Vol.
178,
August
No. 15,
3, 1991
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
OF THE PEROXIDATION PRODUCT HYDROXYSTEARIC LUNG CARCINOMA CELLS G.
of Biological
E. Casali,
Cavalli,
Chemistry,
School
*Department Recei.ved
AND
Pages 1260-1265
IDENTIFICATION
Institute
BIOCHEMICAL
1991
July
A. Spisni, of Medicine,
of Biochemistry,
ACID
IN LEWIS
and L. Masotti* University
of Parma, 43100 Parma, Italy
40127 Bologna, Italy
3, 1991
SUMMARY: Whole cell lipids were extracted from the Lewis lung carcinoma in vitro line C108. The fatty acids were derivatized to methylesters in order to identify endogenous oxidized derivatives by gasmass spectroscopy. The presence of 9-hydroxystearic acid and lo-hydroxystearic acid was thus evidenced for the fiit time in cultured mammalian cells. Moreover a linear correlation was found between the concentration of these products expressed as percentage of total fatty acid methylesters and the cell density in tissue culture flasks. This finding suggests an involvement of hydroxystearic acid in cellular functions related to the cell density in monolayer cultures. o 1991 Academic press, IX.
In the past few years, the significance of primary and secondary lipid peroxidation tumor
cell processes has been the subject of many studies. In particular,
tromboxanes directly
and leucotrienes, the hydroperoxy
or indirectly
through
enzymatic
products in normal and besides prostaglandins,
and hydroxy derivatives of polyene fatty acids formed
or non-enzymatic
mechanisms
have become increasingly
interesting because of their in vitro and in vivo effects on DNA synthesis and cell proliferation Furthermore,
it is well known that lipid peroxidizability
is reduced in tumor cells as compared to normal
cells, and researches carried out by our group showed an inverse correlation peroxidizability peroxidation
(1,2,3,4,5).
between the extent of lipid
and the rate of tumor growth (6,7,8,9). These data prompted us the hypothesis that lipid by-products
may play a role in the control of normal and tumor cell proliferation.
we have been focused on the identification
of endogenous fatty acid peroxidation
systems. In this paper we report the presence of endogenous
9-HSA
Therefore,
products in other celhrlar
and lo-HSA
in the Lewis lung
carcinoma cell line C108.
MateriaLr: FCS was from International PBI. RPM1 1640 Medium and L-Glutamine were from Seromed. Waymouth’s MB 752 and tissue culture flasks were from ICN Nutritionals. TLC silica gel plates The abbreviations used are : 9-HSA, 9-hydroxystearic acid ; lo-HSA, lo-hydroxystearic acid ; BHT, 2,6-di-tert-butyl-p-cresol; ART, Alkaline-reducing transmethylation; HODEs, hydroxyoctadecadienoic acids ; HETEs, hydroxyeicosatetraenoic acids; FCS, Foetal calf serum; 0-TMS, 0-trimethylsilyl; TLC, thin layer chromatography; GLC, gas-liquid chromatography ; GLC-GM, gas-liquid chromatography-gas mass; EDTA, ethylenediaminetetraacetic acid; PBS, Dulbecco’s phosphate buffered saline; FAMES, fatty acid methylesters. 0006-291X/91 Copyright All rights
$1.50
0 1991 by Academic Press, Inc. of reproduction in any form reserved.
1260
Vol.
178,
No.
3, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
were from Merck. All chemicals were of analytical grade. All solvents were of chromatography grade. Oxidized egg lecithin (10) was kindly provided by Prof. M. V. Piretti, University of Bologna, Italy. Cell culture. The Lewis Lung carcinoma cell line Cl08 (11,12) was a generous gift of Dr. G. Zupi, “1st. Regina Elena”, Rome, Italy. Cells were cultured in tissue culture flasks using RPM1 1640 medium supplemented with L-Glutamine (2mM), 20 &ml Streptomycin and 20 Units/ml Penicillin (Eurobio, Paris) and FCS 10%. For subculturing, the monolayers were washed with EDTA 0,7 mM in PBS. The rinse was discarded and cells were incubated with lml/lO cm’ of the same solution for 5’ at room temperature. After stripping, the cells were centrifuged for 5’ at 1200 rpm and resuspended in the culture medium. Seeding was done at a cell density of 16,000 cells/cm*. Lipid extraction. After stripping, the cells were washed two times with PBS and counted. The percentage of dead cells was determined by Trypan Blue exclusion and it was always lower than 10%. Total lipids were then extracted according to Folch et al. (13). BHT was added to the extraction mixture of cloroform:methanol (2:l) at a final concentration of 0.01% (w/v), in order to avoid oxidation artifacts during the extraction procedure. Fatty Acid transmethylation and separation of polar products. Fatty acids were transmethylated according to Piretti ti. (14,15). TLC separation of the material recovered after transmethylation was performed by loading 1 - 1,5 mg/cm at the start line and developing plates at a constant temperature of 4°C with hexane:diethyl ether mixture (4:1, v/v). The solvent was run for about 12 cm in order to obtain a good resolution of the bands containing the oxidation products. Under these conditions, the methylesters without polar groups posses a RF = 0.50, while the hydroxy esters are characterized by a RF = 0.05. TLC plates were sprayed with Rhodamin 6 G (Merck) and the bands, detected by fluorescence (excitation at 366 nm), were scraped and extracted with ethyl ether stabilized with BHT. Tbe solvent was then evaporated under nitrogen stream and the material dissolved in benzene for GLC and/or GLC-MS analysis. The benzene solutions were stored at -20°C. For the GLC-MS analysis the -0TMS derivatives were obtained by using the TRI-SIL reagent (Pierce). GLC and GLC-MS analysis. For GLC analysis, a Nordion Micromat HRGC 412 gas chromatograph, equipped with a bonded 25 m fused silica capillary column (0,32 mm internal diameter, 0,25 pm film thickness, model 007 FFAP, Quadrex Corporation) was employed. Peaks area were recorded by a Kipp&Zonen bd 71 integrator. The injector and detector (flame ionization detector) temperature was 250°C; the column temperature was initially kept at 180°C and then raised to 240°C at a rate of 5°C per minute and maintained at this value until the chromatogram was completed. The carrier gas was H,, employed at a flow rate of 1.7 ml/mm; splitting was set at a ratio of l/50. GLC-MS was carried out using a Finnigan 1020 GLC-MS combination.
The whole cell total lipid extract was subjected to ART in order to avoid artifacts during the various analytical steps (15). This procedure protects from oxidation allows to perform the simultaneous
reduction
of any hydroperoxy
hydroxy methylester, thereby avoiding hydroperoxide
methylester,
reaction and
to the correspondent
loss. Tbe material recovered after ART was applied
to silica gel TLC plates for the separation of oxidation (hexane/diethyl
during the transmethylation
products. A solvent with a low eluting power
ether 4:l; see Methods) was used and it was possible to evidence three different bands.
GLC analysis of the products of the band characterized by an RF = 0.05 indicated the existence of a peak associable with a hydroxy methylester. In fact, the peak co-elutes with the one corresponding 9-HSA
and lo-HSA
methylesters (Fig.la,lb,lc).
In order to confirm our GLC data, we analyzed the -0TMS by GLC-MS.
to the
The mass spectrum and the fragmentation
derivative of the endogenous polar compound scheme are shown in Fig. 2 and 3 respectively.
Despite the lack of some high molecular weight fragments, which is a typical feature of these compounds, the fragmentation
can be interpreted as representative of a mixture of methylesters derived from 9-HSA
and lo-HSA.
1261
Vol.
178,
No.
3, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Figure 1. (a) GLC profile of the material recovered from the TLC band with RF = 0.05; (b) GLC profile of an oxidized standard of egg lecithin containing different hydroxy derivatives of linoleic and oleic acids methylesters; (c) mixture of (a) plus (b). A relative increase of the peak 2 can be observed. Peak identification: 1) hydroxyoleic acid (cis) methylester 2) mixture of 9- and lo-hydroxystearic acid methylesters 3) hydroxyoleic acid (trans) methylester
259 55
229 03 215
273
iir I- 1 50
Figure 2. Mass spectrum of the 0-TMS derivative of the oxidation product corresponding to the peak 2 in figure la. The spectrum has been recorded using a Finnigan 1020 GLC-MS combination.
1262
Vol. 178, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
a
m/z=22g
.
I
0 CH,- (CH,),+CH;(CH,),-C' 'OCH, . .......... m/z=259 ____--------_-_----_------~ b
0 CH,- (CH,),+CH+ (CH,),--CT OCH,
I
Figure 3. Fragmentation scheme of the 0-TMS (b) IO-hydroxystearic acid methylesters.
