INTRODUCTION

Within the order Siluriformes, the family Loricariidae is the most diverse, and several studies have sought to elucidate its phylogenetic relationships (Montoya-Burgos et al. 1998; Schaefer 1998; Armbruster 2004; Cramer et al. 2011; Lujan et al. 2015; Covain et al. 2016; Roxo et al. 2019). Currently, with more than a thousand valid species, the family is organized into six subfamilies: Lithogeninae, Delturinae, Rhinelepinae, Loricariinae, Hypoptopomatinae and Hypostominae (Fricke et al. 2024).

The members of the Loricariinae, known as armored catfish, are distributed throughout the rivers of South America, from the affluents of La Prata River in the south to the coastal rivers of the Caribbean in the north (Ferraris Jr. 2003; Rapp Py-Daniel and Ohara 2013). Fishes of this subfamily are easily distinguished from other loricariids by their very depressed and elongated caudal peduncle, the absence of an adipose fin and the fact that they often have a depressed snout. In addition, they usually have variations in body shape, lip morphology and dentition (Covain and Fisch-Muller 2007; Rapp Py-Daniel and Ohara 2013). Nonetheless, considering the various taxonomic revisions in the systematics of Loricariinae, as evidenced by both morphological and molecular findings (Isbrücker 1979; Rapp Py-Daniel 1997; Covain et al. 2016; Londoño-Burbano and Reis 2021), the validity of several genera remains contentious, resulting in a degree of complexity within the taxonomy of this group (Covain et al. 2016; Londoño-Burbano and Reis 2021).

Covain et al. (2016) organized Loricariinae into two tribes: Harttiini and Loricariini. Within the tribe Loricariini, two subtribes were further delineated: Farlowelliina and Loricariina. Loricariina was diagnosed based on the dentition and structure of the lips, and it encompasses three morphological groups: Loricaria-Pseudohemiodon, Loricariichthys, and Rineloricaria.

The clade Loricariichthys comprises several genera, including LoricariichthysBleeker 1862, Furcodontichthys Rapp Py-Daniel 1981, Hemiodontichthys Bleeker 1862, LimatulichthysIsbrücker & Nijssen 1979, and Pseudoloricaria Bleeker 1862. This clade is characterized predominantly by monotypic or poorly diversified genera (Covain and Fisch-Muller 2007; Covain et al. 2016). In this clade, the genus Pseudoloricaria has undergone some discussions. It was originally designated as a member of Loricaria Linnaeus 1758 with the species Loricaria laeviuscula Valenciennes, 1840, and was later established as Pseudoloricaria by Bleeker in 1862. At first, there were uncertainties regarding its validity as a genus or subgenus of Loricaria. Such doubts arose from the original description, which was made from a single specimen, without additional information on the collection locality. It was only known that it originated in South America (Bleeker 1862; Isbrücker and Nijssen 1976). Despite initial uncertainties, Pseudoloricaria is considered a valid genus that was redescribed and revised by Isbrücker and Nijssen (1976), who classified two species: P. laeviuscula and Pseudoloricaria punctata (Regan, 1904). Nonetheless, P. punctata was relocated to the genus Limatulichthys and it is currently referred to as the species Limatulichthys petleyi (Fowler 1940) (Covain and Fisch-Muller 2007). Thus, Pseudoloricaria is monotypic, containing only P. laeviuscula.

The true diversity of the genus Pseudoloricaria may be underestimated, since exclusive studies on the genus are scarce. Several decades after the studies of Isbrücker and Nijssen (1976), Ohara (2010) conducted a taxonomic review of the genera Pseudoloricaria and Limatulichthys, recognizing, based on samples from different locations in the Amazon basin, two new species within the genus: Pseudoloricaria sp. n. “mucajaí” and Pseudoloricaria sp. n. “madeira”. Ohara’s study was based only on morphological characters, and the author emphasized the need for more detailed investigations of the genus, demonstrating the possibility that Pseudoloricaria may not be monotypic. In other studies, it is possible to note the difficulty in identification, since some authors refer only to Pseudoloricaria sp. (Melo et al. 2004; Lujan et al. 2012) or P. aff. laeviuscula (Collins et al. 2015). Data are limited regarding distribution, and studies indicate the occurrence of the genus and P. laeviuscula in the Amazon and Tocantins-Araguaia basins (Covain and Sleen 2017; SiBBr 2024).

