Phosphorylation reactions in bovine rod outer segments studied by 32P-labelling of intact retina

Biochimica et Biophysica Acta 881 (1986) 185-195

185

Elsevier BBA 22287

Phosphorylation reactions in bovine rod outer segments studied by 32P-labelling of intact retina K.M. Paul Kamps *, Edward A. Dratz **, Frans J.M. Daemen and Willem J. de Grip *** Department of Biochemistry, Universityof Nijmegen, 6500 HB Nijmegen (The Netherlands) (Received December 6th, 1985)

Key words: Protein phosphorylation; Visual transduction; Ca2+; Rod outer segment; Plasma membrane protein; (Bovine retina)

The protein phosphorylation pattern in the intact bovine retina has been investigated by labelling with 32P-phosphate under incubation conditions that preserve the electrical photoresponse of the photoreceptor cells. The phosphorylation of rod outer segment proteins was analysed after isolation of outer segments from the labelled retina. The global influence of light, Ca2+ and the phosphodiesterase inhibitor, isobutylmethylxanthine, on protein phosphorylation in rod outer segments was analysed. A 12 kDa protein is the most prominent phosphorylated species in the intact bovine retina. Its phosphorylation is increased by light and/or Ca2+. Evidence is presented that this strongly phosphorylated protein is not located in the outer segment, and we suggest that it may be a synaptic protein. Retinal rod outer segment membrane proteins with apparent molecular weights of 245, 226, 125, 110, 50, 46, 38 and 20 all show light-stimulated phosphorylation. Lowering the extracellular Ca2÷ levels results in a decrease of the phosphorylation level of some of these proteins, viz. at 125, 50, 38 and probably at 20 kDa. Such proteins, whose phosphorylation level is influenced both by light and by elevated Ca2÷, are candidates for mediators of phototransduction. The phosphorylated species at 245, 226, 110, 50 and 20 kDa are enriched in rod outer segment plasma membrane preparations. These protein species could participate in the light-regulated modulation of the Na÷-conductance of the plasma membrane.

Introduction Visual excitation of the rod photoreceptor cell in the vertebrate retina is mediated by the visual pigment rhodopsin. Absorption of a photon by a rhodopsin molecule, located in the membrane of one of the many disks present in the rod outer * Present address: Department of Chemical Technology, Technical University Twente, 7500 AE Enschede, The Netherlands. ** On leave from: Department of Biochemistry, Division of Natural Sciences, University of California, Santa Cruz, CA 96064, U.S.A. *** To whom correspondence should be addressed.

segment, generates a signal that traverses the rod cytoplasm and causes a transient reduction in sodium permeability of the plasma membrane of the rod outer segment. The resulting hyperpolarization of the plasma membrane spreads to the synapse at the other end of the rod cell, where it modulates neurotransmitter release. The details of the processes by which photoexcited rhodopsin and the sodium-conductance mechanism of the rod outer segment plasma membrane communicate are still a matter of conjecture [1,2]. There is considerable evidence that altered phosphorylation of membrane proteins may be an important biochemical mechanism for regulating

0304-4165/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)

186 membrane-related events in nervous tissue [3,4]. In addition, light-sensitive phosphorylation of several proteins in the rod outer segment has been observed. Photoactivated rhodopsin catalyzes the exchange of GTP for GDP on a GTP-binding protein [5,6] and undergoes multiple phosphorylation by an opsin kinase [7]. The GTP-binding complex activates a cyclic GMP-specific phosphodiesterase, resulting in a rapid hydrolysis of cyclic GMP. The local drop in cyclic GMP may affect protein kinase activity in the outer segment cytosol, since phosphorylation of several proteins in this cytosol is cyclic nucleotide-dependent [8-12]. In order to determine the physiological relevance of the observed regulation of phosphorylation reactions, it is crucial to determine whether the phosphorylation reactions observed in isolated rod outer segments do occur in the intact functional retina. Hence, we have studied the phosphorylation pattern of rod outer segments in intact bovine retinas which have been incubated in vitro with [32P]phosphate. The results are compared to those obtained with rod outer segment membranes incubated with [~,-32p]ATP. We report some effects of light, Ca 2+ levels and isobutylmethylxanthine on 32p-incorporation, since these parameters have been extensively studied in electrophysiological experiments. The electrophysiological integrity of the bovine retina has been checked in parallel experiments by recording electroretinograms under similar conditions in a Sickel chamber. The results support the use of isolated rod outer segments as a suitable system to investigate the effects of light and C a 2+ o n the regulation of the phosphorylation processes. The possible relevance of light-dependent phosphorylation of carriers or conductor proteins is discussed in relation to the findings that the lightsensitive ion-conducting mechanism of the rod plasma membrane is modulated by cyclic GMP, but not by Ca2+, ATP or GTP [13,14]. Furthermore, special attention is given to the location and the phosphorylation pattern of a 12 kDa protein in bovine retina, in view of the role attributed to similar proteins in rod excitation in the frog retina [2].

