Created by:
John M. Mesa, MD

Updated by:
Ross F. Goldberg, MD







































































































































The Plastic Surgery Research Laboratory is located on the sixth floor of the Thier Research Building on the main campus of the Massachusetts General Hospital. The laboratory includes more than 1000 square feet of dedicated wet laboratory space adjacent to the Knight Surgical Suite and the large animal farm. The laboratory conducts investigations into the following areas of interest:

1) tissue engineering of cartilage;
2) mechanisms of neural injury and regeneration;
3) ischemia/reperfusion injuries;
4) transplantation of vascularized limb tissues.

1) Tissue Engineering of Cartilage

Tissue engineering is another approach being investigated in the Plastic Surgery Research Laboratory to overcome the paucity of tissue available for reconstruction. The engineering of cartilage from autologous chondrocytes allows for the reconstruction of craniofacial and other musculoskeletal tissue deformities, including joint resurfacing, without the complications associated with frozen osteochondral allografts or prosthetic materials. In addition, by using tissue culture techniques for expanding cell numbers, only a small number of autologous cells are required, thereby eliminating the donor site morbidity associated with large cartilage autografts.

a) Injectable articular cartilage: Our early work showed that cartilage could be produced when articular chondrocytes were suspended in a three dimensional biodegradable hydrogel polymer, such as alginate, in a nude mouse model. By utilizing polymers that exist in liquid form, it is possible to introduce the polymer containing chondrocytes into various body compartments using minimally invasive techniques. Such liquid polymers could be injected through a needle into the subperiosteal space or even subcutaneously for craniofacial reconstruction.

Subsequent studies have explored other polymers, namely fibrin-based gels and photocurable synthetic polymers, for improved delivery and fixation. In addition, these polymers could be injected under direct vision with the use of endoscopy into the joint space for the purpose of repairing cartilaginous defects in joints injured by osteoarthritis, rheumatoid arthritis, temporomandibular joint dysfunction, acute trauma, and a multitude of other debilitating joint disorders. Being able to create autologous cartilage in situ by employing minimally invasive technology could potentially reduce donor site morbidity, patient recuperation time, hospitalization, and overall costs. Studies are now underway in swine to investigate the potential for injectable cartilage to repair defects in the articular joint surface.

In collaboration with chemical engineers at the Massachusetts Institute of Technology and the University of Colorado, other studies are exploring new polymer mixtures and evaluating the cartilage matrix product.

b) Craniofacial cartilage: The fibrocartilage structures of the cranium are also a focus of our tissue engineered cartilage program. Unlike the avascular nature of the joint, the cartilages in the cranium have a different composition and blood supply. Efforts are directed towards understanding the parameters for engineering craniofacial cartilage. Studies have been performed comparing the matrix products of tissue engineered cartilage formed from ear and costal-derived chondrocytes versus that from articular type cells. Because of the close interaction of the perichondrium and skin in the cranial cartilages, we are studying means to create lamination of soft tissues over the engineered cartilage as well.

c) Meniscal repair: In collaboration with the MGH Orthopaedic service, we have a program investigating repair of meniscal lesions using scaffolds and cells. Tears to the inner portion of the meniscus, referred to as the white or avascular zone, do not heal.

Using a cell-seeded scaffold implanted into a lesion in a nude mouse model, we have demonstrated that this portion of the mensicus can heal. Studies are underway exploring scaffold materials and different cell sources to effect the repair. Initial studies have begun in swine to evaluate these techniques in situ.


2) Mechanisms of Neural Injury and Regeneration

The development of microsurgical techniques in the 1970's raised the level of technical sophistication in peripheral nerve reconstruction to one that has not yet been surpassed. With proximal injuries, however, poor recovery remains the norm. Even in more favorable lesions, such as median or ulnar nerve injuries in the forearm, only 10% of patients who undergo microsurgical repair achieve an excellent recovery, i.e. full muscle strength and normal two-point discrimination. Nerve gaps and nerve deficits further complicate the regeneration process, prompting efforts to develop improved methods of neural repair. Using both small and large animal models of peripheral nerve injury and recovery, several aspects of neural injury and repair are currently being investigated in our laboratory.

a) Transplantation of Embryonic Stem Cell Derived Motor Neurons : We hypothesize that transplantation of embryonic stem (ES) cell derived motor neurons may enhance functional outcomes by better supporting the biological integrity of the injured muscle and nerve. These motor neurons may provide trophic support to the muscle by forming neo-neuromuscular junctions and up-regulating specific growth factors, preserving motor unit integrity. Currently, we examine the functional properties of ES cell derived motor neurons in vitro, and the potential of transplanted motor neurons to revitalize denervated skeletal muscle after denervation in vivo. After inflicting a sciatic or tibial nerve transection, ES cell derived motor neurons will be transplanted into the mouse gastrocnemius muscle. The effect of the transplants on functional recovery following nerve repair will be investigated. Electrical stimulation of the muscle will also be used to stimulate the motor neurons after transplant.

