HSP70 Apoptosis

Transfection of NS0Myeloma Fusion Partner Cells With hsp70Gene Results in Higher Hybridoma Yield by Improving Cellular Resistance to ApoptosisElena sunskaia,1Irina I.Fridlianskaia,2Zoulfia A.Darieva,1Mariele S.R.da Silva,1Milton M.Kanashiro,1Boris A.Margulis21Universidade Estadual do Norte Fluminense,Campos,Rio de Janeiro, Brazil;telephone:+5522-27261501;fax:+5522-27261520;e-mail:elena@uenf.br2Institute of Cytology,Russian Academy of Sciences,St.Petersburg, RussiaReceived11April2002;accepted15July2002DOI:10.1002/bit.10493Abstract:Mouse myeloma NS0cells widely used in hy-bridoma technology lack the expression of a major stress protein Hsp70which is the principal component of the basic cellular defense mechanism.These cells rapidly undergo apoptosis at the late-stationary phase of batch culture following nutrient exhaustion.Since Hsp70was recently demonstrated to protect cells against numerous apoptotic stimuli,the aim of the present study was to examine the protective potential of the protein ex-pression in engineered myeloma NS0cells and in result-ing hybridomas.Myeloma cells were transfected with the hsp70gene under beta-actin gene promoter.To imitate harmful conditions that hybridoma or myeloma cells often experience when cultivated in large scale for an antibody production,NS0wt and NS0hsp70cell cul-tures were maintained without changing the medium for a few days,and the expression of apoptotic markers has been studied.It was found that long-term cultivation in-duced apoptosis in original cells manifested by typical nuclei fragmentation,DNA ladders and activation of caspase-3.In contrast,in transfected cells under the same conditions the outcome of apoptosis was post-poned for24hours.Most relevant was that the fusion of transfected myeloma cells with immune splenocytes re-sulted in twofold hybridomas output compared with wild-type fusion partner.Almost half of the hybridomas continued to be hsp70-positive and maintained higher robustness in culture.The level of monoclonal antibod-ies production by hybridoma cells obtained with the use of NS0wt and NS0hsp70was similar,however,the se-creted product was better preserved in culture superna-tants of Hsp70-positive cells.It is concluded that trans-fection of mouse myeloma cells with the hsp70gene can be a novel means to increase hybridoma yield and re-duce the sensitivity of myeloma and hybridoma cells to culture conditions insults accompanying monoclonal an-tibody production.©2003Wiley Periodicals,Inc.Biotechnol Bioeng81:496–504,2003.Keywords:apoptosis;Hsp70;myeloma;hybridoma;cell cultureINTRODUCTIONThe data reported during the last few years,indicate that the apoptosis plays a central role in controling cell number and viability of cell lines that are employed for the production of biopharmaceuticals.Production of monoclonal antibodies by B-cell hybridoma cultures is also often limited by the rapidness with which the cells undergo apoptosis at the entry into the decline phase.Hybridoma and myeloma cells were shown to be highly susceptible to several stress factors encountered during cell growth in bioreactors.Oxygen de-privation and hydrodynamic stress,depletion of specific nu-trients or growth factors,accumulation of toxic metabolic end products normally observed at decline phase of cell culture cause rapid and massive apoptosis of cultured cells (Al-Rubeai et al.,1990;Singh et al.,1994;Mercille and Massie,1994).Apoptotic death of myeloma and hybridoma cells is followed by secondary necrosis and liberation of proteases and other cellular products that could alter the final product,monoclonal antibodies.The extension of cell survival in batch culture may significantly influence pro-ductivity and cost-effectiveness of monoclonal antibody production.Identification of genes regulating apoptosis made it possible to prevent cell death.The transfection of myeloma and hybridoma cells with anti-apoptotic genes, bcl-xL(Charbonneau and Gauthier,2000),bcl-2(Ray and Diamond,1994;Schwarze and Hawley,1995;Fassnacht et al.,1998)and E1B(Mercille et al.,1999),was found to increase the robustness of