Peripheral noxious stimuli activate neurons in the spinal cord and at supraspinal levels. These activation patterns have been studied extensively by application of staining for the nuclear protein Fos.1–4
The protein Fos is encoded by the gene c-fos, an immediate early gene, which is rapidly expressed in neurons in response to stimulation. The protein Fos reaches its peak level about 2 h after induction of gene transcription. Fos is involved in the signal transduction pathways responsible for the long-term intracellular changes provoked by extracellular events.4
Using Fos expression as a tool for studying the neural basis of nociception has specific advantages. First, Fos expression enables the identification of the exact location of neurons that respond to the applied noxious stimuli. Second, Fos expression can be easily quantified. Third, immunocytochemical labeling of Fos can be performed in combination with other immunocytochemical techniques (e.g., labeling neurotransmitters or their receptors). These double stainings give more insight in the neuronal circuits involved in nociceptive processing.
However, using Fos expression also has some disadvantages. Not all neurons express the gene when they are activated, and Fos does not distinguish between inhibition and excitation activity but only indicates the involvement of the neuron in the investigated process.
In a previous study, Fos expression observed after the noxious stimulation-induced withdrawal reflex (NIWR) was investigated in both the dorsal horn of the spinal cord5 and in various brain regions (in preparation). For the NIWR, an electrical noxious stimulus is applied to the hindpaw of an (usually isoflurane) anesthetized rat. This stimulation evokes a response that can be quantified by the strength of the withdrawal force of the hindpaw.6,7 The response force of the NIWR depends on a combination of the pain processing at spinal and supraspinal levels and the level of anesthesia. The electrical stimulus induced Fos expression in both the dorsal horn of the spinal cord and in various brain regions. In the spinal cord, Fos expression was observed mainly in Laminae I, II, V, and X at the lumbar level (L4-5). The pattern of Fos expression obtained under low levels of isoflurane anesthesia changed when the NIWR was suppressed by a higher dose of isoflurane, or the same low dose of isoflurane in combination with the opiate fentanyl.5
The Fos expression induced by noxious stimulation in the dorsal horns at levels L4/L5 is likely to be the result of the interaction of three effects1: the direct input from activated afferent fibers2; the modulatory effects of local interactions in the same or adjacent spinal segments3; direct or indirect input of descending modulating signals from supraspinal regions.
We sought to distinguish between the effects on Fos expression induced by direct sensory input and local modulation, and the effects originating from modulating supraspinal pathways.
For this purpose, a thoracic spinal block was administered to interrupt the neuronal communication between the lumbar segments and supraspinal regions. This thoracic spinal block was accomplished by administering bupivacaine through an implanted intrathecal catheter.
Thirty-five adult male Wistar rats (weight 250-360 g; local breeding facilities of the Radboud University Nijmegen, The Netherlands) were used in this study. Animals were housed individually in the Central Animal Laboratory of the UMCN St. Radboud with a 12/12-h light/dark cycle. Food (Ssniff R/MH, Ssniff Spezialsdiäten Gmgh, Soest, Germany) and acidified water were available ad libitum. All experiments were approved by the Regional Animal Experiment Committee.
Intrathecal Catheter Implantation
Intrathecal catheters (PT45, ID 0.28 mm, OD 0.61 mm; Medica Europe BV) were placed as described by Yaksh and Rudy.8,9 The rat was placed in a stereotaxic head holder under isoflurane anesthesia. Skin and muscle were incised, and the cisternal membrane exposed at the base of the skull. The dura was then incised and the catheter inserted. Catheters were 5-cm long and reached to the mid-thoracic level of T6-8.
Animals were closely observed after awakening and preceding NIWR application, and if any behavioral or motor abnormalities appeared, the rat was killed and excluded from further experiments. To avoid possible Fos-immunoreactivity (IR) induced by intrathecal catheter insertion, rats were allowed to recover for at least 14 days before experiments were performed.
