Fos induction in lamina I projection neurons in response to noxious thermal stimuli

Neuroscience 131 (2005) 209 –217

FOS INDUCTION IN LAMINA I PROJECTION NEURONS IN RESPONSE TO NOXIOUS THERMAL STIMULI A. J. TODD,a* R. C. SPIKE,a S. YOUNGa AND Z. PUSKÁRa,b

tion as nociceptors or thermoreceptors (Light and Perl, 1979; Sugiura et al., 1986). It contains neurons that project to various parts of the brain, including the thalamus, periaqueductal grey matter, parabrachial area and medullary reticular formation (Trevino and Carstens, 1975; Giesler et al., 1979; Menétrey et al., 1982; Lima and Coimbra, 1989; Lima et al., 1991; Craig, 1995; Villanueva and Bernard, 1999; Todd et al., 2000, 2002; Spike et al., 2003). Many of the nociceptive afferents that terminate in lamina I contain substance P (Hökfelt et al., 1975; Lawson et al., 1997), and this is released in the dorsal horn following noxious stimulation (Kuraishi et al., 1985; Duggan et al., 1988; Mantyh et al., 1995). Projection cells make up around 5% of the neuronal population in lamina I (Spike et al., 2003), and most of them express the neurokinin 1 (NK1) receptor, on which substance P acts (Marshall et al., 1996; Todd et al., 2000). Lamina I neurons that express the NK1 receptor can be selectively destroyed by intrathecal administration of saporin conjugated to substance P, and this leads to a dramatic reduction in hyperalgesia in both inflammatory and neuropathic models (Mantyh et al., 1997; Nichols et al., 1999). This suggests that NK1 receptor-expressing neurons in lamina I play an important role in the development of hyperalgesia. Electrophysiological studies have demonstrated that most neurons in lamina I are activated by noxious stimuli (Christensen and Perl, 1970; Price et al., 1978; Light et al., 1979; Ferrington et al., 1987; Han et al., 1998; Bester et al., 2000; Craig et al., 2001). Unlike cells in deeper dorsal horn laminae, which generally respond to both noxious and innocuous stimulation, the majority of lamina I neurons only respond to stimuli in the noxious range. Activation of lamina I neurons by noxious stimulation has also been revealed with anatomical markers, for example internalisation of the NK1 receptor (Mantyh et al., 1995; Allen et al., 1997) and induction of transcription factors such as Fos (Hunt et al., 1987). Many lamina I projection neurons can be activated by both noxious mechanical stimuli and noxious heating, but not by cold, whilst others respond to noxious mechanical, heat and cold stimuli (Price et al., 1978; Ferrington et al., 1987; Han et al., 1998; Bester et al., 2000; Craig et al., 2001). Han et al. (1998) have defined these two functional classes as “nociceptive-specific” (NS) cells, and “polymodal nociceptive” or heat/pinch/cold (HPC) cells, respectively. Although the majority of lamina I projection neurons respond to noxious stimuli, a significant number are activated by innocuous cooling (Dostrovsky and Craig, 1996; Craig et al., 2001). In addition, there appear to be much smaller populations of projection cells that respond to other stimuli, such as warming or the

a Spinal Cord Group, Institute of Biomedical and Life Sciences, West Medical Building, University of Glasgow, Glasgow G12 8QQ, UK b Department of Anatomy, Histology and Embryology, Faculty of Medicine, Semmelweiss University, Tuzoltó u.58. H-1094, Budapest, Hungary

Abstract—Lamina I of the spinal cord contains many projection neurons: the majority of these are activated by noxious stimulation, although some respond to other stimuli, such as innocuous cooling. In the rat, approximately 80% of lamina I projection neurons express the neurokinin 1 (NK1) receptor, on which substance P acts. Lamina I neurons can be classified into three main morphological classes: pyramidal, fusiform and multipolar cells. It has been reported that in the cat, pyramidal cells respond to innocuous cooling, and whilst both fusiform and multipolar cells are activated by noxious mechanical and heat stimuli, only cells in the latter group respond to noxious cold [Nat Neurosci 1 (1998) 218]. However, we have previously shown that NK1 receptor-immunoreactive projection neurons belonging to each morphological class are equally likely to up-regulate the transcription factor Fos after noxious chemical stimulation, and that the density of innervation by substance P-containing (nociceptive) afferents is similar for cells of each type [J Neurosci 22 (2002) 4103]. This suggests that the morphological–physiological correlation that has been reported in the cat may not apply in the rat. We have tested this further by examining Fos expression in lamina I spinoparabrachial neurons in the rat after application of noxious heat or noxious cold stimuli under general anesthesia. Following noxious heat, 57– 69% of NK1 receptor-immunoreactive spinoparabrachial neurons expressed Fos, and the proportion did not differ significantly between morphological groups. However, after noxious cold stimulation Fos was present in 63% of multipolar neurons, but only 19 –26% of fusiform or pyramidal cells. These results suggest that although most NK1 receptor-expressing spinoparabrachial neurons are activated by noxious stimuli, responsiveness to noxious cold is significantly more common in those of the multipolar type. There therefore appears to be a correlation between morphology and function for lamina I projection neurons in the rat. © 2005 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: noxious heat, noxious cold, NK1 receptor, morphology, spinoparabrachial.

Lamina I of the spinal dorsal horn is densely innervated by fine diameter primary afferent axons, most of which func*Corresponding author. Tel: ⫹44-141-330-5868; fax: ⫹44-141-3302868. E-mail address: [email protected] (A. J. Todd). Abbreviations: CTb, cholera toxin B subunit; CVLM, caudal ventrolateral medulla; HPC, heat/pinch/cold; LPb, lateral parabrachial area; NK1, neurokinin 1; NS, nociceptive-specific.

