Search

EPC-1 (early population doubling level cDNA-1) is a quiescence-specific gene expressed at high levels by early passage WI-38 fibroblasts under conditions of either density-dependent growth arrest or serum deprivation. Late passage WI-38 cells lose the ability to express EPC-1 under all conditions tested. The decline in EPC-1 mRNA is gradual during the replicative life span and correlates inversely with the population doubling level (PDL) of the cells. The objective of this study was to determine whether the decline in EPC-7 mRNA abundance observed during proliferative senescence also occurs in cultures derived from donors of different ages. To address this question, we examined the abundance of EPC-1 mRNA in 28 skin fibroblast lines established from healthy donors of different ages ranging from 12 fetal weeks to 94 years. EPC-1 expression was measured, under conditions of growth arrest, prior to the end of the replicative life span of the cultures. Despite some variability in steady-state transcript levels among the cell lines, EPC-1 expression was significantly lower in cells derived from the fetal donor group (12-20 gestational weeks) than in cells derived from adult donors. An in vitro age-dependent decline in EPC-1 expression was observed in all the skin lines examined, independent of donor age; however, no significant difference was observed between the young adult donor group (17-33 years) and the old adult donor group (78-94 years). Thus, expression of EPC-1 is linked to the replicative age of the cells and whether the cells are derived from fetal skin or adult skin. In adults, EPC-1 expression is independent of donor age.

We have examined the activities and mRNA abundance of two hydrogen peroxide metabolizing enzymes (glutathione peroxidase and catalase), glutathione concentration, and the activities of several enzymes that influence glutathione concentration, including glutathione reductase (GR), glucose-6-phosphate dehydrogenase (G-6-PD), and gamma-glutamylcysteine synthetase (gamma-GCS), in 29 skin fibroblast lines derived from donors ranging in age from 14 gestational weeks to 94 years of age. H2O2 metabolizing enzyme activities and mRNA abundances were greater in skin fibroblast cultures established from postnatal donors than in fetally derived cultures. There were no significant differences in either of these parameters in cell lines established from postnatal donors of different ages. Total glutathione concentration decreased with age, but GR activity appeared to be unaffected by age. In order to estimate the ability of the cultures to produce NADPH (an important component of cellular redox status and a cofactor for GR), we determined glucose-6-phosphate dehydrogenase activity and mRNA abundance. We were unable to directly measure gamma-GCS activity or mRNA abundance in any of the skin lines or in fetal lung fibroblast; however, we were able to indirectly demonstrate the presence of this enzyme by stimulating fetal lung fibroblasts with H2O2 following treatment with L-buthionine-S,R-sulfoximine (BSO), an inhibitor of gamma-GCS activity. These results show that some, but not all, age-associated differences in antioxidant defense levels are maintained in a culture environment and are consistent with the hypothesis that developmental stages of life are associated with lower antioxidant defense levels than are present in postnatal phases of life.

The transcript for EPC-1 (early population doubling level (PDL) cDNA-1) is induced under conditions of growth arrest due to density-dependent contact inhibition and/or serum deprivation in early-passage but not in senescent WI-38 fibroblasts. We have characterized the EPC-1 transcript with respect to its cell-cycle regulation, tissue specificity, and interspecies conservation of related genomic sequences. In low density, quiescent (serum-deprived), early-passage fibroblasts that are stimulated to proliferate with fresh serum, steady-state EPC-1 transcript levels are steadily reduced until they reach a basal level approximately 24 h after stimulation. However, when early-passage fibroblasts are made quiescent by both serum deprivation and density-dependent contact inhibition and then stimulated with serum, steady-state EPC-1 transcript levels remain relatively constant throughout a 36 h period following serum stimulation. Senescent WI-38 cells (> 95% life span completed) do not express EPC-1 under these conditions. We show that differences in the regulation of EPC-1 transcript levels in early-passage cells are due to differences in growth state rather than changes in cell density or contact. We also show that expression of the EPC-1 transcript is limited to specific cell types and that related genomic sequences are found in all mammalian species examined as well as in the chicken.

As normal human cells approach the end of their proliferative lifespan in vitro they lose responsiveness to a variety of growth factors, to which they respond with DNA replication when they are young. Recently it has been reported that the protooncogene c-fos is not expressed in senescent cells (Seshadri and Campisi, 1990). In this study we have found that both c-jun and jun B, partners of c-fos in heterodimeric transactivating complexes, are equivalently expressed in young and senescent cells at both early (1–6 hr) and late (12 or 16 hr) time points following serum stimulation of quiescent cells. We have also investigated the effect of the enforced expression of c-fos in senescent WI-38 cells using an inducible construct carrying the murine c-fos gene under the control of the sheep metallothionein promoter. We have found that the transient transfection and subsequent activation of the conditional promoter with Zn++ stimulated DNA synthesis in a significant fraction of senescent cells which had completed 90%–95% of their proliferative lifespan. However, populations which had completed 100% of their proliferative life span and nondividing cultures which had been selected with BrdU did not respond to the expression of the c-fos gene. These results demonstrate that one of the primary events associated with senescence in human cells is the suppression of c-fos gene expression, but additional phenotypic changes must also occur in order to explain the ultimate loss of proliferative responsiveness of these cells. © 1992 Wiley-Liss, Inc.

