- Email: [email protected]
tyrosine hydroxylase-containing neurons of the rat retina
Brain Research, 214 (1981) 335-347 © Elsevier/North-Holland Biomedical Press
335
T H E ROLE OF GABA IN T H E R E G U L A T I O N OF T H E D O P A M I N E / T Y R O S I N E H Y D R O X Y L A S E - C O N T A I N I N G N E U R O N S OF T H E RAT R E T I N A
PAUL B. MARSHBURN and P. MICHAEL 1UVONE* Departments of Pharmacology and Ophthalmology, Emory University, School of Medicine, Atlanta, Ga. 30322 (U.S.A.)
(Accepted November 6th, 1980) Key words: retina - - dopamine - - tyrosine hydroxylase -- GABA -- muscimol -- picrotoxin --
bicuculline - - amacrine cells
SUMMARY Dopamine (DA) and gamma-aminobutyric acid (GABA) are putative neurotransmitters in two separate populations of amacrine neurons in the mammalian retina. Pharmacological studies have been conducted to determine if GABA neurons regulate the neuronal activity of the neurons that secrete DA. Tyrosine hydroxylase (TH) activity, a biochemical indicator of changes in activity of DA/TH-containing neurons, was low in dark-adapted retinas and high in light-exposed retinas. Muscimol (a GABA receptor agonist) produced a dose-related, biphasic effect on the lightevoked activation of TH, when the drug was injected into the vitreous (intravitreal injection) of dark-adapted rats. At low doses, (35 and 60 pmol) muscimol enhanced the light-evoked activation of TH, but at higher doses (_> 120 pmol) it inhibited the light-evoked increase in enzyme activity. Muscimol had no significant effect on the T H activity of dark-adapted retinas. GABA antagonists, bicuculline and picrotoxin, produced effects on T H activity that were dependent on both dose and light-exposure. At low doses (0.443.5 nmol), bicuculline and picrotoxin both inhibited the lightevoked activation of TH, but had no effect on T H activity of the dark-adapted retinas. At a higher dose (2.0 nmol), both antagonists increased T H activity in the darkadapted retina and attenuated the further activation of the enzyme by light. Rat retinas were dissociated into suspensions of viable cells in order to investigate the direct effects of muscimol and picrotoxin on the DA/TH-containing cells. The process of dissociating dark-adapted retinas resulted in an apparent activation of TH. * To whom correspondence should be addressed.
336 Incubation of the cells with muscimol resulted in a decrease of TH activity in a concentration-dependent manner. Picrotoxin antagonized the inhibitory effect of muscimol, but had no effect when incubated alone. The biphasic effects of GABA agonists and antagonists in vivo suggest that a certain subpopulation of GABA neurons are involved in the activation of the DA/THcontaining neurons by photic stimulation, while another subpopulation of GABA neurons produce a tonic inhibition of the DA/TH-containing neurons in darkness. The experiments with retinal cell suspensions indicate that the tonic inhibition is probably mediated by synapses of GABA neurons directly onto the DA/THcontaining cells. INTRODUCTION Dopamine (DA) is the putative neurotransmitter of a population of amacrine neurons in the retina of several mammalian species11,la,15,26, and appears to be the only catecholamine in retinal neurons of the rat 16,17. Light exposure increases the rate of DA biosynthesis9,16,29 and release 19 from retina, indicating that these neurons are activated by photic stimulation, and are presumably involved in the neuronal processing of visual information. Tyrosine hydroxylase (TH; EC 1.14.16.2), the rate-limiting enzyme of catecholamine biosynthesisza, is activated concomitantly with the light-evoked increase of DA biosynthesis16,17. Retinal TH activity increases rapidly during photic stimulation (within 1 min), and is maximal following 5 min of light exposure. TH activity decreases following termination of photic stimulation. Thus, changes of TH activity appear to be a biochemical indicator of changes in the neuronal activity of DA/TH-containing amacrine cells. The DA/TH-containing amacrine cells apparently receive their synaptic input exclusively from other amacrine cells10. Gamma-aminobutyric acid (GABA) is the putative neurotransmitter of another population of amacrine neurons 2-4,u,~a,ls. Therefore, we examined the effects of GABA agonists and antagonists on the regulation of retinal TH activity to determine if the GABA neurons modulate the neuronal activity of the DA/TH-containing amacrine cells. MATERIALS AND METHODS Animals. Male Sprague-Dawley-derived rats (Zivic-Miller, Allison Park, Pa.),
150-250 g, were housed in clear plastic cages with perforated metal tops under a 12 h light-dark cycle,'and were given free access to food and water. Prior to drug administration, rats were dark-adapted for 3-14 h in a room with a double-door entrance that permitted access to the room without light-leakage. All manipulations of the darkadapted rats were performed under dim red light (Kodak Wratten No. 92); this condition is designated in the text as 'dark'. Materials. Picrotoxin, GABA, reduced nicotinamide adenine dinucleotide phosphate, catalase, bovine serum albumin, deoxyribonuclease I (DN-EP), coUagenase
337 type IA, hyaluronidase type Ili, and papain type llI were purchased from Sigma Chemicals, St. Louis, Mo. Pyridoxal phosphate and OL-6-methyl-5,6,7,8,-tetrahydropterine (6-MPH4) were obtained from Calbiochem, LaJolla, Calif. Bicuculline methiodide was obtained from Pierce Chemicals, Rockford, I11. L-[1-14C]tyrosine, 57 mCi/mmol, was purchased from Amersham, Arlington Heights, I11. Muscimol was obtained from Smith Kline and French Laboratories, Philadelphia, Pa. All other chemicals were obtained in the purest form available. Drug administration. Drugs were dissolved on 0.9 ~ NaC1 immediately before use, and were administered by subcutaneous (s.c.) or intravitreal injections. Subcutaneous injections were delivered via a 25-gauge syringe needle to the subcutaneous space approximately 15 mm posterior to the animal's head on the dorsal surface. For intravitreal injections, rats were anesthetized with ether. Injections were delivered with a Hamilton beveled-syringe needle (26S-gauge) that had a 20-gauge cannula covering all but the last 7 mm at the beveled end. The needle was inserted into the vitreous from the nasal quarter of the eye, approximately 1 mm below the border of the cornea and sclera. The needle was visually located in center of the vitreous and was positioned so that the bevel was directed towards the lens in order to minimize damage to the retina. Five #1 of solution were injected over a period of approximately 10 sec. All injections were administered in the dark. In the intravitreal-drug experiments, one eye of each rat was injected with drug solution and the other eye with saline, to serve as a paired control. Photic stimulation. Rats were exposed to light, following recovery from anesthesia, for the periods indicated in the table and figure legends. Photic stimulation was provided by fluorescent tubes (35 W Sylvania Super-Saver) suspended 1.5 m above the cages. Light-intensity inside the cage was estimated to be approximately 400 lux by measuring the reflected light from a neutral gray card with a photometer. Retinal dissociation. The dissociation of dark-adapted retinas into suspensions of viable single cells was performed by a modification of the technique of Sarthy and Lam3°. Eyes were hemisected and the posterior chambers incubated at 37 °C for 15 min in mammalian Ringer's solution containing hyaluronidase (100 U/ml) and coUagenase (180 U/ml). The eye cups were washed in fresh Ringer's solution and the retinas removed from the eye cups. The retinas were incubated for 30 min at 37 °C in Ca2+-free Ringer's solution containing papain (50/~g/ml). The retinas were washed in normal Ringer's solution and transferred to conical centrifuge tubes containing Ringer's solution (0.125 ml/retina), deoxyribonuclease I (320 U/ml), and 0.1 ~o bovine serum albumin (fraction V). The retinas were dissociated by gently pipetting the tissue through a pasteur pipette. The undissociated pieces of retina settled for 5 min, and the top half of the supernatant containing the dissociated ceils was removed and transferred to a chilled tube on ice. The remaining pieces of retina and supernatant were gently pipetted up and down several times, allowed to settle, and the top half of the supernatant removed and pooled with the supernatant from the previous step. All Ringer's solutions contained 10 mM D-glucose and were gassed with 95 ~ O2-5~ CO2. This procedure gave suspensions in which greater than 95 ~ of the cells were present as single cells.
338 TABLE I
Effect of GABA on retinal tyrosine hydroxylase activity Dark-adapted rats were injected intravitreally with 0.9 ~ NaC1 and GABA (5 mnol) as described in Methods. Rats were sacrificed t5 min after injection, either in the dark, or after 5-9 min (~ -- 7 rain) of light exposure. TH activity is expressed as pmol CO2/min/mg protein; values represent the mean :t: S.E.M. The number in parentheses represent the n.
Injection
Saline GABA
Tyrosine hydroxylase activity Dark
Light
1.4:5 0.1 (5) 1.5 4- 0.1 (5)
4.1 4- 0.3 (4) 3.4 4- 0.2 (4)*
* P < 0.05, 2-tailed, paired Student's t-test, compared to matched saline control.
