0 1992 by The American Society for Biochemistry and Molecular Biology, Inc. THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 267, No. 21, Issue of July 25, pp. 15152-15159,1992

Printed in U.S.A.

Protein Degradation by the Phosphoinositide-specific Phospholipase C-cu Family from Rat Liver Endoplasmic Reticulum*

(Received for publication, February 18,1992)

Reiko Urade, Masayuki Nasu, Tatsuya Moriyama, Kazuteru Wada, and Makoto Kit04 From the Research Institute for Food Science, Kyoto University, Uji, Kyoto 611, Japan

A 60-kDa protein homologous to phosphoinositide- specific phospholipase C - a was purified to apparent homogeneity on sodium dodecyl sulfate-polyacryl- amide gel electrophoresis from the rough endoplasmic reticulum of rat liver through three sequential chro- matographies on DEAE Toyopearl 650, AF-heparin Toyopearl 650M, and TSK gel G3000SW. The purified protein was monomeric, with an M, of 60,000. Eight types of protein were further separated from the 60- kDa protein and named ERGOA-ERGOH according to the order of their elution from a TSK gel DEAE-5PW column. They were essentially identical in terms of immunochemical properties and the NH2-terminal amino acid sequence. The partial amino acid sequence of ERBOF showed homology to that of phosphoinosi- tide-specific phospholipase C - a . ERGOA-ERGOH showed no phosphoinositide-specific phospholipase C activity. However, ERGOA-ERGOH catalyzed cleavage of themselves and the endoplasmic reticulum proteins protein disulfide-isomerase and calreticulin. Proteo- lytic degradation was inhibited by p-chloromercuri- benzoate. These results indicate that ERGOA-ERGOH comprise a group of endoplasmic reticulum resident proteins and show thiol group-related proteolytic ac- tivity.

The lumen of the endoplasmic reticulum (ER)’ contains abundant soluble proteins called reticuloplasmins (1). These are ER resident proteins that are distinguished from newly synthesized secretory proteins by the presence of a COOH- terminal tetrapeptide known as the ER retention signal (2) and play important physiological roles in the ER. Immuno- globulin heavy chain-binding protein, a homologue of the 70- kDa heat shock protein, functions as a molecular chaperon that is involved in the folding and assembly of nascent poly- peptides into oligomers (3). Protein disulfide-isomerase stim- ulates the correct inner- and interdisulfide bond formations in nascent polypeptides (4, 5). Calreticulin, a Ca2+-binding protein, is another abundant protein in the ER that is known

* This work was supported by a grant-in-aid for scientific research from the Ministry of Education, Science, and Culture of Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ To whom correspondence should be addressed. The abbreviations used are: ER, endoplasmic reticulum; PIP2,

phosphatidylinositol 4,5-bisphosphate; PVDF, polyvinylidene difluo- ride; PMSF, phenylmethylsulfonyl fluoride; his-Tris, bis(2-hydro- xyethyl)iminotris(hydroxymethyl)methane; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; HPLC, high performance liquid chromatography; pCMB, p-chloromercuriben- zoate; HEPES, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid; PIPES, piperazine-N,N’-bis(2-ethanesulfonic acid).

as a calcium-storage organelle (6-8). Carboxylesterases are also reticuloplasmins that hydrolyze a variety of xenobiotics and endogenous substrates like mono- and diacylglycerols (9- 11).

Phosphoinositide-specific phospholipase C-a was described as a cytosolic protein by Bennett and Crooke (12). However, this enzyme is assumed to be an ER resident protein (13) since the phosphoinositide-specific phospholipase C-a of rat basophilic leukemia cells reported by Bennett et al. (14) has an NHz-terminal hydrophobic sequence that is post-transla- tionally processed and a COOH-terminal tetrapeptide that can act as an ER retention signal (15). Recently, a protein homologous to phosphoinositide-specific phospholipase C-a was found to be localized in the ER of rat liver as a reticulo- plasmin (16). Srivastava et al. (17) reported that a protein that has a partial amino acid sequence identical to that of phosphoinositide-specific phospholipase C-a was purified from rat liver microsomes as a new isozyme of thio1:protein- disulfide oxidoreductase.

In this paper, we report the purification of eight types of protein from the rough ER of rat liver that are regarded as members of the phosphoinositide-specific phospholipase C-a family on the basis of their immunological characteristics and NHz-terminal amino acid sequences. They did not show any phosphoinositide-specific phospholipase C activity, but showed proteolytic activity that was inhibited by p-chlorom- ercuribenzoate (pCMB).

EXPERIMENTAL PROCEDURES

Materials [in0sitol-2-~H]Phosphatidylinositol 4,5-bisphosphate (110.0 GBs/

mmol) was purchased from Du Pont-New England Nuclear. Phos- phatidylinositol 4,5-bisphosphate (PIP,) was from Sigma. Bacillus cereus phospholipase C (grade I, 2000 units/mg) and casein were purchased from Boehringer Mannheim. Bovine serum albumin was from Miles Laboratories Inc. A DEAE Toyopearl 650 prepacked column, an AF-heparin Toyopearl 650M column, a TSK gel G3000SW high performance liquid chromatography (HPLC) column, and a TSK gel DEAE-5PW HPLC column were obtained from TOSOH (Tokyo). A PBondasphere CIS HPLC column was purchased from Waters. Polyvinylidene difluoride (PVDF) protein sequencing membranes and Bio-Lyte 3/10 were obtained from Bio-Rad. Lysylen- dopeptidase (2 units/mg) was purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Male Japanese white rabbits were obtained from Kitayama LABES Co., Ltd. (Kyoto, Japan). A Proto Blot@ immunoblotting system was purchased from Promega Biotec. All other chemicals were reagent-grade.

