http://www.jhltonline.org
ORIGINAL PRE-CLINICAL SCIENCE
Perfusion decellularization of human and porcine lungs: Bringing the matrix to clinical scale Sarah Elizabeth Gilpin, PhD,a,b Jacques P. Guyette, PhD,a,b Gabriel Gonzalez, PhD,a,b Xi Ren, PhD,a,b John M. Asara, PhD,a,b Douglas J. Mathisen, MD,a,b Joseph P. Vacanti, MD,a,c and Harald C. Ott, MDa,b,d,e From the aCenter for Regenerative Medicine, Massachusetts General Hospital; bHarvard Medical School; cDivision of Pediatric Surgery; and dDivision of Thoracic Surgery, Department of Surgery, Massachusetts General Hospital; and the eHarvard Stem Cell Institute, Boston, Massachusetts.
KEYWORDS: organ engineering; organ scaffolds; acellular matrix; recellularization; decellularization
BACKGROUND: Organ engineering is a theoretical alternative to allotransplantation for end-stage organ failure. Whole-organ scaffolds can be created by detergent perfusion via the native vasculature, generating an acellular matrix suitable for recellularization with selected cell types. We aimed to up-scale this process, generating biocompatible scaffolds of a clinically relevant scale. METHODS: Rat, porcine, and human lungs were decellularized by detergent perfusion at constant pressures. Collagen, elastin, and glycosaminoglycan content of scaffolds were quantified by colorimetric assays. Proteomic analysis was performed by microcapillary liquid chromatography tandem mass spectrometry. Extracellular matrix (ECM) slices were cultured with human umbilical vein endothelial cells (HUVEC), small airway epithelial cells (SAEC), or pulmonary alveolar epithelial cells (PAECs) and evaluated by time-lapse live cell microscopy and MTT (3-[4,5-dimethylthiazol-2-yl]-2, 5-diphenyltetrazolium bromide) assay. Whole-organ culture was maintained under constant-pressure media perfusion after seeding with PAECs. RESULTS: Rat lungs were decellularized using: (1) sodium dodecyl sulfate (SDS), (2) sodium deoxycholate (SDC), or (3) 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS). Resulting scaffolds showed comparable loss of DNA but greatest preservation of ECM components in SDS-decellularized lungs. Porcine (n ¼ 10) and human (n ¼ 7) lungs required increased SDS concentration, perfusion pressures, and time to achieve decellularization as determined by loss of DNA, with preservation of intact matrix composition and lung architecture. Proteomic analysis of human decellularized lungs further confirmed ECM preservation. Recellularization experiments confirmed scaffold biocompatibility when cultured with mature cell phenotypes and scaffold integrity for the duration of biomimetic culture. CONCLUSIONS: SDS-based perfusion decellularization can be applied to whole porcine and human lungs to generate biocompatible organ scaffolds with preserved ECM composition and architecture. J Heart Lung Transplant 2014;33:298–308 r 2014 International Society for Heart and Lung Transplantation. All rights reserved.
In 2012, more than 28,000 organ transplants were performed in the United States, while more than 128,000 Americans Reprint requests: Harald C. Ott, MD, Massachusetts General Hospital, Department of Surgery, Harvard Medical School, 185 Cambridge St, CPZN 4812, Boston, MA 02114. E-mail address: [email protected]
remain on waiting lists with waiting times extending to several years in some instances.1 As the need for alternate sources of donor organs continuously increases, whole-organ engineering based on human or large-animal acellular scaffolds presents a unique opportunity and holds the potential for long-term graft survival. Creating regenerated donor organs by repopulating organ scaffolds with donor-specific mature or
1053-2498/$ - see front matter r 2014 International Society for Heart and Lung Transplantation. All rights reserved. http://dx.doi.org/10.1016/j.healun.2013.10.030
Gilpin et al.
Perfusion Decellularization of Lungs
stem cell-derived populations would allow us to personalize transplantation medicine and reduce the need for long-term immunosuppressive therapy.2 Bioengineering of such wholeorgan grafts requires scaffolds that outline the lung’s structure and composition, thus providing a hospitable and structurally intact niche for cells of interest to attach and engraft.3 Lung decellularization has been successfully achieved using a variety of approaches and protocols,4–7 and the concept has been applied to various other organs across multiple species.8–12 A non-perfusion decellularization approach using Triton-X and sodium deoxycholate (SDC) produced structurally intact lung matrices that were biocompatible with mesenchymal stem cells13 and pre-differentiated murine embryonic stem cells.14 The vascular perfusion of CHAPS (3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate) detergent also successfully decellularized rat lungs, which subsequently were recellularized to recapitulate mechanical and physiologic organ function.5 Our group has reported a perfusion decellularization protocol using sodium dodecyl sulfate (SDS). The acellular rat lung scaffolds were recellularized with carcinomatous human alveolar basal epithelial cells (cell line A549) or rat fetal lung cells, combined with endothelial cells, resulting in partial restoration of function.4,15 Regardless of protocol, once successful decellularization has been accomplished, the challenge of site-specific recellularization remains; the physiologic functions of the lung depend on its complex organ architecture and multiple cell types required. The decellularized scaffold needs to support cell attachment, proliferation, and engraftment to restore the physiologic context of cells and tissues that enable organ integrity and function. To serve as a robust scaffold for regeneration, the matrix must maintain the essential structural proteins for cell integration and resist proteolytic degradation, at least initially. The present study outlines the development of a reproducible procedure to generate clinical-scale porcine and human lung scaffolds as the first step toward generating a transplantable, bioartificial lung. These constructs are devoid of residual donor-derived cells and maintain the hierarchical structural components of lung vasculature and airways necessary for subsequent recellularization and organ regeneration approaches. To determine the most suitable protocol for larger porcine and human organs, we first optimized the method used by comparing matrices generated with 3 published detergents. We did not include protocols involving repeated freezing, freeze-drying, or enzymatic digestion to avoid the inherent extracellular matrix (ECM) damage. We then applied a single, optimized decellularization protocol to organs of clinical scale and analyzed the resulting scaffolds. Finally, we validated the tissue integrity and biocompatibility of these lung scaffolds.
Methods
299 rats (150–200 g, Charles River Laboratories, Inc, Wilmington, MA) were used throughout. Porcine lungs were harvested in-house from Yorkshire swine (40–200 kg) being otherwise used for experiments not involving the chest cavity. Rat donors were pretreated with a dose of heparin (2.0 U/g of body weight) into the inferior vena cava. Porcine donors were not pre-treated with heparin. Human donors were heparinized at the time or organ harvest following standard organ procurement protocols of the New England Organ Bank (NEOB). The rat lung protocols tested used (1) 0.1% SDS,4 (2) 2% SDC,6 or (3) 8 mmol/liter CHAPS (pH 12)5 detergent perfused through the pulmonary artery at a constant pressure of 30 cm H2O. The protocol for all 3 detergents included:
1. a 10-minute initial antegrade wash with phosphate-buffered saline (PBS),
2. detergent perfusion for the time required to visualize an opaque translucent matrix (indicative of decellularization) plus an extra 20% of that initial time (e.g., 70 minutes þ 14 minutes), 3. 15-minute deionized H2O wash, and 4. an additional 72-hour PBS wash with added antibiotics and antimycotics. The SDS protocol included an additional wash of 1% Triton-X following the deionized H2O. The SDC protocol included a 0.1% Triton-X perfusion before SDC and a 1 mol/liter NaCl wash after SDC (see Supplemental Figure 1, available on the jhltonline.org Web site). Human lungs otherwise unsuitable for transplantation were recovered under sterile conditions within 60 minutes of cessation of cardiovascular circulation and obtained from the International Institute for the Advancement of Medicine (IIAM) or the NEOB. Donor criteria included age o 75 years, negative serologies, tobacco smoking of o 10 pack-years, and no known lung disease, pneumonia, aspiration, or trauma. As described for rat lungs, porcine and human lungs were perfused though the pulmonary artery at constant pressure, followed by sequential washing with H2O, 1% Triton-X solution, and PBS. Similar to rat lungs, decellularization was deemed complete upon visual inspection. Variability in the starting organ, mainly due to extensiveness of pre-flushing during harvest and any resulting clots contributed to the required length of perfusion. The time of SDS perfusion varied from 4 to 7 days. Further details of the decellularization protocol are discussed within the Results.
Histologic analysis After decellularization, several areas of the lung (n ¼ 8–12 tissue samples per lung) sampled randomly from proximal and distal areas of each lobe were isolated, fixed with 5% formalin, and embedded in paraffin. Sections cut to 5 μm were deparaffinized and stained with standard hematoxylin and eosin and Masson’s Trichrome staining. Proteoglycans were visualized by Alcian blue staining, and elastin components by VerhoeffVan Gieson staining. Immunohistochemistry for matrix proteins (collagen IV, fibronectin, and laminin) was done following standard protocols.
Whole-organ perfusion decellularization
Quantitative biochemical matrix assays
Animal organs were harvested in accordance with approved animal protocols and institutional guidelines. Outbred Sprague-Dawley
After decellularization, random samples of each lung lobe (n ¼ 8–12 per lung) from proximal and distal areas of multiple lobes
300
The Journal of Heart and Lung Transplantation, Vol 33, No 3, March 2014
were isolated and stored at –801C until used for the following assays.
Liquid chromatography–tandem mass spectrometry proteomic matrix analysis
DNA
Decellularized lung tissue samples isolated from the superior segments of the right (n ¼ 2) and left (n ¼ 2) upper lobes were first lyophilized and then digested in 5 mol/liter urea buffer,17 followed by acetone precipitation. Microcapillary liquid chromatography tandem mass spectrometry (LC-MS/MS), using linear iontrap MS technology, was used to identify and semi-quantify the sample components in isolated protein samples. The protein samples were first digested with trypsin, followed by injection onto a microcapillary C18 column. A reverse-phase gradient was run and the elution injected directly into the MS for detection. Resulting MS/MS spectra were searched using SEQUEST (The Scripps Research Institute La Jolla, CA) against a reversed and concatenated Swiss-Prot (UniProt consortium, United Kingdom, Switzerland, and Washington DC) protein database for identification. Scaffold software was then used to calculate the spectral counts per protein, which can be used as a semiquantitative measure of abundance. Three lungs were analyzed for each rat decellularization protocol. For human tissue, samples from the upper left and upper right lobe from 2 decellularized lungs were analyzed.
Lyophilized tissue samples were first digested in a 200-μg proteinase-K/ml–Tris-HCl buffer for 3 hours at 371C. Extracted DNA was then quantified using the Quant-iT PicoGreen doublestranded (ds)DNA Reagent Kit (Invitrogen/Molecular Probes, Eugene, OR) and normalized to starting dry weight values.
Residual SDS Lyophilized tissue samples were first digested in collagenase buffer (10 mg/ml) at 371C for 48 hours. Solubilized SDS was quantified using the Stains All Dye reagent (Sigma-Aldrich, St. Louis, MO) by absorbance readings at 438 nm16 and then normalized to starting dry weight values.
