OGFOD1, a Member of the 2-Oxoglutarate and Iron Dependent Dioxygenase Family, Functions in Ischemic Signaling

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

FEBS Letters 584 (2010) 3340–3347

journal homepage: www.FEBSLetters.org

OGFOD1, a member of the 2-oxoglutarate and iron dependent dioxygenase family, functions in ischemic signaling

Ken Saito a, Noritaka Adachi b, Hideki Koyama b, Masayuki Matsushita a,c,*

a Mitsubishi Kagaku Institute of Life Sciences (MITILS), 11 Minamiooya, Machida, Tokyo 194-8511, Japan b Graduate School of Nanobioscience, Yokohama City University, 22-2 Seto, Kanazawa-ku, Yokohama 236-0027, Japan c Molecular and Cellular Physiology, University of the Ryukyus, Graduate School of Medicine, 207 Uehara, Nishihara, Okinawa 903-0215, Japan

article info abstract

Article history: The 2-oxoglutarate and iron dependent dioxygenase family are crucial for cellular adaptation to Received 26 February 2010 changes in oxygen concentration. We found that cells with OGFOD1 gene silencing in this family Revised 10 June 2010 showed resistance to cell death under ischemia, and cDNA microarray analysis of OGFOD1 knockout Accepted 11 June 2010 human cells revealed downregulation of ATPAF1. Although reintroduction of the OGFOD1 wild-type Available online 18 June 2010 gene to OGFOD1 KO cells restored ATPAF1 mRNA levels, the catalytically inactive OGFOD1 mutants Edited by Robert Barouki did not. Furthermore, introduction of ATPAF1 gene to OGFOD1 KO cells induced ischemic cell death. Thus, OGFOD1 plays an important role in ischemic cell survival and an OGFOD1 iron binding residue is required for ATPAF1 gene expression. Keywords: Ischemia Ó 2010 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Cell death 2-Oxoglutarate and iron dependent dioxygenase

1. Introduction and cardiomyocyte serve to enhance glucose uptake and metabo- lism for protection from hypoxic/ischemic injury [8,9]. VEGF also Metazoans can sense the oxygen concentration and respond to protects motor neurons from growth factor deprivation and hydro- hypoxia with adaptive changes in gene expression relating to angi- gen peroxide treatment in vitro and from ischemia in vivo [10,11]. ogenesis and glycolysis. This gene expression is induced by hypoxic Drug studies have reported that inhibitor, dimethyloxalylglycine inducible factor (HIF) [1,2]. Under normoxic conditions, HIF (HIF- (DMOG), for PHDs and FIH delay cell death caused by NGF depriva- 1a and HIF-2a) are hydroxylated by prolyl hydroxylase (PHDs) of tion in a stroke model [12,13]. DMOG is 2-oxoglutarate analogs and the 2-oxoglutarate and iron dependent dioxygenase (2-OG-Fe(II) inhibit the activity of 2-oxoglutarate-dependent dioxygenases dioxygenase) family, and then hydroxylated HIF is degraded by a through the competitive binding with 2-oxoglutarate to active site proteasome through interaction with the pVHL E3 ubiquitin ligase [4]. Thus, it is thought that regulation of PHDs and FIH might also complex [3]. Another 2-OG-Fe(II) dioxygenase, factor inhibiting be one of the survival responses under hypoxia/ischemia. HIF (FIH) hydroxylates an asparagine residue of HIF [4,5]. FIH mod- To discover the novel 2-OG-Fe(II) dioxygenases candidate in ulates the transcriptional activation rather than stabilization of ischemic survival, we screened this family using an siRNA library. HIF, preventing its interaction with the transcriptional co-activator In this study, OGFOD1 (FLJ10826) gene-silenced cells showed sig- p300 that potentially reduces HIF transcriptional activity [5,6]. niﬁcant resistance to cell death in ischemia, indicating a role of However, oxygen-dependent hydroxylation of HIF is reduced in OGFOD1 in ischemic cell survival. hypoxia, resulting in the stabilization, activation [3,5,7], and induc- tion of gene expression such as glucose transporters and vascular 2. Materials and methods endothelial growth factor (VEGF). In the hypoxic and ischemic cardiomyocyte, the expression of glucose transporters was increased See Supplementary data.

