Combination of Dexamethasone and Tofacitinib Reduces Xenogeneic MSC-Induced Immune Responses in a Mouse Model of Alzheimer's Disease

Abstract

We have recently reported on how transplantation of human mesenchymal stem cells (MSCs) into the mouse parenchyma generated immune responses. To facilitate the clinical translation of MSC-based AD therapy, the safety and efficacy of human derived MSCs (hMSCs) must be confirmed in the pre-clinical stage. Thus, it is imperative to investigate measures to reduce immune responses exerted via xenotransplantation. In this study, immunosuppressants were co-administered to mice that had received injections of hMSCs into the parenchyma. Prior to performing experiments using transgenic AD mice (5xFAD), varying immunosuppressant regimens were tested in wild-type (WT) mice and the combination of dexamethasone and tofacitinib (DexaTofa) revealed to be effective in enhancing the persistence of hMSCs. According to transcriptome sequencing and immunohistochemical analyses, administration of DexaTofa reduced immune responses generated via transplantation of hMSCs in the parenchyma of 5xFAD mice. Significant mitigation of amyloid burden, however, was not noted following transplantation of hMSCs alone or hMSCs with DexaTofa. The efficacy of the immunosuppressant regimen should be tested in multiple AD mouse models to promote its successful application and use in AD stem cell therapy.

Keywords: Alzheimer’s disease; dexamethasone; immune response; immunosuppressant; tofacitinib.

Figure 1

Examining the effects of various immunosuppressant regimens in WT mice transplanted with hMSCs. (A) Experimental Scheme. Dexamethasone (Dexa) was administered via the IV route at D0, right before and 4 h following hMSC transplantation. Both tacrolimus (Tac) and tofacitinib (Tofa) were administered orally starting at D-2 up to the sacrifice time point (D7). (B) Quantification of residual hMSCs in the WT mouse parenchyma carried out via ALU qPCR (hMSC: n = 7, hMSC+Dexa: n = 8, hMSC+Tac: n = 6, hMSC+DexaTac: n = 7, hMSC+Tofa: n = 9, hMSC+DexaTofa: n = 6). The highest number of residual hMSCs was observed from the hMSC + DexaTofa group. Statistical significance is defined as * p < 0.05 vs. hMSC; mean ± SEM (one-way ANOVA, Dunnett’s multiple comparison). (C) Representative images of CD45-positive leukocytes visualized as a dark brown precipitate (hMSC: n = 3, hMSC+Dexa: n = 3, hMSC+Tac: n = 3, hMSC+DexaTac: n = 3, hMSC+Tofa: n = 2, hMSC+DexaTofa: n = 2). Compared to hMSC transplantation alone, co-administration of immunosuppressants reduced leukocyte infiltration overall. Scale bars: whole brain =2 mm, magnified inset area (solid black box) = 60 µm.

Figure 2

Whole-Transcriptome sequencing analysis revealed that combined administration of dexamethasone and tofacitinib reduced immune response in 5xFAD mice. (A) Differentially Expressed Genes (DEGs) between hMSC vs. MEM and hMSC vs. hMSC + DexaTofa. Red nodes represent genes that are upregulated in the hMSC (n = 2) group, while blue nodes represent genes upregulated by the MEM (top panel, n = 2) or hMSC + DexaTofa (bottom panel, n = 2) groups. Gene Ontology (GO) analysis of DEGs (right panel). (B) Gene Set Enrichment Analysis (GSEA) indicated that compared to MEM media administration, transplantation of hMSCs demonstrated enrichments of gene sets associated with immune response. (C) When compared to the hMSC group via GSEA, gene sets associated with immunodeficiency are revealed from the hMSC + DexaTofa group.

Figure 3

Combined administrations of dexamethasone and tofacitinib reduced PD-1 expression. (A) Representative images of PD-1 positive cells (solid red arrows) in the left hemisphere of hMSC (n = 4) and hMSC + DexaTofa (n = 6) groups. Higher PD-1 expression is evident from the hMSC + DexaTofa group. Statistical significance was defined as ** p < 0.01 vs. hMSC; mean ± SEM (unpaired t-test). Scale bars: whole brain = 2 mm, magnified inset area (solid black box) = 60 µm. (B) Representative images of CD45-positive leukocytes (dark brown precipitate, solid red arrows) for the MEM (n = 3), hMSC (n = 4), and hMSC + DexaTofa (n = 6) groups. Compared to the hMSC group, the hMSC + DexaTofa group displayed reduced levels of CD45-positive leukocytes. Scale bars: whole brain = 2 mm, magnified inset area = 60 µm. Values are expressed as mean ± SEM.

