Tau tangles are a hallmark of Alzheimer’s disease and other tauopathies, but how these aggregates relate to cognitive decline remains a mystery. Writing in the October 10 Nature Medicine, researchers led by Karen Ashe of the University of Minnesota in Minneapolis report that rather than aggregates, a small soluble fragment of tau may orchestrate the neuronal dysfunction that leads to a cognitive downslide. They found that caspase-2 cleaves tau to generate a 35kDa fragment, which they dubbed Δtau314. This fragment appears elevated in the brains of people with mild cognitive impairment or AD. In transgenic mouse models of tauopathy, Δtau314 infiltrated dendritic spines and somehow coerced full-length tau to do the same, the authors proposed. This redistribution of tau disrupted synaptic function and ultimately promoted memory problems in mice, they argued. If this holds up, blocking caspase-2 might restore synaptic function in people suffering from AD or other tauopathies.

Tau normally stabilizes microtubules, but in neurodegenerative disease, the protein forms fibrils that accumulate in neurons, particularly in synapse-laden dendritic compartments. While the presence of these protein tangles in the cortex correlates with neurodegeneration and cognitive impairment in people with AD (see May 2016 news and Aug 2016 news), growing evidence suggests soluble forms of tau, particularly tau oligomers, may be toxic to neurons (see Nov 2010 conference news; Sep 2013 news). In fact, in 2005 Ashe’s group was one of the first to spark interest in soluble forms of toxic tau. They reported that switching off expression of mutant human tau in mice improved synaptic function despite the persistence of neurofibrillary tangles. The findings implied tangles were not the toxic entity (see Jul 2005 news). Ashe has been searching for putative soluble culprits ever since.

To uncover soluble forms of tau that directly associate with synaptic dysfunction, first author Xiaohui Zhao and colleagues turned to three-month-old rTg4510 mice. At this age, the mice, which express human tau with the P301L mutation, start to develop memory deficits but lack obvious tangles, synaptic loss, or neurodegeneration. The researchers tested 14 animals (consisting of roughly equal numbers of males and females) in the Morris water maze. Then, while blinded to the memory data, they measured various known soluble forms and fragments of tau using a panel of antibodies and western blotting. It turned out that while none of the known forms of tau correlated with memory deficits, one 35kDa tau fragment did. When Zhao then switched off tau expression in the animals, she found that insoluble tau species continued to accumulate, but the 35kDa fragment disappeared, and the animals performed better in the water maze. The findings hinted that this soluble fragment may underlie the animals’ memory problems. However, in vitro, the fragment did not fibrillize, and the researchers found no trace of it when they solubilized aggregates of tau.

The researchers used mass spectroscopy to determine the fragment comprised amino acids 299-314 of the full-length protein. Given the aspartate at its carboxy terminus, they predicted that cleavage at this residue was likely the work of a caspase. In in vitro experiments, they found that only caspase-2 could generate Δtau314 from full-length tau, and that mutating Asp314 to glutamate (D314E) prevented this cleavage. When they knocked down caspase-2 in rTg4510 mice, Δtau314 levels dropped, and the animals’ performance in the water maze improved.

Could Δtau314 play a role in cognitive impairment in people? The researchers generated antibodies against the fragment, and used them to probe postmortem brain tissue from 85 elderly people—30 with mild cognitive impairment, 25 with AD, and 30 who were cognitively normal. They detected more Δtau314 in the brains of people with mild cognitive impairment or AD than in the cognitively normal controls. 

Since Δtau314 seems to resists aggregation, how does it weaken memory? Given that Ashe’s lab and others had previously reported that tau triggered synaptic problems when it moved into dendritic spines, the researchers scrutinized the location of tau fragments by labelling them with green fluorescent protein. Dendritic spines lit up in rat hippocampal neurons transfected with the Δtau314 fragment. The same happened in neurons transfected with full-length P301L, as reported previously, however, not in caspase-2 deficient neurons, or when tau Asp314 was mutated to glutamate to ablate the caspase site. Together, these findings indicated that cleavage by caspase-2 facilitated the movement of a tau fragment, or fragments, into dendritic spines.

Interestingly, when the researchers isolated postsynaptic compartments from mice expressing different forms of tau in the hippocampus, they found that full-length tau accumulated in spines from animals expressing P301L tau, but not those expressing D314G P301L tau, suggesting caspase-2-generated fragments somehow drive the full protein into spines. Animals expressing the D314E version of P301L tau were also spared from memory deficits as well as a drop in hippocampal volume, which indicated that caspase-2 cleavage of tau was a crucial component of both synaptic dysfunction and neurodegeneration (see image below). 

Deadly Cleavage.

Animals injected with viruses expressing human P301L tau (left) lost hippocampal volume, but not mice injected with P301L tau that resists caspase-2 (right). [Image courtesy of Zhao et al., 2016.]

