ClC‐c regulates the proliferation of intestinal stem cells via the EGFR signalling pathway in Drosophila

Abstract Objectives Adult stem cells uphold a delicate balance between quiescent and active states, which is crucial for tissue homeostasis. Whereas many signalling pathways that regulate epithelial stem cells have been reported, many regulators remain unidentified. Materials and Methods Flies were used to generate tissue‐specific gene knockdown and gene knockout. qRT‐PCR was used to assess the relative mRNA levels. Immunofluorescence was used to determine protein localization and expression patterns. Clonal analyses were used to observe the phenotype. RNA‐seq was used to screen downstream mechanisms. Results Here, we report a member of the chloride channel family, ClC‐c, which is specifically expressed in Drosophila intestinal stem/progenitor cells and regulates intestinal stem cell (ISC) proliferation under physiological conditions and upon tissue damage. Mechanistically, we found that the ISC loss induced by the depletion of ClC‐c in intestinal stem/progenitor cells is due to inhibition of the EGFR signalling pathway. Conclusion Our findings reveal an ISC‐specific function of ClC‐c in regulating stem cell maintenance and proliferation, thereby providing new insights into the functional links among the chloride channel family, ISC proliferation and tissue homeostasis.


| INTRODUC TI ON
In most adult somatic organs, resident adult stem cells, whose high proliferation and differentiation abilities compensate for cell loss, are responsible for both tissue homeostasis and organ functions throughout the lifespan of the organism. Compensation of cell loss is especially important in tissues with a high turnover rate, such as the intestinal epithelium. 1 Intestinal stem cells (ISCs) undergo cell division and differentiation to rapidly replenish damaged cells, thereby maintaining intestine integrity during physiological turnover or in response to damage. 2,3 Thus, characterizing the regulators and signalling pathways that control stem cell activity and maintain the critical balance between cell generation and degeneration has been one of the main focuses of developmental biology and regenerative medicine. However, how cell-fate transitions occur and how signalling pathways and transcriptional networks control these coordinated cellular changes remain largely unexplored.
The Drosophila melanogaster intestine shares many similarities with its human counterpart and thus has emerged as a powerful system to study the role of intestinal stem cells in adult tissue homeostasis and regeneration because of its relatively simple and well-characterized stem cell lineage and tractable genetic manipulation. 4,5 Drosophila ISCs are small diploid cells scattered along the basement membrane of the midgut epithelium and specifically express the transcription factor Escargot (Esg) and Notch ligand Delta (Dl). 2,3 The ISCs divide asymmetrically to generate new stem cells and either transient post-mitotic progenitor cells named enteroblasts (EBs) or enteroendocrine mother cells (EMCs). EBs progressively differentiate into polyploid enterocytes (ECs), which are responsible for the absorption of nutrients in the midgut. The other type of differentiated cells in the Drosophila intestine is enteroendocrine cells (EEs). These cells emerge from EMCs, which express markers of both ISCs (Esg) and EEs (Prospero). A high level of Notch signalling drives ISCs to produce ECs, whereas a low level drives them to produce EEs ( Figure 1A). 1,4,[6][7][8] Under homeostatic conditions, the ISCs in Drosophila midgut are largely quiescent. The number of ISCs and progenitor cells in the midgut is relatively small and remain stable. However, these ISCs promptly undergo proliferation and differentiation when the tissue is injured. 9,10 This cellular response is essential for maintaining the epithelial homeostasis since a failure to replace lost cells may compromise tissue function and overproduction of cells may lead to cancer. Under both homeostatic and stressed conditions, ISC divisions are modulated by numerous regulators, such as Sox21a, GATAe and Hairless, [11][12][13] and signalling pathways, such as the EGFR, Wnt, mTOR and JAK/STAT pathways, [14][15][16][17][18][19][20] to maintain the critical balance between cell generation and degeneration. However, how these signalling pathways are regulated and integrated by which is unclear.

| Drosophila lines and husbandry
The following fly lines were obtained from the Bloomington  (C) Expression of ClC-c-3 × HA protein (red) in the esg-Gal4 ts > GFP + cells (ISCs and EBs; co-stained with GFP, green) in the R4 region of the Drosophila midgut. (D) Expression of ClC-c-3 × HA protein (red) in the ISCs (labelled by Delta staining, green) in the R4 region of the Drosophila midgut. (E and F) Expression of ClC-c-3 × HA protein (red) in the NRE-Gal4 ts > GFP + cells (EBs; co-stained with GFP, green) in the R4 region of the Drosophila midgut. Yellow arrowhead indicates the NRE-Gal4 ts > GFP + cells with small nuclei (newly formed EBs). (G) Expression of ClC-c-3 × HA protein (red) in the EEs (labelled by Prospero staining, green) in the R4 region of the Drosophila midgut.
The following fly lines were kindly provided as gifts: esg-GFP, esg-Gal4, UAS-lacZ, tub-Gal4 and FTR2A lines by Dr. Allan Spradling; and esg-Gal4 ts , MyolA-Gal4 ts , ISC-Gal4 ts and NRE-Gal4 ts lines by Dr. Benjamin Ohlstein. All the Drosophila lines used in this study are listed in Table S1.
Flies were cultured in the standard medium (50 g cornmeal, 18.75 g yeast, 80 g sucrose, 20 g glucose, 5 g agar and 30 ml propionic acid per 1 L of media) at 25°C and with 65% humidity on a 12 h light/12 h dark cycle. Unless indicated otherwise, only females were used in this study.

