Heart-specific Rpd3 downregulation enhances cardiac function and longevity

Rpd3 downregulation enhances resistance to oxidative stress

To investigate whether Rpd3 is related to the mechanism of stress resistance, a long-lived rpd3 heterozygous (P{PZ}rpd3[04556]/+) mutant fly [3] was incubated with paraquat-induced oxidative stress. Although rpd3−/− homozygous mutant flies are lethal, we found that the rpd3−/+ heterozygous flies (46% of rpd3 expression in Fig. 1A) can increase survivorship up to 31% under the oxidative stress compared to wild type (+/+) flies (Fig. 1B-C). This indicates that the Rpd3 downregulation enhances stress resistance as shown in other long-lived flies such as mth or loco heterozygous mutant flies [6, 17]. The lifespan of this Rpd3 downregulation was extended consistent with the previous report [3] although the content of extension (13% in SFig. 1A-B) was less than the previous observation (33% in [3]), which might be explained due to the different genetic background [18].

In order to validate our findings, we utilized the UAS/Gal4 system [19] which provides an alternate approach to reinvestigate if the Rpd3 downregulation enhances stress resistance. The rpd3 expression was reduced via a UAS-rpd3-dsRNAi transgene (hereafter rpd3Ri in Fig. 1D) under the arm- or act-Gal4 driver, which respectively expresses a target UAS-gene mildly or highly in a whole-body manner. The highly downregulated rpd3 expression in rpd3Ri/actG4 flies induced lethality as shown in the rpd3−/− homozygous mutant flies. However, the rpd3Ri/armG4 flies exhibited 60% of rpd3 expression compared to the single transgene control (+/armG4 in Fig. 1D) and increased survivorship up to 22% under the oxidative stress (Fig. 1E-F). This mild rpd3 downregulation of rpd3Ri/armG4 flies also extended lifespan (16% in SFig. 1C-D), implying that the mild decrease of rpd3 expression in the whole body is favorable for enhancement of both stress resistance and lifespan.

Heart-specific downregulation of Rpd3 increases stress resistance

Wessells et al. (2004) [10] reported that age-related change of heart function is minimized in long-lived flies when systemic levels of Insulin-IGF receptor (InR) signaling are reduced. Moreover, interfering with InR signaling exclusively in the heart prevents the decline in cardiac performance with age. According to those observations, we investigated if heart-specific Rpd3 downregulation affects heart function, stress resistance, and lifespan. Using a heart-specific driver (tinman-Gal4) [20], we examined the effect of Rpd3 downregulation (rpd3Ri) under three different stressors (oxidation, starvation and heat in Fig. 2). It was confirmed that the tinman-Gal4 driver (tinG4) displays the GFP signal in a heart-specific pattern (long heart tube from posterior abdomen in SFig. 2A) and expresses a target UAS-GFP gene that can be detectable in total RNAs purified from a whole-body (SFig. 2B). Although the change of rpd3 expression could not be detected with the heart-specific Rpd3 downregulation (rpd3Ri/tinG4) from the total RNAs of whole-body (data not shown), the rpd3Ri/tinG4 male flies interestingly revealed a 35% increase of the median survival time (hour) in resistance against the oxidative stress (Fig. 2A and D). Similarly, starvation and heat resistances were also enhanced by 17% and 31%, respectively, on average median over single transgene controls (Fig. 2). All three assays were statistically significant (P < 0.01), which exhibit that the heart-specific Rpd3 downregulation improves resistance to all of the stress tests: oxidation, starvation and heat. These enhancements of stress resistance could be generally exhibited in female flies as well (SFig. 3). We found that the heart-specific Rpd3 downregulation in females also improves stress resistance significantly in response to oxidation and starvation stressors, accumulating survival times approximately 47% and 60% longer than the single transgene control, respectively (SFig. 3). Surprisingly, these increased yields in females were higher than those of male flies (Fig. 2), indicating that the stress resistance is more effectively enhanced in females by heart-specific Rpd3 downregulation. However, in contrast to males, it appeared that female tolerance to heat did not increase significantly (data not shown), which may be interpreted as a gender difference. All together these findings indicate that the tissue-specific regulation of Rpd3 can systemically modulate the fly's response to the environment.

