CG4968 positively regulates the immune deficiency pathway by targeting Imd protein in Drosophila

Drosophila strain and husbandry

All of the Drosophila utilized in this study were raised on standard Drosophila medium, which consisted of agar and maize meal. Drosophila were reared at 25 °C and 12 h of alternating light and dark. The control and host for P-element-mediated transformation was the w¹¹¹⁸ strain. In this investigation, the following Gal4 lines and transgenes were employed: (1) P{UAS-CG4968^KD} obtained from Tsinghua Drosophila Centre (No. TH04864.N); (2) P{ppl-Gal4} obtained from Bloomington Stock Center; and (3) P{UAS-CG4968^OE} and P{UAS-CG4968^ΔOTU}, in which the full-length CG4968 and CG4968^ΔOTU were placed under the control of the UASp promoter. Among these, P{ppl-Gal4} could drive the specific high expression of genes in the fat bodies of adult flies.

RNAi knockdown assays in S2 cells

We conducted synthesis of dsRNA using the T7 in vitro transcription kit (QIAGEN, Hilden, Germany) and RNase-Free throughout the experiment. All dsRNAs were synthesized according to the standard protocol. The specific primers of dsRNA were first designed using Primer Premier 5.0 software, and polymerase chain reaction amplification was then carried out. The amplified fragment was subjected to T7 transcription in vitro to synthesize dsRNA. At a density of 1 × 10⁶ cells per ml, S2 cells were harvested, diluted into fresh medium, and immediately treated with dsRNA. S2 cells were treated with dsRNA for 48 h, and then transfected with related plasmids. Primers used for dsRNA production are listed in Table S1.

Drosophila S2 cell transfection and luciferase reporter assays

Drosophila S2 cells were kept in our laboratory for long-term storage (Hua et al., 2022). At 27 °C, S2 cells were grown in insect medium (Gibco, Waltham, MA, USA) with 10% fetal bovine serum (Hyclone, Logan, UT, USA). All of the cells utilized in this investigation were transfected using Lipofectamine 2000 (Invitrogen, Waltham, MA, USA). Transfection was divided into two Groups: in Group 1, plasmid and Opti-MEM were mixed at a ratio of 1 μg:50 μl, and in Group 2, plasmid and Lipofectamine 2000 were mixed at a ratio of 1 μg:2 μl. Then, the liquids of Group 1 and Group 2 were fully mixed. The S2 cells were collected 48 h after plasmid transfection.

For the luciferase reporter assays, we used the Drosomycin-luciferase or Attacin-luciferase constructs in which the luciferase coding sequence was put under the Drosomycin or Attacin promoter, Toll, and IMD signaling reporter assays in S2 cells. The procedure used to measure Firefly luciferase and Renilla luciferase was as follows. In order to perform luciferase experiments using a luciferase assay reagent (Promega), the S2 cells were lysed with 50 ml of Passive lysis buffer (Promega), according to the manufacturer’s instructions, and 20 μl cell lysate of each sample was placed in a 96-well plate containing luciferase substrate. Finally, the ratio of Firefly luciferase and Renilla luciferase was calculated, and three biological repelicates were performed.

qRT-PCR assays

All samples were extracted using Trizol Reagent (Invitrogen, Waltham, MA, USA) to determine total RNA. In the following step, cDNA was synthesized using the first-strand cDNA synthesis kit (Transgen, Beijing, China) according to the manufacturer’s instructions. On a Light Cycler 480, qRT-PCR was performed in triplicate using SYBR Green Master Mix (Thermo Fisher Scientific, Waltham, MA, USA). The PCR reaction procedure was as follows: denaturation at 95 °C for 30 s, annealing at 55 °C for 30 s, and extension at 72 °C for 30 s for a total of 35 cycles. Template concentrations were normalized to an endogenous reference, Rp49. The data used in the detection were all analyzed using the 2^−ΔΔct method. Three biological replicates were performed. The primers for qRT-PCR were designed using Primer Premier 5.0 software. The primers for qRT-PCR production are shown in Table S2.

