Spotlight on ECI researchers

Xiaojing Cong,

Institut de Génomique Fonctionnelle, Montpellier, France

Machine learning of sequence-function relationships of olfactory receptors

Xiaojing Cong,1,* Wenwen Ren,2 Rui Xu,3 Minghong Ma,4 Hiroaki Matsunami,5 Yiqun Yu,2,3* Jérôme Golebiowski6,7*

1 Institut de Génomique Fonctionnelle, University of Montpellier, CNRS, INSERM, 34094 Montpellier, France

2 Fudan University, Shanghai 200031, People's Republic of China

3 School of Life Sciences, Shanghai University, Shanghai 200444, People's Republic of China

4 University of Pennsylvania Perelman School of Medicine, Philadelphia, PA 19104, USA.

5 Duke University Medical Center, Research Drive, Durham, NC 27710, USA

6 Université Côte d’Azur, CNRS, Institut de Chimie de Nice UMR7272, Nice 06108, France

7 Daegu Gyeongbuk Institute of Science and Technology, Daegu 711-873, South Korea

# These authors contributed equally.

Machine learning can predict protein structures and functions from primary sequences, of which AlphaFold is a striking example. However, data scarcity is often the bottleneck. In recent studies, we circumvented this problem by extracting the evolutionary “hotspots” of the protein sequences, using site-directed mutagenesis and functional assays. This reduced the dimensionality for machine learning and obtained good predictions with limited data. We studied olfactory receptors (ORs), which make up half of the human GPCRs and detect innumerous odorants. There are two major obstacles in OR research: heterologous expression of ORs and thus, difficult functional assays to map ORs with odorants.

In one study [1], we trained a machine learning proteochemometric model to map ORs with odorants, which gave the prospective hit rate of up to 50%. The model used residues around the orthosteric pocket (the hotspots) of the ORs and chemical structures of the odorants. In another study [2], we found that residues in the transmembrane helix (TM) 3 and TM4 play key roles in OR heterologous expression. Using the sequences of OR TM3-4, we could predict the expression probability and design mutants to rescue ORs that otherwise fail in heterologous expression.


[1] Cong et al. Large-scale G protein-coupled olfactory receptor-liand pairing. ACS Cent Sci 2022

[2] Xu et al. Interactions among key residues regulate mammalian odorant receptor trafficking. FASEB J. (final revision before acceptance).

Tianyi Ding,

Queen’s University Belfast, Northern Ireland

Ligand binding site mapping at GPCR-Lipid Interface

Targeting G protein-coupled receptors (GPCRs) through allosteric sites offers advantages over orthosteric sites in identifying drugs with increased selectivity and potentially reduced side effects (1-4). We have recently developed a probe confined dynamic mapping protocol to map GPCRs allosteric sites at various locations (5). Here, we propose an improved protocol that allows efficient prediction of allosteric sites at the receptor-lipid interface. The P2Y1 purinergic and CB1 cannabinoid receptors were selected for protocol validation. We show that our protocol together with the sequence analysis is specific in identifying an allosteric site of a compound and works in various receptor conformations. The protocol provides a fast and efficient prediction of key polar interactions in an allosteric cavity at the lipid interface.


1. Hauser AS, Attwood MM, Rask-Andersen M, Schiöth HB, Gloriam DE. Trends in GPCR drug discovery: new agents, targets and indications. Nature reviews Drug discovery. 2017;16(12):829–42.

2. Renault P, Giraldo J. Dynamical Correlations Reveal Allosteric Sites in G Protein- Coupled Receptors. Int J Mol Sci. 2020 Dec 27;22(1):187.

3. Kenakin T. G protein coupled receptors as allosteric proteins and the role of allosteric modulators. J Recept Signal Transduct Res. 2010 Oct;30(5):313–21.

4. Gentry PR, Sexton PM, Christopoulos A. Novel Allosteric Modulators of G Protein-coupled Receptors. J Biol Chem. 2015 Aug 7;290(32):19478–88.

