Roles of ZnT86D in Neurodevelopment and Pathogenesis of Alzheimer Disease in a Drosophila melanogaster Model

Zinc is a fundamental trace element essential for numerous biological processes, and zinc homeostasis is regulated by the Zrt-/Irt-like protein (ZIP) and zinc transporter (ZnT) families. ZnT7 is mainly localized in the Golgi apparatus and endoplasmic reticulum (ER) and transports zinc into these organelles. Although previous studies have reported the role of zinc in animal physiology, little is known about the importance of zinc in the Golgi apparatus and ER in animal development and neurodegenerative diseases. In this study, we demonstrated that ZnT86D, a Drosophila ortholog of ZnT7, plays a pivotal role in the neurodevelopment and pathogenesis of Alzheimer disease (AD). When ZnT86D was silenced in neurons, the embryo-to-adult survival rate, locomotor activity, and lifespan were dramatically reduced. The toxic phenotypes were accompanied by abnormal neurogenesis and neuronal cell death. Furthermore, knockdown of ZnT86D in the neurons of a Drosophila AD model increased apoptosis and exacerbated neurodegeneration without significant changes in the deposition of amyloid beta plaques and susceptibility to oxidative stress. Taken together, our results suggest that an appropriate distribution of zinc in the Golgi apparatus and ER is important for neuronal development and neuroprotection and that ZnT7 is a potential protective factor against AD.


Introduction
Zinc is a fundamental trace element essential for the activity of over 300 enzymes and 1000 transcription factors [1]. It is the most abundant heavy metal in the brain and plays an essential role in normal brain functions [2]. Approximately 90% of zinc in the brain exists in a protein-bound state and confers catalytic activity and structural stability [3]. A significant amount of the chelatable zinc is present in the presynaptic vesicles of glutamatergic neurons [2]. They are secreted into synapses and modulate neurotransmission by interacting with a number of zinc-permeable channels [2]. Zinc is also crucial for normal brain development [4]. When zinc is depleted in the early postnatal period, the developmental apoptosis of neurons necessary for proper formation of the central nervous system decreases remarkably in rats [5,6]. In addition, maternal zinc deficiency during pregnancy reduces nestin levels in offspring, which could cause abnormal neurogenesis in mice [7].
It is also well known that heavy metals, such as zinc and copper, are involved in the pathology of Alzheimer disease (AD) [8]. AD is a chronic neurodegenerative disorder that accompanies dementia and progressively worsens over decades [9,10]. Brain lesions 2 of 17 of patients with AD are characterized by amyloid beta (Aβ) plaques and neurofibrillary tangles. The exact cause of AD remains unknown, and there is no definitive cure. In particular, zinc is colocalized with Aβ plaques in the brain tissue of patients with AD [11], and zinc binds to Aβ peptide and promotes its aggregation in vitro [12,13]. In addition, increased zinc levels are noted in the cortex and serum of patients with AD compared to those in age-matched controls [14,15]. Clioquinol, a drug that chelates zinc and copper, and PBT-2, a second-generation chelator, have shown promising results in clinical trials [2].
Zinc homeostasis is regulated by two major zinc transporter families: the Zrt-/Irt-like protein (ZIP) family and the zinc transporter (ZnT) family [16]. ZIPs transport zinc into the cytosol by importing it from the extracellular space or exporting it into the cytoplasm from intracellular organelles, thereby increasing the intracellular zinc concentration [16]. By contrast, ZnTs, encoded by the solute carrier family 30 (SLC30A) genes, generally decrease the zinc level in the cytoplasm. When the intracellular zinc concentration is elevated, zinc is released into the extracellular matrix or into intracellular compartments through ZnTs [17]. Ten ZnT family members have been identified thus far [17]. ZnT1 is localized to the plasma membrane and releases zinc into the extracellular space [17]. By contrast, ZnT2-8 are embedded in the membrane of intracellular compartments and transport zinc into these compartments [17]. In Drosophila, seven ZnT homologs have been identified based on phylogenetic alignments: ZnT49B, ZnT41F, ZnT35C, ZnT33D, ZnT86D, ZnT77C, and ZnT63C [18]. ZnT63C, a homolog of ZnT1, is primarily embedded in the basolateral membrane of enterocytes and transports zinc into the extracellular space, thereby absorbing dietary zinc [19]. ZnT35C, a homolog of ZnT2-4 and ZnT8, is localized on the apical side of the Malpighian tubules [20] and plays a role in the excretion of zinc into the lumen of the tubules [21]. It is also required for zinc storage granule biogenesis in Malpighian tubules [22].
ZnT7 is mainly localized in the Golgi apparatus, and a recent study suggested that ZnT7 is also present in the endoplasmic reticulum (ER), where it transports zinc into organelles [23,24]. Several studies have investigated the physiological role of ZnT7 in ZnT7-knockout (KO) mice. For example, ZnT7-KO mice were zinc deficient, had low body fat accumulation [25], and manifested impaired glucose tolerance involved with insulin resistance [26]. In addition, ZnT7 was colocalized with Aβ plaques, and there was a significant increase in ZnT7 expression in the hippocampus and cortex of APP/PS1 transgenic mouse mice [27]. Furthermore, ZnT7 colocalized with Aβ deposits in the brains of patients with AD [28].
ZnT86D, a Drosophila ortholog of ZnT7, is located in the Golgi apparatus [29] as in mammals, and a few studies have been conducted to investigate the role of ZnT86D in zinc homeostasis. For instance, ZnT86D overexpression in neurons resulted in complete developmental lethality [30]. In addition, when ZnT86D was overexpressed in the eyes, the eclosed flies displayed depigmentation in the eyes and had flatter eyes than control flies [29]. As such, ZnT86D is thought to play a crucial role in the development of various tissues; therefore, an in-depth study of the role of ZnT86D during development should be conducted.
ZnT86D regulates zinc homeostasis by altering the intracellular zinc distribution. Considering that zinc and AD are closely related, it is highly probable that ZnT86D is involved in AD pathogenesis. For example, the genetic inhibition of dZIP1, an ortholog of ZIP1 located in the cell membrane that transports zinc to the cytoplasm, ameliorated the pathology of a Drosophila AD model [31]. The expression level of dZIP1 was altered in the brains of Aβ42-overexpressing flies, and when it was silenced in the neurons, the amount of Aβ oligomers and fibril deposits was reduced, ameliorating neurodegeneration and improving cognitive function [31]. As zinc promotes the aggregation of Aβ peptides [12,13], these results were attributed to decreased zinc levels in the brain by downregulating dZIP1. Thus, in light of the results of the above study, zinc homeostasis seems to be deeply involved in the pathogenesis of AD. In particular, the zinc distribution inside neurons and the zinc content in the brain are likely to play an important role in this process.
In this study, we examined the role of ZnT86D in neuronal development in Drosophila as well as in neuroprotection against Aβ toxicity. When ZnT86D was silenced in the neurons using ZnT86D RNAi, body size, survival rate, and locomotor activity were reduced, and the RNAi flies manifested abnormal neurogenesis and increased cell death. In addition, the downregulation of ZnT86D in developing wings interrupted normal wing development. Furthermore, silencing ZnT86D in the neurons of a Drosophila AD model increased cell death and exacerbated neurodegeneration. These results indicate that zinc distribution between the cytoplasm and Golgi apparatus is crucial for the development of neurons and wings, as well as the pathogenesis of AD.

