SEARCH

SEARCH BY CITATION

Keywords:

  • Cell growth;
  • IGF-I;
  • longevity;
  • mutations;
  • oxidation;
  • stress

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

Snell dwarf mice live longer than controls, and show lower age-adjusted rates of lethal neoplastic diseases. Fibroblast cells from adult dwarf mice are resistant to the lethal effects of oxidative and nonoxidative stresses, including the carcinogen methyl methanesulfonate. We now report that dwarf-derived fibroblasts are slow to enter the stage of growth arrest induced by culturing normal cells under standard culture conditions at 20% O2. Dwarf cells cultured at 20% O2 resemble control cells cultured at 3% O2 not only in their delayed growth arrest, but also in their rapid growth rates and resistance to both oxidative and nonoxidative forms of cytotoxic stress. Levels of the heat-shock protein HSP-70 respond to serum withdrawal and stress only in control cells, showing that intracellular signals are blunted in dwarf-derived cells. These data suggest a model in which stable epigenetic changes induced in skin fibroblasts by the hormonal milieu of the Snell dwarf lead to resistance to multiple forms of injury, including the oxidative damage that contributes to growth arrest in vitro and neoplasia in intact mice.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

The Snell dwarf mutation (dw) at the Pit1 locus leads to abnormal differentiation of the embryonic pituitary, producing mice with primary defects in production of growth hormone (GH), thyroid-stimulating hormone (TSH), and prolactin, and secondary defects in insulin-like growth factor I (IGF-I) and thyroxine (Bartke et al., 2001; Miller, 2001; Tatar et al., 2003). These mice, like those with the very similar Ames dwarf syndrome (df/df, a mutation at Prop1), live about 40% longer than littermate controls in standard, specific pathogen-free vivarium conditions (Brown-Borg et al., 1996; Flurkey et al., 2001). The dwarf mice are not only long-lived, but they also appear to age more slowly than control mice, with delayed or decelerated changes in tendon collagen, T-cell subsets, cognition, joint and kidney pathology, cataract development and tumor incidence (Silberberg, 1972; Bartke et al., 2001; Flurkey et al., 2001; Kinney et al., 2001a,b; Ikeno et al., 2003; Vergara et al., 2004). Three other mutants with abnormal production or response to IGF-I also show extended longevity. These include the Ghrhr mutant with low GH production (Flurkey et al., 2001), the GHR-KO mutant with high levels of GH but low responses to GH (Coschigano et al., 2003) and the IGF-I receptor KO, with low responses to IGF-I (Holzenberger et al., 2003). Because each of these three mouse mutants have relatively pure defects in the GH/IGF-I pathway, i.e. without the major changes in thyroid and prolactin status seen in Ames and Snell dwarfs, it seems likely that the delay in aging relates to a significant extent to the deficiency in GH/IGF-I signals. It is noteworthy, however, that lifelong treatment of Snell dwarf mice with thyroxine leads to a partial reversal of the lifespan difference, suggesting that both the GH/IGF-I and the TSH/thyroxine defects may contribute to longevity in the Snell mouse (Vergara et al., 2004). The connection between the abnormal hormonal status and the delayed aging has not yet been elucidated.

Extensive study of genetic mutations that extend lifespan in the nematode Caenorhabditis elegans has shown that many such mutations lead to stress resistance; worms with longevity mutations are typically found to be resistant to the toxic effects of heat, ultraviolet (UV) light, oxidants including hydrogen peroxide and paraquat, and heavy metals such as cadmium (Lithgow et al., 1994; Johnson et al., 1996; Lithgow, 2000). Indeed the extent of lifespan extension in a panel of different mutants was found to be highly correlated with the extent of stress resistance. These observations suggest the hypothesis that the exceptional longevity of these mutant worms may be a consequence of stress resistance per se. Studies with Drosophila also point to the same conclusion. For example, Jun N-terminal kinase (JNK) signaling activity confers both increased resistance to oxidative stress and increases lifespan in Drosophila (Wang et al., 2003, 2005).

To test the idea that extended lifespan of the Snell dwarf mouse was the result of intrinsic cellular stress resistance, we evaluated fibroblast cell lines derived from tail skin of adult mice for resistance to cadmium, heat, hydrogen peroxide, paraquat and UV light and found (Murakami et al., 2003a,b) that cell lines derived from dw/dw mice had increased resistance to each of these agents. However, because these tests were conducted on cells that had been passaged for several weeks prior to evaluation, it seemed likely that their unusual stress resistance represented an epigenetic change, induced by the hormonal milieu in the adult mouse, which led to an enduring alteration in cellular stress sensitivity, which was retained during multiple mitotic cycles in culture. Subsequent work demonstrated that cell lines from young adult Ames dwarf mice and GHR-KO mice were similarly resistant to most of the same stresses in culture (Salmon et al., 2005), showing that the effect was not due to background genes or to factors in a single vivarium and further suggesting that thyroid hormone defects were not necessary for the stress resistance phenomenon. Furthermore, Snell dwarf cell lines were found to be resistant to death induced by methyl methanesulfonate (MMS), a DNA alkylating agent; this, along with the resistance of UV-induced cell death to antioxidants, showed that the cell lines were relatively resistant to multiple forms of injury, including oxidant- and nonoxidant-dependent toxicity.

