Repeat-associated heterochromatin expansion in Acyclania tenebrosa, a noctuid with one of the largest lepidopteran genomes

Abstract

Background

Genome size in Lepidoptera is generally constrained, and heterochromatin is typically limited to sex chromosomes. In some cases, the heterochromatin and genome size increased, but the mechanisms underlying extreme genome and heterochromatin expansion remain poorly understood. Here, we investigated the structural and repetitive DNA composition of Acyclania tenebrosa genome, a species with high heterochromatin content, to explore how repetitive elements shape genome architecture.

Results

Acyclania tenebrosa retains the ancestral and modal diploid number for Lepidoptera (male 2n = 62) and lacks major chromosomal fusions. The species exhibits an extremely large genome for Lepidoptera (male 1 C = 2.09 Gb), whose expansion has been driven by multiple repetitive elements that predominantly form large heterochromatic blocks. Moreover, reshuffling of major rDNA was observed. The abundance of the estimated repeats, including satellite DNAs (satDNAs) and transposable elements (TEs) comprise ~ 68% of the genome content, including an unprecedented amplification of satDNAs among Lepidoptera (14.3% of genome content), specially dominated by the AtenSat01 family, which forms multiple heterochromatin blocks. Independent local amplification of satDNAs in terminal and interstitial regions was revealed by FISH mapping. The TEs were also abundant accounting for ~ 54% of the genome, particularly LINEs and LTRs also contributed to genome enlargement, with landscape analyses indicating temporally distinct waves of amplification.

Conclusions

Acyclania tenebrosa represents a striking example of gradual, repeat-mediated genome expansion, without evidence of whole-genome duplication. The integration of cytogenetic and genomic analyses demonstrates that massive repeat-associated heterochromatin accumulation resulted in unprecedent genome architectures in Lepidoptera, challenging the view of limited heterochromatin content in this order.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-026-12600-6.

Introduction

Heterochromatin, first identified by Emil Heitz in 1928, consists of densely packed chromosomal regions that remain stained throughout the cell cycle, distinguishing them from the more loosely packed euchromatin [1, 2]. This genomic fraction is divided into two categories: constitutive and facultative. Constitutive heterochromatin is highly condensed and is enriched in terminal/subterminal areas as well as pericentromeric regions, playing a crucial role in maintaining chromosome stability and ensuring proper chromosome segregation during cell division by anchoring kinetochore proteins [3, 4]. Additionally, interstitial heterochromatin loci have been identified at specific positions on chromosomes, and it is also enriched in particular elements, such as sex and supernumerary B chromosomes [5–8]. In this context, heterochromatin plays key roles in gene regulation, particularly in gene silencing, genome stability, evolution, and the overall three-dimensional spatial organization of the genome, revealing its structural and functional importance [2, 4, 9, 10].

Concerning composition, constitutive heterochromatin is especially enriched with repetitive DNAs, such as satellite DNAs (satDNAs) and transposable elements (TEs), which are major components of eukaryotic genomes [7, 11, 12]. Beyond this structural role, nowadays, heterochromatin is increasingly recognized as a transcriptionally active and regulatory environment, hosting expressed genes and noncoding RNAs (ncRNA) [13–15]. Since the first mapping of satDNAs in the heterochromatin of mouse chromosomes [16], a growing number of satDNAs associated with heterochromatin have been described [7, 17], and similarly, the enrichment of TEs in these regions has also been documented [12]. TEs are sequences primarily interspersed throughout the genome, whereas satDNAs are highly tandemly repeated sequences, forming long arrays, but dispersed satDNAs or clustered TEs also occur [18]. Generally, satDNA sequences exhibit a high turnover rate in abundance and nucleotide sequence, evolving faster than other genomic elements, leading to the emergence of species-specific profiles [19–23]. In this way, satDNAs form libraries with significant variations across species in both abundance and nucleotide sequence, driven by mutations and mechanisms of concerted evolution such as unequal crossing over, replication slippage, and gene conversion [24, 25]. Like satDNAs, the quantity and nucleotide sequences of TEs in genomes vary greatly, potentially generating species-specific elements. However, shared repertoires of TEs and satDNAs have also been observed even among distantly related species [7, 17, 26–28]. Both satDNAs and TEs are crucial for genome organization and are considered key drivers of genome size variation and evolution. The chromosomal position of these elements has been associated with chromosomal rearrangements, emphasizing their role in genome reorganization and karyotype evolution [27, 29–31].

Recent advances in genome sequencing and computational pipelines, such as RepeatExplorer [32, 33] and dnaPipeTE [34, 35], have greatly enhanced our understanding of heterochromatin and its associated repetitive DNAs. These approaches, often combined with cytogenetic analyses, have uncovered remarkable variability in heterochromatic repeats regarding their genomic abundance, chromosomal localization, turnover, and size across multiple orders. By focusing on these previously overlooked genomic fractions, researchers have gained key insights into the structure and dynamics of heterochromatin, including the amplification of specific repeats [36–38], the composition of sex and B chromosomes [39–42], and the role of repeats in centromere divergence and karyotype evolution [43–46]. These studies also provide empirical support for the library hypothesis of satDNAs [19, 47, 48], which proposes that related species share a common repertoire of satellite DNA families inherited from a common ancestor, with differential amplification or loss shaping their current genomic composition [17, 49]. Moreover, they illuminate lineage-specific differentiation driven by repeats [50, 51], while simultaneously assisting genome assembly efforts [52, 53].