0306-4522/05$30.00⫹0.00 © 2005 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2004.11.001

209

210

A. J. Todd et al. / Neuroscience 131 (2005) 209 –217

application of pruritic chemicals (Andrew and Craig, 2001a,b). Lamina I projection neurons can be classified based on the morphology of their cell bodies and primary dendrites. Gobel (1978) originally identified pyramidal and multipolar neurons in lamina I of the cat spinal trigeminal nucleus, and more recent studies in rat, cat and monkey have recognised three major classes: fusiform, pyramidal and multipolar (or flattened) cells (Lima et al., 1991; Zhang et al., 1996; Zhang and Craig, 1997; Todd et al., 2002; Spike et al., 2003). Han et al. (1998) carried out intracellular recording and labelling in lamina I of cat spinal cord, and concluded that morphology was closely related to function. They reported that pyramidal cells responded to innocuous cooling, that fusiform cells were NS, whilst multipolar cells belonged to either the NS or HPC classes. In support of this, Yu et al. (1999) reported that most pyramidal spinothalamic neurons in the monkey did not possess the NK1 receptor, which appears to be expressed only by dorsal horn neurons that respond to noxious stimulation (Henry, 1976). However, there is also evidence to suggest that the correlation between morphology and function may not apply in the rat. Many lamina I pyramidal neurons in rat spinal cord express the NK1 receptor (Cheunsuang and Morris, 2000; Spike et al., 2003), and we have previously reported that both the density of innervation by substance Pcontaining (nociceptive) primary afferents and the frequency of up-regulation of Fos protein following an acute noxious chemical stimulus (s.c. formalin injection) were similar in NK1 receptor-immunoreactive projection neurons belonging to each morphological class (Todd et al., 2002). In the present study, we have tested the hypothesis that morphology of lamina I projection neurons in the rat is related to function by examining the up-regulation of Fos in spinoparabrachial neurons following either noxious heat or noxious cold stimuli.

EXPERIMENTAL PROCEDURES Animals Nine adult male Wistar rats (Harlan, Loughborough, UK; 220 – 390 g) were used in this study. Six of the rats were deeply anaesthetised with a mixture of ketamine and xylazine (7.33 and 0.73 mg/100 g i.p., respectively, supplemented as necessary) and placed in a stereotaxic frame. Each of these rats received an injection of 200 nl 1% cholera toxin B subunit (CTb; Sigma, Poole, Dorset, UK) through a glass micropipette into the lateral parabrachial area (LPb) on the left side. Following a 3 day survival period, they were re-anaesthetised with ketamine and xylazine i.p. and received one of the following noxious stimuli to the right hind-paw: (1) a single immersion of the paw in water at 52 °C for 20 s (3 rats), or (2) repeated immersion of the paw in water at 4 °C for 30 s in every 2 min, over a period of 30 min (3 rats; Doyle and Hunt, 1999b). The animals were maintained under anaesthetic for 2 h after the application of the noxious stimulus, and were then injected with pentobarbitone i.p. and perfused through the left ventricle with 4% freshly depolymerised formaldehyde. To control for the effect of movement of water over the hindpaw and the handling involved in the noxious cold stimulation protocol, a further three rats that had not received stereotaxic injections were anaesthetised with ketamine and xylazine (as described above) and had

their right hindpaws repeatedly immersed in water at room temperature (23 °C) for 30 s in every 2 min, over a period of 30 min. These animals were maintained under anaesthetic for a further 2 h, and then injected with pentobarbitone i.p. and perfused with fixative, as described above. All experiments were approved by the Ethical Review Process Applications Panel of the University of Glasgow, and were performed in accordance with the European Community Directive, 86/609/EC, and the UK Animals (Scientific Procedures) Act 1986. All efforts were made to minimise the number of animals used and their suffering.

Immunocytochemistry Mid-lumbar spinal cord segments (L3–L5) were removed and post-fixed overnight. Horizontal sections through the superficial dorsal horn (70 ␮m thick) were cut with a Vibratome and immersed in 50% ethanol for 30 min to enhance antibody penetration (Llewellyn-Smith and Minson, 1992). The sections from the six rats that had received stereotaxic tracer injection and noxious heat or cold stimulation were incubated for 24 h in a cocktail of the following primary antibodies: goat anti-CTb (List Biological, Campbell, CA, USA; 1:5000), guinea-pig antiserum against NK1 receptor (Polgár et al., 1999; diluted 1:1000) and rabbit anti-Fos (Hunt et al., 1987; diluted 1:5000). These were revealed by overnight incubation in appropriate species-specific secondary antibodies that were labelled with lissamine rhodamine, fluorescein isothiocyanate or cyanine 5.18 (all raised in donkey; Jackson Immunoresearch, West Grove, PA, USA; diluted 1:100). Sections were mounted in anti-fade medium (Vectashield; Vector Laboratories, Peterborough, UK) and stored at ⫺20 °C. Sections from the three rats that had not received stereotaxic injections and that had had the hindpaw immersed in water at room temperature were processed in the same way, except that the anti-CTb antibody was omitted.

