The depletion of ubiquilin in Drosophila melanogaster disturbs neurochemical regulation to drive activity and behavioral deficits

Drosophila melanogaster is a useful and highly tractable model organism for understanding the molecular mechanisms of human diseases. We previously characterized a new dUbqn knockdown model that induces learning-memory and locomotive deficits mediated by impaired proteostasis. Although proteinopathies are the main causes of neurodegenerative diseases, limited information is currently available on the relationship between proteostasis and neurodegenerative-related behavioral perturbations, such as locomotion, wakefulness, and sexual activities. Thus, the present study aimed to elucidate the mechanisms by which dUbqn depletion which is known to cause proteinopathies, affects neurodegenerative-related behavioral perturbations. Pan-neuronal dUbqn-depleted flies showed significantly reduced evening activity along with altered pre- and postsynaptic structural NMJ’s proteins by attenuating signals of Bruchpilot puncta and GluRIIA clustering. In addition, the neurochemical profiles of GABA, glutamate, dopamine, and serotonin were disturbed and these changes also affected courtship behaviors in dUbqn-depleted flies. Collectively, these results extend our understanding on how dUbqn depletion affects neurochemical regulation to drive behavioral disturbances that are generally found in the early stage of neurodegenerative diseases. Moreover, the present study may contribute a novel finding to the design of new agents that prevent disease progression or even treat diseases related to neurodegeneration.

Drosophila melanogaster is a useful and highly tractable model organism for studying and understanding the molecular mechanisms of human diseases due to biologically and physiologically highly conserved pathways between Drosophila and humans. Drosophila has a single homologue of UBQLN in its genome called dUbqn, which shows high similarity to human UBQLN1, 2, and 4 in the major domains as well as functions 13 .
Bedford et al. discovered the regulatory function of UBQLNs in synapses to facilitate GABA-A receptor expression 14 . Moreover, the dysregulation of dUbqn also contributes to defects in postsynaptic growth and synaptic proteostasis 15 . Although previous studies reported a role for UBQLNs in receptor regulation at synapses, its involvement in the control of neurochemical expression remains unclear. Dysfunctional UBQLNs with or without mutations cause ubiquitylated protein accumulation in the early stages of neurodegenerative diseases 16 . At the same stage, declines in motor-cognitive functions and behavioral changes, such as the loss of sexual abilities and sleep-awakening disturbances, are generally observed before the loss of affected neurons 5,[16][17][18] . Therefore, the mechanisms by which dysfunctional UBQLNs affect neurochemical changes to induce motor, cognitive, and behavioral deficits are interesting to investigate in more detail.
Recently, our group took advantage of the novel mass spectrometry imaging (MSI) method to visualize neurochemicals inside the body of Drosophila 19 . In addition, with a newly established dUbqn knockdown model we demonstrated that proteostasis impairments can drive learning-memory and locomotive deficits 20 . Thus, we intend to use the dUbqn knockdown model to gain insights into the changes in more complex behaviors including copulation during proteostasis impairments through investigations on neurochemical regulation.

Drosophila courtship assay. Pairs (virgin male and virgin female) of 3-day adult flies carrying
UAS-dUbqnIR 107-494 /+;elav-GAL4/+, +;elav-GAL4/UAS-dUbqnIR 471-824 and UAS-GFP-IR/+;elav-GAL4/+ were selected under CO 2 anesthesia. Briefly, flies were allowed to recover from anesthesia for at least 1 h, and an aspirator was then used to gently transfer an individual male to the courtship chamber followed by a female 22 . Recordings of behavior were immediately started and conducted for at least 20 minutes.
