Cytotechnology 4: 13-28, 1990. 9 1990KluwerAcademicPublishers. Printedin the Netherlands.
Electron microscopy of hybridoma cells with special regard to monoclonal antibody production
M. A1-Rubeai 1, D. Mills 2 and A.N. Emery 1 1School of Chemical Engineering and 2School of Biochemistry, University of Birmingham, P.O. Box 363, Birmingham B15 2TT, UK Received 10 November1989; acceptedin revisedform 3 January 1990
Key words: monoclonal antibody, hybridoma, electron microscopy, endoplasmic reticulum. Abstract
Electron microscopy of mouse hybridoma cell lines shows that the major difference between non, low and high producer cell lines is the amount of endoplasmic reticulum. Vesicular-tubular or cavernous structures of endoplasmic reticulum, which can survive long after cell death, are particularly abundant in producer cell lines. Immunogold labelling with anti-mouse IgG reveals that antibodies are predominantly located in these structures. The cell membrane undergoes structural changes during the late stages of batch culture with the disappearance of microvilli and the appearance of blebs and deep indentations. Necrosis disrupts the cytoplasmic structures and the nucleus is last to degrade.
Introduction
With the advent of monoclonal antibodies as diagnostic agents, the need to produce large quantifies of hybridoma cells secreting high levels of monoclonal antibodies has attained significant commercial importance and considerable attention has been given to optimizing culture conditions. However, the various cellular steps in the production of monoclonal antibodies - synthesis, intracellular accumulation and export remain largely incompletely understood. Studies of immunoglobulin (Ig) synthesis and secretion by hybridoma cells in culture are important for monoclonal antibody production optimization. Variation exists between cell lines in terms of productivity, production pattern, genetic stability and cellular properties. Electron microscopy can
be a useful tool in studies of cellular activities such as the nature of the intracellular Ig distribution, the mechanism and route by which Ig molecules are exported from cells into the extracellular fluid and for detection and quantitation of viral particles. In a series of experiments, the synthesis and secretion of immunoglobulins in synchronised and asynchronised cultures of hybridoma cells have been studied (A1-Rubeai et al., 1989; A1-Rubeai and Emery, 1989). The aim was underlined by the fact that maximisation of Ig product may well depend on a clear understanding of the control of product formation and the route by which these products are exported from hybridoma cells. In short monoclonal antibody production is primarily a cell biological phenomenon. Understanding such multiple processing requires multiple biological approaches. In this
14 paper we aim, by using ultra-structural techniques to provide a detailed description of hybridoma cells, to determine the extent of subcellular variation between producer and nonproducer cell lines, and to define the relation between the cellular structures and product formarion.
Materials and methods The following murine hybridomas were used in this work: TB]C3 and WC2 (Jefferis et al., 1982) producing antibodies to the C2 region of human IgG. HPV producing antibodies against a defined 8 amino acid epitope of the E4 protein of the Human HPV. 1 Papilloma Virus (Doorbar et al., 1988). PQXB1]2 and PQXB2/2 producing antibodies against paraquat. EBNA, a non-producing cell line. All cell lines were maintained in RPMI 1640 supplemented with 5% FCS and collected by centrifugation at 1000 rpm for 5 minutes. To compare antibody productivity cells were seeded at 2 x 105/ml in 50 ml RPMI + 5% FCS into flasks and incubated at 37~ for 48 hr. Antibodies in the supematants were determined by the sandwich-type Enzyme-linked Immuno Sorbent Assay (ELISA) using mouse IgG coated plates and sheep anti-mouse IgG peroxidase conjugate as the second antibody. Cell number/ml x 10 -5 and antibody concentration (gg/ml) at the exponential phase are as follows: TB/C3, 4.9 (13.3); WC2, 4.2 (6.9); PQXB1, 6.6 (10.4); PQXB2, 7.9 (0.3); HPV, 4.6 (0.5); EBNA, 9.8 (0.01). To follow the destruction of cell structures during and after cell death, cells were immobilised in alginate beads. Calcium Alginate beads were formed by pumping alginate cell suspension through a manifold with nine 21-gauge needles positioned above a 50 mM CaC1 solution contained in a Bellco bioreactor. The beads were washed with buffer and incubated in the bioreactor for 50 days.