5. Ciancetta, A, Gill, A.K.; Ding, T.; Karlov, D.S.; Chalhoub, G.; McCormick P.J.; Tikhonova I.G.; Probe Confined Dynamic Mapping for G Protein-Coupled Receptor Allosteric Site Prediction. ACS Cent. Sci. 2021

Khaled Abd-Elrahman,

University of Ottawa, Canada

A positive allosteric modulator of M1 Acetylcholine receptors improves pathology and cognitive deficits in female APPswe/PSEN1ΔE9 mice

Khaled S. Abd-Elrahman1, Shaarika Sarasija1, Tash-Lynn L. Colson1 and Stephen S. G. Ferguson1

1 University of Ottawa Brain and Mind Research Institute, Department of Cellular and Molecular Medicine, University of Ottawa, Ottawa, ON, Canada

Alzheimer’s disease (AD) is an age-related neurodegenerative disorder characterized by progressive cognitive decline with no effective treatments to slow progression. Beta-amyloid (Aβ) protein is considered the principal neurotoxic species in AD brains. The m1 Acetylcholine receptor (m1 mAChR) plays a key role in memory and learning. m1 mAChR agonists shows pro-cognitive activity but cause many off target adverse effects including seizures. A new m1 mAChR positive allosteric modulator (PAM), VU0486846, is devoid of direct agonist activity and adverse effects but was not tested for efficacy in AD mice. Since women account for more than 60% of cases and most AD research is conducted in male models, we tested the efficacy of VU0486846 in female AD mice first. Here, we treated 9-month-old female APPswe/PSEN1ΔE9 (APPswe) and wild-types with VU0486846 in drinking water (10mg/kg/day) for 4 or 8 weeks. Cognitive function of all mice was assessed after treatment and brains were harvested for biochemical and immunohistochemical assessment. Both 4 and 8 weeks of treatment with VU0486846 improved cognitive function of APPswe mice when tested in novel object recognition and Morris water maze. This was paralleled by a significant reduction in hippocampal Aβ oligomers and plaques. VU0486846 did not change the expression of amyloid precursor protein in APPswe but the reduction in Aβ load in VU0486846-treated mice was due to a shift in the processing of amyloid precursor protein from β-cleavage to non-amyloidogenic cleavage. Specifically, VU0486846 reduced expression of β-secretase 1 (BACE1) whereas enhanced expression of the α-secretase ADAM10 in APPswe hippocampus. Thus, using m1 AChR PAMs can be a viable disease-modifying approach that should be exploited clinically to slow AD.

Susanne Prokop

Institute of Experimental Medicine, Budapest, Hungary

PharmacoSTORM: a novel approach for the nanoscale imaging of drug binding sites

Susanne Prokop 1,2, Péter Ábrányi-Balogh3, Benjámin Barti1,2,9, Márton Vámosi1, Miklós Zöldi 1,2,9, László Barna4, Gabriella M. Urbán1, András Dávid Tóth5, Barna Dudok1,6, Attila Egyed3, Hui Deng7, Gian Marco Leggio8, László Hunyady5, Mario van der Stelt7, György M. Keserű3 and István Katona1,9

1) Momentum Laboratory of Molecular Neurobiology, Institute of Experimental Medicine, Budapest, Hungary; 2) School of Ph.D. Studies, Semmelweis University, Budapest, Hungary; 3) Medicinal Chemistry Research Group, Research Centre for Natural Sciences, Budapest, Hungary; 4) Nikon Center of Excellence for Neuronal Imaging, Institute of Experimental Medicine, Budapest, Hungary; 5) MTA-SE Molecular Physiology Laboratory, Department of Physiology, Faculty of Medicine, Semmelweis University, Budapest, Hungary; 6) Department of Neurosurgery, Stanford University, Stanford, CA, USA; 7) Department of Molecular Physiology, Leiden Institute of Chemistry, Leiden University & Oncode Institute, Leiden, the Netherlands; 8) Department of Biomedical and Biotechnological Sciences, Section of Pharmacology, University of Catania, Catania, Italy; 9) Department of Psychological and Brain Sciences, Indiana University, Bloomington, Indiana, USA