| Generation of the knockout, knock-in and transgenic fly lines
To knock out ClC-c, the following two sgRNAs located in the forepart of the CDS region of ClC-c were designed using CRISPR Optimal Target Finder with the maximum stringency (http://tools.flycr ispr. molbio.wisc.edu/targe tFind er/). Each sgRNA was synthesized in vitro and then subcloned into the PMD18T vector to acquire the U6 promoter. Then, the U6 promoter and sgRNA were recombined into the attB vector. This sgRNA construct was injected into the fly line y [1] w [1118]; VK00037 (BDSC# 9752) to generate ClC-c sgRNA alleles. The To generate ClC-c-3×HA knock-in lines, two constructs were generated, one with two sgRNAs and the other with a homologous recombination sequence. The sgRNA construct was generated as mentioned above. To generate the homologous recombination construct, the 5ʹ homologous arm (~1 KB upstream sequence before the termination codon), 3 × HA and the 3ʹ homologous arm (~1 KB downstream sequence after the termination codon) were inserted into the PASK vector. All the final vectors were verified by sequencing and injected by Fungene Biotechnology (Beijing, China). The two sgRNAs were used to generate DNA double-strand breaks. The 5ʹ and 3ʹ homologous arms were used for homologous recombination repair. 3 × P3-RFP was used for screening. All the sgRNA sequence used are listed in Table S3.
To get UAS-ClC-c transgenic lines, the UAS-ClC-c expression vector was generated first. Full-length cDNA (RT-PCR from total mRNA of w 1118 flies) of ClC-c was subcloned into the pEntry vector by using pEASY-Uni Seamless Cloning and Assembly Kit (TransGen Biotech, CU101-02) and then sub-cloned into pTW vector by using LR recombination reaction. The primers are listed in Table S3.

| RNA-sequencing (RNA-seq) and data analysis
The detailed process was described previously. 35 The Drosophila adult midguts (R1-R5) were first dissected in phosphate-buffered saline (PBS) on ice. They were then immediately placed in a −80°C freezer. Total RNA was isolated using the isothiocyanate-alcohol phenyl-chloroform method. Whole RNA previous sequencing was carried out on the NovaSeq 6000 platform (Illumina) by Berry Genomics Corporation. The quality control was performed using FastQC (v0.11.8).
The read length of the raw RNA-seq data was 150 bp. All the reads were aligned to the Drosophila reference genome (Ensembl BDGP6 release-89). The aligned-read sam files were then converted to bam files and sorted using SAMtools. DESeq2 (v1. 22.2) was used to determine the gene expression profiles of samples. P-value ≤ 0.05 following Benjamini and Hochberg correction for multiple hypothesis testing was considered to indicate differential gene expression.

| Cell sorting and RT-qPCR
One hundred to two hundred midguts were dissected in cold di- Total gut RNA was extracted from dissected midguts by using the Trizol reagent (Invitrogen). This RNA (1 μg) was used to generate cDNA via reverse transcription, and the cDNA was subjected to quantitative polymerase chain reaction (qPCR) in a Q uantStudio 5 System (Thermo Fisher Scientific). The 2 −ΔΔCT method was used to calculate the expression values. The relative expression was normalized to that of Rp49. All the primers used are listed in Table S3.

| Immunofluorescence microscopy
Adult midguts were dissected in PBS and then fixed with 4% Paraformaldehyde for 2 h, followed by washing three times (10 min each) with PBS containing 0.1% Tween-20 (PBST). The midguts were then blocked in 0.1% BSA for 30 min at room temperature, followed by washing with PBST, and then incubated overnight at 4°C with primary antibodies diluted in PBST. After washing three times with PBST, the tissues were incubated with secondary antibodies and DAPI for 2 h at room temperature, followed by the same washing steps above. The sources and dilutions of the primary and secondary antibodies used are listed in Table S2.
Leica TCS-SP8 confocal microscope was used to acquire all the immunofluorescence images. The Leica Application Suite X (LAS X), Adobe Photoshop cc2020 and Adobe Illustrator cc2020 were used to assemble the images.

