Silent cold-sensing neurons contribute to cold allodynia in neuropathic pain

Abstract Patients with neuropathic pain often experience innocuous cooling as excruciating pain. The cell and molecular basis of this cold allodynia is little understood. We used in vivo calcium imaging of sensory ganglia to investigate how the activity of peripheral cold-sensing neurons was altered in three mouse models of neuropathic pain: oxaliplatin-induced neuropathy, partial sciatic nerve ligation, and ciguatera poisoning. In control mice, cold-sensing neurons were few in number and small in size. In neuropathic animals with cold allodynia, a set of normally silent large diameter neurons became sensitive to cooling. Many of these silent cold-sensing neurons responded to noxious mechanical stimuli and expressed the nociceptor markers Nav1.8 and CGRPα. Ablating neurons expressing Nav1.8 resulted in diminished cold allodynia. The silent cold-sensing neurons could also be activated by cooling in control mice through blockade of Kv1 voltage-gated potassium channels. Thus, silent cold-sensing neurons are unmasked in diverse neuropathic pain states and cold allodynia results from peripheral sensitization caused by altered nociceptor excitability.

contralateral hind-paw placed on an adjoining thermoneutral surface. The animal was held vertically such that its forepaws did not contact any surface. The time until ipsilateral hindpaw withdrawal was then measured. This was repeated two to three times and a trial averaged latency to withdrawal was obtained.

Acetone Test:
The acetone evaporation test measures the nociceptive behaviours triggered by evaporative cooling of the hindpaw (Yoon et al., 1994). Mice were habituated for 1 hour in enclosures with a wire mesh floor. Using a home-made syringe, a 50 µl drop of acetone was applied to the ventral side of the ipsilateral hindpaw. The cumulative time where the ipsilateral hindpaw was engaged in nociceptive and nocifensive behaviours (lifting, shaking, licking, guarding, biting) over the ensuing 60 s was then counted. An average of two to three trials was obtained, with at least 10 minutes between trials 1.2 In vitro electrophysiology

DRG neuronal culture
Adult mice were killed by inhalation of a rising CO2 concentration followed by cervical dislocation to confirm death. Dorsal root ganglia (DRG) were dissected from the entire length of the spinal column and then digested in a pre-equilibrated enzyme mix for 45 minutes (37 ˚C, 5% CO2). The enzyme mix consisted of Hanks' balanced salt solution containing collagenase (type XI; 5 mg/ml), dispase (10 mg/ml), HEPES (5 mM) and glucose (10 mM). DRGs were then gently centrifuged for 5 minutes at 300 revolutions per minute, the supernatant was discarded and replaced with warmed Dulbecco's modified Eagle's medium (DMEM), supplemented with L-glutamine (1%), glucose (4.5 g/litre), sodium pyruvate (110 mg/litre) and 10% fetal bovine serum (FBS). Next, DRGs were mechanically triturated with three firepolished glass Pasteur pipettes of gradually decreasing inner diameter. Dissociated cells were then centrifuged again at 300 revolutions per minute, the supernatant was discarded and cells were re-suspended in the required volume of DMEM supplemented with FBS and nerve growth factor (50 ng/ml). Finally, cells were plated onto 12 mm glass coverslips coated with poly-Llysine (1 mg/ml) and laminin (1 mg/ml). Cells were incubated at 37 ˚C in 5% CO2 and recordings were performed at room temperature (18-21 ˚C) between 24 -72 hours after dissociation.

Whole cell patch-clamp
Functional deletion of NaV1.8 was assessed by pharmacological isolation of TTX-resistant currents. Patch pipettes (tip resistance of 3-5 MΩ) were filled with intracellular solution containing: 140 mM CsF, 1 mM EGTA, 5 mM NaCl, 10 mM HEPES. Neurons were perfused with extracellular solution containing in: 70 mM NaCl, 70 mM Choline-Cl, 3 mM KCl, 1 mM MgCl2, 20 mM TEA-Cl, 0.1 mM CdCl2, and 10 mM Glucose. 5nM TTX was included in the extracellular solution to isolate TTX-resistant currents. Whole-cell recordings were obtained using an Axopatch 200B amplifier, filtered at 10 kHz and digitized at 50 kHz via a Digidata 1322A (Axon Instruments). tdTomato-expressing neurons from heterozygous and homozygous NaV1.8-Cre mice were voltage-clamped at -70 mV. Series resistance compensation was at least 60%. To measure the voltage-dependence of sodium channel activation, the holding command was dropped to -120 mV to de-inactivate all sodium channels and then a step-protocol from -80 to 20 mV was applied, in increments of 5 mV, to activate sodium channels. Supplementary Fig. 1. Behavioural and functional effects of oxaliplatin treatment.

