Cytotechnology 8: 57-64, 1992. 9 1992 Kluwer Academic Publishers. Prh~ted h~ the Netherlands.
Viability measurements of hybridoma cells in suspension s Jos6 M. Coco-Martin, Jan W. Oberink, Tiny A.M. van der Velden-de Groot.and E; Coon .Beu,ve~-y National h~stitute o f Public Health and Environmental Protection, P.O. Box 2, 3720 B A Bi!thov, en, 7~he Netherlands Received 14 October 1991; accepted in revised form 1.April 1992
Key ,words: viability, hybridoma, fluorescein diacetate, propid~um-iodide,, flo~v c y t o m e ~
Abstract Several methods were applied to determine the viability o f hybridomacells in suspension,.. These,methods include dye inclusion and exclusion assays such as the classical, trypan blue exclusion ~assay~ thV propidium iodide (PI) exclusion assay and the fluorescein diacetate (FDA)inclusion:,assay,,. F-.u_rthema0~r the relation was studied between release of lactate dehydrogenase {LDH).by ,hybridoma cel.ls.and.!th6i~r viability. Also the ATP content of the cells and cellular heterogeneity as measured with. a, flow cytometer were determined in relation to cellular viability. The dye inclusion and exclusion assays using trypan blue, FDA, PI were shown to:be ~useful-:method~to determine cellular viability. With the FDA and PI methods it was possible to obtain additional information about ceils which are in a transition state between viable and non-viable'., "s .,viabili~,, according to the scatter properties of the cells appears to reflect the overall condition of the ceils, although interpretation of the results is difficult. Measurement of LDH release ,in. the culture ,flt~id~or~;th~ cytoplasmic ATP content could not be used as parameters for cell viability.
Introduction Cultures of mammalian cells are widely used for the production of monoclonal antibodies (Mab) and other biologicals of pharmaceutical and therapeutic interest (Rhodes and Birch, 1988). Culture systems include bioreactors (Birch et al., 1985; Martin et al., 1987), as well as more complex systems, including air lift reactors (Birch e t a l . , 1985) and continuously perfused reactors such as hollow fiber systems (Altshuler et al., 1986). A~. important aspect of these .different culture sys-. terns is the need for reliable and easy to. p e f f o ~ techniques to determine the conditionand viability of the cells during the culture. There are several
methods to determine:cellular.viability,..(C0.ok~and MitcheU, 1989). These maethods, arc. based~ Lon, different., principles like, cellular~ membrane iinteg~; rity (Patterson ~r.,~ :!9.7~)r dcte~rm.inafior~,o~.meta.~ bolic: activity,, morphologicalt changes, :.(I..,aiho.',;e~. aL, 197!)or.cell division (Rope~:an&Dre,v~ko~ 1976). Cellular membrane~int~grity carf(,.l~_.:.dete,r.'-., mined by :dye .ineJusion or.exClusion a'.ssays.,,(S~@i~ et a_l., 1987; Pav.,lik. et ,a!~," 198.3~)~,.~ m~e:.m.l~l_,f'r acfi.vity .to .be:used a#_.ax~iabi~.w,-.p~..~.amc.tr involve m etabolites that~ :~r n~r ~,~c~ at., t987)., and., :the.xate.pf:..Dl~!A~ ~W'~e:~,~k,~r Viability m ea~ .m:em..en.ts9o , 9 ~ b..~i:s~9.~ e ~ p
~
58 phology are based on measurement of membrane blebbing, volume changes caused by cellular osmotic processes or alterations in intracellular structures (Laiho et al., 1971). Morphological features of the cell can also be used to determine the physiological state of the cells during culture (Needham et al., 1991; Goebel et al., 1990). In this paper a comparison is made between different methods to determine cell viability. The object of these investigations was to monitor culture health during continuous culture of hybridoma ceils. The viability measurements included four different permeability assays. Moreover, an assay was used based on the determination of intracellular ATP and another consists of the determination of morphological changes. The permeability assays were based on the exclusion of trypan blue (Patterson Jr., 1979) and propidium iodide (PI) (Sasaki et al., 1987), the inclusion of fluorescein diacetate (FDA) (Schols et al., 1988) and the release of lactate dehydrogenase (LDH) (Racher et al., 1990) by damaged cells.