Specifically, for Lepidoptera (moths and butterflies), despite the extensive amount of available genomic data, our understanding of heterochromatin and the chromosomal organization of repetitive DNAs remains limited [40, 54–57]. With around 160,000 described species [58], Lepidoptera not only exhibit remarkable diversity but also striking chromosomal and genome-size variability. Concerning genome size, Lepidoptera exhibit moderate variation, ranging from 0.199 Gb to over 2 Gb [59–61], with repetitive DNA content spanning from ~ 7.4% to 65.1% [61]. Generally, the TEs are abundant repeats in the lepidopteran genomes [61, 62], while the satDNAs are much less represented [51, 54, 56], with rare exceptions [55]. From a chromosomal point of view, lepidopterans possess holocentric chromosomes, which lack localized centromeres [63], and despite a typical haploid number of n = 31, chromosome counts can vary widely, from as few as n = 5 to as many as n = 229 [61, 64–66]. Furthermore, Lepidoptera is the largest animal group in number of species with stable female heterogamety, and their sex chromosomes are highly dynamic [67–69]. Heterochromatin is generally scarce and often undetectable on autosomes, with visible regions usually restricted to the W chromosome [67, 70]. However, some species exhibit conspicuous interstitial or terminal heterochromatic regions, highlighting the complexity of their genome architecture [68, 71]. Together, these findings characterize Lepidoptera as a complex group, and it is among the groups with the highest number of species whose genomes have been sequenced among all insect orders, many aspects of their heterochromatin organization, diversity, and evolutionary dynamics remain enigmatic.

This complexity and limited understanding of Lepidoptera genomes and chromosomal repetitive DNA organization is particularly evident in Noctuidae moths, one of the most diverse families within the order, comprising more than 12,000 species [58]. Beyond their important ecological role as pollinators [72], some noctuid species are major agricultural pests, with Spodoptera frugiperda recognized as one of the world’s most damaging species [73, 74]. Despite their global biodiversity and significant impact on human welfare, knowledge about their karyotypes and repetitive DNAs remains scarce [51, 75–78], hindering a full understanding of their genome organization and evolutionary dynamics. To address this gap, we recently initiated an investigation into the repetitive DNA content and chromosomal organization in representatives of Noctuidae. Specifically, we analyzed the karyotype, genome size, and repetitive DNA composition of Acyclania tenebrosa by integrating genomic and cytogenetic approaches. This combination allowed us to identify and characterize evolutionary patterns of repetitive elements at both the sequence and chromosomal levels. Large heterochromatic blocks dispersed on all chromosomes were noticed. We hypothesize that the unusually high abundance of heterochromatic regions in A. tenebrosa chromosomes is associated with a massive amplification of specific satDNAs and some TEs. Consistent with this idea, our results revealed an unprecedented accumulation of these repeats forming large heterochromatic blocks. Such extensive amplification of repetitive elements likely underlies the formation of an exceptionally large genome in this species. These findings provide new insights into the evolutionary dynamics of repetitive DNA in Noctuidae and highlight the remarkable genomic plasticity that remains largely unexplored in this group.

Materials and methods

Sample collection and chromosome Preparation

A total of 30 individuals of Acyclania tenebrosa in the larval stage were collected at São Paulo State University (UNESP), Rio Claro campus (São Paulo, Brazil), where they were associated with and defoliating a Byrsonima crassifolia (Malpighiaceae) tree. Some individuals were reared in the laboratory until adulthood, and adults were used for species identification based on classical morphological characters. Males and females were recognized by the presence of testis or ovaries and were used for analysis, being testes and wing imaginal discs dissected in physiological solution. These organs were subjected to hypotonic treatment for 10 min in 75 mM KCl to induce osmotic shock and then fixed in Carnoy I solution (ethanol/chloroform/acetic acid, 6:3:1) for 15 min. Chromosome spreads were prepared as previously described [79, 80]. Pachytene meiotic chromosomes were obtained from the testes of third-instar larvae. Mitotic chromosomes were obtained from the wing imaginal discs of fifth-instar larvae. To obtain chromosomal preparations briefly, the fixed tissues were disaggregated in 40 µl of 50% glacial acetic acid in a microtube by pipetting. The cell suspension was repeatedly spread in three drops on a slide at 42 °C using a heating plate. The slides were dehydrated in a graded ethanol series (70%, 85%, and 100% for 30 s each) and stored at −20 °C until use. All dissected larvae bodies were stored in 100% ethanol in a freezer at −20 °C until the extraction of genomic DNA (gDNA).

To screen the presence or absence of the W sex chromatin, derived from multiple W chromosome copies, polyploid interphase nuclei were prepared from Malpighian tubules of last-instar larvae from both sexes. Malpighian tubules were dissected in physiological solution, briefly fixed in Carnoy’s fixative (~ 1 min) on slides and stained with 1.25% lactic acetic orcein for 5 min. Slides were then analyzed under a light microscope to search the sex chromatin, derived from multiple W chromosome copies [81]. As in other species of Lepidoptera, the heterochromatin distribution was observed using DAPI (4’,6-diamidino-2-phenylindole; Sigma-Aldrich) staining after hypotonic osmotic shock in pachytene chromosomes [68, 71, 82]. DAPI staining highlights A-T-rich domains that remain condensed and resistant to the relatively strong hypotonic shock, indicating the presence of heterochromatic blocks.

Nuclear 1 C value measurement

Nuclear genome size of A. tenebrosa was accessed by flow cytometry from brain ganglion nuclei. The mean 1 C value was measured from 15 males and 8 females, comprising fresh and 70% ethanol-fixed individuals. Cerebral ganglia were extracted from the mouthparts cephalic capsule and dissected in saline solution using a stereomicroscope. A Drosophila melanogaster female individual (1 C = 0.18 pg) was used as the reference standard. Cerebral ganglia from fresh individuals were transferred to a 1.5 mL microtube containing 100 µL of OTTO I nuclear extraction buffer (0.2 M citric acid, 0.5% Tween 20 and 50 µg/mL RNAse, pH = 2.3). Fixed ganglia were transferred to 100 µL of modified OTTO I (0.4 M citric acid, 0.5% Tween 20 and 50 µg/mL RNAse, pH = 2.3). Both A. tenebrosa and D. melanogaster ganglia were simultaneously macerated in nuclear extraction buffer using a plastic pistil until complete cellularization. After maceration, the cell suspension was incubated for 5 min at room temperature (RT). Following incubation period, 1000 µL of OTTO I buffer (modified OTTO I for fixed samples) was added prior to filtration through a 30 μm mesh into a 2.0 mL microtube. The tubes were centrifuged at 100 xg for 5 min, and the supernatant was carefully discarded. The pellet was resuspended in 100 µL of OTTO I buffer (modified OTTO I for fixed samples) using a vortex and incubated for 5 min at RT. Then, nuclei suspension was stained with 500 µL of OTTO II buffer (0.4 µM Na₂HPO₄.2H₂O supplemented with 75 µM propidium iodide and 50 µg/mL RNAse pH 7.8) for 30 min in the dark at RT. Afterwards, stained nuclear suspension was filtered through a 20 μm mesh into an acrylic tube (Sarsted, Nümbrecht, DE) for analysis in a flow cytometer BD Accuri™ C6. Analyses were carried out using monoparametric and biparametric histograms obtained through BD Accuri™ C6 software. Nuclear 1 C value was measured considering G₀/G₁ peaks showing coefficient of variation lower than 5.0%.

