Ablation of progranulin augments microglial activation and accelerates prion progression

Abstract

Mutations or polymorphisms in GRN, encoding the CNS glycoprotein progranulin (PGRN), have been linked to several neurodegenerative diseases. In this study, we explored the role of PGRN in prion diseases. We observed that prion infection upregulated microglial PGRN expression. Following intracerebral inoculation with RML6 prions, Grn^-/- mice exhibited accelerated disease progression compared to Grn^+/- and Grn^+/+ littermates. Histological analysis revealed augmented microglial activation in Grn^-/- mice. Temporal analysis revealed enhanced early microglial activation and prion clearance at 120 dpi, followed by excessive complement activation but inadequate clearance by 150 dpi. Additionally, Grn^-/- brains exhibited exacerbated astrogliosis and vacuolation. RNA-seq analysis indicated that complete PGRN deficiency in prion-infected mice shifted microglia from homeostatic to pro-inflammatory states. Notably, microglia-specific depletion of PGRN did not affect prion pathogenesis, suggesting that PGRN deficiency affects microglial activation and prion progression in a non-cell autonomous manner. These findings suggest that microglia respond to prion infection in a stepwise manner, and PGRN plays a critical role in modulating prion-induced microglial activation. Our results highlight the neuroprotective role of PGRN in prion disease and suggest that supplementation or boosting expression of PGRN could represent a promising therapeutic strategy.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40478-025-02128-3.

Introduction

Progranulin (PGRN), encoded by the GRN gene, is a glycoprotein involved in diverse biological processes, including development, wound healing, and tumorigenesis [1]. In the central nervous system (CNS), PGRN is predominantly expressed by microglia and neurons, where it functions as a neurotrophic factor to promote neuronal development, survival and neurite growth [2, 3]. In 2006, several independent studies identified that loss-of-function mutations in GRN gene, leading to haploinsufficiency, cause frontotemporal lobar degeneration (FTLD) with tau-negative, ubiquitin-positive inclusions [4–6]. These inclusions were later found to contain transactivation response DNA binding protein-43 (TDP-43) [7, 8]. PGRN is not only secreted but also localizes to lysosomes, where it plays a critical role in lysosomal functions [9, 10]. Homozygous GRN mutations have been linked to lysosomal storage disorders, such as neuronal ceroid lipofuscinosis (NCLs) [11–16]. Furthermore, genome-wide association studies (GWAS) have identified GRN variants as risk factors for a broad spectrum of neurodegenerative diseases, including Alzheimer’s disease (AD) [17–26], Parkinson’s disease (PD) [27], amyotrophic lateral sclerosis (ALS) [28], corticobasal syndrome (CBS) [29, 30], limbic-predominant aging-related TDP-43 encephalopathy (LATE) [31], and Lewy body dementia [32]. However, the mechanisms by which GRN mutations or variants contribute to neurodegeneration remain poorly understood.

Grn knockout mice have been widely used to dissect PGRN’s functions and the mechanisms underlying PGRN deficiency-associated neurodegeneration [33–37]. While Grn^+/- mice are neuropathologically indistinguishable from wild type mice, Grn^-/- mice consistently develop microglial activation, astrogliosis, neuronal loss and accelerated lipofuscinosis with aging [37, 38]. These findings suggest that PGRN acts as a brake, suppressing aberrant microglial activation and controlling neuroinflammation during aging or under pathological conditions [34, 37]. Notably, ubiquitinated and phosphorylated TDP-43 aggregates are also observed in aged Grn^-/- mice [36, 37]. Mechanistically, PGRN is implicated in the autophagy-lysosomal pathway, and its deficiency may impair autophagy and lysosomal homeostasis, leading to TDP-43 accumulation [39, 40]. Single-nucleus RNA-sequencing (snRNA-seq) of Grn^-/- mouse brains revealed that microglia are the first cells to exhibit transcriptomic changes, transitioning from a homeostatic to a disease-associated phenotype [41, 42]. Additionally, PGRN deficiency in microglia have been linked to altered lipid metabolism [43–46]. Recent studies using snRNA-seq have also identified neurovascular dysfunction in FTD-GRN patients [47]. In animal models of neurodegenerative disease, PGRN deficiency exacerbates MPTP-induced dopaminergic neuronal loss and accelerates tau deposition and phosphorylation in human tau-expressing mice [34, 48, 49]. Interestingly, in mouse models of AD, PGRN deficiency has been reported to either increase Aβ deposition [50, 51] or decrease diffuse Aβ plaques [49, 52]. The reasons for these discrepancies remain unclear but may be attributed to differences in mouse models and/or the ages of animals studied.

Prion diseases are a group of fatal neurodegenerative disorders affecting both animals and humans [53]. Prion diseases remain incurable, with pathological hallmarks including spongiform changes, deposition of scrapie prion protein (PrP^Sc), and prominent microglial activation [54]. Microglia play an overall neuroprotective role in prion diseases by clearing prions in the brain [55]. Microglia-derived triggering receptor expressed on myeloid cells 2 (TREM2) has been implicated in prion-induced microglial activation [56], but the molecular mechanisms underlying microglial activation and prion clearance remain to be dissected. Autophagic and lysosomal abnormalities have been observed in both patients and animal models of prion diseases [57, 58], yet the contribution of impaired autophagy and lysosomal dysfunction to prion-induced neurodegeneration is not fully elucidated. Given PGRN’s involvement in various neurodegenerative disorders, particularly through its regulation of autophagy and lysosomal functions in microglia, we sought to determine whether PGRN plays a role in prion pathogenesis.

In this study, we first assessed PGRN expression in prion-infected mouse brains and found that it was significantly upregulated in microglia following prion infection. Prion inoculation experiments revealed that complete PGRN deficiency (Grn^-/-) accelerated disease progression compared to heterozygous (Grn^+/-) and wild-type (Grn^+/+) littermates. Histological analysis demonstrated that Grn^-/- microglia exhibited aberrant activation, adopting an amoeboid morphology and displaying altered cytokine profiles. At 120 days post-inoculation (dpi) of prion, Grn^-/- microglia were more activated and efficient at clearing prions, resulting in reduced prion deposition. However, by 150 dpi, overactivated Grn^-/- microglia triggered excessive complement activation and insufficient prion clearance, leading to PrP^Sc levels comparable to those in Grn^+/- and wild-type littermates. RNA-seq analysis revealed that complete PGRN deficiency in prion-infected mice shifted microglia from a homeostatic to a disease-associated phenotype. Interestingly, microglia-specific depletion of Grn did not affect prion pathogenesis. These findings suggest that microglia respond to prion infection in a stepwise manner, and complete PGRN deficiency sensitizes mice to prion-induced neurodegeneration. PGRN restricts prion-induced microglial activation and protects against prion diseases, partly by suppressing excessive activation of complement cascade.

Materials and methods

Ethical statement

All animal experiments were carried out in strict accordance with the Rules and Regulations for the Protection of Animal Rights (Tierschutzgesetz and Tierschutzverordnung) of the Swiss Bundesamt für Lebensmittelsicherheit und Veterinärwesen and were preemptively approved by the Animal Welfare Committee of the Canton of Zürich (permit ZH040/2015).

Animals

Grn^-/- mice [33] carrying a targeted deletion of exons 2–13 were first backcrossed to C57BL/6J to obtain Grn^+/- offspring. Grn^+/- mice were then intercrossed to obtain Grn^-/-, Grn^+/- and Grn^+/+ (WT) littermates, which were used for the experiments described here. CX3CR1-CreERT2 mice expressing tamoxifen inducible Cre under endogenous CX3CR1 promoter were purchased from Jackson’s lab (JAX Stock #021160) [59]. Grn^fl/fl mice with the whole coding sequence floxed [34] were crossed to CX3CR1-CreERT2 mice to generate CX3CR1-CreERT2; Grn^fl/fl mice. Prnp^-/- mice [60] kept on a C57BL/6J background were used as prion-resistant controls.

Mice were maintained in high hygienic grade facility and housed in groups of 3–5, under a 12 h light/12 h dark cycle (from 7 am to 7 pm) at 21 ± 1 °C, with sterilized food (Kliba No. 3242, Provimi Kliba, Kaiseraugst, Switzerland) and water ad libitum.

