In vitro cleavage assay
For in vitro cleavage assays, 400ng of recombinant human progranulin (R&D #CF-2420) was incubated with or without 1µM of each protease. Depending on the pH setpoint, the following buffers were used: 100mM sodium citrate pH 3.4, 50mM sodium acetate pH 4.5 or 5.5, 50mM 2-(N-morpholino)-ethanesulfonic acid (MES) buffer pH 6.5, 100mM phosphate buffer saline (PBS) pH 7.4 with 1mM EDTA and 2mM DTT for 20 or 60 minutes at 37ºC water bath. The cleavage was performed in a total volume to 19.5µl. Protease activity was stopped by adding 7.5µl of NuPAGE 4X LDS (Fisher #NP0007), 3µl of 10X reducing agent (i.e. 50µM) (Fisher #NP0009) and denatured for 10 minutes at 70ºC. Of note, we used these denaturing conditions to reduce the potential for dimerization of fragments (a known possibility with PGRN) [21]. The samples were run on precast NOVEX 4-12% Bis-Tris gels (Fisher #NP0321PK2) using MES buffer (Fisher #NP0002). The gel was then either fixed in 40% ethanol and 10% acetic acid for silver and coomaisse staining or transferred onto nitrocellulose membranes for western blotting analysis.
Silver stain - Silver staining was performed according to manufacturer’s instructions with SilverQuest silver staining kit (Thermo #LC6070).
Coomassie stain - Coomassie staining was done using BIO-RAD QC colloidal coomaisse stain (Bio-rad #160-0803). The gel was stained overnight and was removed by washing in milli-Q water for 3 hours.
Recombinant proteases - Cathepsin E (R&D #1294-AS), Cathepsin D (R&D #1014-AS), Napsin A (R&D #8489-NA), Cathepsin G (Millipore #219873), Cathepsin A (R&D #1049-SE), Pro-X-carboxypeptidase (R&D #7164-SE-010), Cathepsin L (Millipore #219402), Cathepsin B (Millipore #219364), Cathepsin K (Millipore #219461), Cathepsin S (R&D #1183-CY), Cathepsin V (R&D #1080-CY), Asparagine endopeptidase/legumain (R&D #2199-CY), Cathepsin H (R&D #7516-CY-010), Cathepsin C (R&D #1071-CY), Cathepsin O (Abcam #ab267932), Cathepsin F (Abcam #ab240858), and Cathepsin X (R&D #934-CY).
Protease activation and confirmation of activity – Cathepsins E, D, G, L, B, K, S, V, F, and O as well as Asparagine endopeptidase, Napsin A, and Pro-X-carboxypeptidase require no activation, according to the vendor, and were used as purchased. Cathepsin A, C, and H require pre-activation before use. For cathepsin A and cathepsin C activation, 1µM of each protease was incubated with 200nM of cathepsin L at room temperature for 1 hour. After 1 hour, 50µM of benzyloxycarbonyl FY(t-Bu)-DMK, an irreversible, highly specific inhibitor of cathepsin L (Sigma # 219427) was added to quench cathepsin L activity. Once activated, cathepsin A and C were used in the experiments outlined. Similarly, for cathepsin H activation, 1µM of cathepsin H was incubated with 500nM of Thermolysin (R&D # 3097-ZN) at room temperature for 3 hours. After 3 hours, 1mM of Phosphoramidon (Tocris # 6333), a specific thermolysin inhibitor, was added to quench Thermolysin activity. Once activated, cathepsin H was used in the experiments outlined. To confirm each protease was enzymatically active, protease activity control experiments were performed using a FRET-based fluorogenically labeled casein assay (Biovision #K781), according to the vendor's protocol. In brief, 1µM of each protease was added to a 50µL reaction mixture using the recommended concentration of control fluorogenic substrate in a buffer consisting of 50mM sodium acetate buffer at pH 4.5 and in the presence of 1mM EDTA and 2mM DTT. The assay was performed at 37ºC for 60 minutes and readings were taken every 2 minutes using Tecan Infinite M Plex plate reader.
Statistical analysis
Details of the statistical test used for each experiment is in figure legends along with n number and p value. All data is represented as mean ± SD. Statistical analysis was performed using GraphPad Prism 8 (GraphPad Software, La Jolla, California USA). A P-value < 0.05 was considered significant.
