Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Targeted manipulation of the sortilin–progranulin axis rescues progranulin haploinsufficiency

Targeted manipulation of the sortilin–progranulin axis rescues progranulin haploinsufficiency Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 6 1467–1478 doi:10.1093/hmg/ddt534 Advance Access published on October 26, 2013 Targeted manipulation of the sortilin – progranulin axis rescues progranulin haploinsufficiency 1 2 1 1 1 Wing C. Lee , Sandra Almeida , Mercedes Prudencio , Thomas R. Caulfield , Yong-Jie Zhang , 1 1 1 1 1 William M. Tay , Peter O. Bauer , Jeannie Chew , Hiroki Sasaguri , Karen R. Jansen-West , 1 1 1 3 1 Tania F. Gendron , Caroline T. Stetler , NiCole Finch , Ian R. Mackenzie , Rosa Rademakers , 2 1, Fen-Biao Gao and Leonard Petrucelli Department of Neuroscience, Mayo Clinic Jacksonville, 4500 San Pablo Rd S, Jacksonville, FL 32224, USA, 2 3 Department of Neurology, University of Massachusetts Medical School, Worcester, MA 01605, USA and Department of Pathology, University of British Columbia, 2211 Wesbrook Mall, Vancouver V6T 2B5, British Columbia, Canada Received August 23, 2013; Revised October 18, 2013; Accepted October 22, 2013 Progranulin (GRN) mutations causing haploinsufficiency are a major cause of frontotemporal lobar degener- ation (FTLD-TDP). Recent discoveries demonstrating sortilin (SORT1) is a neuronal receptor for PGRN endo- cytosis and a determinant of plasma PGRN levels portend the development of enhancers targeting the SORT1 – PGRN axis. We demonstrate the preclinical efficacy of several approaches through which impairing PGRN’s interaction with SORT1 restores extracellular PGRN levels. Our report is the first to demonstrate the efficacy of enhancing PGRN levels in iPSC neurons derived from frontotemporal dementia (FTD) patients with PGRN deficiency. We validate a small molecule preferentially increases extracellular PGRN by reducing SORT1 levels in various mammalian cell lines and patient-derived iPSC neurons and lymphocytes. We further demonstrate that SORT1 antagonists and a small-molecule binder of PGRN , residues critical for 588 – 593 PGRN – SORT1 binding, inhibit SORT1-mediated PGRN endocytosis. Collectively, our data demonstrate that the SORT1 – PGRN axis is a viable target for PGRN-based therapy, particularly in FTD-GRN patients. pathogenic loss-of-function mutations in GRN reported so far INTRODUCTION account for 4–26% of familial FTD cases and 1–12% of spor- adic cases worldwide (4,8–15). Several disease-related missense Progranulin (PGRN), a multi-functional secreted glycoprotein, mutations have also been identified and appear to be associated plays key roles in various biological processes (1,2) and, when de- with reduced PGRN secretion (16). As such, drug discovery ficient, leads to frontotemporal dementia (FTD). Individuals car- efforts aimed at enhancing PGRN levels in patients with FTD rying a null or loss-of-function allele in GRN, the gene encoding with GRN mutations (FTD-GRN) is of great interest to the scien- PGRN, suffer from PGRN haploinsufficiency, which is a major tific community (17,18). Exciting new research by our group and cause of the most common pathological subtype of FTD, fronto- others demonstrates the interaction between PGRN and sortilin temporal lobar degeneration (FTLD-TDP) (3,4). While the (SORT1), a neuronal receptor that mediates extracellular mechanisms linking loss of PGRN function and disease patho- PGRN clearance via an endocytosis mechanism (19), is a prom- genesis remain unclear, evidence from molecular and cellular ising target. For example, while SORT1 is an important regulator studies suggests that decreased levels of extracellular PGRN of PGRN levels (20), PGRN’s neurotrophic and neuroprotective (exPGRN) are relevant to disease pathogenesis. For example, supplementation of exogenous PGRN in culture medium effects are SORT1 independent (5,21), providing assurance that 2/2 rescues neurite outgrowth deficits observed in Grn neuronal the PGRN–SORT1 axis is a viable target for drug discovery cultures (5), facilitates wound healing by promoting the accumu- efforts aimed at identifying exPGRN enhancers. lation of neutrophils, macrophages and fibroblasts (6) and inhibits Herein, we identify and validate several therapeutic neutrophilic inflammation in vivo (7). In addition, the 69 strategies—the development of SORT1 expression suppressors, To whom correspondence should be addressed. Tel: +1 9049532855; Email: petrucelli.leonard@mayo.edu # The Author 2013. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 1468 Human Molecular Genetics, 2014, Vol. 23, No. 6 Figure 1. MPEP decreases SORT1 expression and increases extracellular PGRN in mammalian cell lines. (A and B) M17 cells were treated with control siRNA (siR-Ctrl) or gene-specific SORT1 siRNA (siR-SORT1). (A) Intracellular levels of PGRN, SORT1 and GAPDH were evaluated by western blot at a 48 h time-point. (B) Suppression of SORT1 levels increased extracellular PGRN levels. (C) Chemical name and structure of MPEP. (D–I) Treatment of M17 cells (D and E), HeLa cells (F and G) or NIH3T3 cells (H and I) with MPEP for 24 h dose dependently reduced SORT1 levels (D, F and H) and increased exPGRN levels (E, G and I) at 10 and ∗∗∗ 20 mM. (J) Under the same conditions, MPEP did not affect levels of SORLA, SORCS1 and ubiquitinated proteins in M17 cells. P , 0.001 versus vehicle control, analysis performed by one-way ANOVA followed by Tukey’s post-test. SORT1 antagonists and small-molecule PGRN-specific To evaluate the capacity of small molecules to suppress binders—to reduce SORT1-mediated endocytosis, thereby SORT1 levels, we screened a commercial compound library enhancing exPGRN levels in relevant disease models. and identified a bioactive compound, 1-[2-(2-tert-butyl-5- methylphenoxy)-ethyl]-3-methylpiperidine, termed MPEP (Fig. 1C), that dose dependently reduces SORT1 levels in a mammalian cell lines. MPEP treatment for 24 h reduced RESULTS SORT1 (Fig. 1D) and increased exPGRN up to 3-fold at a Pharmacological suppression of SORT1 expression 20 mM dose (Fig. 1E). A similar effect was observed in HeLa increases extracellular PGRN in mammalian cell lines cells (Fig. 1F and G) and in NIH3T3 cells, a mouse fibroblast line (Fig. 1H and I). To determine the specificity of MPEP, we Recent genetic evidence implicating SORT1 as an important also examined the effect of MPEP on other sortilin-related pro- regulator of GRN levels in serum (20) and the finding that abla- +/2 teins: sortilin-related LDLR class A repeats-containing receptor tion of Sort1 in Grn mice restores Pgrn in brain to normal (SORLA) and sortilin-related VPS10 domain-containing recep- levels (19) support the notion that pharmacological suppression tor 1 (SORCS1) (Fig. 1J). In addition, we examined levels of of SORT1 expression in the brain may be a potential therapeutic ubiquitinated proteins following MPEP treatment to determine approach for upregulating PGRN levels. Prior to investigating whether this drug influences proteasomal degradation of pro- the use of SORT1 protein suppression as a PGRN enhancer, teins (Fig. 1J). That MPEP did not significantly alter the levels we first confirmed that SORT1 downregulation enhances of any of these targets except for SORT1 provides evidence of PGRN levels in culture. To this end, SORT1 expression was its specificity. Moreover, under the same conditions, MPEP reduced by treating M17 cells with SORT1-specific silencing neither increased GRN mRNA nor suppressed SORT1 mRNA RNA (siRNA) (Fig. 1A), which resulted in a significant increase levels indicating the effect is transcription independent (Supple- in exPGRN levels in a time-dependent manner (Fig. 1B). While mentary Material, Fig. S1A). Rather, the significant decrease in previous reports suggested the major mechanism of such SORT1 protein expression as early as 2 h post-MPEP treatment observed rescue effects was due to a reduction of SORT1- suggests MPEP effectively suppresses SORT1 levels by increas- mediated PGRN endocytosis, additional mechanisms, such as ing SORT1 degradation (Supplementary Material, Fig. S1C). induction of PGRN secretion, had not been ruled out. To Finally, treatment of M17 cells with high doses of rPGRN, address this gap in our knowledge, we also evaluated intracellu- which unlike MPEP does not reduce SORT1 expression, rules lar PGRN levels and detected no significant changes (Fig. 1A), out the possibility that MPEP acts via autocrine regulation which indicates the boostin exPGRN levels is likely due to inhib- (Supplementary Material, Fig. S1D). ition of endocytosis. Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 6 1469 Figure 2. MPEP decreases SORT1 expression and increases extracellular PGRN in cellular models of FTD-GRN.(A–C) MPEP decreased SORT1 levels (A) and increased extracellular PGRN levels (B) but not intracellular PGRN (C) in a recently reported PGRN S116X human neuron model differentiated from FTLD patient- specific iPSCs. (D–G) MPEP reduced intracellular SORT1 levels (D and F) and preferentially increased extracellular PGRN levels (E and G) in lymphoblastoid cell lines (LCLs) from two FTD-GRN families, UBC17 (D and E) and UBC15 (F and G). MPEP at 20 mM restored extracellular PGRN to near normal level in mutation +/2 +/+ ∗∗ ∗∗∗ carrier (GRN ) compared with non-carrier control (GRN ). P , 0.01, P , 0.001 versus vehicle control, analysis performed by one-way ANOVA followed by Tukey’s post-test. Pharmacological response to the small-molecule MPEP PGRN S116X neurons yielded GRN mRNA that are 41% rescues PGRN haploinsufficiency in iPSC-neurons and lower than in control and sporadic FTD neurons (22). lymphoblastoid cells derived from FTD patients To evaluate whether MPEP counteracts the PGRN haploin- sufficiency in the above-mentioned model, the cells were To validate the effect of MPEP in an authentic neuronal model treated with the compound at 5, 10 and 20 mM for 24 h. The of the disease, we used recently developed induced pluripotent result was as follows: in iPSC-neurons carrying a heterozygous stem cells (iPSCs) derived from FTD patients with a novel PGRN S116X mutation, MPEP treatment dose dependently heterozygous GRN mutation (22). Because iPSCs are derived reduced SORT1 levels (Fig. 2A) and selectively increased directly from somatic tissues of patients, the differentiated neur- exPGRN up to 5.5-fold (Fig. 2B). Furthermore, MPEP treatment onal cells represent a more physiologically relevant model of the had no effect on intracellular PGRN levels (Fig. 2C). disease and platform for testing therapeutics. Like many other To further evaluate the efficacy of MPEP, we also tested the neurodegenerative disease iPSCs models, including ALS (23), small molecule on lymphoblastoid cell lines (LCLs) derived AD (24) and PD (25), the FTD-GRN iPSCs neurons display from individuals with and without a disease-associated GRN the expected molecular phenotype caused by the inherited muta- mutation. Our investigators initially generated the chosen tion (i.e. PGRN haploinsufficiency). For example, iPSC lines LCLs, which were first used to validate mutations in GRN in derived from FTD patients with the PGRN S116X mutation FTD patients (3) and were shown to reproduce the PGRN were confirmed to have 50% lower GRN mRNA levels than in haploinsufficiency phenotype (3,4). Consistent with the results non-carrier controls. Differentiation of the iPSC lines into from mammalian cell lines and iPSC-neurons, MPEP treatment Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 1470 Human Molecular Genetics, 2014, Vol. 23, No. 6 Figure 3. Elastase-mediated removal of C-terminal motif of PGRN blocks PGRN endocytosis by SORT1. (A) A synthetic PGRN peptide was enzymatically 574– 593 cleavedbyrecombinantelastase (EL)in a time-dependentmanner as analyzedbyMS-MALDIanalysis. ThePGRN peptidewitha molecularweightof2430 Da 574– 593 was processed into a product with 1806Da indicating the last five residues were removed by EL as shown schematically in (B). (C) WT or EL-site-mutated A588G-rPGRN protein in cell culture media was set up to react with EL in vitro. After reaction, the media was used as input to carry out a cellular endocytosis assay in M17 cells overexpressing SORT1. PGRN endocytosed in cells was detected in cell lysate after treatment (uptake). Unlike WT rPGRN, A588G-rPGRN was more resistant to EL activity as shown by the amount of residual full-length protein in the medium (input), which was endocytosed by cells to a similar extent compared with no EL control. Immunocytofluorescence analysis showed that only full-length rPGRN (FL) (E) but not the carboxyl-terminal-truncated SORT1 SORT1 PGRN (F) was significantly endocytosed by M17 cells. (D) The untreated M17 control was included. PGRN and SORT1 were labeled in red and 1-588 green, respectively. +/+ decreased SORT1 (Fig. 2D and F) and preferentially increased with the vehicle-treated GRN -LCL, indicating a tight exPGRN (Fig. 2E and G) in the LCLs from two FTD-GRN- correlation between SORT1 reduction and exPGRN induction. affected families: one termed UBC17, which carries a p.C31LfsX34 mutation, and one termed UBC15, which carries The PGRN C-terminal motif, PGRN , is essential (589 –593) a p.R418X mutation (3,4). for SORT1-mediated endocytosis In family UBC17 (Fig. 2D and E), 20 mM of MPEP complete- ly restored exPGRN levels in the heterozygous mutation carrier To identify small molecules that interfere with the SORT1– +/2 (GRN -LCL) to a level comparable with the homozygous PGRN interaction via a receptor antagonist approach, first we +/+ non-carrier (GRN -LCL) (Fig. 2E). While MPEP restored aimed to locate the specific PGRN region essential for SORT1 exPGRN levels in the UBC15 family to 80% of that of non- binding and endocytosis. Our group and others have shown +/+ carrier GRN -LCL levels (Fig. 2G), such exPGRN levels rep- C-terminal, and not N-terminal, tagging of recombinant human resent a 3-fold increase compared with the control. Higher PGRN protein completely blocks intracellular uptake (26) (Sup- SORT1 levels were also observed in the corresponding plementary Material, Fig. S2A). In our current study, we identi- UBC17 and UBC15 heterozygous LCL (Fig. 2D and F), which fied a crucial neutrophil elastase (NE) cleavage site between may account for the more severe exPGRN reduction beyond residues Ala-588 and Leu-589 of PGRN by in vitro digestion of the haploinsufficiency caused by the null mutation (20). At a synthetic PGRN peptide by NE, followed by 574–593 20 mM, however, MPEP reduced SORT1 levels and increased MALDI-MS analysis (Fig. 3A and B). In addition, we found dis- +/2 exPGRN in the GRN -LCL model to levels comparable ruption of the A588 elastase cleavage site, through the Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 6 1471 Figure 4. SORT1 ligands competitively inhibit PGRN endocytosis. (A–C) Structure for NTS (A), native human PGRN peptide (B) and mouse Pgrn 588– 593 584 – 589 peptide (C) docked with SORT1 and the corresponding docking scores are shown. (A) The strongly interacting core of NTS peptide, -PYIL, which is resolved in X-ray structure, is showninbolder,darkgreen.SORT1 residueswithin4 Aare renderedingray. SecondarystructureforSORT1isshownas