Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Aug 1.
Published in final edited form as: Plast Reconstr Surg. 2021 Aug 1;148(2):354–365. doi: 10.1097/PRS.0000000000008191

Functional outcomes of nerve allografts seeded with undifferentiated and differentiated mesenchymal stem cells in a rat sciatic nerve defect model

Femke Mathot 1,2, Tiam M Saffari 1,3, Nadia Rbia 1,3, Tim HJ Nijhuis 3, Allen T Bishop 1, Steven ER Hovius 2, Alexander Y Shin 1
PMCID: PMC8373640  NIHMSID: NIHMS1690358  PMID: 34153019

Abstract

Background

Mesenchymal stem cells (MSCs) have the potential to produce neurotrophic growth factors and establish a supportive micro-environment for neural regeneration. The purpose of this study was to determine the effect of undifferentiated and differentiated MSCs dynamically seeded onto decellularized nerve allografts on functional outcomes when used in peripheral nerve repair.

Methods

In 80 Lewis rats a ten millimeter sciatic nerve defect was reconstructed with (i) autograft, (ii) decellularized allograft (iii) decellularized allograft seeded with undifferentiated MSCs, or (iv) decellularized allograft seeded with MSCs differentiated into Schwann cell-like cells. Nerve regeneration was evaluated over time by cross sectional tibial muscle ultrasound measurements, and at 12 and 16 weeks by isometric tetanic force measurements (ITF), compound muscle action potentials (CMAP), muscle mass, histology and immunofluorescence analyses.

Results

At 12 weeks, undifferentiated MSCs significantly improved ITF and CMAP outcomes compared to decellularized allograft alone, while differentiated MSCs significantly improved CMAP outcomes. The autografts outperformed both stem-cell groups histologically at 12 weeks. At 16 weeks, functional outcomes normalized between groups. At both time points, the effect of undifferentiated versus differentiated MSCs was not significantly different.

Conclusions

Undifferentiated and differentiated MSCs significantly improved functional outcomes of decellularized allografts at 12 weeks and were similar to autograft results in the majority of measurements. At 16 weeks, outcomes normalized as expected. Although differences between both cell-types were not statistically significant, undifferentiated MSCs improved functional outcomes of decellularized nerve allografts to a greater extent and have practical benefits for clinical translation by limiting preparation time and costs.

Keywords: mesenchymal stem cells, differentiation, peripheral nerve repair, decellularized nerve allograft, functional recovery

INTRODUCTION

Peripheral nerve defects not amendable to direct end-to-end neurorrhaphy require reconstruction with interposition nerve graft which could be accomplished with autograft, allograft or synthetic bioabsorbable conduits, each with their benefits and controversies.1–3 Decellularized nerve allografts have been proposed as an ideal alternative to overcome donor site morbidity and limited supply of autografts.1,4–7 Improvement of outcomes of decellularized allografts by addition of host derived mesenchymal stem cells (MSCs) has been proposed to overcome the limitations of decellularized allograft nerves by producing trophic factors resulting in a favorable micro-environment for tissue regeneration.8–14 MSCs are hypothesized to not only stimulate tissue regeneration, but potentially form extracellular matrix components, enhance angiogenesis, inhibit scar formation and control immune responses.15 Adipose derived MSCs are easily accessible and proliferate faster than bone marrow derived MSCs, while having a similar effect on nerve regeneration and are thus ideal for translation to clinical use.16–18

In comparison to undifferentiated MSCs, MSCs differentiated into Schwann cell-like cells express neurotrophic and angiogenic genes to a greater extent than undifferentiated MSCs in vitro.16,19–21 Several in vivo studies using different MSC-delivery strategies did not demonstrate clear differences between the outcomes of undifferentiated and differentiated MSCs.22,23 Others reported that differentiated MSCs led to longer regenerating axon distance in vivo19,21,24,25, without resulting in improved functional outcomes.21 The differentiation process of MSCs requires additional preparation time and expensive differentiation factors, which should be considered in translating bench work to clinical application.16

Recent studies have reported a non-traumatic strategy to adhere undifferentiated and differentiated MSCs to the surface of decellularized allografts, leading to a 29-day in vivo survival of seeded MSCs.26–28 The adherence of MSCs to the decellularized allograft has demonstrated an interaction between MSCs and the extracellularly matrix leading to enhanced expression of neurotrophic, angiogenic, extracellular matrix and regulatory cell cycle genes in the first three (differentiated MSCs) to seven (undifferentiated MSCs) days after seeding in vitro, implying a direct effect of differentiated MSCs after implementation while undifferentiated MSCs require time to interact with the environment.14

A comparative study focusing on functional outcomes can elucidate the effect of different cells and their different effective phases on motor nerve regeneration. The purpose of this study was to determine the effect of dynamically seeding undifferentiated and differentiated MSCs onto decellularized nerve allografts7 with respect to functional and histologic outcomes in a rat sciatic defect model.

METHODS

Experimental design

After IACUC institutional review committee and our Institutional Review Board approval (IACUC protocol A2464–00), a 10 mm segmental defect of the sciatic nerve of 80 male Lewis rats weighing 250–300 grams (Envigo, Madison, WI, USA) was repaired with a 10 mm (i) reversed autograft, (ii) decellularized allograft (iii) decellularized allograft seeded with undifferentiated MSCs, or (iv) decellularized allograft seeded with differentiated MSCs. The decellularized allografts originated from Sprague-Dawley rats and were specifically chosen for their histocompatibility mismatch to Lewis rats.29,30 This simulates the clinical setting where an allogenic processed nerve graft is seeded with autologous MSCs. After 12 and 16 weeks, functional, histological and immunofluorescence outcomes were evaluated.

