Julien’s group is interested in non-neuronal ALS pathology. Swarup focused on proteins that TDP-43 interacts with in microglia. There is evidence that these cells, the nervous system’s immune vanguard, are activated in ALS. The signal that initially activates spinal cord microglia is uncertain, Swarup said. They may respond to lipopolysaccharide (LPS, a sugar commonly associated with bacteria coats) reported to circulate in the bloodstream of people with the disease (Zhang et al., 2009; Zhang et al., 2011), or to the still-unexplained presence of viral gene transcripts in the spinal cord (Douville et al., 2011), or to some other ALS risk factor. To mimic the disease state of microglia, Swarup treated BV-2 mouse microglia cultures with LPS. Then, he used antibodies to pull down TDP-43 and its partners, which he identified with mass spectrometry.

TDP-43 interacted with “hundreds of proteins” in the microglia, including previously known and new partners, Swarup told ARF. The lab is following up on several of these hits, including NF-κB. This transcription factor, commonly made up of p65 and p50 subunits, translocates from the cytoplasm to the nucleus to activate the cell’s response to infection. Swarup confirmed that TDP-43 and p65 co-immunoprecipitate not only in cell culture, but also in spinal cord extracts from TDP-43 model mice (Swarup et al., 2011) and from postmortem ALS, but not from control donors. In these mouse and human spinal cord samples, p65 tended to colocalize with TDP-43 in the nuclei of microglia, astrocytes, and neurons.

Since NF-κB works in cooperation with co-activators, Swarup and Julien suspected TDP-43 might be one of them. To test this, they used a reporter combining a promoter with four NF-κB binding sites and a luciferase gene. They transfected this construct into BV-2 microglia. Adding NF-κB p65 boosted the luciferase signal, and adding TDP-43 as well further increased the luciferase expression. Treating BV-2 cells with a small interfering RNA that reduced TDP-43 expression turned down luciferase activity, supporting the hypothesis. Gel-shift assays confirmed that p65 was more likely to bind to the reporter DNA in the presence of TDP-43. TDP-43 overexpression also boosted the cells’ production of proinflammatory cytokines, which were toxic to neurons in further experiments. Notably, fused in sarcoma (FUS), another RNA-binding protein involved in ALS, has also been shown to be a co-activator of NF-κB (Uranishi et al., 2001).

Are these cell culture results relevant to what happens in human disease? Swarup discovered that TDP-43 mRNA was upregulated by 2.5-fold in ALS spinal cords, compared to control cases. Similarly, NF-κB mRNA was upregulated fourfold. He found a 1.8- and 3.5-fold upregulation of the respective proteins in ALS spinal cord extracts. This upregulation of TDP-43 in ALS, which Swarup told ARF has been confirmed in other laboratories, runs “contrary to popular belief,” he noted. At endstage, ALS pathology often includes a loss of TDP-43 from the nucleus, but Swarup believes that during disease, there is sufficient nuclear TDP-43 to interact with NF-κB.

If that model is correct, then dampening NF-κB activity should ameliorate ALS pathology. Swarup tested this idea in the TDP-43 mice using withaferin A, an inhibitor of NF-κB (Oh et al., 2008) derived from the plant Withania somnifera. The herb, also known as “Indian ginseng,” is used in Ayurvedic medicine; people take it as a natural treatment for rheumatoid arthritis, Swarup said. Treating the mice with withaferin A not only reduced p65 nuclear levels compared to untreated TDP-43 mice, but it also improved the animals’ ability to balance on a rotating rod. Once the researchers stopped the treatment, motor skills rapidly deteriorated to match those of untreated animals.

In NF-κB, “this study opens up a new potential target for therapeutic intervention in ALS,” wrote Michal Schwartz of the Weizmann Institute of Science in Rehovot, Israel, in an e-mail to ARF. Schwartz was not involved with the study. While an NF-κB inhibitor such as withaferin A is unlikely to cure the disease—it cannot undo the degeneration of motor neurons—it might help preserve the remaining neurons and extend lifespan, Swarup said. He is now testing the withaferin A treatment in different ALS mouse models, and at different time points in their disease progression.

The work lends solid support to the idea that inflammation and immunity are key to ALS pathology, said Terrence Town of the Cedars-Sinai Medical Center in Los Angeles, who was not part of the study team (e.g., see ARF related news story). One question that remains, he said: “How much of the pathobiology of ALS is being driven by a TDP-43-p65 interaction in neurons, versus…in microglia and astrocytes?” Swarup, who wonders the same thing, is working to selectively knock out p65 in the different cell types.—Amber Dance.

Reference:
Swarup V, Phaneuf D, Dupré N, Petri S, Strong M, Julien JP. Deregulation of TDP-43 in amyotrophic lateral sclerosis triggers nuclear factor kappaB-mediated pathogenic pathways. J Exp Med. 2011 Nov 21;208(12):2429-47. Epub 2011 Nov 14. Abstract