Novel Cytochrome P450, cyp6a17, Is Required for Temperature Preference Behavior in Drosophila

Perception of temperature is an important brain function for organisms to survive. Evidence suggests that temperature preference behavior (TPB) in Drosophila melanogaster, one of poikilothermal animals, is regulated by cAMP-dependent protein kinase (PKA) signaling in mushroom bodies of the brain. However, downstream targets for the PKA signaling in this behavior have not been identified. From a genome-wide search for the genes regulated by PKA activity in the mushroom bodies, we identified the cyp6a17 Cytochrome P450 gene as a new target for PKA. Our detailed analysis of mutants by genetic, molecular and behavioral assays shows that cyp6a17 is essential for temperature preference behavior. cyp6a17 expression is enriched in the mushroom bodies of the adult brain. Tissue-specific knockdown and rescue experiments demonstrate that cyp6a17 is required in the mushroom bodies for normal temperature preference behavior. This is the first study, to our knowledge, to show PKA-dependent expression of a cytochrome P450 gene in the mushroom bodies and its role as a key factor for temperature preference behavior. Taken together, this study reveals a new PKA-Cytochrome P450 pathway that regulates the temperature preference behavior.


Introduction
Behavioral responses to environmental stimuli, including light, humidity, and temperature, are important for survival of all living organisms. Especially for poikilothermal animals, extreme changes in ambient temperature or persistence for a long time at high or low temperature lead to death. These animals adapt to their body temperature changes by using molecular mechanisms to alter metabolism or by behavioral strategies to choose proper temperature conditions. Drosophila has been widely used as a genetic model for studying a variety of behaviors including learning. Recently, it has also been utilized to study the genetic basis of temperature sensation and temperature preference behavior.
In Drosophila, larvae and adult flies show strong temperature preference behavior [1][2][3]. A family of transient receptor potential (TRP) ion channels plays major roles in the sensation of temperature. For example, Painless, one of the TRP channel (TRPA1) superfamily, is required for sensing nociceptive stimuli over 38uC [4]. Another TRP channel, Pyrexia, is involved in protecting flies against noxious temperature over 40uC [5]. In contrast, Drosophila ANKTM1 TRP family channel participates in temperature selection by opening at warm temperature (24-29uC) [6,7].
Despite extensive studies on the role of Drosophila TRP family channels in temperature sensing, it is not well understood how flies perform specific behavior to choose optimal temperature conditions. Interestingly, recent studies have shown that the mushroom bodies in the brain, which plays a critical role in learning and memory, is important for TPB [8]. Furthermore, cAMP-dependent PKA signaling in the mushroom bodies is not only essential for learning and memory but also for TPB. These studies have provided important clues to the mechanism underlying TPB, but the target genes for PKA signaling have been elusive. Hence, we carried out a genome-wide screen for the genes regulated by PKA to obtain insights into the molecular events underlying TPB. From this screen, we found cyp6a17, a cytochrome P450 superfamily gene, as a PKA downstream factor for TPB.
Here, we show that cyp6a17 is regulated by PKA and is required for temperature preference behavior. We demonstrate that cyp6a17 expression in the mushroom bodies is necessary and sufficient for TPB. This study identifies cyp6a17 as an important target of PKA signaling for mediating TPB in the mushroom bodies.

Identification of new genes regulated by PKA in the mushroom bodies
Temperature preference behavior in Drosophila depends on the level of PKA signaling in the mushroom bodies. To identify new components downstream to PKA, we carried out a genome-wide screen for genes regulated by PKA signaling in the mushroom bodies. Using the Gal4-UAS system, we increased or decreased PKA activity in the mushroom bodies by expressing dominantnegative (UAS-PKA DN ) or constitutively active PKA (UAS-PKA CA ), respectively. Expression of PKA transgenes was targeted to the mushroom bodies using the mushroom body-specific MB247-Gal4 driver [8]. PKA expression was induced for 12-16 hours in threeday-old adults by inactivating the temperature-sensitive Gal80 [24] at the restrictive temperature. We then analyzed geneexpression profiles to identify the genes showing altered expression levels in response to the high or low PKA activity. The Drosophila GeneChip (DrosGenome 2.0) was used to obtain gene expression profiles from fly heads of three different groups: (i) the control group with no PKA transgene expression, (ii) the low PKA activity group and (iii) the high PKA activity group. Transcripts that showed more than 2-fold changes from the control expression level were considered for further analysis ( Figure 1A & Table S1).