Abbreviations: 2-OG-Fe(II) dioxygenase, 2-oxoglutarate and iron dependent 3. Results dioxygenase; PHD, prolyl hydroxylase; FIH, factor inhibiting HIF; OGD, oxygen and glucose deprivation; DMOG, dimethyloxalylglycine 3.1. siRNA library screening * Corresponding author. Address: Molecular and Cellular Physiology, University of the Ryukyus, Graduate School of Medicine, 207 Uehara, Nishihara, Okinawa 903- 0215, Japan. Fax: +81 098 895 1402. Ischemia is complex physiological processes, in which multiple E-mail address: [email protected] (M. Matsushita). changes contribute to cellular adaptation. To study cell survival

0014-5793/$36.00 Ó 2010 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2010.06.015 K. Saito et al. / FEBS Letters 584 (2010) 3340–3347 3341 under these component stimuli, HeLa cells were cultured under conditions, but knockdown of EGLN1 and EGLN3 (also known as oxygen and glucose deprivation (OGD) conditions to induce ische- PHD2 and PHD3) induced cell death (Fig. 1C). Among these genes, mia. Cell death under OGD was prevented by the 2-OG-Fe(II) diox- function of OGFOD1 (FLJ10826) is unclear and it was selected as ygenase inhibitor DMOG in a dose-dependent manner (Fig. 1A). the most viable for further validation and characterization. Cells Therefore, we used the siRNA library of the 2-OG-Fe(II) dioxygen- with a silenced OGFOD1 gene survived under both normoxic and ase family (Table S1) to discover signiﬁcant genes responsible for OGD conditions (Fig. 2A and B). We also conﬁrmed that OGFOD1 cell survival under OGD. HeLa cells were added to the well contain- protein levels were decreased by siRNA under our experimental ing mixture of siRNA and transfection reagent (Fig. 1B). After 48 h conditions (Fig. 2C). of culture, cells were exposed to OGD conditions and cell survival was measured. We established a cut-off point (absorbance of 2.3) 3.2. Cellular localization of OGFOD1 in the siRNA library screening based on the value of cell viability using 2 mM DMOG, and obtained knockdown of ALKBH, DEPC-1, OGFOD1 has >80% identity between human (542 amino acids) PHF8, ASPH, FLJ10826, and JMJD2B genes in HeLa cells with the and mouse (545 amino acids). From the NCBI database, the N-ter- same or more viability as 2 mM DMOG-treated cells under OGD minal region in OGFOD1 was predicted to contain the prolyl 4- (Fig. 1A and C). While knockdown of HIF1AN and EGLN2 (also hydoxylase (P4H) domain (Fig. 3A). A BLAST search and alignment known as FIH and PHD1) showed the absorbance of 1.7 and 1.9, analysis indicated that the P4H domain of OGFOD1 shares 35–40% respectively. They were slightly increased survival under OGD identity with theSchizosaccharomyces pombe Ofd1 orSaccharomyces

2.5 HeLa Cells ( 3 x 103 cells / well )

2.0 siRNA (50 nM / well)

1.5 Culture for 48h

1.0 Change to the DMEM (1 g/L glucose)

Cell viability level 0.5 (Absorbance at 490 nm) Culture for 5 days in 21% or 0.1% O2 0.0 DMSO DMSO 0.5 0.75 1.02.0 (mM) ( - ) (0.2%) DMOG Cell death assay

3.0

2.5

2.0

1.5

1.0 Cell viability level (Absorbance at 490 nm) 0.5

0.0 PHF8 PLOD ASPH P4HA1 EGLN3 PLOD3 EGLN2 ALKBH PLOD2 EGLN1 PTDSR HIF1AN GAPDH DEPC-1 LEPRE1 JMJD2B JMJD2A JMJD2C FLJ13491 LEPREL1 FLJ10826 FLJ22222 FLJ20308 FLJ20013 HSPBAP1 Lamin A/C Cyclo B duplex RISC-free siRNA Non-Targeting siRNA siGLO Risc-free siRNA

