Novel bone adhesives: a comparison of bond strengths in vitro.

Fracture fixation using adhesive is a promising alternative in craniofacial surgeries, replacing the plates and screws system. The advantages include the ease of application and avoidance of drilling holes that may weaken the bone and cause fractures. In this study the bond strengths of selected adhesives were evaluated and compared with resorbable plates and screws. Four adhesives, octyl-cyanoacrylate, N-butyl-cyanoacrylate, a novel methyl-methacrylate, and a novel cyanoacrylate derivative, were tested for their microtensile and shear bond strengths. The bone samples were cut into rectangular bars and bonded with selected adhesives for microtensile testing. For the shear bond test, paired bars were bonded at the overlap, while two other sets of bars were attached by a Lactosorb plate using either adhesive or screws. Data were analysed by analysis of variance (ANOVA). The microtensile bond strengths of N-butyl-cyanoacrylate, novel cyanoacrylate derivative, and novel methyl-methacrylate derivative were significantly greater than octyl-cyanoacrylate. When bone sections were fixed with resorbable plates and adhesives, shear bond strength was significantly greater for N-butyl-cyanoacrylate than plate and screws, while the bond strengths of other adhesives were comparable with the plate and screws. N-Butyl cyanoacrylate was shown to have the greatest potential for fixation of fractured bone in craniofacial surgical applications.

Early treatment of severe mandibular hypoplasia with distraction mesenchymogenesis and bilateral free fibula flaps.

The technique of distraction has revolutionized the treatment of mandibular hypoplasia; however, presently large mandibular defects still require bone grafts. Microvascular grafting is commonly used in adults. Conversely, in pediatric reconstruction, nonvascularized rib grafts remain standard. Unfortunately, resorption of nonvascularized bone remains a major issue, particularly when soft tissue is hypoplastic. This case study represents a combination of techniques in the treatment of severe mandibular deficiency, and introduces the concept of distraction mesenchymogenesis. The patient was a 2 1/2-year-old boy with severe bilateral Pruzansky class III mandibular hypoplasia. He had a permanent open mouth posture, an overjet of 23 mm, and was unable to move the lower mandibular segment. His oropharyngeal airway diameter was 2.2 mm and he was tracheostomy dependent. The patient was treated with distraction of the lower jaw mesenchyme followed by bilateral functional free fibular microvascular flaps containing reinnervated muscle. This created a well-vascularized body, ramus, and condyle bilaterally within an adequate soft-tissue envelope. Postoperatively, the overjet was reduced to 5 mm. The patient can now actively move his mandible. Airway diameter increased to 10 mm, and the patient is able to tolerate intermittent tracheostomy plugging. This innovative combination of techniques allows early intervention, limits graft resorption, and improves airway control.

Distraction osteogenesis of the craniofacial skeleton.

Distraction osteogenesis is becoming the treatment of choice for the surgical correction of hypoplasias of the craniofacial skeleton. Its principle is based on the studies of Ilizarov, who showed that osteogenesis can be induced if bone is expanded (distracted) along its long axis at the rate of 1 mm per day. This process induces new bone formation along the vector of pull without requiring the use of a bone graft. The technique also provides the added benefit of expanding the overlying soft tissues, which are frequently deficient in these patients. This article reviews the authors' 11-year clinical and research experience with mandibular distraction osteogenesis. It highlights the indications and contraindications of the technique and emphasizes the critical role that basic science research has played in its evolution.

Remodeling of the temporomandibular joint following mandibular distraction osteogenesis in the transverse dimension.

