Alzheimer and heart disease share a common cast of villains: cholesterol, inflammation, and macrophage activation are involved in both. Could the diseases also have a common treatment? That’s one implication of a growing number of studies on the liver X receptors (LXRs), a pair of sterol-binding nuclear receptors that control cholesterol metabolism and inhibit inflammatory gene expression.

The latest contribution in the story comes from the lab of Peter Tontonoz at University of California, Los Angeles. In a paper in the June 11 online edition of PNAS, Tontonoz and colleagues show that knocking out either of the LXRα or β genes in APP/PS1 mice leaves the animals with higher amyloid plaque load.

“The LXRs play an important role in macrophage activity and inflammation, and are known to regulate cholesterol metabolism. When we looked for physiological disease states where these pathways might be important, AD was an obvious candidate,” Tontonoz said.

Their results, coupled with evidence, from their own and other studies, that activating LXRs with a synthetic agonist represses neuroinflammation while promoting Aβ clearance, provide in vivo target validation to support the idea that LXR agonists, now in development for cardiovascular disease and atherosclerosis, might also be used to treat AD.

The natural ligands of LXRs are cholesterol-derived oxysterols. When occupied with oxysterols, LXRs regulate the expression of genes in the cholesterol metabolism pathway, as well as inhibit the expression of inflammatory cytokines. Their actions in the liver and other tissues have been extensively studied, but what they do in the brain has been less clear.

The new work, done in collaboration with the lab of Gary Landreth at Case Western Reserve University School of Medicine in Cleveland, Ohio, follows on results from several investigators using LXR agonists in cell lines and AD mouse models. According to those studies, activation of LXR both inhibits Aβ production and enhances Aβ clearance (for a recent review, see Koldamova and Lefterov, 2007).

In this study, first author Noam Zelcer and colleagues took a different approach, by crossing LXR knockout mice with AD mice to gauge the effect of LXR on amyloid pathology in vivo. In 32-week-old APP/PS1 transgenic mice lacking either LXRα or LXRβ, the investigators found that cortical amyloid plaques were larger and more numerous. The increase was statistically significant in the LXRα knockouts, but not in the mice lacking LXRβ. The knockouts showed no change in APP levels or processing.

The LXRs could affect Aβ clearance by macrophages, according to the next set of experiments. LXRs are potent inhibitors of inflammation, and experiments with glial cells in vitro suggested that their effects carry over to the inflammation stimulated by fibrillar Aβ. In cultures of mixed glial cells from WT or double knockout mice, the cells responded to fibrillar Aβ with an induction of inflammatory cytokines including RANTES, TNFα, Mip1β and IL-1β. The response was inhibited by an LXR agonist and enhanced in cells from LXR knockout mice. Inflammatory cytokines have been shown to inhibit macrophage phagocytic activity, and in agreement with this idea, the researchers found that IL-1β or LPS treatment of BV2 microglia cells inhibited phagocytosis of fibrillar Aβ. An LXR agonist that inhibited the inflammatory response stimulated phagocytosis of fibrillar Aβ in the cells.

The results suggest that LXRs, by inhibiting neuroinflammation, could help pump up Aβ clearance by activated glia, which would account for the worse pathology in knockout mice. However, there could be other explanations. Loss of either LXRα or β resulted in lowered expression of the cholesterol transporter ABCA1 in young (12-week-old) APP/PS1 mice. A deficiency in ABCA1 was previously shown to result in Aβ accumulation (see ARF related news story).

Both the ABCA1 and inflammatory pathways are likely to contribute to the effects of LXRs on amyloid load, Tontonoz says. “We don’t have definite evidence which is more important, and we speculate that both are important.”

Besides ABCA1, there are many other genes that are involved in cholesterol metabolism, Tontonoz added. “LXR regulates many genes in that pathway, so it may be an advantage to modulate the whole pathway,” he said.

