Drosophila egg-derived tyrosine phosphatase (EDTP): a novel target for improved survivorship to prolonged anoxia and cellular protein aggregates

Drosophila egg-derived tyrosine phosphatase (EDTP), a lipid phosphatase that removes 3-position phosphate at the inositol ring, has dual functions in the oogenesis and the muscle performance during adult stages. A mammalian homologous gene MTMR14, which encodes the myotubularin-related protein 14, negatively regulates autophagy. Mutation of EDTP/MTMR14, however, causes at least three deleterious consequences: (1) lethality in the early embryogenesis in Drosophila; (2) “jumpy” phenotype with apparently impaired motor functions; and (3) association with a rare genetic disorder called centronuclear myopathy. Here we show that flies carrying a heterozygous EDTP mutation had increased survivorship to prolonged anoxia; tissue-specific downregulation of EDTP in non-muscle tissues, particularly motoneurons, extended the lifespan; and tissue-specific downregulation of EDTP in motoneurons improved the survivorship to beta-amyloid peptides (Aβ42) and polyglutamine (polyQ) protein aggregates. MTMR14 expression was evident in the hippocampus and cortex in C57BL/6J and APP/PS1 mice. Compared with C57BL/6J mice, APP/PS1 mice had reduced MTMR14 in the cortex but not in the hippocampus. Hippocampal expression of MTMR14 was increased and plateaued at 9-17 months compared with 2-6 months in C57BL/6J mice. Aβ42 treatment increased the expression of MTMR14 in the primarily cultured hippocampal neurons of Sprague/Dawley rats and mouse Neuro2a neuroblasts. We demonstrated a novel approach of tissue-specific manipulation of the disease-associated gene EDTP/MTMR14 for lifespan extension and the improvement of survivorship to cellular protein aggregates.

The most interesting characteristic of EDTP expression is that there are two peaks, one at oogenesis [4,5], and another at adult stages [6,7]. The transcription of MTMR14, a mouse homolog of EDTP, also shows a peak at day five of differentiation in C2C12 myoblasts, followed by a decline [1]. Human MTMR14 transcripts are detectable at the ages between 19 and 69 years in the tested tissues, including heart, brain, placenta, lung, liver, skeletal muscle, kidney, and pancreas [1]. Levels of MTMR14 are highly coincident with those of Vps34 in the human brain, heart, and skeletal muscles [8], indicating a tightly regulated biological process which is associated with cellular PtdIns3P levels.
The decline of EDTP between early embryogenesis and young adult stages is accompanied with the Drosophila metamorphosis, a process requiring extensive autophagy and apoptosis for histolysis [9]. These observations suggest a role for EDTP in the regulation of autophagy. This is indeed supported by the findings that PtdIns3P stimulates autophagy in human HT-29 cells [10], that MTMR14 negatively controls the autophagosome formation and maturation in mammalian cells [11], and that the EDTP/MTMR14 inhibitor, AUTEN-99, activates autophagy in human HeLa cells and mouse tissues [12]. Interestingly, the latter group also shows that overexpression of a modified EDTP (by skipping the first exon) in the fat body antagonizes the effect of AUTEN-67 in inducing autophagy [13]. 4 Despite the negative regulation of autophagy, the deletion of Drosophila EDTP is lethal during embryogenesis or in the first instar, and germline clones with a null EDTP allele fail to produce mature oocytes [5]. Homozygous flies carrying a hypomorphic EDTP allele are shortlived with impaired motor functions and reduced fecundity [14]. Additionally, muscles of the MTMR14-deficient mice have decreased force production, prolonged relaxation and exacerbated fatigue [15]. A human MTMR14 missense variant (R336Q) is associated with centronuclear myopathy, a rare genetic disorder with muscle weakness and wasting [1]. Therefore, the function of autophagy initiation is likely overwhelmed by the lethality or disease-causing effects due to a ubiquitous loss of EDTP/MTMR14.
There are advantages of autophagy in degrading and recycling disrupted organelles, longlived proteins, and denatured protein aggregates [10,16]. A strategy to maximize the potential benefit of EDTP is to manipulate its downregulation in the favorable tissues while leaving the expression intact in oocytes and as well as in muscles at adult stages. This seems to be feasible by using the Drosophila Gal4/UAS expression system [17]. We thus hypothesized that selective downregulation of EDTP in non-muscle tissues, particularly the central nervous system, extends lifespan and improves the survivorship to cellular protein aggregates in Drosophila.
In the current study, we demonstrate that heterozygous EDTP mutants had improved survival to prolonged anoxia, a condition mimicking the extreme hypoxia that induces autophagy in mammalian and human cells [18]. We also show that selective downregulation of EDTP in the fly motoneurons extended lifespan and increased the survivorship to beta-amyloid peptides (Aβ42) or polyglutamine (polyQ) protein aggregates. The expression of MTMR14 in the hippocampus and cortex in mice was evident, promising a potential application by targeting EDTP/MTMR14 for the treatment of neurodegeneration.
Heterozygous mutants (EDTP DJ694 /+) and their sibling controls (w 1118 ) were prepared by two consecutive crosses between homozygous EDTP DJ694 mutants and w 1118 flies. We chose virgin female offspring (EDTP DJ694 /+) from the first mating and crossed them with w 1118 males. Flies carrying the EDTP mutation (red-eyed) and their siblings (white-eyed) were collected for survival analysis.
C57BL/6J mice and Sprague/Dawley rats were purchased from the Experimental Animal Centre of Wuhan University. The C57BL/6J mice are wild-type whereas APP/PS1 (stock number 004462, Jackson Laboratory) are the transgenic mouse models of Alzheimer's disease.
The APP/PS1 mice express a chimeric mouse/human amyloid precursor protein (Mo/HuAPP695swe) and a mutant human presenilin 1 (PS1-dE9) both directed to CNS neurons.
The mice were housed in a light/dark (12h:12h) cycle in standard group cages (4-5 per cage) 6 with accessible food and water ad libitum. The Sprague/Dawley rats were used for the isolation of primary hippocampal neurons for cell culture. Experimental procedures with mice were approved by the Ethics Committee of Tongji Medical College, Huazhong University of Science and Technology.