genetically modified cells,while the enhancement in monoclonal antibody production was disputable.Although several groups observed elevated monoclonal antibody production by bcl-2-(Itoh et al.,1995;Correspondence to:Elena sunskaiaContract grant sponsors:Fundac¸a˜o de Apoio a Pesquisa do Estado do Rio de Janeiro(FAPERJ);Conselho Nacional de Pesquisa(CNPq); FENORTE©2003Wiley Periodicals,Inc.Simpson et al.,1998)or bcl-2/bag-1-(Terada et al.,1997) transfected hybridoma cells,many others found no differ-ence between resistant and original cell lines(Fassnacht et al.,1998;Mercille et al.,1999).The resistance of hybridoma cells to apoptotic stimuli may be enhanced by direct transfection with an anti-apoptotic gene or by the transfection of a myeloma fusion partner.Although a first approach gave good results(Itoh et al.,1995;Simpson et al.,1998),the second one seems to be more convenient.Attempts to employ myeloma cells trans-fected by anti-apoptotic genes for the fusion with spleno-cytes gave controversial results.Bcl-2-transfected NS0my-eloma cells increased the yield of hybridomas(Ray and Diamond,1994),while E1B-19K gene had no effect on the efficiency of fusion.Furthermore,the resulting hybridomas, though expressing E1B-19K at levels comparable to the NS0myeloma parent,were no longer resistant to apoptosis (Mercille et al.,1999;Nylandsted et al.,2000).Heat shock proteins of hsp70family are well-known mo-lecular chaperones that are involved in protein folding, transport,and degradation machinery.Hsp70plays the de-cisive role in cell protection against a variety of cytotoxic factors that at least partly attributed to the suppression of apoptosis(Jaattela,1999).Hsp70-mediated anti-apoptotic effect against DNA dam-age,UV radiation,serum withdrawal and growth factors deprivation was well documented(Samali and Orrenius, 1998).It seems meaningful that murine myeloma NS0cells lack this unique protective system since they are deficient in expression of Hsp70(Aujame and Firco,1988).The deliv-ery of pure Hsp70to myeloma cells with liposomes(La-sunskaia et al.,1997)or transfection with hsp70gene (Fridlianskaia et al.,2000)made myeloma cells more resis-tant to apoptosis caused by heat shock or serum deprivation. Therefore,correction of the natural defect in hsp70expres-sion could probably enhance anti-apoptotic resistance of myeloma cells maintained in batch culture.In the present study we show that the transfection of mouse myeloma NS0cells with human hsp70gene con-trolled by constitutive promoter is able to protect NS0my-eloma cells against metabolic stress normally observed at the end of the stationary phase.The employment of trans-fected myeloma cells in fusion with splenocytes yielded in twofold increased hybridoma output.Almost40%of re-sulted hybridomas continued to express human hsp70,and were more resistant to apoptosis though did not enhance antibody productivity.MATERIALS AND METHODSMaterialsCell culture media and supplements,Trizol and Lipofectin reagents,G418antibiotic,RT-PCR“Ready-To-Go”kit were purchased from Gibco-BRL Life Technologies(Rio de Janeiro,Brazil).ECL cocktail for chemiluminescent devel-opment of immunoblots kit and reagents for PCR were ob-tained from Pharmacia/Amersham(Rio de Janeiro,Brazil). The reagents for electrophoresis,immunoblotting were from Sigma-Aldrich(Rio de Janeiro,Brazil).Cell CulturesNS0murine myeloma,J774.A1mouse macrophages and U-937human leukemia cells were obtained from the Rus-sian Collection of Animal Cell Cultures(St.Petersburg, Russia).All cell lines were found to be negative for myco-plasma contamination according to Hoechst staining.The parental NS0myeloma,transfected myeloma and hybrid-oma cells were grown in DMEM-F12medium supple-mented with10%fetal calf serum(FCS),50␮M2-mercap-toethanol and50␮g/mL gentamicin.J774and U-937cells were cultivated in DMEM and RPMI-1640respectively with10%FCS.Growth rates of cell lines were determined by plating cells at a starting concentration of5×105cells mL−1to24-well plates with1mL of cell suspension per well and by counting cell concentrations and viability at24h intervals over