The Journal of Neuroscience, April 1995, 15(4): 2733-2747 Complementary Distribution of Receptors for Neurotensin and NPY in Small Neurons in Rat Lumbar DRGs and Regulation of the Receptors and Peptides after Peripheral Axotomy X. Zhang, Z.-Q. Xu, L. Bao, A. Dagerlind, and T. H&felt Department of Neuroscience, Karolinska Institute, 171 77 Stockholm, Sweden Neurotensin (NT) has been reported to have antinocicep- minals is the superficial dorsal horn of the rat spinal cord (Uhl tive effects at the spinal level. In situ hybridization, electro- et al., 1979; Gibson et al., 1981; Hunt et al., 1981; Ninkovic et physiology, immunohistochemistry, and electronmicros- al., 1981; Seybold and Elde, 1982; Seybold and Maley, 1984; copy were used to investigate the distribution of NT Todd et al., 1992; Proudlock et al., 1993). Since NT has not receptors, possible effects of NT on primary sensory neu- been reported to be present in normal dorsal root ganglion rons, and the effect of nerve injury on the expression of NT (DRG) neurons,it has been assumedthat the densenetwork of receptors and NT. NT receptor (Pi) mRNA was observed in NT-IR fibers in the dorsal horn originatesfrom the local neurons. more than 25% of the small dorsal root ganglion (DRG) However, low amounts of NT-like immunoreactivity (LI) may neurons, which lacked neuropeptide Y NPY-R mRNA and be present in a few primary afferent fibers in the dorsal horn essentially other neuropeptide mRNAs. Intracellular re- under normal circumstances(Zhang et al., 1993b). Behavioral, cording using voltage-clamp mode showed that NT evokes neurophysiological,and pharmacologicalstudies have indicated an outward current in NPY-insensitive small neurons, and functional roles for NT in the dorsal horn. Thus, NT has been NPY an outward current in NT-insensitive small neurons. shown to have antinociceptive effects (Clineschmidt and Mc- Both peptides lacked effect on several small DRG neurons. Guffin, 1979; Yaksh et al., 1982; Spampinatoet al., 1988). In- In the superficial dorsal horn NT immunoreactive (IR) ter- trathecal administration of NT results in an increasein the no- minals directly contacted primary afferent terminals with- ciceptive threshold as measuredin the hot plate test and an out synaptic specializations. This new category (> 25%) of increase in the latency in the acetic acid-induced writhing re- the small DRG neurons expressing NT-R mRNA was com- sponse(Yaksh et al., 1982; Spampinatoet al., 1988). Morever, plementary to the around 60% of small neurons expressing after peripheral axotomy, a marked decreasein NT-L1 has been NPY-R mRNA (and also substance P and calcitonin gene- describedin the dorsal horn of the monkey spinal cord (Zhang related peptide mRNAs) and to the rest exhibiting somato- et al., 1993a), suggestingthat NT may be involved in sensory statin mRNA expression. The electrophysiological results mechanismsthat may causechronic pain. In order to better un- support this classification, showing that NT and NPY have derstand the role of NT at the spinal level and its possible in- inhibitory effects on separate subpopulations of small DRG volvement in syndromesaccompanying peripheral nerve injury, neurons. After sciatic nerve transection, a marked de- we have analyzed the distribution of NT receptor mRNA, the crease was observed in (1) the number of NT-R mRNA-pos- effects of NT on sensoryneurons, and the expressionof NT and itive neurons in DRGs, (2) NT mRNA-positive neurons in its receptors after axotomy. the dorsal horn, and (3) NT-IR cell bodies and fibers in lam- NT binding sites have been observed in the brain (see Uhl, inae I-II. Thus, axotomy causes downregulation of several 1990) and also in the superficial dorsal horn of the spinal cord NT systems at the spinal level, suggesting that the possible (Ninkovic et al., 1981; Young and Kuhar, 1981). No change in effects of NT on primary sensory neurons is attenuated af- density or distribution of NT binding siteswas found after dorsal ter peripheral axotomy. root section (Ninkovic et al., 198l), suggestingthat NT receptors [Key words: coexistence, calcitonin gene-related pep- (NT-Rs) are not present on the terminals of primary afferents. tide, pain, plasticity, somatostatin] It has recently been shown that the NT-R is a member of the guanine nucleotide binding, G-protein-coupled receptor super- Neurotensin (NT), a tridecapeptide (Carraway and Leeman, family (Tanaka et al., 1990), and the cloning of its cDNA se- 1973, 1975), is widely distributed in the nervous system (Ko- quence (Tanakaet al., 1990) has allowed analysisof distribution bayashi et al., 1977; Emsonet al., 1985), where it is an important of NT-R mRNA in the brain using in situ hybridization (Elde et regulatory molecule (see Kitabgi and Nemeroff, 1992). One re- al., 1990). gion with many immunoreactive (IR) cell bodies and nerve ter- In the present study, in