Transverse mandibular distraction osteogenesis involves moving the osteotomized segments of the mandible in either a varus or valgus direction. This maneuver allows for widening of the bigonial distance or for a lateral shift of an asymmetric mandibular midline. During this process, a significant amount of torque is placed on the mandibular condyles, because they act as the pivot point for the mandibular translation. Although standard linear distraction osteogenesis induces transient, reversible changes in the temporomandibular joint, it is not known what effect the varus and valgus stresses of transverse distraction have on the temporomandibular joint. We therefore designed a study to document the temporomandibular joint changes following various degrees of transverse distraction. Bilateral transverse mandibular distraction was performed on 10 adult, female mongrel dogs using an external, multiplanar mandibular distraction device. The distraction protocol was as follows: (1) complete osteotomy at the angle of the mandible, (2) 5-day latency period, (3) distraction rate of 1 mm/day, (4) rhythm of one turn per day, (5) linear activation 16 to 30 mm bilaterally, and (6) 8-week consolidation period. A variety of varus and valgus distraction vectors were applied to the mandible only after 10 mm of initial linear distraction had been achieved. Posteroanterior and lateral cephalograms were performed throughout the entire process. Pre-distraction and post-consolidation computed tomographic scans were also performed. Changes in mandibular conformation, axis of rotation, temporomandibular joint structure, and glenoid fossa changes were directly assessed by evaluating the postmortem craniofacial skeleton. The findings were compared with those of normal, age-matched mongrel dog skulls. Significant remodeling changes were observed in the temporomandibular joints of all animals involved in the study. The mandibular condyles demonstrated varying degrees of flattening and erosion at all contact points with the craniofacial skeleton. In some cases, the condyle became part of the distraction regenerate process and was hypertrophied in all dimensions. The condyles were frequently displaced out of the glenoid fossa, particularly on the side in the direction of varus distraction. When the latter occurred, a new fossa was created on the undersurface of the zygomatic arch. Varying degrees of mandibular rotation in the sagittal plane were also observed, which led to abnormal torquing of the condyles in the coronal plane, depending on whether the axis of rotation occurred primarily around the condyle or around the distraction regenerate zone.In conclusion, transverse mandibular distraction is an effective means of producing a varus or valgus shift in the gonion relative to the midsagittal plane. However, unlike linear or angular mandibular distraction, transverse distraction has a multitude of nontransient effects on the temporomandibular joint. Therefore it must be emphasized that in clinical practice, transverse distraction should be used cautiously. One must also be aware that such a maneuver in distraction can have negative effects on the temporomandibular joint.

Nerve dependency in scarless fetal wound healing.

The human fetus is capable of healing cutaneous wounds without scar up to the third trimester of development This process of tissue repair is more akin to newt limb regeneration than classic adult scar forming wound repair. Regeneration of the newt limb is dependent on neural input in its early stages. This study was an attempt to determine whether a similar dependence on neural input exists for mammalian fetal wounds to heal without scar. The left hind limb of six fetal lambs was denervated during the early second trimester of development (day 55; term = 145 days). Two weeks after denervation, the animals were again exposed to create bilateral incisional and 6-mm-diameter excisional wounds on their innervated right and denervated left lower extremities. Five days after creation of these defects, the wounds were examined for alterations in repair. Four fetal lambs survived, and three were suitable for evaluation. There were marked alterations in wound healing seen after denervation. Excisional wounds on the innervated side contracted and decreased their surface area by 14 percent. In contrast, the denervated wounds not only failed to contract, but increased in size by 60 percent. Changes in the incisional wounds were equally distinctive. Innervated incisional wounds healed completely without scar and had a wound breaking strength comparable to that of normal skin (Table I). In contrast, two of the three denervated incisional wounds dehisced and failed to heal, even in the regions where the skin was approximated by suture. The third denervated incisional wound did heal but with a significant amount of scar. Electron microscopy confirmed this finding by clearly demonstrating thickened and irregular collagen deposition in the extracellular matrix of all the denervated incisional specimens. In summary, like the regenerating newt limb, scarless fetal skin wound repair requires neural stimulation for tissue regeneration to occur. Therefore, in the mammal, the primary regulator for this unique type of tissue repair may have a central neural, rather than a local, tissue origin.

Reconstruction of the posttraumatic short upper lip.