Transmission electron microscopy Cells from the exponential phase (2 days) of static culture were fixed in 2.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4 for 1-3 hr. Samples were washed twice with 0.1 M cacodylate buffer; postfixed in Osmium tetroxide for 1 hr; washed 3 times with 0.05 M cacodylate buffer; washed with 1% Tannic acid in 0.05 M cacodylate buffer, pH 7.0 for 30 min; washed 5 times in 1% sodium sulphate in 0.05 M cacodylate buffer, and dehydrated in a graded series of solutions of ethanol and 100% propylene oxide. Thin sections of samples embedded in epoxy resin were stained with uranyl acetate and lead citrate and viewed in a Philips EM301 electron microscope.
Scanning electron microscopy Cells were fixed in 2.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4 for 1-3 hours. After initial fixation cells were sedimented down onto glutaraldehyde fixed gelatine cover slips and washed twice with 0.1 M cacodylate buffer for 15 min. The samples were postfixed in Osmium Tetroxide for 1 hour, washed in buffer and dehydrated in a graded series of acetone. Samples were dried, coated with gold and viewed in a Hitachi scanning electron microscope.
Immunogold labelling Cells were fixed in 1.5% glutaraldehyde in cacodylate buffer (0.1M) containing sucrose (0.25M), pH 7.2. Cells were washed twice after fixation in 0.1 M cacodylate buffer containing 0.25 M sucrose, dehydrated in ethanol series at decreasing temperature and resuspended in Lowacryl. After UV polymerization the blocks were sectioned and stored on grids. Grids were transferred to diluted immunogold label (Goat anti-mouse IgG, 15 nm (BioCell, Cardiff), 1:10 in 0.1 M phosphate buffer, pH 7.2) for 2 hr, washed 6 times in phosphate buffer, washed twice in distilled water,
15
Fig. 1. Electronmicrographof TB/C3 hybridomacell. The cell is characterisedby a large nucleus (n) and a well-developed(and often
distended) endoplasmicreticulum (er). Bar marker represents 1 gm; v, type A virus particle. blotted and stained with uranyl acetate and lead citrate.
Results Hybridoma cells are usually fairly uniform in size (Fig. 1). Most of the cells are round, but ovoid cells are occasionally seen and irregularly shaped cells are not uncommon. Small, finger-like processes or microvilli are commonly formed by the plasma membrane. Nuclei are usually large and round but irregular, and multi-segmented nuclei are frequently seen. Chromatin is clumped and located mainly near the nuclear envelope. The distribution of the chromatin alternates with that of the nuclear pores. One or two vacuole-like structures containing oval-shaped bodies and usually associated with the dense chromatin are frequently seen in the nucleus.
The endoplasmic reticulum (ER) from a single cell line usually has vesicular, tubular or cavernous structures (Figs. 1 and 2a). The amount and appearance of such structures varies markedly from one cell line to another (Fig. 2). In producer cell lines the most frequent arrangement is a vesicular-tubular and cavernous roughened profile. Large cytoplasmic processes are frequently observed at the periphery of the cell. In many cells the endoplasmic reticulum vesicles are greatly dilated leaving little cytoplasmic spaces between them. Sometimes this space is quite narrow (Fig. 3a), so that the ER membrane and the plasma membrane (PM) are immediately adjacent to each other. At times fusion of ER membrane with PM has apparently occurred and then the density within the peripheral ER vesicle is much less than in the cytoplasm and is quite similar to the electron density of the extracellular milieu (Fig. 3b). This suggests that the contents
16
Fig. 2. Electron micrograph of hybridoma cells from producer and non producer cell lines. Typically, the only difference is the amount of the vesicular structures of endoplasmic reticulum. a. PQXB 1/2, a high producer cell line. b. WC2, a high producer cell line.
17
Fig. 2. Electron micrograph of hybridoma cells from producer and non producer celt lines. Typically, the only difference is the amount of the vesicular structures of endoplasmic reticulum. c. HPV, a low producer cell line. d. PQXB2/2, a tow producer cell line.
18
Fig. 2. Electron micrograph of hybridoma cells from producer and non producer cell lines. Typically, the only difference is the amount of the vesicular structures of endoplasmic reticulum. e. EBNA, a non producer cell line.