Aliquots of the retinal cell suspension were transferred into ice-chilled roundbottom centrifuge tubes containing freshly prepared control or test Ringer's solutions, and gently agitated for 30 min at 37 °C in a Dubnoff Metabolic Shaker under an atmosphere of 95 ~ 02-5 ~ CO2. Following the drug incubation period, the tubes were chilled on ice, and centrifuged for 30 min at 250 × g at 0-4 °C. The supernatants were discarded and the pellets were frozen on dry-ice for tyrosine hydroxylase assay. Dissociation and incubation of retinal suspensions were performed under dim red light (Kodak Wratten No. 92). Tyrosine hydroxylase assay. Retinas were dissected and frozen on dry-ice as described previously 17. Frozen retinas or pelletted retinal cells were homogenized in ice-cold 50 m M Tris acetate, p H 6.0, containing 0.2~/o Triton X-100 in glass-teflon homogenizers. Homogenates were centrifuged at 32,000 x g for 15 min at 0-4 °C, and tyrosine hydroxylase was measured in the supernatants by a modification of the D O P A decarboxylase-coupled assay of Waymire et al. 31, as described previously17; [14C-carboxyl]tyrosine was used as substrate, and [14C]CO2 was the product. 6-MPH4 was dissolved at a concentration of 10 mM in 0.01 N HCI and reduced under hydrogen gas in the presence of platinum (IV) oxide. The activation of retinal T H by brief-light expOsure is characterized by a decrease in the apparent Km of the enzyme for the pteridine cofactor17; therefore, in order to demonstrate the light-evoked activation of TH, a subsaturating concentration of 6-MPH4 (0.2 mM) was used. The concentration of L-tyrosine was 0.1 mM and the p H was 7.0. Protein was determined by the method of Lowry et al. 24. RESULTS
Effbct of GABA agonists and antagonists in vivo Retinal T H activity was relatively low in dark-adapted rats and was markedly stinmlated by light exposure (Table I). When administered intravitreally to darkadapted rats, 5 nmoles of G A B A significantly reduced the light-evoked activation of retinal TH, even though it did not alter the activity of the enzyme in the dark (Table I).
339
150-
g, 125Z 0
Z ¢
10075-
5025* p < 0.01
MUSCIMOL (log pmole)
Fig. 1. Effect of muscimol on light-stimulated tyrosine hydroxylase activity. Dark-adapted rats were injected intravitreally with 0.9~ NaC1 and muscimol as described in Methods. Subjects were sacrificed and retinas dissected 40 min after injection, following 16-24 min (~ = 20 min) of photic stimulatiOn. Data are expressed as the mean and S.E.M. of the percentage of TH activity in the muscimol-tr~ated retina relative to the saline-treated retina of each animal. The sample size for each dose was 4-9. The TH activity of all of the control retinas was 5.3 ± 0.25 pmol CO2/min/mgprotein (~ ± S.E.M.).t*P ~ 0.01, 2-tailed, paired Student's t-test, compared to matched control; A dose-response curve for muscimol, a GABA-receptor agonist14,21, 22, was determined to more fully characterize the effects of GABA receptor stimulation on the regulation of retinal T H activity. When administered by intravitreal injection prior to photic stimulation, muscimol had a dose-dependent, biphas!c effect on the lightevoked activation of retinal T H (Fig. 1). Low doses of muscimol (35 and 60 pmol) significantly increased the activity of T H measured following photic stimulation. In contrast, higher doses of the GABA-receptor agonist (_> 120 pmol) significantly inhibited the activation of the enzyme by light. When compared to dark-adapted controls, 300 pmoles of muscimol completely inhibited the light-evoked stimulation of TH. Muscimol had no effect on the activity of retinal T H in the dark-adapted retinas, at the doses tested (60 and 900 pmol). To determine if the inhibitory effect of muscimol was mediated by an interaction with GABA receptors, we examined the effects of systemically administered picrotoxin, a GABA antagonistS, 20, on the inhibition of retinal T H by 400 pmoles of muscimol, administered intravitreally (Fig. 2). Picrotoxin (4 #mol/kg) significantly antagonized the inhibitory effect of muscimol on T H activity. In this experiment, and in attempts to antagonize the stimulatory effect of low doses of muscimol, we observed that systemically administered picrotoxin inhibited the activation of T H by light. Further investigation demonstrated that picrotoxin suppressed the light-evoked stimulation of T H activity in a dose-dependent manner (Fig. 3). To avoid the convulsions produced by systemic administration of picrotoxin, the effects of the GABA antagonist on the activity of T H in the dark and the light were examined following intravitreal administration o f picrotoxin to dark-adapted rats (Table II). Picrotoxin, at a dose of 0.5 nmoles, significantly decreased the activity of T H in the light-exposed retinas, but had no significant effect in the dark. A higher dose of
340
I
3
d~
:¢
O O
2
-6
1
CONTROL
MUSCIMOL
MUSCIMOL & PICROTOXlN
Fig. 2. Antagonism by picrotoxin of the inhibitory effect of muscimol on light-stimulated tyrosine hydroxylase activity• Dark-adapted rats were injected subcutaneously with 0.9 ~ NaCI or picrotoxin (4/~mol/kg). Ten rain after the subcutaneous injection, the subjects were injected intravitreally with 0.9 % NaCI or with muscimol (0.4 nmol). Subjects were sacrificed and retinas dissected 25 min after the intravitreal injections, following 12-18 rain (£ = 15) of light exposure• Data are expressed as mean S.E.M., N = 6. *P < 0.01, Dunnett's test.
picrotoxin, 2.0 nmoles, significantly activated TH in the dark. The activity of the enzyme in the light-exposed retinas was not significantly different from control after the higher dose of picrotoxin; however, the stimulation of enzyme activity by light was reduced in the drug-treated retinas (saline ---- 92 %, picrotoxin = 29 %).
I
c
E
~.
3"
E O O e o
J_ 2-
E
P1
P4
P8
,umole/kg (s.c.)
Fig. 3. Antagonism of light-evoked activation of tyrosine hydroxylase by picrotoxin. Rats were injected subcutaneously with 0.9 % NaCI or picrotoxin (1, 4, or 8 #mol/kg). Subjects were sacrificed and retinas dissected 35 min after injection, following 11-19 min (£ = 15) of photic stimulation. Data are expressed as mean 4- S.E.M., n -- 8-9. *P < 0.05, Dunnett's test.
341 TABLE II
Effect of intravitreal picrotoxin on retinal tyrosine hydroxylase activity Dark-adapted rats were injected intravitreally with 0.9 % NaCI and picrotoxin as described in Methods. Rats were sacrificed 22 min after injection, either in the dark or after 8-14 rain (~ = 11 min) of light exposure. TH activity is expressed as pmol CO2/min/mg protein; values represent the mean and S.E.M. The number in parentheses represents the n.
Injection
Dose (nmol)
Tyrosine hydroxylase activity Dark
Light
Experiment 1 Saline Picrotoxin
-0.5
1.9 4- 0.35 (7) 2.6 4- 0.25 (7)
6.1 ± 0.59 (6) 5.2 4- 0.40 (6)*
Experiment 2 Saline Picrotoxin
-2.0
2.5 -4- 0.09 (5) 4.1 4- 0.17 (5)*
4.8 4- 0.40 (6) 5.3 4- 0.42 (6)
* P < 0.05, 2-tailed, paired Student's t-test, compared to matched saline control.
Bicuculline, another GABA
antagonist6,7, s, h a d a n effect s i m i l a r to t h a t o f
p i c r o t o x i n a f t e r i n t r a v i t r e a l a d m i n i s t r a t i o n ( T a b l e III). Bicuculline, at a d o s e o f 0.4 n m o l e s , significantly r e d u c e d the a c t i v i t y o f T H in the l i g h t - e x p o s e d retinas, a n d h a d n o s i g n i f i c a n t effect in t h e d a r k . A t a d o s e o f 2.0 n m o l e s , b i c u c u l l i n e significantly i n c r e a s e d the e n z y m e a c t i v i t y in t h e d a r k , a n d a t t e n u a t e d the f u r t h e r a c t i v a t i o n o f the e n z y m e b y light.
TABLE III
Effect of bicuculline on retinal tyrosine hydroxylase activity Dark-adapted rats were injected intravitreally with 0.9 % NaCI and bicuculline methiodide as described in Methods. Rats were sacrificed 40 min after injection, either in the dark or after 12-18 rain (~ = 15 min) of light exposure. TH activity is expressed as pmol C02/min/mg protein; values represent the mean 4- S.E.M. The number in parentheses represents the n.
Injection
Dose (nmol)
Tyrosine hydroxylase activity Dark
Light
Experiment I Saline Bicuculline
-0.4
2.1 4- 0.2 (9) 2.3 4- 0.3 (9)
8.3 4- 0.6 (11) 7.4 4- 0.4 (11)*
Experiment 2 Saline Bicuculline
-2.0
2.9 4- 0.5 (8) 5.3 4- 0.5 (8)*
7.9 -4- 0.5 (8) 8.2 4- 0.5 (8)
* P < 0.05, 2-tailed, paired Student's t-test, compared to matched saline control.
