Analysis of Succinyl CoA synthetase alpha subunit
deficiency in Drosophila
47 Summary
Succinyl-CoA synthetase/ligase (SCS) is a mitochondrial enzyme that catalyzes the reversible process from succinyl-CoA to succinate and free coenzyme A in TCA cycle.
SCS deficiencies are implicated in mitochondrial hepatoencephalomyopathy in humans. To investigate the impact of SCS deficiencies in Drosophila, I generated a null mutation in Scs alpha subunit (Scsα) using the CRISPR/Cas9 system. The null mutants, designated ScsαKO, accumulated a high level of succinyl CoA, a substrate for the enzyme, and also affected many other metabolites in TCA cycle and glycolysis, indicating that the energy metabolism was impaired. Unlike SCS deficiencies in humans, there was no reduction in lifespan, indicating that Scsα is not critical for viability in Drosophila. However, they showed developmental delays, locomotor activity defects, and reduced survival rates under starvation. I also found that glycogen
breakdown occurred at the beginning of the third instar larval stage, suggesting that the mutant flies utilized the stored energy for survival. These results suggested that SCSα is essential for proper energy metabolism in Drosophila. The ScsαKO flies should be useful as a model to understand the physiological role of SCS as well as the
pathophysiology of SCSα deficiency.
48 Introduction
Energy metabolism is essential for all organisms, and dysfunctions of metabolic enzymes could impair many aspects of physiology including development, behavior and lifespan. Succinyl CoA synthetase/ligase (SCS) is a mitochondrial enzyme that
catalyzes the reversible process from succinyl-CoA to succinate and free coenzyme A in TCA cycle. SCS is a heterodimeric enzyme composed of an invariant α subunit encoded by SUCLG1 and a β subunit that determines the enzyme's nucleotide specificity: the GTP-specific β subunit encoded by SUCLG2 and the ATP-specific subunit encoded bySUCLA2 (Johnson et al. 1998).
It has been reported that mutations in human SCSβ (SUCLA2) cause mitochondrial DNA depletion syndrome with a severe progressive childhood-onset
encephalomyopathy, and methylmalonic aciduria (Elpeleg et al. 2005; Lamperti et al.
2012; Ostergaard, Hansen, et al. 2007). Patients with mutations in SUCLA2 have Leigh’s or Leigh-like symptoms with hypotonia, muscle weakness, and sensorineural hearing loss (Carrozzo et al. 2007; Morava et al. 2009). Similarly, mutations in human SCSα (SUCLG1) also cause mitochondrial DNA depletion syndrome with severe lactic acidosis, and elevated pyruvate in blood and urine, which results in neonatal
death(Ostergaard et al. 2007). However, the pathophysiology of SCS deficiency remains elusive.
Drosophila is a powerful model organism to study the function of human disease-related genes. Metabolic enzymes in TCA cycle are generally conserved between flies and humans. With respect to the SCS subunits, the Drosophila genome contains Scsα, Sucb and skap, which encode the homologs of human SUCLG1,
49
SUCLG2 and SUCLA2, respectively: the amino acid sequence identities were 70%
(Fig1), 57% and 61%, and the similarities were 90%, 85% and 91%, respectively. In this study, I have generated a Drosophila model of Scsα deficiency using the
CRISPR/Cas9 system(Kondo and Ueda 2013), and characterized their phenotype. I found that the SCSα mutants accumulated a high level of succinyl CoA, a substrate for the enzyme, indicating that SCS is defective. The SCSα deficiency caused
developmental delays, locomotor defects and reduced survival rates under starvation.
The developmental delay was associated with breakdown of glycogen during larval period. These results demonstrate that SCSα is essential for proper energy
metabolism in Drosophila.
.
50
Materials and methods
Fly strains and maintenance
Flies were reared on a standard glucose-yeast-agar medium containing propionic acid and n-butyl p-hydroxybenzoate as a mold inhibitor, in temperature-controlled environmental chambers at 25 °C throughout development. Unless otherwise stated, y1 w67c23 (y w) strain was used as a control.
SsadhKO was generated by the CRISPR/Cas9 system(Kondo and Ueda 2013) as described in Chapter 2. The following primers were used;
5’ UTR _Forward: 5’-CTTCGCATCCGCCACCGTCATGTT -3’, 5’ UTR _Reverse: 5’-AAACAACATGACGGTGGCGGATGC -3’;
3’ UTR _Forward: 5’-CTTCGTCCAGTGGTCTCCATCATA -3’, 3’ UTR _Reverse: 5’-AAACTATGATGGAGACCACTGGAC -3’.
