Understanding lipid and EFA metabolism is crucial for maintaining both physical and mental health. Working with the brain tissues of a rodent model orally intoxicated with PPA could offer new insight into brain disturbances related to the absolute and relative abundance of SCFAs and PUFAs and their relationship to different lipid mediators, such as leukotrienes and prostaglandins. It is well known that PGE2, which is an anti-inflammatory prostanoid, is induced by SCFAs. In addition to classical eicosanoids such as PGE2, other lipid mediators that include lipoxins, resolvins, protectins and maresins are also generated from PUFAs.
It is well known that high concentrations of butyrate prevent NF-kB activation in LPS-stimulated RAW264.7 murine macrophage cells. In addition, other studies have reported that inhibition of NF-kB induced by butyrate in IFNγ-stimulated RAW264.7 is implicated in the butyrate-mediated reduction of TNFα, IL-6 and iNOS expression. All of these actions, which are mediated by butyrate, may contribute to the anti-inflammatory effects of butyrate in inflammatory bowel disease (IBD), both in experimental models and in humans, where it has been reported that butyrate suppresses the activation of HSP70 and NF-kB[15, 16]. As dual effect of butyrate, brain microglial cells were found to be greatly affected. Sodium butyrate concentration of 0.6 mM but not that of 0.2 mM potentiated the LPS-induced secretion of both IL-6 and nitric oxide. Most interestingly, butyrate treatment induced a proinflammatory response in transformed N9 microglial cell line and anti-inflammatory response in primary, brain-derived microglial cells. In contrast to the effect of butyrate on pro-inflammatory, anti-inflammatory cytokines show diverse pattern with dose-dependent dual effects. IL-10 as an important anti-inflammatory cytokine is increased by 0.25 mM concentration but significantly decreased by 1 mM of butyrate. On the other hand, most of pro-inflammatory cytokines are significantly stimulated by high concentration of PPA. A lower butyrate /propionate ratio was reported in the present study in the PPA-intoxicated group (Table 1 and Figure 1). This finding could be related and supported by our previous study in which pro-inflammatory markers (TNFα, INFγ, IL-6, and HSP70) are persistently induced in rat pups that are orally administered the same neurotoxic dose of PPA.
It is well known that acetic acid is utilized by hepatocytes and transformed into Acetyl-CoA, which can act as a precursor for lipogenesis and stimulate gluconeogenesis[19–21]. Propionic acid is mainly metabolized in the liver and has been shown to inhibit gluconeogenesis and increase glycolysis in rat hepatocytes. It has also been proposed that propionic acid may lower plasma cholesterol concentrations by inhibiting hepatic cholesterogenesis. The significantly lower acetic/propionic ratio reported in the present study (Table 1 and Figure 1) together with the associated depletion of glucose (a major energy source for brain cells) and cholesterol[22, 23] can identify PPA-induced neurotoxicity[6, 7]. Based on the fact that a deficiency in cholesterol and glucose greatly reduces glutamate uptake by glutamate transporters and leads to elevated glutamate, the reported lower acetic/PPA ratio could also be related to glutamate excitotoxicity, a mechanism that is strongly involved in the etiology of autism.
It is well documented that AA, EPA, and DHA are essential for the normal development and growth of the brain and of memory[24–27]. AA stimulates glucose uptake in cerebral cortical astrocytes; thus, it plays a critical role in the regulation of energy metabolism in the cerebral cortex. AA and DHA enhance acetylcholine (ACh) release, which guarantees long-term synaptic plasticity, thereby improving learning ability in experimental animals. Several clinical studies have shown that supplementing infants with AA, EPA, and DHA significantly improves their cognitive development and memory[30, 31].
Additionally, DHA promotes neuronal survival by facilitating membrane translocation/activation of Akt, which is a central player in signal transduction pathways and thus controls cellular functions such as proliferation and survival, metabolism, angiogenesis, and exocytosis. The in vivo reduction of DHA by dietary depletion increases the susceptibility of hippocampal neurons to apoptosis[32, 33].
