Structural basis for activation of somatostatin receptor 5 by cyclic neuropeptide agonists

Significance

Somatostatin receptor 5 (SSTR5) is a promising therapeutic target for endocrine, metabolic, and neurological disorders. This study unveils the molecular basis for agonist recognition and activation of SSTR5 by solving cryogenicelectron microscope (cryo-EM) structures of SSTR5 bound to cyclic peptide agonists cortistatin-17 and octreotide. The structures reveal distinct binding modes and conformational changes induced by the two ligands, elucidating principles of agonist selectivity. Key insights include the roles of extracellular loops and a “hydrophobic lock” in mediating ligand-specific interactions and receptor activation. These insights will facilitate developing improved SSTR5 agonists to treat neuroendocrine tumors and pituitary disorders.

Abstract

Somatostatin receptor 5 (SSTR5) is an important G protein–coupled receptor and drug target for neuroendocrine tumors and pituitary disorders. This study presents two high-resolution cryogenicelectron microscope structures of the SSTR5-G_i complexes bound to the cyclic neuropeptide agonists, cortistatin-17 (CST17) and octreotide, with resolutions of 2.7 Å and 2.9 Å, respectively. The structures reveal that binding of these peptides causes rearrangement of a “hydrophobic lock”, consisting of residues from transmembrane helices TM3 and TM6. This rearrangement triggers outward movement of TM6, enabling Gα_i protein engagement and receptor activation. In addition to hydrophobic interactions, CST17 forms conserved polar contacts similar to somatostatin-14 binding to SSTR2, while further structural and functional analysis shows that extracellular loops differently recognize CST17 and octreotide. These insights elucidate agonist selectivity and activation mechanisms of SSTR5, providing valuable guidance for structure-based drug development targeting this therapeutically relevant receptor.

Somatostatin is an important inhibitory hormone that regulates a range of physiological processes including growth hormone release, pancreatic secretion, cognition, and vasoconstriction. Disruption of somatostatin signaling can lead to endocrine, metabolic, and neurological disorders. The actions of somatostatin are mediated through somatostatin receptors (SSTRs), which belong to the G protein–coupled receptor family (1). These receptors are categorized into two subfamilies based on pharmacology and sequence conservation: SSTR family 1 (SSTR2, SSTR3, and SSTR5), and SSTR family 2 (SSTR1 and SSTR4) (2). Despite a sequence variation of 43 to 61% among these subtypes (3, 4) (SI Appendix, Fig. S1), they share common functionalities, primarily inhibiting growth hormone secretion and curtailing neuroendocrine tumor growth (5).

SSTRs are expressed across various tissues, including the central nervous system, anterior pituitary gland, gastrointestinal tract, immune cells, and lymphoid tissue (6, 7). All five SSTRs primarily facilitate the inhibition of hormone release via the heterotrimeric G protein pathway, specifically the G_i/o pathway (Fig. 1A), which is predominantly modulated by somatostatin analogs and related neuropeptides (8, 9).

Fig. 1.

Overall maps and models of the SSTR5-G_i complexes bound to CST17 and octreotide. (A) Schematic overview of G protein–mediated SSTRs signaling. (B and C) Cryo-EM map (B) and structural model (C) of CST17-SSTR5 in complex with G_i. (D and E) Cryo-EM map (D) and structural model (E) of octreotide-SSTR5 in complex with G_i. (F and G) Three-dimensional models of CST17 (F) and octreotide (G), with surrounding density maps shown in mesh.

SSTR5 is highly expressed in the pituitary and plays a crucial role in regulating hormones such as adrenocorticotropic hormone, prolactin, and growth hormone (10). Its expression remains relatively stable irrespective of cortisol fluctuations (11–13). Moreover, SSTR5 uniquely modulates tumor growth through signaling pathways distinct from other members of SSTRs (14–16), making it a prime therapeutic target for conditions like Cushing’s syndrome, acromegaly, and pituitary adenomas.

The endogenous peptide cortistatin-17 (CST17) and the synthetic cyclic analogue octreotide are both known to exhibit strong affinity for SSTR5. While CST17 modulates neuroendocrine and cardiovascular functions, octreotide has proven efficacious in treating neuroendocrine tumors and pituitary disorders via SSTR5 activation.

CST17 is predominantly expressed in the brain cortex and shares a significant sequence overlap with somatostatin-14 (SST14), including the essential SSTR binding motif Phe-Trp-Lys-Thr (FWKT) (17, 18). This overlap elucidates their analogous pharmacological profiles (19, 20), and both CST17 and SST14 have comparable affinity for SSTRs. Studies show protective cardiovascular effects of CST17 relevant to SSTR5 (21–25), yet the structural intricacies governing its interaction with SSTR5 remain elusive.

