Dr. Rebecca Morris Professor, Stem Cells and Cancer Ph.D.
The focus of research in the Morris laboratory is on stem cells and non-melanoma skin cancer. This past year, we have made progress in two areas. First, we have identified a candidate gene of interest in the Ksc2 locus on chromosome 4 of the mouse. Second, we are particularly interested in the CD49f+/CD34+ hair follicle stem cells and mesenchymal and hematopoietic stem cells of the bone marrow in the context of chemically-induced and solar radiation-induced models of skin cancer.
The skin provides an anatomical barrier to physical, chemical, and biological agents. Hence, it is not surprising that it should have well-developed innate immunity. What is surprising, however, is that the CD49f+/CD34+ hair follicle stem cells have an enriched expression profile of so many genes involved in innate immunity. Do these stem cells require extra protection from environmental insults? Or, could there be a new role for these genes in keratinocyte stem cell mobilization? To address these questions, we have designed and performed functional characterization of the gene for Toll-Like Receptor 4 (Tlr4) identified as a gene of interest in the mouse Keratinocyte Stem Cell Locus 2 (Ksc2). We found that a quantitative trait locus (QTL) we previously named keratinocyte stem cell locus 2 (Ksc2), was associated with the largest keratinocyte colonies following genetic studies in mice. Further fine mapping within the Ksc2 locus revealed a candidate stem cell regulatory gene, Tlr4.
We found that, in the Tlr4 gene of the C57BL/6 (allele G) and the BALB/c (allele A), there is a non-synonymous single nucleotide polymorphism (SNP), i.e., rs13489097 (G/A) located at the genomic location, 66502287 of mouse chromosome 4. We, therefore, selected Tlr4 as a relevant candidate gene of interest within the Ksc2 QTL. We discovered that the Tlr4 gene is expressed in freshly harvested keratinocytes from mice. Moreover, Tlr4 is expressed abundantly in enriched CD49f+/CD34+ keratinocyte stem cells from both C57BL/6 and BALB/c mice, and to a much lesser extent in CD34 depleted keratinocytes. We found Tlr4 mRNA and protein expression is much greater in CD49f+/CD34+ keratinocyte stem cells from C57BL/6 mice than from BALB/c mice. This is particularly interesting because keratinocytes from C57BL/6 mice form larger colonies in culture than those from BALB/c mice. Addition of bacterial lipopolysaccharide (LPS), the naturally occurring ligand of Tlr4, to clonal cultures from both strains enhanced in vitro keratinocyte colony formation in a dosedependent manner; however, the effect was greater in colonies from C57BL/6 mice. In contrast, keratinocyte colony formation in C3H/HeJ mice bearing a null mutation in Tlr4, is unaffected by LPS addition Additionally, we found that LPS seems to protect the keratinocytes from apoptosis, as apoptosis was more likely to occur in untreated keratinocytes compared to LPS treated. This finding may be biologically relevant due to a protective effect of Tlr4 signaling on CD49f+/CD34+ keratinocyte stem cells.
We note with interest the recent literature reports that LPS preconditioning enhances the proliferation of mesenchymal stem cells (Wang et al., 2009), and B9 cells (Pedersen et al., 1995). Recently, Arimilli et al. found that exposure to LPS can protect myocytes and dendritic cells from apoptosis (Arimilli et al., 2007; Ha et al., 2008). Additionally, defective LPS signaling has been reported in some mice strains due to Tlr4 gene mutations (Poltorak et al., 1998). Furthermore, an intact Tlr4 signaling mechanism appears to protect mice from chemically induced skin cancer (Yusuf et al., 2008). It, therefore, appears likely that Tlr4 may function in keratinocyte stem cells as a gene regulating proliferation and mobilization as well as a target gene in non-melanoma skin cancer. Going forward, we will continue to investigate the keratinocyte stem cell regulatory genes as target genes in cutaneous carcinogenesis.
In a second area of research, we have hypothesized a relationship between a specific population of hair follicle stem cells and cells of the bone marrow. Hints of such a relationship already exist. First, the notion that bone marrow derived cells can be recruited into adult tissue epithelia and transdifferentiate into epithelial tissues has been documented recently for a number of organ systems. Usually, significant recruitment of bone marrow derived cells is predicated on some sort of severe or chronic damage to the epithelia. The epithelial recruitment is usually attributed to transdifferentiation of the bone marrow derived cells, although another possibility is that it could be dispersed skin stem cells that have “come home.” The mechanism still remains to be determined. Second, it is now generally accepted that cancer cells from the breast and other epithelial tissues can be found in peripheral blood, lymph nodes, and bone marrow. This has been documented by the presence of cytokeratins as well as molecular markers of specific cancers, such as breast cancer in the blood, lymph nodes, and bone marrow. The third hint is the presence of a hematopoietic stem cell marker on a hair follicle stem cell population. Therefore, it is tempting to speculate that the CD34 might be expressed on the hair follicle stem cells because they are related in some way to bone marrow cells. The link among these hints lies perhaps in the use by both tissues of the CXCR4/CXCL12 (SDF1) signaling and chemotactic pathway. The production of SDF1 is known to be involved in the bone marrow-homing of CD34 positive hematopoietic stem cells. Because skin fibroblasts, endothelial cells, and pericytes are known to produce SDF when damaged, they might, therefore, provide a sufficient chemotactic gradient to lure the hair follicle stem cells into the circulation and to home to the bone marrow (a potentially nurturing and protected environment), especially when the skin has been damaged by ultraviolet light.
Therefore, we explored the in vitro migration of primary epidermal keratinocytes towards various conditioned media and cells as bait. For each of three experiments, BD BioCoat Matrigel invasion chambers were activated, according to the manufacturer’s directions, and were seeded with freshly harvested keratinocytes from adult mice. Beneath the chambers, we placed mouse bone marrow cells in conditioned medium (test group) or 3T3 cells, DMEM with 10% fetal bovine serum, DMEM without fetal bovine serum, epidermal keratinocytes from mice, or bone marrow conditioned medium without keratinocytes as negative controls. In the test group, we found a time dependent migration of the epidermal keratinocytes through the matrigel and towards the bone marrow conditioned medium that was significantly greater than their migration towards the controls. These experiments suggest that keratinocytes can follow a chemotactic gradient towards medium conditioned by bone marrow cells. Ongoing studies will determine whether SDF1 made by the bone marrow cells is the active ingredient in the conditioned medium, and whether CD34+ keratinocyte stem cells in vivo have the capacity for extracutaneous migration under certain circumstances.