- Email: [email protected]
Novel genetic link between the ATP-binding cassette subfamily A gene and hippo gene in Drosophila
Journal Pre-proof Novel genetic link between the ATP-binding cassette subfamily A gene and hippo gene in Drosophila Ibuki Ueoka, Akari Takai, Mizuki Yamaguchi, Tomohiro Chiyonobu, Hideki Yoshida, Masamitsu Yamaguchi PII:
S0014-4827(19)30616-0
DOI:
https://doi.org/10.1016/j.yexcr.2019.111733
Reference:
YEXCR 111733
To appear in:
Experimental Cell Research
Received Date: 12 June 2019 Revised Date:
16 October 2019
Accepted Date: 16 November 2019
Please cite this article as: I. Ueoka, A. Takai, M. Yamaguchi, T. Chiyonobu, H. Yoshida, M. Yamaguchi, Novel genetic link between the ATP-binding cassette subfamily A gene and hippo gene in Drosophila, Experimental Cell Research (2019), doi: https://doi.org/10.1016/j.yexcr.2019.111733. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc.
Novel genetic link between the ATP-binding cassette subfamily A gene and hippo gene in Drosophila
Ibuki Ueoka1, 2, Akari Takai3, Mizuki Yamaguchi1, 2, Tomohiro Chiyonobu3, Hideki Yoshida1, 2, *, and Masamitsu Yamaguchi1, 2, *
1
Department of Applied Biology and 2Advanced Insect Research Promotion Center,
Kyoto Institute of Technology, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan. 3
Department of Pediatrics, Kyoto Prefectural University of Medicine, Graduate School
of Medical Science, 465 Kaji-cho, Hirokoji, Kamigyo-ku, Kyoto 602-8566, Japan.
* Co-correspondence: M. Yamaguchi, Department of Applied Biology, Kyoto Institute of Technology, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan. Tel.: +81-75-724-7781; Fax: +81-75-724-7799; E-mail: [email protected] H. Yoshida, Department of Applied Biology, Kyoto Institute of Technology, Matsugasaki, Sakyo-ku, Kyoto, 606-8585, Japan. Tel.: +81-75-724-7787; Fax: +81-75-724-7787; E-mail: [email protected]
1
Abstract The pan-neuron-specific knockdown of dABCA, a Drosophila homologue of the human ATP binding cassette subfamily A member 13 gene, increases social space without affecting climbing ability and induces the early onset of evening activity in adult flies followed by relatively high activity throughout the day. Satellite bouton numbers in the presynaptic terminals of motor neurons are increased in dABCA knockdown flies. In the present study, we further characterized pan-neuron-specific dABCA knockdown flies and found that active zones in the presynaptic terminals of motor neurons increased, whereas learning abilities decreased in larvae. Genetic crossing experiments revealed that the hippo mutation enhanced the hyperactivity phenotype of adults, but suppressed the increased satellite bouton phenotype induced by the dABCA knockdown. Drosophila ABCA is predicted to transport lipid molecules and impair the asymmetric distribution of phospholipids across the plasma membrane, and these local changes are considered to be important for various cellular functions. The disruption of lipid homeostasis in central and peripheral nervous systems by the dABCA knockdown may affect the Hippo-related signaling pathway in order to induce the observed phenotypes. Keywords: autism spectrum disorder; ATP-binding cassette protein A; Drosophila; neuron; hippo gene; activity monitoring Abbreviations: ABCA, ATP-binding cassette subfamily A; ASD, autism spectrum disorder; CNS, central nervous system; NMJ, neuromuscular junction; hpo, hippo
2
1. Introduction Human adenosine triphosphate (ATP)-binding cassette subfamily A member 13 (ABCA13), a member of the ABCA protein family, is predicted to have the typical structure of full-size ABC proteins, including two transmembrane domains (TMDs), each containing six transmembrane α helices, and two nucleotide-binding domains (NBDs), with the latter containing characteristic ATP-binding Walker A and B motifs. Similar to other ABCA protein family members, ABCA13 is predicted to transport lipid molecules [1]. Mutations in the human ABCA13 gene are associated with autism spectrum disorder (ASD) [2-4] as well as schizophrenia and bipolar disorder [1,5]. Recent studies showed that a monkey carrying the heterozygous ABCA13 deletion and a mutation in the serotonin 2C receptor (5HT2c) displayed an impaired ability to monitor the behavior of others in addition to repetitive behavior, which are frequently associated with ASD [6]. This monkey model of ASD with the ABCA13 deletion and a mutation in 5HT2c may have impaired neural maturation in the central nervous system (CNS) [7]. Drosophila has a homologue for human ABCA family genes called dABCA (CG1718), which shows the highest homology to the human ABCA13 gene. The pan-neuron-specific knockdown of dABCA resulted in increased social space with the closest neighbor in adult male flies, but did not affect climbing ability, indicating that the increase in social space is not due to a defect in climbing ability [8]. An activity assay with adult male flies revealed that the knockdown of dABCA in all neurons induced the early onset of evening activity in adult flies followed by relatively high activity during morning peaks, evening peaks, and midday siestas [8]. These phenotypes are similar to the defects observed in human ASD patients, suggesting that the established dABCA knockdown flies are a promising model for ASD. Furthermore, satellite bouton numbers were increased in the presynaptic terminals of motor neurons at the neuromuscular junction (NMJ) in dABCA knockdown third instar larvae, suggesting that dABCA regulates the formation and/or maintenance of the presynaptic terminals of motor neurons [8]. The Hippo pathway is a tumor-suppressive signaling cascade that was originally found in Drosophila. In this pathway, Hippo (Hpo) phosphorylates Warts (Wts) with Salvador (Sav) and Mats, and phosphorylated Wts then phosphorylates Yorkie (Yki), a transcriptional coactivator. This pathway limits cell proliferation by inducing apoptosis and cell cycle arrest [9]. In addition to the role of the Hippo pathway in the regulation of 3
cell proliferation, recent studies reported that the Strip-Hippo pathway is involved in synaptic development and this regulation is independent of Yki [10]. The specific knockdown of strip in motor neurons increased satellite bouton numbers at the NMJ, which is similar to that observed with the knockdown of dABCA [8,10]. This satellite bouton phenotype induced by the knockdown of strip was effectively suppressed by a hpo mutation [10]. More recent studies have implicated hpo in some neurodegenerative diseases, such as amyotrophic lateral sclerosis and Charcot–Marie–Tooth disease [11,12]. In the present study, we therefore investigated whether a genetic interaction exists between dABCA and hpo. The hpo mutation enhanced the hyperactivity phenotype of adults, but suppressed the increased satellite bouton phenotype induced by the knockdown of dABCA. The disruption of lipid homeostasis in central and peripheral nervous systems by the knockdown of dABCA may explain the observed phenotypes that appear to be mediated by some Hippo-related signaling pathways. 2. Materials and methods 2.1. Fly stocks Flies were raised on standard food (0.65% agar, 10% glucose, 4% dry yeast, 5% cone flour, and 3% rice bran) under a 12-h light-dark cycle at 25℃. The following fly stocks obtained from the Bloomington Drosophila Stock Center were used in the present study: UAS-dABCA-IR666-672 (HMS01796, BL38329) targeted to the region corresponding to amino acid residues (aa) 666-672 of dABCA, UAS-strip-IR (HMS01134, BL34657), UAS-ena-IR (HMS01953, BL39034), UAS-GFP-IR (BL9331), elav-Gal4 (third chromosome insertion, BL8760), hpoKS240 (BL25085), hpoKC202 (BL25090), wtsX1 (BL44251), and ykiB5 (BL36290). UAS-dABCA-IR997-1106 (GD3708, VDRC44449) targeted to the region corresponding to aa997-1106 of dABCA, UAS-rho-IR (GD2243, VDRC51952), and UAS-rho-IR (GD2243, VDRC51953) were obtained from the Vienna Drosophila RNAi Center (VDRC). To minimize the effects of the genetic background, the flies used in the present study were backcrossed 6 times with the w strain. 2.2. Odor-taste learning assay for larvae The Pavlovian-type larval learning assay based on odor-taste was performed with some modifications [13,14]. A group of larvae was exposed to n-amyl acetate (AM) (Merck) 4
in the presence of a reward (2 M sucrose, SUC) and this was followed by 1-octanol (OCT) (Nacalai) in the absence of SUC. This training was defined as AM+/OCT, indicating “+” as the reward. In the next step, reciprocal training was performed. A group of larvae was sequentially exposed to OCT in the presence of SUC that was followed by AM exposure in the absence of SUC. The second training step was defined as OCT+/AM. Five minutes of each training allowed flies to establish an association between AM or OCT and the reward. This set of training was repeated three times. Ten microliters of each odorant was added to a 0.2-mL Eppendorf tube with a perforated lid and placed on the opposite site inside the proper Petri dish filled with 1% agarose. AM was provided undiluted, while OCT was used at a 1:75 dilution with liquid paraffin. Two Petri dishes were used for the training period, one for each couple of odorants. Each dish was divided into two distinct zones with a 1-cm so-called neutral zone arranged in the middle. The neutral zone was defined as the area at which larvae receive the same intensity of each odorant. Twenty-four larvae for each corresponding training (AM +/OCT and OCT+/AM) were divided into three groups, and each group of 8 received training in the appropriate dish. After training, larvae were tested by exposure to AM and OCT in the absence of SUC. The odorants were deposited on the opposite ends of each dish and larvae were transferred into a dish in groups of 8 after the training period. The resulting preference indexes were used to calculate the learning index (LI). The AM preference was calculated as: (number of larvae on the AM side-number of larvae on the OCT side)/ (total number of larvae on both sides). The AM preference ranged between 1 (perfect attraction to AM) and −1 (perfect attraction to OCT). The normalized AM LI was calculated as (AM preference-Average of OCT preference)/2. The normalized OCT LI was calculated as (Average AM preference-OCT preference)/2. LI was calculated as (Normalized AM + Normalized OCT)/2. 2.3. Crawling assay for larvae A crawling assay was performed as described previously [15]. Flies were cultured at 28°C. After laying eggs for 24 hours, adult flies were removed from the food vial. Larvae (20~30 larvae per vial) were cultured under not overcrowded conditions. At the beginning of the wandering stage, only third instar larvae that were wandering out of the food were transferred into a 15-cm Petri dish containing 2% agarose at a density of 4 larvae per plate. A 1-min video was acquired and analyzed using ImageJ software with 5
the wrMTrck plug-in [16]. 2.4. Immunostaining An immunohistochemical analysis of the CNS tissues of larvae and adults was performed as described previously [14]. CNS tissues from third instar male larvae or 1-day-old adult male flies were dissected in phosphate-buffered saline (PBS) and fixed in 4% paraformaldehyde in PBS at 25°C for 20 min. After washing with 0.3% Triton X-100 in PBS (PBST), samples were blocked in 0.3% PBST with 2% normal goat serum (NGS) at 25°C for 30 min. An incubation with the primary antibody was performed in 0.15% PBST with 10% NGS at 4°C for 16 h. The anti-Fasciclin II (Fas II) antibody (mouse, 1:20 dilution, DSHB 1D4) was used as a primary antibody. After washing with 0.3% PBST, the secondary antibody incubation was conducted in 0.15% PBST with 10% NGS at 25°C for 2 h. Alexa Fluor 488 conjugated anti-mouse IgG (Invitrogen, all were used at a 1:400 dilution) was used as a secondary antibody. After washing with 0.3% PBST, samples were mounted with Vectashield (Vector Laboratories). In the immunohistochemical analysis of the NMJ, third instar male larvae were dissected in HL3 saline and fixed in 4% paraformaldehyde in PBS at 25°C for 30 min. After washing with 0.3% PBST, samples were blocked in 0.15% PBST with 10% NGS at 25°C for 30 min. The primary antibody incubation was performed in 0.15% PBST with 10% NGS at 4°C for 16 h. The following primary antibodies were used: anti-Discs large (Dlg) (mouse, 1:200 dilution, DSHB 4F3), anti-Bruchpilot (BRP) (mouse, 1:200 dilution, DSHB nc82), and anti-GluRIIA (mouse, 1:80 dilution, DSHB 8B4D2) IgGs. FITC-conjugated
goat
anti-horseradish
peroxidase
(HRP)
IgG
(1:400,
MP
Biochemicals) was used to visualize neurons. After washing with 0.3% PBST, the incubation with the secondary antibody and FITC-conjugated anti-HRP IgG was performed in 0.15% PBST with 10% NGS at 25°C for 2 h. The following secondary antibodies were used (all from Invitrogen and at a 1:400 dilution): Alexa Fluor 594 conjugated anti-mouse IgG and Alexa Fluor 464 conjugated anti-mouse IgG. After washing with 0.3% PBST, samples were mounted with Vectashield (Vector Laboratories) for inspection by a confocal laser scanning microscope (Olympus Fluoview FV10i) or ProLong Diamond (Invitrogen) for inspection by a super resolution microscope (Nikon, N-SIM). Images of the NMJ at muscle 4 of the A3 and A4 6
segments were acquired and processed with ImageJ software. 2.5. Drosophila activity assay Flies were reared for more than several generations at 25°C under a 12-h light-dark cycle. Adult male flies were selected from each group to be analyzed in the Drosophila Activity Monitor (DAM, from Trikinetics) under a 12-h light-dark cycle equipped in a 25°C incubator [8]. One-day-old virgin male or female flies were individually introduced into the DAM glass tube. Fly activity was recorded for at least 4 consecutive days. The activity recorded for each 15-min bin was based on how many times per 15 min each individual fly crossed an infrared light beam that bisected the capillary tube and was perpendicular to the tube. The measured values were plotted in the graph at the endpoint of the 15-min measurement as the average activity bouts of the flies. 2.6. Statistical analysis Statistical analyses were performed using GraphPad Prism 6.0 software and Excel 2016. The criteria for significance were: ns (not significant, p> 0.05), *p< 0.05, **p < 0.01, ***p<0.001, ****p<0.00001. The normality of all data was analyzed with the Shapiro-Wilk normality test and visualization of histograms. Data following a normal distribution were analyzed with a parametric test (one-way ANOVA or two-way ANOVA followed by Dunnett’s or Bonferroni’s post hoc test). Some data followed a non-normal distribution because a number of outliers existed. Data following a non-normal distribution were analyzed with non-parametric tests (the Kruskal-Wallis test or Friedman test followed by Dunn’s post hoc test). The graph representation, error bars, box segments and whiskers are defined in each figure legend, together with which statistical test was performed. 3. Results 3.1. dABCA knockdown in pan neurons exerted only a marginal effect on larval locomotive ability
