J Neurosurg 73:760-767, 1990
Cholesterol uptake by human glioma cells via receptormediated endocytosis of low-density lipoprotein MASAJI MURAKAMI, M.D., YUKITAKA USHIO~ M.D., YOSUKE MIHARA, M.D., JUNICHI KURATSU, M.D., SEIKOH H o m u c m , M.D., AND YOSHIMASA MORINO, M.D. Departments of Neurosurgery and Biochemistry, Kumamoto University Medical School, Kumamoto, Japan L,- Low-density lipoprotein (LDL) is a carrier of the cholesterol found in human plasma. Cells utilize cholesterol for membrane synthesis by taking up LDL via receptor-mediated endocytosis. In the present study, interactions of LDL with human malignant glioma cell lines (U-251MG and KMG-5) were investigated biochemically and morphologically. The LDL, labeled with the fluorescent dyes 1,1 '-dioctadecyl-3,3,Y,Y-tetramethylindocarbocyanine (DiI) and fluorescein isothiocyanate (FITC), was internalized by both cell processes and cell bodies. Reductive methylation of DiI-labeled LDL, which abolishes the ability of the cell to bind to the LDL receptor, prevented the internalization of the cholesterol moiety of LDL. Cellular binding of ~25I-LDLto U-25 I MG cells at 4~ revealed the presence of a specific saturable-associated receptor (dissociation constant (Kd) approximately 38 ~g/ml). Endocytic uptake of ~25I-LDLor 3H-cholesterol oleate-labeled LDL (3H-LDL) at 37~ demonstrated the cell-associated t25I-LDLand 3H-LDL increase. The intraceUular degradation of protein moiety increased linearly with time. Reductive methylation of aH-LDL led to a remarkable decrease in the cell-associated cholesterol moiety of LDL. The difference in uptake of the cholesterol moiety of LDL between U-25 IMG cells and KMG-5 cells showed that the U-25 IMG cells, which proliferate more actively than KMG5 cells, take up more of the cholesterol moiety of LDL than do the KMG-5 cells. Thus, LDL cholesterol seems to be endocytosed predominantly via the LDL receptor present on the plasma membrane of malignant glioma cells. In addition, for growth, these cells may require large amounts of the cholesterol moiety of LDL. KEY WORDS
L
9 glioma
9 lipoprotein
OW-DENSITY lipoprotein (LDL) is an endogenous carrier of cholesterol found in human plasma; 75% of the cholesterol in plasma is transported in L D L ? Cells requiting cholesterol for membrane synthesis are known to take up L D L via receptor-mediated endocytosis or to initiate the de novo synthesis of cholesterol. The LDL-receptor activity is enhanced in rapidly proliferating cells in culture, whereas few receptors are found on nondividing cellsJ '~9 Such being the case, malignant tissue might be expected to have higher LDL-receptor activity than normal tissue with a very low rate of replication.~~ Epidemiological studies have shown an increased risk of death from cancer in subjects with low plasma cholesterol levels, unrelated to nutritional s t a t u s . 6'8'22 In cases of acute leukemia, leukemic blood and bonemarrow cells from most patients have elevated LDLreceptor activities compared with white blood cells and bone-marrow cells from healthy subjects. A high receptor-mediated uptake and degradation of LDL by the 760
9 cholesterol metabolism
9 receptor
leukemic cells were shown to be the cause of hypocholesterolemia in cases of acute leukemia; 14'33hypocholesterolemia is a secondary phenomenon, due to the elevated LDL-receptor activity. Evidence for a high LDLreceptor activity in various tumor cells has also been obtained, both in vitro and in vivo. ~t,~6,26,32 Recently, LDL-receptor activity has also been identified in normal tissues and neoplasms of the central nervous system. Pitas, et al..27 noted that rat astrocytes have the LDL receptor, and Rudling, et al., 3~ showed that a human malignant glioma cell line (U-251MG) accumulates and degrades LDL by a saturable highaffinity process. It has been shown by morphological techniques that neuronal growth cones of pheochromocytoma rapidly take up LDL via L D L receptor-mediated endocytosis. J7 In addition, it was also reported that after a crush injury to the rat sciatic nerve, a cholesterol transfer mechanism was required for rapid membrane biogenesis during axon regeneration and remyelination. 3 Thus, the role of LDL receptor in cholesterol J. Neurosurg. / Volume 73/November, 1990
Lipoprotein cholesterol uptake by gliomas transport in neural tissues has been the focus of particular attention. In the present study, the interactions of LDL with human malignant glioma cell lines (U-251MG and KMG-5) were investigated biochemically and morphologically. Emphasis was placed on the dependency on LDL for the transfer of cholesterol into the human glioma cells. Materials and Methods
Lipoprotein Isolation and Modification Low-density lipoprotein (density (D) = 1.019 to 1.063) and lipoprotein-deficient serum (D > 1.25) were isolated by ultracentrifugation from fresh human plasma of normolipidemic subjects, after an overnight fast, as described. 25 The LDL was iodinated with the Bolton-Hunter reagent,* according to the manufacturer's instructions, to a specific activity of 270 cpm/ng of protein (~25I-LDL); other LDL was labeled with (3H)cholesterol oleate, as described elsewhere 23 to a specific activity of 43 dpm/ng of total cholesterol ester (3H-LDL). In addition, LDL was labeled with fluorescein isothiocyanate (FITC) or with 1,1 '-dioctadecyl3,3,3',3'-tetramethylindocarbocyanine (DiI), as follows. In brief, LDL concentrations of 4 to 5 mg/ml of protein were adjusted to a pH of 9 with bicarbonate buffer, and FITC solution, 5 mg/ml, was added at the ratio of 100 to 150 ug FITC/mg of LDL protein. The mixture was allowed to stand at room temperature for 4 hours, after which it was applied to a G-50 column equilibrated with a Tris-saline solution (pH 8.0) to remove the unbound fraction. The first eluting portion was collected and dialyzed against 0.15 M-NaC1, 0.02% ethylenediaminetetra-acetic acid (pH 7.4) (Buffer A). The molar ratio of FITC/LDL was about 2.3; FITC is covalently linked to the lysine residues of proteins. Therefore, the FITC-LDL indicated behavior of the protein moiety of LDL. Next, LDL (2 mg/ml) was mixed with lipoprotein-deficient serum (D > 1.25) and then DiI (3 mg/ml) in dimethyl sulfoxide was added. The mixture was allowed to stand at 37~ for 12 hours, after which it was ultracentrifuged for 24 hours at a density of 1.063 gm/ml to remove the unbound fraction. After centrifugation, the isolated fraction was dialyzed against Buffer A. Because DiI is a highly lipophilic molecule that can be noncovalently incorporated into LDL, the DiI-LDL indicated behavior of the cholesterol moiety of LDL. This procedure was based on previously reported methods. 9'29
Reductive Methylation of 3H-LDL or DiI-LDL Tritiated LDL or DiI-LDL (3 to 5 mg/ml of protein), dissolved in Buffer A, was diluted with 0.3 M sodium borate buffer (pH 9.0) to 1.5 times the original volume.
* Bolton-Hunter reagent (2200 Ci/mmol) obtained from New England Nuclear, Boston, Massachusetts.
J. Neurosurg. / Volume 73/November, 1990
Reductive melhylation of 5 mg of the samples was performed at 0~ by adding 1 mg of sodium borohydrate (zero tirr.e for reaction sequence) followed by six additions over 30 minutes of 1 ul of 37% aqueous formaldehyde. After the last addition of formaldehyde, the reaction mixture was dialyzed at 4~ against Buffer A for 20 hour,s. The specific activity of reductive methylated 3H-LDL was 48 dpm/ng of total cholesterol ester. This procedure was based on previously reported methods. 34
Binding Assay For the binding assay, 105 U-251MG or KMG-5 malignant glioma cells? were plated in Medium A (Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum) into 24-well (16 • 16 mm) polystyrene dishes on Day 0. On Day 2, the medium wa~,;removed, and the cells were washed three times with l~hosphate-buffered saline (PBS). The medium was then replaced with 500 #1 of fresh Medium B (DMEM supplemented with 1% human serum albumin) with an indicated or fixed concentration of ~2~ILDL, in the absence or presence of excess amounts of unlabeled LDL. After 3 hours of incubation at 4"C, the cells were washed three times with PBS, and 300 ul of 0.1 N NaC1 was added directly to the washed cells to determine the cellular binding of ~25I-LDL and cell protein concentration.