The deletion was confirmed by genomic PCR with the following primers;
Forward (GF): 5’-GCAGATAAGAGAAACGCCTCTAAC -3’, Reverse (GR): 5’-
ATGTCTCTATGATCGGTTTCTTGTC-3’. All fly stocks were each backcrossed to y w strain for six generations before using for all experiments.
Longevity assay
20 flies were reared in a glass vial containing standard glucose-yeast-agar medium.
The number of dead flies was counted every 2 days, meanwhile, the flies were transferred to a fresh media. At least 100 flies were used for longevity assay for each genotype.
51
Quantitative reverse transcription-polymerase chain reaction (RT-qPCR) RT-qPCR was performed as described in Chapter 2. The following primers were used; CG6255-Forward (RF): 5’- -3’, CG6255-Reverse (RR): 5’- -3’, rp49-Forward:
5’-AAGATCGTGAAGAAGCGCAC-3’, rp49-Reverse:
5’-TGTGCACCAGGAACTTCTTG-3’.
Measurement of developmental rate
Flies were allowed to lay eggs for 24 hours in a vial containing the media and the numbers of pupae and eclosed adult flies were counted every 24 hrs. At least 10 vials per genotype were used in these experiments. To measure the organ growth, oviposited eggs were collected within three hours, allowed to develop, and then wing discs were dissected from the third instar larvae at a 24 hr-interval. Dissection was performed in cooled phosphate buffered saline (PBS), and photographed using a CCD camera equipped on a Leica MZ16 F Fluorescence stereomicroscope. Wing disc area was measured using Image J. At least 10 larvae were used to determine an average size of wing discs at a given developmental time after egg laying.
Starvation and diet restriction
To determine the survival rate of flies under starvation, 3-day-old male flies were kept in vials containing 2 % agar, and the number of dead flies was counted every 8 hrs. At least 80 flies per each genotype were used. Diet restriction was performed using a poor media, diluted to 20 % of the normal food.
Statistical analysis
52
Statistical analyses for lifespans were performed using a log-rank test. For all other experiments, means comparisons were analyzed using a Student’s t-test.
53 Results
Lifespan of ScsαKO mutant flies
To investigate the molecular and physiological function of Scsα in vivo, I generated a null mutant ScsαKO with CRISPR/Cas9 system. Since mutations in genes involved in energy metabolism often affect lifespan, I measured lifespan under standard condition.
However, there was no reduction of lifespan in ScsαKO compared to control flies (Fig. 2).
The mutant female flies lived even slightly but significantly longer than wild-type females (Fig. 2A), suggesting that there may be some beneficial effects associated with ScsαKO. Therefore, Scsα is not essential for survival under standard condition.
Effects of SCSα deficiency on TCA cycle and glycolysis metabolites
SCSα is known to be required for the conversion of succinyl-CoA into succinate and CoA in TCA cycle. Therefore, the mutation in Scsα is likely to impair the function of TCA cycle, which may be detectable as altered levels of the relevant metabolites. In addition, glycolytic metabolites may also be affected, since the final product pyruvate is a
substrate for TCA cycle. I measured the levels of intermediates in glycolysis and TCA cycle using a liquid chromatography-mass spectrometry (LC−MS/MS) (Fig. 3). I found that succinyl-CoA, a substrate for SCS was highly accumulated in ScsαKO compared to control flies: the amount of succinyl-CoA was approximately seven-times and four-times higher than those of the controls for females and males, respectively (Figs. 3B and D).
The altered levels of succinyl-CoA indicated that the mutants were defective in
catalyzing the conversion of succinyl-CoA to succinate. However, the level of succinate, another substrate for SCS was also slightly higher than that of control flies for both
54
sexes (Figs. 3B and D). Succinate can be provided by from glutamate through 4-aminobutanoic acid (GABA) and succinic semialdehyde. I found that the level of GABA was decreased in the mutant flies by 50 % compared to control flies, suggesting that this pathway may contribute to provide succinate in the absence of SCSα (Fig. 4).
The mutant flies contain relatively high levels of other metabolites in glycolysis and TCA cycle, which include 3PG, PEP, pyruvate, acetyl-CoA, and 2-oxoglutarate, indicating that the mutation affected glycolysis and TCA cycle (Fig. 3). These results indicate that deficiency of SCSα disrupts energy metabolism.
