Effects of fibroblasts of different origin on long term maintenance of xenotransplanted human epidermal keratinocytes in immunodeficient mice

Expression of CD90 on keratinocyte stem/progenitor cells

BACKGROUND

The identification and purification of keratinocyte stem cells (KSCs) that are capable of self-renewal and maintenance of differentiating cell populations could contribute both to our understanding of the biology of these cells, and to significant clinical applications, such as the culturing of keratinocytes for transplantation to severe burn wounds. Here, we report the detection of CD90(+) cells in cultured normal human epidermal keratinocytes and adult skin.

OBJECTIVES

To investigate the biological function of CD90(+) and CD90(-) keratinocytes.

METHODS

CD90(+) and CD90(-) keratinocytes were purified from adult skin and cultured keratinocytes using fluorescent activated cell sorting, and their biological abilities were analysed using both in vitro and in vivo assays.

RESULTS

Flow cytometry (FCM) analysis identified approximately 18% of post-primary neonatal keratinocytes as CD90(+). However, during expansion of the culture, the expression level of CD90 rapidly decreased to about 2.5% at passage 10, while most of the keratinocytes maintained expression of alpha6 integrin. Purified CD90(+) keratinocytes demonstrated a sixfold higher cell growth rate than CD90(-) cells and the ability to form large (over 3 mm in diameter) colonies. We then quantitatively evaluated both populations using a previously described in vivo human epidermal cyst formation assay. Enhanced green fluorescent protein (EGFP)-labelled CD90(+) or CD90(-) keratinocytes were subcutaneously injected into nonobese diabetic/severe combined immunodeficient (NOD/SCID) mice. Six weeks after transplantation, EGFP(+) cell clusters in human epidermal cysts were evaluated using image analysis software. EGFP(+) cell cluster areas in the basal layer, derived from EGFP(+) CD90(+) cells, were eightfold larger than clusters of EGFP(+) CD90(-) cells. Furthermore, immunohistochemical staining and FCM analysis indicated that CD90 was expressed in most of the basal layer of the normal human epidermis.

CONCLUSIONS

These results indicated that CD90 is a useful marker for the detection of human KSC-enriched populations in cultured human keratinocytes.

Dynamics of keratinocytes in vivo using HO labeling: a sensitive marker of epidermal proliferation state

A heavy water ((2)H(2)O) labeling method recently developed to measure cell proliferation in vivo is applied here to the measurement of murine epidermal cell turnover and to investigate conditions in which keratinocyte proliferation is either inhibited or stimulated. The technique is based on incorporation of (2)H(2)O into the deoxyribose moiety of deoxyribonucleotides in dividing cells. Label incorporation and die-away studies in cells isolated from C57BL/6J mouse epidermis revealed the replacement rate to be 34%-44% per wk (half-life of 1.6-2 wk). The kinetics provided evidence of a non-proliferative subpopulation of cells (10%-15% of the total) within the epidermis. Topical administration of 7,12-dimethylbenz(a)anthracene and 12-O-tetradecanoylphorbol-13-acetate for 3 wk increased epidermal cell proliferation by 55% in SENCAR mice. Topical addition of lunasin, an anti-mitotic agent from soy, decreased epidermal cell proliferation modestly though significantly (16% given alone, 9% given with carcinogens). Caloric restriction (by 33% of energy intake) for 4 wk decreased the epidermal cell proliferation rate by 45% in C57BL/6J mice. In summary, epidermal cell proliferation can be measured in vivo using (2)H(2)O labeling in normal, hyper- and hypo-proliferative conditions. Potential applications of this inherently safe method in humans might include studies of psoriasis, wound healing, chemopreventive agents, and caloric intake.

Induction of vascular endothelial growth factor by fibrin as a dermal substrate for cultured skin substitute

In the initial phase of wound healing, endogenous fibrin clots are known to form a provisional matrix and to promote angiogenesis. Growth factors such as vascular endothelial growth factor (VEGF) increase in wounds to stimulate angiogenesis. However, it remains unknown whether VEGF is induced when fibrin is used as a dermal substrate for cultured skin substitutes. The authors investigated the effect of fibrin gel as a dermal substrate for a cultured skin substitute, using human keratinocytes and dermal fibroblasts. A collagen-cultured skin substitute was also examined for comparison. VEGF in the culture supernatant in both types was measured by enzyme-linked immunosorbent assay, and VEGF mRNA was determined semiquantitatively by reverse-transcriptase polymerase chain reaction after 2 days of incubation. Experiments were performed using 12 cultured skin substitutes: four for histologic examination before transplantation, four for VEGF assay in vitro, and four for the transplantation to athymic mice. Three independent experiments were performed for each step. VEGF concentration in the fibrin-cultured supernatant was 84.3 +/- 11.8 pg/ml, whereas it was 27.8 +/- 4.68 pg/ml in the case of the collagen substrate. The relative levels of VEGF mRNA were 1.088 +/- 0.100 and 0.698 +/- 0.226, respectively. In in vivo transplantation, the fibrin-type cultured skin substitute showed an excellent take on the wound bed, and a normally proliferating keratinocyte layer with emergence of vascular endothelial cells in the transplanted floor was seen 3 days after transplantation. Vascular endothelial cells, which were identified using alkaline phosphatase stain, were significantly increased in the fibrin-type cultured skin substitute. The use of fibrin as a dermal substrate for cultured skin substitute increases the secretion of VEGF, improves regeneration of mature epidermal structure after in vivo transplantation, and promotes the migration of vascular endothelial cells.