Treatment of the posttraumatic, vertically shortened upper lip is a difficult surgical problem. It requires careful evaluation of the underlying injury followed by staged therapeutic interventions. Both surgical and nonsurgical treatments need to be employed to optimize results. The authors present three distinct cases of posttraumatic upper lip reconstruction that utilize a variety of treatment modalities. All patients were treated by the senior author.

Fetal wound repair: where do we go from here?

In contrast to adult wound healing, early-gestation fetal skin wound healing occurs rapidly, in a regenerative fashion, and without scar formation. The accelerated rate of healing, relative lack of an acute inflammatory response, and an absence of neovascularization distinguishes fetal from adult wound healing. However, this remarkable ability of the fetus to heal without scarring still remains poorly understood. The uncertainties include the role of cytokines, extracellular matrix components, homeobox genes, and certain cell types in the scarless wound repair process. Nevertheless, some strides have been made within the last two decades. This report, discusses the current knowledge of the mechanisms and characteristics of scarless fetal wound healing. Furthermore, to shy away from being just another all inclusive review, the authors point out deficiencies in the knowledge base on this important topic. Last, the future direction of research is discussed that may elucidate the mechanisms regulating the scarless repair phenomena.

A long-term, controlled-outcome analysis of in utero versus neonatal cleft lip repair using an ovine model.

Successful open repair of a cleft lip in utero has the advantage of scarless wound healing in the fetus. Unfortunately, no long-term outcome studies have been performed to evaluate the efficacy of these repairs. Moreover, no study to date has compared the long-term results of an in utero cleft lip repair to a similar, control-matched, newborn cleft repair. This study was performed to evaluate the 9-month outcome of in utero cleft lip surgery compared with an identical cleft lip repair performed on infant lambs. In utero epithelialized cleft lips were created through an open hysterotomy in sixteen 65-day-old fetal lambs (term = 140 days) using methods described by Longaker et al. Eight of 16 animals underwent subsequent in utero repair of these clefts at 90 days gestational age. The repair of the remaining eight animals was delayed until 1 week postpartum. At 9 months, the animals were analyzed for changes in lip contour and for the degree of scarring by hematoxylin and eosin and Masson's trichrome collagen staining. Two animals in each group died from preterm labor. Of the animals that survived to term, all repaired lips had some degree of abnormality postoperatively. One of six lips repaired in utero dehisced before delivery. Three of six neonatal repairs dehisced in the first postoperative month. In the remaining animals with intact lip repairs, the vertical lip height on the repaired side was an average of 9 to 12 mm shorter than the normal lip in both the in utero and neonatally repaired animals. Phenotypically, the postnatally repaired animals had more lip distortion and visible notching. Histologically, the in utero repair was scarless and the neonatal repairs had scar throughout the entire vertical height of the lip with an associated loss of hair in this region. Maxillary growth was also evaluated. There was no inhibition of maxillary growth in the animals that underwent in utero cleft lip repair. However, in the neonatal repair group, significant maxillary retrusion was evident. Compared with the cleft side of the maxilla, horizontal growth was decreased by 11 percent (p = 0.01). Compared with the intrauterine repair group, there was a 17-percent decrease in horizontal maxillary width (p = 0.01). Straight-line in utero repair of a cleft lip produces a better long-term result in terms of maxillary growth than a similar repair performed postnatally in the ovine model. There was no diminution in maxillary growth in the animals treated in utero. Histologically, in utero repair of clefts was indeed scarless. However, both lip repairs produced lips that were significantly shorter than their contralateral noncleft sides. This degree of lip shortening would require a secondary lip revision, thereby defeating the purpose of performing an intrauterine repair. Comparisons now need to be made between in utero and neonatal repairs using a Millard-type rotation advancement technique before intrauterine treatment can be considered to be more beneficial than our current treatment modalities.

Vilray P. Blair, his surgical descendants, and their roles in plastic surgical development.

