Differential dysregulation of CREB and synaptic genes in transgenic Drosophila melanogaster expressing shaggy (GSK3), TauWT, or Amyloid-beta

Tau, Amyloid-beta (Aβ42), and Glycogen synthase kinase 3 (GSK3) contribute to synaptic dysfunction observed in Alzheimer's disease (AD), the most common form of dementia. In the current study, the effect of pan-neuronal expression of TauWT, Aβ42, or shaggy (orthologue of GSK3) in Drosophila melanogaster was assessed on the locomotor function, ethanol sensitivity, synaptic genes and CREB expression. The effect of TauWT and Aβ42 on the expression of shaggy was also determined. Gene expression analysis was performed using quantitative real-time RT-PCR method. While syt1, SNAP25 and CREB (upstream transcription factor of syt1 and SNAP25) were upregulated in flies expressing TauWT or Aβ42, a prominent decline was observed in those genes in shaggy expressing flies. Although all transgenic flies showed climbing disability and higher sensitivity to ethanol, abnormality in these features was significantly more prominent in transgenic flies expressing shaggy compared to TauWT or Aβ42. Despite a significant upregulation of shaggy transcription in TauWT expressing flies, Aβ42 transgenic flies witnessed no significant changes. TauWT, Aβ42, and shaggy may affect synaptic plasticity through dysregulation of synaptic genes and CREB, independently. However shaggy has more detrimental effect on synaptic genes expression, locomotor ability and sensitivity to ethanol. It is important when it comes to drug discovery. It appears that CREB is a direct effector of changes in synaptic genes expression as they showed similar pattern of alteration and it is likely to be a part of compensatory mechanisms independent of the GSK3/CREB pathway in TauWT or Aβ42 expressing flies.


Introduction
Alzheimer's disease (AD) is one of the most prevalent forms of dementia in the elderly population. It is a progressive disorder that gradually disrupts the function of different brain regions involved in memory formation and cognition. AD is characterized by widespread synaptic and neuronal loss and the generation of two neuropathological lesions, including extracellular amyloid plaques composed of Aβ peptides and intracellular neurofibrillary tangles composed of hyperphosphorylated tau protein [1]. In addition, GSK3β expression is dysregulated in the brain of AD patients, and the evidence confirms its pivotal role in AD pathogenesis [2].
AD is mainly a disorder of synaptic dysfunction referred to as synaptopathy [3]. There is a robust correlation between synaptic loss and cognitive decline in AD, and the dysfunction occurs in the very early stages of the disease [4]. Remarkable evidence has demonstrated that increased phosphorylated Tau, Aβ, and GSK3 levels can lead to synaptic dysfunction and loss [reviewed in 2,4,5]. In addition, these factors are all involved in the dysregulation of various genes in the brain of AD patients and models of Alzheimer's disease [6][7][8][9].
Previous studies revealed that two presynaptic proteins, Synaptotagmin 1 (SYT1) and Synaptosomal-associated protein of 25 kDa (SNAP25), undergo expression alteration at both RNA and protein levels in AD brain patients [10][11][12]. It is noteworthy that there is a close relationship between altered synaptic genes expression and decreased neuronal activity, cognitive impairment, and memory loss during AD pathogenesis [13].
SYT1 is a primary Ca 2+ sensor for triggering synaptic vesicle exocytosis at central nerve system synapses [14]. It is a potential biomarker of synaptic activity [15] and crucial for regulating synaptic plasticity in different brain regions. This presynaptic protein facilitates learning and memory formation, and its altered levels lead to cognition impairment [16].
SNAP25 is one of the critical components of the soluble N-ethylmaleimide-sensitive factor attachment protein receptors (SNARE) complex, which is essential for calcium-dependent exocytosis of the synaptic vesicle [17]. This protein is also involved in the regulation of synaptic plasticity and is a biomarker for synaptic plasticity impairment [18]. SNAP25 is necessary for learning and memory formation and consolidation, neuronal survival, and cognitive function [17].
Cyclic AMP response element-binding protein (CREB) plays a pivotal role in learning and memory because it is a central regulator of memory gene expression, including synaptic genes. CREB is activated upon phosphorylation [19], and it has been shown that total and active CREB (pCREB) is downregulated in the postmortem brain of AD patients confirming that the CREB signaling pathway is disrupted in this disease [20,21].
Although many aspects of mechanisms underlying AD pathogenesis have been discovered, the disease is still incurable and current therapies only slow down the progression of the disease. To achieve an appropriate cure for the disease, discovering various mechanisms involved in AD seems necessary [22].
As a powerful invertebrate model for studying human diseases, especially neurodegenerative diseases, Drosophila melanogaster was applied in this research [23]. To elucidate the pathogenic mechanisms of Tau WT , Aβ 42, and GSK3 on synaptic dysfunction in Alzheimer's disease, we investigated the expression of syt1 and SNAP25 synaptic genes and CREB in transgenic drosophila flies expressing Tau WT , Aβ 42 , or shaggy (orthologue of GSK3), independently. In addition, the effect of Tau WT and Aβ 42 was assessed on the shaggy expression. Behavioral studies were also performed to compare the effect of Tau, Aβ 42 , or shaggy on locomotor function and alcohol sensitivity.