Developmental delay and locomotor activity defects in ScsαKO
Both ScsαKO and Scsα1 flies were viable, and did not show any obvious morphological defects. Since disruption of energy metabolism may affect physiological traits, I focused on developmental rate, locomotor activity and lifespan. I noticed that there was a significant developmental delay for both ScsαKOand Scsα1: mean egg-to-eclosion periods of ScsαKO, Scsα1 and control flies were 12.17 ± 0.03, 10.97 ± 0.05 and 9.27 ± 0.02 days, respectively (Fig.5A). It has been well known that developmental delays occur under nutrient-poor conditions. When wild-type flies were raised on a poor medium, diluted to 20% of normal food, the mean egg-to-eclosion period was 14.83 ± 0.12 days, which is a 60% increase from that on normal food. The delay was more remarkable for ScsαKO whose mean egg-to-eclosion period was 20.09 ± 0.24 days, a 116% increase from that of control flies reared on normal media. These results indicate that deficiency of Scsα leads to a developmental delay, as it is observed under
nutrient-poor conditions.
55
Next I examined whether ScsαKO affects the climbing activity of adult flies (Fig. 5B).
Consistent with the result of Scsα1 mutants, ScsαKO showed decreased locomotor activity, approximately 50% of that of control flies for both sexes: 5.0 ± 0.3, 10.1 ± 0.3, 6.3 ± 0.4, and 14.3 ± 0.7 cm for female ScsαKO and control, male ScsαKO and control, respectively. The results indicated that the Scsα mutants have defects in locomotor activity.
As previous described, the reduction of longevity was not observed in ScsαKO mutant flies. However, I found that the survival rate under starvation was significantly reduced in ScsαKO flies (Fig. 5C). The mean survival periods were 46.4 ± 0.7 and 53.7 ± 1.4 hours for the mutant and control, respectively.
Analysis of developmental delay in ScsαKO mutants
Since the loss of SCSα extended the egg-to-eclosion period, I examined whether the developmental delay occurred in any specific stage or throughout the development. I found that the larval period, especially the late 2nd – early 3rd instar larval periods were significantly extended in ScsαKO mutants, whereas the duration of pupal period
remained unchanged from that of wild-type flies. The growth rate of imaginal discs was measured by size increase during third instar, which corresponds to the period from 3 to 5 days after egg laying (AEL), and that from 4 to 7 days AEL, for control and ScsαKO mutants, respectively. The wing discs grow dramatically during the third instar period:
the average area of wing disc of control in 3 days AEL was approximately same as that of ScsαKO in 4 days AEL (Fig. 6A and B). There was no significant difference in final size of wing discs and final size of wing in adult stage (Fig. 6C and D). The results indicate
56
that the disruption of Scsαcauses developmental delay, most strikingly during the 3rd larval instar from 4 to 6 day AEL, but does not affect the final size of the organ.
Nutrient storage during larval period is important for holometabolous insects.
Nutrients are stored mainly in the fat body during the 3rd larval instar(Britton et al. 2002) and will be used for survival as well as for metamorphosis during pupal period when they have no food intake. To examine the energy storage in ScsαKO, I measured the level of glycogen, which is the main energy storage form of carbohydrate and provides the energy source during starvation to overcome the survival crisis. It has been reported that limited nutrition seems to delay the attainment of critical weight by reducing growth rate in Drosophila (Shingleton et al. 2005), and the stored energy may be used for the survival of cells (Britton et al. 2002). I found that the glycogen content was increased and reached the maximum level on the 4th day in wild-type, whereas ScsαKO mutants showed a decrease in glycogen level on the 5th day (Fig. 7A). There was no significant difference in final glycogen contents between control and ScsαKO mutant flies (Fig. 7B).
These data demonstrated that the Scsα deficiency impaired the energy storage and metabolism, and suggested that there is a high level of energy demand at this stage of development.
57 Discussion
In this study, I demonstrated the physiological functions of Scsα in vivo using a null mutation ScsαKO. This is the first model of SCSα deficiency in Drosophila. The mutant exhibited drastic changes in the levels of intermediate metabolites of TCA cycle and glycolysis, which include accumulation of succinyl-CoA, one of the substrates for SCS.