Nerve allograft collection, processing and seeding

Sixty sciatic nerve segments from 30 Sprague-Dawley rats (Envigo, Madison, WI, USA) weighing 250–350 grams served as nerve allografts. After anesthesia with isoflurane, rats were euthanized, shaved and sterilely prepped. The sciatic nerve was exposed, removed under an operating microscope (Zeiss OpMi6, Carl Zeiss Surgical GmbH, Oberkochen, Germany) and processed according to a previously published protocol7. After sterilization with γ-irradiation, nerves were stored at 4°C in Phosphate Buffered Saline (PBS) until surgery.

Stem cell preparation and differentiation

MSCs were derived from the inguinal fat pad of inbred Lewis rats according to protocol.16 Cells were previously characterized by plastic adherence, pluripotency towards mesodermal lineages, the expression of mesenchymal stem cell markers CD29 (88.2%) and CD90 (88.3%) and the absence of hematopoetic cell markers CD34 (91.1% absent) and CD45 (86.0% absent).26–28

Both cell-types were cultured in an incubator at 37°C (5% CO2) and the growth medium was changed every 72 hours. Passage six MSCs were used in this experiment for both differentiated and undifferentiated MSCs.

MSCs-culture

The stromal cell pellet was re-suspended in normal growth medium consisting of α-MEM (Advanced MEM (1x); Life Technologies Corporation, NY, USA), 5% platelet lysate (PLTMax®; Mill Creek Life Sciences, MN, USA), 1% Penicillin/Streptomycin (Penicillin-Streptomycin (10.000 U/mL; Life Technologies Corporation), 1% GlutaMAX (GlutaMAX Supplement 100X; Life Technologies Corporation) and 0.2% Heparin (Heparin Sodium Injection, USP, 1.000 USP units per mL; Fresenius Kabi, IL, USA).

MSC differentiation

MSCs were differentiated into Schwann cell-like cells using a differentiation cocktail containing 0.14% Forskolin (Sigma-Aldrich corp., MO, USA), 0.01% basis fibroblast growth factor (bFGF; PeproTech, NJ, USA), 0.005% platelet-derived growth factor (PDGF-AA; PeproTech) and 0.02% Neuregulin-1 ß1 (NRG1-b1; R&D systems Inc, MN, USA).16 Differentiation was assessed by immunocytochemistry for the expression of S100 (S100; ThermoFisher Scientific, MA, USA), Glial fibrillary acidic protein (GFAP, mouse anti-GFAP; ThermoFisher Scientific) and neurotrophin receptor p75 (p75 NTR, rabbit anti-p75 NTR; ThermoFisher Scientific). Goat anti-rabbit fluorescein isothiocyanate (FITC) and goat anti-mouse cyanine 3 (CY3, both ThermoFisher Scientific) were used as secondary antibodies. Cell nuclei were labeled with 4′,6-diamidino-2-phenylindole (DAPI).

Seeding Protocol for Allograft Nerves

To attach the undifferentiated and differentiated MSCs to the decellularized nerve allografts they were dynamically seeded according to a previously described protocol.28 Either 1×10^6 undifferentiated MSCs or 1×10^6 differentiated MSCs in 10mL growth medium were placed in a conical tube containing a decellularized nerve allograft. The conical tube was then placed on a bioreactor that was positioned in an incubator at 37°C (5% CO2). After the bioreactor had rotated for 12 consecutive hour, the nerve grafts with the attached MSCs were taken out of the tubes and directly implemented in the Lewis rats. The dynamic seeding strategy previously resulted in 80% and 95% adherence of cells on the surface of the processed allografts for undifferentiated and differentiated MSCs respectively.27,28

Surgical procedure of the recipient animals

Under isoflurane anesthesia the right sciatic nerve of the Lewis rat was exposed. A 10 mm segment of the sciatic nerve was excised and reconstructed with a 10 mm graft under an operating microscope (Zeiss OpMi6, Carl Zeiss Surgical GmbH, Oberkochen, Germany). The epineurium was sutured with six 10–0 sutures (10–0 Ethilon, Ethicon Inc., USA), the muscle was approximated (6–0 Vicryl Rapide, Ethicon Inc.) and the skin was closed with a continuous subcutaneous suture (5–0 Vicryl Rapide, Ethicon Inc.). All rats received 5mL of 0.9% saline solution, 0.6mg/kg Buprenorphine and one dose of 30mg/kg diluted trimethoprim/sulfadiazine subcutaneously (Tribissen, Five Star Compounding Pharmacy, Clive, IA). Postoperatively, the rats were individually housed and provided with food and water ad libitum with a 12-hour light-dark cycle.

Ultrasound measurements

The cross-sectional tibial muscle area of six randomly selected rats per group was evaluated with ultrasound measurements of both sides at baseline and at two, four, eight, twelve and sixteen weeks after surgery as previously described using a GE Vivid 7 Ultrasound system (General Electric, Fairfield, CT, USA).31,32 Cross-sectional area was calculated with Adobe Photoshop CC 2018 (Adobe Systems Incorporated, San Jose, CA, USA).

