Biochimica et Biopt~vsica Acta. 1084(1991) 21-28 a~ 1991 ElsevierScience PublishersB.V.0005-2760/91/$03.50 ADONIS 000527609100180Y
Ca2+-independent phospholipase A 2 activity associated with secretory granular membranes in rat parotid gland Masako Mizuno, Yasunaga Kameyama and Yutaka Yokota Department of Oral Biochemistry, Asal~i Unwersttv School of Dentistry, Glfu (Japanl
(Received 19 February 1991)
Key words: PhospholipaseA2: Secretorygranule: Granularmembrane:Ar~chidonicacid: (Rat parotid gland) Phospholipase A 2 activity was detected in a secret~ry granular fraction (SG) purified by Percull gradient centrifugation from rat parotid gland using [aH]phosphatidylcholine (PC) as a substrate. High aclivity oi this enzyme was observed at neutral pH. The enzyme was activated by Triton X-100 and did not require Ca2~ for its activity, in the absence of Ca 2+, its apparent K m for exogenous PC was 28/~M while it was slightly increased by adding 5 mM CaCI 2 (73/zM). Furthermore, the enzyme was located essentially in a granular membrane fraction separated from granular lysate. The deacylation activities were also detected in other suheellalar fractions, which showed a different detergent-susceptibility or pH-dependency from that in SG. These results suggest that secretory granules have membrane-bound phospholipase A z which has properties different from that found in other organelles.
Introduction Parotid acinar cells are specialized for saliva secretion and strictly regulated by the autonomic nervous system [1]. On stimulation by various secretagogues, the secretory granules rapidly begin to fuse with apical surface membranes of acinar cells to discharge granular contents such as amylase [I]. This event, viz., exocytosis, is the final key step in stimuius-secretion coupling phenomena and, therefore, studies on its regulatory mechanisms are very important to understand the secretory mechanism in parotid gland. Recently, it has been suggested that exocytosis is associated with the changes in membrane phospholipid compositions [2]. Snowdowne et al. [31 reported that an addition of arachidonic acid, its metabolites by cyclooxygenase and lipoxygenase pathway, or melinin, a potent phospholipase A 2 activator, into isolated parotid cells stimulated amylase release. These effects have also
been obser, ed in other secretory cells such as mast cells [4], pancreatic islets [5] and anterior pituitary prolactin cells [61. Kurihara et al. [7] demonstrated morphologically that multigranular exocytosis was induced by melittin or mastoparan. Creutz [8] reported that arachidonic acid induced fusion between isolated chromaffin granules aggregated by synexin. These observations suggest that the liberation of arachidonic acid is involved in secretory processes, particularly exocytosis. However, there is little knowledge about the arachidonic acid-releasing pathway in parotid gland. Recently, we have demonstrated that secretory granules contain a high level of lysophospholipids [9,10], which may be released together with free fatty acid by phospholipase A action. These data suggest the possible involvement of phospholipase A in the metabolism of granular membrane lipids. In the present study, we demonstrate the presence of phospholipase A z activity, a candidate of the arachidonic acid-releasing enzyme, in secretory granules and characterize its properties to investigate about the possibility of arachidonic acidliberation at the fusion site.
Abbreviations: PC. phosphatidylcholine;SG. purified secretorygranular fraction; SG-M, secretory granular membrane fraction: SG-S, secretory granular soluble fraction: K~-NPPase, K+-dependent pnitropbenyl phospbatase;ER, endoplasmicrcticulum; FFA, free fatty acid, l-acyI-GPC, 1-acyt-sn-glycero-3-phosphocholine.
Materials and Methods
Correspondence: Y. Yokota, Departmentof Oral Biochemistry,Asahi University School of Dentistry. 1851 Hozumi.Motosu, Gifu 501-02, Japan.
