INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Persistent pain has a strong emotional component. Understanding the neurobiological basis of the emotional responses to pain is crucial for novel and improved therapeutic strategies. The amygdala complex is a medial temporal lobe brain structure, which is generally believed to be involved in the neural substrates of emotion. As part of the limbic system, the amygdala plays a key role in the complex affective and autonomic aspects of behavior, the evaluation of the emotional significance of sensory stimuli, emotional learning and memory, fear, anxiety and depression, and stress responses (Aggleton 1992; Cahill 1999; Davidson et al. 1999; Davis 1994, 1998; Gallagher and Chiba 1996; Gallagher and Schoenbaum 1999; LeDoux 1996; Maren 1999; Post et al. 1998; Rasia-Filho et al. 2000; Rogan and LeDoux 1996; Rolls 2000).

The amygdala is directly linked to nociceptive centers in the spinal cord and brain stem through the spino-ponto-amygdaloid pain-pathway from the pontine parabrachial area to the central nucleus of the amygdala (CeA) (see Bernard and Besson 1990; Bernard et al. 1996). This purely nociceptive pathway originates from lamina I neurons in the spinal cord and the trigeminal nucleus caudalis (Bernard and Bandler 1998; Bernard et al. 1996; Jasmin et al. 1997). Polymodal, including nociceptive, information reaches the amygdala from thalamic and cortical areas through connections with the lateral and basolateral amygdaloid nuclei, which then project to the CeA (Doron and LeDoux 1999; Herzog and Van Hoesen 1976; LeDoux et al. 1990; Li et al. 1996; Linke et al. 1999; Pitkänen et al. 1995, 1997; Savander et al. 1995; Shi and Cassell 1998; Shi and Davis 1999; Smith et al. 2000; Stefanacci et al. 1992). In addition, spinal neurons in the deep dorsal horn and/or the area around the central canal form monosynaptic connections with amygdala neurons and may provide sensory, including nociceptive, input to the amygdala (Burstein 1996; Newman et al. 1996; Wang et al. 1999).

The CeA is the common output nucleus for major amygdala functions. The CeA modulates various effector systems involved in the expression of affective and autonomic emotional responses through widespread, mainly reciprocal, connections with the forebrain and brain stem, including the bed nucleus of the stria terminalis, frontal cortical areas, hippocampus, septal nuclei, lateral hypothalamus, parabrachial area, solitary tract nucleus and brain stem areas involved in endogenous pain control (Aggleton 1992; Cassell et al. 1986; Gray 1993; Krettek and Price 1979; LeDoux et al. 1988; Maren 1996; Price and Amaral 1981).

An increasing body of evidence implicates the amygdala in the modulation of pain behavior and pain sensation. Electrical stimulation of the amygdala elicits vocalizations that are accompanied by emotional reactions in monkeys (Jürgens 1982; Jürgens et al. 1967). Conversely, lesions or temporary chemical inactivation of the amygdala, involving the CeA, decrease tonic pain responses (Manning 1998) and emotional pain reactions in tests that involve higher integrated responses (Borszcz 1999; Calvino et al. 1982; Charpentier 1967; Grijalva et al. 1990; Werka 1997), without affecting normal behavior or baseline nociceptive responses (Calvino et al. 1982; Charpentier 1967; Fox and Sorenson 1994; Grijalva et al. 1990; Helmstetter 1992; Helmstetter and Bellgowan 1993; Maier et al. 1993; Pavlovic et al. 1996; Tershner and Helmstetter 2000; Watkins et al. 1993, 1998).

The amygdala has also been implicated in endogenous pain control and opioid analgesia. Microinjections of opioids and opioid receptor agonists, enkephalinase inhibitors, neurotensin, and cholinergic agonists into the amygdala produce antinociceptive behavior that is, in part, mediated through the periaqueductal gray (PAG) (Al-Rodhan et al. 1990; Helmstetter et al. 1993, 1995, 1998; Herz et al. 1970; Kalivas et al. 1982; Klamt and Prado 1991; Olivera and Prado 1994; Pavlovic and Bodnar 1998; Pavlovic et al. 1996; Tershner and Helmstetter 2000; Valverde et al. 1996). Lesions of the amygdala, including the CeA, can attenuate morphine analgesia (Manning 1998; Manning and Maier 1995a,b) and inhibit certain forms of conditioned and stress-induced analgesia (Fox and Sorenson 1994; Helmstetter 1992; Helmstetter and Bellgowan 1993; Pavlovic et al. 1996; Watkins et al. 1993, 1998; Werka 1997). These data suggest that the amygdala plays a significant role in complex pain behavior that involves supraspinal systems and in endogenous pain control mechanisms.