342 TABLE IV
Comparison of tyrosine hydroxylase activity extracted from retinal cell suspensions and intact retina Dark-adapted retina were dissected, frozen and homogenized as described in Methods. Other darkadapted retinas were dissociated into suspensions of cells, incubated at 37 °C under an atmosphere of 95 % 03/5 % COs for 30 min, centrifuged, frozen and homogenized as described in Methods. Data are expressed as ~ 4- S.E.M.
Intact retina Retinal cell suspensions
n
Tyrosine hydroxylase activity (pmol COz/min/mg protein)
16 14
2.3 :t: 0.1 3.9 d- 0.4*
* P < 0.01, 2-tailed Student's t-test.
Effect of muscimol and picrotoxin in retinal cell suspensions Retinas were dissociated into suspensions of cells to investigate the direct effects of muscimol and picrotoxin on the DA/TH-containing cells. Examination of the cell suspensions with inverted-stage phase microscopy indicated that greater than 95 ~ of the cells in the suspension were present as single cells. After the cell suspensions were centrifuged at 250 x g, substantial tyrosine hydroxylase activity was found in homogenates of the pelleted cells. However, tyrosine hydroxylase activity was not found in the supernatant from the 250 x g centrifugation, or in the resuspended pellet obtained from recentrifugation of the 250 × g supernatant at 35,000 x g for 20 min. These results indicate that the majority of enzyme remained in intact cells, in somas, or in large processes, rather than being released into the media by lysis of cells or being localized to free synaptosomes. q¢
f
3O z o
"r" z
20
lO
6
lO'SM
IO';'M
10-61~
MUSCIMOL
Fig. 4. Effect of muscimol on tyrosine hydroxylase activity of retinal cell suspensions. Retinas were dissociated into suspensions of single cells as described in Methods. For each suspension, triplicate aliquots were incubated in Ringer's solution for control, and triplicate aliquots were incubated in Ringer's solution containing a given concentration of muscimol. The per cent inhibition of TH activity induced by a given concentration of muscimol was obtained by comparing the effect of muscmiol/Ringer's solution and control/Ringer's solution from the same cell preparation. The data represent the mean 4- S.E.M. for 4-6 experiments. The TH activity of all of the controls was 3.9 4- 0.4 pmol C02/min/mg protein (~ 4- S.E.M.). Muscimol, at doses of 10-7 M and 10-6 M produced significant decreases in T H activity. *P < 0.05, 2-tailed, paired Student's t-test compared to matched control.
343 TABLE V Picrotoxin-blockade of the effect of muscimol on tyrosine hydroxylase activity of dissociated retinal suspensions
Retinas weredissociatedinto suspensions of singlecellsas describedin Methods. From each suspension, triplicate aliquots were incubated in Ringer's solution or Ringer's solution containing muscimol and picrotoxin at 37 °C under an atmosphere of 95 % 02/5 % CO2.After a 30 rain incubation, the suspensions were centrifuged, and the precipitates were frozen, homogenized and assayed for TH activity as described in Methods. Data are expressed as ~ 4- S.E.M. of 4 separate experiments. Condition
n
Control 4 Muscimol (10-~ M) -~- picrotoxin(2.4 × 10-SM) 4
Tyrosine hydroxylase activity (pmol C02/min/mg protein)
3.5 4- 0.2 3.4 4- 0.2
Muscimol's inhibitory effect on T H activity is probably mediated through GABA receptors that are located on the DA/TH-containing neurons, for the following reasons. The T H activity measured in homogenates of cell suspensions of darkadapted retinas prepared under dim red light was significantly higher than that measured in homogenates of intact dark-adapted retinas (Table IV). When cell suspensions were incubated in the presence of muscimol, T H activity was decreased in a concentration-dependent manner (Fig. 4). In contrast to the effect of muscimol in vivo, only inhibition, and not facilitation, of enzyme activity was observed over a 100fold concentration range. The decrease in T H activity was not caused by a direct inhibition of the enzyme, because muscimol (10 -6 M) added directly to the supernatant of the homogenized cells had no effect on enzyme activity. Furthermore, the inhibitory effect of incubating the cells with muscimol was completely antagonized by picrotoxin (Table V), indicating that the inhibitory effect of muscimol on T H activity was mediated through GABA receptors. Incubation of the cells with picrotoxin alone had no effect on T H activity. DISCUSSION Dopamine (DA) is localized within a population of amacrine cells in the mammalian retina11,lZ,15, 26. Light exposure enhances both DA biosynthesisg, 16,29 and release 19, and also activates retinal tyrosine hydroxylase16,17, indicating that these neurons are activated by photic stimulation. Anatomical evidence suggests that the DA/TH-containing neurons receive their synaptic input exclusively from other amacrine cells 10. Here we have presented pharmacological evidence that GABA, a putative neurotransmitter in another population of amacrine cells 2-4,11,13,18, is involved in the regulation of the DA/TH-containing neurons. GABA and muscimol, a GABA receptor agonist 14,21,z2, altered the light-evoked stimulation of TH, while having no influence on the enzyme activity in the retinas of dark-adapted rats. The light-evoked stimulation of T H was enhanced by low doses of muscimol (20--60 pmol), but inhibited at higher doses (120-900 pmol). The inhibitory
344 effect of high doses of muscimol was apparently mediated by an interaction with GABA receptors, since the GABA antagonist, picrotoxin, blocks this effect of muscimol. Intravitreal administration of two GABA antagonists, picrotoxin and bicuculline, increased TH activity in dark-adapted retinas and attenuated the further activation of the enzyme by light. These biphasic effects of GABA receptor agonists and antagonists suggest two distinct roles for GABA in the regulation of retinal DA/TH-containing neurons. First, these neurons appear to be tonically inhibited in the dark. The tonic inhibition is at least partially mediated by GABA, because intravitreal administration of GABA antagonists elicited an increase in TH activity in the dark-adapted retina, presumably by blocking the effect of endogenously released GABA. Furthermore, both GABA and muscimol block the light-evoked activation of TH when administered prior to photic stimulation. The observation that muscimol has no effect on TH activity in the dark-adapted retinas suggests that the GABA-mediated inhibitory influence on the DA/TH-containing neurons is probably maximal in the dark. A second role of GABA is in the activation of retinal DA/TH-containing neurons as a consequence of light exposure, for the following two reasons. First, picrotoxin and bicuculline, GABA antagonists, blocked the light-evoked stimulation of TH. Secondly, intravitreal administration of low doses of muscimol (20-60 pmol), an agonist, produced an enhancement of the light-evoked enzyme activity. Because GABA receptor agonists produce direct inhibitory effects upon central neuronal activity6,7, 22, the enhancement of TH activity is probably mediated by an action of muscimol on neurons other than the DA/TH-containing cells; the locus or mechanism for this action of muscimol to increase TH activity cannot be determined from our data. Higher doses of intravitreally administered muscimol, which inhibited light-evoked stimulation of TH activity, presumably act at the site where endogenously released GABA exerts its inhibitory influence on DA/TH-containing amacrine cells in the dark. Support for this hypothesis is the observation that the higher doses of muscimol decreased TH activity in light-stimulated retinas to the level of that measured in the retinas of dark-adapted rats. A working hypothesis, based on the data outlined below, is that GABAcontaining neurons synapse directly on the DA/TH-containing neurons and inhibit the neuronal activity of these cells. The dissociation of retinas into suspensions of cells appears to disrupt most connections between cells (more than 95 ~o of the cells were present as single cells). The DA/TH-containing neurons were released from a tonic inhibitory influence in the process of dissociation, as indicated by the observation that the TH activity of cell suspensions was greater than that found in intact retinas when both preparations had been obtained from dark-adapted subjects. This hypothesis is supported by the observation that TH activity was not affected when the cells were incubated with picrotoxin. Muscimol decreased the TH activity of the cell suspensions in a concentration-dependent fashion; this suppression of TH activity was blocked by picrotoxin. Stimulation of enzyme activity by muscimol was not observed in the cell suspensions. The TH activity of the cell suspensions appears to be regulated by the membrane potential of the cells, as indicated by the observation that K+-induced
