Material

Table 1 summarizes the geographic origin, relevant cytogenetic traits and GISH results of the material here analyzed. No specific permissions were required for insect collections performed in this work, and did not involve endangered or protected species.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Geographic origin, relevant cytogenetic traits and GISH results of the eleven Triatominae species analyzed by three genomic probes.

https://doi.org/10.1371/journal.pone.0114298.t001

Chromosome preparations for GISH analyses were obtained from males of 11 triatomine species, included in the two principal tribes of the subfamily: Rhodniini and Triatomini (Table 1), which involve almost 90% of the 140 recognized species [17]. For Rhodniini tribe, which included 19 species in 2 genera, we analyzed one species: Rhodnius prolixus. For Triatomini tribe, which involved 104 species in eight genera, we studied species from the following five genera: Dipetalogaster, Eratyrus, Mepraia, Panstrongylus and Triatoma. For the Triatoma genus, we studied the three main clades or groups: a) the Rubrofasciata Group (from Central and North America and Old World species): T. protracta and T. dimidiata, b) the Dispar Group (west of the Amazon region): Triatoma carrioni and c) the Infestans Group (from south and east of the Amazon region). For this group we included three closely related species of the infestans subcomplex (T. delpontei, T. infestans and T. platensis). In T. infestans we analyzed the two chromosomal groups: Andean and non-Andean with striking differences in the amount and chromosome distribution of the C-heterochromatin, both in autosomes and sex chromosomes (see Table 1).

Chromosome preparations and C-banding

For cytological preparations, testes were removed from adult insects alive, fixed in an ethanol–glacial acetic acid mixture (3∶1) and stored at -20°C. Squashes were made in a 50% acetic acid drop, coverslips were removed after freezing in liquid nitrogen and the slides were air dried and then stored at 4°C. C-banding was performed according to Panzera et al. [8]. The analysis of C-banded preparations was made using a Nikon Eclipse 80i microscope with a DS-5Mc-U2 digital camera.

Genomic DNA Isolation and probe labeling for GISH techniques

Three genomic DNA probes were used from two species. Each one isolated from one adult individual: one from a male of T. infestans (Andean group) collected in Bolivia (Potosí, Palquiza, sylvatic) and the others two probes were obtained from one male and one female of T. delpontei collected from Bolivia (Santa Cruz, Tita, sylvatic). These two species have the highest DNA content in triatomines: T. delpontei (2.90 pg) and T. infestans -Andean group- (1.98 pg) [3]. In both species, C-heterochromatin amount represent approximately 45% of the autosomal complement [18], but with different chromosome localization (see Table 1).

Genomic DNA was purified from adult legs following the NucleoSpin Tissue kit (MACHEREY-NAGEL). For the probes, total genomic DNAs were labeled with biotin-16-dUTP (Roche) using a Nick Translation Kit (Roche), following manufacturer's instructions.

In situ hybridization was carried out as described previously [19]. Hybridization solutions were prepared to a final concentration of 0.5–2 ng probe/mL in 50% formamide. Hybridization was conducted at 37°C overnight. Fluorescence immunological detection was performed using the avidin-FICT/anti-avidin-biotin system with two amplification rounds. Slides were mounted with Vectashield (Vector). DAPI in the antifade solution was used to counterstain chromosomes. The hybridized chromosomes were observed and photographed using a BX51 Olympus fluorescence microscope equipped with a CCD camera (Olympus DP70) and merged using the DPManager software. Hybridization pattern for each species was determined by the chromosomal analyses of at least two individuals.

Self-GISH or Auto-GISH. Genomic DNA from T. delpontei (male or female) on T. delpontei male meiotic chromosomes

All chromatin present scattered hybridization signals. In addition 9 of the 10 bivalents and the X chromosome exhibit strongest hybridization signals in only one chromosomal end, while the Y chromosome appears almost totally hybridized (Figures 1a to 1c). This T. delpontei individual does not have C-block in one autosomal bivalent (data not shown), consequently this chromosome do no have hybridization signals on the chromosomal ends (Figures 1b and 1c). Male meiotic prophase clearly shows the two hybridization configurations: a large heteropycnotic chromocenter with a strong hybridization signal and the rest of the chromatin with weaker signals (Figures 1d and 1e). This meiotic chromocenter is formed by the heterologous association of the C-heteropycnotic regions of all autosomal bivalents and both sex chromosomes [18]. In metaphase II, each autosomal pair has a strong signal in only one chromosomal end and both sex chromosomes (X and Y) appear labeled (Figure 1f). This GISH pattern is very similar compared with those observed with C-banding: figure 1g shows a C-banded metaphase I with a C-block in only one chromosomal end of all bivalents while both sex chromosomes are C-heterochromatic (compare Figure 1c and 1g).