Collagen, elastin, and glycosaminoglycans Collagen, elastin, and glycosaminoglycans (GAGs) matrix components were quantified using the Sircol, Fastin, and Blyscan assays, respectively (Biocolor Ltd, Carrickfergus, United Kingdom), according to the manufacturer’s instructions. Results were normalized to starting dry tissue weight.
Perfusate contamination analysis To assess potential bacterial and fungal contamination of the organ perfusate, samples of the organ perfusate solution (5 ml) were
Figure 1 Comparison of rat lung decellularization protocols. (A) (i) Loss of cellular components in decellularized lung tissue, as visualized by hematoxylin and eosin (H&E). (ii) Maintenance of collagen (blue) visualized by Masson’s trichrome, (iii) glycosaminoglycans (blue) by Alcian blue, and (iv) elastin (blue-black) by Verhoeff-Van Gieson, for the 3 decellularization protocols. Main scale bar ¼ 200 μm, inset scale bar ¼ 20 μm. (B) Loss of residual double-stranded DNA, as quantification by PicoGreen assay, comparing decellularized lung tissue vs cadaveric (n ¼ 3/group). (C) Loss of soluble collagen quantification by Sircol assay, comparing decellularized lung tissue vs cadaveric (n ¼ 3/group). One-way analysis of variance with Dunnett’s post-test. The error bars show the standard deviation. CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate; NS, not significant; SDC, sodium deoxycholate; SDS, sodium dodecyl sulfate.
Gilpin et al.
Perfusion Decellularization of Lungs
301
Figure 2 Peptide analysis of decellularized rat lung matrices by liquid chromatography tandem mass spectrometry. Spectral counts, representing a semi-quantitative measure of peptide abundance, indicate greatest preservation of (A) collagen, (B) Laminin, and (C) several other extracellular matrix components in lungs decellularized by sodium dodecyl sulfate (SDS) perfusion. BM, basement membrane; EMLIN, elastin microfibril interface; (D) Myosin peptides were measured at highest levels in sodium deoxycholate (SDC) decellularized lungs. (E) The largest amount of residual cytoplasmic peptides was quantified after decellularization by CHAPS (3-[(3-cholamidopropyl) dimethylammonio]-1-propanesulfonate), and the least found in SDS decellularized lungs. The error bars indicate the standard deviation.
inoculated into lysogeny (L)-broth (7 ml) and cultured for 96 hours at 371C. Bioburden was quantified by broth absorbance at 425 nm and 600 nm.
Lung slice culture Decellularized rat and human lung tissue was fixed in 30% sucrose, embedded in optimum cutting temperature compound, and stored at –801C. Sections 50-μm thick and 10- to 12-mm long were cut and immobilized on standard tissue culture plates. Sequential washes with 100% and 95% ethanol were applied. and sections were air-dried for 30 minutes. Sections were washed 3 times in PBS and incubated overnight in PBS to remove any residual contaminating detergents. Before cell seeding, slices were pre-conditioned with the appropriate culture media for 3 hours. Small airway epithelial cells (SAECs; Lonza, Walkersville, MD), pulmonary alveolar epithelial cells (PAECs; ScienCell, Carlsbad, CA), or human umbilical vein endothelial cells (HUVECs), at 5 105 per cell type, were seeded directly on top of the lung matrix slice and cultured for 5 days in Small Airway Growth Media (Lonza), Alveolar Epithelial Cell Medium (ScienCell), and Endothelial Growth Media-2 (Lonza), respectively. Calcein AM Viability Dye (12.5 nmol/liter; eBiosciences, San Diego, CA) and nuclear-specific Hoechst 33342 dye was added to the cultures, and images were captured on a Ti Eclipse Microscope (Nikon, Tokyo, Japan). Integrin staining for α2β1 and α3β1 (Abcam, Cambridge, MA) was performed after ice-cold methanol fixation and included a Hoechst 33342 nuclear stain.
MTT assay SAECs (2.5 104 per slice), PAECs (2 105 per slice), or HUVECs (1 106 per slice) were cultured on decellularized lung slices as described above or on standard tissue culture plastic. Each cell type was seeded to 6 individual slices in replicate. On Day 5, MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) reagent (Roche Diagnostics, Mannheim, Germany) was added to culture media for 4 hours, followed by lysis reagent overnight. Total dye was quantified by absorbance at 600 nm for 1 second and compared between matrix and plastic-only culture by t-test. A standard curve of known cell numbers was prepared and analyzed in parallel.
Recellularization and biomimetic lung culture The upper right lobe was isolated from an intact human decellularized lung that had been stored at 41C in PBS with antibiotics after decellularization. The main airway, artery, and vein were identified and cannulated, and the lobe was mounted in the bioreactor under sterile conditions. The pulmonary vein cannula was left open to allow passive drainage. Gravity was used to instill 500 106 cultured human small airway PAEC (ScienCell) through the airway. Alveolar Epithelial Cell Medium (ScienCell) was circulated by pressure-controlled perfusion (30 mm Hg, 100 ml/min) initiated 3-hours after cell delivery. Biomimetic culture was maintained for 96 hours with 1 full media change on Day 2, and then the lung tissue was fixed and processed for histologic analysis.
302
The Journal of Heart and Lung Transplantation, Vol 33, No 3, March 2014
Results We first aimed to analyze 3 published decellularization protocols for rat lung4–6 to determine the most suitable methods for subsequent up-scaling. A perfusion-based decellularization method was adopted for each protocol. The specific protocol details are outlined in the Methods and Supplemental Figure 1 (available in the online version of this article at jhltonline.org). Histologic analysis of rat lungs decellularized by each protocol revealed equivalent removal of intracellular and nuclear components. Maintenance of extracellular lung architecture was confirmed across protocols, including the visualization of intact airway, alveolar, and vascular elements. The presence of elastic fiber and GAG was confirmed in the decellularized tissue (Figure 1A). A significant loss of residual dsDNA was found in SDS and SDC lungs, whereas CHAPS decellularized lungs maintained a higher level of DNA (Figure 1B). The highest level of soluble collagen was measured in SDS decellularized lungs, whereas a decreased amount was found in SDC and CHAPS lungs compared with control tissue (Figure 1C). Microcapillary LC-MS/MS further identified specific peptide components of the rat scaffolds. The generated spectral counts, used as a semi-quantitative measure of protein abundance, identified 341 proteins across all samples. ECM components and cell-associated proteins were compared among the 3 protocols (Figure 2). A higher number of collagen and laminin peptide counts were identified in the SDS decellularized lungs compared with the other 2 protocols (n ¼ 3 lungs per protocol; Figure 2A and B). In addition, we noted the highest level of myosin components remained in SDC decellularized lungs (Figure 2C and D), whereas CHAPS lungs demonstrated a higher level of cytoplasmic proteins, suggesting incomplete decellularization or ineffective washing in this protocol (Figure 2E). We determined from the results of our previous and current rat lung experiments that the SDS protocol was suitable for up-scaling to larger porcine and human lungs. Porcine lungs were harvested on site, whereas human lungs not suitable for transplantation were obtained from consented research organ donors through the IIAM and the NEOB. Sterile surgical technique was maintained throughout the recovery procedure and decellularization process. Lungs were recovered in a standard fashion, transported on ice, and initially underwent a sterile saline retrograde flush of the vasculature to remove any possible blood clots and then cannulated through the pulmonary artery. Lungs were next transferred to the perfusion chamber, which was custom-designed to accommodate human-sized lungs and hold up to 60 liters of fluid used for decellularization daily (Figure 3A). The up-scaling of the decellularization protocol required increasing the concentration of SDS to 0.5%, increasing the perfusion pressure within to a range of 40 to 60 mm Hg, and increasing the total time of SDS perfusion to the order of days (Figure 3B). The decellularization system included a feedback pressure-control mechanism to maintain constant perfusion parameters and continuous recording of pressure,
flow rate, and vascular resistance during the course of decellularization. As the lungs progressed through decellularization, blood and cell debris was observed exiting the pulmonary and veins and trachea, and the organ swelled to fill the chamber while in the SDS solution. Vascular resistance was calculated from recorded pressure and fluid flow data (Figure 3C). Variability between lungs was noted, and on average, porcine lungs had lower resistance. The solutions were changed every 24 hours, and a visual color change was observed moving from pink, to beige, to opaque white (Figure 3D). Once detergent perfusion was completed, the organ was washed by perfusion with several daily changes of PBS, allowing it to return to normal, physiologic size.
Figure 3 The process of perfusion decellularization for whole porcine and human lungs. (A) Lungs are sterilely cannulated through the pulmonary artery and mounted in a custom-designed decellularization chamber that holds 40 liters of perfusate. Vascular perfusion is facilitated by a pressure-controlled pump system, with continuous data acquisition. (B) The up-scaling of the decellularization protocol to large organs included increasing sodium dodecyl sulfate (SDS) concentration, perfusion pressure, and total procedure time. PBS, phosphate-buffered saline. (C) Vascular resistance of porcine (PL) and human lungs (HL) during the first stage of SDS perfusion. Resistance was calculated from recorded pressure and fluid flow values. The horizontal line in the middle of each box indicates the median; the top and bottom borders of the box mark the 75th and 25th percentiles, respectively; and the whiskers mark the 5th and 95th percentiles. (D) Visual changes observed in the lungs during the course of decellularization. Remaining blood and then cellular material are seen draining from the pulmonary vein and trachea as the lungs transition from a pink color to opaque white. Swelling is also observed during SDS perfusion but resolves throughout the washing phase.
Gilpin et al.
Perfusion Decellularization of Lungs
303
Figure 4 Analysis of the decellularized lung matrix. (A) Bacterial and fungal growth after perfusate inoculation into lysogeny (L)-broth (culture for 4 days) indicates minimal contamination. (B) Sodium dodecyl sulfate (SDS) quantification in digested decellularized lung tissue by Stains-All assay confirms detergent removal after washing. The horizontal line in the middle of each box indicates the median; the top and bottom borders of the box mark the 75th and 25th percentiles, respectively; and the whiskers mark the 5th and 95th percentiles. (C) Residual double-stranded DNA quantification in digested decellularized lung tissue by PicoGreen assay confirms effective decellularization for human (HL 1–3) and porcine (PL 5–7) lungs compared with cadaveric lung tissue (HL CAD 4–5; n ¼ 7–10 tissue pieces per lung). One-way analysis of variance with Dunnett’s post-test. The error bars indicate the standard deviation.