Materials and Methods Materials

Carrier free [32p]H3PO4, in dilute HC1 (pH 2-3), was obtained from Amersham International (U.K.). Isobutylmethylxanthine was from Aldrich Chemical Company Inc. (St. Louis, MO; Milwaukee, WI, U.S.A.). Sodium penicillin G and streptomycin were from Gist Brocades N.V., Delft (The Netherlands). Cyclic AMP and cyclic GMP were obtained from Sigma Chemical Co. (St. Louis, MO, U.S.A.); [y-32p]ATP (3000 Ci/nmol) was obtained as a solution in 50% ethanol from New England Nuclear (Boston, MA, U.S.A.); sodium dodecyl sulfate from Pierce Biochemical Company (Rockford, IL, U.S.A.); acrylamide (electrophoresis grade) from Bio-Rad (Richmond, CA, U.S.A.); N,N'-methylene bisacrylamide from Eastman Kodak (Rochester, NY, U.S.A.). Analytical-grade chemicals of the highest purity available were used throughout. The following molecular weight standards were used for polyacrylamide gel electrophoresis: flgalactosidase, 130 kDa (Sigma); phosphorylase A, 93 kDa (Sigma); bovine serum albumin, 68 kDa (Behringwerke, A.G., F.R.G.); glutamate dehydrogenase, 53 kDa (Sigma); ovalbumin, 45 kDa (Sigma); aldolase, 40 kDa (Pharmacia, Uppsala, Sweden); concanavalin A, 27 kDa (Pharmacia); chymotrypsinogen 25.7 kDa (Sigma); aA-crystallin, 20 kDa (purified from bovine lens, provided by Dr. H. Driessen, Nijmegen); myoglobin, 17.8 kDa (Boom, Meppel, The Netherlands); cytochrome c, 13 kDa (Boehringer); cytochrome c (1-65), 7.8 kDa (gift from Dr. P. Boon, Nijmegen). Phosphorylation in the intact isolated retina

Cattle eyes were obtained from the local slaughter house immediately after the death of the animals. The eyes were kept in a light-tight container at room temperature and dissected in dim red light using an RG 665 filter with 1% T cut-off at 659 nm. Two retinas were incubated in a 20 ml polyethylene syringe in 5 ml of an oxygenated Ringer-bicarbonate solution containing (in mM): 125 NaC1, 5 KC1, 1.8 CaC12, 0.5 MgC12, 10 glucose, 25 NaHCO 3, 634 U / m l sodium penicillin G and 4.6 mg/ml streptomycin sulfate. Carrier free [32P]phosphate (1 mCi) was neutralized with 1.0 M

187 NaOH and added to the incubation solution to a final concentration of 0.2 mM phosphate. The retinas were gassed through the syringe orifice with 95% 02/5% CO2 at a rate of 3 ml/min, giving a final pH of 7.5. Various incubation times have been tried. The phosphate donor pools seem to equilibrate with 32po4 within 45 min, since after that time a steady-state level of protein phosphorylation was observed. However, better reproducibility was obtained with 60 min incubation times. A 60 min incubation time has therefore been used routinely. All results presented have been consistently observed in two or more experiments. The following conditions were tested: (1) Low C a 2 + levels (10-z M). This condition produces large changes in the electrical behavior of the photoreceptors within 5 rnin, both in darkness and in the light [18,19], and was therefore imposed during the last 5 min of the incubation period by addition of EGTA to a final concentration of 1.96 mM [2]. (2) Addition of isobutylmethylxanthine (1 rnM). Light-activation of the cyclic GMP-dependent phosphodiesterase and the subsequent decrease in cyclic GMP levels can be partially inhibited with isobutylmethylxanthine [20,21]. Eventually, isobutylmethylxanthine irreversibly depolarizes the rod [22]. Effects of the compound, were studied in darkness by addition of a 10 mM solution in physiological buffer containing 10% ethanol to a final concentration of 1 mM. (3) Saturating illumination. Intact retinas were flash-illuminated with a 20 ms xenon flash equipped with a 530 nm cut-off filter (Schott RG 530), bleaching about 20% of the rhodopsin. Such a saturating bleach strongly hyperpolarizes the rods for a period over 2 min [23]. Illuminated retinas were disrupted in the cold and samples were placed in trichloroacetic acid within this time-period. The electrophysiological performance of the isolated bovine retina under these conditions was checked occasionally in a Sickel chamber with freshly prepared Ag IAgC1 electrodes [24]. Conditions of perfusion and stimulation of dark-adapted bovine retinas, and the recording of the electroretinograms were carried out according to Winkler [25]. Trephinized pieces of retina (9 mm diameter) were placed in a Sickel chamber. B-waves were

irregular but the a-waves of the electroretinograms were found to be large (100 #V) and to agree well with literature data on freshly isolated rat retinas [26].