The physiological phenotype includes developmental delays, glycogen breakdown, impaired locomotor activity, and reduced survival under starvation conditions. I also characterized a hypomorphic mutation Scsα1in chapter 2 and found that their
phenotypes were essentially the same as ScsαKO, supporting the role of SCS in energy metabolism. Since the larval period of Scsα1 was one day shorter than that of ScsαKO, the hypomorophic allele might have functioned to some extent.
The delayed development is observed in several mutants related to mitochondrial function (Fergestad et al. 2006; Rikhy et al. 2003; Toivonen et al. 2001). For example, the gene knockdown (kdn) encodes citrate synthase catalyzing the first reaction of the citric acid cycle to produce citrate from oxaloacetate and acetyl-CoA. The kdn mutants are bang-sensitive with a decreased ATP level, and a delayed development: 4-7 days longer than that of wild-type (Fergestad et al. 2006). The gene stress-sensitive protein B (sesB) encodes ADP/ATP translocase, which catalyzes the exchange of ADP and ATP across the mitochondrial inner membrane, is also bang-sensitive with a hypoactivity and a delayed development (Rikhy et al. 2003). The mutation in technical knockout (tko) encoding a mitochondrial ribosomal protein S12 is involved in the translational capacity of mitochondria and shows deafness, bang-sensitivity, delayed development, 2-3 days longer than that of wild-type, and decreased ATP level (Toivonen et al. 2001). These
58
suggest that the genes involved in the mitochondrial functions including Scsα, are crucial during larval development.
The Drosophila larvae have to manage both maintaining cell growth and storage of energy, such as glycogen and triacylglyceride in the fat body to prepare for future metamorphosis. I have shown that glycogen content was decreased in ScsαKO at the beginning of the third instar. Developmental delays and glycogen breakdown also occur in nutrient-restricted larvae (Ballard, Jarolimova, and Wharton 2010; Mirth and Riddiford 2007). It is likely that the SCS deficiency suffers from a shortage of energy supply, and that glycogen was utilized for survival because of insufficient energy supply.
Interestingly, even the developmental period is extended in ScsαKO mutants, there is no difference in the final size of adult flies. It has been thought that the insulin-signaling pathway controls developmental time, body size and organ size through its effects on the rate of cell growth, and proliferation in different organs (Shingleton et al. 2005).
Furthermore, the growth rate and body size are tightly coupled with the sensitivity to environmental signals including nutrition (Koyama et al. 2013; Nijhout 2003). My results suggested that the nutritional sensitivity is specifically high during this period.
In humans, a mutation in SCSα (SUCLG1) is responsible for the fatal infantile lactic acidosis (Ostergaard et al. 2007). However, the Drosophila SCSα deficient was viable, and lived as long as wild-type flies. Since Scsα was not essential for survival of flies, the role of SCSα could be different between humans and Drosophila. I found that the amount of succinate in ScsαKO mutants was higher than that of control flies, and suggested that succinate was provided by the butanoate metabolic pathway through GABA and succinic semialdehyde. To confirm this hypothesis, I generated a null mutant of Ssadh by CRISPR/Cas9 system (Fig. 8), and was designated it as SsadhKO. Genomic
59
PCR and RT-PCR were conducted to confirm the deletion of CDS in Ssadh locus (Figs.
8B and C). Surprised me again, the double knockout mutant flies (ScsαKO/SsadhKO ) were viable with no morphological defects. There was an additional copy of an
SCSα-like gene in the fly genome, named CG6255 (Fig. 9A), I thought this gene might have complemented the defects of ScsαKO. However, the gene was expressed in the males only (Fig. 9B), and therefore it is unlikely that CG6255 compensated the defects caused by ScsαKO. The reason why flies are viable in the absence of SCSα is still unclear. There might be an unknown mechanism that maintains the level of succinate independent of SCS or SSADH. It is possible that the flies are relatively resistant against nutrient stress: they can adapt to nutrient-poor conditions by slowing down the growth rate while utilizing the stored energy. Therefore, the SCS deficiency is less critical in Drosophila compared to humans in terms of viability. On the other hand, other phenotypes of ScsαKO such as metabolome, locomotor activity, survival rate under starvation can be used to further analyze the enzyme function in vivo. The Scsα mutants should be useful to understand the function of SCSα and the pathophysiology of SCSα deficiency in vivo. In addition, they can serve as a model to investigate the mechanisms of metabolic adaptation to the deficiency and perhaps to nutrient-poor conditions, which also cause developmental delays.