Materials and methods
Cultivation of hybridoma cell line MN12 Cell line MN12 (gift from Dr. J. Poolman, Dept. of Bacterial Vaccines, National Institute for Public Health and Environmental Protection, The Netherlands), produces a murine IgG2a antibody directed against outer membrane protein P. 1.16 of Neisseria Meningitidis. The cells were cultured in a continuous culture system for 50 days. A 3-liter bioreactor was inoculated with MN12 cells at a cell density of 2 • 10-5 cells per ml. The culture medium was Iscove's modified Dulbecco's medium (Gibco Laboratories, Grand Island, NY) supplemented with 0.25% (w/v) Primatone RL (Sheffield Products, Norwich, NY), 5%, 3% and 2% heat-inactivated (56~ for 30 min) fetal bovine serum (FBS) (Flow Laboratories, Woodcockhill, UK), and antibiotics (35,000 U/1 polymyxin, 14,000 U/I neomycin, and 75,000 U/I streptomycin). An ADI500 control unit (Applikon Dependable Instruments, Schiedam, The Nether-
lands) was used to control pH (7.2), temperature (36.5~ dissolved oxygen (50% air saturation), and stirrer speed (100 rpm). The viability during this culture was determined with the following methods: trypan blue exclusion assay, FDA inclusion assay, PI inclusion assay and by the scatter properties as determined by FC. The determinations of LDH and ATP were performed during the culture of cell line MN12 in a homogeneous continuous perfusion culture for 25 days. A 3-1 bioreactor was inoculated with 2 • 105 cells per ml in Iscove's modified Dulbecco's medium (Gibco Laboratories, Grand Island, NY) supplemented with 0.25% (w/v) Primatone RL (Sheffield Products, Norwich, NY), 3% heatinactivated (56~ for 30 rain) fetal bovine serum (FBS; Sanbio, Uden, The Netherlands) and antibiotics (35,000 U/I polymyxin, 14,000 U/I neomycin, and 75,000 U/1 streptomycin). At a cell density of 106 cells per ml, the perfusion of the medium was started at a flow rate equivalent to one culture volume per 24 h. One-hundred-andsixty-two hours after the start of the culture the FBS concentration was reduced from 3% to 2% (v/v) and after 404 h from 2% to 1.5% (v/v). An ADI1000 (Applikon Dependable Instruments, Schiedam, The Netherlands) control unit was used to control pH (7.2), temperature (36.5~ dissolved oxygen concentration (50% air saturation), and stirrer speed (100-200 rpm). Viability assay methods Ti3,pan blue exclusion assay. Cells were incubated for 5 min with 1% (w/v) trypan blue (Merck, Darmstadt, FRG) in phosphate buffered saline (PBS) pH 7.3, and examined under a microscope. Non-viable cells stained purple-violet whereas viable cells remained unstained. FDA inclusion assay. A stock solution of FDA (Sigma, St. Louis, MO) was prepared at 5 m ~ m l in dimethylsulfoxide (DMSO), and further dilutions were made in PBS. In the assay, 10 p.I of a solution of 5 ~g/ml was added to 1 ml of cell suspension, containing 106 cells. The cell suspension was analyzed on a flow cytometer (FC) (Fac-
59 scan; Becton-Dickinson; Mountain View, CA) after the appropriate incubation time (see Results and discussion section). PI exclusion assay. Cell suspensions were washed twice with PBS. Ten microliters PI (1.9 mg/ml) (Sigma; St. Louis, MO) was added to the cell suspension. The cell suspension was analyzed on a flow cytometer (FC) (Facscan; Becton-Dickinson; Mountain View, CA) after the appropriate incubation time (see Results and discussion section). LDH determination. LDH activity was determined in cell-free medium (cell culture supernatants). The samples were stored at -70~ until determination. LDH was assayed spectrophotometrically at 30~ on a centrifugal analyzer (Cobas-Bio, Hoffman-La Roche, Basel, Switzerland) by following the oxidation of NADH at 340 nm. The reaction was initiated by the addition of pyruvate (J.T. Baker BV, Deventer, The Netherlands). The extinction was measured after 45 s. Duplicate assays were done on each sample. ATP determination. The intracellular ATP concentration was measured using the luciferin-luciferase method (Whitehead et al., 1979). 106 Cells per ml were washed twice with PBS, and incubated for 1 h at 4~ in PBS containing 0.02% Triton X-100 (Sigma, St. Louis, MO) to lyse the cells. After centrifugation, 240 gl of the supematant was added to a tube containing 60 pl of the ATP monitoring reagent (Bio-Orbit, Turku, Finland). The ATP level was read on a luminometer (Biocounter, Lumac, Landgraaf, The Netherlands). The ATP concentration was calculated relative to an ATP standard curve. Scatter properties. Cell culture samples were analyzed with a flow cytometer (FACScan; Becton Dickinson BV, Etten-Leur, The Netherlands). An HP310 computer (Hewlett-Packard Corporation, Pittsburgh, PA) with a Consort 30 program (Becton Dickinson) was used for data processing. Dead cells were gated out according to their forward-angle and right-angle scattering..proper-
ties. Forward scatter,(FSC) andsidescatter (SSC), were measured on a flow cytometer without pretreatment of the samples (Coco-Martin et al., 1992).