Cytogenetic studies regarding the Loricariichthys group remain limited, with existing data restricted to the genera Loricariichthys (Scavone and Júlio Jr. 1995; Fenocchio et al. 2003; Rodrigues 2010; Takagui et al. 2014, 2020) and Hemiodontichthys (Carvalho et al. 2018). The 2n within this group ranges between 46 and 56, with 2n=54 predominating. Heterochromatin is observed in a few discrete blocks, and the nucleolus organizer region (NOR) is characterized as simple (Scavone and Júlio Jr. 1995; Carvalho et al. 2018; Takagui et al. 2014, 2020). Additionally, the 18S and 5S ribosomal sites exhibit synteny in Loricariichthys platymetoponIsbrücker & Nijssen 1979, whereas they are located on different chromosomes in Loricariichthys anus (Valenciennes 1835) (Takagui et al. 2020). Considering the existing uncertainties surrounding the taxonomy and the deficiencies in the cytogenetic data for this subfamily, this study aimed to characterize P. laeviuscula through both conventional and molecular cytogenetic approaches to infer its evolutionary relationships within the Loricariichthys group.

MATERIAL AND METHODS

In this study, 20 individuals (seven males and thirteen females) of the species Pseudoloricaria laeviuscula (Figure 1) were collected manually using trawl nets, with nineteen obtained from the Cuieiras River (2°49’03.6” S, 60°29’09.6” W) and one individual from the Negro River, in the Anavilhanas National Park (2°40’27.6” S 60°39’18.4” W), Amazonas, Brazil (Figure 2). The Brazilian Institute of the Environment and Renewable Resources (IBAMA) authorized the collections under a permanent license from the Biodiversity Information System (SISBio No. 28095-1). The experiments followed the ethical guidelines established by the National Council for the Control of Animal Experimentation (CONCEA) and were approved by the Ethics Committee for Animal Research at INPA (CEUA-INPA) under protocol number 030/2020, SEI 01280.001063/2020-07. The specimens were identified by Dra. Lúcia Helena Rapp Py-Daniel and deposited in the fish collection at INPA (INPA-ICT 059686 and INPA-ICT 060695).

Figure 1
Specimen of Pseudoloricaria laeviuscula: A) Dorsal view, B) Ventral view (opened due to cytogenetic procedures), C) Lateral view (mirrored to the left, as it was the better-preserved side). Male individual measuring 260 mm, coloration in alcohol.

Figure 2
Sampling sites: in the circle, the Cuieiras River; in the square, the Anavilhanas National Park, both located in the state of Amazonas, Brazil.

Mitotic chromosomes were obtained according to the protocol of Gold et al. (1990). The heterochromatin pattern was determined using C-banding (Sumner 1972; Lui et al. 2012), and the nucleolus organizer regions were located via impregnation with silver nitrate (Ag-NOR) (Howell and Black 1980).

Genomic DNA extraction was performed using the muscle tissue and liver of the species under study, which were preserved in 100% ethanol, using the Wizard^® extraction kit (Promega), following the manufacturer’s recommendations. The repetitive sequences 18S and 5S rDNA, used as probes, were isolated via PCR, using the primers: 18Sf (5’-CCG CTT TGG TGA CTC TTG AT-3’) and 18Sr (5’-CCG AGG ACC TCA CTA AAC CA-3’) (Gross et al. 2010); 5Sf (5’-TAC GCC CGA TCT CGT CCG ATC-3’) and 5Sr (5’- CAG GCT GGT ATG GCC GTA AGC-3’) (Martins and Galetti Jr. 1999). For the detection of telomeric sequences, the primers (TTAGGG)5 and (CCCTAA)5 were used (Ijdo et al. 1991). The PCR products were labeled using nick translation with the labeling kit dUTP-ATTO-550 (red) for 18S rDNA and the telomeric sequence and dUTP-ATTO-488 (green) for 5S rDNA, following the manufacturer’s instructions (Jena Bioscience). Fluorescent in situ hybridization (FISH) followed the protocol of Pinkel et al. (1986). The slides containing the chromosomes were denatured with 70% formamide and hybridized at 37 °C, overnight, with 77% stringency. Metaphases were stained with DAPI (0.8 ng/µL) in an antifading reagent (Vectashield^®).

