SCIENTIFIC ARTICLE

The Effect of Chondroitinase on Nerve Regeneration

Following Composite Tissue Allotransplantation

Sami Tuffaha, MD, Meghan Quigley, BS, Timothy Ng, BS, Vijay S. Gorantla, MD, PhD,Jaimie T. Shores, MD, Benson Pulikkottil, MD, Christopher Shestak, BS, Gerald Brandacher, MD,

W. P. Andrew Lee, MD

Purpose To improve the degree of functional return and sensibility provided by compositetissue allotransplantation, enhanced nerve regeneration is essential. Chondroitin sulfateproteoglycans are found in the extracellular matrix of nerves and inhibit regenerating axonsafter injury. Treatment with chondroitinase to remove chondroitin sulfate proteoglycans hasbeen shown to improve nerve regeneration in isolated nerve graft and transection-and-repairmodels. This study assesses the efficacy of chondroitinase as a neurotherapeutic agent in thesetting of composite tissue allotransplantation.

Methods Adult Lewis rats received either orthotopic hind limb transplants from BrownNorway rat donors (n � 12) or sciatic nerve transection and repair (n � 6). Followingapproximation of the sciatic nerve, half the animals received intraneural injections ofchondroitinase in saline and the other half received intraneural injections of saline alone.Five weeks after transplantation, we killed the animals and analyzed nerves with nonbiasedquantitative nerve histomorphometry. One day after transection and repair, we killed animalsand harvested sciatic nerves for immunohistochemical staining of cleaved chondroitin sulfateproteoglycans epitope and laminin. We used unpaired t-tests for statistical analysis.

Results Distal to the suture line, chondroitinase-treated animals demonstrated statisticallygreater total number of fibers and nerve density compared with controls. There were nostatistically significant differences in fiber number or nerve density proximal to the sutureline or in fiber widths. We observed staining of cleaved chondroitin sulfate proteoglycanepitopes only in treated animals, with no differences observed in the degree of lamininstaining.

Conclusions Intraneural injection of chondroitinase cleaved inhibitory chondroitin sulfateproteoglycans without disrupting proregenerative laminin and resulted in enhanced nerveregeneration after composite tissue allotransplantation. Studies at later time points are neededto assess whether this enhanced nerve regeneration will produce improved functional return.

Clinical relevance This study identifies a promising therapeutic agent that has the potential toenhance nerve regeneration and functional outcomes for patients undergoing compositetissue allotransplantation. (J Hand Surg 2011;36A:1447–1452. Copyright © 2011 by theAmerican Society for Surgery of the Hand. All rights reserved.)

Key words Chondroitinase, nerve, regeneration, transplantation.

From the Division of Plastic and Reconstructive Surgery, University of Pittsburgh, Pittsburgh, PA; and theDepartment of Plastic and Reconstructive Surgery, Johns Hopkins University School of Medicine, Balti-more, MD.

Received for publication January 25, 2011; accepted in revised form June 3, 2011.

No benefits in any form have been received or will be received related directly or indirectly to the

Corresponding author: Gerald Brandacher, MD, Department of Plastic and Reconstructive Surgery,Johns Hopkins University School of Medicine, Ross 749D, 720 Rutland Avenue, Baltimore, MD 21205;e-mail: brandacher@jhmi.edu.

0363-5023/11/36A09-0002$36.00/0doi:10.1016/j.jhsa.2011.06.007

subject of this article.

© ASSH � Published by Elsevier, Inc. All rights reserved. � 1447

1448 CHONDROITINASE AND NERVE REGENERATION

WITH 67 HAND TRANSPLANTS performed since1998, composite tissue allotransplantation(CTA) has emerged as a means of restoring

function and quality of life to upper extremity ampu-tees. Because CTA is elective and carries serious risksassociated with immunosuppression and rejection, im-proving the risk-benefit ratio is essential to enhancingthe viability of CTA as a treatment option. Functionaloutcomes after hand transplantation have been encour-aging1–4 but there is still much room for improvement.Whereas most agree that the main factor limitingbroader clinical application of hand transplantation isthe need for chronic immunosuppression,5 improvedfunctional outcomes would likely make the risks and costsassociated with CTA more acceptable. Therefore, in addi-tion to continued efforts to minimize the risks associatedwith immunosuppression and rejection, more focus mustbe placed on augmenting the degree of motor function andsensibility that a hand transplant recipient can achieve.

