ARCHIVES

Vol.

OF BIOCHEMISTRY

284, No. 1, January,

AND

BIOPHYSICS

pp. 63-70,

1991

Biogenesis of the Endoplasmic Reticulum in Activated Lymphocytes: Temporal Relationships between the Induction of Protein N-Glycosylation Activity and the Biosynthesis of Membrane Protein and Phospholipid’ Jeffrey

S. Rush,*

Thomas

Sweitzer,?

Claudia

Kent,?

Glenn

L. Decker,$

B

and Charles

J. Waechter”

*Department of Biochemistry, University of Kentucky College of Medicine, A. B. Chandler Medical Center, Lexington, Kentucky 40536; TDepartment of Biochemistry, Purdue University, West Lafayette, Indiana 47907; and SDepartment of Biochemistry and Molecular Biology, University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030

Received

June

5, 1990, and in revised

form

September

8, 1990

An earlier report from this laboratory documented a substantial increase in the rates of dolichol-linked oligosaccharide intermediate synthesis and protein N-glycosylation in purified murine splenic B lymphocytes (B cells) activated by treatment with bacterial lipopolysaccharide (LPS). In this study the developmental patterns for the induction of lipid-mediated protein N-glycosylation, membrane protein, and phosphatidylcholine (PC) biosynthesis were compared during the proliferative response of B cells to LPS. By electron microscopy it could be seen that a distinct endoplasmic reticulum (ER) network began to develop by 24-48 h after exposure of the purified B cells to LPS. The rate of synthesis of membrane protein increased markedly during the first 10 h after activation, reaching a maximum at 30-40 h. The induction of protein N-glycosylation was delayed slightly relative to membrane protein synthesis, with glycoprotein synthesis increasing sharply approximately 20 h after activation. When phospholipid synthesis was monitored by measuring [CH,-3H]choline incorporation into PC, the rate of labeling increased slowly during the first 35 h, but more substantially between 35 and 90 h. The incorporation of labeled choline into PC was drastically reduced by 5’-deoxy-5’-isobutylthio-3-deazaadenosine, an inhibitor of CDP-choline synthesis, indicating that the incorporation of radiolabeled choline is primarily a measurement of the rate of de novo synthesis of PC. In vitro assays revealed that while choline kinase activity was virtually unchanged, CDP-choline synthetase activity increased gradually throughout the activation period. Diacylglycerol cholinephosphotransferase activity, an ER-associated enzyme, was present at low levels between i This work the American

was supported Cancer Society.

by Grant

0003-9861/91$3.00 Copyright 0 1991 by Academic Press, All rights of reproduction in any form

BC-583

awarded

to C.J.W.

by

0 and 35 h, but increased fivefold between 35 and 90 h. On the basis of the developmental patterns for the rates of protein N-glycosylation, membrane protein insertion, and PC biosynthesis determined by metabolic labeling experiments, we tentatively conclude that all of the ERassociated membrane proteins involved in these biosynthetic processes are not induced concurrently during the activation of B cells by LPS. 0 issl academic press,be.

Exposure of resting (G,) splenic B lymphocytes (B cells)* to bacterial lipopolysaccharide (LPS) and other polyclonal mitogens causesthe cells to become “activated” and enter the cell cycle (l-3). During the proliferative response there is an extensive development of an endoplasmic reticulum (ER) network (4, 5) and other membranous organelles. Recently, we have reported that when B cells are activated by treatment with LPS, anti-immunoglobulin M or the combination of phorbol ester and ionomycin, there is a dramatic induction of synthesis of dolichol-linked oligosaccharide intermediates and Nlinked glycoproteins (6, 7). The assembly of new ER and other organelles in activated B cells requires substantial increases in membrane protein and phospholipid synthesis. Since most membrane proteins and phospholipids are synthesized in the rough ER (8-12), the subcellular location of at least 20 proteins ‘Abbreviations used: B cells, B lymphocytes; LPS, bacterial lipopolysaccharide; ER, endoplasmic reticulum; DZ-SIBA, 5’-deoxy-5’.isobutylthio&deazaadenosine; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PI, phosphatidylinositol; CK, choline kinase; CPT, diacylglycerol:cholinephosphotransferase; PBS, phosphate-buffered saline; TCA, trichloroacetate; DMEM, Dulbecco’s modified Eagle’s medium: SDS, sodium dodecyl sulfate. 63

Inc. reserved.

64

RUSH

involved in the synthesis and transfer of dolichol-linked saccharide intermediates, (13-18), it was of interest to compare the developmental patterns for the induction of protein N-glycosylation, membrane protein, and phospholipid biosynthesis during the activation of B cells. The B cell system has the potential to provide new information about the regulation of the N-glycosylation apparatus as well as the biogenesis of the rough ER. Understanding the molecular events in the development of the ER in B cells may also be relevant to the recent evidence for a possible role of the ER in MHC-restricted endogenous antigen presentation (19-21). These studies suggest that processed antigens are associated with MHC class I molecules before they depart the ER. The results reported here indicate that the induction of lipid-mediated protein N-glycosylation is preceded by membrane protein synthesis, but occurs prior to the largest increase in the rate of synthesis of phosphatidylcholine (PC), the major membrane phospholipid. The implications of membrane proteins involved in the three rough ER-associated biosynthetic processes being induced asynchronously during the activation of B cells are discussed. Portions of this work have been presented in preliminary form (22). MATERIALS

