Anterior parietal cortical response to tactile and skin-heating stimuli

JOURNALOFNEUROPHYSIOLOGY RAPID PUBLICATION Vol. 75, No. 6, June 1996. Printed in U.S.A.

Anterior Parietal Cortical Response to Tactile and Skin-Heating Stimuli Applied to the Same Skin Site

M. TOMMERDAHL, K. A. DELEMOS, C. J. VIERCK, JR., 0. V. FAVOROV, AND B. L. WHITSEL Departments of Biomedical Engineering and Physiology, and Dental Research Center, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599; and Department of Neuroscience, University of Florida, Gainesville, Florida 32610

SUMMARY AND CONCLUSIONS INTRODUCTION

1. The response of anterior parietal cortex to skin stimuli was evaluated with optical intrinsic signal imaging and extracellular microelectrode recording methods in anesthetized squirrel monkeys.

2. Nonnoxious mechanical stimulation (vibrotactile or skin tapping) of the contralateral radial interdigital pad was accompanied by a decrease in reflectance (at 833 nm) in sectors of cytoarchitectonic areas 3b and 1. This intrinsic signal was in register with regions shown by previous receptive field mapping studies to receive low-threshold mechanoreceptor input from the radial interdigital pad.

3. A skin-heating stimulus applied to the contralateral radial interdigital pad with a stationary probe/thermode evoked no dis- cernable intrinsic signal in areas 3b and 1, but evoked a signal within a circumscribed part of area 3a. The region of area 3a responsive to skin heating with the stationary probe/thermode was adjacent to the areas 3b and 1 regions that developed an intrinsic signal in response to vibrotactile stimulation of the same skin site. Skin heating with a stationary probe/thermode also evoked intrinsic signal in regions of areas 4 and 2 neighboring the area 3b/ 1 regions activated by vibrotactile stimulation of the contralateral radial interdigital pad.

The idea that anterior parietal cortex consists of two centrally located fields (cytoarchitectonic areas 3b and 1) for the initial processing of input from skin receptors, and two bordering fields (areas 3a and 2) devoted to processing of input from receptors in muscle and joints, has a long history, traceable to the pioneering single-neuron recording studies of Mountcastle (1957) and Powell and Mountcastle ( 1959b). Although the response of anterior parietal cortex to diverse forms of noxious stimulation has been studied repeatedly with the use of neurophysiological recording methods (for comprehensive review see Kenshalo and Willis 199 1) , the evidence obtained in those studies has not led to a universally accepted view of how anterior parietal cortex represents stimuli that activate nociceptors.

4. The intrinsic signal evoked in area 3a by a series of heating stimuli to the contralateral radial interdigital pad (applied with a stationary probe/ thermode) increased progressively in magnitude with repeated stimulation (exhibited slow temporal summation) and remained above prestimulus levels for a prolonged period after termination of repetitive stimulation.

5. Brief mechanical stimuli ( “taps” ) applied to the contralat- era1 radial interdigital pad with a probe/ thermode maintained either at 37OC or at 52OC were accompanied by the development of an intrinsic signal in both area 3a and areas 3b/ 1. For the 52°C stimulus, the area 3a intrinsic signal was larger and the intrinsic signal in areas 3b/l smaller than the corresponding signals evoked by the 37°C stimulus.

6. Spike discharge activity was recorded from area 3a neurons during a repetitive heating stimulus applied with a stationary probe/ thermode to the contralateral radial interdigital pad. Like the area 3a intrinsic signal elicited by repetitive heating of the same skin site, the area 3a neuron spike discharge activity also exhibited slow temporal summation and poststimulus response persistence.

7. The experimental findings suggest 1) a leading role for area 3a in the anterior parietal cortical processing of skin-heating stimuli, and 2) the presence of inhibitory interactions between the anterior parietal responses to painful and vibrotactile stimuli consistent with those demonstrated in recent cortical imaging and psychophysical studies of human subjects.

The availability of approaches for imaging the optical response of large neural networks to afferent drive (Grinvald 199 1, 1994) makes it possible to experimentally evaluate the assumption that areas 3b and 1 play the “leading” role in the anterior parietal cortical representation of all stimuli that activate skin receptors (e.g., gentle mechanical as well as thermal stimuli). With this goal in mind we imaged in anesthetized squirrel monkeys the intrinsic signal evoked in areas 3a, 3b, 1, and 2 in response to stimulation of the same skin site, with the use of three different modes of precisely controlled stimuli: 1) a 25Hz vibrotactile stimulus, 2) a heating stimulus applied with the use of a probe/thermode in stationary contact with the skin, and 3) a thermal-mechanical (T-M) stimulus (brief skin contact- “tap” -with a probe / thermode maintained at a specified temperature) that even at temperatures ~45°C evokes pain only when applied repeatedly at relatively short interstimulus intervals (typically <3 s) (Price 1977; Vierck and Whitsel, unpublished observation). In addition, we used extracellular microelectrode recording methods to assess, in the same subjects that provided optical imaging results, the effects on anterior pari- eta1 neuron spike discharge activity of skin temperature increases delivered with the use of a stationary probe/thermode (stimulus mode 2 above). A novel aspect of the neurophysiological phase of the experiments was that we directed our microelectrode penetrations to sample neurons in the same cortical regions that in the immediately preceding optical imaging phase of the experiment had demonstrated a prominent intrinsic signal in response to skin-heating stimulation.

In this paper we present findings that indicate that area

2662 0022-3077/96 $5.00 Copyright 0 1996 The American Physiological Society

AREA 3A IN NOCICEPTION 2663

3a, and not areas 3b and 1, plays a leading role in the anterior parietal cortical processing of skin-heating stimuli. The view of somatosensory cortical information processing in squirrel monkeys suggested by the experimental findings incorpo- rates recently recognized features of thalamocortical and corticocortical connectional organization, and predicts inten- sity-dependent and time-dependent interactions between anterior parietal regions that respond differentially to nonnoxious mechanical and thermal skin stimulation-interactions leading to outcomes consistent not only with the observations described in this paper, but also with those reported in recent cortical imaging and psychophysical studies of the responses of human subjects to vibrotactile and noxious skin-heating stimuli ( Apkarian 1995; Apkarian et al. 1994; Derbyshire et al. 1996).

