Genetically-engineered Salmonella typhimurium expressing FGF21 promotes neurological recovery in ischemic stroke via FGFR1/AMPK/mTOR pathway

Abstract

Background: Ischemic stroke (IS) remains a leading cause of mortality and disability, with limited therapeutic options due to poor drug delivery to ischemic lesions. To address this challenge, an engineered Salmonella based therapeutic method for targeted drug delivery and long-term treatment is herein designed to mitigate ischemic damage.

Methods: We engineered an attenuated luminescent Salmonella typhimurium (S.t -ΔpG) strain with an L-arabinose-inducible pBAD system to secrete bioactive FGF21. C57BL/6 mice were used to to measure neuron apoptosis and the activity of immune cells following IS induction plus S.t-ΔpG injection. Bioluminescence imaging was applied for bacterial colonization. ELISA and glucose uptake assays were performed to detect FGF21 secretion and the bioactivity. Neurological tests, TTC staining, and TUNEL labeling were used to assess the therapeutic effects of barterially secreted FGF21. Immunofluorescence assay of FGF21/FGFR1 dominant pathway was explored to investigate neuroprotective mechanism, while IBA-1 staining, CD3/CD68 immunostaining, cytokine profiling, and hepatorenal histopathology were detected to evaluate biosecurity.

Results: S.t-ΔpG^FGF21 selectively colonized peri-infarct regions and secreted functional FGF21, reducing neurologic deficits (48%) and infarct volume (46%) versus controls (p < 0.01). Mechanistically, immunofluorescence demonstrated that bacterially secreted FGF21 activated neuronal FGFR1/AMPK/mTOR pathway to enhance autophagy, whereas autophagy inhibition abolished its neuroprotection. Further, bacterial exclusion from neuron was validated via MAP2/NeuN plus Salmonella co-staining in primary neuron cells and brain tissue. Critically, CD3/CD68 immunostaining, serum cytokine profiling, and hepatorenal histopathology confirmed the long-term biosafety of this approach.

Conclusion: Our study presents a novel, Salmonella - based platform for targeted and sustained FGF21 delivery, offering a promising therapeutic strategy for ischemic stroke with robust efficacy and minimal systemic toxicity.

Keywords: Salmonella typhimurium; FGF21; FGFR1/AMPK/mTOR pathway; Hepatorenal histopathology; Ischemic stroke; Neurologic deficit.

Fig. 1

Salmonella (S.t-ΔpG^lux) Targets to cerebral infarct zones. (A) Bioluminescent signal of Salmonella (S.t-ΔpG^lux) on LB plates; (B, C, D) Experimental timeline and bioluminescence images of animals after injection with Salmonella (S.t-ΔpG^lux). Images were captured and quantified on 1 d, 3 d, 5 d, 7 d, 9 d, and 14 d after injection, n = 6; (E) Global scanning of luminous Salmonella captured from the ischemic semi-dark band. Salmonella was labelled in green, and nuclei were stained in blue, scale bar = 1 mm, n = 6

Fig. 2

Engineering and induction of FGF21-expressing bacteria (S.t-ΔpG^FGF21) in vitro. (A) Genomic map of the engineered plasmid pBAD-PelB-FGF21-3*Flag; (B) Immunoblot analysis of bacterially secreted FGF21. Samples were prepared with (+) or without (-) 0.2% L-Arabinose and separated into pellet and supernatant (Sup) fractions. (C) Quantitative transcription analysis of gult1 upon bacterial secreted FGF21 in PC-12 cells, n = 3. (D) Global glucose level determined on 0, 6, 12, and 24 h after bacterial injection, n = 5. ** p < 0.01, and *** p < 0.001; ns, no significance

Fig. 3

S.t-ΔpG^FGF21(+) ameliorates neurological deficits and reduces the volume of cerebral infarct foci in stroke mice. (A) Experiments timeline; (B) ELISA analysis of FGF21 expression in vivo, n = 5. (C) mNSS, n = 12; (D) Rotarod test, n = 12; (E) Representative images of TTC stained brain slices on 5 days after dMCAO; (F) Quantitative analysis of brain infarct volumes, n = 6. * p < 0.05, ** p < 0.01, and *** p < 0.001; ns, no significance

Fig. 4

S.t-ΔpG^FGF21(+) improves neuronal apoptosis after stroke. (A) Representative images of TUNEL and NeuN co-staining in different groups, scale bar = 100 μm; (B) Quantitative analyses of neuronal apoptosis rate in each group, n = 6; (C-F) Expression level and relative quantitative analysis of Bax, Bcl-2, and Cleaved-Caspase3, n = 6. * p < 0.05, ** p < 0.01, *** p < 0.001

Fig. 5

S.t-ΔpG^FGF21(+) promotes long-term stroke recovery in mice model. (A) Representative images of TTC stained brain slices on 28 days after dMCAO; (B) Quantitative analysis of brain infarct volumes, n = 6; (C) Representative images of NeuN staining in different groups, scale bar = 100 μm; (D) Quantitative calculation of surviving neurons in each group, n = 6. ** p < 0.01, *** p < 0.001

