17 October 2011. We know that overindulging in cheeseburgers and fries is bad for our hearts, and growing evidence indicates it may be no better for our brains. A high-fat, high-cholesterol diet has been fingered as a risk factor for sporadic Alzheimer’s disease, with some data suggesting that such a diet increases levels of the membrane lipid ceramide and thereby messes up amyloid precursor protein (APP) processing. A new study strengthens the link among AD, fatty foods, and ceramides, and describes a novel pathway for regulating ceramides through microRNAs. These tiny, non-coding RNAs, about 21 nucleotides long, bind to the tails of specific messenger RNAs to block protein synthesis. In the October 12 Journal of Neuroscience, Hirosha Geekiyanage and Christina Chan at Michigan State University, East Lansing, report that several microRNAs regulate an enzyme that produces ceramide. In cell cultures, inhibiting these miRNAs increased both lipid and Aβ levels, while overexpressing them did the opposite. In a small sample of human AD brains, the researchers found low levels of the miRNAs and high levels of ceramide, hinting that the loss of these miRNAs contributes to pathology in some sporadic AD cases. The results point to a possible pathway for therapeutic intervention and potential for a diagnostic.
“This is an interesting and well-conceived study combining patient, mouse, primary cell, and in-vitro data. It further strengthens the link between microRNA dysregulation and sporadic AD,” said Sébastien Hébert at Centre de Recherche du CHUQ, Québec, Canada. He was not involved in the research. “This paper also validates previous microarray studies suggesting that specific microRNAs, such as miR-29, miR-181, and miR-9, are misregulated in [human] disease.” (See, e.g., Lukiw, 2007; Cogswell et al., 2008; Nunez-Iglesias et al., 2010; and Shioya et al., 2010.)
In the last few years, microRNAs have jumped to the attention of AD researchers, with several studies suggesting they play a role in regulating APP, presenilin, and BACE (see, e.g., ARF related conference story and ARF conference story). A paper published online September 23 in The European Molecular Biology Organization Journal examined the role of miRNAs in learning, memory, and dementia. Researchers led by Andre Fischer at the University of Göttingen, Germany, used a technique called massive parallel sequencing to look for miRNAs enriched in mouse hippocampus. One such miRNA, miR-34c, caught their eye because many of its predicted mRNA targets encode genes upregulated after fear conditioning, a type of associative learning. Among its putative targets is the sirtuin SIRT1, which several studies have implicated in memory processes (see, e.g., ARF related news story and ARF news story). The authors found high levels of miR-34c in aged mice, APP/PS1 transgenic mice, and human AD brains. Overexpressing miR-34c in wild-type mice led to poor memory, while inhibiting it in AD model mice rescued memory. However, some researchers noted that studies by other groups report low levels of miR-34c in control and AD brains, and questioned whether it could be abundant enough to have such profound effects.
Chan and Geekiyanage were interested in the link among high-fat diets, ceramide, and increased risk of AD, which has been shown in animal models and cell cultures (see, e.g., Patil et al., 2007; Shah et al., 2008; Julien et al., 2010; see also ARF related news story). Ceramide levels are high in people with mild cognitive impairment (see ARF related news story). These lipids stabilize BACE1 (see ARF related news story). Ceramide forms the backbone of sphingolipids, which may cluster in lipid rafts and alter APP processing (see ARF related news story; ARF news story; and ARF news story).
The enzyme serine palmitoyltransferase (SPT) catalyzes the rate-limiting step in ceramide production. Since scientists have reported that SPT activity can increase without any change in its mRNA level, the authors reasoned that the enzyme might be regulated post-transcriptionally. They used a predictive algorithm to find miRNAs that could potentially bind SPT mRNA, then tested candidates in a reporter assay and in primary rat cell culture. They identified five miRNAs with activity: miR-181c, miR-137, miR-9, miR-29a, and miR-29b-1. To look for relevance to human disease, the researchers examined seven AD and seven control brains. The average levels of the five miRNAs in AD frontal cortex extracts were about one-quarter of what they were in controls. Correspondingly, SPT and ceramide were about threefold higher in AD than in control brain.
To connect these findings to diet, the authors fed male mice high-fat chow for five months. The mice ended up with more SPT and ceramide, and less miR-137, miR-181c, and miR-9 in their brains than controls. In primary astrocyte cultures made from transgenic Swedish APP mice, overexpressing miR-137 or miR-181c caused a one-third drop in SPT and Aβ levels. Conversely, transfecting the cultures with anti-miR-137 or -181c nearly doubled SPT and Aβ levels. These gain-of-function and loss-of-function experiments suggest a causal relationship between these miRNAs and Aβ levels, the authors write.
In future work, the scientists plan to see if injecting an SPT inhibitor into mice eating fatty foods protects against the diet’s negative effects, Chan told ARF. “Being able to regulate SPT opens up therapeutic possibilities and diagnostic approaches for addressing AD,” she said, adding that she will look at levels of these miRNAs in human and animal serum to investigate their potential as AD biomarkers.
Hébert agrees that microRNAs may be useful as a diagnostic tool, as they are found in all body fluids and are fairly stable. However, he cautions, “I think we have to be careful about microRNA-based therapies, because they can potentially modulate several genes and pathways.” He suggested that a next step should be to find out if the miRNAs identified by Geekiyanage and Chan can modulate pathology in vivo in AD model mice. Converging lines of evidence suggest that these particular miRNAs may play roles in AD, Hébert noted. For example, miR-9 and miR-29 have been shown to suppress BACE1 expression (see ARF related news story). Several studies have indicated a role for miR-137 in regulating processes such as neuronal development and tau splicing (see, e.g., Silber et al., 2008; Smrt et al., 2010; and Smith et al., 2011). Intriguingly, a recent genomewide association study (GWAS) identified a single nucleotide polymorphism (SNP) in the miR-137 locus associated with schizophrenia (see The Schizophrenia Psychiatric GWAS Consortium et al., 2011). Hébert pointed out that current GWAS include few SNPs in microRNA regions, and suggested that more genetic links may remain to be discovered.—Madolyn Bowman Rogers.
Geekiyanage H, Chan C. MicroRNA-137/181c regulates serine palmitoyltransferase and in turn amyloid β, novel targets in sporadic Alzheimer’s disease. J Neurosci. 2011 Oct 12;31(41):14820-30. Abstract
Zovoilis A, Agbemenyah HY, Agis-Balboa RC, Stilling RM, Edbauer D, Rao P, Farinelli L, Delalle I, Schmitt A, Falkai P, Bahari-Javan S, Burkhardt S, Sananbenesi F, Fischer A. MicroRNA-34c is a novel target to treat dementias. EMBO J. 2011 Sep 23. Abstract