Identifying Target Genes of the Transcription Factor Hr38 in Drosophila Clock Neurons

Hillary Yoseph, CAS '25

Bachelor of Science: Global Public Health/Biology

Advised by Dr. Justin Blau

Abstract

Structural plasticity involves neurons modifying their connections in response to stimulation. Immediate early genes are rapidly expressed in response to neuronal firing and are required for structural plasticity. However, studying the functions of immediate early genes is difficult because of their rapid changes in expression. With this, Drosophila small ventral lateral neurons (s-LNvs) are used as a model because of their predictable changes in plasticity and involvement of the immediate early gene Hormone Receptor-like in 38 (Hr38). I monitored the normal expression of Hr38 and its candidate target gene still life (sif) and found that they are transcribed at similar times at dawn. I also induced Hr38 at dusk and found that this is sufficient to upregulate sif transcription. These findings support the hypothesis that sif is a target gene of Hr38. To identify additional Hr38 target genes, I sorted larval LNvs with Hr38 briefly induced and performed RNA sequencing. I analyzed the data on R and found a large pool of genes that had an increased fold change, including sif.

Introduction

Structural plasticity is the ability for neurons to physically alter their connections in response to environmental events (Lymer et al. 2024, Petsakou et al., 2015). Structural plasticity plays a key role in brain development and remains important throughout life for forming, storing, and recalling memories (Bozelos et al., 2017). The plasticity of the brain allows for the modification of neural connections by forming new synapses, remodeling dendritic spines, and removing underutilized connections (Gipson et al., 2017). Apart from its ability to aid cognitive functions, structural plasticity provides a mechanism for the brain to repair itself after minor injury (Gage et al., 2004). The ability to modify connections allows the brain to remain flexible and adaptable, which is fundamental for healthy brain function. Altered structural plasticity mechanisms are seen in conditions like schizophrenia, Autism Spectrum Disorder (ASD), Post-Traumatic Stress Disorder (PTSD), and addiction (Lymer et al., 2024). Therefore, learning more about how structural plasticity can impact neural connections at a molecular level is important.

Activating immediate early genes (IEGs), also known as activity-regulated genes, is one of the initial responses triggered by neuronal firing. IEGs are characterized for their rapid and transient changes in transcription (Sheng et al., 1990). Many IEGs, such as c-fos, encode transcription factors that then regulate downstream genes involved in cell differentiation, learning, motor control, and cognition (Lara Aparicio et al., 2022). Other IEGs, like Arc, function directly in plasticity by controlling synaptic strength via receptor trafficking (Zhang et al., 2021). Since IEGs are highly dynamic, it is difficult to study their functions without a model organism. Therefore, identifying unknown target genes of IEGs in an organism like Drosophila is important to understand how structural plasticity is regulated.

In Drosophila, the small ventral lateral neurons (s-LNvs) provide a model to study how IEGs impact structural plasticity. s-LNvs are circadian pacemaker neurons involved in regulating rhythms in activity and sleep-wake cycles (Dubowy et al., 2017). Additionally, s-LNvs display rhythmic structural plasticity with expanded projections at dawn and retracted projections at dusk (Lymer et al., 2024). This predictable change makes the s-LNvs a good model to evaluate the molecular mechanisms involved in plasticity. Additionally, there are only four s-LNvs per hemisphere of the brain, making them easy to visualize (Lymer et al., 2024).

The expansion of s-LNv projections at dawn requires the IEG Hormone receptor-like in 38 (Hr38). Hr38 is an orphan nuclear receptor in the NR4A subfamily, which means its ligand is unknown or not required for its activation (Gu et al., 2016). Hr38 gets its name from being identified as a mediator of the ecdysteroid pathway at the 38th cytological position (Brody et al., 1999). However, because X-ray crystallography revealed the absence of a classical ligand binding pocket, the term “like” was added to the end of the name. Hr38 expression is triggered by neuronal activity, a hallmark of IEGs (Brody et al., 1999). From there, Hr38 acts as a transcription factor to regulate downstream targets involved in structural remodeling and cell signaling, ultimately leading to expanded projections. Lymer et al. have shown that silencing Hr38 inhibits s-LNv projection expansion at dawn. Conversely, overexpressing Hr38 for two hours at dusk drives expansion when s-LNv projections are normally retracted (Lymer et al., 2024). This highlights Hr38 as a primary regulator of s-LNv structural plasticity; however, the specific target genes Hr38 regulates are unknown.

One potential downstream target regulated by Hr38 is still life (sif). sif encodes a Rac1 Guanine nucleotide Exchange Factor (GEF) that activates the small GTPase, Rac1, which drives synaptic growth required for s-LNV projection expansion. Gundermann et al. showed that knocking down sif for four hours at dawn inhibits projection expansion. In contrast, overexpressing sif for four hours at dusk is sufficient in expanding projections (Gundermann et al., 2023). Normally, sif translation is regulated by the RNA-binding protein Fragile X Mental Retardation Protein (FMRP). FMRP inhibits sif translation at dusk to reduce Rac1 activity and allow projections to retract (Gundermann et al., 2024). Additionally, the mammalian ortholog of Sif, TIAM1, is also required for plasticity in neurons by stabilizing dendrites and synapses (Gundermann et al., 2023). Since Hr38 modulates genes that are involved in projection expansion and sif regulates synaptic growth, I wanted to test whether sif is a target of Hr38.

