- Email: [email protected]
Carbon disulfide-induced modification and cytotoxicity of low-density lipoproteins
Pergamon
To.uicology in Vitro IO (1996) 423429
Carbon Disulfide-induced Modification and Cytotoxicity of Low-density Lipoproteins T. WRONSKA-NOFER*t, tThe
W. LAURMANt, J.-R. and M. WALTERf
NOFERf,
U. SEEDORF$
Nofer Institute of Occupational Medicine, Lodz. Poland and tlnstitut fur Klinische und Laboratoriumsmedizin. WestfZlische Wilhelm+Universitat Miinster, Germany
Chemie
(Accepted IS March 1996) Abstract-The secondary oxidation of biologically modified low-density lipoproteins (LDL) was demonstrated to contribute to the cytotoxicity and thereby to the atherogenicity of modified lipoproteins. Previously we have shown that chemical modification of LDL by carbon disulfide (CS) mimicked the naturally occurring process of LDL modification. In the present study the cytotoxicity of CS,-modified LDL and their susceptibility to the secondary oxidative modification was investigated. The cytotoxicity of CS,-modified LDL did not significantly exceed that of native LDL. However, the Cu’+ -oxidized form of C&modified LDL revealed to be more cytotoxic than oxidized native LDL. Oxidized CS-modified LDL presented with altered physicochemical properties including derivatization of protein amino and - SH groups, increased negative charge, and electrophoretic mobility which exceeded that of oxidized native LDL. The secondary oxidative moditicatton of C&modified LDL involved the process of Cu’+ binding to C&-derived dithiocarhamate -SH groups followed by covalent modification of -SH groups by products of lipid peroxidation. Taken together, these finding suggest that secondary oxidation of C!&-modified LDL may contribute to the atherogenic effect of the chronic occupational exposure to CS,. Copyright 4-S 1996 Elsevier Science Ltd.
INTRODUCTION
The biological modification of low-density lipoproteins (LDL) plays an essential part in the initiation of atherosclerotic processes (Brown and Goldstein, 1990; Steinberg ef ul., 1989). There is evidence that lipid peroxidation is a key event in LDL modification in zkw. Although mechanisms involved in the oxidation of LDL are not fully understood, its occurrence in the extracellular environment of the arterial wall may contribute to the LDL atherogenicity (Palinski et al., 1989). Modified LDL contain ligands recognized by scavenger receptors and subsequently accumulate in macrophages, thus contributing lo the foam cell formation (Brown and Goldstein, 1983). In addition, modified LDL were demonstrated to be highly cytotoxic to endothelial cells, smooth muscle cells, as well as macrophages, and this is thought to have a crucial role in the development of atherosclerotic lesion (Hessler et al., 1983: Liu et al., 1991). *Author for correspondence at: The Nofer Institute of Occupational Medicine, 8 SW. Teresy Str. 90-950 Lodz, Poland. Abhrez~iations: BHT = butylated hydroxytoluene: DTNB = 5.5-dithio-his-2nitrobenzoic HUVEC = acid; human umbilical vein endothelial cells; LDH = lactic dehydrogenase; LDL = low-density lipoprotein; LDLbuffered saline; TBARS = CS! = phosphate substances; T-NBS = thiobarbituric acid-reactive trinitrobenzenesulfonic acid. 0887-2333/96/$15.00 + 0.00 Copyright PIf SO887-2333(96)00029-X
The recent in citro studies of Laurman et a/. (1989) revealed that CS2 is able to modify LDL. CS? still attracts the particular attention of toxicologists as a chemical risk factor for coronary heart disease in workers in the synthetic fibre industry (WronskaNofer and Laurman, 1987). The earlier studies of Wronska-Nofer (1969) and Wronska-Nofer et al. (1978) demonstrated that the atherosclerotic changes in the arterial wall produced by CS, may be due to disturbances in the cholesterol metabolism such as increased cholesterol synthesis in the liver and deposition in the artery wall. Nevertheless, the definite mechanism of C&induced atherosclerosis remained unclear. Laurman ef al. (1989) showed that C&modified LDL (LDL-CS,) physicochemically and functionally resemble oxidatively modified LDL (Ox-LDL). LDL-CS? are taken up by the scavenger receptor and accumulate in macrophages. It is likely that LDL-CS2 undergo further modification. Secondary modification of biologically modified LDL has been demonstrated previously. For example, LDL modified by the glycosylation or by the formation of complexes with intimal proteoglycans become unstable particles more susceptible to secondary modification by oxygen free radicals than native LDL (Camejo et al., 1991; Hunt et al., 1990). In the present study, we examined whether LDL chemically modified by CS2 are cytotoxic. In addition, we investigated whether secondary oxidative modification of LDL-CS2 results in the
,C 1996 Elsevier Science Ltd. Al! rights reserved.
Printed
in Great
Britain
T. Wronska-Nofer rt ai
424
alteration of the physicochemical properties and cytotoxicity. We assumed that both modification of LDL by CS? and the secondary modification of LDL-CS, by oxidation could contribute to the development of atherosclerotic lesions. MATERIALS AND METHODS
Prepuration
of LDL
und LDL-C’S,
LDL (density I ,019-l ,063 g/ml) was isolated from the plasma of healthy subjects by sequential ultracentrifugation according to the method of Have1 et al. (1955). LDL was dialysed against IO mM phosphate buffered saline (PBS), pH 7.4. for 24 hr. sterilized by filtration, and stored under nitrogen at 4“C in the dark. EDTA (0.1%. M.:v) was present throughout all steps of preparation. Modification of LDL by CS2 was performed by dialyzing LDL (5 mg/ml) against C&-saturated PBS, pH 7.4 for 4 hr. Free CSI was removed from the preparation by extensive dialysis against PBS containing EDTA (0.1 %, w/v). Osidution
of LDL
und LDL-C‘S,
Before oxidation LDL and LDL-CSI were dialysed for 24 hr at 4 C against de-aerated and nitrogen-saturated PBS to remove EDTA. The LDL and LDL-CS, were diluted with EDTA free. oxygen saturated IO mM PBS, pH 7.4. Oxidation of LDL was initiated by addition of freshly prepared CuSO, at a room temperature and monitored spectrophotometritally at 234 nm. The final concentrations were 0.25 mg/ml of LDL and 1.66 PM CuSO,. Assessment qf c~~~toto.\-icit~~
Human umbilical vein endothelial cells (HUVEC’) were purchased from PromoCell (Heidelberg, Germany) and maintained in endothelial cell basal medium (PromoCell) supplemented with foetal bovine serum, human basic fibroblast growth factor. human epidermal growth factor. endothelial cell growth supplement/heparin. hydrocortisone, amphotercin B and gentamicin at 37 C in a humidified atmosphere containing 5% CO,. All cell culture supplements were provided by PromoCell and used in concentrations recommended by the supplier. HUVEC were grown to confluency in 96-well dishes. The cytotoxicity was assessed using CytoTox 96’” Heidelberg, assay (Promega/Serva. cytotoxicity Germany) according to the man&cturer’s instrucTable Relative
I.