Larvae tissue preparation. For sample treatment, third instar larvae at similar developmental stages (within 6 hours difference) were picked and dipped in 70% ethanol before embedding. Then, they were embedded in 10% gelatin on base-mold (7 mm × 7 mm × 5 mm; FALMA, Tokyo, Japan) 23,24 . Gelatin powder (Sigma-Aldrich, St. Louis, MO, USA) was dissolved in ultrapure water (Genpure UV-TOC xCAD PLUS, Thermo Fisher Scientific, Waltham, MA, USA) and mixed for 10 min at 42 °C, using a thermomixer (Eppendorf, Hamburg, Germany) until a homogeneous solution was obtained 24 . After aligning the larvae per their body axes in the embedding material, the molds were frozen rapidly in liquid nitrogen. The frozen blocks of gelatin with flies were placed in cryomicrotome (CM1950, Leica, Wetzlar, Germany) at −20 °C for 1 hour 24 . These blocks were then placed on a tissue-holder, fixed with an optimum cutting temperature polymer (Leica, Wetzlar, Germany), and sliced at 30-μm thickness. The tissue sections were thaw-mounted onto Indium-tin-oxide (ITO)-coated glass slides (100 Ω/m 2 without anti-peeling coating, Matsunami Glass, Osaka, Japan) by warming them up by placing a finger on the posterior side of the glass slides 24 . The glass slides were then dried at 40 °C for 5 min and placed in 50-mL conical tubes containing silica gel 24 . On tissue derivatization. 2,4-diphenyl-pyranylium tetrafluoroborate (DPP-TFB) (Sigma-Aldrich, St. Louis, MO, USA) was dissolved in methanol to prepare 10 mg/mL stock solutions 23 . DPP-TFB solutions used for derivatization contained 6 µL of the stock solution, 69 µL of 60% methanol, and 1 µL of triethylamine. After Matrix application. 2,5-dihydroxybenzoic acid (DHB) (Sigma-Aldrich, St. Louis, MO, USA) was applied to tissue sections via spray coating method 24 . DHB solution (50 mg/mL dissolved in 50% methanol) was applied to each section using an airbrush (PS-270, GSI Creos, Tokyo, Japan) 24 . The airbrush was maintained at 10 cm distal from glass slides and an approximately 50 µL DHB solution was sprayed to each section and then completely dried. The spray-dry process was repeated multiple times 24 . Mass spectrometry imaging (MSI) for neurochemicals. MSI was performed using an iMScope TRIO (Shimadzu, Kyoto, Japan). Both optical images and ion distribution were obtained within the same instrument under atmospheric pressure 24 . Nd:YAG laser (λ = 355 nm, 1 kHz) was used as the MALDI laser source, and laser irradiation was repeated 150 times on each data point with a laser power of 45.0 (arbitrary unit in iMScope TRIO) 24 . The voltages of the sample stage and detector were 3.50 kV and 2.10 kV, respectively 24 . A positive ion detection mode was used with an m/z range of 200-400 for detection of neurochemicals. Tandem mass spectrometry (MS/MS) analysis was carried out to visualize DPP-derivatized GABA, glutamate, dopamine, and serotonin with the selected precursor m/z 318, 362, 368, and 391. The ion peak in the product ion mass spectra provided the peak at m/z 232.11, therefore the peak intensity of m/z 232.11 was visualized for neurochemicals imaging. The IMAGEREVEAL (Shimadzu, Kyoto, Japan) was used for the data analyses.

Sample preparation for the Liquid chromatography-tandem mass spectrometry (LC-MS/MS)
analysis. Five third instar larvae at similar developmental stages in 1.5-mL Eppendorf tubes were freeze-dried for 12 h using a freeze dryer (VD-800F, Taitec, Tokyo, Japan). Freeze-dried samples were milled using a ball mill (Retsch, Haan, Germany) at 20 Hz for 1 min. Five hundred microliters of extraction solvent (methanol/water/ chloroform= 5:2:2 v/v/v%) and 20 µL of 1 mM 3,4-dihydroxybenzylamine as an internal standard were added to each tube with the ball-milled sample. The sample was mixed using a vortex and sonicated for 1 minute. After sonication, the sample was centrifuged at 10000×g at 4 °C for 10 minutes. After centrifugation, 400 µL of the supernatant was transferred to a new tube containing 150 µL of water. Then the sample was mixed by vortexing and centrifuged at 10000×g at 4 °C for 10 minutes to separate the polar and non-polar phases. In the GABA, glutamate, and dopamine analysis, the upper polar phase was diluted by 10-fold using water with 0.1% formic acid and syringe filtration (0.2-µm PTFE hydrophilic membrane, Merck) was performed. The filtered solution was transferred to a LC vial. In the serotonin analysis, 300 µL of the upper polar phase was transferred via syringe filtration. The sample was then concentrated by centrifugation for 2 hours and freeze-dried for 12 hours. After dissolving in 30 µL of 30% methanol, the sample was transferred to a LC vial.