Stress resistance is enhanced exclusively by heart-specific Rpd3 downregulation

The Rpd3 gene is ubiquitously expressed in all fly tissues (http://flybase.org), implying that the Rpd3 may perform an important role everywhere as histone deacetylase. Accordingly, we investigated if Rpd3 down-regulation in other non-cardiac tissues also enhances stress resistance systemically as it did in the heart tissue (Fig. 3A). The fat body, the fruit fly analogue of mammalian liver and adipose tissues, is known to function in metabolic homeostasis, stress tolerance, growth and longevity in Drosophila [21, 22]. The fat body-specific modulation of a specific gene, such as overexpressed dFOXO, increases stress resistance and lifespan in flies [23, 24]. It is also reported that activation of JNK (Jun-N-terminal Kinase)-NLaz (Neural Lazarillo) signaling in the fat body promotes stress tolerance and extends lifespan as well [22]. The Rpd3 downregulation in fat body (rpd3Ri/r4G4), however, caused the semi-lethality in male flies (data not shown) and did not enhance stress resistance in females (Fig. 3A). Similarly, eye-specific Rpd3 downregulation (rpd3Ri/GMRG4) also could not affect the stress resistance (Fig. 3A) without any defect of eye morphology (data not shown). Although limited tests were performed, only heart-specific downregulation of the rpd3 gene enhanced the stress resistance (Fig. 3A), implying that the Rpd3 protein functions in the heart for the regulation of stress resistance.

To exclude a possibility of non-specific RNA interference from 482bp target sequences of rpd3Ri transgene, the specific 21bp of another region in the rpd3 gene was tested as a new heart-specific dsRNAi (rpd3RiS/tinG4 in Fig. 3B). Although the 23% increase in stress resistance of the rpd3RiS/tinG4 flies (Figs. 3B-C) was less than the 35% increase of the rpd3Ri/tinG4 flies (Fig. 3A), it confirmed that dsRNAi of any region of the rpd3 gene in the heart enhances the stress resistance systemically (Fig. 3). We also tested whether other long-lived flies can enhance heart-specific stress resistance (Fig. 3C). It has previously been found that the Loco downregulation throughout the entire body increases not only the resistance to three stressors, including oxidation (26-63% increase), but also extends lifespan (17-32% extension) [17]. However, Loco downregulation limited exclusively to the heart (locoRi/tinG4) had no effect on oxidative stress resistance (−0.5% in Fig. 3C), suggesting that the rpd3 gene is specific for heart-specific stress resistance. Sir2, another histone deacetylase, was also investigated in a heart-specific manner. It was reported that the Sir2 upregulation extends lifespan in flies, but is dependent upon the Sir2 dosage [25]. Stress resistance, however, was not enhanced by either mild or high Sir2 upregulation in the heart (Sir2/tinG4 in Fig. 3C), implying that the Rpd3's deacetylase activity is directly related to the stress resistance mechanism within the heart tissue. Although the fat body-specific dFOXO upregulation increases stress resistance and lifespan in flies [23, 24], a heart-specific dFOXO upregulation could not be tested in stress resistance due to semi-lethality in our genetic background (data not shown), further implicating that the heart-specific Rpd3 exclusively contributes to the enhanced stress resistance.

Heart-specific Rpd3 downregulation systemically increases expressions of anti-aging genes

