(C) PLOS One This story was originally published by PLOS One and is unaltered. . . . . . . . . . . Evolutionary history of calcium-sensing receptors unveils hyper/hypocalcemia-causing mutations [1] ['Aylin Bircan', 'Faculty Of Engineering', 'Natural Sciences', 'Sabanci University', 'İstanbul', 'Nurdan Kuru', 'Onur Dereli', 'Berkay Selçuk', 'Ogün Adebali', 'Tübi̇tak Research Institute For Fundamental Sciences'] Date: 2025-01 Despite advancements in understanding the structure and functions of the Calcium Sensing Receptor (CaSR), gaps persist in our knowledge of the specific functions of its residues. In this study, we used phylogeny-based techniques to identify functionally equivalent orthologs of CaSR, predict residue significance, and compute specificity-determining position (SDP) scores to understand its evolutionary basis. The analysis revealed exceptional conservation of the CaSR subfamily, emphasizing the critical role of residues with high SDP scores in receptor activation and pathogenicity. To further enhance the findings, gradient-boosting trees were applied to differentiate between gain- and loss-of-function mutations responsible for hypocalcemia and hypercalcemia. Lastly, we investigated the importance of these mutations in the context of receptor activation dynamics. In summary, through comprehensive exploration of the evolutionary history of the CaSR subfamily, coupled with innovative phylogenetic methodologies, we identified activating and inactivating residues, providing valuable insights into the regulation of calcium homeostasis and its connections to associated disorders. We applied machine learning methods to differentiate between mutations that cause either a gain or loss of function in CaSR, leading to disorders like hypercalcemia and hypocalcemia. Our analysis highlights specific regions of the receptor that are essential for its activity, providing insights into how mutations can disrupt calcium homeostasis. CaSR plays a crucial role in maintaining calcium balance in the body. Our study investigates the evolutionary history of CaSR to better understand its function and the impact of mutations. By using phylogenetic techniques, we identified key residues that are critical for CaSR’s function. These residues are highly conserved, indicating their importance in receptor activation and calcium regulation. Funding: This study was supported by EMBO Installation Grant no:4163 funded by TÜBİTAK (to OA). OA is supported by the BAGEP program of the Science Academy - Türkiye, and the TÜBA-GEBİP program of the Turkish Academy of Sciences. BS is supported by EMBO Installation Grant no:4163 funded by TÜBİTAK. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Here, we show the importance of each residue in CaSR by comparing it with the closely related subfamilies, GPRC6A and TAS1Rs. We identified all orthologous sequences in each subfamily by building phylogenetic trees and manually curating the duplications/speciations on the tree to obtain all functionally equivalent orthologs within each subfamily. To obtain members of a subfamily without requiring computationally expensive phylogenetic tree building and manual curation steps, we generated highly sensitive subfamily-specific profile hidden Markov models (HMMs) by using the functionally equivalent orthologs we determined using phylogenetic tree analysis. We calculated a specificity score for each residue in a subfamily by calculating scores based on a modified version of the PHACT algorithm [ 19 ] scores which considers independent evolutionary events on the phylogenetic tree while scoring the acceptability of an amino acid substitution. We predicted the functional consequence of every potential substitution in CaSR by using the gradient boosting trees machine learning approach. Lastly, we investigated how our predictions relate to the activation mechanism of CaSR. While all subfamily receptors of class C GPCRs share common domains and structural features, the details of responding to different ligands and activating signaling pathways may differ even between closely related receptors [ 6 ]. Gene duplication is the main mechanism that generates new protein functions across GPCRs [ 15 ]. Protein families are evolved by speciation events following gene duplication [ 16 , 17 ]; thus, sequence comparisons of members within a subfamily and between subfamilies can show the evolutionarily conserved domains as well as diverged protein sites that distinguish one subfamily from others. One challenge with this analysis is that excessive gene duplication events complicate the identification of functionally identical orthologs in a subfamily. Moreover, the conservation patterns in paralogs (specifically, outparalogs that have diverged after a speciation event) and distant homologs may help infer the specific roles of a single residue in protein function. Because the evolutionary pressure on paralogs and close orthologs is not the same, substitutions allowed in paralogs may not be acceptable in close orthologs. While orthologs are derived from a single ancestral gene in the last common ancestor of the compared species, paralogs are arised from gene duplication event and generally, paralogs perform