Efficient and selective kidney targeting by chemically modified carbohydrate conjugates

Abstract

We investigated a renal tubule-targeting carbohydrate (RENTAC) that can selectively deliver small-molecule and nucleic acid analogs to the proximal convoluted tubules of the kidney following systemic delivery in mice. We comprehensively evaluated anti-miR-21-peptide nucleic acid-RENTAC, and fluorophore-RENTAC conjugates in cell culture and in vivo. We established that RENTAC conjugates showed megalin- and cubilin-dependent endocytic uptake in the immortalized kidney cell line. In vivo biodistribution studies confirmed the retention of RENTAC conjugates in the kidneys for several days compared with other organs. Immunofluorescence staining confirmed the selective distribution of the RENTAC conjugates in proximal convoluted tubules. We further demonstrated proximal convoluted tubule targeting features of RENTAC conjugates in a folic acid-induced kidney fibrosis mouse model. As a biological readout, we targeted miR-33 using antisense peptide nucleic acid (PNA) 33-RENTAC conjugates in the fibrotic kidney disease model. The targeted delivery of PNA 33-RENTAC resulted in slower fibrosis progression and decreased collagen deposition. We also confirmed that the RENTAC ligand did not exert any adverse reactions. Thus, we established that the RENTAC ligand can be used for broad clinical applications targeting the kidneys selectively.

Keywords: RENTAC; acetylated lactobionic acid; antisense oligonucleotide; cubilin; fibrosis; kidney delivery; megalin; miR-33; peptide nucleic acid; proximal convoluted tubules.

Graphical abstract

Figure 1

Chemical synthesis scheme of RENTAC ligand and solution-phase conjugation with 5-carboxytetramethylrhodamine (A) RENTAC ligand synthesis (1–7). Step one (1–4) comprises the synthesis of the lysine-1,6 diamino hexane backbone, and in step two (5–7), the backbone was functionalized using acetylated lactobionic acid (LBA). (B) Schematic of solution-phase conjugation of 5-carboxytetramethylrhodamine (TAMRA) with RENTAC ligand (7–9). (C) HPLC chromatogram of TAMRA-RENTAC. (D) Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) analysis of TAMRA-RENTAC.

Figure 2

PNA conjugation with RENTAC ligand (A) Chemical illustration of RNA and PNA backbones. (B) Sequences of miR-21-5p and PNA 21 oligomers. PNAs contained lysine (K) at the 3′ end. The 5′ end of the PNAs was functionalized with succinic acid, whereas the 3′ end of PNAs was labeled with TAMRA. -OOO- represents the trioxo-miniPEG flexible spacers. (C) Schematic of conjugation of PNA and RENTAC ligand. (D) Quality control analysis of PNA 21-2-RENTAC by HPLC purity assessment and MALDI-TOF-MS analysis (inset image).

Figure 3

PNA 21-2-RENTAC showed megalin and cubilin receptor-mediated uptake in human renal proximal tubular cells For uptake studies, cells were probed for TAMRA fluorescence following 2 h incubation or as specified. (A) Confocal microscopic images of cell uptake of PNA 21-2 and PNA 21-2-RENTAC at 2 μM concentration. Scale bar, 20 μm. (B) Flow cytometric evaluation of cell uptake of PNA 21-2 and PNA 21-2-RENTAC at 1 μM concentration. (C) Concentration-dependent cell uptake of PNA 21-2-RENTAC by flow cytometry. Each circle in contour plots represents the 5% events, and the adjunct graphs are histogram plots of cellular uptake. (D) Flow cytometry histogram plots depicting cellular uptake of PNA 21-2-RENTAC (2 μM) in HK-2 cells and HEK293 cells. (E) Cumulative distribution function (CDF) plots depict the effect of endocytosis inhibition on cellular uptake of PNA 21-2-RENTAC (1 μM). (F and G) Flow cytometry histogram plots depicting the effect of siRNA-mediated knockdown of megalin and cubilin receptor proteins on uptake of PNA 21-2-RENTAC (1 μM) in HK-2 cells (F) and quantification of PNA 21-2-RENTAC uptake in HK-2 cells following megalin and cubilin knockdown (G). Data are presented as the mean + SEM. (H–J) miR-21 levels after 24 h of PNA 21-3-RENTAC (1 μM) treatment in normal HK-2 cells (H), and with megalin (I), and cubilin (J) knockdown. Data are presented as the mean + SEM. All experiments were performed in triplicates. Statistical significance analysis was performed on GraphPad Prism software using one-way ANOVA analysis followed by Dunnett’s test for multiple comparisons. An unpaired t test was performed for comparison between two groups. A p value <0.05 was considered significant and presented on the graph.

