EXPERIMENTAL
CELL
RESEARCH
189.93-99
(1990)
Transformed SV3T3 Cells Have a Reduced Lysosomal Compartment and Lower Levels of Enzyme Activity than 3T3 Cells V. KUNDRA AND M. F. DEAN Kennedy Institute of Rheumatology, 6 Bute Gardens, Hammer-smith, London W6 70 W, United Kingdom
the intracellular and secreted activities of a range of acid hydrolases present in actively dividing, normal, and SV40-transformed 3T3 cells to determine if there were any differences. The number of lysosomes, their total volume, and the surface area of their membranes were also calculated. Lysosomal membranes were purified from each cell type and their protein composition was examined using gel electrophoresis to determine whether any changes we observed in enzyme activity could be explained solely by alteration in lysosomal number or volume or whether, in addition, there were contributory changes in the structure of their membranes.
The secreted and intracellular activities of a number of lysosomal hydrolases were higher in 3T3 cells than in SV40-transformed cells. The number of lysosomes and their total volume were also much larger in 3T3 cells and the surface area of their lysosomal membranes was almost twice that of SV3T3 cells. These differences alone were not sufllciently large, however, to account for the disparity seen in activity of some enzymes. Gel electrophoresis showed that a number of protein components present in lysosomal membranes purified from 3T3 cells were absent from SV3T3 membrane preparations. The absence of these components may be correlated with the reduced enzyme activity of SV3T3 cells particularly with respect to &glucosidase and acid phosphatase, both of which are normally found associQ 1990 Academic Press, Inc. ated with lysosomal membranes.
MATERIALS
AND METHODS
Cell czzlture. Swiss 3T3 cells (Flow Laboratories, Rickmansworth, Herts) and SV3T3 cells, kindly donated by Dr. Fiona Watt (ICRF, London), were grown at 3’7°C in Falcon T-175 flasks containing HAMS F-12 medium supplemented with 10% heat-inactivated fetal calf serum or bovine serum (Imperial Laboratories), 2 mM L-glutamine, 20 mM Hepes, 0.75% sodium bicarbonate, 50 pg/ml penicillin, 50 pg/ml streptomycin. Cells were passaged when they were 80-90% confluent. Purification of lysosomes. All purification procedures were carried out at 4°C. Cells were disrupted by nitrogen cavitation at 70 psi for 15 min using 10s cells in 3.5 ml isolation buffer (10 mM triethanolamine, 10 mM acetic acid, 1 mM EDTA. 0.25 M sucrose, pH 7.4) according to the method of Rome et al. [S]. The suspension was centrifuged at 8OOg for 10 min in a swing-out rotor (MSE Chilspin) and the postnuclear supernatant saved. After resuspending the pellet in 1 ml isolation buffer by 10 triturations with a Pasteur pipet it was recentrifuged and the last step repeated using 0.5 ml buffer. The combined postnuclear supematants were centrifuged at 33,000g for 20 min in a Sorvall OTD 65 centrifuge (6 X 5 ml swing-out rotor); the pellet was resuspended in 2 ml isolation buffer containing 1 mg BSA and layered onto 39 ml of Percoll (Pharmacia Fine Chemicals), starting density 1.05 g/ml. After centrifugation in an angle-head rotor (Sorval SS-34) for 1 hat 45,OOOg, fractions of approximately 1 ml were collected from the bottom of the gradient and assayed for enzyme activities. The bottom 10 fractions were pooled, diluted 1:4, and centrifuged at 33,000g for 20 min in an angle-head rotor to remove Percoll. This step was repeated, and the bottom 4 ml was diluted 1:lO and centrifuged again at ll,OOOg for 20 min in a swing-out rotor (Sorvall SS-37) to obtain a crude lysosomal pellet. Lysosomal membranes were then separated from contaminating mitochondria using a modification of the technique of Ohsumi et al. [9]. The pellet was resuspended in 1 ml buffer A (0.025 M sucrose, 1 mM EDTA, 1 mM dithiothreitol, 1 mM phenylmethylsulphonyl fluoride), left on ice for 30 min, then centrifuged at 11,OOOgfor 20 min (Biofuge). The resulting mitochondrial pellet was resuspended in 0.6 ml buffer B (0.2 M KCl, 0.025 M sucrose, 1 mM EDTA, 1 mM dithio-
INTRODUCTION Lysosomes are membrane-bound organelles of heterogeneous morphology that contain a large number of acid hydrolases and are thus able to degrade a wide range of macromolecules. Their surrounding membranes are able to prevent the entry and exit of high-molecular-weight materials, are resistant to hydrolysis, and help to maintain an acidic intralysosomal pH by means of an ATPdependent proton pump [l]. Lysosomal membranes have a defined structure and contain a number of wellcharacterized components, including specific glycoproteins of 80, 100, and 120 Kda [2,3] that are not present in other intracellular organelles. Many of the enzymes found within lysosomes are glycosylated [4], and a characteristic of tumor cells irrespective of their origin is that they often contain altered glycoproteins [ 5 1. Furthermore, abnormalities in both the synthesis and the secretion of lysosomal enzymes have been observed frequently in many tumor cells [6]. Mouse 3T3 fibroblasts transformed with Simian virus 40 display many of the characteristics of typical tumorderived cells, including loss of contact inhibition of growth [7]. They can be used therefore as a model cell to determine if viral transformation of cells can affect their lysosomal system. In these experiments we measured 93
All
0014~4S27/90 $3.00 Press, Inc. reserved.
Copyright Q 1990 by Academic of reproduction in any form
rights
KUNDRA
94
threitol, 1 mMphenylmethylsulphony1 fluoride) and centrifuged again at ll,OOOg for 20 min. The two supernatant solutions were combined and spun at 90,OOOgfor 1 h in a Beckman DL-100 ultracentrifuge to obtain a pellet containing purified lysosomal membranes, which was stored at -70°C until used. Enzyme assays. Lysosomal enzymes were measured fluorometritally using 4-methylumbelliferyl substrates as described by Olsen et al. [lo]. In addition, acid phosphatase activity was determined using 4-methylumbelliferyl phosphate in 0.2 M citrate buffer, pH 4.8, as a substrate (20 pl sample and 80 ~1 substrate). Cytochrome e was used as a substrate for cytochrome oxidase activity [ll] and sodium benzylamine for monoamine oxidase [9]. Catalase activity was detected using a modification of the procedure of Baudhuin et al. [12], with 0.05 mit4 HzOz in 10 mM imidazole, 0.1% bovine serum albumin, pH 7.2, as the substrate. Substrate (20 ~1) or buffer without H,O, as a blank was added to 20 ~1 of sample and incubated at room temperature for 30 min, and the reaction was stopped with 120 pl of 15% w/v solution of titanium (IV) sulfate (BDH). The absorbance was read immediately at 414 nm. Activity of 5’-nucleotidase was measured by the method of Avruch et al. [13], and galactosyl transferase determined using the method of Abraham [ 141.Protein concentration was assayed using the Biorad protein determination kit [15]. Electrophoresis. Samples of lysosomal membrane were solubilized by boiling them for 5 min in 0.1% SDS, 1% j3-mercaptoethanol, 10 mM Tris, 1 mM EDTA, and 0.005% bromophenol blue, containing 10% sucrose, and electrophoresed on 7.5% acrylamide gels [40] at 25 mA/ gel for 16 h. Proteins were visualized using a modification of Heukeshoven’s silver stain [ 161. Acridine orange. Cells (4 x lo4 in 1 ml medium) were plated onto coverslips in Linbro 24-well culture dishes and left overnight to monolayer. After a wash with PBS, acridine orange (1 pg/ml in medium) was added, and the cells were incubated for 2 h at 37°C and washed three times in PBS, and the coverslips were mounted onto slides. Fluorescence was observed using a Nikon Diaphot microscope fitted with a blue filter. Pun&ate orange fluorescent vesicles were counted in a single visual plane of individual, nonmitotic cells using a Xl00 objective with a focal depth of 0.6 pM. Total orange fluorescence was determined using a Zeiss microfluorometer fitted with a Bp 436-5 filter to measure at wavelengths between 460 and 475 nm and a X40 objective with focal depth of 1.0 pm. Lysosomal enzyme assays. Cells were plated out at a concentration of 5 X lo6 in 2 ml medium in 6 X 35-mm Linbro culture dishes and allowed to monolayer overnight. The medium was then replaced with an additional 2 ml of fresh medium, which was collected after a 24-h period of incubation. The cells were washed three times with PBS, lysed by freezing and thawing them in 0.5 ml of 0.1% Triton X-100, and centrifuged at ll,OOOg for 30 s. Both media and cell extracts were assayed for lysosomal enzyme activities. Into each well of 6 X 35-mm Uptake of ‘=I-Polyvinylpyrrolidone. Linbro culture dishes, 5 X ld cells were seeded in 2 ml Hams F12 medium and left overnight. They were then washed with PBS and incubated in fresh serum-free medium containing 6 X 10” cpm 1261-polyvinylpyrrolidone per milliliter. After an appropriate period of incubation, the cells were washed three times in PBS, lysed with 0.5 ml of 0.1% Triton X-100, and freeze-thawed. The suspension was centrifuged to remove insoluble material and the radioactivity in 0.25 ml of the supernatant solution was counted in a LKB compugamma counter. RESULTS
Lysosomal Enzyme Activity The intracellular activity of all of the enzymes tested was greater in 3T3 cells than in SV3T3 cells (Table 1) with activity ratios ranging from 1.53 for @-glucuronidase to 14.05 for &galactosidase. To determine whether
AND DEAN
these differences were due to an increased rate of secretion by SV3T3 cells, resulting in a reduction in intracellular levels, the activity of enzymes released during a 24h period of incubation was measured. Only in the case of /I-hexosaminidase was activity in medium from SV3T3 cells slightly higher than that from 3T3 cells (1.28 times greater). There was no significant difference in the activity of ,&glucuronidase, while secreted acid phosphatase and @galactosidase activities were 2.2 and 7.0-fold higher in medium collected from 3T3 cells than in that
from SV3T3 cells. @-glucosi&se me&um from either type of ce]]
wag not &&e&d
in the
Although in general larger amounts of enzyme activity were released by 3T3 cells than by SV3T3 cells, secreted enzyme as a fraction of total activity (secreted plus intrace]]u]~)
was
approximately
twice
as high
in
SV3T3
fibroblasts. Conversely, intracellular activity as a proportion of total activity was greater in 3T3 cells. Although SV3T3 cells did secrete a greater proportion of their total activity than did 3T3 cells, the amounts secreted were not sufficiently large to account for the lower ] evels of activity present intracellularly.