When it comes to LXR agonists, most researchers think of cardiovascular disease and atherosclerosis, and AD would be a new application for these compounds. The compounds are not without side effects, but several companies are fairly far along on finding agonists that have the beneficial effects of reducing cellular cholesterol without the downside of inducing fatty liver and high plasma lipids, Tontonoz says.

In the meantime, look for more data on the effects of LXR agonists in mouse AD models in the near future. Both Tontonoz and the University of Pittsburgh’s Rada Koldamova and Iliya Lefterov report they have papers submitted with much more information on the action and mechanisms of the compounds in additional AD mouse models.—Pat McCaffrey


  1. The paper by Zelcer and coauthors (1) comes from a leading laboratory in the field of LXR research (P. Tontonoz, UCLA), and the results of the study further support the role of LXRs in the pathogenesis and development of Alzheimer disease—a story that began more than 4 years ago. Now, Zelcer et al. demonstrate that APP-expressing mice with global deletion of either LXRα or LXRβ have an increased number of cortical plaques. The paper shows that loss of LXRα/β is correlated with significantly reduced expression of ABCA1 and ABCG1 in the brain, which implies a potential protective role for either or both of these transporters. Surprisingly, whereas the protein level of ApoE was decreased in LXR null mice, there was no difference in ApoE mRNA between LXRα/β-/- and wild-type mice. Interestingly, despite the fact that the ApoE gene is a target for both LXRs, the authors observed no effect of LXR ligands on ApoE expression in whole brain. As the authors note, a possible explanation is that LXR activation may alter the post-translational stability of ApoE by regulating its ABCA1-dependent lipidation. One additional explanation, however, could be that ApoE transcription is regulated differently in different brain cells. Our recent data suggest such a possibility: we found that following LXR treatment, ApoE mRNA is increased in glia but not in neurons (Lefterov and Koldamova, manuscript under review), explaining why there is no or little effect if RNA for RT-QPCR is purified from whole brain.

    A larger part of the article describes in vitro experiments with isolated primary glial cells from LXR knockout and WT animals, transcriptional profiling of BV2 cells treated with LXR agonists and expression level of lipid inducible genes in the brain of WT and LXR knockout mice. At the end of their discussion the authors draw the important conclusion that the data warrant further investigation on the LXR pathway as a potential target for AD treatment.

    The data in the paper further support what has already been published or presented at SfN and AD meetings:

    imageGene profiling using brain RNA from APP-expressing and WT animals treated with LXR agonists for a month has shown upregulation of genes involved in cholesterol efflux and cholesterol transport within the brain and downregulation of genes involved in inflammation (Koldamova et al., presented at 2006 SfN meeting, Atlanta, GA).

    imageIt was demonstrated that LXR ligands decrease Aβ levels in APP transgenic and non-transgenic mice expressing wild-type ABCA1; importantly, if applied to AD animals with global deletion of ABCA1, the effect of the synthetic LXR agonist T0901317 is undetectable (2-5).

    imageIn primary glial cultures and in vitro models of neuronal toxicity and survival, LXR agonists exert inhibitory effects on cytokine induction and increase neuronal survival (Lefterov et al., presented at 2006 SfN meeting, Atlanta, GA).

    imageThe potential of T0901317 to decrease Aβ42 levels and to completely reverse contextual memory deficits was demonstrated in relatively young Tg2576 mice (5).

    What are the implications? The results of all these studies provide the fundamentals, first, to design and develop preventive and therapeutic strategies that hopefully can manipulate and deal with brain apolipoproteins. ApoE has been insufficiently explored so far in the context of the metabolic pathways controlled by ABCA1 and other cholesterol transporters, which mediate cholesterol efflux and lipidation of brain apolipoproteins. We begin to realize that LXR-ABCA1-ApoE/ApoA-I axis, which influences APP processing/Aβ aggregation and clearance, is amenable to pharmacological modulation, and therefore is worth pursuing as a novel therapeutic approach for AD. Second, these studies will hopefully stimulate research to better understand LXR controlled signaling pathways and molecular mechanisms responsible for AD-related neuroinflammation, and thus the design of a novel class of anti-inflammatory agents. Third, but not least important, the expansion of these studies will certainly reveal the role of ABCA1, other transporters (ABCG1, ABCA7), and brain apolipoproteins in glial/neuronal interactions, neuronal regeneration, and synaptic plasticity.

