Drosophila Caliban preserves intestinal homeostasis and lifespan through regulating mitochondrial dynamics and redox state in enterocytes

Summary

Precise regulation of stem cell activity is crucial for tissue homeostasis. In Drosophila, intestinal stem cells (ISCs) maintain the midgut epithelium and respond to oxidative challenges. However, the connection between intestinal homeostasis and redox signaling remains obscure. Here we find that Caliban (Clbn) functions as a regulator of mitochondrial dynamics in enterocytes (ECs) and is required for intestinal homeostasis. The clbn knock-out flies have a shortened lifespan and lose the intestinal homeostasis. Clbn is highly expressed and localizes to the outer membrane of mitochondria in ECs. Mechanically, Clbn mediates mitochondrial dynamics in ECs and removal of clbn leads to mitochondrial fragmentation, accumulation of reactive oxygen species, ECs damage, activation of JNK and JAK-STAT signaling pathways. Moreover, multiple mitochondria-related genes are differentially expressed between wild-type and clbn mutated flies by a whole-genome transcriptional profiling. Furthermore, loss of clbn promotes tumor growth in gut generated by activated Ras in intestinal progenitor cells. Our findings reveal an EC-specific function of Clbn in regulating mitochondrial dynamics, and provide new insight into the functional link among mitochondrial redox modulation, tissue homeostasis and longevity.

Keywords:

Drosophila melanogaster – Cell differentiation – Stem cells – Cloning – Homeostasis – Gastrointestinal tract – DAPI staining

Introduction

Tissue homeostasis needs to be properly maintained to preserve the organismal fitness, especially in tissues with high turnover rate, such as the intestinal epithelium. The digestive tract is constantly challenged by chemicals, toxins, and pathogens in ingested food, and needs to be regenerated continuously. The intestinal stem cells (ISCs) exhibit the ability to self-renew to maintain the stem cell population, generate daughter cells of all mature cell types, and replace damaged and aged cells during normal epithelial regeneration as well as in response to cellular stress and injury. These functions are essential for intestinal homeostasis. Deregulation of ISCs proliferation or differentiation can lead to developmental abnormality and chronic diseases [1].

The Drosophila intestinal system shares anatomical and functional similarities with the mammalian digestive system, and is a facile model for dissecting the signaling mechanisms that control stem cell self-renewal and differentiation [2]. The Drosophila ISCs originate from adult midgut precursors (AMPs), the endodermal progenitors, during the larval stage [3], and reside adjacent to the basement membrane [4, 5]. ISCs self-renewal and differentiation are controlled mainly by differentially regulated Notch pathway activity. The daughter cells with high Notch activity become transient, differentiating precursor cells called enteroblasts (EBs), which terminally differentiate into enterocytes (ECs). The daughter cells with low Notch activity retain ISCs identity and are primed to differentiate into pre-enteroendocrine (pre-EE) cells and further into enteroendocrine (EE) cells when the neuronal transcription factor, Prospero is highly expressed [4–10]. Multiple evolutionarily conserved signaling pathways are required to maintain normal ISCs activities, such as Notch, EGFR, Wnt, JAK-STAT, JNK, BMP, Hippo, Slit/Robo etc [7, 11–17].

Mitochondria are highly dynamic organelles that continually fuse and divide. The morphology of mitochondria is well balanced between fusion and fission reactions and linked with ROS production [18, 19]. Defective mitochondrial fusion and fission are often associated with aging, metabolic disease, and cancer [20]. The regenerative potential of stem cells is closely correlated with the level of intracellular reactive oxygen species (ROS). Normal physiological levels of ROS are essential for stem cells functions and fate decision [21,22]. Mitochondria are the principal source of cellular ROS (mtROS), and a regulated physiological level of mtROS is important to maintain stem cells self-renewal and differentiation. Recent studies suggest that genes regulating mitochondrial fusion-fission dynamics play important roles for the maintenance of stem cells functions, fate decisions and aging processes [23–26].

Similar to mammals, Drosophila intestinal stem cells also have increased proliferation in response to oxidative stress. Commensal bacteria in the Drosophila gut can induce NADPH oxidase 1 (NOX1)-dependent endogenous ROS generation and proliferation of intestine stem cells [27], and enteropathogenic bacteria ECC-15 infection induces oxidative burst, activates JNK and JAK-STAT signaling pathways and increases ISCs proliferation [28]. Increased level of unfolded protein response of the endoplasmic reticulum (UPR^ER) in tissues can activate the JNK pathway by producing ROS and cause hyper-proliferation of ISCs [29]. The master regulator of cellular redox state, Nrf2, maintains intestinal homeostasis through controlling ISCs proliferation [30]. Mitochondrial dynamics also regulate differentiation of ISCs in Drosophila; knock-down of opa1 or mitofusin (mfn) to inhibit mitochondrial fusion causes differentiation failure [31]. Although the roles of mitochondria in stem cell self-renewal and differentiation are becoming more evident, many questions are still open. Further work needs to be done to identify genes related to mitochondria dynamics and metabolism, and to decipher their functions in the maintenance of stem cell homeostasis.

We have previously shown that Drosophila Caliban (Clbn) works as a nuclear exporting mediator for Prospero in S2 cells and a tumor suppressor in human NSCLC cells [32]. We generated clbn gene knock-out fly (a deletion covering the entire clbn protein-coding region) using homologous recombination, and demonstrated that Clbn regulates DNA damage induced p53-dependent and–independent cell apoptosis, and S phase cell cycle checkpoint by antagonizing E2F1 activity [33–34]. A recent study demonstrated that Clbn can bind to fly Ala- and Thr-tRNAs, and loss of clbn rescued the mitochondrial morphology and DA neuron loss phenotypes in muscle of PINK1 mutant, although clbn mutant showed normal phenotype [35]. In the present study, we carried out series of genetic analyses, and provided the direct evidence that Drosophila Clbn is required for the maintenance of adult intestinal homeostasis and lifespan by specifically regulating mitochondrial dynamics and redox balance in enterocytes. Flies loss of clbn have shortened lifespan. Clbn resides in mitochondrial outer membrane in enterocytes. Depletion of clbn in enterocytes results in mitochondrial defects and accumulation of ROS production and cellular damage, which in turn leads to activation of oxidative stress and promotion of ISCs hyper-proliferation. Thus, our results demonstrate the novel function of Clbn in regulating mitochondrial dynamics and intestinal homeostasis and further highlight the importance of mitochondrial dynamics control in regulating somatic stem cells proliferation and tissue homeostasis.