Western blotting and co-immunoprecipitation (Co-IP)

We collected cell samples and lyse in buffer (150 mM NaCl, 50 mM Tris-HCl, pH 7.5, 10% glycerol, 0.5% Triton X-100, 10 μg/ml aprotinin, 10 μg/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride) for 40 min. The processed protein samples were put onto a protein gel for SDS-PAGE analysis, which was then run at 100 V for 30 min followed by 150 V for 30 min. The protein samples were then electrotransferred on ice using PVDF membranes (Thermo Fisher Scientific, Waltham, MA, USA) at a constant voltage of 100 V, blocked with skim milk, and then washed with PBST. Primary antibodies were incubated overnight at 4 °C and secondary antibodies were incubated at room temperature for 2 h. Both were washed with PBST. The following antibodies were used for Western blot: mouse anti-His (1:2,000; HRP), mouse anti-GST (1:3,000; Abcam, Cambridge, UK), mouse anti-Flag (1:3,000; Sigma-Aldrich, St. Louis, MO, USA), rabbit anti-Myc (1:2,000; Medical & Biological Laboratories, Nagoya, Japan), mouse anti-Tubulin (1:2,000; Cowin, Cambridge, MA, USA), rabbit anti-HA (1:1,000; Medical & Biological Laboratories, Nagoya, Japan). Goat anti-mouse IgG H&L (1:2,500; Abcam, Cambridge, UK) and goat anti-rabbit IgG H&L (1:2,500; Abcam, Cambridge, UK) were the secondary antibodies utilized for Western blot.

For the Co-IP experiment, cells were collected after plasmid transfection for 48 h and lysed in buffer-1 (150 mM NaCl, 50 mM Tris-HCl, pH 7.5, 10% glycerol, 0.5% Triton X-100, 10 μg/ml aprotinin, 10 μg/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride) for 40 min. Cell lysate was combined with Anti-Flag affinity gel (Sigma) at 4 °C for 4 h. Buffer-2 (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 5 mM EDTA, and 0.5% Igepal CA-6030) was used to wash the immune complexes three times for 20 min each time. Finally, the corresponding antibody was used for immunoblotting.

Protein purification and GST pull-down assays

The Escherichia coli (E.coli) strain BL21 was used to produce and purify the GST-tagged or His-tagged proteins. Isopropyl-beta-D-thiogalactopyranoside (IPTG) was added to increase the protein expression final dosage of 1 mM into BL21 bacterial medium when the culture’s OD600 reached 0.5–0.7, then inducted at 16 °C in a 300 rpm shaker overnight. Bacteria cells were pelleted and resuspended in lysis buffer (25 mM Tris-HCl, pH 7.5, 100 mM NaCl, and 2 mM EDTA). Each sample was sonicated for 40 min using an EpiShearTM Probe Sonicator (pulse 6s on, 10s off, 40% amplitude) to disturb the integrity of the cells. After centrifuging the sample at 10,000 rpm for 40 min at 4 °C, the protein concentration was extracted from the supernatant. For the purification of GST-tagged or His-tagged proteins, BeyoGoldTM His-tag Purification Resin (Beyotime, Nantong, China) or Glutathione Sepharose (Sigma, St. Louis, MO, USA) were employed, respectively. Finally, loading buffer was added to each sample (1 mg), and the Bradford assay was performed. The appropriate proteins were treated with Glutathione Sepharose at 4 °C for 4 h to perform the GST pull-down. The samples were then Western blot tested after being washed three times with PBS and 1% TritonX-100 for a total of 1 h.

In vivo ubiquitination assay

Transfected S2 cells were collected and lysed in buffer A (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.5% Nonidet P-40, 10% glycerol and 1% SDS) for 40 min. After heating at 95 °C for 10 min and mixed thoroughly with buffer B (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.5% Nonidet P-40, and 10% glycerol). After ultrasonic treatment of the lysate (pulse 4s on, 6s off, 30% amplitude), Anti-Flag beads (Sigma, St. Louis, MO, USA) were used to bind the lysate at 4 °C for 6 h. The conjugate was washed three times for 20 min each with washing buffer (50 mM Tris-HCl, pH 7.5, 500 mM NaCl, 0.5% Nonidet P-40, and 10% glycerol). Western blot analysis was then performed to detect the ubiquitination pattern of the specified protein and analysed using Image J software.