We previously reported that defects in the social interactions of the dABCA knockdown fly were more severe in male flies than in female flies [8]. Furthermore, in activity monitoring, we obtained more consistent data in male flies than in female flies (Supplementary Fig. S1). Therefore, we used male flies in the present study. The 7
pan-neuron-specific knockdown of dABCA exerted no apparent effect on the climbing ability of adult flies [8]. To investigate the effects of the dABCA knockdown on locomotive ability in more detail, we performed a larval crawling assay (Fig. 1). We used two independent RNAi lines targeted to different regions of dABCA mRNA. The effective knockdown of dABCA in these RNAi lines was previously demonstrated by Western immunoblot analyses with male adult head extracts [8]. Previous studies also revealed that these RNAi lines showed a similar defect in social activity [8]. The crawling speed (mm/sec) of elav>dABCA-IR997-1106 was slightly slower than that of control elav>GFP-IR flies, while other dABCA knockdown flies including elav>dABCA-IR997-1106 2copy did not show any significant difference in crawling speed from control flies (Fig. 1C). The distance from start to finish (mm) in a 1-min test of elav>dABCA-IR997-1106 and elav>dABCA-IR997-1106 was slightly shorter than that of control flies, while other dABCA knockdown flies including elav>dABCA-IR997-1106 2copy did not show significant differences in distance from start to finish from the control (Fig. 1A, D). These results indicate that the knockdown of dABCA exerts only a marginal effect on larval locomotive ability. 3.2. dABCA knockdown in pan neurons impaired larval learning ability ASD is sometimes accompanied by a defect in learning ability [17]. To investigate the role of dABCA in complex neuronal functions, we performed a Drosophila larval olfactory learning assay (Fig. 2A) [13]. Control elav>GFP-IR larvae showed the ability to make the correct association between an odorant and the reward because their preference to the reward odorant was stronger than that to the no reward odorant in accordance with training (Fig. 2B). Control larvae after training in the presence of AM and SUC (AM+) exhibited the ability to recognize the reward odorant in the presence of both odorants (AM+/ OCT), while after training in the presence of OCT and SUC (OCT+), they showed a stronger preference for OCT in the presence of both odorants (OCT+/AM) (Fig. 2B). However, dABCA knockdown larvae did not show a significantly higher number of preferred AM in AM+/OCT than in OCT+/AM (Fig. 2B). LI showed that the pan-neuron-specific dABCA knockdown resulted in a reduction in learning abilities (Fig. 2C). The quantification of data revealed that the LI of elav>dABCA-IR666-672, elav>dABCA-IR997-1106, and elav>dABCA-IR997-1106 2copy flies were 23, 79, and 115% lower, respectively, than that of control elav>GFP-IR flies (Fig. 8
2C). Mushroom bodies (MBs) are the centers for olfactory associative learning and elementary cognitive functions in the Drosophila brain, and aberrant morphologies have been observed with other model flies for neurodegenerative diseases [14]. Therefore, we used the antibody to the cell adhesion molecule Fasciclin II (FASII), which is a well-established MB marker [18], to perform the immunostaining of larval brain lobes and the adult brain (Fig. 2D). Control and elav>dABCA-IR997-1106 2copy flies showed a symmetrical distribution of MBs between the two larval brain lobes (thin white dashed line) and adult brains with a sharp separation of the midline and a normal arrangement of structures (Fig. 2D). Thus, there was no detectable difference in the appearance of MB structures between control and dABCA knockdown larvae (Fig. 2D). 3.3. dABCA knockdown in pan neurons increased the number or density of active zones at the NMJ We previously reported that the number of satellite boutons at synapses in the NMJ increased in dABCA knockdown flies, whereas the number of synapse branches, the total branch length of synapses, and mature bouton numbers were not affected by the dABCA knockdown [8]. Drosophila NMJ synapses are glutamatergic and similar to those in the vertebrate CNS [19]. Boutons contain multiple active zones that are neurotransmitter release sites, and each of these apposes a glutamate receptor (GluR) cluster [19]. The presynaptic boutons at larval NMJs are eventually surrounded by a subsynaptic reticulum (SSR), which contains neurotransmitter receptors, scaffolding proteins, and postsynaptic signaling complexes [20]. This glutamatergic NMJ is known to contain five AMPA-like GluR subunits, GluRIIA to E. Therefore, we used the cell adhesion molecule Brp as an active zone marker and GluRIIA to monitor the distribution of GluR in the NMJ (Fig. 3). The BRP-positive area (µm2)/NMJ area (µm2) in elav>dABCA-IR666-672 flies was significantly larger (by 1.63-fold) than that in control elav>GFP-IR flies (Fig. 3C), while those in elav>dABCA-IR997-1106 and elav>dABCA-IR997-11062copy flies were slightly
larger
(Fig.
3C).
The
BRP-positive
area/NMJ
area
(µm2)
in
elav>dABCA-IR997-1106 and elav>dABCA-IR997-11062copy flies were significantly larger than that in control flies (Fig. 3D). These results indicate that the dABCA knockdown increases the number or density of active zones at the NMJ. Moreover, we found that 9
the GluRIIA-positive area (µm2)/NMJ area (µm2) in dABCA RNAi flies appeared to be slightly increased (Fig. 3E). 3.4. Loss-of-function mutation in hpo enhanced the hyperactivity phenotype induced by the pan-neuron-specific dABCA knockdown In analyses of adult flies with DAM (from Trikinetics), we previously reported that the pan-neuron-specific dABCA knockdown induced the early onset of evening activity followed by relatively high activity throughout the day [8]. The circadian wake and sleep rhythm itself established in a 12-h light-dark cycle did not appear to be affected by the dABCA knockdown because circadian rhythmicity was maintained in constant darkness (data not shown). In order to identify the genes that genetically interact with the dABCA gene, we focused on the hyperactivity phenotype of pan-neuron-specific dABCA knockdown flies that may be easily monitored by the Drosophila activity assay. In the activity monitoring assay, we used one-day-old virgin male flies and showed five-day average activity (Fig. 4) because dABCA-knockdown flies show hyperactivity on each day from 1 to 5 days (Supplementary Fig. S1). We performed a genetic test by crossing dABCA knockdown flies and the mutant flies of hpo, a candidate gene, as described in the Introduction section. We also crossed dABCA knockdown flies and the knockdown flies of rhomboid (rho), the epidermal growth factor-receptor (EGFR) pathway-related gene, as a negative control [21-24]. The knockdown of rho did not affect the hyperactivity phenotype induced by the pan-neuron-specific dABCA knockdown, suggesting the lack of a genetic interaction between dABCA and rho (Fig. 4C). In contrast, the strong enhancement in hyperactivity induced by the knockdown of dABCA was observed by crossing with the loss-of-function mutant, hpoKC202 (Fig. 4A, D), while the hpoKC202 mutation alone slightly enhanced this activity (Fig. 4B, E). Enhanced hyperactivity was also observed by crossing with hpoKS240 (Fig. 4A, D), while the hpoKS240 mutation alone suppressed this activity (Fig. 4B, E). These results suggest that dABCA genetically interacts with hpo. 3.5. Loss-of-function mutation in hpo suppressed the increased satellite bouton number phenotype induced by the pan-neuron-specific dABCA knockdown We also examined the effects of the hpo mutation on the aberrant morphology of larval 10
NMJs induced by the dABCA knockdown. The increased number of satellite boutons induced by the dABCA knockdown was effectively suppressed by the loss-of-function mutation in hpoKC202 (Fig. 5A, 5B). The other loss-of-function mutation in hpoKS240 also showed slight suppression (Fig. 5A, 5B). A half-dose reduction in hpo itself did not affect the number of satellite boutons for both alleles (Fig. 5B). Therefore, in this assay system, we confirmed the genetic interaction between dABCA and hpo. 3.6. Loss-of-function mutations in wts and yki slightly suppressed the increased satellite bouton number phenotype induced by the pan-neuron-specific dABCA knockdown To clarify whether canonical Hippo pathway genes other than the hpo gene also genetically interact with dABCA, we examined the effects of the wts and yki mutations on the aberrant bouton morphology phenotype. The loss-of-function mutation in wts slightly suppressed the increased satellite bouton number phenotype induced by the dABCA knockdown (Fig. 6A, B). We also crossed the dABCA knockdown with the loss-of-function mutant of yki, a downstream target of the Hippo pathway that is negatively regulated by the Hippo pathway. Although we expected the enhancement associated with the increased satellite bouton number phenotype, a slight reduction was observed in the satellite bouton number (Fig. 6A, B). The half-dose reduction in wts or yki did not affect the number of satellite boutons (Fig. 6B). These results suggest that dABCA genetically interacts with wts, but not yki. Therefore, the canonical Hippo pathway may not be related to the phenotype induced by the pan-neuron-specific dABCA knockdown. 3.7. Knockdown of ena enhanced the increased satellite bouton number phenotype induced by the pan-neuron-specific dABCA knockdown A recent study reported that the yki-independent Strip-Hippo pathway plays a role in synaptogenesis in Drosophila [10]. Strip, the Drosophila homolog of mammalian Strip 1 and 2, is involved in the strain-interacting phosphatase and kinase (STRIPAK) complex, a negative regulator of hpo. Strip genetically interacts with Enabled (ena), an actin assembly/elongation factor. Therefore, Ena is the presumed downstream target of Hippo signaling to modulate local actin organization at synaptic termini. This regulation occurs independently of the transcriptional coactivator Yki, the canonical downstream 11
target of the Hippo pathway [10]. To assess the possible involvement of dABCA in the Hippo pathway in which Ena rather than Yki is the downstream target of Hippo signaling, we examined the effects of the ena knockdown on the increased satellite bouton number phenotype induced by the dABCA knockdown by crossing dABCA knockdown flies with ena knockdown flies (Fig. 7A). The double knockdown of dABCA and ena resulted in a slightly higher satellite bouton number than the dABCA knockdown alone, indicating a genetic interaction between dABCA and ena (Fig. 7B). The knockdown of ena alone exerted no effect on the number of satellite boutons (Fig. 7B). These results suggest that dABCA is involved in the Hippo-Ena pathway rather than in the Hippo-Yki pathway in synapses at the NMJ. 4. Discussion We previously reported that the knockdown of dABCA in pan neurons induced increases in social space with the closest neighbors and the early onset of evening activity followed by hyperactivity in adult flies, and also increased satellite bouton numbers at the NMJ of third instar larvae [8]. In the present study, we also found increases in the number or density of active zones at the NMJ in dABCA knockdown flies, suggesting increases in neurotransmitter release in dABCA knockdown flies that may result in abnormal increases in neural activity. This may explain the hyperactivity phenotype of dABCA knockdown flies. In contrast, we also found defects in learning ability in dABCA knockdown flies. Since we did not detect an aberrant morphology of MB, the limited learning abilities of dABCA knockdown flies may be caused by a functional defect rather than an abnormal MB morphology. Further analyses are needed to elucidate the underlying mechanisms. In any event, these phenotypes are similar to the defects observed in human ASD patients, suggesting that the established dABCA knockdown fly is a promising model for ASD. The asymmetric distribution of phospholipids across the plasma membrane and its local changes are important for cell functions. Substrates transported by ABCA family members
include
lipids,
glucosylceramide
for
ABCA12,
cholesterol
and
phosphatidylcholine (PC) for ABCA1, PC and phosphatidylglycerol for ABCA3, N-retinylidene-phosphatidylethanolamine for ABCA4, and PC for ABCA7. Considering their amino acid sequence homology, ABCA13 is also predicted to transport lipid 12
molecules [1]. Since the CNS represents the second most lipid-rich area in higher organisms after adipose tissue, we assume that ABCA family transporters are of critical importance for the integrity of the CNS. Recent evidence links ABCA family transporters to the maintenance of brain lipid homeostasis and neurodegenerative diseases, as suggested for other members of the ABC protein family [25]. Although transport substrates for dABCA are not yet known, dABCA, a homologue of ABCA13 appears to transport lipid molecules, as noted for human ABCA family transporters, and may play a role in maintaining brain lipid homeostasis in Drosophila. Genetic interaction studies with Drosophila mutants of the ASD candidate gene rugose (rg), the Drosophila homologue of mammalian neurobeachin, show that rg genetically interacts with components of the EGFR- and Notch-mediated signaling pathways [26]. Based on these findings, we speculated that the disruption of the asymmetric distribution of phospholipids across the plasma membrane in the brain induced by the dABCA knockdown exerts effects on some signaling pathways. In the present study, we identified hpo as a genetic interactant of dABCA. The loss-of-function of wts, another component of the Hippo pathway, slightly suppressed the increased satellite bouton number phenotype induced by the dABCA knockdown, as observed with the loss-of-function mutations in hpo. However, the loss-of-function mutation in yki, one of the most downstream components negatively regulated by the canonical Hippo pathway via its phosphorylation, slightly suppressed the increased satellite bouton number phenotype. We also performed a genetic test between dABCA and ena, a candidate downstream component that is negatively regulated by the Yki-independent Strip-Hippo pathway. The double knockdown of dABCA and, as expected, ena slightly enhanced the increased satellite bouton number phenotype induced by the dABCA knockdown alone. These results suggest that dABCA is involved in the Hippo-Ena pathway rather than in the Hippo-Yki pathway in synapses at the NMJ. However, we cannot exclude the possibility that some unknown downstream component of the Hippo pathway other than yki and ena is involved in this pathway at the NMJ. Further studies are needed to clarify this point. In any event, we herein identified a novel genetic link between dABCA and hpo. These results will contribute to clarifying the pathogenesis of ASD as well as the development of disease-modifying therapy.