345 depolarization of the cells results in an activation of the enzyme (Iuvone and Marshburn, unpublished observation). Thus, the inhibitory action of muscimol on TH activity is probably mediated by an interaction with GABA receptors located on the DA/TH-containing neurons, and is presumably a consequence of hyperpolarization of these cells. Our data are in agreement with the data of Morgan and Camp 29, who found that muscimol inhibited the light-evoked increase in DA turnover in the rat retina, as measured by the rate of decline of the concentration of retinal DA following inhibition of TH by a-methyl-p-tyrosine. However, these investigators found that multiple doses of muscimol, given intravenously, decreased DA turnover below the level of that found in dark-adapted controls. In contrast, we found that muscimol, administered intravitreally, decreased TH activity to the level of that in dark-adapted retinas, but not below that level. One possible explanation for this difference may be related to the different measurements employed in the two studies; perhaps DA turnover can be furthel decreased by inhibition of release or metabolism when TH is in an unactivated state. Another possibility is that the effect observed previously 29 is caused by a metabolite of muscimol, because multiple doses of muscimol were administered intravenously, and because muscimol is rapidly metabolized following intravenous injection 25. The biphasic effects of intravitreally administered GABA agonists and antagonists were dose-dependent. Low doses of agonist and antagonists appeared to mimic and antagonize, respectively, the stimulatory effect of GABA on the activation of the DA/TH-containing neurons by light. High doses of agonist and antagonists appeared to mimic and antagonize, respectively, the inhibitory effect of GABA on the DA/THcontaining neurons. These data suggest that the stimulatory influence of GABA may be mediated via high-affinity receptors, while the inhibitory influence is mediated via low-affinity ~eceptors. In support of this explanation, high- and low-affinity [3H]GABA binding-sites ('receptors') have been observed in the mammalian retina t2. Alternatively, the dose-related effects of GABA agonists and antagonists may be related to different intraretinal locations of the receptors that mediate the presumed stimulatory and inhibitory effects of GABA. Our observation of the biphasic effects of GABA agonists and antagonists on the activity of TH, as well as the pharmacological, biochemical, and electrophysiological studies of other investigators, provide the basis for the following hypothetical model. The mammalian retina appears to contain at least two distinct populations of GABA neurons, one that is active in the dark and one that is active in the light. GABA antagonists increase the spontaneous activity of some retinal ganglion cells 1, increase acetylcholine release27, zs, and increase TH activity in the dark-adapted retina, providing pharmacological evidence for the synaptic release of GABA in the dark. The release of [ZH]GABA from preloaded retinas is enhanced by photic stimulation 2, and GABA antagonists decrease the light-evoked activation of tyrosine hydroxylase, providing both biochemical and pharmacological evidence for synaptic release of GABA in response to photic stimulation. We propose that some of the GABA neurons that are active in the dark synapse directly onto the DA/TH-containing
346 neurons and tonically inhibit the neuronal activity of these cells in darkness. During photic stimulation, G A B A release from these neurons is inhibited, thereby removing the tonic inhibitory influence on the D A / T H - c o n t a i n i n g cells. Photic stimulation m a y also activate the release o f G A B A from other G A B A neurons, and in some way, perhaps by inhibiting an interneuron that inhibits the D A / T H - c o n t a i n i n g cells, augments the activation of the ceils by light. ACKNOWLEDGEMENTS We wish to thank Dr. N o r t o n H. Neff, N I M H , for helpful discussions and for assistance on pilot studies related to this project. We thank Dr. Henry Haigler, E m o r y University, for assistance in the preparation of this manuscript. We thank L. K. Holmes for technical assistance, and Dr. Peter Ridley, Smith Kline and French Laboratories, for the gift o f muscimol. P.B.M. was supported by N I H Training G r a n t 5-TO-1-GM-00179. P.M.I. was partially supported by grants from the Pharmaceutical Manufacturers Association F o u n d a t i o n and the N I H , 5S07-RR-05364.
REFERENCES 1 Ames, A., III and Pollen, D. A., Neurotransmission in central nervous tissue: a study of isolated rabbit retina, J. Neurophysiol., 32 0969) 424-442. 2 Bauer, B. and Ehinger, B., Light evoked release of radioactivity from rabbit retinas preloaded with [aH]GABA, Experimentia (Basel), 33 (1977) 470-471. 3 Brandon, C., Lam, D. M. K. and Wu, J-Y., The y-aminobutyric acid system in rabbit retina: Localization by immunocytochemistry and autoradiography, Proc. nat. Acad. Sci. (Wash), 76 (1979) 3557-3561. 4 Brunn, A. and Ehinger, B., Uptake of certain possible neurotransmitters into retinal neurones of some mammals, Exp. Eye Res., 19 (1974) 435-447. 5 Curtis, D. R. and Johnston, G. A. R., Amino acid transmitters. In A. Lajtha (Ed.), Handbook of Neurochemistry, Vol. 4, Plenum Press, New York, 1970, pp. 115-135. 6 Curtis, D. R., Duggan, A. W., Felix, D. and Johnston, G. A. R., GABA, bicuculline and central inhibition, Nature, (Lond.), 226 (1970) 1222-1224. 7 Curtis, D. R., Duggan, A. W., Felix, D. and Johnston, G. A. R., Bicucullin, an antagonist of GABA and synaptic inhibition in the spinal cord of the cat, Brain Research, 32 (1971) 69-96. 8 Curtis, D. R., Duggan, A. W., Felix, D., Johnston, G. A. R. and McLennan, H., Antagonism between bicuculline and GABA in the cat brain, Brain Research, 33 (1971) 57-73. 9 Da Prada, M., Dopamine content and synthesis in retina and n. accumbens septi: pharmacological and light-induced modifications, Advance Biochem. Psychopharmacol., 16 (1977) 311-319. 10 Dowling, J. E., and Ehinger, B., Synaptic organization of the dopaminergic neurons in the rabbit retina, J. comp. Neurol., 180 0978) 203-220. 11 Ehinger, B. and Falck, B., Autoradiography of some suspected neurotransmitter substances: GABA, glycine, glutamic acid, histamine, dopamine, and L-DOPA, Brain Research, 33 (1971) 157-172. 12 Enna, S. J. and Snyder, S. H., Gamma-aminobutyric acid (GABA) receptor binding in mammalian retina, Brain Research, 115 (1976) 174-179. 13 Graham, L. T., Comparative aspects of neurotransmitters in the retina. In H. Dawson, and L. T. Graham (Eds.) The Eye, Comparative Physiology, Vol. 6, Academic Press, New York, 1974, pp. 283-342. 14 Greenlee, D. V., Van Ness, P. C. and Olsen, R. W., Gamma-aminobutyric acid binding in mammalian brain: receptorqike specificity of sodium-independent sites, J. Neurochem., 31 (1978) 933-938.
347 15 Haggendal, J. and Malmfors, T., Identification and cellular localization of the catecholamines in the retina and choroid plexus of the rabbit, Acta physioL scand., 64 (1965) 58-66. 16 Iuvone, P. M., Galli, C. L., Garrison-Gund, C. K. and Neff, N. H., Light stimulates tyrosine hydroxylase activity and dopamine synthesis in retinal amacrine neurons, Science, 202 (1978) 901-902. 17 Iuvone, P. M., Galli, C. L. and Neff, N. H., Retinal tyrosine hydroxylase: comparison of shortterm and long-term stimulation by light, Molec. PharmacoL, 14 (1978) 1212-1219. 18 Kojima, K., Mizuno, K. and Miyazaki, M., Gamma-amino-butyricacid in ocular tissue, Nature (Lond.), 181 (1958) 1200-1201. 19 Kramer, S. G., Dopamine: A retinal neurotransmitter. I. Retinal uptake, storage and light stimulated release of H3-dopamine in vivo, Invest. OpthalmoL, 10 (1971) 438-452. 20 Krnjevfc, K., Chemical nature of synaptic transmission in vertebrates, PhysioL Rev., 54 (1974) 418-540. 21 Krogsgaard-Larsen, P. and Johnston, G. A. R., Structure-activity studies on the inhibition of GABA binding to rat brain membranes by muscimol and related compounds, J. Neurochem., 30 (1978) 1377-1382. 22 Krogsgaard-Larsen, P., Johnston, G.A.R., Curtis, D. R., Game, C. J. A. and McCulloch, R. M., Structure and biologic activity of a series of conformationally restricted analogues of GABA, J. Neurochem., 25 (1975) 803-809. 23 Levitt, M., Spector, S., Sjoerdsma, A. and Udenfriend, S., Elucidation of the rate-limiting step in noradrenaline biosynthesis in the perfused guinea-pig heart, J. PharmacoL exp. Ther., 148 (1965) 1-8. 24 Lowry, O. H., Rosenbrough, N. J., Farr, A. L., and Randall, R. J., Protein measurement with the Folin phenol reagent, J. bioL Chem., 193 (1951) 265-275. 25 Maggi, A. and Enna, S. J., Characteristics of muscimol accumulation in mouse brain after systemic administration, Neuropharmacology, 18 (1979) 361-366. 26 Malmfors, T., Evidence of adrenergic neurons with synaptic terminals in the retina of rats demonstrated with fluorescence and electron microscopy, Actaphysiol. scand., 58 (1963) 99-100. 27 Massey, S. C. and Neal, N. J., The light evoked release of acetylcholine from rabbit retina in vivo and its inhibition by 7-aminobutyric acid, J. Neurochem., 32 (1979) 1327-1329. 28 Massey, S. C. and Redburn, D. A., GABA inhibition of light evoked ACh release, Invest. Ophthalmol., Suppl. ARVO, (1980) 283. 29 Morgan, W. W. and Kamp, C. W., A GABAergic influence on the light-induced increase in dopamine turnover in the dark-adapted rat retina in vivo, J. Neurochem., 34 (1980) 1082-1086. 30 Sarthy, P. V. and Lam, D. M. K., Isolated cells from a mammalian retina, Brain Research, 176 (1979) 208-212. 31 Waymire, J. C., Bjur, R. and Weiner, N., Assay of tyrosine hydroxylase by coupled decarboxylation of Dopa formed from (1-14C)-L-tyrosine,Analyt. Biochem., 43 (1971) 588-600.
