Journal of Neuroscience Methods, 35 (1990) 19-29 Elsevier
19
N S M 01127
A single-unit recording system, contact thermal probe and electromechanical stimulator for studying cellular mechanisms related to nociception at brain stem level of awake, freely moving rats Jean-Louis Olivrras, GlUes Martin, Bart Vos * a n d Jacqueline M o n t a g n e Unit~ de Neurophysiologie Pharmacologique de I ' I N S E R M (U. 161), 2 rue d'A l~sia, Paris (France) (Received 25 October 1989) (Revised version received 26 February 1990 and 9 May 1990) (Accepted 23 May 1990)
Key words: Single unit recording system; Contact thermal probe; Electromechanical stimulator; Freely moving rat; Nociception The purpose of this paper is to describe a simple, light-weight (3 g) device bearing a fine p l a t i n u m - i r r i d i u m Teflon-coated wire (50 ~ m ) used to record single-unit activity extracellularly at brain stem level in the totally conscious freely moving rat. The up and down movements of the electrode through a guide cannula are insured by a small nut and a spring; the distance between the electrode and the end of the guide cannula is measured with a nut index. The system is directly connected to an amplifier (no FET or preamplifier) and allows for long term recordings necessary for a complete neuronal characterization and pharmacological experiments. The device is easy to make, entirely recoverable, and can be implanted from an animal to another. Further improvements are possible such as tungsten microelectrodes and telemetric or microinjection systems. In order to study some neuronal brain stem mechanisms involved in nociception, we have also designed a contact thermal probe and an electromechanical stimulator. The thermode is stuck to the shaved skin on the back of the rat, allowing heat pulses up to 5 1 ° C to be applied. The mechanical stimulator is used manually and delivers reproducible innocuous stimuli to the skin. The fact that both types of stimulations are driven electrically enables the elaboration of cumulated peristimulus histograms which will reflect the neuronal activities in response to the application of noxious and non noxious stimuli.
Introduction
It has been recognized that the recording of intracerebral single-unit activity in the fully conscious and freely moving animals is a major approach in neurosciences. As a matter of fact, there was soon a growing awareness amongst re-
* Present address: R.J. Maciewicz, Neurology Service, Massachusetts General Hospital, Boston, MA 02114, U.S.A. Correspondence: Dr. J.-L. Ohvrras, Unit6 de Neurophysiologie Pharmacologique de H N S E R M (U.161), 2 rue d'Alrsia, Paris (France)
searchers that such an approach is essential in order to appreciate the real physiological basis behind global CNS functions as for example locomotion or the variations in the state of vigilance. However, it seems paradoxical that although pain and nociception are phenomena with a high degree of CNS integration, implying emotional and cognitive states and the absolute necessity to react to a noxious stimulus, most experiments are performed on animals that are paralyzed and anesthetized, always recorded between constraining and painful ear bars. Refined psychophysiological approaches with simultaneous single-unit recordings have already been
0165-0270/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
20 performed on chair-restrained behaving monkeys at first in the medullary dorsal horn (Dubner et al., 1981; Hayes et al., 1981; Hoffman et al., 1981; Bushnell et al., 1984; Maixner et al., 1988) and later in the thalamus (Bushnell et al., 1987). Simpler studies were also conducted in the thalamus of awake monkeys (Casey and Morrow, 1983) and human beings (Lenz et al., 1987; Lenz et al., 1988). With lower species, even fewer single-unit recording studies in relation to pain or motor systems have been carried out with awake animals and, in all cases, they concerned restrained or paralysed animals: for instance, in the spinal cord of the cat (Collins and Ken, 1987) and in the thalamus or the cortex of the rat (Sapienza et al., 1981; Casella and Davis, 1987; Bassant et al., 1990). In order to avoid such experimental conditions that potentially induce stress and related phenomena able to modify nociceptive processing and other sources of uncontrolled variability, we shall describe in the present study a certain number of methods that allow the appreciation of cellular mechanisms involved in nociception at the brain stem ventromedial medulla level (VMM, a structure implicated in the downward descending control of nociception) in the awake and totally freely moving rat. From a general point of view, most of the single unit recording devices proposed in the literature are not entirely satisfactory. Indeed some can finely drive a microelectrode but are complex and sometimes expensive (Ainsworth and O'Keefe, 1977; Sinnamon, 1977; Deadwyler et al., 1979; Pager, 1984; Veregge, 1985; Wayne-Atdridge et al., 1988; Amos et al., 1989), others show occasionnal electrode rotation (Winson, 1973; Vertes et al., 1975), and finally, some are fixed systems (Strumvasser, 1958; Chorover and Delula, 1972; Palmer, 1977; Fontani, 1981; Yamamoto, 1986), which reduce to a great extent the probability that single units will be isolated (Costa et Delacour, 1976). Hence, in the present study we have designed a simple, inexpensive recording system which can vertically drive a fine platinum wire as the recording electrode with no rotation. Moreover, we have also designed two additional tools in order to apply easily both the noxious and the
innocuous stimuli in the totally freely moving rat. The noxious thermal stimulation was realized with a semi-chronically implanted contact probe less sophisticated than other models (Dubner et al., 1976; Morrow and Casey, 1981) but reliable enough so as to be able to apply reproducible temperatures in the noxious range (up to 51°C), and also determine in some cases whether the units were encoding the temperature intensity (Olivrras et al., 1990). Finally, the innocuous stimulus was delivered via a manually applied electrostimulator.
Methods
Single-unit recording device Principle and realisation In order to isolate and record the VMM single-unit activities for long enough periods of time (several hours a day over 1 or 2 weeks), the rats were chronically implanted with a light-weight system (3 g, excluding the acrylic and silver loops for fixation on the skull: 0.7 g, see further) prior to the recording sessions (Fig. 1). The recording electrode, a fine platinum wire, is supported by the device and driven by a nut-bolt spring loaded system. The wire slides into a guide cannula that protects the electrode while the dura matter is perforated in the course of implantation; it also allows the electrode to be precisely positioned just above the structure studied. An index placed at the top of the driving nut enables the exact location of the electrode tip to be measured taking the end of the guide cannula as a reference. The system is comprised of two major parts both 5 mm in width, which are the fixed base (8) and the movable block (5) (cut from a 3-ram plastic piece) and measuring 17 mm and 11 mm long, respectively. Two holes and a notch are drilled simultaneously through both aligned parts. One of the holes (0.5 mm diameter) is used for the guide (2) insertion. This guide, glued with cyanolit inside the base, prevents the lateral rotations of the movable block. The second hole (2.3 mm diameter) allows the main shaft bolt (3) and its accessories (nut (4) and spring (6)) to settle. Ini-
21
®
lmm
3
ACRYLIC ~ ; ~ G L t .UE 3NNECTION WIRES
;illl; 7
13
12
10
11
Fig. 1. Semi-sectionalview of the single unit recording device in place. 1: plug-in, 2: guide, 3: main shaft bolt, 4: driving nut, 5: movable block, 6: spring, 7: platinum wire protective tube, 8: base, 9: polyethylene tube and gel-foam, 10: guide cannula, 11: platinum wire, 12: fixating screws, 13: reference electrode. The real distance between the base and the cranium has not been respected.
tially, the bolt with its nut fully screwed along, is inserted through the movable block. Then, the spring is placed around the bolt which is finally screwed through the base and receives the fixating nuts (12) further embedded in acrylic at the time of implantation. Secondly, the guide cannula, a beveled (approx. beveling diameter 0.5 mm), varnished stainless steel tube (0.4 m m diameter; 10) is cemented in the notch of the base. Another tube (0.2 m m diameter; 7), used as the electrode wire protection, slides within the cannula. This small tube, cemented in the notch of the movable block, holds the 50-/~m Teflon-coated platinum irridium wire cut at the end by sharp scissors (11).
