Advertisement

Qualitative and Quantitative Evaluation of a Novel Detergent-Based Method for Decellularization of Peripheral Nerves

  • Charlot Philips
  • Fernando Campos
  • Annelies Roosens
  • María del Carmen Sánchez-Quevedo
  • Heidi Declercq
  • Víctor Carriel
Article
  • 118 Downloads

Abstract

Tissue engineering is an emerging strategy for the development of nerve substitutes for peripheral nerve repair. Especially decellularized peripheral nerve allografts are interesting alternatives to replace the gold standard autografts. In this study, a novel decellularization protocol was qualitatively and quantitatively evaluated by histological, biochemical, ultrastructural and mechanical methods and compared to the protocol described by Sondell et al. and a modified version of the protocol described by Hudson et al. Decellularization by the method described by Sondell et al. resulted in a reduction of the cell content, but was accompanied by a loss of essential extracellular matrix (ECM) molecules such as laminin and glycosaminoglycans. This decellularization also caused disruption of the endoneurial tubes and an increased stiffness of the nerves. Decellularization by the adapted method of Hudson et al. did not alter the ECM composition of the nerves, but an efficient cell removal could not be obtained. Finally, decellularization by the method developed in our lab by Roosens et al. led to a successful removal of nuclear material, while maintaining the nerve ultrastructure and ECM composition. In addition, the resulting ECM scaffold was found to be cytocompatible, allowing attachment and proliferation of adipose-derived stem cells. These results show that our decellularization combining Triton X-100, DNase, RNase and trypsin created a promising scaffold for peripheral nerve regeneration.

Keywords

Peripheral nerve regeneration Decellularized nerve allografts Tissue engineering Detergents Extracellular matrix Biomechanics 

Introduction

The peripheral nervous system (PNS) forms a bridge between most tissues and organs and the central nervous system. Because of their high distribution throughout the body, peripheral nerves (PNs) are very vulnerable. Vehicle accidents, gunshot wounds, sports injuries, etc. can all lead to PN injuries, resulting in a prevalence of 2.8% in all trauma cases.34 It is of major importance to adequately treat these injuries to avoid lifelong discomfort and disability of patients.

Although the PNS has the capacity to regenerate spontaneously, full recovery is only observed in rare cases. Surgical intervention is mostly necessary and the preferred method for repair is end-to-end neurorrhaphy or direct repair of the damaged nerve stumps.17,39 However, when injuries are more severe and tensionless repair is no longer possible, the gold standard treatment is the use of a PN autograft. These grafts have the advantage of containing viable Schwann cells (SCs) and a rich extracellular matrix (ECM), providing a suitable microenvironment for PN regeneration. Despite these advantages, PN autografts are only effective in 50% of the cases and disadvantages such as donor site morbidity, possible neuroma formation and limited availability compromise their usefulness.6,18,39

Tissue engineering strategies are becoming more and more interesting to offer an alternative to PN autografts. Nerve conduits (NCs) have been developed to aid in axon guidance and to promote PN regeneration.10,13,23 Both synthetic and natural biomaterials can be used for their design and several NCs already gained approval of the US Food and Drug Administration and Conformit Europe.28,30 These off-the-shelf available nerve substitutes have already been shown to perform as good as PN autografts in short defects, but often fail over longer distances (> 30 mm).31 The lack of appropriate topographical cues results in inefficient regeneration and eventually in loss of function of the target organ.

To overcome the limitations of NCs, an attractive alternative are decellularized PN allografts (DPNAs). The natural nerve architecture and ECM components are retained after decellularization and provide the ideal substrate for newly formed axons.27 In addition, by removing all cellular components, the need for immunosuppression normally associated with allografts can be circumvented. Another advantage is that the size and type of nerve (motor, sensory or mixed) can be adapted to the required needs.

Several approaches have already been described to obtain DPNAs, such as cold preservation,20,24 freeze-thawing22,43 and detergent-based decellularization.25,40,47 Of all reported methods, the latter seems to be the most efficient. One of the first detergent-based decellularization protocols used a combination of Triton X-100 and sodium deoxycholate (SDC).40 This method achieved complete decellularization of the nervous tissue, but was found to be a harsh method in terms of structural preservation. Later on, an optimized decellularization protocol combining sulfobetaine-10 (SB-10), sulfobetaine-16 (SB-16) and Triton X-200 was described.25 This formed the basis for further development of the only commercially available DPNA, the Avance® Nerve Graft of AxoGen, Inc. This commercial graft was shown to improve regeneration in a rat model when compared to silicone conduits32 or the collagen-based NeuraGen® conduit45 in both 14- and 28-mm sciatic nerve defects. Early clinical data also hold promising results for longer gaps up to 50 mm in humans.1 Nevertheless, the Avance® Nerve Graft is only licensed for the American market, making its use in Europe difficult. Regulations concerning donor selection and tissue processing differ between both authorities and therefore, importation is not evident.

Many alternative decellularization protocols have already been proposed in literature, but none of them reached to the clinical stage. Comparison between protocols is not obvious, since many different sizes and types of nerves have already been used for the investigation of these protocols. In addition, no consensus exists on which evaluation methods should be included and therefore, a thorough in vitro characterization is often lacking. In the present study, our main goal was to evaluate our newly developed decellularization protocol in both a qualitative and quantitative manner. Furthermore, we compared our method to two other detergent-based decellularization protocols. By evaluating all decellularization protocols under the same conditions and by including histological as well as ultrastructural, biochemical and mechanical analysis, a clear overview of the impact of the different decellularization protocols on the cellular content and architecture of the nerves is obtained. This will allow for better selection of a superior decellularization protocol for further in vivo follow-up studies.

Materials and Methods

Nerve Isolation and Experimental Groups

Forty-five male Wistar rats weighing 200–250 g were provided by the Experimental Unit of the University Hospital Virgen de las Nieves in Granada (Spain). The animals were housed in a temperature and light controlled environment (21 ± 1 °C and 12 h light/dark cycle, respectively) with ad libitum access to tap water and food. All animals were euthanized by anesthesia overdose and these procedures were conducted according to the European Union and Spanish Government guidelines for ethical care of animals (EU Directive No. 63/2010, RD 53/2013). Furthermore, this project was approved by the local ethical committee of Granada (FIS PI14/01343 and FIS PI17/0393).

Segments of approximately 30 mm of both sciatic nerves were harvested from euthanized animals and cryopreserved at − 80 °C in 10% DMSO in fetal bovine serum until decellularization. For the decellularization experiments, nerves were thawed at room temperature, sectioned into 10-mm segments and randomly assigned to the following experimental groups: (1) the Sondell group, containing nerves decellularized by the method described by Sondell et al.40; (2) the Hudson-based group, containing nerves decellularized by an adaptation of the method described by Hudson et al.25 omitting the detergent Triton X-200 due to unavailability and reported negative impact on laminin activity33; (3) the Roosens group, containing nerves decellularized by a novel protocol recently developed in our lab and which was already successful for decellularization of heart valve leaflets37 and (4) the native group, containing native sciatic nerves as a control for all analyses. In each group, a total of 50 10-mm segments was used.