335
T H E ROLE OF GABA IN T H E R E G U L A T I O N OF T H E D O P A M I N E / T Y R O S I N E H Y D R O X Y L A S E - C O N T A I N I N G N E U R O N S OF T H E RAT R E T I N A
PAUL B. MARSHBURN and P. MICHAEL 1UVONE* Departments of Pharmacology and Ophthalmology, Emory University, School of Medicine, Atlanta, Ga. 30322 (U.S.A.)
(Accepted November 6th, 1980) Key words: retina - - dopamine - - tyrosine hydroxylase -- GABA -- muscimol -- picrotoxin --
bicuculline - - amacrine cells
SUMMARY Dopamine (DA) and gamma-aminobutyric acid (GABA) are putative neurotransmitters in two separate populations of amacrine neurons in the mammalian retina. Pharmacological studies have been conducted to determine if GABA neurons regulate the neuronal activity of the neurons that secrete DA. Tyrosine hydroxylase (TH) activity, a biochemical indicator of changes in activity of DA/TH-containing neurons, was low in dark-adapted retinas and high in light-exposed retinas. Muscimol (a GABA receptor agonist) produced a dose-related, biphasic effect on the lightevoked activation of TH, when the drug was injected into the vitreous (intravitreal injection) of dark-adapted rats. At low doses, (35 and 60 pmol) muscimol enhanced the light-evoked activation of TH, but at higher doses (_> 120 pmol) it inhibited the light-evoked increase in enzyme activity. Muscimol had no significant effect on the T H activity of dark-adapted retinas. GABA antagonists, bicuculline and picrotoxin, produced effects on T H activity that were dependent on both dose and light-exposure. At low doses (0.443.5 nmol), bicuculline and picrotoxin both inhibited the lightevoked activation of TH, but had no effect on T H activity of the dark-adapted retinas. At a higher dose (2.0 nmol), both antagonists increased T H activity in the darkadapted retina and attenuated the further activation of the enzyme by light. Rat retinas were dissociated into suspensions of viable cells in order to investigate the direct effects of muscimol and picrotoxin on the DA/TH-containing cells. The process of dissociating dark-adapted retinas resulted in an apparent activation of TH. * To whom correspondence should be addressed.
336 Incubation of the cells with muscimol resulted in a decrease of TH activity in a concentration-dependent manner. Picrotoxin antagonized the inhibitory effect of muscimol, but had no effect when incubated alone. The biphasic effects of GABA agonists and antagonists in vivo suggest that a certain subpopulation of GABA neurons are involved in the activation of the DA/THcontaining neurons by photic stimulation, while another subpopulation of GABA neurons produce a tonic inhibition of the DA/TH-containing neurons in darkness. The experiments with retinal cell suspensions indicate that the tonic inhibition is probably mediated by synapses of GABA neurons directly onto the DA/THcontaining cells. INTRODUCTION Dopamine (DA) is the putative neurotransmitter of a population of amacrine neurons in the retina of several mammalian species11,la,15,26, and appears to be the only catecholamine in retinal neurons of the rat 16,17. Light exposure increases the rate of DA biosynthesis9,16,29 and release 19 from retina, indicating that these neurons are activated by photic stimulation, and are presumably involved in the neuronal processing of visual information. Tyrosine hydroxylase (TH; EC 1.14.16.2), the rate-limiting enzyme of catecholamine biosynthesisza, is activated concomitantly with the light-evoked increase of DA biosynthesis16,17. Retinal TH activity increases rapidly during photic stimulation (within 1 min), and is maximal following 5 min of light exposure. TH activity decreases following termination of photic stimulation. Thus, changes of TH activity appear to be a biochemical indicator of changes in the neuronal activity of DA/TH-containing amacrine cells. The DA/TH-containing amacrine cells apparently receive their synaptic input exclusively from other amacrine cells10. Gamma-aminobutyric acid (GABA) is the putative neurotransmitter of another population of amacrine neurons 2-4,u,~a,ls. Therefore, we examined the effects of GABA agonists and antagonists on the regulation of retinal TH activity to determine if the GABA neurons modulate the neuronal activity of the DA/TH-containing amacrine cells. MATERIALS AND METHODS Animals. Male Sprague-Dawley-derived rats (Zivic-Miller, Allison Park, Pa.),
150-250 g, were housed in clear plastic cages with perforated metal tops under a 12 h light-dark cycle,'and were given free access to food and water. Prior to drug administration, rats were dark-adapted for 3-14 h in a room with a double-door entrance that permitted access to the room without light-leakage. All manipulations of the darkadapted rats were performed under dim red light (Kodak Wratten No. 92); this condition is designated in the text as 'dark'. Materials. Picrotoxin, GABA, reduced nicotinamide adenine dinucleotide phosphate, catalase, bovine serum albumin, deoxyribonuclease I (DN-EP), coUagenase
337 type IA, hyaluronidase type Ili, and papain type llI were purchased from Sigma Chemicals, St. Louis, Mo. Pyridoxal phosphate and OL-6-methyl-5,6,7,8,-tetrahydropterine (6-MPH4) were obtained from Calbiochem, LaJolla, Calif. Bicuculline methiodide was obtained from Pierce Chemicals, Rockford, I11. L-[1-14C]tyrosine, 57 mCi/mmol, was purchased from Amersham, Arlington Heights, I11. Muscimol was obtained from Smith Kline and French Laboratories, Philadelphia, Pa. All other chemicals were obtained in the purest form available. Drug administration. Drugs were dissolved on 0.9 ~ NaC1 immediately before use, and were administered by subcutaneous (s.c.) or intravitreal injections. Subcutaneous injections were delivered via a 25-gauge syringe needle to the subcutaneous space approximately 15 mm posterior to the animal's head on the dorsal surface. For intravitreal injections, rats were anesthetized with ether. Injections were delivered with a Hamilton beveled-syringe needle (26S-gauge) that had a 20-gauge cannula covering all but the last 7 mm at the beveled end. The needle was inserted into the vitreous from the nasal quarter of the eye, approximately 1 mm below the border of the cornea and sclera. The needle was visually located in center of the vitreous and was positioned so that the bevel was directed towards the lens in order to minimize damage to the retina. Five #1 of solution were injected over a period of approximately 10 sec. All injections were administered in the dark. In the intravitreal-drug experiments, one eye of each rat was injected with drug solution and the other eye with saline, to serve as a paired control. Photic stimulation. Rats were exposed to light, following recovery from anesthesia, for the periods indicated in the table and figure legends. Photic stimulation was provided by fluorescent tubes (35 W Sylvania Super-Saver) suspended 1.5 m above the cages. Light-intensity inside the cage was estimated to be approximately 400 lux by measuring the reflected light from a neutral gray card with a photometer. Retinal dissociation. The dissociation of dark-adapted retinas into suspensions of viable single cells was performed by a modification of the technique of Sarthy and Lam3°. Eyes were hemisected and the posterior chambers incubated at 37 °C for 15 min in mammalian Ringer's solution containing hyaluronidase (100 U/ml) and coUagenase (180 U/ml). The eye cups were washed in fresh Ringer's solution and the retinas removed from the eye cups. The retinas were incubated for 30 min at 37 °C in Ca2+-free Ringer's solution containing papain (50/~g/ml). The retinas were washed in normal Ringer's solution and transferred to conical centrifuge tubes containing Ringer's solution (0.125 ml/retina), deoxyribonuclease I (320 U/ml), and 0.1 ~o bovine serum albumin (fraction V). The retinas were dissociated by gently pipetting the tissue through a pasteur pipette. The undissociated pieces of retina settled for 5 min, and the top half of the supernatant containing the dissociated ceils was removed and transferred to a chilled tube on ice. The remaining pieces of retina and supernatant were gently pipetted up and down several times, allowed to settle, and the top half of the supernatant removed and pooled with the supernatant from the previous step. All Ringer's solutions contained 10 mM D-glucose and were gassed with 95 ~ O2-5~ CO2. This procedure gave suspensions in which greater than 95 ~ of the cells were present as single cells.
338 TABLE I
Effect of GABA on retinal tyrosine hydroxylase activity Dark-adapted rats were injected intravitreally with 0.9 ~ NaC1 and GABA (5 mnol) as described in Methods. Rats were sacrificed t5 min after injection, either in the dark, or after 5-9 min (~ -- 7 rain) of light exposure. TH activity is expressed as pmol CO2/min/mg protein; values represent the mean :t: S.E.M. The number in parentheses represent the n.
Injection
Saline GABA
Tyrosine hydroxylase activity Dark
Light
1.4:5 0.1 (5) 1.5 4- 0.1 (5)
4.1 4- 0.3 (4) 3.4 4- 0.2 (4)*
* P < 0.05, 2-tailed, paired Student's t-test, compared to matched saline control.