The clinical utility of SST14 is constrained by its short half-life (26, 27). To overcome this limitation, synthetic somatostatin analogs (SAs) were developed. Octreotide, a stable SST analog, retains the FWKT core and has been an important drug for hormone-producing tumors since clinical introduction in 1983 (6, 28–30). It can bind to SSTR2 and SSTR5, but with affinities that differ between the two receptors (9, 31). The molecular mechanism of how octreotide recognizes and activates SSTR2 has been outlined already (9, 32). However, the molecular details of how octreotide recognizes and interacts with SSTR5 are still unclear. There remains a pressing need for more selective SSTR5 agonists for better therapeutic targeting and efficacy (33). To facilitate structure-based design of SSTR5 agonists, we determine the cryogenic electron microscope (cryo-EM) structures of SSTR5 in complexes with CST17 and octreotide.

Results

Structure Determination of SSTR5 Complexes.

To obtain stable complexes of SSTR5 with the peptide agonists CST17 and octreotide, we introduced a V253^6.40L mutation in full-length SSTR5. Functional assays showed that this variant exhibited similar activity to wild-type SSTR5, maintaining similar potency and efficacy (SI Appendix, Fig. S2 A and B). We coexpressed the SSTR5 variant in insect cells along with heterotrimeric G_i protein and an antibody fragment (scFv16), which helped stabilize SSTR5 complexes with heterotrimeric G proteins (34). These strategies enabled us to obtain stable CST17-SSTR5-G_i and octreotide-SSTR5-G_i complexes, which were determined by cryo-EM single particle analysis at overall resolutions of 2.7 Å and 2.9 Å, respectively (Fig. 1 B–E and SI Appendix, Figs. S3 and S4 and Table S1). The resulting maps exhibited excellent density, permitting accurate modeling of most receptor regions (SI Appendix, Fig. S5 A–C) and clear visualization of The CST17 and octreotide ligands (Fig. 1 F and G). However, density quality limited modeling of the N and C termini of SSTR5 and parts of CST17 outside of the disulfide bond loop region (Fig. 1 B–G).

While the overall architecture of the two complexes was similar, detailed alignment revealed the RMSD values of 1.43 Å for the receptor and 1.50 Å across the entire complexes (SI Appendix, Fig. S2 C and D). This highlights subtle but important conformational differences between the CST17- and octreotide-bound states.

Molecular Basis for Recognition of CST17 by SSTR5.

In the activated SSTR5 structure, CST17 forms a circular β-hairpin structure stabilized by intracellular disulfide bond that locates in the orthosteric pocket in an upright orientation relative to the membrane plane (Fig. 2 A–C). This pocket is composed of the extracellular regions of the transmembrane helices (TM) and extracellular loops (ECLs), excluding TM1 (Fig. 2C). It resembles a pocket similar to that found in SSTR2 for SST14 (PDB ID: 7WIC) (SI Appendix, Fig. S2 E and F) (35). It is noteworthy that CST17, octreotide, and SST14 share a conserved FWKT motif considered to be a critical pharmacophore for biological activity (Fig. 2B) (30). This motif inserts into the central pocket, establishing crucial interactions with the receptor.

Fig. 2.

Molecular basis of SSTR5 recognition by CST17. (A) Conformational comparison of F^L/DWKT (Superscript L/D means L/D configuration) conserved motif in CST17 and octreotide. Disulfide bonds are shown as yellow sticks. (B) Sequence alignment of the endogenous peptides CST17, SST14, and octreotide. (C) Front view of CST17-bound SSTR5. SSTR5 is shown as cartoon and colored in pale blue; CST17 is shown as cartoon and colored in light coral, with a surrounding density map at a threshold of 0.0115. (D–G) Detailed interactions between CST17 and SSTR5, with interacting residues shown in sticks. Hydrogen bonds are shown with red dashed lines. Observed polar interaction is excluded by black dashed lines. (H) Inhibition of forskolin-stimulated cAMP accumulation of WT SSTR5 and its mutants, induced by CST17. ΔpEC₅₀ represents the difference between pEC₅₀ values of the mutant SSTR5 and the WT SSTR5. Data are shown as means ± SEM from at least three independent experiments performed in triplicate. The significance was determined with one-way ANOVAfollowed by Dunnett’s multiple comparisons test compared with WT. The P value was defined as *P < 0.05; **P < 0.01; and ***P < 0.001; ns, not significant. Detailed statistical evaluation is shown in SI Appendix, Table S2. WT: wild type.

Specifically, K11^C (superscript indicating residues from CST17) of CST17 forms hydrogen bonds with Y89^2.53 and Y294^7.43 of the receptor and also makes a salt bridge interaction with D119^3.32, which function as anchors at the bottom of the binding pocket (Fig. 2D). To validate our structural findings, we performed GloSensor cAMP assays to analyze the impact of these significant interactions on ligand-induced receptor activation. When we substituted Y89^2.53, D119^3.32 and Y294^7.43 with alanine, we observed a significant reduction in the potency of CST17-mediated G_i activation (Fig. 2H). Combined with sequence alignment of SSTRs, D119^3.32 and Y294^7.43 are highly conserved in SSTRs, so these polar interactions are important for receptor-binding peptide agonists (SI Appendix, Fig. S1) (36). Additionally, the side chain of K11^C engages in hydrophobic interactions with L96^2.60, V290^7.39, and F264^6.51 of the receptor (Fig. 2D). W10^C of CST17 is surrounded by a cluster of hydrophobic amino acids, including F124^3.37, F201^5.38, I202^5.39, T205^5.42, F264^6.51, and F265^6.52, resulting in extensive hydrophobic interactions. Furthermore, W10^C can form hydrophobic interactions with Q123^3.36 and N268^6.55 (Fig. 2E). Alanine mutations of Q123^3.36, F124^3.37, F201^5.38, I202^5.39, F264^6.51, and F265^6.52 remarkably reduce the potency of CST17 in activating the G_i pathway (Fig. 2H and SI Appendix, Fig. S6 and Table S2).