AND

METHODS

Materials. [2-3H]Mannose (15 Ci/mmol) and [l-3H]myoinositol (15 Ci/mmol) were obtained from American Radiolabeled Chemicals, Inc. (St. Louis, MO). [2,3,4,5-3H]Leucine (111 Ci/mmol) and [1,2“C]ethanolamine (100 mCi/mmol) were purchased from ICN, Biomedicals Inc. (Irvine, CA). [CH3-i4C]CDP-choline (52.1 mCi/mmol) and [CH,-“Clphosphocholine (50 mCi/mmol) were acquired from New England Nuclear Research Products (Boston, MA). [CH,-3H]Choline (84.78 Ci/mmol) was purchased from Cen Saclay-Molecules Marquees (Gif-Sur-Yvettes, France). [CH3-3H]Methionine (35 mCi/mmol) was obtained from Research Products International Corp. (Elk Grove Village, IL). DBA/P mice (8-12 weeks, female) were obtained from Harlan Sprague-Dawley, Inc. (Indianapolis, IN). Diolein (sn-1,2-dioleoylglycerol), sodium taurocholate, adenosine triphosphate, and choline chloride were purchased from Sigma Chemicals Co. (St. Louis, MO). Bacterial LPS (S. Typhosa 0901) was obtained from Difco (Detroit, MI) and fetal calf serum was acquired from Sterile Systems (Ogden, UT). Agl-X8 (formate) ion exchange resin was purchased from Bio-Rad Laboratories (Richmond, CA), Whatman GF/C glass microfiber filter disks were obtained from American Scientific Products (McGaw Park, IL). Scintillation cocktail 3a70B was from Research Products Int. Corp. (Elk Grove Village, IL). 5’.Deoxy-5’-isobutylthio-3-deazaadenosine (DZ-SIBA) was a generous gift from Dr. Guilio L. Cantoni, The National Institutes of Health and Dr. Peter K. Chiang, Walter Reed Army Inst. of Research. All other chemicals and reagents were purchased from standard commercial sources. Isolation and maintenance of murine splenic B lymphocytes. T-lymphocyte-depleted preparations of B lymphocytes were isolated from female DBA/2 mice by Percoll density gradient centrifugation and maintained in culture by the procedures described previously (6). B Lymphocytes were removed from culture, Electron microscopy. fixed in phosphate buffer containing 3.5% glutaraldehyde, and washed in Sorenson’s phosphate-buffered sucrose. B cells were then postfixed with 1% phosphate-buffered osmium tetroxide, dehydrated by sequential extraction with graded concentrations of ethanol from 30 to lOO%, and

ET

AL.

embedded in Spurr’s resin. Thin sections were cut on an LKB Nova ultramicrotome and mounted on nickel grids. Thin sections were stained with uranyl acetate and lead citrate and examined using a Phillips 410 transmission electron microscope. Assay of protein N-glycosylation activity in cultured B lymphocytes. Typical reaction mixtures consisted of 0.5-2 X lo6 B cells suspended in 0.2 ml of Dulbecco’s modified Eagle’s medium (DMEM) containing 1 mM glucose and 5 &i of [2-3H]mannose in 12 ml conical tubes. Following incubation at 37°C for 30 min, the incorporation of [23H]mannose into glycoprotein was determined by a multiple extraction procedure (6, 23). Assay of protein synthesis. Total protein synthesis was assessed by following the incorporation of [2,3,4,5-3H]leucine by the procedure reported earlier (6). Membrane protein synthesis was assayed by measuring the incorporation of labeled leucine into the particulate fraction obtained after the metabolically labeled cells were homogenized. B cells were incubated in 0.2 ml of DMEM containing 5 &i [2,3,4,5,-3H]leucine. Following the incorporation period, the cells were disrupted by sonication and the particulate fraction collected on GF/C glass fiber filter disks by vacuum filtration. The filters were washed sequentially with ice cold PBS (20 ml), ice cold 10% TCA (10 ml), and diethyl ether (10 ml). The filter containing the membrane residue was then transferred to a scintillation vial and solubilized in 1 ml of 1% SDS (lOO”C, 2 min). The amount of radiolabeled protein in the membrane residue was determined after the addition of 10 ml of 3a70B scintillation cocktail. Assay of phospholipid biosynthesis. The rates of synthesis of PC, phosphatidylinositol (PI), and phosphatidylethanolamine (PE) were determined by metabolic labeling using [CHs-3H]choline, [l-3H]inositol, or [ 1,2-i4C]ethanolamine as the isotopic precursors. Phospholipids were labeled by incubating B cells (l-2 X lo6 cells) in 0.1 ml DMEM containing [CH3-3H]choline (10 &i/ml), [l-3H]inositol (100 pCi/ml), or [1,2i4C]ethanolamine (10 &i/ml, 100 KM) for 1 h at 37°C. The incubation was terminated by the addition of 3 ml of CHCl,:CH,OH (2:l) and the delipidated residue sedimented by centrifugation. The lipid extract was washed with k vol of 0.9% NaCl (24). The aqueous (upper) phase was discarded and the organic layer washed three times with CHCl,-CH,OHH,O (3:48:47). The organic layer was then transferred to a scintillation vial, dried under a stream of air, and analyzed for radioactivity by scintillation spectrometry. The radioactive products were characterized as PC, PI, or PE on the basis of cochromatography with authentic standard phospholipids with solvent mixtures A, B, and C, and their conversion to t,he corresponding glycerophosphoryl ester by mild alkaline methanolysis (25). Preparation of crude membrane fractions. Cell homogenates were prepared from B cells by sonication. Cells were removed from culture, washed two times with ice cold PBS and resuspended at 25 X lo6 cells/ ml in 100 mM Tris-HCl (pH 8.0), 0.1 M sucrose, 10 mM Z-mercaptoethanol, 1 mM EDTA, and leupeptin (25 kg/ml). Following sonication for 1 min, a crude membrane fraction was sedimented by centrifugation at 100,OOOg for 1 h. The cytosol was assayed for choline kinase activity, and the particulate fraction was resuspended in the lysis buffer and assayed for CPT activity. CDP-choline synthetase assays were performed on both fractions. Enzyme assays. Choline kinase assay mixtures contained 20-100 pg of cytosolic protein, 20 mM MgClx, 5 mM ATP, and 200 FM [CH,3H]choline (500 cpm/pmol) and 0.1 M Tris-HCl (pH 8.0) in a total volume of 0.1 ml. Following the incubation period, the reaction was stopped by the addition of 1 ml of ice cold H,O containing 5000 cpm of [CH3-i4C]phosphocholine as recovery standard. The mixture was then loaded onto a l-ml column of Agl-X8 (formate) and washed with four column volumes of H,O (26). The product was eluted directly into scintillation vials by the addition of 1.5 ml of 0.1 M NaOH. The solution was neutralized with 0.15 ml of 1 M HCl and the amount of labeled phosphocholine determined following the addition of 18 ml of scintillation cocktail 3a70B. The radioactive product cochromatographed with authentic phosphocholine on Silica Gel G developed with CHSOH-0.5% NaCl-concentrated NH,OH (50:50:1), and it was converted to free cho-