Fig. 3. Secretory vesicles and protein secretion in hybridoma cells. a. An early stage of protein secretion in WC2 hybridoma cell. Large vesicles in the cytoplasm. The arrow indicates the region where the vesicle membrane fuse with the plasma membrane. V, type A virus particle.
19
Fig. 3. Secretory vesicles and protein secretion in hybridoma cells. b. Portions of TB/C3 hybridoma cell with several endoplasmic vesicles. At the cell surface there are places (arrows) where the vesicle membranes are opened. Note that several of the vesicles have a light electron density which is comparable to the external space. This indicates that the immunoglobulin contents (indicated by the immunogold label) have been released.
Fig. 4. Electron micrographs of TB/C3 hybridoma cells subjected to an immunogold reaction using goat anti-mouse IgG. a. Portion of cell with extreme dilation of vesicle.
20
Fig. 4. Electron micrographs of TB/C3 hybridoma cells subjected to an immunogold reaction using goat anti-mouse IgG. b. Enlarged section of cell. The immunogold label is found only in the dilated vesicles. c. Vesicles restricted to the periphery are about to release their contents into the external space. The membrane of one vesicle (arrow) is apparently ruptured.
21
Fig. 5. The death of a TB/C3 hybridomacell.
a. Condensationof cellular structures during early stage of cell death. The appearance of srnaUvacuolesrepresents the beginning of cytoplasmic fragmentation. of these marginal vesicles have been released from the cells. Using immunoelectron microscopy, gold labels are specifically located in ER structures especially in the large vesicular structures (Figs. 3b and 4), however, the appearance of the labels within the ER structures is quite clearly related to whether the cell line is a producer or not. The labels are only observed in moderately to markedly dilated ER (ie vesicular and cavernous structures) which in turn only appear in producer cell lines. Another feature of the cell lines studied is the presence of many cytoplasmic A-type viral particles (Figs. 1 and 3a). The particles average 75 nm in diameter, with a very low electron density center surrounded by an electron dense envelope. A and C-type particles are observed budding from the plasma membrane, (micrograph is not shown). There are no obvious changes in cell ultra-
structure during the growth phases of batch culture. However, during the decline phase dying cells appear to be darker with condensed ground cyto- and nucleoplasm (Fig. 5). Dying cells show disruption of external and internal membranes and total disorganisation of cytoplasmic organelles. Early in the death process the nuclear morphology is little disturbed with various degrees of chromatin clumps appearing in the periphery. Eventually, the nucleus is destroyed. Sections of immobilised cells in alginate beads reveal that vesicles containing IgG survive long after cell death probably protecting their proteinous contents from proteolytic attack (Fig. 6). However, most vesicles are destroyed with time leaving ghostly structures scattered throughout the cytoplasm. SEM micrographs (Fig. 7) clearly indicate that the cell membrane undergoes some structural changes during various phases of batch culture. Specialised features such as microvilli
22
Fig. 5. The death of a TB/C3 hybridoma cell. b. Section through dying cell showing the degradation of cytoplasmic fine structures and condensation of chromatin. c. Later stage in cell death showing the diffuse degeneration of cellular contents and dilation of the nuclear membrane (arrow).
23
Fig. 6. Portionof dead TB/C3 hybridomacell immobilisedin alginatebead. Note the intactIgG - containingvesicles (arrows) in the
degraded cytoplasm.IgG is labelledwith immunogoldanti-mouseIgG. are gradually lost. Blebs of various sizes and deep indentations in the membrane appear, leading eventually to fragmentation of cells into membrane-bounded small bodies of various sizes. Some cell surfaces at day 1 are totally obscured by microvilli. On the other hand cell surfaces at day 3 are either smooth or extensively blistered. Branching of some microvilli is apparent the latter seeming sometimes to emerge from small flat protuberances of the cell surface.