Gross visual examination showed the scaffold retained pleura integrity, lobar definition, and major vasculature and airways. Sequential samples of the perfusion fluid were tested for bacterial and fungal contamination, confirming that no significant growth was detected after 4 days of culture in L-broth (Figure 4A). Analysis of several samples across the resulting lung scaffold confirmed removal of SDS in the lung tissue at the end of the protocol, with most of the analyzed samples reading undetectable levels of SDS (Figure 4B). The level of measurable dsDNA remaining within the decellularized tissue was significantly reduced compared with cadaveric tissue (Figure 4C). Preservation of decellularized tissue architecture and loss of cells was confirmed by hematoxylin and eosin staining (Figure 5A). Conservation of elastin and GAG components was also confirmed by Alcian blue and Verhoeff-Van Gieson staining, respectively (Figure 5A). Soluble collagen, GAGs, and elastin were measured throughout the decellularized lung tissue in a homogenous manner (Figure 5B–D). An overall loss of soluble collagen and elastin was quantified in the decellularized lung tissue when compared with cadaveric tissues. An enrichment of GAGs was also measured in the decellularized lung, relative to cadaveric tissue. Immunohistochemistry for collagen IV, fibronectin, and laminin further demonstrated the conservation of these ECM components in the decellularized human lung tissue (Figure 5E). Transmission electron microscopy (original magnification 3,000–10,000) showed the basement membrane was also intact (Figure 5F). Although each of these techniques only examines a component of the entire
matrix, taking these data points together, we feel comfortable stating that the ECM scaffolds were well preserved. Further investigation of the matrix by LC-MS/MS revealed additional peptide components were retained in the decellularized lung. We identified 70 peptides in one lung and 75 peptides in a second, with 37 proteins overlapping both samples. High levels of collagen, laminin, and other matrix components were identified in a similar pattern to that identified in SDS-decellularized rat lungs. To a lesser degree, other secreted proteins were detected within the decellularized scaffold (Figure 6). No intracellular peptides were identified within the samples analyzed. To investigate the biocompatibility of the decellularized rat and human lung tissue, a sterile tissue slice culture procedure was developed in conjunction with live microscopy imaging. Adult-derived SAECs, PAECs, or HUVECs cultured on top of decellularized rat or human lung matrix slices confirmed the biocompatibility of the decellularized matrix (bright-field images, Figure 7A, i). After 5 days of standard culture, cell viability was confirmed across cell types, with cells observed adherent to the matrix slices as well as on the surrounding tissue culture plastic (Figure 7A, ii). Images at higher magnification showed interaction of the cells with matrix, demonstrating attachment and elongation (Figure 7A, iii). Staining for integrins α2β1 and α3β1, which recognize collagen I/IV and laminin 5/10/11, respectively,18 demonstrated a specific interaction between the cells and matrix components in the tissue slice cultures (Figure 7A iv and v). Further quantification of cell growth and survival in the context of decellularized human lung slices was achieved by MTT assay after 5 days of culture.
304
The Journal of Heart and Lung Transplantation, Vol 33, No 3, March 2014
Figure 5 Matrix analysis of porcine and human decellularized lung tissue. (A) (i) Loss of cellular components in decellularized human and porcine lung tissue, as visualized by hematoxylin and eosin (H&E). (ii) Preservation of glycosaminoglycans (blue) by Alcian blue, and (iii) elastin (blue-black) by Verhoeff-Van Gieson across species. Main scale bar ¼ 200 μm, inset scale bar ¼ 20 μm. Quantification of (B) soluble collagen, (C) glycosaminoglycan, and (D) elastin in decellularized porcine (PL 5–7) and human (HL 1–3) lung tissue compared with cadaveric tissue (PL and HL CAD; n ¼ 7–10 tissue pieces per lung). Quantitative values all normalized to dry starting tissue weight. One-way analysis of variance with Dunnett’s post-test. (E) Immunohistochemistry staining of decellularized human lung tissue for collagen IV, fibronectin, and laminin. Scale bar ¼ 200 μm (upper) and 20 mm (lower). (F) Transmission electron microscopy demonstrating intact basement membrane in decellularized human lung tissue. (i and ii) Original magnification 3,000, 10-μm scale bar; (iii) original magnification 8,000, 2-μm scale bar; (iv) original magnification 10,000, 2-μm scale bar. The error bars show the standard deviation. *p o 0.05; ***p o 0.001
No difference in cell growth or survival was found between standard tissue culture and culture on the lung matrix, indicating biocompatibility of the decellularized matrix (Figure 7B–D). The decellularized lungs were further validated for their potential to serve as scaffolds for whole organ recellularization. The integrity of the lung matrix was verified by applying positive-pressure inflation by using a modified piston pump (Harvard Apparatus, Holliston, MA) at a stepwise increasing volume of 20 cm3/stroke and 10 strokes/ min until visual inflation without air leakage was observed (Figure 8A). Successful application of cyclic ventilation was achieved at volumes ranging from 15 to 100 cm3/stroke. Decellularized lungs were compliant with inspiration,
whereas elastic recoil was sufficient to enable passive expiration. Biomimetic culture of a decellularized lung lobe further confirmed the capability of the scaffold to serve as a platform for regeneration. After delivery of 500 106 human PAECs to the airways by gravity, the lungs were successfully cultured with media perfusion under constant pressure for 96 hours (Figure 8B). The integrity of the tissue was maintained throughout biomimetic culture, with no tissue damage and no contamination. Continuity of the vascular network was visualized by passive drainage of media from the pulmonary vein. Analysis of the tissue for cell distribution showed cell retention and appearance of flattened epithelial phenotypes consistent with cell attachment
Gilpin et al.
Perfusion Decellularization of Lungs
305
Figure 6 Peptide analysis of decellularized human lung matrix by liquid chromatography tandem mass spectrometry. Total spectral counts, representing a semi-quantitative measure of peptide abundance within the tissue, indicate a preservation of (A) collagen, (B) laminin, and (C) various other extracellular matrix components. BM, basement membrane; EMLIN, elastin microfibril interface located protein. (D) Other secreted, extracellular proteins were retained within the matrix. The error bars show the standard deviation.
and engraftment onto the basement membrane. As expected, the retention pattern was heterogeneous given limited cell number and short duration of culture (Figure 8C). Taken together, these results validate our protocol for the generation of clinical scale acellular lung scaffolds from porcine and human donors and provide initial proof of potential for matrix recellularization.
Discussion More than 1,700 Americans are in need of a donor lung,1 with limited or no available alternative therapies.19 Waiting list mortality rates for lung candidates continue to rise, reported at 15.7/100 waiting list-years in 2011.20 In addition, only 21.0% of potential donor lungs are used for transplantation,20 further highlighting the shortage of adequate organs. A 50% probability of graft failure and more than 40% probability of death at 5 years after transplant remain clinical realities lung transplant recipients continue to face.20 These facts underscore the theoretical clinical utility of a recipient-specific, regenerated organ. Our results demonstrate that porcine and human lungs not suitable for transplantation can be decellularized to generate whole-organ acellular scaffolds as a platform for organ regeneration. The procedure we have described in this report uses a simple, inexpensive detergent and a constant-pressure perfusion system to generate an acellular lung matrix that retains robust extracellular components and architecture. We aimed to use the rat model to compare decellularization detergents and inform the development of a large-scale protocol. Because the comparison was not repeated at the porcine or human level, the optimal perfusate cannot be directly determined, although the approach we have reported proved successful in our hands. When moving our promising small-animal data toward clinical translation, we faced the challenge of up-scaling this technology to generate a consistent, reproducible lung
scaffold. One question we considered was the source of donor lungs, which we aimed to address by applying our technology to human and porcine sources. We did not identify any differences when comparing porcine and human lungs in response to detergent or resulting matrix components. The applicability of our procedure to both species is promising because sourcing porcine lungs may be advantageous with regards to product development and regulatory approval. Porcine lungs could potentially be used as a whole-organ xenograft because porcine biologics are already used in clinical application, such as matrix derived from small-intestinal sub-mucosa for wound care21 and pericardial reconstruction.22 One question not yet addressed is the potential immune response elicited from xenotransplantation and any surface antigens that may remain within the ECM.23 If looking to human lung donors, the concept of human-derived transplantable tissue is supported by products such as biomesh derived from human skin.24 To broadly use human donor lungs, the variability of the donor pool must be addressed, especially amongst an extended criteria pool. The underlying donor pathology imparts potential differences to the native matrix, which may result in altered regenerative capacity of the scaffold.25,26 Our current results show that reproducible ECM scaffolds can be achieved with a variety of lung donors within the donor criteria we tested. We can report that decellularization failed to occur in 1 lung donor with a history of asthma and heavy marijuana use, perhaps due to underlying matrix or vascular damage. This lung showed signs of hyperinflation and parenchymal damage on macroscopic inspection and could not be salvaged as a scaffold. The lung is a particularly challenging donor organ due to its constant contact with the external environment and increased potential for infection or contamination. Lungs are also highly susceptible to injuries after donor brain death, including pulmonary edema, pneumonia, aspiration, or
306
The Journal of Heart and Lung Transplantation, Vol 33, No 3, March 2014
Figure 7 In vitro cell culture on decellularized lung matrix. (A) Live imaging of cells cultured on human lung matrix slices (50 μm). (i) Bright field images and (ii-iii) after treatment with viability dye (green) and Hoechst nuclear dye (blue) on Day 5. Robust cell viability is confirmed for small airway epithelial cells (SAECs), human umbilical vein endothelial cells (HUVECs), and pulmonary alveolar epithelial cells (PAECs) cultured on decellularized human slices. Higher magnification images highlight the interaction of viable cells with the matrix, demonstrating cell elongation and attachment to the matrix. (ii) Scale bar ¼ 100 μm; (iii) scale bar ¼ 50 μm. Integrin expression (red) and interaction with decellularized matrix (green) is confirmed for (iv) α2β1 and (v) α3β1. Scale bars ¼ 50 μm. (B–D) Cells cultured on human lung slices for 5 days survive and proliferate, as quantified by MTT assay, with no overall cytotoxicity associated with matrix-cell interaction. The error bars show the standard deviation.
ventilator-induced lung injury.27 These complications may alter their utility as source organs for decellularization. Similarly, the porcine airway is at risk for infection with variety of viral and bacterial pathogens,28 and more targeted treatment may be required to ensure downstream sterility of the decellularized organ. We have found that treatment with 0.08% peracetic acid solution is effective in reducing the bioburden in tissue before and after decellularization, as has been previously reported.29 Ethylene oxide has been used for the sterilization
of urinary bladder matrix, with minimal effects on structural properties compared with gamma or electron beam radiation.30 Even a moderate gamma dose (7–12 kGy) has been reported to cause human dermis ECM denaturation,31 emphasizing the need for optimized sterilization protocols to preserve essential matrix components. If lung scaffolds were to be produced for off-the-shelf use, protocols enabling the preservation and storage of these matrices will also need to be developed32,33 because these procedures may affect the integrity and utility of the scaffold.30,31,34
Gilpin et al.
Perfusion Decellularization of Lungs
Figure 8 Ventilation and biomimetic culture of human decellularized lungs. (A) (i) Fully deflated and (ii) maximal positive inflation of human decellularized lungs. (B) Upper right decellularized human lobe cannulated and mounted in bioreactor. (C) Hematoxylin and eosin staining of tissue after pulmonary alveolar epithelial cells delivery and 96 hours of biomimetic culture. Scale bar ¼ 100 μm. (D) TUNEL (terminal deoxynucleotide transferase-mediated deoxy uridine triphosphate nick-end labeling) fluorometric assay for apoptotic cells demonstrates minimal cell death (green) in recellularized lung tissue both proximal and distal to the perfusion cannula. Blue ¼ 4',6-diamidino-2phenylindole (DAPI). Scale bar ¼ 50 μm.