Figure 4

PNA-RENTAC targeted the PCT in the mice kidney following subcutaneous administration Biodistribution of PNA 21-1-RENTAC and PNA 21-2-RENTAC in C57BL/6J mice at 2 mg/kg subcutaneous dose. (A) Ex vivo IVIS images of the harvested liver (Li), lungs (Lu), heart (H), kidney (K), and spleen (S) depicting time course biodistribution of PNAs (n = 3 mice for each time point). (B) Quantitative analysis of PNA 21-1/-RENTAC organ distribution at 2 h. Plotted are the TAMRA average radiant efficiency (ps⁻¹ cm⁻² sr⁻¹)/(μW cm⁻²). n = 3 mice and data are presented as the mean + SEM. (C) Confocal stitched images of PNA 21-1/-RENTAC distribution in kidneys at 2 h depicted at magnifications of 2× (left) and 10× (right) illustrating PNA distribution in red and nuclear stain Hoechst dye in blue. Scale bars, 500 μm (for 2× zoom) and 200 μm (for 10× zoom). (D and E) Flow cytometry histogram plots depicting the PNA 21-2-RENTAC distribution for in-vivo-treated mice at 48, 72, and 96 h time points (D) and quantification of kidney cells showing PNA 21-2-RENTAC retention (E). Kidney cells were analyzed for TAMRA fluorescence by flow cytometry. n = 3 mice for each time point and data are presented as the mean + SEM. (F) Confocal immunofluorescence (IF) imaging confirming localization of PNA 21-2-RENTAC in PCTs. Mice were dosed at 5 mg/kg. PNA 21-2-RENTAC distribution in the kidney at 24 h. Red depicts the PNA 21-2-RENTAC distribution, green is the nephron region-specific marker, and blue is nuclear stain DAPI. Scale bar, 25 μm. Statistical significance analysis was performed on GraphPad Prism software using one-way (E) and two-way (B) ANOVA analysis followed by Dunnett’s test for multiple comparisons. A p value <0.05 was considered significant and presented on the graph.

Figure 5

TAMRA-RENTAC showed targeted delivery in PCTs in the mice kidney Biodistribution of TAMRA-RENTAC in C57BL/6J mice at 10 mg/kg subcutaneous dose. (A) Ex vivo IVIS images of the harvested liver (Li), lungs (Lu), heart (H), kidney (K), and spleen (S) showing TAMRA-RENTAC biodistribution at 2 and 4 h time points. n = 3 mice for each time point. (B) Quantitative analysis of TAMRA-RENTAC organ distribution at 2 and 4 h. Data are presented as the mean + SEM. (C) After in vivo treatment, kidney cells were analyzed for TAMRA fluorescence by flow cytometry. Graph shows the percentage of the TAMRA-positive cell population. n = 3 mice for each treatment. (D) Confocal IF imaging of TAMRA-RENTAC-treated kidneys. Depicted are 60× zoom-in images at 4 h. Red depicts the TAMRA-RENTAC distribution, green is the nephron region-specific marker, and blue is nuclear stain DAPI. Scale bar, 25 μm. Statistical significance analysis was performed on GraphPad Prism software using two-way ANOVA analysis. A p value <0.05 was considered significant and presented on the graph.