Post-anoxia survival
Newly emerged flies were collected and aged to 6-8 days in regular culture conditions and exposed to an anoxia (generated by pure argon gas) during the light phase of the photoperiod.
The percent survival at 12 h (% 12-h survival) after anoxia was examined. Replicated groups (15-25 flies per group, n = 3 ~ 7) of flies were examined. We chose the anoxia exposure of 1, 3, or 6 h because flies are highly tolerant to anoxia for hours [20].
For the tests of post-anoxia long-term survivorship, flies were exposed to anoxia for 0 or 6 h.
The dead flies were scored daily during the first week of recovery, and twice a week thereafter until all the flies were counted. Throughout the long-term survival experiments, alive flies were transferred to fresh food vials twice a week.

Lifespan experiments
Newly emerged flies were collected and raised at a density of 20-25 flies per vial. They were transferred into fresh culture media twice a week.

Cell Culture
The rat primary hippocampal neurons were prepared by following a reported method [21].
Briefly, hippocampal neurons from the 18-day-old embryonic (E18) Sprague/Dawley rats were isolated and seeded at 30,000 -40,000 cells per well on 6-well plates coated with Poly-D-Lysine/Laminin (Bioscience) in the neurobasal medium (Invitrogen), which was supplemented with 2% B27/0.5 mM glutamine/25 mM glutamate. Half of the culture medium was changed every 3 days with neurobasal medium supplemented with 2% B27 and 0.5 mM glutamine. All cultures were kept at 37 º C in a humidified 5% CO2 cultural condition.
Neuro2a cells were cultured at 37 °C in a 1:1 mixture of DMEM (Dulbecco's modified Eagle's medium) and OPTI-MEM supplemented with 15% FBS (fetal bovine serum, Gibco), 100 units/ml penicillin, and 100 mg/ml streptomycin. A humidified atmosphere containing 5 % CO2 was provided. The cells were plated on to six-well plates overnight and treated with 5 M A42 or vehicle control (DMSO) for 24 h.

Immunohistochemistry
Mice were anesthetized and perfused through aorta with 0.9% NaCl followed by phosphate buffer containing 4% paraformaldehyde. The brains were removed and post-fixed in perfusate overnight. Tissue sectioning (20 µm) was performed with a vibratome (Leica Biosystems). The sections were fixed with 0.3% H2O2 in the absolute methanol for 30 min and saturated with 8 bovine serum albumin (BSA) for 30 min at room temperature. Tissues were then incubated with polyclonal anti-MTMR14 overnight at 4 °C. After wash, the tissues were incubated with appropriate secondary antibody. Immunoreaction was developed with diaminobenzidine using Histostain-SP Kits (Zymed, CA, USA). Images were taken using a light microscope (Olympus BX60, Tokyo, Japan).
Blots were incubated with anti-rabbit or anti-mouse IgG conjugated to IRDye (800 CW) at room temperature and visualized using the Odyssey Infrared Imaging System (LI-COR Biosciences, Lincoln, NE, USA). Protein bands were analyzed using ImageJ [23].