Preparation of Stripped ER Sixty-four male Sprague-Dawley rats (8-weeks old) were killed by

decapitation under anesthesia. If not specified, all procedures were carried out a t 4 “C. The livers were removed and homogenized in 1.8 liters of 20 mM Tris-HC1, pH 7.4, containing 3 mM MgC12, 0.5 mM phenylmethylsulfonyl fluoride (PMSF), and 0.29 M sucrose with a Potter-Elvehjem homogenizer, with eight strokes at 1200 rpm/min.

15152

ER Protein Degradation 15153

The homogenate was centrifuged for 10 min at 700 X P and then for Amino Acid Sequencing 10 min at-7000 X g. The suiernatant obtained was centrifuged at NH2-terminal Amino Acid Sequen~ing-~roteins were separated

the cytosolic fraction. The pellet was resuspended in 2.4 liters of 20 corresponding to 5-10 pg of protein was cut out and applied to a

0.25 M sucrose. The rough ER was isolated from the suspension by sedimentation for 50 min at 200,000 X g through a sucrose cushion ' 2 ~ ' ~ ~ ~ ~ ~~~~~~$ $.internal Peptide Fragments"One-

between the sample layer and the 1.5 M sucrose cushion was collected solution and then dialyzed twice overnight against liter of water.

to the method of Kreibich et al. (18). The stripped ER obtained was of mM Tris-HC1, pH 8.0, for 48 at 37 oc. The reaction mixture

2~~~~~~ x g for 70 min. The resultant supernatant was removed as by SDS-pAGE and blotted onto a pVDF membrane. A single band

mM Tris-HC1, pH 7.41 containing mM MgC127 0'5 mM PMSFy and Protein Sequencer (Model 477A) equippedwith on-line HPLC (Model

(1.5 sucr'Ose~ mM M&lZ, 0.5 mM PMSF). The interface layer hundred micrograms of purified protein was boiled in a 2% SDS

8s the crude smooth ER. The removal of ribosomes from the rough The dialysate was dried in a centrifugal concentrator. The dried ER Was carried out with 1 mM Puromycin and 0.5 M KC1 according protein was digested with 8 milliunits of lysylendopeptidase in 100 pl

resuspended in lo mM Tris-HC1, pH 8.0, containing 0.5 mM PMSF was diluted with 0.13% trifluoroacetic acid to 0.4 ml and then applied (5 mg Of proteidm1). Five grams Of protein from the stripped ER was to a pBondasphere CI8 HPLC column (0.39 x 15 cm). Peptides were obtained. eluted with a linear madient of 0-60% acetonitrile in 0.1% trifluoro-

Purification of Phosphoinositide-specific Phospholipase C-a Family

One liter of the stripped ER suspension was mixed with 4000 units of B. cerew phospholipase C, followed by incubation for 10 min at 37 "C. The suspension was centrifuged for 1 h at 200,000 X g. The reticuloplasmins released into the supernatant were precipitated by the addition of solid ammonium sulfate (516 g/liter) and stirring for 30 min. The mixture was then centrifuged for 30 min at 13,000 X g. The resulting pellet was dissolved in 50 ml of 20 mM Tris-HC1, pH 7.4, containing 0.2 mM EDTA, 0.5 mM PMSF, and 10% glycerol (buffer A) and then dialyzed twice overnight against 2 liters of buffer A.

The dialysate (660 mg of protein) was applied to a DEAE Toyopearl 650 column (2.2 X 20 cm) equilibrated with buffer A. The column was washed with 100 ml of buffer A and then eluted with a linear gradient of 0-0.25 M NaCl in buffer A at a flow rate of 1 ml/min for 200 min. The fractions were analyzed by sodium dodecyl sulfate- polyacrylamide gel electrophoresis (SDS-PAGE), and the peak frac- tions containing the 60-kDa protein of which the NH2-terminal amino acid sequence was determined to be homologous to that of phosphoi- nositide-specific phospholipase C-a (14) were pooled and precipitated with ammonium sulfate as described above. The precipitate was dissolved in 7.5 ml of 20 mM HEPES/NaOH, pH 6.8, containing 50 mM KCl, 0.5 mM PMSF, and 10% glycerol (buffer B) and then dialyzed overnight against 1 liter of buffer B. The dialysate (31 mg of protein) was applied to an AF-heparin Toyopearl 650M column (1 X 5 cm) equilibrated with buffer B. The column was washed with buffer B until the absorbance reached the base line and then eluted stepwise with buffer B containing 175 and 400 mM KC1 at a flow rate of 0.5 ml/min. The phosphoinositide-specific phospholipase C-a family was eluted from the resin with 400 mM KC1. The eluted fraction (5 ml) was concentrated to 0.8 ml with a Centricon-10 (Amicon Corp.). The concentrated solution (10 mg of protein) was applied to a TSK gel G3000SW column equilibrated with 20 mM Tris-HC1, pH 7.4, con- taining 0.15 M NaCl, 0.2 mM EDTA, 0.5 mM PMSF, and 10% glycerol. Chromatography was carried out at a flow rate of 0.4 ml/min. The peak fractions of the phosphoinositide-specific phospholipase C-a family were collected and dialyzed overnight against 2 liters of 10 mM bis-Tris/HCl, pH 7.0, containing 0.2 mM EDTA, 0.5 mM PMSF, and 10% glycerol (buffer C). The dialysate (6 mg of protein) was applied to a TSK gel DEAE-5PW HPLC column (0.75 X 7.5 cm) equilibrated with buffer C. The column was washed with 24 ml of buffer C containing 0.05 M NaCl and then eluted with a 60-ml linear gradient of 0.05-0.1 M NaCl in buffer C at a flow rate of 0.4 ml/min.

PIP2 Hydrolytic Activity Assay PIP2 hydrolytic activity was assayed as described by Moriyama et

al. (19). Briefly, -7000 cpm of [3H]PIP2 (specific activity, 1400 cpm/ nmol) was incubated for 30 min at 37 "C with 40 mM PIPES/HCl, pH 7.0, containing 100 p~ CaCl,, 0.1% sodium deoxycholic acid, and the sample in a final volume of 100 pl. The radioactive inositol triphosphate formed was extracted and measured with a liquid scin- tillation counter.