Finally, the electrical contacts between the recording electrode and the plugs (1) are assured by a small m a l e - f e m a l e connector placed onto the lateral side of the movable block. The extremity of the electrode wire is directly soldered onto the female element in which is inserted the male part then connected to the plugs via a flexible coated wire. This arrangement allows the free up and down movements of the block and the disconnection of the plugs for a possible dismounting of the device (see next section). For the contact of the reference electrode (13), we have adopted an identical system but, in this case, the m a l e - f e m a l e connector is glued onto the base. In addition, it is also possible to add two other connectors of this type onto the base, available for E M G recordings or subcutaneous electrical stimulation for example. When all the connections are in place,, the plugs and connectors are solidly cemented with acrylic onto the base piece. At this stage, the single-unit recording device is almost set up; only a few details necessary for the implantation remain as follows. First, small amounts of silicone rubber are introduced inside the guide cannula in order to facilitate the up and down movements of the 0.2 m m diameter tube and prevent potential blood leaking inside the cannula which could coagulate and block the tube. For this purpose, the guide cannula is filled with silicone rubber by laying this material with a fine tool at the b o t t o m and at the top, and, then pushed into the cannula by to and fro movements of the electrode protective tube. Second, a piece of sticking plaster is applied to the undersurface of the base in order to protect the device from acrylic allowing to retrieve the system after the perfusion of the rat (see next section). For the same purpose, a small polyethylene tube (9) is put around the guide cannula. Third, since the length of the guide cannula is fixed and the position of the main shaft bolt and its accessories variable, the position of the electrode tip relative to the cannula is adjusted via the main shaft bolt which is more or less screwed inside the base. When the movable block is completely up (nut touching the bolt cap; 3) the electrode tip should be positioned slightly inside the cannula. This rnanoeuver secures the electrode tip so that there is no damage during puncture of
22 the dura and brain penetration. Then, we verify that each turn of the nut (measured by the nut index and the bolt marks; 3) has to correspond to an electrode protrusion of 400 ~ m for a total distance of 1.6 m m (4 complete turns of the nut). Finally, we check the resistance of the electrode and the absence of short-circuits.
Implantation and retrievial The rats are anesthetized with ketamine (150 m g / k g i.p.) and positioned with ear bars into the stereotaxic frame in the horizontal position. In a first approach, solid basal fixation points are made following a technique originally developed in the laboratory: 3 silver loops are delicately placed through and under the skull via small holes (3 pairs) drilled fairly laterally relative to the midline of the cranium in frontal and temporal positions.
When in place, the loops are twisted, cut (except one which served as the reference for the recordings; see further) and embedded in acrylic, This system, different from the screws classically used, offers a better stability and is less vulnerable to bone decalcification (personal observations). Then, the guide cannula attached to the device holder between the base and polyethylene tube, is placed above the V M M (posterior: 2.8, lateral: 0.0) according to the stereotaxic coordinates of Fifkova and Marsala (1967). After drilling the skull, the guide cannula is very slowly lowered at a height of 0.2 m m from the ear bars line. The recording device is then fixed to the cranium via the screws at the bottom of the nut-bolt spring loaded system with acrylic, also embodying the polyethylene tube, gel-foam around it, and the 3 silver loops. In the mean time, the reference loop is soldered to the
Fig. 2. Example of the tracks left by the guide cannula and the platinum wire recording electrode in the brain stem at ventromedial medulla level in the awake, freely moving rat. A: trace of the guide cannula; B: trace of the recording electrode; NRM: nucleus raphe magnus; NRPGC: nucleus reticularis paragigantocellularis.
23
device as shown in Fig. 1 (13). Finally, the total under surface of the base covered with stickingplaster (used to retrieve the system, see further) is also embedded with acrylic. When the animals are returned to their home cage, they receive a protective cap as shown on Fig. 4. This cap (piece of circular plastic and male connectors) is simply fixed to the female connectors of the device (Fig. 1, 1) and entirely protects the device (except at the top). After the perfusion of the rats (10-15 days after the implantation in order to clearly mark the tracks of the guide cannula and the electrode in the brain stem, Fig. 2), the single-unit recording device is retrieved. For this operation, the main shaft bolt is unscrewed, the reference electrode cut and the system gently unstuck from the sticking plaster. During this manoeuvre, the guide cannula
THERMODE
slides inside the polyethylene tube. Once retrieved, the device is cleaned and electrically checked. If there is no damage, the device is used again for another animal after gas sterilization. Otherwise, the device can be entirely taken apart for replacement parts such as tubes, electrode, spring, etc.
Contact thermal probe This thermode (Fig. 3) consists mainly of a small spiraled enameled stainless steel resistance (0.8 cm2; 6) that is embedded in cyanolit and acrylic, and is applied to the shaved skin. Heat pulses are produced by delivering a DC voltage (up to 20 V) for a short period of time (0.5 s) through an electrical relay. The heat pulse intensity is controlled by a rheostat and a copper-con-
ELECTROMECH STIM 24v
24v
2mm
Fig. 3. Section view of contact thermal probe and the electromechanical stimulator. Thermode = 1 : thermocouple connection screws, 2: connection wires of the resistance coil, 3: rubber support, 4: biological glue, 5: thermocouple, 6: heating resistance coil. Electromechanical stimulator = 1: piece of rubber, 2: capacitor, 3: spring, 4: solenoid, 5: solenoid core, 6: flat brush, 7: skin ring.
24
stantan thermocouple (CT = 5 ms) glued onto the heating surface of the probe (5). The thermocoupie is connected to a digital thermometer and an oscilloscope in order to be able to read the temperature variations, visualizing their time course. An example is given in Fig. 6. It can be seen that the temperature rise is rather low (about 8 o C/s), and that the passive return to the baseline is very progressive. The probe is built as follows. In a first instance, a 20-cm piece of enameled stainless steel wire (0.2 mm diameter) is cut and further wound around a stick for 16 cm. Next, the spires are flattened, arranged as a spiral (6) and glued onto a ceramic block in order to flatten the heating surface and protect it from acrylic. Then the connectors are soldered to the extremities of the wire (2). Second, a circular rubber support (3) used for sticking the probe to the skin, is placed around the coil and, finally, the whole probe is body mounted with acrylic for a distance of about 2 cm. Third, the thermode is unstuck from the ceramic block and the thermocouple is glued with cyanolit to the center of the spiral. The thermocouple wires are wrapped around the body of the thermode for easier replacement and also to facilitate the connections with 2 small bolts embedded in the acrylic (1). Finally, the connections probe-rheostat and thermocouple screws-thermometer are made long enough to allow free movement of the rat when tested in the recording chamber. Before being used, the rheostat-thermode couple is calibrated by taking into account the fact that the basal temperature of the shaved skin of the rat is 36°C. More precisely, we established reference curves (like that shown in Fig. 6) via the oscilloscope and also took into account direct measurements from the digital thermometer. In our experiments the heat pulse intensity could vary in the noxious range from 43 to 51°C. We obtained an excellent accuracy in the temperature appreciation and thus we were able to demonstrate good stimulus-dependent effects with 2 ° C steps (Oliv6ras et al., 1990). In our experiments, the heat pulses were delivered every 3 min. With the experimenters, the application of pulses reaching values of 4 6 ° C and more induced the well-known double pain sensation (Lewis and
i I~i
/.///
,/
-~
...
2/
~_ _
.....
Fig. 4. Rat in the recording chamber. Note the protective shield of the single unit device, the connection wires and the thermode stuck on top of the back.