Low-pass genome sequencing, identification, and analysis of repetitive DNAs

Since our observations indicated that all chromosomes in males, including autosomes and sex elements, occur in pairs, we sequenced and analyzed the repetitive DNA from a male genome to ensure equal dosage across chromosomes. We adopted this strategy because we could not clearly determine the sex chromosome system of the species (i.e., whether it is Z0/ZZ or Z₁Z₂W/Z₁Z₁Z₂Z₂, females and males, respectively), given the difference in diploid number and genome size between the sexes (see results). The gDNA was extracted from larva head using the “Wizard Genomic DNA Purification Kit” (Promega, Madison, WI) following the manufacturer’s protocol. The genome of A. tenebrosa was sequenced using paired-end sequencing (2 × 150 bp) on the Illumina HiSeq 4000 platform from a library constructed with DNA fragments of 350 bp, performed by Novogene (HK) Co., Ltd. (Hong Kong, China). The genome reads have been deposited in NCBI/Sequence Read Archive under the accession number SRR35430731. Reads were processed, filtered, interlaced, and analyzed using the tools available in RepeatExplorer2/TAREAN [33], being a total of 644,680 reads analyzed, that correspond to ~ 0.05× coverage of the species genome (see results). After clustering with RepeatExplorer2, the satDNAs identified by the TAREAN tool were selected for further analysis. While the satDNA analysis using RepeatExplorer2/TAREAN was performed exclusively for the identification of distinct families, quantification was carried out using RepeatMasker (see below). Furthermore, the tandem arrangement of satDNAs was confirmed using dot plots and Tandem Repeats Finder (TRF) [83]. Python script “rm_homology.py” (available at: https://github.com/fjruizruano/satminer/blob/master/rm_homology.py) were used to perform an all-against-all comparison of monomers from all recovered satDNAs to analyze similarities and define satDNA families, and their variants [84]. To identify potential similarities with known sequences, the satDNA library was submitted to CENSOR, available at GIRI RepBase [85]. The parameters to consider one satDNA with similarity with a TE was the minimum of 50% monomer coverage and exceeding 65% sequence similarity [86]. Satellite DNA nucleotide sequences were deposited in GenBank under the accession numbers PX393033- PX393046.

The abundance and divergence of each satDNA family were calculated using RepeatMasker, based on a total of 1,000,000 randomly selected reads. Genomic abundance for each satDNA family was estimated by calculating the genome fraction, which was determined from the number of nucleotides mapped relative to the total library size. Additionally, the “calcDivergenceFromAlign.pl” script from the RepeatMasker utility tools was used to estimate the average Kimura-2-parameter (K2P) distances. Finally, repeat landscapes were generated, displaying the relative abundance of the satDNAs on the Y-axis and the K2P distance from the consensus on the X-axis.

The overall TE library was identified using dnaPipeTE [34] based on higher classifications, i.e. TE class/superfamilies. This analysis utilizes Trinity [87] to assemble the most abundant repetitive sequences, followed by homology-based characterization [34]. For automated annotation, we employed a custom TE library as described by [88]. To perform a detailed quantification of each TE class, the dnaPipeTE analysis was supplemented with an additional RepeatMasker [89] quantification. For this purpose, we used a custom script, “processing_db_dnapipete.py,” to create a database in the appropriate format for RepeatMasker, using outputs from dnaPipeTE, specifically “one_RM_hit_per_Trinity_contigs” and “Trinity.fasta.” Subsequently, we ran CD-HIT-EST [90] to concatenate contigs and reduce redundancy, ensuring the sequences maintained at least 80% similarity (parameters: -c 0.80, -n 4). Following this, RepeatMasker was used, and K2P divergence and normalization were estimated as previously described, leading to the generation of repeat landscapes. In order to identify more complete TE sequences with specific classification (specific elements), additionally, we analyzed the output from RepeatExplorer2 and selected for analysis the six most abundant clusters with similarity to TEs. To recover the consensus sequence for each cluster, all contigs were processed in Geneious v11.1.5 software (https://www.geneious.com) for de novo assembly. These sequences were then annotated using the CENSOR tool from the Repbase/GIRI database [85] to confirm the similarity with TEs. Only hits with more than 50% coverage and over 60% similarity were included in further analysis. The abundance of these specific elements was also estimated using RepeatMasker.

