[CANCER RESEARCH 49, 2722-2728, May 15, 1989]
Association of Lysosomal Activity with Sensitivity and Resistance to Tumor
Necrosis Factor in Murine L929 Cells1
James D. Liddil, Robert T. Dorr,2 and Philip Scuderi
Arizona Cancer Center [J. D. L., R. T. D., P. S.¡,and Departments of Pharmacology and Toxicology [J. D. L., R. T. D.J and Microbiology and Immunology [P. S.J,
University of Arizona, Tucson, Arizona 85724
ABSTRACT
The cytotoxic mechanism of action of tumor necrosis factor (TNF)
was examined using murine L929 fibrosarcoma cells in vitro. Two cell
lines were evaluated: parental TNF sensitive (L929S) (50% cytotoxic
concentration, 2-6 ng/ml); and TNF resistant (L929R) (50% cytotoxic
concentration, > 10,000 ng/ml). The latter resistant cell line was devel
oped by serial passage in increasing concentrations of recombinant human
TNF. Sensitive cells demonstrated cytolytic and cytostatic effects at TNF
concentrations between 2 and 6 ng/ml, respectively. However, TNF failed
to show any selective depression of RNA, DNA, or protein synthesis or
ATP content in these cells until general cell death was apparent, as
defined by the cell rounding and lifting off the plastic surface. The
cytokine also failed to cause DNA single-strand breaks, as detected by
alkaline elution techniques. TNF was also found to be no more active in
glutathione-depleted cells than in target cells containing normal glutathione levels. In contrast, various nonspecific lysosomotropic agents such
as ammonium chloride and D-saccharic acid lactone led to a marked
inhibition of the cytotoxic action of TNF in vitro. Furthermore, significant
differences in lysosomal enzyme activity were noted between L929S and
L929R cells. The changes in L929R cells involved a 50% reduction in
total lysosomal protein levels and a marked depression of /?-glucuronidase
activity. In contrast, L929R lysosomal hexosaminidase activity was sig
nificantly elevated over the L929S cells. From these studies it is con
cluded that the antitumor activity of TNF does not involve specific
inhibition of macromolecular synthesis, ATP production, or the level of
reduced thiols. Instead, TNF cytotoxicity appears to require functional
lysosomes, which are altered when TNF resistance develops in vitro.
INTRODUCTION
In 1975, it was reported that the sera of rodents infected with
Bacillus Calmette-Guérinand subsequently treated with endotoxin contained a substance which caused tumor necrosis in
dependent of endotoxin action (1). This substance was subse
quently named TNF.3 In addition to causing necrosis of trans
planted murine tumors, TNF also inhibits the growth of human
and murine neoplastic cell lines in culture (2). Production of
large quantities of pure human TNF has been made possible by
cloning of the gene sequence encoding TNF (3).
Numerous investigators have described cytotoxic and/or cy
tostatic effects of TNF on tumor cells in culture (1, 2, 4).
However, many tumor cell lines and most normal cells are
resistant to the cytotoxic effects of TNF (4). This resistance to
the effects of the cytokine are not due to a decreased number
of TNF receptors or to the expression of low affinity receptors
(4, 5). Furthermore, the cytotoxic action of TNF is undiminished by the inhibition of DNA, RNA, or protein synthesis
using doxorubicin, dactinomycin, and cycloheximide, respecReceived 7/14/88; revised 12/27/88; accepted 2/7/89.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
' Supported in part by a grant (CA 17094) from the Department of Health
and Human Services. N1H, Bethesda, MD.
2 To whom requests for reprints should be addressed, at Arizona Cancer Center,
Rm. 4947, Tucson, AZ 85724.
1The abbreviations used are: TNF, tumor necrosis factor: rHuTNF-a, recom
binant human tumor necrosis factor-«; NPSH, non-protein sulfhydryl; SSB,
single-strand break; IC50,50% cytotoxic concentration.
lively. Thus, it is reasonable to speculate that TNF does not
induce the synthesis of new proteins which mediate the cytolytic
process. Indeed, increased TNF cytotoxicity is associated with
the inhibition of RNA or protein synthesis, suggesting that cells
can actively block TNF-induced damage by synthesizing specific
new mRNAs.
A number of hypotheses have been forwarded to explain the
mechanism by which TNF cytotoxicity occurs. These have
included (a) the reduction of glutathione levels, (b) free radical
generation followed by lipid peroxidation, and (c) mitochondrial injury and concomitant interference with energy produc
tion (4, 6-8). Recent reports have demonstrated that the cyto
toxic effects of TNF on L929 cells can be inhibited by lysosom
otropic agents such as chloroquine and cytoskeleton-disrupting
agents such as colchicine and cytochalasin B (9, 10). These
findings suggest that the cellular uptake of TNF involves internalization by receptor-mediated endocytosis followed by the
fusion of pinosomes with lysosomes. Further support for the
role of lysosomes in the action of TNF comes from studies
which demonstrate that TNF is rapidly degraded into small
fragments after internalization and that chloroquine inhibits
this process (11). Niitsu et al. (10) have also demonstrated that
direct microinjection of TNF into the cytoplasm or nucleus
fails to cause cell death, suggesting that if TNF does not enter
the cell via the conventional receptor/lysosome route it is
inactive.
Due to the current lack of understanding of the mechanisms
by which tumor necrosis factor mediates its cytotoxic effect, we
performed a number of studies to examine the correlation
between cell death and various biochemical effects in cytokinesensitive and -resistant tumor cells. In addition, lysosomal
activity and inhibition by specific agents was compared in TNFsensitive and -resistant cells using both histochemical and bio
chemical techniques. The results suggest that lysosomes may
play a role in mediating TNF cytotoxicity and resistance. These
findings are discussed in terms of other competing explanations
for TNF-induced cytotoxicity.