In addition, the flexible conformation of CST17 contributes to diverse orientations of the benzene rings of F8^C, F9^C, and F13^C. F8^C extends towards TM6 and TM7, positioning its large benzene ring parallel to TM7. This orientation allows F8C to form extensive hydrophobic interactions with TM6 and TM7, including N268^6.55, N271^6.58, Y286^7.35, F287^7.36, and V290^7.39. F8^C also forms π–π interactions with F13^C to stabilize the active conformation of CST17. F9^C is surrounded by the hydrophobic pocket formed by W190^ECL2, F201^5.38, and I202^5.39 (Fig. 2F), and alanine mutations of these residues markedly diminish the activity of SSTR5.

Apart from the hydrophobic pocket, ECL2 also plays a crucial role in CST17 recognition of the receptor. T12^C makes hydrogen bond interaction with N187^45.51 in ECL2 (Fig. 2G). When the residue N187^45.51 was mutated to alanine, it significantly reduced CST17-induced receptor activation, resulting in approximately a 110-fold decrease in CST17-mediated G_i signaling (Fig. 2H and SI Appendix, Table S2). T12^C can also interact with L96^2.60 and Q99^2.63 (Fig. 2G). L96^2.60A and Q99^2.63A mutations significantly reduce the signaling of CST17 to SSTR5 (Fig. 2H and SI Appendix, Table S2). In addition, F13^C can form polar contact with T185^ECL2 (Fig. 2G), and the mutation of T185^ECL2A markedly decreases the potency of CST17 (Fig. 2H and SI Appendix, Table S2). Collectively, these results delineate the extensive polar and hydrophobic interactions mediating distinct recognition of CST17 by SSTR5. Together with previous studies, the hydrophobic and charged amino acids within ECL2, TM2, TM3, TM6, and TM7 play an indispensable role in CST17 recognition and subsequent activation of the receptor (5).

Recognition Mechanism of Octreotide by SSTR5.

Unlike the endogenous peptides SST14 and CST17, octreotide is a short 8-residue synthetic cyclic analogue, but it retains the key bioactive motif of SST14 and CST17. Furthermore, octreotide substitutes 2 amino acids with D-form at positions 7 and 10 to improve its pharmaceutical properties, with the F^DWKT (superscript D indicating D-configuration) conserved motif nested between the disulfide bond residues C8^O and C13^O (Fig. 2B). In the activated SSTR5, octreotide also adopts a disulfide-stabilized β hairpin structure with the key pharmacophore (F^DWKT) and binds upright the pocket of SSTR5 (Fig. 3 A and B). The conserved F^DWKT motif inserts at the bottom of the orthosteric binding pocket of SSTR5, where the core region ^DWK forms extensive hydrophobic interactions with the receptor, stabilizing the ligand at the pocket of the receptor (Fig. 3C).

Fig. 3.

Molecular basis of SSTR5 recognition by octreotide. (A) The ligand-binding pocket in octreotide-SSTR5 structure. (B) Schematic representation of the interactions between SSTR5 and octreotide analyzed using the LigPlot+ program (37). Atoms are shown as circle and colored by black (carbon), red (oxygen), and blue (nitrogen). The disulfide bond is indicated by the yellow line. (C) The orthosteric octreotide-binding pocket of SSTR5 shown as surface. (D) Inhibition of forskolin-stimulated cAMP accumulation of WT SSTR5 and its mutants induced by octreotide. ΔpEC₅₀ represents the difference between pEC₅₀ values of the mutant SSTR5 and the WT SSTR5. Data are shown as means ± SEM from at least three independent experiments performed in triplicate. The significance was determined with one-way ANOVAfollowed by Dunnett’s multiple comparisons test compared with WT. The P value was defined as *P < 0.05; **P < 0.01; and ***P < 0.001; ns: not significant, nd: not detected. Detailed statistical evaluation is shown in SI Appendix, Table S3.