Long-term remodeling of a bilayered living human skin equivalent (Apligraf) grafted onto nude mice: immunolocalization of human cells and characterization of extracellular matrix

Type I collagen is a clinically approved biomaterial largely used in tissue engineering. It acts as a regenerative template in which the implanted collagen is progressively degraded and replaced by new cell-synthesized tissue. Apligraf, a bioengineered living skin, is composed of a bovine collagen lattice containing living human fibroblasts overlaid with a fully differentiated epithelium made of human keratinocytes. To investigate its progressive remodeling, athymic mice were grafted and the cellular and the extracellular matrix components were studied from 0 to 365 days after grafting. Biopsies were analyzed using immunohistochemistry with species-specific antibodies and electron microscopy techniques. We observed that this bioengineered tissue provided living and bioactive cells to the wound site up to 1 year after grafting. The graft was rapidly incorporated within the host tissue and the bovine collagen present in the graft was progressively replaced by human and mouse collagens. A normal healing process was observed, i.e., type III collagen appeared transiently with type I collagen, the major collagen isoform present at later stages. New molecules, such as elastin, were produced by the living human cells contained within the graft. This animal model combined with species-specific immunohistochemistry tools is thus very useful for studying long-term tissue remodeling of bioengineered living tissues.

Injection of CD4+ and CD8+ cells with donor or host accessory cells induces acute graft-vs-host disease in human skin in immunodeficient mice

OBJECTIVE

We examined cell subsets with respect to cutaneous graft-vs-host disease by cell sorting selection of subsets of human mononuclear cells and injecting the subsets subcutaneously in a mouse model.

MATERIALS AND METHODS

Cell suspensions containing cultured human epidermal cells and dermal fibroblasts from a single donor mixed with lymphoid cell subsets positively selected using the FACSVantage cell sorting instrument and/or MACS cell isolation kits from unrelated individuals were injected into immunodeficient mice. This model is known to generate human skin with histologic findings similar to human graft-vs-host disease.

RESULTS

Donor T-cell subsets CD4(+) and CD8(+) plus either host or donor CD14(+) cells were necessary to cause acute cutaneous graft-vs-host disease. Although graft-vs-host disease can result from recognition of class I antigens expressed on human cutaneous cells by donor peripheral blood mononuclear cells, additional recognition of class II antigens expressed on host mononuclear cells resulted in more severe histologic manifestations. Dendritic cells that differentiated from donor and host monocytes also showed competent accessory cell function in this system.

CONCLUSIONS

Based on this model, human cutaneous graft-vs-host disease was caused by donor CD4(+) cells and CD8(+) cells activated through recognition of host antigens, including class I and class II antigens presented by either donor or host CD14(+) cells or dendritic cells.

Transfer of SV40 temperature-sensitive early gene into human epidermal keratinocytes by the recombinant adenovirus vector

We constructed a recombinant adenovirus vector that contained the origin-defective SV40 early gene, coding temperature-sensitive T antigen. This vector transferred the SV40 early gene into human epidermal keratinocytes with high efficiency. T antigen conferred the ability of keratinocytes to grow with limited differentiation in the presence of serum and high calcium concentration at the permissive temperature (34 degrees C), although normal keratinocytes were induced to differentiate and stop growing under the same conditions. The serum/Ca++-resistant cells did not proliferate at the nonpermissive temperature (40 degrees C), indicating that they depended on T antigen for their proliferation. The temperature-sensitive T antigen dissociated from the tumor suppressor gene products, p53, at 40 degrees C. The serum/Ca++-resistant cells still had the ability to proceed to terminal differentiation when injected into SCID mice as cultured keratinocytes. However, they did not form an apparent basal layer. This indicated that the tissue remodeling process in the serum/Ca++-resistant keratinocytes was abnormal. All of these epidermoid cysts disappeared within 8 wk and no tumor developed for 6 mo. We consider that deltaE1/SVtsT is a useful tool to examine multistep carcinogenesis of human epithelial cells in vitro.