Generation of Grn^-/- BV2 cells

Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM; MeilunBio, MA0212) supplemented with 10% fetal bovine serum (FBS; MeilunBio, PWL217), 1% L-glutamine (MeilunBio, PWL031), 1% penicillin-streptomycin (MeilunBio, MA0110) at 37 °C with 5% CO₂.

To generate Grn^-/- BV2 cells, single guide RNAs (sgRNAs) targeting the mouse Grn gene were designed using CRISPOR (http://crispor.tefor.net/). The sgRNA oligonucleotides (forward: 5´-CAC CGC TGG CTG GCC TTC GCG GCA GT-3´, reverse༚5´-TAA AAC TGC CGC GAA GGC CAG CCA GC-3´) were synthesized and cloned into the LentiV2 CRISPR/Cas9 vector. Lentiviral particles were produced in HEK293T cells by co-transfecting LentiV2-Grn plasmid (6 µg) with the packaging plasmids PMD2.G (4 µg) and pSPAX2 (3 µg), and applied to BV2 cells that had been pre-seeded in 6-well plates. 48 h post-transduction, the culture medium was replaced with complete medium containing puromycin (2 µg/mL) for selection for 5 days. Successful generation of Grn^-/- BV2 cells were verified by DNA sequencing of the Grn target site and Western blot.

Intracerebral prion inoculation

Grn^-/- (8 female, 10 male), Grn^+/- (14 female, 11 male) and Grn^+/+ (WT) (11 female, 4 male) mice at 1.5-2 months old were intracerebrally (i.c) inoculated with 30 µl of 0.1% (w/v) brain homogenate diluted in PBS with 5% BSA and containing 3 × 10⁵ LD50 units of the RML6. For CX3CR1-CreERT2; Grn^fl/fl mice (13 female, 12 male) and Grn^fl/fl controls (12 female, 13 male), mice at 1 month old were first fed with Tamoxifen diet (ENVIGO, TD.55125IC.I Tam400/CreER, Tamoxifen Citrate 400 ppm) for 4 weeks, followed by normal food (Kliba No. 3242) to rest for 2 weeks. CX3CR1-CreERT2; Grn^fl/fl mice (12 female, 12 male) and Grn^fl/fl controls (13 female, 11 male) without Tamoxifen diet feeding were also used for control. Then mice were i.c inoculated with 30 µl RML6. Scrapie was diagnosed according to clinical criteria (ataxia, kyphosis, priapism, and hind leg paresis) as previously described [61]. Briefly, mice were observed every other day after RML6 inoculation for clinical signs including gait, grooming, activity, rough hair coat, limb paresis and ataxia. Once the mice showed the first sign of scrapie (grade 1: wadding gait, mild signs of reduced grooming, rough hair coat, limb weakness, front leg paresis, etc.), they were monitored every day and wet food was supplied in the cage. When the mice reached score grade 2 (ataxia, reduced grooming and activity, paralysis and rolling, etc.) that hamper the mice reaching water bottle, they were defined as terminally sick. Mice were sacrificed at various time points for analysis at pre-clinical and clinical stages, and on the day of onset of terminal clinical signs of scrapie. Non-infectious brain homogenates (NBH) prepared from uninfected mice and diluted with the same concentration (0.1%) as RML6 were used as negative controls.

Quantitative real-time PCR (qRT-PCR)

Total RNA from brain was extracted using TRIzol (Invitrogen Life Technologies) according to the manufacturer’s instruction. The quality of RNA was analyzed by Bioanalyzer 2100 (Agilent Technologies), RNAs with RNA integrity number > 7 were used for cDNA synthesis. cDNA was synthesized from ~ 1 µg total RNA using QuantiTect Reverse Transcription kit (QIAGEN) according to the manufacturer’s instruction. Quantitative real-time PCR (qRT-PCR) was performed using the SYBR Green PCR Master Mix (Roche) on a ViiA7 Real-Time PCR system (Applied Biosystems).

The following primer pairs were used: Gapdh sense 5´-CCA CCC CAG CAA GGA GAC T-3´; antisense, 5´-GAA ATT GTG AGG GAG ATG CT-3´. Grn sense 5´- GTG TTG TGA GGA TCA CAT TC -3´; antisense, 5´- CTA TGA CCT TCT TCA TCC AG -3´. Tnfα sense, 5´-CAT CTT CTC AAA ATT CGA GTG ACA A-3´; antisense, 5´-TGG GAG TAG ACA AGG TAC AAC CC-3´. Il-1β sense, 5´-CAA CCA ACA AGT GAT ATT CTC CAT G-3´; antisense, 5´-GAT CCA CAC TCT CCA GCT GCA-3´. Il-6 sense, 5´-TCC AAT GCT CTC CTA ACA GAT AAG-3´; antisense, 5´-CAA GAT GAA TTG GAT GGT CTT G -3´. C1qa sense, 5´-AAA GGC AAT CCA GGC AAT ATC A-3´; antisense, 5´-TGG TTC TGG TAT GGA CTC TCC-3´. C3 sense, 5´-CCA GCT CCC CAT TAG CTC TG-3´; antisense, 5´-GCA CTT GCC TCT TTA GGA AGT C-3´. Cd22 sense, 5′-CCA CTC CTC AGG CCA GAA ACT-3′; antisense, 5′-TGC CGA TGG TCT CTG GAC TG-3′.

Western blot analysis

Forebrains of one hemisphere from prion-infected mice were homogenized in sterile 0.32 M sucrose in PBS. Total protein concentration was determined using the bicinchoninic acid assay (Pierce). ~20 µg proteins were loaded and separated on a 12% Bis-Tris polyacrylamide gel (NuPAGE, Invitrogen) and then blotted onto a nitrocellulose membrane. Membranes were blocked with 5% Topblock (LuBioScience) (w/v) in PBS supplemented with 0.05% Tween 20 (v/v) and incubated with primary antibodies in 1% Topblock overnight at 4 °C. After washing, the membranes were then incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies for 2 h at room temperature. Blots were developed using Luminata Crescendo Western HRP substrate (Millipore) and visualized using the FUJIFILM LAS-3000 system. Primary antibodies used in this study included anti-PGRN antibody (1:400, R&D system, AF2557), POM1 (200 ng ml^- 1), anti-Iba1 (1:1,000, Wako laboratory chemicals, 019-19741), anti-GFAP (1:2,000, Cell signaling technology, 12389 S), anti-C1qa (1:1,000, Abcam, ab71089), anti-C3 (1:500, Abcam, ab11862), anti-LC3B (1:1,000, Cell Signaling Technology, 2775 S), anti-P62 (1:1,000, MBL international, PM045). To avoid variation in loading, the same blots were stripped and incubated with an anti-actin (1:10,000, Millipore, MAB1501R) or anti-GAPDH (1:5,000, Millipore, MAB374) antibody. The signals were normalized to actin or GAPDH as a loading control. Secondary antibodies were goat anti-rabbit IgG (1:10,000, Jackson ImmunoResearch, 111-035-045), goat anti–mouse IgG (1:10,000, Jackson ImmunoResearch, 115-035-003), goat anti-rat IgG (1:2,000, Invitrogen, 62-9520) and donkey anti-sheep IgG (1:5,000, Abcam, ab150178).

To detect PrP^Sc in prion-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) mouse brains, prion-infected forebrains were homogenized in sterile 0.32 M sucrose in PBS. Total protein concentration was determined using the bicinchoninic acid assay (Pierce). Samples were adjusted to 20 µg protein in 20 µl and digested with 25 µg ml^- 1 proteinase K in digestion buffer (PBS containing 0.5% wt/vol sodium deoxycholate and 0.5% vol/vol Nonidet P-40) for 30 min at 37 °C. PK digestion was stopped by adding loading buffer (Invitrogen) and boiling samples at 95 °C for 5 min. Proteins were then separated on a 12% Bis-Tris polyacrylamide gel (NuPAGE, Invitrogen) and blotted onto a nitrocellulose membrane. POM1 (200 ng ml^- 1) and horseradish peroxidase (HRP)-conjugated goat anti–mouse IgG (1:10,000, Jackson ImmunoResearch, 115-035-003) were used as primary and secondary antibodies, respectively. Blots were developed using Luminata Crescendo Western HRP substrate (Millipore) and visualized using the FUJIFILM LAS-3000 system.