A total of 103 transcripts showing more than 2-fold changes were classified into three categories based on their responses to PKA activity: (i) 44 transcripts responded to PKA DN but not by PKA CA . Among these, 14 showed downregulation while 30 resulted in upregulation. (ii) 39 transcripts showed altered expression by PKA CA (23 increases and 16 decreases). (iii) 20 transcripts showed changes by both PKA DN and PKA CA conditions. Among them, 8 showed decreased and 11 showed increased transcript levels by both PKA DN and PKA CA conditions ( Figure 1B).
Only one gene, cyp6a17, showed upregulation by PKA CA and downregulation by PKA DN . We were most interested in this gene since its expression was correlated with both higher and lower PKA activities ( Figure 1C). To confirm these microarray data, we examined the PKA-dependent expression of cyp6a17 by RT-PCR and real-time-PCR. The expression level of cyp6a17 mRNA in the head was decreased to 12% of the wild-type level by inhibiting PKA activity with mushroom body-specific expression of PKA DN . On the contrary, the transcript level was increased 5.6361.32 times by PKA CA overexpression ( Figure 1D). Hence, cyp6a17 expression in the head is regulated by PKA in an activitydependent manner.

cyp6a17 mutations affect temperature preference behavior
To test whether cyp6a17 is required for normal TPB, we analyzed the effects of cyp6a17 mutations on TPB. We generated a deletion allele (cyp6a17 D173 ) by inducing imprecise excision of the P-element P{SUPor-P}KG04448 inserted at 133 bp downstream from the translation start site (Figure 2A). The cyp6a17 D173 mutation is a deletion of 1.1 kb sequence from the P-insertion site and affects only the cyp6a17 gene ( Figure 2A). Consistent with the deletion mapping, the transcript for cyp6a17 was not detected in the cyp6a17 D173 mutant by RT-PCR analysis, suggesting that cyp6a17 D173 is a null mutation ( Figure 2B). Homozygous cyp6a17 D173 mutant flies are viable and fertile with no visible morphological defects, suggesting that cyp6a17 is not essential for normal development and fertility.
Next, we characterized the temperature preference in this deletion mutant. For TPB assays, all mutant and transgenic lines were outcrossed to the w 1118 line at least four times. cyp6a17 D173 mutant flies showed reduced AI (avoidance index) for low temperature (AI Low = 0.5660.06, p,0.001) ( Figure 2C and Table  S2). Immunocytochemcial analysis with anti-FasII antibody marker for the mushroom bodies did not reveal significant morphological defects in the brain of this mutant ( Figure S1). Thus, the abnormal TPB shown in cyp6a17 loss-of-function mutant is probably not caused by morphological defects in the mushroom bodies.
Specificity of the cyp6a17 gene function in TPB was also confirmed by rescue of the cyp16a17 mutant phenotypes. Abnormal TPB phenotypes of cyp6a17 null mutant flies were rescued to the wild-type level by a 4.2-kb genomic DNA fragment (cyp6a17 GR in Figure 2A) spanning the cyp6a17 locus (AI-High = 0.8460.04; AI Low = 0.8860.01) (Figure 2A, B, & D and Table S2). These data indicate that cyp6a17 is required for normal TPB.

Expression of cyp6a17 is enriched in the mushroom bodies
To visualize the expression pattern of cyp6a17, we generated a cyp6a17 MB -Gal4 transgenic reporter strain in which Gal4 is induced by a cyp6a17 genomic region containing the promoter. The translation start site for cyp6a17 is only 612 bp apart from the adjacent upstream gene cyp6a22. cyp6a17 MB -Gal4 was created by cloning an 1,201 bp fragment upstream of the cyp6a17 translation start site into pCaSpeR vector. This fragment contains the entire 59 region of cyp6a17 including a 39 part of the adjacent cyp6a22 gene. Therefore, the 59 region of cyp6a17 includes regulatory sequences necessary for the cyp6a17 function in TPB (Figure 2A).