Arcridine OrUnge Acridine orange was used as a marker with which to determine the number of low-pH vesicles present in each type of cell. After uptake, it was seen as a punctate, bright orange fluorescence, concentrated in low-pH vesicles generally perinuclear in distribution in 3T3 cells but more peripheral in SV3T3 cells. Fluorescence became optimal after a 2-h incubation. The single visual plane that resolved the greatest number of vesicles was chosen for each cell and counting completed quickly before overexposure to ultraviolet light induced rupture of the vesicles and quenched their fluorescence. We assumed that the low-pH vesicles in each visual plane were randomly distributed and that determining the ratio of 3T3: SV3T3 vesicles minimized error since every plane should contain the same ratio. Frequency histograms of vesicle numbers showed a bell-shaped distribution of low-pH vesicles in both cell types (Fig. 1). Although there was some degree of overlap between the two, the mean number of vesicles in 3T3 cells was higher than that in SV3T3 cells by a factor of 1.6 (P < 0.001, n = 124) as shown in Table 2. Overall orange fluorescence was measured in addition to vesicle number to exclude the possibility that vesicles too small to be resolved had been omitted during counting. The microfluorometer was blanked using dead cells which fluoresced green due to staining of their nucleic acids with acridine orange. The same criteria as those employed for vesicle counting were used to select cells. When frequency histograms of arbitrary fluorescence units per cell were plotted (Fig. 2), they showed that 3T3 cells had a fluorescence greater than that of SV3T3 cells (n = 76, P < 0.01). Since this semiquantitative method
LYSOSOMAL
ENZYMES
IN SV3T3
TABLE Comparison
Intracellular activities fi-glucuronidase fl-galactosidase &hexosaminidase &glucosidase Acid phosphatase
27.41* 86.14 f 360.12 + 20.74+ 259.1 +
Secreted activities @-glucuronidase @-galactosidase @hexosaminidase @-glucosidase Acid phosphatase
0.88 5.71 5.36 1.88 12.86
1.99 + 1.83 + 36.03 + nd 10.65 z!z
0.28 0.41 3.07 3.52
95
3T3 CELLS
1
of Enzyme Activities 3T3 cells (1)
AND
in 3T3 and SV-3T3 Cells Ratio (1:2)
SV-3T3 cells (2)
17.93 6.13 126.02 6.98 50.51
f 1.43 + 0.38 k 4.52 f 0.51 z!z5.89
1.53 14.05 2.86 2.97 5.12
2.27 2 0.32 0.26 2 0.13 46.23 f 7.3 nd 4.5 iI 0.59
0.87 7.0 0.78 2.37
Note. Medium was collected from cell monolayers after 24 h of culture; enzyme activities were measured and expressed as &/h/ml of medium. The remaining cells were washed and lysed in 0.1% Triton X-100, and their activity was expressed as r&f/h/ml of solution. All values represent the means of eight separate experiments.
measured total orange fluorescence, it showed not only that the number of low-pH vesicles in each cell type differed, but also that their area and hence their probable volume were also different. Uptake of ‘251-Polyvinylpyrrolidone
Since uptake of ‘25-I-PVP is not receptor-mediated and it is not broken down within cells, it can be used to determine the amount of liquid taken up by fluid-phase pinocytosis [ 171and thus provide a measure of lysosomal volume when this has reached equilibrium. lz51-PVP is known to accumulate within lysosomes [ 18-201 and does not affect the rate of fluid-phase pinocytosis [ 171. It has been used previously to determine the volume of the endocytic compartment in cells [21].
In our experiments the initial rates of uptake of “‘IPVP into both 3T3 and SV3T3 cells were equivalent (Fig. 3), suggesting that the rate of fluid-phase pinocytosis was the same for each cell type. However, after 12 h of uptake the amount of ‘251-PVP present in 3T3 cells was almost double that found in SV3T3 cells. When cells were allowed to take up this labeled material for 24 h the amount of radioactivity present within both types of cell remained constant between the 12- and 24-h time points, indicating that their respective endocytic compartments consisting primarily of lysosomes had been saturated. At this equilibrium the ratio of 3T3 to SV3T3 lysosomal volume was calculated to be 1.9 f 0.3 (Table 2). Lysosomal Surface Area
In order to compare the lysosomal surface areas of 3T3 cells and SV3T3 cells, lysosomal shape was assumed to be spherical. The formula TABLE
2
Physical Characteristics of Lysosomes from 3T3 and SV40-3T3 Cells Ratio 3T3/SV40-3T3
20
40
6G
*o
100
120
140
160
180
200
220~240
numberoflysosomes
FIG. 1. Frequency distribution of fluorescent vesicles. After 2 h incubation with acridine orange coverslips were washed and mounted on slides, and the number of pun&ate, orange fluorescent vesicles in each cell was counted. Figure represents the results obtained from 124 separate 3T3 cells (solid bars) and from 124 SV3T3 cells (crosshatched bars).
Volume (iz51 uptake/24 h) Number/visual plane/cell Surface area/cell
1.9 + 0.3 (4) 1.6 + 0.2 (124) 1.8 + 0.2
Note. Ratios were calculated from the number of separate experiments or number of cells observed, given in parentheses and expressed as a mean value *SD. Cells were allowed to phagocytose ‘%I-polyvinylpyrrolidone until cell-associated radioactivity reached a constant value after which the lysosomal compartment was considered to be
saturated. Acridine orange was ueed as a marker with which to count the number of lysosomes per cell and lysosomes were assumed to be spherical when membrane surface area was calculated.
96
KUNDRA
20
40
60
80
100
120
140
160
180
200
220
240
260
Fluorescence (UnitS)
FIG. 2. Total fluorescence per cell. After uptake of acridine orange the overall fluorescence per cell was measured using a microdensitometer as described in the text and expressed as arbitrary units of fluorescence. Numbers represent the values recorded for 124 separate 3T3 cells (solid bars) and for 124 SV3T3 cells (cross-hatched bars).
