Somatic Cell Geneiics, Vol. 5, No. 6, 1979, pp. 1013-1029
Complementation of Genetic Disease: A Velocity Sedimentation Procedure for the Enrichment of Heterokaryons Lynda Karig Hohmann and Thomas B. Shows Biochemical Genetics Section, Roswell Park Memorial Institute, New York State Department o f Health, Buffalo, New York 14363 Received 17 July 1979
Abstract--Methodology is described to enrich for heterokaryons after mammalian cell fusion. A heterogeneous cell mixture can be separated on a Sta-Put apparatus into fractions of uniform size cells by sedimentation through a 1% bovine serum albumin-5% Ficoll gradient. Unfused RAG and LM/TK- cells, differing by 10% in diameter, have been sorted by size; following fusion, larger and faster sedimenting cells were shown to be hybrids. This methodology can be utilized in genetic complementation studies of human genetic diseases where selection procedures for proliferating hybrids do not exist. When fibroblasts from individuals with Tay-Sachs disease [deficient in hexosaminidase A (HEX A-)] and Sandhoff-Jatzkewitz disease (HEX A - and H E X B-) are fused, HEX A is generated, demonstrating complementation of two different mutations. After Sta-Put fractionation, the HEX A complementation product was associated with the faster sedimenting multinuclear cells and not with the mononuclear parental cells. This methodology will facilitate detection of genetic differences in fibroblasts from related inherited disorders.
INTRODUCTION Multistep pathways in human metabolism enhance the possibility that different genetic mutations can result in the same or similar disease phenotype. For example, several broadly classified inherited disorders, such as the GMI gangliosidoses, GM2 gangliosidoses, and mucolipidoses, have been subdivided into distinct diseases based on biochemical and clinical properties (1-8). Clinical and biochemical characteristics also suggest internal heterogeneity for other less well-defined syndromes. For counseling and treatment 1013 0098-0366/79/1100-1013503.00/0 9 1979 Plenum PublishingCorporation
1014
Karig Hohmann and Shows
purposes, it is important to know whether this variation is indeed genetic and if it is caused by mutations at different genes or at alleles of the same locus. In only rare instances are Mendelian genetic family studies useful in defining different subtypes within heterogeneous disorders; however, the ability to fuse tissue culture cells derived from affected individuals has made it possible to investigate these questions using classical genetic complementation analyses (2, 5, 9, 10). In several diseases, e.g,, GM2 gangliosidoses (5) and xeroderma pigmentosum (11), the observed heterogeneity has been shown to be due to different mutations since cell fusion of different deficient fibroblasts has resulted in the appearance of a previously absent gene product or function. The observed number of complementation groups also represents the minimum number of steps in the involved pathway, information which is useful for characterizing the biochemical defect (12). Cell fusion with current techniques (13, 14) produces a cell population consisting of heterokaryons containing one or multiple copies of the genomes of both parents as well as homokaryons containing the genome of one parent and the unfused parental cells. Since the proportion of cells that are heterokaryons is small, it is essential to isolate these heterokaryons or the resulting proliferating cell hybrids in order to concentrate the complemented gene product sufficiently for biochemical characterization. For human fibroblast cell lines, selective growth techniques developed to isolate cell hybrids from heterokaryons of permanent cell lines are generally not applicable. Few naturally occurring selectable traits exist in fibroblasts, and their limited lifespan make mutagen-induced traits difficult to obtain. Since a constant property of fused cells is their increased size, significant enrichments of human-human fibroblast heterokaryons can be achieved by velocity sedimentation at unit gravity using the Sta-Put apparatus (15). The apparatus has been previously useful for the fractionation of spleen cells (15, 16) and spermatogonia (17) where cell size has varied with different functions and different stages of differentiation. Because cell size varies during the cell cycle, the Sta-Put can fractionate cells at the several stages (16, 18). Additionally, in human fibroblast populations, aging is associated with an increase in cell size (19). Because of these variables affecting cell size and the complex mixture of cells in the fusion mixture, a pure heterokaryon population cannot be isolated; however, significant enrichments for the large heterokaryons can be obtained. All heterokaryons, whether they are dikaryons or polyheterokaryons, continue their metabolic functions (20) and can be used to demonstrate genetic complementation. Sufficient quantities of heterokaryons can be isolated to determine if the recovered activity possesses normal or abnormal biochemical characteristics, which is an added advantage over procedures which determine activity in single heterokaryons (2, 21). This paper describes a protocol for the enrichment of heterokaryons based on cell
Velocity Sedimentation of Mammalian Heterokaryons
1015
size that is sensitive and of proven use in genetic complementation studies. An abstract of this work has been previously published (22). MATERIALS AND M E T H O D S
