Involvement of eNAMPT/TLR4 signaling in murine radiation pneumonitis: protection by eNAMPT neutralization

Abstract

Therapeutic strategies to prevent or reduce the severity of radiation pneumonitis are a serious unmet need. We evaluated extracellular nicotinamide phosphoribosyltransferase (eNAMPT), a damage-associated molecular pattern protein (DAMP) and Toll-Like Receptor 4 (TLR4) ligand, as a therapeutic target in murine radiation pneumonitis. Radiation-induced murine and human NAMPT expression was assessed in vitro, in tissues (IHC, biochemistry, imaging), and in plasma. Wild type C57Bl6 mice (WT) and Nampt^+/- heterozygous mice were exposed to 20Gy whole thoracic lung irradiation (WTLI) with or without weekly IP injection of IgG₁ (control) or an eNAMPT-neutralizing polyclonal (pAb) or monoclonal antibody (mAb). BAL protein/cells and H&E staining were used to generate a WTLI severity score. Differentially-expressed genes (DEGs)/pathways were identified by RNA sequencing and bioinformatic analyses. Radiation exposure increases in vitro NAMPT expression in lung epithelium (NAMPT promoter activity) and NAMPT lung tissue expression in WTLI-exposed mice. Nampt^+/- mice and eNAMPT pAb/mAb-treated mice exhibited significant histologic attenuation of WTLI-mediated lung injury with reduced levels of BAL protein and cells, and plasma levels of eNAMPT, IL-6, and IL-1β. Genomic and biochemical studies from WTLI-exposed lung tissues highlighted dysregulation of NFkB/cytokine and MAP kinase signaling pathways which were rectified by eNAMPT mAb treatment. The eNAMPT/TLR4 pathway is essentially involved in radiation pathobiology with eNAMPT neutralization an effective therapeutic strategy to reduce the severity of radiation pneumonitis.

Figure 1.. Radiation-induced NAMPT expression in human cells and murine lung tissues.

A. 8 Gy radiation exposure (24 hours) significantly increased NAMPT promoter luciferase reporter activities in lung endothelial cells (HPAEC), epithelial cells (HBE and BEAS), and fibroblasts (IMR90) (**p<0.01 each). B. 8 Gy-mediated increases in NAMPT promoter activities in lung epithelial cells (HBE) are time-dependent, peaking at 4 hours but remaining elevated at 48 hours (**p<0.01 each). C. Increased NAMPT promoter activities were accompanied by increased NAMPT protein content in human epithelial cells (HBE) exposed to 8 Gy radiation for 1, 4, 24 and 48 hrs. Cells were lysed with RIPA buffer and NAMPT protein detected by immunoblotting with rabbit anti-human NAMPT 1:10000 (BLR058F, Bethyl, Montgomery, TX), as we previously reported [25, 34]. Densitometry studies confirmed significantly increased NAMPT protein levels in HBEs after exposure to 8 Gy irradiation at 24 and 48 hours, compared non-irradiated controls. D. Immunohistochemical lung tissue staining with Image J quantification of NAMPT detected by immunoblotting with rabbit anti-human NAMPT (BLR058F, Bethyl, Montgomery, TX) as we previously reported [21] in C57BL/6J mice exposed to 20Gy WTLI. E. Western blotting of immunoreactive NAMPT in lung homogenates from control and 20 Gy WTLI-exposed murine lung tissues at 1, 2 and 4 weeks with densitometric measurements. Together, these results demonstrate significantly increased NAMPT expression at 1 week persisting at 2 and 4 weeks after WTLI compared to controls with expression intensity summarized by Image J software imaging or by densitometry.

Figure 2.. In vivo imaging of NAMPT expression in preclinical radiation pneumonitis.