a5–6-day period using trypan blue exclusion assay and haemacytometer counts.Transfection of Myeloma CellsPlasmid p␤-actin-neo-hsp70was a generous gift of Dr.R. Morimoto(Northwestern University,USA);it contains hu-man hsp70cDNA under the transcriptional control of the human␤-actin promoter and SV2neo sequence encoding resistance to geneticin(G418).The transfection was per-formed with Lipofectin Reagent according to protocol of manufacturer.Briefly,15–20␮g of p␤-actin-neo-hsp70 plasmid and25␮L of Lipofectin formed a lipid-DNA com-plex that was added to60-mm dish with exponentially growing NS0cells in the presence of5%of fetal calf serum. After20h of incubation the medium was replaced with4 mL of growth medium and cells were incubated for another 48h.Then the cells were subcultured at low density in the growth medium with0.6mg/mL geneticin.The resulting clones were isolated and tested for hsp70expression by Western blotting or by RT-PCR assay.Generation of HybridomasSix to eight week BALB/c female mice were immunized with a number of antigens including pure bovine heat shock protein hsc70,recombinant rice protein SALT;phospholi-pase A from snake venom;vitelogenin from Rhodnius pro-lixos oocytes and T.cruzi DAF protein.Three days after a booster primary spleen cells were fused with either parental NS0wt or transfected NS0hsp70myeloma partners according to standard protocol(Barrett,1994).Overall,eight fusions were performed.The wells of96-well plates were scored for hybridoma growth10days after fusion.The number of the specific hybridoma clones producing monoclonal antibodiesLASUNSKAIA ET AL.:HSP70PREVENTS APOPTOSIS IN MYELOMA AND HYBRIDOMA CELLS497of interest was evaluated by immunoenzyme assay.Thirty randomly selected hybridomas originated from NS0wt fu-sion partner and thirty NS0hsp70-based hybridomas were tested for hsp70expression by RT-PCR assay.PCR and Western BlottingTotal RNA was extracted from cells by Trizol reagent ac-cording to the manufacturer’s instruction and applied to two-step RT-PCR using the Ready To Go kit.Briefly,2␮g RNA were added to Ready-To-Go beads and supplemented with1,5␮g Pd(N)6random primer and MgCl2to obtain the final concentration of1m M,in a total volume of50␮L. The reaction was incubated at42°C for30min,followed by 5min at95°C to inactivate enzyme.At the second step7,5 m M of the human hsp70A(hsp70-1)stress gene PCR primer pair(StressGen Biotechnol.Corp.,Canada):(sense) 5Ј-TGTTCCGTTTCCAGCCCCCAA-3Јand(antisense)5Ј-GGGCTTGTCTCCGTCGTTGAT-3Ј.were added to the re-action mixture.PCR reaction conditions were as follows: 1-min denaturation at95°C,1-min annealing at60°C,with a final extension of a1min at72°C for30cycles to generate a product of360bp.Both steps of the reaction were per-formed with the use of PTC-100MJ Res.Inc.thermal cycler (USA).An aliquot of the PCR reaction was electrophoresed through a1.0%agarose gel containing0.5␮g/mL ethidium bromide.The analysis of Hsp70expression in myeloma and hy-bridoma cells was assessed by Western blotting,using2H9 monoclonal antibody specific to inducible isoform of mam-malian Hsp70(Lasunskaia et al.,1997).Briefly,cells were lysed in RIPA buffer(150m M NaCl;2m M EDTA;0.5% Triton X-100;20m M Tris-HCl,pH7.5;1m M Na ortho-vanadate;50m M NaF;1m M PMSF).100␮g of protein was electrophoresed through10%SDS-PAGE(Laemmli,1970) and transferred onto a nitrocellulose filter.Blots were first incubated with5%skim milk and then treated with anti-hsp70antibody followed by peroxidase-conjugated goat anti-mouse Ig antibody.Peroxidase activity was finally de-veloped with the aid of chemoluminenscent cocktail,ECL. Growth Curves and Quantification of Apoptosisin Cell CulturesOriginal NS0wt and transfected NS0hsp70myeloma cells as well as hybridomas resulting from above myeloma cells(ten randomly selected hsp70-positive and ten hsp70-negative hybridomas)were cultured for6days.Cell number and viability were measured using trypan blue exclusion assay and haemacytometer counts.For the apoptosis quantifica-tion cells were stained by acridine orange/ethidium bromide DNA fluorochroms(Martin and Lenardo,1998).Viable