Human umbilical cord mesenchymal stem cell-derived exosomes loaded into a composite conduit promote functional recovery after peripheral nerve injury in rats

Haoshuai Tang ,Junjin Li ,Hongda Wang,Jie RenHan DingJun ShangMin Wang,Zhijian Wei,Shiqing Feng

Abstract Complete transverse injury of peripheral nerves is challenging to treat.Exosomes secreted by human umbilical cord mesenchymal stem cells are considered to play an important role in intercellular communication and regulate tissue regeneration.In previous studies,a collagen/hyaluronic acid sponge was shown to provide a suitable regeneration environment for Schwann cell proliferation and to promote axonal regeneration.This three-dimensional (3D) composite conduit contains a collagen/hyaluronic acid inner sponge enclosed in an electrospun hollow poly (lactic-co-glycolic acid) tube.However,whether there is a synergy between the 3D composite conduit and exosomes in the repair of peripheral nerve injury remains unknown.In this study,we tested a comprehensive strategy for repairing long-gap (10 mm) peripheral nerve injury that combined the 3D composite conduit with human umbilical cord mesenchymal stem cell-derived exosomes.Repair effectiveness was evaluated by sciatic functional index,sciatic nerve compound muscle action potential recording,recovery of muscle mass,measuring the cross-sectional area of the muscle fiber,Masson trichrome staining,and transmission electron microscopy of the regenerated nerve in rats.The results showed that transplantation of the 3D composite conduit loaded with human umbilical cord mesenchymal stem cell-derived exosomes promoted peripheral nerve regeneration and restoration of motor function,similar to autograft transplantation.More CD31-positive endothelial cells were observed in the regenerated nerve after transplantation of the loaded conduit than after transplantation of the conduit without exosomes,which may have contributed to the observed increase in axon regeneration and distal nerve reconnection.Therefore,the use of a 3D composite conduit loaded with human umbilical cord mesenchymal stem cell-derived exosomes represents a promising cell-free therapeutic option for the treatment of peripheral nerve injury.

Key Words: axon growth;collagen;exosome;human umbilical cord mesenchymal stem cells;hyaluronic acid;muscular atrophy;nerve guidance conduits;peripheral nerve regeneration

Introduction

Peripheral nerve injury (PNI) is a complex condition associated with a variety of signs and symptoms,such as loss of motor and/or sensory function,myophagism,numbness,burning,or sharp pain (Gordon,2020;Hussain et al.,2020).Peripheral nerves are present throughout most of the body and are fragile,which means that they are easily damaged as a result of injury to nearly any area (Midha and Grochmal,2019).Because nerve cells are highly differentiated,nerve regeneration post-injury is slow and incomplete,so patients often experience unpredictable recovery patterns and may develop lifelong disabilities (Herrup et al.,2004).Although current research shows that injured peripheral nerves have the potential for axon growth,there are no effective therapies for full recovery of nerve function when the nerve defect is too long and end-to-end suturing cannot be performed (Singh et al.,2022).In these cases,it is necessary to carry out autologous nerve graft,which is considered the gold-standard treatment for long-gap nerve defects.However,there are some problems with this approach,such as donor nerve deficiency,neuroma formation,axonal misdirection,secondary surgery,and additional costs,which generally lead to unsatisfactory clinical results (Grinsell and Keating,2014;Wang et al.,2019).Nerve guidance conduits based on biomaterials have become a promising alternative to traditional autologous nerve graft.

Various artificial nerve guidance conduits have been fabricated to bridge broken nerve stumps when autologous nerve transplantation is impossible (Korus et al.,2016).Nevertheless,the traditional hollow tube conduits are unable to provide adequate biomechanical support or to regulate the inhibitory microenvironment after injury.Natural biomaterials (such as hyaluronic acid (HA) and collagen) have many advantages compared with artificial biomaterials,such as an appropriate biodegradation rate,no adverse reaction to the degradation products,and good biocompatibility,as well as promotion of normal cell biological activity (survival,adhesion,and migration) (Fornasari et al.,2020;Jiang et al.,2022).Both HA and collagen are important components of the extracellular matrix (ECM) surrounding adult peripheral nerves (Gregory and Phillips,2021;Poongodi et al.,2021).Type I and III collagen are the main collagens present in the ECM of the peripheral nerve system,and they play a crucial role in ECM remodeling and tissue development (Fujii et al.,1986;Deister et al.,2007).HA decreases scarring,improves fibrin matrix formation,and prevents peripheral nerve adhesion (Seckel et al.,1995;Ikeda et al.,2003;Özgenel,2003;Djoudi et al.,2022).HA not only has excellent biocompatibility and safety,but is also commonly used to deliver additional therapeutic factors (Ma et al.,2007;Xu et al.,2021).Therefore,three-dimensional (3D) composite conduit based on collagen and HA (Col/HA) has certain advantages in mimicking the microstructure and function of the ECM of peripheral nerves.As the mechanical strength of Col/HA sponges is insufficient to support peripheral nerve repair,electrospun poly (lactic-co-glycolic acid) (PLGA) has been used as a shell to increase the strength of the Col/HA core (Entekhabi et al.,2021).PLGA is a highly biocompatible copolymer that biodegradable via hydrolysis of its ester bonds,and has strong mechanical and prominent electrospinning properties (Gentile et al.,2014).In short,we assembled a 3D composite conduit with Col/HA inner sponge and electrospun PLGA nanofibers outer shell,but it is still necessary to load a therapeutic factor to achieve the efficacy of autograft.