A. Representative whole-body iQID images collected at 240 min post ^99mTc-ProNamptor™ monoclonal antibody (Pro-mAb) and ^99mTc-IgG (IgG) injection, respectively, in sham-IR, non-radiated mice (Ctrl) and 20 Gy-exposed (WTLI) C57BL/6J mice at 1 week. The entire chest region of interest (ROI) is outlined by the white dashed lines and the cardiac blood pool is outlined by the red dashed lines. The ROI volumes lying between the white and red dashed lines represents the lung volumes. These images represent specific increased WTLI-induced activity and NAMPT lung expression. B. A summary of the results of in vivo quantitative image analysis of radiolabeled Pro-mAb activity in control sham-IR mice (Ctrl, n=6) and in WTLI-exposed mice 1 week post WTLI exposure (n=7). Results were compared to ^99mTc-IgG -exposed WTLI-exposed mice (n=5) 1 week post exposure. Results are expressed as a ratio of left or right lung ROI activity to the soft-tissue background activity (measured over the thigh region) (* P < 0.05 compared to Ctrl; Φ P < 0.05 compared to RILI 1-wk Pro-mAb for either left or right lungs). C. Ex vivo lung autoradiograph imaging confirmed the in vivo imaging findings, showing marked increased Pro-mAb uptake in the WTLI lungs, as compared to lungs of sham-IR (Ctrl Pro-mAb) or mice receiving non-specific ^99mTc-IgG 1 week post WTLI exposure radiation. D. Ex vivo quantitative total-lung radioactivity of Pro-mAb and IgG measured and expressed as a percent of total injected dose per gram of tissue (%ID/g) at 1 week post WTLI exposure showing statistical differences in Pro-mAb lung uptake between the WTLI and control lungs. A 2.6-fold increase of Pro-mAb activity was observed in the mice at 1 week after radiation compared to Pro-mAb in Ctrl non-irradiated mice, and 1.9-fold higher than non-specific IgG lung activity at the same time after WTLI (P <0.05). The difference between non-specific IgG uptake in the 1 week post-irradiation mice and Pro-mAb in non-irradiated controls is statistically nonsignificant.* P < 0.05 compared to Ctrl; Φ P < 0.05 compared to RILI 1-wk Pro-mAb.

Figure 3.. eNAMPT neutralization reduces the severity of preclinical radiation pneumonitis.

A/B. Compared to sham-IR exposed C57BL/6J mice (inset), H&E staining revealed marked increases in vascular leakage, leukocyte infiltration and inflammatory lung injury in WT mice exposed to 20Gy radiation at 1 and 4 weeks post WTLI exposure (quantified by quantified by Image J analysis). WTLI 20Gy-exposed WT mice receiving either the eNAMPT-neutralizing pAb (4mg/kg, weekly intraperitoneally) or humanized mAb (0.4mg/kg, weekly intraperitoneally) demonstrated significant histologic reductions in WTLI-mediated inflammatory lung injury with the humanized eNAMPT mAb significantly more protective than the eNAMPT pAb compared to the PBS/IgG₁ control. C/D. Both WTLI-mediated BAL protein levels and BAL cells were increased at 1 week post WTLI with further significant increases observed at 4 weeks. Consistent with the histologic protection noted in panel A/B, eNAMPT-neutralizing pAb and mAb treatments produced marked reductions in WTLI-mediated BAL protein levels and BAL cells assessed 4 weeks post WTLI exposure. E. Evaluation of the Radiation-Induced Lung Injury Severity Score (RILISS), integrating inflammatory indices (H&E staining, BAL protein and cells), demonstrates an overall 60% reduction in RILISS by eNAMPT-neutralizing pAb and mAb treatments. The humanized mAb again provided significantly greater protection against WTLI-induced pneumonitis than the eNAMPT pAb. F/G/H. The MesoScale ELISA-based Discovery platform was utilized to assess eNAMPT neutralization effects on plasma cytokine levels in preclinical radiation pneumonitis. Depicted are mouse plasma levels of eNAMPT (F), and the inflammatory cytokines IL-6 (G) and IL-1b (H) obtained in control sham-IR exposed mice (n=5), untreated 20 Gy WTLI-exposed mice receiving weekly IP PBS/IgG₁ (n=7, 4 weeks) and WTLI-exposed mice receiving the eNAMPT-neutralizing mAb weekly IP (n=7, 4 weeks). Exposure to 20 Gy WTLI marked increases plasma eNAMPT, IL-6 and IL-1b levels which are significantly reduced by the eNAMPT-neutralizing mAb.

Figure 4.. Nampt^+/− heterozygous mice are protected from preclinical radiation pneumonitis.