cells with intact membrane were ethidium bromide-negative and presented well-preserved nucleus morphology.Ethid-ium bromide-negative cells with the condensed chromatin and fragmented nuclei were qualified as primary apoptotic cells.Cells stained by ethidium bromide were considered to be necrotic:either primary(intact nuclei)or secondary (apoptotic nucleus morphology).Both,primary apoptotic and secondary necrotic,were quantified as apoptotic cells. DNA fragmentation assay was employed to evaluate in-ternucleosomal degradation of cellular DNA(Lasunskaia et al.,1997).Briefly,2×106myeloma or hybridoma cells were washed in PBS,lysed in10m M Tris pH7,5,2m M EDTA,1%Triton X-100for20min on ice and then treated with20␮g/mL RNAse A at37°C for1h and0.5mg/mL Proteinase K for an additional hour.The DNA was precipitated in ethanol,centrifuged for20 min at11,000g and dissolved in loading buffer(10m M Tris, pH7.5,2m M EDTA,3%Ficoll400,0.1%bromphenol blue-xylene cyanole)and electrophoresed on a1.2%aga-rose gel for1h at85V.To measure the activity of caspase-3cells were lysed in 50m M HEPES,5m M EGTA,2m M MgCl2,1m M DTT,1 m M PMSF,1␮g/mL leupeptin and100␮g of total protein was incubated with5␮g Ac-DEVD-7-amino4-methylcou-marin caspase-3fluorogenic substrate(Calbiochem)in0.5 mL of100m M HEPES,10%sucrose,0.1%CHAPS,1m M DTT for40min at37°C(Han et al.,1999).The fluores-cence was measured at380nm(excitation)and460nm (emission)with the aid of Fluorescence Spectrophotometer F-4500(Hitachi,Japan).Specificity of caspase-3measure-ment was confirmed by inhibitory analysis in the presence of the specific caspase-3inhibitor Z-DEVD-fmk,20␮M, (Stratagene,USA).Quantification of Monoclonal Antibody(mAb)ProductionThe titer of murine immunoglobulins(Ig)in cell superna-tants was measured by an immunoenzyme assay.MAbs produced by hybridoma were sandwiched by rabbit anti-mouse Ig polyclonal antibody(Sigma-Aldrich,Rio de Ja-neiro,Brazil)and biotinilated goat anti-mouse Ig antibody (Dako,Denmark)followed by incubation with streptavidin-peroxidase conjugate(Dako,Denmark).The amount of mAb was determined by measuring absorbance at490nm using microplate reader(Dynatech MR5000).A standard curve was generated by titration of purified mAb.Anti-mouse Ig Ab titers were determined from the linear portion of the standard curve.Statistical AnalysisThe results were given as means±standard errors of the means.The significance of the observed differences was calculated by Student’s t test.A value of p<0.05was considered to be significant.RESULTSHsp70Expression in Transfected NS0CellsThe transfection of NS0cells by human hsp70gene driven by constitutive promoter resulted in a few viable clones.498BIOTECHNOLOGY AND BIOENGINEERING,VOL.81,NO.4,FEBRUARY20,2003One of them,NS0hsp70,with growth characteristics resem-bling that of parental cells was chosen for further experi-ments.The results of Western blotting showed that these cells unlike parental NS0wt cells synthesized Hsp70(Fig.1A).Monoclonal antibodies employed specifically recog-nized inducible isoform of both human and rodent Hsp70,but not its constitutive isoform (Lasunskaia et al.,1997).Mild heat shock significantly increased Hsp70content in the cells of human U-937and murine J774.A1lines natu-rally expressing the protein,but not in NS0wt cells,that confirmed the lack of Hsp70expression in these cells de-scribed earlier (Aujame and Firco,1988;Lasunskaia et al.,1997).Transfected NS0hsp70cells expressed Hsp70consti-tutively.To our surprise,mild heating at 43°C for 15min substantially increased Hsp70amount in transfected cells.To prove that the Hsp70expression was not a result of a spontaneous selection of Hsp70-positive myeloma cells cel-lular mRNA was analyzed by RT-PCR using specific primer for human hsp70.The data showed that hsp70ex-pressed in NS0hsp70cells was of human origin (Fig.1B).The specificity of primer for human,but not for mice gene,was confirmed by negative results of RT-PCR analysis of other murine cells lines,J774.A1,RAW264.7,and P388.D1,known to express hsp70(the data are shown only for J774.A1cells).The level of hsp70message