Self-GISH or Auto-GISH. Genomic DNA from T. infestans Andean Group (male) on T. infestans Andean Group male meiotic chromosomes

All chromosomes show hybridization signals but with differences in size and intensity. The strongest signals are observed in the terminal regions of 6 bivalents, in the whole Y chromosome and more scarcely on the X chromosome (Figure 1h). Also in this case the hybridization pattern is very similar to the C-banding pattern: T. infestans (Andean group) exhibit 7–9 autosomal bivalents with C-blocks of different size in one or both chromosomal ends. Both sex chromosomes (X and Y) appear C-heterochromatic (Figure 1i).

Inter-specific GISH

Genomic probes produce a specific pattern on each one of the 11 species analyzed. However, since each probe recognizes the same repeated sequences, the chromosome hybridizations results on each species are the same with the three applied probes. For this reason, to illustrate the patterns of the 11 analyzed species we combine different genomic probes in figures 1, 2 and 3. Genomic DNA probe of T. infestans (Andean group) on T. delpontei chromosomes and the reverse hybridization, i.e., T. delpontei gDNA probe on T. infestans (Andean group) chromosomes shows the same patterns described in the corresponding self-GISH.

Genomic DNA probe of T. infestans (Andean group) on T. infestans (non-Andean group) chromosomes shown strong hybridization signals on the Y chromosome (almost entirely) and on the three largest autosomal bivalents (Figures 2a and 2b). These signals are localized in both chromosomal ends of the two largest bivalents (I and II, Figure 2b) and in one end of the third largest autosomal pair (III, Figure 2b). The remaining seven autosomal bivalents and X chromosome not display strong labeling. Similar pattern is observed with C-banding (Figure 2c). On the mitotic chromosomes, more detail about the chromosomal localization of the hybridization signals can be seen (Figure 2d). The Y chromosome appears almost completely labeled. Terminal or telomeric positions of the autosomal hybridization signals are clearly observed. Furthermore a subterminal region (negative for GISH and DAPI) is observed between the telomeric hybridization signal and the euchromatin (Figure 2d, arrowheads). This pattern is similar to C-banding, in which the Y chromosome and six autosomes (three pairs) appear with C-blocks in one or both chromosomal ends (Figure 2e). The size of the C-heterochromatic regions is larger than those obtained with GISH, and they are formed by a darker subtelomeric region (arrowheads) and a clearer telomeric region with Giemsa staining. These results indicated that telomeric C-bands are strongly hybridized with the GISH probes while the subtelomeric regions are not.

In T. platensis, the three largest autosomal bivalents present small but intense signals on only one chromosomal end and both sex chromosomes (X and Y) are almost entirely hybridized (Figure 2f). This GISH pattern is very similar to the described by C-banding (Table 1). In M. spinolai, all chromatin appear labeled but the ten autosomal pairs exhibit strong hybridization dots on chromosomal ends, clearly observed during early meiotic prophase and metaphase II (Figures 2g to 2k). The Y chromosome appears strongly and totally labeled (Figures 2h to 2k). This hybridization pattern is similar to C-banding pattern (Figure 2l). In the other seven Triatomine tribe species, including Triatoma, Dipetalogaster, Eratyrus, and Panstrongylus genera, strong hybridization signal is only observed on the C-heterochromatic Y chromosome (Figures 3a to 3d, and 3f to 3h). However, some of these species exhibit autosomal C-heterochromatin (Table 1), which clearly is not labeled with our GISH probes, as observed in autosomes and one X chromosome of T. protracta (compare Figures 3d with 3e).