To detect cyp6a17 expression, we crossed cyp6a17 MB -Gal4 transgenic flies to a reporter line UAS-mCD8.GFP. Confocal analysis of GFP expression in the adult brain revealed a complex pattern of expression in the mushroom bodies ( Figure 3A). GAL4 is expressed in the mushroom body neurons that project to a-, band c-lobes ( Figure 3D & E), although it is not clear whether the cyp6a17 MB -Gal4 is expressed in all or a subset of mushroom body neurons. In addition to the mushroom bodies, GFP was also expressed in the ellipsoid bodies (EB), pars intercerebralis (PI) and several unknown lateral neurons ( Figure 3B). Abnormal TPB phenotypes in the cyp6a17 D173 mutant was rescued by expressing wild-type cyp6a17 using the cyp6a17 MB -Gal4 driver (AI High = 0.7460.03, p,0.001; AI Low = 0.9060.03) ( Figure 3F and Table  S2). These data suggest that the cyp6a17 MB -Gal4 reporter represents the pattern of endogenous cyp6a17 expression.
cyp6a17 is required in the mushroom bodies for temperature preference behavior To test whether cyp6a17 is required for TPB in specific tissues, we examined the effects of reduced cyp6a17 expression by targeted knockdown using cyp6a17-specific UAS-RNAi transgenic flies. To minimize potential off-target effects on other Drosophila CYP genes, we selected a relatively small unconserved region (222 base pairs) of cyp6a17 as an RNAi target. Expression of cyp6a17-RNAi in the neuronal cells using a pan-neural Elav-Gal4 driver resulted in a significant reduction of TPB (AI High = 0.9060.02; AI Low = 0.2260.14, p,0.001) ( Figure 4A and Table S2). In contrast, cyp6a17 RNAi expression in the whole body except the head by using c290-Gal4 showed no effect on TPB ( Figure 4B). These data suggest that cyp6a17 function for normal TPB is required in neurons of the brain.
In the adult brain, cyp6a17 MB -Gal4 expression is enriched in the mushroom bodies as well as EB and PI ( Figure 3A). cyp6a17-RNAi  Table S2). This result is consistent with the previous report that EB and PI are not essential for TPB [8]. To exclude the possibility that cyp6a17 RNAi affects locomotor activity, we examined the motor activity of the flies by climbing behavior test. In these control assays, locomotor activity was normal at all temperature conditions tested.
Next, we induced cyp6a17-RNAi in the cyp6a17-expressing cells by using cyp6a17 MB -Gal4. These flies showed a reduced AI index, as seen in cyp6a17 mutants (AI High = 0.5760.07, p,0.001; AI Low = 0.1160.07, p,0.001) ( Figure 5A and Table S2). Thus, cyp6a17 function is indeed required for TPB in the cells identified by the cyp6a17 reporter. To determine whether cyp6a17 is required in the mushroom bodies, we induced cyp6a17-RNAi preferentially in the mushroom bodies using MB247-Gal4. These flies with knockdown of cyp6a17 in the mushroom bodies showed a reduced avoidance index in the TPB assay (AI High = 0.8960.03; AI Low = 0.1460.09, p,0.001) ( Figure 5B and Table S2). In the heads of these flies, cyp6a17 mRNA expression level was reduced compared with wild-type (0.5660.06, p,0.001) ( Figure S2A). cyp6a17 MB -Gal4 and MB247 show additional expression in surface glia [25]. To check whether cyp6a17 in surface glia is required for TPB, we induced cyp6a17-RNAi in surface glia by cyp6a17 SG -Gal4 which drives strong Gal4 expression in surface glia but very weakly in the mushroom bodies ( Figure S3). These flies showed normal The abnormal TPB of cyp6a17 D173 was restored by expression of the cyp6a17 genomic transgene (cyp6a17 GR ). Black bars, AI High ; gray bars, AI Low . One asterisk, P,0.005 and two asterisks, P,0.001. doi:10.1371/journal.pone.0029800.g002 TPB ( Figure 6A and Table S2). In addition, cyp6a17 RNAi knockdown by other mushroom body Gal4 lines like OK107and c772-Gal4 also resulted in slightly weaker but significant TPB defects: OK107-Gal4 (AI High = 0.8060.02; AI Low = 0.4060.20, p,0.001) ( Figure 6B and Table S2) and c772-Gal4 (AI High = 0.8160.05; AI Low = 0.4360.13, p,0.001) ( Figure 6C & Table S2). Furthermore, cyp6a17 mutant phenotype was rescued by expressing the wild-type cyp6a17 cDNA using the MB247-Gal4 mushroom body driver ( Figure 5C and Table S2). Taken together, these data suggest that cyp6a17 expression in the mushroom bodies is necessary for normal low temperature TPB.
