BIOLOGICAL OVERVIEW

Even a cursory look at the Drosophila head [Images] reveals an incredible elaboration of numerous structures: discrete, diverse and unique. What produces these complex yet well organized and differentiated results? Early in head differentiation the anterior posterior axis is marked by a subdivision into seven segments. The element primarily responsible for this subdivision is the action of gap genes, expressed in well defined anterior-posterior positions. Among the gap genes, buttonhead regulates cnc activation, and subsequently cnc regulates genes responsible for labral and mandibular development, specifically in the dorsal portion of the labral segment and the posterior lateral and ventral portion of the mandibular segment. cnc also functions in conjunction with Deformed, a homeotic gene expressed in the head.

In the trunk, segment polarity genes are activated by pair-rule genes, but this is not the case in the anterior of the embryo. Here gap genes activate segment polarity genes. The segment polarity genes hedgehog and wingless are two important targets of cnc and forkhead, expressed in the anterior and posterior gut anlagen. cnc is expressed in the labral region of the foregut, fated to give rise to the dorsal pharynx and fkh is expressed in the adjacent esophagus. fkh is responsible for the maintenance but not the initiation of wg synthesis in the invaginating esophageal primordium. cnc is responsible for the maintenance of wg in the dorsal pharyngeal domain of wingless expression. Expression of hedgehog is similarly affected in cnc and fkh mutants. It is not known whether the actions of cnc and fkh on hh and wg are direct or indirect (Mohler, 1995).

Deletion mutants of cnc coding sequences indicate that cnc functions are required for the normal development of both labral and mandibular structures (Mohler, 1995). In place of the missing mandibular structures, some maxillary structures - mouth hooks and cirri - are ectopically produced (Harding, 1995; Mohler, 1995). The genetic function of the homeotic gene Deformed (Dfd) is required in the cnc mutant background to produce ectopic mouth hooks, and Mohler (1995) have proposed that Dfd and cnc function in combination to specify mandibular identity. A protein isoform (CncB) from the Drosophila cap ‘n’ collar locus has been characterized that selectively represses cis-regulatory elements that are activated by the Hox protein Deformed. Analysis of the cnc gene reveals the presence of three isoforms: cncA, cncB, and cncC. The expression patterns of the three transcript isoforms were analyzed using exon-specific probes both on wild-type and EMS-induced cnc mutants. In wild-type embryos, a cncB probe detects cytoplasmic transcripts limited to the mandibular and labral segments from cellular blastoderm to the end of embryogenesis. The cncB transcripts are expressed throughout both anterior and posterior regions of the mandibular lobes. In contrast, a cncA-specific probe detects a ubiquitous distribution of presumably maternal RNA at syncytial and early cellular blastoderm stages. After cellular blastoderm, cncA transcripts are not detectable until stage 14, when the level of ubiquitous cytoplasmic transcript increases and remains high for the remainder of embryogenesis. cncC-specific probes also detect a ubiquitous distribution of mRNA in syncytial stage embryos and a low level ubiquitous expression pattern in embryos after stage 14 (McGinnis, 1998).

Based on the above results, the labral and mandibular stripes of transcription that were detected by Mohler, (1991) using a probe including the cnc common exons (A2 and A3), correspond primarily to cncB transcripts. Since cncB is the transcript isoform that is expressed throughout the entire mandibular segment during mid-embyronic stages, cncB is likely to encode the principal function that modulates Dfd function in the mandibular segment (Harding, 1995). To further test this hypothesis, an assay was carried out to see whether cncB transcript or protein abundance is altered in embryos homozygous for the cnc^2E16 and cnc^C7, mutations known to reveal interaction with Dfd. The pattern of zygotic RNA expression detected with a cncB probe is unaltered in the EMS-induced cnc mutant embryos. The signal due to cncA and cncC transcripts is also unchanged in these mutants. However, the use of polyclonal antiserum raised against the common domain of the cnc isoforms (anti-Cnc) indicates that CncB protein expression is strikingly reduced in both cnc^2E16 and cnc^C7 mutant embryos. In wild-type embryos, the anti-Cnc antiserum exhibits a low-level ubiquitous staining in syncytial embryos, presumably due to maternally deposited CncA and CncC isoforms. From cellular blastoderm (stage 5) until stage 14, the staining detected by the anti-Cnc antiserum is localized in the nuclei of mandibular and labral cells. Although the anti-Cnc antiserum used in these experiments cross-reacts with all three Cnc proteins, only cncB RNA expression is localized in mandibular and hypopharyngeal regions from stages 6 through 14. cnc^2E16 mutants (and cnc^C7 mutants) accumulate much lower levels of Cnc antigen in both mandibular and labral cells of stage 11 embryos. These results provide further evidence that the cnc^2E16 and cnc^C7 mutations result in a loss of cncB function, and is consistent with the idea that CncB protein is required to prevent the maxillary-promoting function of Dfd from being active in mandibular cells (McGinnis, 1998).

In another test of the functions of the Cnc protein isoforms, each of the cncA, cncB and cncC open reading frames were placed under the control of the heat-shock promoter in P-element vectors and transgenic fly strains were generated carrying these constructs. Using the Cnc common-region antiserum to stain heat-shocked embryos, it appears that all three isoforms are produced at similar levels, localized in nuclei and possess similar stabilities after ectopic expression. However, their morphogenetic and regulatory effects are quite dissimilar. Heat-shock-induced ectopic expression of CncA during embryogenesis has no effect on embryonic morphology. Nearly all of the hs-cncA embryos hatch and proceed through larval development, and many eclose as viable adults. In contrast, ectopic expression of CncB at mid-stages (4-10 hours) of embryonic development is lethal. When ectopic expression is induced at 6 to 8 hours after egg lay, a defective embryonic head phenotype, which resembles the mutant phenotype of strong Dfd hypomorphs is produced. These hs-cncB embryos develop with rudimentary mouth hooks, H-piece and cirri. In addition, the anterior portion of the lateralgräten are truncated. All of these structures are components of the head skeleton that are absent or abnormal in Dfd mutant embryos. The head defects seen in the hs-cncB embryos also include an absent or abnormal dorsal bridge, a structure that is usually unaffected in Dfd mutant embryos. Many other head structures that develop in a Dfd-independent manner, such as the antennal sense organ, vertical plates and T-ribs develop normally in the hs-cncB embryos. The hs-cncB head defects are produced at high penetrance (>95%) by heat shocks in mid-embryogenesis (4-10 hours). In 10%-70% of these embryos, depending on the stage of heat shock, abdominal denticles near the ventral midline are replaced with naked cuticle. Ectopic induction of hs-cncC at 6-8 hours of development also results in highly penetrant defects in head development that include the loss of maxillary mouth hooks and cirri as well as head involution defects that are more profound than those induced by hs-cncB. In addition to the morphological defects described for CncB, ectopic CncC induces the formation of an abnormal head sclerite that develops as an extension of the normal lateralgräten. The position and appearance of this extra fragment of head skeleton suggests that it might correspond to ectopic production of lateralgräten or longitudinal arms of the H-piece (McGinnis, 1998).

Since CncB encodes a function that is required and sufficient to antagonize the maxillary-promoting effects of the Hox gene Dfd, it is reasonable to ask if CncB protein acts upstream to repress Dfd transcription, or in parallel to inhibit Dfd protein function? It is possible for CncB to do both, since Dfd protein function is required to establish an autoactivation circuit that provides persistent Dfd transcription in maxillary and mandibular cells. In wild-type embryos at stage 9, both Dfd and CncB proteins are expressed throughout the entire mandibular segment. By stage 11, Dfd protein is present at lower levels in the anterior, when compared to posterior mandibular nuclei, while CncB protein persists at relatively high levels throughout the segment. Finally, at stage 13, Dfd protein expression is no longer detected in anterior mandibular nuclei, although it is still abundant in posterior nuclei. cnc is required for this progressive repression of Dfd expression in the anterior mandibular segment, since cnc null mutants as well as the EMS-induced mutants show inappropriate persistence of Dfd transcripts and protein after stage 11 in anterior mandibular cells. All of these data suggest that CncB is not capable of repressing Dfd expression before stage 11. But after this stage, CncB represses the maintenance phase of Dfd transcription in mandibular cells, perhaps by repressing the autoactivation circuit that is normally established during stages 9 and 10 (Zeng et al., 1994). CncB is found to be sufficient to repress Dfd transcription outside the mandibular segment. When CncB is ectopically expressed in embryos, Dfd transcript levels in the maxillary segment are reduced, especially in the anterior region of the segment. Only the CncB isoform is capable of repressing Dfd transcription. Neither the ectopic expression of CncA nor CncC have an effect on the abundance or pattern of Dfd transcripts in the maxillary epidermis. Since the phenotypic effect of hs-cncC in epidermal cells strongly resembles that of hs-cncB, this indicates that the effect of Cnc gene products on maxillary epidermal development may not require repression of Dfd transcription per se. However, various experiments show that the maxillary-promoting function of Dfd protein is reduced in the presence of CncB; this could either be due to CncB-mediated repression of the Dfd autoactivation circuit in ectopic positions or to CncB repression of downstream target elements of Dfd protein, or to both of these effects. It is concluded that CncB provides a mechanism to modulate the specificity of Hox morphogenetic outcomes, which results in an increase in the segmental diversity in the Drosophila head. (McGinnis, 1998).

Cytosolic O-GlcNAcylation and PNG1 maintain Drosophila gut homeostasis by regulating proliferation and apoptosis

It remains unknown how intracellular glycosylation, O-GlcNAcylation, interfaces with cellular components of the extracellular glycosylation machinery, like the cytosolic N-glycanase NGLY1. This study utilized the Drosophila gut and uncovered a pathway in which O-GlcNAcylation cooperates with the NGLY1 homologue PNG1 to regulate proliferation in intestinal stem cells (ISCs) and apoptosis in differentiated enterocytes. Further, the CncC antioxidant signaling pathway and ENGase, an enzyme involved in the processing of free oligosaccharides in the cytosol, interact with O-GlcNAc and PNG1 through regulation of protein aggregates to contribute to gut maintenance. These findings reveal a complex coordinated regulation between O-GlcNAcylation and the cytosolic glycanase PNG1 critical to balancing proliferation and apoptosis to maintain gut homeostasis (Na, 2022).

Intestinal stem cells regulate tissue homeostasis by balancing self-renewal, proliferation, and differentiation all of which are supported by elevated flux through the hexosamine biosynthetic pathway (HBP). Both N-linked glycosylation and intracellular O-GlcNAc modifications are regulated by the HBP pathway in a nutrient-sensing manner. However, how NGLY1 is utilized to control stem cell homeostasis and differentiation in cells remains largely unknown. This is a critical question as patients with NGLY1-deficiency display global developmental delay, movement disorder and growth retardation. Elevation of NGLY1 was observed in patients' tumor samples, suggesting a function in oncogenic signaling. In Drosophila, PNG1 mutants had severe developmental defects and reduced viability, with the surviving adults frequently sterile (Funakoshi, 2010). This study has identified a pathway by which PNG1 regulates ISC homeostasis in vivo. This study shows that PNG1 levels increased in ISC/EBs concomitant with O-GlcNAc. This interaction between PNG1 and O-GlcNAcylation is critical for maintaining normal ISC proliferation and differentiation. Thus, through their mutual regulation, OGT and PNG1 have key roles in both progenitor (ISCs/EBs) and differentiated cells (ECs) contributing to tissue homeostasis. Previous reports indicated that PNG1 null larvae have specific developmental abnormalities in their midgut that contributes to their lethality. Further, intestinal inflammation in Crohn's disease is associated with increased O-GlcNAc modification. A previous study also showed that increased O-GlcNAc promotes gut dysplasia through regulation of DNA damage (Na, 2020). Thus, PNG1 or O-GlcNAc might still be associated with gut dysfunction in a disease context (Na, 2022).

