- Email: [email protected]
A telemetry system for single neuronal discharge recording from behaving monkey
0361~923oiE4$3.00 + .oo
Brain Research Bulletin, Vol. 12, pp. 129-132, 1984. Q Ankho International Inc. Printed in the U.S.A.
BRIEF COMMUNICATION
A Telemetry System for Single Neuronal Discharge Recording from Behaving Monkey’ TOMOYA YAMAMOTO, YUTAKA OOMURA,* SHUJI AOU, YASUHIKO NAKANO AND SEIJI NEMOTO
Department
of Biological Control System, National Institute for Physiological Sciences Okazaki 444 and Department of Physiology, Faculty of Medicine, Kyusyu University, Fukuoka 812, (Japan) Received YAMAMOTO, discharge
T., Y. OOMURA,
recording
from
behaving
S. AOU, monkey.
30 June 1983
Y. NAKANO AND S. NEMOTO. BRAIN RES BULL 12(l) 129-132,
for
A telemetry system single neuronal 1984-A telemetry system for neuronal
discharge recording from behaving monkeys positioned in a primate chair is described. Using FM telemetry system single neuronal activity was recorded from the monkey brain through five teflon-coated platinum-iridium microwire electrodes (25 ~1 in dia.). This system permits stable long term neuronal discharge recording during feeding behavior with minimal artifacts even during mastication. The system has a broadcasting distance of more than 30 m. The frequency response and the S/N ratio obtained from this system is almost comparable with that of the usual direct wire procedures, and the elimination of artifacts is superior. This report describes the procedure for constructing the microelectrode assembly and
FM telemetry system. Telemetry
Unit recording
Feeding
Microwire electrodes
Monkey
ing. Microwires should protrude 5 mm from the tip of the tubing. The opposite ends of the microwires are then stripped of its teflon insulation and soldered to the DIP switch as shown in Fig. 1. The tubing containing microwires and the DIP switch are cemented together with dental acrylic. The outputs of the DIP switch are wired to the 4P connector. Prior to implantation of this electrode assembly, the protruding bundle of microwires is fused and coated with melted polyethylene glycol (melting point, 60°C) for the penetration to the brain tissue [3, 4, 5, lo].
IN recent years, there has been an increasing interest in recording the activity of neurons from unanesthetized and unrestrained animals [l, j,4, 5,6,9, 10, 12, 13, 15, 161. The use of conventional metal or glass electrodes requires physical restraint of the experimental animals. These restraint can be stressful for animals especially during long experimental sessions. This report describes a telemetry for single neuronal discharge recording from monkeys under feeding behavior. This system is composed of a microwire electrode assembly and an FM telemetry system. ELECTRODE ASSEMBLY
FM TRANSMITTER
The electrode assembly is composed of five teflon-coated 90% platinum-lo% iridium microwires, (25 p in dia. Medwire Co. NY), a 25 gauge stainless steel tubing, a dual-in-line-
The circuit diagram for the FM transmitter is shown in Fig. 2A, its perspective bottom view is shown in Fig. 2B. The FET Ql in the input stage is used for a high impedence input. Transistors 42 and Q3 in Darlington connection provide sufficient voltage gain and low power consumption. Transisters Q4 and QS form the modified Colpitts-type oscillator. This type oscillator needs no special oscillator coil and is easy to set the broadcasting frequency. The broadcasting frequency of 90 MHz is used because the frequency range of FM receiver is usually 76 MHz to 90 MHz in Japan. It can be changed by the trimmer capacitor C5 and by adjusting the
pack (DIP) switch (Type SPC-010, Saiden Co. Tokyo), and a 4 pin connector (Type AC4S, Echo Co. Tokyo) as shown in Fig. 1. Construction of this assembly proceeds as follows. The 25 gauge stainless steel tubing is cut to an appropriate length for a target site (e.g., for the orbitofrontal cortex (OBF) the length is approximately 30 mm). Five tefloncoated microwires with the length 20 mm longer than that of the stainless tubing are passed through the lumen of the tub-
‘This work was supported partly by Grant-in-Aid (Y.O.) 57440085 and 5637ooO5from the Ministry of Education, Science and Culture in Japan. “Requests for reprints should be addressed to Y. Oomura.