Immunohistochemistry and immunofluorescence staining

For immunohistochemistry of prion-infected brains, formalin-fixed tissues were treated with concentrated formic acid for 60 min to inactivate prion infectivity and embedded in paraffin. Paraffin Sect. (2 μm) of brains were stained with hematoxylin/eosin (H&E). After deparaffinization through graded alcohols, Iba-1 antibody (1:1000; Wako Chemicals GmbH, Germany, 019-19741) was used for highlighting microglial cells, GFAP antibody (1:300; DAKO, Carpinteria, CA, z0334) was used for astrocytes. Stainings were visualized using an IVIEW DAB Detection Kit (Ventana), with a hematoxylin counterstain applied subsequently. For the histological detection of partially proteinase K-resistant prion protein deposition, deparaffinized sections were pretreated with formaldehyde for 30 min and with 98% formic acid for 6 min, and then washed in distilled water for 30 min. Sections were incubated in Ventana buffer and stains were performed on a NEXEX immunohistochemistry robot (Ventana instruments, Switzerland) using an IVIEW DAB Detection Kit (Ventana). After incubation with protease 1 (Ventana) for 16 min, sections were incubated with anti-PrP SAF84 (1:200, SPI bio, A03208) for 32 min. Sections were counterstained with hematoxylin. Sections were imaged using a Zeiss Axiophot light microscope. Quantification of vacuolation (H&E staining), SAF84, GFAP and Iba-1 staining and Sholl analysis were performed on acquired images using Image J software (National Institutes of Health). Briefly, for cell number counting, images were converted into 8-bit grayscale and the thresholds were adjusted to distinguish cells from the background. The particles were then analyzed, and the cell counts and area were measured. The cell counts were also validated by manual counting. For SAF84 signal calculation, images were color deconvoluted and H-DAB was chosen as the stain. The thresholds were adjusted and the mean gray values were measured and analyzed. For Sholl analysis of microglia, images of Iba-1 staining were first converted into 8-bit grayscale and the thresholds were adjusted. After defining a center point in soma, the parameters of Sholl analysis were adjusted. The number of dendritic intersections were counted with different distances from the soma. The operator was blind to the genotype and treatment of the analyzed tissues. 4 ~ 17 sections per group (depending on the samples collected) were used for quantification.

For immunofluorescent staining of PGRN on deparaffinized mouse brain sections, anti-progranulin antibody (10ug ml^- 1, R&D systems AF2557) and anti-Iba1 (1:1000, Wako laboratory chemicals, 019-19741) were used as primary antibodies and Alexa Fluor^® 488 goat anti-rabbit IgG (1:500, Invitrogen, A11008) and Alexa FluorⓇ 594 donkey anti-sheep IgG (1:500, Abcam, ab150180) were used as secondary antibodies. DAPI was used for fluorescent nuclear counterstaining. Images were captured using a confocal microscope (Leica, SP8).

ELISA

PGRN levels in tamoxifen-treated Grn^fl/fl and CX3CR1-CreERT2;Grn^fl/fl mouse brain samples were determined by an ELISA kit (Abcam, ab213473) according to the manufacturer’s protocol. Briefly, forebrain samples of one hemisphere were homogenized in 1X cell extraction buffer PTR to prepare 10% brain homogenates. Total protein concentration was determined using the bicinchoninic acid assay (Pierce). After adjustment of the protein concentration by 1:250 dilution, 50µL samples were loaded to the 96-well plate. The antibody cocktail (50µL) was then added and incubated for 1 h at room temperature. After washing, TMB development solution (100µL) was added to the wells. The reactions were then stopped by stop solution (100µL) and the plate was read at 450 nm. Grn^-/- mouse brains were used as negative control and background for normalization.

Magnetic activated cell sorting

Mice were anesthetized and perfused with ice-cold PBS. Half brain tissue were collected and immediately put into 10 mL cold D-PBS buffer for microglia isolation. Brains were chopped into small pieces by forceps and transferred to a 70 μm strainer. Tissue were dissociated with a pestle and washed with more D-PBS to collect all cells. Cells were collected by centrifugation at 300×g for 10 min at 4 °C. Pellets were resuspended with 3100 µL of D-PBS + 900 µL Debris Removal Solution, and then slowly and gently overlaid with 4 mL of D-PBS on the top, followed by centrifugation at 4 °C and 3000×g for 10 min with full acceleration and brake. Three phases were formed and the two top phases were aspirated completely to remove myelin and debris. The lower phase was filled with cold D-PBS to a final volume of 15 mL and centrifuged at 4 °C and 1000×g for 10 min with full acceleration and brake. Supernatant were removed completely. Cell pellets were resuspended in 90 µL of cold PB buffer per 10⁷ total cells. 10 µL of CD11b (Microglia) MicroBeads (Miltenyi Biotech) were added to the cell suspension and incubated for 15 min in the dark in the refrigerator (2–8 °C). Cells were washed by adding 1 mL of cold PB buffer per 10⁷ cells and centrifuged at 300×g for 5 min. Supernatants were removed and cells were resuspended up to 10⁷ cells in 500 µL of PB buffer. CD11b⁺ cells were separated in a magnetic field by MS columns according to the manufacturer’s instruction (Miltenyi Biotech). The purity of isolated microglia was analyzed by flow cytometry with CD11b staining. Isolated microglia were immediately lysed in RLT buffer and total RNAs were extracted using RNeasy kit (Qiagen) for further analysis.

RNA sequencing

The quality of RNA extracted from isolated microglia was analyzed by Bioanalyzer 2100 (Agilent Technologies), RNAs with RNA integrity number >8 were further processed. The TruSeq Stranded mRNA Sample Prep Kit (Illumina, Inc, California, USA) was used in the succeeding steps. Briefly, total RNA samples (500 ng) were poly-A selected and then reverse-transcribed into double-stranded cDNA with Actinomycin added during first-strand synthesis. The cDNA samples were fragmented, end-repaired and adenylated before ligation of TruSeq adapters. The adapters contain the index for multiplexing. Fragments containing TruSeq adapters on both ends were selectively enriched with PCR. The quality and quantity of the enriched libraries were validated using Qubit^® (1.0) Fluorometer (Life Technologies, California, USA) and the Bioanalyzer 2100 (Agilent Technologies). The product is a smear with an average fragment size of approximately 360 bp. The libraries were normalized to 10nM in Tris-Cl 10 mM, pH8.5 with 0.1% Tween 20. The TruSeq SR Cluster Kit v4-cBot-HS (Illumina, Inc, California, USA) was used for cluster generation using 8 pM of pooled normalized libraries on the cBOT. Sequencing was performed on the Illumina HiSeq 4000 single end 125 bp using the TruSeq SBS Kit v4-HS (Illumina, Inc, California, USA). Reads were quality-checked with FastQC. For RNAseq data analysis, we used RSEM to perform RNAseq read alignment and expression estimation [62]. As reference we used the GRCm38 genome assembly and the corresponding gene annotations provided by Ensembl. We computed differential expression with the Bioconductor package DESeq2 [63]. A gene was considered as differentially expressed by applying a threshold of < 0.05 for the p-value and >0 for the |log2 ratio| (fold change). Additionally, we filtered away genes that had very low counts. Specifically, we did not consider a gene as expressed if it did not exceed in at least one condition an average read count of 10 in the samples. All clusterings and visualizations were performed with R/Bioconductor (R version 3.2). Raw RNA-seq data were deposited in NCBI Sequence Read Archive (SRA) under BioProject accession number PRJNA1260716.

Statistical analysis

Results are presented as the mean of replicas ± SEM. Unpaired Student’s t-test was used for comparing two groups. Two-way ANOVA was used for the microglial sholl analysis. For survival curves of the in vivo experiments, all groups were compared by Log-rank (Mantel-Cox) test. p-values < 0.05 were considered statistically significant.