Perhaps one of the most historically well-known plastic surgeons is Vilray P. Blair. As commander of the U.S. Army corps of head and neck surgeons during World War I, he became well known for his work in posttraumatic reconstruction. Blair's efforts in the early part of this century helped to develop plastic surgery as a distinct surgical subspecialty in the United States. His prowess as a surgeon allowed him to build one of the largest plastic surgery centers in the country and to train many of the top young American surgeons. Blair excelled as a teacher. He produced academic surgeons such as James Barrett Brown and Bradford Cannon, who took the lead in the care of wartime injuries during World War II. At Valley Forge General Hospital, Blair's trainees dedicated themselves to the reconstruction of injured patients and trained other young plastic surgeons in the care of postwar trauma. This exceptional level of patient care resulted in the U.S. government recognizing plastic surgery as a subspecialty following World War II. Since that time, Blair's surgical descendants at Washington University have led the country in the development of new training concepts and ideals and have gone on to become leaders in plastic surgery worldwide.

Distraction osteogenesis of the mandible: a ten-year experience.

Mandibular distraction has been performed at the authors' institution for the past 10 years on a variety of craniofacial anomalies. This article reviews the experience with distraction and outlines the authors' treatment algorithms based on patient age and pathology. The roles of distraction versus conventional orthognathic surgery are reviewed. The need for preoperative surgical planning and postoperative orthodontic therapy is emphasized.

Adverse outcomes following endoscopic repair of a fetal cleft lip using an ovine model.

OBJECTIVE: The purpose of this study was to determine if endoscopic techniques could be used to repair an epithelialized lip cleft with accuracy and with an outcome comparable to fetuses treated through an open hysterotomy. INTERVENTIONS AND RESULTS: In contrast to previous open fetal cleft lip repairs in the same model, none of the five fetuses reported here had a good aesthetic result. Although there was no evidence of scar histologically, the edges of the lip were poorly approximated. The epithelial lining and underlying dermis of the wound margins were notably inverted. The orbicularis oris muscle, which had been reapproximated, appeared thin and hypoplastic. Most of the vermilion elements were poorly aligned, and in one animal, there was a complete dehiscence of the repair. CONCLUSIONS: In a more representative model of cleft lip that is not an acute lip wound, in utero endoscopic suture repair of the ovine lip gave a poor result using current technology. Only a meticulously performed, multilayered, open repair of a cleft appears to give a good cosmetic and functional outcome. Further studies to improve the endoscopic repair as our technology advances are therefore warranted.

Modulation of the human homeobox genes PRX-2 and HOXB13 in scarless fetal wounds.

Scarless healing of cutaneous wounds occurs in humans during the first two trimesters of development, but by birth all wounds are repaired with scar formation. To search for transcriptional regulatory genes that might mediate fetal tissue regeneration, we surveyed homeobox gene expression in proliferating fetal fibroblasts and in wounded and unwounded skin. Two novel human homeobox genes, PRX-2 and HOXB13, were identified that were differentially expressed during fetal versus adult wound healing. Both genes were predominantly expressed in proliferating fetal fibroblasts and developing dermis, and PRX-2 was downregulated in adult skin. In a model of scarless fetal skin regeneration, PRX-2 expression was strongly increased compared with unwounded skin and the signal was localized to the wounded dermis, the site of scarless repair. Conversely, in adult skin weak epidermal PRX-2 expression was observed, mRNA levels were not increased by wounding, and no dermal expression was detected. HOXB13 expression was decreased in wounded fetal tissue relative to unwounded fetal controls or wounded adult skin. Thus both HOXB13 and PRX-2 are expressed in patterns consistent with roles in fetal skin development and cutaneous regeneration.

Intracranial migration of microplates versus wires in neonatal pigs after frontal advancement.

