Impairment of DHA synthesis alters the expression of neuronal plasticity markers and the brain inflammatory status in mice

Abstract

Docosahexaenoic acid (DHA) is a ω-3 fatty acid typically obtained from the diet or endogenously synthesized through the action of elongases (ELOVLs) and desaturases. DHA is a key central nervous system constituent and the precursor of several molecules that regulate the resolution of inflammation. In the present study, we questioned whether the impaired synthesis of DHA affected neural plasticity and inflammatory status in the adult brain. To address this question, we investigated neural and inflammatory markers from mice deficient for ELOVL2 (Elovl2^-/- ), the key enzyme in DHA synthesis. From our findings, Elovl2^-/- mice showed an altered expression of markers involved in synaptic plasticity, learning, and memory formation such as Egr-1, Arc1, and BDNF specifically in the cerebral cortex, impacting behavioral functions only marginally. In parallel, we also found that DHA-deficient mice were characterized by an increased expression of pro-inflammatory molecules, namely TNF, IL-1β, iNOS, caspase-1 as well as the activation and morphologic changes of microglia in the absence of any brain injury or disease. Reintroducing DHA in the diet of Elovl2^-/- mice reversed such alterations in brain plasticity and inflammation. Hence, impairment of systemic DHA synthesis can modify the brain inflammatory and neural plasticity status, supporting the view that DHA is an essential fatty acid with an important role in keeping inflammation within its physiologic boundary and in shaping neuronal functions in the central nervous system.

Keywords: PUFA; anti-inflammatory molecules; brain plasticity; microglia; omega-3.

Figure 1

Biosynthesis of fatty acids in wild‐type (WT) and Elovl2^−/− (KO) mice brains. A‐B, Representative spectra of WT and KO mice fed standard chow diet (no DHA) and determined by extraction, derivatization, and LC‐MS/MS analysis of acyl‐etheno‐CoAs fatty acids. C, Composition of key n‐6 and n‐3 fatty acids in the brain of WT and KO. Results shown are means ± SEM from seven animals per group. *P < .05, **P < .01, and ***P < .001

Figure 2

Synaptic plasticity markers expression is impaired in the cerebral cortex of KO mice. mRNA expression of Egr‐1, Arc‐1, and BDNF in the total brain and their mRNA and protein in the cortex of WT and KO mice by RT‐PCR and western blotting. Data are shown as mean ± SEM of seven animals per group (WT and KO), each in duplicate. *P < .05 and ***P < .001 vs WT by unpaired Student's t test

Figure 3

Inflammatory markers expression is impaired in the cerebral cortex of KO mice. A‐B, mRNA expression of TNF, IL‐1β, iNOS, and Casp1 as well as protein expression of Casp1(p20) in total brain and cortex of WT and KO mice by qRT‐PCR and western blotting. Data are shown as mean ± SEM of seven (qRT‐PCR) or four (western blotting) animals per group (WT and KO), each in duplicate. *P < .05 vs WT by unpaired Student's t test

Figure 4

DHA deficiency alters microglia response without affecting astrogliosis. A, Immunofluorescence staining of microglia (Iba1) in the cerebral cortex of WT and KO mice. Double‐labeled and merged confocal images of Iba1 (red) plus NeuroTrace Nissl staining (green) (scale bars = 100 μm). Insets: Single‐labeled images of Iba1 staining (scale bars = 10 μm). B, Western blotting and its densitometry of Iba1. Immunoblot images are cropped from different parts of the same gel. C, Histogram of the number of Iba1 positive cells in the cerebral cortex of WT and KO mice. D, Immunofluorescence staining of astrocytes (GFAP) in the cerebral cortex of WT and KO mice. In the lower portion of the sections it is well visible also the corpus callosum (cc). Double‐labeled and merged confocal images of GFAP (red) plus NeuroTrace Nissl staining (green) (scale bars = 100 μm). E, Western blotting and its densitometry of GFAP. Immunoblot images are cropped from different parts of the same gel. F, Histogram of the number of GFAP positive cells in the cerebral cortex of WT and KO mice. G‐K, Histograms of the Sholl analysis of microglia in the cerebral cortex of WT and KO mice showing the perimeter (G) and the cross‐sectional area of the Iba1 + cells (H), the number of intersections (I), and the number of nodes/branch points at the different radii (J). All results are mean ± SEM or representative of six animals per group (WT and KO) *P < .05, **P < .01, and ***P < .001 vs WT by unpaired Student's t test

Figure 5

DHA deficiency does not alter behavioral functions. A, Individual exploration trajectories showing the change in the exploratory activity of WT and KO mice. Brown lines indicate locomotion (horizontal movements), blue lines indicate rears (vertical movements), and green lines indicate center area and corners. Exploratory activity was calculated as the number of visits to the center zone, percentage of time spent in the center zone, and percentage of distance in the center zone. Data are shown as mean ± SEM of 12 WT and 16 KO mice and analyzed by the unpaired Student's t test. B, Working memory consolidation with two objects (novel and familiar) evaluated as the percentage of time spent at exploring the objects. Data are shown as mean ± SEM of seven animals per group (WT and KO) and analyzed by the unpaired Student's t test. C, Spatial and learning memory evaluated as the ability of animals to seek out for escape holes (time spent in quadrants of target, opposite, negative or positive holes). Data are shown as mean ± SEM of 8 WT and 7 KO and analyzed by the unpaired Student's t test

Figure 6

Dietary supplementation with DHA restores neuroplasticity and microglia status in the cerebral cortex. A, mRNA expression of Arc‐1, IL‐1β, and Casp1 in the cortex by qRT‐PCR. Data are shown as mean ± SEM of four animals per group (WT, KO, and KO + DHA), each in duplicate. ***P < .001 vs WT and *P < .05 vs KO by one‐way ANOVA. B, Protein expression of Arc‐1 by western blotting and (C) level of IL‐1β by ELISA. Data are shown as mean ± SEM of four animals per group (WT, KO, and KO + DHA), each in duplicate. ***P < .001 vs WT and *P < .05 vs KO by one‐way ANOVA followed by Bonferroni post‐hoc test. D, Immunofluorescence staining of microglia in the cortex of WT, KO, and DHA‐supplemented KO (KO + DHA) mice. Double‐labeled and merged confocal images of Iba1 (red) plus NeuroTrace Nissl staining (green) in the cerebral cortex (scale bars = 100 μm). Inset: Single‐labeled images of Iba1 staining (scale bars = 10 μm). E, Histogram of the number of Iba1 positive cells in the cerebral cortex. Data are shown as mean ± SEM of six animals per group (WT, KO, and KO + DHA). ***P < .001 and **P < .01 by one‐way ANOVA followed by the Bonferroni post‐hoc test (F‐I). Histograms of the Sholl analysis of microglia in the cerebral cortex of WT, KO, and KO + DHA mice showing the perimeter (F) and the cross‐sectional area of the cells (G), the number of intersections (H), and the number of nodes/branch points of intersections at the different radii (I). Results are mean ± SEM or representative of six animals per group (WT, KO, and KO + DHA). **P < .01 and ***P < .001 by one‐way ANOVA followed by Bonferroni post‐hoc test. J, Schematic representation of DHA deficiency‐induced impact on brain plasticity and inflammation markers