Credit: CORBIS

A hallmark of Alzheimer's disease (AD) is the accumulation of amyloid-β (Aβ) peptides, which are released from neurons after cleavage of amyloid precursor protein (APP). Soluble oligomeric Aβ peptides, particularly Aβ42, have neurotoxic functions; however, the molecular mechanisms underlying their generation and toxicity are poorly understood. Reporting in Neuron, Yoon and colleagues have dissected the steps of Aβ42 generation and toxicity, and demonstrate that AD is a metabolic disease in which Aβ42 peptides induce a translational block that leads to neurotoxicity and cleavage of APP in a complex feed-forward loop that is tightly controlled by JUN N-terminal kinase 3 (JNK3). Deletion of JNK3 in a mouse model of AD was shown to alleviate symptoms.

These results indicate that AD is a metabolic disease that is under tight control by JNK3.

The idea that Aβ42 oligomers might induce a block of protein translation was based on the fact that treatment of mice with Aβ42 causes similar long-term potentiation (LTP) and memory impairments as treatment with compounds that induce a translational block. Using cultures of rat hippocampal neurons, the authors discovered that Aβ42 induces the phosphorylation of AMP-activated protein kinase (AMPK), a kinase that responds to energy imbalance in the cell. AMPK, in turn, inhibits the mammalian target of rapamycin (mTOR) pathway, inducing endoplasmic reticulum (ER) stress and a rapid translational block that initiates the unfolded protein response (UPR). The UPR had previously been shown to induce the activation of the JNK pathway. In the nervous system, JNK3 is the predominant JNK isoform, and further in vitro experiments revealed that activated JNK3 can phosphorylate APP at position T668P. This, in turn, was shown to lead to the internalization of APP and its cleavage into pathogenic Aβ42 — thereby forming a positive feedback loop.

The authors then set out to explore where and how this vicious cycle can be interrupted. Further in vitro experiments showed that the increase in Aβ42 production in response to a translational block is dependent on JNK3 — thus implying that JNK3 can control this process.

Increased JNK3 activation had previously been reported in brains of patients with AD and in the mouse model of familial AD (FAD), but its role was not well understood. To gain insight into the role of JNK3 in AD pathogenicity in vivo, the authors introduced a JNK3 deletion into FAD mice. Brain samples of FAD:JNK3−/− mice showed dramatically reduced levels of insoluble Aβ42; at 6 months of age, the mice had an 87% reduction in Aβ42 plaque load compared to FAD:JNK3+/+ mice. Moreover, the number of neurons in the frontal cortex was higher in FAD:JNK3−/− mice compared to their FAD:JNK+/+ counterparts (although it did not reach the levels seen in non-FAD mice), and they performed significantly better in memory tasks.

Taken together, these results indicate that AD is a metabolic disease that is under tight control by JNK3, and that JNK3 perpetuates the cycle of an Aβ42-induced translational block via ER stress, JNK3 activation and further production of Aβ42. JNK3 could therefore be a promising new target for the treatment of AD.