- Email: [email protected]
Significantly different fatty acid profiles in various phospholipid head groups of neuroblastoma cell lines
Life Sciences, Vol. 58, No. Is, pp. 1X5-1290,19% Cop@ght 0 19% F&vier Science Inc. Printed in the USA. All rights tesetvcd 0024-3205/% $15.00 + .a0
ELSEVIER
PII SOO24-3205(96)00090-2
SIGNIFICANTLY DIFFERENT FATTY ACID PROFILES IN VARIOUS PHOSPHOLIPID HEAD GROUPS OF NEUROBLASTOMA CELL LINES Amitava Dasgupta and Probal Banerjee Department of Pathology, University of New Mexico School of Medicine, Albuquerque, NM, and Department of Chemistry, College of Staten Island/City University of New York, Staten Island, NY (Received in final form February 9, 1996)
We studied lipid profiles of hybrid cells derived from fusion of sympathetic ganglia and neuroblastoma cells (NCB 20, F 11) and also CHO (Chinese hamster ovary) cells. The proportion of saturated to unsaturated fatty acids changed significantly in different phospholipid fractions. We observed 35.4 A of total fatty acids as saturated fatty acids in the phosphatidyl ethanolamine (PE) fraction of NCB-20 cell lines while 47.2% of total fatty acids were saturated in the phosphatidyl inositol (PI) fraction. In general, in neuroblastoma cell lines, we observed the lowest proportion of saturated fatty acids in the PE fraction while the other lipid fraction showed a much higher proportion of saturated fatty acids. On the other hand the PE fraction of a non neuronal cell line, CHO, showed 57.4% saturated fatty acids in the PE fraction. This significant difference in saturated to unsaturated fatty acid ratios in different phospholipid head groups may be linked to the different biological functions of those hybrid cells. Key Wordr:
fatty acid, phospholipid, neuroblastoma cells
Recently, many investigators have utilized tissue culture methodology to study the molecular aspects of information processing by cells from the nervous system. Interactions between projecting neurons and their targets mediate the development and maintenance of specific neural connections. Sperry et al proposed the chemoaffinity hypothesis of synapse formation and suggested that neurons distinguish appropriate from inappropriate synaptic partner cells by interactions between molecules that code cell recognition (1). In order to study the molecular mechanism of such interactions, several investigators used clonal cell lines of CNS lineage (2-5). Hybrid cells derived from somatic cell fusion may manifest properties of differentiated cells and may provide a model system for molecular analysis (6). Hybrid derived from fusion of sympathetic ganglia and neuroblastoma cells are capable of synthesizing dopamine (7). Lee et al
Address correspondence to Dr. Amitava Dasgupta, Pathology Services, University of New Mexico Hospital, 2211 Lomas Blvd, N.E. Albuquerque, NM 87106.
12%
Fatty Acid Profiles in Neuroblastoma Cells
Vol. 58, No. l5,19!X
reported that fusion of hippocampal cells to N18TG2 neuroblastoma cells resulted in phenotypically stable hybrid cell lines that exhibit electrophysiological behavior typical of hippocampal neurons. They also synthesize high levels of nerve growth factors (3). Although the role of acetylcholine and dopamine have been established in signal processing by the neurons, the role of lipids and especially the fatty acid moiety have not been studied. We have previously shown that 5hydroxytryptamine serotonin 1A receptor (5HT r,,) enrichment leads to the enrichment of phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylserine (PS), phosphatidylinositol (PI), and phosphatidic acid (PA) at the expense of sphingolipids and cholesterol. In addition, the composition of phospholipid in the vicinity of G-protein linked receptors could differ from that of the lipid of other membranes. During detergent solubilization of sheep brain gray matter, the overall proportion of saturated fatty acids in PE, PI, PC (major lipid) and PS increased from 50&O% in sheep brain phospholipid to 70-75% in 1.5 % 3-[(3cholamidopropyl) dimethylamino-l-propane sulfonate (CHAPS) solubilized reconstituted and biologically active serotonin 5-HT,, preparation (8). We investigated phospholipid and fatty acid profiles in two hybrid neuroblastoma cell lines (NCB-20 and F-l 1) and one non neuroblastoma cell line (CHO). In order to understand the role of lipid in neuronal cells, we needed a non neuronal control and chose CHO cell lines as our control.
