provided as supplementary materials.

Consent for publication: All authors approve the submission of this manuscript.

Ethics approval consent to participate: Not applicable.

References

  1. Andrews KR, Good JM, Miller MR, Luikart G, and Hohenlohe PA. 2016. Harnessing the power of RADseq for ecological and evolutionary genomics. Nat Rev Genet 17:81–92. doi:10.1038/nrg.2015.28. [DOI] [PMC free article] [PubMed]
  2. Andújar C, Arribas P, Ruiz C, Serrano J, Gómez-Zurita J. 2014. Integration of conflict into integrative taxonomy: fitting hybridization in species delimitation of Mesocarabus (Coleoptera: Carabidae). Mol Ecol 23:4344–4361. doi:10.1111/mec.12793. . [DOI] [PubMed]
  3. Beebe W. 1947. Notes on the Hercules beetle, Dynastes hercules (Linn.), at Rancho Grande, Venezuela, with special reference to combat behavior. Zoologica 32:109–116. doi:10.5962/p.203537.
  4. Burbrink FT, Crother BI, Murray CM, Smith BT, Ruane S, Myers EA, and Pyron RA. 2022. Empirical and philosophical problems with the subspecies rank. Ecol Evol 12:e9069. doi:10.1002/ece3.9069. . [DOI] [PMC free article] [PubMed]
  5. Burma BH. 1949. The species concept: a semantic review. Evolution 3:369–370. . [DOI] [PubMed]
  6. Castillo DM. 2017. Factors contributing to the accumulation of reproductive isolation: a mixed model approach. Ecol Evol 7:5808–5820. doi:10.1002/ece3.3093. . [DOI] [PMC free article] [PubMed]
  7. Coyne JA, Orr HA. 1989. Patterns of speciation in Drosophila. Evolution 43:362–381. doi:10.1111/j.1558-5646.1989.tb04233.x. . [DOI] [PubMed]
  8. Dong Z, Wang Y, Li C, Li Lm and Men X. 2021. Mitochondrial DNA as a molecular marker in insect ecology: current status and future prospects. Ann Entomol Soc Am 14:470–476. doi:10.1093/aesa/saab020.
  9. Eaton DAR. 2014. PyRAD: assembly of de novo RADseq loci for phylogenetic analyses. Bioinformatics 30:1844–1849. doi:10.1093/bioinformatics/btu121. [DOI] [PubMed]
  10. Eaton DAR, Overcast I. 2020. ipyrad: interactive assembly and analysis of RADseq datasets. Bioinformatics 36:2592–2594. doi:10.1093/bioinformatics/btz966. [DOI] [PubMed]
  11. Edelman MB, Frandsen PB, Miyagi M, Clavijo B, Davey J et al. 2019. Genomic architecture and introgression shape a butterfly radiation. Science 366:594–599. doi:10.1126/science.aaw2090. . [DOI] [PMC free article] [PubMed]
  12. Edgar RC. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nuc Acid Res 32:1792–1797. doi:10.1093/nar/gkh340. [DOI] [PMC free article] [PubMed]
  13. Emlen DJ. 2001. Costs and the diversification of exaggerated animal structures. Science 291:1534–1536. doi:10.1126/science.1056607. [DOI] [PubMed]
  14. Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R. 1994. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol Marine Biol Biotech 3:294–299. . [PubMed]
  15. García-Morales LJ, Moctezuma V, Sánchez-Huerta JL. 2022. New northernmost record for Dynastes maya Hardy and sympatry with Dynastes hyllus Chevrolat (Coleoptera: Scarabaeidae: Dynastinae). Coleopterists Bulletin 76:136–139. doi:10.1649/0010-065X-76.1.136.