situ hybridization was used to analyze the localization of NT-R mRNA in DRG neuronsof normal rats and after peripheral axotomy. Furthermore, it was attempted to Received Apr. 12, 1994; revised Sept. 29, 1994; accepted Oct. 3, 1994. This work was supported by the Swedish MRC (04X-2887), the Bank of define the relation of DRG neuronsexpressing NT-R mRNA to Sweden Tercentenary Foundation, Marianne and Marcus Wallenbergs Stiftelse those recently described to contain neuropeptide tyrosine and Gustav V:s and Drottning Victorias Stiftelse. We thank Dr. l? Frey, Sandoz (NPY)-R mRNA (Jazin et al., 1993; Zhang et al., 1994a) as well Research Institute, Bern, Switzerland, for generous supply of NT antiserum. Correspondence should be addressed to T. HGkfelt at the above address. as certain neuropeptidemRNAs. The effect of NT and NPY on Copyright 0 1995 Society for Neuroscience 0270-6474/95/152733-15$05.00/O DRG neuronswas monitored with intracellular recordingsusing 2734 Zhang et al. Neurotensin Receptors in Sensory Neurons voltage-clamp mode. Finally, NT-L1 and NT mRNA were stud- for 3 min, fixed in Kodak 3000A&B for 6 min, and rinsed for 30 min ied in the superficial dorsal horn of normal rats and after pe- in running water. Developed slides were mounted with glycerol and coverslipped for analysis in a Nikon Microphot-FX microscope ripheral axotomy. equipped with a dark-field condenser or stained with Toluidine blue, The present results show that NT-R mRNA-positive neurons mounted with Entellan (Merck, Darmstadt, Germany) and a coverslip represent a unique subpopulation of small DRG neurons. To- for viewing under bright-field. gether with two other subpopulationscontaining NPY-R and so- For control hybridizations were carried out with an excess of cold probe (loo-fold) together with the labeled probe. matostatin (SOM), respectively, they account for virtually all Qunnti$cation. To determine the percentage of labeled neurons, small DRG neurons. In agreement, small DRG neurons were counts were done on sections stained with Toluidine blue. Four to eight sensitive either to NT or NPY or insensitive to both of these sections (14 pm thick) of L5 DRGs from three control animals and peptides. In the dorsal horn, NT-positive boutons of local neu- three injured animals of each time point were processed. The sections rons had direct, nonsynaptic contacts with primary afferent ter- were examined under bright-field using a 20X objective lens. We di- vided the size of DRG neuron profiles into small, medium-sized, and minals. Finally, axotomy reduced NT-R mRNA levels in DRG large ones. Small neuron profiles were 22-32 pm in diameter with neuronsand NT-L1 and NT mRNA in the superficialdorsal horn, nucleus and dark cytoplasm, medium-sized ones were 32-50 pm in suggestingthat peripheral nerve injury causesa general down- diameter (longest diameter) with a nucleus and dark cytoplasm, and regulation of several NT systemsat the spinal cord level. large ones had a diameter larger than 50 km with a nucleus or a di- ameter of 40-50 p.m without nucleus and with light cytoplasm. Neurons Materials and Methods with three times more grains than mean background grain densities were counted. Mean background grain densities (O-5 grains per 100 ym2) In situ hybridization. Eighteen adult male Sprague-Dawley rats (body were determined by averaging grain counts over defined areas of the weight, 200-250 gm; ALAB, Stockholm, Sweden) were deeply anes- neuropil devoid of positively labeled cell bodies. Five hundred and fifty thetized with sodium pentobarbital (Mebumal; 60 mg/kg, i.p.), and the to 820 neurons from the sections of each DRG of each animal of the left sciatic nerve was transected above trochanter major. In all cases, a two groups were analyzed. The total number of labeled neuron profiles 5 mm portion of the nerve was resected. The rats were allowed to was divided by the total number of toluidine blue-stained neuron pro- survive for 2, 7, 21, and 28 d (three rats in each group) or 14 d (six files. rats). Lesioned animals and three normal control animals were deeply To determine the size of DRG neurons positive for NT-R or NPY-R anesthetized (as above) and perfused via the aorta with 50 ml warm mRNAs, 200 positive neurons for each receptor mRNA and 500 DRG (37°C) saline to clear the blood, followed by rapid dissection and freez- ing of L4 and L5 DRGs and L4-5 spinal cord segments. Before sec- neurons were randomly selected and measured on lightly toluidine blue- tioning, experimental and normal DRGs or spinal cords were fused by stained sections. A Macintosh 11x equipped with a DAGE-MT1 CCD- saline on the same blocks, so that all groups could be processed on the 72 series camera connected with a Nikon Microphot-FX microscope same slide. Sections (14 pm) were cut in-a cryostat (&Iicrom, Heidel- equipped with a 20X objective lens was used to collect the data.