PROTEIN

FIG. 1. Reference

Electron micrographs bars = 1 pm.

of representative

cultured

N-GLYCOSYLATION

B cells exposed

line by treatment with alkaline phosphatase. Diacylglycerol:cholinephosphotransferase activity was assayed essentially as described by Percy et al. (27). Incubation mixtures contained 20-100 pg of membrane protein, 0.1 M Tris-HCl (pH 8.0), 20 mM MgCl*, 200 pM diolein, 200 pM CDP-[CHs-i4C]choline (1000-2000 cpm/pmol), and 0.5% sodium taurocholate in a total volume of 0.1 ml. After incubation for 10 min at 37”C, reactions were stopped by the addition of 3 ml of CHC&-CHsOH (2:l). The membrane residue was sedimented by centrifugation and the lipid extract was saved. After partitioning with t vol of 0.9% NaCl, the lower phase was washed three times with 2 ml of CHC13-CHsOH-0.9% NaCl (3:48:47). The washed lipid extract was then transferred to a scintillation vial, dried under a stream of air, and the amount of radiolabeled PC formed was determined by scintillation spectrometry. The enzymatically labeled product cochromatographed with authentic PC on EDTA-treated SG-81 paper developed with solvent mixtures A, B, and C, and it was converted to a product that was chromatographically identical to glycerophosphorylcholine by mild alkaline methanolysis (25). CDP-choline synthetase activity was determined by the method of Wright et al. (28). Assay mixtures contained enzyme (525 fig protein), 20 mM Tris-succinate (pH 7.8), 6 mM MgCl,, 5 mM CTP, 4 mM phospho[CH,-3H]choline (500 pCi/mmol), and 0.2 mM PC and 0.2 mM oleic acid in a total volume of 0.05 ml. Assay for the incorporation and CDP-choline. B cells

of radiolabeled choline into phosphocholine were incubated for 1 h at 37°C in 100 ~1 of DMEM with [CHs-3H]choline as described above. Following the incorporation period, cells were washed twice with ice cold PBS and extracted with 70% ethanol containing a known amount of i4C-labeled phosphocholine and CDP-choline as internal standards. Choline-containing metabolites were separated by thin layer chromatography on Silica Gel G developed with CH,OH-0.5% NaCl-concentrated NH,OH (50:50:1). The radioactive zones were located by autoradiography and eluted with 1% SDS and the amount of radioactivity in each zone was measured by scintillation counting. The loss of 3H-labeled products during the analytical procedure was calculated from the recovery of the ‘*C-labeled standard compounds.

IN

to LPS

65

B CELLS

for 0 h (A, X9460),

24 h (B, X9460),

and 48 h (C, X15,180).