Results

Flies loss of clbn have shortened lifespan and dysfunctional intestinal barrier

Flies loss of clbn gene are viable, show no external morphological defects and small developmental delays under physiological conditions [33]. However, clbn knock-out flies exhibited a reduced lifespan compared with control flies, and most of clbn mutated flies died at eight weeks after birth (Fig 1A). Consistently, the clbn knockdown flies had reduced lifespan, and restoration of clbn expression can fully rescue the lifespan of clbn knock-out flies (Fig 1A). As defects in the maintenance of intestinal homeostasis could result in shortened lifespan in flies [36], we assessed whether intestinal morphology and integrity were damaged in clbn mutant. We did not observe a morphology change in guts from clbn mutant adults at 15-day-old. We next examined integrity of the intestinal barrier by feeding flies with non-absorbable blue food dye. In wild-type flies, blue food dye was restricted to the pro-boscis and in digestive tract, whereas blue dye spread to other body parts in clbn mutant, indicating the intestinal barrier was dysfunctional in clbn mutant (Fig 1B–1E).

Loss of clbn causes over-proliferation of intestinal stem cells

To explore the cause of defective intestinal barrier in clbn mutant, we defined the number of ISCs, EBs, EEs and ECs in the fly midgut. To distinguish the different cell types in intestinal epithelium in female adult flies, we used esgGal4>GFP to label progenitor cells (ISCs and EBs), Myo1A>GFP to label ECs, and anti-Prospero antibody staining to label EEs. In young flies (5-day-old), loss of clbn did not significantly affect the number of ISCs, EBs and EEs (S1 Fig). However, clbn mutant at day 15 or 30 after birth had significantly increased number of ISCs and EBs when compared with that of control flies (S1 Fig), while the number of progenitor cells was comparable between clbn heterozygous flies and control flies, suggesting increased proliferation of progenitor cells and rapid epithelial turnover in clbn mutant (Fig 1F–1F" and 1G). We further used the ISC marker delta-LacZ and the EB marker Su(H)GBE>GFP to distinguish ISCs from EBs, and observed a marked increase of both ISCs and EBs in the clbn mutant (S2 Fig). Furthermore, the number of EE cells (Pros+) was significantly increased in the clbn mutant (Fig 1F", 1H and S1 Fig), however the number of EC cells was reduced (Fig 1I–1I" and 1J). Immunostaining with anti-phospho-histone H3 (PH3) antibody revealed a significant increase of PH3⁺ cells in clbn mutant, and most of PH3⁺ cells were esgGal4>GFP⁺ (Fig 1K′, 1L and S1 Fig), indicating the over-proliferation of intestine stem cells in clbn mutant.

To address the possibility that the changed number of ISCs, EBs, EEs and ECs in clbn mutant resulted from a block to differentiation, we performed the mosaic analysis with a repressible cell marker (MARCM), and detected the presence of EEs with anti-Prospero antibody and ECs with anti-Pdm1 antibody in clbn mutant clones, respectively. The EEs and ECs were presented within clbn mutant clones without significant changes in cell numbers (S3A–S3F Fig), indicating that clbn is not required for epithelial cell differentiation in the midgut. Moreover, the cell numbers in MARCM clones were comparable between wild-type and clbn mutant, suggesting that Clbn might function non-autonomously in regulating ISC proliferation (S3H Fig). Together, these findings indicated that clbn is essential for the control of ISCs proliferation and intestinal homeostasis, but dispensable for cell differentiation.

Clbn resides in enterocytes to modulate intestinal homeostasis

Next we detected the cellular localization of Clbn in the fly intestinal epithelium using anti-Clbn antibody staining. Clbn protein was highly expressed in Myo1A>GFP ⁺ ECs, and most of Clbn protein was cytoplasmic (Fig 2A and 2A'). However, the esg>GFP⁺ progenitor cells or Pros⁺ EEs had little or no expression of Clbn protein (Fig 2B–2C'). The expression of clbn was confirmed by qPCR on sorted ISC/EB, EE and EC cells using fluorescence-activated cell sorting (FACS) (Fig 2D). To further explore the site that Clbn acts in the midgut, we used a panel of Gal4 drivers to drive knock-down of clbn in the different cell types in midgut. The number of progenitor cells and EE cells were significantly increased when clbn was knocked-down in the ECs under Myo1A-Gal4 (S4A, S4B, S4F and S4G Fig), while knock-down of clbn in the ISC and EB cells under esg-Gal4, or knock-down of clbn in the EE cells under pros-Gal4, had no significant effect on the number of ISC, EB and EE cells (S4C, S4D, S4F and S4G Fig). Moreover, knock-down of clbn in the smooth muscle surrounding the gut with mef2-Gal4 had no effect on the number of ISC, EB and EE cells (S4E–S4G Fig), suggesting that Clbn performs its functions specifically in ECs to regulate intestinal homeostasis.

