Cells have more than one way to get rid of tau, the tangled protein in the brains of people with Alzheimer disease. Writing in the February 18 Journal of Neuroscience, researchers from Kenneth Kosik’s laboratory at the University of California, Santa Barbara, describe an ubiquitin-independent route for tau to reach the proteasome. The cochaperone BAG2, along with Hsp70, can capture tau and send it to the garbage disposal without the need for ubiquitin tags. This pathway gives scientists another possible place to tackle the pathology of AD.

Tau normally sits astride the microtubules, stabilizing these intracellular tracks. In AD, however, excess phosphorylation of tau likely drives the protein off of the microtubules and induces it to form large aggregates. Tau, then, is an appealing—but challenging—drug target. Kinase inhibitors might block aggregation, Kosik said, but it is difficult to find inhibitors that would specifically target improper tau phosphorylation. “In the field of neurofibrillary pathology, we now have a new target” in the BAG2 tau degradation pathway, Kosik said.

When a tau molecule is in trouble, it attracts a complex including Hsp70, the ubiquitin ligase CHIP, and other chaperones. This is the “triage” point, Kosik said. “It represents a critical decision node for whether or not tau will be salvaged…or the cell just gives up on it and says, ‘It’s over for you, you’re off to the proteasome.’” Kosik’s laboratory had previously found an association between CHIP and tau (Shimura et al., 2004), and another group discovered that CHIP is often found linked to BAG2 (Dai et al., 2005). That led Kosik, along with joint first authors Daniel Carrettiero and Israel Hernandez, to seek a link between BAG2 and tau in cultured cells.

The scientists found that when BAG2 is overexpressed in primary neurons from fetal hippocampus, the levels of insoluble, aggregated tau decreased. The two proteins also co-immunoprecipitated, confirming their association. BAG2 blocks CHIP from attaching ubiquitin to damaged protein (Arndt et al., 2005; Dai et al., 2005), so more BAG2 should, logically, lead to more tau. Since they observed the opposite, the researchers concluded that BAG2 might be sending tau toward destruction, but without ubiquitination. Sure enough, dominant-negative ubiquitin mutations did not prevent BAG2-mediated tau degradation in COS-7 cells.

With ubiquitin out of the picture, the scientists next asked whether the proteasome was necessary for tau degradation. They treated cells with lactacystin to inhibit proteasome activity, and found that levels of phosphorylated tau increased, suggesting that the proteasome is the tau destroyer in the BAG2, non-ubiquitin pathway.

Next, the researchers brought BAG2 gene regulation into the picture. Among the microRNAs dysregulated in Alzheimer disease (Lukiw, 2007), miR-128a was predicted to interact with the BAG2 gene. When the scientists added miR-128a to cultured neurons, BAG2 levels dropped fourfold compared to cells with a scrambled control miRNA. The miRNA treatment also doubled the amount of phosphorylated tau present in the cells. “That part really sold me on the paper,” said Leonard Petrucelli of the Mayo Clinic in Jacksonville. “It really shows a physiological point.”

The results suggest to the authors that there are three possible ways to get rid of troublesome tau. The BAG2 system is more efficient, Hernandez thinks, than the ubiquitin pathway, so may be the best choice for neurons. He compares the cell’s options to two doors: BAG2, a wide-open gate that allows plenty of tau through; and ubiquitin, a smaller door that the tau can squeeze through if necessary. “If you block the big gate, all the proteins have to go through that small door,” Hernandez said. The addition of miR-128 artificially locked the big gate, and mutations in pathway participants might do the same thing, causing disease, he speculated. It is also possible that phosphorylated, tangled tau is a poor candidate for the primary BAG2 degradation pathway, and builds up because the ubiquitin route cannot accommodate all the protein. The third route to get rid of damaged tau, and the road least taken, is to activate caspases to clear up the problematic protein. “This is really a last-ditch effort of the cell,” Kosik said. “When all else fails, you take this very risky route in which you are on the border of apoptosis.”