The slides that used fluorochromes (C-bands and FISH) were analyzed under an epifluorescence photomicroscope (Olympus, BX-51) with an appropriate filter. At least 30 metaphases per individual were analyzed, and the best ones had their image captured using the DPController image capture system and were processed using the DPManager program. To assemble the karyotypes, Adobe Photoshop 7.0 (version CS6) was used, via which the chromosomes in mitotic metaphase were cut, paired, measured in the DPManager program, and placed in descending order of size. The chromosomes were classified according to Levan et al. (1964).

RESULTS

Pseudoloricaria laeviuscula showed a diploid number equal to 54 chromosomes (14m + 10sm + 4st + 26a) and the fundamental number (FN) was equal to 82, with no evidence of chromosomal sexual heteromorphism (Figure 3a).

Figure 3
Karyotype of Pseudoloricaria laeviuscula: A) conventional staining with Giemsa; B) C-banding; Ag-NOR highlighted in pair 13; C) chromosome mapping using fluorescent in situ hybridization (FISH) of the 18S rDNA (red) and 5S rDNA (green); and D) telomeric sequence (red). Bar equal to 20 µm.

C-banding revealed blocks of constitutive heterochromatin in the pericentromeric and centromeric regions of most chromosomes. Additionally, a pair with heterochromatic short arms (pair 8), an interstitial block in the long arm (pair 13), and terminal markings in the long arms of pairs 18, 19 and 26 were observed (Figure 3b), highlighting conspicuous blocks in the centromeric region of pair 27. The NOR was evidenced in the long arms, in the interstitial position of pair 13, coincident with the secondary constriction and the C⁺ band (Figure 3b) and confirmed via FISH with the 18S rDNA probe (Figure 3c), while the 5S rDNA site is in pair 18, in a pericentromeric position, coincident with heterochromatin (Figure 3c). The telomeric sequence was observed only in the terminal portions of all the chromosomes (Figure 3d).

DISCUSSION

The karyotype analysis of P. laeviuscula reinforces the idea of a plesiomorphic chromosomal arrangement in the family, which is characterized by a diploid number of 54 chromosomes. This chromosomal configuration is similarly observed in species of the genera HarttiaSteindachner 1877 and Loricariichthys (Scavone and Júlio Jr. 1995; Blanco et al. 2017; Takagui et al. 2014, 2020; Sassi et al. 2020, 2021). Species of the other genera of Loricariinae present diverse karyotypic macrostructures, which result from Robertsonian and non-Robertsonian rearrangements (Rosa et al. 2012; Ferreira et al. 2014; Primo et al. 2016; Glugoski et al. 2018, 2023). Some studies consider 2n=54 to be the ancestral number, i.e., plesiomorphic of Loricariidae. This assertion is based on the observation that this chromosomal number occurs in species belonging to the sister group Trichomycteridae and the subfamilies Hypostominae, Hypoptopomatinae and Loricariinae (Artoni and Bertollo 2001; Kavalco et al. 2005; Alves et al. 2012; Ziemniczak et al. 2012; Takagui et al. 2020).

Within Loricariinae, the species belonging to the morphological group Loricariichthys have a predominance of 2n=54. An exception is noted for Hemiodontichthys acipenserinus (Kner 1853), which has been reported to possess two diploid numbers, specifically 2n=46 and 2n=58 (Carvalho et al. 2018). Additionally, Loricariichthys maculatus (Bloch 1794) is characterized by a 2n=56 (Fenocchio et al. 2003). Furthermore, P. laeviuscula, which also has a diploid number of 54, shares the same FN (82) as L. anus and L. platymetopon (Scavone and Júlio Jr. 1995; Takagui et al. 2014, 2020). However, the karyotypic formulas differ, indicating the presence of non-Robertsonian rearrangements. A striking difference is the presence of subtelocentric chromosomes in P. laeviuscula, which is absent in L. anus (Takagui et al. 2014, 2020), and in certain populations of L. platymetopon (Takagui et al. 2014, 2020). In contrast, a predominance of a higher number of acrocentric chromosomes is observed in all species of this group.