Results and discussion Trypan blue excl,tsion assay In this study we have compared different methods to determine cellular viability. These viability measurements are based on different principl'es. Four different methods were based on .the' cel~ lular permeability assessed by inclusion or exclusion of dyes or release of LDH. The f r s t exclusion determination was the conventional trypan blue assay: under a microscope, membrane-dam: aged cells stain purple-violet whereas undamaged cells remain translucid. This method is applied by many investigators as a standard method. Table 1 shows the correlation coefficients of the trypan blue exclusion assay with the other viability' measurements (except LDH and ATP determination). FDA inclusion and PI exclusion assays The FDA inclusion assay is based on the active uptake of FDA by viable cells. All cells take Up FDA which is hydrolysed in the cytoplasm" by non-specific esterase activity to fluorescein (ROtman and Papermaster, 1966). The cleavage prod -~ uct fluorescein is highly fluorescent and' doesnot escape from viable cell~ because Of its polar nature. This results in the intracellular accumUia~ tion of fluorescein that can be determined With an FC. The PI exclusion estimation, employed' Pt .as fluorescent dye that intercalates 'between the. basepairs of double-stranded DNA of.membrane-: damaged cel.ls and becomes highly fhiorescent. The fluorescence Can be determined.using an. FC. Using the FC a large. number ~of:~c611'~ can be evaluated with a good-statistiCal .accu~racy. To compare the.performance~ .of;'the, v.i:abit#y assays using FDAand;PI, the.viabi,lity',w/ts~'~l&ef:
6O Table 1. Correlation coefficients between the different viability measurements methods (n = 43 for each method)
FDA Inclusion PI Exclusion Trypan blue exclusion Scatter properties
FDA Inclusion
PI Exclusion
Trypan blue exclusion
Scatter properties
X 0.83 0.93 0.86
0.83 X 0.79 0.88
0.93 0.79 X 0.80
0.86 0.88 0~80 X
mined as a function o f the incubation time (Fig. 1) at r o o m temperature. Using F D A the m a x i m u m viability value was reached after 5 rain of incubation and remained constant up to 90 rain o f incubation. In contrast, using PI the m a x i m u m viability was observed after 1 rain, but a sharp decrease in viability took place after 10 min o f incubation. So the measurement of the fluorescence using PI should be after about 5 min of incubation. Using FDA, the fluorescence should be measured after at least 5 rain of incubation. The rapid decrease in viability after 10 min using PI could be due to a toxic effect of PI on hybridoma cells. Figure 2 shows the typical dot plots using FDA (Fig. 2A) and PI (Fig. 2B). Three populations can be distinguished. In the case of FDA, population A represents the viable ceils. The percentage o f viable cells represented by this population is defined as the viability. Due to the active incorporation of F D A and its hydrolyzation to fluorescein these cells have become highly fluorescent. The cells o f population B exhibit cellular fluorescence, although lower compared with population A. This lower cellular fluorescence could indicate a lower esterase activity o f these cells. In addition, a partially damaged cell membrane could cause leakage o f fluorescein or of the esterases. The cells in population B seems to be in a transition state between viable and non-viable cells. Population C represents the non-viable cells, which exhibit no cellular fluorescence. Using PI, similar populationscan be distinguished (Fig. 2B). W h e n PI is used the non-viable ceils are highly fluorescent, since PI passes through the permeable cell m e m b r a n e o f the non-viable cells only and subsequent intercalates between the base pairs of the D N A (population C). As in t h e c a s e Of FDA,
there is a population o f cells (B) with a lower cellular fluorescence, indicating the existence o f cells with a partially damaged cell membrane. Finally, population A represents the viable cells, with a high membrane integrity, so PI cannot pass through the cell membrane. The percentage o f viable cells represented by this population is defined as the viability. LDH determination
A release assay (or permeability assay) was used based on the measurement of the e n z y m e L D H in the culture fluid. Figure 3 shows the growth curve (non-viable cells were determined by trypan blue exclusion) and the L D H levels in the culture fluid during the cultivation o f a hybridoma cell line. The L D H level shows a similar pattern as the viable and non-viable cells. The correlation coefficients between the L D H in the culture fluid and the viable 100
0 , ~ ~ "7~ : - ~ - ~ . . - : ~
~-~
.
C
80
60
\
40
20
x , "~-----I
0 0
20
40
-Ip-
n
60
80
I 100
incubation time (man.)