Having more BAG2, then, might help cells struggling to control tau. It might be possible to get that effect by dampening the effect of miR-128a. One potential method to block miR-128a would be microRNA sponges, said co-author Thales Papagiannakopoulos, who did the miRNA work. These transcripts are driven by strong promoters and they contain multiple binding sites for the target miRNA. The sponges thus sequester the miRNA so it cannot downregulate its target genes (Ebert et al., 2007). “They suck all the miRNA to a certain area in the cell,” Papagiannakopoulos said. However, there are many questions to be answered before he is ready to start sopping up all of a neuron’s miR-128a. Putting the sponge gene into a person would require gene therapy, with all the accompanying risks. In addition, a single miRNA might target dozens of genes, he noted, and “even a subtle difference in their levels might affect a lot of downstream targets.” Such a therapy, then, might have unintended side effects.—Amber Dance


  1. This is an interesting set of results that seems to be suggesting that providing additional Bag2 somehow promotes a ubiquitin-independent proteasomal turnover of tau. I'm wondering how strongly expressed Bag2 might be in this context, relative to its basal level. Regardless, this is probing some interesting circuitry that deserves close attention.

  2. This is a very informative paper from Carretierro et al. describing a novel relationship between BAG2 and tau. It further demonstrates the growing complexity of the chaperone network, moving us away from the idea that the chaperones are merely housekeeping genes that act in an unregulated, automated fashion. It seems that the route of tau clearance will be as complex, if not more complex, than the road to its hyperphosphorylation.

    Members of the degradation process may also offer us more appropriate drug targets for therapeutic intervention in tauopathies and perhaps other diseases of protein misfolding. Identifying which chaperones are most specific for tau degradation could provide us with very novel clinical strategies for Alzheimer disease.

    It should be noted that KNK437 does not inhibit Hsp70 activity but rather its levels via transcriptional repression of not only the Hsp70 gene, but also other heat shock genes (Yokota et al., 2000; Koishi et al., 2001). Changing the expression of Hsp70 levels could have very different consequences from directly inhibiting its ATPase function when clients are already bound.

    One concern from many of these studies, including our own, is that the majority of models we work with overexpress tau. This is quite different from endogenous levels becoming dysfunctional, which is what occurs in human disease. It is quite possible that overexpressed chaperone clients, like tau, are processed quite differently from those that become dysfunctional at endogenous levels. Defining this could be critical for our understanding of the nature of chaperone biology in all protein misfolding diseases.


    . Benzylidene lactam compound, KNK437, a novel inhibitor of acquisition of thermotolerance and heat shock protein induction in human colon carcinoma cells. Cancer Res. 2000 Jun 1;60(11):2942-8. PubMed.

    . The effects of KNK437, a novel inhibitor of heat shock protein synthesis, on the acquisition of thermotolerance in a murine transplantable tumor in vivo. Clin Cancer Res. 2001 Jan;7(1):215-9. PubMed.

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . CHIP-Hsc70 complex ubiquitinates phosphorylated tau and enhances cell survival. J Biol Chem. 2004 Feb 6;279(6):4869-76. PubMed.
  2. . Regulation of the cytoplasmic quality control protein degradation pathway by BAG2. J Biol Chem. 2005 Nov 18;280(46):38673-81. PubMed.
  3. . BAG-2 acts as an inhibitor of the chaperone-associated ubiquitin ligase CHIP. Mol Biol Cell. 2005 Dec;16(12):5891-900. PubMed.
  4. . Micro-RNA speciation in fetal, adult and Alzheimer's disease hippocampus. Neuroreport. 2007 Feb 12;18(3):297-300. PubMed.
  5. . MicroRNA sponges: competitive inhibitors of small RNAs in mammalian cells. Nat Methods. 2007 Sep;4(9):721-6. PubMed.

Further Reading


  1. . In vivo evidence of CHIP up-regulation attenuating tau aggregation. J Neurochem. 2005 Sep;94(5):1254-63. PubMed.
  2. . The high-affinity HSP90-CHIP complex recognizes and selectively degrades phosphorylated tau client proteins. J Clin Invest. 2007 Mar;117(3):648-58. PubMed.
  3. . CHIP and HSPs interact with beta-APP in a proteasome-dependent manner and influence Abeta metabolism. Hum Mol Genet. 2007 Apr 1;16(7):848-64. PubMed.
  4. . BAG-1 associates with Hsc70.Tau complex and regulates the proteasomal degradation of Tau protein. J Biol Chem. 2007 Dec 21;282(51):37276-84. PubMed.
  5. . CHIP and Hsp70 regulate tau ubiquitination, degradation and aggregation. Hum Mol Genet. 2004 Apr 1;13(7):703-14. PubMed.

Primary Papers

  1. . The cochaperone BAG2 sweeps paired helical filament- insoluble tau from the microtubule. J Neurosci. 2009 Feb 18;29(7):2151-61. PubMed.