Nanoparticle Transport in Proximal Tubules with Rhabdomyolysis-Induced Necrosis

Abstract

Renal-clearable engineered nanoparticles are being explored for their potential to deliver therapeutic agents for kidney disease treatment. A fundamental understanding of how these nanoparticles accumulate in diseased kidneys at the cellular level is essential to enhance their effectiveness and minimize side effects on adjacent healthy tissues. Herein, we report that the accumulation of glutathione-coated, near-infrared emitting gold nanoparticles (GS-AuNPs) correlates strongly with the necrotic stages of injured proximal tubular cells. Using a rhabdomyolysis-induced acute kidney injury (AKI) mouse model, we observed that GS-AuNPs were significantly accumulated in the extracellular lumen of proximal tubular epithelial cells (PTECs) at advanced necrotic stage, where cellular debris and released intracellular contents impeded their clearance. In contrast, during early necrosis, GS-AuNPs were still cleared through the unobstructed lumen. Additionally, intracellular uptake of GS-AuNPs was significantly reduced across all necrotic stages. These findings underscore the need for new strategies to design nanoparticles that can effectively target and be taken up by the diseased tubular cells before extensive necrosis occurs; so that nanoparticle-mediated drug delivery for kidney disease treatment can be achieved with desired efficacy and precision.

Introduction

Due to their unique clearance pathway through the kidneys, renal-clearable nanoparticles have great potential for targeted delivery of therapeutic agents to treat kidney diseases.^[1] However, the intra-kidney physiological barriers change dramatically in diseased kidneys,^[2] depending on the disease stage, thereby affecting the kidney transport kinetics and urine clearance efficiency of renal-clearable nano-particles. In general, prolonged retention and reduced clearance of renal-clearable nanoparticles have been widely observed in various kidney disease models.^[3] For instance, using renal-clearable glutathione-coated gold nanoparticles (GS-AuNPs) as multimodality imaging nanoprobes,^[3b,4] we have previously demonstrated that the intra-kidney transport kinetics of GS-AuNPs varied with the disease stage in a unilateral ureteral obstruction (UUO) mouse model. The kidney retention of GS-AuNPs was enhanced in mild UUO kidneys due to obstructed urine clearance, whereas it decreased in severe UUO kidneys due to reduced blood perfusion.^[3b] Additionally, in an acute kidney injury (AKI) mouse model, the retention of pegylated dye macromolecules in the kidneys increased while their urine clearance decreased with increasing disease severity.^[5] In a cisplatin-induced AKI model, increased accumulation of GS-AuNPs in the inner strip of outer medulla was also observed.^[6] Enhanced accumulation of nanoparticles has also been harnessed for kidney disease treatment. For instance, a series of antioxidant nanoparticles^{[3d,e, 7]} have been used for treatment of AKI via scavenging elevated reactive oxygen species (ROS). While prolonged kidney retention of renal-clearable nanoparticles has been widely observed in different kidney diseases, the correlation between nanoparticle retention and kidney injury stage at the cellular level has not been extensively investigated, which is critical to the targeted delivery of therapeutic agents to injured kidney tissue for optimized treatment efficacy of kidney diseases.

Rhabdomyolysis, a serious syndrome caused by skeletal muscle injury due to trauma, excessive exercise and alcohol abuse, leads to the release of muscle cell contents such as myoglobin into blood circulation.^[8] The released myoglobin is filtered by the glomerulus and causes AKI through increasing oxidative stress and vasoconstriction.^[9] AKI is the most serious complication of rhabdomyolysis and represents ~7 to 10% of all causes of AKI in the United States.^[10] The proximal tubule (PT) is the primary target of injury in AKI.^[11] Proximal tubular epithelial cells (PTECs) underwent different injury stages, characterized by different histopathological features^[12] such as cytoplasmic alteration, nuclear morphology change, loss of brush border and formation of cell debris and protein casts. These histopathological characteristics substantially affect cell functions and change local flow dynamics in tubular lumen. Previously, we and others have found that renal-clearable nanoparticles can be taken up by healthy PTECs from the luminal side once they reach the lumen of PTs after glomerular filtration^[13] and Coulombic interaction between nanoparticles and the charged surface of PTECs played a crucial role in regulating the uptake efficiency.^[14] This proximal tubular uptake offered opportunities to unravel how renal-clearable nanoparticles interact differently with healthy and injured PTs and to correlate the cellular injury stage of PTECs with nanoparticle interaction.

