Recently, hypotheses regarding the evolutionary patterns of ribosomal genes in ant chromosomes have been under discussion. One of these hypotheses proposes a relationship between chromosomal location and the number of rDNA sites, suggesting that terminal locations facilitate the dispersion of rDNA clusters through ectopic recombination during meiosis, while intrachromosomal locations restrict them to a single chromosome pair. Another hypothesis suggests that the multiplication of rDNA sites could be associated with an increase in the chromosome number in Hymenoptera due to chromosomal fissions. In this study, we physically mapped rDNA sites in 15 new ant species and also reviewed data on rDNA available since the revision by Teixeira et al. (2021a). Our objectives were to investigate whether the new data confirm the relationship between chromosomal location and the number of rDNA sites, and whether the increase in the chromosome number is significant in the dispersion of rDNA clusters in ant karyotypes. Combining our new data with all information on ant cytogenetics published after 2021, 40 new species and nine new genera were assembled. Most species exhibited intrachromosomal rDNA sites on a single chromosome pair, while three species showed these genes in terminal regions of multiple chromosome pairs. On one hand, the hypothesis that the chromosomal location of rDNA clusters may facilitate the dispersion of rDNA sites in the ant genome, as previously discussed, was strengthened, but, on the other hand, the hypothesis of chromosomal fission as the main mechanism for dispersion of ribosomal genes in ants is likely to be refuted. Furthermore, in certain genera, the location of rDNA sites remained similar among the species studied, whereas in others, the distribution of these genes showed significant variation between species, suggesting a more dynamic chromosomal evolution.

Formicidae, FISH, karyotype, molecular cytogenetics, rDNA sites

In Formicidae, molecular cytogenetic studies involving fluorescence in situ hybridization (FISH) for physical mapping of major ribosomal genes, 45S ribosomal DNA (rDNA), here referred to as rDNA clusters, were first conducted in Australian ants of the genus Myrmecia Fabricius, 1804 (Imai et al. 1992; Hirai et al. 1994, 1996). In recent years, FISH has been widely employed in several ant species, particularly in the Neotropical region (Santos et al. 2016; Aguiar et al. 2017; Micolino et al. 2019, 2022; Barros et al. 2021a, b, 2022a, b; Murakami et al. 2021; Silveira 2022; Teixeira et al. 2022, 2023; Jacintho et al. 2023). These molecular cytogenetic studies have provided valuable insights into various biological aspects of these insects, including evolution, taxonomy, and reproduction.

For instance, the physical mapping of ribosomal genes in certain ant genera has enabled the proposal of chromosomal rearrangements during their karyotypic evolution, such as the occurrence of inversions in Myrmecia (Hirai et al. 1996), Dolichoderus Lund, 1831 (Santos et al. 2016), Mycetophylax Emery, 1913 (Micolino et al. 2019), and Acromyrmex Mayr, 1865 (Barros et al. 2016; Teixeira et al. 2021a). Additionally, chromosomal polymorphisms involving the rDNA clusters, with homozygous and heterozygous karyotypes, which may arise from duplications/deletions due to unequal crossing-over or the formation of extrachromosomal circular DNA (eccDNA) have been observed in Gnamptogenys regularis Mayr, 1870 (Teixeira et al. 2020a) and Odontomachus bauri Emery, 1892 (Teixeira et al. 2021a). These may arise from duplications/deletions due to unequal crossing-over or the formation of extrachromosomal circular DNA (eccDNA) in these species. EccDNA can replicate via a rolling circle mechanism and then either get reintegrated into the genome or deleted from it, respectively causing duplications or deletions of these sequences (Teixeira et al. 2020a, 2021a).

Regarding ant taxonomy, mapping the chromosomal distribution of rDNA clusters has been important in helping to delimit specific boundaries between taxa, as is the case of the ants Camponotus renggeri Emery, 1894 and Camponotus rufipes (Fabricius, 1775) (Aguiar et al. 2017). These two species were subjects of discussion regarding taxonomic synonymization. However, the number of chromosomes bearing the rDNA clusters differs between them, with C. rufipes possessing one pair and C. renggeri possessing two pairs, a hereditary characteristic capable of distinguishing these two Camponotus Mayr, 1861 species (Aguiar et al. 2017). Additionally, C. rufipes and C. renggeri differ in ecological, molecular, and behavioral traits, and this further confirms their status as valid species (Ronque et al. 2015).

