Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2010 Oct 1.
Published in final edited form as: J Neurochem. 2009 Jul 17;111(1):15–25. doi: 10.1111/j.1471-4159.2009.06290.x

Lipidomic Profiling in Mouse Brain Reveals Differences Between Ages and Genders, with Smaller Changes Associated with α-Synuclein Genotype

Irit Rappley 1,3, David S Myers 2,3, Stephen B Milne 2, Pavlina T Ivanova 2, Matthew J LaVoie 1, H Alex Brown 2,*, Dennis J Selkoe 1,*
PMCID: PMC2752313  NIHMSID: NIHMS135516  PMID: 19627450

Abstract

Advances in lipidomics technology have facilitated the precise detection, identification and profiling of lipid species within tissues. Mass spectrometry allows for identification of lipids as a function of the total number of carbons and double bonds in their acyl chains. Such detailed descriptions of lipid composition can provide a basis for further investigation of cell signaling and metabolic pathways, both physiological and pathological. Here, we applied phospholipid profiling to mouse models relevant to Parkinson's disease (PD), using mice that were transgenic for human α-synuclein (αSyn) or deleted of endogenous αSyn. Proposed functions of αSyn include phospholipid binding, regulation of membrane composition, and regulation of vesicular pools. We investigated whether αSyn gene dosage interacts with differences in phospholipid composition across brain regions or with age-related changes in brain phospholipid composition. The most dramatic phospholipid changes were observed in αSyn wild-type animals as a function of age and gender. αSyn genotype-specific changes were also observed in aged, but not young, mice. Our results provide a detailed and systematic characterization of brain phospholipid composition in mice and identify age-related changes relevant both to PD and to normal aging.

Keywords: Lipidomics, Lipid profiling, α-Synuclein, Brain phospholipid composition

INTRODUCTION

α-synuclein (αSyn) is a small, cytosolic protein that is expressed in several tissues including heart, lung, kidney, and placenta, but is highly enriched in the brain and particularly at presynaptic terminals (Lavedan, 1998; Tofaris and Spillantini, 2005). Although the precise biochemical pathway in which αSyn normally functions is unknown, altered post-translational processing and accumulation of αSyn are likely to play a critical role in the pathogenesis of Parkinson's disease (PD) (for review, see Dauer and Przedborski, 2003). At the genetic level, point mutations in the gene encoding αSyn, or simple multiplication of the wild-type locus, can cause autosomal dominant, early-onset familial PD (Tofaris and Spillantini, 2007). At the protein level, αSyn is the major component of Lewy bodies and Lewy neurites, the insoluble aggregates that are neuropathological hallmarks of both familial and sporadic PD (Spillantini et al., 1997). Proposed functions of αSyn include lipid binding and regulation of membrane composition (Jo et al., 2000; Kubo et al., 2005; Gitler and Shorter, 2007); in particular, αSyn may interact with specific fatty acids and/or modulate fatty acid metabolism in the brain (Sharon et al., 2003; Assayag et al., 2007; Barcelo-Coblijn et al., 2007; Golovko et al., 2008). αSyn has also been implicated in the regulation of neurotransmitter release and/or of the reserve pool of synaptic vesicles (Abeliovich et al., 2000; Murphy et al., 2000; Cabin et al., 2002; Larsen et al., 2006), perhaps through interactions with specific lipids or fatty acids.

The brain is known to undergo dramatic changes even within the scope of healthy aging, and alterations in membrane composition, in particular, have been under investigation for more than 20 years (for review, see Giusto et al., 2002). Many studies have examined postmortem human brain tissue (Soderberg et al., 1990, 1991), although some degradation of the lipids is almost certain in these analyses due to the relatively protracted postmortem interval. Studies using rodent brain (Calderini et al., 1983) can minimize such deterioration. However, until recently, such studies often relied on chromatographic techniques which allowed only for investigation of broadly defined questions, such as the fatty acid composition of total phospholipid pools or the total phospholipid mass in brain samples. Due to the limitations inherent in these techniques, it has hitherto been impossible to search for finely detailed changes within individual lipid species across the whole breadth of phospholipid species found in the brain.

Recent developments in the field of lipid mass spectrometry (MS) have enabled a much more detailed analysis of lipid composition in biological samples than was previously possible (Forrester et al., 2004; Roberts et al., 2008). This new lipidomics approach has been applied to cultured cells (Rouzer et al., 2006; Ivanova et al., 2007; Cheng et al., 2008), animal models (Han et al., 2000; Yetukuri et al., 2007; Zeng et al., 2009), and human tissues (Han et al., 2002; Adibhatla et al., 2006; Sanchez-Mejia et al., 2008). Mass spectrometry, in particular electrospray ionization mass spectrometry (ESI-MS), provides high resolving power, superb sensitivity, reproducibility, and the ability to detect individual lipid species within a complex mixture such as a biological sample (Ivanova et al., 2007). MS has also been utilized for tissue imaging, providing a detailed and sensitive description of the spatial distribution of individual lipid species (Hankin et al., 2007). Several studies have recently analyzed lipid composition of brain extracts, including extracts from human brain. However, these studies focused on highly specific subcellular fractions (Kiebish et al., 2008) or pre-determined lipid species (Han et al., 2002; Isaac et al., 2003).

In order to determine whether and how αSyn gene dosage interacts with aging-related effects on phospholipid composition in the brain, we analyzed brain phospholipid profiles from mice at different ages and with various levels of αSyn expression. We used young (3-4 mo) or aged (12-14 mo) αSyn wild-type (WT), heterozygous (HET), or homozygous knockout (KO) mice in the BL6/129 genetic background, as well as young or aged non-transgenic (i.e., wild-type) or αSyn transgenic (TG) mice expressing wild-type human αSyn in the BL6/DBA2 genetic background. The experimental design of our study afforded several important advantages: (1) all of the animals used in the study were raised under identical conditions, as it is well known that differences in nutritional intake can affect brain phospholipid composition (Barzanti et al., 1994); (2) all samples were handled optimally to avoid lipid degradation due to extended postmortem intervals; and (3) phospholipid composition was compared in an unbiased manner across brain regions, ages, and genders in two different strains of mice with varying αSyn gene dosage. In contrast to previous studies of brain lipid composition, we report an unbiased phospholipid MS analysis (lipidomics) that provide details about the relative levels of many individual phospholipid species (e.g., 22:6 LPA or 36:1 PC). Phospholipid species are denoted as xx:y based on the total number of carbons (xx) and total double bonds (y) in their two acyl side chains.