METHODS

Subjects and preparation

Adult squirrel monkeys (Saimiri sciureus; males and females; y1 = 6) were studied because in this species one can directly visual- ize most of the anterior parietal region that receives input from the contralateral hand. In addition, because areas 3b and 1 in squirrel monkey have been investigated with the use of both microelectrode recording (Sur et al. 1982) and optical imaging methods (Tommer- dahl and Whitsel 1996), a wealth of information exists about the response of anterior parietal cortex of these animals to nonnoxious mechanical skin stimulation and its sensitivity to general anesthe- tics.

Preparation involved 1) induction of general anesthesia, 2) inser- tion of a soft tube into the trachea and polyethylene cannula into the femoral vein, 3) making a small opening in the skull, 4) instal- lation with dental acrylic of a recording chamber over the skull opening, and 5) removal of the dura overlying anterior parietal cortex. The surgical procedures were carried out under deep general anesthesia ( l-4% halothane in a 50-50 mixture of oxygen and nitrous oxide), and all wound margins were infiltrated with a long- lasting local anesthetic, closed with sutures, and bandaged. Throughout the recording session, subjects were immobilized with Norcuron and ventilated with a gas mixture (a 50-50 mix of oxygen and nitrous oxide; supplemented with O.l- 1 .O% halothane when necessary) with a positive pressure respirator. Respirator rate and volume were adjusted to maintain end-tidal CO, between 3.0 and 4.0%, and electroencephalogram and autonomic signs (slow wave content; heart rate and pupillary signs) were monitored and titrated by adjustments in the anesthetic gas mixture to maintain values consistent with light general anesthesia. Rectal temperature was monitored and maintained at 37.5OC with the use of a heating pad.

Euthanasia was accomplished by intravenous injection of pento- barbital sodium (45 mg/kg) and intracardial perfusion with saline followed by fixative ( 10% Formalin). Fiducial marks were placed to guide removal, blocking, and subsequent histological sectioning of the region of cortex studied. All procedures involving animal subjects were reviewed and approved in advance by an institutional committee and are in full compliance with current National Insti- tutes of Health policy on animal welfare.

Stimulation/stimulus control procedures and protocols

All stimuli were applied to the contralateral radial interdigital pad. Sinusoidal vertical displacement (vibrotactile) stimuli were applied with the use of a servocontrolled transducer (Cantek Enter- prises, Canonsburg, PA). Parameters of vibrotactile stimulation used were as follows: frequency was 25 Hz, peak-to-peak ampli- tude was 400 pm, duration was 7 s, and intertrial interval was 35

s. The stimulator made contact with the skin via a 2-mm-diam plastic probe that, in the absence of vibrotactile stimulation, indented the skin 500 pm.

Skin-heating stimuli were applied with the use of a cylindrical probe that carried a 12.5-mm-diam Peltier-type device (thermode) mounted to its end. Two different approaches to delivery of skin- heating stimuli were used. In the first, the probe/thermode was positioned so that it indented the skin by 500 ,um; it remained stationary throughout the period when triangle-shaped changes ( “ramps” ) in thermode temperature were delivered with the use of a servocontroller. In the second approach the thermode was maintained at a preselected temperature, and contact between thermode and skin was controlled by a solenoid, i.e., in the rest position the thermode/probe was 5 mm above skin contact and solenoid activation caused the probe to move rapidly (in - 100 ms) from the rest position to a point where it indented the skin by l-2 mm. Solenoid inactivation led to equally rapid probe retraction to the rest position. This second mode for application of skin-heating stimuli is referred to as “T-M tapping.”

The perceptual effects of the stimulus conditions, protocols, and apparatus used to deliver stationary skin-heating and T-M tapping stimuli to the radial interdigital pad of the hands of monkeys were evaluated (with the use of a magnitude estimation procedure) in preliminary psychophysical studies in which the investigators and their colleagues were subjects. Only stimulus conditions or protocols that evoked sensations that could be tolerated by the investigators (conditions/protocols that did not elicit escape from the heating stimulus) and were not associated with skin damage were employed to study the response of anterior parietal cortex in anesthetized squirrel monkeys.

Cortical intrinsic signal imaging

Near-infrared (IR; 833 nm) cortical intrinsic signal imaging was carried out with the use of an oil-filled chamber capped with an optical window (Tommerdahl and Whitsel 1996). Images of the exposed anterior parietal cortical surface were acquired 200 ms before stimulus onset (reference images) and every 1.4 s thereafter (poststimulus images) for 28 s after stimulus onset. Exposure time was 200 ms. Difference images were generated by subtracting each prestimulus image from its corresponding poststimulus image. Averaged difference images typically show regions of both increased absorption of IR light (decreased reflectance) and decreased absorption of light (increased reflectance) that, on the basis of the findings of others (e.g., Grinvald 1985; Grinvald et al. 1991, 1994; for a recent comprehensive review see Ebner and Chen 1995)) are accompanied by increases and decreases in neuronal activation, respectively. Thus dark regions shown in the difference images in this paper (such as in Fig. 1) indicate regions of increased neuronal activity and light regions indicate decreased neuronal activity. Difference images of the anterior parietal response to stimulation of the same skin site with nonnoxious mechanical (skin tapping) and skin-heating stimulation (temperature ramps and/or T-M taps) routinely were obtained in the same experimental run. The different stimulus conditions applied with the use of the solenoid-controlled probe/ thermode were interleaved (alternated) in the same run to control for temporal changes in cortical “state” unrelated to stimulus conditions that, if unrecognized, might intro- duce or obscure differences in the IR reflectance patterns evoked by different conditions of skin stimulation.

Neurophysiological recording

Extracellular recordings of the spike discharge activity of single anterior parietal neurons and local neuron populations were obtained with the use of electrolytically etched, glass-insulated tung- sten wires (impedance 300-500 kSn at a test frequency of 10 kHz).