S, 4X7rXrfXN, -= S2 4 X T X r; X N2 was used where r is the half-diameter of each lysosome, S is the lysosomal surface area/cell, V is the lysosomal volume/cell (ml), and N is the lysosomal number/visual plane/cell. If r is the average half-diameter of lysosomes then 113
(V,/(+T X N&)2’3 X Nl “* ”
= (v,/( 2~ X N2))2’3 x N2 ’
AND
DEAN
ent (Fig. 4) maximal enzyme activity was found at a density of 1.06-1.07 g/ml. Similar distribution patterns (data not shown) were obtained for the other lysosomal enzymes, /?-galactosidase, &glucuronidase, and &glucosidase. Lysosomes were well separated from the plasma membrane marker 5’-nucleotidase and from the golgi complex marker galactosyl transferase, both of which were present primarily in the top third of the gradient. Catalase activity formed a broad peak in the middle of the gradient with a maximum at a density of 1.045 g/ml. Mitochondrial cytochrome oxidase activity was more buoyant than that of lysosomes, peaking at a density of 1.055 g/ml, although almost 50% of the activity did overlap the lysosomal fractions. Lysosomal membranes were separated from contaminating mitochondria by differential centrifugation and lysed using buffers A and B (see Methods). Almost 90% of the inner mitochondrial membrane marker, cytochrome oxidase, and 73% of monoamine oxidase, an outer mitochondrial membrane marker, were present in the final pellet, with 70% of the acid phosphatase and fihexosaminidase activities remaining in the supernatant solution (Table 3). Purified lysosomal membranes were isolated from this supernatant by a final centrifugation at 90,OOOg. After electrophoresis on 7.5% acrylamide gels the pattern of components resolved from these membranes was clearly different from those obtained for mitochondrial and endosomal fractions (data not shown). In addition a number of differences between lysosomal membranes from both types of cell were distinguishable (Fig. 5). In particular a number of components with apparent molecular weights of 110,51,40, and 33 Kda present in 3T3 cells were absent from SV3T3 cells. These bands were not detected in the lysosomal membranes purified from SV3T3 cells during three separate experiments. DISCUSSION
This reduces to S1 _ V12’3x Nl”3 s,- V22/3x N 2 113This formula takes into account the ratio of both lysosomal number and volume in order to compare total lysosomal surface area per cell. Calculation showed that the area per cell was 1.8 + 0.2-fold greater in 3T3 cells than in SV3T3 cells (Table 2).
The biochemical and physical properties of lysosomes from these two types of 3T3 cell showed considerable
24r
Lysosomal Membranes Both @-hexosaminidase and acid phosphatase were used as markers for lysosomes during purification. Almost 65% of the acid phosphatase activity and 80% of the /3-hexosaminidase activity from each cell type were found in the first 10 ml of the gradient. After centrifugation through a l.O&g/ml starting-density Percoll gradi-
I 0
1
a
16
24
time (hr)
FIG. 3. Uptake of ‘251-labeled polyvinylpyrrolidone. The rate of accumulation of radiolabeled polyvinylpyrrolidone with time was measured in 3T3 (0) and SV3T3 (m) cells over a period of 24 h. Data represent the means + standard error for four separate experiments.
LYSOSOMAL
ENZYMES
IN SV3T3
AND
97
3T3 CELLS
l.osr A
volume (ml)
volume (ml)
FIG. 4. Purification of lysosomes. Cells were disrupted by nitrogen cavitation and fractionated g/ml, as described in the text: (A) 3T3 cells; (B) SV3T3 cells. Fractions of 1 ml were collected nucleotidase (6), cytochrome oxidase (m), and fi-galactosyl transferase (A).
differences. Intracellular enzyme activity, lysosomal number, volume, and lysosomal membrane surface area were all greater in 3T3 cells than in SV3T3 cells. In addition a number of proteins present in the lysosomal membrane of 3T3 cells were absent from SV3T3 cells. In a previous study, Bosmann [22] reported that both P-glucosidase and /3-galactosidase activities were greater in SV40-transformed than in normal 3T3 cells. However, their cells had been cultured in Eagle’s medium supplemented with 10% serum and ascorbic acid. Our enzyme activities were measured fluorometrically using freshly disrupted cells grown in Hams F-12 medium containing heat-inactivated serum. Both cell types grew better in Hams F-12 than in modified Eagle’s medium. Furthermore, the samples analyzed by Bosman were extracted for 16 h at 4”C, lyophilized, rehydrated, and stored frozen before assay. This long extraction time might inactivate some enzymes. In contrast, Orkin et al. TABLE
on a Percoll gradient, starting density 1.05 and assayed for &bexosaminidase (O), 5’-
[23] found that intracellular activities of both &glucuronidase and P-hexosaminidase were higher in 3T3 than in SV3T3 cells. However, they did not find significant differences in the activity of secreted p-hexosaminidase although they did use a less sensitive spectrophotometric method of enzyme assay, and their cells had been grown in serum-free medium. Although we found that secreted enzyme activities were higher in 3T3 cells, SV3T3 cells secreted a much greater proportion of their total activity than they stored in lysosomes. Other similar studies using Novikoff hepatoma cells and cells from animals bearing this tumor have also shown that their acid phosphatase activity and their lysosomal number were reduced [24]. Secretion of a lower proportion of their lysosomal enzymes, however, could not by itself account for the higher activity we observed in 3T3 cells. When acridine orange was used as a marker to determine the number of lysosomes [2,25] we 3
Purification of Lysosomal Membranes Fraction
Hexosaminidase
Acid phosnhatase
Cytochrome oxidase
Monoamine oxidase
SV40-3T3 cells Crude lysosomal fraction ll,OOOg spin supernatant ll,OOOg spin pellet Membrane supernatant Membrane pellet
100 77 23 38 37
100 73 27 24 50
100 10 90 3 7
100 27 73 3 24
3T3 cells Crude lysosomal fraction ll,OOOg spin supernatant ll,OOOg spin pellet Membrane supematant Membrane pellet
100 73 27 38 35
100 70 30 19 51
100 12 88 4 8
100 27 73 2 25
Note. The lysosomes obtained by fractionation of Percoll were lysed and centrifuged as described in the text. Enzyme measurements were made at each stage of the purification to confirm that lysosomal membranes were free of mitochondrial contamination. Lysosomal enzyme activities show mean percentage recoveries from six separate experiments, cytochrome and monoamine oxidases means from three separate experiments.
98
KUNDRA
Mr x1o-3 ~ -116
'-26
FIG. 6. Gel electrophoresis of lysosomal membranes. Lane 1 shows a representative sample of solubilized membranes from 3T3 cells and lane 2 a sample from SV3T3 cells stained with silver. The position of molecular weight markers is shown alongside for comparison. Protein components present in 3T3 cells but not in SV3T3 cells are indicated by arrows.