Analysis To compare the numbers of Fos-immunoreactive neurons in the superficial dorsal horn in animals that had had the hindpaw repeatedly immersed in water at 4 °C or 23 °C, sections from the L4 segment of these animals were scanned with a confocal laser scanning microscope (Bio-Rad Radiance 2100, Bio-Rad, Hemel Hempstead, UK) through a 60⫻ oil-immersion lens. For each of these six rats, two z-series (201⫻201 ␮m) consisting of 48 optical sections at 1 ␮m z-separation were scanned from randomly selected regions in the medial half of the dorsal horn. These series included lamina I and the adjacent part of lamina II. The total number of Fos-immunoreactive nuclei was counted and averaged across the two z-series for each animal. In order to analyse Fos expression in lamina I spinoparabrachial neurons that resulted from noxious heat or noxious cold stimulation, sections of the L3–L5 segments that showed high levels of Fos expression in lamina I were selected (between one and three sections were examined from each of these six rats). The sections were initially viewed at low magnification to determine the region within the medial part of lamina I that contained numerous Fos-immunoreactive nuclei (Todd et al., 2002). Sections were then scanned through this region with a confocal microscope (Bio-Rad MRC1024 or Radiance 2100) and an oilimmersion lens to reveal CTb, NK1 receptor and Fos. Numerous z-series from each section were acquired in this way, and these were analysed with Neurolucida for Confocal (MicroBrightField, Inc., Colchester, VT, USA). Files corresponding to CTb and NK1 receptor immunoreactivity were used to draw the cell bodies and proximal dendrites of retrogradely labelled neurons that were NK1 receptor-immunoreactive (Todd et al., 2002). In this way, 30 NK1 receptor-immunoreactive lamina I projection neurons of each morphological class (multipolar, fusiform, pyramidal) were selected from each animal, using criteria described by Zhang et al. (1996)

A. J. Todd et al. / Neuroscience 131 (2005) 209 –217

211

Fig. 1. Diagrams to show the spread of tracer (shaded area) in each of the six experiments. Each vertical column represents a single experiment, and the extent of CTb labelling is shown at different rostrocaudal levels through the brainstem. Numbers to the left of the drawings in the first column give the approximate position of the section anterior or posterior (⫺) to the ear bar. The first three columns show the injection sites in animals stimulated with water at 52 °C (Heat) and the last three are from animals that were stimulated at 4 °C (Cold). Drawings are based on those of Paxinos and Watson (1997). CN, cuneiform nucleus; IC, inferior colliculus; KF, Kölliker-Fuse nucleus; MPb, medial parabrachial area; scp, superior cerebellar peduncle.

and Zhang and Craig (1997). The morphological classification of each cell was agreed by two observers. In addition, 20 retrogradely labelled lamina I neurons that were not NK1 receptorimmunoreactive were also selected (except in one experiment, in which only 18 such neurons could be identified within the area of Fos labelling). No attempt was made to classify neurons that lacked NK1 receptors into morphological groups, for two reasons. Firstly, it is not possible to identify enough neurons of this type within the appropriate part of the dorsal horn to give adequate populations if these were further subdivided. Secondly, the filling of these cells with CTb is often relatively weak, compared with projection neurons that express the NK1 receptor, and this (together with the lack of receptor on their surface) means that it is not always possible to classify them on morphological grounds. Once the classification of lamina I projection neurons was completed, the files containing Fos-immunostaining were examined, and the presence or absence of Fos in the nucleus of each neuron was determined. The region of the brainstem that included the injection site was cryoprotected with 30% sucrose overnight and cut into 100␮m-thick coronal sections with a freezing microtome. Every fifth section was reacted with goat antiserum against CTb (List; 1:50,000) using an immunoperoxidase method (Todd et al., 2000). The spread of tracer from the injection site in each case was plotted onto drawings of the brainstem (Paxinos and Watson, 1997).

(Spike et al., 2003). Even though all rats had undergone noxious stimulation of the ipsilateral hindpaw 2 h previously, there was relatively little internalisation of NK1 receptors on neurons in lamina I on the right side of the spinal cord (Figs. 3, 4), compared with that observed at shorter intervals (e.g. Mantyh et al., 1995; Allen et al., 1997). Since noxious thermal stimuli similar to those used in this study are known to internalise NK1 receptors on lamina I neurons (Allen et al., 1997), it is likely that receptors lost from the plasma membrane due to internalisation had been replaced by the time of perfusion. As described previously (Doyle and Hunt, 1999a; Spike et al., 2002), Fos-immunoreactive nuclei were numerous in the medial part of lamina I in the L3 to L5 segments of rats that had had a hind-paw immersed in water at 52 °C, and were virtually restricted to the ipsilateral side. In rats that had had the hindpaw immersed in

RESULTS The CTb injections filled most or all of the LPb (Figs. 1, 2). There was invariably spread of tracer into the medial parabrachial area, and sometimes into the Kolliker-Fuse and/or cuneate nuclei, but never into the periaqueductal grey matter. The distribution of retrogradely labelled neurons in lamina I, as well as their morphological appearances and pattern of NK1 receptor expression, was the same as that reported previously following tracer injections into the LPb

Fig. 2. Photomicrograph of a representative section through the injection site in one of the experiments (corresponding to the sixth column in Fig. 1). Scale bar⫽1 mm.

212

A. J. Todd et al. / Neuroscience 131 (2005) 209 –217

Fig. 3. Expression of Fos protein in lamina I projection neurons in response to noxious heat. All images are from a single horizontal section through lamina I. In each row the left panel (a, d, g, j) shows immunostaining for CTb (red) in a projected image from a confocal series, and the centre panel (b, e, h, k) shows the equivalent field scanned for NK1 receptor-immunoreactivity (green). The right panel (c, f, i, l) is a single optical section from the confocal series showing CTb (red), NK1 receptor (green) and Fos protein (blue). (a– c, d–f, g–i) NK1 receptor-immunoreactive retrogradely labelled lamina I neurons belonging to fusiform (f), pyramidal (p) and multipolar (m) classes, respectively. In each case the nucleus is Fos-immunoreactive. (j–l) Two retrogradely labelled neurons that were not NK1 receptor-immunoreactive. One of these (n1) has a nucleus that is not Fos-immunoreactive, whilst the nucleus of the other (n2) is Fos-immunoreactive. Projections in a, d, g and j are from 10, nine, 11 and six optical sections, respectively, at 1 ␮m z-separation. Scale bar⫽20 ␮m.