129
130
ELECTRODE ASSEMBLY
I
/ +3v
A
CONNECTOR
ANCHOR
SKULL TEFLON-COATED Pt-k bliCflOWlRE
POLYETHYLENE
GLYCOL
COATING
FIG. 1. Schematic illustration of electrode assembly. Five tefloncoated PT-Ir microwires contained in stainless cannula is fused and coated with polyethylene glycol prior to implantation. Combination of microwires, selected by DIP switch.
spacing of the coil Ll . After all the parts are wired to the DIP pitch circuit board (25x 15 mm) as shown in Fig. 3, the transmitter assembly is coated and filled with common household epoxy resin. When multi-channel simultaneous single neuronal discharge recordings are required, the broadcasting frequency of one discharge should have difference of at least 2 MHz in each other. The cost of all the components in one unit is less than $5. PERFORMANCE
5rlnl FIG. ?A. Transmitter circuit. A: circuit diagram. B: perspective bottom view. Transistor Ql. P-channel junction type FET 2SK3OGR (Toshiba. for audio frequency amplification). Transistors Q2. Q3 and Q4, 2SC1815 (Toshiba. for audio frequency amplification). Transistor QS. 2X828 (Matsushita. for radio frequency amplification, fr=220 MHz). Resistors RI 10 R5. 0.25 W 10% values. RI: 10 K0; R2; 20 MR; R3; 10 KfI; R4: 2 MR; R5: I K0. Capacitors Cl and C2. tantalum type; C3 and C4 ceramic type: CS ceramic type trimmer. Cl, C2; 2.2 pF; C3; SO0 pF; C4: 10 pF: CS; 20 pF (max). Power source. Li-Mn battery type CR-l:3N (Sanyo. 3 V. dia. = I 1.6~ 10.5 mm). Oscillator coil L I, 6 mm in dia. 8 mm long. six turns of copper wire (dia. = I mm) with antenna tap 2 turns from the end leading to QS. Transmitting antenna. 200 mm long copper wire (dia. = 1 mm).
OF FM TRANSMITTER
The input impedance of this transmitter is more than lo9 Ma at 1 kHz and the frequency response is 5 Hz to 15 kHz (-3 dB). The frequency response can be modified by changing the value of capacitors Cl and C2. Input signal levels from 5 PV to 20 mV can be transmitted without noticeable distortion. The voltage gain can be changed by using high hPE transistor as Q2 and Q3. The frequency deviation is 2400 kHz with an input signal of 10 mV p-p at 1 kHz. The current drain is 1.2 mA and the life of the Li-Mn battery is usually more than 25 hr without remarkable change in oscillating frequency and voltage gain. RECEIVING
ANTENNAE
AND RECEIVER
The receiving antenna is a l/2 A vertical dipole type. The receiver is a stereo FM receiver in mono mode (frequency range, 76 MHz to 90 MHz, sensitivity; 1.2 PV IHF 1958 75 a, Type KT-31 Torio Co. Toyko) and can be substituted for usual FM receiver on the market. Overall voltage gain
IC
m
FIG. 3. Photograph of FM tranbmitter before coating with epoxy resin. Transmitting antenna is removed. Li-Mn battery is fixed by wires and can be easily replaced.
131
TELEMETRY FOR NEURONAL DISCHARGE RECORDING B
A TRANSMrllED
DIRECT ws?E
-I
rooyv
lmsec.
tooyv -I 20msec.
FIG. 4. Example of single neuronal activity recorded through FM telemetry system (upper) and through conventional direct wire procedure (lower). A: fast sweep. B: slow sweep.
A OIlF
LtlA
B.P.