Endocytic Uptake of Radiolabeled LDL Fixed a~nounts of ~25I-LDL were incubated with the cells in 500 ul of Medium B at 37"C and, at various times, cellular binding and intracellular degradation were determined as described. ~5'24 The cells were also incubated with fixed amounts of 3H-LDL and reductive methylated 3H-LDL over time, or with indicated concentrations for 3 hours at 37~ After extensive washing of the cells, 300 ul of 0.1 N NaC1 was added for determination of cellular uptake and cell protein concentration, as described elsewhere. 23
Morphological Behavior of LDL The U-251MG glioma cells in Medium A were plated in an eight-well chamber, and on Day 2 the medium was replaced with Medium B. These cells were then incubated with 50 ug/ml of FITC-LDL and 15 tsg/ml of DiI-LDL in the absence or presence of excess amounls of unlabeled LDL for studies of protein moiety and the cholesterol moiety of LDL. After 30 minutes to 6 hoars of incubation at 37~ the cells were washed three times for 1 minute with PBS supplemented with 1% bovine serum albumin, then fixed with 4% paraformaldehyde for 1 hour at room temperature. Other U251MG glioma cells were incubated with 15 ug/ml of t The U-251MG cell line was kindly provided by Dr. J Yoshida, Department of Neurosurgery, Nagoya University Medical School, Nagoya, Japan; the KMG-5 cell line was established in our laboratories. 2~
761
//"
7OO0 .IS
%,,,,,, '" 9
e
\
c'~ _.1
10(X
i
0
i i i 1
125 I-
.100
LDL(.IJg/!11i)
9
200
'0
i 2 3 4 Bound(IJg/mg eeIlprotem/3hr)
FIG, 1, Binding of ~251-labeledlow-density lipoprotein (LDL) to U-251MG cells as a function of concentration. Left: The cells were incubated at 4~ for 3 hours with increasing amounts of J25I-LDL. After a wash with ice-cold phosphate-buffered saline, total binding (solid circles) was determined as described in the Materials and Methods section. Nonspecific binding (open circles) was determined by parallel incubation in the presence of more than 100-fold of unlabeled LDL. Specific binding (triangles) was plotted after correction for nonspecific binding, Right." Scatchard plot of the specific binding illustrated left. Each point represents the mean value of duplicate assays.
E 2C
/
e
jo.-----o
./7
/.oP1
0
2
3
.
~
[ncubcttion Time(hr)
5
6
FIG. 2. Endocytic uptake of ~25I-low-density lipoprotein (LDL) in U-251MG cells. The cells were incubated at 37"C with 20 pg-protein of ~25I-LDL in 500 t~l of Medium B, and examined at various times. Cellular binding (closed circles) and intracellular degradation (open circles) of ~25I-LDLwere determined, as described in the Materials and Methods section. Each point represents the mean value of duplicate assays. 762
F~o. 3. Behavior of cholesterol moiety of native and reductive methylated low-density lipoprotein (LDL) in the U251MG cells. The cells were incubated at 37~ with 10 ugprotein of 3H-labeled native LDL (dosed circles) or reductive methylated LDL (open circles), as a function of concentration for 3 hours in 500 ul of Medium B. CE = cholesterol ester. J. Neurosurg. / Volume 73/November, 1990
Lipoprotein cholesterol uptake by gliomas reductively methylated DiI-LDL for 6 hours. All of the cells were examined using fluorescence microscopy.
binding (Bmax) of 3.4 ug protein of LDL/mg cell protein.
LDL-Receptor Localization To study LDL-receptor localization, U-251MG cells were preincubated with Medium B overnight then fixed with 4% paraformaldehyde for 20 minutes at room temperature. They were then incubated at 1:50 with monoclonal anti-bovine LDL-receptor antibody (clone 7, which reacts with samples of human origin ~) as primary antibody for 1 hour at room temperature, and extensively washed in PBS. These cells were then incubated with FITC-labeled anti-mouse immunoglobulin G antibody for 40 minutes at room temperature. Fluorescence microscopy was used to examine LDL-receptor localization.
Endocytic Uptake of LDL in U-251MG Cells Endocytic uptake, an event occurring subsequent to the initial binding, was examined by incubating 125ILDL at 37~ with U-251MG cells. As shown in Fig. 2, the cell-associated 125I-LDL increased with time and reached equilibrium within 6 hours. The intracellular degradation of the ligand, as determined by an increase in trichloroacetic acid-soluble radioactivity in the medium, began at around 20 minutes after adding the radiolabeled ligand and continued to increase linearly with time thereafter. To further determine behavior of the cholesterol moiety of LDL, the U-251MG cells were incubated at 37~ as a function of concentration, or with a fixed concentration of 3H-cholesterol oleate-labeled LDL, followed by determination of the cell-associated radioactivity. Consequently, the amounts of the cell-associated cholesterol moiety of LDL increased as a function of concentration (Fig. 3) or with time (data not shown). The reductive methylation of LDL, which abolishes its ability to bind to the LDL receptor, resulted in a remarkable decrease in the cell association of the cholesterol moiety of LDL in the cells as a function of concentration (Fig. 3). These findings indicated that the uptake of cholesterol moiety of LDL was mediated via the LDL receptor present on the plasma membranes of the cells.
Results
Binding of LDL to U-251 MG Cells Figure 1 shows the binding of 125I-LDL to U-251MG cells at 4"C, as a function of concentration. The binding was inhibited by more than 60% by an excess of unlabeled LDL. The specific binding, exhibiting a typical saturation curve, indicated the presence of a receptor specific for LDL on the plasma membranes of U251MG cells. Scatchard analysis of this binding process gave a straight line, thereby suggesting a single mode of binding with an apparent dissociation constant (Kd) of approximately 38 ug/ml (1.7 • 10-8 M) and maximum
FIG. 4. Labeling of U-251MG cells with FITC-labeled low-density lipoprotein (LDL). The cells were incubated with 50 ug/ml of FITC-labeled LDL in Medium B at 37~ for 6 hours. Note the intense granular fluorescence pattern within the cell bodies as well as processe~;(left), and that the fluorescence labeling was blocked by incubating the cells with an excess amount of unlabeled LDL (right). • 500.
J. Neurosurg. / Volume 73 /November, 1990
763
M. Murakami, et aL
FIG. 5. •abe•ing•fU-25•MGce••swithDi•-•abe•ed••wdensity•ip•pr•tein(LDL).Thece••swereincubated with 15 ~g/ml of Dil-labeled LDL in Medium B at 37*C for 30 minutes (A) or 6 hours (B), with DiI-labeled LDL in the presence of an excess amount of unlabeled LDL (C), or with reductive methylated Dil-labeled LDL (D). After 30 minutes of incubation, the granular fluorescence pattern was definitely observed within the body of the cells as well as their processes (A and B). The fluorescence labeling was blocked in C and D. x 500.
