Alteration of Hybridoma Viability and Antibody Secretion in

Biotechnol. Prog. 1995, 1 1, 565-574

565

Alteration of Hybridoma Viability and Antibody Secretion in Transfectomas with Inducible Overexpression of Protein Disulfide Isomerase? Kirsten Kitchid and Michael C. Flickinger*,* Department of Biochemistry and Institute for Advanced Studies in Biological Process Technology, University of Minnesota, St. Paul, Minnesota 55108-6106

Monoclonal antibody (mAb)-secreting transfectomas with dexamethasone inducible expression of the mammalian endoplasmic reticulum foldase and chaperone protein disulfide isomerase (PDI, ERp59) were generated from the murine 9.2.27 hybridoma in order to obtain in vivo evidence of whether alteration of the level of PDI, believed to be involved in immunoglobulin (Ig) assembly, results in alteration of mAb secretion kinetics. Using a n RNase refolding assay, the specific activity of endogenous PDI in the 9.2.27 hybridoma was found to be constant during batch growth. An expression vector for glucocorticoid-inducibleoverexpression of PDI, pMMTVPDI, was constructed from pMAMneo using a rat PDI cDNA. Cell lysates of stable transfectomas contained 2-4-fold higher levels of PDI mRNA and increased levels of PDI protein, detected by immunoblotting, following induction with 0.1 pM dexamethasone. Monoclonal antibody secretion kinetics were evaluated in 12.5 mL shake flasks, a 100 mL spinner, and a 1 L aerated batch reactor. A transfectoma was found with altered mAb secretion kinetics during cell growth following dexamethasone induction of PDI overexpression. Specific mAb secretion rate was not significantly increased following dexamethasone induction; however, hybridoma viability was sustained longer during the stationary phase of cell growth and hence total antibody yield was increased in comparison t o the parent 9.2.27 hybridoma.

Introduction In recent years, optimization of large-scale culture of lymphoid mammalian cells by increasing either cell density or specific secretion rate has become of increasing importance for efficient production of antibodies and other recombinant proteins. Improvements have been made both in design of expression systems, notably the glutamine synthetase amplifiable system (Bebbington et al.,1992; Brown et al.,1993) and in development of a variety of high cell density bioreactors (Bliem et al., 1991; de la Broise et al., 1992). An understanding of the protein secretory pathway in lymphoid cells at the molecular level is important for future improvement in production of proteins from hybridoma and myeloma cells (Edgington, 1992). Despite a growing body of knowledge on the constitutive pathway of protein secretion in a variety of mammalian cells (Rothblatt et al., 19941, much is still not understood about the post-translational factors which control the assembly and folding of immunoglobulins in the endoplasmic reticulum (ER) and how regulation of the ER environment could be altered to increase secretion rate and yield (Tuite and Freedman, 1994). + Portions of this work were presented at the Third Engineering Foundation Conference on Cell Culture Engineering, Palm Coast, FL, May 1992,and the Fourth Engineering Foundation Conference on Cell Culture Engineering, San Diego, CA, March 1994. Institute for Advanced Studies in Biological Process Technology. Present address: R & D Systems, Inc., Minneapolis, MN 55413. 9 Department of Biochemistry and Institute for Advanced Studies in Biological Process Technology. * Corresponding author: telephone, (612)625-2782;FAX, (612) 625-1700.

*

A variety of approaches are being investigated for altering the rate of protein folding/assembly and secretion from the ER. One approach involves developing an understanding of physiochemical conditions within the ER such as the redox environment (Hwang et al.,1992; Braakman et al., 1992) and ATP concentration (Clairmont et al., 1992; Mayinger and Meyer, 1993). Much research has gone into defining the conditions and identification of ER proteins which influence protein folding in vitro (Bardwell and Beckwith, 1993; Rothblatt et al., 1994). Valuable in vivo evidence of the importance of specific ER proteins on the regulation of folding and assembly of proteins destined for secretion can be obtained by altering levels of those proteins in the ER which have been implicated by in vitro evidence to play a role in protein folding or polypeptide chain assembly. The two major classes of ER proteins involved in polypeptide folding or disulfide bond formation are chaperones and foldase enzymes (Gething and Sambrook, 19921, although recent evidence suggests that these classes are not mutually exclusive (Wang and TSOU,1993; Lamantia and Lennarz, 1993; Puig and Gilbert, 1994). Chaperone proteins which are present in the ER in concentrations stoichiometric to newly synthesized proteins include members of the heat shock or stress protein families which function to prevent misfolding or aggregation of newly synthesized proteins within the ER. Two important members of this family are the immunoglobulin heavy chain binding protein (BiP, glucose-regulated protein, grp78) (Haas and Wabl, 1983; Munro and Pelham, 1986) and grp 94 (ERp99, p100, endoplasmin) (Mazzarella and Green, 1987; Koch et al., 1988). Both grp94 and BiP are induced by physiological stress such as glucose starvation, calcium ionophore treatment, or

8756-7938/95/301I -0565$09.00/0 0 1995 American Chemical Society and American Institute of Chemical Engineers

566

addition of glycosylation inhibitors (Lee, 1987). An in vivo example of the effect of alteration of BiP levels in Chinese hamster ovary cells on several secreted proteins (Dorner et al., 1988) suggests that, when the activity of this major mammalian ER chaperone was altered, protein folding became unbalanced and secretion could be either increased (Dorner et al., 1988) or decreased (Dorner et al., 1992). It is now believed that reduction of secretion with increased BiP levels may not be generally applicable to many secreted recombinant proteins (Edgington, 1992; Robinson et al., 1994). BiP has been shown to bind to immunoglobulin heavy chains (Haas and Wabl, 1983) as well as associate with Ig light chains (Knittler and Haas, 1992). A recent study shows that grp94 also binds to light and heavy Ig chains perhaps in concert with BiP although the significance of this is not known (Melnick et al., 1994). The other class of proteins which play a role in protein folding and assembly in the ER are foldases, which generally function i n vitro in catalytic amounts to accelerate protein folding and disulfide bond formation. Mammalian ER foldases include the multifunctional protein disulfide isomerase (PDI, ERp59, EC 5.3.4.1) (Freedman, 1989; Noiva and Lennarz, 1992) and ERp72 (Mazzarella et al., 1990; Urade et al., 1993; Srinivasan et al., 1993; Rupp et al., 1994; Nigam et al., 1994). Originally classified as a foldase, PDI is an abundant ER protein present in levels comparable to that of BiP (Zapun et al., 1992). Despite the carboxy-terminal KDEL ER retention signal (Pelham, 19901, secretion of PDI has been reported from exocrine cells CYoshimori et al., 1990). In vitro evidence has shown PDI to function as a thiol: disulfide exchange protein, catalyzing oxidation or reduction and hence isomerization of disulfide bonds in vitro (Hillson et al., 1984). The two thioredoxin-like domains of PDI, conserved in a wide variety of mammalian species and yeast (Lu et al., 1992; Farquhar et al., 1991; Lamantia et al., 1991; Tachikawa et al., 1991), function independently to catalyze disulfide bond formation and rearrangement reactions (Lyles and Gilbert, 1991; Hawkins and Freedman, 1991; Hawkins et al., 1991a; Freedman et al., 1994). Catalysis of disulfide bond formation appears to be essential for folding of disulfide-containing proteins in the mammalian ER (Bulleid and Freedman, 1988). PDI has been shown to have multiple functions as a foldase (Freedman, 1989; Bassuk and Berg, 1989; Gething and Sambrook, 1992; Noiva and Lennarz, 1992). PDI has at least one nonspecific peptiddprotein binding site which may be involved in binding of unfolded proteins and polypeptides (Morjana and Gilbert, 1991; Noiva et al., 1991) resulting in PDI having chaperonelike activity (Wang and TSOU,1993; Puig and Gilbert, 1994). This protein has recently been shown to function in vitro as a chaperone by accelerating the folding of ~-glyceraldehyde-3-phosphate dehydrogenase, a protein with no disulfide bonds (Cai et al., 1994). Evidence for PDI binding and catalyzing immunoglobulin disulfide bond formation comes from demonstration of cross-linking of PDI to IgG and IgM in intact lymphocytes (Roth and Pierce, 1987) and correlations between increases in secretion of disulfide bond-containing proteins and PDI levels in a variety of tissues and cell types (Freedman, 1984). In this study we investigate i n vivo manipulation of the ER environment that might alter the folding and assembly of a monoclonal antibody (mAb) as an initial approach to design of lymphoid cells with enhanced secretion. The stable, high mAb secreting murine hybridoma cell line 9.2.27, which produces an antibody of the IgGz, subclass against a human melanoma antigen