Fig. 6

Salmonella was excluded from Neuronal Cells. (A) Co-culture of PC-12 cells and Salmonella (S.t-ΔpG ^FGF21). PKH26 (red) labelling of cell membranes and anti-Salmonella typhimurium LPS antibody labelling Salmonella (green), n=5, scale bar = 100 μm; (B-C) Immunofluorescence staining and quantitative analyses of FGF21 protein (green), n=5, scale bar = 100 μm; (D-E) Immunofluorescence staining and quantitative analyses of Flag fragment (green), n=5, scale bar = 100 μm; (F) Identification of primary neuron with MAP2 (red) and NeuN (green). Positive cells are greater than 90%; (G) Co-culture of primary neuron cells and U251 cells with Salmonella (S.t-ΔpG ^FGF21). MAP2 (red) labelling of neuronal cytoplasm and anti-Salmonella typhimurium LPS antibody labelling Salmonella (green), n=5, scale bar = 100 μm; (H-I) Immunofluorescence staining and quantitative analyses of FGF21 protein (green) in primary neuron cells, n=5, scale bar = 100 μm; (J) Immunofluorescence analysis of co-stained neurons (NeuN) with Salmonella in cortex. n=5, scale bar = 100 μm.

Fig. 7

S.t-ΔpG^FGF21(+) enhances autophagy through binding to the FGFR1 receptor to activate the AMPK-mTOR pathway. (A-B) Immunofluorescence staining and quantitative analyses of p-FGFR1 (green) in PC-12 cells, n = 5, scale bar = 100 μm; (C-D) Immunofluorescence staining and quantitative analyses of p-FGFR1 (green) in primary neuron cells, n = 5, scale bar = 100 μm; (E-F) Representative images of p-FGFR1 (green) and NeuN (red) staining on day 5 in the cerebral cortex, n = 6, scale bar = 100 μm. (G-J) Expression level and relative quantitative analysis of FGFR1, p-FGFR1, AMPK, p-AMPK, and p-mTOR in the cerebral cortex on day 5 post stroke, n = 6; (K-N) Representative western blot images and relative quantification analysis of autophagy relative protein including p62, ATG5, and LC 3 in the cerebral cortex of each group, n = 6. * p < 0.05, ** p < 0.01, *** p < 0.001; ns: no significant difference

Fig. 8

Autophagy inhibition reverses the effect of bacterially secreted FGF21 on neuroprotection after stroke. (A-D) Expression level and relative quantification of p62, ATG5, and LC3 expression in the cerebral cortex of different groups, n = 6. (E-H) Expression level and relative quantification of Bax, Bcl-2, and Cleaved-Caspase3, n = 6; (I-J) Representative images and quantitative analysis of apoptosis rate of neuronal cells in each group, scale bar = 100 μm, n = 6. * p < 0.05, ** p < 0.01, *** p < 0.001

Fig. 9

Assessment of Immune cells infiltration around cerebral infarction and detection of systemic levels of inflammatory cytokines in mice. (A-B) Representative images and quantitative analysis of Iba1 staining on 3 days after dMCAO, n = 6, scale bar = 100 μm; (C-D) Representative images and quantitative analysis of Iba1 staining on 14 days after dMCAO, scale bar = 100 μm; (E-J) Representative images and quantitative analysis of CD3 staining on 1, 3, 14 days after dMCAO, n = 6, scale bar = 100 μm; (K-P) Representative images and quantitative analysis of CD68 staining on 14 days after dMCAO, n = 6, scale bar = 100 μm; (Q-U) Quantification of IL-1β, IL-6, TNF-α,MCP-1 and IFN-γ in each group was assessed by ELISA, n = 5. * p < 0.05, ** p < 0.01, *** p < 0.001; ns: no significant difference

Fig. 10

Effect of S.t-ΔpG^FGF21 infection on body weight and quantification of liver and kidney functions. (A) Body weights of different groups of mice on 3, 6, 9, and 12 days after bacterial injection, n = 12; (B) Fresh morphology of liver and kidney tissue, n = 6; (C) Histopathological analysis of liver and kidney tissue on 1, 3, 14 days after bacterial injection, n = 6, scale bar = 100 μm; (D-E) Liver function determined on 1,2,3,5 days after bacterial injection, n = 9. The normal reference values of ALT were 10.06–96.47 U/L, and the normal reference values of AST were 36.31-235.48 U/L; (F-G) Kidney function of different groups of mice determined on 1,2,3,5 days after bacterial injection, n = 9. Normal reference range of CREA: 10.91–85.09 umol/L; Normal reference range of BUN: 10.81–34.74 mg/dL.

Fig. 11

Schematic illustration of therapeutic mechanism of engineered Salmonella (S.t-ΔpG^FGF21) in cerebral infarct foci