Virginia W. Sykes

SETA is a multifunctional adapter protein with three SH3 domains that binds Grb2, Cbl, and the novel SB1 proteins

Expression of the src homology 3 (SH3)-encoding, expressed in tumorigenic astrocytes (SETA) gene is associated with astrocyte transformation in culture and tumors in the adult brain. SETA binds to the apoptosis regulator apoptosis-linked gene 2 (ALG-2) interacting protein 1 (AIP1), and modulates apoptosis in astrocytes. The predicted protein structure of SETA revealed two SH3 domains, while related proteins were reported to have three. Here we report the identification of an additional SH3 domain N-terminal to the previously identified SETA sequence. Yeast two-hybrid screening of a p53(-/-) astrocyte cDNA library with this SH3 domain identified a novel gene, SETA binding protein 1 (SB1), with 55% amino acid identity to the renal tumor antigen, NY-REN-45. In vitro confrontation and co-immunoprecipitation experiments confirmed the binding of SB1 to SETA. Evidence that SETA binds to the CD2 protein, the proto-oncogene c-Cbl, and the signal transduction molecule Grb2, and can dimerize via its C-terminal coiled coil (CC) domain is also presented.

SETA: A novel SH3 domain-containing adapter molecule associated with malignancy in astrocytes

Differential display polymerase chain reaction analysis was used to compare five differentiation states of the O-2A progenitor-like cell line CG4: progenitor cells and cells at 12 h or 4 days after the induction of differentiation into oligodendrocytes or astrocytes. This led to the identification of 52 sequence tags that were expressed differentially with cellular phenotype. One sequence was upregulated during differentiation of CG4 cells and represented a novel gene that we named SETA (SH3 domain-containing gene expressed in tumorigenic astrocytes). This gene encodes an SH3 domain-containing adapter protein with sequence similarity to the CD2AP (CD2 adapter protein) and CMS (Cas ligand with multiple Src homology) genes. SETA mRNA was expressed at high levels in the developing rat brain but was barely detectable in the normal adult rat or human brain. However, SETA mRNA was found in approximately one half of the human gliomas tested, including astrocytomas grades II, III, and IV, as well as oligodendrogliomas, mixed oligoastrocytomas, and human glioma-derived cell lines. A rat glioma generated by treatment with the alkylating carcinogen ethylnitrosourea on postnatal day 1 and a derived cell line also expressed SETA mRNA. Furthermore, in an in vitro model of astrocytoma progression based on p53-/- astrocytes, expression of SETA was restricted to cells that are tumorigenic.

Differential Expression of Procollagen Genes Between Mid- and Late-Gestational Fetal Fibroblasts

BACKGROUND Previously, we have shown that cutaneous wounds in mid-gestational (E15) mice heal in a scarless manner with decreased procollagen 1 and increased procollagen 3 production compared with wounds in late-gestational (E18) mice, which heal with scars. The aim of the current work was to determine whether E15 and E18 fibroblasts respond to stimulation in culture with differential procollagen expression, suggesting they may preserve their phenotype in vitro. Further, we wanted to determine if fetal fibroblast gene expression patterns persisted in tissue culture. We measured expression of procollagen types 1alpha1 and 3 in response to TGF-beta1 stimulation. We theorized that E15 fibroblasts would respond with a pattern of procollagen that would contribute to a more easily remodeled collagen. METHODS Mid- and late-gestational fetal fibroblasts were obtained from dorsal skin harvested from fetuses of time-dated CD-1 mice. Cells were grown to confluence in culture plates overnight. Cell monolayers were treated with 0.01% bovine serum albumin (BSA) plus 10 ng/well of TGF-beta1. Cells were harvested at 6 and 24 h following treatment. Additional groups were treated with BSA alone (vehicle controls) and collected at 6 and 24 h. Another group without treatment was harvested after reaching confluence (0 time point). In a separate experiment to determine if gene expression patterns persisted, cells were treated with 0.01% BSA plus 10 ng/well of TGF- beta1 for 24 h, then harvested. A second group of cells were treated again at 24 h and harvested at 48 h. Additional cells were treated with BSA alone for 24 and 48 h, and another group without treatment was harvested after reaching confluence (0 time point). Cells were processed to obtain mRNA, cDNA was made, and then samples analyzed by QPCR. Results were analyzed by ANOVA and Holm-Sidak method. RESULTS Procollagen 1alpha1 gene expression was decreased in E15 cells at 6 and 24 h following TGF-beta1 treatment, P<0.05. In contrast, procollagen 1alpha1 was increased in E18 cells, P<0.05. Procollagen 3 gene expression was decreased in E18 cells at 6 and 24 h following treatment with TGF-beta1, P<0.05, whereas levels in E15 cells were unchanged at 6 h, and only trended lower at 24 h. We evaluated whether this differential expression of procollagen 3 persisted at 24 and 48 h. At 24 and 48 h, E15 control groups had increased procollagen 3 expression compared with E18 groups, P<0.05. E15 and E18 cells in TGF-beta1-treated groups had decreased procollagen 3 at 48h compared with their respective BSA control groups, P<0.05. However, the degree of difference appeared to be greater in the E15 group than the E18 group. CONCLUSIONS Our results from this in vitro work demonstrate a differential pattern of gene expression for procollagen 1alpha1 and 3 in E15 and E18 fibroblasts in response to TGF-beta1. E15 cells showed decreased expression of procollagen 1alpha1, while E18 cells showed increased procollagen 1alpha1 and decreased procollagen 3 expression. These patterns of expression in E15 cells are suggestive of increased type 3 to 1 collagen ratio seen in scarless fetal wounds. Interestingly, treatment of either E15 or E18 cells with TGF-beta1 significantly decreased procollagen 3 expression by 48 h, yet this was more profound in E15 groups. This suggests that after 24 h, E15 cells may transition towards an E18 phenotype and corresponding signaling.

