Processing of pain and itch information by modality-specific neurons within the anterior cingulate cortex in mice

Abstract

Pain and itch are aversive sensations with distinct qualities, processed in overlapping pathways and brain regions, including the anterior cingulate cortex (ACC), which is critical for their affective dimensions. However, the cellular mechanisms underlying their processing in the ACC remain unclear. Here, we identify modality-specific neuronal populations in layer II/III of the ACC in mice involved in pain and itch processing. Using a synapse labeling tool, we show that pain- and itch-related neurons selectively receive synaptic inputs from mediodorsal thalamic neurons activated by pain and itch stimuli, respectively. Chemogenetic inhibition of these neurons reduced pruriception or nociception without affecting the opposite modality. Conversely, activation of these neurons did not enhance stimulus-specific responses but commonly increased freezing-like behavior. These findings reveal that the processing of itch and pain information in the ACC involves activity-dependent and modality-specific neuronal populations, and that pain and itch are processed by functionally distinct ACC neuronal subsets.

Fig. 1. ACC neurons are required for pain and itch responses without showing somatotopy.

a Representative image showing microinjection site in the ACC. Scale bar, 1 mm. b Scratching behavior after histamine injection (n = 4 mice in sal, n = 4 mice in CNQX/TTX; two-tailed unpaired t-test, t₆ = 2.704, *p = 0.0354). c Nociceptive behavior after formalin injection (n = 4 mice in sal, n = 4 mice in CNQX/TTX; two-tailed unpaired t-test, t₆ = 3.338, *p = 0.0157). d, e Antinociceptive and antipruriceptive responses induced by histamine injection in the nape of the neck (d n = 6 mice in measuring antinociceptive and antipruriceptive responses; two-tailed unpaired t-test, t₁₀ = 6.565, ***p < 0.0001) or formalin injection in the left hind sole (e n = 6 mice in measuring antinociceptive and antipruriceptive responses; two-tailed unpaired t-test, t₁₀ = 4.610, ***p = 0.0010). f The distribution of fos (+) cells in the whole ACC. Six mice per group were injected with histamine in the nape of the neck or formalin in the left hind sole. The distribution of fos (+) cells reconstructed in heatmaps. Scale bars, 200 μm. g Representative double immunolabeling images of the ACC showing CaMKII and Fos (left), and GABA and Fos (right) in histamine-injected mice. Scale bars, 100 μm. h Percentage of cells co-stained with CaMKII or GABA among total fos (+) cells (n = 7 mice injected with His; two-tailed unpaired t-test, t₁₂ = 20.32, ***p < 0.0001). i Representative double immunolabeling images of the ACC showing CaMKII and Fos (left), and GABA and Fos (right) in formalin-injected mice. Scale bars, 100 μm. j Percentage of cells co-stained with CaMKII or GABA among total fos (+) cells (n = 8 mice injected with For; two-tailed unpaired t-test, t₁₄ = 20.20, ***p < 0.0001). sal Saline, His Histamine, For Formalin. Data presented as mean ± SEM. Source data are provided as a Source Data file.

Fig. 2. Labeling of neurons activated by pruritogenic and algogenic stimuli over a time interval.

a Experimental scheme. b, c Representative immunolabeling images of the ACC in mice that waited 6 h or 3 days between histamine and formalin stimuli (b) or between histamine and histamine stimuli (c). White arrowheads indicate H2BGFP and fos double (+) cells. Scale bars, 50 μm. d The proportion of neurons activated by both itch- and pain-inducing stimuli, given either with a 6-h or a 3-day interval (n = 9 mice in 6-h interval, n = 8 mice in 3-day interval; two-tailed paired t-test; 6-h interval, t₈ = 4.301, **p = 0.0026; 3-day interval, t₇ = 1.247, p = 0.2524). e The proportion of neurons activated by both consecutive itch-inducing stimuli, given with a 6-h or 3-day interval (n = 7 mice in 6-h interval, n = 7 mice in 3-day interval; two-tailed unpaired t-test; 6-h interval, t₆ = 4.008, **p = 0.0071; 3-day interval, t₆ = 5.860, **p = 0.0011). f Levels of reactivation index (overlap/chance) were above chance except for Itch → Pain at the 3-day interval (n = 9 mice in Itch → Pain 6-h interval, n = 8 mice in Itch → Pain 3-day interval, n = 7 mice in Itch → Itch 6-h interval, n = 7 mice in Itch → Itch 3-day interval; one sample t-test, chance = 1; Itch → Pain at the 6-h interval, t₈ = 4.568, ## p = 0.0018; Itch → Pain at the 3-day interval, t₇ = 1.207, p = 0.2666; Itch → Itch at the 6-h interval, t₆ = 4.173, ## p = 0.0059; Itch → Itch at the 3-day interval, t₆ = 3.440, # p = 0.0138). The overlap/chance at the 3-day interval was higher in the Itch → Itch group compared to Itch → Pain group (two-tailed unpaired t-test; t₁₃ = 2.356, *p = 0.0349). The dotted line indicates chance level. Data presented as mean ± SEM. Source data are provided as a Source Data file.