Pochin, 1938; Landau and Bishop, 1953; Sinclair and Stokes, 1964; Price et al., 1977). At the time of experimentation, the probe is firmly applied to the shaved skin and the sticking rubber firmly fixed in place with a special biological glue (Fig. 4) (Rapid Colle, Alvar). When the experiment is completed, the probe is removed with alcohol. Electromechanical stimulator This device was originally designed in order to apply a light tactile innocuous stimulus by means of a flat brush mounted on a solenoid (Fig. 3). However, the brush can easily be replaced with a fine needle in order to deliver a brief noxious pin-prick stimulus (Oliv6ras et al., 1989). The solenoid is activated from a 24-V DC source controlled by a classical electrical pulse generator driving a relay. In our experimental conditions, we deliver 6 brush displacements within 300 ms. The tool (brush, (6) or needle) is simply connected to the core (5) of the solenoid (4). The amplitude and smoothness of the to and fro movements of the core are ensured by a spring (3) and a piece of rubber (1) in contact with the metal box containing the solenoid. This box is used to hold the device and as a protective shield a ~ n s t even-
25 tual electromagnetic emissions. For the same purpose, a capacitor (2) is interposed on the 24-V DC line. The ring (7) is there to adjust and keep constant the distance between the skin and the brush or needle. At the time of experimentation, the ring is gently manually laid upon the shaved back skin, and after a short period of adaptation, the stimulus is delivered at a rate of 1 every 10 s. When applied to the experimenters, the first sensation due to the ring contact rapidly disappears and the stimulation itself is felt as a saliant light tactile fluttering sensation.
Usefulness of the single-unit recording device, thermode and electromechanical stimulator in the awake, freely moving rat In Fig. 4 the rat is wearing the recording device and the thermode is in place. The picture shows that the recording system does not take much room and is not too high on the rat's head. Besides, the daily manipulation of the animals revealed that it is relatively easy to use a pair of forceps to handle the driving nut and check the position of the nut index. As is also shown in Fig. 4, the thermode is stuck on the back and thus the animal has no easy access to the probe and its wires. On the whole, the rats perfectly tolerated the presence of the probe on their skin and attempts to remove it were rare. However, when they occured, a few gentle taps on the table upon which they were standing was sufficient to distract and discourage them. The innocuous stimulation was also well tolerated and in general, the animals did not try to escape the electromechanical stimulator. However, because the animals kept moving, the wires from the thermode and the recording device could be twisted together, but this potential complication is minimized by familiarizing the rat with the experimenters and the recording chamber.
Results and discussion
More than 2 years of experiments carried out with about 150 awake, totally freely moving rats
(400 neurons were fully studied with the reproducible non-noxious and noxious stimuli) has emphasized the efficiency of the single-unit recording system described in the present study. In most cases, we obtained recordings of very good quality, all performed at the VMM level (example of Fig. 5). In this example, the spikes amplitude turns around 300 /xV (generally between 100 and 400 laV) and the signal/noise ratio around 10 (generally between 5 and 10). These signals were obtained by simply directly connecting the animal (1 m shielded cables) to the amplifier (Tektronix AM 502, band pass: 0.1-3 kHz). Such a level of quality eliminated the need to use a FET (Field Effect Transistor for movement artefacts elimination) or a head preamplifier. For subsequent studies, the action potentiels were stored on a digital magnetic tape, discriminated (window discriminator Cemi INSERM) and histograms were produced with a IBM PC compatible microcomputer, Furthermore, we were able to sustain the recording of most units for several hours (4 h in Fig. 5) and sometimes more. When we lost or killed the neurons, it was always because of mechanical shocks between the recording device and the recording chamber. Fig. 6 demonstrates the possibility of generating peristimulus histograms after innocuous and noxious stimulations applied by the electromechanical stimulator and the thermode, respectively. These histograms, showing remarkable responses to both types of stimuli, have led to further quantitative comparisons between the responses and the elaboration of stimulus-response functions for noxious heat (Olivdras et al., 1989, 1990). We also showed that our sytems are reliable enough to initiate neuropharmacological studies, as appears with the physiological responses in Fig. 5. From these preliminary experiments, clear effects due to the administration of brevital (100 m g / k g i.p.) or morphine (1 and 3 m g / k g i.v.) were observed. In addition, these compounds did not induce any neuronal loss or spike amplitude modifications even after repeated i.p. and i.v. injections in the course of long periods of recording. Finally, we have managed the chronic implantation of a bipolar stimulating electrode inside the dorsolateral funiculus of the spinal cord (contain-
26
BREVITAL
CONTROL
60mn
lOmn
LT
51"C 200pv 0.2s
MORPHINE 3mg
MORPHINE lmg 30mn L~ ' ,it~J ..... Illl: . ' ~
60ran
60mn l..lIII :_.'~ ~I I t L ' Z ~ - ' : : ~
lal ~:::~l.]]~,~t] II Fig. 5. Example of a VMM single unit recorded for 4 h in the awake, freely moving rat. This figure displays the neuronal responses consecutive to the two types of cutaneous stimuli: light touch delivered by the electromechanical stimulator (LT) and noxious heat via the contact thermal probe (NH, heat pulses from 36 to 51°C). The action potentials were directly amplified and filtered (Tektronix AM 502, gain: 20 K, band pass: 0.1-3 kHz). One can see the quality of the recordings all along the session and the effects of a short-duration barbiturate (brevital, 100 mg/kg i.p.) and 2 doses of the opiate analgesic morphine. Taking into account that this neuron was not spontaneously active, brevital caused total disappearence of both responses and recovery 1 h after administration. Alternatively, 3 m g / k g i.v. morphine totally abolished the noxious heat response and possibly enhanced the innocuous one.
ing many axons from the VMM). We confirmed the presence of bulbospinal neurons at VMM level as revealed by the classical method of antidromic activation (example of Fig. 6). This shows that such a method is possible in totally freely moving rats. Considering that these data give only a few examples of the possibility offered by our systems, we think that the techniques we developed represent a real progress for the present and future studies of nociceptive mechanisms in the rat, a species widely used. Removing the anesthesia
should already have been the ultimate aim of microphysiological studies on pain in the rat and our work is of a pioneering approach, In fact, we have recently shown that such a barbiturate as pentobarbital strongly modifies the VMM neuronal properties, revealing the presence of different neuronal classes not found in the conscious animal (Oliv6ras et al., in preparation). The use of freely moving animals limits a priori the application of reproducible calibrated peripheral non noxious and noxious stimuli, but the tools we have designed can easily be used, particu-
27
5resp L~GHT TOUCH
500hz
2O-
BOOhz
FREQ
2ms ~
~
COLL 16
E - 2ot lO 0
5*
--
J
05s
Fig. 6. Responses of identified VMM bulbospinal neuron to light touch and noxious heat: cumulated peristimulus histograms (5 responses). This unit was antidromically activated from the dorsolateral funiculus of the spinal cord (see text) following the classical criteria: fixed latency (LAT), high frequency following (FREQ) and collision with an orthodromic spike (COLL). The numbers represent the interval between the antidromic spike and the occurence of the antidromic stimulation. For the physiological responses, one can observe the intense initial burst of activation due to light touch and the adaptation of the noxious heat response.
larly the thermal probe. Indeed, although this thermode is rather rough compared to other systems (Dubner et al., 1981; Morrow and Casey, 1981) - there is neither a temperature feed-back control leading to a temperature plateau nor cooling procedure for a fast return to baseline temperature - we managed to obtain authentic thermal responses and were able to show a temperature encoding under particular conditions in a totally freely moving, non-anesthetized rat (Oliv6ras et al., 1990). The electromechanical stimulator is cheap and of a simple design. However, this tool is not as elegant as the thermode since it requires a manual intervention on the animal. In addition, the ring application necessary for keeping the distance brush-skin constant is a source of stimulation which can potentially disturb the physiologic response to the brush stimulation. However, as the rats are well habituated to manipulation inside the recording chamber and the responses to ring application rapidly habituated, there is no major problem for delivering the reproducible innocuous stimulation. So far, it is not possible to vary the
intensity of such a stimulation with our electromechanical stimulator, but one intensity is sufficient in order to compare the neuronal responses produced by non-noxious and noxious stimuli onto the same neurons. In any case, both the electromechanical stimulator and the thermal contact probe trigger off remarkable VMM neuronal responses which can be cumulated in peristimulus histograms. The single unit recording device which is close to the drive proposed by Diana et al. (1987) offers a better security for the electrode penetration and allows to check with precision the position of the electrode within the brain stem. This device is very cheap and all the parts are commercially available. Building this tool requires neither particular skills nor any sophisticated machinery (only a good power drill). The major disadvantage could be the inability to perform penetrations at different places in the same rat, but we generally got a reasonable number of units with only one spot per animal in which it was possible to realise several penetrations. Finally, this device can be positioned over any region of the brain and receive a telemetric system for very long term recordings, e.g. in the case of chronic pain. It can also be modified to receive other electrode types such as tungsten and glass microelectrodes or a microinjection system.