Decellularization of Sciatic Nerves

Nerves of the Sondell group were first placed in distilled water for 7 h, then transferred to 3% Triton X-100 (Sigma-Aldrich; T8787) in distilled water for overnight incubation and subsequently immersed in 4% SDC (Sigma-Aldrich; D6750) in distilled water for 24 h, all at room temperature (RT) and under constant agitation (24 rpm; 20° tilt angel; VWR® Nutating Mixer). These steps were repeated and after a final wash with distilled water at RT, the nerves were kept in phosphate buffered saline (PBS) at 4 °C until further analyses.

Nerves of the Hudson-based group were also placed in distilled water first for 7 h at RT. The solution was then replaced with 125 mM SB-10 (Sigma-Aldrich; D4266) in Extraction Buffer (10 mM Na phosphate buffer with 150 mM NaCl) for 15 h at RT and under constant agitation. After rinsing with Rinse Buffer (10 mM Na phosphate buffer) at RT, the nerves were transferred to 0.6 mM SB-16 (Sigma-Aldrich; H6883) in Extraction buffer for 24 h at RT under constant agitation. These steps were repeated with incubation in SB-10 and SB-16 for 7 and 15 h respectively. After a final wash with Rinse buffer, the nerves were kept in PBS at 4 °C until further analyses.

Decellularization of the nerves of the Roosens group started with overnight incubation at 4 °C in 50 mM Tris buffer (pH = 8). The solution was then replaced with 1% Triton-X100 in 50 mM Tris buffer for 24 h at 4 °C under constant agitation. After rinsing with Hank’s Balanced Salt Solution (HBSS), the nerves were subjected to an enzymatic treatment with 40,000 units/L DNase (Roche; Cat. No. 04536282001), 20 mg/L RNase (Sigma-Aldrich; R4875) and 100 mg/L trypsin (Sigma-Aldrich; T9201) in HBSS at 37 °C for 2 × 45 min under constant agitation. Subsequently, nerves were agitated again overnight in 1% Triton-X100 in 50 mM Tris buffer at 4 °C. After several washes in HBSS, the nerves were stored in HBSS at 4 °C until further analyses.

Transparency Analysis

Macroscopic images of native and decellularized nerves were used to determine the level of transparency. For each group, 4 nerves were placed on a black and white background and images were further processed with Image J. The intensity was measured on 5 different spots in each nerve for both the black and white area as previously described.7 Average values were calculated and the difference between the black and white area of the nerve was expressed as a percentage of the maximal difference between the black and white background. In this way, low percentages correlate to non-transparent nerves, while high percentages indicate high levels of nerve transparency where the black background is visible through the nerve.

Histological Analysis

Native and decellularized nerves (n = 5 for each group) were fixed in 10% formalin, embedded in paraffin and cut into 5 µm thick sections. Routine staining with haematoxylin–eosin (HE) was performed to evaluate the general morphology and to determine the presence of cell nuclei. In addition, fluorescent staining with 4’,6-diamidino-2-phenylindole (DAPI) was carried out for identification of any nuclear remnants. Several histochemical methods were used to analyze the ECM of the nerves. Picrosirius Red, Alcian Blue, Orcein and the MCOLL method were used as previously described to visualize collagen fibers, acid proteoglycans, elastic fibers and myelin respectively.8,11,35

For immunohistochemistry, sections were pretreated with citrate buffer (pH = 6) at 95 °C and/or pepsin at 37 °C for antigen retrieval. Next, sections were incubated with 3% (v/v) H2O2 in 0.1 M PBS for 10 min to block endogenous peroxidase activity. Nonspecific staining was further blocked with Casein solution (Vector® Laboratories) for 15 min. Primary and secondary antibodies were used according to previously standardized protocols10,12,13 and technical information is summarized in Table 1. Subsequently, the peroxidase reaction was visualized with the ImmPACT™ DAB Peroxidase Substrate Kit (Vector® Laboratories; SK-4100). After counterstaining with Harris haematoxylin, slides were dehydrated, cleared and mounted and visualized with a Nikon Eclipse 90i light microscope.
Table 1

Antibodies used for immunohistochemical analysis.

Antibody

Dilution

Pretreatment

Incubation time

Cat. no.

Rabbit polyclonal anti-S100

1:400

Citrate buffer, pH 6, 95 °C, 25 min

30 min at RT

Dako; Z0311

Mouse monoclonal anti-vimentin

1:200

Citrate buffer, pH 6, 95 °C, 25 min

60 min at RT

Sigma-Aldrich; V6630

Mouse monoclonal anti-neurofilament

1:500

Citrate buffer, pH 6, 95 °C, 25 min

60 min at RT

Sigma-Aldrich; N2912

Rabbit polyclonal anti-collagen I

1:500

Pepsin, 37 °C for 10 min

90 min at RT

Acris; R1038

Mouse monoclonal anti-laminin

1:1000

Citrate buffer, pH 6, 95 °C, 25 min

Pepsin, 37 °C for 5 min

Overnight at 4 °C

Sigma-Aldrich; L8271

ImmPRESS™ anti-rabbit IgG (peroxidase)

Ready-to-use

30 min at RT

Vector® Laboratories; MP-7401

ImmPRESS™ anti-mouse IgG (peroxidase)

Ready-to-use

30 min at RT

Vector® Laboratories; MP-7402

Semiquantitative analysis of the histological images was performed as previously described.19,21 Three independent researchers scored the histological reactions as absent (−), weak reaction (+), moderate reaction (++) or strong reaction (+++).

Scanning and Transmission Electron Microscopy

To evaluate the ultrastructure of the native and decellularized nerves, samples (n = 3 for each group) were prepared for scanning electron microscopy (SEM) and transmission electron microscopy (TEM). After overnight fixation at 4 °C with 2.5% glutaraldehyde in 0.05 M cacodylate buffer (pH 7.2), samples were washed three times, post-fixed for 1 h at room temperature with 2% OsO4 and washed again with 0.05 M cacodylate buffer (pH 7.2).

Samples for SEM were further dehydrated with increasing alcohol series (50–75–85–96–100% ethanol) and finally, hexamethyldisilazane (Acros Organics; Cat. No. 120585000) was used to completely dry the samples. After coating with gold, samples were visualized with a Quanta 200 FEI environmental scanning electron microscope.

Samples for TEM were dehydrated with increasing aceton series and embedded in epoxy resin. Ultrathin Sections (60 nm) were made using a Leica ultracut UCT ultramicrotome (Leica Microsystems GmbH). Sections were mounted in 200 mesh copper grids (Gilder; G200), contrasted with uranyl acetate and lead citrate and evaluated with a JEOL 1200 EX II transmission electron microscope operating at 80 kV.

Quantitative Biochemical Analysis

Biochemical analysis was performed to determine the amount of DNA, glycosaminoglycans (GAGs) and collagen in both native and decellularized nerves. Total DNA was extracted from tissue samples using the QIAamp DNA Mini Kit (Qiagen) and further quantified with the Quant-iT PicoGreen dsDNA Assay Kit (Invitrogen). Fluorescence was measured at an excitation wavelength of 480 nm and emission wavelength of 520 nm with the Paradigm™ Detection Platform (Beckman Coulter). For quantification of sulfated GAGs, the Blyscan Sulfated Glycosaminoglycan Assay (Biocolor) was used according to the manufacturer’s protocol. Collagen content was determined by measuring the amount of hydroxyproline with the Hydroxyproline Assay Kit (Sigma-Aldrich). Since hydroxyproline is largely restricted to collagen, it serves as an indicator for the total amount of collagen. All biochemical measurements were normalized to tissue dry weight and 5 samples were used for each group.