  16. Gouy M, Tannier E, Comte N, Parsons DP. 2021. Seaview Version 5: a multiplatform software for multiple sequence alignment, molecular phylogenetic analyses, and tree reconciliation, pp. 241–260. In: Katoh K (eds) Multiple Sequence Alignment. Methods in Molecular Biology, vol 2231. NY, Humana. doi:10.1007/978-1-0716-1036-7_15. . [DOI] [PubMed]
  17. Guindon S, Dufayard J-F, Lefort V, Anisimova M, Hordijk W, Gascuel O. 2010. New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0. Syst Biol 59:307–321. doi:10.1093/sysbio/syq010. . [DOI] [PubMed]
  18. Hedrick PW. 2013. Adaptive introgression in animals: examples and comparison to new mutation and standing variation as sources of adaptive variation. Mol Ecol 22:4606–4618. doi:10.1111/mec. 12415. . [DOI] [PubMed]
  19. Hinton HE, Jarman GM. 1973. Physiological colour change in the elytra of Hercules beetle, Dynastes hercules. J Insect Physiol 19:533–549. doi:10.1038/238160a0.
  20. Huang J-P. 2016. Parapatric genetic introgression and phenotypic assimilation: testing conditions for introgression between Hercules beetles (Dynastes, Dynastinae). Mol Ecol 25:5513–5526. doi:10.1111/mec.13849. . [DOI] [PubMed]
  21. Huang J-P. 2017. The Hercules beetles (subgenus Dynastes, genus Dynastes, Dynastidae): a revisionary study based on the integration of molecular, morphological, ecological, and geographic analyses. Misc Pub Mus Zool Univ Mich NO 206.
  22. Huang J-P. 2018. What have been and what can be delimited as species using molecular data under the multi-species coalescent model? A case study using Hercules beetles (Dynastes; Dynastidae). Insect Syst Diversity 2:3. doi:10.1093/isd/ixy003.
  23. Huang J-P. 2019. Holocene population decline and conservation implication for the western Hercules beetle, Dynastes grantii (Coleoptera, Scarabaeidae). J Hered 110:629–637. doi:10.1093/jhered/esz036. . [DOI] [PubMed]
  24. Huang J-P. 2020. Is population subdivision different from speciation? From phylogeography to species delimitation. Ecol Evol 10:6890–6896. doi:10.1002/ece3.6524. . [DOI] [PMC free article] [PubMed]
  25. Huang J-P. 2021. The genealogical divergence index across a speciation continuum in Hercules beetles. Insect Syst Diversity 5:1. doi:10.1093/isd/ixab021.
  26. Huang J-P, Knowles LL. 2016. The species versus subspecies conundrum: quantitative delimitation from integrating multiple data types within a single Bayesian approach in Hercules beetles. Syst Biol 65:685–699. doi:10.1093/sysbio/syv119. . [DOI] [PubMed]
  27. Jarman GM, Hinton HE. 1974. Some defense mechanisms of the Hercules beetle, Dynastes hercules. J Entomol A General Entomol 49:71–80. doi:10.1111/j.1365-3032.1974.tb00070.x.
  28. Jennings JH, Snook RR, Hoikkala A. 2014. Reproductive isolation among allopatric Drosophila montana populations. Evolution 68:3095–3108. doi:10.1111/evo.12535. . [DOI] [PubMed]
  29. Jombart T. 2008. adegenet: a R package for the multivariate analysis of genetic markers. Bioinformatics 24:1403–1405. doi:10.1093/bioinformatics/btn129. . [DOI] [PubMed]
  30. Krell F-T, Branco T, Ziani S. 2012. Scarabaeus Linnaeus, 1758, Dynastes MacLeay, 1819, scarabaeinae Latreille, 1802, and dynastinae MacLeay, 1819 (Insecta, Coleoptera, scarabaeoidea): proposed conservation of usage. Bulletin Zool Nomenclature 63:182–190.