Nonsurvival procedure

At 12 and 16 weeks, ten rats of each group underwent a non-survival procedure. Anesthesia was induced by isoflurane, followed by intraperitoneal injection of Ketamine (80mg/kg) and Xylazine (10mg/kg) and maintained by additional doses of Ketamine (40mg/kg).

Compound Muscle Action Potentials (CMAP) - A miniature bipolar electrode was clamped around the sciatic nerve proximal to the nerve graft. One ground electrode was placed in surrounding musculature and two recording electrodes were superficially placed in the anterior tibial muscle. The CMAP was measured using a VikingQuest portable electromyelogram (Nicolet Biomedical, Madison, WI). A non-recurrent single stimulation with a duration of 0.02ms at an intensity level of 2.7mA was applied. Maximal amplitude measurements were obtained bilaterally.33,34

Isometric Tetanic Force (ITF) - The ITF was measured bilaterally per the protocol of Shin and colleagues.35 The peroneal nerve and tibial muscle were exposed and the hind limb was secured to a testing platform with K-wires through the femur and ankle. The tibial tendon was secured to a clamp in anatomical position and attached to a force transducer (MDB-50; Transducer Techniques, Temecula, CA, USA) whose signals were processed using LabView (National instruments, Austin, Texas). A miniature electrode (Harvard Apparatus, Holliston, MA, USA), stimulated by a bipolar stimulator (Medtronic, Minneapolis, MN, USA) was clamped around the peroneal nerve branch of the sciatic nerve. The muscle tension and the stimulator frequency were optimized after which the maximal ITF was obtained. The tibial muscle was kept moist with warm saline.

Wet tibial muscle mass - Rats were euthanized with an overdose of pentobarbital (Fatal Plus, 390 mg/mL, Vortech, Dearborn, MI, USA) intraperitoneally. Tibial muscles were harvested bilaterally and wet muscle mass was determined after removing the tendon.

Histology - A three millimeter segment of both peroneal nerves of all rats were collected and placed into Trumps solution. Specimens were processed with 0.1M Phosphate Buffer, 1% Osmium tetroxide in buffer, graded series of alcohol and acetone. The samples were infiltrated in a 50%, 75% and finally 100% epoxy resin and polymerized at 65°C for 12–18 hours. Samples were cut in sections at 0.6 microns, placed on slides and stained on a warming plate with Toluidine blue for 2–2.5 minutes. The total tissue cable area (nerve area), axon area, axon count and myelin area were obtained using a Nikon Eclipse 50i microscope and Image Pro Plus Software. The N-ratio was calculated by dividing the myelinated fiber area (axon area and myelin area) by the tissue cable area.36

Immunofluorescence - Both sciatic nerves of five randomly selected rats per group were dissected and fixed in 10% formalin for 48 hours. Nerves samples were vertically embedded in paraffin and sections from the exact middle were stained for Schwann cell marker S100 and protein gene product 9.5 (PGP9.5), a pan neuronal marker. Immunohistochemical staining was performed at the Pathology Research Core (Mayo Clinic, Rochester, MN, USA) using the Leica Bond RX stainer (Leica, Buffalo Grove, IL, USA). The S100 (rabbit polyclonal; Dako, Agilent Technologies Inc., Carpinteria, CA, USA) and PGP9.5 primary antibody (rabbit polyclonal; Dako, Agilent Technologies Inc.) were diluted to 1:5000 in Background Reducing Diluent (Dako, Agilent Technologies Inc.) and incubated for 60 minutes with the samples, prior to staining with the appropriate secondary antibody (Alexa Goat-Anti-Rabbit 488, 1:300, for S100 and Alexa Goat-Anti-Rabbit 568, 1:200 for PGP9.5) and counterstained with Hoechst 33342 (all ThermoFisher Scientific, MA, USA). Images of the stained slides were obtained with a fluorescence laser confocal microscope (Zeiss LSM 780, Carl Zeiss Surgical GmbH, Oberkochen, Germany). The mean fluorescent density of both stains was measured using ImageJ software.

Statistical analysis

All obtained images were blinded and all outcomes were expressed as a percentage of the contralateral side to correct for biological variability between rats. Non-physiologic outcomes were excluded from analysis after review by a statistician and an independent researcher. Two-way analysis of variance (ANOVA) was used for the cross-sectional tibial area measurements. One-way ANOVA was used to compare all other outcome measures between groups. Post-hoc Bonferroni was used to correct for multiple comparisons. Outcomes were expressed as the mean and the standard error of the mean (SEM). Outcomes of cross-sectional tibial muscle area were expressed as mean difference and the standard error of the mean difference. The level of significance was set at α≤0.05.

RESULTS

MSC differentiation

Differentiated MSCs showed immunofluorescence for the markers S100, GFAP and p75 NTR, corresponding to Schwann cells that served as positive controls. Undifferentiated MSCs did not show expression of these markers (figure 1).

Figure 1.

Figure 1.

Open in a new tab

Differentiation of MSCs into Schwann-like cells. Comparison of immunocytochemistry between undifferentiated MSCs (A-D-G), differentiated MSCs (B-E-H) and Schwann cells (C-F-I). Cells are tested for the presence of Schwann cell marker S100 (green, A-B-C), glial cell marker GFAP (red, D-E-F) and neurotrophin Receptor p75 (green, G-H-I). Cell nuclei are DAPI-stained (blue). 40X magnificantion, white scale bar = 40μm.

