130

BWchimwa et Bioplo'~ica A¢'fa, 1082(1991) 130-135 1991 ElsevierScience Publishers BV. 0005-2760/91/$03.50 A D O N I S 0D0527609100|123

Subcellular localization of rat gastric phospholipase A R e n ~ e G r a t a r o l i i, Elise T e r m i n e n, H e n r i P o r t u g a l 2 A n n e - M a r i e Huguette Lafont t and Gilles Nalbone : Instttu¢ Natiomd de l~ Sunt~

P a u l i 2,

¢':.rela Recht, r~he M~:dicale,

Unit~ 130, Murseille (Frame) and ~' H~pital Sic-Marguerite, l.aboratoire Uentral, Marseitte (France)

(Received 30 OcltJber 1990) Key words: PhospholipaseAz; Gaslric mucosa; Subcellularlocalization;(Rat} In the present study, we have performed experiments to gain some insight into the suheellular localization and biochemical properties of gastric mucosal phospholipase A z. After classical snbcelhilar fractionation of whole glandular stomach mucosa, we found that gastric phospholipase A z was essentially enriehod in the 105000 x g pellet that contains microsumes and plasma membranes. Except for the cytosol0 all the suheelhilar fractions exhibited similar phospholipase A 2 activity (i.e., optimum of pH, calcium dependence, apparent K m and positional specificity). The high-speed pellet was further characterized by ultracentTifugation on a sucrose gradient. Data showed that the sedimentation profile of phospholipase A z was quite similar to those of plasma membrane markers and more specifically to an apical membrane marker. These results, taken together, showed that a gastric phosphoUpase A z is distriboted among the various subcellular fractions (as a result of cross-contamination) together with the membrane fraction on which it is assodated. It is proposed that this fraction is the apical plasma membrane which would he the main slte of plt,~l~ollpase A± action for arachidonic acid release. Lysophospholipase showed the same sedimentation profile as phospholipase Az, whereas acyl CoA-lysophosphatidylcholine:acyltransferase mainly sedimented with heavy microsomes. The substrate specificity of the enzyme was assessed by endogenous hydlr¢tysis of gastric mucosal phespholipids. We were able to show that the enzyme acts at nearly the same rate on two major gastric membrane phospholipids, namely phosphatidyleholian and phosphatidylethanolamlne.

Introduction One of the most important functions of intracenular phospholipase A , (EC 3.'1,1.4) is undoubtedly its involvement in the synthesis of prostaglandins (PG) through the release of arachidonic acid from the s n - 2 position of membrane phospholipids. This imporlant physiological event takes place in the stomacil where PG play a role in cytoprotection against gastric ulceration [1]. Phospholipase A 2 activity has been previously described in the stomach [2-5] where it is thought to be involved in gastric prostaglandin synthesis via arachidonic acid release. Its specific activ-

Abbreviations: PcJ. Prostaglandins; PC. Phosphatidylcholines; PE, Pho~phafidy1¢thano[aminc~; Sph. Sphinsomyclins; LPC, Lys~-

phosphat[dylcholines; LPE, LysophosphatidyIclhanolamincs; ACLAT. acyl CoA-Lysophosphatidylcboline: acyltransfer~e. Correspondence: G. Nalbonc, lnserm Unite 130, 18 Avenue Mozart. 13009 Mar~eille,France.

ity is one of the highest compared to those measured in other organs [6] which may be related to the high PG turn.over in the gastric mucosa. This is illustrated by the results of Takano et al. [7] who showed a concomitant decrease in phospholipase A z activity and PG levels in stomachs of rats stressed by water immersion. According to the authors, these events were responsible for the higher formation of gastric ulcers in these rats. Problems related to the subcellular localization of the initial step of PG synthesis (i.e., arachidonic acid release by phospholipase A2) have been encountered because PG are not usually stored in the tissue due to their rapid metabolizatlon [81. Also, information related to the cellular or subcellular localization o[ gastric intracellular phospholipase A z are. according to our knowledge, very scarce. Phospholipase A 2 activity was essentially located in the corpus mucosa [3]. Phospholipase A 2 activity was localized by cytoehcmical staining on the plasma membrane of malignant gastric adenocarcinoma cells, but this was not the case in cells beyond the margin of the tumor [9]. Our previous results clearly indicated that on the basis of a classical subcellular fractionation, gastric

131 phospholipase A~ is a 95% membrane-bound enzyme [4]. In the present study we tried to better characterize the subcellular localization of gastric phospholipase A zThis was achieved by a subeellular fractionation followed by gradient nltracentrifugation of the 105000 × g pellet. Some of the biochemical properties of this enzyme are described. Lysophospholipid-metabolizing enzymes were also studied. Materials and Methods