Results

Hr38 and sif transcription increases at dawn

To investigate when sif and Hr38 are normally transcribed, I used in situ hybridization to monitor the transcription levels of both genes. Control yellow white (y w) larvae were grown and entrained in Light: Dark (LD) cycles at 25°C. ZT (Zeitgeber time) refers to the time in a 12 hour: 12 hour LD cycle with lights on from ZT0 to ZT12 and lights off from ZT12 to ZT24. Larval brains were dissected at three different timepoints: ZT23.75, ZT0.5, and ZT2. Since Hr38 is an activity-regulated gene, these timepoints were selected to monitor the rapid change in transcription after neuronal firing at dawn (Lymer et al., 2024). I used intronic sif and Hr38 probes to measure the transcription levels. Pre-mRNA transcripts contain introns because splicing had not yet occurred. Since splicing happens during transcription, the level of the pre-mRNA transcript can be used to measure transcription. Pigment Dispersing Factor (Pdf) encodes a neuropeptide specific to the LNvs, so I used an exonic Pdf probe to visualize larval LNvs (Gundermann et al., 2023). Since Pdf RNA levels remain constant throughout these timepoints, I normalized sif and Hr38 transcription levels to Pdf RNA to compare levels across time.

The data in Figure 1 show that Hr38 transcription levels were undetectable at ZT23.75. At ZT0.5 (“dawn”) Hr38 transcription levels were strongly detected and by ZT2, Hr38 transcription levels had become undetectable again. These data are consistent with Hr38’s behavior as an immediate early gene, which is characterized by a rapid increase and decrease in transcription. Figure 1 also shows that sif transcription was undetectable at ZT23.75. By ZT0.5 sif transcription is at high levels. In contrast to Hr38, sif transcription was still detectable at ZT2, although less than at ZT0.5. The results show that there is a correlation in the timing of the increase of Hr38 and sif transcription. This data are consistent with Hr38 regulating sif transcription, but this relationship needs further testing.

Figure 1. Time course of sif and Hr38 transcription in larval LNvs

A) Confocal images of larval LNvs dissected from control y w flies at ZT23.75, ZT0.5, and ZT2 for in situ hybridization. Images on the left show samples hybridized with intronic Hr38 (yellow), intronic sif (purple), and exonic Pdf (green) probes. Images on the right are the same as on the left, but they have a white dotted line instead of the Pdf channel to better visualize the signals from the intronic sif and Hr38 probes in the nucleus.

B) Quantification of the sif and Hr38 transcription levels normalized to Pdf. Error bars show SEM. These images are representative of 2 independent experiments, with at least 4 brains per time point in each independent experiment. Statistics used One-way ANOVA followed by Tukey’s multiple comparisons test. **** indicates p < 0.0001.

Inducing Hr38 at dusk upregulates sif transcription

To test if Hr38 expression modulates sif transcription levels, I performed an experiment to induce Hr38 transcription at dusk when it is not normally expressed. Since both Hr38 and sif are normally expressed at dawn, monitoring their transcription at dusk will allow us to learn if Hr38 transcription is sufficient to upregulate sif transcription. To perform the induction we used the Gal4/UAS system combined with a temperature sensitive repressor. Gal4 is a yeast transcription factor that homodimerizes and binds upstream activating sites (UAS) to drive gene expression. To restrict Gal4 expression to the s-LNvs, we used the Pdf-Gal4 transgene, which places Gal4 under the control of the Pdf promoter (Lymer et al., 2024). Additionally, the tubulin promoter is used to drive expression of the repressor Gal80^TS. At temperatures below 25°C, Gal80^TS is active and represses Gal4 activity, preventing Hr38 expression. Above 25°C, Gal80^TS is inactive and Gal4 is active, inducing Hr38 expression.

I used two ways to express Hr38: a UAS-Hr38 transgene to drive Hr38 expression when Gal4 binds to the UAS site or Hr38^EY fliesーwhich contains a P-element with Gal4 binding sites inserted ~2kb upstream of the start site of Hr38 transcription (Brody et al., 1999). Pdf-Gal4, tubulin-Gal80^TS flies were crossed to y w flies as controls, or to UAS-Hr38, or Hr38^EY flies. Larvae were grown and entrained in LD cycles at 18°C. At ZT12, the temperature was raised to 30°C for one hour to induce expression of Hr38. ZT12 was chosen as the induction start time because s-LNv projections are retracted and Hr38 is not endogenously expressed at dusk. Larval brains were dissected at ZT12 and ZT13. Hr38 and sif transcription levels were measured using the same in situ hybridization protocol as in Figure 1.