found this was 1.6 times higher in 3T3 cells than in SV3T3 cells. A difference of this magnitude could account for the increase of 1.53-fold in the /3-glucuronidase activity of 3T3 cells, since this enzyme was secreted in approximately equal amounts by both cell types. It would not, however, account for the differences in other lysosomal enzymes, which were more than 1.6 fold greater in 3T3 cells. It has been argued previously that increased secretion of lysosomal enzymes may lead eventually to alteration of plasma membrane glycoproteins and hence to uncontrolled growth [26, 271. Mitosis is associated with rearrangement and decrease in lysosomal number [28] and to avoid any effect this may have had in our observations, all measurements of the number of low-pH vesicles were carried out using individual, nonmitotic cells. Low-pH vesicles were observed in cytoplasmic projections and at the periphery of SV3T3 cells but were mainly perinuclear in 3T3 cells. A similar peripheral distribution of low-pH vesicles has been observed previously up to 10 days after nonlytic SV40 virus infection of 3T3 cells [29]. Because of the uneven distribution of these vesicles and the possibility that some may have been too small to observe during counting, total acridine orange fluorescence was also measured. 3T3 cells showed a greater degree of fluorescence than SV3T3 cells, which confirmed the data obtained from vesicle counting and also indicated that the area of low-pH vesicles was also greater in 3T3 cells. Although the volume of the lysosomal compartment in 3T3 cells was almost twice that of SV3T3 cells when measured by uptake of ‘251-polyvinylpyrrolidone, this difference was less than the differences in activity for all enzymes tested except b-glucuronidase. The slightly reduced rate of secretion of &hexosaminidase (1.26 fold) and the greater lysosomal volume (1.9-fold) of 3T3
AND DEAN
cells could, by themselves, account for the increase in intracellular activity seen in this enzyme. Neither secretion nor simple physical properties, such as lysosomal volume and number, could, however, account for the large differences in activity of &galactosidase and acid phosphatase. The higher activities of these enzymes in 3T3 cells may have been due to an increased rate in synthesis and/or greater substrate specificity since both stored and secreted enzyme activities were elevated. The lack of @-glucosidase activity in medium collected from both types of cell can be explained by the fact that this enzyme is normally tightly bound to lysosomal membranes [30]. A major proportion of leukocyte acid phosphatase and 80% of hepatocyte acid phosphatase is also found associated with lysosomal membranes [9, 301. In our preparations of purified lysosomal membranes, 70% of the acid phosphatase was bound. The ratios of lysosomal volume and number in 3T3 cells and SV3T3 cells were used to derive the ratio of total lysosomal membrane surface area per cell (Table 2), but this ratio was considerably lower that those calculated for @-glucosidaseand acid phosphatase activities (Table 1). Thus, the relative activity of these two enzymes does not appear to be dependent solely on lysosomal surface area. It is possible that there was a greater concentration of these enzymes in membranes from 3T3 cells or that the turnover number of each enzyme molecule was greater. Alternatively, our estimate of the ratio of surface area may be conservative because lysosomes are of various sizes and do not always approximate a sphere [31], while our calculation assumed that all lysosomes were spherical and had the same radius. Lysosomes were purified using Percoll at 1.05 g/ml because denser gradients, such as those used previously to isolate lysosomes from NIH 3T3 cells [32] and Balb/c 3T3 cells [33], did not separate our lysosomes completely from other organelles. A number of distinct membrane components with molecular weights of lOO110 and 120 kDa [2, 3, 34, 351 have been described in other lysosomal preparations, and different types of cell have been shown to contain lysosomal components of slightly differing sizes [34]. Lysosomal membranes from our 3T3 cells contained a number of predominent proteins which were totally absent from SV3T3 cells. This finding, along with the dissimilarity we observed in Bglucosidase and acid phosphatase activities, showed that not only lysosomal contents but also the composition of their membranes differed significantly between 3T3 and SV3T3 cells. It is possible, therefore, that these observations are related and that alterations in membrane proteins may affect the anchorage or retention of lysosomal enzymes. There have been a number of reports indicating that lysosomal stabilizing agents can retard certain types of cancer [35,36] and that agents which stabilize lysosomal membranes can also inhibit cell division [37]. With cessation of growth, 3T3 cells display a twofold increase in
99
LYSOSOMAL ENZYMES IN SV3T3 AND 3T3 CELLS the number of early autophagic vacuoles and an increase in their protease activity [38]. Wharton et al. [39] suggested that the increase in lysosomal enzymes associated with cell aging could be due to the increase in population doubling time of aged cells, which in turn may cause lysosomes to accumulate. Mitosis is associated with lysosome rearrangement and loss, and 3T3 cells transformed with SV40 virus undergo mitosis more frequently than normal 3T3 cells [28]. Therefore, the decreased enzyme activity, lysosome number, volume, and membrane surface area seen in SV3T3 cells may all result from the greater rate of division associated with their transformed phenotype. Alternatively, transformed cells frequently have reduced growth requirements, and a reduction in the need for external nutrients may in turn be reflected in a smaller lysosomal compartment. The authors gratefully acknowledge the help given by Dr. Helen Muir and Sister Helen Christensen and wish to thank Dr. J. Chayen and Dr. L. Bitensky for use of their microfluorometer. We also acknowledge the helpful discussion with Drs. Fojo, Keefer, Graham, Cunningham Dobler, and Swanson. This work was supported by Fulbright-Haves Grant PL87-256 awarded to V.K.
14. 15.
19.
Heukeshoven, J., and Dernick, R. (1985) Ekctrophresis 6,103112. Williams, K. E., Kidston, E. M., Beck, F., and Lloyd, J. B. (1975) J. Cell Biol. 64,113-122. Allison, A. C., and Black, P. H. (1967) J. Natl. Cancer Inst. 39, 775-787. Meijer, A. E. F. H., and Willinghagen, R. G. J. (1963) Biochem.
20.
England, I. G., Naess, L., Bromhoff,
16. 17. 18.
Pharmncol. 12,973-980.
1. 2. 3. 4.
5.
Ohkuma, S., and Poole, B. (1978) Proc. Natl. Acad. Sci. USA 76, 3327-3331. Reggio, H., Bainton, D., Harms, E., Coudrier, E., and Louvard, D. (1984) J. Cell Biol. 99,1511-1526. Lewis, V., Green, S. A., Marsh, M., Vihko, P., Helenius, A., and Mellman, I. (1985) Cell Bial. 100,1839-1847. Dean, R. T. (1975) in Lysosomes in Biology and Pathology (Dingle, J. T., and Dean, R. T., Eds.), Vol. 4, pp. 349-382, North Holland, Amsterdam. Lenti, L., Lipari, M., Lombardi, D., Zicari, A., Dotta, A., and Pontieri, G. A. (1986) Brit. J. Exp. Pathal. 67,793-799.
6. 7.
&ego, C. M., and Pietras, R. J. (1984) Znt. Reu. Cytol. 88.1-301. Lockwood, T. D., and Shier, W. T. (1977) Nature (London) 267, 252-254.
8.
Rome, L. H., Garvin, A. J., Allieta, M. M., and Neufeld, E. F. (1979) 7,143-153. Ohsumi,Y., Ishikawa, T., andKato, K. (1983) J. Biockm. 93(2), 547-556.
9. 10. 11. 12.
13.
Olsen, I., Dean, M. F., Muir, H., Smith, R., Jenne, B. M., and Hand, C. (1983) Cell B&hem. Fun&. l(2), 103-108. Cooper&in, S. J., and Lazarow, A. (1951) J. Biol. C&m. 189, 665-670. Baudhuin, P., Beaufay, H., Rahman-Li, U., Sellinger, 0. Z., Wattiaux, R., Jacques, P., and DeDuve, C. (1963) Biachem. J. 92, 179-184. Avruch, J., and Wallach, 233,334-347.
D. F. H. (1976) Biochem. Biophys. Acta
Received November 23,1989 Revised version received March 6,199O
R., and Berg, T. (1986) Bio-
them. Pharmacol. 36,201-208. 21.
McKinley,
D. N., and Wiley,
H. S. (1988) J. Cell. Physiol.
136,
389-397. 22. 23. 24. 25. 26. 27.
REFERENCES
Abraham, D., Bou-Gharios, G., Muir, H., Olsen, I., and Smith, R. (1986) Eur. J. Cell Biol. 40,16’7-175. Bradford, M. (1976) Anal. Biochem. 72,243-254.
28.
Bosmann, H. B. (1969) Exp. Cell Res. 64,217-221. Orkin, R. W., Underhill, C. B., and Toole, B. P. (1982) J. Biol. Chem. 267(10), 5821-5826. Urban, J. L., and Unsworth, B. R. (1977) Erperientiu 30,12171219. Moriyana, Y., Takamo, T., and Okhuma, S. (1982) J. Biochem. 92,1333-1336. Henson, P. M. (1976) (Dingle, J. T. and Dean, R. T., Eds.), pp. 99-126, North Holland, Amsterdam. Poole, A. R. (1973) in Lysosomes in Biology and Pathology (Dingle, J. T. Ed.), Vol. 3, pp. 303-337, North Holland, Amsterdam. Sylven, B., Snellman, O., and Strauli, P. (1974) Virchws Arch.
B:Cell Pathol. 17,143-153. 29.
Allison,
A. C., Harington,
J. S., and Birbeck,
M. (1966) J. Exp.
Med. 124,141-154.
32.
Avila, J. L., and Casanova, M. A. (1981) Acta Cien Venez 32,8187. Ahlberg, J., Marzella, L., and Glaumann, H. (1982) Z.&I. Invest. 47(6), 523-532. Gal, S., Willingham, M. C., and Gottesman, M. M. (1985) J. Cell
33.
Merion,
30. 31.
Biol. 100,535-544. M., and Poretz, R. D. (1981) J. Supramol. Struck Cell.
Biochem. 17,337-346. 34. 35.
Chen, J. W., Murphy, T. L., Willingham, M. C., Pastan, I., and August, J. T. (1985) J. Cell BioL 101,85-95. Bagwell, C. E., and Ferguson, W. W. (1980) J. Surg. Res. 28, 410-415.
36.
Gross, J., Azizikhan, R. G., Biswas, C., Bruns, R. R., Hsieh, D. S. T., and Folkman, J. (1981) Proc. Natl. Acad Sci. USA 78, 1176-1180.
37.
Redai, I., and Halmy, 27(l), 41-45.
38.
Papadopoulos, 121.
39.
Wharton, S. A., and Riley, P. A. (1984) in Lysosomes in Biology and Pathology (DingIe, J. T., Dean, R. T., and Sly, W., Eds.), Vol. 7, pp 401-420, North Holland, Amsterdam.
40.
Fairbanks,
M. (1980) Acta Microbial. Acad. Sci. Hung.
T., and Ulrich,
P. (1987) Exp. Cell Res. 171, llO-
G., Steck, T. L., and Wollach,
chemistry 10,2606-2616.