    The paper published by Zelcer et al. leaves enough unanswered questions, though, to keep the readers curious about other endpoints and issues that on one side could have been addressed, but on another will certainly require additional research. For example, how does the global deletion of LXRs influence APP processing? How does the actual amyloid plaque load (including plaques in the hippocampus) change in PS1APP mice with or without LXRα or LXRβ? Is there a difference in the amount of Aβ deposited in diffuse and compact plaques in animals with disrupted LXR? Is there a difference in the amount of soluble and insoluble Aβ peptides and how, if at all, does LXR deletion influence the formation and amount of soluble oligomers (which will inevitably lead to behavioral experiments), etc.? We hope that with more and more laboratories working on LXR-controlled genes and AD, the answers to these questions are just a matter of time.


    . Attenuation of neuroinflammation and Alzheimer's disease pathology by liver x receptors. Proc Natl Acad Sci U S A. 2007 Jun 19;104(25):10601-6. PubMed.

    . The effects of ABCA1 on cholesterol efflux and Abeta levels in vitro and in vivo. J Neurochem. 2006 Aug;98(3):792-800. PubMed.

    . Role of LXR and ABCA1 in the pathogenesis of Alzheimer's disease - implications for a new therapeutic approach. Curr Alzheimer Res. 2007 Apr;4(2):171-8. PubMed.

    . The liver X receptor ligand T0901317 decreases amyloid beta production in vitro and in a mouse model of Alzheimer's disease. J Biol Chem. 2005 Feb 11;280(6):4079-88. PubMed.

    . The LXR agonist TO901317 selectively lowers hippocampal Abeta42 and improves memory in the Tg2576 mouse model of Alzheimer's disease. Mol Cell Neurosci. 2007 Apr;34(4):621-8. PubMed.

  2. In their paper, Zelcer et al. show that LXR-null mice bred to a mouse model of Alzheimer disease have increased amyloid deposition. Using a murine microglial cell line and primary murine microglial cultures, they also demonstrate that LXR agonists decrease some markers of inflammation in response to fibrillar Aβ40 and may increase phagocytosis. They hypothesize that LXR-null mice may develop increased amyloid deposition because they have a greater inflammatory response and inhibition of phagocytosis. However, the LXR-null mice have an approximately 40 percent reduction in ABCA1, which previous studies suggest could completely account for the increased amyloid deposition that Zelcer and colleagues found. ABCA1 lipidates ApoE, which likely protects ApoE from rapid catabolism and affects the chaperone-like binding of ApoE to Aβ. This explains why ABCA1 knockout mice crossed to mouse models of Alzheimer disease have low levels of lipid-poor ApoE that may directly promote aggregation of Aβ.

    Independent of the reason why the LXR-null mice had increased amyloid deposition, the authors' data showing that LXR is an important regulator of inflammation in microglial cells is intriguing. A maladaptive inflammatory response of the brain to amyloid is likely key to the development of neurodegeneration in Alzheimer disease. The finding that LXR agonists may modulate the immune response in a beneficial manner is exciting and should provoke more research into whether LXR is an appropriate therapeutic target.

Make a Comment

To make a comment you must login or register.


News Citations

  1. ABCA1 Loss Lowers ApoE, Not Amyloid; New ApoE Immunology

Paper Citations

  1. . Role of LXR and ABCA1 in the pathogenesis of Alzheimer's disease - implications for a new therapeutic approach. Curr Alzheimer Res. 2007 Apr;4(2):171-8. PubMed.

Further Reading

Primary Papers

  1. . Attenuation of neuroinflammation and Alzheimer's disease pathology by liver x receptors. Proc Natl Acad Sci U S A. 2007 Jun 19;104(25):10601-6. PubMed.