Biotechnol. Prog., 1995,Vol. 11, No. 5 (Morgan et al., 19811, is used as a host cell line for generation of transfectoma clones with inducible overexpression of PDI. A structured kinetic model has been developed (Bibila and Flickinger, 1992a,b; Flickinger et al., 1992) which describes the flux of the antibody secretory pathway of this hybridoma as a function of growth rate. This model confirms that antibody folding, assembly, or the rate of exit from the ER may be the overall rate-limiting step for antibody secretion during exponential growth. Our in vivo approach to posttranslational manipulation of antibody assembly/folding, or rate of exit from the ER, is to increase the levels of PDI since correlative data linking increases in PDI levels to increases in antibody secretion from lymphoid cells has been reported (Roth and Koshland, 1981; Paver et al., 1989; Kronig et al., 1991). By creating stable 9.2.27 transfectomas capable of inducibly overexpressingprotein disulfide isomerase without PDI secretion, in vivo evidence confirming a correlation between PDI levels and antibody secretion may be obtained which may be applicable to improving secretion of antibodies or disulfide bonded recombinant proteins from other lymphoid cell lines.

Materials and Methods Cell Culture. The 9.2.27 cell line was grown either in Dulbecco's Modified Eagle's medium (DMEM)containing 4.5 g/L glucose and 4 mM L-glutamine (JRH Biosciences, Lenexa, KS) supplemented with 5%fetal bovine serum (FBS) (Hyclone Laboratories, Logan, UT) and 25 mM Hepes (Research Organics, Cleveland, OH)or in a protein-free hybridoma medium (PFHM 11)(Gibco-BRL, Grand Island, NY). The P3X63-Ag8.653mouse myeloma cell line (Kearney et al., 19791, the nonsecreting parent of the 9.2.27 hybridoma, was grown in RPMI 1640 supplemented with 10% heat-inactivated FBS, in T flasks. All cells were grown in a humidified atmosphere supplemented with 5% COZat a temperature of 37 "C. Cell counts and viability determinations were made using 0.4% erythrosin B as a vital stain and a hemocytometer (Kruse and Patterson, 1973); an average error of f1015% was typical for these measurements. Measurement of Secreted Antibody. Secreted antibody concentration in the culture medium was measured from 0.2-1.0 mL samples removed from flask or bioreactor cultures and centrifuged at lOOOg for 3 min. The cell-free culture media was stored at -20 "C prior to assay by ELISA (Abrams et al., 1984). Specific antibody secretion rate was calculated using the differential method (Ozturk and Palsson, 1991). The first and last samples taken on each growth curve were not used for calculation of specific antibody secretion rate. Transfectoma clones grown in 12.5 mL shake flask cultures in 5%FBS DMEM containing 0.25 mg/mL G418 (Gibco-BRL) were used to inoculate duplicate, 100 mL spinner cultures (25 rpm) or 12.5 mL shake flask cultures (75 rpm) at a density of 1.5 x lo5 viable celldml for growth in 5% FBS DMEM without G418. Dexamethasone (Sigma Cell Culture, St. Louis, MO) was dissolved in water to produce a 100 pM stock solution and stored in aliquots at -20 "C. For induction, dexamethasone was added to cultures at a final concentration of 1 x lo-' M approximately 20 h after inoculation. Transfectomas were also grown in a 1.0 L glassjacketed bioreactor (Applikon, Foster City, CAI, sparged continuously with COZ, air, or 02 (to maintain the level of dissolved oxygen at 20.03 atm) and equipped with a low-shear four-blade impeller as previously described (Flickinger et al., 1990) but without an internal mem-

567

Biotechnol. Pfog., 1995,Vol. 11, No. 5

brane oxygenator. A working volume of 1.0 L was used with an agitation rate of 150 rpm. Dissolved oxygen was monitored using a galvanic probe, and pH was maintained by buffer between pH 6.9 and 7.4. Fermentation medium consisted of 5% FBS DMEM, 4 m M L-glutamine, 4.5 g/L D-glucose, and 25 mM Hepes. Dexamethasone induction was carried out by dilution of 1.0 mL of M dexamethasone in 10 mL of media which was added through the bioreactor feed line resulting in a final concentration of lo-' M a t a cell density of 4-5 x lo5 viable cells/mL. Protein Disulfide Isomerase Activity Assay. The activity assay used was based on the procedure of Myllyla and Oikarinen (1983). Standard procedures were used for creating a ribonuclease-free environment in which to prepare reagents and run the assay (Sambrook et al., 1989). Purified PDI was prepared from fresh frozen bovine liver according to the methods of Lambert and Freedman (1983) and Hillson et al. (1984). Commercially available purified bovine PDI (Takara Biochemicals, Berkeley, CA) was used as standard for the PDI activity assay. Samples of 2 x lo6 viable 9.2.27 hybridoma cells were removed from culture and centrifuged at 600g for 5 min. The cell pellet was washed twice with 10 mL of PBS and either used immediately or frozen at -20 "C for up to 2 weeks. The cell pellet was resuspended in 0.5 mL of 0.05 M Tris-HC1 (pH 7.51, and the cells were lysed either by three cycles of -70 to 37 "C freeze-thawing or by using 50 strokes of a 1.0 mL ground glass homogenizer (Wheaton, Millville, NJ). The cell lysate was centrifuged in an Eppendorf centrifuge (Brinkmann Instruments, Westbury, NY)a t 14 000 rpm for 30 min (4 "C, 16 OOOg). The crude postnuclear supernatant (PNS), was assayed immediately or frozen overnight a t -80 "C. Aliquots of the PNS were assayed for protein (Bio-Rad Laboratories, Hercules, CA) with bovine serum albumin (BSA, Sigma Chemical, St. Louis, MO) as the standard. Randomly reoxidized ribonuclease (sRNase) was prepared according to Myllyla and Oikarinen (19831, lyophilized, gassed with nitrogen, and stored a t -80 or 4 "C. Before use, sRNase was reconstituted by adding ribonuclease-free water to a concentration of 1 mg/mL and stored in 20 pL aliquots at -20 "C. These aliquots were thawed immediately prior to use and never refrozen. Labeled RNA was prepared by feeding 9.2.27 cells with [5-3H]cytosinetriphosphate (CTP), a modification of the method of Myllyla and Oikarinen (1983). A 100 pCi amount of [5-3H]CTP (Dupont-NEN, Wilmington, DE) was fed to a 10 mL culture containing 1.5 x lo7 exponentially growing 9.2.27 cells. M e r 12 h of incubation, cells were harvested and total cytoplasmic RNA was extracted using the RNAzol B reagent (Tel-Test, Inc., Friendswood, TX). The specific activity of the labeled RNA was measured in Ecolume (ICN, Costa Mesa, CA) using a LS 7000 scintillation counter (Beckman Instruments, Fullerton, CAI. Labeled RNA had a specific activity of 7 x lo7 dpmlmg and was stored in aliquots at -20 "C. The RNA was diluted with 0.05 M Tris-HC1 prior to use. Crude PNS containing 1-2 pg protein was preincubated with 0.5 pg of sRNase in 0.05 M Tris-HC1 (pH 7.5) and 2 mM EDTA at 30 "C for 30 min in a total volume of 100 pL. A 10 mL sample of this preincubation mixture was transferred to a tube containing 90 p1 of a mixture containing 120 pg of yeast RNA (Boehringer Mannheim, Indianapolis, IN) as carrier and 0.2 pg (14 000 dpm) of [5-3HlCTP RNA in 0.05 M Tris-HC1 (pH 7.5). This mixture was incubated a t 30 "C for 10 min. The reaction tube was placed on ice and the reaction stopped by