EPC-1/PEDF expression is closely associated with reversible growth arrest in normal human diploid fibroblast-like (HDF) cells and is diminished with proliferative senescence in vitro. EPC-1 expression in HDF cells is induced under conditions of density-dependent contact inhibition and growth factor deprivation. Antiserum generated against EPC-1 recognizes a secreted protein of approximately 50 kDa from medium conditioned by early passage HDF cells, but not from senescent cells. The addition of EPC-1 antiserum to early population doubling level (PDL) cultures near the plateau phase of growth significantly increases the number of cells entering DNA synthesis. Affinity purified EPC-1 antibodies alone enhance the ability of near plateau-phase early PDL WI-38 cells to synthesize DNA by as much as threefold. Further, the addition of recombinant EPC-1 (rEPC-1) to logarithmically growing cells resulted in a marked decrease in the ability of these cells to enter DNA synthesis. We also demonstrate the loss of EPC-1 expression in WI-38 and IMR-90 HDF cell lines with both senescence and simian virus 40 (SV40) transformation. The loss of EPC-1 expression with SV40 transformation occurs at the level of steady-state mRNA and protein accumulation with genomic EPC-1 sequences grossly intact. Taken together, these results suggest that EPC-1 may play a role in the entry of early passage fibroblasts into a G0 state or the maintenance of such a state once reached. © 2003 Wiley-Liss, Inc.

We have determined the activities, protein, and mRNA abundances as well as the level of transcriptional activation of two intracellular forms of the free radical metabolizing enzyme superoxide dismutase in 29 human skin fibroblast lines established from donors of different ages. SOD-1 (a copper and zinc containing form of SOD) and SOD-2 (a manganese containing form of the enzyme) activities were both observed to be significantly lower in cell lines derived from fetal skin than in lines established from postnatal skin (ages 17-94 years). The percent of total activity contributed by SOD-1 decreased in an age-associated manner from approximately 50% in the fetal lines to less than 20% in lines established from old tissue donors. All of the cell lines were screened to exclude the possibility that they contained a polymorphism known to influence SOD-2 activity. Northern blot analysis revealed three SOD-1 mRNA transcripts that were 0.5, 0.7, and 1.9 kb in length. Although SOD-1 protein abundance was lower in fetal lines than in lines derived from postnatal donors, SOD-1 mRNA abundance did not differ between fetal cells and cell lines derived from young donors. SOD-2 protein abundance and mRNA abundance were both significantly lower in fetal lines than in postnatal lines. No postnatal age-dependent differences were observed in any of the SOD-2 parameters examined. Nuclear run-on analysis revealed that fetal cell lines exhibited a lower level of transcriptional initiation for SOD-1 than postnatal lines. The transcription of SOD-2 was readily detected in postnatal lines, but undetectable in fetal lines. These results are consistent with multiple levels of control for SOD-1 expression and with a strong transcriptional influence on SOD-2 expression.

Transcriptional activation of c-fos in response to both serum stimulation and DNA damage requires the serum response element. The inability of in vitro aged or senescent fibroblasts to proliferate in response to serum has been shown to be associated with repressed c-fos expression and reduced AP-1 binding activity. In contrast, we have observed similar levels of c-fos mRNA and protein expression in young (early passage) and old (late passage) cells following their treatment with ultraviolet (UV) irradiation or methyl methanesulfonate (MMS). Thus, the early events in the signal transduction pathway leading to transcriptional activation of c-fos following DNA damage are distinct from those mediating the gene's expression in response to mitogenic stimulation. Despite normal levels of c-fos expression, we observed a reduced level of AP-1 binding activity in old cells relative to young cells treated with UV irradiation or MMS. Reduced AP-1 binding activity is associated with reduced expression of the AP-1-dependent gene, collagenase, in old cells treated with DNA damaging agents. Since other DNA damage-inducible genes also contain an AP-1 regulatory element presumed to play a role in their expression, reduced AP-1 binding activity is likely to have a major impact on the old cell's ability to respond appropriately to DNA damage. © 1995 Wiley-Liss, Inc.

Insulin-like growth factor-I (IGF-I) (13 nM) can replace insulin (0.8 microM) in a serum-free medium containing epidermal growth factor (EGF) (16 nM) and dexamethasone (DEX) (140 nM) and stimulate DNA synthesis in young cultures of WI-38 cells, similar to the stimulation of serum-supplemented medium. By contrast, senescent cells become unresponsive to all of these hormones. The effect of IGF-I, EGF, and DEX is synergistic in stimulating multiple rounds of low density cell division. Total specific binding of [125]IGF-I per cell in monolayer culture does not change with age, which indicates, in light of increased cell size with age, an actual decrease in specific binding per micron2 of cell surface area. Binding can be traced to two separate cell proteins. Binding to the alpha subunit of the IGF-I transmembrane receptor may increase slightly with age while the 50% displacement remains unchanged. The remainder of the IGF-I specific binding (five- to thirty-fold more) is to a low molecular weight, cell-associated binding protein whose 50% displacement is 10 times higher, but also remains unchanged with age. Specific binding to the lower affinity sites decreases slightly with age at equal cell densities. IGF-I binding to the alpha subunit of the transmembrane receptor is independent of cell density, while binding to the low molecular weight binding protein is inversely proportional to cell density and may vary by as much as tenfold.