Text Box: Figure: Fluorescence image shows eGFP expression in motor neurons derived from GFP/HB9+ ES cell.  ES cell derived motor neurons can extend neurites in vitro.

b) Photochemical Tissue Bonding : An optimal nerve repair technique would be quick, atraumatic, and would minimize inflammation while maximizing axonal recovery. Currently, we are collaborating with the Wellman Centre for Photomedicine ( ) to develop such a technique. Utilizing the process of non-thermal collagen cross-linking, we are developing alternatives to standard microsurgical anastomosis. This watertight tissue seal has been shown to enhance recovery of the regenerating nerve in a rat model. We are currently investigating its effects in a large animal model.

We are also investigating alternate applications of this technology, specifically in the area of microsurgery. We have shown it's superiority over standard microsurgical vessel anastamosis in a rat model using the femoral vessels. We also hypothesize that the unique properties of this repair technique will allow for decreased bulk, increased gliding and decreased adhesion formation around a tendon repair site. We are evaluating this in both in vivo and ex vivo models of flexor and rotator cuff tendon injuries.

c) Real time in vivo neural microscopy : The necessity of surgical intervention for peripheral nerve damage often depends on the severity of the injury. In the field of peripheral nerve reconstruction, there is currently no very accurate method that can be implemented in a real time manner for predicting outcome following injury. Recovery from peripheral nerve injury usually requires restoration of the internal neural architecture by resolution of edema in cases of neurapraxia and by regrowth of axons to distal targets in more severe cases of crush or stretch or actual division. Electrical studies in the early phase of injury cannot distinguish between nerves with minor or severe disruption internally. Furthermore, only after recovery has occurred over a period of several months can such a determination be made. This prolongs the period of muscle denervation distally, prolongs the time to ultimate recovery if surgical reconstruction is required, and ultimately hastens the time after which meaningful reconstruction, particularly of motor neuron lesions, is no longer possible. We are currently evaluating two imaging devices, in collaboration with the Wellman Centre for Photomedicine ( ) with a goal of providing vital prognostic and diagnostic information following nerve injury.

1) C.A.R.S. (Coherent Anti-Stokes Raman Scattering) Microscopy is a non-contact, minimally invasive method of imaging tissues in-vivo. It consists of a nonlinear optical process that uses ultrashort laser pulses at multiple frequencies to probe molecular vibrational structures and conformations in samples. This generates strong, molecule-specific signals that can be used for molecular detection and high resolution imaging. The unique characteristics of these Microscopes provides detailed images of the peripheral nerve micro-environment in vivo, allowing serial cataloging of nerve regeneration following various injuries.

2) OCT (Optical Coherence Tomography) is a non-contact, minimally-invasive technology which uses infrared light to image tissue in a cross-sectional manner. Spectral-domain OCT is a technique that provides in vivo imaging of sub-surface structures with high resolution in three-dimensions. This capability can be augmented by functional extensions of the technique. Phase-resolved optical Doppler tomography enables depth-resolved imaging of flow by observing differences in phase between successive depth scans, and has been utilized to observe blood flow patterns in port-wine stains and retinal blood flow patterns. Polarization-sensitive OCT utilizes change in the polarization state of detected light to determine the depth-resolved light-polarization changing properties of a sample. The most significant of these properties for clinical imaging is birefringence, which is exhibited by fibrous or organized structures such as collagen, muscle, tendon, and nerves. PS-OCT can be used to identify, characterize, and detect changes in these structures, as has been demonstrated in studies mapping the birefringence of the nerve fiber layer and correlating the extent of thermal injury with a decrease in dermal collagen birefringence. We are currently evaluating OCT as both a tool to characterize peripheral nerve injury and as an intra-operative surgical aid.


3) Ischemia/Reperfusion Injuries

Ischemia-reperfusion injury (I/R) is the inflammatory response to restoration of blood flow after an ischemic event. This inflammatory response is associated with significant morbidity and mortality in many clinical settings, such as free flap transfers, transplantation and replantation surgery, and crush injuries. Locally, this can lead to tissue loss and muscle necrosis. Systemically it can lead to multi-system organ dysfunction and death.

In many clinical settings, the duration of ischemia is beyond the surgeon's control, and preventive measures are required to reduce the extent of reperfusion injury. There are many studies investigating methods to reduce injury to organs, most of which have been done in rat, rabbit, and larger animal models. These models focus on ischemic insult to tissue including musculoskeletal, gastrointestinal, renal and cardiovascular system.

Although episodes of ischemia can cause organ damage and dysfunction, subjecting myocytes to brief episodes of transient myocardial ischemia increases the capacity of the heart to endure subsequent prolonged periods of ischemic injury. This phenomenon is termed preconditiong. The protection conferred by preconditioning was shown to be endogenous with myocardial protection but later shown to be also conveyed at a distance. where myocardial protection was shown to be conferred with ischemic preconditiong of the gastrocnemius of the rabbit. The current belief is that humoral factors are responsible for this local protection and also its remote protective effects. However these mechanisms are not completely understood.

