|
| Specimen of Archocentrus sajica, a Mesoamerican heroine cichlid. Photo courtesy of Vinod Kutty. |
[To be completed]
The tribe Heroini constitutes is the second largest (more than 150 species) and a very diverse assemblage of Neotropical cichlids. This tribe is the one with the widest distribution of all the Neotropical cichlid tribes, with a range that extends from Southern Buenos Aires Province in Argentina (close to the Northern border of Patagonia) to the Río Grande river basin in Texas, USA; and includes most of South America, all of Central America, some of the Caribbean islands, and Southern North America. Currently, it comprises the South American genera Heroina, Heros, Hoplarchus, Hypselecara, Mesonauta, Pterophyllum, Symphysodon and Uaru; the NW South American species referred to Cichlasoma by earlier authors and the SE South American `Cichlasoma' facetum species group, along with all the Caribbean, Central, and North American species referred to the genus Cichlasoma Swainson, 1839 in the past (now genera Archocentrus, Amphilophus, Astatheros, Herichthys, Hypsophrys, Nandopsis, Parachromis, Paraneetroplus, Paratheraps, Petenia, Theraps, Thorichthys, Tomocichla, Vieja, and several yet unassigned species), as well as the Mesoamerican genera Herotilapia and Neetroplus.
In spite of their scientific importance, the taxonomy of the heroines is in need of an urgent revision, and the evolutionary history of the group is unclear. Traditional classifications have relied mainly on tropic morphology, but these traits turned out to be unreliable for the group, since this fish are characterized by very plastic morphology. Efforts to use molecular markers have not involved any comprehensive treatment of the group, and were usually limited to a few species and left many recognized groups unrepresented.
Heroini constitutes an important component of the Neotropical fish fauna, specially in Mesoamerica, where they are the only cichlids present, with a few minor exceptions, all in southern Panamá: one `Aequidens' species of the rivulatus group, one `Geophagus' species of the steindachneri group, and one Retroculus species; all of them belonging to South American genera. The biogeography of the heroine group, and specially the number of times the group colonized Central, North America, and the Caribbean, as well as the expansion routes and the time at which the colonization took place, are not known. One point of special interest is the question of the monophyly of the non-South American heroines.
The group is an important one as it is used as for many studies on ecology, behavior, biogeography, and speciation. The heroines are also the standard group used as a comparison in studies involving the radiation of the cichlids of the Great Lakes in Africa. All these studies need a good phylogeny of the Heroini.
Cichlidae is one of the very few families of fishes which monophyly is virtually uncontested. Recent advances in the phylogeny of Cichlidae, especially those using molecular tools, have elucidated the picture of the major groups of cichlids and their relationships. But, while the major groups within the family Cichlidae as a whole and those of the Neotropical clade are clear, the relationships within these clades are not known. In spite of their scientific importance, the taxonomy of the heroines is in need of an urgent revision, and the evolutionary history of the group is unclear. There is presently no well-supported comprehensive phylogeny for the Heroini tribe, but only old ones based largely on trophic morphology which is a bad character for this group due to their well documented morphological plasticity. Phylogenies for the group based on molecular characters are partial and not comprehensive, usually part of a phylogeny of larger groups of cichlids or of Cichlidae in general.
My best guess at the species of the Heroini.
The molecular markers to be used will be the DNA sequence of the mitochondrial gene cytochrome b and the single-copy nuclear gene ependymin. Cytochrome b is a good molecular marker for the scales of time and phylogenetic separation of the group to be studied, less than 100 MY. It has also already been used in our laboratory following well-established technique, so it will not be necessary to acquire new know-how. Besides, many of the sequences required for this study are available at the lab itself form previous work. Finally, there are published sequences of this marker for many of the taxa of the Heroini, thus reducing the sheer complexity of sampling such a large and geographically extended group.
Ependymin is a glycoprotein secreted by meningeal cells that constitutes a major component of the cerebrospinal fluid in many orders of teleost fish. It was originally identified in goldfish brain. The gene for ependymin is presumed to be a good molecular marker for this group since it is a single-copy nuclear gene and previous studies in other groups of fish have shown that ependymin has a good phylogenetic signal over a wide range of taxonomic distance and divergence time. The use of a nuclear marker is important as a complement of studies performed with mitochondrial markers. The use of this gene will also constitute a novel approach of the study of phylogeny of the Cichlidae, and will complement the use of the mitochondrial marker, cytochrome b.
