The identification of autosomal dominant mutations in the GRN gene as a common cause for FTLD

| The discovery that heterozygous and homozygous mutations in the gene encoding progranulin are causally linked to frontotemporal dementia and lysosomal storage disease, respectively, reveals previously unrecognized roles of the progranulin protein in regulating lysosome biogenesis and function. Given the importance of lysosomes in cellular homeostasis, it is not surprising that progranulin deficiency has pleiotropic effects on neural circuit development and maintenance, stress response, innate immunity and ageing. This Progress article reviews recent advances in progranulin biology emphasizing its roles in lysosomal function and brain innate immunity, and outlines future avenues of investigation that may lead to new therapeutic approaches for neurodegeneration. NATURE REVIEWS | NEUROSCIENCE ADVANCE ONLINE PUBLICATION | 1 PROGRESS © 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved. directions (progranulin ↓, granulin ↑). Progranulin also undergoes regulated glycosylation18,19, protecting it from lysosomal enzyme-mediated degradation and adding a layer of complexity to its localization and half-life. Progranulin expression patterns. Progranulin is broadly expressed by many cell types, although the expression is low in muscles or resting endothelium20,21. Within the CNS, GRN mRNA is produced by a wide range of cell types including neurons, microglia, endothelial cells and astrocytes22. Although its putative role is in the lysosome, an amino-terminal secretion signal directs progranulin to secretory vesicles in which it undergoes regulated exocytosis23. This confers both potential autocrine and endocrine roles to progranulin. A case in point is progranulin in microglia, in which basal progranulin levels and secretion are low; upon activation, microglial progranulin expression increases significantly and is likely to have an impact on neuronal function and synaptic density24. Thus, when examining progranulin levels, it is important to consider the source of progranulin and whether it is derived from an intracellular or an extracellular source, or some combination of both. Indeed, results from a large-scale study show poor correlations of progranulin levels in plasma and cerebrospinal fluid (CSF)25. dendrites and axons of cultured hippocampal neurons in both anterograde and retrograde manners, and the secretion of progranulin seems to be modulated by neuronal activity23. Despite these observations, one major challenge in validating the neurotrophic property of progranulin is the lack of evidence for a definitive receptor that interacts with progranulin and initiates signal transduction pathways to promote survival and neurite outgrowth in neurons. Although tumour necrosis factor (TNF) receptors and sortilin have been implicated as progranulin receptors31,32, evidence supporting these proteins as the bona fide receptors that transduce progranulin signals is still elusive. Sortilin traffics extracellular progranulin to the lysosome rather than serving as a signalling hub. Further, the reported interactions between progranulin and TNF receptor were not reproduced by co-immunoprecipitation or surface plasmon resonance33. Thus, it is unclear how progranulin could regulate neuronal survival and neurite outgrowth via the TNF-mediated mechanism. Curiously, recombinant progranulin continues to promote neurite outgrowth in hippocampal neurons lacking sortilin in vitro29, and inhibition of the granulin E–sortilin interaction does not diminish the ability of granulin E to promote neurite outgrowth34. Interestingly, a recent study showed that progranulin binds to ephrin type-A receptor 2 (EPHA2) on the cell surface and that such interaction activates the tyrosine kinase activity of EPHA2 and the downstream kinase AKT35. The interaction between progranulin and EPHA2 promotes capillary morphogenesis and the autoregulation of GRN mRNA, which does not require sortilin. Although these new results provide novel insights into the role of progranulin as a growth factor that promotes angiogenesis, the role of EPHA2 in neurite outgrowth and in neurodegeneration remains to be determined. Progranulin dosage and disease Several lines of evidence indicate that progranulin has gene-dosage-dependent effects on neurodegeneration. As previously indicated, GRN haploinsufficiency causes familial forms of FTD2,3. GRN mutations that are associated with neurodegenerative disease (for a list of mutations, see http://www. molgen.ua.ac.be/FTDmutations/) largely result in premature stop codons and nonsense mediated decay, although a few familial mutations impair secretion or produce full-length protein with loss of a key cysteine residue27,36,37. GRN-mutation carriers have Progranulin as a neurotrophic factor. In addition to its potential roles as a growth factor, structural similarities between progranulin and other secreted proteins have led to the hypothesis that progranulin, and perhaps the cleaved granulins, may function as neurotrophic factors to promote neuronal survival and differentiation. In support of this idea, progranulin levels are reduced in the CSF of patients with FTD with GRN mutations25. Further, some reports found that human recombinant full-length progranulin and granulin E peptide (amino acids 494–594) can seemingly promote survival and neurite outgrowth in rat spinal motor neurons, cortical neurons and hippocampal neurons in vitro26–29, whereas exogenous progranulin rescues the neurite outgrowth phenotypes in Grn−/− neurons. By contrast, FTD-associated pathogenic mutations in progranulin interfere with the neurite outgrowth promoting activity27–29. Although these results suggest that progranulin functions as an exogenous trophic factor to promote neurite outgrowth and neuronal survival, it is interesting to note that knocking down progranulin in hippocampal neurons and SH-SY5Y cells also compromises their survival, neurite outgrowth and synapse formation28,30, further suggesting that progranulin functions via autocrine mechanisms. Indeed, progranulin has been shown to be co-transported with brain-derived neurotrophic factor (BDNF) in Box 1 | Evolutionary dynamics of granulin domains and progranulin genes Progranulins are evolutionary ancient proteins that emerged about 1.5 billion years ago. Granulin-domain-containing proteins are found in unicellular eukaryotes, plants, metazoan animals and vertebrates, and they are among the first extracellular regulatory proteins still used by multicellular animals. Across species, the granulin domains share conserved cysteine residues. However, the number of granulin domains per progranulin is highly variable from species to species. A single granulin domain is found in basal eukaryotes such as Dictyostelium discoideum, whereas Caenorhabditis elegans has 3 granulin repeats, Xenopus tropicalis has 14 and the purple sea urchin has 27 granulin domains. This domain diversity is achieved by the duplication and/or loss of the combination of exons encoding granulin peptides. In some lineages, granulins are also associated with other domains, such as a cysteine protease domain in many plants including Arabidopsis thaliana. No progranulin gene has been identified in