Optimization and critical evaluation of decellularization strategies to develop renal extracellular matrix scaffolds as biological templates for organ engineering and transplantation

Abstract

The ability to generate patient-specific cells through induced pluripotent stem cell (iPSC) technology has encouraged development of three-dimensional extracellular matrix (ECM) scaffolds as bioactive substrates for cell differentiation with the long-range goal of bioengineering organs for transplantation. Perfusion decellularization uses the vasculature to remove resident cells, leaving an intact ECM template wherein new cells grow; however, a rigorous evaluative framework assessing ECM structural and biochemical quality is lacking. To address this, we developed histologic scoring systems to quantify fundamental characteristics of decellularized rodent kidneys: ECM structure (tubules, vessels, glomeruli) and cell removal. We also assessed growth factor retention--indicating matrix biofunctionality. These scoring systems evaluated three strategies developed to decellularize kidneys (1% Triton X-100, 1% Triton X-100/0.1% sodium dodecyl sulfate (SDS) and 0.02% Trypsin-0.05% EGTA/1% Triton X-100). Triton and Triton/SDS preserved renal microarchitecture and retained matrix-bound basic fibroblast growth factor and vascular endothelial growth factor. Trypsin caused structural deterioration and growth factor loss. Triton/SDS-decellularized scaffolds maintained 3 h of leak-free blood flow in a rodent transplantation model and supported repopulation with human iPSC-derived endothelial cells and tubular epithelial cells ex vivo. Taken together, we identify an optimal Triton/SDS-based decellularization strategy that produces a biomatrix that may ultimately serve as a rodent model for kidney bioengineering.

Keywords: animal models: murine, bioengineering, kidney biology, stem cells; basic (laboratory) research/science; regenerative medicine; tissue/organ engineering.

Figure 1. Rat kidney decellularization strategies and macroscopic evaluation of kidney decellularization

(A) Schedules of reagents (quantity and duration) that were sequentially perfused through the renal artery of each rat kidney in three different decellularization protocols: Triton, Triton/SDS, or Trypsin-EGTA/Triton. (B) Representative images of kidneys are shown at each stage throughout the decellularization process as indicated. In the Triton/SDS protocol kidneys became transparent after perfusion of SDS. Using Trypsin-EGTA/Triton, kidneys became progressively transparent early in the decellularization process. Scale bars: 5 mm.

Figure 2. H&E staining of native and decellularized kidneys

Kidneys decellularized using Triton (C–D), Triton/SDS (E–F), or Trypsin-EGTA/Triton (G–H) were stained with H&E and compared to native kidneys (A–B). Decellularization with Triton resulted in retention of endothelial and smooth muscle cells in vessels and mesangial cells within the glomerulus. Triton/SDS and Trypsin-EGTA/Triton removed all cells as assessed by H&E. Representative images are shown at 10× (A, C, E, G) and 40× (B, D, F, H) magnification. Scale bars: 50 µm.

Figure 3. Scanning electron micrographs of native and decellularized kidneys

Kidneys decellularized using Triton (C–D), Triton/SDS (E–F), or Trypsin-EGTA/Triton (G–H) were imaged using scanning electron microscopy and compared to native kidneys (A–B). Representative images are shown at 250× (A, C, E, G) and 850× (B, D, F, H) magnification. Decellularization with Triton/SDS or Trypsin-EDTA/Triton removed parenchyma cells resulting in a honeycomb appearance of the renal ECM (E, G) and preservation of glomerular structures (F, H) which are a benchmark for renal decellularization. Scale bars: 50 µm.

Figure 4. Nuclear basophilia score of decellularized kidneys

This scoring system assesses the removal of cells (loss of nuclear basophilia) from components of the decellularized kidney ECM as evaluated by a pathologist blinded to each protocol used. A score of 1 is optimal. 1- Complete (100%) removal of cells; 2- Substantial (70%) loss; 3- Moderate (50%) loss; 4- Minimal (30%) loss; 5- No (0%) loss. Results are presented as mean ± standard deviation. # indicates a significant difference in means compared with all other groups (p<0.05), * indicates a significant difference in means between conditions specified by horizontal and vertical bars (p<0.05), and N.S. indicates no significant difference in means as specified.