Isotopically labeled phospholipids Chromatographic procedures. formed by B cells were identified by chromatography on EDTA-impregnated SG-81 paper (29) developed by solvent mixtures (A) CHC13CHBOH-HP0 (65:25:4) (30); (B) CHCl,-CH,OH-concentrated NH,OH (65:25:4); or (C) CHCl,-CH,OH-glacial acetic acid (65:25:4). Radiolabeled phosphatides were located by autoradiography and standard phospholipids were detected by spraying with a molybdate reagent of Dittmer and Lester (31). The water-soluble glycerophosphoryl derivatives were produced by mild alkaline methanolysis (25) and resolved by two-dimensional chromatography on cellulose thin-layer plates as described elsewhere (32). Chemical analyses. Cellular phospholipids were extracted as described above for the [CH,-3H]choline-labeling experiments, and lipidphosphorus was determined by the method of Bartlett (33). Protein in membrane suspensions was measured by the method of Lowry et al. (34).

For each experiment radioactivity Determination of radioactivity. was measured by liquid scintillation spectrometry in a Packard TriCarb in the presence of 3a70B. All data are average values calculated after counting each sample for 10 min.

RESULTS

Electron microscopy of B cells at various times after exposure to LPS. The time course for the development of the rough ER in the purified B cell preparations used for these studies was examined by low magnification electron microscopy after treatment with the polyclonal mitogen, LPS. As seen in Fig. lA, a typical resting B cell contained only a very sparse ER network that is primarily associated with the nuclear envelope. Characteristic morphological changes began to occur by 24 h after treatment with LPS (Fig. 1B). There was an enlargement of many B cells with

66

RUSH

Y F 02 p

GLYCOPROTEIN

I 0

I I 25 50 TIME OF EXPOSURE

I 75 TO LPS

1 100

OF z

(h)

FIG. 2. Time course for the induction of protein N-glycosylation (0) and membrane protein synthesis (0). B cells (2 X lo6 cells/ml) were cultured in complete medium in the presence of LPS (50 pg/ml). At the indicated times, cells were removed from culture, labeled metabolically with either [2-3H]mannose or [2,3,4,5-“Hlleucine for 30 min at 37’C. The incorporation of the isotopic precursors into glycoprotein or membrane protein was assayed as described under Materials and Methods.

a marked development of the rough ER and an increased number of mitochondria. By 48 h (Fig. 1C) there was extensive development of a prominent rough ER network, and a large fraction of the cell population appears to be plasmablastic. During activation the ultrastructural changes observed in Fig. 1 are reflected in substantial changes in the membrane protein and phospholipid composition of the B cells. The cellular content of membrane protein and phospholipid began to rise lo-20 h after the addition of LPS as newly synthesized ER appeared (data not included). As anticipated, the ratio of phospholipid to membrane protein (approximately 200 nmol lipid-P/ mg membrane protein) remained relatively constant throughout the time course of activation. Induction of protein N-glycosylation and membrane protein synthesis in activated B cells. To correlate the rates of protein N-glycosylation with the synthesis of new membrane proteins, B cells were pulse-labeled for 30 min at various stages in the activation process with either [2,3,4,5-3H]leucine (0) or [2-3H]mannose (0). As seen in Fig. 2, the rate of membrane protein synthesis rose sharply within 10 h after exposure to LPS, reaching a maximum at 35 h. Since the incorporation of labeled leucine into membrane protein was inhibited at least 85% by the addition of 2 PM cycloheximide (data not included), protein synthesis within the mitochondria could account for only a minor fraction of the metabolic labeling. On the basis of the relative abundance of ER seen in the electron micrographs after 24 h of exposure to LPS, it is also likely that polypeptides synthesized in the cytoplasmic compartment and subsequently incorporated into mitochon-

ET

AL.

dria would represent only a small fraction of the leucinelabeled membrane proteins. The induction of protein N-glycosylation was slightly delayed relative to membrane protein synthesis. Activity was maximal at approximately 60 h after addition of LPS, and both biosynthetic rates declined, but remained high relative to resting cells, between 60 and 84 h of activation. The induction pattern for dolichol-bound oligosaccharide formation was virtually coincident with N-linked glycoprotein biosynthesis (6). Induction of protein N-glycosylation and phosphatidylcholine biosynthesis in activated B cells. To compare the induction pattern of protein N-glycosylation with the rates of phospholipid biosynthesis, B cells were pulselabeled at various times of exposure to LPS with [23H]mannos e o r [ CH3-3H]choline. PC was chosen for this metabolic labeling experiment because it is the major membrane phospholipid. As can be seen in Fig. 3, the labeling of PC was stimulated fairly soon after the addition of LPS, but its rate of synthesis increased more slowly than the rate of synthesis of N-linked glycoproteins. The rate of PC synthesis continued to rise throughout the activation process. The largest increase in the rate of PC synthesis was seen between 36 and 90 h of exposure to the polyclonal mitogen. In this experiment protein N-glycosylation peaked at an earlier time than in the comparison depicted in Fig. 2. While variations in the developmental patterns for protein N-glycosylation occur

z k g 9

12

100 [%I JGLYCOPROTEIN

TIME

OF EXPOSURE

TO LPS

(h)

FIG. 3. Temporal relationship between the induction of protein Nglycosylation (0) and phosphatidylcholine biosynthesis (0). B cells (2 X lo6 cells/ml) were cultured in the presence of LPS (50 jig/ml). At the indicated times, cells were removed from culture, metabolically radiolabeled with either [2-3H]mannose (30 min) or [CH3-aH]choline (60 min) at 37”C, and the incorporation into either glycoprotein or PC was assayed as described under Materials and Methods. All data are average values of duplicate analyses and are representative of three separate experiments.