Unlike in solid organ transplantation, composite tis-sue graft function depends on adequate host nerve re-generation into the graft. Axons, however, regenerateslowly relative to the distance they must travel, andirreversible motor end plate denervation can occur be-fore axons reach their targets. Therefore, to maximizegraft function, the goal must be to optimize the quantity,quality, and speed of graft end organ innervation byhost neurons.

Chondroitinase is a promising agent in this regard, asit has been shown to promote nerve regeneration innerve repair models.6–8 Chondroitinase is a glycosidasewith strong specificity for glycosaminoglycans. Themechanism by which it enhances nerve regeneration isthought to be through inactivation of chondroitin sulfateproteoglycans (CSPGs), which are known to be majorinhibitory regulators of axonal regeneration after nerveinjury.9–11 Chondroitin sulfate proteoglycans promoteaxon death and block access to Schwann cell (SC) basallamina tubes and proregenerative laminin molecules.12

After peripheral nerve injury, CSPGs are rapidly pro-duced by SCs and macrophages within the nerve distalto the injury site.13 Simultaneously, during the processof Wallerian degeneration, SCs and macrophages mi-grating into the distal nerve express metalloproteinases,which degrade the inhibitory CSPGs, thus facilitatingaxonal regeneration.13–17 It is postulated that theCSPGs are upregulated throughout the nerve matrix butare only degraded within the SC basal lamina tubes,thus preventing aberrant axonal growth outside basallamina tubes and facilitating proper targeting of regen-erating axons within the nerve ultrastructure.8,18 The

aim of this study was to test the hypothesis that intra-

JHS �Vol A, Se

neural injection of chondroitinase during limb trans-plantation degrades CSPGs while sparing laminin, re-sulting in enhanced regeneration of host neurons intothe donor limb nerve.

MATERIALS AND METHODS

Animals

We obtained adult 8- to 12-week-old male Lewis(LEW; RT11) and Brown Norway (BN; RT1n) ratsfrom Harlan Sprague Dawley (Indianapolis, IN). Allanimals were housed in a pathogen-free animal facilityand maintained in accordance with Institutional AnimalCare and Use Committee and the National Institutes ofHealth and national law guidelines on the care and useof laboratory animals.

Hind limb transplantation

We used a rat orthotopic hind limb transplant model totest the effects of chondroitinase treatment on nerveregeneration after CTA. Animals were anesthetizedwith 40 mg/kg pentobarbital intraperitoneally. All ani-mals in the control group and experimental group (n �6 per group) received orthotopic hind limb transplanta-tion from BN to LEW, as previously described, withend-to-end anastomosis of the femoral vessels and sci-atic nerve.19 After re-establishing blood flow, the hostand donor sciatic nerve stumps were tightly approxi-mated with epineurial neurorrhaphy using 10-0 nylonsuture. Each animal of the experimental group thenreceived an injection of 1.5 U/20 �L chondroitinaseABC in vehicle (1% bovine serum albumin in phos-phate-buffered saline) directly into the sciatic nerve atthe suture line using a 30-gauge needle. The animals ofthe control group received intraneural injections of 20�L vehicle alone. The muscle and skin were thenclosed with 4-0 Vicryl (Ethicon, Somerville, NJ) andnylon suture, respectively. All animals received FK-506monotherapy (1 mg/kg per day intraperitoneally) for 5weeks until they were killed to prevent allograft rejec-tion.

Nerve transection and repair

We used a nerve transection-and-repair model (LEW;n � 3 per group) to determine the efficacy andselectivity of enzymatic degradation by chondroiti-nase treatment. Animals were anesthetized with 40mg/kg pentobarbital intraperitoneally. With the ani-mals in the prone position, the sciatic nerve on theright was exposed with a muscle-splitting incisionand sharply transected with microsurgical scissors.The 2 nerve stumps were then tightly reapproximated

with epineurial neurorrhaphy using 10-0 nylon su-

ptember

CHONDROITINASE AND NERVE REGENERATION 1449

ture. Each animal in the experimental group receivedan injection of 1.5 U/20 �L chondroitinase ABC invehicle (1% bovine serum albumin in phosphate-buffered saline) directly into the sciatic nerve at thesuture line using a 30-gauge needle, and the animalsof the control group received intraneural injections of20 �L vehicle alone. The concentration, volume, anddelivery technique of chondroitinase into the repairednerves were identical to those used in the hind limbtransplant model. There were no adverse events as-sociated with the survival surgeries or immunosup-pressive therapy.