Phwcochemlcal
electrophoretic
mobtlity
tions. This assay quantitatively measures activity of lactate dehydrogenase (LDH) released on cell lysis following incubation with the cytotoxic agents. The absorbance data were collected at 490 nm with the Dynatech MR600 96 well plate reader and expressed as percentage of total LDH activity released from cells following treatment with Lysis Solution supplied by the manufacturer. In addition, the cell viability was estimated using Trypan blue exclusion assay. Briefly, HUVEC grown to confluency in 24-well dishes were incubated with the cytotoxic agents for the appropriate time. The cells were then trypsinized and the cell suspension was incubated for 5 min with Trypan blue (0.25%, v/v). The number of dye-excluding cells was determined in the haemocytometer. The results are presented as percent of viable cells. Other rr1ethod.s
LDL-protein
was assayed by the method of Lowry serum albumin as the standard. Electrophoresis of LDL was performed in I % agarose gels in barbital buffer. pH 8.6 (Laurman <>iui., 1989). Lipoproteins were visualized by staining with Fat Red 7B (Sigma, Deisenhofen, Germany). Free amino groups of LDL were determined by the trinitrobenzenesulfonic acid (TNBS) calorimetric method (Habeeb. 1966) using HCI-lysine as a standard. Thiobarbituric acid-reactive substances (TBARS) in the LDL samples were determined by the method of Yagi (1982) using malondialdehyde bis-dimethylacetal as a standard. The concentration of -SH groups in LDL-CS? was determined by the method of Cardin et al. (1982) with 5,5-dithiobis-2nitrobenzoic acid (DTNB).
et 01. (1951) using bovine
RESULTS
The physicochemical properties of modified LDL are presented in Table I. LDL were modified with CS, by dialysis against CS1-saturated buffer in neutral pH (7.4). Ox-LDL presented with an increased negative charge as demonstrated by the enhanced electrophoretic mobility in a 1% agarose gel. CS1-modified LDL migrated further towards the anode than the native LDL though not so far away as Ox-LDL. The oxidized LDL-CS? exhibited the most pronounced electrophoretic mobility. However. prooert~es of modified _~
Free amino groups
LDL Free -SH nmoljmg
LDL (nattve) Ox-LDL LDL-CS: Oxidized LDL-CS: *P .z 0.05: **P < 0.01: ***P
1.00 2.07 2 0.17 I .37 + 0.09 2.24 5 0.21 <
509 176 312 154
ND = not detectable 0.001 I‘ native LDL. Student’s r-test.
* + * i
78 21** 53’ 27’
groups
TBARS
protein
7*2 Ski 131 * 7”’ ND
ND 16.2 * I.1 ND 14.7 k 1.8
425
Oxidation of C&-modified LDL propagation phase
lag phase
degradation phase
2.4 1 E 2.2E 2.0 ZJ N al 18” c : G n P a
1
.6-
1
.4 -8
1.0
-
I
0
60
I time
120 [min]
I
1
180
240
Fig. 1. Time course of the CL?+-catalysed oxidation of LDL. LDL (0.25 mg/ml) was incubated with 10 PM CuSOl at room temperature. The absorbance at 234 nm reflecting conjugated dienes formation during lipid peroxidation was monitored. Values presented arc computed from triplicate determinations.
the migration distance of LDL-CS! did not represent the addition of the migration distances of Ox-LDL and LDL-CSZ. Thus, the oxidative and C&mediated modifications were apparently not self dependent. The content of free amino groups in LDL-CS, was 30% lower than in native LDL, whereas in oxidatively modified LDL their amount decreased by 75%. In oxidized LDL-CS, the decrease of free amino groups was comparable to that found in Ox-LDL. The decrease of free amino groups in LDL-CS, was paralleled by an increase in -SH groups. The newly-formed -SH groups most likely originated from dithiocarbamates of CS, origin (Lam and di Stephano, 1986). No changes of the -SH group level in oxidatively modified LDL were observed, whereas the content of -SH groups in oxidized LDL-CS, was reduced below the lower detection level. These data indicate, that oxidative and C.‘$-mediated modifications of LDL are not self-dependent and that dithiocarbamate-derived -SH groups of LDL-CSr are further modified/derivatized by oxidation. Additionally, native and CS? modified LDL were analysed for TBARS, an index of lipid peroxidation. The results demonstrated similar increase in TBARS values in native and C&modified LDL following oxidation indicating that the degree of lipid oxidation was similar in both instances. The kinetics qf’ the oxidation
qf LDL-CS,
To further characterize LDL-CSr, its susceptibility to the Cur + -stimulated oxidation was evaluated. The kinetics of the oxidation was followed by the continuous monitoring of the absorbance at 234 nm, which reflects the extent of conjugated dienes formation during oxidation of polyunsaturated fatty acids in LDL lipids to fatty acid hydroperoxides with
double bonds. The recorded time-kinetics of the native LDL oxidation presented three typical successive phases (Fig. 1): (I) a lag phase with slow increase of the 234-nm absorbance reflecting the minimal rate of lipid peroxidation and conjugated diene formation due to protective effects of LDL antioxidants; (2) a propagation phase characterized by the rapid increase of the absorbance related to the burst of lipid peroxidation and diene formation; and (3) a decomposition phase, in which the labile lipid hydroperoxides with conjugated double bonds decompose to a great variety of products including reactive aldehydes with cytotoxic properties. As shown in Fig. 2, the profile of the LDL-CS, oxidation monitored by conjugated diene formation differed from that of native LDL. Addition of Cu’+ to LDL-CSI caused rapid increase of the 234-nm absorption reflecting the formation of the short-lived Cu-mercaptide chromophore (Fig. 2). This was followed by the gradual decline of the absorption during the first hour of the reaction. Approximately 40% of the initial -SH group content disappeared in LDL-CS?, when Cu” was added (Fig. 3). As the experiment was done in the nitrogen-saturated medium, the observed decrease in -SH groups should be ascribed to Cu’ + binding rather than to covalent modification by lipid peroxidation products. This was further confirmed by the observation that addition of EDTA (O.l%, w/v) to LDL-CS>-Cu’ + complex partially recovered free -SH groups, most likely due to removal of copper from LDL-CS?. By contrast, in the oxygen-containing medium, the content of -SH groups in LDL-CSr particles incubated with CL?+ decreased below the lower detection limit during the first hour of the reaction (Fig. 4). This process was blocked by the conjugated
T Wronska-Nofer et al.