Acknowledgement This work was supported by 1NSERM.
References Ainsworth, A. and O'Keefe, J. (1977) A lightweight microdrive for the simultaneous recording of several units in the awake. freely moving rat, J. Physiol. (Lond.), 269: 8-10. Amos, A., Armstrong, D.M., Harris, R. and Marple-Horvat, D.E. (1989) A lightweight hybrid microdrive for use with awake, unrestrained animals, J. Neurosci. Methods. 28: 219-224. Bassant, M.H., Baleyte, J.M. and Lamour, Y. (1990) Electrophysiological and microiontophoretic studies of single cortical somatosensory neurons in the unasthetized rat, Neuroscience, in press. Bushnell, M.C., Duncan, G.H., Dubner, R. and He, L . F
28 (1984) Activity of trigeminothalamic neurons in medullary dorsal horn of awake monkeys trained in a thermal discrimination task, J. Neurophysiol., 52: 170-187. Bushnell, M.C. and Duncan, G.H. (1987) Mechanical response properties of ventroposterior medial thalamic neurons in the alert monkey, Exp. Brain Res., 67: 603-614. Casey, K.L. and Morrow, T.J. (1983) Ventral posterior thalamic neurons differentially responsive to noxious stimulation of the awake monkey, Science, 221: 675-677. Cassella, J.V. and Davis, M. (1987) A technique to restrain awake rats for recording single unit activity with glass micropipettes and conventional microdrives, J. Neurosci. Methods, 19: 105-113. Chorover, S.L. and Delula A.M. (1972) A sweet new multiple electrode for chronic single unit recording in the moving animal, Physiol. Behav., 9: 671-674. Collins, J.G. and Ren, K. (1987) WDR response profiles of spinal dorsal horn neurons may be unmasked by barbiturate anesthesia, Pain, 28: 369-378. Costa, J.C. and Delacour, J. (1976) Enregistrements d'activitrs unitaires et multiunitaires chez le rat chronique, Physiol. Behav., 16: 497-499. Deadwyler, S.A., Bicla J., Rose. G., West, M. and Lynch, G. (1979) A microdrive for use with glass or metal microelectrodes in recording from freely moving rats, Electroenceph. Clin. Neurophysiol., 47: 752-754. Diana, M., Garcia-Munoz, M. and Freed, C.R. (1987) Wire electrodes for Chronic single unit recording of dopamine cells in substantia nigra pars compacta of awake rats, J. Neurosci. Methods, 21: 71-79. Dubner, R., Beitel, R.E. and Brown, FJ. (1976) A behavioral model for the study of pain mechanisms in primates. In: M. Weisenberg and B. Tursky (Eds.), Pain: New Perspectives in Therapy and Research, Plenum Press, New York, pp. 155-170. Dubner, R., Hoffman, D.S. and Hayes, R.L. (1981) Neuronal activity in medullary dorsal horn of awake monkeys trained in a thermal discriminative task. III. Task-related responses and their functionnal role, J. Neurophysiol., 46: 444-463. Fifkova, E. and Marsala, J. (1967) Stereotaxic atlas for the cat. rabbit and rat. In: J. BureL M. Petran and J. Zacher (Eds./, Electrophysiological Methods in Biological Research, Academic Press, New York, pp. 653-731. Fontani, G. (1981) A technique for long-term recording from single neurons in unrestrained behaving animals, Physiol. Behav., 26: 331-333. Hayes, R.L., Dubner. R. and Hoffman, D.S. (1981) Neuronal activity in medullary dorsal horn of awake monkeys trained in a thermal discrimination task. II. Behavioral modulation of responses to thermal and mechanical stimuli. J. Neurophysiol., 46: 428-443. Hoffman, D.S., Dubner. R., Hayes, R.L. and Medlin, T.P. (1981) Neuronal activity in medullary dorsal horn of awake monkeys trained in a thermal discrimination task. I. responses to innocuous and noxious thermal stimuli, J. Neurophysiol., 46: 409-427. Landau, W. and Bishop, G.G. (1953) Pain of non-dermal,
periosteal, and fascial endings and from inflammation, Arch. Neurol. Psychiatry, 69: 490-504. Lenz, F.A., Tasker, R.R., Dostrovsky, J.O.. Kwan, H.(_ Gorecki, J.. Hirayama, T. and Murphy, J.J. (1987J Abnormal single-unit recorded in the somatosensory thalamus of a quadriplegic patient with central pain, Pain, 31: 225-236. Lenz, F.A., Dostrovsky, J.O., Tasker, R.R, Yamashiro, K., Kwan, H.C, and Murphy, J.J. (1988) Single-unit analysis of the human ventral thalamic nuclear group: somatosensory responses, J. Neurophysiol., 59: 299-316. Lewis, T. and Pochin, E.E. (1938) The double response of the human skin to a single stimulus, Clin. Sci., 3: 67-76. Maixner, W., Dubner, R., Bushnell, M.C., Kenshalo, D.R., Jr. and Olivrras, J.L. (1988) Wide-dynamic-range dorsal horn neurons participate in the encoding process by which monkeys perceive the intensity of noxious heat stimulus, Brain Res., 374: 385-388. Morrow, T.J. and Casey, K.L. (1981) A contact thermal stimulator for neurobehavioral research on temperature sensation, Brain Res. Bull., 6: 281-284. Olivrras, J.I_.., Vos, B., Martin, G. and Montagne, J. (I989) Electrophysiological properties of ventromedial medulla neurons in response to noxious and non-noxious stimuli in the awake, freely moving rat: a single unit study, Brain Res., 486: 1-14. Oliv6ras, J.L., Martin, G., Montagne, J. and Vos. B. (1990l Single unit activity at ventromedial medulla level in the awake, freely moving rat: effects of noxious heat and light tactile stimuli onto convergent neurons, Brain Res., 506: 19-30. Pager, J. (1984) A removable head-mounted microdrive for unit recording in the free behaving rat, Physiol. Behav., 33: 843-848. Palmer, C. (1977) A microdrive technique for recording single neurons in unrestrained animals, Brain Res. Bull., 3: 285289. Price, D.D., Hu, W.J., Dubner, R. and Gracely, R. (1977) Peripheral suppression of first pain and central summation of second pain evoked by heat pulses. Pain. 3: 57-68. Sapienza, S., Talbi, B., Jacquemin, J. and Albe-Fessard, D. (1981) Relationship between input and output of cells in motor and somatosensory cortices of the chronic awake rat, Exp. Brain Res., 43: 47-56. Sinclair, D.C. and Stokes, B.A.R. (1964) The production and characteristics of second pain. Brain, 87: 609-618. Sinnamon, H.M. and Woodward, D.J. (. 1977) Microdrive and method for single unit recording in the active rat, Physiol. Behav., 19: 451-453. Strumvasser, F. (1958) Long-term recording from single neurons in brain of unrestrained mammals, Science, 127: 469470. Veregge, S. and Frost, J.D., Jr. (1985) A simple, inexpensive, hydraulic microdrive for recording neocortical unit activity in the unanesthetized rat, Electroenceph. Clin. Neurophysiol., 61: 94-97. Vertes, R.P. (1975) A device for recording single unit activity
29 in freely moving rats by movable fine wire microelectrode, Electroenceph. Clin. Neurophysiol., 38: 90-92. Wayne-Aldridge, J., Gilman, S. and Jones, D. (1988) A microdrive positionning adapter for chronic single unit recording, Physiol. Behav., 44: 821-823. Winson, J. (1973) A compact microelectrode assembly for
recording from the freely moving rat. Electroenceph. Clin. Neurophysiol., 35: 215-217. Yamamoto, T. (1986) Easy construction of an improved fine wire electrode for chronic single neurons recording in freely moving animals, Physiol. Behav. 39: 649-652.