Statistics
Statistical analysis was conducted by using R [24] and the following packages: gdata, survival, survminer, and ggplot2. A two-way ANOVA with Tukey's test was applied for the analysis of 12-h survival of different genotypes of flies which were subject to anoxia with varying durations. A log-rank test was performed for the comparison of two survival curves. A log-rank test with the Benjamini -Hochberg ("BH") adjustment was used for the comparison of three survival curves. One-way ANOVA with Bonferroni's multiple comparisons was performed to analyze the MTMR14 expression among three groups of cells. A student t-Test was conducted to compare the levels of MTMR14 between two groups of cells. A value of P < 0.05 was considered statistical significance.

Improved 12-h survival to prolonged anoxia in male flies of EDTP mutant
EDTP DJ694 is an enhancer-trap line carrying a transposable element P{GawB} in the first intron with the orientation opposite to the affected EDTP gene [6,7]. Flies homozygous for EDTP DJ694 are hypomorphic (reduced EDTP transcription, short-lived, motor defective, and reduced fecundity) [14]. To avoid the potential confounding effect, we started to examine the flies heterozygous for EDTP DJ694 allele.

Increased post-exposure survivorship to prolonged anoxia in EDTP mutant
The increased 12-h survival to prolonged anoxia raised an immediate question: whether EDTP mutants have increased post-exposure survivorship throughout the life. Young flies (6-8 days old) were exposed to a 6-h anoxia and the post-exposure survivorship was examined. We used the 6-h exposure because the % 12-h survival was markedly different between heterozygous mutant and controls.
There was also no significant difference in the survivorship between female EDTP DJ694 /+

RNAi knockdown of EDTP in motoneurons extended lifespan
Extreme hypoxia induces autophagy in mammalian and human cells [18]. The mammalian homologous MTMR14 negatively regulates the autophagic removal of protein aggregates [11].
These data together with our findings suggested that downregulation of EDTP could promote autophagy and improve the survivorship. We used the Gal4/UAS binary system [17] for selective downregulation of EDTP in motoneurons and examined the lifespan. The rationale for targeting motoneurons was to avoid affecting the abundant expression of EDTP in muscles, ovaries, spermatheca, and oocytes. The motoneuron-specific driver D42-Gal4 [25] was used for RNAi knockdown.
Therefore, EDTP downregulation in motoneurons improved the survivorship to the expression of Aβ42 or polyQ protein aggregates.

Expression of MTMR14 in the hippocampus and cortex in C57BL/6J and APP/PS1 mice
Human MTMR14 is transcribed ubiquitously at relatively low levels in the brain [27]. Mouse MTMR14 is expressed in the muscles, liver, and fat but there is little evidence for the brain expression [28]. We examined the expression of MTMR14 in the hippocampus and cortex in C57BL/6J and APP/PS1 mice. C57BL/6J is wild-type, whereas APP/PS1 is a transgenic strain with constitutive expression of a chimeric mouse/human amyloid precursor protein (Mo/HuAPP695swe) and a mutant human presenilin 1 (PS1-dE9) in CNS neurons [29].
MTMR14 expression was observable in the CA3 pyramidal neurons and dentate gyrus (DG) in the hippocampus of a 3-month-old C57BL/6J mouse (Fig. 5a). MTMR14 expression was detectable in the hippocampus in both C57BL/6J and APP/PS1 mice. The levels in two strains were similar with no significant difference (Fig. 5b). MTMR14 expression was also detectable in the cortex in both strains. However, relative levels of MTMR14 in APP/PS1 mice were lower than those in C57BL/6J mice (P < 0.05, t-Test). These data indicated an association between the transgenic expression of amyloid precursor protein/presenilin 1 and MTMR14 level.
14 Relative levels of MTMR14 in the C57BL/6J hippocampus increased and plateaued at 9, 12 and 17 months compared with the levels at 2 and 6 months (Fig. 5c), suggesting an agedependent kinetics of MTMR14 expression.

Neuro2a cells
Aβ42 causes microRNA deregulation in primary hippocampal neurons in C57BL/6J mice [30] and inhibits the viability of the mouse neuroblastoma Neuro2a cells [31]. We explored the effect of Aβ42 treatment on MTMR14 expression in primary hippocampal neurons and Neuro2a cells.
The primary hippocampal neurons from 18-day-old embryonic (E18) Sprague/Dawley rats were prepared and treated with Aβ42. Relative MTMR14 levels were increased in the primary cultures after a 72-h incubation with 10 µM Aβ42 compared with controls (P < 0.01, One-way ANOVA with Bonferroni's multiple comparisons) (Fig. 6a). There was no significant increase of relative MTMR14 levels in the cultures incubated with 5 µM Aβ42 for 72 h. Data indicated a dosage-dependent induction of MTMR14 by Aβ42 treatment. We examined the MTMR14 expression in Neuro2a cells with Aβ42 treatment. MTMR14 expression was increased in Neuro2a cells treated with Aβ42 at 5 µM for 24 h compared with controls (P < 0.01, t-Test) (Fig.   6b). Therefore, Aβ42 treatment increased the expression of MTMR14 in both the primary hippocampal neurons of rats and Neuro2a cells. 15 We report in Drosophila a novel approach of fine-tuning the disease-associated gene EDTP/MTMR14 for lifespan extension and improved survivorship to cellular protein aggregates.