The regulation of O-GlcNAc by PNG1 and the interaction between PNG1 and O-GlcNAc has been implicated previously. In fact, GlcNAc supplementation partially rescued lethality associated with PNG1 knockdown (Funakoshi, 2010). Although the mechanism by which GlcNAc supplementation rescued these mutant flies has not been fully worked out, Gfat1 transcript levels were downregulated in PNG1 knockdown flies (Funakoshi, 2010). Gfat1 is the enzyme that controls the rate limiting step in the HBP to produce UDP-GlcNAc. Thus, PNG1 through regulation of Gfat1 could impact levels of UDP-GlcNAc and ultimately O-GlcNAc. Additionally, it has been hypothesized that the loss of PNG1 could increase the presence of intracellular N-GlcNAc modification, potentially interfering with O-GlcNAc mediated signaling. Therefore, alterations in UDP-GlcNAc levels or presence of intracellular N-GlcNAc upon PNG1-deficiency can interact with O-GlcNAc to regulate stem cell homeostasis (Na, 2022).

Previous reports have shown that Nrf1 undergoes NGLY1-mediated deglycosylation, followed by proteolytic cleavage and translocation into the nucleus as an active transcription factor. Loss of NGLY1 caused Nrf1 dysfunction, as evidenced by an enrichment of deregulated genes encoding proteasome components and proteins involved in oxidation reduction. Proteasome activity can induce an apoptotic cascade that leads to growth arrest and, subsequently, cell death. The current data indicated that PNG1 or OGT knockdown suppressed ISC proliferation, which was rescued by Oltipraz (CncC activation) treatment in ISCs/EBs. Furthermore, there was increased apoptosis in PNG1 or OGT knockdown with treatment compared to non-treated groups. Interestingly, it was found O-GlcNAc-induced intestinal dysplasia was rescued by knockdown of PNG1 in ISCs/EBs through regulation of ROS levels. Similarly, increases in global O-GlcNAcylation in embryos of diabetic mice caused an overproduction of ROS and subsequent oxidative and ER stress. It is known that activation of SKN-1A/Nrf1 also requires deglycosylation by PNG-1/NGLY1 in C. elegans. Further, SKN-1 is O-GlcNAc modified and translocates to the nucleus in ogt-1(ok430)-null worms. Together, these studies all suggest conserved functional connections between O-GlcNAc and Nrf family transcription factors. This study also showed EC-specific OGT or PNG1 knockdown-induced hyperproliferation and cell death was decreased by CncC activation. This data indicated OGT or PNG1 can be regulated by CncC activity in ISCs/EBs and ECs. CncC has high activity within ISCs/EBs of unstressed as well young ISCs and quiescent ISCs but decreases with age and damage. These data indicated that CncC acts to properly balance between signaling and damage responses necessary for tissue homeostasis. CncC activation increased ISC proliferation in ISCs/EBs and decreased ISC proliferation in ECs of OGT or PNG1 knockdown contributing towards tissue homeostasis. Another study showed that inhibition of NGLY1 resulted in Nrf1 being misprocessed, mislocated, and inactive, thus indicating that functional NGLY1 is essential for Nrf1 processing, nuclear translocation, and transcription factor activity (Tomlin, 2017). Therefore, the data suggests that PNG1 and OGT modulated by CncC activation contribute to ISC proliferation and ultimately regulating tissue homeostasis. Nrf2 activation was able to rescue the developmental growth of NGLY1 deficiency in worm and fly models. In cancer-initiating cells, ER stress-dependent (ROS-independent) CncC induction is an event necessary to maintain stemness. The data showed that PNG1 knockdown-induced Poly-UB accumulation and 26S proteasome expression that was rescued by CncC overexpression and chemical activation. Through functioning as a sensor of cytosolic proteasome activity and an activator of aggresomal formation, Nrf2 alleviates cell damages caused by proteasomal stress. Expression of proteasome subunit genes and mitophagy-related genes were broadly enhanced after sulforaphane (Keap1 inhibitor) treatment and pharmacologically induction of Nrf2 promotes mitophagy and ameliorates mitochondrial defect in Ngly1-/- cells. Thus, it is believed that the sensitized background of the OGT or PNG1 mutant provides an environment where CncC activation promotes proliferation to the normal level through regulation of proteasome activity and protein aggregation (Na, 2022).

In a previously published paper (Ha, 2020), it was shown that OGT overexpression and OGA knockdown in ISCs/EBs both increased O-GlcNAc levels and induced hyperproliferation of the stem cells, whereas OGT knockdown decreased proliferation. However, in differentiated ECs, OGT overexpression and OGA knockdown phenotypes were similar to the normal gut, whereas OGT knockdown elevated proliferation and cell death. In general, EC death promotes proliferation in order to maintain gut homeostasis. Here, NGLY1 knockdown in ISCs/EBs decreased proliferation and clone size but NGLY1 knockdown in ECs induced hyperproliferation and cell death and importantly decreased O-GlcNAc levels. Thus, the phenotypes of OGT and NGLY1 were similar, demonstrating that maintenance of OGT and NGLY1 protein expression is highly interdependent for the maintenance of tissue homeostasis. It is interesting that the progenitor and differentiated cell types within the gut respond differently to changes in O-GlcNAc. It is possible that a certain level of O-GlcNAcylation is needed to maintain stem cells and promote proliferation and self-renewal, however, differentiated cells that do not have the same energy and growth requirements are not as reliant on high levels of O-GlcNAc. On the other hand, both ISC/EBs and ECs require some level of O-GlcNAc and without OGT there is decreased proliferation in progenitor cells and increased cell death of ECs. There are a few possibilities how NGLY1 and OGT can collaboratively work, however, it is unlikely that they share protein targets. First, a previous publication showed that additional deletion of ENGase, another N-deglycosylating enzyme that leaves a single GlcNAc residue, alleviates some of the lethality of Ngly1-deficient mice. Thus, it is possible with the accumulation of aggregation prone intracellular N-GlcNacylated proteins, there is disruption of normal O-GlcNac signaling. These data also showed increased protein aggregation in OGT or NLGY1 knockdown that was rescued by ENGase knockdown. In addition, MYC-OGT protein levels in OGT overexpression fly guts were decreased by PNG1 knockdown. It is possible that loss of NGLY1 disrupts normal OGT degradation and thus impacts levels global of O-GlcNAcylation (Na, 2022).

This study has shown that ENGase levels increased in PNG1 or OGT knockdown ISCs/EBs and ECs. PNGase is involved in the process of endoplasmic reticulum associated degradation (ERAD), acting as a deglycosylating enzyme that cleaves N-glycans attached to ERAD substrates. The small molecule ENGase inhibitors have potential to treat pathogenesis associated with NGLY1 deficiency. Rabeprazole, a proton pump inhibitor, was identified as a potential ENGase inhibitor. It was demonstrated that the consequences of knockdown of OGT or PNG1 on ISC proliferation and ENGase activity was rescued by Rabeprazole treatment in ISCs/EBs or ECs. The data showed that cell death was elevated in ISCs/EBs-specific PNG1/OGT knockdown with Rabeprazole treatment compared to non-treated groups concomitant with an increase in ISC proliferation. On the other hand, cell death decreased in EC-specific PNG1 knockdown treated with Rabeprazole resulting in a decrease in ISC proliferation. It is known that loss of PNG1 function in cells can cause the accumulation of aberrant proteins in the cytosol and the interruption of ERAD. Further, downregulation of ER stress-related genes has been reported in B-cell-specific OGT mutant mice. The protective effects of O-GlcNAc are not limited to mitochondrial function but also rescue injury caused by ER stress. Therefore, NGLY1/OGT seems to be functionally associated with the ERAD machinery. More recently, using a model ERAD substrate, it was reported that the ablation of Ngly1 causes a disruption in the ERAD process in mouse embryonic fibroblast (MEF) cells. Moreover, lethality of mice bearing a knockout of the Ngly1-gene was partially rescued by the additional deletion of the Engase gene. Interestingly, this study showed that OGA knockdown rescued ENGase levels of PNG1 knockdown ISCs/EBs. Hence, these findings suggest that there is a correlation between OGT/PNG1 and ENGase contributing to tissue maintenance (Na, 2022).

Taken together, these findings implicate O-GlcNAc and PNG1 as key regulators of tissue maintenance. PNG1 can impact stem cell homeostasis through regulation of O-GlcNAc both in ISCs/EBs or ECs. Of significance is the finding that PNG1 and OGT phenotypes are rescued by modulating CncC and ENGase activity in ISCs/EBs or ECs. Thus, these findings reveal that nutrient-driven glycosylation contribute towards control of ISC and progenitor cell proliferation and EC cell death via regulation of CncC and ENGase. This study provides a platform for future designs of interventions in which changes in O-GlcNAc can be utilized as a therapeutic for stem-cell-derived diseases like cancer. This study also presents a molecular mechanism and unexpected pathway that can be targeted for treating NGLY1-deificient patients (Na, 2022).

Loss of a proteostatic checkpoint in intestinal stem cells contributes to age-related epithelial dysfunction

A decline in protein homeostasis (proteostasis) has been proposed as a hallmark of aging. Somatic stem cells (SCs) uniquely maintain their proteostatic capacity through mechanisms that remain incompletely understood. This study describes and characterizes a 'proteostatic checkpoint' in Drosophila intestinal SCs (ISCs). Following a breakdown of proteostasis, ISCs coordinate cell cycle arrest with protein aggregate clearance by Atg8-mediated activation of the Nrf2-like transcription factor cap-n-collar C (CncC). CncC induces the cell cycle inhibitor Dacapo and proteolytic genes. The capacity to engage this checkpoint is lost in ISCs from aging flies, and it can be restored by treating flies with an Nrf2 activator, or by over-expression of CncC or Atg8a. This limits age-related intestinal barrier dysfunction and can result in lifespan extension. These findings identify a new mechanism by which somatic SCs preserve proteostasis, and highlight potential intervention strategies to maintain regenerative homeostasis (Rodriguez-Fernandez, 2019).

Protein Homeostasis (Proteostasis) encompasses the balance between protein synthesis, folding, re-folding and degradation, and is essential for the long-term preservation of cell and tissue function. It is achieved and regulated by a network of biological pathways that coordinate protein synthesis with degradation and cellular folding capacity in changing environmental conditions. This balance is perturbed in aging systems, likely as a consequence of elevated oxidative and metabolic stress, changes in protein turnover rates, decline in the protein degradation machinery, and changes in proteostatic control mechanisms. The resulting accumulation of misfolded and aggregated proteins is widely observed in aging tissues, and is characteristic of age-related diseases like Alzheimer's and Parkinson's disease. The age-related decline in proteostasis is especially pertinent in long-lived differentiated cells, which have to balance the turnover and production of long-lived aggregation-prone proteins over a timespan of years or decades. But it also affects the biology of somatic stem cells (SCs), whose unique quality-control mechanisms to preserve proteostasis are important for stemness and pluripotency (Rodriguez-Fernandez, 2019).