Nerve harvest

At 5 weeks (orthotopic hind limb transplant) or 1 day(nerve transection and repair), animals were anesthe-tized with 40 mg/kg pentobarbital intraperitoneally. Thesciatic nerve was harvested and placed in cold fixativein preparation for analysis. The animals were thenkilled by cervical dislocation.

Immunohistochemistry

Sciatic nerves harvested 1 day after transection andrepair were fixed in 3% paraformaldehyde overnight,mounted in paraffin, and sectioned just distal to thesuture line.

To assess for the presence of intact laminin, westained all samples with polyclonal rabbit anti-ratlaminin antibody (Sigma Aldrich, St. Louis, MO) di-luted 1:25. To assess for the presence of the cleavedchondroitin sulfate epitope, all samples were stainedwith mouse anti-rat chondroitin-4 sulfate antibody (Mil-lipore, Billerica, MA) diluted 1:50. Stained sectionswere evaluated by light microscopy and qualitativelyanalyzed by a blinded observer.

Quantitative nerve histomorphometry

We analyzed nerves harvested from animals undergo-ing orthotopic hind limb transplantation by quantitativenerve histomorphometry to measure regenerating my-elinated axons. After fixation in a cold, buffered solu-tion of 3% glutaraldehyde, the sciatic nerves were cut at2 sites: 0.5 cm proximal and 0.5 cm distal to the sutureline. The nerve segments were then postfixed with 1%osmium tetroxide, dehydrated in ethanol, and embed-ded in Epon resin (Miller-Stephenson; Sylmar, CA).We cut 1-�m sections and stained them for myelin with1% toluidine blue for light microscopy. We examinedcross-sections using a digital microscopy image-analy-sis system linked to morphometric software (IA32;Leco Instruments, St. Joseph, MI).20 For each nervesection, a blinded observer analyzed 6 randomly se-

lected fields at � 1,000 magnification.

JHS �Vol A, Se

Statistics

Using binary imaging software, we measured fibercount and overall nerve area. Total fiber number wascalculated as [(fibers/high powered field)/(total nerve area/field constant)], where the field constant is 7,095.278�m. Fiber density was calculated by the total fibernumber per total nerve area. We compared histomor-phometric parameters between groups using the un-paired t-test.

RESULTS

Quantitative nerve histomorphometry

Assessment of histological sections distal to the sutureline demonstrated significantly greater total fiber num-ber in chondroitinase-treated animals compared withcontrols (chondroitinase: 13,802 � 6,522; control:2,290 � 860; P � .01, unpaired t-test). There was nosignificant difference in fiber number proximal to thegraft (chondroitinase: 9,945 � 1,441; control: 8,877 �2,103; P � .20, unpaired t-test).

Nerve fiber density (fibers per square millimeter)distal to the suture line was also greater in the chon-droitinase-treated group compared with the controlgroup (chondroitinase: 12,160 � 4,588; control:2,459 � 1,198; P � .01, unpaired t-test). No signif-icant difference was seen in nerve fiber density prox-imal to the suture line (chondroitinase: 13,456 �1,441; controls � 12,325 � 1,287; P � .24, unpairedt-test). Figure 1 shows representative toluidine bluestaining for myelin.

No significant differences were observed in fiberwidth either proximal or distal to the suture line.

Laminin staining

Immunohistochemical staining for laminin demon-strated relative maintenance of the basal laminin in thechondroitinase-treated nerves compared with controls(Fig. 2A, B).

Cleaved chondroitin sulfate epitope staining

Positive immunohistochemical staining for the cleavedchondroitin sulfate epitope in the chondroitinase-treatedgroup confirmed digestion of chondroitin sulfate side-chains on CSPGs (Fig. 2C). No staining was observedin the untreated controls.