The cellular processing of nociceptive information, particularly from deep tissue, in the amygdala is less clear. One previous study using extracellular recordings in anesthetized rats suggested that a majority (80%) of neurons in the lateral and capsular divisions of the CeA were exclusively or preferentially excited or inhibited by brief noxious stimulation of the skin (Bernard et al. 1992). In the present study, we addressed the processing of nociceptive information from deep tissue (joints and muscles) in individual CeA neurons. The analysis of response characteristics of different types of CeA neurons to brief noxious stimulation of deep tissue will provide a reference for the study of persistent and chronic forms of pain arising from deep tissue, such as arthritis or myositis. Our focus on CeA neurons with knee-joint input will allow a comparison of nociceptive processing in other parts of the nervous system, with articular afferents and spinal neurons with knee joint input, that have been well characterized in previous studies (Neugebauer and Schaible 1990; Neugebauer et al. 1989, 1993; Schaible and Schmidt 1988; Schaible et al. 1987).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal preparation and anesthesia

Adult male Sprague-Dawley rats (240-350 g) were anesthetized with pentobarbital sodium (50 mg/kg ip). A cannula was inserted into the trachea for artificial respiration and to measure end-tidal CO₂ levels; a catheter was placed in the jugular vein for continuous administration of anesthesia; the carotid artery was catheterized for blood pressure monitoring. Depth of anesthesia was assessed by regularly testing the corneal blink and hindpaw withdrawal reflexes and by continuously monitoring the end-tidal CO₂ levels (kept at 4.0 ± 0.2%) and arterial blood pressure (kept at 135 ± 5 mmHg). Core body temperature was measured with a rectal thermometer and maintained at 37°C by means of a homeothermic blanket system.

Animals were mounted in a stereotaxic frame, paralyzed with pancuronium (0.3-0.5 mg iv) and artificially ventilated (3-3.5 ml; 55-65 strokes/min). Constant levels of anesthesia and paralysis of the musculature were maintained by intravenous infusion of a mixture of pentobarbital sodium (50 mg) and pancuronium (5 mg) in 30 ml NaCl (at ~40 µl/min). A unilateral craniotomy was performed at the sutura fronto-parietalis level for the recording of amygdala neurons and at the sutura occipito-parietalis level for electrical stimulation in the lateral pontine parabrachial area, where the monosynaptic connections of the spino-ponto-amygdaloid pain pathway to the CeA originate (Bernard et al. 1996). The dura mater was opened and reflected; the pia mater was removed over the recording site to allow smooth insertion of the recording electrodes.

Electrophysiological recording and identification of amygdala neurons

Extracellular recordings were made from single neurons in the CeA with glass-insulated carbon filament electrodes (3-5 M) using the following stereotaxic coordinates (cf. Paxinos and Watson 1998): 1.6-3.2 mm caudal to bregma; 3.8-4.4 mm lateral to midline; depth of 7,000-9,000 µM. The recorded signals were amplified and displayed on analog and digital storage oscilloscopes. Signals were also fed into a window discriminator, whose output was processed by an interface (CED 1401) connected to a Pentium III PC. Spike2 software (CED, version 3) was used to create peristimulus rate histograms on-line and to store and analyze digital records of single-unit activity off-line. Spike size and configuration were continuously monitored on the storage oscilloscopes and with the use of Spike-2 software.

CeA neurons were orthodromically activated by electrical stimulation (square-wave current pulse, 50-500 µA, 150 µs) in the lateral pontine parabrachial area, where the monosynaptic connections of the spino-ponto-amygdaloid pathway to the CeA originate (Bernard et al. 1996). The stereotaxic coordinates of the monopolar stimulation electrode were: 1-2 mm rostral to the lambda and 2.2 mm lateral to midline at the depth of 7.3 mm. Once an individual CeA neuron was identified and its spike size optimized, we carefully searched for a receptive field in the knee joint(s) and determined size and threshold of its total receptive field in the deep tissue and skin.

Experimental protocol

Background activity was recorded for 10 min to calculate mean ± SE and 95% confidence intervals (CI), using Prism 3.0 software (GraphPad Software, San Diego, CA). Size and thresholds of the receptive fields in deep tissue and skin were mapped. Response thresholds for mechanical stimulation of the knee joint and other deep tissue (e.g., ankle joint and muscles) were determined as follows: mechanical stimuli of gradually increasing intensity (steps of 50 g/30 mm²) were applied to the deep tissue (joints and muscles) by means of a forceps with a force transducer, whose calibrated output was amplified and displayed in g on a LCD screen. The output signal was also fed into the CED interface and recorded on the Pentium III PC for on- and off-line analysis.

The mechanical threshold was defined as the minimum stimulus intensity that evoked an excitatory response [spike frequency higher than the upper 95% confidence interval (CI) of background activity] or an inhibitory response (spike frequency less than the lower 95% CI of background activity). The threshold stimulus intensity was then tested again three times to verify the presence of a response in 50% of trials. A neuron was classified as receiving input from deep tissue if careful stimulation of overlying skin evoked no response or a clearly distinct response from that produced by stimulation of the deep tissue. Stimulus-response relationships were measured by applying graded mechanical test stimuli of 500-3,000 g/30 mm² intensity in increments of 500 g/30 mm² (15-s duration each; 15-s intervals).