Localization of F I T C - L D L or DiI-LDL in U-251MG Cells To determine behavior of the protein moiety and cholesterol moiety of LDL in U-251MG cells morphologically, FITC-LDL and DiI-LDL were used. An intense granular fluorescence pattern was observed within the cell body and processes of U-251MG cells after 6 hours of incubation of the cells with FITC-LDL at 37~ (Fig. 4 left). Fluorescence labeling could be blocked by incubating the cells with a 30-fold excess of unlabeled
764
LDL along with FITC-LDL (Fig. 4 right), thereby indicating that the FITC-LDL was internalized via the LDL receptor in the cells. To further determine the behavior of the cholesterol moiety of LDL, DiI-LDL was incubated with U251MG cells over time at 37~ Granular staining was seen in the body of the cells as well as in their processes after 30 minutes of incubation (Fig. 5A). An intense granular fluorescence pattern was observed within the body of the cells as well as their processes after 6 hours
J. Neurosurg. / Volume 73/November, 1990
Lipoprotein cholesterol uptake by gliomas Thus, the uptake of cholesterol moiety of LDL was mediated via the LDL receptor present on the plasma membrane of U-251MG cells.
LDL.Receptor Visualization in U-251MG Cells Indirect immunofluorescence studies using the monoclonal anti-LDL-receptor antibody showed that the LDL-:eceptor was visible within the cell body and on the surface of the cells (Fig. 6). In addition, a strongly positive immunoreaction was observed, especially in the dividing neoplastic cells.
FIG. 6. Low-density lipoprotein (LDL)-receptor visualization in U-251MG cells. After fixation with 4% paraformaldehyde for 20 minutes at room temperature, the cells were incubated with monoclonal anti-LDL-receptor antibody fbr 1 hour at room temperature. Note that the LDL receptor can be visualized within the cell body and on the surface of the cells, x 420.
Uptake of L D L Cholesterol in U-251MG and in K34G..5 Cells To observe differences in the uptake of LDL cholesterol in the U-251MG and KMG-5 cells (the doubling time of which is approximately 25 and 100 hours, respeclively), endocytic uptake of LDL cholesterol was examined by biochemical techniques (Fig. 7). Two kinds of cells were incubated at 37~ with 3H-LDL over time, followed by determination of the cell-associated cholesterol moiety of LDL. Uptake of the cholesterol moiety of LDL in the U-251MG cells was over threefold higher than in the KMG-5 cells, after incubation. Discussion
FIG, 7. Comparable uptake of cholesterol moiety of lowdensity lipoprotein (LDL) in U-251MG cells and KMG-5 cells. U-251MG cells (circles) or KMG-5 cells (triangles) were incubated with t0 ug-protein of 3H-LDL at 37~ over time in 500 ~1 of Medium B. Uptake of the cholesterol moiety of LDL was determined as described in the Materials and Methods section. Each point represents the mean value of duplicate assays. CE = cholesterol ester.
of incubation (Fig. 5B). Fluorescence labeling could be blocked by incubating the cells with a 30-fold excess of unlabeled LDL (Fig. 5C). Furthermore, reductive methylation of the apoprotein moiety of LDL, a procedure known to prevent receptor-mediated binding of LDL, prevented most of the binding and internalization of the DiI-LDL (Fig. 5D).
J. Neurosurg. / Volume 73/November, 1990
The results of this study gave evidence that the LDL recepto:: is present on the plasma membranes of malignant glioma cells (U-251MG), and that the cholesterol moiety of LDL is endocytosed predominantly via the LDL receptor in the cells. The literature on LDL metabolism in tumors of the central nervous system is sparse. ~'~~Rudling, et al., 3~found that a human malignant glioma cell line (U-251MG) accumulates and degrades the apoprotein moiety of LDL by a saturable high-affinity process; however, little is known of the cholesterol metabolism of LDL in human glial neoplasms. Hence, we focused our study on the cholesterol metabolism of LDL in human malignant gtioma cell lines. It is well known that regulation of cellular cholesterol metabolism depends on sequential cell-surface binding, internalization, and intracellular degradation of LDL. ~2.~3
L D L Binding in Glioma The cell-surface binding of ~2~I-LDL was studied at 4~ as sl:Lown in Fig. 1. The data provide evidence that the plasma membranes of a human malignant glioma cell line (U-251MG) possess the high-affinity LDL receptor (Kd approximately 38 ug/ml). This is in good agreement with the study of Rudling, eta/.; 3~the temperature at which the experiments were conducted could probably produce the different values of Kd o f the LDL receptor in U-251MG cells seen between their data and ours. Evidence that the LDL receptor was visualized within the cell bcdy as well as on the surface of the cells (Fig. 6) is interpreted to mean that the internalization of the 765
M. Murakami, et al. LDL receptor into the cells occurs after binding to LDL, and/or that the binding of LDL receptor to the cell surface occurs following production of the LDL receptor in the Golgi apparatus. 2~
LDL Cholesterol Uptake in Glioma It is of interest to know the extent to which transport of the cholesterol moiety of LDL into the human malignant glioma cells depends on the LDL receptor. For elucidation, lysine residues of LDL were modified by reductive methylation, a procedure known to block the interaction of LDL with the LDL receptor. 34 Modification of lysine residues by reductive methylation was shown to be an important procedure to probe the role of lysine in the recognition site since, unlike other modifications such as diketone and carbamoylation, this reaction did not alter the charge of the lysine or the net charge of LDL and yet was effective in blocking the binding activity of LDL. The lipid compositions were essentially identical for untreated LDL and for LDL modified by reductive methylation (data not shown). 34 In addition, DiI-lipoproteins were shown to have the same chemical composition and physical properties as the native lipoproteins. 28'29 Consequently, biochemical and morphological evidence using reductively methylated LDL demonstrated that the amount of transport of cholesterol moiety of modified LDL into malignant glioma cells significantly decreased. Thus, it seems likely that more than 80% of the cholesterol moiety of LDL is endocytosed via the LDL receptor (Fig. 3). In vivo experiments revealed that about 80% of plasma LDL is cleared by the LDL receptor, and the
100
,
,
,
,
,
8O P
g u
60
Implications of LDL Cholesterol Uptake in Glioma It is also of interest to know the implications o f uptake of cholesterol moiety of LDL in malignant glioma cells. For this study, we used other malignant glioma cells (the KMG-5 cell line 2~) which are known to divide less rapidly than U-251MG cells in culture systems. The doubling time of KMG-5 and U-251MG cells is 100 and 25 hours, respectively. 2'2~ The binding of 125I-LDL to KMG-5 at 4"C was also inhibited in a dose-related manner by unlabeled LDL; however, the rate of inhibition was higher in the U-251MG than in the KMG-5 cells, suggesting that LDL-receptor activity might be higher in U-251MG than in KMG-5 cells (Fig. 8). As shown in Fig. 7, uptake of the cholesterol moiety of LDL in the U-251MG cells was three times than in the KMG-5 cells. Thus, malignant glioma cells that grow more rapidly and have a higher activity of LDL receptor might need a larger amount of the cholesterol moiety of LDL. The cholesterol moiety of LDL is rapidly endocytosed in growth cones of pheochromocytoma cells, 17 and the rate of LDL uptake in cervical cancer cells ~ or leukemia cells33 is much greater than that in corresponding non-neoplastic cells. In addition, it is shown that the extracytoplasmic domain of the LDL receptor contains a region that is 38% identical with a 96 amino acid sequence in the precursor to the epidermal growth factor?' Thus, large amounts of receptor-mediated uptake of the cholesterol moiety of LDL in malignant glioma cells may correlate with cell growth. Acknowledgments
0 139
We are grateful to Drs. K. Takata and T. Ohta for valuable discussions and suggestions, and to M. Ohara for editorial assistance.
_J t23
-~, 20
160 ~
~
~0
Unlobeled LDL(pg/rnl) FIG. 8. Competition by unlabeled low-density lipoprotein (LDL) for ~:5I-LDL binding to U-251MG cells or KMG-5 cells. U-251MG cells (closed circles) or KMG-5 cells (open circles) were incubated with 10 ug-protein of ~25I-LDLin 500 ul of Medium B for 3 hours at 4~ The cell-associated radioactivity was determined as described in the Materials and Methods section.