Survival and bacterial CFU assays

Bacterial infections were carried out using the following method. A small needle dipped in a concentrated overnight culture of bacteria was used to puncture 6-day-old adult flies. Flies were transferred to a new medium each day after infection. In the survival study score, 30 flies per group were kept in the same environment. The number of flies that survived was counted daily. Flies that died within 3 h of bacterial infection were not included in this study. For the CFU counting test, 10 infected flies were taken at the appropriate time after infection and crushed in PBS. We plated 100 μl of each dilution on LB agar plates. After overnight growth in an incubator at 30 °C, bacterial colonies were counted.

Statistical analysis

GraphPad Prism 8 was used to perform all statistical analyses. The mean and standard errors of the data were displayed. Except for lifetime assays, which employed the log-rank test for statistical analysis, significance was assessed using two-tailed Student’s t tests. p < 0.05 was considered statistically significant (ns, not significant; *p < 0.05; **p < 0.01; ***p < 0.001).

CG4968 is a positive regulator of the IMD pathway

Recent research has found that Dubs play an important role in regulating the immune pathway in Drosophila (Hua et al., 2022; Ji et al., 2019). Our excavation of the Drosophila genome resulted in the discovery of a functionally undefined Dub gene, CG4968 (Gene ID: 34384). Analysis of the domain of CG4968 through the Flybase online website (https://flybase.org/) showed that it has an OTU domain, indicating that CG4968 is a typical member of the OTU Dubs family (Fig. 1A). To determine whether CG4968 is involved in the regulation of innate immunity in Drosophila, we first designed dsRNAs targeting the CG4968 coding sequence and the 3′-untranslated region (dsRNA 49-1, dsRNA 49-3′UTR). The qRT-PCR results showed that both of the two groups of dsRNAs significantly knocked down the expression of CG4968 in Drosophila S2 cells (Fig. 1B). We then constructed an Attacin-Luciferase reporter system (Att-Luc), and the firefly luciferase gene was placed under the control of the Attacin promoter. The IMD pathway was activated by expressing the active form of IMD, resulting in the activation of luciferase expression driven by the Attacin gene promoter, and showed dosage and temporal trends (Fig. S1). Att-Luc reporter assay revealed that the knockdown of CG4968 expression in S2 cells significantly reduced Att-Luc activity activated by IMD overexpression (Fig. 1C). Additionally, we examined the effect of the knockdown of CG4968 in S2 cells on the expression of antimicrobial peptides (AMPs) of the IMD immune pathway. qRT-PCR results showed that the knockdown of CG4968 significantly reduced the expression of AMPs Attacin (Att) and Cecropin (Cec) downstream of the IMD pathway compared to the control group, which was similar to the results of the Att-Luc reporter assay (Figs. 1D and 1E).

To further explore the association of CG4968 with the IMD pathway, CG4968 was overexpressed in S2 cells. The results showed that CG4968 overexpression significantly increased Att-Luc activity induced by the active form of IMD (Fig. 1F). Additionally, the expression levels of Att and Cec in the IMD pathway increased significantly with the increase of CG4968 concentration (Figs. 1G and 1H). The above results suggested that the Dub CG4968 positively regulated the IMD pathway in S2 cells.

CG4968 regulates IMD immune pathway via the OTU domain

In order to gain more insight into the mechanism of CG4968 regulating the IMD pathway, two truncated forms of CG4968 expression plasmids (CG4968^1-26 and CG4968^OTU) were constructed based on domain analysis (Fig. 1A). The CG4968, CG4968^1-26, and CG4968^OTU each transfected Drosophila S2 cells. The Att-Luc reporter assay showed that the expression of CG4968 and CG4968^OTU significantly enhanced induced Att-Luc activity compared to the control group. After transfection with CG4968^1-26, Att-Luc activity was not significantly different from that of the control group (Fig. 2A). Also, the qRT-PCR results showed that the expression of CG4968 and CG4968^OTU in S2 cells significantly elevated the expression of Att and Cec of the IMD pathway compared to that of the control. However, the expression of CG4968^1-26 showed no difference in the expression of Att and Cec compared to that of the control group (Figs. 2B and 2C).