2664 TOMMERDAHL, DELEMOS, VIERCK, FAVOROV, AND WHITSEL

1 mm

Subject #Z 1 3b 3a 1 3b 3a

1 mm

FIG. 1. Average infrared (IR) reflectance change (“response”) evoked by 2 different conditions of skin stimulation in two subjects. Left: vascular patterns. CS, central sulcus. Middle: average poststimulus-prestimulus difference images (120 trials; 1 stimulus/trial; 35 s intertrial interval) obtained with the use of 7-s, 25-Hz, 400~km vibrotactile stimulus applied with the use of a 2-mm probe advanced 500 pm into me skin. Images were generated by averaging the differences between the prestimulus image (obtained 200 ms before stimulus onset) and poststimulus images acquired between 2 and 8 s after stimulus onset. Right: average difference images obtained with the use of a skin-heating ramp applied with the thermode advanced 500 pm into the skin (ramp parameters: temperature increased from 36 to 52°C and then decreased back to 36°C; rates of temperature increase and decrease 19”/s), 1 stimulus/trial; 180 s intertrial interval; 15 trials. Images were generated by averaging the differences between prestimulus image and poststimulus images acquired between 2 and 8 s after thermal stimulus onset. Lines on images: cytoarchitectonic boundaries. The pixel values for the averaged difference images obtained with the use of different conditions of skin stimulation in each subject were normalized and resealed to maximally reflect within-subject differences in intrinsic signal magnitude evoked by different stimuli.

Microelectrode penetrations were performed subsequent to the op- optical imaging and neurophysiological recording methods. The tical imaging phase of the experiments under closed-chamber con- maps display the cytoarchitectonic boundaries as they appear from ditions-the recording chamber was filled with artificial cerebro- the cortical surface. The locations of microelectrode tracks and spinal fluid and covered with a glass plate containing an “0’‘-ring electrolytic lesions in the cortical sections were projected radially through which the microelectrode could be advanced under direct to the pial surface and transferred to the map of cytoarchitectonic visual control. Microelectrode penetrations were inserted at cortical boundaries reconstructed from the same sections. As the final step, sites distinguished on the basis of the magnitude and direction the cytoarchitectonic boundaries (along with the locations of mi- (increase or decrease) of the IR reflectance change associated with croelectrode tracks and lesions whenever present) identified in either vibrotactile or skin-heating stimulation. Conventional ap- each subject were mapped onto the images of the stimulus-evoked proaches and apparatus were used to amplify, filter, display, store. intrinsic signal obtained from the same subject, with the use of and analyze records (both analog and digital) of neuronal spike fiducial marks as well as morphological landmarks (e.g., blood trains collected before, during, and after application of the vibrotac- vessels and sulci evident both in optical images and in histological tile and skin-heating stimuli. Stimulus conditions and protocols sections). To ensure that deternnnations of the locations of cytoar- similar to those used to evoke changes in IR reflectance were chitectonic boundaries were uninfluenced by information about the employed to study the effects of skin stimulation on the spike location of intrinsic signal, the experimental identity of the sections discharge activity of cortical neurons. Electrolytic lesions in cortex was concealed from the two investigators (Whitsel and Favorov) were created by passing direct current through the recording micro- who identified the boundaries, and the relationship between cytoar- electrode. Such microlesions were placed at the terminal locations chitecture and the stimulus-evoked IR reflectance patterns for a of microelectrode tracks and also at sites where neurophysiological subject was not evaluated until those investigators had committed recordings of particular interest were obtained. to a final map of cytoarchitectonic boundaries for that subject.

Histological procedures/generation of maps relating cytoarchitectural boundaries to functional observations RESULTS

Cortical blocks were postfixed, cryoprotected, frozen, and sec- tioned serially in the sagittal plane at 30 pm. Sections were stained with cresyl fast violet and inspected microscopically to distinguish anterior parietal regions on the basis of established cytoarchitectonic criteria (Powell and Mountcastle 1959a; also Jones and Porter 1980). The boundaries between adjacent cytoarchitectonic areas were identified and plotted with the use of a microscope with a drawing tube, and the plots used to reconstruct a two-dimensional map of cytoarchitectonic boundaries within the region studied with

Four squirrel monkeys were studied using only the IR imaging approach. These first studies provided information about the effects of systematic variations in the parameters of thermal skin stimulation on the optical response of anterior parietal cortex. Both neurophysiological recordings and IR imaging observations were obtained from the other two subjects. This rapid publication presents only those findings that bear directly on the major question that motivated the


experiments, i.e., do areas 3b and 1 play the principal role in the processing and representation of skin-heating stimuli?

IR imaging observations

LOCUS OF INTRINSIC SIGNAL EVOKED BY VIBROTACTILE VER- SUS SKIN-HEATING STIMULATION. Figure 1 shows IR difference images obtained under two different stimulus conditions in two animals. These images are representative of those obtained from four subjects, each studied with the use of the same two conditions of skin stimulation. The imaged cortical region (cortical surface is shown in Fig. 1, left) included extensive portions of both the precentral and postcentral gyri at the lateral end of the central sulcus (CS) . The average poststimulus-prestimulus difference images in Fig. 1, middle, show the pattern of average IR reflectance change associated with delivery of a 25Hz, 400+m-ampli- tude vibrotactile stimulus to a 2-mm site located centrally on the contralateral radial interdigital pad. Note that in both subjects 1 and 2 there is a single prominent region of reflectance decrease -in the former it occupies neighboring parts of areas 3b and 1, in the latter it occupies the posterior part of area 3b. For both subjects 1 and 2, the region of decreased reflectance associated with vibrotactile stimulation of the contralateral radial interdigital pad is, to a first approx- imation, centered on the area 3b territory identified as the representation of the radial interdigital pad in the map of forelimb skin regions generated from receptive field mapping observations (Sur et al. 1982). In each subject; however, the region occupied by the stimulus-evoked decrease in IR reflectance is substantially larger ( lo-50 times) (Tom- merdahl et al. 1994, 1996; Tommerdahl and Whitsel 1996) than the zone indicated as the radial interdigital pad representation in the map of Sur et al. ( 1982).