Several case reports have demonstrated that microplates migrate intracranially when used in the reconstruction of the neonatal craniofacial skeleton. We conducted a study to analyze the validity of this finding, to calculate the rate of intracranial plate migration, and to compare the migration of microplates to standard surgical wire. We hypothesized that, because of constant bone generation and resorption in the developing cranial skeleton, both microplates and wires would migrate and resorption in the developing cranial skeleton, both microplates and wires would migrate intracranially. Bilateral supraorbital osteotomies were performed in five 6-week-old Yucatan minipigs. The right side was repaired by fixating the bone segment with 32-gauge stainless steel wire. The left side was fixed with two microplates (Leibinger Corp, Dallas, TX). The pigs were monitored for 6 months. During this time each pig was injected with a fluorescent bone dye on the first day of each month. One pig was killed at 3 months and used for midpoint comparison. We found that both microplates and wires migrated intracranially in these neonatal pigs. None of the plates or wires penetrated the dura, but several pieces of hardware from each group migrated far enough to rest directly on the dural surface. The rate of plate and wire migration also varied slightly among animals, but the mean rate remained relatively constant at 0.91 mm/month. Computed tomographic scans and morphometric analysis revealed no significant difference between the fracture healing and facial growth of the microplate and wire sides. We conclude that the biology of bone deposition in the calvarium causes any stationary foreign body on the surface of the skull to migrate intracranially in the neonate.

A new in utero sheep model for unilateral coronal craniosynostosis.

Several animal models have been designed in the past to analyze the pathophysiology and management of craniosynostosis, very few of which were intrauterine. Those that were interuterine had problems with either a short gestation or limited availability that prevented most researchers from using them in treatment analysis. We desired to create a biologically sound intrauterine model of craniosynostosis, using an animal with a long gestation and an early calvarial bone formation, which was easy to manipulate in utero, that could be created by any researcher studying this disorder. Using biologic data available regarding growth factors thought to be involved in bone growth and cranial suture closure, we developed a new in utero fetal lamb model for the study of craniosynostosis. Ten 70-day gestation fetal lambs (term gestation 140 days) received a midline coronal incision to expose both coronal sutures. The entire right coronal suture was then excised along with a 4-mm bony margin. In each animal, the site was packed with 25 mg of demineralized sheep bone powder augmented with 50 microg of bone morphogenetic protein-2 (BMP-2) and 1 microg of poly-transforming growth factor-beta. The scalp was closed, and the sheep were returned to the uterus until either 90 or 140 days of gestation. Complete fusion of the right coronal suture occurred in all fetuses by 90 days gestation. In every animal, right-sided frontal bone flattening and supraorbital rim elevation were evident. Histologic analysis showed bony synostosis at the suture site without evidence of suture regeneration. By 140 days, this isolated suture fusion led to marked craniofacial abnormalities including right supraorbital rim elevation, significant frontal bone flattening, a decrease in the anterior-posterior length of the cranial vault, and flattening of the cranial base. In conclusion, we have developed a new model for the study of the secondary effects induced by the process of cranial suture fusion, which produces abnormalities seen in naturally occurring cases of isolated right coronal suture synostosis. In addition, this model confirms that isolated coronal suture fusion alone can lead to the multiple cranial and facial abnormalities seen with this disorder, even in the absence of associated cranial base suture fusions.

The in utero correction of unilateral coronal craniosynostosis.

We performed the first in utero correction of a unilateral right coronal craniosynostosis using 70-day gestation fetal lambs. The craniosynostosis was created in eight fetuses by excising their right coronal sutures, and then placing demineralized bone powder, transforming growth factor-beta, and bone morphogenetic protein-2 into the defect. Twenty-one days later, after suture fusion had occurred, four of the eight sheep were treated with a 4 mm x 12 mm strip craniectomy to open the entire synostosed right coronal suture. The edges of the excision were wrapped with 100-microm-thick Gore-Tex (W. L. Gore & Associates, Flagstaff, Ariz.) sheets to prevent bony refusion. All eight lambs then progressed to term (140 days). The skulls of four normal, unoperated, term lambs were used as controls. At 140 days, all four treated lambs had a widely patent strip craniectomy site without any evidence of bone regeneration. This in utero correction led to a marked improvement in craniofacial morphology of three of four animals when compared with the uncorrected controls with significant (p < 0.01) correction in orbital position, skull length, and shape of the frontal bone. This was in sharp contrast to the uncorrected animals, which had marked orbital elevation, compression of the anteroposterior length of the cranial vault, frontal bone flattening, and shortening of the cranial base. The fourth corrected animal also showed evidence of improvement but had some abnormal calvarial changes secondary to the development of horns, which displaced the calvaria in a downward vector. We conclude that the in utero correction of craniosynostosis is feasible and provides a significant benefit by decreasing the severity of many of the associated deformities seen with this disorder.

