InteractiveFly: GeneBrief

HIF prolyl hydroxylase: Biological Overview | References

BIOLOGICAL OVERVIEW

The Hypoxia Inducible Factor (HIF) mediates cellular adaptations to low oxygen. Prolyl-4-hydroxylases are oxygen sensors that hydroxylate the HIF alpha-subunit (Similar in Drosophila), promoting its proteasomal degradation in normoxia. Three HIF-prolyl hydroxylases, encoded by independent genes, PHD1, PHD2, and PHD3, occur in mammals. PHD2, the longest PHD isoform includes a MYND domain, whose biochemical function is unclear. PHD2 and PHD3 genes are induced in hypoxia to shut down HIF dependent transcription upon reoxygenation, while expression of PHD1 is oxygen-independent. The physiologic significance of the diversity of the PHD oxygen sensors is intriguing. This study has analyzed the Drosophila PHD locus, fatiga, which encodes 3 isoforms, FgaA, FgaB and FgaC that are originated through a combination of alternative initiation of transcription and alternative splicing. FgaA includes a MYND domain and is homologous to PHD2, while FgaB and FgaC are shorter isoforms most similar to PHD3. Through a combination of genetic experiments in vivo and molecular analyses in cell culture, it was shown that that fgaB but not fgaA is induced in hypoxia, in a Sima-dependent manner, through a HIF-Responsive Element localized in the first intron of fgaA. The regulatory capacity of FgaB is stronger than that of FgaA, as complete reversion of fga loss-of-function phenotypes is observed upon transgenic expression of the former, and only partial rescue occurs after expression of the latter. It is concluded that diversity of PHD isoforms is a conserved feature in evolution. As in mammals, there are hypoxia-inducible and non-inducible Drosophila PHDs, and a fly isoform including a MYND domain co-exists with isoforms lacking this domain. These results suggest that the isoform devoid of a MYND domain has stronger regulatory capacity than that including this domain (Acevedo, 2010).

In response to oxygen deprivation cells, tissues and whole organisms induce the expression of a wide range of genes that tend to restore energy homeostasis. Hypoxic gene induction is mainly mediated by the Hypoxia Inducible Factor (HIF), a heterodimeric α/β transcription factor composed of two basic-Helix-Loop-Helix-PAS (bHLH-PAS) subunits. Whereas the HIFβ subunit is constitutive, HIFα is tightly regulated by oxygen levels through various mechanisms that include protein stability, transcription coactivator recruitment and subcellular localization. The molecular mechanism that controls HIFα protein stability has been characterized in detail: In normoxia, HIFα is ubiquitinated and degraded at the 26S proteasome, while in hypoxia the protein is stabilized. HIFα ubiquitination in nomoxia is mediated by the Von Hippel Lindau (VHL) tumor suppressor factor which is the substrate recognition subunit of a multimeric E3 ubiquitin ligase complex. Physical interaction between VHL and HIFα requires hydroxylation of 2 key prolyl residues in the HIFα sequence (P402 and P564 in human HIF-1α), which is catalyzed by specific prolyl-4-hydroxylases, named PHD1- PHD2 and PHD3 (Epstein, 2001). These enzymes are members of the Fe (II) and 2-oxoglutarate dependent dioxygenase superfamily that utilizes O₂ as a co-substrate for catalysis (Jaakkola, 2001; Epstein, 2001). Under hypoxia, PHD hydroxylase activity is reduced, HIFα escapes hydroxylation and proteolysis, leading to HIF nuclear accumulation and transcriptional induction of target genes. HIF-dependent transcription involves direct binding to Hypoxia Response Elements (HREs) that are characterized by an invariant 5′CGTG 3′ core consensus. Interestingly, a negative feed back loop, limiting HIFα activity in chronic hypoxia or upon re-oxygenation has been reported: PHD2 and PHD3 mRNAs are induced by low oxygen in a HIF-dependent manner to shut-down HIF activity; PHD1 transcription is oxygen-independent (Epstein, 2001; Appelhoff, 2004; Acevedo, 2010 and references therein).

The occurrence of three mammalian PHD isoforms encoded by three independent genes (PHD 1, PHD2 and PHD3) has opened the question of how each of these enzymes contributes to HIF regulation. It has been shown that all three PHDs can hydroxylate HIFα in vitro, and that upon over-expression, they can all suppress HRE-reporter induction (Bruick, 2001; Epstein, 2001). Cell culture analysis revealed that, PHD2 has a dominant role in controlling HIF-1α in normoxia (Berra 2003), while PHD3 is important for regulating HIF in hypoxia or upon re-oxygenation (Appelhoff, 2004). Furthermore, in vivo studies showed that PHD2, but not PHD1 or PHD3 knockout mice, exhibit enhanced angiogenesis and erythropoiesis (Fong, 2008; Takeda, 2008), whereas PHD1 knockout mice display metabolic differences under ischemic conditions (Aragones, 2008; Acevedo, 2010 and references therein).

Previous work has led to the identification of a hypoxia response system in Drosophila that is homologous to mammalian HIF, in which the bHLH-PAS protein Similar (Sima), and the prolyl-4-hydroxylase Fatiga (Fga) are the homologues of HIFα and PHDs, respectively (Lavista-Llanos, 2002). sima null mutant individuals are unable to carry out transcriptional responses to hypoxia, although they are fully viable in normoxia. fga loss-of-function alleles showed different levels of Sima accumulation in normoxia, as well as tracheal defects and lethality at different developmental stages. Interestingly, sima loss-of-function mutations rescued viability and tracheal defects of fatiga mutants, suggesting that Sima protein over-accumulation accounts for these phenotypes (Centanin, 2005; Acevedo, 2010 and references therein).

This study describes a characterization of the single fatiga locus. The locus encodes three Fatiga variants, FgaA, FgaB and FgaC that originate from a combination of alternative transcription initiation and alternative mRNA splicing. FgaA includes a MYND domain, so it is homologous to PHD2, while both FgaB and FgaC are shorter isoforms that lack the MYND domain, and are similar to PHD3. Expression pattern of FgaA and FgaB were analyzed, as well as their transcriptional induction in hypoxia. Whereas FgaA expression remains constant and relatively low throughout the life cycle, FgaB is strongly upregulated in adult stages. FgaB but not FgaA is induced in hypoxia in a Sima dependent manner, both in cell culture and in vivo. Cell culture studies revealed that an HRE lying 759 to 756 base pairs upstream of the FgaB transcription initiation site accounts for FgaB induction in hypoxia. Finally, the ability of FgaA and FgaB to shut down Sima-dependent gene expression was explored; although the two isoforms are active, the regulatory capacity of FgaB is clearly stronger than that of FgaA (Acevedo, 2010).

Three PHD variants occur in mammals, and one single PHD gene, named EGL9, has been reported in Caenorabditis elegans. In Drosophila, previous studies on Fatiga, the Drosophila PHD homologous gene, have focused on its role in the regulation of Sima protein abundance (Lavista-Llanos, 2002; Centanin, 2005) and CyclinD-dependent cellular growth (Frei, 2004). In these functional studies, however, the occurrence of diverse Fga isoforms has not been addressed. This study has analyzed the fatiga locus, revealing that three different PHD isoforms occur in the fruit fly, which are generated through a combination of alternative splicing and alternative initiation of transcription. One of the isoforms, FgaA, includes a MYND domain, so it is homologous to mammalian PHD2, and the other two isoforms, FgaB and FgaC, lack a MYND domain, and are similar to PHD3. Thus, the diversity of PHD isoforms, including or not a MYND domain, seems to be an ancestral condition in evolution maintained in phylogenetically distant phyla such as insects and mammals. The occurrence of a single PHD isoform including a MYND domain in C. elegans might be due to evolutionary loss of shorter PHD variants (Acevedo, 2010).

In mammals PHD2 and PHD3, but not PHD1 mRNAs, are HIF-inducible (Epstein, 2001). This work has shown that FgaB, but not FgaA, is hypoxia-inducible, and that this induction depends on Drosophila HIF/Sima. A HIF Responsive Element (HRE) that mediates hypoxic transcriptional activation of fgaB mRNAs is localized at the position –759 to –756 with respect to the transcription initiation site of fgaB. Most HREs of hypoxia inducible genes of various organisms localize at their 5' regulatory region no more than 1 Kb upstream to the transcription initiation site. The identified HRE upstream to the fgaB open reading frame adjusts to this general rule. Due to the structure of the fga locus, the 5' regulatory region of fgaB lies in the large (8630 bp) first intron of fgaA (Acevedo, 2010).