MATERIALS
AND METHODS
Cell Lines and Culture. L929S cells (Certified Cell Line 1), a murine
tumorigenic fibroblast line, were obtained from the American Type
Culture Collection (Rockville, MD). L929 cells were grown as adherent
monolayers on plasticware (Costar, Cambridge, MA) in RPMI 1640
(Irvine Scientific, Santa Ana, CA) supplemented with 7.5% fetal bovine
serum (M. A. Bioproducts, Inc., Walkersville, MD). The cells were
maintained in a humidified incubator at 37°Cin an atmosphere of 95%
air-5% carbon dioxide. L929R cells resistant to tumor necrosis factor
at a concentration of 1 x IO5 units/ml were produced by continuous
exposure to increasing TNF concentrations over 4 months. Cells were
passaged every 3 days (doubling time, 18 h).
Reagents. rHuTNF-a was the gift of Genentech, Inc. (San Francisco,
CA). rHuTNF-a had an activity of 5 x IO7units/mg, a protein concen
tration of 0.49 mg/ml, and an endotoxin level of <0.123 endotoxin
unit/ml by Limulus amebocyte lysate assay, as specified by the manu
facturer. Culture medium containing fetal bovine serum was used as a
diluent for rHuTNF-a. All other reagents and inhibitors were obtained
2722
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1989 American Association for Cancer Research.
LYSOSOMES AND TUMOR NECROSIS FACTOR
from Sigma Chemical Co. (St. Louis, MO) except where noted. The
lysosomal enzyme inhibitors are listed in Table 1 (12-14).
Cytotoxicity Assay. Test materials were serially diluted in medium
and added to microtiter wells containing 2 x IO4 L929 cells. After 48
h of incubation, the number of living cells was assessed by the Coomassie G250 method (15). The absorbance of protein in each well was
measured at 590 nm using a microtiter plate reader (Dynatech MR600).
Enzyme Inhibition Studies. In studies of the various enzyme inhibi
tors, 2 x IO4 cells were seeded in microtiter wells, along with the
appropriate concentration of inhibitor, and incubated for 24 h. The
wells were then washed and fresh medium, along with the inhibitor and
tumor necrosis factor, was added. After cytokine addition the cells were
incubated at 37°Cfor an additional 48 h. At the end of this incubation
period, the plates were processed for protein determination as previ
ously described. The enzyme inhibitors were used throughout these
studies at noncytotoxic concentrations (>95% survival) as determined
in protein dye cytotoxicity assays.
A significant change in the cytotoxicity induced by TNF alone versus
the cytotoxicity of TNF combined with the various inhibitors was
defined as a 1 logio or greater difference in the respective dose-response
curves. This criterion was used to assure that the differences in cytotox
icity were not due only to variability in the assays.
Macromolecular Synthesis Inhibition Studies. The inhibition of DNA,
RNA, and protein synthesis was investigated using a modification of
the radionuclide substrate incorporation method of Li et al. (16). During
the last hour of a 24-h TNF exposure, 1.0 ¿¿Ci
[3H)thymidine, ['H]uridine, or ['4C]valine (ICN Radiochemicals, Irvine, CA) was added for
the measurement of DNA, RNA, or protein synthesis, respectively.
Vials were counted in a scintillation counter and percentages of control
of [3H]thymidine, ['4C]valine, or [3H]uridine cpm were calculated. As
positive controls, actinomycin D was used for RNA synthesis inhibition
( 1 jjg/ml), cycloheximide for protein synthesis inhibition (5 ¿ig/ml),and
doxorubicin for DNA synthesis inhibition (1 /jg/ml).
ATP Determination. The cellular content of ATP is a reliable indi
cator of general cell viability and of cellular energy metabolism (17).
ATP was measured by a modification of the firefly luciferin-luciferase
reaction (18). ATP content was then determined using the ATP lu
ciferin-luciferase assay with a luminometer (Turner Designs, Mountain
View, CA).
Cytochemical Staining. TNF-sensitive and -resistant L929 cells were
prepared on slides by cytocentrifugation (Shandon Cytospin 2; Pitts
burgh. PA) at 600 rpm for 10 min. The slides were stained for acid
phosphatase and ff-glucuronidase as described previously (19). Cells
(100) were visualized at random using light microscopy and cellular
enzyme content determined based on staining intensity. A staining
intensity score ranging from 3, which indicated the darkest staining
cells, down to 0, which signified no staining was used in these assays.
Alkaline Elution Experiments. The potential for TNF-induced DNA
damage was assessed by quantitation of DNA SSBs as described (20).
Cellular DNA was labeled by logarithmic cell growth in complete
medium containing [2-14C]thymidine (0.1 nCi/ml), followed by growth
tetrapropylammonium hydroxide, which was pumped through the fil
ters at a rate of 2.5 ml/h for 15 h. Fractions were collected hourly for
[I4C]DNA quantitation using scintillation counting.
Lysosomal Purification. Cellular material was kept at 4"( ' throughout
the lysosome isolation procedures and during subsequent enzyme analy
sis. The lysosomal isolation was carried out as previously reported with
some modifications (21). The suspension was centrifugea for 90 min at
40,000 x #., in a Beckman T70.1 rotor. The density of the fractions
was determined using calibrated density marker beads (Pharmacia,
Uppsala, Sweden).
Enzyme Assays. All enzyme activities were determined after solubilization of samples in 0.1% Triton X-100, which did not interfere with
any of the assays. Protein levels were determined spectrophotometrically by the method of Smith et al. (22) using a commercially available
kit (Pierce Chemical Co., Rockford, IL).
For plasma membranes, alkaline phosphodiesterase I was determined
usingp-nitrophenyl-S'-thymidylate
as a substrate (23). Hexosaminidase
(24), acid phosphatase (25), and /i-glucuronidase (26) were used as
lysosomal markers. Succinate-/>-iodonitrotetrazolium violet reducÃ-ase
was used as a marker for mitochondria! enzyme activity (27). For the
Golgi apparatus, a-D-mannosidase II was measured withp-nitrophenyl«-D-mannopyranoside as the substrate (28). All assays were performed
in triplicate and enzyme activities were expressed as milliunits/mg or
units/mg cellular protein.