Mechanical Testing

A tensile test was performed to measure the mechanical strength of the decellularized nerves compared to native tissue. For each group, 4 or 5 samples (~ 30 mm) were loaded on an electromechanical material testing instrument (Instron, Model 3345-K3327) orientated with their length along the direction of the tension. The samples were gripped at both ends, leaving a constant distance of 10 mm of nerve between both clamps as previously described for rat sciatic nerves.13 The tests were run at a constant strain rate of 10 mm/min at room temperature. Parameters such as the Young’s Modulus, stress at fracture and strain at fracture were obtained from a stress–strain curve (N/mm2) and serve as measures for elasticity and tensile strength.36

Ex Vivo Cytocompatibility

To assess the cytocompatibility of the novel decellularization protocol described here, a cryoculture assay was performed.5,38 Cryosections of 20 µm were cut from nerves of the Roosens group and were subsequently used for culturing of adipose-derived stem cells (ADSCs). The ADSCs were isolated from Wistar rats and expanded in DMEM Glutamax (Gibco™; Invitrogen) supplemented with 10% fetal calf serum and 1% penicillin/streptomycin in a humidified atmosphere at 37 °C and 5% CO2. ADSCs of passage 3 were seeded at a density of 10,000 cells/sample. After 2, 4 and 6 days, a live/dead assay, MTT assay and DNA quantification were carried out in fivefold. For each analysis, ADSCs cultured on glass coverslips served as positive control, while ADSCs cultured on glass coverslips and incubated with 2% Triton X-100 served as negative control.

For the live/dead assay, culture medium was replaced by 1 mL PBS containing 2 µL calcein-AM (Anaspec; 89201; 1 mg/mL) and 2 µL propidium iodide (Sigma-Aldrich; P4170; 1 mg/mL). After 10 min incubation at RT, the samples were rinsed with PBS and analyzed with an Olympus IX81 fluorescence microscope.

The metabolic activity of the ADSCs was evaluated with an MTT assay following the manufacturer’s recommendations (Calbiochem™; 475989-1GM). After rinsing with PBS, the culture medium was replaced by the MTT solution (5 mg MTT powder in 10 mL culture medium) and incubated for 4 h at 37 °C. The supernatant was then discarded and lysis buffer (50% ethanol and 50% DMSO) was added. After transferring 200 µL of the dissolved formazan solution to a 96-well plate, the absorbance was measured at 580 nm with a spectrophotometer (Universal microplate reader EL 800, Biotek Instruments).

The DNA released from the samples into the culture medium, which indicates irreversible cell-membrane damage, was quantified as a measure for the extent of damaged cells.3,4 The absorbance at 260 nm was measured with a UV–Vis Spectrophotometer (NanoDrop™ 1000; Thermo Scientific). Distilled water was used as a blank and the value of culture medium without cells was subtracted from the experimental data.

Statistical Analysis

For transparency, biochemical and mechanical analysis and ex vivo cytocompatibility, differences between groups were statistically analyzed with the commercially available software package SPSS Statistics 24 (SPSS GmbH Software). Normality of the data was assessed by using the Shapiro–Wilk test and homogeneity of variance was determined by means of a Levene’s test. The data was further analyzed with a one-way ANOVA test followed by a post hoc Tukey’s (equal variances) or with a Welch F test followed by a post hoc Dunett’s T3 (unequal variances). A value of p < 0.05 was considered statistically significant.

Results

Transparency Analysis

The impact of the different decellularization protocols on the nerves was first assessed macroscopically (Fig. 1a). Native nerves appeared bright white and were not transparent, whereas nerves of the Hudson-based and Roosens group were not as bright anymore. Nerves of the Sondell group, on the other hand, became more transparent and the black background was visible through the nerves. These observations are supported by the quantitative analysis of transparency (Fig. 1b). Native nerves were only 1.10 ± 2.30% transparent, while nerves of the Sondell, Hudson-based and Roosens group were 16.39 ± 5.42, 5.92 ± 2.27 and 5.30 ± 2.20% transparent, respectively.
Figure 1

Transparency analysis of native and decellularized nerves. (a) Macroscopic pictures of nerves representative for each group. Notice the black lines of the background visible through the nerves of the Sondell group. (b) Quantification of the transparency of native and decellularized nerves. *p < 0.05 compared to native nerves, #p < 0.05 compared to other decellularization groups. (n = 4).

Histological Analysis

Characterization of the native and decellularized nerves was carried out by means of histological analysis to assess the cellular content as well as the structural integrity. First of all, general morphology was evaluated with HE staining (Fig. 2a). Native nerves showed the typical organization of endoneurial structures dispersed with cell nuclei and an epineurium surrounding the fascicles. After decellularization, the epineurium was well preserved in all groups. Furthermore, in nerves of the Sondell group, no more visible cell nuclei could be detected, but the structure of the endoneurium seemed disrupted. Nerves of the Hudson-based group also showed a disorganized endoneurium and, in addition, there could still be cell nuclei detected. In contrast, nerves of the Roosens group seemed to be cell-free and showed good preservation of the endoneurium with cylindrical structures remaining.
Figure 2

Histological analysis of the cellular content of native and decellularized nerves. (a) HE staining reveals the presence of cell nuclei in nerves of the Hudson-based group, while these are absent in the other decellularization groups. (b) DNA remnants are also demonstrated with DAPI staining in nerves of the Hudson-based group. (c) S100 immunohistochemical staining shows an intense reaction in nerves of the Hudson-based group. (d) Vimentin is only weakly present in nerves of the Hudson-based group as demonstrated with immunohistochemical staining. (e) Remnants of axonal proteins can still be found in nerves of the Sondell and Hudson-based group as shown with immunohistochemical staining for neurofilament. (f) The MCOLL histochemical method demonstrates complete removal of myelin in all decellularization groups. Scale bar = 100 µm.

To further evaluate the cellular content after decellularization, DAPI staining (Fig. 2b) was carried out and similar results were obtained as with HE staining. In the Hudson-based group, many DNA remnants were revealed, while this was not the case for nerves of the Sondell or Roosens group. Immunohistochemical staining for S100 (Fig. 2c) and vimentin (Fig. 2d), which are considered markers for SCs and fibroblasts respectively, also confirmed these observations. Nerves of the Hudson-based group showed presence of SCs and some remnants of fibroblasts, while nerves of the Sondell and the Roosens group showed absence of either cell type. Furthermore, immunohistochemical staining for neurofilament (Fig. 2e) was carried out to visualize any remaining axonal material. Only nerves of the Roosens group showed complete removal of axons, while nerves of the Sondell and the Hudson-based group still contained traces of axonal proteins. Finally, the MCOLL histochemical method (Fig. 2f) revealed complete removal of myelin in nerves of all decellularization groups.

