Allosteric antagonism of insect odorant receptor ion channels

Abstract

Background: At a molecular level, insects utilize members of several highly divergent and unrelated families of cell-surface chemosensory receptors for detection of volatile odorants. Most odors are detected via a family of odorant receptors (ORs), which form heteromeric complexes consisting of a well-conserved OR co-receptor (Orco) ion channel and a non-conserved tuning OR that provides coding specificity to each complex. Orco functions as a non-selective cation channel and is expressed in the majority of olfactory receptor neurons (ORNs). As the destructive behaviors of many insects are principally driven by olfaction, Orco represents a novel target for behavior-based control strategies. While many natural and synthetic odorants have been shown to agonize Orco/Or complexes, only a single direct Orco modulator, VUAA1, has been described. In an effort to identify additional Orco modulators, we have investigated the structure/activity relationships around VUAA1.

Results: A search of our compound library identified several VUAA1 analogs that were selected for evaluation against HEK cells expressing Orco from the malaria vector Anopheles gambiae (AgOrco). While the majority of compounds displayed no activity, many of these analogs possess no intrinsic efficacy, but instead, act as competitive VUAA1 antagonists. Using calcium mobilization assays, patch clamp electrophysiology, and single sensillum in vivo recording, we demonstrate that one such candidate, VU0183254, is a specific allosteric modulator of OR signaling, capable of broadly inhibiting odor-mediated OR complex activation.

Conclusions: We have described and characterized the first Orco antagonist, that is capable of non-competitively inhibiting odorant-evoked activation of OR complexes, thereby providing additional insight into the structure/function of this unique family of ligand-gated ion channels. While Orco antagonists are likely to have limited utility in insect control programs, they represent important pharmacological tools that will facilitate the investigation of the molecular mechanisms underlying insect olfactory signal transduction.

Figure 1. VUAA1 analogs can antagonize AgOr responses.

A. Structures of the Orco agonist VUAA1, Orco antagonist VU0183254, and control VU0450667. B. IC₅₀ curve of the response curves of AgOrco-only expressing HEK cells in the presence of pre-loaded antagonist (VU0183254 or VU0450667) followed by an addition of VUAA1 to 100 µM; measurements were taken on an FDS6000 using Fluo4-AM as described in methods C. IC₅₀ curve of AgOrco+AgOr65 expressing HEK cells as in B., followed by addition of eugenol to 1 µM.

Figure 2. VU0183254 reduces OR-mediated currents.

A–B. Whole-cell patch clamp recordings of odorant- and VUAA1-induced currents in AgOr-expressing cells (n = 3). A. VU0183254 (100 µM) decreased responses to delta-undecalactone (1 µM) and VUAA1 (100 µM) in HEK cells expressing AgOrco+AgOr48. B. Responses to eugenol (1 µM) and VUAA1 (100 µM) in AgOrco+AgOr65 cells were reduced by VU0183254 (100 µM). C–D. Cells expressing Orco from either An. gambiae (C) or Harpegnathos saltator (D) had reduced VUAA1 (100 µM) currents in the presence of 100 µM VU0183254. E. Capsaicin (10 µM) currents in HEK cells expressing rat TRPVI were not reduced with by 100 µM VU0183254. Holding potential for all figures is −60 mV.

Figure 3. VU0183254 is an allosteric antagonist.

A. Concentration response curves of AgOrco+AgOr65 expressing HEK cells in the presence of a series of steady concentrations (expressed as logM) of pre-loaded VU0183254 (different colored lines, see inset) followed by increasing concentrations of eugenol as measured using Fluo-4AM and an FDSS6000. B. As in A with VUAA1. C. As in A, with AgOrco+AgOr48 and delta-undecalactone. D. As in C with VUAA1. In all cases results are shown as means+/−SEM, n = 4.

Figure 4. VU0183254 reduces Orco-mediated activity in vivo.

A. Representative traces of Cp neuron activity in response to vehicle (DMSO) or VU0183254 as measured by single-sensillum electrophysiology. Activity was allowed to stabilize for 10 seconds and then recorded for 60 seconds. CpA spikes can be distinguished by their larger amplitude in expansions (right panel). CpB/C spikes (counts below traces) are reduced in the presence of VU0183254. B. Spontaneous CpA spike rates are unaffected by the presence of VU0183254 (n = 8). Spikes were counted for each of the first 10 seconds and then averaged across the remaining 10-second intervals. Normal CpA activity is confirmed by CO₂ pulse delivered at the end of the recording period. C. Spontaneous firing rates of CpB/C neurons are reduced by VU0183254 in a dose-dependent manner (n = 8). All spikes were distinguished and quantified using AutoSpike software (Syntech).

Figure 5. VU0183254 antagonizes odor-evoked ORN activity in vivo.

A. Representative traces of Cp neuron activity in response to a 1 s pulse of 1-octen-3-ol (dark bar above traces). Vertical lines under each raw trace indicates spike counter for CpB and CpC spikes only. Vehicle-only trace (top) responds strongly to 1-octen-3-ol, while preparations exposed to VU0183254 (bottom) do not respond. B. Quantification of traces as in (A). Spike frequencies were binned every second for 5 seconds prior to 1-octen-3-ol pulse and for 2 seconds afterwards. In the presence of vehicle alone (blue circles, n = 11) CpB/C neurons respond to a 1 s pulse of 1-octen-3-ol with increased activity, while preparations including VU0183254 (red squares)(n = 11) do not respond as strongly (*, p<0.05; **, p<0.005).