accumula-tion in intact and heat-shocked transfected NS0cells was similar that opposed to the results of Western blotting.We suggest that the increase in Hsp70accumulation in trans-fected cells after heat shock is not a result of higher expres-sion rate of the protein but of its lower degradation rate.Hsp70Expression Protects Myeloma Cells From ApoptosisNaive,NS0wt ,and transfected,NS0hsp70,myeloma cells were cultivated during 5-day period without medium change and tested for viability and apoptosis.Growth curves demonstrated the enhanced survival of transfected cells (Fig.2A).Stationary-growth phase of NS0hsp70cells was 24h longer than that of the original myeloma.Cell death in both myeloma cultures started,when cell density reached 2.0×106mL −1.Fluorescence microscopy analysis of the cells stained with DNA dyes revealed that the decrease in viability coincided with an increase in the proportion of apoptotic cells in culture (Fig.2B).The early appearance of secondary necrosis was typical of apoptotic myeloma cells.Apoptotic cells detected in cul-tures starting from 72h were predominantly membrane-damaged or secondary necrotic as they were permeable for ethidium bromide.The number of primary necrotic cells remained extremely low (less than 5%)throughout the whole time of the experiment (data not shown).Naive NS0wt cells were much more susceptible to apoptosis than NS0hsp70:The proportion of apoptotic cells in the original myeloma culture was at least twofold higher at all stages of observation,reaching 80%on the fifth day,in comparison with 40%in culture of transfected cells.Delay of apoptotic death was confirmed by higher level of DNA fragmentation clearly observed in NS0wt cells on the third day in culture,while transfected cells were intact up to the fourth day (Fig.2C).Activation of the effector caspase-3started in the naive myeloma cells after 2days cultivation,rising to amaximumFigure 1.Hsp70expression in native and transfected NS0myeloma cells.Native (NS0wt )or transfected (NS0hsp70)murine myeloma cells and human (U-937)or murine (J774.A1)cells were heated (HS)or not (C)at 43°C,for 15min,recovered for 18h,and then total cellular proteins or RNA were extracted and analyzed for hsp70expression.(A)Western blotting analysis using monoclonal antibody,2H9,specific for inducible form of mammalian Hsp70.Amounts of total cellular protein loaded to each lane:50␮g/lane -U-937and J774.A1cells,100␮g/lane-myeloma cells.(B)RT-PCR analysis using human hsp70A primer,resulting in a 360bp PCR product.1:“Ready-to-go”kit positive control;2,3:human mono-blastoid U-937cells;4,5:murine macrophage-like J774.A1cells;6:human peripheral blood lymphocytes;7,8:NS0hsp70cells;9,10:NS0wtcells.Figure 2.Growth curves and kinetics of apoptotic death of native and transfected myeloma cells throughout 5-d culture.Viable cells density (A)and percentage of apoptotic cells cultured in complete medium (B)quan-tified under fluorescence microscopy of acridine orange/propidium iodide-stained cells according to morphological observations.DNA fragmentation evaluated by gel electrophoresis of cell samples obtained at day 3and 4of culture (C)and caspase-3activity in myeloma cells cultured either with or without caspase-3inhibitor and measured by fluorometric assay.Data are representative of three independent experiments expressed as the mean ±SD.LASUNSKAIA ET AL.:HSP70PREVENTS APOPTOSIS IN MYELOMA AND HYBRIDOMA CELLS 499at the fourth day,and then dropped,that was probably con-nected with the profound secondary necrosis and degrada-tion of the cells.Caspase-3activation in NS0hsp70cells de-veloped with the24h delay and was significantly less pro-found(Fig.2D).NS0hsp70Fusion Partner Employment Enhances Hybridoma YieldTo test the fusion efficiency of wild-type and transfected myeloma cells eight fusions with spleen cells from mice immunized with various antigens were performed.All fu-sions with transfected myeloma cells resulted in nearly two-fold increase in total number of hybrids as compared to NS0wt fusions(Table I).Although,the rate of specific,an-tigen-positive hybridomas production was the