To further confirm that Clbn works in the ECs to regulate intestinal homeostasis, we performed rescue experiments. Forced expression of clbn in ECs using Myo1A-Gal4 completely rescued the abnormal increased number of progenitor cells and EE cells in the clbn mutant, while expression of clbn in progenitor cells using esg-Gal4 partially rescued dysplasia in the clbn mutant (Fig 2E–2J), indicating that Clbn functions mainly in ECs to regulate intestinal homeostasis. Furthermore, the shortened lifespan of clbn mutant can be largely rescued by expressing clbn in ECs (Fig 2K), while no rescue effect was observed when expressing clbn in ISCs and EBs (Fig 2L).

Clbn regulates mitochondrial dynamics in ECs

The cytoplasmic localization of Clbn in ECs shows a “perinuclear” pattern, and Clbn protein is highly enriched around nucleus (Fig 2A and 2A'), which suggested that Clbn might be localized to mitochondria. To test this hypothesis, we used a mito-GFP reporter driven by Myo1A-Gal4 to label the mitochondria in the ECs. The Clbn protein predominantly colocalized with the mito-GFP in the ECs (Fig 3A–3C‴). Furthermore, by performing cell fractionation experiments using intestinal extract, we observed highly enriched Clbn protein in the fraction containing mitochondria marker ATP5A (Fig 3D). To determine the sub-mitochondrial localization of Clbn, the mitochondria from midgut was purified and then digested with protease K. Clbn was completely removed by protease K treatment (Fig 3E), as that of Tom20, a marker of the outer membrane protein. In comparison, Cytochrome C and SOD2, which reside at the inter-membrane space and the matrix, respectively, were resistant to protease K treatment (Fig 3E). Thus, Clbn localizes to the mitochondrial outer membrane. Taken together, we concluded that Clbn resides at the outer membrane of mitochondria in ECs to perform its functions.

We next investigated the roles of Clbn in regulating mitochondrial morphology in ECs. In control flies, mitochondria form the net-like structure around the nucleus (Fig 4A), while in clbn mutant, mitochondria were highly fragmented (Fig 4B). In young flies (5-day-old), loss of clbn did not significantly affect the mitochondrial morphology (S5A Fig). However, clbn mutants at day 15 after birth exhibited fragmented mitochondria morphology (S5B Fig). Moreover, we analyzed the ultrastructure of mitochondria in ECs by TEM and found that loss of clbn led to fragmented mitochondria (Fig 4G and 4H), indicating that loss of clbn in flies change the mitochondrial dynamics in ECs. To determine whether Clbn generally regulate mitochondrial morphology in other tissues, we examined the mitochondrial morphology in flight muscle of the clbn mutant. Interestingly, no obvious changes of mitochondrial morphology were detected in flight muscle of clbn mutant when compared with the control (S5C and S5D Fig), indicating the function of Clbn in regulating the mitochondrial dynamics is EC-specific. To further confirm the role that Clbn plays for the maintenance of mitochondria is EC-specific, we performed rescue experiments by over-expressing clbn in ECs. Restoration of clbn expression in ECs successfully rescued the abnormal mitochondria morphology observed in clbn mutant (Fig 3C–3C"). Taken together, clbn is essential for the maintenance of mitochondria structure specifically in ECs.

Mitochondrial morphology is regulated by continuous fusion and fission. These dynamic processes are controlled by a group of GTPases conserved from yeast to human. Mitofusins play important roles in mediating fusion of outer mitochondrial membrane. The Drosophila mitofusin, also named Marf (mitochondrial assembly regulatory factor), plays a more general role to regulate mitochondrial fusion. While dynamin-related protein 1 (Drp1) is a major player responsible for mitochondrial fission [37–38]. To define whether Clbn and Drp1/Mfn work in the same pathway, we evaluated mitochondrial morphology in ECs upon knock-down of drp1 or over-expression of marf in combination with clbn knock-out. Both drp1 knock-down and marf over-expression in ECs can largely rescue mitochondrial fragmentation in clbn mutant (Fig 4C–4F, 4I and 4J). Knock-down of drp1 can also partially rescue the ISC overproliferation in clbn mutant (S6 Fig). Furthermore, knock-down of drp1 or over-expression of marf partially rescued the shortened lifespan of the clbn mutant (Fig 4K), indicating Clbn functions upstream of Drp1 and Mfn in regulating mitochondrial dynamics in ECs.

Clbn is the fly homolog of yeast RQC2 or Tae2. RQC2/Tae2 is a subunit of the ribosome-associated quality control (RQC) complex, which recruits the alanine- and threonine-charged tRNA to stalled ribosomes and is required for CAT- tailing in yeast [39]. In Drosophila, Clbn can also bind to Ala- and Thr-tRNAs, and loss of clbn rescued mitochondrial morphology, ATP production, and DA neuron loss phenotypes of PINK1 mutant [35]. To examine whether the mitochondrial defects in clbn mutant is associated with the quality control of ribosomes, we over-expressed dNOT4, dUPF1, deRF1 and dXRN1 in clbn mutant, these RQC pathway genes work in the same pathway as PINK1 to control mitochondrial function in ECs [40]. Over-expression of these RQC factors in ECs had no effect on ISCs over-proliferation and mitochondrial fragmentation phenotypes of clbn mutant (S7 Fig and Fig 4L–4Q), indicating that defect in intestinal homeostasis in clbn mutant is independent of quality control of ribosomes.

Multiple genes related with mitochondria are differentially expressed in clbn mutant

To investigate the intrinsic mechanisms of Clbn functions, we systemically analyzed genes expression in clbn mutant using a microarray assay. A whole genome transcripts with 18,952 probe sets representing 12,874 genes were evaluated, which is more than 72.4% of fly genes (17,772 genes, dmel_r6.25_FB2018_06 from flybase). Among them, 35 genes were up-regulated and 61 genes were down-regulated in clbn mutant (S1 Table) using the parameters of adjusted P < 0.05 and fold-change ≥ 2. To identify the functional differences that reflect the transcriptional difference between the clbn mutant and wild-type fly, a gene set enrichment analysis (GSEA) with GO biological process was performed, gene sets with P < 0.01 were chosen to indicate significance. The GO biological processes correlated with top 15 up-regulated gene sets (n = 112) were associated with stem cell differentiation, DNA repair, RNA processing, ribosome biogenesis etc, while top 15 down-regulated genes sets (n = 150) were associated with monocyte chemotaxis, amino acid transportation, muscle contraction and metabolisms etc (S8 Fig, S2 Table).