A comparative study of fibroblast cell lines derived from mouse embryos and from human embryonic lung samples (Busuttil et al., 2003) has shown an interesting species-specific difference in the effects of O2 concentration on continuous cultures. At 20% O2, the concentration most commonly used for tissue culture work, mouse fibroblasts were found to enter a period of growth arrest after 8–10 population doublings (Parrinello et al., 2003), accompanied by genomic instability, low levels of apoptosis and an enlarged, flattened morphology. A small fraction of the mouse cells emerged from the growth arrest with an aneuploid karyotype, rapid continuous growth and altered morphology. Human diploid fibroblasts grown at 20% O2, however, continued to grow at a steady rate for more than 30 population doublings, eventually entering a period of growth arrest, triggered by telomere shortening, from which proliferating variants never emerge. This group found (Parrinello et al., 2003), strikingly, that the growth arrest in mouse cell cultures was prevented when cultures were maintained at a lower level (3%) of O2, a level closer to that characteristic of internal tissues and capillary beds. Notably, mouse embryonic fibroblasts (MEFs) grown in 20% O2 had many more oxidative lesions in DNA and chromosome breaks than human fibroblasts grown at 20% O2. Growth in 3% O2 significantly reduced these abnormalities in the mouse embryo fibroblasts (Busuttil et al., 2003). These authors thus postulated that the growth cessation of mouse cell lines in 20% O2 was a result of oxygen-mediated DNA damage and that human cells were relatively resistant to such damage in comparison with cells from mouse embryos.

Because cells from dwarf mice were relatively resistant to the acute injury caused by high levels of hydrogen peroxide or paraquat, we hypothesized that they might differ from control cell lines in their behavior in continuous culture.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

Proliferation is enhanced and growth arrest delayed in fibroblasts from Snell dwarf mice

Fibroblasts were grown from tail skin of young adult Snell dwarf (dw/dw) mice and from their normal littermates (dw/+). To examine the growth characteristics of the dwarf cells we counted the cells at each passage starting at passage 2 and recorded the cumulative cell numbers. Motivated by the report of Parrinello et al. (2003), each experiment included parallel cultures set up at either 20% or 3% oxygen from each dwarf and control mouse. Three independent experimental sets were conducted, each involving either three or four pairs of dwarf and littermate control donors. Growth curves for all three sets are shown in Fig. 1. The cell lines of Set A (four per condition) were passaged every 5 days; those of Set B (four per condition) were passaged every 4 days; and those of Set C (three per condition) were passaged every 4–7 days, depending on cell density. The N20 (normal cells at 20% O2) cultures in each case show a pattern consistent with previous results (Parrinello et al., 2003) on mouse embryonic fibroblasts, i.e. a period of logarithmic growth, followed by a period of lower net cell accumulation, followed in turn by a resumption of higher net accumulation rates and the appearance of cells with high proliferative capacity.

image

Figure 1. Control cells accumulate slower than dwarf cells at 20% oxygen and slower than both control and dwarf cells at 3% oxygen. (A–C) Normal and dwarf dermal fibroblasts were cultured in 20% or 3% oxygen, as indicated, and log of cumulative cell numbers determined at each passage. The average of three (Set C) or 4 (Sets A and B) independent cultures is shown for each condition. Statistical analysis for passage 8 cells is shown in Table 1. (D) Unlike normal dermal fibroblasts, MEF grown at 3% show no signs of decline, throughout 18 passages (4 individual cell lines shown of each type).

Download figure to PowerPoint

In each of the three experimental sets, culture of normal cells at 3% O2 (‘N3’ lines) delayed but did not entirely prevent the period of diminished cell accumulation, in contrast to the previous study of MEFs (Parrinello et al., 2003). To see if this difference in growth characteristics depended on differences between cells derived from embryos or from adult skin, we conducted a separate analysis of MEF cell lines, shown in Fig. 1(D). None of the four MEF lines cultured at 3% O2 showed any evidence of growth diminution over 18 passages, confirming the observations of Parrinello et al. (2003) and suggesting that adult skin fibroblasts are relatively susceptible to growth arrest even when cultured in a 3% O2 environment.