13
Competing financial interests The authors declare no competing interests in relation to the work described. Acknowledgments We thank the Bloomington Drosophila Stock Center and Vienna Drosophila Genetic Resource Center for the fly lines. This research was partially supported by the JSPS Core-to-Core Program, Asia-Africa Science Platforms B, and JSPS KAKENHI Grant Number JP19K06659. We also thank Medical English Service for the English language review.
14
References [1] M. Tomioka, Y. Toda, J. Kurisu, Y. Kimura, M. Kengaku, K. UEDA, The effects of neurological disorder-related codon variations of ABCA13 on the function of the ABC protein. Biosci. Biotechnol. Biochem. 76 (2012) 2289–2293. [2] I.Iossifov, M. Ronemus, D. Levy, Z. Wang, I. Hakker, J. Rosenbaum, B. Yamrom, Y.H. Lee, G. Narzisi, A. Leotta, J. Kendall, E. Grabowska, B. Ma, S. Mark, L. Rodgers, A. Stepansky, J. Troge, P. Andrews, M. Bekritsky, K. Pradhan, E. Ghiban, M. Kramer, J. Parla, R. Demeter, L.L. Fulton, R.S. Fulton, V.J. Magrini, K. Ye, J.C. Darnell, R.B. Darnell, E.R. Mardis, R.K. Wilson, M.C. Schatz, W.R. McCombie, M. Wigler, De Novo gene disruptions in children on the autistic spectrum. Neuron 74 (2012) 285–299. [3] I.Iossifov, B.J. O’Roak, S.J. Sanders, M. Ronemus, N. Krumm, D. Levy, H.A. Stessman, K.T. Witherspoon, L. Vives, K.E. Patterson, J.D. Smith, B. Paeper, D.A. Nikerson, J. Dea, S. Dong, L.E. Gonzalez, J.D. Mandell, S.M. Mane, M.T. Murtha, C.A. Sullivan, M.F. Walker, Z. Waqar, L. Wei, A.J. Willsey, B. Yamrom, Y.H. Lee, E. Grabowska, E. Dalkic, Z. Wang, S. Marks, P. Andrews, A. Leotta, J. Kendall, I. Hakker, J. Rosenbaum, B. Ma, L. Rodgers, J. Troge, G. Narzisi, S. Yoon, M.C. Schatz, K. Ye, W.R. McCombie, J. Shendure, E.E. Eichler, M.W. State, M. Wigler, The contribution of de novo coding mutations to autism spectrum disorder. Nature 515 (2014) 216–221. [4] S. De Rubeis, X. He, A.P. Goldberg, C.S. Poultney, K. Samocha, A.E. Cicek, Y. Kou, L. Liu, M. Fromer, S. Walker, et al. Synaptic, transcriptional and chromatin genes disrupted in autism. Nature 515 (2014) 209–215. [5] H.M. Knight, B.S. Pickard, A. Maclean, M.P. Malloy, D.C. Soares, A.F. McRae, A. Condie, A. White, W. Hawkins, K. McGhee, M. van Beck, D.J. Maclntyre, J.M. Starr, I.J. Deary, P.M. Visscher, D.J. Deary, P.M. Visscher, D.J. Porteous, R. E. Cannon, D. St. Clair, W.J. Muir, D.H. Blackwood, A cytogenetic abnormality and rare coding variants identify ABCA13 as a candidate gene in schizophrenia, bipolar disorder, and depression. Am. J. Hum. Genet. 85 (2009) 833–846. [6] K. Yoshida, Y. Go, I. Kushima, A. Toyoda, A. Fujiyama, H. Imai, N. Saito, A Iriki, N. Ozaki, M. Isoda, Single-neuron and genetic correlates of autistic behavior in macaque. Sci. Adv. 2 (2016) e1600558. 15
[7] S. Iritani, Y. Torii, C. Habuchi, H. Sekiguchi, H. Fujishiro, M. Yoshida, Y. Go, A. Iriki, M. Isoda, N. Ozaki, The neuropathological investigation of the brain in a monkey model of autism spectrum disorder with ABCA13 deletion. Int. J. Dev. Neurosci. 71 (2018) 130–139. [8] I. Ueoka, H. Kawashima, A. Konishi, M. Aoki, R. Tanaka, H. Yoshida, T. Maeda, M. Ozaki, M. Yamaguchi, Novel Drosophila model for psychiatric disorders including autism spectrum disorder by targeting of ATP-binding cassette protein A. Exp. Neurol. 300 (2018) 51–59. [9] B. Zhao, K. Tumaneng, K.L. Guan, The Hippo pathway in organ size control, tissue regeneration and stem cell self-renewal. Nat. Cell Biol. 13 (2011) 877–883. [10] C. Sakuma, Y. Saito, T. Umehara, K. Kamimura, N, Maeda, T.J. Mosca, M. Miura, T. Chihara. The Strip-Hippo pathway regulates synaptic terminal formation by modulating actin organization at the Drosophila neuromuscular synapses. Cell Rep. 16 (2016), 2289-2297. [11] Y. Azuma, T. Tokuda, Y. Kushimura, I. Yamamoto, I. Mizuta, T. Mizuno, M. Nakagawa, M. Ueyama, Y. Nagai, Y. Iwasaki, M. Yoshida, D. Pan, H. Yoshida, M. Yamaguchi, Hippo, Drosophila MST, is a novel modifier of motor neuron degeneration induced by knockdown of Caz, Drosophila FUS, Exp. Cell Res. 371 (2018) 311–321. [12] Y. Kushimura, Y. Azuma, I. Mizuta, Y. Muraoka, A. Kyotani, H. Yoshida, T. Tokuda, T. Mizuno, M. Yamaguchi, Loss-of-function mutation in Hippo suppressed enlargement of lysosomes and neurodegeneration caused by dFIG4 knockdown. Neuroreport 29 (2018) 856–862. [13] B. Gerber, R. Biernacki, J. Thum, Odor-taste learning assays in Drosophila larvae, Cold Spring Harb. Protoc. 8 (2013) 213–223. [14] S. Jantrapirom, L. Lo Piccolo, H. Yoshida, M. Yamaguchi, A new Drosophila model of Ubiquilin knockdown shows the effect of impaired proteostasis on locomotive and learning abilities, Exp. Cell Res. 362 (2017) 461–471. [15] D.S. Brooks, K. Vishal, J. Kawakami, S. Bouyain, R. Erika, Optimization of wrMTrck to monitor Drosophila larval locomotor activity, J. Insect Physiol. 93-94 (2016) 11–17. [16] C.I. Nussbaum-Krammer, M.F. Neto, R.M. Brielmann, J.S. Pederson, R.I. Morimoto, Investigating the spreading and toxicity of prion-like proteins using the 16
metazoan model organism C. elegans, J. Vis. Exp. 95 (2015) 52321. [17] F. Foti, F. Piras, S. Vicari, L. Mandolesi, L. Petrosini, D. Menghini, Observational learning in low-functioning children with autism spectrum disorders: A behavioral and neuroimaging study. Front. Psychol. 9 (2019) 1–13. [18] M. Kurusu, T. Awasaki, L.M. Masuda-Nakagawa, H. Kawauchi, K. Ito, K. Furukubo-Tokunaga, Embryonic and larval development of the Drosophila mushroom bodies: concentric layer subdivisions and the role of fasciclin II. Development 129 (2002) 409–419. [19] K.P. Menon, R.A. Carrillo, K. Zinn, Development and plasticity of the Drosophila larval neuromuscular junction. Wiley Interdiscip. Rev. Dev. Biol. 2 (2013) 647– 670. [20] G. Lee, T.L. Schwarz, Filamin, a synaptic organizer in Drosophila, determines glutamate receptor composition and membrane growth. Elife 5 (2016) e19991. [21] P. Wee, Z. Wang, Epidermal growth factor receptor cell proliferation signaling pathways. Cancers (Basel) 9 (2017) 1–45. [22] J.D. Wasserman, S. Urban, M. Freeman, A family of rhomboid-like genes: Drosophila rhomboid-1 and roughoid/rhomboid-3 cooperate to activate EGF receptor signaling. Genes Dev. 14 (2000) 1651–1663. [23] B.-Z. Shilo, Regulating the dynamics of EGF receptor signaling in space and time. Development 132 (2005) 4017–4027. [24] S. Yogev, E.D. Schejter, B.Z. Shilo, Drosophila EGFR signalling is modulated by differential compartmentalization of Rhomboid intramembrane proteases. EMBO J. 27 (2008) 1219–1230. [25] A.P. Piehler, M. Özcurumez, W.E. Kaminski, A-subclass ATP-binding cassette proteins in brain lipid homeostasis and neurodegeneration. Front. Psychiatry 3 (2012) 1–16. [26] A. Wise, L. Tenezaca, R.W. Fernandez, E. Schatoff, J. Flores, A. Ueda, X. Zhong, C.F. Wu, A.F. Simon, T. Venkatesh, Drosophila mutants of the autism candidate gene neurobeachin (rugose) exhibit neuro-developmental disorders, aberrant synaptic properties, altered locomotion, and impaired adult social behavior and activity patterns. J. Neurogenet. 29 (2015) 135-143.
