Evolution of the α-Subunit of Na/K-ATPase from Paramecium to Homo sapiens: Invariance of Transmembrane Helix Topology
Na/K-ATPase is a key plasma membrane enzyme involved in cell signaling, volume regulation, and maintenance of electrochemical gradients. The α-subunit, central to these functions, belongs to a large family of P-type ATPases. Differences in transmembrane (TM) helix topology, sequence homology, helix–helix contacts, cell signaling, and protein domains of Na/K-ATPase α-subunit were compared in fungi (Beauveria), unicellular organisms (Paramecia), primitive multicellular organisms (Hydra), and vertebrates (Xenopus, Homo sapiens), and correlated with evolution of physiological functions in the α-subunit. All α-subunits are of similar length, with groupings of four and six helices in the N- and C-terminal regions, respectively. Minimal homology was seen for protein domain patterns in Paramecium and Hydra, with high correlation between Hydra and vertebrates. Paramecium α-subunits display extensive disorder, with minimal helix contacts. Increases in helix contacts in Hydra approached vertebrates. Protein motifs known to be associated with membrane lipid rafts and cell signaling reveal significant positional shifts between Paramecium and Hydra vulgaris, indicating that regional membrane fluidity changes occur during evolution. Putative steroid binding sites overlapping TM-3 occurred in all species. Sites associated with G-protein-receptor stimulation occur both in vertebrates and amphibia but not in Hydra or Paramecia. The C-terminus moiety “KETYY,” necessary for the Na+ activation of pump phosphorylation, is not present in unicellular species indicating the absence of classical Na+/K+-pumps. The basic protein topology evolved earliest, followed by increases in protein domains and ordered helical arrays, correlated with appearance of α-subunit regions known to involve cell signaling, membrane recycling, and ion channel formation.
KeywordsNa/K-ATPase α-subunit Evolution Protein domains Transmembrane helix Cell signaling Helix–helix interactions
Na/K-ATPase, a highly conserved integral membrane protein, is key to maintenance of both cell volume and electrochemical gradients (Kaplan 2002; Blanco 2005; Geering 2008), as well as cell signaling (Reinhard et al. 2013), has origins that go back to the prokaryotes (reviewed in Saez et al. 2010; Chan et al. 2010). Its α-subunit, responsible for many of its known functions, belongs to a large family of P-type ATPases (reviewed in [Kaplan 2002; Bublitz et al. 2011]). The P-type ATPases, also known as E1–E2 ATPases, are a large group of evolutionarily related ion and lipid transporters that are found in bacteria, archaea, and eukaryotes (Saez et al. 2010). Each consists of bundles of 10 transmembrane (TM) α-helices and is referred to as a P-type ATPase because they catalyze autophosphorylation of a key conserved aspartate residue within the α-subunit. Thever and Saier (2009) analyzed the fully sequenced genomes of 26 eukaryotes for P-type ATPases and reported probable topologies and conserved motifs of 9 functionally characterized families and 13 uncharacterized families of these transporters. Studies by Saez et al. (2010) indicate that the likely origin of many of the P-type (subfamily IIC) proteins is prokaryotic, and that many are present in non-metazoans, such as algae, protozoans, or fungi. They also propose that early deuterostomes presented a single IIC gene, from which all the extant diversity of vertebrate IIC proteins originated by gene and genome duplications.
In this study, we compare the differences in TM helix topology, sequence homology, helix–helix contacts, cell signaling, disordered protein regions, and protein domains of Na/K-ATPase α-subunits in several organisms: (1) Beauveria bassiana, a fungus, (2) Paramecium tetraurelia, a unicellular ciliate protozoa, (3) Hydra vulgaris, a primitive multicellular organism with two main body layers separated by a gel, and (4) both lower (Xenopus laevis) and higher vertebrates (Homo sapiens). Since the P-type ATPases appear to have evolved into multifunctional integral plasma membrane enzymes during evolution from Paramecium to H.sapiens, it may be possible to correlate sequential structural changes in the primitive Na+-pump (Na/K-ATPase) with the appearance of specialized functions (e.g., cellular signaling, membrane recycling) and in turn with specific sequence changes. As indicated, each α-subunit contains about 1000 residues and 10 TM helices. Structure–function analysis indicates that evolution was due in large part to progressive increases in membrane protein domains and helix–helix interactions. Thus the basic protein topology evolved very early, followed by selective amino acid sequence changes over millions of years that resulted in a complex system that controls cell division, growth, and differentiation.