The data in Figure 2 shows that sif transcription levels were undetectable for the control at both ZT12 and ZT13. For Pdf-Gal4, tubulin-Gal80^TS > UAS-Hr38 and Pdf-Gal4, tubulin-Gal80^TS > Hr38^EY flies, sif transcription was not detected at ZT12, but had increased to detectable levels by ZT13. These results show that inducing Hr38 expression at dusk upregulates sif expression. However, the data in Figure 2 show that Hr38 transcription was undetectable at ZT12 and ZT13 in all genotypes. This result was expected for the UAS-Hr38 line because the transgene expresses an Hr38 cDNA, which lacks introns. Since the Hr38 probe used targets intronic sequences, it would not detect Hr38 from this construct. One possible reason for the lack of detectable Hr38 transcription in the Hr38^EY line is that inducing Hr38 triggered a negative feedback loop where elevated Hr38 levels led to suppression of Hr38 within the 1 hour time frame of the experiment.

Figure 2. Inducing Hr38 expression at dusk with Pdf-Gal4 increases sif transcription

A) Confocal images of larval LNvs dissected from Pdf-Gal4, tubulin-Gal80^TS > y w (Control), and Experimental Pdf-Gal4, tubulin-Gal80^TS > UAS-Hr38, and Pdf-Gal4, tubulin-Gal80^TS > Hr38^EY larvae at ZT12 and ZT13 for in situ hybridization. Control and experimental larvae were grown and entrained at 18°C. The temperature was induced to 30°C for one hour. Images show samples hybridized with intronic Hr38 (yellow), intronic sif (purple), and exonic Pdf (green) probes. These images are representative of 2 independent experiments, with at least 4 brains per time point in each experiment.

B) Quantification shows sif/Pdf or Hr38/Pdf transcription levels for the Control larvae and for Experimental Pdf-Gal4, tubulin-Gal80^TS > UAS-Hr38, and Pdf-Gal4, tubulin-Gal80^TS > Hr38^EY at ZT12 and ZT13. Error bars show SEM. Statistics used t-test. sif transcription significantly increased from ZT12 to ZT13 in both UAS-Hr38 and Hr38^EY genotypes (p < 0.0001). Hr38 transcription was not significantly different between time points in both experimental groups. No significant differences were observed in sif or Hr38 transcription levels between ZT12 and ZT13 in the control group.

As an alternative, I repeated this experiment using the Pura-Gal4 driver instead of Pdf-Gal4.

Figure 3 shows that when using the Pura-Gal4 construct, sif transcription levels were undetectable at ZT12 for all three genotypes. As with Pdf-Gal4, sif transcription becomes detectable at ZT13 for Pura-Gal4, tubulin-Gal80^TS > UAS-Hr38 and Pura-Gal4, tubulin-Gal80^TS > Hr38^EY flies. Additionally, Hr38 transcription levels were also not detected at ZT12 and ZT13 for all three Pura-Gal4, tubulin-Gal80^TS fly lines. These results show that inducing Hr38 is sufficient in upregulating sif transcription.

Figure 3. Inducing Hr38 expression at dusk with Pura-Gal4 increases sif transcription

A) Confocal images of LNvs dissected from Control Pura-Gal4, tubulin-Gal80^TS > y w, and Experimental Pura-Gal4, tubulin-Gal80^TS > UAS-Hr38, and Pura-Gal4, tubulin-Gal80^TS > Hr38^EY larvae at ZT12 and ZT13 for in situ hybridization. Larvae were entrained and induced as in Figure 2. Images show samples hybridized with intronic Hr38 (yellow), intronic sif (purple), and exonic Pdf (green) probes. These images are representative of 1 experiment, with at least 4 brains per time point in each experiment.

B) Quantification shows sif/Pdf or Hr38/Pdf transcription levels for the control, Pdf-Gal4, tubulin-Gal80^TS > UAS-Hr38, and Pdf-Gal4, tubulin-Gal80^TS > Hr38^EY at ZT12 and ZT13. Error bars show SEM. Statistics used t-test. sif transcription significantly increased from ZT12 to ZT13 in both UAS-Hr38 and Hr38^EY genotypes (p < 0.0001). Hr38 transcription was not significantly different between time points in both experimental groups. No significant differences were observed in sif or Hr38 transcription levels between ZT12 and ZT13 in the control group.

RFP-Positive LNvs were isolated using Fluorescent Activated Cell Sorting (FACS)

To identify additional target genes of the transcription factor Hr38 apart from sif, I used Fluorescent Activated Cell Sorting (FACS) to isolate and sort larval LNvs based on RFP fluorescence. This isolation was done in preparation for RNA sequencing, with the goal of identifying target genes regulated by Hr38. As a transcription factor, Hr38 regulates downstream genes in response to neuronal activity. By inducing Hr38 we expected these activity-related genes to be upregulated.

Pdf-Gal4, Pdf-RFP; tubulin-Gal80^TS flies were crossed to control y w and experimental Hr38^EY flies. Control and experimental larvae were grown and entrained at 18°C. The temperature was induced to 30°C for one hour. Larval brains were dissected at ZT13 following the one hour temperature induction. y w larval brains were also dissected to act as a control for FACS gates. The brains were enzymatically and mechanically dissociated to separate brains into a single cell suspension in preparation for FACS.