Autoradiography is the traditional method-of-choice to put pharmacological interactions into an anatomical context. However, the generally low-resolution of autoradiographs only allows the visualization of drugs at the regional level, and the cell-type specific target engagement of clinical drugs remains unrevealed. On the other hand, the advent of super-resolution microscopy allows the localization of fluorophores with nanoscale precision. Here, we introduce PharmacoSTORM super-resolution imaging that combines the complementary advantages of these approaches and applies fluorescently labeled drugs for the visualization of their target engagement in a cell-type- and subcellular context-dependent manner within complex tissue preparations. We demonstrated the broad applicability of our approach by the rational design of fluorophore-tagged high-affinity receptor ligands and an enzyme inhibitor. In addition, we took advantage of PharmacoSTORM to investigate cariprazine, a clinically approved antipsychotic and antidepressant drug, with an elusive mechanism of action. We provide in vivo evidence that cariprazine predominantly binds to D3 dopamine receptors on Islands of Calleja granule cell axons in the mouse basal forebrain. Our findings show that PharmacoSTORM can be as useful for the visualization of the regional distribution of drug binding as autoradiography but has the unique capacity to zoom beyond the diffraction limit and to detect single drug molecules at the nanoscale level at their native tissue localization.

Nipuna Weerasinghe,

University of Arizona, USA

The Modulation of Conformational Energetics of GPCR Activation by Water

Nipuna Weerasinghe1 (Presentor), Steven D. E. Fried1, Andrey V. Struts1,2, Suchithranga M. D. C. Perera1, Michael F. Brown1,3

1 Department of Chemistry and Biochemistry, University of Arizona, Tucson, AZ 85721, USA. 2 Laboratory of Biomolecular NMR, St. Petersburg State University, St. Petersburg 199034, Russia. 3 Department of Physics, University of Arizona, Tucson, AZ 85721, USA.

G-protein-coupled receptors (GPCRs) are signal transducers that enable transmembrane communication in cells and are the largest family of targets (~40%) for currently approved drugs. They exist as dynamic conformational ensembles with multiple inactive and active conformational substates described by an energy landscape model. Although three-dimensional structures show the presence of structural water molecules within GPCRs, the role that solvent molecules play in the receptor dynamics and signaling behavior remains unexplored. Our previous work has further shown a bulk influx of water stabilizing the active rhodopsin conformation [1]. Here we broadened the experimental approach to investigate how the conformational energetics of GPCR activation are modulated by water using a series of hydrophilic polymers including polyethylene glycols (PEGs), dextrans, polyvinylpyrrolidones, and their monomers. Our hypothesis was that the thermodynamic stability of conformational substates of photoactivated rhodopsin is hydration mediated. Accordingly, we changed the osmotic stress on rhodopsin with different concentrations of various polymer solutes and used UV-visible spectroscopy to monitor the shifting of the metarhodopsin equilibrium under controlled hydration. An opposite effect of large and small osmolytes was observed on the metarhodopsin equilibrium, in which larger osmolytes (e.g., PEG 1500-6000) promoted the inactive metarhodopsin-I (MI) state, while small osmolytes increased the active metarhodopsin-II (MII) fraction. We propose that these size effects are a result of the degree of penetration of osmolytes into the protein core of rhodopsin. Small polymers penetrate the protein core and favor the open MII state of rhodopsin, while the large polymers behave like ideal osmolytes and dehydrate rhodopsin. Interestingly PEGs and dextrans with extremely large molar masses deviated in their trends from other large polymers, which we attribute to a combined result of crowding and penetration effects. Our results provide us with new insights into the interplay between osmotic stress, penetration, and crowding effects and further our understanding of the vital role of water in the GPCR activation mechanisms.


Chawla U, Perera SMDC, Fried SDE, Eitel AR, Mertz B, Weerasinghe N, Pitman MC, Struts AV & Brown MF (2021) Activation of the G-Protein-Coupled Receptor Rhodopsin by Water. Angewandte Chemie International Edition 60: 2288–2295

Lina Verbakel,

Universität Freiburg, Germany

Prothoracicostatic Activity of the Ecdysis-Regulating Neuropeptide Crustacean Cardioactive Peptide (CCAP) in the Desert Locust

Lina Verbakel1,2, Cynthia Lenaerts1, Rania Abou El Asrar1,3, Caroline Zandecki1,4, Evert Bruyninckx1, Emilie Monjon1, Elisabeth Marchal1,5 and Jozef Vanden Broeck1

1 Molecular Developmental Physiology and Signal Transduction, KU Leuven, Naamsestraat 59, P.O. Box 2465, B-3000 Leuven, Belgium.