Results

Prion infection upregulates PGRN expression in mouse brain

To investigate whether the PGRN expression is altered by prion infection, we first performed quantitative reverse-transcription PCR (qRT-PCR) on forebrain mRNA isolated from terminally scrapie-sick wild-type (WT) C57BL/6J mice infected with the Rocky Mountain Laboratory strain of mouse-adapted scrapie prions (passage 6, therefore named RML6) (B6 + RML). Age-matched WT C57BL/6J mice inoculated with non-infectious brain homogenates (NBH) served as controls (B6 + NBH). We found that Grn mRNA was significantly increased in RML6 infected mouse forebrains (Fig. 1a). As an additional control, we included RML6-inoculated Prnp^-/- mice (Prnp^-/- +RML), which are resistant to prion infection and do not develop prion disease. These mice exhibited Grn mRNA levels similar to those in NBH-treated WT C57BL/6J mice (Fig. 1a), indicating that the upregulation of Grn mRNA is specifically associated with prion-induced pathology. Western blot analysis further confirmed that PGRN protein levels were significantly elevated in RML6-infected C57BL/6J mouse forebrains (Fig. 1b, full uncropped blot image in Suppl Fig. 1a, 1a’).

Fig. 1.

Upregulation of PGRN expression in prion-infected mouse brains. (a) qRT-PCR for Grn mRNA from terminally sick RML6-infected C57BL/6J mice (B6 + RML) and age-matched NBH-inoculated C57BL/6J mice (B6 + NBH) and RML6-inoculated Prnp^-/- mice (Prnp^-/-+RML) (n = 4). Relative expression was normalized to Gapdh expression and represented as percentage of average values in NBH-treated mice. (b) Left, Western blot for PGRN and Actin on brains collected from terminally sick RML6-infected C57BL/6J mice and age-matched NBH-inoculated C57BL/6J mice (n = 4 ~ 5). Right, densitometric quantification of the PGRN Western blot (n = 4 ~ 5). (c) RNA-Seq data of Grn expression in hippocampi at different time points after prion inoculation (n = 3). (d) Ribosomal profiling of Grn expression in CX3CR1-positive microglia at different time points after prion inoculation (n = 3, except n = 2 at terminal stage). (e) RNA-Seq data of Grn expression in sorted microglia (CD11b⁺) at different time points after prion inoculation (n = 5). (f) Immunofluorescence staining of PGRN on brain sections from RML6- and NBH-inoculated mice. Scale bar: 20 μm. ns: P > 0.05; *: P < 0.05; ***: P < 0.001; ****: P < 0.0001

Additionally, leveraging a time course RNA sequencing (RNA-Seq) study of hippocampal transcriptome in prion-infected mice [64], we compared the Grn expression in RML6 and NBH-inoculated C57BL/6J mouse hippocampi. We observed a significant increase in Grn mRNA levels in prion-infected mice from 18 weeks post-inoculation (wpi) until the terminal stage, compared to the NBH inoculated mice (Fig. 1c).

To identify the cell type responsible for the upregulation of Grn expression during prion infection, we analyzed data from cell-type-specific ribosomal profiling using CX3CR1-CreER, CamKIIa-Cre, and GFAP-Cre mice [65]. This analysis revealed that Grn expression was markedly upregulated in microglia at 24 wpi and the terminal stage (Fig. 1d), whereas no significant changes were observed in neurons or astrocytes (Suppl Fig. 2). To confirm that the increased Grn mRNA level reflected an absolute upregulation of Grn gene expression in microglia -rather than merely a consequence of prion-induced microglial proliferation [54]- we performed magnetic activated cell sorting (MACS) to isolate CD11b⁺ microglia from RML6-infected or NBH-treated C57BL/6J mouse brains. RNA-Seq analysis of these microglia confirmed that Grn mRNA levels were significantly elevated from 18 wpi until the terminal stage (Fig. 1e). Furthermore, immunostaining of brain sections from RML6 or NBH-inoculated mice demonstrated that the increased PGRN expression were predominantly localized to microglia (Fig. 1f).

Notably, a recent microarray analysis of sporadic Creutzfeldt-Jakob disease (sCJD) and non-neurological controls revealed that GRN transcription levels were significantly higher in sCJD brains [66]. Collectively, these findings indicate that Grn expression in the brain, particularly in microglia, is robustly upregulated by prion infection, suggesting that PGRN may play a role in prion pathogenesis and associated neuroinflammation.

Complete depletion, but not haploinsufficiency, of PGRN accelerates prion progression

To investigate whether PGRN deficiency influences prion pathogenesis, we intracerebrally inoculated 1.5-2-month-old Grn^-/-, Grn^+/- and Grn^+/+ (WT) littermates with RML6 prions. The incubation time was defined as the interval between prion inoculation and the onset of terminal scrapie symptoms, at which point mice were euthanized. We found that haploinsufficiency of PGRN (Grn^+/-) did not affect prion progression, whereas complete depletion of PGRN (Grn^-/-) significantly accelerated the disease. Grn^-/- mice exhibited significantly earlier onset of clinical symptoms and shorter incubation times compared to isogenic WT littermates (median survival: 166.5 days post inoculation (dpi) for Grn^-/-, n = 18, vs. 177 dpi for WT littermates, n = 15, ***p = 0.0002) (Fig. 2a). Grn^+/- mice showed incubation times similar to WT littermates (median survival: 174 dpi, n = 25). Interestingly, there was no difference in the terminal clinical signs or intensity between different groups. These results indicate that complete PGRN deficiency, but not haploinsufficiency, accelerates prion progression, underscoring the neuroprotective role of PGRN in prion pathogenesis.

Fig. 2.

PGRN deficiency accelerated prion progression. (a) Survival curves of RML6 infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) mice. The median survival of Grn^-/- mice was 166.5 dpi (n = 18) whereas that of Grn^+/+ (WT) females was 177 dpi (n = 15). The median survival of Grn^+/- mice was 174 dpi (n = 25). (b-d) Representative histology of terminally sick mouse brains from Grn^-/-, Grn^+/- and Grn^+/+ (WT) littermates stained for Hematoxylin/Eosin (H&E) (b), SAF84 (c) and GFAP (d). Scale bar: 100 μm in (b) and (c), 25 μm in (d). (e): Quantitation of the vacuoles (upper panel), SAF84 signal (middle panel) and GFAP⁺ cells (lower panel). n = 5 ~ 6 for each group and each brain region. ns: P > 0.05; *: P < 0.05; ***: P < 0.001

To evaluate the impact of PGRN deficiency on prion-induced neuropathology and astrogliosis, brains were collected from prion-infected Grn^-/-, Grn^+/- and Grn^+/+ mice at the terminal stage (when mice were ~ 7–8 months old). Histological analysis revealed similar lesion patterns, visualized as vacuolation, across all groups, with slightly more vacuoles in the thalamus of Grn^-/- mice (n = 5 ~ 6, Fig. 2b, e). PrP^Sc deposition, assessed by SAF84 staining, was also comparable in different brain regions at the terminal stage (n = 5 ~ 6, Fig. 2c, e). GFAP (glial fibrillar acidic protein) staining showed no significant differences in reactive astrocytes among the three groups (n = 5 ~ 6, 401 ~ 472 cells/mm², Fig. 2d, e), suggesting that astrogliosis at the terminal stage is not influenced by PGRN deficiency. H&E and SAF84 staining on NBH-inoculated Grn^-/- and Grn^+/+ mice at the same age (7 ~ 8 months old) demonstrated no vacuolation or PrP^Sc deposition in the brains (Suppl Fig. 3a, b), GFAP staining revealed similar astrocyte numbers in NBH-inoculated Grn^-/- and Grn^+/+ mouse brains (Suppl Fig. 3c).

It has been previously reported that Grn^-/- mice developed TDP-43 proteinopathy by 24 months of age [42]. However, we did not detect obvious TDP-43 pathology in the brains of prion-infected Grn^+/- or Grn^-/- mice at the terminal stage (data not shown), indicating that prion infection does not accelerate TDP-43 proteinopathy induced by PGRN insufficiency.