To characterize how heart-specific Rpd3 down-regulation systemically enhances stress resistance, expression of representative anti-aging genes was examined using total RNAs purified from whole flies. Although sod1 (Cu/ZnSOD) expression did not change with heart-specific Rpd3 downregulation, sod2 (MnSOD) expression increased by 75% in comparison with control rpd3Ri/+ flies (Fig. 4A), which may explain how heart-specific Rpd3 downregulation more efficiently removes the superoxide produced from paraquat in mitochondria. This increase of SOD expression was observed in several long-lived flies such as InR or Loco downregulated flies [17, 26]. It is reported that the long-lived rpd3−/+ mutants show a 134% increase in sir2 expression and are related to caloric restriction [3]. Interestingly, heart-specific Rpd3 downregulation still resulted in increased expression of sir2 gene by 93% (Fig. 4A). In Drosophila, moderate increase of Sir2 (2-5 folds) extends lifespan, although higher levels of Sir2 decrease the lifespan with a cellular toxicity [25]. When expression of foxo, another anti-aging gene, was measured, the heart-specific Rpd3 downregulation also increased the foxo expression by a significant 63% (Fig. 4A). It was reported that the dFOXO upregulation in brain fat body induces oxidative stress resistance and extends lifespan in flies, and is related to insulin signaling [23, 24]. These increased expressions of anti-aging genes did not result from Rpd3 downregulation in other tissues such as eye tissue (rpd3Ri/GMRG4 in Fig. 4A), indicating that only heart-specific Rpd3 downregulation systemically induces a significant increase of the anti-aging genes.

Furthermore, with heart-specific Rpd3 downregulation, the transcriptional level of representative stress-related genes was measured, which were selected from previously published stress response genes (http://flybase.org) [27, 28]. Expressional changes of Hexokinase A (HexA) and takeout (to) genes were not detected (Fig. 4B) although those proteins are related to starvation-tolerance mechanisms. However, the Thor gene (d4E-BP: eIF4E-binding protein) that mediates starvation and oxidative stress resistances was upregulated up to 40%, significantly (p<0.002 in Fig. 4B) with the heart-specific Rpd3 downregulation. This may partially explain how heart-specific Rpd3 downregulation can enhance resistance against starvation and oxidative stressors (Fig. 2 and SFig. 3). During oxidative stress and aging, antimicrobial peptide (AMP) genes were induced [28]. One of AMP genes, DptB, was 99% upregulated with the heart-specific Rpd3 downregulation (Fig. 4B). Heat shock protein (Hsp) and Turandot (Tot) genes are transcriptionally upregulated during heat stress and overexpression of TotA increases stress resistance in flies [28, 29]. Interestingly, heart-specific Rpd3 downregulation exhibited 343% expression of TotM compared to the control (Fig. 4B). However, heart-specific Rpd3 downregulation did not change gene expression of JNK and p53 in stress responsive signaling pathway (Fig. 4B) although their protein activities have not yet been tested.

Heart-specific Rpd3 downregulation enhances resistance against oxidative stress with higher cardiac function

Having observed increased stress resistance from heart-specific Rpd3 downregulation, we performed further experiments to determine whether there was an observable physiological change in cardiac function.

The Drosophila heart exhibits physiological changes, such as lower heart rate, as it ages [10]. Interestingly, long-lived flies reducing an InR signaling have been shown to eliminate these manifestations of the aging process, prolonging younger cardiac function into old age as keeping higher heart rate [10]. We measured the heart rate of flies concurrent with paraquat-induced oxidative stress in the different time points: 0, 18, 24, and 42 hours (Figs. 5A-B). Our findings indicated that heart-specific Rpd3 downregulation (Rpd3Ri/tinG4) displays a gradual enhancement of oxidative stress resistance as a stressed period increases (Figs. 5A and C). The Rpd3Ri/tinG4, which yielded a 91% survival rate at 24 hours, was at least 20% higher than the survival rates for both single transgene controls: rpd3Ri/+ (71%) and +/tinG4 (47%) (Fig. 5A). Furthermore, the difference between the survival rates continued to steadily increase thereafter, up to 33% at 42 hours (Figs. 5A and C). Interestingly, a notable enhancement in cardiac function was detected even at zero hours, showing that the genotype with Rpd3 downregulation displays a 16% initial increase of heart rate over the rpd3Ri/+ single transgene control (327 vs. 283 HBM) with persistent augmentation throughout the period of oxidative stress (Figs. 5B-C). Consequently, after 42 hours of oxidative stress the heart rate in flies with Rpd3 downregulation was still 12% higher than the control, possibly correlating improved cardiac function to survival against oxidative stress (Fig. 5C).