60 References
Ballard, S.L., Jarolimova, J., Wharton, K.A. (2010) Gbb/BMP signaling is required to maintain energy homeostasis in Drosophila. Dev Biol 337, 375-385.
Britton, J. S., Lockwood, W. K., Li, L., Cohen, S. M., and Edgar, B. A. (2002) Drosophila ’ s insulin / PI3-kinase pathway coordinates cellular metabolism with nutritional conditions. Dev Cell 2, 239–249.
Carrozzo, R., Dionisi-vici, C., et al. (2007) SUCLA2 mutations are associated with mild methylmalonic aciduria , Leigh-like encephalomyopathy , dystonia and deafness. Brain 130, 862–874.
Elpeleg, O., Miller, C., Hershkovitz, E., et al. (2005) Deficiency of the ADP-forming succinyl-CoA synthase activity is associated with encephalomyopathy and mitochondrial DNA depletion. Am J Hum Genet 76, 1081–1086.
Fergestad, T., Bostwick, B. and Ganetzky, B. (2006) Metabolic disruption in Drosophila bang-sensitive seizure mutants. Genetics 173(3), 1357–1364.
Johnson, J.D., Mehus, J.G., Tews, K., Milavetz, B.I., and Lambeth, D.O. (1998) Genetic evidence for the expression of ATP- and GTP-specific succinyl-CoA synthetase in multicellular eucaryotes. J Biol Chem 273(42), 27580-27586.
Kondo, S., and Ueda, R. (2013) Highly improved gene targeting by germline-specific Cas9 expression in Drosophila . Genetics 195, 715–721.
Koyama, T., Mendes, C. C., and Mirth, C. K. (2013) Mechanisms regulating
nutrition-dependent Developmental plasticity through organ-specific effects in insects.
Front Physiol 4:263, 1–12.
61
Lamperti, C. et al. (2012) A novel homozygous mutation in SUCLA2 gene identified by exome sequencing. Mol Genet Metab 107, 403-408.
Mirth, C.K., and Riddiford, L.M. (2007) Size assessment and growth control: how adult size is determined in insects. BioEssays 29, 344-355.
Morava, E., Steuerwald, U., Carrozzo, R., Kluijtmans, L.A., Joensen, F., Santer, R., Dionisi-Vici, C. and Wevers, R.A. (2009) Mitochondrion dystonia and deafness due to SUCLA2 defect; clinical course and biochemical markers in 16 Children. Mitochondrion 9, 438-442.
Nijhout, H. F. (2003) The control of body size in insects. Dev Biol 261(1), 1–9.
Ostergaard, E. et al. (2007) Mitochondrial encephalomyopathy with elevated methylmalonic acid is caused by SUCLA2 mutations. Brain 130, 853-861.
Ostergaard, E., Christensen, E., Kristensen, E., Mogensen, B., Duno, M., Shoubridge, E.A. and Wibrand, F. (2007) Deficiency of the alpha subunit of succinate-coenzyme A ligase causes fatal infantile lactic acidosis with mitochondrial DNA depletion. Am J Hum Genet 81, 383-387.
Rikhy, R., Ramaswami, M., and Krishnan, K. S. (2003) A temperature-sensitive allele of Drosophila sesB reveals acute functions for the mitochondrial adenine nucleotide translocase in synaptic transmission and dynamin regulation. Genetics 165(3), 1243-1253.
Shingleton, A. W., Das, J., Vinicius, L., and Stern, D. L. (2005) The temporal requirements for insulin signaling during development in Drosophila. Plos Biol 3(9), e289.
62
Toba, G., Ohsako, T., Miyata, N., Ohtsuka, T., Seong, K.H. and Aigaki, T. (1999) The gene search system. A method for efficient detection and rapid molecular identification of genes in Drosophila melanogaster. Genetics 151, 725-737.
Toivonen, J. M., Dell, K. M. C. O., Petit, N., Irvine, S. C., Knight, G. K., Lehtonen, M., Longmuir, M., et al. (2001) technical knockout , a Drosophila model of mitochondrial deafness, Genetics 159(1), 241-254.
63 Figures
Figure 1. Alignment of amino acids between SUCLG1 and SCSα
Black box indicated the conserved region of amino acids sequence between SUCLG1 and SCSα. The Genetyx software was used for the arrangement.
64
Figure 2. Lifespan of control (black) and ScsαKO (grey) adult flies.