In contrast to CST17, K11^O (superscript O indicating residues from octreotide) is positioned slightly farther away from the polar amino acids at the bottom of the SSTR5 pocket. As a result, K11^O forms polar interactions with D119^3.32 and Y89^2.53 (Fig. 3C). The cAMP assays show that D119^3.32A mutation abolishes the potency of octreotide -induced receptor activation and Y89^2.53A mutation also reduces the potency of octreotide (Fig. 3D). In addition, the side chain of K11^O can form a hydrophobic interaction with F264^6.51 (Fig. 3C), and alanine mutation of F264^6.51 remarkably diminishes the activity by ~115-fold (Fig. 3D and SI Appendix, Fig. S7 and Table S3). Both F9^O and ^DW10^O bury themselves into distinct hydrophobic pockets formed by the receptor residues Q123^3.36, F124^3.37, F201^5.38, I202^5.39, F264^6.51, F265^6.52, N268^6.55 and W190^ECL2, G198^5.35, F201^5.38, I202^5.39, respectively, causing ECL2 to be pulled towards the ligand (Fig. 3A). T12^O makes hydrophobic interaction with L96^2.60 and Q99^2.63 (Fig. 3 A and B). C8^O and C13^O make hydrophobic constructs with Y286^7.35and N187^45.51 (Fig. 3 A and B), and alanine mutations of Y286^7.35 and N187^45.51 remarkably reduce octreotide-induced bioactivity (Fig. 3D and SI Appendix, Table S3). Apart from ECL2, ECL3 also contributes to extensive interactions with octreotide. ^DF7^O is covered by L276^ECL3, P277^ECL3, Q278^ECL3, and E279^ECL3 at ECL3 of SSTR5. T14^O establishes hydrophobic interaction with A283^7.32 and Y286^7.35 (Fig. 3 A and B). In summary, octreotide primarily utilizes hydrophobic interactions for binding and recognition of SSTR5. In addition to this, ECL2 and ECL3, alongside TM2, TM3, and TM5-TM7, play pivotal roles in ligand recognition.

Distinct Binding Modes of CST17 and Octreotide in SSTR5.

Compared to the binding mode of CST17 and octreotide in SSTR5, they both share a conserved F^L/DWKT motif (Fig. 2B) and exhibit a similar set of interactions with SSTR5 (SI Appendix, Fig. S8). However, due to the differing positions of the disulfide bonds within the two peptides, there are structural conformation variances within the pocket, leading to distinct interaction modes with the receptor (SI Appendix, Fig. S8).

Structural alignment revealed a vertical shift in the WK motif of CST17 when contrasted with the ^DWK motif of octreotide (Fig. 2A). Consequently, the distance between K11^O and the polar amino acids is greater than 3.2 Å and K11^O fails to establish hydrogen bond with Y89^2.53 (Fig. 3A). Given this, alanine mutation of Y89^2.53 has a more severe impact on reducing ligand-induced activity for CST17 compared to octreotide (Figs. 2H and 3D and SI Appendix, Tables S2 and S3).

It is noteworthy that the CST17, with a disulfide bond between C5^C and C16^C, protrudes outward from the binding pocket, while the disulfide bond of octreotide (C8^O to C13^O) is situated inside the pocket and oriented toward TM7. The position corresponding to the disulfide bond of octreotide (C8^O to C13^O) is occupied by F8^C and F13^C in CST17. Because of the bulkier side chain of F8^C and F13^C in CST17 than a disulfide bond in octreotide, Y286^7.35 in the CST17-bound SSTR5 moves outward in comparison to that in the octreotide-bound SSTR5, resulting in the rearrangement of F287^7.36 and the formation of hydrophobic interactions (Fig. 4A).

Fig. 4.

The different binding modes of CST17 and octreotide in SSTR5. (A) Structural comparison of active SSTR5 bound by CST17 and octreotide. (B) The detailed TM and ECL variations of SSTR5 activated by CST17 and octreotide. All TMs are shown in tube. (C) Differences in the conformation of ECL2 of two complex structures. The polarity network is shown in black dashed lines. (D) Differences in the conformation of ECL3 of two complex structures. (E and F) Dose–response curves of SSTR5 for CST17 and octreotide. WT and mutants of SSTR5, measuring the inhibition of cAMP production upon forskolin stimulation, induced by CST17 (E) or octreotide (F). The response data were normalized by WT receptor within each individual experiment. Data from three independent experiments, each of which was performed in triplicate, are presented as mean ± SEM. Detailed statistical evaluation is shown in SI Appendix, Tables S2 and S3.

In addition, the extracellular receptor loops play distinct functional roles in recognizing two peptides, octreotide and CST17. Specifically, when octreotide binds to SSTR5, the extracellular receptor loops undergo an upward shift compared to their positioning in CST17-bound SSTR5 (Fig. 4 B–D). This suggests that the binding of SSTR5 to these two peptides involves conformational changes in the extracellular receptor loops, which may be crucial for the specific recognition and binding of each peptide. In the CST17-SSTR5 structure, N187^45.51 and T185^ECL2 form polar interactions with the T12^C and F13^C backbones, respectively, and both N187^45.51A and T185^ECL2A mutants markedly reduce the ligand-induced activity (Figs. 2G and 4 C and E and SI Appendix, Table S2). Despite ECL2 not forming polar interactions with octreotide, an essential intrareceptor polar network comprising Q99^2.63, T185^ECL2, and N187^45.51 emerges in octreotide binding (Fig. 4C). This network plays a crucial role in stabilizing the receptor’s conformation, a feature not observed with CST17. Single-point mutations of Q99^2.63A and N1874^5.51A resulted in varying degrees of reduction in receptor activation (Figs. 3D and 4F and SI Appendix, Table S3), while either N187^45.51T or T185^ECL2W mutation in SSTR5, making them have SSTR2’s ECL2 residues, significantly enhanced the efficacy of octreotide in activating SSTR5 (SI Appendix, Fig. S7 and Table S3). Both amino acid mutations disrupted the polar network on the receptor, and we assumed that the polar network formed by Q99^2.63, T185^ECL2, and N187^45.51 may be a hindrance to activation of SSTR5 by octreotide. This also indicates that ECL2 is important for octreotide selectivity (32).

