Phylogenetic sequence analysis, recombinant expression, and tissue distribution of a channel catfish estrogen receptor beta.

An estrogen receptor beta (ERbeta) cDNA fragment was amplified by RT-PCR of total RNA extracted from liver and ovary of immature channel catfish. This cDNA fragment was used to screen an ovarian cDNA library made from an immature female fish. A clone was obtained that contained an open reading frame encoding a 575-amino-acid protein with a deduced molecular weight of 63.9 kDa. Maximum parsimony and Neighbor Joining analyses were used to generate a phylogenetic classification of channel catfish ERbeta on the basis of 25 full-length teleost and tetrapod ER sequences. The consensus tree obtained indicated the existence of two major vertebrate ER subtypes, alpha and beta. Within each subtype, and in accordance with established phylogenetic relationships, teleost and tetrapod ER were monophyletic confirming the results of a previous analysis (Z. Xia et al., 1999, Gen. Comp. Endocrinol. 113, 360-368). Extracts of COS-7 cells transfected with channel catfish ERbeta cDNA bound estrogen with high affinity (K(d) = 0.21 nM) and specificity. The affinity of channel catfish ERbeta for estrogen was higher than previously reported for channel catfish ERalpha. As determined by qualitative RT-PCR, the tissue distributions of ERalpha and ERbeta were similar but not identical. Both ER subtypes were present in ovary and testis. ERalpha was found in all other tissues examined from juvenile and mature fish of both sexes. ERbeta was also found in most tissues except, in most cases, whole blood and head kidney. Interestingly, the pattern of expression of ER subtypes in head kidney always corresponded to the pattern in whole blood. In conclusion, we isolated a channel catfish ERbeta with ligand-binding affinity and tissue expression patterns different from ERalpha. Also, we confirmed the validity of our previously proposed general classification scheme for vertebrate ER into alpha and beta subtypes and within each subtype, into teleost and tetrapod clades.

Cloning, in vitro expression, and novel phylogenetic classification of a channel catfish estrogen receptor.

We obtained two channel catfish estrogen receptor (ccER) cDNA from liver of female fish using RT-PCR. The two fragments were identical in sequence except that the smaller one had an out-of-frame deletion in the E domain, suggesting the existence of ccER splice variants. The larger fragment was used to screen a cDNA library from liver of a prepubescent female. A cDNA was obtained that encoded a 581-amino-acid ER with a deduced molecular weight of 63.8 kDa. Extracts of COS-7 cells transfected with ccER cDNA bound estrogen with high affinity (Kd = 4.7 nM) and specificity. Maximum parsimony and Neighbor Joining analyses were used to generate a phylogenetic classification of ccER on the basis of 18 full-length ER sequences. The tree suggested the existence of two major ER branches. One branch contained two clearly divergent clades which included all piscine ER (except Japanese eel ER) and all tetrapod ERalpha, respectively. The second major branch contained the eel ER and the mammalian ERbeta. The high degree of divergence between the eel ER and mammalian ERbeta suggested that they also represent distinct piscine and tetrapod ER. These data suggest that ERalpha and ERbeta are present throughout vertebrates and that these two major ER types evolved by duplication of an ancestral ER gene. Sequence alignments with other members of the nuclear hormone receptor superfamily indicated the presence of 8 amino acids in the E domain that align exclusively among ER. Four of these amino acids have not received prior research attention and their function is unknown. The novel finding of putative ER splice variants in a nonmammalian vertebrate and the novel phylogenetic classification of ER offer new perspectives in understanding the diversification and function of ER.

Crocodilian evolution: insights from immunological data.

The quantitative immunological technique of microcomplement fixation was used to examine serum albumin evolution among members of the order Crocodylia. The cross-reactivity of the albumin antisera and antigens employed in this study had been examined previously using the qualitative technique of immunodiffusion. The phylogenetic conclusions derived from these two data sets are highly congruent, including support of the families Alligatoridae and Crocodylidae, with the placement of Gavialis as the sister taxon of Tomistoma. Both methods provide similar information on the relative amounts of amino acid sequence divergence between albumin molecules; however, the data obtained from microcomplement fixation comparisons are more discriminating than those derived from immunodiffusion. The estimated divergence times within the Crocodylia derived from the fossil record are examined in light of divergence times predicted by the microcomplement fixation-based albumin clock. The traditional phylogenetic placement of Gavialis outside the remaining extant crocodilians is inconsistent with all molecular data sets and we suggest that a careful reexamination of both the extant and the fossil morphological data is warranted.

Length variation and heteroplasmy are frequent in mitochondrial DNA from parthenogenetic and bisexual lizards (genus Cnemidophorus).

Samples of mtDNA isolated from each of 92 lizards representing all color pattern classes of Cnemidophorus tesselatus and two populations of C. tigris marmoratus were digested with the restriction endonucleases MboI, TaqI, RsaI and MspI. The mtDNA fragment sizes were compared after radioactive labeling and gel electrophoresis. Three features were notable in the comparisons: there was little variation due to gain or loss of cleavage sites, two fragments varied noticeably in length among the samples, one by a variable amount up to a maximum difference of approximately 370 base pairs (bp) and the other by a discrete amount of 35 bp, these two fragments occasionally varied within, as well as between, samples. Two regions that corresponded in size to these variants were identified by restriction endonuclease cleavage mapping. One of these is adjacent to the D-loop. Heteroplasmy, heretofore rarely observed, occurred frequently in these same two regions. Variability in the copy number of a tandemly repeated 64-bp sequence appears to be one component of the variation, but others (e.g., base substitutions or small additions/deletions) must also be involved. The frequent occurrence of these length variations suggests either that they can be generated rapidly or that they were inherited from a highly polymorphic ancestor. The former interpretation is favored.