In contrast, Rpd3 downregulation within other non-cardiac tissues had no observable impact in either stress resistance or heart function (Figs. 3A and 5C). For example, Rpd3 downregulation in the eye, Rpd3Ri/GMRG4, did not display an increase of survival in the oxidative stress over the Rpd3Ri/ + control during whole time period stressed (Fig. 5C). After 24 hours, the controlled +/GMRG4 and Rpd3Ri/+ cohorts had survival rates of 19% and 65% respectively, yet only 58% of the rpd3Ri/GMRG4 had survived (SFig. 4A). Corroborating this assertion, heart rate was also reflective of being in between both controls; at 24 hours there was a 12% increase in heart rate over the +/GMRG4 genotype, but a 4% decrease over the Rpd3Ri/+ genotype (SFig. 4B). Finally, we compared the differences between heart-specific and eye-specific Rpd3 downregulation through the common control genotype, Rpd3Ri/+ (Fig. 5C). During the oxidative stress, the comparison showed that heart-specific Rpd3 downregulation provides an average enhancement of 13% in cardiac activity, with a gradual increase in oxidative stress resistance up to a maximum 33%. In contrast, eye-specific Rpd3 downregulation averaged only −0.1% and −1.3% changes, respectively (Fig. 5C).

An assessment of Rpd3 downregulation in non-cardiac regions (Figs. 3A and 5C) suggests that reduced expression of the rpd3 gene translates to enhanced stress survival and cardiac function only when it is specifically targeted at the heart. To validate this assertion, the heart failure and recovery assays were performed via the heart pacing protocol as described by Wessells et al. (2004) [10] with applying a sustained 80V electric shock during 30 seconds. Those cardiac functions were measured after a 24 hour treatment of oxidative stress in order to characterize how the cardiac function and stress resistance are related (Figs. 5D-F). At 24 hours point, a survival rate of heart-specific Rpd3 downregulation (Rpd3Ri/tinG4) revealed 19% increase over the control (Rpd3Ri/+ in Fig. 5D). In contrast, eye-specific Rpd3 downregulation (Rpd3Ri/GMRG4) had no effect from the controls (Fig. 5D). It was found that the heart-specific Rpd3 downregulation showing higher stress survivalship yields 14% lower heart failure rate at 4 minutes after electric shock, compared to the control (Rpd3Ri/+ in Fig. 5E). At the 0 and 2 minute points, the heart failure rates were also lower than the control (14% and 18%, respectively). The eye-specific Rpd3 downregulation (Rpd3Ri/GMRG4), however, exhibited only −0.6% change in heart failure (Fig. 5E). A heart recovery rate from heart failure at 0 minute until 4 minutes after electric shock was increased by 18% with the heart-specific Rpd3 downregulation (Fig. 5F). Even in other time periods (0 - 2 min and 2 - 4 min), the higher rates of heart recovery were still observed (14% and 13%, respectively). In a trial that the oxidative stress resistance was not enhanced, interestingly, improved cardiac functions also was not detected (data not shown). All together, these results suggest a positive and exclusive correlation between heart-specific rpd3 expression and cardiac function/stress tolerance.

Heart-specific Rpd3WT upregulation decreases both stress resistance and cardiac function, but the no-phosphorylated mutant Rpd3ADA restores them

Our findings, which the heart-specific Rpd3 downregulation enhances stress tolerance and cardiac function, let us investigate whether rpd3 upregulation in the heart causes the opposite effect. Using a UAS-rpd3WT transgene (Fig. 6A) the stress tolerance was measured under the heat stressor. The heart-specific Rpd3WT upregulation (rpd3WT/tinG4), unsurprisingly, revealed a 15% decrease of the median survival time (hour) in resistance against the heat (Figs. 6A and C) when compared to the control (+/tinG4). Similarly, the heart rate in cardiac function was also reduced by an average of 5% in the different time points of a heat-stressed period (Figs. 6B-C). All assays were statistically significant (P < 0.007), which indicate that the heart-specific Rpd3 upregulation decreases both stress resistance and cardiac function.