307 As the next step toward creation of viable tissue, we must address the challenges of recellularization. This will require the replacement of many different lung cell phenotypes, including airway and alveolar epithelium along with endothelium. Our results suggest that mature cell phenotypes from these 3 lineages can attach and survive in culture with decellularized lung matrix. The tissue slice model is a useful tool to test possible cell types in terms of cell-matrix attachment, proliferation, and overall biocompatibility in a high-throughput, 2-dimensional experimental design13,36–38 and can be applied to various cell types and matrices. Ultimately, the next steps toward clinical translation will be the generation of 3-dimensional, mature, functional tissue. This progress will require moving from the tissue culture dish to a whole-organ system and the design of appropriate bioreactors to recapitulate the body’s natural environment and physiology.5,6 Although we were able to maintain tissue integrity and cell retention for 96 hours of biomimetic culture, further refinements will be required to optimize ventilation and perfusion strategies to create an ideal environment for cell attachment, proliferation, and full tissue regeneration. Current advances in pulmonary stem cell differentiation have contributed to the ultimate goal of functional organ regeneration from patient-specific, pluripotent cells.37,39 The elucidation of specific milestones in lung development allows for directed differentiation of stem cells before lung reseeding and stage-specific investigation of lung progenitor populations in terms of attachment, proliferation, and further differentiation capacity. Once seeded, the native decellularized matrix may act through biologic signaling to influence stem cell differentiation pathways,40,41 further assisting in the regeneration process. Our proteomic analysis of the scaffold demonstrates the retention of critical basement membrane components, including collagen IV and laminin, which are required for epithelial cell attachment. To achieve the goal of truly patient-specific organs regenerated with autologous cells, identification of the optimal developmental stage of seeded cells and adaptation of subsequent organ culture protocols will be critical. Ultimately, the generation of human-scale lung scaffolds brings the goal of organ regeneration one step closer to clinical application. The field must now tackle the larger challenges of recellularization and restoring organ function to ultimately create transplantable organs for clinical use.
Disclosure statement Functional regeneration of lung scaffolds for transplantation will require the conservation of scaffold mechanics after decellularization. Our results demonstrated an average of greater than 60% conservation of elastic fibers in porcine scaffolds and greater than 45% conservation in human scaffolds; however, we have yet to develop a standardized battery of tests to ensure consistency throughout the decellularization process. Real-time testing of isolated lung tissue, including mechanical stress and strain evaluation, may provide a tool for continuous scaffold assessment.35
This study was supported by the United Therapeutics Corporation and the National Institutes of Health (NIH) Director’s New Innovator Award (DP2-OD008749-01). This work was presented at the Thirty-third Annual Meeting and Scientific Sessions of the International Society for Heart and Lung Transplantation, Montreal, Québec, Canada, April 24–27, 2013. HC Ott is founder and stockholder of IVIVA Medical Inc. This relationship did not influence design, execution, and/or interpretation of the present study. None of the authors has a financial relationship with a commercial entity that has an interest in the subject of the presented manuscript or other conflicts of interest to disclose.
308
The Journal of Heart and Lung Transplantation, Vol 33, No 3, March 2014
Supplementary data Supplementary data are available in the online version of this article at jhltonline.org.
References 1. Health Resources and Services Administration. Organ Procurement and Transplantation Network O. National Data Reports, as of April 5, 2013. Available from: http://optn.transplant.hrsa.gov/. Accessed December 4, 2013. 2. Soto-Gutierrez A, Wertheim JA, Ott HC, Gilbert TW. Perspectives on whole-organ assembly: moving toward transplantation on demand. J Clin Invest 2012;122:3817-23. 3. Badylak SF, Weiss DJ, Caplan A, Macchiarini P. Engineered whole organs and complex tissues. Lancet 2012;379:943-52. 4. Ott HC, Clippinger B, Conrad C, et al. Regeneration and orthotopic transplantation of a bioartificial lung. Nat Med 2010;16:927-33. 5. Petersen TH, Calle EA, Zhao L, et al. Tissue-engineered lungs for in vivo implantation. Science 2010;329:538-41. 6. Price AP, England KA, Matson AM, Blazar BR, Panoskaltsis-Mortari A. Development of a decellularized lung bioreactor system for bioengineering the lung: the matrix reloaded. Tissue Eng Part A 2010;16:2581-91. 7. Wallis JM, Borg ZD, Daly AB, et al. Comparative assessment of detergent-based protocols for mouse lung de-cellularization and re-cellularization. Tissue Eng Part C Methods 2012;18:420-32. 8. Ott HC, Matthiesen TS, Goh SK, et al. Perfusion-decellularized matrix: using nature’s platform to engineer a bioartificial heart. Nat Med 2008;14:213-21. 9. Crapo PM, Gilbert TW, Badylak SF. An overview of tissue and whole organ decellularization processes. Biomaterials 2011;32:3233-43. 10. Macchiarini P, Jungebluth P, Go T, et al. Clinical transplantation of a tissue-engineered airway. Lancet 2008;372:2023-30. 11. Sullivan DC, Mirmalek-Sani SH, Deegan DB, et al. Decellularization methods of porcine kidneys for whole organ engineering using a highthroughput system. Biomaterials 2012;33:7756-64. 12. Remlinger NT, Wearden PD, Gilbert TW. Procedure for decellularization of porcine heart by retrograde coronary perfusion. J Vis Exp 2012:e50059. 13. Daly AB, Wallis JM, Borg ZD, et al. Initial binding and recellularization of decellularized mouse lung scaffolds with bone marrow-derived mesenchymal stromal cells. Tissue Eng Part A 2012;18:1-16. 14. Jensen T, Roszell B, Zang F, et al. A rapid lung de-cellularization protocol supports embryonic stem cell differentiation in vitro and following implantation. Tissue Eng Part C Methods 2012;18:632-46. 15. Song JJ, Kim SS, Liu Z, et al. Enhanced in vivo function of bioartificial lungs in rats. Ann Thorac Surg 2011;92:998-1005; discussion 1006. 16. Rusconi F, Valton E, Nguyen R, Dufourc E. Quantification of sodium dodecyl sulfate in microliter-volume biochemical samples by visible light spectroscopy. Anal Biochem 2001;295:31-7. 17. Ngoka LC. Sample prep for proteomics of breast cancer: proteomics and gene ontology reveal dramatic differences in protein solubilization preferences of radioimmunoprecipitation assay and urea lysis buffers. Proteome Sci 2008;6:30. 18. Sheppard D. Functions of pulmonary epithelial integrins: from development to disease. Physiol Rev 2003;83:673-86. 19. Bousquet J, Kaltaev N. Global surveillance, prevention and control of chronic respiratory diseases: a comprehensive approach. Geneva: World Health Organization; 2007. p. vii, 146 p. 20. Scientific Registry of Transplant Recipients S. OPTN and SRTR annual data report, 2011. Available from: http://srtr.org/. Accessed April 5, 2013. 21. Mostow EN, Haraway GD, Dalsing M, Hodde JP, King D. Effectiveness of an extracellular matrix graft (OASIS Wound Matrix) in the treatment of chronic leg ulcers: a randomized clinical trial. J Vasc Surg 2005;41:837-43.
22. Boyd WD, Johnson WE 3rd, Sultan PK, Deering TF, Matheny RG. Pericardial reconstruction using an extracellular matrix implant correlates with reduced risk of postoperative atrial fibrillation in coronary artery bypass surgery patients. Heart Surg Forum 2010; 13:E311-6. 23. Cooper DK, Ekser B, Burlak C, et al. Clinical lung xenotransplantation —what donor genetic modifications may be necessary? Xenotransplantation 2012;19:144-58. 24. Jansen LA, De Caigny P, Guay NA, Lineaweaver WC, Shokrollahi K. The evidence base for the acellular dermal matrix AlloDerm: a systematic review. Ann Plast Surg 2013;70:587-94. 25. Booth AJ, Hadley R, Cornett AM, et al. Acellular normal and fibrotic human lung matrices as a culture system for in vitro investigation. Am J Respir Crit Care Med 2012;186:866-76. 26. Sokocevic D, Bonenfant NR, Wagner DE, et al. The effect of age and emphysematous and fibrotic injury on the re-cellularization of decellularized lungs. Biomaterials 2013;34:3256-69. 27. Avlonitis VS, Fisher AJ, Kirby JA, Dark JH. Pulmonary transplantation: the role of brain death in donor lung injury. Transplantation 2003;75:1928-33. 28. Opriessnig T, Gimenez-Lirola LG, Halbur PG. Polymicrobial respiratory disease in pigs. Anim Health Res Rev 2011;12:133-48. 29. Kemp PD. Peracetic acid sterilization of collagen or collagenous tissue. In: Office UP, editor. USA: Organogenesis Inc.; 1995. 30. Freytes DO, Stoner RM, Badylak SF. Uniaxial and biaxial properties of terminally sterilized porcine urinary bladder matrix scaffolds. J Biomed Mater Res B Appl Biomater 2008;84:408-14. 31. Sun WQ, Leung P. Calorimetric study of extracellular tissue matrix degradation and instability after gamma irradiation. Acta Biomater 2008;4:817-26. 32. Nonaka PN, Campillo N, Uriarte JJ, et al. Effects of freezing/thawing on the mechanical properties of decellularized lungs. J Biomed Mater Res A 2013, http://dx.doi.org/10.1002/jbm.a.34708: [E-pub ahead of print]. 33. Bonenfant NR, Sokocevic D, Wagner DE, et al. The effects of storage and sterilization on de-cellularized and re-cellularized whole lung. Biomaterials 2013;34:3231-45. 34. Dijkman PE, Driessen-Mol A, Frese L, Hoerstrup SP, Baaijens FP. Decellularized homologous tissue-engineered heart valves as off-theshelf alternatives to xeno- and homografts. Biomaterials 2012; 33:4545-54. 35. Petersen TH, Calle EA, Colehour MB, Niklason LE. Matrix composition and mechanics of decellularized lung scaffolds. Cells Tissues Organs 2012;195:222-31. 36. Bonvillain RW, Danchuk S, Sullivan DE, et al. A nonhuman primate model of lung regeneration: detergent-mediated decellularization and initial in vitro recellularization with mesenchymal stem cells. Tissue Eng Part A 2012;18:2437-52. 37. Longmire TA, Ikonomou L, Hawkins F, et al. Efficient derivation of purified lung and thyroid progenitors from embryonic stem cells. Cell Stem Cell 2012;10:398-411. 38. Shamis Y, Hasson E, Soroker A, et al. Organ-specific scaffolds for in vitro expansion, differentiation and organization of primary lung cells. Tissue Eng Part C Methods 2011;17:861-70. 39. Mou H, Zhao R, Sherwood R, et al. Generation of multipotent lung and airway progenitors from mouse ESCs and patient-specific cystic fibrosis iPSCs. Cell Stem Cell 2012;10:385-97. 40. Hosokawa T, Betsuyaku T, Nishimura M, Furuyama A, Katagiri K, Mochitate K. Differentiation of tracheal basal cells to ciliated cells and tissue reconstruction on the synthesized basement membrane substratum in vitro. Connect Tissue Res 2007;48:9-18. 41. Gong J, Sagiv O, Cai H, Tsang SH, Del Priore LV. Effects of extracellular matrix and neighboring cells on induction of human embryonic stem cells into retinal or retinal pigment epithelial progenitors. Exp Eye Res 2008;86:957-65.