Augmented microglial activation in prion-infected Grn^-/- mice

Given that PGRN has been shown to regulate microglial activation in various pathological contexts [34, 37] and during aging [40, 46], we investigated the effect of PGRN deficiency on prion-induced microglial activation. Iba1 (ionized calcium-binding adaptor molecule 1) staining on brain sections from terminally sick Grn^-/-, Grn^+/- and Grn^+/+ mice revealed significantly augmented microglial activation in the cortex, hippocampus and thalamus of Grn^-/- mice compared to Grn^+/- and Grn^+/+ littermates (n = 5 ~ 6, Fig. 3a). Consistent with previous observations in aging brains, the augmentation was most pronounced in thalamus. Interestingly, microglial activation in the cerebellum was similar across all genotypes, suggesting that the enhanced microglial activation is brain region-specific (Fig. 3a). Iba-1 staining on NBH-inoculated Grn^-/- and Grn^+/+ mouse brains revealed similar microglia numbers (Suppl Fig. 3d). Although 7-month-old Grn^-/- mice have been reported to exhibit slightly increased microglial activation compared to age-matched Grn^+/- and Grn^+/+ mice [38], prion infection drastically exacerbated this activation. Morphologically, Grn^-/- microglia in thalamus displayed an ameboid shape with enlarged cell bodies and shortened processes. Sholl analysis of Iba1⁺ microglial morphology confirmed that thalamic Grn^-/- microglia have significant fewer processes (Fig. 3b). In contrast, while microglial proliferation was enhanced in hippocampus and cortex of Grn^-/- mice, their morphology remained similar to that of Grn^+/+ and Grn^+/- microglia (Fig. 3c-d).

Fig. 3.

Augmented microglial activation and altered cytokine profiles in prion-infected Grn^-/- mice. (a) Left, representative histology of Iba1 immunohistochemistry of various brain regions from RML6-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) littermates at terminal stage. Scale bar: 25 μm. Right, quantification of Iba1 staining. n = 5 ~ 6 for each group and each brain region. (b-d) Sholl analysis of Iba⁺ microglia at cortex (b), hippocampus (c) and thalamus (d) of Grn^-/-, Grn^+/- and Grn^+/+ (WT) mice at terminal stage. n = 5 ~ 6 mice per group, 5–6 microglia/region/mouse were analyzed. (e) qRT-PCR of Tnfα, Il-1β and Il-6 mRNA in RML6-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) brains at terminal stage. n = 5. ns: P > 0.05; *: P < 0.05; **: P < 0.01; ***: P < 0.001; ****: P < 0.0001

Cytokine analysis revealed that TNFα were significantly elevated in Grn^-/- mouse brains compared to Grn^+/- and Grn^+/+littermates, whereas IL-1β and IL-6 was downregulated or showed a decreasing trend (Fig. 3e). These findings suggest that complete PGRN deficiency enhances prion-induced microglial activation and alters cytokine profiles, indicating that PGRN acts as a brake on prion-induced microglial activation.

Enhanced prion clearance in Grn^-/- mice at 120 Dpi but not at 150 Dpi

To further explore the impact of PGRN deficiency on prion-induced microglial activation, we collected brains from prion-infected Grn^-/-, Grn^+/- and Grn^+/+ mice at 120 dpi (when mice were ~ 6 months old), when mice began developing scrapie symptoms. Iba1 staining revealed significantly enhanced microglial activation in Grn^-/- mouse brains (primarily thalami) compared to Grn^+/- and Grn^+/+ littermates (Fig. 4a, upper panel; Suppl Fig. 4a). Interestingly, histological and Western blot analysis of forebrains showed that PrP^Sc deposition at this stage was significantly reduced in Grn^-/- mice (Fig. 4b upper panel, 4c; Suppl Fig. 4b; full uncropped blot image in Suppl Fig. 1b), although the total PrP are similar between different groups (Suppl Fig. 4c). These results suggest that microglia at this stage exhibit a phagocytic phenotype, and the enhanced microglial activation enables more efficient prion clearance. Since vacuolation induced by prion infection was minimal at this stage, no differences were observed between genotypes (Suppl Fig. 4d). Similarly, GFAP staining revealed no significant differences in astrogliosis among different genotypes at this stage (Suppl Fig. 4e).

Fig. 4.

Enhanced prion clearance in Grn^-/- mice at 120dpi, but not at 150dpi. (a) Upper panel: left, representative histology of Iba1 immunohistochemistry of thalamus from RML6-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) littermates at 120 dpi. Scale bar: 25 μm. Right, quantification of Iba1⁺ cells at thalamus. n = 6 ~ 17. Lower panel: left, representative histology of Iba1 immunohistochemistry of thalamus from RML6-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) littermates at 150 dpi. Scale bar: 25 μm. Right, quantification of Iba1⁺ cells at thalamus. n = 4 ~ 6. (b) Upper panel: left, representative histology of SAF48 immunohistochemistry of thalamus from RML6-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) littermates at 120 dpi. Scale bars: 100 μm. Right, quantification of SAF84 staining. n = 5 ~ 6. Lower panel: left, representative histology of SAF48 immunohistochemistry of thalamus from RML6-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) littermates at 150 dpi. Scale bars: 100 μm. Right, quantification of SAF84 staining. n = 4 ~ 6. (c) Left, PrP^Sc Western blot of homogenates prepared from RML6-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) mouse brains at 120 dpi. NBH control was WT brain homogenates inoculated with NBH at the same age. Samples were digested with PK as indicated and detected with POM1. Right, densitometric quantification of the PrP^Sc Western blot. n = 3. (d) Left, PrP^Sc Western blot of homogenates prepared from RML6-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) mouse brains at 150 dpi. NBH control was WT brain homogenates inoculated with NBH at the same age. Samples were digested with PK as indicated and detected with POM1. Right, densitometric quantification of the PrP^Sc Western blot. n = 3. (e) Left, representative histology of GFAP immunohistochemistry of thalamus from RML6-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) littermates at 150 dpi. Scale bar:25 μm. Right: quantification of GFAP⁺ cells at thalamus. n = 4 ~ 6. (f) Left, representative histology of H&E staining of thalamus from RML6-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) littermates at 150 dpi. Scale bars: 100 μm. Right: quantification of vacuoles at thalamus. n = 4 ~ 6. ns: P > 0.05; **: P < 0.01; ***: P < 0.001; ****: P < 0.0001

At 150 dpi (when mice were ~ 7 months old), when mice were severely sick and nearing the terminal stage, microglial activation remained significantly enhanced in Grn^-/- mouse thalami compared to Grn^+/- and Grn^+/+ littermates (Fig. 4a, lower panel; Suppl Fig. 4a). However, forebrain PrP^Sc deposition in Grn^-/- mice at this stage was similar to that of Grn^+/- and Grn^+/+ mice (Fig. 4b lower panel, 4d; Suppl Fig. 4b; full uncropped blot image in Suppl Fig. 1c). This suggests that excessive microglial activation in Grn^-/- mice at 150 dpi was less effective in clearing prions compared to the 120 dpi stage. These findings imply that prion-induced microglial activation occurs in a stepwise manner: at 120 dpi, the activated microglia exhibit a phagocytic phenotype, while at 150 dpi, they transition to a less efficient state with reduced prion clearance capacity.

Interestingly, GFAP staining revealed enhanced astrogliosis in Grn^-/- thalami at 150 dpi (Fig. 4e), consistent with recent reports that PGRN deficiency leads to aberrant astrocytic phenotype promoting synaptic degeneration [67]. Importantly, Grn^-/- thalami exhibited more severe vacuolation at this stage compared to Grn^+/- and Grn^+/+ littermates (Fig. 4f). These results indicate that complete deficiency of PGRN exacerbates prion pathology as the disease progresses, correlating with the shortened incubation time observed in Grn^-/- mice.