Considering that the ancestral diploid number for Loricariidae is established as 2n=54, pericentric inversions are believed to have played a significant role in the karyotypic diversification of the Loricariichthys group, thus originating different karyotypic formulas, but preserving the diploid number at 54 in the species L. anus (Takagui et al. 2014, 2020), L. platymetopon (Scavone and Júlio Jr. 1995; Takagui et al. 2014, 2020) and P. laeviuscula. On the other hand, centric fissions are thought to have resulted in karyotypes with 2n=56 in L. maculatus (Fenocchio et al. 2003) and 2n=58 in H. acipenserinus (Carvalho et al. 2018). Additionally, the fusion events are believed to give rise to the karyotype with 2n<54 in H. acipenserinus (2n=46) (Carvalho et al. 2018).

The heterochromatic pattern found in P. laeviuscula exhibits similarities to that which is observed for most of Loricariinae, characterized by a limited number of small heterochromatic blocks. This configuration is also considered plesiomorphic in Loricariidae, as evidenced in species of the genera Harttia, Loricariichthys and RineloricariaBleeker 1862, among others (Kavalco et al. 2005; Ziemniczak et al. 2012; Blanco et al. 2017; Takagui et al. 2014, 2020; Sassi et al. 2020). However, despite the few blocks, the distribution of heterochromatin in these species reveals distinct chromosomal characteristics, with each species presenting a unique and exclusive pattern, indicating the occurrence of chromosomal rearrangements. For instance, in L. anus, two populations sharing the same diploid number (2n = 54) and the same FN (82) have been found to exhibit pericentric inversions involving both metacentric and acrocentric chromosomes. These inversions, detected by the heterochromatin distribution pattern, account for the variations observed in karyotypic formulas within the same species (Takagui et al. 2014).

In P. laeviuscula, the presence of terminal heterochromatic blocks in three pairs (18, 19 and 26, acrocentric) corresponds to the pattern also found in some species of the clades Harttiini and Farlowellina, which are considered basal clades of Loricariinae (Kavalco et al. 2004; Fernandes et al. 2015, 2021; Blanco et al. 2014, 2017; Marajó et al. 2018; Sassi et al. 2021). This observation suggests that, despite P. laeviuscula being classified within a derived clade according to the phylogenetic analysis of Covain et al. (2016), it may still retain karyotype characteristics that are reminiscent of the basal groups of the subfamily.

Another character considered plesiomorphic in the family is the simple and interstitial NOR (Oliveira and Gosztonyi 2000; Kavalco et al. 2005; Alves et al. 2012; Ziemniczak et al. 2012). This character is also present in P. laeviuscula. However, there is a variation in the chromosome location and karyotype position of the NOR among species within the Loricariichthys group. Specifically, in P. laeviuscula, the NOR was observed in a subtelocentric chromosome pair, whereas in L. anus, and L. platymetopon (Takagui et al. 2014, 2020), the NOR is located on acrocentric chromosomes. This pattern is similar to that found in species of the clade Harttiini (Centofante et al. 2006; Blanco et al. 2012, 2014, 2017).

Nonetheless, when comparing the nucleolar pair of P. laeviuscula in relation to the characteristics of the Loricariichthys group, it is possible to suggest that the presence of an interstitial NOR in a subtelocentric pair may be a consequence of a pericentric inversion. This hypothesis is supported by the findings in some species of Harttia, such as H. longipinna Langeani, Oyakawa & Montoya-Burgos 2001, H. gracilis Oyakawa 1993, H. punctata Rapp Py-Daniel & Oliveira 2001, H. torrenticola Oyakawa 1993, and H. carvalhoi Miranda Ribeiro 1939, in which the NOR is located in the proximal region of the first pair of acrocentric chromosomes (Centofante et al. 2006; Blanco et al. 2012, 2014, 2017). In contrast, in H. absaberi Oyakawa, Fichberg & Langeani 2013 (Rodrigues 2010) and H. kronei Miranda Ribeiro 1908 (Blanco et al. 2017), the NOR is located on metacentric chromosomes. This difference in the NOR pattern is suggested to result from pericentric inversions.