biologically distinct functions [ 18 ]. Thus, using functionally identical orthologs in sequence comparisons is crucial to inferring the role of each residue in a protein family. Variants in CaSR may cause malfunctions that result in Ca 2+ homeostasis diseases in humans. More than 400 germline loss/gain-of-function (i.e., LoF and GoF, respectively) mutations cause hypercalcaemic disorders, neonatal severe hyperparathyroidism (NSHPT), familial hypocalciuric hypercalcemia type-1 (FHH1), and autosomal dominant hypocalcemia type-1 (ADH1), respectively [ 2 ]. Many more CaSR variants are anticipated to be identified as more population-level genetic data become available [ 2 ]. Gaining insight into the function of individual residues within the receptor structure and their involvement in activation mechanisms has the potential to enhance our understanding of the probability of variant pathogenicity and the signaling processes of the CaSR. The examination of receptors within a family and across different families allows for the identification of the specific function of each residue in a receptor. However, the comprehensive understanding of the structure and activation mechanisms of several families within the class C GPCRs remains elusive. This is particularly true for the G-protein coupled receptor family C group 6 member A (GPRC6A) and the type 1 taste receptors (TAS1Rs; specifically members 1, 2, and 3), which are the most closely related subfamilies to the CaSR. Crystal structures of the ECD [ 4 , 7 ] and cryo-electron microscopy structures of the full-length CaSR [ 3 , 5 , 8 – 10 ] reveal the structural basis for activation mechanisms and ligand binding sites. L-amino acid binding sites at the interdomain cleft of LB1-LB2 [ 3 – 5 , 11 – 13 ] and multiple Ca2+ binding sites on the VFT domain are shown in the literature. [ 3 – 5 ]. While Ca 2+ serves as the primary agonist for the CaSR, L-amino acids enhance receptor activation in conjunction with Ca 2+ . However, it’s important to note that L-amino acids alone are insufficient to activate the receptor independently [ 14 ]. Even though Ca 2+ alone activates the receptor in functional assays [ 14 ], whether it activates the CaSR in the absence of L-amino acid is still controversial [ 3 , 5 ]. Calcium sensing receptor (CaSR) is a class C G-protein-coupled receptor (GPCR) that maintains extracellular Ca 2+ homeostasis by sensing calcium ions in the blood and regulating parathyroid hormone release and urinary calcium [ 1 , 2 ]. The CaSR is activated by Ca 2+ and L-amino acids such as L-Phe and L-Trp as well as polyamines and polypeptides [ 3 – 5 ]. Like the other class C GPCRs such as metabotropic glutamate receptors, ligands bind to the extracellular Venus flytrap (VFT) domain of the receptor [ 6 ]. Results Functionally equivalent orthologs and evolutionary history of CaSR in Class C To reveal the evolutionary constraints on protein families, we developed a strategy to precisely define a protein subfamily. A precise subfamily definition can be achieved by revealing the evolutionary history of the superfamily. The evolutionary history of gene families can only be established by reconstructing high-quality phylogenetic trees, which can be used to pinpoint gene duplication events. Discrimination between gene duplication and speciation nodes enabled us to define the paralogous and orthologous protein sequences. We further analyzed the phylogenetic trees to classify the orthologous sequences that are likely equivalent in function. A subfamily is defined by the human receptor and its orthologs. Within class C GPCRs, there are twenty-two distinct subfamilies: CaSR, GPRC6A, three taste receptors (TAS1R1-3), eight metabotropic glutamate receptors (mGluR1-8), GPRC5A-D, GABBR1, GABBR2, GPR156, GPR158, and GPR179. We used functionally-equivalent orthologs in comparative analyses between subfamilies, which eventually yielded subfamily-specific signatures that can be used to define that particular subfamily and its function. Finally, the association between the signature and function would enable a better understanding of specific molecular mechanisms and the effects of variants, particularly for the protein subfamily of interest. Here, we aim to reveal the signatures of the CaSR subfamily that is implied in the specific function of calcium-sensing and downstream signaling. We have retrieved the complete proteomes of 478 species from the NCBI database. To identify proteins that belong to the class C GPCR family, we performed a hmmsearch using the seven transmembrane domain profile (Pfam: 7tm_3) (Fig 1) against the proteomes. While this search allowed us to retrieve the entire class C GPCRs hitting the hmm profile, it did not provide subfamily annotations for the 22 subfamilies in class C GPCRs. To select canonical isoforms, we performed a profile hmmscan of the PfamA profile against Class C GPCR. To generate a general HMM profile for each subfamily, we first applied a BLAST search using each human class C GPCR as a query [20]. For each subject, we blasted them against the human proteome and retrieved the bidirectional best hits (Core subfamily assignment, Fig 1). PPT PowerPoint slide PNG larger image TIFF original image Download: Fig 1. Summary of the Methodological Framework. 