Discussion
Cell-specific downregulation of EDTP in non-muscle tissues circumvents the deleterious consequences of ubiquitous loss of EDTP. Specifically, downregulation of EDTP in motoneurons extends lifespan and increases the survivorship to Aβ42 or polyQ protein aggregates. The expression of the mouse MTMR14 in the hippocampus and cortex, together with its age-dependent kinetics and the Aβ42-induced increase of MTMR14 in the rat's cultured hippocampal neurons, promises a potential application by targeting EDTP/MTMR14 in a tissuespecific manner for the treatment of neurodegeneration.
The proposed cell-specific downregulation of EDTP/MTMR14 for beneficial consequences is formulated largely from the preliminary observation that heterozygous flies carrying the EDTP DJ694 allele have increased survivorship to prolonged anoxia. Homozygous flies carrying this allele display a "jumpy" phenotypeimpaired motor function associated with shortened lifespan and reduced fecundity [14]. These findings indicate that downregulation of EDTP in flies homozygous for EDTP DJ694 is overall deleterious, whereas moderate downregulation of EDTP in heterozygous flies causes no obvious motor defect and becomes protective if flies are exposed to a prolonged anoxia. Such a conditional benefit highlights the importance of subtle manipulation of spatiotemporal expression of EDTP/MTMR14 for the desired effect.
EDTP/MTMR14 has two main functions: negative regulation of autophagy and disease-causing effect if deficient in the muscles [1,[11][12][13][14][15]. Drosophila EDTP has two peaks of expression, one at oogenesis and another at adult stages around 20-30 days [4][5][6][7]. Both male and female adult flies have abundant EDTP in muscles and additionally, female flies express rich amounts of EDTP in the spermatheca and ovaries. The differential expression of EDTP in female flies might 16 also be responsible for the enhanced short-term survival (i.e. % 12-h survival) to prolonged anoxia. The dual functions, multiple peaks over time, broad expressing tissues, and the sexual dimorphism of expression pattern together do not likely allow a complete deletion of EDTP/MTMR14 to be associated with beneficial consequences. Precise control of EDTP/MTMR14 expression in the favorable cells or tissues could be essential for eliciting the protective effects. The current study presents a strategy of cell-specific downregulation of Drosophila EDTP in motoneurons with beneficial effects in lifespan extension and improved survivorship to cellular protein aggregates.
Motoneurons are highly specialized and terminally differentiated with reduced function of global genome repair [32,33]. More importantly, neuronal cells utilize autophagy as an essential process for normal turnover of cytoplasmic contents [34]. Drosophila motoneuronal expression of a human Cu-Zn superoxide dismutase (SOD1) leads to a marked extension of lifespan by up to 40% [35]. Motoneuronal overexpression of the heat shock protein 70 results in structural plasticity of axonal terminals which is associated with increased larval thermotolerance [36].
Motoneurons innervate muscle cells through neuromuscular junctions, making them an ideal target for EDTP downregulation while leaving muscular expression intact. The findings that motoneuronal downregulation of EDTP extends lifespan and improves the survivorship to Aβ42 or polyQ protein aggregates firmly support motoneurons as a favorable targeting tissue. Notably, our findings greatly rely on the putative motoneuronal driver D42-Gal4, which also shows a prominent expression pattern in the peripheral sensory neurons [37]. Whether sensory neurons utilize the EDTP/MTMR14-associated autophagy for removing cellular wastes is unclear, but a study has shown the connection between the sensory neurons and lifespan extension [38].
EDTP/MTMR14 possesses a function opposite to Vps34 in the regulation of the PtdIns3P pool. Vps34 is the only phosphatidylinositol 3-kinases in yeast [2] and has been evolutionarily conserved through mammals. Vps34 plays an essential role in the process of autophagy [39,40].
Downregulation of EDTP/MTMR14 shifts the focus from Vps34 to its functional opponent for the regulation of PtdIns3P and thus represents a novel approach to manipulate the process of autophagy.     Presented data are from male flies. *** P < 0.001 from Log-rank test with "BH" adjustment.