Phosphorylation in recombined outer segment preparations Rod outer segments and subfractions, i.e. disk and plasma membrane fractions, have been isolated as previously described [15,16]. In order to extract soluble kinase activity, isolated rod outer segments are suspended in a strongly hypotonic medium of 1 mM EDTA and 1 mM dithiothreitol adjusted with Tris to pH 8.1 at 0°C, and passed several times through a 19-gauge needle [27]. The suspension is kept on ice for 60 min for optimal solubilization of peripheral membrane-associated proteins and then centrifuged at 100000 × g for 60 min at 2°C. The stripped membrane fractions were washed twice in the same medium. The supernatant fraction, which contains about 1 mg protein/ml, as determined by a modified Lowry procedure [27], is immediately adjusted to pH 7.4 with 5 vol.% 2 M Tris-HC1 buffer containing 0.1 M MgC12. The resulting soluble rod outer segment fraction is either used within 10 h or quickly frozen on dry ice-ethanol, stored at - 80°C and used the next day. The standard reaction mixture for in vitro phosphorylations contains dark-adapted rod outer segment membranes suspended in 50 mM Tris-HCl buffer containing 5 mM MgC12 (pH 7.4). Samples of 50 #1 (12 nmol rhodopsin/ml) are transferred to Eppendorf tubes and supplemented in the case of stripped membrane fractions with 15 #g protein of the soluble protein fraction, which is equivalent to the amount extracted from isolated rod outer segments. The reaction is started by adding [32p]ATP. Relatively low concentrations of ATP (50-150 /~M) are used in order to obtain a high specific activity. In order to investigate light effects on the phosphorylation pattern, immediately after addition of the labelled triphosphates the samples are illuminated for 25 s through an orange filter (~ > 530 nm, Schott-Jena OG 530 filter) by a 75 W tungsten bulb at a distance of 20 cm. This bleaches about 20% of the rhodopsin present. After a 10 min incubation at 25°C the reaction is stopped by addition of 200 ttl ice-cold 10%

188 trichloroacetic acid containing 10 mM ATP and 5 m M H3PO 4. The precipitated membranes are collected by centrifugation (10 min, 1 2 8 0 0 × g ) , washed once with 0.5 ml 5% trichloroacetic acid in 5 mM H3PO 4 and collected by centrifugation in 1.5 ml Eppendorf vials.

Analysis of phosphorylation pattern After a 60 min incubation with [32P]phosphate, the retinas were transferred to 2 ml ice-cold isolation medium, containing 67 mM sodium phosphate buffer (pH 7.4), 2 mM EDTA, 10 mM adenosine and 600 mM sucrose. Phosphate buffer and adenosine were added to inhibit phosphatase activity [7,28] during isolation of the rod outer segments. The mixture was vortexed for 45 s in order to homogenize the retinas, and was then filtered through 125-mesh nylon gauze. The filtrate was layered on a 23-36% (w/w) sucrose density gradient [16] in 67 mM sodium phosphate buffer, 2 mM EDTA (pH 7.4). The rod outer segments were diluted with a half volume of the same buffer and were collected by centrifugation at 5000 x g for 10 min at 2°C. In some experiments the bovine rod outer segments have been isolated by means of 5-16% (w/w) Ficoll-400 density gradients made up in 600 mM sucrose and 20 mM Tris-HC1 buffer (pH 7.4) according to Schnetkamp et al. [29], or by means of density gradients of pure Ficoll according to Bownds et al. [30]. For analysis of the total retinal protein population, 200 /zl of ice-cold 10% trichloroacetic acid was added to 100 /~1 samples of retinal homogenate or filtrate. The mixture was kept on ice for 10 min in order to precipitate all proteins present. The precipitate was collected by centrifugation at 10000 × g for 2 min, and the pellet was washed once with water. All manipulations were carried out under dim red light (~ > 659 nm, Schott R G 665 filter), until the sample was dispersed in trichloroacetic acid. The protein content of the retinal and outer segment samples is determined according to a modified Lowry procedure [27]. For analysis by SDS-polyacrylamide gel electrophoresis, pellets are suspended in distilled water to a protein concentration of about 5 m g / m l and are solubilized by addition of one volume of twice-concentrated sample buffer [31]. Electrophoresis is performed

according to Laemmli [31]. After destaining, the slab gels are dried as described by Berns and Bloemendal [32]. For autoradiography, gels are placed on X-ray film (Kodak Royal X-Omat AR) and exposed for 10-14 days at - 7 0 ° C . Densitometric scanning of autoradiograms is performed by means of an LKB 2202 Ultroscan Laser Densitometer. Results and Discussion