Discussion Most cells from producer cell lines show clearly the presence of large numbers of various sizes of vesicles. These vesicles are shown to contain IgG molecules by specific immunogold labelling. In some cells the vesicles are so large that they occupy most of the cytoplasm which would suggest that the transport of these vesicles is a rate
limiting step in mltibody secretion. This observation is in contrast to claims that secretion in hybridoma is mainly constitutive (i.e. protein is released immediately after synthesis) (Bienkowski, 1983). Why do cells have to store so much of the synthesized IgG as would be the case in regulated secretion? Furthermore the secretion rate has not been found to be maintained at constant levels but in a fashion typical of programmed cells (A1-Rubeai et al., 1989). Although we failed to observe any significant changes in the ultrastrucmre of cells during the growth phase of batch culture, both TEM and SEM studies indicate a process of cell death during the decline phase induced by environmental changes to which the cell cannot respond. Of course, full statistically sound quantification differences in cell populations might be obtained, but only at the expense of an unrealistically large number of sections and analyses. The ultrastructural and morphological changes during the con-
24
Fig. 7. Series of scanning electron micrographs of TB/C3 hybridoma cell in batch suspension culture. a and b Day 1 in culture. (mi), microvilli.
25
Fig. 7. Series of scanning electron micrographs of TB/C3 hybridoma cell in batch suspension culture. c and d Day 2 in culture.
26
Fig. 7. Series of scanning electron micrographs of TB/C3 hybridoma cell in batch suspension culture. e and f Day 3 in culture. (nc), necrotic cell; (mbv), membrane-bounded vesicle.
27
Fig. 7. Series of scanningelectronmicrographsof TB/C3 hybridomacell in batch suspensionculture.
g Day 3 in culture. tinuous progression from growth to death processes in batch culture appear to be more consistent with the apoptosis pattern of cell death than with the necrosis one (Wyllie, 1981; Wyllie et al., 1984), although the death process may well include both patterns. Wyllie (1981) presented a detailed description of both patterns of cell death. In short, a cell undergoing necrosis is characterised by cellular oedema and culminates in dissolution of cytoplasmic organelles, rupture of plasma and internal membranes and leakage of cellular contents into the extracellular space. However in epoptosis the cell lose their microvilli, the cytoplasm becomes condensed and nuclear chromatin marginates into one or several large masses. The cell transiently adopts a deeply convoluted outline and subsequently breaks up into several membrane-bounded smooth-surfaced 'apopotic bodies' The shedding of membranous material by mammalian cells was reported earlier by Biberfield (1971). He found that lymphocytes bleb off
vesiculated buds when activated in immunological reactions or as a result of phytohaemagglutinin stimulation in vivo or in vitro. Sheldrake (1974) suggested that the loss of membranous material might be of considerable importance in enabling cells to rid themselves of the accumulated deleterious breakdown products of membrane lipids resulting from cellular senescence. In hybridoma culture the morphological changes during cell death are accompanied by a fall-off in incorporation of tritiated thymidine and 358methionine which may be taken as an indication of the loss of reproductive viability and protein metabolism (A1-Rubeai & Emery, to be published).
References 1. Al-Rubeai MA, Rookes S and Emery AN (1989) Flow cytometric studies during synchronous and asynchronous
28
2.
3.
4. 5.
suspension cultures of hybridoma cells. In: Spier RE, Griffiths JB, Stephenne J and Crooy PJ (eds.) Advances in Animal Cell Biology and Technology for Bioprocesses. Butterworths, London, pp. 241-245. A1-RubeaiMA and Emery AN (1989) Monoclonal antibody accumulation and release observed in sub-cellular structures in synchronous and asynchronous hybridoma culture. Cytotechnology $9 (Abstract). Biberfield P (1971) Uropod formation in phytohaemagglutinin (PHA) stimulated lymphocytes, Expl. Cell Res. 66: 433-445. Bienkowski RS (1983) Intracellular degradation of newly synthesised secretory proteins. Biochem. J. 214: 1-10. Doorbar J, Evans HS, Coneron J, Crawford LV and Gallimore PH (1988) Analysis of HPV-1 E4 gene expression using epitope defined antibodies. EMBO J. 7: 825-832.
6. Sheldrake AR (1974) The ageing, growth and death of ceils. Nature 250: 381-385. 7. Wyllie AH (1981) Cell death: a new classification separating apoptosis from necrosis. In: Cell death in biology and pathology. Bowen ID and Lockshin RA (eds.) Chapman and Hall, London, pp. 9-34. 8. Wyllie AH, Dnvall E and Blow JJ (t984) Intracellular mechanisms in cell death in normal and pathological tissues. In: Cell Ageing and Cell Death. Davies I and Siegg DC (eds.) Cambridge University Press, Cambridge, pp. 269294.