Surprisingly, the most dramatic differences were observed in wild-type animals as a function of age; differences in αSyn gene dosage, in particular the overexpression of wild-type human αSyn, had striking effects on some phospholipid classes in certain subsets of animals, but these differences were much less profound than those observed as a function of age and gender-specific aging.

METHODS

Mice

αSyn knockout (KO), heterozygous (HET), or wild-type (WT) mice in a BL6/129 background (Abeliovich et al., 2000) or non-transgenic or αSyn transgenic (TG) mice expressing wild-type human αSyn under the PDGF-β promoter in a BL6/DBA2 background (Masliah et al., 2000) were housed according to Harvard University protocols and fed standard chow. All mice were bred and raised in-house, under identical conditions. Four cohorts of mice were analyzed: young αSyn KO and littermate controls in the BL6/129 genetic background at 3-4 months old (mo); young αSyn TG and littermate controls in the BL6/DBA2 genetic background at 3-4 mo; aged αSyn KO and littermate controls at 12-14 mo; and aged αSyn TG and littermate controls at 12-14 mo. A total of 96 mice were used in this study.

Sample preparation and glycerophospholipid analysis by mass spectrometry

Mice were rapidly asphyxiated with CO2 followed by cervical dislocation and decapitation, as per Harvard University animal use guidelines. Cortex, hippocampus, striatum, and cerebellum were bilaterally dissected on ice and flash-frozen in liquid nitrogen. Global lipid extracts were prepared via a modified Bligh/Dyer extraction procedure as described previously (Ivanova et al., 2007). Mass spectral (MS) analysis was performed on a Finnigan TSQ Quantum triple quadrupole mass spectrometer (ThermoFinnigan, San Jose, CA) equipped with a Harvard Apparatus syringe pump and electrospray source. Data were collected with the Xcalibur software package (ThermoFinnigan) and analyzed as previously described (Rouzer et al., 2006; Ivanova et al., 2007). A detailed description is found in the Supplementary Methods.

Statistical analyses

Global glycerophospholipid analyses used the intensities of the internal standards (see above) in each ionization mode for normalization. Then, peak heights relative to the standards were calculated for each species to account for variability in MS signal across samples. Preliminary analyses utilized principal components analysis (PCA) to visualize clustering of broad multivariate lipid patterns of variation which recur across samples and conditions. For a detailed description of this statistical analysis, see Supplementary Methods. Where appropriate, data were analyzed by ANOVA with Bonferroni corrections for multiple comparisons as appropriate. For all bar graphs, error bars represent standard error of the mean.

Two different types of statistical analysis were conducted to aid in the interpretation of the lipidomics data on multiple levels. Ratiomics analysis (Rouzer et al., 2006) was used to examine the share of the signal of each species within the major classes of glycerophospholipids. These results are represented by the relative (percent) contribution of each phospholipid species (e.g., 34:1 PC or 22:6 LPA) to the total signal from its class (in this example, PC or LPA, respectively). The analyses for each lipid class are conducted independently, such that the total signal from each phospholipid or lysolipid class sum separately to 100%. Ratiomic analysis provides a species-level view of differences within each lipid class; such differences are often masked in conventional analyses across lipid classes because they account for only a small part of the overall variance in the data set. However, systematic differences within a particular lipid class could indicate physiologically important changes in lipid uptake and/or metabolism.

Separately, the data were also analyzed as summed intensities across each phospholipid class. As with ratiomic analysis, each lipid class (e.g., PA, LPA) was treated as a separate and independent group; the sum of the normalized signal intensities of all species within each lipid class was compared across conditions to identify general trends in phospholipid composition profiles.

RESULTS

Mouse strains must be evaluated separately for lipid profile differences

In order to identify general trends in the data in an unbiased way, we used principal components analysis (PCA) of ratiomic data from diacyl glycerophospholipids to reveal differences between genotypes, brain regions, and/or ages on a multivariate basis across all lipid species. Analysis was restricted to diacyl phospholipids because ratiomic data within lysolipid classes contained far less statistical information, with as few as five species identified for some headgroups. The first two principal components accounted for 50.5% of the variance within the BL6/DBA2 data, and the corresponding principal components accounted for 66.3% of the variance in the BL6/129 strain. Thus, in both mouse strains and across all of the lipidomic data collected, variance was not random; just two modes of variability of co-varying lipids accounted for more than half of the total variance in each strain.

These initial PCA analyses showed striking and unexpected differences between the patterns of aging in the mouse strains, in spite of their similar genetic background and the fact that the animals were simultaneously raised in a single facility under identical conditions of housing and diet (Figure S1). However, these differences were rarely statistically significant when further investigated by summed intensities. When analysis was restricted to the WT BL6/129 and non-TG BL6/DBA2 mice alone, the leading principal components showed only slight differences between strains (Figure S1c; p=0.029 for the distance between the centroids of the clusters for each strain). Due to the observed differences in patterns of variance between the two mouse strains, all further analyses were conducted for each strain separately.

Non-TG BL6/DBA2 mice show age-dependent changes in phospholipid profile

We first investigated whether lipid profiles in WT mice within each strain varied by age or gender. There were no obvious differences between genders within the principal components analyses of either strain. However, analysis by summed intensities revealed more age-dependent variability in males than in females from both strains.

In non-TG (i.e., WT) animals in the BL6/DBA2 background (Figure 1), levels of PA and LPA were 1.6- and 2.4-fold higher, respectively, in aged (12-14 mo) males than in young (3-4 mo) males. Levels of PE, LPE, PS, LPS, PI, and PC were all significantly reduced in aged males to ratios of 0.6, 0.5, 0.4, 0.5, 0.4, and 0.8, respectively, compared with young males. Of the phospholipid / lysolipid pairs, only PG and LPG showed aged-dependent variation in opposite directions in males: PG decreased to 0.5 in aged males relative to young males, whereas LPG increased by 3.3-fold. Phospholipid composition in WT females also tended to decrease, albeit much less dramatically, with age. Levels of PG, LPG, PI, LPI, PS, LPS, PC, and LPE all decreased significantly in aged females compared with young females, to ratios of 0.7, 0.5, 0.7, 0.4, 0.8, 0.7, 0.9, and 0.8, respectively (Figure 1).

Figure 1. Non-TG BL6/DBA2 mice show marked changes in broad phospholipid classes with age.

Figure 1

Open in a new tab

Sums of intensities by class show broad differences with age, and many of these age-dependent changes are driven by differences in data from males only. Error bars represent SEM. Samples from all brain regions were pooled, n=20-24 per group. *, p<0.05 compared with young mice of same gender.