Rats were anesthetized with isoflurane (induction at 5.0%, maintenance dose 1.2%). The animal was placed in a box on a Peltier heating/cooling plate. Rectal temperature, heart rate, oxygen saturation, and end-expiratory CO2 were monitored; temperature was maintained at 37.0°C ± 0.5°C and CO2 concentration at 4.0% ± 0.5%. A stimulation shoe was placed on the left hindpaw to administer an electrical stimulus.
At the start of the experiment, depending on the experimental group of the animal, 15 μL 2.5% bupivacaine (AstraZeneca BV, Zoetermeer, The Netherlands) or saline solution was administered through the intrathecal catheter, followed by 10 μL of saline. Seven minutes later, the rat received noxious electrical stimulation (pulse trains of 100 Hz, pulse width 4 ms, 7.5 mA, train duration 500 ms, repeated each 80 s) for 10 min. Hereafter, the animal was maintained under anesthesia for 2 h.
Rats were then deeply anesthetized with 5.0% isoflurane and transcardially perfused with 250 mL 0.1 M phosphate buffered saline (PBS) (pH 7.3), followed by 400 mL 4% buffered paraformaldehyde (pH 7.2). After examining the exact location of the tip of the intrathecal catheter, the catheter was removed. Spinal cords were dissected, postfixed for 24 h at 4°C and then transferred to 30% buffered sucrose (pH 7.2), in which they were stored for 4-7 days at 4°C.
After 4-7 days, the spinal cords were cut in the frontal plane into 40-μm sections (divided over six parallel series; distance 240 μm between successive sections in each series) using a freezing microtome (HM440E, Microm, Walldorf, Germany) and collected in 0.1 M PBS.
Five experimental groups were examined. In all animals, a catheter was placed in the vertebral canal. On the day of the experiments, the rats in the control catheter group (CC) were anesthetized and immediately perfused. All other rats were placed in the experimental arrangement (under anesthesia). Two groups were not stimulated. In one, a saline solution was administered through the catheter (saline solution group, not stimulated, SSN). The second group received bupivacaine through the catheter (bupivacaine block group, not stimulated, BBN). The last two groups were given an electrical noxious stimulation on the left hindpaw, wherein one group received saline through the catheter (saline solution group, stimulated, SSS) and the last group received bupivacaine (bupivacaine block group, stimulated, BBS). After 2 h in the experimental arrangement, all animals in the last four groups were perfused.
From each animal, one series of about 20 lumbar spinal cord sections (about 5 mm at L4/L5 level) were immunostained for Fos. The sections were incubated overnight with rabbit-anti-fos, a primary polyclonal antibody against Fos, raised in rabbits (Santa Cruz Biotechnology, Santa Cruz, CA; 1:20,000). They were then incubated for 90 min with donkey-anti-rabbit, a secondary, biotin-conjugated antibody, raised in donkeys (Jackson Immunoresearch, Westgrove, PA; 1:1500). Hereafter, the sections were incubated for 90 min in Vector ABC-Elite (Vectastain, Brunschwig Chemie, Amsterdam, The Netherlands; 1:800).
To visualize the Fos/antibody complex, the diaminobenzidine tetrahydrochloride method was used. Sections were preincubated for 10 min with 3,3′-diaminobenzidine tetrahydrochloride-nickel solution (0.05 M Tris buffer, containing 0.02% diaminobenzidine tetrahydrochloride [Sigma Aldrich, St. Louis, MO] and 0.03% ammonium nickel sulfate, pH 7.6), and subsequently incubated for 10 min with 3,3′-diaminobenzidine tetrahydrochloride-nickel solution containing 0.02% perhydrol. This treatment results in a black-blue staining of the FOS/antibody complex.
Sections were extensively rinsed with 0.1 M PBS between successive steps. The sections were then mounted on gelatin-coated object glasses and dried overnight at 37°C. Finally, the sections were dehydrated in a graded alcohol series, cleared in xylol, and mounted in entellan.
All incubations were performed at room temperature.
