Identification of a new allele of O-fucosyltransferase 1 involved in Drosophila intestinal stem cell regulation

Stem cells are responsible for maintaining the homeostasis of adult tissues, where the frequently lost cells are constantly replenished by stem cell progeny. The proliferation and differentiation of adult stem cells must be tightly controlled. Disruption of this balance will lead to stem cell accumulation or stem cell depletion, eventually resulting in various diseases, such as cancer and aging (Jasper, 2020; Lin, 2008; Morrison and Spradling, 2008; Radtke and Clevers, 2005; Singh et al., 2019). Therefore, understanding of the mechanisms controlling stem cell proliferation and differentiation will provide insight into the development of therapeutics to treat human diseases.

The posterior midgut of the adult Drosophila intestine has proven to be an excellent system to study the regulation of stem cell proliferation and differentiation. Drosophila intestines show marked similarities with their mammalian counterparts in terms of cellular make-up, development and genetic control (Banerjee et al., 2019; Casali and Batlle, 2009; Gervais and Bardin, 2017; Jasper, 2020; Stainier, 2005; Wang and Hou, 2010). ISCs are distributed along the basement membrane of the Drosophila adult midgut epithelium (Micchelli and Perrimon, 2006; Ohlstein and Spradling, 2006). Initial studies proposed that ISCs constantly undergo asymmetric divisions and produce non-dividing daughter cells, enteroblasts (EBs) (Micchelli and Perrimon, 2006; Ohlstein and Spradling, 2006, 2007). Delta (Dl), one of the Notch (N) ligands, is specifically expressed in ISCs, while Notch receptor is expressed in both ISCs and EBs. ISCs signal via Dl to activate Notch signaling in EBs, which terminally differentiate into either absorptive enterocytes (ECs) or secretory enteroendocrine cells (EE) depending on their signaling environments (Beebe et al., 2010; Micchelli and Perrimon, 2006; Ohlstein and Spradling, 2006; Perdigoto et al., 2011; Yeung et al., 2011). Recent studies show that a significant proportion of ISCs divides symmetrically in response to differentiation and subsequent loss of a neighboring ISC (or vice versa) (de Navascués et al., 2012; Goulas et al., 2012; O'Brien et al., 2011). Moreover, EE cells may not be generated from EBs, but directly from ISCs or EE progenitor cells (EEPs) (Biteau and Jasper, 2014; Chen et al., 2018; Zeng et al., 2015). Interestingly, unlike other systems in which differentiated cells could be de-differentiated into stem cells, no new ISCs could be re-generated after all progenitors were ablated, indicating that fully differentiated intestinal cells could not de-differentiate into ISCs in the absence of any progenitors (Brawley and Matunis, 2004; Lu and Li, 2015a; Raff, 2003).

Previous studies have shown that the proliferation and differentiation of ISCs under physiological and stressed conditions are regulated by many signaling pathways including the Notch, Wingless (Wg), Janus Kinase/Signal Transducer and Activator of Transcription (JAK/STAT), Epidermal Growth Factor Receptor (EGFR), Hippo (Hpo), Insulin, Hedgehog (Hh), and Bone Morphogenetic Protein (BMP) pathways (see reviews by Colombani and Andersen, 2020; Gervais and Bardin, 2017; Guo et al., 2016; Jiang et al., 2016; Joly and Rousset, 2020; and references therein). The evolutionarily conserved Notch signaling pathway plays essential roles in the control of cell proliferation and specification/differentiation during animal development (Artavanis-Tsakonas and Muskavitch, 2010). Before being presented on the plasma membrane, the Notch receptor undergoes post-translational modifications such as fucosylation in the ER and Golgi apparatus where the S1 proteolytic cleavage of the Notch receptor occurs. Upon binding of its ligand (Dl/Ser) from the signal sending cell on the plasma membrane, the Notch receptor in the signal receiving cell undergoes two consecutive proteolytic cleavages (S2 and S3 cleavages), producing Notch extracellular truncate (NEXT, the S2 cleavage product) and finally releasing the intracellular domain of Notch (NICD, the S3 cleavage product) from the plasma membrane to regulate the expression of downstream genes in the nucleus (Artavanis-Tsakonas and Muskavitch, 2010; Artavanis-Tsakonas et al., 1999; Bray, 2006; Okajima and Irvine, 2002; Pandey et al., 2021). Notch signaling is essential for the proliferation and differentiation of ISCs in Drosophila, loss of Notch signaling in ISCs leads to increased ISC proliferation and progeny differentiation defects, generating ISC and/or EE tumors, while constitutive activation of Notch signaling forces direct differentiation of ISCs into ECs, resulting in ISC loss (Micchelli and Perrimon, 2006; Ohlstein and Spradling, 2006, 2007).