17
Figure legends Fig. 1. The knockdown of dABCA exerted a marginal effect on larval locomotive ability evaluated by the crawling assay. (A) Diagram of path parameters, including the length (blue) and distance (grey) from start to finish measured in wrMTrck. (B) Traces of larval path routes. Third instar larvae of the indicated genotypes were monitored for 60 sec and path lengths were analyzed using ImageJ and wrMTrck. Speed (mm/sec) (C) and distance from start to finish (mm) (D). (C) Speed ware analyzed with one-way ANOVA followed by Dunnet’s post hoc test. (D) Distance from start to finish ware analyzed with Kruscal-Wallis test followed by Dunn’s post hoc test. Bar graph represents the mean ± SEM. Box and whisker plots represent the minimum, 25th percentile, median, 75th percentile and maximum. *p < 0.05. The genotype of each fly was
as
follows:
elav>dABCA-IR666-672 elav>dABCA-IR997-1106 elav>dABCA-IR997-1106
elav>GFP-IR (w/Y; (w/Y; 2copy
(w/Y;
(w/Y;
UAS-GFP-IR/+;
+;
elav-Gal4/UAS-dABCA-IR666-672);
UAS-dABCA-IR997-1106/+;
elav-Gal4/+); elav-Gal4/+);
UAS-dABCA-IR997-1106/UAS-dABCA-IR997-1106;
elav-Gal4/+). Fig. 2. The knockdown of dABCA induced defects in larval learning ability. (A) Principle of the odor-taste learning test. TRAINING 1: In one of the groups, n-amyl acetate (AM) was added with a sucrose reward (+) and 1-octanol (OCT) was subsequently added without a reward (AM+/OCT). TRAINING 2: The other group received reciprocal training (AM/OCT+). After training exposure, larvae were tested for their choice between AM and OCT. Stronger preferences for AM after AM+/OCT training than after AM/OCT+ training reflected associative learning and were quantified by the learning index (LI). (B) Larval preference for AM versus OCT (1 = all prefer AM; −1 = all prefer OCT) was analyzed with the Friedman test followed by Dunn’s post hoc test. Box and whisker plots represent the minimum, median, and maximum values. (C) Normalized AM and OCT learning indexes were averaged for LI and analyzed with a one-way ANOVA followed by Dunnett’s post hoc test. Bar Graph plots the mean ± SEM. *p<0.05, **p<0.01, ***p<0.001 and ns means no significance (p>0.05). (D) Larval and adult mushroom bodies (MBs). Third instar larval brains and adult brains stained with anti-FASII IgG (gray). Brains are marked with a white line and larval MBs are marked with a white dashed line. α, adult dorsal lobes; β and γ, adult 18
medial lobes; cx, calyx; DL, dorsal lobe; KCs, Kenyon cells; ML, medial lobe; OL, optic lobe; Ped, peduncle; VNC, ventral nerve cord. Reconstruction of optical sections. Bars indicate 50 µm. The genotype of each fly was as follows: elav>GFP-IR (w/Y; UAS-GFP-IR/+;
elav-Gal4/+);
elav>dABCA-IR666-672
elav-Gal4/UAS-dABCA-IR666-672); UAS-dABCA-IR997-1106/+;
(w/Y;
elav>dABCA-IR997-1106
elav-Gal4/+);
elav>dABCA-IR997-1106
+; (w/Y;
2copy
(w/Y;
UAS-dABCA-IR997-1106/UAS-dABCA-IR997-1106; elav-Gal4/+). Fig. 3. Knockdown of dABCA increased the number or density of active zones in synapses at the NMJ. (A) Super-resolution images of synapses at muscle 4 NMJs of third instar larvae. A representative image of anti-HRP staining, anti-BRP staining, and merged images. Bars indicate 1 µm. (B) Confocal images of synapses at muscle 4 NMJs of third instar larvae. A representative image of anti-HRP staining, anti-GluRIIA staining, and merged images. Bars indicate 10 µm. The total BRP-positive area per synapse area at the NMJ (C), total BRP-positive signal number per synapse area at the NMJ (D), and GluIIA-positive area per synapse area at the NMJ (E). (C, D, E) Each data set was analyzed with a one-way ANOVA followed by Dunnett’s post hoc test. Graphs plot the mean ± SEM. *p<0.05, **p<0.01, ***p<0.001 and ns means no significance (p>0.05). The genotype of each fly was as follows: elav>GFP-IR (w/Y; UAS-GFP-IR/+;
elav-Gal4/+);
elav>dABCA-IR666-672
elav-Gal4/UAS-dABCA-IR666-672); UAS-dABCA-IR997-1106/+;
(w/Y;
elav>dABCA-IR997-1106
elav-Gal4/+);
elav>dABCA-IR997-1106
+; (w/Y;
2copy
(w/Y;
UAS-dABCA-IR997-1106/UAS-dABCA-IR997-1106; elav-Gal4/+). Fig. 4. Mutations in hpo enhanced the hyperactivity phenotype induced by the dABCA knockdown, whereas the knockdown of rho did not. Five-day average activity plotted as the number of infrared beam crossings measured per 15 min of elav>dABCA-IR997-1106/GFP-IR elav-Gal4/+),
(w/Y;
UAS-dABCA-IR997-110/UAS-GFP-IR;
elav>dABCA-IR997-1106/rho-IR
UAS-dABCA-IR997-1106/+;
elav-Gal4/UAS-rho-IR
elav>dABCA-IR997-1106/rho-IR elav-Gal4/UAS-rho-IR UAS-dABCA-IR997-1106/+;
(VDRC51952) (VDRC51952)),
(VDRC51953)(w/Y;
(VDRC51953))
(A),
elav-Gal4/+), 19
(w/Y; and
UAS-dABCA-IR997-1106/+;
elav>dABCA-IR997-1106 KS240
elav>dABCA-IR997-1106/hpo
(w/Y; (w/Y;
UAS-dABCA-IR997-110/ hpoKS240; elav-Gal4/+), and elav>dABCA-IR997-1106/hpoKC202 (w/Y; UAS-dABCA-IR997-110/ hpoKC202; elav-Gal4/+) (B), and elav>+ (w/Y; +; elav-Gal4/+), elav>hpoKS240 (w/Y; hpoKS240/+; elav-Gal4/+), and elav>hpoKC202 (w/Y; hpoKC202/+; elav-Gal4/+) (C). The average total crossing activities during the day (21:00-21:00)
of
elav>dABCA-IR997-1106/GFP-IR,
elav>dABCA-IR997-1106/rho-IR
(VDRC51952), and elav>dABCA-IR997-1106/rho-IR (VDRC51953) (D) were analyzed with the Kruskal-Wallis test followed by Dunn’s post hoc test, while those of elav>dABCA-IR997-1106/hpoKS240
elav>dABCA-IR997-1106, elav>dABCA-IR997-1106/hpo
KC202
(E), and elav>+, elav>hpo
KS240
and elav>hpo
and KC202
(F)
were analyzed with a one-way ANOVA followed by Dunnett’s post hoc test. Box and whisker plots represent the minimum, 25th percentile, median, 75th percentile and maximum. Bar graphs represent the mean ± SEM. *p<0.05, **p<0.01 and ns means no significance (p>0.05). Fig. 5. Mutations in hpo suppressed the increased satellite bouton phenotype induced by the dABCA knockdown. (A) Confocal images of synapses at muscle 4 NMJs of third instar larvae. A representative merged image of anti-HRP staining and anti-Dlg staining. Bars indicate 10 µm. The inset in each panel shows a higher magnification image of the boxed area. (B) The number of satellite boutons was analyzed with the Kruskal-Wallis test followed by Dunn’s post hoc test. Box and whisker plots represent the minimum, 25th percentile, median, 75th percentile and maximum. **p<0.01, ****p<0.0001 and ns means no significance (p>0.05). The genotype of each fly is as follows: elav>GFP-IR (w/Y; UAS-GFP-IR/+; elav-Gal4/+); elav>dABCA-IR997-1106/hpoKS240 (w/Y; UAS-dABCA-IR997-110/ hpoKS240; elav-Gal4/+); elav>dABCA-IR997-1106/hpoKC202 (w/Y; UAS-dABCA-IR997-110/ hpoKC202; elav-Gal4/+); elav>dABCA-IR997-1106 (w/Y; UAS-dABCA-IR997-1106/+; elav-Gal4/+); elav>hpoKS240 (w/Y; hpoKS240/+; elav-Gal4/+); elav>hpoKC202 (w/Y; hpoKC202/+; elav-Gal4/+). Fig. 6. Mutation in wts suppressed the increased satellite bouton phenotype induced by the dABCA knockdown, whereas that of yki did not. (A) Confocal images of synapses at muscle 4 NMJs of third instar larvae. A representative merged image of anti-HRP staining and anti-Dlg staining. Bars indicate 10 µm. The inset in each panel shows a higher magnification image of the boxed area. (B) The number of 20
satellite boutons was analyzed with the Kruskal-Wallis test followed by Dunn’s post hoc test. Box and whisker plots represent the minimum, 25th percentile, median, 75th percentile and maximum. ****p<0.0001 and ns means no significance (p>0.05). The genotype of each fly was as follows: elav>GFP-IR (w/Y; UAS-GFP-IR/+; elav-Gal4/+);
elav>dABCA-IR997-110/ X1
elav-Gal4/wts );
wtsX1
elav>dABCA-IR997-110/yki
B5
(w/Y; (w/Y;
UAS-dABCA-IR997-110/
+; B5
UAS-dABCA-IR997-110/yki ;
elav-Gal4/+); elav>dABCA-IR997-1106 (w/Y; UAS-dABCA-IR997-1106/+; elav-Gal4/+); elav>wtsX1 (w/Y; +; elav-Gal4/ wtsX1); elav>ykiB5 (w/Y; ykiB5/+; elav-Gal4/+). Fig. 7. Knockdown of ena enhanced the increased satellite bouton phenotype induced by the dABCA knockdown. (A) Confocal images of synapses at muscle 4 NMJs of third instar larvae. A representative merged image of anti-HRP staining and anti-Dlg staining. Bars indicate 10 µm. The inset in each panel shows a higher magnification image of the boxed area. (B) The number of satellite boutons was analyzed with the Kruskal-Wallis test followed by Dunn’s post hoc test. Box and whisker plots represent the minimum, 25th percentile, median, 75th percentile and maximum. *p<0.05 and ns means no significance (p>0.05). The genotype of each fly is as
follows:
elav>GFP-IR
(w/Y;
UAS-GFP-IR/+;
elav-Gal4/+);
elav>dABCA-IR997-110/ena-IR (w/Y; UAS-dABCA-IR997-110/+; elav-Gal4/ UAS-ena-IR); elav>dABCA-IR997-110/GFP-IR (w/Y; UAS-dABCA-IR997-110/UAS-GFP-IR; elav-Gal4/+); elav>ena-IR (w/Y; +; elav-Gal4/UAS-ena-IR). Supplementary Fig. S1. Hyperactivity phenotype induced by the dABCA knockdown. Five-day activity plotted as the number of infrared beam crossings measured per 15 min by male (A) or female (B) flies with elav>GFP-IR (w/Y; UAS-GFP-IR/+;
elav-Gal4/+)
or
elav>dABCA-IR997-1106
2copy
(w/Y;
UAS-dABCA-IR997-1106/UAS-dABCA-IR997-1106; elav-Gal4/+). Average total crossing activity during the indicated day (21:00-21:00) for male (C) or female (D) was analyzed with two-way ANOVA followed by Dunnett’s post hoc test. Graphs were represented by the mean ± SEM. **p<0.01, **** p<0.0001.
21
Highlights
・Neuron-specific knockdown of dABCA increases active zone at NMJ. ・Hippo mutation enhanced the hyperactivity phenotype induced by dABCA knockdown. ・Hippo mutation suppressed the decreased bouton number phenotype induced by dABCA knockdown. ・Novel genetic link between psychiatric disorder-causing gene and hippo.
