Overexpression of CBS and CSE genes affects lifespan, stress resistance and locomotor activity in Drosophila melanogaster

Overexpression of transgenes

Using FlyBase search [17] we identified 3 genes responsible for H₂S production in Drosophila, CG12264 coding for mitochondria-located enzyme with CSE activity [18], CG5345 that codes the cytoplasm-located orthologue of CSE [11] and CG1753 that is the single homolog of CBS [4]. Using cDNA of these genes 3 UAS lines were generated (Table S1).

In their previous report, Kabil [4] showed that ubiquitous (driven by tubulin-GAL4) or neuron-specific (driven by Elav-GAL4) transgenic overexpression of CBS enhanced Drosophila longevity, while CBS overexpression in abdominal fat body (driven by GeneSwitch S1-106) and in gut (driven by GeneSwitch TIGS-2) had no effect on lifespan [4]. To further study the role of genes affecting H₂S production in longevity of Drosophila, we tested whether ubiquitous overexpression of CBS and CSE under the control of da-GAL4 driver also increase resistance to stress factors, prolong lifespan and improve locomotor performance.

The expression levels of transgenes were examined by qRT-PCR assays before further analysis (Figure S1). We found 2.4-, 6.1- and 1.9-fold induction (p<0.05) of CSE[LD22661], CSE[LD22255], CBS[LD21426], respectively compared with their isogenic UAS controls (Figure S1, Table S3). The level of transgene expression was found to depend on the UAS transgene.

Lifespan

The activation of constitutive CSE[LD22661] overexpression leaded to prolongation by 15.6% relative to da-GAL4 parental control line (p<0.05). At the same time, we observed the decline of median and maximum lifespan in males with CSE[LD22661] overexpression compared with maternal control line UAS-CSE[LD22661] by 24.4% (p<0.0001) and 3.1% (p<0.01), respectively (Table 1, Figure 1A). The increase in median lifespan by 11.6% (p<0.05) was caused by upregulation of CSE[LD22255] relative to da-GAL4 parental control line, but the lifespan differences in comparison with the UAS-CSE[LD22255] maternal control line were not statistically significant (Table 1, Figure 1B). Thus, overexpression of both forms of the CSE gene did not lead to an increase in lifespan relative to the both paternal and maternal control lines. As we know the effect of CSE overexpression on the longevity was not previously studied. As it was established by Kabil [4] pharmacological inhibition of the CSE using propargylglycine reversed lifespan extension induced by DR, and there was no effect of CSE inhibition on fully fed flies [4]. Overexpression of CSE in Drosophila SCA3-model restored protein persulfidation, decreased oxidative stress, dampened the immune response and improved SCA3-associated tissue damage and neurodegeneration [19]. Despite these published beneficial effects on aging-related processes, we did not observe the positive effect of CSE overexpression on the lifespan.

The CBS[LD21426] overexpression increased median (by 15.1-28.3%, p<0.05) and maximum (by 6.9-12.5%, p<0.05) lifespan relative to UAS-CBS[LD21426] maternal and da-GAL4 parental control lines, respectively (Table 1, Figure 1C). In addition, CBS[LD21426] overexpression decreased MRDT (by 17% and 21.4%, respectively) suggesting that geroprotective effect of CBS in associated with prolongation of healthspan. The observed lifespan extending effect of CBS ubiquitous overexpression is consistent with the results of Kabil [4].

Stress resistance

We also analyzed the resistance to different kinds of stress factors of flies with overexpression of the CBS and CSE genes and the control lines. The survival curves and times of 25%, 50%, 75% and 90% mortality in flies under conditions of paraquat (oxidative stress), hyperthermia, and combination of arid and food-free conditions were estimated (Table S4, Figure 2). The CSE[LD22661] overexpression leaded to marginal decrease (p<0.05) of resistance to paraquat (Figure 2A), while CSE[LD22255] and CBS[LD21426] overexpression had no effect (p>0.05) on resistance to oxidative stress (Figures 2B and C). According to published data exogenous and endogenous H₂S has been shown to improve the capacity of model animals to survive in conditions of oxidative stress. Pharmacological H₂S donor (GYY4137) protected wild-type C. elegans against paraquat poisoning [20]. In D. melanogaster the X-ray radiation-induced embryonic mortality were found to be reduced by H₂S treatment [21]. Our data demonstrated that overexpression of the CBS and CSE is not sufficient to increase tolerance to oxidative stress in Drosophila.