Male Wistar rats (10-12 weeks old) were maintained ad libitum on Oriental MF solid chow (Oriental Yeast Co., Tokyo) and water. After fasting overnight, all
Materials
animals were killed by bleeding under light diethyl ether anesthesia. Immediately, the parotid gland was isolated and connective tissue trimmed off. [~H]Phosphatidylcholine (L-a-dipalmitoyl, [2-palmitoyl-9,10-~H(N)] (1.85 TBq/mmol)) and [t4C]phosphatidylcholine (L-a-dipalmitoyl, [choline-methyl-~ac] (5.7 GBq/mmol)) were purchased from Du Pont/New England Nuclear Research Products (Boston, MA). They were diluted with unlabeled egg yolk phosphatidylcholine (Sigma, St. Louis, MO) to give an appropriate specific radioactivity for phospholipase assay (7000080 000 dpm/nmol) and then purified by thin-layer chromatography with c h l o r o f o r m / m e t h a n o l / w a t e r (65 : 25 : 4, v/v). The radiochemical purity of the labeled phosphatidylcholine (PC) was more than 99% verified by thin-layer chromatography. The positional distribution of [3H]palmitoyl moiety in phosphatidylcholine was determined by enzymatic hydrolysis with phospholipase A z from Naja naja venom (Sigma). it was confirmed that more than 99.5% of [3H]palmitic acid was located at sn-2 position of phosphatidylcholine. Pcrcoll was purchased from Pharmacia (Uppsala, Sweden). The various substrates used in the assay for marker enzymes were of the highest purity available from Sigma.
Subcellular fractionation of parotid gland All procedures were carried out at 0 - 4 ° C . The minced parotid glands were homogenized in ice-cold 2 mM sodium 4-morpholinepropanesulphonate buffer (Mops-NaOH, pH 7.0) containing 0.3 M sucrose and 0.2 mM MgCI 2 with a Potter-Elvehjem Teflon homogenizer (four to five strokes at 1300-1600 rpm) to give a 15% (w/v) tissue homogenate. The homogenate was centrifuged at 70 × g for 30 s to remove unbroken cells and the supernatant was further diluted to 7.5% (w/v) with the same buffer. After removal of nuclei (Pt) by centrifugation at 940 × g for 10 min, EDTA was added to the resulting supernatant to give a final concentration of 1.2 raM. A crude secretory granular fraction (P2) was obtained from this supernatant by centrifugation at 1900 × g for 10 min. In order to separate secretory granules from other organelles such as mitochondria, 1"2 was further fractionated by Percoll gradient centrifugation as essentially described by Arvan and Castle [11]; briefly, Pz was resuspended in Percoll buffer A (9 mM Mops-NaOH (pH 7.3), containing 57% Percoll, 0,3 M sucrose, 0.07 mM MgCI 2 and 1.1 mM EDTA), loaded into polycarbonate centrifuge tubes containing a 2 ml cushion of Percoll buffer B (12 mM Mops-NaOH (pH 7.3), containing 86% Percoll, 0.3 M sucrose and 1 mM EDTA) and centrifuged in a Hitachi RP 70T rotor at 14600 rpm for 30 min to obtain two major bands, The secretory granule rich fraction was located in the lower quarter of the self-formed gradients, and the other (P3), which consisted of the other organelles, was located
near the top. In order to purify further, the granular fraction was resuspended in 3 vol, of Percoll buffer A and recentrifuged in the same rotor at 24700 rpm for 30 min. The white granular fraction was collected and washed twice with buffer C (20 mM Mops-NaOH (pH 7.0), containing 0.3 M sucrose). The purified secretory granular fraction (SG) and P~were resuspended in buffer C and stored at - 8 0 ° C . Furthermore, the 1 9 0 0 × g supernatant described above was also fractionated by differential centrifugations at 10000 × g, 20000 × g for 20 min and 105000 x g for 60 rain and the resulting precipitates were named Pa, P6 and P7, respectively. A lysosome rich fraction ( ~ ) was prepared from 10000 × g precipitate (P4), by Percoll gradient eentrifugation as essentially described by Pertoft and W~irmeg~trd [12]. All precipitates obtained from these centrifugations were washed, resuspended in buffer C and stored at -80°C.
Preparation for secretory granular membranes The purified granules were lysed by freezing and thawing (three repetitions) in a hypotonic buffer (2 mM Mops-NaOH, pH 7.0). Then a granular membrane fraction (SG-M) was collected by centrifugation at 150000 × g for 60 min. The supernatant was saved as a secretory granular soluble fraction (SG-S) and SG-M was washed with 2 mM Mops-NaOH (pH 7.0) ar, d resuspended in the buffer C. They were stored at -80°C.