Fig. 1. Effect of the incubation time on the viab~ity as deter-
mined with the F"DAinclusionmethod (O) and the PI exclusion method (n).
non-viableby la:ypan :blUeex elusi0n~,;~i~ ot)~b~ii~.g, able to. release the L D H into., flue: cu,lture ft..did, because .the molecular, wei.ght ~of:~DH' ig :lf4~',.00~ D while : t h a t o f trypan bltae~is: 960),,~:D~ ,R.,'acl~er,,~t al. -.(,"1.990) and. Geaugey:. et.,ali ,(.1,990): ,found'-,:a: significan~ correlation,, between, the,.-~I~-I~,i'n3~.th'e culture' f l u i d and. the.numbcrof,~non'.-,,4'abl~.:cDH'w This, Could be due ,tO a.reduced.synthesis,:o~.~DHO n l y little is: known about the ~synthesiS~o!?>Ll~,':ttI. 'This enzyme plays a key. role,in,".the ,'.path~a,2~-~:Of lactate format, ion~(Geaugey :e~.al,., 'lt9~0):.:Its..aeti;v~ ity in the cytoplasm is likely.:to,beaffeeted)'by parameters, that. influence..the .kine~ies~,of !glueos~ metabolism, such, as.. sugar concentratri0ia, " d ~ -~ solved oxygen, and, pH o f ,the ca,liar, e m~dium,.~ Furthermore; the' intracel;lu,lar prote01ytie~a'~ti~v~, of the cells, should also. be taken",int0; .hOc0tii~.~ (Bienkowski, 1983)'....Thus, the, co,clad'on, be'-.tween ..the L D H .concentration. in 'the. cu,ltur~" flt~d~ and the number of ,non-v, iable cells is not clear.
Fig. 2. Typical dotplots as obtained.after 5 min incubation with FDA (A) and PI (B) at room temperature.
ceils and non-~/iable cells are respectively 0.76 and 0.77. These.results indicate that it is difficult to use L D H levels in the culture fluid as.a viability index. This could be due to cells which release.. their L D H into the :culture .fluid and .then :lyse. Another explanation is the.discrimination>of being,
Fig. 3. Gmw.thcu.ry.e and,LDH )r i,n.th~..cuJtu_r~,.fl~d~u~Dg~ the cu!tivation Of~cl[ line I~INI,?ina~bo.mog~neous.r ' .... :, .. ,"~',:~-6~-.'";( ~ "';~'~'~ ~ | ~ . peffusnon:culture system.Vmble cells~(Q) ~cells~ r ml~ non-.
9
-
~
viable cells {+)(cells per ml) and LDH levels ~'~
- ~ 1
6,'2. A T P determination 20
:The ATP content of cells was determined. Since the. intracellular level of ATP reflects the energetic,status and metabolic activityofcells, ithas been suggested that it is correlated with the biomass of living cells in culture (Kangas et al., 1984). ATP is synthesized within the respiratory cycle (glycolysis, citric acid-cycle, oxidative phosphorylation) and is ,used as basic source of intracellular energy. Thus, the intracellular ATP concentration can be: taken as a parameter of cell viability. Figure,4 shows the growth curve (non-viable cells were determined by trypan blue exclusion) and the intracellular ATP concentration during the cultivation of a hybridoma cell line. The intracellular ATP contentdecreased rapidly from 22 fmol per cell to 4 fmol per cell during exponentional growth and decreased slightly from 4 fmol per cell to 1.5 fmol per cell until the end of the cultivation. However, the ATP content of hybridoma cells showed a poor correlation (r = 0.45) with the trypan blue exclusion assay. Kuzmits et al. (1986) reported a good correlation (r = 0.819) between the ATP content of HeLa cells in suspension and :the percentage of viable cells as determined by the trypan blue exclusion assay. The ATP content of living cells can be dependent upon intra- or extracellular parameters such as metabolic state of the cell, cell cycle stage or culture conditions. Moreover, ATPase activity can cause a reduction of Cytoplasmic ATP content of living cells (Kangas et al., 1984). Thus, it is not to be expected that cytoplasmicATP levels are a useful parameter of cell. viability,
Scatter properties FC was used to determine cellular viability according to the scatter properties m e a s u r e d b y F C . The forward scatter is a measure for cell size and side scatter is a measure of the cell heterogeneity (Coco-Martin et al., 1992). Since viable ceils have a larger relative cell size and are less heterogeneous ~ with respect to cell morphology than non,viable cells (Coco-Martin et al., 1992;~Loken et all, 1976).
25~
"1""20.
mA
.!