Common mechanisms to surveil, protect from, and respond to proteotoxic stress are the heat shock response (HSR) and the organelle-specific unfolded protein response (UPR). When activated, both stress pathways lead to the upregulation of molecular chaperones that are critical for the refolding of damaged proteins and for avoiding the accumulation of toxic aggregates. If changes to the proteome are irreversible, misfolded proteins are degraded by the proteasome or by autophagy. While all cells are capable of activating these stress response pathways, SCs deal with proteotoxic stress in a specific and state-dependent manner. Embryonic SCs (ESCs) exhibit a unique pattern of chaperone expression and elevated 19S proteasome activity, characteristics that decline upon differentiation. ESCs share elevated expression of specific chaperones (e.g., HspA5, HspA8) and co-chaperones (e.g., Hop) with mesenchymal SCs (MSCs) and neuronal SCs (NSCs), and elevated macroautophagy (hereafter referred to as autophagy) with hematopoietic SCs (HSCs), MSCs, dermal, and epidermal SCs. Defective autophagy contributes to HSC aging. It has further been proposed that SCs can resolve proteostatic stress by asymmetric segregation of damaged proteins, a concept first described in yeast (Rodriguez-Fernandez, 2019).

While these studies reveal unique proteostatic capacity and regulation in SCs, how the proteostatic machinery is linked to SC activity and regenerative capacity, and how specific proteostatic mechanisms in somatic SCs ensure that tissue homeostasis is preserved in the long term, remains to be established. Drosophila intestinal stem cells (ISCs) are an excellent model system to address these questions. ISCs constitute the vast majority of mitotically competent cells in the intestinal epithelium of the fly, regenerating all differentiated cell types in response to tissue damage. Advances made by numerous groups have uncovered many of the signaling pathways regulating ISC proliferation and self-renewal. In aging flies, the intestinal epithelium becomes dysfunctional, exhibiting hyperplasia and mis-differentiation of ISCs and daughter cells. This age-related loss of homeostasis is associated with inflammatory conditions that are characterized by commensal dysbiosis, chronic innate immune activation, and increased oxidative stress. It further seems to be associated with a loss of proteostatic capacity in ISCs, as illustrated by the constitutive activation of the unfolded protein response of the endoplasmic reticulum (UPR-ER), which results in elevated oxidative stress, and constitutive activation of JNK and PERK kinases. Accordingly, reducing PERK expression in ISCs is sufficient to promote homeostasis and extend lifespan (Rodriguez-Fernandez, 2019).

ISCs of old flies also exhibit chronic inactivation of the Nrf2 homologue CncC. CncC and Nrf2 are considered master regulators of the antioxidant response, and are negatively regulated by the ubiquitin ligase Keap1. In both flies and mice, this pathway controls SC proliferation and epithelial homeostasis. It is regulated in a complex and cell-type specific manner. Canonically, Nrf2 dissociates from Keap1 in response to oxidative stress and accumulates in the nucleus, inducing the expression of antioxidant genes. Drosophila ISCs, in turn, exhibit a 'reverse stress response' that results in CncC inactivation in response to oxidative stress. This response is required for stress-induced ISC proliferation, including in response to excessive ER stress, and is likely mediated by a JNK/Fos/Keap1 pathway (Rodriguez-Fernandez, 2019).

The Nrf2 pathway has also been linked to proteostatic control: 'Non-canonical' activation of Nrf2 by proteostatic stress as a consequence of an association between Keap1 and the autophagy scaffold protein p62 has been described in mammals. A similar non-canonical activation of Nrf2 has been described in Drosophila, where CncC activation is a consequence of the interaction of Keap1 with Atg8a, the fly homologue of the autophagy protein LC3. Nrf2/CncC activation induces proteostatic gene expression, including of p62 in mammalian cells and of p62/Ref2P and LC3/Atg8a in flies. Nrf2 is further a central transcriptional regulator of the proteasome in both Drosophila and mammals. Whether and how Nrf2 also influences proteostatic gene expression in somatic SCs remains unclear (Rodriguez-Fernandez, 2019).

This study shows that Drosophila CncC links cell cycle control with proteostatic responses in ISCs via the accumulation of dacapo, a p21 cell cycle inhibitor homologue, as well as the transcriptional activation of genes encoding proteases and proteasome subunits. This study establish that this program constitutes a transient 'proteostatic checkpoint', which allows clearance of protein aggregates before cell cycle activity is resumed. In old flies, this checkpoint is impaired and can be reactivated with a CncC activator (Rodriguez-Fernandez, 2019).

The central role of Nrf2/CncC in the proteostatic checkpoint is consistent with its previously described and evolutionarily conserved influence on longevity and tissue homeostasis, and is likely to be conserved in mammalian SC populations, as Nrf2 has for example been shown to influence proliferative activity, self-renewal and differentiation in tracheal basal cells. It may be unique to somatic SCs, however, as CncC or Nrf2-mediated inhibition of cell proliferation is not observed during development (such as in imaginal discs) or in other dividing cells. Assessing the existence of an Nrf2-induced proteostatic checkpoint in mammalian SC populations will be an important future endeavor (Rodriguez-Fernandez, 2019).

Mechanistically, the results support a model in which the presence of protein aggregates activates CncC through Atg8a-mediated sequestration of Keap1. In mammals, Nrf2 activation can also be achieved through the interaction of Keap1 with the Atg8a homologue LC3 and p62, and ref2p/p62 contributes to the degradation of polyQ aggregates in Drosophila, suggesting that a conserved Atg8a/p62/Keap1 interaction may be involved in the activation of the proteostatic checkpoint. The activation of CncC after cytosolic proteostatic stress described in this study thus differs mechanistically and in its consequence from the regulation of CncC after other types of protein stress in ISCs: in response to unfolded protein stress in the ER, CncC is specifically inactivated by a ROS/JNK-mediated signaling pathway. This mechanism allows ISC proliferation to be increased in response to tissue damage, but can also contribute to the loss of tissue homeostasis in aging conditions. The activation of CncC after cytosolic protein stress, in turn, allows arresting ISC proliferation during protein aggregate clearance. The distinct responses of ISCs to cytosolic or ER-localized proteostatic stress has interesting implications for understanding of the maintenance of tissue homeostasis. While the XBP1-mediated UPR-ER allows the expansion of the ER and the induction of ER chaperones to deal with a high load of unfolded proteins in the ER, it also stimulates ISC proliferation through oxidative stress and the activation of PERK and JNK. It is tempting to speculate that the sequestration of unfolded proteins within the ER allows ISCs to proceed through mitosis without the possibility of major misregulation, while the presence of cytosolic protein aggregates may be a unique danger to the viability of the cell and its daughters. It seems likely that constitutive activation of autophagy and proteasome pathways during the clearance of cytosolic aggregates is incompatible with the need for intricate regulation of these same pathways during the cell cycle in proliferating ISCs. It will be of interest to explore this hypothesis further in the future (Rodriguez-Fernandez, 2019).

The data suggest that the coordination of cell cycle arrest and aggregate clearance is achieved by the simultaneous induction of the cell cycle inhibitor Dacapo and a battery of genes encoding proteins involved in proteolysis. While it was possible to detect dacapo transcript expression in ISCs by fluorescent in situ hybridization at 24h after HttQ138 expression, it remains unclear whether Dacapo is induced directly by CncC or via the action of a CncC target gene. It is surprising that transcriptional induction of autophagy genes in was not seen in a RNAseq experiment, but it is possible that this is due to the fact that only one timepoint was sampled after induction of protein aggregates. Since the transcriptional response of autophagy genes is likely very dynamic, a more time-resolved transcriptome analysis during aggregate formation and clearance may have captured such a response (Rodriguez-Fernandez, 2019).

It is further notable that dap deficient ISC clones exhibit a significantly higher aggregate load in these experiments than wild-type ISC clones. This suggests that the induction of proteolytic genes and of cell cycle regulators is not only coincidentally linked by CncC, but that aggregate clearance and the cell cycle arrest mediated by Dacapo need to be tightly coordinated for effective ISC proteostasis. It will be interesting to explore the mechanism of this requirement in the future. It is tempting to speculate that, as the elimination of protein aggregates requires an increase in proteasome activity, and proteasome activity can influence cell cycle timing, cell cycle inhibition is a critical safeguard against de-regulation of normal cell cycle progression (Rodriguez-Fernandez, 2019).

The data suggest that Atg8a induction in ISCs experiencing proteostatic stress may serve a dual purpose: sustained activation of the proteostatic checkpoint as well as increased autophagy flux. This dual role is distinct from other autophagy components like Atg1, since Atg1 over-expression, an efficient way of promoting autophagy in Drosophila cells, counteracts the checkpoint rather than promoting it. Exploring the relative kinetics of Atg8a and Atg1 induction in ISCs after proteostatic stress is likely to provide deeper mechanistic insight into the regulation of the proteostatic checkpoint (Rodriguez-Fernandez, 2019).

Critically, the proteostatic checkpoint is reversible. Based on the current data and previous studies, it is proposed that upon clearance of aggregates, the Keap1/Atg8a interaction is decreased, thus releasing Keap1 to inhibit CncC. Lineage-tracing studies show that this allows re-activation of ISC proliferation and recovery of normal regenerative responses (Rodriguez-Fernandez, 2019).

The loss of proteostatic checkpoint efficiency in ISCs of old guts is likely a consequence of the age-related inactivation of CncC in these cells (possibly caused by chronic oxidative stres. Accordingly, reactivating Nrf2/CncC in the gut by overexpressing CncC is sufficient to restore epithelial homeostasis in the intestine of old flies, and this study found that exposing animals to Otipraz intermittently late in life promotes epithelial barrier function and extends lifespan (Rodriguez-Fernandez, 2019).

Since Nrf2/CncC and other components required for the proteostatic checkpoint are conserved across species, it is anticipated that the current findings will be relevant to homeostatic preservation of adult SCs in vertebrates. Supporting this view, mammalian Cdkn1a (p21) has been described as an Nrf2 target gene. Transient activation of Nrf2 may thus be a viable intervention strategy to improve proteostasis and maintain regenerative capacity in high-turnover tissues of aging individuals (Rodriguez-Fernandez, 2019).

Roles of C/EBP class bZip proteins in the growth and cell competition of Rp ('Minute') mutants in Drosophila

Reduced copy number of ribosomal protein (Rp) genes adversely affects both flies and mammals. Xrp1 encodes a reportedly Drosophila-specific AT-hook, bZIP protein responsible for many of the effects including the elimination of Rp mutant cells by competition with wild type cells. Irbp18, an evolutionarily conserved bZIP gene, heterodimerizes with Xrp1 and with another bZip protein, dATF4. This study shows that Irbp18 is required for the effects of Xrp1, whereas dATF4 does not share the same phenotype, indicating that Xrp1/Irbp18 is the complex active in Rp mutant cells, independently of other complexes that share Irbp18. Xrp1 and Irbp18 transcripts and proteins are upregulated in Rp mutant cells by auto-regulatory expression that depends on the Xrp1 DNA binding domains and is necessary for cell competition. Xrp1 is conserved beyond Drosophila, although under positive selection for rapid evolution, and that at least one human bZip protein can similarly affect Drosophila development (Blanco, 2020).