Cresyl Violet Staining
To examine the exact pathway of, and possible damage induced by, the intrathecal catheter, cervical, and thoracic spinal cord sections were stained with cresyl violet. Sections were mounted on gelatin-coated object glasses and dried overnight at 37°C. They were then dehydrated in an alcohol series and stained in 0.1% cresyl violet solution. Sections were dehydrated in an alcohol series again, cleared in xylol, and mounted in entellan.
The lumbar Fos stained spinal cord sections were examined with a Zeiss Axioscop microscope combined with a NeuroLucida system (MicroBrightField, Williston, VT).
The three L4 sections of each animal containing the highest number of Fos-IR cells were selected.10,11 The distribution of Fos-IR neurons over laminae was determined according to the figure of the L4 segment from the Paxinos atlas of the rat brain.12 Fos-IR cells were counted in Laminae I/II on both sides, the side with the highest amount of Fos-IR was then defined ipsilateral to the noxious stimulus. This classification was maintained when all further laminae were counted. The numbers of labeled cells in the three sections were averaged for each animal.
Data were analyzed with one-way analysis of variance and Scheffe post hoc test, using SPSS 16.0 statistical software. A P value of <0.05 was considered significant.
Intrathecal Catheter and Bupivacaine Block
Two of the 35 animals showed behavioral abnormalities directly after recovery from catheter implantation surgery. They were killed and excluded from further study. All other rats did not show behavioral abnormalities from the implanted catheter.
Dissection of the spinal cord was done in most cases to confirm that the catheter tip had been placed approximately in the T6-8 segments. The spread of the bupivacaine block had been tested by adding a small amount of methylene blue to the bupivacaine in some pilot experiments. After administration of 15 + 10-μL solution, the rostrocaudal spread was about 1.5 cm in the rostral direction and only about 0.5 cm in the caudal direction.
Cresyl violet staining showed that the pathways followed through and alongside the spinal cord, and the location of the tip of the intrathecal catheter in the transverse plane varied greatly among animals. However, no correlation was found between the catheter tip location (ventral or dorsal) and the number of Fos labeled cells in the lumbar spinal cord sections.
Implantation of the catheter appeared much more invasive than expected, causing considerable malformation and compression of the spinal cord. However, it was well tolerated by the animals and did not seem to overtly affect motor or sensory pathways.
Noxious Stimulation-Induced Withdrawal Reflex
A normal withdrawal response was only observed in the SSS group. The thoracic spinal block (in group BBS) completely suppressed the NIWR elicited by electrical stimulus.
Eight animals were used in pilot experiments for testing the necessary bupivacaine concentration and dose, and the completeness and dynamics of the resulting block. One animal was excluded because of problems with the immunostaining. Thus, 24 animals were included for analysis: group CC, SSN, BBN, BBS, n = 5 each, group SSS, n = 4.
Fos-IR cells were most abundant on the ipsilateral side, mostly in Laminae I, II, V, and VI. The mean numbers of labeled cells in the different laminae are presented in Table 1 and visualized in Figures 1A and B.
Low basal Fos expression can be observed in CC animals, in which the distinction of ipsilateral/contralateral could not be made.
The unstimulated animals (SSN and BBN) showed some increase in Fos labeled cells compared with CC animals in Laminae I/II (not significant), and a decrease in Fos-IR in Laminae III and IV, which reached significance in Lamina III (SSN: P = 0.037; BBN: P = 0.025).
Stimulated animals (SSS and BBS) showed a significant increase compared with animals in their corresponding unstimulated groups in almost all of Laminae I-VII (SSS: P < 0.05 in Laminae I-V and VII; BBS: P < 0.05 in Laminae I-VI).
Bupivacaine block in the stimulated animals (group BBS versus SSS) caused a significant increase in Lamina V (P = 0.004). When comparing both unstimulated groups (BBN versus SSN), the bupivacaine block as such did not cause any differences.
Figure 2 shows the dorsal horn activation pattern at level L4 for each of the experimental groups.