DNA probes and chromosomal characterization through fluorescence in situ hybridization

To define the general characteristics of the species karyotype, the chromosomes were stained with DAPI, and the slides were mounted in VECTASHIELD (Vector, Burlingame, CA). To identify the physical location of repetitive DNAs and their relationship with the heterochromatin of A. tenebrosa, the satDNAs identified in the genome of A. tenebrosa, several abundant TEs (see results), 18 S rDNA, and the telomeric motif (TTAGG)_n were mapped on the chromosomes. The 18 S rDNA, satDNAs, and TE probes were amplified by PCR from male gDNA and labeled by nick-translation with digoxigenin-11-dUTP (Roche, Mannheim, Germany). The major ribosomal DNA was amplified using primers described in [91]. Primers for satDNA amplification were designed using Geneious software, based on the consensus sequence of each satDNA and TE. For satDNAs primers were all divergent, allowing the amplification of entire monomers, for TEs they were convergent and more conserved regions were amplified to be used as probes (Supplementary Table 1). Each PCR reaction consisted of 10× PCR Rxn Buffer, 0.2 mM MgCl₂, 0.16 mM dNTPs, 2 µM of each primer, 1 U of Taq Platinum DNA Polymerase (Invitrogen), and 50–100 ng/µL of template DNA. The PCR conditions included an initial denaturation at 94 °C for 5 min, followed by 30 cycles at 94 °C for 30 s, 55 °C for 30 s, and 72 °C for 80 s, with a final extension at 72 °C for 5 min. The insect telomeric probe was amplified using a non-template PCR method, following the protocol of [92], with modifications [93]. This was achieved using self-complementary primers (TTAGG)₅ and (CCTAA)₅. The PCR products were then separated on a 1% agarose gel using electrophoresis.

Chromosome preparations from testes were used for fluorescence in situ hybridization (FISH) following the protocol described by [94]. All probes were detected using anti-digoxigenin-Rhodamine (Roche). Chromosomes were counterstained with DAPI, and the slides were mounted in VECTASHIELD. Images of hybridized chromosomes were captured using a cooled DP70 digital camera attached to an Olympus BX51 fluorescence microscope with appropriate filter sets (Olympus, Hamburg, Germany). The images were pseudocolored in green or red, merged (chromosomes and signals), and optimized for brightness and contrast using Adobe Photoshop CC6.

Results

Unusual heterochromatin rich chromosomes and large genome size in Acyclania tenebrosa

The chromosome counting analysis of mitotic cells from wing imaginal discs of A. tenebrosa using a telomeric FISH probe, that allows to see clearly each individual chromosome, revealed a diploid number of 2n = 62 in males (Fig. 1a) and 2n = 61 in females (Fig. 1b), consistent with a sex chromosome system with absence of a W chromosome ZZ (males)/Z0 (females), or a multiple system, like Z₁Z₁Z₂Z₂ (males)/Z₁Z₂W (females). The chromosomes are holokinetic, lacking primary constrictions, as typically occur in Lepidoptera. In addition, they vary slightly in size (Fig. 1a, b). The absence of a W chromosome in females was further supported by the lack of sex chromatin, usually derived from the W chromosome [67], in Malpighian tube cells (Supplementary Fig. 1), but we take care with this data as absence of visible sex chromatin can sometimes be misinterpreted as the absence of a W chromosome [95]. The analysis of pachytene chromosomes from cells obtained after hypotonic osmotic shock of the testes revealed the unusual presence of multiple DAPI-positive heterochromatic blocks of varying size and DAPI brightness intensity, a pattern rarely observed in Lepidoptera, where heterochromatin is typically absent or restricted to specific chromosomes with low fluorescent signal uniformity [67, 70]. The heterochromatin blocks were distributed across all chromosomes in distinct positions (Fig. 1c, d), evidencing large amount of heterochromatin content. Probably due to the high heterochromatin content, the pachytene chromosomes, including non-homologous ones, tended to remain closely associated, likely reflecting the strong physical cohesion promoted by repetitive DNA clustering and heterochromatin-binding proteins (Fig. 1c). As result of this closely association, we were also unable to detect in female pachytene cells the expected univalent of a sex system with no W chromosome or occurrence of multivalent in the case of multiple sex chromosome system. Additionally, no metaphase I or nurse cell was observed, despite the analysis of multiple individuals in distinct ages. Chromosome mapping via FISH with 18 S rDNA revealed signals on six mitotic chromosomes of varying sizes, all located at terminal positions (Fig. 1e). This pattern was consistent with observations in pachytene bivalents, where the 18 S rDNA probe hybridized exclusively to three elements (Fig. 1f), suggesting that each labeled mitotic chromosome corresponds to one homologous pair in meiosis and confirming the terminal localization of these rDNA sites.

Fig. 1.

General characteristics of the chromosomes of Acyclania tenebrosa (a, b) Mitotic chromosomes from the wing imaginal disk probed with the (TTAGG)_n telomeric probe, revealing a chromosome count of 2n = 62 in males (a) and 2n = 61 in females (b). (c, d) High heterochromatin content, indicated by multiple heterochromatic blocks (arrowheads) distributed almost along the entire length of the chromosomes: (c) testis pachytene chromosomes, (d) selected bivalent from a pachytene. (e, f) Mapping of 18 S rDNA in mitotic metaphase (e) and on pachytene chromosomes (f). In (c, d), arrowheads highlight heterochromatin blocks, while in (e, f), arrows point to the 18 S rDNA clusters. Bar = 5 μm

We measured the nuclear 1 C value, revealing that the unusually high heterochromatin content observed in cytological analyses is accompanied by the large genome of A. tenebrosa. This suggests a link between the amplification of repeat-rich DNA regions (heterochromatin) and genome size, a relationship further supported by our abundance estimates of repetitive DNAs (see below). In addition, 1 C values estimated from 15 males and 8 females consistently differed between the sexes, with males exhibiting higher values (1 C = 2.09 Gb) than females (1 C = 1.97 Gb) (Supplementary Fig. 2). This is consistent with the fact that females have one chromosome fewer than males.