Silencing ZnT86D in Neurons or Wings Exerted Toxic Effects on the Development of Each Tissue
To investigate the role of ZnT86D during Drosophila development, we silenced it in the neurons, developing wings and eyes, using several GAL4 drivers. When this gene was silenced in the neurons using elav-GAL4, the proportion of individuals emerging from embryos to adults decreased significantly ( Figure 1A). In addition, the survived ZnT86D RNAi flies had wing expansion defects and reduced body size ( Figure 1B,C). [12,13], these results were attributed to decreased zinc levels in the brain by downregulating dZIP1. Thus, in light of the results of the above study, zinc homeostasis seems to be deeply involved in the pathogenesis of AD. In particular, the zinc distribution inside neurons and the zinc content in the brain are likely to play an important role in this process.
In this study, we examined the role of ZnT86D in neuronal development in Drosophila as well as in neuroprotection against Aβ toxicity. When ZnT86D was silenced in the neurons using ZnT86D RNAi, body size, survival rate, and locomotor activity were reduced, and the RNAi flies manifested abnormal neurogenesis and increased cell death. In addition, the downregulation of ZnT86D in developing wings interrupted normal wing development. Furthermore, silencing ZnT86D in the neurons of a Drosophila AD model increased cell death and exacerbated neurodegeneration. These results indicate that zinc distribution between the cytoplasm and Golgi apparatus is crucial for the development of neurons and wings, as well as the pathogenesis of AD.