water at 4 °C, Fos-immunoreactive nuclei were present throughout the whole rostrocaudal length of the L3–L5 segments on the ipsilateral side in the superficial laminae, and were largely confined to the medial half of the dorsal horn. As reported by Doyle and Hunt (1999b), some Fos-

labelled neurons were also seen on the side contralateral to the stimulated limb; however, these were much less numerous than those on the ipsilateral side. Very few Fos-labelled neurons were present in the superficial dorsal horn on either side in the rats that had had the hindpaw

Fig. 4. Expression of Fos protein in a multipolar lamina I projection neuron in response to noxious cold. (a) Immunostaining in a projected image of 27 optical sections at 1 ␮m z-spacing from a horizontal section of lamina I. (b) The corresponding field scanned for NK1 receptor. Four NK1 receptor-immunoreactive projection neurons are visible: two fusiform cells (f1, f2), one multipolar cell (m) and one pyramidal cell (p). (c) Single optical sections through the nuclei of the four cells scanned to reveal CTb (red), NK1 receptor (green) and Fos protein (blue). Only the multipolar cell has Fos-immunoreactivity in its nucleus. Note that the nuclei of f2, m and p were all at the same vertical level, whilst that of f1 was at a different level. A different optical section has therefore been used to illustrate the nucleus of f1. Arrow in c indicates the nucleus of f2. Scale bar⫽20 ␮m.

A. J. Todd et al. / Neuroscience 131 (2005) 209 –217

213

Table 1. Fos-immunoreactivity in lamina I projection neurons following noxious thermal stimulationa Type of cell

NK1⫹ multipolar NK1⫹ fusiform NK1⫹ pyramidal NK1⫺

Sample size per experiment

30 30 30 20 (*18)

Number Fos-immunoreactive

Mean % Fos-immunoreactive

Expt 1

Expt 2

Expt 3

21 21 19 4

15 18 20 3

15 17 23 *1

57 62 69 14

a

Comparison of Fos expression among different morphological types of NK1 receptor-immunoreactive projection neuron and in projection cells that lacked the receptor, following immersion of the ipsilateral hindpaw in water at 52 °C. * Note that in experiment 3 only 18 cells without the NK1 receptor were analysed. No significant difference in Fos-expression was found between the different morphological classes of NK1 receptor-immunoreactive cell, but cells that lacked the receptor showed significantly lower Fos-expression (Pⱕ0.001; one-way ANOVA with Tukey’s test post hoc).

immersed in water at 23 °C. Although lamina I projection neurons can be readily identified in horizontal sections, the boundaries of lamina I cannot be accurately defined in this plane of section. For this reason it was not possible to compare the number of lamina I neurons that expressed Fos in each group. However, to confirm that the Fos expression in animals that were stimulated at 4 °C was due to the cold stimulus, rather than to the repeated immersion of the hindpaw in water or the associated handling, we compared the numbers of Fos-immunoreactive nuclei in the superficial dorsal horn in these rats with the numbers in rats that had had the hindpaw immersed in water at 23 °C. The mean numbers of Fos-immunoreactive nuclei in the z-series scanned from the three animals stimulated at 4 °C ranged from 16 to 20 (mean⫽17), whilst the equivalent results for the three animals stimulated at 23 °C were 0 –3 (mean⫽1). This difference was highly significant (P⬍ 0.001, t-test). In the rats that had received a noxious heat stimulus, Fos-immunoreactivity was present in between 57 and 69% of NK1 receptor-expressing projection neurons belonging to the different morphological populations (Fig. 3, Table 1), and there was no significant difference between the proportions of neurons in each morphological class that had Fos-immunoreactive nuclei (one-way ANOVA). Fos-immunoreactivity was much less common in projection neurons that lacked the NK1 receptor (mean 14%) than in any of the other three groups (Table 1), and this difference was highly significant (Pⱕ0.001; Tukey’s test post hoc). Following a noxious cold stimulus, Fos was present in nuclei belonging to some spinoparabrachial neurons in each class. However, it was much more frequently seen in NK1 receptor-immunoreactive neurons that had been classified as multipolar (Table 2). An example of a Fos-labelled multipolar cell is shown in Fig. 4. ANOVA confirmed that the difference between groups was significant (P⬍0.01), and Tukey’s test revealed that Fos expression was significantly more common in multipolar NK1 receptor-immunoreactive neurons than in any of the other groups (P⬍0.05).

DISCUSSION The aim of this study was to investigate the induction of Fos in different types of lamina I projection neuron (classified by

NK1 receptor expression and morphology) resulting from noxious thermal stimuli. The main finding was that both noxious heat and noxious cold stimuli (immersion of the ipsilateral hindpaw in water at 52 °C or 4 °C, respectively) caused up-regulation of Fos in some lamina I spinoparabrachial neurons that expressed the NK1 receptor, but that the pattern of Fos-expression in relation to morphology differed between the two stimuli. For noxious heat, between 57 and 69% of NK1 receptor-immunoreactive projection neurons expressed Fos, and the proportion did not differ significantly between morphological groups. However, following the noxious cold stimulus, NK1 receptor-immunoreactive spinoparabrachial neurons that were multipolar were significantly more likely to express Fos (63%) than those that were fusiform (19%) or pyramidal (26%). With both types of noxious thermal stimulus, a relatively small proportion of spinoparabrachial neurons that lacked the NK1 receptor showed Fos-immunoreactivity (14 –15%). Choice of projection target The LPb was used for retrograde labelling as it receives a substantial projection from lamina I neurons in the rat (Cechetto et al., 1985; Hylden et al., 1989; Feil and Herbert, 1995; Slugg and Light, 1994; Villanueva and Bernard, 1999; Todd et al., 2000; Spike et al., 2003), and also because there are electrophysiological data concerning the response properties of rat spinoparabrachial neurons (Bester et al., 2000). In our previous study of Fos expression and substance P innervation of lamina I neurons, we used injections into the caudal ventrolateral medulla (CVLM) to identify projection cells (Todd et al., 2002). We have recently shown that injections of tracer into the LPb or the CVLM label similar numbers of lamina I neurons (approximately nine to 10 cells per 70 ␮m transverse section in the L4 segment) and that there is a substantial overlap between these two populations (Spike et al., 2003). Our findings in that study led us to conclude that injections into either LPb or CVLM would label approximately 85% of all projection neurons in lamina I. It is therefore likely that the populations of neurons examined here are very similar to those investigated in our previous Fos study (Todd et al., 2002) and are representative of all lamina I projection neurons.