Fig. 1. Screening of genes for cell survival under ischemia. (A) Cells were treated with the DMSO or dimethyloxalylglycine (DMOG). Cell viability under normoxic (black bars) and ischemic (white bars) conditions was measured by the absorbance of formazan products at 490 nm. Results represent mean ± S.D (n = 3). (B) Outline of the transfection and screening condition. (C) Ischemic cell survival in the knockdown of 2-OG-Fe(II) dioxygenase family genes was measured. Cut-off point (absorbance of 2.3) was indicated by dotted line. Results represent mean ± S.D. (n = 3). 3342 K. Saito et al. / FEBS Letters 584 (2010) 3340–3347

Control siRNA OGFOD1 siRNA 3.5 * 3.0

2.5

2.0 Phase

1.5

1.0 Cell viability level 0.5 (Absorbance at 490 nm) 0.0 PI siRNA siRNA siRNA siRNA Control Control OGFOD1 OGFOD1

Normoxia Ischemia Control siRNA OGFOD1 (kDa) siRNA 80 OGFOD1 60

50 Actin 40

Fig. 2. Cell viability with OGFOD1 knockdown. (A) OGFOD1 knockdown cells were exposed to ischemia for four days. Cell viability under normoxia (black bars) and ischemia (white bars) was measured by the absorbance of formazan products. Results are represented as means ± S.D. (n =6,*P < 0.01). (B) Cell death under ischemia was indicated by morphology of cells and PI staining. Scale bar = 50 lm. (C) OGFOD1 knockdown was detected by western blotting.

cerevisiae TPA1 genes (Fig. 3B); regions other than the P4H domain (OGFOD1hyg/puro) compared to wild-type cells (Table S2, GEO acces- have little homology (<20% identity). The N-terminal region of OG- sion number: GSE21557). Quantification of mRNA levels using FOD1 also contains a positively charged amino acid sequence qRT-PCR showed a 1.5-fold change that corresponded to the micro- (Fig. 3A, underlined), most likely correlating to a nuclear localiza- array data, except for the NAALADL2 and COX1 genes (Table S3). tion signal. To determine the cellular localization of OGFOD1, Taken together, the expression levels of many genes were down- EGFP-OGFOD1 (wild-type) and nuclear localization signal-deleted regulated in OGFOD1 KO cells. EGFP-OGFOD1 (DNLS) expression plasmids were transfected into HeLa cells. Wild-type expression was localized to the nucleus, while 3.4. Regulation of ATPAF1 expression by OGFOD1 the mutant OGFOD1 (DNLS) expression was not (Fig. 3C). OGFOD1 translocation in response to changes in oxygen concentration re- To confirm whether OGFOD1 modulates the downregulation of sulted in similar nuclear localization of endogenous OGFOD1 under genes, we reintroduced a OGFOD1 wild-type gene into the OGFOD1 both normoxic and ischemic conditions, indicating that OGFOD1 KO cells (revWT) and found that revWT expression recovered only functions in the nucleus (Fig. 3D). ATPAF1 mRNA levels among the downregulated genes (Table S4). This result indicates that the intact OGFOD1 appears to dominantly 3.3. Downregulated genes in OGFOD1 KO cells regulate ATPAF1 mRNA levels over the other downregulated genes. Therefore, we focused on the ATPAF1 gene for further studies of To investigate OGFOD1 function, we generated an OGFOD1 OGFOD1 function and examined whether the structure and regula- knockout cell line using human Nalm6 cells. The Nalm6 cell line tion of OGFOD1 are important for expression of the ATPAF1 gene. provides convenient and high-efficiency gene targeting [14].We We generated an OGFOD1 H155A mutant that substitutes the disrupted exons 5–8 of the P4H domain (Fig. 4A) by substitution conserved His with Ala, a residue that along with Asp is located with both hygromycin and puromycin resistance genes. In South- within the P4H domain and is important for iron coordination in ern blot analysis, OGFOD1hyg/+ and OGFOD1hyg/puro cells revealed the active site (Fig. 5A) [15,16]. We also generated an OGFOD1 C bands of 7.0 kb and 8.6 kb, respectively (Fig. 4B, left). Wild-type and OGFOD1 N mutants (Fig. 5A). Although these mutants were cells, but not OGFOD1 KO (OGFOD1hyg/puro) cells expressed OG- stably introduced into OGFOD1 KO cells, re-expression of ATPAF1 FOD1 (Fig 4B, right). This OGFOD1 KO cells had more viability than was not restored by revH155A, revN, or revC (Fig. 5B); the N-termi- wild-type cells under ischemia (Fig. 4C). nal domain of OGFOD1 (revN) is an unstable protein fragment Next, to understand how OGFOD1 KO results in ischemic cell (Fig. 5C). Furthermore, we examined the cell survival of revartant survival, we analyzed gene expression by cDNA microarray and cell lines under OGD conditions. The cells of revC or revH155A sig- qRT-PCR. Our cDNA microarray analysis showed genes that were nificantly reduced the cell death under OGD compared with wild- up- and downregulated by at least threefold in OGFOD1 KO cells type cells (Fig. 5D). The cell survival of revC was the same as that of K. Saito et al. / FEBS Letters 584 (2010) 3340–3347 3343