Interestingly, multiple differentially expressed genes in clbn mutant have been reported or inferred with functions in mitochondria (Fig 4R and 4S), although the top 15 up-regulated and down-regulated GO pathways do not include mitochondria (S8 Fig). Among these genes, sfl is inferred to regulate mitochondrial organization [41], Aplip1 regulates axonal transport of mitochondria and JNK cascade [42], and 312, also named MALSU1, is inferred as a component of mitochondrial assembly of ribosomal large subunit 1 [41]. Some genes regulate metabolic functions related with mitochondria, i.e. UQCR-11 and CG8199 are inferred as a component of ubiquinol-cytochrome c reductase complex or mitochondrial alpha-ketoglutarate dehydrogenase complex respectively [41], and CG9149 is inferred with mitochondria with unknown functions [41]. Moreover, VhaM9.7-c and CG33970 have ATPase activity [41], Calx is a Na/Ca-exchange protein and regulates calcium ion binding [43], AOX3 [44], CG18547, CG5493, and CG6910 are inferred to regulate oxidation-reduction process and electron transport [41, 45] (Fig 4R and 4S). We further validated a number of these genes using RT-qPCR analysis. Indeed, most of genes we detected showed similar relative expression pattern as in microarray data (Fig 4T).

Loss of clbn leads to mitochondrial defect

The functional output of mitochondria can be monitored by the mitochondrial membrane potential and ATP production. We used 5,5',6,6'-tetrachloro-1,1',3,3'- tetraethylbenzimidazolyl-carbocyanine iodide (JC-1), a more mitochondria specific cationic fluorescent dye, to quantify the mitochondrial membrane potential. At low membrane potential, JC-1 forms the green-fluorescent monomer, while at higher potential, JC-1 forms aggregate and shows red-fluorescence. We observed reduced JC-1 aggregate in clbn mutant (Fig 5A–5A”), indicating a mitochondrial depolarization in clbn mutant. Furthermore, the ATP level was significantly reduced in clbn mutant, which can also be rescued by restoration of clbn expression in ECs (Fig 5B).

Mitochondrial dynamics are closely connected with the level of ROS, and mitochondrial dysfunction could induce the production of ROS and activation of oxidative stress. We next monitored the ROS level in ECs using the ROS reporter gstD1-GFP and dihydroethidium (DHE) probe. As expected, increased expression of gstD1-GFP and DHE fluorescent intensity were observed in clbn mutant (Fig 5C–5D”). The accumulation of ROS in clbn mutant can be partially rescued by knock-down of drp1 (S9 Fig), which is consistent with that knock-down of drp1 partially rescued the ISC over-proliferation in clbn mutant (S6 Fig). Furthermore, we observed more γ-H2Av foci in ECs in clbn mutant (Fig 5E–5E”), the equivalent of mammalian γ-H2AX, indicating increased DNA damage in clbn mutant. All these data indicated that Clbn is an essential component of mitochondria in ECs and required for normal mitochondrial morphology and functions.

Increasing evidence suggests that the gut microbiota plays a vital role in mediating intestinal homeostasis. We also observed a high bacterial load in clbn mutant (S10A Fig), which is consistent with the loss of intestinal homeostasis in clbn mutant. To address the possibility that microbiota change contributes to the loss of intestinal homeostasis in clbn mutant, we reared germ-free clbn mutant and examined gut phenotype. The germ-free clbn mutant exhibited similar ISCs over-proliferation, mitochondrial fragmentation and ATP production phenotypes as the conventionally raised flies (S10B–S10H Fig), which confirmed that disrupted intestinal homeostasis in clbn mutant was caused by loss of clbn gene.

Activation of JNK and JAK/STAT signaling pathways contributes to over-proliferation of ISCs in clbn mutant

We further investigated the cellular signaling pathways that are responsible for over-proliferation of ISCs in clbn mutant. As reactive oxygen species activate the Jun N-terminal kinase (JNK) signaling pathway in Drosophila [27, 28], we monitored JNK activity by using an enhancer-trap line, puc-lacZ, as a reporter and examined its expression in ECs. In control flies, puc-LacZ expression was low (Fig 6A). However, a significant increased expression of puc-lacZ was observed in clbn mutant (Fig 6A'). Moreover, the expression level of puc-LacZ in clbn MARCM clones was greatly increased when compared with that in surrounding control ECs (Fig 6B–6B').

In Drosophila, JAK-STAT, EGFR, WNT and Dpp/TGF-β signaling pathways play major roles to regulate ISCs proliferation in midgut, at physiological condition or in response to stress [36, 46, 47]. We therefore performed RT-qPCR and examined expression level of ligands for these pathways. We observed the ~ 8-fold increase of unpaired 2 (upd2) expression, and ~ 19-fold increase of unpaired 3 (upd3) expression, the two ligands of the JAK/STAT pathway, whereas expression level of keren (krn, a ligand for EGFR) and dpp (a ligand for TGF-β) was not significantly changed (Fig 6C). Consistently, expression of upd-lacZ reporter was greatly increased in clbn mutant clones (Fig 6D–6D'), while staining signal of anti-diphosopho-ERK antibody (dpERK), a direct readout of MAPK signaling activity, was not changed in clbn mutant (S11 Fig).