Based on this information, the significant decrease in α LA, EPA, DHA, γ-LNA and AA reported in the present study can be easily related to PPA-neurotoxicity. Depletion of these PUFAs could indicate impairment of energy metabolism, loss of synaptic plasticity and increased neuronal susceptibility to apoptosis. These persistent biochemical autistic features were induced in rat pups that were orally administered the same neurotoxic dose of PPA. The observed variation in the absolute and relative concentrations of FFA could be also related to the pro-inflammatory effects of PPA. While AA is considered a pro-inflammatory lipid, γ-LNA, EPA and DHA have been viewed as anti-inflammatory molecules due to their capacity to reduce the production of pro-inflammatory cytokines via the NF-κB signaling pathway[34, 35]. Moreover, EPA is itself a substrate for cyclooxygenase and lipoxygenase, giving rise to mediators that often have biological effects opposite to those of AA.
Linoleic acid (LA) is an essential dietary fatty acid that is crucial to neonatal development. It is necessary for the growth and development of the brain and other body tissues that are dependent on AA, the central n-6 eicosanoid precursor synthesized from LA via elongation-desaturation. Desaturase and elongase enzymes insert double bonds and elongate carbon chains to create PUFA from essential fatty acid (EFA) precursors. It appears that the same enzymes catalyze the conversion of both omega-6 and omega-3 fatty acid precursors into PUFAs. Lower desaturase activity, which is estimated by the ratio of AA to LA, might also be a neurotoxic effect of PPA. A 56.85% reduction in AA/LA could confirm the toxic effect of PPA on the brains of treated rat pups. A link between fatty acid desaturation genes and attention-deficits was previously suggested. The non-significant change observed in αLA/EPA and αLA/DHA could be attributed to differences in Km values and affinity between the fatty acid desaturase enzyme and the substrates LA, ALA, EPA and DHA. This suggestion could be supported by the remarkable lower αLA/LA ratio, which are precursors for ω3 and ω 6 respectively.
Furthermore, PUFAs could be related to the neurochemical activity of the brain. These acids could control the expression, properties, and action of dopamine, serotonin (5-HT), and Ach[39, 40], especially during the perinatal period. During this period the growth and development of brain is at a maximum and ACh, in turn, regulates the release of PUFAs. Thus, ω-3 PUFAs and AA modulate neural function, including neurotransmission, membrane fluidity, ion channel, enzyme regulation and gene expression and prevent inflammation. They could therefore be of significant benefit in the prevention and management of autism.
Fatty acids may exert structural effects on membranes, either as free entities (i.e. FFA) or as part of other molecules such as phospholipids and triacylglycerides. Omega-3 fatty acids have a curved shape, allowing gaps between molecules when they are incorporated into cell membranes. These gaps increase the fluidity of the membrane, enabling cell-to-cell communication with the aid of ion channels. In contrast, omega-6 fatty acids are straighter and narrower and therefore reduce the fluidity of the membrane due to a remarkable decrease in the gaps between cells. Consequently, it is important that the ratio of the PUFA types remains balanced to preserve optimal functioning of the cell membrane.
The lower α-Linolenic acid and DHA reported in the present study (Table 2 and Figure 2) can also be related to impaired serotonin transmission, which was previously reported in rat pups either orally administered (the same toxic dose) or intraventricularly infused with PPA[6, 7]. α-Linolenic and DHA deficient rats exhibited significantly lower prefrontal cortex 5-HT content, greater 5-hydroxyindoleacetic acid 5-HIAA content, a significantly greater 5-HIAA/5-HT ratio, and a decrease in midbrain tryptophan hydroxylase-2 expression, which is a rate limiting enzyme in the 5-HT biosynthesis pathway. This finding suggests that a relationship exists between dysregulation in central 5-HT neurotransmission and omega-3 fatty acid deficiency.