Figure 1 also shows that adult skin-derived fibroblasts from dwarf mice, whether grown at 20% (‘D20’) or at 3% (‘D3’) O2, show the same resistance to growth arrest exhibited by control cell lines at 3% only. To evaluate this quantitatively we conducted an analysis of variance, using data from three sets of experiments with 11 donors of each genotype, in which cumulative cell number (on a logarithmic scale) was expressed as a multiple of the mean cumulative cell number of the N20 cultures in the same experimental set (Table 1). At passage 8, cumulative cell number was found to be 16-fold higher (P < 0.001) in D20 cell lines compared with N20; thus we concluded that cell accumulation is significantly higher in dwarf than in control cultures at 20% oxygen. Accumulation in N3 cultures was on average 15-fold higher than in N20 cultures (P < 0.001); we concluded therefore that 20% O2 diminished cell accumulation for cell lines from control mice. Accumulation in D3 cell lines was 42-fold higher at passage 8 than in N20 lines (P < 0.001), but was not significantly different from the D20 or N3 result. This same pattern – significantly lower accumulation in N20 lines than in N3, D20, or D3 lines – was seen using data from passages 6 through 9; analysis at later passages is confounded by the appearance of rapidly growing cells in many of the N20 cultures.

Table 1.  Statistical analysis of cell accumulation levels at passage 8: comparisons to N20 cell cultures
Cell lineAdjusted mean ± SD (log10 scale)Fold changeP-value to N20 (anova)
  1. Notes: summary of an analysis of variation for cell accumulation data evaluated at passage 8. anova calculation was done on normalized data, in which accumulation for each of the 44 cell lines (11 of each type) was expressed as the difference between the raw value and the mean values for N20 mice of the same experimental set. Because the data are log (10), a difference of 1 unit corresponds to a tenfold difference in cell number accumulated. The post hoc P-values represent differences from the value in the N20 cultures, using the Sidak adjustment for multiple comparisons. Similar results, i.e. significantly lower accumulation for N20 cells compared with D20, D3, and N3 cells, were also seen in data obtained at passages 6, 7, and 9 (not shown).

N20 0.0 ± 0.12(NA)    0
D20 1.2 ± 0.2416×< 0.001
N31.17 ± 0.1615×< 0.001
D31.62 ± 0.1942×< 0.001

We also evaluated cell lines from Set B at passage 9 for the presence of senescence-associated β-galactosidase. At 4 days after subculturing, all four of the N20 cell lines showed strong staining for SA-βGal, with > 90% positive cells. The proportion of β-Gal positive cells was less than 20% for each of the N3, D3 and D20 cell lines (four lines examined for each condition). In addition, most of the cells in the N20 group had a flattened morphology at this stage, but the unstained cells in the N3, D20 and D3 had the spindle-shaped form typical of early passage cells. This suggests that the growth arrest that we observe may be due to senescence. Assessment of gene expression levels in dwarf and control cell lines, now in progress, will help to delineate further the extent to which the senescence process, induced by 20% O2 in cells from normal mice, is modulated in the dwarf-derived cell lines.

We evaluated cell number and incorporation of 3H-labeled thymidine in the first 72 h after subculturing to provide additional insights into the higher rates of accumulation in D20 cultures compared with control N20 cultures. Figure 2 shows the results of experiments conducted on 14 cell lines of each type at passages 3, 4, or 5, evaluating thymidine incorporation at 60 h after subculturing. Cultures at 3% O2 had significantly higher incorporation than cultures at 20% O2 for both dwarf and normal donors. In addition, dwarf cells had higher levels of thymidine incorporation compared with normal cells; this difference was statistically significant at 3% but not at 20% O2. In addition, we obtained cell counts from four pairs of N20 and D20 cultures evaluated at 1, 2, or 3 days after subculture at 20% and 3% oxygen. At 20% O2, dwarf cell counts were significantly higher at all time points (Fig. 2B). At 3% O2 dwarf cell numbers were significantly higher at days 2 and 3 (not shown).

image

Figure 2. Tests of proliferation using thymidine incorporation (left) and cell counts (right). (A) Higher levels of DNA synthesis at low O2 levels and in cells from Snell dwarf mice, estimated by incorporation of 3H-thymidine into DNA at 60 h after subculturing. Each bar represents mean ± standard error for N = 14 cell lines, normalized so that the mean level of N20 cultures was 1.0 for each experimental set. Cell lines were tested at passage 3, 4, or 5, balanced across experimental conditions for each day's testing. (*) indicates significant difference from the 20% cultures of the same genotype. (+) indicates significant difference between dwarf and control cultures grown in 3% O2. (B) Cell counts at 1, 2, or 3 days after subculturing N20 and D20 cells (Set A at passage 3); initial concentration was at 10 × 103 cells/well. Each bar shows mean and standard error for N = 4 lines, comparing N20 to D20 cultures. Differences between N20 and D20 are significant (P < 0.05) at each time point. A parallel experiment (not shown) found equivalent, statistically significant (P < 0.01) differences between N3 and D3 cells on Days 2 and 3 after subculture.

