421
Fatty Acid Composition of Brain Phospholipids in Aging and in Alzheimer's Disease M. Soderberg*, C. Edlund, K. Kristensson and G. Dallner Division of Medical Cell and Neurobiology, Clinical Research Center, Huddinge University Hospital, Karolinska Institutet, S-141 86 Huddinge, Sweden
The two major phospholipid classes, namely, phosphatidylethanolamines {PE} and phosphatidylcholines {PC), were studied in four different regions of human brain,/.e., in frontal gray matter, frontal white matter, hippocampus and in pons. The fatty acid (FA) compositions of these phospholipids were found to be specific for the different regions. PC contains mostly saturated and 18:1 FA, while PE is rich in polyunsaturated FA. Aging has no influence on the FA compositions, while in Alzheimer's disease {AD) PE is modified in all four regions, particularly in frontal gray matter and in hippocampus. The abundance of the major monounsaturated FA of PE, 18:1, is not significantly altered in Alzheimer's disease, but there is a substantial increase in the relative amounts of the saturated components 14:0, 16:0 and 18:0. This is paralleled by a decrease in the polyunsaturated FA 20:4, 22:4 and 22:6. It is not clear whether the changes observed are specific for AD. Changes in saturated/polyunsaturated FA ratio are likely to influence cellular function, which in turn may cause certain neural deficiencies. The findings do not sup~ port the hypothesis that AD reflects an accelerated aging process. Lipids 26, 421-425 (199D.
study a decreased n-6 PUFA content was observed in PC from the frontal cortex of AD patients, while no significant changes in the FA composition of PE were found. In light of these facts, it appeared timely and important to initiate a study of the lipid compositions of different parts of the brain in AD. The study is concerned with changes in the FA composition of PE and PC in four different regions of the brain.
MATERIALS AND METHODS
Brain samples were obtained during autopsy approximately 24 hr after death. The contralateral brain hemisphere was fixed for histopathological examination. In histological preparations various silver staining techniques were employed. Patients with clinical dementia were diagnosed according to the criteria established for AD by the National Health Institute/American Association of Retired People (NHI/AARP} working group (10). All patients included in this study showed substantial quantities of neuritic plaques, well above the minimum number required to verify the diagnosis by the abovementioned criteria. In addition, neurofibrillary tangles were present in the neocortex in all patients with AD. Patients from which control samples were taken died Phospholipids (PL} are structurally and functionally im- from causes unrelated to the nervous system and showed portant constituents of membranes. They regulate mem- no clinical manifestations of dementia or other mental brane structure and permeability, as well as the activities disorders. Microscopic investigation of the control brains of membrane~bound enzymes. The lipid composition of the revealed normal cytoarchitecture in the neocortex. In some brain--especially in regard to the high concentrations of of the older patients small numbers of neuritic plaques cholesterol, of the polyisoprenoid compounds dolichol and were found as is usual at advanced ag~ ubiquinone, and of glycolipids--differs significantly from Brain regions chosen for lipid analysis were removed and those of other tissues {1,2}. It is generally assumed that homogenized in 0.25 M sucrose with an Ultra-Turrax cholesterol in model membranes preferentially associates blender {Janke & Kunkel, Staufen, Germany} for 2 rain with phosphatidylcholine (PC) {3} and that isoprenoids {24000 rpm}. The