478 complete proteomes were retrieved from the NCBI database. Each sequence was searched by hmmsearch against the Pfam 7tm_3 domain profile to retrieve all class C GPCRs. Domain architectures of class C GPCRs were determined by hmmscan against Pfam. A profile was created to identify canonical isoforms. Species-specific BLAST databases of the canonical isoforms were built. Bi-directional mutual best hits were detected by blasting each canonical sequence against the species databases (core subfamily assignment). TM domains of core subfamily sequences were aligned, and ML trees were built to make subfamily profile HMMs. By hmmsearch against subfamily profile HMMs, other sequences in the subfamilies were found (subfamily extension). Sequences in each subfamily were aligned, and ML trees were built. Based on the ML trees, paralogs were filtered, and functionally identical groups were identified. https://doi.org/10.1371/journal.pcbi.1012591.g001 For proteins that did not have bidirectional mutual best hits, we assigned them to a subfamily based on their homology search against the HMM profiles generated in the previous step (Subfamily Extension, Fig 1). We produced maximum likelihood (ML) trees of extended subfamilies and filtered inparalogous sequences to obtain functionally identical groups. The CaSR subfamily produced over five thousand hits, which included vomeronasal and olfactory receptors that have never been shown to sense calcium. Previous research has shown that CaSR is classified in the pheromone/olfactory cluster of class C GPCRs [21]. In species that had multiple proteins assigned to the CaSR subfamily, we constructed an ML tree using these hits and other human class C GPCR protein sequences. These trees revealed that a significant number of duplication events occurred in the species after the clade (a group of organisms that includes a common ancestor and all of its descendants) diverged from CaSR. As a result, we defined this diverged clade as a new subfamily named CaSR-likes. In this subfamily, there are diverse sequences including vomeronasal, olfactory receptors that are unlikely to maintain calcium homeostasis, and therefore should not be annotated as calcium-sensing receptors. We selected representative sequences from different species for each subfamily of 22 different receptor subfamilies and 264 CaSR-like sequences and built an ML tree (Fig 2A). Also, we built the ML trees of all proteins from CaSR, GPRC6A, taste receptors and merged these trees to the representative tree of class C GPCRs (Fig 2A). We used 1000 transfer bootstrap to ensure statistical reliability of the phylogenetic tree. We rooted the ML tree using the GPR158 and GPR179 as outgroups. There are 22 different class C GPCRs in humans. The resulting phylogeny shows that are five major clades: CaSR-related, GABA, mGluR, orphans, and retinoic acid-induced (RAIG). Orphan receptors, GPR158 and GPR179, formed a clade that was diverged from other receptors consistent with previous trees [22] and had a 0.95 transfer bootstrap expectation (TBE) value. γ-aminobutyric acid-B receptors (GABBR1 and GABBR2) formed another clade diverging from GPR156 with 0.97 TBE. γ-aminobutyric acid-B receptors evolved earliest and have a common ancestor with the highest taxonomic rank (33213−Bilateria) compared to other subfamilies. The CaSR group (CaSR, CaSR-likes, GPRC6A and taste receptors) was diverged from metabotropic glutamate receptors (mGluR1-8) and RAIG receptors (GPRC5A, GPRC5B, GPRC5C, GPRC5D) with 1 and 0.98 TBE values, respectively. Within the CaSR group, clade CaSRs and CaSR-likes were diverged from GPRC6A and taste receptors with 1 TBE. Except for TAS1R1 and TAS1R2, all CaSR group subfamilies have a common ancestor from taxonomy clade 7776-Gnathostomata. TAS1R1 and TAS1R2 were more specific than other CaSR group subfamilies that evolved from 117571-Euteleostomi. In a phylogenetic tree, the length of branches signifies genetic changes over time, where longer branches indicate greater divergence. Employing Welch’s t-Test in our statistical analysis, we measured these branch lengths to gain insights into the evolutionary conservation of receptor subfamilies, revealing genetic similarities and differences. Our comparison analysis of branch lengths [22] among common species in CaSR, GPRC6A, and taste receptors highlights a significant conservation trend within the CaSR subfamily compared to its closest counterparts (Fig 2B). PPT PowerPoint slide PNG larger image TIFF original image Download: Fig 2. Evolution of Class C GPCRs. (A) The maximum likelihood phylogenetic tree of Class C GPCRs, spanning representative species from each subfamily, is shown. Subfamilies are represented as circular layers around the ML tree. All twenty-two Class C GPCR subfamilies are shown in the inner circle. In addition to these subfamilies, vomeronasal and other orphan receptors are represented as CaSR-like receptors. All proteins in CaSR, GPRC6A and TAS1Rs are merged into this representative species tree. (B) Branch lengths from leaf to root of the common species that exist in all CaSR, GPRC6A and TAS1Rs are taken from the subfamily trees. Welch’s t-Test