same for both fusion partners since approximately45%of hybrids ob-tained were specific for the immunization antigen in both cases,the total yield of specific hybridomas obtained using transfected myeloma was significantly higher.NS0hsp70fu-sions resulted in560antigen-specific hybridomas,while employment of NS0wt gave in the same conditions261hy-bridomas.The data demonstrate twofold improvement of hybridoma generation and establish the advantage of the transfected NS0myeloma cells as a fusion partner for hy-bridoma technology.Hsp70Expression in Hybridoma Cells Improves Cellular ViabilityTo study whether hybridomas resulted from fusion with transfected myeloma cells continued to express recombinant human hsp70we tested the protein expression by Western blotting and RT-PCR in30randomly selected hybridomas obtained from NS0wt and30hybridomas obtained from NS0hsp70cells.All the NS0wt-based hybridomas tested were hsp70-negative,while11hybridomas(37%)generated from NS0hsp70fusions expressed hsp70.RT-PCR analysis proved that the hsp70gene expressed in positive hybridoma cells originated from NS0hsp70cells because PCR primer used in our experiments was specific for human but not rodent hsp70gene(Fig.3).The level of hsp70expression in posi-tive hybridoma cells was monitored by Western blotting up to5days in culture and was demonstrated to be stable(data not shown).Ten randomly selected hsp70-positive and ten hsp70-negative hybridomas were compared for cell viability and occurrence of apoptosis during long-term cultivation.Vi-ability of hsp70expressing hybridomas did not change sig-nificantly up to96h in culture whereas viability of hsp70-negative cells started to drop after48h(Fig.4A).The in-creased robustness of hsp70-positive cells in culture was mediated by more pronounced resistance to apoptosis(Fig. 4B).Table I.Yields of total and specific hybridomas from splenic fusions with NSO wt and NSO hsp70.Fusion number Antigen Fusion partnerWells with hybridomas:Total Specificn°/N°a%b n°/N°c%d1hsc73NSO wt99/2883426/9926NSO hsp70143/2885040/143282hsc73NSO wt79/2882713/7917NSO hsp70100/2883525/100253hsc73NSO wt137/2884811/1378NSO hsp70172/2886037/172224SALT NSO wt74/192395/747NSO hsp70134/1927029/134225phospho-NSO wt107/3842838/10735 lipase A NSO hsp70300/3847853/300176vitelogenin NSO wt58/2402458/58100NSO hsp70121/24050118/121987vitelogenin NSO wt99/2404165/9965NSO hsp70205/24085156/205768T.cruzi DAF NSO wt51/2881845/5188NSO hsp70137/28848102/13775 Fusion efficiency(%)NSO wt32,4±3,5bNSO hsp7059,5±6,0b(p<0.001)*Specific hybridomas NSO wt261/704c43,2±12,9dyield NSO hsp70560/1312c(p<0.001)*48,6±10,8d(pס0.63) a Number of confluent wells(n°)/Number of wells plated(N°).Wells containing hybridoma cells were considered confluent when cell density covered at least70%of the well surface.b Percent fusion efficiency was calculated as the number of wells with confluent cell growth divided by the number of wells plated×100.c Number of wells containing ELISA+antigen-specific hybridomas(n°)/Number of confluent wells(N°).d Percent of the specific hybridomas yield was calculated as the number of wells with antigen-specific hybridomas divided by the number of confluent hybridoma wells×100.*p<0.001vs.hybridomas generated from wild-type myeloma(Student’s t test).500BIOTECHNOLOGY AND BIOENGINEERING,VOL.81,NO.4,FEBRUARY20,2003Hsp70Expression in Hybridoma Cells Does Not Affect Antibody Production But Assists the Preservation of the Produced Antibodies in Culture SupernatantsTo compare monoclonal antibody (mAb)productivity of hsp70-positive or negative hybridomas culture supernatants10hybridomas in each group were tested for antibody con-centration throughout 6day cultivation.The level of mAb production varied significantly in individual hybridomas of both groups with a maximum concentration ranged between 15and 85␮g mL −1(data not shown).To exclude individual variation in mAb secretion by hybridomas and verify the contribution of enhanced cell viability to the final output of the antibodies we defined the productivity index.The pro-ductivity index is the ratio between the titer value