Parental Cells. The mouse cells were RAG (23), deficient in hypoxanthine phosphoribosyltransferase ( H P R T - ) , and L M / T K - (24), deficient in thymidine kinase ( T K ) . These were grown in Dulbecco's minimum essential medium ( D M E M ) (Grand Island Biological Company, Grand Island, New York) with 10% fetal calf serum and antibiotics. Cell hybrids of RAG and L M / T K were selected after fusion in D M E M supplemented with H A T (hypoxanthine, aminopterin, thymidine) medium which kills the parental cells (23). The human cells used were Tay-Sachs 408 (5) and SandhoffJatzkewitz 421 skin fibroblasts (5). Tay-Sachs 408 is deficient in lysosomal hexosaminidase A (HEX A), and Sandhoff-Jatzkewitz 421 is deficient in lysosomal hexosaminidase A and B (HEX A and HEX B). These fibroblasts were grown in Eagle's BME-basal medium (Gibco) with 10% fetal calf serum and antibiotics. Cell Fusions. Equal numbers of cells were cocultivated and, when approximately 80% confluent, were treated with polyethylene glycol (PEG), 5 0 : 5 0 with DMEM containing 3x bicarbonate, and rinsed thoroughly according to the cell fusion method of Davidson et al. (14). Control flasks untreated with PEG were also prepared. Mouse cells were treated with oneand two-minute exposures to PEG 1000 or 6000. The cells were allowed to recover for three days and then fractionated on the Sta-Put apparatus. The human cells were fused with 30- to 45-sec exposures to PEG 1000 and allowed to recover. For heterokaryon counts, cells were fractionated the next day on the Sta-Put apparatus. For complementation studies, at least five days after fusion were allowed for enzyme expression. Cell Separation Using the Sta-Put. The Sta-Put apparatus (Johns Scientific, Toronto) is illustrated in Fig. 1. Theoretical considerations and general protocol for operation are discussed by Miller and Phillips (15). The sedimentation chamber A was siliconized with Siliclad (Clay Adams) to decrease spurious cell attachment, and the whole apparatus was autoclaved when necessary. Cells are loaded into the bottom of the sedimentation chamber (A), and a gradient is formed underneath the cells. The sedimentation velocity of a cell through the gradient is directly related to the square of the cell's radius, gravity, and the density difference between the cell and gradient, and inversely related to the viscosity of the gradient. Under constant conditions, the larger cells will sediment to the lower sections of the sedimentation chamber more rapidly than smaller cells. Mixtures of human fibroblasts after PEG treatment were very hetero-
1016
Karig Hohmann and Shows
e
I
1
UU Fig. 1. Schematic diagram of STA-PUT apparatus. The labeled parts are: A, sedimentation chamber; B, cell or buffer loading chamber; C1 and C2, two-chambered gradient maker; D, Sta-Put baffle; E, three-way micrometering stopcock; F, magnetic stirring motor. The protocol is described in Materials and Methods.
geneous, with cells ranging from mononuclear parental cells to large multinucleated heterokaryons. This size range could not be satisfactorily fractionated with published protocols (15, 16). In these protocols the gradient, which stabilizes against convection currents especially during loading and unloading, was either bovine serum albumin (0.3-2%) or fetal calf serum (0-30%) in phosphate-buffered saline. The gradient viscosity was approximately that of water (1.56 7/) (15) and its density was 1.006, sufficiently less than that of most mammalian cells (1.06) to preclude separation by density. The gradient was formed as either a "shear step gradient" or a "buffered step gradient." The latter requires a slower initial loading of the cells and gradient, and has a higher "streaming limit." The shear step gradient allowed an increased loading rate but lowered the "streaming limit." The streaming limit (15, 25), which is inversely proportional to cell size, is a cell concentration which, if exceeded, results in some cells sedimenting at rates faster than predicted by size, thus decreasing the accuracy of cell separations. The streaming limit for our unfused parental cells was approximately 5 x 105 cells/ml or 20 x 106 cells/40 ml loaded with the buffered step gradient. Using these protocols, the slow filling rate was less than the sedimentation rate of the larger and denser multinuclear fused cells which, therefore, never rose above the bottom of the sedimentation chamber. To counter these very fast sedimentation rates, the density and viscosity of the gradient were increased by using a gradient of 1% BSA-5% Ficoll. The density of 5% Ficoll is 1.015 g/cm 3, with a viscosity of
Velocity Sedimentationof Mammalian Heterokaryons
1017
4 ~ at 4~ (26). Additionally, the initial filling rate was increased, which necessitated the use of the shear step gradient to maintain gradient stability. Even below the theoretical streaming limit, it appeared that the significantly larger multinuclear cells drag smaller parental cells down into the lower section of the Sta-Put. The number of cells loaded had to be less than the streaming limit and at a density which minimized "drag." Optimal cell numbers were determined to be 5 x 10 6 cells/40 ml, initial loading volume. The protocol for enrichment of fused cells is as follows: All solutions are made in Puck's saline GM, pH 7.2 (saline G minus CaC12 and MgSO4), to decrease cell aggregation (27). The procedure is carried out in a cold room (4~ The first chamber of the gradient maker (C 1) is filled with 750 ml 1% BSA in saline GM, and the second (C 2) with 5% Ficoll in saline GM. The top layer of 50 ml of saline GM is loaded into the sedimentation chamber (A) through the loading chamber (B). The baffle (D) in the sedimentation chamber must be checked for correct placement. Optimally, for these studies, 3-5 X 1 0 6 monodisperse cells are loaded next; however, even with 20 x 10 6 cells, definite enrichments are possible. Cells are loaded in 40 ml of 0.3% BSA in saline GM into the sedimentation chamber (A) through the loading chamber (B) at the rate of 25 ml/min. As the loading chamber drains, the sides are rinsed with more (80%) and cell viability does not change during separation. Particularly important, adequate numbers of heterokaryons are quickly obtained to biochemically compare the recovered complemented enzyme with the control enzyme for kinetic, electrophoretic, and immunological properties. Several types of apparatus have been described for sedimentation at unit gravity to overcome the problems of cell streaming and to ensure a sharp starting band of cells (15, 36, 37); however, the general principle of all the various methods is the same. Individual cells at the top of a lower density column will sediment at a rate based on size, with larger ceils sedimenting faster. The protocol presented here for the commercially available Sta-Put apparatus results in significant enrichments (at least five-fold) of heterokaryon fractions from fused populations of cells. The Sta-Put enrichment protocol is applicable to most cell hybridizations, particularly when a suitable cell hybrid selection system is lacking and when heterokaryons will yield a complementation product. If cell hybrids are desired, the multinucleated cell fractions can be plated ih culture dishes for isolating proliferating clones; however, each clone needs further testing to identify those clones derived from heterokaryons rather than homokaryons. Using the same apparatus, Chang and Davidson (38) have presented another protocol designed to enrich for fused human fibroblasts for the purpose of isolating proliferating hybrids. If the goal is to determine genetic complementation as well as to characterize to some extent the complemented gene product, then a sufficient number of heterokaryons rather than proliferating cell hybrids is adequate and less time consuming to acquire. Two major limitations were apparent for separating human fibroblasts by the Sta-Put procedure. The first is the limit on the number of cells that can be loaded because of cell streaming and cellular drag. In our system, the range of 3-5 • 1 0 6 cells was optimum for separation, as discussed in Materials and Methods; however, in several studies using up to 20 x 106 cells, good separations and significant enrichments for heterokaryons could still be achieved. If more cells are desired, a larger sedimentation chamber is available. The second limitation is the lack of uniformity in cell size in a population of human fibroblasts (compare Figs. 4 and 5). Cell size varies during the cell cycle, and as human fibroblasts age, they become larger. The unsyncbror~ized RAG and L M / T K - cells were considerably more uniform than the human disease fibroblasts. These human disease fibroblasts age more rapidly than normal fibroblasts, presumably as a result of their disease state. The disease associated with the cells, in this case a lysosomal storage disease, may also play a role in increasing the size of the cells. To reduce size heterogeneity associated with cell cycle changes, cells were grown to
1028
Karig Hohmannand Shows
confluency; however, synchronization of cells to reduce this heterogeneity can also be employed. Variation in size due to age and disease state is best avoided by using young cultures of cells, which is often difficult with human genetic disease fibroblasts. Fractionation of the parental cells on the Sta-Put prior to fusion is also a possibility to obtain a more uniform cell population. Despite these limitations, significant enrichments for heterokaryons were achieved at a level permitting biochemical analysis of the complemented product. While none of the systems designed to isolate heterokaryons based on physical properties is ideal, the protocol described here is least dependent on high technology and is least disruptive to the cells while still providing a high yield of viable cells per experiment. Two cell lines, RAG and L M / T K - , differing by 10% in size, can be separated on the Sta-Put with the same sensitivity as possible on a fluorescent activated cell sorter using the same parameter, cell size, to sort. The heterokaryons isolated from fusion of RAG and L M / T K - by Sta-Put separation have proven to be viable, and even a suitable source for RAG-LM/TK- cell hybrids, as demonstrated by growth on HAT medium. Motivation for the methodology described here was to develop a simple protocol to enrich human-human heterokaryons for identifying and biochemically characterizing complementation products and for proving the association of the product with the heterokaryon fraction. The earliest fractions from the Sta-Put and the multinucleated wash fractions have proven to be an adequate source of heterokaryons for complementation testing. With the known complementing system of fibroblasts from individuals with Tay-Sachs and Sandhoff-Jatzkewitz diseases, the HEX A band arising from complementation is clearly associated with the heterokaryon fractions, and not with the parental fraction. The Sta-Put protocol presented here will be of continuting use for the enrichment of heterokaryons from other human genetic diseases for which selection systems are otherwise absent and for which observed biochemical and clinical data suggest heterogeneity. ACKNOWLEDGEMENTS We gratefully acknowledge the assistance of R. Eddy, L. Haley, and C. Young. We especially appreciate the expert assistance of I. Mink in the study using the Coulter Channelyzer, and G. Hausner in the studies performed on the FACS II cell sorter. This work was supported by grants HD 05196, GM 07093, and MOD 1-485.
LITERATURE CITED 1. O'Brien,J. S. (1978). In The Metabolic Basis oflnherited Disease, (eds.) Stanbury, J. B., Wyngaarden, J. B., and Fredrickson, D. S. (Mc Graw-Hill,New York), pp. 841-865.