In recent years,studies have shown that human umbilical cord mesenchymal stem cells (hUCMSCs) have regenerative potential and may promote restoration of neurological function after injury (Yousefi et al.,2019;Bojanic et al.,2020).hUCMSCs have become one of the most promising types of mesenchymal stem cells with clinical potential owing to their painless collection,high proliferation ability,low immunogenicity,potent immunomodulatory properties,and the absence of ethical issues (Secco et al.,2008;Ding et al.,2015;Kubiak et al.,2020).In addition,hUCMSCs secrete bioactive compounds into their surrounding environment,affecting nearby cells through paracrine activity (Guo et al.,2015).However,directly transplanting hUCMSCs into lesions,which have an adverse pathological microenvironment,is detrimental to hUCMSC survival and results in low transplantation efficiency (Stoll et al.,2002;Davies et al.,2020).Recent research has shown that the pleiotropic effects of MSCs are mainly mediated by the secretion of soluble paracrine factors,especially exosomes (Nikfarjam et al.,2020).Exosomes,which are nanosized extracellular vesicles with diameters ranging between 50 and 200 nm (100 nm on average) and a lipid bilayer membrane,are important paracrine mediators (Joo et al.,2020).Exosomes contain substances including proteins,lipids,and RNA that are involved in mediating intercellular communication (Zhang and Yang,2018;Dong et al.,2019).Furthermore,circRNAs and microRNAs carried by exosomes participate in several pivotal biological processes,such as angiopoiesis and nerve regeneration (Qing et al.,2018).In general,hUCMSCderived exosomes promote nerve regeneration with similar efficacy to hUCMSCs and avoid the shortcomings of hUCMSC transplantation therapy.

Efficiency and fast action are the primary goals of clinical treatment,especially for PNI,because long-term denervation can lead to target organ atrophy.Although transplanting hUCMSCs with artificial nerve guidance conduits may be a more effective way to promote repair of sciatic nerve defects than transplanting hUCMSCs alone,peripheral nerve cell proliferation and differentiation take a long time,and overloading the injured area with cells accelerates consumption of limited nutrients,which is not conducive to nerve regeneration.Thus,for repairing peripheral nerve defects,the use of hUCMSC-derived exosomes is preferable because it is a cell-free regenerative medicine.Therefore,in this study we designed and tested the efficacy of a comprehensive treatment strategy that combines a PLGA@Col/HA conduit with hUCMSC-derived exosomes (PLGA@Col/HA+exo) to promote peripheral nerve repair.

Methods

Preparation and characterization of composite conduits

First,PLGA was added to a glass bottle,and an appropriate amount of hexafluoroisopropanol was added to prepare a 10% (w/v) concentration of PLGA electrospinning solution.Spinning was performed at 40°C,30% humidity using a 22# needle at a feed rate of 2.0 mL/hour and a voltage of 18 kV.When the spinning step was complete,the material on the receiver was removed and dried in an oven at 50°C for 1 hour to remove the residual solvent,yielding a PLGA shell with a diameter of 1.5 mm.Next,a 1% collagen solution and a 4% hyaluronic acid solution were combined in a 1:1 volume ratio,and the resulting Col/HA mixture was used to fill the electrospun PLGA tube.Finally,the prepared 3D composite conduit was pre-frozen at -20°C,then vacuum freeze-dried at -80°C for 24 hours.SEM (Hitachi SU8100,Tokyo,Japan) was used to analyze the PLGA tube morphology and the porosity of the Col/HA filler.

The fiber diameter and average pore size of the Col/HA filler was calculated using the Nano measure software (version 1.2,Fudan University,Shanghai,China).The diameters of all the pores observed in the SEM images were measured,and the average value was defined as the pore size of the Col/HA sponge.The PLGA shell,Col,HA,and assembled conduit were analyzed using a Fourier transform infrared (FTIR) spectrophotometer (Thermo Scientific Nicolet iN10,Thermo Fisher Scientific,Waltham,MA,USA) with a scan range of 400-4000 cm-1.The mechanical properties of the 3D composite conduits were assessed using a uniaxial testing instrument (INSTRON 3343,Instron,Shanghai,China).First,the conduits were cut into 12.0-mm lengths to confirm that they were regular and cylindrical.Then,a single segment was carefully fixed in the wedge grip,and force was gradually applied until the segment broke.Finally,the stress-strain curves were plotted based on the force needed to break the conduit samples.

Preparation and identification of hUCMSCs

The hUCMSCs used in our study were provided by Changhe Biotechnology Inc.(Tianjin,China).To confirm the multipotent nature of the hUCMSCs,the cells were cultured in osteogenic,adipogenic,and chondrogenic media.When the differentiation process was complete,the cells were fixed.The differentiated chondrocytes,adipocytes,and osteoblasts were stained with Alcian blue,Oil Red O,and alizarin red (all from Sigma-Aldrich,St.Louis,MO,USA) at room temperature for 60 minutes according to the manufacturer’s protocols (Peng et al.,2011).After rinsing with phosphate-buffered saline (PBS) to wash away the staining solution,the specimens were observed under a light microscope (Olympus,Tokyo,Japan).

hUCMSCs were detached from the culture dish by digesting with trypsin,centrifuged,and resuspended to yield a single-cell suspension.Then,the hUCMSCs (1×105cells) were incubated with the following fluorescein isothiocyanate (FITC)-conjugated or phycoerythrin (PE)-conjugated monoclonal antibodies for 30 minutes in the dark at 4°C: anti-human CD14 (mouse,1×106cells/µL,Abcam,Cambridge,UK,Cat# ab36595,RRID: AB_726114),anti-human CD45 (mouse,1×106cells/µL,Abcam,Cat# ab123522,RRID: AB_10975507),anti-human CD90 (mouse,1×106cells/µL,Abcam,Cat# ab23894,RRID: AB_447748),and anti-human CD105 (mouse,1×106cells/µL,Abcam,Cat# ab114052,RRID: AB_10900113).The hUCMSC cell surface antigens were then analyzed by flow cytometry (CytoFLEX;Beckman Coulter Inc.,Brea,CA,USA).