The negative y w sample, which does not express RFP fluorescence, was used to set FACS gates, as shown in the y w FACS plot in Figure 4. Once the gates were defined, Pdf-RFP; tubulin-Gal80^TS > y w and Pdf-RFP; tubulin-Gal80^TS > Hr38^EY samples were sorted to isolate RFP-positive LNvs. Figure 4A shows that RFP-positive LNvs fall within the gated region of both the control and experimental samples. The results show that RFP-positive LNvs were isolated from both Pdf-RFP; tubulin-Gal80^TS > y w and Pdf-RFP; tubulin-Gal80^TS > Hr38^EY larvae. We next performed RNA sequencing to compare expression levels between control and experimental RFP-positive LNvs to identify genes regulated by Hr38.

cDNA libraries prepared for RNA sequencing

To identify genes regulated by Hr38, Dr. Jennifer Lennon, a postdoc in the lab, performed RNA sequencing on the RNA isolated from RFP-positive LNvs. RNA extraction was performed to separate the RNA located in the aqueous layer from the organic layer. The RNA was then cleaned and concentrated and a binding buffer was used to elute purified RNA. The NEBNext Single Cell/Low Input Kit was then used to create cDNA libraries. Afterwards, the cDNA concentration of the experimental and control samples were measured and barcodes were added for identification. The quality of the libraries were then analyzed on the Agilent 4200 TapeStation. Figure 4B shows the TapeStation results for the control and experimental libraries. The results show three main peaks. The sharp Upper and Lower peaks represent the ladder and the broader middle peak represents the libraries. For the control, the library size was between 95 bp and 744 bp. For the experimental, the library is between 110 bp and 688 bp. The concentration of the control and experimental libraries were 0.0690 ng/µL and 0.0630 ng/µL, respectively. The libraries were then pooled then sequenced using the NextSeq 500 System. I then analyzed the data on R using DesSeq2. Overall, approximately 5,000 genes exhibited increased fold changes in the experimental sample in comparison to the control. From the 5,000 genes, sif was upregulated by 4.5 fold and Hr38 was downregulated by 15 fold. The sif results support sif as a candidate gene and the Hr38 results support the negative feedback loop mechanism mentioned in the induction experiment. Although both results support previous findings, one replicate with a large pool of genes expressed makes it premature to draw definitive conclusions without additional replicates.

Figure 4: RFP Sorted LNvs using FACS and preparation of y w and Hr38^EY cDNA libraries

A) On the left is a representative whole population FACS plot of single cell suspension of 15 non-fluorescent y w larval brains. Autofluorescence was removed with a BFP- gate. The middle is a representative whole population FACS plot of 48 Pdf-RFP; tubulin-Gal80^TS > y w larval brains. 1.3 million events were run on FACS-S8. The top right is a representative whole population FACS plot of 46 Pdf-RFP; tubulin-Gal80^TS > Hr38^EY larval brains. 1.9 million events were run on FACS-S8. The x-axis represents RFP detectors and the y-axis represents GFP intensity for all plots. RFP cells are located within the black gated outline. All sorting was performed by the BD FACSDiscover S8 Cell Sorter.

B) TapeStation results of control and experimental cDNA libraries performed on Agilent 4200 TapeStation. The x-axis represents size in base pairs and the y-axis represents normalized fluorescent sample intensity. Peaks labeled “Upper” and “Lower” represent the ladder. The left is RFP-positive cells from the Pdf-RFP; tubulin-Gal80^TS > y w FACS sort in 4A. The right is the RFP-positive cells from the Pdf-RFP; tubulin-Gal80^TS > Hr38^EY FACS sort in 4A.

Discussion

I have shown that transcription of both Hr38 and sif increases at dawn in LNvs and that Hr38 is sufficient to upregulate sif transcription at dusk. This supports the idea that sif is regulated by Hr38 on a transcription level. Additionally, Gundermann et al. had shown that sif is also regulated post-transcriptionally by FMRP. FMRP represses sif translation at dusk by binding sif mRNA (Gundermann et al., 2023). Additionally, knocking down the gene encoding FMRP, FMr1, at dusk can be rescued by simultaneously reducing sif expression (Gundermann et al., 2023). This double layer of sif regulation ensures that sif expression is tightly controlled, allowing effective modulation of structural plasticity mechanisms. Post-translation, sif genetically interacts with cell adhesion molecule Fas2, a protein required for synaptic growth at the neuromuscular junction (Sone et al., 2000). This suggests that regulating sif transcription is not only important for the timing of projection changes, but it can also influence molecular interactions that contribute to synaptic remodeling. Overall, this shows that sif is a key regulator in signaling networks and confirming if sif is a direct target gene of Hr38 is relevant in understanding structural plasticity.