Heterozygous mutation of ribosomal protein genes lead to cell-autonomous, deleterious phenotypes in both flies and mammals and provide the classic example of a genotype that is eliminated from mosaics by competition. There is increasing interest in the potential roles of cell competition in mammalian development, cancer development, and in regenerative medicine. A remarkable recent finding from Drosophila is that many of the phenotypic effects of mutating ribosomal protein genes are mediated by a putative transcription factor, Xrp1, rather than as a direct consequence of altered ribosome number. Accordingly, Xrp1 plays a key role in the elimination of Rp mutant cells by cell competition. Xrp1 transcription and protein expression are elevated in Rp mutant cells, restricting translation, cellular growth rate, and the rate of organismal development, and enabling cell competition with nearby wild type cells. Xrp1 had previously been implicated in the DNA damage response downstream of p53 and in the transposition of P elements, and contributes to the pathology of a Drosophila model of Amyotrophic Lateral Sclerosis, as well as to the coordination of organ growth in flies with Rp gene knockdowns (Blanco, 2020).

Xrp1 has been reported not to have homologs in other eukaryotes. This seems surprising given the highly conserved and fundamental roles of ribosomal proteins, and is a barrier to investigating the potential conservation of cell competition mechanisms and the roles of cell competition in mammals, for example in the development of cancer. Xrp1 binds to DNA as a heterodimer with Irbp18, the Drosophila homolog of the C/EBP protein family, which is a conserved protein and co-purifies with it in cultured cells. Irbp18 in turn heterodimerizes with the conserved protein dATF4, encoded by the crc gene in Drosophila). This led to an investigation od whether it is the Xrp1 heterodimer with the conserved Irbp18 protein that functions in Rp+/- cells, and if so whether Xrp1/Irbp18 acts positively; alternatively, Xrp1 could act as a competitive inhibitor of Irbp18 function with its other partner, dAtf4/Crc, in which case Xrp1 could represent a Drosophila-specific regulator of a more conserved pathway (Blanco, 2020).

The current data provide overwhelming genetic evidence that Xrp1 does function along with Irbp18. Like Xrp1 mutations, irbp18 mutation suppressed multiple effects of Rp mutations, including the elimination of Rp+/- mutant cells by competitive apoptosis in the proximity of wild type cells and the reduced growth of Rp+/- wing cells. Like Xrp1, irbp18 was also required for the prompt disappearance of Rp-/- cell clones, which survived in the irbp18 mutant background. All these data were consistent with the model that Xrp1/Irbp18 heterodimers are the active species in Rp mutant cells and were inconsistent with the idea that Xrp1 might act as a competitive inhibitor of other Irbp18-containing species, as this would have predicted that Irb18 mutations would have had phenotypes opposite to those of Xrp1 (Blanco, 2020).

The phenotype of Crc knockdown is different from those of Xrp1 and irbp18 mutations. Whereas Xrp1 and irbp18 mutations enhance the growth and competitiveness of Rp+/- cells, crc knockdown greatly diminished growth and survival of Rp+/- cells. If Xrp1 was a Drosophila-specific competitive inhibitor of a conserved Crc/Irbp18 heterodimer that was required for growth, both irbp18 and crc mutants would show reduced growth, similar to Rp+/- genotypes. In contrast to this, irbp18 mutants have little phenotype except in Rp+/- genotypes, where their effects closely resemble those of Xrp1 mutants. Also, whereas crc knockdown strongly and cell-autonomously affected the growth of Rp+/- cells, it had less effect on Rp+/+ cells (Blanco, 2020).

In addition to these findings in loss-of-function experiments, it was also found that Xrp1 over-expression phenotypes depended on IRBP18, as would be expected if these proteins function together. It was also found that Xrp1 over-expression at higher temperatures resulted in a still stronger eye phenotype where simultaneous irbp18 mutation restored normal eye size but did not completely restore eye morphology. This is consistent with some IRBP18-independent component to ectopic Xrp1 function that is either cold-sensitive or only apparent at the highest ectopic expression levels. There is not yet any evidence whether these over-expression effects are physiologically relevant (Blanco, 2020).

Taken together, these findings suggest that the Xrp1/IRBP18 and Crc/IRBP18 heterodimers have independent and perhaps unrelated functions. Consistent with this, ectopic expression of IRBP18 had no phenotypic effect, suggesting that IRBP18 is normally made in excess, so that it is Xrp1 that is limiting for the growth inhibiting activities of the Xrp1/IRBP18 heterodimers, which do not impact IRBP18 availability sufficiently to affect Crc/Irbp18 functions (Blanco, 2020).

As expected if Xrp1 functions in a heterodimer, the basic and Leucine Zipper domains were important for Xrp1 function, as was the AT hook. In over-expression assays only, there could be reduced activity of proteins deleted for any of these domains individually, but not of a truncation that precedes them all. When encoded from the endogenous locus, basic and AT-hook domains appeared absolutely required (Blanco, 2020).

Mutual auto-regulation may be a significant feature of Xrp1 and IRBP18 function. As noted previously, the elevated Xrp1 and irbp18 transcription observed in Rp+/- wing discs is dependent on Xrp1 function. This study shows that IRBP18 protein levels are also elevated in Rp+/- cells in an Xrp1-dependent fashion, and that irbp18 is also required for the autoregulation. In principle, autoregulation could have been the major or indeed the only transcriptional function of the Xrp1/IRBP18 heterodimer, ie perhaps these proteins could control cell competition through other mechanisms once levels were sufficient. This cannot be completely correct, however, because Xrp1 was still substantially dependent on the irbp18 gene and on the Leucine Zipper and DNA binding domains when expressed using GAL4-driven transgenes that are independent of auto-regulation, so by-passing the requirement of auto-regulation does not relieve the requirements for heterodimerization and DNA binding domains. It is also worth noting that Xrp1 and irbp18 are both required to promptly eliminate Rp-/- cells, where their expression does not require auto-regulation. Although implicating other transcriptional targets of Xrp1/IRBP18 in Rp+/- and Rp-/- cells, these studies do not rule out other functions besides transcription (Blanco, 2020).

Previously it was thought that Xrp1 was restricted to the genus Drosophila, a surprising finding for a protein that has an important cellular function. This study found, however, that Xrp1 genes have been under strong positive selection for rapid evolutionary change. Recurrent positive selection is often the sign of an evolutionary arms race, such as are often driven by host-pathogen interactions, sexual competition, or intra-genomic conflict. Possibly pathogens target Xrp1 to promote growth and survival of infected cells. It is interesting that Xrp1 is already documented to interact with one transposable element, the P element. However, none of these scenarios for positive selection, or indeed additional possibilities, can yet be ruled out (Blanco, 2020).

Rapid divergence makes homology difficult to detect, and accordingly this study has identify divergent Xrp1 homologs in other Dipteran insects that have not previously been annotated because their sequence similarities are restricted to the key DNA-binding portion of the protein, and to a more amino-terminal Xrp1-homology domain. The failure to identify still more distant homologs might be genuine, or might reflect further divergence beyond ones ability to recognize homology. Mammals do contain other members of the C/EBP protein family without identified Drosophila homologs, and this study shows that DDIT3 (aka CHOP and C/EBP-Z) can generate a similar phenotype to Xrp1 when expressed in Drosophila. Interestingly C/EBP-α, one of the mammalian proteins more related to Irbp18, has been implicated in a cell competition-like phenomenon, the elimination of cells from the multipotent hematopoietic stem cell niche following irradiation (Blanco, 2020).

Xrp1 and Irbp18 trigger a feed-forward loop of proteotoxic stress to induce the loser status

Cell competition induces the elimination of less-fit 'loser' cells by more fit 'winner' cells. In Drosophila, cells heterozygous mutant in ribosome genes, Rp/+, known as Minutes, are outcompeted by wild-type cells. Rp/+ cells display proteotoxic stress and the oxidative stress response, which drive the loser status. Minute cell competition also requires the transcription factors Irbp18 and Xrp1, but how these contribute to the loser status is partially understood. This study provided evidence that initial proteotoxic stress in RpS3/+ cells is Xrp1-independent. However, Xrp1 is sufficient to induce proteotoxic stress in otherwise wild-type cells and is necessary for the high levels of proteotoxic stress found in RpS3/+ cells. Surprisingly, Xrp1 is also induced downstream of proteotoxic stress, and is required for the competitive elimination of cells suffering from proteotoxic stress or overexpressing Nrf2. These data suggests that a feed-forward loop between Xrp1, proteotoxic stress, and Nrf2 drives Minute cells to become losers (Langton, 2021).

Cells within a tissue may become damaged due to spontaneous or environmentally induced mutations, and it is beneficial to organismal health if these cells are removed and replaced by healthy cells. During cell competition, fitter cells, termed winners, recognise and eliminate less-fit cells, termed losers, resulting in restoration of tissue homoeostasis. Cell competition therefore promotes tissue health and is thought to provide a level of protection against developmental aberrations and against cancer by removing cells carrying oncoplastic mutations. However, an increasing body of evidence indicates that cell competition can also promote growth of established tumours, enabling them to expand at the expense of surrounding healthy cells (Langton, 2021).

Minute cell competition was discovered through the study of a class of Drosophila ribosomal mutations called Minutes and initial work suggests that it is conserved in mammals. While homozygous Rp mutations are mostly cell lethal, heterozygosity for most Rp mutations gives rise to viable adult flies that exhibit a range of phenotypes including developmental delay and shortened macrochaete bristles. Rp/+ tissues display a higher cell-autonomous death frequency than wild-type tissues, and competitive interactions further elevate cell death in Rp/+ cells bordering wild-type cells, contributing to progressive loss of Rp/+ cells over time (Langton, 2021).

It was suggested that Rp/+ cells are eliminated by cell competition due to their reduced translation rate. However, it has beem shown that Rp/+ cells experience significant proteotoxic stress and this is the main driver of their loser status. Rp/+ cells have a stoichiometric imbalance of ribosome subunits, which may provide the source of proteotoxic stress. The autophagy and proteasomal machineries become overloaded and protein aggregates build up in Rp/+ cells, leading to activation of stress pathways. This includes activation of Nuclear factor erythroid 2-related factor 2 (Nrf2) and of the oxidative stress response, whichia sufficient to cause the loser status. Restoring proteostasis in Rp/+ cells suppresses the activation of the oxidative stress response and inhibits both autonomous and competitive cell death (Langton, 2021).

Genetic screening for suppressors of cell competition led to the identification of Xrp1, a basic leucine Zipper (bZip) transcription factor. Loss of Xrp1 rescues both the reduced growth and competitive cell death of Rp/+ cells in mosaic tissues. Consistently, loss of Xrp1 restores translation rates and abolishes the increased JNK pathway activity characteristic of Rp/+ cells. Xrp1 forms heterodimers with another bZip transcription factor called Inverted repeat binding protein 18kDa (Irbp18), and removal of Irbp18 also strongly suppresses the competitive elimination of Rp/+ cells in mosaic tissues. Irbp18 and Xrp1 are transcriptionally upregulated and mutually required for each other's expression in Rp/+ cells, suggesting they function together in Minute cell competition. Irbp18 forms heterodimers with another bZip transcription factor, ATF4. However, knockdown of ATF4 in Rp/+ cells reduces their survival in mosaic tissues, which is the opposite effect to knockdown of Xrp1 or Irbp18. This has been interpreted to suggest that the ATF4-Irbp18 heterodimer acts independently to the Xrp1-Irbp18 heterodimer (Langton, 2021).