E o}
15
I
E ,~
moo E
..Q
10
E
13_ I--
Viability measurements of hybridoma cells in suspension s Jos6 M. Coco-Martin, Jan W. Oberink, Tiny A.M. van der Velden-de Groot.and E; Coon .Beu,ve~-y National h~stitute o f Public Health and Environmental Protection, P.O. Box 2, 3720 B A Bi!thov, en, 7~he Netherlands Received 14 October 1991; accepted in revised form 1.April 1992
Key ,words: viability, hybridoma, fluorescein diacetate, propid~um-iodide,, flo~v c y t o m e ~
Abstract Several methods were applied to determine the viability o f hybridomacells in suspension,.. These,methods include dye inclusion and exclusion assays such as the classical, trypan blue exclusion ~assay~ thV propidium iodide (PI) exclusion assay and the fluorescein diacetate (FDA)inclusion:,assay,,. F-.u_rthema0~r the relation was studied between release of lactate dehydrogenase {LDH).by ,hybridoma cel.ls.and.!th6i~r viability. Also the ATP content of the cells and cellular heterogeneity as measured with. a, flow cytometer were determined in relation to cellular viability. The dye inclusion and exclusion assays using trypan blue, FDA, PI were shown to:be ~useful-:method~to determine cellular viability. With the FDA and PI methods it was possible to obtain additional information about ceils which are in a transition state between viable and non-viable'., "s .,viabili~,, according to the scatter properties of the cells appears to reflect the overall condition of the ceils, although interpretation of the results is difficult. Measurement of LDH release ,in. the culture ,flt~id~or~;th~ cytoplasmic ATP content could not be used as parameters for cell viability.
Introduction Cultures of mammalian cells are widely used for the production of monoclonal antibodies (Mab) and other biologicals of pharmaceutical and therapeutic interest (Rhodes and Birch, 1988). Culture systems include bioreactors (Birch et al., 1985; Martin et al., 1987), as well as more complex systems, including air lift reactors (Birch e t a l . , 1985) and continuously perfused reactors such as hollow fiber systems (Altshuler et al., 1986). A~. important aspect of these .different culture sys-. terns is the need for reliable and easy to. p e f f o ~ techniques to determine the conditionand viability of the cells during the culture. There are several
methods to determine:cellular.viability,..(C0.ok~and MitcheU, 1989). These maethods, arc. based~ Lon, different., principles like, cellular~ membrane iinteg~; rity (Patterson ~r.,~ :!9.7~)r dcte~rm.inafior~,o~.meta.~ bolic: activity,, morphologicalt changes, :.(I..,aiho.',;e~. aL, 197!)or.cell division (Rope~:an&Dre,v~ko~ 1976). Cellular membrane~int~grity carf(,.l~_.:.dete,r.'-., mined by :dye .ineJusion or.exClusion a'.ssays.,,(S~@i~ et a_l., 1987; Pav.,lik. et ,a!~," 198.3~)~,.~ m~e:.m.l~l_,f'r acfi.vity .to .be:used a#_.ax~iabi~.w,-.p~..~.amc.tr involve m etabolites that~ :~r n~r ~,~c~ at., t987)., and., :the.xate.pf:..Dl~!A~ ~W'~e:~,~k,~r Viability m ea~ .m:em..en.ts9o , 9 ~ b..~i:s~9.~ e ~ p
~
58 phology are based on measurement of membrane blebbing, volume changes caused by cellular osmotic processes or alterations in intracellular structures (Laiho et al., 1971). Morphological features of the cell can also be used to determine the physiological state of the cells during culture (Needham et al., 1991; Goebel et al., 1990). In this paper a comparison is made between different methods to determine cell viability. The object of these investigations was to monitor culture health during continuous culture of hybridoma ceils. The viability measurements included four different permeability assays. Moreover, an assay was used based on the determination of intracellular ATP and another consists of the determination of morphological changes. The permeability assays were based on the exclusion of trypan blue (Patterson Jr., 1979) and propidium iodide (PI) (Sasaki et al., 1987), the inclusion of fluorescein diacetate (FDA) (Schols et al., 1988) and the release of lactate dehydrogenase (LDH) (Racher et al., 1990) by damaged cells.