Colony-forming Assay in Soft Agar. Inhibition of tumor cell colony
formation was determined using the soft agar cloning method of Ham
burger and Salmon (29). All drug exposures were performed in tripli
cate, with the plates being incubated at 37°Cin a humidified incubator
with an atmosphere of 5% COj/air for 7-9 days. At the end of this
time the tumor cell colonies (>60 urn in diameter) were enumerated
using an automated image analysis instrument (FAS II, Omnicon;
Bausch and Lomb, Rochester, NY) (30).
Non-Protein Sulfhydryl Bound Determination. NPSH concentrations
were determined as previously described (31). Results were expressed
as nmol/106 cells.
Statistics. Experiments for determination of enzyme activity in both
whole cells and lysosomal fractions were carried out with triplicate
samples in all instances. Data are expressed as mean ±SD. Differences
in enzyme activity between the fractions for TNF-sensitive and -resist
ant cells were analyzed using a Student t test for paired samples with
significance being designated for P < 0.01.
Dose-response curves for TNF cytotoxicity were analyzed for varia
tion in the percentage of survival using the formula for variance of a
ratio.4
RESULTS
TNF Cytotoxicity Studies. Previous studies have demon
strated the need for continuous exposure to tumor necrosis
factor for >15 h to achieve maximal cytotoxic responses in vitro
(9). Therefore, all cytotoxicity assays were carried out with cells
for 4 h in nonradioactive medium. The DNA was eluted from the filter
continually exposed to TNF throughout the 5-7-day colonywith a solution of 1% sodium dodecyl sulfate adjusted to pH 12.1 with
forming assays or for 48 h for the dye cytotoxicity assays. Fig.
1 compares the dose response curves for TNF against sensitive
Table 1 Lysosomotropic agents
inhibited"Nonspecific
L929 cells using both assays. It is apparent that, with each
InhibitorAmmonium
procedure, the dose-response curve was sigmoidal but relatively
lysosomal*Nonspecific
chlorideChloroquineMonensinVerapamilAntipainBestatinLeupeptinPepstatinD-Saccharic
lysosomalNonspecific
shallow. This is unlike the steeper survival curves produced by
lysosomalNonspecific
most cytotoxic antitumor agents but not unlike those seen for
lysosomalCathepsins
glucocorticosteroids (32). The flat secondary phase of sensitiv
BAminopeptidase
A and
ity may be due to the existence of TNF-resistant subpopulations
leucineaminopeptidaseCathepsin
B.
of cells (33).
BCathepsin
Using either the soft agar cloning method or the Bio-Rad dye
Drf-GlucuronidaseCathepsins
lactoneE-64DexamethasoneEnzyme(s)
acid
assay, a ICso of TNF was found to be in the range of 2-6 ng/
LMembrane B. H, and
ml continuous exposure. The ICso was about 20% lower with
stabilizerRef.131244,454646464646234737
the colony-forming assays. Thus, the colony-forming assay
" Specific enzyme inhibitors defined as interfering with a single enzyme func
tion at a concentration 100-fold less than that which inhibits other enzyme
appears to be a slightly more sensitive method of measuring
activities.
'' An agent which inhibits lysosomal enzymes by raising intralysosomal pH
above 5.
4 D. Slyman. Biostatistics Department, Arizona Cancer Center, personal com
munication.
2723
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1989 American Association for Cancer Research.
LYSOSOMES AND TUMOR NECROSIS FACTOR
or
i
io
DO
1000
TNF CONCENTRAT ION (ng/ml)
Fig. 1. Comparison of L929S cell survival following rHuTNF exposure. Two
methods of analysis were used, a clonogenic assay involving colony formation in
soft agar and a cytotoxicity assay using a Bio-Rad assessment of protein (n = 6
determinations per point; error bars are within symbol size).
.0015
.0010
.0005
100
200
300
400
BOUND(fM,TNF)
500
Fig. 2. TNF binding curves for L929S and L929R cells. The slope and Ka for
the two cell lines were not significantly different.
the response of cells to TNF. This may be due to the ability of
colony formation to measure both cytostatic and cytolytic ef
fects of TNF on the cells (34). Because of the simplicity and
time savings involved in its use, the Bio-Rad dye cytotoxicity
assay was chosen for measurement of cell killing by TNF in
subsequent mechanistic studies.
TNF Resistance. The L929R subclone was produced by serial
passage in increasing concentrations of recombinant human
tumor necrosis factor over several months. This resistant line
showed no response to the cytolytic effects of TNF at doses up
to 10,000 ng/ml, as determined in the dye cytotoxicity assays.
TNF receptor content was found to be equivalent in the sensitive
and resistant lines, with approximately 150 receptors/cell as
determined by Scatchard analysis (35) of radioiodinated TNF
binding (Fig. 2). TNF receptor dissociation constants were not
different in the two L929 cell lines. These ranged from 3.6 to
4.2 x 10~'°M for L929S and L929R cells, respectively.
Macromolecular synthesis experiments were done by expos
ing L929S cells continually to a cytotoxic concentration, 100
ng/ml, of TNF. Fig. 3 illustrates that 5-10 h are required before
any perturbation is seen in the synthesis of DNA, RNA, or
protein or in ATP content. After this initial lag time, there is a
proportional decrease in all parameters of macromolecular
synthesis (Fig. 3). Thus, TNF simultaneously suppresses the
synthesis of each macromolecule, including ATP. There was no
evidence of an early or "selective" inhibition of DNA, RNA, or
protein synthesis or cellular ATP content in the L929S cells
IO 15 20
25
30 35 40
45
50
TIME (MRS)
Fig. 3. Effect of TNF on macromolecular synthesis and ATP content of TNFsensitive L929 cells. Effect of TNF and specific macromolecular synthesis inhib
itors on DNA, RNA, and protein synthesis and ATP content of TNF-sensitive
L929 cells. TNF and inhibitors were present throughout the course of the
experiment.
