In vivo imaging of C. elegans ASH neurons: cellular response and adaptation to chemical repellents

Abstract

ASH sensory neurons are required in Caenorhabditis elegans for a wide range of avoidance behaviors in response to chemical repellents, high osmotic solutions and nose touch. The ASH neurons are therefore hypothesized to be polymodal nociceptive neurons. To understand the nature of polymodal sensory response and adaptation at the cellular level, we expressed the calcium indicator protein cameleon in ASH and analyzed intracellular Ca(2+) responses following stimulation with chemical repellents, osmotic shock and nose touch. We found that a variety of noxious stimuli evoked strong responses in ASH including quinine, denatonium, detergents, heavy metals, both hyper- and hypo-osmotic shock and nose touch. We observed that repeated chemical stimulation led to a reversible reduction in the magnitude of the sensory response, indicating that adaptation occurs within the ASH sensory neuron. A key component of ASH adaptation is GPC-1, a G-protein gamma-subunit expressed specifically in chemosensory neurons. We hypothesize that G-protein gamma-subunit heterogeneity provides a mechanism for repellent-specific adaptation, which could facilitate discrimination of a variety of repellents by these polymodal sensory neurons.

Figure 1

(A) Soluble repellent stimulation conditions. An adult hermaphrodite is glued on a 2% agarose pad placed into a perfusion chamber and bathed in extracellular saline (direction of main flow is indicated with a large black arrow at the bottom of the panel). Soluble repellents are applied via a stream of liquid (small black arrows) from a glass needle. The stimulus is applied and removed by changing the position of the needle (white arrow). Anterior is at left and dorsal is up in this and in all following panels. Scale bar, 200 μm. (B) Diagram of the animal's head with one of the two symmetrical ASH neurons highlighted. From the ASH cell body, the dendrite runs anteriorly until the tip of the head ending with a sensory cilium. On the opposite side of the cell body, the axon makes a turn and enters the nerve ring. The blue box shows the approximate position of the recording field. (C) Calcium transients in ASH visualized with cameleon under control of the sra-6 promoter. Individual frames taken before and during application of 10 mM Cu²⁺ are shown. Colors indicate YFP/CFP ratio, where high ratio (red) corresponds to high calcium. Color bar indicates ratio scale. Field of view is 70 × 35 μm (corresponding approximately to the blue box in panel B). (D) Calcium imaging in ASH. ASH imaging during a 3 s application of 10 mM Cu²⁺ (black bar) reveals a change in the YFP/CFP ratio (red line, quantified as % of the baseline level) that can be seen as a reciprocal change in YFP and CFP intensities (yellow and cyan lines). The ratio change does not correspond to motion of the sample (gray line), indicating that it reflects an increase in the calcium transient in ASH. (E) Calcium transients in the ASH cell body (red line) and dendrite (purple line) have similar profiles in response to a 3 s 10 mM Cu²⁺ stimulus (black bar).

Figure 2

(A) Noxious stimuli trigger calcium transients in ASH. Shown are typical calcium transients in ASH evoked by the noxious stimuli 10 mM Cu²⁺, 1 M glycerol, 0.1% SDS, 10 mM quinine and 10 mM denatonium; by nose touch with and without 2.5 mM serotonin, in wild type; by nose touch in the presence of 2.5 mM serotonin in unc-13 (synaptic transmission defective) backgrounds; and by the innocuous stimuli of extracellular saline buffer and 50 mM NaCl. Response is plotted as fractional fluorescence ratio change over baseline. Black bars indicate duration of application of each stimulus. (B) Quantification of ASH responses to selected noxious stimuli in N2 and unc-13 backgrounds, and in N2 in the presence of 2.5 mM serotonin. Small diamonds indicate the amplitude of YFP/CFP ratio change during a single application of a repellent (black bar). Horizontal lines indicate mean±s.e.m. The asterisk indicates a significant decrease in Cu²⁺ response compared to N2 without serotonin (P<0.01). N2 sample sizes (n=number of animals): Cu²⁺ (n=19); glycerol (n=12); SDS (n=15); quinine (n=11); unc-13 sample sizes: Cu²⁺ (n=10); glycerol (n=7); 5-HT sample sizes: Cu²⁺ (n=5); glycerol (n=6).

Figure 3

(A) ASH response is maintained throughout persistent stimuli (less than 10 s). Animals were stimulated for 1 s (thin bar), 3 s (gray bar) or 10 s (black bar) with 10 mM Cu²⁺ and corresponding ratio changes in ASH were recorded (thin line, gray line, black line). (B) ASH adapts to 10 mM Cu²⁺ within 60 s. Recordings are made as in (A), but stimulation lasted 60 s (black bar).