Sequence conservation of the HRE lying upstream of Drosophila fgaB transcription initiation site and the mammalian PHD3 HRE -- localized in its first intron -- is remarkable, and extends beyond CGTG HRE invariant core. Fourteen out of 17 nucleotides around the fgaB HRE (CTGGGCTACGTGAGCAT) are conserved in the PHD3 regulatory region. This observation supports the notion that oxygen-dependent induction of PHD isoforms is important for adaptation of organisms to changing oxygen conditions (Acevedo, 2010).

The fact that a single Drosophila PHD locus encodes different isoforms that parallel two of the mammalian PHD variants encoded by independent genes is remarkable, and argues in favor that a combination of PHDs including or not a MYND domain is functionally relevant. The role of the MYND domain in HIF prolyl-4-hydroxylases is intriguing. Although PHD2 is the most abundant mammalian isoform and hence, has a dominant role in controlling HIFα in normoxia, PHD3 has been reported to have stronger intrinsic hydroxylation capacity than PHD2, which includes the MYND domain (Appelhoff, 2004; Berra, 2003). Consistent with this, the MYND domain has been proposed to mediate inhibition of PHD2 hydroxylase activity, as deletion of this domain led to increased activity of the enzyme (Choi, 2005). Supporting the notion of the MYND domain provoking reduction of PHD regulatory capacity, it has been shown that direct interaction of the peptidyl cis/trans isomerase FKBP38 with the MYND domain of PHD2 negatively regulates PHD2 protein stability (Barth, 2009). FKBP38 does not interact with the hydroxylase isoforms PHD1 or PHD3, which lack the MYND domain (Barth, 2007). Some reports, however, weigh in favor of a model of a MYND domain enhancing PHD negative regulation of HIF, as PHD2 but not PHD1 or PHD3 have the capacity to inhibit HIF transcriptional activity through a hydroxylation-independent mechanism (Ozer, 2005; To, 2005). Consistent with this, proteins including a MYND domain have been reported to mediate transcriptional inhibition of other transcription factors, so it is conceivable that transcription inhibitory capacity is a general feature of this domain. Thus, it is still unclear as whether the MYND domain increases or decreases the regulatory capacity of PHDs. The results in Drosophila support the latter possibility, as the PHD isoform that lacks the MYND domain has stronger regulatory capacity than the isoform that includes this domain. Detailed biochemical and functional studies are required to define the precise role of this protein domain in transcriptional responses to hypoxia (Acevedo, 2010).

A fat-tissue sensor couples growth to oxygen availability by remotely controlling insulin secretion

Organisms adapt their metabolism and growth to the availability of nutrients and oxygen, which are essential for development, yet the mechanisms by which this adaptation occurs are not fully understood. This study describes an RNAi-based body-size screen in Drosophila to identify such mechanisms. Among the strongest hits is the fibroblast growth factor receptor homolog breathless necessary for proper development of the tracheal airway system. Breathless deficiency results in tissue hypoxia, sensed primarily in this context by the fat tissue through HIF-1a prolyl hydroxylase (Hph). The fat relays its hypoxic status through release of one or more HIF-1a-dependent humoral factors that inhibit insulin secretion from the brain, thereby restricting systemic growth. Independently of HIF-1a, Hph is also required for nutrient-dependent Target-of-rapamycin (Tor) activation. These findings show that the fat tissue acts as the primary sensor of nutrient and oxygen levels, directing adaptation of organismal metabolism and growth to environmental conditions (Texada, 2019).

This report identifies one tissue in particular, the fat body, which senses internal oxygen levels and regulates growth rate accordingly. The data show that, as an adaptive response to oxygen limitation, the fat tissue releases into the circulation one or more factors that inhibit the secretion of insulin from the brain to reduce systemic growth. The ability of oxygen to reduce systemic body growth through downregulation of insulin signaling requires Hph-dependent HIF-1a/Sima activity in the fat tissue. Furthermore, hypoxia and AA deprivation both reduce Hph activity in the fat tissue, and this reduction leads to suppression of Tor signaling, independently of HIF-1a/Sima. This is consistent with a requirement of Hph for cell growth. In other contexts, Sima is known to regulate Tor-pathway activity via the protein Scylla/REDD1. However, this pathway does not appear to be responsible for the effects of hypoxia on Tor activity observed in this study, as sima mutation does not block hypoxia- or starvation-induced Tor suppression. Likewise, Tor suppression is not necessary for the systemic growth reduction induced by hypoxia. The data suggest that Hph is involved in AA sensing, in addition to its well-described role in oxygen sensing, and that HIF-1a is not involved in this process. Together, this suggests that AA and oxygen sensing converge through Hph in the fat body to modulate systemic growth in response to environmental conditions (Texada, 2019).

Many of the known Drosophila adipokines that affect insulin secretion from the IPCs are regulated by Tor-pathway activity in the fat body, including CCHa-2, Egr, FIT, GBP and GBP2, and Sun. The finding that AA availability regulates Hph activity, and that Hph activity modulates Tor signaling, thus places Hph upstream of these known factors, in addition to the separate Sima-dependent and Tor-independent humoral factor(s) that modulate insulin secretion under hypoxia. Several routes by which AA availability regulates Tor have been investigated and Hph may modulate some of these and not others, thereby allowing for different responses to AA starvation and hypoxia. Indeed, the current work shows that HIF-1a/Sima is required for the growth-suppressive effect of hypoxia, but not for growth responses to varied dietary AA input, although Hph is involved in both. The mechanisms by which Hph activity, which simultaneously requires AAs and oxygen, allows or promotes Tor signaling is an interesting topic to investigate in future studies, as is the identity of the Tor-independent humoral factor(s) downstream of Sima (Texada, 2019).

The results show that hypoxia or loss of fat-body Hph activity Sima dependently represses Dilp3 and Dilp5 transcription, while having little or no suppressive effect on Dilp2 expression. This suggests that specific transcriptional regulation of Dilp3 and Dilp5 is an important component of the response to hypoxia. Consistent with this observation, previous studies have shown that the transcription of Dilp2, -3, and -5 are independently regulated. Nutrient deprivation reduces expression of Dilp3 and -5, while having no effect on Dilp2 expression, similar to the effects of exposure to hypoxia. This is consistent with the finding that both AA deprivation and hypoxia suppress Hph activity, although the downstream pathways involved appear likely to be different, as at least some aspects of nutrient deprivation are relayed through the Tor pathway, whereas the hypoxia-specific signal(s) shown here is not. This observed transcriptional response could conceivably arise secondarily to DILP-release inhibition via autocrine feedback regulation that operates in both Drosophila and mammals. However, lower secretion of DILPs, which were observed under hypoxic conditions, generally feeds back to induce an increase in the expression of Dilp3 and -5 rather than the decrease that was observed. Therefore, the hypoxia-induced alterations of Dilp3 and -5 expression appear to be specific transcriptional responses rather than feedback effects. Thus, beyond the identity of the fat-body factor involved, the mechanisms operating in the IPCs by which it regulates insulin-like gene expression and peptide release will be important to study in future experiments (Texada, 2019).

In mammals, several adipokines regulate β-cell function and insulin secretion, including Leptin, which conveys information about fat storage and is a functional analog of the Drosophila fat-derived cytokine Unpaired. Interestingly, strong increases were observed in fat-body LD size induced by Tor inhibition downstream of AA starvation, hypoxia, or Hph mutation, indicating a change in lipid metabolism within the fat tissue. Although this phenotype is Tor-dependent and not upstream of the particular HIF-1a-dependent factor described above, it is possible that additional signals related to lipid metabolism may be released by the fat body in response to hypoxia or starvation, such as a lipid-binding protein or even a lipid per se. For example, the mammalian fatty acid-binding protein 4 is an insulin-modulating adipokine that is influenced by obesogenic conditions that lead to adipose tissue hypoxia, and orthologous proteins are encoded by the Drosophila genome (Texada, 2019).

