Reversibly permeable hepatoma cells in culture

Biochimica et Biophysica Acta, 721 (1982) 253-261 Elsevier Biomedical Press

253

BBA 11084

REVERSIBLY PERMEABLE HEPATOMA CELLS IN CULTURE MALGORZATA BALINSKA, WILLIAM A. SAMSONOFF and JOHN GALIVAN

Center for Laboratories and Research, New York State Department of Health, Albanv, N Y 12201 (U.S.A.) (Received April 26th, 1982)

Key words: Permeability," Enzyme activity; Folate depletion," (Hepatoma cell)

A brief treatment of H35 hepatoma cells with lysolecithin resulted in a cell population which is permeable to low-molecular weight charged molecules that cannot normally cross the plasma membrane. These include deoxynucleotide and nucleotide triphosphates, folyl and methotrexate polyglutamates, and trypan blue. As a result dTTP can be incorporated into the DNA of the permeable cells, providing the required nucleotides and deoxynucleotides are added to the medium. This result, combined with only a slight observed loss (20-25%) in total cell protein, lactate dehydrogenase (EC 1.1.1.27) activity and tyrosine aminotransferase (EC 2.6.1.5) activity, demonstrated that permeation of the cells does not extensively disrupt membrane integrity. Further support for this view comes from the fact that the permeable cells could seal when placed in enriched medium. The process of sealing was inhibited by cycloheximide and tunicamycin. The sealed cells, whose surfaces appeared identical to those of untreated cells by scanning electron microscopy, were fully capable of cell division when exposed to serum. Values for several other parameters, including dexamethasone-dependent tyrosine aminotransferase induction, thymidine incorporation into DNA, leucine incorporation into protein and folate coenzyme transport, supported the conclusion that sealed cells and untreated H35 cells have identical properties. Based on the characteristics of the permeable and sealed 1-135 cells, a discussion of the experimental potential of these preparations for studying macromolecular synthesis, investigating enzymes in situ and depleting cells of folate coenzymes is presented.

Introduction

The study of hepatic cell function in vitro has been approached by the use of two types of cultured cell systems: primary isolates of hepati c parenchymal cells, either as fresh suspensions or as monolayer cultures, and stable transformed cell lines derived from hepatomas. Primary cultures of hepatocytes have the advantage of retaining many, but not all of the hepatic specific functions present in the intact animal [1]. The major disadvantage of these cells has been the difficulty in developing Abbreviations: ( +)5-CH3 H4 PteGlu, (+)5-methyl 5,6,7,8-tetrahydropteroylglutamic acid; methotrexate, 4-amino-10-methylpteroylglutamic acid. 0167-4889/82/0000-0000/$02.75 © 1982 Elsevier Biomedical Press

conditions which will promote long-term viability and retention of differentiated hepatic properties. In some instances hepatoma cells have served in the study of liver-specific processes, although they do not exhibit all of the properties of the highly differentiated cells in normal liver [2]. Deployment of these cell-culture systems has contributed significantly to the understanding of the physiological, biochemical and metabolic properties of hepatic tissue. The object of numerous studies with cultured cell systems has been to investigate metabolic processes under conditions that more closely approximate physiological conditions than does the use of isolated enzymes. A major impediment to this end with cultured cells is the impermeability

254

of charged substrates for which there is no transport system. A primary example is the study of D N A synthesis from nucleoside triphosphates. To circumvent such difficulties, a number of investigators have employed a variety of techniques for cell permeation [3-15]. An early report indicated that under certain conditions the permeabilized cells could seal and form intact cells [7]. These investigators subsequently published a more detailed study demonstrating the general utility of lysolecithin as permeabilizing agent and also demonstrated that many of these cell lines could seal following permeation [10]. None of the later investigations were conducted with hepatic cell lines in culture. Because a permeable hepatic cell line would be a useful experimental tool to investigate hepatic function in vitro, we undertook such a study with H-4-II-E-C3 hepatoma cells, which are derived from the Reuber H35 hepatoma [16,17]. The results indicated that a short incubation of these cells with lysolecithin causes the cells to become permeable to low-molecular weight charged molecules, which cannot normally cross the cell membrane. After permeation the cells can seal in enriched medium. This report includes a description of these processes and a comparison of the properties of untreated and sealed H35 cells. Materials and Methods