, data for TNF exposure;
, data for inhibitor exposure. See
Table 2 for details (n = 3 determinations per point).
Table 2 Time course for a 50% decrease in DNA, RNA, and protein synthesis and
ATP content in L929S cells
Time at which 50% decrease occurs (h)°
ParameterDNA
TNF12.5
+
+ specific synthesis in
hibitors (inhibitor concentra
tion)0.5
synthesis
(doxorubicin, 1 pg/ml)
RNA synthesis
19
0.5 (dactinomycin, 1 ¿ig/ml)
Protein synthesis
14
0.5 (cycloheximide, 5 Mg/ml)
ND*
ATP contentL929
16.5L929
" Cells were exposed continuously to 100 ng/ml TNF during the entire exper
iment.
* ND, not determined.
(Table 2). The use of specific macromolecular synthesis inhib
itors (the positive controls actinomycin D, doxorubicin, and
cycloheximide) each caused greater than 90% inhibition of
macromolecular synthesis within 1 h. In contrast, TNF did not
2724
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1989 American Association for Cancer Research.
LYSOSOMES AND TUMOR NECROSIS FACTOR
decrease any one of the parameters in a similar rapid selective
fashion. The use of these positive controls supports the inter
pretation that TNF does not kill cells by specifically inhibiting
macromolecular synthesis.
Effect of Reduced Thiols on TNF. The results of thiol deple
tion experiments showed that the glutathione synthesis inhibi
tor L-buthionine sulfoximine reduced the ICso for TNF in
L929R cells by just over 0.5 logio. This represents only a slight
increase in TNF sensitivity. In addition, the L929S and L929R
cells both contained similar levels of NPSHs, thereby discount
ing a role for NPSH levels in mediating TNF resistance in vitro
(Table 3).
DNA Damage Experiments. L929S cells were exposed for 24
h to 1, 10, 100, and 1000 ng/ml TNF and then the incidence
of SSBs was determined using the method of Kohn et al. (20).
Fig. 4 illustrates the DNA elution pattern from these experi
ments. It is apparent that there is no significant increase in the
rate of DNA elution for the TNF-treated cells versus that of the
controls. Thus, even at a TNF concentration of 1000 ng/ml
(which is cytotoxic to over 90% of the cells), there is no increase
Table 3 Non-protein sulfhydryl group determination of TNF-sensitive and
-resistant L929 cells
NPSH (nmol/10'cells)"
Cell line
TNF-sensitive L929
5.7 ±2.7
TNF-resistant L929
5.3 ±0.7
1Mean of three determinations ±SD on 5 x 10' cells.
0.6
Table 4 Lysosomal enzyme cytochemical staining scores for TNF-sensitive and
-resistant L929 cells
05
Scores represent the total from 100 cells which were assessed for enzyme
content by light microscopy.
CONTROL
TNF Ing/ml
C.4
O
in DNA elution, indicating no significant production of DNA
SSBs.
Cytochemical Staining. The data in Table 4 show that both
L929S and L929R cells have equal staining scores for total
acid phosphatase, traditionally a marker of general lysosomal
activity. However, the sensitive and resistant cells differed sub
stantially in their staining scores for 0-glucuronidase activity.
The L929R cells contained no stainable enzyme, whereas
L929S cells possessed an average staining intensity of 124,
indicating high 0-glucuronidase activity. The results of this
experiment led us to pursue other studies which examine the
role /3-glucuronidase might play in TNF resistance.
Lysosomal Inhibitor Studies. The nonspecific lysosomotropic
agents verapamil, ammonium chloride, and chloroquine each
inhibited the cytolytic activity of TNF to varying degrees (Table
5). Ammonium chloride was the most effective and chloroquine
least effective when used at maximal noncytotoxic concentra
tions. In contrast, various cathepsin inhibitors did not decrease
TNF activity to a similar extent (Table 5). Only D-saccharic
acid lactone, which inhibits 0-glucuronidase, and leupeptin,
which blocks cathepsin B, produced any appreciable change in
TNF sensitivity. This finding supports the view that the lyso
somal enzymes which are important in the action of TNF
probably include /3-glucuronidase but few, if any, cathepsins.
Dexamethasone was also moderately effective in reducing the
toxicity of TNF, presumably due to its membrane stabilization
properties (36).
Lysosomal Enzyme Activity in TNF-sensitive and -resistant
Cells. Table 6 describes the lysosomal enzymes and their relative
activities in both the L929S and L929R cells. In the whole-cell
lysates there was a slightly greater amount of hexosaminidase
activity per mg of protein in the L929R cells (P < 0.01). There
was also a higher level of /3-glucuronidase activity in the cell
lysates from the sensitive cell line: enzyme activity was twice
that in the resistant cell line (P < 0.01). The other lysosomal
enzyme marker, acid phosphatase, was slightly elevated in the
Enzyme score
TNF 10 ngrtnl
TNF lOOng/Vnl
o—o TNF
typeSensitive
Cell
IOOOng/hil
250 cGy
UJ
phos
phatase"300
L929
Resistant L929Acid
300/3-Glucuronidase6124 0
" Substrate for acid phosphatase was naphthol AS-BI phosphoric acid.
* Substrate for /3-glucuronidase was naphthol AS-BI glucuronide. Both reaction
products were visualized using fast red diazonium dye.