NCB -20 cell line was N18TG2 neuroblastoma cell line hybridized with Chinese hamster brain explant cells (4), while F-l 1 cell line was N18TG2 neuroblastoma cells hybridized with rat’s dorsal root ganglion cells (2). CHO cell line was Chinese hamster ovary type cell line. The cells were maintained in dulbeccos modified eagle medium (DMEM) containing 10% fetal bovine serum and 1% penicillin/streptomycin (v/v). The cells were seeded at 2.5Xld cells per 10 cm dish and allowed to grow for 5 days before harvesting. The NCR20 and F-l 1 cells were harvested by trituration whereas CHO cells were released from the plate by trypsinization. Detached cells were condensed into pellets, washed twice with phosphate buffered saline and then extracted with 2: 1 chloroform/methanol (by vol). The mixture after blending was centrifuged and the supematant obtained was evaporated under nitrogen to obtain a mixture of total lipid which was again dissolved in chloroform/methanol (2: 1 by vol) for storage and further experiments. The high performance thin layer chromatography (HPTLC) analysis of lipid was conducted by putting commercially available lipid standards (Sigma Chemical Company, St. Louis, MO) on the same plate along with lipid extracted from NCB-20, F-l 1 and CHO cell lines. The plates (10 cm X 10 cm) were developed approximately to 7.5 cm from the origin using a solvent system composed of ethyl acetate/l-propanol/chloroform/methanol/0.25% KC1 (25:25:20: 10:9 by volume), dried at room temperature and then developed to full length with hexane/diethyl ether/acetic acid (75:21:4 by vol). After drying, the HPTLC plates were partially exposed to iodine and the respective lipid bands were marked, scraped off and collected from the unexposed parts of the plates as described earlier (8). Transesterification of phospholipid was carried out in 5 mL react&vials with a cone capacity of 0.9 mL (Pierce, Rockford, IL). The scraped band of silica gel containing the appropriate phospholipid fraction was transferred directly to reacti-vials and 2 mL of 14% boron trifluoride in methanol (Sigma Chemical Company, St. Louis, MO) was added. The react&vials were capped under nitrogen with mini inert valves (Pierce, Rockford, IL). The vials were heated in a boiling
Vol. 58, No. 15, 19%
Fatty Acid Profdes in Neuroblastoma Cells
1287
water bath for 15 min and then allowed to equilibrate at room temperature for 20 min. Fatty acid methyl esters derived from individual phospholipids were extracted with hexane and purified using small silica gel columns. The fatty acid methyl esters were eluted from the column using hexane/diethyl ether (80:20 by vol). The gas chromatographic/mass spectrometric analyses (GUMS) were performed using a model 5890 gas chromatogmph coupled to a model 5970 mass selective detector (Hewlett Packard, Palo Alto, CA). The capillary column used for the analysis of fatty acid methyl esters was fused silica cross linked with 5% phenyl methyl silicone with a 0.33 nm film thickness. The initial oven temperature of the gas chromatograph was maintained at 13O”C, and after 1 min of injection, the temperature was raised at a rate of 2”C/min to 2OO”C, following which the oven temperature was increased at a rate of 9”C/min to reach a final oven temperature of 280°C. The mass spectral analysis was carried out using electron impact (scan 40-800 m/z). The individual fatty acids were identified based on the gas chromatographic retention times as well as mass spectral fragmentation patterns. We repeated all experiments three times in order to verify the precision of our results. We expressed fatty acid profiles as mean and standard deviation obtained from triplicate measurements. We used independent t-test in order to determine statistical significance of the differences observed in fatty acid profiles.