In addition, we performed rescue experiments to explore the role of the CG4968 OTU domain in the positive regulation of the IMD pathway. S2 cells were first treated with dsRNA 49-1 to attenuate IMD pathway activity and then transfected with plasmids CG4968, CG4968^1-26, and CG4968^OTU to determine the rescue effects of CG4968, CG4968^1-26, and CG4968^OTU on the decrease of IMD pathway signal activity. The results showed that during the expression of dsRNA 49-1, the activity of Att-Luc decreased due to the downregulation of CG4968 expression. At the same time during the expression of CG4968 and CG4968^OTU, Att-Luc activity was significantly increased compared with the control group. However, CG4968^1-26 did not significantly elevate Att-Luc activity reduced by knockdown of CG4968 expression due to its lack of OTU domain (Fig. 2D). We also detected the expression of AMPs in each group using qRT-PCR and found that the expression of CG4968^OTU effectively rescued the decreased expression of the IMD pathway Att and Cec caused by the knockdown effect of dsRNA 49-1, similarly to that of the expression of CG4968. However, the expression of CG4968^1-26 did not have a rescue effect on the decline of the IMD pathway AMPs compared to the expression of CG4968^OTU (Figs. 2E and 2F). This suggests that the OTU domain of CG4968 is indispensable for its positive regulation of the Drosophila IMD pathway.

CG4968 is non-essential for the Toll immune pathway

Given the regulatory role of CG4968 on the IMD pathway, we also used the Drosomycin promoter-driven luciferase reporter system (Drs-Luc) to explore the regulatory role of CG4968 on the Toll immune pathway. The expression of Drs-Luc activated overexpression of the active form of Toll^ΔLRR and showed dosage and temporal trends (Fig. S2). We found that knockdown of CG4968 expression in S2 cells had no significant effect on the increased expression of Toll^ΔLRR activated Drs-Luc activity (Fig. 3A). We also analysed whether knockdown of CG4968 in S2 cells affected the expression levels of the AMPs Drosomycin (Drs) and Metchnikowin (Mtk) of the Toll immune pathway. The results showed that the knockdown of CG4968 did not affect the expression of Drs and Mtk compared with the control group (Figs. 3B and 3C).

Meanwhile, we used overexpressed CG4968 in S2 cells at a concentration gradient to investigate whether overexpression of CG4968 affected the Drs-Luc activity activated by Toll^ΔLRR and the expression level of AMPs of the Toll immune pathway. The Drs-Luc reporter assay showed that the gradient overexpression of CG4968 in S2 cells did not affect Drs-Luc activity compared to the control (Fig. 3D). Similarly, overexpression of CG4968 had no effect on the expression of the Toll immune pathway Drs and Mtk (Figs. 3E and 3F). Based on the above results, our study shows that although CG4868 can positively regulate the IMD pathway, it has no effect on the transduction of the Toll immune pathway.

CG4968 regulates the K48-linked deubiquitination of Imd via the OTU domain

Imd protein is an important connector protein that plays an upstream and downstream role in the IMD pathway (Russell et al., 2021). Our study found that IMD pathway activity was significantly higher in S2 cells cotransfected with IMD and CG4968 plasmids than with IMD plasmids alone (Figs. 1F–1H). To determine whether there is an interaction between CG4968 and Imd, Co-IP experiments were performed in S2 cells with co-transfected IMD and CG4968 plasmids (empty plasmids expressing GFP were used as controls). The Co-IP results showed a clear interaction between CG4968 and Imd (Fig. 4A). In order to further explore the interaction between CG4968 and Imd, we purified GST-tagged CG4968 and His-tagged Imd from E. coli (Fig. S3). As shown in the GST pull-down assays, we found a significant interaction between purified GST-CG4968 and His-Imd (Fig. 4B). This further confirms the interaction between CG4968 and Imd.

To explore which type of deubiquitination of Imd is mediated by CG4968, we used a series of ubiquitin mutants for analysis. Our ubiquitination experiments showed that CG4968 overexpression in S2 cells significantly reduced ubiquitination modifications associated with the K48-linkage of Imd (all lysines in ubiquitin except K48 were mutated to arginine) compared to the control group’s expression of GFP. However, the expression of CG4968^1-26 had no effect on the K48-linked ubiquitination modifications of Imd compared to the control (Figs. 5A and 5D). To further confirm that CG4968 regulates the K48-linked deubiquitination modification of Imd, we treated S2 cells with dsRNA 49-1 to explore the effects of CG4968 knockdown and overexpression in S2 cells on the K48-linked deubiquitination modification of the Imd. The results showed that overexpression of CG4968 significantly reduced the K48-linked ubiquitination modification of the Imd compared to the control but knockdown of CG4968 expression in S2 cells significantly elevated the level of K48-linked ubiquitination modification of Imd (Figs. 5B and 5E). We also tested whether CG4968 affected the level of K63-linked (all lysines in ubiquitin were mutated to arginine except for K63) ubiquitination modification of Imd. In vivo ubiquitination experiments showed that CG4968 had no effect on the level of K63-linked ubiquitination modification of the Imd compared to controls (Figs. 5C and 5F).