Fig. 3. Localization of OGFOD1 in the nucleus. (A) Domain structure of OGFOD1 was shown. (B) Sequence alignment of the P4H domain of TPA1, Ofd1, and OGFOD1. Residues predicted to be required for iron coordination are indicated by an asterisk. Conserved residues are shown in the gray box. (C) The localization of OGFOD1 (wild-type) and OGFOD1 (DNLS) was shown. Nuclei were stained with DAPI. Scale bar = 10 lm. (D) Endogenous OGFOD1 under normoxia and ischemia was shown. Scale bar = 10 lm.

OGFOD1 KO cells. These results confirm that the iron-coordinating 4. Discussion residue is required for ATPAF1 expression and cell death under OGD conditions. We identified the OGFOD1 gene, a member of the 2-OG-Fe(II) Under OGD conditions, endogenous ATPAF1 and OGFOD1 dioxygenase family, that plays an important role in ischemic sur- mRNA levels decreased in wild-type cells (Fig. S1A). To examine vival. It is well known that loss of PHDs and FIH (EGLN1, 2, 3 and whether decreased ATPAF1 mRNA levels under OGD conditions HIF1AN) stabilize the HIF for hypoxic adaptation. However, knock- correlate with OGFOD1 protein expression levels, FLAG-OGFOD1 down of EGLN1, 2, 3 and HIF1AN genes in our siRNA library screen- was overexpressed in HeLa cells (Fig. S1B). We found that the ing had no significant effect on ischemic survival compared to expression levels of FLAG-OGFOD1 did not increase ATPAF1 knockdown of Ogfod1 gene (Fig. 1). We found that HIF-1a expres- mRNA levels in normal and OGD condition (Fig. S1C). In the cell sion in OGFOD1 KO cells was slightly lower compared to wild-type survival assay under OGD, increased expression of FLAG-OG- cells and PHDs expression were not difference between OGFOD1 FOD1 had no effect on cell survival (Fig. S1D). Thus, we con- wild-type and KO cells (Fig. S2A), suggesting that PHDs, FIH, and cluded that ATPAF1 mRNA expression required the activity of OGFOD1 might recognize different substrates and play different OGFOD1 P4H domain rather than OGFOD1 protein expression roles in cell survival. levels. The S. pombe ortholog of OGFOD1, Ofd1 degrades the Sre1 ma- The yeast homolog of ATPAF1, ATP11p, has been reported to be ture form (Sre1N) and an Ofd1-deficient strain accumulates Sre1N a mitochondrial F1-ATPase assembly factor [17,18]. We confirmed [19], which is an ortholog of the human sterol regulatory element that ATPAF1 was co-localized with the mitochondrial maker, binding protein (SREBP). Ofd1 has a molecular mechanism where- DsRed-Mito, in HeLa cells (Fig. 6A, left). Moreover, introduction by the C-terminal region has catalytic activity under normoxia and of the ATPAF1 gene into OGFOD1 KO cells (Fig. 6A, right) signifi- the activity was suppressed by the binding of Nro1 protein under cantly increased cell death under OGD compared with OGFOD1 hypoxia [20]. In our study, OGFOD1 KO cells had little influence KO cells (Fig. 6B). on the accumulation of SREBP (Fig. S2B) and loss of OGFOD1 3344 K. Saito et al. / FEBS Letters 584 (2010) 3340–3347

