Lymphocyte-Activation Gene 3 Facilitates Pathological Tau Neuron-to-Neuron Transmission

Abstract

The spread of prion-like protein aggregates is a common driver of pathogenesis in various neurodegenerative diseases, including Alzheimer's disease (AD) and related Tauopathies. Tau pathologies exhibit a clear progressive spreading pattern that correlates with disease severity. Clinical observation combined with complementary experimental studies has shown that Tau preformed fibrils (PFF) are prion-like seeds that propagate pathology by entering cells and templating misfolding and aggregation of endogenous Tau. While several cell surface receptors of Tau are known, they are not specific to the fibrillar form of Tau. Moreover, the underlying cellular mechanisms of Tau PFF spreading remain poorly understood. Here, it is shown that the lymphocyte-activation gene 3 (Lag3) is a cell surface receptor that binds to PFF but not the monomer of Tau. Deletion of Lag3 or inhibition of Lag3 in primary cortical neurons significantly reduces the internalization of Tau PFF and subsequent Tau propagation and neuron-to-neuron transmission. Propagation of Tau pathology and behavioral deficits induced by injection of Tau PFF in the hippocampus and overlying cortex are attenuated in mice lacking Lag3 selectively in neurons. These results identify neuronal Lag3 as a receptor of pathologic Tau in the brain，and for AD and related Tauopathies, a therapeutic target.

Keywords: Tau; Tau preformed fibrils; cell‐to‐cell transmission; lymphocyte‐activation gene 3; receptor.

Figure 1

Lag3 binds with Tau PFF, but not Tau monomer. a) Tau‐biotin PFF binds to SH‐SY5Y cells transiently transfected with Lag3 in a saturable manner, as a function of Tau‐biotin total concentration (monomer equivalent for PFF preparations) while SH‐SY5Y cells that are transfected with empty vector exhibit Tau‐biotin PFF binding at high concentrations. b) Scatchard analysis. Data are the means ± SEM, n = 3 independent experiments. c) The binding of Tau‐biotin PFF to mouse primary cortical neurons is reduced by Lag3 knockout (Lag3^−/−). Tau‐biotin PFF WT‐KD = 251 nm, Lag3^−/−‐KD = 604 nm, estimated KD for neuronal Lag3 (dotted line, ΔLag3 = wild type minus Lag3^−/−) is 100 nm. Data are the means ± SEM, n = 3 independent experiments. d) Schematics of deletion mutants of Lag3 ectodomains. e) Tau‐biotin PFF binding to Lag3 and deletion mutants of Lag3: extracellular domains (ΔD1–ΔD4), intracellular domain (ΔICD, and subdomains of D1 domain (del1—5‐D1) (n = 3). ****P < 0.05, ****P < 0.0001 compared to FL, n.s., not significant, one‐way ANOVA followed by Dunnett's correction. Data are the means ± SEM.

Figure 2

Lag3 is required for neuronal uptake of Tau PFF. a) Live image analysis of the endocytosis of Tau PFF‐pHrodo in WT, WT+Lag3, Lag3^−/−, and Lag3^−/−+Lag3 neurons at indicated time points. Increased red fluorescence indicates higher endocytosis. Scale bar, 10 µm. b) Quantification of Tau PFF‐pHrodo, n = 3 independent experiments. c) Lag3^−/− reduced the neuronal uptake of Tau‐biotin PFF co‐localized with endosome marker (Rab7). WT or Lag3^−/− neurons were cultured, and received transient transfection with Lag3. WT, WT+Lag3, Lag3^−/−, and Lag3^−/−+Lag3 neurons were treated with 500 nm Tau‐biotin PFF for 2 h. Scale bar, 10 µm. d) Quantification of (c) from three independent experiments. One‐way analysis of variance (ANOVA) with Tukey's correction. e) Quantification of time series imaging of calcium‐dependent fluorescence of WT and Lag3^−/− neurons. Neurons were treated with 1 εm Fluo‐2 acetoxymethyl (AM) ester for 30 min followed by 500 nm Tau PFF and live images were acquired for the indicated time period. Error bars represent SEM. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 3