In the two subjects that provided the difference images shown in Fig. 1, middle, exposure of the same skin locus to a ramp-shaped change in the temperature imposed with a stationary thermode (from 36 to 52°C and then back to 36°C) led to an IR reflectance decrease in a location very different from the region occupied by the intrinsic signal evoked by vibrotactile stimulation. In both subjects I and 2 (as in all subjects studied) the intrinsic signal associated with the skin-heating stimulus occupied a region anterior and medial to the location of the signal evoked by vibrotactile stimulation (see Fig. 1, right). In subjects 1 and 2 (as in all 6 subjects studied with vibrotactile and skin-heating stimuli), the principal anterior parietal region that responded to skin heating lies within cytoarchitectural area 3a.

In four of the six subjects the skin-heating stimulus also evoked (in addition to the response in area 3a) a prominent intrinsic signal within a sector of area 4, and weaker signals at both a posterior location in the vicinity of the area l-2 boundary (e.g., Fig. 1, top right) and laterally in the upper bank of the lateral sulcus. The intrinsic signal that appeared in the upper bank of the lateral sulcus in response to skin heating was located outside the fields imaged in Fig. 1.

In summary, the IR imaging data obtained from all six subjects reveal that 1) regions centered on, but more extensive than, the sectors of areas 3b and 1 known to receive their principal input from the contralateral radial interdigital pad respond with an intrinsic signal to 25Hz vibrotactile

stimulation of the contralateral interdigital pad, 2) areas 3b and 1 do not contribute significantly to the anterior parietal intrinsic signal observed when the same skin site is exposed to a transient heating stimulus perceived as noxious by con- scious human subjects, and 3) the most prominent anterior parietal intrinsic signal that accompanies transient heating of the skin of radial interdigital pad occurs in a circumscribed region of area 3a located anterior and medial to the region ( s ) of areas 3b and 1 that yielded an intrinsic signal in response to 25-Hz vibrotactile stimulation.

INTRINSIC SIGNALS EVOKED BY THERMAL VERSUS NONNOX- IOUS MECHANICAL SKIN STIMULATION: EVIDENCE FOR IN- TERAREAL INTERACTIONS? The dependency of the locus of the anterior parietal intrinsic signal on conditions of skin stimulation was evaluated. The four different conditions of T-M stimulation of the radial interdigital pad selected for study can be ordered in terms of the relative strengths of the mechanoreceptor afferent (MRA) and thermoreceptor afferent drive they evoke. Specifically, the 25-Hz vibrotactile stimulus (delivered with the probe at room temperature 27°C) is regarded as approximating a strong, “pure mechanical” skin stimulus that vigorously activates both the rapidly adapting (RA) and the slowly adapting (SA) classes of skin MRAs in the neighborhood of the stimulator probe. Because the probe of the vibrotactile stimulator is at or near to resting skin temperature, this stimulus presumably evokes little or no change in the activity of skin thermoreceptors. In contrast, skin heating (from 36 to 52°C and back to 36°C) with the stationary probe/thermode is regarded as a strong and “predominantly thermal’ ’ skin stimulus that, although it evokes substantial skin warming receptor and heat-nociceptor activation, along with some SA-type MRA activity due to probe/ thermode contact with the skin, elicits little or no RA-type MRA activity because the probe/thermode device was stationary. By the same rationale, a skin tap with the solenoid- driven probe/thermode at 36°C is considered a “moderately strong mechanical/ weak thermal’ ’ stimulus because it effectively elicits both RA and SA skin MRA spike discharge activity, as well as relatively weak skin thermoreceptor activity, whereas a skin tap applied with the probe/thermode at 52°C is a “moderately strong mechanical/strong thermal’ ’ skin stimulus because it not only effectively activates both SA and RA MRAs, but produces vigorous activation of both the C fiber and A delta afferent populations that innervate the warming and heat-sensitive nociceptors in the vicinity of the stimulator probe.

Figure 2 shows IR difference images obtained from the same subject (subject 2) for the four different stimulus conditions described above. Figure 2, top left, shows that in this subject the vibrotactile stimulus evoked intrinsic signal virtually exclusively in area 3b. In contrast, skin taps with the temperature of the probe/thermode at 36°C resulted in the appearance of intrinsic signal in area 3a and also in area 3b (Fig. 2, top right); and taps with the probe/thermode at 52°C led the intrinsic signal in area 3a to become even stronger relative to the signal evoked in area 3b (Fig. 2, bottom left). Finally, when the stationary probe was used to apply a skin heating ramp (the predominantly thermal stimulus condition), the intrinsic signal that was evoked was con- fined to areas 3a and 4 (Fig. 2, bottom right). An interesting

2666 TOMMERDAHL. DELEMOS. VIERCK, FAVOROV, AND WHITSEL

1 3b 3a 1 3b 3a

0 5

0 E e >

1 mm

FIG. 2. Anterior parietal IR response to 4 different stimulus conditions in the same subject. All stimuli were delivered to the same site on the contralateral interdigital pad. Top left: response to 25.Hz vibrotactile stimulation. The probe/thermode of the vibrotactile stimulator was at room temperature (27°C). Top right: response to 1.2-s skin taps applied with solenoid-controlled probe/thermode at 36°C. Bottom left: response to 1.2-s skin taps applied with solenoid-controlled probeithermode at 52°C. Bottom right: response to 1.2-s skin-heating ramps (from 36 to 52°C and then back to 36°C) applied with probe/thermode in stationary contact with skin.

aspect of the results shown in Fig. 2 is that the response of area 3b to the 36°C taps is much larger than its response to the 52°C taps, even though the mechanical component of the tapping stimulus was identical for the two conditions. Each of the three other subjects studied in this way yielded a similar result.

The variation in the magnitude of the area 3b/l intrinsic signal that accompanied the change in probe/thermode temperature is interpreted to indicate that the area 3b/ 1 response to the nonnoxious mechanical component of skin stimulation and the area 3a response to heating the same skin site are not independent. Instead, they appear to interact so that a T-M skin stimulus whose thermal component evokes substantial activity in area 3a suppresses the response of areas 3b/l to the component of thalamocortical excitatory drive attributable to skin MRA activitation.