Address for offprints: M. A1-Rubeai,School of Chemical Engineering, University of Birmingham, P.O. Box 363, Birmingham B15 2TT, UK
Electron microscopy of hybridoma cells with special regard to monoclonal antibody production
M. A1-Rubeai 1, D. Mills 2 and A.N. Emery 1 1School of Chemical Engineering and 2School of Biochemistry, University of Birmingham, P.O. Box 363, Birmingham B15 2TT, UK Received 10 November1989; acceptedin revisedform 3 January 1990
Key words: monoclonal antibody, hybridoma, electron microscopy, endoplasmic reticulum. Abstract
Electron microscopy of mouse hybridoma cell lines shows that the major difference between non, low and high producer cell lines is the amount of endoplasmic reticulum. Vesicular-tubular or cavernous structures of endoplasmic reticulum, which can survive long after cell death, are particularly abundant in producer cell lines. Immunogold labelling with anti-mouse IgG reveals that antibodies are predominantly located in these structures. The cell membrane undergoes structural changes during the late stages of batch culture with the disappearance of microvilli and the appearance of blebs and deep indentations. Necrosis disrupts the cytoplasmic structures and the nucleus is last to degrade.
Introduction
With the advent of monoclonal antibodies as diagnostic agents, the need to produce large quantifies of hybridoma cells secreting high levels of monoclonal antibodies has attained significant commercial importance and considerable attention has been given to optimizing culture conditions. However, the various cellular steps in the production of monoclonal antibodies - synthesis, intracellular accumulation and export remain largely incompletely understood. Studies of immunoglobulin (Ig) synthesis and secretion by hybridoma cells in culture are important for monoclonal antibody production optimization. Variation exists between cell lines in terms of productivity, production pattern, genetic stability and cellular properties. Electron microscopy can
be a useful tool in studies of cellular activities such as the nature of the intracellular Ig distribution, the mechanism and route by which Ig molecules are exported from cells into the extracellular fluid and for detection and quantitation of viral particles. In a series of experiments, the synthesis and secretion of immunoglobulins in synchronised and asynchronised cultures of hybridoma cells have been studied (A1-Rubeai et al., 1989; A1-Rubeai and Emery, 1989). The aim was underlined by the fact that maximisation of Ig product may well depend on a clear understanding of the control of product formation and the route by which these products are exported from hybridoma cells. In short monoclonal antibody production is primarily a cell biological phenomenon. Understanding such multiple processing requires multiple biological approaches. In this
14 paper we aim, by using ultra-structural techniques to provide a detailed description of hybridoma cells, to determine the extent of subcellular variation between producer and nonproducer cell lines, and to define the relation between the cellular structures and product formarion.
Materials and methods The following murine hybridomas were used in this work: TB]C3 and WC2 (Jefferis et al., 1982) producing antibodies to the C2 region of human IgG. HPV producing antibodies against a defined 8 amino acid epitope of the E4 protein of the Human HPV. 1 Papilloma Virus (Doorbar et al., 1988). PQXB1]2 and PQXB2/2 producing antibodies against paraquat. EBNA, a non-producing cell line. All cell lines were maintained in RPMI 1640 supplemented with 5% FCS and collected by centrifugation at 1000 rpm for 5 minutes. To compare antibody productivity cells were seeded at 2 x 105/ml in 50 ml RPMI + 5% FCS into flasks and incubated at 37~ for 48 hr. Antibodies in the supematants were determined by the sandwich-type Enzyme-linked Immuno Sorbent Assay (ELISA) using mouse IgG coated plates and sheep anti-mouse IgG peroxidase conjugate as the second antibody. Cell number/ml x 10 -5 and antibody concentration (gg/ml) at the exponential phase are as follows: TB/C3, 4.9 (13.3); WC2, 4.2 (6.9); PQXB1, 6.6 (10.4); PQXB2, 7.9 (0.3); HPV, 4.6 (0.5); EBNA, 9.8 (0.01). To follow the destruction of cell structures during and after cell death, cells were immobilised in alginate beads. Calcium Alginate beads were formed by pumping alginate cell suspension through a manifold with nine 21-gauge needles positioned above a 50 mM CaC1 solution contained in a Bellco bioreactor. The beads were washed with buffer and incubated in the bioreactor for 50 days.