Isolation and characterization of hUCMSC-derived exosomes

After 3 days of hUCMSC growth in serum-free medium,the cell culture medium was harvested and ultracentrifuged,and the collected exosomes were stored at -80°C for further experiments (Chen et al.,2020).Exosome morphology and structure were evaluated by transmission electron microscopy (Hitachi HT7700).Nanoparticle tracking analysis (NTA;NanoSight NS300;Malvern Panalytical,Malvern,UK) was used to measure the size distribution of the exosomes.Total protein was extracted from hUCMSCderived exosomes and hUCMSC supernatant and quantified by western blot.The membrane was incubated at 4°C overnight with the following primary antibodies: anti-CD63 (rabbit,1:1000,Abcam,Cat# ab134045,RRID: AB_2800495),anti-CD9 (rabbit,1:1000,Abcam,Cat# ab236630,RRID: AB_2922400),and anti-tumor susceptibility gene 101 (TSG101;mouse,1:1000,Abcam,Cat# ab83,RRID: AB_306450) (Yang et al.,2021).Then the membrane was incubated with the following secondary antibodies at room temperature for 2 hours: HRP-conjugated anti-rabbit IgG (goat,1:2000,Abcam,Cat# ab205718,RRID: AB_2819160) and HRP-conjugated anti-mouse IgG (goat,1:2000,Abcam,Cat# ab205719,RRID: AB_2755049).

Animal experiments and surgical implantation procedures

Female Sprague-Dawley (SD) rats were purchased from Beijing Vital River Laboratory Animal Technology Co.,Ltd.,Beijing,China (license No.SCXK (Jing) 2021-0006).All rats were allowed access to food and waterad libitum.All experimental animal procedures were approved by the Institutional Animal Care and Use Committee of Tianjin Medical University General Hospital (Tianjin,China) (approval No.IRB2022-DW-43,approval date: August 31,2022).Eighteen SD rats (8 weeks old,200-220 g) were randomly assigned to three experimental groups (n=6/group): PLGA@Col/HA group,PLGA@Col/HA+Exo group,and autograft group.All rats were housed in an animal room with a 12/12-hour light/dark cycle,a temperature of 18-24°C,humidity of 40-60%,and a ventilation rate of eight times per hour.To ensure the reliability of the test results,the animal room conditions were kept stable throughout the duration of the experiment.Each rat was anesthetized by inhalation of 4% isoflurane (R510-22,RWD,Shenzhen,China) until unconsciousness was achieved,and then 2% isoflurane was delivered by continuous inhalation during the surgery.The surgical procedure involved blunt dissection of the muscle via an incision made in the epidermis until the sciatic nerve was exposed.Subsequently,a 10-mm femoral segment was removed from the sciatic nerve using a scalpel to create a long-gap peripheral nerve injury.Next,the nerve gap was bridged with PLGA@Col/HA,PLGA@Col/HA+Exo,or autologous nerve inserted using 8-0 sutures.The isolated exosomes were diluted to a concentration of 10 mg/mL in PBS.PBS containing 10 µL of exosomes is injected into the PLGA@Col/HA+Exo group.Three conduits were loaded with exosomes (PLGA@Col/HA+Exo),and the remaining three conduits were loaded with an equal volume of PBS (PLGA@Col/HA) as a control.The repaired nerve,which was approximately the same length as the original uninjured nerve,was then sutured into its original position.For the autograft group,tension-free suturing was used to insert a 10-mm autograft nerve into the nerve gap.In the PLGA@Col/HA+Exo group,after suturing one end of the conduit,hUCMSC-derived exosomes were injected into the opposite end using a needle,and the free end was subsequently sutured,resulting in the formation of a relatively enclosed chamber for nerve regeneration.Finally,penicillin was administered by intramuscular injection to prevent infection after surgical procedures.

Examination of nerve electrophysiology

To determine the compound muscle action potential (CMAP),rats were anesthetized by inhalation of 4% isoflurane,while the body temperature was maintained at 37.0°C with an electric heating blanket.After the sciatic nerve was exposed on the operative side as described above,a neural needle electrode,which was used to deliver the electrical stimulation,was placed 5 mm from the proximal part of conduit suture site,and the reference electrode was placed in the gluteus maximus.For CMAP waveforms recording,two recording electrodes were inserted at the two ends of the gastrocnemius muscle.A single electrical pulse,with an intensity of 1 mA and a duration of 0.1 ms,was used to stimulate the trunk of the regenerated nerve.CMAP latency and amplitude were recorded using an electrophysiological monitoring device (YRKJ-G2008;Zhuhai Yiruikeji Co,Ltd,Guangdong,China).