Figure 4

Quantification of residual hMSCs in the 5xFAD parenchyma. (A). Representative images of STEM121 positive cells (residual hMSCs). Non-specific signals are present (solid black arrowhead). Residual hMSCs (solid red arrows) were barely detected at D7 for both the hMSC and hMSC + DexaTofa groups. Scale bars: whole brain = 2 mm, magnified inset area (solid black box) = 60 µm. (B) ALU-qPCR was used to quantify the number of residual hMSCs present in the parenchyma for both the hMSC (n = 3) and hMSC + DexaTofa (n = 2) groups. Human DNA was not detected from the MEM (n = 2) group and compared to the hMSC group, higher MSC quantity was revealed (difference is not statistically significant) from the hMSC + DexaTofa group. Values are expressed as mean ± SEM.

Figure 5

Comparison of amyloid burden between the injected (left caudate putamen) and non-injected (right caudate putamen) areas of 5xFAD mice. Representative images of 6E10 stained sections to assess amyloid deposition in the left (injected area) and right caudate putamen of the MEM (n = 3), hMSC (n = 4), and hMSC + DexaTofa groups (n = 6). L: left, R: right. Beta amyloid was visualized as dark brown precipitates. The injected area (left putamen) displayed reduced levels of amyloid deposition for both the hMSC and hMSC + DexaTofa groups. but the differences between the two hemispheres were statistically insignificant (two-way ANOVA, Tukey’s multiple comparison). Scale bars: whole brain = 2 mm, magnified inset area = 300 µm. Values are expressed as mean ± SEM.

Figure 6

Assessment of changes in amyloid deposition in the cortex, hippocampus, and thalamus. (A) Representative 5xFAD brain tissue sections acquired from the following groups: MEM (n = 3), hMSC, and hMSC (n = 4) + DexaTofa (n = 6) were stained using the 6E10 antibody to examine differences in amyloid burden (dark brown precipitate) among the groups. Scale bar: whole brain = 2 mm. (B) The amyloid burden of the left and right hemispheres was evaluated first in the following three regions: hippocampus, and thalamus. Significant differences between the hemispheres were not noted for all three regions. mean ± SEM (two-way ANOVA, Tukey’s multiple comparison). (C) Data acquired from the left and right hemispheres were combined for each of the three regions. Out of the three groups, the hMSC + DexaTofa group displayed the highest level of amyloid deposition for all three regions. Statistical significance is defined as * p < 0.05 vs. hMSC; mean ± SEM (one-way ANOVA, Dunnett’s multiple comparison).

Figure 7

Thioflavin-S staining of the left and right caudate putamen. Representative Thioflavin-S-stained tissue images of the MEM (n = 3), hMSC (n = 4), and DexaTofa (n = 6) groups acquired via Vectra (whole brain) and confocal microscopy (magnified image). Green dots (fluorescence signal) indicate location of amyloid plaques. Scale bars = whole brain: 2 mm, magnified inset area = 100 µm. (L vs. R) mean ± SEM (two-way ANOVA, Tukey’s multiple comparison).

Figure 8

Thioflavin-S staining of the cortex, hippocampus, and thalamus of 5xFAD mice. (A) Representative Thioflavin-S-stained tissue images of the MEM (n = 3), hMSC (n = 4), and hMSC+DexaTofa (n = 6) groups were acquired via a confocal microscope. Green dots (fluorescence signal) indicate the location of amyloid plaques. Scale bar for each image = 100 µm. (B) Thioflavin-S burden % of the left (L) and right (R) hemispheres were quantitated for the cortex, hippocampus, and thalamus. Differences between the left and right hemispheres were not statistically significant for all 3 regions (mean ± SEM, two-way ANOVA, Tukey’s multiple comparison). (C) Data collected from the left and right hemispheres were combined and averaged for each region Statistical significance is defined as * p < 0.05, ** p < 0.01, *** p < 0.001 vs. hMSC; mean ± SEM (one-way ANOVA, Dunnett’s multiple comparison).