HOX homeobox genes exhibit spatial and temporal changes in expression during human skin development.

The spatial and temporal deployment of HOX homeobox genes along the spinal axis and in limb buds during fetal development is a key program in embryonic pattern formation. Although we have previously reported that several of the HOX homeobox genes are expressed during murine skin development, there is no information about developmental expression of HOX genes in human skin. We have now used reverse transcriptase polymerase chain reaction, in conjunction with a set of degenerate oligonucleotide primers, to identify a subset of HOX genes that are expressed during human fetal skin development. In situ hybridization analyses demonstrated that there were temporal and spatial shifts in expression of these genes. Strong HOXA4 expression was detected in the basal cell layers of 10 wk fetal epidermis and throughout the epidermis and dermis of 17 wk skin, whereas weak signal was present in the granular layer of newborn and adult skin. The expression patterns of HOXA5 and HOXA7 were similar, but their expression was weaker. In situ hybridization analysis also revealed strong HOXC4 and weaker HOXB7 expression throughout fetal development, whereas HOXB4 was expressed at barely detectable levels. Differential HOX gene expression was also observed in developing hair follicles, and sebaceous and sweat glands. None of the HOX genes examined were detected in the adult dermis.

Bone morphogenetic protein-2 induces scar formation and skin maturation in the second trimester fetus.

Fetal mammals heal skin wounds through the second trimester of development without evidence of scar. We have investigated the role of bone morphogenetic protein 2 (BMP-2), which is a member of the TGF-beta superfamily, in normal skin development and fetal wound healing. We first used RNA in situ hybridization to demonstrate that BMP-2 was expressed at low levels in the developing hair follicles and in the epidermis of normal human fetal skin. We then created an in vivo model to test how exogenous BMP-2 would affect fetal skin development and wound healing. Fifty micrograms of BMP-2 was implanted into the subcutis of five 70-day-old fetal lambs through a full-thickness linear incision. The BMP-2 was placed beneath the right half of the wound, whereas the left half served as an untreated control. In two of the five animals 1 microgram of TGF-beta was placed into the same position in addition to the 50 micrograms of BMP-2. Twenty days later (90 days gestation, term = 140 days) all the fetal wounds were examined for evidence of cellular hyperproliferation and scar formation. BMP-2 induced massive dermal and epidermal growth when compared with controls. This finding was characterized by marked epidermal thickening and keratinization, a dramatic increase in the number of hair follicles, and more than 50 percent thickening of the dermis. The dermal thickening was the result of both increased cellularity and deposition of large irregular collagen bundles. Wounds treated with both BMP-2 and TGF-beta healed also with an adult-like pattern of scar formation. Surprisingly, the wounds with BMP-2 alone healed with an equal pattern of scar, indicating that there was not an additive effect of combining BMP-2 and TGF-beta. We conclude that BMP-2 is a pleomorphic growth factor that induces cellular growth, maturation, and fibroplasia in both the dermis and epidermis. Further analysis of this growth factor in both fetal and adult wound healing may lead to important discoveries regarding the control of scar formation and fibrosis in many adult tissues.

The in utero repair of Tessier number 7 lateral facial clefts created by amniotic band-like compression.