Silencing ZnT86D in Neurons or Wings Exerted Toxic Effects on the Development of Each Tissue
To investigate the role of ZnT86D during Drosophila development, we silenced it in the neurons, developing wings and eyes, using several GAL4 drivers. When this gene was silenced in the neurons using elav-GAL4, the proportion of individuals emerging from embryos to adults decreased significantly ( Figure 1A). In addition, the survived ZnT86D RNAi flies had wing expansion defects and reduced body size ( Figure 1B   Furthermore, to examine the role of ZnT86D in the wings during development, ZnT86D RNAis were expressed under the control of MS1096-GAL4, ptc-GAL4, and en-GAL4. MS1096-GAL4 drives gene expression in the dorsal wing pouch of the wing imaginal disc during development, ptc-GAL4 in the anterior/posterior boundary, and en-GAL4 in the posterior compartment [32][33][34]. When ZnT86D was downregulated using the three drivers, crumpled and shrunken wings were observed ( Figure 1D). In particular, the ZnT86Di BL line was lethal when crossed with ptc-GAL4 or en-GAL4 lines ( Figure 1D).
As zinc is a fundamental trace metal essential for the activity of over 1,000 transcription factors [1], changes in intracellular zinc concentration due to ZnT86D knockdown may affect the expression of genes that play a pivotal role in wing development, such as engrailed (en) and wingless (wg). Therefore, we investigated whether changes in the expression patterns of en or wg could be the cause of these results but found no significant changes ( Figure 1E,F). This suggests that defects in wing development are not the result of the abnormal expression of genes that are crucial for wing compartmentation. Interestingly, no change in eye shape or size was observed when ZnT86D was silenced using GMR-GAL4 ( Figure 1G). Taken together, these data show that ZnT86D is essential for the development of neurons and wings, but not eyes.

Neuron-Specific Knockdown of ZnT86D Reduced Locomotor Activity and Lifespan, Accompanied by Abnormal Neurogenesis
To further characterize the role of ZnT86D in neurons, the locomotor activity of 2-dayold flies was measured using a climbing assay. Surprisingly, the climbing ability of the neuron-specific ZnT86D RNAi flies was drastically reduced ( Figure 2A). Interestingly, the RNAi flies manifested movements similar to tremors, which can be defined as rhythmic and involuntary movements of any body part [35]. In addition, the flies died within 9 days after eclosion ( Figure 2B).
(MS1096 > ZnT86Di BL and MS1096 > ZnT86Di KK ) flies. (G) Eye phenotypes of control (GMR-GAL4/+) and eye-specific ZnT86D RNAi (GMR > ZnT86Di BL and GMR > ZnT86Di KK ) flies. Data are expressed as the mean ± standard error of the mean (SEM). Furthermore, to examine the role of ZnT86D in the wings during development, ZnT86D RNAis were expressed under the control of MS1096-GAL4, ptc-GAL4, and en-GAL4. MS1096-GAL4 drives gene expression in the dorsal wing pouch of the wing imaginal disc during development, ptc-GAL4 in the anterior/posterior boundary, and en-GAL4 in the posterior compartment [32][33][34]. When ZnT86D was downregulated using the three drivers, crumpled and shrunken wings were observed ( Figure 1D). In particular, the ZnT86Di BL line was lethal when crossed with ptc-GAL4 or en-GAL4 lines ( Figure 1D).
As zinc is a fundamental trace metal essential for the activity of over 1,000 transcription factors [1], changes in intracellular zinc concentration due to ZnT86D knockdown may affect the expression of genes that play a pivotal role in wing development, such as engrailed (en) and wingless (wg). Therefore, we investigated whether changes in the expression patterns of en or wg could be the cause of these results but found no significant changes ( Figure 1E,F). This suggests that defects in wing development are not the result of the abnormal expression of genes that are crucial for wing compartmentation. Interestingly, no change in eye shape or size was observed when ZnT86D was silenced using GMR-GAL4 ( Figure 1G). Taken together, these data show that ZnT86D is essential for the development of neurons and wings, but not eyes.

Neuron-Specific Knockdown of ZnT86D Reduced Locomotor Activity and Lifespan, Accompanied by Abnormal Neurogenesis
To further characterize the role of ZnT86D in neurons, the locomotor activity of 2day-old flies was measured using a climbing assay. Surprisingly, the climbing ability of the neuron-specific ZnT86D RNAi flies was drastically reduced ( Figure 2A). Interestingly, the RNAi flies manifested movements similar to tremors, which can be defined as rhythmic and involuntary movements of any body part [35]. In addition, the flies died within 9 days after eclosion ( Figure 2B).  Neurodevelopmental disorders often involve impaired motor function and an increased risk of premature death [36,37]. Therefore, we examined whether the defects caused by the downregulation of ZnT86D were associated with neurogenesis by immunostaining mushroom bodies, a center for learning and memory in Drosophila brain. Surprisingly, ZnT86D RNAi flies had missing α-lobes, implying that neuronal development did not occur normally ( Figure 2C). Taken together, these data strongly suggest that ZnT86D is essential for neurodevelopment.