ORIGINAL PRE-CLINICAL SCIENCE
Perfusion decellularization of human and porcine lungs: Bringing the matrix to clinical scale Sarah Elizabeth Gilpin, PhD,a,b Jacques P. Guyette, PhD,a,b Gabriel Gonzalez, PhD,a,b Xi Ren, PhD,a,b John M. Asara, PhD,a,b Douglas J. Mathisen, MD,a,b Joseph P. Vacanti, MD,a,c and Harald C. Ott, MDa,b,d,e From the aCenter for Regenerative Medicine, Massachusetts General Hospital; bHarvard Medical School; cDivision of Pediatric Surgery; and dDivision of Thoracic Surgery, Department of Surgery, Massachusetts General Hospital; and the eHarvard Stem Cell Institute, Boston, Massachusetts.
KEYWORDS: organ engineering; organ scaffolds; acellular matrix; recellularization; decellularization
BACKGROUND: Organ engineering is a theoretical alternative to allotransplantation for end-stage organ failure. Whole-organ scaffolds can be created by detergent perfusion via the native vasculature, generating an acellular matrix suitable for recellularization with selected cell types. We aimed to up-scale this process, generating biocompatible scaffolds of a clinically relevant scale. METHODS: Rat, porcine, and human lungs were decellularized by detergent perfusion at constant pressures. Collagen, elastin, and glycosaminoglycan content of scaffolds were quantified by colorimetric assays. Proteomic analysis was performed by microcapillary liquid chromatography tandem mass spectrometry. Extracellular matrix (ECM) slices were cultured with human umbilical vein endothelial cells (HUVEC), small airway epithelial cells (SAEC), or pulmonary alveolar epithelial cells (PAECs) and evaluated by time-lapse live cell microscopy and MTT (3-[4,5-dimethylthiazol-2-yl]-2, 5-diphenyltetrazolium bromide) assay. Whole-organ culture was maintained under constant-pressure media perfusion after seeding with PAECs. RESULTS: Rat lungs were decellularized using: (1) sodium dodecyl sulfate (SDS), (2) sodium deoxycholate (SDC), or (3) 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS). Resulting scaffolds showed comparable loss of DNA but greatest preservation of ECM components in SDS-decellularized lungs. Porcine (n ¼ 10) and human (n ¼ 7) lungs required increased SDS concentration, perfusion pressures, and time to achieve decellularization as determined by loss of DNA, with preservation of intact matrix composition and lung architecture. Proteomic analysis of human decellularized lungs further confirmed ECM preservation. Recellularization experiments confirmed scaffold biocompatibility when cultured with mature cell phenotypes and scaffold integrity for the duration of biomimetic culture. CONCLUSIONS: SDS-based perfusion decellularization can be applied to whole porcine and human lungs to generate biocompatible organ scaffolds with preserved ECM composition and architecture. J Heart Lung Transplant 2014;33:298–308 r 2014 International Society for Heart and Lung Transplantation. All rights reserved.
In 2012, more than 28,000 organ transplants were performed in the United States, while more than 128,000 Americans Reprint requests: Harald C. Ott, MD, Massachusetts General Hospital, Department of Surgery, Harvard Medical School, 185 Cambridge St, CPZN 4812, Boston, MA 02114. E-mail address: [email protected]
remain on waiting lists with waiting times extending to several years in some instances.1 As the need for alternate sources of donor organs continuously increases, whole-organ engineering based on human or large-animal acellular scaffolds presents a unique opportunity and holds the potential for long-term graft survival. Creating regenerated donor organs by repopulating organ scaffolds with donor-specific mature or
1053-2498/$ - see front matter r 2014 International Society for Heart and Lung Transplantation. All rights reserved. http://dx.doi.org/10.1016/j.healun.2013.10.030
Gilpin et al.
Perfusion Decellularization of Lungs
stem cell-derived populations would allow us to personalize transplantation medicine and reduce the need for long-term immunosuppressive therapy.2 Bioengineering of such wholeorgan grafts requires scaffolds that outline the lung’s structure and composition, thus providing a hospitable and structurally intact niche for cells of interest to attach and engraft.3 Lung decellularization has been successfully achieved using a variety of approaches and protocols,4–7 and the concept has been applied to various other organs across multiple species.8–12 A non-perfusion decellularization approach using Triton-X and sodium deoxycholate (SDC) produced structurally intact lung matrices that were biocompatible with mesenchymal stem cells13 and pre-differentiated murine embryonic stem cells.14 The vascular perfusion of CHAPS (3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate) detergent also successfully decellularized rat lungs, which subsequently were recellularized to recapitulate mechanical and physiologic organ function.5 Our group has reported a perfusion decellularization protocol using sodium dodecyl sulfate (SDS). The acellular rat lung scaffolds were recellularized with carcinomatous human alveolar basal epithelial cells (cell line A549) or rat fetal lung cells, combined with endothelial cells, resulting in partial restoration of function.4,15 Regardless of protocol, once successful decellularization has been accomplished, the challenge of site-specific recellularization remains; the physiologic functions of the lung depend on its complex organ architecture and multiple cell types required. The decellularized scaffold needs to support cell attachment, proliferation, and engraftment to restore the physiologic context of cells and tissues that enable organ integrity and function. To serve as a robust scaffold for regeneration, the matrix must maintain the essential structural proteins for cell integration and resist proteolytic degradation, at least initially. The present study outlines the development of a reproducible procedure to generate clinical-scale porcine and human lung scaffolds as the first step toward generating a transplantable, bioartificial lung. These constructs are devoid of residual donor-derived cells and maintain the hierarchical structural components of lung vasculature and airways necessary for subsequent recellularization and organ regeneration approaches. To determine the most suitable protocol for larger porcine and human organs, we first optimized the method used by comparing matrices generated with 3 published detergents. We did not include protocols involving repeated freezing, freeze-drying, or enzymatic digestion to avoid the inherent extracellular matrix (ECM) damage. We then applied a single, optimized decellularization protocol to organs of clinical scale and analyzed the resulting scaffolds. Finally, we validated the tissue integrity and biocompatibility of these lung scaffolds.
Methods
299 rats (150–200 g, Charles River Laboratories, Inc, Wilmington, MA) were used throughout. Porcine lungs were harvested in-house from Yorkshire swine (40–200 kg) being otherwise used for experiments not involving the chest cavity. Rat donors were pretreated with a dose of heparin (2.0 U/g of body weight) into the inferior vena cava. Porcine donors were not pre-treated with heparin. Human donors were heparinized at the time or organ harvest following standard organ procurement protocols of the New England Organ Bank (NEOB). The rat lung protocols tested used (1) 0.1% SDS,4 (2) 2% SDC,6 or (3) 8 mmol/liter CHAPS (pH 12)5 detergent perfused through the pulmonary artery at a constant pressure of 30 cm H2O. The protocol for all 3 detergents included:
1. a 10-minute initial antegrade wash with phosphate-buffered saline (PBS),
2. detergent perfusion for the time required to visualize an opaque translucent matrix (indicative of decellularization) plus an extra 20% of that initial time (e.g., 70 minutes þ 14 minutes), 3. 15-minute deionized H2O wash, and 4. an additional 72-hour PBS wash with added antibiotics and antimycotics. The SDS protocol included an additional wash of 1% Triton-X following the deionized H2O. The SDC protocol included a 0.1% Triton-X perfusion before SDC and a 1 mol/liter NaCl wash after SDC (see Supplemental Figure 1, available on the jhltonline.org Web site). Human lungs otherwise unsuitable for transplantation were recovered under sterile conditions within 60 minutes of cessation of cardiovascular circulation and obtained from the International Institute for the Advancement of Medicine (IIAM) or the NEOB. Donor criteria included age o 75 years, negative serologies, tobacco smoking of o 10 pack-years, and no known lung disease, pneumonia, aspiration, or trauma. As described for rat lungs, porcine and human lungs were perfused though the pulmonary artery at constant pressure, followed by sequential washing with H2O, 1% Triton-X solution, and PBS. Similar to rat lungs, decellularization was deemed complete upon visual inspection. Variability in the starting organ, mainly due to extensiveness of pre-flushing during harvest and any resulting clots contributed to the required length of perfusion. The time of SDS perfusion varied from 4 to 7 days. Further details of the decellularization protocol are discussed within the Results.
Histologic analysis After decellularization, several areas of the lung (n ¼ 8–12 tissue samples per lung) sampled randomly from proximal and distal areas of each lobe were isolated, fixed with 5% formalin, and embedded in paraffin. Sections cut to 5 μm were deparaffinized and stained with standard hematoxylin and eosin and Masson’s Trichrome staining. Proteoglycans were visualized by Alcian blue staining, and elastin components by VerhoeffVan Gieson staining. Immunohistochemistry for matrix proteins (collagen IV, fibronectin, and laminin) was done following standard protocols.
Whole-organ perfusion decellularization
Quantitative biochemical matrix assays
Animal organs were harvested in accordance with approved animal protocols and institutional guidelines. Outbred Sprague-Dawley
After decellularization, random samples of each lung lobe (n ¼ 8–12 per lung) from proximal and distal areas of multiple lobes
300
The Journal of Heart and Lung Transplantation, Vol 33, No 3, March 2014
were isolated and stored at –801C until used for the following assays.
Liquid chromatography–tandem mass spectrometry proteomic matrix analysis
DNA
Decellularized lung tissue samples isolated from the superior segments of the right (n ¼ 2) and left (n ¼ 2) upper lobes were first lyophilized and then digested in 5 mol/liter urea buffer,17 followed by acetone precipitation. Microcapillary liquid chromatography tandem mass spectrometry (LC-MS/MS), using linear iontrap MS technology, was used to identify and semi-quantify the sample components in isolated protein samples. The protein samples were first digested with trypsin, followed by injection onto a microcapillary C18 column. A reverse-phase gradient was run and the elution injected directly into the MS for detection. Resulting MS/MS spectra were searched using SEQUEST (The Scripps Research Institute La Jolla, CA) against a reversed and concatenated Swiss-Prot (UniProt consortium, United Kingdom, Switzerland, and Washington DC) protein database for identification. Scaffold software was then used to calculate the spectral counts per protein, which can be used as a semiquantitative measure of abundance. Three lungs were analyzed for each rat decellularization protocol. For human tissue, samples from the upper left and upper right lobe from 2 decellularized lungs were analyzed.
Lyophilized tissue samples were first digested in a 200-μg proteinase-K/ml–Tris-HCl buffer for 3 hours at 371C. Extracted DNA was then quantified using the Quant-iT PicoGreen doublestranded (ds)DNA Reagent Kit (Invitrogen/Molecular Probes, Eugene, OR) and normalized to starting dry weight values.
Residual SDS Lyophilized tissue samples were first digested in collagenase buffer (10 mg/ml) at 371C for 48 hours. Solubilized SDS was quantified using the Stains All Dye reagent (Sigma-Aldrich, St. Louis, MO) by absorbance readings at 438 nm16 and then normalized to starting dry weight values.
Collagen, elastin, and glycosaminoglycans Collagen, elastin, and glycosaminoglycans (GAGs) matrix components were quantified using the Sircol, Fastin, and Blyscan assays, respectively (Biocolor Ltd, Carrickfergus, United Kingdom), according to the manufacturer’s instructions. Results were normalized to starting dry tissue weight.