19
N S M 01127
A single-unit recording system, contact thermal probe and electromechanical stimulator for studying cellular mechanisms related to nociception at brain stem level of awake, freely moving rats Jean-Louis Olivrras, GlUes Martin, Bart Vos * a n d Jacqueline M o n t a g n e Unit~ de Neurophysiologie Pharmacologique de I ' I N S E R M (U. 161), 2 rue d'A l~sia, Paris (France) (Received 25 October 1989) (Revised version received 26 February 1990 and 9 May 1990) (Accepted 23 May 1990)
Key words: Single unit recording system; Contact thermal probe; Electromechanical stimulator; Freely moving rat; Nociception The purpose of this paper is to describe a simple, light-weight (3 g) device bearing a fine p l a t i n u m - i r r i d i u m Teflon-coated wire (50 ~ m ) used to record single-unit activity extracellularly at brain stem level in the totally conscious freely moving rat. The up and down movements of the electrode through a guide cannula are insured by a small nut and a spring; the distance between the electrode and the end of the guide cannula is measured with a nut index. The system is directly connected to an amplifier (no FET or preamplifier) and allows for long term recordings necessary for a complete neuronal characterization and pharmacological experiments. The device is easy to make, entirely recoverable, and can be implanted from an animal to another. Further improvements are possible such as tungsten microelectrodes and telemetric or microinjection systems. In order to study some neuronal brain stem mechanisms involved in nociception, we have also designed a contact thermal probe and an electromechanical stimulator. The thermode is stuck to the shaved skin on the back of the rat, allowing heat pulses up to 5 1 ° C to be applied. The mechanical stimulator is used manually and delivers reproducible innocuous stimuli to the skin. The fact that both types of stimulations are driven electrically enables the elaboration of cumulated peristimulus histograms which will reflect the neuronal activities in response to the application of noxious and non noxious stimuli.
Introduction
It has been recognized that the recording of intracerebral single-unit activity in the fully conscious and freely moving animals is a major approach in neurosciences. As a matter of fact, there was soon a growing awareness amongst re-
* Present address: R.J. Maciewicz, Neurology Service, Massachusetts General Hospital, Boston, MA 02114, U.S.A. Correspondence: Dr. J.-L. Ohvrras, Unit6 de Neurophysiologie Pharmacologique de H N S E R M (U.161), 2 rue d'Alrsia, Paris (France)
searchers that such an approach is essential in order to appreciate the real physiological basis behind global CNS functions as for example locomotion or the variations in the state of vigilance. However, it seems paradoxical that although pain and nociception are phenomena with a high degree of CNS integration, implying emotional and cognitive states and the absolute necessity to react to a noxious stimulus, most experiments are performed on animals that are paralyzed and anesthetized, always recorded between constraining and painful ear bars. Refined psychophysiological approaches with simultaneous single-unit recordings have already been
0165-0270/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
20 performed on chair-restrained behaving monkeys at first in the medullary dorsal horn (Dubner et al., 1981; Hayes et al., 1981; Hoffman et al., 1981; Bushnell et al., 1984; Maixner et al., 1988) and later in the thalamus (Bushnell et al., 1987). Simpler studies were also conducted in the thalamus of awake monkeys (Casey and Morrow, 1983) and human beings (Lenz et al., 1987; Lenz et al., 1988). With lower species, even fewer single-unit recording studies in relation to pain or motor systems have been carried out with awake animals and, in all cases, they concerned restrained or paralysed animals: for instance, in the spinal cord of the cat (Collins and Ken, 1987) and in the thalamus or the cortex of the rat (Sapienza et al., 1981; Casella and Davis, 1987; Bassant et al., 1990). In order to avoid such experimental conditions that potentially induce stress and related phenomena able to modify nociceptive processing and other sources of uncontrolled variability, we shall describe in the present study a certain number of methods that allow the appreciation of cellular mechanisms involved in nociception at the brain stem ventromedial medulla level (VMM, a structure implicated in the downward descending control of nociception) in the awake and totally freely moving rat. From a general point of view, most of the single unit recording devices proposed in the literature are not entirely satisfactory. Indeed some can finely drive a microelectrode but are complex and sometimes expensive (Ainsworth and O'Keefe, 1977; Sinnamon, 1977; Deadwyler et al., 1979; Pager, 1984; Veregge, 1985; Wayne-Atdridge et al., 1988; Amos et al., 1989), others show occasionnal electrode rotation (Winson, 1973; Vertes et al., 1975), and finally, some are fixed systems (Strumvasser, 1958; Chorover and Delula, 1972; Palmer, 1977; Fontani, 1981; Yamamoto, 1986), which reduce to a great extent the probability that single units will be isolated (Costa et Delacour, 1976). Hence, in the present study we have designed a simple, inexpensive recording system which can vertically drive a fine platinum wire as the recording electrode with no rotation. Moreover, we have also designed two additional tools in order to apply easily both the noxious and the
innocuous stimuli in the totally freely moving rat. The noxious thermal stimulation was realized with a semi-chronically implanted contact probe less sophisticated than other models (Dubner et al., 1976; Morrow and Casey, 1981) but reliable enough so as to be able to apply reproducible temperatures in the noxious range (up to 51°C), and also determine in some cases whether the units were encoding the temperature intensity (Olivrras et al., 1990). Finally, the innocuous stimulus was delivered via a manually applied electrostimulator.
Methods
Single-unit recording device Principle and realisation In order to isolate and record the VMM single-unit activities for long enough periods of time (several hours a day over 1 or 2 weeks), the rats were chronically implanted with a light-weight system (3 g, excluding the acrylic and silver loops for fixation on the skull: 0.7 g, see further) prior to the recording sessions (Fig. 1). The recording electrode, a fine platinum wire, is supported by the device and driven by a nut-bolt spring loaded system. The wire slides into a guide cannula that protects the electrode while the dura matter is perforated in the course of implantation; it also allows the electrode to be precisely positioned just above the structure studied. An index placed at the top of the driving nut enables the exact location of the electrode tip to be measured taking the end of the guide cannula as a reference. The system is comprised of two major parts both 5 mm in width, which are the fixed base (8) and the movable block (5) (cut from a 3-ram plastic piece) and measuring 17 mm and 11 mm long, respectively. Two holes and a notch are drilled simultaneously through both aligned parts. One of the holes (0.5 mm diameter) is used for the guide (2) insertion. This guide, glued with cyanolit inside the base, prevents the lateral rotations of the movable block. The second hole (2.3 mm diameter) allows the main shaft bolt (3) and its accessories (nut (4) and spring (6)) to settle. Ini-
21
®
lmm
3
ACRYLIC ~ ; ~ G L t .UE 3NNECTION WIRES
;illl; 7
13
12
10
11
Fig. 1. Semi-sectionalview of the single unit recording device in place. 1: plug-in, 2: guide, 3: main shaft bolt, 4: driving nut, 5: movable block, 6: spring, 7: platinum wire protective tube, 8: base, 9: polyethylene tube and gel-foam, 10: guide cannula, 11: platinum wire, 12: fixating screws, 13: reference electrode. The real distance between the base and the cranium has not been respected.
tially, the bolt with its nut fully screwed along, is inserted through the movable block. Then, the spring is placed around the bolt which is finally screwed through the base and receives the fixating nuts (12) further embedded in acrylic at the time of implantation. Secondly, the guide cannula, a beveled (approx. beveling diameter 0.5 mm), varnished stainless steel tube (0.4 m m diameter; 10) is cemented in the notch of the base. Another tube (0.2 m m diameter; 7), used as the electrode wire protection, slides within the cannula. This small tube, cemented in the notch of the movable block, holds the 50-/~m Teflon-coated platinum irridium wire cut at the end by sharp scissors (11).