Silencing ZnT86D in Neurons Increased Cell Death, Which Was Not Associated with ER Stress or Hyperglycemia
To investigate whether the defects caused by the downregulation of ZnT86D were associated with increased cell death, we performed acridine orange (AO) staining in the larval brains. As shown in Figure 3A,B, the number of AO-positive cells in the larval brains was increased in the ZnT86D-downregulated groups compared to that in the control group, suggesting that the proper level of ZnT86D is critical for brain cell survival.
Neurodevelopmental disorders often involve impaired motor function and a creased risk of premature death [36,37]. Therefore, we examined whether the de caused by the downregulation of ZnT86D were associated with neurogenesis by munostaining mushroom bodies, a center for learning and memory in Drosophila b Surprisingly, ZnT86D RNAi flies had missing α-lobes, implying that neuronal dev ment did not occur normally ( Figure 2C). Taken together, these data strongly sugges ZnT86D is essential for neurodevelopment.

Silencing ZnT86D in Neurons Increased Cell Death, Which Was Not Associated with E Stress or Hyperglycemia
To investigate whether the defects caused by the downregulation of ZnT86D associated with increased cell death, we performed acridine orange (AO) staining i larval brains. As shown in Figure 3A,B, the number of AO-positive cells in the l brains was increased in the ZnT86D-downregulated groups compared to that in the trol group, suggesting that the proper level of ZnT86D is critical for brain cell surviv  A recent study suggested that ZnT7 is also present in the ER [24], carrying zinc into the organelle, and zinc has been implicated in ER stress and unfolded protein response (UPR) [38]. ER stress occurs when misfolded proteins accumulate and aggregate in the ER, and if these aggregates cannot be properly cleared by UPR, it leads to apoptosis [39]. To test the hypothesis that increased cell death occurred owing to increased protein aggregation, we used an FK2 antibody to verify whether polyubiquitinated protein aggregates were present in the brains of adult flies, but no sign of abnormal protein accumulation was noted ( Figure 3C). To confirm that the increased cell death was caused by ER stress, we used xbp1-EGFP, a marker of UPR, in which EGFP is expressed in frame only upon the occurrence of ER stress [40], and there was no significant change in the GFP signal between the ZnT86D-downregulated groups and the control group ( Figure 3D). Flies overexpressing Aβ42 in the neurons were used as positive controls in both experiments ( Figure 3C,D).
One of other possible causes of increased brain cell death in neuron-specific ZnT86D knockdown flies is increase in blood glucose level. Previous studies have reported metabolic abnormalities in ZnT7-KO mice such as glucose intolerance and insulin resistance [26,41,42]. As glucose intolerance is accompanied by hyperglycemia, which increases the risk of diabetes and cardiovascular disease, as well as apoptosis [43], we measured the amount of glucose in the larva hemolymph to determine whether the increased brain cell death is due to increased blood glucose. However, unlike in mice, there was no significant change in the blood glucose levels of ZnT86D RNAi larvae compared to control ( Figure 3E).
Taken together, neuron-specific knockdown of ZnT86D increased cell death in the developing brain, which was not associated with ER stress or high blood glucose level.

Downregulation of ZnT86D Exacerbated Aβ Toxicity in a Drosophila AD Model
It is known that one of the pathological traits of AD is neuronal apoptosis, implicated in neurodegeneration [44,45]. Neuronal apoptosis is induced by the toxic aggregation of Aβ, which eventually leads to neurodegeneration in AD [46,47]. As neuron-specific silencing of ZnT86D increased cell death, we investigated the role that ZnT86D may play in AD pathogenesis. First, to investigate whether there is a genetic interaction between ZnT86D and Aβ42, we simultaneously overexpressed Aβ42 and downregulated ZnT86D in the eyes. Interestingly, the Aβ42-induced rough eye phenotype was exacerbated by silencing ZnT86D (Figure 4A), but single expression of ZnT86D RNAi did not alter the eye phenotype ( Figure 1G), suggesting that ZnT86D deficiency enhanced Aβ42 cytotoxicity.
Additionally, we investigated the role that ZnT86D might have in Aβ-induced apoptosis and neurodegeneration. We used the GeneSwitch (GS) system, which allows temporally and spatially restricted expression of the desired genes [48], thereby excluding the effects of ZnT86D RNAi during development. We overexpressed Aβ42 and downregulated ZnT86D simultaneously for 20 days after eclosion in the neurons of adult flies by using elav-GS-GAL4 and detected cleaved death caspase 1 (active DCP-1) by immunohistochemistry. Unlike when ZnT86D was silenced during development, no prominent apoptosis was observed in this model system. However, we found that cell death induced by Aβ42 was exacerbated by downregulating ZnT86D (Figure 4B,C). In addition, the degree of neurodegeneration was investigated using histological analysis, and the area of neuronal loss was increased by silencing ZnT86D in the brain of 30-day-old flies ( Figure 4D,E). Taken together, these data demonstrate that neuron-specific ZnT86D downregulation exacerbates Aβ-induced cell death and neurodegeneration.