cartoonribbons.The carboxylate ofterminalLeu is visible in the electrostatic surface renderingof the binding pocket.As with NTS,(B) the humanPGRN (-ALRQLL) or (C) the mousePgrn 588 –593 584– (-VPRPLL) is shown docked with the SORT1-binding pocket (electrostatic surface rendered) with the residues ,4A indicated. Carboxylate of Leu is in the iden- TM tical position as NTS. (D and E) A quantitative cell-based assay has been established to measure PGRN endocytosis by SORT1. DyLight 594-labeled rPGRN was SORT1 endocytosed dose dependently in COS-1 cells. (D) Images were captured by BD-pathway system. (E) Quantitative cellular endocytosis of DyL-rPGRN was SORT1 measured by fluorescence signal from treated cells. (F) NTS at 0.1, 1 and 5 mM dose dependently inhibited PGRN endocytosis. Untreated COS-1 cells were included as negative control (2ve). (G) Quantification of signal from (F). (H and I) SORT1 ligands at 10 mM, NTS, human PGRN peptide and mouse 588 – 593 ∗∗∗ PGRN peptide competitively inhibited DyL-rPGRN endocytosis as compared with vehicle control, respectively. P , 0.001 versus vehicle control, analysis 584 – 589 performed either by (G) one-way ANOVA followed by Tukey’s post-test or (I) by unpaired student t-tests. A588G introduction of an Ala-to-Gly mutation, protected PGRN usually discontinuous, which further reduces the probability of from elastase processing, hence preserving the SORT1-inter- identifying effective disruptors of protein–protein interactions. A588G action motif and PGRN endocytosis (Fig. 3C). To demon- To address these challenges in a time-efficient manner, we strate that the PGRN residues 588–593 are critical for employed the use of computer-assisted modeling. As our SORT1-mediated PGRN endocytosis, we produced recombinant search for a physiological ligand derived from the previously truncated PGRN 1–588 (rPGRN ). The truncated protein implicated carboxyl-terminus (C-terminus) of PGRN(26) shed 1-588 failed to be endocytosed in M17 cells overexpressing SORT1 new light on the importance of PGRN residues 588–593, we (Fig. 3F), unlike the full-length rPGRN control (Fig. 3E), which generated models of SORT1 bound to human PGRN , 588 – 593 presented an endolysosomal localization (Supplementary Mater- neurotensin (NTS), which is a high-affinity SORT1 ligand ial, Fig. S2B), thereby validating the PGRN region as an and mouse Pgrn (-ALRQLL, -ELYENKPRRPYIL and 589–593 584 – 589 essential motif for SORT1-mediated PGRN endocytosis. -VPRPLL, respectively). We used crystal structure data of SORT1 complexed with NTS(27) to model peptide sequences into the binding cleft of Computer-assisted modeling confirms ligands SORT1 using NTS fragment -PYIL as a template, to deter- 10-13 share same SORT1-binding pocket mine a GRID for docking, and then to optimize the interactions through energy minimization (Fig. 4A–C). The model of the sub- Identification of small molecules to target protein–protein inter- strate neurotensin (-PYIL fragment) obtains a docked position action interfaces is considered extremely challenging owing to relative to the crystallographic structure 3F6 K within 0.8 A the size disparity between small molecules and large contact sur- RMSD; NTS residues proline and tyrosine fill a largely faces on proteins. In addition, protein contact surfaces are Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 1472 Human Molecular Genetics, 2014, Vol. 23, No. 6 hydrophobic cleft between the flanking clusters of positive mimetics, may serve as potential exPGRN enhancers through in- charge, with the NTS Leu side chain embedded in the binding hibition of extracellular clearance. site forming numerous hydrophobic interactions with Phe273, Phe281, Ile294 and Ile320. For each peptide that docked, the carb- Use of small-molecule and antibody binders of the oxylate of the terminal Leu formed strong interactions with PGRN motif inhibit SORT1-mediated PGRN 588–593 SORT1, as shown in Figure 4A–C. The peptides docked with endocytosis the SORT1-receptor gain stabilization via charge complementar- ity from electrostatic interactions between the carboxylate anions Given that restoring exPGRN levels using SORT1 antagonists and Arg292 cations, which is stabilized by Ser283 and Ser319 at can potentially trigger off-target effects related to NTS (29), carbonyl interactions and bridged by H-bonds. Additional favor- lipoprotein lipase (30) and LDL-receptor-associated protein able close interactions between the peptides and SORT1 are com- (31) function and cause undesirable clinical side-effects, modu- prised of hydrophobic and van der Waals interactions and a series lating the PGRN–SORT1 interaction via PGRN-specific of hydrogenbondpartnerswithinthe bindingpocket(Fig.4A–C). binders may be a promising approach. To identify small mole- The docking results show NTS with an average docking score of cules that can selectively bind to the PGRN C-terminal motif 27.86 kcal/mol (Fig. 4A), human PGRN of 29.041 kcal/ to inhibit PGRN–SORT1 interactions, we screened 4800 com- 588–593 mol (Fig. 4B) and mouse Pgrn of 26.85 kcal/mol (Fig. pounds from a commercial library against a custom-synthesized 584– 589 4C). As such, the human PGRN has the highest binding PGRN peptide using a 384-well format, biochemical 588–593 588 – 593 affinity followedbyNTS and thenthe mouse Pgrn . Higher- binding assay that utilizes resonant waveguide grating biosensor 584–589 affinity binding of human PGRN peptide to SORT1 com- detection (Fig. 5A) (Supplementary Material, Method). We 588– 593 pared with that of NTS/SORT1 is derived from strong interaction identified a small molecule, 4-[2-(3-bromophenyl)vinyl]-6- pairs between the terminal few residues and key side chains and (trifluoromethyl)-2(1H)-pyrimidinone termed BVFP (Fig. 5B), backbone interactions from SORT1. Additionally, the mouse that binds to the PGRN peptide at a low micromolar con- 588 – 593 mPgrn /SORT1 binding suffers a loss in interaction centration (K ¼ 20 mM, Fig. 5C). Upon mixing and titrating 584–589 d from a distinct repositioning of mPgrn residues, -VPR, BVFP with the PGRN peptide, we detected a shift in 584–586 588 – 593 without significant SORT1 contacts. The details of the docking the ultraviolet(UV)-absorption spectra of BVFP and distinct models and decomposition of individual interactions between changes of absorption intensities, which in addition to the pres- SORT1 and NTS, human PGRN and mouse Pgrn ence of an isosbestic point at 313 nm, validated the BVFP and 588 –593 584– 589 are described in the Supplementary Material. PGRN peptide interaction (Fig. 5D). 588 – 593 To determine whether BVFP, when bound to PGRN, disrupts full-length PGRN and SORT1 binding, we devised a SORT1- SORT1 ligands competitively inhibit full-length PGRN dependent rPGRN precipitation assay utilizing the Meso-Scale endocytosis Discovery (MSD) system (Supplementary Material, Fig. S4A Given our computer-assisted modeling confirmed and provided and B). Immobilized SORT1 generated a specific binding new information of how NTS (28,29) and PGRN share a signal from the amino-terminal-tagged PGRN (N-rP) but not 588 – 593 similar SORT1-binding site (19), we sought to determine from the C-terminal-tagged PGRN, validating once again the whether the use of these ligands as antagonists could be an effect- PGRN–SORT1 binding is C-terminal dependent. A dose- ive strategy to inhibit PGRN endocytosis in cell culture. Using dependent inhibition of rPGRN binding to immobilized SORT1 TM DyLight -594 fluorescence-labeled rPGRN (DyL-rPGRN) was observed in the presence of 20 and 200 mM ofBVFP, respect- SORT1 and SORT1-expressing COS-1 (COS-1 ) cells, we estab- ively, proving the small molecule’s capacity to disrupt PGRN– lished a cell-based assay to quantitatively analyze PGRN endo- SORT1 binding (Supplementary Material, Fig. S4C). To TM cytosis by measuring DyLight -594 emission from includeahigher-affinity PGRN motif binder as a positive 588–593 endocytosed DyL-rPGRN upon treatment. Addition of titrated control, an antibody targeting the motif, PGRN-CT antibody, was SORT1 amounts of DyL-rPGRN (0 to 50 nM) to COS-1 cells pro- made for validating the approach (Fig. 5F). Using our quantitative duced a linear, non-saturated endocytosis response (Fig. 4D and endocytosis assay (Fig. 5E and G), we found that BVFP inhibited E). As expected, co-treatment of DyL-rPGRN with NTS (i.e. 0.5, DyL-rPGRN endocytosis by 23% when treated at 5 mM, com- 1 and 5 mM) dose dependently inhibited PGRN endocytosis pared with 43% inhibition by PGRN-CT antibody when treated (Fig. 4F and G). At the same concentrations, NTS, as well as at 80 nM (Fig. 5E and G). Note that a negative control antibody human PGRN and mouse PGRN , respectively, targeting GRN-A had no effect on DyL-rPGRN endocytosis. 588 – 593 584 – 589 inhibited full-length PGRN endocytosis by 90, 93 and 73% To show that the effects of BVFP and the PGRN-CT antibody (Fig. 4H and I). The fact that the mouse ortholog peptide of are SORT1 dependent, the PGRN endocytosis experiment was PGRN also significantly inhibited full-length PGRN repeated in human embryonic stem cell lines (hESCs) with and 588 – 593 SORT1 endocytosis by human SORT1 in COS-1 cells, albeit to a without SORT1 expression established by using the transcription lesser extent, suggests that the PGRN-CT motif is evolutionarily activator-like nucleases genome-editing system (TALENs) (32). conserved. In a protein co-immunoprecipitation assay, we First, we found SORT1 was abundantly expressed in the wild- SORT1 further validated that human PGRN peptide, NTS and type hESCs (Fig. 5H). Similar to the results from COS-1 588 – 593 mouse Pgrn peptide can each inhibit the physical inter- cells, BVFP and PGRN-CT antibody inhibited endocytosis of 584 – 589 action between full-length PGRN and SORT1 (Supplementary DyL-rPGRN by 22 and 39%, respectively, under the same experi- Material, Fig. S3). These results support the notion that the mental conditions in wild-type hESCs (Fig. 5I left). Treatment of KO development of SORT1 antagonists, such as stabilized forms DyL-rPGRN to the isogenic hESCs-SORT1 line, which has an of chemically modified human PGRN peptide or NTS identical genetic background and epigenetic state compared with 588 –593 Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 6 1473 Figure 5. Small-molecule and antibody binders of the PGRN motif inhibit SORT1-mediated PGRN endocytosis. (A) Schematic diagram illustrating the de- 588 – 593 tection mechanism of Epic biochemical binding assay. When screening for compound binders of PGRN peptide (target), the specific binding events are 588 – 593 detected by changes in reflected resonant wavelength before (l) and after (l ) addition of a potential compound binder. (B) Structure and chemical name of BVFP, an identified compound that binds the PGRN peptide. (C) Saturated binding curve of BVFP to PGRN peptide (n ¼ 4). (D) Wavelength red- (588 – 593) (588 – 593) shifting of UV-absorption spectra of BVFP (20 mM) upon titration with the PGRN peptide. The interaction between BVFP and the peptide is represented by (588 – 593) distinct changes in the absorption intensities (black arrows). The presence of an isosbestic point at 313 nm (inset red arrow) also confirmed the interactions. (E) The SORT1 PGRN binders, BVFP (5 mM) and PGRN-CT antibody (80 nM) inhibited rPGRN endocytosis as tested by the quantitative cell-based assay in COS-1 (588 – 593) cells. Both binders were pre-incubated with rPGRN for an hour and then added to the cells for an hour to allow endocytosis. A GRN-A specific antibody (80 nM) 1-593 was used as a negative control. (F) Vehicle control (2ve), full-length protein (rPGRN ) or truncated protein (rPGRN ) was analyzed by western blot using 1-593 1-588 PCDGF or PGRN-CT antibody for detection. The PGRN-CT antibody was confirmed to be PGRN dependent. (G) Fluorescence signal quantification of (588 – 593) KO (E). (H) Western blot analysis confirmed the absence of SORT1 protein in the SORT1 -hESCs. (I) BVFP and the PGRN-CT antibody inhibited PGRN endocytosis KO ∗ ∗∗ ∗∗∗ in WT-hESCs but not in SORT1 -hESCs indicating that the effect was SORT1 dependent. P , 0.05, P , 0.01, P , 0.001 versus vehicle control, analysis performed by unpaired student t-tests. the parent hESC-WT line, generated a significantly lower signal pathology (i.e. FTLD-TDP), therapeutically restoring PGRN compared with the WT-hESCs, indicating impaired PGRN endo- levels may be a promising therapeutic strategy. In fact, evi- cytosis in the absence of SORT1 (Fig. 5H). Both BVFP and dence from numerous studies suggests that PGRN acts as a PGRN-CT antibody had no effect on DyL-rPGRN endocytosis protective neurotrophic factor by regulating neuronal survival KO in the hESCs-SORT1 line (Fig. 5I right), confirming that and neurite growth in cortical/motor neurons, immortalized their effect on PGRN endocytosis is SORT1 dependent. cell lines and zebra fish (33–35) and that PGRN is protective against insults induced by TDP-43 (36). Given that there is currently no feasible way to pharmacologically manipulate PGRN levels in the brain and that recombinant PGRN is DISCUSSION too large to cross the blood–brain barrier (BBB) for protein replacement, the development of bio-available, BBB Given that partial loss of PGRN, owing to mutations in the permeable PGRN enhancers will be a valuable tool to PGRN gene (GRN), is causative of FTLD with TDP-43 Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 1474 Human Molecular Genetics, 2014, Vol. 23, No. 6 determine whether therapeutically modulating or increasing PGRN levels can alleviate the pathogenesis associated with FTD-GRN. While new strategies designed to upregulate PGRN levels have emerged, challenges remain in terms of limiting off-target effects. For example, Cenik et al. recently identified the histone deacetylase (HDAC) inhibitor suberoylanilide hydroxamic acid as an enhancer of GRN expression (18). Christian Haass’ group also has recently demonstrated four independent and highly se- lective inhibitors of vacuolar ATPase (V-ATPase) significantly elevate intracellular and secreted PGRN (17). However, because HDAC and V-ATPase inhibitors can potentially affect a wide- range of genes at transcriptional and post-translational levels, re- spectively, thereby increasing the likelihood that the inhibitors will also trigger undesirable, clinical side-effects and liabilities, the development of a strategy to improve target specificity may hold more promise. Here, we demonstrate the preclinical efficacy of several approaches through which impairing PGRN’s interaction with SORT1, a neuronal receptor that mediates PGRN endocytosis and degradation, restores extracellular PGRN levels in FTD patient-derived iPSC-neurons and lymphocytes (Fig. 6). To the best of our knowledge, our report is the first to demonstrate the efficacy of enhancing PGRN levels in iPSC-neurons derived Figure 6. Schematic diagram summarizing the strategies applied to inhibit from FTD patients with PGRN deficiency. We also validate SORT1-mediated endocytosis in the current study. (A) Under normal conditions, that the bioactive compound MPEP preferentially increases extracellular PGRN interacts with the b-propeller tunnel structure of SORT1 exPGRN levels by suppressing or reducing intracellular