Finally in R. prolixus from the Rhodniini tribe, a species with C-band only on the Y chromosome, the hybridization signals are scattered throughout all chromosomes (Figure 3i). No chromosomal region was observed with specific labeling, including the heterochromatic Y chromosome.

Correlation between GISH and DAPI positive C-band

Genome-specific repeats have frequently a non-random distribution, forming clusters within heterochromatin blocks. In most organisms, including many plant species, the GISH hybridization signals often coincide with C-bands and also with DAPI positive regions, suggesting the presence of AT-rich DNA sequences at these regions [14]. Our GISH results in triatomines are partially consistent with that observed in plants, when evolutionarily related species are compared. However, the correspondence between C and DAPI patterns with GISH hybridizations are not observed when more evolutionarily distant species are analyzed.

In the three closely related infestans subcomplex species (T. infestans, T. delpontei, and T. platensis) and Mepraia spinolai, the strong hybridization signals mainly co-localized with their C-banding patterns (Figures 1, 2 and 3). However, more detailed analysis in T. infestans demonstrated that GISH signals coincide exactly with autosomal DAPI positive regions described for this species (Figure 2d). Hence, our hybridization signals are not able to label the subterminal CMA₃ positive C-heterochromatic regions (GC-rich) observed in the largest autosomal pairs of T. infestans (arrowheads Figures 2d and 2e). This concordance between GISH patterns and DAPI regions in the infestans subcomplex and M. spinolai species is not observed when we analyze more evolutionarily distant species, even within the genus Triatoma. For example, the DAPI positive C-heterochromatin regions in all autosomes and X chromosome of T. protracta do not show strong hybridization signals (compare Figures 3d and 3e). Similar results were obtained in other species with autosomal C-heterochromatin such as T. dimidiata and T. carrioni (Figures 3b and 3c). Our GISH results are in agreement with the Triatominae molecular phylogeny, which shown that M. spinolai is closed relater with South American than North American Triatoma species [29]. Hence, repetitive sequences shared between M. spinolai and infestans subcomplex species is showing this evolutionary relationship.

Recent analyses of the AT-rich satDNA portion of T. infestans using reassociation kinetics (C₀t) found two repetitive arrays located on the terminal regions of autosomal C-heterochromatin but not on the sex chromosomes heterochromatin [13]. Due to the similar chromosome localization, our GISH probes probably identify these repetitive arrays (Figure 2d) and others A–T rich sequences present on the C-heterochromatin of the X and Y sex chromosomes of T. infestans, not identified by C₀t studies (Figures 1h, 2b and 2d).

These results clearly show that: (a) infestans subcomplex closely related species share the majority of their repetitive sequences, in spite of their different size and chromosome distribution, (b) our genomic probes identify different repetitive DNA families, which are A-T rich, (c) autosomal and X heterochromatin regions from other evolutionary distant species are integrated by other divergent repetitive DNA, without homology to our genomics probes, hence not identified by GISH. Isolation and characterization of repeated sequences, either by GISH or others methodological approaches, seems to be appropriate genetic markers to infer evolutionary relationships between different species groups.

Evolution of sex chromosomes inferred by GISH

In all triatomine species, sex chromosomes are very well differentiated from autosomes by their particular behavior during meiosis. They are considered asynaptic and achiasmatic during male meiotic division [30], [31] showing an inverted meiosis: in first meiotic division the sister chromatids of each sex chromosome separate equationally and in the second division the sex chromosomes segregate to opposite poles [4].

In insect groups such as Diptera, Orthoptera and Lepidoptera distinct DNAs classes have been mapped on sex chromosomes [32]–[35]. However in Heteroptera there is a lack of knowledge about the molecular composition and evolution of sex chromosomes. In all of the 80 cytogenetically Triatominae studied species, the Y chromosome is almost entirely C-heterochromatic [3]. Molecular studies on the Y chromosome sequences are limited to fluorescence banding, showing that it is constituted by DAPI positive AT-rich sequences [10], [12], [13]. Our GISH studies reveal that in the ten Triatomini species, the Y chromosome is mainly constituted by accumulated repetitive sequences. It is well known that most Y chromosomes (from mammals, fish, insects and plants) are enriched with repeats in a process of heterochromatin accumulation. However, a striking result is that 10 species of triatomines, including 5 different genera, present and share the same types of repeated DNA sequences. In Diptera and Lepidoptera, unlike triatomines, Y chromosome (or their equivalent W chromosome) sequence composition appears to have a rather high turnover evolution rate because even in related species, the Y chromosome exhibit different repeated sequences or gene content [34], [36]. For this reason the Y chromosome repeated sequences conservation in Triatomini tribe species here analyzed is a very uncommon phenomenon and probably these repeated sequences represent an ancestral character of this tribe (Figures 1, 2 and 3).