Having established that differential levels of rpd3 expression in the heart modulate stress resistance and cardiac function, we further investigated whether the enzymatic activity of Rpd3 is the responsible mechanism for such enhancement. It was reported that the phosphorylation of histone deacetylase 1 (HDAC1), the mammalian homologue of the Rpd3 gene, promotes enzymatic activity and complex formation in mammals [14-16]. The phosphorylation at Rpd3 S419/S421 sites is evolutionarily conserved and known to promote mammalian HDAC1 activity. Through site-directed mutagenesis, a non-phosphorylated rpd3ADA clone (S419A/S421A) was produced from the rpd3WT construct and the stability of Rpd3ADA protein was similar to the Rpd3WT levels (data not shown).

Interestingly, we found that the impaired stress resistance induced by heart-specific Rpd3WT upregulation disappears when the rpd3ADA clone is upregulated instead of rpd3WT (Fig. 6A), indicating that phosphorylated Rpd3 in the heart functions as a primary regulator of stress resistance. This heart-specific Rpd3ADA upregulation (rpd3ADA/tinG4) also restored the reduced heart rate caused by heart-specific Rpd3WT upregulation (Fig. 6B). Surprisingly, stress resistance and heart rate in the rpd3ADA/tinG4 genotype were higher than those of the common control +/tinG4 (Fig. 6C), suggesting a possibility that the Rpd3ADA protein competes with endogenous Rpd3WT in the heart and therefore interferes with the function of Rpd3WT in the regulation of stress resistance and cardiac activity.

Lifespan is extended when Rpd3 is downregulated in a heart continuously during aging

Downregulating the rpd3 gene in the whole body enhanced both stress resistance and lifespan in the rpd3−/+ and rpd3Ri/armG4 flies (Fig. 1 and SFig. 1) [3]. However, the flies that downregulated rpd3 expression in the heart (rpd3Ri/tinG4) displayed no significant improvement of lifespan (Fig. 7E) in spite of showing higher stress resistance (Fig. 2). Additionally, the improvement of cardiac function in young age diminished progressively throughout the lifespan (SFig. 6A and Fig. 7D). Further experiments revealed that there was insufficient Gal4 expression from the tinman-Gal4 transgene in older aged flies to downregulate rpd3 expression in the heart at effective levels (Fig. 7A). Therefore, a recombinant strain of tinG4 and UASG4 was introduced, in which Gal4 expression can be auto-amplified in the heart. At the age of 2-days, the tinG4,UASG4 flies increased the Gal4 expression up to 46-fold compared to the flies with only the tinG4 transgene (Fig. 7A). This tinG4,UASG4 driver was confirmed to display stronger UAS-GFP signal in the heart than the tinG4 driver (SFig. 5). Oxidative stress resistance of tinG4,UASG4 flies was still 36% higher from the common rpd3Ri/+ control (Fig. 7B) and interestingly, the expression of multiple anti-aging genes were more increased with further downregulation of the heart-specific rpd3 gene (Fig. 7C). During aging, the tinG4,UASG4 flies also showed a gradual decline in Gal4 expression (Fig. 7A). However, the Gal4 expression of tinG4,UASG4 flies at 7-week old was still higher than that of tinG4 flies at 2-day old by 3.1 fold (Fig. 7A). In addition, the cardiac function improved in young age of the tinG4,UASG4 strain was constantly sustained throughout the lifespan (SFig. 6B and Fig. 7D). Continual heart-specific downregulation of Rpd3 during aging, finally, extended the lifespan up to 24% (Fig. 7E), revealing that the Rpd3 in heart influences a variety of anti-aging mechanism, in turn increasing stress resistance and augmenting lifespan (Fig. 7).

Fly genotypes and cross

The heart-specific tinC∆4-Gal4 (tinG4) flies were kindly provided by Rolf Bodmer [20]. The rpd3−/+ (P{PZ}rpd3[04556]), UAS-rpd3-dsRNAi, UAS-loco-dsRNAi, UAS-sir2, and Gal4 driver flies were obtained from Bloomington Drosophila stock center except for the UAS-loco-dsRNAi flies (V110275, Vienna Drosophila RNAi Center) [17]. The lab stock strain y¹ w¹ (Bloomington) were used as wild-type control and the flies obtained outside were six times isogenized with y¹ w¹ before the stress resistance and aging tests. Virgin flies were collected from a bottle in which larval density was controlled in standard cornmeal medium, and were used for all fly experiments including stress response and aging studies.