addition of 200 pL of 15%(w/v) trichloroacetic acid and 12 pL (60 pg) of yeast RNA, After 30 min the sample was centrifuged at 14 000 rpm (16,OOOg) for 20 min in an Eppendorf centrifuge (4 "C). Then 200 pL of the supernatant was added to 1.5 mL of water and 10 mL of ecolume (ICN) and counted. PDI activity was expressed as dpm per milligram of total cell protein (assuming a constant counting efficiency of 50% for tritium). All activity levels were corrected for the reappearance in RNase activity observed when sRNase was incubated in the absence of cell lysate. Immunodetection of Protein Disulfide Isomerase. Bovine PDI purified in this laboratory was used to raise polyclonal antisera in chickens (Gassman et al., 1990) which was isolated from hen's egg yolks by the method of Polson (1990). This chicken anti-bovine PDI polyclonal detected 15 ng of purified bovine PDI (Takara) on a nitrocellulose membrane (Bio-Rad). The Nonidet-P40 (NP-40, Sigma) method of Dorner and Kaufman (1990) was used for 9.2.27 hybridoma cell lysis starting with a frozen cell pellet containing (5-10) x lo6 9.2.27 cells. An aliquot of cleared cell lysate was assayed for protein using the BCA protein assay (Pierce Chemical Co., Rockford, IL) with bovine serum albumin as the standard. Reducing, or nonreducing, denaturing SDS polyacrylamide gel electrophoresis was performed using the discontinuous buffer system of Laemmli (1970). Proteins were transferred to a nitrocellulose membrane, pore size 0.24 pm (Bio-Rad), using an electroblotter (Hoefer Scientific Instruments). Transfer was carried out at 75 V for 2 h in 25 mM Tris, 150 mM glycine, and 20% methanol. The membrane was blocked using 5% dried milk powder in Tris-buffered saline (TBS: 20 mM Tris-HC1, pH 7.5, 500 mM NaC1) for 1 h at room temperature or overnight a t 4 "C. The chicken anti-PDI antisera was diluted 1in 500 in antibody buffer (1X TBS, 0.05% Tween 20,1% gelatin) and incubated with the membrane for 3 h or overnight at room temperature. Following washing, a goat anti-chicken IgG (H L) alkaline phosphatase conjugate (Bio-Rad) was diluted 1 in 2000 in antibody buffer and incubated with the membrane for 1h a t room temperature. The blot was developed using a colorimetric alkaline phosphatase detection kit (Bio-Rad). Construction of PDI Expression Vector. The Escherichia coli cloning strain DH5aF' (Gibco-BRL)was cultured in LB medium (5 g/L yeast extract (Difco Laboratories, Detroit, MI), 10 g/L Bacto-tryptone (Difco), and 10 g/L sodium chloride). Rat PDI cDNA cloned into the Sca 1 site of pBR322 was obtained from plasmid ppdilOO (Edman et al., 1985). Standard molecular biology techniques were used during construction of the PDI expression vector pMMTVPDI (Sambrook et al., 1989; Ausubel et al., 1989) (Figure 1). Enzymes were purchased from Gibco-BRL. The mammalian expression vector pMAMneo (Clontech Labs, Palo Alto, CA) was used for construction of pMMTVPDI. pMAMneo contains a Rous sarcoma virus (RSV) enhancer element directly 5' to the MMTV LTR, SV40-derived splicing and polyadenylation signals immediately 3' of the inserted cDNA, and an expression cassette for aminoglycoside phosphotransferase (neo) which inactivates G418 (Southern and Berg, 1982). The SV40 early promoter drives expression of the neo resistance gene, and this promoter is active in cell lines from a wide variety of tissues and species (Banerji et al., 1981). The MMTV LTR functions as a glucocorticoid-responsive promoter (Lee et al., 1981) and can be induced by dexamethasone.

+

Biotechnol. Prog,, 1995, Vol. 11, No. 5

568

rind91 n

RSVMMTV h LTR

1 € 1

t

4000

o I 1560

3000

I

'ZI

3 2000

3 1000 SV40 (splicinglpolyad

Figure 1. Map of plasmid pMMTVPDI.

Rat PDI cDNA was excised from ppdilOO on a 2.3 kb Apa I-Sma I fragment and treated with T4 DNA polymerase to blunt the Apa I sticky end. The blunt-ended cDNA was gel-purified and annealed to phosphorylated Xho I linkers (New England Biolabs, Beverly, MA). The PDI cDNA with its Xho I sticky ends was ligated into Xho I-digested pMAMneo. The ligation mixture was used to transform E.coZi DH5aF'. Transformants which had a 2.3 kb insert in pMAMneo were identified first by digestion with Xho I, and the orientation of the PDI cDNA was confirmed by digestion with BamH I and subsequently with BgZ I1 and Hind I11 (Figure 1). Generation of 9.2.27 "ransfectomas. An alkaline lysis procedure coupled with a poly(ethy1ene glycol) (PEG) precipitation step was used for large-scale pMMTVPDI DNA preparation (Krieg and Melton, 1986). A known amount of supercoiled plasmid DNA was ethanol precipitated, air-dried in a vertical laminar flow biohood (Baker, Sanford, ME), then resuspended in sterile, deionized water. The 9.2.27 hybridoma was transfected using lipofection (Lipofectin, Gibco-BRL) using the manufacturer's recommendations and a procedure for T cell and macrophage lipofection (Dorman and Young, 1989). Exponentially growing 9.2.27 cells were centrifuged, rinsed in PFHM 11, and resuspended in prewarmed PFHM I1 at a concentration of 1.25 x lo6 viable celldml. The cell suspension was plated out at 0.4 muwell into 12 well tissue culture plates and incubated at 37 "C in 5% CO2. A DNALipofectin complex containing 25 pg/mL Lipofectin and 10 pglmL plasmid DNA was prepared according to the manufacturer's directions. A single 100 pl aliquot of the complex was added to each well. The plates were cultured without shaking for 9 h at 37 "C in 5% CO2, when 2.0 mL of 5%FBS DMEM were added to each well. Approximately 72 h after DNA-Lipofectin addition, transfected cells were diluted into selection medium [5% FBS DMEM supplemented with 1.0 mg/mL G418, 100 unitdml penicillin G, and 0.1 mg/mL streptomycin (Gibco-BRL)]at a viable cell concentration of 2 x lo5 per milliliter. When culture viability dropped to 20-40%, dead cells were removed using Ficoll-Paque (Pharmacia) density centrifugation and live cells resuspended in fresh selection media at a concentration of 2 x lo5 per milliliter. As soon as culture viability increased, generally after 2 weeks of selection, a final live cell enrichment was performed and a portion of the viable cells were subcloned by limiting dilution in selection medium. ARer 7-10 days, colonies of cells could be observed in some wells. These clones were fed fresh selection medium and expanded until sufficient cells were available for analysis or storage in liquid nitrogen. RNA Analysis of 9.2.27 Transfectomas Using Northern Blotting. For each transfectoma, a 10 mL shake flask culture was inoculated at a viable cell

