Fig 1.
S1 invasion to the brain and its effects on learning and memory.
(A) Intranasally injected S1 (His-tagged) in the olfactory bulb, striatum, and hippocampus was examined at 3 hr post S1 administration (0.5 μg/per animal) using immunoblotting. Representative images are presented. (B) Schematic of behavior tests. (C-G) Episodic-like memory task. (C) The episodic-like memory task consists of two sample phases and a test phase. In each sample phase, rats encountered two sets of four identical novel objects (old objects, sample 1; recent objects, sample 2). In the test phase, objects were mixed together. One of the old familiar objects was placed in new location (b: old familiar displaced). The other three objects were placed in same location as in sample phases (a: old familiar stationary, c: recent familiar stationary). (D) Distribution of exploratory time for ‘what-when’ (old familiar objects vs. recent familiar objects). (E) Recency discrimination of exploratory times for ‘what-when’ (old stationary objects vs. recent stationary objects) (F) Distribution of exploratory time for ‘what-where’ (old familiar displaced vs. recent familiar stationary). (G) Total exploratory distance for each group in the test phase. (H-M) Water maze test. (H) Representative swimming paths of the control and S1 injected group during the probe trial. (I-J) The escape latencies in the two groups over five consecutive training days (I) and on the 5th day of the spatial acquisition session (J). (K-L) The percentage of time spent and distance traveled in the target quadrant during the probe trial (K), the average crossing number over the platform-site, and the latency of the first target-site crossover (L). (M) The average swimming speed of two groups. (O-R) Open field test. (O) Representative track sheets of the control and S1 injected group during the test trial. (P-R) Graphs showing alterations in time spent in central zone (P), distance traveled in central zone (Q), and overall distance traveled (R). Values are presented as means±SEMs. * p < 0.05, ** p < 0.01, ***p < 0.001 vs. control. Con, control group (n = 12); S1, S1 injected group (n = 12).
Fig 2.
S1-mediated transcriptome changes associated with synapse activity, immune response or ROS regulation.
(A) Schematic of gene expression profiling in animals injected with S1 (0.5 μg/ per animal, intranasal). (B) Gene ontology (GO) analysis of differentially expressed genes (DEGs) between control and S1 injected group showing the top enriched pathways of DEGs. (C) Chord plot showing the overlapped genes among positive regulation of apoptotic process (GO:0043065), response to hypoxia (GO:0001666), excitatory postsynaptic potential (GO:0060079), chemical synaptic transmission (GO:0007268), cellular response to type II interferon (GO:0071346). Red indicates upregulated and blue for downregulated. (D-L) Gene Set Enrichment Analysis (GSEA) for the samples of control vs. S1 injected group revealed that: genes involved in synapse activity (D-I) that were downregulated, and genes involved in immune response (J-K) and ROS regulation (L) were upregulated in the S1 injected group. NES (normalized enriched score), p-value, and FDR q-value are indicated.
Fig 3.
S1-mediated expression change of NMDAR2A and JPH3 in the hippocampus.
(A) Schematic of immunohistochemistry and western blot in animals injected with S1 (0.5 μg/per animal, intranasal) (B-D) Hippocampus tissues were obtained at 1 or 6 weeks after S1 injection (0.5 μg/ animal, intranasal). The protein levels of NMDAR2A and JPH3 were measured by immunoblotting (B) and protein levels of NMDAR2A (C) and JPH3 (D) are presented as means±SEMs (n=3). (E-H) Coronal brain sections of hippocampus were obtained at 1 week after S1 injection and stained using anti-NMDAR2A (E-F) and anti-JPH3 (G-H) antibodies. The insets (E-H) are high-magnification photographs of the white boxes. Images are representative of three independent experiments. Scale bars in E-H represent 300 μm and those in insets represent 50 μm. * p < 0.05 vs. control group. Con, control group; S1, S1 injected group.
Fig 4.
The effect of S1 to the stabilization of HIF-1α protein and gene expression related to synaptic function.