Supplementary Figures
(A) Comparison of the behavioural effects of oxaliplatin on different sensory modalities (cold, mechanical and heat) in male and female mice. For vehicle, n=5 males and n=3 females. For oxaliplatin, n=5 males and n=4 females. Means were before and after treatment were compared by 2-way ANOVA followed by post-hoc Sidak's test. Error bars denote 95% confidence interval.
(B) Cumulative probability plots of cross-sectional areas for cells responding to each stimulus modality, compared using Kolmogorov-Smirnov test.
(C) Effect of oxaliplatin on acetone-evoked pain behaviour. Means before and after treatment were compared using repeated measures 2-way ANOVA with post-hoc Sidak's test. n=7.
(D) Violin plots showing the peak responses evoked by cold stimuli in the vehicle group and separately in the basal and silent cold-sensing neurons from the oxaliplatin group. Ice water: n=51 for vehicle, n=40 for basal, and n=41 for silent. Acetone: n=58 for vehicle, n=57 for basal, and n=88 for silent. Medians were compared by Kruskall-Wallis test followed by Dunn's multiple comparison's test.
(E) Bar plots showing the proportion of all responding neurons responding to heat or mechanical stimuli, compared using χ 2 test. Error bars denote 95% confidence interval.
(F) Violin plots showing peak response evoked by each stimulus modality, compared using Mann-Whitney test.
(G) Proportion of mechanically-sensitive neurons also responding to noxious heat, compared using χ 2 test. Error bars denote 95% confidence interval.

Supplementary Fig. 2. Silent cold-sensing neurons are not low-threshold mechanoreceptors.
(A) Representative traces of DRG neuron calcium signals in an oxaliplatin-treated mouse in response to cold and 2g Von Frey stimuli.
(B) Bar plot showing the percentage of basal and silent cold-sensing neurons in oxaliplatin-treated mice that show a response to repeated 2 g Von Frey stimulation. n=421 cold-sensing neurons from 19 oxaliplatin-treated animals (10♂ & 9♀).
(C) Example traces of DRG neuron calcium signals in a naive mouse in response to cold and brush stimuli. Example traces of a low-threshold mechanoreceptor (top) and a cold-sensing neurons (middle) are shown. A rare neuron responding to both brushing and cooling is also shown (bottom).
(D) Bar plot showing the percentage of basal and silent cold-sensing neurons in oxaliplatin-treated mice that show any response to a wide range of low-threshold mechanical stimuli (cotton swap/brush was applied to glabrous skin and to hairy skin with and against grain). n=67 cold-sensing neurons from 3 oxaliplatin-treated animals (2♂ & 1♀).
(E) Bar plot showing the percentage of basal and silent cold-sensing neurons in mice with partial nerve ligation that show a response to brushing of the paw. n=105 cold-sensing neurons from 6 PNL-operated animals (3♂ & 3♀).
(F) Bar plot showing the percentage of basal and silent cold-sensing neurons in P-CTX-2-treated mice that show a response to brush either before or after treatment. n=227 cold-sensing neurons from 10 P-CTX-2-injected mice (4♂ & 6♀).
(B) Comparison of the behavioural effects of partial nerve ligation (PNL) on different sensory modalities (cold, mechanical and heat) in male and female mice. For PNL, n=3 males and n=3 females. Means were before and after treatment were compared by 2-way ANOVA followed by post-hoc Sidak's test, or using a two-tailed unpaired t test. Error bars denote standard error of the mean.
(C) Cumulative probability plots of cross-sectional areas for cells responding to each stimulus modality, compared using Kolmogorov-Smirnov test.
(D) Violin plots showing the peak responses evoked by cold stimuli in the sham group and separately in the basal and silent cold-sensing neurons for the PNL group. Ice water: n=64 for sham, n=46 for basal, and n=25 for silent. Acetone: n=95 for sham, n=42 for basal, and n=31 for silent. Medians were compared by Kruskall-Wallis test followed by Dunn's multiple comparison's test.
(F) Bar plots showing the proportion of all responding neurons responding to heat or mechanical stimuli, compared using χ 2 test. Error bars denote 95% confidence interval.
(G) Violin plots showing peak response evoked by each stimulus modality, compared using Mann-Whitney test.
(H) Proportion of mechanically-sensitive neurons also responding to noxious heat, compared using χ 2 test.
(B) Violin plots of cross-sectional areas for cells responding to each stimulus modality, compared using Kruskall-Wallis test followed by Dunn's multiple comparisons test.
(C) Line plots showing the median response magnitude of basally cold-sensitive neurons before and after treatment, compared using Kruskall-Wallis test followed by Dunn's multiple comparisons test.
(D) Violin plots showing the response magnitude of all silent cold-sensing neurons unmasked by P-CTX-2 (n=127 for ice-water, and n=60 for acetone) compared to all basally-active neurons recorded from naïve mice (n=105 for both). Medians were compared using Mann-Whitney test.
(E) Box plots showing the median response magnitude of all mechanical and heat-responsive neurons before and after treatment.
(F) Proportion of mechanically-sensitive neurons also responding to noxious heat, before and after treatment, compared using χ 2 test. Error bars denote 95% confidence intervals.
(G) Quantification of the proportion of neurons responding acetone that were also sensitive to either mechanical (i.) or heat (ii.) before and after treatment. (iii.) Comparison of the proportion of silent cold-sensing neurons activated by acetone that were responsive to other modalities before and after the induction of cold-sensitivity by P-CTX-2. n=60. The proportion of polymodal neurons was compared using χ 2 test, and error bars denote 95% confidence intervals.