426
2.4
-
E2.2 c 02.0 0 (u al.8 0 c zl.6
-
0
::I.4 m
-
1.2
-
1.0
’
’
I
0
60
I time
120 (m in]
I
1
180
240
Fig. 2. Cu’ -catalysed oxldatlon ot’ <‘ST-modllied LDL I native LDL. CS,modified LDL (0.25 mg/ml) and native LDL (0.25 mg;ml) were Incubated with IO FM CuSO, at room temperature. Oxidation was followed by monitoring the ahsorhance at 234 nm. Values presented are computed from triplicate determinations. Note in the case oP LDL-C’S, a rapid increase of the absorbance on the addition of Cu’(Cu-mercaptide chromophore) followed b) gradual decline and elongation of the lag phase of conjugated dienek generation.
addition of 25 PM butylated hydroxytoluene (BHT) which inhibits lipid peroxidation by way of free radical scavenging. These findings suggest that to -SH groups in LDL-C’S: ih binding of Cu” followed by the gradual ~ SH group modification by lipid peroxidation products. Cytototosicit~~
of
o.uidrrtiwl~~ nzod$frrtl LDL
rrtrd
LDL-CS:
Because oxidized LDL were shown to be cytotoxic and since this process is likely to be involved in the pathogenesis of atherosclerosis, the cytotoxicity of modified LDL towards endothelial cells was examined. HUVEC grown to confluency were incubated with 0.3 m&/ml native LDL, LDL-CS,. Ox-LDL. oxidized LDL-CS? or with buffer as a control. The
”
LDH release in control cells was 7.4% + 0.16% (mean i: SEM; n = 8) and was not significantly changed in the presence of native LDL or LDL-CS2. By contrast. oxidatively modified LDL revealed to be highly cytotoxic (Fig. 5). In addition, the ability of oxidatively modified LDL-C& to exert cytotoxic effect towards HUVEC exceeded significantly that of oxidatively modified native LDL. To further investigate the cytotoxic effect of modified LDL, the cell viability was estimated using Trypan blue exclusion assay. The percentage of viable cells at the begining of the experiment was 92 k 1.5 (mean i SEM; n = 4). and was not significantly changed after incubation with buffer or with 0.3 mg;ml native LDL or C&modified LDL. By contrast, both Ox-LDL and oxidized LDL-CSI
100
g 0’ 0
80
;
80
g a
40
5b
20 0
LDL LDL-CS, Fig. 3. Effect of EDTA on Cu’ _-mediated depletion of the content of free - SH groups in CS>-modified LDL. Native and C&-modified LDL (0.25 mg/ml) were ureincubated with 10 UM CuSO, in the medium saturated with nitrogen to prevent lipih peroxidation and next supplemented with 0.1% EDTA (w/v) or buffer as reference. Content of - SH group was determined using DTNB reagent and expressed as relative value Y. untreated, native LDL. Values are mean ) SEM for three separate determinations.
Oxidation of C&-modified LDL .E
421
(Brown and Goldstein, 1990; Luck and Fruchart, 1991; Steinberg ef al., 1989). The oxidation of LDL has been shown to occur in vivo, and the positive correlation between the level of Ox-LDL and extent of atherosclerotic lesions as well as an inverse correlation with blood level of antioxidants has been found (Lavy et al., 1991). Previously, we reported that carbon disulfide can modify LDL (Laurman et al., 1989) yielding a particle that physicochemically and functionally resembles oxidized native LDL. In this paper we present evidence that Cu’ +-induced oxidation of LDL-CS, results in a new particle with altered physicochemical properties and more cytotoxic than oxidatively modified native LDL. In this respect, LDL-CSZ resembles glycosylated or proteoglycanmodified LDL which also display increased cytotoxicity and atherogenicity after secondary oxidative modification (Camejo et al., 1991; Hunt ef nl., 1990). The changes of the functional properties of oxidized LDL-CS2 may be linked to the considerably increased negative charge of the particles as inferred from an increased electrophoretic mobility and decreased level of positively charged amino groups. It is possible that Cu’ +-oxidized LDL-CS, particles generate epitopes which are ligands for the macrophage scavenger receptor. In this context, it is noteworthy that the degree of positively charged amino group modification in LDL determines the extent of the oxidized LDL particle recognition by scavenger receptors and its delivery in an unregulated manner to the target cell (Brown and Goldstein, 1983). In comparison with native LDL, the time course of CL?+-catalysed oxidation of C&modified LDL is characterized by an altered profile of conjugated of atherosclerosis
0
‘*O
time [mm] -1~0
phase 2
Fig. 4. Protective effect of BHT against depletion of the - SH group content in Cu’ +-oxidized LDL-CS?. LDL-CS? (0.25 mg/ml) was incubated in the presence or absence of BHT (25 PM) in the medium saturated with oxygen. Oxidation of LDL-CS? was initiated by addition of CuS04 (10 PM). At the indicated time -SH groups content was determined. Value represents mean f SD from three separate determinations.
considerably decreased cell viability (Fig. 6). Again, the effect of oxidatively modified LDL-CS? exceeded significantly that of oxidatively modified native LDL. DISCUSSION
Increasing evidence suggests that oxidative modification of LDL plays a crucial part in the pathogenesis
m D
El J 0
3-l
diene formation. This may be related to the high content of dithiocarbamate-derived - SH groups in the LDL-CS, particle. The rapid binding of Cu’+ by
Ox-LDL oxidized LDL-CS,
6h
9h
12h
time Fig. 5. Cytotoxicity of modified LDL. Ox-LDL and oxidized LDL-CS, (0.3 mg/ml) were incubated with HUVEC grown to confluency at 37°C for 3,6,9 and I2 hr, and the activity of LDH released to the medium was determined as described in Materials and Methods. Results were expressed as percent of total LDH activity released to medium from the cells treated with a Lysis Solution. Values are the mean + SEM for eight determinations in one experiment and are representative for three independent experiments. Note the significant difference between the value for oxidized LDL-CS? and oxidized native LDL at *P < 0.05 and **P < 0.01 (Student’s /-test).
428
T. Wronska-Nofer
3h
6h
cl al
9h
12h
time Fig. 6. Effect of Ox-LDL and oxidized LDL-CSI on HUVEC vlabdity. Ox-LDL and oxidzed LDL-CS? (0.3 mg/ml) were incubated with HUVEC grown to confluency at 37 C for 3. 6, 9 and 12 hr, and the cell viability was assessed with Trypan blue exclusion test as described in Materials and Methods. Values are the mean f SEM for four determinations In one experiment and are representative for three independent experiments. Note the significant difference between the value for oxidized LDL-CSI and oxidized native LDL at *P < 0.05 and **P < 0.01 (Student’s r-test).
dithiocarbamate-derived
chromophore may
be
thiols
with a pronounced suggested
that
yields
a short-lived
UV absorbance.