The 9-OTMS-stearate
is characterized by the two peaks at m/z=229 and m/z=259 (Fig. 3a) and it seems
to be the main component. m/z=273 Similarly,
derivatives of: (a) 9-hydroxystearic acid and
The
lo-OTMS-stearate,
characterized
by the two peaks at m/z=215 and
(Fig. 3b), is probably the minor component of the mixture. the band identified by RF=OSO was recovered and analyzed by GLC. As expected the results
confirmed the presence of non polar FAMES’. As a second step we investigated the possible existence of a lipid composition
dependence from cell density.
Table 1 shows that the relative amount of each fatty
Table 1. fatty acid composition in samples characterized by different cell densities (87%, 68%, 49% and 32%) Fatty acid 14:o 16:0 16:l 18:0 18:lA 18:lA 18:2 2O:l 20:3 20:4 22:4 22:s 22:6
9 11
Composition % 1.4 17.6 4.6 18.3 32.9 6.2 3.0 1.1 1.2 5.7 1.2 3.3 3.4
2.2 21.9 4.9 17.0 29.5 6.1 2.8 0.7 1.3 5.9 1.4 3.2 3.5
1263
1.7 18.8 4.6 17.1 28.2 6.2 3.3 0.7 1.5 6.5 1.1 3.5 5.5
1.9 19.1 4.4 18.1 27.6 5.9 3.4 0.9 1.2 6.3 1.1 3.4 5.4
Vol.
178,
No.
BIOCHEMICAL
3, 1991
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
0 06 30
40
50
60
70
60
90
100
% of confluence
Figure 4. Correlation between the relative amount of endogenous HSA and the % of cell confluence. Cell confluence was 600000 cells/cm* ; the relative amount of the HSA methylester (9-HSA plus loHSA) was calculated dividing the area of peak 2 in the chromatogram @and with RF=0.05; see figure la) over the area of total FAMES; (band with RF=O.S). The response correction factor for HSA was assumed equal to 1. The linear correlation coefficient is r= 0.997 @
178,
August
No. 15,
3, 1991
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
OF THE PEROXIDATION PRODUCT HYDROXYSTEARIC LUNG CARCINOMA CELLS G.
of Biological
E. Casali,
Cavalli,
Chemistry,
School
*Department Recei.ved
AND
Pages 1260-1265
IDENTIFICATION
Institute
BIOCHEMICAL
1991
July
A. Spisni, of Medicine,
of Biochemistry,
ACID
IN LEWIS
and L. Masotti* University
of Parma, 43100 Parma, Italy
40127 Bologna, Italy
3, 1991
SUMMARY: Whole cell lipids were extracted from the Lewis lung carcinoma in vitro line C108. The fatty acids were derivatized to methylesters in order to identify endogenous oxidized derivatives by gasmass spectroscopy. The presence of 9-hydroxystearic acid and lo-hydroxystearic acid was thus evidenced for the fiit time in cultured mammalian cells. Moreover a linear correlation was found between the concentration of these products expressed as percentage of total fatty acid methylesters and the cell density in tissue culture flasks. This finding suggests an involvement of hydroxystearic acid in cellular functions related to the cell density in monolayer cultures. o 1991 Academic press, IX.
In the past few years, the significance of primary and secondary lipid peroxidation tumor
cell processes has been the subject of many studies. In particular,
tromboxanes directly
and leucotrienes, the hydroperoxy
or indirectly
through
enzymatic
products in normal and besides prostaglandins,
and hydroxy derivatives of polyene fatty acids formed
or non-enzymatic
mechanisms
have become increasingly
interesting because of their in vitro and in vivo effects on DNA synthesis and cell proliferation Furthermore,
it is well known that lipid peroxidizability
is reduced in tumor cells as compared to normal
cells, and researches carried out by our group showed an inverse correlation peroxidizability peroxidation
(1,2,3,4,5).
between the extent of lipid
and the rate of tumor growth (6,7,8,9). These data prompted us the hypothesis that lipid by-products
may play a role in the control of normal and tumor cell proliferation.
we have been focused on the identification
of endogenous fatty acid peroxidation
systems. In this paper we report the presence of endogenous
9-HSA
Therefore,
products in other celhrlar
and lo-HSA
in the Lewis lung
carcinoma cell line C108.
