G protein-coupled receptors (GPCRs) are one of the largest groups of membrane proteins encoded by the human genome. These essential membrane proteins are involved in a myriad of different cellular and (patho-) physiological processes (Ali et al., 2020, Pierce et al., 2002, Venkatakrishnan et al., 2019). Signal transduction by GPCRs is facilitated by binding of various ligands to the extracellular orthosteric site (Rosenbaum et al., 2009) of the receptor followed by the interaction with their intracellular partner proteins, such as G proteins (Mahoney and Sunahara, 2016) or ß-arrestins (Luttrell and Lefkowitz, 2002). The involvement of GPCRs in diverse signaling processes makes them attractive and promising targets for pharmaceutical drug development (Shimada et al., 2019). In the last two decades, X-ray crystallography (Rasmussen et al., 2011) and cryo-EM (Draper-Joyce et al., 2018) provided atomic-resolution structures of GPCRs in complex with G proteins and ß-arrestins (Huang et al., 2020). Despite these insights, it has been noted that static structures cannot fully explain the activation pathway of these highly dynamic proteins (Hilger, 2021, Weis and Kobilka, 2018).

GPCRs are seven transmembrane helical proteins with a high intrinsic conformational flexibility, enabling the population of inactive and active states (Draper-Joyce and Furness, 2019, Weis and Kobilka, 2018). Nuclear magnetic resonance (NMR) spectroscopy proved to be a versatile tool to decipher GPCR conformational states and dynamics in solution (Ueda et al., 2019) using amino-acid selective (Eddy et al., 2018, Isogai et al., 2016) and fluorine labeling (Liu et al., 2012), or the introduction of methyl probes by chemical modification of lysine (Sounier et al., 2015) or cysteine residues (Goba et al., 2021). Nevertheless, a high-resolution NMR study with a uniformly isotope labeled GPCR has not been reported so far. Production of isotope labeled receptors in different expression hosts (Abiko et al., 2021, Berger et al., 2011, Clark et al., 2017, Werner et al., 2008, Xu et al., 2019) and in a cell-free setup (Shilling et al., 2017) has been achieved and optimized over the past years. In addition, directed evolution strategies made it possible to enhance the expression levels of GPCRs in E. coli (Dodevski and Pluckthun, 2011; Sarkar et al., 2008, Scott et al., 2014) and enabled the production of these challenging membrane proteins for NMR studies (Schuster et al., 2020). Compared to eukaryotic expression hosts, such as insect cells or mammalian cells, E. coli has the major advantage of tolerating high levels of deuterium in the growth medium. The stabilized neurotensin receptor subtype 1 (NTR1) variant HTGH4 can be produced in E. coli at a yield of 2-4 mg/l cell culture and exhibits high thermal stability even in detergent solution (Scott et al., 2014). Crystal structures of HTGH4 and other stabilized NTR1 variants obtained by directed evolution in E. coli (Egloff et al., 2014) bound to the agonist peptide neurotensin-1 (NTS-1) are almost identical to the structure of another stabilized neurotensin receptor obtained from insect cells (White et al., 2012). Structures are known for NTR1 in the active and inactive states, as well as in the G protein bound form (Deluigi et al., 2021, Kato et al., 2019, Krumm et al., 2016, Zhang et al., 2021). Even with the availability of abundant structural information of the inactive and active states, the mechanistic details of the activation pathway of such a peptide receptor by high-resolution NMR methods remain only poorly investigated.

Here, we optimized the sample conditions for the stabilized NTR1 variant HTGH4 and obtained high-resolution NMR spectra of sufficient quality for the collection of multidimensional NMR experiments for sequence-specific NMR backbone resonance assignments. Furthermore, we used NMR chemical shift perturbation and hydrogen deuterium (HDX) mass spectrometry experiments to probe structural changes in the orthosteric binding site of HTGH4 bound to the agonist neurotensin or the antagonist SR142948 (Gully et al., 1997). In addition, we could show that large parts of the transmembrane region of HTGH4 are strongly protected from the solvent, prohibiting the observation of NMR backbone resonances in this region if fully deuterated protein is used. To address this issue, we use partial HTGH4 unfolding by chemical denaturants to enhance amide proton exchange in transmembrane helices and obtain a larger number of amide resonances for further NMR experiments. However, this procedure also led to enhanced sample heterogeneity, suggesting that isotope labeling strategies that omit backbone amide deuteration will be required for further NMR studies of the entire protein. Taken together, these data show that high-resolution NMR studies of GPCRs are facilitated by directed evolution, enabling the investigation of GPCR allostery at atomic resolution.