D. F. H. (1971) Bio-
CELL
RESEARCH
189.93-99
(1990)
Transformed SV3T3 Cells Have a Reduced Lysosomal Compartment and Lower Levels of Enzyme Activity than 3T3 Cells V. KUNDRA AND M. F. DEAN Kennedy Institute of Rheumatology, 6 Bute Gardens, Hammer-smith, London W6 70 W, United Kingdom
the intracellular and secreted activities of a range of acid hydrolases present in actively dividing, normal, and SV40-transformed 3T3 cells to determine if there were any differences. The number of lysosomes, their total volume, and the surface area of their membranes were also calculated. Lysosomal membranes were purified from each cell type and their protein composition was examined using gel electrophoresis to determine whether any changes we observed in enzyme activity could be explained solely by alteration in lysosomal number or volume or whether, in addition, there were contributory changes in the structure of their membranes.
The secreted and intracellular activities of a number of lysosomal hydrolases were higher in 3T3 cells than in SV40-transformed cells. The number of lysosomes and their total volume were also much larger in 3T3 cells and the surface area of their lysosomal membranes was almost twice that of SV3T3 cells. These differences alone were not sufllciently large, however, to account for the disparity seen in activity of some enzymes. Gel electrophoresis showed that a number of protein components present in lysosomal membranes purified from 3T3 cells were absent from SV3T3 membrane preparations. The absence of these components may be correlated with the reduced enzyme activity of SV3T3 cells particularly with respect to &glucosidase and acid phosphatase, both of which are normally found associQ 1990 Academic Press, Inc. ated with lysosomal membranes.
MATERIALS
AND METHODS
Cell czzlture. Swiss 3T3 cells (Flow Laboratories, Rickmansworth, Herts) and SV3T3 cells, kindly donated by Dr. Fiona Watt (ICRF, London), were grown at 3’7°C in Falcon T-175 flasks containing HAMS F-12 medium supplemented with 10% heat-inactivated fetal calf serum or bovine serum (Imperial Laboratories), 2 mM L-glutamine, 20 mM Hepes, 0.75% sodium bicarbonate, 50 pg/ml penicillin, 50 pg/ml streptomycin. Cells were passaged when they were 80-90% confluent. Purification of lysosomes. All purification procedures were carried out at 4°C. Cells were disrupted by nitrogen cavitation at 70 psi for 15 min using 10s cells in 3.5 ml isolation buffer (10 mM triethanolamine, 10 mM acetic acid, 1 mM EDTA. 0.25 M sucrose, pH 7.4) according to the method of Rome et al. [S]. The suspension was centrifuged at 8OOg for 10 min in a swing-out rotor (MSE Chilspin) and the postnuclear supernatant saved. After resuspending the pellet in 1 ml isolation buffer by 10 triturations with a Pasteur pipet it was recentrifuged and the last step repeated using 0.5 ml buffer. The combined postnuclear supematants were centrifuged at 33,000g for 20 min in a Sorvall OTD 65 centrifuge (6 X 5 ml swing-out rotor); the pellet was resuspended in 2 ml isolation buffer containing 1 mg BSA and layered onto 39 ml of Percoll (Pharmacia Fine Chemicals), starting density 1.05 g/ml. After centrifugation in an angle-head rotor (Sorval SS-34) for 1 hat 45,OOOg, fractions of approximately 1 ml were collected from the bottom of the gradient and assayed for enzyme activities. The bottom 10 fractions were pooled, diluted 1:4, and centrifuged at 33,000g for 20 min in an angle-head rotor to remove Percoll. This step was repeated, and the bottom 4 ml was diluted 1:lO and centrifuged again at ll,OOOg for 20 min in a swing-out rotor (Sorvall SS-37) to obtain a crude lysosomal pellet. Lysosomal membranes were then separated from contaminating mitochondria using a modification of the technique of Ohsumi et al. [9]. The pellet was resuspended in 1 ml buffer A (0.025 M sucrose, 1 mM EDTA, 1 mM dithiothreitol, 1 mM phenylmethylsulphonyl fluoride), left on ice for 30 min, then centrifuged at 11,OOOgfor 20 min (Biofuge). The resulting mitochondrial pellet was resuspended in 0.6 ml buffer B (0.2 M KCl, 0.025 M sucrose, 1 mM EDTA, 1 mM dithio-
INTRODUCTION Lysosomes are membrane-bound organelles of heterogeneous morphology that contain a large number of acid hydrolases and are thus able to degrade a wide range of macromolecules. Their surrounding membranes are able to prevent the entry and exit of high-molecular-weight materials, are resistant to hydrolysis, and help to maintain an acidic intralysosomal pH by means of an ATPdependent proton pump [l]. Lysosomal membranes have a defined structure and contain a number of wellcharacterized components, including specific glycoproteins of 80, 100, and 120 Kda [2,3] that are not present in other intracellular organelles. Many of the enzymes found within lysosomes are glycosylated [4], and a characteristic of tumor cells irrespective of their origin is that they often contain altered glycoproteins [ 5 1. Furthermore, abnormalities in both the synthesis and the secretion of lysosomal enzymes have been observed frequently in many tumor cells [6]. Mouse 3T3 fibroblasts transformed with Simian virus 40 display many of the characteristics of typical tumorderived cells, including loss of contact inhibition of growth [7]. They can be used therefore as a model cell to determine if viral transformation of cells can affect their lysosomal system. In these experiments we measured 93
All
0014~4S27/90 $3.00 Press, Inc. reserved.
Copyright Q 1990 by Academic of reproduction in any form
rights
KUNDRA
94
threitol, 1 mMphenylmethylsulphony1 fluoride) and centrifuged again at ll,OOOg for 20 min. The two supernatant solutions were combined and spun at 90,OOOgfor 1 h in a Beckman DL-100 ultracentrifuge to obtain a pellet containing purified lysosomal membranes, which was stored at -70°C until used. Enzyme assays. Lysosomal enzymes were measured fluorometritally using 4-methylumbelliferyl substrates as described by Olsen et al. [lo]. In addition, acid phosphatase activity was determined using 4-methylumbelliferyl phosphate in 0.2 M citrate buffer, pH 4.8, as a substrate (20 pl sample and 80 ~1 substrate). Cytochrome e was used as a substrate for cytochrome oxidase activity [ll] and sodium benzylamine for monoamine oxidase [9]. Catalase activity was detected using a modification of the procedure of Baudhuin et al. [12], with 0.05 mit4 HzOz in 10 mM imidazole, 0.1% bovine serum albumin, pH 7.2, as the substrate. Substrate (20 ~1) or buffer without H,O, as a blank was added to 20 ~1 of sample and incubated at room temperature for 30 min, and the reaction was stopped with 120 pl of 15% w/v solution of titanium (IV) sulfate (BDH). The absorbance was read immediately at 414 nm. Activity of 5’-nucleotidase was measured by the method of Avruch et al. [13], and galactosyl transferase determined using the method of Abraham [ 141.Protein concentration was assayed using the Biorad protein determination kit [15]. Electrophoresis. Samples of lysosomal membrane were solubilized by boiling them for 5 min in 0.1% SDS, 1% j3-mercaptoethanol, 10 mM Tris, 1 mM EDTA, and 0.005% bromophenol blue, containing 10% sucrose, and electrophoresed on 7.5% acrylamide gels [40] at 25 mA/ gel for 16 h. Proteins were visualized using a modification of Heukeshoven’s silver stain [ 161. Acridine orange. Cells (4 x lo4 in 1 ml medium) were plated onto coverslips in Linbro 24-well culture dishes and left overnight to monolayer. After a wash with PBS, acridine orange (1 pg/ml in medium) was added, and the cells were incubated for 2 h at 37°C and washed three times in PBS, and the coverslips were mounted onto slides. Fluorescence was observed using a Nikon Diaphot microscope fitted with a blue filter. Pun&ate orange fluorescent vesicles were counted in a single visual plane of individual, nonmitotic cells using a Xl00 objective with a focal depth of 0.6 pM. Total orange fluorescence was determined using a Zeiss microfluorometer fitted with a Bp 436-5 filter to measure at wavelengths between 460 and 475 nm and a X40 objective with focal depth of 1.0 pm. Lysosomal enzyme assays. Cells were plated out at a concentration of 5 X lo6 in 2 ml medium in 6 X 35-mm Linbro culture dishes and allowed to monolayer overnight. The medium was then replaced with an additional 2 ml of fresh medium, which was collected after a 24-h period of incubation. The cells were washed three times with PBS, lysed by freezing and thawing them in 0.5 ml of 0.1% Triton X-100, and centrifuged at ll,OOOg for 30 s. Both media and cell extracts were assayed for lysosomal enzyme activities. Into each well of 6 X 35-mm Uptake of ‘=I-Polyvinylpyrrolidone. Linbro culture dishes, 5 X ld cells were seeded in 2 ml Hams F12 medium and left overnight. They were then washed with PBS and incubated in fresh serum-free medium containing 6 X 10” cpm 1261-polyvinylpyrrolidone per milliliter. After an appropriate period of incubation, the cells were washed three times in PBS, lysed with 0.5 ml of 0.1% Triton X-100, and freeze-thawed. The suspension was centrifuged to remove insoluble material and the radioactivity in 0.25 ml of the supernatant solution was counted in a LKB compugamma counter. RESULTS
Lysosomal Enzyme Activity The intracellular activity of all of the enzymes tested was greater in 3T3 cells than in SV3T3 cells (Table 1) with activity ratios ranging from 1.53 for @-glucuronidase to 14.05 for &galactosidase. To determine whether
AND DEAN
these differences were due to an increased rate of secretion by SV3T3 cells, resulting in a reduction in intracellular levels, the activity of enzymes released during a 24h period of incubation was measured. Only in the case of /I-hexosaminidase was activity in medium from SV3T3 cells slightly higher than that from 3T3 cells (1.28 times greater). There was no significant difference in the activity of ,&glucuronidase, while secreted acid phosphatase and @galactosidase activities were 2.2 and 7.0-fold higher in medium collected from 3T3 cells than in that
from SV3T3 cells. @-glucosi&se me&um from either type of ce]]
wag not &&e&d
in the
Although in general larger amounts of enzyme activity were released by 3T3 cells than by SV3T3 cells, secreted enzyme as a fraction of total activity (secreted plus intrace]]u]~)
was
approximately
twice
as high
in
SV3T3
fibroblasts. Conversely, intracellular activity as a proportion of total activity was greater in 3T3 cells. Although SV3T3 cells did secrete a greater proportion of their total activity than did 3T3 cells, the amounts secreted were not sufficiently large to account for the lower ] evels of activity present intracellularly.
