Diseases such as Alzheimer’s and amyotrophic lateral sclerosis come with a hefty dose of neuronal stress. Could blocking just one stress response activator, the dual leucine zipper kinase (DLK), help? Yes, according to scientists at Genentech led by Joseph Lewcock. In the August 16 Science Translational Medicine, they report that in mouse models of both diseases, knocking out or inhibiting this kinase protected against neurodegeneration and even led to some behavioral benefits. “Our findings suggest the potential for inhibition of DLK as a therapeutic for these diseases,” said Lewcock, who is now at Denali Therapeutics in South San Francisco. Because DLK seems to be active only in neurons and only when they are under duress, blocking it may prove beneficial, other researchers agreed.
“This is a really exciting paper,” said George Perry, University of Texas at San Antonio, who was not involved in the study. “It’s a new avenue looking at stress responses and how they link to aging, disease, and the normal physiology of the brain.” He agreed that this may offer new opportunities for pharmaceutical intervention.
Neurons make most of the DLK in the body. The kinase senses many types of acute neuronal injury, such as being crushed or severed. When active, it phosphorylates the c-Jun N-terminal kinase (JNK), which kicks off a stress response cascade. That leads to the downstream phosphorylation of c-Jun, which upregulates genes that are proapoptotic. Lewcock and colleagues wondered if DLK was chronically activated in neurodegenerative diseases, and if blocking it would rescue neurons from neurodegeneration and associated deficits.
To find out, co-first authors Claire Le Pichon and William Meilandt examined phosphorylated c-Jun (p-c-Jun) in several animal models of ALS and AD. In 14-week-old SOD1G93A mice, there were more p-c-Jun-positive motor neurons in the lumbar spinal cord than in nine-month-old animals, suggesting an age-related increase in pathway activity. TDP-43 A315T mouse models of ALS also had more p-c-Jun in the cortex than did wild-type animals. The pathway appeared to be overactive in amyloid and tau models as well; PS2APP and TauP301L mice both had more than double the p-c-Jun in hippocampal neurons at six months than non-transgenic controls.
The pathway also seemed to be upregulated in people with neurodegenerative disease. Patients with sporadic ALS had five times more p-c-Jun in the lumbar spinal cord than did healthy controls, while in postmortem brain samples from early and late-stage AD patients, levels of p-c-Jun in hippocampal neurons topped those in healthy control samples by 200 percent. In AD patients, the elevated p-c-jun correlated with the amount of neurofibrillary tangles in the superior frontal gyrus, suggesting a link with pathology.
Would knocking down DLK help these various models of disease? The researchers used the tamoxifen-induced Cre recombination system to knock out DLK in the brain after SOD1G93A, PS2APP, and TauP301L mice had reached adulthood but before they developed pathology. The knockout was about 80 percent complete, and reduced p-c-Jun in the neurons by about 40 percent. Whereas SOD1G93A mice with the normal DLK signaling lost about 40 percent of their spinal cord motor neurons seven weeks after tamoxifen was started, the conditional DLK knockouts (cKO) lost just 13 percent (see image above). And while axonal lumen area shrank in the SOD1G93A mice, it held steady in the knockouts. Both microglia and astrocytes were less reactive and the cKOs lived on average 162 days, eight days longer than the 154-day lifespan of controls. This may seem modest, said Lewcock, but it has proven consistently difficult to increase lifespan in this particular model. The cKO mice also maintained their strength longer, as they were better able than SOD1G93A control mice to hang from a wire at 14 weeks old, he added.
Knocking out DLK also helped AD mouse models. Again, the authors used the tamoxifen-cre system to turn off DLK in 10-week-old PS2APP mice. These animals lose dendritic spines, especially near plaques. Meilandt used confocal two-photon microscopy to count synapses in the superficial layers of the cortex in nine-month-old mice. He found the cKOs lost 30 percent fewer near plaques in the somatosensory cortex (see image at left). While the two-photon technique did not allow the researchers to probe deeper regions of the brain, such as the hippocampus, Lewcock said he expects synapses there were spared as well. In keeping with this, the cKOs also learned an active avoidance task, in which they were taught to move to a trained location to avoid a mild electric shock, better than did control PS2APP mice. This improvement occurred even if the researchers waited until the animals were six months old to delete DLK, at which point the mice had accumulated plaques. As for the tau P301L mice, DLK knockouts lost 20 percent fewer neurons from the subiculum than did the controls.