Additionally, we have performed FACS and RNAseq to identify the direct target genes of Hr38 in LNvs. Approximately 30% of genes in the Drosophila genome underwent a fold change increase. However, one limitation of the RNA sequencing results is that I only performed one replicate, so we were not able to perform statistical analysis. Additionally, I found that Hr38 had a 15 fold decrease, which was not what I expected. However, this aligns with the idea presented after the induction experiments that Hr38 underwent a negative feedback loop since its mRNA levels decreased greatly in the experimental versus the control. Alternatively, this could also be due to any leaky expression of Hr38 prior to the induction. This could affect the apparent fold change in Hr38 after induction as well as the amount of genes that experienced increased expression. If the latter occurred, then this means that not only direct target genes of Hr38 were expressed, but also other downstream genes beyond Hr38 targets. One method to reduce the genes that were amplified would be to shorten the induction time. In the first experiment, Hr38 transcription levels increased rapidly 30 minutes after dawn. Additionally, another activity-regulated gene, c-fos, showed mRNA accumulation 30 minutes after stimulation (Perrin-Terrin, 2016). With this, performing a 30 minute induction may allow us to narrow down the RNA sequencing results to just the direct target genes of Hr38. Another issue could be that it took 45 minutes for 3 people to dissect 48 and 46 brains for the control and experimental, respectively. This means that some brains underwent longer inductions than only one hour, which can also impact the number of genes expressed. Performing a 30 minute induction may also help minimize the effects of longer dissection times. I also found that sif increased 4.5 fold from the control to experimental. These results support the induction experiment results that sif was upregulated by Hr38 expression. However, without more RNAseq replicates and modifying the induction time to be shorter, we cannot yet conclude if sif is a direct target gene. Furthermore, additional experiments to confirm that sif is a direct target gene would include identifying Hr38 binding sites in sif’s regulatory region and testing if deleting the binding site inhibits sif transcription.

Other genes that showed an increased fold change included Mef2, Dpr8, Pde9, unc-13, and syt1. Mef2 is a transcription factor involved in activity-dependent transcription that is required for expanding s-LNv projections (Gundermann et al., 2023, Sivachenko et al., 2013). Additionally, in ethanol Mef2 activates Hr38 (Adhikari et al., 2019). Dpr8 is involved in synapse organization and acts as a cell adhesion molecule with Dpr Interacting Proteins (DIP) (Cheng et al., 2019). This protein interaction controls synaptic connectivity during development (Sergeeva et al., 2020) and therefore may similarly regulate synaptic growth in the LNvs. Pde9 regulates cGMP and cAMP levels by enabling cGMP and cAMP phosphodiesterase activity. Moreover, Pde9 inhibition was found to strengthen synaptic plasticity at the level of long-term potentiation (Dorner-Ciossek et al., 2017). Moving on to unc-13, it is involved in synaptic vesicle tethering in vesicle exocytosis. Lastly, syt1 is involved in calcium ion regulated exocytosis of neurotransmitters (Bouazza-Arostegui et al., 2022). All of the genes mentioned above are involved in mechanisms that are centered around synaptic changes, which mean they influence plasticity and may drive changes in s-LNv projections. However, since approximately 5,000 genes were up-regulated in the experimental RNAseq analysis, it is not surprising that many synaptic genes were upregulated.

To identify direct target genes of Hr38, future experiments will need a shorter induction, such as the 30 minutes suggestion above. This way, secondary target genes will not be expressed and a narrow number of genes will experience fold changes. Once this smaller pool of genes is analyzed, experiments can be performed to test whether these genes affect structural plasticity. The first experiment will mirror the induction experiments described previously by testing whether activating a target gene at the wrong time changes the 3D spread of s-LNv projections. This will be performed by gain of function experiments at dusk to see if projections will expand and appear in a dawn-like state. Conversely, performing loss of function experiments at dawn will test if projections will appear retracted as in the dusk-like state. This approach is connected to prior loss of function experiments that tested whether sif and Hr38 were sufficient in expanding projections at dusk. In Gundermann et al., inducing sif expression for four hours at dusk expanded projections. Similarly, Lymer et al. showed that inducing Hr38 expression for two hours at dusk expanded projections. These experiments are important because identifying target genes relevant to structural plasticity can inform potential therapeutics on how to better modulate this network in conditions with altered plasticity mechanisms.

Conclusion

This study focused on identifying target genes of the transcription factor Hr38. By monitoring normal sif and Hr38 transcription, I found that transcription of both genes increased at dawn. Inducing Hr38 at dusk showed a causal relationship between Hr38 and sif expression. These results support Hr38 activating sif transcription. To further test this, disrupting Hr38 binding sites on sif’s regulatory region can confirm sif as a direct target gene. Identifying additional target genes was done using FACS and RNAseq; however, more replicates are needed to validate the results. Identifying the target genes of Hr38 is important to understand plasticity on a molecular level. Investigating this mechanism can help develop potential targets for therapeutics for neurological disorders.