Amino Acid Composition-Proteins were separated by SDS-PAGE and blotted onto a PVDF membrane. The blotted protein bands were cut out and then hydrolyzed in 6.0 N HCl for 24,48, or 72 h at 110 "C in evacuated sealed tubes. Amino acids were converted to phenylthio- carbamyl-derivatives according to the method of Bidlingmeyer et al. (20) and then analyzed with a Pic0 Tag amino acid analysis system (Waters). Cysteine was determined as cysteic acid after performic acid oxidation (21). Tryptophan was estimated fluorometrically by the method of Pajot (22).

acetic acid at 25 "C at a flow rate of 1 rnl/min for 4 h. Peptides were monitored for absorbance at 214 nm. The isolated peptides were analyzed with the Protein Sequencer.

Preparation of Polyclonal Antibody

The purified protein (100 pg) emulsified with Freund's complete adjuvant was injected intradermally into a male rabbit weighing -2 kg. The first and second booster injections (50 pg each) of complete adjuvant were given 4 weeks after the preceding injection, respec- tively. The third booster (50 pg) of incomplete adjuvant was given 4 weeks after the second one. Serum was collected 7 days after the last injection.

Immunoblot Analysis

Proteins separated by two-dimensional electrophoresis or SDS- PAGE were electrophoretically blotted onto a PVDF membrane and then immunostained with the polyclonal antibody and an immuno- blotting kit. For analysis of the stripped ER, 600 pg of protein from the isolated stripped ER was suspended in 10 p1 of 10 mM Tris-HC1, pH 7.5, containing 1% Triton X-100 and 0.5 mM PMSF and then was allowed to stand for 30 min at 4 "C. The suspension was centri- fuged for 10 min at 10,000 X g, and then 0.5 p1 of the supernatant was subjected to two-dimensional electrophoresis.

Immunoelectron Microscopy

The liver of a male Sprague-Dawley rat (350 g of body weight) was perfused in situ through the portal vein with a mixed fixative of 2% paraformaldehyde and 0.1% glutaraldehyde in 0.1 M sodium phos- phate buffer, pH 7.4. Small pieces of the liver were embedded in LR- White resin (London Resin Co., Ltd., Hamshire, United Kingdom) at 60 "C. Ultrathin sections were cut and blocked with Block Ace (Dainippon Seiyaku Co., Ltd., Tokyo). The sections were incubated for 2 h a t room temperature with the anti-ERGOF antibody, washed, and then incubated for 1 h with anti-rabbit IgG-gold (15-nm gold particles; E-Y Laboratories, Inc., CA). Then the sections were stained with uranyl acetate and lead citrate.

Assay of Proteolytic Degradation

The purified phosphoinositide-specific phospholipase C-a family (ER60) was dialyzed overnight against 1 liter of 10 mM bis-Tris/HCl, pH 7.0. The dialysate (3-5 pg of protein), supplemented with 0.1 M 0-mercaptoethanol, was allowed to stand for the indicated times and at the indicated temperatures in the absence or presence of urea with or without a drug for determination of autocatalytic degradation. Protein disulfide-isomerase, calreticulin, carboxylesterase El , bovine serum albumin or casein, and ER60 were incubated for 3 h at 37 "C with 10 mM bis-Tris/HCl, pH 6.5, supplemented with 0.1 M 0- mercaptoethanol in the presence or absence of an inhibitor in a final volume of 9 pl. The reaction products were analyzed by SDS-PAGE.

Preparation of Protein Disulfide-Isomerase, Calreticulin, and Carboxylesterase

Protein disulfide-isomerase, calreticulin, and carboxylesterase E l were prepared from an extract of the stripped ER of rat liver by B. cereus phospholipase C treatment as described under "Purification of Phosphoinositide-specific Phospholipase C-a Family." The extract was applied to a DEAE Toyopearl 650 column equilibrated with buffer A. The eluted fractions were analyzed by SDS-PAGE, and the NHz-terminal amino acid sequences of the 60-kDa proteins were

15154 ER Protein Degradation determined. Carboxylesterase E l was eluted in the flow-through fractions. Protein disulfide-isomerase and calreticulin were eluted with a linear gradient of 0-0.25 M NaCl in buffer A. The peak fractions of protein disulfide-isomerase and calreticulin and the flow-through fractions were pooled separately and then precipitated with ammo- nium sulfate (516 g/liter). The precipitates were dissolved and di- alyzed overnight against buffer B. The dialysates were applied to an AF-heparin Toyopearl 650M column equilibrated with buffer B. The flow-through fractions of protein disulfide-isomerase and carboxyl- esterase E l and the fractions eluted with 0.175 M KC1 in buffer B for calreticulin were collected separately and applied to a TSK gel G3000SW column. The peak fractions of each protein were collected and dialyzed overnight against 1 liter of 10 mM bis-Tris/HCl, pH 7.0.

Miscellaneous Protein concentrations were determined using a protein assay kit

from Bio-Rad with y-globulin as a standard. SDS-PAGE was performed according to the method of Laemmli

(23) using 10 or 12.5% acrylamide. All SDS-PAGE analyses were carried out under reducing conditions. Two-dimensional gel electro- phoresis was carried out by the method of O’Farrell(24). The proteins were stained with Coomassie Brilliant Blue R-250 or a silver staining kit (Daiichikagaku, Tokyo). For the detection of glycoproteins, pro- teins were blotted onto a PVDF membrane and then stained by the periodic acid-Schiff method (25).

Phosphoamino acid analysis was carried out by the method of Murthy and Iqbal(26) using the Pic0 Tag amino acid analysis system.