  31. Lai J, Ko H-P. 2008. For the Love of Rhinoceros and Stag beetles. 2nd ed., Taipei, Taiwan, Morning Star Publisher.
  32. Lawson DJ, van Dorp L, Falush D. 2018. A tutorial on how not to over-interpret STRUCTURE and ADMIXTURE bar plots. Nature Comm 9:3258. doi:10.1038/s41467-018-05257-7. . [DOI] [PMC free article] [PubMed]
  33. Lefort V, Longueville J-E, Gascuel O. 2017. SMS: smart model selection in PhyML. Mol Biol Evol 34:2422–2424. doi:10.1093/molbev/msx149. . [DOI] [PMC free article] [PubMed]
  34. Linnen CR, O’Quin CT, Shackleford T, Sears CR, Lindstedt C. 2018. Genetic basis of body color and spotting pattern in redheaded pine sawfly larvae (Neodiprion lecontei). Genetics 209:291–305. doi:10.1534/genetics.118.300793. . [DOI] [PMC free article] [PubMed]
  35. Malinsky M, Trucchi E, Lawson DJ, Falush D. 2018. RADpainter and fineRADstructure: population inference from RADseq data. Mol Biol Evol 35:1284–1290. doi:10.1093/molbev/msy023. . [DOI] [PMC free article] [PubMed]
  36. Mallet J. 2008. Hybridization, ecological races and the nature of species: empirical evidence for the ease of speciation. Phil Trans Roy Soc B Biol Sci 363:2971–2986. doi:10.1098/rstb.2008.0081. . [DOI] [PMC free article] [PubMed]
  37. Mallet J. 2020. Alternative views of biological species: reproductively isolated units or genotypic clusters? Nat Sci Rev 7:1401–1407. doi:10.1093/nsr/nwaa116. . [DOI] [PMC free article] [PubMed]
  38. Mayr E. 1942. Systematics and the Origin of Species from the Viewpoint of a Zoologist. New York, Columbia University Press.
  39. Mayr E. 1949. The species concept: semantics versus semantics. Evolution 3:371–372. doi:10.1111/j.1558-5646.1949.tb00038.x.
  40. Maxwell SJ, Bordon AV, Rymer TL, Congdon BC. 2019. The birth of a species and the validity of hybrid nomenclature demonstrated with a revision of hybrid taxa within Strombidae (Neostromboidae). Proc Biol Soc Washington 132:119–130. doi:10.2988/19-00007.
  41. McCullough EL, Tobalske BW, Emlen DJ. 2014. Structural adaptations to diverse fighting styles in sexually selected weapons. Proc Nat Acad Sci USA 111:14484–14488. doi:10.1073/pnas.1409585111. . [DOI] [PMC free article] [PubMed]
  42. Meier JI, Marques DA, Mwaiko S, Wagner CE, Excoffier L, Seehausen O. 2017. Ancient hybridization fuels rapid cichlid fish adaptive radiations. Nature Comm 8:14363. doi:10.1038/ncomms14363. . [DOI] [PMC free article] [PubMed]
  43. Moctezuma V, Sánchez-Huerta JL. 2018. New state and habitat records for Dynastes maya Hardy (Coleoptera: Scarabaeidae: Dynastinae) in Mexico. Coleopterists Bulletin 72:184–185. doi:10.1649/0010-065X-72.1.184.
  44. Morgan B, Wang T-Y, Chen Y-Z, Moctezuma V, Burgos O, Le MH, Huang J-P. 2022. Long-read sequencing data reveals dynamic evolution of mitochondrial-genome size and the phylogenetic utility of mitochondrial DNA in Hercules beetles (Dynastes; Scarabaeidae). Genome Biol Evol 14:evac147. doi:10.1093/gbe/evac147. . [DOI] [PMC free article] [PubMed]
  45. Morón MÁ. 1987. Los estados inmaduros de Dynastes hyllus Chevrolat (Coleoptera: Melolonthidae; Dynastinae); con observaciones sobre su biologia y el crecimiento alometrico del imago. Folia Entomológica Mexicana 72:33–74.