Functional index testing of rat sciatic nerves

Rats were placed in the Catwalk XT System (version 10.6,Noldus,Wageningen,Netherlands) and acclimated for 3 days prior to testing motor function.Twelve weeks after surgery,the rats were recorded walking across the length of the Catwalk XT System.Sciatic function index (SFI) measurements from three steps were averaged for each rat.Some key kinematic parameters such as the print length (PL,indicates distance from the heel to the third toe),toe spread (TS,indicates distance from the first to the fifth toes),and intermediary toe spread (IT,indicates distance from the middle of the second and fourth toes) were recorded by the Catwalk XT System.The SFI values ranged from -100 to 0,with a lower value indicating poorer motor function,and were used to dynamically evaluate recovery of lower extremity muscle strength and coordination among lower extremity muscles (Shen and Zhu,1995).Kinematic parameters were collected for both the experimental and the normal hind legs.The values from the experimental side were recorded as EPL,ETS,and EIT,and the values from the normal side were recorded as NPL,NTS,and NIT.The following formula was used to calculate the SFI values: SFI=-38.3×(EPL -NPL)/NPL+109.5×(ETS -NTS)/NTS+13.3×(EIT -NIT)/NIT -8.8.

Gastrocnemius muscle assessment

The gastrocnemius muscles on the experimental side (ES) and the normal side (NS) of rats were removed and weighed after electrophysiological assessment (12 weeks after sciatic nerves surgery).The ES and NS gastrocnemius muscles were removed carefully by blunt dissection,washed with normal saline,dried on filter paper,and weighed using an electronic balance.The recovery rate of gastrocnemius muscle wet weight (Wr) is the ratio of wet weight of the muscle on the operative side (Wo) to that on the normal side (Wn).The Wr was calculated according to the following equation: Wr(%)=(Wo/Wn)×100.After weighing the muscles,the gastrocnemius tissues were then immersed in 4% PFA for no less than 24 hours.Then,after ethanol gradient dehydration,xylene,and paraffin embedding,5-µm-thick paraffin sections were made with a microtome (Leica RM2255,Köln,Germany).Next,Masson trichrome staining (Solarbio,Beijing,China) was carried out.The muscle fiber area (Sm) and collagen fiber area (Sc) were quantified using ImageJ software (version 1.8.0,NIH,Bethesda,MD,USA).The percentage of collagen fiber area (P) was calculated according to the following equation: P (%)=[Sc/(Sc+Sm)]×100.

Transmission electron microscopy analysis

Twelve weeks after surgery,the 2-mm distal regions of the sciatic nerves were collected and immersed in 2.5% glutaraldehyde solution (MilliporeSigma,Burlington,MA,USA) at 4°C for 24 hours,followed by post-fixation in 1% osmium tetroxide (MilliporeSigma) for 1.5 hours.Next,the samples were sliced into 60-nm ultrathin sections and dyed with lead citrate and uranyl acetate (MilliporeSigma).The microstructure of the nerves was assessed by transmission electron microscopy (TEM) imaging (Hitachi HT7700).Three randomly selected visual fields were photographed at 2000× magnification under the electron microscope at 80 kV,and the diameter of the myelinated axon and the thickness of the myelin sheath for all axons in each image were measured using ImageJ.

Immunofluorescence

Immunofluorescence was used to examine the biological effects of hUCMSCderived exosomes on axons.In this study,the mature primary cortical neurons were isolated from postnatal 1-day SD rats through sequential enzymatic digestion.For each group,mature primary cortical neurons were seeded in 6-well plates.The negative control group was left untreated,the neurons from the PLGA@Col/HA group were treated with 10 µL of PBS,and the neurons from the PLGA@Col/HA+Exo group were treated with 10 µL of hUCMSCderived exosomes.Then,the neurons were co-cultured separately with DMEM,PLGA@Col/HA,and PLGA@Col/HA+Exo for 7 days.On days 1 and 7 of co-culturing,the primary neurons were washed with PBS three times and fixed with 4% paraformaldehyde for 30 minutes at room temperature.Then,they were incubated with anti-beta III tubulin (Tuj-1,rabbit,1:500,Abcam,Cat# ab18207,RRID: AB_444319) overnight at 4°C.Next,the neurons were incubated with goat anti-rabbit IgG Alexa Fluor 488 (goat,1:1000,Invitrogen,Cat# A-11008,RRID: AB_143165) in the dark for 2 hours at room temperature.Images were acquired using a Leica inverted fluorescence microscope (Leica DMi8,Wetzlar,Germany),and the axonal length in each group was measured using ImageJ.

Assessment of vascularization

Gastrocnemius muscle and distal regenerated nerve tissues were collected 12 weeks after surgery.The 2-mm regenerated nerve samples were taken from the distal end of the sciatic nerve,which corresponded to the same site used for TEM,and were fixed with 4% paraformaldehyde.A 6-mm cross-section was taken and stained with anti-Laminin (rabbit,1:200,Abcam,Cat# ab7463,RRID: AB_305933) and anti-CD31 (rabbit,1:500,Abcam,Cat# ab222783,RRID: AB_2905525) primary antibodies and a goat anti-rabbit IgG Alexa Fluor 488 (goat,1:1000,Invitrogen,Cat# A-11008,RRID: AB_143165) secondary antibody.The incubation conditions and times were the same as described above.Finally,the gastrocnemius muscle fibers and nerve microvessels were observed under a fluorescence microscope (Leica),measured using ImageJ,and statistically analyzed.

Comprehensive safety assessment of the composite conduit loaded with exosomes

Upon excision of the sciatic nerve and gastrocnemius muscle,the lung,spleen,liver,and kidney were harvested and preserved in a 4% paraformaldehyde fixative solution.Subsequently,the tissues were dehydrated in an alcohol gradient (75% ethanol,85% alcohol,95% alcohol,100% alcohol,each for 2 hours),followed by embedding in paraffin and sectioning at 4-µm thickness.The morphology of each organ was evaluated via standard hematoxylin and eosin (Solarbio,Beijing,China) staining (UNIC,PRECICE 500,Suzhou,China).