This suggests that CG4968 significantly inhibited the K48-linked ubiquitination modification of the Imd, with the OTU domain acting as the key to CG4968 inhibiting the K48-linked ubiquitination modification of Imd.

Drosophila CG4968 executes its physiological function via OTU domain

Next, to investigate the physiological function of CG4968 in vivo, we constructed the overexpression of CG4968 flies P{Uasp-CG4968^OE} and CG4968^1-26 flies P{Uasp-CG4968^ΔOTU} and obtained the knockdown of CG4968 expression flies P{Uasp-CG4968^KD}. Because flies’s fat body serves as an important organ (Ramirez-Corona et al., 2021), we crossed the flies with P{ppl-gal4}, which is a fat body-specific driver. Characterization of transgenic flies revealed that P{ppl-gal4} effectively reduced and increased the expression of CG4968 in the fat body (Fig. S4). We infected 6-day-old control and transgenic flies with Erwinia carototovora ASP 15 (Ecc15), a commonly-used bacterium that can induce the response of the IMD pathway and the expression of AMPs (Hua et al., 2022). The qRT-PCR results showed that the expression of Att and Cec of the IMD pathway in ppl>CG4968^KD was significantly lower than that in the control ppl>+ after infection with Ecc15. The expression of Att and Cec was significantly higher in ppl>CG4968^OE after infection with Ecc15 compared to the control ppl>+. Interestingly, we also detected no significant difference in the expression of Att and Cec in ppl>CG4968^ΔOTU compared to ppl>+ due to the deletion of the OTU domain (Figs. 6A and 6B). This suggests that CG4968 is dependent on the OTU domain for positive regulation of the IMD pathway in flies.

Similarly, we used the Gram-positive bacterium Micrococcus luteus (M.L) to infect 6-day-old flies to activate Toll immune pathway responses (Ji et al., 2014). qRT-PCR results showed that the expression of the Toll immune pathways Mtk and Drs was not significantly different in transgenic flies (ppl>CG4968^KD, ppl>CG4968^OE, ppl>CG4968^ΔOTU) compared to ppl>+ after infection with M.L (Figs. 6C and 6D). This suggests that CG4968 has no effect on the Toll immune pathway in flies.

To gain more insight into the physiological function of CG4968 in vivo, we examined the role of CG4968 in resistance to an exogenous pathogenic infestation in flies. Gram-negative pathogenic bacteria Serratia marcescens (S. marcescens) were used to infect ppl>CG4968^KD, ppl>CG4968^OE, and ppl>CG4968^ΔOTU. The effect of CG4968 on the growth of S. marcescens bacteria in host flies was examined using bacterial load assay. When infected with S. marcescens pathogenic bacteria, the number of S. marcescens colonies in ppl>CG4968^KD was significantly higher than that of the control ppl>+, and the number of S. marcescens colonies in ppl>CG4968^OE was significantly lower than the control ppl>+. Due to the absence of the OTU domain, ppl>CG4968^ΔOTU showed no difference in the number of S. marcescens colonies in comparison to the control (Fig. 7A). Additionally, we measured the survival rate after infection of flies with S. marcescens. The results showed that the survival rate after infection of ppl>CG4968^OE flies with S. marcescens was significantly higher than the control ppl>+. On the contrary, the survival rate of ppl>CG4968^KD after infection with S. marcescens was significantly lower than that of the control group. The ppl>CG4968^ΔOTU survival rate did not differ compared to the control group (Fig. 7B). Taken together, we concluded that CG4968 inhibits the growth of Gram-negative pathogens by promoting the expression of AMPs of the IMD pathway, thereby increasing the survival rate of flies. It is noteworthy that the OTU domain is the key to the physiological function of CG4968 in flies.