Flies lines and maintenance
Flies were raised on standard oats-agar media at 24 ± 1 °C, at 60-70% relative humidity, and a photoperiod of 12 h. As Aβ 42 is the most toxic form of Amyloid-beta peptides and there are no known mutations in Tau associated with AD, we applied Aβ 42 and Tau WT transgenic flies [24,25]. UAS-Tau WT , UAS-Aβ 42 (Bloomington Stock NO.33769), and UAS-shaggy (Bloomington Stock NO.5255) carry human Tau WT , human Aβ 42, and S9A shaggy (orthologue of GSK3 in Drosophila) gene, respectively. nSyb-gal4 (Bloomington Stock NO.51945), the neuron-specific transcription driver, was applied for the pan-neuronal expression of transgenes. Human Tau expressing flies were a gift from Feany's lab (Harward Medical School, Boston, USA), and all other lines were from the Bloomington drosophila stock center (https:// bdsc. india na. edu/). UAS parental lines were applied as control groups. In all experiments, 15-day-old flies were assessed. RT-PCR was employed using specific primers detecting Tau WT , Aβ 42 [7], or shaggy ( Fig S1) in order to verify the expression of transgenes in each line of flies.

Climbing assay
The climbing assay was performed according to the procedure described by Ali et al. [26]. The detailed description is given in the supplemental experiment procedure.

Alcohol sedation and recovery assay
The alcohol sedation assay was accomplished according to the protocol explained by Maples and Rottenfluh [27]. The detailed procedure is explained in the supplemental experiment procedure.

RNA extraction and cDNA synthesis
RNA extraction was performed from a homogenate of 100 flies' heads using RNXplus kit (Cinnagene). RNA concentration and purity were measured using Nanodrop (Thermo Fisher Scientific), and the integrity was evaluated by 1.5% agarose gel electrophoresis. 2.5 µg of each RNA sample was converted to cDNA using Yekta Tajhiz Azma (YKT) cDNA synthesis kit following the provider's protocol. A separate reaction was performed without RT enzyme (noRT control). Three biological replicates were performed for each transgenic and control flies.

Primer designing and Quantitative Real-time PCR
The exon-exon junction primers or intron-inclusion primers were designed using AlleleID software version 7. Designed primers were specific for target mRNA and did not amplify genomic DNA. The primers detected all known isoforms of target genes.
Real-time PCR reactions were performed with 50 nanogram cDNA from each sample using the SYBR green PCR master mix (ampliqon) on an ABI prism 7500 real-time PCR system (Applied Biosystems). A negative control lacking cDNA was also included. A melt curve verified a single product was generated for each transcript. All Real-time PCR reactions were run in triplicate. Relative gene expression levels of interested genes were calculated after normalizing against endogenous reference gene EF1α by 2 -ΔΔCt method [28]. The list of primers' sequences is available in Supplementary Table S1.