Statistical analysis

No statistical methods were used to predetermine sample sizes;however,our sample sizes are similar to those reported in previous publications (Yang et al.,2023;Zhao et al.,2023).The data in this study are presented as mean ± standard deviation.Statistical differences between different multiple groups were detected by one-way analysis of variance followed by Tukey’s multiple comparisons test.Significant differences in neurite length were analyzed by two-way analysis of variance and Tukey’s multiple comparisons test.The data were analyzed using GraphPad Prism 9 software,version 9.4.1 (GraphPad Software,San Diego,CA,USA,www.graphpad.com).AP-value of 0.05 or lower was considered to indicate a statistically significant difference.

Results

Characterization of the PLGA@Col/HA conduits

The nerve conduit had a tubular structure with an outer diameter of 2.4 mm and an inner diameter of 1.2 mm,the tube wall was 0.6 mm thick,and the total length used for transplantation was 12 mm (Figure 1A).The scanning electron micrographs showed that the microporous inner Col/HA sponge and the outer electrospun PLGA nanofibrous layer formed a tubular lumen,which could provide a guide channel for axon regeneration (Figure 1B).As shown inFigure 1C,the lyophilized inner Col/HA sponge exhibited an interconnected porous microstructure,which could facilitate nutrient and metabolite transport.The pore sizes in the PLGA@Col/HA conduit ranged from 13.14 to 64.18 µm,and 77% of the pores were 18.2-49.4 µm in diameter (Figure 1D).The diameter of the electrospun PLGA nanofiber ranged from 0.13 to 0.80 µm,and 68% of the PLGA@Col/HA conduit exhibited a diameter of 0.3-0.5 µm (Figure 1EandF).To ascertain the secondary structure of the PLGA@Col/HA conduit,chemical analysis was performed by FTIR spectroscopy.Characteristic C-H (2952.1 and 2949.7 cm-1),amide A (3279.9,3261.6,and 3428.9 cm-1),amide B (2892.3,2879.8,and 2873.5 cm-1),amide I (1606.9,1629.6,and 1648.4 cm-1),amide II (1559.2,1531.7,and 1540.9 cm-1),and amide III (1308.5,1323.4,and 1343.4 cm-1) peaks were observed,indicating CH2 and CH3 groups in Col/HA and C-H and C=O bonds in PLGA,showing that Col and HA were preserved in the final conduits (Figure 1G).Analysis of the mechanical properties of the conduits showed that the ultimate tensile strength of the PLGA@Col/HA conduit was 9.47 ± 0.29 MPa,the elongation at break was 44.41 ± 3.23%,and the tensile modulus was 296.43 ± 109.25 MPa (Figure 1H).These results indicate that the PLGA shell provides sufficient mechanical support for the Col/HA sponge core.

Figure 1｜Characterization of PLGA@Col/HA conduits.

Characterization of hUCMSCs and hUCMSC-derived exosomes

To assess the ability of hUCMSCs to differentiate into multiple lineages,chondrocytes,adipocytes,and osteocytes differentiated from hUCMSCs were detected by staining with Alcian blue,Oil Red O,and alizarin red,respectively (Figure 2A).Flow cytometry analysis detected expression of the MSC phenotypic markers CD90 and CD105,whereas expression of the hematopoietic phenotypic markers CD14 and CD45 was not detected (Figure 2B).Taken together,these findings confirmed that the cells used in this experiment were human umbilical cord MSCs.TEM imaging showed that purified human umbilical cord MSC-derived exosomes exhibited a cup-shaped morphology (Figure 2C).Furthermore,nanoparticle tracking analysis (NTA) showed that the average diameter of the exosomes was 159.9 ± 56.7 nm,and the peak diameter was 136 nm (Figure 2D).To further investigate the specific surface markers of the exosomes,we performed western blot analysis,which indicated that the hUCMSC-derived exosomes expressed the exosome-specific markers CD63,CD9,and TSG101 (Figure 2E).These results indicate that we successfully isolated human umbilical cord MSCs and their corresponding exosomes.

Figure 2｜Identification of hUCMSCs and hUCMSCderived exosomes.

hUCMSC-derived exosomes promote neurite growth

Next,neurons were co-cultured with different conduits to explore the effects of the conduits and hUCMSC-derived exosomes on neurite outgrowth.Immunofluorescence staining (Tuj-1) was performed to examine the effect of hUCMSC-derived exosomes on neurites (Figure 3AandB).We found that neurites were longer on day 7 compared with day 1 of co-culture.The average neurite length of neurons incubated with hUCMSC-derived exosomes was significantly greater compared with the negative control and non-exosome group on day 1 and day 7 (P<0.0001;Figure 3C).These results suggest that the hUCMSC-derived exosomes promoted neurite elongation within 24 hours.There was no significant difference in neurite growth between the negative control group and the PLGA@Col/HA group,indicating that the conduits are not toxic to cells (P>0.05).