Moreover, ECL3 of SSTR5 is drawn closer to octreotide upon binding, enabling extensive hydrophobic interactions between the two. Specific residues L276^ECL3, P277^ECL3, Q278^ECL3, and E279^ECL3 located on ECL3 participate in hydrophobic contacts with octreotide. In particular, Q278^ECL3 forms cation–π interaction with octreotide (Fig. 4D), thereby enhancing the stabilization of the receptor with the binding pocket. The substitution of Q278^ECL3 in SSTR5 with P as found in SSTR2, increased potency of stimulating SSTR5 signaling (SI Appendix, Figs. S7 and S9 and Table S3).The enhancement is probably due to its special side-chain loop forming a small cap that provides a hydrophobic environment for the phenylalanine at the N-terminal end of octreotide. In contrast, in the CST17-bound SSTR5 structure, only E279^ECL3 interacts with CST17 (Fig. 4D). These findings imply that SSTR5 adopts a different structural adaptation in recognition of CST17 versus octreotide. Collectively, these results suggest that ECL2 and ECL3 plays a critical role of in agonist binding and consequent receptor activation.

Activation Mechanism of SSTR5.

Structural comparison of the active-state SSTR5 in complex with either CST17 or octreotide against the inactive-state SSTR2 reveals conserved structural features characteristic of class A GPCR activation. Compared to inactive SSTR2 (PDB ID: 7XNA) (38), the structure of activated state SSTR5 undergoes distinct conformational changes, including an outward movement of the cytoplasmic tail of TM6 and an inward movement of TM7, which highlights common mechanisms of receptor activation (Fig. 5A). Interestingly, while we observed shared mechanisms of receptor activation, we also identified subtle deviations in the activation mechanism of SSTR5 compared to established knowledge about class A GPCRs.

Fig. 5.

Activation mechanism of SSTR5. (A) Superposition of activated SSTR5 (pale blue) bound to CST17 with active SSTR5 (medium turquoise) bound to octreotide and inactive SSTR2 (dark gray; PDB code: 7XNA). Notable conformational changes occur at intracellular ends of TM6 and TM7 upon receptor activation, front view and top view. (B) Compared with inactive SSTR2 bound to CYN154806, “hydrophobic lock” in SSTR5 exhibits rotameric rearrangements upon agonist binding. (C) Inhibition of forskolin-stimulated cAMP accumulation of WT and mutants of SSTR5 induced by CST17 and octreotide. ΔpEC₅₀ represents the difference between pEC₅₀ values of the mutant SSTR5 and the WT SSTR5. Data are shown as means ± SEM from at least three independent experiments performed in triplicate. The significance was determined with one-way ANOVA followed by Dunnett’s multiple comparisons test compared with WT. The P value was defined as *P < 0.05; **P < 0.01; and ***P < 0.001. Detailed statistical evaluation is shown in SI Appendix, Tables S2 and S3. (D–F) The key P^5.50V^3.40F^6.44 (D) motif (PIF motif in common GPCRs), D^3.49R^3.50Y^3.51 (E) motif, and N^7.49P^7.50xxY^7.53 (F) motif displayed conformational rearrangements in activated SSTR5 by CST17 or octreotide.

Unlike the loss of the hydrogen bond between Q126^3.36-Y273^6.52 of SSTR2 that allows TM3 to be displaced away from TM6 to active the receptor, the ligand-bound active structures of SSTR5 reveal a distinct mechanism (SI Appendix, Fig. S10). In both CST17-bound and octreotide-bound activated SSTR5, hydrophobic interactions between tryptophan in the 10th position and residues Q123^3.36, F124^3.37, F264^6.51, and F265^6.52 on TM3 and TM6 lead to the rearrangement of these amino acids and the cascade of conformational changes (Fig. 5B and SI Appendix, Fig. S10C). Mutation of these residues to alanine significantly reduces potency for both ligands (Fig. 5C).