Transmission electron microscopy Cells from the exponential phase (2 days) of static culture were fixed in 2.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4 for 1-3 hr. Samples were washed twice with 0.1 M cacodylate buffer; postfixed in Osmium tetroxide for 1 hr; washed 3 times with 0.05 M cacodylate buffer; washed with 1% Tannic acid in 0.05 M cacodylate buffer, pH 7.0 for 30 min; washed 5 times in 1% sodium sulphate in 0.05 M cacodylate buffer, and dehydrated in a graded series of solutions of ethanol and 100% propylene oxide. Thin sections of samples embedded in epoxy resin were stained with uranyl acetate and lead citrate and viewed in a Philips EM301 electron microscope.
Scanning electron microscopy Cells were fixed in 2.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4 for 1-3 hours. After initial fixation cells were sedimented down onto glutaraldehyde fixed gelatine cover slips and washed twice with 0.1 M cacodylate buffer for 15 min. The samples were postfixed in Osmium Tetroxide for 1 hour, washed in buffer and dehydrated in a graded series of acetone. Samples were dried, coated with gold and viewed in a Hitachi scanning electron microscope.
Immunogold labelling Cells were fixed in 1.5% glutaraldehyde in cacodylate buffer (0.1M) containing sucrose (0.25M), pH 7.2. Cells were washed twice after fixation in 0.1 M cacodylate buffer containing 0.25 M sucrose, dehydrated in ethanol series at decreasing temperature and resuspended in Lowacryl. After UV polymerization the blocks were sectioned and stored on grids. Grids were transferred to diluted immunogold label (Goat anti-mouse IgG, 15 nm (BioCell, Cardiff), 1:10 in 0.1 M phosphate buffer, pH 7.2) for 2 hr, washed 6 times in phosphate buffer, washed twice in distilled water,
15
Fig. 1. Electronmicrographof TB/C3 hybridomacell. The cell is characterisedby a large nucleus (n) and a well-developed(and often
distended) endoplasmicreticulum (er). Bar marker represents 1 gm; v, type A virus particle. blotted and stained with uranyl acetate and lead citrate.
Results Hybridoma cells are usually fairly uniform in size (Fig. 1). Most of the cells are round, but ovoid cells are occasionally seen and irregularly shaped cells are not uncommon. Small, finger-like processes or microvilli are commonly formed by the plasma membrane. Nuclei are usually large and round but irregular, and multi-segmented nuclei are frequently seen. Chromatin is clumped and located mainly near the nuclear envelope. The distribution of the chromatin alternates with that of the nuclear pores. One or two vacuole-like structures containing oval-shaped bodies and usually associated with the dense chromatin are frequently seen in the nucleus.
The endoplasmic reticulum (ER) from a single cell line usually has vesicular, tubular or cavernous structures (Figs. 1 and 2a). The amount and appearance of such structures varies markedly from one cell line to another (Fig. 2). In producer cell lines the most frequent arrangement is a vesicular-tubular and cavernous roughened profile. Large cytoplasmic processes are frequently observed at the periphery of the cell. In many cells the endoplasmic reticulum vesicles are greatly dilated leaving little cytoplasmic spaces between them. Sometimes this space is quite narrow (Fig. 3a), so that the ER membrane and the plasma membrane (PM) are immediately adjacent to each other. At times fusion of ER membrane with PM has apparently occurred and then the density within the peripheral ER vesicle is much less than in the cytoplasm and is quite similar to the electron density of the extracellular milieu (Fig. 3b). This suggests that the contents
16
Fig. 2. Electron micrograph of hybridoma cells from producer and non producer cell lines. Typically, the only difference is the amount of the vesicular structures of endoplasmic reticulum. a. PQXB 1/2, a high producer cell line. b. WC2, a high producer cell line.
17
Fig. 2. Electron micrograph of hybridoma cells from producer and non producer celt lines. Typically, the only difference is the amount of the vesicular structures of endoplasmic reticulum. c. HPV, a low producer cell line. d. PQXB2/2, a tow producer cell line.
18
Fig. 2. Electron micrograph of hybridoma cells from producer and non producer cell lines. Typically, the only difference is the amount of the vesicular structures of endoplasmic reticulum. e. EBNA, a non producer cell line.
Fig. 3. Secretory vesicles and protein secretion in hybridoma cells. a. An early stage of protein secretion in WC2 hybridoma cell. Large vesicles in the cytoplasm. The arrow indicates the region where the vesicle membrane fuse with the plasma membrane. V, type A virus particle.