Loading PLGA@Col/HA nerve conduits with hUCMSC-derived exosomes improves motor function and electrophysiological muscle parameters after sciatic nerve injury

To investigate the therapeutic effects of hUCMSC-derived exosomes,we evaluated motor function after sciatic nerve injury using the Catwalk system (Figure 4A).The SFI is a widely accepted quantitative method used to assess the function of muscles innervated by the sciatic nerve through analysis of toe spread.As shown inFigure 4Band4C,the footprint parameters were clearly measured,and the results were calculated using the SFI formula.The SFI values from the PLGA@Col/HA+Exo group indicated a remarkable degree of recovery of motor function (P<0.0001),close to but still lower than that seen in the autograft group (P=0.0029),whereas this effect was not observed in the PLGA@Col/HA group (Figure 4D).To evaluate electrical conduction by the regenerated nerves,CMAP recordings were taken from anesthetized rats (Figure 4E).As shown inFigure 4Fand4G,compared with the PLGA@Col/HA group,the latency of the CMAP waveforms in the PLGA@Col/HA+Exo group was significantly improved (P=0.0005),as was the peak amplitude of the CMAP waveforms (P=0.0133).These results indicate that recovery of muscle contraction strength and speed was significantly promoted by the inclusion of hUCMSC-derived exosomes.Taken together,these findings suggest that the PLGA@Col/HA+Exo conduit improved the electroneurographic conduction of regenerated nerves and reinnervation of the gastrocnemius muscle.

Figure 4｜Effect of PLGA@Col/HA+Exo on motor function in rats with nerve defects.

Loading PLGA@Col/HA nerve conduits with hUCMSC-derived exosomes alleviates gastrocnemius muscle atrophy after sciatic nerve injury

After sciatic nerve injury,contraction of the denervated muscles becomes difficult,and this typically leads to chronic muscle atrophy.Although the gastrocnemius muscles in each group exhibited atrophy,the degree of atrophy improved over time in the PLGA@Col/HA+Exo group.The gastrocnemius muscles were harvested from both legs of rats in each group after the electrophysiological testing was complete.Next,the gastrocnemius muscles were weighed and photographed (Figure 5A).The ES gastrocnemius was visibly smaller than the NS gastrocnemius muscle in rats from the PLGA@Col/HA group,and this size difference was even more pronounced in the other two groups.There was no significant difference in the wet weight ratio between the PLGA@Col/HA+Exo group and the autograft group (P=0.0855).However,the gastrocnemius wet weight ratio was significantly greater in the PLGA@Col/HA+Exo group than in the PLGA@Col/HA group (P<0.0001;Figure 5B).As shown inFigure 5CandD,there was no significant difference in average cross-sectional muscle fiber area between the PLGA@Col/HA+Exo group and the autograft group (P=0.3739),and the values for both were significantly higher than that of the PLGA@Col/HA group (P<0.0014).Abnormal collagen fiber deposition around the muscle implies decreased motor function.Masson trichrome staining of muscle cross-sections demonstrated widespread collagen deposition around atrophied muscle fibers in the PLGA@Col/HA group (Figure 5E).The average percentage of collagen area in the PLGA@Col/HA+Exo group and the autograft group was significantly smaller than that in the PLGA@Col/HA group (Figure 5F).These results indicate that treatment with hUCMSCderived exosomes reduced abnormal deposition of collagen fibers in rats,which is consistent with the motor function recovery results.

Figure 5｜Effect of PLGA@Col/HA+Exo on gastrocnemius muscle atrophy in rats with nerve defects.

Loading PLGA@Col/HA nerve conduits with hUCMSC-derived exosomes promotes nerve regeneration and remyelination after sciatic nerve injury

Twelve weeks after surgery,the axon diameter and myelin sheath thickness of regenerated sciatic nerve fibers were examined by transmission electron microscopy (Figure 6A).In the transverse sections,clusters of small-diameter unmyelinated nerve bundles were seen around the myelinated axons,and the regenerated myelinated axons were surrounded by distinct layers of myelin sheath.In the autograft group and the PLGA@Col/HA+Exo group,the regenerated axons were wrapped in clear,thick myelin sheaths.Compared with the PLGA@Col/HA group,the diameters of the axons were significantly higher in the PLGA@Col/HA+Exo group (P=0.046;Figure 6B).As shown inFigure 6C,the myelin sheaths were thicker in the PLGA@Col/HA+Exo group than in the PLGA@Col/HA group (P=0.0483).The structure of the regenerated myelinated axons in the PLGA@Col/HA+Exo group was similar to that in the autograft group.There were no significant differences between the autograft group and the PLGA@Col/HA+Exo group in terms of either myelin thickness or diameter (P>0.05).These findings confirm that the efficacy of the PLGA@Col/HA conduit loaded with hUCMSC-derived exosomes is close to that of standard autograft in promoting axonal and myelin regeneration.

Figure 6｜Effect of PLGA@Col/HA+Exo on axonal regeneration in rats with nerve defects.

Loading PLGA@Col/HA nerve conduits with hUCMSC-derived exosomes facilitates angiogenesis in regenerated nerves

To evaluate angiogenesis in each group,immunofluorescent staining of cross-sections of the regenerated nerves with an anti-CD31 antibody was performed.InFigure 7A,the green areas indicated by white arrows are CD31+cells,and the green closed circles are blood vessels.The microvessel density for each group was calculated based on the immunofluorescence results (Figure 7B).The microvessel density was greater in the PLGA@Col/HA+Exo and autograft groups than in the PLGA@Col/HA group (P<0.0001),and there was no significant difference in microvessel density between the PLGA@Col/HA+Exo group and the autograft group (P=0.0686).Thus,the 3D composite conduit loaded with hUCMSC-derived exosomes promoted recovery after peripheral nerve injury by improving angiogenesis and vascularization,which mediate oxygen and nutrient transport.

Figure 7｜Effect of PLGA@Col/HA+Exo on neurovascular regeneration in rats with sciatic nerve injury.