Arcridine OrUnge Acridine orange was used as a marker with which to determine the number of low-pH vesicles present in each type of cell. After uptake, it was seen as a punctate, bright orange fluorescence, concentrated in low-pH vesicles generally perinuclear in distribution in 3T3 cells but more peripheral in SV3T3 cells. Fluorescence became optimal after a 2-h incubation. The single visual plane that resolved the greatest number of vesicles was chosen for each cell and counting completed quickly before overexposure to ultraviolet light induced rupture of the vesicles and quenched their fluorescence. We assumed that the low-pH vesicles in each visual plane were randomly distributed and that determining the ratio of 3T3: SV3T3 vesicles minimized error since every plane should contain the same ratio. Frequency histograms of vesicle numbers showed a bell-shaped distribution of low-pH vesicles in both cell types (Fig. 1). Although there was some degree of overlap between the two, the mean number of vesicles in 3T3 cells was higher than that in SV3T3 cells by a factor of 1.6 (P < 0.001, n = 124) as shown in Table 2. Overall orange fluorescence was measured in addition to vesicle number to exclude the possibility that vesicles too small to be resolved had been omitted during counting. The microfluorometer was blanked using dead cells which fluoresced green due to staining of their nucleic acids with acridine orange. The same criteria as those employed for vesicle counting were used to select cells. When frequency histograms of arbitrary fluorescence units per cell were plotted (Fig. 2), they showed that 3T3 cells had a fluorescence greater than that of SV3T3 cells (n = 76, P < 0.01). Since this semiquantitative method
LYSOSOMAL
ENZYMES
IN SV3T3
TABLE Comparison
Intracellular activities fi-glucuronidase fl-galactosidase &hexosaminidase &glucosidase Acid phosphatase
27.41* 86.14 f 360.12 + 20.74+ 259.1 +
Secreted activities @-glucuronidase @-galactosidase @hexosaminidase @-glucosidase Acid phosphatase
0.88 5.71 5.36 1.88 12.86
1.99 + 1.83 + 36.03 + nd 10.65 z!z
0.28 0.41 3.07 3.52
95
3T3 CELLS
1
of Enzyme Activities 3T3 cells (1)
AND
in 3T3 and SV-3T3 Cells Ratio (1:2)
SV-3T3 cells (2)
17.93 6.13 126.02 6.98 50.51
f 1.43 + 0.38 k 4.52 f 0.51 z!z5.89
1.53 14.05 2.86 2.97 5.12
2.27 2 0.32 0.26 2 0.13 46.23 f 7.3 nd 4.5 iI 0.59
0.87 7.0 0.78 2.37
Note. Medium was collected from cell monolayers after 24 h of culture; enzyme activities were measured and expressed as &/h/ml of medium. The remaining cells were washed and lysed in 0.1% Triton X-100, and their activity was expressed as r&f/h/ml of solution. All values represent the means of eight separate experiments.
measured total orange fluorescence, it showed not only that the number of low-pH vesicles in each cell type differed, but also that their area and hence their probable volume were also different. Uptake of ‘251-Polyvinylpyrrolidone
Since uptake of ‘25-I-PVP is not receptor-mediated and it is not broken down within cells, it can be used to determine the amount of liquid taken up by fluid-phase pinocytosis [ 171and thus provide a measure of lysosomal volume when this has reached equilibrium. lz51-PVP is known to accumulate within lysosomes [ 18-201 and does not affect the rate of fluid-phase pinocytosis [ 171. It has been used previously to determine the volume of the endocytic compartment in cells [21].
In our experiments the initial rates of uptake of “‘IPVP into both 3T3 and SV3T3 cells were equivalent (Fig. 3), suggesting that the rate of fluid-phase pinocytosis was the same for each cell type. However, after 12 h of uptake the amount of ‘251-PVP present in 3T3 cells was almost double that found in SV3T3 cells. When cells were allowed to take up this labeled material for 24 h the amount of radioactivity present within both types of cell remained constant between the 12- and 24-h time points, indicating that their respective endocytic compartments consisting primarily of lysosomes had been saturated. At this equilibrium the ratio of 3T3 to SV3T3 lysosomal volume was calculated to be 1.9 f 0.3 (Table 2). Lysosomal Surface Area
In order to compare the lysosomal surface areas of 3T3 cells and SV3T3 cells, lysosomal shape was assumed to be spherical. The formula TABLE
2
Physical Characteristics of Lysosomes from 3T3 and SV40-3T3 Cells Ratio 3T3/SV40-3T3
20
40
6G
*o
100
120
140
160
180
200
220~240
numberoflysosomes
FIG. 1. Frequency distribution of fluorescent vesicles. After 2 h incubation with acridine orange coverslips were washed and mounted on slides, and the number of pun&ate, orange fluorescent vesicles in each cell was counted. Figure represents the results obtained from 124 separate 3T3 cells (solid bars) and from 124 SV3T3 cells (crosshatched bars).
Volume (iz51 uptake/24 h) Number/visual plane/cell Surface area/cell
1.9 + 0.3 (4) 1.6 + 0.2 (124) 1.8 + 0.2
Note. Ratios were calculated from the number of separate experiments or number of cells observed, given in parentheses and expressed as a mean value *SD. Cells were allowed to phagocytose ‘%I-polyvinylpyrrolidone until cell-associated radioactivity reached a constant value after which the lysosomal compartment was considered to be
saturated. Acridine orange was ueed as a marker with which to count the number of lysosomes per cell and lysosomes were assumed to be spherical when membrane surface area was calculated.
96
KUNDRA
20
40
60
80
100
120
140
160
180
200
220
240
260
Fluorescence (UnitS)
FIG. 2. Total fluorescence per cell. After uptake of acridine orange the overall fluorescence per cell was measured using a microdensitometer as described in the text and expressed as arbitrary units of fluorescence. Numbers represent the values recorded for 124 separate 3T3 cells (solid bars) and for 124 SV3T3 cells (cross-hatched bars).