Velocity Sedimentation of Mammalian Heterokaryons
1029
2. Galjaard, H., Hoogeven, A., de Wit-Verbeek, H. A., Reuser, A. J. J., Keijzer, W., Westerveld, A., and Bootsma, D. (1974).Exp. Cell Res. 87:444-448. 3. Okada, S., and O'Brien, J. S. (1968). Science 165:698-700. 4. Sandhoff, K., Andreae, U., and Jatzkewitz, H. (1968). Life Sci. 7:278-285. 5. Rattazzi, M. C., Brown, J. A., Davidson, R. G., and Shows, T. B. (1976). Am. J. Human Genet. 28:143-154. 6. McKusiek, V. A. (1972). Heritable Disorders o f Connective Tissue, 4th ed. (Mosby, St. Louis), pp. 521-686. 7. Leroy, J. G., Spranger, J. W., Feingold, M., Opitz, J. M., and Crocker, A. C. (1971). J. Pediatr. 79:360-365. 8. Kelly, T. E., Thomas, G. H., Taylor, H. A., McKusick, V. A., Sly, W. S., Glaser, J. H., Robinow, M., Luzzatti, L., Espiritu, C., Feingold, M., Bull, M. J., Asenhurt, E. M., and Ives, E. J. (1975). Johns Hopkins Med. J. 137:156-175. 9. Fineham, J. R. S. (1966). Genetic Complementation (W. A. Benjamin, Inc., New York). 10. Ephrussi, B. (1972). Hybridization o f Somatic Cells (Princeton University Press, Princeton, New Jersey). 11. Kraemer, K. H., Coon, H. G., Petinga, R. A., Barrett, S. F., Rake, A. E., and Robbins, J. H. (1975). Proc. Natl. Acad. Sci. U.S.A. 72:59-63. 12. Kao, F.-T., Chasin, L., and Puck, T. T. (1969). Proc. Natl. Acad. Sci. U.S.A. 64:12841291. 13. Okada, Y. (1962). Exp. Cell Res. 26:98-107, 14. Davidson, R. L., O'Malley, K. A., and Wheeler, T. B. (1976). Somat. Cell Genet. 2:271-280. 15. Miller, R. G., and Phillips, R. A. (1969). J. Cell. Physiol. 73:191-202. 16. Miller, R. G. (1973). In New Techniques in Biophysics and Cell Biology Vol. 1, (eds.) Pain, R. H., and Smith, B. J. (John Wiley and Sons, New York), pp. 87-113. 17. Meistrich, M. L. (1977). Methods CellBiol. 15:15-54. 18. MacDonald, H. R., and Miller, R. G. (1970). Biophysics J. 10:834-842. 19. Simons, J. W. (1967). Exp. Cell Res. 45:336-350. 20. Harris, H., and Watkins, J. F. (1965). Nature 205:640-646. 21. Gravel, R. A., Mahoney, M. J., Ruddle, F. H., and Roseberg, L. E. (1975). Proc. Natl. Acad. Sci. U.S.A. 72:3181-3185. 22. Karig Hohmann, L., and Shows, T. B. (1978). J. Cell Biol. 79:387a. 23. Klebe, R. J., Chen, T. R., and Ruddle, F. H. (1970). J. Cell Biol. 45:74-82. 24. Kit, S., Dubbs, D. R., Pietzarski, L. J., and Hsu, T. C. (1963). Exp. Cell Res. 31:297-312. 25. Peterson, E. A., and Evans, W. H. (1967). Nature 214:824-825. 26. Pretlow, T. G., Boone, C. W., Shrager, R. I., and Weiss, G. H. (1969). Anal. Biochem. 29:230-237. 27. Crissman, H. A., Mullaney, P. F., and Steinkamp, J. A. (1975). Methods Cell Biol. 9:179-246. 28. Crissman, H. A., and Tobey, R. A. (1974). Science 184:1297-1298. 29. Champion, M. J., and Shows, T. B. (1977). Proc. Natl. Acad. Sci. U.S.A. 74:2968-2972. 30. Shows, T. B., Ruddle, F. H., and Roderick, T. H. (1969). Biochem. Genet. 3:25-35. 31. Carter, W. D., Parr, C. W. (1967) Nature 216:511. 32. Roderick, T. H., Ruddle, F. H., Chapman, V. M., and Shows, T. B. (1971). Biochem. Genet. 5:457~J,66. 33. Shows, T. B. (1978). In Isozymes: Current Topics in Biological and Medical Research, Vol. 2, (eds.) Rattazzi, M., Scandalios, J., and Whitt, G., (Alan R. Liss, New York), pp. 107-158. 34. Sanderson, R. J., and Bird, K. E. (1977). Methods Cell Biol. 15:1-14. 35. Hoehn, H., Bryant, E. M., Johnston, P., Norwood, T. H., and Martin, G. M. (1975). Nature 258:608-610. 36. Bont, W. S., and Hilgers, J. H. M. (1977). Prep. Biochem. 7:45-60. 37. Schor, S. L., Johnson, R. T., and Mulliger, A. M. (1975). J. Cell Sci. 19:281-303. 38. Chang, P. L., Joubert, G. I., and Davidson, R. G. (1979). Nature 278:168-170.
