Tingaud-Sequeira A et al. (2008), Structural and functional divergence of two fis...

XB-ART-38426

BMC Evol Biol 2008 Sep 23;8:259. doi: 10.1186/1471-2148-8-259.

Show Gene links Show Anatomy links

Structural and functional divergence of two fish aquaporin-1 water channels following teleost-specific gene duplication.

Tingaud-Sequeira A , Chauvigné F , Fabra M , Lozano J , Raldúa D , Cerdà J .

???displayArticle.abstract???
BACKGROUND: Teleost radiation in the oceans required specific physiological adaptations in eggs and early embryos to survive in the hyper-osmotic seawater. Investigating the evolution of aquaporins (AQPs) in these vertebrates should help to elucidate how mechanisms for water homeostasis evolved. The marine teleost gilthead sea bream (Sparus aurata) has a mammalian aquaporin-1 (AQP1)-related channel, termed AQP1o, with a specialized physiological role in mediating egg hydration. However, teleosts have an additional AQP isoform structurally more similar to AQP1, though its relationship with AQP1o is unclear. RESULTS: By using phylogenetic and genomic analyses we show here that teleosts, unlike tetrapods, have two closely linked AQP1 paralogous genes, termed aqp1a and aqp1b (formerly AQP1o). In marine teleosts that produce hydrated eggs, aqp1b is highly expressed in the ovary, whereas in freshwater species that produce non-hydrated eggs, aqp1b has a completely different expression pattern or is not found in the genome. Both Aqp1a and Aqp1b are functional water-selective channels when expressed in Xenopus laevis oocytes. However, expression of chimeric and mutated proteins in oocytes revealed that the sea bream Aqp1b C-terminus, unlike that of Aqp1a, contains specific residues involved in the control of Aqp1b intracellular trafficking through phosphorylation-independent and -dependent mechanisms. CONCLUSION: We propose that 1) Aqp1a and Aqp1b are encoded by distinct genes that probably originated specifically in the teleost lineage by duplication of a common ancestor soon after divergence from tetrapods, 2) Aqp1b possibly represents a neofunctionalized AQP adapted to oocytes of marine and catadromous teleosts, thereby contributing to a water reservoir in eggs and early embryos that increases their survival in the ocean, and 3) Aqp1b independently acquired regulatory domains in the cytoplasmatic C-terminal tail for the specific control of Aqp1b expression in the plasma membrane.

???displayArticle.pubmedLink??? 18811940
???displayArticle.pmcLink??? PMC2564943
???displayArticle.link??? BMC Evol Biol

Species referenced: Xenopus laevis
Genes referenced: aqp1 csnk1a1 csnk2b krt8.1 thoc1

???attribute.lit??? ???displayArticles.show???