MateriaLr: FCS was from International PBI. RPM1 1640 Medium and L-Glutamine were from Seromed. Waymouth’s MB 752 and tissue culture flasks were from ICN Nutritionals. TLC silica gel plates The abbreviations used are : 9-HSA, 9-hydroxystearic acid ; lo-HSA, lo-hydroxystearic acid ; BHT, 2,6-di-tert-butyl-p-cresol; ART, Alkaline-reducing transmethylation; HODEs, hydroxyoctadecadienoic acids ; HETEs, hydroxyeicosatetraenoic acids; FCS, Foetal calf serum; 0-TMS, 0-trimethylsilyl; TLC, thin layer chromatography; GLC, gas-liquid chromatography ; GLC-GM, gas-liquid chromatography-gas mass; EDTA, ethylenediaminetetraacetic acid; PBS, Dulbecco’s phosphate buffered saline; FAMES, fatty acid methylesters. 0006-291X/91 Copyright All rights
$1.50
0 1991 by Academic Press, Inc. of reproduction in any form reserved.
1260
Vol.
178,
No.
3, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
were from Merck. All chemicals were of analytical grade. All solvents were of chromatography grade. Oxidized egg lecithin (10) was kindly provided by Prof. M. V. Piretti, University of Bologna, Italy. Cell culture. The Lewis Lung carcinoma cell line Cl08 (11,12) was a generous gift of Dr. G. Zupi, “1st. Regina Elena”, Rome, Italy. Cells were cultured in tissue culture flasks using RPM1 1640 medium supplemented with L-Glutamine (2mM), 20 &ml Streptomycin and 20 Units/ml Penicillin (Eurobio, Paris) and FCS 10%. For subculturing, the monolayers were washed with EDTA 0,7 mM in PBS. The rinse was discarded and cells were incubated with lml/lO cm’ of the same solution for 5’ at room temperature. After stripping, the cells were centrifuged for 5’ at 1200 rpm and resuspended in the culture medium. Seeding was done at a cell density of 16,000 cells/cm*. Lipid extraction. After stripping, the cells were washed two times with PBS and counted. The percentage of dead cells was determined by Trypan Blue exclusion and it was always lower than 10%. Total lipids were then extracted according to Folch et al. (13). BHT was added to the extraction mixture of cloroform:methanol (2:l) at a final concentration of 0.01% (w/v), in order to avoid oxidation artifacts during the extraction procedure. Fatty Acid transmethylation and separation of polar products. Fatty acids were transmethylated according to Piretti ti. (14,15). TLC separation of the material recovered after transmethylation was performed by loading 1 - 1,5 mg/cm at the start line and developing plates at a constant temperature of 4°C with hexane:diethyl ether mixture (4:1, v/v). The solvent was run for about 12 cm in order to obtain a good resolution of the bands containing the oxidation products. Under these conditions, the methylesters without polar groups posses a RF = 0.50, while the hydroxy esters are characterized by a RF = 0.05. TLC plates were sprayed with Rhodamin 6 G (Merck) and the bands, detected by fluorescence (excitation at 366 nm), were scraped and extracted with ethyl ether stabilized with BHT. Tbe solvent was then evaporated under nitrogen stream and the material dissolved in benzene for GLC and/or GLC-MS analysis. The benzene solutions were stored at -20°C. For the GLC-MS analysis the -0TMS derivatives were obtained by using the TRI-SIL reagent (Pierce). GLC and GLC-MS analysis. For GLC analysis, a Nordion Micromat HRGC 412 gas chromatograph, equipped with a bonded 25 m fused silica capillary column (0,32 mm internal diameter, 0,25 pm film thickness, model 007 FFAP, Quadrex Corporation) was employed. Peaks area were recorded by a Kipp&Zonen bd 71 integrator. The injector and detector (flame ionization detector) temperature was 250°C; the column temperature was initially kept at 180°C and then raised to 240°C at a rate of 5°C per minute and maintained at this value until the chromatogram was completed. The carrier gas was H,, employed at a flow rate of 1.7 ml/mm; splitting was set at a ratio of l/50. GLC-MS was carried out using a Finnigan 1020 GLC-MS combination.