A/B. Shown is histologic evidence (H&E staining) of moderately severe inflammatory lung injury in WTLI (20Gy)-exposed C57BL/6J wild type mice at 4 weeks compared to sham-IR exposed mice (inset). In contrast, WTLI-exposed Nampt^+/− heterozygous mice exhibited significantly reduced histologic inflammation and injury compared to WT mice, with quantification by Image J software. C/D. WTLI-exposed Nampt^+/− heterozygous mice (4 weeks) demonstrated significant increases in total BAL protein levels and BAL cells (* p< 0.05). When compared to WTLI-exposed WT mice, the total number of BAL cells was significantly reduced (** p<0.01) as was total BAL protein without reaching statistical significance. E. Comparison of WTLI-induced lung injury scores (RILISS) (4 weeks) revealed significantly increased RILISS in both WT and Nampt^+/− heterozygous mice compared with sham-IR exposed mice (* p<0.01) but with RILISS significantly reduced in Nampt^+/− heterozygous mice compared to WT mice (** p<0.01).

Figure 5.. Genomic interrogation of dysregulated genes/signaling pathways in preclinical radiation pneumonitis: effect of eNAMPT neutralization.

A. RNA sequencing of WTLI-exposed lung tissues (4 week) was performed and compared to non-radiated control mice. Shown is a volcano plot depicting the differentially-expressed genes (DEGs) with the X axis representing log2 transformed fold change (FC). A total of 2325 DEGs exhibiting a FDR < 0.01 and 336 exhibited a FC >1.0 (289 genes) or l <1.0 (47 genes). The top 4 upregulated genes with FC >1.5: MMP12 (−log10padj =82.2), Phida3 (−log10padj =60.8, Aen (−log10padj =46), and Eda2r (−log10padj =43.6) are not depicted. B. Shown are the top GO biologic processes (molecular function) identified by STRING analysis of the 336 DEGs. Highlighted are the MAP kinase signaling and cytokine- and chemokine-mediated signaling. C. The role of circulating eNAMPT in driving dysregulated lung inflammatory signaling and injury/repair processes was assessed by RNA sequencing of lung tissues from WTLI-exposed mouse lung tissues (4 week) with and without eNAMPT mAb treatment. Shown is a volcano plot depicting the 207 DEGs filtered from exhibiting a FDR < 0.05 (far fewer DEGs than in Panel A), with 55 genes exhibiting FCs >1.0 and 12 with FCs <1.0). D. STRING interaction networks of the 67 significant eNAMPT-influenced DEGs (adjusted p<0.05, FC +/− 1). Network nodes represent proteins and the edges indicate the functional association, the line thickness indicates the strength of the data support. These analyses highlighted a very strong signal for IL-1b as a hub of eNAMPT mAb-influenced interacting genes.

Figure 6.. eNAMPT neutralization attenuates dysregulated signaling pathways in preclinical radiation pneumonitis.

A. Compared to sham-IR-exposed mice (n=3), Western blot studies of lung tissue homogenates from WTLI-exposed, PBS/IgG-treated mice (n=5, 4 weeks) showed prominent phosphorylation of NFKB compared to total NFkB protein content (total β-actin as loading control). In contrast, phospho-NFKB protein content was markedly reduced in WTLI-exposed mice (n=5) receiving the eNAMPT-neutralizing mAb (0.4mg/kg, weekly) compared with PBS/IgG-treated mice. These results were captured by densitometric evaluation of the ratio of p-NFkB/NFkB (fold change) (*p<0.05 control vs. untreated WTLI; **p<0.05 mAb WTLI vs. untreated WTLI). B. Similar to Panel 7A, lung tissue homogenates from control and WTLI-exposed mice (n=5, 4 weeks) were evaluated for MAP kinase family activation by levels of ERK, JNK p38 phosphorylation. These studies demonstrated robust MAP kinase family pathway activation as evidenced by protein phosphorylation (pp-p38, pp-JNK, pp-42/44 ERK) which was reversed by treatment with the eNAMPT-neutralizing mAb, results captured by densitometric evaluation. *p<0.05 control vs. untreated WTLI; **p<0.05 mAb WTLI vs. untreated WTLI.