Interspecies Chromosome Mapping in Caprimulgiformes, Piciformes, Suliformes, and Trogoniformes (Aves): Cytogenomic Insight into Microchromosome Organization and Karyotype Evolution in Birds

Interchromosomal rearrangements involving microchromosomes are rare events in birds. To date, they have been found mostly in Psittaciformes, Falconiformes, and Cuculiformes, although only a few orders have been analyzed. Hence, cytogenomic studies focusing on microchromosomes in species belonging to different bird orders are essential to shed more light on the avian chromosome and karyotype evolution. Based on this, we performed a comparative chromosome mapping for chicken microchromosomes 10 to 28 using interspecies BAC-based FISH hybridization in five species, representing four Neoaves orders (Caprimulgiformes, Piciformes, Suliformes, and Trogoniformes). Our results suggest that the ancestral microchromosomal syntenies are conserved in Pteroglossus inscriptus (Piciformes), Ramphastos tucanus tucanus (Piciformes), and Trogon surrucura surrucura (Trogoniformes). On the other hand, chromosome reorganization in Phalacrocorax brasilianus (Suliformes) and Hydropsalis torquata (Caprimulgiformes) included fusions involving both macro- and microchromosomes. Fissions in macrochromosomes were observed in P. brasilianus and H. torquata. Relevant hypothetical Neognathae and Neoaves ancestral karyotypes were reconstructed to trace these rearrangements. We found no interchromosomal rearrangement involving microchromosomes to be shared between avian orders where rearrangements were detected. Our findings suggest that convergent evolution involving microchromosomal change is a rare event in birds and may be appropriate in cytotaxonomic inferences in orders where these rearrangements occurred.


Diploid Number and Karyotype Description
At least 20 metaphase spreads conventionally stained (Giemsa 5%) were analyzed to determine the diploid chromosome number and chromosomal morphologies for all studied species. Chromosomes were numbered consecutively based on their size and centromere position [33]. We performed a detailed karyotype description for P. brasilianus because this is the first cytogenetic analysis in this species.

Cross-Species FISH-Based Chromosome Maps
For the purpose of interspecies chromosome mapping, two BAC probes, selected from the genomic library of chicken or zebra finch and positioned as close as possible to the end of each microchromosome arms, were chosen for microchromosomes GGA10-28 (except GGA16) and applied to metaphases of P. brasilianus, H. torquata, P. inscriptus, R. t. tucanus, and T. s. surrucura (Table S1). The GGA16 and 29-38 were not tested here because there are no BAC probes available for these chromosomes. Two BAC probes per macrochromosome corresponding to GGA1-9 were also applied to metaphases of P. brasilianus and H. torquata (Table S1) since there were no prior studies of molecular cytogenetics in these species. Although we used BAC probes from chicken and zebra finch libraries, our results were compared with the chicken. Most of the BAC probes used were from the chicken. However, for some chromosomes, the chicken BACs did not work successfully in all bird species [24]; in these cases, we used BAC probes from the zebra finch. The BAC clone isolation, amplification, labeling and hybridization were performed following the O'Connor et al. [15] procedure. The FISH results were confirmed by analyzing at least 10 metaphase preparations per experiment.

Reconstruction of the Neognathae/Neoaves Ancestral Karyotypes
Datasets resulted from cross-species FISH-based mapping were used as input files for the software-assisted reconstruction of the hypothetical Neognathae and Neoaves ancestral karyotypes using the maximum-likelihood algorithm. For this purpose, we employed the Maximum Likelihood for Gene Order Analysis (MLGO) webserver [34]. The MLGO reconstruction algorithm built up a Neoaves ancestor (NAA) using the maximumlikelihood scenario for a certain number of common BACs among the five avian species studied plus the chicken reference karyotype. Being quite flexible, the algorithm can handle mixed datasets with missing/failed hybridization information for chromosome location of a particular BAC in a single species. For macrochromosome datasets in P. brasilianus, H. torquata and chicken, information about physical positions/orders of BACs relative to each other, p/q arms and centromeres was taken into account, enabling to reconstruct ancestral macrochromosomes in more detail.
To run the MLGO web tool, a phylogenetic tree for the 12 species was recreated from a comprehensive avian phylogeny described by Prum et al. [36]. The correct tree topology (in Newick format) was tested using the ETE v3 toolkit [37] ( Figure S1). As we established in our previous reconstruction studies [6,19,20,23], MLGO outputs were, by inference and manually, curated and adjusted further to interpret and correct software-assisted reconstruction results using the most parsimonious explanation of the available data.