Statistical analysis
The normality test, Shapiro-Wilk, was carried out to ensure the normal distribution of all data. Mean comparison between flies expressing transgene and their controls was performed using independent sample t-test. Comparative analysis between different transgenes was done using oneway analysis of variance followed by Tukey HSD posthoc test for pairwise group comparison.
All statistical analyses were conducted by IBM SPSS statistic software version 21 (SPSS, Chicago, IL, US). The minimum level of significance was set at p-value of 0.05.

Expression of shaggy caused more locomotor deficits compared to human Tau or human Aβ 42
Changes in locomotion happen in very early stages of AD [29]. Locomotor assay has been previously described to evaluate neuronal dysfunction in the fly models of Alzheimer's disease [30]. Therefore, we compared locomotor activity in our transgenic flies. Flies expressing human Tau, human Aβ 42 , or shaggy in neuronal brain displayed a significant reduction in climbing ability (39%, 29%, and 72%, respectively) compared to their control flies (p values < 0.001) (Fig. 1). As it can be seen from Fig. 1, shaggy expressing flies showed the highest reduction compared to the other transgenic flies and there was no significant difference between Tau WT and Aβ 42 expressing flies in reducing locomotor ability (Table S2).

More sensitivity to ethanol was observed in flies expressing shaggy compared to flies expressing human wild-type Tau or human Aβ 42
The genes involved in alcohol metabolizing pathway are shared between flies and mammals [31]. Moreover, there is an overlapping between genes contributing to alcohol use disorder and Alzheimer's disease [32]. Interestingly, recent studies showed similarity between genes involved in memory formation and ethanol-related behaviors in Drosophila melanogaster [reviewed in 33]. So we investigated the influence of overexpression of Tau WT , Aβ 42 , or shaggy on ethanol sensitivity using ethanol sedation assay. The ST50 of flies expressing Tau WT , Aβ 42 , or shaggy was significantly lower (37%, 30%, and 57%, respectively) than the ST50 of control genotypes (p values < 0.001) (Fig. 2a), while these transgenic flies showed a significant elevated RC50 value (36%, 27%, and 67%, respectively) compared to their control groups (P-value < 0.001, p-value < 0.01, p-value < 0.001) (Fig. 2b). Interestingly, shaggy expressing flies had more impact on the sensitivity to ethanol compared to the other two transgenic groups and Tau WT and Aβ 42 transgenic flies did not indicate any remarkable difference in increasing sensitivity to ethanol (Table S3 and S4).

While wild-type Tau and Aβ 42 upregulated the expression of syt1 and SNAP25, shaggy overexpression downregulated these synaptic genes
To elaborate on the effect of Tau WT , Aβ 42, or shaggy on synaptic dysfunction in AD, we measured syt1 and SNAP25 mRNA levels in 15-day-old adult flies expressing Tau WT , Aβ 42 , or shaggy in the central nerve system (CNS) neurons using relative real-time RT-PCR.

CREB was differently dysregulated in shaggy overexpressing flies compared to wild-type Tau or Aβ 42 transgenic flies
In the next step, we examined the expression of CREB, the upstream transcription factor of syt1 and SNAP25 in our transgenic flies [34]. We perceived a striking increase of 128% in the expression of CREB in Tau WT transgenic flies compared to their control (Fig. 4) (p-value < 0.01). Similarly, with regard to Aβ 42 expressing flies, CREB expression was notably elevated by 114% compared to their counterpart control (Fig. 4) (p-value < 0.001). By contrast, shaggy transgenic flies witnessed a plummet of 86% in mRNA level of CREB in comparison to the control group. (Fig. 4) (p-value < 0.01). No remarkable difference was spotted between Tau WT and Aβ 42 expressing flies concerning CREB mRNA level (Table S5).

While wild-type Tau upregulated shaggy expression, no change was observed in shaggy levels in Aβ 42 transgenic flies
As pointed out before, GSK3 has a detrimental effect on synaptic plasticity and is dysregulates in AD [2,35]. Also,  GSK3 is a regulator of CREB activity [36]; therefore, we determined the mRNA level of shaggy in Tau WT or Aβ 42 transgenic flies to investigate if the change in CREB expression is through GSK3/CREB pathway or not. As it can be seen from Fig. 5, there was an increase in transcription levels of shaggy in Tau WT expressing flies by 28% in comparison to the control group (p-value < 0.05). However, we could not detect any significant difference in shaggy levels between Aβ 42 transgenic flies and their control.