I

0

I

0

I

I

0.04 0.08 0.12 0.16 PDI concentration (pM)

0.2

Figure 2. sRNase protein disulfide isomerase activity assay. Purified PDI (Takara) was used as the standard. Open and solid symbols indicate different assays under identical reaction conditions. The sRNase concentration in the assay mixture is 0.37 pM.

concentration of 2 x lo5 per milliliter in 5% FBS DMEM. In order to perform dexamethasone induction, cultures were grown in duplicate, and after 20 h, dexamethasone was added to one at a final concentration of 1 x M. In all cases cultures were counted and harvested in late exponential phase, approximately 50-60 h after inoculation. Each culture was divided in half and centrifuged. One cell pellet was used immediately for RNA extraction, while the other was stored at -20 "C for subsequent protein analysis. RNA extraction was carried out using the RNAzol B reagent (Tel-Test Inc.). The Genius 1DNA Labelling and Detection Kit (Boehringer Mannheim) was used for preparing doublestranded digoxigenin (DIG)-labeled DNA probes. The DNA template used was the 2.3 kb Xho I fragment from pMMTVPDI containing the PDI cDNA. The scaled-up random primed DNA labeling reaction was performed as described by the vendor (Boehringer Mannheim). Denaturing formaldehyde agarose gels were used for size fractionation of total RNA prior to Northern blotting (Ausubel et aZ.,1989). Each sample was loaded in duplicate so that half of the gel could be stained with ethidium bromide (0.5 mg/mL in 0.1 M ammonium acetate). The distance migrated by the 28S, 18sand 5s ribosomal RNA bands was measured. The other half of the gel was prepared for Northern transfer (Ausubel et aZ., 1989). Double-stranded digoxigenin (DIG)-labeled probes were used according to the manufacturers directions for Northern hybridizations (Boehringer Mannheim). Probe bound to the membrane was detected using the chemiluminescent alkaline phosphatase substrate Lumi-Phos 530 (Boehringer Mannheim). '

Results and Discussion Protein Disulfide Isomerase Activity in the 9.2.27 Hybridoma during Batch Culture. The assay of Myllyla and Oikarinen (1983) was chosen for measuring PDI activity because its ease and sensitivity made it potentially suitable for analysis of multiple, crude cell samples. When purified PDI (Takara) was used, sRNase refolding was reproducible and proportional to the PDI concentration in the refolding mixture (Figure 2). However adding large quantities (tens of milligrams) of protein to the assay in the form of PDI-containing cell lysate, or sRNase, inhibited RNase refolding. In order to minimize this nonlinearity and improve the reproducibility, care was taken to add equal amounts of total protein to the assay when measuring PDI activity in crude cell samples. Upon the basis of the work of Roth and Koshland (1981), it was expected that the Igsecreting 9.2.27 hybridoma should have higher PDI activity levels than its non-Ig-secreting parent myeloma P3X63-Ag8.653. Comparison of PDI specific activity was

Biofecbnd. Prog., 1995, Vol. 11, No. 5

1 0’

I

A

Tlme (hours)

Figure 3. Specific protein disulfide isomerase activities during batch growth of the 9.2.27 hybridoma in 5% FBS, DMEM. Similar hybridoma growth and PDI activity was also obtained in PFHM 11. (A) Cell growth: total cells, 0; viable cells, 0. (B) PDI specific activity. (C) Monoclonal antibody accumulation.

made in the parent myeloma and the 9.2.27 hybridoma; however, no significant difference in PDI levels was found (data not shown). PDI specific activity levels were monitored during batch growth of the 9.2.27 hybridoma and found to remain essentially constant through batch culture until culture degeneration in late stationary phase (Figure 3) (Kitchin, 1993). Similar results of constant PDI activity during batch culture were obtained when the 9.2.27 hybridoma was grown in protein free medium (PFI-IM11). This correlated with the non-growth-associated kinetics of antibody secretion by this hybridoma and contrasts with the up to 4-fold increase in PDI and prolyl 4-hydroxylase activity observed as fibroblasts entered the stationary phase during batch growth of the murine collagen-secretingfibroblast cell line 3T6 (Myllyla et al., 1983); however, this earlier study used a different PDI activity assay (Ibbetson and Freedman, 1976). Quantitative measurement of PDI activity using the assay of Myllyla and Oikarinen (1983) on 9.2.27 hybridoma lysates varied using the same batch of randomly refolded ribonuclease. A recent in vitro study by Puig and Gilbert (1994) of lysozyme refolding by purified PDI showed that PDI may function as either an antichaperone or a chaperone depending on the relative concentrations of PDI and lysozyme and their order of addition to the reaction mixture. Although the authors state that they did not observe similar effects during refolding of ribonuclease, purified PDI was used in their system. Our results suggest that the activity assay of