FIG. 5A. ~imuI~neous recording in the o~itofron~ cortex (OBF) and in the lateral hypothalamic area (LHA). A: neuronal activity in the OBF (upper) and the LHA (lower) B: frequency histogram of neuronal activity in the OBF (upper) and the LHA (middle) duringhighfmed ratio (F.R.=20) bar press food task. Arrow (t), time of cue lamp on; filled triangle (A), time of reward (raisin). Neuronal activity during bar press (B.P., lower) in the OBF decreases, while that in the LHA increases.
through the FM transmitter and the receiver is approximately 30 dB at 1 kHz and the frequency response is 10 Hz to 13 kHz (-3 dB). The broadcasting rauge is usually more than 30 m. RECORDINGPROCEDURE
The electrode assembly was stereotaxically implanted in the orbitofrontal cortex (OBF, A 33, L 9 and H 7) and the lateral hypothalamic area (LHA, A 19, L 3 and H 2) of A4ucaca Fuscuta (body weight 5.5 kg.) according to the atlas of Kusama and Mabuchi under ~ntob~bit~ anesthesia (35 mg&g IP) and aseptic conditions 171. The impfanted sites were assured by X-ray films in the autero-posterior and lateral axes. The electrode assembly and the 4P connector were fixed to the skull with dental acrylic. After recovery, a combination of the microelectrodes was selected by the DIP switch for single unit recordings in both implanted sites. As shown in Pig. 1, microelectrodes A and C can be selected by depressing the individual switches numbered 1 and 6. Signals
through the two channel telemetry system were amplified, monitored on an oscilloscope, stored on magnetic tape, and analyzed using an on-line and off-line signal processor 7T- 17 (Niho~enkis~ei Co. Tokyo). The criteria described in a previous report was adopted in order to insure that the same unit was successively recorded; units were characterized by the amplitude and the duration of the action potential, their responses to sensory inputs or motor activity, response latency and pattern of discharges. If these parameters did not change, the same neuron was assumed to be successively recorded [5, 10, 151. Usually activity from the same single unit was recorded for more than a couple of days. RESULTS AND DISCUSSION
Figure 4 A and B shows an example of single neuron activity in the monkey OBF recorded both through the FM telemetry system described and through the more conventional direct wire procedure using an FET preamplifier [lo].
132
~‘:\X1:\MO’IO
Signals through this telemetry system were comparable to those through direct wire procedures and yield an excellent
S/N ratio. Simultaneous recording in the OBF and in the LHA through this telemetry system is shown in Fig. 5A. The frequency histogram of neuronal activity in the OBF and in the LHA during high fixed-ratio bar press feeding task is shown in Fig. 5B. In this task, a monkey presses a bar 20 times after a cue lamp on and then a reward (raisin) is presented immediately following a cue tone stimulus. The neural activity during the bar press decreased in the OBF, while that in the LHA increased. Compared with conventional methods, our system has the following advantages: (I) no wire cable between the behaving animal and a recording system, (2) long term stable single unit recording with minimal artifacts, (3) easy selection of the combination of microwire electrodes to be used. (4) easy construction and low cost. Other reports of chronically implanted microwire electrodes permitting long term single neuron activity with less artifact than the usual metal and glass microelectrode have been described [3,4, 5, 6, 10, IS], but recording with these electrodes sometimes suffers from vibration and myoelectric potentials during masticatory and body movements. However, employing the method in which one of the microwires serves as a recording electrode and the other as a indifferent electrode can easily eliminate these common-mode artifacts [ 111. Although it is ordinarily troublesome to select one combination of electrodes. the use of
the DIP switch makes the selection
quite c;tsy.