Functional outcome measurements

Cross-sectional tibial muscle area (figure 2)

Figure 2.

Figure 2.

Open in a new tab

Cross-sectional tibial muscle area ratios (R/L) over time. No significant differences were found between groups. Autografts (+15.77 ±6.56%) and unseeded allografts (+11.33 ±9.22%) had the strongest increase in muscle area between 8 and 12 weeks, while allografts seeded with undifferentiated MSCs (+16.13 ±3.83%) and differentiated MSCs (+10.87 ±9.29%) experienced their strongest increase between 4 and 8 weeks after surgery.

uMSCs = undifferentiated MSCs

dMSCs = differentiated MSCs

Error bars = Standard error of the mean

No significant differences between the groups were found. Within group comparisons only showed significant differences between the consecutive time points zero and two weeks after surgery for autografts, allografts and allografts seeded with differentiated MSCs. The lowest tibial muscle area in all groups was reached at two weeks (40–60% of the unoperated side) and improved upto 16 weeks, with a cross-sectional tibial muscle area ratio of approximately 75%.

Compound Muscle Action Potential (CMAP) (figure 3)

Figure 3.

Figure 3.

Open in a new tab

Compound muscle action potential ratios (CMAP, R/L) at 12 and 16 weeks. CMAP recovery of unseeded allografts was significantly inferior compared to all other groups at 12 weeks

uMSCs = undifferentiated MSCs

dMSCs = differentiated MSCs

* = p<0.05, ** = p<0.01, *** = p<0.001

Error bars = Standard error of the mean

At 12 weeks, CMAP ratio of unseeded allografts (13.48 ±5.00%) was significantly inferior to autografts (53.78 ±5.82%) (p<0.001), allografts seeded with undifferentiated MSCs (44.32 ±7.20%) (p=0.004) and differentiated MSCs (48.89 ±5.37%) (p<0.001). At 16 weeks, CMAP ratio was normalized between all groups, with 57.51 ±7.54% for autografts, 52.26 ±5.80% for allografts, 66.04 ±7.28% for allografts with undifferentiated MSCs and 61.49±8.16% for allografts with differentiated MSCs.

Isometric Tetanic Force (ITF) (figure 4)

Figure 4.

Figure 4.

Open in a new tab

Isometric Tetanic Force ratios (R/L) at 12 and 16 weeks. ITF recovery of unseeded allografts were significantly inferior compared to allografts seeded with undifferentiated MSCs at 12 weeks

uMSCs = undifferentiated MSCs

dMSCs = differentiated MSCs

* = p<0.05, ** = p<0.01, *** = p<0.001

Error bars = Standard error of the mean

The ITF ratio of allografts seeded with undifferentiated MSCs (49.74 ±6.80%) was significantly higher compared to unseeded allografts (26.32 ±4.36%) (p=0.017) at 12 weeks. The ratio in autografts (44.16 ±3.32%) and allografts seeded with differentiated MSCs (43.10 ±4.59%) did not demonstrate significant differences with any of the other groups. At 16 weeks, the ITF ratio of autografts (51.11 ±4.98%), allografts (56.22 ±4.44%), allografts with undifferentiated MSCs (56.12 ±6.51%) and allografts with differentiated MSCs (53.86 ±4.47%) did not significantly differ.

Muscle mass (figure 5)

Figure 5.

Figure 5.

Open in a new tab

Wet tibial muscle mass ratios (R/L) at 12 and 16 weeks. Autografts showed a significantly higher muscle mass recovery compared to unseeded allografts at 12 weeks, and allografts seeded with undifferentiated MSCs at 16 weeks.

uMSCs = undifferentiated MSCs

dMSCs = differentiated MSCs

* = p<0.05, ** = p<0.01, *** = p<0.001

Error bars = Standard error of the mean

At 12 weeks, unseeded allografts measured a significantly lower tibial muscle mass ratio (49.54 ±2.30%) compared to autografts (59.84 ±1.64%) (p=0.021). Allografts with undifferentiated and differentiated MSCs measured a muscle mass ratio of 57.68 ±2.87% and 55.21 ±2.36% respectively. At 16 weeks, the muscle mass ratio of allografts seeded with undifferentiated MSCs was 59.96 ±3.79%, which significantly differed from autografts (74.13 ±1.90%) (p=0.002). Unseeded allografts and allografts seeded with differentiated MSCs had a muscle mass ratio of 69.09 ±1.54% and 70.09 ±2.60% respectively.

Histology

All obtained histology and immunofluorescence values are displayed in table 1. Figure 6 provides representative nerve sections of the different groups. At 12 weeks, autografts had a significant larger axon area ratio compared to unseeded allografts (p<0.001), allografts seeded with undifferentiated MSCs (p<0.001) and allografts seeded with differentiated MSCs (p=0.004). At 16 weeks, no significant differences in axon area ratio between groups were found. The axon count, myelin area and nerve area measures did not demonstrate any significant differences between groups at any of the time points. Autografts had a significant higher N-ratio compared to unseeded allografts (p=0.023) and allografts seeded with undifferentiated MSCs (p=0.040) at 12 weeks. At 16 weeks, autografts had a significantly better N-ratio compared to unseeded allografts (p=0.003), allografts with undifferentiated MSCs (p=0.025) and allografts with differentiated MSCs (p=0.002) (figure 7).

Table 1.

Histology and immunofluorescence outcomes obtained at 12 and 16 weeks for all groups. Outcomes are displayed as the mean and the standard error of the mean (SEM).