Preservation of important ECM components was evaluated with both histochemical and immunohistochemical methods. Picrosirius Red staining (Fig. 3a) showed the preservation of collagen in the epineurium, perineurium and endoneurium of nerves of all decellularization groups as also seen with the MCOLL method. This was further confirmed by the immunohistochemical staining for collagen type I, which was especially abundant in the endoneurial compartment (Fig. 3b). In addition, Alcian Blue staining (Fig. 3c) showed the presence of acid proteoglycans in all stromal layers of each experimental group, except in the epineurium of nerves of the Sondell group. Presence of elastin was evaluated with the Orcein staining (Fig. 3d). A reduced amount of elastin was detected after decellularization, irrespective of the protocol used. Finally, laminin was preserved in the basal lamina of PN fibers and blood vessels, but not in the perineurium of nerves of the Hudson-based and the Roosens group as shown with immunohistochemical staining (Fig. 3d). In contrast, in nerves of the Sondell group most of the laminin seemed to be lost.
Figure 3

Histological analysis of ECM components of native and decellularized nerves. (a and b) Preservation of collagen in all stromal layers of each experimental group is demonstrated by Picrosirius Red histochemical and collagen I immunohistochemical staining. (c) Alcian Blue staining shows that acid proteoglycans are well preserved in all decellularization groups. (d) Orcein staining demonstrates the presence of elastin in native nerves, but not in decellularized nerves. (e) Laminin can be found in all groups as shown with immunohistochemical staining, but nerves of the Sondell group have a decreased amount compared to native nerves. Scale bar = 100 µm.

Scanning and Transmission Electron Microscopy

The ultrastructure of the native and decellularized nerves was visualized with electron microscopy. SEM showed that both axons and myelin were efficiently removed in nerves of the Sondell group, leaving only the ECM (Fig. 4a). This ECM, however, seems to have been affected by the detergents, leading to a disrupted appearance. In nerves of the Hudson-based group, on the other hand, the ECM was well preserved with the presence of the typical endoneurial tubes. For nerves of the Roosens group, SEM revealed that not all myelin was removed after decellularization. TEM, on the other hand, was not evident for nerves of the Sondell group, since the remaining structural components provided little contrast to be visualized (Fig. 4b). In nerves of the Hudson-based and Roosens group, the detergents clearly affected the axons and myelin, but there is still debris remaining. Finally, Fig. 4c shows that the endoneurial collagen-rich ECM was relatively well preserved in all groups.
Figure 4

Ultrastructural analysis of the nerve architecture of native and decellularized nerves. (a) SEM of cross-sections of native and decellularized nerves; (b) TEM of axons in native and decellularized nerves; (c) Detail of the ECM between axons. The myelin sheath around the axons of native nerves is clearly visible and collagen can be distinguished between the axons. After decellularization, nerves of the Roosens group still contain a lot of myelin, in contrast to nerves of the Sondell and Hudson-based group. Collagen is preserved in all decellularization groups. Scale bar = 50 µm (a), 1 µm (b) and 200 nm (c).

Quantitative Biochemical Analysis

To further evaluate the impact of the different decellularization protocols on DNA, GAG and hydroxyproline (HYP) content in a quantitative way, biochemical analyses were performed as summarized in Table 2 and Fig. 5. DNA content (Fig. 5a) was significantly (p < 0.05) reduced in nerves of the Sondell and Roosens group compared to native nerves. Nerves of the Hudson-based group, on the other hand, did not differ significantly from native nerves and had a significantly (p < 0.01) higher DNA content than nerves of the Sondell and Roosens group. GAG content (Fig. 5b) in nerves of the Sondell group fell below the detection limit of the assay and was therefore set to zero, resulting in a significant (p < 0.01) lower value compared to all other groups. When comparing the Hudson-based and Roosens group to native nerves, no obvious differences could be detected. Finally, HYP content (Fig. 5c) was found to be significantly (p < 0.05) higher in nerves of the Sondell group compared to native nerves. Both in nerves of the Hudson-based and Roosens group, HYP content seemed to be slightly elevated compared to native nerves, but not in a significant manner.
Table 2

Biochemical quantification of DNA, GAG and HYP content of native and decellularized nerves.

 

DNA (ng/mg)

GAG (µg/mg)

HYP (µg/mg)

Mean

(SD)

Mean

(SD)

Mean

(SD)

Native

852.24

(302.28)

2.49

(0.28)

14.83

(1.55)

Sondell

195.85

(53.34)

0.00

(0.00)

21.64

(3.18)

Hudson-based

1257.77

(236.99)

2.66

(0.42)

17.34

(0.42)

Roosens

35.15

(7.11)

2.10

(0.47)

16.58

(1.02)

Figure 5

Quantitative biochemical analysis of native and decellularized nerves. (a) DNA content in nerves of the Sondell and Roosens group is significantly lower than in native nerves and nerves of the Hudson-based group. (b) GAG content is similar in all groups, except in nerves of the Sondell group, of which the GAG content fell below the detection limit of the assay. (c) HYP content is higher in all decellularization groups compared to native nerves, but this is only significant for nerves of the Sondell group. *p < 0.05 compared to native nerves, #p < 0.05 compared to other decellularization groups. (n = 5).

Mechanical Testing

To evaluate the mechanical properties of native and decellularized nerves, a tensile test was carried out. Young’s Modulus, stress at fracture and strain at fracture were calculated from the resulting stress–strain curves and are summarized in Table 3 and Fig. 6. Both the Young’s Modulus (Fig. 6a) and stress at fracture (Fig. 6b) were significantly (p < 0.01) higher in nerves of the Sondell and Hudson-based group compared to native nerves. Interestingly, nerves of the Roosens group, on the other hand, had similar overall values as native nerves. Additionally, the strain at fracture (Fig. 6c) was comparable between all groups.
Table 3

Mechanical analysis of native and decellularized nerves.

 

Young’s Modulus (MPa)

Stress at fracture (MPa)

Strain at fracture (mm/mm)

Mean

(SD)

Mean

(SD)

Mean

(SD)

Native

8.34

(1.13)

2.44

(0.26)

0.70

(0.38)

Sondell

22.07

(3.50)

4.82

(0.58)

0.68

(0.15)

Hudson-based

18.40

(1.52)

5.33

(0.39)

0.63

(0.07)

Roosens

8.48

(2.35)

2.65

(0.51)

0.65

(0.16)

Figure 6

Mechanical analysis of native and decellularized nerves. (a) The Young’s Modulus is significantly higher in nerves of the Sondell and Hudson-based group compared to native nerves and nerves of the Roosens group. (b) Stress at fracture is also significantly higher in nerves of the Sondell and Hudson-based group compared to native nerves and nerves of the Roosens group. (c) Strain at fracture is similar for all groups. *p < 0.05 compared to native nerves, #p < 0.05 compared to other decellularization groups. (n = 5).

Overall Impact of the Decellularization Protocols

To have a clear overview of the effect of the different decellularization methods, the most important findings are summarized in Table 4. Each of the combinations of decellularizing agents resulted in a different outcome in terms of cell removal, ECM preservation and mechanical integrity. The Sondell protocol seemed to be effective in removal of nuclear remnants and myelin, but did not remove all cytoplasmic remnants. In addition, several ECM components were lost and the mechanical integrity was affected. The Hudson-based protocol did not remove the nuclear and cytoplasmic content, but retained all ECM components, except for elastin. The mechanical integrity was also affected in a similar manner as for the Sondell protocol. The Roosens protocol seemed to efficiently remove nuclear and cytoplasmic remnants, but did not successfully remove all myelin. Elastin was lost due to the decellularization, but all other investigated ECM components were well-preserved and the mechanical properties remained similar to native nerves.
Table 4

Main impact of the different decellularization protocols.