Material and Methods

In-situ hybridization and imaging

Larval dissections were performed in PBS and the brains were immediately transferred to 2% PFA. Once all dissections were complete, the brains were further fixed in 4% PFA for 20 minutes. To dehydrate the brains, I used a methanol series on ice, starting from 25% methanol and ending in 100% methanol. The brains were washed in each concentration for 5 minutes. After the final wash in 100% methanol the brains were stored at -20°C. To prepare for permeabilization, the brains were rehydrated by repeating the methanol series in reverse order. To permeabilize the brains, 5% acetic acid was added for 5 minutes then the brains were washed with PBT on ice. To refix the brains, 4% PFA is added at room temperature.

To prepare for the hybridization, the brains are washed in 5X SSCT, then placed in a hybridization solution made by Molecular Instruments at 37°C for 30 minutes. To monitor transcription levels of the s-LNvs, 2µL of intronic Hr38 and sif probes that recognized the first and 14th intron, respectively were added to the hybridization buffer. Additionally, 2µL of an exonic Pdf probe that recognizes the single exon was also added into the buffer and placed into an incubator at 37.5°C for 24 hours. Afterwards, the probe/buffer mix was removed and the brains were washed with a wash solution (Molecular Instruments). The brains were equilibrated in an amplification buffer (Molecular Instruments) prior to adding the amplification probes. To promote efficient hybridization and amplify the sif, Hr38, and Pdf signal, HCR amplification probes synthesized by Molecular Instruments were used. To ensure a stronger signal, the brains were left overnight in the probe solution in dark at 25°C. To remove excess unbound probes, the amplification probe solution were removed and the brains were washed with SSCT. The brains were then mounted with SlowFade and imaged using a Leica Stellaris confocal microscope. All probes were from Molecular Instruments with specific details provided below.

In-situ induction experiment

To test whether Hr38 expression at dusk impacts sif expression, we performed a temperature-mediated induction experiment. Control y w flies and experimental UAS-Hr38 and Hr38^EY flies were crossed to Pura-Gal4; tub-Gal80^TS flies. Flies are usually grown at 25°C unless there is a temperature-sensitive transgene tubulin-Gal80^TS. With this, the progeny were raised and entrained under a LD cycle for 4 days at 18°C. At ZT12, the flies underwent a temperature shift to 30°C for 1 hour, allowing the tub-Gal80^TS repressor to be inactivated and induce Gal4 activity. To compare sif and Hr38 expression prior to Hr38 activation, fly brains from all three lines mentioned were dissected and fixed prior to induction at ZT12 and immediately after the 1 hour induction at ZT13. The brains were then taken through the in-situ hybridization protocol mentioned prior and imaged using the Leica SP8 confocal microscope.

Fluorescence-Activated Cell Sorting

I used Fluorescence-Activated Cell Sorting (FACS) to sort larval LNvs. Control y w flies and experimental Hr38^EY flies were both crossed to Pdf-Gal4, Pdf-Rfp; tub-Gal80^TS flies. The progeny were grown and induced in the same conditions as the induction experiment. At ZT13, we dissected brains from both fly lines in cold 1x PBS and transferred them to low-bind eppendorf tubes (Fisher Scientific) containing 1mL of 1X PBS. After PBS removal, brains were dissociated in a solution of trypsin in EDTA and incubated in a ThermoMixer at 25°C at 500 rpm for 30 minutes. Afterwards, the enzyme solution was removed and the brains were washed 3 times with 1x DPBS, making sure the brains rested on ice between each wash. Then the brains are washed with 1x DPBS + 0.04% BSA 3 times to minimize cell clumping and protect the s-LNvs, also resting the brains on ice in between. Afterwards, the solution is removed and resuspended using fresh 1x DPBS + 0.04% BSA. The brains were then pipetted up and down 50 times for mechanical dissociation. This was repeated with fresh 1x DPBS + 0.04% BSA 2 more times. The samples were then brought to a final volume of 500 µL with the 1 DPBS + 0.04% BSA solution and filtered through a 20µL pluriStrainer (pluriSelect) into a low-bind eppendorf tube. The samples were then centrifuged at 600 rpm for 1 minute. The samples were then transferred to low-bind test tubes on ice and sorted by the BD FACSDiscover S8 Cell Sorter using RFP fluorescence to isolate the s-LNvs. After performing FACS, stringent gates that only included RFP positive LNvs were made.

RNA extraction and library preparation

The following protocol was performed by Dr. Jennifer Lennon, a postdoc in the lab. The cells were sorted into TRIzol (Life Technologies, Cat. No 15596026). A TRIzol/chloroform RNA extraction was performed, creating a clear aqueous layer containing RNA, and a pink organic layer. After separating the clear aqueous layer, the RNA was cleaned and concentrated using the RNA Clean & Concentrator kit (Zymo Research, Cat. No R1013). This led to the elution of 6µL of purified RNA.