Aliquots of the retinal cell suspension were transferred into ice-chilled roundbottom centrifuge tubes containing freshly prepared control or test Ringer's solutions, and gently agitated for 30 min at 37 °C in a Dubnoff Metabolic Shaker under an atmosphere of 95 ~ 02-5 ~ CO2. Following the drug incubation period, the tubes were chilled on ice, and centrifuged for 30 min at 250 × g at 0-4 °C. The supernatants were discarded and the pellets were frozen on dry-ice for tyrosine hydroxylase assay. Dissociation and incubation of retinal suspensions were performed under dim red light (Kodak Wratten No. 92). Tyrosine hydroxylase assay. Retinas were dissected and frozen on dry-ice as described previously 17. Frozen retinas or pelletted retinal cells were homogenized in ice-cold 50 m M Tris acetate, p H 6.0, containing 0.2~/o Triton X-100 in glass-teflon homogenizers. Homogenates were centrifuged at 32,000 x g for 15 min at 0-4 °C, and tyrosine hydroxylase was measured in the supernatants by a modification of the D O P A decarboxylase-coupled assay of Waymire et al. 31, as described previously17; [14C-carboxyl]tyrosine was used as substrate, and [14C]CO2 was the product. 6-MPH4 was dissolved at a concentration of 10 mM in 0.01 N HCI and reduced under hydrogen gas in the presence of platinum (IV) oxide. The activation of retinal T H by brief-light expOsure is characterized by a decrease in the apparent Km of the enzyme for the pteridine cofactor17; therefore, in order to demonstrate the light-evoked activation of TH, a subsaturating concentration of 6-MPH4 (0.2 mM) was used. The concentration of L-tyrosine was 0.1 mM and the p H was 7.0. Protein was determined by the method of Lowry et al. 24. RESULTS
Effbct of GABA agonists and antagonists in vivo Retinal T H activity was relatively low in dark-adapted rats and was markedly stinmlated by light exposure (Table I). When administered intravitreally to darkadapted rats, 5 nmoles of G A B A significantly reduced the light-evoked activation of retinal TH, even though it did not alter the activity of the enzyme in the dark (Table I).
339
150-
g, 125Z 0
Z ¢
10075-
5025* p < 0.01
MUSCIMOL (log pmole)
Fig. 1. Effect of muscimol on light-stimulated tyrosine hydroxylase activity. Dark-adapted rats were injected intravitreally with 0.9~ NaC1 and muscimol as described in Methods. Subjects were sacrificed and retinas dissected 40 min after injection, following 16-24 min (~ = 20 min) of photic stimulatiOn. Data are expressed as the mean and S.E.M. of the percentage of TH activity in the muscimol-tr~ated retina relative to the saline-treated retina of each animal. The sample size for each dose was 4-9. The TH activity of all of the control retinas was 5.3 ± 0.25 pmol CO2/min/mgprotein (~ ± S.E.M.).t*P ~ 0.01, 2-tailed, paired Student's t-test, compared to matched control; A dose-response curve for muscimol, a GABA-receptor agonist14,21, 22, was determined to more fully characterize the effects of GABA receptor stimulation on the regulation of retinal T H activity. When administered by intravitreal injection prior to photic stimulation, muscimol had a dose-dependent, biphas!c effect on the lightevoked activation of retinal T H (Fig. 1). Low doses of muscimol (35 and 60 pmol) significantly increased the activity of T H measured following photic stimulation. In contrast, higher doses of the GABA-receptor agonist (_> 120 pmol) significantly inhibited the activation of the enzyme by light. When compared to dark-adapted controls, 300 pmoles of muscimol completely inhibited the light-evoked stimulation of TH. Muscimol had no effect on the activity of retinal T H in the dark-adapted retinas, at the doses tested (60 and 900 pmol). To determine if the inhibitory effect of muscimol was mediated by an interaction with GABA receptors, we examined the effects of systemically administered picrotoxin, a GABA antagonistS, 20, on the inhibition of retinal T H by 400 pmoles of muscimol, administered intravitreally (Fig. 2). Picrotoxin (4 #mol/kg) significantly antagonized the inhibitory effect of muscimol on T H activity. In this experiment, and in attempts to antagonize the stimulatory effect of low doses of muscimol, we observed that systemically administered picrotoxin inhibited the activation of T H by light. Further investigation demonstrated that picrotoxin suppressed the light-evoked stimulation of T H activity in a dose-dependent manner (Fig. 3). To avoid the convulsions produced by systemic administration of picrotoxin, the effects of the GABA antagonist on the activity of T H in the dark and the light were examined following intravitreal administration o f picrotoxin to dark-adapted rats (Table II). Picrotoxin, at a dose of 0.5 nmoles, significantly decreased the activity of T H in the light-exposed retinas, but had no significant effect in the dark. A higher dose of
340
I
3
d~
:¢
O O
2
-6
1
CONTROL
MUSCIMOL
MUSCIMOL & PICROTOXlN
Fig. 2. Antagonism by picrotoxin of the inhibitory effect of muscimol on light-stimulated tyrosine hydroxylase activity• Dark-adapted rats were injected subcutaneously with 0.9 ~ NaCI or picrotoxin (4/~mol/kg). Ten rain after the subcutaneous injection, the subjects were injected intravitreally with 0.9 % NaCI or with muscimol (0.4 nmol). Subjects were sacrificed and retinas dissected 25 min after the intravitreal injections, following 12-18 rain (£ = 15) of light exposure• Data are expressed as mean S.E.M., N = 6. *P < 0.01, Dunnett's test.
picrotoxin, 2.0 nmoles, significantly activated TH in the dark. The activity of the enzyme in the light-exposed retinas was not significantly different from control after the higher dose of picrotoxin; however, the stimulation of enzyme activity by light was reduced in the drug-treated retinas (saline ---- 92 %, picrotoxin = 29 %).
I
c
E
~.
3"
E O O e o
J_ 2-
E
P1
P4
P8
,umole/kg (s.c.)
Fig. 3. Antagonism of light-evoked activation of tyrosine hydroxylase by picrotoxin. Rats were injected subcutaneously with 0.9 % NaCI or picrotoxin (1, 4, or 8 #mol/kg). Subjects were sacrificed and retinas dissected 35 min after injection, following 11-19 min (£ = 15) of photic stimulation. Data are expressed as mean 4- S.E.M., n -- 8-9. *P < 0.05, Dunnett's test.
341 TABLE II
Effect of intravitreal picrotoxin on retinal tyrosine hydroxylase activity Dark-adapted rats were injected intravitreally with 0.9 % NaCI and picrotoxin as described in Methods. Rats were sacrificed 22 min after injection, either in the dark or after 8-14 rain (~ = 11 min) of light exposure. TH activity is expressed as pmol CO2/min/mg protein; values represent the mean and S.E.M. The number in parentheses represents the n.
Injection
Dose (nmol)
Tyrosine hydroxylase activity Dark
Light
Experiment 1 Saline Picrotoxin
-0.5
1.9 4- 0.35 (7) 2.6 4- 0.25 (7)
6.1 ± 0.59 (6) 5.2 4- 0.40 (6)*
Experiment 2 Saline Picrotoxin
-2.0
2.5 -4- 0.09 (5) 4.1 4- 0.17 (5)*
4.8 4- 0.40 (6) 5.3 4- 0.42 (6)
* P < 0.05, 2-tailed, paired Student's t-test, compared to matched saline control.
Bicuculline, another GABA
antagonist6,7, s, h a d a n effect s i m i l a r to t h a t o f
p i c r o t o x i n a f t e r i n t r a v i t r e a l a d m i n i s t r a t i o n ( T a b l e III). Bicuculline, at a d o s e o f 0.4 n m o l e s , significantly r e d u c e d the a c t i v i t y o f T H in the l i g h t - e x p o s e d retinas, a n d h a d n o s i g n i f i c a n t effect in t h e d a r k . A t a d o s e o f 2.0 n m o l e s , b i c u c u l l i n e significantly i n c r e a s e d the e n z y m e a c t i v i t y in t h e d a r k , a n d a t t e n u a t e d the f u r t h e r a c t i v a t i o n o f the e n z y m e b y light.
TABLE III
Effect of bicuculline on retinal tyrosine hydroxylase activity Dark-adapted rats were injected intravitreally with 0.9 % NaCI and bicuculline methiodide as described in Methods. Rats were sacrificed 40 min after injection, either in the dark or after 12-18 rain (~ = 15 min) of light exposure. TH activity is expressed as pmol C02/min/mg protein; values represent the mean 4- S.E.M. The number in parentheses represents the n.
Injection
Dose (nmol)
Tyrosine hydroxylase activity Dark
Light
Experiment I Saline Bicuculline
-0.4
2.1 4- 0.2 (9) 2.3 4- 0.3 (9)
8.3 4- 0.6 (11) 7.4 4- 0.4 (11)*
Experiment 2 Saline Bicuculline
-2.0
2.9 4- 0.5 (8) 5.3 4- 0.5 (8)*
7.9 -4- 0.5 (8) 8.2 4- 0.5 (8)
* P < 0.05, 2-tailed, paired Student's t-test, compared to matched saline control.