On the side contralateral to the noxious stimulus, small amounts of cells with Fos expression were observed in all experimental groups. The only significant differences were in Laminae III and IV, where all experimental groups exhibited decreased amounts of Fos labeled cells when compared with CC animals (Lamina III: P < 0.05 in all groups; Lamina IV: P = 0.014 in SSN).
The noxious electric stimulation in the present experiment resulted in a general and strong increase in Fos expression in the ipsilateral lumbar spinal cord, mainly in Laminae I, II, and V. This is in full agreement with data from the literature (for review see Ref. 13). The presence of the thoracic block caused a remarkable increase in the amount of Fos-IR after stimulation in Lamina V, but had no effect on Fos-IR in Laminae I and II. It is not likely that this increase was caused by the injected bupivacaine itself, because no significant differences were found in Fos-IR comparing the two unstimulated groups (SSN and BBN).
These findings suggest the existence of supraspinal mechanisms differentially affecting the amount of Fos-IR in Laminae I/II versus Lamina V. Supraspinal mechanisms controlling lumbar dorsal horn activity have been described extensively. Mostly known is the circuitry involving the periaqueductal gray (PAG) and descending serotonergic pathways14; this has been described as inhibiting neuronal activity in the superficial laminae.15 The present data, however, do not show any effect on Laminae I/II, but strongly suggest a specific effect on the Lamina V-neurons. This raises the question: has any descending modulatory system been described affecting neuronal activity in Lamina V versus Laminae I/II in a specific and different way?
Such a system has been described, indeed, descending from the anterior pretectal nucleus (APtN), as a nonserotonergic pain modulating pathway.16–20 Mayer et al.21 were the first to observe that electrical stimulation of the APtN caused antinociception. In further studies, it was demonstrated that stimulation of the APtN not only inhibits the noxious stimulus-induced excitation of deep neurons but also excites Laminae I and II neurons.18 Despite the fact that many aspects of APtN pain modulation need further classification, we conclude that differential modulatory effects of the APtN in combination with the inhibitory effects from the PAG form the best hypothesis for the observed differences in Lamina V versus Laminae I/II effects of bupivacaine-induced thoracic block. Further experiments by means of double staining, for example, are needed to differentiate the role of the Fos-expressing neurons in this pain-modulating pathway.
A few other studies have addressed the same question by blocking the spinal-supraspinal communication using another method.11,22,23 In these studies, the dorsolateral funiculus (DLF) of rats was transected unilaterally, followed by formalin injections into both hindpaws. The lesioned side of the spinal cord exhibited more Fos-IR in both the superficial and the deep laminae,11 after which the authors suggested that an inhibitory pathway descending in the DLF affected both superficial and deep laminae. This result differs from our present results, in which we observed only in Lamina V were there changes in the numbers of Fos-IR neurons. This could suggest that not all descending pain modulating pathways are running through the DLF.
However, there are many methodological differences between these studies and ours. One of the advantages of the lesioning method is that it allows paired comparisons. A drawback of this method is that the exact size and location of the lesion is difficult to examine, and it is not exactly clear which spinal-supraspinal connections are cut or which are still intact after DLF lesioning. A major difference is that by performing a thoracic spinal block, all spinal-supraspinal communication, both ascending and descending, is blocked, whereas it is not possible to be certain that the DLF lesioning method will not block all ascending communication and will block all descending communication. Therefore, a straightforward comparison of the results from these different methods is not possible.
The presence of an intrathecal catheter apparently does not influence Fos-IR patterns at the level we investigated (L4) or the effects had disappeared at the time of these experiments, because in the CC group only low levels of Fos-IR were observed. These were similar to the Fos-IR in rats without an intrathecal catheter in earlier studies (data not shown). An intrathecal catheter has been previously used in Fos experiments.24–27 Tong et al.26 also found no difference in Fos immunostaining in animals with or without an intrathecal catheter. Luo et al.24 described an increase in Fos-IR in the lumbar spinal cord. However, in their experiment the catheter reached the lumbar level, and the Fos experiments were done 5-7 days after catheter implantation. These methodological differences could have caused the different results in our experiment. Our results suggest that, despite upper spinal cord damage, potentially induced by the catheter, well-controlled experiments aimed at differences in distribution of Fos-IR at the lumbar level are fully feasible.