To create cDNA libraries, the NEBNext Single Cell/Low Input Kit (New England Biolabs, Cat. No E6420S) was used. To anneal cDNA primers NEBNEXT RT Primer mix was added to the lysate and heated to 70°C for 5 minutes. Reverse transcription was then performed to generate cDNA, which was amplified by 18 cycles of PCR. The cDNA cleanup was performed using AMPure XP bead purification and washing the sample with 80% ethanol. The cDNA was eluted from the beads in 0.1x TE buffer. The cDNA concentration was measured using the Qubit Fluorimeter (ThermoFisher Scientific). The cDNA concentration of Hr38^EY was 0.0630 ng/µL and the y w control was 0.0690 ng/µL. Afterwards the barcodes i702 and i502 were added to the Hr38^EY cDNA and i701 and i501 were added to control cDNA to identify the sequences. The quality and quantity of the libraries were checked on an Agilent TapeStation 4200 and the final cDNA concentration was measured using the Qubit. The final Hr38^EY concentration was 0.72 ng/µL and the final y w concentration was 0.684ng/µL. To prepare for next-generation sequencing, 3nM of each library (4µL of each) was pooled in a 24µL solution containing 10mM of Tris. The library/solution was then sent to be sequenced using the NextSeq 500 System (2x100 bp read configuration, mid output version 2.5, Illumina).

Supplementary Information

Acknowledgements

I would like to thank my PI, Justin Blau, for all of the support you have given me throughout this project. All of the advice whether it was for research or academics helped me grow and challenge myself outside of the lab. To Jenny Lennon, thank you for teaching me the ins-and-outs of the lab and answering all of my questions without judgment. Thank you for always helping me in dissections, planning experiments, and imaging brains even when it takes all day. To all my fellow undergrads, thank you for helping me dissect even when it was early and for all of the support you all have shown me. I have learned so much from all of you and would not have been able to do this without you all.

Specific aims - December 2024

Circadian rhythm controls structural plasticity, which is the ability for neurons to physically alter their connections. In the model organism used, Drosophila, small ventral lateral neurons (s-LNvs) are under circadian control and expand and retract every 24 hours. At dusk the neural projections retract and at dawn the projections expand. Structural plasticity is key in encoding memories, learning, and brain development. Health conditions associated with altered structural plasticity networks include autism, PTSD, and schizophrenia. However, the mechanism behind structural plasticity is not fully understood. Understanding the regulators behind this process can help us learn more about how to control structural plasticity and develop future treatments. Our long-term goal is to understand the molecular mechanisms behind structural plasticity. The specific objective of this proposal is to identify the target genes of the transcription factor Hr38 in small ventral lateral neurons (s-LNvs) in Drosophila. Hr38 (Hormone receptor-like in 38) is an activity-regulated gene, which means its transcription is activated by neuronal activity rapidly. This also means that Hr38 is involved in activating later genes involved in expanding s-LNvs projections. The central hypothesis is that sif is a potential target gene of Hr38 that controls structural plasticity? sif encodes Rac1 GEF required for s-LNvs plasticity rhythms by activating Rac1, which leads to actin polymerization and later structural expansion. This hypothesis was formulated in part based on preliminary data which shows that overexpression of Hr38 upregulates sif transcription within a 1-2 hour period, leading to s-LNv expansion. The outcome will tell us if Hr38 is a direct or indirect transcription factor of sif and if there are other target genes. Additionally, we will further test if the other potential target genes affect s-LNv plasticity. We will pursue these studies in three Specific Aims.

Aim 1: To investigate whether still life (sif) is transcribed at similar time as Hr38

We will test the prediction that sif is transcribed at similar time as Hr38. To monitor sif transcription within the 1-2 hour window from preliminary data we will perform in-situ hybridizations time points ZT23.75, ZT0.5, ZT2. This would allow us to determine whether sif transcription is correlated with Hr38. A potential outcome would be that shortly after Hr38 transcription occurs, sif transcription follows, which would support our model that sif is a candidate direct target gene of Hr38.

Aim 2: Test whether inducing Hr38 increases sif expression at dusk

We will test whether the inducing Hr38 increases sif expression at dusk by using the Gal4-UAS binding system to induce expression of Hr38. This system will use a temperature sensitive repressor that will repress Hr38 at 18°C but become inactive at 30°C. We will induce Hr38 at 30°C at ZT12 for one hour. We will dissect larvae at ZT12 and ZT13. Then we will perform an in situ hybridization to monitor sif and Hr38 transcription levels at both timepoints. This protocol will be performed using 3 fly genotypes: Pdf-Gal4, tubulin-Gal80^TS > UAS-Hr38 and Pdf-Gal4, tubulin-Gal80^TS > Hr38^EY flies as the experimental and Pdf-Gal4, tubulin-Gal80^TS > y w as the control.

Aim 3: To identify additional target genes of Hr38 using an unbiased approach

In this aim, we will expand our knowledge of Hr38 target genes in s-LNv neurons. To identify Hr38 specific targets we will isolate s-LNv neurons from yellow white (yw) control flies with normal hr38 transcription levels and Hr38EY flies, which is the overexpression of Hr38. To purify RNA from the s-LNvs from the cell supernatant we will use Fluorescent Activated Cell Sorting. Then we will make RNAseq libraries and compare the control yw target genes to the Hr38EY target genes. To compare the RNAseq results we will quantify this data using DESeq2. These studies will complement our approach in Aim 2 to study Hr38’s role as a transcription factor in structural plasticity mechanisms.

