Two valid ant species, Camponotus rufipes and Camponotus renggeri, have recently been the subject of a broad discussion with reference to taxa synonymization. Both species are quite common among the Neotropical myrmecofauna and share some unique traits, such as the shape of the scape and the pilosity patterns of the tibiae and scapes. A single morphological trait can help distinguish these species; however, only a combination of different approaches can enlighten our view of the complex phylogenetic relationships prevailing in the different populations of these two taxa. Therefore, focusing on the taxonomic issues concerning these two species, a cytogenetic survey including 10 populations of C. rufipes and two populations of C. renggeri was performed. In order to better understand the extent of the relationship between C. rufipes and C. renggeri, two common Neotropical Camponotus species, C. atriceps and C. cingulatus were taken as outgroups. All four species of Camponotus that were studied had 2n = 40 chromosomes (4sm+34st+2t); however, the abundance of chromosome rearrangements observed, combined with several chromosome markers, suggest that C. rufipes and C. renggeri are two good distinct species although closely related. The already reported chromosome translocation 2n = 39 (1m+4sm+32st+2t) for C. rufipes has been found in different populations as in the unprecedented chromosome inversions found both in C. rufipes and in C. renggeri populations. Within the C. renggeri chromosome inversions, both the heterozygous state 2n = 40 (1m+3sm+34st+2t) and the homozygous state, 2n = 40 (2m+2sm+34st+2t) were identified. However, only heterozygous specimens for chromosome inversions were found among C. rufipes, with karyotype configurations distinct from those found in C. renggeri, with 2n = 40 (1m+4sm+34st+2t). None of the populations studied showed signs of mosaic individuals. With respect to rDNA clusters, the 18S rDNA seemed to be more restricted inside the genome, as C. renggeri showed four 18S rDNA clusters, whereas, C. rufipes, C. atriceps, and C. cingulatus showed only two clusters. The chromosome locations of the 5S rDNA clusters were pointed for the first time in Formicidae, and showed itself to be more widely spread over the genome. By combining different chromosome banding approaches it was possible to demonstrate the crucial importance that chromosome inversions played on the karyotype evolution within these ants. The results also showed that chromosome translocations might be a consequence of the chromatin dynamic condition observed among Camponotus species. The homozygosis condition found in a C. renggeri from a Brazilian savanna population for chromosome inversions and the contrasting heterozygous condition for a different kind of chromosome inversion in C. rufipes from the Brazilian coastal rainforest, opens the window for a chromosome race hypothesis within the group C. renggeri and C. rufipes. The wide distribution, rich ecological interactions, genetic diversity, and morphological variability among C. renggeri and C. rufipes justify questioning of the actual taxonomic status of these species. The answer of this puzzle is clear when observing the number of 18S rDNA clusters of these ants, as C. rufipes has only two clusters whereas C. renggeri has four.
Citation: Aguiar HJACd, Barros LAC, Alves DR, Mariano CdSF, Delabie JHC, Pompolo SdG (2017) Cytogenetic studies on populations of Camponotus rufipes (Fabricius, 1775) and Camponotus renggeri Emery, 1894 (Formicidae: Formicinae). PLoS ONE 12(5): e0177702. https://doi.org/10.1371/journal.pone.0177702
Editor: Kunbo Wang, Chinese Academy of Agricultural Sciences Cotton Research Institute, CHINA
Received: July 12, 2016; Accepted: May 2, 2017; Published: May 16, 2017
Copyright: © 2017 Aguiar et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The minimal underlying data set is available within the paper and its Supporting Information file.
Funding: This research was part of the DSc. thesis to author HJACA, and was supported by the Fundação de Amparo à Pesquisa de Minas Gerais (FAPEMIG) process No. APQ-00259-13. The authors wish to thank the Conselho Nacional de Pesquisa (CNPq) for the scholarship granted to HJACA and the research grant to JHCD. We also acknowledge the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES/PNPD) for the postdoctoral grant provided to LACB. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
There are many species of ants that form complexes of species and it is not rare to find several morphotypes (formerly called varieties) within many groups of taxa in a similar context as that discussed by Mayr . It is important then, to distinguish sister species from ecologically or geographically differentiated morphotypes. According to the Biological Species Concept, two “good” species must have well-defined prezygotic or postzygotic barriers. This condition is not verified between morphotypes, races, varieties, or whatever the terminology used . In keeping with this idea, Bickford et al.  defined a cryptic species as a group of two or more morphologically similar species hidden under one single nominal taxon. These authors do not seem to agree with the possibility of hybridism, as cryptic species are “good” species, and the level of morphological relationship would not reflect genetic distinctness. Therefore, the term species complex may be a little slippery because one should first decide if the argument is related to the topic of cryptic species or another kind of individual grouping, based on morphology. In ants the utilization of the terms varieties and races are uncommon –although largely used in other groups. They are deeply studied under the context of population, some of them involving cytogenetic studies .
At present, Camponotus Mayr 1861 is the ant genus with the highest number of described species and subspecies, that is, 1,099 described taxa inserted in 43 subgenera . Taxonomic relationships between subgenera and species complexes of Neotropical Camponotus have been intensively discussed and reviewed by Mackay . Many species of this genus are widely distributed and are found in different biomes . Following data accumulation on the Camponotus species, several discussions involving taxonomic implications became popular, such as, species synonymization [7, 8], or the occurrence of cryptic species complexes . Although Camponotus is considered one of the most prevalent genera  with vast ecological diversity, there is still much to discover about the morphological variability within the valid Camponotus taxa, mainly relating to those with wide geographic distributions. Following this thought, the synonymization of Camponotus rufipes (Fabricius 1775) and C. renggeri Emery 1894, suggested by Mackay , gains attention because both species are abundant in several regions of South America and usually have highly contrasting color patterns. In dealing with the synonymization problem, Mackay  has the following to state about C. rufipes: “This species can be distinguished from others as the scape is flattened at the base (majors and females), the scape and tibiae are covered with erect hairs, the majority of the ant is dark (nearly black) with orange legs, and all surfaces are densely punctate. No other New World ant has this combination of characters. Camponotus renggeri is obviously a synonym [of C. rufipes].”