Most organisms stop growing after reaching a genetically predetermined species-characteristic size. Although insight from genetic studies in Drosophila into the mechanisms that regulate body growth with regard to nutrition helps to explain how organisms modulate their growth rate according to nutritional conditions, a mechanism that allows organisms to assess their size and stop their growth when they have reached an optimum has remained elusive. However, recent evidence suggests that body size in insects may be determined by a mechanism that involves oxygen sensing, and oxygen availability is known to place limits on insect body size. According to this recent insight, the limited growth ability of the tracheal system during development may limit overall body size via downstream oxygen sensing. The size of the tracheal system is established at the beginning of each developmental stage and remains largely fixed, aside from terminal branching, as the body grows until it eventually reaches the limit of the system's ability to deliver oxygen. This allows the body to assess its size by sensing internal oxygen concentrations and to terminate growth at a characteristic size that is determined by the size of the tracheal system. An RNAi screen shows that the FGF receptor Btl, which is a key factor essential for tracheal growth during development, is a main determinant of body size. Indeed, btl was a stronger hit than known size-governing genes. The data therefore support the notion that the tracheal system and oxygen sensing may be part of a size-assessment mechanism (Texada, 2019).

Oxygen homeostasis also requires the coordination of growth between the tissues that consume oxygen and those that deliver it. The development of the oxygen delivery system is therefore oxygen sensitive in both mammals and Drosophila. In mammals, local tissue hypoxia promotes angiogenesis via induction of many pro-angiogenic factors, including FGF. In Drosophila, tissue hypoxia induces expression of the FGF-like ligand Bnl, leading to branching of the tracheal airway tubes toward oxygen-deficient areas. This study shows that this mechanism operates independently of insulin, as reduced insulin signaling in the trachea has no effect on overall body growth. This system therefore allows an adaptive response to low oxygen by reducing overall body growth via suppression of insulin signaling, while promoting hypoxia-induced FGF-dependent tracheal growth to increase oxygen delivery (Texada, 2019).

Cell and tissue hypoxia are also observed in human conditions of obesity and cancer. The insect fat body performs the functions of mammalian fat and liver tissues. Accordingly, perturbation of systemic insulin signaling by adipose and hepatic tissue hypoxia is also observed in mammalian systems. In mammals, obesity induces hypoxia within adipose tissue due to the rarefaction of vascularization of this tissue, leading to the release of inflammatory mediators and other adipokines that are associated with the pathophysiology of obesity-related metabolic disorders including diabetes. Although loss of normal β-cell activity is considered a main factor in diabetes, the mechanism by which tissue hypoxia affects insulin secretion is poorly understood. The finding of one or more hypoxia-induced fat-body-derived insulinostatic factors may lead to insights into the role of adipose-tissue hypoxia in obesity and its impact on diabetes. Furthermore, a link is shown between oxygen and AA availability in the adipose tissue through Hph-dependent regulation of the Tor pathway, linking these pathways in a common metabolic response to oxygen limitation and nutrient scarcity (Texada, 2019).

Obesity also causes physical and hormonal changes that affect breathing patterns, leading to apnea and thus intermittent episodes of systemic hypoxia. These hypoxic periods can induce changes in the liver, leading to fatty liver disease and dyslipidemia. The alterations to fat-body lipid metabolism observed in this study may thus be relevant to human health as well. Furthermore, hypoxia-induced programs play important roles in tumor formation. During cancer development, tumor cells undergo a metabolic reprogramming, the so-called Warburg effect, in which their metabolism shifts from oxidative phosphorylation to glycolysis, and activation of HIF-1a is believed to play a key role in this shift. As the hypoxia-sensing mechanism and the insulin-signaling system are conserved between flies and mammals, understanding the effects of hypoxia on the fat body could thus provide insight into many human disease states. It will be of interest to study whether tissue hypoxia also inhibits Tor-pathway activity in mammalian adipocytes (Texada, 2019).

In conclusion, this study unravels a mechanism that allows organisms to adapt their metabolism and growth to environments with low oxygen. Hypoxia activates a fat-tissue oxygen sensor that remotely controls the secretion of insulin from the brain by inter-organ communication. This involves the inhibition of Hph activity, leading to the activation of a HIF-1a-dependent genetic program within the fat tissue, which then secretes one or more humoral signals that alter insulin-gene expression and repress insulin secretion, thereby slowing growth. AA scarcity, like oxygen deficiency, is shown to inhibit Hph activity, and the activity of Hph, but not of HIF-1a, is required for Tor activity in the fat body. Thus, in addition to its role in regulating the as-yet unidentified fat-body hypoxia signal via HIF-1a, Hph connects both oxygen and AA levels to the Tor pathway through an unknown HIF-1a-independent mechanism. Given the conservation of oxygen-sensing and growth-regulatory systems, and the influence of oxygen on growth between Drosophila and mammals, a similar adaption response may operate in mammals via adipose tissue oxygen sensing to maintain homeostasis (Texada, 2019).

Dose-dependent modulation of HIF-1alpha/sima controls the rate of cell migration and invasion in Drosophila ovary border cells

The role of the hypoxic response during metastasis was analysed in migrating border cells of the Drosophila ovary. Acute exposure to 1% O₂ delayed or blocked border cell migration (BCM), whereas prolonged exposure resulted in the first documented accelerated BCM phenotype. Similarly, manipulating the expression levels of sima, the Drosophila hypoxia-inducible factor (HIF)-1alpha ortholog, revealed that Sima can either block or restore BCM in a dose-dependent manner. In contrast, over-expression of Vhl (Drosophila von Hippel-Lindau) generated a range of phenotypes, including blocked, delayed and accelerated BCM, whereas over-expression of hph (Drosophila HIF prolyl hydroxylase) only accelerated BCM. Mosaic clone analysis of sima or tango (HIF-1beta ortholog) mutants revealed that cells lacking Hif-1 transcriptional activity were preferentially detected in the leading cell position of the cluster, resulting in either a delay or acceleration of BCM. Moreover, in sima mutant cell clones, there was reduced expression of nuclear slow border cells (Slbo) and basolateral DE-cadherin, proteins essential for proper BCM. These results show that Sima levels define the rate of BCM in part through regulation of Slbo and DE-cadherin, and suggest that dynamic regulation of Hif-1 activity is necessary to maintain invasive potential of migrating epithelial cells (Doronkin, 2010).

Cell autonomy of HIF effects in Drosophila: tracheal cells sense hypoxia and induce terminal branch sprouting

Drosophila tracheal terminal branches are plastic and have the capacity to sprout out projections toward oxygen-starved areas, in a process analogous to mammalian angiogenesis. This response involves the upregulation of FGF/Branchless in hypoxic tissues, which binds its receptor Breathless on tracheal cells. This study show that extra sprouting depends on the Hypoxia-Inducible Factor (HIF)-α homolog Sima and on the HIF-prolyl hydroxylase Fatiga that operates as an oxygen sensor. In mild hypoxia, Sima accumulates in tracheal cells, where it induces breathless, and this induction is sufficient to provoke tracheal extra sprouting. In nontracheal cells, Sima contributes to branchless induction, whereas overexpression of Sima fails to attract terminal branch outgrowth, suggesting that HIF-independent components are also required for full induction of the ligand. It is proposed that the autonomous response to hypoxia that occurs in tracheal cells enhances tracheal sensitivity to increasing Branchless levels, and that this mechanism is a cardinal step in hypoxia-dependent tracheal sprouting (Centanin, 2008).

This study has analyzed the role of the Drosophila HIF-α homolog Sima and the oxygen-sensing prolyl-4-hydroxylase Fga in tracheal terminal branching. It is assumed that during embryonic stages, tracheal development depends on hard-wired developmental cues, and, later, in larval stages, tracheal terminal branching is driven by local hypoxia in the target tissues. The observations carried out in this study indicate that the tracheal system of sima mutant third-instar larvae is indistinguishable from that of wild-type individuals, including the pattern of terminal branches. Thus, the results imply that if terminal branching during normal development was mediated by tissue hypoxia, the mechanism involved in such a local response should be Sima independent. This is a remarkable difference between Drosophila tracheogenesis and the development of the mammalian vascular system, in which HIF proteins are critically required for both vasculogenesis and developmental angiogenesis (Centanin, 2008).