Phospholipase assay The standard incubation mixture for a phospholipase assay contained, in 0.1 ml: 220/~M [3H]phosphatidyicholine, 2 m g / m l enzyme protein, 10 mM EDTA, 0.05% Triton X-100 and 100 mM Mops-NaOH (pH 7.0), Labeled substrate was added as a iiposomal suspension (2.2 p,mol/ml) which was prepared by vigorously mixing with small glass beads (0.17-0.18 mmo, B. Braun, Melsungen) for more than a total of 3 min at room temperature. The reaction was initiated by the addition of enzyme protein. Incubation was carried out at 37°C for 60 min and terminated by adding 1.85 ml of chlorof o r m / m e t h a n o l / w a t e r (1 : 2 : 0.7, v/v). The lipids were extracted according to the method of Bligh and Dyer [13]. They were separated on a Silica Gel 60 plate (Merck, Darmstadt) with chloroform/methanol/acetic acid/water ( 2 5 : 1 5 : 4 : 2 , v/v, for polar lipids) as a solvent system, or on a 0.4 M boric acid impregnated Silica Gel 60 plate with chloroform/acetone (9 : 1, v/v, for neutral lipids). Lipid spots were visualized by the brief exposure of the plates to iodine vapors and identified by comparison with standards run in parallel on the same plate. The radioactivity of each area was measured in an Aloka LSC-900 liquid scintillation counter as described previously [14].
When a mixture of [3HI and [Z'*C]phosphatidylcholines was used as a substrate, radioactivity of the aqueous phase after lipid extraction was also measured in order to investigate the formation of water soluble products. The values for nonenzymatic hydrolysis were subtracted to give net hydrolysis. The recovery of labeled compounds through this assay is usually 85-95%. Biochemical analysis Amylase was used as a marker enzyme of secretory granules and granular lysate, its activity was assayed with a commercially obtained assay kit using blue starch polymer (Pharmacia Diagnostics, Sweden) based on the method of Ceska et al. [15]. The following enzymes were used as markers of the organelles and assayed as described in the references: fl-N-acetyl glucosaminidase (lysosomes) [16]; K+-dependent p-nitrophenyl phosphatase (plasma membranes) [17]; N A D P H - c y t o c h r o m e c reduetase (endoplasmic reticulum) [181 and succinate dehydrogenase (mitochondria) [19]. Protein concentration was determined by the method of Lowry et al. [201 using bov'.'ne serum albumin as a standard. Phospholipid phosphorus was measured by the method of Bartlett [21] as m.odified by Marinetti [22]. Results
Distribution o f marker enzymes in subcellular fractions Table I shows the distribution of marker enzymes in the various subcellular fractions. The highest specific activity for amylase, a marker enzyme for parotid secretory granules [23], was obtained in SG. On the other hand, the specific activity for K+-dependent pnitrophenyl phosphatase (K+-NPPase) in SG (0.03
n m o l / m i n per mg otein) was the lowest a m o n g all subcellular fractions, and it was only 0.05% of the highest one shown in P7 (54.8 n m o l / m i n per mg protein). Therefore, the contamination of plasma membranes into SG is estimated at less than 0.05%. When the same calculation was performed with respect to succinate dehydrogenase, /~-N-acetylglucosaminidase and N A D P H - c y t o c h r o m e c reductase, the maximal possible contamination with mitochondria, lysosomes and endoplasmic reticulum (ER) into SG was judged to be less than 0.9%, 3.5% and 0.9%, respectively. Judging from the marker enzyme distribution, the fractions richest in mitochondria, lysosomes, E R and plasma membranes were the upper band of Percoll gradient (PO, the lysosome rich fraction (Ps), the 2 0 0 0 0 × g precipitate (P6) and 105000 x g precipitate (PT). respectively. Ht'dro(vsis of exogenou.v phosphatidylchofine by SG under t'arious conditions Using [3H]phosphatidylcholine (PC) labeled at sn-2 position as a substrate, significant formation of [3H]free fatty acid (FFA) was detected in SG under various conditions shown in Table 11 although most (more than 99%) of the radioactivity was recovered in PC fraction (data not shown). The formation of [3HIFFA was remarkably activated by addition of 0,05% Triton X-100. On the other hand, addition of E D T A was !ess effective on this reaction either in the presence or absence of the detergent although the removal of C a : + actually decreased the deacylation activity of rat liver mitochondria (used as a Ca2+-dependent phospholipase) by about 70%. Thus, the FFA-releasing pathway in SG does not require Ca '-+. From these data, it is estim,qted
TABLE I
Typical distributionof marker enzyme activities i n t.ariotL~"subcellularfractionsfrom rut parotid gland The assay method for each marker enzyme is described in Materials and Methods. Values are the mean of three to seven determinations-+S.E.
