World J Urol (2008) 26:333–339 DOI 10.1007/s00345-008-0283-y
TOPIC PAPER
Cavernous nerve regeneration using acellular nerve grafts Stephen S. Connolly · James J. Yoo · Mohamed Abouheba · Shay Soker · W. Scott McDougal · Anthony Atala
Received: 7 April 2008 / Accepted: 19 May 2008 / Published online: 2 July 2008 © Springer-Verlag 2008
Abstract Introduction The restoration of erectile function following complete transection of nerve tissue during surgery remains challenging. Recently, graft procedures using sural nerve grafts during radical prostatectomy have had favorable outcomes, and this has rekindled interest in the applications of neural repair in a urologic setting. Although nerve repair using autologous donor graft is the gold standard of treatment currently, donor nerve availability and the associated donor site morbidity remain a problem. In this study, we investigated whether an “oV-the-shelf” acellular nerve graft would serve as a viable substitute. We examined the capacity of acellular nerve scaVolds to facilitate the regeneration of cavernous nerve in a rodent model. Materials and methods Acellular nerve matrices, processed from donor rat corporal nerves, were interposed across nerve gaps. A total of 80 adult male Sprague-Dawley rats were divided into four groups. A 0.5-cm segment of cavernosal nerve was excised bilaterally in three of the four groups. In the Wrst group, acellular nerve segments were inserted bilaterally at the defect site. The second group underwent autologous genitofemoral nerve grafts at the
S. S. Connolly · J. J. Yoo (&) · M. Abouheba · S. Soker · A. Atala Department of Urology, Wake Forest Institute for Regenerative Medicine, Medical Center Boulevard, Winston-Salem, NC 27157, USA e-mail:
[email protected]
same site, and the third group had no repair. The fourth group underwent a sham procedure. Serial cavernosal nerve function assessment was performed using electromyography (EMG) at 1 and 3 months following initial surgery. Histological and immunocytochemical analyses were performed to identify the extent of nerve regeneration. Results Animals implanted with acellular nerve grafts demonstrated a signiWcant recovery in erectile function when compared with the group that received no repair, both at 1 and 3 months. EMG of the acellular nerve grafts demonstrated adequate intracavernosal pressures by 3 months (87.6% of the normal non-injured nerves). Histologically, the retrieved regenerated nerve grafts demonstrated the presence of host cell inWltration within the nerve sheaths. Immunohistochemically, antibodies speciWc to axons and Schwann cells demonstrated an increase in nerve regeneration across the grafts over time. No organized nerve regeneration was observed when the cavernous nerve was not repaired. Conclusion These Wndings show that the use of nerve guidance channel systems allow for accelerated and precise cavernosal nerve regeneration. Acellular nerve grafts represent a viable alternative to fresh autologous grafts in a rodent model of erectile dysfunction. Keywords Nerve regeneration · Cavernous nerve · Erectile dysfunction · Radical prostatectomy · Acellular tissue scaVold
Introduction W. S. McDougal Department of Urology, Massachusetts General Hospital and Harvard Medical School, 55 Fruit St. GRB1102, Boston, MA 02114, USA e-mail:
[email protected]
Current statistics show that cancer of the prostate accounts for approximately 29% of all cancers diagnosed in men and approximately 30% of these men eventually require radical
123
334
prostatectomy [1, 2]. While radical prostatectomy is eVective in eradicating the primary tumor, procedural complications including erectile dysfunction are prevalent after this surgery. A population-based study indicates that erectile dysfunction occurs in 82 and 79% of patients undergoing radical prostatectomy at 2 and 5 years post-surgery, respectively, leading to a decrease in quality of life for these patients [3].To this end, nerve-sparing surgical techniques have been developed in an attempt to preserve sexual function, but it has been found that even when the cavernous nerves are preserved, nerve damage may still occur [4, 5]. Inevitably, there is some trauma to the nerves during surgery, and recent experimental studies have determined that simply exposing the cavernous nerve can lead to erectile dysfunction [6]. Therefore, methods of restoring nerve function in these patients are required. Recent investigations show that restoration of autonomic erectile function can be achieved successfully with autologous