LC-MS/MS analytical conditions. The LC-MS/MS analysis was performed using the Nexera UPLC System
(Shimadzu, Kyoto, Japan) connected to LCMS-8050 (Shimadzu, Kyoto, Japan) with the ion source of electrospray ionization in the multiple reaction monitoring (MRM) mode 24 . Chromatographic separation was performed using a PFP column (2.1×100 mm, 3 µm, Shimadzu GLC, Tokyo, Japan) at 40 °C with an injection volume of 1 µL. The mobile phases were water with 0.1% formic acid (eluent A) and acetonitrile with 0.1% formic acid (eluent B) with a gradient program: 0.01 minutes, 2% B; 4.0 minutes, 2% B; 6.0 minutes, 80% B; 7.0 minutes, 100% B; 8.3 minutes, 100% B; 8.31 minutes, 2% B; and 10 minutes, 2% B. The flow rate was 0.3 mL/minute with a total run time of 10 minutes. The ion source was operated in the positive ion mode and interface temperatures were set as follows: nebulizer gas: 3.0 L/minute; heating gas: 10 L/minute; drying gas: 10 L/minute; interface temperature: 300 °C; desolvation line temperature: 250 °C; and heat block temperature: 400 °C. MRM conditions are summarized in Table 1. Data acquisition and processing were performed using LabSolutions (Shimadzu, Kyoto, Japan).
Data analysis. All statistical analyses were performed using GraphPad Prism version 6.02. The Mann-Whitney U test was used to assess the significance of differences between two independent groups 20 . The Kruskal-Wallis test followed by Dunnett's multiple comparison analysis was used to assess the significance of differences between three or more independent groups. All data were expressed as a mean ± standard deviation (S.D.). p < 0.05 was considered statistically significant.

Results
pan-neuronal dUbqn depletion decreases evening activity in D. melanogaster. We previously characterized dUbqn RNAi flies and found that the knockdown of dUbqn in all neurons caused an accumulation of ubiquitylated proteins and affected negative geotaxis in flies 19,20 . The proteostasis impairments caused by dUbqn depletion appeared to negatively affect neuronal homeostasis and associated with neurodegeneration. Although a progressive reduction in the neuronal proteostasis capacity is known to generate a normal neuronal aging, it may result in an occurrence of neurodegenerative diseases. Some of the disease developing patients might suffer from a difficulty in sleeping which certainly affects their circadian rhythm [25][26][27] . Thus, we hypothesized that ubiquilins might be one of the key players in a circadian rhythm alteration and this relationship is worth to investigate. Due to the fact that, Drosophila exhibit daily cycles in their behavioral rhythms that are governed by endogenous circadian rhythms 28 . Therefore, we examined whether the all-day activity or behavioral rhythms of Drosophila are disturbed by the knockdown of dUbqn. Flies carrying UAS-dUbqnIR 107-494 /+;elav-GAL4/+ (dUbqn depletion) and UAS-GFP-IR/+;elav-GAL4/+ (control), 3 days after eclosion were used to monitor daily activities by counting the number of times in 30 minutes that each individual fly crossed an infrared light beam. The dUbqn RNAi lines with different targeted sequences were used to confirm and minimize the possibility of the off-target effects, as demonstrated previously in various assays 19,20 and also in the present study (Figs. S1-S3). Both RNAi lines carrying elav > dUbqnIR 107-494 and elav > dUbqn-IR 471-824 similarly shortened the synapse branch length at NMJ (Fig. S1) and reduced the levels of the dUbqn protein and mRNA to the similar extent (Fig. S2). Control flies showed a regular pattern of activity based on peak activities twice a day between 6.30 to 10.00 am and between 9.00 to 10.00 pm, which are defined as the morning and evening peaks, respectively, and resting activities twice a day between 10.30 am to 6.00 pm and between 10.00 pm to 6.00 am, which are defined as the midday siesta and nighttime sleep, respectively (Fig. 1A).