214

A. J. Todd et al. / Neuroscience 131 (2005) 209 –217

Table 2. Fos-immunoreactivity in lamina I projection neurons following noxious cold stimulationa Number Fos-immunoreactive Type of cell

Sample size per experiment

Expt 1

Expt 2

Expt 3

Mean % Fos-immunoreactive

NK1⫹ multipolar NK1⫹ fusiform NK1⫹ pyramidal NK1 ⫺

30 30 30 20

21 2 2 5

18 4 9 1

18 11 12 3

63 19 26 15

a

Comparison of Fos expression among different morphological types of NK1 receptor-immunoreactive projection neuron and in projection cells that lacked the receptor, following immersion of the ipsilateral hindpaw in water at 4 °C. The incidence of Fos-expression was significantly higher in multipolar NK1 receptor-immunoreactive neurons than in the other three groups (P⬍0.05; one-way ANOVA with Tukey’s test post hoc).

Induction of Fos in lamina I in response to noxious thermal stimuli There have been numerous reports of Fos expression in the superficial dorsal horn of the spinal cord or spinal trigeminal nucleus following noxious heat stimulation with temperatures from 46 to 54 °C (e.g. Hunt et al., 1987; Bullitt, 1990; Williams et al., 1990; Strassman et al., 1993; Abbadie et al., 1994b; Doyle and Hunt, 1999a; Bester et al., 2001; Dai et al., 2001; Spike et al., 2002). The threshold for induction appears to between 42 and 46 °C, and the number of Fos-immunoreactive neurons increases with temperature, up to approximately 52 °C. Relatively brief applications of noxious heat (10 –20 s) are able to produce substantial Fos-labelling in the superficial dorsal horn. Four previous studies have investigated Fos expression in the spinal cord or spinal trigeminal nucleus in response to noxious cold stimulation of the skin of the hindlimb (Abbadie et al., 1994a; Doyle and Hunt, 1999b; Dai et al., 2001) or the face (Strassman et al., 1993). In three of these studies, several (10 –15) applications of the cold stimulus (4 –10 °C) were applied at 2 min intervals, and in each case moderate numbers of Fos-immunoreactive nuclei were seen in lamina I (Strassman et al., 1993; Doyle and Hunt, 1999b; Dai et al., 2001). However, Fos-positive neurons in this lamina were less numerous than those seen after application of noxious heat, for example Doyle and Hunt (1999a,b) found approximately 16 Fos-labelled lamina I cells per 40 ␮m transverse section following immersion of the hindpaw in water at 52 °C, but only seven such cells per section after repeated immersion of the paw in water at 4 °C. This difference presumably reflects the fact that noxious heat activates a larger proportion of lamina I neurons than does noxious cold (Han et al., 1998; Bester et al., 2000; Craig et al., 2001). In the study of Abbadie et al. (1994a), only a single immersion (for periods of between 1 and 5 min) was used, and with this protocol virtually no Fos was induced at temperatures above ⫺10 °C. This indicates that unlike the situation with noxious heating, repeated stimuli at temperatures between 4 and 10 °C are required to produce Fos, even though these stimuli are clearly within the noxious range. Bester et al. (2000) reported that ⬎90% of lamina I spinoparabrachial neurons were activated by noxious heat (50 °C for 20 s) and that this resulted in a high firing frequency (mean 40⫾4.4 Hz), whereas only 35% of these cells responded to noxious cold stimuli and this resulted in a much lower firing rate in these cells (7.6⫾4.4 Hz). This suggests that activation of Fos requires either a relatively brief stimulus

that evokes a high firing frequency (e.g. noxious heat) or else prolonged stimulation that causes firing at a lower frequency (e.g. noxious cold). Fos induction in NK1 receptor-expressing projection neurons in lamina I Doyle and Hunt (1999a,b) demonstrated that some of the neurons in lamina I that showed Fos in response to noxious thermal stimuli were immunoreactive for the NK1 receptor. They reported that around 80% of NK1 receptorexpressing neurons in lamina I were Fos-positive in response to a 10 s immersion of the hindpaw in water at 52 °C, whilst around 50% were labelled by a noxious cold stimulus similar to that used in the present study. The proportions of NK1 receptor-expressing neurons that were Fos-labelled in our experiments are somewhat lower (63% for heat, 36% for cold, data pooled from all three morphological classes shown in Tables 1, 2). This discrepancy may reflect differences in the selection of cells between our study and those of Doyle and Hunt (1999a,b), since we only examined neurons that were retrogradely labelled from the LPb, and we also selected cells on the basis of their morphology. We have previously reported that 45% of lamina I neurons express the NK1 receptor (Todd et al., 1998), whereas we estimate that only 5% of neurons in this lamina project to the brain (Spike et al., 2003). This indicates that many NK1 receptor-expressing neurons in lamina I are likely to be interneurons, and these probably correspond to the small cells with relatively weak NK1 receptor-immunoreactivity described by Cheunsuang and Morris (2000; for discussion, see Polgár et al., 2002). NK1 receptor-expressing interneurons in lamina I may have been included in the populations sampled by Doyle and Hunt (1999a,b). In a study of lamina I spinoparabrachial neurons in the rat, Bester et al. (2000) reported that all of the neurons examined were activated by noxious heat stimuli, whereas 35% responded to noxious cold. The figure for cold responsiveness is consistent with our results, but it appears that in our experiments, immersion of the hindpaw in water at 52 °C evoked Fos expression in only a proportion of the neurons that were capable of responding to a noxious heat stimulus. Interestingly, Allen et al. (1997) found that between 50 and 70% of NK1 receptor-immunoreactive lamina I neurons internalised the receptor in response to a 30 s heat stimulus at 48 or 55 °C, whilst 18 –28% showed internal-