Microscope Analysis and Image Capturing
Images were captured under a 100× objective using an Olympus BX61 epifluorescence microscope with a cooled CCD camera and SmartCapture (Digital Scientific UK, Cambridge, UK) system. Final image processing was performed using Adobe Photoshop 7.0 (accessed on 22 March 2021).

Statistical Analysis
Data of microchromosome rearrangements obtained in our study and from the literature were analyzed using a heterogeneity chi-squared test followed by analyses of residuals using WinPEPI software [38]. A statistically significant positive residual indicates an excess of species with rearrangements in a particular microchromosome. Residuals' p-values were adjusted for multiple tests using Benjamini and Hochberg's [39] false discovery rate (FDR) approach, with p ≤ 0.05 being considered statistically significant.

Karyotype Description
The karyotype of P. brasilianus is 2n = 74 and is described here for the first time. The first, second, third and fifth pairs are submetacentric, sixth and eighth are metacentric, and the remaining autosomal pairs are acrocentric. The Z sex chromosome is submetacentric and equivalent in size to the third autosomal pair (Figure 1a). Corroborating previous studies, we found 2n = 74 in H. torquata [40], 2n = 112 in P. inscriptus and R. t. tucanus [11], and 2n = 82 in T. s. surrucura [32] (Figure 1b-e, respectively). rate (FDR) approach, with p ≤ 0.05 being considered statistically significant.

Karyotype Description
The karyotype of P. brasilianus is 2n = 74 and is described here for the first time. The first, second, third and fifth pairs are submetacentric, sixth and eighth are metacentric, and the remaining autosomal pairs are acrocentric. The Z sex chromosome is submetacentric and equivalent in size to the third autosomal pair (Figure 1a). Corroborating previous studies, we found 2n = 74 in H. torquata [40], 2n = 112 in P. inscriptus and R. t. tucanus [11], and 2n = 82 in T. s. surrucura [32] (Figure 1b,c,d and e, respectively).

FISH Mapping Results and Reconstruction of the Neognathae/Neoaves Ancestors
The selected BAC probes were successfully hybridized in all the five species included in our cross-species FISH mapping study, except the ones for chromosome pair 25 in T. s.

FISH Mapping Results and Reconstruction of the Neognathae/Neoaves Ancestors
The selected BAC probes were successfully hybridized in all the five species included in our cross-species FISH mapping study, except the ones for chromosome pair 25 in T. s. surrucura. Examples of the respective FISH results are shown in Figures 2-4.
Using a web-based MLGO tool, the relevant input files for BAC order/orientation maps in the whole set of 12 species, and the relevant input phylogenetic tree ( Figure S1), we performed an overall estimation of the hypothetical Neognathae (NGA) and Neoaves (NAA) ancestors ( Figure 5). This was possible thanks to the chromosome maps for the ostrich (infraclass Palaeognathae) available from the O'Connor et al. [16] study and used as an outgroup. In the case of macrochromosomes, the orientation of each BAC relative to its neighbors on a particular chicken chromosome and relative to p/q arms was an additional advantage for the reconstruction of bird ancestors.      Using a web-based MLGO tool, the relevant input files for BAC order/orientation maps in the whole set of 12 species, and the relevant input phylogenetic tree ( Figure S1), we performed an overall estimation of the hypothetical Neognathae (NGA) and Neoaves (NAA) ancestors ( Figure 5). This was possible thanks to the chromosome maps for the ostrich (infraclass Palaeognathae) available from the O'Connor et al. [16] study and used as an outgroup. In the case of macrochromosomes, the orientation of each BAC relative to its neighbors on a particular chicken chromosome and relative to p/q arms was an additional advantage for the reconstruction of bird ancestors.
As a result of our ancestral karyotype reconstruction, a similar pattern of chromosome organization in the presumable NGA and NAA was observed ( Figure 5). Overall, according to this gross estimate and using datasets for the 12 species produced in this and few other published studies, NGA and NAA are likely to have 29 chromosomes (autosomes), including 10 macrochromosomes (i.e., autosomes 1-9 + 4A) and 19 microchromosomes (i.e., autosomes 10-28). Compared to the chicken karyotype, the only difference between these two karyotypes and the chicken one was that chromosome GGA4 was split into two separate chromosomes (4 and 4A) in NGA and NAA. Considering that the infraclasses Palaeognathae (ratites and tinamous) and Neognathae (superorders Galloanserae and Neoaves) diverged about 100 million years ago (Mya) and the Neognathae diverged into the evolutionary lineages of Galloanserae and Neoaves about 88 Mya [41], we have used the most ancestral NGA karyotype to compare with our results obtained through the FISH experiments.