The hydrophobic residues F124^3.37, F264^6.51, and F265^6.52 form a densely packed “hydrophobic lock” (Fig. 5B) that effectively stabilizes the packing between TM3 and TM6, which contributes to SSTR5 activation (Fig. 5C). Due to the lower position of W10^C in CST17 compared to octreotide, the phenyl groups of F264^6.51 and F265^6.52 tilt toward W261^6.48, acting as a “toggle switch” to trigger receptor activation. In contrast, with ^DW10^O positioned relatively higher in octreotide, F264^6.51 and F265^6.52 tilt upward, and W261^6.48 also shifts slightly upward (Fig. 5B). These conformational changes cause the outward movement of F257^6.44, which in turn leads to outward displacement of the cytoplasmic segment of TM6 in the octreotide-bound state (SI Appendix, Fig. S10C). The conserved P^5.50 V (I) ^3.40 F^6.44 motif also undergoes rearrangement (Fig. 5D). The extensive hydrophobic interactions formed by these conserved conduction motifs (3.40, 6.51, 6.52, 6.44, and 6.48) stabilize active receptor conformations. Notably, W^6.48 is a conserved residue in class A GPCRs, commonly referred to as a “toggle switch” element that undergoes conformational changes upon activation of most class A GPCRs (39). W261^6.48 acts as a toggle switch in CST17-bound SSTR5 similar to W272^6.48 in the active-state SSTR4 induced by SST14. However, W261^6.48 in octreotide-bound SSTR5 remains largely unchanged, resembling W269^6.48 in SST14 induced active-state SSTR2 conformation (Fig. 5B and SI Appendix, Fig. S10) (32, 38).

In active SSTR5, the D^3.49R^3.50Y^3.51 motif is rearranged, breaking a hydrogen bond between D136^3.49 and R137^3.50. This facilitates the insertion of α5 helix of G_αi into the cytoplasmic end of SSTR5 to propagate downstream signaling (Fig. 5E). Additionally, Y89^2.53 in active SSTR5 undergoes downward movement, unlike F92^2.53 in inactive SSTR2. This exerts pressure on TM3, disrupting interactions between D86^2.50, N122^3.35, and S297^7.46. Consequently, the N^7.49P^7.50xxY^7.53 motif is reorganized, shifting the cytoplasmic end of TM7 inward (Fig. 5F). In summary, upon activation, SSTR5 undergoes rearrangements of crucial motifs. These structural changes break stabilizing interactions, triggering larger conformational changes that enable the ligand-induced receptor activation and signal transduction.

Discussion

SSTRs are important therapeutic targets for neuroendocrine tumors, and SSTR5 has emerged as a promising target for metabolic diseases. In this study, we elucidated the molecular mechanisms underlying agonist selectivity and activation of SSTR5 using high-resolution cryo-EM structures of SSTR5 bound to the peptide agonists CST17 and octreotide.

Our structural and functional analyses reveal that while CST17 and octreotide share a conserved FWKT motif, they exhibit distinct binding modes and recognition by SSTR5 due to differences in their disulfide bond positions. The distinct disulfide bond locations result in conformational variations between CST17 and octreotide when bound to SSTR5. Consequently, the conserved lysine in the WK motif forms stronger polar contacts in CST17 compared to octreotide. Mutagenesis studies validate the significance of these polar interactions unique to CST17 binding.

In addition, our findings highlight the critical roles of the ECLs, especially ECL2 and ECL3, in mediating differential recognition of CST17 and octreotide. ECL2 forms crucial polar contacts with CST17 but adopts a distinct conformation and intrareceptor polar network upon octreotide binding. Meanwhile, ECL3 undergoes more extensive hydrophobic interactions with octreotide compared to CST17. Collectively, these extracellular elements enable the receptor to distinguish between the two agonists.

Structural comparison between inactive SSTR2 and active SSTR5 confirms hallmarks of class A GPCR activation like outward TM6 movement. However, our octreotide-bound structure reveals variations from the canonical “toggle switch” mechanism mediated by W^6.48. Instead, octreotide binding induces rearrangements of a conserved “hydrophobic lock,” triggering the cascading conformational changes leading to activation.

In summary, these high-resolution structures provide valuable insights into the molecular determinants governing agonist selectivity and activation of SSTR5. Our findings guide ongoing structure-based drug design efforts to develop improved SSTR5-selective agonists for treating pituitary tumors and other neuroendocrine disorders.

Materials and Methods

Constructs.

The human SSTR5 was modified with an HA signal peptide at its N terminus and a LgBiT subunit at its C terminus (40, 41). To promote protein solubility and proper folding, OMBP and MBP fusion tags were inserted at both ends of WT SSTR5, respectively, using homologous recombination (CloneExpress One Step Cloning Kit, Vazyme). One mutation, V253^6.40L, was introduced (42) to increase protein yield and promote complex formation. The engineered Gα_i was designed as a chimera based on the mini-G_s/i143 skeleton. It contains the G_i1 N-terminal 1-18 amino acids and the GαAH domain. This engineered Gα_i can bind NB35 and scFv16, respectively (34, 43). Gβ₁ was fused with a SmBiT subunit (peptide 86) at a C terminus (40, 41) by a 15-residue linker. SSTR5, engineered Gα_i (43), His8-Gβ₁, Gγ₂, and a single-chain antibody scFv16 (SI Appendix, Fig. S11) were cloned into pFastBac vectors (34). These constructs were generated in insect cell expression vectors.

Insect Cell Expression.