Perfusate contamination analysis To assess potential bacterial and fungal contamination of the organ perfusate, samples of the organ perfusate solution (5 ml) were
Figure 1 Comparison of rat lung decellularization protocols. (A) (i) Loss of cellular components in decellularized lung tissue, as visualized by hematoxylin and eosin (H&E). (ii) Maintenance of collagen (blue) visualized by Masson’s trichrome, (iii) glycosaminoglycans (blue) by Alcian blue, and (iv) elastin (blue-black) by Verhoeff-Van Gieson, for the 3 decellularization protocols. Main scale bar ¼ 200 μm, inset scale bar ¼ 20 μm. (B) Loss of residual double-stranded DNA, as quantification by PicoGreen assay, comparing decellularized lung tissue vs cadaveric (n ¼ 3/group). (C) Loss of soluble collagen quantification by Sircol assay, comparing decellularized lung tissue vs cadaveric (n ¼ 3/group). One-way analysis of variance with Dunnett’s post-test. The error bars show the standard deviation. CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate; NS, not significant; SDC, sodium deoxycholate; SDS, sodium dodecyl sulfate.
Gilpin et al.
Perfusion Decellularization of Lungs
301
Figure 2 Peptide analysis of decellularized rat lung matrices by liquid chromatography tandem mass spectrometry. Spectral counts, representing a semi-quantitative measure of peptide abundance, indicate greatest preservation of (A) collagen, (B) Laminin, and (C) several other extracellular matrix components in lungs decellularized by sodium dodecyl sulfate (SDS) perfusion. BM, basement membrane; EMLIN, elastin microfibril interface; (D) Myosin peptides were measured at highest levels in sodium deoxycholate (SDC) decellularized lungs. (E) The largest amount of residual cytoplasmic peptides was quantified after decellularization by CHAPS (3-[(3-cholamidopropyl) dimethylammonio]-1-propanesulfonate), and the least found in SDS decellularized lungs. The error bars indicate the standard deviation.
inoculated into lysogeny (L)-broth (7 ml) and cultured for 96 hours at 371C. Bioburden was quantified by broth absorbance at 425 nm and 600 nm.
Lung slice culture Decellularized rat and human lung tissue was fixed in 30% sucrose, embedded in optimum cutting temperature compound, and stored at –801C. Sections 50-μm thick and 10- to 12-mm long were cut and immobilized on standard tissue culture plates. Sequential washes with 100% and 95% ethanol were applied. and sections were air-dried for 30 minutes. Sections were washed 3 times in PBS and incubated overnight in PBS to remove any residual contaminating detergents. Before cell seeding, slices were pre-conditioned with the appropriate culture media for 3 hours. Small airway epithelial cells (SAECs; Lonza, Walkersville, MD), pulmonary alveolar epithelial cells (PAECs; ScienCell, Carlsbad, CA), or human umbilical vein endothelial cells (HUVECs), at 5 105 per cell type, were seeded directly on top of the lung matrix slice and cultured for 5 days in Small Airway Growth Media (Lonza), Alveolar Epithelial Cell Medium (ScienCell), and Endothelial Growth Media-2 (Lonza), respectively. Calcein AM Viability Dye (12.5 nmol/liter; eBiosciences, San Diego, CA) and nuclear-specific Hoechst 33342 dye was added to the cultures, and images were captured on a Ti Eclipse Microscope (Nikon, Tokyo, Japan). Integrin staining for α2β1 and α3β1 (Abcam, Cambridge, MA) was performed after ice-cold methanol fixation and included a Hoechst 33342 nuclear stain.
MTT assay SAECs (2.5 104 per slice), PAECs (2 105 per slice), or HUVECs (1 106 per slice) were cultured on decellularized lung slices as described above or on standard tissue culture plastic. Each cell type was seeded to 6 individual slices in replicate. On Day 5, MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) reagent (Roche Diagnostics, Mannheim, Germany) was added to culture media for 4 hours, followed by lysis reagent overnight. Total dye was quantified by absorbance at 600 nm for 1 second and compared between matrix and plastic-only culture by t-test. A standard curve of known cell numbers was prepared and analyzed in parallel.
Recellularization and biomimetic lung culture The upper right lobe was isolated from an intact human decellularized lung that had been stored at 41C in PBS with antibiotics after decellularization. The main airway, artery, and vein were identified and cannulated, and the lobe was mounted in the bioreactor under sterile conditions. The pulmonary vein cannula was left open to allow passive drainage. Gravity was used to instill 500 106 cultured human small airway PAEC (ScienCell) through the airway. Alveolar Epithelial Cell Medium (ScienCell) was circulated by pressure-controlled perfusion (30 mm Hg, 100 ml/min) initiated 3-hours after cell delivery. Biomimetic culture was maintained for 96 hours with 1 full media change on Day 2, and then the lung tissue was fixed and processed for histologic analysis.
302
The Journal of Heart and Lung Transplantation, Vol 33, No 3, March 2014
Results We first aimed to analyze 3 published decellularization protocols for rat lung4–6 to determine the most suitable methods for subsequent up-scaling. A perfusion-based decellularization method was adopted for each protocol. The specific protocol details are outlined in the Methods and Supplemental Figure 1 (available in the online version of this article at jhltonline.org). Histologic analysis of rat lungs decellularized by each protocol revealed equivalent removal of intracellular and nuclear components. Maintenance of extracellular lung architecture was confirmed across protocols, including the visualization of intact airway, alveolar, and vascular elements. The presence of elastic fiber and GAG was confirmed in the decellularized tissue (Figure 1A). A significant loss of residual dsDNA was found in SDS and SDC lungs, whereas CHAPS decellularized lungs maintained a higher level of DNA (Figure 1B). The highest level of soluble collagen was measured in SDS decellularized lungs, whereas a decreased amount was found in SDC and CHAPS lungs compared with control tissue (Figure 1C). Microcapillary LC-MS/MS further identified specific peptide components of the rat scaffolds. The generated spectral counts, used as a semi-quantitative measure of protein abundance, identified 341 proteins across all samples. ECM components and cell-associated proteins were compared among the 3 protocols (Figure 2). A higher number of collagen and laminin peptide counts were identified in the SDS decellularized lungs compared with the other 2 protocols (n ¼ 3 lungs per protocol; Figure 2A and B). In addition, we noted the highest level of myosin components remained in SDC decellularized lungs (Figure 2C and D), whereas CHAPS lungs demonstrated a higher level of cytoplasmic proteins, suggesting incomplete decellularization or ineffective washing in this protocol (Figure 2E). We determined from the results of our previous and current rat lung experiments that the SDS protocol was suitable for up-scaling to larger porcine and human lungs. Porcine lungs were harvested on site, whereas human lungs not suitable for transplantation were obtained from consented research organ donors through the IIAM and the NEOB. Sterile surgical technique was maintained throughout the recovery procedure and decellularization process. Lungs were recovered in a standard fashion, transported on ice, and initially underwent a sterile saline retrograde flush of the vasculature to remove any possible blood clots and then cannulated through the pulmonary artery. Lungs were next transferred to the perfusion chamber, which was custom-designed to accommodate human-sized lungs and hold up to 60 liters of fluid used for decellularization daily (Figure 3A). The up-scaling of the decellularization protocol required increasing the concentration of SDS to 0.5%, increasing the perfusion pressure within to a range of 40 to 60 mm Hg, and increasing the total time of SDS perfusion to the order of days (Figure 3B). The decellularization system included a feedback pressure-control mechanism to maintain constant perfusion parameters and continuous recording of pressure,
flow rate, and vascular resistance during the course of decellularization. As the lungs progressed through decellularization, blood and cell debris was observed exiting the pulmonary and veins and trachea, and the organ swelled to fill the chamber while in the SDS solution. Vascular resistance was calculated from recorded pressure and fluid flow data (Figure 3C). Variability between lungs was noted, and on average, porcine lungs had lower resistance. The solutions were changed every 24 hours, and a visual color change was observed moving from pink, to beige, to opaque white (Figure 3D). Once detergent perfusion was completed, the organ was washed by perfusion with several daily changes of PBS, allowing it to return to normal, physiologic size.
Figure 3 The process of perfusion decellularization for whole porcine and human lungs. (A) Lungs are sterilely cannulated through the pulmonary artery and mounted in a custom-designed decellularization chamber that holds 40 liters of perfusate. Vascular perfusion is facilitated by a pressure-controlled pump system, with continuous data acquisition. (B) The up-scaling of the decellularization protocol to large organs included increasing sodium dodecyl sulfate (SDS) concentration, perfusion pressure, and total procedure time. PBS, phosphate-buffered saline. (C) Vascular resistance of porcine (PL) and human lungs (HL) during the first stage of SDS perfusion. Resistance was calculated from recorded pressure and fluid flow values. The horizontal line in the middle of each box indicates the median; the top and bottom borders of the box mark the 75th and 25th percentiles, respectively; and the whiskers mark the 5th and 95th percentiles. (D) Visual changes observed in the lungs during the course of decellularization. Remaining blood and then cellular material are seen draining from the pulmonary vein and trachea as the lungs transition from a pink color to opaque white. Swelling is also observed during SDS perfusion but resolves throughout the washing phase.
Gilpin et al.
Perfusion Decellularization of Lungs
303
Figure 4 Analysis of the decellularized lung matrix. (A) Bacterial and fungal growth after perfusate inoculation into lysogeny (L)-broth (culture for 4 days) indicates minimal contamination. (B) Sodium dodecyl sulfate (SDS) quantification in digested decellularized lung tissue by Stains-All assay confirms detergent removal after washing. The horizontal line in the middle of each box indicates the median; the top and bottom borders of the box mark the 75th and 25th percentiles, respectively; and the whiskers mark the 5th and 95th percentiles. (C) Residual double-stranded DNA quantification in digested decellularized lung tissue by PicoGreen assay confirms effective decellularization for human (HL 1–3) and porcine (PL 5–7) lungs compared with cadaveric lung tissue (HL CAD 4–5; n ¼ 7–10 tissue pieces per lung). One-way analysis of variance with Dunnett’s post-test. The error bars indicate the standard deviation.