589

Myllyla and Oikarinen (1983)is linear with purified PDI; however, it can be very nonlinear when using crude cell lysates due to concentration-dependent effects of other proteins on PDI foldase activity. These difficulties may explain why there are few reports in the literature which quantify PDI activity levels in cultured cells. Inducible PDI Expression In 9.2.27 ’I’ransfectomas. Transfection of the 9.2.27 hybridoma with the PDI expression vector pMMTVPDI resulted in isolation of stable IgG-secreting transfectomas (PDI 1-14). The vedor pMAMneo, which lacked the PDI cDNA insert was also used to transfect the 9.2.27 hybridoma. Of these clones, Mneo 1was chosen as a negative control. The levels of endogenous PDI mRNA were examined in the 9.2.27 parent cell line. A hybridizing band was detected in each lane loaded with 9.2.27 RNA. The distance migrated by this RNA species corresponds to a size of approximately 3 kb which agrees with the size of murine PDI mRNA observed by Mazzarella et al. (1990). The detection of PDI appears quantitative, as the bands in lanes loaded with 10 pg of total RNA were twice as intense as those loaded with 5 pg of total RNA. The same hybridization conditions and PDI DNA probe were used to examine total RNA isolated from 9.2.27 transfectomas. Northern analyses of the untransfected 9.2.27 hybridoma, PDI 4, and PDI 14, all with and without dexamethasone induction, were performed (Figure 4B). Densitometry of the Northern blots indicated that clones PDI 4 and 14 show approximately a 2-fold increase in PDI mRNA following dexamethasone induction. Only one hybridizing band is seen for each sample. Although it might be expected that the recombinant rat PDI mRNA would be a different size from the endogenous mouse PDI “A, estimations of the size of the recombinant mRNA are 2.8-3 kb, so that the endogenous and recombinant PDI mRNA would be detected as a single band. Transfectomas Mneo 1,PDI 4 and PDI 14 were assayed for PDI protein levels, with or without dexamethasone induction, using Western blotting (Figure 5). PDI protein was detected in all the cell lysates running at a slightly lower molecular weight than the bovine PDI. In addition, the PDI antisera also bound to a high molecular weight species-possibly high molecular weight PDI aggregates (Hillson et al., 1984). There was no evidence for a dexamethasone-induced increase in PDI protein levels in Mneo 1 or PDI 14. There was a slight increase in PDI protein level when PDI 4 was induced with dexamethasone. Overexpression of PDI may result in PDI saturating the secretion retention system and leaking from the cell (Dorner et al., 1990). Samples of media were taken from exponentially growing dexamethasone-induced and uninduced cultures and assayed for secreted PDI using Western blotting. No evidence of PDI secretion was found (data not shown). Antibody Secretion Kinetics in ’I’ransfectomas with Inducible PDI Expression. Two transfectomas, PDI 4 and 14, which showed increased PDI mRNA following dexamethasone induction were chosen for analysis of antibody secretion kinetics. Mneo 1was used as a negative control. Transfectomas PDI 14 and Mneo 1 were grown in 100 mL spinner cultures, with and without dexamethasone induction. Dexamethasone treatment had no effect on culture viability but did seem to slightly decrease the final cell number for both PDI 14 and Mneo 1. Dexamethasone induction resulted in a slightly increased antibody yield for transfectoma PDI 14 but not Mneo 1. Dexamethasone induction alone had no effect on specific antibody secretion rate in either transfectoma (data not shown).

Biotechnol. Prog., 1995,Vol. 11, No. 5

570 A

B

I

---

18s rRNA, (1.77 kb)

28s rRNA, (4.74 kb)

1 2 3 4 1 2 3 4 5 6 Figure 4. Detection of PDI mRNA in the 9.2.27 hybridoma and transfectomas. (A) Endogenous PDI mRNA in the 9.2.27 hybridoma: lanes 1and 2,5 pg of total RNA, lanes 3 and 4,lO pg of total RNA. (B)PDI &A in transfectomas under dexamethasoneinduced and uninduced conditions. (10 pg of total RNA loaded in lanes 1-6): lane 1,9.2.27 uninduced; lane 2,9.2.27 induced; lane 3, PDI 4 uninduced; lane 4, PDI 4 induced; lane 5,PDI 14 uninduced; lane 6, PDI 14 induced. 1

2

3

4

5

6

7

8

9 1 0 1 1 12

106.0 kD 80.0 kD 49.5 kD

32.5 kD 27.5 kD 18.5 kD

Figure 5. Immunodetectionof PDI in lysates of transfectomas (10% SDS PAGE, nonreducing, primary antibody is chicken anti-bovinePDI). Cell lysate containing 25 pg of protein loaded in lanes 2-8: lanes 1 and molecular weight markers; lane 2, PDI 14 induced; lane 3, PDI 14 uninduced; lane 4, PDI 4 induced; lane 5,PDI 4 uninduced; lane 6, MSVPDI 1(Kitchin, 1993); lane 7, Mneo 1induced; lane 8, Mneo 1uninduced; lanes 9-11,15,30, and 100 ng of bovine PDI (Takara), respectively.

Transfectomas PDI 4 and Mneo 1 were compared in 12.5 mL shake cultures. Dexamethasone induction did not have any effect on cell number or viability. However, dexamethasone induction of transfectoma PDI 4 increased the final yield of antibody. This was not the case for Mneo 1 (Figure 6A). Dexamethasone induction did not significantly increase the specific antibody secretion rate of either transfectoma but did result in clone PDI 4 maintaining a higher antibody secretion rate into the late stationary phase (Figure 6B). Transfectomas PDI 4 and Mneo 1 were also grown in a 1.0 L bioreactor, with or without dexamethasone induction. Dexamethasone induction did not alter growth or viability of Mneo 1 (Figure 7). However, dexamethasone induction extended the viability of PDI 4 compared to both uninduced PDI 4 and Mneo 1. Dexamethasone induction of PDI 4 resulted in an increase in antibody yield compared to the uninduced PDI 4 culture; this was

not observed for transfectoma Mneo 1 (Figure 8). Dexamethasone induction did not significantly increase the specific antibody secretion rate in either PDI 4 or Mneo 1 during exponential growth; however, it resulted in a higher specific antibody secretion rate being maintained into the stationary phase (Figure 8). Transfectoma PDI 4 had increased PDI mRNA and protein following dexamethasone induction. This clone demonstrated altered antibody secretion kinetics compared to clones which were not overexpressing PDI. Similar altered mAb secretion kinetics following dexamethasone induction were observed in 12.5 mL shake flask cultures of transfectoma PDI 4 and in the 1 L bioreactor. The observation that dexamethasone induction of PDI 4 increased antibody yield compared to the uninduced PDI 4 transfectoma without increasing specific secretion rate during exponential growth has several implications for our understanding of the antibody secretory pathway in the 9.2.27 hybridoma. The low level of increase in PDI overexpression using this promoter may not have been sufficient to alter antibody assembly rate during exponential growth as originally postulated (Bibila and Flickinger, 1992a,b). The increased longevity of PDI 4 cultures induced with dexamethasone may be significant. Increased longevity resulted in a higher specific antibody secretion rate being maintained later into the stationary phase. In the highly oxidizing conditions of the ER, PDI may function as both a chaperone and a catalyst of antibody chain disulfide bonds, and thus, elevated levels of PDI may improve the efficiency of folding of many unfolded polypeptides, some of which also affect cell viability (Puig and Gilbert, 1994;Freedman et al., 1994). In order to make a definitive correlation between PDI expression and antibody secretion in the 9.2.27 hybridoma, it will be necessary to analyze more transfectomas. PDI overexpression using the weak MMTV LTR inducible promoter was an initial approach in case overexpression of PDI had a negative effect on cell survival or resulted in PDI secretion, as has been reported in yeast. This study confirms that rat PDI overexpression in a hybridoma is not toxic (reduce the growth rate); therefore,

Biotechnol. Pro& 1995, Vol. 11, No. 5

571

s! B

A

-3 I

250

p1

e

o

48 72 96 Tlme (hours)

24

I

31

24

120 144

48

72

96

120

Time (hours)

Figure 6. (A)Comparison of monoclonal antibody yield from clones PDI 4 and Mneo 1 in a 12.5 mL shake flask culture with and without 0.1 pM dexamethasone: Mneo 1no dexamethasone, 0;Mneo 1 0.1 pM dexamethasone, B; PDI 4 no dexamethasone, 0; PDI 4 0.1 pM dexamethasone, 0.(B)Specific monoclonal antibody secretion rate of clones PDI 4 and Mneo 1determined in a 12.5 mL shake flask culture with and without 0.1 pM dexamethasone: Mneo 1no dexamethasone, 0 ; Mneo 1 0.1 pM dexamethasone, W; PDI 4 no dexamethasone, 0; PDI 4 0.1 pM dexamethasone, 0.