Protein Sequence Sources
The amino acid sequences of the steroid binding proteins were downloaded from the ExPASy Proteomic Server of the Swiss Institute of Bioinformatics (http://www.expasy.org; http://www.uniprot.org). About 98 % of the protein sequences provided by UniProtKB are derived from the translation of the coding sequences (CDS) which have been submitted to the public nucleic acid databases, the EMBL-Bank/Genbank/DDBJ databases (INSDC). The present study uses the complete amino acid sequences of the α-subunits of the Na/K-ATPase from B. bassiana (BEAB2, Accession #J5JMV7), Paramecium tetraurelia (ATP1A_PARTE, Accession #Q6BGF7), H. vulgaris (AT1A_HYDVU, Accession #P35317), X. laevis (AT1A1_XENLA, Accession #Q92123), and H. sapiens (AT1A1_HUMAN, Accession #P05023).
Secondary Structure Predictions
TM helices were predicted using (1) MEMSAT-SVM (Nugent and Jones 2009), (http://www.bioinf.cs.ucl.ac.uk/psipred/), (2) TMpred (Krogh et al. 2001), (http://www.ch.embnet.org/software/tmbase/TMBASF.doc,html), (3) SPOCTOPUS algorithm (Viklund et al. 2008) (http://octopus.cbr.su.se), and (4) MemBrain, which integrates a number of recent bioinformatic approaches including the optimized evidence-theoretic K-nearest neighbor algorithm (Shen and Chou 2008) available at http://chou.med.harvard.edu/bioinf/MemBrain/. Pore-lining regions in TM protein sequences were predicted using the method of Nugent and Jones (2012).
The contribution of intrinsic disorder to protein function and identification of functional sites in disordered regions was estimated by the method of Cozzetto and Jones (2013). Protein domain boundary prediction was estimated using the DomPred server (Bryson et al. 2007) and the Membrain server (Yang et al. 2013). The methods of Bryson et al. (2007) are available at http://bioinf.cs.ucl.ac.uk/software.html. The method of Yang et al. is available at http://www.csbio.sjtu.edu.cn/bioinf/MemBrain .
TMKink: A Method to Predict Transmembrane Helix Kinks
Meruelo et al. (2011) have identified distinct residue preferences in kinked versus non-kinked helices and have exploited these differences and residue conservation to predict kinked helices using a neural network. The kink predictor, TMKink, is available at http://tmkinkpredictor.mbi.ucla.edu/.
Helical Packing Arrangement Predictions
The MEMPACK prediction server (firstname.lastname@example.org) was used to predict lipid exposure, residue contacts, helix–helix interactions, and helical packing arrangement, in addition to TM topology. The MemBrain method (http://csbio.sjto.edu.cn/bioinf/MemBrain/) was used to derive TM inter-helix contacts from amino acid sequences by combining correlated mutations and multiple machine-learning classifiers (Yang et al. 2013). The TOPCONS web interface (http://topcons.cbr.su.se/pred/result/rst_j7ZjE1) allows for constraining parts of the protein sequence to a known inside/outside location to be displayed both graphically and in text format (Bernsal et al. 2009).
Results and Discussion
Comparison of Transmembrane Topology of Na/K-ATPase α-Subunits in Paramecia, Hydravulgaris, and Homo sapiens
As shown in Fig. 1, the MemBrain algorithm predicts 10 TM helices in H. sapiens and Hydravulgaris but only 9 TM helices plus 2 “half helix” TMs in Paramecia (PARTE). For comparison, other servers [e.g., SPOCTOPUS, (Viklund et al. 2008) MEMSAT-SVM, (Nugent and Jones 2012) and Phobius, (Kall et al. 2007)] predict that each species contains 10 TM helices. As can be seen, there is more variability in the C-terminal region, with apparent shifts in position of TM helices 5, 6, and 7 during the evolution of early multicellular organisms (i.e., Paramecium to Hydra). A similar MemBrain analysis of the topology of other mammalian P-type IIC enzymes Ca-ATPase (SERCA), H-ATPase, H/K-ATPase, and phospholipid flippase predicted that all contain 10 TM helices, with two pairs of TM helices in the N-terminal region, and a group of 6 helices at the C-terminal end (data not shown).