Morphology of rat tissues after transplantation of PLGA@Col/HA conduit loaded with hUCMSC-derived exosomes

To verify the safety of hUCMSC-derived exosomes and PLGA@Col/HA conduitsin vivo,with a view to clinical use in the future,12 weeks after transplantation we harvested and stained sections of major rat organs.We selected organs that are prone to tumorigenesis or toxic injury,namely the spleen,liver,kidney,and lung,and subjected them to hematoxylin and eosin staining.As shown inFigure 8A,spleen sections from all groups showed normal structure.No sinus dilatation,necrosis,or inflammation were seen in the liver sections (Figure 8B).Likewise,there were no histopathological findings of congestion,necrosis,or vascular dilation in the kidney sections from all groups (Figure 8C).As expected,no tumors or abnormal nodules were seen in lung sections from any of the groups,which was probably avoided by the application of small local doses of exosomes (Figure 8D).In summary,no morphological abnormalities were found in these tissues,indicating that the hUCMSCderived exosomes and PLGA@Col/HA conduits did not induce any detectable toxicity or tumor formation.

Figure 8｜Histopathological changes in rats tissues 12 weeks after transplantation.

Discussion

The peripheral nervous system is located throughout the body,and reaches almost all tissues and organs to provide motor and/or sensory innervation (Geuna et al.,2009).At present,autologous nerve transplantation is a common treatment strategy for complete nerve transection injury (O’Brien et al.,2022).However,this traditional treatment option has some shortcomings.Long-gap nerve defects limit regenerative ability,and neurorrhaphy cannot be performed (Burks et al.,2021).In previous studies,both hollow tube conduits and porous scaffold conduits were shown to support axon regeneration to some extent,but the results were not satisfactory because they did not achieve the same level of recovery as nerve autograft (Kaplan and Levenberg,2022).Composite tissue scaffolds used to carry cells or therapeutic agents for treating nerve damage are currently a focus of clinical research.In this study,we evaluated the efficacy of a 3D composite nerve conduit loaded with hUCMSC-derived exosomes for peripheral nerve repair.This comprehensive strategy not only provides biomechanical support,but also provides a favorable microenvironment for axon extension and nerve regeneration.

Bridging the gap between proximal and distal nerve stumps with a nerve conduit can guide the growth of new axons;however,traditional hollow conduits did not contain inner structures that mimic the collagenous membrane of nerve tissues.Some studies have found that filling the inner lumen of conduits can provide better biomechanical support and therapeutic effects than using hollow conduits (Sun et al.,2017;Carvalho et al.,2019).Many previous studies have demonstrated that grafts engineered from natural biomaterials have good biocompatibility,and that their decomposition products are non-irritating (Wieringa et al.,2018).Collagen and HA are the principal constituents of the extracellular matrix,and thus hydrogels prepared from a mixture of collagen and HA have inherent advantages in mimicking the structure and function of the extracellular matrix,as well as being easy to load with biologically active substances (Xu et al.,2021).However,the mechanical properties of hydrogels are insufficient to mediate repair of injured sciatic nerves.In this study we used electrospun PLGA as the outer shell to wrap a Col-HA lyophilized sponge inner core,given that the mechanical properties of PLGA shells are similar to those of the rat sciatic nerve (Borschel et al.,2003).We found that both PLGA@Col/HA conduits and PLGA@Col/HA+Exo conduits displayed excellent biocompatibility.Hence,this type of conduit exhibits potential for clinical application and meets the requirements forin vivotransplantation.In addition,a previousin vitrostudy showed that about 78% of exosomes are released from PLGA@Col/HA conduits over 14 days,with the fastest release rate during the first 2 days.This slow-release process is beneficial for initiating and maintaining peripheral nerve regeneration (Zhao et al.,2020).

Numerous studies have demonstrated that hUCMSCs were known with the potential for neurological function restoration and gained attention as a treatment option for peripheral nerve injury (Bojanic et al.,2020;Cheng et al.,2022).Cell transplantation is a potential treatment option for repairing peripheral nerve defects.Some studies have attempted to transplant hUCMSCs into sciatic nerve lesions (Hei et al.,2017;Cui et al.,2018).However,the therapeutic effect is limited not only by the low viability of the transplanted cells but also because the volume of the conduits limits the number of cells that can be transplanted.Here,we collected hUCMSC exosomes and loaded them into the composite nerve guidance conduits to assess the effects of sustained release of hUCMSC-derived exosomes in the injured area.Crucially,we found that hUCMSC-derived exosomes loaded into PLGA@Col/HA nerve guidance conduits promote recovery of neurological and motor function in rats with complete sciatic nerve transection.Electrophysiological analysis demonstrated a decrease in latency and an increase in maximum amplitude,suggesting improvement in axonal electrical signal conduction.These results correlate well with the observed improvement in motor function.There were no significant differences in most of the nerve regeneration and functional recovery parameters between the PLGA@Col/HA+Exo and autograft groups.This suggests that the effectiveness of hUCMSC-derived exosomes approaches that of autologous nerve transplants,providing a rationale for further exploration of the underlying molecular mechanisms.

In our study,we first evaluated the multi-lineage differentiation potential of hUCMSCs and detected the specific phenotypic markers on the surfaces of the differentiated cells by flow cytometry.We successfully extracted exosomes derived from hUCMSCs,and these exosomes exhibited a cupshaped membrane structure upon TEM imaging.The morphologies and sizes of the hUCMSC-derived exosomes obtained in this study are consistent with those of hUCMSC-derived exosomes used in a previous study (Ma et al.,2021).Our results showed that neuronal neurite growth improved during 7 days of treatment with hUCMSC-derived exosomes.Our sustained-release experiment showed that,after 7 days,the conduits still contained about 66% of the exosomes that they were loaded with (Additional Figure 1).Our findings indicate that the our composite conduits have good biological safety,and that neurite growth can be effectively promoted by sustained release of hUCMSC-derived exosomes from loaded conduitsin vitro.However,it is unclear whether this approach would be effective for nerve regenerationin vivo.