Human SSTR5-LgBiT, human Gα_i, rat His8-Gβ₁-SmBiT, and bovine Gγ₂ and scFv16 were coexpressed in High Five insect cells (Thermo Fisher) cultured in serum-free medium ESF921 using the baculovirus method (Expression Systems) in the ratio of 1:1:1:1:1. The cells were infected at a concentration of 2.5-3.0 ×10⁶ cells/mL at 27 °C, and cultures were collected after 48 h. Cells were harvested and stored at −80 °C.

Complex Purification.

For purification of CST17-SSTR5-G_i and octreotide-SSTR5-G_i complexes, cell pellets were lysed in a buffer containing 20 mM HEPES pH 7.4, 100 mM NaCl, 5 mM MgCl₂, 5 mM CaCl₂, 0.1 mM TCEP (Sigma-Aldrich), 10% (v/v) glycerol, and EDTA-free protease inhibitor cocktail (TargetMol). 10 µM CST17 or octreotide and 25 mU/mL apyrase (Sigma) were added, and cells were lysed by homogenization using a glass dounce tissue grinder. The suspension was incubated at room temperature for 1 h, and then, the complexes were solubilized from the membrane with 0.5% (w/v) lauryl maltose neopentylglycol (LMNG) and 0.1% (w/v) cholesteryl hemisuccinate TRIS salt (CHS, Anatrace) at 4 °C for 3 h. The supernatant was collected by centrifugation at 65,000 g for 35 min and incubated with MBP beads overnight at 4 °C. After centrifugation at 500 g for 5 min, MBP beads were loaded onto a gravity flow column and washed with 20 column volumes of buffer containing 20 mM HEPES pH 7.4, 100 mM NaCl, 5 mM MgCl₂, 5 mM CaCl₂, 10 µM CST17 or octreotide, 0.05% (w/v) (LMNG), 0.005%(w/v) glycol-diosgenin (GDN, Anatrace), and 0.011% (w/v) (CHS). The complexes were eluted with the buffer containing 30 mM maltose and concentrated using an Amicon Ultra Centrifugal Filter (MWCO, 100 kDa). The complexes were then loaded onto a pre-equilibrated Superose 6 Increase 10/300 GL column (GE Healthcare) with size buffer containing 20 mM HEPES pH 7.4, 100 mM NaCl, 0.00075% (w/v) LMNG, 0.00025% (w/v) GDN (Anatrace) and 0.0002% CHS, and 20 μM corresponding ligand. The fractions containing complex were collected and evaluated by SDS-PAGE (SI Appendix, Fig. S11 B–E) and then concentrated for cryo-EM analysis.

G_i-Mediated cAMP Inhibition Assay for SSTRs.

The intracellular 3′, 5′-cyclic AMP (cAMP) levels were measured using the GloSensor cAMP assay (Promega) as described in the previous studies (36, 44). The WT and mutants of SSTR5 were cloned into the pcDNA6.0 vector (Invitrogen) with a 3×FLAG tag at the N terminus (45). First, AD293 cells were seeded in a 12-well plate with DMEM containing 10% FBS and allowed to grow and adhere overnight. Cells were cotransfected with 1,100 ng total plasmid mixture consisting of SSTR5 (WT or mutants) and the GloSensor22F cAMP biosensor (Promega) at a 1:1.5 ratio. After 24 h incubation at 37 °C with 5% CO2, the cells were harvested using trypsin, washed once with PBS, and collected by centrifugation at 100 × g for 5 min. The collected cells were then resuspended in CO₂-independent medium with 2% GloSensor cAMP Reagent (Promega) and distributed into a white 384-well plate at 10 μL per well and then incubated in the dark at 37 °C for 1 h. Agonist dilutions at gradient concentrations in CO₂-independent medium containing 3 μM forskolin (Sigma) were then added at 5 μL per well. After 15 min incubation at room temperature, luminescence was measured using an Envision multilabel plate reader (PerkinElmer). Finally, sigmoidal dose–response nonlinear regression analysis was performed in GraphPad Prism to calculate half-maximum effective concentration (EC₅₀) and Emax values. The data presented were means from at least three independent experiments performed in triplicate.

Detection of Surface Expression of Mutants.

The wild-type SSTR5 and mutants were cloned into the pcDNA6.0 vector (Invitrogen) with a 3×FLAG tag at the N terminus. AD293 cells were cultured in DMEM supplemented with 10% (v/v) FBS at 37 °C with 5% CO₂ incubator. The day before transfection, AD293 cells were seeded at a density of 1.6 × 10⁵ cells per well in a 12-well plate and allowed to grow and adhere. After overnight incubation, a total of 1,100 ng of SSTR5 wild-type or mutants were transfected into each well. After 24 h of transfection, the cells were harvested using cell dissociation buffer enzyme-free PBS. The cells were blocked with 5% (w/v) BSA (dissolved in PBS) for 15 min at room temperature and then incubated with primary anti-Flag antibody (diluted 1:300 in PBS containing 5% BSA, ABclonal) for 1 h at room temperature. The cells were then washed 3 times with PBS containing 1% (w/v) BSA before incubating with anti-mouse Alexa-488-conjugated secondary antibody (diluted 1:1,000 in PBS containing 5% BSA, ABclonal) at 4 °C in the dark for 1 h. After washing another three times, the cells were resuspended in PBS, and fluorescence intensity was measured using the Guava easy-Cyte flow cytometer system (Luminex) at excitation 488 nm and emission 519 nm. Approximately 10,000 cellular events per sample were collected, and data were normalized from WT. All data presented are means at least three independent experiments.