Knockdown of ZnT86D Did Not Increase the Deposition of Aβ Plaques
Deposition of excess Aβ aggregates is a major hallmark of AD [49], and it is known that zinc binds to Aβ peptides and promotes their aggregation in vitro [12,13]. In addition, Lang and colleagues demonstrated that silencing dZIP1, an ortholog of ZIP1 located in the cell membrane that transports zinc to the cytoplasm, in the neurons of a Drosophila AD model reduced the accumulation of Aβ deposits, and the authors attributed the results to the decreased zinc level in the brain by the downregulation [31]. Based on this, we hypothesized that silencing ZnT86D would affect the level of Aβ deposits in the brain of the Drosophila AD model. Therefore, we conducted thioflavin S staining and measured the amount of Aβ deposits after downregulating ZnT86D for 30 days in the neurons of adult flies. However, there was no significant difference between ZnT86D RNAi flies and control flies, indicating that knockdown of ZnT86D did not affect the accumulation of Aβ plaques ( Figure 5A,B).
apoptosis was observed in this model system. However, we found that cell death induced by Aβ42 was exacerbated by downregulating ZnT86D (Figure 4B,C). In addition, the degree of neurodegeneration was investigated using histological analysis, and the area of neuronal loss was increased by silencing ZnT86D in the brain of 30-day-old flies ( Figure  4D,E). Taken together, these data demonstrate that neuron-specific ZnT86D downregulation exacerbates Aβ-induced cell death and neurodegeneration.

Knockdown of ZnT86D Did Not Increase the Deposition of Aβ Plaques
Deposition of excess Aβ aggregates is a major hallmark of AD [49], and it is known that zinc binds to Aβ peptides and promotes their aggregation in vitro [12,13]. In addition, Lang and colleagues demonstrated that silencing dZIP1, an ortholog of ZIP1 located in the cell membrane that transports zinc to the cytoplasm, in the neurons of a Drosophila AD model reduced the accumulation of Aβ deposits, and the authors attributed the results to the decreased zinc level in the brain by the downregulation [31]. Based on this, we hypothesized that silencing ZnT86D would affect the level of Aβ deposits in the brain of the Drosophila AD model. Therefore, we conducted thioflavin S staining and measured the amount of Aβ deposits after downregulating ZnT86D for 30 days in the neurons of adult flies. However, there was no significant difference between ZnT86D RNAi flies and control flies, indicating that knockdown of ZnT86D did not affect the accumulation of Aβ plaques ( Figure 5A,B).

Silencing ZnT86D Did Not Affect the Susceptibility of Aβ42-Overexpressing Flies to Oxidative Stress
Zinc is known to induce oxidative stress in the brain, and oxidative damage caused by an altered redox balance is a key pathophysiological feature of AD [50][51][52][53][54][55]. Therefore, we speculated that changes in the distribution of zinc in neurons by downregulating ZnT86D would affect the susceptibility of the Drosophila AD model to hydrogen peroxide (H 2 O 2 ), which may mediate the enhanced Aβ42 cytotoxicity observed in ZnT86D RNAi flies. As shown in Figure 6 and Table 1, when 20-day-old flies were exposed to H 2 O 2 , the survival rate of ZnT86Di BL flies was higher than that of control flies, while that of ZnT86Di KK flies was not. These data suggest that the enhanced cytotoxicity of Aβ42 observed in ZnT86D RNAi flies was not the result of increased sensitivity to H 2 O 2 .
ZnT86D would affect the susceptibility of the Drosophila AD model to hydrogen peroxide (H2O2), which may mediate the enhanced Aβ42 cytotoxicity observed in ZnT86D RNAi flies. As shown in Figure 6 and Table 1, when 20-day-old flies were exposed to H2O2, the survival rate of ZnT86Di BL flies was higher than that of control flies, while that of ZnT86Di KK flies was not. These data suggest that the enhanced cytotoxicity of Aβ42 observed in ZnT86D RNAi flies was not the result of increased sensitivity to H2O2.