Gross visual examination showed the scaffold retained pleura integrity, lobar definition, and major vasculature and airways. Sequential samples of the perfusion fluid were tested for bacterial and fungal contamination, confirming that no significant growth was detected after 4 days of culture in L-broth (Figure 4A). Analysis of several samples across the resulting lung scaffold confirmed removal of SDS in the lung tissue at the end of the protocol, with most of the analyzed samples reading undetectable levels of SDS (Figure 4B). The level of measurable dsDNA remaining within the decellularized tissue was significantly reduced compared with cadaveric tissue (Figure 4C). Preservation of decellularized tissue architecture and loss of cells was confirmed by hematoxylin and eosin staining (Figure 5A). Conservation of elastin and GAG components was also confirmed by Alcian blue and Verhoeff-Van Gieson staining, respectively (Figure 5A). Soluble collagen, GAGs, and elastin were measured throughout the decellularized lung tissue in a homogenous manner (Figure 5B–D). An overall loss of soluble collagen and elastin was quantified in the decellularized lung tissue when compared with cadaveric tissues. An enrichment of GAGs was also measured in the decellularized lung, relative to cadaveric tissue. Immunohistochemistry for collagen IV, fibronectin, and laminin further demonstrated the conservation of these ECM components in the decellularized human lung tissue (Figure 5E). Transmission electron microscopy (original magnification 3,000–10,000) showed the basement membrane was also intact (Figure 5F). Although each of these techniques only examines a component of the entire
matrix, taking these data points together, we feel comfortable stating that the ECM scaffolds were well preserved. Further investigation of the matrix by LC-MS/MS revealed additional peptide components were retained in the decellularized lung. We identified 70 peptides in one lung and 75 peptides in a second, with 37 proteins overlapping both samples. High levels of collagen, laminin, and other matrix components were identified in a similar pattern to that identified in SDS-decellularized rat lungs. To a lesser degree, other secreted proteins were detected within the decellularized scaffold (Figure 6). No intracellular peptides were identified within the samples analyzed. To investigate the biocompatibility of the decellularized rat and human lung tissue, a sterile tissue slice culture procedure was developed in conjunction with live microscopy imaging. Adult-derived SAECs, PAECs, or HUVECs cultured on top of decellularized rat or human lung matrix slices confirmed the biocompatibility of the decellularized matrix (bright-field images, Figure 7A, i). After 5 days of standard culture, cell viability was confirmed across cell types, with cells observed adherent to the matrix slices as well as on the surrounding tissue culture plastic (Figure 7A, ii). Images at higher magnification showed interaction of the cells with matrix, demonstrating attachment and elongation (Figure 7A, iii). Staining for integrins α2β1 and α3β1, which recognize collagen I/IV and laminin 5/10/11, respectively,18 demonstrated a specific interaction between the cells and matrix components in the tissue slice cultures (Figure 7A iv and v). Further quantification of cell growth and survival in the context of decellularized human lung slices was achieved by MTT assay after 5 days of culture.
304
The Journal of Heart and Lung Transplantation, Vol 33, No 3, March 2014
Figure 5 Matrix analysis of porcine and human decellularized lung tissue. (A) (i) Loss of cellular components in decellularized human and porcine lung tissue, as visualized by hematoxylin and eosin (H&E). (ii) Preservation of glycosaminoglycans (blue) by Alcian blue, and (iii) elastin (blue-black) by Verhoeff-Van Gieson across species. Main scale bar ¼ 200 μm, inset scale bar ¼ 20 μm. Quantification of (B) soluble collagen, (C) glycosaminoglycan, and (D) elastin in decellularized porcine (PL 5–7) and human (HL 1–3) lung tissue compared with cadaveric tissue (PL and HL CAD; n ¼ 7–10 tissue pieces per lung). Quantitative values all normalized to dry starting tissue weight. One-way analysis of variance with Dunnett’s post-test. (E) Immunohistochemistry staining of decellularized human lung tissue for collagen IV, fibronectin, and laminin. Scale bar ¼ 200 μm (upper) and 20 mm (lower). (F) Transmission electron microscopy demonstrating intact basement membrane in decellularized human lung tissue. (i and ii) Original magnification 3,000, 10-μm scale bar; (iii) original magnification 8,000, 2-μm scale bar; (iv) original magnification 10,000, 2-μm scale bar. The error bars show the standard deviation. *p o 0.05; ***p o 0.001
No difference in cell growth or survival was found between standard tissue culture and culture on the lung matrix, indicating biocompatibility of the decellularized matrix (Figure 7B–D). The decellularized lungs were further validated for their potential to serve as scaffolds for whole organ recellularization. The integrity of the lung matrix was verified by applying positive-pressure inflation by using a modified piston pump (Harvard Apparatus, Holliston, MA) at a stepwise increasing volume of 20 cm3/stroke and 10 strokes/ min until visual inflation without air leakage was observed (Figure 8A). Successful application of cyclic ventilation was achieved at volumes ranging from 15 to 100 cm3/stroke. Decellularized lungs were compliant with inspiration,
whereas elastic recoil was sufficient to enable passive expiration. Biomimetic culture of a decellularized lung lobe further confirmed the capability of the scaffold to serve as a platform for regeneration. After delivery of 500 106 human PAECs to the airways by gravity, the lungs were successfully cultured with media perfusion under constant pressure for 96 hours (Figure 8B). The integrity of the tissue was maintained throughout biomimetic culture, with no tissue damage and no contamination. Continuity of the vascular network was visualized by passive drainage of media from the pulmonary vein. Analysis of the tissue for cell distribution showed cell retention and appearance of flattened epithelial phenotypes consistent with cell attachment
Gilpin et al.
Perfusion Decellularization of Lungs
305
Figure 6 Peptide analysis of decellularized human lung matrix by liquid chromatography tandem mass spectrometry. Total spectral counts, representing a semi-quantitative measure of peptide abundance within the tissue, indicate a preservation of (A) collagen, (B) laminin, and (C) various other extracellular matrix components. BM, basement membrane; EMLIN, elastin microfibril interface located protein. (D) Other secreted, extracellular proteins were retained within the matrix. The error bars show the standard deviation.
and engraftment onto the basement membrane. As expected, the retention pattern was heterogeneous given limited cell number and short duration of culture (Figure 8C). Taken together, these results validate our protocol for the generation of clinical scale acellular lung scaffolds from porcine and human donors and provide initial proof of potential for matrix recellularization.
Discussion More than 1,700 Americans are in need of a donor lung,1 with limited or no available alternative therapies.19 Waiting list mortality rates for lung candidates continue to rise, reported at 15.7/100 waiting list-years in 2011.20 In addition, only 21.0% of potential donor lungs are used for transplantation,20 further highlighting the shortage of adequate organs. A 50% probability of graft failure and more than 40% probability of death at 5 years after transplant remain clinical realities lung transplant recipients continue to face.20 These facts underscore the theoretical clinical utility of a recipient-specific, regenerated organ. Our results demonstrate that porcine and human lungs not suitable for transplantation can be decellularized to generate whole-organ acellular scaffolds as a platform for organ regeneration. The procedure we have described in this report uses a simple, inexpensive detergent and a constant-pressure perfusion system to generate an acellular lung matrix that retains robust extracellular components and architecture. We aimed to use the rat model to compare decellularization detergents and inform the development of a large-scale protocol. Because the comparison was not repeated at the porcine or human level, the optimal perfusate cannot be directly determined, although the approach we have reported proved successful in our hands. When moving our promising small-animal data toward clinical translation, we faced the challenge of up-scaling this technology to generate a consistent, reproducible lung
scaffold. One question we considered was the source of donor lungs, which we aimed to address by applying our technology to human and porcine sources. We did not identify any differences when comparing porcine and human lungs in response to detergent or resulting matrix components. The applicability of our procedure to both species is promising because sourcing porcine lungs may be advantageous with regards to product development and regulatory approval. Porcine lungs could potentially be used as a whole-organ xenograft because porcine biologics are already used in clinical application, such as matrix derived from small-intestinal sub-mucosa for wound care21 and pericardial reconstruction.22 One question not yet addressed is the potential immune response elicited from xenotransplantation and any surface antigens that may remain within the ECM.23 If looking to human lung donors, the concept of human-derived transplantable tissue is supported by products such as biomesh derived from human skin.24 To broadly use human donor lungs, the variability of the donor pool must be addressed, especially amongst an extended criteria pool. The underlying donor pathology imparts potential differences to the native matrix, which may result in altered regenerative capacity of the scaffold.25,26 Our current results show that reproducible ECM scaffolds can be achieved with a variety of lung donors within the donor criteria we tested. We can report that decellularization failed to occur in 1 lung donor with a history of asthma and heavy marijuana use, perhaps due to underlying matrix or vascular damage. This lung showed signs of hyperinflation and parenchymal damage on macroscopic inspection and could not be salvaged as a scaffold. The lung is a particularly challenging donor organ due to its constant contact with the external environment and increased potential for infection or contamination. Lungs are also highly susceptible to injuries after donor brain death, including pulmonary edema, pneumonia, aspiration, or
306
The Journal of Heart and Lung Transplantation, Vol 33, No 3, March 2014
Figure 7 In vitro cell culture on decellularized lung matrix. (A) Live imaging of cells cultured on human lung matrix slices (50 μm). (i) Bright field images and (ii-iii) after treatment with viability dye (green) and Hoechst nuclear dye (blue) on Day 5. Robust cell viability is confirmed for small airway epithelial cells (SAECs), human umbilical vein endothelial cells (HUVECs), and pulmonary alveolar epithelial cells (PAECs) cultured on decellularized human slices. Higher magnification images highlight the interaction of viable cells with the matrix, demonstrating cell elongation and attachment to the matrix. (ii) Scale bar ¼ 100 μm; (iii) scale bar ¼ 50 μm. Integrin expression (red) and interaction with decellularized matrix (green) is confirmed for (iv) α2β1 and (v) α3β1. Scale bars ¼ 50 μm. (B–D) Cells cultured on human lung slices for 5 days survive and proliferate, as quantified by MTT assay, with no overall cytotoxicity associated with matrix-cell interaction. The error bars show the standard deviation.
ventilator-induced lung injury.27 These complications may alter their utility as source organs for decellularization. Similarly, the porcine airway is at risk for infection with variety of viral and bacterial pathogens,28 and more targeted treatment may be required to ensure downstream sterility of the decellularized organ. We have found that treatment with 0.08% peracetic acid solution is effective in reducing the bioburden in tissue before and after decellularization, as has been previously reported.29 Ethylene oxide has been used for the sterilization
of urinary bladder matrix, with minimal effects on structural properties compared with gamma or electron beam radiation.30 Even a moderate gamma dose (7–12 kGy) has been reported to cause human dermis ECM denaturation,31 emphasizing the need for optimized sterilization protocols to preserve essential matrix components. If lung scaffolds were to be produced for off-the-shelf use, protocols enabling the preservation and storage of these matrices will also need to be developed32,33 because these procedures may affect the integrity and utility of the scaffold.30,31,34
Gilpin et al.
Perfusion Decellularization of Lungs
Figure 8 Ventilation and biomimetic culture of human decellularized lungs. (A) (i) Fully deflated and (ii) maximal positive inflation of human decellularized lungs. (B) Upper right decellularized human lobe cannulated and mounted in bioreactor. (C) Hematoxylin and eosin staining of tissue after pulmonary alveolar epithelial cells delivery and 96 hours of biomimetic culture. Scale bar ¼ 100 μm. (D) TUNEL (terminal deoxynucleotide transferase-mediated deoxy uridine triphosphate nick-end labeling) fluorometric assay for apoptotic cells demonstrates minimal cell death (green) in recellularized lung tissue both proximal and distal to the perfusion cannula. Blue ¼ 4',6-diamidino-2phenylindole (DAPI). Scale bar ¼ 50 μm.