SORT1 using its C-terminal end binding motif as shown in red color. SORT1 facilitates levels without affecting sortilin-related family members endocytosis of exPGRN and directs it to the endolysosomal pathway for degrad- SORLA and SORCS1. To understand whether MPEP might in- ation. (B) High-affinity SORT1 ligands such as NTS or the PGRN (588 – 593) peptide competitively limits the access of exPGRN to SORT1-binding sites, crease exPGRN through protease(s) inhibition, we performed a thereby inhibiting PGRN endocytosis. (C) To improve target specificity, we database search using the similarity ensemble approach tool have also identified a small-molecule binder, BVFP, targeting the PGRN (588 – (37). No protease inhibitors in the database were found to have motif that is essential for PGRN–SORT1 interaction. We demonstrated 593) significant structural similarity toMPEP, suggesting that mechan- that pretreatment of BVFP to rPGRN significantly reduced the amount of rPGRN captured by SORT1 in vitro and inhibited SORT1-mediated rPGRN isms other than protease inhibition account for the SORT1- endocytosis. (D) Suppressors of SORT1 expression, such as MPEP, reduce targeted exPGRN upregulation by MPEP. SORT1-mediated endocytosis, thereby increasing extracellular PGRN levels. Given SORT1 s role in regulating exPGRN levels, we further The above-mentioned strategies, used alone or in combination with others, are demonstrate that SORT1 antagonists and PGRN binders 588 – 593 potential avenues for discovery of SORT1-dependent PGRN enhancers for the inhibit SORT1-mediated PGRN endocytosis. Our novel cell treatment of FTD-GRN. culture data reveal the PGRN peptide is the critical 588 – 593 region for SORT1–PGRN binding; the ‘hot spot’ is confirmed by our computational model of the PGRN–SORT1 interaction. serving as a regulatory mechanism to boost extracellular As mentioned, the feasibility of finding a small molecule that PGRN and granulins levels that are involved in the inflammation specifically targets protein–protein interactions is very much signaling cascade. case dependent(38), but identifying the crucial site improved While we demonstrate BVFP prevents SORT1-medited endo- our ability to hone in on high-affinity, small-molecule binders cytosis and increases exPGRN levels, an alternate supplemen- targeting either binding partner. To reduce undesirable effects tary approach would be to identify small molecules targeting on other SORT1 ligands, we concentrated our efforts on a com- SORT1, specifically at the Arg-292-, Ser-283- and Ser-319- pound library screen and biochemical binding assay that allowed binding pockets. A structure-based virtual high-throughput usto identify a small molecule, BVFP,that targets the C-terminal screening approach (vHTS) using the available NTS-complexed PGRN motif. As the motif is located on a linker region SORT1 crystallographic data could be employed to accelerate 588 – 593 and not on the functional granulin regions, binding of BVFP to drug discovery (27). The vHTS hits can be validated by a func- the extreme C-terminal motif is unlikely to significantly impair tional assay such as the quantitative endocytosis assay described PGRN function (1). Moreover, the discovery of a potential NE in the current study. cleavage site at the C-terminal Ala-588 residue of PGRN sug- We also evaluated the use of high-affinity SORT1 ligands as gests that a surge in extracellular NE levels would increase competitive inhibitors to effectively block exPGRN’s access C-terminal truncation at Ala-588 leading to accumulation of to SORT1 s binding site. Our computer-assisted modeling the PGRN protein and granulins generated by increasing sheds new light on why we observed a dramatic reduction in 1-588 NE proteolytic activity. As NE is an innate immune response PGRN endocytosis when ligands, such as NTS and human mediator that triggers downstream molecular events required PGRN , were present in vitro: the binding pocket and 588 – 593 for the inflammation process (39), the Ala-588 site is likely protein–protein interactions of NTS and PGRN’s C-terminal Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 6 1475 binding motif significantly overlap, hence explaining the com- given neutralizing PGRN to counteract its possible tumorigenic petitive inhibition effect. Although both NTS and full-length effect (46) is a proposed cancer treatment (47,48). In addressing PGRN induce endocytosis upon binding to SORT1, the two such concerns, and further exploring the relationship between ligands have been reported to traffic to different subcellular loca- TDP-43, SORT1 and PGRN (49), the clinical development of lizations, namely endolysosomal compartment (19) and the PGRN-based therapy for FTD-GRN patients will and should trans-Golgi network (40). In both cases, SORT1 is dissociated continue. from the bound ligands in early endosomes and then recycled back to the cell surface to drive the endocytosis process. While MATERIALS AND METHODS the use of a SORT1 antagonist approach is enticing owing to af- finity for and efficacy in blocking PGRN uptake, this approach is Recombinant PGRN protein, plasmids and antibodies also more likely to trigger off-target effects by disrupting Wild-type or mutant human recombinant PGRN protein binding between SORT1 and other ligands that occupy a (rPGRN) with a 6-histidine tag on the amino-terminus (N-rP) similar binding pocket. or the carboxyl-terminus (C-rP) was purified from culture As with many other ligands such as prosaposin (41), acid media of stable HEK293 cells secreting the corresponding sphingomyelinase (42) and cathepsins D and H (43), evidence protein. Purification procedures were the same as previously suggests SORT1 is a sorting receptor of PGRN for lysosome de- published (5). All the plasmids used in this study are listed as livery and degradation (19). Our group (5) and another (21) also follows: pCDNA4-PGRN(1–593) (N-terminal 6His tag), showed that SORT1 is not essential for stimulating neurite out- pCDNA4-PGRN(1–588) (N-terminal 6His tag), pCMV- growth in cultured neurons, suggesting that SORT1 might not be SORT1 (OriGene), pCDNA6-SORT1 (C-terminal Flag tag) involved in the PGRN-mediated signaling for neurotrophic and pEGFP-N1 (Clontech). All the antibodies used in this function. Nevertheless, inhibition of PGRN endocytosis might study are listed as follows: PCDGF (or Grn-E specific) antibody influence PGRN’s intracellular functions or unidentified intra- (1:1000 for western blot) (Invitrogen), human PGRN monoclo- cellular signaling pathways, especially in cells that rely on its nal antibody (1:200 for immunofluorescence) (R&D Systems), uptake from the extracellular space or neighbor cells. It is Grn-A antibody (80 nM for blocking PGRN endocytosis) known that complete deficiency in PGRN is a cause of clinical (Novus; 26320002), PGRN-CT antibody (1:1000 for western neuronal ceroid lipofuscinosis (NCL) (44), a rare lysosomal blot; 80 nM for blocking PGRN endocytosis) (21st Century; af- storage disease characterized by intraneuronal accumulation of finity purified rabbit polyclonal immunized with PGRN 574 – 593 autofluorescence lipofuscins. Based on this finding, it is reason- peptide), sortilin antibody (1:3000 for western blot; 1:1000 for able to speculate that PGRN has a lysosomal function related to immunofluorescence) (Abcam; ab16640), SORLA antibody turnover of lipofuscin-related substrates. While blocking PGRN (1:200) (R&D Systems; AF5699), SORCS1 antibody (1:1000) endocytosis will reduce level of PGRN in lysosomes, intracellu- (R&D Systems; AF3457), ubiquitin antibody (1:1000) lar PGRN provided from the biosynthetic pathway to lysosomes (DAKO; Z0458), EEA1 antibody (1:100 for immunofluores- is expected to be enough to maintain normal lysosomal function. cence) (BD Biosciences; 610457), giantin antibody (1:1000 At least for most lysosomal storage disorders, clinical onset is for immunofluorescence) (Abcam; ab24586), Flag M2 monoclo- often associated with.90% loss of activity of the corresponding nal antibody (1:500 for immunoprecipitation) (Sigma), GAPDH enzyme (45). However, further studies will be required to under- antibody (1:10 000 for western blot), and b-actin antibody stand whether abrogation of PGRN endocytosis can cause NCL (1:10 000 for western blot). or other deleterious biological consequences. Given the challenges associated with the development of ther- apies and clinical trials for patients suffering from neurodegen- Western blot analyses erative diseases, the need is urgent and the time is now for the PGRN, SORT1, GFP, GAPDH and b-actin proteins were medical community to rally behind a promising target. The analyzed using 10% Tris–glycine polyacrylamide gel by increased excitement surrounding PGRN-based therapies is SDS–PAGE. After electrophoresis, proteins on gel were trans- warranted: the specific patient population that could benefit is blotted to PVDF membranes followed by standard immunoblot- easily identifiable thanks to genetic testing. Moreover, the feasi- ting protocols. bility of restoring PGRN to normal levels has been shown by our group and others. The fact that genetic ablation of SORT1 restores and normalizes PGRN levels in brain uniquely positions SORT1 Analysis of PGRN endocytosis in M17 cells the SORT1–PGRN axis as an ideal target for PGRN-based therapy in FTD-GRN (19). Herein, we validated multiple strat- M17 cells seeded in 12-well culture plates (western blot) or on egies to target the SORT1–PGRN axis withthe aim ofenhancing glass coverslips in 24-well culture plate (immunofluorescence) PGRN levels. We show that SORT1 antagonists and PGRN were transfected with pEGFP-N1 (Clontech) or pCMV-SORT1 588 – binders inhibit SORT1-mediated PGRN endocytosis and a (OriGene) vectors. Next day, the cells were changed into serum- SORT1 suppressor restores exPGRN in patient-derived cell free medium 1 h before addition of rPGRN (500 ng/ml) and models. In addition, both small molecules identified in this treated for 60 min. The cells were then put on ice and washed study, namely MPEP and BVFP, are in compliance with the with cold PBS. For western blot, cells were lysed in cold Lipinski’s Rule of five to predict drug-likeness properties, M-Per protein extraction buffer followed by standard procedures which should facilitate in vivo studies in the near future. as mentioned. For immunofluorescence analyses, cells were Further studies to understand the dosage effect of fixed in 4% paraformaldehyde solution followed by PBS PGRN-enhancing reagents to treat FTD-GRN will be essential, washes. The cells were then incubated in blocking buffer Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 1476 Human Molecular Genetics, 2014, Vol. 23, No. 6 (PBS, 5% goat serum, 0.1% saponin) for 1 h at room temperature Determination of UV-absorption spectrum of BVFP followed by incubation in primary antibody overnight at 48C. Stock solutions of compound BVFP (20 mM) and PGRN (588 – 593) The cells were incubated with secondary antibodies (1:1000) peptide (12.96 mM) were prepared in DMSO and water, respect- for 1 h at room temperature. The coverslips were then ively. For the UV absorption experiment, both components were mounted onto slides and waited overnight before analysis. diluted to give final concentrations of 20 mM BVFP and 324 mM Image analysis was performed by Zeiss LSM 510 META con- peptide. The interaction between BVFP and the PGRN (588 – 593) focal microscope using 60× magnification setting. peptide was monitored by scanning over the UV absorbance range of 200–800 nm on a Cary 3 Bio UV–visible spectropho- tometer (Varian), as a small amount of KKGRN peptide (0.1 MALDI-MS analysis of NE-processed PGRN (584–593) equivalence/1 ml per addition) was titrated into the 20 mM peptide BVFP sample. In addition to baseline correction with DMSO, An enzymatic reaction was set up with 7 mg of custom- a control titration experiment using water in place of the synthesized PGRN peptide (Mayo peptide synthesis (584 – 593) PGRN peptide was performed under identical condi- (588 – 593) core) and 20 mgof NEina25 ml reaction at 378C. Every tions in order to correct for potential spectral changes owing to 10 min, 4 ml of the mixture was collected and diluted 10 times sample dilution. in a solvent mixture (50% acetonitrile:50% water:0.1% acetic acid) until 1 h had passed. For MALDI-MS analysis, 1 mlof the diluted reaction mixture was dried on a MS gold chip and then layered with a concentrated sinipinic acid matrix on top iPSC neuronal culture for crystallization. Afterwards, the samples were analyzed by the Bio-Rad MALDI-MS system. An induced pluripotent stem cell (iPSC) line from an FTD patient carrying a GRN nonsense mutation (p.S116X) was differ- entiated into neurons as described previously (22). Two-week- Quantitative endocytosis cell-based assay old neurons were incubated with 5–20 mM MPEP or DMSO COS-1 cells seeded on a 96-well black plate 1 day before were (in fresh culture medium) for 24 h. The medium was then col- transfected with pCMV-SORT1 vector. After 24 h, cells were lected, and cells were lysed with 1% NP-40 lysis buffer. treated with fluorescence-tagged rPGRN (DyL-rPGRN) pre- PGRN levels in total cell lysates and culture medium were deter- TM labeled by DyLight 594 antibody labeling kit (Thermo Scien- mined with an ELISA kit (Alexis Biochemicals, San Diego, CA, tific). The DyL-rPGRN was diluted in OptiMEM to tested USA) according to the manufacturer’s instructions. concentrations and incubated with the cells for 1 h for endocyto- sis. The cells were then washed with cold PBS and then fixed by 4% paraformaldehyde. After washing twice, each well was filled with PBS prior scanning. The total PGRN fluorescent signal MSD system-based immunoassay to analyze human PGRN from the cells was obtained by a plate reader with 593/618 nm The assay was performed according to manufacturer’s instruc- (Ex/Em) settings. The endocytosis signal was normalized by tions. A goat anti-PGRN antibody (R&D Systems) was used as the total nuclei signal obtained by staining with a Hoechst capture antibody. PGRN standards (from 0.5 to 50 ng/ml) 33342 dye. Baseline endocytosis signal was defined as a signal were from purified N-6His-rPGRN protein. For detection, a from medium-only-treated cells, whereas the 100% endocytosis 1:1 ratio of an anti-PCDGF antibody (Invitrogen) and a level was set as the signal obtained from cells treated with 50 nM SULFO-TAG-labeled anti-rabbit IgG antibody were applied. DyL-rPGRN. For testing SORT1 ligands, DyL-rPGRN and the The MSD Sector Imager 2400 was used to read assay signal. peptide were added simultaneously to the cells. For testing of PGRN binders, the binder was pre-incubated with DyL-rPGRN for 1 h before adding to the cells. For qualitative analysis, images from each well were captured by BD Pathway 855 system using a A MSD-based SORT1–PGRN interaction assay 20× magnification setting. M17 cells were transfected with pCMV-SORT1-Flag for 48 h. Then, the cell lysate was harvested by the same method used PGRN co-immunoprecipitation in the co-immunoprecipitation assay. Expression of SORT1- HEK293T cells were transfected empty vector or pCMV- Flag protein was confirmed by western blot. MSD plate was SORT1-Flag vector for 48 h. Then, cells were lysed by using first coated with the Flag M2 antibody (Sigma). Following an Co-IP buffer. Pre-incubation was performed with 300 mgof overnight incubation at 48C, the plate was blocked for 2 h and lysate protein mixed with 20 mM NTS, human PGRN then washed prior to addition of 5 mg/well of lysate protein con- (588 – 593) peptide or mouse PGRN peptide for 1 h. Then, rPGRN taining SORT1-Flag protein. After overnight incubation at 48C, (584 – 589) (100 nM) was added into the protein G beads pre-cleared super- it was then washed and added with 200 nM of either N-6His- or natant and mixed for 30 min. Next, anti-Flag M2 agarose C-6His-tagged rPGRN pre-incubated with either vehicle, (Sigma) was added and mixed for another 4 h. The agarose 20 mM or 200 mM of BVFP, for 30 min at room temperature. was collected by centrifugation at 1000 g for 3 min and Then plate was then incubated for 2 h, washed and incubated washed with Co-IP buffer six times. Captured protein was with a mixture of PCDGF antibody and a SULFO-TAG-labeled eluted from the beads using loading buffer and analyzed by anti-rabbit IgG antibody for 2 h. After final wash, the plate was western blot. read by the MSD Sector Imager 2400. Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 6 1477 4. Cruts, M., Gijselinck, I., van der Zee, J., Engelborghs, S., Wils, H., Pirici, D., Quantitative reverse transcription–polymerase chain Rademakers, R., Vandenberghe, R., Dermaut, B., Martin, J.J. et al. (2006) reaction Null mutations in progranulin cause ubiquitin-positive frontotemporal dementia linked to chromosome 17q21. Nature, 442, 920–924. Quantitative PCR (qPCR) was performed using an HT7900 ana- 5. Gass, J., Lee, W.C., Cook, C., Finch, N., Stetler, C., Jansen-West, K., Lewis, lyzer (Applied Biosystems, Carlsbad, CA, USA). Each 5 ml re- J., Link, C.D., Rademakers, R., Nykjaer, A. et al.. (2012) Progranulin action contained: 2 ml of pre-diluted (1:50) cDNA, 0.5 ml regulates neuronal outgrowth independent of Sortilin. Mol. Neurodegener., primer mix (1 mM for each primer) and 2.5 ml SYBR green 7, 33. (Invitrogen). The thermal cycling conditions were as follows: 6. He, Z., Ong, C.H., Halper, J. and Bateman, A. (2003) Progranulin is a mediator of the wound response. Nat. Med., 9, 225–229. 508C for 2 min, 958C for 2 min, 40 cycles of 958C for 15 s and 7. Kessenbrock, K., Frohlich, L., Sixt, M., Lammermann, T., Pfister, H., 608C for 1 min. The mRNA levels of SORT1 and PGRN were Bateman, A., Belaaouaj, A., Ring, J., Ollert, M., Fassler, R. et al. (2008) normalized to GAPDH, an endogenous transcript control. The Proteinase 3 and neutrophil elastase enhance inflammation in mice by data were analyzed by using Software RQ Manager 1.2 inactivatingantiinflammatory progranulin. J. Clin. Invest., 118, 2438–2447. (Applied Biosystems). Primer sequences for amplification of 8. Gijselinck, I., Van Broeckhoven, C. and Cruts, M. (2008) Granulin ′ ′ ′ mutations associated with frontotemporal lobar degeneration and related PGRN were 5 CCCTTCTGGACAAATGGC 3 and 5 disorders: an update. Hum. Mutat., 29, 1373–1386. GGAAGTCCCTGAGACGGTAA 3 . Primer sequences for 9. Gass, J., Cannon, A., Mackenzie, I.R., Boeve, B., Baker, M., Adamson, J., ′ ′ GAPDH were 5 CTTTGGTATCGTGGAAGGACTCATG 3 Crook, R., Melquist, S., Kuntz, K., Petersen, R. et al. (2006) Mutations in ′ ′ and 5 CCAGTAGAGGCAGGGATGATGTTC 3 . Primer progranulin are a major cause of ubiquitin-positive frontotemporal lobar sequences for SORT1 were 5 GGCCTGTGGGTGTCCAA- degeneration. Hum. Mol. Genet., 15, 2988–3001. ′ ′ ′ GAA 3 and 5 GCAGGAGCCATTTGCATAGGTT 3 . 10. Pickering-Brown, S.M., Rollinson, S., Du Plessis, D., Morrison, K.E., Varma, A., Richardson, A.M., Neary, D., Snowden, J.S. and Mann, D.M. (2008) Frequency and clinical characteristics of progranulin mutation carriers in the Manchester frontotemporal lobar degeneration cohort: Statistical analysis comparison with patients with MAPT and no known mutations. Brain, 131, 721–731. Representative data of at least three independent experiments 11. Le Ber, I., van der Zee, J., Hannequin, D., Gijselinck, I., Campion, D., Puel, with reproducible results were shown. All statistical analysis M., Laquerriere, A., De Pooter, T., Camuzat, A., Van den Broeck, M. et al. was performed by unpaired student t-test. One-way analysis of (2007) Progranulin null mutations in both sporadic and familial variance (ANOVA) followed by Tukey’s post-test was applied frontotemporal dementia. Hum. Mutat., 28, 846–855. in multi-doses experiments. 12. Le Ber, I., Camuzat, A., Hannequin, D., Pasquier, F., Guedj, E., Rovelet-Lecrux, A., Hahn-Barma, V., van der Zee, J., Clot, F., Bakchine, S. et al. (2008) Phenotype variability in progranulin mutation carriers: a clinical, neuropsychological, imaging and genetic study. Brain, 131, SUPPLEMENTARY MATERIAL 732–746. 13. Bronner, I.F., Rizzu, P., Seelaar, H., van Mil, S.E., Anar, B., Azmani, A., Supplementary Material is available at HMG online. Donker Kaat, L., Rosso, S., Heutink, P. and van Swieten, J.C. (2007) Progranulin mutations in Dutch familial frontotemporal lobar degeneration. Eur. J. Hum. Genet., 15, 369–374. ACKNOWLEDGEMENTS 14. Huey, E.D., Grafman, J., Wassermann, E.M., Pietrini, P., Tierney, M.C., Ghetti, B., Spina, S., Baker, M., Hutton, M., Elder, J.W. et al. (2006) KO We thank Chad A. Cowan for providing the WT and SORT1 - Characteristics of frontotemporal dementia patients with a Progranulin hESC lines. mutation. Ann. Neurol., 60, 374–380. 15. Bruni,A.C.,Momeni, P.,Bernardi,L.,Tomaino,C.,Frangipane,F., Elder,J., Kawarai, T., Sato, C., Pradella, S., Wakutani, Y. et al. (2007) Heterogeneity Conflict of Interest statement. None declared. within a large kindred with frontotemporal dementia: a novel progranulin mutation. Neurology, 69, 140–147. 16. Shankaran, S.S., Capell, A., Hruscha, A.T., Fellerer, K., Neumann, M., FUNDING Schmid, B. and Haass, C. (2008) Missense mutations in the progranulin gene linked to frontotemporal lobar degeneration with ubiquitin-immunoreactive This work was supported by Mayo Clinic Foundation (L.P.), inclusions reduce progranulin production and secretion. J. Biol. Chem., 283, NIH R01 AG026251, R01 NS 063964-01, R01 NS077402, 1744–1753. 17. Capell, A., Liebscher, S., Fellerer, K., Brouwers, N., Willem, M., Lammich, R01 ES20395-01, R21 NS84528-01J (L.P.), NS065782 (R.R.), S., Gijselinck, I., Bittner, T., Carlson, A.M., Sasse, F. et al. (2011) Rescue of NIH R01 AG026251, R01 NS057553 (F.B.G.) and the Consor- progranulin deficiency associated with frontotemporal lobar degeneration tium for Frontotemporal dementia Research (R.R.) (F.B.G.). by alkalizing reagents and inhibition of vacuolar ATPase. J. Neurosci., 31, Funding to pay the Open Access publication charges for this 1885–1894. article was provided by the Mayo Clinic. 18. Cenik, B., Sephton, C.F., Dewey, C.M., Xian, X., Wei, S., Yu, K., Niu, W., Coppola, G., Coughlin, S.E., Lee, S.E. et al. (2011) Suberoylanilide hydroxamic acid (vorinostat) up-regulates progranulin transcription: rational therapeutic approach to frontotemporal dementia. J. Biol. Chem., REFERENCES 286, 16101–16108. 19. Hu, F., Padukkavidana, T., Vaegter, C.B., Brady, O.A., Zheng, Y., 1. Toh, H., Chitramuthu, B.P., Bennett, H.P. and Bateman, A. (2011) Structure, Mackenzie, I.R., Feldman, H.H., Nykjaer, A. and Strittmatter, S.M. (2010) function, and mechanism of progranulin; the brain and beyond. J. Mol. Sortilin-mediated endocytosis determines levels of the frontotemporal Neurosci., 45, 538–548. dementia protein, progranulin. Neuron, 68, 654–667. 2. Bateman, A.and Bennett,H.P.(2009) The granulingene family:fromcancer 20. Carrasquillo, M.M., Nicholson, A.M., Finch, N., Gibbs, J.R., Baker, M., to dementia. Bioessays, 31, 1245–1254. Rutherford, N.J., Hunter, T.A., DeJesus-Hernandez, M., Bisceglio, G.D., 3. Baker, M., Mackenzie, I.R., Pickering-Brown, S.M., Gass, J., Rademakers, Mackenzie, I.R. et al. (2010) Genome-wide screen identifies rs646776 near R., Lindholm, C., Snowden, J., Adamson, J., Sadovnick, A.D., Rollinson, S. sortilin as a regulator of progranulin levels in human plasma. Am. J. Hum. et al. (2006) Mutations in progranulin cause tau-negative frontotemporal dementia linked to chromosome 17. Nature, 442, 916–919. Genet., 87, 890–897. Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 1478 Human Molecular Genetics, 2014, Vol. 23, No. 6 21. De Muynck, L., Herdewyn, S., Beel, S., Scheveneels, W., Van Den Bosch, Progranulin functions as a neurotrophic factor to regulate neurite outgrowth L., Robberecht, W. and Van Damme, P. (2013) The neurotrophic properties and enhance neuronal survival. J. Cell Biol., 181, 37–41. of progranulin depend on the granulin E domain but do not require sortilin 35. Chitramuthu, B.P., Baranowski, D.C., Kay, D.G., Bateman, A. and Bennett, binding. Neurobiol. aging, 34, 2541–2547. H.P. (2010) Progranulin modulates zebrafish motoneuron development in 22. Almeida, S., Zhang, Z., Coppola, G., Mao, W., Futai, K., Karydas, A., vivo and rescues truncation defects associated with knockdown of Survival Geschwind, M.D., Tartaglia, M.C., Gao, F., Gianni, D. et al. (2012) Induced motor neuron 1. Mol. Neurodegener., 5, 41. pluripotent stem cell models of progranulin-deficient frontotemporal 36. Laird, A.S., Van Hoecke, A., De Muynck, L., Timmers, M., Van den Bosch, dementia uncover specific reversible neuronal defects. Cell Rep., 2, L., Van Damme, P. and Robberecht, W. (2010) Progranulin is neurotrophic 789–798. in vivo and protects against a mutant TDP-43 induced axonopathy. PLoS 23. Dimos, J.T., Rodolfa, K.T., Niakan, K.K., Weisenthal, L.M., Mitsumoto, H., One, 5, e13368. Chung, W., Croft, G.F., Saphier, G., Leibel, R., Goland, R. et al. (2008) 37. Keiser, M.J., Roth, B.L., Armbruster, B.N., Ernsberger, P., Irwin, J.J. and Induced pluripotent stem cells generated from patients with ALS can be Shoichet, B.K. (2007) Relating protein pharmacology by ligand chemistry. differentiated into motor neurons. Science, 321, 1218–1221. Nat. biotechnol., 25, 197–206. 24. Israel, M.A., Yuan, S.H., Bardy, C., Reyna, S.M., Mu, Y., Herrera, C., 38. Wells, J.A. and McClendon, C.L. (2007) Reaching for high-hanging fruit in Hefferan, M.P., Van Gorp, S., Nazor, K.L., Boscolo, F.S. et al. (2012) drug discovery at protein–protein interfaces. Nature, 450, 1001–1009. Probingsporadicand familialAlzheimer’s disease usinginducedpluripotent 39. Meyer-Hoffert, U. (2009) Neutrophil-derived serine proteases modulate stem cells. Nature, 482, 216–220. innate immune responses. Front. Biosci., 14, 3409–3418. 25. Nguyen, H.N., Byers, B., Cord, B., Shcheglovitov, A., Byrne, J., Gujar, P., 40. Morinville, A., Martin, S., Lavallee, M., Vincent, J.P., Beaudet, A. and Kee, K., Schule, B., Dolmetsch, R.E., Langston, W. et al. (2011) LRRK2 Mazella, J. (2004) Internalization and trafficking of neurotensin via NTS3 mutant iPSC-derived DA neurons demonstrate increased susceptibility to receptors in HT29 cells. Int. J. Biochem. Cell Biol., 36, 2153–2168. oxidative stress. Cell Stem Cell, 8, 267–280. 41. Lefrancois, S., Zeng, J., Hassan, A.J., Canuel, M. and Morales, C.R. (2003) 26. Zheng,Y.,Brady,O.A.,Meng,P.S.,Mao,Y.andHu, F.(2011)C-terminus of The lysosomal trafficking of sphingolipid activator proteins (SAPs) is progranulin interacts with the beta-propeller region of sortilin to regulate mediated by sortilin. EMBO J., 22, 6430–6437. progranulin trafficking. PLoS One, 6, e21023. 42. Ni, X., Canuel, M. and Morales, C.R. (2006) The sorting and trafficking of 27. Quistgaard, E.M., Madsen, P., Groftehauge, M.K., Nissen, P., Petersen, lysosomal proteins. Histol. histopathol., 21, 899–913. C.M. and Thirup, S.S. (2009) Ligands bind to Sortilin in the tunnel of a 43. Canuel, M., Korkidakis, A., Konnyu, K. and Morales, C.R. (2008) Sortilin ten-bladed b-propeller domain. Nat. Struct. Mol. Biol., 16, 96–98. mediates the lysosomal targeting of cathepsins D and H. Biochem. Biophys. 28. Mazella, J., Chabry, J., Zsurger, N. and Vincent, J.P. (1989) Purification of Res. Commun., 373, 292–297. the neurotensin receptor from mouse brain by affinity chromatography. 44. Smith, K.R., Damiano, J., Franceschetti, S., Carpenter, S., Canafoglia, L., J. Biol. Chem., 264, 5559–5563. Morbin, M., Rossi, G., Pareyson, D., Mole, S.E., Staropoli, J.F. et al. (2012) 29. Mazella, J., Zsurger, N., Navarro, V., Chabry, J., Kaghad, M., Caput, D., Strikingly different clinicopathological phenotypes determined by Ferrara, P., Vita, N., Gully, D., Maffrand, J.P. et al. (1998) The 100-kDa progranulin-mutation dosage. Am. J. Hum. Genetics, 90, 1102–1107. neurotensin receptor is gp95/sortilin, a non-G-protein-coupled receptor. 45. Gieselmann, V. (2005) What can cell biology tell us about heterogeneity in J. Biol. Chem., 273, 26273–26276. lysosomal storage diseases? Acta Paediatr. Suppl., 94, 80–86; discussion 30. Nielsen, M.S.,Jacobsen,C., Olivecrona, G., Gliemann, J. and Petersen, C.M. (1999) Sortilin/neurotensin receptor-3 binds and mediates degradation of 46. Serrero, G. (2003) Autocrine growth factor revisited: PC-cell-derived lipoprotein lipase. J. Biol. Chem., 274, 8832–8836. growth factor (progranulin), a critical player in breast cancer tumorigenesis. 31. Petersen, C.M., Nielsen, M.S., Nykjaer, A., Jacobsen, L., Tommerup, N., Biochem. Biophys. Res. Commun., 308, 409–413. Rasmussen, H.H., Roigaard, H., Gliemann, J., Madsen, P. and Moestrup, 47. Ho, J.C., Ip, Y.C., Cheung, S.T., Lee, Y.T., Chan, K.F., Wong, S.Y. and Fan, S.K. (1997) Molecular identification of a novel candidate sorting receptor S.T. (2008) Granulin-epithelin precursor as a therapeutic target for purified from human brain by receptor-associated protein affinity hepatocellular carcinoma. Hepatology, 47, 1524–1532. chromatography. J. Biol. Chem., 272, 3599–3605. 48. Kamrava, M., Simpkins, F., Alejandro, E., Michener, C., Meltzer, E. and 32. Ding, Q., Lee, Y.K., Schaefer, E.A., Peters, D.T., Veres, A., Kim, K., Kohn, E.C. (2005) Lysophosphatidic acid and endothelin-induced Kuperwasser, N., Motola, D.L., Meissner, T.B., Hendriks, W.T. et al. (2013) proliferation of ovarian cancer cell lines is mitigated by neutralization of A TALEN genome-editing system for generating human stem cell-based granulin-epithelin precursor (GEP), a prosurvival factor for ovarian cancer. disease models. Cell Stem Cell, 12, 238–251. Oncogene, 24, 7084–7093. 33. Ryan, C.L., Baranowski, D.C., Chitramuthu, B.P., Malik, S., Li, Z., Cao, M., 49. Prudencio, M., Jansen-West, K.R., Lee, W.C., Gendron, T.F., Zhang, Y.J., Minotti, S., Durham, H.D., Kay, D.G., Shaw, C.A. et al.. (2009) Progranulin Xu, Y.F., Gass, J., Stuani, C., Stetler, C., Rademakers, R. et al. (2012) is expressedwithinmotor neuronsand promotesneuronalcell survival.BMC Neurosci., 10, 130. Misregulation of human sortilin splicing leads to the generation of a 34. Van Damme, P., Van Hoecke, A., Lambrechts, D., Vanacker, P., Bogaert, E., nonfunctional progranulin receptor. Proc. Natl. Acad. Sci. USA, 109, van Swieten,J., Carmeliet, P., VanDen Bosch,L.and Robberecht,W. (2008) 21510–21515. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Human Molecular Genetics Oxford University Press

Targeted manipulation of the sortilin–progranulin axis rescues progranulin haploinsufficiency

Download PDF
12 pages

Loading next page...