+

+

+

+

B

A

F-

100 80

3

I

n

60

40 20

0 : 0

4

24

I

I

1

48

72

96

120

Time (hours)

o

24

7 2 96 120 144 168 Time (hours)

48

Figure 7. Growth of transfectomas Mneo 1and PDI 4 in a 1L bioreactor and induction with dexamethasone. (A)Mneo 1: uninduced, D; dexamethasone induced, W. (B)PDI 4: uninduced, 0; dexamethasone induced, 0.

transfectomas with much higher levels of PDI can now be generated by replacing the MMTV LTR with a stronger inducible or constitutive promoter. A strong inducible promoter would be useful to examine correlations between PDI levels and antibody production of single transfectomas as an internal control for changes in antibody secretion kinetics which might have occurred during transfection and selection. This is the focus of current work in our laboratory. One implication from this study is the general usefulness of this approach to improving secretion productivity

in other hybridomas or recombinant myeloma cell lines expressing antibodies or other proteins with multiple disulfide bonds. By comparing the shake flask data and bioreactor data on transfectoma PDI 4 and Mneo 1,it is clear that PDI 4 expresses less antibody (even following dexamethasone induction) than the negative control transfectoma Mneo 1. Transfectoma PDI 4 is not only expressing the protein encoded by the ne0 gene but also PDI. This may reduce the yield of antibody, or it may be a transfection (insertion) effect specific to PDI 4. If a positive correlation between PDI levels and protein

Biotechnol. Prog., 1995, Vol. 11, No. 5

572

A

B 100,

I

1

PI

g . 75 9

Y

r

P

.-n

5

50

25

f$--b" 24

0 0

40

72

9 6 1 2 0 144 168

Time (hours)

e a .g g 2.52 1.5 1

Induction

I !

0.

0

0

' 24

48 72 Time (hours)

96

120

Figure 8. Comparison of specific antibody secretion rate and antibody yield of transfectomas Mneo 1 and PDI 4 during growth in a 1 L bioreactor. (A) Mneo 1: uninduced, 0; dexamethasone induced, m. (B) PDI 4: uninduced, 0; dexamethasone induced, 0.

secretion is confirmed by further study, rather than transfecting every antibody-producing cell line with a PDI expression vector, it may be more useful to increase expression of that cell line's endogenous PDI gene. However, such an approach is not simple because, although correlations between PDI levels and Ig secretion have been described, very little in vivo data exist about PDI regulation (Freedman et al., 1994). Estrogen inhibits PDI i n vitro at physiologically relevant concentrations, but the in vivo significance, if any, of this observation is not known (Tsibris et al., 1989). It might be expected that PDI levels would be regulated coordinately with those of BiP and grp94, especially as the promoters of these genes have some regulatory elements in common (McCauliffe et al., 1992). Confirming evidence that PDI expression is not stress-regulated in the 9.2.27 hybridoma has recently been obtained in our laboratory (Yang, 1994);however, Dorner et al. (1990) show that calcium ionophore treatment of CHO cells produces a small increase in PDI mRNA which may or may not correspond to an increase in protein levels. Tasanen et al. (1991, 1993) are characterizing the 5' regulatory regions of the human PDI gene, and this work should produce a clearer understanding of the control of PDI expression. Further investigations of the regulation of PDI expression and whether higher levels of this chaperone/foldasein the ER of lymphoid cells correlates with increased mAb secretion may allow development of strategies for alteration of PDI expression at the level of transcription.

Acknowledgment Financial support of this project was provided by R & D Systems, Minneapolis, MN, and from a graduate fellowship for K.K. from Merck, Sharp and Dohme Research Laboratories, Rahway, NJ. Plasmid ppdilOO containing PDI cDNA was a gift from Dr. William J. Rutter and Jeffery C. Edman, Hormone Research Institute, University of California, San Francisco. Rabbit anti-bovine PDI antibody was kindly supplied by Dr. Robert B. Freedman, Biological Laboratory, University of Kent, Canterbury, U.K. The technical assistance of

Nancy Goebel, Xianhui Yang, and Hong Lam for portions of this work is gratefully acknowledged.

Literature Cited Abrams, P. G.; Ochs, J. J; Giardina, S. L.; Morgan, A. C.; Wilburn, S. B.; Wilt, A. R.; Oldham, R. K.; Foon, K. A. Production of large quantities of human immunoglobulin in the ascites of athymic mice: implications for the development of anti-idiotype monoclonal antibodies. J . Immunol. 1984, 132, 1611-1613. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., Struhl, K., Eds. Current Protocols in Molecular Biology; John Wiley & Sons: New York, 1989. Baneji, J.; Rusconi, S.; Schaffner, W. Expression of a betaglobin gene is enhanced by remote SV40 DNA sequences. Cell 1981,27,299-308. Bardwell, J. C. A.; Beckwith, J. C. The bonds that tie: Catalyzed disulfide bond formation. Cell 1993,74,769-771. Bassuk, J. A.; Berg, R. A. Protein disulfide isomerase, a multifunctional endoplasmic reticulum protein. Matriz 1989, 9, 244-258. Bebbington, C. R.; Renner, G.; Thomson, S.; King, D.; Abrams, D.; Yaranton, G. T. High-level expression of a recombinant antibody from myeloma cells using a glutamine synthetase gene as an amplifiable selectable marker. Bio / Technology 1992,10,169-175. Bibila, T.A.; Flickinger, M. C. Use of a structured kinetic model of antibody synthesis and secretion for optimization of antibody production systems: I steady-state analysis. Biotechnol. Bioeng. 1992a,39, 251-261. Bibila, T. A.; Flickinger, M. C. Use of a structured kinetic model of antibody synthesis and secretion for optimization of antibody production systems: I1 transient analysis. Biotechnol. Bioeng. 1992b,39, 262-272. Bliem, R.; Konoptsky, K.; Katinger, H. Industrial animal cell reactor systems: Aspects of selection and evaluation. Adu. Biochem. Eng. Bwtechnol. 1991,44, 1-26. Braakman, I.; Helenius, J.; Helenius, A. Manipulating disulfide bond formation and protein folding in the endoplasmic reticulum. EMBO J. 1992,11, 1717-1722. Brown, M. E.; Renner, G.; Field, R. P.; Hassell, T. Process development for the production of recombinant antibodies using the glutamine synthetase system. Cytotechnology 1992, 9, 231-236.