Fig. 3. In vivo imaging of pain and itch-responsive populations.

a Schematic for the miniscope recordings. b Representative images captured via miniscope before (upper) and after (lower) stimulus injection. Brown arrowhead indicates neurons activated only before stimulus. Cyan arrowhead indicates neuron activated both before and after the stimulus. The blue arrowhead indicates neurons activated only after the stimulus. Calcium imaging traces of the neurons marked with arrowheads are shown in matching colors on the right. Scale bar, 100 μm. c Proportion of neurons activated only before, both before and after, and only after histamine or formalin injection. d Representative rendering images of neurons activated only after stimulus. Neurons activated by itch-inducing or pain-inducing stimuli are marked in green or red, respectively. Neurons activated by both stimuli at 6-h or 3-day intervals are marked in orange. Black arrowheads indicate neurons activated by both stimuli at 6-h or 3-day intervals. Scale bar, 100 μm. e Percentage of overlapping neurons responding to histamine and formalin given at 6-h or 3-day intervals (n = 7 mice in 6-h interval, n = 7 mice in 3-day intervals; two-tailed unpaired t-test, t₁₂ = 2.219, *p = 0.0465). f Representative images of neurons activated only after stimulus injection. Neurons activated by the first or second itch-inducing stimulus are marked with green or red, respectively. Neurons activated by both stimuli at 6-h or 3-day intervals are marked in orange. Black arrowheads indicate neurons activated by both stimuli at 6-h or 3-day intervals. Scale bar, 100 μm. g Percentage of overlapping neurons responding to two consecutive histamine injections given at 6-h or 3-day intervals (n = 5 mice in 6-h intervals, n = 5 mice in 3-day intervals; two-tailed paired t-test, t₄ = 1.267, p = 0.2738). h Machine learning classification of neurons activated only after stimulation (histamine or formalin) (above: overview, below: confusion matrix). The top-left or bottom-right quadrants: the percentage (number) of correctly classified neurons. The top-right or bottom-left quadrants: the percentage (number) of incorrectly classified neurons. Data presented as mean ± SEM. Source data are provided as a Source Data file.

Fig. 4. Paired synaptic connections between modality-specific neurons in the ACC and MD.

a Illustration of the viral injection site in a TRAP2 mouse. b Schematic of the AAV constructs and their combinations injected in the MD and ACC. c Experimental scheme. d Representative images showing virus expression in the ACC among 4 mice used in dual e-GRASP experiments. Expression of iRFP670 indicates itch-activated neurons (middle), while expression of mScarlet-I indicates pain-activated neurons (right). A color-merged image marking the ACC area with a dotted line appears in the left panel. Scale bar, 100 μm. Inset images showed a magnified itch-activated (middle upper) or pain-activated (right upper) neuron, respectively. Scale bar, 3 μm. e Representative images showing dual eGRASP in itch-activated and pain-activated ACC dendrites with cyan eGRASP puncta (synapses originating from itch-activated MD neurons) and yellow eGRASP puncta (synapses originating from pain-activated MD neurons). A total of 29 z-stack images were acquired from ACC slices of 4 mice. Above: color-merged raw image. Middle: raw image with only eGRASP signals extracted. Below: 3D reconstruction of dendrites and spines overlaid with eGRASP signals. Cyan and yellow arrowheads indicate cyan and yellow eGRASP puncta, respectively. Scale bar, 4 μm. f The percentage of eGRASP puncta located on iRFP- and mScarlet-I-expressing ACC dendrites. g The percentage of iRFP- and mScarlet-I-expressing dendrites located on cyan or yellow eGRASP signals. h The relative ratio of these GRASP signals indicates that synaptic inputs to ACC neurons are predetermined by their presynaptic MD neurons (n = 21 images of 4 mice in C/Y ratio, n = 18 images of 4 mice in Y/C ratio; two-tailed Mann-Whitney U test, ****p < 0.0001). C Cyan eGRASP puncta, Y Yellow eGRASP puncta. Data presented as mean ± SEM. Source data are provided as a Source Data file.