Materials and methods
Cultivation of hybridoma cell line MN12 Cell line MN12 (gift from Dr. J. Poolman, Dept. of Bacterial Vaccines, National Institute for Public Health and Environmental Protection, The Netherlands), produces a murine IgG2a antibody directed against outer membrane protein P. 1.16 of Neisseria Meningitidis. The cells were cultured in a continuous culture system for 50 days. A 3-liter bioreactor was inoculated with MN12 cells at a cell density of 2 • 10-5 cells per ml. The culture medium was Iscove's modified Dulbecco's medium (Gibco Laboratories, Grand Island, NY) supplemented with 0.25% (w/v) Primatone RL (Sheffield Products, Norwich, NY), 5%, 3% and 2% heat-inactivated (56~ for 30 min) fetal bovine serum (FBS) (Flow Laboratories, Woodcockhill, UK), and antibiotics (35,000 U/1 polymyxin, 14,000 U/I neomycin, and 75,000 U/I streptomycin). An ADI500 control unit (Applikon Dependable Instruments, Schiedam, The Nether-
lands) was used to control pH (7.2), temperature (36.5~ dissolved oxygen (50% air saturation), and stirrer speed (100 rpm). The viability during this culture was determined with the following methods: trypan blue exclusion assay, FDA inclusion assay, PI inclusion assay and by the scatter properties as determined by FC. The determinations of LDH and ATP were performed during the culture of cell line MN12 in a homogeneous continuous perfusion culture for 25 days. A 3-1 bioreactor was inoculated with 2 • 105 cells per ml in Iscove's modified Dulbecco's medium (Gibco Laboratories, Grand Island, NY) supplemented with 0.25% (w/v) Primatone RL (Sheffield Products, Norwich, NY), 3% heatinactivated (56~ for 30 rain) fetal bovine serum (FBS; Sanbio, Uden, The Netherlands) and antibiotics (35,000 U/I polymyxin, 14,000 U/I neomycin, and 75,000 U/1 streptomycin). At a cell density of 106 cells per ml, the perfusion of the medium was started at a flow rate equivalent to one culture volume per 24 h. One-hundred-andsixty-two hours after the start of the culture the FBS concentration was reduced from 3% to 2% (v/v) and after 404 h from 2% to 1.5% (v/v). An ADI1000 (Applikon Dependable Instruments, Schiedam, The Netherlands) control unit was used to control pH (7.2), temperature (36.5~ dissolved oxygen concentration (50% air saturation), and stirrer speed (100-200 rpm). Viability assay methods Ti3,pan blue exclusion assay. Cells were incubated for 5 min with 1% (w/v) trypan blue (Merck, Darmstadt, FRG) in phosphate buffered saline (PBS) pH 7.3, and examined under a microscope. Non-viable cells stained purple-violet whereas viable cells remained unstained. FDA inclusion assay. A stock solution of FDA (Sigma, St. Louis, MO) was prepared at 5 m ~ m l in dimethylsulfoxide (DMSO), and further dilutions were made in PBS. In the assay, 10 p.I of a solution of 5 ~g/ml was added to 1 ml of cell suspension, containing 106 cells. The cell suspension was analyzed on a flow cytometer (FC) (Fac-
59 scan; Becton-Dickinson; Mountain View, CA) after the appropriate incubation time (see Results and discussion section). PI exclusion assay. Cell suspensions were washed twice with PBS. Ten microliters PI (1.9 mg/ml) (Sigma; St. Louis, MO) was added to the cell suspension. The cell suspension was analyzed on a flow cytometer (FC) (Facscan; Becton-Dickinson; Mountain View, CA) after the appropriate incubation time (see Results and discussion section). LDH determination. LDH activity was determined in cell-free medium (cell culture supernatants). The samples were stored at -70~ until determination. LDH was assayed spectrophotometrically at 30~ on a centrifugal analyzer (Cobas-Bio, Hoffman-La Roche, Basel, Switzerland) by following the oxidation of NADH at 340 nm. The reaction was initiated by the addition of pyruvate (J.T. Baker BV, Deventer, The Netherlands). The extinction was measured after 45 s. Duplicate assays were done on each sample. ATP determination. The intracellular ATP concentration was measured using the luciferin-luciferase method (Whitehead et al., 1979). 106 Cells per ml were washed twice with PBS, and incubated for 1 h at 4~ in PBS containing 0.02% Triton X-100 (Sigma, St. Louis, MO) to lyse the cells. After centrifugation, 240 gl of the supematant was added to a tube containing 60 pl of the ATP monitoring reagent (Bio-Orbit, Turku, Finland). The ATP level was read on a luminometer (Biocounter, Lumac, Landgraaf, The Netherlands). The ATP concentration was calculated relative to an ATP standard curve. Scatter properties. Cell culture samples were analyzed with a flow cytometer (FACScan; Becton Dickinson BV, Etten-Leur, The Netherlands). An HP310 computer (Hewlett-Packard Corporation, Pittsburgh, PA) with a Consort 30 program (Becton Dickinson) was used for data processing. Dead cells were gated out according to their forward-angle and right-angle scattering..proper-
ties. Forward scatter,(FSC) andsidescatter (SSC), were measured on a flow cytometer without pretreatment of the samples (Coco-Martin et al., 1992).