Discussion
ZnT7 is located mainly in the Golgi apparatus where it transports zinc from the cytoplasm into the Golgi apparatus [23]. Its physiological role has been actively studied using various models since its discovery in 2003. Interestingly, ZnT7-KO mice manifest metabolic abnormalities, including reduced body zinc status, body fat accumulation, glucose intolerance, and insulin resistance [25,26,41]. This suggests that ZnT7 plays a crucial role in development, and in-depth studies on the role of ZnT7 in this process are needed. Therefore, we silenced ZnT86D, a Drosophila ortholog of ZnT7, in a specific tissue by using the GAL4/UAS system, a powerful genetic tool in the field of research using Drosophila melanogaster as a model organism.
Surprisingly, neuron-specific ZnT86D knockdown had fatal effects on development, and RNAi flies showed extremely toxic phenotypes. In particular, the proportion of individuals eclosed to adults from the embryo, and the survival rate after eclosion decreased dramatically (Figures 1A and 2B). These results are noteworthy in that adult flies did not eclosed from larvae when ZnT86D was overexpressed in neurons, which was attributed to localized zinc toxicity by the manipulation [30]. Therefore, the data suggest that the zinc distribution between the Golgi apparatus and cytoplasm is pivotal for normal neuronal development. Consistent with these results, the mushroom body, a center for learning and memory in Drosophila, was not formed normally in ZnT86D RNAi flies ( Figure 2C). Indeed, zinc has been known to be strongly implicated in neurogenesis. For example, maternal zinc deficiency in mice impaired the expression of nestin, a marker of neural stem cells, in the offspring [7]. However, zinc intake during pregnancy boosted neuronal proliferation in the developing fetus [56]. Likewise, zinc homeostasis is crucial for normal neuronal development.
In addition, the locomotor activity and lifespan of ZnT86D RNAi flies were dramatic ally decreased, consistent with the fact that neurodevelopmental disorders often accompany impaired motor function and increase the risk of premature death [36,37]. The RNAi flies also displayed symptoms similar to those of tremors, defined as rhythmic and involuntary movements of any body part [35]. According to a case report, 14-month-old male infants with infantile tremor syndrome showed zinc deficiency, suggesting an association between tremor and zinc [57]. Therefore, further studies are required to determine whether ZnT86D RNAi flies are zinc-deficient, thereby manifesting abnormal neurodevelopment and motility.
As shown in Figure 3A,B, we conducted AO staining to investigate whether the various defects caused by silencing ZnT86D were associated with cell death, and it was confirmed that cell death increased in the brains of ZnT86D RNAi flies. According to a recent study, ZnT7 knockdown resulted in ER stress, and subsequently, JNK was activated in a Drosophila model of the malignant tumor Raf GOF scrib −/− [58]. It is well known that JNK can induce apoptosis by upregulating pro-apoptotic genes [59]. Based on the results of this study we examined whether the increased cell death was due to ER stress through xbp1-EGFP, a marker of UPR, and anti-polyubiquitin immunostaining, but there were no significant changes in the ZnT86D-downregulated groups compared to those in the control group. This suggests that the increased cell death was not caused by ER stress, and the discrepancy between the results of Wei et al. (2021) and this study might be attributed to the difference in the model flies and investigated tissues.
In 2012, Lang et al. (2012) demonstrated that genetic inhibition of dZIP1 in neurons of a Drosophila AD model ameliorated Aβ pathology. When dZIP1 was downregulated, brain zinc levels and the accumulation of Aβ42 fibril deposits were reduced [31]. Because zinc binds to the Aβ peptide and promotes its aggregation [12,13], Lang et al. (2012) concluded that the decrease in Aβ deposits was due to the reduced zinc concentration in the brain by dZIP1 knockdown. In other words, silencing dZIP1 in the neurons of the Drosophila AD model decreased the zinc level in the brain, leading to less oligomer formation, which in turn resulted in a decrease in the amount of Aβ plaques. By contrast, downregulation of ZnT86D in the neurons had no effect on Aβ accumulation ( Figure 5A,B), suggesting that silencing ZnT86D does not elevate the level of cytosolic zinc enough to enhance Aβ aggregation.
As the increase in apoptosis and exacerbation of neurodegeneration by ZnT86D downregulation were not associated with the accumulation of Aβ42 plaques, we hypothesized that the lethal effects would have been manifested by increased cellular stress. In particular, since zinc is known to be associated with oxidative stress in the brain [49,51,53,54] and ZnT7 is known to protect mouse osteoblast MC3T3-E1 cells from oxidative stress-induced apoptosis through the PI3K/Akt and MAPK/ERK pathways [60], we postulated that the manipulation made neurons susceptible to oxidative stress. However, there was no significant change in the survival rate after exposure to H 2 O 2 between ZnT86D RNAi flies and control flies, suggesting that the increase in cell death and exacerbation of neurodegeneration in the fly model brain were not associated with oxidative stress.