The Camponotus genus is the richest ant group and is possibly not monophyletic , with the phylogenetic relationship of its species based on morphological traits . Several efforts have been made to organize the Camponotus subgenera into natural groups, but it seems that something is always out of place. The Neotropical subgenus Myrmothrix, which includes C. rufipes, C. renggeri, C. atriceps, and C. cingulatus is considered by Hashmi  to be monophyletic, by using morphological traits. However, Brady et al. , through COI mitochondrial DNA sequences presented a paraphyly possibility between Myrmothrix and Tanaemyrmex. Lastly, Mackay  suggested a new phylogenetic organization for Camponotus species, whereas the group (Tanaemyrmex + Myrmothrix) would be subdivided into natural subgroups, initially considered species complexes. Most of the Myrmothrix species, such as, C. rufipes, C. renggeri, and C. atriceps would belong to the Atriceps species complex, while C. cingulatus (a Myrmothrix species) would be part of the Picipes species complex alongside different species of Tanaemyrmex subgenus. Several taxonomic suggestions proposed by Mackay  are not well supported by the myrmecologists, including the synonymization of C. rufipes and C. renggeri . These two species are commonly sympatric and both share several morphological and behavioral traits. So far, the dull/shiny aspect of the body and the color of the coxae are the main morphological traits that allow distinguishing of the two.
The recent advances from the so-called “integrative taxonomy” represent a promise of solving some issues by combining morphology with different fields [13, 14]. Through cytogenetics, it is possible to investigate not only the amount of genetic variability [15, 16], but also the speciation process itself [5, 17], by finding possible postzygotic barriers [18–20]. Therefore, this field of genetics may be an important tool for better understanding both the complex phylogenetic relationships and the mechanisms involved, with the creation of high levels of variability present within different ant species.
The cytogenetic information available for more than 750 ant morphospecies  help taxonomic discussions, such as those involving cryptic species [22–24]. Only 4% of the valid Camponotus species have any kind of cytogenetic information available  and only recently C. renggeri has been described by cytogenetic means  through chromosome morphology and number. On the other hand, its presupposed sister species, C. rufipes accounts with two distantly related populations studied: one from Uruguay  and the second from southeastern Brazil where a characteristic chromosome polymorphism has been observed .
Camponotus rufipes and C. renggeri sympatrically inhabit some areas from Cerrado (South American savannas) and Mata Atlântica (South American Atlantic coastal rainforests). The chromosome dynamics of these two taxa must be better investigated, to clarify several questions pertinent to recent taxonomic discussions about the boundaries of these species. Therefore, the present study intends to compare the chromosome configuration of different Brazilian populations of C. rufipes and C. renggeri, picking up pieces that may help to solve the complex puzzle which is the real taxonomic status of these species. The possibility of synonymization was investigated by observing the peculiarities of distinct populations from contrasting South American environments, such as Cerrado, Amazon rainforest, and Mata Atlântica (including the sandy open vegetation along the Brazilian coastline), and also the karyotypes of the closely related Camponotus species (subgenus Myrmothrix, sensu Hashmi ): Camponotus atriceps (Fabricius, 1804), and Camponotus cingulatus Mayr, 1862.
Materials and methods
Field work, specimen collecting and nest maintenance
The sampling of specimens was carried out in several distinct Brazilian environments, such as rainforests (Amazon and Mata Atlântica), and Savanna between 2011 and 2015 (Table 1, S1 Table). Two other Camponotus species were also collected: C. atriceps and C. cingulatus. All four species belong to the Myrmothrix subgenus . In this study, C. atriceps and C. cingulatus species were considered external to the clade (C. renggeri + C. rufipes). Both C. rufipes and C. renggeri are widely distributed over South America, and therefore, the samplings were carried out in different localities allowing further environmental comparisons. As this study has qualitative purposes, only the presence of chromosome variations within these populations was observed.
Information connecting species; sample sites; geographic coordinates; biome and diploid chromosome number. Karyotype information: W (wild, 2n = 40), I (inversion, 2n = 40), T (translocation, 2n = 39). Cytogenetic techniques information: BC (C-banding), FL (fluorochromes CMA3/DAPI), NOR (NOR banding), FISH (Fluorescence In Situ Hybridization). Locality details: U—Uruguay; Brazillian States: MG—Minas Gerais, RJ—Rio de Janeiro, MT—Mato Grosso, AP—Amapá, PR—Paraná.
Specimen collection was authorized by the Brazilian Instituto Chico Mendes de Conservação da Biodiversidade (ICMBio) according to a special collecting permit (SISBio accession number 34567–4). Ant vouchers (workers) were deposited in the reference collection of the Laboratório de Mirmecologia, Centro de Pesquisas do Cacau (CPDC/Brazil), under the record #5724.
Parts of colonies of Camponotus were kept alive under laboratory conditions in the Laboratório de Citogenética de Insetos of Universidade Federal de Viçosa, Brazil. Rearing colonies allowed obtaining eggs and larvae in their initial stages of development, aiming to obtain metaphasic chromosomes.
