Targeting RAGE with Nanobodies for Molecular Imaging of Cancers and Alzheimer's Disease

Abstract

The receptor for advanced glycation end products (RAGE) is a multifunctional cell surface receptor implicated in aging and the progression of chronic diseases, including cancer and Alzheimer's disease. Its interaction with advanced glycation end products (AGEs) promotes cellular stress and inflammation, underscoring the diagnostic and therapeutic relevance of targeting RAGE. In this study, we explored the potential of nanobodiessingle-domain antibodies known for high specificity, strong affinity, and deep tissue penetrationas molecular tools for RAGE-targeted applications. Using a phage display library, a panel of RAGE-specific nanobodies were isolated and characterized. Binding activity and affinity were evaluated through enzyme-linked immunosorbent assay (ELISA) and surface plasmon resonance (SPR) assays. Among them, nanobody NbF8 demonstrated the highest affinity and specificity toward RAGE. In vitro, NbF8 selectively bound RAGE-expressing cells, while in vivo imaging in renal carcinoma and Alzheimer's disease mouse models confirmed its targeted accumulation in RAGE-overexpressing tumors and brain tissues. These findings highlight NbF8 as a promising molecular imaging agent for RAGE-associated diseases. This study supports the potential of RAGE-targeting nanobodies in both diagnostic imaging and therapeutic development, offering a novel approach for precision medicine in conditions driven by RAGE signaling.

Keywords: Alzheimer's disease; RAGE (Receptor for Advanced Glycation End Products); in vivo imaging; molecular diagnostics; nanobodies; renal carcinoma.

Figure 1

RAGE is overexpressed in renal cancer and Alzheimer's disease. A) RAGE RNA expression (TPM, RNA seq) in renal cancer and normal kidney tissues. n = 523 for renal cancer and n = 72 for normal group, ^* p < 0.05. B) The correlation between RAGE expression levels and the prognosis of renal cancer patients. C) The expression of RAGE in different renal cancer cell lines was detected by Western blot. D) The expression levels of RAGE in Alzheimer's disease model cell lines were detected by Western blot. E) The differential expression of RAGE in the hippocampal tissue of AD mice and wild‐type (WT) mice was detected by immunofluorescence (IF). F) Images were processed using ImageJ to quantify the average area of RAGE. Scale bar:50 µm. Results are quantified as mean ± SEM. Data are representative from n = 3 mice in each group. Statistical comparisons across groups were performed using two‐tailed Student's t‐test. ^** p < 0.01.

Figure 2

Screening of RAGE nanobodies. A) SDS‐PAGE analysis of purified human RAGE (hRAGE), showing a molecular weight between 40 and 45 kDa. B) Three‐round screening of nanobodies against RAGE using phage display technology. C) SDS‐PAGE analysis of purified nanobodies (Left, 15–20 kDa), confirmation of nanobody binding with HA antibody (Right, 15–20 kDa). D) The binding activity of NbH4, NbF8, NbF9, NbA12, NbD3, and control nanobody to hRAGE was evaluated using ELISA. E) Quantitative analysis of the binding activity of NbH4, NbF8, and NbF9 to hRAGE was performed again using ELISA. Results are quantified as mean ± SEM. Statistical comparisons across groups were performed using two‐way ANOVA. ^**** p < 0.0001. F–H) The binding affinity of nanobodies to hRAGE was measured by surface plasmon resonance (SPR) analysis. The equilibrium dissociation constants (KD) were 4.18 nm (NbF8), 26 nm (NbH4), and 21.6 nm (NbF9).

Figure 3

Characterization of RAGE nanobodies in renal cancer cells. A,C) The binding activity of NbF8, NbH4 and NbC9 nanobodies for RAGE‐positive cells (OSRC2 and KMRC1) was analyzed by Cell ELISA. B,D) Quantitative analysis showed that both NbF8 and NbH4 can recognize RAGE protein expressed on the membrane of renal cancer cells, with NbF8 exhibiting higher activity compared to the control NbC9 and NbH4. Data are present as mean ± SEM. Statistical comparisons across groups were performed using two‐way ANOVA. ^**** p < 0.0001. E) The binding activity of NbF8 to RAGE in OSRC2 and KMRC1 cells was detected by immunofluorescence (blue: DAPI; red: RAGE; green, nanobody NbF8, Scale bar:50 µm).