Download figure to PowerPoint

Analyses of acute stress resistance

Previous work (Murakami et al., 2003b; Salmon et al., 2005) had shown that cells from dwarf mice were more resistant to the acute effects of heat, cadmium, UV, paraquat and H2O2, than cells from normal controls; all such tests were done at 20% O2. The protocol incorporates a 24-h period of serum deprivation prior to stress exposure, because factors in serum increase stress resistance sufficiently to obscure the differences between dwarf and control cultures (Murakami et al., 2003a,b). Table 2 shows LD50 values for stress sensitivity to cadmium, UV, paraquat and H2O2 for cell lines at passage 3 or 4 grown at either 3% or 20% O2 prior to stress exposure. Consistent with previous results, which were conducted using cell lines grown in 20% O2, the D20 cell lines were relatively resistant, compared with N20 lines, to all four forms of injury, although in the current data set the difference for paraquat exposure did not reach statistical significance (P = 0.09). Culturing normal cells in 3% O2 led to a significant increase in resistance to all four forms of stress (compare N20 vs. N3). The effect of culture in 3% O2 was less consistent in cells from dwarf mice (comparing D20 to D3): dwarf cells grown at 3% O2 were more resistant than D20 cells to cadmium and to paraquat, but not to UV or to peroxide. A comparison of N3 to D3 cells showed little difference, except that D3 cells were more resistant to cadmium toxicity. This clearly shows that the cells are compromised by growth in 20% oxygen prior to addition of the cytotoxic stresses, especially the normal cells.

Table 2.  Comparisons of resistance to four forms of acute stress
A Means ± standard errors
ConditionCadmiumUVParaquatH2O2
N201.0 ± 0.091.0 ± 0.131.0 ± 0.131.0 ± 0.04
D202.4 ± 0.271.7 ± 0.271.4 ± 0.171.4 ± 0.17
N32.2 ± 0.361.8 ± 0.145.0 ± 0.531.4 ± 0.16
D34.9 ± 0.552.1 ± 0.374.6 ± 0.711.4 ± 0.18
B P-values for Student's t-tests
ComparisonCadmiumUVParaquatH2O2
  1. Dose–response curves for each stress were used to calculate LD50 values as in (Murakami et al., 2003a, b), and each LD50 was then expressed as a multiple of the mean LD50 level seen for N20 cultures in the same experimental set. N = 14 donors of each type (N20, D20, N3, and D3) for all agents, except N = 7 donors per condition for the paraquat analysis. Boldface font indicates comparisons for which P < 0.05.

N20 vs. D20P < 0.001P = 0.02P = 0.09P = 0.05
N20 vs. N3P = 0.002P < 0.001P < 0.001P = 0.04
D20 vs. D3P < 0.001P > 0.1P = 0.001P > 0.1
N3 vs. D3P < 0.001P > 0.1P > 0.1P > 0.1

Levels of heat-shock protein HSP-70

We hypothesized that the resistance of cells from dwarf mice might represent a generalized induction of stress-resistance defense pathways. Because the heat-shock protein HSP-70 often participates in protection of cells against multiple forms of injury (Jaattela et al., 1992; Mosser et al., 1997), we measured HSP-70 levels by immunoblotting using lysates from seven pairs of N20 and D20 cell lines. Figure 3 shows a summary of the results, in which each HSP-70 level is expressed as a multiple of the level of HSP-70 seen in confluent cells of the N20 sample evaluated in parallel. Because the standard stress exposure protocol includes a 24-h preincubation without serum, we evaluated cells in three conditions: at confluence, or after 24 h without serum, or after both serum removal and 2-h exposure to H2O2 at 100 µm. The data show that at confluence D20 cells do not have significantly more HSP-70 than N20 cells (D20/N20 ratio = 1.3 ± 0.3, P = 0.4). Incubation in serum-free medium induces a significant increase in HSP-70 levels of N20 cells of 2.6-fold (±0.4, P = 0.01) and these levels are significantly higher than those seen in D20 cells after serum withdrawal (P = 0.04). HSP-70 levels after 2-h exposure to H2O2 remain significantly higher (P = 0.01) in N20 cells compared with D20 cells. The levels of HSP-70 in peroxide-treated cultures are not significantly above those produced by serum withdrawal for either normal or dwarf cells. Thus the resistance to injury exhibited by D20 cells does not seem to reflect unusually high resting or induced levels of HSP-70 and indeed the signals that induce HSP after serum withdrawal and peroxide exposure in N20 cells seem to be muted in dwarf-derived cell lines.