DISCUSSIONThe results of this study suggest that intraneural admin-istration of chondroitinase degrades CSPGs and in ahind limb transplant model produces increased axonalregeneration. These results are consistent with the hy-

pothesis that artificial degradation of CSPGs bypasses

ptember

erve

1450 CHONDROITINASE AND NERVE REGENERATION

the time normally required for physiological degrada-tion of CSPGs, allowing robust, rapid axonal regener-ation. This is important because the speed of axonalregeneration is thought be the most important determi-nant of functional outcomes.21 As time elapses, dener-vated skin, muscle, and SC undergo atrophic changesthat prevent optimal reinnervation.21–24 Therefore, bydecreasing the latency of axonal regeneration and end-organ reinnervation, chondroitinase treatment has thepotential to positively affect functional outcomes afterlimb transplantation.

To assess the effects of enzymatic degradation, weused a nerve transection-and-repair model to approxi-mate the hind limb transplant model. The surgical ap-proximation of nerve stumps as well as the concentra-tion, volume, and delivery method of chondroitinasewere identical in both models. Because analysis wasdone at 1 day postoperatively, there would not yet havebeen an alloimmune response in a hind limb transplantmodel that would cause the intraneural environment todiffer from that of a nerve transection-and-repair model.Our immunohistochemical data confirmed the specific-ity of chondroitinase for CSPGs. We observed CSPGdegradation in treated nerves as evidenced by positivestaining for the cleaved chondroitin sulfate epitope.Importantly, we also observed qualitatively similarstaining for laminin in treated and untreated nerves,indicating sparing of this important proregenerative gly-coprotein. Nerve regeneration depends largely on bind-ing of infiltrating SCs and axons to laminin within thebasal lamina tubes.12,25,26 There is a balance betweeninhibitory cues from compounds such as CSPGs and

FIGURE 1: Representative nerve histology distal to suture line,(scale bars � 15 �m; magnification �100). A greater total numin the chondointinase-treated nerve compared with the control n

growth-promoting compounds such as laminin that ul-

JHS �Vol A, Se

timately dictate whether adequate regeneration will oc-cur. It has been shown that a high enough concentrationof laminin can overwhelm the inhibitory effects ofCSPGs.27 Because chondroitinase inactivates CSPGswithout degrading laminin, it effectively tips the bal-ance between inhibition and promotion toward im-proved regeneration.

Our histomorphometric data demonstrated signifi-cantly greater total fiber number and fiber density inchondroitinase-treated animals undergoing hind limbtransplantation compared with untreated controls. Pre-vious studies in a similar nerve transection-and-repairmodel also demonstrated enhanced nerve regenerationwith chondroitinase treatment6,8; nevertheless, the dis-parities observed between treatment and control groupsin this study were substantially more pronounced. Thegreater differences could be explained by the greatervolume of vehicle used in this study. Whereas previousstudies used 1 U chondroitinase in 2 �L vehicle,6,8 ourstudy used 1.5 U in 20 �L. Despite the lower concen-tration, our immunohistochemical data demonstratedthorough digestion of CSPGs compared with controls.At the same time, the greater volume of vehicle used inour study may have allowed for greater drug infiltrationthrough the length of the nerve. During intraoperativeinjection of chondroitinase, we observed ballooning ofthe entire visible length of the nerve both proximal anddistal to the injection site. In a study by Zuo et al8 using1 U/2 �L, the authors observed a large difference infiber number between treatment and control groupsnear the suture line, but the differences became minimal

ed with toluidine blue for myelin: A chondroitinase, B controland density of regenerating myelinated axons can be observed5 weeks after limb transplantation.

stainber

moving distally through the nerve. It is possible that

ptember

CHONDROITINASE AND NERVE REGENERATION 1451

those findings were due to diminished drug delivery inthe distal nerve, given the small volume of the vehicle.