The whole cell total lipid extract was subjected to ART in order to avoid artifacts during the various analytical steps (15). This procedure protects from oxidation allows to perform the simultaneous
reduction
of any hydroperoxy
hydroxy methylester, thereby avoiding hydroperoxide
methylester,
reaction and
to the correspondent
loss. Tbe material recovered after ART was applied
to silica gel TLC plates for the separation of oxidation (hexane/diethyl
during the transmethylation
products. A solvent with a low eluting power
ether 4:l; see Methods) was used and it was possible to evidence three different bands.
GLC analysis of the products of the band characterized by an RF = 0.05 indicated the existence of a peak associable with a hydroxy methylester. In fact, the peak co-elutes with the one corresponding 9-HSA
and lo-HSA
methylesters (Fig.la,lb,lc).
In order to confirm our GLC data, we analyzed the -0TMS by GLC-MS.
to the
The mass spectrum and the fragmentation
derivative of the endogenous polar compound scheme are shown in Fig. 2 and 3 respectively.
Despite the lack of some high molecular weight fragments, which is a typical feature of these compounds, the fragmentation
can be interpreted as representative of a mixture of methylesters derived from 9-HSA
and lo-HSA.
1261
Vol.
178,
No.
3, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Figure 1. (a) GLC profile of the material recovered from the TLC band with RF = 0.05; (b) GLC profile of an oxidized standard of egg lecithin containing different hydroxy derivatives of linoleic and oleic acids methylesters; (c) mixture of (a) plus (b). A relative increase of the peak 2 can be observed. Peak identification: 1) hydroxyoleic acid (cis) methylester 2) mixture of 9- and lo-hydroxystearic acid methylesters 3) hydroxyoleic acid (trans) methylester
259 55
229 03 215
273
iir I- 1 50
Figure 2. Mass spectrum of the 0-TMS derivative of the oxidation product corresponding to the peak 2 in figure la. The spectrum has been recorded using a Finnigan 1020 GLC-MS combination.
1262
Vol. 178, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
a
m/z=22g
.
I
0 CH,- (CH,),+CH;(CH,),-C' 'OCH, . .......... m/z=259 ____--------_-_----_------~ b
0 CH,- (CH,),+CH+ (CH,),--CT OCH,
I
Figure 3. Fragmentation scheme of the 0-TMS (b) IO-hydroxystearic acid methylesters.
The 9-OTMS-stearate
is characterized by the two peaks at m/z=229 and m/z=259 (Fig. 3a) and it seems
to be the main component. m/z=273 Similarly,
derivatives of: (a) 9-hydroxystearic acid and
The
lo-OTMS-stearate,
characterized
by the two peaks at m/z=215 and
(Fig. 3b), is probably the minor component of the mixture. the band identified by RF=OSO was recovered and analyzed by GLC. As expected the results
confirmed the presence of non polar FAMES’. As a second step we investigated the possible existence of a lipid composition
dependence from cell density.
Table 1 shows that the relative amount of each fatty
Table 1. fatty acid composition in samples characterized by different cell densities (87%, 68%, 49% and 32%) Fatty acid 14:o 16:0 16:l 18:0 18:lA 18:lA 18:2 2O:l 20:3 20:4 22:4 22:s 22:6
9 11
Composition % 1.4 17.6 4.6 18.3 32.9 6.2 3.0 1.1 1.2 5.7 1.2 3.3 3.4
2.2 21.9 4.9 17.0 29.5 6.1 2.8 0.7 1.3 5.9 1.4 3.2 3.5
1263
1.7 18.8 4.6 17.1 28.2 6.2 3.3 0.7 1.5 6.5 1.1 3.5 5.5
1.9 19.1 4.4 18.1 27.6 5.9 3.4 0.9 1.2 6.3 1.1 3.4 5.4
Vol.
178,
No.
BIOCHEMICAL
3, 1991
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
0 06 30
40
50
60
70
60
90
100
% of confluence
Figure 4. Correlation between the relative amount of endogenous HSA and the % of cell confluence. Cell confluence was 600000 cells/cm* ; the relative amount of the HSA methylester (9-HSA plus loHSA) was calculated dividing the area of peak 2 in the chromatogram @and with RF=0.05; see figure la) over the area of total FAMES; (band with RF=O.S). The response correction factor for HSA was assumed equal to 1. The linear correlation coefficient is r= 0.997 @