image

Figure 3. HSP-70 protein levels increased by serum removal in control fibroblast cell lines but not in dwarf-derived cells. Each symbol shows mean and standard error for 7 independent cell lines, either for confluent cells, after serum withdrawal for 24 h, or after serum withdrawal followed by 2-h exposure to 100 µm H2O2. The (*) indicates a significant difference between normal (N20) and dwarf (D20) cell lines.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

Our principal result is that the growth arrest induced in mouse fibroblast cell lines by 20% O2 is delayed in cells derived from the long-lived Snell dwarf mutant mouse. Because of the evidence that the growth arrest induced in mouse cells by 20% O2 is accompanied by point mutations and chromosomal abnormalities (Busuttil et al., 2003), our observation that dwarf cells grown at 20% O2 resemble those of normal mouse cells at 3% O2 suggests a cellular basis for the delay in lethal neoplastic diseases that accompanies and contributes to the remarkable longevity of mutant mice with abnormalities of the GH/IGF-I axis (Flurkey et al., 2001; Ikeno et al., 2003; Vergara et al., 2004).

Our data on MEF cell lines (Fig. 1D) confirm previous reports (Busuttil et al., 2003; Parrinello et al., 2003) that decline in net cell accumulation typically seen at early passage of mouse primary fibroblasts (Karatza et al., 1984; Smith & Pereira-Smith, 1996) is not an inevitable consequence of in vitro proliferation, but represents a toxic response to concentrations of O2 that are far higher than those typically encountered in the tissues. Continuous culture of MEFs at 20% O2 levels leads to higher levels of somatic mutations (Busuttil et al., 2003) than are seen in MEFs grown at 3% O2. The resistance of human fibroblast cell lines to mutation and growth crisis even at 20% O2 has prompted the speculation that resistance to oxidative damage might contribute to the much greater resistance of human cells in spontaneous transformation in vitro and to the longer cancer-free lifespan of humans compared with mice.

We find that fibroblast cell lines from tail skin of young adult mice, unlike MEF lines, do reproducibly show a decline in cell accumulation, even when grown in 3% O2 (Fig. 1), although crisis is delayed by several passages compared with cells grown at 20% O2, so that by passage 8 the total cell accumulation is, on average, about 15-fold higher in N3 than in N20 cells. We cannot tell without additional work whether this distinction between MEF and tail skin fibroblast cell lines represents an effect of maturation between embryo and young adult, or is instead attributable to differences between fibroblasts derived from dermis and those derived from whole embryos, which must represent a mixture of multiple cell types from diverse organs. It would be of interest to compare fibroblasts from several organs of young adult mice and from mice of a wider range of ages to address the question of whether resistance to this form of oxidative damage varies systematically with age and tissue source.

Cells from Snell dwarf mice at 20% O2 resemble cells from normal mice grown at 3% O2 in their growth curves (Fig. 1). This apparent resistance to the effects of toxic (i.e. 20%) levels of oxygen in culture is consistent with our previous report that these fibroblasts are also relatively resistant to the acute lethal effects of exposure to hydrogen peroxide and to the free radical generator paraquat. The resistance to acute lethal stresses is not limited to oxidative agents: cell lines from Snell dwarf mice are also relatively resistant to death induced by the heavy metal cadmium, by UV irradiation, by heat and by the DNA alkylating agent methyl methanesulfonate (Salmon et al., 2005). Stress resistance is not seen in cell lines derived from mice less than 7 days of age and thus presumably involves an effect of cellular differentiation in the peculiar hormonal environment of the dwarf mouse, with its low concentrations of growth hormone, IGF-I, prolactin and thyroid hormones. Low IGF-I levels appear to be particularly important, in that a similar (although not identical) pattern of multiplex stress resistance develops in skin fibroblasts from GH receptor-knockout mice, which have high levels of GH, low levels of IGF-I and only minor abnormalities in thyroid hormone levels (Bartke et al., 2003; Salmon et al., 2005). The stress resistance involves a stable, epigenetic change, which continues to affect cell properties even after several weeks of growth in a serum-supplemented medium. Our data are consistent with the idea that the resistance of dwarf-derived cells to oxidative damage is responsible for their unusual growth properties.

Our data show two other differences between dwarf and normal fibroblasts, in addition to the increase in the number of cell divisions prior to the period of growth arrest at 20% O2. Dwarf fibroblasts proliferate more rapidly after subculturing, as indicated by cell counts and by thymidine uptake and they also show lower levels of the heat-shock protein HSP-70 after serum withdrawal. The differences in growth rates might represent an early sign of the oxidative damage to DNA that at later passages becomes manifest as a decline in cell accumulation rate; more detailed analyses of cell cycle progression and clone size distribution would be needed to address this issue. Altered growth kinetics might also reflect differences in cellular metabolic pathways. Our unpublished data (Leiser & Miller, in preparation) shows differences between dwarf and normal cells in metabolic responses to low glucose levels and to mitochondrial inhibitors.