Complementation of Genetic Disease: A Velocity Sedimentation Procedure for the Enrichment of Heterokaryons Lynda Karig Hohmann and Thomas B. Shows Biochemical Genetics Section, Roswell Park Memorial Institute, New York State Department o f Health, Buffalo, New York 14363 Received 17 July 1979
Abstract--Methodology is described to enrich for heterokaryons after mammalian cell fusion. A heterogeneous cell mixture can be separated on a Sta-Put apparatus into fractions of uniform size cells by sedimentation through a 1% bovine serum albumin-5% Ficoll gradient. Unfused RAG and LM/TK- cells, differing by 10% in diameter, have been sorted by size; following fusion, larger and faster sedimenting cells were shown to be hybrids. This methodology can be utilized in genetic complementation studies of human genetic diseases where selection procedures for proliferating hybrids do not exist. When fibroblasts from individuals with Tay-Sachs disease [deficient in hexosaminidase A (HEX A-)] and Sandhoff-Jatzkewitz disease (HEX A - and H E X B-) are fused, HEX A is generated, demonstrating complementation of two different mutations. After Sta-Put fractionation, the HEX A complementation product was associated with the faster sedimenting multinuclear cells and not with the mononuclear parental cells. This methodology will facilitate detection of genetic differences in fibroblasts from related inherited disorders.
INTRODUCTION Multistep pathways in human metabolism enhance the possibility that different genetic mutations can result in the same or similar disease phenotype. For example, several broadly classified inherited disorders, such as the GMI gangliosidoses, GM2 gangliosidoses, and mucolipidoses, have been subdivided into distinct diseases based on biochemical and clinical properties (1-8). Clinical and biochemical characteristics also suggest internal heterogeneity for other less well-defined syndromes. For counseling and treatment 1013 0098-0366/79/1100-1013503.00/0 9 1979 Plenum PublishingCorporation
1014
Karig Hohmann and Shows
purposes, it is important to know whether this variation is indeed genetic and if it is caused by mutations at different genes or at alleles of the same locus. In only rare instances are Mendelian genetic family studies useful in defining different subtypes within heterogeneous disorders; however, the ability to fuse tissue culture cells derived from affected individuals has made it possible to investigate these questions using classical genetic complementation analyses (2, 5, 9, 10). In several diseases, e.g,, GM2 gangliosidoses (5) and xeroderma pigmentosum (11), the observed heterogeneity has been shown to be due to different mutations since cell fusion of different deficient fibroblasts has resulted in the appearance of a previously absent gene product or function. The observed number of complementation groups also represents the minimum number of steps in the involved pathway, information which is useful for characterizing the biochemical defect (12). Cell fusion with current techniques (13, 14) produces a cell population consisting of heterokaryons containing one or multiple copies of the genomes of both parents as well as homokaryons containing the genome of one parent and the unfused parental cells. Since the proportion of cells that are heterokaryons is small, it is essential to isolate these heterokaryons or the resulting proliferating cell hybrids in order to concentrate the complemented gene product sufficiently for biochemical characterization. For human fibroblast cell lines, selective growth techniques developed to isolate cell hybrids from heterokaryons of permanent cell lines are generally not applicable. Few naturally occurring selectable traits exist in fibroblasts, and their limited lifespan make mutagen-induced traits difficult to obtain. Since a constant property of fused cells is their increased size, significant enrichments of human-human fibroblast heterokaryons can be achieved by velocity sedimentation at unit gravity using the Sta-Put apparatus (15). The apparatus has been previously useful for the fractionation of spleen cells (15, 16) and spermatogonia (17) where cell size has varied with different functions and different stages of differentiation. Because cell size varies during the cell cycle, the Sta-Put can fractionate cells at the several stages (16, 18). Additionally, in human fibroblast populations, aging is associated with an increase in cell size (19). Because of these variables affecting cell size and the complex mixture of cells in the fusion mixture, a pure heterokaryon population cannot be isolated; however, significant enrichments for the large heterokaryons can be obtained. All heterokaryons, whether they are dikaryons or polyheterokaryons, continue their metabolic functions (20) and can be used to demonstrate genetic complementation. Sufficient quantities of heterokaryons can be isolated to determine if the recovered activity possesses normal or abnormal biochemical characteristics, which is an added advantage over procedures which determine activity in single heterokaryons (2, 21). This paper describes a protocol for the enrichment of heterokaryons based on cell
Velocity Sedimentation of Mammalian Heterokaryons
1015
size that is sensitive and of proven use in genetic complementation studies. An abstract of this work has been previously published (22). MATERIALS AND M E T H O D S