Previous studies proposed that the PKA signaling cascade is important for TPB [8]. According to our microarray data, cyp6a17 expression is regulated by PKA in an activity-dependent manner. Thus we tested epistatic relationship between PKA and cyp6a17. It is noteworthy that TPB phenotypes induced by PKA CA overexpression were strongly suppressed by cyp6a17-RNAi (AI High = 0.7260.06, p = 0.08; AI Low = 20.0560.12, p,0.001) ( Figure 7A & Table S2). Furthermore, the transcript of cyp6a17 was significantly decreased to 0.04460.026 (p,0.001, t test) in dCrebB-17A s162 mutant ( Figure 7B). In rut 1 , which is a loss-of-function mutant for rutabaga, encoding calcium-calmodulin responsive adenylate cyclase, cyp6a17 expression level was reduced compared with the wild-type level (0.05460.008, p,0.001, t test) ( Figure 7C). On the contrary, in the dnc 1 /dnc M14 transheterozygote, which can increase cAMP level due to a defect in Dunce phosphodiesterase, cyp6a17 expression level was slightly increased (1.4860.05, p,0.001, t test) ( Figure 7D). These results suggest that cyp6a17 transcription is regulated by the cAMP-PKA signaling pathway and cyp6a17 functions downstream to PKA for TPB.

Discussion
Temperature preference is an important behavioral strategy for optimizing the body temperature in poikilothermal animals like Drosophila. TPB involves temperature sensing and processing of the sensory input at higher levels in the brain. Recent studies have implicated that PKA signaling in the mushroom bodies is required for TPB. However, it is largely unknown how PKA signaling pathway contributes to the behavioral response. In this study, we identified cyp6a17 as a new gene regulated by PKA signaling. Our analysis provides several pieces of evidence that cyp6a17 is required in the mushroom bodies for TPB. First, mutations in the cyp6a17 gene showed reduced TPB. Second, abnormal TPB phenotypes in cyp6a17 mutants were rescued with a wild-type cyp6a17 transgene.  Fig. 3E). It also shows weak circular staining in three ring structures that appear to be ab lobes. Note that there is no staining inside the rings that may correspond to the ab core.  Third, tissue-specific RNAi knockdown of cyp6a17 expression in the mushroom bodies resulted in TPB defects. TPB assays are measured by two indices, AI Low and AI High , for avoiding low and high temperatures, respectively. It is interesting to note that loss of cyp6a17 results in a preferential reduction of the AI Low value compared with the AI High score. This asymmetric effect of cyp6a17 mutations suggests that cyp6a17 might be selectively required for avoidance of low temperatures. We noted that the transgenic expression by MB247.cyp6a17 from the cyp6a17 mutant significantly lowers the AI High index. However, one of the control strains for this test, cyp6a17 D173 ; UAS-cyp6a17/+, shows a very low AI-High value in the absence of any Gal4 driver. Thus, the reduction of AI High in this strain is unlikely to be due to cyp6a17 overexpression. cyp6a17 mutant and this transgene were cyp6a17 Is Required for TPB in Drosophila outcrossed at least four times, but the effect of UAS-cyp6a17/+ in reducing the AI High still remained. It is unknown why this strain shows non-specific low AI High . Importantly, however, transgenic overexpression of Cyp6a17 in the presence of Gal4 restores the low AI Low defect of the cyp6a17 D173 mutant (Fig. 5C). Previous studies have shown that increased levels of cAMP or PKA signaling cause stronger avoidance from low temperature whereas lower cAMP-PKA signaling levels result in the opposite trend with reduced avoidance. This is consistent with our data that cyp6a17 mutants show lower AI Low . cyp6a17 is necessary for TPB but may not be sufficient to induce AI High . Since cyp6a17 is the only one of multiple downstream factors for PKA CA , it may need additional PKA downstream factors to induce AI High .