It was also shown that Sima does play a cardinal role in hypoxia-dependent tracheal terminal branch sprouting, as well as in the formation of terminal branches that compensate for poor oxygenation in exceptional situations in which a neighboring branch is missing. Sima-dependent extra sprouting is negatively regulated by the oxygen-sensing prolyl-4-hydroxylase Fga, since fga mutants displayed an extra sprouting phenotype that was even stronger than that observed in wild-type individuals exposed to hypoxia. This extra sprouting phenotype is the first demonstration that loss of function of a HIF-prolyl hydroxylase can provoke an angiogenic-like phenotype. Thus, it seems reasonable to expect that conditional knockdown of mammalian PHDs in an appropriate cell type will promote angiogenesis (Centanin, 2008).

The long-standing paradigm for mammalian angiogenesis is that low oxygen levels trigger HIF accumulation in target tissues, which, in turn, mediates VEGF induction that, upon binding to VEGF receptors on endothelial cells, attracts the outgrowth of newly formed blood capillaries. Nevertheless, this apparently passive role of endothelial cells has recently been challenged. It has been demonstrated that in endothelial cell-specific HIF-α knockout mice the angiogenic response is impaired, highlighting a central role of the oxygen-sensing machinery in endothelial cells (Centanin, 2008).

This study has shown that the specialized Drosophila tracheal cells that respond to hypoxia by projecting angiogenic-like subcellular processes -- i.e., the terminal branches -- are apparently more sensitive to hypoxia than any other cell type in the larva. The sensory threshold to induce Sima-driven gene activation in these cells is shifted to near-normoxic oxygen tension. An alternative interpretation of the data is that tracheal terminal cells are similarly sensitive but more hypoxic than other cells, thereby inducing hypoxia-dependent transcription with higher sensitivity. In either case, the results suggest that Sima-dependent transcription within the tracheal terminal cells is part of the mechanism of oxygen sensing and tracheal extra sprouting (Centanin, 2008).

To test this hypothesis directly, EGFP-labeled sima homozygous mutant terminal cells were generated, and it was found that the ability of these cells to ramify upon a hypoxic stimulus is largely impaired. Furthermore, whether overexpression of Sima in the tracheae can provoke the angiogenic-like response was examined, and it was found that, indeed, expression of Sima restricted to the tracheal system is sufficient to induce extra sprouting. In contrast, overexpression of Sima -- or of a nondegradable variant of Sima -- in flip-out random clones outside the tracheae failed to provoke a similar phenotype, suggesting that accumulation of Sima in these cells is not sufficient for extra sprouting. Interestingly, in these Sima flip-out clones, a cell-autonomous response was observed, in which long subcellular processes projected from the cells that overexpressed Sima. Thus, although it is clear that bnl is induced in hypoxia and attracts the extension of terminal branches, the data support the notion that Sima is necessary, but not sufficient, for bnl induction in hypoxia (Centanin, 2008).

This study investigated which Sima target genes might be responsible for tracheal extra sprouting in fga mutants or upon exposure of wild-type larvae to hypoxia. Northern blot analyses indicated that bnl and btl are both upregulated in mildly hypoxic larvae or fga mutants. However, bnl homozygous EGFP-labeled terminal cells of larvae exposed to hypoxia retained their branching capacity, suggesting that extra sprouting in hypoxia is not mediated by an autocrine effect of Bnl, upon Sima-dependent induction in tracheal cells. In contrast, btl is directly induced by Sima in tracheal cells, and, consistent with this, overexpression of Btl in tracheal cells is sufficient to mimic the phenotypes of larvae exposed to hypoxia. Thus the data suggest that Sima-dependent transcriptional induction of btl in tracheal terminal cells is a critical step of the angiogenic-like response of the tracheal system in hypoxic larvae (Centanin, 2008).

In summary, it is proposed that tracheal cells respond to hypoxia in an autonomous manner, by promoting the accumulation of Sima, which induces expression of the receptor Btl, thereby increasing sensitivity of these cells to the ligand Bnl. Concomitantly, Bnl is induced in hypoxic target tissues through a mechanism that also involves the participation of Sima, and serves to cue the outgrowth of terminal branches toward O₂-starved areas (Centanin, 2008).

During angiogenesis, vertebrate VEGF receptors are upregulated in endothelial cells of blood vessels that invade hypoxic tissues, and, particularly, Flt-1 induction is HIF dependent. Endothelial-specific overexpression of VEGF receptors might reveal to what extent this induction is a cardinal step in the angiogenic response to hypoxia (Centanin, 2008).

Reversion of lethality and growth defects in Fatiga oxygen-sensor mutant flies by loss of Hypoxia-Inducible Factor-alpha/Sima

Hypoxia-Inducible Factor (HIF) prolyl hydroxylase domains (PHDs) have been proposed to act as sensors that have an important role in oxygen homeostasis. In the presence of oxygen, they hydroxylate two specific prolyl residues in HIF-alpha polypeptides, thereby promoting their proteasomal degradation. So far, however, the developmental consequences of the inactivation of PHDs in higher metazoans have not been reported. This study describes novel loss-of-function mutants of fatiga (HIF prolyl hydroxylase), the gene encoding the Drosophila PHD oxygen sensor, that manifest growth defects and lethality. A null mutation in dHIF-alpha/similar (sima) is reported, that is unable to adapt to hypoxia but is fully viable in normoxic conditions. Strikingly, loss-of-function mutations of sima rescue the developmental defects observed in fatiga mutants and enable survival to adulthood. These results indicate that the main functions of Fatiga in development, including control of cell size, involve the regulation of dHIF/Sima (Centanin, 2005).

Recent work has led to the definition of widely operative signalling systems that control the transcriptional response to hypoxia through hypoxia-inducible factor (HIF). HIF proteins are a family of alpha/ß-heterodimers in which the common ß-subunit is constitutive and the alpha-subunits are oxygen-regulated by mechanisms that include transcriptional co-activator recruitment, subcellular localization and protein stabilization. The regulation of proteasomal degradation of alpha-subunits has been well characterized in cell culture and in in vitro systems. In the presence of oxygen, a series of 2-oxoglutarate and iron-dependent dioxygenases termed PHDs (prolyl hydroxylase domains) hydroxylate specific prolyl residues in the HIF-alpha oxygen-dependent degradation domain (ODDD), enabling its ubiquitination and proteasomal degradation. As molecular oxygen is absolutely required in the prolyl hydroxylation reaction and enzyme activity is sensitive to mild hypoxia, the PHDs have suitable characteristics that enable them to function as bona fide oxygen sensors that determine the half-life of HIF-alpha proteins, thereby controlling hypoxia-dependent transcription. Analyses of 'knockout' mouse strains have shown developmental roles of mammalian HIF proteins. They are required for the normal formation of the heart, brain, vasculature, cartilage and placenta, suggesting that fetal oxygen availability might have a role in these processes. However, this question remains open, and the developmental effects of genetic inactivation of the oxygen-sensitive PHD pathways have not yet been defined (Centanin, 2005).

The Drosophila bHLH-PAS proteins Similar (Sima) and Tango (Tgo) are, respectively, the functional homologues of HIF-alpha and HIF-ß in the fly. A lethal P-element insertional mutation (l(3)02255) is described in the Drosophila PHD gene (CG1114 in FlyBase) that fails to downregulate Sima protein in normoxia, thus driving constitutive activation of the transcriptional response to hypoxia. The aim of the present work was to investigate the developmental role of Drosophila PHD, which has been termed fatiga (fga; Spanish for 'fatigue') after its lack-of-oxygen phenotype (Centanin, 2005).