Chromosome preparation, karyotype analysis, banding techniques and FISH
Mitotic metaphases were obtained according to Imai et al. , by dissecting the cerebral ganglia or testes of the larvae (after meconium elimination). A total of 77 nests of C. rufipes, 28 of C. renggeri, two of C. cingulatus and two of C. atriceps were analyzed including about 15 individuals per colony (S1 Table for details). Conventional Giemsa staining was used to determine the chromosome number and morphology.
The metaphases were observed and photographed using a Olympus® BX 60 microscope with a 100X objective, coupled with a Q-Color3 Olympus® image capture system. Chromosomes were measured and organized using Image Pro Plus® and Adobe Photoshop CS®software packages, respectively. Non-overlapping metaphases with similar degrees of condensation were organized by pairing the chromosomes in order of size, measuring and classifying them according to Levan et al. , based on the chromosome arm ratio (r) and organized as metacentric (m), submetacentric (sm), subtelocentric (st) and telocentric (t). This chromosome classification allowed distinguishing minor differences in the chromosome morphology in ants with similar karyotypes [30, 31].
The C-banding technique was performed according to Sumner  with minor clock adaptations to detect heterochromatin suggested by Barros et al. . It was noteworthy that the heterochromatin pattern could also be observed using Giemsa staining as a result of structural heterochromatin differences .
Metaphases of individuals from different localities were stained with sequential fluorochromes Chromomycin A3 (CMA3) and 4’,6-diamidin-2-phenylindole (DAPI), to detect regions rich in GC and AT base pairs, respectively, according to the protocol of Schweizer . The metaphases were analyzed with an epifluorescence microscope using the filters WB (450–480 nm) and WU (330–385 nm) for CMA3 and DAPI, respectively.
The NOR banding technique was carried out to detect Nucleolus Organizer Regions (NORs) according to Howell and Black . Additionally, ribosomal gene clusters were detected by Fluorescence In Situ Hybridization (FISH) according to Pinkel et al.  with the use of 18S rDNA probe  and 5S rDNA , isolated from the bee Melipona quinquefasciata and the ant C. rufipes, respectively. The probes 18S and 5S rDNA were labeled by an indirect method using digoxigenin-11-dUTP (Roche Applied Science), and the signal was detected with antidigoxigenin-rhodamine (Roche Applied Science). The metaphases were analyzed with an epifluorescence microscope using WU (330–385 nm) and WG filters (510–550 nm), for DAPI and rhodamine, respectively.
All the populations studied showed a diploid chromosome number of 2n = 40, including the species C. atriceps and C. cingulatus, used as outgroup. The karyotype formula observed for the four species was 2n = 40 (4sm+34st+2t) (Fig 1). However, chromosomal variation in structure and number were seen (Fig 2, Table 1).
a) C. atriceps from Lavras; b) C. cingulatus from Viçosa; c) C. cingulatus from Rio de Janeiro; d) C. renggeri from Macapá; e) C. renggeri from Nova Mutum; f) C. rufipes from Lavras; g) C. rufipes from Viçosa; h) C. rufipes from Rio de Janeiro; i) C. rufipes from Petrópolis; j) C. rufipes from Ponte Nova; k) C. rufipes from Ubá; l) C. rufipes from Curitiba; m) C. rufipes from Urucânia. Bar = 5μm.
a) C. renggeri with heterozygous chromosome inversion from Nova Mutum (2n = 40); b) C. renggeri with homozygous chromosome inversion from Nova Mutum (2n = 40); c) C. rufipes with heterozygous chromosome translocation from Lavras (2n = 39); d) C. rufipes with heterozygous chromosome translocation from Ponte Nova (2n = 39); e) C. rufipes with heterozygous chromosome inversion from Ponte Nova (2n = 40); f) C. rufipes with heterozygous chromosome translocation from Ubá (2n = 39); g) C. rufipes with heterozygous chromosome translocation from Viçosa (2n = 39); h) C. rufipes with heterozygous chromosome translocation from Curitiba (2n = 39). ✳—Rearranged chromosomes. Bar = 5μm.
In the populations of C. rufipes from the following localities: Lavras, Ponte Nova, Ubá, Viçosa, and Curitiba it was observed individuals with 2n = 39 (1m+4sm+32st+2t) (Fig 2c, 2d, 2f, 2g and 2h), a condition that characterizes a chromosome polymorphism. This will henceforth be denominated as “type I polymorphism”, hypothesized here as a Robertsonian translocation. Another kind of chromosome rearrangement was detected within C. rufipes in the population of Ponte Nova (Fig 2e) and also in C. renggeri at Nova Mutum (Fig 2a and 2b). This polymorphism was not yet described and was characterized by chromosome inversion maintaining the diploid number of wild individuals (2n = 40) changing the morphology of one of their chromosomes. It will be denominated “type II polymorphism” from now on. At Nova Mutum, nests of C. renggeri presented a heterozygous condition with 2n = 40 chromosomes (1m+3sm+34st+2t) (Fig 2a); but one colony showed the type II polymorphism in homozygous condition with 2n = 40 chromosomes (2m+2sm+34st+2t) (Fig 2b). In C. rufipes, only heterozygous individuals for this kind of rearrangement were observed with the karyotype formulae 2n = 40 (1m+4sm+33st+2t) (Fig 2e). The trait of mosaicism, the presence of two different chromosome configurations in an individual, was not found among any of the Camponotus individuals studied. Males of C. cingulatus (Rio de Janeiro population), C. renggeri (Macapá population) and C. rufipes (Rio de Janeiro, Ubá, Viçosa, and Curitiba populations) showed haploid karyotype n = 20 (2sm+17st+1t). Males of C. rufipes (Viçosa population) showed haploid karyotypes with n = 19 (1m+2sm+15st+1t), bearing the type I polymorphism (Fig 3). Polymorphisms involving the morphology of the chromosomes in C. atriceps and C. cingulatus were not observed. The four Camponotus species of the present study showed minor morphological differences between the homologous chromosomes bearing the NORs. This kind of heteromorphic condition is found, to a greater or lesser extent, in differently studied populations, as observed in Fig 1. Apart from this, an accentuated heteromorphic condition involving markings of CMA3 and clusters of 18S rDNA were observed in C. cingulatus.
a) C. cingulatus from Rio de Janeiro; b) C. renggeri from Macapá; c) C. rufipes from Rio de Janeiro; d) C. rufipes from Ubá; e) C. rufipes from Curitiba; f) C. rufipes with heterozygous translocation from Viçosa (n = 19). The box points a chromosome result of a translocation. Bar = 5μm.