An in vivo model of human skin acute graft-versus-host disease: transplantation of cultured human epidermal cells and dermal fibroblasts with human lymphocytes into SCID mice

The ability of mixed epidermal cell-lymphocyte reactions to detect allogeneic reactivities in an in vivo model was investigated by developing an in vivo model of acute graft-versus-host disease (GVHD), using SCID mice with a C.B-17 background in which human skin structures were generated by transplantation of cultured human epidermal cells (HEC) with dermal fibroblasts (HDFC). Suspensions containing cultured HEC and HDFC from a single donor were mixed with autologous peripheral blood mononuclear cells (PBMNC) or with PBMNC from unrelated individuals, and were injected into the flanks of C.B-17-SCID mice. Ten and 21 days after injection, subcutaneous nodules generated in the mice were examined histologically and immunohistochemically. Cystic structures developing after injection of HEC and HDFC without human PBMNC showed normal epidermislike tissue. Human skin generated in SCID mice injected with HEC and HDFC with auto-PBMNC showed no graft-versus-host reaction (GVHR) histologically, whereas those mice injected with PBMNC from siblings that shared an HLA haplotype showed mild GVHR. Human skin in SCID mice injected with HEC and HDFC with histoincompatible unrelated PBMNC showed moderate to severe GVHR. The severity of GVHR paralleled the dose of unrelated PBMNC, and GVHR was prevented by peroral treatment with cyclosporine A. Immunohistochemically, inflammatory cells infiltrating human cutaneous tissue formed in the SCID mice were stained by an anti-human CD45RO antibody that reacts with human T cells but not with murine lymphocytes, and most T cells were stained by an anti-human CD8 antibody recognizing HLA class I antigens. These findings are similar to those in clinical skin graft-versus host disease (GVHD) observed in patients undergoing allogeneic bone marrow transplantation. This experimental system should be useful as an in vivo model of human skin GVHD.

Cultivation and transplantation of epidermal keratinocytes

Transplantation of autologous cultured keratinocytes is the most advanced area of tissue engineering which has clinical application in restoration of skin lesions. In vitro, disaggregated keratinocytes undergo activation and after adhesion and histogenic aggregation form three-dimensional epithelial sheets suitable for grafting on prepared wounds that provide a reparative environment. Epidermal stem cells survive and proliferate in culture, retaining their potential to differentiate and to produce neoepidermis. Reconstructed skin is physiologically compatible to split-thickness autografts. Autotransplantation of cultured keratinocytes is a promising technique for gene therapy. In many cases allografting of cultured keratinocytes promotes wound healing by stimulation of epithelialization. Banking of cryopreserved keratinocytes is a significant improvement in usage of cultured keratinocytes for wound healing. Skin substitutes reconstructed in vitro that have morphological, biochemical, and functional features of the native tissue are of interest as model systems that enable extrapolation to situations in vivo.

Interaction of cultured keratinocytes and fibroblasts from human psoriatic and normal skin in immunodeficient mice

Clarification of the pathogenesis of psoriasis requires separate studies of the epidermis, dermis, and inflammatory cells. We previously subcutaneously transplanted a mixture of cultured human keratinocytes and fibroblasts into mice to develop cysts with human skin structures. Using this method, we separately cultured psoriatic and normal keratinocytes and fibroblasts. Four mixtures were prepared: normal keratinocytes and normal fibroblasts (NK/NF); psoriatic keratinocytes and normal fibroblasts (PK/NF); normal keratinocytes and psoriatic fibroblasts (NK/PF); and psoriatic keratinocytes and psoriatic fibroblasts (PK/PF). Each mixture was transplanted into immunodeficient mice to observe formation of cysts and histological changes. The cysts varied in structure depending on the mixture, which suggests that psoriatic keratinocytes and fibroblasts had some abnormalities. Psoriatic fibroblasts may be partially responsible for thickening of the epidermis. Cell differentiation might have been accelerated in psoriatic keratinocytes after transplantation, resulting in the loss of epidermis structures.

Experimental study of epithelialization of the muscle-only flap in the oral cavity

PURPOSE

The purpose of this study was to observe the epithelialization process of the muscle-only flap used for reconstruction of the oral mucosal defects.

MATERIALS AND METHODS

Forty-three male adult Japanese rabbits were used. A superiorly based cleidomastoid muscle flap was designed after vascular assessment. The flap was transferred into the oral cavity to cover a mucoperiosteal defect made in the mandibular alveolus. Epithelialization of the flap was histologically evaluated at designated intervals.

RESULTS

The flaps survived without ischemic necrosis. By 8 days postoperation, the flap was infiltrated by acute inflammatory cells and being replaced by granulation tissue originating from the adjacent tissues. The oral epithelial cells advanced onto this granulating muscle flap, with eventual coverage by 21 days. The granulation tissue matured to fibrous tissue with significant contraction by 2 months. At 6 months postoperation, abnormally hyperkeratinized epithelium was seen on the flap. This differed from the surrounding parakeratinized oral epithelium.

CONCLUSIONS

The muscle-only flap in the oral cavity epithelializes after the granulation process.