Marker enzyme activity (nmol/min per mg protein) amylage Total homogenate 940× g, ppt b (PO 1900x g, ppt (P2) purified secretory granular fraction (SG) upper band of Percoll gradient (P3) lO000×g, ppt(P4) lysosome rich fraction (Ps) 20000× g, ppt (P¢,) 105000X g, ppt (Pv)
~-N-acetyl glucosaminidase
cytochrome c reductase 1.28_+0.13 0.48_+0.09 0.78_+0.15 0.05_+0.05
K+-NPPase a
19.0_+4.6 16.5_+0.4 60.3_+8.9 1.6+0.9
12.4_+ 0.8 10.04- 0.6 22.5 _+ 2.3 8.0_+ 0.8
2570_+360
185 0+9.2
20.4_+ 3.0
1.69-+0A4
14.0_+1.1
1700_+ 90 7400-+ 820
117.5_+5.4 < O.I
52.5_+ 2.7 227.4_+44.3
2.85_+0.20 < O.Ol
24.7_+1.3 O.I 4-O.1
560-+ 70 320_+ 40
24.8-+ 2.4 25.1 -+2.6
24.3 -+ 1.7 17.0_+ 1.1
5.68-+0.41 5.10_+0.35
41.9-+ 2.8 54.8-+ 2.5
K+-dependent p-nitrophenyl phosphatas¢. b Precipitate.
succinate dehydrogenase
3970_+650 7870_+480 5660_+350 10240+560
13.1 _+0.7 11.0+0.8 8.2-+ 1.0
Ca2+-independent phospholipase A 2 activity associated with secretory granular membranes in rat parotid gland Masako Mizuno, Yasunaga Kameyama and Yutaka Yokota Department of Oral Biochemistry, Asal~i Unwersttv School of Dentistry, Glfu (Japanl
(Received 19 February 1991)
Key words: PhospholipaseA2: Secretorygranule: Granularmembrane:Ar~chidonicacid: (Rat parotid gland) Phospholipase A 2 activity was detected in a secret~ry granular fraction (SG) purified by Percull gradient centrifugation from rat parotid gland using [aH]phosphatidylcholine (PC) as a substrate. High aclivity oi this enzyme was observed at neutral pH. The enzyme was activated by Triton X-100 and did not require Ca2~ for its activity, in the absence of Ca 2+, its apparent K m for exogenous PC was 28/~M while it was slightly increased by adding 5 mM CaCI 2 (73/zM). Furthermore, the enzyme was located essentially in a granular membrane fraction separated from granular lysate. The deacylation activities were also detected in other suheellalar fractions, which showed a different detergent-susceptibility or pH-dependency from that in SG. These results suggest that secretory granules have membrane-bound phospholipase A z which has properties different from that found in other organelles.
