Protein cold adaptation strategy via a unique seven-amino acid domain in the icefish (Chionodraco hamatus) PEPT1 transporter

Abstract

Adaptation of organisms to extreme environments requires proteins to work at thermodynamically unfavorable conditions. To adapt to subzero temperatures, proteins increase the flexibility of parts of, or even the whole, 3D structure to compensate for the lower thermal kinetic energy available at low temperatures. This may be achieved through single-site amino acid substitutions in regions of the protein that undergo large movements during the catalytic cycle, such as in enzymes or transporter proteins. Other strategies of cold adaptation involving changes in the primary amino acid sequence have not been documented yet. In Antarctic icefish (Chionodraco hamatus) peptide transporter 1 (PEPT1), the first transporter cloned from a vertebrate living at subzero temperatures, we came upon a unique principle of cold adaptation. A de novo domain composed of one to six repeats of seven amino acids (VDMSRKS), placed as an extra stretch in the cytosolic COOH-terminal region, contributed per se to cold adaptation. VDMSRKS was in a protein region uninvolved in transport activity and, notably, when transferred to the COOH terminus of a warm-adapted (rabbit) PEPT1, it conferred cold adaptation to the receiving protein. Overall, we provide a paradigm for protein cold adaptation that relies on insertion of a unique domain that confers greater affinity and maximal transport rates at low temperatures. Due to its ability to transfer a thermal trait, the VDMSRKS domain represents a useful tool for future cell biology or biotechnological applications.

Fig. 1.

Sequence and expression analysis of icefish PEPT1. (A) Model of icefish PEPT1. The picture was obtained using TeXtopo (47). TMDs I–XII were predicted using TMHMM (transmembrane prediction using hidden Markov models) 2.0. N-glycosylation (yellow ●), cAMP-dependent protein kinase A (PKA; purple ●), PKC (azure ●), and overlapped PKA/PKC (orange ●) sites; PTR2 family proton/oligopeptide symporters signature 1 (blue) and 2 (green) motifs were identified using PROSITE 20.24. The six VDMSRKS domains close to the COOH terminus are indicated alternately in red and pink. (B) Unrooted phylogenetic tree of vertebrate PEPT1 transporters. Alignments were performed using ClustalW2. The unrooted tree (1,000 times bootstrap) was constructed using neighbor-joining analysis with data corrected for multiple hits through Poisson distribution. (C) Tissue distribution of icefish pept1 mRNA. Expression levels of icefish pept1 mRNA in different tissues determined by qualitative RT-PCR (Upper) and real-time PCR (Lower) analysis. Relative values are expressed with respect to the intestinal signal. (D) Comparison of the COOH-terminal regions of PEPT1 from four Antarctic teleosts. The alignment (ClustalW2) indicates TMD XI and XII and the conserved seven-amino acid domains (D1–D6).

Fig. 2.

Functional characterization of icefish PEPT1. (A) Steady-state current–voltage relationships for GQ. Icefish PEPT1-expressing oocytes were perfused at pH 6.5 in the presence of increasing (0.05–5 mM) GQ concentrations. (B) Kinetic parameters of GQ transport. Apparent GQ affinity (K_0.5) and maximal transport current (I_max) were calculated at −60 and −120 mV by least-squares fitting to the Michaelis-Menten equation. I_max values are expressed as the percentage of the value calculated at pH 6.5 in the same experiment. n, number of oocytes; N, number of frogs. (C and D) Membrane potential and pH dependence of K_0.5 and I_max of GQ transport. I_max values have been normalized to those measured at −160 mV.

Fig. 3.

Temperature sensitivity of wild-type and mutant icefish PEPT1. (A) Schematic representation of wild-type and mutant PEPT1 transporters. (B) Q₁₀ values of wild-type and mutant transporters. Q₁₀ was calculated over the temperature range 280–300 K (7–27 °C) in the presence of saturating concentrations of GQ (20 mM). (C) Arrhenius plots of wild-type and mutant transporters. Transport currents were calculated at −60 mV and different temperatures (280–300 K). Currents were normalized to that at 300 K. Lines represent the linear regression of the transformed data. (D) Activation energy values (E_a) of wild-type and mutant transporters. Data were calculated from the slope of the curves reported in C. For all analyses, values were means ± SE (n = 6–9). a, difference vs. icefish PEPT1; b, difference vs. icefish PEPT1-∆6 (one-way ANOVA/Bonferroni post hoc test). **P < 0.01.

Fig. 4.

Temperature dependence of chimeric rabbit–icefish PEPT1. (A) Schematic representation, (B) Q₁₀ values, (C) Arrhenius plots, and (D) activation energy values (E_a) of wild-type and chimeric PEPT1 transporters. Experimental details are the same as for Fig. 3. a, difference vs. icefish PEPT1; b, difference vs. rabbit PEPT1 (one-way ANOVA/Bonferroni post hoc test).