In this work, we correlated the interaction between GS-AuNPs and PTs at different injury stages at the cellular level. we found that the kidney accumulation of GS-AuNPs was enhanced by ~8-fold in a rhabdomyolysis-induced AKI mouse model. By distinguishing the injury stages of necrotic PTECs through cytoplasmic alteration and nuclear morphology change, we identified early-stage necrotic PTs, characterized by cytoplasmic vacuolization and relatively normal nuclei, and late-stage necrotic PTECs, characterized by flattened eosinophilic epithelium with pyknosis and karyorrhexis (Figure 1). We found that the enhanced accumulation of GS-AuNPs in AKI was due to their selective retention in the extracellular lumen of late-stage necrotic PTECs as a result of lumen obstruction by cell debris, which, however, was not observed in early-stage necrotic PTECs with unobstructed lumen (Figure 1). In addition, the cellular uptake of GS-AuNPs decreased in both early and late necrotic PTECs compared to that in normal PTECs regardless of the injury stage (Figure 1). Our findings highlight the significance of the correlation between cell injury stage and nanoparticle interaction and offer new insights to selectively target injured kidney tissues in kidney disease.

Figure 1.

Selective extracellular retention of renal-clearable gold nanoparticles (AuNPs) in obstructed lumen of late-stage necrotic proximal tubules (PTs). Compared to normal PTs, early-stage necrotic PTs were characterized by cytoplasmic vacuolization. However, they showed relatively normal nuclei, recognizable microvilli (MV) layer and unobstructed lumen. In contrast, late-stage necrotic PTs were characterized by flattened eosinophilic epithelium with pyknotic (pyknosis), fragmented (karyorrhexis) or disappeared (karyolysis) nuclei. The microvilli layer was detached, and more cell debris was accumulated in the lumen due to progressed necrosis, causing tubular obstruction. As a result, renal-clearable AuNPs were selectively retained in the obstructed lumen of late-stage necrotic PTs after glomerular filtration. Moreover, the proximal tubular uptake of AuNPs decreased in both early- and late-stage necrotic PTs compared to that in normal PTs regardless of the injury stage.

Results and Discussion

Establishment of Rhabdomyolysis-Induced AKI Model

To unravel the interaction of renal-clearable AuNPs with diseased kidney, we first established a clinically relevant rhabdomyolysis-induced AKI mouse model (Figure 2a), in which the kidneys are injured by myoglobin released from damaged muscle cells.^[10] To establish the model,^[3e] the mice were first deprived of water but had free access to food for 15 h. After water restriction, 8 mL/kg 50% glycerol was intramuscularly injected into each hind limb of mice to damage the muscle cells. The disease model was allowed to progress for 2 h after the injection of glycerol. Blood urea nitrogen (BUN) and serum creatinine (sCr) levels were both significantly elevated at 2 h after the injection of glycerol (Figure 2b and 2c), indicating the successful establishment of the AKI model. PTs are the major injury sites in AKI.^[11,15] PTECs undergo necrosis in AKI.^[16] There are three death steps of the cell nucleus^[17] as a result of cell necrosis, including pyknosis,^[18] the irreversible condensation of chromatin and shrinkage of cell nucleus, followed by karyorrhexis,^[19] which is the fragmentation of the nucleus, and eventually to karyorlyis,^[20] which is the “melting” of the nucleus due to complete dissolution of chromatin (Figure 2d). In normal kidney tissue (Figure 2e), the luminal side of normal PTECs is distinctively covered by a densely packed microvilli layer (Figure 2f), which dramatically increases the surface area of PTECs and enhances the absorption function towards substances filtered by the glomerulus.^[21] However, in rhabdomyolysis-induced AKI, unlike the focal injury in cisplatin-induced AKI,^[13c] diffusive proximal tubular injury was observed (Figure 2g). In addition, two types of proximal tubular injuries were observed in rhabdomyolysis-induced AKI: early-stage necrotic PTECs with cytoplasmic vacuolization but relatively normal nuclei and recognizable microvilli layer (Figure 2g and 2h, early-stage necrotic PTECs were labelled by brown diamond shapes) and late-stage necrotic PTECs showing flattened eosinophilic epithelium with pyknotic and fragmented nuclei (Figure 2g and 2i, late-stage necrotic proximal tubules were labelled by blue triangle shapes). Due to more severe injury, the microvilli layer in late-stage necrotic proximal tubules was detached and not recognizable anymore (Figure 2i). Instead, more cell debris was observed in the lumen of late-stage necrotic proximal tubules (Figure 2i) due to progressed necrosis, causing tubular obstruction. In addition, protein casts (Figure 2g to 2i, protein casts were labelled by green star shapes), a typical pathological feature of rhabdomyolysis-induced AKI, were also observed. The glomerulus in AKI kidney showed dilated Bowman’s space compared to the open Bowman’s space in normal kidney (Figure S1). In comparison to proximal tubules, only early-stage necrosis showing intracellular vacuolization was observed in the epithelial cells of the downstream microvilli-free renal tubules including loop of Henle, distal tubules and collecting ducts (Figure S1). However, lumen obstruction was not observed in these early-stage necrotic renal tubules (Figure S1).