Swims medium S-77, fetal calf serum and horse serum were obtained from Grand Island Biological Co., Grand Island, NY. [3',5',7,9-3H]Folic acid, [4,5-3H]leucine, [5-3H]thymidine and [methyl3H]dTTP were purchased from Amersham Corp., Arlington Heights, IL. All other compounds were obtained from Sigma Chemical Corp., St. Louis, MO. Cell culture and permeabilization. H- 11-EC3 cells (hereafter called H35 cells) were derived from the Reuber H35 hepatoma and grown in monolayer culture as described previously [16-18]. The cells were made permeable by treatment with lysolecithin as follows. The medium was removed from cells that had been in culture 96 h (late log to early stationary phase of growth), and 1.5 ml of solution A (1.0 m g / m l lysolecithin in 0.25 M sucrose containing 10 m M Tris-HC1, p H 7.4/10 m M

E D T A / 4 . 0 mM MgC12) was added for 1.5 min. Solution A was then removed, and the cells were washed three times with Swims S-77 medium. To obtain sealed cells, the permeable preparation was incubated at 37°C in 5% CO2/95% air for 16 h in 2 ml of Swims S-77 medium containing 10 m U / m l insulin, unless otherwise noted. Permeabilization and sealing of the cells were measured with trypan blue by a modification [19] of the method described by Williams et al. [20]. Cell growth studies were conducted as described by using ZBI Coulter Counter [18]. Macromolecular Synthesis. Incorporation of tritiated leucine, thymidine or d T T P was conducted by a modification of the procedure of Miller et al. [81. Leucine incorporation was measured as follows. The medium was removed from the plates and the cells were washed twice with 4 ml Hanks balanced salt solution and incubated 1 h at 37°C. [3H]Leucine (5 C i / m m o l ) was added at a final concentration of 4 ~ M , and the incubation was continued for 60 min at 37°C. To terminate the reaction, the plates were cooled and washed four times with 4-ml washes of ice-cold 0.85% NaC1 with 10 m M potassium phosphate, p H 7.4. The cells were removed with two 1-ml successive washes of 1 N N a O H . An aliquot was assayed for protein by the method of Lowry et al. [21], and counted by liquid scintillation spectrometry, with the radioactivity expressed as pmol leucine/106 cells. For [methyl-3H]thymidine 5'-phosphate incorporation into D N A the medium was removed, and the plates were washed twice with 2 ml Hanks balanced salt solution and incubated 1 h at 37°C. The Hanks solution was replaced with 2 ml incubation mixture, as described by Miller et al. [8], containing 150 m M sucrose/80 m M KC1/35 mM Hepes (pH 7.4)/5 m M potassium phosphate (pH 7.4)/5 mM MgC12/0.5 mM C a C l z / 2 0 mM phosphoenolpyruvate/1.25 m M ATP/0.1 m M CTP, G T P and U T P / 0 . 2 5 m M [3H]dTTP (0.2 C i / mmol)/0.25 mM dATP, dCTP and dGTP. The reaction was conducted at 37°C for 1 h and was terminated by cooling to 0°C and then washing three times with 4 ml ice-cold 0.85% NaC1 with 10 mM potassium phosphate (pH 7.4). The cells were pooled with two 0.5-ml aliquots of the saline solution, and 1 ml 10% ice-cold trichloroacetic acid

255

was added immediately. An aliquot was applied to a 0.45 # m nitrocellulose filter, which was then washed extensively with 5% cold trichloroacetic acid, dried and counted by liquid scintillation spectrometry. Incorporation of dTTP was expressed as pmol d T T P / 1 0 6 cells. For thymidine incorporation the cells were prepared as for [3H]dTTP incorporation and incubated with [3H]thymidine (0.5/~Ci/ml) in Hanks balanced salt solution or in the incubation mixture used for dTTP incorporation. Reaction termination and analysis of incorporation were identical to those described for [3H]dTTP. Enzyme assays. Lactate dehydrogenase (EC 1.1.1.27) activity was measured by the method of Bergmeyer and Brent [22]. The activity of tyrosine aminotransferase (EC 2.6.1.5) was measured by the procedure of Diamondstone [23]. Uptake and efflux studies. The rate of ( + ) 5 CH3H4PteGlu uptake was determined as described previously [18,24] as were the accumulation and efflux of folic acid [25]. Results Of the several techniques we tried for inducing permeability to trypan blue in H35 cells, the most useful and reproducible was the use of lysolecithin (Table I). A brief incubation (1.5 min) of monolayer cultures of H35 cells with 0.1% lysolecithin resulted in a preparation retaining 80% of the original cells, which were 90% permeable to trypan blue (Table I). This preparation is termed 'permeable' cells. Despite a number of attempted variations in the conditions of lysolecithin treatment, a small portion ( 1 0 -+ 5%) of the treated cells remained impermeable. Replacement of the lysolecithin buffer with serum-free Swims S-77 medium after premeation resulted in a gradual reversal of membrane permeability. After 16h all of the remaining cells were trypan blue-negative as were the untreated and control cultures. Recovery of attached cells from treated cultures after the 16-h incubation was 63% of the control cultures. The kinetics of sealing are shown in Fig. 1. Hanks balanced salt solution does not support this process, but enriched medium does. Inclusion of insulin in the enriched medium results in margi-