RESULTS

Purification of Phosphoimsitide-specific Phospholipase C-a Family-We first discovered, in the rough ER fraction of rat liver, the existence of 60-kDa protein(s) of which the internal peptide fragment amino acid sequences were identical to those of phosphoinositide-specific phospholipase C-a. Then the pu- rification of this protein (ER60) was attempted by monitoring as to an M, of 60,000, by SDS-PAGE, and the NH2-terminal sequence SDVLELTDEN, which is the NH2-terminal amino acid sequence of phosphoinositide-specific phospholipase C-a (14). In the first step, the stripped ER (ribosome-depleted rough ER) of rat liver was treated with phosphatidylcholine- specific phospholipase C from B. cereus. Then 12% of the PIP2 hydrolytic activity and 18.7% of the proteins from the stripped ER were released into the supernatant on centrifu- gation at 200,000 X g. The rest of the activity was recovered in the sedimented fraction. Since PIP2 hydrolytic activity was not exhibited by the B. cereus phospholipase C preparation used in this experiment, the released activity was thought to be due to endogenous phosphoinositide-specific phospholipase C. As shown in Fig. 1, 60-kDa proteins in the extract were eluted with a linear gradient of NaCl from a DEAE column as three peaks (peaks 1-111) (Fig. 1B). ER60 was detected in peak I eluted at an NaCl concentration between 60 and 110 mM. Proteins that possessed the NH2-terminal amino acid sequences of two abundant reticuloplasmins, protein disul- fide-isomerase (5) and calreticulin (8), were distributed in peaks I1 and 111. A protein exhibiting PIP, hydrolytic activity was eluted at an NaCl concentration between 120 and 170 mM (Fig. lA). However, no phosphoinositide-specific phos- pholipase C activity was detected in the peak I fraction. The proteins in peak I were applied to a heparin column and eluted with 0.4 M KC1 (Fig. 2, lane 4 ) . In the next purification step, which involved a TSK gel G3000SW column, ER60 was eluted at a position corresponding to an apparent M , of 60,000, suggesting that ER60 is a monomeric protein (Fig. 3). After this gel filtration column chromatography, the ER60 fraction was apparently homogeneous as judged by SDS-PAGE (Fig. 4, lane 6). However, the ER60 in this fraction was separated into seven spots with different PI values (5.70,5.95,6.10,6.20, 6.35, 6.50, and 6.65) by two-dimensional gel electrophoresis (Fig. 5A). Then ER60 was fractionated into eight peaks by

A ,

Fraction Number

R

koa

41 . 34-

- ” I II 111

FIG. 1. DEAE Toyopearl 650 chromatography of crude ex- tract of stripped ER of rat liver. A, the crude extract of the stripped ER of rat liver was chromatographed on the DEAE column. Protein was monitored for absorbance at 280 nm (--). PIPp hydro- lytic activity (.--.) was determined as described under “Experi- mental Procedures.” The bar indicates the pooled ER60 fraction. B, SDS-PAGE (10% gel) analysis with 2.5-pl aliquots of each fraction from the DEAE column. Proteins were stained with Coomassie Bril- liant Blue R-250. Numbers above the lanes correspond to the fraction numbers. The Roman numerals (1-111) denote the peaks of proteins with M, values of 60,000.

1 2 3 4 kDa ““ kDa

117- 92 - 73 - 58-

41- 34 -

- 60

FIG. 2. AF-heparin Toyopearl 650M column chromatogra- phy of ER60 fraction. The ER60 fraction eluted from the DEAE column (lane I ) was applied to the heparin column as described under “Experimental Procedures.” The unabsorbed fraction (2 pl) (lane 2 ) and the fractions eluted with 0.175 M (2 pl) (lane 3) and 0.4 M (2 pl) (lane 4 ) KC1 were analyzed by SDS-PAGE (10% gel). Proteins were stained with Coomassie Brilliant Blue R-250.

TSK gel DEAE-5PW chromatography (Fig. 6). The proteins in these eight peaks were named ER6OA-ERGOH according to the order of their elution. ERGOF, one of the major proteins in the ER60 fraction, was collected and rechromatographed.

ER Protein Degradation 15155

I 1 I

3 0 4 0 5 0 6 0

Fraction Number

42 43 44 45 48 47 40 49 50

73 - 58-

- ”- -60

FIG. 3. TSK gel G3000SW gel filtration chromatography of ER60 fraction. A, the ER60 fraction eluted from the heparin column was applied to the column as described under “Experimental Proce- dures.” Protein was monitored for absorbance at 280 nm (--). The elution positions of the standards are indicated by arrows: blue dextran (M, 2,000,000), alcohol dehydrogenase (150,000), ovalbumin (45,0001, carbonic anhydrase (29,000), and cytochrome c (12,400). The bar indicates pooled ER60. B, SDS-PAGE (10% gel) analysis with 0.25-p1 aliquots of each fraction (0.4 ml) from the TSK gel G3000SW column. Proteins were stained with Coomassie Brilliant Blue R-250. Numbers above the lanes correspond to the fraction numbers.

1 2 3 4 5 6 kDa tDs 191

117 92

41 - - 34 -

FIG. 4. SDS-PAGE analysis of pooled fractions at various stages of purification of ER60. The whole stripped ER (75 pg of protein) (lane I ) ; precipitated protein (51 pg of protein) (lane 2); released protein after B. cereus phospholipase C treatment of the stripped ER (24 pg of protein) (lane 3); and fractions obtained by DEAE Toyopearl 650 column chromatography (20 pg of protein) (lane 41, heparin column chromatography (6 pg of protein) (lane 5), and TSK gel G3000SW column chromatography (3 pg of protein) (lane 6) were electrophoresed on a 10% polyacrylamide gel. The gel was stained with Coomassie Brilliant Blue R-250.

kDa

117 92

73 58

41

kDm

t (’1

92

5a 73

D

8 7 8 5 4 3 8 7 6 5 4 3 . I . .

pH pH

FIG. 5. Western blotting of ER60 and Triton X-100 extract of stripped ER after two-dimensional gel electrophoresis. After two-dimensional gel electrophoresis of ER60 obtained by TSK gel G3000SW column chromatography (1 pg of protein) ( A and B ) and a Triton X-100 extract of the stripped ER (C and D), protein was stained with silver ( A and C) or transferred onto a PVDF membrane and then immunostained with anti-ERGOF antibody ( B and D) as described under “Experimental Procedures.” First dimen- sion, isoelectric focusing ( I E F ) on a 5% polyacrylamide gel containing 2% Bio-Lyte 3/10; second dimension, SDS-PAGE (10% gel).