The method we used for cutting and staining resulted in free-floating sections, making it impossible to distinguish the original left and right side. Therefore, we had to redefine ipsilateral and contralateral. This is, however, the standard method for side differentiation after this cutting technique, because most Fos is expressed ipsilateral to the stimulus.
The stimulus used in NIWR experiments is an electrical stimulus. Although electrical stimulation in general results in a nonspecific excitation of different types of nerve fibers, it is possible to distinguish between c-fos activation from noxious versus nonnoxious afferent activation. This distinction is related to the fact that in normal adult rats, maximal c-fos activation is induced when stimulation intensity is at Aδ/C-fiber strength; electrical stimulation at Aα/Aβ strength will only induce minimal c-fos activation.1–3 Therefore, we are convinced that most of the measured Fos in this study resulted from noxious Aδ/C-fiber stimulation.
Summarizing the present findings in the lumbar spinal cord, we conclude that bupivacaine block in combination with stimulation caused a significant increase in Fos-IR in Lamina V, whereas the block exerted no effect on Fos expression in Laminae I and II.
The peripheral noxious stimulation in the present experiment possibly activated both supraspinal modulation originating from the PAG and from the APtN. To confirm this hypothesis, further experiments are required, in which one of the two systems is specifically eliminated.
Our experimental design using a bupivacaine-induced thoracic block enabled the demonstration of supraspinal effects on nociception-induced lumbar Fos-IR. The effects, however, were not similar to studies in which the DLF was transected. Although we cannot explain the exact mechanism by which these methods produce different effects, we find the bupivacaine block preferable because it does not produce Fos-IR by itself and seems less invasive than transection of the spinal cord. In any case, Fos appears to be a useful tool to investigate different analgesic effects.
We thank Mrs. F. van der Pol for her biotechnical assistance and Mr. P. J. Dederen for his assistance with the immunohistochemical procedures.
1. Herdegen T, Kovary K, Leah J, Bravo R. Specific temporal and spatial distribution of JUN, FOS, and KROX-24 proteins in spinal neurons following noxious transsynaptic stimulation. J Comp Neurol 1991;313:178–91
2. Herdegen T, Leah JD. Inducible and constitutive transcription factors in the mammalian nervous system: control of gene expression by Jun, Fos and Krox, and CREB/ATF proteins. Brain Res Brain Res Rev 1998;28:370–490
3. Hunt SP, Pini A, Evan G. Induction of c-fos-like protein in spinal cord neurons following sensory stimulation. Nature 1987;328:632–4
4. Harris JA. Using c-fos as a neural marker of pain. Brain Res Bull 1998;45:1–8
5. Sommers MG, Nguyen NK, Veening JG, Vissers KC, Ritskes-Hoitinga M, Van Egmond J. Suppression of noxious-induced c-fos expression in the rat lumbar spinal cord by isoflurane alone or combined with fentanyl. Anesth Analg 2008;106:1303–8
6. Dirksen R, Lerou J, Lagerwerf A, Vree T, Nijhuis GM, Booij LH, Jurna I. A small animal model for pharmacological studies of general anesthetic agents. Eur J Anaesthesiol 1990;7:285–98
7. Duysens J, Gybels J. The automated measurement of hindlimb flexor reflex of the rat as a substitute for the tail-flick assay. J Neurosci Methods 1988;24:73–9
8. Malkmus SA, Yaksh TL. Intrathecal catheterization and drug delivery in the rat. Methods Mol Med 2004;99:109–21