PROTEIN

N-GLYCOSYLATION

occasionally with different B cell preparations, the maximal rate of protein N-glycosylation consistently preceded the major increase in PC labeling. Effect of DZ-SIBA on the incorporation of labeled choline into phosphatidylcholine. To demonstrate that the metabolic labeling of PC is a reliable measure of de nouo synthesis, the effect of DZ-SIBA, an inhibitor of CDP-choline synthesis (35), on the rate of incorporation of [CHs3H]choline (0) into the phospholipid was assessed. DZSIBA produced a concentration-dependent inhibition of the metabolic labeling of PC in activated B cells (Fig. 4). While 500 PM DZ-SIBA suppressed the labeling of PC by 95%, it did not appreciably affect the incorporation of labeled inositol (A) or ethanolamine (0) into their respective phosphatidyl esters. When 200 yM DZ-SIBA reduced the incorporation of [CH,-3H]choline into CDPcholine by 74%, there was a corresponding 64% reduction of labeling of PC, without affecting the incorporation into phosphocholine (Table I). The stepwise methylation of PE does not appear to be a quantitatively significant pathway for PC biosynthesis in the B cell system since virtually no label was incorporated into PC when cells were incubated with [CH,-14C]methionine. Moreover, it is unlikely that significant amounts of labeled choline are incorporated into PC by an exchange mechanism because no radiolabeled PC was detected after B cell membrane preparations were incubated with [CH3-3H]choline in the presence of Ca*+. All of these results are consistent with

Jt

PHIIf--

Y

I

‘1/ 60-

30~H]CHOLI

NE \

O*0

250

500 [DZ-SIBA]

750

*

IO00

(PM)

FIG. 4. Effect of DZ-SIBA on the biosynthesis of PC (O), PE (O), and PI (a) in B cells. B cells were cultured in complete medium in the presence of LPS (50 pg/ml). After 72 h, cells were removed from culture and metabolically labeled for 1 h at 37°C with [CH,-3H]choline, [1,2“Clethanolamine, or [2-3H]inositol and increasing concentrations of DZ-SIBA. The incorporation of the isotopic precursors into phospholipid was assayed as described under Materials and Methods. Control values expressed as pmol/h/lOs cells were: PC = 39.4, PE = 2.6, and PI = 1.5.

IN

67

B CELLS TABLE

Effect

I

of DZ-SIBA on the Incorporation of [ 3H]Choline into Phosphocholine,CDP-Choline, and PC [3H]Choline

Addition to culture

Phosphocholine

CDP-choline (cpm

None 200 PM DZ-SIBA

166 193 (116)

incorporated

X lo-s/lo6

into PC

cells)

6.9 1.6 (24)

46.1 15.8 (34)

Note. B cells were cultured in complete medium in the presence of LPS (50 pg/ml) for 72 h, and then incubated with [CH,-3H]choline (50 &i/ml) for 1 h at 37°C in the presence and absence of 200 PM DZSIBA. The amounts of radiolabeled choline incorporated into phosphocholine, CDP-choline, and PC were assayed as described under Materials and Methods. All data are average values from duplicate analyses and are representative of two separate experiments. The numbers in parentheses are the data expressed as percentages of control values.

the conclusion that [CH,-3H]choline is incorporated into PC in these studies primarily uia the CDP-choline pathway, and the initial rate is a reflection of de nouo phospholipid synthesis. Comparison of the developmental pattern for choline kinase, CDP-choline synthetase, and CPT activities with the induction of phosphatidylcholine biosynthesis. To investigate further the developmental increase in the rate of synthesis of PC, the levels of the enzymes involved in the biosynthetic pathway were assessedin vitro at various times during the activation process. While choline kinase (A) activity remained relatively constant, the free cytosolic form of CDP-choline synthetase (0) began to rise as early as 10 h after the addition of LPS, increasing twofold during the first 60 h of LPS treatment (Fig. 5). Thus, the higher level of CDP-choline synthetase activity could contribute to the initial rise in the rate of PC synthesis (Fig. 5B). The translocation of cytosolic CDP-choline synthetase activity to a membrane-bound form was not detected in this study. CPT (0) activity did not change significantly during the first 30 h, but increased markedly from 30 to 90 h of activation during the period when the rate of PC synthesis increased approximately threefold. The rates of PC synthesis from Fig. 3 are presented in Fig. 5B for direct comparison with the changes in the pertinent enzyme activities. On the basis of these results it appears that the largest increase in the rate of PC synthesis and the induction of CPT, an ER-associated enzyme, occurs several hours after the major developmental increase in protein N-glycosylation. Kinetic analysis of developmental changes in CDP-choline synthetase and CPT activity during activation of B cells. A kinetic analysis of the substrate concentration dependencies for CDP-choline synthetase indicated that the amount of enzyme in the cytosolic compartment had

68

RUSH

A

60 -

ET

AL.