Cell culture, treatment, and lysis
Cell culture - SH-SY5Y human neuroblastoma cells were obtained from ATCC (CRL-2266) and maintained 1:1 EMEM/F12 media (ATCC #30-2003/Thermo #11765062) supplemented with 10% FBS and 1% Penicillin-Streptomycin (Thermo #15140122). SH-SY5Y cells were differentiated by treating with 10µM all-trans retinoic acid (Sigma #R2625) for 6 days in EMEM/F12 media supplemented with 10% FBS and 1% Penicillin-Streptomycin, followed by 4 days of treatment with 50ng/mL of BDNF (Peprotech #450-02B) in EMEM/F12 media supplemented with only 1% Penicillin-Streptomycin. HEK293FT were maintained in high glucose DMEM media with 10%FBS and 1% Penicillin-Streptomycin.
Cell treatments - For protease inhibition studies, day 10 post-differentiation SH-SY5Y cells were treated with 10µM E64D (Tocris Bioscience #4545), 20µM cathepsin L inhibitor II (Santa Cruz #sc-3132), and 15µM of CA074_methyl ester (Caymen Chemical Company #18469) or DMSO for 24 hours. For siRNA studies, day 10 post-differentiation SH-SY5Y cells were transfected according to the manufacturer’s protocol using Oligofectamine (Invitrogen #12252011). Sixty hours post transfection, cells were collected for western blot analysis, as described below. siRNA scramble (Thermo Fisher #4390843), siRNA Legumain (Santa Cruz #sc-60930) were used. For overexpression studies, pCMV6 mammalian expressing vectors carrying the cDNA of AEP were ordered from Origene (#RC224975). XtremeGene HP (Sigma, # 06366236001) was used to transiently transfect HEK293FT cells for 24hours following the manufactures suggestions. The cells were then pelleted and prepared for western blotting as described below.
Cell Lysis - The cells were collected post-treatment and prepared for western blot analysis. Cells were washed once with PBS, trypsinized, and pelleted. The pellets were lysed in RIPA buffer (Fisher #89900) with protease and phosphatase inhibitors (Sigma #4693124001, #4906837001) and centrifuged at 15000 rpm at 4ºC for 20 minutes. The supernatant containing the soluble proteins were transferred into a new tube. Estimated protein concentration with a Peirce BCA (Fisher # PI23225) protein assay kit following manufacturer’s instructions. The nitrocellulose membrane was blocked with 5% milk (Fisher #NC9121673) or Odyssey buffer (Li-cor #927-50010) for 1-2hrs. The membranes were cut between 60-50kD and between 20-15kD to obtain PGRN and granulin-specific bands respectively and incubated with primary antibodies overnight [19]. Li-cor secondary antibodies at 1:5000 dilution was used and imaged using Odyssey CLx imager. Western blots were quantified using FIJI software.
Cloning CRISPR guide and Viral packaging
Using a protocol similar to Sanjana et, 2014, CRISPR guide (sgRNA), GGTCCAGGCAGCAGGCCACAGGG targeting the PGRN gene was cloned into pLentiV2 (Addgene, #52961) at the BsmBI site and confirmed by sequencing [22]. For viral production, 3x106 HEK293FT cells were seeded in a 10cm dish. 24 hours post seeding, cells were transfected with either pLentiV2 with guide sequence or only the backbone (mock), psPax2 (Addgene, #12260) and pCMV- VSV-G (Addgene, #8454) using XtremeGene HP (Sigma, # 06366236001) following manufacturer’s guidelines. 12-16 hours post transfection, fresh media was added. 48 hours post transfection, virus containing media was collected and filtered using a 0.45µM syringe filter.
Generation of PGRN knock-out SH-SY5Y line
To generate a PGRN KO, Human SH-SY5Y cells were infected with the virus and polybrene (4µg/mL) in 6 well plates. 24 hours post infection, cells were selected using 1µg/mL puromycin. To select the PGRN KO SY-SY5Y cells, single clones were isolated and genomic DNA from each clonal population was extracted with the QuickExtract kit (Epicentre). The region of interest was amplified from genomic DNA by PCR using primers (forward -TGTAAAACGACGGCCAGTGGGCTAGGGTACTGAGTGAC
and reverse -CAGGAAACAGCTATGACCCTGATGGCACGCTCACCTCT) and sequenced to determine the mutation. Cell lysates from each clone was extracted (see below) and PGRN and granulin levels were determined by western blots. With this method, a successful KO clone was identified and validated. These cells were then expanded and frozen for the experiments outlined.
Antibodies
Antibody generation - The following epitopes were used to generate the rabbit polyclonal anti-bodies:
paragranulin (p-Gran) -TRCPDGQFCPVACCLDPGGASYSCCRPLLD,
granulin F (Gran-F) - QCPDSQFECPDEST, and
granulin E (Gran-E) - ECGEGHFCHDNQTCCR.