Figure 2.

Identification of early- and late-stage necrotic PTs based on histopathological characteristics in a rhabdomyolysis-induced AKI mouse model. a) Establishment of the rhabdomyolysis-induced AKI mouse model through intramuscular injection of glycerol. b) Serum blood urea nitrogen level of normal and AKI mice. c) Serum creatinine level of normal and AKI mice. N=4 mice in each group in b and c. Data are presented as mean ± standard deviation and two-sided Student’s t-test was performed at the 0.05 significance level in b and c. d) Nuclear morphology change in cell necrosis. Pyknosis: shrinkage and condensation of cell nuclei. Karyorrhexis: fragmentation of the condensed pyknotic nuclei. Karyorlysis: melting of nuclei due to complete dissolution of chromatin. e) Representative image of hematoxylin and eosin (H&E)-stained normal kidney tissue. f) Magnified image of the square area in e showing the histopathological structure of normal PTs. g) Representative image of hematoxylin and eosin (H&E)-stained AKI kidney tissue showing diffusive kidney injury. Early-stage necrotic PTs were labelled by brown diamond shapes. Late-stage necrotic PTs were labelled by blue triangle shapes. Protein casts were labelled by green star shapes. h) Magnified image of an early-stage necrotic PT in g showing cytoplasmic vacuolization, relatively normal nuclei, recognizable microvilli layer and unobstructed lumen. i) Magnified image of a late-stage necrotic PT in g showing flattened eosinophilic epithelium with pyknotic, fragmented or disappeared nuclei. The lumen was obstructed with cell debris due to progressed necrosis.

Enhanced Accumulation of GS-AuNPs in AKI

Renal-clearable luminescent GS-AuNPs were synthesized according to our previously reported method.^[22] GS-AuNPs are known for their resistance to serum protein binding and have been widely used as multimodality probes to investigate the nano-bio interactions in the kidneys.^[23] The core size of GS-AuNPs is 2.4±0.3 nm and the hydrodynamic diameter is 3.8±1.0 nm (Figure S2), which is below the kidney filtration threshold (~5.5 nm). GS-AuNPs exhibited intrinsic luminescence with near-infrared emission at 810 nm (Figure S3), enabling in vivo fluorescence imaging after systemic injection. In addition, GS-AuNPs showed no binding to serum proteins (Figure S4), which ensured the ultra-small size of GS-AuNPs in blood circulation and the efficient glomerular filtration and urine clearance.