Cutaneous receptive fields were mapped using the following stimuli: brush (brushing the skin with a soft-hair artist's brush in a stereotyped manner), press (firm pressure using a large arterial clip to apply 1,005 g/8 mm², which is marginally painful when applied to the skin in humans), and pinch (using a small arterial clip to apply 2,660 g/4 mm², which is clearly painful without causing overt damage to the skin). The most responsive site of the receptive field was then stimulated using a series of von Frey monofilaments with bending forces ranging from 60 mg to 178.5 g to measure stimulus-responses relationships. Each filament was applied repeatedly for a period of 15 s followed by a 15-s pause.

Thermal stimuli of innocuous and noxious intensity (37-53°C) were applied by a feedback-controlled contact Peltier thermode with an active area of 36 mm². Adapting temperature was set to 35°C; cycles of 5-s stimuli were delivered at intervals of 35 s. The temperature at the thermode was continuously measured and, with the use of the CED interface, recorded on the Pentium III PC for on- and off-line analysis of the stimulus-responses relationships.

Heterosensory (visual and auditory) input was also tested by shining a bright light into each pupil, snapping fingers, clapping hands, and whistling.

Classification of neurons and thresholds

CeA neurons were classified as nociceptive specific (NS), multireceptive, low threshold (LT), inhibited (INH), or nonresponsive neurons. The classification was primarily based on the neurons' responses to mechanical stimulation of the knee joint and other deep tissue although the responses to cutaneous and other natural somesthetic stimuli (see preceding text) were also characterized. A multireceptive neuron consistently responded to low-intensity stimuli (deep tissue, <500 g/30 mm²; skin, <1.3 g von Frey filament and/or brush) but was more strongly activated by noxious stimuli (deep tissue, >1,500 g/30 mm²; skin, >4.8 g von Frey filament, press, pinch). A stimulus intensity of 500 g/30 mm² applied to the knee and other deep tissue was considered innocuous; it did not evoke hindlimb withdrawal reflexes in awake rats (unpublished observations) and was not felt painful when tested on the experimenters. Pressure stimuli >1,500 g/30 mm² applied to the knee joint and other deep tissue were considered noxious; they evoked hindlimb withdrawal reflexes in awake rats (unpublished observations) and were felt distinctly painful when applied to the experimenters. A neuron that exclusively responded to noxious stimuli was called a NS neuron. LT neurons responded to innocuous stimuli without increasing the magnitude of their responses for noxious stimuli. INH neurons were inhibited by mechanical stimuli. Nonresponsive neurons were not activated by any mechanical or thermal stimuli.

Histology

At the end of each experiment the recording site in the CeA was marked by injecting DC current (250 µA for 3 min) through the carbon filament recording electrode. The brain was removed and submerged in 10% formalin and potassium ferrocyanide. Tissues were stored in 20% sucrose before they were frozen sectioned at 50 µM. Recording sites were later identified histologically and plotted on standard diagrams (from Paxinos and Watson 1998) of coronal brain sections (see Fig. 1).

View larger version (41K): [in this window] [in a new window]   Fig. 1. Localization of 61 amygdala neurons with knee-joint input included in this study. The medial, lateral, and lateral capsular subdivisions of the central nucleus of the amygdala (CeM, CeL, and CeC, respectively) are shown in the diagrams B-F, which are enlargements of coronal rat brain sections like the one in A (from Paxinos and Watson 1998). Numbers indicate caudal distance from bregma in mm. Histologically identified (see METHODS) recording sites of the different types of neurons with knee-joint input are represented by different symbols as indicated. Also included are 15 neurons recorded at different depths in the same electrode tracks as the neurons whose recording sites were actually lesioned (n = 46). NS, nociceptive specific (n = 18); multireceptive (n = 30); LT, low threshold (n = 3); INH, inhibited (n = 10).

