6 February 2011. For people with mutations in progranulin that lead to frontotemporal dementia, the problem is one of production—they have one good progranulin gene, but it cannot make enough of the protein. In the February 2 Journal of Neuroscience, scientists present drugs that alter a cell’s biology so that it revs up progranulin production. The trick, they show, is to neutralize the pH of the cell’s secretory system, triggering an unknown pathway that boosts synthesis of the protein. Although it is far too early to market the drugs as treatment for this type of dementia, the researchers are investigating the therapeutic potential, said senior author Christian Haass, who with first author Anja Capell led the study at the Ludwig-Maximilians-University in Munich, Germany.
Progranulin is neurotrophic (Van Damme et al., 2008), although its exact role in disease is far from clear. A single loss-of-function or missense mutation in progranulin (GRN) causes frontotemporal lobar degeneration with pathological inclusions in the brain containing the RNA-binding protein TDP-43 (FTLD-TDP). These GRN mutations account for some 5 to 10 percent of FTLD cases (reviewed in Gijselinck et al., 2008). Carriers make some progranulin, produced by the one good gene copy, but only at about one-third of normal levels (Sleegers et al., 2009; Finch et al., 2009; Ghidoni et al., 2008). Progranulin may also be a factor in some cases of Alzheimer’s disease (see ARF related news story on Kelley et al., 2010; reviewed in Sleegers et al., 2010).
Capell and Haass suspected that low progranulin levels might be due to excessive degradation rather than a failure to make the protein in sufficient quantities. To discover compounds that might increase its levels, they screened several inhibitors of proteolytic degradation. Using both HeLa human cervical cancer cells and SH-SY5Y human neuroblastoma cells, they looked for an increase in progranulin secreted into the culture media. Only one chemical hit the target: bafilomycin A1. It inhibits the vacuolar ATPase proton pump that acidifies cellular compartments such as the lysosome (V-ATPase reviewed in Forgac, 2007). The results hinted that bafilomycin works by preventing lysosomal degradation of progranulin; however, none of the lysosomal protease inhibitors the researchers tried altered progranulin levels. In addition, they observed, via immunofluorescence, that progranulin did not reside in lysosomes.
The scientists ruled out several other ways bafilomycin might affect progranulin expression. If the drug were boosting GRN transcription, then blocking transcription should neutralize its effects—but that made no difference to the bafilomycin-induced rise in progranulin. Could bafilomycin be reducing autophagy? Autophagy inhibitors made no difference to progranulin levels in bafilomycin-treated cells, either, nor did the scientists observe progranulin in autophagosomes. The ubiquitin protease system, perhaps? Another negative. And bafilomycin had no effect on total protein secreted by the cell, only on progranulin, suggesting it wasn’t simply boosting overall protein output. Capell and Haass concluded that the only place left for bafilomycin to act was on translation of GRN mRNA.
As an inhibitor of V-ATPase, bafilomycin ought to neutralize the pH of cellular compartments. Cells that light up with acid-sensitive LysoSensor go dark in the presence of bafilomycin, for example. Alkalizing agents such as ammonium chloride and cloroquine had the same effect, preventing cellular acidification. They also raised progranulin levels. Together, these experiments suggest that the neutralized pH somehow promotes progranulin translation.
Next, the researchers wondered how long the effects of bafilomycin would last. When Capell removed bafilomycin from the culture medium, the cellular compartments rapidly returned to their normal, acidic pH. But a funny thing happened on the way—progranulin levels stayed high, even after the pH normalized. The increase in progranulin persisted for at least three days after the drug was gone. “That means we trigger something, and what we trigger is independent of the actual pH in the cell…somehow,” Capell said. Exactly what pathway responds to neutral lysosomes by persistently boosting progranulin translation is unknown. The researchers hope that by pursuing this pathway, they will learn more about what progranulin does in the cell.
The work also suggests a potential treatment for FTLD-TDP. “Just changing progranulin levels a little bit could be sufficient to have a protective effect,” Haass said, citing genetic studies that show how small changes in progranulin expression influence a person’s risk for disease. People who have a minor single nucleotide polymorphism (SNP) near the gene for sortilin—which downregulates its binding partner progranulin—have 16 percent less progranulin than people with the major allele (see ARF related news story on Carrasquillo et al., 2010 and Hu et al., 2010). Another polymorphism lying near the gene for transmembrane protein 106B (TMEM106B) (see ARF related news story on Van Deerlin et al., 2010) boosts TMEM106B levels and is linked to higher risk for FTLD, possibly because it alters progranulin expression (Finch et al., 2010).
The work’s significance may extend beyond FTLD and into the realm of Alzheimer’s disease. It seems that a lack of presenilin-1 prevents V-ATPase assembly and throws off cellular pH (see ARF related news story on Lee et al., 2010). And pH-raising treatments block β-secretase, reducing the production of amyloid-β (see ARF related news story on Mitterreiter et al., 2010).
The researchers tested the treatment potential of bafilomycin as well as the alkalizing agent chloroquine, already in common use to prevent and treat malaria, in two models for FTLD. One model was based on mice heterozygous for progranulin. Although these animals show no symptoms of neurodegeneration, they do have reduced progranulin protein. In cortical slice cultures from the mice, bafilomycin boosted progranulin expression. The researchers observed a weaker effect with chloroquine. The progranulin pickup lasted even after the drugs were removed.
To test their drugs on human tissue, the researchers collaborated with Christine Van Broeckhoven at the University of Antwerp in Belgium. She provided lymphoblasts from six people with FTLD-TDP and four healthy controls. Again, the drugs increased progranulin production. Currently, the researchers are testing the same treatment on living mice, Haass said. He is also in talks with clinical researchers to do a small pilot study with people who have progranulin-based FTLD. Initially the scientists will just look for increased progranulin in people’s plasma.
The results from cortical slices and patient lymphoblasts are promising, commented Andrew Bateman of McGill University in Montréal, Canada. But researchers agree that whether chloroquine or similar drugs will alleviate symptoms, especially in people who have already lost neurons, is another question. “My feeling would be that once the damage is done, it may not be reversible,” Bateman speculated. However, he said, even a drug that could slow or stave off the onset of disease in people carrying a progranulin mutation would be “pretty wonderful.”
The work is exciting, said Rosa Rademakers of the Mayo Clinic in Jacksonville, Florida, but she noted there would be challenges in developing a treatment protocol. “To reach a threshold which is enough, but not too much, will be critical,” she said, noting that progranulin levels are high in some cancers (reviewed in Bateman and Bennett, 2009). And, she added, scientists have to look out for other proteins that might be affected by the cellular neutralization.—Amber Dance.
Capell A, Liebscher S, Fellerer K, Brouwers N, Willem M, Lammich S, Gijselinck I, Bittner T, Carlson AM, Sasse F, Kunza B, Steinmetz H, Jansen R, Dormann D, Sleegers K, Cruts M, Herms J, Van Broeckhoven C, Haass C. Rescue of progranulin deficiency associated with frontotemporal lobar degeneration by alkalizing reagents and inhibition of vacuolar ATPase. J Neurosci. 2011 Feb 2;31(5):1885-94. Abstract