transfer
of
Cu”
It to
in apoprotein moleties of LDLCS, located close to LDL-lipids is essential for copper-catalysed oxidation of LDL-CSI. The profile of changes in the 234-nm absorbance of conjugated dienes in the course of Cu’ +-catalysed peroxidation of LDL-CSZ accounts for this assumption. The Cu’- -catalysed oxidative modificatron of LDL-CS, consists in the secondary derivatization of dithiocarbamate - SH groups. This derivatization may be explained by binding of CL?. to thiols. We demonstrated that in the nitrogen-saturated medium a number of SH groups in LDL-CS, is significantly reduced on addition of Cu’ + and that this is partially reverted by addition of EDTA. It is well established that -SH groups of dithiocarbamates can avidly bind transition metals (Sunderman. 1992). Morcovcr. copper ions attached to proteins, enzymes or cellular surfaces can still retain their highly pro-oxldative redox potential. In this context, it is worth noting that thiol-containing agents including cysteine, homocysteine and glutathione were demonstrated to induce LDL modification by a reaction requiring the presence of copper. The process of LDL modification includes oxygen- and sulfur-centred free radicals generated during metal-catalysed oxidation of thiols (Heinecke rf ul.. 1993; Sparrow and Olszewski, 1993). It has been postulated that formation of LDL-Cu complexes has an essential role not only in the oxidative modification but also in the cytotoxicity of Ox-LDL (Kuzuya et ul.. 1991; Thomas, 1992). Hence, it is possible that increased cytotoxicity of the oxidatively modified LDL-CS2 is related to the extensive binding of Cu’+ to LDL-CS, and the subsequent - SH group derivatization. The attack of copper-generated reactive oxygen species on dithioweak-binding
sites
carbamate-derived -SH groups in LDL-CS, may lead to the creation of thiol radicals and their cytotoxic derivatives. Another explanation for thiol derivatization comes from the observation that it can be prevented by BHT, the lipid peroxidation inhibitor (cf. Fig. 4). Among the products of lipid peroxidation, which could secondarily react with -SH groups, the aldehydes are the most likely candidates. Aldehydes, which possess a$-unsaturated CH = CHCHO units react with nucleophiles such as -SH to give S-conjugates, CHSCH&HO. These conjugates have been shown to be cytotoxic and to exist in native LDL particles albeit at a low concentration (Hessler et al., 1983; Luck and Fruchart, 1991). The toxicity of oxidatively modified LDL-CS! might thus be ascribed to the presence of the S-conjugates in the modified lipoprotein particle. In summary. we have shown that oxidation of CSI-modified LDL results in further changes in its physicochemical and functional properties. These changes lead in turn to the increased cytotoxicity of the modified particle. Our results give further evidence that modification of LDL by carbon disulfide plays a role in the C&induced development of the atherosclerosis. Ac,knolc,lr~lRc,,nr,rts~The authors wish to thank Mrs A. Kubiak and Mrs Z. Rudnicka for their expert technical assistance. This work was supported by Polish State Committee for Scientific Research, Grant No. PB/2395/4/ 91.
REFERENCES Brown M. S. and Goldstein J. L. (1983) Lipoprotein metabolism in the mdcrophage: implications for cholesterol deposition in atherosclerosis. Atmuol Reritw~ q/’ Biochemistr~~ 52, 223-26 I Brown M. S. and Goldstein J. L. (1990) Scavenging for receptors. Norur~ 343, 508-509.
Oxidation
of CS2 -modified
Camejo G., Hurt-Cameyo E., Rosengren B., Wiklund O., Lopez F. and Bondjers G. (1991) Modification of copper-catalysed oxidation of low density lipoprotein by proteoglycans and glycosaminoglycans. Journal of Lipid Research 32, 1983-1991. Cardin A. D., Witt K. R., Barnhart C. L. and Jackson R. L. (1982) Sulfhydryl chemistry and solubility properties of human plasma apolipoprotein B. Eiochemisfry 21, 4.503451 I. Habeeb A. F. S. A. (1966) Determination of free amino groups in proteins by trinitrobenzensulfonic acid. Analytical Biochemistry 14, 328-336. Have1 R. J., Elder M. A. and Bragdon J. H. (1955) The distribution and chemical composition of ultracentrifugaily separated lipoproteins in human serum. Journal of Clinical Investigation 34, 1345-l 353. Heinecke J. W., Kawamura M., Suzuki L. and Chait A. (1993) Oxidation of low density lipoprotein by thiols: superoxide-dependent and -independent mechanisms, Journal of Lipid Research 34, 2051-2061. Hessler J. R. D., Morel D. W., Levis L. J. and Chisolm G. M. (1983) Lipoprotein oxidation and lipoprotein-induced cytotoxicity. Arteriosclerosis and Thrombosis 3, 213-216. Hunt J. V., Smith C. C. T. and Wolf S. P. (1990) Autooxidative glycosylation and possible involvement of peroxides and free radical in LDL modification by glucose. Diabetes 39, 1420-1425. Kuzuya M., Naito M., Funaki Ch., Hayashi T., Asai K. and Kuzuva F. (1991) Linid neroxide and transition metals are required for’ the* toxicity of oxidized low density lipoprotein to cultured endothelial cells. Biochimica et Biophysics Acta 1096, 1555161. Lam C.-W. and Di Stephano V. (1986) Characterization of carbon disulfide binding in blood and other biological substances. Toxicology and Applied Pharmaco1og.y 86, 235-239. Laurman W., Salmon A., Maziere C., Maziere J.-C., Auclair M., Theron L. and Santus R. (1989) Carbon disulfide modification and impaired catabolism of low density lipoprotein. Atherosclerosis 78, 2 I l-2 18. Lavy A., Brook G. J., Dankner G., Amotz A. B. and Aviram M. (1991) Enhanced in vifro oxidation of plasma lipoproteins derived from hypercholesterolemic patients. Metabolism 8, 194-799. Liu K., Ramjiawan B., Kutryk M. J. B. and Pierce G. N. (1991) Effects of oxidative modification of cholesterol in
LDL
429
isolated low density lipoprotein on cultured smooth cells. Molecular and Cellular Biochemistry 108, 49-56. Lowry 0. H., Rosebrough N. J., Farr A. L. and Randall R. J. (1951) Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry 193, 2651215. Luck G. and Fruchart J. C. (1991) . , Oxidation of lioouroteins and atherosclerosis. American Journal of Clinkal Nutrition 53, 206s-209s. Palinski W., Rosenfeld M. E., Yla-Herttuala S., Gurtner G. C.. Socher S. S., Butler S. W., Parthasarathy S., Carew T. E., Steinberg D. and Witztum J. L. (1989) Low density lipoproteins undergoes oxidative modification in ciao. Proceedings of the National Academy of Sciences qf the U.S.A. 86, 1372-1376. Sparrow C. P. and Olszewski J. (1993) Cellular oxidation of low density lipoproteins is caused by thiol production in media containing transition metal ions. Journal of Lipid Research 34, 219-1228. Steinberg D., Parthasarathy S., Carew T. E., Khoo J. C. and Witzum J. L. (1989) Beyond cholesterol. Modification of LDL that increase its atherogenicity. Neir England Journal of Medicine 320, 9 15-924. Sunderman F. (1992) The extended therapeutic role of dithiocarb(sodium diethyldithiocarbamate) from nickel poisoning to Aids. Annals qf Clinical and Laboratory Science 22, 245-248. Thomas C. E. (1992) The influence of medium components on CL?+ -dependent oxidation of low-density lipoproteins and its sensitivity to superoxide dismutase. Biochimica et Biophysics Acta 1128,50-57. Wronska-Nofer T. (1969) The incorporation of ‘C-acetate into cholesterol in rats exposed to carbon disulfide. Biochemical Pharmacology 18, 925-926. Wronska-Nofer T. and Laurman W. (1987) The lipid risk factor of coronary heart disease in human exposure to carbon disulfide. In Occupational and Biochemical Indices ,for Monitoring Toxicity. Edited by V. Foa, E.A. Emmett, M. Maroni and A. Colombi. pp. 1855191. Ellis Horwood, Chichester. Wronska-Nofer T.. Szendzikowski S. and Laurman W. (1978) The effect of carbon disulfide and atherogenic diet on the development of atherosclerosis changes in rabbits. Atherosclerosis 31, 33339. Yagi K. (1982) Assay for serum lipid peroxide level and its clinical significance. In Lipid Peroxides in Biology and Medicine. Edited by K. Yagi. Academic Press, New York.