S, 4X7rXrfXN, -= S2 4 X T X r; X N2 was used where r is the half-diameter of each lysosome, S is the lysosomal surface area/cell, V is the lysosomal volume/cell (ml), and N is the lysosomal number/visual plane/cell. If r is the average half-diameter of lysosomes then 113
(V,/(+T X N&)2’3 X Nl “* ”
= (v,/( 2~ X N2))2’3 x N2 ’
AND
DEAN
ent (Fig. 4) maximal enzyme activity was found at a density of 1.06-1.07 g/ml. Similar distribution patterns (data not shown) were obtained for the other lysosomal enzymes, /?-galactosidase, &glucuronidase, and &glucosidase. Lysosomes were well separated from the plasma membrane marker 5’-nucleotidase and from the golgi complex marker galactosyl transferase, both of which were present primarily in the top third of the gradient. Catalase activity formed a broad peak in the middle of the gradient with a maximum at a density of 1.045 g/ml. Mitochondrial cytochrome oxidase activity was more buoyant than that of lysosomes, peaking at a density of 1.055 g/ml, although almost 50% of the activity did overlap the lysosomal fractions. Lysosomal membranes were separated from contaminating mitochondria by differential centrifugation and lysed using buffers A and B (see Methods). Almost 90% of the inner mitochondrial membrane marker, cytochrome oxidase, and 73% of monoamine oxidase, an outer mitochondrial membrane marker, were present in the final pellet, with 70% of the acid phosphatase and fihexosaminidase activities remaining in the supernatant solution (Table 3). Purified lysosomal membranes were isolated from this supernatant by a final centrifugation at 90,OOOg. After electrophoresis on 7.5% acrylamide gels the pattern of components resolved from these membranes was clearly different from those obtained for mitochondrial and endosomal fractions (data not shown). In addition a number of differences between lysosomal membranes from both types of cell were distinguishable (Fig. 5). In particular a number of components with apparent molecular weights of 110,51,40, and 33 Kda present in 3T3 cells were absent from SV3T3 cells. These bands were not detected in the lysosomal membranes purified from SV3T3 cells during three separate experiments. DISCUSSION
This reduces to S1 _ V12’3x Nl”3 s,- V22/3x N 2 113This formula takes into account the ratio of both lysosomal number and volume in order to compare total lysosomal surface area per cell. Calculation showed that the area per cell was 1.8 + 0.2-fold greater in 3T3 cells than in SV3T3 cells (Table 2).
The biochemical and physical properties of lysosomes from these two types of 3T3 cell showed considerable
24r
Lysosomal Membranes Both @-hexosaminidase and acid phosphatase were used as markers for lysosomes during purification. Almost 65% of the acid phosphatase activity and 80% of the /3-hexosaminidase activity from each cell type were found in the first 10 ml of the gradient. After centrifugation through a l.O&g/ml starting-density Percoll gradi-
I 0
1
a
16
24
time (hr)
FIG. 3. Uptake of ‘251-labeled polyvinylpyrrolidone. The rate of accumulation of radiolabeled polyvinylpyrrolidone with time was measured in 3T3 (0) and SV3T3 (m) cells over a period of 24 h. Data represent the means + standard error for four separate experiments.
LYSOSOMAL
ENZYMES
IN SV3T3
AND
97
3T3 CELLS
l.osr A
volume (ml)
volume (ml)
FIG. 4. Purification of lysosomes. Cells were disrupted by nitrogen cavitation and fractionated g/ml, as described in the text: (A) 3T3 cells; (B) SV3T3 cells. Fractions of 1 ml were collected nucleotidase (6), cytochrome oxidase (m), and fi-galactosyl transferase (A).
differences. Intracellular enzyme activity, lysosomal number, volume, and lysosomal membrane surface area were all greater in 3T3 cells than in SV3T3 cells. In addition a number of proteins present in the lysosomal membrane of 3T3 cells were absent from SV3T3 cells. In a previous study, Bosmann [22] reported that both P-glucosidase and /3-galactosidase activities were greater in SV40-transformed than in normal 3T3 cells. However, their cells had been cultured in Eagle’s medium supplemented with 10% serum and ascorbic acid. Our enzyme activities were measured fluorometrically using freshly disrupted cells grown in Hams F-12 medium containing heat-inactivated serum. Both cell types grew better in Hams F-12 than in modified Eagle’s medium. Furthermore, the samples analyzed by Bosman were extracted for 16 h at 4”C, lyophilized, rehydrated, and stored frozen before assay. This long extraction time might inactivate some enzymes. In contrast, Orkin et al. TABLE
on a Percoll gradient, starting density 1.05 and assayed for &bexosaminidase (O), 5’-
[23] found that intracellular activities of both &glucuronidase and P-hexosaminidase were higher in 3T3 than in SV3T3 cells. However, they did not find significant differences in the activity of secreted p-hexosaminidase although they did use a less sensitive spectrophotometric method of enzyme assay, and their cells had been grown in serum-free medium. Although we found that secreted enzyme activities were higher in 3T3 cells, SV3T3 cells secreted a much greater proportion of their total activity than they stored in lysosomes. Other similar studies using Novikoff hepatoma cells and cells from animals bearing this tumor have also shown that their acid phosphatase activity and their lysosomal number were reduced [24]. Secretion of a lower proportion of their lysosomal enzymes, however, could not by itself account for the higher activity we observed in 3T3 cells. When acridine orange was used as a marker to determine the number of lysosomes [2,25] we 3
Purification of Lysosomal Membranes Fraction
Hexosaminidase
Acid phosnhatase
Cytochrome oxidase
Monoamine oxidase
SV40-3T3 cells Crude lysosomal fraction ll,OOOg spin supernatant ll,OOOg spin pellet Membrane supernatant Membrane pellet
100 77 23 38 37
100 73 27 24 50
100 10 90 3 7
100 27 73 3 24
3T3 cells Crude lysosomal fraction ll,OOOg spin supernatant ll,OOOg spin pellet Membrane supematant Membrane pellet
100 73 27 38 35
100 70 30 19 51
100 12 88 4 8
100 27 73 2 25
Note. The lysosomes obtained by fractionation of Percoll were lysed and centrifuged as described in the text. Enzyme measurements were made at each stage of the purification to confirm that lysosomal membranes were free of mitochondrial contamination. Lysosomal enzyme activities show mean percentage recoveries from six separate experiments, cytochrome and monoamine oxidases means from three separate experiments.
98
KUNDRA
Mr x1o-3 ~ -116
'-26
FIG. 6. Gel electrophoresis of lysosomal membranes. Lane 1 shows a representative sample of solubilized membranes from 3T3 cells and lane 2 a sample from SV3T3 cells stained with silver. The position of molecular weight markers is shown alongside for comparison. Protein components present in 3T3 cells but not in SV3T3 cells are indicated by arrows.