The proposed work is innovative because it utilizes in situ hybridizations and RNAseq to uncover Hr38 target genes which are currently unknown. At the completion of this project, we expect that the combined work in Aims 1 and 2 will demonstrate an interaction of Hr38 and sif . Moreover, Aim 3 further investigates additional target genes, and confirms whether sif is a direct target gene. This knowledge will open a new field of research and lead to potential applications in treatments regarding strengthening, weakening, or modulating structural plasticity networks.

References

​​Adhikari, P., D. Orozco, H. Randhawa and F. W. Wolf, 2019 Mef2 induction of the immediate early gene Hr38/Nr4a is terminated by Sirt1 to promote ethanol tolerance. Genes Brain Behav 18: e12486.

Blanchard, F. J., B. Collins, S. A. Cyran, D. H. Hancock, M. V. Taylor et al., 2010 The transcription factor Mef2 is required for normal circadian behavior in Drosophila. J Neurosci 30: 5855-5865.

Bouazza-Arostegui, B., M. Camacho, M. M. Brockmann, S. Zobel and C. Rosenmund, 2022 Deconstructing Synaptotagmin-1's Distinct Roles in Synaptic Vesicle Priming and Neurotransmitter Release. J Neurosci 42: 2856-2871.

Bozelos, P., and P. Poirazi, 2017 Chapter 15 - Impact of Structural Plasticity on Memory Capacity, pp. 319-341 in The Rewiring Brain, edited by A. van Ooyen and M. Butz-Ostendorf. Academic Press, San Diego.

Brody, T., 1999 The Interactive Fly: gene networks, development and the Internet. Trends Genet 15: 333-334.

Cheng, S., J. Ashley, J. D. Kurleto, M. Lobb-Rabe, Y. J. Park et al., 2019 Molecular basis of synaptic specificity by immunoglobulin superfamily receptors in Drosophila. Elife 8.

Dorner-Ciossek, C., K. S. Kroker and H. Rosenbrock, 2017 Role of PDE9 in Cognition. Adv Neurobiol 17: 231-254.

Drysdale, R., 2008 FlyBase : a database for the Drosophila research community. Methods Mol Biol 420: 45-59.

Dubowy, C., and A. Sehgal, 2017 Circadian Rhythms and Sleep in Drosophila melanogaster. Genetics 205: 1373-1397.

Gage, F. H., 2004 Structural plasticity of the adult brain. Dialogues Clin Neurosci 6: 135-141.

Gipson, C. D., and M. F. Olive, 2017 Structural and functional plasticity of dendritic spines - root or result of behavior? Genes Brain Behav 16: 101-117.

Gu, S.-H., Y.-C. Hsieh and P.-L. Lin, 2016 Stimulation of orphan nuclear receptor HR38 gene expression by PTTH in prothoracic glands of the silkworm, Bombyx mori. Journal of Insect Physiology 90: 8-16.

Gundermann, D. G., S. Lymer and J. Blau, 2023 A rapid and dynamic role for FMRP in the plasticity of adult neurons. bioRxiv.

Lara Aparicio, S. Y., Á. d. J. Laureani Fierro, G. E. Aranda Abreu, R. Toledo Cárdenas, L. I. García Hernández et al., 2022 Current Opinion on the Use of c-Fos in Neuroscience. NeuroSci 3: 687-702.

Lymer, S., K. Patel, J. Lennon and J. Blau, 2024 Circadian clock neurons use activity-regulated gene expression for structural plasticity. bioRxiv.

Perrin-Terrin, A. S., F. Jeton, A. Pichon, A. Frugiere, J. P. Richalet et al., 2016 The c-FOS Protein Immunohistological Detection: A Useful Tool As a Marker of Central Pathways Involved in Specific Physiological Responses In Vivo and Ex Vivo. J Vis Exp.

Petsakou, A., T. P. Sapsis and J. Blau, 2015 Circadian Rhythms in Rho1 Activity Regulate Neuronal Plasticity and Network Hierarchy. Cell 162: 823-835.

Sergeeva, A. P., P. S. Katsamba, F. Cosmanescu, J. J. Brewer, G. Ahlsen et al., 2020 DIP/Dpr interactions and the evolutionary design of specificity in protein families. Nat Commun 11: 2125.

Sheng, M., and M. E. Greenberg, 1990 The regulation and function of c-fos and other immediate early genes in the nervous system. Neuron 4: 477-485.

Sivachenko, A., Y. Li, K. C. Abruzzi and M. Rosbash, 2013 The transcription factor Mef2 links the Drosophila core clock to Fas2, neuronal morphology, and circadian behavior. Neuron 79: 281-292.

Sone, M., E. Suzuki, M. Hoshino, D. Hou, H. Kuromi et al., 2000 Synaptic development is controlled in the periactive zones of Drosophila synapses. Development 127: 4157-4168.

Zhang, H., and C. R. Bramham, 2021 Arc/Arg3.1 function in long-term synaptic plasticity: Emerging mechanisms and unresolved issues. Eur J Neurosci 54: 6696-6712.