Results and discussion Trypan blue excl,tsion assay In this study we have compared different methods to determine cellular viability. These viability measurements are based on different principl'es. Four different methods were based on .the' cel~ lular permeability assessed by inclusion or exclusion of dyes or release of LDH. The f r s t exclusion determination was the conventional trypan blue assay: under a microscope, membrane-dam: aged cells stain purple-violet whereas undamaged cells remain translucid. This method is applied by many investigators as a standard method. Table 1 shows the correlation coefficients of the trypan blue exclusion assay with the other viability' measurements (except LDH and ATP determination). FDA inclusion and PI exclusion assays The FDA inclusion assay is based on the active uptake of FDA by viable cells. All cells take Up FDA which is hydrolysed in the cytoplasm" by non-specific esterase activity to fluorescein (ROtman and Papermaster, 1966). The cleavage prod -~ uct fluorescein is highly fluorescent and' doesnot escape from viable cell~ because Of its polar nature. This results in the intracellular accumUia~ tion of fluorescein that can be determined With an FC. The PI exclusion estimation, employed' Pt .as fluorescent dye that intercalates 'between the. basepairs of double-stranded DNA of.membrane-: damaged cel.ls and becomes highly fhiorescent. The fluorescence Can be determined.using an. FC. Using the FC a large. number ~of:~c611'~ can be evaluated with a good-statistiCal .accu~racy. To compare the.performance~ .of;'the, v.i:abit#y assays using FDAand;PI, the.viabi,lity',w/ts~'~l&ef:
6O Table 1. Correlation coefficients between the different viability measurements methods (n = 43 for each method)
FDA Inclusion PI Exclusion Trypan blue exclusion Scatter properties
FDA Inclusion
PI Exclusion
Trypan blue exclusion
Scatter properties
X 0.83 0.93 0.86
0.83 X 0.79 0.88
0.93 0.79 X 0.80
0.86 0.88 0~80 X
mined as a function o f the incubation time (Fig. 1) at r o o m temperature. Using F D A the m a x i m u m viability value was reached after 5 rain of incubation and remained constant up to 90 rain o f incubation. In contrast, using PI the m a x i m u m viability was observed after 1 rain, but a sharp decrease in viability took place after 10 min o f incubation. So the measurement of the fluorescence using PI should be after about 5 min of incubation. Using FDA, the fluorescence should be measured after at least 5 rain of incubation. The rapid decrease in viability after 10 min using PI could be due to a toxic effect of PI on hybridoma cells. Figure 2 shows the typical dot plots using FDA (Fig. 2A) and PI (Fig. 2B). Three populations can be distinguished. In the case of FDA, population A represents the viable ceils. The percentage o f viable cells represented by this population is defined as the viability. Due to the active incorporation of F D A and its hydrolyzation to fluorescein these cells have become highly fluorescent. The cells o f population B exhibit cellular fluorescence, although lower compared with population A. This lower cellular fluorescence could indicate a lower esterase activity o f these cells. In addition, a partially damaged cell membrane could cause leakage o f fluorescein or of the esterases. The cells in population B seems to be in a transition state between viable and non-viable cells. Population C represents the non-viable cells, which exhibit no cellular fluorescence. Using PI, similar populationscan be distinguished (Fig. 2B). W h e n PI is used the non-viable ceils are highly fluorescent, since PI passes through the permeable cell m e m b r a n e o f the non-viable cells only and subsequent intercalates between the base pairs of the D N A (population C). As in t h e c a s e Of FDA,
there is a population o f cells (B) with a lower cellular fluorescence, indicating the existence o f cells with a partially damaged cell membrane. Finally, population A represents the viable cells, with a high membrane integrity, so PI cannot pass through the cell membrane. The percentage o f viable cells represented by this population is defined as the viability. LDH determination
A release assay (or permeability assay) was used based on the measurement of the e n z y m e L D H in the culture fluid. Figure 3 shows the growth curve (non-viable cells were determined by trypan blue exclusion) and the L D H levels in the culture fluid during the cultivation o f a hybridoma cell line. The L D H level shows a similar pattern as the viable and non-viable cells. The correlation coefficients between the L D H in the culture fluid and the viable 100
0 , ~ ~ "7~ : - ~ - ~ . . - : ~
~-~
.
C
80
60
\
40
20
x , "~-----I
0 0
20
40
-Ip-
n
60
80
I 100
incubation time (man.)