(A-B) S1 (0.5 μg/ animal, intranasal) were injected and HIF-1α was assessed by immunoblotting at 1 or 6 weeks after S1 administration. Representative images are presented. The protein levels of HIF-1α at 1 or 6 weeks after S1 injection were measured by immunoblotting (A) and protein levels at each time points (B) are presented as means±SEMs (n = 3). * p < 0.05, ** p < 0.01 vs. PBS-treated control group. (C) Coronal brain sections of the hippocampus (CA1 and CA3) were obtained at 1 or 6 weeks after S1 administration and stained using anti-HIF-1α, anti-NeuN antibodies and DAPI. Photographs are representative of three independent experiments. Scale bars in C represent 100 μm. (D-E) N2a cells were incubated with S1 (0.5 μg/ mL) for 3, 6, 9, 24 hr and HIF-1α was examined by immunoblotting (D). β-actin was used as a loading control. The protein levels at each time points (E) are presented as means±SEMs (n = 3). * p < 0.05 vs. PBS-treated N2A cells. (F-I) Schematic of HIF-1α knockdown analysis in N2A cell line (F). N2a cells were transfected with murine HIF-1α siRNA (siHIF-1α, 75 pM per 106 cells) or control siRNA (siCon), and then treated with S1 (0.5 μg/ mL) for 24 hr. The protein level of HIF-1α was examined using immunoblotting (G), and the quantitative data are presented as means±SEMs (n = 3) in (H). The gene expressions of GRIN2A, JPH3, and SHANK1 were assessed using a realtime PCR analysis (I). mRNA levels are presented as means ± SEMs. * p < 0.05, ** p < 0.01, ***p < 0.001 vs. control+siCon treated cells, #p < 0.05, ## p < 0.01 vs. S1 + siCon treated cells. Con, control group; S1, S1 injected group.
Fig 5.
Accumulation of Alzheimer’s disease or Parkinson’s disease-related proteins and neural damage in the hippocampus following S1 administration.
(A-D) Coronal sections of hippocampus (CA3) were obtained at 1 or 6 weeks after S1 injection (0.5 μg/ animal, intranasal) and stained using anti-NeuN antibody (A-C). The number of NeuN+ cells in CA3 region of hippocampus in control and S1-injected rats (6 weeks) are presented as means ± SEMs (n = 3) in (D). Photographs are representative of three independent experiments. Scale bars in A-C represent 200 μm. (E-N) Hippocampus of 6 weeks after S1 injection and control group were stained using anti-p-tau (T205) (E-H) and anti-aggregated α-synuclein (K-L) antibodies. The insets (G-H and K-L) are high magnification images of the black boxes in E-F and K-L. Photographs are representative of three independent experiments. Scale bars in E-F and K-L represent 100 μm; those in the high magnification photographs (G-H, K-L) and insets represent 20 μm. The protein levels of p-tau, total tau (I-J) and aggregated α-synuclein (M-N) were measured. Representative immunoblots and protein levels presented as means ± SEMs (n = 3). (O-Q) Brain sections of the rats at 6 weeks post S1 injection were prepared and processed for TUNEL assay and staining with anti-NeuN antibody. Representative images are shown (O-P), and the numbers of TUNEL-/ NeuN+ , TUNEL+/ NeuN+, p-tau+ , and α-syn + cells were counted in the CA3 region (0.1 mm2)(Q) and presented as means±SEMs (n = 21 from three animals). Con, control group; S1, S1 injected group.
Fig 6.
The effects of metformin to the expression of synaptic function-related genes and neuropathological protein aggregation induced by S1 protein.
(A) Schematic of metformin treatment in N2A or HGS1 cell lines. (B) Effects of metformin on the expression of GRIN2A, JPH3, SHANK1, and GRIA2 in N2A cells were examined by real-time PCR after co-treatment with metformin (10 mM) and S1 (0.5 μg/ mL) for 24 hr. (C-F) Effects of metformin on the neuropathological protein aggregation were evaluated by immunoblotting after co-treatment of metformin and S1 protein for 24 hr. α-synuclein (C) and p-tau and total tau (E) were measured. The protein levels of α-synuclein (D) and p-tau/total tau (F) are presented as means±SEMs (n = 3). (G-I) Effects of metformin on HIF-1α stabilization and sumoylation were performed with H4 and HGS1 cells using immunoblotting (G). The protein levels of HIF-1α (H) and Sumo-1 (I) are presented as means±SEMs (n = 3). Representative images are shown. * p < 0.05, **p < 0.01 vs. control group, #p < 0.05, ##p < 0.01, ###p < 0.001 vs. S1 only treated group. Con, control group; S1, S1 injected group.
Fig 7.
Schematic illustration of SARS-CoV-2 S1 spike protein–induced cognitive impairment and `the protective role of metformin.
Intranasally administered SARS-CoV-2 S1 protein enters the rat brain and leads to cognitive decline associated with learning and memory deficits. S1 induces hypoxia-independent stabilization of HIF-1α, which alters synaptic plasticity–related gene expression. S1 exposure also promotes the accumulation of pathological proteins such as phosphorylated tau and α-synuclein through distinct pathways, collectively contributing to neuronal dysfunction. Metformin alleviates these effects by suppressing HIF-1α stabilization, thereby restoring the expression of synaptic plasticity-related genes, and by reducing the accumulation of pathological proteins, ultimately contributing to recovery from cognitive dysfunction.