SatDNAs and transposable elements role in heterochromatin and genome expansion of Acyclania tenebrosa

Our analysis of repetitive DNA, including satDNAs and TEs, through bioinformatic approaches and chromosomal mapping, provided insights into the sequences driving genome expansion in A. tenebrosa and their organization within the large heterochromatic blocks. Consistent with the species’ high heterochromatin content (which is typically enriched in tandem repeats) and its large genome size, satDNAs represented a substantial fraction of the genome, accounting for approximately 14.3% of the male genome, as revealed by RepeatMasker estimation. Although highly abundant, few satDNA families were identified, accounting for six elements, with relatively small monomer lengths varying from 82 bp to 267 bp, as evidenced by TAREAN output. All satDNA sequences were A + T-rich, with A + T content ranging from 54.8% to 70.5% (mean 62.7%), consistent with the strong DAPI staining observed in heterochromatic blocks, suggesting that these repeats contribute substantially to the composition of heterochromatin. The satDNA families were ranked in decreasing order of abundance, and none showed consistent sequence similarity with previously described satDNAs or other repetitive elements. The most abundant satDNA family, AtenSat01-90, consisted of three closely related variants (AtenSat01.1–91, AtenSat01.2–92, and AtenSat01.3–89) differing slightly in length (91, 92, and 89 bp) and nucleotide composition (Table 1). Despite these minor differences, the variants shared ~ 84% sequence homology, indicating they likely arose from a common ancestral repeat and may contribute collectively to the expansion and structural organization of heterochromatic regions.

Table 1.

Main characteristics of satellite DNAs (satDNAs) identified in the genome of Acyclania tenebrosa. For each repeat, abundance, Kimura 2-parameter divergence (K2P), monomer length (base pairs, bp), and A + T richness are provided

SatDNA	Abundance (%)	K2P divergence (%)	Length (bp)	A + T (%)
AtenSat01-1	5.99	8.81	91	63.8
AtenSat01-2	5.70	12.87	92	60.8
AtenSat01-3	0.34	18.67	89	61.8
AtenSat02	1.87	5.85	180	62.6
AtenSat03	0.22	16.27	267	62.2
AtenSat04	0.08	9.53	246	65.4
AtenSat05	0.06	17.63	82	54.8
AtenSat06	0.02	0.53	190	70.5
Total abundance	14.28
Mean divergence		11.27

The variants AtenSat01.1–91 and AtenSat01.2–92 were similarly abundant, each comprising ~ 6% of the genome, whereas AtenSat01.3–89 was much less represented at only 0.34%, reflecting differential amplification within this satDNA family. The second most abundant family, AtenSat02-180, accounted for 1.87% of the genome, while the remaining satDNAs were collectively rare, totaling just 0.38%. These patterns suggest that specific variants of a satDNA family can undergo markedly unequal expansion, potentially influencing the structure and evolution of heterochromatic regions. Regarding sequence divergence, K2P values ranged from 0.53% for AtenSat06-190 to 18.67% for AtenSat01.3–89, that can reflect the patterns of amplification or homogenization of these elements. Notably, the three AtenSat01-90 variants also displayed distinct K2P values (Table 1).

To better understand the evolutionary dynamics of each satDNA family, we constructed repeat landscapes (abundance versus divergence), based on the output from RepeatMasker analysis. This approach allows us to visualize patterns of sequence accumulation and homogenization, providing insights into their relative ages, amplification history, and potential turnover, or homogenization. For the most abundant family, AtenSat01-90, the three variants displayed divergence peaks in distinct K2P values: AtenSat01.1–91 showed its highest abundance at 5–10% K2P divergence (Fig. 2a), AtenSat01.2–92 peaked around 15% (Fig. 2b), and AtenSat01.3–89 peaked between 15 and 20% (Fig. 2c). The remaining families also exhibited variable K2P divergence patterns: AtenSat02-180 peaked at ~ 5% (Fig. 2d); AtenSat03-267 displayed two peaks, one at 5–10% and another near 18% (Fig. 2e); AtenSat04-246 peaked at 5–10% (Fig. 2f); AtenSat05-82 showed a major peak at ~ 18% (Fig. 2g); and AtenSat06-190 peaked at 0% (Fig. 2h). This data reflects heterogeneous amplification histories, variable levels of sequence homogenization, and differences in relative age, with some elements showing recent expansions and others representing older, more diverged sequences.

Fig. 2.

Individual landscapes (abundance vs. divergence) for each satellite DNA or its variants observed in the genome ofAcyclania tenebrosa The distinct peaks of abundance correspond to different K2P divergence values for each satDNA, including the variants of AtenSat01-90. These differences in K2P divergence highlight the variations in the sequence homogenization of each element. (a) AtenSat01.1–91, (b) AtenSat01.2–92, (c) AtenSat01.3–89, (d) AtenSat02-180, (e) AtenSat03-267, (f) AtenSat04-246, (g) AtenSat05-82, and (h) AtenSat06-190. For detailed information, see the results

The chromosomal mapping through FISH of the satDNAs allowed us to explore the chromosomal organization of these repeats and provided insights into whether different satDNA families contribute broadly to heterochromatin composition across the genome or are restricted to specific chromosomal regions. The mapping revealed two main patterns depending on the repeat: (i) multiple distinct and intense hybridization signals widely distributed across the chromosomes, mostly corresponding to heterochromatic regions, or (ii) accumulation in specific chromosomes as distinct, localized clusters. The most abundant satDNAs, AtenSat01-90, revealed to be an important constituent of heterochromatin, as it revealed multiple discrete and intense signals virtually decorating most heterochromatic blocks across all chromosomes, independent of their position, interstitial or terminal. Due to their high sequence similarity (~ 84%), the specific chromosomal distribution of each AtenSat01-90 variants could not be distinguished by FISH, and independent hybridizations of each variant produced identical patterns (Fig. 3a–c). AtenSat02-180 was also mostly enriched on heterochromatic regions, although presented overall much weaker signals than AtenSat01-90, in accordance with its lower genomic abundance. On the other hand, in some terminal heterochromatic blocks larger signals were detected for AtenSat02-180 in comparison to AtenSat01-90 (Fig. 3d, e), suggesting differential local amplification or accumulation in specific chromosomal regions for these repeats. AtenSat03-267 was enriched in heterochromatin but not uniformly present across all heterochromatic blocks (Fig. 4a), being absent from certain regions. It is important to note that the boundaries between heterochromatic and euchromatic regions are difficult to define precisely, and minor or faint signals can sometimes be observed in the euchromatic areas. AtenSat04-246, AtenSat05-82 and AtenSat06-190 showed more restricted chromosomal distribution patterns, in accordance with their lower genomic abundances. AtenSat04-246 labeled two chromosomes: one bivalent carried a single interstitial block, while another harbored four (Fig. 4b). AtenSat05-82 was mapped on two chromosomes, each with a single interstitial cluster (Fig. 4c). AtenSat06-190 was restricted to a single interstitial block on one chromosome (Fig. 4d). In all cases, hybridization signals coincide with heterochromatic regions. This reveals the minor contribution of these lower-abundance elements to overall heterochromatin organization.