/;1
\I
I hc \clsctcon
of two out of iive electrodes provides ten Lombinations. ‘I‘hc DIP switch used here is compact t 1Ox8k 76 mm). ;tnd c;~\il\ mounted on the head of a monkey. Although other devices have been dc\cribed in the liter+ ture [?. 4. 8. 14). telemetry is not popular because ordinarilh the performance. e.g.. input impedance. voltage gain. S,‘< ratio. dynamic range, covering distance and batrcry lift 14 not sifftcient for single neuronal discharge recording. In atldition. other transmitters are expensive and difficult to ohfain or construct. The FM transmitter described here has been modified from a previous report 131. The oscillator stage has been altered to the C‘olpitts-type for stable oscillation and easy construction. The operating voltage has been changed to 3 V for increasing dynamic mnge. voltage gain. and coverage distance and low distortion. All of the parts used here are popular radio components and arc easy to obtain. Artifacts obtained from the wiring from the animal to the recording instrument and from the contact hetwern the animal and the instrument. which have been difficult to eliminate with the use ofconventional pre-amplifiers 1I. 8. I I. I?. Ih]. can now be easily eliminated by using the FM telemetry 5yhiem.
ACKNOWl.EIX;EMEN
C-G
WC thank Dr. A. Simpson and Dr. H. Grill for help in preparatton
of this manuscript.
REFERENCES 1. Brake], S., T. Babb. J. Mahnke and M. Verzeano. A compact amplifier for extracellular recording. Ph>~sio/&&I* 6: 731-733. 1971. 2. Carson, V. G., R. T. Kado and B. M. Wenzei. A telemeter for moni~o~ng the ele~troc~d~ograms of freely moving mice. Fhvsiof Brhav 8: 561-563, 1972. 3. Chorover, S. L. and A. M. Deluca. A sweet new multiple electrode for chronic single unit recording in moving animals.
Physiol Brhav 9: 671-674, 1972. 4. Eichenbaum.
5.
6.
7. 8.
H., D. Pettijohn, A. M. De&a and S. L. Chorover. Compact miniature mi~roelectrode-telemetry system. Phfsiol B&at* IS: f 175-I 178, 1977. Fontani, G. A technique for long term recording from single neurons in unrestrained behaving animals. Physiol Behav 26: 331-333, 1981. Heym, J., G. F. Steinfels and B. L. Jacobs. Activity of serotonin-containing neurons in the nucleus raphe pallidus of freely moving cats. Brain Res 251: 259-276, 1982. Kusama, T. and M. Mabuchi. ~?e~~fjfaxi~ Atlas qf’rke Brain qf Maccaca Fuscata. Tokyo: University Tokyo Press, 1970. MarshaB, D. A. and G. Celebi. A tunable subminiture biotelemetry transmittter. Physiol Behav 5: 709-712. 1970.
9. Oomura, Y.. H. Ooyama and K. Yoneda. Miniaturized high input impedance preamplifier. Ph~.sh~/Rcho~~2: 93-95. 1%7. IO. Palmer. C. A microwire technique for recording single neurons in unrestrained animals. Brain Rcs Bdl 3: 285-289. 1978. 1 I. Rossetto, M. A. and D. H. VanDercar. Lightweight FET circuit for differential or single-ended recoding in free-moving animals. Phy.tio/ Beho\, 9: 105-106. 1972. I?. Sasaki, K.. T. Ono. H. Nishino. M. Fukuda and K. Muramoto. A method for long-term artifact-free recording of single unit activity in freely moving. eating and drinking animals. .I ,Vcltr~,c~i :Clcrhc& 7: 4.3-47. 1983. 13. Sia. 1.. D. A. MacNeil and E. B. Sigg. A miniature threechannel preamplifier for unit recording in freely moving animals. Ph~siol Beha\, 7: 121-122. 1971. 14. Sperry. C. J.. Jr., C. P. Gadsden, C. Rodriguez and L. M. N. Bach. Miniature subcutaneous frequency-modulated transmitter for brain potentials. .S(.ienc.c,134: 1423-1424, l%l. 15. Strumwasser. F. Long-term recording from single neurons in brain of unrestrained mammals. Scic,rrc.c 127: 469-470. IYS8. 16. Suzuki. H. and M. Takahashi. A method for single unit recording from the free-moving cat. Ph,vsiol Brha1. 13: 331-334. 1974.