As a first step in the study of the functions of fga in development, new loss-of-function mutations were generated by mobilizing the l(3)02255 P-element, which is located between the second and third exons of the fga gene. Precise excisions of the transposon led to a reconstitution of the wild-type hypoxic response, as shown by hypoxia-inducible expression of transcriptional reporters that are based on the murine LDH-A enhancer. Imprecise excisions of the P-element resulted in three novel fga alleles (fga¹, fga⁹ and fga⁶⁴) that were characterized at the molecular level by Southern blot analysis and PCR experiments. fga⁹ conserved a 1.4 kb fragment of the original transposon; in fga⁶⁴, a large genomic portion upstream of the insertion site was removed, and fga¹ conserved a fragment of about 9 kb of the original P-element. In normoxic wild-type embryos, Sima protein and induction of the Ldh-Gal4 reporter are not detected; upon exposure to hypoxia (5% O₂), both Sima protein and reporter expression are observed, mainly in the tracheal system. The molecular basis of this pattern of induction is now under investigation. Interestingly, fga loss-of-function alleles show different levels of accumulation of Sima protein in normoxia, which was widely expressed in fga¹ and fga⁶⁴, and shows some prevalence in the tracheal system in fga⁹, correlating with the constitutive induction of the hypoxic reporter. fga⁰²²⁵⁵, fga¹ and fga⁶⁴ are lethal at the first larval instar and fga⁹ die in the pupal stage. Because overexpression of Sima through a ubiquitous Gal4 driver provokes lethality in the larval stages, it was reasoned that lethality in fga mutants could be due to overaccumulation of Sima protein in normoxia. To test this hypothesis and to determine if Fga is a dedicated regulator of Sima or whether, alternatively, it might modulate other molecular targets, attempts were made to analyse fga phenotypes in a sima-free genetic background. Loss-of-function mutations of sima have not been reported so far, but two different P-element insertions mapping within the sima locus were available from the Public Stock Centers. One of these insertion lines was able to respond to hypoxia and, thus, was indistinguishable from the wild type. In contrast, embryos homozygous for the other insertion, sima⁰⁷⁶⁰⁷, did not express sima mRNA and failed to induce the Ldh-LacZ reporter in hypoxia. Introduction of a UAS-Sima transgenic element under the control of an hs-Gal4 driver was able to rescue induction of reporter expression, which was expressed in a wild-type pattern, indicating that the absence of Sima was indeed responsible for the lack of hypoxic response. Altogether, these results indicate that sima⁰⁷⁶⁰⁷ is a sima loss-of-function allele (Centanin, 2005).

To explore whether the absence of Sima protein, and thus the inability to respond to hypoxia, affects developmental progression, phenotypes were analysed in sima⁰⁷⁶⁰⁷ mutants. Homozygous mutant embryos developed without any obvious difference from the controls, and the first-instar larvae looked healthy and motile. Next, homozygous mutant or control larvae were placed in vials containing fresh food, which were then exposed to 21% or 5% O₂ until individuals attained the pupal stage. sima⁰⁷⁶⁰⁷ mutants were viable and fertile in normoxia, but virtually unable to develop in hypoxia. Precise excision of the P-element totally reverted hypoxia-dependent lethality, which indicated that the insertion was indeed responsible for this phenotype. Thus, it is concluded that, unlike Tango that participates as a common bHLH-PAS partner in several developmental processes in normoxia, Sima is necessary for developmental progression in hypoxia but not in normoxia (Centanin, 2005).

fga mutations caused a reduction in cell size, but it is unclear whether this effect depends on overaccumulation of Sima. The availability of sima⁰⁷⁶⁰⁷ as a sima loss-of-function allele enabled this particular question to be answered and, more generally, the extent to which the developmental defects of fga loss-of-function mutations are due to the de-regulated accumulation of Sima protein to be addressed. As expected, in fga¹sima⁰⁷⁶⁰⁷ double homozygous mutants, Sima protein is undetectable and embryos do not show any expression of hypoxia-inducible reporters in normoxia. Consistent with previously described growth defects of fga mutants, fga⁹ pupae are smaller than their heterozygous siblings and, interestingly, they exhibited a delay in larval development, taking 2 additional days to reach the pupariation stage. Strikingly, fga¹sima⁰⁷⁶⁰⁷ double homozygous mutants are indistinguishable from the controls, both in their pupal weight and in the duration of larval development. Thus, the loss of Sima provokes the complete reversion of growth defects occurring in fga mutants. To answer whether overaccumulation of Sima is sufficient to account for the autonomous reduction in cell size reported for fga mutant cells, Sima protein was overexpressed in random clones using the flipase-induced recombination (FLP-OUT) technique, and the effect on cell size was analysed. Overexpression of Sima in isolated cells causes a marked autonomous reduction of cell size, which correlated with smaller nuclei. Taken together, these results indicate that Sima is a downstream effector of Fga as a regulator of cell growth. Further analyses were carried out on the tracheal system; once again defects (particularly, air-filling impairment) that are observed in fga mutants, are corrected in fga sima double mutants (Centanin, 2005).

Given the reversion of the analysed fga phenotypes in fga sima double mutants, it was of interest to test whether lethality that occurred following fga loss-of-function is also due to overaccumulation of Sima. This is indeed the case, since, in normoxia, fga¹ sima⁰⁷⁶⁰⁷ double homozygous mutants are viable to adulthood, even when many of these adults fail to complete emergence from the pupal case, and those that emerged looked weak and frequently die shortly afterwards. As expected, in hypoxia, this reversion of lethality does not occur, and fga or sima single mutants, as well as fga sima double mutant flies, die in the first larval stage. Overall, these results show that Drosophila development can proceed in the absence of PHD oxygen sensors, provided that the HIF-alpha subunit is absent and oxygen availability is not compromised. Thus, it is concluded that the most fundamental functions of Fatiga/PHD in development probably involve the downregulation of Sima protein levels. However, fga sima exarate adults show defects in wing and ovary development, which may imply that Fga is involved in patterning these organs in a Sima-independent manner. Detailed genetic and molecular analyses of fga sima double mutants should help to define Sima-independent developmental functions of the oxygen sensor in Drosophila (Centanin, 2005).

Drosophila Cyclin D/Cdk4 requires Hif-1 prolyl hydroxylase, potentially independently of Sima/Hif, to drive cell growth

The Drosophila cyclin-dependent protein kinase complex Cyclin D/Cdk4 induces cell growth (accumulation of mass) as well as proliferation (cell cycle progression). To understand how CycD/Cdk4 promotes growth, a screen was performed for modifiers of CycD/Cdk4-driven overgrowth in the eye. Loss-of-function mutations in Hif-1 prolyl hydroxylase (Hph), an enzyme involved in the cellular response to hypoxic stress, dominantly suppress the growth but not the proliferation function of CycD/Cdk4. hph mutant cells are defective for growth, and, remarkably, ectopic expression of Hph is sufficient to increase cellular growth. Epistasis analysis places Hph downstream of CycD/Cdk4. Overexpressed CycD/Cdk4 causes an increase in Hph protein in tissues where Hph induces growth, suggesting a mechanism whereby Hph levels are regulated posttranscriptionally in response to CycD/Cdk4. These data suggest that Hph, in addition to its function in hypoxic response, is a regulator of cellular growth and that it is a key mediator for CycD/Cdk4. Nevertheless, observations weigh against an important role for Hif-1 in Hph-driven growth, suggesting a potential Hif-1-independent function of Hph (Frei, 2004).

The synthetic promoter element Glass multiple reporter (GMR) is expressed in the eye imaginal disc posterior to the morphogenetic furrow, where only cells in the second mitotic wave undergo one synchronized cell division. GMR can be used to drive expression of the yeast transcription factor Gal4. Gal4 can be used to direct transcription of transgenes placed next to the UAS binding site of Gal4. Therefore, UAS transgenes driven by Glass-activated GMR-Gal4 are expressed predominantly in postmitotic cells. Under these circumstances, expression of CycD/Cdk4 leads to an enlargement of the adult eye, bigger ommatidia and bristles, and a general rough appearance. Although some ommatidia have additional cells, the main cause of the enlargement is an increase in cell size leading to 61% larger ommatidia. In order to identify loss-of-function mutants that modify this phenotype, CycD/Cdk4 was expressed in a deficiency collection background and screened under the light microscope for modifiers. Out of 162 deficiencies that cover 60%-70% of the genome, four deficiencies were isolated that dominantly suppress CycD/Cdk4. Df(3R)6-7, which deletes polytene segments 82D3/8-F3/6, led to a decrease in the enlargement of the eye and bristle size. Most strikingly, under these conditions, CycD/Cdk4 led to an increase in ommatidia size of only 17. Subsequently, partial overlapping deficiencies were tested: Df(3R)3-4 showed the same suppression phenotype but not a third deficiency, Df(3R)110, demonstrating that the gene of interest is between 82F3 and F7. All available mutants in this region were tested, and two lethal P element insertions were identified that showed an identical suppression phenotype compared to the deficiencies: l(3)02255 is inserted 104 bp, and l(3)S030304 is inserted 1111 bp upstream of the putative transcription start site of Hph/dmHph/CG1114 (Frei, 2004).