Rearrangements
The rearrangements (Table 2) were identified, comparing our results with the NGA karyotype. According to these comparisons, chromosome evolution in P. brasilianus involved macrochromosomal fissions, a fusion of macrochromosomes, a fusion between a macrochromosome and a microchromosome and fusions of microchromosomes. The chromosomes homologous to NGA5 and 6 were split into two pairs each in the karyotype of P. brasilianus. Four fusions overall were found: NGA5seg/7, NGA8/12, NGA9/10, and NGA11/13. Chromosomes NGA14-28 were conserved as discrete units in this species, following the ancestral pattern. Figure 6 shows the homology between the NGA karyotype and P. brasilianus.
Fissions and fusions involving macro-and microchromosomes were also observed in H. torquata. Chromosomes homologous to NGA1, 2, and 5 are split into two pairs in the karyotype of H. torquata. Three associations were found: NGA6/10, NGA8/14, and NGA9/13. Chromosomes NGA11, 12, and 15-28 were conserved as respective microchromosomes in this species. Figure 7 shows the homology between the NGA karyotype and H. torquata.
Although a high diploid number is present in P. inscriptus and R. t. tucanus (2n = 112), no interchromosomal rearrangements involving the microchromosomes (NGA10-28) were found in these Piciformes species. Likewise, no interchromosomal rearrangements involving these microchromosomes were found in T. s. surrucura (2n = 82). Unlike the other species analyzed, no hybridization signs were produced when the BAC probes for chromosome 25 were applied to metaphase spreads of T. s. surrucura. The homology maps for these species are not shown because no interchromosomal rearrangements involving the microchromosomes were found, and the macrochromosome homology with Gallus gallus has been previously reported [11,32].

Comparison of Microchromosome Organization in Birds
We added our data to those previously published in the literature about chromosomal rearrangements in 18 microchromosomes in birds. In a total of 34 avian species (Supplementary Table S2), nine presented interchromosomal rearrangements ( Table 3). The average frequency of species with interchromosomal rearrangements along the 18 microchromosomes was 10.3%, but we found significant heterogeneity among chromosomes (χ 2 = 40.927, df = 17, p = 0.001). The frequency was higher in NGA10 (26%, p = 0.001), NGA13 (21%, p = 0.042) and NGA14 (23%, p = 0.009) ( Table 3). No interchromosomal rearrangements were observed in microchromosomes NGA22, NGA24, NGA26 and NGA27 (all at p < 0.05). However, when adjusting for the number of tests performed, only in NGA10 the excess of species with rearrangements reached statistical significance (p = 0.025) ( Table 3).  Fissions and fusions involving macro-and microchromosomes were also observed in H. torquata. Chromosomes homologous to NGA1, 2, and 5 are split into two pairs in the karyotype of H. torquata. Three associations were found: NGA6/10, NGA8/14, and NGA9/13. Chromosomes NGA11, 12, and 15-28 were conserved as respective microchromosomes in this species. Figure 7 shows the homology between the NGA karyotype and H. torquata.
Although a high diploid number is present in P. inscriptus and R. t. tucanus (2n = 112), no interchromosomal rearrangements involving the microchromosomes (NGA10-28) were found in these Piciformes species. Likewise, no interchromosomal rearrangements involving these microchromosomes were found in T. s. surrucura (2n = 82). Unlike the other species analyzed, no hybridization signs were produced when the BAC probes for chromosome 25 were applied to metaphase spreads of T. s. surrucura. The homology maps for these species are not shown because no interchromosomal rearrangements involving the microchromosomes were found, and the macrochromosome homology with Gallus gallus has been previously reported [11,32].