Biotechnol. Prog., 1995, Vol. 11, No. 5 Bulleid, N. J.; Freedman, R. B. Defective co-translational formation of disulfide bonds in protein disulfide isomerasedeficient microsomes. Nature 1988,335,649-651. Cai, H.; Wang, C. C.; Tsou, C. L. Chaperone-like activity of protein disulfide isomerase in the refolding of a protein with no disulfide bonds. J . Biol. Chem. 1994,269,24550-24562. Clairmont, C. A.; De Maio, A.; Hirschberg, C. Translocation of ATP into the lumen of the rough endoplasmic reticuhmderived vesicles and its binding to luminal proteins including BiP (GRP78) and GRP94. J . Biol. Chem. 1992,267,39833990. de la Brosie, D.; Noiseux, M.; Massie, B.; Lemieux, R. Hybridoma perfusion systems: A comparison study. Biotechnol. Bioeng. 1992,40,25-32. Dorman, L.; Young, H. Cationic liposome-mediatedtransfection of suspension cell cultures. Focus (Gibco-BRL) 1989,11,37. Dorner, A. J.; Kaufman, R. J. Analysis of synthesis, processing and secretion of proteins expressed in mammalian cells. Methods Enzymol. 1990,185,577-596. Dorner, A. J.; Krane, M. G.; Kaufman, J. R. Reduction of endogenous GRP78 levels improves secretion of a heterologous protein in CHO cells. Mol. Cell. Biol. 1988,8, 40634070. Dorner, A. J.;Wasley, L. C.; Raney, P.; Haugejorden, S.; Green, M.; Kaufman, R. J. The stress response in Chinese hamster ovary cells. J. Biol. Chem. 1990,265,22029-22034. Dorner, A. J.; Wasley,.L. C.; Kaufman, R. J. Overexpression of GRP78 mitigates stress induction of glucose regulated proteins and blocks secretion of selective proteins in Chinese hamster ovary cells. EMBO J. 1992,11, 1563-1571. Edgington, S. M. Rites of passage: moving biotech proteins through the ER. Biol Technology 1992,10,1413-1420. Edman, J. C.; Ellis, L.; Blacher, R. W.; Roth, R. A; Rutter, W. J. Sequence of protein disulfide isomerase and implications of its relationship to thioredoxin.Nature 1985,317,267-270. Farquhar, R.; Honey, N.; Murant, S. J.; Bossier, P.; Schultz, L.; Montgomery, D.; Ellis, R. W.; Freedman, R. B. Protein disulfide isomerase is essential for viability in Saccharomyces cerevisiae. Gene 1990,108,81-89. Flickinger, M. C.; Goebel, N. K.; Bohn. M. A. Determination of specific monoclonal antibody secretion rate during very slow hybridoma growth. Bioproc. Eng. 1990,5,155-164. Flickinger, M. C.; Bibila, T. A.; Kitchin, K. Prediction and experimental investigation of the effect of altered peptide chain assembly rate on antibody secretion using a structured model of antibody synthesis and secretion. In Harnessing Biotechnology for the 21st Century; Ladisch, M. R., Bose, A., Eds.; American Chemical Society: Washington DC, 1992; pp 345-350. Freedman; R. B. Native disulfide bond formation in protein biosynthesis: evidence for the role of protein disulfide isomerase. TZBS 1984,9,438-441. Freedman, R. B. Protein disulfide isomerase: multiple roles in the modification of nascent secretory proteins. Cell 1989,57, 1069-1072. Freedman, R. B.; Hirst, T. R.; Tuite, M. F. Protein disulphide isomerase: building bridges in protein folding. TZBS 1994, 19,331-336. Gassman, M.; Thimmes, P.; Weiser, T.; Hubsher, U. Efficient production of chicken egg yolk antibodies against a conserved mammalian protein. FASEB J . 1990,4, 2528-2532. Gething, M. J.;Sambrook, J. Protein folding in the cell. Nature 1992,355,33-45. Haas, I. G.; Wabl, M. Immunoglobulin heavy chain binding protein. Nature 1983,306, 387-389. Hawkins, H. C.; Freedman, R. B. The reactivities and ionization properties of the active-site dithiol groups of mammalian protein disulfide-isomerase.Biochem J . 1991,275,335-339. Hawkins, H. C.; De Nardi, M.; Freedman, R. B. Redox properties and cross-linking of the dithiolldisulfide active sites of mammalian protein disulfide-isomerase. Biochem J . 1991a,275, 341-348. Hawkins, H. C.; Blackburn, E. C.; Freedman, R. B. Comparison of the activities of protein disulfide-isomeraseand thioredoxin in catalyzing disulfide isomerization in a protein substrate. Biochem. J . 1991b,275, 349-353.

573 Hilleon D. A.; Lambert, N.; Freedman, R. B. Formation and isomerizationof disulfide bonds in proteins: protein disulfideisomerase. Methods Enzymol. 1984,107,281-294. Hwang, C.;Sinskey, A. J.; Lodish, H. F. Oxidized redox state of glutathione in the endoplasmic reticulum. Science 1992,257, 1496-1502. Ibbetson, A. L.; Freedman, R. B. Assay of microsomal membranebound glutathione-insulin transhydrogenase and comparison with protein disulfide-isomerase.Biochem. J . 1976,159,377384. Kearney, J. F.; Radbruch, A.; Liesegang, B.; Rajewsky, K. A new mouse myeloma cell line that has lost immunoglobulin expression but permits the construction of antibody-secreting hybrid cell lines. J . Zmmuml. 1979,123, 1548-1550. Kitchin, K. Genetic manipulation of protein disulfide isomerase levels in a murine hybridoma: Altering antibody assembly and secretion. M. S. Thesis, University of Minnesota, St. Paul, MN, 1993. Knittler, M. R.; Haas, I. G. Interaction of BiP with newly synthesized immunoglobulin light chain molecules: cycles of sequential binding and release. EMBO J. 1992,11, 15731581. Koch, G. L. E.; Macer, D. R. J.; Wooding, F. B. P. endoplasmin is a reticuloplasmin. J. Cell Sci. 1988,90,485-491. Krieg, P.; Melton, D. Double stranded DNA template preparation. In Protocols and Applications Guide; Titus, D., Ed.; Promega Corp.: Madison, WI, 1986; p 106. Kronig, H.; Thiel, U.; Ansorge, S. Induction of protein disulfide isomerase and immunoglobulins by pokeweed mitogen in human lymphocytes. Hybridoma 1991,10,651-657. Kruse, P. F.; Patterson, M. K. (Eds.) Tissue Culture Methods and Applications; Academic Press: New York, 1973; pp 406. Laemmli, U. K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970,227, 680-685. Lamantia, M. L.; Lennarz, W. J. The essential function of yeast protein disulfide isomerase does not reside in its isomerase activity. Cell 1993,74,899-908. Lamantia, M.; Miura, T.; Tachikawa, H.; Kaplan, H. A.; Lennarz, W. J.; Mizunaga, T. Glycosylation site binding and protein disulfide isomerase are identical and essential for cell viability in yeast. PNAS 1991,88,4453-4457. Lambert, N.;Freedman, R. B. Structural properties of homogeneous protein disulfide-isomerasefrom bovine liver purified by a rapid high-yielding procedure. Biochem. J . 1983,213, 225-234. Lee, A. S. Coordinated regulation of a set of genes by glucose and calcium ionophores in mammalian cells. TZBS 1987,12, 20-23. Lee, F.; Mulligan, R.; Berg, P.; Ringold, G. Glucocorticoids regulate expression of dihydrofolatereductase DNA in mouse mammary tumor virus chimeric plasmids. Nature 1981,294, 228-232. Lu, X.;Gilbert, H. F.; Harper, J. W. Conserved residues flanking the thiol disulfide centers of protein disulfide isomerase are not essential for catalysis of thiolldisulfide exchange. Biochemistry 1992,31,4205-4210. Lyles, M.M.; Gilbert, H. Catalysis of the oxidative folding of ribonuclease A by protein disulfide isomerase: dependence of the rate on the composition of the redox buffer. Biochemistry 1991,30, 613-619. Mayinger, P.; Meyer, D. I. An ATP transporter is required for protein translocation into the yeast endoplasmic reticulum. EMBO J . 1993,12,659-666. Mazzarella, R. A.; Green, M. ERp99, an abundant, conserved glycoprotein of the endoplasmic reticulum, is homologous t o the 90-kDa heat shock protein (hsp90) and the 94-kDa glucose regulated protein (GRP94).J . Biol. Chem. 1987,262,88758883. Mazzarella, R. A.; Srinivasan, M.; Haugejorden, S. M.; Green, M. ERp72, an abundant luminal endoplasmic reticulum protein, contains three copies of the active site sequences of protein disulfide isomerase. J . Biol. Chem. 1990,265,10941101. McCauliffe, D. P.; Yang, Y-S.; Wilson, J.; Sontheimer, J. D.; Capra, J. D. The 5'-flanking region of the human calreticulin gene shares homology with the human grp78, grp94, and