resulting 10% homogenate was stored located in the hydrophobic portion of phospholipid bi- at -20~ until assayed, which was usually within 1-2 wk. layers prefer phosphatidylethanolamine {PE) {4). Changes White matter was taken from the frontal lob~ in the amounts and types of PL could thus, in turn, also To extract lipids, 6 mL methanol and 6 mL chloroform affect the neutral lipid components of membranes. were added to 2 mL of the homogenate to obtain a final Changes in the composition and metabolism of brain chloroform/methanol/water ratio of 1:1:0.3 (v/v/v}. Extraclipids have been proposed to contribute to the deteriora- tion was performed at 40~ for 60 min with magnetic stirtion of central nervous system function with increasing ring. Phase separation was obtained by addition of chlorc~ age {2,5}. It has also been proposed that phosphoinositide form and water to give a final chloroform/methanol/water levels in brain, and of some of the metabolites of PE and ratio of 3:2:1 {v/v/v}.The tubes were then centrifuged for PC, are involved in the processes that cause various types 5 rain at 250 • g. The lower phase was washed 4 times of senile dementia {6,7}. In one study, the content of with theoretical upper phase (chloroform/methanol/water, polyunsaturated fatty acids {FA}{PUFA) in total PL from 3:48:47, v/v/v} and evaporated under nitrogen. The lipid frontal brain cortex was found to decrease with increas- residue was dissolved in 5 mL chloroform and placed onto ing age, while the PUFA content in Alzheimer's disease a Silica-SepPak cartridge {Millipore, Waters Associates, {AD} varied {8}. In contrast, another investigation found Milford, MA}, which was then washed with 25 mL chlorono differences in FA composition in aging {9}. In this latter form to remove neutral lipids. PL were eluted from the Silica-SepPak with 25 mL methanol. The solvent was *To whom correspondence should be addressed. evaporated under N2 and lipids were dissolved in 200 ~L Abbreviations: AD, Alzheimer'sdisease; FA, fatty acids; GLC,gasliquid chromatography;NADH, nicotinamideadeninedinucleotide; chloroform/methanol 2:1 {v/v}. Twenty gL of the extract PC, phosphatidylcholine; PE, phosphatidylethanolamine; PL, was applied to a Silica gel 60 plate {Merck, Darmphospholipids; PUFA, polyunsaturatedfatty acids; TLC,thin-layer stadt, Germany} for two-dimensional thin-layer chrochromatography. matography (TLC}. The plates were developed with LIPIDS,Vol. 26, No. 6 (1991)
422
M. SODERBERG ET AL. TABLE 1
Phospholipid Compositions of Different Regions of the Human Brain a
Region
Age (yrs)
Total phospholipid amount (mg/g)
Phospholipidb (mg/g) PE
PC
PC/PE ratio
Frontal gray
33-36 54-57 69-72 89-92
36.4 35.4 34.8 32.8
__. 3.0 _+ 2.6 -4--3.8 +_ 3.2 c
13.3 12.6 12.3 12.0
+__ 1.5 _+ 1.6 __ 1.4 ___ 1.0
16.1 15.5 14.7 13.9
_ 1.5 _+ 1.8 __ 1.7 _+ 1.2c
1.21 1.23 1.20 1.16
_+ 0.08 ___ 0.07 _+ 0.09 + 0.08
White matter
33-36 54-57 69-72 89-92 33-36 54-57 69-72 89-92 33-36 54-57 69-72 89-92
70.0 69.0 59.8 57.6 46.0 42.8 37.0 35.0 64.8 62.2 56.4 53.2
+_ 4.6 _+ 8.6 + 8.4c _+ 7.6 c + 4.0 -4-_4.4 +- 4.2 c -!-_4.4c -4--4.6 +_ 5.0 +_ 6.2 c _+ 7.2 c
29.5 29.9 25.1 24.7 18.1 17.0 14.4 13.4 31.4 30.4 27.1 25.4
_+ 3.9 _+ 4.3 _+ 2.5 c __ 3.5 c +_ 2.4 +_ 2.0 +_ 1.5c +_ 1.8c + 3.4 + 2.2 ___ 2.7 c _ 2.7 c
19.4 20.1 17.2 16.5 19.3 17.4 15.0 14.0 20.0 19.5 17.5 15.8
_ 2.4 _+ 2.0 -!-_2.4c +__2.6c +_ 1.8 + 1.5 __ 1.6c _+ 1.0c _+ 1.7 _+ 2.5 + 1.8c + 1.6c
0.66 0.67 0.69 0.67 1.07 1.02 1.04 1.05 0.64 0.64 0.65 0.66
+_ 0.05 +_ 0.06 _ 0.04 _+ 0.06 +_ 0.09 _+ 0.05 __ 0.06 ___ 0.07 _+ 0.05 +_ 0.04 -!-_0.05 -!--0.04
Hippocampus
Pons
aThe values are the means + SD of 8-10 measurements. bpE, phosphatidylethanolamine; PC, phosphatidylcholine. c Significantly decreased compared with the youngest age group ip