Indeed, the researchers found caspase-2-driven synaptic deficits in neurons expressing P301L tau. They had weaker miniature excitatory postsynaptic currents, and expressed less GluR1, a subunit of AMPA receptors, than did normal neurons. Neurons expressing the caspase-2-resistant Asp314Glu variant of P301L tau had normal synapses. The findings hinted that cleavage of tau at Asp314 somehow caused synaptic and cognitive deficits. However, the fragment alone was incapable of causing this damage, as wild-type animals injected with Δtau314 performed normally in the water maze and had a normal complement of synaptic proteins, despite the accumulation of Δtau314 in dendritic spines. Ashe thinks Δtau314 enlists other proteins, perhaps other species of tau, such as full-length P301L tau, to do its dirty work. “When tau is cleaved at Asp314, it actually seems to elicit a broader redistribution of full-length tau within the cell,” Ashe told Alzforum. She said her lab is investigating how Δtau314 promotes this molecular feng shui. 

The findings mesh with previous studies in J20 mice, an AD model in which synaptic and cognitive deficits could be restored by knocking out caspase-2 (see Pozueta et al., 2013). Deleting the caspase also protects a mouse model of Huntington’s disease from behavioral deficits (see Caroll et al., 2011). 

The work adds to a long list of studies linking cleavage of tau to toxicity. Caspase-3 reportedly unleashes tau fragments that form neurofibrillary tangles, and a recent report indicated that these fragments damaged synapses in people with progressive supranuclear palsy (see de Calignon et al., 2010, and Sep 2015 news). Caspase-6 and -9 also have been implicated in dicing up tau (see Guo et al., 2004; Rohn et al., 2002). Other fragments, produced by asparagine endopeptidase (AEP), form tangles and trigger synaptic loss, rather than synaptic dysfunction as Δtau314 does (see Oct 2014 news). 

Keqiang Ye of Emory University School of Medicine in Atlanta, who discovered the AEP tau fragments, proposed that AEP and caspase-2 generate fragments that attack neurons in different ways, as exemplified by the synaptic dysfunction caused by Δtau314 and outright synaptic loss caused by AEP-generated tau fragments.

However, Ye expressed doubts about the feasibility of caspase-2 as a therapeutic target. “Due to the high homology among the caspase family members, it will be extremely challenging to selectively block caspase-2 without inhibiting other cousins,” he wrote.

Huaxi Xu of Sanford Burnham Prebys Medical Discovery Institute in La Jolla, California, whose previous studies helped uncover the toxicity of caspase-3 cleaved tau, proposed a broader approach. “Together with previous evidence showing beneficial effects by inhibiting caspase-3 or AEP in AD, it is worthy to explore a therapeutic strategy targeting caspase-2/3 and AEP altogether or targeting post-translational cleavage of tau,” he wrote.

Ashe added that one of the benefits of targeting caspase-2 over the other proteases is that it may actually reverse synaptic dysfunction, rather than merely suppressing further synaptic loss. “This is a mechanism by which you can block the process that’s damaging synaptic connections, and allow them to repair themselves,” Ashe said. —Jessica Shugart


No Available Comments

Make a Comment

To make a comment you must login or register.


News Citations

  1. On Multiple Marker Analysis, Tangles Track Best With Functional Decline
  2. Brain Imaging Suggests Aβ Unleashes the Deadly Side of Tau
  3. San Diego: Tau Oligomer Antibodies Relieve Motor Deficits in Mice
  4. Chaperone “Saves” Tau, Turning it into Toxic Oligomers
  5. No Toxicity in Tau’s Tangles?
  6. Killer Cleavage: Appoptosin Stokes Tauopathy through Caspase 3
  7. Does Novel Tau Protease Promote Pathology?

Research Models Citations

  1. rTg(tauP301L)4510
  2. J20 (PDGF-APPSw,Ind)

Mutations Citations

  1. MAPT P301L

Paper Citations

  1. . Caspase-2 is required for dendritic spine and behavioural alterations in J20 APP transgenic mice. Nat Commun. 2013;4:1939. PubMed.
  2. . Mice lacking caspase-2 are protected from behavioral changes, but not pathology, in the YAC128 model of Huntington disease. Mol Neurodegener. 2011 Aug 19;6:59. PubMed.
  3. . Caspase activation precedes and leads to tangles. Nature. 2010 Apr 22;464(7292):1201-4. PubMed.
  4. . Active caspase-6 and caspase-6-cleaved tau in neuropil threads, neuritic plaques, and neurofibrillary tangles of Alzheimer's disease. Am J Pathol. 2004 Aug;165(2):523-31. PubMed.
  5. . Caspase-9 activation and caspase cleavage of tau in the Alzheimer's disease brain. Neurobiol Dis. 2002 Nov;11(2):341-54. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Caspase-2 cleavage of tau reversibly impairs memory. Nat Med. 2016 Nov;22(11):1268-1276. Epub 2016 Oct 10 PubMed.