DNA sequences for both genes will be obtained in the lab, and added to sequences available from published sources and potential collaboration with other researchers. Sequences will be obtained in the lab from available tissue samples by PCR amplification and sequenced in an automatic sequencer, following standard protocols. In the case of ependymin, the primer to be used will be designed after the consensus sequence to be obtained in a current pilot study, using cDNA for 16 species of cichlids. Tissue samples will be obtained from samples already available in Dr. Ortí's laboratory, donated by researchers doing field work and captive breeding programs, form specimens collected by ourselves in the field, and from fish purchased from the aquarium trade, this as a last resource. As far as possible, voucher specimens for the samples and sequences will be secured or identified as publicly available. This preserved material will be used for accurate taxonomic identification, and for future reference and checking as new taxonomic information becomes available, since new developments in cichlid taxonomy are almost certain to occur.
Sequences will be aligned using the program Clustal-W. Phylogenetic trees will be reconstructed using PHYLIP and PAUP* for parsimony and maximum likelihood, and PUZZLE for quartet-puzzling maximum likelihood. Seeking to facilitate reproducibility by other researchers, public software will be favored for this study.
After the basic, computerized phylogenetic estimation is completed, the resulting clades will be compared with the current taxonomy and published work on the morphology, ecology, and geographical ranges of the involved taxa, in order to draw monophyletic groups that are also meaningful biological generalizations. If possible and deemed necessary, recommendations will be made on the nomenclature of these fishes.
The obtained phylogeny will be superimposed with geological maps of the present and ancient configuration of the Americas, following the guidelines suggested by Avise. The current distribution of the primitive Mesoamerican cichlids will be correlated with the position of these locations in geological times, and correspondence between the branching pattern of the phylogenetic tree and the past and present geographical locations of the clades. These would test for the first time the several competing `informal' hypothesis on the colonization routes on a phylogenetic framework, as well as estimate the age of the colonization(s); a crucial information, given the very dynamic nature of geology the Mesoamerican isthmus, that has caused the isolation of South and North America at different times, as well as their interconnection through at least two different land bridges.
The tribe Heroini is recognized as one in dire need of an urgent thorough taxonomical revision. Given the relatively large number of species and their diversity, and the poor reliability of morphological traits for inferring evolutionary relationships within the group, the reconstruction of their phylogeny and a new revised taxonomy will be problematic. Although the proposed research is not of a revisionary nature, a well supported phylogeny based on molecular markers will constitute a good framework for any subsequent attempt at this much needed revision of the group. Given the importance of this group of fishes for many kinds of research, from reproductive ethology to studies speciation, this will surely have a wide impact in many areas of science.
A better understanding of the relationship between morphological and molecular evolution, and on the mechanism of speciation, are also probable indirect outcomes of the proposed research.
Geography-related evolution and speciation mechanisms are currently areas of intensive research, and cichlids are one of the standard models for these kinds of studies. An improvement if the knowledge of the biogeography of cichlids in the Americas will add to these efforts.
Finally, many of these cichlids have restricted ranges, and are therefore vulnerable to the impact of human activities. A better understanding of their distribution and phylogenetic uniqueness of the different species and clades will be helpful for conservation efforts, as will likely identify taxa and geographical areas of particular interest for the preservation of biodiversity.
![[Crenicichla proteus]](http://www.geocities.com/NapaValley/5491/proteusfemale.jpg) |
A female Crenicichla proteus guarding fry. Although morphologically very different, molecular data show that the predatory cichlids of the genus  Crencichla are part of the geophagine clade along with the genus Geophagus. Photo courtesy of Vinod Kutty. |
Farias, I.P., G. Ortí, and A. Meyer. 2000. Total evidence: molecules, morphology and the phylogenetics of cichlid fishes. Journal of Experimental Zoology (Molecular and Developmental Evolution) 288: 76-92.
Farias, I.P., G. Orttí, I. Sampaio, H. Schneider, and A. Meyer. 1999. Mitochondrial DNA phylogeny of the family cichlidae: monophyly and fast molecular evolution of the Neotropical assemblage. Journal of Molecular Evolution 48:703-711.