To.uicology in Vitro IO (1996) 423429
Carbon Disulfide-induced Modification and Cytotoxicity of Low-density Lipoproteins T. WRONSKA-NOFER*t, tThe
W. LAURMANt, J.-R. and M. WALTERf
NOFERf,
U. SEEDORF$
Nofer Institute of Occupational Medicine, Lodz. Poland and tlnstitut fur Klinische und Laboratoriumsmedizin. WestfZlische Wilhelm+Universitat Miinster, Germany
Chemie
(Accepted IS March 1996) Abstract-The secondary oxidation of biologically modified low-density lipoproteins (LDL) was demonstrated to contribute to the cytotoxicity and thereby to the atherogenicity of modified lipoproteins. Previously we have shown that chemical modification of LDL by carbon disulfide (CS) mimicked the naturally occurring process of LDL modification. In the present study the cytotoxicity of CS,-modified LDL and their susceptibility to the secondary oxidative modification was investigated. The cytotoxicity of CS,-modified LDL did not significantly exceed that of native LDL. However, the Cu’+ -oxidized form of C&modified LDL revealed to be more cytotoxic than oxidized native LDL. Oxidized CS-modified LDL presented with altered physicochemical properties including derivatization of protein amino and - SH groups, increased negative charge, and electrophoretic mobility which exceeded that of oxidized native LDL. The secondary oxidative moditicatton of C&modified LDL involved the process of Cu’+ binding to C&-derived dithiocarhamate -SH groups followed by covalent modification of -SH groups by products of lipid peroxidation. Taken together, these finding suggest that secondary oxidation of C!&-modified LDL may contribute to the atherogenic effect of the chronic occupational exposure to CS,. Copyright 4-S 1996 Elsevier Science Ltd.
INTRODUCTION
The biological modification of low-density lipoproteins (LDL) plays an essential part in the initiation of atherosclerotic processes (Brown and Goldstein, 1990; Steinberg ef ul., 1989). There is evidence that lipid peroxidation is a key event in LDL modification in zkw. Although mechanisms involved in the oxidation of LDL are not fully understood, its occurrence in the extracellular environment of the arterial wall may contribute to the LDL atherogenicity (Palinski et al., 1989). Modified LDL contain ligands recognized by scavenger receptors and subsequently accumulate in macrophages, thus contributing lo the foam cell formation (Brown and Goldstein, 1983). In addition, modified LDL were demonstrated to be highly cytotoxic to endothelial cells, smooth muscle cells, as well as macrophages, and this is thought to have a crucial role in the development of atherosclerotic lesion (Hessler et al., 1983: Liu et al., 1991). *Author for correspondence at: The Nofer Institute of Occupational Medicine, 8 SW. Teresy Str. 90-950 Lodz, Poland. Abhrez~iations: BHT = butylated hydroxytoluene: DTNB = 5.5-dithio-his-2nitrobenzoic HUVEC = acid; human umbilical vein endothelial cells; LDH = lactic dehydrogenase; LDL = low-density lipoprotein; LDLbuffered saline; TBARS = CS! = phosphate substances; T-NBS = thiobarbituric acid-reactive trinitrobenzenesulfonic acid. 0887-2333/96/$15.00 + 0.00 Copyright PIf SO887-2333(96)00029-X
The recent in citro studies of Laurman et a/. (1989) revealed that CS2 is able to modify LDL. CS? still attracts the particular attention of toxicologists as a chemical risk factor for coronary heart disease in workers in the synthetic fibre industry (WronskaNofer and Laurman, 1987). The earlier studies of Wronska-Nofer (1969) and Wronska-Nofer et al. (1978) demonstrated that the atherosclerotic changes in the arterial wall produced by CS, may be due to disturbances in the cholesterol metabolism such as increased cholesterol synthesis in the liver and deposition in the artery wall. Nevertheless, the definite mechanism of C&induced atherosclerosis remained unclear. Laurman ef al. (1989) showed that C&modified LDL (LDL-CS,) physicochemically and functionally resemble oxidatively modified LDL (Ox-LDL). LDL-CS? are taken up by the scavenger receptor and accumulate in macrophages. It is likely that LDL-CS2 undergo further modification. Secondary modification of biologically modified LDL has been demonstrated previously. For example, LDL modified by the glycosylation or by the formation of complexes with intimal proteoglycans become unstable particles more susceptible to secondary modification by oxygen free radicals than native LDL (Camejo et al., 1991; Hunt et al., 1990). In the present study, we examined whether LDL chemically modified by CS2 are cytotoxic. In addition, we investigated whether secondary oxidative modification of LDL-CS2 results in the
,C 1996 Elsevier Science Ltd. Al! rights reserved.
Printed
in Great
Britain
T. Wronska-Nofer rt ai
424
alteration of the physicochemical properties and cytotoxicity. We assumed that both modification of LDL by CS? and the secondary modification of LDL-CS, by oxidation could contribute to the development of atherosclerotic lesions. MATERIALS AND METHODS
Prepuration
of LDL
und LDL-C’S,
LDL (density I ,019-l ,063 g/ml) was isolated from the plasma of healthy subjects by sequential ultracentrifugation according to the method of Have1 et al. (1955). LDL was dialysed against IO mM phosphate buffered saline (PBS), pH 7.4. for 24 hr. sterilized by filtration, and stored under nitrogen at 4“C in the dark. EDTA (0.1%. M.:v) was present throughout all steps of preparation. Modification of LDL by CS2 was performed by dialyzing LDL (5 mg/ml) against C&-saturated PBS, pH 7.4 for 4 hr. Free CSI was removed from the preparation by extensive dialysis against PBS containing EDTA (0.1 %, w/v). Osidution
of LDL
und LDL-C‘S,
Before oxidation LDL and LDL-CSI were dialysed for 24 hr at 4 C against de-aerated and nitrogen-saturated PBS to remove EDTA. The LDL and LDL-CS, were diluted with EDTA free. oxygen saturated IO mM PBS, pH 7.4. Oxidation of LDL was initiated by addition of freshly prepared CuSO, at a room temperature and monitored spectrophotometritally at 234 nm. The final concentrations were 0.25 mg/ml of LDL and 1.66 PM CuSO,. Assessment qf c~~~toto.\-icit~~
Human umbilical vein endothelial cells (HUVEC’) were purchased from PromoCell (Heidelberg, Germany) and maintained in endothelial cell basal medium (PromoCell) supplemented with foetal bovine serum, human basic fibroblast growth factor. human epidermal growth factor. endothelial cell growth supplement/heparin. hydrocortisone, amphotercin B and gentamicin at 37 C in a humidified atmosphere containing 5% CO,. All cell culture supplements were provided by PromoCell and used in concentrations recommended by the supplier. HUVEC were grown to confluency in 96-well dishes. The cytotoxicity was assessed using CytoTox 96’” Heidelberg, assay (Promega/Serva. cytotoxicity Germany) according to the man&cturer’s instrucTable Relative
I.