Phosphorylation pattern in the intact bovine retina The present experiments were carried out in order to study protein phosphorylation in intact retinas under conditions where normal photoreceptor potentials of the rod cells are observed. When 9 mm trephinized disks of bovine retina were perfused with the standard Ringer bicarbonate/phosphate medium and illuminated with 20 ms light flashes, normal a-waves of 80-100 #V were obtained for at least 2.5 h after the death of the animal i.e., during a 1.5 h incubation. The b-wave was variable, of low amplitude, and highly sensitive to perfusion rate. It decreased dramatically after 1.5 h of incubation of the retina. The b-wave originates in higher-order neurons which are more susceptible to postmortem trauma. The normal size of the a-wave indicates that the photoreceptors retained normal function reasonably well. The phosphorylation pattern of the intact retina appears to be very complex. It was difficult to resolve single peaks above an apparent M r of 30 kDa (Fig. 1). Except for the phosphorylation of light-activated rhodopsin, only small changes in the protein phosphorylation pattern of whole retinas were observed both upon illumination and upon lowering extracellular Ca z+ (Fig. l a - d ) . Under all conditions tested, the stronger distinct band of phosphorylation was observed as a protein of approx. 12 kDa. The 12 kDa protein: location and phosphorylation The low-molecular-weight region of the retinal homogenate shows three strong Coomassie bluestained protein bands at 8, 13 and 14 kDa (Fig. 1, large arrows) and two weaker ones at 12 and 16 kDa (Fig. 1, small arrows). The corresponding autoradiogram shows four phosphorylated proteins in this region with apparent molecular masses

189 rho

5o

125

245

h~,Ca2+

+ , 1.8mM -7

"1-, 10 M

-

-

M W

xlO-

3

i I 8 12

12

I 20

I 28

I 45

I 50

i 75

I 93

,1.8mM

,

107M

I 'i 145 185 245 340

16

of 8, 12, 13.5 a n d 16 k D a . The three p h o s p h o r y lated p r o t e i n s of 8, 12 a n d 16 k D a c o r r e s p o n d to Coomassie blue-stained protein bands. The s t r o n g l y C o o m a s s i e b l u e - s t a i n e d b a n d s of 8, 13 a n d 14 k D a , as well as a strongly p h o s p h o r y l a t e d p r o t e i n of 12 k D a , are m a r k e d l y enriched in the d e n s e retinal fraction (P > 1.15 g / m l : Fig. 2A, l a n e 5; Fig. 2B, lanes c a n d c'). R o d outer seg-

Fig. 1. Protein and phosphorylation pattern of the intact bovine retina labelled with [a2p]phosphate. All retinas were isolated in dim red light (Scott-Jena RG 665 filter) and incubated at 37°C in complete darkness, a, c: incubation for 60 min in 1.8 mM Ca2+; b, d: incubation for 55 min in 1.8 mM Ca2+, followed by 5 min in 10 -7 M Ca2+; a, b: a saturating light flash (bleaching approx. 20% rhodopsin) is given at the end of the incubation period; c, d: dark-adapted samples. Gels contain 28 /Lg protein. Lower gels (a-d): Coomassie blue-stained protein pattern of an SDS-polyacrylamide gradient gel (8-19% acrylamide); upper gels (a'-d'): autoradiograms presenting the phosphorylation pattern; traces (a"-d"): densitometric scans of the corresponding lanes on autoradiograms.

m e n t s purified on sucrose d e n s i t y g r a d i e n t s contain only m i n o r a m o u n t s of low m o l e c u l a r mass C o o m a s s i e b l u e - s t a i n a b l e p r o t e i n s (8, 13 a n d 14 k D a ; Fig. 2A, lane 2b; Fig. 2B, lane b). A similar p a t t e r n is o b t a i n e d when the sucrose d e n s i t y gradient is replaced b y a 5 - 1 6 % ( w / w ) Ficoll g r a d i e n t m a d e up in 600 m M sucrose, which is c l a i m e d to preserve the functional integrity of the isolated rod

190

23 w

sucrose

36%

ä

1

<

a bed

a’ b’ c’