Data analysis

Recorded activity was analyzed off-line from peristimulus rate histograms using Spike2 software (CED, version 3). The neurons' responses to mechanical and thermal stimuli were measured and expressed as spikes per second (Hz). Background activity, if present, was subtracted from the evoked responses. Stimulus-response relationships for mechanical and thermal inputs were measured for each neuron and then averaged across a sample of neurons. Stimulus-response functions were analyzed using models of linear and nonlinear regression (Prism 3.0, GraphPad Software). Sigmoid curves were fitted to the stimulus-response data using the following "four parameter logistic equation" for nonlinear regression (Prism 3.0, GraphPad Software): y = A + (B  A)/[1 + (10^C/10^X)^D], where A = bottom plateau, B = top plateau, C = log(half-maximal intensity), and D = slope coefficient. The following F test (Prism 3.0 software) was used to compare the fits of the linear and nonlinear models and to determine which equation was more appropriate: F = [(SS1 - SS2)/(DF1 - DF2)]/(SS2/DF2), where SS1 and SS2 = sum or squares of the linear and nonlinear model, respectively; DF1 and DF2 = degrees of freedom for the linear and nonlinear model, respectively. Stimulus-response functions of NS and multireceptive neurons were compared using a two-way ANOVA (Prism 3.0 software). Means of half-maximal stimulus intensities were compared using a two-tailed unpaired t-test with Welch correction (Instat 3.1, GraphPad Software). Means of mechanical and thermal thresholds of NS and multireceptive neurons were compared using a one-way ANOVA followed by Newman-Keuls multiple comparison test (Prism 3.0 software). All averaged values are given as the means ± SE. Statistical significance was accepted at the level P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Extracellular single-unit recordings were made from 119 neurons in the CeA in 46 anesthetized rats. Most neurons were recorded in the posterior portion of the CeA (2.2-3.2 mm caudal to bregma) and, particularly, in the lateral capsular subdivision (see Fig. 1 and insets in Figs. 3-5) (nomenclature according to Paxinos and Watson 1998). The CeA neurons responded to orthodromic electrical stimulation in the lateral pontine parabrachial area with latencies ranging from 6.3 to 16.5 ms (11 ± 0.3 ms; see Fig. 2D and examples in Figs. 3 and 4). The mean threshold for orthodromic activation was 340 ± 30 µA (50-500 µA; 150 µs). No apparent differences were found in the distribution of latencies and electrical thresholds for the individual types of CeA neurons. Ongoing activity was present in most of the neurons (98/113) and ranged from 0.4 to 21.5 Hz (mean = 4.8 ± 2.1 Hz). Excitatory or inhibitory input from the knee joint was detected in 77 neurons. None of the CeA neurons with knee joint input responded to visual or auditory stimuli (see METHODS).

Classification of nociceptive CeA neurons with knee joint input

EXCITATORY INPUT. Sixty-two neurons in the CeA received excitatory input from the knee joint; 58 of these neurons responded preferentially or exclusively to noxious stimulation. Primarily based on their responses to mechanical stimulation of the knee joint and other deep tissue (see METHODS), 25 neurons were NS, i.e., activated exclusively by noxious stimulation of the deep tissue and, in some cases, the skin (see following text, Receptive fields); 33 neurons were classified as multireceptive neurons responding to innocuous but more strongly to noxious stimuli. NS and multireceptive CeA neurons had significantly different response thresholds for mechanical stimulation of the knee joint and the skin and for thermal stimulation (Fig. 2, A-C; P < 0.001, Newman-Keuls multiple comparison test, see METHODS; 2C). Only four neurons were activated preferentially by innocuous stimuli and were classified as LT neurons.

View larger version (33K): [in this window] [in a new window]   Fig. 2. Functional properties of amygdala neurons. Response thresholds of different types of amygdala neurons with knee-joint input to mechanical stimulation of the knee joint (A) and skin (B) and to thermal stimulation (C). Thresholds were measured as described in METHODS. Statistical analysis of the mechanical and thermal response thresholds of NS and multireceptive neurons showed they were significantly different (A-C, P < 0.001, Newman-Keuls multiple comparison test). NS, nociceptive specific neurons (A, n = 25; B, n = 8; C, n = 7); multireceptive neurons (A, n = 33; B, n = 17; C, n = 14); LT (A, n = 4; B, n = 4; C, n = 4); INH (A, n = 15; B, n = 6; C, n = 9). D: latencies following electrical stimulation in the lateral pontine parabrachial area (see METHODS) for all 119 CeA neurons. No apparent differences were found in the distribution of latencies for the individual types of CeA neurons. Latencies were measured from records like those shown in Figs. 3, B and C, and 4, B and C. Bin width = 0.5 ms.

View larger version (36K): [in this window] [in a new window]   Fig. 3. Nociceptive specific (NS) neuron recorded extracellularly in the CeA of the right hemisphere. A: noxious, but not innocuous, mechanical stimulation (15-s duration) of the left knee joint evoked responses; binwidth of histogram: 1 s. Intensities of the applied pressure stimuli were recorded on-line (top; see METHODS). B: action potential evoked orthodromically by electrical stimulation (450 µA; 150 µs) in the lateral parabrachial area. C: latencies of the evoked responses ranged from 5-7 ms (histogram shows 50 sweeps); *, stimulus artifacts. D: the neuron's receptive field was symmetrical and located in the deep tissue only (). E: coronal section shows histologically identified (see METHODS) recording site in the lateral capsular part of the CeA.

View larger version (33K): [in this window] [in a new window]   Fig. 4. Multireceptive neuron recorded extracellularly in the right CeA. A: noxious mechanical stimulation (15-s duration) of the left knee joint excited the neuron more strongly than innocuous pressure stimuli; binwidth of histogram: 1 s. Top: stimulus intensities recorded on-line (see METHODS). B: action potential evoked orthodromically by electrical stimulation (250 µA; 150 µs) in the lateral parabrachial area. C: latencies of the evoked responses ranged from 14 to 16 ms (histogram shows 50 sweeps); *, stimulus artifacts. D: the large receptive field was symmetrical and located in the deep tissue of all 4 extremities () and in the skin of the trunk (). E: coronal section shows histologically identified (see METHODS) recording site in the lateral capsular part of the CeA.