766
remaining 20% is removed by nonspecific mechanisms, determined using glucosylated L D L ) 8 In addition, in vitro experiments raised the possibility of dual mechanisms for the uptake and regulation of LDL; in fibroblasts, almost all the LDL was endocytosed and degraded by a receptor-mediated mechanism while, in hepatocytes, even LDL modified by acetoacetylation, acetylation, and reductive methylation was endocytosed and degraded at 30% to 50% of the level of native LDL. 7 Therefore, the mechanism for the uptake and regulation of LDL in glioma cells seems to be similar to that in fibroblasts.
References 1. Beisiegel U, Schneider WJ, Goldstein JL, et al: Monoclonal antibodies to the low density receptor as probes for study of receptor-mediated endocytosis and the genetics of familial hypercholesterolemia. J Biol Chem 256: 11923-11931, 1981 2. Bigner SH, Bullard DE, Pegram CN, et al: Relationship of in vitro morphologic and growth characteristics of established human glioma-derived cell lines to their tu-
J. Neurosurg. / Volume 73/November, 1990
Lipoproteincholesterol uptake by gliomas 3.
4.
5. 6. 7.
8. 9. t0.
11.
12. t3.
14. 15. 16. 17. 18.
19.
morigenicity in athymic nude mice. J Neuropathol Exp Neurol 40:390-409, 1981 Boyles JK Zoellner CD, Anderson LJ, et al: A role for apolipoprotein E, apolipoprotein A-l, and low density lipoprotein receptors in cholesterol transport during regeneration and remyelination of the rat sciatic nerve. J Clin Invest 83:1015-1031, 1989 Brown MS, Goldstein JL: Analysis of a mutant strain of human fibroblasts with a defect in the internalization of receptor-bound low density lipoprotein. Cell 9:663-674, 1976 Brown MS, Goldstein JL: The low-density lipoprotein pathway and its relation to atherosclerosis. Annu Rev Biochem 46:897-930, 1977 Chao FC, Efron B, Wolf P: The possible prognostic usefulness of assessing serum proteins and cholesterol in malignancy. Cancer 35:1223-1229, 1975 Edge SB, Hoeg JM, Triche T, et al: Cultured human hepatocytes. Evidence for metabolism of low density lipoproteins by a pathway independent of the classical low density lipoprotein receptor. J Biol Chem 261: 3800-3806, 1986 Feinleib M: Review of the epidemiological evidence for a possible relationship between hypocholesterolemia and cancer. Cancer Res 43:2503-2507, 1983 Fothergill JE: Huoroehromes and their conjugation with proteins, in Nairn RC (ed): Fluorescent Protein Tracing, ed 3. London: E & S Livingstone, 1969, pp 5-34 Gal D, MacDonald PC, Porten JC, et al: Effect of cell density and confluency on cholesterol metabolism in cancer cells in monolayer culture. Cancer Res 41: 473-477, 1981 Gal D, Ohashi M, MacDonald PC, et al: Low density lipoprotein as a potential vehicle for chemotherapeutic agents and radionucleotides in the management of gynecological neoplasms. Am J Obstet Gynecol 139: 877-885, 1981 Goldstein JL, Basu SK, Brunschede GY, et al: Release of low density lipoprotein from its cell surface receptor by sulfated glycosaminoglycans. Cell 7:85-95, 1976 Goldstein JL, Brown MS: Binding and degradation of low density lipoproteins by cultured human fibroblasts: comparison of cells from a normal subject and from a patient with l~omozygous familial hypercholesterolemia. J Biol Chem 249:5153-5162, 1974 Ho YK, Smith RG, Brown MS, et al: Low density lipoprotein (LDL) receptor activity in human acute myelogenous leukemia cells. Blood 52:1099-I 1I4, I978 Horiuchi S, Murakami M, Takata K, et al: Scavenger receptor for aldehyde-modified proteins. J Biol (?hem 261:4962-4966, 1986 Hynds SA, Welsh J, Stewart JM, et al: Low-density lipoprotein metabolism in mice with soft tissue tumours. Biochim Biophys Acta 795:589-595, 1984 Ignatius MJ, Shooter EM, Pitas RE, et al: Lipoprotein uptake by neuronal growth cones in vitro. Science 236: 959-962, 1987 Kesaniemi YA, Witztum JL, Steinbrecher UP: Receptormediated catabolism of low density lipoprotein in man. Quantitation using glycosylated low density lipoprotein. J Clin Invest 71:950-959, 1983 Kruth HS, Arigan J, Gamble W, et al: Effect of cell density on binding and uptake of low density lipoprotein by human fibroblasts. J Cell Biol 83:588-594, 1979
J. Neurosurg. / Volume 73/November, 1990
20. Mahley RW: Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science 40: 622-630. 1988 21. Mihara "(: Establishment and characterization of 5 cell lines derived from human brain tumors. Kumamoto Med J 39:25-41, 1986 22. Miller SR, Tartter PI, Papatestas AE, et al: Serum cholesterol and human cancer. JNCI 67:297-300, 1981 23. Murakami M, Horiuchi S, Takata K, et al: Distinction in the mode of receptor-mediated endocytosis between high density lipoprotein and acetylated high density lipoprorein: evklence for high density lipoprotein receptor-mediated cholesterol transfer. J Biochem 101:729-741, 1987 24. Murakami M, Horiuchi S, Takata K, et al: Scavenger receptor for malondialdehyde-modified high density lipoprotein on rat sinusoidal liver cells. Biochem Biophy Res Commun 137:29-35, 1986 25. Murakami M, Ushio Y, Morino Y, et al: Immunohistochemical localization of apolipoprotein E in human glial neoplasms. J Clin Invest 82:t77-188, 1988 26. Norata G, Canti G, Ricci L, et al: In vivo assimilation of low density lipoproteins by a fibrosarcoma tumour line in mice. Cancer Lett 25:203-208, 1984 27. Pitas RE, Boyles JK, Lee SH, et al: Astrocytes synthesize apolipogrotein E and metabolize apolipoprotein E-containing lipoproteins. Biochim Biophys Acta 917: 148-161, 1987 28. Pitas RE, Innerarity TL, Mahley RW: Foam cells in explant ~of atherosclerotic rabbit aortas have receptors for beta-ve~:y low density lipoproteins and modified low density lipoproteins. Arteriosclerosis 3:2-12, 1983 29. Pitas RE, Innerarity TL, Weinstein JN, et al: Acetoacetylated lipoproteins used to distinguish fibroblasts from macrol:,hages in vitro by fluorescence microscopy. Arteriosclerosis 1:177- 185, 1981 30, Rudling M J, Collins VP, Peterson CO: Delivery of aclacinomycin A to human glioma cells in vitro by the lowdensity lipoprotein pathway. Cancer Res 43:4600-4605, 1983 31. Russell DW, Schneider W J, Yamamoto T, et al: Domain map o:~ the LDL receptor: sequence homology with the epidermal growth factor precursor. Cell 37:577-585, 1984 32. Sato JD, Kawamoto T, Okamoto T: Cholesterol requirement of P3-X63-Ag8 and X63-AgS.653 mouse myeloma cells fi~r growth in vitro. J Exp Med 165:1761-1766, 1987 33. Vitols S, Gahrton G, Ost A, et al: Elevated low density lipoprotein receptor activity in leukemic cells with monocytic differentiation. Blood 63:1186-1193, 1984 34. Weisgraber KH, Innerarity TL, Mahley RW: Role of the lysine residues of plasma lipoproteins in high affinity bindirg to cell surface receptors on human fibroblasts. J Biol Chem 253:9053-9062, 1978 Manuscript received September 29, 1989. Accepted in final form April 19, 1990. This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture of Japan. Address reprint requests to: Masaji Murakami, M.D., Department of Neurosurgery, Kumamoto University Medical School, Honjo 1-1-I, Kumamoto 860, Japan.