Whether l(3)02255 and l(3)S030304 are alleles of hph and whether their loss of function causes the suppression of CycD/Cdk4 were addressed. Northern blot experiments revealed that both heterozygous mutants l(3)02255 and l(3)S030304 have reduced expression of hph compared to the endogenous levels and that l(3)02255/l(3)S030304 transheterozygotes lack all detectable hph expression. In addition to these loss-of-function mutants, the EP3200 line has an EP insertion 382 bp upstream of Hph. Expression of hph using this EP element, by the hs-Flp Act>CD2>Gal4 system, led to a weak expression of hph in the absence of a heat shock, due to leakage of the system. A further increase in hph expression occured upon heat shock (Frei, 2004).

A full-length Hph cDNA was cloned under the control of a UAS promoter and injected into flies. These Hph transgenes suppress at least partially the cell growth phenotype of homozygous hph mutants and completely suppress the reduced viability of l(3)02255/l(3)S030304 transheterozygotes. Furthermore, when UAS-Hph transgenes were coexpressed with CycD/Cdk4 using the GMR-Gal4 driver, the dominant suppression of CycD/Cdk4-driven overgrowth by l(3)02255 was inhibited and the overgrowth phenotype was restored. It is concluded that l(3)02255 and l(3)S030304 are alleles of Hph and that the reduction of Hph can suppress CycD/Cdk4-induced overgrowth (Frei, 2004).

Whether growth or proliferation would be suppressed by hph in the eye imaginal disc was tested. GFP was expressed either alone or together with CycD/Cdk4 in wild-type, hph⁰²²⁵⁵/+, or Df(3R)3-4/+ backgrounds using the GMR-Gal4 driver. Imaginal eye discs from wandering third instar larvae were dissected, trypsinized to single cells, and analyzed by FACS for their cell size by using the forward scatter (FCS). Expression of CycD/Cdk4 in a wild-type background led to an increase in the forward scatter of 20%-30%, which was reduced to 10%-15% in an hph⁰²²⁵⁵/+ or Df(3R)3-4/+ background. Acridine orange staining showed that the suppression phenotype was not due to an increase in cell death. Furthermore, the cell cycle distribution was analyzed of eye imaginal discs from wandering larvae or pupae 48 hr after prepupae formation. At both time points, the increase in cells entering S and G2/M phases of the cell cycle due to ectopic expression of CycD/Cdk4 was not altered in an hph⁰²²⁵⁵/+ background. Taken together, these data demonstrate that the cell size and proliferation functions of CycD/Cdk4 can be separated. Furthermore, Hph is required for the increase in cell size but not required for proliferation, suggesting that Hph functions downstream of CycD/Cdk4 in a growth-specific manner (Frei, 2004).

In the experiments described above, CycD/Cdk4 was induced in mostly postmitotic cells of the eye imaginal disc. To test suppression by hph in mitotically dividing cells, CycD/Cdk4 was induced during larval development, and wing discs cells were analyzed. Ectopic expression of CycD/Cdk4 shows a distinctive induction of growth: cells divide at a faster rate but are otherwise indistinguishable from control cells from the same disc. Therefore, when single clones are measured, the clone area is increased, and the clone consists of more cells with no change in cell size or cell cycle phasing. Since columnar cells of wing discs form a single cell layer, measuring the clone area gives an accurate estimation of the amount of mass that was accumulated during the growth of the clone. CycD/Cdk4 was overexpressed together with GFP in random clones using the hs-Flp Act>CD2>Gal4 system and analyzed after a 48 hr growth period. Compared to external control clones expressing only GFP, expression of CycD/Cdk4 caused a 75% increase in the median clone size. This phenotype depends on Hph, since the median clone size was reduced to control level in a heterozygous hph⁰²²⁵⁵ mutant background. The suppression did not correlate with an increase in apoptosis, since coexpression of the cell death inhibitor p35 gave identical phenotypes. When cell size and cell cycle phasing were analyzed by FACS, there was no difference between cells expressing CycD/Cdk4 and internal control cells in either wild-type or hph/+ mutant backgrounds. These results demonstrate that the induction of growth by CycD/Cdk4 depends on normal levels of Hph. Furthermore, since hph suppressed growth but not proliferation in the eye imaginal discs, expression of CycD/Cdk4 in the wing should lead to a change in cell size if only growth but not proliferation were suppressed. However, no difference in cell size was detected, suggesting that the increase in proliferation caused by CycD/Cdk4 is secondary to the induction of growth (Frei, 2004).

Whether hph function was required for normal rates of cell growth was tested. Most hph⁰²²⁵/hph⁰²²⁵⁵ or hph⁰²²⁵⁵/Df(3R)3-4 animals die during embryogenesis, and only a few larvae hatch. These mutant larvae have severe growth defects and die within 2 to 3 days. Transheterozygotic hph^S030304/hph⁰²²⁵⁵ mutants develop normally until pupariation, but very few escaper adults eclose. These escapers are smaller than their heterozygous siblings but have normal body proportions. Weight measurements showed that hph^S030304/hph⁰²²⁵⁵ mutant flies are 18% lighter than heterozygotes. Therefore, hph mutant animals show a phenotype similar to homozygous cdk4³ flies or wild-type flies reared at low oxygen (Frei, 2004).

To test whether hph mutant cells are autonomously defective for growth, homozygous mutant clones were induced in the fat body using ionizing radiation. hph⁰²²⁵⁵ was crossed to flies expressing GFP under the control of a constitutively active promoter inserted on the same arm of the chromosome as Hph (3R). The progeny were irradiated during embryogenesis, emerging larvae were grown in regular food, wandering third instar larvae were dissected, and their fat bodies were fixed and mounted. Homozygous hph⁰²²⁵⁵ mutant cells lacked GFP, whereas heterozygous mutant cells expressed GFP. hph⁰²²⁵⁵/hph⁰²²⁵⁵ cells were smaller and contain less DNA than heterozygous neighboring cells. Importantly, the presence of a UAS-Hph transgene partially suppresses this phenotype, indicating that loss of Hph is the cause of the growth defect (Frei, 2004).

Whether ectopic expression of Hph is sufficient to stimulate growth was tested. EP3200 or UAS-Hph transgenes were used to induce Hph expression. Cell clones expressing Hph were induced in wing imaginal discs, and the median clone size was measured. Expression of Hph led to an increase in clone area very similar to CycD/Cdk4. Surprisingly, expression of Hph together with CycD/Cdk4 stimulated clonal growth to the same extent as Hph alone. However, in the presence of the apoptosis inhibitor p35, an additive phenotype was detected when both growth drivers were coexpressed. As for CycD/Cdk4, overexpressed Hph did not change cell size or cell cycle phasing, as assayed by FACS (Frei, 2004).

To test whether Hph functions downstream of CycD/Cdk4 also in this tissue, clones expressing Hph were induced in a homozygous cdk4³ mutant background, and the median clone size was measured. Under these conditions, Hph led to a very similar induction of growth as in a wild-type background. FACS analysis indicates that there are no detectable changes in cell size or cell cycle phasing. These data show that Hph is sufficient to stimulate growth, and the finding that this stimulation is independent of Cdk4 suggests that Hph functions downstream of CycD/Cdk4 (Frei, 2004).