TABLE I H35 CELL PERMEABILIZATION A N D SEALING H35 cells were permeabilized with 0. I% lysolecithin for 1.5 min and sealed in Swims medium. Permeable cells were counted by the trypan blue technique 1 h after lysolecithin treatment and sealed cells 16 h after lysolecithin treatment. Untreated cells were incubated with Swims medium for 16 h; controls were incubated for 1.5 min in solution A lacking lysolecithin and then for 16 h in Swims medium. Each mean +S.D. is based upon a minimum of 20 observations. Cell status

Cells × 106/plate

Trypan blue negative

(%) Prior to lysolecithin Permeable Sealed Untreated Control

4 . 8+4.0 ± 2.4 3.8 + 3.8 +

1.2 1.2 O.5 1.0 1.0

100 10--- 5 1O0 100 100

nally higher yields and a cell preparation which more closely resembles the parent cell culture in appearance. In the absence of insulin the cytoplasm of the sealed cells is more granular and exhibits vacuolization. Inclusion of serum during

3O

t.o

g_ ~", 20

o

_o x

!,o I

0

I

8 16 h Fig. I. Sealing o f lysolecithin-treated H35 cells. H35 cells were permeabilized and then placed in Swims S-77 medium (O C)), Swims with 10 m U / m l insulin (O O), Swims with 20% horse serum and 5% fetal bovine serum (E3 r-l), or Hanks balanced salt solution (A ~). The number of viable cells in each culture was measured during the ensuing 16 h by the trypan blue method.

256 sealing gives even higher cell yields. All three cultures d e p i c t e d in Fig. 1 were 100% t r y p a n bluenegative at the end of 16h. A l t h o u g h we used Swims S-77 m e d i u m s u p p l e m e n t e d with insulin, except as noted, serum can be included if the e x p e r i m e n t a l design is not j e o p a r d i z e d b y the presence of this c o m p l e x mixture of u n d e f i n e d factors. E v i d e n c e was sought to show the extent to which sealed cells r e s e m b l e d u n t r e a t e d cells and to d e m o n s t r a t e that p e r m e a b l e cells exhibited p r o p e r ties c o n s i s t e n t with the partial loss of m e m b r a n e integrity. By s c a n n i n g electron m i c r o s c o p y the control a n d sealed cultures were virtually identical, b u t the p e r m e a b l e cells e x h i b i t e d a vastly altered surface structure. C o n t r o l cells h a d a long microvilli on their surface (Fig. 2a). P e r m e a b l e cells lacked microvilli, e x h i b i t e d m a n y m e m b r a n e blebs a n d were c o n t r a c t e d , as e v i d e n c e d b y their smaller a p p a r e n t size and the cylindrical processes (arrows, Fig. 2b) a t t a c h i n g t h e m to the s u b s t r a t u m . Sealed cells (Fig. 2c) were replete with microvilli a n d were i n d i s t i n g u i s h a b l e from c o n t r o l cells. P l a c e m e n t of the sealed cells in m e d i u m with s e r u m caused t h e m to divide until confluent (Fig. 3). T r y p s i n i z a t i o n of p e r m e a b i l i z e d cell cultures, which had b e c o m e sealed in Swims S-77 m e d i u m , a n d d i l u t i o n into s e r u m - s u p p l e m e n t e d m e d i u m resulted in a growth curve similar to that for cultures m a i n t a i n e d in serum b u t with a m o r e e x t e n d e d lag phase. These d a t a d e m o n s t r a t e the c o m p e t e n c e of the sealed cells for n o r m a l cell division. A l t h o u g h several e x a m p l e s of the sealing of p e r m e a b l e cells have been d e s c r i b e d [10], little is k n o w n a b o u t the characteristics of this process. W e therefore investigated the effects of inhibitors to gain some insight into this process in H35 cells (Fig. 4). N e i t h e r h y d r o x y u r e a nor colchicine, which inhibit D N A synthesis [7] a n d m i c r o t u b u l e aggreg a t i o n [26], respectively, i m p a i r e d sealing. Inhibi-