A. J. Todd et al. / Neuroscience 131 (2005) 209 –217

isation following a 30 s cold stimulus at 0 or 10 °C. This indicates that substance P is released in response to both noxious heat and noxious cold stimuli. However, substance P may not contribute significantly to the Fos upregulation (at least for the noxious heat stimulus), since Bester et al. (2001) have shown that Fos expression in lamina I of the lumbar spinal cord is similar in NK1 receptor knock-out mice and in wild-type mice following immersion of the ipsilateral hindpaw in water at 50 °C. The low level of Fos expression by neurons that lack the NK1 receptor in response to noxious thermal stimuli (present study) or injection of formalin (Todd et al., 2002) is difficult to interpret, since the results of Bester et al. (2000) suggest that all (or virtually all) spinoparabrachial neurons in lamina I respond to noxious stimulation. It is possible that these cells were more weakly activated by the noxious stimuli than were NK1 receptor-expressing cells, and thus did not reach the threshold for activation of Fos. Alternatively, cells of this type may be intrinsically less likely to produce Fos even when adequately stimulated. Correlation between morphology and function in lamina I projection neurons The results of Han et al. (1998) and Yu et al. (1999) suggest that in cat and monkey, there is a close relationship between morphology and physiology for lamina I neurons. The majority of pyramidal spinothalamic cells in the monkey lacked the NK1 receptor (Yu et al., 1999), and pyramidal lamina I neurons recorded in the cat were activated by innocuous cooling (Han et al., 1998). In addition, although both fusiform and multipolar cells were found to respond to noxious mechanical and heat stimuli, responses to noxious cold were only seen in multipolar cells (Han et al., 1998). However, data obtained from the rat suggest that the situation may be different in this species. We have reported that amongst lamina I projection neurons in the rat, 80% of pyramidal cells are NK1 receptor-immunoreactive, and the proportion of strongly, moderately and weakly immunoreactive cells is similar to that seen in multipolar and fusiform populations (Spike et al., 2003). We have also found that for NK1 receptor-immunoreactive projection cells labelled from the CVLM, the density of innervation by substance P-containing afferents did not differ between pyramidal, multipolar and fusiform types, and that approximately 80% of cells of each type showed Fos-labelling after injection of formalin into the ipsilateral hindpaw (Todd et al., 2002). These results suggest that in the rat, most (if not all) pyramidal neurons are activated by nociceptors. Consistent with this view, we have found that approximately 30% of lamina I neurons projecting to the LPb are pyramidal cells (Spike et al., 2003), whilst Bester et al. (2001) reported that all lamina I spinoparabrachial neurons responded to noxious stimuli. The results of the present study show that many pyramidal cells also respond to noxious heat. The high frequency of Fos expression by multipolar NK1 receptor-immunoreactive cells in response to noxious cold stimuli suggests that, as reported in the cat (Han et al.,

215

1998), many multipolar cells belong to the HPC class. It is more difficult to interpret negative results with Fos, since failure to express Fos does not mean that a cell has not been activated (see above). However, the fact that most NK1 receptor-immunoreactive fusiform and pyramidal cells were Fos-labelled following noxious heat (present study) or s.c. formalin (Todd et al., 2002), but not after noxious cold stimulation suggests that the majority of these cells are likely to be unresponsive (or only weakly responsive) to noxious cold stimuli, and would thus belong to the NS class, as defined by Han et al. (1998). In conclusion, although our results do not support the view that pyramidal cells in lamina I are activated exclusively by innocuous stimuli (as has been proposed for the cat), they do suggest a functional difference between projection neurons of the multipolar class (most of which are activated by noxious cold stimuli, and presumably correspond to HPC cells) and those belonging to fusiform and pyramidal types (most of which are likely to belong to the NS class of Han et al., 1998). Acknowledgments—We thank Mrs. M. M. McGill, Mrs. Christine Watt and Mr. Robert Kerr for expert technical assistance, and Dr. D. Andrew for helpful discussion. The work was supported by grants from the Wellcome Trust, the Hungarian Ministry of Education (grant number FKFP 0121/2001) and the Hungarian Science and Technology Foundation (GB 2/03).