r ho

121kD6

I3

Fig. 2. Protein and phosphorylation pattern of retinal fractions separated by sucrose density-gradient centrifugation. (A) SDS-polyacrylamide gradient gel (8-198 acrylamide) stained with Coomassie blue. Lane 1. upper band from sucrose gradient; lane 2, rod outer segment band; lane 3, lower band, just below rod outer segments; lane 4, colorless layer on top of the pellet; lane 5. brownish pellet, lane 6, marker proteins. Lanes designated (b) represent membrane-associated material precipitated by centrifugation for 10 min at 10000X g, at 4°C after two-fold dilution with sucrose-free buffer. Lanes designated (a) represent trichloroacetic acid-precipitated material from the resulting supematants. (B) Protein and phosphorylation pattern of illuminated 32P-labelled retinal fractions, obtained by sucrose density centrifugation and analysed by SDS-polyacrylamide-gradient gel electrophoresis (7-14% acrylamide). a-c, Coomassie blue-stained protein patterns; a’-c’, corresponding autoradiogram presenting the phosphorylation patterns. Lane a, retinal homogenate (85 pg protein); lane b, rod outer segments (approx. 20% rhodopsin bleached, 35 pg); lane c, pellet (p > 1.15 g/ml) of retinal fraction in sucrose density gradient (63 pg); lane d, marker proteins. Note the almost exclusive presence of the strongly 32P-labeiled low-molecular-weight proteins in the pellet fraction (lane c’).

outer segments better than do pure sucrose density gradients [29]. The 12 kDa component shows a higher phosphate incorporation than opsin in retinal homogenates (Fig. la and b). In darkness, low Ca2+ levels (lo-’ M) decrease the phosphorylation level of the 12 kDa species to 80% of its original level (SE. = 5, n = 3), while at a normal Ca’+-level (1.8 mM) its phosphorylation level is enhanced by illumination up to 50% (S.E. = 8, n = 3). In rod

outer segments the 12 kDa protein is usually not detectable by Coomassie blue staining, but only by autoradiography when phosphorylated (Fig. 2B). The amount of the 12 kDa protein remaining in the isolated rod outer segment fraction was estimated to be less than 2%. This figure was calculated from the ratio of the amount of 12 kDa phosphoprotein in rod outer segments and retina as obtained by densitometry of autoradiograms after correction for the percentage of rod outer

191

segment protein present in retinal homogenates. The 12 kDa species in bovine retina resembles the 12 kDa phosphoprotein observed in the frog retina by Bownds and co-workers [33,34]. They have reported the presence of two phosphorylated proteins of 12 and 13 kDa in frog retina as well as in frog rod outer segments, isolated by means of a discontinuous Ficoll density gradient [30]. However, these frog proteins are highly phosphorylated in darkness, the reaction is stimulated by low Ca2÷ levels (10 -9 M), and both proteins are dephosphorylated upon illumination [33]. When bovine rod outer segments are isolated in a similar way by means of discontinuous Ficoll density gradients, the rod outer segment band shows heavy contamination with other membrane material and retains much more of the 8-16 kDa species (data not shown). Since these proteins do not appear to co-purify with rod outer segments, we suggest that the strongly phosphorylated 12 kDa protein in the bovine retina and probably the other low M r species (at 8, 13 and 16 kDa) as well, are not endogenous to rod outer segments. Rather, they

may be associated with some other light-responsive part of the rod photoreceptor cell, i.e., synaptic membranes or synaptosomes. The phosphorylatable low-molecular-weight proteins in bovine retina probably do not bear any relationship to proteins of similar molecular weight in frog, in view of their different response to light and Ca2÷ and their different localization.

Phosphorylation pattern of rod outer segments The protein and phosphorylation pattern of bovine rod outer segments, isolated from [32p]. phosphate labelled retinas by means of a sucrose density gradient, is shown in Fig. 3, and densitometric scans are shown in Fig. 4. Table I compiles the phosphorylated proteins which have been observed. The levels of [32p]phosphate incorporation into outer segment membrane proteins of 245, 226, 125, 110, 50 and 20 kDa are enhanced upon illumination of the retina (Figs. 3 and 4). The weak Coomassie blue-stained bands in the high molecular weight region (340, 295, 265 kDa) are not reproducibly detected in outer segments and

TABLE I P H O S P H O R Y L A T E D PROTEINS IN R O D O U T E R SEGMENTS ISOLATED F R O M 32pO4-LABELLED R E T I N A Apparent M r (kDa)

Attributed to disk membrane (DM) plasma membrane (PM)

Stimulating factors

245 226 125 110 85 68 66 56 50 46 38 (rhodopsin) 32 25 20 18

PM > D M PM DM, PM PM cytosol(?) cytosol cytosol cytosol PM cytosol DM, PM cytosol cytosol PM cytosol

light light light light light

Inhibition by 10-7 M Ca 2+

+ + +

light

+

light light light

+ + +

light light 10 - 7 M Ca 2+

+

a In isolated rod outer segments, phosphorylation of these proteins is not always detectable.