0.3
Ul
tr
Table 5 Effect of enzyme and metabolic inhibitors on TNF-mediated cytotoxicity
on L929S cells
oc
(logchange)*-0.6+0.6+6.0+1.3+3.0+0.60.0
TreatmentL-Buthionine
sulfoximineDexamethasoneNH..C1ChloroquineVerapamilLeupeptinAntipainBestatinPepstatinD-Sacch
MMS/lM20
MM10
MM22
pM50
MM50
0.
J
I
I
I
I
I
I
23456789
HOURS OF ELUTION
I
J
10
MM50
MM5
MM5
I
12 13 14
Fig. 4. Effect of TNF on the incidence of DNA single-strand breaks in TNFsensitive L929 cells. Effect of TNF on the incidence of DNA single-strand breaks
in TNF-sensitive L929 cells, as assessed by alkaline elution. Cells were exposed
to TNF for 24 h prior to the start of the experiment. The fraction of 14C-labeled
DNA retained on the filter is plotted against the time of elution.
lactoneE-64MonensinConcentration"10
acid
mM100
1.30.0+0.35
MM1
I1MSensitivity
°Concentrations used were noncytotoxic to cells over the course of the 48-h
assay.
* -, increase in sensitivity; +, decrease in sensitivity; both determined at IC5(,.
A significant change is defined as a 1 log or greater difference in sensitivity.
2725
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1989 American Association for Cancer Research.
LYSOSOMES AND TUMOR NECROSIS FACTOR
Table 6 Activity of enzymes from TNF-sensitive and -resistant L929 cells
Enzyme levels in whole cells
(milliunits/mg protein)"
±0.50*
Hexosaminidase
±0.07
11.71 ±1.52*
/3-Glucuronidase
27.90 ±0.71
2.53 ±0.05*
Acid phosphatase
2.34 ±0.71
Succinate-/>-iodonitrotelrazolium
0.032.51
1.00 +
±0.062.70
1.09
violet reducÃ-ase
±0.04*
a-Mannosidase
±0.02
Alkaline phosphodiesterase
0.24 ±0.01
0.21 ±0.03
Cellular protein (mg/108 cells)L929S5.93
27.5L929R7.29
25.8
* Mean of three determinations ±SD on 10* cells in milliunits of enzyme
activity.
4 P < 0.01 by Student's t test for difference between sensitive and resistant
cells.
Table 7 Cellular enzyme activities of lysosomal fractions from TNF-sensitive and
-resistant L929 cells
Enzyme levels in lysosomal
fraction (milliunits/mg pro
tein)"
±1.46*
+ 0.06
Hexosaminidase
22.52 + 6.20*
ft-Glucuronidase
52.61 + 2.70
3.09 ±0.05*
Acid phosphatase
0.012.45
2.07 ±
0.74
0.39*2.01
±
Succinate-p-iodonitrotetrazolium
0.772.08
±
violet reducÃ-ase
a-Mannosidase
+ 0.00
±0.02
0.06 ±0.01*
Alkaline Phosphodiesterase
0.12 ±0.01
Lysosomal Protein (mg/108 cells)L929S13.03 1.94L929R27.90 0.97
' Mean of Ihree determinations ±SD on IO8cells.
*/" < 0.01 by Studenl's / lest for difference belween sensilive and resistant
cells.
resistant cell line. The enzyme markers for mitochondria (succinate-p-iodonitrotetrazolium
violet reducÃ-ase),Golgi appara
tus (a-mannosidase II), and plasma membrane (alkaline phos
phodiesterase I) did not show any significant differences be
tween the sensitive and resistant cells.
Lysosomal fractions obtained by differential density centrifugation (density, 1.050 g/liter) showed even greater differences
in enzyme activity between the sensitive and resistant subclones
(Table 7). Importantly, the lysosomal fraction obtained from
the L929R cells had twice the level of hexosaminidase activity
as did the sensitive cells but only one-half the /3-glucuronidase
activity when quantitated on a basis of total protein. In contrast,
the lysosomal fraction from the L929S cells contained twice
the level of alkaline phosphodiesterase and over 3 times the
succinate-/Modonitrotetrazolium
violet reducÃ-aseactivily indiealive of mitochondria. The level of a-mannosidase was approx
imately equal for both fractions.
DISCUSSION
It has been suggested lhal TNF may cause a iransienl increase
in RNA synlhesis in cells which are ultimately killed by Ihis
cylokine (37). This speculation has been confirmed in ine
present study, although it is not of stalislical significance. The
currenl findings are unique, however, in ihe comprehensive
evaluation of cellular DNA, RNA, and prolein synlhesis, as
well as ATP conlenl, using highly purified rHuTNF ralher lhan
Ihe crude serum supernalanls used in earlier mechanistic studies
(37).
In Ihe experimenls reported here, TNF failed lo fil Ihe
classical pallern of a specific inhibilor of macromolecular syn
lhesis which blocks produclion of a single class of macromolecules. TNF also failed lo produce cyloloxic effecls over short
exposure periods (<1 h), which is characlerislic of such synlhe
sis inhibilors. In addilion, TNF did noi inhibil oxidalive phosphorylalion or olher cellular processes which lead lo ATP
produclion. These findings serve lo conlrasl TNF wilh Ihe
classical inhibilors, such as 2,4-dinilrophenol, which produce a
reduclion in cellular ATP conlenl wilhin l h of exposure. Wilh
TNF exposure, there was only a slow loss of energy, which
paralleled Ihe general decrease in cell viabilily. Thus, in Ihe
case of TNF, cellular ATP conlenl appears lo simply be anolher
indicalor of cell viabilily.
Colleclively, Ihese dala confirm previous observalions show
ing lhal TNF does noi direclly reduce DNA, RNA, or prolein
synthesis (38). The current results also suggest lhal a hypolhesis
lhal Ihe primary mechanism of aclion of TNF involves a
dislurbance in cellular energy metabolism (6) is incorrect
II has also been postulated lhal TNF could cause mobilizalion
of polyunsaluraled fatty acids and subsequently lead lo lipid
peroxidalion, which would influence free radical produclion
and cause cell dealh (39). Olher researchers have argued lhal
cells with lower levels of glulalhione would have decreased
abilities lo scavenge free radicals and would, Iherefore, be more
sensilive lo Ihe cylotoxic aclions of TNF. This laller speculation
suggesled lhal cells selecled for resistance lo TNF mighl have
higher levels of reduced ihiols than Iheir sensilive counterparts.