We also found that CSE[LD22661], CSE[LD22255] and CBS[LD21426] upregulation provoked significant elevation of resistance to hyperthermia (Figures 2D-F). The obtained results are consistent with the published effects of exogenous H₂S. Exposure of C. elegans increased concentrations of H₂S has been shown to enhance thermotolerance [15].

The significant decrease of resistance to combination of arid and food-free conditions was observed (p<0.001, Figures 2G-I). As known from the published data exposure to H₂S emanated from a K₂S donor significantly increased survival of flies under arid and food-free conditions, but not under humid and food-free conditions, demonstrating that H₂S plays a role in desiccation tolerance but not in nutritional stress alleviation [22]. On the other hand, overexpression of the CBS, was sufficient to promote fat deposition in flies [4]. It is well known that Drosophila are better able to resist starvation by sequestering more body lipid reserves [23]. Thus the final role of endogenous H₂S in the formation of resistance to starvation and desiccation has to be clarified.

Locomotor activity

The influence of CBS and CSE overexpression on locomotor activity was estimated in flies at the age of 5, 15 and 25 days (Figure 3). It was found that UAS-CSE[LD22661] overexpression induced decline of spontaneous locomotor activity in the first part of life and increase of this parameter in the second part of male’s life (Figure 3A). In UAS-CSE[LD22255] overexpressing males locomotor activity is higher than in control group in the first and second third of life but it is lower in the last one. (Figure 3B). The UAS-CBS[LD21426] overexpressing males were characterized by high longlife locomotor activity (Figure 3C).

Drosophila melanogaster lines

Full length sequenced cDNA Clones from Drosophila Genomics Resource Center (DGRC, USA) were used for UAS lines creation. Cloning of cDNAs into pUAST-attB vector, sequence verification of each clone, and φC31 integrase-mediated site-specific germline transformation via embryo injection were ordered in GenetiVision (Houston, Texas, USA) (Table S1). UAS-CSE[LD22661] contains an additional copy of CG12264 gene coding for mitochondria-located enzyme with CSE activity [18]. UAS-CSE[LD22255] carries an additional copy of CG5345 gene that is the Drosophila orthologue of CSE (cytoplasm-located) [11]. UAS-CBS[LD21426] contains an additional copy of CG1753 gene that is the single homolog of CBS in Drosophila [4].

The GAL4 line (GAL4 driver) was used for overexpression of UAS transgenes. da-GAL4 expresses GAL4 ubiquitously and strongly under the control of daughterless [28]. This driver expresses throughout development and in most adult tissues [29]. da-GAL4 line was kindly provided by Dr. Laurent Seroude (Queen’s University, Kingston, Canada).

In order to match the genetic background of UAS and GAL4 lines they were backcrossed into white-eyed standard w¹¹¹⁸ (#3605, Bloomington Drosophila Stock Center, USA) for 6-8 times.

Activation of transgenes overexpression

The GAL4-UAS binary transgenic system was used to activate the expression of the studied genes [30]. We used constitutively active (da-GAL4) driver of GAL4 that activates the gene overexpression ubiquitously. The flies with overexpressed transgenes were obtained by mating the females from UAS lines and males from GAL4 driver lines.