Furthermore, cytogenetic data, including the chromosome location of rDNA sites, in the fungus-growing ant Mycocepurus smithii (Forel, 1893), have contributed to enhancing the understanding of cytological mechanisms associated with thelytokous parthenogenesis in this species (Barros et al. 2022a). Karyotypic variations were observed in the asexual population (2n=9, 10, and 11) with a decay of the diploid structure in the absence of meiosis and genetic recombination, whereas in the sexual population, the karyotype remained stable (2n=14) with appropriate homologous pairing. The data mapping of rDNA sites in M. smithii shows a single chromosome pair bearing these genes in the sexual population and in the karyomorphs 2n=9 and 2n=11 of the asexual population, supporting the idea that asexual individuals are indeed diploids. However, these data demonstrate the decay of the diploid structure, particularly in the 2n=11 karyomorph, in which there is a variation in size between the homologs of the pair bearing rDNA sites (Barros et al. 2022a).

Recently, based on new and previously published data regarding the chromosomal mapping of ribosomal genes from 63 species, 19 genera and six subfamilies of ants, Teixeira et al. (2021a) proposed important insights into the general patterns of these genes in ant chromosomes. These authors showed that rDNA clusters have a non-random distribution within the ant genome in which there is a relationship between chromosomal location and the number of rDNA sites. Most ant species have a single intrachromosomal (pericentromeric/interstitial) rDNA site, whereas species with multiple rDNA sites have these genes located in the terminal regions. Based on Hirai’s model (2020), Teixeira et al. (2021a) argued that the terminal location of rDNA sites in ants would facilitate association with other non-homologous chromosome terminal sequences during the meiotic bouquet, forming affinity systems, which would lead to the occurrence of ectopic recombination and dispersion of rDNA clusters to other chromosomes. However, the intrachromosomal location of rDNA sites would hinder interaction with other chromosomes, restricting these genes to a single chromosome pair.

Alternatively, Menezes et al. (2021) proposed that the multiplication of rDNA sites could be linked to an increase of the chromosome number in most groups of Hymenoptera (ants, wasps and bees), suggesting that chromosomal fissions play a pivotal role in the dispersal of rDNA clusters in the karyotypes of these insects.

Despite notable advances in molecular cytogenetic data in ants, entire genera and even subfamilies have not yet been studied in this respect. Thus, in this study, we performed chromosomal mapping of ribosomal genes through FISH in 15 new ant species belonging to 9 genera, and also reviewed molecular cytogenetic data involving rDNA sites available since the paper by Teixeira et al. (2021a) was published. Our goal was to verify whether the chromosomal distribution of ribosomal genes in these ant species follows a relationship between the chromosomal location and the number of rDNA sites, and whether the increase in the chromosome number is significant in the dispersion of these genes in ant karyotypes.

Field campaigns to collect ant colonies were performed in French Guiana and Brazil in regions of Amazonian and Atlantic rainforests (Table 1) from the following locations: Campus Agronomique, Kourou (5.17312°N, 52.65480°W) and Petit Saut route (5.13051°N, 52.94385°W), both in French Guiana; Oiapoque, Amapá State (3.84151°N, 51.84112°W) and Viçosa, Minas Gerais State (20.75696°S, 42.87314°W), both in Brazil. Sampling license in Brazil was provided by the Instituto Chico Mendes de Conservação da Biodiversidade (ICMBio) (SISBIO accession numbers 87049-1). Adult specimens were deposited in the Coleção Entomológica do Laboratório de Coleoptera (CELC), at the Universidade Federal de Viçosa (UFV), Viçosa, Brazil.

In addition, for comparative purposes, we performed a survey of molecular cytogenetic data involving chromosomal mapping of ribosomal genes through FISH in ants since the last review by Teixeira et al. (2021a). The cytogenetic data are shown in Table 2 and the following traits were considered for each species: general number of chromosomes, number of chromosomes bearing rDNA clusters, and location of rDNA clusters on chromosomes.