To investigate the transport of GS-AuNPs in rhabdo-myolysis-induced AKI, we intravenously injected GS-AuNPs into mice at 2 h post intramuscular injection of glycerol. At 7 h post intravenous injection of GS-AuNPs, we conducted in vivo fluorescence imaging of mice and sacrificed mice to harvest organs for ex vivo imaging (Figure 3a). We found that GS-AuNPs exhibited significantly higher kidney accumulation in rhabdomyolysis-induced AKI than in normal mice (Figure 3b and Figure S5). The kidney fluorescence intensity in rhabdomyolysis-induced AKI was ~8.4 times higher than that in the control group (Figure 3c). Ex vivo organ fluorescence imaging further showed that GS-AuNPs were selectively accumulated in injured kidneys in rhabdomyolysis-induced AKI mice (Figure 3d and Figure S6). The ex vivo fluorescence intensity of the kidneys was significantly higher than other organs in rhabdomyolysis-induced AKI mice (Figure 3e). And the ex vivo kidney fluorescence intensity in the AKI group was ~15.2 times higher than that in the normal control group (Figure 3e). We also correlated the ex vivo kidney fluorescence intensity of AKI mice to their sCr and BUN levels. We found that the kidney fluorescence intensity strongly correlated with sCr and BUN levels (Figure S7), suggesting that kidney fluorescence intensity can be used to indicate different stages of kidney injury. Moreover, we kept monitoring the kidney fluorescence intensity within 24 h post injection of GS-AuNPs. The fluorescence intensity of AKI kidneys was significantly higher than that of normal kidneys at 2, 7 and 24 h post injection of GS-AuNPs (Figure S8). And no significant decay of the kidney fluorescence intensity was observed within 24 hours (Figure S8). The ex vivo fluorescence intensity of the kidneys was still significantly higher than other organs in rhabdomyolysis-induced AKI mice and was ~29 times higher than that in the normal control group 24 hours post injection (Figure S9). Together these results clearly demonstrated that the accumulation of renal-clearable GS-AuNPs was significantly enhanced in rhabdomyolysis-induced AKI.

Figure 3.

Enhanced kidney accumulation of GS-AuNPs in rhabdomyolysis-induced AKI mice. a) GS-AuNPs were intravenously injected into mice at 2 h post injection of glycerol. At 7 h post injection of GS-AuNPs, in vivo fluorescence imaging and ex vivo organ fluorescence imaging were conducted. b) Whole-body in vivo fluorescence imaging of mice at 7 h post intravenous injection of GS-AuNPs. The kidney fluorescence intensity of AKI mice was significantly higher than that of healthy mice. BF: bright field. FL: fluorescence field. N=3 mice in healthy group. N=4 mice in AKI group. c) Quantification of the in vivo kidney fluorescence intensity in b. The kidney fluorescence intensity of AKI mice was ~8.4 times higher than that of healthy mice. d) Ex vivo fluorescence imaging of organs harvested from healthy and AKI mice at 7 h post intravenous injection of GS-AuNPs. BF: bright field. FL: fluorescence field. Me: membrane, Mu: muscle, Bo: bone, Br: brain, Ki: kidneys, St: stomach, In: intestine, Sk: skin, He: heart, Lu: lung, Li: liver, Sp: spleen. N=3 mice in healthy group. N=4 mice in AKI group. e) Quantification of the ex vivo organ fluorescence intensity in d. Data are presented as mean ± standard deviation and two-sided Student’s t-test was performed at the 0.05 significance level in c and e.