The heterochromatin pattern obtained with C-banding, and also with Giemsa staining, showed itself to be restricted to the pericentromeric region of all chromosomes, especially in their long arms (Fig 4). Some interstitial and discontinuous blocks of heterochromatin could be seen on longer chromosomes when less condensed.
a) C. atriceps from Lavras; b) C. cingulatus from Rio de Janeiro (haploid male); c) C. renggeri from Nova Mutum; d) C. rufipes from Lavras. Bar = 5μm.
Regions rich in GC base pairs using the CMA3 were observed in C. renggeri on the second submetacentric pair, and also on the subtelocentric pair of medium size from Macapá and Nova Mutum populations (Fig 5c). This submetacentric chromosome pair was GC-rich on the pericentromeric region of the short arm major extension, including a secondary constriction. The subtelocentric pair showed its CMA3+ region in the totality of the short arm. All the populations of C. atriceps (Fig 5a), C. cingulatus (Fig 5b), and C. rufipes (Fig 5d) had only one pair of chromosomes, the second submetacentric pair, bearing GC-rich regions. Even in the metaphases bearing rearrangements, the difference in the GC marking pattern between C. rufipes and C. renggeri was observed, indicating that they were not involved with the reported rearrangements (Fig 6). However, C. rufipes from Urucânia (MG) showed three chromosomes with blocks of rich chromatin in GC base pairs (Fig 6c and 6d). The relationship among CMA3+ regions and clusters of 18S rDNA was confirmed by the FISH technique, which showed four markings in C. renggeri (Fig 7c) and two, both in C. cingulatus and in C. rufipes (Fig 7a and 7e). A heteromorphic condition observed in C. cingulatus was confirmed by the observation of 18S rDNA genes (Fig 7a).
(a, b) Camponotus rufipes; (c, d) C. renggeri; (e, f) C. atriceps; (g, h) C. cingulatus stained with DAPI and CMA3 respectively. Arrows point GC-rich/AT-poor regions. Bar = 5μm.
All images are stained respectively with DAPI and CMA3. (a, b) C. renggeri with homozygous inversion (2n = 40); (c, d) C. rufipes with heterozygous translocation or hybridism (2n = 40); (e, f) C. renggeri with heterozygous inversion (2n = 40); (g, h) C. rufipes with heterozygous translocation (2n = 40). Arrows point GC-rich/AT-poor regions. Arrowheads point the rearranged chromosome. Bar = 5μm.
(a, b) C. cingulatus from Viçosa; (c, d) C. renggeri from Nova Mutum; (e, f) C. rufipes from Viçosa showing 18S rDNA and 5S rDNA respectively; g) C. renggeri from Macapá showing 18S rDNA clusters; h) C. rufipes with heterozygous chromosome translocation (2n = 39) from Viçosa showing 18S rDNA clusters. Bar = 5μm.
The detection of 5S probe showed multiple marks at the pericentromeric region of all the chromosomes of C. renggeri, C. rufipes and C. cingulatus (Fig 7b, 7d and 7f). Similar results were obtained using the NOR banding technique with silver nitrate staining (Fig 8).
Searching the taxonomic borders of complex biological units through cytogenetic surveys has demonstrated significant advancement in better understanding species complexes [23, 39], as chromosome variations are potentially strong postzygotic barriers [40, 41]. The chromosome configuration of all the four Camponotus species studied here suggests a phylogenetic closeness. The novel type II polymorphism in both studied populations resulted from inversion rearrangements, but with different origins. In C. renggeri, the origin of the rearranged chromosome was from the higher submetacentric to the metacentric. In C. rufipes, the putative origin of the rearranged chromosome was from one subtelocentric to the metacentric. These mutated metacentric chromosomes were easily detected, as the four species of Camponotus studied here had a majority of their chromosomes with terminal centromeres.
The patterns of heterochromatin do not show measurable variations among populations or taxa. The majority of chromosomes have minor amounts of pericentromeric heterochromatin and also small blocks in the short arms of the chromosomes. Small interstitial blocks of heterochromatin dispersed within the long arms of the largest chromosomes were observed in the four Camponous species. As these interstitial heterochromatic blocks are not continuous in the chromosome long arms, the chromosome evolution of this group of ants may not be restricted to the fission cycle followed by tandem heterochromatin growth, such as that described by Imai et al. . The uniformity of the chromosome numbers within the Myrmothrix subgenus clarifies the absence of chromosome fission/fusion, but chromosome inversions may have a great impact on karyotype evolution in these ants.
The combined patterns of C-banding, NOR-banding and 5S FISH technique helped in the understanding of the terminal heterochromatin of the species studied. Both C-banding and NOR-banding indicated that the terminal heterochromatin regions are argentophilic. This characteristics has been reported by Sumner  and also recently by Cristiano et al.  on the Anthophoridae bee Melitoma segmentaria (Fabricius, 1804). However, for the first time, this trait was linked to the presence of 5S rDNA clusters. As this is the first study, to our knowledge, which has successfully detected 5S rDNA clusters in karyotypes of Formicidae, further observations will be useful in understanding this special pattern of heterochromatin.