Introduction Parotid acinar cells are specialized for saliva secretion and strictly regulated by the autonomic nervous system [1]. On stimulation by various secretagogues, the secretory granules rapidly begin to fuse with apical surface membranes of acinar cells to discharge granular contents such as amylase [I]. This event, viz., exocytosis, is the final key step in stimuius-secretion coupling phenomena and, therefore, studies on its regulatory mechanisms are very important to understand the secretory mechanism in parotid gland. Recently, it has been suggested that exocytosis is associated with the changes in membrane phospholipid compositions [2]. Snowdowne et al. [31 reported that an addition of arachidonic acid, its metabolites by cyclooxygenase and lipoxygenase pathway, or melinin, a potent phospholipase A 2 activator, into isolated parotid cells stimulated amylase release. These effects have also
been obser, ed in other secretory cells such as mast cells [4], pancreatic islets [5] and anterior pituitary prolactin cells [61. Kurihara et al. [7] demonstrated morphologically that multigranular exocytosis was induced by melittin or mastoparan. Creutz [8] reported that arachidonic acid induced fusion between isolated chromaffin granules aggregated by synexin. These observations suggest that the liberation of arachidonic acid is involved in secretory processes, particularly exocytosis. However, there is little knowledge about the arachidonic acid-releasing pathway in parotid gland. Recently, we have demonstrated that secretory granules contain a high level of lysophospholipids [9,10], which may be released together with free fatty acid by phospholipase A action. These data suggest the possible involvement of phospholipase A in the metabolism of granular membrane lipids. In the present study, we demonstrate the presence of phospholipase A z activity, a candidate of the arachidonic acid-releasing enzyme, in secretory granules and characterize its properties to investigate about the possibility of arachidonic acidliberation at the fusion site.
Abbreviations: PC. phosphatidylcholine;SG. purified secretorygranular fraction; SG-M, secretory granular membrane fraction: SG-S, secretory granular soluble fraction: K~-NPPase, K+-dependent pnitropbenyl phospbatase;ER, endoplasmicrcticulum; FFA, free fatty acid, l-acyI-GPC, 1-acyt-sn-glycero-3-phosphocholine.
Materials and Methods
Correspondence: Y. Yokota, Departmentof Oral Biochemistry,Asahi University School of Dentistry. 1851 Hozumi.Motosu, Gifu 501-02, Japan.
Male Wistar rats (10-12 weeks old) were maintained ad libitum on Oriental MF solid chow (Oriental Yeast Co., Tokyo) and water. After fasting overnight, all
Materials
animals were killed by bleeding under light diethyl ether anesthesia. Immediately, the parotid gland was isolated and connective tissue trimmed off. [~H]Phosphatidylcholine (L-a-dipalmitoyl, [2-palmitoyl-9,10-~H(N)] (1.85 TBq/mmol)) and [t4C]phosphatidylcholine (L-a-dipalmitoyl, [choline-methyl-~ac] (5.7 GBq/mmol)) were purchased from Du Pont/New England Nuclear Research Products (Boston, MA). They were diluted with unlabeled egg yolk phosphatidylcholine (Sigma, St. Louis, MO) to give an appropriate specific radioactivity for phospholipase assay (7000080 000 dpm/nmol) and then purified by thin-layer chromatography with c h l o r o f o r m / m e t h a n o l / w a t e r (65 : 25 : 4, v/v). The radiochemical purity of the labeled phosphatidylcholine (PC) was more than 99% verified by thin-layer chromatography. The positional distribution of [3H]palmitoyl moiety in phosphatidylcholine was determined by enzymatic hydrolysis with phospholipase A z from Naja naja venom (Sigma). it was confirmed that more than 99.5% of [3H]palmitic acid was located at sn-2 position of phosphatidylcholine. Pcrcoll was purchased from Pharmacia (Uppsala, Sweden). The various substrates used in the assay for marker enzymes were of the highest purity available from Sigma.