Figure 4

(A, B) The TRPV-related channel OSM-9 is required for repellent-induced calcium transients. (A) Typical responses of N2 (gray line) and osm-9 (black line) animals to 10 mM Cu²⁺ and (B) 1 M glycerol are shown. No significant repellent-induced calcium transients have been observed in osm-9 with any repellent. Black bars indicate duration of application of the stimulus. (C) OSM-9 is also required for calcium transients induced by nose touch. (D) Mutation in the G-protein odr-3 reduces response to all repellents. Individual responses are indicated as diamonds. Horizontal bars represent the mean±s.e.m. of the repellent response as a fraction of the mean wild-type response; n⩾6 animals for each condition; reduction is significant in each case (^*P<0.001). (E) L-type channels are a major source of observed calcium transients. Individual responses to 1 M glycerol (diamonds) and mean±s.e.m. (horizontal bars) are shown for wild-type, egl-19(ad1006) and unc-2(mu74) strains. A significant reduction is observed for the loss-of-function allele of the L-type channel EGL-19 (P<0.001, asterisk), but not for a putative null of the worm non-L-type channel UNC-2. n⩾7 animals for each condition. (F) Wild-type worms give reliable behavioral responses to 1 M glycerol and nose touch (left bars). egl-19(ad1006) worms give significantly reduced behavioral responses (P<0.01, asterisk), consistent with the reduced calcium transient observed in (E). A minimum of 60 trials were used for each condition, except glycerol on N2 for which 30 trials were used. Error bars indicate expected s.e.m. given observed response rate. (G) osm-10 mutants have a specific deficit for response to high-osmotic shock (1 M glycerol) but not Cu²⁺. Individual responses (diamonds) and mean±s.e.m. (horizontal lines) are shown. The asterisk indicates a significant difference from wild type (P<0.001). Sample sizes: n=19, wild-type Cu²⁺; n=5, osm-10 Cu²⁺; n=12, wild-type gly; n=7, osm-10 gly. (H) Nose touch is diminished in osm-10 mutants. Typical wild-type (gray line) and osm-10 (black line) traces are shown.

Figure 5

(A) Repeated repellent application causes adaptation of the behavioral avoidance response. The fraction of animals reversing in response to successive drops of 10 mM Cu²⁺ delivered at 10 s ISI is shown. n=25 animals. (B) Typical ASH traces obtained with successive 10 mM copper stimuli are shown. The second trace (ISI 10 s) was overlaid with the first so that the difference between them is more readily appreciated. Bar indicates the duration of the stimulus. (C) Repeated repellent application causes adaptation of ASH response. Cu²⁺ (10 mM) was delivered to wild-type animals in 3 s puffs with a 10 s ISI. Plotted is the ratio of the second response to the first for individual trials (diamonds) and mean±s.e.m. (horizontal bars). n=17 animals for the first two stimuli and n=6 for stimuli 3–5. All responses after the first are significantly reduced (P<0.01). (D) Adaptation of ASH response is maintained in the synaptic-transmission-defective unc-13 background. Recording conditions are as in (C). n=8 animals for the first two stimuli and n=4 thereafter. All responses after the first are significantly reduced (P<0.01). (E) Prolonged repellent application causes repellent-specific behavioral adaptation. The fraction of animals reversing in response to a drop of 10 mM Cu²⁺ before, 1 min after and 5 min after a 1 min exposure to 10 mM Cu²⁺ is shown. n=24 animals per condition. The fraction of animals reversing in response to a drop of 1 M glycerol before and 1 min after a 1 min exposure to 10 mM Cu²⁺ is also shown. n>10 animals per condition. The asterisk indicates significant reduction (P<0.001). (F) Prolonged repellent application causes repellent-specific adaptation in ASH response. Copper (left columns): individual calcium transients (diamonds) and means±s.e.m. (horizontal bars) before, 1 and 5 min after a 1 min exposure to 10 mM Cu²⁺. n=19 before, n=18 1 min after and n=5 5 min after. Glycerol (right columns): individual calcium transients (diamonds) and means±s.e.m. (horizontal bars) before and 1 min after a 1 min exposure to 10 mM Cu²⁺. n=12 before and n=5 1 min after. Reduction in response to Cu²⁺ after 1 min is significant (P<0.001) compared to before and 5 min after; 5 min after is also reduced compared to before (P<0.01). (G) Typical calcium transients in response to 10 mM Cu²⁺ are shown before (thick line), 1 min after (gray line) and 5 min after (thin line) a 1 min exposure to 10 mM Cu²⁺ (wild-type animals). Bar indicates duration of application of the stimulus.

Figure 6

(A) gpc-1 animals are behaviorally defective in adaptation to prolonged repellent stimuli. Wild-type and mutants in the G-protein γ-subunit gpc-1 were tested with a copper test stimulus 1 min after a 1 min exposure to 10 mM Cu²⁺. gpc-1 animals show significantly less (P<0.01, asterisk) reduction of the avoidance index compared to the wild-type animals. n=24 animals for each condition. (B) gpc-1 animals are defective for ASH adaptation to prolonged stimuli. Wild-type and gpc-1 animals were stimulated before and 1 min after a 1 min exposure to 10 mM Cu²⁺. gpc-1 animals show significantly less reduction compared with the wild type (P<0.01, asterisk; n=10 gpc-1, n=12 N2). Ratio of second response/first response for individual trials (diamonds) and mean±s.e.m. (horizontal bars) are shown. (C) gpc-1 animals are behaviorally defective for adaptation to repeated stimuli. Wild-type and gpc-1 animals were given 10 successive 10 mM Cu²⁺ stimuli with an ISI of 10 s. By the second stimulus, gpc-1 animals show a significant difference from wild type (P<0.01, Fisher's exact test). n=15 animals for gpc-1; n=25 for wild type. (D) gpc-1 animals are defective in ASH adaptation to repetitive repellent stimulation. Ca²⁺ responses in ASH were recorded for two successive stimuli of 3 s with an ISI of 10 s. The ratio of second response to first response for individual trials (diamonds) and mean±s.e.m. (horizontal bars) are shown for the following repellents: 1 M glycerol, 10 mM Cu²⁺, 10 mM quinine, 0.1% SDS (columns, left to right). gpc-1 showed no significant decrease in response magnitude and was significantly different from wild type (^*P<0.001).