Fig. 2. Scanning electron micrographs of (a) normal. (b) permeable and (c) sealed H35 hepatoma cells. All cells were fixed in situ for 45 min at 20°C with 2% ghitaraldehyde in a 0.1 M sodium cacodylate buffer at pH 7.2. After washing with the same buffer, cells were postfixed for 45 rain at 20°C with 1% osmium tetroxide prepared in the same buffer, dehydrated in a

graded ethanol series and critical-point-driedwith liquid carbon dioxide as the transition fluid. Portions of the culture dishes were mounted on aluminum stubs and coated with gold with a sputter coater for viewing in the scanning electron microscope. Arrows (b) indicate cylindrical processes attaching the cells to the substratum.

257

0

./"~

Ioo[

2o

/

04

!'° ~

2

d

I

48

96

0

h h Fig. 3. Growth of lysolecithin-permeabilizedcells. Permeable cells were allowed to seal for 16 h and then placed directly in Swims S-77 medium with 20% horse serum and 5% fetal bovine serum (@ @) or trypsinized and subcultured at 2.105 cells/60 mm culture dish in serum-supplemented Swims S-77 medium ( O O). Untreated cultures (O O). Fig. 4. Effect of inhibitors on sealing of permeabilized H35 cells. Cell permeabilization and sealing was conducted as described in Fig. 1 in Swims medium with 10 mU insulin/ml. After removal of lysolecithin inhibitors were included as follows: no addition (0 0), 1 mM hydroxyurea (A A), 1 /~M colchicine (11 II), 10 ,uM cycloheximide ( [ ] - [2]), or 0.5 / ~ g / m l tunicamycin

(o

o).

tors of protein synthesis (cycloheximide) and glycoprotein biosynthesis (tunicamycin) caused a marked inhibition of the sealing process, suggesting that these two processes are involved in the sealing of H35 cells after lysolecithin treatment. In

studies of [3H]leucine incorporation 1 mM cycloheximide caused an 80% inhibition of protein synthesis; but tunicamycin, an inhibitor of glycosylation [27], caused no inhibition of protein synthesis at 0.5 ffg/ml (M. Balinska, unpublished data). Analyses of protein and enzyme content demonstrated the similarity between control and sealed cells and suggested that permeable cells allow little translocation of proteins. Cell protein was only slightly reduced upon permeabilization (Table II), and the protein content of sealed cells was nearly identical to that of the control. Similarities were also observed in lactate dehydrogenase activity and the basal activity of tyrosine aminotransferase. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of control, permeable and sealed cells demonstrated no differences among their cellular protein profiles. Tyrosine aminotransferase induction by glucocorticoids, a characteristic of both normal and transformed hepatic cells [28], was not impaired by lysolecithin treatment, and the induced activities were somewhat higher in sealed cells. Permeable cells treated with dexamethasone for 20 h showed elevations in tyrosine aminotransferase similar to those in untreated cultures. However, the cells had become sealed by that time, and as a result the data are not included. To further demonstrate the properties of permeable H35 cells, DNA synthesis was examined.

TABLE II PROTEIN C O N T E N T A N D ENZYME ACTIVITIES IN H35 CELLS Protein content and lactate dehydrogenase activity were measured in confluent cells after 16 h in Swims medium (untreated), 1 h after lysolecithin treatment of the same cells (permeable) and 20 h after lysolecithin treatment of the cells (sealed). When dexamethasone was included, a 20-h incubation was conducted in Swims medium prior to tyrosine aminotransferase assay. For sealed cells the dexamethasone was added 16 h after lysolecithin treatment. Cell status

Untreated Plus dexamethasone (1 t.tM) Permeable Sealed Plus dexamethasone (1 ~tM)

Protein content ~ g / l O 5 cells

Activity (/~ m o l / m i n / m g ) Lactate dehydrogenase

Tyrosine aminotransferase

53

2.84

39 48

2.2 2.62

0.1 0.5 0.08 0.11 1.6

258 0.6

Z 0

U

z

0.4

0.2

u.I Z

>-

~

0

30

60

min

Fig. 5. Thymidine incorporation by H35 cells. Incorporation of [ 3 H ] - t h y m i d i n e in u n t r e a t e d (O O), p e r m e a b l e (~ ~ ) and sealed (© ©) H35 cells was measured.