Parental Cells. The mouse cells were RAG (23), deficient in hypoxanthine phosphoribosyltransferase ( H P R T - ) , and L M / T K - (24), deficient in thymidine kinase ( T K ) . These were grown in Dulbecco's minimum essential medium ( D M E M ) (Grand Island Biological Company, Grand Island, New York) with 10% fetal calf serum and antibiotics. Cell hybrids of RAG and L M / T K were selected after fusion in D M E M supplemented with H A T (hypoxanthine, aminopterin, thymidine) medium which kills the parental cells (23). The human cells used were Tay-Sachs 408 (5) and SandhoffJatzkewitz 421 skin fibroblasts (5). Tay-Sachs 408 is deficient in lysosomal hexosaminidase A (HEX A), and Sandhoff-Jatzkewitz 421 is deficient in lysosomal hexosaminidase A and B (HEX A and HEX B). These fibroblasts were grown in Eagle's BME-basal medium (Gibco) with 10% fetal calf serum and antibiotics. Cell Fusions. Equal numbers of cells were cocultivated and, when approximately 80% confluent, were treated with polyethylene glycol (PEG), 5 0 : 5 0 with DMEM containing 3x bicarbonate, and rinsed thoroughly according to the cell fusion method of Davidson et al. (14). Control flasks untreated with PEG were also prepared. Mouse cells were treated with oneand two-minute exposures to PEG 1000 or 6000. The cells were allowed to recover for three days and then fractionated on the Sta-Put apparatus. The human cells were fused with 30- to 45-sec exposures to PEG 1000 and allowed to recover. For heterokaryon counts, cells were fractionated the next day on the Sta-Put apparatus. For complementation studies, at least five days after fusion were allowed for enzyme expression. Cell Separation Using the Sta-Put. The Sta-Put apparatus (Johns Scientific, Toronto) is illustrated in Fig. 1. Theoretical considerations and general protocol for operation are discussed by Miller and Phillips (15). The sedimentation chamber A was siliconized with Siliclad (Clay Adams) to decrease spurious cell attachment, and the whole apparatus was autoclaved when necessary. Cells are loaded into the bottom of the sedimentation chamber (A), and a gradient is formed underneath the cells. The sedimentation velocity of a cell through the gradient is directly related to the square of the cell's radius, gravity, and the density difference between the cell and gradient, and inversely related to the viscosity of the gradient. Under constant conditions, the larger cells will sediment to the lower sections of the sedimentation chamber more rapidly than smaller cells. Mixtures of human fibroblasts after PEG treatment were very hetero-
1016
Karig Hohmann and Shows
e
I
1
UU Fig. 1. Schematic diagram of STA-PUT apparatus. The labeled parts are: A, sedimentation chamber; B, cell or buffer loading chamber; C1 and C2, two-chambered gradient maker; D, Sta-Put baffle; E, three-way micrometering stopcock; F, magnetic stirring motor. The protocol is described in Materials and Methods.
geneous, with cells ranging from mononuclear parental cells to large multinucleated heterokaryons. This size range could not be satisfactorily fractionated with published protocols (15, 16). In these protocols the gradient, which stabilizes against convection currents especially during loading and unloading, was either bovine serum albumin (0.3-2%) or fetal calf serum (0-30%) in phosphate-buffered saline. The gradient viscosity was approximately that of water (1.56 7/) (15) and its density was 1.006, sufficiently less than that of most mammalian cells (1.06) to preclude separation by density. The gradient was formed as either a "shear step gradient" or a "buffered step gradient." The latter requires a slower initial loading of the cells and gradient, and has a higher "streaming limit." The shear step gradient allowed an increased loading rate but lowered the "streaming limit." The streaming limit (15, 25), which is inversely proportional to cell size, is a cell concentration which, if exceeded, results in some cells sedimenting at rates faster than predicted by size, thus decreasing the accuracy of cell separations. The streaming limit for our unfused parental cells was approximately 5 x 105 cells/ml or 20 x 106 cells/40 ml loaded with the buffered step gradient. Using these protocols, the slow filling rate was less than the sedimentation rate of the larger and denser multinuclear fused cells which, therefore, never rose above the bottom of the sedimentation chamber. To counter these very fast sedimentation rates, the density and viscosity of the gradient were increased by using a gradient of 1% BSA-5% Ficoll. The density of 5% Ficoll is 1.015 g/cm 3, with a viscosity of
Velocity Sedimentationof Mammalian Heterokaryons
1017
4 ~ at 4~ (26). Additionally, the initial filling rate was increased, which necessitated the use of the shear step gradient to maintain gradient stability. Even below the theoretical streaming limit, it appeared that the significantly larger multinuclear cells drag smaller parental cells down into the lower section of the Sta-Put. The number of cells loaded had to be less than the streaming limit and at a density which minimized "drag." Optimal cell numbers were determined to be 5 x 10 6 cells/40 ml, initial loading volume. The protocol for enrichment of fused cells is as follows: All solutions are made in Puck's saline GM, pH 7.2 (saline G minus CaC12 and MgSO4), to decrease cell aggregation (27). The procedure is carried out in a cold room (4~ The first chamber of the gradient maker (C 1) is filled with 750 ml 1% BSA in saline GM, and the second (C 2) with 5% Ficoll in saline GM. The top layer of 50 ml of saline GM is loaded into the sedimentation chamber (A) through the loading chamber (B). The baffle (D) in the sedimentation chamber must be checked for correct placement. Optimally, for these studies, 3-5 X 1 0 6 monodisperse cells are loaded next; however, even with 20 x 10 6 cells, definite enrichments are possible. Cells are loaded in 40 ml of 0.3% BSA in saline GM into the sedimentation chamber (A) through the loading chamber (B) at the rate