It has been reported that many cytochrome P450 proteins are expressed in the midgut, malpighian tubules and fat body whereas only one gene cyp4g15 is detected in the brain of developing larvae   [26]. We found that cyp6a17 mRNA is expressed in the adult body including hindgut, but about 30% more expression is detected in the head ( Figure S2B). Using a reporter construct regulated by a 59 region of the cyp6a17 gene, we showed that cyp6a17 expression is highly enriched in the mushroom bodies, especially ab and c lobes. Three observations strongly support that this reporter reflects the pattern of cyp6a17 gene expression. Firstly, cyp6a17 RNAi driven by mushroom body-specific Gal4 lines causes TPB defects. Secondly, cyp6a17 RNAi using cyp6a17 MB -Gal4 also results in similar TPB phenotypes. Thirdly, cyp6a17 expression induced by cyp6a17 MB -Gal4 rescues cyp6a17 null mutant phenotypes. These data indicate that the mushroom body regions labeled by cyp6a17 reporter are the sites where cyp6a17 function is required for TPB. To our knowledge, cyp6a17 is the first gene of this family shown to be expressed in the mushroom bodies with specific function in TPB. However, it is possible that other cell types in addition to the mushroom bodies might also be involved in the TPB phenotype.
Cytochrome P450 proteins are known to participate in the synthesis of steroid hormone, thereby regulating homeostasis and thermal responses [27,28]. Because cyp6a17 is also expressed in PI and EB regions that have been implicated in neuroendocrine regulation [29], cyp6a17 may play a role in the neuroendocrine system. However, it has been shown that the PI and EB are not required for TPB [8]. Furthermore, our data show that the cyp6a17 mutant deficits in TPB are rescued by cyp6a17 expression in the mushroom bodies but not in subsets of PI and EB neurons. Thus, it appears that the cyp6a17 function for TPB is required mainly in the mushroom bodies rather than the PI and EB cells.
There are 83 CYP family genes in Drosophila, excluding 7 pseudogenes. Of these, only several CYP genes have been characterized for their expression patterns and functions. Some CYP proteins function together in a protein complex system for catalyzing oxidative reactions [30]. However, many CYP genes show tissue-specific expression profiles, suggesting that CYP genes may play distinct functions in diverse tissues. This possibility is supported by our finding that cyp6a17 has a specific function in the mushroom bodies as a necessary component for controlling the temperature preference behavior. In addition to cyp6a17, other CYP genes might also play important roles in TPB.
cAMP-PKA signaling in the mushroom bodies is important for learning and memory in Drosophila [31], but downstream events are not well understood. Our study identified a cytochrome P450 gene as a downstream factor for PKA signaling in TPB. Given the potential relationship between temperature and stress related genes [32][33][34], this work could provide clues to understanding the relationship between stress response and temperature preference behavior. Although we focused on characterizing the function of cyp6a17 in this study, our genome-wide screen identified many other genes that showed strong response to either PKA CA or PKA DN (Table S1). Further analysis of these genes might identify additional genetic factors involved in PKA-dependent TPB and their relationship with cyp6a17.