How the Xrp1/Irbp18 complex contributes to the loser status is not clear. Given the recently identified role of proteotoxic stress in cell competition this study sought to establish whether Xrp1/Irbp18 and proteotoxic stress act independently or in the same pathway to contribute to cell competition in Rp/+ cells. This study identified a feed-forward loop between Xrp1/Irbp18 and proteotoxic stress, which is required for downstream activation of the oxidative stress response and the loser status. The data suggests a model in which the initial insult in RpS3/+ cells is ribosomal imbalance-induced proteotoxic stress, which is Xrp1 independent. Xrp1 is then transcriptionally activated downstream of proteotoxic stress, by increased phosphorylated-eukaryotic Initiation Factor 2α (p-eIF2α), and possibly by Nrf2. The Xrp1-Irbp18 complex then induces further proteotoxic stress, completing the feed-forward loop. This work provides new insight into the interactions between the stress signalling pathways active in Rp/+ cells and provides a mechanism for how the Xrp1-Irbp18 heterodimer mediates the competitive elimination of Rp/+ cells by wild-type cells (Langton, 2021).

This study provided evidence that a feed-forward loop between proteotoxic stress, Nrf2 and the Xrp1/Irbp18 complex is operational in RpS3/+ cells (including in the absence of cell competition) and contributes to reducing their fitness during cell competition. The data suggests that an imbalance between SSU and LSU Ribosomal proteins generates an initial source of proteotoxic stress, independently of Xrp1. This leads to xrp1 transcriptional upregulation, likely via p-eIF2α. Xrp1, together with Irbp18, generates further proteotoxic stress, in a feed-forward loop. This causes LSU ribosome proteins to accumulate, exacerbating the stoichiometric imbalance between LSU and SSU subunit components in RpS3/+ cells. Knockdown of Xrp1 or Irbp18 rescues proteotoxic stress in RpS3/+ cells, suggesting that this feed-forward loop is essential for build-up of proteotoxic stress and to reduce the competitiveness of Rp/+ cells. It is noted that during the revision of this manuscript two other independent studies have reported relevant and complementary findings. Nrf2 is also activated by proteotoxic stress and contributes to this feedback loop, either independently of p-eIF2α, or downstream of p-eIF2α. The data cannot distinguish between these two possibilities (Langton, 2021).

The data indicate that xrp1 upregulation is likely mediated by increased p-eIF2α levels. p-eIF2α accumulates in Rp/+ cells, and increasing p-eIF2α in wild-type cells (by knocking down GADD34) leads to increased xrp1 transcription, suggesting that p-eIF2α does, at least partially, contribute to xrp1 transcription in Rp/+ cells. p-eIF2α induces many transcriptional targets via stabilization of the transcription factor ATF4. This suggested that ATF4 may activate xrp1. Consistent with this, it was found that ATF4 overexpression is sufficient to upregulate an xrp1 transcriptional reporter in wing disc cells. However, it was surprising to find that xrp1 upregulation does not seem to depend on ATF4 in RpS3/+ cells. Indeed, ATF4 knockdown did not reduce xrp1 transcription in RpS3/+ cells. Furthermore, it was not possible to detect stabilization of ATF4 in RpS3/+ cells using a translational reporter. These observations suggest that p-eIF2α upregulates xrp1 transcription in Rp/+ cells by an unknown, ATF4 independent, mechanism. Alternatively, the role of ATF4 may be masked by other inputs onto the xrp1 promoter. For example, ATF4 knockdown could increase proteotoxic stress in Rp/+ cells, by inhibiting the UPR, and this may upregulate other pathways that act on the xrp1 promoter, thus masking any effect of ATF4 knockdown. This mechanism could involve Nrf2, since Nrf2 is also induced by proteotoxic stress and since it has been shown that Nrf2 induces cellular toxicity via xrp1. However, it is also possible that other factors activate Xrp1 in Rp/+ cells (Langton, 2021).

Nrf2 plays a pro-survival role in many contexts, by activating a battery of genes that enable the metabolic adaptation to oxidative stress. It is therefore counterintuitive that Nrf2 overexpression should induce the loser status and, at high expression levels, cell death. The current work suggests that the toxicity of Nrf2 is at least in part due to Xrp1 function, as elimination of Nrf2 expressing cells is rescued by Xrp1 knockdown. Whether additional Nrf2 target genes contribute to the loser status remains to be established (Langton, 2021).

Besides Xrp1 or Irbp18 knockdown, the only other condition known thus far to rescue xrp1 transcriptional upregulation in Rp/+ cells is an RpS12 point mutation, RpS1297D. However, the mechanism by which RpS12 affects xrp1 transcription remains elusive. It will be important in future work to establish whether RpS12 mutations rescue xrp1 transcriptional activation upstream or downstream of proteotoxic stress (Langton, 2021).

The results provide compelling evidence that Xrp1 and Irbp18 are responsible for inducing proteotoxic stress in RpS3/+ cells. Firstly, knockdown of Xrp1 or Irbp18 rescues the accumulation of p62 labelled aggregates and rescues the increased p-eIF2α in RpS3/+ cells. Secondly, overexpression of Xrp1 is sufficient to upregulate markers of proteotoxic stress in wild-type cells. Third, the presence of Xrp1 in RpS3/+ cells worsens the imbalance of Ribosomal proteins, causing LSUs to accumulate. It will be crucial in future work to identify the relevant targets of Xrp1 that cause proteotoxic stress in Rp/+ cells. Xrp1 may alter expression of a single target, for example a gene encoding a component or regulator of the autophagy or proteasomal systems, which deregulates cellular proteostasis. Alternatively, several target genes may contribute to enhancing proteotoxic stress: if several subunits of multi-protein complexes are deregulated by increased Xrp1, this could lead to unassembled complexes, increasing the burden on the cellular degradation machinery in already stressed Rp/+ cells. There may also be Xrp1 targets that contribute to the loser status without affecting proteotoxic stress. It is remarkable that, in addition to rescuing competitive elimination of Rp/+ cells, loss of Xrp1 can rescue elimination of mahj deficient cells and Nrf2 overexpressing cells. In mahj deficient cells, loss of Xrp1 was able to rescue the upregulation of p-eIF2α, suggesting that Xrp1 also promotes proteotoxic stress in mahj cells. It will be interesting to establish whether this is the case for Nrf2 expressing cells (Langton, 2021).

Xrp1 has been shown to play a role in a Drosophila model of Amyotrophic lateral sclerosis (ALS), a debilitating and lethal neurodegenerative disorder that can be caused by aggregogenic mutations in genes encoding RNA binding proteins, including TDP-43 and FUS, a member of the FET family of proteins. TDP-43 and FUS also form cytoplasmic, ubiquitinated aggregates, in several other neurodegenerative disorders. Drosophila cabeza (caz) is the single ortholog of the human FET proteins. Xrp1 is upregulated in caz mutants, and the pupal lethality, motor defects and dysregulated gene expression of caz mutants is rescued by xrp1 heterozygosity [51]. Therefore, it is possible that the feed-forward loop that this study has uncovered is also active in this context: formation of cytoplasmic proteotoxic aggregates could stimulate xrp1 expression, which could then induce further proteotoxic stress in a feed forward loop, resulting in neuronal toxicity. Understanding the relationship between Xrp1, proteotoxic stress and oxidative stress may thus be beneficial for the study of human proteinopathies (Langton, 2021).

Injury activates a dynamic cytoprotective network to confer stress resilience and drive repair

In healthy individuals, injured tissues rapidly repair themselves following damage. Within a healing skin wound, recruited inflammatory cells release a multitude of bacteriocidal factors, including reactive oxygen species (ROS), to eliminate invading pathogens. Paradoxically, while these highly reactive ROS confer resistance to infection, they are also toxic to host tissues and may ultimately delay repair. Repairing tissues have therefore evolved powerful cytoprotective 'resilience' machinery to protect against and tolerate this collateral damage. This study used in vivo time-lapse imaging and genetic manipulation in Drosophila to dissect the molecular and cellular mechanisms that drive tissue resilience to wound-induced stress. This study identified a dynamic, cross-regulatory network of stress-activated cytoprotective pathways, linking calcium, JNK, Nrf2, and Gadd45, that act to both 'shield' tissues from oxidative damage and promote efficient damage repair. Ectopic activation of these pathways confers stress protection to naive tissue, while their inhibition leads to marked delays in wound closure. Strikingly, the induction of cytoprotection is tightly linked to the pathways that initiate the inflammatory response, suggesting evolution of a fail-safe mechanism for tissue protection each time inflammation is triggered. A better understanding of these resilience mechanisms-their identities and precise spatiotemporal regulation-is of major clinical importance for development of therapeutic interventions for all pathologies linked to oxidative stress, including debilitating chronic non-healing wounds (Weavers, 2019).

Reactive oxygen species (ROS) are universal injury-induced signals, produced by NADPH oxidases as an immediate response to tissue damage. At low levels, ROS can function as attractants for the recruitment of innate immune cells and to promote efficient wound angiogenesis; however, incoming inflammatory cells generate additional ROS in a 'respiratory burst' to eliminate invading pathogens and confer resistance to infection. Although this bacteriocidal response is clearly beneficial, excessive ROS levels can cause substantial bystander damage to host tissue; indeed, excessive oxidative stress is thought to be a key player in the pathogenesis of chronic non-healing wounds of patients in the clinic (Weavers, 2019).

To counter inflammatory stress, host tissues must employ powerful cytoprotective machinery to limit the 'collateral' damage and prevent immunopathology. Mammalian wound studies have identified a number of signaling pathways that may promote protection against oxidative stress, but such investigations have been complicated by the intricacy of the protection machinery and relative genetic intractability of vertebrate models. Nevertheless, a better understanding of these protective mechanisms will be crucial to enable the development of improved therapeutic interventions for a wide range of oxidative stress-related diseases, including chronic non-healing wounds. Also in the context of wound repair, therapeutic activation of cytoprotective pathways in the clinic could also offer an exciting approach to 'precondition' patient tissues prior to elective surgery (Weavers, 2019).

This study has characterized the temporal and spatial dynamics of the stress 'resilience' mechanisms that are induced downstream of wounding and dissect the underlying molecular and cellular mechanisms driving tissue protection. A complex cross-regulatory network of cytoprotective pathways were identified, involving calcium, JNK, Nrf2, and Gadd45, which collectively 'shield' tissues from ROS-induced damage and promote efficient damage repair. RNAi-mediated inhibition of either Nrf2 or Gadd45 delays wound repair, which is further exacerbated if both pathways are inhibited. Interestingly, it was found that these cytoprotective pathways are activated downstream of the same calcium signaling pathway that initiates the inflammatory response, suggesting the existence of a 'fail-safe' mechanism for cytoprotection whenever inflammation is triggered. Finally, ectopic activation of these protective pathways can confer stress resilience to naive unwounded tissue, and in the case of Gadd45, can even accelerate the rate of wound repair. Prolonged activation of Nrf2, however, caused marked delays in wound repair, suggesting that the optimal level of cytoprotection required for the most efficient tissue repair will be a finely tuned spatiotemporal balance of cytoprotective signaling (Weavers, 2019).

Until now, research on cytoprotective factors in wound repair has mainly focused on how antioxidant systems directly minimize ROS-induced damage following injury. However, tissues will undoubtedly have evolved a diverse range of 'resilience' mechanisms acting on different cellular targets and working in a highly coordinated manner to collectively reduce damage. This study shows that injury activates a cytoprotective signaling network that targets multiple different components to protect the repairing epithelial tissue, including both the upregulation of antioxidant defense machinery and DNA repair mechanisms. In this way, tissue resilience mechanisms can both shield the tissue from damage by directly dampening ROS levels and enhance DNA repair mechanisms (thus making wounded tissues more tolerant to any DNA damage caused by residual ROS). The presence of multiple, partially redundant protective mechanisms ensures effective resilience and thus minimizes delays in tissue repair; indeed, this study found that simultaneous knockdown of Nrf2 and Gadd45 exaggerates wound repair defects compared to individual knockouts alone (Weavers, 2019).