In comparisons with control flies, dUbqn-depleted flies showed a significant decrease in evening activity (evening peak) in the 1-5 days of activity measurements (Fig. 1A,B), but only a slight reduction in morning activity (Fig. 1A,C). On days 3 and 5, we observed a significant increase in activity in dUbqn-depleted flies during the resting period, such as the midday siesta, and activity at nighttime also appeared to increase (Fig. 1A,D,E). These results suggest that dUbqn-depleted flies exhibited decreased activity, particularly in the evening, which is consistent with our previous findings showing defects in the locomotive activity of dUbqn-depleted flies 20 . Another possibility to explain lower evening activity in dUbqn-deficient flies is that their higher activity during resting (the midday siesta) might affect their activity in the next period of time (evening); however, further studies are needed to clarify this issue. dUbqn plays a role in regulating the presynaptic terminals of the Drosophila NMJ. AZ is an important structure for the exchange of chemicals at synapses in which the release of neurochemicals onto postsynaptic receptors is precisely controlled 29,30 . Alterations in AZs have been observed in various Drosophila disease models, particularly in neurological disorders, such as ALS, FTD, and ASD 21,31,32 . The immunostaining signals of Bruchpilot (Brp), one of the synaptic proteins located at the AZs of the Drosophila NMJs, were attenuated in the Drosophila model of ALS and Pitt-Hopkins syndrome 33,34 . Based on previous findings showing defects in the locomotive activity of dUbqn-depleted flies 19,20 and the present results in which flies showed marked reductions in evening peak activity, we hypothesized that reductions in the locomotive activity of dUbqn-depleted flies may be caused by alterations in AZs at the presynaptic terminals of the NMJ.
The majority of adult fly motor neurons originally develop from larval motor neurons 35 . Therefore, we evaluated the expression of Brp by quantifying the density of Brp-positive puncta on NMJs. The Brp puncta were generally distributed throughout control NMJs ( Fig. 2A-C), but were reduced in dUbqn-depleted NMJs (Fig. 2D-F). Due to the differences in the size and length of NMJs between groups 20 , we decided to quantify the number of Brp puncta by normalization to the related bouton area. We found that the number of Brp puncta was significantly lower in dUbqn-depleted NMJs than in control NMJs (Fig. 2G), suggesting that locomotive defects in dUbqn-depleted flies might be possibly caused by presynaptic alterations.
Theoretically, neurons compensate for changes in excitatory synaptic outputs to maintain a normal synaptic output 36 . Reductions in Brp expression in the Drosophila NMJs may reflect a deficiency in the presynaptic output, and, thus, compensatory processes may take function to optimize a stability of neurotransmission by increasing the highly abundant glutamate receptor (GluR) in the postsynaptic area of the Drosophila NMJs 37 . Vice versa, when GluRIIA receptors are reduced, homeostasis retrospectively responds by increasing presynaptic glutamate release from AZs 38,39 . The distribution of GluR in the NMJ can be monitored by one of its subunits, GluRIIA. Therefore, we examined the level of GluRIIA at NMJs in dUbqn-depleted larvae. The results revealed that postsynaptic GluRIIA signals normalized to the bouton area were markedly increased at the NMJs of dUbqn-depleted larvae (Fig. 3A-G). On the other hand, the GluRIIA cluster area was reduced in dUbqn-depleted larvae, possibly reflecting partial homeostatic compensation in dUbqn depletion (Fig. 3A-F,H). Since the failure of neuronal compensatory mechanisms has been commonly observed in PD, AD, and other neurogenerative diseases, we speculated that the locomotive defects we found in dUbqn-depleted flies were possibly caused by presynaptic alterations together with a failure in synaptic compensation.