DISCUSSION Previously, we have shown that there are dramatic increases in the rates of dolichol-linked oligosaccharide intermediate synthesis and protein N-glycosylation when B cells are activated by LPS, anti-immunoglobulin M, or the combination of phorbol ester and ionomycin (6, 7). Since the lipid intermediate pathway for protein N-glycosylation is located in the rough ER (13-18), it was of interest to compare the patterns for the induction of protein N-glycosylation and other ER-associated biosynthetic processes involved in membrane biogenesis. The

20-

#’ 50-

pi]

A

.O

PC //

rPPARENl

Km =

p’ ,

i

25 -

.O’ ,O’# 0’

-0-O’

Otl 0

TIME

25 50 OF EXPOSURE

75 TO LPS (h)

0.04 11 [CD+CHOLINE]

IOC

FIG. 5. Comparison of the developmental patterns for choline kinase (A, A), CDP-Choline synthetase (0, A) and diacylglycerol cholinephosphotransferase (0, A) activities with the time course for the induction of PC biosynthesis (0, B). Zn vitro enzyme activities were determined from appropriate subcellular fractions prepared as described under Materials and Methods. The data from Fig. 4 for the induction of PC synthesis are also presented in B for direct comparison with enzyme activities in A.

increased but that the apparent K,,, values for CTP and phosphocholine had not changed (data not presented). To determine if the developmental increase in CPT activity was due to the presence of larger amounts of enzyme or the formation of a more active enzyme with altered affinity for its substrates, a similar kinetic analysis was conducted with membrane fractions from resting (0) and activated (0) cells (Fig. 6). From the double-reciprocal plots it can be seen that although the V,,, was sixfold higher in membrane preparations from activated cells, the apparent K,,, values for both substrates were similar if not identical in the resting and activated cells. The simplest conclusion from these kinetic analyses is that activated cells contain a sixfold higher level of the ERassociated protein.

(uhf-‘)

El

’5-

APPARENT

RESTING

Km= Il4pM

ACTIVATED

110

CELLS

-a05

0

CELLS

005 I/ [DIOLEIN]

0.K (#+I-’

1

FIG. 6. Kinetic analysis of the developmental increase in CPT activity in membrane fractions from resting B cells (0) and B cells that have been exposed to LPS for 72 h (0). CDP-choline concentration was varied in the presence of 200 pM CDP-choline. The procedures for the preparation of membranes and the assay of CPT are described under Materials and Methods. l/V is expressed as (nmol/min/10~6 cells)-‘.

PROTEIN

N-GLYCOSYLATION

development of the ER in activated B cells is clearly pertinent to the induction of the lipid-mediated mechanism for the N-glycosylation of membrane-bound and secretory immunoglobulins, and possibly MHC-restricted endogenous antigen presentation. By electron microscopy it could be seen that resting cells used in these studies are occupied predominantly by a relatively large nucleus and contain low numbers of other cytoplasmic organelles. The active development of the rough ER begins within 24 h after LPS is added to the purified B cell cultures. Similar observations have been reported by Shohat et al. (5). Indicative of an early response in membrane biogenesis, the cellular content of membrane protein and phospholipid increases within 12 h after the cells are exposed to the polyclonal mitogen. It is, therefore, not surprising that the rate of membrane protein synthesis also increases significantly during this early stage in the proliferative response to LPS. Since 85% of the leucine labeling of membrane protein was sensitive to cycloheximide, it is reasonable to conclude that the metabolic labeling primarily reflects protein synthesis initiated in the cytoplasm and completed at the ER and other potential membrane insertion sites. The results of the comparative study on developmental changes in the rates of membrane protein and glycoprotein synthesis suggest that some proteins are inserted during an early stage in the assembly of the ER prior to the full expression of lipid-mediated protein N-glycosylation. This is not totally unexpected because at least 20 ER proteins are required to complete the biosynthesis and transfer of dolichol-linked oligosaccharide intermediates. The rate of PC synthesis was evaluated by metabolic labeling with radiolabeled choline at different periods during B cell activation. Since De Blas et al. (35) have shown DZ-SIBA blocks PC synthesis by inhibiting the formation of CDP-choline in NGlOB-15 cells, the effect of this inhibitor was tested. The inhibitory effect of the adenosine derivative on the metabolic labeling of PC by [CH,-3H]choline in B cells supports the conclusion that the isotopic precursor is incorporated predominantly via the de nouo pathway. On the basis of metabolic-labeling with [CH3-3H]choline, the biosynthesis of PC, the major membrane phospholipid, increased steadily for 90 h after B cells were treated with LPS. The major increase in PC synthesis was seen after protein N-glycosylation reached maximum levels and began to decline. Related studies on mixed lymphocyte preparations derived from a variety of organs (36-41) have revealed increases in PC synthesis following exposure to several different mitogens. The stimulatory effect of a combination of pokeweed mitogen and T lymphocytes on PC synthesis in purified peripheral blood B cells has also been documented (42). Considering that the ratio of membrane protein/phospholipid was essentially constant throughout B cell development, it was anticipated that there would be coordinate increases in the rates of synthesis of the two membrane components. Surprisingly, while the rate of