| TUNEL assay
Adult midguts were dissected in PBS and fixed with 4% paraformaldehyde for 2 h, followed by washing with PBST. Apoptosis was assessed using the Apoptag Kit (Millipore) according to the manufacturer's instructions.

| EdU incorporation assay
For EdU labelling, a chromatography paper was cut into 4 × 6 cm strips and saturated with 5% sucrose and 100 µM EdU (Baseclick).
After being starved in empty vials for 1 h, flies were transferred into vials with the EdU solution-saturated chromatography paper for 24 h. The guts of the flies were then dissected in PBS and fixed with 4% paraformaldehyde for 2 h. Edu incorporation was evaluated following the manufacturer's instructions. The immunostaining procedure has previously been described.

| Detection of reactive oxygen species (ROS) by using dihydroethidium (DHE)
Adult midguts were dissected in PBS and then incubated in 30 μM DHE (Invitrogen) for 5 min in the dark at room temperature. The guts were then washed twice with PBS, mounted and immediately examined using a confocal microscope.

| Assessment of necrosis by using propidium iodide
Adult midguts were dissected in PBS and then stained with 1.5 μM propidium iodide (PI; Invitrogen) at room temperature for 15 min.
The guts were then fixed with 4% formaldehyde for 20 min, washed three times with PBST, rinsed twice with PBS, mounted in Vectashield with DAPI and examined using a confocal microscope.

| Fluorescence intensity statistics
Immunofluorescence images were analysed via confocal microscopy, and the fluorescence intensity statistics in the region of interest (ROI) were calculated using ImageJ. The detailed process was described previously. 35 Mean fluorescence intensity = Integrated density of the back-

| ClC-c is specifically expressed in the stem and progenitor cells of the adult Drosophila intestinal epithelium
To investigate the possible role of ClC-c in the regulation of intestinal stem cells, a ClC-c-3×HA knock-in fly line was generated ( Figure 1B).
We used the conditional temperature-sensitive driver esg-Gal4 (esg-Gal4 ts ) to express the dsRNA construct (BDSC # 27034) against ClC-c and observed that the HA signal was strongly reduced ( Figure   S1A-C). We detected ClC-c expression in R1-R5 regions of the adult Drosophila midgut ( Figure S1D). Using specific cell markers to distinguish the cell-specific expression pattern of the endogenous ClC-c protein, we found that ClC-c is specifically expressed in esgpositive cells ( Figure 1C). By using esg-Gal4, UAS-GFP; Gal80 ts and NRE-Gal80 (ISC-Gal4 ts ) flies, we found that ClC-c is expressed in ISCs ( Figure S1E). Additionally, co-staining the ClC-c-3 × HA midguts for the Dl protein and HA tag confirmed that ClC-c is expressed in ISCs ( Figure 1D). Next, we found that ClC-c is expressed in some newly formed EBs with small nuclei ( Figure 1E, yellow arrowhead), whereas no ClC-c expression was detected in mature EBs with large nuclei ( Figure 1F). These results imply that the expression of ClC-c is gradually lost during the differentiation from EBs to ECs. As predicted, no ClC-c expression was detected in mature ECs ( Figure 1H) or EEs ( Figure 1G). Taken together, these results indicate that ClC-c is specifically expressed in the stem/progenitor cells of the Drosophila intestinal epithelium.

| ClC-c is required for ISC proliferation in the Drosophila midguts
Since the expression of ClC-c is uniquely restricted to ISCs and some newly formed progenitor cells, we first depleted ClC-c in ISCs Taken together, these results strongly suggest a cell-autonomous role for ClC-c in regulating ISC proliferation.

| ClC-c is required for midgut regeneration
Intestinal stem cells confer a high regenerative capacity to the intestinal epithelium. 36  These results suggest that ClC-c is essential for inducing ISC proliferation in response to damage to the intestinal epithelium.

| Reducing ClC-c inhibits the proliferation of progenitor cells, but does not lead to apoptosis or necrosis
Knocking down ClC-c in the esg + progenitor cells and clones decreased the number of esg + cells and resulted in small clones respectively. Apoptosis is a common reason for cell loss. To assess the possibility that the esg + cells undergo apoptosis upon knocking down ClC-c, the TUNEL assay was performed to detect apoptotic signals after 7 days of ClC-c RNAi. We did not observe a significant increase in the number of apoptotic esg + cells, compared with the number in the control flies ( Figure 4A,B,H). Forced expression of rpr (an inducer of apoptosis) in esg + cells was used as a positive control ( Figure 4C,H).