INHIBITORY INPUT. Another 15 neurons were inhibited by noxious mechanical stimulation of the knee joint and other deep tissue. A typical example of an INH neuron is shown in Fig. 5. Noxious mechanical stimulation (Fig. 5A, 1,500-3,000 g/30 mm²; 15-s duration) of the left knee joint inhibited the background activity of this neuron. The inhibitory receptive field in the deep tissue was symmetrical and located in all four extremities; a cutaneous inhibitory receptive field was detected in the dorsal pelvis area (Fig. 5B). Histological analysis (see METHODS) identified the recording site on the border between the lateral and lateral capsular subdivisions of the CeA.

View larger version (44K): [in this window] [in a new window]   Fig. 5. INH neuron recorded extracellularly in the right CeA. A: noxious mechanical stimulation (15-s duration) of the left knee joint inhibited the background activity of this neuron; binwidth of histogram: 1 s. Top: stimulus intensities recorded on-line (see METHODS). Bottom: corresponding recording of original spikes. B: the inhibitory receptive field in the deep tissue () was symmetrical and located of all 4 extremities; there was also an inhibitory receptive field in the skin over the dorsal pelvis (). C: coronal section shows histologically identified (see METHODS) recording site in the area of the lateral and lateral capsular divisions of the CeA.

CEA NEURONS WITHOUT KNEE-JOINT INPUT. A group of 15 neurons had no receptive field in the knee but responded to noxious stimulation of other parts of the body. Another 27 nonresponsive neurons were not activated by any mechanical or thermal stimuli.

Receptive fields

CeA neurons with excitatory nociceptive input from the knee joint (n = 58) typically had large symmetrical receptive fields in the deep tissue of both hindlimbs (n = 19) or in all four extremities (n = 36). Only rarely was the receptive field confined to one knee joint (n = 3). Noxious mechanical stimulation of the skin excited 30 of these neurons (9 NS neurons and 21 multireceptive neurons); 17 multireceptive neurons were also weakly activated by brush (see METHODS). Noxious heat activated 21 of 41 CeA neurons (7 of 18 NS neurons and 14 of 23 multireceptive neurons).

Similarly, the 15 INH neurons with knee joint input typically had large receptive fields in the deep tissue of the hindlimbs or in all four extremities. Noxious cutaneous stimuli inhibited the background activity in 6 of 15 INH neurons. A response to thermal stimulation was not detected in any of nine INH neurons tested.

Figure 6 summarizes the various sizes and arrangements of receptive fields in the deep tissue and skin for NS neurons (Fig. 6A), multireceptive neurons (Fig. 6B), and INH neurons (Fig. 6C). The percentages of NS, multireceptive, and INH neurons with receptive fields in all four extremities were 52, 70, and 67%, respectively. Receptive fields confined to the hindlimb(s) were found in 48% of NS neurons, 30% of multireceptive neurons, and 33% of INH neurons.

View larger version (34K): [in this window] [in a new window]   Fig. 6. Size and arrangement of the receptive fields of NS neurons (A, n = 25), multireceptive neurons (B, n = 33), and INH neurons (C, n = 15). Innocuous and noxious somesthetic stimuli (see METHODS) were used to map the receptive fields in the deep tissue () and skin (). Typical receptive field configurations are shown and the percentage of neurons in each category is given below the drawings. Receptive fields of individual neurons showed slight variations. The recording sites in relation to the receptive fields are also indicated. Overall, the majority of NS, multireceptive and INH neurons had large symmetrical receptive fields in all four extremities (A-C, 2 right columns). Only NS neurons were occasionally found to have a receptive field restricted to the knee-joint area.

Stimulus-responses relationships

MECHANO-NOCICEPTIVE INPUT. A sigmoid nonlinear regression model rather than a monotonically increasing linear function best described stimulus-response curves for graded mechanical stimulation of the deep tissue and the skin. Figure 7 shows the stimulus-response functions of CeA neurons that responded exclusively (NS neurons, ) or more strongly (multireceptive neurons, ) to noxious than innocuous mechanical stimulation of the deep tissue (7A, pressure stimuli applied to the knee joint by a calibrated forceps) and the skin (7B, stimulation of the skin of the lower back with von Frey monofilaments, see METHODS). An F test (see METHODS) revealed that the nonlinear sigmoid regression model yielded a significantly better fit than a linear equation [7A, NS neurons, n = 25, F(2,2) = 31.90, P < 0.05; multireceptive neurons, n = 33, F(2,2) = 41.54, P < 0.05; 7B, NS neurons, n = 8, F(2,5) = 125.0, P < 0.0001; multireceptive neurons, n = 17, P < 0.005, F(2,5) = 23.26]. The sigmoid stimulus-response function with its bottom and top plateaus emphasizes the two distinct levels of stimulus encoding, i.e., innocuous and noxious intensities. The plateau and shallow slope of the stimulus-response curves for high-intensity stimuli may suggest limited intensity encoding at the higher end of the noxious stimulation range (Fig. 7A, >2,000 g/30 mm²; Fig. 7B, >12 g).