For cytogenetic analysis, mitotic metaphase chromosomes were obtained from the cerebral ganglia of larvae after meconium elimination according to the methods described by Imai et al. (1988). The 18S rDNA probes were amplified via polymerase chain reaction (PCR) using primers 18SF1 (5`-TCATATGCTTGTCTCAAAG-3`) and 18SR1.1 (3`-TCTAATTTTTTCAAAGTAAACG-5`) designed for Melipona quinquefasciata Lepeletier, 1836 (Pereira 2006) in the genomic DNA from the ant Camponotus rufipes. Gene amplification was performed following Pereira (2006). The probes were labeled with digoxigenin-11-dUTP using Dig-Nick-Translation Mix (Roche Applied Science, Mannheim, Germany), and the FISH signals were detected with anti-digoxigenin-rhodamine (Roche Applied Science), following the manufacturer’s protocol.

The rDNA sites were mapped on the chromosomes of Neotropical ant species using FISH according to Pinkel et al. (1986) with modifications described in Teixeira et al. (2021a): the slides were treated with RNase A (100 μg/ml) and kept in a moist chamber at 37 °C for 1 h. After that, they were washed in 2×SSC for 5 min, incubated in 5 μg/ml pepsin in 0.01 N HCl for 10 min, washed in 1× PBS for 5 min, and dehydrated in 50%, 70% and 100% alcohol series for 2 min each. After this pretreatment, metaphase chromosomes were denatured in 70% formamide/2×SSC at 75 °C for 5 min, and 20 μl of hybridization mix including 200 ng of labeled probe, 2×SSC, 50% formamide, and 10% dextran sulfate was denatured for 10 min at 85 °C and added on preparations. The slides were kept in a moist chamber up to 37 °C overnight. Then, the slides were washed in 2×SSC for 5 minutes; the detection solution including anti-digoxigenin-rhodamine was added on slides that were kept in a moist chamber at 37 °C for 1 h. The slides were washed three times in 4×SSC/Tween 20 (4×SSC, 0.05% [v/v] Tween 20) and dehydrated in an alcohol series. Finally, counterstaining with DAPI (DAPI Fluoroshield, Sigma Aldrich) was performed.

Slides subjected to FISH with the 18S rDNA probes were photographed using an epifluorescence microscope Olympus BX60 attached to an Olympus DP23M camera, and CellSens image capture software, using the filters WG (510–550 nm) for the rhodamine, and WU (330–385 nm) for DAPI. Images of the chromosomes were arranged using Adobe Photoshop® CS6. At least 20 metaphases for each species were analyzed to determine the FISH patterns.

In this study, we physically mapped rDNA genes in the karyotypes of 15 species from 9 genera and 4 subfamilies (data for six genera have not been previously published) (Table 1). Among these species, 13 exhibit a single chromosomal pair bearing rDNA clusters, which are located in the pericentromeric region: in Brachymyrmex admotus Mayr, 1887 (Fig. 2A), B. heeri Forel, 1874 (Fig. 2B), Nylanderia sp. (Fig. 2D), Cephalotes cordatus (Smith, 1853) (Fig. 3A), Cyphomyrmex laevigatus Weber, 1938 (Fig. 3B), Megalomyrmex aff. incisus Smith, 1947 (Fig. 3D), Pheidole jelskii Mayr, 1884 (Fig. 3E), P. vorax (Fabricius, 1804) (Fig. 3F), Strumigenys schulzi Emery, 1894 (Fig. 3G), Neoponera unidentata (Mayr, 1862) (Fig. 4A), and Pseudoponera stigma (Fabricius, 1804) (Fig. 4B). In Azteca andreae Guerrero, Delabie et Dejean, 2010 (Fig. 1) and Cephalotes minutus (Fabricius, 1804) (Fig. 3C), the rDNA sites are located on the short arms. Furthermore, two species showed rDNA clusters on the short arms of more than one chromosome pair: Camponotus cameroni Forel, 1892 in two submetacentric and two subtelocentric pairs (Fig. 2C) and Solenopsis saevissima (Smith, 1855) in two submetacentric pairs (Fig. 3H).