Constructs and RT-PCR

To construct rpd3WT and rpd3ADA transgenes (Fig. 6), the rpd3 wild-type cDNA (1,563 bp) was cloned into the pCRII-TOPO vector (Invitrogen) and, from the rpd3WT plasmid the rpd3ADA clone (S419A/S421A) was produced using QuikChange Site-Directed Mutagenesis Kit (Agilent Technologies). After sequencing was completed, the plasmids were subcloned into a XhoI/XbaI digested pUASTattB vector.

To check transcriptional expression of the genes, 5 μg of the total RNA purified from adult flies (TRIzol, Invitrogen) were treated with DNase I (RNase-free, Roche) and used to produce oligo dT-primed cDNAs (SuperScript II RT, Invitrogen), which were then used as templates for quantitative real-time PCR [39]. The rp49 gene was used as an internal reference for normalizing the quality of total RNAs. Real-time PCR was performed with SYBR green using ABI7300 Real-time PCR Instrument (Applied Biosystems). Expressional fold of the various genes were determined by the comparative C_T method (ABI Prism 7700 Sequence Detection System User Bulletin #2, Applied Biosystems).

Stress response and aging assays

To measure responses for each of the three stressors (starvation, oxidation and heat), 100 newborn male flies (20 flies per vial) were kept on standard cornmeal medium at 25°C for two days [6, 40]. For the starvation stress-response assay, these two-day-old flies were transferred into clean vials (2.5 × 9.3 cm) that contain two-filter circle (2.4-cm diameter, Fisher Scientific) soaked with 300 μl of distilled water. Then, the vials were maintained at 25°C under moist conditions with 100 μl of distilled water added every 12 hours. For the oxidative stress assay, the two-day-old adult flies were starved for 6 hours at 25°C as described above. Then after the 6 hours starvation period, the flies were transferred to new vials containing two filter circles wetted with 300 μl of 20 mM methyl viologen hydrate (Paraquat, Fisher Scientific) in a 5% sucrose solution and maintained at 25°C. For the heat test, the two-day-old flies were transferred into vials with standard cornmeal medium and were kept in a 37°C incubator with 30% humidity. The survival number for each of the experiment was manually counted with four hour intervals. To test for females, five-day-old adult flies were used following the same procedure for each of the assays. For the aging test, 200 virgin flies (20 flies per vial) were counted and transferred to fresh standard cornmeal vials every three to four days [6, 40].

Heart rate assay

To measure the heart rate, the flies in a FlyNap chamber were anesthetized with 40 μl of FlyNap (Carolina) for 3 minutes [41]. Then, the flies were placed on a microscopic slide with tape on it and the wings of the flies were brushed off to the side so the heart was clearly visible. Using an Olympus microscope, three videos consisting of 20 seconds for each fly were recorded and the heart rate was manually counted and averaged. For each independent experiment, four to eight flies were used. During aging, this procedure was repeated biweekly from one-week old flies until they became seven-week old flies [10].

Heart failure and recovery rate assays

To observe heart failure in the flies, a microscope slide with two electrodes on opposite ends was prepared [42]. Conductive electrode jelly was spread between the two electrodes, leaving a minuscule gap in the middle to provide an electric shock through the flies. Flies of several genotypes were anesthetized with 40 μl of FlyNap for three minutes and then placed onto the microscope slide oriented so that the head and abdomen of the fly are facing opposite electrodes and are perpendicular to the gap in the conductive jelly. The wings of the flies were pushed onto the side of the flies on the conductive jelly to get a better view of the heart. After a maximum of 12 flies were placed onto one microscope slide, the slide was placed under the OLYMPUS microscope to observe the heartbeat. Then, the flies were electrically paced with a Grass SD9 stimulator at 80V, 6Hz, 0 Delay, 30ms Duration for 30 seconds. Cardiac arrest was measured at 0, 2 and 4 minutes after the pacing stimulation. Heart failure rate was calculated, among all the flies by counting the flies in a state of cardiac arrest until 4 minutes have passed. To obtain heart recovery rate, the amount of flies that restores their heart beat at 4 minutes after being in a state of cardiac-arrested at 0 minute was measured [10].