Antibody validation – The sensitivity, specificity, and reproducibility of results of the novel p-Gran, Gran-F, and Gran-E antibodies were validated using a protocol similar to Bordeaux, et al. 2010 [23]. In brief, antibody specificity for each antigen was first measured by dot blot. For these experiments, 2.5µg of each granulin peptide was pipetted onto a nitrocellulose membrane defined into grids. Each antibody was incubated with the membrane overnight at 1:1000 dilution. Signals from the primary antibodies were amplified using species-specific antisera conjugated with horseradish peroxidase (Cell Signaling #7074) and detected with a chemiluminescent substrate detection system ECL and scanned the images with a densitometer. Following successful confirmation of antibody sensitivity, the specificity was evaluated via a peptide blocking experiment. Recombinant human progranulin was incubated with CTSL, a known PGRN protease, at pH 4.5 for 20 minutes at 37C as previously described. Samples were denatured and western blots were performed for peptide blocking analysis. Each primary antibody (p-Gran, Gran-F, Gran-E) was pre-incubated with its respective granulin peptide for 2 hours at room temperature with gentle mixing. Each primary antibody (p-Gran, Gran-F, Gran-E) was pre-incubated with its respective granulin peptide for 2 hours at room temperature with gentle mixing. Optimal primary antibody concentration was incubated with two different peptide amounts and on its own. Primary antibody to peptide ratio was determined by molarity at 10X, 100X, and no peptide. Following incubation, solutions were used to blot membranes overnight at 4C. Normal western blot procedure was completed as described.
Commercial antibodies - Anti-progranulin C-terminus (Thermo Fisher #40-3400, 1:200), Anti-granulin F (Sigma #HPA008763, 1:250), Anti-AEP (R&D #AF2199, 1:500), anti-Actin (Sigma #MAB1501R, 1:5000), and anti-FLAG (Sigma #F3165)
Human brain tissue lysis and analysis
Brain tissue was prepared as previously described in Salazar et al, 2015 [12]. Briefly, adjacent tissue blocks were fixed, embedded in paraffin wax, sectioned, stained for hematoxylin and eosin, and rated for astrogliosis (0–3 scale). Absent or low gliosis regions were selected with a score of 0-1, while high levels of gliosis regions with a score of 2-3. From this, we selected the middle frontal gyrus (MFG) with severe degeneration and inferior occipital cortex (IOC) with no degeneration for all assays. The same region was collected from control subjects. Brain tissue samples were weighed, diluted 25-fold with PBS (Thermo Fisher #14190250) and 1% triton-X (Millipore Sigma, #T9284), homogenized for 1 minute (pestle pellet motor) and sonicated for 5 rounds of 30 seconds on and 1 minute off (BioRuptor, Diagenode). The lysate was centrifuged at max for 20 minutes and the soluble supernatant was used for western blot assays. BCA was performed to determine protein concentration according to manufacturer’s instructions. 25µg of total protein from each sample was loaded for western blot analysis. The nitrocellulose membrane was blocked with 5% milk (Fisher #NC9121673) or Odyssey buffer (Li-cor, #927-50010) for 1-2hrs. The membranes were cut between 60-50kD and between 20-15kD to obtain PGRN and granulin-specific bands respectively and incubated with primary antibodies overnight [19]. Li-cor secondary antibodies at 1:5000 dilution was used and imaged using Odyssey CLx imager. Western blots were quantified using FIJI software.
qPCR
Differentiated SH-SY5Y cells were pelleted and RNA was extracted using standard phenol:chloroform extraction techniques. 1µg of RNA was reverse transcribed to cDNA using Superscript III reverse transcription kit following manufacturer’s protocol (Thermo Fisher #18080044), random primers (Thermo Fisher #48190011) and dNTPs (Sigma #11969064001). RT-qPCR was performed using TaqMan Fast Advanced Master Mix (Thermo Fisher #4444557) in a ABI Prism 7900HT Sequence Detection System (Applied Biosystems), with TaqMan FAM probes for human CTSB (Hs00947433_m1), CTSL (Hs009964650_m1), CTSK (Hs00166156_m1), CTSS (Hs00175407_m1), AEP/LGMN (Hs00271599_m1), CTSV (Hs00426731_m1), CTSE (Hs00157213_m1), CTSG (Hs01113415_g1) and the house keeping gene GAPDH (Hs02758991_g1) as a control. Four biological samples were analyzed, each with three technical replicates. mRNA levels of target genes were normalized to the mean of the house keeping gene GAPDH. Data is displayed as relative values compared to GAPDH.