Lastly, in SOD1G93A mice the researchers tried tuning down DLK signaling with two inhibitors developed by Genentech—GNE-8505 and GNE-3511. Not only did animals treated acutely with the inhibitors have fewer p-c-Jun-positive cells than vehicle-treated controls, but after four weeks of treatment they lost 10 percent fewer neuromuscular junctions. The researchers did not test the effect of long-term treatment on lifespan.
The results suggest that blocking DLK signaling could protect neurons, the authors wrote. Lewcock said that Genentech has completed safety studies on yet a related DLK inhibitor, GDC-0134, and is now running a Phase 1 trial in ALS patients.
“It was exciting to see that this pathway could potentially be a relevant target for neurodegenerative diseases,” said Catherine Collins, University of Michigan, Ann Arbor. She has studied DLK signaling extensively and speculated that it plays a role, but was intrigued to see it finally tested in several mouse models. She has data in press at eLife that points to a role for DLK signaling in synapse loss, a result that complements the current paper, she said.
“This is a nice, carefully done paper,” wrote Thomas Sudhof, Stanford School of Medicine, California, to Alzforum. “The data are compelling; DLK inhibition clearly benefits the animal models tested.” However, that DLK is central to many processes raises the question of whether it can be targeted therapeutically, he noted. Kinases are notoriously hard to target safely, as they have pleiotropic effects. However, the authors saw no adverse effects in the knockouts, only minimal weight loss. That this kinase is specific for neurons should help, Lewcock told Alzforum. Collins agreed. “As far as kinases go, this is a logical one to target because of its specificity to neurons and conditions of stress,” she said. “You might not want to completely inhibit it because there are also beneficial aspects to the stress response. But if you could tune it, that could be helpful.” Genentech researchers are now doing similar experiments in models of other neurodegenerative diseases.—Gwyneth Dickey Zakaib
Research Models Citations
- Welsbie DS, Mitchell KL, Jaskula-Ranga V, Sluch VM, Yang Z, Kim J, Buehler E, Patel A, Martin SE, Zhang PW, Ge Y, Duan Y, Fuller J, Kim BJ, Hamed E, Chamling X, Lei L, Fraser ID, Ronai ZA, Berlinicke CA, Zack DJ. Enhanced Functional Genomic Screening Identifies Novel Mediators of Dual Leucine Zipper Kinase-Dependent Injury Signaling in Neurons. Neuron. 2017 Jun 21;94(6):1142-1154.e6. PubMed.
- Larhammar M, Huntwork-Rodriguez S, Jiang Z, Solanoy H, Sengupta Ghosh A, Wang B, Kaminker JS, Huang K, Eastham-Anderson J, Siu M, Modrusan Z, Farley MM, Tessier-Lavigne M, Lewcock JW, Watkins TA. Dual leucine zipper kinase-dependent PERK activation contributes to neuronal degeneration following insult. Elife. 2017 Apr 25;6 PubMed.
- Le Pichon CE, Meilandt WJ, Dominguez S, Solanoy H, Lin H, Ngu H, Gogineni A, Sengupta Ghosh A, Jiang Z, Lee SH, Maloney J, Gandham VD, Pozniak CD, Wang B, Lee S, Siu M, Patel S, Modrusan Z, Liu X, Rudhard Y, Baca M, Gustafson A, Kaminker J, Carano RA, Huang EJ, Foreman O, Weimer R, Scearce-Levie K, Lewcock JW. Loss of dual leucine zipper kinase signaling is protective in animal models of neurodegenerative disease. Sci Transl Med. 2017 Aug 16;9(403) PubMed.