One interesting feature of the differential fatty acid profile of different phospholipid head groups was the higher percentage of saturated fatty acids in the PI, PC, and PS fractions compared to the PE fraction in both NCR20 and F-l 1 cell lines. For example, in the lipid extracted from NCB-20 cell line, the PE fraction showed 35.4% saturated fatty acids. However, we observed 47.2%, 48.0% and 45.7% saturated fatty acids in the PI, PS and PC fractions respectively (Table 1). The increases in percentages of saturated fatty acids in PI, PS and PC fractions compared to the PE fraction were statistically significant. A similar trend was observed with the F-l 1 cell line where the PE fraction contained 33.8% saturated fatty acids, the lowest percentage of saturated fatty acids compared to the other phospholipid fractions. In sharp contrast, we observed 57.4% saturated fatty acids in the PE fraction obtained from the CHO cell line, the control. Another interesting feature was the significant increase in percentage of palmitic acid (16:0) in the PC fraction compared to the PE fraction in lipid extracted from NCB-20 and F-l 1 cell lines. The concentration of another major saturated fatty acid, stearic acid (18:0), was reduced in the PC fraction compared to the PE fraction. In contrast, the percentages of stearic acid in the PI and PS fractions were significantly higher than that of the PE fraction while the percentages of palmitic acid were comparable between PI, PS and PE fractions. For example, we observed 28.6% palmitic acid and 13.3% stearic acid in the PC fraction of lipid extracted from NCB-20 cells, while the PE fraction showed 17.0% palmitic acid and 18.4% stearic acid. On the other hand, we observed 17.7% palmitic acid and 26.4% stearic acid in the PI fraction. We also observed 20.4% palmitic acid and 24.6% stearic acid in the PS fraction of lipid extracted from NCB-20 cells (Table 1). A similar trend was also observed with the F-l 1 cell lines where the percentage of palmitic acid was also increased in the PC fraction compared to the PE fraction, while the percentage of stearic acid was reduced in the PC fraction compared to the PE fraction. Interestingly, in the CHO cell line, the control, we observed a comparable percentage of palmitic
1288
Fatty Acid
TABLE
Profiles in Neuroblastoma Cells
Vol. 58, No. 15, 19%
1. Fatty acid profiles in different phaspholipid head groups of lipid extracted from NCB20. F-11 and TCHO ceils
Cells
Fatty Acid
PE
PI
PC
PS
Percent of total fatty acids, Mean (SD), n=3 14:o 16:l 16:0 18:2 18:l 18:0 20:4 2O:l 20:o 2216 22:s 23:l 24: 1 % Saturated % Unsaturated l&.Ll
% Saturated % Unsaturated
14:o 16:l 16:0 18:2 18:l 18:0 20:4 20~3 20:2 20: 1 20:o 22:6 22:s 22:o
14:o 16:l 16:O 18:l 18:O 20:4 22~6 22:s 23:l % Saturated % Unsaturated
1.3(0.4) 5.6(1.0) 17.7(2.6)
2.2(0.5) 7.5(1.0) 28.6(1.8)
32.8(2.9) 26.4(4.2) 3.1(0.3) 3.7(0.8) l.g(O.4) 3.0(0.3) 3.2(0.7)
35.7(2.4) 13.3(2.4) 4.6(1.0) 1.6(0.2) 3.1(0.6) 2.5(0.5)
0.4(0.1) 7.5(0.8) 20.4(2.9) 4.4(0.6) 29.7(3.4) 24.6(3.8) 2.20.4) 4.1(0.6) 2.6(1.1) 2.1(0.4) 1.9(0.3)
47.2(5.3)* 51.4(5.3)*
45.7(2.9)* 53.4(2.9)*
48.q4.8)* 51.9(4.9)Y
2.3(0.6) 5.0(1.3) 27.q3.8)
3.5(0.5) 9.4(1.4) 22.2(2.6) 3.5(1.2) 24.1(4.8) 23.5(3.6) 1.6(0.4)
2.7(0.4) 2.9(0.2)
2.0(1.2) 6.7(0.6) 16.7(3.4) 3.9(0.6) 21.3(3.4) 24.1(4.6) 1.4 (0.3) 5.5(1.5) 6.2(1.3) 3.4(0.2) 0.9(0.4) 3.5(0.3) 2.5(0.3)
33.8(5.4) 65.1(5.5)
43.7(4.1)^ 54.4(4.1)^
1.9(0.4) 1.9(0.5) 25.3(1.7) 40.6(2.9) 30.3(1.2)
1.2(0.4) 2.6(1.0) 17.3(2.0) 23.8(0.8) 21.3(2.7) O.g(O.2) 12.9(0.4) 12.q0.6) g.l(O.9) 39.8(3.5)* 60.1(3.6)*
9.8(0.8) 17.0(3.3) 26.5(1.4) 18.4(1.8) 4.7(0.7) 4.8(0.4) 5.2(0.8) 3.7(0.3) 4.9(1.0) 4.8(l.l) 35.4(2.9) 64.5(3.0) 1.6(0.6) 5.6(1.3) 14.9(4.1) 33.3(4.8) 17.4(2.5) 3.6(0.7) 5.4(0.4) 5.7(0.4) 5.9(1.3)
57.4(1.4) 42.5(1.4)
38.7(1.2) 11.2(0.8) 2.6(0.5) 1.7(0.3) 2.2(0.2) 1.3(0.4) 3.qo. 1) 2.9(0.1) 1.4(0.4) 43.2(2.9)56.2(2.9)-
6.q1.6) 1.9(0.4) 2. I(O.2) 2.1(0.2) 51.1(6.2)* 48.8(6.3)*
5.3(0.4) 9.5( 1.0) 25.9(1.5) 43.3(2.5) 15.9(0.8)
i.l(O.5) 23.3(2.4) 29.1(2.5) 42.4(3.2)
47.1(2.9)* 52.8(2.8)*
65.6(2.3)* 34.2(2.3)*
*Statisticallysignificant difference from PE fraction by t-test, two tailed @<0.05), difference from PE fraction by t-test, one tailed (pCO.05).