Lag3 is required for Tau pathology propagation and neuron‐to‐neuron transmission. a) Depletion of Lag3 reduces abnormal Tau aggregation (T49). WT, WT+Lag3, Lag3^−/−, and Lag3^−/−+Lag3 primary cortical neurons at 7 DIV were treated with Tau PFF or PBS for 14 days. Neurons were fixed with methanol to remove soluble Tau as described^{[ 2d ]} and immunostained with T49, a mouse Tau‐specific monoclonal antibody. Abnormal Tau aggregates were assessed by immunostaining with T49 antibody. Scale bar, 20 µm. b) Quantification of (a), n = 3 independent experiments. Aggregates Tau levels were normalized with DAPI signal. Error bars represent SEM. Statistical significance was determined using one‐way ANOVA followed with Tukey's correction, c) Depletion of Lag3 reduced hyper‐phosphorylated Tau at Ser202, Thr205 (AT8). Scale bar, 20 µm. d) Quantification of panel (c), n = 3 independent experiments. AT8 intensities were normalized with neuronal marker MAP2. Error bars represent SEM. Statistical significance was determined using one‐way ANOVA followed with Tukey's correction. e) Schematic of microfluidic neuron device with three chambers (chamber 1‐2‐3, C1‐C2‐C3), to separate neuron cultures. f) Transmission of aggregated Tau, 14 days after the administration of Tau PFF in C1. Different neurons in C2, listed as C1‐(C2)‐C3, are: WT‐(WT)‐WT, WT‐(WT+Lag3)‐WT, WT‐(Lag3^−/−)‐WT, WT‐(Lag3^−/−+Lag3)‐WT. Scale bar, 100 µm. g) Quantification of (f); error bar represents SEM, n = 3 independent experiments. Aggregates Tau levels were normalized with DAPI signal. Statistical significance was determined using one‐way ANOVA followed by Tukey's correction, *P < 0.01, **P < 0.01, ***P < 0.001, ****P < 0.0001, n.s., not significant.

Figure 4

Lag3 antibodies block Tau PFF binding to Lag3, and subsequent pathologic propagation and neuron‐to‐neuron transmission. a) Anti‐Lag3 410C9 blocks the binding of Tau‐biotin PFF to Lag3‐expressing SH‐SY5Y cells. b) Quantification of (a); Error bars represent means ± SEM, n = three independent experiments, Student's t‐test. ****P < 0.0001. c) AT8 phosphorylated Tau (P‐Tau) was reduced by 410C9 (30 nm) in primary cortical neurons. d) Quantification of c) Error bars represent means ± SEM, n = 3 independent experiments, hyper‐phosphorylated Tau at Ser202, Thr205 (AT8) levels were normalized with neuronal marker MAP2 signal. Student's t‐test. *P < 0.05, Scale bar, 50 µm. e) Schematic of a microfluidic device experimental design. The different combinations of neurons in C2, listed as C1‐(C2)‐C3, are: WT‐(WT+mIgG)‐WT, WT‐(WT+410C9)‐WT. f) Transmission of Tau pathology 14 days after the administration of Tau PFF in C1. 410C9 significantly reduces the neuron‐to‐neuron transmission of Tau pathology (P‐Tau). g) Quantification of (f). Error bars represent means ± SEM, n = 3 independent experiments, hyper‐phosphorylated Tau at Ser202, Thr205 (AT8) levels were normalized with DAPI intensity. Student's t‐test. ***P < 0.001, n.s., not significant. Scale bar, 50 µm.

Figure 5

Tau pathologic spreading is reduced in Lag3 neuronal conditional knockout (Lag3^{L/L‐N‐/−)} mice and reduction of Tau PFF‐induced behavioral deficits. a) Schema of in vivo experiment. Stereotaxic injection of Tau PFF was performed into the dorsal hippocampus and overlying cortex in Lag3^{L/L‐N‐/−} mice and Lag3^L/L mice. Pathology and behavioral assessment were performed 9 months after Tau PFF injection. b–g) Immunostaining of P‐Tau in the ventral Hippocampus (b,c), entorhinal cortex (d,e), and amygdala (f,g) 9 months after Tau PFF injection. Error bars represent SEM, and P‐values were calculated with two‐tailed paired student t‐test. **P < 0.01, ***P < 0.001, n = 5 mice, Scale bar, 100 µm. h–j) Lag3^{L/L‐N‐/−} reduces the behavioral deficits in the open‐field test, total horizontal activity (h), % central area (i), and sniffing time in the social interaction test (j). Error bars represent SEM, and P values were calculated with one‐way ANOVA followed by Tukey's correction. n.s., not significant, **P < 0.01, ***P < 0.001.

Figure 6

Proposed Model. Conditional Lag3 deletion or antibody‐dependent inhibition delays Tau PFF transmission: Tau PFF binding to Lag3 promotes transmission. Binding and endocytosis of Tau PFF are significantly reduced upon Lag3^−/− or binding to Lag3 antibodies, leading to delayed pathological transmission and toxicity (Image is prepared with Biorender).