Autografts Allografts Allografts + undifferentiated MSCs Allografts + differentiated MSCs
Axon area
 12 weeks 27.58 ±2.93% 10.14 ±1.89% 11.76 ±1.62% 15.48 ±2.27%
 16 weeks 35.95 ±4.63% 21.90 ±3.41% 20.07 ±3.18% 20.84 ±5.65%
Axon count
 12 weeks 56.87 ±7.47% 32.42 ±5.38% 39.61 ±5.11% 47.39 ±5.55%
 16 weeks 50.93 ±7.20% 38.94 ±2.67% 42.04 ±6.91% 43.37 ±13.49%
Myelin area
 12 weeks 57.12 ±4.77% 38.43 ±7.18% 40.11 ±3.58% 46.86 ±4.06%
 16 weeks 69.63 ±8.12% 57.53 ±6.25% 50.69 ±5.72% 41.93 ±7.33%
Nerve area
 12 weeks 77.97 ±7.58% 63.77 ±5.74% 61.58 ±3.53% 71.09 ±6.23%
 16 weeks 77.37 ±8.55% 79.18 ±8.09% 66.27 ±7.40% 60.74 ±8.77%
N-ratio
 12 weeks 59.06 ±2.48% 42.62 ±5.89% 46.58 ±3.59% 46.58 ±3.59%
 16 weeks 72.22 ±2.93% 53.57 ±3.04% 57.53 ±2.74% 51.71 ±5.34%
S100 density
 12 weeks 96.48 ±1.16% 97.84 ±1.65% 92.85 ±3.17% 94.88 ±3.11%
 16 weeks 102.37 ±6.94% 93.54 ±2.38% 94.61 ±3.86% 95.92 ±5.90%
PGP9.5 density
 12 weeks 106.95 ±16.30% 105.26 ±4.88% 101.13 ±9.55% 98.62 ±7.97%
 16 weeks 92.81 ±5.15% 101.89 ±12.22% 98.13 ±13.88% 76.68 ±6.25%
Open in a new tab

Figure 6.

Figure 6.

Open in a new tab

Examples of obtained images of peroneal nerve sections stained with toluidine blue at 12 weeks of follow-up. Scale bar upper images = 1mm, lower images = 0.5mm.

Figure 7.

Figure 7.

Open in a new tab

N- ratios (R/L) at 12 and 16 weeks. Autografts had a significant better N-ratio than unseeded allografts and allografts seeded with undifferentiated MSCs at 12 weeks and compared to other groups at 16 weeks.

uMSCs = undifferentiated MSCs

dMSCs = differentiated MSCs

* = p<0.05, ** = p<0.01, *** = p<0.001

Error bars = Standard error of the mean

Immunofluorescence

S100 and PGP9.5 density outcomes are displayed in table 1. Examples of obtained immunofluorescence images are displayed in figure 8. Between group comparisons showed no significant differences in S100 and PGP9.5 density ratio at both time points.

Figure 8.

Figure 8.

Open in a new tab

An overview of examples of the obtained images of nerve sections stained with S100 (Schwann cells, green) or PGP9.5 (axons, red) (both immunohystochemistry) at 12 weeks. Cell nuclei are DAPI stained and displayed in blue. The displayed images are obtained with 20X magnification, white error bar = 0.2mm.

DISCUSSION

Despite advances in decellularization techniques for allograft nerves, nerve autograft remain the gold standard for segmental defect reconstruction of critical motor or sensory nerves.1,6 To overcome the limitations of decellularized allograft nerves, MSCs have been hypothesized to improve outcomes of decellularized allograft nerves7 by producing proteins and cytokines that establish a micro-environment favorable for neural regeneration.8,12–14,37 Differentiated MSCs have been demonstrated to exert their neurotrophic effect immediately after implementation by expressing increased levels of neurotrophic genes, while undifferentiated MSCs require additional time to interact with the surrounding tissue prior to expressing neurotrophic genes.14 The purpose of this study was to determine the effect of dynamically seeding undifferentiated and differentiated MSCs onto decellularized nerve allografts7 with respect to functional and histologic outcomes in a rat sciatic defect model, in order to determine which cell-type has greatest clinical potential.

In this study, MSCs were successfully differentiated into Schwann cell-like cells16 and dynamically seeded onto decellularized nerve allografts.27 Compared to unseeded allografts, undifferentiated MSCs led to significant improvement of both ITF and CMAP (p=0.017 and p=0.004) outcomes at 12 weeks, while differentiated MSCs only led to significant improved CMAP outcomes (p<0.001). These findings correspond to the study of Hou and colleagues whom observed that (differentiated) MSC-seeded grafts recovered earlier than acellular grafts when measuring electrophysiology, with significant results at 12 weeks.38 Differences between groups normalized at 16 weeks which is consistent with the study of Tang and colleagues, that demonstrated normalizing ITF measurements at 16 weeks of follow-up.39 Functional assessment did not result in any significant differences between both cell-types for all functional outcome measures at 12 and 16 weeks, which is in line with published studies of Orbay and Watanabe.22,23 The hypothesized consequences of different effective phases of both cell-types could not be confirmed in this study.