19
Fig. 3. Secretory vesicles and protein secretion in hybridoma cells. b. Portions of TB/C3 hybridoma cell with several endoplasmic vesicles. At the cell surface there are places (arrows) where the vesicle membranes are opened. Note that several of the vesicles have a light electron density which is comparable to the external space. This indicates that the immunoglobulin contents (indicated by the immunogold label) have been released.
Fig. 4. Electron micrographs of TB/C3 hybridoma cells subjected to an immunogold reaction using goat anti-mouse IgG. a. Portion of cell with extreme dilation of vesicle.
20
Fig. 4. Electron micrographs of TB/C3 hybridoma cells subjected to an immunogold reaction using goat anti-mouse IgG. b. Enlarged section of cell. The immunogold label is found only in the dilated vesicles. c. Vesicles restricted to the periphery are about to release their contents into the external space. The membrane of one vesicle (arrow) is apparently ruptured.
21
Fig. 5. The death of a TB/C3 hybridomacell.
a. Condensationof cellular structures during early stage of cell death. The appearance of srnaUvacuolesrepresents the beginning of cytoplasmic fragmentation. of these marginal vesicles have been released from the cells. Using immunoelectron microscopy, gold labels are specifically located in ER structures especially in the large vesicular structures (Figs. 3b and 4), however, the appearance of the labels within the ER structures is quite clearly related to whether the cell line is a producer or not. The labels are only observed in moderately to markedly dilated ER (ie vesicular and cavernous structures) which in turn only appear in producer cell lines. Another feature of the cell lines studied is the presence of many cytoplasmic A-type viral particles (Figs. 1 and 3a). The particles average 75 nm in diameter, with a very low electron density center surrounded by an electron dense envelope. A and C-type particles are observed budding from the plasma membrane, (micrograph is not shown). There are no obvious changes in cell ultra-
structure during the growth phases of batch culture. However, during the decline phase dying cells appear to be darker with condensed ground cyto- and nucleoplasm (Fig. 5). Dying cells show disruption of external and internal membranes and total disorganisation of cytoplasmic organelles. Early in the death process the nuclear morphology is little disturbed with various degrees of chromatin clumps appearing in the periphery. Eventually, the nucleus is destroyed. Sections of immobilised cells in alginate beads reveal that vesicles containing IgG survive long after cell death probably protecting their proteinous contents from proteolytic attack (Fig. 6). However, most vesicles are destroyed with time leaving ghostly structures scattered throughout the cytoplasm. SEM micrographs (Fig. 7) clearly indicate that the cell membrane undergoes some structural changes during various phases of batch culture. Specialised features such as microvilli
22
Fig. 5. The death of a TB/C3 hybridoma cell. b. Section through dying cell showing the degradation of cytoplasmic fine structures and condensation of chromatin. c. Later stage in cell death showing the diffuse degeneration of cellular contents and dilation of the nuclear membrane (arrow).
23
Fig. 6. Portionof dead TB/C3 hybridomacell immobilisedin alginatebead. Note the intactIgG - containingvesicles (arrows) in the
degraded cytoplasm.IgG is labelledwith immunogoldanti-mouseIgG. are gradually lost. Blebs of various sizes and deep indentations in the membrane appear, leading eventually to fragmentation of cells into membrane-bounded small bodies of various sizes. Some cell surfaces at day 1 are totally obscured by microvilli. On the other hand cell surfaces at day 3 are either smooth or extensively blistered. Branching of some microvilli is apparent the latter seeming sometimes to emerge from small flat protuberances of the cell surface.
Discussion Most cells from producer cell lines show clearly the presence of large numbers of various sizes of vesicles. These vesicles are shown to contain IgG molecules by specific immunogold labelling. In some cells the vesicles are so large that they occupy most of the cytoplasm which would suggest that the transport of these vesicles is a rate
limiting step in mltibody secretion. This observation is in contrast to claims that secretion in hybridoma is mainly constitutive (i.e. protein is released immediately after synthesis) (Bienkowski, 1983). Why do cells have to store so much of the synthesized IgG as would be the case in regulated secretion? Furthermore the secretion rate has not been found to be maintained at constant levels but in a fashion typical of programmed cells (A1-Rubeai et al., 1989). Although we failed to observe any significant changes in the ultrastrucmre of cells during the growth phase of batch culture, both TEM and SEM studies indicate a process of cell death during the decline phase induced by environmental changes to which the cell cannot respond. Of course, full statistically sound quantification differences in cell populations might be obtained, but only at the expense of an unrealistically large number of sections and analyses. The ultrastructural and morphological changes during the con-
24
Fig. 7. Series of scanning electron micrographs of TB/C3 hybridoma cell in batch suspension culture. a and b Day 1 in culture. (mi), microvilli.