Conflict of Interest Form The authors declare no conflict interest in relation to the work described.
S0014-4827(19)30616-0
DOI:
https://doi.org/10.1016/j.yexcr.2019.111733
Reference:
YEXCR 111733
To appear in:
Experimental Cell Research
Received Date: 12 June 2019 Revised Date:
16 October 2019
Accepted Date: 16 November 2019
Please cite this article as: I. Ueoka, A. Takai, M. Yamaguchi, T. Chiyonobu, H. Yoshida, M. Yamaguchi, Novel genetic link between the ATP-binding cassette subfamily A gene and hippo gene in Drosophila, Experimental Cell Research (2019), doi: https://doi.org/10.1016/j.yexcr.2019.111733. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc.
Novel genetic link between the ATP-binding cassette subfamily A gene and hippo gene in Drosophila
Ibuki Ueoka1, 2, Akari Takai3, Mizuki Yamaguchi1, 2, Tomohiro Chiyonobu3, Hideki Yoshida1, 2, *, and Masamitsu Yamaguchi1, 2, *
1
Department of Applied Biology and 2Advanced Insect Research Promotion Center,
Kyoto Institute of Technology, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan. 3
Department of Pediatrics, Kyoto Prefectural University of Medicine, Graduate School
of Medical Science, 465 Kaji-cho, Hirokoji, Kamigyo-ku, Kyoto 602-8566, Japan.
* Co-correspondence: M. Yamaguchi, Department of Applied Biology, Kyoto Institute of Technology, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan. Tel.: +81-75-724-7781; Fax: +81-75-724-7799; E-mail: [email protected] H. Yoshida, Department of Applied Biology, Kyoto Institute of Technology, Matsugasaki, Sakyo-ku, Kyoto, 606-8585, Japan. Tel.: +81-75-724-7787; Fax: +81-75-724-7787; E-mail: [email protected]
1
Abstract The pan-neuron-specific knockdown of dABCA, a Drosophila homologue of the human ATP binding cassette subfamily A member 13 gene, increases social space without affecting climbing ability and induces the early onset of evening activity in adult flies followed by relatively high activity throughout the day. Satellite bouton numbers in the presynaptic terminals of motor neurons are increased in dABCA knockdown flies. In the present study, we further characterized pan-neuron-specific dABCA knockdown flies and found that active zones in the presynaptic terminals of motor neurons increased, whereas learning abilities decreased in larvae. Genetic crossing experiments revealed that the hippo mutation enhanced the hyperactivity phenotype of adults, but suppressed the increased satellite bouton phenotype induced by the dABCA knockdown. Drosophila ABCA is predicted to transport lipid molecules and impair the asymmetric distribution of phospholipids across the plasma membrane, and these local changes are considered to be important for various cellular functions. The disruption of lipid homeostasis in central and peripheral nervous systems by the dABCA knockdown may affect the Hippo-related signaling pathway in order to induce the observed phenotypes. Keywords: autism spectrum disorder; ATP-binding cassette protein A; Drosophila; neuron; hippo gene; activity monitoring Abbreviations: ABCA, ATP-binding cassette subfamily A; ASD, autism spectrum disorder; CNS, central nervous system; NMJ, neuromuscular junction; hpo, hippo
2
1. Introduction Human adenosine triphosphate (ATP)-binding cassette subfamily A member 13 (ABCA13), a member of the ABCA protein family, is predicted to have the typical structure of full-size ABC proteins, including two transmembrane domains (TMDs), each containing six transmembrane α helices, and two nucleotide-binding domains (NBDs), with the latter containing characteristic ATP-binding Walker A and B motifs. Similar to other ABCA protein family members, ABCA13 is predicted to transport lipid molecules [1]. Mutations in the human ABCA13 gene are associated with autism spectrum disorder (ASD) [2-4] as well as schizophrenia and bipolar disorder [1,5]. Recent studies showed that a monkey carrying the heterozygous ABCA13 deletion and a mutation in the serotonin 2C receptor (5HT2c) displayed an impaired ability to monitor the behavior of others in addition to repetitive behavior, which are frequently associated with ASD [6]. This monkey model of ASD with the ABCA13 deletion and a mutation in 5HT2c may have impaired neural maturation in the central nervous system (CNS) [7]. Drosophila has a homologue for human ABCA family genes called dABCA (CG1718), which shows the highest homology to the human ABCA13 gene. The pan-neuron-specific knockdown of dABCA resulted in increased social space with the closest neighbor in adult male flies, but did not affect climbing ability, indicating that the increase in social space is not due to a defect in climbing ability [8]. An activity assay with adult male flies revealed that the knockdown of dABCA in all neurons induced the early onset of evening activity in adult flies followed by relatively high activity during morning peaks, evening peaks, and midday siestas [8]. These phenotypes are similar to the defects observed in human ASD patients, suggesting that the established dABCA knockdown flies are a promising model for ASD. Furthermore, satellite bouton numbers were increased in the presynaptic terminals of motor neurons at the neuromuscular junction (NMJ) in dABCA knockdown third instar larvae, suggesting that dABCA regulates the formation and/or maintenance of the presynaptic terminals of motor neurons [8]. The Hippo pathway is a tumor-suppressive signaling cascade that was originally found in Drosophila. In this pathway, Hippo (Hpo) phosphorylates Warts (Wts) with Salvador (Sav) and Mats, and phosphorylated Wts then phosphorylates Yorkie (Yki), a transcriptional coactivator. This pathway limits cell proliferation by inducing apoptosis and cell cycle arrest [9]. In addition to the role of the Hippo pathway in the regulation of 3
cell proliferation, recent studies reported that the Strip-Hippo pathway is involved in synaptic development and this regulation is independent of Yki [10]. The specific knockdown of strip in motor neurons increased satellite bouton numbers at the NMJ, which is similar to that observed with the knockdown of dABCA [8,10]. This satellite bouton phenotype induced by the knockdown of strip was effectively suppressed by a hpo mutation [10]. More recent studies have implicated hpo in some neurodegenerative diseases, such as amyotrophic lateral sclerosis and Charcot–Marie–Tooth disease [11,12]. In the present study, we therefore investigated whether a genetic interaction exists between dABCA and hpo. The hpo mutation enhanced the hyperactivity phenotype of adults, but suppressed the increased satellite bouton phenotype induced by the knockdown of dABCA. The disruption of lipid homeostasis in central and peripheral nervous systems by the knockdown of dABCA may explain the observed phenotypes that appear to be mediated by some Hippo-related signaling pathways. 2. Materials and methods 2.1. Fly stocks Flies were raised on standard food (0.65% agar, 10% glucose, 4% dry yeast, 5% cone flour, and 3% rice bran) under a 12-h light-dark cycle at 25℃. The following fly stocks obtained from the Bloomington Drosophila Stock Center were used in the present study: UAS-dABCA-IR666-672 (HMS01796, BL38329) targeted to the region corresponding to amino acid residues (aa) 666-672 of dABCA, UAS-strip-IR (HMS01134, BL34657), UAS-ena-IR (HMS01953, BL39034), UAS-GFP-IR (BL9331), elav-Gal4 (third chromosome insertion, BL8760), hpoKS240 (BL25085), hpoKC202 (BL25090), wtsX1 (BL44251), and ykiB5 (BL36290). UAS-dABCA-IR997-1106 (GD3708, VDRC44449) targeted to the region corresponding to aa997-1106 of dABCA, UAS-rho-IR (GD2243, VDRC51952), and UAS-rho-IR (GD2243, VDRC51953) were obtained from the Vienna Drosophila RNAi Center (VDRC). To minimize the effects of the genetic background, the flies used in the present study were backcrossed 6 times with the w strain. 2.2. Odor-taste learning assay for larvae The Pavlovian-type larval learning assay based on odor-taste was performed with some modifications [13,14]. A group of larvae was exposed to n-amyl acetate (AM) (Merck) 4
in the presence of a reward (2 M sucrose, SUC) and this was followed by 1-octanol (OCT) (Nacalai) in the absence of SUC. This training was defined as AM+/OCT, indicating “+” as the reward. In the next step, reciprocal training was performed. A group of larvae was sequentially exposed to OCT in the presence of SUC that was followed by AM exposure in the absence of SUC. The second training step was defined as OCT+/AM. Five minutes of each training allowed flies to establish an association between AM or OCT and the reward. This set of training was repeated three times. Ten microliters of each odorant was added to a 0.2-mL Eppendorf tube with a perforated lid and placed on the opposite site inside the proper Petri dish filled with 1% agarose. AM was provided undiluted, while OCT was used at a 1:75 dilution with liquid paraffin. Two Petri dishes were used for the training period, one for each couple of odorants. Each dish was divided into two distinct zones with a 1-cm so-called neutral zone arranged in the middle. The neutral zone was defined as the area at which larvae receive the same intensity of each odorant. Twenty-four larvae for each corresponding training (AM +/OCT and OCT+/AM) were divided into three groups, and each group of 8 received training in the appropriate dish. After training, larvae were tested by exposure to AM and OCT in the absence of SUC. The odorants were deposited on the opposite ends of each dish and larvae were transferred into a dish in groups of 8 after the training period. The resulting preference indexes were used to calculate the learning index (LI). The AM preference was calculated as: (number of larvae on the AM side-number of larvae on the OCT side)/ (total number of larvae on both sides). The AM preference ranged between 1 (perfect attraction to AM) and −1 (perfect attraction to OCT). The normalized AM LI was calculated as (AM preference-Average of OCT preference)/2. The normalized OCT LI was calculated as (Average AM preference-OCT preference)/2. LI was calculated as (Normalized AM + Normalized OCT)/2. 2.3. Crawling assay for larvae A crawling assay was performed as described previously [15]. Flies were cultured at 28°C. After laying eggs for 24 hours, adult flies were removed from the food vial. Larvae (20~30 larvae per vial) were cultured under not overcrowded conditions. At the beginning of the wandering stage, only third instar larvae that were wandering out of the food were transferred into a 15-cm Petri dish containing 2% agarose at a density of 4 larvae per plate. A 1-min video was acquired and analyzed using ImageJ software with 5
the wrMTrck plug-in [16]. 2.4. Immunostaining An immunohistochemical analysis of the CNS tissues of larvae and adults was performed as described previously [14]. CNS tissues from third instar male larvae or 1-day-old adult male flies were dissected in phosphate-buffered saline (PBS) and fixed in 4% paraformaldehyde in PBS at 25°C for 20 min. After washing with 0.3% Triton X-100 in PBS (PBST), samples were blocked in 0.3% PBST with 2% normal goat serum (NGS) at 25°C for 30 min. An incubation with the primary antibody was performed in 0.15% PBST with 10% NGS at 4°C for 16 h. The anti-Fasciclin II (Fas II) antibody (mouse, 1:20 dilution, DSHB 1D4) was used as a primary antibody. After washing with 0.3% PBST, the secondary antibody incubation was conducted in 0.15% PBST with 10% NGS at 25°C for 2 h. Alexa Fluor 488 conjugated anti-mouse IgG (Invitrogen, all were used at a 1:400 dilution) was used as a secondary antibody. After washing with 0.3% PBST, samples were mounted with Vectashield (Vector Laboratories). In the immunohistochemical analysis of the NMJ, third instar male larvae were dissected in HL3 saline and fixed in 4% paraformaldehyde in PBS at 25°C for 30 min. After washing with 0.3% PBST, samples were blocked in 0.15% PBST with 10% NGS at 25°C for 30 min. The primary antibody incubation was performed in 0.15% PBST with 10% NGS at 4°C for 16 h. The following primary antibodies were used: anti-Discs large (Dlg) (mouse, 1:200 dilution, DSHB 4F3), anti-Bruchpilot (BRP) (mouse, 1:200 dilution, DSHB nc82), and anti-GluRIIA (mouse, 1:80 dilution, DSHB 8B4D2) IgGs. FITC-conjugated
goat
anti-horseradish
peroxidase
(HRP)
IgG
(1:400,
MP
Biochemicals) was used to visualize neurons. After washing with 0.3% PBST, the incubation with the secondary antibody and FITC-conjugated anti-HRP IgG was performed in 0.15% PBST with 10% NGS at 25°C for 2 h. The following secondary antibodies were used (all from Invitrogen and at a 1:400 dilution): Alexa Fluor 594 conjugated anti-mouse IgG and Alexa Fluor 464 conjugated anti-mouse IgG. After washing with 0.3% PBST, samples were mounted with Vectashield (Vector Laboratories) for inspection by a confocal laser scanning microscope (Olympus Fluoview FV10i) or ProLong Diamond (Invitrogen) for inspection by a super resolution microscope (Nikon, N-SIM). Images of the NMJ at muscle 4 of the A3 and A4 6
segments were acquired and processed with ImageJ software. 2.5. Drosophila activity assay Flies were reared for more than several generations at 25°C under a 12-h light-dark cycle. Adult male flies were selected from each group to be analyzed in the Drosophila Activity Monitor (DAM, from Trikinetics) under a 12-h light-dark cycle equipped in a 25°C incubator [8]. One-day-old virgin male or female flies were individually introduced into the DAM glass tube. Fly activity was recorded for at least 4 consecutive days. The activity recorded for each 15-min bin was based on how many times per 15 min each individual fly crossed an infrared light beam that bisected the capillary tube and was perpendicular to the tube. The measured values were plotted in the graph at the endpoint of the 15-min measurement as the average activity bouts of the flies. 2.6. Statistical analysis Statistical analyses were performed using GraphPad Prism 6.0 software and Excel 2016. The criteria for significance were: ns (not significant, p> 0.05), *p< 0.05, **p < 0.01, ***p<0.001, ****p<0.00001. The normality of all data was analyzed with the Shapiro-Wilk normality test and visualization of histograms. Data following a normal distribution were analyzed with a parametric test (one-way ANOVA or two-way ANOVA followed by Dunnett’s or Bonferroni’s post hoc test). Some data followed a non-normal distribution because a number of outliers existed. Data following a non-normal distribution were analyzed with non-parametric tests (the Kruskal-Wallis test or Friedman test followed by Dunn’s post hoc test). The graph representation, error bars, box segments and whiskers are defined in each figure legend, together with which statistical test was performed. 3. Results 3.1. dABCA knockdown in pan neurons exerted only a marginal effect on larval locomotive ability