TIME COURSE OF INTRINSIC SIGNAL EVOKED BY SKIN HEAT- ING. Information about the detailed time course of the development of the intrinsic signal evoked by heating stimulation of the same skin site was obtained in two subjects. The stationary probe/thermode, advanced 500 pm into the skin, was used to deliver series of skin-heating stimuli to the radial interdigital pad. Each series, or trial, lasted 13.2 s and consisted of a succession of triangle-shaped, 1.2-s ramps of thermode temperature from 36 to 52°C and back to 36°C. The ramps were repeated six times per trial, one ramp every 2.4 s. Multiple IR optical images of the anterior parietal cortex were taken immediately preceeding, during, and after each series of six ramps in order to track the temporal development of the IR reflectance change during stimulus application. Ten such series of six heating stimuli were delivered in each experiment, with 180 s of waiting time (the intertrial interval) between successive series.

Figure 3 shows a representative time series of IR difference images, each an average of the difference images obtained in 10 stimulus series. Figure 3 shows that the intrinsic cortical signal developed very slowly over the period during which the skin was subjected to six applications of the 52°C heating stimulus. For example, the signal is barely evident at 2.6 s after stimulus onset (after 1 stimulus), is still poorly developed at 4 s after onset (after 2 stimuli), and only reaches its maximum 8.2 s after the onset of stimulation (between the 3rd and 4th heating stimuli). It also is evident that after termination of the stimulus series the intrinsic signal decays very gradually (it can be seen even at 18 s after stimulus onset-4.8 s after the last heating stimulus).

Observations obtained in microelectrode penetrations

The slow temporal summation characteristic of the intrinsic signal evoked by the skin-heating stimulus suggested that the neuroelectrical events this same stimulus evokes in area 3a (and in area 4) might also exhibit this property. This prediction was evaluated in two subjects in combined experiments in which extracellular single (n = 12) and multiunit recordings (n = 10) were carried out subsequent to the acquisition of IR imaging data.

The data in Fig. 4 were obtained from a representative neuron in area 3a isolated 650 pm below the cortical surface (near layer IV) during a microelectrode penetration inserted at a site within the region that responded with a large intrinsic signal to a heating stimulus applied to the contralateral interdigital pad. For this area 3a neuron, both the spike trains recorded during each skin-heating stimulus (the data for 10 stimulus presentations, labeled l- 10, are shown in Fig. 4, top; the triangular skin-heating stimulus is shown immediately under the spike trains) and the stimulus-by-stimulus plot of average rate of spike discharge within the entire 40- s stimulus-nostimulus period (OMFR, Fig. 4, bottom) reveal appreciable slow temporal summation, as well as a prominent persistence of the response after stimulus termination. Specifically, the responses of this area 3a neuron to the first two to three skin-heating stimuli are extremely weak. With continued stimulation, however, the neuron’s average spike discharge activity increased progressively during the period of skin heating, as well as in the interval between the end of one stimulus and the beginning of the next.

DISCUSSION

The findings lead us to conclude that of the four anterior parietal cortical cytoarchitectural fields, it is not areas 3b or 1 but area 3a that plays the most prominent role in the processing of information from distal forelimb skin receptors sensitive to elevations in skin temperature perceived as painful. Our conclusion is based on two observations. The first is that a skin-heating stimulus evokes activity (visualized as an intrinsic signal-a decrease in reflectance-with the use of the IR imaging approach) in cytoarchitectural area 3a but not in areas 3b and 1. The second, reinforcing observation is that both the area 3a stimulus-evoked intrinsic signal and the spike discharge activity of single area 3a neurons increase progressively during repetitive applications of a brief skin-heating stimulus. Temporal summation of both the area

8.2

AREA 3A IN NOCICEPTION

2.6

12.4 16.6 18.0

2667

FIG. 3. Average IR difference images showing the responses of the same subject to a repetitive skin-heating stimulus delivered with the thermode in stationary contact with the skin (in each stimulus the thermode increased from 36 to 52°C and then returned to 36°C). Each panel shows the cortical response at a different time after onset of repetitive stimulation. Numbers located above or below panels indicate time (in s) after onset of repetitive stimulation. Solid lines on image at 25 s after stimulus onset: cytoarchitectonic boundaries. Dotted lines: area displayed in images of Figs. 1 and 2.

3a optical and neural response to repetitive skin heating appears consistent with the published reports that in normal human subjects the magnitude of pain evoked by a brief noxious skin-heating stimulus increases progressively if the stimulus is applied at rates equal to or greater than one every 3 s (this psychophysical phenomenon has been referred to as slow temporal summation; Price et al. 1977; for recent reviews of the perceptual effects of repeated skin-heating stimulation see Price 1991; Price et al. 1977, 1994).

Areas 3b and 1 frequently have been regarded as the anterior parietal areas that play the leading role in the processing of input from cutaneous receptors, and area 3a as an area devoted to processing of input from receptors located in skeletal muscle. The findings obtained in this study are in direct conflict with such a parcelation of anterior parietal sensory information processing. The findings are compatible, however, with multiple lines of experimental evidence sug- gesting that, at least for certain modes of skin stimulation (e.g., noxious skin heating), activity in cortical areas other than areas 3b and 1 appears more directly associated with the stimulus-evoked sensory experience. First, positron emission tomography and single photon emission computed tomography imaging studies in normal human subjects indicate that although a brief and only modestly painful noxious skin- heating stimulus strongly and reliably activates multiple neocortical regions, such a stimulus is accompanied by either relatively weak or no significant activation of a cortical territory (referred to as ‘ ‘SI” ) comprised principally of areas 3b and 1 (Apkarian 1995; Apkarian et al. 1992; Bushnell et al. 1995; Casey et al. 1994; Coghill et al. 1994; Derbyshire et al. 1994, 1996; Jones et al. 1991; Talbot et al. 1991). Second, in normal subjects a more prolonged (“tonic”) noxious skin-heating stimulus associated with modest to in-

tense pain is accompanied by frank inhibition of “SI,” and by elevation of the threshold for detection of vibrotactile stimuli (Apkarian 1995; Apkarian et al. 1992, 1994). Inter- estingly, decreases in “SI” and ventrobasal thalamic activity appear to be especially prominent in patients experiencing severe chronic pain (for recent reviews of clinical imaging studies see Apkarian et al. 1995; Derbyshire et al. 1996). Third, modern neuroanatomic tracing studies identify two distinct cortical termination patterns for the immunocyto- chemically distinct dorsal thalamic neurons that are the tar- gets of spinothalamic tract or caudal trigeminal nuclear projections (Craig 1991, 1996a,b; Craig and Kniffki 1996; Craig et al. 1994; Jones 1991; Kniffki 1989; Rausell and Jones 1991a,b; Rausell et al. 1992). Although the majority of thalamic neurons that receive input from the spinothalamic tract or caudal trigeminal nucleus project preferentially to layer I in areas 3b and 1 (a pattern unlike that of “typical” thalamocortical sensory projections, Rausell et al. 1992), others project selectively to layer IV in area 3a (Craig 1996a,b-a lamina IV termination pattern typical of ‘ ‘primary sensory” thalamocortical projections; White 1986, 1989). And fourth, the human clinical neurological evidence has been interpreted to indicate that a region buried deeply in the central sulcus (the position occupied by area 3a) might play a role in pain perception (Per1 1984, personal communication).