However, Ihe currenl sludies of NPSH levels in TNF-sensilive and -resistanl L929 cells show lhal bolh cell lines do noi
significantly differ in NPSH conlenl. Furthermore, L-bulhionine sulfoximine (which depletes Ihe level of reduced Ihiols)
failed lo significanlly augmenl TNF sensilivily in normal
L929S cells and il did noi reverse Ihe resislance of TNFresislanl L929R cells. These observalions suggesl thai Ihe
mechanism of aclion of TNF is noi dependenl on Ihe amounl
of reduced Ihiols present in largel cells.
The possibilily lhat TNF caused dealh by DNA slrand break
age had been suggesled by several groups (40, 41), although
ihere is lillle evidence lo support Ihis hypolhesis. A previous
reporl demonstrated thai lympholoxin, a cylokine wilh 30%
homology wilh TNF, caused fragment at ion of largel cell DNA
(41). Due lo Ihe partial identity with TNF, Ihe dala of lympho
loxin suggesl lhal Ihe cyloloxic mechanism of action of TNF
mighl similarly involve damage lo nuclear DNA (40). Our
resulls, however, show lhal exposing largel cells to high concenlralions of TNF over a period of lime sufficient for adequate
inlernalizalion and cyloloxicily does noi produce significant
DNA SSBs. These resulls, however, do no exclude the unlikely
possibilily lhal olher lypes of DNA lesions such as DNA-DNA
or DNA-prolein cross-linking may lead lo cell dealh.
Various nonspecific lysosomolropic agents are known to
reduce inlralysosomal pH and each agenl blocked Ihe cyloloxic
aclion of TNF by 1 log or greater. Of Ihese, ammonium chloride
produced Ihe mosl substantial inhibilion of TNF cylotoxicity.
Surprisingly, verapamil, which concenlrales in lysosomes (42),
also decreased Ihe cyloloxicily of TNF by 3 logs. This was noi
enlirely expected, allhough some of Ihis effecl may be due lo
Ihe general membrane-altering properties of verapamil (14).
Chloroquine was leasl effeclive al blocking TNF cyloloxicity,
possibilily due lo ils inherenl cyloloxicily in L929S cells, which
obviated ihe use of high concenlralions. Similarly, ihe carboxylic sodium ionophore monensin only weakly blocked TNF
cyloloxicily, possibly due again lo ils inherenl cyloloxic aclivily,
which precluded lesling high concenlralions.
In contrasl lo Ihe effect of nonspecific lysosomolropic agenls
on ihe cyloloxic aclion of TNF, several specific lysosomal
enzyme inhibilors did noi significantly reduce cytokine aclivily.
Similar resulls for Ihe effect of leupeplin on TNF aclivily have
2726
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1989 American Association for Cancer Research.
LVSOSOMES AND TUMOR NECROSIS FACTOR
also been reported (9). Of interest is the observation that Dsaccharic acid-l,4-lactone produced the greatest (1.3-fold) in
hibition of the cytotoxic acid of TNF for a specific enzyme
inhibitor. This is intriguing, considering the fact that this agent
inhibits 0-glucuronidase and that the TNF-resistant L929R
cells had significantly lower levels of this lysosomal enzyme.
/3-Glucuronidase cleaves nonreducing terminal 0-glucuronosyl residues from glycosaminoglycans and conjugated steroids,
drugs, and other xenobiotics (43). One possible explanation for
the decrease in 0-glucuronidase in the TNF-resistant cells is
that the enzyme is involved in membrane digestion (43). Thus,
with a lower amount of enzyme activity, cells may be protected
from lysosomal autodigestion of their own membranes as a
result of TNF exposure.
In order to further characterize the differences between the
L929S and L929R cells, experiments were performed to com
pare various enzymes in both whole cells and isolated lysosomes. Such purified organelle preparations are free of cellular
protein inhibitors of lytic enzyme activity. Thus, they have less
protein per unit of enzyme activity than do whole cells. The
level of 0-glucuronidase activity was found to be decreased in
L929R cells in both whole cells and isolated lysosomes when
compared to the sensitive line. In contrast, both hexosaminidase
and acid phosphatase levels were significantly elevated in the
L929R lysosomal fraction, as compared to the sensitive cell
fraction (Tables 6 and 7). It is tempting to speculate that the
TNF-associated increase in hexosaminidase and the decrease
in 0-glucuronidase in resistant cells facilitates a protective effect
against TNF-mediated cytotoxicity by enhancing TNF degra
dation and/or blocking intralysosomal processing. TNF may
be endocytosed by cells and taken up into the lysosomes, where
it may cause a decrease in lysosomal pH. This could ultimately
lead to cell death and TNF-resistant cells may resist the uptake
of TNF into the lysosome. Alternatively, TNF may inhibit
lysosomal functions which detoxify TNF and resistant cells
may have these in an overabundance.
In summary, the current studies have shown that TNF does
not specifically inhibit RNA, DNA, protein synthesis content
or ATP in tumor cells. It was also shown that reduced thiol
levels play no role in mediating the cytotoxic actions of TNF.
A further examination of the effect of TNF on DNA structure,
as assessed by single-strand break formation, failed to show any
correlations. The data show that there are consistent enzymatic
differences between TNF-sensitive and -resistant cells and that
these differences were reflected primarily in lysosomal enzymes.