25
Fig. 7. Series of scanning electron micrographs of TB/C3 hybridoma cell in batch suspension culture. c and d Day 2 in culture.
26
Fig. 7. Series of scanning electron micrographs of TB/C3 hybridoma cell in batch suspension culture. e and f Day 3 in culture. (nc), necrotic cell; (mbv), membrane-bounded vesicle.
27
Fig. 7. Series of scanningelectronmicrographsof TB/C3 hybridomacell in batch suspensionculture.
g Day 3 in culture. tinuous progression from growth to death processes in batch culture appear to be more consistent with the apoptosis pattern of cell death than with the necrosis one (Wyllie, 1981; Wyllie et al., 1984), although the death process may well include both patterns. Wyllie (1981) presented a detailed description of both patterns of cell death. In short, a cell undergoing necrosis is characterised by cellular oedema and culminates in dissolution of cytoplasmic organelles, rupture of plasma and internal membranes and leakage of cellular contents into the extracellular space. However in epoptosis the cell lose their microvilli, the cytoplasm becomes condensed and nuclear chromatin marginates into one or several large masses. The cell transiently adopts a deeply convoluted outline and subsequently breaks up into several membrane-bounded smooth-surfaced 'apopotic bodies' The shedding of membranous material by mammalian cells was reported earlier by Biberfield (1971). He found that lymphocytes bleb off
vesiculated buds when activated in immunological reactions or as a result of phytohaemagglutinin stimulation in vivo or in vitro. Sheldrake (1974) suggested that the loss of membranous material might be of considerable importance in enabling cells to rid themselves of the accumulated deleterious breakdown products of membrane lipids resulting from cellular senescence. In hybridoma culture the morphological changes during cell death are accompanied by a fall-off in incorporation of tritiated thymidine and 358methionine which may be taken as an indication of the loss of reproductive viability and protein metabolism (A1-Rubeai & Emery, to be published).
References 1. Al-Rubeai MA, Rookes S and Emery AN (1989) Flow cytometric studies during synchronous and asynchronous
28
2.
3.
4. 5.
suspension cultures of hybridoma cells. In: Spier RE, Griffiths JB, Stephenne J and Crooy PJ (eds.) Advances in Animal Cell Biology and Technology for Bioprocesses. Butterworths, London, pp. 241-245. A1-RubeaiMA and Emery AN (1989) Monoclonal antibody accumulation and release observed in sub-cellular structures in synchronous and asynchronous hybridoma culture. Cytotechnology $9 (Abstract). Biberfield P (1971) Uropod formation in phytohaemagglutinin (PHA) stimulated lymphocytes, Expl. Cell Res. 66: 433-445. Bienkowski RS (1983) Intracellular degradation of newly synthesised secretory proteins. Biochem. J. 214: 1-10. Doorbar J, Evans HS, Coneron J, Crawford LV and Gallimore PH (1988) Analysis of HPV-1 E4 gene expression using epitope defined antibodies. EMBO J. 7: 825-832.
6. Sheldrake AR (1974) The ageing, growth and death of ceils. Nature 250: 381-385. 7. Wyllie AH (1981) Cell death: a new classification separating apoptosis from necrosis. In: Cell death in biology and pathology. Bowen ID and Lockshin RA (eds.) Chapman and Hall, London, pp. 9-34. 8. Wyllie AH, Dnvall E and Blow JJ (t984) Intracellular mechanisms in cell death in normal and pathological tissues. In: Cell Ageing and Cell Death. Davies I and Siegg DC (eds.) Cambridge University Press, Cambridge, pp. 269294.
Address for offprints: M. A1-Rubeai,School of Chemical Engineering, University of Birmingham, P.O. Box 363, Birmingham B15 2TT, UK