sural nerve grafts to the divided neurovascular bundles during radical prostatectomy [7, 8]. However, use of fresh sural nerve requires a second operative procedure and the sacriWce of a healthy functioning nerve, with inherent morbidity [9, 10]. The rapidly evolving Weld of tissue engineering may provide one solution, as the use of alternative, bioengineered nerve grafts could avoid such morbidity and minimize risk of associated complications. Current strategies for developing tissue engineered constructs for nerve regeneration generally require the production of a nerve conduit or scaVold that can guide and support axonal regeneration [11]. This scaVold may be the most important constituent of nerve regeneration strategy [11]. The scaVold is believed to provide a guide for promoting the growth of the regenerating axon [12]. Regenerating axons migrate from the proximal cut surface distally, and use the scaVold surface as a tactile stimulus upon which to move. Schwann cells may also use the scaVold to migrate in the opposite direction, from distal to proximal, to myelinate the regenerating axons [13]. Numerous biomaterials have been used to facilitate nerve regeneration. The ideal biomaterial for nerve regeneration should have appropriate physical and mechanical properties. It should also be non-immunogenic and biocompatible, promote tissue development and appropriate cellular interaction, and be non-toxic, biodegradable and bioresorbable [14]. Both natural and synthetic materials have been used as nerve guidance channels for regeneration. Synthetic materials, such as the polymers poly (L-glycolic) acid (PGLA), and poly (L-lactic) acid (PLLA), have advantages of controlling fabrication and physical properties. However, these materials result in limited success due to various reasons, including lack of biocompatibility [15–18]. Naturally derived materials such as Wbronectin, laminin, and collagen are also in use, and these have the advantage of superior biocompatibility [11, 19].
123
World J Urol (2008) 26:333–339
Acellular matrices derived from donor tissues are commonly used in many tissue engineering applications [12, 20, 21]. These are primarily collagen-rich scaVolds produced through decellularizing a particular tissue, which is a process of removing cellular elements from donor tissues, leaving only extracellular matrices. As such, decellularized nerve matrix has been used in many peripheral nerve applications experimentally. This matrix is known to be biocompatible and provides a microarchitecture inherently native to axons. In addition, this type of matrix displays reduced immunogenicity, which results from the removal of the cellular component of any tissue [22]. Acellular nerve grafts have been reported to be beneWcial in the repair of sciatic nerve defects, however, there are no reports of their use in this autonomic pelvic setting [22]. In this study, we investigated the feasibility of using acellular nerve scaVolds obtained and processed from donor cavernous nerve tissue as an alternative to autografting in the treatment of erectile dysfunction. We examined the capacity of acellular nerve scaVolds to facilitate the regeneration of autonomic nerve in a rodent model.
Materials and methods Acellular nerve grafts All cavernous nerve grafts were obtained from rat cadaveric donors. The grafts were decellularized according to a protocol previously used in our laboratory for a variety of tissue types, including the nerve [23]. BrieXy, the nerves were washed repeatedly with cold normal saline upon harvesting. Nerves were then immersed and agitated in decellularization medium [0.5% (v/v) Triton X100, 47.6 mM ammonium hydroxide] for 72 h at 4°C. After decellularization, the nerve segments were washed again repeatedly with distilled water, frozen at ¡85°C and lyophilized. Histological examination conWrmed the absence of cells on the resulting collagen based nerve matrix grafts. The grafts were sterilized by a cool cycle of ethylene oxide. Prior to implantation, the nerve scaVolds were soaked overnight in standard Dulbecco’s ModiWed Eagle’s Medium (DMEM) at 37°C. The scaVolds were cut to 5 mm immediately prior to implantation (Fig. 1a). All chemicals and media were purchased from Sigma Chemical Co., St. Louis, MO, USA. Animal studies A total of 80 Sprague-Dawley rats (Charles River, Inc., Boston, MA) weighing approximately 250 g were used in this study under a protocol approved by the Institutional Animal Care and Use Committee. Regimens of 12 h light and 12 h darkness were followed daily.