of 25 ml/min. As the loading chamber drains, the sides are rinsed with more (80%) and cell viability does not change during separation. Particularly important, adequate numbers of heterokaryons are quickly obtained to biochemically compare the recovered complemented enzyme with the control enzyme for kinetic, electrophoretic, and immunological properties. Several types of apparatus have been described for sedimentation at unit gravity to overcome the problems of cell streaming and to ensure a sharp starting band of cells (15, 36, 37); however, the general principle of all the various methods is the same. Individual cells at the top of a lower density column will sediment at a rate based on size, with larger ceils sedimenting faster. The protocol presented here for the commercially available Sta-Put apparatus results in significant enrichments (at least five-fold) of heterokaryon fractions from fused populations of cells. The Sta-Put enrichment protocol is applicable to most cell hybridizations, particularly when a suitable cell hybrid selection system is lacking and when heterokaryons will yield a complementation product. If cell hybrids are desired, the multinucleated cell fractions can be plated ih culture dishes for isolating proliferating clones; however, each clone needs further testing to identify those clones derived from heterokaryons rather than homokaryons. Using the same apparatus, Chang and Davidson (38) have presented another protocol designed to enrich for fused human fibroblasts for the purpose of isolating proliferating hybrids. If the goal is to determine genetic complementation as well as to characterize to some extent the complemented gene product, then a sufficient number of heterokaryons rather than proliferating cell hybrids is adequate and less time consuming to acquire. Two major limitations were apparent for separating human fibroblasts by the Sta-Put procedure. The first is the limit on the number of cells that can be loaded because of cell streaming and cellular drag. In our system, the range of 3-5 • 1 0 6 cells was optimum for separation, as discussed in Materials and Methods; however, in several studies using up to 20 x 106 cells, good separations and significant enrichments for heterokaryons could still be achieved. If more cells are desired, a larger sedimentation chamber is available. The second limitation is the lack of uniformity in cell size in a population of human fibroblasts (compare Figs. 4 and 5). Cell size varies during the cell cycle, and as human fibroblasts age, they become larger. The unsyncbror~ized RAG and L M / T K - cells were considerably more uniform than the human disease fibroblasts. These human disease fibroblasts age more rapidly than normal fibroblasts, presumably as a result of their disease state. The disease associated with the cells, in this case a lysosomal storage disease, may also play a role in increasing the size of the cells. To reduce size heterogeneity associated with cell cycle changes, cells were grown to
1028
Karig Hohmannand Shows
confluency; however, synchronization of cells to reduce this heterogeneity can also be employed. Variation in size due to age and disease state is best avoided by using young cultures of cells, which is often difficult with human genetic disease fibroblasts. Fractionation of the parental cells on the Sta-Put prior to fusion is also a possibility to obtain a more uniform cell population. Despite these limitations, significant enrichments for heterokaryons were achieved at a level permitting biochemical analysis of the complemented product. While none of the systems designed to isolate heterokaryons based on physical properties is ideal, the protocol described here is least dependent on high technology and is least disruptive to the cells while still providing a high yield of viable cells per experiment. Two cell lines, RAG and L M / T K - , differing by 10% in size, can be separated on the Sta-Put with the same sensitivity as possible on a fluorescent activated cell sorter using the same parameter, cell size, to sort. The heterokaryons isolated from fusion of RAG and L M / T K - by Sta-Put separation have proven to be viable, and even a suitable source for RAG-LM/TK- cell hybrids, as demonstrated by growth on HAT medium. Motivation for the methodology described here was to develop a simple protocol to enrich human-human heterokaryons for identifying and biochemically characterizing complementation products and for proving the association of the product with the heterokaryon fraction. The earliest fractions from the Sta-Put and the multinucleated wash fractions have proven to be an adequate source of heterokaryons for complementation testing. With the known complementing system of fibroblasts from individuals with Tay-Sachs and Sandhoff-Jatzkewitz diseases, the HEX A band arising from complementation is clearly associated with the heterokaryon fractions, and not with the parental fraction. The Sta-Put protocol presented here will be of continuting use for the enrichment of heterokaryons from other human genetic diseases for which selection systems are otherwise absent and for which observed biochemical and clinical data suggest heterogeneity. ACKNOWLEDGEMENTS We gratefully acknowledge the assistance of R. Eddy, L. Haley, and C. Young. We especially appreciate the expert assistance of I. Mink in the study using the Coulter Channelyzer, and G. Hausner in the studies performed on the FACS II cell sorter. This work was supported by grants HD 05196, GM 07093, and MOD 1-485.
LITERATURE CITED 1. O'Brien,J. S. (1978). In The Metabolic Basis oflnherited Disease, (eds.) Stanbury, J. B., Wyngaarden, J. B., and Fredrickson, D. S. (Mc Graw-Hill,New York), pp. 841-865.