Phwcochemlcal
electrophoretic
mobtlity
tions. This assay quantitatively measures activity of lactate dehydrogenase (LDH) released on cell lysis following incubation with the cytotoxic agents. The absorbance data were collected at 490 nm with the Dynatech MR600 96 well plate reader and expressed as percentage of total LDH activity released from cells following treatment with Lysis Solution supplied by the manufacturer. In addition, the cell viability was estimated using Trypan blue exclusion assay. Briefly, HUVEC grown to confluency in 24-well dishes were incubated with the cytotoxic agents for the appropriate time. The cells were then trypsinized and the cell suspension was incubated for 5 min with Trypan blue (0.25%, v/v). The number of dye-excluding cells was determined in the haemocytometer. The results are presented as percent of viable cells. Other rr1ethod.s
LDL-protein
was assayed by the method of Lowry serum albumin as the standard. Electrophoresis of LDL was performed in I % agarose gels in barbital buffer. pH 8.6 (Laurman <>iui., 1989). Lipoproteins were visualized by staining with Fat Red 7B (Sigma, Deisenhofen, Germany). Free amino groups of LDL were determined by the trinitrobenzenesulfonic acid (TNBS) calorimetric method (Habeeb. 1966) using HCI-lysine as a standard. Thiobarbituric acid-reactive substances (TBARS) in the LDL samples were determined by the method of Yagi (1982) using malondialdehyde bis-dimethylacetal as a standard. The concentration of -SH groups in LDL-CS? was determined by the method of Cardin et al. (1982) with 5,5-dithiobis-2nitrobenzoic acid (DTNB).
et 01. (1951) using bovine
RESULTS
The physicochemical properties of modified LDL are presented in Table I. LDL were modified with CS, by dialysis against CS1-saturated buffer in neutral pH (7.4). Ox-LDL presented with an increased negative charge as demonstrated by the enhanced electrophoretic mobility in a 1% agarose gel. CS1-modified LDL migrated further towards the anode than the native LDL though not so far away as Ox-LDL. The oxidized LDL-CS? exhibited the most pronounced electrophoretic mobility. However. prooert~es of modified _~
Free amino groups
LDL Free -SH nmoljmg
LDL (nattve) Ox-LDL LDL-CS: Oxidized LDL-CS: *P .z 0.05: **P < 0.01: ***P
1.00 2.07 2 0.17 I .37 + 0.09 2.24 5 0.21 <
509 176 312 154
ND = not detectable 0.001 I‘ native LDL. Student’s r-test.
* + * i
78 21** 53’ 27’
groups
TBARS
protein
7*2 Ski 131 * 7”’ ND
ND 16.2 * I.1 ND 14.7 k 1.8
425
Oxidation of C&-modified LDL propagation phase
lag phase
degradation phase
2.4 1 E 2.2E 2.0 ZJ N al 18” c : G n P a
1
.6-
1
.4 -8
1.0
-
I
0
60
I time
120 [min]
I
1
180
240
Fig. 1. Time course of the CL?+-catalysed oxidation of LDL. LDL (0.25 mg/ml) was incubated with 10 PM CuSOl at room temperature. The absorbance at 234 nm reflecting conjugated dienes formation during lipid peroxidation was monitored. Values presented arc computed from triplicate determinations.
the migration distance of LDL-CS! did not represent the addition of the migration distances of Ox-LDL and LDL-CSZ. Thus, the oxidative and C&mediated modifications were apparently not self dependent. The content of free amino groups in LDL-CS, was 30% lower than in native LDL, whereas in oxidatively modified LDL their amount decreased by 75%. In oxidized LDL-CS, the decrease of free amino groups was comparable to that found in Ox-LDL. The decrease of free amino groups in LDL-CS, was paralleled by an increase in -SH groups. The newly-formed -SH groups most likely originated from dithiocarbamates of CS, origin (Lam and di Stephano, 1986). No changes of the -SH group level in oxidatively modified LDL were observed, whereas the content of -SH groups in oxidized LDL-CS, was reduced below the lower detection level. These data indicate, that oxidative and C.‘$-mediated modifications of LDL are not self-dependent and that dithiocarbamate-derived -SH groups of LDL-CSr are further modified/derivatized by oxidation. Additionally, native and CS? modified LDL were analysed for TBARS, an index of lipid peroxidation. The results demonstrated similar increase in TBARS values in native and C&modified LDL following oxidation indicating that the degree of lipid oxidation was similar in both instances. The kinetics qf’ the oxidation
qf LDL-CS,
To further characterize LDL-CSr, its susceptibility to the Cur + -stimulated oxidation was evaluated. The kinetics of the oxidation was followed by the continuous monitoring of the absorbance at 234 nm, which reflects the extent of conjugated dienes formation during oxidation of polyunsaturated fatty acids in LDL lipids to fatty acid hydroperoxides with
double bonds. The recorded time-kinetics of the native LDL oxidation presented three typical successive phases (Fig. 1): (I) a lag phase with slow increase of the 234-nm absorbance reflecting the minimal rate of lipid peroxidation and conjugated diene formation due to protective effects of LDL antioxidants; (2) a propagation phase characterized by the rapid increase of the absorbance related to the burst of lipid peroxidation and diene formation; and (3) a decomposition phase, in which the labile lipid hydroperoxides with conjugated double bonds decompose to a great variety of products including reactive aldehydes with cytotoxic properties. As shown in Fig. 2, the profile of the LDL-CS, oxidation monitored by conjugated diene formation differed from that of native LDL. Addition of Cu’+ to LDL-CSI caused rapid increase of the 234-nm absorption reflecting the formation of the short-lived Cu-mercaptide chromophore (Fig. 2). This was followed by the gradual decline of the absorption during the first hour of the reaction. Approximately 40% of the initial -SH group content disappeared in LDL-CS?, when Cu” was added (Fig. 3). As the experiment was done in the nitrogen-saturated medium, the observed decrease in -SH groups should be ascribed to Cu’ + binding rather than to covalent modification by lipid peroxidation products. This was further confirmed by the observation that addition of EDTA (O.l%, w/v) to LDL-CS>-Cu’ + complex partially recovered free -SH groups, most likely due to removal of copper from LDL-CS?. By contrast, in the oxygen-containing medium, the content of -SH groups in LDL-CSr particles incubated with CL?+ decreased below the lower detection limit during the first hour of the reaction (Fig. 4). This process was blocked by the conjugated
T Wronska-Nofer et al.