REFERENCES Abbadie C, Honoré P, Besson J-M (1994a) Intense noxious cold stimulation of the rat hindpaw induces c-fos expression in lumbar spinal cord neurons. Neuroscience 59:457– 468. Abbadie C, Honore P, Fournie-Zaluski MC, Roques BP, Besson JM (1994b) Effects of opioids and non-opioids on c-Fos-like immunoreactivity induced in rat lumbar spinal cord neurons by noxious heat stimulation. Eur J Pharmacol 258:215–227. Allen BJ, Rogers SD, Ghilardi JR, Menning PM, Kuskowski MA, Basbaum AI, Simone DA, Mantyh PW (1997) Noxious cutaneous thermal stimuli induce a graded release of endogenous substance P in the spinal cord: imaging peptide action in vivo. J Neurosci 17:5921–5927. Andrew D, Craig AD (2001a) Spinothalamic lamina I neurons selectively sensitive to histamine: a central neural pathway for itch. Nat Neurosci 4:72–77. Andrew D, Craig AD (2001b) Spinothalamic lamina I neurones selectively responsive to cutaneous warming in cats. J Physiol 537:489 – 495. Bester H, Chapman V, Besson J-M, Bernard J-F (2000) Physiological properties of the lamina I spinoparabrachial neurons in the rat. J Neurophysiol 83:2239 –2259. Bester H, De Felipe C, Hunt SP (2001) The NK1 receptor is essential for the full expression of noxious inhibitory controls in the mouse. J Neurosci 21:1039 –1046. Bullitt E (1990) Expression of c-fos-like protein as a marker for neuronal activity following noxious stimulation in the rat. J Comp Neurol 296:517–530. Cechetto DF, Standaert DG, Saper CB (1985) Spinal and trigeminal dorsal horn neuron projections to the parabrachial nucleus in the rat. J Comp Neurol 240:153–160. Cheunsuang O, Morris R (2000) Spinal lamina I neurons that express neurokinin 1 receptors: morphological analysis. Neuroscience 97:335–345.

216

A. J. Todd et al. / Neuroscience 131 (2005) 209 –217

Christensen BN, Perl ER (1970) Spinal neurons specifically excited by noxious or thermal stimuli: marginal zone of the dorsal horn. J Neurophysiol 33:293–307. Craig AD (1995) Distribution of brainstem projections from spinal lamina I neurons in the cat and monkey. J Comp Neurol 361:225–248. Craig AD, Krout K, Andrew D (2001) Quantitative response characteristics of thermoreceptive and nociceptive lamina I spinothalamic neurons in the cat. J Neurophysiol 86:1459 –1480. Dai Y, Iwata K, Kondo E, Morimoto T, Noguchi K (2001) A selective increase in Fos expression in spinal dorsal horn neurons following graded thermal stimulation in rats with experimental mononeuropathy. Pain 90:287–296. Dostrovsky JO, Craig AD (1996) Cooling-specific spinothalamic neurons in the monkey. J Neurophysiol 76:3656 –3665. Doyle CA, Hunt SP (1999a) Substance P receptor (neurokinin-1)expressing neurons in lamina I of the spinal cord encode for the intensity of noxious stimulation: a c-fos study in rat. Neuroscience 89:17–28. Doyle CA, Hunt SP (1999b) A role for spinal lamina I neurokinin-1positive neurons in cold thermoreception in the rat. Neuroscience 91:723–732. Duggan AW, Hendry IA, Morton CR, Hutchison WD, Zhao ZQ (1988) Cutaneous stimuli releasing immunoreactive substance P in the dorsal horn of the cat. Brain Res 451:261–273. Feil K, Herbert H (1995) Topographic organization of spinal and trigeminal somatosensory pathways to the rat parabrachial and Kölliker-Fuse nuclei. J Comp Neurol 353:506 –528. Ferrington DG, Sorkin LS, Willis WD (1987) Responses of spinothalamic tract cells in the superficial dorsal horn of the primate lumbar spinal cord. J Physiol 388:681–703. Giesler GJ, Menétrey D, Basbaum AI (1979) Differential origins of spinothalamic tract projections to medial and lateral thalamus in the rat. J Comp Neurol 184:107–125. Gobel S (1978) Golgi studies of the neurons in layer I of the dorsal horn of the medulla (trigeminal nucleus caudalis). J Comp Neurol 180:375–394. Han Z-S, Zhang E-T, Craig AD (1998) Nociceptive and thermoreceptive lamina I neurons are anatomically distinct. Nat Neurosci 1:218 –225. Henry JL (1976) Effects of substance P on functionally identified units in cat spinal cord. Brain Res 114:439 – 451. Hökfelt T, Kellerth JO, Nilsson G, Pernow B (1975) Substance P: localization in the central nervous system and in some primary sensory neurons. Science 190:889 – 890. Hunt SP, Pini A, Evan G (1987) Induction of c-fos-like protein in spinal cord neurons following sensory stimulation. Nature 328:632– 634. Hylden JLK, Anton F, Nahin RL (1989) Spinal lamina I projection neurons in the rat: collateral innervation of parabrachial area and thalamus. Neuroscience 28:27–37. Kuraishi Y, Hirota N, Sato Y, Hino Y, Satoh M, Takagi H (1985) Evidence that substance P and somatostatin transmit separate information related to pain in the spinal dorsal horn. Brain Res 325:294 –298. Lawson SN, Crepps BA, Perl ER (1997) Relationship of substance P to afferent characteristics of dorsal root ganglion neurones in guinea-pig. J Physiol 505:177–191. Llewellyn-Smith IJ, Minson JB (1992) Complete penetration of antibodies into vibratome sections after glutaraldehyde fixation and ethanol treatment: light and electron microscopy for neuropeptides. J Histochem Cytochem 40:1741–1749. Light AR, Perl ER (1979) Spinal termination of functionally identified primary afferent neurons with slowly conducting myelinated fibres. J Comp Neurol 186:133–150. Light AR, Trevino DL, Perl ER (1979) Morphological features of functionally defined neurons in the marginal zone and substantia gelatinosa of the spinal dorsal horn. J Comp Neurol 186:151–172. Lima D, Coimbra A (1989) Morphological types of spinomesencephalic neurons in the marginal zone (lamina I) of the rat spinal cord, as