Although we found evidence of pronounced en-hancement of axonal regeneration in chondroitinase-treated animals, it is unclear whether this will ultimatelytranslate into improved function. One possibility is thatthe greater axon counts in the treated animals resultedfrom a burst of nonproductive axonal sprouting near theinjury site. Strong end-organ reinnervation depends onsustained growth of the main axonal processes frommultiple cell bodies, and we cannot rule out the possi-bility that our histomorphometric findings were theresult of abundant but short-lived sprouting originatingfrom a limited number of cell bodies. Our treatmentprotocol involved 1 injection of chondroitinase at the

FIGURE 2: Immunohistochemistry. A, B Scale bars � 10 �m,A Chondroitinase-treated nerves, stained for laminin. B Controlor distribution of laminin staining between the chondroitinase-tC Positive staining for chondroitin-4 sulfate (cleaved chontransection and repair (not seen in control nerves).

time of transplantation, but there may be continued

JHS �Vol A, Se

production of inhibitory CSPGs in the distal denervatednerve.28 For this reason, it may prove beneficial todevise a protocol for subsequent chondroitinase injec-tion after initial transplantation to allow sustained re-generation. We must also consider the possibility thatthe removal of CSPGs from the extracellular matrixwill result in poorer targeting of the axons to theirproper end-organ sites. The physiological functionserved by inhibitory CSPGs is not fully clear, but theylikely serve as a means of regulating and directingaxonal regeneration. Were it the case that axonal tar-geting is jeopardized by chondroitinase treatment, thiscould adversely affect functional outcomes.

We chose an early time point for this initial studybecause early time points are crucial for detecting his-

nification �100. C Scale bars � 100 �m, magnification �10.ned for laminin. No difference can be observed in the intensityd nerves and control nerves 1 day after transection and repair.in epitope) in the chondroitinase-treated nerve 1 day after

mag, staireatedroit

tological differences in nerve regeneration when using a

ptember

1452 CHONDROITINASE AND NERVE REGENERATION

rodent model. At longer end points, it has been shownthat the superlative regenerative capacities of rodentswill mask real differences between treatment groups,with axonal counts eventually equalizing over time.29

Having demonstrated early histological evidence of im-proved nerve regeneration, for the reasons mentionedearlier, follow-up studies using longer time points arenow necessary to assess functional return. However, theresults from this study serve to highlight the promise ofchondroitinase as an agent to enhance nerve regenera-tion and functional outcomes after hand transplantation.

REFERENCES1. Lanzetta M, Petruzzo P, Margreiter R, Dubernard JM, Schuind F,

Breidenbach W, et al. The International Registry on Hand andComposite Tissue Transplantation. Transplantation 2005;79:1210–1214.

2. Lanzetta M, Petruzzo P, Dubernard JM, Margreiter R, Schuind F,Breidenbach W, et al. Second report (1998–2006) of the Interna-tional Registry of Hand and Composite Tissue Transplantation.Transpl Immunol 2007;18:1–6.

3. Ravindra KV, Buell JF, Kaufman CL, Blair B, Marvin M,Nagubandi R, et al. Hand transplantation in the United States:experience with 3 patients. Surgery 2008;144:638–643; discussion643–634.

4. Petruzzo P, Lanzetta M, Dubernard JM, Landin L, Cavadas P,Margreiter R, et al. The International Registry on Hand and Com-posite Tissue Transplantation. Transplantation 2010;90:1590–1594.

5. Mathes DW, Schlenker R, Ploplys E, Vedder N. A survey of northamerican hand surgeons on their current attitudes toward hand trans-plantation. J Hand Surg 2009;34A:808–814.

6. Graham JB, Neubauer D, Xue QS, Muir D. Chondroitinase appliedto peripheral nerve repair averts retrograde axonal regeneration. ExpNeurol 2007;203:185–195.

7. Neubauer D, Graham JB, Muir D. Chondroitinase treatment in-creases the effective length of acellular nerve grafts. Exp Neurol2007;207:163–170.

8. Zuo J, Neubauer D, Graham J, Krekoski CA, Ferguson TA, Muir D.Regeneration of axons after nerve transection repair is enhanced bydegradation of chondroitin sulfate proteoglycan. Exp Neurol 2002;176:221–228.

9. Muir D, Engvall E, Varon S, Manthorpe M. Schwannoma cell-derived inhibitor of the neurite-promoting activity of laminin. J CellBiol 1989;109:2353–2362.

10. Zuo J, Neubauer D, Dyess K, Ferguson TA, Muir D. Degradation ofchondroitin sulfate proteoglycan enhances the neurite-promoting po-tential of spinal cord tissue. Exp Neurol 1998;154:654–662.