The 18S rDNA clusters of C. rufipes, C. renggeri, and C. cingulatus have been carried out through the observation of GC-rich chromatin regions. These clusters were further confirmed by the FISH technique. The studied populations of C. renggeri have four clusters of 18S rDNA, whereas all populations of the species C. rufipes and C. cingulatus have only two clusters of this kind of 18S rDNA. Among eukaryotes, the GC-rich chromatin usually corresponds to the NORs , as successfully reported in other ant species, where the FISH and NOR banding were performed [46–49]. The type I chromosome rearrangement is probably not an isolated event among C. rufipes, and was detected in different populations. Chromosome translocations may be connected to the origin of the additional 18S rDNA pair of the C. renggeri. This hypothesis is supported by the absence of a secondary constriction in the chromosome pair, which carries both the extra 18S rDNA clusters and the GC-rich chromatin. Among C. rufipes, C. atriceps, and C. cingulatus, only a single chromosome pair has a secondary constriction, and this same chromosome pair also carries both the 18S rDNA clusters and the GC-rich chromatin. Chromosome heteromorphic conditions connected to the NORs are not uncommon (e.g., [50, 51]).
According to the Minimum Interaction Theory (MIT) [28, 42], the evolution of chromosome morphology is directional and the selection is disruptive due to the selection against the heterozygous individuals. High numbers of chromosomes with small sizes would reduce the possibility of deleterious chromosome interactions inside the nucleus, assuming the consistency of the interphase nucleus volume, and that the chromosomes remain anchored to the nuclear membrane by their telomeres [52, 53]. These chromosome interactions produce at first heterozygous individuals. Imai  indicates that the growth of the heterochromatin blocks play a central role in changing the chromosome morphology, but none of the four Camponotus species studied have any large heterochromatin blocks at the chromosome telomeres.
It is suggested that both paracentric and pericentric inversions occur simultaneously with chromosome evolution in this group of species. According to Hoffmann et al.  and Hoffmann and Rieseberg  gene clusters that display epistatic interactions tend to remain together. Therefore, the fitness of a karyotype which undergoes chromosome inversion depends on the protection of such a group of genes from crossing-over. This seems to be a hypothesis to explain the absence of metacentric chromosomes among the species of ants from Myrmothrix subgenus, a chromosome configuration that allows the fission-inversion chromosome cycle to continue, as predicted by the MIT. The abundance of chromosomes with terminal centromeres may be a result of specific epistatic interactions between genes or groups of genes present in the long arms of the chromosomes. An alternative explanation could be that it is due to the centromere drive . This karyotype pattern is completely different from that observed in the ant Camponotus (Myrmobrachys) crassus Mayr, 1862 (2n = 20) characterized by lower chromosome numbers, all of them metacentric [25, 27]. This demonstrates the rich chromosome diversity and the distinct array of evolutionary chromosome paths existing inside the Camponotus genus.
The chromosome rearrangements detected in the present study, translocations or inversions, are potentially able to enhance the genetic variability among the populations of these ant species. The occurrence of males with such rearrangements only increases the possibilities of chromosome exchange between populations (Fig 3f). However, these rearrangements lead to an inconvenient problem, because heterozygotes that bear these rearranged chromosomes would normally produce only 50% of balanced gametes. Some rodent species show high hybridism levels, leading to an F1 heterozygous for Robertsonian translocations. The negative effects on the fertility among them are variable but not necessarily high [58, 59]. The nests of C. rufipes and C. renggeri are large and populous and according to Ronque et al. , they are polygynous, which means that workers of a single colony can be the offspring of different reproductive females. Therefore, the nest would be able to survive even with heterozygous queens, producing heterozygous alates for chromosome rearrangements. However the effects on the fertility among reproductive castes of Camponotus species bearing rearranged chromosomes are yet unknown and should be investigated in future studies.
Camponotus rufipes and C. renggeri exist in sympatry in different localities over the range of their distribution, and at Urucânia, the chromosome marker CMA3/DAPI showed individuals with intermediary cytogenetic configurations. C. rufipes have a single pair of chromosomes marked with CMA3, whereas C. renggeri have two pairs marked. Intermediary individuals presented three chromosomes marked with this fluorochrome. These individuals were collected in nests with ants morphologically identified as C. rufipes. There are two hypotheses to explain this new finding: the first one suggests that the queen of the nest bears a third kind of chromosome rearrangement in a heterozygous condition, characterized by a chromosome translocation involving the rDNA 18S cluster; the second hypothesis deals with hybridization events. Unfortunately, it was not possible to obtain cytogenetic data of C. renggeri from Urucânia, which means these questions need to be discussed in further studies. Besides, Ronque et al  did not find data that leads to the hybridism hypothesis, and therefore, further approaches in different sympatric populations may enlighten this discussion.
Nowadays, Camponotus is the ant genera with the highest number of species closely followed by the Myrmicinae Pheidole, according to Bolton , and many of Camponotus, like the four species studied in this survey, are widely distributed over extensive geographic areas. The discussion about the synonymization of C. renggeri and C. rufipes is a result of the accumulation of morphological data on several populations. These discussions reveal the presence of “lumper” behaviors [60, 61], uncommon among myrmecologists since the emphatic study of Wilson & Brown . Formicidae is considered an “ultra-diverse” group, and some may believe that the pursuit of the 20,000 valid species proposed by Hölldobler & Wilson  may actually represent what Mallet  called “taxonomic inflation”. Comprehending the many Camponotus taxa by organizing them in species complexes seems welcome, mainly when we focus on species with high heritable variability and large geographic distribution, such as C. rufipes and C. renggeri. The cytogenetic variability and population dynamics of these two species can be compared with those found in some well-known chromosome races of rodents, in which there is a possibility of genetic interchange, but with constraints , as nests with the type II rearrangement in homozygosis were found at Nova Mutum. The possibility of chromosome races within and between these taxa may also be connected to the origin of all rearrangements described in this study. This hypothesis encompasses not only the possibility of hybridization, but also deals with the role played by chromosome rearrangements in the formation of postzygotic barriers during the speciation process.