Brain Research Bulletin, Vol. 12, pp. 129-132, 1984. Q Ankho International Inc. Printed in the U.S.A.
BRIEF COMMUNICATION
A Telemetry System for Single Neuronal Discharge Recording from Behaving Monkey’ TOMOYA YAMAMOTO, YUTAKA OOMURA,* SHUJI AOU, YASUHIKO NAKANO AND SEIJI NEMOTO
Department
of Biological Control System, National Institute for Physiological Sciences Okazaki 444 and Department of Physiology, Faculty of Medicine, Kyusyu University, Fukuoka 812, (Japan) Received YAMAMOTO, discharge
T., Y. OOMURA,
recording
from
behaving
S. AOU, monkey.
30 June 1983
Y. NAKANO AND S. NEMOTO. BRAIN RES BULL 12(l) 129-132,
for
A telemetry system single neuronal 1984-A telemetry system for neuronal
discharge recording from behaving monkeys positioned in a primate chair is described. Using FM telemetry system single neuronal activity was recorded from the monkey brain through five teflon-coated platinum-iridium microwire electrodes (25 ~1 in dia.). This system permits stable long term neuronal discharge recording during feeding behavior with minimal artifacts even during mastication. The system has a broadcasting distance of more than 30 m. The frequency response and the S/N ratio obtained from this system is almost comparable with that of the usual direct wire procedures, and the elimination of artifacts is superior. This report describes the procedure for constructing the microelectrode assembly and
FM telemetry system. Telemetry
Unit recording
Feeding
Microwire electrodes
Monkey
ing. Microwires should protrude 5 mm from the tip of the tubing. The opposite ends of the microwires are then stripped of its teflon insulation and soldered to the DIP switch as shown in Fig. 1. The tubing containing microwires and the DIP switch are cemented together with dental acrylic. The outputs of the DIP switch are wired to the 4P connector. Prior to implantation of this electrode assembly, the protruding bundle of microwires is fused and coated with melted polyethylene glycol (melting point, 60°C) for the penetration to the brain tissue [3, 4, 5, lo].
IN recent years, there has been an increasing interest in recording the activity of neurons from unanesthetized and unrestrained animals [l, j,4, 5,6,9, 10, 12, 13, 15, 161. The use of conventional metal or glass electrodes requires physical restraint of the experimental animals. These restraint can be stressful for animals especially during long experimental sessions. This report describes a telemetry for single neuronal discharge recording from monkeys under feeding behavior. This system is composed of a microwire electrode assembly and an FM telemetry system. ELECTRODE ASSEMBLY
FM TRANSMITTER
The electrode assembly is composed of five teflon-coated 90% platinum-lo% iridium microwires, (25 p in dia. Medwire Co. NY), a 25 gauge stainless steel tubing, a dual-in-line-
The circuit diagram for the FM transmitter is shown in Fig. 2A, its perspective bottom view is shown in Fig. 2B. The FET Ql in the input stage is used for a high impedence input. Transistors 42 and Q3 in Darlington connection provide sufficient voltage gain and low power consumption. Transisters Q4 and QS form the modified Colpitts-type oscillator. This type oscillator needs no special oscillator coil and is easy to set the broadcasting frequency. The broadcasting frequency of 90 MHz is used because the frequency range of FM receiver is usually 76 MHz to 90 MHz in Japan. It can be changed by the trimmer capacitor C5 and by adjusting the
pack (DIP) switch (Type SPC-010, Saiden Co. Tokyo), and a 4 pin connector (Type AC4S, Echo Co. Tokyo) as shown in Fig. 1. Construction of this assembly proceeds as follows. The 25 gauge stainless steel tubing is cut to an appropriate length for a target site (e.g., for the orbitofrontal cortex (OBF) the length is approximately 30 mm). Five tefloncoated microwires with the length 20 mm longer than that of the stainless tubing are passed through the lumen of the tub-
‘This work was supported partly by Grant-in-Aid (Y.O.) 57440085 and 5637ooO5from the Ministry of Education, Science and Culture in Japan. “Requests for reprints should be addressed to Y. Oomura.