‘i o.20 I Y T o.20 - E ””-

””- s 0.15 -””” N

m c $ 0.10 -

2 0.05 - n

0 0 20 40 60 80 1 0 0 120 140 160

, 0 0 20 40 60 80 1 0 0 120 140 160

Time (min) FIG. 6. TSK gel DEAE-5PW chromatography of ER60. The

ER60 fraction eluted from the TSK gel G3000SW column was sub- jected to TSK gel DEAE-5PW column chromatography as described under “Experimental Procedures.” Protein was monitored for absorb- ance at 280 nm (--). ER6OA-ERGOH fractions were separately pooled as indicated.

This ERGOF fraction gave a single spot with a PI value of 6.10 on two-dimensional gel electrophoresis, followed by silver staining (Fig. 7A). Finally, 680 pg of protein from ERGOF was obtained.

Amino Acid Composition of ERGOF-The amino acid com- position of ERGOF was similar to that calculated from the cDNA of phosphoinositide-specific phospholipase C-a (Table I). This suggests that the ERGOF protein is homologous to the phosphoinositide-specific phospholipase C-a protein.

Purtiul Amino Acid Sequence of ERGOF-The NH2-terminal amino acid sequence of ERGOF was determined to be SDVLELTDEN, which is identical to the NH2-terminal se- quence of phosphoinositide-specific phospholipase C-a (Table 11). After digestion of ERGOF with lysylendopeptidase, the digest was subjected to HPLC on a pBondasphere Cl8 column and thus separated into peptide fragments. The amino acid sequences of the three peptides obtained were determined to

15156 ER Protein Degradation

A IEF

kDa

117. 92. 73. 58-

B

kDa 1 117- U) 92- g 73-

I 58” 9

41 - 41- 34-

I I I I l l 34- 8 7 6 5 4 3

pH - FIG. 7. Two-dimensional gel electrophoresis and SDS-

PAGE of ERGOF. A , purified ERGOF was subjected to two-dimen- sional gel electrophoresis (0.5 pg of protein) as described in the legend to Fig. 5. B, the sample was prepared in the presence of 1 mM pCMB to prevent contamination by a trace amount of autocatalytic degra- dation products. Purified ERGOF (5 pg of protein) was subjected to SDS-PAGE (10% gel). Proteins were stained with silver. IEF, iso- electric focusing.

TABLE I Amino acid composition of ERGOF

Amino acids ER60F PIPLC-nb

Asx Glx Ser GIY His Arg Thr Ala Pro TY r Val Met Ile Leu Phe LYS %Cys Trp

mol %

10.5 11.8 5.7 8.4 2.0 4.1 5.7 7.6 5.2 4.1 6.2 1.5 3.9 7.8 5.2 8.5 1.4 0.4

mol %

11.9 10.2 5.6 6.5 1.9 4.4 5.6 7.7 5.0 4.0 6.0 1.4 4.0 7.7 5.8

10.2 1.5 0.6

“Values are from direct analysis and are the average of three determinations (24-, 48-, and 72-h hydrolyses). Half-cysteine and tryptophan were determined as described under “Experimental Pro- cedures.”

Values were calculated from the deduced amino acid sequence of phosphoinositide-specific phospholipase C-n (PIPLC-a) (14).

TABLE I1 Partial amino acid sequence of ERGOF

The partial amino acid sequence of ERGOF was determined as described under “ExDerimental Procedures.”

Peptide ERGOF amino fragment acid sequence

Corresponding PIPLC-n” sequenceb

NH, terminus SDVLELTDEN Ser*5-Asn34 1 AASNLRDNYRFAHTNVESLV Ala’74-Val’9’ 2 MDATANDVPSPYEV Met4”-Va1446 3 GFPTIYFSPAN G l ~ ~ ‘ ~ - A s n ‘ ~

PIPLC-n, phosphoinositide-specific phospholipase C-n. ~ ~~ ~~

*Data are from Bennett et al. (14).

be identical to that of phosphoinositide-specific phospholipase C-a (Table 11).

Microheterogeneity of ERGO-Purified ERGOA-ERGOH cross-immunoreacted with the anti-ERGOF antibody (data not shown). On immunoblotting after two-dimensional gel elec-

1 2 3 4 5 6 kDa kDa

117 92

73 sJ3

-80

41

34

FIG. 8. Subcellular distribution of ER60. The stripped ER (100 and 5 pg of protein; lanes 1 and 4, respectively), smooth ER (100 and 5 pg of protein; lanes 2 and 5, respectively), and cytosol (100 and 5 pg of protein, respectively; lanes 3 and 6, respectively) were subjected to SDS-PAGE (10% gel). Protein was stained with Coomassie Bril- liant Blue R-250 (lanes 1-3) or transferred and immunostained (lanes 4-6) as described in the legend to Fig. 5.