A similar analysis for WT BL6/129 mice revealed age-dependent changes on a similar scale to those seen in the BL6/DBA2 mice (Figure S2), but the trends were not identical. For example, PC, LPE, and LPS all decreased with age in BL6/DBA2 males but increased with age in BL6/129 males. The gender differences in aging seemed confined to the BL6/DBA2 strain. Thus, such age-related changes are specific to certain genetic backgrounds. Importantly, these age-related differences were identified in analyses that included only WT animals from each strain, raised under identical conditions of housing and diet, and thus reveal the effect of aging per se.

To further investigate the age-related changes in the lipid profiles of the non-TG BL6/DBA2 mice, we analyzed the ratiomic species-level differences within each lipid class and examined the aging-related data both for males and females pooled and for each gender separately (Figure 2). Even with pooling, some statistically significant changes were observed. For example, 18:1LPC and 20:1LPC are increased by 13% and 80%, respectively, while 18:0LPC and 22:6LPC are decreased by 45% and 36%, respectively, in data from both genders pooled. When each gender is examined separately, somewhat different patterns emerge. In males 18:1LPC and 20:1LPC are increased by 29% and 55%, respectively, while in females these species are increased by 25% and 148%, respectively. Thus, the increase in 18:1LPC in the pooled data is driven by increases in males and females equally, whereas the increase in 20:1LPC in the pooled data is driven strongly by data from females. Conversely, the decrease in 22:6LPC in the pooled data is driven by a greater decrease in levels from males (41% decrease) than in levels from females (18% decrease). We also noted a decrease in PUFA-containing PG species with age and a complementary increase in LPG species (Figure S3b). Representative graphs are shown in Figure 2, and the full data sets for PC and PG are shown in Figure S3.

Figure 2. Non-TG BL6/DBA2 mice show age-dependent lipid profile changes within phospholipid classes.

Figure 2

Open in a new tab

Ratiomic analysis of changes in phospholipid composition within each class revealed significant age-dependent differences in certain phospholipid species in non-TG mice. Data from all brain regions were pooled. Error bars represent SEM; n=16-24 per gender group. Each diacyl phospholipid or lysophospholipid class independently sums to 100%. *, Bonferroni adjusted p<0.05; #, p<10-5. Representative species of PC and LPC are shown for males and females pooled (left), males alone (middle), and females alone (right). See Figure S3 for full data sets.

Limited variability in phospholipid composition among brain regions in wild-type mice

PCA analysis (Figure S1) revealed differences in variance among the different brain regions in WT mice of both strains, and greater variability in data from aged animals than in data from young animals. In order to investigate the extent of these differences, we analyzed the summed intensities of the major phospholipid classes in each mouse strain. Generally, the aged mice in both strains showed little change between regions on the level of phospholipid classes (Figure S4). One exception was the LPC class, which had elevated levels in the hippocampus (10.176 ± 0.597) and striatum (11.572 ± 1.013) relative to the cortex (6.650 ± 0.723) and cerebellum (7.535 ± 1.039) in a one-way ANOVA (p<0.01).

Ratiomic analyses within each phospholipid and associated lysolipid class further elucidated the small differences across brain regions that were seen more broadly in the summed intensities in Figure S4, and also revealed more narrowly defined differences that were masked within the region-pooled data. In general, the cerebellum was the most distinct of the brain regions in both the BL6/129 (Figure 3a) and BL6/DBA2 (Figure 3b) strains. Some of the larger region-specific phospholipid differences were found among subspecies of PI (Figure 3), PE, and PG (Figure S5). In addition, sizeable ratiomic differences were seen for certain lysolipids (e.g., 18:1LPE, 18:0LPI). For instance, LPI regional differences were considerably larger than those for PI, and saturated PI and LPI species increased relative to other PI/LPI species. Some long chain polyunsaturated fatty acid (PUFA)-containing PG species showed substantial regional differences relative to other PGs (Figure S5).

Figure 3. Brain regions differ in composition within phospholipid classes in WT mice.

Figure 3

Open in a new tab

Ratiomic analysis within each phospholipid class shows region-dependent differences in individual phospholipid species. Data from young and aged mice of both genders were pooled. Each diacyl phospholipid or lysophospholipid class independently sums to 100%. Representative graphs are shown and error bars indicate SEM; see Figure S4 for additional data. n=15-20 for each brain region. *, Bonferroni adjusted p<0.05; #, p<10-5. (a) WT mice from the BL6/129 strain show region-dependent differences in PA/LPA (top) and PI/LPI (bottom) composition. (b) In non-TG mice from the BL6/DBA2 strain, none of the region-dependent differences in PA/LPA reach statistical significance when corrected for multiple comparisons (top), but many significant differences in species of PI/LPI are apparent (bottom).

αSyn genotype-dependent differences in phospholipid profile emerge with age

We next investigated the effects of αSyn gene dosage on phospholipid composition in the brain. Pooling all available data from each strain separately, PCA revealed no broad genotype-related trends in the data (not shown). However, as noted above, this analysis can overlook more narrowly-defined trends in specific subsets of data, which might be masked by the pooling required for PCA. Therefore, we analyzed summed intensities across the lipid classes to identify genotype-dependent changes in each phospholipid class separately. In the BL6/DBA2 strain, young (3-4 mo) non-TG and TG mice showed remarkably similar phospholipid profiles (Figure 4a). However, the profiles for aged (12-14 mo) mice differed consistently from those of the young mice: for both genotypes, total lipid signal in the diacyl phospholipid classes tended to decrease in the aged mice compared with the young mice. For example, comparison of data from aged non-TG mice with young non-TG mice showed that total signal decreased by 43% in PG, by 45% in PI, and by 41% in PS. Only aged mice showed significant αSyn-dependent differences. Similar comparisons among the data from TG mice showed that aging led to a decrease of 23% in PA, 28% in PG, and 28% in PI.

Figure 4. Phospholipid differences between non-TG and αSyn TG mice become apparent with age.

Figure 4

Open in a new tab

Data from all brain regions were pooled, and summed intensities were calculated for the broad phospholipid classes. (a) Brain phospholipid composition in young (3-4 mo) TG and non-TG mice, compared with that in aged (12-14 mo) mice. *, p < 0.05 compared with non-TG. (b) Within the aged (12-14 mo) group, males and females were compared separately. *, p < 0.05 compared with young mice of same gender. For all panels, error bars indicate SEM; n=16-24 per group.

