A Biomimetic C-Terminal Extension Strategy for Photocaging Amidated Neuropeptides

Abstract

Photoactivatable neuropeptides offer a robust stimulus-response relationship that can drive mechanistic studies into the physiological mechanisms of neuropeptidergic transmission. The majority of neuropeptides contain a C-terminal amide, which offers a potentially general site for installation of a C-terminal caging group. Here, we report a biomimetic caging strategy in which the neuropeptide C-terminus is extended via a photocleavable amino acid to mimic the proneuropeptides found in large dense-core vesicles. We explored this approach with four prominent neuropeptides: gastrin-releasing peptide (GRP), oxytocin (OT), substance P (SP), and cholecystokinin (CCK). C-terminus extension greatly reduced the activity of all four peptides at heterologously expressed receptors. In cell type-specific electrophysiological recordings from acute brain slices, subsecond flashes of ultraviolet light produced rapidly activating membrane currents via activation of endogenous G protein-coupled receptors. Subsequent mechanistic studies with caged CCK revealed a role for extracellular proteases in shaping the temporal dynamics of CCK signaling, and a striking switch-like, cell-autonomous anti-opioid effect of transient CCK signaling in hippocampal parvalbumin interneurons. These results suggest that C-terminus extension with a photocleavable linker may be a general strategy for photocaging amidated neuropeptides and demonstrate how photocaged neuropeptides can provide mechanistic insights into neuropeptide signaling that are inaccessible using conventional approaches.

Figure 1

Biomimetic approach to photocaging C-terminally amidated neuropeptides. The biosynthesis of amidated neuropeptides involves enzymatic conversion of glycine to glyoxylate by peptidylglycine alpha-amidating monoxygenase (PAM). An adjacent dibasic motif initiates proteolytic processing prior to the oxidative cleavage by PAM to produce the active, C-terminally amidated neuropeptide. To mimic this process, the photocaged neuropeptide is C-terminally extended with a photocleavable amino acid followed by a dibasic motif and several charged, sterically bulky amino acids. Exposure to UV light removes the caging group to release the active, C-terminally amidated neuropeptide.

Figure 2

Chemical structures of gastrin releasing peptide (14-27), oxytocin, substance P, and cholecystokinin (8S). All four peptides contain a C-terminal amide but otherwise exhibit diverse structural features.

Scheme 1. Synthesis of NPP-Caged Variants of Gastrin-Releasing Peptide (14-27), Oxytocin, Substance P, and Cholecystokinin (8S)

(A) Fmoc-protected 2-(4,5-dimethoxy-2-nitrophenyl) beta-alanine (4) was prepared from 4,5-dimethoxybenzenaldehyde via a Mannich reaction followed by functional group protection and nitration. (B) NPP-caged peptides were prepared from 4 via solid phase peptide synthesis (SPPS).

Figure 3

In vitro characterization of NPP-caged neuropeptides using a functional assay of G protein signaling. (A) Dose–response curves at the gastrin-releasing peptide receptor (GRPR) using a GloSensor assay of cAMP signaling in HEK293T cells (n = 10 wells per data point). Data were normalized to the maximal response to GRP(14-27) (300 nM) and are expressed as the mean ± SEM. (B) Same as panel A, but using the oxytocin receptor (OTR) and oxytocin (OT, 300 nM) for normalization. (C) Same as panel A, but using the neurokinin 1 receptor (NK1R) and substance P (SP, 300 nM) for normalization. (D) Same as panel A, but using the cholecystokinin receptors (CCK1R and CCK2R) and cholecystokinin-8S (CCK(8S), 300 nM) for normalization.

