Journal of Neuroehemistry Raven Press, Ltd., New York 0 1992 International Society for Neurochernistry

Effect of Brain Ischemia on Protein Kinase C Krystyna Domaiiska-Janik and Teresa Zalewska Department of Neurochernistry, Medical Research Centre, Polish Academy of Sciences, Warsaw, Poland

resulted in a further increase of its relative activity up to 162% of control values. In vitro experiments using a synaptoneurosomal particulate fraction were performed to clarify the mechanism of the rapid PKC inhibition observed in cerebral tissue after ischemia. These experiments showed a progressive, Ca2+-dependent,antiprotease-insensitivedownregulation of PKC during incubation. This down-regulation was significantly enhanced by prior phorbol (PDBu) treatment. It can be postulated that in the ischemic gerbil hippocampus model as well as during phorbol stimulation of the synaptoneurosomal fraction in vitro, the common phenomenon was an initial PKC activation (translocation) followed by a subsequent enhancement of Ca2+-dependentenzyme inhibition. Experimentaldata suggest that the inhibition is localized in the membrane compartment. Key Words: Protein kinase C-Translocation-Brain-IschemiaPhorbol ester. Domanska-Janik K. and Zalewska T. Effect of brain ischemia on protein kinase C. J. Neurochern. 58, 1432-1439 (1992).

Abstract: We examined the influence of brain ischemia on the activity and subcellular distribution of protein kinase C

(PKC). Two different models of ischemic brain injury were used: postdecapitative ischemia in rat forebrain and transient (6-min) cerebral ischemia in gerbil hippocampus. In the rat forebrain model, at 5 and 15 min postdecapitation there was a steady decrease of total PKC activity to 60% of control values. This decrease occurred without changes in the proportion of the particulate to the soluble enzyme pools. Isolated rat brain membranes also exhibited a concomitant decrease of [3H]phorbol 12,13-dibutyrate ([3H]PDBu)binding with an apparent increase of the ligand affinity to the postischemic membranes. On the other hand, the ischemic gerbil hippocampus model displayed a 40% decrease of total PKC activity, which was accompanied by a relative increase of PKC activity in its membrane-bound form. This resulted in an increase in the membrane/total activity ratio, indicating a possible enzyme translocation from cytosol to the membranes after ischemia. Moreover, after 1 day of recovery, a statistically significant enhancement of membrane-bound PKC activity

generation of 50-kDa protein kinase (PKM), which is a product of proteolytic degradation of PKC and is also catalytically active in the absence of Ca2’ and phospholipids (Melloni et al., 1985). It has been generally accepted that at least two mechanisms of ischemic brain injury exist. The first results from a prolonged depletion of cellular high-energy reserves and leads to generalized tissue necrosis, termed “infarction.” The other, initiated by comparatively moderate ischemia, consists of progressive postischemic membrane perturbation in which, over several days, selective neuronal death occurs in only vulnerable brain areas (Kirino, 1982; Kirino and Sano, 1984). It has been proposed that PKC is a link that couples short-term reactions initiated by excitatory neurotransmission and calcium mobilization with long-term

Among the many kinases, Ca2+/phospholipid-dependent protein kinase-known as protein kinase C (PKC)-is most abundant in brain tissue (Nishizuka, 1986). A wide spectrum of neuronal processes, including the signal transduction system (Sugden and Klein, 1988; Weiss et al., 1989), ion channels (Kaczmarek, 1986), and synaptic plasticity (Akers and Routtenberg, 1987; Routtenberg, 1987), has been demonstrated to be sensitive to PKC. In most tissues, enzyme activation (translocation) (Takai et al., 1979) is followed by increased binding of specific PKC ligands-phorbol esters-to the plasma membrane. Another feature of phorbol ester action on the intact cell is the rapid activation (translocation) and subsequent loss of PKC activity (Shenolikar et al., 1986; Matthies et al., 1987). In several tissues this down-regulation of PKC by phorbol esters leads to the ~~

~

Received May 2, 1991;revised manuscript received August 8, 1991; accepted September 17, 1991. Address correspondence and reprint requests to Dr. K. DomanskaJanik at Department of Neurochemistry, Medical Research Centre, Polish Academy of Sciences, ul. Dworkowa 3, 00-784 Warszawa, Poland.

Abbreviations used: KRH buffer, Krebs-Ringer-Henseleit buffer; PDBu, phorbol 12,13-dibutyrate: PKC, protein kinase C: PKM, 50kDa protein kinase; PMSF, phenylmethylsulfonyl fluoride.