Detectable upon labelling of isolated outer segments with [ y 32P]ATP + + + + + a + + + + + + + + + + a

192

mw

(xlo

- 245 -226

2

whole retinas than in isolated rod outer segment fractions, (Figs. 4 and 5, Table I). This could be due to: (1) the experimental conditions still being suboptimal for the isolated outer segments (e.g., the levels of trinucleotides applied are lower than in the intact retina); (2) the use of a more general phosphate source for the incubated retina ([ 32p ] p o 4 instead of [y-32 P]ATP); (3) selective loss of kinases or cofactors upon isolation of stripped membrane fractions; or (4) structural and physiological dis-

-125 -110

ffrho

jt~, dimer --

-

~0

a

-

226

to hv, IBI~X

50

",

/

20

+ , 1 mH

-- ~lmM J

v,.:i ' ; i

/

2i°

/~ '

340

'

a b c d a'b'c'd' Fig. 3. Protein and phosphorylation pattern of rod outer segments isolated from 32p-labelled retinas. Outer segments samples (35 #g protein) are analysed on an SDS-polyacrylamidegradient gel (6-15% acrylamide), a - d , protein patterns (Coomassie blue); a ' - d ' autoradiograms. Conditions: a, 1.8 m M Ca 2+ extracellularly, approx. 20% rhodopsin bleached in the retina before isolation of outer segments; b, 10 -7 M Ca 2+, approx. 20% rhodopsin bleached in the retina before isolation of outer segments; c, 1.8 m M Ca 2+, outer segments isolated from dark-adapted retinas; d, 10 -7 M Ca 2+, outer segments first isolated from dark-adapted retinas, then illuminated (approx. 15% rhodopsin bleached).

may originate from material denser than the rod outer segment band (cf. Fig. 2A, lane 3b), or represent cytoskeletal proteins. Therefore they are not included in Table I.

Light effects on phosphorylation Illumination leads to more pronounced phosphorylation of rod outer segment proteins of 245, 226 and 125 kDa when performed on 32p-labelled

~

F......

'

At-, 1,8 rlnM ,1.8ram

-

-

+ ,10-' H

+,10-' H

MW

1~ 2~ ~s

& 1o 6~ rl . b , ~ ~

22~,,s ~oo

Fig. 4. Phospho~lation pattern of outer segment proteins isolated from 32p-labelled retinas, incubated with low Ca 2+ or isobutylmethylxant~ne. Densitometric scans of autoradiograms are averages from two gradient gels (6-15% acrylamide). a, b: retinas incubated in the presence of isobutylmethylxanthine (1 mM); a: outer segments from illuminated retinas; b: outer segments from dark-adapted retinas; c, d: control incubation with normal Ca 2+ (1.8 mM); c: outer segments isolated from illuminated retinas; d: outer segments from dark-adapted retinas; e, f: Ca 2+ is lowered from 1.8.10 -3 to 10 - 7 M during the last 5 rain of incubation; e: outer segments from illuminated retinas; f: outer segments illuminated after isolation by sucrose density-gradient centfifugation.

193

connection of rod outer segments from intact retina, with concomitant disassembly of cytomatrix elements.

MW (xlO"

130 93 68 53 a5 4O

25.7 2O 1 7.8 13 7.8'

Fig. 5. Comparison of the phosphorylation pattern with ATP as phosphate donor in outer segment membrane fractions (0.6 nmol rhodopsin) recombined with cytosolic proteins (15 # g). A plasma membrane fraction recombined with cytosolic proteins (lane c: Coomassie blue pattern) and depleted total outer segment membranes, recombined with cytosolic proteins, (lane d: Coomassie blue pattern), are partially bleached (5% of the rhodopsin present), while incubated under standard conditions in the presence of 2-10 - s M cyclic GMP and 2-10 - s M cAMP with 25.10 -6 M [32plATP (7.2 Ci/nmol) as a phosphate donor. Lane b shows cytosolic proteins. Lane a shows calibration proteins. Lanes c', d' show autoradiograms of lanes c and d respectively. Arrows indicate regions of the gel where proteins enriched in the plasma membrane fraction are detected.

Effects of Ca 2 + and isobutylmethylxanthine Both lowering extracellular Ca 2+ levels by EGTA addition and treatment with isobutylmethylxanthine in the presence of normal calcium levels (1.8 mM) increase the cyclic GMP levels in isolated retina and depolarize the rod [35,36]. However, effects on the receptor potential, peak amplitude and sensitivity are different [37,38]. We also find different effects on the phosphorylation patterns of rod outer segment proteins isolated from retinas incubated in low Ca2+ (10 -7 M) as compared to isobutylmethylxanthine. Treatment of retinas with 1 mM isobutylmethylxanthine during 60 min reduces the amount of phosphate incorporation to less than 40% of normal levels. It nevertheless results in a fairly normal dark phosphorylation pattern (Fig. 4, scan b), except that rhodopsin shows much stronger dark phosphorylation (although well below the normal light-level). Upon illumination, no significant increase in phosphate incorporation is observed. Thus, prolonged treatment with isobutylmethylxanthine abolishes light responses in the major phosphorylation routes. It has been reported that incubation with isobutylmethylxanthine over long periods of time (30-60 min), decreases the light-sensitivity of the rod cell: the increase of the receptor potential observed initially is reversed within 1 h of treatment [29], and the cells seem to become irreversibly poisoned [39,40]. The relatively enhanced dark-phosphorylation of rhodopsin which we observe upon isobutylmethylxanthine treatment could be one of the mechanisms which desensitize the rod cell. However, the effects of shorter isobutylmethylxanthine treatments should be evaluated before the results can be interpreted in terms of relevant electrophysiological models. Lowering the extracellular Ca2+ levels by EGTA addition depolarizes the rod [18,22,40] by an increase in the membrane conductance for sodium ions. Hence, the response amplitude increases when the sodium conductance is again decreased upon illumination. This has led us to study the effect of Ca 2+ on phosphorylation levels. Low extracellular Ca 2÷ (during the final 5 min) seems to decrease the dark-phosphorylation levels of proteins at 125, 85, 66, 46, 38 and 20 kDa in the intact retina and appears to increase phosphorylation of a 18 kDa

