Figure 1.
Agonist-treatment enables the selective removal of TRPV1(+) neurons.
(A) Immunolabeling of the neuronal marker UCHL1 and TRPV1 in frozen DRG sections and cultured sensory neurons from rat. TRPV1 is selectively expressed in a subpopulation of sensory neurons. (B) Representative view field showing the automated image analysis to quantify TRPV1 expression in cultured sensory neurons. Green encircled objects represent sensory neurons marked by UCHL1. (C) Distribution of TRPV1 immunofluorescence intensities in frozen DRG sections (red line) and cultured sensory neurons (blue line). The scattered line indicates the threshold used to determine the number of TRPV1(+) neurons. (D) Work flow to remove TRPV1(+) neurons from freshly isolated sensory neurons. (E) Immunoblot showing the reduction of TRPV1 protein following depletion of TRPV1(+) neurons with 10 µM capsaicin for 30–120 min. (F) Time-dependent reduction of TRPV1 mRNA after removal of TRPV1(+) neurons with 10 µM capsaicin or 100 nM RTX determined by qPCR. (G) Dose-dependent reduction of TRPV1 mRNA following depletion of TRPV1(+) neurons with RTX for 30 min determined by qPCR. (H) Agonist-treated neurons were cultured overnight, immunostained for TRPV1, and analyzed by quantitative microscopy. Both agonists effectively reduced the number of TRPV1(+) neurons. Scattered lines indicate the threshold used to determine the number of TRPV1(+) neurons. (I) Quantification of TRPV1(+) neurons following treatment with 10 µM Cap or 100 nm RTX, respectively (n = 3, p<0.01, one-way ANOVA with Bonferroni's multiple comparisons test).
Figure 2.
Transcriptome analysis reveals novel pain targets enriched in TRPV1(+) neurons.
(A) qPCR quantification of TRPV1 mRNA levels in samples hybridized on microarrays (n = 4, p<0.001, One-way ANOVA). (B) Relative expression levels of TRPV1 mRNA determined by microarray hybridizations (n = 4, p<0.001, One-way ANOVA with Bonferroni's multiple comparisons test). (C) Correlation matrix of the expression level of all genes (normalized fluorescence intensities) detected using micorarrays. Clustering of overall gene expression is visible between RTX and capsaicin treated samples. (D) qPCR quantification of TRPV1 mRNA levels in samples used for RNA-Seq (n = 3, p<0.001, paired two-tailed t-test). (E) Relative expression levels of TRPV1 mRNA determined by RNA-Seq (n = 3, p<0.001, One-way ANOVA). (F) Overview of the number of target transcripts identified by microarray hybridizations and RNA-Seq. Raw fluorescence intensities of microarrays were background corrected, log2 transformed, normalized, and filtered for expressed genes (Illumina detection p-value <0.01 in at least one of the samples, see Material and Methods sections for details). Sequencing was performed using a 5500xl SOLiD System resulting in 664 million reads in total (see Material and Methods sections for details). The reads were filtered to remove ribosomal RNA, tRNAs, and vector sequences. The remaining reads mapped to 16590 genes of the reference genome (rn5). Read counts were transformed to RPKM values (Reads per kilo base per million), normalized, and filtered to remove weakly expressed transcripts (RPKM>0.1). P-values of differentially expressed genes identified with both methods were adjusted for multiple testing with Benjamini and Hochberg's method, adjusted p-values <0.05 were considered significant.
Figure 3.
Validation of transcriptome data.
(A) Correlation of fold-changes obtained by microarray hybridizations and RNA-Seq (Spearmans ρ = 0.66, p<0.0001). (B) qPCR validation of 14 transcripts identified as differentially expressed by microarray hybridizations or RNA-Seq. The depletion of TRPV1(+) neurons was performed with capsaicin (1 and 10 µM) for 30 min. The reduction of all target transcripts is dose-dependent.
Table 1.
Top 50 transcripts found with higher expression levels in TRPV1(+) neurons by microarrays and/or RNA-Seq (adj. p<0.05, fold-change>1.5, RPKM>0.5).
Figure 4.
Validation of the transcriptome data by single cell based quantitative High Content Screening (HCS) microscopy focusing on selected signaling-relevant proteins.
(A) Triple staining of the neuronal marker UCHL1 and two different TRPV1 antibodies derived from rabbit and goat, respectively, to facilitate the analysis of various targets. The staining intensities obtained with both TRPV1 antibodies correlated significantly (Spearmans ρ = 0.96, p<2.2e-16). (B-E, G) Co-labeling of TRPV1 and CART (B), Nos1 (C), KChIP1 (D), KChIP2 (E), and CaMKIIα (G). Plots of respective controls are shown in S1 Fig. (F) Average fluorescence intensities of TRPV1 and the indicated targets in TRPV1-negative (grey) and -positive (black) neurons. Signal intensities of all analyzed targets were significantly higher within the TRPV1(+) population (n = 3 with>3000 analyzed neurons per experiment, paired two-tailed t-tests).
Figure 5.
PGD2 is a paracrine mediator synthesized in myelinated large-diameter neurons that acts on TRPV1(+) neurons.
(A) Dose-dependent induction of RII phosphorylation in sensory neurons after 1 min stimulation with PGD2 (EC50 = 377 nM, n = 3,>2000 neurons/condition; one-way ANOVA with Bonferroni's multiple comparisons test). (B) PGD2 did not induce pRII in non-neuronal cells of the same cultures shown in A. (C) Time course of RII phosphorylation indicating long-lasting effects of PGD2 (10 µM) on sensory neurons. (D) Stimulation with PGD2 also results in phosphorylation of the ERK1/2 measured in the same cultures shown in D. (E) Representative experiment demonstrating that induction of RII phosphorylation is enhanced in TRPV1(+) neurons (total of 3664 neurons). Plots of respective controls are shown in S2 Fig. (F) Fold changes of pRII intensities in TRPV1(−) (grey bars) and TRPV1(+) (black bars) neurons after 1 min stimulation with 10 µM PGD2 (n = 3,>2000 neurons/condition, one-way ANOVA with Bonferroni's multiple comparisons test). (G) Co-labeling of TRPV1 and PTGDS revealing that PTGDS is expressed in neurons lacking TRPV1 (total of 9951 neurons, also refer to S2 Fig. for control plots). (H) Co-labeling of NF200 and PTGDS showing that PTGDS(+) neurons express NF200 (total of 12966 neurons, also refer to S2 Fig. for control plots).(I) Size distribution of PTGDS(+) (green), NF200(+) (red), and all sensory neurons (black) indicating that PTGDS(+) neurons are larger than other neurons. (J) Suggested pathway of interneuronal communication between subgroups of sensory neurons. Large-diameter mechanosensitive neurons express PTGDS resulting in the production of PGD2, which activates DP1 receptors present on nociceptive neurons expressing TRPV1.