Further inspection of the data from the BL6/DBA2 strain revealed an interaction of age and genotype which was dominated by males. When we analyzed genotype-dependent differences in aged animals by gender, it became clear that the differences observed in Figure 4a were driven by differences among the males (Figure 4b). Thus, none of the phospholipid classes showed statistically significant differences by αSyn genotype in the aged females, whereas PA was reduced by 50% and PE was increased by 60% in aged TG males compared with aged non-TG males. In addition, trends were observed in the aged males in PG (increased by 52%, p=0.053) and PS (increased by 54% p=0.066). Similar analyses in the BL6/129 strain (i.e., αSyn WT, HET, and KO) showed different trends from those seen in the BL6/DBA2 strain, though in this strain the age-dependent effects of genotype were not limited to males (Figure S6).

Ratiomic changes in lipid species within each phospholipid class were also examined for evidence of genotype-specific differences. Because the aged males showed the greatest variability in the summed intensities across the broad phospholipid classes, in particular in the BL6/DBA2 strain (above), we analyzed ratiomic changes within each phospholipid class and associated lysolipid class only for the aged males of each strain. Figure 5 shows examples of such analyses for PA; the complete data set is represented in Figure S7. BL6/DBA2 males (Figure 5b) showed less variability overall than BL6/129 males (Figure 5a), though in both cases the αSyn genotype-dependent differences appeared sporadic and did not appear to directly imply changes to any single lipid metabolic pathway. It is notable that we saw striking decreases in 36:0 PA and 18:0LPA in the TG males compared with the non-TG males in the BL6/DBA2 strain; a genotype-dependent increase in 36:0PA and 18:0LPA was seen in the BL6/129 males, but the increase was not gene-dose dependent because the HET males showed greater increases than either the WT or KO males. For a few individual phospholipid species, the genotype-dependent differences seemed to truly depend on gene dosage across the two mouse strains. For example, the relative levels of 40:6PA decrease with decreasing αSyn gene dosage in the BL6/129 strain (Figure 5a), and correspondingly increase in the TG mice compared with non-TG in the BL6/DBA2 strain (Figure 5b).

Figure 5. Genotype-dependent differences within phospholipid class are limited.

Figure 5

Open in a new tab

Ratiomic analysis within phospholipid classes revealed few genotype-dependent changes. Data from aged males from each strain, with all regions pooled, were used because these showed the greatest differences in previous analyses. Each diacyl phospholipid or lysophospholipid class independently sums to 100%. Representative graphs are shown and error bars indicate SEM; see Figure S7 for additional data. n=12-24 per group. *, Bonferroni adjusted p<0.05 (a) Although statistically significant differences can be seen between WT, HET, and KO mice in the BL6/129 strain for PA species, these are limited to the phospholipid species shown (see Figure S7 for complete data set). (b) Few differences can be seen between non-TG and TG mice in the BL6/DBA2 strain in PA species (see Figure S7 for complete data set).

Because we observed a relatively large number of lysolipid species in our samples, we were able to investigate changes to these lipids in addition to their diacyl phospholipid counterparts. Many of the αSyn genotype-dependent changes that we identified were found in lysolipids. For example, in the BL6/129 strain we saw consistent and significant decreases of approximately 25% in LPA, LPE, LPG, and LPC in KO mice compared with WT mice when data from all brain regions and both genders were pooled and analyzed by summed intensities (Table 1). Summed intensities signals in all of the lysolipid classes approximately doubled in aged mice, compared with young mice in this strain (Table 1). Similar analyses revealed less profound changes in the BL6/DBA2 strain (Table S1).

Table 1. Brain lysophospholipid composition of αSyn WT, HET, and KO mice varies with age and with genotype.

P-values and ratios (knockout/wild-type and aged/young) are presented for both age-related and genotype-related differences. Data from both genders and all brain regions have been pooled. A comparable table for BL6/DBA2 mice can be found as Supplemental Table 1.

p (Genotype) 3-Way Comparison Ratio KO / WT Ratio HET / WT P (Age) 2-Way Comparison Ratio Aged / Young
LPA 0.0387 0.73 0.93 <10-5 2.19
LPC 0.0086 0.73 0.90 <10-5 1.81
LPE NS 0.78 0.97 <10-5 2.10
LPG 0.0092 0.80 0.89 <10-5 1.48
LPI NS 1.06 1.03 0.0106 0.81
LPS NS 0.81 0.96 <10-5 1.81
Open in a new tab

To further investigate the age-genotype interaction in the data from the BL6/129 strain, we compared the summed intensities for each age and gender, in each lysolipid class, to the summed intensities value for young WT mice in the respective class. Thus, values for aged WT mice, young KO mice, and aged KO mice were each compared with young WT mice. The resulting ratios are presented in Table 2. The interactions of age and genotype for LPE (p<0.05), LPC (p<0.05), and LPA (p<0.01) are all statistically significant. A similar comparison in the BL6/DBA strain revealed differences of smaller magnitude (Table S2).

Table 2. Brain lysophospholipid composition of αSyn WT, HET, and KO males shows age-enotype interaction.

Ratios of changes, relative to the young wild-type mice, are presented for all other combinations of age and genotype. Data from all brain regions have been pooled, for male mice only. A comparable table for BL6/DBA2 mice can be found as Supplemental Table 2.

Young WT Aged WT Young HET Aged HET Young KO Aged KO P (Interaction)
LPA 1.00 3.51 0.96 1.68 1.09 2.45 0.0060
LPC 1.00 2.17 0.82 1.09 0.86 1.60 0.0224
LPE 1.00 3.31 0.99 1.81 1.03 2.66 0.0387
LPG 1.00 1.97 0.90 1.38 0.99 1.49 NS
LPI 1.00 0.67 0.97 0.82 0.87 0.75 NS
LPS 1.00 2.74 0.91 1.59 0.99 2.25 NS
Open in a new tab

DISCUSSION

Utilizing state-of-the-art lipid mass spectrometry, we were able to establish phospholipid profiles of mouse brain samples and to analyze broad trends across lipid classes by summed intensities, as well as narrowly defined changes in individual phospholipid or lysolipid species by ratiomic composition. We analyzed cortex, hippocampus, striatum, and cerebellum from young (3-4 mo) and aged (12-14 mo) male and female mice from two strains with different αSyn gene dosage: αSyn WT, HET, and KO mice in the BL6/129 genetic background, and αSyn non-TG or TG mice in the BL6/DBA2 genetic background. To our knowledge, this study represents the first such detailed and systematic analyses of phospholipid composition in mouse brain across ages, genders, and brain regions.