426
2.4
-
E2.2 c 02.0 0 (u al.8 0 c zl.6
-
0
::I.4 m
-
1.2
-
1.0
’
’
I
0
60
I time
120 (m in]
I
1
180
240
Fig. 2. Cu’ -catalysed oxldatlon ot’ <‘ST-modllied LDL I native LDL. CS,modified LDL (0.25 mg/ml) and native LDL (0.25 mg;ml) were Incubated with IO FM CuSO, at room temperature. Oxidation was followed by monitoring the ahsorhance at 234 nm. Values presented are computed from triplicate determinations. Note in the case oP LDL-C’S, a rapid increase of the absorbance on the addition of Cu’(Cu-mercaptide chromophore) followed b) gradual decline and elongation of the lag phase of conjugated dienek generation.
addition of 25 PM butylated hydroxytoluene (BHT) which inhibits lipid peroxidation by way of free radical scavenging. These findings suggest that to -SH groups in LDL-C’S: ih binding of Cu” followed by the gradual ~ SH group modification by lipid peroxidation products. Cytototosicit~~
of
o.uidrrtiwl~~ nzod$frrtl LDL
rrtrd
LDL-CS:
Because oxidized LDL were shown to be cytotoxic and since this process is likely to be involved in the pathogenesis of atherosclerosis, the cytotoxicity of modified LDL towards endothelial cells was examined. HUVEC grown to confluency were incubated with 0.3 m&/ml native LDL, LDL-CS,. Ox-LDL. oxidized LDL-CS? or with buffer as a control. The
”
LDH release in control cells was 7.4% + 0.16% (mean i: SEM; n = 8) and was not significantly changed in the presence of native LDL or LDL-CS2. By contrast. oxidatively modified LDL revealed to be highly cytotoxic (Fig. 5). In addition, the ability of oxidatively modified LDL-C& to exert cytotoxic effect towards HUVEC exceeded significantly that of oxidatively modified native LDL. To further investigate the cytotoxic effect of modified LDL, the cell viability was estimated using Trypan blue exclusion assay. The percentage of viable cells at the begining of the experiment was 92 k 1.5 (mean i SEM; n = 4). and was not significantly changed after incubation with buffer or with 0.3 mg;ml native LDL or C&modified LDL. By contrast, both Ox-LDL and oxidized LDL-CSI
100
g 0’ 0
80
;
80
g a
40
5b
20 0
LDL LDL-CS, Fig. 3. Effect of EDTA on Cu’ _-mediated depletion of the content of free - SH groups in CS>-modified LDL. Native and C&-modified LDL (0.25 mg/ml) were ureincubated with 10 UM CuSO, in the medium saturated with nitrogen to prevent lipih peroxidation and next supplemented with 0.1% EDTA (w/v) or buffer as reference. Content of - SH group was determined using DTNB reagent and expressed as relative value Y. untreated, native LDL. Values are mean ) SEM for three separate determinations.
Oxidation of C&-modified LDL .E
421
(Brown and Goldstein, 1990; Luck and Fruchart, 1991; Steinberg ef al., 1989). The oxidation of LDL has been shown to occur in vivo, and the positive correlation between the level of Ox-LDL and extent of atherosclerotic lesions as well as an inverse correlation with blood level of antioxidants has been found (Lavy et al., 1991). Previously, we reported that carbon disulfide can modify LDL (Laurman et al., 1989) yielding a particle that physicochemically and functionally resembles oxidized native LDL. In this paper we present evidence that Cu’ +-induced oxidation of LDL-CS, results in a new particle with altered physicochemical properties and more cytotoxic than oxidatively modified native LDL. In this respect, LDL-CSZ resembles glycosylated or proteoglycanmodified LDL which also display increased cytotoxicity and atherogenicity after secondary oxidative modification (Camejo et al., 1991; Hunt ef nl., 1990). The changes of the functional properties of oxidized LDL-CS2 may be linked to the considerably increased negative charge of the particles as inferred from an increased electrophoretic mobility and decreased level of positively charged amino groups. It is possible that Cu’ +-oxidized LDL-CS, particles generate epitopes which are ligands for the macrophage scavenger receptor. In this context, it is noteworthy that the degree of positively charged amino group modification in LDL determines the extent of the oxidized LDL particle recognition by scavenger receptors and its delivery in an unregulated manner to the target cell (Brown and Goldstein, 1983). In comparison with native LDL, the time course of CL?+-catalysed oxidation of C&modified LDL is characterized by an altered profile of conjugated of atherosclerosis
0
‘*O
time [mm] -1~0
phase 2
Fig. 4. Protective effect of BHT against depletion of the - SH group content in Cu’ +-oxidized LDL-CS?. LDL-CS? (0.25 mg/ml) was incubated in the presence or absence of BHT (25 PM) in the medium saturated with oxygen. Oxidation of LDL-CS? was initiated by addition of CuS04 (10 PM). At the indicated time -SH groups content was determined. Value represents mean f SD from three separate determinations.
considerably decreased cell viability (Fig. 6). Again, the effect of oxidatively modified LDL-CS? exceeded significantly that of oxidatively modified native LDL. DISCUSSION
Increasing evidence suggests that oxidative modification of LDL plays a crucial part in the pathogenesis
m D
El J 0
3-l
diene formation. This may be related to the high content of dithiocarbamate-derived - SH groups in the LDL-CS, particle. The rapid binding of Cu’+ by
Ox-LDL oxidized LDL-CS,
6h
9h
12h
time Fig. 5. Cytotoxicity of modified LDL. Ox-LDL and oxidized LDL-CS, (0.3 mg/ml) were incubated with HUVEC grown to confluency at 37°C for 3,6,9 and I2 hr, and the activity of LDH released to the medium was determined as described in Materials and Methods. Results were expressed as percent of total LDH activity released to medium from the cells treated with a Lysis Solution. Values are the mean + SEM for eight determinations in one experiment and are representative for three independent experiments. Note the significant difference between the value for oxidized LDL-CS? and oxidized native LDL at *P < 0.05 and **P < 0.01 (Student’s /-test).
428
T. Wronska-Nofer
3h
6h
cl al
9h
12h
time Fig. 6. Effect of Ox-LDL and oxidized LDL-CSI on HUVEC vlabdity. Ox-LDL and oxidzed LDL-CS? (0.3 mg/ml) were incubated with HUVEC grown to confluency at 37 C for 3. 6, 9 and 12 hr, and the cell viability was assessed with Trypan blue exclusion test as described in Materials and Methods. Values are the mean f SEM for four determinations In one experiment and are representative for three independent experiments. Note the significant difference between the value for oxidized LDL-CSI and oxidized native LDL at *P < 0.05 and **P < 0.01 (Student’s r-test).
dithiocarbamate-derived
chromophore may
be
thiols
with a pronounced suggested
that
yields
a short-lived
UV absorbance.