Permeable cells had little ability to incorporate thymidine into DNA, but sealed cells regained this capacity (Fig. 5). This loss in permeable cells was due to depletion of nucleoside triphosphates (Table III), since inclusion of a nucleotide mixture [8] reinstated thymidine incorporation in sealed cells. Further evidence for the facile translocation of nucleoside triphosphates comes from the measurement of ATP [29] in H35 cells. Untreated cells have 0.76 nmol/106 cells and this value is reduced in permeable cells to 0.016 nmol/106 cell 1 h after lysolecithin treatment. The ATP lost by the cells is quantitatively recovered in the medium. Cells which have sealed following lysolecithin treatment

have 0.5 nmol ATP/106 cells. Little loss of ATP is noted with control cells since there is 0.63 nmol A T P / 1 0 6 cells following treatment with solution A lacking lysolecithin. Because permeation allows nucleotides to be depleted from H35 cells, it should also permit their entry. With inclusion of the appropriate factors, d T T P should be incorporated into DNA in permeable cells but not in untreated or sealed cells. The results in Table III confirm this prediction. Little incorporation of [3H]dTTP was observed in untreated or sealed cells in the presence or absence of added nucleotides, thus verifying the reestablishment of membrane integrity in sealed cells, as demonstrated initially with trypan blue. Permeable cells readily incorporated [3H]dTTP, providing the necessary factors were included in the medium. The data for leucine incorporation into protein, measured in untreated, permeable and sealed cells (Table III), support the results from analogous experiments on thymidine incorporation into DNA. Untreated and sealed H35 cells exhibited nearly identical leucine incorporation; the permeable cells were deficient. Identical results are obtained when leucine incorporation is measured by acid precipitation of the cells and isolation of the ~dioactively-labeled protein by nitrocellulose filtration as in the case of thymidine and dTTP. Permeation with lysolecithin should also prevent concentrative transport of metabolites, since the altered plasma membrane is unable to restrict the movement of low-molecular weight charged molecules and cannot maintain a concentration

T A B L E III T H Y M I D I N E , dTTP A N D L E U C I N E I N C O R P O R A T I O N I N T O H35 CELLS Incorporation of [ 3H]thymidine, [3 ]dTTP and [3 H]leucine was measured in untreated, permeable (1 h after lysolecithin treatment) and sealed (20 h after lysolecithin treatment) cells. Cell status

Untreated Permeable Plus incubation mixture a Sealed Plus incubation mixture

Incorporation (pmol/106 cells/60 min) [ 3 H]thymidine [ 3 H]dTTP 22.1 0.5 9.8 21.2 19.3

0.01 0.2 5.3 0 0.083

[ 3 H]leucine 3.52 0.13 3.35

a Includes all nucleotides and deoxynucleotides described by Miller et al. [8] for D N A synthesis in permeable cells.

259

500

E

Eo_ 2 5 0 u

7

-i,

uJ

•

• I0

, 20 mln

i 50

oL 0

20 FRACTION

Fig. 6. Uptake of (+)5-CH3H4PteGlu by untreated (O e), permeable ( O - ©) and sealed (~3 ~) H35 cells. Uptake was measured as previously described [231. Fig. 7. Loss of folate coenzymes by permeable H35 cells. H35 cells were incubated with 5 /LM [3H]folic acid (6.104 dpm/nmol) for 24 h. The cells were extracted as described previously [17,24] and the contents chromatographed by Sephadex G-25 gel filtration [24,31]. Folic acid-eluted in fractions 19-23 and the folyl polyglutamates in fractions 5-13. The cellular composition of folate coenzymes in the cell was analyzied after the 24-h incubation (O O). Identical cultures were permeabilized with lysolecithin and incubated for 2 h in Swims S-77 medium lacking folic acid (O ©). Control cultures were treated with solution A lacking lysolecithin and then incubated for 2 h in Swims S-77 medium lacking folic acid (o ......

o).