FIG. 9. Immunoelectron microscopic localization of ER60 in rat hepatocyte by anti-rabbit IgG-colloidal gold technique. M , mitochondria. Original magnification X 50,000. Bar = 0.1 pm. Arrows denote gold particles localized in the cisternal space of the ER.

trophoresis of purified ERG0 obtained by TSK gel G3000SW column chromatography, positive spots were seen at the same positions as the’ spots obtained by protein staining (Fig. 5, A and B ) . No spot was observed to cross-immunoreact with preimmune control serum (data not shown). The NH,-ter- minal amino acid sequences of ERGOA-E, ERGOG, and ERGOH were determined to be identical to that of ERGOF. These results suggest that the ER60 subtypes are proteins of which the amino acid sequences are essentially identical to each other. I t seems unlikely that the microheterogeneity was artificially caused during the purification procedures since

Urea(M) 0 0 1 Timdh) 5 5 5 Temp("C) 25 4 4

koa

117- 92 - 73- - - - - 58-

41- 34-

1 2 3 4

6 7 0

ER Protein Degradation

3 5 4

kD8

5

FIG. 10. Interautocatalytic degradation of ER60. ER60 (3 pg of protein19 pl for lanes 1-5 and 9 and 5 pg of protein115 pl for lanes 6-67, which was eluted from a TSK gel G3000SW column, was allowed to stand in the presence of 0.1 M P-mercaptoethanol under the following conditions: lanes 2 and 7, 5 h at 25 "C without urea; lanes 3 and 8, 5 h a t 4 "C without urea; lane 4, 5 h a t 4 "C with 1 M urea; lanes 5 and 9, 5 h at 4 "C with 3 M urea. Then the samples were subjected to SDS-PAGE (12.5% gel). Proteins were stained with Coomassie Brilliant Blue R-250 (lanes 1-5) or transferred and im- munostained (lanes 6-9) as described in the legend to Fig. 5. Lunes I and 6 contained untreated ER60 (3 pg of protein).

1 2 3 4 5 6 7 8 kDa

=a FIG. 11. Interautocatalytic degradation of ER6OA-ERGOH.

Two micrograms of protein from each of the ER6OA-ERGOH proteins (lanes 1-8, respectively) was allowed to stand for 5 h a t 4 "C in the presence of 3 M urea and 0.1 M P-mercaptoethanol and then subjected to SDS-PAGE (12.5% gel). Protein was stained with silver.

immunoblotting after two-dimensional gel electrophoresis of a Triton X-100 extract of the stripped ER showed the same profiles as those of purified ER60 (Fig. 5, B and D). Among Triton X-100-extracted proteins from the stripped ER, only ER60 proteins were immunostained, suggesting that the anti- serum was highly specific. I t is likely that ER60 proteins were abundant in the stripped ER (Fig. 5 C ) . In addition, phospho- serine, phosphothreonine, and phosphotyrosine were hardly detected in purified ER6OA-ERGOH, although 20 pg of protein was used for the assays with the Pic0 Tag system and no sugar was detected with 0.5 pg of protein.

ER60 was detected in the rough and smooth ER fractions, but not in the cytosol fraction of rat liver on immunoblotting analysis (Fig. 8). In an electron micrograph, immunoreactivity appeared to be localized in the cisternal space of the ER of a rat hepatocyte (Fig. 9). The gold particles indicated with arrows were localized in the cisternal sides of the ER. The rest of the particles appeared also to be in the ER rather than in cytosol, which is by no means certain due to blurring of the photograph.

kD8

117- 92 - 73- 58-

41 - 34 -

15157

1 . . ~ 2 " 3 -.i. 4 ".. 5-6 kD8

FIG. 12. Effects of pCMB, diisopropyl fluorophosphate, and PMSF on interautocatalytic degradation of ER60. ER60 (3 pg of protein/9 pl) obtained by TSK gel G3000SW column chromatog- raphy was allowed to stand for 5 h at 4 "C in the presence of 3 M urea under the following conditions: lane 2, in the presence of 0.1 M 8- mercaptoethanol; lane 3, without a drug; lane 4, in the presence of 1 mM pCMB; lane 5, in the presence of 1 mM diisopropyl fluorophos- phate, and 0.1 M 8-mercaptoethanol; lane 6, in the presence of 1 mM PMSF and 0.1 M P-mercaptoethanol. Then the samples were sub- jected to SDS-PSGE (12.5% gel). Protein was stained with Coomassie Brilliant Blue R-250. Lune 1 contained untreated ER60 (3 pg of protein).

PCMB - - - + " " + - - + - - kDa

-117 - 92 - 13 -58 -1

-41 - 34

1 Z J 4 5 0 I 0 Y 1U l l l Z 1 J 1 4

FIG. 13. Proteolytic degradation of ER proteins by ER60. Bovine serum albumin (3 pg of protein) (lanes 2-4), casein (3 pg of protein) (lanes 5 and 6 ) , protein disulfide-isomerase ( P D I ) (8 pg of protein) (lanes 7-9), calreticulin (5 pg of protein) (lanes 10-12), and carboxylesterase E l (5 pg of protein) (lanes 13 and 14) were incubated with (lanes 3, 4, 6, 8, 9, 11, 12, and 14) or without (lanes 2, 5, 7, 10, and 13) 1.5 pg of ER60, which was eluted from a TSK gel G3000SW column under the following conditions: lanes 2,3,5-8, 10, 11, 13 and 14 in the presence of 0.1 M 0-mercaptoethanol; lanes 4 ,9 , and 12, in the presence of 1 mM pCMB. Then the samples were subjected to SDS-PAGE (12.5% gel). Lane 1 contained ER60 (1.5 pg), which was incubated in the presence of 0.1 M 8-mercaptoethanol for 3 h at 37 "C. Asterisks (lanes 8 and 11) denote peptide fragments for which the NH2-terminal amino acid sequences were determined.