transfer
of
Cu”
It to
in apoprotein moleties of LDLCS, located close to LDL-lipids is essential for copper-catalysed oxidation of LDL-CSI. The profile of changes in the 234-nm absorbance of conjugated dienes in the course of Cu’ +-catalysed peroxidation of LDL-CSZ accounts for this assumption. The Cu’- -catalysed oxidative modificatron of LDL-CS, consists in the secondary derivatization of dithiocarbamate - SH groups. This derivatization may be explained by binding of CL?. to thiols. We demonstrated that in the nitrogen-saturated medium a number of SH groups in LDL-CS, is significantly reduced on addition of Cu’ + and that this is partially reverted by addition of EDTA. It is well established that -SH groups of dithiocarbamates can avidly bind transition metals (Sunderman. 1992). Morcovcr. copper ions attached to proteins, enzymes or cellular surfaces can still retain their highly pro-oxldative redox potential. In this context, it is worth noting that thiol-containing agents including cysteine, homocysteine and glutathione were demonstrated to induce LDL modification by a reaction requiring the presence of copper. The process of LDL modification includes oxygen- and sulfur-centred free radicals generated during metal-catalysed oxidation of thiols (Heinecke rf ul.. 1993; Sparrow and Olszewski, 1993). It has been postulated that formation of LDL-Cu complexes has an essential role not only in the oxidative modification but also in the cytotoxicity of Ox-LDL (Kuzuya et ul.. 1991; Thomas, 1992). Hence, it is possible that increased cytotoxicity of the oxidatively modified LDL-CS2 is related to the extensive binding of Cu’+ to LDL-CS, and the subsequent - SH group derivatization. The attack of copper-generated reactive oxygen species on dithioweak-binding
sites
carbamate-derived -SH groups in LDL-CS, may lead to the creation of thiol radicals and their cytotoxic derivatives. Another explanation for thiol derivatization comes from the observation that it can be prevented by BHT, the lipid peroxidation inhibitor (cf. Fig. 4). Among the products of lipid peroxidation, which could secondarily react with -SH groups, the aldehydes are the most likely candidates. Aldehydes, which possess a$-unsaturated CH = CHCHO units react with nucleophiles such as -SH to give S-conjugates, CHSCH&HO. These conjugates have been shown to be cytotoxic and to exist in native LDL particles albeit at a low concentration (Hessler et al., 1983; Luck and Fruchart, 1991). The toxicity of oxidatively modified LDL-CS! might thus be ascribed to the presence of the S-conjugates in the modified lipoprotein particle. In summary. we have shown that oxidation of CSI-modified LDL results in further changes in its physicochemical and functional properties. These changes lead in turn to the increased cytotoxicity of the modified particle. Our results give further evidence that modification of LDL by carbon disulfide plays a role in the C&induced development of the atherosclerosis. Ac,knolc,lr~lRc,,nr,rts~The authors wish to thank Mrs A. Kubiak and Mrs Z. Rudnicka for their expert technical assistance. This work was supported by Polish State Committee for Scientific Research, Grant No. PB/2395/4/ 91.
REFERENCES Brown M. S. and Goldstein J. L. (1983) Lipoprotein metabolism in the mdcrophage: implications for cholesterol deposition in atherosclerosis. Atmuol Reritw~ q/’ Biochemistr~~ 52, 223-26 I Brown M. S. and Goldstein J. L. (1990) Scavenging for receptors. Norur~ 343, 508-509.
Oxidation
of CS2 -modified
Camejo G., Hurt-Cameyo E., Rosengren B., Wiklund O., Lopez F. and Bondjers G. (1991) Modification of copper-catalysed oxidation of low density lipoprotein by proteoglycans and glycosaminoglycans. Journal of Lipid Research 32, 1983-1991. Cardin A. D., Witt K. R., Barnhart C. L. and Jackson R. L. (1982) Sulfhydryl chemistry and solubility properties of human plasma apolipoprotein B. Eiochemisfry 21, 4.503451 I. Habeeb A. F. S. A. (1966) Determination of free amino groups in proteins by trinitrobenzensulfonic acid. Analytical Biochemistry 14, 328-336. Have1 R. J., Elder M. A. and Bragdon J. H. (1955) The distribution and chemical composition of ultracentrifugaily separated lipoproteins in human serum. Journal of Clinical Investigation 34, 1345-l 353. Heinecke J. W., Kawamura M., Suzuki L. and Chait A. (1993) Oxidation of low density lipoprotein by thiols: superoxide-dependent and -independent mechanisms, Journal of Lipid Research 34, 2051-2061. Hessler J. R. D., Morel D. W., Levis L. J. and Chisolm G. M. (1983) Lipoprotein oxidation and lipoprotein-induced cytotoxicity. Arteriosclerosis and Thrombosis 3, 213-216. Hunt J. V., Smith C. C. T. and Wolf S. P. (1990) Autooxidative glycosylation and possible involvement of peroxides and free radical in LDL modification by glucose. Diabetes 39, 1420-1425. Kuzuya M., Naito M., Funaki Ch., Hayashi T., Asai K. and Kuzuva F. (1991) Linid neroxide and transition metals are required for’ the* toxicity of oxidized low density lipoprotein to cultured endothelial cells. Biochimica et Biophysics Acta 1096, 1555161. Lam C.-W. and Di Stephano V. (1986) Characterization of carbon disulfide binding in blood and other biological substances. Toxicology and Applied Pharmaco1og.y 86, 235-239. Laurman W., Salmon A., Maziere C., Maziere J.-C., Auclair M., Theron L. and Santus R. (1989) Carbon disulfide modification and impaired catabolism of low density lipoprotein. Atherosclerosis 78, 2 I l-2 18. Lavy A., Brook G. J., Dankner G., Amotz A. B. and Aviram M. (1991) Enhanced in vifro oxidation of plasma lipoproteins derived from hypercholesterolemic patients. Metabolism 8, 194-799. Liu K., Ramjiawan B., Kutryk M. J. B. and Pierce G. N. (1991) Effects of oxidative modification of cholesterol in
LDL
429
isolated low density lipoprotein on cultured smooth cells. Molecular and Cellular Biochemistry 108, 49-56. Lowry 0. H., Rosebrough N. J., Farr A. L. and Randall R. J. (1951) Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry 193, 2651215. Luck G. and Fruchart J. C. (1991) . , Oxidation of lioouroteins and atherosclerosis. American Journal of Clinkal Nutrition 53, 206s-209s. Palinski W., Rosenfeld M. E., Yla-Herttuala S., Gurtner G. C.. Socher S. S., Butler S. W., Parthasarathy S., Carew T. E., Steinberg D. and Witztum J. L. (1989) Low density lipoproteins undergoes oxidative modification in ciao. Proceedings of the National Academy of Sciences qf the U.S.A. 86, 1372-1376. Sparrow C. P. and Olszewski J. (1993) Cellular oxidation of low density lipoproteins is caused by thiol production in media containing transition metal ions. Journal of Lipid Research 34, 219-1228. Steinberg D., Parthasarathy S., Carew T. E., Khoo J. C. and Witzum J. L. (1989) Beyond cholesterol. Modification of LDL that increase its atherogenicity. Neir England Journal of Medicine 320, 9 15-924. Sunderman F. (1992) The extended therapeutic role of dithiocarb(sodium diethyldithiocarbamate) from nickel poisoning to Aids. Annals qf Clinical and Laboratory Science 22, 245-248. Thomas C. E. (1992) The influence of medium components on CL?+ -dependent oxidation of low-density lipoproteins and its sensitivity to superoxide dismutase. Biochimica et Biophysics Acta 1128,50-57. Wronska-Nofer T. (1969) The incorporation of ‘C-acetate into cholesterol in rats exposed to carbon disulfide. Biochemical Pharmacology 18, 925-926. Wronska-Nofer T. and Laurman W. (1987) The lipid risk factor of coronary heart disease in human exposure to carbon disulfide. In Occupational and Biochemical Indices ,for Monitoring Toxicity. Edited by V. Foa, E.A. Emmett, M. Maroni and A. Colombi. pp. 1855191. Ellis Horwood, Chichester. Wronska-Nofer T.. Szendzikowski S. and Laurman W. (1978) The effect of carbon disulfide and atherogenic diet on the development of atherosclerosis changes in rabbits. Atherosclerosis 31, 33339. Yagi K. (1982) Assay for serum lipid peroxide level and its clinical significance. In Lipid Peroxides in Biology and Medicine. Edited by K. Yagi. Academic Press, New York.