Ancient origins of RGK protein function: modulation of voltage-gated calcium channels preceded the protostome and deuterostome split

Abstract

RGK proteins, Gem, Rad, Rem1, and Rem2, are members of the Ras superfamily of small GTP-binding proteins that interact with Ca2+ channel β subunits to modify voltage-gated Ca2+ channel function. In addition, RGK proteins affect several cellular processes such as cytoskeletal rearrangement, neuronal dendritic complexity, and synapse formation. To probe the phylogenetic origins of RGK protein-Ca2+ channel interactions, we identified potential RGK-like protein homologs in genomes for genetically diverse organisms from both the deuterostome and protostome animal superphyla. RGK-like protein homologs cloned from Danio rerio (zebrafish) and Drosophila melanogaster (fruit flies) expressed in mammalian sympathetic neurons decreased Ca2+ current density as reported for expression of mammalian RGK proteins. Sequence alignments from evolutionarily diverse organisms spanning the protostome/deuterostome divide revealed conservation of residues within the RGK G-domain involved in RGK protein--Cavβ subunit interaction. In addition, the C-terminal eleven residues were highly conserved and constituted a signature sequence unique to RGK proteins but of unknown function. Taken together, these data suggest that RGK proteins, and the ability to modify Ca2+ channel function, arose from an ancestor predating the protostomes split from deuterostomes approximately 550 million years ago.

Figure 1. Heterologous expression of zebrafish RGK protein orthologs reduces I_Ca density in rat sympathetic neurons.

A. Current-voltage (I-V) plots in which mean ± SEM I_Ca density (pA/pF) is plotted versus test potential (mV). I_Ca was evoked from a holding potential of −80 mV to the test potentials indicated. I_Ca amplitude, determined 10 ms after initiation of the test pulse, was normalized to membrane capacitance (C_m). Control neurons (open circles) were injected with EGFP cDNA (n = 6). Neurons previously injected with Danio rerio RGK protein cDNA clones (100 ng/µl approximately 18–24 hours prior to recording) are depicted with filled symbols: dr_Gem (triangle, n = 11); dr_Rad (diamond, n = 7), dr_Rem1 (inverted triangle, n = 7), and dr_Rem2 (square, n = 9). B. Category plot of data shown in panel A for I_Ca density at +10 mV. Mean I_Ca density for neurons expressing RGK protein clones differed significantly (P<0.05) from control (one-way ANOVA followed by Dunnett's multiple comparisons test). C. Normalized I-V plots for control (open circle) and dr_Rem2 expressing (red filled square) neurons. Data, from panel A, was normalized to the maximal I_Ca density and plotted to illustrate similarity of voltage-dependence. D. Exemplar I_Ca traces acquired at +10 mV from control (black) and dr_Rem2 (red) expressing neurons. Traces are depicted without (left) and with normalization (right) to maximal current. Dotted line represents zero current level.

Figure 2. Fruit fly RGK-like protein homologs reduce I_Ca density in rat sympathetic neurons.

A. I-V plots in which mean ± SEM I_Ca density is plotted versus test potential. I_Ca was evoked and acquired as described for Fig. 1. Control neurons (open circles) were not injected with cDNA (n = 17). Neurons previously injected with Drosophila melanogaster RGK protein cDNA clones (50 ng/µl approximately 18–24 hours prior to recording) are depicted with filled symbols: dm_RGK1 (triangle, n = 13); dm_ RGK2t (diamond, n = 8), dm_RGK3 (inverted triangle, n = 8), and dm_RGK3L (square, n = 9). B. Category plot of data shown in panel A for I_Ca density at +10 mV. Mean I_Ca density for neurons expressing RGK protein clones differed significantly (P<0.05) from control (one-way ANOVA followed by Dunnett's multiple comparisons test). C. Normalized I-V plots for control (open circle) and dm_RKG2t expressing (red filled diamond) neurons. Data, from panel A, was normalized to the maximal I_Ca density and plotted to illustrate similarity of voltage-dependence. D. Exemplar I_Ca traces acquired at +10 mV from control (black) and dm_RGK2t (red) expressing neurons. Traces are depicted without (left) and with normalization (right) to maximal current. Dotted line represents zero current level.

Figure 3. Phylogenetic tree for RGK protein orthologs.

Phylogenetic tree layout was adapted from Yau and Hardie . The deuterostome branch is labeled in green with dark green representing the phylum chordata. Tunicates and cephalochordates are subdivisions within this phylum with vertebrates representing the major group. Echinoderms (e.g., starfish, sea urchins, sea cucumbers) represent the second largest grouping of deuterostomes and are labeled in light green. The protostome branch is depicted below the dotted line in blue. Major phyla for which RGK protein orthologs/homologs were identified are presented in blue and the others in gray. Organisms with two letter abbreviation and common name (in parentheses) are depicted on the right.

Figure 4. Alignment of RGK protein G-domains.

G-domain sequences were aligned using the ClustalW algorithm in MacVector (version 12.7.5). Sequence labels (left) are color coded with vertebrates in green, non-vertebrate deuterostomes in pink, and protostomes in blue. The black horizontal line divides the deuterostomes/protostome sequences. Organism genus and species abbreviations (e.g., hs for Homo sapiens) are depicted in Figure 3. The method for parsing the RGK protein G-domain sequence is documented in the text and supplement (table S1) along with GenBank accession numbers for each sequence. The G-motifs (G1–5) along with canonical RGK protein sequences for G1, G3, G4, and G5 are depicted above the alignments. The residue numbers are for human Gem (top row) and are right justified. The inverted black triangles represent key residues participating in RGK protein–Ca_vβ interaction. Secondary structural domains (α-helices and β-sheets) are depicted schematically below the sequences. Sequence identity color scale is depicted in the lower right corner.

Figure 5. Alignment of RGK protein C-termini.

A. The last 40 residues of RGK and RGK-like protein sequences were aligned using the ClustalW algorithm in MacVector (version 12.7.5). Sequence labels and identity color-coding are as shown in Figure 4. Previously defined functional domains identified in vertebrate RGK proteins are depicted schematically above the sequences. Residue labeling is for hs Gem (lower numbering) or in residues from the stop codon (upper numbering). B. Sequence logo depiction of the terminal 20 amino acids including the RGK protein signature C-7 domain. Web logo was created with WebLogo 3 (http://weblogo.threeplusone.com) .

Figure 6. Location of RGK protein structural determinants.

Shown are two perspectives, related by a 90° rotation along the drawn axis, of the RGK G-domain in surface representation. Secondary structure is depicted along with labeled residues underneath the surface. Nucleotide (GDP) is shown in a bond representation and Mg²⁺ is seen as a magenta sphere. The red residues are the amino acids essential for RGK-Ca_V β association as denoted by inverted black triangles in Figure 4. Alpha helix 5 is shown in pink and the absolutely conserved Ala from alpha helix 4 is shown in orange. Residue numbering is based on Homo sapiens Gem (PDB code: 2HT6).