To identify factors regulating the proliferation and differentiation of ISCs, we carried out a small-scale forward genetic screen using some p-element insertions by MARCM technique in the posterior midgut (data not shown) (Lee and Luo, 2001). We were surprised to find that there are two types of ISC MARCAM clones generated upon induction of stum^MB01421 mutant (Fig. 1A,B). A p-element is inserted in the coding region of stumble (stum, stum^MB01421), which is required for locomotion and mechanical sensing in proprioceptive neurons by transducing dendrite stretching into cellular responses (Desai et al., 2014). The type I stum^MB01421 ISC MARCM clones are similar as those of wild-type (WT) control (Fig. 1B,D,E). While the type II stum^MB01421 ISC MARCM clones grow much faster than those of WT clones, and the cell size in these clones is tiny and quite uniform compared to the varied cell size of control and type I clones, suggesting that ISC proliferation and/or differentiation is affected in these clones (Fig. 1C–E). We purified the mutant for the type II clones by consecutive backcrosses and named this mutant as stum 9-3 (data not shown).

We examined the cell identity in stum 9-3 mutant clones. One to two Dl+ cells could be observed in control clones, while the number of Dl+ cells is dramatically increased in stum 9-3 mutant clones, suggesting that stum 9-3 mutant affects ISC proliferation and/or differentiation (Fig. 2A–D). Meanwhile, EE cell (Pros+) could be observed in control clones (Fig. 2A). On the contrary, two types of stum 9-3 mutant clones are observed: the number of EE cells is significantly increased and increased number of EE cells could be observed adjacent to some clones; while the majority of the other type clones are EE cells (Fig. 2B,C,E). These data suggest that stum 9-3 mutant likely affects ISC proliferation and differentiation. The phenotype of stum 9-3 mutant is reminiscent of Notch loss of function (Fig. S1) (Micchelli and Perrimon, 2006; Ohlstein and Spradling, 2006, 2007). We then examined whether Notch signaling is affected in stum 9-3 mutant ISC MARCM clones. Notch signaling is activated [by Gbe+Su(H)-lacZ] in control clones (Fig. 2F) (Furriols and Bray, 2001). However, no Gbe+Su(H)-lacZ+ cells are observed in stum 9-3 ISC clones, indicating that Notch signaling is abolished in stum 9-3 mutant (Fig. 2G–I). In supporting of this, the expression of Notch downstream target Cut during wing development is also abolished in stum 9-3 mutant clones, indicating that stum 9-3 mutant is generally required for the activation of Notch signaling (Fig. S2). Furthermore, we find that both NECD and NICD are highly accumulated in stum 9-3 ISC MARCM clones, suggesting that the full-length Notch receptor is accumulated in stum 9-3 mutant clones (Fig. S3). Altogether, these data indicate that stum 9-3 mutant affects the activation of Notch signaling.