 

Evaluation

Native

Sondell

Hudson-based

Roosens

Decellularizing agents

  

Triton X-100

SDC

SB-10

SB-16

Triton X-100

DNase, RNase, trypsin

Cellular content

 Nuclear remnants

Biochemistry

852.24 ± 302.28 ng/mg

195.85 ± 53.34 ng/mg

1257.77 ± 236.99 ng/mg

35.15 ± 7.11 ng/mg

 Cytoplasmic remnants

Histology

+++

++

 Myelin

Histology

+++

 

TEM

+++

+

+

ECM composition

 Collagen

Histology

+++

+++

++

++

 GAGs

Biochemistry

14.83 ± 1.55 µg/mg

21.64 ± 3.18 µg/mg

17.34 ± 0.42 µg/mg

16.58 ± 1.02 µg/mg

Histology

++

+++

+++

+++

Biochemistry

2.49 ± 0.28 µg/mg

Not detectable

2.66 ± 0.42 µg/mg

2.10 ± 0.47 µg/mg

 Elastin

Histology

++

+

+

+

 Laminin

Histology

+++

+

++

++

 Mechanical integrity

  

Increased stiffness

Increased stiffness

Similar to native nerves

Histological reactions were scored as absent (−), weak reaction (+), moderate reaction (++) or strong reaction (+++)

Ex Vivo Cytocompatibility

The cellular interaction of ADSCs with nerves decellularized according to our novel decellularization protocol was investigated through an ex vivo cytocompatibility study. Figure 7 shows the results of the live/dead assay, the MTT assay and DNA quantification of ADSCs cultured on cryosectioned decellularized nerves. The viability of the ADSCs remained high after 6 days of culturing on the cryosections and was comparable to the positive control (Fig. 7a). The ADSCs were growing both around and upon the cryosections, making close contact with the nervous tissue. In some cases, the ADSCs exhibited an elongated morphology directional to the endoneurium. The MTT assay revealed that ADSCs show good proliferation on cryosections of nerves of the Roosens group with values similar to the positive control (Fig. 7b). For the negative control, no proliferation could be observed as expected. The high DNA release from the negative control confirms the effective incubation with 2% Triton X-100 causing disruption of the cell membrane (Fig. 7c). Both for the positive control and the ADSCs cultured on the cryosections almost no DNA release could be measured, indicating preservation of the cell membrane integrity and thus a high survival rate.
Figure 7

Ex vivo cytocompatibility of nerves of the Roosens group. (a) Live/dead assay shows good viability for the positive control (pos control) and for the ADSCs cultured on cryosections of nerves of the Roosens group (Decell Roosens) on all timepoints. The negative control (neg control) demonstrates the effective incubation with Triton X-100 resulting in dead cells. (b) The MTT assay reveals that the proliferation of ADSCs on cryosections of nerves of the Roosens group is comparable to the positive control. (C) DNA release from ADSCs of the neg control is significantly higher than from ADSCs of the pos control and ADSCs cultured on the cryosections of nerves of the Roosens group. *p < 0.05 compared to the other experimental conditions of the same timepoint. Scale bar = 500 µm. (n = 5).

Discussion

DPNAs are emerging alternatives for the gold standard PN autografts. They offer the advantage of being off-the-shelf available and provide a supporting role for the migrating SCs and sprouting axons during PN regeneration. In this study, our goal was to evaluate the effectiveness of our newly developed decellularization protocol in comparison with two other detergent-based decellularization protocols. To the best of our knowledge, it is the first time that three different protocols for decellularization of PN allografts are this extensively evaluated, by including not only histochemistry and biochemistry, but also ultrastructural and mechanical analyses. Based on these evaluations, we could demonstrate that our decellularization protocol incorporating Triton X-100, DNase, RNase and trypsin efficiently removed the cellular content and preserved most of the ECM.

The first objective of tissue decellularization is to obtain natural scaffolds that no longer contain cellular material. Not only the removal of nuclear material, but also cytoplasmic remnants need to be taken into account when evaluating DPNAs.36 Therefore, we combined several histochemical (HE, DAPI and MCOLL) and immunohistochemical (S100, vimentin and neurofilament) stainings and complemented these with biochemical quantification of the DNA content. Investigation of the removal of cellular material already revealed marked differences between the decellularization protocols. Nerves of the Sondell group, which were decellularized with a combination of Triton X-100 and SDC, appeared almost completely cell free based on histology. Furthermore, the DNA content drastically decreased from 852.24 ± 302.28 ng/mg for native nerves to 195.85 ± 53.34 ng/mg. However, this decrease of almost 80% is not sufficient to reach the criterion of Crapo et al., stating that less than 50 ng/mg dsDNA may remain after decellularization.16 In contrast, nerves of the Hudson-based group were clearly not cell free after decellularization with SB-10 and SB-16 as demonstrated with histology. Quantification of the DNA content confirmed this finding, since no significant difference could be detected compared to native nerves. These results are surprising, since the protocol described by Hudson et al. formed the basis for the development of the commercially available Avance® Nerve Graft.25,45 In this regard, it is important to mention that Triton X-200 is no longer on the market and therefore, could not be used in this study. The synergistic effect of SB-16 combined with Triton X-200 could not be achieved and indeed, as described in the original study of Hudson et al., SB-16 alone is not an efficient detergent for cellular removal.25 Finally, nerves of the Roosens group were decellularized with a combination of Triton X-100, DNase, RNase and trypsin, which resulted in successful removal of nuclear material. Quantitative biochemical analysis revealed that only 35.15 ± 7.11 ng/mg dsDNA could be detected, meeting the above-mentioned criterion for acceptable decellularization. The beneficial effects of including nucleases in the decellularization protocol have been demonstrated before in a study of Sridharan et al., where the combination of Triton X-100 and SDC led to a twofold decrease of DNA content, while addition of DNase and RNase resulted in a 5-fold decrease.41 Despite the efficient removal of the nuclear content, SEM analysis demonstrated that the myelin was only partially removed from nerves of the Roosens group. Due to the impact of the decellularization process, the myelin, which is partially preserved in formaldehyde-fixed and paraffin-embedded material,8,9,11 was severely extracted by dehydrating and clearing agents used during tissue processing, and thus undetectable by conventional histological techniques. The post-fixation with OsO4 during tissue processing for SEM analysis optimally fixed and stabilized the myelin’s lipids, allowing for accurate demonstration of myelin remnants still present after tissue decellularization. Since remaining myelin might elicit an immune response and can delay or impede regeneration by inducing Wallerian degeneration,42 further finetuning of our promising decellularization protocol is desirable.

In the second part of the study, the preservation of the ECM and molecular composition was investigated, keeping the recently proposed guidelines for structural evaluation of DPNAs in mind.36 Differences between the experimental groups were already obvious when evaluating the nerves macroscopically. Quantification of the transparency demonstrated that native nerves are almost not transparent, while nerves of the Hudson-based and Roosens group are slightly, but not significantly more transparent. In contrast, nerves of the Sondell group showed a significant increase of transparency, which was clearly visible with the naked eye. An increase of transparency has previously been observed in the original study of Sondell et al. as well.40 This observation could indicate that structural components were removed from these nerves.