Different approaches using morphology , ecology, and molecular techniques  deal with the taxonomic problem of C. rufipes and C. renggeri, and now by observing the number of chromosomes bearing 18S rDNA clusters, it is possible to detect a discrete heritable trait, which is able to differentiate these two Camponotus species in different geographic locations and contrasting environments.
We would like to thank Gisele Amaro Teixeira for the technical support provided to us.
This research was part of the DSc. thesis of the first author, and was supported by Fundação de Amparo à Pesquisa de Minas Gerais (FAPEMIG) process No. APQ-00259-13. The authors wish to thank the Conselho Nacional de Pesquisa (CNPq) for the scholarship granted to HJACA and the research grant to JHCD. We also acknowledge Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES/PNPD) the postdoctoral grant provided to LACB
- Conceptualization: HJACA LACB JHCD CSFM SGP.
- Data curation: JHCD.
- Formal analysis: HJACA LACB SGP.
- Funding acquisition: SGP.
- Investigation: HJACA LACB DRA SGP.
- Methodology: HJACA LACB DRA.
- Project administration: HJACA LACB SGP.
- Resources: HJACA LACB DRA JHCD CSFM SGP.
- Supervision: HJACA LACB SGP.
- Visualization: HJACA LACB.
- Writing – original draft: HJACA LACB CSFM JHCD SGP.
- Writing – review & editing: HJACA LACB.
- 1. Mayr E. Populations, species, and evolution: an abridgment of animal species and evolution. Harvard University Press; 1970.
- 2. Mayr E. What Evolution Is? Clays ltd, St Ives pic Press, Great Britain; 2001.
- 3. Bickford D, Lohman DJ, Sodhi NS, Ng PK, Meier R, Winker K, et al. Cryptic species as a window on diversity and conservation. Trends Ecol Evol. 2007; 22(3): 148–155. dx.doi.org/10.1016/j.tree.2006.11.004 pmid:17129636
- 4. Wilson EO, Brown WL The subspecies concept and its taxonomic application. Syst Zool. 1953; 2(3): 97–111.
- 5. King M. Species Evolution: The role of Chromosome Change. Cambridge University Press. Cambridge; 1993.
- 6. Bolton B. An online catalog of the ants of the world. http://antcat.org. (accessed on 05 May 2016); 2016.
- 7. Mackay WP. The systematics and biology of the New World carpenter ants of the hyperdiverse genus Camponotus (Hymenoptera: Formicidae), The University of Texas, http://www.utep.edu/leb/ants/Camponotus.htm (accessed 21 January 2015); 2004
- 8. Hashmi AA. A revision of the Neotropical ant subgenus Myrmothrix of genus Camponotus. Studia Entomol. 1973; 16: 1–140.
- 9. Groc S, Orivel J, Dejean A, Martin JM, Etienne MP, Corbara B, et al. Baseline study of the leaf-itter ant fauna in a French Guianese forest. Insect Conserv Divers. 2009; 2(3): 183–193.
- 10. Wilson EO. Which are the most prevalent ant genera? Studia Entomol. 1976; 19: 187–200.
- 11. Brady SG, Gadau J, Ward PS. Systematics of the ant genus Camponotus (Hymenoptera: Formicidae): A preliminary analysis using data from the mitochondrial gene cytochrome oxidase I. In: Austin A & Dowton M (ed). Hymenoptera: Evolution, Biodiversity and Biological Control. Csiro Publishing London; 2000.
- 12. Ronque MUV, Azevedo-Silva M, Mori GM, Souza AP, Oliveira PS. Three ways to distinguish species: using behavioural, ecological, and molecular data to tell apart two closely related ants, Camponotus renggeri and Camponotus rufipes (Hymenoptera: Formicidae) Zool J Linnean Soc. 2015; 176: 170–181.
- 13. Seifert B. Cryptic species in ants (Hymenoptera: Formicidae) revisited: we need a change in the alpha-taxonomic approach. Myrmecol News. 2009; 12: 149–166.
- 14. Schlick-Steiner BC, Steiner FM, Seifert B, Stauffer C, Christian E, Crozier RH. Integrative taxonomy: a multisource approach to exploring biodiversity. Annu Rev Entomol. 2010; 55: 421–438. pmid:19737081
- 15. John B, Lewis KR. Chromosome variability and geographic distribution in insects. Science. 1966; 152: 711–721. pmid:17797432
- 16. Dubinin NP, Sokolov NN, Tiniakov GG. Intraspecific chromosome variability. Comp Cytogenet. 2010; 4: 207–214.
- 17. White MJD. Animal Cytology and Evolution. 3rd Ed. Cambridge University Press, Cambridge, Massachusetts; 1978.
- 18. Shaw DD, Lightfood DC, Weissman DB Chromosomes as isolating mechanisms in Trimerotropine grasshoppers. J Orthoptera Res. 1998; 7: 157–163.