We then examined which step of Notch receptor processing that stum 9-3 mutant is required for. We performed rescue experiments by expressing full-length Notch (S1 cleavage product), NEXT (the S2 cleavage product) and NICD (the S3 cleavage product) in stum 9-3 mutant clones, respectively. We found that expression of full-length Notch could not rescue defects observed in stum 9-3 mutant clones (Fig. 3A,B; Fig. S4). While expression of either NEXT or NICD dramatically reduced the size of stum 9-3 mutant clones, likely causing precocious ISC differentiation due to ectopic activation of Notch signaling (Fig. 2C–E; Fig. S4) (Micchelli and Perrimon, 2006; Ohlstein and Spradling, 2006). These data suggest that stum 9-3 mutant affects Notch signaling upstream of the S2 cleavage of the Notch receptor, with accumulation of full-length Notch receptors in stum 9-3 mutant clones. The retromer complex is reported to directly regulate Notch receptor retrograde trafficking in Drosophila neuroblast lineages to prevent aberrant Notch signaling activation (Li et al., 2018). However, depleting the retromer complex could not rescue defects observed in stum 9-3 mutant clones, indicating that stum 9-3 mutant may act upstream of endocytic and retrograde trafficking of the Notch receptor (Fig. S5). Taken together, these data show that stum 9-3 mutant is involved in Notch signaling upstream of the S2 cleavage of the Notch receptor.

As stum 9-3 mutant is derived from stum^MB01421 mutant, which disrupts stum function, we then examined whether the defects observed in stum 9-3 mutant are resulted from stum loss of function. However, we found that stum is not expressed in the intestines (by stumGal4) (Fig. S6A–B′) (Desai et al., 2014). Furthermore, no obvious defects were detected in the adult intestines of homozygous stum null mutants (Fig. S6C,D) (Desai et al., 2014). Moreover, GFP-Stum is localized in the cell periphery when expressed in the progenitors and expression of GFP-stum could not rescue defects observed in stum 9-3 mutant (Fig. S6E–J) (Desai et al., 2014). Altogether, these data demonstrate that the defects observed in stum 9-3 mutant are not caused by stum loss of function.

We went on to identify the gene responsible for the observed defects observed in stum 9-3 mutant. Initial examinations showed that stum 9-3 mutant is not allele of known genes involved in the Notch signaling pathway located on the right arm of the 2nd chromosome (2R), like presenilin enhancer (pen-2, encoding the key subunit of the γ-secretase responsible for the S3 cleavage of the Notch receptor) and Caf1-p105 [encoding the subunit of the chromatin assembly factor 1 (CAF-1)] (Fig. S7 and data not shown). We then carried out detailed mapping of stum 9-3 mutant using different deficiency kits for 2R and other mutants (Table S1). The lesion was finally mapped to the region of O-fut1, which is responsible for catalyzing the reaction that attaches fucose through an O-glycosidic linkage to a conserved serine or threonine residue found in the consensus sequence of EGF domains of the Notch receptor (Table S1) (Okajima and Irvine, 2002). To further confirm that the defects observed in 9-3 mutant is caused by mutations of O-fut1, we performed rescue experiments. Restoring O-fut1 function by either over-expressing O-fut1 or endogenous O-fut1 under its own promoter (O-fut1^+t3.8) completely rescued defects observed in 9-3 mutant (Fig. 4A–C and data not shown) (Okajima and Irvine, 2002). These data indicate that mutation of O-fut1 in stum 9-3 mutant is responsible for the defects observed. We further examined the nature of lesion in O-fut1 region. We found that the lesion occurs in the 2nd coding exon of O-fut1, probably caused by insertion of an unknown p-element or DNA fragment (Fig. S8). We thus named this O-fut1 allele as O-fut1st^um9-3.