Quantitative real time PCR

To confirm overexpression of studied genes in the whole body 10-15 imagoes were used in every variant of the experiment. Gene expression levels were analyzed in flies at the age of 5-6 days after imago hatching. Experiments were performed in 3-4 replicates. Whole flies or heads were homogenized with the Silent Crusher-S homogenizer (Heidolph, Germany) in TRIzol Reagent (Invitrogen, USA). RNA was separated using BCP (Invitrogen, USA), in accordance with the manufacturer’s protocol. To test that RNA samples were DNA-free, control PCR experiments without the reverse transcription step were performed with primers for the β-Tubulin gene. Reverse transcription was performed using an Oligo(dT)₂₀ primer (Invitrogen, USA) and SuperScript III Reverse Transcriptase (Invitrogen, USA), according to manufacturer’s instructions.

Quantitative real-time PCR (qRT-PCR) assays were performed using SYBRGreen PCR Master Mix (Applied Biosystems, USA). The list of primers is presented in Table S2. All reactions were performed using a CFX96 real-time PCR detection system (Bio-Rad Laboratories, USA). The thermal cycle conditions were: initial denaturation step at 95°С for 10 min, followed by 50 cycles of 95°С for 15 s (denaturation), 60°С for 30 s (annealing) and 60°С for 30 s (elongation). Expression levels were normalized against the housekeeping gene β-Tubulin. All target genes and β-Tubulin were amplified in separate PCR tubes. Four measurements were performed for each version of the experiment.

Lifespan analysis

Control and experimental fly males were collected during 24 h after imago hatching and maintained in constant conditions on a sugar-yeast medium at 25°C and 60% humidity in a 12:12 h light-dark cycle. The flies were housed in Drosophila vials at a density of 30 individuals per vial, with 5 vials per experimental variant. Dead flies were recorded daily. Fresh medium vials were provided two times per week. Experiments were performed in 2 replicates. The median lifespan, the age of 90% mortality (maximum lifespan), and the mortality rate doubling time (MRDT) were calculated.

Analysis of locomotor activity

Age-dependent changes in spontaneous locomotor activity were measured using the Drosophila Population Monitor (TriKinetics Inc., USA). Activity was defined as total number of sensor crosses during 3 min by 30 flies. Measurements were carried out every 10 days, starting from the age of 5 days, until 30 flies were alive in control and experimental variants (at the age of 5, 15 and 25 days).

Stress resistance analysis

The resistance to acute stress conditions was evaluated in 5 days old imagoes. The exposure to paraquat (oxidative stress), hyperthermia, and combination of arid and food-free conditions were used. Before as exposed to paraquat flies were deprived of food and water for 3 h and then transferred into vials containing 3-layer filter paper saturated with 300 ml of a 5% sucrose solution with 20 mM paraquat (Sigma-Aldrich, USA). At the variant of hyperthermia, flies were kept in vials on agar-yeast medium at 35 °C. For combination of arid and food-free conditions flies were placed into empty vials. Flies were transferred into new vials every two days, and kept under stressful conditions until the death.

For evaluating the resistances to oxidative stress and heat shock the amount of dead flies were counted two times per day. For evaluating the resistances to combination of arid and food-free conditions dead flies were counted every 3 hours. The survival time for 25%, 50%, 75% and 90% of populations were estimated.

Statistical analysis

Relative levels of expression were calculated using 2^-ΔΔСt method [31]. ΔΔCt was calculated according to equation ΔΔCt = ΔCt (experiment) – ΔCt (control), where ΔCt = Ct (target gene) – Ct (β-Tubulin). Statistical significance of expression differences was estimated using Mann-Whitney U-test [32]. To compare the statistical differences in survival functions and median lifespan between control and experimental groups, the modified Kolmogorov-Smirnov and log-rank [33] tests were used, respectively [34,35]. A Wang-Allison test was used to estimate the differences in the age of 90% mortality [36]. The χ² test was used for locomotor activity and fecundity analysis. To assess the statistical significance of differences in resistance to stress factors, the Fisher's exact test was used [37]. Statistical analyses of the data were performed using R, version 2.15.1 (The R Foundation) and OASIS 2: Online Application for Survival Analysis 2 [37].