Given the analytical power of the lipidomic methods we employed, and given the known interactions between αSyn and phospholipids, we designed this study to search for αSyn genotype-dependent differences in aged mice. We were surprised to find broad differences in lipid profiles with aging in WT mice of the two genetic strains, as revealed by principal components analysis (Figure S1), in spite of their shared and identical housing and feeding conditions. This made pooling of data across mouse strains inadvisable, even for WT mice of similar ages and genders. Therefore, our analyses were conducted for each mouse strain separately. Although very similar across young mice of different genders and genotypes, the profiles clearly diverged with age. Thus, aging itself is a variable in determining the specific phospholipid composition of the brain.

Our initial analyses focused on identifying global trends in WT mice. Summed MS intensities across the phospholipid classes in WT mice from both strains identified marked differences between young and aged mice, which were driven primarily by differences in the males in the BL6/DBA2 strain (Figures 1 and S2). The breadth and scope of changes in summed intensities over phospholipid classes were striking. Furthermore, our ability to detect a relatively large number of lysolipid species enabled us to analyze trends in these lipids, which in some preparations are difficult to detect and thus generally less well characterized than their diacyl counterparts. Further analyses of these trends on a ratiomic level within each phospholipid and associated lysolipid class revealed large age-dependent changes in certain phospholipid and lysolipid species (Figures 2 and S3). In contrast, differences across brain regions were limited, with no changes that might implicate specific known pathways of lipid metabolism (Figures 3, S4, and S5).

It is important to note that these age-related differences were identified in WT mice housed under identical conditions and diets and, therefore, reflect the effects of aging per se. Although beyond the scope of our current research, the trends that we have reported could be expanded to explore age-related phenomena caused by changes in the lipid environment of membrane proteins or the availability of lipid substrates for phospholipases or other enzymes.

Because αSyn has been shown by several labs, including ours, to interact with lipid membranes and is hypothesized to affect phospholipid composition and/or fatty acid metabolism (Sharon et al., 2003; Golovko et al., 2005; Golovko et al., 2006; Assayag et al., 2007; Barcelo-Coblijn et al., 2007; Golovko et al., 2007; Golovko et al., 2008), one of our original goals in undertaking this study was to determine whether αSyn gene dosage affects specific brain phospholipid profiles. Thus, we investigated whether αSyn WT, HET, and KO mice in the BL6/129 strain, or αSyn non-TG and TG mice in the BL6/DBA2 strain, differed in their phospholipid profiles. Young mice in both strains showed little αSyn genotype-dependent phospholipid variability. The most striking effect of αSyn overexpression in the BL6/DBA2 mice seemed to be a dampening of the age-dependent variability in phospholipid levels among non-TG males, such that aged TG males were more similar to young TG males than were aged non-TG males to their young counterparts (Figure 4b). Marked αSyn-dependent decreases were seen in the lysophospholipids in the BL6/129 αSyn WT, HET, and KO mice, where we also detected a strong interaction of age and αSyn gene dosage (Tables 1 and 2).

Previously published results of lipid composition have focused on classes resolvable by thin layer chromatography (TLC), and lack detailed inventories based on the species-level profiling which are included in this study. Recent work with male αSyn KO mice at various ages, in a genetic background different from those tested here, highlighted only differences in PG and cardiolipin by TLC with the knockout of αSyn (Ellis et al., 2005; Barcelo-Coblijn et al., 2007). A comparison of the PG levels reported in the WT mice of those two studies with the data presented here shows that differences between age groups and/or strain-dependent changes in aging are likely to be at least as large as any differences between αSyn genotypes. While aging effects on some lipid classes (PA, LPA, PC, LPG, PS) were similar between BL6/129 and BL6/DBA2, others were not (Figures 1, 4, S2, and S6). Moreover, species-level changes within the PG class with aging appear at least as strong as αSyn genotype-dependent effects, as demonstrated by the observed relative decreases of PUFA PG species and relative increases of saturated PG species (Figure S3b).

Within each lipid class, ratiomic analysis revealed certain αSyn genotype-dependent differences in aged mice of both mouse strains. Oddly, diacyl phospholipid profiles of the HET mice in the BL6/129 strain were most consistently different from both the WT and KO, which were often similar to each other (Figures 5a and S7a). This result was surprising because we predicted that complete loss of αSyn in the KO mice would lead to the greatest changes in brain phospholipid composition, as was seen for lysolipids (Table 1). It is possible that compensatory mechanisms are active in the KO mice to return phospholipid profiles to near normal, whereas such mechanisms are not activated in the HET mice due to the incomplete loss of αSyn (Figure S8). An obvious candidate as a mediator of such compensatory mechanisms could be β-synuclein (βSyn), a closely related synuclein family member with similar tissue and subcellular distribution to αSyn and whose physiological function, like that of αSyn, is not yet known. In order to test this hypothesis, we probed brain lysates of mice from each genotype for both αSyn and βSyn (Figure S8). While αSyn protein levels mirrored the expected levels based on genotype, there were no differences in βSyn protein levels between the mice.

Overall, the differences detected among WT mice of different strains, genders, and ages were more profound than the limited differences among mice based on αSyn gene dosage, except for certain diacyl phospholipid species in the aged BL6/DBA2 mice and the lysophospholipid species distribution in the aged BL6/129 mice. We conclude that phospholipid profiles change substantially in the brain with age, that these changes are generally more pronounced in males than in females, and that the effects of αSyn gene dosage are relatively small and limited to a few lipid species when compared with the effects of age and gender alone.

Although αSyn gene dosage did not markedly alter brain phospholipid composition, the age- and gender-dependent lipid changes observed in this study could have substantial effects on the function of αSyn. Because αSyn binding to phospholipid membranes is thought to be important in its normal function and in its role in PD pathogenesis, it is possible that potentially pathological changes in αSyn function would manifest primarily in aged male mice as a result of their altered lipid profiles. Currently available mouse models of αSyn function and dysfunction have been disappointing due to their limited replication of the human PD phenotype (Litvan et al., 2007); however, these studies pooled data from male and female mice together. It is possible that a re-examination of data from aged male mice alone would yield insights about changes in αSyn function as a result of its interaction with the altered phospholipid environment that we delineate in the brains of male mice in particular. Additionally, PD is more prevalent in men than in women, and the symptoms may be more severe in men (Lyons et al., 1998; Van Den Eeden et al., 2003). If the effects of age and gender on brain phospholipid profiles in mice can be extrapolated to humans, then it is possible that gender-dependent lipidomic differences in aged men and resulting alterations in αSyn function could contribute to the observed gender differences in the PD patient population.