129
130
ELECTRODE ASSEMBLY
I
/ +3v
A
CONNECTOR
ANCHOR
SKULL TEFLON-COATED Pt-k bliCflOWlRE
POLYETHYLENE
GLYCOL
COATING
FIG. 1. Schematic illustration of electrode assembly. Five tefloncoated PT-Ir microwires contained in stainless cannula is fused and coated with polyethylene glycol prior to implantation. Combination of microwires, selected by DIP switch.
spacing of the coil Ll . After all the parts are wired to the DIP pitch circuit board (25x 15 mm) as shown in Fig. 3, the transmitter assembly is coated and filled with common household epoxy resin. When multi-channel simultaneous single neuronal discharge recordings are required, the broadcasting frequency of one discharge should have difference of at least 2 MHz in each other. The cost of all the components in one unit is less than $5. PERFORMANCE
5rlnl FIG. ?A. Transmitter circuit. A: circuit diagram. B: perspective bottom view. Transistor Ql. P-channel junction type FET 2SK3OGR (Toshiba. for audio frequency amplification). Transistors Q2. Q3 and Q4, 2SC1815 (Toshiba. for audio frequency amplification). Transistor QS. 2X828 (Matsushita. for radio frequency amplification, fr=220 MHz). Resistors RI 10 R5. 0.25 W 10% values. RI: 10 K0; R2; 20 MR; R3; 10 KfI; R4: 2 MR; R5: I K0. Capacitors Cl and C2. tantalum type; C3 and C4 ceramic type: CS ceramic type trimmer. Cl, C2; 2.2 pF; C3; SO0 pF; C4: 10 pF: CS; 20 pF (max). Power source. Li-Mn battery type CR-l:3N (Sanyo. 3 V. dia. = I 1.6~ 10.5 mm). Oscillator coil L I, 6 mm in dia. 8 mm long. six turns of copper wire (dia. = I mm) with antenna tap 2 turns from the end leading to QS. Transmitting antenna. 200 mm long copper wire (dia. = 1 mm).
OF FM TRANSMITTER
The input impedance of this transmitter is more than lo9 Ma at 1 kHz and the frequency response is 5 Hz to 15 kHz (-3 dB). The frequency response can be modified by changing the value of capacitors Cl and C2. Input signal levels from 5 PV to 20 mV can be transmitted without noticeable distortion. The voltage gain can be changed by using high hPE transistor as Q2 and Q3. The frequency deviation is 2400 kHz with an input signal of 10 mV p-p at 1 kHz. The current drain is 1.2 mA and the life of the Li-Mn battery is usually more than 25 hr without remarkable change in oscillating frequency and voltage gain. RECEIVING
ANTENNAE
AND RECEIVER
The receiving antenna is a l/2 A vertical dipole type. The receiver is a stereo FM receiver in mono mode (frequency range, 76 MHz to 90 MHz, sensitivity; 1.2 PV IHF 1958 75 a, Type KT-31 Torio Co. Toyko) and can be substituted for usual FM receiver on the market. Overall voltage gain
IC
m
FIG. 3. Photograph of FM tranbmitter before coating with epoxy resin. Transmitting antenna is removed. Li-Mn battery is fixed by wires and can be easily replaced.
131
TELEMETRY FOR NEURONAL DISCHARGE RECORDING B
A TRANSMrllED
DIRECT ws?E
-I
rooyv
lmsec.
tooyv -I 20msec.
FIG. 4. Example of single neuronal activity recorded through FM telemetry system (upper) and through conventional direct wire procedure (lower). A: fast sweep. B: slow sweep.
A OIlF
LtlA
B.P.