Intermolecular Degradation of ERGO-Three peptides with apparent M , values of 30,000, 29,000, and 28,000 were de- tected, with a concomitant decrease in the ER60 protein, by SDS-PAGE after ER60, which had been eluted from a TSK gel G3000SW column, was allowed to stand for 5 h at 4 "C in the presence of 3 M urea and 0.1 M P-mercaptoethanol (Fig. 10, lane 5). Under these conditions, small amounts of a 57- kDa fragment and smaller peptide fragments were also formed. Degradation was highly enhanced in the presence of 3 M urea, but not with less than this concentration (Fig. 10, lane 4 ) . In the absence of urea, degradation occurred to a very

15158 ER Protein Degradation

slight extent at 25 "C since the products could not be detected by Coomassie Brilliant Blue staining, but by immunostaining (Fig. 10, lanes 2 and 7). The immunoreactivity of the 28-kDa peptide fragment was very weak compared with that of the 30-kDa fragment. The cleavage of ER60 optimally occurred between pH 6 and 7 in the presence of 3 M urea. Degradation was hardly observed above pH 7.5 or below pH 4.5 (data not shown). When each of the ERGOA-ERGOH proteins was in- cubated for 5 h at 4 "C in the presence of 3 M urea and 0.1 M P-mercaptoethanol, intermolecular degradation was similarly observed (Fig. 11). The NH2-terminal amino acid sequence of the 30-kDa peptide fragment, one of the major products, was determined to be SDVLELTDEN, which was identical to the NH,-terminal sequence of ERGOA-ERGOH. On the other hand, the NH,-terminal amino acid sequence of the 28-kDa peptide fragment was KTFLDAGXXL, which was identical to the Lys287-Le~296 sequence of phosphoinositide-specific phospholipase C-a. Therefore, the 28-kDa peptide fragment may be generated through endoproteolytic cleavage between AlazE6 and LysZE7 of ER60 subtypes.

Cleavage was stimulated in the presence of P-mercaptoeth- anol (Fig. 12, lanes 2 and 3 ) and was inhibited by pCMB (lane 4), suggesting that the thiol group of ER60 is essential for autocatalytic cleavage, whereas diisopropyl fluorophosphate and PMSF did not show any effect (lanes 5 and 6).

Degradation of ER Proteins by ERGO-Protein disulfide- isomerase, calreticulin, and carboxylesterase E l prepared from the ER of rat liver were used as substrates. Protein disulfide-isomerase was cleaved to produce 48- and 54-kDa peptide fragments (Fig. 13, lane 8 ) of which the NH,-terminal amino acid sequences were determined to be the same se- quence, DALEEEDNVL, which was identical to the NH2- terminal amino acid sequence of protein disulfide-isomerase (5). Calreticulin was cleaved to produce a 55-kDa peptide fragment (Fig. 13, lune 11) of which the NHa-terminal amino acid sequence was determined to be FEPFSNXGQT, which was identical to the Phesl-Thrgo sequence of calreticulin (8). Hence, the 55-kDa peptide fragment may be generated through cleavage between Argo and Phe" of calreticulin. Carboxylesterase E l was resistant to the digestion (Fig. 13, lane 14). Bovine serum albumin was cleaved (Fig. 13, lane 3 ) . However, casein could not be degraded (Fig. 13, lune 6). Degradation was inhibited by pCMB (Fig. 13, lanes 4 , 9, and 12).

DISCUSSION

We described the purification of ER60 from the rough ER fraction of rat liver; its NH2-terminal amino acid sequence was identical to that of phosphoinositide-specific phospholi- pase C-a from rat basophilic leukemia cells (14). ER60 was separated into eight proteins (ERGOA-ERGOH) according to differences in PI value. ERGOA-ERGOH were highly similar to each other since all of them showed positive immunoreactivity with the anti-ERGOF antibody and had identical NH,-termi- nal amino acid sequences. However, it was difficult to find differences in their molecular structure since ERGOA-ERGOH were neither phosphorylated proteins nor glycoproteins. Some other modification should be considered.

PIP, hydrolytic activity was detected in the fractions re- leased from the stripped ER on treatment with B. cereus phospholipase C. However, the activity was completely sepa- rated from ER60 on the first DEAE column chromatography. ER60 did not show any phosphoinositide-specific phospholi- pase C activity even though a high degree of identity was found between ER60 and phosphoinositide-specific phospho- lipase C-a. Recently, Martin et al. (16) isolated a 58-kDa

protein from rat liver microsomes. Although this protein showed 99% homology in its partial amino acid sequence to phosphoinositide-specific phospholipase C-a, it did not show phosphoinositide-specific phospholipase C activity. The mi- crosomal protein of rat liver, Q-2, purified by Srivastava et al. (17) with a partial amino acid sequence identical to that of phosphoinositide-specific phospholipase C-a also exhibited no phosphoinositide-specific phospholipase C activity, but it exhibited thio1:protein-disulfide oxidoreductase activity. It cannot be excluded that the 58-kD and Q-2 proteins may each correspond to one of the ERGOA-ERGOH proteins. It has been questioned whether the phosphoinositide-specific phospholi- pase C-a sequenced by Bennett et aE. (14) is a true phosphoi- nositide-specific phospholipase C protein. It is possible that phosphoinositide-specific phospholipase C-a is an activated form of these ER proteins, with the activation occurring through an unknown post-translational modification, consid- ering that Q-2 was cross-immunoreactive with a polyclonal antibody to guinea pig uterine cytosolic phosphoinositide- specific phospholipase C-a (17) and that phosphoinositide- specific phospholipase C purified from bovine brain by Tomp- kins and Moscarello (27) exhibited considerable sequence homology to the phosphoinositide-specific phospholipase C-a reported by Bennett et al. (14). The possibility cannot be excluded that ERGOA-ERGOH were degradation products of phosphoinositide-specific phospholipase C-a. However, the molecular weights of ERGOA-ERGOH were similar to that of phosphoinositide-specific phospholipase C-a (12).