On the other hand, the R. prolixus Y chromosome, species which belongs to the Rhodniini tribe, does not shared the same conserved repetitive DNA sequences observed in the analyzed Triatomini species (Figure 3i). This result suggests an early evolutionary differentiation or dichotomy within Triatominae subfamily in two clearly distinct clades: the Rhodniini and Triatomini tribes. Our results are in agreement with phylogenetic studies with nuclear and mitochondrial DNA sequences which suggested a Rhodniini–Triatomini node as the oldest known split within the triatomine bugs [39], [40]. Another explanation could be a different origin of the Triatomini and Rhodniini Y chromosomes, either deriving from different autosomal pairs or B chromosomes as suggested in homopteran and Drosophila species [41].

In triatomines, sex chromosomes are achiasmatic, so surely completely differentiated between each other, which implicate a scenario with an old Y chromosome. As has been described in Drosophila and mammals, old Y chromosome evolution is mostly driven by genetic hitchhiking, where evidence of positive selection evidence was found [36], [37]. In addition, the Y chromosome is a very important determinant of male fitness [38], so conservation in triatomines could be caused by a selective pressure on important male fitness genes. Maybe genome project data on Y-linked genes would bring light on this question.

One of the most striking results of this paper is that, in infestans subcomplex and M. spinolai species, the sex chromosomes (Xs and Y) with each other and with some autosomes share highly repeated sequences (Figures 1 and 2). A distinctive cytogenetic feature of these species is that during meiotic prophase chromosomal associations among autosomes with both sex chromosomes occur [18], [42]. These associations would facilitate the sequences exchange among these non-homologous chromosomes [9]. If we consider that the highly repeated sequences localized in the Y chromosome are the plesiomorphic state, a possible scenario could be a sequences transfer from Y chromosome to the autosomes and X chromosomes. Consequently, the sequence homology between them could be considered a secondary character (apomorphic) restricted to closely related species groups. Comparative analyses on the 45S rDNA clusters chromosome position in several triatomine species also suggested the existence of chromatin exchange between sex chromosomes [6], in contrast to the widely accepted idea that the achiasmatic sex chromosomes of Heteroptera do not interchange sequences [31]. Furthermore, the simultaneous presence of 45S ribosomal clusters in autosomes and the X chromosome observed in T. infestans and T. delpontei also suggests the occurrence of sequences exchanges among autosomes and sex chromosomes [5], [9]. It has been suggested that transposable elements could play an important role in the mobilization of repeat sequences among chromosomes [23]. Nevertheless, in order to advance in the knowledge of sex chromosomes evolution, a molecular characterization of their sequences will be necessary.

Phylogenetic origin of the subfamily Triatominae

The phylogenetic origin of blood-feeding Triatominae has received considerable attention due to the epidemiological significance as vectors of Chagas disease. Conflicting hypotheses support Triatominae as monophyletic [29], [39], [43], polyphyletic [17], [44] or paraphyletic group [40]. The answer to this question goes beyond the academic interest, since it could represent a very important issue to know if hematophagy in Triatominae arose from a single evolutionary event or as multiple independent evolutionary processes. According to the polyphyletic and paraphyletic proposals, the Triatomini and Rhodniini tribes are derived from quite different reduviid subfamilies. The sister group of the Triatomini seems most likely to be the Reduviinae, while the sister group for the Rhodniini may possibly be the Stenopodainae or Salyavatinae [17], [43]. Considering the extreme Y chromosome molecular differentiation between the two tribes described here, the application of GISH methodology on the suspected sister species of Triatominae could help elucidate the monophyletic or polyphyletic origin of this subfamily.