We also measured activity of JAK-STAT, WNT and Notch signaling pathways using reporter lines. Compared with control flies, expression of 10 × stat-GFP, a Stat92E reporter, was greatly increased in the clbn mutant (Fig 6E–6E'), indicating that flies loss of clbn had aberrant activation of JAK-STAT signaling pathway. Moreover, knock-down of stat92E in ISCs/EBs can rescue the ISC over-proliferation phenotype in clbn mutant (Fig 6F–6F'). We did not observe any significant change of Notch or Wnt signaling pathways in clbn mutant, using a Notch responsive element (NRE) reporter NRE-GFP and a wingless reporter wg-lacz (S12 Fig). To provide further evidence that Clbn regulates ISCs proliferation non-autonomously, the wild-type control clones or clbn mutant clones were induced during adulthood. In the control clones, ISCs and EBs were distributed evenly along the midgut with high level of membrane-associated Armadillo/β-catenin protein (Fig 6G), and in clbn mutant clones, the aberrantly increased number of progenitor cells were present in proximity to the ECs (Fig 6G').

Intestinal epithelial junctions are disrupted in clbn mutant

As the clbn mutant displayed defects in intestinal barrier (Fig 1B–1E) and increased proliferation of ISCs, we therefore assessed whether intestinal epithelial junctions were damaged in clbn mutant. Immunostaining with anti-Armadillo antibody to mark cell junctions revealed that Armadillo was strongly expressed around progenitor cells and presented at the cell membrane of ECs, and all cells in intestinal epithelium were well organized and assembled in the control flies (Fig 7A and 7A'). However, in clbn mutant, the cellular architecture became aberrant and disorganized, and Armadillo protein presented at ECs membrane was decreased (Fig 7B and 7B'). The defect of cellular architecture in clbn mutant can be largely rescued through the EC-specific blocking of JNK activity by expression of a dominant negative form of Bsk (UAS-bsk^DN) (Fig 7C and 7C').

Flies loss of clbn are susceptible to stress

Recent studies suggest that intestinal regeneration makes an important contribution to fly viability and life span [36]. Flies with intestinal dysplasia are short-lived, and animals with impaired ISCs proliferation or daughter cells differentiation die faster when challenged with pathogens, genotoxins, or ROS inducing compounds [48]. We therefore investigated the ability of clbn mutant to resist stress. Under low nutrition condition, the clbn mutant had a significantly reduced lifespan: all clbn knock-out flies were dead at day-42 after eclosion, while more than 70% clbn heterozygous flies were still viable (Fig 7D). Moreover, the clbn mutant had greatly reduced viability when treated with dextran sodium sulfate (DSS) or enteropathogenic bacteria ECC-15 (Fig 7E and 7F). Consistent with shortened lifespan, flies with loss of clbn had more progenitor cells in the midgut at low nutrient condition or when treated with DSS or ECC-15 (S13 Fig). The increased mortality and over-proliferation of progenitor cells seen with low nutrition, DSS treatment or bacterial infection of clbn mutant can be largely rescued by inhibiting JNK signaling in ECs (Fig 7E and 7F and S13 Fig). On the other hand, blocking JNK signaling didn’t rescue the defective mitochondrial morphology in ECs (Fig 7G–7I), indicating that mitochondrial fragmentation is an upstream event to Clbn depletion.

Loss of clbn promotes tumor progression in gut

The sustained proliferative signaling and the reprogramming of energy metabolism are two of the hallmarks of cancer [49]. Our previous study demonstrated that over-expression of clbn in A549 lung cancer cell line can repress colony formation and invasiveness [32]. In addition, flies with loss of clbn have activated JAK-STAT signaling pathway which is responsible for cell proliferation (Fig 6), and changed mitochondrial metabolism (Fig 4). These findings promoted us to consider whether loss of clbn contributes to the tumor growth in response to tumorigenic stimuli in gut, although clbn knock-out flies generate no spontaneous tumors under physiological conditions. Over-activation of the mitogenic EGFR signaling pathway can establish a tumor model in the fly midgut [2, 50, 51], therefore, a constitutively active form of Ras oncogene at 85D (also known as Ras^V12) was induced by esg^ts>Gal4 to activate EGFR signaling pathway in the progenitor cells. In the control flies, progenitor cells were randomly scattered along the basement membrane in the midgut (S14A–S14A’ Fig). Consistent with aforementioned data, we observed robust proliferation of ISCs in clbn mutant (S14B Fig). The ISCs accumulated and attached to the basement membrane and had normal apical-basal polarity (S14B’ Fig). Forced expression of Ras^V12 in progenitor cells for 3 days resulted in hyper-proliferation of ISCs and tumor formation, with accumulation of aberrantly differentiated progenitor cells and loss of epithelial cell polarity (S14C and S14C′ Fig). Depletion of clbn enhanced this effect, promoting tumor mass growth and invasion induced by Ras^V12 expression (S14D and S14E Fig). This suggests that dysfunctional mitochondria dynamics and metabolisms and disrupted intestinal homeostasis caused by loss of clbn might contribute to ISC tumor initiation or progression.

Discussion

Modulation of mitochondrial dynamics and redox balance is vital for tissue homeostasis. Mitochondrial dynamics between the fusion and fission state play a critical role in determining mitochondrial morphology and functions. In this study, we established Clbn as an essential mitochondrial component in enterocytes that regulates mitochondrial dynamics, redox balance and intestinal homeostasis. Loss of clbn causes mitochondrial fragmentation, and blocking mitochondrial fission by knock-down of drp1 or over-expression of marf can largely rescue the mitochondrial morphology in clbn mutant. These data suggest that Clbn promotes mitochondrial fusion and/or inhibits fusion by negatively regulating Drp1, and/or positively regulating Mfn. As the main source of intracellular ROS, mitochondrial damage could trigger ROS over-production and leads to oxidative stress and tissue damage [52]. Recent studies indicate that progenitor-specific mitochondrial dysfunction has significant impacts on ISCs proliferation, differentiation and tissue homeostasis in Drosophila [30, 31]. Loss of Nrf2 in ISCs causes accumulation of reactive oxygen species and deterioration of gut homeostasis [30]. Prevention of mitochondrial fusion by knock-down of opa-1 or marf in intestinal progenitor cells can block differentiation into enterocytes [31]. Previous studies have shown that maintenance of mitochondria in ISCs is essential for tissue homeostasis [30, 31]; our data suggest that maintenance of mitochondrial dynamics and redox balance in ECs is also critical for ISCs proliferation and intestinal homeostasis. Mitochondrial dysfunction in ECs in clbn mutant results in lower mitochondrial membrane potential, less ATP production and higher level of ROS, which leads to ISCs over-proliferation and loss of intestinal homeostasis (Fig 8).