Enzyme activity assays
AEP activity assay - AEP substrate Z-Ala-Ala-Asn-AMC (Bachem #I1865) was used to measure activity. 50mM sodium acetate buffer pH 5.5 with 2mM DTT, 1mM EDTA and 0.1% CHAPS was used as the assay buffer. 100µM of the substrate was used per reaction. The assay was performed at 37ºC for 60 minutes and readings were taken every 2 minutes using Tecan Infinite M Plex plate reader. Activity was measured using excitation wavelength 380nm and emission wavelength 460nm. Assays were performed in triplicates for each sample.
CTSL activity assay - Biovision CTSL fluorometric assay kit (#K142) was used according to manufacturer’s protocol with some modifications. Since CTSB can also cleave the assay substrate, the assay was performed with 1mM CTSB inhibitor (CA074_ME, Caymen Chemical Company #18469) to specifically assay CTSL activity. The assay was performed at 37ºC for 60 minutes and reading were taken every 2 minutes using a Tecan Infinite M Plex plate reader. Activity was measured using excitation wavelength 400nm and emission wavelength 505nm. This assay was performed in triplicates for each sample.
Identification of AEP cleavage sites
In-gel Digestion - 400 ng of recombinant human progranulin (R&D #CF-2420) was incubated at 37ºC for 60 minutes with recombinant human AEP (R&D #2199-CY) at pH 4.5. The cleavage was stopped by adding 7.5µl of NuPAGE 4X LDS (Fisher #NP0007), 3µl of 10X reducing agent (Fisher #NP0009) and denatured for 10 minutes at 70ºC. The samples were run on precast NOVEX 4-12% Bis-Tris gels (Fisher #NP0321PK2) using MES buffer (Fisher #NP0002). The gel was fixed, and silver stained as described above. Protein bands were excised and digested in-gel with Endoproteinase Asp-N (Sigma-Aldrich #11054589001) as described previously [24, 25]. The extracted digests were vacuum-evaporated, resuspended in 20µl of 0.1% formic acid, and desalted using C18 ZipTips (Millipore #ZTC18M096).
Mass Spectrometry - Asp-N peptides were analyzed by on-line LC-MS/MS using an Orbitrap Fusion Lumos (Thermo Scientific, San Jose, CA) coupled with a NanoAcquity M-Class UPLC system (Waters, Milford, MA). Peptides were separated over a 15cm x 75μm ID 3μm C18 EASY-Spray column (Thermo Scientific #ES800). Precursor ions were measured from 375 to 1500 m/z in the Orbitrap analyzer (resolution: 120,000; AGC: 4.0e5). Ions charged 2+ to 7+ were isolated in the quadrupole (selection window: 1.6 m/z units; dynamic exclusion window: 30 s; MIPS Peptide filter enabled), fragmented by HCD (Normalized Collision Energy: 30%) and measured in the Orbitrap (resolution: 30,000; AGC; 5.0e4). The cycle time was 3 seconds.
Peaklists were generated using PAVA (UCSF) and searched using Protein Prospector 5.23.0 against the SwissProt database (downloaded 9/6/2016) and a randomized concatenated database. Cleavage specificity was set as Asp-N/AEP (Asn-C) allowing 2 mis-cleavages. Carbamidomethylation of Cys was set as a constant modification and two of the following variable modifications were allowed per peptide: acetylation of protein N-termini, oxidation of Met, oxidation and acetylation of protein N-terminal Met, cyclization of N-terminal Gln, protein N-terminal Met loss, protein N-terminal Met loss and acetylation. Precursor mass tolerance was 20ppm and fragment mass tolerance was 30ppm. A subsequent search using the above parameters but limiting the search to the following accession numbers and a user defined PGRN amino acid sequence (the human sequence minus the first 18 amino acids and with a C-terminal 6xHis tag) was used for further analysis: B2FQP3 O08692 P02533 P02666 P02754 P02769 P04264 P28799 P35527 P35908 P85945 Q8PC00 Q91FI1 Q9R4J4 P07711 Q99538. Spectra containing AEP cleavage sites can be viewed in MS-Viewer with the search key “besr1qb6q4” or the following link: http://msviewer.ucsf.edu/prospector/cgi-bin/mssearch.cgi?report_title=MS-Viewer&search_key=besr1qb6q4&search_name=msviewer [26].