Drosophila DNA/RNA methyltransferase contributes to robust host defense in aging animals by regulating sphingolipid metabolism

Abstract

Drosophila methyltransferase (Mt2) has been implicated in the methylation of both DNA and tRNA. In this study, we demonstrate that loss of Mt2 activity leads to an age-dependent decline of immune function in the adult fly. A newly eclosed adult has mild immune defects that are exacerbated in a 15 day old Mt2^-/- fly. The age-dependent effects appear to be systemic, including disturbances in lipid metabolism, changes in cell shape of hemocytes and significant fold-changes in levels of transcripts related to host defense. Lipid imbalance, as measured by quantitative lipidomics, correlates with immune dysfunction, with high levels of immunomodulatory lipids, sphingosine-1-phosphate (S1P) and ceramides, along with low levels of storage lipids. Activity assays on fly lysates confirm the age-dependent increase in S1P and concomitant reduction of S1P lyase activity. We hypothesize that Mt2 functions to regulate genetic loci such as S1P lyase and this regulation is essential for robust host defense as the animal ages. Our study uncovers novel links between age--dependent Mt2 function, innate immune response and lipid homeostasis.

Keywords: Ceramide; Hemocyte; Innate immunity; Invertebrate; Lipid homeostasis; Mt2; Signaling; Sphingosine-1-phosphate (S1P).

Fig. 1. Life span and bacterial clearance assays suggest a decline in immune function for Mt2^-/- flies.

(A) Mock infected Mt2^-/- flies show a shorter lifespan when compared to mock infected Mt2^+/+ animals. When infected with E. coli, Mt2^-/- flies show enhanced lethality at 25 °C and 29 °C (data not shown), when compared to infected Mt2^+/+ flies. Infected Mt2^-/- flies show increased mortality as compared to mock infected Mt2^-/- flies. Experiments were in biological triplicates, with life spans curves analyzed using Log-rank (Mantel-Cox) Test, in GraphPad Prism version 5. Analysis indicates that the curves Mt2^-/- (mock) vs Mt2^-/- (infected) as well as the Mt2^-/- (infected) vs Mt2^+/+ (infected) differ significantly (p<0.0001). (B) Total bacterial colony forming unit (cfu) count at 0h and 6h post infection for 2 day old Mt2^+/+, Mt2^-/- and Mt2-TG males at 25 °C. The Mt2^-/- flies fail to clear bacterial load to the same extent as wild type, 6h post infection. The data represents three independent biological replicates. Data analyzed by 1-way ANOVA. For all figures henceforth, ‘*’, indicates p<0.01, ‘**’, p<0.05 and ‘***’, p<0.001. (C) Cfu count for ageing flies at day 1, day 15 and day 30 post eclosion. The 6h post-infection cfu was normalized to the mean 0h cfu in each case. 2-way ANOVA post arcsine transformation was used to test significance. 1 day old Mt2^-/- flies showed a mild deficiency in their ability to clear bacteria, which worsened dramatically with age. N (biological replicates) =3, n (number of flies, for each day/time-point and each genotype) =4.

Fig. 2. Hematopoiesis is disturbed in Mt2^-/- animals.

(A) The total hemocyte count for 1 day and 15 day old Mt2^+/+, Mt2^-/- and Mt2-TG flies indicates increase in plasmatocyte number with age for Mt2^-/- animals, a trend opposite to that of the controls. The data shown represents three independent biological replicates with 15 males per replicate. 1Way ANOVA followed by tukey test was performed for statistical analysis. N=3, n=15. (B) Plasmatocytes from 15 day old flies imaged using SEM at 5K magnification show that Mt2^-/- show ‘sickle-cell’ morphology, as compared to nearly round cells seen in wild type flies. The bar indicates a linear scale of 2 μM. (C) The linear dimensions of individual cells from SEM images were analyzed using ImageJ and the roundness for each cell was plotted in GraphPad Prism version 5. N=3; n=4. Student’s t-test was used for statistical analysis. (D) serpent(srp) transcript levels, as measured by real-time qPCR were reduced by half in Mt2^-/- animals, without infection. 1WAY ANOVA followed by Tukey’s test were performed as a test of significance. N=3, n=5. (E) Real time qPCR for eater and u-shaped was carried out for 15 day old flies’ pre and post infection. The data is a mean of three independent biological replicates (N=3), with 5 animals per experiment (n=5). Interestingly, the production of AMPs and cellular immunity players appear to be lowered with age in Mt2^-/- flies in comparison with Mt2^+/+.