View larger version (17K): [in this window] [in a new window]   Fig. 7. Stimulus-responses relationships of CeA neurons with excitatory input from the knee joint for nociceptive mechanical input from the deep tissue (A, NS neurons, , n = 25; multireceptive neurons, , n = 33) and from the skin (B, NS neurons, n = 8; multireceptive neurons, n = 17). A: for mechanical stimulation of the knee joint a calibrated forceps was used to apply pressure of the indicated intensities. B: graded mechanical stimuli were applied to the skin of the lower back using a series of calibrated von Frey monofilaments (see METHODS). The neurons' responses to mechanical stimuli were measured over the 15-s stimulation period at each stimulus intensity (see METHODS) and were used to calculate the frequency of the evoked response (expressed as spikes per second). Background activity, if present, was subtracted from the total activity during stimulus application. The responses of individual NS and multireceptive neurons to different stimulus intensities were averaged across a sample of neurons. Sigmoid curves were fitted to the stimulus-response data using the following formula for nonlinear regression that provided the best fit (Prism 3.0, GraphPad Software; see METHODS and RESULTS): y = A + (B  A)/[1 + (10C/10X)D], where A = bottom plateau, B = top plateau, C = log(half-maximal intensity), D = slope coefficient. All averaged values are given as the means ± SE (see METHODS and RESULTS for statistical analysis).

THERMO-NOCICEPTIVE INPUT. Similar to the processing of mechano-nociceptive information, the stimulus-response relationship of CeA neurons for thermal stimuli followed a sigmoid nonlinear regression model rather than a monotonically increasing linear function. Figure 8A shows the responses of an individual multireceptive neuron recorded in the lateral capsular part of the CeA. The receptive field of this neuron is illustrated in Fig. 8B. Thermal stimuli in the innocuous range evoked clear responses (see Fig. 8B for location of the stimulation site), but noxious heat (49-53°C) activated this neuron more strongly. Figure 8C summarizes the stimulus-response data of CeA neurons that responded exclusively (NS neurons, , n = 7) or predominantly (multireceptive neurons, , n = 14) to thermal stimuli in the noxious range. An F test (see METHODS) revealed that the nonlinear sigmoid regression model yielded a significantly better fit than a linear equation (NS neurons, n = 7, F(2,8) = 69.63, P < 0.0001; multireceptive neurons, n = 14, F(2,8) = 8.97, P < 0.01). The sigmoid shape of the stimulus-responses curves suggests two distinct levels of stimulus encoding in the innocuous and noxious range, respectively. Like in the case of mechano-nociceptive processing, the stimulus-response curves for thermal heat stimuli (Fig. 8C, >49°C) reach a plateau, which may suggest limited intensity encoding in the high noxious heat range (>50°C).

View larger version (30K): [in this window] [in a new window]   Fig. 8. Processing of nociceptive thermal information in CeA neurons with excitatory input from the knee joint. A: responses of 1 multireceptive neuron in the lateral capsular part of the CeA to graded thermal stimuli applied to the skin of the lower back; binwidth: 1 s. B: receptive field of the same neuron and site of thermal stimulation. C: stimulus-responses functions of CeA neurons (NS neurons, , n = 7; multireceptive neurons, , n = 14). The neurons' responses to thermal stimuli were measured over the 5-s stimulation period at each stimulus intensity (see METHODS) and were used to calculate the frequency of the evoked response (expressed as spikes per second). Background activity, if present, was subtracted from the total activity during stimulation. The responses of individual NS and multireceptive neurons to different stimulus intensities were averaged across a sample of neurons. Sigmoid curves were fitted to the stimulus-response data using the following formula for nonlinear regression that provided the best fit (Prism 3.0, GraphPad Software; see METHODS and RESULTS): y = A + (B - A)/[1 + (10C/10X)D], where A = bottom plateau, B = top plateau, C = log (half-maximal intensity), D = slope coefficient. All averaged values are given as the means ± SE (see METHODS and RESULTS for statistical analysis).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study is the first to address the processing of nociceptive information in CeA neurons with input from the knee joint(s). The major findings are as follows. 1) A large proportion of CeA neurons receive input from the knee joint (65%). 2) Excitation is the predominant effect of brief painful stimulation of somatic tissue in CeA neurons with knee joint input (62 of 77 neurons). 3) The high percentage of multireceptive neurons (53%) suggests that a substantial population of CeA neurons receive and process innocuous information in addition to purely nociceptive input from the lateral parabrachial area (see Bernard and Besson 1990; Bernard et al. 1994), electrical stimulation of which activated the neurons in the present study. And 4) NS and multireceptive CeA neurons have large symmetrical receptive fields and sigmoid rather than monotonically increasing linear stimulus-response functions, suggesting a role in other than sensory-discriminative aspects of pain.