Taken together, the results of this study and the imaging observations obtained from human subjects (described in preceding paragraph) indicate very different roles for areas 3a and for areas 3b and 1 in the processing of information arising in skin receptors, i.e., whereas areas 3b and 1 play the leading role in the processing of input from low-threshold mechanoreceptors, area 3a serves as the primary anterior parietal field for the processing of information from thermo-

2668 TOMMERDAHL, DELEMOS, VIERCK, FAVOROV, AND WHITSEL

2::: 0 5 10 15 20 25 30 35 40

time (see)

4 I

0 5 10 15 2iJ 25 30 35 40 time (see)

I; 1 2 3 4 5 6 7 8 9 10

Presentation X

FIG. 4. Stimulus-evoked activity of neuron recorded in middle layers of region of area 3a that exhibited decreased IR reflectance in response to repetitive heating of the radial interdigital pad. Inset at top: difference image showing IR response to repetitive skin-heating stimulus and site (indicated by white arrow) at which recording microelectrode contacted cortical surface. Bottom 3 panels (from top to bottom): 1) spike trains recorded from the neuron during 10 ramps of skin temperature; the stimulus ramp is shown below the spike trains 2) peristimulus time histogram showing average spike discharge activity (OMFR) computed for all responses to the 10 stimuli; and 3) trend plot showing that average rate of spike discharge associated with each stimulus (computed for each 40 s stimulus-poststimulus period) increased progressively with repetitive stimulation.

receptors. Our data also suggest that although these same two anterior parietal territories (area 3a vs. areas 3b and 1) are distinguishable on the basis of the mode of skin stimulation that activates them most effectively (noxious heating is most effective for area 3a, whereas thermally neutral mechanical stimuli are best for areas 3b and 1 ), they do not operate independently-this suggestion derives from our

discovery that the IR response of areas 3b and 1 to a mechanical skin tapping stimulus is maximal when the stimulus is near to skin temperature, and becomes progressively weaker as the temperature of the stimulating probe is elevated relative to resting skin temperature. Both of these suggestions appear wholly consistent with the demonstrated inhibitory effect of tonic noxious skin heating on SI activity ( Apkarian et al. 1992)) and with the demonstration that pain “gates” human subjects’ capacity to detect vibrotactile stimuli (Apk- arian et al. 1995). Because it is area 3a that exhibits intrinsic signal in response to painful skin heating, and because layer IV has been shown to receive input from thalamic neurons receiving spinothalamic projections (Craig 1996a,b), we consider the spinothalamic pathway as the most likely con- veyor of the influences evoked in area 3a by skin heating.

Where along the neuraxis might such inhibitory effects arise? Although it cannot be denied that some inhibitory effects arise at subcortical levels (for discussion see Apk- arian et al. 1995), we favor the suggestion (Tommerdahl et al. 1996) that intense area 3a activation leads to an inhibition of areas 3b and 1 mediated by the long-distance horizontal connections that link these anterior parietal fields (Burton and Fabri 1995; DeFelipe et al. 1986). Although the long- range glutaminergic corticocortical connections that interre- late the different anterior parietal fields are excitatory in their synaptic action, it has only recently been realized that they exert predominantly inhibitory effects on their target cell columns whenever they are tonically and strongly active (see Tommerdahl et al. 1996 for full description of proposed mechanism; see also Hirsch and Gilbert 1991) . Our view is that corticocortical inhibitory effects evoked by painful skin heating could account, at least in part, for the finding (Apk- arian et al. 1994) that a tonic painful skin-heating stimulus elevates human vibrotactile detection threshold at the same and at neighboring skin sites. For example, if in normal subjects a painful thermal skin stimulus evokes thalamocortical drive that leads to strong layer IV neuron activation in area 3a (relative to that occurring simultaneously in areas 3b/ 1)) then the corticocortical interactions triggered by such a stimulus should suppress (for details see Tommerdahl et al. 1996) the response of areas 3b and 1 to vibrotactile stimulation.

Although the precise functional role of interactions among the different anterior parietal areas remains to be ascertained for any stimulus and experimental condition, the findings obtained in this study and the considerable evidence available in the published literature indicate that the stimulus sensitivity and receptive field properties of large anterior parietal neuron populations (and presumably of individual anterior parietal neurons) are not determined solely by the classes of thalamocortical afferents they receive (e.g., see Stemmler et al. 199.5; also Douglas et al. 1995; Singer 1995). Instead, they appear to be influenced in ways important for normal functioning by stimulus-directed dynamic interactions occurring among cytoarchitectonically distinguishable anterior parietal territories. From this perspective, the current tendency to regard the four cytoarchitectonically distinguishable areas in anterior parietal cortex as separate and function- ally independent entities seems both unjustified and unneces- sarily limiting.


The authors acknowledge the expert technical assistance provided by C. Metz (histological processing) and C. Wong (experimental support) ; they are also indebted to Dr. W. Maixner for making available to this study a thermal-mechanical skin stimulator.

The research was supported, in part, by National Institutes of Health Grants R29 NS-32358 to M. Tommerdahl, ROl MH-48654 to B. L. Whitsel and 0. Favorov, and PO1 DE-07509 to K. A. Delemos, and by Whitehall Foundation Grant S93-10 to M. Tommerdahl.

Address for reprint requests: M. Tommerdahl, 155 Medical Research Bldg. CB# 7545, Dept. of Physiology, University of North Carolina, Chapel Hill, NC 27599.