* statistically significant
Vol. 58, No. 15, 1996
Fatty Acid Profiles in Neuroblastoma Cells
1289
acid in both the PC and PE fraction, although the percentage of stearic acid was significantly reduced in the PC fraction compared to the PE fraction (Table. 1). We did not observe any arachidonic acid ( < 0.5 % of total) in the PE fraction of lipid extracted from CHO cells, while a small amount of amchidonic acid was present in the PE fraction of lipid extracted from NCB-20 and F-l 1 cells. Another interesting observation was the association of a significant amount of very long chain fatty acids (23: 1 and 24: 1) in the PE fraction of lipid extracted from NCB-20 cells. The plasma membrane of most cells is attached to the intracellular organelles. In addition, there is a dynamic process of endocytosis which allows internalization of parts of the plasma membrane. Therefore, the large pool of phospholipid in the plasma membrane is continuously undergoing exchange with the intracellular organelles. This study was carried out on whole cell lipids of which a large portion was from plasma membranes. Since our present data show that the neuronderived cell lines have a discrete pattern of phospholipid-linked fatty acids which is quite different from that of the non-neural CHO cells, our future research will entail isolation of purified plasma membranes followed by similar detailed analysis of the fatty acid composition of each phospholipid. The presence of a higher proportion of palmitic acid in PC could be significant for the following reasons. Intracellular and cytosolic proteins use palmitic acid or myristic acid anchor for remaining attached to the membrane, whereas stearic acid anchor is not known. PC is found in both the inner and outer membrane and it is also used by phospholipase C as a substrate to release diacylglycerol which stimulates protein kinase C. PI is a well known intracellular lipid which plays an important role in signal transduction and synthesis of both phosphatidylinositol 1,4,5triphosphate (PI, )and inositol 1,4,5triphosphate (IP, ). The presence of a higher proportion of stearic acid in PI could help target the otherwise uncharged lipid to both cytoskeletal as well as plasma membrane Recently, lipid peroxidation has been implicated in the mechanism of several diseases. Knight recently published an excellent review on this subject (9). Moreover, Knight et al also demonstrated peroxidation of unsaturated fatty acids in the presence of transition metals (10). A higher proportion of unsaturated fatty acids in the PE fraction may result in more susceptibility of PE to lipid peroxidation since transitional metal cations are present in the biological system. The significant increases in saturated fatty acids in PS, PI and PC fractions compared to PE fraction were observed only in NCB-20 and F-l 1 cell lines which were hybrid cell lines derived from neuroblastoma cells. Therefore, fatty acid moieties of phospholipid may play an indirect role in signal transduction or other possible roles of neurons. We are currently investigating other cell lines and receptor systems to explore whether such trend could also be present in other systems.
1. 2. 3. 4.
R.W. SPERRY, Proc Natl Acad Sci, U.S.A. 50 703-710 (1963) D. PLATIKA, M.H. BOULOS, L. BAIZER, M.C. FISHMAN, Proc Natl Acad Sci, U.S.A. 82 3499-3503 (1985) H.J. LEE, D.N. HAMMOND, T.H. LARGE, J.D. RODBACK, J.A. SIM, D.A. BROWN, U.H. O’ITEN, B.H. WARNER, J Neuroscience Xl 1779-1787 (1990) M. NIRENBERG, M. Wilson, H. HIGASHIDA, A. ROTTER, K. KRUEGER, N.
1290
5. 6. 7. 8. 9. 10.