Velocity Sedimentation of Mammalian Heterokaryons
1029
2. Galjaard, H., Hoogeven, A., de Wit-Verbeek, H. A., Reuser, A. J. J., Keijzer, W., Westerveld, A., and Bootsma, D. (1974).Exp. Cell Res. 87:444-448. 3. Okada, S., and O'Brien, J. S. (1968). Science 165:698-700. 4. Sandhoff, K., Andreae, U., and Jatzkewitz, H. (1968). Life Sci. 7:278-285. 5. Rattazzi, M. C., Brown, J. A., Davidson, R. G., and Shows, T. B. (1976). Am. J. Human Genet. 28:143-154. 6. McKusiek, V. A. (1972). Heritable Disorders o f Connective Tissue, 4th ed. (Mosby, St. Louis), pp. 521-686. 7. Leroy, J. G., Spranger, J. W., Feingold, M., Opitz, J. M., and Crocker, A. C. (1971). J. Pediatr. 79:360-365. 8. Kelly, T. E., Thomas, G. H., Taylor, H. A., McKusick, V. A., Sly, W. S., Glaser, J. H., Robinow, M., Luzzatti, L., Espiritu, C., Feingold, M., Bull, M. J., Asenhurt, E. M., and Ives, E. J. (1975). Johns Hopkins Med. J. 137:156-175. 9. Fineham, J. R. S. (1966). Genetic Complementation (W. A. Benjamin, Inc., New York). 10. Ephrussi, B. (1972). Hybridization o f Somatic Cells (Princeton University Press, Princeton, New Jersey). 11. Kraemer, K. H., Coon, H. G., Petinga, R. A., Barrett, S. F., Rake, A. E., and Robbins, J. H. (1975). Proc. Natl. Acad. Sci. U.S.A. 72:59-63. 12. Kao, F.-T., Chasin, L., and Puck, T. T. (1969). Proc. Natl. Acad. Sci. U.S.A. 64:12841291. 13. Okada, Y. (1962). Exp. Cell Res. 26:98-107, 14. Davidson, R. L., O'Malley, K. A., and Wheeler, T. B. (1976). Somat. Cell Genet. 2:271-280. 15. Miller, R. G., and Phillips, R. A. (1969). J. Cell. Physiol. 73:191-202. 16. Miller, R. G. (1973). In New Techniques in Biophysics and Cell Biology Vol. 1, (eds.) Pain, R. H., and Smith, B. J. (John Wiley and Sons, New York), pp. 87-113. 17. Meistrich, M. L. (1977). Methods CellBiol. 15:15-54. 18. MacDonald, H. R., and Miller, R. G. (1970). Biophysics J. 10:834-842. 19. Simons, J. W. (1967). Exp. Cell Res. 45:336-350. 20. Harris, H., and Watkins, J. F. (1965). Nature 205:640-646. 21. Gravel, R. A., Mahoney, M. J., Ruddle, F. H., and Roseberg, L. E. (1975). Proc. Natl. Acad. Sci. U.S.A. 72:3181-3185. 22. Karig Hohmann, L., and Shows, T. B. (1978). J. Cell Biol. 79:387a. 23. Klebe, R. J., Chen, T. R., and Ruddle, F. H. (1970). J. Cell Biol. 45:74-82. 24. Kit, S., Dubbs, D. R., Pietzarski, L. J., and Hsu, T. C. (1963). Exp. Cell Res. 31:297-312. 25. Peterson, E. A., and Evans, W. H. (1967). Nature 214:824-825. 26. Pretlow, T. G., Boone, C. W., Shrager, R. I., and Weiss, G. H. (1969). Anal. Biochem. 29:230-237. 27. Crissman, H. A., Mullaney, P. F., and Steinkamp, J. A. (1975). Methods Cell Biol. 9:179-246. 28. Crissman, H. A., and Tobey, R. A. (1974). Science 184:1297-1298. 29. Champion, M. J., and Shows, T. B. (1977). Proc. Natl. Acad. Sci. U.S.A. 74:2968-2972. 30. Shows, T. B., Ruddle, F. H., and Roderick, T. H. (1969). Biochem. Genet. 3:25-35. 31. Carter, W. D., Parr, C. W. (1967) Nature 216:511. 32. Roderick, T. H., Ruddle, F. H., Chapman, V. M., and Shows, T. B. (1971). Biochem. Genet. 5:457~J,66. 33. Shows, T. B. (1978). In Isozymes: Current Topics in Biological and Medical Research, Vol. 2, (eds.) Rattazzi, M., Scandalios, J., and Whitt, G., (Alan R. Liss, New York), pp. 107-158. 34. Sanderson, R. J., and Bird, K. E. (1977). Methods Cell Biol. 15:1-14. 35. Hoehn, H., Bryant, E. M., Johnston, P., Norwood, T. H., and Martin, G. M. (1975). Nature 258:608-610. 36. Bont, W. S., and Hilgers, J. H. M. (1977). Prep. Biochem. 7:45-60. 37. Schor, S. L., Johnson, R. T., and Mulliger, A. M. (1975). J. Cell Sci. 19:281-303. 38. Chang, P. L., Joubert, G. I., and Davidson, R. G. (1979). Nature 278:168-170.