Figure 1.

FISH with 18S rDNA probe (red signals) performed in ant Azteca andreae (Dolichoderinae). Scale bar: 5 µm.

Figure 2.

FISH with 18S rDNA probe (red signals) performed in different ant species of the subfamily Formicinae. Scale bars: 5 µm.

Figure 3.

FISH with 18S rDNA probe (red signals) performed in different ant species of the subfamily Myrmicinae. Scale bars: 5 µm.

Figure 4.

FISH with 18S rDNA probe (red signals) performed in different ant species of the subfamily Ponerinae. Scale bars: 5 µm.

Chromosome mapping data from rDNA sites, which have been available since the review conducted by Teixeira et al. (2021a), encompassing 25 species across 10 genera (including three new genera), were reviewed (Table 2). This new information focused on ants from the Neotropics. However, some invasive populations of Solenopsis invicta Buren, 1972 from eastern Asia were also investigated. In most taxa, a single chromosomal pair bearing rDNA sites, located in the pericentromeric or interstitial regions, was observed. However, populations of S. invicta exhibited multiple rDNA terminal sites: the native population from Argentina had two chromosome pairs carrying rDNA clusters, while invasive populations from Argentina, United States, and Taiwan showed large variation on this pattern, from 1 to 11 chromosomes carrying rDNA sites depending on the ploidy (Table 2).

Molecular cytogenetic data, which involve chromosomal mapping of rDNA clusters in ants, are now available for 103 species/subspecies, 28 genera and 6 subfamilies (this study, Table 2; reviewed in Teixeira et al. 2021a). Considering the number of chromosome pairs bearing rDNA clusters observed among the ant taxa analyzed since the review by Teixeira et al. (2021a) was published, 36 of them showed only a single pair bearing these genes while three species, namely S. saevissima and S. invicta, had two pairs, and C. cameroni had four pairs (Table 2). These patterns indicate that a diploid genome with a single chromosome pair carrying ribosomal genes should be considered an ancestral feature in Formicidae, as previously discussed by Teixeira et al. (2021a), and later suggested for the Hymenoptera in general (Menezes et al. 2021).

The rDNA physical mapping also strongly reinforces the relationship between the number of rDNA sites and their location discussed by Teixeira et al. (2021a) since the species C. cameroni, S. saevissima, and S. invicta have multiple rDNA sites located on short arms (including terminal regions) of the chromosomes while most other species studied show a single intrachromosomal rDNA site (Table 2). Based on the Hirai’s model (2020), terminal location of rDNA sites in ants would facilitate association with other non-homologous chromosome terminal sequences during the meiotic bouquet, forming affinity systems, which would lead to the occurrence of ectopic recombination and dispersion of rDNA clusters to other chromosomes. Hirai’s (2020) model for the dispersion of rDNA clusters in karyotypes does not address the presence of haploid males (haplodiploid reproductive system), such as those found in ants. We hypothesize that males could inherit rearrangements involved in the dispersal of rDNA clusters that occurred in the queens’ genomes. Due to the haploid nature of hymenopteran males, the occurrence of ectopic recombination is restricted to females which are diploid. Therefore, it is possible that the evolution of repetitive DNA sequences (such as rDNA clusters), through the mechanisms described by Hirai (2020) may be slower in Hymenoptera when compared to other insect orders that reproduce in diplodiploid fashion.