194

band (not shown). Low C a 2+ c a u s e s a moderate decrease in the phosphorylation levels of the 125, 85 and 50 kDa bands in the light (Fig. 3a vs. b and Fig. 4 c vs. e).

Plasma membrane proteins of rod outer segments Six of the phosphorylated proteins that we detect in rod outer segments are relatively strongly labelled when only a plasma membrane fraction is supplied with the soluble protein fraction (Fig. 5, lane c' vs. d'). These proteins migrate with molecular masses of 245, 226, 110, 50, 46 and 20 kDa. Two of these proteins (226, 110) are not detectable in disk membrane fractions and seem to be unique to the plasma membrane [15]. These two proteins are probably identical to proteins iodinated from the extracellular space in intact outer segments [41]. Light-stimulated phosphorylation of two high molecular weight proteins of 240 and 220 kDa in frog outer segments has also been reported by Szuts [42]. Our evidence does not indicate that the 226 kDa protein that we detect in bovine rod outer segments corresponds to the large protein described by Papermaster et al. [43]. The latter is a glycoprotein, migrated with a slightly lower molecular mass (220 kDa, Figs. 3 and 5) and does not show a high level of phosphate incorporation.

Conclusions A 12 kDa protein is the most prominent phosphorylated protein in the bovine retina and shows phosphorylation changes in response to light, Ca 2 + and isobutylmethylxanthine. In bovine retina, the phosphorylation of this protein is stimulated by light and calcium, in contrast to a similar component in frog retina. The 12 kDa protein does not seem to purify with bovine rod outer segments and we suggest that it is a synaptic protein. The phosphorylation pattern in bovine rod outer segments is rather similar, whether initiated by labelling intact retina or by labelling isolated rod outer segments (Table I). These results support the use of isolated outer segment fractions to investigate separately the effects of Ca 2 ÷ or cyclic GMP, since in the intact rod, intracellular Ca 2÷ levels and cyclic G M P levels seem to have some negative mutual interaction [44]. The results of Table I further show that phosphate incorporation in most

phosphorylated proteins is reduced by the presence of low extracellular Ca 2+. Whether a protein kinase C activity [45] is involved in these phosphorylation reactions remains to be established. The strongly light-dependent phosphorylation of the two putative plasma membrane proteins of 226 and 110 kDa indicates that these could be candidates for light-regulated carrier proteins. Taken together with the observation that a cyclic GMPsensitive sodium-conductance mechanism is present in the plasma membrane, which is not modulated by calcium ions, ATP or GTP [13,14], our results are suggestive of direct or indirect regulation of this or another conductance mechanism by light-modulated phosphorylation. Proteins of 50 and 20 kDa, that according to Coomassie blue staining as well as to phosphate incorporation are 2-5-fold enriched in plasma membrane fractions as compared to whole rod outer segment membranes, could actually represent modulatory subunits of light-dependent channel or carrier proteins or intermediate signal-transducing proteins.

Acknowledgements We thank Dr. P. Boon for valuable comments. This investigation was supported by the Netherlands Organization for the Advancement of Basic Research (ZWO) through the Netherlands Foundation of Chemical Research (SON).