Finally, the electrical contacts between the recording electrode and the plugs (1) are assured by a small m a l e - f e m a l e connector placed onto the lateral side of the movable block. The extremity of the electrode wire is directly soldered onto the female element in which is inserted the male part then connected to the plugs via a flexible coated wire. This arrangement allows the free up and down movements of the block and the disconnection of the plugs for a possible dismounting of the device (see next section). For the contact of the reference electrode (13), we have adopted an identical system but, in this case, the m a l e - f e m a l e connector is glued onto the base. In addition, it is also possible to add two other connectors of this type onto the base, available for E M G recordings or subcutaneous electrical stimulation for example. When all the connections are in place,, the plugs and connectors are solidly cemented with acrylic onto the base piece. At this stage, the single-unit recording device is almost set up; only a few details necessary for the implantation remain as follows. First, small amounts of silicone rubber are introduced inside the guide cannula in order to facilitate the up and down movements of the 0.2 m m diameter tube and prevent potential blood leaking inside the cannula which could coagulate and block the tube. For this purpose, the guide cannula is filled with silicone rubber by laying this material with a fine tool at the b o t t o m and at the top, and, then pushed into the cannula by to and fro movements of the electrode protective tube. Second, a piece of sticking plaster is applied to the undersurface of the base in order to protect the device from acrylic allowing to retrieve the system after the perfusion of the rat (see next section). For the same purpose, a small polyethylene tube (9) is put around the guide cannula. Third, since the length of the guide cannula is fixed and the position of the main shaft bolt and its accessories variable, the position of the electrode tip relative to the cannula is adjusted via the main shaft bolt which is more or less screwed inside the base. When the movable block is completely up (nut touching the bolt cap; 3) the electrode tip should be positioned slightly inside the cannula. This rnanoeuver secures the electrode tip so that there is no damage during puncture of
22 the dura and brain penetration. Then, we verify that each turn of the nut (measured by the nut index and the bolt marks; 3) has to correspond to an electrode protrusion of 400 ~ m for a total distance of 1.6 m m (4 complete turns of the nut). Finally, we check the resistance of the electrode and the absence of short-circuits.
Implantation and retrievial The rats are anesthetized with ketamine (150 m g / k g i.p.) and positioned with ear bars into the stereotaxic frame in the horizontal position. In a first approach, solid basal fixation points are made following a technique originally developed in the laboratory: 3 silver loops are delicately placed through and under the skull via small holes (3 pairs) drilled fairly laterally relative to the midline of the cranium in frontal and temporal positions.
When in place, the loops are twisted, cut (except one which served as the reference for the recordings; see further) and embedded in acrylic, This system, different from the screws classically used, offers a better stability and is less vulnerable to bone decalcification (personal observations). Then, the guide cannula attached to the device holder between the base and polyethylene tube, is placed above the V M M (posterior: 2.8, lateral: 0.0) according to the stereotaxic coordinates of Fifkova and Marsala (1967). After drilling the skull, the guide cannula is very slowly lowered at a height of 0.2 m m from the ear bars line. The recording device is then fixed to the cranium via the screws at the bottom of the nut-bolt spring loaded system with acrylic, also embodying the polyethylene tube, gel-foam around it, and the 3 silver loops. In the mean time, the reference loop is soldered to the
Fig. 2. Example of the tracks left by the guide cannula and the platinum wire recording electrode in the brain stem at ventromedial medulla level in the awake, freely moving rat. A: trace of the guide cannula; B: trace of the recording electrode; NRM: nucleus raphe magnus; NRPGC: nucleus reticularis paragigantocellularis.
23
device as shown in Fig. 1 (13). Finally, the total under surface of the base covered with stickingplaster (used to retrieve the system, see further) is also embedded with acrylic. When the animals are returned to their home cage, they receive a protective cap as shown on Fig. 4. This cap (piece of circular plastic and male connectors) is simply fixed to the female connectors of the device (Fig. 1, 1) and entirely protects the device (except at the top). After the perfusion of the rats (10-15 days after the implantation in order to clearly mark the tracks of the guide cannula and the electrode in the brain stem, Fig. 2), the single-unit recording device is retrieved. For this operation, the main shaft bolt is unscrewed, the reference electrode cut and the system gently unstuck from the sticking plaster. During this manoeuvre, the guide cannula
THERMODE
slides inside the polyethylene tube. Once retrieved, the device is cleaned and electrically checked. If there is no damage, the device is used again for another animal after gas sterilization. Otherwise, the device can be entirely taken apart for replacement parts such as tubes, electrode, spring, etc.
Contact thermal probe This thermode (Fig. 3) consists mainly of a small spiraled enameled stainless steel resistance (0.8 cm2; 6) that is embedded in cyanolit and acrylic, and is applied to the shaved skin. Heat pulses are produced by delivering a DC voltage (up to 20 V) for a short period of time (0.5 s) through an electrical relay. The heat pulse intensity is controlled by a rheostat and a copper-con-
ELECTROMECH STIM 24v
24v
2mm
Fig. 3. Section view of contact thermal probe and the electromechanical stimulator. Thermode = 1 : thermocouple connection screws, 2: connection wires of the resistance coil, 3: rubber support, 4: biological glue, 5: thermocouple, 6: heating resistance coil. Electromechanical stimulator = 1: piece of rubber, 2: capacitor, 3: spring, 4: solenoid, 5: solenoid core, 6: flat brush, 7: skin ring.
24
stantan thermocouple (CT = 5 ms) glued onto the heating surface of the probe (5). The thermocoupie is connected to a digital thermometer and an oscilloscope in order to be able to read the temperature variations, visualizing their time course. An example is given in Fig. 6. It can be seen that the temperature rise is rather low (about 8 o C/s), and that the passive return to the baseline is very progressive. The probe is built as follows. In a first instance, a 20-cm piece of enameled stainless steel wire (0.2 mm diameter) is cut and further wound around a stick for 16 cm. Next, the spires are flattened, arranged as a spiral (6) and glued onto a ceramic block in order to flatten the heating surface and protect it from acrylic. Then the connectors are soldered to the extremities of the wire (2). Second, a circular rubber support (3) used for sticking the probe to the skin, is placed around the coil and, finally, the whole probe is body mounted with acrylic for a distance of about 2 cm. Third, the thermode is unstuck from the ceramic block and the thermocouple is glued with cyanolit to the center of the spiral. The thermocouple wires are wrapped around the body of the thermode for easier replacement and also to facilitate the connections with 2 small bolts embedded in the acrylic (1). Finally, the connections probe-rheostat and thermocouple screws-thermometer are made long enough to allow free movement of the rat when tested in the recording chamber. Before being used, the rheostat-thermode couple is calibrated by taking into account the fact that the basal temperature of the shaved skin of the rat is 36°C. More precisely, we established reference curves (like that shown in Fig. 6) via the oscilloscope and also took into account direct measurements from the digital thermometer. In our experiments the heat pulse intensity could vary in the noxious range from 43 to 51°C. We obtained an excellent accuracy in the temperature appreciation and thus we were able to demonstrate good stimulus-dependent effects with 2 ° C steps (Oliv6ras et al., 1990). In our experiments, the heat pulses were delivered every 3 min. With the experimenters, the application of pulses reaching values of 4 6 ° C and more induced the well-known double pain sensation (Lewis and
i I~i
/.///
,/
-~
...
2/
~_ _
.....
Fig. 4. Rat in the recording chamber. Note the protective shield of the single unit device, the connection wires and the thermode stuck on top of the back.