Our previous works have shown that Clbn has diversified functions, including nuclear exporting [32], DNA damage induced apoptosis [33] and cell cycle checkpoint [34]. A recent study also demonstrated that Clbn can bind to fly Ala- and Thr-tRNAs and involved in PINK1 associated RQC pathway [35]. In this report, we identify a novel function of clbn in maintaining intestinal homeostasis by regulating mitochondrial dynamics in ECs. This function seems not to be general, but EC-specific, because we did not observe abnormal mitochondria in other tissues, such as flight muscle, from clbn knock-out flies. This idea is further supported by the data that EC-specific restoration of clbn expression can largely rescue the defect of gut and shortened lifespan of clbn mutant. The mitochondrial and gut defect observed in the clbn mutant should be a direct consequence of abnormal mitochondrial dynamics in ECs caused by clbn depletion. Several lines of evidence support this notion. Firstly, Clbn is localized to mitochondrial outer membrane in ECs, indicating a role in mitochondrial regulation. And blocking JNK activity in ECs partially rescued the gut defect and longevity of clbn mutant, but didn’t rescue the defective mitochondrial morphology, indicating that mitochondrial fragmentation is an upstream event to Clbn depletion, but not a secondary effect caused by other defective processes. Secondly, ECs are terminally differentiated cells under cell cycle arrest, so the mitochondrial defect is less likely associated with the functions of Clbn in regulating G1/S transition. Lastly, over-expression of several factors in RQC pathways did not rescue mitochondrial morphology defect and ISCs hyper-proliferation in clbn mutant, which excludes the involvement of RQC in intestinal defects caused by clbn depletion.

Stem cell aging is a cause of organismal aging especially in high-turnover tissues such as intestine, and dysregulation of aging processes could promote tumor formation and progression. The incidence of spontaneous tumorigenesis in intestine is largely unaffected by mutation of clbn gene, however, the growth of intestine-tumor induced by activation of Ras gene was enhanced in clbn knock-out flies, indicating that increased oxidative stress induced by mitochondrial impairment could enhance Ras-driven epithelial tumorigenesis. Consistent with our results, a recent study demonstrated that blocking autophagy enhances Ras^V12-driven epithelial tissue over-growth via the accumulation of reactive oxygen species and activation of the JNK pathway [53]. Considering the versatility of Clbn functions, particularly its importance in regulating redox balance and mitochondrial dynamics, the relative contribution of each of these processes in modulating stress response and tissue homeostasis remains to be determined. It is likely that Clbn could be a coordinating regulator that links these processes together in response to various stresses. It will be interesting to determine whether SDCCAG1, a homologue of Clbn, also regulates tissue homeostasis in high-turnover tissues in mammals. Our work may provide new insights in the mechanisms that control redox balance and tissue homeostasis.

Materials and methods

Fly genetics

All flies were maintained at 25°C on standard corn meal unless specified. Fly lines used in this study were as follows: clbn ^1Q (clbn ^KO) and UAS-clbn were described in [32]; esg-Gal4, UAS-GFP; Su(H)GBE-Gal4, UAS-GFP; Myo1A-Gal4; Gal80^ts; delta-lacZ; upd-lacZ; wg-lacZ and 10 × stat-GFP were gifts from Dr. Lei Zhang, Shanghai Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences; pros-Gal4, mef2-Gal4, gstD1-GFP and mito-GFP were gifts from Dr. Wanzhong Ge, Zhejiang University; puc-lacZ was gift from Dr. Renjie Jiao, Guangzhou Medical University; act-Gal4, bsk ^DN; NRE-GFP; UAS-Ras ^V12; esg-lacZ; yw, hs-flp; UAS-GFP; tubGal4, FRT^82B, tubGal80; stat RNAi, drp1 RNAi, UAS-marf, UAS-dNOT4, UAS-dUPF1, UAS-deRF1, UAS-dXRN1 were obtained from Bloomington Stock Center.

MARCM analysis

To generate MARCM clones during adulthood, progenies of the desired genotypes were collected within 3–5 days after eclosion and were heat-shocked in a 37°C water bath for two 60-min heat shocks over 2 days. The adult clones were examined 7–10 days later.

Generation of anti-Pdm1 antibody

The cDNA fragment of Pdm1-aa 332–588 was amplified with primers pdm1-F: CCCAAGCTTGCCAACAGCCACGCAGTTCCA and pdm1-R: CCGCTCGAGTTTAGGGACTGTCCAGGGAGGGAT, and PCR product was constructed into Hind III/Xho I sites of pET-28a plasmid (Novagen), and protein expression was induced in BL21 competent cells. The gel slice corresponding to Pdm1 fusion protein was cut, crushed, emulsified with Freund's adjuvant and injected into rabbits to generate anti-Pdm1 antibody (Abgent Biotechnology, Suzhou, China). Sera were collected over a period of 2 months and purified with Protein A beads.