References 1 Hubbell, W.L. and Bownds, M.D. (1979) Annu. Rev. Neurosci. 2, 17-34 2 Bownds, M.D. (1981) Curr. Top. Membranes Transp. 15, 203-214 3 Greengard, P. (1978) Science 199, 146-152 4 Greengard, P. (1980) J. Biol. Chem. 255, 11619-11629 5 Fung, B.K.-K. and Stryer, L. (1980) Proc. Natl. Acad. Sci. USA 77, 2500-2504 6 Fung, B.K.-K., Hurley, J.B. and Stryer, L. (1981) Proc. Natl. Acad. Sci. USA 78, 152-156 7 Wilden, U. and K~bn, H. (1982) Biochemistry 21, 3014-3022 8 Lolley, R.N., Brown, B.M. and Farber, D.B. (1977) Biochem. Biophys. Res. Commun. 78, 572-578 9 Shuster, T.A. and Farber, D.B. (1984) Biochemistry 23, 515-521 10 Lee, R.H, Brown, B.M. and Lolley, R.N. (1984) Biochemistry 23, 1972-1977

195 11 Fesenko, E.E. and Orlov, N.Ya. (1980) Mol. Biol. 14, 617-622 12 Lee, R.M., Farber, D.B. and Lolley, R.N. (1982) Methods Enzymol. 81,496-506 13 Fesenko, E.E., Kolesnikov, S.S. and Arkadiy, A.L. (1985) Nature 313, 310-313 14 Cobbs, W.H. and Pug/a, E.N., Jr. (1985) Nature 313, 585-587 15 Kamps, K.M.P., De Grip, W.J. and Daemen, F.J.M. (1982) Biochim. Biophys. Acta 687, 296-303 16 De Grip, W.J., Daemen, F.J.M. and Bonting, S.L. (1980) Methods Enzymol. 67, 301-320 17 Wheeler, G.L. and Bitensky, M.W. (1977) Proc. Natl. Acad. Sci. USA 74, 4238-4242 18 Bastian, B.L. and Fain, G.L. (1979) J. Gen. Physiol. 297, 493-520 19 Woodruff, M.L., Fain, G.L. and Bastian, B.L. (1982) J. Gen. Physiol. 80, 517-536 20 Miller, J.P., Boswell, K.H., Muneyama, K., Simon, L.N., Robins, K.H. and Shuman, D.A. (1973) Biochemistry 12, 5210-5219 21 Miki, N., Baraban, J.M., Keirns, J.J., Boyce, J.J. and Bitchsky, M.W. (1975) J. Biol. Chem. 250, 6320-6327 22 Lipton, S.A., Rasmussen, H. and Dowling, J.E. (1977) J. Gen. Physiol. 70, 771-779 23 Albani, C., N611, G.N. and Yoshikami, S. (1980) Photochem. Photobiol. 32, 515-520 24 Siekel, W. (1965) Science 148, 648-651 25 Winkler, B.S. (1972) Vision Res. 12, 1183-1198 26 Winkler, B.S., Simson, V. and Benner, J. (1977) Invest. Ophthalmol. Visual SCi. 16, 766-768 27 Peterson, G.L. (1977) Anal. Biochem. 83, 346-356 28 Fairbanks, G., Avruch, J., Dino, J.E. and Patel, V.P. (1972) J. Supramol. Struct. 9, 97-1112

29 Schnetkamp, P.P.M., Klompmakers, A.A. and Daemen, F.J.M. (1979) Biochim. Biophys. Acta 552, 379-389 30 Bownds, M.D., Gordon-Walker, A., Gaide Huguenin, A.C. and Robinson, W. (1971) J. Gen. Physiol. 58, 225-237 31 Laemmli, U.K. (1970) Nature 227, 680-685 32 Berns, A.J.M. and Bloemendal, H. (1974) Methods Enzymol. 30, 675-694 33 Polans, A.S., Hermolin, J. and Bownds, M.D. (1979) J. Gen. Physiol. 74, 595-613 34 Hermolin, J., Karell, M.A., Hamin, H.E. and Bownds, M.D. (1982) J. Gen. Physiol. 79, 633-655 35 Brown, J.E. and Waloga, G. (1981) Curr. Top. Membranes Transp. 15, 369-380 36 Ostroy, S.E., Meyertholen, E.P., Stein, P.J., Svoboda, R.A. and Wilson, M.J. (1981) Curr. Top. Membranes Transp. 15, 383-403 37 Waloga, G. and Brown, J.E. (1979) Invest. Ophthalmol. ARVO Abstr. No. 3, 5-6 38 Brown, J.E., Coles, J.A. and Pinto, L.H (1977) J. Physiol. (London) 269, 702-722 39 Leser, K.H. (1981) Z. Naturforsch. 36, 597-603 40 Brown, J.E. and Pinto, L.H. (1974) J. Physiol. (London) 236, 575-591 41 Clark, V.M. and Hall, M.O. (1982) Exp. Eye Res. 34, 847-859 42 Szuts, E.Z. (1985) Biochemistry 24, 4176-4184 43 Papermaster, D.S., Schneider, B.G., Zorn, M.A. and Kraehenbuhl, J.P. (1978) J. Cell Biol. 78, 415-425 44 Cohen, A.I. (1982) J. Neurochem. 38, 781-796 45 Kapoor, C.L. and Chader, G.J. (1984) Biochem. Biophys. Res. Commun. 122, 1397-1403