Received 7 December 1995; accepted in final form 23 February 1996.

REFERENCES

APKARIAN, A. V. Functional imaging of pain: new insights regarding the role of the cerebral cortex in human pain perception. Semin. Neurosci. 7: 279-293, 1995.

APKARIAN, A. V. AND SHI, T. Squirrel monkey lateral thalamus. I. Somatic nociresponsive neurons and their relation to spinothalamic terminals. J. Neurosci. 14: 6779-6795, 1994.

APKARIAN, A. V., STEA, R. A., AND BOLANOWSKI, S. J. Heat-induced pain diminishes vibrotactile perception: a touch gate. Somatosens. Mot. Res. 11: 259-267, 1994.

APKARIAN, A. V., STEA, R. A., MANGLOS, S. H., SZEVERENYI, N. M., KING, R. B., AND THOMAS, F. D. Persistent pain inhibits contralateral somatosensory cortical activity in humans. Neurosci. Lett. 140: 14 1 - 147, 1992.

BURTON, H. AND FABRI, M. Ipsilateral intracortical connections of physio- logically defined cutaneous representations in areas 3b and 1 of macaque monkeys: projections in the vicinity of the central sulcus. J. Comp. Neurol. 355: 508-538, 1995.

BUSHNELL, M. C., CRAIG, A. D., REIMAN, E. M., YUN, L.-S., AND EVANS, A. Cerebral activation in the human brain by pain, temperature, and an illusion of pain. Sot. Neurosci. Abstr. 2 1: 1637, 1995.

CASEY, K. L., MINOSHIMA, S., BERGER, K. L., KOEPPE, R. A., MORROW, T. J., AND FREY, K. A. Positron emission tomographic analysis of cerebral structures activated specifically by repetitive noxious heat stimuli. J. Neurophysiol. 71: 802-807, 1994.

COGHILL, R. C., TALBOT, J. D., EVANS, A. C., MEYER, E., GJEDDE, A., BUSHNELL, M. C., AND DUNCAN, G. H. Distributed processing of pain and vibration by the human brain. J. Neurosci. 14: 4095 -4108, 1994.

CRAIG, A. D. Supraspinal pathways and mechanisms relevant to central pain. In: Pain and Central Nervous System Disease: The Central Pain Syndromes, edited by K. Casey. New York: Raven, 1991, chapt. 12, p. 157-170.

CRAIG, A. D. Supraspinal projections of lamina I neurons. In: Forebrain Processing of Pain, edited by J.-M. Besson. Paris: Libbey, 1996a. In press.

CRAIG, A. D. Pain, temperature, and the sense of the body. In: Somesthesis and the Neurobiology of Somatosensory Cortex, edited by 0. Franzen. Boston, MA: Birkhauser, 1996b. In press.

CRAIG, A. D., BUSHNELL, M. C., ZANG, E.-T., AND BLOMQVIST, A. A thalamic nucleus specific for pain and temperature sensation. Nature Lond. 372: 770-773, 1994.

CRAIG, A. D. AND KNIFFKI, K. D. Spino-thalamo-cortical mechanisms of nociception. In: Current Trends In Pain Research and Therapy: Basic Mechanisms and Clinical Applications, edited by K. N. Sharma and U. Nayar. New Delhi. 1996, vol. I. In press.

DEFELIPE, J., CONLEY, M., AND JONES, E. G. Long-range focal collateraliza- tion of axons arising from corticocortical cells in monkey sensory-motor cortex. J. Neurosci. 6: 3749-3766, 1986.

DERBYSHIRE, S. W. G., CHEN, A. C. N., AND JONES, A. K. P. Somatosensory and frontal cortical processing during pain experience. In: Somesthesis and the Neurobiology of Somatosensory Cortex, edited by 0. Franzen. Boston, MA: Birkhauser, 1996. In press.

DERBYSHIRE, S. W. G., JONES, A. K. P., DEVANI, P., FRISTON, K. J., FEIN- MANN, C., HARRIS, M., PEARCE, S., WATSON, J. D. G., AND FRACKOWIAK, R. S. J. Cerebral responses to pain in patients with atypical facial pain measured by positron emission tomography. J. Neurol. Neurosurg. Psy- chiatry 57: 1166- 1173, 1994.

DOUGLAS, R. J., KOCH, C., MAHOWALD, M., MARTIN, K. A. C., AND SUAREZ, H. H. Recurrent excitation in neocortical circuits. Science Wash. DC 269: 981-985, 1995.

EBNER, T. J. AND CHEN, G. Use of voltage-sensitive dyes and optical recordings in the central nervous system. Prog. Neurobiol. 46: 463-506, 1995.

GRINVALD, A. Real-time optical mapping of neuronal activity: from single growth cones to the intact mammalian brain. Annu. Rev. Neurosci. 8: 263-305, 1985.

GRINVALD, A., FROSTIG, R. D., SIEGAL, R. M., AND BARFIELD, E. High resolution optical imaging of functional architecture in the awake monkey. Proc. Natl. Acad. Sci. USA 88: 11559-11563, 1991.

GRINVALD, A., LIEKE, E. E., FROSTIG, R. D., AND HILDESHEIM, R. Cortical point-spread function and long-range lateral interactions revealed by real- time optical imaging of macaque monkey primary visual cortex. J. Neu- rosci. 14: 2545-2568, 1994.

HIRSCH, J. A. AND GILBERT, C. D. Synaptic physiology of horizontal connections in the cat’s visual cortex. J. Neurosci. 11: 1800- 1809, 1991.

JONES, A. K. P., BROWN, W. D., FRISTON, K. J., QI, L. Y., AND FRACKOWIAK, R. S. J. Cortical and subcortical localization of response to pain in man using positron emission tomography. Proc. R. Sot. Lond. B Biol. Sci. 244: 39-44, 1991.

JONES, E. G. Transmitter chemistry in the thalamus. In: Pain and Central Nervous System Disease: The Central Pain Syndromes, edited by K. Casey. New York: Raven, 1991, chapt. 15, p. 201-218.

JONES, E. G. AND PORTER, R. What is area 3a? Brain Res. Rev. 2: l-43, 1980.