Similarities between helices of the α-subunit of Na/K-ATPase in Xenopus laevis, Hydra vulgaris, and Paramecium tetraurelia compared to Homo sapiens
Transmembrane (TM) helices
% Identity of amino acid sequences of individual transmembrane helices (% Amino acid similarities indicated in parenthesis)
The Emboss Water protocol (version 36.3.5e Nov, 2012; preload8) used here employs the Smith–Waterman algorithm (with modified enhancements) to estimate the local alignment of two sequences (Huang and Miller 1991). Comparison of H.sapiens and H. vulgaris α-subunits revealed a Waterman–Eggert score of 4825 with 70.9 % identity (88.5 % similar) in 1020 amino acid overlap (8-1023:14-1031). Comparison of H. sapiens and Paramecium α-subunits indicated a Waterman–Eggert score of 2404 with 46.6 % identity (74.0 % similar) in 882 amino acid overlap (40-886:99-969).
Putative Domain Boundaries in Human, Hydra, and Paramecium α-Subunits
The shortest sequence of amino acids in proteins that contains functional and structural information is termed a “motif,” whereas a conserved part of a given protein that can evolve, function, and exist independently is termed a “domain.” Domains form the structural basis of the physiological functions of proteins and each domain can be considered as a semi-independent structural unit of a protein capable of folding independently (Wetlaufer 1973; Richardson 1981; Vogel et al. 2004). A variety of different methodologies have been employed to predict domains but many are fraught with problems since domain assignment is difficult even when the structures are known. Bryson et al. (e.g., 2007) have developed a useful method for computer-assisted protein domain boundary prediction, using the DomPred server (see Methods). This server uses the results from two completely different categories (DPS and DomSSEA). Each is individually compared against one of the latest domain prediction benchmarks to determine their respective reliabilities.
Evolution of Na/K-ATPase as a Signal Transducer
Comparison of different regulatory systems associated with cell signaling in the Na/K-ATPase α-subunit
Activation of phosphatidylinositide 3-Kinase 1A/Akt
Putative progesterone cell surface binding site first external loop
C-terminal contacts that stabilize Na-pump conformations
Homo sapiens P05023
Xenopus laevis Q92123
Hydra vulgaris P35317
Beauveria bassiana J5JMV7
Vicia faba Q43131
Studies using isolated plasma-vitelline membranes from Rana pipiens oocytes have shown that progesterone may act as a meiotic agonist by binding to the external loop sequence (QAATEEEPQN) between TM-1 and TM-2 helices of the α-subunit (Morrill et al. 2008) and rapidly activate lipid enzymes such as N-methyl transferase and sphingomyelin synthase (Morrill et al. 2010). As shown in Table 2, both H. sapiens and X. laevis contain the putative progesterone binding site, whereas Hydra vulgaris and lower forms do not. Voltage-clamp measurements by Vedovato and Gadsby (2010), using intact Xenopuslaevis oocytes indicate that the two C-terminal tyrosines (YY) of the α-subunit stabilize Na/K-pump conformations. Deletion of the last five residues (KETYY) of the α-subunit markedly lowers the apparent affinity for Na+ activation of pump phosphorylation (Vedovato and Gadsby 2010). Table 2 further indicates that humans, frogs, and hydra all contain the C-terminal KET(or S)YY contacts, whereas paramecia and the plasma membrane ion transporter ATPase of plant (Vicia faba) cells do not. This indicates that both the P13K1A-PDK-Akt pathway and the C-terminal YY residues play important roles in primitive multicellular species such as H. vulgaris (column 2), but not in lower species. Table 2 also indicates that progesterone binding does not contribute to H. vulgaris physiology, suggesting that a steroid response system(s) arose at a later stage of evolution.
Predicting Helix–Helix Interactions in the α-Subunits Based on Residue Contacts
Comparison of Pore-Lining Regions, Hydrophobic Cores, and Disordered Regions in α-Subunits of Several Species
Comparison of pore-lining regions and predicted kinks by position in Paramecia tetraurelia,Hydra vulgaris, and Homo sapiens
Transmembrane (TM) helix
The emerging pattern suggests that the catalytic α-subunit of Na/K-ATPase must have evolved as a result of the sequential introduction of protein domains within the C-terminal region of a ~ 1000 residue precursor polypeptide. The α-subunit of a fungus, B. bassiana exhibits a similar number of amino acid residues (1107 and 1023) and the same topology as that of H. sapiens; e.g., 2 pairs of TM helices in the N-terminal region and a cluster of 6 TM helices in the C-terminal region (Figs. 1, 6, 7). The TM helices and pore-lining regions (Figs. 6, 7) probably evolved early and the topology has remained unchanged over millions of years. A primary difference between Beauveria, Paramecium, and the early multicellular organism (H. vulgaris) is the progressive increase in the number and complexity of the protein domains (Figs. 2, 7).