Depletion of O-fut1 in progenitors resulted in identical defects as depletion of the Notch receptor (Fig. S9A–G). We further examined the consequences of depleting O-fut1 in different types of intestinal cells using cell-type specific drivers [ISC: DlGal4, EB: Gbe+Su(H)Gal4 and EE: ProsGal4]. Depletion of O-fut1 in ISCs results the same defects as that of Notch depletion, generating clusters of ISCs and EE cells, indicating that Notch signaling is required in ISCs for their proliferation and differentiation (Fig. S9H–K). Unexpectedly, no defects were observed upon depletion of either O-fut1 or Notch in EBs, in contrast to the observation that Notch signaling is activated in EBs for progeny differentiation (Fig. S9L–O). The discrepancy is likely caused by the nature of the EB driver used. The Gal4 used is directed by Gbe+Su(H), the responsive element of Notch signaling activation, in EBs (Furriols and Bray, 2001; Lu and Li, 2015b). Thus genes begin to be depleted only after the expression of this EB driver, i.e. after the activation of Notch signaling in EBs. Moreover, EB cell is in a transient differentiating state. Therefore, these facts impede the detection of the requirement of genes in EBs. Interestingly, depletion of both O-fut1 and Notch in EE cells produced no obvious defects (Fig. S9P–S). O-fut1 attaches fucose through an O-glycosidic linkage to the EGF domains of the Notch receptor using GDP-fucose as donor substrate in the ER (Okajima and Irvine, 2002; Okajima et al., 2005). Consistently, both NECD and NICD are accumulated in the ER upon depletion of O-fut1 in progenitors (Fig. 4D–H). Taken together, our data show that the defects observed in O-fut1st^um9-3 mutant are caused by defective O-fucosylation of the full-length Notch receptor.

Two classic methods are generally deployed to identify factors involved in specific developmental process or signaling pathways: forward genetic screen and reverse genetic screen. Both methods have advantages and shortcomings. Factors involved in specific processes can be identified unbiasedly through forward genetic screen; however, the pinpoint of the lesion responsible for the defects observed is often tedious and time consuming, especially when the genetic background of the mutant is not clean, or the required deficiency (Df) lines are not available, or some of the Df lines are not correctly annotated, and it will be quite disappointing to find out that the identified mutants with interesting phenotype are new alleles of known genes after tedious mapping. The mapping of stum 9-3 mutant is a good example for these drawbacks as we found that the genetic background of the p-element line obtained is not clean and carries multiple lethal hits, and some Df lines are not correctly annotated (Table S1). The genes responsible for specific processes can be easily identified from reverse genetic screens; however, the off-target effects of RNAi (RNA interference), especially for library generated by dsRNA (double strand RNA), must be carefully excluded. Thus, a combination of different experiments should be carried out to confirm that the identified gene is responsible for the defects observed.

Flies were maintained on standard media at 25°C. Crosses were raised at 18°C in humidity controlled incubators, or as otherwise noted. Flies hatched in 18°C incubators (2–3 days old) were picked and transferred to 29°C incubator, unless otherwise specified. Flies were transferred to new vials with fresh food every day, and dissected at time points specified in the text. In all experiments, only the female posterior midgut was analyzed. Information for alleles and transgenes used in this study can be found either in FlyBase or as noted: stum^MB01421 (BL23736), stum²⁰⁴ BL58774), stum⁴⁴⁸⁷ (BL58775), stumGal4 (BL58776 and BL58776), Cyo, 2XTb (gift from A. Zhu) (BL36336), UAS-GFP-stum (BL58778) (Desai et al., 2014), pen-2^MI02639 (BL36019), UAS-Notch^FL (BL26820), UAS-NICD (Cooper and Bray, 2000), Ote^+t5.9 (OteP, gift from Chen DH) (Jiang et al., 2008), Df(2R) deficiency kit collections (gift from J. Pastor and from Bloomington), O-fut1^SH2260 (BL51666), O-fut1^+t3.8 (BL44244), UAS-O-fut1 (BL9376) (Okajima and Irvine, 2002), esgGal4, UAS-GFP, tubGal80^ts (esg^ts, gift from N. Perrimon), DlGal4, UAS-GFP, tubGal80^ts (Dl^ts, gift from S. Hou and R. Xi) (Zeng et al., 2010), GBE+Su(H)Gal4, UAS-GFP, tubGal80^ts (GBE+Su(H)^ts) (Lu and Li, 2015b), vps26^RNAi (THU3819/HMS01769), vps35^RNAi (THU3886/HMS01858), Notch^RNAi (BL33611), O-fut1^RNAi (THU2167/JF02052), FRT19A-Notch^264-39, Gbe+Su(H)-lacZ (gift from S. Bray) (Furriols and Bray, 2001), Caf1-p105³⁶ (gift from R. Jiao) (Yu et al., 2013), hsFlp, ActGal4, UAS-GFP; FRT42D-tubGal80 (for MARCM clonal analysis), UAS- GFP.KDEL (BL9898).