Mitochondrial fractionation

The guts from 5-day-old female adults were dissected in cold PBS and cut into small pieces, then treated with 1× trypsin in EDTA (Thermo Scientific) for 2 h at room temperature and homogenized. Cytosolic and mitochondrial fractions were isolated following the manufacturer′s instructions (Mitochondria Isolation Kit for Tissue, 89801, Thermo Scientific) and analyzed by Western blotting. The submitochondrial localization assay was performed as previously described [54]. The intact mitochondria were resuspended in 20 mM HEPES-KOH, pH 7.4, 0.6 M sorbitol. Protease K (100 μg/ml) was added to mitochondria preparation and kept on ice for 20 min. The reaction was stopped by the addition of 2 mM phenylmethylsulfonyl fluoride. Mitochondria were collected by centrifugation at 16,000 g for 5 min at 4°C and followed by SDS–PAGE and Western blot analyses.

Immunostaining and imaging

The intestines from female adults were dissected in cold PBS and fixed immediately in PBS containing 4% (weight/vol) paraformaldehyde. Samples were then washed in PBS with 0.1% Triton X-100 (vol/vol) three times, blocked in PBTB [PBT containing 5% (vol/vol) normal goat serum], and incubated with primary antibodies overnight. The following primary antibodies were used: rabbit anti-Clbn (1:400), mouse anti-Prospero (1:200, Developmental Studies Hybridoma Bank), rabbit anti-Pdm1 (1:400), mouse anti-Armadillo (1:50, Developmental Studies Hybridoma Bank), rabbit anti-β-galactosidase (1:200, Promega), rabbit anti-phospho–histone-H3 (1:200, Millipore), rabbit anti-H2AvD pS137 (1:200, Rockland), rabbit anti-dpERK (1:200, Cell Signaling Technology), Alexa-Fluor-555-conjugated Phalloidin (1:500, Thermo Fisher Scientific). After three washes with PBT, secondary anti-rabbit or anti-mouse fluorescence antibodies, including Alexa 488 and 555 (1:400, Cell Signaling Technology), were used. Samples were mounted and analyzed on a Olympus FV1000 confocal laser-scanning microscope. The images were processed using Adobe Photoshop, Illustrator, and ImageJ.

Transmission electron microscopy

The intestines from 15-day-old female adults were dissected in cold PBS and fixed immediately with 2.5% (vol/vol) glutaraldehyde with Phosphate Buffer (PB) (0.1 M, pH 7.4), washed four times in PB. Then tissues were immersed in 1% (weight/vol) OsO₄ and 1.5% (weight/vol) potassium ferricyanide aqueous solution at 4°C for 2 h. After washing with PB, tissues were dehydrated through graded alcohol (30, 50, 70, 80, 90, 100, 100%, 10min each) into pure acetone (2 × 10 min). Samples were infiltrated in graded mixtures (3:1, 1:1, 1:3) of acetone and SPI-PON812 resin (19.6 ml SPI-PON812, 6.6ml DDSA and 13.8ml NMA), then changed pure resin. Finally, tissues were embedded in pure resin with 1.5% BDMA and polymerized for 12 h at 45°C, 48 h at 60°C. The ultrathin sections (70 nm thick) were sectioned with microtome (Leica EM UC6), stained by lead citrate, and examined by a transmission electron microscope (FEI Tecnai Spirit120kV).

JC-1 assay

Whole guts from 15-day-old female adults were dissected and transferred into 5 μM JC-1 (Thermo Scientific) in DMSO and incubated in dark for 30 min at room temperature. Samples were washed 2 times and mounted in PBS. Images were taken in the 568 nm channel using the Olympus FV1000 confocal laser-scanning microscope and the number of JC-1 red fluorescent aggregates was counted using ImageJ software.

ATP measurement

For each measurement, 10 guts from 15-day-old female adults were dissected and the ATP level was measured using a luciferase-based bioluminescence assay (ATP Bioluminescence Assay Kit HS II; Roche Applied Science) according to manufacturer's instructions.

DHE assay

Dihydroethidium (DHE) staining was performed as previously described [28]. Briefly, guts from 15-day-old female adults were dissected and handled in Schneider′s medium throughout whole procedure. After incubation in 30 uM DHE (Invitrogen) for 5 min in dark at room temperature, samples were washed three times and mounted in Schneider′s medium. Images were captured immediately with a Olympus FV1000 confocal laser-scanning microscope.

Western blot

Protein extracts from mitochondrial fractionation were subjected to SDS/PAGE electrophoresis and transferred to polyvinylidene fluoride membrane. Membranes were immunoblotted with rabbit anti-Clbn (1:2,000), mouse anti-ATP5A (1:1,000, ab14748, Abcam), anti-KDEL (1:1,000, ab69659, Abcam), α-RpS6 (1:300, CST), α-Tom20 (1:800, #13929, CST), α-SOD2 (1:800, NB100-1992, Novus Biologicals), α-Cytochrome c (1:1,000, 7H8.2C12, Novus Biologicals), and mouse anti-tubulin (1:1,000, ab44928, Abcam) antibodies separately. Blots were subsequently incubated with horseradish peroxidase-conjugated goat-anti-rabbit or goat-anti-mouse secondary antibodies (GE Healthcare), and processed for chemiluminescence (GE Healthcare).

FACS

FACS were performed as described previously [55]. Briefly, 200 guts were dissected in RNase-free PBS and dissociated for 45 min using elastase. Cells were then sorted using FACS Aria III (BD Biosciences) and sorted straight into RNA extraction buffer. The following fly lines were used to select different cell populations: esg-Gal4 UAS-GFP; tub-Gal80ts (ISCs and EBs), tub-Gal80ts UAS-GFP; Pros-Gal4 (EEs), Myo1A-Gal4; tub-Gal80ts UAS-GFP (ECs).