The diminished induction of HSP-70 to serum withdrawal is of particular interest. We have shown previously (Murakami et al., 2003b) that the amount of fetal bovine serum present in standard culture medium increases LD50 to multiple acute stresses and that differences in stress resistance between normal and dwarf fibroblasts appear only when this serum-dependent influence is removed by 24 h of culture in serum-free medium. Because heat-shock proteins protect cells against many forms of stress, we suspected that dwarf-derived cell lines would show relatively high levels of HSP-70, either before or both before and after serum withdrawal. Our results show, in contrast, that there is no difference between dwarf and control cells prior to serum withdrawal and that only the control cells respond to serum withdrawal by elevation of HSP-70 levels. There are many other members of the heat-shock protein family and it is possible that some of these may prove to be elevated in dwarf cells. In any case, the HSP-70 data suggest that the signals that lead to induction of HSP-70 levels after removal of serum in control cells are diminished or absent in cells from dwarf mice. It would be of interest to discover how normal fibroblasts sense the change from high to low serum levels and to see if the differences between dwarf and control cells in this response are related to the factors that protect dwarf cells from lethal injury.

The data on resistance to acute stress (Table 2) replicate our previous observations (Murakami et al., 2003a,b; Salmon et al., 2005) that dwarf cells are more resistant than normal cells to multiple forms of stress when cultured in standard conditions. Furthermore, the comparison of N20 to N3 cells now shows that normal cells cultured in low-O2 medium resemble dwarf cells in that they are more resistant than control cells (at 20% O2) to cadmium, UV, paraquat and hydrogen peroxide. We had hypothesized that culture in 20% O2 might lead to induction of antioxidant defense mechanisms and thus render N20 cells more stress resistant than N3 cells, but the data support the opposite conclusion. It seems possible that culture in 20% O2 might lead to a sufficient level of injury that N20 cells might be susceptible to relatively small additional injuries in an acute stress test. If so, however, it is puzzling to note susceptibility of N20 cells not only to oxidant stresses with peroxide and paraquat, but also to UV, which we have shown in this system not to reflect oxidation damage (Salmon et al., 2005). It seems plausible that the similarity between N3 and D20 cells in resistance to acute lethal stresses may provide the basis for the similarities in their relative resistance to the cell growth arrest demonstrated in Fig. 1.

In MEFs, cell growth crisis is accompanied by chromosomal aberrations which contribute to the eventual outgrowth of spontaneously immortalized cell clones. A preliminary analysis of cell lines at passage 24 documented increased chromosome numbers, consistent with aneuploidy, in N3, D20 and D3 cell lines, in addition to the aneuploidy expected in N20 cells (D. Ferguson et al., unpublished data). Thus the delay in the occurrence of growth arrest seen in dwarf cells and in N3 cells (Fig. 1) does not represent a complete avoidance of aneuploidy and immortalization. Additional analyses of the time course for development of chromosomal abnormalities in early passages will be helpful in documenting the relationship between oxygen levels, dwarf origin and karyotypic change in adult skin fibroblasts.

Ikeno and colleagues (2003) have shown that the longer lifespan of the Ames dwarf mice is accompanied by and presumably caused by the delay in development of lethal neoplasia in these mutants. Our own necropsy data, although more limited, support the same conclusion for the hormonally similar Snell dwarf mice (Vergara et al., 2004). Our new data lead to a model that could explain delays in tumorigenesis in these mutant mice in which resistance to oxidative (and perhaps also other kinds of) damage prevents, in vivo, the development of chromosomal abnormality and somatic cell mutations that contribute to neoplasia. There are many gaps in this model, notably including the current absence of information on stress resistance in other dwarf-derived cell types in vitro or in any neoplasia-prone cell type in the intact mouse. Data on the spectrum of mutations seen in dwarf-derived cell lines, analogous to data now available for MEF cells grown in 3% and 20% O2 (Busuttil et al., 2003), would help to support or refute this model and provide further insights into the molecular basis for delayed neoplasia in dwarf mice.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