Subcellular fractionation of parotid gland All procedures were carried out at 0 - 4 ° C . The minced parotid glands were homogenized in ice-cold 2 mM sodium 4-morpholinepropanesulphonate buffer (Mops-NaOH, pH 7.0) containing 0.3 M sucrose and 0.2 mM MgCI 2 with a Potter-Elvehjem Teflon homogenizer (four to five strokes at 1300-1600 rpm) to give a 15% (w/v) tissue homogenate. The homogenate was centrifuged at 70 × g for 30 s to remove unbroken cells and the supernatant was further diluted to 7.5% (w/v) with the same buffer. After removal of nuclei (Pt) by centrifugation at 940 × g for 10 min, EDTA was added to the resulting supernatant to give a final concentration of 1.2 raM. A crude secretory granular fraction (P2) was obtained from this supernatant by centrifugation at 1900 × g for 10 min. In order to separate secretory granules from other organelles such as mitochondria, 1"2 was further fractionated by Percoll gradient centrifugation as essentially described by Arvan and Castle [11]; briefly, Pz was resuspended in Percoll buffer A (9 mM Mops-NaOH (pH 7.3), containing 57% Percoll, 0,3 M sucrose, 0.07 mM MgCI 2 and 1.1 mM EDTA), loaded into polycarbonate centrifuge tubes containing a 2 ml cushion of Percoll buffer B (12 mM Mops-NaOH (pH 7.3), containing 86% Percoll, 0.3 M sucrose and 1 mM EDTA) and centrifuged in a Hitachi RP 70T rotor at 14600 rpm for 30 min to obtain two major bands, The secretory granule rich fraction was located in the lower quarter of the self-formed gradients, and the other (P3), which consisted of the other organelles, was located
near the top. In order to purify further, the granular fraction was resuspended in 3 vol, of Percoll buffer A and recentrifuged in the same rotor at 24700 rpm for 30 min. The white granular fraction was collected and washed twice with buffer C (20 mM Mops-NaOH (pH 7.0), containing 0.3 M sucrose). The purified secretory granular fraction (SG) and P~were resuspended in buffer C and stored at - 8 0 ° C . Furthermore, the 1 9 0 0 × g supernatant described above was also fractionated by differential centrifugations at 10000 × g, 20000 × g for 20 min and 105000 x g for 60 rain and the resulting precipitates were named Pa, P6 and P7, respectively. A lysosome rich fraction ( ~ ) was prepared from 10000 × g precipitate (P4), by Percoll gradient eentrifugation as essentially described by Pertoft and W~irmeg~trd [12]. All precipitates obtained from these centrifugations were washed, resuspended in buffer C and stored at -80°C.
Preparation for secretory granular membranes The purified granules were lysed by freezing and thawing (three repetitions) in a hypotonic buffer (2 mM Mops-NaOH, pH 7.0). Then a granular membrane fraction (SG-M) was collected by centrifugation at 150000 × g for 60 min. The supernatant was saved as a secretory granular soluble fraction (SG-S) and SG-M was washed with 2 mM Mops-NaOH (pH 7.0) ar, d resuspended in the buffer C. They were stored at -80°C.
Phospholipase assay The standard incubation mixture for a phospholipase assay contained, in 0.1 ml: 220/~M [3H]phosphatidyicholine, 2 m g / m l enzyme protein, 10 mM EDTA, 0.05% Triton X-100 and 100 mM Mops-NaOH (pH 7.0), Labeled substrate was added as a iiposomal suspension (2.2 p,mol/ml) which was prepared by vigorously mixing with small glass beads (0.17-0.18 mmo, B. Braun, Melsungen) for more than a total of 3 min at room temperature. The reaction was initiated by the addition of enzyme protein. Incubation was carried out at 37°C for 60 min and terminated by adding 1.85 ml of chlorof o r m / m e t h a n o l / w a t e r (1 : 2 : 0.7, v/v). The lipids were extracted according to the method of Bligh and Dyer [13]. They were separated on a Silica Gel 60 plate (Merck, Darmstadt) with chloroform/methanol/acetic acid/water ( 2 5 : 1 5 : 4 : 2 , v/v, for polar lipids) as a solvent system, or on a 0.4 M boric acid impregnated Silica Gel 60 plate with chloroform/acetone (9 : 1, v/v, for neutral lipids). Lipid spots were visualized by the brief exposure of the plates to iodine vapors and identified by comparison with standards run in parallel on the same plate. The radioactivity of each area was measured in an Aloka LSC-900 liquid scintillation counter as described previously [14].