1432

PROTEIN KINASE C DURING BRAIN ISCHEMIA alteration in cell function, by regulating the expression of specific target proteins, which function as nuclear “third messenger” molecules (Hunter et al., 1988). The expression of these “third messengers,” which directly or indirectly induce the transcription of genes encoding proteins contributing to the adaptative responses, was recently reported in postischemic and other related brain pathologies (Dragunow and Robertson, 1987; Herrera and Robertson, 1989; Onodera et al., 1989b; Nowak et al., 1990). However, the mediating role of PKC in this response still remains controversial. The aim of this work was to compare PKC activity in two distinct types of brain ischemia using either a postdecapitativerat model or a transient ischemic gerbil model. Preliminary data showing a decrease of enzyme activity after these ischemic insults have been presented previously (Domaliska-Janik and Zalewska, 1988). Besides measuring PKC‘s phosphorylating enzyme activity, we also estimated the binding of phorbol 12,13-dibutyrate (PDBu) to postischemic rat brain membranes. Particular attention was paid to the possible conversion of PKC to the activator-insensitive enzyme form, PKh4. Furthermore, to understand better the mechanism of the observed decrease of enzyme activity after ischemia, additional in vitro experiments, using a rat synaptoneurosomal particulate fraction, were performed in which calcium- and phorbol-dependent activation and deactivation of PKC were measured.

MATERIALS AND METHODS Animal models Postdecapitative global ischemia was produced using 3month-old Wistar rats of both sexes. The heads were quickly sealed in plastic bags and kept at 37°C for the indicated time. Ischemia of the cerebral hemispheres in gerbils (Meriones unguiculutus) was produced by bilateral ligation of the common carotid arteries in the midcervical region through a midline incision with the animal under light ether anesthesia. The animals were decapitated after occlusion for 6 min, or the clips were removed. The animals were then allowed to recover for the indicated time. Sham-operated animals awakening from anesthesia served as controls.

Preparation of brain homogenate or particulate fraction Homogenates (5% wt/vol) were prepared after decapitation from quickly (90% of this activity was recovered in fractions 2-6, these fractions were combined and used for enzymatic determination. In some experiments, the columns were further eluted with 0.4 M NaCl in buffer A, and 0.5-ml fractions were collected and used for determination of Ca’+/phospholipid-independent (PKM) kinase activity as recommended by Melloni et al. (1985). PKC activity was measured as the difference in histone kinase activity observed in the presence and absence of phosphatidylserine, calcium, and phorbol 12-myristate I 3-acetate. The enzyme eluted from DE-52 columns (30 pl) was incubated in 20 mMTns (pH 7 3 , 5 mM MgC12, 1.O mMCaClz (or 1.O m M EGTA), 15 pg of histone type I1 A, 10 p M [y3ZP]ATP(0.25 pCi), 10 p M phosphatidylserine, 1.5 p M phorbol 12-myristate 13-acetate, and a trace of PMSF and 2-mercaptoethanol in a total volume of 0.125 ml. Because the protein sample usually contained EDTA/EGTA in Tris buffer (see above), the net Ca2+concentration was 0.375 mM in these conditions. The mixture was routinely incubated for 3 rnin at 30°C; when extrapolated, however, the reaction proceeded linearly for up to 9 min. The reaction was stopped by addition of 0.5 ml of an ice-cold mixture of 0.4% albumin with 20 mM Na3P04and 20 mM Na4P207in 5% trichloroacetic acid. Each sample was immediately filtered through presoaked Whatman GF/C filters positioned under vacuum in a multiple-sample filtration Millipore manifold. The filters were washed twice with 5 ml of stopping solution without

J. Neurochem., Vol. 58, No. 4, 1992

K. DOMANSKA-JANIK AND T. ZALE WSKA

1434

albumin, and radioactivity was counted in Bray's mixture with a Beckman LS 9000 counter. PKC activity, expressed in units (U = picomoles of 3 2 P 0 4 3 bound to substrate per minute), was measured and normalized either to the amount of protein applied to the column or to the total brain wet weight.

rH]PDBu binding to membranes Rat brain membranes were prepared from 10% homogenate in 20 mM Tris (pH 7.5) with 0.2 mM PMSF. Where indicated, 0.5 mM EGTA and 2 mM EDTA were added to the preparation buffer. Homogenates were centrifuged for 3 rnin at 2,400 g, and the pellet was discarded. Supernatants were centrifuged at 50,000 g for 10 min, and the pellets were washed twice with the appropriate buffer and sedimented again at the same speed. [3H]PDBu (specific activity, 23 Ci/mmol) was added to the 0.5 ml of sample (containing -50 pg of protein) to a concentration of 25-50 nM. Incubations were performed in triplicate for 30 rnin at 0°C. The binding was stopped by addition of 1 ml of 0.4% albumin in 20 mM Tris, pH 7.5. The material was then filtered through Whatman GF/C filters and washed twice with 5 ml of 20 mM Tris, pH 7.4. Nonspecific binding was determined in the presence of 1 p M PDBu (unlabeled) and subtracted from the total counts to yield specific binding. The nonspecific binding did not exceed 30%of the total binding, which was always < 10%of the total activity in the sample. The KD(affinity constant)and B,, (total number of binding sites) values were calculated using Scatchard analysis, and regression lines were fitted to the data points by least squares analysis. The activity of lactate dehydrogenase was assayed spectrophotometrically according to the procedure of Bergmeyer (1963). Protein content was assayed by the method of Lowry et al. (1951). For statistical analysis the t test for small samples (Bailey, 1975) was used to evaluate differences between the expenmental groups.