Figure 4

Imaging and tissue analysis of RAGE nanobodies NbF8 and NbC9 in renal cancer. A) Purified nanobodies NbF8 and NbC9 (left) and labeling efficiency analysis of nanobodies NbF8 and NbC9 via fluorescent imaging (right). B) In vivo imaging of renal cancer mice using Cy5‐labeled nanobodies NbF8 and NbC9 at different time points. C) Quantitative analysis revealed that the fluorescence signal of NbF8‐Cy5 was stronger than that of NbC9‐Cy5 at the indicated time points in tumor tissues. Results are quantified as mean ± SEM. Data are representative from n = 3 mice in each group. Statistical comparisons across groups were performed using two‐way ANOVA. ^* p < 0.05, ^** p < 0.01. D) Ex vivo imaging of tumors and major organs from mice. E) Quantitative analysis for the fluorescent signal of NbF8‐Cy5 and NbC9‐Cy5 in various tissues. Results are presented as mean ± SEM. Group data were compared using students’ t‐test. ^* p < 0.05, ^** p < 0.01. F) Immunohistochemical staining images of RAGE in tumor tissues (Scale bar: 20 µm). G) Immunohistochemical staining images of NbF8 and NbC9 in tumor tissues collected after in vivo imaging (Scale bar: 20 µm). H) Representative immunofluorescence colocalization (merge) images of NbF8 (green) and Cy5 (red) in tumor tissues (Scale bar: 100 µm). I) Representative immunofluorescence colocalization (merge) images of the vascular marker CD31 (green) and NbF8‐Cy5 (red) in tumor tissues (Scale bar: 100 µm).

Figure 5

The binding activity of NbF8 to mouse RAGE protein. A) SDS‐PAGE analysis of purified mRAGE (40–45 kDa). B) ELISA analysis of the binding capacity of nanobodies NbF8 and NbC9 to mRAGE. Results are presented as mean ± SEM. Group data were compared using two‐way ANOVA test, ^**** p < 0.0001 C,E) The binding activity of NbF8 and NbC9 nanobodies to mRAGE‐positive cells (Renca and N2a) was analyzed by Cell ELISA (C: Renca cells, E: N2a cells). D,F) Quantitative analysis showing that NbF8 can recognize mRAGE protein expressed on the cell membrane. Results are presented as mean ± SEM. Group data were compared using a two‐way ANOVA test, ^**** p < 0.0001.

Figure 6

Cy5‐labeled RAGE nanobody NbF8 could effectively image the brains of AD mice. A) in vivo imaging for AD mice with RAGE NbF8 at different time points (n = 3, blue circles represent ROI for imaging). B) Ex vivo organ imaging of brain, liver, heart, spleen, lung, and kidney dissected from AD mice after in vivo imaging shown in panel A (n = 3, the first row is labeled as 1): Control, the second row is designated as 2): NbC9‐Cy5, and the third row is labeled as 3): NbF8‐Cy5). C) Imaging quantitation analysis for A at each time point. The signal generated by NbF8‐Cy5 in the brains of AD mice was significantly higher than that of NbC9‐Cy5. The data are expressed as mean ± SEM. Data are representative from n = 3 mice in each group. Statistical comparisons across groups were performed using a two‐way ANOVA test. ^**** p < 0.0001. D) Quantitative analysis for various organs revealed that the fluorescence signal generated by NbF8‐Cy5 in ex vivo brain and lung tissues was significantly higher than that of NbC9‐Cy5. Data are representative from n = 3 in each group. Statistical comparisons across groups were performed using two‐tailed Student's t‐test. ^* p < 0.05.

Figure 7

Fluorescence intensity analysis of Cy5‐labeled nanobodies Nbf8 and Nbc9 in various organs from AD mice after imaging. A) Ex vivo fluorescence scanning of Cy5‐labeled nanobodies NbF8 and NbC9 in different organs (brain, liver, heart, spleen, lung, kidney) of AD mice 90 min postinjection (n = 3). B) Quantitative analysis indicated that the signal produced by NbF8 in the brain and lung of AD mice was significantly higher than that of NbC9. The data are presented as mean ± SEM. Data are representative of n = 3 in each group. Statistical comparisons across groups were performed using two‐tailed Student's test. ^* p < 0.05, ^** p < 0.01. C) Detection of NbF8‐Cy5 and NbC9‐Cy5 fluorescence signals in mouse brain tissues. Scale bar: 50 µm. D) Quantitative analysis for C showed that the fluorescence signal in the brain tissues of AD mice treated with NbF8‐Cy5 was stronger than that of NbC9‐Cy5. The data are expressed as mean ± SEM. Data are representative from n = 3 in each group. Statistical comparisons across groups were performed using two‐tailed Student's test. ^*** p < 0.001.