The membrane potential of most fungal cells is about −150 mV [cf. (Shi et al. 2002)]. This high membrane potential is the driving force that supports the function of many uniporters, symporters, and antiporters that fungi have evolved to balance excessive Na+ uptake. The limited distribution of protein domains to the C-terminal region of fungi indicates that the uniporters, symporters, and/or antiporters are largely associated with protein domains within the first 250 residues of the N-terminal region (compare Figs. 2, 7). Paramecia lack the classical Na/K-ATPase, since K+ does not stimulate Na+ efflux (Hansma 1979), consistent with the absence of the C-terminal KETYY sequence (Table 2). In Paramecium, ionic currents control swimming behavior (Eckert and Brehm 1979). When its membrane potential is at the resting level, Paramecium swims forward. When its membrane is hyperpolarized it swims backward. A closely related ciliate, Tetrahymena, also appears to lack a classical pump Na/K-ATPase, since the specific inhibitor, ouabain, has no effect on Na+ and K+ transport (Dunham and Kropp 1983). The absence of the C-terminal KETYY sequence in both fungi and Paramecium (Table 2) is evidence for limited α-subunit functionality in lower species.
As shown in Fig. 2, the catalytic subunit of a primitive multicellular organism, H. vulgaris, contains many of the protein domains characteristic of the ATP1A1 α-subunit of H. sapiens, indicating that much of Na-pump evolution occurred during the development of multicellular organisms. The ectodermal epithelium of the fresh water H. vulgaris is the main site of ionic regulation in these organisms (Chain 1980). It actively transports Na+ from the environment into the enteron and extracellular fluids, which are the two milieus that act as these animals’ only “internal environment.” A major step in electrical potential occurs across the inner ectodermal membrane; produced by an inwardly directed electrogenic pump which is sensitive to ethacrynic acid but not to ouabain (Chain 1980). As indicated in Table 2, the C-terminus of the H. vulgaris α-subunit contains the moiety “KETYY,” reported to be necessary for the Na+ activation of pump phosphorylation from ATP in vertebrates (Vedovato and Gadsby 2010). The lack of “KETYY” in Paramecium and B. bassiana indicates that neither contains classical Na+/K+-pumps.
Comparison of caveolin-binding motifs and leucine-rich repeat sequences in the α-subunit of various animal and plant cells
Homo sapiens (P05023)
89WIKFCRQLFGGFSMLLW105 Overlaps TM-1
987WWFCAFPYSLLIFVY1001 Overlaps TM-10
Xenopus laevis (Q92123)
91WVKFCRQLFGGF102 Overlaps TM-1
989WWFCAFPYSLIIFIY1003 Overlaps TM-10
Hydra vulgaris (P35317)
95WVKFCKQMFGGF106 Overlaps TM-1
Paramecium tetraurelia (Q6BGF7)
650FKLEGFTF657 Between TM-4 and TM-5
1052YYDLRYIFVYYDQNYQRW1069 Between TM-7 and TM-8
Vicia faba (Q43131)
91WVKFCRQLFGGF102 Overlaps TM-1
989WWFCAFPYSLIIFIY1003 Overlaps TM-10
The observations presented here indicate that a number of molecular traits of the Na/K-ATPase catalytic subunit became determined by the single-cell stage. These include protein sequence length, TM helix topology, and protein motifs such as the Leucine-rich repeats. Subsequent evolutionary changes appear to be due largely to introduction of protein domains into the catalytic subunits during evolution. A protein domain has been defined as a conserved part of a given protein sequence that can evolve, function, and exist independently (reviewed in Vogel et al. 2004). Each domain forms a compact three-dimensional structure and often can be independently stable and folded to create proteins serving different functions. Domains vary in length from about 25 amino acids to about 500 amino acids in length. Because they are independently stable, they may be exchanged between proteins leading to the evolution of protein families such as the P-type ATPases, which by the end of 2014 had 493 confirmed and unique members in the Swiss-Prot Database (George et al. 2004, Prosite motif PS00154).
This research was supported in part by National Institutes of Health research Grants HD-10463, GM-071324, and NHLBI Grant HL-36573.
GM and AK conceived of the study. AA, RG, and LL contributed protein structural concepts and analyzed data, and GM wrote the paper. All authors read and approved the final manuscript.
Compliance with Ethical Standards
Conflict of Interest
The authors declare that they have no conflict of interest.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.