Histological, biochemical and ultrastructural evaluation methods were employed to further assess the impact of the different decellularization protocols on the nerve architecture and composition. The basal lamina of PNs plays an important role during regeneration and contains the glycoprotein laminin. Preservation of this ECM component in DPNAs is particularly essential to support SC proliferation and migration and to guide newly-formed axons.15,26 However, a previous study showed that laminin was strongly reduced after decellularization with Triton X-100 and SDC, which is the method described by Sondell et al..29 In our study, we also observed, through immunohistochemistry, a decrease of laminin in nerves of the Sondell group. On the other hand, in nerves of the Hudson-based and Roosens group, laminin was well preserved.

Another group of interesting ECM molecules are the GAGs. They form part of proteoglycans, which can be mainly found in the endoneurial tubes and the epineurium. Biochemical quantification with the Blyscan assay revealed a significant decrease of sulfated GAGs in nerves of the Sondell group. This effect on the GAG content has been observed before in a study of Sridharan et al..41 Interestingly, histochemical staining with Alcian Blue demonstrated that most of the sulfated proteoglycans were removed in the epineurium, but not in the endoneurium of nerves of the Sondell group. We hypothesize that the detrimental effect on the GAG content is mainly due to the treatment with SDC and not Triton X-100, since nerves of the Roosens group, which were also treated with Triton X-100, contained a similar level of GAGs after decellularization compared to native nerves.

The most important fibrillar components of the ECM of PNs are collagen fibers. Together with elastic fibers, they are involved in the mechanical properties of PNs. Histological analysis revealed that most of the elastic fibers were removed in all decellularization groups. Collagen fibers, on the other hand, were well preserved as demonstrated with both collagen I immunostaining and Picrosirius Red histochemical staining, as well as with biochemical quantification. Remarkably, the HYP content of the decellularized nerves was even higher than native nerves. Since the HYP content is expressed as a ratio to the tissue dry weight, this can be explained by a reduced total weight of each 10-mm segment while collagen is unaffected, resulting in higher collagen ratios. This observation is further confirmed by TEM, clearly showing the presence of collagen fibers in the interaxonal space in all experimental groups. Possibly, the tight organization of the collagen fibers provides resistance to the decellularizing agents, whereas the loose organization of the elastic fibers makes them more susceptible for solubilization and degradation.37

The impact of the decellularizing agents on the nerve ultrastructure also holds implications for the mechanical properties of the tissues. Up until now, there is no consensus on the effects caused by each detergent separately. Previous studies reported no significant differences in stiffness compared to native nerves when decellularizing with Triton X-100 and SDC.44,46 On the other hand, a more recent study showed an increased stiffness after decellularization with the same detergents.41 In our study, we also found an increased stiffness in nerves of the Sondell group. A possible explanation for the differences observed, is the variation in technical parameters, more specifically the strain rate applied during tensile testing. It has been described that when applying a low strain rate, PNs will be more capable of withstanding morphological changes.2 Interestingly, nerves of the Roosens group showed no significant differences in mechanical properties compared to native nerves. This might indicate that mostly the treatment with SDC is responsible for the altered mechanical properties of nerves of the Sondell group and that the washout of GAGs in this group is strongly related to this effect. Additionally, the repeated exposure to detergents seems to cause more structural damage than the combination of a single detergent with enzymes. Also in nerves of the Hudson-based group the stiffness increased after alternating decellularization with SB-10 and SB-16. Although it has been described that trypsin can disrupt protein–protein interactions as well,16 it seems that the limited exposure in the protocol of Roosens did not significantly affect the nerve ultrastructure and the mechanical properties.

An important challenge for decellularized tissues is the complete removal of the decellularizing agents. Residual detergents could cause cytotoxicity, leading to impaired recellularization and eventually poor regeneration.14,36 To evaluate if any toxic residues were present after decellularization with our novel protocol, we carried out an ex vivo cytocompatibility study. The ADSCs exhibited a good cell-scaffold interaction with the DPNAs generated with the Roosens method. Some cells even adopted an elongated morphology and started aligning along the direction of the endoneurium. No dead cells or significant release of DNA into the culture medium could be observed, indicating that the decellularizing agents were properly washed away during the decellularization process and no leaching of toxic residues has occurred.

In summary, histological, biochemical, ultrastructural and mechanical evaluation methods were employed in this study to thoroughly investigate a novel decellularization protocol for the generation of DPNAs. Based on those results, we can conclude that the addition of nucleases results in more efficient decellularization. Moreover, the nerve ultrastructure and mechanical properties are better preserved when limiting the exposure to detergents and including enzymes such as trypsin. The protocol developed in our lab using Triton X-100, DNase, RNase and trypsin resulted in a cytocompatible scaffold and holds promise for the future as a suitable nerve substitute for peripheral nerve repair.

There are still some limitations associated to the current study. Due to the unavailability of Triton X-200, a direct comparison with the protocol of Hudson et al. was not possible and the true difference with our novel decellularization protocol remains unknown. Furthermore, our protocol makes use of DNase, RNase and trypsin which can be expensive and could raise the costs for upscale manufacturing. However, these enzymes are only needed in very small quantities, minimizing the additional costs. Moreover, in comparison to the protocol of Hudson et al., which is currently being commercialized, the total cost to decellularize an equal amount of nerves is 15 times lower. This is mainly due to the expensive detergent SB-10 which is required in high quantities in the Hudson protocol. Finally, further in vivo studies are warranted to elucidate the clinical potential of our novel decellularization protocol.

Notes

Acknowledgments

This study was supported by the Grant FIS PI14/1343 and FIS PI17/0393 of the Spanish Plan Nacional de Investigación Científica, Desarrollo e Innovación Tecnológica from the Ministerio de Economía y Competitividad, Instituto de Salud Carlos III (co-financed by FEDER funds, European Union) and by the Special Research Fund (BOF 14/IOP/045) from Ghent University, Belgium. The authors would like to thank Dr. Víctor Domingo Roa, Amalia de la Rosa and Concepción Villegas (Experimental Unit of the University Hospital Virgen de las Nieves, Granada) for their assistance with the laboratory animals, Leen Pieters (Ghent University) for the technical assistance with the TEM and Lisa Van Vlaenderen (Ghent University) for the assistance with the ex vivo cytocompatibility.

