Age- and temperature-dependent somatic mutation accumulation in Drosophila melanogaster

Abstract

Using a transgenic mouse model harboring a mutation reporter gene that can be efficiently recovered from genomic DNA, we previously demonstrated that mutations accumulate in aging mice in a tissue-specific manner. Applying a recently developed, similar reporter-based assay in Drosophila melanogaster, we now show that the mutation frequency at the lacZ locus in somatic tissue of flies is about three times as high as in mouse tissues, with a much higher fraction of large genome rearrangements. Similar to mice, somatic mutations in the fly also accumulate as a function of age, but they do so much more quickly at higher temperature, a condition which in invertebrates is associated with decreased life span. Most mutations were found to accumulate in the thorax and less in abdomen, suggesting the highly oxidative flight muscles as a possible source of genotoxic stress. These results show that somatic mutation loads in short-lived flies are much more severe than in the much longer-lived mice, with the mutation rate in flies proportional to biological rather than chronological aging.

Figure 1. Comparative analysis of somatic mutations in mice and flies.

(A) Schematic depiction of the LacZ-plasmid model in Mus musculus and D. melanogaster. Note that the mouse line 60 contains about 10 head-to-tail organized plasmid copies per integration site, but only two plasmid copies are depicted. For the transgenic flies, each line harbors only 1 copy of the pUR288 plasmid. Individual plasmids can be rescued by excision of genomic DNA with Hind III (H) or Pst I (P). After purification from the mouse or fly genomic DNA, self-ligation and transformation into Escherichia coli C (ΔlacZ, galE ^-) host cells, individual plasmids are recovered in the form of ampicillin-resistant colonies. A small amount of transformants is plated on medium containing X-gal, to determine the total number of plasmids rescued (titer plate). The remainder is plated on media supplemented with the lactose-analogue p-gal, to select only the cells harboring a mutant lacZ (selective plate). The mutant frequency is the ratio of the colonies on the selective plate versus colonies on the titer plate (times the dilution factor). Location and direction of the integration site of the pP[CaSper]vector is shown for line 11. For the mouse, the location and direction of the integrated pUR288 concatamers is shown for line 60. LacZ =  lacZ reporter gene; P5′ = 5′ end of (pP[CaSper]); P3′ = 3′ end of (pP[CaSper]); white =  the white selection marker. (B) Direct comparison of spontaneous somatic mutation frequency in mice (3-month old; liver) and flies (1–2-days old; whole body). White bars represent the frequency of point mutations and black bars the frequency of DNA rearrangements. The frequency of all mutations in the fly is greater than that of mouse liver (one-tailed Welch Two Sample t-test, p = 9.04e−05). Error bars are Standard Deviations.

Figure 2. Somatic mutations as a function of life span at different temperatures.

(A) Survivalship of female flies grown at 18, 25 and 29°C (p<0.0001 for all comparisons by log-rank test). (B) The same for male flies (p<0.0001 for all comparisons by log-rank test). (C) Mutation frequency in flies maintained at different temperatures with age. Through linear regression analysis, the slopes of all mutation frequencies with chronological time are significant at the p<.01 level except for female flies at 18°C (see Figure S2). The slopes of the different temperatures are significant for females (p = 3.33e−05) and males (p = 4.89e−05). (D) Mutation accumulation at 18 and 29°C in male flies (data are independent of Figure 2C). (E) Mutation frequency of flies as a function of remaining survival time at a given temperature (data are from Figure 2A–2C). Black = 18°C, green = 25°C, red = 29°C, circle = female, triangle = male. Error bars are standard deviations.

Figure 3. LacZ mutation spectra in D. melanogaster at different temperatures.

(A) Frequencies of point mutations (white) and DNA rearrangements (black). Each determination point is based on 60 mutants, taken from the plates used to generate the data in Figure 2C. The accumulation of genome rearrangements was statistically significant (p = 0.0043 in males and p = 0.0260 in females). No significant accumulation of point mutations was detected. (B) Physical maps of the second breakpoint for the genome rearrangements in the fly at 18 and 29°C.

Figure 4. Mutation frequency in DNA from the head, thorax, and abdomen.

Of young and old, female (A) and male (B) flies maintained at 29°C. *p<0.05; **p<0.01, ***p<1e−5 (t-test of old vs. young). Mutation frequency of old male thorax was significantly different from old male abdomen (p = 9.13e−05) and old male head (p = 0.01092). Mutation frequency of old female thorax was different only from old head p = 0.03318. Error bars are Standard Deviations.