Then, what is the cause of the various toxic phenotypes manifested by ZnT86D downregulation in neurons? One possibility is that the increased cell death caused by ZnT86D knockdown in the developing neurons or in the neurons of the Drosophila AD model could have been associated with changes in metabolism. Indeed, ZnT7-KO mice show diet-induced glucose intolerance and insulin resistance [26,41], and it is known that ZnT7 deficiency inhibits insulin-dependent Akt activation and glucose uptake [42]. Furthermore, ZnT7 is expressed in the islets of Langerhans in the mouse pancreas, and it has been demonstrated that ZnT7 overexpression in RIN5mf cells (rat insulinoma cells) increased insulin mRNA expression, insulin protein synthesis, and insulin secretion [61]. Insulin has antiapoptotic activity [62], and insulin resistance is known to induce apoptosis [63]. Moreover, it is well known that the glucose is able to affect neurogenesis and hyperglycemia increases the risk of AD [64]. The insulin signaling pathway is conserved in Drosophila: IPCs located in the brain play the role of the pancreas, and DILP secreted from the IPCs play the role of insulin [65]. However, we found that the blood glucose level of ZnT86D RNAi larvae was not significantly different from that of the control group. This implies that brain cell death in our Drosophila AD model was not due to hyperglycemia. Furthermore, it raises the possibility that the function of ZnT86D in Drosophila metabolism is not completely identical to that of its mammalian orthologue. Future studies of ZnT86D function are expected to broaden our understanding of the role of zinc in animal cell metabolism as well as AD.
Another possible scenario is that perturbation of intracellular zinc distribution induced by ZnT86D knockdown affects cellular signaling, and consequently, adversely affects neuronal survival. A recent study on the importance of zinc fluctuations in cellular signaling strongly supports this possibility [66]. This study shows that at the single cell level, changes in intracellular zinc levels occur in response to subtle extracellular perturbations, and these changes directly correlate with changes in ERK and Akt activity [66]. Since ZnT7 is important for the movement of zinc from the cytoplasm to the Golgi [23,24], the cellular distribution of zinc would be changed by ZnT86D knockdown, which may affect the activity of the ERK and Akt signaling pathways. In general, ERK pathway is associated with cell growth, differentiation, and survival [67]. However, chronic ERK activation in pathological conditions is crucial to the pathology of AD and contributes to death signaling [68,69]. Meanwhile, Akt is a PI3K downstream factor that is involved in survival of various cells including neurons [70]. As chronic activation of ERK observed in neurons of AD patients [71], or decreased Akt activity are closely related to neuronal death [72][73][74], the decrease in survival of ZnT86D knockdown neurons in the development process or in stressful situations such as the presence of Aβ could be related to changes in ERK or Akt signaling pathways. The study about the alteration of signaling pathways in ZnT86D RNAi Drosophila is expected to provide important clues elucidating the neuroprotective role of ZnT86D.
Meanwhile, proteins that directly interact with ZnT86D may have been disrupted by ZnT86D downregulation. For example, ZnT7 activates alkaline phosphatases, zincrequiring enzymes that are glycosylphosphatidylinositol-anchored to the cytoplasmic membrane, along with ZnT5 [75]. In addition, ZnT7 binds to CD40 and influences CD154triggered p38 MAPK activity in B-lymphocytes [76]. Therefore, investigating the proteins that interact with ZnT7 will provide insights into this issue.
In conclusion, we demonstrated that ZnT86D plays a pivotal role in neuronal and wing development and in the pathogenesis of AD in Drosophila. When ZnT86D was downregulated in neurons, the embryo-to-adult survival rate, locomotor activity, and lifespan were dramatically reduced. These toxic phenotypes were accompanied by abnormal neurogenesis and neuronal cell death, which were not associated with ER stress and increased blood glucose level. In addition, knockdown of ZnT86D in neurons of a Drosophila AD model increased apoptosis and exacerbated neurodegeneration, but there was no significant change in the degree of Aβ plaque deposition and susceptibility to oxidative stress. Further studies need to be conducted on how the manipulation caused such fatal effects, and this endeavor would provide insights into the design of treatments for neurodevelopmental disorders and AD.