Fig. 3.

Chromosome mapping through Fluorescence in situ Hybridization of the two most abundant satellite DNAs in the male genome of Acyclania tenebrosa. (a-c) AtenSat01-90 (red), mapped in a pachytene nucleus (a), and on two selected pachytene bivalents, one large (b) and one small (c), showing the occurrence of this repeat across most heterochromatin blocks. (d-f) AtenSat02-180 (green), mapped in mitotic chromosomes from the wing imaginal disk (d), in a pachytene nucleus (e), and in a selected pachytene bivalent (f), illustrating the enrichment of this repeat in the terminal regions of the chromosomes. Some smaller interstitial blocks, coincident with heterochromatin, are also observed, which are less prominent than those seen for AtenSat01-90. Bar = 5 μm

Fig. 4.

Chromosomal location revealed through Fluorescence in situ Hybridization of four satellite DNAs in male pachytene chromosomes of Acyclania tenebrosa. (a) AtenSat03-267 is found in some heterochromatic blocks, but certain blocks lack this repeat, as shown in the selected bivalent in the inset (blue arrowheads). (b) AtenSat04-246 is located on two bivalents, with four clusters observed in one of them (inset). In the bivalent with multiple clusters, note the absence of signals in some heterochromatic blocks. (c) AtenSat05-82 is interstitially located in two bivalents, with distinct cluster sizes. (d) AtenSat06-190 is exclusively located interstitially in one bivalent. In (b-d), the signals are indicated by arrowheads. Bar = 5 μm

As revealed by combined analysis of dnaPipeTE followed by RepeatMasker quantification, in comparison to satDNAs, the TEs accounted to more genome abundance, corresponding to 53.67% of the genome content. The LINEs represented the largest fraction (19.62%) of TE sequences, followed by LTRs (14.78%), revealing they as important contributor to genome size enlargement in A. tenebrosa. The remaining classes contributed less to genome size of the species and were much less represented, with abundances ranging from 4.01% to 6.25%. The mean K2P divergence of TEs was approximately 14.37%, ranging from 9.83% to 20.96% (Table 2), reflecting a spectrum of sequence variation within this genomic component. TE landscapes revealed distinct waves of activity for different classes, mirrored in their patterns of homogenization and divergence. DNA transposons showed increased abundance at K2P divergences below 20% (Fig. 5a), suggesting relatively recent or ongoing amplification. LINE elements displayed a more uniform abundance across the 0–20% divergence range (Fig. 5b), indicative of continuous accumulation over time. In contrast, LTRs (Fig. 5c), Helitrons (Fig. 5d), and SINEs (Fig. 5e) peaked around 20% divergence, suggesting that these elements represent older, more diverged copies, with limited recent activity. Collectively, these patterns indicate that different TE classes have contributed to genome evolution through waves of amplification occurring at different times.

Table 2.

Main characteristics of transposable elements (TEs) identified in the genome of Acyclania tenebrosa through DnaPipeTE (TE class) and by repeatexplorer along with CENSOR classification (TE). For each repeat, abundance and Kimura 2-parameter divergence (K2P) are provided. Repeat length (base pairs, bp) is shown for TEs from repeatexplorer analysis

TE classification	Abundance (%)	K2P divergence (%)	Length (bp)	A + T (%)
dnaPipeTE
DNA	4.01	11.32
LINE	19.62	14.27
LTR	14.78	9.83
Helitron	6.25	15.65
SINE	4.94	20.96
unknown	4.07	11.34
Total abundance	53.67
Man divergence		13.895
RepeatExplorer/CENSOR
LINE-RTE	6.4	13.78	3,462
Unclassified/REP-3_Lmi	1.52	9.93	1,588
LTR-BEL/Pao	1.11	7.16	6,938
LINE-Jockey	1.91	15.80	5,123
LTR-Gypsy	1.00	8.87	5,086
LINE-Jockey/Daphne	0.23	12.16	2,954

Fig. 5.

Transposable elements landscapes (abundance vs. divergence) from male for elements identified in the genome of Acyclania tenebrosa. (a) DNA, (b) LINE, (c) LTS, (d) Helitron, and (e) SINE. The increased abundance for DNA elements and LINEs is associated with lower K2P divergence, while for LTRs, Helitrons, and SINEs, the peak of abundance occurs around 20% K2P divergence. The names of each element are labeled directly in each landscape

Through RepeatExplorer2 analysis, in addition to the identification of satDNAs, we identified relatively abundant TEs. A more specific classification of these TEs was confirmed by comparison with previously identified elements deposited using CENSOR from Repbase. A total of six TE elements were identified, including three LINEs, two LTR elements, and one unclassified TE described as REP-3_Lmi, a TE found in the genome of the grasshopper Locusta migratoria. These elements ranged in size from 1,588 to 6,938 bp. Their abundance varied from 0.23% for LINE-Jockey/Daphne to 6.4% for LINE-RTE. Except for REP-3_Lmi, which was described in the grasshopper genome, the other five TEs were previously identified in other species of Lepidoptera (Table 2; Supplementary Table 2). These elements were used in FISH mapping, providing insights into their association with, and potential contribution to, the composition of heterochromatic regions. The chromosomal mapping of the six TE elements revealed variable labeling intensities across the chromosomes. These elements were predominantly enriched in the heterochromatic blocks, although the euchromatic regions also exhibited few dispersed signals (Fig. 6), particularly for REP-3_Lmi (Fig. 6b) and LDTI LINE-Jockey (Fig. 6d). The FISH signal intensities were similar for the four most abundant TEs, while the two least abundant elements showed weaker signals and virtually lower number of loci (Fig. 6), reflecting their lower genomic representation, and suggesting differential contributions to chromosomal heterochromatin organization.