The difference between the genotypes was significant for females (A) (log rank test,
**p<0.01), but not significant for males (B) (log rank test, p>0.05).
65
Figure 3. Quantification of intermediate metabolites in glycolysis and TCA cycle.
Dark (control) and grey (ScsαKO) bars indicate the mean values (±SEM) of metabolites in glycolysis (A, C) and TCA cycle (B, D). Statistical significance levels: *; p<0.05, **;
p<0.01, ***; p<0.001.
66
Figure 4. Quantification of GABA in 3 day-old adult flies.
Bars represent relative amounts of GABA (mean±SEM) in control (dark grey) and ScsαKO (light grey). Statistical significance (t-test): *; p<0.05.
67
Figure 5. Phenotypic characterization of Scsαmutants.
(A) Egg-to-eclosion periods of control (closed circle), Scsα1 (grey circle) and ScsαKO (open circle). Number of eclosed flies on each day against total number of eclosed flies were indicated for each genotype. (B) Climbing activity of adult flies. Bars indicate mean (±SEM) climbing activities of control (dark grey) and ScsαKO (light grey).
Statistical significance (t-test): ***; p<0.001 (C) Survival curves of control (black) and ScsαKO (grey) flies under starvation. Statistical significance (log rank test): ***; p<0.001.
68
Figure 6. Delayed developmental larval stage in ScsαKO mutant flies.
(A) Images of wing discs dissected from third instar larvae at 3-5 day AEL and 4-7 day AEL for control and ScsαKO, respectively. Bar: 200μm. (B) Growth of wing disc. Areas of wing discs at each day are indicated. Control: closed circle, ScsαKO: grey circle. (C) Comparison of final size of wing discs between control and ScsαKO. (B) Comparison of wing size between control (dark grey) and ScsαKO (light grey).
69
Figure 7. Altered glycogen accumulation pattern in ScsαKO larvae.
(A) Glycogen contents in control (closed circle) and ScsαKO (grey circle) larvae.
Glycogen content showed sharp decrease on the 5th day(AEL). (B) Comparison of final glycogen content between control and ScsαKO. No significant difference was observed between two groups.
70
Figure 8. Genomic organization of Ssadh locus and the structure of a null mutation, SsadhKO .
(A) Schematic representation of the Ssadh locus. Black and grey boxes are the protein coding and untranslated regions, respectively. The CRISPR/Cas9 system was used to generate SsadhKO. Two scissors indicate the targeted sites of guide RNA. (B) Genomic DNA PCR using primers GF and GR. A 5.0kb and a 0.8kb fragments
represent wild-type (control) and ScsαKO allele, respectively. (C) RT-PCR using primers RF and RR. Amplified fragments derived from Ssadh mRNA were detected in
wild-type (control), but not in SsadhKO. rp49 was used as an internal control.
71
Figure 9. Paralog CG6255 is only expressed in male flies.
(A) Alignment of amino acids sequence between SCSαand CG6255. Black box indicated the conserved region of amino acids. The Genetyx software was used for arrangement. (B) Quantitative RT-PCR result revealed that CG6255 was only expressed in male flies.
72
Acknowledgements
I am most grateful to Prof. Toshiro Aigaki for his constant help, guidance and support throughout my Ph.D. course. He provided me the valuable suggestion and comments on my research, and always encouraged me to persist through challenges to build a strong mindset.
I owe my special thanks to Dr. Yukiko Sato-Miyata, Dr. Satomi Takeo, Dr. Manabu Tsuda, Dr. Tsunaki Asano and Dr. Takaomi Sakai, for their friendly concerns on my study and the foreign life. A lot of advices and technique supports allow me to complete the speculation of research. All members of the Cellular Genetic Lab should be
appreciated for their encouragement and kind help. I feel so lucky to be one of the members in this lab.
Fruit flies, my best partner, I should thank you for your contributions to my research and experiments.
Also, I would like to show my gratitude to my friends, C.Z., M.K., J.D., for sharing my joy and sorrow these years.
Finally, my deep appreciation goes to my family, my strong backing, Hedong Quan, Jishan Quan and Chunming Quan. They worked as hard as they can to support my study, I cannot finish my Ph.D. course without their continuous encouragement and unparalleled love. The last special words of thanks are for my grandfather, Yuanchun Quan, for naming me with the most beautiful name.
73
Appendix
GABA
Supplemental Fig.1
Supplemental Fig.2
Female
Male
B
Supplemental Fig.3
A
Days Days