Pochin, 1938; Landau and Bishop, 1953; Sinclair and Stokes, 1964; Price et al., 1977). At the time of experimentation, the probe is firmly applied to the shaved skin and the sticking rubber firmly fixed in place with a special biological glue (Fig. 4) (Rapid Colle, Alvar). When the experiment is completed, the probe is removed with alcohol. Electromechanical stimulator This device was originally designed in order to apply a light tactile innocuous stimulus by means of a flat brush mounted on a solenoid (Fig. 3). However, the brush can easily be replaced with a fine needle in order to deliver a brief noxious pin-prick stimulus (Oliv6ras et al., 1989). The solenoid is activated from a 24-V DC source controlled by a classical electrical pulse generator driving a relay. In our experimental conditions, we deliver 6 brush displacements within 300 ms. The tool (brush, (6) or needle) is simply connected to the core (5) of the solenoid (4). The amplitude and smoothness of the to and fro movements of the core are ensured by a spring (3) and a piece of rubber (1) in contact with the metal box containing the solenoid. This box is used to hold the device and as a protective shield a ~ n s t even-
25 tual electromagnetic emissions. For the same purpose, a capacitor (2) is interposed on the 24-V DC line. The ring (7) is there to adjust and keep constant the distance between the skin and the brush or needle. At the time of experimentation, the ring is gently manually laid upon the shaved back skin, and after a short period of adaptation, the stimulus is delivered at a rate of 1 every 10 s. When applied to the experimenters, the first sensation due to the ring contact rapidly disappears and the stimulation itself is felt as a saliant light tactile fluttering sensation.
Usefulness of the single-unit recording device, thermode and electromechanical stimulator in the awake, freely moving rat In Fig. 4 the rat is wearing the recording device and the thermode is in place. The picture shows that the recording system does not take much room and is not too high on the rat's head. Besides, the daily manipulation of the animals revealed that it is relatively easy to use a pair of forceps to handle the driving nut and check the position of the nut index. As is also shown in Fig. 4, the thermode is stuck on the back and thus the animal has no easy access to the probe and its wires. On the whole, the rats perfectly tolerated the presence of the probe on their skin and attempts to remove it were rare. However, when they occured, a few gentle taps on the table upon which they were standing was sufficient to distract and discourage them. The innocuous stimulation was also well tolerated and in general, the animals did not try to escape the electromechanical stimulator. However, because the animals kept moving, the wires from the thermode and the recording device could be twisted together, but this potential complication is minimized by familiarizing the rat with the experimenters and the recording chamber.
Results and discussion
More than 2 years of experiments carried out with about 150 awake, totally freely moving rats
(400 neurons were fully studied with the reproducible non-noxious and noxious stimuli) has emphasized the efficiency of the single-unit recording system described in the present study. In most cases, we obtained recordings of very good quality, all performed at the VMM level (example of Fig. 5). In this example, the spikes amplitude turns around 300 /xV (generally between 100 and 400 laV) and the signal/noise ratio around 10 (generally between 5 and 10). These signals were obtained by simply directly connecting the animal (1 m shielded cables) to the amplifier (Tektronix AM 502, band pass: 0.1-3 kHz). Such a level of quality eliminated the need to use a FET (Field Effect Transistor for movement artefacts elimination) or a head preamplifier. For subsequent studies, the action potentiels were stored on a digital magnetic tape, discriminated (window discriminator Cemi INSERM) and histograms were produced with a IBM PC compatible microcomputer, Furthermore, we were able to sustain the recording of most units for several hours (4 h in Fig. 5) and sometimes more. When we lost or killed the neurons, it was always because of mechanical shocks between the recording device and the recording chamber. Fig. 6 demonstrates the possibility of generating peristimulus histograms after innocuous and noxious stimulations applied by the electromechanical stimulator and the thermode, respectively. These histograms, showing remarkable responses to both types of stimuli, have led to further quantitative comparisons between the responses and the elaboration of stimulus-response functions for noxious heat (Olivdras et al., 1989, 1990). We also showed that our sytems are reliable enough to initiate neuropharmacological studies, as appears with the physiological responses in Fig. 5. From these preliminary experiments, clear effects due to the administration of brevital (100 m g / k g i.p.) or morphine (1 and 3 m g / k g i.v.) were observed. In addition, these compounds did not induce any neuronal loss or spike amplitude modifications even after repeated i.p. and i.v. injections in the course of long periods of recording. Finally, we have managed the chronic implantation of a bipolar stimulating electrode inside the dorsolateral funiculus of the spinal cord (contain-
26
BREVITAL
CONTROL
60mn
lOmn
LT
51"C 200pv 0.2s
MORPHINE 3mg
MORPHINE lmg 30mn L~ ' ,it~J ..... Illl: . ' ~
60ran
60mn l..lIII :_.'~ ~I I t L ' Z ~ - ' : : ~
lal ~:::~l.]]~,~t] II Fig. 5. Example of a VMM single unit recorded for 4 h in the awake, freely moving rat. This figure displays the neuronal responses consecutive to the two types of cutaneous stimuli: light touch delivered by the electromechanical stimulator (LT) and noxious heat via the contact thermal probe (NH, heat pulses from 36 to 51°C). The action potentials were directly amplified and filtered (Tektronix AM 502, gain: 20 K, band pass: 0.1-3 kHz). One can see the quality of the recordings all along the session and the effects of a short-duration barbiturate (brevital, 100 mg/kg i.p.) and 2 doses of the opiate analgesic morphine. Taking into account that this neuron was not spontaneously active, brevital caused total disappearence of both responses and recovery 1 h after administration. Alternatively, 3 m g / k g i.v. morphine totally abolished the noxious heat response and possibly enhanced the innocuous one.
ing many axons from the VMM). We confirmed the presence of bulbospinal neurons at VMM level as revealed by the classical method of antidromic activation (example of Fig. 6). This shows that such a method is possible in totally freely moving rats. Considering that these data give only a few examples of the possibility offered by our systems, we think that the techniques we developed represent a real progress for the present and future studies of nociceptive mechanisms in the rat, a species widely used. Removing the anesthesia
should already have been the ultimate aim of microphysiological studies on pain in the rat and our work is of a pioneering approach, In fact, we have recently shown that such a barbiturate as pentobarbital strongly modifies the VMM neuronal properties, revealing the presence of different neuronal classes not found in the conscious animal (Oliv6ras et al., in preparation). The use of freely moving animals limits a priori the application of reproducible calibrated peripheral non noxious and noxious stimuli, but the tools we have designed can easily be used, particu-
27
5resp L~GHT TOUCH
500hz
2O-
BOOhz
FREQ
2ms ~
~
COLL 16
E - 2ot lO 0
5*
--
J
05s
Fig. 6. Responses of identified VMM bulbospinal neuron to light touch and noxious heat: cumulated peristimulus histograms (5 responses). This unit was antidromically activated from the dorsolateral funiculus of the spinal cord (see text) following the classical criteria: fixed latency (LAT), high frequency following (FREQ) and collision with an orthodromic spike (COLL). The numbers represent the interval between the antidromic spike and the occurence of the antidromic stimulation. For the physiological responses, one can observe the intense initial burst of activation due to light touch and the adaptation of the noxious heat response.
larly the thermal probe. Indeed, although this thermode is rather rough compared to other systems (Dubner et al., 1981; Morrow and Casey, 1981) - there is neither a temperature feed-back control leading to a temperature plateau nor cooling procedure for a fast return to baseline temperature - we managed to obtain authentic thermal responses and were able to show a temperature encoding under particular conditions in a totally freely moving, non-anesthetized rat (Oliv6ras et al., 1990). The electromechanical stimulator is cheap and of a simple design. However, this tool is not as elegant as the thermode since it requires a manual intervention on the animal. In addition, the ring application necessary for keeping the distance brush-skin constant is a source of stimulation which can potentially disturb the physiologic response to the brush stimulation. However, as the rats are well habituated to manipulation inside the recording chamber and the responses to ring application rapidly habituated, there is no major problem for delivering the reproducible innocuous stimulation. So far, it is not possible to vary the
intensity of such a stimulation with our electromechanical stimulator, but one intensity is sufficient in order to compare the neuronal responses produced by non-noxious and noxious stimuli onto the same neurons. In any case, both the electromechanical stimulator and the thermal contact probe trigger off remarkable VMM neuronal responses which can be cumulated in peristimulus histograms. The single unit recording device which is close to the drive proposed by Diana et al. (1987) offers a better security for the electrode penetration and allows to check with precision the position of the electrode within the brain stem. This device is very cheap and all the parts are commercially available. Building this tool requires neither particular skills nor any sophisticated machinery (only a good power drill). The major disadvantage could be the inability to perform penetrations at different places in the same rat, but we generally got a reasonable number of units with only one spot per animal in which it was possible to realise several penetrations. Finally, this device can be positioned over any region of the brain and receive a telemetric system for very long term recordings, e.g. in the case of chronic pain. It can also be modified to receive other electrode types such as tungsten and glass microelectrodes or a microinjection system.