Fig. 6.

Fluorescence in situ Hybridization of showing the chromosomal distribution of male for the six abundant transposable elements identified in Acyclania tenebrosa genome through combined analysis using RepeatExplorer and CENSOR (Repase), (a) LINE-RTE, (b) unclassified element, (c) LTR-BEL/Pao, (d) LINE-Jockey, (e) LTR-Gypsy, and (f) LINE/Daphne. It is shown entire pachytenes and selected chromosomes. The hybridization signals are primarily enriched in the heterochromatic blocks of the chromosomes, with scattered signals observed in the euchromatin, particularly for (b) REP-3_Lmi and (d) LDTI LINE-Jockey. Each panel shows the chromosome stained with DAPI (gray), the hybridization signals (red), and the merged image. Arrowheads in (b) highlight some signals located in euchromatin regions. Bar = 5 μm

Discussion

The integrative structural and genomic analysis of repetitive DNA in A. tenebrosa revealed a striking example of genome size enlargement and unprecedented heterochromatin amplification, primarily driven by the expansion of satDNAs and TEs. The male karyotype (2n = 62) matches the most common chromosome number in Lepidoptera, suggesting that fusions of fissions did not play a major role in shaping this genome. However, massive amplification of the heterochromatin-associated sequences took place, causing the emergence of numerous heterochromatic blocks across all chromosomes. Repetitive DNA dynamism is further supported by the multiplication of rDNA clusters. While a single terminal cluster is thought to represent the ancestral state in Lepidoptera [96, 97], A. tenebrosa exhibits three terminal rDNA sites. This result emphasizes the plasticity of rDNA, which in other noctuids (such as Spodoptera spp. and Mamestra brassicae) can also relocate to interstitial positions [57, 96, 97]. The presence of multiple terminal rDNA sites in A. tenebrosa is consistent with proposed mechanisms of rDNA dispersion that putatively involved satDNA-mediated ectopic recombination or extrachromosomal circular DNA [57].

With a male genome size of 2.09 Gb, according to our knowledge, A. tenebrosa harbors the second-largest genome reported in Lepidoptera to date [59, 61], surpassed only by Euclidia mi (~ 2.32 Gb; [60]). This 1 C value is ~ 10.5× larger than the smallest lepidopteran genome documented for Heliconius xanthocles (199 Mb; [59]). Such expansion is exceptional in Lepidoptera, where most genomes, including noctuids, remain below 1 Gb [59, 61, 98]. The few Noctuidae exceeding 1 Gb, e.g., Orthosia gothica [99], Polia nebulosa [100], Sesamia nonagrioides [101], and Tholera decimalis [52] do not approach the large size of A. tenebrosa, highlighting its particularity. For the last two species, the repeat content was estimated to be around 65% of the genome [61, 101]. Similar to what has been proposed for A. tenebrosa, the larger genome size in S. nonagrioides appears to be driven by the abundance of repetitive elements, as in other Noctuidae with smaller genomes, the proportion of repetitive DNA is much lower [61, 101]. The disproportionate enlargement of this genome therefore raises important questions about the evolutionary forces and genomic mechanisms behind repeat-driven expansion in Lepidoptera, which remains poorly understood.

Here, the integration of bioinformatic and chromosomal analyses provided detailed insights into the genome of A. tenebrosa, allowing us to hypothesize that its extraordinary genome enlargement resulted from the expansion of heterochromatic and repetitive-rich genomic regions, which together account for ~ 68% of the genome, with no evidence of polyploidization events. This interpretation is supported by the presence of large, multiple heterochromatic blocks distributed across the chromosomes, as well as by the exceptionally high abundance of satDNAs and TEs enriched in heterochromatin, and the occurrence of the conserved diploid number for lepidopterans. Genome enlargement driven by repetitive DNAs without evidence of whole-genome duplication has also been reported in other groups, such as grasshoppers [102, 103], rotifers [104], tunicates [105], and Antarctic krill [106]. In A. tenebrosa, the amplification of repeats is further evident from their overall proportion in comparison to other species, being among the highest reported in Lepidoptera [61]. However, it is important to note that in many genome analyses in Lepidoptera the satDNAs abundance are not properly addressed and estimation of TE content used distinct approaches, making these estimates somewhat controversial and study-dependent. Heterochromatin amplification of this magnitude is highly unusual in Lepidoptera. Typically, heterochromatin is restricted to the W chromosome, occasionally found on the NOR-bearing chromosome, and only rarely observed in other autosomes. Exceptional cases of large heterochromatin blocks have been reported in Leptidea (Pieridae; [68]) and Abraxas glossulariata (Geometridae; [71]). Yet, compared to A. tenebrosa, the heterochromatin in these species is more limited in number and distribution, making A. tenebrosa the most striking case documented so far in Lepidoptera. Interestingly, heterochromatin amplification has occurred independently across unrelated families, as Noctuidae, Pieridae, and Geometridae [68, 71]. Unlike Leptidea, where heterochromatin expansion coincided with extensive chromosomal reshuffling, A. tenebrosa retained the ancestral diploid number, suggesting that repetitive DNA can drive genome expansion without necessarily altering diploid number. Still, the role of repetitive DNAs in facilitating chromosomal evolution is well recognized, as these regions are particularly prone to rearrangements [31, 107, 108]. Although rare in Lepidoptera, massive heterochromatin expansion is common across other eukaryotes, including insects, as seen in Melipona bees [109], Scarabaeinae [91] and Tenebrio molitor [110] beetles, and several hemipterans [111, 112]. Therefore, A. tenebrosa genome provides a rare but powerful model to study how repetitive DNA can reshape genome composition and organization in Lepidoptera.