Lack of progranulin in prion-infected mice results in abnormal immune response and altered microglial States

To investigate how complete PGRN deficiency augments microglial activation and accelerates prion progression, we isolated microglia (CD11b⁺) from prion-infected Grn^-/-, Grn^+/-and Grn^+/+ (WT) mouse forebrains at 150 dpi (~ 7 months old) using MACS and performed whole-transcriptome analyses by RNA-seq. The purities of sorted microglia were analyzed by flow cytometry and >90% cells were positive for CD11b (data not shown), consistent with previous reports [68]. The multidimensional scaling (MDS) plot showed that Grn^+/- (Het) and Grn^+/+ (WT) microglial replicates clustered closely, whereas Grn^-/- (KO) microglia exhibited a distinct transcriptomic profile (Suppl Fig. 5a). These results suggest that haploinsufficiency of PGRN has a modest effect on microglial transcriptomes, while complete deficiency of PGRN leads to profound alterations in microglial gene expression.

We compared the gene expression changes between Grn^-/-, Grn^+/- and Grn^+/+ microglia (Suppl Table 1). Grn^-/- microglia showed 2,300 differentially expressed genes (DEGs) (1,129 upregulated, 1,171 downregulated) compared to Grn^+/+ microglia (p < 0.05), whereas Grn^+/- microglia exhibited only 152 DEGs (30 upregulated, 122 downregulated) relative to Grn^+/+ microglia (Fig. 5a). Among these, 10 upregulated and 40 downregulated (including Grn itself) DEGs overlapped between Grn^-/- v.s Grn^+/+ and Grn^+/- v.s Grn^+/+ microglia (Fig. 5b). Gene enrichment analysis of the 2,300 DEGs in Grn^-/- microglia by Metascape online tool (http://metascape.org/) revealed significant involvement in pathways related to positive and negative regulation of response to external stimulus, positive regulation of cell migration, regulation of cell activation and innate immune response (Fig. 5c). These findings align with previous studies demonstrating abnormal immune responses caused by PGRN deficiency.

Fig. 5.

PGRN deficiency shifted microglia from homeostatic to disease-associated states in prion-infected mouse brains. (a) Volcano plots of RNAseq data for microglia (CD11b⁺) from prion-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) mouse brains at 150 dpi. Left, Grn^-/- v.s Grn^+/+. Right, Grn^+/- v.s Grn^+/+. Upregulated (red dots) and downregulated (blue dots) are genes that show fold change (fc) > 1 or <-1 with P < 0.05. P > 0.05: nonsignificant. Genes that are not detected: absent. n = 3 for Grn^-/- and Grn^+/+, n = 4 for Grn^+/-. (b) Venn diagram of overlapping differentially expressed genes (DEGs) between Grn^-/- v.s Grn^+/+ and Grn^+/- vs. Grn^+/+. Left: significantly upregulated genes; Right: significantly downregulated genes. (c) GO analysis of DEGs between Grn^-/- and Grn^+/+. (d) Heat map of disease-associated markers and homeostatic markers of RNAseq data for microglia (CD11b⁺) from prion-infected Grn^-/- mouse brains at 150 dpi. (e, f) Read counts and qRT-PCR of phagocytosis negative regulator Cd22 in CD11b⁺ microglia from prion-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) mouse brains at 150 dpi. n = 3 for Grn^-/- and Grn^+/+, n = 4 for Grn^+/-. ns: P > 0.05; *: P < 0.05

Interestingly, while PGRN deficiency alone showed moderate changes (45 DEGs) in microglial gene expression at 7 months of age in untreated mice revealed by single-nucleus RNA sequencing (snRNA-seq) [42], prion-infected Grn^-/- mice at the same age exhibited more profound transcriptional alterations in microglia. There was an overlap of 28 DEGs between Grn^-/- microglia isolated from prion-infected mice or untreated mice at 7 months old (Suppl Fig. 5b, overlapping genes are highlighted in yellow and orange in suppl Table 1). Notably, 19 of the 21 consistently dysregulated genes in Grn^-/- microglia across different ages (7 months, 12 months and 19 months of age) were also differentially expressed in Grn^-/- microglia from prion-infected mice (Suppl Fig. 5c, overlapping genes are highlighted in orange in suppl Table 1). These results suggest synergistic effects between prion infection and PGRN deficiency.

Although the transcriptional profiles of aged Grn^-/- microglia diverge significantly from those of disease-associated microglia (DAM) [42, 69], microglia from prion-infected Grn^-/- mice showed increased expression of most DAM markers (e.g. Apoe, Ctsb, Ctsd, Itgax, Cd63 and Cd68), and decreased expression of homeostatic markers (e.g. Cx3cr1, P2ry12, P2ry13, Tmem119) (Fig. 5d). These results suggest that complete PGRN deficiency accelerates the transition of microglia from a homeostatic state to a disease-associated phenotype in prion-infected mice. Importantly, Grn^-/- microglia at 150 dpi exhibited significantly upregulated expression of CD22 (Fig. 5e, f), a recently identified negative regulator of microglia phagocytosis [70, 71]. This may explain the impaired prion clearance observed at this stage. Notably, we observed that prion infection alone significantly upregulated CD22 expression in both bulk brain tissue and isolated microglial after 18 wpi (Suppl Fig. 5d, e). Moreover, CD22 levels were already elevated in noninfected Grn^-/- brains (7 months old) and Grn^-/- microglia compared to wild-type controls (Suppl Fig. 5f, g). Together, these findings suggest a synergistic effect of prion infection and PGRN deficiency on CD22 upregulation at 150 dpi.

PGRN deficiency in prion-infected mice results in aggravated complement activation

Innate immune response emerged as one of the top dysregulated pathways in prion-infected Grn^-/- mice. Previous studies have reported aberrant activation of complement cascade in Grn^-/- mice, leading to excessive synaptic pruning, TDP-43 neuropathology, and neurotoxicity [40, 42]. To determine whether complement activation contributes to the accelerated prion progression in Grn^-/- mice, we first assessed complement cascade activation in prion-infected mouse brains. qRT-PCR and WB analysis revealed that C1qa and C3 mRNA levels were upregulated in RML6 infected mice (Fig. 6a). Similarly, the protein levels of C1qa and C3 cleavage products iC3b were increased in prion-infected brains (Fig. 6b, full uncropped blots images in Suppl Fig. 1d-e). Temporal RNA-seq analysis confirmed upregulation of C1qa and C3 in both bulk tissue and microglia after 18 wpi (Suppl Fig. 6a-d).

Fig. 6.

Aggravated complement activation in prion-infected Grn^-/- mouse brains. (a) qRT-PCR of C1qa (left) and C3 (right) mRNA in prion-infected C57BL/6J mouse brains (B6 + RML). n = 4 (b) Left, Western blot of C1qa and C3 in prion-infected C57BL/6J mouse brains. Right, densitometric quantification of the C1qa and C3 Western blot. n = 5. (c) qRT-PCR of C1qa and C3 in prion-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) mouse brains at 120 dpi. n = 3 ~ 5. (d) Left, Western blot of C1qa and C3 in prion-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) mouse brains at 120 dpi. NBH control was WT brain homogenates inoculated with NBH at the same age (B6 + NBH). Right, densitometric quantification of the C1qa and C3 Western blot. n = 3. (e) qRT-PCR of C1qa and C3 in prion-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) mouse brains at 150 dpi. n = 4 ~ 5. (f) Left, Western blot of C1qa and C3 in prion-infected Grn^-/-, Grn^+/- and Grn^+/+ (WT) mouse brains at 150 dpi. NBH control was WT brain homogenates inoculated with NBH at the same age. Right, densitometric quantification of the C1qa and C3 Western blot. n = 3. ns: P > 0.05; *: P < 0.05; **: P < 0.01; ***: P < 0.001; ****: P < 0.0001

We then evaluated complement components in Grn^-/-, Grn^+/- and Grn^+/+ mouse forebrains infected with RML6 at 120dpi and 150dpi. While C1qa and C3 levels were unchanged in Grn^-/- mice at 120dpi, both were significantly upregulated in Grn^-/- mice at 150dpi (Fig. 6c-f, full uncropped blots images in Suppl Fig. 1f-i). This aggravated complement activation in Grn^-/- mice at 150dpi may partially explain the exacerbated neuronal lesion and accelerated prion progression observed in these mice. Interestingly, RNA-seq did not find significant increase of C1qa transcripts in Grn^-/- microglia (Suppl Table 1), suggesting that greater microglial numbers and/or other cell types may contribute to the increased complement activation. Although depletion of C1qa or C3 alone does not alter prion progression following intracerebral infection [72], further studies are needed to determine whether ablation or antagonism of these complement components could mitigate prion acceleration induced by PGRN-deficiency.