At 16 weeks, no significant differences in functional outcomes between groups were found, except for muscle mass recovery that was significantly better in autografts than in allografts seeded with differentiated MSCs (p=0.002). Although muscle mass is easily obtainable, it is an indirect measurement of motor outcome as enlarged muscle fibers do not neccesarily feature improved contractility.36 ITF has been described to objectively quantify contractility of muscle fibers and is easily reproducible.36 The vulnerability of CMAP measurements, which is affected by the placement of all individual electrodes, may explain why the CMAP outcomes are greater than the ITF measures.40

Histologically, the autografts had significantly better N-ratios in the peroneal nerves at both time points compared to all other groups. Although not examined, this could be explained by less formation of fibrosis in autografts.41 Due to small groups and insufficient sensitivity of density measures, the histology outcomes could not be significantly confirmed by immunofluorescence outcomes, but unseeded nerve allografts subjectively seem to contain less Schwann cells and axons compared to all other groups.

Autografts were used as control group to test whether MSCs could improve outcomes of decellularized allografts up to a level equal to that of autografts. While an additional control group in which sham surgery is performed would also be interesting to have, it would require the undesirable and precious use of additional animals. Alternatively, outcomes of the operated side were normalized to the unoperated control side in order to relate the test-outcomes to normal nerve and muscle function.

The significant differences between groups presented at 12 weeks and normalized after 16 weeks, insinuates that nerve regeneration in motor nerves in rats will occur after 12 weeks independently from the type of nerve repair. This finding might be correlated to the demonstrated finite survival of MSCs up to 29 days in vivo; it is suggested that MSCs significantly enhance nerve regeneration up to 12 weeks after which the superlative neuroregenerative capacity of rats takes over, due to the apoptosis of the MSCs.42 The superlative neuroregenerative capacity of rats is a commonly described explanation and can be mitigated in a larger animal model.26,39 Absent significant differences when comparing cross-sectional tibial muscle areas is also a likely consequence of using a small animal model with small cross-sectional nerve areas, relatively leading to larger standard errors and less significant differences between groups.31,32 Future research should be performed on multiple time points in larger animal models with larger nerve gaps to potentially translate outcomes to humans.

Considering the overall goal to improve outcomes of decellularized nerve allografts in clinical practice, clinical applicability should be considered when interpretating results. The use of autologous differentiated MSCs requires approximately 4–5 weeks of preparation time, against 2–3 weeks for undifferentiated MSCs.16 Moreover, the costs of the differentiation cocktail required to differentiate MSCs into Schwann Cell-like cells are high and add to the costs of extended cell culture. Differences between undifferentiated an differentiated MSCs were not statistically significant in light of the analyzed factors, but undifferentiated MSCs improved functional outcomes of decellularized nerve allografts to a greater extent than differentiated MSCs. Taking all this in consideration, undifferentiated MSCs have the greatest potential for bench-to-bedside application. Hypothetically, at the day of presentation in a clinical setting, adipose tissue can be obtained using minimally invasive techniques from the patient with nerve injury, MSCs can then be derived from this tissue and cultured for approximately 2 weeks after which the MSCs can be dynamically seeded onto an off-the-shelf commercially available nerve allografts, 12 hours in advance of the nerve repair. Translation to a larger animal model to ensure the enhanced functional outcomes, study of the capacity of human MSCs to be seeded on clinically available nerve allografts and FDA approval are potential hurdles that need to be addressed prior to application of the presented strategy in clinical practice.

CONCLUSION

Undifferentiated and differentiated MSCs significantly improved functional outcomes of decellularized allografts at 12 weeks in motor nerves and equaled the autograft results in the majority of outcome measurements. At 16 weeks, outcome measures normalized as expected. Considering clinical applicability, undifferentiated MSCs are more attractive as outcomes did not significantly differ between both cell-types, and differentiation requires increased time and cost.

Acknowledgements

We thank Roman Thaler, PhD, for assisting with cell culture and differentiation of the Mesenchymal Stem Cells.

We thank Patricia F. Friedrich for assistance with the preparations of the experiments.

Financial Disclosure Statement

The authors have nothing to disclose. Research reported in this publication was supported by the National Institute of Neurological Disorders and Stroke of the National Institutes of Health under Award Number RO1 NS 102360. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Footnotes