	Figure 1. Phylogenetic relationships of AQP1-like proteins in vertebrates. Bayesian majority rule consensus phylogenetic tree for the amino acid alignment of teleost and tetrapod AQP1-like sequences. Nodes with â¥70% Bayesian posterior probabilities are shown. Branch lengths are proportional to BI estimates of numbers of amino acid substitutions. The GenBank accession number, scaffold, or chromosome group are indicated in parenthesis for each sequence.
	Figure 2. Genomic organization of human AQP1 and teleost aqp1a and aqp1b genes. Schematic representation of human AQP1 [67], and zebrafish, fugu, medaka and sea bream aqp1a and aqp1b gene loci. White (human AQP1 and teleost aqp1a) and grey (teleost aqp1b) boxes indicate exons with coding regions only. Downstream genes from human AQP1 and teleost aqp1b are growth hormone releasing hormone receptor (GHRHR) and THO complex subunit 1 (thoc1), respectively.
	Figure 3. Amino acid sequence alignment of human AQP1 and teleost Aqp1a and Aqp1b. The six transmembrane (TM) domains and connecting loops A-B of human AQP1 are indicated by brackets and horizontal arrows, respectively. The vertical arrows above human AQP1 show the conserved residues Phe56, His180 and Arg195 (human AQP1 numbering) in water-selective AQPs. Identical residues between human AQP1 and teleost AQP1-like sequences are indicated with an asterisk, whereas conserved amino acid substitutions and substitutions with similar amino acids are indicated by a double or single dot, respectively. Residues conserved in human and most teleost sequences are shaded in black, and residues different from human but conserved between most of the teleost Aqp1a and Aqp1b sequences are shaded in grey.
	Figure 4. Gene expression pattern and functional characterization of teleost Aqp1b. (A-D) Representative RT-PCR analysis of aqp1b (upper panels) and bactin (lower panels) transcripts in sea bream (A), European eel (B), Senegalese sole (C) and zebrafish (D) tissues. PCR products were detected by Southern blot. Minus indicates absence of RT during cDNA synthesis. The size (kb) of PCR products and molecular markers are indicated on the left and right, respectively. (E) Pf and Hg2+ inhibition of X. laevis oocytes expressing teleost Aqp1a or Aqp1b. Oocytes were injected with cRNAs encoding sea bream Aqp1a or Aqp1b (1 ng), eel Aqp1b (10 ng), Senegalese sole Aqp1b (10 ng) or zebrafish Aqp1b (10 ng), or with 50 nl of water (control). The Pf was assayed in the presence or absence of 0.7 mM HgCl2. Some oocytes treated with HgCl2 were incubated with 5 mM Î²-mercaptoethanol (Î²ME) for 15 min before swelling measurements. Values represent the mean Â± SEM (n = 6â10 oocytes) from a representative experiment.
	Figure 5. Differential localization of sea bream Aqp1a and Aqp1b expressed in X. laevis oocytes. (A) Pf of oocytes expressing increasing amounts of Aqp1a or Aqp1b cRNA. Values represent the mean Â± SEM (n = 6â10 oocytes) from 3 independent experiments. (B) Immunoblots of total and plasma membrane equivalents (TM and PM, respectively) of oocytes expressing 1 ng of Aqp1a or Aqp1b. The arrows indicate two very close Aqp1b reactive bands. (C-E) Immunofluoresence microscopy of water-injected and Aqp1a- or Aqp1b-expressing oocytes. Arrows show localization of the protein at the plasma membrane. The arrowhead indicates Aqp1b in the cytoplasm below the plasma membrane. Bar, 50 Î¼m. (F) Immunoblot of total membrane fraction of Aqp1b-expressing oocytes incubated with or without alkaline phosphatase (AP) for 6 h. In B and F, the apparent molecular mass of a 29-kDa marker is indicated on the left and right, respectively.
	Figure 6. Functional properties and subcellular localization of wild-type (WT) sea bream Aqp1a and Aqp1b, and of their chimeric proteins, in oocytes. (A) Membrane topology of AQP family members showing the six transmembrane helices with five connecting loops (A-E), and two conserved Asn-Pro-Ala (NPA) motifs in loops B and E. (B-E) WT Aqp1a (B) and Aqp1b (C), Aqp1a chimera in which the C-terminus of Aqp1a was exchanged with that of Aqp1b (Aqp1a-Ct1b; D), and Aqp1b chimera in which the C-terminus of Aqp1b was exchanged with that of Aqp1a (Aqp1b-Ct1a; E). (F) Pf of oocytes expressing 1 ng of cRNA encoding WT or chimeric proteins. Values represent the mean Â± SEM (n = 5â8 oocytes) from a representative experiment. The asterisk denotes statistically significant differences (p < 0.01). (G-J) Immunofluorescence microscopy of oocytes localizing WT Aqp1a and Aqp1b-Ct1a exclusively at the plasma membrane, whereas WT Aqp1b and Aqp1a-Ct1b are also in the cytoplasm. Sections shown in G and J were probed with the anti-Aqp1a antisera, whereas the sections in H and I were probed with the anti-Aqp1b antisera. Bar, 50 Î¼m. (K-L) Immunoblots of total and plasma membrane equivalents (TM and PM, respectively) of oocytes expressing the differents cRNAs. Blots were probed as indicate above. The apparent molecular mass of a 29-kDa marker is indicated on the left.
	Figure 7. Amino acid sequence alignment of the C termini of human AQP1, and Aqp1a and Aqp1b from representative teleosts. At the bottom, identical residues are indicated by asterisks, whereas conserved amino acid substitutions and substitutions with similar amino acids are indicated by double or single dots, respectively. Residues of teleost Aqp1a and Aqp1b conserved in human AQP1 are in bold, and conserved residues in Aqp1a sequences are boxed. Double underlined residues indicate typical potential sorting and internalization sequences, whereas those single underlined indicate other sorting-like motifs. In each sequence, potential Ser, Thr and Tyr phosphorylation sites (score â¥ 0.9) are indicated by arrowheads. Consensus sites for potential kinases are indicated: grey, candidate Pro-directed kinase and preceding docking domain; arrows, casein kinase I (CK1); circles, CK2. Other candidate phosphorylation sites shown do not match any eukaryotic linear functional motif included in the ELM resource.
	Figure 8. Role of specific residues in the sea bream Aqp1b C-terminal tail for intracellular trafficking in oocytes. (A) Water permeability of oocytes expressing wild-type (WT) or mutant Aqp1b. Oocytes were injected with cRNAs encoding WT Aqp1b (0.25 or 1 ng), Aqp1b-T229A (1 ng), Aqp1b-L234A/L235A (1 ng), Aqp1b-S254A (0.25 ng) or Aqp1b-S254D (0.25 ng). Permeability is expressed in % related to oocytes injected with WT Aqp1b. Values are the mean Â± SEM of 3â5 experiments (n = 10â15 oocytes per treatment). The asterisks denote statistically significant differences (, p < 0.05; *, p < 0.01). (B-E) Immunoblots of total and plasma membrane equivalents (TM and PM, respectively) of oocytes expressing WT or mutant Aqp1b. The apparent molecular mass of a 29-kDa marker is indicated on the left. (F-J) Localization of Aqp1b mutants in oocytes. The plasma membrane is indicated by arrows, and retention of Aqp1b-L234A/L235A proteins possibly in the ER is indicated by arrowheads (H). Bars, 100 Î¼m.

References [+] :

Abascal, ProtTest: selection of best-fit models of protein evolution. 2005, Pubmed