767
Cholesterol uptake by human glioma cells via receptormediated endocytosis of low-density lipoprotein MASAJI MURAKAMI, M.D., YUKITAKA USHIO~ M.D., YOSUKE MIHARA, M.D., JUNICHI KURATSU, M.D., SEIKOH H o m u c m , M.D., AND YOSHIMASA MORINO, M.D. Departments of Neurosurgery and Biochemistry, Kumamoto University Medical School, Kumamoto, Japan L,- Low-density lipoprotein (LDL) is a carrier of the cholesterol found in human plasma. Cells utilize cholesterol for membrane synthesis by taking up LDL via receptor-mediated endocytosis. In the present study, interactions of LDL with human malignant glioma cell lines (U-251MG and KMG-5) were investigated biochemically and morphologically. The LDL, labeled with the fluorescent dyes 1,1 '-dioctadecyl-3,3,Y,Y-tetramethylindocarbocyanine (DiI) and fluorescein isothiocyanate (FITC), was internalized by both cell processes and cell bodies. Reductive methylation of DiI-labeled LDL, which abolishes the ability of the cell to bind to the LDL receptor, prevented the internalization of the cholesterol moiety of LDL. Cellular binding of ~25I-LDLto U-25 I MG cells at 4~ revealed the presence of a specific saturable-associated receptor (dissociation constant (Kd) approximately 38 ~g/ml). Endocytic uptake of ~25I-LDLor 3H-cholesterol oleate-labeled LDL (3H-LDL) at 37~ demonstrated the cell-associated t25I-LDLand 3H-LDL increase. The intraceUular degradation of protein moiety increased linearly with time. Reductive methylation of aH-LDL led to a remarkable decrease in the cell-associated cholesterol moiety of LDL. The difference in uptake of the cholesterol moiety of LDL between U-25 IMG cells and KMG-5 cells showed that the U-25 IMG cells, which proliferate more actively than KMG5 cells, take up more of the cholesterol moiety of LDL than do the KMG-5 cells. Thus, LDL cholesterol seems to be endocytosed predominantly via the LDL receptor present on the plasma membrane of malignant glioma cells. In addition, for growth, these cells may require large amounts of the cholesterol moiety of LDL. KEY WORDS
L
9 glioma
9 lipoprotein
OW-DENSITY lipoprotein (LDL) is an endogenous carrier of cholesterol found in human plasma; 75% of the cholesterol in plasma is transported in L D L ? Cells requiting cholesterol for membrane synthesis are known to take up L D L via receptor-mediated endocytosis or to initiate the de novo synthesis of cholesterol. The LDL-receptor activity is enhanced in rapidly proliferating cells in culture, whereas few receptors are found on nondividing cellsJ '~9 Such being the case, malignant tissue might be expected to have higher LDL-receptor activity than normal tissue with a very low rate of replication.~~ Epidemiological studies have shown an increased risk of death from cancer in subjects with low plasma cholesterol levels, unrelated to nutritional s t a t u s . 6'8'22 In cases of acute leukemia, leukemic blood and bonemarrow cells from most patients have elevated LDLreceptor activities compared with white blood cells and bone-marrow cells from healthy subjects. A high receptor-mediated uptake and degradation of LDL by the 760
9 cholesterol metabolism
9 receptor
leukemic cells were shown to be the cause of hypocholesterolemia in cases of acute leukemia; 14'33hypocholesterolemia is a secondary phenomenon, due to the elevated LDL-receptor activity. Evidence for a high LDLreceptor activity in various tumor cells has also been obtained, both in vitro and in vivo. ~t,~6,26,32 Recently, LDL-receptor activity has also been identified in normal tissues and neoplasms of the central nervous system. Pitas, et al..27 noted that rat astrocytes have the LDL receptor, and Rudling, et al., 3~ showed that a human malignant glioma cell line (U-251MG) accumulates and degrades LDL by a saturable highaffinity process. It has been shown by morphological techniques that neuronal growth cones of pheochromocytoma rapidly take up LDL via L D L receptor-mediated endocytosis. J7 In addition, it was also reported that after a crush injury to the rat sciatic nerve, a cholesterol transfer mechanism was required for rapid membrane biogenesis during axon regeneration and remyelination. 3 Thus, the role of LDL receptor in cholesterol J. Neurosurg. / Volume 73/November, 1990
Lipoprotein cholesterol uptake by gliomas transport in neural tissues has been the focus of particular attention. In the present study, the interactions of LDL with human malignant glioma cell lines (U-251MG and KMG-5) were investigated biochemically and morphologically. Emphasis was placed on the dependency on LDL for the transfer of cholesterol into the human glioma cells. Materials and Methods
Lipoprotein Isolation and Modification Low-density lipoprotein (density (D) = 1.019 to 1.063) and lipoprotein-deficient serum (D > 1.25) were isolated by ultracentrifugation from fresh human plasma of normolipidemic subjects, after an overnight fast, as described. 25 The LDL was iodinated with the Bolton-Hunter reagent,* according to the manufacturer's instructions, to a specific activity of 270 cpm/ng of protein (~25I-LDL); other LDL was labeled with (3H)cholesterol oleate, as described elsewhere 23 to a specific activity of 43 dpm/ng of total cholesterol ester (3H-LDL). In addition, LDL was labeled with fluorescein isothiocyanate (FITC) or with 1,1 '-dioctadecyl3,3,3',3'-tetramethylindocarbocyanine (DiI), as follows. In brief, LDL concentrations of 4 to 5 mg/ml of protein were adjusted to a pH of 9 with bicarbonate buffer, and FITC solution, 5 mg/ml, was added at the ratio of 100 to 150 ug FITC/mg of LDL protein. The mixture was allowed to stand at room temperature for 4 hours, after which it was applied to a G-50 column equilibrated with a Tris-saline solution (pH 8.0) to remove the unbound fraction. The first eluting portion was collected and dialyzed against 0.15 M-NaC1, 0.02% ethylenediaminetetra-acetic acid (pH 7.4) (Buffer A). The molar ratio of FITC/LDL was about 2.3; FITC is covalently linked to the lysine residues of proteins. Therefore, the FITC-LDL indicated behavior of the protein moiety of LDL. Next, LDL (2 mg/ml) was mixed with lipoprotein-deficient serum (D > 1.25) and then DiI (3 mg/ml) in dimethyl sulfoxide was added. The mixture was allowed to stand at 37~ for 12 hours, after which it was ultracentrifuged for 24 hours at a density of 1.063 gm/ml to remove the unbound fraction. After centrifugation, the isolated fraction was dialyzed against Buffer A. Because DiI is a highly lipophilic molecule that can be noncovalently incorporated into LDL, the DiI-LDL indicated behavior of the cholesterol moiety of LDL. This procedure was based on previously reported methods. 9'29
Reductive Methylation of 3H-LDL or DiI-LDL Tritiated LDL or DiI-LDL (3 to 5 mg/ml of protein), dissolved in Buffer A, was diluted with 0.3 M sodium borate buffer (pH 9.0) to 1.5 times the original volume.
* Bolton-Hunter reagent (2200 Ci/mmol) obtained from New England Nuclear, Boston, Massachusetts.
J. Neurosurg. / Volume 73/November, 1990
Reductive melhylation of 5 mg of the samples was performed at 0~ by adding 1 mg of sodium borohydrate (zero tirr.e for reaction sequence) followed by six additions over 30 minutes of 1 ul of 37% aqueous formaldehyde. After the last addition of formaldehyde, the reaction mixture was dialyzed at 4~ against Buffer A for 20 hour,s. The specific activity of reductive methylated 3H-LDL was 48 dpm/ng of total cholesterol ester. This procedure was based on previously reported methods. 34
Binding Assay For the binding assay, 105 U-251MG or KMG-5 malignant glioma cells? were plated in Medium A (Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum) into 24-well (16 • 16 mm) polystyrene dishes on Day 0. On Day 2, the medium wa~,;removed, and the cells were washed three times with l~hosphate-buffered saline (PBS). The medium was then replaced with 500 #1 of fresh Medium B (DMEM supplemented with 1% human serum albumin) with an indicated or fixed concentration of ~2~ILDL, in the absence or presence of excess amounts of unlabeled LDL. After 3 hours of incubation at 4"C, the cells were washed three times with PBS, and 300 ul of 0.1 N NaC1 was added directly to the washed cells to determine the cellular binding of ~25I-LDL and cell protein concentration.
