The effects of reduced rpd3 levels on fly physiology

1Introduction

Aging is characterized with progressive decline of physiological responses. Dietary restriction (DR) without malnutrition delays age-related pathophysiology and increases mean and maximal lifespan in a number of species [1, 2]. DR affects many physiological processes, including mitochondrial function, preserved protein homeostasis, enhanced genomic stability, increased insulin sensitivity, and nutrient metabolism. DR improves organismal health, and it delays the development of Type 2 diabetes, cardiovascular disease, neurodegeneration, and cancer. Similar health benefits have been observed in humans on a strict DR regimen without malnutrition, as well as in short-term, randomized clinical trials [3]. The effects of DR depend on the timing and particular nutrients in diet [2]. Some of the possible mediators of beneficial effects of DR include the nutrient-sensing insulin/insulin-like signaling (IIS), the Tor signaling pathway, AMPK, HSF, and sirtuins [2].

Members of class I of the zinc-dependent histone deacetylase (HDAC) family are vital regulators of chromatin structure and gene expression [4–6]. Class I HDACs have a role in development, as well as in pathophysiology of human disorders. Mutations that affect the normal function of HDAC1 have been implicated in development of human cancers, Alzheimer’s, Huntington’s, and Parkinson’s diseases. Thus HDAC inhibitors have the potential to be used in therapy for cancer, neurodegenerative disorders, muscular atrophy, and heart disease [6–9]. Underexpression of Rpd3, the Drosophila HDAC1 homologue, is associated with a longer lifespan in yeast and fruit flies [10–15]. Males heterozygous for either the hypomorphic or null rpd3 mutations live 33% and 41% longer compared to controls, respectively [11]. Female flies heterozygous for the hypomorphic allele have a 52% longer lifespan, while females carrying the null mutation have a longer maximum but not median lifespan [11]. Longer lifespan was also observed in flies fed 4-phenylbutyrate (PBA), a histone deacetylase inhibitor [16]. Others and our lab have shown that longevity extension associated with reduced rpd3 overlaps with longevity effects of a dietary restricted diet [11, 13–15, 17, 18]. This hypothesis was supported by findings that DR and rpd3 under-expression are not additive in their lifespan effects [11]. They also cause a similar change in transcription levels of dSir2, takeout, 4E-BP, and members of the insulin/insulin-like signaling (IIS) pathway such as InR, chico, and dFoxo [11, 13, 15, 17]. It was also reported that levels of rpd3 mRNA are decreased and dSir2 mRNA levels are increased in starved flies, while the opposite was found in flies when re-fed; they have increased levels of rpd3 and reduced dSir2 mRNA levels [19]. Flies mutant for both rpd3 and dSir2 live shorter compared to rpd3 mutant flies, suggesting that dSir2 mediates some effects of rpd3’s effects on fly physiology and longevity [13]. However, when longevity studies were done on diets with varying caloric content, rpd3 mutant flies live longer on some DR diet levels suggesting that rpd3 and DR have distinct but interacting effects on fly longevity [14]. This suggests that the relationship between the mechanism of longevity observed in rpd3 flies and DR is more complex than previously thought.

2Materials and methods

3Results

3.1The effect of rpd3 reduction on fly spontaneous locomotor activity

To test if the mechanism of rpd3 reduction on fly longevity is similar to DR we investigated various aspects of Drosophila physiology in rpd3 mutant flies and their genetic controls. We used two different heterozygous rpd3 alleles due to embryonic lethality of homozygous rpd3 mutant flies. We used rpd3 deficient, an rpd3 null line, (rpd3^def/+) and their genetic control, F1 progeny of rpd3^def/+ littermates. The levels of rpd3 mRNA in rpd3^def/+ is 50% lower compared to controls. We also used rpd3^P-UTR/+ flies, an rpd3 hypomorph, and rpd3^P-1.8/+, which are their genetic controls and have rpd3 reduction only in the eyes [20]. The presence of one copy of a P-element in rpd3^P-UTR/+ flies results in 60% of rpd3 mRNA levels compared to heterozygous rpd3^P-1.8/+ control flies [11]. Locomotor activity reflects the functional activity of the nervous system. There is an age-related decline in the locomotor activity of flies [21, 22]. One of the hallmarks of DR is increased spontaneous locomotor activity, and it is observed in multiple species [21, 22, 25, 26]. This is thought to be due to an increased scavenging for food. In mice and flies Sirt1 and dSir2 mediate the increased locomotor activity associated DR, respectively [25–27]. To test if there is an overlap in the mechanism of DR and rpd3 mutation, we examined the spontaneous physical activity of rpd3 mutants and their controls. We used population monitors, which allow continuous monitoring of spontaneous physical activity of the population of 10 flies during longer time periods [21]. At 10 days of age, we found an increase in peak activity in rpd3^def/+ male flies (Fig. 1A, Table 1). However, at an older age the peak activity in male rpd3^def/+ flies was not different compared to controls (Fig. 1B, Table 1). Total 24 hour activity was not different at age 10 or 40 (Fig. 1C). Females rpd3^def/+ flies showed little difference compared to their controls at 10 days of age (data not shown). A similar tendency toward increased peaks of spontaneous activity but no significant effect on the peak or on total 24 hour activity was observed in rpd3^P-UTR/+ male and female flies when compared to controls at 10 days of age (Fig. 1D-F).