342 TABLE IV
Comparison of tyrosine hydroxylase activity extracted from retinal cell suspensions and intact retina Dark-adapted retina were dissected, frozen and homogenized as described in Methods. Other darkadapted retinas were dissociated into suspensions of cells, incubated at 37 °C under an atmosphere of 95 % 03/5 % COs for 30 min, centrifuged, frozen and homogenized as described in Methods. Data are expressed as ~ 4- S.E.M.
Intact retina Retinal cell suspensions
n
Tyrosine hydroxylase activity (pmol COz/min/mg protein)
16 14
2.3 :t: 0.1 3.9 d- 0.4*
* P < 0.01, 2-tailed Student's t-test.
Effect of muscimol and picrotoxin in retinal cell suspensions Retinas were dissociated into suspensions of cells to investigate the direct effects of muscimol and picrotoxin on the DA/TH-containing cells. Examination of the cell suspensions with inverted-stage phase microscopy indicated that greater than 95 ~ of the cells in the suspension were present as single cells. After the cell suspensions were centrifuged at 250 x g, substantial tyrosine hydroxylase activity was found in homogenates of the pelleted cells. However, tyrosine hydroxylase activity was not found in the supernatant from the 250 x g centrifugation, or in the resuspended pellet obtained from recentrifugation of the 250 × g supernatant at 35,000 x g for 20 min. These results indicate that the majority of enzyme remained in intact cells, in somas, or in large processes, rather than being released into the media by lysis of cells or being localized to free synaptosomes. q¢
f
3O z o
"r" z
20
lO
6
lO'SM
IO';'M
10-61~
MUSCIMOL
Fig. 4. Effect of muscimol on tyrosine hydroxylase activity of retinal cell suspensions. Retinas were dissociated into suspensions of single cells as described in Methods. For each suspension, triplicate aliquots were incubated in Ringer's solution for control, and triplicate aliquots were incubated in Ringer's solution containing a given concentration of muscimol. The per cent inhibition of TH activity induced by a given concentration of muscimol was obtained by comparing the effect of muscmiol/Ringer's solution and control/Ringer's solution from the same cell preparation. The data represent the mean 4- S.E.M. for 4-6 experiments. The TH activity of all of the controls was 3.9 4- 0.4 pmol C02/min/mg protein (~ 4- S.E.M.). Muscimol, at doses of 10-7 M and 10-6 M produced significant decreases in T H activity. *P < 0.05, 2-tailed, paired Student's t-test compared to matched control.
343 TABLE V Picrotoxin-blockade of the effect of muscimol on tyrosine hydroxylase activity of dissociated retinal suspensions
Retinas weredissociatedinto suspensions of singlecellsas describedin Methods. From each suspension, triplicate aliquots were incubated in Ringer's solution or Ringer's solution containing muscimol and picrotoxin at 37 °C under an atmosphere of 95 % 02/5 % CO2.After a 30 rain incubation, the suspensions were centrifuged, and the precipitates were frozen, homogenized and assayed for TH activity as described in Methods. Data are expressed as ~ 4- S.E.M. of 4 separate experiments. Condition
n
Control 4 Muscimol (10-~ M) -~- picrotoxin(2.4 × 10-SM) 4
Tyrosine hydroxylase activity (pmol C02/min/mg protein)
3.5 4- 0.2 3.4 4- 0.2
Muscimol's inhibitory effect on T H activity is probably mediated through GABA receptors that are located on the DA/TH-containing neurons, for the following reasons. The T H activity measured in homogenates of cell suspensions of darkadapted retinas prepared under dim red light was significantly higher than that measured in homogenates of intact dark-adapted retinas (Table IV). When cell suspensions were incubated in the presence of muscimol, T H activity was decreased in a concentration-dependent manner (Fig. 4). In contrast to the effect of muscimol in vivo, only inhibition, and not facilitation, of enzyme activity was observed over a 100fold concentration range. The decrease in T H activity was not caused by a direct inhibition of the enzyme, because muscimol (10 -6 M) added directly to the supernatant of the homogenized cells had no effect on enzyme activity. Furthermore, the inhibitory effect of incubating the cells with muscimol was completely antagonized by picrotoxin (Table V), indicating that the inhibitory effect of muscimol on T H activity was mediated through GABA receptors. Incubation of the cells with picrotoxin alone had no effect on T H activity. DISCUSSION Dopamine (DA) is localized within a population of amacrine cells in the mammalian retina11,lZ,15, 26. Light exposure enhances both DA biosynthesisg, 16,29 and release 19, and also activates retinal tyrosine hydroxylase16,17, indicating that these neurons are activated by photic stimulation. Anatomical evidence suggests that the DA/TH-containing neurons receive their synaptic input exclusively from other amacrine cells 10. Here we have presented pharmacological evidence that GABA, a putative neurotransmitter in another population of amacrine cells 2-4,11,13,18, is involved in the regulation of the DA/TH-containing neurons. GABA and muscimol, a GABA receptor agonist 14,21,z2, altered the light-evoked stimulation of TH, while having no influence on the enzyme activity in the retinas of dark-adapted rats. The light-evoked stimulation of T H was enhanced by low doses of muscimol (20--60 pmol), but inhibited at higher doses (120-900 pmol). The inhibitory
344 effect of high doses of muscimol was apparently mediated by an interaction with GABA receptors, since the GABA antagonist, picrotoxin, blocks this effect of muscimol. Intravitreal administration of two GABA antagonists, picrotoxin and bicuculline, increased TH activity in dark-adapted retinas and attenuated the further activation of the enzyme by light. These biphasic effects of GABA receptor agonists and antagonists suggest two distinct roles for GABA in the regulation of retinal DA/TH-containing neurons. First, these neurons appear to be tonically inhibited in the dark. The tonic inhibition is at least partially mediated by GABA, because intravitreal administration of GABA antagonists elicited an increase in TH activity in the dark-adapted retina, presumably by blocking the effect of endogenously released GABA. Furthermore, both GABA and muscimol block the light-evoked activation of TH when administered prior to photic stimulation. The observation that muscimol has no effect on TH activity in the dark-adapted retinas suggests that the GABA-mediated inhibitory influence on the DA/TH-containing neurons is probably maximal in the dark. A second role of GABA is in the activation of retinal DA/TH-containing neurons as a consequence of light exposure, for the following two reasons. First, picrotoxin and bicuculline, GABA antagonists, blocked the light-evoked stimulation of TH. Secondly, intravitreal administration of low doses of muscimol (20-60 pmol), an agonist, produced an enhancement of the light-evoked enzyme activity. Because GABA receptor agonists produce direct inhibitory effects upon central neuronal activity6,7, 22, the enhancement of TH activity is probably mediated by an action of muscimol on neurons other than the DA/TH-containing cells; the locus or mechanism for this action of muscimol to increase TH activity cannot be determined from our data. Higher doses of intravitreally administered muscimol, which inhibited light-evoked stimulation of TH activity, presumably act at the site where endogenously released GABA exerts its inhibitory influence on DA/TH-containing amacrine cells in the dark. Support for this hypothesis is the observation that the higher doses of muscimol decreased TH activity in light-stimulated retinas to the level of that measured in the retinas of dark-adapted rats. A working hypothesis, based on the data outlined below, is that GABAcontaining neurons synapse directly on the DA/TH-containing neurons and inhibit the neuronal activity of these cells. The dissociation of retinas into suspensions of cells appears to disrupt most connections between cells (more than 95 ~o of the cells were present as single cells). The DA/TH-containing neurons were released from a tonic inhibitory influence in the process of dissociation, as indicated by the observation that the TH activity of cell suspensions was greater than that found in intact retinas when both preparations had been obtained from dark-adapted subjects. This hypothesis is supported by the observation that TH activity was not affected when the cells were incubated with picrotoxin. Muscimol decreased the TH activity of the cell suspensions in a concentration-dependent fashion; this suppression of TH activity was blocked by picrotoxin. Stimulation of enzyme activity by muscimol was not observed in the cell suspensions. The TH activity of the cell suspensions appears to be regulated by the membrane potential of the cells, as indicated by the observation that K+-induced