Biotechnol. Pfog., 1995, Vol. 11, No. 5

574 protein disulfide isomerase promoters. J. Biol. Chem. 1992, 267,2557-2562. Melnick, J.; Dul, J. L.; Argon, Y. Sequential interaction of the chaperones BiP and Grp94 with immunoglobulin chains in the endoplasmic reticulum. Nature 1994,370,373-375. Morgan, A. C.; Galloway, D. R.; Reisfeld, R. A. Production and characterization of monoclonal antibody to a melanoma specific glycoprotein. Hybridoma 1981,1, 27-36. Morjana, N. A.; Gilbert, H. F. Effect of protein and peptide inhibitors on the activity of protein disulfide isomerase. Biochemistry 1991,30,4985-4990. Munro, S.;Pelham, H. R. B. An hsp7O-like protein in the E R Identity with the 78kD glucose-regulated protein and immunoglobulin heavy chain binding protein. Cell 1986,46,291300. Myllyla, R.; Oikarinen, J. A rapid and sensitive assay for protein disulfide isomerase activity. J. Biochem. Biophys. Methods 1983,7, 115-121. Myllyla, R.;Koivu, J.; Pihlajaniemi, T.; Kivirikko, K. I. Protein disulfide isomerase activity in various cells synthesizing collagen. Eur. J. Biochem. 1983,134,7-11. Nigam, S. K.; Goldberg, A. L.; Ho, S.; Rohde, M. F.; Bush, K. T.; Sherman, M. Y. A set of endoplasmic reticulum proteins possessing properties of molecular chaperones includes Ca+2binding proteins and members of the thioredoxin superfamily. J . Biol. Chem. 1994,269,1744-1749. Noiva, R.; Lennarz, W. J. Protein disulfide isomerase. J. Biol. Chem. 1992,267,3553-3556. Noiva, R.; Kimura, H.; Roos, J.;Lennarz, W. J. Peptide binding by protein disulfide isomerase, a resident protein of the endoplasmic reticulum. J . Biol. Chem. 1991,266, 1964519649. Ozturk, S. S.; Palsson, B. 0. Growth, metabolic, and antibody production kinetics of hybridoma cell culture: 1.Analysis of data from controlled batch reactors. Biotechnol. Prog. 1991, 7, 471-480. Paver, J. L.; Freedman, R. B.; Parkhouse, R. M. E. Induction of expression of protein disulfide isomerase during lymphocyte maturation stimulated by bacterial lipopolysaccharide. FEBS Lett. 1989,242,357-362. Pelham, H. R. B. The retention signal for soluble proteins of the endoplasmic reticulum. TIBS 1990,15,483-486. Polson, A.Isolation of IgY from the yolks of eggs by a chloroform polyethylene glycol procedure. Zmmunol. Invest. 1990, 19, 253-258. Puig, A.; Gilbert, H. F. Protein disulfide isomerase exhibits chaperone and anti-chaperone activity in the oxidative refolding of lysozyme. J. Biol. Chem. 1994,269,7764-7771. Robinson, A. S.; Hines, V.; Wittrup, K. D. Protein disulfide isomerase overexpression increases secretion of foreign proteins in Saccharomyces cerevisiae. Bio I Technology 1994,12, 381-384. Roth, R. A,; Koshland, M. E. Role of disulfide interchange enzyme in immunoglobulin synthesis. Biochemistry 1981,20, 6594-6599. Roth, R. A.; Pierce, S. B. Zn vivo cross-linkingof protein disulfide isomerase to immunoglobulins.Biochemistry 1987,26,41794182. Rothblatt, J., Novick, P., Stevens, T. H., Eds. Guidebook to the Secretory Pathway; Oxford University Press: New York, 1994.

Rupp, K.; Birnbach, U.; Lundstrom, J.; Van, P. N.; Soling, H. D. Effects of CaBP2, the rat analog of ERp72, and of CaBPl on the refolding of denatured proteins. Comparison with protein disulfide isomerase. J.Biol. Chem. 1994,269,25012507. Sambrook, J.;Fritsch, E. F.; Maniatis, T. Molecular Cloning: A Laboratory Manual, 2nd ed.; Cold Spring Harbor Laboratory Press: New York, 1989. Southern, P. J.; Berg, P. Transformation of mammalian cells to antibiotic resistance with a bacterial gene under control of the SV40 early region promoter. J. Mol. Appl. Genet. 1982, 1, 327-341. Srinivasan, M.; Lenny, N.; Green, M. Identification of genomic sequences that mediate the induction of the endoplasmic reticulum protein, ERp72, by protein traffic. DNA Cell Biol. 1993,12,807-822. Tachikawa, H.; Miura, T.; Katakura, Y.; Mizunaga, T. Molecular structure of a yeast gene, PDIl, encoding protein disulfide isomerase that is essential for cell growth. J. Biochem. 1991, 110,306-313. Tasanen, T.;Oikarinen, J.; Kivirikko, K. I.; Pihlajaniemi, T. Promoter of the gene for the multifunctional protein disulfide isomerase polypeptide. J. Biol. Chem. 1991,267, 1151311519. Tasanen, T.; Oikarinen, J.; Kivirikko, K. I.; Pihlajaniemi, T. Interaction of transcription factor S p l with the promoter of the gene for the multifunctional protein disulfide isomerase polypeptide. Biochem. J . 1993,292,41-45. Tsibris, J. C. M.; Hunt, L. T.; Ballejo, G.; Barker, W. C.; Toney, L. J.;Spellacy, W. N. Selective inhibition of protein disulfide isomerase by estrogens. J . Biol. Chem. 1989,264, 1396713970. Tuite, M. F.; Freedman, R. B. Improving secretion of recombinant proteins from yeast and mammalian cells: rational or empirical design. TIBTECH 1994,12,432-434. Urade, R.; Takenaka, Y.; Kito, M. Protein degradation by ERp72 from rat and mouse liver endoplasmic reticulum. J. Biol. Chem. 1993,268,22004-22009. Wang, C. C.; Tsou, C-L. Protein disulfide isomerase is both an enzyme and a chaperone. FASEB J . 1993,7, 1515-1517. Yang, X.Investigation of antibody assembly kinetics in vivo in a mouse hybridoma and simulations of antibody synthesis and secretion using a modified structured model. M.S. Thesis, University of Minnesota, St. Paul, MN, 1994. Yoshimori, T.; Semba, T.; Takemoto, H.; Akagi, S.; Yamamoto, A.; Tashiro, Y. Protein disulfide isomerase in rat exocrine pancreas cells is exported from the endoplasmic reticulum despite processing the retention signal. J. Biol. Chem. 1990, 265, 15984-15990. Zapun, A,; Creighton, T. E.; Rowling, P. J. E.; Freedman, R. B. Folding in vitro of bovine pancreatic trypsin inhibitor in the presence of proteins of the endoplasmic reticulum. Proteins 1992,14,10-15. Accepted April 3, 1995.@ BP950023V @

Abstract published in Advance ACS Abstracts, June 15,1995.