Fig.1

Effects of rpd3 mutations on fly spontaneous physical activity. (A-C) Average (A, B) and total (C) 24 hours spontaneous physical activity of rpd3^def/+ male flies and their genetic controls (+/+) at 10 (A) and 40 (B) days of age. (A-C) rpd3^def/+ male flies have increased peak spontaneous physical activity compared to their genetic controls but the total locomotor activity of rpd3^def/+ male flies over the course of 24 hours is simialr to their genetic controls at age 10 and 40 days. (D,E) Average spontaneous physical activity of rpd3^P-UTR/+ male (D) and female (E) flies and their genetic controls rpd3^P-1.8/+ at 10 days of age. (F) The total spontaneous physical activity of rpd^P-UTR/+male and female flies and their genetic controls rpd3^P-1.8/+ over the course of 24 hours at age 10 days. (A, B, D, E) The data are collected in 10 minutes bins and represent an average of sponteneous locomotor activity of three biological replicates of 10 female or 10 male flies (each replicate) of experimental (rpd3^def /+ (A-C); rpd3^P-UTR (D-F)) or control (+/+(A-C); rpd3^P-1.8 (D-F)) flies. An asterix on the x-axis marks the time when the light was switched off and the transition to dark cycle (6 : 00PM or 720 minutes).

Table 1

Spontaneous physical activity of rpd3^def and rpd3^P-UTR heterozygous flies and controls

Gender	Genotype	Age	Peak activity						Total activity	(SE)
			1	p	2	p	3	p
M	rpd3def/+	10	141		388		192		1347	112
M	Control	10	223	0.3	171	0.004*	90	0.18	1390	305
M	rpd3def/+	40	193	85	93	879	172
M	Control	40	154	0.28	98	0.57	76	0.58	749	156
M	rpd3P-UTR/+	10	116	328	5	775	93
M	rpd3P-1.8/+	10	137	0.6	331	0.95	6	0.08	677	79
F	rpd3P-UTR/+	10	14	199	0	351		51
F	rpd3P-1.8/+	10	49	0.08	153	0.5	0	1.0	355	59

The peaks and the total spontaneous physical activity of rpd3^def/+ and their genetic controls+/+, rpd^P-UTR/+ and their genetic controls rpd3^P-1.8/+. The peak 1 is sum of the total activity between 6AM–7AM, peak 2 between 5PM–7PM, and 3 between 5AM–6AM. The total spontenaous activity is sum of the activity over the course of 24 hours at age 10 or 40 days. The data were collected for at least three biological replicates of 10 female or 10 male flies (each replicate) of experimental or control flies. Student T test analysis was used to determine the difference in the peaks of activity. ^*= a statistical significant result.

3.2The effect of reduced rpd3 on fly fecundity

We have previously reported that rpd3 reduction does not affect egg production in virgin rpd3^P-UTR /+ compared to rpd3^P-1.8 /+ female flies aged on a regular diet [11]. DR reduces fly fecundity due to re-allocation of the resources from reproduction to maintenance and repair [22, 28]. Here we examined if feeding rpd3^P-UTR /+ flies a high (1.5 N) or a low (0.5 N) calorie diet would affect their fecundity. We counted the number of eggs laid by flies heterozygous for wild type Canton S (CS) and rpd3^P-UTR /+ or rpd3^P-1.8/+ (genetic control) daily during the first 10 days of life. CS female flies were included for an additional comparison since they share 50% of the genetic background, and to illustrate the negative effect of 0.5 N diet on female fecundity. If rpd3 reduction mimics DR completely, we would expect rpd3^P-UTR/+ females to lay fewer eggs compared to controls on a 1.5 N diet, and even fewer on a 0.5 N diet. However, there was no difference in total egg number laid by rpd3^P-UTR/CS females compared to rpd3^P-1.8 on 1.5 N diet, and only a minor tendency toward a lower number on 0.5 N diet (p = 0.053) (Fig. 2A, B, Table 2). Furthermore, no differences in fecundity between rpd3^P-UTR/CS and CS were observed. Thus rpd3 reduction does not affect female fecundity on either 0.5 N or 1.5 N diet.