Fatty Acid Profdes in Neuroblastoma Cells
Vol. 58, No. 15, 19%
BUSIS, R. RAY, J.G. KENIMER, M. ALDER, Science. 222 794-799 (1983) M. BENEDEl’TI, A. LEVI, M.V. CHAO, Proc Nat1 Acad Sci, U.S.A. M 7859-7863 (1993) R.L. DAVIDSON, Nat1 Cancer Inst Monogr 48 21-30 (1978) L.A. GREENE, W. SHAIN, A. CHALAZONlTIS, X. BREAKFIELD, J. MINNA, H.G. COON, M. NIRENBERG. Proc Natl Acad Sci, U.S.A. 22 4923-4927 (1975) P. BANERJEE, G. DAWSON, A. DASGUPTA, B&him, Biophys Acta .LLlO 65-74 (1992) J.A. KNIGHT, Annal Clin Lab Sci 25 111-121 (1995) J.A. KNIGHT, L. MCCLELLAN, Annal Clin Lab Sci 19 377-384 (1989)
ELSEVIER
PII SOO24-3205(96)00090-2
SIGNIFICANTLY DIFFERENT FATTY ACID PROFILES IN VARIOUS PHOSPHOLIPID HEAD GROUPS OF NEUROBLASTOMA CELL LINES Amitava Dasgupta and Probal Banerjee Department of Pathology, University of New Mexico School of Medicine, Albuquerque, NM, and Department of Chemistry, College of Staten Island/City University of New York, Staten Island, NY (Received in final form February 9, 1996)
We studied lipid profiles of hybrid cells derived from fusion of sympathetic ganglia and neuroblastoma cells (NCB 20, F 11) and also CHO (Chinese hamster ovary) cells. The proportion of saturated to unsaturated fatty acids changed significantly in different phospholipid fractions. We observed 35.4 A of total fatty acids as saturated fatty acids in the phosphatidyl ethanolamine (PE) fraction of NCB-20 cell lines while 47.2% of total fatty acids were saturated in the phosphatidyl inositol (PI) fraction. In general, in neuroblastoma cell lines, we observed the lowest proportion of saturated fatty acids in the PE fraction while the other lipid fraction showed a much higher proportion of saturated fatty acids. On the other hand the PE fraction of a non neuronal cell line, CHO, showed 57.4% saturated fatty acids in the PE fraction. This significant difference in saturated to unsaturated fatty acid ratios in different phospholipid head groups may be linked to the different biological functions of those hybrid cells. Key Wordr:
fatty acid, phospholipid, neuroblastoma cells
Recently, many investigators have utilized tissue culture methodology to study the molecular aspects of information processing by cells from the nervous system. Interactions between projecting neurons and their targets mediate the development and maintenance of specific neural connections. Sperry et al proposed the chemoaffinity hypothesis of synapse formation and suggested that neurons distinguish appropriate from inappropriate synaptic partner cells by interactions between molecules that code cell recognition (1). In order to study the molecular mechanism of such interactions, several investigators used clonal cell lines of CNS lineage (2-5). Hybrid cells derived from somatic cell fusion may manifest properties of differentiated cells and may provide a model system for molecular analysis (6). Hybrid derived from fusion of sympathetic ganglia and neuroblastoma cells are capable of synthesizing dopamine (7). Lee et al
Address correspondence to Dr. Amitava Dasgupta, Pathology Services, University of New Mexico Hospital, 2211 Lomas Blvd, N.E. Albuquerque, NM 87106.
12%
Fatty Acid Profiles in Neuroblastoma Cells
Vol. 58, No. l5,19!X
reported that fusion of hippocampal cells to N18TG2 neuroblastoma cells resulted in phenotypically stable hybrid cell lines that exhibit electrophysiological behavior typical of hippocampal neurons. They also synthesize high levels of nerve growth factors (3). Although the role of acetylcholine and dopamine have been established in signal processing by the neurons, the role of lipids and especially the fatty acid moiety have not been studied. We have previously shown that 5hydroxytryptamine serotonin 1A receptor (5HT r,,) enrichment leads to the enrichment of phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylserine (PS), phosphatidylinositol (PI), and phosphatidic acid (PA) at the expense of sphingolipids and cholesterol. In addition, the composition of phospholipid in the vicinity of G-protein linked receptors could differ from that of the lipid of other membranes. During detergent solubilization of sheep brain gray matter, the overall proportion of saturated fatty acids in PE, PI, PC (major lipid) and PS increased from 50&O% in sheep brain phospholipid to 70-75% in 1.5 % 3-[(3cholamidopropyl) dimethylamino-l-propane sulfonate (CHAPS) solubilized reconstituted and biologically active serotonin 5-HT,, preparation (8). We investigated phospholipid and fatty acid profiles in two hybrid neuroblastoma cell lines (NCB-20 and F-l 1) and one non neuroblastoma cell line (CHO). In order to understand the role of lipid in neuronal cells, we needed a non neuronal control and chose CHO cell lines as our control.