Lysosomotropic agents, both specific and nonspecific, can in
hibit the cytotoxic actions of TNF in vitro and support the
theory that lysosomes are involved in TNF cytotoxicity. Over
all, these observations suggest that TNF requires interactions
with cytoplasmic targets (probably the lysosomes). The fact that
agents which block TNF cytotoxicity in this study are known
to reverse multidrug resistance is an intriguing area for further
study.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
REFERENCES
31.
1. Carswell, E. A., Old, L. J., Kasse!, R. C, Green, S., Fiore, N., and William
son, B. An endotoxin-induced serum factor that causes necrosis of tumors.
Proc. Nati. Acad. Sci. USA, 72: 3666-3670, 1975.
2. Fiore, N., Green, S., Williamson, B.. Carswell, E. A., Old L. J., and Hlinka,
J. Tumor necrosis factor: further studies. Proc. Am. Assoc. Cancer Res., 16:
125, 1975.
3. Wang, A. M., Creasey, A. A., Ladner, M. B., Lin, L. S., Strickler, J., Van
Arsdell, J. N., Yamamoto, R., and Mark, D. F. Molecular cloning of the
complementary DNA from human tumor necrosis factor. Science (Wash.
DC), 228: 149-154, 1985.
4. Sugarman, B. J., Aggarwal, B. B., Hass, P. E., Figari, I. S., Palladino, M.
32.
33.
34.
A., and Shepard, H. M. Recombinant human tumor necrosis factor-«:effects
on proliferation of normal and transformed cells in vitro. Science (Wash.
DC), 230: 943-945, 1985.
Tsujimoto, M., Yip, Y. K., and Vilcek, J. Interferon-i enhances expression
of cellular receptors for tumor necrosis factor. J. Immunol. ¡36:2441-2444,
1986.
Jones. G. R. N. Free radicals in immunological killing: the case of tumour
necrotising factor (TNF). Med. Hypnoanal., 21: 267-271, 1986.
Dealtry, G. B., Naylor, M. S., Fiers, W., and Balkwill, F. R. DNA fragmen
tation and cytotoxicity caused by tumor necrosis factor is enhanced by
interferon-7. Eur. J. Immunol., 17:689-693, 1987.
Maury, C. P. J. Tumour necrosis factor—an overview. Acta Med. Scand.,
220:387-394,1986.
Kull, F. C., Jr., and Cuatrecasas, P. Possible requirement of internalization
in the mechanism of in vitro cytotoxicity in tumor necrosis serum. Cancer
Res., 47:4885-4890, 1981.
Niitsu, Y., Watanabe, N., Soné,
H., Neda, H., Yamauchi, N., and Urushizaki,
I. Mechanism of the cytotoxic effect of tumor necrosis factor. Jpn. J. Cancer
Res., 76: 1193-1197, 1985.
Tsujimoto, M., Yip, Y. K., and Vilcek, J. Tumor necrosis factor: specific
binding and internalization in sensitive and resistant cells. Proc. Nati. Acad.
Sci. USA, 82: 7626-7630, 1985.
Gonzalez-Noriega, A., Grubb, J., Talkad, V., and Sly, W. S. Chloroquine
inhibits lysosomal enzyme pinocytosis and enhances lysosomal enzyme se
cretion by impairing receptor recycling. J. Cell Biol., 85: 839-852, 1980.
de Duve, C., de Barsy, T., Poole, B., Trouet, A., Tulke, P., and Van Hoff, F.
Lysosomotropic agents. Biochem. Pharmacol. 23: 2495-2531, 1974.
Akiyama, S., Gottesman, M. M., Hanover, J. A., Fitzgerald, D. J. P., and
Pastan, I. Verapamil enhances the toxicity of conjugates of epidermal growth
factor with Pseudomonas exotoxin and antitransferrin receptor with Pseudomonas exotoxin. J. Cell. Physiol., 120: 271-279, 1984.
Bearden, J. C. Quantitation of submicrogram quantities of protein by an
improved protein-dye binding assay. Him him. Biophys. Acta, 533: 525-529,
1978.
Li, L. H., Timmins, L. G., Wallace, T. L., Knieger, W. C., Prairie, M. D.,
and Im, W. B. Mechanism of action of didemnin B, a depsipeptide from the
sea. Cancer Lett., 23: 279-288, 1984.
Strehler, B. L., and Totter, J. R. Firefly luminescence in the study of energy
transfer mechanisms. I. Substrate and enzyme determination. Arch. Biochem.
Biophys., 40: 28-41, 1952.
Lundin, A., and Thore, A. Analytical information obtainable by evaluation
of the time course of firefly bioluminescence in the assay of ATP. Anal.
Biochem., 66:47-63, 1975.
Culling, C. F. A. Enzymes. In: C. F. A. Culling, R. T. Allison, and W. T.
Barr (eds.), Cellular Pathology Technique, pp. 300-330. London: Butterworths, 1985.
Kohn, K. W., Erickson, L. C., Ewig, R. A. S., and Friedman, C. A. Fractionation of DNA from mammalian cells by alkaline elution. Biochemistry, 14:
4629-4637, 1976.
Jonas, A. J., Smith, M. L., Allison, W. S., Laikind, P. K., Greene, A. A.,
and Schneider, J. A. Proton-translocating ATPase and lysosomal cystine
transport. J. Biol. Chem., 25«:11727-11730, 1983.
Smith, P. K., Krohn, R. I., Hermanson, G. T., Mallia, A. K., Gartner, F. H.,
Provenzone, M. D., Fujimoto, E. K., Goeke, N. M., Olson, B. J., and Klenk,
D. C. Measurement of protein using bicinchoninic acid. Anal. Biochem., 150:
76-85, 1985.
Aronson, N. N., and Touster, O. Isolation of rat liver plasma membrane
fragments in isotonic sucrose. Methods Enzymol., 31: 90-102, 1974.
Li, Y. T., and Li, S. T. a-Mannosidase, /3-/V-acetylhexosaminidase and ßgalactosidase from jack bean meal. Methods Enzymol., 28: 702-713, 1972.