Fig.2

Effect of rpd3 reduction on fly fecundity. (A) The average number of eggs per day laid by wild type Canton-S (CS), rpd3^P-UTR/CS, and genetic controls rpd3^P-1.8/CS on 0.5 N and 1.5 N diets. (B) The total number of eggs per 10 day period laid by CS, rpd3^P-UTR/CS, and rpd3^P-1.8/CS on 0.5 N and 1.5 N diets. rpd3^P-UTR/CS flies laid similar numbers of eggs on 1.5 N but lower numbers on 0.5 N compared to the controls, a difference that did not reach significance (p = 0.053). 10 vials with a single male and a single female were used per genotype and diet.

Table 2

Effect of rpd3 reduction on female fecundity

Genotype		Food		Total eggs		SE
rpd3P-UTR/+		0.5 N		62		6.2
rpd3P-1.8/+		0.5 N		86		10.9
CS		0.5 N		65		11.6
rpd3P-UTR/+		1.5 N		269		26.1
rpd3P-1.8/+		1.5 N		292		28.7
CS		1.5 N		230		20.0

The total number of eggs per 10 day period laid by CS, rpd3^P-UTR/CS, and rpd3^P-1.8/CS flies aged on 0.5 N or 1.5 N diets. 10 vials with a single male and a single female were used per genotype and diet.

3.3Reduction in rpd3 increases starvation resistance of male and female flies

Increased starvation resistance is a phenotype found in many long-lived animals, yet its modification by DR remains a debatable topic. The first Drosophila DR experiment found that both genders of flies had increased starvation resistance on low-calorie food, yet the increase was not as great in males [28]. Reducing the calorie content in food, especially protein, increases starvation resistance in flies due to lipid accumulation [28, 29, 30]. Nevertheless, extended DR reduces starvation resistance in flies [28]. In addition, some studies indicate that genetic manipulation that mimics DR decreases starvation resistance because of the reduction in energy stores [31]. rpd3^def/+ and control (+/+) flies were raised on different food levels until 10 or 40 days of age and then exposed to starvation. We used 0.7 N as a DR diet and 1.5 N as a high calorie diet. We first compared the starvation resistance of aging control flies on 0.7 N to controls aged on 1.5 N food. At 10 days of age male and female control flies aged on 1.5 N food had a higher starvation resistance compared to flies kept on 0.7 N food (Fig. 3A,B, Table 3). However, at 40 days of age, there were no differences in starvation resistance for flies aged on 0.7 N or 1.5 N food (Fig. 3C,D, Table 3). Another interesting observation was that control male flies aged on 0.7 N or 1.5 N had a much lower starvation resistance compared to control females aged on the same food at 10 days of age. However, both male and female control flies had an almost identical starvation resistance at 40 days of age regardless of the content of the food on which they were aged (Fig. 3E). Male and female rpd3^def/+ flies were more resistant to starvation than +/+males at 0.7 N and 1.5 N (Fig. 3A-D, Table 3). In addition, starvation resistance in female rpd3^def/+ was much higher compared to males at both 10 and 40 days of age. Thus rpd3 reduction increases starvation resistance in rpd3^def/+ flies when aged on a standard, a low, or a high calorie diet.

Fig.3

Reduction of rpd3 increases resistance to starvation of male and female flies on a low and a high calorie diet. (A, B) Survival curves of rpd3^def/+ and +/+ male (A) and female (B) flies aged on 0.7 N or 1.5 N food levels and exposed to starvation at 10 days of age. (C, D) Survival curves of rpd3^def/+ and+/+male (C) and female (D) flies aged on 0.7 N or 1.5 N food levels and exposed to starvation at 40 days of age.