For standard immunostaining, intestines were dissected in 1 X PBS (10 mmol/l NaH₂PO₄/Na₂HPO₄, 175 mmol/l NaCl, pH7.4), and fixed in 4% paraformaldehyde for 25 min at room temperature. Samples were rinsed, washed with 1 X PBT (0.1% Triton X-100 in 1 X PBS) and blocked in 5% horse serum in 1 X PBT for 45 min. Embryos were fixed and stained following standard protocol. Primary antibodies were added to the samples and incubated at 4°C overnight. The following primary antibodies were used: mouse mAb anti-Dl [C594.9B, 1:50, developed by S. Artavanis-Tsakonas, Developmental Studies Hybridoma Bank (DSHB)], mouse mAb anti-Prospero (MR1A, 1:100, developed by C.Q. Doe, DSHB), mouse mAb anti-Arm (N2 7A1, 1:100, developed by E. Wieschaus, DSHB), mouse mAb anti-Cut (2B10, 1:50, developed by G.M. Rubin, DSHB), mouse mAb anti-NECD (C458.2H, 1:30, developed by S. Artavanis-Tsakonas, DSHB), mouse mAb anti-NICD (C17.9C6, 1:30, developed by S. Artavanis-Tsakonas, DSHB), rabbit anti-β-glactosidase (Cappel, 1:5000), mouse anti-β-glactosidase (Cell Signaling Technology, 1:1000). The primary antibodies were detected by fluorescent-conjugated secondary antibodies from Jackson ImmunoResearch Laboratories. Secondary antibodies were incubated for 2 h at room temperature. DAPI (Sigma-Aldrich, 0.1 μg/ml) was added after secondary antibody staining. The samples were mounted in mounting medium (70% glycerol containing 2.5% DABCO). All images were captured by a Zeiss LSM780 inverted confocal microscope, and were processed in Adobe Photoshop and Illustrator.

The clonal analyses were achieved using the MARCM system (Lee and Luo, 2001). The ISC clones were induced by heat shocking 3–5 day-old adult flies at 37°C for 60 min. The flies were maintained at 25°C incubator and transferred to new vials with fresh food every day. The sizes of the marked clones were assayed at 6–8 days after clone induction (6–8D ACI, clones from at least ten midguts for each genotype were assayed).

The genomic coding region of pen-2 (with stop codon omitted) was cloned into the XhoI and XbaI sites of a pUAST-cV5 vector to be fused with a V5 tag at its C-terminus. The N-terminus of Notch protein (1-160 aa, containing the signal peptide) was fused to the NEXT part of Notch protein (the product after S2 cleavage, 1714-2703 aa, deleting the extracellular domain) and was cloned into a pUAST vector. Transgenic flies were obtained by standard P-element-mediated germline transformation.

The size of ISC clones was determined using Image-Pro Plus software from each confocal image. The number of specific cells per clone was determined manually from each confocal image. The cell size in MARCM clones was determined using Image-Pro Plus software. Confocal images of 40×lens/1.0 zoom from a defined posterior midgut region between the hindgut and the copper cells of different genotypes indicated were acquired. The number of intestines scored is indicated in the text. Statistical analysis was done using the Student's t-test. PEMS 3.1 software was used for s.d. analyses and Sigma Plot software for graph generation. The graphs were further modified using Adobe Photoshop and Illustrator. ****P<0.0001.

We are grateful to Jose Pastor, Norbert Perrimon, Sarah Bray, Steven Hou, Rongwen Xi, Dahua Chen, Alan Zhu, Yu Cai and Renjie Jiao for generous gifts of reagents, the Bloomington Stock Center, VDRC, NIG-FLY Center, TRiP at Harvard Medical School (NIH/NIGMS R01-GM084947) and Tsinghua Fly Center (THFC) for fly stocks, and DSHB for antibodies. We apologize to the colleagues whose work could not be cited because of space restrictions.