using the ggstatsplot package results are shown on the graph. https://doi.org/10.1371/journal.pcbi.1012591.g002 The higher diversity of CaSR-likes relative to CaSRs is reflected in the ML tree (Fig 2A). Branch lengths of CaSR-likes are longer in contrast to shorter branch lengths in CaSR. Longer branch lengths show that more variation, and thus divergence, occurred in the CaSR-like clade. Moreover, extensive gene duplication events occurred in this clade. For instance, rodents such as Dipodomys ordii (taxid: 10020), Octodon degus (taxid:10160) and snakes such as Notechis scutatus (taxid: 8663) have more than a hundred receptors matching the CaSR profile. However, these matches include type 2 vomeronasal receptors (V2R) and V2R-likes. Among mammals, V2R genes exhibit significant variation. While dogs, cows, and primates except prosimians do not have functional V2Rs, rodents, reptiles, and fish have multiple intact V2Rs [23]. Since these receptors do not have functional orthologs in mammals, they are likely to be functionally diverged, and it is crucial to separate them from functionally-equivalent CaSRs. Precision in subfamily identification: Constructing subfamily-specific profile hmms for class c gpcrs based on phylogenetic analysis In the class C GPCR family, gene duplication events give rise to new specificity, and each duplicated gene with a new function evolves through further speciation events, producing a set of orthologous sequences [16,17]. Each subfamily of class C GPCRs shares a relatively conserved membrane-spanning region but also exhibits a degree of variability that underlies functional differences. At the molecular level, residues that are responsible for certain functional characteristics such as ligand and coupling selectivity are called specificity-determining residues [16]. Conservation analysis from multiple sequence alignments (MSAs) can be used to find residues that are conserved in all subfamilies through evolution as well as specificity-determining residues that are only conserved in a subfamily and differ in other subfamilies. However, the success of this method depends on the sequences that are used to build alignments. Therefore, it is vital to use functionally identical orthologs in the analysis. The seven-transmembrane domains of class C GPCRs are used to build a class-specific general profile for this family (Pfam:7tm_3). However, this domain does not contain enough information to differentiate subfamilies further. Moreover, excessive gene duplication events, as seen in the CaSR-like clade, require precise phylogenetic analysis to differentiate between CaSR and CaSR-like sequences. Also, subfamily specific profile HMMs are shown to be promising methods to detect protein sequences belonging to a protein subfamily, as well as separation of homologs and non-homologs [24,25]. In this paper, we present a novel approach to constructing subfamily-specific profile HMMs based on the precisely produced subfamily alignments and trees. The general idea of the approach is given in Fig 3. PPT PowerPoint slide PNG larger image TIFF original image Download: Fig 3. Subfamily Specific HMM Models. Based on the phylogenetic tree, the target, the closest, and the rest groups were determined. Initial HMMs were built without using priors. Representative amino acids in each group are selected, and their scores are calculated. According to groups, representative amino acids, and conservation scores, we calculated weights to change the emission probabilities of initial HMMs. https://doi.org/10.1371/journal.pcbi.1012591.g003 To construct subfamily specific profile HMMs, we first define the target family, its closest family (phylogenetic neighboring clade), and the rest based on the phylogenetic tree. We weight the identity score of each amino acid to calculate the emission probabilities. The highest weight is given to the residues which are only conserved in the target subfamily; hence, they differentiate one subfamily from the others. A minimum weight is given to the residues that are conserved both in the target subfamily and its closest clade. We downloaded complete proteomes from UNIPROT [26] and we tested our subfamily-specific profile HMMs’ performance on independent sequences retrieved from UNIPROT [26]. These sequences were from different species and they were not used in calculating the position weights. We assigned sequences to their corresponding subfamilies by following the same steps as the NCBI dataset [27] used to build these models. We selected new taxa that were not in the NBCI dataset to test the performance of our profiles, and our subfamily specific profile HMMs correctly identified all members of a subfamily while avoiding hits to proteins from other subfamilies (Table 1). PPT PowerPoint slide PNG larger image TIFF original image Download: Table 1. Subfamily Specific Profile HMM’s Performance. https://doi.org/10.1371/journal.pcbi.1012591.t001 [END] --- [1] Url: https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1012591 Published and (C) by PLOS One Content appears here under this condition or license: Creative Commons - Attribution BY 4.0. via Magical.Fish Gopher News Feeds: gopher://magical.fish/1/feeds/news/plosone/