- 19. Perrin A, Morel F, Douet-Guilbert N, Le Bris MJ, Amice J, Amice V, et al. A study of meiotic segregation of chromosomes in spermatozoa of translocation carriers using fluorescent in situ hybridization. Andrologia. 2010; 42: 27–34. pmid:20078513
- 20. Kouru KH, Malmgreen H, White I, Blennow E. Hidden mosaicism for a structural chromosome rearrangement: a rare explanation for recurrent miscarriages and affected offspring? Fertil Steril. 2011; 95: 806–808. pmid:20951377
- 21. Lorite P, Palomeque T. Karyotype evolution in ants (Hymenoptera: Fomicidae), with a review of the known ant chromosome numbers. Myrmecol News. 2010; 13: 89–102.
- 22. Mariano CS, Pompolo SG, Borges DS, Delabie JHC. Are the Neotropical ants Pachycondyla crenata (Roger, 1861) and Pachycondyla mesonotalis (Santschi, 1923) (Hymenoptera: Formicidae: Ponerinae) good species? A cytogenetic approach. Myrmecol Nachr. 2006; 8: 277–280.
- 23. Mariano CSF, Pompolo SG, Silva JG, Delabie JHC. Contribution of cytogenetics to the debate on paraphyly of Pachychondyla spp. (Hymenoptera, Formicidae, Ponerinae). Psyche. 2012; 1–9.
- 24. Delabie JHC, Mariano CSF, Mendes LF, Pompolo SG, Fresneau D. Problemas apontados por estudos morfológicos, ecológicos e citogenéticos no gênero Pachycondyla na região neotropical: o caso do complexo apicalis. In: Vilela EF, Santos IA, Schoereder JH, Neto JL, Serrão JE, Campos LAO, Eds. Insetos Sociais: Da Biologia à Aplicação. Viçosa: Editora UVF; 2008. p. 197–222.
- 25. Aguiar HJAC, Barros LAC, Soares FAF, Carvalho CR, Pompolo SG. Estimation of nuclear genome size of three species of Camponotus (Mayr, 1861) (Hymenoptera: Formicidae: Formicinae) and their cytogenetic relationship. Socobiology. 2016; 63(2): 777–782.
- 26. Goñi B, Zolessi LC, Imai HT. Karyotypes of thirteen ant species from Uruguay (Hymenoptera, Formicidae). Caryologia. 1983; 36: 363–371.
- 27. Mariano CSF, Pompolo SG, Delabie JHC, Campos LAO. Estudos cariotípicos de algumas espécies neotropicais de Camponotus Mayr (Hymenoptera: Formicidae). Rev Bras Entomol. 2001; 45: 267–274.
- 28. Imai HT, Taylor RW, Crosland MWJ, Crozier RH. Modes of spontaneous chromosomal mutation and karyotype evolution in ants with reference to the minimum interaction hypothesis. Jpn J Genetics. 1988; 63: 159–185.
- 29. Levan A, Fredga K, Sandberg AA. Nomenclature for centromeric position on chromosomes. Hereditas. 1964; 52: 201–220.
- 30. Barros LAC, Teixeira GA, Aguiar HJAC, Mariano CSF, Delabie JHC, Pompolo SG. Banding patterns of three leafcutter ant species of the genus Atta (Formicidae: Myrmicinae) and chromosomal inferences. Flo Entomol. 2014; 97(4): 1694–1701.
- 31. Barros LAC, Aguiar HJAC, Mariano CSF, Andrade-Souza V, Costa MA, Delabie JHC, et al. Cytogenetic data on six leafcutter ants of the genus Acromyrmex Mayr, 1865 (Hymenoptera, Formicidae, Myrmicinae): insights into chromosome evolution and taxonomic implications. Comp Cytogenet. 2016; 10(2): 229–243. pmid:27551345
- 32. Sumner AT. A simple technique for demonstrating centromeric heterochromatin. Exp Cell Res. 1972; 75: 304–306. pmid:4117921
- 33. Barros LAC, Mariano CSF, Pompolo SG. Cytogenetic studies of five taxa of the tribe Attini (Formicidae: Myrmicinae). Caryologia. 2013; 66: 1–5.
- 34. Schweizer D. Simultaneous fluorescent staining of R bands and specific heterocromatic regions (DA/DAPI-bands) in human chromosomes. Cytogenet Cell Genet. 1980; 27: 190–193. pmid:6156801
- 35. Howell WM, Black DA. Controlled silver-staining of nucleolus organizer regions with a protective colloidal developer: a 1-step method. Experientia. 1980; 36:1014–1015. pmid:6160049
- 36. Pinkel D, Straume T, Gray JW. Cytogenetic analysis using quantitative, high-sensitivity, fluorescence hybridization. Proc Natl Acad Sci USA. 1986; 83: 2934–2938. pmid:3458254
- 37. Pereira JOP, Diversidade genética da abelha sem ferrão Melipona quinquefasciata baseada no seqüenciamento das regiões ITS1 parcial e 18S do DNA ribossômico nuclear, D.Sc. Thesis, Programa de Pós-graduação em Zootecnia, Universidade Federal do Ceará, Fortaleza-CE, Brazil, 2006.
- 38. Martins C, Ferreira IA, Oliveira C, Foresti F, Galetti PM Jr. A tandemly repetitive centromeric DNA sequence of the fish Hoplias malabaricus (Characiformes: Erythrinidae) is derived from 5S rDNA. Genetica. 2006; 127(1–3): 133–141. pmid:16850219
- 39. Kavalco KF, Brandão KO, Pazza R, Almeida-Toledo LF. Astyanax hastatus Myers 1928 (Teleostei, Characidae): A new species complex within the genus Astyanax? Genet Mol Biol. 2009; 32: 477–483. pmid:21637509
- 40. Dobzhansky T, Pavlovsky O. An extreme case of heterosis in a Central American population of Drosophila tropicalis. Proc Natl Acad Sci USA. 1955; 41: 289–295. pmid:16589666
- 41. Navarro A, Barton H. Accumulating postzygotic isolation genes in parapatry: a new twist on chromosomal speciation. Intern Evolution. 2003; 57: 447–459.