 
/lp/oxford-university-press/targeted-manipulation-of-the-sortilin-progranulin-axis-rescues-xtQnl0oc5C

References (64)

Publisher
Oxford University Press
Copyright
Copyright © 2022 Oxford University Press
Subject
ARTICLES
ISSN
0964-6906
eISSN
1460-2083
DOI
10.1093/hmg/ddt534
pmid
24163244
Publisher site
See Article on Publisher Site

Abstract

Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 6 1467–1478 doi:10.1093/hmg/ddt534 Advance Access published on October 26, 2013 Targeted manipulation of the sortilin – progranulin axis rescues progranulin haploinsufficiency 1 2 1 1 1 Wing C. Lee , Sandra Almeida , Mercedes Prudencio , Thomas R. Caulfield , Yong-Jie Zhang , 1 1 1 1 1 William M. Tay , Peter O. Bauer , Jeannie Chew , Hiroki Sasaguri , Karen R. Jansen-West , 1 1 1 3 1 Tania F. Gendron , Caroline T. Stetler , NiCole Finch , Ian R. Mackenzie , Rosa Rademakers , 2 1, Fen-Biao Gao and Leonard Petrucelli Department of Neuroscience, Mayo Clinic Jacksonville, 4500 San Pablo Rd S, Jacksonville, FL 32224, USA, 2 3 Department of Neurology, University of Massachusetts Medical School, Worcester, MA 01605, USA and Department of Pathology, University of British Columbia, 2211 Wesbrook Mall, Vancouver V6T 2B5, British Columbia, Canada Received August 23, 2013; Revised October 18, 2013; Accepted October 22, 2013 Progranulin (GRN) mutations causing haploinsufficiency are a major cause of frontotemporal lobar degener- ation (FTLD-TDP). Recent discoveries demonstrating sortilin (SORT1) is a neuronal receptor for PGRN endo- cytosis and a determinant of plasma PGRN levels portend the development of enhancers targeting the SORT1 – PGRN axis. We demonstrate the preclinical efficacy of several approaches through which impairing PGRN’s interaction with SORT1 restores extracellular PGRN levels. Our report is the first to demonstrate the efficacy of enhancing PGRN levels in iPSC neurons derived from frontotemporal dementia (FTD) patients with PGRN deficiency. We validate a small molecule preferentially increases extracellular PGRN by reducing SORT1 levels in various mammalian cell lines and patient-derived iPSC neurons and lymphocytes. We further demonstrate that SORT1 antagonists and a small-molecule binder of PGRN , residues critical for 588 – 593 PGRN – SORT1 binding, inhibit SORT1-mediated PGRN endocytosis. Collectively, our data demonstrate that the SORT1 – PGRN axis is a viable target for PGRN-based therapy, particularly in FTD-GRN patients. pathogenic loss-of-function mutations in GRN reported so far INTRODUCTION account for 4–26% of familial FTD cases and 1–12% of spor- adic cases worldwide (4,8–15). Several disease-related missense Progranulin (PGRN), a multi-functional secreted glycoprotein, mutations have also been identified and appear to be associated plays key roles in various biological processes (1,2) and, when de- with reduced PGRN secretion (16). As such, drug discovery ficient, leads to frontotemporal dementia (FTD). Individuals car- efforts aimed at enhancing PGRN levels in patients with FTD rying a null or loss-of-function allele in GRN, the gene encoding with GRN mutations (FTD-GRN) is of great interest to the scien- PGRN, suffer from PGRN haploinsufficiency, which is a major tific community (17,18). Exciting new research by our group and cause of the most common pathological subtype of FTD, fronto- others demonstrates the interaction between PGRN and sortilin temporal lobar degeneration (FTLD-TDP) (3,4). While the (SORT1), a neuronal receptor that mediates extracellular mechanisms linking loss of PGRN function and disease patho- PGRN clearance via an endocytosis mechanism (19), is a prom- genesis remain unclear, evidence from molecular and cellular ising target. For example, while SORT1 is an important regulator studies suggests that decreased levels of extracellular PGRN of PGRN levels (20), PGRN’s neurotrophic and neuroprotective (exPGRN) are relevant to disease pathogenesis. For example, supplementation of exogenous PGRN in culture medium effects are SORT1 independent (5,21), providing assurance that 2/2 rescues neurite outgrowth deficits observed in Grn neuronal the PGRN–SORT1 axis is a viable target for drug discovery cultures (5), facilitates wound healing by promoting the accumu- efforts aimed at identifying exPGRN enhancers. lation of neutrophils, macrophages and fibroblasts (6) and inhibits Herein, we identify and validate several therapeutic neutrophilic inflammation in vivo (7). In addition, the 69 strategies—the development of SORT1 expression suppressors, To whom correspondence should be addressed. Tel: +1 9049532855; Email: petrucelli.leonard@mayo.edu # The Author 2013. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 1468 Human Molecular Genetics, 2014, Vol. 23, No. 6 Figure 1. MPEP decreases SORT1 expression and increases extracellular PGRN in mammalian cell lines. (A and B) M17 cells were treated with control siRNA (siR-Ctrl) or gene-specific SORT1 siRNA (siR-SORT1). (A) Intracellular levels of PGRN, SORT1 and GAPDH were evaluated by western blot at a 48 h time-point. (B) Suppression of SORT1 levels increased extracellular PGRN levels. (C) Chemical name and structure of MPEP. (D–I) Treatment of M17 cells (D and E), HeLa cells (F and G) or NIH3T3 cells (H and I) with MPEP for 24 h dose dependently reduced SORT1 levels (D, F and H) and increased exPGRN levels (E, G and I) at 10 and ∗∗∗ 20 mM. (J) Under the same conditions, MPEP did not affect levels of SORLA, SORCS1 and ubiquitinated proteins in M17 cells. P , 0.001 versus vehicle control, analysis performed by one-way ANOVA followed by Tukey’s post-test. SORT1 antagonists and small-molecule PGRN-specific To evaluate the capacity of small molecules to suppress binders—to reduce SORT1-mediated endocytosis, thereby SORT1 levels, we screened a commercial compound library enhancing exPGRN levels in relevant disease models. and identified a bioactive compound, 1-[2-(2-tert-butyl-5- methylphenoxy)-ethyl]-3-methylpiperidine, termed MPEP (Fig. 1C), that dose dependently reduces SORT1 levels in a mammalian cell lines. MPEP treatment for 24 h reduced RESULTS SORT1 (Fig. 1D) and increased exPGRN up to 3-fold at a Pharmacological suppression of SORT1 expression 20 mM dose (Fig. 1E). A similar effect was observed in HeLa increases extracellular PGRN in mammalian cell lines cells (Fig. 1F and G) and in NIH3T3 cells, a mouse fibroblast line (Fig. 1H and I). To determine the specificity of MPEP, we Recent genetic evidence implicating SORT1 as an important also examined the effect of MPEP on other sortilin-related pro- regulator of GRN levels in serum (20) and the finding that abla- +/2 teins: sortilin-related LDLR class A repeats-containing receptor tion of Sort1 in Grn mice restores Pgrn in brain to normal (SORLA) and sortilin-related VPS10 domain-containing recep- levels (19) support the notion that pharmacological suppression tor 1 (SORCS1) (Fig. 1J). In addition, we examined levels of of SORT1 expression in the brain may be a potential therapeutic ubiquitinated proteins following MPEP treatment to determine approach for upregulating PGRN levels. Prior to investigating whether this drug influences proteasomal degradation of pro- the use of SORT1 protein suppression as a PGRN enhancer, teins (Fig. 1J). That MPEP did not significantly alter the levels we first confirmed that SORT1 downregulation enhances of any of these targets except for SORT1 provides evidence of PGRN levels in culture. To this end, SORT1 expression was its specificity. Moreover, under the same conditions, MPEP reduced by treating M17 cells with SORT1-specific silencing neither increased GRN mRNA nor suppressed SORT1 mRNA RNA (siRNA) (Fig. 1A), which resulted in a significant increase levels indicating the effect is transcription independent (Supple- in exPGRN levels in a time-dependent manner (Fig. 1B). While mentary Material, Fig. S1A). Rather, the significant decrease in previous reports suggested the major mechanism of such SORT1 protein expression as early as 2 h post-MPEP treatment observed rescue effects was due to a reduction of SORT1- suggests MPEP effectively suppresses SORT1 levels by increas- mediated PGRN endocytosis, additional mechanisms, such as ing SORT1 degradation (Supplementary Material, Fig. S1C). induction of PGRN secretion, had not been ruled out. To Finally, treatment of M17 cells with high doses of rPGRN, address this gap in our knowledge, we also evaluated intracellu- which unlike MPEP does not reduce SORT1 expression, rules lar PGRN levels and detected no significant changes (Fig. 1A), out the possibility that MPEP acts via autocrine regulation which indicates the boostin exPGRN levels is likely due to inhib- (Supplementary Material, Fig. S1D). ition of endocytosis. Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 6 1469 Figure 2. MPEP decreases SORT1 expression and increases extracellular PGRN in cellular models of FTD-GRN.(A–C) MPEP decreased SORT1 levels (A) and increased extracellular PGRN levels (B) but not intracellular PGRN (C) in a recently reported PGRN S116X human neuron model differentiated from FTLD patient- specific iPSCs. (D–G) MPEP reduced intracellular SORT1 levels (D and F) and preferentially increased extracellular PGRN levels (E and G) in lymphoblastoid cell lines (LCLs) from two FTD-GRN families, UBC17 (D and E) and UBC15 (F and G). MPEP at 20 mM restored extracellular PGRN to near normal level in mutation +/2 +/+ ∗∗ ∗∗∗ carrier (GRN ) compared with non-carrier control (GRN ). P , 0.01, P , 0.001 versus vehicle control, analysis performed by one-way ANOVA followed by Tukey’s post-test. Pharmacological response to the small-molecule MPEP PGRN S116X neurons yielded GRN mRNA that are 41% rescues PGRN haploinsufficiency in iPSC-neurons and lower than in control and sporadic FTD neurons (22). lymphoblastoid cells derived from FTD patients To evaluate whether MPEP counteracts the PGRN haploin- sufficiency in the above-mentioned model, the cells were To validate the effect of MPEP in an authentic neuronal model treated with the compound at 5, 10 and 20 mM for 24 h. The of the disease, we used recently developed induced pluripotent result was as follows: in iPSC-neurons carrying a heterozygous stem cells (iPSCs) derived from FTD patients with a novel PGRN S116X mutation, MPEP treatment dose dependently heterozygous GRN mutation (22). Because iPSCs are derived reduced SORT1 levels (Fig. 2A) and selectively increased directly from somatic tissues of patients, the differentiated neur- exPGRN up to 5.5-fold (Fig. 2B). Furthermore, MPEP treatment onal cells represent a more physiologically relevant model of the had no effect on intracellular PGRN levels (Fig. 2C). disease and platform for testing therapeutics. Like many other To further evaluate the efficacy of MPEP, we also tested the neurodegenerative disease iPSCs models, including ALS (23), small molecule on lymphoblastoid cell lines (LCLs) derived AD (24) and PD (25), the FTD-GRN iPSCs neurons display from individuals with and without a disease-associated GRN the expected molecular phenotype caused by the inherited muta- mutation. Our investigators initially generated the chosen tion (i.e. PGRN haploinsufficiency). For example, iPSC lines LCLs, which were first used to validate mutations in GRN in derived from FTD patients with the PGRN S116X mutation FTD patients (3) and were shown to reproduce the PGRN were confirmed to have 50% lower GRN mRNA levels than in haploinsufficiency phenotype (3,4). Consistent with the results non-carrier controls. Differentiation of the iPSC lines into from mammalian cell lines and iPSC-neurons, MPEP treatment Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 1470 Human Molecular Genetics, 2014, Vol. 23, No. 6 Figure 3. Elastase-mediated removal of C-terminal motif of PGRN blocks PGRN endocytosis by SORT1. (A) A synthetic PGRN peptide was enzymatically 574– 593 cleavedbyrecombinantelastase (EL)in a time-dependentmanner as analyzedbyMS-MALDIanalysis. ThePGRN peptidewitha molecularweightof2430 Da 574– 593 was processed into a product with 1806Da indicating the last five residues were removed by EL as shown schematically in (B). (C) WT or EL-site-mutated A588G-rPGRN protein in cell culture media was set up to react with EL in vitro. After reaction, the media was used as input to carry out a cellular endocytosis assay in M17 cells overexpressing SORT1. PGRN endocytosed in cells was detected in cell lysate after treatment (uptake). Unlike WT rPGRN, A588G-rPGRN was more resistant to EL activity as shown by the amount of residual full-length protein in the medium (input), which was endocytosed by cells to a similar extent compared with no EL control. Immunocytofluorescence analysis showed that only full-length rPGRN (FL) (E) but not the carboxyl-terminal-truncated SORT1 SORT1 PGRN (F) was significantly endocytosed by M17 cells. (D) The untreated M17 control was included. PGRN and SORT1 were labeled in red and 1-588 green, respectively. +/+ decreased SORT1 (Fig. 2D and F) and preferentially increased with the vehicle-treated GRN -LCL, indicating a tight exPGRN (Fig. 2E and G) in the LCLs from two FTD-GRN- correlation between SORT1 reduction and exPGRN induction. affected families: one termed UBC17, which carries a p.C31LfsX34 mutation, and one termed UBC15, which carries The PGRN C-terminal motif, PGRN , is essential (589 –593) a p.R418X mutation (3,4). for SORT1-mediated endocytosis In family UBC17 (Fig. 2D and E), 20 mM of MPEP complete- ly restored exPGRN levels in the heterozygous mutation carrier To identify small molecules that interfere with the SORT1– +/2 (GRN -LCL) to a level comparable with the homozygous PGRN interaction via a receptor antagonist approach, first we +/+ non-carrier (GRN -LCL) (Fig. 2E). While MPEP restored aimed to locate the specific PGRN region essential for SORT1 exPGRN levels in the UBC15 family to 80% of that of non- binding and endocytosis. Our group and others have shown +/+ carrier GRN -LCL levels (Fig. 2G), such exPGRN levels rep- C-terminal, and not N-terminal, tagging of recombinant human resent a 3-fold increase compared with the control. Higher PGRN protein completely blocks intracellular uptake (26) (Sup- SORT1 levels were also observed in the corresponding plementary Material, Fig. S2A). In our current study, we identi- UBC17 and UBC15 heterozygous LCL (Fig. 2D and F), which fied a crucial neutrophil elastase (NE) cleavage site between may account for the more severe exPGRN reduction beyond residues Ala-588 and Leu-589 of PGRN by in vitro digestion of the haploinsufficiency caused by the null mutation (20). At a synthetic PGRN peptide by NE, followed by 574–593 20 mM, however, MPEP reduced SORT1 levels and increased MALDI-MS analysis (Fig. 3A and B). In addition, we found dis- +/2 exPGRN in the GRN -LCL model to levels comparable ruption of the A588 elastase cleavage site, through the Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 6 1471 Figure 4. SORT1 ligands competitively inhibit PGRN endocytosis. (A–C) Structure for NTS (A), native human PGRN peptide (B) and mouse Pgrn 588– 593 584 – 589 peptide (C) docked with SORT1 and the corresponding docking scores are shown. (A) The strongly interacting core of NTS peptide, -PYIL, which is resolved in X-ray structure, is