Procedures ]'or samples Male Wistar rats weighing between 200 and 300 g were used. The animals were fasted overnight and killed by cervical fracture. The stomachs were removed, opened, extensively washed with NaCI 0.15 M and incubated in 10 mM K2HPO4 buffer (pH 7.4), NaCI 0.15 M, sucrose 0.25 M containing hyaluronidase 2 m g / m l , for 10 rain at 37°C. in order to disrupt the mucosal coat. After washing with the same buffer, the upper portion of the stomach, the forestomach, was discarded and the glandular part (antrum plus fundus) was gently scraped onto an ice-cold glass plate. The mucosa was weighed and suspended in the homogenization buffer (10 mM Tris-HCl (pH 7.4), sucrose 0.25 M, EDTA 1 raM, 10%, w / v ) and homogenized by using a polytron tissue processor (Kinematica PT lfl, Luzern, Switzerland) for 3 s at a rheostat setting of 6.0. Subsequent centrifugations were performed at 4°C according to the procedure previously described 14]. The 105000 × g (high-speed) pellet was re.suspended in the homogenization buffer and layered on a discontinuous sucrose gradient consisting of 1.95 ml 50~ sucrose (w/v), 3.25 mi 35% sucrose, 2.6 ml of 25% sucrose and 1.3 ml of 10% sucr3se, in 10 mM Tris-HC1 buffer at pH 7.4 according to the method of Tepperman et al. [10]. Centfifugation was performed at 50000 x g for 18 h , 4 ° C using a SW 40 Ti rotor (Beckman, Palo Alto, CA) in a L5 75 B ultracentrifuge (Beckman). The gradient was collected in 1 ml fractions from the top. Linearity of the sucrose gradient was checked gravimetricall~, Protein and marker enzyme assays Proteins were measured by the dye-binding me~hod of Bradford [11] using rabbit "y-globulin as a standard (protein assay kit Bio.Rad, Richmond, CA). The localization of gastric mocosa phospholipase A 2 was assessed via the assay of conventional marker enzymes. Succinate INT reductase [12] ~vas used as a mitochondrial marker enzyme, and rotenone insen!itive NADPH-cytochrome-c reductase [13] (EC 1.6.2.4) for endoplasmic reticulum. For the plasma membrane, 5"nueleotidase (EC 3.1.3.5) was determined [14] by an enzymatic method with a 5"-uacleotidase kit (Sigma No. 265 B) at 300C. For the apical plasma membrane,

alkaline phosphatase [15] (EC 3.1.3.1) was determined ~vith p-nitrophenyl phosphate (Sigma 104 phosphalase substratel by using a Technicon-Smac analyzer (Technicon. U.S.~..) at 37°C. For the basolateral membrane, ouabain sertsiti~e Na+/K*-ATPase (EC 3.6.l.3) activity was assayed as previously described [16].

Phosphobpase A: and (vsophospholipM-metaboli:ing enzyme a~says Gastric phospholipase A,_ was assayed as previously described [4]. Incubation mixtures contained 0.2 mM 1-palmitoyl-2-11 - 14C]oleoylphosphat idylcholine (Amersham. France) with a specific radioactivity of 59.2 K B q / ~ m o l and 50 mM sodium cacodylate at pH 7.0 in a final volume of 0.2 ml. Final calcium concentration in the assay was 5 mM or 10 mM if EDTA was present in the sample. Radioactivity was counted by liquid scintillation spectrometry (Beckman LS 9000) in 10 ml of Readysolve MP (Beckman) scintillation fluid. The resuits were corrected using a control incubated without protein. All assays were done in duplicate. Lysophospholipase (EC 3.1.1.5) and acyl CoA-lysophospbatidylcholine : ac3,1transferase (ACLAT: EC 2.3.1.23) were assayed with [l-t4C]palmitoyl-lysophosphatidylcholine (New England Nuclear, France) as previously described [17,181. Endogenous hydrolysis of gastric membrane phospholipids In some experiments, we investigated the endogenous hydrolysis rate of gastdc membrane phospholipids by gastric phospholipase A 2. To this end, homogenization of the mucosa was performed with a standard buffer containing 10 mM EDTA to prevent any membrane degradation during this process. Then, 0.88 ml of the homogenate (16 to 22 mg of protein) was incubated at 37 ° C after adding CaCI 2 for a final concentration of 20 mM. Incubation was for 1, 2 and 5 min. The reaction was stopped by addition of chloroform/methanol (2:1, v/v) and lipids were extracted according to Folch et al. [191. The lipid extract was evaporated to dryness under nitrogen and taken up with chloroform/methanol (2 : I. v/v). From this lipid extract (40 45 ttg of lipidic phosphorus), phosphatidylcholine, phosphatidylethanolamine and their respective lysoderivatives, and sphingomyelin were separated by HPLC (Water Associate) as previously described [20] and then quantified by phosphorus determination [21]. Results

Subcellular dtstribution and biochemical properties of phospholipase A, in the various sulx'ellular fractions of rat gastric mucosa As Table I shows, a high percemage of the enzymes studied sedimented with the low-speed pellet. Differents procedures were tested to lower these percentages. A

132

~o~

oc,

~ $. . . .

q

'~

i~

~;

+1 -~

+1 +1

+1 +1

~4N~ ~1 ÷1

+1

gZ.

m o r e vigorous h o m o g e n i z a t i o n was successful b u t resuited in a greater recovery of e n z y m e activities in the soluble fraction, certainly due to a significant breakd o w n of organelles ( n o t shown). T h e n we tried to enzymatically disrupt the m u c u s by increasing the t i m e of incubation with hyaluronidase at 3 7 ° C . N o m a r k e d i m p r o v e m e n t of the p a t t c r n of fractionation was obtained. T h e high-speed pellet a p p e a r s to be essentially composed of microsomes a n d plasma m e m b r a n e s as j u d g e d by the activities of their respective m a r k e r enzymes. W e studied s o m e of the biochemical characteristics of gastric p h o s p h o l i p a s e A 2 in the subcellular fractions (Table ll). As T a b l e II shows, all the fractions exhibited similar p H d e p e n d e n c e w i t h a m a x i m u m centered at 7.0 except in the cytosol in which a n o t h e r peak of activity

25

~.

1.2

Proteln

g +1 +1

+1 ÷1

r~,~

No

1.5

{NQ" * K*t A~Pase

~

~e

'2g

÷'÷'

n5

~ 21 3 4 5' 0 7 8° 9 1 011 1

123

4 5 e 7 n 91011

31AlkOline phOSphatose

~

~~

+1 +1

N