FIG. 5A. ~imuI~neous recording in the o~itofron~ cortex (OBF) and in the lateral hypothalamic area (LHA). A: neuronal activity in the OBF (upper) and the LHA (lower) B: frequency histogram of neuronal activity in the OBF (upper) and the LHA (middle) duringhighfmed ratio (F.R.=20) bar press food task. Arrow (t), time of cue lamp on; filled triangle (A), time of reward (raisin). Neuronal activity during bar press (B.P., lower) in the OBF decreases, while that in the LHA increases.
through the FM transmitter and the receiver is approximately 30 dB at 1 kHz and the frequency response is 10 Hz to 13 kHz (-3 dB). The broadcasting rauge is usually more than 30 m. RECORDINGPROCEDURE
The electrode assembly was stereotaxically implanted in the orbitofrontal cortex (OBF, A 33, L 9 and H 7) and the lateral hypothalamic area (LHA, A 19, L 3 and H 2) of A4ucaca Fuscuta (body weight 5.5 kg.) according to the atlas of Kusama and Mabuchi under ~ntob~bit~ anesthesia (35 mg&g IP) and aseptic conditions 171. The impfanted sites were assured by X-ray films in the autero-posterior and lateral axes. The electrode assembly and the 4P connector were fixed to the skull with dental acrylic. After recovery, a combination of the microelectrodes was selected by the DIP switch for single unit recordings in both implanted sites. As shown in Pig. 1, microelectrodes A and C can be selected by depressing the individual switches numbered 1 and 6. Signals
through the two channel telemetry system were amplified, monitored on an oscilloscope, stored on magnetic tape, and analyzed using an on-line and off-line signal processor 7T- 17 (Niho~enkis~ei Co. Tokyo). The criteria described in a previous report was adopted in order to insure that the same unit was successively recorded; units were characterized by the amplitude and the duration of the action potential, their responses to sensory inputs or motor activity, response latency and pattern of discharges. If these parameters did not change, the same neuron was assumed to be successively recorded [5, 10, 151. Usually activity from the same single unit was recorded for more than a couple of days. RESULTS AND DISCUSSION
Figure 4 A and B shows an example of single neuron activity in the monkey OBF recorded both through the FM telemetry system described and through the more conventional direct wire procedure using an FET preamplifier [lo].
132
~‘:\X1:\MO’IO
Signals through this telemetry system were comparable to those through direct wire procedures and yield an excellent
S/N ratio. Simultaneous recording in the OBF and in the LHA through this telemetry system is shown in Fig. 5A. The frequency histogram of neuronal activity in the OBF and in the LHA during high fixed-ratio bar press feeding task is shown in Fig. 5B. In this task, a monkey presses a bar 20 times after a cue lamp on and then a reward (raisin) is presented immediately following a cue tone stimulus. The neural activity during the bar press decreased in the OBF, while that in the LHA increased. Compared with conventional methods, our system has the following advantages: (I) no wire cable between the behaving animal and a recording system, (2) long term stable single unit recording with minimal artifacts, (3) easy selection of the combination of microwire electrodes to be used. (4) easy construction and low cost. Other reports of chronically implanted microwire electrodes permitting long term single neuron activity with less artifact than the usual metal and glass microelectrode have been described [3,4, 5, 6, 10, IS], but recording with these electrodes sometimes suffers from vibration and myoelectric potentials during masticatory and body movements. However, employing the method in which one of the microwires serves as a recording electrode and the other as a indifferent electrode can easily eliminate these common-mode artifacts [ 111. Although it is ordinarily troublesome to select one combination of electrodes. the use of
the DIP switch makes the selection
quite c;tsy.