2 Neuronal circuits and behavior, Université de Fribourg, Ch. Du Musée 14, PER 02 bu. 1.405, 1700 Fribourg, Switzerland

3 Laboratory of Cellular Transport Systems, ON I, Herestraat 49, P.O. Box 802, B-3000 Leuven, Belgium.

4 Developmental Neurobiology, KU Leuven, Naamsestraat 61, P.O. Box 2464, B-3000 Leuven, Belgium.

5 Imec, Kapeldreef 75, B-3001 Leuven, Belgium.

Accurate control of innate behaviors associated with developmental transitions requires functional integration of hormonal and neural signals. Insect molting is regulated by a set of neuropeptides, which trigger periodic pulses in ecdysteroid hormone titers and coordinate shedding of the old cuticle during ecdysis. In the current study, we demonstrate that crustacean cardioactive peptide (CCAP), a structurally conserved neuropeptide described to induce the ecdysis motor program, also exhibits a previously unknown prothoracicostatic activity to regulate ecdysteroid production in the desert locust, Schistocerca gregaria. We identified the locust genes encoding the CCAP precursor and three G protein-coupled receptors that are activated by CCAP with EC50 values in the (sub)nanomolar range. Spatiotemporal expression profiles of the receptors revealed expression in the prothoracic glands, the endocrine organs where ecdysteroidogenesis occurs. RNAi-mediated knockdown of CCAP precursor or receptors resulted in significantly elevated transcript levels of several Halloween genes, which encode ecdysteroid biosynthesis enzymes, and in elevated ecdysteroid levels one day prior to ecdysis. Moreover, prothoracic gland explants exhibited decreased secretion of ecdysteroids in the presence of CCAP. Our results unequivocally identify CCAP as the first prothoracicostatic peptide discovered in a hemimetabolan species and reveal the existence of an intricate interplay between CCAP signaling and ecdysteroidogenesis.

Infant Sagayaraj Ravhe,

Indian Institute of Technology Madras, India

Evolutionary history of histamine receptors: Early vertebrate origin and expansion of the H3 - H4 subtypes

Infant Sagayaraj Ravhe, Arunkumar Krishnan, and Manoj Narayanan, Indian Institute of Technology Madras, India

Histamine receptors belonging to the superfamily of G protein-coupled receptors (GPCRs) mediate the diverse biological effects of biogenic histamine. They are classified into four phylogenetically distinct subtypes H1-H4, each with a different binding affinity for histamine and divergent downstream signaling pathways. Here we present the evolutionary history of the histamine receptors using a phylogenetic approach complemented with comparative genomics analyses of the sequences, gene structures, and synteny of gene neighborhoods. The data indicate the earliest emergence of histamine-mediated GPCR signaling by a H2 in a prebilaterian ancestor. The analyses support a revised classification of the vertebrate H3-H4 receptor subtypes. We demonstrate the presence of the H4 across vertebrates, contradicting the currently held notion that H4 is restricted to mammals. These non- mammalian vertebrate H4 orthologs have been mistaken for H3. We also identify the presence of a new H3 subtype (H3B), distinct from the canonical H3 (H3A), and propose that the H3A, H3B, and H4 likely emerged from a H3 progenitor through the 1R/2R whole genome duplications in an ancestor of the vertebrates. It is apparent that the ability of the H1, H2, and H3-4 to bind histamine was acquired convergently. We identified genomic signatures suggesting that the H1 and H3-H4 shared a last common ancestor with the muscarinic receptor in a bilaterian predecessor whereas, the H2 and the α-adrenoreceptor shared a progenitor in a prebilaterian ancestor. Furthermore, site-specific analysis of the vertebrate subtypes revealed potential residues that may account for the functional divergence between them.

References: Ravhe IS, Krishnan A, Manoj N. Evolutionary history of histamine receptors: Early vertebrate origin and expansion of the H3-H4 subtypes. Mol Phylogenet Evol. 2021 Jan;154:106989.