Until now, there is no report to our knowledge that Drosophila CYP is involved in temperature preference behavior. It is an intriguing question how PKA-dependent regulation of cyp6a17 is functionally related to TPB. Since steroid hormone can modulate several kinds of behavior, neuronal homeostasis could be an important factor for TPB. Consistent with this idea, it has been shown that Drosophila dystroglycan plays a role in modulating energy homeostasis and the thermal responses [35]. Cyp6a17 is known to have monooxygenase activity [13]. Hence, it is plausible that cyp6a17 might be involved in the synthesis of steroid hormones, thereby modulating temperature preference behavior. cyp6a17 may also participate in the regulation of ATP level to modulate energy homeostasis in neurons. Detailed analysis of these possibilities in the future will help provide mechanistic insights into the function of cyp6a17 for TPB.

Drosophila strains and transgenic lines
Fly stocks were raised on standard cornmeal food at 25uC and 40-50% relative humidity. w 1118 and Canton-S were used as wildtype strains. UAS-PKA DN and PKA CA were provided by D. Kalderon; other lines were provided as follows: MB247, c772 (T. Zars), c309 (L. C. Griffith), MB247; tub-GAL80 ts (G. Roman), P{SUpor-P}KG04448, c232, c290 and dilp2-Gal4 (Bloomington stock center). All mutant and transgenic alleles used in this study were outcrossed to the w 1118 line at least four times.

Microarray analysis
Total RNA was extracted from adult heads. After extracting total RNA, microarrays were generated by Digital Genomics Inc. (Seoul, Korea). Modified cDNA from total RNA was hybridized to GeneChipH Drosophila Genome 2.0 Array. Image acquisition and data extraction from microarray experiments were analyzed by Affymetrix GeneChipH Scanner 3000 7G and Affymetrix GCOS1.4 software, respectively. Data were normalized by MAS5 and RMA (Robust Multi-array Average), and differentially expressed genes were selected by comparing control and transgenic lines. The functional category of selected genes was classified using the DAVID2007 (http://david.abcc.ncifcrf.gov/ home.jsp) program.
Temperature preference behavior assay and climbing assay TPB assays were performed as described previously [8,36]. 1-3 day old adult files of mixed gender were collected without anesthetization in vials containing fresh food during morning hours. After 3 days under a 12 hr light/dark incubation cycle, the collected flies were placed in the test device. A temperature gradient from 15 to 45uC with a slope of 0.75uC/cm was produced in an aluminum block (42 length624 width67 cm heights). Electronic thermal sensors were embedded in the block every 3 cm, and the gradient was established using cold and hot circulating water chambers at each end. A glass plate with five separate lanes was placed 2.5 mm above the block, creating suitable space for flies to walk. The glass plate was coated with quinine sulfate powder, an aversive stimulus for Drosophila, to prevent flies from escaping the temperature gradient by resting on the walls or roof of the lane. 1% yeast paste was applied on the block for flies to stay on the block. Adult flies were placed between the aluminum block and the glass plate, allowed to move for 25 min in the dark, and photographed with a digital camera with 2-5 msec exposure at the end of each assay. Avoidance indices against low temperature (AI Low ) and high temperature (AI High ) were calculated from the following formulae: AI Low = N Int 2N Low / N Int +N Low and AI High = N Int 2N High /N Int +N High . N Low , N High and N Int are the numbers of flies in the region below 23.4uC, above 26.7uC, and between 23.4 and 26.7uC, respectively. A population of mixed sexes was tested for TPB. The number of flies was recorded by using digital photographs and processed with Microsoft Excel, and one-way ANOVA followed by Dunnett's post test was performed. All data points in the figures are represented as the mean and the standard error of the mean (s.e.m.), unless mentioned otherwise.
For climbing test to check the locomotor activity, groups of ten flies were transferred into vials and incubated for 1 hr at 25uC for environmental acclimatization. Then flies were transferred to 18uC and 32uC incubator for 15 min. After tapping the flies down to the bottom, we counted the number of flies that climb more than 10 cm from the bottom in 1 min. Three trials were performed for each genotype and repeated with 3 different batches. All mutant strains used in our TPB assays showed similar climbing rates as the wild-type. A horizontal locomotor activity test was done as described previously [37]. It showed that cyp6a17 D173 mutant flies are normal in the horizontal locomotor activity.