KENSHALO, D. R., JR. AND WILLIS, W. D., JR. The role of the cerebral cortex in pain sensation. In: The Cerebral Cortex, edited by A. Peters and E. G. Jones. New York: Plenum, 1991, vol. 9, chapt. 5, p. 153-212.

KNIFFKI, K.-D. Input to and output from lamina I neurones in the cat spinal cord. In: Processing of Sensory Information in the Superficial Dorsal Horn of the Spinal Cord, edited by F. Cevero, G. J. Bennett, and P. M. Headley. New York: Plenum, 1989, p. 129-144.

MOUNTCASTLE, V. B. Modality and topographic properties of single neurons of cat’s somatic sensory cortex. J. Neurophysiol. 20: 408-434, 1957.

PERL, E. R. Pain and Nociception. In: Handbook of Physiology. The Nervous System. Sensory Processes. Bethesda, MD: Am. Physiol. Sot., 1984, sect. 1, vol. III, p. 915-975.

POWELL, T. P. S. AND MOUNTCASTLE, V. B. The cytoarchitecture of the post-central gyrus of the monkey Macaca mulatta. Bull. Johns Hopkins Hosp. 105: 108-131, 1959a.

POWELL, T. P. S. AND MOUNTCASTLE, V. B. Some aspects of the functional organization of the postcentral gyrus of the monkey: a correlation of findings obtained in a single unit analysis with cytoarchitecture. Bull. Johns Hopkins Hosp. 105: 133-162, 1959b.

PRICE, D. D. Characterizing central mechanisms of pathological pain states by sensory testing and neurophysiological analysis. In: Pain and Central Nervous System Disease: The Central Pain Syndromes, edited by K. Casey. New York: Raven, 1991, chapt. 8, p. 103-l 15.

PRICE, D. D., Hu, J. W., DUBNER, R., AND GRACELY, R. H. Peripheral suppression of first pain and central summation of second pain evoked by noxious heat pulses. Pain 3: 57-68, 1977.

PRICE, D. D., MAO, J., AND MAYER, D. J. Central neural mechanisms of normal and abnormal pain states. In: Progress In Pain Research and Management, edited by H. L. Fields and J. C. Liebeskind. Seattle, WA: IASP Press, 1994, vol. 1, chapter 6, p. 61-84.

RAUSELL, E., BAE, C. S., VINUELA, A., HUNTLEY, G. W., AND JONES, E. G. Calbindin and parvalbumin cells in monkey VPL thalamic nucleus: distribution, laminar cortical projections, and relations to spinothalamic terminations. J. Neurosci. 12: 4088-4111, 1992.

RAUSELL, E. AND JONES, E. G. Histochemical and immunocytochemical compartments of the thalamic VPM nucleus in monkeys and their relationship to the representational map. J. Neurosci. 11: 2 10-225, 199 1 a.

RAUSELL, E. AND JONES, E. G. Chemically distinct compartments of the thalamic VPM nucleus in monkeys relay principal and spinal trigeminal pathways to different layers of the somatosensory cortex. J. Neurosci. 11: 226-237, 1991b.

RAUSELL, E. AND JONES, E. G. Extent of arborization of thalamocortical axons as a determinant of representational plasticity in monkey somatic sensory cortex. J. Neurosci. 15: 4270-4288, 1995.

SHI, T., STEVENS, R. T., TESSIER, J., AND APKARIAN, A. V. Spinothalamo- cortical inputs preferentially innervate the superficial and deep cortical layers of SI. Neurosci. Lett. 160: 209-213, 1993.

SINGER, W. Development and plasticity of cortical processing architectures. Science Wash. DC 270: 758-764, 1995.

STEMMLER, M., USHER, M., AND NIEBUR, E. Lateral interactions in primary

TOMMERDAHL, DELEMOS, VIERCK, FAVOROV, AND WHITSEL

visual cortex: a model bridging physiology and psychophysics. Science matosensory Cortex, edited by 0. Franzen. Boston, MA: Birkhauser, Wash. DC 269: 18774880, 1995. 1996.

SUR, M., NELSON, R. J., AND KAAS, J. H. Representations of the body TOMMERDAHL, M., WHITSEL, B., VIERCK, C. J., JR ., FAVOROV, O., JULIANO, surface in cortical areas 3b and 1 of squirrel monkeys: comparisons with other primates. J. Comp. Neural. 211: 177 - 192, 1982.

S., COOPER, B., METZ, C., AND NAHKLE, B. Effects of spinal dorsal

TALBOT, J. D., MARRETT, S., EVANS, A. C., MEYER, E., BUSHNELL, M. C., column transection on the response of monkey anterior parietal cortex

AND DUNCAN, G. H. Multiple representations of pain in human cerebral to repetitive skin stimulation. Cereb. Cortex 6: 13 1 - 155, 1996.

cortex. Science Wash. DC 251: 1355-1358, 1991. VIERCK, C. J., JR., COHEN, R. H., AND COOPER, B. Y. Effects of spinal

TOMMERDAHL, M., FAVOROV, O., AND WHITSEL, B. L. Optical imaging of lesions on temporal resolution of cutaneous sensations. Somatosens. Res.

stimulus-evoked intrinsic signals in somatosensory cortex. Sot. Neurosci. 3: 45-56, 1985.

Abstr. 20: 57.6, 1994. WHITE, E. L. Termination of thalamic afferents in the cerebral cortex.

TOMMERDAHL, M., FAVOROV, O., WHITSEL, B. L., NAHKLE, B., AND In: The Cerebral Cortex, Sensory-Motor Areas and Aspects of Cortical

GONCHAR, Y. Minicolumnar activation patterns in cat and monkey SI Connectivity, edited by E. G. Jones and A. Peters. New York: Plenum,

cortex. Cereb. Cortex 3: 399-411, 1993. 1986, vol. 5, p. 271-289.

TOMMERDAHL, M. AND WHITSEL, B. L. Optical imaging of intrinsic signals WHITE, E. L. Cortical Circuits; Synaptic Organization of the Cerebral in somatosensory cortex. In: Somesthe&s and the Neurobiology of So- Cortex; Structure, Function and Theory. Boston, MA: Birkhauser, 1989.

Documents

Anterior parietal cortical response to tactile and skin-heating stimuli