3.4The effect of reduced rpd3 mRNA levels on paraquat resistance

Increased resistance to oxidative stress is often associated with increased longevity and it has been suggested that improved stress response could be one of the contributors to longer life [24]. DR does not affect oxidative stress resistance of young flies but has a negative effect on aged flies [28]. We have reported that rpd3 reduction increases resistance to hydrogen peroxide, but this resistance depends on fly age and sex. Resistance is higher in females compared to males, and more pronounced at older ages [18]. Increased resistance to paraquat is also observed in flies with heart-specific rpd3 reduction, as well as whole body reduction in 2 days old flies [15]. However, the effect of whole body rpd3 reduction on paraquat resistance during aging has not been determined. There is an age-associated decline in resistance to paraquat in wild type flies, thus we examined if rpd3 reduction could postpone the known age-associated decline in resistance [32]. The response of rpd3^def/+ and rpd3^P-UTR/+ flies to paraquat depends on fly sex, age, and allele. rpd3^P-UTR/+ females had increased resistance to paraquat at both 10 and 40 days, while males had no difference compared to controls (Fig. 4A, B). In contrast, rpd3^def/+ females were less resistant to paraquat at 10 days while there was no difference at age 40. There was no difference in resistance to paraquat in rpd3^def/+ males at both ages (Fig. 4C, D). These data illustrate how genetic background, level of rpd3 reduction, age, and sex affect fly resistance to paraquat.

Fig.4

rpd3 reduction affects stress resistance in flies. (A, B) Survival curves for male and female rpd3^P-UTR/+ (experimental) and rpd3^P-1.8/+ (control) flies exposed to 20 mM paraquat at age 10 (A) or 40 (B) days. rpd3^P-UTR/+ females are more resistant to paraquat at 10 and 40 days, but no changes in resistance were observed in male rpd3^P-UTR/+ flies. (C, D) Survival curves for male and female rpd3^def/+ and control flies exposed to 20 mM paraquat at age 10 (C) or 40 (D). Male rpd3^def/+ flies are more resistant and female rpd3^def/+ are less resistant compared to control flies at age 10. No difference in paraquat resistance was observed between rpd3^def/+ male and female and control flies at 40 days.

4Discussion

DR has multiple effects on physiology. In flies, DR reduces fecundity, increases fly spontaneous physical activity, increases mitochondrial biogenesis, and slows down age-related mortality [22, 28, 33]. The effects of rpd3 reduction on lifespan in flies appear to at least partially overlap with the effects of DR, as DR and rpd3 whole body under-expression share a key effect on dSir2, takeout, chico, InR, and dFoxo levels of transcription [11, 13–15, 17–19]. Heart-specific rpd3 reduction increases the levels of Sod2, dSir2, dFoxo, Thor, and DptB in 2 day old flies [15]. dSir2 has been thought to mediate longevity extension in DR flies, illustrated by a lack of DR longevity effects in dSir2 mutant flies. Flies double mutant for dSir2 and rpd3 live shorter compared to single rpd3 mutant flies suggesting that dSir2 mediates at least partially the longevity effect in rpd3 flies [13]. In addition, rpd3 reduction and DR are not additive in their effects on lifespan [11]. Therefore, it was hypothesized that the mechanism of longevity extension in DR and rpd3 flies is the same or related [11, 13]. However, rpd3 mutants live longer under conditions close to starvation and during starvation, suggesting that DR and rpd3 may have distinct but interacting or overlapping effects on fly longevity [14, 18]. To gain further insight into the mechanism of lifespan extension observed in rpd3 flies, we determined if similar changes in physiology associated with DR can be observed in flies with reduced rpd3 levels. Particularly, we investigated the effects of decreased levels of rpd3 on spontaneous physical activity, fecundity, and stress resistance. rpd3^def/+ male flies had an increased peak in spontaneous activity at a younger age most likely mediated by increased dSir2 levels. Fly spontaneous physical activity decreases with age, and so do the differences in the peak activity. No difference was observed in rpd3^def/+ males at age 40. Although peak activity appears to be associated with rpd3 reduction in male rpd3^def/+ flies at a younger age, no clear difference in peak or total spontaneous physical activity was associated with rpd3 reduction in rpd3^def/+ females or rpd3^P-UTR/+ male and female flies. The two rpd3 alleles have different levels of rpd3 mRNA that could explain difference in the peak of spontaneous physical activity or resistance to paraquat. While rpd3 mRNA expression is reduced to 50% in the rpd3^def/+ heterozygous flies compared to controls, the level of rpd3 mRNA is reduced by 40% in rpd3^P-UTR/+ compared to rpd3^P-1.8/+ flies. The levels of rpd3 mRNA are linked to different physiological consequences, as well as longevity effects. For instance, while males of both rpd3 alleles and female rpd3^P-UTR/+ flies have dramatically extended longevity, rpd3^def/+ females only have extended maximal lifespan but not median lifespan [11]. Similarly, different effects on starvation resistance were observed in flies with whole body rpd3 reduction using RNAi mediated silencing compared to reduction only in fly heart tissue [15]. Likewise, the two rpd3 alleles respond differently to paraquat. However, we show that H₂O₂ and starvation resistance were similar in both rpd3 alleles (discussed below).