Impaired autophagosome functions have been reported in aged Grn^-/- mice. To investigate whether Grn deficiency exacerbates autophagosome dysfunction after prion infection, we performed Western blots analysis of LC3B and P62. However, no significant changes in these markers were observed in prion-infected Grn^-/-, Grn^+/- and Grn^+/+ mice (Suppl Fig. 6e, f), suggesting that impaired autophagosome function is not responsible for the accelerated prion progression in Grn^-/- mice.

Microglia-specific depletion of PGRN does not alter prion pathogenesis

Given that PGRN is primarily upregulated in microglia upon prion infection and PGRN deficiency leads to augmented microglial activation, we investigated whether microglia-specific deletion of Grn affects prion pathogenesis. We generated CX3CR1-CreERT;Grn^fl/fl mice and treated them with Tamoxifen (Tx) for 4 weeks. RT-qPCR and ELISA confirmed significant reduction of Grn mRNA and PGRN protein levels in CX3CR1-CreERT;Grn^fl/fl mice compared to Tx-treated Grn^fl/fl control mice (Fig. 7a, b).

Fig. 7.

Microglia specific Grn depletion did not alter prion pathogenesis. (a) qRT-PCR for Grn mRNA and (b) ELISA for PGRN levels in tamoxifen-treated CX3CR1-CreERT;Grn^fl/fl and Grn^fl/fl mouse brains. n = 5 ~ 6. (c) Survival curves of RML6-infected CX3CR1-CreERT;Grn^fl/fl and Grn^fl/fl mice untreated or treated with Tamoxifen. The median survival of Grn^fl/fl+Tx mice was 172 dpi (n = 25), CX3CR1-CreERT;Grn^fl/fl+Tx mice was 170.5 dpi (n = 25), Grn^fl/fl-Tx mice was 166 dpi (n = 24), CX3CR1-CreERT;Grn^fl/fl-Tx mice was 174dpi (n = 24). There was no significant difference between the groups. (d-g) Left, representative histology of cortex (d), hippocampus (e), thalamus (f) and cerebellum (g) of RML6-infected CX3CR1-CreERT;Grn^fl/fl and Grn^fl/fl mice treated with Tamoxifen. Scale bar: 100 μm in (d) and (e), 25 μm in (f) and (g). Right, quantification of vacuoles, SAF84 signals, GFAP⁺ cells and Iba⁺ cells at various brain regions. n = 6. ns: P > 0.05; *: P < 0.05

To assess the impact of microglia-specific PGRN depletion on prion pathogenesis, CX3CR1-CreERTxGrn^fl/fl and Grn^fl/fl mice were fed with tamoxifen diet (diet containing tamoxifen citrate 400 ppm, +Tx) for 4 weeks, followed by a normal diet for 2 weeks, before intracerebral inoculation with RML6. Control groups without tamoxifen diet feeding (-Tx) were also inoculated. Unexpectedly, prion progression was similar between CX3CR1-CreERT;Grn^fl/fl and Grn^fl/fl mice, regardless of tamoxifen treatment (median survival: 170 dpi for Grn^fl/fl mice + Tx (n = 25), 170 dpi for CX3CR1-CreERT;Grn^fl/fl +Tx (n = 25), 168 dpi for Grn^fl/fl mice -Tx (n = 24), 173.5 dpi for CX3CR1-CreERT;Grn^fl/fl -Tx (n = 24), p = 0.62) (Fig. 7c). Histological analysis also revealed no differences in lesion pattern, PrP^Sc deposition, astrogliosis, or microglial activation between CX3CR1-CreERT;Grn^fl/fl and Grn^fl/fl mice (Fig. 7d-g). Complement activation and CD22 expression were comparable between CX3CR1-CreERT;Grn^fl/fl and Grn^fl/fl mice (Suppl Fig. 7a, b, c).

These observations suggest that, unlike complete PGRN ablation, microglia-specific depletion of PGRN does not alter prion progression. This indicates that the augmented microglial activation observed in prion-infected Grn^-/- mice is not cell-autonomous. Instead, PGRN expressed and released by other cell types, such as neurons, may regulate prion-induced microglial activation in a non-cell-autonomous manner. The crosstalk between microglia and other neural cells in prion pathogenesis warrants further investigation. Additionally, residual PGRN taken up by microglia from other cells or incomplete microglial depletion of Grn by Cx3cr1-CreERT-mediated recombination indicated by PGRN staining may mitigate the effects of microglia-specific Grn deficiency (Suppl Fig. 7d).

Discussion

The mechanisms by which PGRN deficiency leads to neurodegeneration are a subject of intense investigation. Our study demonstrates that PGRN expression in microglia is upregulated in response to prion infection. Complete depletion of Grn, but not haploinsufficiency, accelerates prion progression. PGRN ablation enhances prion-induced microglial activation, with microglia exhibiting increased phagocytic capacity at 120 dpi but losing this capability by 150 dpi, suggesting a phenotypic transition during this period. Both prion infection and PGRN-deficiency independently promote microglial activation, the augmented microglial response observed in prion-infected Grn^-/- mice likely reflects a synergistic interaction between these two factors. Importantly, PGRN deficiency results in augmented complement cascade activation at 150 dpi, which may contribute to the accelerated prion disease progression. These findings highlight a neuroprotective role for PGRN in prion pathogenesis and expand our understanding of PGRN’s functions in neurodegenerative diseases.

Most, if not all, pathological GRN mutations or polymorphisms result in reduced PGRN levels due to premature stop codons or mRNA decay. Consequently, plasma and/or cerebrospinal fluid (CSF) PGRN levels serve as important diagnostic biomarkers for GRN-related FTLD. Indeed, low PGRN levels reliably predict GRN mutations in FTLD patients [73–78]. Notably, reduced CSF PGRN levels have also been observed in FTLD patients without GRN mutations [79]. Interestingly, while AD patients carrying GRN mutations exhibit lower PGRN levels [17, 20, 23, 80], PGRN levels are elevated in other AD patients [81, 82]. In mouse models of AD, PGRN levels increase in microglia clustering around amyloid plaques [83]. However, in other AD models, PGRN levels decrease in the early disease stages but increase at later stages [50]. These findings suggest that PGRN expression varies dynamically across different neurodegenerative diseases, depending on the context and disease stage. Interestingly, our study observed an upregulation of microglial PGRN at the late stage of prion infection, suggesting a potential role of PGRN in prion-induced neuroinflammation and neurodegeneration.

The effect of PGRN deficiency on prion disease was observed only in Grn^-/- but not in Grn^+/- mice, consistent with previous reports that lipofuscinosis and transcriptional changes are detected only in the complete absence of PGRN [38, 40, 44]. Interestingly, in contrast to prion-infected mice, PGRN haploinsufficiency has been shown to influence Aβ load and deposition in AD mouse models [52]. The time-dependent effects of PGRN deficiency on microglial function may explain conflicting observations across AD models. In some studies, Grn^-/- mice exhibit increased amyloid deposits due to overactivated microglia with defective phagocytic capacity [50, 51], while others report reduced plaque burden due to more efficient phagocytosis and clearance by microglia at early stages [49, 52, 84].

While our data demonstrate enhanced microglial PrP^Sc clearance in Grn^-/- mice at 120 dpi, we considered alternative explanations for reduced deposition, including impaired PrP^Sc propagation due to altered neuronal membrane dynamics or autophagic-lysosomal dysfunction. The temporal glial response progresses through distinct phases: early microglial activation and hyperphagocytosis coincides with moderate astrocyte activation, synergistically enhancing prion clearance. This protective phase was followed by late microglial overactivation and impaired prion clearance that drives increased astrogliosis. The eventual convergence of astrocyte marker and PrP^Sc deposition at the terminal stage likely reflects the different incubation periods of Grn^-/- and Grn^+/+ mice. Collectively, these temporal shifts suggest that PGRN ablation initially boosts clearance capacity but ultimately drives maladaptive gliosis, consistent with our RNA-seq showing microglial polarization toward inflammatory states at 150 dpi.