Since both Nrf2 and Gadd45α are upregulated within mammalian skin wounds, similar networks of wound-induced resilience mechanisms are likely to be well conserved from flies to man. Drosophila, with its advanced genetic tractability, capacity for live-imaging, and opportunity for large-scale genetic screening, thus offers an exciting new model for dissecting the mechanisms governing tissue resilience to stress, particularly those during wound repair. These studies may also have important implications for cancer therapy, as cancer cells could hijack this resilience machinery to protect the tumor from host immune attack, as well as confer resistance to clinical therapies such as chemo- or radio-therapy. Indeed, it is known that Gadd45α deficiency sensitizes epithelial cancer cells to ionizing radiation in vivo, implicating cytoprotective genes such as Gadd45a as potential drug targets in management of cancer radiotherapy treatments (Weavers, 2019).

For nearly 30 years, experimental biologists and clinicians have observed the remarkable but mysterious phenomenon of 'preconditioning,' whereby a brief period of sub-lethal tissue damage triggers adaptive mechanisms that confer subsequent cytoprotection against further insult, either within the same tissue or more remotely. Indeed, recent work in zebrafish has shown that superficial insult (via thoracotomy) preconditions adjacent cardiac tissue and renders it more resilient to subsequent cryoinjury (modeling an infarct) by upregulation of cardioprotective factors. Remarkably, activation of cardioprotective signaling by injection of exogenous ciliary neurotrophic factor just prior to ventricular cryoinjury had beneficial regenerative effects and rendered the heart more resilient to injury. In this regard, therapeutic activation of some or all of these resilience pathways could offer exciting 'pre-conditioning' strategies in the clinic to protect patient tissues during surgery or following organ transplant (Weavers, 2019).

A better understanding of resilience pathways and their long-term effects (including an analysis of 'cost') is clearly crucial for their full application in a clinical setting, given that excessive and long-term activation of resilience machinery could potentially have adverse effects. Indeed, while this study found that ectopic expression of Gadd45 prior to wounding could accelerate wound repair, long-term overexpression of dNrf2 within the epithelium caused marked delays in wound closure. Previous work suggests that prolonged Nrf2 activation may make cells less 'competitive' than their neighbors and can also induce certain skin defects (such as hyperkeratosis) and fibroblast senescence. Given the role for wound-induced ROS in inflammatory cell recruitment and angiogenesis, it is envisioned that achieving an optimal transient and balanced activation of this endogenous resilience machinery will be the key to unlocking its enormous therapeutic benefits, conferring valuable stress resilience without reaching levels that might otherwise be detrimental to repair or later tissue health (Weavers, 2019).

Increasing autophagy and blocking Nrf2 suppress laminopathy-induced age-dependent cardiac dysfunction and shortened lifespan

Mutations in the human LMNA gene cause a collection of diseases known as laminopathies. These include myocardial diseases that exhibit age-dependent penetrance of dysrhythmias and heart failure. The LMNA gene encodes A-type lamins, intermediate filaments that support nuclear structure and organize the genome. Mechanisms by which mutant lamins cause age-dependent heart defects are not well understood. This study modeled human disease-causing mutations in the Drosophila Lamin C gene and expressed mutant Lamin C exclusively in the heart. This resulted in progressive cardiac dysfunction, loss of adipose tissue homeostasis, and a shortened adult lifespan. Within cardiac cells, mutant Lamin C aggregated in the cytoplasm, the CncC(Nrf2)/Keap1 redox sensing pathway was activated, mitochondria exhibited abnormal morphology, and the autophagy cargo receptor Ref2(P)/p62 was upregulated. Simultaneous over-expression of the autophagy kinase Atg1 gene and an RNAi against CncC eliminated the cytoplasmic protein aggregates, restored cardiac function, and lengthened lifespan. These data suggest that simultaneously increasing rates of autophagy and blocking the Nrf2/Keap1 pathway are a potential therapeutic strategy for cardiac laminopathies (Bhide, 2018).

Mutations in the human LMNA gene are associated with a collection of diseases called laminopathies in which the most common manifestation is progressive cardiac disease. This study has generated Drosophila melanogaster models of age-dependent cardiac dysfunction. In these models, mutations synonymous with those causing disease in humans were introduced into Drosophila LamC. Cardiac-specific expression of mutant LamC resulted in (1) cardiac contractility, conduction, and physiological defects, (2) abnormal nuclear envelope morphology, (3) cytoplasmic LamC aggregation, (4) nuclear enrichment of the redox transcriptional regulator CncC (mammalian Nrf2), (5) and upregulation of autophagy cargo receptor Ref(2)P (mammalian p62). These cardiac defects were enhanced with age and accompanied by increased adipose tissue in the adult fat bodies and a shortened lifespan (Bhide, 2018).

To understand the mechanistic basis of cardiolaminopathy and identify genetic suppressors, advantage was taken of powerful genetic tools available in Drosophila. The presence of cytoplasmic LamC aggregates prompted a determination of whether increasing autophagy would suppress the cardiac defects. Cardiac-specific upregulation of autophagy (Atg1 OE) suppressed G489V-induced cardiac defects. Consistent with this, decreased autophagy due to expression of Atg1 DN resulted in enhanced deterioration of G489V-induced cardiac dysfunction. Interestingly, cardiac-specific Atg5 OE and Atg8a OE, two factors that also promote autophagy, showed little to no suppression of G489V-induced heart dysfunction, suggesting that Atg1 might be rate limiting in this context. These findings are consistent with studies in mouse laminopathy models in which rapamycin and temsirolimus had beneficial effects on heart and skeletal muscle through inhibition of AKT/mTOR signaling. These findings are depicted in a model (see Model for the interactions between the autophagy and CncC/Keap1 signaling pathway in mutant lamin-induced cardiac disease) in which cytoplasmic aggregation of mutant LamC results in upregulation of p62, which in turn inhibits autophagy via activation of TOR and inactivation of AMPK. AMPK inactivation leads to the activation of PI3K/Akt/mTOR pathway and inhibition of autophagy Atg1 OE promoted clearance of the LamC aggregates and restored proteostasis in these Drosophila models. Thus, the data suggest that mutant LamC reduces autophagy, resulting in impairment of cellular proteostasis that leads to cardiac dysfunction (Bhide, 2018).

Cardiac-specific expression of mutant LamC altered CncC subcellular localization. Previously, Drosophila larval body wall muscles expressing G489V were shown to experience reductive stress, an atypical redox state characterized by high levels of reduced glutathione and NADPH, and upregulation CncC target genes (Dialynas, 2015). Cardiac-specific CncC RNAi in the wild-type LamC background did not produce major cardiac defects. Consistent with this, Nrf2 deficiency in mice does not compromise cardiac and skeletal muscle performance. Cardiac-specific CncC RNAi suppressed G489V-induced cardiac dysfunction and reduced cytoplasmic LamC aggregation, but not R205W-induced defects. However, cardiac-specific RNAi against CncC did not affect G489V-induced adipose tissue accumulation and lifespan shortening. Similar to the nuclear enrichment of CncC in hearts expressing G489V, human muscle biopsy tissue from an individual with a point mutation in the LMNA gene that results in G449V (analogous to Drosophila G489V) showed nuclear enrichment of Nrf2 (Dialynas, 2015). Disruption of Nrf2/Keap1 signaling has also been reported for Hutchinson-Gilford progeria, an early-onset aging disease caused by mutations in LMNA. In this case, however, the thickened nuclear lamina traps Nrf2 at the nuclear envelope that results in a failure to activate Nrf2 target genes, leading to oxidative stress. In these studies, CncC nuclear enrichment was observed; however, a redox imbalance was not readily observed at the three-time points investigated. This might indicate that there is a window of time in disease progression in which redox imbalance occurs and that mechanisms are in place to re-establish homeostasis (Bhide, 2018).

It has been postulated that there is cross-talk between autophagy and Nrf2/Keap1 signaling. This was tested by manipulating autophagy and CncC (Nrf2) alone and in combination. CncC RNAi suppressed the cardiac defects caused by G489V, but not the lipid accumulation and lifespan shortening, suggesting the latter two phenotypes are not specifically due to loss of cardiac function. In contrast, Atg1 OE suppressed the cardiac and adipose tissue defects and lengthened the lifespan. The double treatment (simultaneous Atg1 OE and RNAi knockdown of CncC) gave the most robust suppression of the mutant phenotypes and completely restored the lifespan. Interestingly, Atg1 DN and RNAi knockdown of CncC simultaneously did not further deteriorate or improve the mutant phenotypes. Taken together, these data suggest that autophagy plays a key role in suppression of the G498V-induced phenotypes and that knockdown on CncC enhances this suppression (Bhide, 2018).

These findings support a model whereby autophagy and Nrf2 signaling are central to cardiac health. It is proposed that cytoplasmic aggregation of LamC increases levels of Ref(2)P (p62), which competitively binds to Keap1, resulting in CncC (Nrf2) translocation to the nucleus. Inside the nucleus, Nrf2 regulates genes involved in detoxification. Continued expression of antioxidant genes results in the disruption of redox homeostasis, defective mitochondria, and dysregulation of energy homeostasis/energy sensor such as AMPK and its downstream targets. Simultaneously, upregulation of Ref(2)P (p62) causes inhibition of autophagy via activation of TOR, which leads to the inactivation of AMPK. AMPK inactivation in combination with activation of the TOR pathway causes cellular and metabolic stress that leads to cardiomyopathy. In support of this model, transcriptomics data from muscle tissue of an individual with muscular dystrophy expressing Lamin A/C G449V (analogous to Drosophila G489V) showed (1) upregulation of transcripts from Nrf2 target genes, (2) upregulation of genes encoding subunits of the mTOR complex, and (3) downregulation of AMPK, further demonstrating relevance of the Drosophila model for providing insights on human pathology (Bhide, 2018).

Second order regulator Collier directly controls intercalary-specific segment polarity gene expression

In Drosophila, trunk metamerization is established by a cascade of segmentation gene activities: the gap genes, the pair rule genes, and the segment polarity genes. In the anterior head, metamerization requires also gap-like genes and segment polarity genes. However, because the pair rule genes are not active in this part of the embryo, the question of which gene activities fulfill the role of the second order regulators still remains to be solved. This study provides first molecular evidence that the Helix-Loop-Helix-COE transcription factor Collier fulfills this role by directly activating the expression of the segment polarity gene hedgehog in the posterior part of the intercalary segment. Collier thereby occupies a newly identified binding site within an intercalary-specific cis-regulatory element. Moreover, a direct physical association has been identified between Collier and the basic-leucine-zipper transcription factor Cap'n'collar B, which seems to restrict the activating input of Collier to the posterior part of the intercalary segment and to lead to the attenuation of hedgehog expression in the intercalary lobes at later stages (Ntini, 2011b).