In addition, ω -3 fatty acids may modulate biochemical and physiological responses that are implicated in neurodevelopment via their effects on nuclear transcription factors, especially those involved in immunologic dysfunction. In vitro neuronal cell studies have revealed DHA to be a potent ligand for peroxisome proliferator-activated receptor γ (PPAR γ), which results in a suppression of proinflammatory genes that encode various interleukins and tumor necrosis factor (TNF)-α.
The significantly lower ω 3, ω6 and ω 6/ω 3 ratios reported in the present study (Table 2 and Figure 2) are consistent with the previous work of Thomas et al., in which PPA infusion in rats decreased the total levels of monounsaturates, ω 6 fatty acids, and phosphatidylethanolamine plasmalogens and decreased the ω 6/ω 3 ratio, providing evidence of a relationship between changes in brain lipid profiles and autism-like behaviors in a rodent model.
Changes in behavior and brain integrity in adult rodents following transient selective serotonin (5-HT) reuptake inhibitors (SSRIs) such as fluoxetine may be related to disturbed AA neurotransmission and metabolism because AA is released from synaptic membrane phospholipids during neurotransmission involving 5-HT2A/2C receptors[47–50]. As a second messenger, AA can modify multiple aspects of brain function and structure. It is a precursor of a large number of bioactive eicosanoid products within the brain’s AA metabolic cascade[51, 52].
Anti-inflammatory ω-3 derived EPA and DHA block sodium (Na+) channels in a dose and time dependent manner in neonatal rat ventricular myocytes[53, 54]. This in turn limits the activity of the Na+/Ca+2 exchanger, which regulates intracellular Ca+2 influx. In vivo cell culture studies demonstrate that both EPA and DHA increase cellular inactivation in the Cornu Ammonis area (CA1) region of the hippocampus, thus reducing Ca+2 influx-associated cellular activity. Thus, PUFAs are likely to reduce intracellular Ca+2 influx-associated excitotoxicity.
In the present study, the impaired relative and absolute fatty acid levels could be related to mitochondrial dysfunction as an etiological mechanism in autism. PPA is thought to affect mitochondrial fatty acid metabolism by binding to propionyl Coenzyme A and by sequestering carnitine[56, 57]. This in turn induces potential metabolic disturbance by affecting mitochondrial β-oxidation and bioenergetics. On the other hand, impaired fatty acid oxidation activates uncoupling protein-2 (UCP-2), which results in heat generation that does not contribute to ATP production[58, 59].
The receiver operating characteristic (ROC) analysis (Tables 3 and4) showed that most of the measured relative values for fatty acids can be used as biomarkers for PPA neurotoxicity, recording satisfactory specificity, sensitivity and AUC. While elevated propionic acid is significantly correlated with levels of acetic, butyric, stearidonic and linoleic acid, it is inversely associated with the rest of the absolute and relative values of PUFAs (Table 5). Omega 6 fatty acids, represented by γ-LA and AA, were more significantly related to PPA-neurotoxicity than ω-3 fatty acids (ALA, EPA, and DHA). γ-LA, AA, DHA and EPA all play key roles in brain function, especially via the synthesis of eicosanoids that have anti-inflammatory, anti-thrombotic, and vasodilatory properties. Therefore, the inverse association between PPA level and the levels of these acids confirm that the neurotoxic dose of this short chain fatty acid used in this study (250 mg/kg body weight/day for three days) was effective in inducing an alteration in the brain fatty acid profile. These results provide evidence for a relationship between changes in brain lipid profiles and the occurrence of ASD-like biochemical alterations in a rodent autism model. This can be easily supported by the positive associations between brain malondialdehyde (MD) and short chain fatty acids (PA and acetic acid), and the negative correlations between polyunsaturated fatty acids and MD as marker of increased oxidative stress, a status recently related to the aetiology and clinical presentation of autism[6, 61]. Another support can be easily found in our recent work in which impaired plasma phospholipids and relative amounts of essential polyunsaturated fatty acids were recorded as predictive biomarkers in Saudi patients with autism. Based on this, we propose that altered absolute and relative amounts of fatty acids may contribute to ASD and can be used as biomarkers for early detection of neurotoxicity related to this disorder.