Fatty Acid Composition of Brain Phospholipids in Aging and in Alzheimer's Disease M. Soderberg*, C. Edlund, K. Kristensson and G. Dallner Division of Medical Cell and Neurobiology, Clinical Research Center, Huddinge University Hospital, Karolinska Institutet, S-141 86 Huddinge, Sweden
The two major phospholipid classes, namely, phosphatidylethanolamines {PE} and phosphatidylcholines {PC), were studied in four different regions of human brain,/.e., in frontal gray matter, frontal white matter, hippocampus and in pons. The fatty acid (FA) compositions of these phospholipids were found to be specific for the different regions. PC contains mostly saturated and 18:1 FA, while PE is rich in polyunsaturated FA. Aging has no influence on the FA compositions, while in Alzheimer's disease {AD) PE is modified in all four regions, particularly in frontal gray matter and in hippocampus. The abundance of the major monounsaturated FA of PE, 18:1, is not significantly altered in Alzheimer's disease, but there is a substantial increase in the relative amounts of the saturated components 14:0, 16:0 and 18:0. This is paralleled by a decrease in the polyunsaturated FA 20:4, 22:4 and 22:6. It is not clear whether the changes observed are specific for AD. Changes in saturated/polyunsaturated FA ratio are likely to influence cellular function, which in turn may cause certain neural deficiencies. The findings do not sup~ port the hypothesis that AD reflects an accelerated aging process. Lipids 26, 421-425 (199D.
study a decreased n-6 PUFA content was observed in PC from the frontal cortex of AD patients, while no significant changes in the FA composition of PE were found. In light of these facts, it appeared timely and important to initiate a study of the lipid compositions of different parts of the brain in AD. The study is concerned with changes in the FA composition of PE and PC in four different regions of the brain.
MATERIALS AND METHODS
Brain samples were obtained during autopsy approximately 24 hr after death. The contralateral brain hemisphere was fixed for histopathological examination. In histological preparations various silver staining techniques were employed. Patients with clinical dementia were diagnosed according to the criteria established for AD by the National Health Institute/American Association of Retired People (NHI/AARP} working group (10). All patients included in this study showed substantial quantities of neuritic plaques, well above the minimum number required to verify the diagnosis by the abovementioned criteria. In addition, neurofibrillary tangles were present in the neocortex in all patients with AD. Patients from which control samples were taken died Phospholipids (PL} are structurally and functionally im- from causes unrelated to the nervous system and showed portant constituents of membranes. They regulate mem- no clinical manifestations of dementia or other mental brane structure and permeability, as well as the activities disorders. Microscopic investigation of the control brains of membrane~bound enzymes. The lipid composition of the revealed normal cytoarchitecture in the neocortex. In some brain--especially in regard to the high concentrations of of the older patients small numbers of neuritic plaques cholesterol, of the polyisoprenoid compounds dolichol and were found as is usual at advanced ag~ ubiquinone, and of glycolipids--differs significantly from Brain regions chosen for lipid analysis were removed and those of other tissues {1,2}. It is generally assumed that homogenized in 0.25 M sucrose with an Ultra-Turrax cholesterol in model membranes preferentially associates blender {Janke & Kunkel, Staufen, Germany} for 2 rain with phosphatidylcholine (PC) {3} and that isoprenoids {24000 rpm}. The