EM Data Collection and Processing.

The purified complexes (CST17-SSTR5-G_i-scFv16 and octreotide-SSTR5-G_i-scFv16) were applied onto the freshly glow-discharged Quantifoil holey carbon grid (R1.2/1.3, Au, 300 mesh) using a Vitrobot at 4 °C in 100% humidity. The prepared grids of the CST17-SSTR5-Gi-scFv16 complex were transferred to a Titan Krios G4 microscope (Thermo Fisher Scientific), operated at an operated at an accelerating voltage of 300 kV with a Gatan Quantum-LS Energy Filter (GIF) and a Gatan K3 Summit direct electron detector. Images were collected at a nominal magnification of 105 K, corresponding to a calibrated pixel size of 0.824 Å per pixel. The dataset was acquired using the EPU software, with a defocus range of −1.0 to −2.0 μm. Each image was dose-fractionated to 36 frames at a dose rate of 15 e⁻ Å⁻¹ s⁻¹ to accumulate a total dose of 50 e⁻ Å⁻². The 5,043 dose-fractionated movies were subjected to beam-induced motion correction using RELION 4.0 (46), and the contrast transfer function and the defocus parameters were estimated using CTFFIND4.1 (47). The 2,247,486 particles were initially picked from the 5,043 movies using the reference-based template-matching picking function. These particles were subjected to several rounds of 2D and 3D classifications. Finally, 320,649 particles were subjected to 3D refinement. The homogenous subset was subjected to per-particle defocus refinement, beam-tilt refinement, Bayesian polishing, and 3D refinement. The final 3D refinement and postprocessing produced a map with global resolution at 2.7 Å according to the Fourier shell correlation = 0.143 criterion. The final density was further processed with DeepEMhancer package (48).

The prepared grids of the octreotide-SSTR5-G_i-scFv16 complex were transferred to a Titan Krios G4 microscope (Thermo Fisher Scientific), operated at an accelerating voltage of 300 kV with a Gatan Quantum-LS Energy Filter (GIF) and a Gatan Falcon IV Summit direct electron detector. Images were collected at a nominal magnification of 165 K, corresponding to a calibrated pixel size of 0.73 Å per pixel. The dataset was acquired using the EPU software, with a defocus range of −0.8 to −2.8 μm. Each image was dose-fractionated to 784 frames at a dose rate of 15.4 e⁻ Å⁻¹ s⁻¹ to accumulate a total dose of 50 e⁻ Å⁻². The 8,058 dose-fractionated movies were subjected to beam-induced motion correction using RELION 4.0, and the contrast transfer function and the defocus parameters were estimated using CTFFIND4.1. The 3,255,556 particles were initially picked from the 8,058 movies using the reference-based template-matching picking function. These particles were subjected to several rounds of 2D and 3D classifications. Finally, 235,599 particles were subjected to 3D refinement. The homogenous subset was subjected to per-particle defocus refinement, beam-tilt refinement, Bayesian polishing, and 3D refinement. The final 3D refinement and postprocessing produced a map with global resolution at 2.9 Å according to the Fourier shell correlation = 0.143 criterion. The final density was also further processed with the DeepEMhancer package.

Model Building.

The predicted SSTR5 structure from AlphaFold2 was used as the starting reference model for the CST17-SSTR5-G_i-scFv16 model building (49). Structures of Gα_i derived from PDB entry 7E2Z (50), Gβ₁, Gγ₂, and scFv16 derived from PDB entry 7W53 (51) were rigid body fit into the density. All models were fitted into the EM density map using UCSF Chimera (52) followed by iterative rounds of manual adjustment and automated rebuilding in COOT (53) and PHENIX (54), respectively. The model was finalized by rebuilding in ISOLDE (55) followed by refinement in PHENIX with torsion-angle restraints to the input model. The final model statistics were validated using Comprehensive validation for cryo-EM structures in PHENIX and provided in SI Appendix, Table S1. The octreotide-SSTR5-G_i-scFv16 model is built using the CST17-SSTR5-G_i-scFv16 model as the initial reference. All structural figures were prepared using Chimera X (56).

Supplementary Material

Appendix 01 (PDF)

Data, Materials, and Software Availability

Cryo-EM maps have been deposited in the Electron Microscopy Data Bank under accession codes: EMD-38148 (57) (CST17-SSTR5-G_i complex) and EMD-38150 (58) (octreotide-SSTR5-G_i complex). The atomic coordinates have been deposited in the Protein Data Bank under accession codes: 8X8L (59) (CST17-SSTR5-G_i complex) and 8X8N (60) (octreotide-SSTR5-G_i complex). All other data are included in the manuscript and/or SI Appendix.