Presented at: ASSH Annual meeting 2018, Boston, MA, USA

REFERENCES

  • 1.Giusti G, Willems WF, Kremer T, Friedrich PF, Bishop AT, Shin AY. Return of motor function after segmental nerve loss in a rat model: comparison of autogenous nerve graft, collagen conduit, and processed allograft (AxoGen). J Bone Joint Surg Am. 2012;94(5):410–417. [DOI] [PubMed] [Google Scholar]
  • 2.Millesi H Progress in peripheral nerve reconstruction. World J Surg. 1990;14(6):733–747. [DOI] [PubMed] [Google Scholar]
  • 3.Mackinnon SE, Hudson AR, Falk RE, Hunter DA. The nerve allograft response--an experimental model in the rat. Ann Plast Surg. 1985;14(4):334–339. [DOI] [PubMed] [Google Scholar]
  • 4.Whitlock EL, Tuffaha SH, Luciano JP, et al. Processed allografts and type I collagen conduits for repair of peripheral nerve gaps. Muscle & Nerve. 2009;39(6):787–799. [DOI] [PubMed] [Google Scholar]
  • 5.Cho MS, Rinker BD, Weber RV, et al. Functional outcome following nerve repair in the upper extremity using processed nerve allograft. J Hand Surg. 2012;37(11):2340–2349. [DOI] [PubMed] [Google Scholar]
  • 6.Rbia N, Shin AY. The Role of Nerve Graft Substitutes in Motor and Mixed Motor/Sensory Peripheral Nerve Injuries. J Hand Surg Am. 2017;42(5):367–377. [DOI] [PubMed] [Google Scholar]
  • 7.Hundepool CA, Nijhuis TH, Kotsougiani D, Friedrich PF, Bishop AT, Shin AY. Optimizing decellularization techniques to create a new nerve allograft: an in vitro study using rodent nerve segments. Neurosurg Focus. 2017;42(3). [DOI] [PubMed] [Google Scholar]
  • 8.Caplan AI. Adult Mesenchymal Stem Cells: When, Where, and How. Stem Cells Int’l. 2015;2015:628767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Mushtaq M, DiFede DL, Golpanian S, et al. Rationale and design of the Percutaneous Stem Cell Injection Delivery Effects on Neomyogenesis in Dilated Cardiomyopathy (the POSEIDON-DCM study): a phase I/II, randomized pilot study of the comparative safety and efficacy of transendocardial injection of autologous mesenchymal stem cell vs. allogeneic mesenchymal stem cells in patients with non-ischemic dilated cardiomyopathy. J Cardiovasc Translat Res. 2014;7(9):769–780. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Shen H, Wang Y, Zhang Z, Yang J, Hu S, Shen Z. Mesenchymal Stem Cells for Cardiac Regenerative Therapy: Optimization of Cell Differentiation Strategy. Stem cells Int’l. 2015;2015:524756. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Nam HY, Karunanithi P, Loo WC, et al. The effects of staged intra-articular injection of cultured autologous mesenchymal stromal cells on the repair of damaged cartilage: a pilot study in caprine model. Arthritis Res & Therap. 2013;15(5):R129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Castro-Manrreza ME, Montesinos JJ. Immunoregulation by mesenchymal stem cells: biological aspects and clinical applications. J Immunol Res. 2015;2015:394917. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Rbia N, Bulstra LF, Lewallen EA, Hovius SER, van Wijnen AJ, Shin AY. Seeding decellularized nerve allografts with adipose-derived mesenchymal stromal cells: An in vitro analysis of the gene expression and growth factors produced. JRAS. 2019;72(8):1316–1325. [DOI] [PubMed] [Google Scholar]
  • 14.Mathot F, Rbia N, Thaler R, Bishop AT, Van Wijnen AJ, Shin AY. Gene expression profiles of differentiated and undifferentiated adipose derived mesenchymal stem cells dynamically seeded onto a processed nerve allograft. Gene. 2020;724:144151. [DOI] [PubMed] [Google Scholar]
  • 15.Mathot F, Shin AY, Van Wijnen AJ. Targeted stimulation of MSCs in peripheral nerve repair. Gene. 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Kingham PJ, Kalbermatten DF, Mahay D, Armstrong SJ, Wiberg M, Terenghi G. Adipose-derived stem cells differentiate into a Schwann cell phenotype and promote neurite outgrowth in vitro. Exper Neurol. 2007;207(2):267–274. [DOI] [PubMed] [Google Scholar]
  • 17.Yoshimura H, Muneta T, Nimura A, Yokoyama A, Koga H, Sekiya I. Comparison of rat mesenchymal stem cells derived from bone marrow, synovium, periosteum, adipose tissue, and muscle. Cell Tis Res. 2007;327(3):449–462. [DOI] [PubMed] [Google Scholar]
  • 18.di Summa PG, Kingham PJ, Raffoul W, Wiberg M, Terenghi G, Kalbermatten DF. Adipose-derived stem cells enhance peripheral nerve regeneration. Journal of plastic, reconstructive & aesthetic surgery : JPRAS. 2010;63(9):1544–1552. [DOI] [PubMed] [Google Scholar]
  • 19.Kingham PJ, Kolar MK, Novikova LN, Novikov LN, Wiberg M. Stimulating the neurotrophic and angiogenic properties of human adipose-derived stem cells enhances nerve repair. Stem Cells Develop. 2014;23(7):741–754. [DOI] [PubMed] [Google Scholar]
  • 20.Tomita K, Madura T, Sakai Y, Yano K, Terenghi G, Hosokawa K. Glial differentiation of human adipose-derived stem cells: implications for cell-based transplantation therapy. Neurosci. 2013;236:55–65. [DOI] [PubMed] [Google Scholar]
  • 21.Ladak A, Olson J, Tredget EE, Gordon T. Differentiation of mesenchymal stem cells to support peripheral nerve regeneration in a rat model. Exper Neurol. 2011;228(2):242–252. [DOI] [PubMed] [Google Scholar]