IN

B CELLS

69

synthesis of membrane protein and phospholipid both increase soon after exposure to LPS, the developmental changes are neither concurrent nor proportional during the proliferative response to LPS. These apparent discrepancies could be explained by differential developmental changes in the rates of turnover of the two membrane components. It is also plausible that the apparent gradual increase in the rate of incorporation of labeled choline into PC is due to a decrease in the pool size of phosphocholine, affecting the intracellular specific activity of the isotopic precursor. Because developmental changes in the intracellular pools of biosynthetic intermediates could affect the specific activities of isotopic precursors, the results of metabolic labeling experiments must be interpreted cautiously. In this regard, calculations of the rates of protein N-glycosylation based on the specific activity of the intracellular pool of GDP-[3H]mannose indicate that the maximum rate of synthesis may actually be slightly later than the peak observed by following the incorporation of [2-3H]mannose (Rush and Waechter, unpublished observation). An in vitro analysis of the enzymes involved in the de nouo pathway was conducted to understand the possible regulatory mechanisms responsible for the enhanced rate of synthesis of PC in LPS-activated B cells. Choline kinase activity was relatively constant during the activation period and probably does not play a regulatory role in B cells. The cytosolic form of CDP-choline synthetase increased almost immediately after activation until 60 h and could contribute to the increase in PC biosynthesis, as proposed in other systems (43-46). The B cell cytidylyltransferase is apparently not activated by translocation to membrane-binding sites as observed in other mammalian cells (47-49). Watkins and Kent (50) have also reported that there is no appreciable translocation of the cytidylyltransferase when the incorporation of labeled choline into PC is increased fivefold by phorbol ester in HeLa cells. There was an impressive developmental increase in CPT, an ER-associated enzyme, between 30 and 95 h after exposure to LPS. Since the rate of PC synthesis increases prior to the major induction of CPT, it is probably not responsible for the stimulation of PC synthesis during the early stages of activation, but could contribute to the enhanced rate observed during the later period of the activation process. The results of a kinetic analysis are consistent with the conclusion that there is a sixfold increase in the amount of CPT in the ER compartment of activated B cells. It is interesting that the induction of CPT, an ER protein, occurs considerably after the major increase in lipid-mediated protein N-glycosylation, another ERassociated process. This study suggests that not all of the resident membrane proteins are induced synchronously during the development of new rough ER in B cells. Work by Green and co-workers (51-53) also indicates that there is an ordered expression of proteins during the development of

70

RUSH ET AL.

the ER in activated murine splenic B cells, and that the proteins are synthesized at different rates. More recently in an interesting study, Wiest et al. (54) have reported that the majority of rough microsomal proteins, including BiP and ribophorins I and II, increased proportionally to the size of the rough ER in LPS-stimulated CH12 cells, a murine B cell line. However, it was not determined if the time courses for the induction of individual proteins were coincident during the expansion of the ER in these differentiating cells. In summary, these results provide evidence that the ER proteins involved in the lipid intermediate pathway for protein N-glycosylation, membrane protein, and phospholipid synthesis are not all induced concurrently during the activation of murine splenic B cells. A future aim will be to determine if all of the enzymes synthesizing dolichol-linked saccharide intermediates are induced coordinately in activated B cells. ACKNOWLEDGMENTS The authors gratefully Bruce Maley and Mary presented in this paper.

acknowledge the valuable assistance Gail Engle in the electron microscopy

of Dr. studies

REFERENCES 1. Andersson, J., Melchers, F., Galanos, C., and Luderitz, 0. (1973) J. Ezp. Med. 137, 943-953. K. (1983) Proc. N&l. 2. Raetz, C. R. H., Purcell, S., and Takayama, Acad. Sci. USA 80, 4624-4628. 3. Melchers,

F., and Andersson,

4. Shands, J. W., Peavy, 70,1-24. 5. Shohat, M., Janossy, Immunol. 3,680-687.

J. (1984)

Cell 37,

D. L., and Smith,

G., and Dourmashkin,

6. Rush, J. S., Snow, E. C., and Waechter, Biophys. 259, 567-575. 7. Rush, J. S., and Waechter, 267,305-316. 8. Palade, G. E. (1975) Science 9. Sabatini, D. D., Kreibich, J. Cell Biol. 92, l-22. Pfeffer, S. R., and Rothman, 829-852.

11

Rose, J. K., and Doms, 288.

12. Bishop, 610.

W. R., and Bell,

C. J. (1987) Arch.

Arch.

Biochem.

J. Path& Eur.

J.

Eiochem. Biophys.

T., and Adesnik,

J. E. (1987)

Annu.

M. (1982)

Reu. Biochem.

56,

R. W. (1988)

Annu.

Reu. Cell Biol.

4, 257-

R. M. (1988)

Annu.

Rev. Cell Biol.

4,579-

13. Czichi, U., and Lennarz, W. J. (1977) J. Biol. Chem. 252, 79017904. 14. Nilsson, 0. S., DeTomas, M. E., Peterson, E., Bergman, A., Dallner, G., and Hemming, F. (1978) Eur. J. Biochem. 89, 619-628. 15. Ravoet, A. M., Amar-Costesec, A., Godelaine, D., and Beaufay, H. (1981) J. Cell Biol. 91, 679-688. 16. Eggens, I., and Dallner, G. (1982) Biochim. Biophys. Acta 686,7793. G. H., and Waecbter, C. J. (1984) Arch. 17. Scher, M. G., DeVries, Biochem. Biophys. 23 1, 293-302. 18. Hirschberg, C. B., and Snider, M. D. (1987) Anna Reu. Biochem. 56,63-87.