A number of neurochemicals were disturbed in whole bodies of dUbqn-depleted flies. Identifying the expression of neurochemicals is critical for interpreting neuronal and behavioral activities [40][41][42] . Based on the present results, we hypothesized that neurochemical dysregulation may be one of the factors affecting the changes observed in sleep-awake activity and locomotive ability in dUbqn-depleted flies. Thus, different kinds of neurochemicals, such as GABA, glutamate, dopamine, and serotonin, have been selected for detection using MSI to examine the distribution patterns of neurochemicals with the combination of LC-MS/MS for the relative quantitation of neurochemical levels.

Scientific RepoRtS |
(2020) 10:5689 | https://doi.org/10.1038/s41598-020-62520-y www.nature.com/scientificreports www.nature.com/scientificreports/ Each of the neurochemicals showed a different pattern of distribution under control and dUbqn-depleted conditions. GABA and glutamate were detected alongside the larval bodies (Figs. 4A,B,D,E and S3A,B,D,E), whereas dopamine generally diffused throughout the body (Figs. 4G,H and S3G,H). Serotonin showed a pattern that differed from the others due to its dispersal near the internal organs of larvae (Figs. 4J,K and S3J,K). When we quantified the signal intensity of each neurochemical, the data revealed that dUbqn-depleted larvae (elav > dUbqn-IR 107-494 ) markedly showed an increase of GABA intensity (Fig. 4C) whereas the other neurochemicals seemed to be declined as comparison to the control (Fig. 4F,I,L). Similarly, we have confirmed the significant decrease in glutamate and dopamine intensity with another dUbqn RNAi line (elav > dUbqn-IR 471-824 ) along with a slightly increase and decrease of GABA and serotonin intensity, respectively compared to the control (Fig. S3A-L).
We have also quantified levels of each neurochemical by using LC-MS/MS. Under dUbqn-depleted conditions, the level of the inhibitory neurochemical, GABA was significantly higher (Fig. 5A) than under control conditions. The levels of excitatory neurochemicals, such as glutamate and dopamine, were significantly decreased (Fig. 5B,C), while those of serotonin were slightly decreased (Fig. 5D). These results suggested that the dysregulation of neurochemicals might be participated in order to reduce the activity in dUbqn-depleted flies.  www.nature.com/scientificreports www.nature.com/scientificreports/ dUbqn-depleted conditions. These results prompted us to examine whether dUbqn-depleted flies have dysfunctions in a fundamental behavior, such as their courtship behaviors.
Pairs of 3-day-old virgin flies were used to perform courtship behaviors. Most of the control and the dUbqn-depleted flies have completed copulation processes (Figs. 6B and S4A). During copulation latency, the male started orientation behaviors by turning toward and trying to catch the female. The male then tapped the female's body using his foreleg. The male made wing vibrations, called the "courtship song", extended his proboscis, and licked the female's genitalia. Courtship was considered to be successful following acceptance by the female to mounting and the male ejaculating his seminal fluid into the female (Fig. 6A). The time after mounting to separation was generally approximately 10-20 minutes 22 and we observed approximately 12-17 minutes of mounting in control and dUbqn-deficient pairs (Fig. 6D). An alteration of either copulation processes such as an extension of copulation latency or a shortening of copulation period might reveal some defects in courtship behaviors. The dUbqn-depleted pairs (elav > dUbqn-IR 107-494 ) had a longer copulation latency of nearly double www.nature.com/scientificreports www.nature.com/scientificreports/ that of control pairs (Fig. 6C). It means that the affected flies required more time to recognize and respond to mating songs. In addition, a shortening of copulation period was also observed with another dUbqn RNAi line (elav > dUbqn-IR 471-824 ) (Fig. S4C). Cumulatively, our results suggested that an alteration in related neurochemical regulations such as dopamine and serotonin might be one of the causes of courtship behavioral defects in dUbqn-depleted flies.