11. Carulli D, Laabs T, Geller HM, Fawcett JW. Chondroitin sulfateproteoglycans in neural development and regeneration. Curr OpinNeurobiol 2005;15:116–120.

12. Salonen V, Peltonen J, Roytta M, Virtanen I. Laminin in traumatizedperipheral nerve: basement membrane changes during degeneration

and regeneration. J Neurocytol 1987;16:713–720.

JHS �Vol A, Se

13. Zuo J, Hernandez YJ, Muir D. Chondroitin sulfate proteoglycan withneurite-inhibiting activity is up-regulated following peripheral nerveinjury. J Neurobiol 1998;34:41–54.

14. Tomita K, Kubo T, Matsuda K, Yano K, Tohyama M, Hosokawa K.Myelin-associated glycoprotein reduces axonal branching and en-hances functional recovery after sciatic nerve transection in rats. Glia2007;55:1498–1507.

15. Braunewell KH, Martini R, LeBaron R, Kresse H, Faissner A,Schmitz B, et al. Up-regulation of a chondroitin sulphate epitopeduring regeneration of mouse sciatic nerve: evidence that the immu-noreactive molecules are related to the chondroitin sulphate pro-teoglycans decorin and versican. Eur J Neurosci 1995;7:792–804.

16. Ferguson TA, Muir D. MMP-2 and MMP-9 increase the neurite-promoting potential of schwann cell basal laminae and are upregu-lated in degenerated nerve. Mol Cell Neurosci 2000;16:157–167.

17. Zuo J, Ferguson TA, Hernandez YJ, Stetler-Stevenson WG, Muir D.Neuronal matrix metalloproteinase-2 degrades and inactivates a neu-rite-inhibiting chondroitin sulfate proteoglycan. J Neurosci 1998;18:5203–5211.

18. Dou CL, Levine JM. Inhibition of neurite growth by the NG2chondroitin sulfate proteoglycan. J Neurosci 1994;14:7616–7628.

19. Sacks JM, Kuo YR, Taieb A, Breitinger J, Nguyen VT, ThomsonAW, et al. Prolongation of composite tissue allograft survival byimmature recipient dendritic cells pulsed with donor antigen andtransient low-dose immunosuppression. Plast Reconstr Surg 2008;121:37–49.

20. Hunter DA, Moradzadeh A, Whitlock EL, Brenner MJ, MyckatynTM, Wei CH, et al. Binary imaging analysis for comprehensivequantitative histomorphometry of peripheral nerve. J Neurosci Meth-ods 2007;166:116–124.

21. Guth L, Albers RW, Brown WC. Quantitative changes in cholines-terase activity of denervated muscle fibers and sole plates. ExpNeurol 1964;10:236–250.

22. Sunderland S. Factors influencing the course of regeneration and thequality of the recovery after nerve suture. Brain 1952;75:19–54.

23. Terenghi G, Calder JS, Birch R, Hall SM. A morphological study ofSchwann cells and axonal regeneration in chronically transectedhuman peripheral nerves. J Hand Surg 1998;23B:583–587.

24. Hoffer JA, Stein RB, Gordon T. Differential atrophy of sensory andmotor fibers following section of cat peripheral nerves. Brain Res1979;178:347–361.

25. Lander AD, Fujii DK, Reichardt LF. Laminin is associated with the“neurite outgrowth-promoting factors” found in conditioned media.Proc Natl Acad Sci U S A 1985;82:2183–2187.

26. Wang GY, Hirai K, Shimada H. The role of laminin, a component ofSchwann cell basal lamina, in rat sciatic nerve regeneration withinantiserum-treated nerve grafts. Brain Res 1992;570:116–125.

27. Snow DM, Letourneau PC. Neurite outgrowth on a step gradient ofchondroitin sulfate proteoglycan (CS-PG). J Neurobiol 1992;23:322–336.

28. Hoke A. Proteoglycans in axonal regeneration. Exp Neurol 2005;195:273–277.

29. Brenner MJ, Moradzadeh A, Myckatyn TM, Tung TH, Mendez AB,Hunter DA, et al. Role of timing in assessment of nerve regeneration.

Microsurgery 2008;28:265–272.

ptember