/;1
\I
I hc \clsctcon
of two out of iive electrodes provides ten Lombinations. ‘I‘hc DIP switch used here is compact t 1Ox8k 76 mm). ;tnd c;~\il\ mounted on the head of a monkey. Although other devices have been dc\cribed in the liter+ ture [?. 4. 8. 14). telemetry is not popular because ordinarilh the performance. e.g.. input impedance. voltage gain. S,‘< ratio. dynamic range, covering distance and batrcry lift 14 not sifftcient for single neuronal discharge recording. In atldition. other transmitters are expensive and difficult to ohfain or construct. The FM transmitter described here has been modified from a previous report 131. The oscillator stage has been altered to the C‘olpitts-type for stable oscillation and easy construction. The operating voltage has been changed to 3 V for increasing dynamic mnge. voltage gain. and coverage distance and low distortion. All of the parts used here are popular radio components and arc easy to obtain. Artifacts obtained from the wiring from the animal to the recording instrument and from the contact hetwern the animal and the instrument. which have been difficult to eliminate with the use ofconventional pre-amplifiers 1I. 8. I I. I?. Ih]. can now be easily eliminated by using the FM telemetry 5yhiem.
ACKNOWl.EIX;EMEN
C-G
WC thank Dr. A. Simpson and Dr. H. Grill for help in preparatton
of this manuscript.
REFERENCES 1. Brake], S., T. Babb. J. Mahnke and M. Verzeano. A compact amplifier for extracellular recording. Ph>~sio/&&I* 6: 731-733. 1971. 2. Carson, V. G., R. T. Kado and B. M. Wenzei. A telemeter for moni~o~ng the ele~troc~d~ograms of freely moving mice. Fhvsiof Brhav 8: 561-563, 1972. 3. Chorover, S. L. and A. M. Deluca. A sweet new multiple electrode for chronic single unit recording in moving animals.
Physiol Brhav 9: 671-674, 1972. 4. Eichenbaum.
5.
6.
7. 8.
H., D. Pettijohn, A. M. De&a and S. L. Chorover. Compact miniature mi~roelectrode-telemetry system. Phfsiol B&at* IS: f 175-I 178, 1977. Fontani, G. A technique for long term recording from single neurons in unrestrained behaving animals. Physiol Behav 26: 331-333, 1981. Heym, J., G. F. Steinfels and B. L. Jacobs. Activity of serotonin-containing neurons in the nucleus raphe pallidus of freely moving cats. Brain Res 251: 259-276, 1982. Kusama, T. and M. Mabuchi. ~?e~~fjfaxi~ Atlas qf’rke Brain qf Maccaca Fuscata. Tokyo: University Tokyo Press, 1970. MarshaB, D. A. and G. Celebi. A tunable subminiture biotelemetry transmittter. Physiol Behav 5: 709-712. 1970.
9. Oomura, Y.. H. Ooyama and K. Yoneda. Miniaturized high input impedance preamplifier. Ph~.sh~/Rcho~~2: 93-95. 1%7. IO. Palmer. C. A microwire technique for recording single neurons in unrestrained animals. Brain Rcs Bdl 3: 285-289. 1978. 1 I. Rossetto, M. A. and D. H. VanDercar. Lightweight FET circuit for differential or single-ended recoding in free-moving animals. Phy.tio/ Beho\, 9: 105-106. 1972. I?. Sasaki, K.. T. Ono. H. Nishino. M. Fukuda and K. Muramoto. A method for long-term artifact-free recording of single unit activity in freely moving. eating and drinking animals. .I ,Vcltr~,c~i :Clcrhc& 7: 4.3-47. 1983. 13. Sia. 1.. D. A. MacNeil and E. B. Sigg. A miniature threechannel preamplifier for unit recording in freely moving animals. Ph~siol Beha\, 7: 121-122. 1971. 14. Sperry. C. J.. Jr., C. P. Gadsden, C. Rodriguez and L. M. N. Bach. Miniature subcutaneous frequency-modulated transmitter for brain potentials. .S(.ienc.c,134: 1423-1424, l%l. 15. Strumwasser. F. Long-term recording from single neurons in brain of unrestrained mammals. Scic,rrc.c 127: 469-470. IYS8. 16. Suzuki. H. and M. Takahashi. A method for single unit recording from the free-moving cat. Ph,vsiol Brha1. 13: 331-334. 1974.