- 42. Imai HT, Taylor RW, Crozier RH. Experimental bases for the minimum interaction theory. I. Chromosome evolution in ants of the Myrmecia pilosula species complex (Hymenoptera: Formicidae: Myrmeciinae). Jpn J Genetics. 1994; 69: 137–182.
- 43. Sumner AT. Chromosome Banding. Hyman U. ed. London, Boston, Sydney, Wellington, 1990.
- 44. Cristiano MP, Simões TG, Lopes DM & Pompolo SG. Cytogenetics of Melitoma segmentaria (Fabricius, 1804)(Hymenoptera, Apidae) reveals differences in the characteristics of heterochromatin in bees. Comp Cytogenet. 2014; 8(3): 223. pmid:25349673
- 45. Phillips RB, Reed KM. Localization of repetitive DNAs to zebrafish (Danio rerio) chromosomes by fluorescence in situ hybridization (FISH). Chromosome Res. 2000; 8: 27–35. pmid:10730586
- 46. Mariano CSF, Pompolo SG, Barros LAC, Mariano-Neto E, Campiolo S, Delabie JHC. A biogeographical study of the threatened ant Dinoponera lucida Emery (Hymenoptera: Formicidae: Ponerinae) using a cytogenetic approach. Insect Conserv Divers. 2008; 1: 161–168.
- 47. Barros LAC, Mariano CSF, Pompolo SG, Delabie JHC. Hsc-FA and NOR bandings on chromosomes of the giant ant Dinoponera lucida Emery, 1901 (Hymenoptera: Formicidae). Comp Cytogenetics. 2009; 3: 97–102. http://dx.doi.org/10.3897/compcytogen.v3i2.16
- 48. Lorite P, Aránega AE, Luque F, Palomeque T. Analysis of the nucleolar organizing regions in the ant Tapinoma nigerrimum (Hymenoptera, Formicidae). Heredity. 1997; 78: 578–582. pmid:9203352
- 49. Barros LAC, Aguiar HJAC, Teixeira GA, Mariano CSF, Teixeira MC, Delabie JHC, Pompolo SG. Cytogenetic data on the threatened leafcutter ant Atta robusta Borgmeier, 1939 (Formicidae: Myrmicinae: Attini). C R Biol. 2015; 338: 660–665. pmid:26315727
- 50. Cardoso DC, Cristiano MP, Barros LAC, Lopes DM, Pompolo SG. First cytogenetic characterization of a species of the arboreal ant genus Azteca Forel, 1978. Comp Cytogenet. 2012; 62: 107–114.
- 51. Boisvert FM, Koningsbruggen SV, Navascués J, Lamond AI. The multifunctional nucleolus. Nature. 2007; 8: 574–585.
- 52. Okazaki S, Tsuchida K, Maekawa H, Ishikawa H, Fujiwara H. Identification of pentanucleotide telomeric sequence, (TTAGG)n, in the silkworm Bombyx mori and in other insects. Mol Cell Biol. 1993; 13:1424–1432. pmid:8441388
- 53. Rothballer A, Kutay U. The diverse functional LINCs of the nuclear envelope the cytoskeleton and chromatin. Chromosoma. 2013; 122: 415–429. pmid:23736899
- 54. Imai HT. Mutability of constitutive heterochromatin (C-bands) during eukaryotic evolution and their cytological meaning. Jpn J Genetics. 1991; 66: 635–661.
- 55. Hoffmann AA, Sgrò CM, Weeks AR. Chromosomal inversion polymorphisms and adaptation. Trends Ecol Evol. 2004; 19: 482–488. pmid:16701311
- 56. Hoffmann AA, Rieseberg LH. Revisiting the impact of inversions in evolution: from population genetic markers to drivers of adaptive shifts and speciation? Annu Rev Ecol Evol Syst. 2008; 39: 21–42. pmid:20419035
- 57. Malik HS. The Centromere-Drive Hypothesis: A Simple Basis for Centromere Complexity. Prog Mol Subcell Biol. 2009; 48: 33–52. pmid:19521811
- 58. Castiglia R, Corti M, Tesha P, Scanzani A, Fadda C, Capanna E, et al. Cytogenetics of the genus Arvicanthis (Rodentia, Muridae) 3. Comparative cytogenetics of A. neumanni and A. nairobae from Tanzania. Genetica. 2003; 118: 33–39. pmid:12733662
- 59. Sharma T, Bardhan A, Bahadur M. Reduced meiotic fitness in hybrids with heterozygosity for heterochromatin in the speciating Mus terricolor complex. J Biosci. 2003; 28(2): 189–198. pmid:12711811
- 60. Kell LT, Dickey-Collas M, Hintzen NT, Nash RDM, Pilling GM, Roel BA. Lumpers or Splitters? Evaluating recovery and management plans for metapopulations of herring. ICES J Mar Sci ICES. 2009; 66: 1776–1783.
- 61. Bertrand Y, Pleijel F, Rouse GW. Taxonomic surrogacy in biodiversity assessments, and the meaning of Linnaean ranks. System Biodivers. 2006; 4: 149–159.
- 62. Hölldobler B, Wilson EO. The superorganism: the beauty, elegance, and strangeness of insect societies. WW Norton & Company; 2009.
- 63. Mallet J. Concepts of species. Encyclopedia of Biodivers. 2007; 1–15.