showninbolder,darkgreen.SORT1 residueswithin4 Aare renderedingray. SecondarystructureforSORT1isshownas cartoonribbons.The carboxylate ofterminalLeu is visible in the electrostatic surface renderingof the binding pocket.As with NTS,(B) the humanPGRN (-ALRQLL) or (C) the mousePgrn 588 –593 584– (-VPRPLL) is shown docked with the SORT1-binding pocket (electrostatic surface rendered) with the residues ,4A indicated. Carboxylate of Leu is in the iden- TM tical position as NTS. (D and E) A quantitative cell-based assay has been established to measure PGRN endocytosis by SORT1. DyLight 594-labeled rPGRN was SORT1 endocytosed dose dependently in COS-1 cells. (D) Images were captured by BD-pathway system. (E) Quantitative cellular endocytosis of DyL-rPGRN was SORT1 measured by fluorescence signal from treated cells. (F) NTS at 0.1, 1 and 5 mM dose dependently inhibited PGRN endocytosis. Untreated COS-1 cells were included as negative control (2ve). (G) Quantification of signal from (F). (H and I) SORT1 ligands at 10 mM, NTS, human PGRN peptide and mouse 588 – 593 ∗∗∗ PGRN peptide competitively inhibited DyL-rPGRN endocytosis as compared with vehicle control, respectively. P , 0.001 versus vehicle control, analysis 584 – 589 performed either by (G) one-way ANOVA followed by Tukey’s post-test or (I) by unpaired student t-tests. A588G introduction of an Ala-to-Gly mutation, protected PGRN usually discontinuous, which further reduces the probability of from elastase processing, hence preserving the SORT1-inter- identifying effective disruptors of protein–protein interactions. A588G action motif and PGRN endocytosis (Fig. 3C). To demon- To address these challenges in a time-efficient manner, we strate that the PGRN residues 588–593 are critical for employed the use of computer-assisted modeling. As our SORT1-mediated PGRN endocytosis, we produced recombinant search for a physiological ligand derived from the previously truncated PGRN 1–588 (rPGRN ). The truncated protein implicated carboxyl-terminus (C-terminus) of PGRN(26) shed 1-588 failed to be endocytosed in M17 cells overexpressing SORT1 new light on the importance of PGRN residues 588–593, we (Fig. 3F), unlike the full-length rPGRN control (Fig. 3E), which generated models of SORT1 bound to human PGRN , 588 – 593 presented an endolysosomal localization (Supplementary Mater- neurotensin (NTS), which is a high-affinity SORT1 ligand ial, Fig. S2B), thereby validating the PGRN region as an and mouse Pgrn (-ALRQLL, -ELYENKPRRPYIL and 589–593 584 – 589 essential motif for SORT1-mediated PGRN endocytosis. -VPRPLL, respectively). We used crystal structure data of SORT1 complexed with NTS(27) to model peptide sequences into the binding cleft of Computer-assisted modeling confirms ligands SORT1 using NTS fragment -PYIL as a template, to deter- 10-13 share same SORT1-binding pocket mine a GRID for docking, and then to optimize the interactions through energy minimization (Fig. 4A–C). The model of the sub- Identification of small molecules to target protein–protein inter- strate neurotensin (-PYIL fragment) obtains a docked position action interfaces is considered extremely challenging owing to relative to the crystallographic structure 3F6 K within 0.8 A the size disparity between small molecules and large contact sur- RMSD; NTS residues proline and tyrosine fill a largely faces on proteins. In addition, protein contact surfaces are Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 1472 Human Molecular Genetics, 2014, Vol. 23, No. 6 hydrophobic cleft between the flanking clusters of positive mimetics, may serve as potential exPGRN enhancers through in- charge, with the NTS Leu side chain embedded in the binding hibition of extracellular clearance. site forming numerous hydrophobic interactions with Phe273, Phe281, Ile294 and Ile320. For each peptide that docked, the carb- Use of small-molecule and antibody binders of the oxylate of the terminal Leu formed strong interactions with PGRN motif inhibit SORT1-mediated PGRN 588–593 SORT1, as shown in Figure 4A–C. The peptides docked with endocytosis the SORT1-receptor gain stabilization via charge complementar- ity from electrostatic interactions between the carboxylate anions Given that restoring exPGRN levels using SORT1 antagonists and Arg292 cations, which is stabilized by Ser283 and Ser319 at can potentially trigger off-target effects related to NTS (29), carbonyl interactions and bridged by H-bonds. Additional favor- lipoprotein lipase (30) and LDL-receptor-associated protein able close interactions between the peptides and SORT1 are com- (31) function and cause undesirable clinical side-effects, modu- prised of hydrophobic and van der Waals interactions and a series lating the PGRN–SORT1 interaction via PGRN-specific of hydrogenbondpartnerswithinthe bindingpocket(Fig.4A–C). binders may be a promising approach. To identify small mole- The docking results show NTS with an average docking score of cules that can selectively bind to the PGRN C-terminal motif 27.86 kcal/mol (Fig. 4A), human PGRN of 29.041 kcal/ to inhibit PGRN–SORT1 interactions, we screened 4800 com- 588–593 mol (Fig. 4B) and mouse Pgrn of 26.85 kcal/mol (Fig. pounds from a commercial library against a custom-synthesized 584– 589 4C). As such, the human PGRN has the highest binding PGRN peptide using a 384-well format, biochemical 588–593 588 – 593 affinity followedbyNTS and thenthe mouse Pgrn . Higher- binding assay that utilizes resonant waveguide grating biosensor 584–589 affinity binding of human PGRN peptide to SORT1 com- detection (Fig. 5A) (Supplementary Material, Method). We 588– 593 pared with that of NTS/SORT1 is derived from strong interaction identified a small molecule, 4-[2-(3-bromophenyl)vinyl]-6- pairs between the terminal few residues and key side chains and (trifluoromethyl)-2(1H)-pyrimidinone termed BVFP (Fig. 5B), backbone interactions from SORT1. Additionally, the mouse that binds to the PGRN peptide at a low micromolar con- 588 – 593 mPgrn /SORT1 binding suffers a loss in interaction centration (K ¼ 20 mM, Fig. 5C). Upon mixing and titrating 584–589 d from a distinct repositioning of mPgrn residues, -VPR, BVFP with the PGRN peptide, we detected a shift in 584–586 588 – 593 without significant SORT1 contacts. The details of the docking the ultraviolet(UV)-absorption spectra of BVFP and distinct models and decomposition of individual interactions between changes of absorption intensities, which in addition to the pres- SORT1 and NTS, human PGRN and mouse Pgrn ence of an isosbestic point at 313 nm, validated the BVFP and 588 –593 584– 589 are described in the Supplementary Material. PGRN peptide interaction (Fig. 5D). 588 – 593 To determine whether BVFP, when bound to PGRN, disrupts full-length PGRN and SORT1 binding, we devised a SORT1- SORT1 ligands competitively inhibit full-length PGRN dependent rPGRN precipitation assay utilizing the Meso-Scale endocytosis Discovery (MSD) system (Supplementary Material, Fig. S4A Given our computer-assisted modeling confirmed and provided and B). Immobilized SORT1 generated a specific binding new information of how NTS (28,29) and PGRN share a signal from the amino-terminal-tagged PGRN (N-rP) but not 588 – 593 similar SORT1-binding site (19), we sought to determine from the C-terminal-tagged PGRN, validating once again the whether the use of these ligands as antagonists could be an effect- PGRN–SORT1 binding is C-terminal dependent. A dose- ive strategy to inhibit PGRN endocytosis in cell culture. Using dependent inhibition of rPGRN binding to immobilized SORT1 TM DyLight -594 fluorescence-labeled rPGRN (DyL-rPGRN) was observed in the presence of 20 and 200 mM ofBVFP, respect- SORT1 and SORT1-expressing COS-1 (COS-1 ) cells, we estab- ively, proving the small molecule’s capacity to disrupt PGRN– lished a cell-based assay to quantitatively analyze PGRN endo- SORT1 binding (Supplementary Material, Fig. S4C). To TM cytosis by measuring DyLight -594 emission from includeahigher-affinity PGRN motif binder as a positive 588–593 endocytosed DyL-rPGRN upon treatment. Addition of titrated control, an antibody targeting the motif, PGRN-CT antibody, was SORT1 amounts of DyL-rPGRN (0 to 50 nM) to COS-1 cells pro- made for validating the approach (Fig. 5F). Using our quantitative duced a linear, non-saturated endocytosis response (Fig. 4D and endocytosis assay (Fig. 5E and G), we found that BVFP inhibited E). As expected, co-treatment of DyL-rPGRN with NTS (i.e. 0.5, DyL-rPGRN endocytosis by 23% when treated at 5 mM, com- 1 and 5 mM) dose dependently inhibited PGRN endocytosis pared with 43% inhibition by PGRN-CT antibody when treated (Fig. 4F and G). At the same concentrations, NTS, as well as at 80 nM (Fig. 5E and G). Note that a negative control antibody human PGRN and mouse PGRN , respectively, targeting GRN-A had no effect on DyL-rPGRN endocytosis. 588 – 593 584 – 589 inhibited full-length PGRN endocytosis by 90, 93 and 73% To show that the effects of BVFP and the PGRN-CT antibody (Fig. 4H and I). The fact that the mouse ortholog peptide of are SORT1 dependent, the PGRN endocytosis experiment was PGRN also significantly inhibited full-length PGRN repeated in human embryonic stem cell lines (hESCs) with and 588 – 593 SORT1 endocytosis by human SORT1 in COS-1 cells, albeit to a without SORT1 expression established by using the transcription lesser extent, suggests that the PGRN-CT motif is evolutionarily activator-like nucleases genome-editing system (TALENs) (32). conserved. In a protein co-immunoprecipitation assay, we First, we found SORT1 was abundantly expressed in the wild- SORT1 further validated that human PGRN peptide, NTS and type hESCs (Fig. 5H). Similar to the results from COS-1 588 – 593 mouse Pgrn peptide can each inhibit the physical inter- cells, BVFP and PGRN-CT antibody inhibited endocytosis of 584 – 589 action between full-length PGRN and SORT1 (Supplementary DyL-rPGRN by 22 and 39%, respectively, under the same experi- Material, Fig. S3). These results support the notion that the mental conditions in wild-type hESCs (Fig. 5I left). Treatment of KO development of SORT1 antagonists, such as stabilized forms DyL-rPGRN to the isogenic hESCs-SORT1 line, which has an of chemically modified human PGRN peptide or NTS identical genetic background and epigenetic state compared with 588 –593 Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 6 1473 Figure 5. Small-molecule and antibody binders of the PGRN motif inhibit SORT1-mediated PGRN endocytosis. (A) Schematic diagram illustrating the de- 588 – 593 tection mechanism of Epic biochemical binding assay. When screening for compound binders of PGRN peptide (target), the specific binding events are 588 – 593 detected by changes in reflected resonant wavelength before (l) and after (l ) addition of a potential compound binder. (B) Structure and chemical name of BVFP, an identified compound that binds the PGRN peptide. (C) Saturated binding curve of BVFP to PGRN peptide (n ¼ 4). (D) Wavelength red- (588 – 593) (588 – 593) shifting of UV-absorption spectra of BVFP (20 mM) upon titration with the PGRN peptide. The interaction between BVFP and the peptide is represented by (588 – 593) distinct changes in the absorption intensities (black arrows). The presence of an isosbestic point at 313 nm (inset red arrow) also confirmed the interactions. (E) The SORT1 PGRN binders, BVFP (5 mM) and PGRN-CT antibody (80 nM) inhibited rPGRN endocytosis as tested by the quantitative cell-based assay in COS-1 (588 – 593) cells. Both binders were pre-incubated with rPGRN for an hour and then added to the cells for an hour to allow endocytosis. A GRN-A specific antibody (80 nM) 1-593 was used as a negative control. (F) Vehicle control (2ve), full-length protein (rPGRN ) or truncated protein (rPGRN ) was analyzed by western blot using 1-593 1-588 PCDGF or PGRN-CT antibody for detection. The PGRN-CT antibody was confirmed to be PGRN dependent. (G) Fluorescence signal quantification of (588 – 593) KO (E). (H) Western blot analysis confirmed the absence of SORT1 protein in the SORT1 -hESCs. (I) BVFP and the PGRN-CT antibody inhibited PGRN endocytosis KO ∗ ∗∗ ∗∗∗ in WT-hESCs but not in SORT1 -hESCs indicating that the effect was SORT1 dependent. P , 0.05, P , 0.01, P , 0.001 versus vehicle control, analysis performed by unpaired student t-tests. the parent hESC-WT line, generated a significantly lower signal pathology (i.e. FTLD-TDP), therapeutically restoring PGRN compared with the WT-hESCs, indicating impaired PGRN endo- levels may be a promising therapeutic strategy. In fact, evi- cytosis in the absence of SORT1 (Fig. 5H). Both BVFP and dence from numerous studies suggests that PGRN acts as a PGRN-CT antibody had no effect on DyL-rPGRN endocytosis protective neurotrophic factor by regulating neuronal survival KO in the hESCs-SORT1 line (Fig. 5I right), confirming that and neurite growth in cortical/motor neurons, immortalized their effect on PGRN endocytosis is SORT1 dependent. cell lines and zebra fish (33–35) and that PGRN is protective against insults induced by TDP-43 (36). Given that there is currently no feasible way to pharmacologically manipulate PGRN levels in the brain and that recombinant PGRN is DISCUSSION too large to cross the blood–brain barrier (BBB) for protein replacement, the development of bio-available, BBB Given that partial loss of PGRN, owing to mutations in the permeable PGRN enhancers will be a valuable tool to PGRN gene (GRN), is causative of FTLD with TDP-43 Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 1474 Human Molecular Genetics, 2014, Vol. 23, No. 6 determine whether therapeutically modulating or increasing PGRN levels can alleviate the pathogenesis associated with FTD-GRN. While new strategies designed to upregulate PGRN levels have emerged, challenges remain in terms of limiting off-target effects. For example, Cenik et al. recently identified the histone deacetylase (HDAC) inhibitor suberoylanilide hydroxamic acid as an enhancer of GRN expression (18). Christian Haass’ group also has recently demonstrated four independent and highly se- lective inhibitors of vacuolar ATPase (V-ATPase) significantly elevate intracellular and secreted PGRN (17). However, because HDAC and V-ATPase inhibitors can potentially affect a wide- range of genes at transcriptional and post-translational levels, re- spectively, thereby increasing the likelihood that the inhibitors will also trigger undesirable, clinical side-effects and liabilities, the development of a strategy to improve target specificity may hold more promise. Here, we demonstrate the preclinical efficacy of several approaches through which impairing PGRN’s interaction with SORT1, a neuronal receptor that mediates PGRN endocytosis and degradation, restores extracellular PGRN levels in FTD patient-derived iPSC-neurons and lymphocytes (Fig. 6). To the best of our knowledge, our report is the first to demonstrate the efficacy of enhancing PGRN levels in iPSC-neurons derived Figure 6. Schematic diagram summarizing the strategies applied to inhibit from FTD patients with PGRN deficiency. We also validate SORT1-mediated endocytosis in the current study. (A) Under normal conditions, that the bioactive compound MPEP preferentially increases extracellular PGRN interacts with the