Fig. 3. Transcriptional response by AMP genes to infection is weaker in 15-day old Mt2^-/- flies.

Real Time qPCR was used to measure levels of dipt, att-D and drs in response to infection for 3 day and 15 day old adult Mt2^+/+, Mt2^-/-and MT2-TG flies. Flies were infected and transcript levels measured at 0 and 6h post infection. Transcripts were normalized to rp49 and relative fold values (6 hrs / 0 hrs) were plotted. 1way ANOVA followed by Tukey’s test was performed as a test of significance. N=3, n=5. (A) Three day old flies show strong activation of all three AMPs. Activation of AMPs in Mt2^-/- animals is stronger, with dipt and drs levels being statistically significant. (B) Fifteen day flies show significantly lower levels of activation for all three AMPs, post infection.

Fig. 4. Sphingosine-1-Phosphate, ceramide levels increase while triacylglycerol levels fall with age in Mt2^-/- flies.

(A) Age dependent drop in TAGs as measured by decrease in band intensities, separated by thin layer chromatography. Chi square test for trend was used for analysis. N=1, n=15. (B) Heat map plot summarizing the relative fold changes for different lipid classes compared between Mt2^-/- and Mt2-TG, when normalized to Mt2^+/+, for 15 day old adult male flies. Red and blue color indicates increase and decrease of respective lipid class relative to Mt2^+/+ flies. S1P and ceramide levels are 4 fold and 2 fold higher, respectively in Mt2^-/- flies, while sphingomyelin, sphingosine, free fatty acids and overall phospholipid levels do not change significantly. TAGs, PE and PA show ~2 fold decreased in levels in Mt2^-/- flies. Data for individual lipid moieties can be found in Fig. S1B and Table S1. N=5, n=2. (C) Lipidomics measures age dependent S1P levels in adult flies from day 1 to day 15. When compared to Mt2^+/+ or Mt2-TG, S1P levels in Mt2^-/- flies start accumulating 5 days post eclosion. By day 10, the levels are approximately 5-fold higher than that of controls. N=4, n=2. (D) Lipase assay shows change in Sply activity in Mt2^-/- adult whole body extracts from day 1 to day 15. Sply activity does not increase in Mt2^-/- flies with age as compared to controls. N=3, n=5.

Fig. 5. Mt2 has a systemic role in providing a robust, age dependent immune function in flies.

(A) Sphingolipid metabolism in Drosophila. Metabolites are in black font while enzymes that are implicated in their conversion are in blue. Lipidomics data suggest age dependent changes in Mt2^-/- flies, with increase in S1P levels, a result in agreement with decrease in activity of Sply. Decrease in Sply activity may also explain the reduction in levels of PE. TAG levels also fall with age. (B) Model for a role for Mt2 in immunity and aging. Mt2 appears to function by regulating both the cellular and humoral arms of the innate immune response in adult flies, with lipid metabolism being a critical component for a robust response. At the molecular level, this effect would be via methylation of DNA which will regulate transcription or via methylation of tRNA, which would regulate tRNA stability and thus affect total protein activity. The model incorporates data from this study (arrows) as well as interactions found in literature (dotted arrows). The number on the dotted line indicates the source of the data. 1(Oskouian et al., 2005), 2 (Adada et al., 2015; Kraft, 2016), 3 (Hinkovska-Galcheva et al., 2003; Tafesse et al., 2015), 4 (Ramet et al., 2002; Shlyakhover et al., 2018).