Ectopic expression of CycD/Cdk4 in the posterior compartment of the wing imaginal disc using the en-Gal4 promoter leads to an enlargement of the posterior compartment in adult wings with no change in trichome (hair) density. Since the trichome density is proportional to the number of cells per area, the increase in compartment size is due to more cells of the same size. When Hph was expressed under en-Gal4 control using EP3200 or UAS-Hph transgenes, a similar result was obtained: posterior compartments were bigger and contained more cells of the same size. Thus, Hph induces growth in a similar manner to CycD/Cdk4 in wing imaginal discs. However, when Hph was expressed in the eye imaginal disc using the GMR-Gal4 driver, no increase in cell size, as assayed by FACS, was observed. Furthermore, adult eyes were not enlarged. Further experiments are required to understand why Hph expression is not sufficient to increase growth in the eye imaginal disc (Frei, 2004).

Little is known about how Hph RNA or protein levels are regulated. In Drosophila embryos, hph is expressed uniformly and does not seem to be subject to patterning. Vertebrate cells have three Hph orthologs (Taylor, 2001): HPH-3 protein localizes to the nucleus, HPH-2 exclusively to the cytoplasm, and HPH-1 mainly in the cytoplasm with a little staining in the nucleus (Huang, 2002; Metzen, 2003; Frei, 2004).

To test whether the subcellular localization or levels of Hph are altered in response to CycD/Cdk4, polyclonal antibodies were raised to full-length Hph. To test the specificity of the antiserum, Hph was overexpressed in the posterior compartment of the wing using the en-Gal4 driver. Hph staining increased in posterior regions, both in the peripodial and columnar epithelium. Homozygous hph mutant cells, marked by the absence of GFP, lacked detectable Hph staining. Therefore, the serum is specific for Hph and is able to detect endogenous levels of Hph. Furthermore, in third instar imaginal wing discs, Hph staining was uniform throughout the disc and specific for the nucleoplasm of the cells. Very little staining was detectable in the cytoplasm or the nucleolus (Frei, 2004).

When Hph staining was analyzed in cells expressing ectopic CycD/Cdk4, increased Hph levels were observed. Antiserum staining in the peripodial epithelium shows that these cells had an increase in nucleoplasmic as well as cytoplasmic Hph. In contrast, when homozygous cdk4³ mutant cells, marked by the absence of GFP, were analyzed, only background levels were detected. Therefore, Hph protein levels are regulated in response to CycD/Cdk4 gain and loss of function. When hph expression was analyzed by RT-PCR from wing discs expressing ectopic CycD/Cdk4, no effect on Hph RNA levels was detected. Furthermore, when whole third instar larvae were analyzed by RT-PCR or microarray analysis, no change in Hph expression was observed. Taken together, these observations suggest that CycD/Cdk4 affects Hph levels posttranscriptionally (Frei, 2004).

The hydroxylation activity of HPHs depends of Fe²⁺ bound to the active site (Epstein, 2001). Therefore, iron chelators like deferoxamine mesylate (DFO) are commonly used to experimentally mimic hypoxic conditions. When Drosophila larvae are raised on regular food supplemented with 2 mM DFO, they show an induction in Hif-1α/β, as assayed with a reporter construct (Lavista-Llanos, 2002), that is very similar to that seen under hypoxic conditions. When CycD/Cdl4 or Hph are expressed in the posterior compartment of the wing imaginal disc, the increase in compartment areas is suppressed by DFO. Moreover, when DFO is added to flies expressing CycD/Cdk4 in the postmitotic eye using the GMR-Gal4 driver, the enlargement of the adult eye as well as the rough appearance is suppressed, however, not to the same extent as in the heterozygous hph mutant backgrounds. This suggests that the hydroxylation activity of Hph is required for its growth function and that Hph is a major growth effector of CycD/Cdk4 (Frei, 2004).

The finding that Drosophila Hph functions downstream of CycD/Cdk4 and is sufficient to increase growth when overexpressed suggests that CycD/Cdk4 and Hph work in a common pathway. Consistent with this, heterozygous hph mutants do not suppress the extra growth induced by components of the insulin pathway or dMyc. Moreover, increases in cell size and changes in cell cycle phasing induced by the insulin signaling pathway, dMyc, or Ras in wing imaginal disc cells, do not depend on Cdk4. Taken together, these results suggest that the CycD/Cdk4-Hph pathway functions separately from these other growth regulatory pathways (Frei, 2004).

Since the kinase activity of Cdk4 is required for the induction of growth and proliferation (Meyer, 2000), Hph could be a phosphorylation target of Cdk4. The consensus sequence of vertebrate pocket proteins, the only known targets of cyclin D1/Cdk4, can be different from the classical CDK sequence. In Drosophila Rbf1, two potential sites have been found that disrupt its regulation by CycD/Cdk4 and CycE/Cdk2: T³⁵⁶PLTR and S⁷²⁸PHPK (Xin, 2002). Both sites are different from the vertebrate consensus sequence. Therefore, a search for putative consensus sequences on Hph is difficult. However, there are three sites that have the minimal requirement of a serine or threonine residue followed by a proline: T⁹¹PDAP, T²⁰⁴PGTT, and T²⁸⁵PPAA. None of these resemble the consensus sequences recognized by either the vertebrate or Drosophila complex. Nevertheless, future experiments should address whether CycD/Cdk4 phosphorylates Hph on these or other sites and how this affects Hph function (Frei, 2004).

It is proposed that in wing discs, Hph protein levels are regulated in response to CycD/Cdk4. Although it cannot be excluded that growth is also induced in an Hph-independent manner, the findings that overgrowth driven by CycD/Cdk4 and Hph is suppressed nearly completely by the iron chelator DFO or by heterozygosity for hph suggest that this is a major mechanism. Moreover, it is proposed that the small size of flies reared at low oxygen concentrations is caused at least partially by a decrease in Hph activity due to the absence of oxygen (Frei, 2004).

How does Hph induce growth? Since the hydroxylation inhibitor DFO suppressed the increases in growth caused by CycD/Cdk4 or Hph, Hph's hydroxylation activity is probably required. The only characterized hydroxylation target of Hph is Hif-1α, a mediator of the transcriptional response to hypoxia (Semenza, 2001; Kaelin, 2002; Bruick, 2003). Although mutant alleles of the Drosophila Hif-1α ortholog sima are not available, a partial loss-of-function allele of the Hif-1β ortholog, tango¹, was available. To test the potential role of Hif1 in growth control, the ey-Flp/FRT method was used to generate flies in which the eyes were >80% homozygous mutant for tango¹. If Hph stimulates growth by hydroxylating Hif-1α and targeting it for degradation, then loss of Hif-1 activity might be expected to result in overgrowth phenotypes. Contrary to this expectation, overgrowth was not observed in tango¹/tango¹ eyes. Moreover, GMR-driven expression of CycD/Cdk4 led to the same degree of overgrowth in tango¹/tango¹ eyes as in wild-type controls. Although these observations weigh against an important role for Hif-1 in Hph-driven growth, it is important to note that tango¹ is not a null allele and that Tango is thought to be expressed in excess over its binding partner, Sima. Thus, further analysis using sima mutants and overexpression will be required to definitively test whether Hph drives cell growth via a Hif-1-dependent mechanism or through hydroxylation of novel targets. The finding that only one of the three vertebrate Hph orthologs is required for regulation of Hif-1α levels in vivo (Berra, 2003) further suggests that additional targets may be important (Frei, 2004).

There is little data that suggest a growth function for vertebrate HPH. Rat HPH-1/SM-20 was identified first as a gene upregulated by growth factors or serum. The induction is very fast and peaks at 60 min after stimulation. Remarkably, this induction does not require de novo protein synthesis, as it is not blocked by the translation inhibitor cyclohexamide. The effect on growth upon deregulation of mouse Falkor/HPH-3 is controversial: whereas expression of a C-terminal fragment induced cells to grow faster and to a higher density, expression of a wild-type construct had no effect. An antisense oligonucleotide specific for Falkor induced cells to grow faster. Thus, the function of vertebrate HPH family member in growth control is still ambiguous (Frei, 2004).