Discussion
In the current research, first, the effect of TauWT, Aβ42, or shaggy was examined on the locomotor activity and ethanol sensitivity. Then, the impact of these transgenes was assessed on the expression of synaptic genes, CREB, and shaggy. We expressed Tau WT , Aβ 42, or shaggy in Drosophila melanogaster using nSyb-Gal4 driver, which drives protein expression only in adult neurons. Therefore, the developmental effects of transgenes were avoided (http:// flyba se. org/ repor ts/ FBtp0 041245. html).
Locomotor activity correlates with the efficiency of synaptic transmission in flies [37]. Previous studies highlighted that a number of synaptic proteins are involved in ethanol responses such as ethanol sensitivity, tolerance and preference [reviewed in 38]. In the present study, all 15-day-old transgenic flies exhibited a reduction in locomotor ability and a surge in sensitivity to ethanol. However shaggy revealed a higher impact, which may implicate that shaggy has a more adverse effect on the efficiency of neurotransmission and synaptic proteins.
As mentioned earlier, Tau WT , Aβ 42, and GSK3β play crucial roles in synaptic dysfunction [reviewed in 2,4,5]. Moreover, changes in synaptic proteins, SYT1 and SNAP25, have been observed in AD patients' brains [10][11][12]. Therefore, in the current study, the effect of Tau WT , Aβ 42, or shaggy overexpression in neurons of Drosophila was investigated on syt1 and SNAP25 mRNA expression. The data obtained in our research exhibited that Tau WT and Aβ 42 significantly upregulated syt1 and SNAP25 transcriptions. This result was in accordance with the previous results showing that the expression of genes involved in synaptic vesicle trafficking and neurotransmitter release, including SYT1 and SNAP25, was elevated in individuals with mild cognitive impairment (MCI) compared to age-matched controls or AD patients [10,11] Our 15-day-old flies are also probably at the early stages of AD, as it was reported by Sofola et al. [39].
The increase in syt1 and SNAP25 expression might reflect a compensatory response to synaptic defects and loss. Synaptic dysfunction and loss occur in the very early stages of AD. In MCI, the remaining presynaptic terminals are sprouted and expanded to compensate for the absence of lost synapses [40].
The increased expression of genes involved in synaptic vesicle trafficking and release presumably increases neurotransmitters' release in the brain. The FMRI results confirm that the brain activity has increased during the MCI stage [41]. Although increasing the release of neurotransmitters is expected to improve cognitive and perceptual functions, neuronal hyperactivity causes excitotoxicity leading to a number of devastating consequences such as generation of ROS and oxidative stress, and apoptosis which finally causes memory impairment [42]. It can be suggested that the neuronal cells try to compensate for the effects of these pathological lesions in Tau WT or Aβ 42 transgenic flies by increasing synaptic genes expression (syt1 and SNAP25) 15 days after eclosion. However, it is likely to impair synaptic plasticity and networks and ultimately leads to inability to record new information. Defect in memory function observed in 15 days old Tau WT or Aβ 42 expressing flies in our previous study [7] and deficit in locomotor ability and less sensitivity to ethanol in the current study may testify that the presumed compensatory mechanisms were inefficient.
Next, we assessed the relative expression of CREB in the brain of transgenic flies to find out if the observed changes in synaptic genes expression can reflect this transcription factor dysregulation. It has been already shown that CREB positively regulates the expression of genes involved in memory consolidation such as SYT1 and SNAP25 [19,34], and has been dysregulated in AD [20,21].
According to our results, pan-neuronal expression of Tau WT or Aβ 42 in nerve cells dramatically increased the level of CREB mRNA, which may suggest that CREB is likely a part of compensatory mechanisms in our flies. Previous studies reported the reduction of CREB expression in cell culture treated with Aβ 42 [21,43] which is in contrast with the current results. In addition, investigation of the total and activated form of CREB protein in AD postmortem brain at late stages of the disease revealed a reduction [20,21]. However, to the best of current authors' knowledge, no studies at the early stages or at transcription level of CREB have been yet conducted.