Finally, the function of many membrane-associated proteins can be dramatically affected by their lipid microenvironment. For example, Lyn kinase, Akt, protein kinase C, phospholipase D, and platelet-derived growth factor receptors each require specific phospholipids in their immediate environment, and changes in this lipid environment can reduce or alter protein function (Epand and Lester, 1990; Liscovitch et al., 1994; Franke et al., 1997; Liu et al., 2000; Young et al., 2003). Thus, the age-dependent changes in brain phospholipid profiles could affect the function of important membrane-associated proteins and thereby help to explain aging-related changes in the associated signaling pathways. In particular, such lipidomic changes could lead to alterations in post-translational modifications of PD-related proteins, including αSyn. Additional studies will be required to determine whether aging-related differences in the composition of phospholipid membranes can affect the cleavage, phosphorylation state, or other post-translational modifications of these proteins in cells and in model organisms.

Supplementary Material

Methods (supp)
Sfig 05b2
Sfig 06
Sfig 07a1
Sfig 07a2
Sfig 07b1
Sfig 07b2
Sfig 08
Sfig 01
Sfig 02
Sfig 03a
Sfig 03b
Sfig 04
Sfig 05a1
Sfig 05a2
Sfig 05b1

ACKNOWLEDGEMENT

The authors wish to thank Michelle Armstrong and Andrew Goodman for excellent technical assistance with the mass spectrometry work. This work was supported by a National Defense Science and Engineering Graduate Fellowship (IR), the Brigham and Women's Hospital Udall Center of Excellence for Parkinson's Disease Research, NIH/NINDS P50 NS038375 (DJS), R01 NS051318 (DJS), and the NIH Large Scale Collaborative Initiative LIPID MAPS, U54 GM069338 (HAB).

Glossary

Abbreviations

αSyn

α-synuclein

βSyn

β-synuclein

HET

heterozygous

KO

knock-out

MS

mass spectrometry

PA

phosphatidic acid

PC

phosphatidylcholine

PCA

principal component analysis

PD

Parkinson's disease

PE

phosphatidylethanolamine

PG

phosphatidylglycerol

PI

phosphatidylinositol

PS

phosphatidylserine

TG

transgenic

WT

wild-type

Footnotes

The authors have no conflicts of interest relevant to this work.