World J Urol (2008) 26:333–339
335
Fig. 1 a Processed acellular nerve graft. b Schematic diagram depicting cavernosal nerve
Experimental groups The rats were allocated evenly among four study groups (A–D). All rats underwent bilateral surgical dissection of the cavernous nerve. In groups A, B and C, the cavernous nerve was divided and a 5 mm segment on each side was excised. Group A underwent microsurgical repair of the 5 mm defect using the prepared acellular allografts. In group B, fresh autologous genitofemoral nerve was harvested as part of the surgical procedure, and this nerve tissue was interposed into the cavernous nerve defect as the primary method of repair. In group C (negative control) there was no attempt to repair the defect leaving a gap between the stumps. In group D (sham control), continuity of the nerve was not disrupted in any way during the procedure.
end-to-end fashion using a taper needle and 10-nylon sutures (Ethilon, Somerville, NJ). One simple suture was used per anastomosis. An operating microscope (12.5£, Olympus Corporation, Tokyo) was used for all steps of the procedure. All animals were given a single intramuscular dose of a broad-spectrum antibiotic (cephalosporin) at the end of the procedure. All animals were followed for either 1 or 3 months. At each time point, a second surgical procedure was performed in which the erectile response was quantiWed by measurement of intracorporal pressure (ICP). The erectile response was induced by bilateral electrical stimulation of the cavernous nerve, proximal to the site of injury. The results were compared with the positive sham control group (D). The entire nerve was excised prior to animal sacriWce. Intracorporal pressure monitoring
Surgical techniques Surgical procedures were performed under sterile conditions. Anesthesia consisting of medetomidine (300 g/kg, Orion Corp., Espoo, Finland) and ketamine (30 mg/kg, Phoenix Pharm. Inc., St. Joseph, MO) was administered intramuscularly. Each animal was placed in the supine position on a regulated thermal heating pad. A lower midline (umbilicus to pubis) incision was made and the peritoneal cavity was exposed. The testes were removed from the scrotum through the inguinal canal. The scrotal attachment was divided, and the testes were wrapped in moist swabs. The space of Retzius was then exposed on each side. The cavernous nerves were identiWed easily in close proximity to the dorsolateral lobe of the prostate (Fig. 1b). Electrical stimulation of each cavernous nerve was performed after minimal dissection using a custom-made bipolar electrode (Harvard Apparatus Inc., Holliston, MA) which was applied directly to the cavernous nerve. Visual veriWcation of a tumescent penile response was taken as adequate evidence of pre-operative potency. The nerves were then severed creating a defect sized 5 mm in length. The animals in each group were treated with their respective regimen described above. All nerveto-graft anastomoses were accomplished in a delicate
During the initial surgery, the inferior aspect of the midline incision was extended to expose the base of the penis to allow visual veriWcation of tumescence. For surgeries performed at 1 or 3 months, the incision was extended inferiorly and the penis was circumcised and degloved. A heparinized (250 /ml) 25G ¾⬙ pediatric cannula was inserted into the left cavernous body just proximal to the angulation of the penis. This was attached to a pressure transducer with a computer display monitor. This provided a graphical representation of the change in intracorporal pressure (ICP) upon electrostimulation of the cavernous nerve. Nerve stimulation A custom-made bipolar microelectrode with 32G platinum wires separated by 1–2 mm (Harvard Apparatus Inc., Holliston, MA) was attached to an S48 pulse stimulator (Grass Inc., West Warwick, RI). This was used to provide electrical stimulation to the cavernous nerves. Stimuli of 5 V, 20 Hz frequency, and 5 ms in length were applied to each nerve for at least 3 min (or until maximal ICP was reached). Changes in ICP were continually monitored during stimulation. Full response was deWned as greater than 80% of the
123
336
World J Urol (2008) 26:333–339