gradient. We have previously established that a carrier-mediated active transport p a t h w a y exists in H35 cells for the serum form of folic acid, ( + ) 5 C H 3 H 4 P t e G l u [24]. The inability of the permeable cells to concentrate this metabolite is consistent with the loss of m e m b r a n e integrity (Fig. 6). Sealed cells accumulate ( + ) 5 - C H 3 H 4 P t e G l u to the same extent as untreated cells, again demonstrating the reversibility of the permeation process. W h e n H35 cells are incubated with the folate analog methotrexate, extensive conversion to methotrexate polyglutamates containing primarily 3 - 5 glutamate residues occurs [18,25]. W h e n the cells are then incubated in m e d i u m lacking methotrexate, the derivatives are retained more avidly than methotrexate which rapidly equilibrates with the extracellular space [25]. Treatm e n t of such cells with lysolecithin causes an immediate release of the methotrexate polyglutamates, indicating that their selective retention rela-

tive to methotrexate is due to reduced permeability [251. Folate coenzymes are also converted to polyglutamates by the liver, y-Glutamylation imparts two properties to the reduced folates that allow them to be utilized as coenzymes in cells: (i) the polyglutamates are less permeable and can be retained in sufficiently high concentrations to serve as coenzymes in one-carbon metabolism [30] and (ii) they are, generally more efficient substrates than the m o n o g l u t a m a t e s [30]. Our experiments with methotrexate polyglutamates indicated that lysolecithin treatment might be a useful technique to generate folate-deficient cells. This possibility was verified since the folate pool was quickly lost following lysolecithin treatment, while cells treated with solution A lacking lysolecithin retained virtually all of the cellular folates when placed in folate-free Swims medium (Fig. 7). The small a m o u n t of folates retained in the treated cell preparation corresponds to the m i n o r fraction of the culture that does not permeabilize during lysolecithin treatment (Table I and Fig. 1). Discussion

The technique described can be used to reversibly permeabilize a stable hepatocarcinoma cell line grown in m o n o l a y e r culture. Analogous procedures have been devised to permeabilize other cell lines [3-15]. Miller et al. [10] have demonstrated the utility of lysolecithin in permeabilizing a number of cell lines and reversibility of this process in certain cases. In our study with H35 cells more than 80% of the original cells are present in permeable cell preparation, and 90% of these are permeable. U p o n sealing, which takes approx. 16 h, all of the cells exclude trypan blue, with 63% recovery c o m p a r e d to controls. The procedure for permeabilizing H35 cells was also applied to p r i m a r y monolayer cultures of rat hepatocytes. While these are readily permabilized in equal or better yield than H35 cells, conditions have not been f o u n d which support sealing. After a 16-h incubation in enriched m e d i u m containing serum, approx. 80% of the permeable cells were attached to the culture plates, but they remained trypan blue-permeable. These cells, like the H35

260

cells were depleted of their folate pools (M. Balinska, unpublished data). In our various attempts permeabilize H35 cells hypo- and hypertonic salts were tried, but lysoelcithin treatment was the most effective. In every parameter tested the untreated and sealed H35 cells were identical. The permeable cells differed in allowing transmembrane passage of low-molecular weight charged molecules, such as trypan blue and nucleoside triphosphates, to which the cells are normally impermeable. There was very little loss of cellular protein or specific enzymes in either permeable or sealed cells. The utility of the procedure to develop folatedepleted cells in culture has been demonstrated. This had previously been done by growing the cells in the absence of folates but that approach required several weeks and the medium had to be supplemented with purines and pyrimidines [31,32]. The lengthy period required for depletion is due largely to the impermeable nature of the folyl polyglutamates which are slowly lost by mammalian cells [30]. Lysolecithin treatment rapidly depleted the cellular folate pool except for the small proportion of cells which did not permeabilize. The permeable cells can then seal in folate-free Swims medium, and the sealed cells can be used to study folate requirements for growth or folyl polyglutamate biosynthesis. The marked advantages of using depleted cells to study the latter problem are that (a) since the folate pools are negligible, quantitative synthesis of folyl polyglutamates can be measured directly with radioactively-labeled precursors; (b) the cells are devoid of cellular folates, so that folate species which may alter or regulate y-glutamylation can be selectively added; and (c) potential inhibitors which are impermeable can be introduced into the cells during the permeable phase. These same properties allow them to be used in the study of macromolecular synthesis in liver-derived cells [4-8,12] or of enzymes in situ [11,15]; as has been suggested with other permeabilized cell lines. The sealing process, which is impaired by inhibitors of protein synthesis and glcyosylation, invites speculation concerning its relation to membrane function. It may be that further analysis of sealing could be utilized as a model to elucidate certain aspects of membrane structure and biosynthesis.