Drosophila Hph has at least two functions: response to hypoxia and regulation of growth. How are they linked? In response to hypoxia, Sima/Tango activity is strongly induced in endoreplicative tissues like trachea, gut and fat body, and to a much lesser extent, in imaginal discs (Lavista-Llanos, 2002). Although endopreplicative cells lacking Hph are impaired for growth, ectopic overexpression of Hph in these cells does not increase their size. In contrast, in imaginal discs, Hph can increase growth when overexpressed. It is speculated that in endoreplicative tissues, Hph's main function is to regulate the hypoxic response and, to a minor extent, growth, whereas in imaginal tissues, Hph's main function is to regulate growth. Taken to the environment of wild Drosophila, this suggests that hypoxic conditions, which are often found in fermenting fruit, may induce a strong hypoxic response in endoreplicative tissues. Since these tissues are metabolically highly active, this response may be required for the generation of sufficient ATP by the induction of glycolysis. In imaginal discs, cell cycle progression is not controlled primarily by extrinsic factors but by disc intrinsic growth cues. Therefore, even under hypoxic stress, growth and development of imaginal discs continues but may be slowed down, presumably by inactivation of Hph activity, in order to ensure the formation of adult animals (Frei, 2004).

In fat body cells, Hph is a nuclear protein, and homozygous Cdk4 mutant cells lack detectable Hph levels. Moreover, ectopic expression of CycD/Cdk4 leads to more Hph protein in the cytoplasm and/or the nucleus. Surprisingly, a reporter line showed an increase, rather than a decrease, in Sima activity upon expression of CycD/Cdk4. It is proposed that in the fat body, Hph induced by CycD/Cdk4 is not sufficient to hydroxylate Hif-1α. In addition to the cofactors oxygen and iron, hydroxylation activity requires the binding of 2-oxoglutarate to the active site of HPH (Epstein, 2001; Bruick, 2001). 2-oxoglutarate is an intermediate of the citrate cycle, and its levels might correlate with the metabolic activity of the cell. Therefore, Hph protein may be induced by CycD/Cdk4 but may require 2-oxoglutarate and oxygen for catalytic activity in the fat body. In this model, Hph would integrate the regulation of growth by CycD/Cdk4 and its upstream regulators, with the regulation of growth by the metabolic activity, mediated by oxygen and 2-oxoglutarate (Frei, 2004).

Search PubMed for articles about Drosophila Fatiga

Acevedo, J. M., Centanin, L., Dekanty, A. and Wappner, P. (2010). Oxygen sensing in Drosophila: multiple isoforms of the prolyl hydroxylase fatiga have different capacity to regulate HIFalpha/Sima. PLoS One 5(8): e12390. PubMed ID: 20811646

Appelhoff, R. J., et al. (2004). Differential function of the prolyl hydroxylases PHD1, PHD2, and PHD3 in the regulation of hypoxia-inducible factor. J. Biol. Chem. 279: 38458-38465. PubMed ID: 15247232

Aragones, J., et al. (2008). Deficiency or inhibition of oxygen sensor Phd1 induces hypoxia tolerance by reprogramming basal metabolism. Nat. Genet. 40: 170-180. PubMed ID: 18176562

Barth, S., et al. (2007). The peptidyl prolyl cis/trans isomerase FKBP38 determines hypoxia-inducible transcription factor prolyl-4-hydroxylase PHD2 protein stability. Mol. Cell. Biol. 27: 3758-3768. PubMed ID: 17353276

Barth, S., et al. (2009). Hypoxia-inducible factor prolyl-4-hydroxylase PHD2 protein abundance depends on integral membrane anchoring of FKBP38. J. Biol. Chem. 284: 23046-23058. PubMed ID: 19546213

Berra, E., Benizri, E., Ginouves, A., Volmat, V., Roux, D. and Pouyssegur, J. (2003). HIF prolyl-hydroxylase 2 is the key oxygen sensor setting low steady-state levels of HIF-1alpha in normoxia. EMBO J. 22: 4082-4090. 12912907

Bruick, R. K. and McKnight, S. L. (2001). A conserved family of prolyl-4-hydroxylases that modify HIF. Science 294: 1337-1340. 11598268

Centanin, L., Ratcliffe, P. J. and Wappner, P. (2005). Reversion of lethality and growth defects in Fatiga oxygen-sensor mutant flies by loss of hypoxia-inducible factor-alpha/Sima. EMBO Rep. 6: 1070-1075. PubMed ID: 16179946

Centanin, L., et al. (2008). Cell autonomy of HIF effects in Drosophila: tracheal cells sense hypoxia and induce terminal branch sprouting. Dev. Cell 14(4): 547-58. PubMed ID: 18410730

Choi, K. O., et al. (2005). Inhibition of the catalytic activity of hypoxia-inducible factor-1alpha-prolyl-hydroxylase 2 by a MYND-type zinc finger. Mol. Pharmacol. 68: 1803-1809. PubMed ID: 16155211

Doronkin, S., et al. (2010). Dose-dependent modulation of HIF-1alpha/sima controls the rate of cell migration and invasion in Drosophila ovary border cells. Oncogene 29(8): 1123-34. PubMed ID: 19966858

Epstein, A. C., et al. (2001). Caenorhabditis elegans EGL-9 and mammalian homologs define a family of dioxygenases that regulate HIF by prolyl hydroxylation. Cell 107: 43-54. 11595184

Frei, C. and Edgar, B. A. (2004). Drosophila Cyclin D/Cdk4 requires Hif-1 prolyl hydroxylase to drive cell growth. Dev. Cell 6: 241-251. 14960278

Fong, G. H. and Takeda, K. (2008). Role and regulation of prolyl hydroxylase domain proteins. Cell Death Differ. 15: 635-641. PubMed ID: 18259202

Huang, J., Zhao, Q., Mooney, S. M. and Lee, F. S. (2002). Sequence determinants in hypoxia-inducible factor-1alpha for hydroxylation by the prolyl hydroxylases PHD1, PHD2, and PHD3. J. Biol. Chem. 277: 39792-39800. 12181324

Jaakkola, P., et al. (2001). Targeting of HIF-alpha to the Von Hippel-Lindau ubiquitylation complex by O₂-regulated prolyl hydroxylation. Science 292: 468-472. 11292861

Kaelin, W.G., Jr. (2002). How oxygen makes its presence felt. Genes Dev. 16: 1441-1445. 12080083

Lavista-Llanos, S., Centanin, L., Irisarri, M., Russo, D. M., Gleadle, J. M., Bocca, S. N., Muzzopappa, M., Ratcliffe, P. J. and Wappner, P. (2002). Control of the hypoxic response in Drosophila melanogaster by the basic helix-loop-helix PAS protein Similar. Mol. Cell. Biol. 22: 6842-6853. 12215541

Metzen, E., et al. (2003). Intracellular localisation of human HIF-1 alpha hydroxylases: implications for oxygen sensing. J. Cell Sci. 116: 1319-1326. 12615973

Meyer, C. A., et al. (2000). Drosophila Cdk4 is required for normal growth and is dispensable for cell cycle progression. EMBO J. 19(17): 4533-42. PubMed ID: 10970847

Ozer, A., Wu, L. C. and Bruick, R. K. (2005). The candidate tumor suppressor ING4 represses activation of the hypoxia inducible factor (HIF). Proc. Natl. Acad. Sci. 102: 7481-7486. PubMed ID: 15897452

Semenza, G.L. (2001). HIF-1, O₂, and the 3 PHDs: how animal cells signal hypoxia to the nucleus. Cell 107: 1-3. 11595178

Takeda, K., et al. (2008). Regulation of adult erythropoiesis by prolyl hydroxylase domain proteins. Blood 111: 3229-3235. PubMed ID: 18056838

Texada, M. J., Jorgensen, A. F., Christensen, C. F., Koyama, T., Malita, A., Smith, D. K., Marple, D. F. M., Danielsen, E. T., Petersen, S. K., Hansen, J. L., Halberg, K. A. and Rewitz, K. F. (2019). A fat-tissue sensor couples growth to oxygen availability by remotely controlling insulin secretion. Nat Commun 10(1): 1955. PubMed ID: 31028268

Taylor, M. S. (2001). Characterization and comparative analysis of the EGLN gene family. Gene 275: 125-132. 11574160

To, K. K. and Huang, L. E. (2005). Suppression of hypoxia-inducible factor 1alpha (HIF-1alpha) transcriptional activity by the HIF prolyl hydroxylase EGLN1. J. Biol. Chem. 280: 38102-38107. PubMed ID: 16157596

date revised: 16 August 2019

Home page: The Interactive Fly © 2008 Thomas Brody, Ph.D.

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