NCB -20 cell line was N18TG2 neuroblastoma cell line hybridized with Chinese hamster brain explant cells (4), while F-l 1 cell line was N18TG2 neuroblastoma cells hybridized with rat’s dorsal root ganglion cells (2). CHO cell line was Chinese hamster ovary type cell line. The cells were maintained in dulbeccos modified eagle medium (DMEM) containing 10% fetal bovine serum and 1% penicillin/streptomycin (v/v). The cells were seeded at 2.5Xld cells per 10 cm dish and allowed to grow for 5 days before harvesting. The NCR20 and F-l 1 cells were harvested by trituration whereas CHO cells were released from the plate by trypsinization. Detached cells were condensed into pellets, washed twice with phosphate buffered saline and then extracted with 2: 1 chloroform/methanol (by vol). The mixture after blending was centrifuged and the supematant obtained was evaporated under nitrogen to obtain a mixture of total lipid which was again dissolved in chloroform/methanol (2: 1 by vol) for storage and further experiments. The high performance thin layer chromatography (HPTLC) analysis of lipid was conducted by putting commercially available lipid standards (Sigma Chemical Company, St. Louis, MO) on the same plate along with lipid extracted from NCB-20, F-l 1 and CHO cell lines. The plates (10 cm X 10 cm) were developed approximately to 7.5 cm from the origin using a solvent system composed of ethyl acetate/l-propanol/chloroform/methanol/0.25% KC1 (25:25:20: 10:9 by volume), dried at room temperature and then developed to full length with hexane/diethyl ether/acetic acid (75:21:4 by vol). After drying, the HPTLC plates were partially exposed to iodine and the respective lipid bands were marked, scraped off and collected from the unexposed parts of the plates as described earlier (8). Transesterification of phospholipid was carried out in 5 mL react&vials with a cone capacity of 0.9 mL (Pierce, Rockford, IL). The scraped band of silica gel containing the appropriate phospholipid fraction was transferred directly to reacti-vials and 2 mL of 14% boron trifluoride in methanol (Sigma Chemical Company, St. Louis, MO) was added. The react&vials were capped under nitrogen with mini inert valves (Pierce, Rockford, IL). The vials were heated in a boiling
Vol. 58, No. 15, 19%
Fatty Acid Profdes in Neuroblastoma Cells
1287
water bath for 15 min and then allowed to equilibrate at room temperature for 20 min. Fatty acid methyl esters derived from individual phospholipids were extracted with hexane and purified using small silica gel columns. The fatty acid methyl esters were eluted from the column using hexane/diethyl ether (80:20 by vol). The gas chromatographic/mass spectrometric analyses (GUMS) were performed using a model 5890 gas chromatogmph coupled to a model 5970 mass selective detector (Hewlett Packard, Palo Alto, CA). The capillary column used for the analysis of fatty acid methyl esters was fused silica cross linked with 5% phenyl methyl silicone with a 0.33 nm film thickness. The initial oven temperature of the gas chromatograph was maintained at 13O”C, and after 1 min of injection, the temperature was raised at a rate of 2”C/min to 2OO”C, following which the oven temperature was increased at a rate of 9”C/min to reach a final oven temperature of 280°C. The mass spectral analysis was carried out using electron impact (scan 40-800 m/z). The individual fatty acids were identified based on the gas chromatographic retention times as well as mass spectral fragmentation patterns. We repeated all experiments three times in order to verify the precision of our results. We expressed fatty acid profiles as mean and standard deviation obtained from triplicate measurements. We used independent t-test in order to determine statistical significance of the differences observed in fatty acid profiles.