shown after retrograde labelling with cholera toxin subunit B. J Comp Neurol 279:327–339. Lima D, Mendes-Ribeiro JA, Coimbra A (1991) The spino-lateroreticular system of the rat: projections from the superficial dorsal horn and structural characterization of marginal neurons involved. Neuroscience 45:137–152. Mantyh PW, DeMaster E, Malhotra A, Ghilardi JR, Rogers SD, Mantyh CR, Liu H, Basbaum AI, Vigna SR, Maggio JE, Simone DA (1995) Receptor endocytosis and dendrite reshaping in spinal neurons after somatosensory stimulation. Science 268:1629 –1632. Mantyh PW, Rogers SD, Honore P, Allen BJ, Ghilardi JR, Li J, Daughters RS, Lappi DA, Wiley RG, Simone DA (1997) Ablation of lamina I spinal neurons expressing the substance P receptor profoundly inhibits hyperalgesia. Science 278:275–279. Marshall GE, Shehab SAS, Spike RC, Todd AJ (1996) Neurokinin-1 receptors on lumbar spinothalamic neurons in the rat. Neuroscience 72:255–263. Menétrey D, Chaouch A, Besson JM (1982) The origins of the spinomesencephalic tract in the rat: an anatomical study using the retrograde transport of horseradish peroxidase. J Comp Neurol 206:193–207. Nichols ML, Allen BJ, Rogers SD, Ghilardi JR, Honore P, Luger NM, Finke MP, Li J, Lappi DA, Simone DA, Mantyh PW (1999) Transmission of chronic nociception by spinal neurons expressing the substance P receptor. Science 286:1558 –1561. Paxinos G, Watson C (1997) The rat brain in stereotaxic coordinates. 3rd ed. San Diego: Academic Press. Polgár E, Shehab SAS, Watt C, Todd AJ (1999) GABAergic neurons that contain neuropeptide Y selectively target cells with the neurokinin 1 receptor in laminae III and IV of the rat spinal cord. J Neurosci 19:2637–2646. Polgár E, Puskár Z, Watt C, Matesz C, Todd AJ (2002) Selective innervation of lamina I projection neurones that possess the neurokinin 1 receptor by serotonin-containing axons in the rat spinal cord. Neuroscience 109:799 – 809. Price DD, Hayes RL, Ruda M, Dubner D (1978) Spatial and temporal transformations of input to spinothalamic tract neurons and their relation to somatic sensations. J Neurophysiol 41:933–947. Slugg RM, Light AR (1994) Spinal cord and trigeminal projections to the pontine parabrachial region in the rat as demonstrated with Phaseolus vulgaris leucoagglutinin. J Comp Neurol 339:49 – 61. Spike RC, Puskár Z, Sakamoto H, Stewart W, Watt C, Todd AJ (2002) MOR-1-immunoreactive neurons in the dorsal horn of the rat spinal cord: evidence for non-synaptic innervation by substance Pcontaining primary afferents and for selective activation by noxious thermal stimuli. Eur J Neurosci 15:1306 –1316. Spike RC, Puskár Z, Andrew D, Todd AJ (2003) A quantitative and morphological study of projection neurons in lamina I of the rat lumbar spinal cord. Eur J Neurosci 18:2433–2448. Strassman AM, Vos BP, Mineta Y, Naderi S, Borsook D, Burstein R (1993) Fos-like immunoreactivity in the superficial medullary dorsal horn induced by noxious and innocuous thermal stimulation of facial skin in the rat. J Neurophysiol 70:1811–1821. Sugiura Y, Lee CL, Perl ER (1986) Central projections of identified unmyelinated (C) afferent fibers innervating mammalian skin. Science 234:358 –361. Todd AJ, Spike RC, Polgár E (1998) A quantitative study of neurons which express neurokinin 1 or somatostatin sst2a receptor in rat spinal dorsal horn. Neuroscience 85:459 – 473. Todd AJ, McGill MM, Shehab SAS (2000) Neurokinin 1 receptor expression by neurons in laminae I, III and IV of the rat spinal dorsal horn that project to the brainstem. Eur J Neurosci 12:689–700. Todd AJ, Puskár Z, Spike RC, Hughes C, Watt C, Forrest L (2002) Projection neurons in lamina I of rat spinal cord with the neurokinin 1 receptor are selectively innervated by substance P-containing afferents and respond to noxious stimulation. J Neurosci 22: 4103– 4113.

A. J. Todd et al. / Neuroscience 131 (2005) 209 –217 Trevino DL, Carstens E (1975) Confirmation of the location of spinothalamic neurons in the cat and monkey by the retrograde transport of horseradish peroxidase. Brain Res 98:177–182. Villanueva L, Bernard J-F (1999) The multiplicity of ascending pain pathways. In: Handbook of behavioral state control: cellular and molecular mechanisms (Lydic R, Baghdoyan H, eds), pp 569 –585. Boca Raton, FL: CRC Press. Williams S, Evan GI, Hunt SP (1990) Changing patterns of c-fos induction in spinal neurons following thermal cutaneous stimulation in the rat. Neuroscience 36:73– 81.

217

Yu XH, Zhang ET, Craig AD, Shigemoto R, Ribeiro-da-Silva A, De Koninck Y (1999) NK-1 receptor immunoreactivity in distinct morphological types of lamina I neurons of the primate spinal cord. J Neurosci 19:3545–3555. Zhang ET, Craig AD (1997) Morphology and distribution of spinothalamic lamina I neurons in the monkey. J Neurosci 17: 3274 –3284. Zhang ET, Han Z-S, Craig AD (1996) Morphological classes of spinothalamic lamina I neurons in the cat. J Comp Neurol 367:537–549.

(Accepted 3 November 2004) (Available online 24 December 2004)