Figure 5. Architectural score of decellularized kidneys

This scoring system evaluates retention of microscopic ECM components evaluated by a pathologist blinded to each protocol used. A score of 5 is optimal. 5-Outlines visible, architecture intact; 4-Outlines visible/Minimal disruption; 3-Outlines visible/Moderate disruption; 2-Outlines visible/marked disruption; 1-No outlines visible/marked disruption/breakdown of tissue. Results are presented as mean ± standard deviation. * indicates a significant difference in means between the conditions specified by the horizontal and vertical bars (p<0.05).

Figure 6. Collagen IV and laminin staining of normal and decellularized kidneys using three decellularization strategies

Native kidneys (A, E) or kidneys decellularized using Triton (B, F), Triton/SDS (C, G), or Trypsin-EGTA/Triton (D, H) were stained for collagen IV/cell nuclei (top row) or laminin (bottom row). Collagen IV (red, top row) is retained in the glomeruli (pictured) and other structures (e.g., renal capsule and tubules, not shown) in all protocols though there was some qualitative loss of protein. Only Triton/SDS caused complete loss of nuclei as seen by the absence of DAPI staining (blue). Trypsin-EGTA/Triton yielded DNA debris within the matrix, as evidenced by residual DAPI staining. Laminin (bottom row) was retained to a degree in each protocol. In Triton/SDS and Trypsin-EGTA/Triton decellularized kidneys, laminin was mostly found within the glomeruli and, to a lesser extent, tubules. Representative images are shown at 40× magnification for collagen IV (top row, 25 µm scale bars) and 10× magnification for laminin (bottom row, 50 µm scale bars).

Figure 7. Evaluation of growth factor retention in normal and decellularized kidneys

Normal kidneys or kidneys decellularized using Triton, Triton/SDS, or Trypsin-EGTA/Triton were lyophilized, digested enzymatically and analyzed for bFGF (A) or VEGF (B) content using ELISA. Growth factor retention was normalized by initial dry weight of native, control kidneys for comparison across all protocols. Results are presented as minimum, median, and maximum values (n=3 kidneys per group). Asterisks indicate significant differences in means as specified (p<0.05). No significant differences were found in retention of VEGF.

Figure 8. Analysis of vasculature, transplantation, and endothelialization of decellularized rodent kidneys

Corrosion casting was used to assess the integrity of the vasculature in Triton/SDS-decellularized kidneys. An intact vasculature is illustrated by polymers injected through the renal artery (red) and vein (blue) (A) and by bright-field microscopy (B). Gross aspect of transplanted Triton/SDS-decellularized kidney before (C) and after (D) 30 minutes of reperfusion of blood within Sprague-Dawley recipient rat. Human iPSC-derived endothelial cells, labeled with CFSE (green) were injected through the renal artery. Fluorescence microscopy shows incorporation of cells into vascular structures, which were well distributed throughout the kidney (E, F).

Figure 9. Recellularization of whole-kidney decellularized ECM scaffolds with tubular epithelial cells

(A) Human RCTE cells were infused into Triton/SDS-decellularized kidneys through the renal artery and cultured within a perfusion bioreactor. Representative images of kidney sections stained with H&E at days 1, 3, and 7, and elastin and PAS at day 3 are shown. Weigert’s resorcin-fuchsin (WRF) staining of elastin demonstrates retention of arteriolar structure and indicates that although RCTE cells were infused through the renal artery they did not remain in this location (arrows depict elastin staining). PAS stain indicates retention of the basement membrane depicted in dark pink, indicating that RCTE cells resided on the basement membrane and formed what appear to be tubular structures. Scale bars: 50 µm. (B) % Area coverage was quantified using ImageJ and decellularized (Decell) and recellularized (Recell-D1) kidneys were compared to native, untreated kidneys. The renal ECM makes up 12.0±0.5% of the total area of decellularized kidneys while normal kidneys have cells and ECM covering 92.7±2.6% with the remaining ~7% taken up by intraluminal (vascular and tubule) open space. Renal cells and adjacent ECM occupied ~50% of the surface area of recellularized kidneys at day 1 as determined in 5 high-powered fields each at 3 separate levels inside the 3D scaffold (15 fields total per kidney). Resazurin reduction assay revealed a significant increase in RCTE metabolic activity over time (p<0.05). (C) Glucose consumption declined after day 1, and remained stable through day 7. Lactate production steadily increased; however, indicating a shift toward glycolytic metabolism. # indicates a significant difference in means compared with all other groups (p<0.05), * indicates a significant difference in means between conditions specified by horizontal and vertical bars (p<0.05), and N.S. indicates no significant difference in means as specified.