DiPietro, D. L., and Zengerle, F. S. Separation and properties of three acid
phosphatases from human placenta. J. Biol. Chem., 242: 3391-3396, 1967.
Fishman, W. H., Kato, K., Antiss, C. L., and Green, S. Human serum /3glucuronidase: its measurement and some of its properties. Clin. Chim. Acta,
75:435-447, 1967.
Moore, D. J. Isolation of Golgi apparatus. Methods Enzymol., 22: 130-148,
1971.
Tulsiani, D. R. P., Opheim, D. J., and Touster, O. Purification and charac
terization of a-D-mannosidase from rat liver Golgi membranes. J. Biol.
Chem., 252: 3227-3233, 1977.
Hamburger, A. W., and Salmon, S. E. Primary bioassay of human tumor
stem cells. Science (Wash. DC), 797:461-463, 1977.
Salmon, S. E., Young, L., Lebowitz, A. Thomson, S., Einspahr, J., Tong, T.,
and Moon, T. Evaluation of an automated image analysis system for counting
human tumor colonies. Int. J. Cell Cloning, 2: 142-160, 1984.
Sedlak, J., and Lindsay, R. H. Estimation of total, protein-bound and nonprotein sulfhydryl groups in tissue with Ellman's reagent. Anal. Biochem.,
25: 192-205, 1968.
Alberts, D. S., Salmon, S. E., Chen, H. S. G., Moon, T. E., Young, L., and
Sm«it, E. A. Pharmacologie studies of anticancer drugs with the human
tumor stem cell assay. Cancer Chemother. Pharmacol., 6: 253-264, 1981.
Nakano, K., Abe, S., and Sohmura, Y. Recombinant human tumor necrosis
factor. I. Cytotoxic activity in vitro. Int. J. Immunopharmacol., 8: 347-355,
1986.
Leonessa, F., Coialbu, T., and Toma, S. Different techniques for drug
cytotoxicity evaluation on MCF-7 human breast cancer cell lines. Anticancer
Res., 6: 1291-1296, 1986.
2727
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1989 American Association for Cancer Research.
LYSOSOMES AND TUMOR NECROSIS FACTOR
35. Scatchard, G. The attractions of proteins for small molecules and ions. Ann.
NY Acad. Sci. SI: 660-672, 1949.
36. Weissmann, G. The effects of steroids and drugs on lysosomes. In: J. T.
Dingle and H. B. Fell (eds.), Lysosomes in Biology and Pathology, Vol. 1,
pp. 276-295. Amsterdam: North Holland, Biomedicai Press, 1969.
37. Ostrove, J. M., and Gifford, G. E. Stimulation of RNA synthesis in L929
cells by rabbit tumor necrosis factor. Proc. Soc. Exp. Biol. Med., 160: 354358 1979
38. Darzynkiewics, Z., Williamson, B., Carswell, E. A., and Old, L. J. Cell cyclespecific effects of tumor necrosis factor. Cancer Res., «.-83-90, 1984.
t€\Fiers,
rn, Brouckaert,
n
i
i> Devos,
r,
i> Fransen,
r, Leroux-Roels,i G.,
r* Remaut,
r,
39.
W.,
P.,
R.,
L
E., SufTys, P., Tavernier, J., Van der Heyden, J., and Van Roy F. Lympnokines and monokines in anti-cancer therapy. Cold Spring Harbor Symp.
Quant. Biol., 51: 587-595, 1986.
40. Ruddle, N. H. Tumor necrosis factor and related cytotoxins. Immunol.
Today, S: 129-130, 1987.
41. Schmid, D. S., Tite, J. P., and Ruddle, N. H. DNA fragmentation: manifestation of target cell destruction mediated by cytotoxic I cell lines, lymph«>
42.
43.
44'
..
45.
46
47.
toxin-secreting helper I-cell clones, and cell-free lymphotoxin-containing
supernatant. Proc. Nati. Acad. Sci. USA, 83: 1881-1885, 1986.
Aklyama, S., Tornita, K., and Kuwano, M. The effect of calcium antagonists
on the proteolytic degradation of low-density lipoprotein in HeLa cells. Exp.
Cell Res., 158: 192-204, 1985.
Barrett, A. J., and Heath, M. F. Lysosomal enzymes. In: J. T. Dingle (ed.),
Lysosomes: a Laboratory Handbook, Ed. 2, pp. 19-145. Amsterdam: NorthHolland Biomedicai Press, 1977.
T"takoff. A- M- ***£!**£
£s8icV'aar
«rafllcwilh the carb°*J'lieion°w?'6 ^T^K'f
B il
H w H n r r i , •¿ f •¿
Wilcox,
Kitson, R.
P., andwjth
Widnell,
C. C. j.Inhibition
of pinocytosis
M| emb D. K.,
fibroblasts
,reatcd
monensin.
Cell Biol.
92: 859-864,in
jcg2
Aoyag¡T Ishizuka, M., Jakeuchi, T., and Umezawa, H. Enzyme inhibitors
in re|at¡onto cancer therapy. J. Antibiotics, 30: 121-132, 1977.
Barrett, A. J., Kembhavi, A. A., Brown, M. A., Kirschke, H., Knight, C. G.,
Tamai, M., and Hanada, K. L-frans-Epoxysuccinyl-leucylamido(4-guanidino)butane (E-64) and its analogues as inhibitors of cysteine proteinases
including cathepsins B, H and L. Biochem. J., 201: 189-198, 1982.
2728
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1989 American Association for Cancer Research.
Association of Lysosomal Activity with Sensitivity and
Resistance to Tumor Necrosis Factor in Murine L929 Cells
James D. Liddil, Robert T. Dorr and Philip Scuderi
Cancer Res 1989;49:2722-2728.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/49/10/2722
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1989 American Association for Cancer Research.