It has been reported that the mushroom bodies are involved in suppressing hyperactivity of flies [38]. We found that test flies of the same genotype placed at high and low temperatures show different locomotor activity at the beginning of the trial but results in the same distribution profiles at the end of the assay. This indicates that the difference in the initial locomotor activity does not affect the temperature preference behavior in the standard assay conditions that allow time for test flies to choose and settle at their preferred temperature.

Construction of transgenic flies
cyp6a17-RNAi construct was generated as described [39]. cyp6a17-RNAi flies contain two copies of a 222 bp fragment of cyp6a17 cDNA. PCR primers used to create the target region for cyp6a17 knockdown were 59-TGCTCTAGACAGCTTGTAC-GATCCAAA-39 and 59-TGCTCTAGAAATCATTTCGCTT-TTCCT-39. A PCR fragment was cloned into pWIZ vector. To construct UAS-cyp6a17, 1.5 kb fragment of a cyp6a17 cDNA obtained from the Drosophila Genome Research Center was cloned into pUAST. A cyp6a17-Gal4 construct was generated using pCaSper vector and 1,201 bp sequence upstream from the transcription start site of the cyp6a17 gene. cyp6a17 SG -Gal4 was inserted at 78A1. Unlike cyp6a17 MB , Gal4 expression in the mushroom bodies was strongly repressed, thus showing a preferential expression in the surface glia. To make a cyp6a17 genomic rescue construct, a 4.2 kb SacI-digested genomic fragment containing about 1.2 kb upstream from the transcription start site and about 1.5 kb downstream from the transcription stop site, was cloned into pCaSper vector. Transgenic lines were generated in the w 1118 background using standard germline transformation techniques.
Relative quantitative RT-PCR 4-6 day old adult flies were collected and frozen immediately in liquid nitrogen. Heads were removed from bodies by vortexmixing and sorting through a chilled sieve. Total RNA was extracted from heads with a Trizol (Invitrogen) and reverse transcribed using the AMV reverse transcriptase system (Promega). RT-PCR was performed with the following primers: for cyp6a17, 59-GTTGTTACTGGCGCTAATCGT-39 and 59-TT-CCATCACAACCTGCTCCA-39; for cyp6a22, 59-TTCGGG-AAACTGTGAAGCAG-39 and 59-GATAAAAACATTTGATT-GTGT-39; for cyp6a23, 59-TCGCTGTTACTAACGTTGAT-39 and 59-TTTCCATGACGACCTGCTCCA-39; for gapdh1, 59-ACCGTCGACGGTCCCTCT-39 and 59-GTGTAGCCCAGG-ATTCCCT-39. Real-time PCR was performed with iCycler iQ (Bio-Rad) using the KAPA SYBR FAST Bio-Rad iCycler qPCR kit (Kapa Biosystems). The level of gapdh1 mRNA was measured as an internal control for the RNA amount in each sample. Data processing was performed with Microsoft Excel. All quantitative analysis were done more than three times.

Immunohistochemistry
Adult brains expressing transgenic cyp6a17-Gal4.UAS-mCD8.GFP were removed from head capsules and fixed in 4% paraformaldehyde in PBS for 30 min, and rinsed in PBS containing 0.5% Triton X-100. For staining with anti-FasII (Hybridoma Bank), adult brains were fixed for 30 min in 4% paraformaldehyde in PBS containing 0.5% Triton X-100. Fixed samples were washed for 10 min three times and incubated with 5% normal goat serum. Antibody dilution was 1:50 for mouse anti-FasII and 1:250 for goat anti-mouse Alexa Fluor 568 (Molecular Probe). Confocal analysis was performed on a Zeiss LSM710 microscope. Confocal stacks were processed with ZEN 2009 Light Edition and Adobe Photoshop.