for a particular day and a maximum mAb titer obtained for a certain hybridoma.This value raised more than fourfold from the beginning of exponential phase (day 1)to the end of the stationary phase (day 3—for hsp70-negative and day 4—for hsp70-positive cells;Fig.5).There was no statisti-cally significant difference in mAb production between hsp70-positive and negative hybridoma groups (Fig.5A).MAb productivity of hsp70-positive hybridomas at different stages of stationary phase (day 3and day 4)wassimilarFigure 4.Kinetics of apoptotic death of hybridoma cells originated from native and transfected myeloma cells throughout 5-d culture.Percentage of viable (A)and apoptotic cells in culture (B)quantified under fluorescence microscopy of acridine orange/ethidium bromide-stained cells according to morphological observations.Data are mean values ±SD of 10represen-tative hybridomas from each group.The data from one of two independent experiments with similarresults.Figure 3.Hsp70expression in hybridomas originated from NS0wt or NS0hsp70fusion partner.(A)Expression of human hsp70mRNA in hy-bridoma cells.Total RNA was extracted from indicated hybridoma cells and submitted to RT-PCR using human hsp70A primer,resulting in a 360bp PCR product.B.Western blotting analysis of Hsp70expression in indicated hybridoma cells using monoclonal antibody 2H9,specific for inducible form of mammalian Hsp70.100␮g of total cellular protein were loaded to eachlane.Figure 5.Effect of Hsp70expression on antibody secretion and hybrid-oma cell survival in culture.Hsp70-positive and negative hybridomas cul-tures were tested for antibody concentration in culture supernatants pre-sented as a percentage of maximum titer of mAb secreted by each hybrid-oma (A)and viable (B)and total cells density (C)throughout a 6-d culture.Data are mean values ±SD of 10representative hybridomas from each group.LASUNSKAIA ET AL.:HSP70PREVENTS APOPTOSIS IN MYELOMA AND HYBRIDOMA CELLS 501suggesting no significant secretion during the remaining plateau stage.The drop of mAb concentration observed in the culture of hsp70-negative cells to the end of the declin-ing phase(day6)coincided with started cell disintegration and loss(Fig.5C)but not with gradual decrease of viability of cell population(Fig.5B)and probably reflects antibody degradation by proteolytic enzymes released from disinte-grating cells.Antibody titer in more robust hsp70-positive hybridoma cultures was not altered up to the day6in culture and demonstrated the better preservation of the secreted product in culture supernatant.DISCUSSIONMyeloma and hybridoma cells are very sensitive to changes in cell culture conditions occurring during long-term culti-vation.High cell concentration and nutrient exhaustion nor-mally accompanying the decline phase of cell culture are known to reduce cell viability and cause the process of programmed cell death,apoptosis.Major heat shock protein,Hsp70,inducible isoform,was convincingly demonstrated to protect cells from various cy-totoxic factors that are predominantly apoptotic stimuli. NS0murine myeloma cells are rare natural cell variants lacking this protein.Since myeloma NS0cells is an impor-tant fusion partner for hybridoma production we studied whether the transfection with hsp70gene is able to protect myeloma cells and generated hybridomas against apoptosis and by this improve hybridoma yield and mAb production. NS0cells were transfected by vector containing the entire human hsp70gene under the transcriptional control of the human beta-actin gene promoter.One of the clones that expressed hsp70observed by Western blotting and RT-PCR analysis,NS0hsp70,was selected and employed in further studies.Interestingly,in transfected cells heat shock in-creased Hsp70amount as detected by Western blotting and one might speculate that NS0hsp70cells were selected from spontaneous cell variants expressing murine Hsp70.How-ever,the results of RT-PCR analysis with the primer spe-cific for the human hsp70gene showed that the Hsp70 expressing in transfected cells was human.In contrast to protein content,the level of hsp70mRNA