Comparison of Microchromosome Organization in Birds
We added our data to those previously published in the literature about chromosomal rearrangements in 18 microchromosomes in birds. In a total of 34 avian species (Supplementary Table S2), nine presented interchromosomal rearrangements ( Table 3). The average frequency of species with interchromosomal rearrangements along the 18 microchromosomes was 10.3%, but we found significant heterogeneity among chromosomes (χ 2 = 40.927, df = 17, p = 0.001). The frequency was higher in NGA10 (26%, p = 0.001),

Discussion
In previous studies, BAC probes have been used for inter-cross FISH mapping to comprehend the structure and organization of the chromosomes in species from 12 orders (Table S2), and interchromosomal rearrangements were reported only in Falconiformes, Psittaciformes and Cuculiformes species [15,16,[24][25][26][27][28] (Table S2). Here we cytogenomically analyzed five species from four Neoaves orders (Caprimulgiformes, Piciformes, Suliformes, and Trogoniformes), expanding the results to a total of 16 orders. Overall, the results demonstrated that interchromosomal rearrangements involving macro-and microchromosomes had an important role in the karyotype evolution of species of Caprimulgiformes and Suliformes, while the microchromosomes remained highly conserved in Piciformes and Trogoniformes. Our results suggest that the microchromosomes NGA10, NGA13 and NGA14 are involved in multiple rearrangements. However, only in NGA10, the frequency of rearrangements was supported statistically. In addition, this is the first molecular cytogenetic study in species of Suliformes and Caprimulgiformes.

Karyotype of H. torquata (Caprimulgiformes)
Fusion and fission events were also observed in H. torquata (Caprimulgiformes) (2n = 74). Chromosomal fissions were found in ancestral chromosome pairs 1, 2 and 5 (NGA1, 2 and 5). The breakpoints involved in these fissions are probably reused in bird chromosome evolution since fissions in these chromosomes were reported in several orders of birds [3][4][5]16]. However, the use of a higher number of BACs probes covering these chromosomes is necessary to confirm this hypothesis. In H. torquata, we found fusions between macrochromosomes (NGA6/10) and between macrochromosomes and microchromosomes (NGA9/13 and NGA8/14). Based on the fact that conventional cytogenetic analyses in Caprimulgiformes species revealed an interesting range of diploid number, from 2n = 68 in Chordeiles pusillus [50] to 2n = 86 in Nyctibius griseus [40], it is plausible to infer that fusions involving microchromosomes and macrochromosomes appear to have played an important role in the chromosome evolution of this group.

Karyotype of T. s. surrucura (Trogoniformes)
Chromosomal analysis of Trogoniformes species is still rare and is based only on the karyotype description and chromosome painting in T. s. surrucura, with 2n = 82 [32]. Although a microchromosome fusion was proposed in this species based on chicken macrochromosome painting [32], we did not find any evidence of this rearrangement in our analysis. However, given that there are no probes available for chicken chromosomes 16 and 29-38, we cannot entirely discard the occurrence of microchromosome fusions.

Karyotype of P. inscriptus and R. t. tucanus (Piciformes)
The diploid numbers found in P. inscriptus and R. t. tucanus, 2n = 112 in both species, raise questions concerning the rearrangements that may have led to this high diploid number. Comparative chromosome painting with chicken macrochromosome probes (GGA1-10, homologous to NGA1-10) has been performed, and extensive chromosomal fissions were found in the first five ancestral chromosome pairs (NGA1-5) [11]. However, the results presented here demonstrate the conservation of the ancestral patterns of microchromosomes in both species, suggesting that the fission events exclusively involved macrochromosomes. This finding suggests that the high diploid number observed in Ramphastidae species is a result of macrochromosomal fission only. Another interesting feature observed in Piciformes species is that the Z sex chromosome is the largest element of the karyotype [11,51]. Our study excludes the possibility that a fusion event took place between the Z chromosome and any of the microchromosomes tested here, corroborating a previous hypothesis that fissions of the macrochromosomes and the accumulation of repetitive sequences are the most likely mechanism responsible for the appearance of this enlarged sex chromosome in Piciformes species [11].