found this was 1.6 times higher in 3T3 cells than in SV3T3 cells. A difference of this magnitude could account for the increase of 1.53-fold in the /3-glucuronidase activity of 3T3 cells, since this enzyme was secreted in approximately equal amounts by both cell types. It would not, however, account for the differences in other lysosomal enzymes, which were more than 1.6 fold greater in 3T3 cells. It has been argued previously that increased secretion of lysosomal enzymes may lead eventually to alteration of plasma membrane glycoproteins and hence to uncontrolled growth [26, 271. Mitosis is associated with rearrangement and decrease in lysosomal number [28] and to avoid any effect this may have had in our observations, all measurements of the number of low-pH vesicles were carried out using individual, nonmitotic cells. Low-pH vesicles were observed in cytoplasmic projections and at the periphery of SV3T3 cells but were mainly perinuclear in 3T3 cells. A similar peripheral distribution of low-pH vesicles has been observed previously up to 10 days after nonlytic SV40 virus infection of 3T3 cells [29]. Because of the uneven distribution of these vesicles and the possibility that some may have been too small to observe during counting, total acridine orange fluorescence was also measured. 3T3 cells showed a greater degree of fluorescence than SV3T3 cells, which confirmed the data obtained from vesicle counting and also indicated that the area of low-pH vesicles was also greater in 3T3 cells. Although the volume of the lysosomal compartment in 3T3 cells was almost twice that of SV3T3 cells when measured by uptake of ‘251-polyvinylpyrrolidone, this difference was less than the differences in activity for all enzymes tested except b-glucuronidase. The slightly reduced rate of secretion of &hexosaminidase (1.26 fold) and the greater lysosomal volume (1.9-fold) of 3T3
AND DEAN
cells could, by themselves, account for the increase in intracellular activity seen in this enzyme. Neither secretion nor simple physical properties, such as lysosomal volume and number, could, however, account for the large differences in activity of &galactosidase and acid phosphatase. The higher activities of these enzymes in 3T3 cells may have been due to an increased rate in synthesis and/or greater substrate specificity since both stored and secreted enzyme activities were elevated. The lack of @-glucosidase activity in medium collected from both types of cell can be explained by the fact that this enzyme is normally tightly bound to lysosomal membranes [30]. A major proportion of leukocyte acid phosphatase and 80% of hepatocyte acid phosphatase is also found associated with lysosomal membranes [9, 301. In our preparations of purified lysosomal membranes, 70% of the acid phosphatase was bound. The ratios of lysosomal volume and number in 3T3 cells and SV3T3 cells were used to derive the ratio of total lysosomal membrane surface area per cell (Table 2), but this ratio was considerably lower that those calculated for @-glucosidaseand acid phosphatase activities (Table 1). Thus, the relative activity of these two enzymes does not appear to be dependent solely on lysosomal surface area. It is possible that there was a greater concentration of these enzymes in membranes from 3T3 cells or that the turnover number of each enzyme molecule was greater. Alternatively, our estimate of the ratio of surface area may be conservative because lysosomes are of various sizes and do not always approximate a sphere [31], while our calculation assumed that all lysosomes were spherical and had the same radius. Lysosomes were purified using Percoll at 1.05 g/ml because denser gradients, such as those used previously to isolate lysosomes from NIH 3T3 cells [32] and Balb/c 3T3 cells [33], did not separate our lysosomes completely from other organelles. A number of distinct membrane components with molecular weights of lOO110 and 120 kDa [2, 3, 34, 351 have been described in other lysosomal preparations, and different types of cell have been shown to contain lysosomal components of slightly differing sizes [34]. Lysosomal membranes from our 3T3 cells contained a number of predominent proteins which were totally absent from SV3T3 cells. This finding, along with the dissimilarity we observed in Bglucosidase and acid phosphatase activities, showed that not only lysosomal contents but also the composition of their membranes differed significantly between 3T3 and SV3T3 cells. It is possible, therefore, that these observations are related and that alterations in membrane proteins may affect the anchorage or retention of lysosomal enzymes. There have been a number of reports indicating that lysosomal stabilizing agents can retard certain types of cancer [35,36] and that agents which stabilize lysosomal membranes can also inhibit cell division [37]. With cessation of growth, 3T3 cells display a twofold increase in
99
LYSOSOMAL ENZYMES IN SV3T3 AND 3T3 CELLS the number of early autophagic vacuoles and an increase in their protease activity [38]. Wharton et al. [39] suggested that the increase in lysosomal enzymes associated with cell aging could be due to the increase in population doubling time of aged cells, which in turn may cause lysosomes to accumulate. Mitosis is associated with lysosome rearrangement and loss, and 3T3 cells transformed with SV40 virus undergo mitosis more frequently than normal 3T3 cells [28]. Therefore, the decreased enzyme activity, lysosome number, volume, and membrane surface area seen in SV3T3 cells may all result from the greater rate of division associated with their transformed phenotype. Alternatively, transformed cells frequently have reduced growth requirements, and a reduction in the need for external nutrients may in turn be reflected in a smaller lysosomal compartment. The authors gratefully acknowledge the help given by Dr. Helen Muir and Sister Helen Christensen and wish to thank Dr. J. Chayen and Dr. L. Bitensky for use of their microfluorometer. We also acknowledge the helpful discussion with Drs. Fojo, Keefer, Graham, Cunningham Dobler, and Swanson. This work was supported by Fulbright-Haves Grant PL87-256 awarded to V.K.
14. 15.
19.
Heukeshoven, J., and Dernick, R. (1985) Ekctrophresis 6,103112. Williams, K. E., Kidston, E. M., Beck, F., and Lloyd, J. B. (1975) J. Cell Biol. 64,113-122. Allison, A. C., and Black, P. H. (1967) J. Natl. Cancer Inst. 39, 775-787. Meijer, A. E. F. H., and Willinghagen, R. G. J. (1963) Biochem.
20.
England, I. G., Naess, L., Bromhoff,
16. 17. 18.
Pharmncol. 12,973-980.
1. 2. 3. 4.
5.
Ohkuma, S., and Poole, B. (1978) Proc. Natl. Acad. Sci. USA 76, 3327-3331. Reggio, H., Bainton, D., Harms, E., Coudrier, E., and Louvard, D. (1984) J. Cell Biol. 99,1511-1526. Lewis, V., Green, S. A., Marsh, M., Vihko, P., Helenius, A., and Mellman, I. (1985) Cell Bial. 100,1839-1847. Dean, R. T. (1975) in Lysosomes in Biology and Pathology (Dingle, J. T., and Dean, R. T., Eds.), Vol. 4, pp. 349-382, North Holland, Amsterdam. Lenti, L., Lipari, M., Lombardi, D., Zicari, A., Dotta, A., and Pontieri, G. A. (1986) Brit. J. Exp. Pathal. 67,793-799.
6. 7.
&ego, C. M., and Pietras, R. J. (1984) Znt. Reu. Cytol. 88.1-301. Lockwood, T. D., and Shier, W. T. (1977) Nature (London) 267, 252-254.
8.
Rome, L. H., Garvin, A. J., Allieta, M. M., and Neufeld, E. F. (1979) 7,143-153. Ohsumi,Y., Ishikawa, T., andKato, K. (1983) J. Biockm. 93(2), 547-556.
9. 10. 11. 12.
13.
Olsen, I., Dean, M. F., Muir, H., Smith, R., Jenne, B. M., and Hand, C. (1983) Cell B&hem. Fun&. l(2), 103-108. Cooper&in, S. J., and Lazarow, A. (1951) J. Biol. C&m. 189, 665-670. Baudhuin, P., Beaufay, H., Rahman-Li, U., Sellinger, 0. Z., Wattiaux, R., Jacques, P., and DeDuve, C. (1963) Biachem. J. 92, 179-184. Avruch, J., and Wallach, 233,334-347.
D. F. H. (1976) Biochem. Biophys. Acta
Received November 23,1989 Revised version received March 6,199O
R., and Berg, T. (1986) Bio-
them. Pharmacol. 36,201-208. 21.
McKinley,
D. N., and Wiley,
H. S. (1988) J. Cell. Physiol.
136,
389-397. 22. 23. 24. 25. 26. 27.
REFERENCES
Abraham, D., Bou-Gharios, G., Muir, H., Olsen, I., and Smith, R. (1986) Eur. J. Cell Biol. 40,16’7-175. Bradford, M. (1976) Anal. Biochem. 72,243-254.
28.
Bosmann, H. B. (1969) Exp. Cell Res. 64,217-221. Orkin, R. W., Underhill, C. B., and Toole, B. P. (1982) J. Biol. Chem. 267(10), 5821-5826. Urban, J. L., and Unsworth, B. R. (1977) Erperientiu 30,12171219. Moriyana, Y., Takamo, T., and Okhuma, S. (1982) J. Biochem. 92,1333-1336. Henson, P. M. (1976) (Dingle, J. T. and Dean, R. T., Eds.), pp. 99-126, North Holland, Amsterdam. Poole, A. R. (1973) in Lysosomes in Biology and Pathology (Dingle, J. T. Ed.), Vol. 3, pp. 303-337, North Holland, Amsterdam. Sylven, B., Snellman, O., and Strauli, P. (1974) Virchws Arch.
B:Cell Pathol. 17,143-153. 29.
Allison,
A. C., Harington,
J. S., and Birbeck,
M. (1966) J. Exp.
Med. 124,141-154.
32.
Avila, J. L., and Casanova, M. A. (1981) Acta Cien Venez 32,8187. Ahlberg, J., Marzella, L., and Glaumann, H. (1982) Z.&I. Invest. 47(6), 523-532. Gal, S., Willingham, M. C., and Gottesman, M. M. (1985) J. Cell
33.
Merion,
30. 31.
Biol. 100,535-544. M., and Poretz, R. D. (1981) J. Supramol. Struck Cell.
Biochem. 17,337-346. 34. 35.
Chen, J. W., Murphy, T. L., Willingham, M. C., Pastan, I., and August, J. T. (1985) J. Cell BioL 101,85-95. Bagwell, C. E., and Ferguson, W. W. (1980) J. Surg. Res. 28, 410-415.
36.
Gross, J., Azizikhan, R. G., Biswas, C., Bruns, R. R., Hsieh, D. S. T., and Folkman, J. (1981) Proc. Natl. Acad Sci. USA 78, 1176-1180.
37.
Redai, I., and Halmy, 27(l), 41-45.
38.
Papadopoulos, 121.
39.
Wharton, S. A., and Riley, P. A. (1984) in Lysosomes in Biology and Pathology (DingIe, J. T., Dean, R. T., and Sly, W., Eds.), Vol. 7, pp 401-420, North Holland, Amsterdam.
40.
Fairbanks,
M. (1980) Acta Microbial. Acad. Sci. Hung.
T., and Ulrich,
P. (1987) Exp. Cell Res. 171, llO-
G., Steck, T. L., and Wollach,
chemistry 10,2606-2616.
D. F. H. (1971) Bio-