Scarless Fetal Mouse Wound Healing May Initiate Apoptosis Through Caspase 7 and Cleavage of PARP

INTRODUCTION Apoptotic mechanisms are thought to be important in wound healing for the removal of inflammatory cells and evolution of granulation tissue. However, little is understood about the signal, propagation, and mechanisms responsible for triggering cell death in tissue injury, particularly during fetal wound repair. Understanding these signals may lead to insights regarding scarless wound healing. We hypothesized that differences in apoptosis would exist in mid- (E15) compared with late-gestational (E18) mice subjected to cutaneous wounds. We examined early apoptotic signals that may be initiated following tissue injury. METHODS Pregnant, time-dated mice underwent laparotomy and hysterotomy on embryonic day 15 (E15) and 18 (E18). Full-thickness, excisional cutaneous wounds were made on the dorsum of the fetuses and dorsal skin harvested 15 and 45 min after wounding. Unwounded dorsal skin from additional fetuses collected at the same time points served as controls. The skin was processed to obtain protein, then levels of caspase 3, caspase 7, and poly ADP-ribose polymerase (PARP) were measured by Western blot. Cyclophilin levels were measured to ensure equal loading of protein. Histone-associated DNA complex formation was examined to provide further evidence of cellular apoptosis. RESULTS There were no differences in total caspase 3 levels between E15 and E18 fetal tissue with or without wounding, nor was any cleavage of caspase 3 noted in any group. However, cleaved caspase 7 was present in the E15 skin with a >2-fold increase following wounding at both 15 and 45 min, yet absent in the E18 groups. Furthermore, levels of cleaved PARP were also increased by >2-fold at both 15 and 45 min in E15 wound groups, whereas a relatively small amount was only seen in the E18 wound groups at 45 min. DNA-histone fragmentation ELISA assay showed a 5-fold increase in the enrichment of histone-associated DNA fragments in the E15 wounded tissue compared with the time-matched controls at 45 min. This was not seen with the E18 tissue. CONCLUSIONS Previously, we demonstrated that cutaneous wounds in E15 fetal mice heal in a scarless manner, while similar wounds in E18 mice heal with scar formation. Results from our current work demonstrate differences in apoptosis in mid- compared with late-gestational mouse skin as well as shortly after wounding. Our results suggest that in mid-gestational wounds, activation of apoptotic pathways may be mediated through effector caspase 7 signals with inactivation of PARP. This initiation of apoptotic signals following tissue injury may play a role in scarless wound repair.

Homeobox Genes Hoxd3 and Hoxd8 Are Differentially Expressed in Fetal Mouse Excisional Wounds

BACKGROUND Cell signaling pathways underlying wound repair are under extensive investigation; however, there is still a poor understanding of the mechanisms orchestrating these processes. Hox genes, which are a subgroup of homeobox genes, encode for a family of transcription factors that play a critical role in tissue migration and cell differentiation during embryogenesis and may also serve as master regulatory genes of postnatal wound repair. We have developed a fetal excisional wound healing model whereby mid-gestational wounds heal in a regenerative manner while late-gestational wounds display scar formation. We theorize that Hoxd3 and Hoxd8 will be differentially expressed in mid- and late-gestational wounds compared with normal skin. MATERIALS AND METHODS Pregnant FVB mice underwent hysterotomy at mid (E15)- or late (E18)-gestational time points, and 3-mm excisional wounds were made on the dorsum of each fetus. Wound samples (w) were collected at the site of injury as well as near wound normal skin (nwc) on the same fetus. Control (c) skin samples were also obtained from unwounded adjacent fetuses. Samples were harvested at 3 and 6 h and real-time polymerase chain reaction was performed for Hoxd3 and Hoxd8 and normalized to glyceraldehyde-3-phosphate dehydrogenase. Data were analyzed by analysis of variance with statistical significance of P < 0.05. RESULTS Hoxd3 levels were increased in all of the mid-gestational groups, with a significant increase at 3 h compared with late-gestational control groups. In the 3-h time group, Hoxd8 is increased in mid-gestational wounds compared with late-gestational control skin. This is repeated in the 6-h time group, where Hoxd8 is increased in mid-gestational wounds compared with all groups. Also, Hoxd8 in the mid-gestational near wound controls is significantly greater than that in the late-gestational near wound control and control groups. CONCLUSIONS These data suggest that Hoxd3 is constitutively expressed in the skin of mid-gestational mice. However, Hoxd8 expression is increased in the mid-gestational wounds compared with normal control groups and late gestational wounds, suggesting that it may play a role in scarless wound repair.

Assignment of Seta to distal mouse X chromosome by radiation hybrid mapping

The Seta (SH3 containing, expressed in tumorigenic astrocytes) gene, originally isolated from rat, is expressed in association with malignant transformation in astrocytes and in human gliomas (Bögler et al., 2000). It is part of a new family of adapter molecules with three SH3 domains, which includes CD2AP (Dustin et al., 1998) and CMS (Kirsch et al., 1999). These molecules interact with cytoskeletal and cell signaling proteins. In order to identify Seta’s chromosome location we have mapped it in the mouse genome using radiation hybrids.