Acknowledgement This work was supported by 1NSERM.
References Ainsworth, A. and O'Keefe, J. (1977) A lightweight microdrive for the simultaneous recording of several units in the awake. freely moving rat, J. Physiol. (Lond.), 269: 8-10. Amos, A., Armstrong, D.M., Harris, R. and Marple-Horvat, D.E. (1989) A lightweight hybrid microdrive for use with awake, unrestrained animals, J. Neurosci. Methods. 28: 219-224. Bassant, M.H., Baleyte, J.M. and Lamour, Y. (1990) Electrophysiological and microiontophoretic studies of single cortical somatosensory neurons in the unasthetized rat, Neuroscience, in press. Bushnell, M.C., Duncan, G.H., Dubner, R. and He, L . F
28 (1984) Activity of trigeminothalamic neurons in medullary dorsal horn of awake monkeys trained in a thermal discrimination task, J. Neurophysiol., 52: 170-187. Bushnell, M.C. and Duncan, G.H. (1987) Mechanical response properties of ventroposterior medial thalamic neurons in the alert monkey, Exp. Brain Res., 67: 603-614. Casey, K.L. and Morrow, T.J. (1983) Ventral posterior thalamic neurons differentially responsive to noxious stimulation of the awake monkey, Science, 221: 675-677. Cassella, J.V. and Davis, M. (1987) A technique to restrain awake rats for recording single unit activity with glass micropipettes and conventional microdrives, J. Neurosci. Methods, 19: 105-113. Chorover, S.L. and Delula A.M. (1972) A sweet new multiple electrode for chronic single unit recording in the moving animal, Physiol. Behav., 9: 671-674. Collins, J.G. and Ren, K. (1987) WDR response profiles of spinal dorsal horn neurons may be unmasked by barbiturate anesthesia, Pain, 28: 369-378. Costa, J.C. and Delacour, J. (1976) Enregistrements d'activitrs unitaires et multiunitaires chez le rat chronique, Physiol. Behav., 16: 497-499. Deadwyler, S.A., Bicla J., Rose. G., West, M. and Lynch, G. (1979) A microdrive for use with glass or metal microelectrodes in recording from freely moving rats, Electroenceph. Clin. Neurophysiol., 47: 752-754. Diana, M., Garcia-Munoz, M. and Freed, C.R. (1987) Wire electrodes for Chronic single unit recording of dopamine cells in substantia nigra pars compacta of awake rats, J. Neurosci. Methods, 21: 71-79. Dubner, R., Beitel, R.E. and Brown, FJ. (1976) A behavioral model for the study of pain mechanisms in primates. In: M. Weisenberg and B. Tursky (Eds.), Pain: New Perspectives in Therapy and Research, Plenum Press, New York, pp. 155-170. Dubner, R., Hoffman, D.S. and Hayes, R.L. (1981) Neuronal activity in medullary dorsal horn of awake monkeys trained in a thermal discriminative task. III. Task-related responses and their functionnal role, J. Neurophysiol., 46: 444-463. Fifkova, E. and Marsala, J. (1967) Stereotaxic atlas for the cat. rabbit and rat. In: J. BureL M. Petran and J. Zacher (Eds./, Electrophysiological Methods in Biological Research, Academic Press, New York, pp. 653-731. Fontani, G. (1981) A technique for long-term recording from single neurons in unrestrained behaving animals, Physiol. Behav., 26: 331-333. Hayes, R.L., Dubner. R. and Hoffman, D.S. (1981) Neuronal activity in medullary dorsal horn of awake monkeys trained in a thermal discrimination task. II. Behavioral modulation of responses to thermal and mechanical stimuli. J. Neurophysiol., 46: 428-443. Hoffman, D.S., Dubner. R., Hayes, R.L. and Medlin, T.P. (1981) Neuronal activity in medullary dorsal horn of awake monkeys trained in a thermal discrimination task. I. responses to innocuous and noxious thermal stimuli, J. Neurophysiol., 46: 409-427. Landau, W. and Bishop, G.G. (1953) Pain of non-dermal,
periosteal, and fascial endings and from inflammation, Arch. Neurol. Psychiatry, 69: 490-504. Lenz, F.A., Tasker, R.R., Dostrovsky, J.O.. Kwan, H.(_ Gorecki, J.. Hirayama, T. and Murphy, J.J. (1987J Abnormal single-unit recorded in the somatosensory thalamus of a quadriplegic patient with central pain, Pain, 31: 225-236. Lenz, F.A., Dostrovsky, J.O., Tasker, R.R, Yamashiro, K., Kwan, H.C, and Murphy, J.J. (1988) Single-unit analysis of the human ventral thalamic nuclear group: somatosensory responses, J. Neurophysiol., 59: 299-316. Lewis, T. and Pochin, E.E. (1938) The double response of the human skin to a single stimulus, Clin. Sci., 3: 67-76. Maixner, W., Dubner, R., Bushnell, M.C., Kenshalo, D.R., Jr. and Olivrras, J.L. (1988) Wide-dynamic-range dorsal horn neurons participate in the encoding process by which monkeys perceive the intensity of noxious heat stimulus, Brain Res., 374: 385-388. Morrow, T.J. and Casey, K.L. (1981) A contact thermal stimulator for neurobehavioral research on temperature sensation, Brain Res. Bull., 6: 281-284. Olivrras, J.I_.., Vos, B., Martin, G. and Montagne, J. (I989) Electrophysiological properties of ventromedial medulla neurons in response to noxious and non-noxious stimuli in the awake, freely moving rat: a single unit study, Brain Res., 486: 1-14. Oliv6ras, J.L., Martin, G., Montagne, J. and Vos. B. (1990l Single unit activity at ventromedial medulla level in the awake, freely moving rat: effects of noxious heat and light tactile stimuli onto convergent neurons, Brain Res., 506: 19-30. Pager, J. (1984) A removable head-mounted microdrive for unit recording in the free behaving rat, Physiol. Behav., 33: 843-848. Palmer, C. (1977) A microdrive technique for recording single neurons in unrestrained animals, Brain Res. Bull., 3: 285289. Price, D.D., Hu, W.J., Dubner, R. and Gracely, R. (1977) Peripheral suppression of first pain and central summation of second pain evoked by heat pulses. Pain. 3: 57-68. Sapienza, S., Talbi, B., Jacquemin, J. and Albe-Fessard, D. (1981) Relationship between input and output of cells in motor and somatosensory cortices of the chronic awake rat, Exp. Brain Res., 43: 47-56. Sinclair, D.C. and Stokes, B.A.R. (1964) The production and characteristics of second pain. Brain, 87: 609-618. Sinnamon, H.M. and Woodward, D.J. (. 1977) Microdrive and method for single unit recording in the active rat, Physiol. Behav., 19: 451-453. Strumvasser, F. (1958) Long-term recording from single neurons in brain of unrestrained mammals, Science, 127: 469470. Veregge, S. and Frost, J.D., Jr. (1985) A simple, inexpensive, hydraulic microdrive for recording neocortical unit activity in the unanesthetized rat, Electroenceph. Clin. Neurophysiol., 61: 94-97. Vertes, R.P. (1975) A device for recording single unit activity
29 in freely moving rats by movable fine wire microelectrode, Electroenceph. Clin. Neurophysiol., 38: 90-92. Wayne-Aldridge, J., Gilman, S. and Jones, D. (1988) A microdrive positionning adapter for chronic single unit recording, Physiol. Behav., 44: 821-823. Winson, J. (1973) A compact microelectrode assembly for
recording from the freely moving rat. Electroenceph. Clin. Neurophysiol., 35: 215-217. Yamamoto, T. (1986) Easy construction of an improved fine wire electrode for chronic single neurons recording in freely moving animals, Physiol. Behav. 39: 649-652.