Endocytic Uptake of Radiolabeled LDL Fixed a~nounts of ~25I-LDL were incubated with the cells in 500 ul of Medium B at 37"C and, at various times, cellular binding and intracellular degradation were determined as described. ~5'24 The cells were also incubated with fixed amounts of 3H-LDL and reductive methylated 3H-LDL over time, or with indicated concentrations for 3 hours at 37~ After extensive washing of the cells, 300 ul of 0.1 N NaC1 was added for determination of cellular uptake and cell protein concentration, as described elsewhere. 23
Morphological Behavior of LDL The U-251MG glioma cells in Medium A were plated in an eight-well chamber, and on Day 2 the medium was replaced with Medium B. These cells were then incubated with 50 ug/ml of FITC-LDL and 15 tsg/ml of DiI-LDL in the absence or presence of excess amounls of unlabeled LDL for studies of protein moiety and the cholesterol moiety of LDL. After 30 minutes to 6 hoars of incubation at 37~ the cells were washed three times for 1 minute with PBS supplemented with 1% bovine serum albumin, then fixed with 4% paraformaldehyde for 1 hour at room temperature. Other U251MG glioma cells were incubated with 15 ug/ml of t The U-251MG cell line was kindly provided by Dr. J Yoshida, Department of Neurosurgery, Nagoya University Medical School, Nagoya, Japan; the KMG-5 cell line was established in our laboratories. 2~
761
//"
7OO0 .IS
%,,,,,, '" 9
e
\
c'~ _.1
10(X
i
0
i i i 1
125 I-
.100
LDL(.IJg/!11i)
9
200
'0
i 2 3 4 Bound(IJg/mg eeIlprotem/3hr)
FIG, 1, Binding of ~251-labeledlow-density lipoprotein (LDL) to U-251MG cells as a function of concentration. Left: The cells were incubated at 4~ for 3 hours with increasing amounts of J25I-LDL. After a wash with ice-cold phosphate-buffered saline, total binding (solid circles) was determined as described in the Materials and Methods section. Nonspecific binding (open circles) was determined by parallel incubation in the presence of more than 100-fold of unlabeled LDL. Specific binding (triangles) was plotted after correction for nonspecific binding, Right." Scatchard plot of the specific binding illustrated left. Each point represents the mean value of duplicate assays.
E 2C
/
e
jo.-----o
./7
/.oP1
0
2
3
.
~
[ncubcttion Time(hr)
5
6
FIG. 2. Endocytic uptake of ~25I-low-density lipoprotein (LDL) in U-251MG cells. The cells were incubated at 37"C with 20 pg-protein of ~25I-LDL in 500 t~l of Medium B, and examined at various times. Cellular binding (closed circles) and intracellular degradation (open circles) of ~25I-LDLwere determined, as described in the Materials and Methods section. Each point represents the mean value of duplicate assays. 762
F~o. 3. Behavior of cholesterol moiety of native and reductive methylated low-density lipoprotein (LDL) in the U251MG cells. The cells were incubated at 37~ with 10 ugprotein of 3H-labeled native LDL (dosed circles) or reductive methylated LDL (open circles), as a function of concentration for 3 hours in 500 ul of Medium B. CE = cholesterol ester. J. Neurosurg. / Volume 73/November, 1990
Lipoprotein cholesterol uptake by gliomas reductively methylated DiI-LDL for 6 hours. All of the cells were examined using fluorescence microscopy.
binding (Bmax) of 3.4 ug protein of LDL/mg cell protein.
LDL-Receptor Localization To study LDL-receptor localization, U-251MG cells were preincubated with Medium B overnight then fixed with 4% paraformaldehyde for 20 minutes at room temperature. They were then incubated at 1:50 with monoclonal anti-bovine LDL-receptor antibody (clone 7, which reacts with samples of human origin ~) as primary antibody for 1 hour at room temperature, and extensively washed in PBS. These cells were then incubated with FITC-labeled anti-mouse immunoglobulin G antibody for 40 minutes at room temperature. Fluorescence microscopy was used to examine LDL-receptor localization.
Endocytic Uptake of LDL in U-251MG Cells Endocytic uptake, an event occurring subsequent to the initial binding, was examined by incubating 125ILDL at 37~ with U-251MG cells. As shown in Fig. 2, the cell-associated 125I-LDL increased with time and reached equilibrium within 6 hours. The intracellular degradation of the ligand, as determined by an increase in trichloroacetic acid-soluble radioactivity in the medium, began at around 20 minutes after adding the radiolabeled ligand and continued to increase linearly with time thereafter. To further determine behavior of the cholesterol moiety of LDL, the U-251MG cells were incubated at 37~ as a function of concentration, or with a fixed concentration of 3H-cholesterol oleate-labeled LDL, followed by determination of the cell-associated radioactivity. Consequently, the amounts of the cell-associated cholesterol moiety of LDL increased as a function of concentration (Fig. 3) or with time (data not shown). The reductive methylation of LDL, which abolishes its ability to bind to the LDL receptor, resulted in a remarkable decrease in the cell association of the cholesterol moiety of LDL in the cells as a function of concentration (Fig. 3). These findings indicated that the uptake of cholesterol moiety of LDL was mediated via the LDL receptor present on the plasma membranes of the cells.
Results
Binding of LDL to U-251 MG Cells Figure 1 shows the binding of 125I-LDL to U-251MG cells at 4"C, as a function of concentration. The binding was inhibited by more than 60% by an excess of unlabeled LDL. The specific binding, exhibiting a typical saturation curve, indicated the presence of a receptor specific for LDL on the plasma membranes of U251MG cells. Scatchard analysis of this binding process gave a straight line, thereby suggesting a single mode of binding with an apparent dissociation constant (Kd) of approximately 38 ug/ml (1.7 • 10-8 M) and maximum
FIG. 4. Labeling of U-251MG cells with FITC-labeled low-density lipoprotein (LDL). The cells were incubated with 50 ug/ml of FITC-labeled LDL in Medium B at 37~ for 6 hours. Note the intense granular fluorescence pattern within the cell bodies as well as processe~;(left), and that the fluorescence labeling was blocked by incubating the cells with an excess amount of unlabeled LDL (right). • 500.
J. Neurosurg. / Volume 73 /November, 1990
763
M. Murakami, et aL
FIG. 5. •abe•ing•fU-25•MGce••swithDi•-•abe•ed••wdensity•ip•pr•tein(LDL).Thece••swereincubated with 15 ~g/ml of Dil-labeled LDL in Medium B at 37*C for 30 minutes (A) or 6 hours (B), with DiI-labeled LDL in the presence of an excess amount of unlabeled LDL (C), or with reductive methylated Dil-labeled LDL (D). After 30 minutes of incubation, the granular fluorescence pattern was definitely observed within the body of the cells as well as their processes (A and B). The fluorescence labeling was blocked in C and D. x 500.