Nociceptive processing in individual amygdala neurons has only been addressed in two previous electrophysiological studies, which focused on CeA neurons with cutaneous receptive fields and found them to be very large (Bernard et al. 1990, 1992). In the present study, we found a high percentage of CeA neurons with knee-joint input (65%), which also had large symmetrical receptive fields in the limbs and the trunk (see Fig. 6). CeA neurons with primarily deep tissue input could be classified as: excited neurons of the NS, multireceptive and LT types; INH neurons; and nonresponsive neurons without a detectable receptive field.

NS, LT, INH, and nonresponsive CeA neurons with deep tissue input in this study may have corresponding counterparts in amygdala neurons with predominantly cutaneous input described previously (see Bernard et al. 1990, 1992). The classification of multireceptive neurons, however, is more complex and deserves further consideration. Multireceptive CeA neurons with knee-joint input may or may not correspond to "wide-dynamic-range" (WDR) amygdala neurons with predominantly cutaneous inputs (Bernard et al. 1990, 1992). Previous studies of CeA neurons used the term WDR neurons to describe neurons "which were preferentially activated by noxious stimuli" but "were also driven to a lesser extent by innocuous stimuli (pressure or in a few cases touch and/or warm water <44°C)" (Bernard et al. 1992). Innocuous pressure applied with a calibrated forceps was 3-5 N/cm², whereas noxious pinch and squeeze were 6-9 and 10-18 N/cm², respectively (Bernard et al. 1992). It would appear that these values are somewhat lower than the stimulus intensities required for activation of multireceptive CeA neurons in the present study. Applying graded pressure stimuli with a calibrated forceps we found that the thresholds for innocuous mechanical stimulation of the knee joint were 256 ± 41.0 g/30 mm² (mean ± SE, n = 33; see Fig. 2 and METHODS). A clear and consistent response was evoked in multireceptive neurons by stimulus intensities of 500 g/30 mm² (see Fig. 7A), which is a relatively firm, albeit innocuous, stimulus when tested on the experimenters. Pressure stimuli >1,500 g/30 mm² applied to the knee joint were considered noxious for the following reasons: pressure stimuli of 1,500 g/30 mm² consistently evoked hindlimb withdrawal reflexes in awake rats (unpublished observations; see METHODS); when applied to the experimenters, this stimulus intensity was felt distinctly painful; this intensity correlates well with the threshold for activation of NS neurons (1,680.0 ± 49.0 g/30 mm², mean ± SE, n = 25; see Fig. 2); it is also close to the stimulus intensity for half-maximal responses of NS neurons (1,702 ± 50.3 g/30 mm², n = 25) and multireceptive neurons (1,642 ± 43.9 g/30 mm², n = 33; see RESULTS and see Fig. 7) to noxious mechanical stimuli.

These findings and the fact that only half of the multireceptive CeA neurons with knee joint input showed a response to brushing the skin (see Receptive fields in RESULTS), may distinguish them from classical WDR neurons. Originally, the term "WDR neuron" referred to a subset of spinal neurons with cutaneous inputs ranging from low to high threshold (see Willis 1985; Willis and Coggeshall 1991). We have also applied this term to spinal neurons with knee-joint input in our previous studies (see Neugebauer and Schaible 1990; Neugebauer et al. 1993, 1996). There we used the term WDR neurons to describe neurons with knee-joint input that responded to a "wider range" of stimulus intensities (including innocuous pressure) than NS neurons that responded exclusively to noxious stimulus intensities. Innocuous mechanical stimuli that activated spinal WDR neurons with knee-joint input (190-300 g/40 mm²) (Neugebauer et al. 1994-1996) are comparable with the mechanical thresholds of multireceptive CeA neurons (~250 g/30 mm²) but somewhat lower than the innocuous test stimuli that clearly and consistently activated multireceptive CeA neurons (500/30 mm²) in the present study. It should be noted that mechanical thresholds have not been reported explicitly for amygdala WDR neurons with cutaneous input (Bernard et al. 1990, 1992) or spinal WDR neurons with knee joint input (Neugebauer et al. 1993-1996).

Therefore in the present study, which is the first to characterize amygdala neurons with knee-joint input, we selected the term "multireceptive" neurons as proposed by a Working Party on Terminology (see Brown and Rethelyi 1981) to describe neurons with input from more than one category, i.e., mechanoreceptive and nociceptive. Functionally, multireceptive CeA neurons may be positioned between WDR and NS neurons or may be a variant of either type. This distinction in terminology will also allow for the possibility that WDR CeA neurons with predominant input from the skin play different roles than multireceptive CeA neurons primarily identified by deep tissue input, particularly from the knee joint.

