The report is the first demonstration of TDP-43 aggregation and toxicity in a mammalian model, said Aaron Gitler of the University of Pennsylvania, who was not involved in the work. Gitler has made a similar demonstration in yeast (ARF related news story and Johnson et al., 2008). “A big focus in the field now is just to try to figure out how TDP-43 contributes to disease,” he said, which would guide therapeutic approaches.

Petrucelli, first author Yong-Jie Zhang and colleagues previously found that TDP-43 is chopped by caspase-3 into carboxyl-terminal fragments of 25 and 35 kD (ARF related news story and Zhang et al., 2007). In the current work, they sought the downstream effects of that truncated TDP-43, expressing the same carboxyl-terminal fragments, tagged with GFP, in cell culture. In human embryonic kidney cells, full-length TDP-43 localized to the nucleus, but the fragments moved into the cytoplasm and formed aggregates including ubiquitin, as the protein does in people with TDP-43 proteinopathies. The scientists focused further experiments on the shorter, 25-kDa piece, since it has been linked to ALS pathology (ARF related news story and Neumann et al., 2006).

The fragment was a killer. Expressed in differentiated neuroblastoma cells, the shortened TDP-43 led to fragmented nuclei and activation of caspase-3, both signs of apoptosis. Zhang and colleagues then asked whether the fragment required phosphorylation to exert its toxic effects. TDP-43 fragments from the brains of people who had ALS or FTLD with ubiquitin inclusions (the form that includes TDP-43 proteinopathy) are phosphorylated (Hasegawa et al., 2008). Using a phospho-TDP-43 antibody, the researchers discovered that in cells transfected with the truncate, the fragment was phosphorylated, but they found that a mutant 25 kD fragment lacking the serines necessary for the modification aggregated as well, showing that phosphorylation is not required for inclusion formation.

Zhang and colleagues reasoned that the fragment could cause cell death by either exerting its own toxic influence, or by binding and sequestering full-length TDP-43, dragging it away from the nucleus where it regulates transcription and RNA splicing. To distinguish between these two possibilities, they tested endogenous TDP-43 function. The scientists transfected HeLa cells with the cystic fibrosis transmembrane conductance regulator (CFTR) as a marker for normal TDP-43 activity. Wild-type TDP-43 prevents exon 9 expression in this gene. Exon 9 transcripts were not increased in cells co-transfected with the TDP-43 fragment, suggesting endogenous TDP-43 nuclear localization and activity were unaffected by the truncate. The fragment, then, must have its own toxic mechanism.

The paper provides evidence for a TDP-43 toxic gain of function, but Petrucelli noted that they only focused on a single piece of the protein. “We are not arguing that the caspase fragment is the only fragment that is found in ALS and FTLD,” he said. “Clearly other fragments are possible,” and those could have other detrimental effects.

In the course of their research, Zhang and colleagues engineered an antibody specific for the 25 kDa TDP-43 fragment. “I think that was the most exciting part of the paper, in some respects,” Petrucelli said. Like the phospho-TDP-43 antibody, this one could be useful in distinguishing FTLD with ubiquitin and TDP-43 inclusions from the tau-based form of the disease, Petrucelli suggested. That might help doctors match treatments to different kinds of FTLD, he said.

The current study seems to solidify the toxic gain of function for TDP-43. “This shows that these carboxyl-terminal fragments might be playing a direct role in the disease pathogenesis,” Gitler said, although he noted the fragment’s toxicity should be confirmed in animals. But another study, published online April 17 in FEBS Letters, found that fruit flies lacking TDP-43 had locomotion problems, abnormal neuromuscular junctions, and reduced lifespan (Feiguin et al., 2009). That work adds points to the loss of normal function column. Ultimately, there may not be a simple distinction between acquired toxicity and lost utility. “I think it’s kind of both,” Gitler said.

To further elucidate the role of TDP-43 in disease, scientists are racing to discover or create models of TDP-43 pathology in animals (ARF related news story and Tatom et al., 2009). In an April 17 paper in Neuroscience Letters, scientists report that TDP-43 aggregates with ubiquitin in a common ALS model system—mice overexpressing mutant human superoxide dismutase (see Shan et al., 2009). However, these researchers did not report carboxyl-terminal fragments or hyperphosphorylated TDP-43, even in end-stage animals, adding a new layer of complexity to the TDP-43 puzzle.—Amber Dance.

Reference:
Zhang Y-J, Xu Y-F, Cook C, Gendron T, Roettges P, Link CD, Lin W-L, Tong J, Castanedes-Casey M, Ash P, Gass J, Rangachari V, Buratti E, Baralle F, Golde T, Dickson DW, Petrucelli L. Aberrant cleavage of TDP-43 enhances aggregation and cellular toxicity. PNAS Early Edition April, 2009. Abstract