estimated by RT-PCR was not affected by heat shock.The discrepancy of the results obtained in two assays may be explained by an enhanced stability of the protein occurred due to its binding with misfolded polypeptides accumulated in heat shocked cells(Guzhova et al.,1997).The analysis of cell growth kinetics and frequency of apoptosis during the long-term cultivation demonstrated that the population of transfected cells was characterized by the prolonged stationary phase and reduced number of apoptotic cells.Higher viability of transfected myeloma cells in culture encouraged us to try the cells as a fusion partner for hy-bridoma production.BALB/c mice were immunized by various antigens and resulted spleen cells were fused with parental and transfected myeloma cells.The total yield of hybrids as well as of antigen-specific hybridomas was two-fold higher when transfected fusion partner was used.Con-siderable part of hybridomas obtained in our experiments (37%)expressed hsp70and the protein content in the cells was demonstrated to be stable along the long-term culture. We tested NS0hsp70-originated hybridomas for cell viability during long-term cultivation.Hsp70-positive hybridomas were demonstrated to be more resistant to spontaneous ap-optosis in culture,while the viability of hsp70-negative hy-bridomas was similar to nontransfected myeloma cells. The protective anti-apoptotic effect of hsp70overexpres-sion observed in our experiments was less pronounced than that described for bcl-2(Simpson et al.,1998)or E1B-19K adenoviral gene(Mercille et al.,1999)overexpressing my-eloma or hybridoma cells.In both cases anti-apoptotic genes protected cells from apoptosis and prolonged stationary phase for additional5–7days in culture in contrast to1–2 days observed in hsp70-transfected cells.However, hsp70—as well as bcl-2—(Ray and Diamond,1994)modi-fied NS0myeloma cells when used as fusion partners pro-vided higher yield of hybridomas while E1B-19K had no effect on hybridoma output(Mercille et al.,1999).Hybri-domas generated from E1B-19K-transfected fusion partner also continued to express viral anti-apoptotic protein,but were no longer resistant to apoptosis(Mercille et al.,1999). Our results proved the protective anti-apoptotic effect of hsp70expression in myeloma and resulted hybridoma cells. The improved cell robustness of hybridoma cells derived from NS0hsp70has likely contributed to their preferential surveillance and selection during fusion and subsequent cloning procedures that resulted in enhanced hybridoma output.During cloning the cells suffer the lack of autocrine growth factors that prone them to undergo apoptosis,there-fore apoptosis-resistant cells have an advantage.The mechanism of Hsp70involvement in apoptosis regu-lation is not yet well established,although recent observa-tions indicate that the protein influences different parts of apoptotic signaling machinery.The data obtained employ-ing the model of heat shock-induced stress demonstrated that inducible Hsp70exerts its protective anti-apoptotic ef-fect by suppressing the activity of stress-kinase JNK(Gabai et al.,1998;Volloch et al.,2000),that could inhibit pro-apoptotic signals mediated by this kinase(Tournier et al., 2000),potentially involving the phosphorylation and inac-tivation of Bcl-2(Yamamoto et al.,1999).It was shown that Hsp70functions to suppress apoptosis predominantly within the mitochondria-mediated pathway to cell death, downstream of cytochrome c release and upstream of caspase-3activation(Li C-Y et al.,2000),specifically,at the level of apoptosome.Hsp70was found to bind APAF-1 and prevent recruitment of caspase-9to the apoptosome complex(Beere et al.,2000;Saleh et al.,2000).To study the effect of enhanced hybridoma viability on mAb production we measured immunoglobulin concentra-tion in culture supernatants of hsp70-positive and negative hybridomas,and compared the increase of mAb titer at the early and later phases of the viable period in culture of each individual hybridoma.There was no difference in mAb pro-502BIOTECHNOLOGY AND BIOENGINEERING,VOL.81,NO.4,FEBRUARY20,2003。