One interesting feature of the differential fatty acid profile of different phospholipid head groups was the higher percentage of saturated fatty acids in the PI, PC, and PS fractions compared to the PE fraction in both NCR20 and F-l 1 cell lines. For example, in the lipid extracted from NCB-20 cell line, the PE fraction showed 35.4% saturated fatty acids. However, we observed 47.2%, 48.0% and 45.7% saturated fatty acids in the PI, PS and PC fractions respectively (Table 1). The increases in percentages of saturated fatty acids in PI, PS and PC fractions compared to the PE fraction were statistically significant. A similar trend was observed with the F-l 1 cell line where the PE fraction contained 33.8% saturated fatty acids, the lowest percentage of saturated fatty acids compared to the other phospholipid fractions. In sharp contrast, we observed 57.4% saturated fatty acids in the PE fraction obtained from the CHO cell line, the control. Another interesting feature was the significant increase in percentage of palmitic acid (16:0) in the PC fraction compared to the PE fraction in lipid extracted from NCB-20 and F-l 1 cell lines. The concentration of another major saturated fatty acid, stearic acid (18:0), was reduced in the PC fraction compared to the PE fraction. In contrast, the percentages of stearic acid in the PI and PS fractions were significantly higher than that of the PE fraction while the percentages of palmitic acid were comparable between PI, PS and PE fractions. For example, we observed 28.6% palmitic acid and 13.3% stearic acid in the PC fraction of lipid extracted from NCB-20 cells, while the PE fraction showed 17.0% palmitic acid and 18.4% stearic acid. On the other hand, we observed 17.7% palmitic acid and 26.4% stearic acid in the PI fraction. We also observed 20.4% palmitic acid and 24.6% stearic acid in the PS fraction of lipid extracted from NCB-20 cells (Table 1). A similar trend was also observed with the F-l 1 cell lines where the percentage of palmitic acid was also increased in the PC fraction compared to the PE fraction, while the percentage of stearic acid was reduced in the PC fraction compared to the PE fraction. Interestingly, in the CHO cell line, the control, we observed a comparable percentage of palmitic
1288
Fatty Acid
TABLE
Profiles in Neuroblastoma Cells
Vol. 58, No. 15, 19%
1. Fatty acid profiles in different phaspholipid head groups of lipid extracted from NCB20. F-11 and TCHO ceils
Cells
Fatty Acid
PE
PI
PC
PS
Percent of total fatty acids, Mean (SD), n=3 14:o 16:l 16:0 18:2 18:l 18:0 20:4 2O:l 20:o 2216 22:s 23:l 24: 1 % Saturated % Unsaturated l&.Ll
% Saturated % Unsaturated
14:o 16:l 16:0 18:2 18:l 18:0 20:4 20~3 20:2 20: 1 20:o 22:6 22:s 22:o
14:o 16:l 16:O 18:l 18:O 20:4 22~6 22:s 23:l % Saturated % Unsaturated
1.3(0.4) 5.6(1.0) 17.7(2.6)
2.2(0.5) 7.5(1.0) 28.6(1.8)
32.8(2.9) 26.4(4.2) 3.1(0.3) 3.7(0.8) l.g(O.4) 3.0(0.3) 3.2(0.7)
35.7(2.4) 13.3(2.4) 4.6(1.0) 1.6(0.2) 3.1(0.6) 2.5(0.5)
0.4(0.1) 7.5(0.8) 20.4(2.9) 4.4(0.6) 29.7(3.4) 24.6(3.8) 2.20.4) 4.1(0.6) 2.6(1.1) 2.1(0.4) 1.9(0.3)
47.2(5.3)* 51.4(5.3)*
45.7(2.9)* 53.4(2.9)*
48.q4.8)* 51.9(4.9)Y
2.3(0.6) 5.0(1.3) 27.q3.8)
3.5(0.5) 9.4(1.4) 22.2(2.6) 3.5(1.2) 24.1(4.8) 23.5(3.6) 1.6(0.4)
2.7(0.4) 2.9(0.2)
2.0(1.2) 6.7(0.6) 16.7(3.4) 3.9(0.6) 21.3(3.4) 24.1(4.6) 1.4 (0.3) 5.5(1.5) 6.2(1.3) 3.4(0.2) 0.9(0.4) 3.5(0.3) 2.5(0.3)
33.8(5.4) 65.1(5.5)
43.7(4.1)^ 54.4(4.1)^
1.9(0.4) 1.9(0.5) 25.3(1.7) 40.6(2.9) 30.3(1.2)
1.2(0.4) 2.6(1.0) 17.3(2.0) 23.8(0.8) 21.3(2.7) O.g(O.2) 12.9(0.4) 12.q0.6) g.l(O.9) 39.8(3.5)* 60.1(3.6)*
9.8(0.8) 17.0(3.3) 26.5(1.4) 18.4(1.8) 4.7(0.7) 4.8(0.4) 5.2(0.8) 3.7(0.3) 4.9(1.0) 4.8(l.l) 35.4(2.9) 64.5(3.0) 1.6(0.6) 5.6(1.3) 14.9(4.1) 33.3(4.8) 17.4(2.5) 3.6(0.7) 5.4(0.4) 5.7(0.4) 5.9(1.3)
57.4(1.4) 42.5(1.4)
38.7(1.2) 11.2(0.8) 2.6(0.5) 1.7(0.3) 2.2(0.2) 1.3(0.4) 3.qo. 1) 2.9(0.1) 1.4(0.4) 43.2(2.9)56.2(2.9)-
6.q1.6) 1.9(0.4) 2. I(O.2) 2.1(0.2) 51.1(6.2)* 48.8(6.3)*
5.3(0.4) 9.5( 1.0) 25.9(1.5) 43.3(2.5) 15.9(0.8)
i.l(O.5) 23.3(2.4) 29.1(2.5) 42.4(3.2)
47.1(2.9)* 52.8(2.8)*
65.6(2.3)* 34.2(2.3)*
*Statisticallysignificant difference from PE fraction by t-test, two tailed @<0.05), difference from PE fraction by t-test, one tailed (pCO.05).