Regarding the composition of heterochromatin blocks, we revealed an exceptionally high amount of repetitive DNA in the genome of A. tenebrosa, including the largest proportion of satDNAs reported so far in Lepidoptera (14.3%, ~ 300 Mb). This represents a rare exception in the order, as substantial satDNA fractions have previously been documented only in the genus Leptidea, with the highest value observed in L. morsei (~ 11%; [55]). By contrast, most lepidopterans harbor less than 1% of satDNA in their genomes [51, 54, 56]. As in Leptidea, the satDNA landscape of A. tenebrosa is dominated by a single family, AtenSat01, which, with its three subfamilies, accounts for ~ 84% of the total satDNA content. Interestingly, despite its massive amplification, this family shows evidence of distinct amplification waves, as indicated by variable K2P divergence values among subfamilies, even between those of similar abundance. FISH mapping further confirmed that AtenSat01 is the major tandem repeat composing the multiple heterochromatic blocks and likely plays a significant role in their expansion. The second most abundant satDNA family also putatively plays a key role, acting as the important contributor to terminal heterochromatin amplification, as shown by strong hybridization signals in chromosome termini. A similar pattern is found in Leptidea, where the two most abundant satDNAs are likewise preferentially localized in heterochromatin [55]. The preferential position along the chromosomes of different satDNA families in terminal versus interstitial regions of A. tenebrosa suggests independent amplification processes. While the abundant satDNAs were predominantly enriched in heterochromatin, faint dispersed signals were also detected in euchromatic regions. The presence of satDNAs in euchromatin has similarly been documented in other animals [7, 17]. For the less abundant satDNAs, our data suggest localized amplification restricted to specific loci, emphasizing the remarkable dynamism and heterogeneity of satDNA organization and amplification patterns in the A. tenebrosa genome. In particular, AtenSat06-190 forms a single cluster confined to one chromosome pair, composed of highly homogeneous copies, as indicated by its low K2P divergence (0.53%) and peaks of abundance between 0% and 1% divergence. This pattern indicates a relatively young satDNA undergoing local amplification, highlighting the distinct evolutionary trajectories of satDNA families in A. tenebrosa.

Along with satDNAs, a strong enrichment of TEs in the heterochromatin of A. tenebrosa, indicates that, in addition to satDNAs, these repeats also contributed substantially to heterochromatin amplification. The overall TE content in this species is exceptionally high, comprising ~ 54% of the genome (~ 1.13 Gb), with LINEs and LTRs being the predominant elements. A similar pattern has been reported in other lepidopterans with enlarged genomes, such as the saturniid Hylesia metabus [62] and Parnassius apollo [113], where LINEs and LTRs also dominate. Landscape analyses further suggest that the amplification of these elements occurred in different evolutionary periods. LTRs show peaks at higher K2P values, consistent with ancient bursts of activity, whereas LINEs expanded more gradually over time, lacking a sharp peak but showing a continuous increase toward lower K2P values, indicative of more recent amplification. Based on landscape analysis, although less abundant, other TEs also exhibit evidence of independent waves of expansion. Together, these findings support the view that the massive genome increase in A. tenebrosa resulted from multiple, temporally distinct waves of repetitive DNA amplification. This contrasts with the case of Leptidea, where genome enlargement arose over a short evolutionary timescale due to a burst of TE hyperactivity [114] and also putatively recent satDNA amplification [55]. In A. tenebrosa, by contrast, genome expansion reflects a more gradual and layered process of repeat accumulation.

Unfortunately, in most Lepidoptera with enlarged genomes, cytogenetic studies addressing heterochromatin detection or chromosomal mapping of repetitive DNAs are still lacking, limiting our ability to better understand the relationship between genome expansion and the chromosomal dynamics of heterochromatin and its repeats. The genome of A. tenebrosa provides an exceptional example of how repetitive DNA can drive both structural and compositional diversification in Lepidoptera. The unprecedented amplification of repeats, together with the massive accumulation of heterochromatin, has not only shaped one of the largest genomes described for the order, but, along with findings from Leptidea species [68] and Abraxas glossulariata [71], also challenges the traditional view of Lepidoptera as an order with limited heterochromatin content. The extraordinary case of A. tenebrosa broadens the current framework of lepidopteran genome evolution and underscores the value of integrating cytogenetic and genomic approaches to reveal hidden layers of genome plasticity inaccessible to genomic tools alone. Altogether, our results demonstrate that A. tenebrosa constitutes a highly peculiar case within Lepidoptera, where massive repeat-driven heterochromatin expansion has generated one of the most remarkable genome architectures known in the order.

Conclusions

Our cytogenomic investigation demonstrates that the exceptional genome enlargement of A. tenebrosa is driven almost entirely by massive and multi-layered amplification of repetitive DNA, rather than polyploidization, and independently of major karyotype changes that altered 2n through chromosome fusions or fissions. SatDNAs and TEs together constitute ~ 68% of the genome, that undergone successive expansion waves contributing to extensive heterochromatin content across all chromosomes. This repeat proliferation also observed for rDNA, that has increased number of terminal clusters and reflecting high structural plasticity. The magnitude and distribution of heterochromatin in A. tenebrosa surpass all previously reported cases in Lepidoptera, revealing a singular extreme repeat-driven genome expansion. Overall, A. tenebrosa provides an interesting species for understanding how repetitive DNA alone can profoundly reshape genome architecture and drive major compositional shifts in lepidopteran genomes.

Authors’ contributions

AEG, ABSMF, RTA-G, FAFS, AO, WC: did the experimental work; DCC-d-M and AEG: analyzed the data, interpreted the results, and wrote the manuscript; DCC-d-M: conceived and coordinated the study; all authors read, revised, and approved the manuscript.

Funding

The study was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP 2023/02581-2), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq 309979/2023-4 to DCC-d-M), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brazil (CAPES, financial code 001 to DCC-d-M and AEG).

Data availability

The datasets generated and/or analysed during the current study are available in the GenBank repository, accession numbers: SRR35430731, PX393033- PX393046.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.