The purpose of this study was to determine the effectiveness of intrauterine reconstruction of lateral facial clefts caused by surgically simulated amniotic band compression. The authors hypothesized that intrauterine lysis of these bands could: 1) prevent the progression of cleft formation; 2) normalize facial development; and 3) allow for the scarless repair of these lateral atypical facial clefts. In a prospective randomized trial, eight 65-day gestational fetal lambs (term, 140 days) were exposed via a maternal hysterotomy. A Tessier 7 facial cleft was made by applying a 2.0 nylon suture as a constriction band to the growing face. The sutures were attached to the zygomatic arch and looped circumferentially into the oral commissure. Two weeks after surgery, four of eight animals were re-exposed. All four animals developed phenotypic changes consistent with Tessier 7 lateral facial clefts. These animals were treated in utero by 1) excising the synthetic constricting band, and 2) opening the epithelialized edges of the lateral facial clefts and performing a primary repair. At birth, the bands on the four untreated animals were cut. The lambs developed to 3 months of age, at which time they were evaluated anthropomorphically for changes in soft tissue measurements, and histologically using hematoxylin and eosin and Mason trichrome stain for the degree of scar at the repair site. Tessier 7 lateral facial clefts were produced in all the untreated animals. Each developed macrostomia with an average 2.3-cm lateral displacement of the oral commissure and partial maxillary clefting induced by the pressure of the restriction band. No evidence of tissue necrosis, maceration, or ulceration was noted. In contrast, animals treated in utero were phenotypically normal. There was no evidence of macrostomia and all anthropomorphic measurements were symmetrical. No bony clefts were seen on gross examination. Histologically, there was no evidence of scar formation at the site of the cleft repair. These findings suggest that like the standard cleft lip, atypical lateral facial clefts can be repaired successfully in utero without scarring. Furthermore, lysis of a restriction band in utero can prevent the progression of this facial deformity, leading to normal development of the face.

The human homeobox genes MSX-1, MSX-2, and MOX-1 are differentially expressed in the dermis and epidermis in fetal and adult skin.

In order to identify homeobox genes which may regulate skin development and possibly mediate scarless fetal wound healing we have screened amplified human fetal skin cDNAs by polymerase chain reaction (PCR) using degenerate oligonucleotide primers designed against highly conserved regions within the homeobox. We identified three non-HOX homeobox genes, MSX-1, MSX-2, and MOX-1, which were differentially expressed in fetal and adult human skin. MSX-1 and MSX-2 were detected in the epidermis, hair follicles, and fibroblasts of the developing fetal skin by in situ hybridization. In contrast, MSX-1 and MSX-2 expression in adult skin was confined to epithelially derived structures. Immunohistochemical analysis of these two genes suggested that their respective homeoproteins may be differentially regulated. While Msx-1 was detected in the cell nucleus of both fetal and adult skin; Msx-2 was detected as a diffuse cytoplasmic signal in fetal epidermis and portions of the hair follicle and dermis, but was localized to the nucleus in adult epidermis. MOX-1 was expressed in a pattern similar to MSX early in gestation but then was restricted exclusively to follicular cells in the innermost layer of the outer root sheath by 21 weeks of development. Furthermore, MOX-1 expression was completely absent in adult cutaneous tissue. These data imply that each of these homeobox genes plays a specific role in skin development.

A new in vivo model for the study of fetal wound healing.

We have developed a new in vivo model for the study of fetal wound healing. Fetal ICR mice (total gestation, 21 days) received a full-thickness incisional wound in the hind limb at gestational day 14 (N = 100). The wound was made with a 28.5-gauge needle that was passed transplacentally into the amniotic cavity. The wounds were analyzed histologically on postoperative days 0, 1, 3, and 5 by hematoxylin-eosin and Mallory's trichrome stains. Once the wounding technique was mastered, the overall mortality rate for this model was 20% by postwounding day 5. Each fetus healed their wound without scar by postwounding day 3. In 3 animals, 5 microliters of human transforming growth factor beta 1 (25 micrograms per microliter) was injected into the wound site, resulting in scar and an inflammatory cell infiltrate, indicating that the 14-day-gestation fetal mouse can be manipulated if necessary. This model offers the advantages of an in vivo system that can be studied at an early gestational age. Furthermore, it is inexpensive, easy to manipulate, and can be studied with commercially available murine probes.