* statistically significant
Vol. 58, No. 15, 1996
Fatty Acid Profiles in Neuroblastoma Cells
1289
acid in both the PC and PE fraction, although the percentage of stearic acid was significantly reduced in the PC fraction compared to the PE fraction (Table. 1). We did not observe any arachidonic acid ( < 0.5 % of total) in the PE fraction of lipid extracted from CHO cells, while a small amount of amchidonic acid was present in the PE fraction of lipid extracted from NCB-20 and F-l 1 cells. Another interesting observation was the association of a significant amount of very long chain fatty acids (23: 1 and 24: 1) in the PE fraction of lipid extracted from NCB-20 cells. The plasma membrane of most cells is attached to the intracellular organelles. In addition, there is a dynamic process of endocytosis which allows internalization of parts of the plasma membrane. Therefore, the large pool of phospholipid in the plasma membrane is continuously undergoing exchange with the intracellular organelles. This study was carried out on whole cell lipids of which a large portion was from plasma membranes. Since our present data show that the neuronderived cell lines have a discrete pattern of phospholipid-linked fatty acids which is quite different from that of the non-neural CHO cells, our future research will entail isolation of purified plasma membranes followed by similar detailed analysis of the fatty acid composition of each phospholipid. The presence of a higher proportion of palmitic acid in PC could be significant for the following reasons. Intracellular and cytosolic proteins use palmitic acid or myristic acid anchor for remaining attached to the membrane, whereas stearic acid anchor is not known. PC is found in both the inner and outer membrane and it is also used by phospholipase C as a substrate to release diacylglycerol which stimulates protein kinase C. PI is a well known intracellular lipid which plays an important role in signal transduction and synthesis of both phosphatidylinositol 1,4,5triphosphate (PI, )and inositol 1,4,5triphosphate (IP, ). The presence of a higher proportion of stearic acid in PI could help target the otherwise uncharged lipid to both cytoskeletal as well as plasma membrane Recently, lipid peroxidation has been implicated in the mechanism of several diseases. Knight recently published an excellent review on this subject (9). Moreover, Knight et al also demonstrated peroxidation of unsaturated fatty acids in the presence of transition metals (10). A higher proportion of unsaturated fatty acids in the PE fraction may result in more susceptibility of PE to lipid peroxidation since transitional metal cations are present in the biological system. The significant increases in saturated fatty acids in PS, PI and PC fractions compared to PE fraction were observed only in NCB-20 and F-l 1 cell lines which were hybrid cell lines derived from neuroblastoma cells. Therefore, fatty acid moieties of phospholipid may play an indirect role in signal transduction or other possible roles of neurons. We are currently investigating other cell lines and receptor systems to explore whether such trend could also be present in other systems.
1. 2. 3. 4.
R.W. SPERRY, Proc Natl Acad Sci, U.S.A. 50 703-710 (1963) D. PLATIKA, M.H. BOULOS, L. BAIZER, M.C. FISHMAN, Proc Natl Acad Sci, U.S.A. 82 3499-3503 (1985) H.J. LEE, D.N. HAMMOND, T.H. LARGE, J.D. RODBACK, J.A. SIM, D.A. BROWN, U.H. O’ITEN, B.H. WARNER, J Neuroscience Xl 1779-1787 (1990) M. NIRENBERG, M. Wilson, H. HIGASHIDA, A. ROTTER, K. KRUEGER, N.
1290
5. 6. 7. 8. 9. 10.
Fatty Acid Profdes in Neuroblastoma Cells
Vol. 58, No. 15, 19%
BUSIS, R. RAY, J.G. KENIMER, M. ALDER, Science. 222 794-799 (1983) M. BENEDEl’TI, A. LEVI, M.V. CHAO, Proc Nat1 Acad Sci, U.S.A. M 7859-7863 (1993) R.L. DAVIDSON, Nat1 Cancer Inst Monogr 48 21-30 (1978) L.A. GREENE, W. SHAIN, A. CHALAZONlTIS, X. BREAKFIELD, J. MINNA, H.G. COON, M. NIRENBERG. Proc Natl Acad Sci, U.S.A. 22 4923-4927 (1975) P. BANERJEE, G. DAWSON, A. DASGUPTA, B&him, Biophys Acta .LLlO 65-74 (1992) J.A. KNIGHT, Annal Clin Lab Sci 25 111-121 (1995) J.A. KNIGHT, L. MCCLELLAN, Annal Clin Lab Sci 19 377-384 (1989)