REFERENCES

  1. Abeliovich A, Schmitz Y, Farinas I, Choi-Lundberg D, Ho WH, Castillo PE, Shinsky N, Verdugo JM, Armanini M, Ryan A, Hynes M, Phillips H, Sulzer D, Rosenthal A. Mice lacking alpha-synuclein display functional deficits in the nigrostriatal dopamine system. Neuron. 2000;25:239–252. doi: 10.1016/s0896-6273(00)80886-7. [DOI] [PubMed] [Google Scholar]
  2. Adibhatla RM, Hatcher JF, Dempsey RJ. Lipids and lipidomics in brain injury and diseases. Aaps J. 2006;8:E314–321. doi: 10.1007/BF02854902. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Assayag K, Yakunin E, Loeb V, Selkoe DJ, Sharon R. Polyunsaturated fatty acids induce alpha-synuclein-related pathogenic changes in neuronal cells. Am J Pathol. 2007;171:2000–2011. doi: 10.2353/ajpath.2007.070373. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Barcelo-Coblijn G, Golovko MY, Weinhofer I, Berger J, Murphy EJ. Brain neutral lipids mass is increased in alpha-synuclein gene-ablated mice. J Neurochem. 2007;101:132–141. doi: 10.1111/j.1471-4159.2006.04348.x. [DOI] [PubMed] [Google Scholar]
  5. Barzanti V, Battino M, Baracca A, Cavazzoni M, Cocchi M, Noble R, Maranesi M, Turchetto E, Lenaz G. The effect of dietary lipid changes on the fatty acid composition and function of liver, heart and brain mitochondria in the rat at different ages. Br J Nutr. 1994;71:193–202. doi: 10.1079/bjn19940126. [DOI] [PubMed] [Google Scholar]
  6. Cabin DE, Shimazu K, Murphy D, Cole NB, Gottschalk W, McIlwain KL, Orrison B, Chen A, Ellis CE, Paylor R, Lu B, Nussbaum RL. Synaptic vesicle depletion correlates with attenuated synaptic responses to prolonged repetitive stimulation in mice lacking alpha-synuclein. J Neurosci. 2002;22:8797–8807. doi: 10.1523/JNEUROSCI.22-20-08797.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Calderini G, Bonetti AC, Battistella A, Crews FT, Toffano G. Biochemical changes of rat brain membranes with aging. Neurochem Res. 1983;8:483–492. doi: 10.1007/BF00965104. [DOI] [PubMed] [Google Scholar]
  8. Cheng H, Mancuso DJ, Jiang X, Guan S, Yang J, Yang K, Sun G, Gross RW, Han X. Shotgun lipidomics reveals the temporally dependent, highly diversified cardiolipin profile in the mammalian brain: temporally coordinated postnatal diversification of cardiolipin molecular species with neuronal remodeling. Biochemistry. 2008;47:5869–5880. doi: 10.1021/bi7023282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dauer W, Przedborski S. Parkinson's disease: mechanisms and models. Neuron. 2003;39:889–909. doi: 10.1016/s0896-6273(03)00568-3. [DOI] [PubMed] [Google Scholar]
  10. Ellis CE, Murphy EJ, Mitchell DC, Golovko MY, Scaglia F, Barcelo-Coblijn GC, Nussbaum RL. Mitochondrial lipid abnormality and electron transport chain impairment in mice lacking alpha-synuclein. Mol Cell Biol. 2005;25:10190–10201. doi: 10.1128/MCB.25.22.10190-10201.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Epand RM, Lester DS. The role of membrane biophysical properties in the regulation of protein kinase C activity. Trends Pharmacol Sci. 1990;11:317–320. doi: 10.1016/0165-6147(90)90234-y. [DOI] [PubMed] [Google Scholar]
  12. Forrester JS, Milne SB, Ivanova PT, Brown HA. Computational lipidomics: a multiplexed analysis of dynamic changes in membrane lipid composition during signal transduction. Mol Pharmacol. 2004;65:813–821. doi: 10.1124/mol.65.4.813. [DOI] [PubMed] [Google Scholar]
  13. Franke TF, Kaplan DR, Cantley LC, Toker A. Direct regulation of the Akt proto-oncogene product by phosphatidylinositol-3,4-bisphosphate. Science. 1997;275:665–668. doi: 10.1126/science.275.5300.665. [DOI] [PubMed] [Google Scholar]
  14. Gitler AD, Shorter J. Prime time for alpha-synuclein. J Neurosci. 2007;27:2433–2434. doi: 10.1523/JNEUROSCI.0094-07.2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Giusto NM, Salvador GA, Castagnet PI, Pasquare SJ, Ilincheta de Boschero MG. Age-associated changes in central nervous system glycerolipid composition and metabolism. Neurochem Res. 2002;27:1513–1523. doi: 10.1023/a:1021604623208. [DOI] [PubMed] [Google Scholar]
  16. Golovko MY, Rosenberger TA, Feddersen S, Faergeman NJ, Murphy EJ. Alpha-synuclein gene ablation increases docosahexaenoic acid incorporation and turnover in brain phospholipids. J Neurochem. 2007;101:201–211. doi: 10.1111/j.1471-4159.2006.04357.x. [DOI] [PubMed] [Google Scholar]
  17. Golovko MY, Faergeman NJ, Cole NB, Castagnet PI, Nussbaum RL, Murphy EJ. Alpha-synuclein gene deletion decreases brain palmitate uptake and alters the palmitate metabolism in the absence of alpha-synuclein palmitate binding. Biochemistry. 2005;44:8251–8259. doi: 10.1021/bi0502137. [DOI] [PubMed] [Google Scholar]
  18. Golovko MY, Barcelo-Coblijn G, Castagnet PI, Austin S, Combs CK, Murphy EJ. The role of alpha-synuclein in brain lipid metabolism: a downstream impact on brain inflammatory response. Mol Cell Biochem. 2008 doi: 10.1007/s11010-008-0008-y. [DOI] [PubMed] [Google Scholar]
  19. Golovko MY, Rosenberger TA, Faergeman NJ, Feddersen S, Cole NB, Pribill I, Berger J, Nussbaum RL, Murphy EJ. Acyl-CoA synthetase activity links wild-type but not mutant alpha-synuclein to brain arachidonate metabolism. Biochemistry. 2006;45:6956–6966. doi: 10.1021/bi0600289. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Han X, Abendschein DR, Kelley JG, Gross RW. Diabetes-induced changes in specific lipid molecular species in rat myocardium. Biochem J. 2000;352(Pt 1):79–89. [PMC free article] [PubMed] [Google Scholar]
  21. Han X, D MH, McKeel DW, Jr., Kelley J, Morris JC. Substantial sulfatide deficiency and ceramide elevation in very early Alzheimer's disease: potential role in disease pathogenesis. J Neurochem. 2002;82:809–818. doi: 10.1046/j.1471-4159.2002.00997.x. [DOI] [PubMed] [Google Scholar]
  22. Hankin JA, Barkley RM, Murphy RC. Sublimation as a method of matrix application for mass spectrometric imaging. J Am Soc Mass Spectrom. 2007;18:1646–1652. doi: 10.1016/j.jasms.2007.06.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Isaac G, Bylund D, Mansson JE, Markides KE, Bergquist J. Analysis of phosphatidylcholine and sphingomyelin molecular species from brain extracts using capillary liquid chromatography electrospray ionization mass spectrometry. J Neurosci Methods. 2003;128:111–119. doi: 10.1016/s0165-0270(03)00168-7. [DOI] [PubMed] [Google Scholar]
  24. Ivanova PT, Milne SB, Byrne MO, Xiang Y, Brown HA. Glycerophospholipid identification and quantitation by electrospray ionization mass spectrometry. Methods Enzymol. 2007;432:21–57. doi: 10.1016/S0076-6879(07)32002-8. [DOI] [PubMed] [Google Scholar]
  25. Jo E, McLaurin J, Yip CM, George-Hyslop P, Fraser PE. alpha-Synuclein membrane interactions and lipid specificity. J Biol Chem. 2000;275:34328–34334. doi: 10.1074/jbc.M004345200. [DOI] [PubMed] [Google Scholar]