Interestingly in the present study, the majority (81%) of CeA neurons with knee-joint input were excited rather than inhibited by peripheral stimuli compared with only ~58% of CeA neurons with mainly cutaneous input in a previous study (see Bernard et al. 1992). Furthermore, we found a larger proportion of multireceptive neurons (57%) than NS neurons (43%) among the nociceptive CeA neurons with excitatory input from the knee joint(s), whereas more CeA neurons with primarily cutaneous input (see Bernard et al. 1992) were NS type neurons (75%) than WDR neurons (25%). These differences may in part be explained by differences in classification and terminology (see preceding text). They may also suggest that nociceptive CeA neurons with major cutaneous input and CeA neurons with receptive fields predominantly in the deep tissue represent two distinct populations and that nociceptive information from deep tissue and skin is processed differently in the CeA. Alternatively, differences in the sampling techniques may account for the larger number of excited neurons recorded in this study. Differently than the other studies (Bernard et al. 1990, 1992), we used orthodromic electrical stimulation in the lateral parabrachial area (see METHODS) to identify CeA neurons that belonged to the spino-parabrachio-amygdaloid pain pathway (Bernard and Bandler 1998; Bernard et al. 1996; Jasmin et al. 1997). A previous study (Bernard et al. 1994) described a large percentage of excited (87%) nociceptive neurons in the parabrachial area, which is strikingly close to the 81% excited neurons in our study.

Finally, we cannot rule out that differences in the types and percentages of CeA neurons between this study and the previous ones (Bernard et al. 1990, 1992) may be due to different anesthetics used (barbiturate versus halothane/nitrous oxide) and/or different levels of anesthesia achieved in the two studies. We are confident that a stable level of anesthesia was achieved with the continuous intravenous infusion of pentobarbital and maintained throughout the experiment by monitoring and tightly regulating heart rate, blood pressure, body temperature, and CO₂ levels. In addition, in experiments where several neurons were identified and characterized, we sampled the different neuron types (NS, multireceptive, LT, INH, nonresponsive) in random order but never observed a neuron changing from one type to another, e.g., NS neurons becoming multireceptive neurons.

The considerable number of multireceptive CeA neurons (53%) with input from the knee joint input and from the parabrachial area in this study may have functional implications. Neurons in the parabrachial area are either NS neurons, which do not respond to innocuous stimuli, or nonresponsive neurons, which are not activated at all by somatosensory and visceral stimuli (Bernard et al. 1994, 1996). Therefore multireceptive neurons in the CeA, which respond to noxious and certain innocuous stimuli, are likely to receive additional inputs that provide nonnociceptive information. Highly integrated multimodal sensory information from thalamic and cortical areas reaches the CeA through the lateral and basolateral amygdaloid nuclei (see Doron and LeDoux 1999; LeDoux et al. 1990; Li et al. 1996; Linke et al. 1999; Pitkänen et al. 1995, 1997; Savander et al. 1995; Shi and Cassell 1998; Shi and Davis 1999; Smith et al. 2000; Stefanacci et al. 1992). Those multimodal sensory inputs and the purely nociceptive inputs from the lateral parabrachial area may then converge onto multireceptive neurons in the CeA. The integration of nociceptive and nonnociceptive information by multireceptive neurons may play an important role in the emotional evaluation of sensory stimuli, a key function of the amygdala (Aggleton 1992; Davis 1994, 1998; Gallagher and Chiba 1996; Gallagher and Schoenbaum 1999; LeDoux 1996; Maren 1999; Rasia-Filho et al. 2000; Rogan and LeDoux 1996). The role of NS neurons may consist in attaching the label "nociceptive" to events occurring in the CeA in conjunction with a noxious stimulus. The INH and nonresponsive CeA neurons could play important roles in neuroplastic changes induced in chronic pain states, e.g., through reduced inhibition/disinhibition of INH neurons and recruitment of unresponsive neurons.

Consistent with a role of the CeA in other than sensory-discriminative aspects of pain is the observation in our study that NS and multireceptive CeA neurons had large symmetrical receptive fields and sigmoid rather than monotonically increasing linear stimulus-response functions. The stimulus-response data for mechanical stimulation of deep tissue and skin and for thermal stimuli were fitted best to a sigmoid nonlinear regression model. The bottom and top plateaus of the sigmoid curve emphasize the ability of CeA neurons to identify and encode the nociceptive character of various types of afferent information. Interestingly, the poor intensity coding in the high noxious range would also be consistent with a role of the CeA in other than sensory-discriminative components of pain. In agreement with our data, a previous study identified different slopes of the stimulus-response relationships of CeA neurons for thermal stimuli (Bernard et al. 1992). The steepest portion of the slope was between 46 and 48°C, which correlates very well with the half-maximal stimulus intensity calculated from the stimulus-response curves in the present study (47°C). The other study also measured shallow slopes for stimulus intensities in the range of 40-44 and 50-52°C (25), which resembles the bottom and top plateaus of the sigmoid stimulus-response curves in our study.

In summary, this study is the first to analyze receptive fields and encoding properties of CeA neurons with nociceptive input from the knee joint(s). Our data show that nociceptive CeA neurons are able to encode nociceptive information but do so in a fashion that suggests a role in other than sensory-discriminative aspects of pain. Furthermore a substantial proportion of CeA neurons (multireceptive neurons) may not only receive purely nociceptive input from the spino-parabrachio-amygdaloid pain pathway but also process nonnociceptive information that reaches the CeA most likely through multimodal sensory inputs from thalamic and cortical areas.