In the context of an analysis to identify cis-regulatory elements controlling expression of segment polarity genes in the embryonic head, an intercalary-specific cis-regulatory element of hh—ic-CRE—was isolated within the upstream 6.43 kb region (Ntini, 2011a). The ~ 1 kb enhancer fragment (− 4085 to − 3077 bp) mediates reporter expression in the hh expressing cells of the posterior part of the intercalary segment, when combined with the endogenous hh promoter (− 120 to + 99 bp;). Further functional dissection of this element showed that the 450 bp ?1mF5 subfragment (− 3914 to − 3465 bp) mediates the intercalary-specific expression with slightly delayed onset, while the 335 bp F5_R4 subfragment (− 3799 to − 3465 bp) constitutes the minimum sequence required for the intercalary expression, but mediates an additional spotty metameric pattern in the trunk (Ntini, 2011). Because a high degree of phylogenetic conservation in non-coding DNA sequence implicates a functional role in vivo, such as recognition and DNA-binding by sequence-specific transcription factors, the sequence of the ic-CRE was subjected to phylogenetic conservation analysis within the genome of twelve Drosophila species, and different in silico analyses were performed to detect putative transcription factor binding sites. The minimum 335 bp ic-CRE consists of six highly conserved sequence blocks. A series of complete block deletions designed in the context of the minimum ic-CRE in combination with the endogenous hh promoter resulted in non-functional elements. This could be either because individual binding motifs were disrupted or inter-motif distances crucial for transcription factor binding and operation were disturbed. A point mutagenesis screen was conducted in the context of the 450 bp ic-CRE to extract crucial cis-regulatory information in respect to the conserved in silico identified transcription factor binding sites (Ntini, 2011b).

The ic-CRE responds to the homeotic transformation of the mandibular into an intercalary segment resulting from ectopic ems expression by a duplication of its expression pattern. However, despite this and the fact that the Hox gene labial is active in the intercalary segment, disrupting the homeodomain binding sites in conserved sequence blocks III or IV by point mutations did not abolish the ic-CRE mediated reporter expression. In contrast, disrupting a putative binding site for the fork head transcription factor Sloppy paired 1 (Slp1) in block IV eliminated the ic-CRE-mediated reporter expression. This is consistent with the reduced reporter expression in an RNAi-mediated knock-down of slp1, which is a proposed head gap-like and pair rule segmentation gene (Ntini, 2011b).

Another in silico prediction was found in conserved block II at position − 3771 to − 3755 bp that scores the binding matrix of the mammalian COE factor Olf1. Disrupting this site by point-mutation resulted in the complete abolishment of the ic-CRE mediated reporter expression, indicating that the site is absolutely required for the function of the 450 bp ic-CRE. Olf1 is the mammalian COE homolog of Collier and the endogenous hh expression in the intercalary segment is abolished in a col loss-of-function mutant (col¹. Likewise, the ic-CRE-mediated expression pattern is abolished in col¹ or col knock-down. In addition, the DNA-binding domain of Collier displays a high degree of primary sequence identity (86%) to the mammalian homolog. High degree of primary sequence identity in the DNA-binding domain, shared among the members of the COE family allows for a similar DNA-binding specificity: both Collier and the Xenopus homologs recognize the mammalian DNA target sequences in vitro. Therefore, the Olf1 prediction identified in silico within the ic-CRE is regarded as a putative Collier binding site and referred to as a Collier recognition site (Ntini, 2011b).

Apart from this functionally required Collier recognition site at − 3773 to − 3751 bp, scanning in silico the 6.43 kb upstream hh enhancer using MatInspector with a similarity cut-off of 1, 0.8 (core, matrix) identifies one more Olf1 prediction within the ic-CRE at position − 3967 to − 3945 bp. The 6.43 kb upstream enhancer of hh was also submitted to rVISTA using the nucleotide positions 3–19 of the binding matrix of Olf1. When setting the highest possible similarity cut-off 0.95, 0.85 (core, matrix), so that at least one prediction is generated, then only the functionally required Collier recognition site CAATTCCCCAATGGCAT (at − 3771 to − 3755) within the ic-CRE is detected. Lowering the matrix similarity threshold by 0.05, using cut-off 0.95, 0.8, generates three additional predictions. These are two distant sites, GAGACACTTGGGATGAG at − 3963 to − 3947 and CACACCACGGGGAAGCG at − 2872 to − 2856, and one promoter-proximal site CACTTCCCTTGCGCATA at − 212 to − 196. The first distant site is within the ic-CRE, 190 bp upstream of the functionally required Collier recognition site, and is also predicted by the MatInspector. Interestingly, in contrast to the functionally required Collier recognition site within the ic-CRE, none of the other predicted sites are phylogenetically conserved among the twelve Drosophila species. Considering the displayed short-range homotypic clustering (within 200 bp), it is, however, possible that the weaker predictions may contribute to the transcriptional outcome of the ic-CRE, even though they might be recognized with minor affinity by Collier in vivo (Ntini, 2011b).

In order to verify that the in silico identified and functionally required Collier recognition site within the ic-CRE is indeed occupied by Collier in vivo, chromatin immunoprecipitations (ChIP) from Drosophila embryonic nuclear extracts were performed with an antibody against Collier. In the anti-Col ChIPs, the functionally required Collier binding site within the ic-CRE was specifically enriched in comparison to mock ChIPs, which indicates that the site is indeed occupied by Collier in vivo (Ntini, 2011b).

In the case of the mammalian COE homolog of Collier, it was previously deciphered that the mouse transcription factor EBF contains two distinct and functionally independent transcription activation domains, the second one within the C-terminal region. Although Drosophila Collier has been genetically implicated as an activator of downstream segment polarity gene expression, its transcriptional activation potential had not yet been analyzed. In Drosophila two Collier isoforms are expressed from the col gene locus. The cDNAs encoding Collier A (also termed Col2) and Collier B (Col1) differ from each other by 465 bp due to alternative splicing. The two protein isoforms share the same first 528 N-terminal amino acids and differ in the C-terminal 29 amino acids for Collier A and 47 amino acids for Collier B. No specific expression pattern of collier A could be detected by double in situ hybridization using an RNA probe specific for collier B and a probe that hybridizes with both transcripts (Ntini, 2011b).

Therefore the transcriptional activation potential of each of the two Collier isoforms was examined by reporter assays in Drosophila S2 R+ cell transfections. In the reporter construct the functionally required and in vivo occupied Collier site was cloned in a single copy upstream of the endogenous hh promoter (− 120 to + 99 bp) driving luciferase gene expression. Both Collier isoforms activate luciferase expression when independently co-transfected with the reporter construct, indicating that both isoforms possess transcriptional activation potential. A truncated form of ColA lacking the last 23 C-terminal amino acids (ColA 1–534) displays a significantly reduced activation potential (~ 84% decrease), which indicates that a transcriptional activation domain must reside within either C-terminal region of both isoforms. Disrupting the Collier recognition site by point mutations decreased the mediated reporter activation by ~ 48% in the case of Collier A and ~ 44% in the case of Collier B. Taking into consideration that disrupting the Collier binding site in the context of the ic-CRE resulted in a complete abolishment of the mediated reporter expression in vivo, and that the same mutation does not support Collier DNA-binding in vitro, it is assumed that part of the reporter activation assessed in cell transfection may be achieved by Collier transactivating via unknown system-provided DNA-binding activities on the regulatory sequences of the reporter plasmid. Moreover, Collier carries a perfect SUMOylation motif within the N-terminus, predicted with the highest threshold value. The protein sequence TSLKEEP at amino acid position 44-50 matches the SUMOylation motif. Additional members of the COE transcription factor family contain also a SUMOylation motif at this conserved position. Apart from antagonizing ubiquitin-mediated degradation, sumoylation has been implicated in modifying transcriptional activation/repression potential of transcription factors. Mutant versions of Collier A and Collier B where the K within the SUMOylation motif is mutated towards R (ColA RK and ColB RK) display reduced activation potential, implying a possible role for sumoylation in regulation of Collier transcriptional activity (Ntini, 2011b).

Data is presented consistent with the cap-n-collar isoform CncB performing as a sequestering factor or inhibitor of Collier DNA-binding to its cognate site found within the ic-CRE. Furthermore, fluorescent immunostaining revealed that only a small fraction of the expressed Collier protein is nuclear localized in vivo. Conversely, CncB protein greatly accumulates in the nuclei. Prediction of nuclear localization signals (NLS) in silico generates no results for Collier, while CncB contains an NLS within the bZIP domain (aa 617–680). Interestingly, Collier carries a perfect SUMOylation motif in the very N-terminus, predicted with the highest threshold value. Apart from antagonizing ubiquitin-mediated degradation and modifying transcriptional activation/repression potential of transcription factors, sumoylation has also been implicated in protein nucleo-cytoplasmic translocation. Alternatively, in the absence of a nuclear localization signal, Collier import in the nucleus may be realized by heterodimerization with a protein that carries an NLS. This would increase the probability that Collier is recruited into combinatorial control mechanisms, which has already been implicated in muscle specification. Furthermore, nuclear accumulation of CncB, in converse to a relatively low concentration of nuclear Collier protein, indicated by the fluorescent immunostainings, may facilitate the sequestering function of CncB to antagonize and overcome the DNA-binding activity of Collier on the ic-CRE in the cells of the anterior most part of the mandibular segment during the establishment of procephalic hh expression, and at later stages in the hh expressing cells of the intercalary lobes (Ntini, 2011b).

In this respect it is interesting to note that despite the intrinsic transcriptional activation properties of the Cnc homologs, CncB acts to suppress both the expression and the homeotic selector (maxillary structures promoting) function of Deformed (Dfd) in the mandibular segment. In particular, CncB represses the maintenance phase of Dfd transcription in the mandibular cells, most probably by interfering with the positive regulatory function of Deformed within the Dfd autoactivation circuit. Overexpression of CncB partially represses Dfd-responsive transcriptional target elements in vivo. Interestingly, interaction between CncB and Dfd proteins has been reported. Perhaps the negative regulation of Dfd expression and function caused by CncB results from CncB interfering with Dfd binding to its cognate target cis-regulatory elements in vivo, as a consequence of a direct physical interaction at protein level with a sequestering effect similar to the interaction with Collier reported in this study (Ntini, 2011b).

The isolation of an intercalary-specific cis-regulatory element from the hh upstream region supports a unique mode for anterior head segment-specific transcriptional control of segment polarity gene expression. Thus, not only cross-regulatory interactions among segment polarity genes during the maintenance phase, but also the initial establishment of procephalic segment polarity gene expression seems to be unique for each of the anterior head segments. The previously proposed mode of second order regulation in anterior head patterning, resulting in activation of hh in the posterior part of the intercalary segment, is mediated by the HLH-COE factor Collier evidently via direct DNA binding. The reported physical interaction between Collier and CncB is likely to attenuate the activating function of Collier in the hh expressing cells of the posterior part of the intercalary segment at a later developmental stage, and it might also be involved in eliminating the potential of target activation by Collier in the anterior most part of the mandibular segment where the two factors are co-expressed (Ntini, 2011b).