Fibroblast isolation and cell culture

The dw/dw mice used in this study were bred as the progeny of (DW/J × C3H/HeJ) − dw/+females and (Dw/J × C3H/HeJ) F1-dw/dw males. Dermal fibroblasts were isolated as follows: tail skin biopsies were minced using a scalpel, placed into a 35-mm dish with 1.5 mL of complete medium (DMEM, 10% FCS, 100 U mL−1 penicillin, 100 U mL−1 streptomycin and 0.25 µg mL−1 fungizone) and 1 mL of collagenase solution (1000 U mL−1), then incubated in 3% or 20% oxygen incubators, at 10% CO2. After 24 h the cells were pipetted vigorously to separate cells, passed through netting to remove large particles and centrifuged for 5 min at 200 g. The cell pellet was resuspended in 3 mL of complete medium and transferred to a 25-cm2 flask and cultured to 90% confluence. At this point 0.75 × 106 cells were subcultured into 75-cm2 flasks; this is considered ‘passage 1’. Each culture was passaged at 0.75 × 106 cells per 75-cm2 flask every 4–6 days, depending on the experiment. On the second or third day after each subculturing two-thirds of the medium was replaced with fresh complete medium. Cell numbers were determined by hemocytometer. Each set of experiments made use of cells grown from three or four different dw/dw mice and an equal number of littermate controls, with each cell line propagated in parallel in both 3% and 20% O2.

One experiment used mouse embryo fibroblasts (MEF) prepared from UM-HET3 mice, bred as the offspring of (BALB/cJ × C57BL/6J)F1 females and (C3H/HeJ × DBA?2J)F1 males. For this experiment, 13-day-old embryos were washed and minced in 2 mL of PBS using a scalpel. This suspension was strained through netting, placed into a 25-cm2 flask containing 3 mL of complete medium and incubated in 3% and 20% oxygen incubators, at 10% CO2. Cells that grew from these tissue fragments were allowed to reach 90% confluence (4 days) and then passed to 75-cm2 flasks and cultured to 90% confluency. These cells were seeded at 0.75 × 106 cells into 75-cm2 flasks and considered as passage 2. Each of the tested MEF cell lines was derived from a single embryo and passaged at 0.75 × 106 cells per 75-cm2 flask every 5 days.

Stress assay

Assessment of cytotoxicity after exposure to stress (H2O2, cadmium, paraquat, or UV), was performed as described previously (Murakami et al., 2003a). Cells grown at both 3% and 20% oxygen were assayed in parallel.

Thymidine uptake assay

Fibroblasts were plated at 2500 cells per well (100 µL per well in complete media) in 96-well plates for 24, 60 and 72 h, at which point 3H-thymidine was added to each well (0.5 µCi in 10 µL complete media). After 12 h cells were harvested onto glass fiber filters and scintillation events counted using the TopCount NXT Microplate Scintillation and Luminescence counter (PerkinElmer, Boston, MA, USA).

Western blotting analysis

Total cellular protein extract (30 µg) was prepared in SDS sample buffer, subjected to SDS-polyacrylamide electrophoresis and transferred to PVDF membrane (Millipore, Billerica, MA, USA). The membrane was blocked in 3% BSA, 0.1% Tween 20 in PBS. The Hsp70 (Stressgen, SPA-810), antibodies were used at a dilution of 1/1000 in 1% BSA, 0.1% Tween 20 in PBS. Bands were detected using alkaline phosphatase conjugated secondary antibodies (antirabbit, Jackson ImmunoResearch, West Grove, PA, USA; antimouse, Sigma, St. Louis, MO, USA) and fluorescent scanning (Storm System, Molecular Dynamics, Piscataway, NJ, USA).

Staining for senescence-associated β-galactosidase

The assay was done 4 days after seeding 105 cells into 35-mm 6-well dishes. The staining was performed using senescence β-galactosidase staining kit (Cell Signaling #9860) according to the manufacturer's protocol.