Selective Retention of GS-AuNPs in Late-Stage Necrotic PTs

To further reveal the original cause for the enhanced kidney accumulation GS-AuNPs in AKI mice, we investigated the intra-kidney distribution of GS-AuNPs in both normal and AKI mice at organ and tissue level. At 7 h post intravenous injection of GS-AuNPs, fluorescence imaging of kidney longitudinal section showed that, while GS-AuNPs exhibited low accumulation in the normal kidneys, in AKI kidneys, GS-AuNPs were efficiently accumulated in renal cortex (Figure 4a), which is mainly composed of glomeruli and proximal tubules. At the tissue level, GS-AuNPs were found to be taken up by normal PTECs after glomerular filtration at 7 h post intravenous injection (Figure 4b and 4c. Figure S10). No obvious accumulation of GS-AuNPs was found in other nephron parts including glomeruli, loop of Henle, distal tubules and collecting ducts (Figure S11). The selective proximal tubular uptake is possibly due to the presence of densely packed microvilli layer in PTs, which dramatically increases the surface area of PTECs and enhances their absorption function against glomerular filtrates, as well as the primary physiological role of PTs in metabolizing glomerular filtrates through secretion, uptake and reabsorption through high expression of receptors and transporters and high density of mitochondria.^[21,24] Surprisingly, in rhabdomyolysis-induced AKI, GS-AuNPs were found to be selectively retained in the obstructed extracellular lumen of late-stage necrotic proximal tubules rather than the early-stage ones (Figure 4d and 4e. Figure S12). The late stage necrotic PTECs exhibited flattened eosinophilic epithelium with pyknotic and fragmented nuclei. In contrast, no retention of GS-AuNPs was observed in early-stage necrotic proximal tubules, characterized by cytoplasmic vacuolization but with relatively normal nuclei, and unobstructed lumen (Figure 4f and Figure S12). And no accumulation of GS-AuNPs was found in other injured nephron parts including glomeruli, loop of Henle, distal tubules and collecting ducts was well (Figure S11). These results indicate that intracellular contents released from cytoplasm are responsible for the blockage of GS-AuNPs. More importantly, harnessing the intrinsic fluorescence of GS-AuNPs, we also conducted kidney tissue fluorescence imaging, which showed the morphology of GS-AuNPs in normal and AKI kidney tissue. We found that while the fluorescence of the endocytosed GS-AuNPs by normal PTECs in the control group was completely diminished (Figure S13) possibly due to intracellular metabolism, GS-AuNPs retained extracellularly in the AKI group remained highly fluorescent and aggregated into large nanoparticles in late-stage necrotic PTs (Figure S13), which presumably should contribute to the entrapment of GS-AuNPs in obstructed PTs. However, the uptake of GS-AuNPs in both early- (Figure 4f) and late-stage (Figure 4e) necrotic PTECs was significantly reduced compared to that in normal PTECs (Figure 4c), which is consistent with our previous finding that the cellular uptake of renal-clearable pegylated AuNPs was significantly reduced in injured PTs in cisplatin-induced AKI mice.^[13c] Intracellular mitochondria are primarily damaged in AKI,^[25] which disrupts cellular energy metabolism and therefore reduces the endocytosis capability of cells. In addition, kidney pathology exhibited no significant difference between nanoparticle-injected group and control group (Figure S14), indicating that GS-AuNPs didn’t cause additional kidney injury in the AKI model. All these results together showed that renal-clearable AuNPs were selectively retained in the obstructed lumen of late-stage proximal tubules due to more severe cellular damage. However, the cellular uptake of renal-clearable AuNPs by injured proximal tubules was significantly reduced compared to normal ones regardless of the injury stage.

Figure 4.

Selective retention of GS-AuNPs in obstructed lumen of late-stage necrotic PTs. a) Fluorescence imaging of kidney longitudinal section showing accumulation of GS-AuNPs in renal cortex in AKI kidneys. b) Representative image of silver-enhanced, hematoxylin and eosin (H&E)-stained normal kidney tissue. c) Magnified image of the square area in b showing the uptake of GS-AuNPs by normal PTs. Silver-enhanced GS-AuNPs were labelled by white arrows. d) Representative image of silver-enhanced, hematoxylin and eosin (H&E)-stained AKI kidney tissue showing selective accumulation of GS-AuNPs in late-stage necrotic PTs. Early-stage necrotic PTs were labelled by brown diamond shapes. Late-stage necrotic PTs were labelled by blue triangle shapes. Protein casts were labelled by green star shapes. e) Magnified image of a late-stage necrotic PT in d showing retention of GS-AuNPs in obstructed lumen. The late-stage necrotic PT was characterized by flattened eosinophilic epithelium with pyknotic, fragmented or disappeared nuclei. The lumen was obstructed with cell debris due to progressed necrosis, thus trapping the filtered GS-AuNPs. f) Magnified image of an early-stage necrotic PT in d showing no accumulation of GS-AuNPs in the lumen and in PTECs. The late-stage necrotic PT was characterized by cytoplasmic vacuolization, relatively normal nuclei, recognizable microvilli layer and unobstructed lumen, enabling intratubular flow of filtered GS-AuNPs.

Conclusion

In summary, we unraveled the critical role of injury stage in mediating the proximal tubular transport of renal-clearable AuNPs at the cellular level and revealed that GS-AuNPs were selective retained in the extracellular lumen of late-stage necrotic proximal tubules because of lumen obstruction by cell debris. Unlike cancer cells which more efficiently internalize nanoparticles than normal cells due to elevated metabolism rate,^[26] injured kidney cells showed reduced nanoparticle uptake efficiency than healthy kidney cells because of cell injury, making it more challenging to target injured kidney cells with nanoparticles in kidney disease. In addition, in this work, the observed selective accumulation of GS-AuNPs in the necrotic proximal tubular cells at the late stage due to the tubular blockage created another barrier for treating the kidney diseases before tubular cell death. Thus, to utilize nanoparticles for targeted delivery of drugs to the diseased kidney cells, new strategies for designing renal-clearable nanoparticles are highly demanded.

Supplementary Material

The data that support the findings of this study are available from the submitted Supporting Information file.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.