We previously reported that defects in the social interactions of the dABCA knockdown fly were more severe in male flies than in female flies [8]. Furthermore, in activity monitoring, we obtained more consistent data in male flies than in female flies (Supplementary Fig. S1). Therefore, we used male flies in the present study. The 7
pan-neuron-specific knockdown of dABCA exerted no apparent effect on the climbing ability of adult flies [8]. To investigate the effects of the dABCA knockdown on locomotive ability in more detail, we performed a larval crawling assay (Fig. 1). We used two independent RNAi lines targeted to different regions of dABCA mRNA. The effective knockdown of dABCA in these RNAi lines was previously demonstrated by Western immunoblot analyses with male adult head extracts [8]. Previous studies also revealed that these RNAi lines showed a similar defect in social activity [8]. The crawling speed (mm/sec) of elav>dABCA-IR997-1106 was slightly slower than that of control elav>GFP-IR flies, while other dABCA knockdown flies including elav>dABCA-IR997-1106 2copy did not show any significant difference in crawling speed from control flies (Fig. 1C). The distance from start to finish (mm) in a 1-min test of elav>dABCA-IR997-1106 and elav>dABCA-IR997-1106 was slightly shorter than that of control flies, while other dABCA knockdown flies including elav>dABCA-IR997-1106 2copy did not show significant differences in distance from start to finish from the control (Fig. 1A, D). These results indicate that the knockdown of dABCA exerts only a marginal effect on larval locomotive ability. 3.2. dABCA knockdown in pan neurons impaired larval learning ability ASD is sometimes accompanied by a defect in learning ability [17]. To investigate the role of dABCA in complex neuronal functions, we performed a Drosophila larval olfactory learning assay (Fig. 2A) [13]. Control elav>GFP-IR larvae showed the ability to make the correct association between an odorant and the reward because their preference to the reward odorant was stronger than that to the no reward odorant in accordance with training (Fig. 2B). Control larvae after training in the presence of AM and SUC (AM+) exhibited the ability to recognize the reward odorant in the presence of both odorants (AM+/ OCT), while after training in the presence of OCT and SUC (OCT+), they showed a stronger preference for OCT in the presence of both odorants (OCT+/AM) (Fig. 2B). However, dABCA knockdown larvae did not show a significantly higher number of preferred AM in AM+/OCT than in OCT+/AM (Fig. 2B). LI showed that the pan-neuron-specific dABCA knockdown resulted in a reduction in learning abilities (Fig. 2C). The quantification of data revealed that the LI of elav>dABCA-IR666-672, elav>dABCA-IR997-1106, and elav>dABCA-IR997-1106 2copy flies were 23, 79, and 115% lower, respectively, than that of control elav>GFP-IR flies (Fig. 8
2C). Mushroom bodies (MBs) are the centers for olfactory associative learning and elementary cognitive functions in the Drosophila brain, and aberrant morphologies have been observed with other model flies for neurodegenerative diseases [14]. Therefore, we used the antibody to the cell adhesion molecule Fasciclin II (FASII), which is a well-established MB marker [18], to perform the immunostaining of larval brain lobes and the adult brain (Fig. 2D). Control and elav>dABCA-IR997-1106 2copy flies showed a symmetrical distribution of MBs between the two larval brain lobes (thin white dashed line) and adult brains with a sharp separation of the midline and a normal arrangement of structures (Fig. 2D). Thus, there was no detectable difference in the appearance of MB structures between control and dABCA knockdown larvae (Fig. 2D). 3.3. dABCA knockdown in pan neurons increased the number or density of active zones at the NMJ We previously reported that the number of satellite boutons at synapses in the NMJ increased in dABCA knockdown flies, whereas the number of synapse branches, the total branch length of synapses, and mature bouton numbers were not affected by the dABCA knockdown [8]. Drosophila NMJ synapses are glutamatergic and similar to those in the vertebrate CNS [19]. Boutons contain multiple active zones that are neurotransmitter release sites, and each of these apposes a glutamate receptor (GluR) cluster [19]. The presynaptic boutons at larval NMJs are eventually surrounded by a subsynaptic reticulum (SSR), which contains neurotransmitter receptors, scaffolding proteins, and postsynaptic signaling complexes [20]. This glutamatergic NMJ is known to contain five AMPA-like GluR subunits, GluRIIA to E. Therefore, we used the cell adhesion molecule Brp as an active zone marker and GluRIIA to monitor the distribution of GluR in the NMJ (Fig. 3). The BRP-positive area (µm2)/NMJ area (µm2) in elav>dABCA-IR666-672 flies was significantly larger (by 1.63-fold) than that in control elav>GFP-IR flies (Fig. 3C), while those in elav>dABCA-IR997-1106 and elav>dABCA-IR997-11062copy flies were slightly
larger
(Fig.
3C).
The
BRP-positive
area/NMJ
area
(µm2)
in
elav>dABCA-IR997-1106 and elav>dABCA-IR997-11062copy flies were significantly larger than that in control flies (Fig. 3D). These results indicate that the dABCA knockdown increases the number or density of active zones at the NMJ. Moreover, we found that 9
the GluRIIA-positive area (µm2)/NMJ area (µm2) in dABCA RNAi flies appeared to be slightly increased (Fig. 3E). 3.4. Loss-of-function mutation in hpo enhanced the hyperactivity phenotype induced by the pan-neuron-specific dABCA knockdown In analyses of adult flies with DAM (from Trikinetics), we previously reported that the pan-neuron-specific dABCA knockdown induced the early onset of evening activity followed by relatively high activity throughout the day [8]. The circadian wake and sleep rhythm itself established in a 12-h light-dark cycle did not appear to be affected by the dABCA knockdown because circadian rhythmicity was maintained in constant darkness (data not shown). In order to identify the genes that genetically interact with the dABCA gene, we focused on the hyperactivity phenotype of pan-neuron-specific dABCA knockdown flies that may be easily monitored by the Drosophila activity assay. In the activity monitoring assay, we used one-day-old virgin male flies and showed five-day average activity (Fig. 4) because dABCA-knockdown flies show hyperactivity on each day from 1 to 5 days (Supplementary Fig. S1). We performed a genetic test by crossing dABCA knockdown flies and the mutant flies of hpo, a candidate gene, as described in the Introduction section. We also crossed dABCA knockdown flies and the knockdown flies of rhomboid (rho), the epidermal growth factor-receptor (EGFR) pathway-related gene, as a negative control [21-24]. The knockdown of rho did not affect the hyperactivity phenotype induced by the pan-neuron-specific dABCA knockdown, suggesting the lack of a genetic interaction between dABCA and rho (Fig. 4C). In contrast, the strong enhancement in hyperactivity induced by the knockdown of dABCA was observed by crossing with the loss-of-function mutant, hpoKC202 (Fig. 4A, D), while the hpoKC202 mutation alone slightly enhanced this activity (Fig. 4B, E). Enhanced hyperactivity was also observed by crossing with hpoKS240 (Fig. 4A, D), while the hpoKS240 mutation alone suppressed this activity (Fig. 4B, E). These results suggest that dABCA genetically interacts with hpo. 3.5. Loss-of-function mutation in hpo suppressed the increased satellite bouton number phenotype induced by the pan-neuron-specific dABCA knockdown We also examined the effects of the hpo mutation on the aberrant morphology of larval 10
NMJs induced by the dABCA knockdown. The increased number of satellite boutons induced by the dABCA knockdown was effectively suppressed by the loss-of-function mutation in hpoKC202 (Fig. 5A, 5B). The other loss-of-function mutation in hpoKS240 also showed slight suppression (Fig. 5A, 5B). A half-dose reduction in hpo itself did not affect the number of satellite boutons for both alleles (Fig. 5B). Therefore, in this assay system, we confirmed the genetic interaction between dABCA and hpo. 3.6. Loss-of-function mutations in wts and yki slightly suppressed the increased satellite bouton number phenotype induced by the pan-neuron-specific dABCA knockdown To clarify whether canonical Hippo pathway genes other than the hpo gene also genetically interact with dABCA, we examined the effects of the wts and yki mutations on the aberrant bouton morphology phenotype. The loss-of-function mutation in wts slightly suppressed the increased satellite bouton number phenotype induced by the dABCA knockdown (Fig. 6A, B). We also crossed the dABCA knockdown with the loss-of-function mutant of yki, a downstream target of the Hippo pathway that is negatively regulated by the Hippo pathway. Although we expected the enhancement associated with the increased satellite bouton number phenotype, a slight reduction was observed in the satellite bouton number (Fig. 6A, B). The half-dose reduction in wts or yki did not affect the number of satellite boutons (Fig. 6B). These results suggest that dABCA genetically interacts with wts, but not yki. Therefore, the canonical Hippo pathway may not be related to the phenotype induced by the pan-neuron-specific dABCA knockdown. 3.7. Knockdown of ena enhanced the increased satellite bouton number phenotype induced by the pan-neuron-specific dABCA knockdown A recent study reported that the yki-independent Strip-Hippo pathway plays a role in synaptogenesis in Drosophila [10]. Strip, the Drosophila homolog of mammalian Strip 1 and 2, is involved in the strain-interacting phosphatase and kinase (STRIPAK) complex, a negative regulator of hpo. Strip genetically interacts with Enabled (ena), an actin assembly/elongation factor. Therefore, Ena is the presumed downstream target of Hippo signaling to modulate local actin organization at synaptic termini. This regulation occurs independently of the transcriptional coactivator Yki, the canonical downstream 11
target of the Hippo pathway [10]. To assess the possible involvement of dABCA in the Hippo pathway in which Ena rather than Yki is the downstream target of Hippo signaling, we examined the effects of the ena knockdown on the increased satellite bouton number phenotype induced by the dABCA knockdown by crossing dABCA knockdown flies with ena knockdown flies (Fig. 7A). The double knockdown of dABCA and ena resulted in a slightly higher satellite bouton number than the dABCA knockdown alone, indicating a genetic interaction between dABCA and ena (Fig. 7B). The knockdown of ena alone exerted no effect on the number of satellite boutons (Fig. 7B). These results suggest that dABCA is involved in the Hippo-Ena pathway rather than in the Hippo-Yki pathway in synapses at the NMJ. 4. Discussion We previously reported that the knockdown of dABCA in pan neurons induced increases in social space with the closest neighbors and the early onset of evening activity followed by hyperactivity in adult flies, and also increased satellite bouton numbers at the NMJ of third instar larvae [8]. In the present study, we also found increases in the number or density of active zones at the NMJ in dABCA knockdown flies, suggesting increases in neurotransmitter release in dABCA knockdown flies that may result in abnormal increases in neural activity. This may explain the hyperactivity phenotype of dABCA knockdown flies. In contrast, we also found defects in learning ability in dABCA knockdown flies. Since we did not detect an aberrant morphology of MB, the limited learning abilities of dABCA knockdown flies may be caused by a functional defect rather than an abnormal MB morphology. Further analyses are needed to elucidate the underlying mechanisms. In any event, these phenotypes are similar to the defects observed in human ASD patients, suggesting that the established dABCA knockdown fly is a promising model for ASD. The asymmetric distribution of phospholipids across the plasma membrane and its local changes are important for cell functions. Substrates transported by ABCA family members
include
lipids,
glucosylceramide
for
ABCA12,
cholesterol
and
phosphatidylcholine (PC) for ABCA1, PC and phosphatidylglycerol for ABCA3, N-retinylidene-phosphatidylethanolamine for ABCA4, and PC for ABCA7. Considering their amino acid sequence homology, ABCA13 is also predicted to transport lipid 12
molecules [1]. Since the CNS represents the second most lipid-rich area in higher organisms after adipose tissue, we assume that ABCA family transporters are of critical importance for the integrity of the CNS. Recent evidence links ABCA family transporters to the maintenance of brain lipid homeostasis and neurodegenerative diseases, as suggested for other members of the ABC protein family [25]. Although transport substrates for dABCA are not yet known, dABCA, a homologue of ABCA13 appears to transport lipid molecules, as noted for human ABCA family transporters, and may play a role in maintaining brain lipid homeostasis in Drosophila. Genetic interaction studies with Drosophila mutants of the ASD candidate gene rugose (rg), the Drosophila homologue of mammalian neurobeachin, show that rg genetically interacts with components of the EGFR- and Notch-mediated signaling pathways [26]. Based on these findings, we speculated that the disruption of the asymmetric distribution of phospholipids across the plasma membrane in the brain induced by the dABCA knockdown exerts effects on some signaling pathways. In the present study, we identified hpo as a genetic interactant of dABCA. The loss-of-function of wts, another component of the Hippo pathway, slightly suppressed the increased satellite bouton number phenotype induced by the dABCA knockdown, as observed with the loss-of-function mutations in hpo. However, the loss-of-function mutation in yki, one of the most downstream components negatively regulated by the canonical Hippo pathway via its phosphorylation, slightly suppressed the increased satellite bouton number phenotype. We also performed a genetic test between dABCA and ena, a candidate downstream component that is negatively regulated by the Yki-independent Strip-Hippo pathway. The double knockdown of dABCA and, as expected, ena slightly enhanced the increased satellite bouton number phenotype induced by the dABCA knockdown alone. These results suggest that dABCA is involved in the Hippo-Ena pathway rather than in the Hippo-Yki pathway in synapses at the NMJ. However, we cannot exclude the possibility that some unknown downstream component of the Hippo pathway other than yki and ena is involved in this pathway at the NMJ. Further studies are needed to clarify this point. In any event, we herein identified a novel genetic link between dABCA and hpo. These results will contribute to clarifying the pathogenesis of ASD as well as the development of disease-modifying therapy.