Fig. 5. Itch- or pain-specific neurons are required for scratching or nocifensive behavior, respectively.

a Experimental scheme. b Representative images of emGFP and n-mCherry expression among the ACC of total of 68 mice. Scale bar, 200 μm. c, d Scratching behavior (c) and the proportion of hM4Di-expressing neurons (d) (n = 9 mice in F → H sal, n = 10 mice in F → H CNO. c: two-tailed unpaired t-test, t₁₇ = 0.03335, p = 0.9738. d two-tailed unpaired t-test, t₁₇ = 1.491, p = 0.1544). e, f Scratching behavior (e) and the proportion of hM4Di-expressing neurons (f) (n = 8 mice in H → H sal, n = 11 mice in H → H CNO. e: two-tailed unpaired t-test, t₁₇ = 2.160, *p = 0.0454. f two-tailed unpaired t-test, t₁₇ = 0.7668, p = 0.4537). g, h Nocifensive responses (g) and the proportion of hM4Di-expressing neurons (h) (n = 7 mice in H → F sal, n = 7 mice in H → F CNO. g two-way repeated measures ANOVA: effect of time, F_(1,12) = 81.58, p < 0.0001; effect of CNO, F_{(1, 12)} = 0.1758, p = 0.6824; interaction effect, F_{(1, 12)} = 0.01248, p = 0.9129. h two-tailed unpaired t-test, t₁₂ = 0.3180 1, p = 0.7559). i, j Nocifensive responses (i) the proportion of hM4Di-expressing neurons (j) (n = 8 mice in F → F sal, n = 8 mice in F → F CNO. i two-way repeated measures ANOVA followed by Bonferroni post-test: effect of time, F_(1,14) = 114.6, p < 0.0001; effect of CNO, F_{(1, 14)} = 11.42, p = 0.0045; interaction effect, F_{(1, 14)} = 15.16, p = 0.0016; post-test, ***p < 0.001. j two-tailed unpaired t-test, t₁₄ = 1.148, p = 0.2703). k, l Schematic diagram of an itch (k) and pain (l) on suppression of modality-specific ACC neurons. sal Saline, F Formalin. H Histamine. Data presented as mean ± SEM. Source data are provided as a Source Data file.

Fig. 6. Activation of itch- or pain-specific neurons contribute to inducing pruritogen- and algogen-induced anxious states.

a Experimental scheme. b Representative images showing mCherry expression in the ACC of a total of 39 mice. Scale bar, 100 μm. c–e Scratching bouts (c, d) and freezing-like behavior (e) when neurons previously activated by histamine were activated using CNO (n = 9 mice in sal, n = 12 mice in CNO. c two-way repeated measures ANOVA: effect of time, F_{(5, 95)} = 2.022, p = 0.0825; effect of CNO, F_{(1, 95)} = 1.834, p = 0.1915; interaction effect, F_{(5, 95)} = 1.277, p = 0.2802. d two-tailed unpaired t-test, t₁₉ = 1.354, p = 0.1915. e two-tailed unpaired t-test, t₁₉ = 1.223, p = 0.2361). f–h Nocifensive responses (f, g) and freezing-like behavior (h) when neurons previously activated by formalin were activated using CNO (n = 10 mice in sal, n = 8 mice in CNO. f two-way repeated measures ANOVA: effect of time, F_{(11, 176)} = 1.188, p = 0.2986; effect of CNO, F_{(1, 176)} = 2.029, p = 0.1736; interaction effect, F_{(11, 176)} = 1.533, p = 0.1231. g: two-tailed unpaired t-test, t₁₆ = 1.424, p = 0.1736. h: two-tailed unpaired t-test, t₁₆ = 2.384, *p = 0.0298). i–k Weak histamine-induced scratching bouts (i, j) and freezing-like behavior (k) when neurons previously activated by histamine were activated using CNO (n = 9 mice in sal, n = 12 mice in CNO. i: two-way repeated measures ANOVA followed by Bonferroni post-test: effect of time, F_{(5, 95)} = 11.62, p < 0.0001; effect of CNO, F_{(1, 95)} = 4.592, p = 0.0453; interaction effect, F_{(5, 95)} = 4.093, p = 0.0021; post-test, ***p < 0.001. j: two-tailed unpaired t-test, t₁₉ = 2.143, *p = 0.0453. k: two-tailed unpaired t-test, t₁₉ = 3.368, **p = 0.0032). l–n, Weak formalin-induced nocifensive responses (l, m) freezing-like behavior (n) and when neurons previously activated by formalin were activated using CNO (n = 10 mice in sal, n = 8 mice in CNO. l two-way repeated measures ANOVA: effect of time, F_{(11, 176)} = 1.884, p = 0.0442; effect of CNO, F_{(1, 176)} = 0.006581, p = 0.9364; interaction effect, F_{(11, 176)} = 0.5102, p = 0.8948. m two-tailed unpaired t-test, t₁₆ = 0.08112, p = 0.9364. n two-tailed unpaired t-test; two-tailed unpaired t-test, t₁₆ = 3.739, **p = 0.0018). sal Saline, H Histamine, F Formalin. Data presented as mean ± SEM. Source data are provided as a Source Data file.