307 As the next step toward creation of viable tissue, we must address the challenges of recellularization. This will require the replacement of many different lung cell phenotypes, including airway and alveolar epithelium along with endothelium. Our results suggest that mature cell phenotypes from these 3 lineages can attach and survive in culture with decellularized lung matrix. The tissue slice model is a useful tool to test possible cell types in terms of cell-matrix attachment, proliferation, and overall biocompatibility in a high-throughput, 2-dimensional experimental design13,36–38 and can be applied to various cell types and matrices. Ultimately, the next steps toward clinical translation will be the generation of 3-dimensional, mature, functional tissue. This progress will require moving from the tissue culture dish to a whole-organ system and the design of appropriate bioreactors to recapitulate the body’s natural environment and physiology.5,6 Although we were able to maintain tissue integrity and cell retention for 96 hours of biomimetic culture, further refinements will be required to optimize ventilation and perfusion strategies to create an ideal environment for cell attachment, proliferation, and full tissue regeneration. Current advances in pulmonary stem cell differentiation have contributed to the ultimate goal of functional organ regeneration from patient-specific, pluripotent cells.37,39 The elucidation of specific milestones in lung development allows for directed differentiation of stem cells before lung reseeding and stage-specific investigation of lung progenitor populations in terms of attachment, proliferation, and further differentiation capacity. Once seeded, the native decellularized matrix may act through biologic signaling to influence stem cell differentiation pathways,40,41 further assisting in the regeneration process. Our proteomic analysis of the scaffold demonstrates the retention of critical basement membrane components, including collagen IV and laminin, which are required for epithelial cell attachment. To achieve the goal of truly patient-specific organs regenerated with autologous cells, identification of the optimal developmental stage of seeded cells and adaptation of subsequent organ culture protocols will be critical. Ultimately, the generation of human-scale lung scaffolds brings the goal of organ regeneration one step closer to clinical application. The field must now tackle the larger challenges of recellularization and restoring organ function to ultimately create transplantable organs for clinical use.
Disclosure statement Functional regeneration of lung scaffolds for transplantation will require the conservation of scaffold mechanics after decellularization. Our results demonstrated an average of greater than 60% conservation of elastic fibers in porcine scaffolds and greater than 45% conservation in human scaffolds; however, we have yet to develop a standardized battery of tests to ensure consistency throughout the decellularization process. Real-time testing of isolated lung tissue, including mechanical stress and strain evaluation, may provide a tool for continuous scaffold assessment.35
This study was supported by the United Therapeutics Corporation and the National Institutes of Health (NIH) Director’s New Innovator Award (DP2-OD008749-01). This work was presented at the Thirty-third Annual Meeting and Scientific Sessions of the International Society for Heart and Lung Transplantation, Montreal, Québec, Canada, April 24–27, 2013. HC Ott is founder and stockholder of IVIVA Medical Inc. This relationship did not influence design, execution, and/or interpretation of the present study. None of the authors has a financial relationship with a commercial entity that has an interest in the subject of the presented manuscript or other conflicts of interest to disclose.
308
The Journal of Heart and Lung Transplantation, Vol 33, No 3, March 2014
Supplementary data Supplementary data are available in the online version of this article at jhltonline.org.
References 1. Health Resources and Services Administration. Organ Procurement and Transplantation Network O. National Data Reports, as of April 5, 2013. Available from: http://optn.transplant.hrsa.gov/. Accessed December 4, 2013. 2. Soto-Gutierrez A, Wertheim JA, Ott HC, Gilbert TW. Perspectives on whole-organ assembly: moving toward transplantation on demand. J Clin Invest 2012;122:3817-23. 3. Badylak SF, Weiss DJ, Caplan A, Macchiarini P. Engineered whole organs and complex tissues. Lancet 2012;379:943-52. 4. Ott HC, Clippinger B, Conrad C, et al. Regeneration and orthotopic transplantation of a bioartificial lung. Nat Med 2010;16:927-33. 5. Petersen TH, Calle EA, Zhao L, et al. Tissue-engineered lungs for in vivo implantation. Science 2010;329:538-41. 6. Price AP, England KA, Matson AM, Blazar BR, Panoskaltsis-Mortari A. Development of a decellularized lung bioreactor system for bioengineering the lung: the matrix reloaded. Tissue Eng Part A 2010;16:2581-91. 7. Wallis JM, Borg ZD, Daly AB, et al. Comparative assessment of detergent-based protocols for mouse lung de-cellularization and re-cellularization. Tissue Eng Part C Methods 2012;18:420-32. 8. Ott HC, Matthiesen TS, Goh SK, et al. Perfusion-decellularized matrix: using nature’s platform to engineer a bioartificial heart. Nat Med 2008;14:213-21. 9. Crapo PM, Gilbert TW, Badylak SF. An overview of tissue and whole organ decellularization processes. Biomaterials 2011;32:3233-43. 10. Macchiarini P, Jungebluth P, Go T, et al. Clinical transplantation of a tissue-engineered airway. Lancet 2008;372:2023-30. 11. Sullivan DC, Mirmalek-Sani SH, Deegan DB, et al. Decellularization methods of porcine kidneys for whole organ engineering using a highthroughput system. Biomaterials 2012;33:7756-64. 12. Remlinger NT, Wearden PD, Gilbert TW. Procedure for decellularization of porcine heart by retrograde coronary perfusion. J Vis Exp 2012:e50059. 13. Daly AB, Wallis JM, Borg ZD, et al. Initial binding and recellularization of decellularized mouse lung scaffolds with bone marrow-derived mesenchymal stromal cells. Tissue Eng Part A 2012;18:1-16. 14. Jensen T, Roszell B, Zang F, et al. A rapid lung de-cellularization protocol supports embryonic stem cell differentiation in vitro and following implantation. Tissue Eng Part C Methods 2012;18:632-46. 15. Song JJ, Kim SS, Liu Z, et al. Enhanced in vivo function of bioartificial lungs in rats. Ann Thorac Surg 2011;92:998-1005; discussion 1006. 16. Rusconi F, Valton E, Nguyen R, Dufourc E. Quantification of sodium dodecyl sulfate in microliter-volume biochemical samples by visible light spectroscopy. Anal Biochem 2001;295:31-7. 17. Ngoka LC. Sample prep for proteomics of breast cancer: proteomics and gene ontology reveal dramatic differences in protein solubilization preferences of radioimmunoprecipitation assay and urea lysis buffers. Proteome Sci 2008;6:30. 18. Sheppard D. Functions of pulmonary epithelial integrins: from development to disease. Physiol Rev 2003;83:673-86. 19. Bousquet J, Kaltaev N. Global surveillance, prevention and control of chronic respiratory diseases: a comprehensive approach. Geneva: World Health Organization; 2007. p. vii, 146 p. 20. Scientific Registry of Transplant Recipients S. OPTN and SRTR annual data report, 2011. Available from: http://srtr.org/. Accessed April 5, 2013. 21. Mostow EN, Haraway GD, Dalsing M, Hodde JP, King D. Effectiveness of an extracellular matrix graft (OASIS Wound Matrix) in the treatment of chronic leg ulcers: a randomized clinical trial. J Vasc Surg 2005;41:837-43.
22. Boyd WD, Johnson WE 3rd, Sultan PK, Deering TF, Matheny RG. Pericardial reconstruction using an extracellular matrix implant correlates with reduced risk of postoperative atrial fibrillation in coronary artery bypass surgery patients. Heart Surg Forum 2010; 13:E311-6. 23. Cooper DK, Ekser B, Burlak C, et al. Clinical lung xenotransplantation —what donor genetic modifications may be necessary? Xenotransplantation 2012;19:144-58. 24. Jansen LA, De Caigny P, Guay NA, Lineaweaver WC, Shokrollahi K. The evidence base for the acellular dermal matrix AlloDerm: a systematic review. Ann Plast Surg 2013;70:587-94. 25. Booth AJ, Hadley R, Cornett AM, et al. Acellular normal and fibrotic human lung matrices as a culture system for in vitro investigation. Am J Respir Crit Care Med 2012;186:866-76. 26. Sokocevic D, Bonenfant NR, Wagner DE, et al. The effect of age and emphysematous and fibrotic injury on the re-cellularization of decellularized lungs. Biomaterials 2013;34:3256-69. 27. Avlonitis VS, Fisher AJ, Kirby JA, Dark JH. Pulmonary transplantation: the role of brain death in donor lung injury. Transplantation 2003;75:1928-33. 28. Opriessnig T, Gimenez-Lirola LG, Halbur PG. Polymicrobial respiratory disease in pigs. Anim Health Res Rev 2011;12:133-48. 29. Kemp PD. Peracetic acid sterilization of collagen or collagenous tissue. In: Office UP, editor. USA: Organogenesis Inc.; 1995. 30. Freytes DO, Stoner RM, Badylak SF. Uniaxial and biaxial properties of terminally sterilized porcine urinary bladder matrix scaffolds. J Biomed Mater Res B Appl Biomater 2008;84:408-14. 31. Sun WQ, Leung P. Calorimetric study of extracellular tissue matrix degradation and instability after gamma irradiation. Acta Biomater 2008;4:817-26. 32. Nonaka PN, Campillo N, Uriarte JJ, et al. Effects of freezing/thawing on the mechanical properties of decellularized lungs. J Biomed Mater Res A 2013, http://dx.doi.org/10.1002/jbm.a.34708: [E-pub ahead of print]. 33. Bonenfant NR, Sokocevic D, Wagner DE, et al. The effects of storage and sterilization on de-cellularized and re-cellularized whole lung. Biomaterials 2013;34:3231-45. 34. Dijkman PE, Driessen-Mol A, Frese L, Hoerstrup SP, Baaijens FP. Decellularized homologous tissue-engineered heart valves as off-theshelf alternatives to xeno- and homografts. Biomaterials 2012; 33:4545-54. 35. Petersen TH, Calle EA, Colehour MB, Niklason LE. Matrix composition and mechanics of decellularized lung scaffolds. Cells Tissues Organs 2012;195:222-31. 36. Bonvillain RW, Danchuk S, Sullivan DE, et al. A nonhuman primate model of lung regeneration: detergent-mediated decellularization and initial in vitro recellularization with mesenchymal stem cells. Tissue Eng Part A 2012;18:2437-52. 37. Longmire TA, Ikonomou L, Hawkins F, et al. Efficient derivation of purified lung and thyroid progenitors from embryonic stem cells. Cell Stem Cell 2012;10:398-411. 38. Shamis Y, Hasson E, Soroker A, et al. Organ-specific scaffolds for in vitro expansion, differentiation and organization of primary lung cells. Tissue Eng Part C Methods 2011;17:861-70. 39. Mou H, Zhao R, Sherwood R, et al. Generation of multipotent lung and airway progenitors from mouse ESCs and patient-specific cystic fibrosis iPSCs. Cell Stem Cell 2012;10:385-97. 40. Hosokawa T, Betsuyaku T, Nishimura M, Furuyama A, Katagiri K, Mochitate K. Differentiation of tracheal basal cells to ciliated cells and tissue reconstruction on the synthesized basement membrane substratum in vitro. Connect Tissue Res 2007;48:9-18. 41. Gong J, Sagiv O, Cai H, Tsang SH, Del Priore LV. Effects of extracellular matrix and neighboring cells on induction of human embryonic stem cells into retinal or retinal pigment epithelial progenitors. Exp Eye Res 2008;86:957-65.