Discussion and conclusion
The present results demonstrated that pan-neuronal dUbqn depletion, which is known to cause proteostasis impairments, reduced evening time activity and altered pre-and postsynaptic structural regulation in Drosophila NMJs. Moreover, the depletion of dUbqn, particularly in neurons, dysregulated neurochemicals, likely resulting in sexual behavioral alterations.  www.nature.com/scientificreports www.nature.com/scientificreports/ We previously reported that dUbqn-depleted flies exhibit locomotive defects and aberrant NMJ macrostructures 20 . The Drosophila activity monitoring assay, which follows the all-day activity of individual flies over many days, showed the daily rhythms of flies driven by endogenous circadian systems 45 . Consistent with our previous findings, we confirmed a reduction in locomotive activities of dUbqn-depleted flies, particularly in the evening peak, whereas normal flies performed peak activities during the same period. Moreover, dUbqn-depleted flies showed slightly increased activities during the resting period for many days, suggesting a disturbance in their circadian rhythms.
Brp is an important synaptic protein located in the AZs of neurotransmitter release sites 46,47 . Alterations in AZs have been found in various types of animal models related to neurological disorders, such as ALS, Pitt-Hopkins syndrome, and ASD 21,33,34 . Since dUbqn-depleted flies showed locomotive defects 20 , reductions in Brp may reflect synaptic defects in dUbqn-depleted flies that already exhibit difficulty with movement.
Homeostatic systems are considered to maintain the stability of neurotransmission. Synapses at the Drosophila NMJ are glutamatergic synapses that contain functional GluRIIA, GluRIIB, and other subunits 38,48,49 . Theoretically, the loss of both GluRIIA and GluRIIC subunits triggers a homeostatic increase in the quantal content to maintain sufficient neuronal activity 38,39,48,50 . On the other hand, the observed increase in GluR numbers in the present study may have been triggered by a compensatory mechanism to counteract reductions in presynaptic glutamate levels. Clustered GluRs are normally large and immobilized, whereas newly forming clusters are observed as the diffused form; therefore, reductions in the GluRIIA cluster area in dUbqn-depleted NMJs may reflect a sign of postsynaptic developmental defects that ultimately affect fly's movements.
Similar to humans, a number of neurochemicals are involved in the regulation of Drosophila activities and behaviors, such as glutamate, GABA, dopamine, and serotonin 44,51,52 . In the present study, we found a decrease in glutamate, an excitatory neurotransmitter, and increase in GABA, an inhibitory neurotransmitter, resulting in less activity, which is consistent with the results obtained in dUbqn-depleted flies. Considering the dopamine which is important for behavioral modulation in Drosophila especially an arousal 53 and courtship behaviors 54 . Previous studies have reported that the dopamine-depleted female flies showed an increase in copulation latency whereas D1 dopamine receptor mutants in male flies found a delay in courtship initiation 55,56 . Moreover, we have found that dUbqn-depleted flies showed an alteration of dopamine and serotonin regulations that are confirmed the impact of the neurochemicals on sexual behaviors observed in our study. The sexual behavioral defects in humans might be caused by both mental and physical mechanisms. However, due to the limitation of the invertebrates. We cannot clearly justify whether the defects caused by mental or physical problems. Considering our experimental design, the courtship parameters that we have chosen such as the copulation latency and the copulation period, persuaded us to investigate the physical state rather than the mental state. However, the mammalian models are needed for further investigation.
Ubiquitylation is a critical mechanism controlling synaptic plasticity and architecture 57,58 , and its dysregulation has been implicated in numerous neurological disorders and psychiatric diseases [59][60][61] . Due to the roles of ubiquilins, which are closely related to ubiquitylation, further studies are warranted to clarify the mechanisms by which ubiquilin regulates synapses as well as neurochemicals. Despite the neurochemicals in both central and peripheral nervous system has been detected in our study, an exploration in the specific area is necessary and meaningful. An accumulation of the neurochemicals in some area outside brains might be caused by the specific release of the neurochemicals from the neurons in the peripheral nervous system. However, we cannot exclude the possibility that they are released in the process of preparing tissues, although the signals could be detected only inside larval bodies. Further analysis is necessary to understand the meaning of these observations. The present study is the first to reveal the role of dUbqn in the regulation of synapses and neurochemicals, and impairments in its function may ultimately affect fundamental activities including locomotive and sexual activities. The dUbqn-deficient model with proteostasis impairments may be an important model for identifying early-stage markers of neurological disorders and developing promising therapies in the future.

Data availability
All of the original data are available on request.