  26. Kiebish MA, Han X, Cheng H, Lunceford A, Clarke CF, Moon H, Chuang JH, Seyfried TN. Lipidomic analysis and electron transport chain activities in C57BL/6J mouse brain mitochondria. J Neurochem. 2008;106:299–312. doi: 10.1111/j.1471-4159.2008.05383.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kubo S, Nemani VM, Chalkley RJ, Anthony MD, Hattori N, Mizuno Y, Edwards RH, Fortin DL. A combinatorial code for the interaction of alpha-synuclein with membranes. J Biol Chem. 2005;280:31664–31672. doi: 10.1074/jbc.M504894200. [DOI] [PubMed] [Google Scholar]
  28. Larsen KE, Schmitz Y, Troyer MD, Mosharov E, Dietrich P, Quazi AZ, Savalle M, Nemani V, Chaudhry FA, Edwards RH, Stefanis L, Sulzer D. Alpha-synuclein overexpression in PC12 and chromaffin cells impairs catecholamine release by interfering with a late step in exocytosis. J Neurosci. 2006;26:11915–11922. doi: 10.1523/JNEUROSCI.3821-06.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lavedan C. The synuclein family. Genome Res. 1998;8:871–880. doi: 10.1101/gr.8.9.871. [DOI] [PubMed] [Google Scholar]
  30. Liscovitch M, Chalifa V, Pertile P, Chen CS, Cantley LC. Novel function of phosphatidylinositol 4,5-bisphosphate as a cofactor for brain membrane phospholipase D. J Biol Chem. 1994;269:21403–21406. [PubMed] [Google Scholar]
  31. Litvan I, Chesselet MF, Gasser T, Di Monte DA, Parker D, Jr., Hagg T, Hardy J, Jenner P, Myers RH, Price D, Hallett M, Langston WJ, Lang AE, Halliday G, Rocca W, Duyckaerts C, Dickson DW, Ben-Shlomo Y, Goetz CG, Melamed E. The etiopathogenesis of Parkinson disease and suggestions for future research. Part II. J Neuropathol Exp Neurol. 2007;66:329–336. doi: 10.1097/nen.0b013e318053716a. [DOI] [PubMed] [Google Scholar]
  32. Liu P, Wang P, Michaely P, Zhu M, Anderson RG. Presence of oxidized cholesterol in caveolae uncouples active platelet-derived growth factor receptors from tyrosine kinase substrates. J Biol Chem. 2000;275:31648–31654. doi: 10.1074/jbc.M004599200. [DOI] [PubMed] [Google Scholar]
  33. Lyons KE, Hubble JP, Troster AI, Pahwa R, Koller WC. Gender differences in Parkinson's disease. Clin Neuropharmacol. 1998;21:118–121. [PubMed] [Google Scholar]
  34. Masliah E, Rockenstein E, Veinbergs I, Mallory M, Hashimoto M, Takeda A, Sagara Y, Sisk A, Mucke L. Dopaminergic loss and inclusion body formation in alpha-synuclein mice: implications for neurodegenerative disorders. Science. 2000;287:1265–1269. doi: 10.1126/science.287.5456.1265. [DOI] [PubMed] [Google Scholar]
  35. Murphy DD, Rueter SM, Trojanowski JQ, Lee VM. Synucleins are developmentally expressed, and alpha-synuclein regulates the size of the presynaptic vesicular pool in primary hippocampal neurons. J Neurosci. 2000;20:3214–3220. doi: 10.1523/JNEUROSCI.20-09-03214.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Roberts LD, McCombie G, Titman CM, Griffin JL. A matter of fat: an introduction to lipidomic profiling methods. J Chromatogr B Analyt Technol Biomed Life Sci. 2008;871:174–181. doi: 10.1016/j.jchromb.2008.04.002. [DOI] [PubMed] [Google Scholar]
  37. Rouzer CA, Ivanova PT, Byrne MO, Milne SB, Marnett LJ, Brown HA. Lipid profiling reveals arachidonate deficiency in RAW264.7 cells: Structural and functional implications. Biochemistry. 2006;45:14795–14808. doi: 10.1021/bi061723j. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Sanchez-Mejia RO, Newman JW, Toh S, Yu GQ, Zhou Y, Halabisky B, Cisse M, Scearce-Levie K, Cheng IH, Gan L, Palop JJ, Bonventre JV, Mucke L. Phospholipase A2 reduction ameliorates cognitive deficits in a mouse model of Alzheimer's disease. Nat Neurosci. 2008;11:1311–1318. doi: 10.1038/nn.2213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sharon R, Bar-Joseph I, Frosch MP, Walsh DM, Hamilton JA, Selkoe DJ. The formation of highly soluble oligomers of alpha-synuclein is regulated by fatty acids and enhanced in Parkinson's disease. Neuron. 2003;37:583–595. doi: 10.1016/s0896-6273(03)00024-2. [DOI] [PubMed] [Google Scholar]
  40. Soderberg M, Edlund C, Kristensson K, Dallner G. Lipid compositions of different regions of the human brain during aging. J Neurochem. 1990;54:415–423. doi: 10.1111/j.1471-4159.1990.tb01889.x. [DOI] [PubMed] [Google Scholar]
  41. Soderberg M, Edlund C, Kristensson K, Dallner G. Fatty acid composition of brain phospholipids in aging and in Alzheimer's disease. Lipids. 1991;26:421–425. doi: 10.1007/BF02536067. [DOI] [PubMed] [Google Scholar]
  42. Spillantini MG, Schmidt ML, Lee VM, Trojanowski JQ, Jakes R, Goedert M. Alpha-synuclein in Lewy bodies. Nature. 1997;388:839–840. doi: 10.1038/42166. [DOI] [PubMed] [Google Scholar]
  43. Tofaris GK, Spillantini MG. Alpha-synuclein dysfunction in Lewy body diseases. Mov Disord. 2005;20(Suppl 12):S37–44. doi: 10.1002/mds.20538. [DOI] [PubMed] [Google Scholar]
  44. Tofaris GK, Spillantini MG. Physiological and pathological properties of alpha-synuclein. Cell Mol Life Sci. 2007 doi: 10.1007/s00018-007-7217-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Van Den Eeden SK, Tanner CM, Bernstein AL, Fross RD, Leimpeter A, Bloch DA, Nelson LM. Incidence of Parkinson's disease: variation by age, gender, and race/ethnicity. Am J Epidemiol. 2003;157:1015–1022. doi: 10.1093/aje/kwg068. [DOI] [PubMed] [Google Scholar]
  46. Yetukuri L, Katajamaa M, Medina-Gomez G, Seppanen-Laakso T, Vidal-Puig A, Oresic M. Bioinformatics strategies for lipidomics analysis: characterization of obesity related hepatic steatosis. BMC Syst Biol. 2007;1:12. doi: 10.1186/1752-0509-1-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Young RM, Holowka D, Baird B. A lipid raft environment enhances Lyn kinase activity by protecting the active site tyrosine from dephosphorylation. J Biol Chem. 2003;278:20746–20752. doi: 10.1074/jbc.M211402200. [DOI] [PubMed] [Google Scholar]
  48. Zeng X, Zheng X, Xiang Y, Plumley H, Jessen JR, Zhong TP, Solnica-Krezel L, Brown HA. Phospholipase D1 is required for angiogenesis of intersegmental blood vessels in zebrafish. Dev Biol. 2009;328:363–376. doi: 10.1016/j.ydbio.2009.01.035. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Methods (supp)
NIHMS135516-supplement-supp_methods.tif (1.1MB, tif)
Sfig 05b2
NIHMS135516-supplement-Sfig_05b2.tif (2.5MB, tif)
Sfig 06
NIHMS135516-supplement-Sfig_06.tif (3MB, tif)
Sfig 07a1
NIHMS135516-supplement-Sfig_07a1.tif (2.4MB, tif)
Sfig 07a2
NIHMS135516-supplement-Sfig_07a2.tif (2.5MB, tif)
Sfig 07b1
NIHMS135516-supplement-Sfig_07b1.tif (2.2MB, tif)
Sfig 07b2
NIHMS135516-supplement-Sfig_07b2.tif (2.2MB, tif)
Sfig 08
NIHMS135516-supplement-Sfig_08.tif (1.1MB, tif)
Sfig 01
NIHMS135516-supplement-Sfig_01.tif (1.1MB, tif)
Sfig 02
NIHMS135516-supplement-Sfig_02.tif (1MB, tif)
Sfig 03a
NIHMS135516-supplement-Sfig_03a.tif (1.7MB, tif)
Sfig 03b
NIHMS135516-supplement-Sfig_03b.tif (1.7MB, tif)
Sfig 04
NIHMS135516-supplement-Sfig_04.tif (2.1MB, tif)
Sfig 05a1
NIHMS135516-supplement-Sfig_05a1.tif (2.4MB, tif)
Sfig 05a2
NIHMS135516-supplement-Sfig_05a2.tif (2.7MB, tif)
Sfig 05b1
NIHMS135516-supplement-Sfig_05b1.tif (2.6MB, tif)

RESOURCES