Although larger animal models have been extensively utilized when examining I/R injury, we use a murine model. This allows us to use may generate tools and technologies associated with mice, such as knockout animals. This also enables us to isolate specific proteins that we believe are involved in the mechanisms of protection. Over the past five years our laboratory has investigated ischemia reperfusion using many different mouse models (hindlimb, intestinal and skin flap). We are currently using our armamentarium of ischemia reperfusion knowledge to unravel a variety of problems facing plastic surgeons today.


4) Transplantation of Vascularized Limb Tissues

The limited availability of autologous tissue for reconstructing large bone and joint defects after tumor resection, traumatic injury, or congenital deformities remains a challenging clinical problem. For example, available sites for bone graft material are primarily limited to calvarium, iliac crest, rib, and fibula. Although rib and fibula grafts can be transplanted on a vascular pedicle, there is little flexibility in their dimensions and configurations and donor site morbidity can also be problematic. Current approaches to overcome autologous tissue shortage include the use of frozen cadaveric bone or osteochondral grafts. The results with large frozen allografts have been equivocal, as the lack of vascularization predisposes the grafts to nonunion and fracture, as well as soft tissue healing problems. The recent hand transplants performed in the US and France demonstrate the technical feasibility of allotransplantation of limb tissues in humans. However, the need for chronic immunosuppression to prevent graft rejection may pose unnecessarily high risk to patients for treating non life-threatening deficiencies. Thus, our laboratory, in collaboration with the Transplantation Biology Research Center, is investigating means for inducing tolerance to vascularized limb tissue allografts.

a) In utero induction of tolerance: Along with the multiple new means for detecting congenital defects in utero, new therapeutic approaches could be developed to begin treatment during gestation. Fetuses identified with inborn errors of metabolism or defective organs could be inoculated in utero to induce tolerance to allogeneic transplants could be performed after birth. During the fetal stage of immune development, it is theorized that the fetus cannot determine self from non-self. Therefore, inoculating the fetus with alloantigen could induce a state of tolerance. Our laboratory has explored this by injecting adult bone marrow from genetically defined miniature swine into the portal vein of fetuses in pregnant sows at mid-gestation (50-55 days.)

Click HERE to view video clip.

Rather than use intraperitoneal injections as other investigators have in sheep and primate models, we inject directly into the portal vein delivering the cell inoculum to the fetal liver--the primary hematopoietic organ in the developing fetus. Eleven animals have been born that demonstrated the presence of donor cells in the peripheral blood after birth as determined by FACS analysis (ranging from 0.17-1.00 %). This chimerism, defined as the coexistence of cells from donor and recipient, declines over time, but this group of animals has demonstrated acceptance of kidney allografts for greater than 200 days. This is significant in that these transplants are across a major histocompatibility barrier and immunosuppression is not required. Our attention is now focusing on the possible mechanisms involved in generating the tolerant state in these offspring.

b) Limb tissue allografts in miniature swine: Whereas recipients of organ allotransplants require life-loing immunosuppressive regimens to maintain their allograft, the use of chronic immunosuppressive therapy for limb tissue transplants is far less likely. Our attention has focused on using the MGH miniature swine as a model

for understanding allotransplantion of limb tissues in adults. We have demonstrated that animals receiving vascularized musculoskeletal transplants from donors that are matched for major histocompatibility antigens and having minor antigen differences can accept their grafts for greater than one year using only a 12-day course of cyclosporine postoperatively.

If cyclosporine is not given, the animals can reject the grafts transplanted across this minor barrier. We are exploring transplantation across increasingly stronger histocompatibility barriers such as class I MHC only or single haplotype class I and II MHC barriers. Another approach for achieving tolerance could be to generate mixed chimerism in the adult. Work in the TBRC has developed mixed chimerism protocol to establish stable multilineage chimerism across minor and major histocompatibility barriers using non-myeloablative regimens. Our future goals are to use such approaches to determine whether similar protocols can work for limb tissue allotransplantation.


c) Limb tissue allografts in rodents: In addition to large animal studies, we are actively working on three tolerance strategies in rodent models: 1) the in utero induction of tolerance, 2) neonatal induction of tolerance, and 3) tolerance through costimulatory blockade. Our protocol for the in utero induction of tolerance in rats parallels that of our swine model. These rodents demonstrate similar levels of chimerism by FACS analysis and have received limb grafts across a complete MHC barrier. The rodent neonatal immune system is still permissive in the first few days after birth. The injection of alloantigen during this period has led to chimerism and long-term acceptance of limb allografts across a complete MHC barrier. New monoclonal antibodies and immunosuppressive agents have made costimulatory blockade an exciting area of transplantation research. In order to mount an immunological (rejection) response against a foreign graft, the host T cells require at least two signals for specific activation. By blocking these signals, it may be possible to induce a state of tolerance to transplanted tissue. All of these models have been developed to avoid the mortality and morbidity of chronic immunosuppression. Another significant hindrance of transplantation is the short supply of donor tissue. Xenogeneic tissue is essentially an unlimited source for musculoskeletal allografts. Our in utero induction model and the costimulatory blockade model are currently being tested across a xenogeneic barrier (mouse to rat).