'"t

cantd

5'

-ischemia

15'

+

FIG. 1. PKC activity in rat brain after postdecapitative ischemia. Enzyme activity was measured in cytosol(0) and membranous(B) fractions according to the procedure described in Materials and Methods. Data are mean f SD (bars) values from four animals. The groups significantly different from controls are indicated: *p < 0.002.

J . Neurochem., Vol. 58, No. 4, 1992

TABLE 1. f 'H]PDBu binding to rat brain membranes after ischemia Protocol, preparation

Bm, (pmollmg of protein)

KD

(a)

A Control Ischemia

16.2 k 1.7 11.3 k 1.7"

15.0 +- 2.5 10.5 & 1.14"

Control Ischemia

2.8 f 0.32 2.5 k 0.30

7.5 f 1.81 5.9 +- 1.76

Control Ischemia

6.8 f 0.31 7.1 f 0.48

5.5 ? 0.61 5.6 ? 0.92

B

C ~

Data are mean f SD values obtained in three separate experiments. The protocols were as follows: The differences in KD and Bmaxare significant (p < 0.001) for all groups. A, membranes prepared and washed three times in Tris buffer; B, membranes washed three times in Tris buffer containing 2 mM EDTA and 0.5 mM EGTA, and C, as for B but supplemented with calcium before binding. Specific 13H]PDBu binding (mean k SEM, in cpm per sample X lo3) at a ligand concentration of 25 nM was as follows: in A, 10.90 k 0.42 for control and 8.72 0.48 for ischemia; in B, 2.56 f 0.08 and 2.47 f 0.05, respectively; and in C, 5.97 k 0.48 and 5.32 k 0.22, respectively. p < 0.05 for difference from reference control.

*

RESULTS Postdecapitative rat brain ischemia resulted in a significant, time-dependent decrease in histone type I1 Aphosphorylating PKC activity. In the cytosolic and membrane fractions, a similar degree of enzyme inactivation was observed: -70% of control values after 5 min and 60%of control values after 15 min of blood flow interruption (Fig. 1). Ca*+/phospholipid-independent (PKM) activity was negligible in the rat brain fractions and did not change after postdecapitative ischemia (data not shown). The binding of the specific PKC ligand [3H]PDBu to postischemic rat brain membranes changed proportionally to the phosphorylating enzyme activity, decreasing to -70% of control values (Fig. 1 and Table 1). Scatchard equilibrium analysis revealed a concomitant increase in the affinity for [3H]PDBu binding after ischemia. There were, however, differencesbetween the protocols of the membrane preparations for catalytic and ligand-binding PKC estimations. For the former purpose the brain was homogenized in the presence of EDTA/EGTA (buffer A), and the pellet was only separated by one-step centrifugation. This procedure has been recommended for stabilizing the enzyme translocation because of in situ intracellular Ca2+ level changes. On the other hand, extensive washing of the membrane is necessary for obtaining reproducible results in the PDBu binding. Therefore, the membnnes were prepared in Tris buffer and washed three times with the same buffer (see Materials and Methods), or to this same mixture calcium chelators (buffer A) were added.

PROTEIN KINASE C DURING BRAIN ISCHEMIA

I435

TABLE 2. Efect of ischemia and recovery on PKC activity in gerbil brain ~~

~

Hippocampus

Cerebral cortex

Recovery

(n) Control Ischemia (6 min)

0

(5)

(4) 1 h (4) 24 h ( 5 )

M/M

+C

Membranes

Cytosol

Membranes

26.8 k 1.3

120.6 2 13.5

0.2 1

24.6

21.8 ? 5.9 26.0 -t 3.1 36.8 ? 5.1 a

66.1 k 14.7" 100.5 k 10.7b 84.5 2 4.8"

0.26 0.26 0.34

18.8 k 2.5' 24.0 +_ 3.0 21.4 k 3.4

* 3.9

Cytosol

M/M

+C

54.8 ? 4.9

0.30

46.0 t 8.6 46.3 k 1.15 55.2 _+ 6.6

0.28 0.29 0.21

~

+

Data are mean SD values, in units per milligram wet weight, from the number of experiments given in parentheses. M/M to total enzyme activity ratio. a p

Effect of brain ischemia on protein kinase C.

We examined the influence of brain ischemia on the activity and subcellular distribution of protein kinase C (PKC). Two different models of ischemic b...
905KB Sizes 0 Downloads 0 Views