9. Yaksh TL, Rudy TA. Chronic catheterization of the spinal subarachnoid space. Physiol Behav 1976;17:1031–6
10. Zeng X, Huang H, Hong Y. Effects of intrathecal BAM22 on noxious stimulus-evoked c-fos expression in the rat spinal dorsal horn. Brain Res 2004;1028:170–9
11. Zhang RX, Wang R, Chen JY, Qiao JT. Effects of descending inhibitory systems on the c-Fos expression in the rat spinal cord during formalin-induced noxious stimulation. Neuroscience 1994;58:299–304
12. Paxinos G, Watson C. The rat brain in stereotaxic coordinates. 5th ed. San Diego: Elsevier Academic Press, 2005
13. Coggeshall RE. Fos, nociception and the dorsal horn. Prog Neurobiol 2005;77:299–352
14. Mantyh PW, Hunt SP. Setting the tone: superficial dorsal horn projection neurons regulate pain sensitivity. Trends Neurosci 2004;27:582–4
15. Rohde DS, McKay WR, Chang DS, Abbadie C, Basbaum AI. The contribution of supraspinal, peripheral and intrinsic spinal circuits to the pattern and magnitude of Fos-like immunoreactivity in the lumbar spinal cord of the rat withdrawing from morphine. Neuroscience 1997;80:599–612
16. Rees H, Roberts MH. Anterior pretectal stimulation alters the responses of spinal dorsal horn neurones to cutaneous stimulation in the rat. J Physiol 1987;385:415–36
17. Rees H, Roberts MH. The anterior pretectal nucleus: a proposed role in sensory processing. Pain 1993;53:121–35
18. Rees H, Terenzi MG, Roberts MH. Anterior pretectal nucleus facilitation of superficial dorsal horn neurones and modulation of deafferentation pain in the rat. J Physiol 1995;489(Part 1):159–69
19. Villarreal CF, Del Bel EA, Prado WA. Involvement of the anterior pretectal nucleus in the control of persistent pain: a behavioral and c-Fos expression study in the rat. Pain 2003;103:163–74
20. Zagon A, Terenzi MG, Roberts MH. Direct projections from the anterior pretectal nucleus to the ventral medulla oblongata in rats. Neuroscience 1995;65:253–72
21. Mayer DJ, Wolfle TL, Akil H, Carder B, Liebeskind JC. Analgesia from electrical stimulation in the brainstem of the rat. Science 1971;174:1351–4
22. Wang H, Wang R, Nie H, Zhang R, Qiao JT. Neurokinin A, calcitonin gene-related peptide, and dynorphin A (1-8) in spinal dorsal horn contribute to descending inhibition evoked by nociceptive afferent pathways: an immunocytochemical study. Regul Pept 2000;89:7–12
23. Zhang RX, Mi ZP, Xie YF, Qiao JT. Morphological evidence for the activation of descending modulatory control by nociceptive afferent pathways: an immunocytochemical study. Brain Res 1993;603:162–5
24. Luo L, Ji RR, Zhang Q, Iadarola MJ, Hokfelt T, Wiesenfeld-Hallin Z. Effect of administration of high dose intrathecal clonidine or morphine prior to sciatic nerve section on c-Fos expression in rat lumbar spinal cord. Neuroscience 1995;68:1219–27
25. Seki S, Erickson KA, Seki M, Nishizawa O, Igawa Y, Ogawa T, de Groat WC, Chancellor MB, Yoshimura N. Elimination of rat spinal neurons expressing neurokinin 1 receptors reduces bladder overactivity and spinal c-fos expression induced by bladder irritation. Am J Physiol Renal Physiol 2005;288:F466–F473
26. Tong C, Ma W, Shin SW, James RL, Eisenach JC. Uterine cervical distension induces cFos expression in deep dorsal horn neurons of the rat spinal cord. Anesthesiology 2003;99:205–11
27. Yang LC, Orendacova J, Wang V, Ishikawa T, Yaksh TL, Marsala M. Transient spinal cord ischemia in rat: the time course of spinal FOS protein expression and the effect of intraischemic hypothermia (27 degrees C). Cell Mol Neurobiol 2000;20:351–65