b-propeller tunnel structure of SORT1 exPGRN levels by suppressing or reducing intracellular SORT1 using its C-terminal end binding motif as shown in red color. SORT1 facilitates levels without affecting sortilin-related family members endocytosis of exPGRN and directs it to the endolysosomal pathway for degrad- SORLA and SORCS1. To understand whether MPEP might in- ation. (B) High-affinity SORT1 ligands such as NTS or the PGRN (588 – 593) peptide competitively limits the access of exPGRN to SORT1-binding sites, crease exPGRN through protease(s) inhibition, we performed a thereby inhibiting PGRN endocytosis. (C) To improve target specificity, we database search using the similarity ensemble approach tool have also identified a small-molecule binder, BVFP, targeting the PGRN (588 – (37). No protease inhibitors in the database were found to have motif that is essential for PGRN–SORT1 interaction. We demonstrated 593) significant structural similarity toMPEP, suggesting that mechan- that pretreatment of BVFP to rPGRN significantly reduced the amount of rPGRN captured by SORT1 in vitro and inhibited SORT1-mediated rPGRN isms other than protease inhibition account for the SORT1- endocytosis. (D) Suppressors of SORT1 expression, such as MPEP, reduce targeted exPGRN upregulation by MPEP. SORT1-mediated endocytosis, thereby increasing extracellular PGRN levels. Given SORT1 s role in regulating exPGRN levels, we further The above-mentioned strategies, used alone or in combination with others, are demonstrate that SORT1 antagonists and PGRN binders 588 – 593 potential avenues for discovery of SORT1-dependent PGRN enhancers for the inhibit SORT1-mediated PGRN endocytosis. Our novel cell treatment of FTD-GRN. culture data reveal the PGRN peptide is the critical 588 – 593 region for SORT1–PGRN binding; the ‘hot spot’ is confirmed by our computational model of the PGRN–SORT1 interaction. serving as a regulatory mechanism to boost extracellular As mentioned, the feasibility of finding a small molecule that PGRN and granulins levels that are involved in the inflammation specifically targets protein–protein interactions is very much signaling cascade. case dependent(38), but identifying the crucial site improved While we demonstrate BVFP prevents SORT1-medited endo- our ability to hone in on high-affinity, small-molecule binders cytosis and increases exPGRN levels, an alternate supplemen- targeting either binding partner. To reduce undesirable effects tary approach would be to identify small molecules targeting on other SORT1 ligands, we concentrated our efforts on a com- SORT1, specifically at the Arg-292-, Ser-283- and Ser-319- pound library screen and biochemical binding assay that allowed binding pockets. A structure-based virtual high-throughput usto identify a small molecule, BVFP,that targets the C-terminal screening approach (vHTS) using the available NTS-complexed PGRN motif. As the motif is located on a linker region SORT1 crystallographic data could be employed to accelerate 588 – 593 and not on the functional granulin regions, binding of BVFP to drug discovery (27). The vHTS hits can be validated by a func- the extreme C-terminal motif is unlikely to significantly impair tional assay such as the quantitative endocytosis assay described PGRN function (1). Moreover, the discovery of a potential NE in the current study. cleavage site at the C-terminal Ala-588 residue of PGRN sug- We also evaluated the use of high-affinity SORT1 ligands as gests that a surge in extracellular NE levels would increase competitive inhibitors to effectively block exPGRN’s access C-terminal truncation at Ala-588 leading to accumulation of to SORT1 s binding site. Our computer-assisted modeling the PGRN protein and granulins generated by increasing sheds new light on why we observed a dramatic reduction in 1-588 NE proteolytic activity. As NE is an innate immune response PGRN endocytosis when ligands, such as NTS and human mediator that triggers downstream molecular events required PGRN , were present in vitro: the binding pocket and 588 – 593 for the inflammation process (39), the Ala-588 site is likely protein–protein interactions of NTS and PGRN’s C-terminal Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 6 1475 binding motif significantly overlap, hence explaining the com- given neutralizing PGRN to counteract its possible tumorigenic petitive inhibition effect. Although both NTS and full-length effect (46) is a proposed cancer treatment (47,48). In addressing PGRN induce endocytosis upon binding to SORT1, the two such concerns, and further exploring the relationship between ligands have been reported to traffic to different subcellular loca- TDP-43, SORT1 and PGRN (49), the clinical development of lizations, namely endolysosomal compartment (19) and the PGRN-based therapy for FTD-GRN patients will and should trans-Golgi network (40). In both cases, SORT1 is dissociated continue. from the bound ligands in early endosomes and then recycled back to the cell surface to drive the endocytosis process. While MATERIALS AND METHODS the use of a SORT1 antagonist approach is enticing owing to af- finity for and efficacy in blocking PGRN uptake, this approach is Recombinant PGRN protein, plasmids and antibodies also more likely to trigger off-target effects by disrupting Wild-type or mutant human recombinant PGRN protein binding between SORT1 and other ligands that occupy a (rPGRN) with a 6-histidine tag on the amino-terminus (N-rP) similar binding pocket. or the carboxyl-terminus (C-rP) was purified from culture As with many other ligands such as prosaposin (41), acid media of stable HEK293 cells secreting the corresponding sphingomyelinase (42) and cathepsins D and H (43), evidence protein. Purification procedures were the same as previously suggests SORT1 is a sorting receptor of PGRN for lysosome de- published (5). All the plasmids used in this study are listed as livery and degradation (19). Our group (5) and another (21) also follows: pCDNA4-PGRN(1–593) (N-terminal 6His tag), showed that SORT1 is not essential for stimulating neurite out- pCDNA4-PGRN(1–588) (N-terminal 6His tag), pCMV- growth in cultured neurons, suggesting that SORT1 might not be SORT1 (OriGene), pCDNA6-SORT1 (C-terminal Flag tag) involved in the PGRN-mediated signaling for neurotrophic and pEGFP-N1 (Clontech). All the antibodies used in this function. Nevertheless, inhibition of PGRN endocytosis might study are listed as follows: PCDGF (or Grn-E specific) antibody influence PGRN’s intracellular functions or unidentified intra- (1:1000 for western blot) (Invitrogen), human PGRN monoclo- cellular signaling pathways, especially in cells that rely on its nal antibody (1:200 for immunofluorescence) (R&D Systems), uptake from the extracellular space or neighbor cells. It is Grn-A antibody (80 nM for blocking PGRN endocytosis) known that complete deficiency in PGRN is a cause of clinical (Novus; 26320002), PGRN-CT antibody (1:1000 for western neuronal ceroid lipofuscinosis (NCL) (44), a rare lysosomal blot; 80 nM for blocking PGRN endocytosis) (21st Century; af- storage disease characterized by intraneuronal accumulation of finity purified rabbit polyclonal immunized with PGRN 574 – 593 autofluorescence lipofuscins. Based on this finding, it is reason- peptide), sortilin antibody (1:3000 for western blot; 1:1000 for able to speculate that PGRN has a lysosomal function related to immunofluorescence) (Abcam; ab16640), SORLA antibody turnover of lipofuscin-related substrates. While blocking PGRN (1:200) (R&D Systems; AF5699), SORCS1 antibody (1:1000) endocytosis will reduce level of PGRN in lysosomes, intracellu- (R&D Systems; AF3457), ubiquitin antibody (1:1000) lar PGRN provided from the biosynthetic pathway to lysosomes (DAKO; Z0458), EEA1 antibody (1:100 for immunofluores- is expected to be enough to maintain normal lysosomal function. cence) (BD Biosciences; 610457), giantin antibody (1:1000 At least for most lysosomal storage disorders, clinical onset is for immunofluorescence) (Abcam; ab24586), Flag M2 monoclo- often associated with.90% loss of activity of the corresponding nal antibody (1:500 for immunoprecipitation) (Sigma), GAPDH enzyme (45). However, further studies will be required to under- antibody (1:10 000 for western blot), and b-actin antibody stand whether abrogation of PGRN endocytosis can cause NCL (1:10 000 for western blot). or other deleterious biological consequences. Given the challenges associated with the development of ther- apies and clinical trials for patients suffering from neurodegen- Western blot analyses erative diseases, the need is urgent and the time is now for the PGRN, SORT1, GFP, GAPDH and b-actin proteins were medical community to rally behind a promising target. The analyzed using 10% Tris–glycine polyacrylamide gel by increased excitement surrounding PGRN-based therapies is SDS–PAGE. After electrophoresis, proteins on gel were trans- warranted: the specific patient population that could benefit is blotted to PVDF membranes followed by standard immunoblot- easily identifiable thanks to genetic testing. Moreover, the feasi- ting protocols. bility of restoring PGRN to normal levels has been shown by our group and others. The fact that genetic ablation of SORT1 restores and normalizes PGRN levels in brain uniquely positions SORT1 Analysis of PGRN endocytosis in M17 cells the SORT1–PGRN axis as an ideal target for PGRN-based therapy in FTD-GRN (19). Herein, we validated multiple strat- M17 cells seeded in 12-well culture plates (western blot) or on egies to target the SORT1–PGRN axis withthe aim ofenhancing glass coverslips in 24-well culture plate (immunofluorescence) PGRN levels. We show that SORT1 antagonists and PGRN were transfected with pEGFP-N1 (Clontech) or pCMV-SORT1 588 – binders inhibit SORT1-mediated PGRN endocytosis and a (OriGene) vectors. Next day, the cells were changed into serum- SORT1 suppressor restores exPGRN in patient-derived cell free medium 1 h before addition of rPGRN (500 ng/ml) and models. In addition, both small molecules identified in this treated for 60 min. The cells were then put on ice and washed study, namely MPEP and BVFP, are in compliance with the with cold PBS. For western blot, cells were lysed in cold Lipinski’s Rule of five to predict drug-likeness properties, M-Per protein extraction buffer followed by standard procedures which should facilitate in vivo studies in the near future. as mentioned. For immunofluorescence analyses, cells were Further studies to understand the dosage effect of fixed in 4% paraformaldehyde solution followed by PBS PGRN-enhancing reagents to treat FTD-GRN will be essential, washes. The cells were then incubated in blocking buffer Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 1476 Human Molecular Genetics, 2014, Vol. 23, No. 6 (PBS, 5% goat serum, 0.1% saponin) for 1 h at room temperature Determination of UV-absorption spectrum of BVFP followed by incubation in primary antibody overnight at 48C. Stock solutions of compound BVFP (20 mM) and PGRN (588 – 593) The cells were incubated with secondary antibodies (1:1000) peptide (12.96 mM) were prepared in DMSO and water, respect- for 1 h at room temperature. The coverslips were then ively. For the UV absorption experiment, both components were mounted onto slides and waited overnight before analysis. diluted to give final concentrations of 20 mM BVFP and 324 mM Image analysis was performed by Zeiss LSM 510 META con- peptide. The interaction between BVFP and the PGRN (588 – 593) focal microscope using 60× magnification setting. peptide was monitored by scanning over the UV absorbance range of 200–800 nm on a Cary 3 Bio UV–visible spectropho- tometer (Varian), as a small amount of KKGRN peptide (0.1 MALDI-MS analysis of NE-processed PGRN (584–593) equivalence/1 ml per addition) was titrated into the 20 mM peptide BVFP sample. In addition to baseline correction with DMSO, An enzymatic reaction was set up with 7 mg of custom- a control titration experiment using water in place of the synthesized PGRN peptide (Mayo peptide synthesis (584 – 593) PGRN peptide was performed under identical condi- (588 – 593) core) and 20 mgof NEina25 ml reaction at 378C. Every tions in order to correct for potential spectral changes owing to 10 min, 4 ml of the mixture was collected and diluted 10 times sample dilution. in a solvent mixture (50% acetonitrile:50% water:0.1% acetic acid) until 1 h had passed. For MALDI-MS analysis, 1 mlof the diluted reaction mixture was dried on a MS gold chip and then layered with a concentrated sinipinic acid matrix on top iPSC neuronal culture for crystallization. Afterwards, the samples were analyzed by the Bio-Rad MALDI-MS system. An induced pluripotent stem cell (iPSC) line from an FTD patient carrying a GRN nonsense mutation (p.S116X) was differ- entiated into neurons as described previously (22). Two-week- Quantitative endocytosis cell-based assay old neurons were incubated with 5–20 mM MPEP or DMSO COS-1 cells seeded on a 96-well black plate 1 day before were (in fresh culture medium) for 24 h. The medium was then col- transfected with pCMV-SORT1 vector. After 24 h, cells were lected, and cells were lysed with 1% NP-40 lysis buffer. treated with fluorescence-tagged rPGRN (DyL-rPGRN) pre- PGRN levels in total cell lysates and culture medium were deter- TM labeled by DyLight 594 antibody labeling kit (Thermo Scien- mined with an ELISA kit (Alexis Biochemicals, San Diego, CA, tific). The DyL-rPGRN was diluted in OptiMEM to tested USA) according to the manufacturer’s instructions. concentrations and incubated with the cells for 1 h for endocyto- sis. The cells were then washed with cold PBS and then fixed by 4% paraformaldehyde. After washing twice, each well was filled with PBS prior scanning. The total PGRN fluorescent signal MSD system-based immunoassay to analyze human PGRN from the cells was obtained by a plate reader with 593/618 nm The assay was performed according to manufacturer’s instruc- (Ex/Em) settings. The endocytosis signal was normalized by tions. A goat anti-PGRN antibody (R&D Systems) was used as the total nuclei signal obtained by staining with a Hoechst capture antibody. PGRN standards (from 0.5 to 50 ng/ml) 33342 dye. Baseline endocytosis signal was defined as a signal were from purified N-6His-rPGRN protein. For detection, a from medium-only-treated cells, whereas the 100% endocytosis 1:1 ratio of an anti-PCDGF antibody (Invitrogen) and a level was set as the signal obtained from cells treated with 50 nM SULFO-TAG-labeled anti-rabbit IgG antibody were applied. DyL-rPGRN. For testing SORT1 ligands, DyL-rPGRN and the The MSD Sector Imager 2400 was used to read assay signal. peptide were added simultaneously to the cells. For testing of PGRN binders, the binder was pre-incubated with DyL-rPGRN for 1 h before adding to the cells. For qualitative analysis, images from each well were captured by BD Pathway 855 system using a A MSD-based SORT1–PGRN interaction assay 20× magnification setting. M17 cells were transfected with pCMV-SORT1-Flag for 48 h. Then, the cell lysate was harvested by the same method used PGRN co-immunoprecipitation in the co-immunoprecipitation assay. Expression of SORT1- HEK293T cells were transfected empty vector or pCMV- Flag protein was confirmed by western blot. MSD plate was SORT1-Flag vector for 48 h. Then, cells were lysed by using first coated with the Flag M2 antibody (Sigma). Following an Co-IP buffer. Pre-incubation was performed with 300 mgof overnight incubation at 48C, the plate was blocked for 2 h and lysate protein mixed with 20 mM NTS, human PGRN then washed prior to addition of 5 mg/well of lysate protein con- (588 – 593) peptide or mouse PGRN peptide for 1 h. Then, rPGRN taining SORT1-Flag protein. After overnight incubation at 48C, (584 – 589) (100 nM) was added into the protein G beads pre-cleared super- it was then washed and added with 200 nM of either N-6His- or natant and mixed for 30 min. Next, anti-Flag M2 agarose C-6His-tagged rPGRN pre-incubated with either vehicle, (Sigma) was added and mixed for another 4 h. The agarose 20 mM or 200 mM of BVFP, for 30 min at room temperature. was collected by centrifugation at 1000 g for 3 min and Then plate was then incubated for 2 h, washed and incubated washed with Co-IP buffer six times. Captured protein was with a mixture of PCDGF antibody and a SULFO-TAG-labeled eluted from the beads using loading buffer and analyzed by anti-rabbit IgG antibody for 2 h. After final wash, the plate was western blot. read by the MSD Sector Imager 2400. Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 6 1477 4. Cruts, M., Gijselinck, I., van der Zee, J., Engelborghs, S., Wils, H., Pirici, D., Quantitative reverse transcription–polymerase chain Rademakers, R., Vandenberghe, R., Dermaut, B., Martin, J.J. et al. (2006) reaction Null mutations in progranulin cause ubiquitin-positive frontotemporal dementia linked to chromosome 17q21. Nature, 442, 920–924. Quantitative PCR (qPCR) was performed using an HT7900 ana- 5. Gass, J., Lee, W.C., Cook, C., Finch, N., Stetler, C., Jansen-West, K., Lewis, lyzer (Applied Biosystems, Carlsbad, CA, USA). Each 5 ml re- J., Link, C.D., Rademakers, R., Nykjaer, A. et al.. (2012) Progranulin action contained: 2 ml of pre-diluted (1:50) cDNA, 0.5 ml regulates neuronal outgrowth independent of Sortilin. Mol. Neurodegener., primer mix (1 mM for each primer) and 2.5 ml SYBR green 7, 33. (Invitrogen). The thermal cycling conditions were as follows: 6. He, Z., Ong, C.H., Halper, J. and Bateman, A. (2003) Progranulin is a mediator of the wound response. Nat. Med., 9, 225–229. 508C for 2 min, 958C for 2 min, 40 cycles of 958C for 15 s and 7. Kessenbrock, K., Frohlich, L., Sixt, M., Lammermann, T., Pfister, H., 608C for 1 min. The mRNA levels of SORT1 and PGRN were Bateman, A., Belaaouaj, A., Ring, J., Ollert, M., Fassler, R. et al. (2008) normalized to GAPDH, an endogenous transcript control. The Proteinase 3 and neutrophil elastase enhance inflammation in mice by data were analyzed by using Software RQ Manager 1.2 inactivatingantiinflammatory progranulin. J. Clin. Invest., 118, 2438–2447. (Applied Biosystems). Primer sequences for amplification of 8. Gijselinck, I., Van Broeckhoven, C. and Cruts, M. (2008) Granulin ′ ′ ′ mutations associated with frontotemporal lobar degeneration and related PGRN were 5 CCCTTCTGGACAAATGGC 3 and 5 disorders: an update. Hum. Mutat., 29, 1373–1386. GGAAGTCCCTGAGACGGTAA 3 . Primer sequences for 9. Gass, J., Cannon, A., Mackenzie, I.R., Boeve, B., Baker, M., Adamson, J., ′ ′ GAPDH were 5 CTTTGGTATCGTGGAAGGACTCATG 3 Crook, R., Melquist, S., Kuntz, K., Petersen, R. et al. (2006) Mutations in ′ ′ and 5 CCAGTAGAGGCAGGGATGATGTTC 3 . Primer progranulin are a major cause of ubiquitin-positive frontotemporal lobar sequences for SORT1 were 5 GGCCTGTGGGTGTCCAA- degeneration. Hum. Mol. Genet., 15, 2988–3001. ′ ′ ′ GAA 3 and 5 GCAGGAGCCATTTGCATAGGTT 3 . 10. Pickering-Brown, S.M., Rollinson, S., Du Plessis, D., Morrison, K.E., Varma, A., Richardson, A.M., Neary, D., Snowden, J.S. and Mann, D.M. (2008) Frequency and clinical characteristics of progranulin mutation carriers in the Manchester frontotemporal lobar degeneration cohort: Statistical analysis comparison with patients with MAPT and no known mutations. Brain, 131, 721–731. Representative data of at least three independent experiments 11. Le Ber, I., van der Zee, J., Hannequin, D., Gijselinck, I., Campion, D., Puel, with reproducible results were shown. All statistical analysis M., Laquerriere, A., De Pooter, T., Camuzat, A., Van den Broeck, M. et al. was performed by unpaired student t-test. One-way analysis of (2007) Progranulin null mutations in both sporadic and familial variance (ANOVA) followed by Tukey’s post-test was applied frontotemporal dementia. Hum. Mutat., 28, 846–855. in multi-doses experiments. 12. Le Ber, I., Camuzat, A., Hannequin, D., Pasquier, F., Guedj, E., Rovelet-Lecrux, A., Hahn-Barma, V., van der Zee, J., Clot, F., Bakchine, S. et al. (2008) Phenotype variability in progranulin mutation carriers: a clinical, neuropsychological, imaging and genetic study. Brain, 131, SUPPLEMENTARY MATERIAL 732–746. 13. Bronner, I.F., Rizzu, P., Seelaar, H., van Mil, S.E., Anar, B., Azmani, A., Supplementary Material is available at HMG online. Donker Kaat, L., Rosso, S., Heutink, P. and van Swieten, J.C. (2007) Progranulin mutations in Dutch familial frontotemporal lobar degeneration. Eur. J. Hum. Genet., 15, 369–374. ACKNOWLEDGEMENTS 14. Huey, E.D., Grafman, J., Wassermann, E.M., Pietrini, P., Tierney, M.C., Ghetti, B., Spina, S., Baker, M., Hutton, M., Elder, J.W. et al. (2006) KO We thank Chad A. Cowan for providing the WT and SORT1 - Characteristics of frontotemporal dementia patients with a Progranulin hESC lines. mutation. Ann. Neurol., 60, 374–380. 15. Bruni,A.C.,Momeni, P.,Bernardi,L.,Tomaino,C.,Frangipane,F., Elder,J., Kawarai, T., Sato, C., Pradella, S., Wakutani, Y. et al. (2007) Heterogeneity Conflict of Interest statement. None declared. within a large kindred with frontotemporal dementia: a novel progranulin mutation. Neurology, 69, 140–147. 16. Shankaran, S.S., Capell, A., Hruscha, A.T., Fellerer, K., Neumann, M., FUNDING Schmid, B. and Haass, C. (2008) Missense mutations in the progranulin gene linked to frontotemporal lobar degeneration with ubiquitin-immunoreactive This work was supported by Mayo Clinic Foundation (L.P.), inclusions reduce progranulin production and secretion. J. Biol. Chem., 283, NIH R01 AG026251, R01 NS 063964-01, R01 NS077402, 1744–1753. 17. Capell, A., Liebscher, S., Fellerer, K., Brouwers, N., Willem, M., Lammich, R01 ES20395-01, R21 NS84528-01J (L.P.), NS065782 (R.R.), S., Gijselinck, I., Bittner, T., Carlson, A.M., Sasse, F. et al. (2011) Rescue of NIH R01 AG026251, R01 NS057553 (F.B.G.) and the Consor- progranulin deficiency associated with frontotemporal lobar degeneration tium for Frontotemporal dementia Research (R.R.) (F.B.G.). by alkalizing reagents and inhibition of vacuolar ATPase. J. Neurosci., 31, Funding to pay the Open Access publication charges for this 1885–1894. article was provided by the Mayo Clinic. 18. Cenik, B., Sephton, C.F., Dewey, C.M., Xian, X., Wei, S., Yu, K., Niu, W., Coppola, G., Coughlin, S.E., Lee, S.E. et al. (2011) Suberoylanilide hydroxamic acid (vorinostat) up-regulates progranulin transcription: rational therapeutic approach to frontotemporal dementia. J. Biol. Chem., REFERENCES 286, 16101–16108. 19. Hu, F., Padukkavidana, T., Vaegter, C.B., Brady, O.A., Zheng, Y., 1. Toh, H., Chitramuthu, B.P., Bennett, H.P. and Bateman, A. (2011) Structure, Mackenzie, I.R., Feldman, H.H., Nykjaer, A. and Strittmatter, S.M. (2010) function, and mechanism of progranulin; the brain and beyond. J. Mol. Sortilin-mediated endocytosis determines levels of the frontotemporal Neurosci., 45, 538–548. dementia protein, progranulin. Neuron, 68, 654–667. 2. Bateman, A.and Bennett,H.P.(2009) The granulingene family:fromcancer 20. Carrasquillo, M.M., Nicholson, A.M., Finch, N., Gibbs, J.R., Baker, M., to dementia. Bioessays, 31, 1245–1254. Rutherford, N.J., Hunter, T.A., DeJesus-Hernandez, M., Bisceglio, G.D., 3. Baker, M., Mackenzie, I.R., Pickering-Brown, S.M., Gass, J., Rademakers, Mackenzie, I.R. et al. (2010) Genome-wide screen identifies rs646776 near R., Lindholm, C., Snowden, J., Adamson, J., Sadovnick, A.D., Rollinson, S. sortilin as a regulator of progranulin levels in human plasma. Am. J. Hum. et al. (2006) Mutations in progranulin cause tau-negative frontotemporal dementia linked to chromosome 17. Nature, 442, 916–919. Genet., 87, 890–897. Downloaded from https://academic.oup.com/hmg/article/23/6/1467/733282 by DeepDyve user on 14 July 2022 1478 Human Molecular Genetics, 2014, Vol. 23, No. 6 21. De Muynck, L., Herdewyn, S., Beel, S., Scheveneels, W., Van Den Bosch, Progranulin functions as a neurotrophic factor to regulate neurite outgrowth L., Robberecht, W. and Van Damme, P. (2013) The neurotrophic properties and enhance neuronal survival. J. Cell Biol., 181, 37–41. of progranulin depend on the granulin E domain but do not require sortilin 35. Chitramuthu, B.P., Baranowski, D.C., Kay, D.G., Bateman, A. and Bennett, binding. Neurobiol. aging, 34, 2541–2547. H.P. (2010) Progranulin modulates zebrafish motoneuron development in 22. Almeida, S., Zhang, Z., Coppola, G., Mao, W., Futai, K., Karydas, A., vivo and rescues truncation defects associated with knockdown of Survival Geschwind, M.D., Tartaglia, M.C., Gao, F., Gianni, D. et al. (2012) Induced motor neuron 1. Mol. Neurodegener., 5, 41. pluripotent stem cell models of progranulin-deficient frontotemporal 36. Laird, A.S., Van Hoecke, A., De Muynck, L., Timmers, M., Van den Bosch, dementia uncover specific reversible neuronal defects. Cell Rep., 2, L., Van Damme, P. and Robberecht, W. (2010) Progranulin is neurotrophic 789–798. in vivo and protects against a mutant TDP-43 induced axonopathy. PLoS 23. Dimos, J.T., Rodolfa, K.T., Niakan, K.K., Weisenthal, L.M., Mitsumoto, H., One, 5, e13368. Chung, W., Croft, G.F., Saphier, G., Leibel, R., Goland, R. et al. (2008) 37. Keiser, M.J., Roth, B.L., Armbruster, B.N., Ernsberger, P., Irwin, J.J. and Induced pluripotent stem cells generated from patients with ALS can be Shoichet, B.K. (2007) Relating protein pharmacology by ligand chemistry. differentiated into motor neurons. Science, 321, 1218–1221. Nat. biotechnol., 25, 197–206. 24. Israel, M.A., Yuan, S.H., Bardy, C., Reyna, S.M., Mu, Y., Herrera, C., 38. Wells, J.A. and McClendon, C.L. (2007) Reaching for high-hanging fruit in Hefferan, M.P., Van Gorp, S., Nazor, K.L., Boscolo, F.S. et al. (2012) drug discovery at protein–protein interfaces. Nature, 450, 1001–1009. Probingsporadicand familialAlzheimer’s disease usinginducedpluripotent 39. Meyer-Hoffert, U. (2009) Neutrophil-derived serine proteases modulate stem cells. Nature, 482, 216–220. innate immune responses. Front. Biosci., 14, 3409–3418. 25. Nguyen, H.N., Byers, B., Cord, B., Shcheglovitov, A., Byrne, J., Gujar, P., 40. Morinville, A., Martin, S., Lavallee, M., Vincent, J.P., Beaudet, A. and Kee, K., Schule, B., Dolmetsch, R.E., Langston, W. et al. (2011) LRRK2 Mazella, J. (2004) Internalization and trafficking of neurotensin via NTS3 mutant iPSC-derived DA neurons demonstrate increased susceptibility to receptors in HT29 cells. Int. J. Biochem. Cell Biol., 36, 2153–2168. oxidative stress. Cell Stem Cell, 8, 267–280. 41. Lefrancois, S., Zeng, J., Hassan, A.J., Canuel, M. and Morales, C.R. (2003) 26. Zheng,Y.,Brady,O.A.,Meng,P.S.,Mao,Y.andHu, F.(2011)C-terminus of The lysosomal trafficking of sphingolipid activator proteins (SAPs) is progranulin interacts with the beta-propeller region of sortilin to regulate mediated by sortilin. EMBO J., 22, 6430–6437. progranulin trafficking. PLoS One, 6, e21023. 42. Ni, X., Canuel, M. and Morales, C.R. (2006) The sorting and trafficking of 27. Quistgaard, E.M., Madsen, P., Groftehauge, M.K., Nissen, P., Petersen, lysosomal proteins. Histol. histopathol., 21, 899–913. C.M. and Thirup, S.S. (2009) Ligands bind to Sortilin in the tunnel of a 43. Canuel, M., Korkidakis, A., Konnyu, K. and Morales, C.R. (2008) Sortilin ten-bladed b-propeller domain. Nat. Struct. Mol. Biol., 16, 96–98. mediates the lysosomal targeting of cathepsins D and H. Biochem. Biophys. 28. Mazella, J., Chabry, J., Zsurger, N. and Vincent, J.P. (1989) Purification of Res. Commun., 373, 292–297. the neurotensin receptor from mouse brain by affinity chromatography. 44. Smith, K.R., Damiano, J., Franceschetti, S., Carpenter, S., Canafoglia, L., J. Biol. Chem., 264, 5559–5563. Morbin, M., Rossi, G., Pareyson, D., Mole, S.E., Staropoli, J.F. et al. (2012) 29. Mazella, J., Zsurger, N., Navarro, V., Chabry, J., Kaghad, M., Caput, D., Strikingly different clinicopathological phenotypes determined by Ferrara, P., Vita, N., Gully, D., Maffrand, J.P. et al. (1998) The 100-kDa progranulin-mutation dosage. Am. J. Hum. Genetics, 90, 1102–1107. neurotensin receptor is gp95/sortilin, a non-G-protein-coupled receptor. 45. Gieselmann, V. (2005) What can cell biology tell us about heterogeneity in J. Biol. Chem., 273, 26273–26276. lysosomal storage diseases? Acta Paediatr. Suppl., 94, 80–86; discussion 30. Nielsen, M.S.,Jacobsen,C., Olivecrona, G., Gliemann, J. and Petersen, C.M. (1999) Sortilin/neurotensin receptor-3 binds and mediates degradation of 46. Serrero, G. (2003) Autocrine growth factor revisited: PC-cell-derived lipoprotein lipase. J. Biol. Chem., 274, 8832–8836. growth factor (progranulin), a critical player in breast cancer tumorigenesis. 31. Petersen, C.M., Nielsen, M.S., Nykjaer, A., Jacobsen, L., Tommerup, N., Biochem. Biophys. Res. Commun., 308, 409–413. Rasmussen, H.H., Roigaard, H., Gliemann, J., Madsen, P. and Moestrup, 47. Ho, J.C., Ip, Y.C., Cheung, S.T., Lee, Y.T., Chan, K.F., Wong, S.Y. and Fan, S.K. (1997) Molecular identification of a novel candidate sorting receptor S.T. (2008) Granulin-epithelin precursor as a therapeutic target for purified from human brain by receptor-associated protein affinity hepatocellular carcinoma. Hepatology, 47, 1524–1532. chromatography. J. Biol. Chem., 272, 3599–3605. 48. Kamrava, M., Simpkins, F., Alejandro, E., Michener, C., Meltzer, E. and 32. Ding, Q., Lee, Y.K., Schaefer, E.A., Peters, D.T., Veres, A., Kim, K., Kohn, E.C. (2005) Lysophosphatidic acid and endothelin-induced Kuperwasser, N., Motola, D.L., Meissner, T.B., Hendriks, W.T. et al. (2013) proliferation of ovarian cancer cell lines is mitigated by neutralization of A TALEN genome-editing system for generating human stem cell-based granulin-epithelin precursor (GEP), a prosurvival factor for ovarian cancer. disease models. Cell Stem Cell, 12, 238–251. Oncogene, 24, 7084–7093. 33. Ryan, C.L., Baranowski, D.C., Chitramuthu, B.P., Malik, S., Li, Z., Cao, M., 49. Prudencio, M., Jansen-West, K.R., Lee, W.C., Gendron, T.F., Zhang, Y.J., Minotti, S., Durham, H.D., Kay, D.G., Shaw, C.A. et al.. (2009) Progranulin Xu, Y.F., Gass, J., Stuani, C., Stetler, C., Rademakers, R. et al. (2012) is expressedwithinmotor neuronsand promotesneuronalcell survival.BMC Neurosci., 10, 130. Misregulation of human sortilin splicing leads to the generation of a 34. Van Damme, P., Van Hoecke, A., Lambrechts, D., Vanacker, P., Bogaert, E., nonfunctional progranulin receptor. Proc. Natl. Acad. Sci. USA, 109, van Swieten,J., Carmeliet, P., VanDen Bosch,L.and Robberecht,W. (2008) 21510–21515.

Journal

Human Molecular Genetics – Oxford University Press

Published: Mar 15, 2014

There are no references for this article.