Some proteolytic processes thought to occur within the ER include the cleavage of signal peptides (28), the removal of carboxyl-terminal extensions of glycosylphosphatidylinositol- anchored proteins (29), and ER degradation (30-33). Except for signal peptidase, however, no other ER protease has been isolated and characterized to date. ER60 showed protease activity. The cleavage of a protein was observed at the specific peptide bond. The sites susceptible to proteolytic degradation were determined to be the peptide bonds between AlazE6 and LysZR7 of ER60 and between Ar$" and Phe" of calreticulin. The proteolytic activity appeared not to be due to contami- nation by other proteolytic enzymes, but to ERGOA-ERGOH themselves since purified ERGOA-ERGOH each showed the activity (Fig. 11). The SDS-PAGE profile of heavily loaded ERGOF is shown in Fig. 7B. No protein bands other than that of ERGOF were detected even on silver staining. Similar profiles were obtained with ERGOA-ERGOE, ERGOG, and ERGOH (data not shown). Hence, we think they may be free from contamination by any other proteolytic enzymes. The proteolytic activity of ER60 required P-mercaptoethanol and was inhibited by pCMB. On the other hand, diisopropyl fluorophosphate and PMSF did not inhibit the activity. This suggests that ER60 catalyzes proteolytic degradation as a thiol protease-like enzyme, i.e. not as a serine protease-like enzyme.

At the present time it is not clear what is the physiological role of any member of this class of proteins. Stafford and Bonifacino (34) suggested that some enzymes involved in ER degradation belong to the class of cysteine proteases. Simoni and co-workers (32, 33, 35) indicated that cysteine protease is related to the degradation of 3-hydroxy-3-methylglutaryl- CoA reductase in the ER. ER60 cleaved protein disulfide- isomerase and calreticulin, but not carboxylesterase El , among ER proteins. Therefore, it is possible to assume that this phosphoinositide-specific phospholipase C-a family is involved in the degradation of proteins in the ER.

Acknowledgments-We are grateful to Dr. Y. Urade and M. Taki-

ER Protein Degradation 15159

mot0 (International Research Laboratories, Ciba-Geigy Japan Ltd.) for preparation of the immunoelectron photomicrograph.

REFERENCES

2. Munro, S., and Pelham, H. R. B. (1987) Czll48,899-907 1. Koch, G. L. E. (1987) J. Cell Sci. 87,491-492

3. Munro, S., and Pelham, H. R. B. (1986) Cell 46,291-300 4. Mills, E. N. C., Lambert, N., and Freedman, R. B. (19831 Biochem. J. 213,

5. Edman, J. C., Ellis, L., Blacher, R. W., Roth, R. A., and Rutter, W. J.

6. Macer, D. R. J., and Koch, G. L. E. (1988) J. Cell Sci. @1,61-70

8. Murthy, K. K., Banville, D., Srikant, C. B., Carrier, F., Holmes, C., Bell, 7. Smith, M. J., and Koch, G. L. E. (1989) EMBO J. 8,3581-3586

9. Mentlein, R., Ronai, A., Rohbi, M., Heymenn, E., and Deimling, 0. V.

10. Mentlem, R., Rix-Matzen, H., and Heymann, E. (1988) Biochim. Biophys.

11. Robbi, M., Beaufay, H., and Octave, J.-N. (1990) Biochem. J. 269, 451-

12. Bennett, C. F., and Crooke, S. T. (1987) J . Biol. Chem. 262,13789-13797 13. Pelham, H. R. B. (1990) Trends Biochem. Sci. 15,483-486 14. Bennett, C. F., Balcarek, J. M., Varrichio, A,, and Crooke, S. T. (1988)

15. Haugejorden, S. M., Srinivasan, M., and Green, M. (1991) J. Biol. Chcm.

16. Martin, J. L., Pumford, N. R., LaRosa, A. C., Martin, B. M., Gonzaga, H.

245-248

(1985) Nature 317,267-270

A., and Patel, Y. C. (1990) Nucleic Acids Res. 18,4933

(1987) Biochzm. Biophys. Acta 913, 27-38

Acto 964,319-328

458

Nature 334, 268-270

266,6015-6018

M. S., Beaven, M. A., and Pohl, L. R. (1991) Biochem. Biophys. Res. Cornmun. 178,679-685

17. Srivastava, S. P., Chen, N., Liu, Y., and Holzman, J. L. (1991) J. Biol.

18. Krelblch, G., Ulrich, B. L., and Sabatini, D. D. (1978) J. Cell Biol. 77,464- cbem. 266, 20337-20344

dR7 19. M&ama, T., Narita, H., Oki, M., Matsuura, T., and Kito, M. (1990) J.

20. Bidlingmeyer, B. A., Cohen, S. A,, and Tarvin, T. L. (1984) J . Chromatogr. Biochem. (Tokyo) 108,414-419

.?%X 93-1 Ad 21. Moore, S. (1963) J. Biol. Chem. 238,235-237 22. Pajot, P. (1976) Eur. J . Biochem. 6 3 , 263-269 23. Laemmli U. K. (1970) Nature 227,680-685 24. OFarrelC P. H. (1975) J. Bwl. Chem. 260,4007-4021 25. Kaschnitz, R., Peterlik, M., and Weiss, H. (1969) A d . Biochern. 30, 148-

-","I " _

1 Fi<? 26. MGithy L. R., and I bal, K. (1991) Anal. Biochem. 193,299-305 27. Tompkins, T. A., an! Moscarello, M. A. (1991) J. Bid. Chem 266,4228-

28. Shelness, G . S., Kanwar, Y. S., and Blobel, G. (1988) J. Biol. Chem. 263,

29. Ferguson, M. A. J., and Williams, A. F. (1988) Annu. Reu. Biochem. 57,

4236

17063-17070

30. Klausner, R. D., and Sitia, R. (1990) Cell 62,611-614 31. Wirkstrom, L., and Lodish, H. F. (199!) J. Cell Biol. 113, 997-1007 32. Chun, K. T., Bar-Nun, S., and Simonl, R. D. (1990) J. Biol. Chem. 265,

285-320

33. Inoue, S., Bar-Nun, S., Roitelman, J. , and Simoni, R. D. (1991) J. Biol.

34. Stafford, J. S., and Bonifacino, J. S. (1991) J. Cell BioL 115, 1225-1236 35. Inoue, S., and Simoni, R. D. (1992) J. Biol. Chem. 267,9080-9086

22004-22010

Chem. 266, 13311-13317