  46. Morón MÁ. 2009. El género Dynates MacLeay, 1819 en la zona de transición Mexicana (Coleoptera: Melolonthidae: Dynastinae). Boletín de la Sociedad Entomológica Aragonesa 45:23–38.
  47. Nosil P. 2008. Speciation with gene flow could be common. Mol Ecol 17:2103–2106. doi:10.1111/j.1365-294x.2008.03715.x. . [DOI] [PubMed]
  48. Peterson BK, Weber JN, Kay EH, Fisher HS, Hoekstra HE. 2012. Double digest RADseq: an inexpensive method for de novo SNP discovery and genotyping in model and non-model species. PLoS ONE 7:e37135. doi:10.1371/journal.pone.0037135. . [DOI] [PMC free article] [PubMed]
  49. Puechmaille SJ. 2016. The program STRUCTURE does not reliably recover the correct population structure when sampling is uneven: subsampling and new estimators alleviate the problem. Mol Ecol Res 16:608–627. doi:10.1111/1755-0998.12512. . [DOI] [PubMed]
  50. Simpson GG. 1951. The species concept. Evolution 5:285–298. doi:10.2307/2405675.
  51. Simmons LW. 2002. Sperm Competition and Its Evolutionary Consequences in the Insects. Princeton University Press, NJ, USA. doi:10.1111/j.1469-185X.1970.tb01176.x.
  52. Slon V, Mafessoni F, Vernot B, de Filippo C, Grote S, Viola B, Hajdinjak M, Peyrégne S, Nagel S, Brown S, Douka K, Higham T, Kozlikin MB, Shunkov MV, Derevianko AP, Kelso J, Prüfer K, Pääbo S. 2018. The genome of the offspring of a Neanderthal mother and a Denisovan father. Nature 561:113–116. doi:10.1038/s41586-018-0455-x. [DOI] [PMC free article] [PubMed]
  53. Stelkens RB, Schmid C, Selz O, Seehausen O. 2009. Phenotypic novelty in experimental hybrids is predicted by the genetic distance between species of cichlid fish. BMC Evol Biol 9:283. doi:10.1186/1471-2148-9-283. . [DOI] [PMC free article] [PubMed]
  54. Tobler A, Kapan D, Flanagan NS, Gonzalez C, Peterson E, Jiggins CD, Johnston JS, Heckel DG, McMillan WO. 2005. First-generation linkage map of the warningly colored butterfly Heliconius erato. Hered 94:408–417. doi:10.1038/sj.hdy.6800619. . [DOI] [PubMed]
  55. True JR, Liu J, Stam LF, Zeng Z-B, Laurie CC. 1997. Quantitative genetic analysis of divergence in male secondary sexual traits between Drosophila simulans and Drosophila mauritiana. Evolution 51:816–832. doi:10.1111/j.1558-5646.1997.tb03664.x. . [DOI] [PubMed]
  56. Turelli M, Lipkowitz JR, Brandvain Y. 2014. On the Coyne and Orr-igin of species: effects of intrinsic postzygotic isolation, ecological differentiation, x chromosome size, and sympatry on Drosophila speciation. Evolution 68:1176–1187. doi:10.1111/evo.12330. . [DOI] [PMC free article] [PubMed]
  57. Wilson EO, Brown Jr. WL. 1953. The subspecies concept and its taxonomic application. Syst Zool 2:97–111. doi:10.2307/2411818.
  58. Wu C-I. 2001. The genic view of the process of speciation. J Evol Biol 14:851–865. doi:10.1046/j.1420-9101.2001.00335.x.

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

SM. 1.

co1.nex.

zoolstud-62-013-s001.nex (12KB, nex)
SM. 2.

hyb.alleles.

zoolstud-62-013-s002.alleles (38MB, alleles)
SM. 3.

hyb.str.

zoolstud-62-013-s003.str (5.1MB, str)
SM. 4.

hyb_stats.txt.

zoolstud-62-013-s004.txt (6.4KB, txt)

Articles from Zoological Studies are provided here courtesy of Biodiversity Research Center, Academia Sinica

RESOURCES