Localization of F I T C - L D L or DiI-LDL in U-251MG Cells To determine behavior of the protein moiety and cholesterol moiety of LDL in U-251MG cells morphologically, FITC-LDL and DiI-LDL were used. An intense granular fluorescence pattern was observed within the cell body and processes of U-251MG cells after 6 hours of incubation of the cells with FITC-LDL at 37~ (Fig. 4 left). Fluorescence labeling could be blocked by incubating the cells with a 30-fold excess of unlabeled
764
LDL along with FITC-LDL (Fig. 4 right), thereby indicating that the FITC-LDL was internalized via the LDL receptor in the cells. To further determine the behavior of the cholesterol moiety of LDL, DiI-LDL was incubated with U251MG cells over time at 37~ Granular staining was seen in the body of the cells as well as in their processes after 30 minutes of incubation (Fig. 5A). An intense granular fluorescence pattern was observed within the body of the cells as well as their processes after 6 hours
J. Neurosurg. / Volume 73/November, 1990
Lipoprotein cholesterol uptake by gliomas Thus, the uptake of cholesterol moiety of LDL was mediated via the LDL receptor present on the plasma membrane of U-251MG cells.
LDL.Receptor Visualization in U-251MG Cells Indirect immunofluorescence studies using the monoclonal anti-LDL-receptor antibody showed that the LDL-:eceptor was visible within the cell body and on the surface of the cells (Fig. 6). In addition, a strongly positive immunoreaction was observed, especially in the dividing neoplastic cells.
FIG. 6. Low-density lipoprotein (LDL)-receptor visualization in U-251MG cells. After fixation with 4% paraformaldehyde for 20 minutes at room temperature, the cells were incubated with monoclonal anti-LDL-receptor antibody fbr 1 hour at room temperature. Note that the LDL receptor can be visualized within the cell body and on the surface of the cells, x 420.
Uptake of L D L Cholesterol in U-251MG and in K34G..5 Cells To observe differences in the uptake of LDL cholesterol in the U-251MG and KMG-5 cells (the doubling time of which is approximately 25 and 100 hours, respeclively), endocytic uptake of LDL cholesterol was examined by biochemical techniques (Fig. 7). Two kinds of cells were incubated at 37~ with 3H-LDL over time, followed by determination of the cell-associated cholesterol moiety of LDL. Uptake of the cholesterol moiety of LDL in the U-251MG cells was over threefold higher than in the KMG-5 cells, after incubation. Discussion
FIG, 7. Comparable uptake of cholesterol moiety of lowdensity lipoprotein (LDL) in U-251MG cells and KMG-5 cells. U-251MG cells (circles) or KMG-5 cells (triangles) were incubated with t0 ug-protein of 3H-LDL at 37~ over time in 500 ~1 of Medium B. Uptake of the cholesterol moiety of LDL was determined as described in the Materials and Methods section. Each point represents the mean value of duplicate assays. CE = cholesterol ester.
of incubation (Fig. 5B). Fluorescence labeling could be blocked by incubating the cells with a 30-fold excess of unlabeled LDL (Fig. 5C). Furthermore, reductive methylation of the apoprotein moiety of LDL, a procedure known to prevent receptor-mediated binding of LDL, prevented most of the binding and internalization of the DiI-LDL (Fig. 5D).
J. Neurosurg. / Volume 73/November, 1990
The results of this study gave evidence that the LDL recepto:: is present on the plasma membranes of malignant glioma cells (U-251MG), and that the cholesterol moiety of LDL is endocytosed predominantly via the LDL receptor in the cells. The literature on LDL metabolism in tumors of the central nervous system is sparse. ~'~~Rudling, et al., 3~found that a human malignant glioma cell line (U-251MG) accumulates and degrades the apoprotein moiety of LDL by a saturable high-affinity process; however, little is known of the cholesterol metabolism of LDL in human glial neoplasms. Hence, we focused our study on the cholesterol metabolism of LDL in human malignant gtioma cell lines. It is well known that regulation of cellular cholesterol metabolism depends on sequential cell-surface binding, internalization, and intracellular degradation of LDL. ~2.~3
L D L Binding in Glioma The cell-surface binding of ~2~I-LDL was studied at 4~ as sl:Lown in Fig. 1. The data provide evidence that the plasma membranes of a human malignant glioma cell line (U-251MG) possess the high-affinity LDL receptor (Kd approximately 38 ug/ml). This is in good agreement with the study of Rudling, eta/.; 3~the temperature at which the experiments were conducted could probably produce the different values of Kd o f the LDL receptor in U-251MG cells seen between their data and ours. Evidence that the LDL receptor was visualized within the cell bcdy as well as on the surface of the cells (Fig. 6) is interpreted to mean that the internalization of the 765
M. Murakami, et al. LDL receptor into the cells occurs after binding to LDL, and/or that the binding of LDL receptor to the cell surface occurs following production of the LDL receptor in the Golgi apparatus. 2~
LDL Cholesterol Uptake in Glioma It is of interest to know the extent to which transport of the cholesterol moiety of LDL into the human malignant glioma cells depends on the LDL receptor. For elucidation, lysine residues of LDL were modified by reductive methylation, a procedure known to block the interaction of LDL with the LDL receptor. 34 Modification of lysine residues by reductive methylation was shown to be an important procedure to probe the role of lysine in the recognition site since, unlike other modifications such as diketone and carbamoylation, this reaction did not alter the charge of the lysine or the net charge of LDL and yet was effective in blocking the binding activity of LDL. The lipid compositions were essentially identical for untreated LDL and for LDL modified by reductive methylation (data not shown). 34 In addition, DiI-lipoproteins were shown to have the same chemical composition and physical properties as the native lipoproteins. 28'29 Consequently, biochemical and morphological evidence using reductively methylated LDL demonstrated that the amount of transport of cholesterol moiety of modified LDL into malignant glioma cells significantly decreased. Thus, it seems likely that more than 80% of the cholesterol moiety of LDL is endocytosed via the LDL receptor (Fig. 3). In vivo experiments revealed that about 80% of plasma LDL is cleared by the LDL receptor, and the
100
,
,
,
,
,
8O P
g u
60
Implications of LDL Cholesterol Uptake in Glioma It is also of interest to know the implications o f uptake of cholesterol moiety of LDL in malignant glioma cells. For this study, we used other malignant glioma cells (the KMG-5 cell line 2~) which are known to divide less rapidly than U-251MG cells in culture systems. The doubling time of KMG-5 and U-251MG cells is 100 and 25 hours, respectively. 2'2~ The binding of 125I-LDL to KMG-5 at 4"C was also inhibited in a dose-related manner by unlabeled LDL; however, the rate of inhibition was higher in the U-251MG than in the KMG-5 cells, suggesting that LDL-receptor activity might be higher in U-251MG than in KMG-5 cells (Fig. 8). As shown in Fig. 7, uptake of the cholesterol moiety of LDL in the U-251MG cells was three times than in the KMG-5 cells. Thus, malignant glioma cells that grow more rapidly and have a higher activity of LDL receptor might need a larger amount of the cholesterol moiety of LDL. The cholesterol moiety of LDL is rapidly endocytosed in growth cones of pheochromocytoma cells, 17 and the rate of LDL uptake in cervical cancer cells ~ or leukemia cells33 is much greater than that in corresponding non-neoplastic cells. In addition, it is shown that the extracytoplasmic domain of the LDL receptor contains a region that is 38% identical with a 96 amino acid sequence in the precursor to the epidermal growth factor?' Thus, large amounts of receptor-mediated uptake of the cholesterol moiety of LDL in malignant glioma cells may correlate with cell growth. Acknowledgments
0 139
We are grateful to Drs. K. Takata and T. Ohta for valuable discussions and suggestions, and to M. Ohara for editorial assistance.
_J t23
-~, 20
160 ~
~
~0
Unlobeled LDL(pg/rnl) FIG. 8. Competition by unlabeled low-density lipoprotein (LDL) for ~:5I-LDL binding to U-251MG cells or KMG-5 cells. U-251MG cells (closed circles) or KMG-5 cells (open circles) were incubated with 10 ug-protein of ~25I-LDLin 500 ul of Medium B for 3 hours at 4~ The cell-associated radioactivity was determined as described in the Materials and Methods section.