Acknowledgements The authors deeply appreciate the excellent technical assistance of Zenia Nimec. The investigation was supported by N I H research grant AG00207 awarded by the National Institute on Aging, P H S / D H H S , and grant CA25933 from the National Cancer Institute, P H S / D H E W .

References l Bissell, D.M. (1981) Fed. Proc. 40, 2469-2473 2 Thompson, E.B. (1979) Methods Enzymol. 58, 544-552 3 Hilderman, R.H., Goldblatt, P.J. and Deutscher, M.P (1975) J. Biol. Chem. 250, 4796-4801 4 Hilderman, R.H. and Deutscher, M.P. (1976) Arch. Biochem. Biophys. 175, 534-540 5 Burger, N.A., Erickson, W.P. and Weber, G. (1976) Biochim. Biophys. Acta 447, 65-75 6 Seki, S. and Oda, T. (1977) Biochim. Biophys. Acta 476, 24-31 7 Costellot, J.J. Jr., Miller, M.R. and Pardee, A.B. (1978) Proc. Natl. Acad. Sci. U.S.A. 75, 351-355 8 Miller, M.R., Castellot, J.J., Jr. and Pardee, A.B. (1978) Biochemistry 17, 1073-1079 9 Lewis, W.H., Kuzik, B.A. and Wright, J.A. (1978) J. Cell Physiol. 94, 287-298 10 Miller, M.R., Castellot, J.J., Jr. and Pardee, A.B. (1979) Exp. Cell Res. 120, 421-425 11 Jorgenson, R.A. and Nordlie, R.C. (1980) J. Biol. Chem. 255, 5907-5915 12 Alonso, M.A. and Carrasco, L. (1980) Eur. J. Biochem. 109, 535-540 13 Johnson, P.A. and Johnstone, R.A. (1981) Can. J. Biochem. 59, 668-675 14 Rudick, M., Rudick, V., Magie, S. and Jacobson, E. (1981) In Vitro 17, 173-177 15 Ganlsema, H.S., Laanen, E., Gwen, A.K. and Tager, J.M. (1981) Eur. J. Biochem. 119, 409-414 16 Reuber, M.D. (1961) J. Natl. Cancer Inst. 26, 891-899 17 Pitot, H., Periano, C., Morse, P. and Potter, V.R. (1964) Natl. Cancer Inst. Monogr. 13, 229-245 18 Gali~an, J. (1979) Cancer Res. 39, 735-743 19 Tarentino, A.L. and Galivan, J.H. (1980) In Vitro 16. 833 846 20 Williams, G.M., Bermudez, E. and Scarameuzino, D. (1977) In Vitro 13, 809-813 21 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275 22 Bergmeyer, H.U. and Brent, E. (1974) in Methods of Enzymatic Analysis (Bergemyer, H.U., ed.), Vol.. 2, pp. 574579, Academic Press, New York 23 Diamondstone, T.L. (1966) Anal. Biochem. 16, 395-401 24 Galivan, J. (1981) Cancer Res. 41, 1757-1762 25 Balinska, M., Galivan, J. and Coward, J.K. (1981) Cancer Res. 41, 2751 2756 26 Dustin, P. (1978) Microtubules, pp. 167-215, Springer Verlag, Berlin

261 27 Elbein, A.D. (1981) Trends Biochem. Sci. 6, 219-221 28 Krawitt, E.L., Baril, E.F., Becker, J.G. and Potter, V.R. (1970) Science 16, 294-296 29 Lamprecht, W. and Troutschold, I. (1974) in Methods of Enzymatic Analysis (Bergmeyer, H.U., ed.), Vol. 4, pp. 2101-2110

30 McGuire, J.J. and Bertino, J.R. (1981) Mol. Cell Biochem. 38, 19-48 31 Moran, R.G., Werkheiser, W.C. and Zakrzewski, S.I. (1976) J. Biol. Chem. 251, 3569-3575 32 Hilton, J.G., Cooper, B.A. and Rosenblatt, D.A. (1979) J. Biol. Chem. 254, 8398-8403