345 depolarization of the cells results in an activation of the enzyme (Iuvone and Marshburn, unpublished observation). Thus, the inhibitory action of muscimol on TH activity is probably mediated by an interaction with GABA receptors located on the DA/TH-containing neurons, and is presumably a consequence of hyperpolarization of these cells. Our data are in agreement with the data of Morgan and Camp 29, who found that muscimol inhibited the light-evoked increase in DA turnover in the rat retina, as measured by the rate of decline of the concentration of retinal DA following inhibition of TH by a-methyl-p-tyrosine. However, these investigators found that multiple doses of muscimol, given intravenously, decreased DA turnover below the level of that found in dark-adapted controls. In contrast, we found that muscimol, administered intravitreally, decreased TH activity to the level of that in dark-adapted retinas, but not below that level. One possible explanation for this difference may be related to the different measurements employed in the two studies; perhaps DA turnover can be furthel decreased by inhibition of release or metabolism when TH is in an unactivated state. Another possibility is that the effect observed previously 29 is caused by a metabolite of muscimol, because multiple doses of muscimol were administered intravenously, and because muscimol is rapidly metabolized following intravenous injection 25. The biphasic effects of intravitreally administered GABA agonists and antagonists were dose-dependent. Low doses of agonist and antagonists appeared to mimic and antagonize, respectively, the stimulatory effect of GABA on the activation of the DA/TH-containing neurons by light. High doses of agonist and antagonists appeared to mimic and antagonize, respectively, the inhibitory effect of GABA on the DA/THcontaining neurons. These data suggest that the stimulatory influence of GABA may be mediated via high-affinity receptors, while the inhibitory influence is mediated via low-affinity ~eceptors. In support of this explanation, high- and low-affinity [3H]GABA binding-sites ('receptors') have been observed in the mammalian retina t2. Alternatively, the dose-related effects of GABA agonists and antagonists may be related to different intraretinal locations of the receptors that mediate the presumed stimulatory and inhibitory effects of GABA. Our observation of the biphasic effects of GABA agonists and antagonists on the activity of TH, as well as the pharmacological, biochemical, and electrophysiological studies of other investigators, provide the basis for the following hypothetical model. The mammalian retina appears to contain at least two distinct populations of GABA neurons, one that is active in the dark and one that is active in the light. GABA antagonists increase the spontaneous activity of some retinal ganglion cells 1, increase acetylcholine release27, zs, and increase TH activity in the dark-adapted retina, providing pharmacological evidence for the synaptic release of GABA in the dark. The release of [ZH]GABA from preloaded retinas is enhanced by photic stimulation 2, and GABA antagonists decrease the light-evoked activation of tyrosine hydroxylase, providing both biochemical and pharmacological evidence for synaptic release of GABA in response to photic stimulation. We propose that some of the GABA neurons that are active in the dark synapse directly onto the DA/TH-containing
346 neurons and tonically inhibit the neuronal activity of these cells in darkness. During photic stimulation, G A B A release from these neurons is inhibited, thereby removing the tonic inhibitory influence on the D A / T H - c o n t a i n i n g cells. Photic stimulation m a y also activate the release o f G A B A from other G A B A neurons, and in some way, perhaps by inhibiting an interneuron that inhibits the D A / T H - c o n t a i n i n g cells, augments the activation of the ceils by light. ACKNOWLEDGEMENTS We wish to thank Dr. N o r t o n H. Neff, N I M H , for helpful discussions and for assistance on pilot studies related to this project. We thank Dr. Henry Haigler, E m o r y University, for assistance in the preparation of this manuscript. We thank L. K. Holmes for technical assistance, and Dr. Peter Ridley, Smith Kline and French Laboratories, for the gift o f muscimol. P.B.M. was supported by N I H Training G r a n t 5-TO-1-GM-00179. P.M.I. was partially supported by grants from the Pharmaceutical Manufacturers Association F o u n d a t i o n and the N I H , 5S07-RR-05364.
REFERENCES 1 Ames, A., III and Pollen, D. A., Neurotransmission in central nervous tissue: a study of isolated rabbit retina, J. Neurophysiol., 32 0969) 424-442. 2 Bauer, B. and Ehinger, B., Light evoked release of radioactivity from rabbit retinas preloaded with [aH]GABA, Experimentia (Basel), 33 (1977) 470-471. 3 Brandon, C., Lam, D. M. K. and Wu, J-Y., The y-aminobutyric acid system in rabbit retina: Localization by immunocytochemistry and autoradiography, Proc. nat. Acad. Sci. (Wash), 76 (1979) 3557-3561. 4 Brunn, A. and Ehinger, B., Uptake of certain possible neurotransmitters into retinal neurones of some mammals, Exp. Eye Res., 19 (1974) 435-447. 5 Curtis, D. R. and Johnston, G. A. R., Amino acid transmitters. In A. Lajtha (Ed.), Handbook of Neurochemistry, Vol. 4, Plenum Press, New York, 1970, pp. 115-135. 6 Curtis, D. R., Duggan, A. W., Felix, D. and Johnston, G. A. R., GABA, bicuculline and central inhibition, Nature, (Lond.), 226 (1970) 1222-1224. 7 Curtis, D. R., Duggan, A. W., Felix, D. and Johnston, G. A. R., Bicucullin, an antagonist of GABA and synaptic inhibition in the spinal cord of the cat, Brain Research, 32 (1971) 69-96. 8 Curtis, D. R., Duggan, A. W., Felix, D., Johnston, G. A. R. and McLennan, H., Antagonism between bicuculline and GABA in the cat brain, Brain Research, 33 (1971) 57-73. 9 Da Prada, M., Dopamine content and synthesis in retina and n. accumbens septi: pharmacological and light-induced modifications, Advance Biochem. Psychopharmacol., 16 (1977) 311-319. 10 Dowling, J. E., and Ehinger, B., Synaptic organization of the dopaminergic neurons in the rabbit retina, J. comp. Neurol., 180 0978) 203-220. 11 Ehinger, B. and Falck, B., Autoradiography of some suspected neurotransmitter substances: GABA, glycine, glutamic acid, histamine, dopamine, and L-DOPA, Brain Research, 33 (1971) 157-172. 12 Enna, S. J. and Snyder, S. H., Gamma-aminobutyric acid (GABA) receptor binding in mammalian retina, Brain Research, 115 (1976) 174-179. 13 Graham, L. T., Comparative aspects of neurotransmitters in the retina. In H. Dawson, and L. T. Graham (Eds.) The Eye, Comparative Physiology, Vol. 6, Academic Press, New York, 1974, pp. 283-342. 14 Greenlee, D. V., Van Ness, P. C. and Olsen, R. W., Gamma-aminobutyric acid binding in mammalian brain: receptorqike specificity of sodium-independent sites, J. Neurochem., 31 (1978) 933-938.
347 15 Haggendal, J. and Malmfors, T., Identification and cellular localization of the catecholamines in the retina and choroid plexus of the rabbit, Acta physioL scand., 64 (1965) 58-66. 16 Iuvone, P. M., Galli, C. L., Garrison-Gund, C. K. and Neff, N. H., Light stimulates tyrosine hydroxylase activity and dopamine synthesis in retinal amacrine neurons, Science, 202 (1978) 901-902. 17 Iuvone, P. M., Galli, C. L. and Neff, N. H., Retinal tyrosine hydroxylase: comparison of shortterm and long-term stimulation by light, Molec. PharmacoL, 14 (1978) 1212-1219. 18 Kojima, K., Mizuno, K. and Miyazaki, M., Gamma-amino-butyricacid in ocular tissue, Nature (Lond.), 181 (1958) 1200-1201. 19 Kramer, S. G., Dopamine: A retinal neurotransmitter. I. Retinal uptake, storage and light stimulated release of H3-dopamine in vivo, Invest. OpthalmoL, 10 (1971) 438-452. 20 Krnjevfc, K., Chemical nature of synaptic transmission in vertebrates, PhysioL Rev., 54 (1974) 418-540. 21 Krogsgaard-Larsen, P. and Johnston, G. A. R., Structure-activity studies on the inhibition of GABA binding to rat brain membranes by muscimol and related compounds, J. Neurochem., 30 (1978) 1377-1382. 22 Krogsgaard-Larsen, P., Johnston, G.A.R., Curtis, D. R., Game, C. J. A. and McCulloch, R. M., Structure and biologic activity of a series of conformationally restricted analogues of GABA, J. Neurochem., 25 (1975) 803-809. 23 Levitt, M., Spector, S., Sjoerdsma, A. and Udenfriend, S., Elucidation of the rate-limiting step in noradrenaline biosynthesis in the perfused guinea-pig heart, J. PharmacoL exp. Ther., 148 (1965) 1-8. 24 Lowry, O. H., Rosenbrough, N. J., Farr, A. L., and Randall, R. J., Protein measurement with the Folin phenol reagent, J. bioL Chem., 193 (1951) 265-275. 25 Maggi, A. and Enna, S. J., Characteristics of muscimol accumulation in mouse brain after systemic administration, Neuropharmacology, 18 (1979) 361-366. 26 Malmfors, T., Evidence of adrenergic neurons with synaptic terminals in the retina of rats demonstrated with fluorescence and electron microscopy, Actaphysiol. scand., 58 (1963) 99-100. 27 Massey, S. C. and Neal, N. J., The light evoked release of acetylcholine from rabbit retina in vivo and its inhibition by 7-aminobutyric acid, J. Neurochem., 32 (1979) 1327-1329. 28 Massey, S. C. and Redburn, D. A., GABA inhibition of light evoked ACh release, Invest. Ophthalmol., Suppl. ARVO, (1980) 283. 29 Morgan, W. W. and Kamp, C. W., A GABAergic influence on the light-induced increase in dopamine turnover in the dark-adapted rat retina in vivo, J. Neurochem., 34 (1980) 1082-1086. 30 Sarthy, P. V. and Lam, D. M. K., Isolated cells from a mammalian retina, Brain Research, 176 (1979) 208-212. 31 Waymire, J. C., Bjur, R. and Weiner, N., Assay of tyrosine hydroxylase by coupled decarboxylation of Dopa formed from (1-14C)-L-tyrosine,Analyt. Biochem., 43 (1971) 588-600.