Furthermore, our results seem to refute the hypothesis of chromosomal fission as the main mechanism for dispersion of ribosomal genes in ants proposed by Menezes et al. (2021) since: (i) species with the same chromosome number show differences in the number of rDNA sites, such as in Solenopsis geminata (Fabricius, 1804) (2n=32; one pair with rDNA), S. saevissima (2n=32; two pairs with rDNA) and the native population of S. invicta from Argentina (2n=32; two pairs with rDNA), in addition to C. rufipes (2n=40; one pair with rDNA) and C. renggeri (2n=40; two pairs with rDNA); (ii) in Camponotus Mayr, 1861, C. cameroni has 2n=36 chromosomes and four pairs bearing rDNA sites compared to other Camponotus species with 2n=40 chromosomes and only one or two pairs with rDNA clusters; (iii) in the fungus-growing ant M. smithii, asexual populations had a certain degree of relaxed chromosome stability (2n=9 and 11) when compared to sexual populations (2n=14) as discussed by Barros et al. (2022a), and still both populations had a single chromosome pair bearing the intrachromosomal rDNA sites; and (iv) in several ant genera studied, a wide variation in chromosome number is observed between species that present only a single pair with rDNA sites, for example, Cephalotes Latreille, 1802 (2n=24 to 44), Pheidole Westwood, 1839 (2n=20 to 42), and Strumigenys Smith, 1860 (2n=4 to 40) (this study, Table 2, reviewed in Teixeira et al. 2021a). As discussed by Hirai (2020), centric fissions could generate chromosomes with very short heterochromatic arms (acrocentrics) and rDNA clusters in terminal/subterminal positions, which could facilitate affinity associations between these genes and other terminal chromosomal sequences promoting dispersion of rDNA sites in the karyotype of ant species.

For several ant genera studied, only single species have any kind of molecular cytogenetic data on rDNA clusters available, as in Nylanderia Emery, 1906, Megalomyrmex Forel, 1885, and Neoponera Emery, 1901 (this study, Table 2, reviewed in Teixeira et al. 2021a), which limits comparisons and in-depth discussion, however, these data are important to start understanding the chromosomal evolution of ribosomal genes in these genera. For other genera, such as Brachymyrmex Mayr, 1868, Camponotus, Solenopsis Westwood, 1840, Pseudoponera Emery, 1900, Strumigenys, Cyphomyrmex Mayr, 1862, Pheidole, and Cephalotes ribosomal gene mapping data is available for some species (this study, Table 2, reviewed in Teixeira et al. 2021a), which allowed interspecific comparisons and the observation of some karyotypic patterns, as discussed below.

In summary, the molecular cytogenetic data from this study, as well as those available after the publication of the revision by Teixeira et al. (2021a), describe chromosome patterns for 40 new species and nine new genera. These new data strengthen the hypothesis suggesting that a single rDNA site per haploid genome represents the ancestral condition in ants. Furthermore, the data reinforce the observed non-random chromosomal distribution of ribosomal genes in Formicidae karyotypes, in which the chromosomal location (terminal or intrachromosomal) of these genes possibly influences the dispersion of rDNA sites in the ant genome. In certain genera, the location of rDNA sites in relation to which chromosomal pair carries the rDNA sites and whether it is located on the short or long arm, remained similar among the species studied, however in others, the distribution of these genes exhibited significant variation between species, suggesting a more dynamic chromosomal evolution. The expansion of molecular cytogenetic studies encompassing other ant subfamilies will continue to enhance our understanding of the chromosomal evolution of ribosomal genes in the genomes of these insects.

We are grateful to Dr. Júlio Chaul for the identification of most of the specimens, Dr. Alexandre Ferreira for the identification of the two Pheidole species, and Frédéric Petitclerc for the identification of Azteca andreae. We are thankful to the reviewing team, Dr. Cléa Mariano, Dr. Pedro Lorite, and Dr. Vladimir Gokhman for the important contributions and improvements to the manuscript. Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) funded MTSD, RBL and PCF and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) funded GAT. This study was also supported by Fundação de Amparo à Pesquisa do Amapá (FAPEAP, #286), Conselho Nacional de Desenvolvimento Científico e Tecnológico (403986/2021-4).

Monique Telcia dos Santos Damasceno https://orcid.org/0000-0003-0373-1389

Gisele Amaro Teixeira https://orcid.org/0000-0002-7106-5798

Paulo Castro Ferreira https://orcid.org/0009-0000-1640-9757

Rodrigo Batista Lod https://orcid.org/0000-0002-3710-5354

Luísa Antônia Campos Barros https://orcid.org/0000-0002-1501-4734

Hilton Jeferson Alves Cardoso de Aguiar https://orcid.org/0000-0001-7738-1460