The mechanism underlying gene expression dysregulation in flies expressing shaggy seemed different. CREB expression and also syt1 and SNAP25 showed downregulation in shaggy transgenic flies. Intriguingly, it has been demonstrated that neuronal overexpression of shaggy leads to a decrease in the number and size of synapses and presynaptic terminals and a decline in neurotransmitter release [44][45][46]. Our finding is in agreement with earlier research indicated that there is a decrease in transcription of genes involved in exocytosis of neurotransmitters like SYT1 and SNAP25 in AD patients compared to agematched control or MCI [10,11]. Furthermore, there is a decrease in the protein level of CREB in the postmortem brain of individuals suffering from AD at later stages [20,21]. Moreover, it has been demonstrated that overexpression of GSK3β in neurons of mice results in downregulation of Syt1 transcription in the hippocampus [6].
Phosphorylation of CREB on serine 129 by GSK3 leads to its inactivation [36]. As CREB is a self-regulator, its inactivation can lead to a decrease in its mRNA level [19]. Therefore, it can be suggested that shaggy, likely through the inactivation and downregulation of CREB in flies' brains, leads to a decrease in the target genes of this transcription factor.
At last, to examine whether the change in CREB transcript level is through GSK3/CREB pathway, we spotted the expression of shaggy in Tau WT or Aβ 42 transgenic flies. The current study showed that ectopic expression of Tau WT in the drosophila nerve system upregulated transcription of shaggy, which is consistent with the finding of a preceding study demonstrating that mRNA level of GSK3β has been elevated in the hippocampus at the early stages of Alzheimer's disease [35]. However, we observed that Aβ 42 did not affect shaggy expression. Our data proposed that the upregulation of CREB as a part of the compensatory mechanism in Tau WT or Aβ 42 transgenic flies is probably independent of GSK3/CREB pathway.
It has been exhibited that GSK3β can trigger abnormal hyperphosphorylation and aggregation of Tau WT [2], but there is no study on the effect of Tau WT on shaggy levels or activation. So we are the first to report that the expression of Tau WT may affect the expression of shaggy, GSK3β orthologue. It would be more interesting to investigate the impact of Tau WT on protein level and activity of GSK3β, as well. In spite of the upregulation of shaggy in Tau WT expressing flies, there was an increase in synaptic genes expression, which may put forward that in the early stages of the illness, the compensatory mechanisms are more powerful than the effect of GSK3.
Increased production of Aβ 42 and higher activation of GSK3β are also in a vicious circle [2]. 15-day-old flies that expressed Arctic Aβ 42 only in adult neurons revealed a reduction in the levels of inhibitory ser9 phosphorylation of shaggy, leading to an increase in its activity [39]. As we did not observe any changes in the mRNA levels of shaggy in Aβ 42 expressing flies, it seems that Aβ 42 probably dysregulates GSK3β activity without affecting its transcriptional levels. Further investigation is required to verify the mechanisms involved in GSK3 dysregulation in AD.
To sum up, here we have shown, for the first time, that Tau WT , Aβ 42 , and shaggy differentially and independently alter the expression of synaptic genes (syt1 and SNAP25) and transcription factor of CREB in transgenic flies at an early time point of their life cycle. While both Tau WT and Aβ 42 had similar effects on synaptic gene expression, shaggy had a different impact on the expression of those genes. It seems that dysregulation of synaptic genes occurs as a consequence of changes in CREB expression and is likely independent of GSK3/CREB pathway. While Tau WT increased the levels of shaggy, no change was observed in Aβ 42 expressing flies.
In addition, shaggy has a more significant impact on ethanol sensitivity and motor dysfunction in flies. This could be due to more dramatic effect of shaggy on synaptic dysfunction compared to Tau WT and Aβ 42 , as we observed in the current research.
With regard to the result of this study, it appears that GSK3 has more adverse effect on Alzheimer's disease's symptoms such as locomotor and synaptic dysfunction. It seems that using GSK3 inhibitors in treating Alzheimer disease or improving patient's quality of life, accompanied by drugs that target Tau and Amyloid beta, would be essential.