resulting 10% homogenate was stored located in the hydrophobic portion of phospholipid bi- at -20~ until assayed, which was usually within 1-2 wk. layers prefer phosphatidylethanolamine {PE) {4). Changes White matter was taken from the frontal lob~ in the amounts and types of PL could thus, in turn, also To extract lipids, 6 mL methanol and 6 mL chloroform affect the neutral lipid components of membranes. were added to 2 mL of the homogenate to obtain a final Changes in the composition and metabolism of brain chloroform/methanol/water ratio of 1:1:0.3 (v/v/v}. Extraclipids have been proposed to contribute to the deteriora- tion was performed at 40~ for 60 min with magnetic stirtion of central nervous system function with increasing ring. Phase separation was obtained by addition of chlorc~ age {2,5}. It has also been proposed that phosphoinositide form and water to give a final chloroform/methanol/water levels in brain, and of some of the metabolites of PE and ratio of 3:2:1 {v/v/v}.The tubes were then centrifuged for PC, are involved in the processes that cause various types 5 rain at 250 • g. The lower phase was washed 4 times of senile dementia {6,7}. In one study, the content of with theoretical upper phase (chloroform/methanol/water, polyunsaturated fatty acids {FA}{PUFA) in total PL from 3:48:47, v/v/v} and evaporated under nitrogen. The lipid frontal brain cortex was found to decrease with increas- residue was dissolved in 5 mL chloroform and placed onto ing age, while the PUFA content in Alzheimer's disease a Silica-SepPak cartridge {Millipore, Waters Associates, {AD} varied {8}. In contrast, another investigation found Milford, MA}, which was then washed with 25 mL chlorono differences in FA composition in aging {9}. In this latter form to remove neutral lipids. PL were eluted from the Silica-SepPak with 25 mL methanol. The solvent was *To whom correspondence should be addressed. evaporated under N2 and lipids were dissolved in 200 ~L Abbreviations: AD, Alzheimer'sdisease; FA, fatty acids; GLC,gasliquid chromatography;NADH, nicotinamideadeninedinucleotide; chloroform/methanol 2:1 {v/v}. Twenty gL of the extract PC, phosphatidylcholine; PE, phosphatidylethanolamine; PL, was applied to a Silica gel 60 plate {Merck, Darmphospholipids; PUFA, polyunsaturatedfatty acids; TLC,thin-layer stadt, Germany} for two-dimensional thin-layer chrochromatography. matography (TLC}. The plates were developed with LIPIDS,Vol. 26, No. 6 (1991)
422
M. SODERBERG ET AL. TABLE 1
Phospholipid Compositions of Different Regions of the Human Brain a
Region
Age (yrs)
Total phospholipid amount (mg/g)
Phospholipidb (mg/g) PE
PC
PC/PE ratio
Frontal gray
33-36 54-57 69-72 89-92
36.4 35.4 34.8 32.8
__. 3.0 _+ 2.6 -4--3.8 +_ 3.2 c
13.3 12.6 12.3 12.0
+__ 1.5 _+ 1.6 __ 1.4 ___ 1.0
16.1 15.5 14.7 13.9
_ 1.5 _+ 1.8 __ 1.7 _+ 1.2c
1.21 1.23 1.20 1.16
_+ 0.08 ___ 0.07 _+ 0.09 + 0.08
White matter
33-36 54-57 69-72 89-92 33-36 54-57 69-72 89-92 33-36 54-57 69-72 89-92
70.0 69.0 59.8 57.6 46.0 42.8 37.0 35.0 64.8 62.2 56.4 53.2
+_ 4.6 _+ 8.6 + 8.4c _+ 7.6 c + 4.0 -4-_4.4 +- 4.2 c -!-_4.4c -4--4.6 +_ 5.0 +_ 6.2 c _+ 7.2 c
29.5 29.9 25.1 24.7 18.1 17.0 14.4 13.4 31.4 30.4 27.1 25.4
_+ 3.9 _+ 4.3 _+ 2.5 c __ 3.5 c +_ 2.4 +_ 2.0 +_ 1.5c +_ 1.8c + 3.4 + 2.2 ___ 2.7 c _ 2.7 c
19.4 20.1 17.2 16.5 19.3 17.4 15.0 14.0 20.0 19.5 17.5 15.8
_ 2.4 _+ 2.0 -!-_2.4c +__2.6c +_ 1.8 + 1.5 __ 1.6c _+ 1.0c _+ 1.7 _+ 2.5 + 1.8c + 1.6c
0.66 0.67 0.69 0.67 1.07 1.02 1.04 1.05 0.64 0.64 0.65 0.66
+_ 0.05 +_ 0.06 _ 0.04 _+ 0.06 +_ 0.09 _+ 0.05 __ 0.06 ___ 0.07 _+ 0.05 +_ 0.04 -!-_0.05 -!--0.04
Hippocampus
Pons
aThe values are the means + SD of 8-10 measurements. bpE, phosphatidylethanolamine; PC, phosphatidylcholine. c Significantly decreased compared with the youngest age group ip