766
remaining 20% is removed by nonspecific mechanisms, determined using glucosylated L D L ) 8 In addition, in vitro experiments raised the possibility of dual mechanisms for the uptake and regulation of LDL; in fibroblasts, almost all the LDL was endocytosed and degraded by a receptor-mediated mechanism while, in hepatocytes, even LDL modified by acetoacetylation, acetylation, and reductive methylation was endocytosed and degraded at 30% to 50% of the level of native LDL. 7 Therefore, the mechanism for the uptake and regulation of LDL in glioma cells seems to be similar to that in fibroblasts.
References 1. Beisiegel U, Schneider WJ, Goldstein JL, et al: Monoclonal antibodies to the low density receptor as probes for study of receptor-mediated endocytosis and the genetics of familial hypercholesterolemia. J Biol Chem 256: 11923-11931, 1981 2. Bigner SH, Bullard DE, Pegram CN, et al: Relationship of in vitro morphologic and growth characteristics of established human glioma-derived cell lines to their tu-
J. Neurosurg. / Volume 73/November, 1990
Lipoproteincholesterol uptake by gliomas 3.
4.
5. 6. 7.
8. 9. t0.
11.
12. t3.
14. 15. 16. 17. 18.
19.
morigenicity in athymic nude mice. J Neuropathol Exp Neurol 40:390-409, 1981 Boyles JK Zoellner CD, Anderson LJ, et al: A role for apolipoprotein E, apolipoprotein A-l, and low density lipoprotein receptors in cholesterol transport during regeneration and remyelination of the rat sciatic nerve. J Clin Invest 83:1015-1031, 1989 Brown MS, Goldstein JL: Analysis of a mutant strain of human fibroblasts with a defect in the internalization of receptor-bound low density lipoprotein. Cell 9:663-674, 1976 Brown MS, Goldstein JL: The low-density lipoprotein pathway and its relation to atherosclerosis. Annu Rev Biochem 46:897-930, 1977 Chao FC, Efron B, Wolf P: The possible prognostic usefulness of assessing serum proteins and cholesterol in malignancy. Cancer 35:1223-1229, 1975 Edge SB, Hoeg JM, Triche T, et al: Cultured human hepatocytes. Evidence for metabolism of low density lipoproteins by a pathway independent of the classical low density lipoprotein receptor. J Biol Chem 261: 3800-3806, 1986 Feinleib M: Review of the epidemiological evidence for a possible relationship between hypocholesterolemia and cancer. Cancer Res 43:2503-2507, 1983 Fothergill JE: Huoroehromes and their conjugation with proteins, in Nairn RC (ed): Fluorescent Protein Tracing, ed 3. London: E & S Livingstone, 1969, pp 5-34 Gal D, MacDonald PC, Porten JC, et al: Effect of cell density and confluency on cholesterol metabolism in cancer cells in monolayer culture. Cancer Res 41: 473-477, 1981 Gal D, Ohashi M, MacDonald PC, et al: Low density lipoprotein as a potential vehicle for chemotherapeutic agents and radionucleotides in the management of gynecological neoplasms. Am J Obstet Gynecol 139: 877-885, 1981 Goldstein JL, Basu SK, Brunschede GY, et al: Release of low density lipoprotein from its cell surface receptor by sulfated glycosaminoglycans. Cell 7:85-95, 1976 Goldstein JL, Brown MS: Binding and degradation of low density lipoproteins by cultured human fibroblasts: comparison of cells from a normal subject and from a patient with l~omozygous familial hypercholesterolemia. J Biol Chem 249:5153-5162, 1974 Ho YK, Smith RG, Brown MS, et al: Low density lipoprotein (LDL) receptor activity in human acute myelogenous leukemia cells. Blood 52:1099-I 1I4, I978 Horiuchi S, Murakami M, Takata K, et al: Scavenger receptor for aldehyde-modified proteins. J Biol (?hem 261:4962-4966, 1986 Hynds SA, Welsh J, Stewart JM, et al: Low-density lipoprotein metabolism in mice with soft tissue tumours. Biochim Biophys Acta 795:589-595, 1984 Ignatius MJ, Shooter EM, Pitas RE, et al: Lipoprotein uptake by neuronal growth cones in vitro. Science 236: 959-962, 1987 Kesaniemi YA, Witztum JL, Steinbrecher UP: Receptormediated catabolism of low density lipoprotein in man. Quantitation using glycosylated low density lipoprotein. J Clin Invest 71:950-959, 1983 Kruth HS, Arigan J, Gamble W, et al: Effect of cell density on binding and uptake of low density lipoprotein by human fibroblasts. J Cell Biol 83:588-594, 1979
J. Neurosurg. / Volume 73/November, 1990
20. Mahley RW: Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science 40: 622-630. 1988 21. Mihara "(: Establishment and characterization of 5 cell lines derived from human brain tumors. Kumamoto Med J 39:25-41, 1986 22. Miller SR, Tartter PI, Papatestas AE, et al: Serum cholesterol and human cancer. JNCI 67:297-300, 1981 23. Murakami M, Horiuchi S, Takata K, et al: Distinction in the mode of receptor-mediated endocytosis between high density lipoprotein and acetylated high density lipoprorein: evklence for high density lipoprotein receptor-mediated cholesterol transfer. J Biochem 101:729-741, 1987 24. Murakami M, Horiuchi S, Takata K, et al: Scavenger receptor for malondialdehyde-modified high density lipoprotein on rat sinusoidal liver cells. Biochem Biophy Res Commun 137:29-35, 1986 25. Murakami M, Ushio Y, Morino Y, et al: Immunohistochemical localization of apolipoprotein E in human glial neoplasms. J Clin Invest 82:t77-188, 1988 26. Norata G, Canti G, Ricci L, et al: In vivo assimilation of low density lipoproteins by a fibrosarcoma tumour line in mice. Cancer Lett 25:203-208, 1984 27. Pitas RE, Boyles JK, Lee SH, et al: Astrocytes synthesize apolipogrotein E and metabolize apolipoprotein E-containing lipoproteins. Biochim Biophys Acta 917: 148-161, 1987 28. Pitas RE, Innerarity TL, Mahley RW: Foam cells in explant ~of atherosclerotic rabbit aortas have receptors for beta-ve~:y low density lipoproteins and modified low density lipoproteins. Arteriosclerosis 3:2-12, 1983 29. Pitas RE, Innerarity TL, Weinstein JN, et al: Acetoacetylated lipoproteins used to distinguish fibroblasts from macrol:,hages in vitro by fluorescence microscopy. Arteriosclerosis 1:177- 185, 1981 30, Rudling M J, Collins VP, Peterson CO: Delivery of aclacinomycin A to human glioma cells in vitro by the lowdensity lipoprotein pathway. Cancer Res 43:4600-4605, 1983 31. Russell DW, Schneider W J, Yamamoto T, et al: Domain map o:~ the LDL receptor: sequence homology with the epidermal growth factor precursor. Cell 37:577-585, 1984 32. Sato JD, Kawamoto T, Okamoto T: Cholesterol requirement of P3-X63-Ag8 and X63-AgS.653 mouse myeloma cells fi~r growth in vitro. J Exp Med 165:1761-1766, 1987 33. Vitols S, Gahrton G, Ost A, et al: Elevated low density lipoprotein receptor activity in leukemic cells with monocytic differentiation. Blood 63:1186-1193, 1984 34. Weisgraber KH, Innerarity TL, Mahley RW: Role of the lysine residues of plasma lipoproteins in high affinity bindirg to cell surface receptors on human fibroblasts. J Biol Chem 253:9053-9062, 1978 Manuscript received September 29, 1989. Accepted in final form April 19, 1990. This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture of Japan. Address reprint requests to: Masaji Murakami, M.D., Department of Neurosurgery, Kumamoto University Medical School, Honjo 1-1-I, Kumamoto 860, Japan.
767