mean intracorporal pressure achieved by nerve stimulation in a normal population of rats. Partial response was deWned as intracorporal pressure between 20 and 80% of the mean achieved in a population of normal rats. No response was deWned as the maximum ICP with electrostimulation less than 20% of the mean achieved in a population of normal rats. Histological evaluation All harvested tissues were Wxed immediately with 4% paraformaldehyde in phosphate-buVered saline (PBS). Tissues were dehydrated and embedded in paraYn, and 4 m sections were cut through each specimen. Hematoxylin and eosin (H&E) staining and immunohistochemical techniques were used for analysis. Monoclonal mouse anti-human neuroWlament protein (DAKO, Carpenteria, California) was used to examine axonal sprouting into acellular nerve grafts, and polyclonal rabbit anti-cow S-100 (DAKO, Carpenteria, California) was used to detect the presence of Schwann cells. DeparaYnized and rehydrated tissue sections were incubated with each primary antibody. After removing excess primary antibody and washing with PBS, the sections were incubated with appropriate biotinylated secondary antibodies (Vector Laboratories, Inc., Burlingame, California), washed, and incubated with streptavidinperoxidase (Vector Laboratories). Gill’s Hematoxylin was used as a counterstain. Normal rabbit serum was used as a negative control for immunohistochemistry. Statistical analysis Outcomes were evaluated with a Fisher’s exact test to compare the study groups and the negative controls. Statistical signiWcance was deWned as a P < 0.05.
Results All animals survived the procedures without noticeable complications and remained viable until their predeter-
mined time points. Electrostimulation of the cavernous nerve in normal rats produced a 10-fold elevation of intracorporal pressure under normal circumstances. The mean ICP at maximum stimulation of normal rat cavernosal nerve was 117 cm H2O. These values were used as a reference point for functional experiments, and the ratio between the intracorporal pressures obtained in experimental animals and these normal values were used as the “restoration index”. Complete restoration of nerve function was determined as an ICP greater than 93 cm H2O (>80% of normal rat ICP). Partial nerve function restoration was considered to be an ICP greater than 23 cm H2O and less than 93 cm H2O (20% of normal rat ICP). ICP less than 23 cm H2O was considered as no response. None of the animals in group C (no repair of the nerve injury) achieved restoration at any time. Within group A (acellular grafts), 25% of the rats had complete nerve function restoration by 1 month and this increased to 55% at 3 months (Fig. 4). Group B (autologous grafts) had a similar outcome, with 23% demonstrating complete nerve function restoration at 1 month and 50% at 3 months. There was no statistically signiWcant diVerence between groups A and B. Partial functional restoration was observed in 90% of the rats in groups A and B at 3 months. These results indicate that acellular nerve grafts may achieve a functional outcome that is equivalent to that obtained with fresh autologous nerve grafts (Table 1). Gross or microscopic evidence of spontaneous nerve regeneration was not detected in rats that did not receive any treatment for nerve injury (group C). Histologically, the acellular grafts (group A) contained normal appearing cellularity, and the structure and cellular organization were similar to that of the fresh nerve graft at both time points. However, a mild inXammatory inWltrate was noted in all specimens from the acellular group (Fig. 2). This was particularly evident at the periphery of the nerve tissue and was most obvious at 1 month post-surgery. Groups A and B demonstrated an increase in the density of nerve Wbers crossing the grafts sites at the 3-month time point. This was best visualized on transverse sections (Fig. 3). At the 3-month time point, the histological features of both the
Table 1 Nerve function recovery in response to electrical stimulation at 3 months Response
Acellular graft
Autologous
No repair
Sham operation
Complete/full
11/20 (55%)
10/20 (50%)
0/17 (0%)
19/20 (95%)
Partial
7/20 (35%)
8/20 (40%)
1/17 (6%)
1/20 (5%)
No response
2/20 (10%)
2/20 (10%)
16/17 (94%)
0/20 (0%)
Total response
18/20 (90%)
18/20 (90%)
1/17 (6%)
20/20 (100%)
The mean intracorporal pressure at maximum stimulation of cavernosal nerve in normal rat population was 117 cm H2O Complete/full response = ICP at maximum stimulation of nerve >93 cm H2O, Partial response = ICP >23 cm H2O and
