Molecular Visualization of Early-Stage Acute Kidney Injury with a DNA Framework Nanodevice

Abstract

DNA nanomachines with artificial intelligence have attracted great interest, which may open a new era of precision medicine. However, their in vivo behavior, including early diagnosis and therapeutic effect are limited by their targeting efficiency. Here, a tetrahedral DNA framework (TDF)-based nanodevice for in vivo near-infrared (NIR) diagnosis of early-stage AKI is developed. This nanodevice comprises three functional modules: a size-tunable TDF nanostructure as kidney-targeting vehicle, a binding module for the biomarker kidney injury molecule-1 (Kim-1), and a NIR signaling module. The cooperation of these modules allows the nanodevice to be selectively accumulated in injured kidney tissues with high Kim-1 level, generating strong NIR fluorescence; whereas the nanodevice with the proper size can be rapidly cleared in healthy kidneys to minimize the background. By using this nanodevice, the early diagnosis of AKI onset is demonstrated at least 6 h ahead of Kim-1 urinalysis, or 12 h ahead of blood detection. It is envisioned that this TDF-based nanodevice may have implications for the early diagnosis of AKI and other kidney diseases.

Keywords: DNA nanodevice; acute kidney injury; early diagnosis; kidney injury molecule-1; tetrahedral DNA framework.

Figure 1

Design and characterization of the engineering TDF nanodevice (Kim‐TDF). A) Schematic illustration of the construction of the Kim‐TDF. The Kim‐TDF₁₇ were analyzed by AFM. B) Characterization of Kim‐TDF₁₇ by native PAGE gel electrophoresis. C) The schematic illustration of fluorescence imaging of Kim‐1 with enhanced AKI‐to‐normal contrast by using Kim‐TDF nanodevice.

Figure 2

The characterizations of Kim‐TDF nanodevice's functionalities. A) The schematic illustration of degradation of nanodevice. B–D) The degradation kinetics of Kim‐TDF₇, Kim‐TDF₁₇ and Kim‐TDF₃₇ via FRET analysis. Data represent mean ± standard deviation (S.D.) (n = 3). E–G) The Kim‐1‐binding ability of Kim‐TDF. E) The CLSM images of the HK‐2 cells incubated with Cy3‐labelled nanodevice. Cell membrane were stained with 3,3'‐dioctadecyloxacarbocyanineperchlorate (DiO). The concentration of Kim‐TDF and corresponding TDF‐3R probe was 200 nM. Scale bars: 30 µm. F) FCM analysis of Kim‐1‐binding ability of Cy3‐labelled Kim‐TDF. The concentration of Kim‐TDF and corresponding TDF‐3R probe was 200 nM. G) Analysis of the binding affinity of Kim‐TDF with different targeting valence (1, 2, 3) for HK‐2 cells treated with H₂O₂. Data represent mean ± S.D. (n = 3). H) NIR fluorescence images of different groups at different scanning intervals (equivalent fluorophore concentration: 200 nM). I) The photobleaching efficiency of Kim‐TDF.

Figure 3

Renal clearance studies of TDFs. A) Schematic illustration of the excretion of TDFs probes through the urinary tract. B) RCE of TDFs at different post‐injection time. Data represent mean ± S.D. (n = 4). C) The amount of TDFs excreted from renal pathway (blue bar) and enterohepatic pathway (gray bar) of mice at 24 h post‐injection. Data represent mean ± S.D. (n = 4). D) Blood concentration (% ID) decay of nanodevices with different module configurations in healthy mice. Data represent mean ± S.D. (n = 4). E) Analysis of the amount of TDFs with different module configurations excreted from kidneys into urine and residual TDFs in major organs of mice after 24 h injection of TDFs. Data represent mean ± S.D. (n = 4). F) NIR images of healthy mice at different post‐treatment time points of Kim‐TDFs. G) NIR fluorescence intensities of kidneys and H) bladder at different post‐treatment time points of Kim‐TDFs in healthy mice. Data represent mean ± S.D. (n = 3). I) Percentage of kidney contrast enhancement at different post‐treatment time points of Kim‐TDFs in healthy mice. Data represent mean ± S.D. (n = 3).

Figure 4

In vivo renal clearance pathways of Kim‐TDFs in health mice. A) Schematic illustration of two distinct renal clearance pathways in the kidneys. B) The t _1/2β of Kim‐TDFs in living mice treated with probenecid or cimetidine. Data represent mean ± S.D. (n = 4). Statistical analysis: *p < 0.05, ****p < 0.0001. C,D,E) Fluorescence images of glomerulus and tubules at tissue level at 6 min post‐injection of Cy3‐labelled Kim‐TDFs (red signal). Nuclei were stained with 2‐(4‐amidinophenyl)‐1H‐indole‐6‐carboxamidine (DAPI; blue). Blood vessel stained with anti‐CD31 antibody (green). “G” denotes glomeruli and “T” denotes renal tubular.

Figure 5

Dependency of AKI‐to‐normal contrast on the TDF size. A) NIR images of living mice after injected with Kim‐TDFs at 12 h post‐induction with 50% glycerol. B) NIR fluorescence intensities of kidneys at different post‐injection time points of Kim‐TDFs. C) The percentage of kidney contrast enhancement at different post‐injection time points of Kim‐TDFs in living mice at 12 h post‐induction with 50% glycerol. D) AKI‐to‐normal contrast at different post‐injection time points of Kim‐TDFs in living mice at 12 h post‐induction with 50% glycerol. Statistical analysis in (D) Kim‐TDF₁₇ versus other groups; mean ±  S.D. (n = 3); *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 6

Real‐time in vivo NIR imaging of AKI mice. A) Schematic illustration of glycerol‐pretreated (5 mL kg⁻¹) mice and NIR imaging at different post‐treatment time points. B) NIR images of living mice after injected with Kim‐TDF₁₇ at different post‐induction time points with 50% glycerol. C) NIR fluorescence intensities of kidneys and D) bladder at different post‐injection time points of Kim‐TDF₁₇ in living mice at 6, 12, 24 h post‐induction with 50% glycerol. E) Percentage of kidney contrast enhancement and F) the ratios of BTK intensity at different post‐injection time points of Kim‐TDF₁₇ in living mice at 6, 12, 24 h post‐induction with 50% glycerol. (C–F) Data represent mean ±  S.D. (n = 3).

Figure 7

Diagnosis of glycerol‐induced AKI in living mice. A) The schematic illustration of diagnosis in vivo by NIR imaging and Kim‐TDF‐based urinalysis. B) Kim‐TDF‐based urinalysis of living mice after injected by Kim‐TDF with different valence (1, 2, 3) of targeting at different time points post‐induction with 50% glycerol. (The reporting fluorophore in Kim‐TDF was IR 800CW). C) Kim‐TDF‐based urinalysis of living mice after injected by Kim‐TDF with different valence (1, 2, 3) of targeting at different time points post‐induction with 50% glycerol. (The reporting fluorophore in Kim‐TDF was FAM). D) Receiver operating characteristic analysis for living mice after injected by Kim‐TDF with different valence (1, 2, 3) of targeting at 12 points post‐induction with 50% glycerol. E,F) Change in sCr, urinary Kim‐1 in living mice at different time points post‐induction with 50% glycerol. Data represent mean ±  S.D. (n = 5). G) The BTK ratios of fluorescence intensity in living mice after injected by Kim‐TDF at different time points post‐induction with 50% glycerol. Data represent mean ±  S.D. (n = 3). Statistical analysis in (E–G) 0 h versus other groups; *p < 0.05, **p < 0.01, ****p < 0.0001; ns, no significance.