Overview of Interchromosomal Rearrangements Involving Microchromosomes in Birds
So far, no patterns of interchromosomal rearrangement have been reported as being shared among the species that exhibit rearrangements involving microchromosomes, e.g., from Cuculiformes [27], Psittaciformes [16], Falconiformes [15,25], Caprimulgiformes, and Suliformes (present study; Table S2). However, while microchromosomal fusions were shared by three Falconiformes species [15,25], each of the four Psittaciformes species studied exhibited a different pattern of microchromosomal fusions [15]. This would suggest that the convergent evolution of microchromosomal rearrangements seems to be a rare event in birds and may be an appropriate tool for phylogenetic analyses in the taxa where these rearrangements are present.
While it is evident that there are karyotypes highly rearranged and interchromosomal rearrangements involving microchromosomes were expected in Cuculiformes, Psittaciformes and Falconiformes [15,16,25,27,28], such rearrangements were not evident in the Giemsa-staining karyotypes of P. brasilianus and H. torquata, despite the fact that their diploid numbers are slightly lower than the putative avian ancestral karyotype (2n = 80).
Hence, even in species with conserved karyotypes at first glance, microchromosomal fusions may have played an important role in their karyotype evolution.
Some microchromosome syntenies are involved in multiple rearrangements; for example, the frequency of species with interchromosomal rearrangements was higher in three syntenic groups, homologous to microchromosomes NGA10, NGA13 and NGA14 (Table 2). However, only in NGA10, the frequency of rearrangements was statistically higher than the average (Table 2). Why this microchromosome is more prone to interchromosomal rearrangement than others remains unclear and deserves futures studies. We did not observe rearrangements in pairs NGA22, 24, 26 and 27, but this result was not supported by the statistical analysis. A possible explanation could be the small sample size studied (34 species), but the real stability of these microchromosomes to rearrangements cannot be ruled out until new studies are done.
Our results corroborate the recent suggestion that microchromosomes are not prone to breakage [15] since no fissions in these elements were observed to date. However, it is unclear why interchromosomal rearrangements involving microchromosomes are quite common in some orders (e.g., Falconiformes, Psittaciformes, Caprimulgiformes, Cuculiformes, and Suliformes), while in most other avian orders, they have remained largely unchanged [15]. Nevertheless, as more chromosome mapping of BAC probes is performed, the list of orders with microchromosome fusions is likely to increase, with some species-specific rearrangements being detected. Future cytogenomic studies using this approach will provide greater clarity on why microchromosomes remain conserved as discrete units in some species while they are prone to interchromosomal rearrangement in others.
Concluding, in this cross-species FISH mapping study, we have reported and characterized the organization of microchromosomes in species from four different Neoaves orders. The results have further contributed to avian cytogenomics, revealing that microchromosome fusions are not exclusive to the orders Cuculiformes, Falconiformes and Psittaciformes but are also inherent in representatives of the orders Caprimulgiformes and Suliformes. Additionally, our findings suggested that some microchromosomes are more likely to undergo interchromosomal rearrangements than others and that convergent evolution of microchromosomal rearrangements is a rare event in birds.

Supplementary Materials:
The following are available online at https://www.mdpi.com/article/ 10.3390/cells10040826/s1, Figure S1: The MLGO input phylogenetic tree, Table S1: Identification of BAC probes used in this study, Table S2: List of avian species analyzed with microchromosome BAC probes.  Institutional Review Board Statement: The biological material from Hydropsalis torquata and Trogon surrucura surrucura were obtained from individuals captured in their natural environment using mist nets following permissions from Sistema de Autorização e Informação em Biodiversidade (SISBIO 44173-3). The biological material from Phalacrocorax brasilianus, Pteroglossus inscriptus, and Ramphastos tucanus tucanus were obtained from ex situ individuals kept at Parque Zoobotânico do Museu Paraense Emilio Goeldi/PA or Parque Mangal das Garças/PA, with the assistance from the staff from these institutions and following permissions from Sistema de Autorização e Informação em Biodiversidade (SISBIO 68443-1). The experiments followed protocols approved by the ethics committee from Universidade Federal do Pará (no. 170/2013) and Universidade Federal do Pampa (no. 018/2014).

Informed Consent Statement: Not applicable.
Data Availability Statement: All the data supporting our findings are contained within the manuscript.