Analysis of Survivability
Embryos from each genotype were collected on grape juice agar plates. Fifty agematched embryos from each group were transferred to a vial containing a standard cornmeal medium and raised at 25 • C. The number of eclosed flies was counted, and the experiment was repeated three times. To measure the survival rate of adult flies, 20 flies of each genotype were kept in a vial at 25 • C. The flies were transferred to fresh media every 3 days, and the number of living flies was counted every 12 h.

Climbing Assay
The climbing assay was conducted as previously described [77] with minor modifications. Ten flies were collected in a vial prepared for the climbing assay and incubated for 1 h at 25 • C for environmental acclimation. After tapping the vial to move the flies to the bottom, individuals climbing to the midpoint of the vial within 10 s were counted. Ten trials were conducted for each vial and repeated four times for each genotype. Therefore, 40 flies were analyzed for each independently derived transgenic line. Climbing scores (the ratio of flies that reached the midpoint of the vial among all flies) were obtained, and the average scores of each group were compared.
To detect cleaved caspase 1 (active DCP-1), the brains of adult flies were dissected in cold PBS and fixed with 4% paraformaldehyde for 30 min. After washing with PBS, samples were incubated with PBST containing 0.1% NGS for 3 h at 25 • C. The brains were then incubated with rabbit anti-cleaved Drosophila DCP-1 antibody [1:200; #9578; Cell Signaling Technology, Danvers, MA, USA] for 2 days at 4 • C. After washing with PBST, the samples were incubated with Alexa-Fluor-555-labeled anti-rabbit IgG secondary antibody (1:200; A-21429; Invitrogen) overnight at 4 • C. After washing, the brains were mounted using Vectashield mounting medium. All the samples were observed under a confocal microscope (LSM800; Carl Zeiss, Oberkochen, Germany).

Acridine Orange Staining
AO staining was performed as previously described [78]. The brains of third instar larvae were dissected in PBS and incubated with 1.6 × 10 −6 M AO (318337; Sigma-Aldrich, St. Louis, MO, USA) for 5 min. The samples were rinsed twice with PBS and observed under a fluorescence microscope (Axiopot2; Carl Zeiss, Oberkochen, Germany).

Thioflavin S Staining
Thioflavin S staining was performed as previously described [79] with minor modifications. Adult flies were fixed in 4% paraformaldehyde containing 0.5% Triton X-100 for 3 h and washed with PBST. Brains were then dissected and incubated in 50% ethanol containing 0.125% thioflavin S (T1892; Sigma-Aldrich) overnight at 4 • C. After incubation in 50% ethanol for 10 min at 25 • C, the samples were washed with PBST and mounted using Vectashield mounting medium. Brains were observed under a confocal microscope (Carl Zeiss).

Histology
The flies were fixed in Carnoy's fixative (ethanol:chloroform:glacial acetic acid at a ratio of 6:3:1) for 4 days at 4 • C and dehydrated with ethanol. The samples were washed with xylene three times for 30 min each and infiltrated with xylene:paraffin solution (1:1) for 1 h; the solution was changed every 20 min at 73 • C. After incubating for 2 h in paraffin changing the solution every 30 min at 73 • C, the brains were embedded and sectioned serially at 5 µm intervals to obtain frontal sections. The sections were stained with hematoxylin and eosin and examined under a light microscope (BX50; Olympus, Tokyo, Japan).

Oxidative Stress Test
The susceptibility of each group to oxidative stress was estimated using hydrogen peroxide (H 2 O 2 ). The flies of each genotype were collected immediately after eclosion and fed RU486 for 20 days. They were then starved for 6 h and transferred to vials with a medium containing 1% H 2 O 2 and 5% sucrose. The number of live flies was counted every 12 h.

Measurement of the Levels of Glucose and Protein in Larval Hemolymph
Quantification of glucose levels in the larval hemolymph was performed as described [80,81]. Seven male larvae of each group were washed twice with water and dried on tissue paper. The cuticles at the mouth part of the larvae were then carefully torn and placed in a perforated 0.5 mL tube within a 1.5 mL tube. The tube was centrifuged at 12,000 rpm for 10 s using a microcentrifuge to obtain hemolymph. The hemolymph was diluted 1:30 in PBS and heat treated at 70 • C for 5 min. Glucose level was measured using the Glucose Assay Kit (GAGO20; Sigma-Aldrich) and glucose concentration was calculated from a standard curve generated with glucose standards. The amounts of glucose were normalized to the protein level measured by detergent compatible (DC) protein assay. The DC Protein Assay Kit (500-0111; Bio-Rad, Hercules, CA, USA) was used to measure the protein level in hemolymph before heat treatment. Protein concentration was calculated from a standard curve generated with the dilutions of BSA standards.