13
Competing financial interests The authors declare no competing interests in relation to the work described. Acknowledgments We thank the Bloomington Drosophila Stock Center and Vienna Drosophila Genetic Resource Center for the fly lines. This research was partially supported by the JSPS Core-to-Core Program, Asia-Africa Science Platforms B, and JSPS KAKENHI Grant Number JP19K06659. We also thank Medical English Service for the English language review.
14
References [1] M. Tomioka, Y. Toda, J. Kurisu, Y. Kimura, M. Kengaku, K. UEDA, The effects of neurological disorder-related codon variations of ABCA13 on the function of the ABC protein. Biosci. Biotechnol. Biochem. 76 (2012) 2289–2293. [2] I.Iossifov, M. Ronemus, D. Levy, Z. Wang, I. Hakker, J. Rosenbaum, B. Yamrom, Y.H. Lee, G. Narzisi, A. Leotta, J. Kendall, E. Grabowska, B. Ma, S. Mark, L. Rodgers, A. Stepansky, J. Troge, P. Andrews, M. Bekritsky, K. Pradhan, E. Ghiban, M. Kramer, J. Parla, R. Demeter, L.L. Fulton, R.S. Fulton, V.J. Magrini, K. Ye, J.C. Darnell, R.B. Darnell, E.R. Mardis, R.K. Wilson, M.C. Schatz, W.R. McCombie, M. Wigler, De Novo gene disruptions in children on the autistic spectrum. Neuron 74 (2012) 285–299. [3] I.Iossifov, B.J. O’Roak, S.J. Sanders, M. Ronemus, N. Krumm, D. Levy, H.A. Stessman, K.T. Witherspoon, L. Vives, K.E. Patterson, J.D. Smith, B. Paeper, D.A. Nikerson, J. Dea, S. Dong, L.E. Gonzalez, J.D. Mandell, S.M. Mane, M.T. Murtha, C.A. Sullivan, M.F. Walker, Z. Waqar, L. Wei, A.J. Willsey, B. Yamrom, Y.H. Lee, E. Grabowska, E. Dalkic, Z. Wang, S. Marks, P. Andrews, A. Leotta, J. Kendall, I. Hakker, J. Rosenbaum, B. Ma, L. Rodgers, J. Troge, G. Narzisi, S. Yoon, M.C. Schatz, K. Ye, W.R. McCombie, J. Shendure, E.E. Eichler, M.W. State, M. Wigler, The contribution of de novo coding mutations to autism spectrum disorder. Nature 515 (2014) 216–221. [4] S. De Rubeis, X. He, A.P. Goldberg, C.S. Poultney, K. Samocha, A.E. Cicek, Y. Kou, L. Liu, M. Fromer, S. Walker, et al. Synaptic, transcriptional and chromatin genes disrupted in autism. Nature 515 (2014) 209–215. [5] H.M. Knight, B.S. Pickard, A. Maclean, M.P. Malloy, D.C. Soares, A.F. McRae, A. Condie, A. White, W. Hawkins, K. McGhee, M. van Beck, D.J. Maclntyre, J.M. Starr, I.J. Deary, P.M. Visscher, D.J. Deary, P.M. Visscher, D.J. Porteous, R. E. Cannon, D. St. Clair, W.J. Muir, D.H. Blackwood, A cytogenetic abnormality and rare coding variants identify ABCA13 as a candidate gene in schizophrenia, bipolar disorder, and depression. Am. J. Hum. Genet. 85 (2009) 833–846. [6] K. Yoshida, Y. Go, I. Kushima, A. Toyoda, A. Fujiyama, H. Imai, N. Saito, A Iriki, N. Ozaki, M. Isoda, Single-neuron and genetic correlates of autistic behavior in macaque. Sci. Adv. 2 (2016) e1600558. 15
[7] S. Iritani, Y. Torii, C. Habuchi, H. Sekiguchi, H. Fujishiro, M. Yoshida, Y. Go, A. Iriki, M. Isoda, N. Ozaki, The neuropathological investigation of the brain in a monkey model of autism spectrum disorder with ABCA13 deletion. Int. J. Dev. Neurosci. 71 (2018) 130–139. [8] I. Ueoka, H. Kawashima, A. Konishi, M. Aoki, R. Tanaka, H. Yoshida, T. Maeda, M. Ozaki, M. Yamaguchi, Novel Drosophila model for psychiatric disorders including autism spectrum disorder by targeting of ATP-binding cassette protein A. Exp. Neurol. 300 (2018) 51–59. [9] B. Zhao, K. Tumaneng, K.L. Guan, The Hippo pathway in organ size control, tissue regeneration and stem cell self-renewal. Nat. Cell Biol. 13 (2011) 877–883. [10] C. Sakuma, Y. Saito, T. Umehara, K. Kamimura, N, Maeda, T.J. Mosca, M. Miura, T. Chihara. The Strip-Hippo pathway regulates synaptic terminal formation by modulating actin organization at the Drosophila neuromuscular synapses. Cell Rep. 16 (2016), 2289-2297. [11] Y. Azuma, T. Tokuda, Y. Kushimura, I. Yamamoto, I. Mizuta, T. Mizuno, M. Nakagawa, M. Ueyama, Y. Nagai, Y. Iwasaki, M. Yoshida, D. Pan, H. Yoshida, M. Yamaguchi, Hippo, Drosophila MST, is a novel modifier of motor neuron degeneration induced by knockdown of Caz, Drosophila FUS, Exp. Cell Res. 371 (2018) 311–321. [12] Y. Kushimura, Y. Azuma, I. Mizuta, Y. Muraoka, A. Kyotani, H. Yoshida, T. Tokuda, T. Mizuno, M. Yamaguchi, Loss-of-function mutation in Hippo suppressed enlargement of lysosomes and neurodegeneration caused by dFIG4 knockdown. Neuroreport 29 (2018) 856–862. [13] B. Gerber, R. Biernacki, J. Thum, Odor-taste learning assays in Drosophila larvae, Cold Spring Harb. Protoc. 8 (2013) 213–223. [14] S. Jantrapirom, L. Lo Piccolo, H. Yoshida, M. Yamaguchi, A new Drosophila model of Ubiquilin knockdown shows the effect of impaired proteostasis on locomotive and learning abilities, Exp. Cell Res. 362 (2017) 461–471. [15] D.S. Brooks, K. Vishal, J. Kawakami, S. Bouyain, R. Erika, Optimization of wrMTrck to monitor Drosophila larval locomotor activity, J. Insect Physiol. 93-94 (2016) 11–17. [16] C.I. Nussbaum-Krammer, M.F. Neto, R.M. Brielmann, J.S. Pederson, R.I. Morimoto, Investigating the spreading and toxicity of prion-like proteins using the 16
metazoan model organism C. elegans, J. Vis. Exp. 95 (2015) 52321. [17] F. Foti, F. Piras, S. Vicari, L. Mandolesi, L. Petrosini, D. Menghini, Observational learning in low-functioning children with autism spectrum disorders: A behavioral and neuroimaging study. Front. Psychol. 9 (2019) 1–13. [18] M. Kurusu, T. Awasaki, L.M. Masuda-Nakagawa, H. Kawauchi, K. Ito, K. Furukubo-Tokunaga, Embryonic and larval development of the Drosophila mushroom bodies: concentric layer subdivisions and the role of fasciclin II. Development 129 (2002) 409–419. [19] K.P. Menon, R.A. Carrillo, K. Zinn, Development and plasticity of the Drosophila larval neuromuscular junction. Wiley Interdiscip. Rev. Dev. Biol. 2 (2013) 647– 670. [20] G. Lee, T.L. Schwarz, Filamin, a synaptic organizer in Drosophila, determines glutamate receptor composition and membrane growth. Elife 5 (2016) e19991. [21] P. Wee, Z. Wang, Epidermal growth factor receptor cell proliferation signaling pathways. Cancers (Basel) 9 (2017) 1–45. [22] J.D. Wasserman, S. Urban, M. Freeman, A family of rhomboid-like genes: Drosophila rhomboid-1 and roughoid/rhomboid-3 cooperate to activate EGF receptor signaling. Genes Dev. 14 (2000) 1651–1663. [23] B.-Z. Shilo, Regulating the dynamics of EGF receptor signaling in space and time. Development 132 (2005) 4017–4027. [24] S. Yogev, E.D. Schejter, B.Z. Shilo, Drosophila EGFR signalling is modulated by differential compartmentalization of Rhomboid intramembrane proteases. EMBO J. 27 (2008) 1219–1230. [25] A.P. Piehler, M. Özcurumez, W.E. Kaminski, A-subclass ATP-binding cassette proteins in brain lipid homeostasis and neurodegeneration. Front. Psychiatry 3 (2012) 1–16. [26] A. Wise, L. Tenezaca, R.W. Fernandez, E. Schatoff, J. Flores, A. Ueda, X. Zhong, C.F. Wu, A.F. Simon, T. Venkatesh, Drosophila mutants of the autism candidate gene neurobeachin (rugose) exhibit neuro-developmental disorders, aberrant synaptic properties, altered locomotion, and impaired adult social behavior and activity patterns. J. Neurogenet. 29 (2015) 135-143.