  • 22.Orbay H, Uysal AC, Hyakusoku H, Mizuno H. Differentiated and undifferentiated adipose-derived stem cells improve function in rats with peripheral nerve gaps. Journal of plastic, reconstructive & aesthetic surgery : JPRAS. 2012;65(5):657–664. [DOI] [PubMed] [Google Scholar]
  • 23.Watanabe Y, Sasaki R, Matsumine H, Yamato M, Okano T. Undifferentiated and differentiated adipose-derived stem cells improve nerve regeneration in a rat model of facial nerve defect. J Tis Eng Regen Med. 2017;11(2):362–374. [DOI] [PubMed] [Google Scholar]
  • 24.Keilhoff G, Goihl A, Stang F, Wolf G, Fansa H. Peripheral nerve tissue engineering: autologous Schwann cells vs. transdifferentiated mesenchymal stem cells. Tis Eng. 2006;12(6):1451–1465. [DOI] [PubMed] [Google Scholar]
  • 25.Kappos EA, Engels PE, Tremp M, et al. Peripheral Nerve Repair: Multimodal Comparison of the Long-Term Regenerative Potential of Adipose Tissue-Derived Cells in a Biodegradable Conduit. Stem Cells Develop. 2015;24(18):2127–2141. [DOI] [PubMed] [Google Scholar]
  • 26.Rbia N, Bulstra LF, Thaler R, Hovius SER, van Wijnen AJ, Shin AY. In Vivo Survival of Mesenchymal Stromal Cell-Enhanced Decellularized Nerve Grafts for Segmental Peripheral Nerve Reconstruction. J Hand Surg Am. 2019;44(6):514.e1–514.e11. [DOI] [PubMed] [Google Scholar]
  • 27.Mathot F, Rbia N, Bishop AT, Hovius SER, Van Wijnen AJ, Shin AY. Adhesion, distribution, and migration of differentiated and undifferentiated mesenchymal stem cells (MSCs) seeded on nerve allografts. JPRAS. 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Rbia N, Bulstra LF, Bishop AT, van Wijnen AJ, Shin AY. A simple dynamic strategy to deliver stem cells to decellularized nerve allografts. JPRAS. 2018;142(2):402–413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Hudson TW, Zawko S, Deister C, et al. Optimized acellular nerve graft is immunologically tolerated and supports regeneration. Tis Eng. 2004;10(11–12):1641–1651. [DOI] [PubMed] [Google Scholar]
  • 30.Kumta S, Yip K, Roy N, Lee SK, Leung PC. Revascularisation of bone allografts following vascular bundle implantation: an experimental study in rats. Arch Orthop Trauma Surg. 1996;115(3–4):206–210. [DOI] [PubMed] [Google Scholar]
  • 31.Hundepool CA, Nijhuis TH, Rbia N, Bulstra LF, Selles RW, Hovius SE. Noninvasive Ultrasound of the Tibial Muscle for Longitudinal Analysis of Nerve Regeneration in Rats. PRS. 2015;136(5):633e–639e. [DOI] [PubMed] [Google Scholar]
  • 32.Bulstra LF, Hundepool CA, Friedrich PF, Nijhuis TH, Bishop AT, Shin AY. Motor Nerve Recovery in a Rabbit Model: Description and Validation of a Noninvasive Ultrasound Technique. J Hand Surg Am. 2016;41(1):27–33. [DOI] [PubMed] [Google Scholar]
  • 33.Hundepool CA, Bulstra LF, Kotsougiani D, et al. Comparable functional motor outcomes after repair of peripheral nerve injury with an elastase-processed allograft in a rat sciatic nerve model. Microsurg. 2018;38(7):772–779. [DOI] [PubMed] [Google Scholar]
  • 34.Giusti G, Lee JY, Kremer T, Friedrich P, Bishop AT, Shin AY. The influence of vascularization of transplanted processed allograft nerve on return of motor function in rats. Microsurg. 2016;36(2):134–143. [DOI] [PubMed] [Google Scholar]
  • 35.Shin RH, Vathana T, Giessler GA, Friedrich PF, Bishop AT, Shin AY. Isometric tetanic force measurement method of the tibialis anterior in the rat. Microsurg. 2008;28(6):452–457. [DOI] [PubMed] [Google Scholar]
  • 36.Vleggeert-Lankamp CL. The role of evaluation methods in the assessment of peripheral nerve regeneration through synthetic conduits: a systematic review. Laboratory investigation. J Neurosurg. 2007;107(6):1168–1189. [DOI] [PubMed] [Google Scholar]
  • 37.Ma S, Xie N, Li W, Yuan B, Shi Y, Wang Y. Immunobiology of mesenchymal stem cells. Cell Death Different. s2014;21(2):216–225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Hou SY, Zhang HY, Quan DP, Liu XL, Zhu JK. Tissue-engineered peripheral nerve grafting by differentiated bone marrow stromal cells. Neurosci. 2006;140(1):101–110. [DOI] [PubMed] [Google Scholar]
  • 39.Tang P, Whiteman DR, Voigt C, Miller MC, Kim H. No Difference in Outcomes Detected Between Decellular Nerve Allograft and Cable Autograft in Rat Sciatic Nerve Defects. J Bone Joint Surg Am. 2019;101(10):e42. [DOI] [PubMed] [Google Scholar]
  • 40.Saffari S, Saffari TM, Moore AM, Shin AY. Peripheral Nerve Basic Science Research – What is important for hand surgeons to know? J Hand Surg Am. (invited submission) 2020. [DOI] [PubMed] [Google Scholar]
  • 41.den Dunnen WF, van der Lei B, Schakenraad JM, et al. Poly(DL-lactide-epsilon-caprolactone) nerve guides perform better than autologous nerve grafts. Microsurg. 1996;17(7):348–357. [DOI] [PubMed] [Google Scholar]
  • 42.Rbia N, Bulstra LF, Thaler R, Hovius SER, van Wijnen AJ, Shin AY. In vivo survival of mesenchymal stromal cell-enhanced decellularized nerve grafts for segmental peripheral nerve reconstruction. J Hand Surg.2019;44(6):514.e511. [DOI] [PubMed] [Google Scholar]

RESOURCES