Figure 4

Electrophysiological validation of NPP-caged neuropeptides at endogenous receptors in acute brain slices. (A) Average inward currents over time after bath application of GRP(14-27)-NPP (3 μM, n = 5 cells from 4 mice) or GRP(14-27) (300 nM, n = 5 from 4 mice), recorded from fluorescently labeled VIP interneurons in layer 1 of the motor cortex. Data are expressed as the mean ± SEM. (B) Average inward currents evoked by photoactivation of GRP(14-27)-NPP (3 μM) with an 84 mW light flash in the absence (green, n = 10 cells from 4 mice) and presence of the GRPR antagonist BW1023U90 (1 μM) (black, n = 6 cells from 2 mice). (C) Summary of peak current amplitudes for the data shown in panels A and B. GRP(14-27) −62.8 ± 12.8; GRP(14-27)-NPP −4.8 ± 8.4; GRP(14-27)-NPP + UV −54.6 ± 14.4; GRP(14-27)-NPP + UV + BW1023U90 1.5 ± 0.7. Data are expressed as the mean ± SEM. Ordinary one-way ANOVA F (3, 22) = 6.37, p = 0.0028, p-values determined using Sidak’s multiple comparison’s test. (D) Average inward currents over time after bath application of OT-NPP (3 μM, n = 7 cells from 5 mice) or OT (300 nM, n = 7 from 4 mice), recorded from fluorescently labeled PV interneurons in the CA1 and CA2 regions of hippocampus. (E) Average inward currents evoked by photoactivation of OT-NPP (3 μM) with an 84 mW light flash in the absence (light blue, n = 8 cells from 4 mice) and presence of the OTR antagonist (OTA) (d(CH2)51,Tyr(Me)2,Thr4,Orn8,des-Gly-NH29)-Vasotocin (1 μM) (black, n = 5 cells from 2 mice). (F) Summary of peak current amplitudes for the data shown in panels D and E. OT −35.8 ± 5.7; OT-NPP 2.4 ± 6.5; OT-NPP + UV −35.1 ± 5.3; OT-NPP + UV + OTA 5.5 ± 5.5. Data are expressed as the mean ± SEM. Ordinary one-way ANOVA F (3, 23) = 14.83, p <0.0001, p-values determined using Sidak’s multiple comparison’s test. (G) Average inward currents over time after bath application of SP-NPP (1 μM, n = 6 cells from 5 mice) or SP (500 nM, n = 7 from 2 mice), recorded from fluorescently labeled cholinergic interneurons in the dorsal striatum. (H) Average inward currents evoked by photoactivation of SP-NPP (1 μM) with an 84 mW light flash in the absence (purple, n = 5 cells from 5 mice) and presence of the NK1R antagonist RP67580 (10 μM) (black, n = 6 cells from 3 mice). (I) Summary of peak current amplitudes for the data shown in panels G and H. SP −197.8 ± 20.1; SP-NPP −5.2 ± 13.6; SP-NPP + UV −265.5 ± 19.7; SP-NPP + UV + RP67580 −30. 9 ± 11.7. Data are expressed as the mean ± SEM. Ordinary one-way ANOVA F (3, 21) = 53.13, p <0.0001, p-values determined using Sidak’s multiple comparison’s test. (J) Average inward currents over time after bath application of CCK(8S)-NPP (3 μM, n = 8 cells from 6 mice) or CCK(8S) (500 nM, n = 6 from 3 mice), recorded from fluorescently labeled PV interneurons in the CA1 region of hippocampus. (K) Average inward currents evoked by photoactivation of CCK(8S)-NPP (3 μM) with an 84 mW light flash in the absence (blue, n = 5 cells from 3 mice) and presence of the CCK2R antagonist YM022 (1 μM) (black, n = 6 cells from 2 mice). (L) Summary of peak current amplitudes for the data shown in panels J and K. CCK −41.1 ± 7.6; CCK-NPP −0.4 ± 4.1; CCK-NPP + UV −69.6 ± 9.2; CCK-NPP + UV + 1 uM YM 022 −13.0 ± 2.0. Data are expressed as the mean ± SEM. Ordinary one-way ANOVA F (3, 21) = 26.59, p <0.0001, p-values determined using Sidak’s multiple comparison’s test.

Figure 5

CCK(8S)-NPP photouncaging unmasks temporal features of CCK signaling. (A) Average inward currents evoked by photoactivation of CCK(8S)-NPP (3 μM) with a 20 ms, 84 mW light flash in the absence (black, n = 6 cells from 2 mice) and presence of a cocktail of peptidase inhibitors (phosphoramidon (1 μM), bestatin (20 μM), butabindide ((2 μM)) (blue, n = 4 cells from 2 mice), recorded from fluorescently labeled PV interneurons in the CA1 region of hippocampus. Data are expressed as the mean ± SEM. (B) Summary of current activation time constants for the data shown in panel A. control 0.84 ± 0.30; PIs 0.73 ± 0.17. Data are expressed as the mean ± SEM. Mann–Whitney test. (C) Summary of peak current amplitudes for the data shown in panel A. control −36.8 ± 16.0; PIs −36.6 ± 16.9. Data are expressed as the mean ± SEM. Mann–Whitney test. (D) Summary of the area under the curve (0–4.5 min postflash) for the data shown in panel A. control −56.6 ± 19.9; PIs −146.5 ± 61.7. Data are expressed as the mean ± SEM. Mann–Whitney test. (E) Average outward currents evoked by bath application of leucine-enkephalin (Enk, 1 μM) in the absence (black, n = 5 cells from 2 mice) and presence of subsequent CCK(8S)-NPP (3 μM) uncaging (1 × 200 ms, 84 mW flash, teal, n = 5 cells from two mice). (F) Average inward current evoked by photoactivation of CCK(8S)-NPP in the absence (blue (control), same data as Figure 4K) and presence of Enk (purple, n = 5 cells from 2 mice), subtractively isolated from the data shown in panel E. (G) Summary of peak current amplitudes for the data shown in panel F. Control −69.6 ± 20.6; in Enk −64.5 ± 22.6. Data are expressed as the mean ± SEM. Mann–Whitney test. (H) Summary of percent current loss 5 min after Enk addition for the data shown in panel F. Enk only 31.0 ± 19.8; Enk + CCK photorelease 84.1 ± 38.8. Data are expressed as the mean ± SEM. Mann–Whitney test.