When a mixture of [3HI and [Z'*C]phosphatidylcholines was used as a substrate, radioactivity of the aqueous phase after lipid extraction was also measured in order to investigate the formation of water soluble products. The values for nonenzymatic hydrolysis were subtracted to give net hydrolysis. The recovery of labeled compounds through this assay is usually 85-95%. Biochemical analysis Amylase was used as a marker enzyme of secretory granules and granular lysate, its activity was assayed with a commercially obtained assay kit using blue starch polymer (Pharmacia Diagnostics, Sweden) based on the method of Ceska et al. [15]. The following enzymes were used as markers of the organelles and assayed as described in the references: fl-N-acetyl glucosaminidase (lysosomes) [16]; K+-dependent p-nitrophenyl phosphatase (plasma membranes) [17]; N A D P H - c y t o c h r o m e c reduetase (endoplasmic reticulum) [181 and succinate dehydrogenase (mitochondria) [19]. Protein concentration was determined by the method of Lowry et al. [201 using bov'.'ne serum albumin as a standard. Phospholipid phosphorus was measured by the method of Bartlett [21] as m.odified by Marinetti [22]. Results
Distribution o f marker enzymes in subcellular fractions Table I shows the distribution of marker enzymes in the various subcellular fractions. The highest specific activity for amylase, a marker enzyme for parotid secretory granules [23], was obtained in SG. On the other hand, the specific activity for K+-dependent pnitrophenyl phosphatase (K+-NPPase) in SG (0.03
n m o l / m i n per mg otein) was the lowest a m o n g all subcellular fractions, and it was only 0.05% of the highest one shown in P7 (54.8 n m o l / m i n per mg protein). Therefore, the contamination of plasma membranes into SG is estimated at less than 0.05%. When the same calculation was performed with respect to succinate dehydrogenase, /~-N-acetylglucosaminidase and N A D P H - c y t o c h r o m e c reductase, the maximal possible contamination with mitochondria, lysosomes and endoplasmic reticulum (ER) into SG was judged to be less than 0.9%, 3.5% and 0.9%, respectively. Judging from the marker enzyme distribution, the fractions richest in mitochondria, lysosomes, E R and plasma membranes were the upper band of Percoll gradient (PO, the lysosome rich fraction (Ps), the 2 0 0 0 0 × g precipitate (P6) and 105000 x g precipitate (PT). respectively. Ht'dro(vsis of exogenou.v phosphatidylchofine by SG under t'arious conditions Using [3H]phosphatidylcholine (PC) labeled at sn-2 position as a substrate, significant formation of [3H]free fatty acid (FFA) was detected in SG under various conditions shown in Table 11 although most (more than 99%) of the radioactivity was recovered in PC fraction (data not shown). The formation of [3HIFFA was remarkably activated by addition of 0,05% Triton X-100. On the other hand, addition of E D T A was !ess effective on this reaction either in the presence or absence of the detergent although the removal of C a : + actually decreased the deacylation activity of rat liver mitochondria (used as a Ca2+-dependent phospholipase) by about 70%. Thus, the FFA-releasing pathway in SG does not require Ca '-+. From these data, it is estim,qted
TABLE I
Typical distributionof marker enzyme activities i n t.ariotL~"subcellularfractionsfrom rut parotid gland The assay method for each marker enzyme is described in Materials and Methods. Values are the mean of three to seven determinations-+S.E.
Marker enzyme activity (nmol/min per mg protein) amylage Total homogenate 940× g, ppt b (PO 1900x g, ppt (P2) purified secretory granular fraction (SG) upper band of Percoll gradient (P3) lO000×g, ppt(P4) lysosome rich fraction (Ps) 20000× g, ppt (P¢,) 105000X g, ppt (Pv)
~-N-acetyl glucosaminidase
cytochrome c reductase 1.28_+0.13 0.48_+0.09 0.78_+0.15 0.05_+0.05
K+-NPPase a
19.0_+4.6 16.5_+0.4 60.3_+8.9 1.6+0.9
12.4_+ 0.8 10.04- 0.6 22.5 _+ 2.3 8.0_+ 0.8
2570_+360
185 0+9.2
20.4_+ 3.0
1.69-+0A4
14.0_+1.1
1700_+ 90 7400-+ 820
117.5_+5.4 < O.I
52.5_+ 2.7 227.4_+44.3
2.85_+0.20 < O.Ol
24.7_+1.3 O.I 4-O.1
560-+ 70 320_+ 40
24.8-+ 2.4 25.1 -+2.6
24.3 -+ 1.7 17.0_+ 1.1
5.68-+0.41 5.10_+0.35
41.9-+ 2.8 54.8-+ 2.5
K+-dependent p-nitrophenyl phosphatas¢. b Precipitate.
succinate dehydrogenase
3970_+650 7870_+480 5660_+350 10240+560
13.1 _+0.7 11.0+0.8 8.2-+ 1.0
