Hypoalgesia in mice with a targeted deletion of the tachykinin 1 gene

Abstract

The tachykinin neuropeptides, substance P and substance K, are produced in nociceptive primary sensory neurons and in many brain regions involved in pain signaling. However, the precise role and importance of these neuropeptides in pain responses has been debated. We now show that mice that cannot produce these peptides display no significant pain responses following formalin injection and have an increased pain threshold in the hotplate test. On the other hand, the mutant mice react normally in the tail flick assay and acetic acid-induced writhing tests. These results demonstrate that substance P and/or substance K have essential functions in specific responses to pain.

Figure 1

Targeted mutagenesis of the Tac1 gene. (A), Map of the wild-type genomic locus and cDNA from the mouse strains 129SV/J (Tac1¹²⁹) and the C57BL/6J (Tac1^C57), the targeting construct, and the recombinant locus (Tac1^tTAneo). The DNA probe used for Southern blot analysis and the primers used for PCR analysis are indicated. H, HpaII; N, Nsi I; K, KpnI. (B) Southern blot analysis of a recombinant ES cell after digestion with Nsi I or KpnI. (C) Genotyping of offspring from Tac1^tTAneo/C57 × Tac1^tTAneo/C57 matings by PCR analysis using primer P1, P2, and P3. (D), Genotyping of offspring from Tac1^129/C57 × Tac1^129/C57 matings by PCR amplification with primers P4 and P5, and subsequent digestion with HpaII. Note that HpaII cuts only the PCR fragment derived from the 129 allele, but not the C57BL/6 allele. Homozygous mutant Tac1^{tTAneo/tTAneo} and wild-type Tac1^129/129 are referred to as Tac1^−/− and Tac1^+/+ mice, respectively, throughout the text.

Figure 2

Expression of tachykinin neuropeptides and the substance P receptor NK1 mRNA. (A) Substance P peptide levels (mean ± SEM.) were determined by RIAs in whole brain extracts after HPLC fractionation (Tac1^+/+, n = 4; Tac1^−/−, n = 4). (B) Immunohistochemistry reveals disappearance of SP immunoreactivity in the brain of Tac1−/− mice. The figure shows immunostaining of nerve fibers in the central amygdala (arrow) of the Tac1+/+ brain in brightfield and darkfield. The darkfield image of the Tac1−/− brain shows the complete lack of immunoreactivity in the identical region. CA, central amygdala; IC, internal capsule; OT, optic tract. (C) NK1 receptor mRNA levels were analyzed by in situ hybridization of coronal sections through the striatum and subsequent quantitation by using National Institutes of Health image (cell number: Tac1^+/+, n = 37; Tac1^−/−, n = 58; mRNA levels are shown on an arbitrary scale). The microscopic darkfield image also demonstrates that the area over cells covered by silver grains is larger in the striatum of Tac1−/− animals.

Figure 3

Responses to noxious heat was determined in the (A) tail flick and (B) hotplate tests. tail flick latency (mean ± SEM) were similar between the two genotypes at high- and low-light beam-intensities. After swim stress, analgesia was observed as a significant delay in tail flick latency, which also did not differ between the genotypes. In contrast, pain-response latency in the hotplate test was highly significantly increased in mice with the Tac1^−/− genotype. Responses to noxious chemical stimuli were determined in the (C) acetic acid-induced abdominal constriction assay and in the (D) formalin test. No significant difference between the genotypes was found in the number of abdominal constriction, but Tac1^−/− mice did not show any significant pain responses in the early or late phase of the formalin test. Data were analyzed with the Mann–Whitney U test, except the stress-induced analgesia was analyzed with the Wilcoxon signed rank test. ∗, P < 0.05; ∗∗∗, P < 0.0005. The number at the bottom of each bar indicates the number of animals analyzed.

Figure 4

Substance P producing cells in DRGs and spinal cord. (A) Neurons were marked using a specific neuronal nuclear marker (NeuN) [chemicon (1:2000)], and the number of positive nuclei in several sections of 4–6 DRGs per mouse were counted using one square inch grid. The number of neurons in the DRGs from Tac1^+/+ (n = 3) and Tac1^−/− (n = 3) mice. (B) Double staining of DRGs from Tac1^+/+ and Tac1^−/− mice with substance P and CGRP specific immunosera. Substance P-positive cells typically have small diameters and are labeled green (fluorescein isothiocyanate), while CGRP positive cells are labeled red (CY3). Double-positive cells therefore appear yellow. Arrows point to small-diameter CGRP-positive and substance P-negative cells in the DRG of Tac1^−/− mice. (C) Analysis of substance P and CGRP immunostaining in the spinal cord. The presence of both peptides, mainly in lamina I and II, overlaps in wild-type Tac1^+/+ mice. Note the absence of substance P immunoreactivity in Tac1^−/− animals, while the CGRP-staining is normal. C, central canal; D, dorsal columns; M, marginal zone of the dorsal horn.