26 kb 1960 b

exon 2 3 4 5 6 7 8 9 10 11 12 13 OGFOD1 locus

5’Arm 3’Arm pOGFOD1-Puro D-TA Puror Targeting vector 5’Arm 3’Arm pOGFOD1-Hyg D-TA Hygr

9.8 kb 8.6 kb

Puror Targeted locus EcoR I 7.0 kb Hygr EcoR I EcoR I probe EcoR I + /

+ / + ( ) + / Hyg Puro / Hyg (kb) Puro Hyg / kDa 80 10.0 9.8 kb OGFOD1 8.6 kb 7.5 7.0 kb 60

50 5.0 Actin 40

WT KO

*** 50

40 EGFP 30

% Cell death 10

0 (PI stained cells / total cells)

PI WT KO

Ischemia

Fig. 4. OGFOD1 KO cell line. (A) The human OGFOD1 gene is located on chromosome 16q13 and is composed of 13 exons (black boxes). Two targeting vectors are shown as pOGFOD1-Hyg and pOGFOD1-Puro. The diphtheria toxin A (DT-A), puromycin (Puror), and hygromycin (Hygr) resistance genes are marked by white boxes and arrows. Triangles represent loxP sequences. (B) The left panel shows southern blot of wild-type (+/+), heterozygous (+/hyg) and homozygous (puro/hyg) mutants. The right panel shows OGFOD1 expressions in wild-type and KO cells. (C) OGFOD1 wild-type and KO cells were stably transfected with EGFP expression plasmid. The cells were cultured under ischemia for five days. Cell death was indicated by PI staining (left panels). Bar = 50 lm. The right panel shows the number of PI stained cells. Results represent means ± S.D. (n =9,***P < 0.0001). decreased the ATPAF1 expression as well as the revH155A and revC nucleus (Fig. 3). Many 2-OG-Fe(II) dioxygenases in the nucleus func- cell lines (Fig. 5). Thus, we found that the iron-coordinating residue tion as demethylases for the methylated DNA, RNA and histones is required for ATPAF1 gene expression. If OGFOD1 has the same [21]. Studies of TPA1, an OGFOD1 ortholog in S. cerevisiae, have sug- intramolecular or intermolecular mechanism with Ofd1, it is gested control of translation termination, mRNA poly(A) tail length, thought that OGFOD1 accumulate a negative regulator of ATPAF1 and mRNA stability [22]. TPA1 has also been co-purified with the expression under ischemia. Further investigation is needed to NuA3 histone H3 HAT complex [23]. Thus, OGFOD1 may cooperates determine the molecular mechanism underlying the decrease in with the translation termination or histone modification complex. ATPAF1 in OGFOD1 KO cells. As shown in Fig. S2C, we found that Lys 9 methylation on histone Although the substrate of OGFOD1 is currently unclear, several H3 levels in OGFOD1 KO cells was lower compared to wild-type reports provide speculative evidence that OGFOD1 functions in cells. Therefore, OGFOD1 may regulates transcription through the K. Saito et al. / FEBS Letters 584 (2010) 3340–3347 3345

OGFOD1 WT P4H 1 545 a.a.

OGFOD1 H155A P4H 1 His155 Ala 545 a.a. OGFOD1 N P4H 1 238 a.a.

OGFOD1 C 1 63 238 545 a.a.

*** 1.2 rev WT rev N rev C rev H155A KO 1.0 (kDa) WT 210 0.8 120 100 OGFOD1 WT 80 OGFOD1 H155A 0.6 60 OGFOD1 C 0.4 50 40 (ATPAF1 / Actin) 0.2 Relative mRNA level OGFOD1 N 0.0 30 KO WT 50 rev N rev C Actin rev WT 40 rev H155A