On the basis of the promisingin vitroresults,we next performedin vivoexperiments in rats with sciatic nerve injury.Twelve weeks after the conduits were transplanted,the surgical area was re-exposed,which revealed no neuroma formation at the suture sites,as well as no obvious adhesion of the conduits to the surrounding soft tissue.The results from the Catwalk system analysis showed that the PLGA@Col/HA+Exo group exhibited higher SFI scores compared withPLGA@Col/HA group,close to those seen in the autograft group.Compared with the conduit-only group,the exosome group exhibited shorter latency and better amplitude in the electrophysiological test,indicating a positive therapeutic effect of hUCMSC-derived exosomes.After nerve transection,denervated muscle fibers begin to atrophy,often leading to chronic myasthenia (Gordon and Fu,2021).Our results showed that muscle weight and muscle fiber cross-sectional area also markedly recovered in the exosome group.Masson trichrome staining showed a lower proportion of collagen in the exosome and autograft groups compared with the PLGA@Col/HA group.Abnormal accumulation of collagen around the muscle is detrimental to motor function.We postulate that hUCMSCderived exosomes expedite recovery of nerve-muscle connectivity,resulting in faster recovery of muscle contraction function.Gastrocnemius muscle wet weight ratios and morphology demonstrated the efficacy of hUCMSC-derived exosome in promoting nerve regeneration.Furthermore,we observed thicker myelin sheaths and larger axon diameters in the exosome group compared with the non-exosome group.The angiogenesis results suggest that newly formed microvessels in the nerves play an important role in promoting axon regeneration.The formation of new blood vessels is believed to be a key factor in promoting a regenerative environment following an injury becuase it can facilitate improved oxygen delivery and metabolic exchange.

The dose of exosomes delivered to the injured area is key to promoting peripheral nerve regeneration after injury (Xue et al.,2017).In the case of nerve transection,intravenously injected exosomes do not reach the distal nerve stump.Intravenous administration requires many more exosomes compared with local delivery,which necessitates the use of more donor cells and leads to greater expense.The median dose used for systemic administration route is 6.75 mg EV protein/kg of body weight,whereas the local dose is 0.5 mg EV protein/kg of body weight (Gupta et al.,2021).However,some studies indicate that hUCMSC-derived exosomes contribute to the development of lung adenocarcinoma (Dong et al.,2018).To investigate whether the hUCMSC-derived exosomes cause tumor formation,hematoxylin and eosin staining was used to examine kidney,spleen,liver,and lung sections.Our results showed no morphological abnormalities in these organs.It is possible that systemic application leads to exosome enrichment in some organs,and the effects of exosome may be context-dependent;further investigation is needed to fully understand the mechanisms underlying these observations.Therefore,local application of a small dose of exosomes through our composite conduit not only reduces costs,but also lowers the risk of tumor formation.hUCMSC-derived exosomes are emerging as a promising alternative to traditional cell transplantation for the treatment of peripheral nerve injuries,and are expected to play an increasingly significant role in this field.

This study had some limitations that should be noted.First,we only measured the sciatic functional index at the 12-week timepoint,and assessing a single timepoint does not account for changes over time.This means that we were unable to detect the earliest timepoint at which the exosome-filled conduit achieved a similar recovery rate compared with the autograft group.Accordingly,we plan to examine the correlation between the degree of behavioral improvement and axon growth by taking frequent measurements at multiple timepoints in a future study.Second,in this study we only explored the effects of hUCMSC-derived exosomes on angiogenesis.Other mechanisms that improve the microenvironment,such as anti-inflammatory or anti-pyroptotic effects,should be investigated in future research.Third,owing to the differences betweenin vivoandin vitroenvironments,we were only able to determine the total amount of exosomes loaded into the conduit,and not amount of exosomes released after composite conduit implantation.Alternative approaches are needed to address this issue in the future.

In this study,we tested the efficacy of a comprehensive strategy of combining a 3D composite conduit with hUCMSC-derived exosomes for peripheral nerve repair,and found that this approach promoted nerve fiber regeneration and muscle reinnervation.Our findings highlight the clinical potential of sustained release of hUCMSC-derived exosomes at the site of peripheral nerve injuries.

Acknowledgments:The graphical abstract in this article was created with BioRender.com,and we are grateful for it.

Author contributions:HT,ZW and SF conceived the project,designed theexperiments.HT and JL conducted surgery and behavioral tests.HW and JL reviewed the data and provided recommendations for the image analysis.JR,HD,JS and MW performed literature search and manuscript preparation.All authors read and approved the final manuscript.

Conflicts of interest:The authors declare that they have no conflicts of interest.

Data availability statement:All relevant data are within the paper and its Additional files.

Open access statement:This is an open access journal,and articles are distributed under the terms of the Creative Commons AttributionNonCommercial-ShareAlike 4.0 License,which allows others to remix,tweak,and build upon the work non-commercially,as long as appropriate credit is given and the new creations are licensed under the identical terms.

Additional file:

Additional Figure 1:Sustained release of human umbilical cord mesenchymal stem cell-derived exosomes from the composite conduit.