DR has a negative effect on egg production in female flies due to a shift in use of available resources from reproduction to maintenance [25, 28]. If rpd3 reduction mimics DR we would expect lower fecundity in rpd3 females aged on a DR diet. However, no difference in fecundity was associated with rpd3 reduction. No difference in fecundity could be explained by higher energy storage and/or increased adaptation to DR found in rpd3 flies, as discussed below [18].

The effects on starvation and oxidative resistance caused by DR of wild type flies changes with age [28]. While DR increases starvation resistance early in life, the opposite was found in aged DR flies [28]. Increased resistance to starvation of young flies is associated with increased lipid content. However, this increased lipid storage is not sustained over longer time, which results in lower starvation resistance in older flies [28]. We previously reported that rpd3^def/+ and rpd3^P-UTR/+ flies have increased starvation resistance [18]. Heart-specific rpd3 reduction also increases starvation resistance in flies at 2 days of age [15]. Nevertheless, the same authors did not find increased starvation resistance in flies when rpd3 was reduced in other tissues. Here we examined how aging rpd3 or control flies on 0.5 N or 1.5 N affect their starvation resistance. Our control flies aged on 1.5 N food had a small increase in starvation resistance compared to flies aged on 0.5 N food at an early age. However, no difference in starvation resistance between control male or female flies was observed at age 40 days, and diet did not have an effect at this age either. We observed increased starvation resistance in both rpd3 alleles, with females showing a larger resistance compared to male rpd3 flies. An increase in lipid storage caused by an increase in the activity of the enzymes associated with lipid biogenesis is often associated with increased starvation resistance [28]. Increased metabolic reserves in the form of both lipids and carbohydrates are associated with increased starvation resistance in an outbred population of Drosophila melanogaster. We recently reported that rpd3 reduction is associated with increased levels of triglycerides before and after starvation, which is at least partially mediated by reduced IIS [18]. Consistently, flies with ablated insulin-producing cells have increased levels of lipids and starvation resistance [35]. Flies with reduced rpd3 levels have also increased levels of glucose, glycogen, and trehalose at older ages, illustrating increased energy storage [18]. It is also possible that increased resistance to starvation observed in rpd3 flies results from metabolic adaptation to nutrient availability. Such changes could be mediated by changes in acetylation of metabolic enzymes, which could affect their enzymatic activity allowing a shift from one metabolic pathway to another [36, 37]. It was shown that histone modifying enzymes link changes in nutrient availability to changes in intermediary metabolism by affecting the activity of the enzymes involved in glycogenesis, glycolysis, gluconeogenesis, and β-oxidation through acetylation [36, 37].

DR has no effect on oxidative stress resistance early in life but decreases resistance to paraquat as flies age [28]. We showed that rpd3 reduction increases resistance to oxidative stress, as measured by increased resistance to H₂O₂. This resistance is mediated by reduced IIS, illustrated by increased dFoxo mRNA levels found in rpd3 mutant flies, and lack of increased resistance to H₂O₂ and shorter lifespan observed in flies double mutant for dFoxo and rpd3, compared to single rpd3 mutant flies [18]. This is in agreement with findings that overexpression of nuclear localized dFoxo mediates increased resistance to oxidative stress in flies [38]. Furthermore, flies treated with PBA, a HDAC1 inhibitor, have increased expression of genes that have been implicated in response to oxidative stress such as glutathione S-transferase and SOD [16]. PBA-treated flies have increased climbing activity, increased resistance to starvation and to paraquat, and extended lifespan [16].

Several findings presented here further confirm that rpd3 reduction extends longevity by a mechanism that partially overlaps with DR but has also distinct characteristics. rpd3^def male flies have an increased peak of spontaneous physical activity. However, rpd3 flies have no changes in fecundity, total spontaneous physical activity, and are more resistance to starvation and oxidative stress compared to controls. Our data illustrates how complex interactions between reduced rpd3 levels and its downstream targets results in various phenotypic changes at the organismallevel.