RT-qPCR

Total RNA was extracted from 10 guts of 15-day-old female adults using TRIzol (Invitrogen), and cleaned with RNAeasy kit (QIAGEN). The complementary DNA (cDNA) was synthesized using the iScript cDNA synthesis kit (Bio-Rad). Quantitative PCR was performed using the iScript one step RT-PCR SYBR green kit (Bio-Rad) for three independent biological replicates. The ribosomal gene rp49 was used as the normalization control. All results were presented as mean ± SEM. Primers used for RT-qPCR were in S3 Table.

Microarray assay and analysis

Embryos of Oregon R and clbn knock-out flies at 18 hr were collected and aged 2–4 h, and mRNA were isolated using Qiagen Oligotex Direct mRNA isolation kit. The microarray assay was performed with a Affymetrix CHP Analysis, and gene expression was evaluated using Genechip Drosophila Genome 2.0 Array (Affymetrix), which includes 18,952 probe sets representing 12,874 genes (R package drosophila2.db). Gene expression data was preprocessed using rma (R package affy). Differential expression analysis was performed using limma (R package limma) to identify individual genes demonstrating enrichment. Differentially expressed genes with adjusted P < 0.05 and fold-change ≥ 2 were considered statistically significant. Volcano plot was used to visualize differential expression distribution. Functional enrichment analysis was performed to determine differences in gene sets between phenotypes (GSEA) using gsea function in R package clusterProfiler [56] with Gene Ontology biological process in msigdbr package. One thousand gene set permutations were run for each pathway. Minimal size of each gene set for analyzing is 10, and maximal size of genes annotated for testing is 500. Gene sets with P < 0.01 were chosen to indicate significance. Top 15 gene sets up-regulated and down-regulated were shown in bubble plot. Expression pattern of differentially expressed genes related with mitochondria, ATP, electron transport was shown in heat map.

Smurf assay

Intestinal integrity was determined using the Smurf assay as described [57]. Briefly, female flies of each genotype were transferred onto fresh medium containing FD&C blue dye #1 (2.5% w/v) for 8 h. A fly was counted as a Smurf when dye coloration was observed outside of the digestive tract.

Bacterial infection and DSS feeding assays

Erwinia carotovora ssp. carotovora 15 (Ecc15) was cultured in LB broth at 29°C for 16 h. Fiften-day-old female adults were starved in empty vials for 2 h at 25°C, then transferred to vials containing a piece of 2.5 cm × 3.75 cm chromatography paper. Chromatography paper was wetted with 2.5% sucrose solution (control), Ecc15 bacteria (final OD600 = 100) in 2.5% sucrose, or 5% Dextran sulfate sodium (DSS) in 2.5% sucrose solution as feeding medium. Flies were transferred to new vials with freshly prepared feeding medium every day. After 3 days, guts were dissected and examined.

Lifespan assay

The control flies and clbn ^KO flies were collected at eclosion. Fifteen flies were reared per tube to avoid overcrowding, and flies were transferred to a new tube every 3 days. Survival flies were checked on a daily base until natural death. At least 12 tubes per genotype were scored, and more than 100 flies were scored per genotype. Survival curves were plotted using Excel.

Bacterial load

Flies were surface sterilized with 70% ethanol and then dissected on ice in sterile PBS. Dissected intestines were stored in sterile eppendorf tubes at -80°C prior to DNA extraction and were then extracted using the PowerSoil DNA isolation kit (MoBio). The q-PCR reactions were conducted in three independent biological replicates using SYBR Green PCR master mix (Takara). Universal primers for the 16 S ribosomal RNA gene were against variable regions 1 (V1F) and 2 (V2R). The Drosophila Actin 5C gene was used for normalization. PCR primers are shown in S1 Table.

Generation of axenic flies

Germ-free flies were generated following the previous report with minor modification [58]. Three to 5 days old flies were transferred on fresh fruit juice agar plates. After 1 day of habituation, flies were allowed to lay eggs for 4–6 hours. Eggs were washed with 70% EtOH, sterile water and dechorionated using 10% bleach for ~ 10 min. Eggs were then transferred under a sterile flow hood and maintained on axenic food until adult stage.

Statistics

All statistical comparisons were performed using data collected from three biological replicates with origin 9.0. Unpaired two-tailed Student’s t-test was used to assess statistical significance between two genotypes/conditions and one-way ANOVA for comparison of multiple samples. The significance level was indicated as * for P < 0.05, ** for P < 0.01, and *** for P < 0.001.

Supporting information

S1 Fig [red]
Loss of causes increased number of progenitor cells and enteroendocine cells in flies older than 15 days.

S2 Fig [h]
Loss of causes increased number of ISCs and EBs.

S3 Fig [green]
Clbn is dispensable for cell differentiation in midgut.

S4 Fig [a]
EC-specific knock-down of results in increased ISCs proliferation.

S5 Fig [a]
Clbn is localized to mitochondrial outer membrane in ECs and loss of leads to mitochondrial fragmentation in ECs but not in flight muscle.

S6 Fig [green]
Knockdown of rescues ISC over-proliferation in clbn mutants.

S7 Fig [a]
Loss of leads to ISCs over-proliferation and mitochondrial fragmentation independent of ribosome-associated quality control pathways.

S8 Fig [tif]
Bubble plot based on GSEA analysis shows top 15 gene sets up-regulated and down-regulated.

S9 Fig [dhe]
Knockdown of rescues ROS accumulation in mutants.

S10 Fig [a]
Microbiota does not contribute to the intestinal defects in flies.

S11 Fig [dperk]
The epidermal growth factor receptor (EGFR) pathway is not induced in the mutant.

S12 Fig [tif]
The Notch and Wnt signaling activity is not changed in the mutant.

S13 Fig [green]
Loss of results in more progenitor cells in response to various stress.

S14 Fig [green]
Loss of promotes ISCs tumor growth.

S1 Table [xlsx]
Up and down-regulated genes in mutant.

S2 Table [xlsx]
List of gene sets for GSEA analysis.

S3 Table [xlsx]
Primers for RT-qPCR.