Fig. 1. Effect of the incubation time on the viab~ity as deter-
mined with the F"DAinclusionmethod (O) and the PI exclusion method (n).
non-viableby la:ypan :blUeex elusi0n~,;~i~ ot)~b~ii~.g, able to. release the L D H into., flue: cu,lture ft..did, because .the molecular, wei.ght ~of:~DH' ig :lf4~',.00~ D while : t h a t o f trypan bltae~is: 960),,~:D~ ,R.,'acl~er,,~t al. -.(,"1.990) and. Geaugey:. et.,ali ,(.1,990): ,found'-,:a: significan~ correlation,, between, the,.-~I~-I~,i'n3~.th'e culture' f l u i d and. the.numbcrof,~non'.-,,4'abl~.:cDH'w This, Could be due ,tO a.reduced.synthesis,:o~.~DHO n l y little is: known about the ~synthesiS~o!?>Ll~,':ttI. 'This enzyme plays a key. role,in,".the ,'.path~a,2~-~:Of lactate format, ion~(Geaugey :e~.al,., 'lt9~0):.:Its..aeti;v~ ity in the cytoplasm is likely.:to,beaffeeted)'by parameters, that. influence..the .kine~ies~,of !glueos~ metabolism, such, as.. sugar concentratri0ia, " d ~ -~ solved oxygen, and, pH o f ,the ca,liar, e m~dium,.~ Furthermore; the' intracel;lu,lar prote01ytie~a'~ti~v~, of the cells, should also. be taken",int0; .hOc0tii~.~ (Bienkowski, 1983)'....Thus, the, co,clad'on, be'-.tween ..the L D H .concentration. in 'the. cu,ltur~" flt~d~ and the number of ,non-v, iable cells is not clear.
Fig. 2. Typical dotplots as obtained.after 5 min incubation with FDA (A) and PI (B) at room temperature.
ceils and non-~/iable cells are respectively 0.76 and 0.77. These.results indicate that it is difficult to use L D H levels in the culture fluid as.a viability index. This could be due to cells which release.. their L D H into the :culture .fluid and .then :lyse. Another explanation is the.discrimination>of being,
Fig. 3. Gmw.thcu.ry.e and,LDH )r i,n.th~..cuJtu_r~,.fl~d~u~Dg~ the cu!tivation Of~cl[ line I~INI,?ina~bo.mog~neous.r ' .... :, .. ,"~',:~-6~-.'";( ~ "';~'~'~ ~ | ~ . peffusnon:culture system.Vmble cells~(Q) ~cells~ r ml~ non-.
9
-
~
viable cells {+)(cells per ml) and LDH levels ~'~
- ~ 1
6,'2. A T P determination 20
:The ATP content of cells was determined. Since the. intracellular level of ATP reflects the energetic,status and metabolic activityofcells, ithas been suggested that it is correlated with the biomass of living cells in culture (Kangas et al., 1984). ATP is synthesized within the respiratory cycle (glycolysis, citric acid-cycle, oxidative phosphorylation) and is ,used as basic source of intracellular energy. Thus, the intracellular ATP concentration can be: taken as a parameter of cell viability. Figure,4 shows the growth curve (non-viable cells were determined by trypan blue exclusion) and the intracellular ATP concentration during the cultivation of a hybridoma cell line. The intracellular ATP contentdecreased rapidly from 22 fmol per cell to 4 fmol per cell during exponentional growth and decreased slightly from 4 fmol per cell to 1.5 fmol per cell until the end of the cultivation. However, the ATP content of hybridoma cells showed a poor correlation (r = 0.45) with the trypan blue exclusion assay. Kuzmits et al. (1986) reported a good correlation (r = 0.819) between the ATP content of HeLa cells in suspension and :the percentage of viable cells as determined by the trypan blue exclusion assay. The ATP content of living cells can be dependent upon intra- or extracellular parameters such as metabolic state of the cell, cell cycle stage or culture conditions. Moreover, ATPase activity can cause a reduction of Cytoplasmic ATP content of living cells (Kangas et al., 1984). Thus, it is not to be expected that cytoplasmicATP levels are a useful parameter of cell. viability,
Scatter properties FC was used to determine cellular viability according to the scatter properties m e a s u r e d b y F C . The forward scatter is a measure for cell size and side scatter is a measure of the cell heterogeneity (Coco-Martin et al., 1992). Since viable ceils have a larger relative cell size and are less heterogeneous ~ with respect to cell morphology than non,viable cells (Coco-Martin et al., 1992;~Loken et all, 1976).
25~
"1""20.
mA
.!
E o}
15
I
E ,~
moo E
..Q
10
E
13_ I--