Acknowledgments

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

We thank Maggie Vergara for technical assistance, Adam Salmon for valuable discussions, Vince Cristofalo and Judith Campisi for helpful critique of the manuscript and David Ferguson for information about chromosomal abnormalities in the skin-derived fibroblast cell lines. This work was supported by NIH grants AG023122 and AG08808.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  • Bartke A, Chandrashekar V, Dominici F, Turyn D, Kinney B, Steger R, Kopchick JJ (2003) Insulin-like growth factor 1 (IGF-1) and aging: controversies and new insights. Biogerontology 4, 18.
  • Bartke A, Coschigano K, Kopchick J, Chandrashekar V, Mattison J, Kinney B, Hauck S (2001) Genes that prolong life: relationships of growth hormone and growth to aging and life span. J. Gerontol. A Biol. Sci. Med. Sci. 56, B340B349.
  • Brown-Borg HM, Borg KE, Meliska CJ, Bartke A (1996) Dwarf mice and the ageing process. Nature 384, 33.
  • Busuttil RA, Rubio M, DollA C, Campisi J, Vijg J (2003) Oxygen accelerates the accumulation of mutations during the senescence and immortalization of murine cells in culture. Aging Cell 2, 287294.
  • Coschigano KT, Holland AN, Riders ME, List EO, Flyvbjerg A, Kopchick JJ (2003) Deletion, but not antagonism, of the mouse growth hormone receptor results in severely decreased body weights, insulin, and insulin-like growth factor I levels and increased life span. Endocrinology 144, 37993810.
  • Flurkey K, Papaconstantinou J, Miller RA, Harrison DE (2001) Lifespan extension and delayed immune and collagen aging in mutant mice with defects in growth hormone production. Proc. Natl Acad. Sci. USA 98, 67366741.
  • Holzenberger M, Dupont J, Ducos B, Leneuve P, Geloen A, Even PC, Cervera P, Le Bouc Y (2003) IGF-1 receptor regulates lifespan and resistance to oxidative stress in mice. Nature 421, 182187.
  • Ikeno Y, Bronson RT, Hubbard GB, Lee S, Bartke A (2003) Delayed occurrence of fatal neoplastic diseases in ames dwarf mice: correlation to extended longevity. J. Gerontol. A Biol. Sci. Med. Sci. 58, 291296.
  • Jaattela M, Wissing D, Bauer PA, Li GC (1992) Major heat shock protein hsp70 protects tumor cells from tumor necrosis factor cytotoxicity. EMBO J. 11, 35073512.
  • Johnson TE, Lithgow GJ, Murakami S (1996) Hypothesis: interventions that increase the response to stress offer the potential for effective life prolongation and increased health. J. Gerontol. A Biol. Sci. Med. Sci. 51, B392B395.
  • Karatza C, Stein WD, Shall S (1984) Kinetics of in vitro ageing of mouse embryo fibroblasts. J. Cell Sci. 65, 163175.
  • Kinney BA, Coschigano KT, Kopchick JJ, Steger RW, Bartke A (2001a) Evidence that age-induced decline in memory retention is delayed in growth hormone resistant GH-R-KO (Laron) mice. Physiol. Behav. 72, 653660.
  • Kinney BA, Meliska CJ, Steger RW, Bartke A (2001b) Evidence that Ames dwarf mice age differently from their normal siblings in behavioral and learning and memory parameters. Horm. Behav. 39, 277284.
  • Lithgow GJ (2000) Stress response and aging in Caenorhabditis elegans. Results Probl. Cell Differ. 29, 131148.
  • Lithgow GJ, White TM, Hinerfeld DA, Johnson TE (1994) Thermotolerance of a long-lived mutant of Caenorhabditis elegans. J. Gerontol. 49, B270B276.
  • Miller RA (2001) Genetics of increased longevity and retarded aging in mice. In Handbook of the Biology of Aging (MasoroEJ, AustadSN, eds). San Diego, CA: Academic Press, pp. 369395.
  • Mosser DD, Caron AW, Bourget L, DenIs-Larose C, Massie B (1997) Role of the human heat shock protein hsp70 in protection against stress-induced apoptosis. Mol. Cell. Biol. 17, 53175327.
  • Murakami S, Salmon A, Miller RA (2003a) Multiplex stress resistance in cells from long-lived dwarf mice. FASEB Journal express article 10.1096/fj.02–1092fje. Published online 3 June 2003.
  • Murakami S, Salmon A, Miller RA (2003b) Multiplex stress resistance in cells from long-lived dwarf mice. FASEB J. 17, 15651566.
  • Parrinello S, Samper E, Krtolica A, Goldstein J, Melov S, Campisi J (2003) Oxygen sensitivity severely limits the replicative lifespan of murine fibroblasts. Nat. Cell Biol. 5, 741747.
  • Salmon AB, Murakami S, Bartke A, Kopchick J, Yasumura K, Miller RA (2005) Fibroblast cell lines from young adult mice of long-lived mutant strains are resistant to multiple forms of stress. Am. J. Physiol. Endocrinol. Metab. 289, E23E29.
  • Silberberg R (1972) Articular aging and osteoarthritis in dwarf mice. Pathol. Microbiol. (Basel) 38, 417430.
  • Smith JR, Pereira-Smith OM (1996) Replicative senescence: implications for in vivo aging and tumor suppression. Science 273, 6367.
  • Tatar M, Bartke A, Antebi A (2003) The endocrine regulation of aging by insulin-like signals. Science 299, 13461351.
  • Vergara M, Smith-Wheelock M, Harper JM, Sigler R, Miller RA (2004) Hormone-treated Snell dwarf mice regain fertility but remain long-lived and disease resistant. J. Gerontol. Biol. Sci. 59, 12441250.
  • Wang MC, Bohmann D, Jasper H (2003) JNK signaling confers tolerance to oxidative stress and extends lifespan in Drosophila. Dev Cell 5, 811816.
  • Wang MC, Bohmann D, Jasper H (2005) JNK extends life span and limits growth by antagonizing cellular and organism-wide responses to insulin signaling. Cell 121, 115125.