References

  1. 1.
    Brooks, D. N., R. V. Weber, J. D. Chao, B. D. Rinker, J. Zoldos, M. R. Robichaux, S. B. Ruggeri, K. A. Anderson, E. E. Bonatz, S. M. Wisotsky, M. S. Cho, C. Wilson, E. O. Cooper, J. V. Ingari, B. Safa, B. M. Parrett, and G. M. Buncke. Processed nerve allografts for peripheral nerve reconstruction: a multicenter study of utilization and outcomes in sensory, mixed, and motor nerve reconstructions. Microsurgery 32:1–14, 2012.CrossRefPubMedGoogle Scholar
  2. 2.
    Bueno, F. R., and S. B. Shah. Implications of tensile loading for the tissue engineering of nerves. Tissue Eng. Part B 14:219–233, 2008.CrossRefGoogle Scholar
  3. 3.
    Campos, F., A. B. Bonhome-Espinosa, L. Garcia-Martinez, J. D. Duran, M. T. Lopez-Lopez, M. Alaminos, M. C. Sanchez-Quevedo, and V. Carriel. Ex vivo characterization of a novel tissue-like cross-linked fibrin-agarose hydrogel for tissue engineering applications. Biomed. Mater. 11:055004, 2016.CrossRefPubMedGoogle Scholar
  4. 4.
    Campos, F., A. B. Bonhome-Espinosa, G. Vizcaino, I. A. Rodriguez, D. Duran-Herrera, M. T. Lopez-Lopez, I. Sanchez-Montesinos, M. Alaminos, M. C. Sanchez-Quevedo, and V. Carriel. Generation of genipin cross-linked fibrin-agarose hydrogel tissue-like models for tissue engineering applications. Biomed. Mater. 13:025021, 2018.CrossRefPubMedGoogle Scholar
  5. 5.
    Carbonetto, S., D. Evans, and P. Cochard. Nerve fiber growth in culture on tissue substrata from central and peripheral nervous systems. J. Neurosci. 7:610–620, 1987.CrossRefPubMedGoogle Scholar
  6. 6.
    Carriel, V., M. Alaminos, I. Garzon, A. Campos, and M. Cornelissen. Tissue engineering of the peripheral nervous system. Expert Rev. Neurother. 14:301–318, 2014.CrossRefPubMedGoogle Scholar
  7. 7.
    Carriel, V. S., J. Aneiros-Fernandez, S. Arias-Santiago, I. J. Garzon, M. Alaminos, and A. Campos. A novel histochemical method for a simultaneous staining of melanin and collagen fibers. J. Histochem. Cytochem. 59:270–277, 2011.CrossRefPubMedPubMedCentralGoogle Scholar
  8. 8.
    Carriel, V., A. Campos, M. Alaminos, S. Raimondo, and S. Geuna. Staining methods for normal and regenerative myelin in the nervous system. Methods Mol. Biol. 1560:207–218, 2017.CrossRefPubMedGoogle Scholar
  9. 9.
    Carriel, V., F. Campos, J. Aneiros-Fernandez, and J. A. Kiernan. Tissue Fixation and Processing for the Histological Identification of Lipids. Methods Mol. Biol. 1560:197–206, 2017.CrossRefPubMedGoogle Scholar
  10. 10.
    Carriel, V., J. Garrido-Gomez, P. Hernandez-Cortes, I. Garzon, S. Garcia-Garcia, J. A. Saez-Moreno, M. Del Carmen Sanchez-Quevedo, A. Campos, and M. Alaminos. Combination of fibrin-agarose hydrogels and adipose-derived mesenchymal stem cells for peripheral nerve regeneration. J. Neural Eng. 10:026022, 2013.CrossRefPubMedGoogle Scholar
  11. 11.
    Carriel, V., I. Garzon, M. Alaminos, and A. Campos. Evaluation of myelin sheath and collagen reorganization pattern in a model of peripheral nerve regeneration using an integrated histochemical approach. Histochem. Cell Biol. 136:709–717, 2011.CrossRefPubMedGoogle Scholar
  12. 12.
    Carriel, V., I. Garzon, A. Campos, M. Cornelissen, and M. Alaminos. Differential expression of GAP-43 and neurofilament during peripheral nerve regeneration through bio-artificial conduits. J Tissue Eng Regen. Med. 11:553–563, 2014.CrossRefPubMedGoogle Scholar
  13. 13.
    Carriel, V., G. Scionti, F. Campos, O. Roda, B. Castro, M. Cornelissen, I. Garzon, and M. Alaminos. In vitro characterization of a nanostructured fibrin agarose bio-artificial nerve substitute. J. Tissue Eng. Regen. Med. 11:1412–1426, 2017.CrossRefPubMedGoogle Scholar
  14. 14.
    Cebotari, S., I. Tudorache, T. Jaekel, A. Hilfiker, S. Dorfman, W. Ternes, A. Haverich, and A. Lichtenberg. Detergent decellularization of heart valves for tissue engineering: toxicological effects of residual detergents on human endothelial cells. Artif. Organs 34:206–210, 2010.CrossRefPubMedGoogle Scholar
  15. 15.
    Chernousov, M. A., W. M. Yu, Z. L. Chen, D. J. Carey, and S. Strickland. Regulation of Schwann cell function by the extracellular matrix. Glia 56:1498–1507, 2008.CrossRefPubMedGoogle Scholar
  16. 16.
    Crapo, P. M., T. W. Gilbert, and S. F. Badylak. An overview of tissue and whole organ decellularization processes. Biomaterials 32:3233–3243, 2011.CrossRefPubMedPubMedCentralGoogle Scholar
  17. 17.
    Dahlin, L. B. Techniques of peripheral nerve repair. Scand. J. Surg. 97:310–316, 2008.CrossRefPubMedGoogle Scholar
  18. 18.
    Daly, W., L. Yao, D. Zeugolis, A. Windebank, and A. Pandit. A biomaterials approach to peripheral nerve regeneration: bridging the peripheral nerve gap and enhancing functional recovery. J. R. Soc. Interface 9:202–221, 2012.CrossRefPubMedGoogle Scholar
  19. 19.
    Díaz-Moreno, E., D. Durand-Herrera, V. Carriel, M.-Á. Martín-Piedra, M. D. C. Sánchez-Quevedo, I. Garzón, A. Campos, R. Fernández-Valadés, and M. Alaminos. Evaluation of freeze-drying and cryopreservation protocols for long-term storage of biomaterials based on decellularized intestine. J. Biomed. Mater. Res. Part B 106:488–500, 2018.CrossRefGoogle Scholar
  20. 20.
    Evans, P. J., S. E. Mackinnon, A. D. Levi, J. A. Wade, D. A. Hunter, Y. Nakao, and R. Midha. Cold preserved nerve allografts: changes in basement membrane, viability, immunogenicity, and regeneration. Muscle Nerve 21:1507–1522, 1998.CrossRefPubMedGoogle Scholar
  21. 21.
    Garcia-Martinez, L., F. Campos, C. Godoy-Guzman, M. Del Carmen Sanchez-Quevedo, I. Garzon, M. Alaminos, A. Campos, and V. Carriel. Encapsulation of human elastic cartilage-derived chondrocytes in nanostructured fibrin-agarose hydrogels. Histochem. Cell Biol. 147:83–95, 2017.CrossRefPubMedGoogle Scholar
  22. 22.
    Gulati, A. K. Evaluation of acellular and cellular nerve grafts in repair of rat peripheral nerve. J. Neurosurg. 68:117–123, 1988.CrossRefPubMedGoogle Scholar
  23. 23.
    Haastert-Talini, K., S. Geuna, L. B. Dahlin, C. Meyer, L. Stenberg, T. Freier, C. Heimann, C. Barwig, L. F. Pinto, S. Raimondo, G. Gambarotta, S. R. Samy, N. Sousa, A. J. Salgado, A. Ratzka, S. Wrobel, and C. Grothe. Chitosan tubes of varying degrees of acetylation for bridging peripheral nerve defects. Biomaterials 34:9886–9904, 2013.CrossRefPubMedGoogle Scholar