To investigate how PGRN depletion enhances microglial activation during prion infection, we generated microglia-specific Grn knockout mice by crossing CX3CR1-CreERT2 with Grn^fl/fl mice. Intriguingly, in contrast to systemic PGRN deficiency, microglia-restricted depletion of PGRN did not alter prion pathogenesis. Thus, the heightened microglial activation seen in prion-infected Grn^-/- mice likely involves non-cell-autonomous mechanisms and the complete PGRN ablation may be necessary to impact prion pathogenesis. This aligns with previous findings that neuronal-specific Grn knockout mice do not develop any neuropathological phenotypes observed in complete Grn knockout mice, and microglia-specific PGRN depletion does not exacerbate behavioral or pathological phenotypes in neuronal PGRN-insufficient mice [85, 86]. Further studies are needed to explore how different PGRN-expressing cell types interact in the brain and whether PGRN released from these cells influence prion pathogenesis.

While impaired autophagy and TDP-43 accumulation have been reported in aged Grn^-/- mice [39], we observed no significant differences in autophagy-related proteins between prion-infected Grn^-/- and control mice. This suggests that prion and TDP-43 employ distinct mechanisms for intracellular clearance and that accelerated prion progression in Grn^-/- mice is not due to exacerbated autophagic impairment.

Dysfunctional lysosome, increased complement production, and enhanced synaptic pruning have been observed in aged Grn^-/- mice [40]. Additionally, Grn^-/- microglia promote TDP-43 granule formation, nuclear pore defects, and excitatory neuronal death via complement activation [42]. These phenotypes can be mitigated by depleting complement genes or blocking the complement cascade, indicating that PGRN suppresses excessive complement activation. In prion-infected Grn^-/- mice, we observed aggravated complement activation at 150dpi, which may contribute to accelerated disease progression. Further studies are warranted to determine whether blocking of complement cascade (such as intercrossing C1qa^-/- or C3^-/- mice to Grn^-/- mice) could mitigate the effects of PGRN depletion on prion disease.

We previously reported that TREM2 deficiency attenuates microglial activation in prion infected mice [56]. Together with the present findings, these results suggest that TREM2 and PGRN have opposing effects on prion-induced microglial activation. Indeed, loss of TREM2 and PGRN result in opposite microglial activation states [87]. While TREM2 ablation or antagonism can rescue the microglial hyperactivation in Grn^-/- mice, it does not alleviate lysosomal dysfunction or neurotoxicity [88]. Furthermore, the transition of microglia from a homeostatic state to a disease-associated phenotype stage 2 depends on TREM2 signaling [69]. Thus, it would be interesting to investigate whether TREM2 depletion could dampen microglial hyperactivation and rescue the accelerated prion progression in Grn^-/- mice.

Several strategies to increase PGRN levels have shown promise in treating FTD-GRN patients [1]. These include blocking PGRN degradation using anti-sortilin antibodies, supplementing recombinant PGRN via protein transport vehicle (PTV)-PGRN biologics, gene therapy employing adeno-associated virus (AAV)-mediated PGRN expression and small molecules to enhance expression of PGRN. However, further research is needed to determine whether these approaches could also benefit prion diseases.

Limitations of the study

While our study demonstrates a neuroprotective role for PGRN in a mouse model of prion infection, further research is needed to elucidate the role of PGRN in prion-induced neurodegeneration. Prion diseases are highly heterogenous, with different strains exhibiting unique characteristics, including variations in clinical symptoms, incubation periods, disease progression rates and brain regional lesion profiles [89]. In mouse models of scrapie prions, the widely studied RML, ME7 and 22 L strains display differences in cellular tropism of PrP^Sc accumulation and neuropathological distribution [90]. RML strain was selected for this study due to its thalamic preference, a brain region particularly vulnerable to PGRN deficiency [40]. While our findings demonstrate that ablation of PGRN alters prion pathology in RML-infected mice, the extent to which PGRN deficiency influences pathogenesis induced by other prion strains warrants further investigation. Notably, it remains unclear whether loss-of-function GRN variants are associated with an increased risk of prion diseases in humans, partly due to the rarity of conditions like Creutzfeldt-Jakob disease (CJD), Gerstmann-Sträussler-Scheinker syndrome (GSS), and fatal familial insomnia (FFI). Therefore, the relevance of GRN variants and PGRN dysfunction in human prion diseases warrants further investigation.

Conclusions

In summary, our study reveals a neuroprotective role for PGRN in prion disease. Microglia respond to prion infection in a stepwise manner, and PGRN restricts prion-induced microglial activation and decelerates prion pathology partly by suppressing excessive complement cascade activation. Given that PGRN overexpression or exogenous delivery has shown therapeutic benefits in other neurogenerative diseases [45, 50, 91–93], we propose that elevating PGRN levels - whether through small molecules, anti-sortilin antibodies, supplementation (e.g., PTV: PGRN biologics), or boosting expression (e.g., AAV-mediated gene therapy) - could represent a potential therapeutic strategy for prion disease.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AAV

Adeno-associated virus

AD

Alzheimer’s disease

ALS

Amyotrophic lateral sclerosis

CBS

Corticobasal syndrome

CJD

Creutzfeldt-Jacob disease

CNS

Central nervous system

dpi

Days post-inoculation

FFI

Fatal familial insomnia

FTLD

Frontotemporal lobar degeneration

GRN/Grn

Gene encoding progranulin in human or mouse

GSS

Gerstmann-Straussler-Scheinker syndrome

GWAS

Genome-wide association studies

LATE

Limbic-predominant age-related TDP-43 encephalopathy

NCL

Neuronal ceroid lipofuscinosis

PD

Parkinson’s disease

PGRN

Progranulin

PrP

Prion protein

PTV

Protein transport vehicle

qRT-PCR

Quantitative reverse transcription polymerase chain reaction

RML6

Rocky Mountain Laboratory strain of mouse-adapted scrapie prions, passage 6

TDP-43

Transactivation response DNA binding protein-43

TREM2

Triggering receptor expressed on myeloid cells 2

Author contributions

C.Z. and A.A. planned and conceived study. B.L., Y.S., W.H., H.G., J.L., T.Y., W.L., D.C. and C.Z. performed and analyzed the biochemical and histological experiments. P.S. performed prion inoculation experiments. RNA-seq was performed in Functional Genomic Center Zurich (FGCZ). C.Z. and A.A. supervised the execution of the project. B.L., C.Z. and A.A. wrote the manuscript with input and revisions from all authors.

Funding

C. Zhu is sponsored by Research Startup Funds of Fudan University, National Natural Science Foundation of China (No. 82271476 and No. 82071436), Shanghai Pujiang Program (No. 20PJ1401100), and The Program for Oriental Scholars of Shanghai Universities (Distinguished Professor) (No. TP2022050). B. Li is supported by National Natural Science Foundation of China (No. 82101502) and China Postdoctoral Science Foundation (No. 2021M690036). A. Aguzzi is the recipient of an Advanced Grant of the European Research Council (ERC, No. 670958) and is supported by grants from the European Union (PRIORITY, NEURINOX), the Swiss National Foundation (SNF, including a Sinergia grant), the Swiss Initiative in Systems Biology, SystemsX.ch (PrionX, SynucleiX), and a Distinguished Scientist Award of the Nomis Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data availability

Additional information is available upon reasonable request.

Declarations

Ethics approval and consent to participate

All animal experiments were carried out in strict accordance with the Rules and Regulations for the Protection of Animal Rights (Tierschutzgesetz and Tierschutzverordnung) of the Swiss Bundesamt für Lebensmittelsicherheit und Veterinärwesen and were preemptively approved by the Animal Welfare Committee of the Canton of Zürich (permit ZH040/2015).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.