J. G., Biddison, 343,74-76.

W. E., and Klausner,

R. D. (1990)

Nature

20. Cox, J. H., Yewdell, J. W., Eisenlohr, L. C., Johnson, P. R., and Bennink, J. M. (1990) Science 247, 715-718. 21. Long, E. 0. (1989) Zmmunol. Today 10,232-234. 22. Rush, J. S., Sweitzer, T., Kent, C., and Waechter, C. J. (1988) J. Cell Eiol. 107, Abstr. No. 1095. 23 Waechter, C. J., Schmidt, J. W., and Catterall, W. A. (1983) J. Biol. C/rem. 258,5117-5123. 24 Folch, J. M., Lees, M., and Sloane-Stanley, G. H. (1957) J. Eiol. Chem. 226,497-509. 25 Waechter, C. J., and Lester, R. L. (1973) Arch. Biochem. Biophys.

158,401-410. 26. Weinhold, 5141.

P. A., and Rethy,

V. B. (1974)

29. Steiner,

S. M., and Lester,

30. Wagner, 175-184.

H., Horhammer,

31. Dittmer,

J. C., and Lester,

R. L. (1972)

Arch. Eiochem. J. Biol.

P. (1961)

R. L. (1964)

J. Lipid

Chem.

109, 81-68.

J. Eacteriol.

L., and Wolff,

13, 5135-

Biockmistry

27. Percy, A. K., Moore, J. F., and Waechter, C. J. (1984) Biophys. 235, 18-25. 28. Wright, P. S., Morand, J. N., and Kent, C. (1985) 260, 7919-7926.

Eiochem.

Z. 334,

Res. 5, 126-127.

32. Waechter, C. J., Steiner, M. R., and Lester, R. L. (1969) J. Eiol. Chem. 244, 3419-3422. 33. Bartlett, G. R. (1959) J. Eiol. Chem. 234, 466-468. 34. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Eiol. Chem. 193, 265-275. 35. De Blas, A., Adler, A., Shih, M., Chiang, P. K., Cantoni, G. L., and Nirenberg, M. (1984) Proc. Nutl. Acad. Sci. USA 81, 4353-4357. 36. Fisher, D. B., and Mueller, G. C. (1968) Proc. Natl. 60, 1396-1402. 37. Kay, J. E. (1968) Nature (London) 219, 172-173. 38. Fisher,

347-358.

G., Morimoto,

10

Amer.

R. R. (1973)

C. J. (1988) 189,

715-720.

R. T. (1973)

19. Nuchtern, CLondon)

D. B., and Mueller,

G. C. (1969)

Acad.

Eiochim.

Sci. USA

Biophys.

Acta

176,316-323. 39. Resch, K., and Ferber, E. (1972) Eur. J. Eiochem. 27, 153-161. 40. Resch, K., Gelfund, E. W., Hansen, K., and Ferber, E. (1972) Eur. J. Immunol. 2,598-601. 41. Ferber, E., de Pasquale, Actu 398,364-376.

G. C., and Resch, K. (1975)

Eiochim.

Biophys.

130,256%2573.

42. Chien, M. M., and Ashman, R. F. (1983) J. Zmmunol. 43. Sleight, R., and Kent, C. (1980) J. Eiol. Chem. 255,

10,644-10,650.

44. Weinhold, P. A., Feldman, D. A., Quade, M. M., Miller, J. C., and Brooks, R. L. (1981) Biochim. Biophys. Actu 665, 134-144. 45. Post, M., Batenburg, J. J., Schuurmans, E. A. J. M., and Van Golde, L. M. G. (1982) Eiochim. Eiophys. Acta 712, 390-394. 46. Pelech, S. L., and Vance, 217-251. 47. Sleight, 48. Sleight, 49. Vance, 50. Watkins,

D. E. (1984)

Biochim.

Biophys.

Acta 779,

R., and Kent,

C. (1983)

J. Eiol.

Chem.

258,831-835.

R., and Kent,

C. (1983)

J. Eiol.

Chem.

258,836-839.

D. E., and Pelech, S. L. (1984) J. D., and Kent,

51. Lewis, M. J., Mazzarella, 260,3050-3057. 52. Lewis, M. J., Turco, 6926-6931.

C. (1990)

J. Biol.

R. A., and Green,

S. J., and Green,

53. Lewis, M. J., Mazzarella, Biophys. 245, 389-403.

Trends

E&hem. Chem.

M. (1985)

M. (1985)

R. A., and Green,

54. Wiest, D. L., Burkhardt, J. K., Hester, D. I., and Argon, Y. (1990) J. Cell Eiol.

Sci. 9, 17-20. 265,2190-2197. J. Biol. Chem.

J. Biol. Chem.

M. (1986)

Arch.

S., Hortsch,

110, 1501-1511.

M.,

260,

B&hem. Meyer,