These results indicate that the cell loss caused by knocking down
ClC-c in esg + cells is not due to an increase in apoptosis. Necrosis is characterized by early plasma-membrane rupture and accumulation of ROS, which can be assessed through PI staining and DHE staining respectively. 38 We detected no increase in PI + or DHE + signals after ClC-c depletion, compared with the signals in the control ( Figure 4D-G,I,J). showed that the proportion of EC cells in the ClC-c mutant clone was higher than that in the control group ( Figure S3H).
Therefore, we believed that the decrease in the number of stem cells caused by the loss of ClC-c is due to the other fact that ISC could not maintain stemness thus undergoing differentiation.

| Depletion of ClC-c downregulates the members of the EGFR signalling in the Drosophila midgut
To identify the mechanism whereby ClC-c regulates ISC proliferation, RNA-seq was performed on dissected midguts of the flies with  Figure 5A and Figure S4A,B). Among these genes, Pnt, Sox21a and Ets21C have been reported to function downstream of the EGFR signalling pathway and regulate ISC proliferation. [39][40][41] To confirm these RNA-seq results, RT-qPCR analyses were performed. The RT-qPCR results showed similar expression patterns to those observed in the RNA-seq data ( Figure 5D-G). These findings strongly suggest that ClC-c regulates ISC proliferation through the EGFR signalling pathway.

| ClC-c regulates ISC proliferation through the EGFR signalling
Previous studies have shown that the EGFR signalling pathway is vital in ISC proliferation during homeostasis or stress-induced regeneration. 10,42 The expression pattern of the EGFR effector mitogen-

Rab5-labelled early endosomes and Rab7-labelled late endosomes
Our results indicated that ClC-c regulates the proliferation of ISCs through the EGFR signalling; however, how it affects this signalling pathway is unclear. Many studies have demonstrated that endocytosis affects the EGFR signalling. Ligand stimulation causes EGFR to internalize and be transported through the endocytic pathway.
Therefore, endocytosis not only regulates the rate of EGFR degradation and circulation but also regulates the EGFR-mediated signal transduction. 43,44 By analysing our RNA-seq data, we found that knocking down ClC-c downregulates several endocytosisrelated genes, including Rab3, Rabex-5, Rab3-GEF, Rab26 and Rab40 ( Figure 7A). These findings strongly suggest that ClC-c might affect the vesicle transport system mediated by RAB family proteins. To test this hypothesis, RAB5-labelled early endosomes and RAB7labelled late endosomes were visualized in esg + cells. Under the homeostatic conditions, the RAB5-GFP signal was remarkably lower in ClC-c-depleted group than in the control ( Figure 7B-D). There was no difference in the level of the RAB7-GFP signal after knocking down ClC-c, compared with the control level ( Figure 7E-G). These results indicate that ClC-c depletion affects the activity of early endosomes instead of late endosomes under homeostatic conditions. BLM treatment for 24 h considerably increased the RAB5-GFP signal compared with the level in the untreated group ( Figure 7B,D,H,J).
Similar results were observed regarding the RAB7-GFP signal after 24 h of BLM treatment ( Figure 7K-M). These results indicate that knocking down ClC-c might inhibit endocytosis. Many studies have shown that ClC-3 is involved in the regulation of the cell cycle in many cancers. 45 A recent study has shown that, in DU145 prostate cancer cells, ClC-3 also acts as a signalling molecule that directly interacts with the stem cell factor SOX2, and then, the two co-regulate the cell cycle. 46  of ERK signalling that is required to promote proliferation in vivo. 55 In addition, endocytic also affects signalling pathways that interact with EGFR signalling pathway, such as the JAK/STAT signalling pathway. Injury-induced JAK/STAT signalling promotes cell proliferation by activating EGFR signalling. Previous studies have shown that blocking trafficking in distinct endosomal compartments leads to an inhibition of the JAK/STAT pathway, 56 and the internalization and endocytic trafficking of activated Dome allows for compartmentalized signalling to regulate subsets of Drosophila JAK/STAT transcriptional targets. 57 These results strongly supports that the internalization and trafficking are both required for JAK/STAT activity. Therefore, endocytosis may also affect the EGFR signalling pathway through indirect means. In summary, different stages of endocytosis inhibition can inhibit EGFR signalling activity in different ways. Endocytosis is involved in the transduction of the EGFR signal, and our experimental results showed that ClC-c deficiency inhibited endocytosis. These results suggest that ClC-c deficiency may suppress the EGFR signalling by inhibiting endocytosis. How ClC-c deficiency inhibits endocytosis is still unclear, and thus, further research is needed.

ClC-3 is an intracellular ClC protein that functions as a Cl
In conclusion, this study revealed that ClC-c regulates the proliferation of Drosophila midgut ISCs by inducing the EGFR signalling pathway ( Figure 7N), thereby providing an important basis for further exploration of the regulatory role of ClC-c in adult stem cells.

ACK N OWLED G EM ENTS
We thank BDSC, VDRC, Dr. Allan Spradling and Dr. Benjamin