Myc-driven overgrowth requires unfolded protein response-mediated induction of autophagy and antioxidant responses in Drosophila melanogaster

Autophagy, a lysosomal self-degradation and recycling pathway, plays dual roles in tumorigenesis. Autophagy deficiency predisposes to cancer, at least in part, through accumulation of the selective autophagy cargo p62, leading to activation of antioxidant responses and tumor formation. While cell growth and autophagy are inversely regulated in most cells, elevated levels of autophagy are observed in many established tumors, presumably mediating survival of cancer cells. Still, the relationship of autophagy and oncogenic signaling is poorly characterized. This study shows that the evolutionarily conserved transcription factor Myc (dm), a proto-oncogene involved in cell growth and proliferation, is also a physiological regulator of autophagy in Drosophila melanogaster. Loss of Myc activity in null mutants or in somatic clones of cells inhibits autophagy. Forced expression of Myc results in cell-autonomous increases in cell growth, autophagy induction, and p62 (Ref2P)-mediated activation of Nrf2 (cnc), a transcription factor promoting antioxidant responses. Mechanistically, Myc overexpression increases unfolded protein response (UPR), which leads to PERK-dependent autophagy induction and may be responsible for p62 accumulation. Genetic or pharmacological inhibition of UPR, autophagy or p62/Nrf2 signaling prevents Myc-induced overgrowth, while these pathways are dispensable for proper growth of control cells. In addition, the autophagy and antioxidant pathways are required in parallel for excess cell growth driven by Myc. Deregulated expression of Myc drives tumor progression in most human cancers, and UPR and autophagy have been implicated in the survival of Myc-dependent cancer cells. These data obtained in a complete animal show that UPR, autophagy and p62/Nrf2 signaling are required for Myc-dependent cell growth. These novel results give additional support for finding future approaches to specifically inhibit the growth of cancer cells addicted to oncogenic Myc (Nagy, 2013).

Earlier genetic studies have established that Myc is required for proper expression of hundreds of housekeeping genes and is therefore essential for cell growth and proliferation. Myc is a typical example of a nuclear oncogene: a transcription factor that drives tumor progression if its expression is deregulated in mammalian cells. Its mechanisms of promoting cell growth are likely different in many ways from that of cytoplasmic oncogenes such as kinases encoded by PI3K and AKT genes, also frequently activated in various cancers. Overexpression of these drives cell growth in Drosophila as well, but Myc also increases the nuclear:cytoplasmic ratio in hypertrophic cells, unlike activation of PI3K/AKT signaling. PI3K and AKT suppress basal and starvation-induced autophagy, while their inactivation strongly upregulates this process. In contrast, this study shows that both basal and starvation-induced autophagy requires Myc, and that overexpression of Myc increases UPR, leading to PERK-dependent induction of autophagy, and presumably to accumulation of cytoplasmic p62 that activates antioxidant responses. Autophagy deficiency predisposes to cancer at least in part through accumulation of the selective autophagy cargo p62, resulting in activation of antioxidant responses and tumor formation. These analyses show that both of these cytoprotective pathways can be activated simultaneously, and are required in parallel to sustain Myc-induced overgrowth in Drosophila cells (Nagy, 2013).

Autophagy and antioxidant responses have been considered to act as tumor suppressor pathways in normal cells and during early stages of tumorigenesis, while activation of these processes may also confer advantages for cancer cells. Lack of proper vasculature in solid tumors causes hypoxia and nutrient limitation. These stresses in the tumor microenvironment have been suggested to elevate UPR and autophagy to promote survival of cancer cells. This study demonstrates that genetic alterations similar to those observed in cancer cells (that is, deregulated expression of Myc) can also activate the UPR, autophagy and antioxidant pathways in a cell-autonomous manner in Drosophila. These processes are likely also activated as a consequence of deregulated Myc expression in human cancer cells based on a number of recent reports, similar to the findings in Drosophila presented in this study. First, chloroquine treatment that impairs all lysosomal degradation pathways is sufficient to reduce tumor volume in Myc-dependent lymphoma models. Second, ER stress and autophagy induced by transient Myc expression increase survival of cultured cells, and PERK-dependent autophagy is necessary for tumor formation in a mouse model. Data suggest that UPR-mediated autophagy and antioxidant responses may also be necessary to sustain the increased cellular growth rate driven by deregulated expression of Myc (Nagy, 2013).

Myc has proven difficult to target by drugs. Myc-driven cancer cell growth could also be selectively prevented by blocking cellular processes that are required in cancer cells but dispensable in normal cells, known as the largely unexplored non-oncogene addiction pathways. Previous genetic studies establish that autophagy is dispensable for the growth and development of mice, although knockout animals die soon after birth due to neonatal starvation after cessation of placental nutrition. Tissue-specific Atg knockout mice survive and the animals are viable, with potential adverse effects only observed in aging animals. Genetic deficiencies linked to p62 are also implicated in certain diseases, but knockout mice grow and develop normally and are viable. Similarly, Nrf2 knockout mice are viable and adults exhibit no gross abnormalities, while these animals are hypersensitive to oxidants. Mice lacking PERK also develop normally and are viable. All these knockout studies demonstrate that these genes are largely dispensable for normal growth and development of mice, and that progressive development of certain diseases is only observed later during the life of these mutant animals. There are currently no data regarding the effects of transient inhibition of these processes, with the exception of the non-specific lysosomal degradation inhibitor chloroquine, originally approved for the treatment of malaria, which is already used in the clinic for certain types of cancer (Nagy, 2013).

Based on these knockout mouse data, UPR, autophagy and antioxidant responses may be considered as potential non-oncogene addiction pathways: strictly required for Myc-dependent overgrowth (this study) and tumor formation, but dispensable for the growth and viability of normal cells, both in Drosophila and mammals. One can speculate that the transient inactivation of these pathways will have even more subtle effects than those observed in knockout mice, but this needs experimental testing. While it is difficult to extrapolate data obtained in Drosophila (or even mouse) studies to human patients, it is tempting to speculate that specific drugs targeting UPR, autophagy and antioxidant responses may prove effective against Myc-dependent human cancers, perhaps without causing adverse side-effects such as current, less specific therapeutic approaches. Notably, widely used anticancer chemotherapy treatments are known to greatly increase the risk that cancer survivors will develop secondary malignancies. Moreover, the autophagy and antioxidant pathways appear to be required in parallel during Myc-induced overgrowth in Drosophila cells. If a similar genetic relationship exists in Myc-dependent human cancer cells, then increased efficacy may be predicted for the combined block of key enzymes acting in these processes (Nagy, 2013).

Elucidation of the genetic alterations behind increased UPR, autophagy and antioxidant responses observed in many established human cancer cells may allow specific targeting of these pathways, and potentially have a tremendous benefit for personalized therapies. In addition to non-specific autophagy inhibitors such as chloroquine, new and more specific inhibitors of selected Atg proteins are being developed. Given the dual roles of autophagy during cancer initiation and progression, a major question is how to identify patients who would likely benefit from taking these drugs. For example, no single test can reliably estimate autophagy levels in clinical samples, as increases in autophagosome generation or decreases in autophagosome maturation and autolysosome breakdown both result in accumulation of autophagic structures. Based on this study's data and recent mammalian reports, elevated Myc levels may even turn out to be useful as a biomarker before therapeutic application of inhibitors for key autophagy, UPR or antioxidant proteins in cancer patients (Nagy, 2013).

Three EMS-induced mutant alleles of cnc (cnc^2E16, cnc^C7 and cnc^C14) in a screen for mutations that interact with the Hox gene Dfd (Harding, 1995). Embryos homozygous for these EMS-induced alleles have ectopic duplications of maxillary mouth hooks and cirri, but retain normal labral structures and some normal mandibular structures, e.g. the lateralgräten and median tooth. This contrasts with the phenotype of deletion mutants of cnc, which lack all mandibular and labral derivatives. The difference between the phenotypes of the EMS-induced alleles when compared to the deletion alleles prompted a consideration of the possibility that multiple functions are encoded in the cnc locus. Previous studies detected one transcript isoform at cnc, but the current molecular analyses of the locus indicates that three transcript and protein isoforms are produced from the cnc gene. A probe homologous to the region that encodes the b-ZIP region of cnc detects three different sizes of polyadenylated RNAs on embryonic northern blots. These will be referred to as the cncA, cncB and cncC transcripts. The 3.3 kb cncA transcript is present in 0-2 hour embryos, presumably from maternal stores and is also abundantly expressed in 12-24 hour embryos. The 5.4 kb cncB transcript is absent from 0-2 hour embryos, but present at all other embryonic stages. The 6.6 kb cncC transcript is present in 0-2 hour embryos, is barely detected in 2-12 hour embryos and is detected at relatively higher levels in 12-24 hour embryos. The sequence of all of the coding exons and exon/intron boundaries for all isoforms on the cnc^2E16 and cnc^C7 mutant chromosomes was determined in an attempt to find the molecular lesion responsible for the decreased amount of CncB protein in the mutant embryos. However, no nucleotide substitutions were detected when the coding and splice site sequences were compared with parental chromosome sequence. Though the location of the mutations that alter CncB protein expression are not yet known, they could plausibly reside in translational regulatory sequences for cncB (McGinnis, 1998).

To identify cDNAs corresponding to the cncA, cncB and cncC transcripts, 212 cDNA clones from libraries covering all stages of Drosophila embryonic development were isolated and characterized. The first class of cDNAs corresponds to the cncA transcript. This is the same class characterized by Mohler, 1991, and is distinguished by the incorporation of exon A1. Exons A2 and A3, which encode the CNC and b-ZIP domains, are present in cncA and the other two isoforms of cnc. A probe containing exon A1 sequences specifically hybridizes the 3.3 kb cncA transcript on northern blots. The cncA open reading frame begins with an ATG codon near the 5' end of exon A2 and is predicted to encode a 533 amino acid protein (Mohler, 1991). A second class of cDNAs from the locus corresponds to cncB transcripts. Such cDNAs lack sequences from exon A1, but contain five additional exons (B1-B5) spliced onto the 5' end of exon A2. Each of these additional exons is found upstream of exon A1. A probe containing the B1-B4 exons detects the 5.4 kb cncB transcript and the 6.6 kb cncC transcript on Northern blots. The total extent of the cncB transcription unit is approximately 17 kb. Since exon A2 sequences contain no stop codons upstream of the initiating ATG for the CncA codons, the open reading frame in cncB transcripts includes the entirety of the CncA protein, as well as an additional 272 codons from exons B3, B4, B5 and A2. The predicted 805 amino acid CncB protein thus is distinguished from CncA by a 272 amino acid region that includes His-Pro repeats, Ala-repeats, a Pro-repeat and Val-Gly repeats, but the region exhibits no extended sequence similarity to other proteins in database searches, other than the CNC/b-ZIP domain that it shares with CncA. The third class of cDNAs from the locus corresponds to cncC transcripts. These cDNAs have identical sequences as the cncB cDNAs, except that exon B1 is absent, and five additional exons (C1-C5) are spliced onto the 5' end of exon B2. Each of these five additional exons are found upstream of the B1 exon. A probe containing the C1-C4 exons detects the 6.6 kb cncC transcript on Northern blots. Since exon B2 and the 5' end of exon B3 contain no stop codons upstream of the initiating ATG for the CncB codons, the ATG-initiated open reading frame in cncC transcripts includes the entirety of the CncB protein, as well as an additional 491 codons that derive from the C3, C4, C5, B2 and B3 exons. The extent of the entire cncC transcription unit is approximately 39 kb. The 491 amino acid CncC-specific domain at the N terminus of the predicted 1296 residue CncC protein includes regions that are rich in Ser and Thr residues, other regions with abundant concentrations of Glu and Asp residues, but exhibits no extended sequence similarity to other proteins in database searches. Interestingly, the fuzzy onions gene, which encodes a testis protein required for mitochondrial fusion in Drosophila spermatids (Hales, 1997), is encoded in the sequence interval between the C5 and B1 exons (McGinnis, 1998).

Bases in 5' UTR - 94

Bases in 3' UTR - 1028

PROTEIN STRUCTURE

Amino Acids - 533

Structural Domains

The leucine zipper of CNC is longer than bZIP proteins and contains six heptad repeats. The function of the leucine zipper is considered as a protein interaction domain. The leucine zipper of CNC is divergent from the typical sequence (Mohler, 1991).

date revised: 18 February 2024

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.