* * KO WT *

70 60 50 40

revWT 30 revH155A

% Cell death 20 10

(PI stained cell / total cells) 0 KO WT rev N rev C revN revC rev WT rev H155A

Fig. 5. Regulation of ATPAF1 mRNA level by OGFOD1. (A) The construction of 3 Â FLAG-tagged OGFOD1 variants. (B) ATPAF1 mRNA levels in OGFOD1 revertant cells (revWT, revH155A, revN, and revC) were detected by qRT-PCR. Results represent means ± S.D. (n =6,***P < 0.0001). The same results were also obtained in other revertant clones. (C) Expression levels of OGFOD1 variants in OGFOD1 revertant cells were detected by western blotting using anti-FLAG antibody. (D) Cell survival of OGFOD1 revertant cell lines under ischemia for ﬁve days were shown. Cell death was indicated by PI staining (left panels). Bar = 100 lm. The right panel shows the number of PI stained cells. Results represent means ± S.D. (n =3,*P < 0.05). chromatin structure. It will be interesting to investigate whether Finally, we propose that intact OGFOD1 iron binding residues histone H3 modiﬁcations can regulate ATPAF1 transcription, and are required for ATPAF1 gene expression and that changes in the whether mitochondrial ATPAF1 expression contributes to the reduc- activity/structure of OGFOD1 induce cell survival under ischemic tion of ADP/ATP ratio (Fig. S3) and cell death (Fig. 6). We also could conditions. These results and further studies of OGFOD1 will im- not rule out the possibility that loss of OGFOD1 reduces cell death by prove the understanding of cell survival under ischemic conditions mechanism other than controlling ATPAF1 expression, since ATPAF1 including brain stroke, as OGFOD1 was found to be more abundant levels in revN cell lines were not coincident with cell death (Fig. 5). in the brain (Fig. S4). 3346 K. Saito et al. / FEBS Letters 584 (2010) 3340–3347

***

ATPAF1-EGFP DsRed-Mito 3.5 3.0 2.5 2.0 1.5 1.0

(ATPAF1 / Actin) 0.5 Relative mRNA level 0.0 WT KO ATPAF1

KO ATPAF1 ** 80

60 Phase

% Cell death 20

0 (PI stained cells / total cells) PI KO ATPAF1

Ischemia

Fig. 6. Cell death in the expression of ATPAF1. (A) Localization of ATPAF1 was shown (left panel). Scale bar = 50 lm. The right panel shows ATPAF1 mRNA levels in ATPAF1- overexpressing OGFOD1 KO cells (ATPAF1). Results represent means ± S.D. (n =6, ***P < 0.0001). The same results were also obtained in other clones. (B) PI staining of OGFOD1 KO (KO) and ATPAF1-overexpressing OGFOD1 KO cells cultured under ischemic conditions for four days. Cell death is indicated by PI staining (left) and the number of PI stained cells are shown (right). Bar = 50 lm. Results represent means ± S.D. (n =6,**P < 0.001).