17
Figure legends Fig. 1. The knockdown of dABCA exerted a marginal effect on larval locomotive ability evaluated by the crawling assay. (A) Diagram of path parameters, including the length (blue) and distance (grey) from start to finish measured in wrMTrck. (B) Traces of larval path routes. Third instar larvae of the indicated genotypes were monitored for 60 sec and path lengths were analyzed using ImageJ and wrMTrck. Speed (mm/sec) (C) and distance from start to finish (mm) (D). (C) Speed ware analyzed with one-way ANOVA followed by Dunnet’s post hoc test. (D) Distance from start to finish ware analyzed with Kruscal-Wallis test followed by Dunn’s post hoc test. Bar graph represents the mean ± SEM. Box and whisker plots represent the minimum, 25th percentile, median, 75th percentile and maximum. *p < 0.05. The genotype of each fly was
as
follows:
elav>dABCA-IR666-672 elav>dABCA-IR997-1106 elav>dABCA-IR997-1106
elav>GFP-IR (w/Y; (w/Y; 2copy
(w/Y;
(w/Y;
UAS-GFP-IR/+;
+;
elav-Gal4/UAS-dABCA-IR666-672);
UAS-dABCA-IR997-1106/+;
elav-Gal4/+); elav-Gal4/+);
UAS-dABCA-IR997-1106/UAS-dABCA-IR997-1106;
elav-Gal4/+). Fig. 2. The knockdown of dABCA induced defects in larval learning ability. (A) Principle of the odor-taste learning test. TRAINING 1: In one of the groups, n-amyl acetate (AM) was added with a sucrose reward (+) and 1-octanol (OCT) was subsequently added without a reward (AM+/OCT). TRAINING 2: The other group received reciprocal training (AM/OCT+). After training exposure, larvae were tested for their choice between AM and OCT. Stronger preferences for AM after AM+/OCT training than after AM/OCT+ training reflected associative learning and were quantified by the learning index (LI). (B) Larval preference for AM versus OCT (1 = all prefer AM; −1 = all prefer OCT) was analyzed with the Friedman test followed by Dunn’s post hoc test. Box and whisker plots represent the minimum, median, and maximum values. (C) Normalized AM and OCT learning indexes were averaged for LI and analyzed with a one-way ANOVA followed by Dunnett’s post hoc test. Bar Graph plots the mean ± SEM. *p<0.05, **p<0.01, ***p<0.001 and ns means no significance (p>0.05). (D) Larval and adult mushroom bodies (MBs). Third instar larval brains and adult brains stained with anti-FASII IgG (gray). Brains are marked with a white line and larval MBs are marked with a white dashed line. α, adult dorsal lobes; β and γ, adult 18
medial lobes; cx, calyx; DL, dorsal lobe; KCs, Kenyon cells; ML, medial lobe; OL, optic lobe; Ped, peduncle; VNC, ventral nerve cord. Reconstruction of optical sections. Bars indicate 50 µm. The genotype of each fly was as follows: elav>GFP-IR (w/Y; UAS-GFP-IR/+;
elav-Gal4/+);
elav>dABCA-IR666-672
elav-Gal4/UAS-dABCA-IR666-672); UAS-dABCA-IR997-1106/+;
(w/Y;
elav>dABCA-IR997-1106
elav-Gal4/+);
elav>dABCA-IR997-1106
+; (w/Y;
2copy
(w/Y;
UAS-dABCA-IR997-1106/UAS-dABCA-IR997-1106; elav-Gal4/+). Fig. 3. Knockdown of dABCA increased the number or density of active zones in synapses at the NMJ. (A) Super-resolution images of synapses at muscle 4 NMJs of third instar larvae. A representative image of anti-HRP staining, anti-BRP staining, and merged images. Bars indicate 1 µm. (B) Confocal images of synapses at muscle 4 NMJs of third instar larvae. A representative image of anti-HRP staining, anti-GluRIIA staining, and merged images. Bars indicate 10 µm. The total BRP-positive area per synapse area at the NMJ (C), total BRP-positive signal number per synapse area at the NMJ (D), and GluIIA-positive area per synapse area at the NMJ (E). (C, D, E) Each data set was analyzed with a one-way ANOVA followed by Dunnett’s post hoc test. Graphs plot the mean ± SEM. *p<0.05, **p<0.01, ***p<0.001 and ns means no significance (p>0.05). The genotype of each fly was as follows: elav>GFP-IR (w/Y; UAS-GFP-IR/+;
elav-Gal4/+);
elav>dABCA-IR666-672
elav-Gal4/UAS-dABCA-IR666-672); UAS-dABCA-IR997-1106/+;
(w/Y;
elav>dABCA-IR997-1106
elav-Gal4/+);
elav>dABCA-IR997-1106
+; (w/Y;
2copy
(w/Y;
UAS-dABCA-IR997-1106/UAS-dABCA-IR997-1106; elav-Gal4/+). Fig. 4. Mutations in hpo enhanced the hyperactivity phenotype induced by the dABCA knockdown, whereas the knockdown of rho did not. Five-day average activity plotted as the number of infrared beam crossings measured per 15 min of elav>dABCA-IR997-1106/GFP-IR elav-Gal4/+),
(w/Y;
UAS-dABCA-IR997-110/UAS-GFP-IR;
elav>dABCA-IR997-1106/rho-IR
UAS-dABCA-IR997-1106/+;
elav-Gal4/UAS-rho-IR
elav>dABCA-IR997-1106/rho-IR elav-Gal4/UAS-rho-IR UAS-dABCA-IR997-1106/+;
(VDRC51952) (VDRC51952)),
(VDRC51953)(w/Y;
(VDRC51953))
(A),
elav-Gal4/+), 19
(w/Y; and
UAS-dABCA-IR997-1106/+;
elav>dABCA-IR997-1106 KS240
elav>dABCA-IR997-1106/hpo
(w/Y; (w/Y;
UAS-dABCA-IR997-110/ hpoKS240; elav-Gal4/+), and elav>dABCA-IR997-1106/hpoKC202 (w/Y; UAS-dABCA-IR997-110/ hpoKC202; elav-Gal4/+) (B), and elav>+ (w/Y; +; elav-Gal4/+), elav>hpoKS240 (w/Y; hpoKS240/+; elav-Gal4/+), and elav>hpoKC202 (w/Y; hpoKC202/+; elav-Gal4/+) (C). The average total crossing activities during the day (21:00-21:00)
of
elav>dABCA-IR997-1106/GFP-IR,
elav>dABCA-IR997-1106/rho-IR
(VDRC51952), and elav>dABCA-IR997-1106/rho-IR (VDRC51953) (D) were analyzed with the Kruskal-Wallis test followed by Dunn’s post hoc test, while those of elav>dABCA-IR997-1106/hpoKS240
elav>dABCA-IR997-1106, elav>dABCA-IR997-1106/hpo
KC202
(E), and elav>+, elav>hpo
KS240
and elav>hpo
and KC202
(F)
were analyzed with a one-way ANOVA followed by Dunnett’s post hoc test. Box and whisker plots represent the minimum, 25th percentile, median, 75th percentile and maximum. Bar graphs represent the mean ± SEM. *p<0.05, **p<0.01 and ns means no significance (p>0.05). Fig. 5. Mutations in hpo suppressed the increased satellite bouton phenotype induced by the dABCA knockdown. (A) Confocal images of synapses at muscle 4 NMJs of third instar larvae. A representative merged image of anti-HRP staining and anti-Dlg staining. Bars indicate 10 µm. The inset in each panel shows a higher magnification image of the boxed area. (B) The number of satellite boutons was analyzed with the Kruskal-Wallis test followed by Dunn’s post hoc test. Box and whisker plots represent the minimum, 25th percentile, median, 75th percentile and maximum. **p<0.01, ****p<0.0001 and ns means no significance (p>0.05). The genotype of each fly is as follows: elav>GFP-IR (w/Y; UAS-GFP-IR/+; elav-Gal4/+); elav>dABCA-IR997-1106/hpoKS240 (w/Y; UAS-dABCA-IR997-110/ hpoKS240; elav-Gal4/+); elav>dABCA-IR997-1106/hpoKC202 (w/Y; UAS-dABCA-IR997-110/ hpoKC202; elav-Gal4/+); elav>dABCA-IR997-1106 (w/Y; UAS-dABCA-IR997-1106/+; elav-Gal4/+); elav>hpoKS240 (w/Y; hpoKS240/+; elav-Gal4/+); elav>hpoKC202 (w/Y; hpoKC202/+; elav-Gal4/+). Fig. 6. Mutation in wts suppressed the increased satellite bouton phenotype induced by the dABCA knockdown, whereas that of yki did not. (A) Confocal images of synapses at muscle 4 NMJs of third instar larvae. A representative merged image of anti-HRP staining and anti-Dlg staining. Bars indicate 10 µm. The inset in each panel shows a higher magnification image of the boxed area. (B) The number of 20
satellite boutons was analyzed with the Kruskal-Wallis test followed by Dunn’s post hoc test. Box and whisker plots represent the minimum, 25th percentile, median, 75th percentile and maximum. ****p<0.0001 and ns means no significance (p>0.05). The genotype of each fly was as follows: elav>GFP-IR (w/Y; UAS-GFP-IR/+; elav-Gal4/+);
elav>dABCA-IR997-110/ X1
elav-Gal4/wts );
wtsX1
elav>dABCA-IR997-110/yki
B5
(w/Y; (w/Y;
UAS-dABCA-IR997-110/
+; B5
UAS-dABCA-IR997-110/yki ;
elav-Gal4/+); elav>dABCA-IR997-1106 (w/Y; UAS-dABCA-IR997-1106/+; elav-Gal4/+); elav>wtsX1 (w/Y; +; elav-Gal4/ wtsX1); elav>ykiB5 (w/Y; ykiB5/+; elav-Gal4/+). Fig. 7. Knockdown of ena enhanced the increased satellite bouton phenotype induced by the dABCA knockdown. (A) Confocal images of synapses at muscle 4 NMJs of third instar larvae. A representative merged image of anti-HRP staining and anti-Dlg staining. Bars indicate 10 µm. The inset in each panel shows a higher magnification image of the boxed area. (B) The number of satellite boutons was analyzed with the Kruskal-Wallis test followed by Dunn’s post hoc test. Box and whisker plots represent the minimum, 25th percentile, median, 75th percentile and maximum. *p<0.05 and ns means no significance (p>0.05). The genotype of each fly is as
follows:
elav>GFP-IR
(w/Y;
UAS-GFP-IR/+;
elav-Gal4/+);
elav>dABCA-IR997-110/ena-IR (w/Y; UAS-dABCA-IR997-110/+; elav-Gal4/ UAS-ena-IR); elav>dABCA-IR997-110/GFP-IR (w/Y; UAS-dABCA-IR997-110/UAS-GFP-IR; elav-Gal4/+); elav>ena-IR (w/Y; +; elav-Gal4/UAS-ena-IR). Supplementary Fig. S1. Hyperactivity phenotype induced by the dABCA knockdown. Five-day activity plotted as the number of infrared beam crossings measured per 15 min by male (A) or female (B) flies with elav>GFP-IR (w/Y; UAS-GFP-IR/+;
elav-Gal4/+)
or
elav>dABCA-IR997-1106
2copy
(w/Y;
UAS-dABCA-IR997-1106/UAS-dABCA-IR997-1106; elav-Gal4/+). Average total crossing activity during the indicated day (21:00-21:00) for male (C) or female (D) was analyzed with two-way ANOVA followed by Dunnett’s post hoc test. Graphs were represented by the mean ± SEM. **p<0.01, **** p<0.0001.
21
Highlights
・Neuron-specific knockdown of dABCA increases active zone at NMJ. ・Hippo mutation enhanced the hyperactivity phenotype induced by dABCA knockdown. ・Hippo mutation suppressed the decreased bouton number phenotype induced by dABCA knockdown. ・Novel genetic link between psychiatric disorder-causing gene and hippo.
Conflict of Interest Form The authors declare no conflict interest in relation to the work described.