  24. 24.
    Hess, J. R., M. J. Brenner, I. K. Fox, C. M. Nichols, T. M. Myckatyn, D. A. Hunter, S. R. Rickman, and S. E. Mackinnon. Use of cold-preserved allografts seeded with autologous Schwann cells in the treatment of a long-gap peripheral nerve injury. Plast. Reconstr. Surg. 119:246–259, 2007.CrossRefPubMedGoogle Scholar
  25. 25.
    Hudson, T. W., S. Y. Liu, and C. E. Schmidt. Engineering an improved acellular nerve graft via optimized chemical processing. Tissue Eng. 10:1346–1358, 2004.CrossRefPubMedGoogle Scholar
  26. 26.
    Ide, C., T. Osawa, and K. Tohyama. Nerve regeneration through allogeneic nerve grafts, with special reference to the role of the Schwann cell basal lamina. Prog. Neurobiol. 34:1–38, 1990.CrossRefPubMedGoogle Scholar
  27. 27.
    Ide, C., K. Tohyama, R. Yokota, T. Nitatori, and S. Onodera. Schwann cell basal lamina and nerve regeneration. Brain Res. 288:61–75, 1983.CrossRefPubMedGoogle Scholar
  28. 28.
    Kehoe, S., X. F. Zhang, and D. Boyd. FDA approved guidance conduits and wraps for peripheral nerve injury: a review of materials and efficacy. Injury 43:553–572, 2012.CrossRefPubMedGoogle Scholar
  29. 29.
    Kvist, M., M. Sondell, M. Kanje, and L. B. Dahlin. Regeneration in, and properties of, extracted peripheral nerve allografts and xenografts. J. Plast. Surg. Hand. Surg. 45:122–128, 2011.CrossRefPubMedGoogle Scholar
  30. 30.
    Meek, M. F., and J. H. Coert. US Food and Drug Administration/Conformit Europe-approved absorbable nerve conduits for clinical repair of peripheral and cranial nerves. Ann. Plast. Surg. 60:466–472, 2008.CrossRefPubMedGoogle Scholar
  31. 31.
    Moore, A. M., R. Kasukurthi, C. K. Magill, H. F. Farhadi, G. H. Borschel, and S. E. Mackinnon. Limitations of conduits in peripheral nerve repairs. Hand 4:180–186, 2009.CrossRefPubMedGoogle Scholar
  32. 32.
    Moore, A. M., M. MacEwan, K. B. Santosa, K. E. Chenard, W. Z. Ray, D. A. Hunter, S. E. Mackinnon, and P. J. Johnson. Acellular nerve allografts in peripheral nerve regeneration: a comparative study. Muscle Nerve 44:221–234, 2011.CrossRefPubMedPubMedCentralGoogle Scholar
  33. 33.
    Muir D. F. Method for decellularization of tissue grafts. Google Patents, 2014.Google Scholar
  34. 34.
    Noble, J., C. A. Munro, V. S. Prasad, and R. Midha. Analysis of upper and lower extremity peripheral nerve injuries in a population of patients with multiple injuries. J. Trauma 45:116–122, 1998.CrossRefPubMedGoogle Scholar
  35. 35.
    Oliveira, A. C., I. Garzon, A. M. Ionescu, V. Carriel, J. Cardona, M. Gonzalez-Andrades, M. del Perez, M. Alaminos, and A. Campos. Evaluation of small intestine grafts decellularization methods for corneal tissue engineering. PLoS ONE 8:e66538, 2013.CrossRefPubMedPubMedCentralGoogle Scholar
  36. 36.
    Philips, C., R. Cornelissen, and V. Carriel. Evaluation methods as quality control in the generation of decellularized peripheral nerve allografts. J. Neural Eng. 15:021003, 2018.CrossRefPubMedGoogle Scholar
  37. 37.
    Roosens, A., P. Somers, F. De Somer, V. Carriel, G. Van Nooten, and R. Cornelissen. Impact of detergent-based decellularization methods on porcine tissues for heart valve engineering. Ann. Biomed. Eng. 44:2827–2839, 2016.CrossRefPubMedGoogle Scholar
  38. 38.
    Sandrock, Jr., A. W., and W. D. Matthew. Identification of a peripheral nerve neurite growth-promoting activity by development and use of an in vitro bioassay. Proc. Natl. Acad. Sci. USA 84:6934–6938, 1987.CrossRefPubMedGoogle Scholar
  39. 39.
    Siemionow, M., and G. Brzezicki. Chapter 8 current techniques and concepts in peripheral nerve repair. In: International Review of Neurobiology, edited by S. Geuna, P. Tos, and B. Battiston. Cambridge: Academic Press, 2009, pp. 141–172.Google Scholar
  40. 40.
    Sondell, M., G. Lundborg, and M. Kanje. Regeneration of the rat sciatic nerve into allografts made acellular through chemical extraction. Brain Res. 795:44–54, 1998.CrossRefPubMedGoogle Scholar
  41. 41.
    Sridharan, R., R. B. Reilly, and C. T. Buckley. Decellularized grafts with axially aligned channels for peripheral nerve regeneration. J. Mech. Behav. Biomed. Mater. 41:124–135, 2015.CrossRefPubMedGoogle Scholar
  42. 42.
    Stocum, D. L. Regenerative Biology and Medicine. Boston: Elsevier Academic Press, 2006.Google Scholar
  43. 43.
    Walsh, S., J. Biernaskie, S. W. Kemp, and R. Midha. Supplementation of acellular nerve grafts with skin derived precursor cells promotes peripheral nerve regeneration. Neuroscience 164:1097–1107, 2009.CrossRefPubMedGoogle Scholar
  44. 44.
    Wang, Q., C. Zhang, L. Zhang, W. Guo, G. Feng, S. Zhou, Y. Zhang, T. Tian, Z. Li, and F. Huang. The preparation and comparison of decellularized nerve scaffold of tissue engineering. J. Biomed. Mater. Res. A 102:4301–4308, 2014.PubMedGoogle Scholar
  45. 45.
    Whitlock, E. L., S. H. Tuffaha, J. P. Luciano, Y. Yan, D. A. Hunter, C. K. Magill, A. M. Moore, A. Y. Tong, S. E. Mackinnon, and G. H. Borschel. Processed allografts and type I collagen conduits for repair of peripheral nerve gaps. Muscle Nerve 39:787–799, 2009.CrossRefPubMedGoogle Scholar
  46. 46.
    Zhao, Z., Y. Wang, J. Peng, Z. Ren, L. Zhang, Q. Guo, W. Xu, and S. Lu. Improvement in nerve regeneration through a decellularized nerve graft by supplementation with bone marrow stromal cells in fibrin. Cell Transplant. 23:97–110, 2014.CrossRefPubMedGoogle Scholar
  47. 47.
    Zilic, L., S. P. Wilshaw, and J. W. Haycock. Decellularisation and histological characterisation of porcine peripheral nerves. Biotechnol. Bioeng. 113:2041–2053, 2016.CrossRefPubMedPubMedCentralGoogle Scholar

Copyright information

© Biomedical Engineering Society 2018

Authors and Affiliations

  1. 1.Tissue Engineering and Biomaterials Group, Department of Basic Medical Sciences, Faculty of Medicine and Health SciencesGhent UniversityGhentBelgium
  2. 2.Tissue Engineering Group, Department of Histology, Faculty of MedicineUniversity of GranadaGranadaSpain
  3. 3.Instituto de Investigación Biosanitaria, Ibs.GRANADAGranadaSpain

Personalised recommendations