Acknowledgements [8] Malhotra, R. and Brosius 3rd, F.C. (1999) Glucose uptake and glycolysis reduce hypoxia-induced apoptosis in cultured neonatal rat cardiac myocytes. J. Biol. Chem. 274, 12567–12575. We thank the members of the Matsushita Lab for suggestions [9] Brosius 3rd, F.C., Liu, Y., Nguyen, N., Sun, D., Bartlett, J. and Schwaiger, M. on experimental design, critical evaluation of this manuscript, (1997) Persistent myocardial ischemia increases GLUT1 glucose transporter and technical assistance (Kaminaga, Kobayashi, Sasaki). expression in both ischemic and non-ischemic heart regions. J. Mol. Cell. Cardiol. 29, 1675–1685. [10] Jin, K.L., Mao, X.O. and Greenberg, D.A. (2000) Vascular endothelial growth factor: direct neuroprotective effect in in vitro ischemia. Proc. Natl. Acad. Sci. Appendix A. Supplementary data USA 97, 10242–10247. [11] Sun, Y., Jin, K., Xie, L., Childs, J., Mao, X.O., Logvinova, A. and Greenberg, D.A. Supplementary data associated with this article can be found, in (2003) VEGF-induced neuroprotection, neurogenesis, and angiogenesis after focal cerebral ischemia. J. Clin. Invest. 111, 1843–1851. the online version, at doi:10.1016/j.febslet.2010.06.015. [12] Siddiq, A. et al. (2005) Hypoxia-inducible factor prolyl 4-hydroxylase inhibition. A target for neuroprotection in the central nervous system. J. Biol. Chem. 280, 41732–41743. References [13] Lomb, D.J., Straub, J.A. and Freeman, R.S. (2007) Prolyl hydroxylase inhibitors delay neuronal cell death caused by trophic factor deprivation. J. Neurochem. [1] Semenza, G.L. (1999) Perspectives on oxygen sensing. Cell 98, 281–284. 103, 1897–1906. [2] Semenza, G.L. (2001) Hypoxia-inducible factor 1: oxygen homeostasis and [14] Adachi, N., So, S., Iiizumi, S., Nomura, Y., Murai, K., Yamakawa, C., Miyagawa, K. disease pathophysiology. Trends Mol. Med. 7, 345–350. and Koyama, H. (2006) The human pre-B cell line Nalm-6 is highly proficient [3] Ivan, M. et al. (2001) HIFalpha targeted for VHL-mediated destruction by in gene targeting by homologous recombination. DNA Cell Biol. 25, 19–24. proline hydroxylation: implications for O2 sensing. Science 292, 464–468. [15] Bruick, R.K. and McKnight, S.L. (2001) A conserved family of prolyl-4- [4] Lando, D., Peet, D.J., Gorman, J.J., Whelan, D.A., Whitelaw, M.L. and Bruick, R.K. hydroxylases that modify HIF. Science 294, 1337–1340. (2002) FIH-1 is an asparaginyl hydroxylase enzyme that regulates the [16] McDonough, M.A. et al. (2006) Cellular oxygen sensing: crystal structure of transcriptional activity of hypoxia-inducible factor. Genes Dev. 16, 1466– hypoxia-inducible factor prolyl hydroxylase (PHD2). Proc. Natl. Acad. Sci. USA 1471. 103, 9814–9819. [5] Lando, D., Peet, D.J., Whelan, D.A., Gorman, J.J. and Whitelaw, M.L. (2002) [17] Ackerman, S.H. and Tzagoloff, A. (1990) Identification of two nuclear genes Asparagine hydroxylation of the HIF transactivation domain a hypoxic switch. (ATP11, ATP12) required for assembly of the yeast F1-ATPase. Proc. Natl. Acad. Science 295, 858–861. Sci. USA 87, 4986–4990. [6] Freedman, S.J., Sun, Z.Y., Poy, F., Kung, A.L., Livingston, D.M., Wagner, G. and [18] Wang, Z.G., Schmid, K.J. and Ackerman, S.H. (1999) The Drosophila gene 2A5 Eck, M.J. (2002) Structural basis for recruitment of CBP/p300 by hypoxia- complements the defect in mitochondrial F1-ATPase assembly in yeast lacking inducible factor-1 alpha. Proc. Natl. Acad. Sci. USA 99, 5367–5372. the molecular chaperone Atp11p. FEBS Lett. 452, 305–308. [7] Jaakkola, P. et al. (2001) Targeting of HIF-alpha to the von Hippel-Lindau [19] Hughes, B.T. and Espenshade, P.J. (2008) Oxygen-regulated degradation of ubiquitylation complex by O2-regulated prolyl hydroxylation. Science 292, fission yeast SREBP by Ofd1, a prolyl hydroxylase family member. EMBO J. 27, 468–472. 1491–1501. K. Saito et al. / FEBS Letters 584 (2010) 3340–3347 3347

[20] Lee, C.Y., Stewart, E.V., Hughes, B.T. and Espenshade, P.J. (2009) Oxygen- deadenylation, and mRNA turnover in Saccharomyces cerevisiae. Mol. Cell. Biol. dependent binding of Nro1 to the prolyl hydroxylase Ofd1 regulates SREBP 26, 5237–5248. degradation in yeast. EMBO J. 28, 135–143. [23] John, S., Howe, L., Tafrov, S.T., Grant, P.A., Sternglanz, R. and Workman, J.L. [21] Loenarz, C. and Schoﬁeld, C.J. (2008) Expanding chemical biology of 2- (2000) The something about silencing protein, Sas3, is the catalytic subunit of oxoglutarate oxygenases. Nat. Chem. Biol. 4, 152–156. NuA3, a yTAF(II)30-containing HAT complex that interacts with the Spt16 [22] Keeling, K.M., Salas-Marco, J., Osherovich, L.Z. and Bedwell, D.M. (2006) Tpa1p subunit of the yeast CP (Cdc68/Pob3)-FACT complex. Genes Dev. 14, 1196– is part of an mRNP complex that inﬂuences translation termination, mRNA 1208.