Figure 6

TAMRA-RENTAC targeted PCTs in folic acid-induced fibrosis mouse model Mice were injected with folic acid (250 mg/kg) intraperitoneally for kidney fibrosis induction, and biodistribution of TAMRA-RENTAC (10 mg/kg, subcutaneous) was studied after 2 weeks. (A and B) Biodistribution of TAMRA and TAMRA-RENTAC in moderate (A) and severe fibrotic mice (B). Ex vivo IVIS images of the harvested liver (Li), lungs (Lu), heart (H), kidney (K), and spleen (S) showing TAMRA and TAMRA-RENTAC biodistribution at 4 and 24 h time points. n = 3 mice for each time point. (C) Quantitative analysis of TAMRA-RENTAC organ distribution in severe fibrotic mice at 4 and 24 h. (D) Comparison of kidney accumulation of TAMRA-RENTAC in normal, moderate fibrotic, and severe fibrotic mice at 4 h. n = 3 mice for each time point, and data are presented as the mean + SEM. (E) Kidney cell uptake of TAMRA-RENTAC in moderate fibrotic mice. Each circle in contour plots represents the 5% events, and the adjunct graphs are histogram plots of kidney cell uptake. (F) Confocal IF imaging indicates the localization of TAMRA-RENTAC in different nephron regions of severe fibrotic mice. Depicted are 60× zoom-in images at 24 h in which red depicts the TAMRA-RENTAC distribution, green is the nephron region-specific marker, and blue is nuclear stain DAPI. Scale bar, 25 μm. Statistical significance analysis was performed on GraphPad Prism software using one-way ANOVA analysis. A p value <0.05 was considered significant and presented on the graph.

Figure 7

Kidney-targeted delivery of anti-miR33 prevented kidney fibrosis in folic acid-treated mice ScrPNA 33-RENTAC, PNA 33, and PNA 33-RENTAC were assessed for their efficacy in folic acid-induced kidney fibrosis in C57BL/6J mice. A single-dose safety study of PNA 33-RENTAC was performed in healthy C57BL/6J mice at 2 mg/kg (subcutaneous). (A) Hematoxylin and eosin (H&E) staining on kidney, liver, and spleen sections for safety evaluation. Scale bar, 100 μm (for 10× zoom). (B) The plasma levels of indicated cytokines. Data are presented as mean + SEM (n = 4 for saline and n = 6 for PNA 33-RENTAC). (C) Nucleotide sequences of miR-33-5p, anti-miR PNA, and scrambled PNA used in this study. The 5′ end of the PNAs was functionalized with SA for ligand conjugation. -OOO- represents a trioxo-miniPEG flexible spacer. (D) Experimental workflow of PNA treatment and fibrosis induction timeline. A single dose of folic acid (250 mg/kg) was injected intraperitoneally on day 1. PNA 33, PNA 33-RENTAC, and ScrPNA 33-RENTAC were dosed at 2 mg/kg (subcutaneously) every other day (days −2, 1, 3, and 5). n ≥ 6 mice for each group. (E) H&E, picrosirius red, and periodic acid-Schiff (PAS) staining on kidney sections of various treatment groups. In H&E staining, tubular injury is highlighted by asterisks representing dilated tubules and arrows indicating infiltrated mononuclear cells in the interstitium. Black arrows in the picrosirius staining depict collagen deposition in red. H&E and picrosirius red staining images were captured at 40× magnification. Scale bar, 50 μm. While PAS staining images were taken at 60× magnification. Scale bar, 25 μm. (F) Quantification of picrosirius red positive area in kidney sections. Five cortical region frames (10× zoom) per mouse from 3 to 5 mice in each treatment group were quantified using ImageJ software, and percentage positive areas (red) were compared. Data are presented as a box and whisker plot showing all data points. (G) Kidney mRNA levels of fibrotic marker collagen type I (Col1a1), fibronectin-1 (Fn-1), and transforming growth factor beta (Tgfb). The data presented are mean + SEM (n = 6–7). (H) Different treatment groups' plasma blood urea nitrogen (BUN) levels at day 7. The data presented are mean + SEM (n = 6–7). (I and J) Kidney levels of miR-33 and its downstream target genes Cpt1a, Hadhb, and Crot by RT-qPCR analysis. Data are presented as mean + SEM (n = 6–7). (K and L) CROT (K) and HADHB (L) protein levels and quantification in the kidney by western blotting. The bands intensities were quantified using ImageJ 1.54g software and cyclophilin B as a sample loading control. Blots represent n = 3; for quantification, data are presented as mean + SEM (n = 6). Statistical significance analysis was performed on GraphPad Prism software using one-way ANOVA followed by Dunn’s test for multiplicity comparison. An unpaired t test was performed for comparison between the two groups. A p value <0.05 was considered significant and presented on the graph.