In contrast to the syntenic arrangement of 18S and 5S rDNA, which is considered plesiomorphic within Loricariidae (Mariotto et al. 2011; Ziemniczak et al. 2012; Blanco et al. 2017; Takagui et al. 2020), and observed in species of Harttia and Farlowella Eigenmann & Eigenmann 1889 (Centofante et al. 2006; Blanco et al. 2013, 2017; Deon et al. 2020; Fernandes et al. 2021), P. laeviuscula exhibits the presence of 18S and 5S rDNA sites on distinct chromosomal pairs, indicating a derived character. The non-synteny of ribosomal genes in fishes is regarded as an evolutionary benefit, as it prevents the potential for detrimental rearrangements between these sites (Martins and Galetti Jr. 1999, 2001). Our results indicate that the 18S and 5S rDNA sites are associated with constitutive heterochromatin. This association may serve to protect these gene sequences from selective pressures, facilitating the differential evolution of various genomic regions (Gross et al. 2010; Favarato et al. 2019).

The 5S rDNA site in P. laeviuscula exhibits a simple configuration, located in a mid-acrocentric pair in the pericentromeric region. This pattern contrasts with those found in other genera of Loricariinae. In this subfamily, the 5S rDNA demonstrated considerable variability, both in the number of sites and in their chromosomes. It can be simple, as in P. laeviuscula (present work) and H. kronei (Blanco et al. 2017), or as multiple sites, as evidenced in L. platymetopon, Rineloricaria cadeae (Hensel 1868), Rineloricaria pentamaculata Langeani & de Araujo 1994 and Farlowella hahni Meinken 1937 (Porto et al. 2011; Takagui et al. 2020; Fernandes et al. 2021; Venturelli et al. 2021). In the context of Loricariinae, the 5S rDNA is significantly more variable and unstable compared to the 18S rDNA, which exhibits relative conservation in the number of sites. This characteristic suggests that the 5S rDNA may serve as a cytotaxonomic marker for P. laeviuscula and other species within Loricariinae. This marker has previously been employed to separate species of Rineloricaria, which are characterized by considerable karyotypic diversity (Venturelli et al. 2021). Regarding the telomeric sequence (TTAGGG)n, P. laeviuscula showed sites that are exclusively located in the terminal regions of its chromosomes.

Thus, cytogenetic analysis reveals that P. laeviuscula is a species with most of its chromosomal characteristics conserved, and the data presented here corroborate that this species is more closely related to the Loricariichthys group than to the Loricaria-Pseudohemiodon group (Rapp Py-Daniel 1997; Covain et al. 2016; Roxo et al. 2019). The diploid number found for Loricaria is 64 (Porto et al. 2014; Benitez et al. 2016; Takagui et al. 2020), while for the Loricariichthys group it is predominantly 2n=54 (Scavone and Júlio Jr. 1995; Rodrigues 2010; Takagui et al. 2014, 2020). This fact should be considered when discussing its taxonomic position in Loricariinae. Thus, we believe that Pseudoloricaria may still have its diversity underestimated, considering the possible new species (Ohara 2010) and uncertainties in identification (Melo et al. 2004; Lujan et al. 2012; Collins et al. 2015).

CONCLUSIONS

The analysis of the karyotypic macrostructure of Pseudoloricaria laeviuscula revealed the conservation of the diploid number (2n=54), the C-banding pattern and the NOR localization, placing the species in a plesiomorphic context within the family Loricariidae. However, the location of the 18S and 5S rDNA sites on distinct chromosomes represents a derived characteristic of the family. This study establishes a basis for future research, suggesting an integrative taxonomic revision across the entire distribution of the genus.

ACKNOWLEDGMENTS

This study was carried out with the support of Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance code 001, Fundação de Amparo à Pesquisa do Estado do Amazonas (FAPEAM) - Programa Institucional de Apoio à Pós-graduação Stricto Sensu (FAPEAM- POSGRAD), and funding from the project INCT/CNPq/FAPEAM (phase II) - Centro de Estudos de Adaptações da Biota Aquática da Amazônia - ADAPTA II (INCT/CNPq/FAPEAM 573976/2008-2). EF received a productivity grant from CNPq - Process No. 302421/2014-9. AJAS received a FAPEAM scholarship. AJAS is grateful to Patrik Ferreira Viana for the suggestions during the construction of this study. This manuscript was translated by a professional translator and proofreader who is a native speaker of English.

REFERENCES

Data availability

The data that support the findings of this study are available, upon reasonable request, from the corresponding author [Ana Júlia Alegria Serra].

Edited by

Publication Dates

History