 |
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
Department of Anatomy and Cell Biology, University of Kansas Medical Center, Kansas City, Kansas.
Correspondence to Dale R. Abrahamson, Department of Anatomy and Cell Biology, University of Kansas Medical Center, 3901 Rainbow Boulevard, Kansas City, KS 66160-7400. Phone: 913-588-7000; Fax: 913-588-2710;
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and HIF-2 proteins bind HIF-1
and initiate expression of genes that alleviate hypoxic stress, including those promoting neovascularization. Both HIF-1 and HIF-2 stimulate transcription of vascular endothelial growth factor (VEGF), a crucial regulator of vascular development. Because VEGF is highly expressed by metanephric podocytes and collecting ducts, developing mouse kidney was examined for the presence and distribution of HIF-1, HIF-2, and HIF-1. The expression of HIF-1 and HIF-2 mRNAs in newborn mouse kidney was confirmed by RT-PCR and Northern blot analysis. By in situ hybridization, HIF-1 and HIF-2 mRNAs were highly expressed in the nephrogenic zone of newborn kidney cortex and in the medulla. Particularly intense hybridization was found in podocytes of developing glomeruli and in medullary collecting ducts. Both HIF-1 and HIF-2 heterodimers were identified in newborn kidney lysates by immunoprecipitation with HIF-1, HIF-2, and HIF-1 antibodies and Western blots. Immunofluorescence analysis of the hypoxia marker, pimonidazole, showed that collecting ducts and many developing tubules undergo severe hypoxia in developing kidney. Immunohistochemistry of newborn kidney demonstrated widespread expression of HIF-1 protein in nuclei of glomeruli and all tubular segments, whereas HIF-2 protein expression was more restricted and localized chiefly to podocytes of developing glomeruli and developing tubules. HIF-1 and HIF-2 protein and VEGF mRNA were all strongly induced in embryonic kidneys maintained in hypoxic organ cultures. Collectively, these data suggest that HIF stabilization, by hypoxia and/or by other means, may be critical for VEGF production and kidney vascular development. E-mail: dabrahamson@kumc.edu |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The HIF-1 heterodimer consists of HIF-1 and HIF-1 (ARNT) (3), whereas HIF-2 contains HIF-2 (EPAS1, HLF, HRF, MOP2) and HIF-1 (4). HIF bioactivities are regulated by stability of the subunits, which act as direct sensors of intracellular oxygen concentrations. When oxygen levels are sufficient to meet cellular demand (normoxia), HIF subunits are subjected to an oxygen-dependent prolyl hydroxylation, which promotes an interaction with Von Hipple Lindau protein (VHL), an E3 ubiquitin ligase (5). Ubiquitinated HIF- is then rapidly destroyed by 26S proteasomes (6). Conversely, when oxygen levels are insufficient (hypoxia), prolyl hydroxylation and subsequent HIF- destruction are avoided (5). Remarkably, an oxygen-dependent hydroxylation reaction also controls the transactivation properties of HIF-1. In normoxia, a conserved asparagine residue within the HIF-1 C-terminal transactivation domain, is hydroxylated, which blocks the recruitment of p300/CBP transcriptional co-activator. In hypoxia, asparagine hydroxylation is blocked and HIF-1 is free to engage p300 and initiate transcription (7).
Disturbance of oxygen homeostasis and abnormal HIF activity contribute to the pathogenesis of a number of disorders, including heart and chronic obstructive pulmonary diseases, preeclampsia, and many types of cancer (for reviews, see references 8 and 9). HIF function during embryogenesis, however, has yet to be examined in detail. Nevertheless, the vascular abnormalities and embryonic lethality observed in mice with targeted disruption of the HIF-1 (10,11 ), HIF-2 (12,13 ), and HIF-1 (14) genes suggest that adaptation to environmental hypoxia, including induction of HIF, is crucial for normal vascular development. That hypoxia exists in developing embryonic tissues has been confirmed by studies with the hypoxia marker pimonidazole and its monoclonal antibody (15). The hypoxia marker is detected in areas where cells are undergoing high rates of proliferation and in tissues with incomplete or developing vascular networks, including the developing kidney (15).
Among the genes strongly induced by HIF and highly expressed during kidney microvascular development is vascular endothelial growth factor (VEGF) (16,17 ), a potent endothelial mitogen and chemoattractant required for normal embryonic angiogenesis and vasculogenesis (18,19 ). Both VEGF tyrosine kinase receptors, Flk1 and Flt1, are also essential for proper vascular development, although they play different roles. Mice with targeted Flk1 mutations lack mature endothelial cells and do not undergo vasculogenesis (20), whereas Flt1 mutants contain mature endothelial cells but have malformed blood vessels (21). Tie-1 and Tie-2 receptor tyrosine kinases, which promote endothelial cell survival and branching morphogenesis, are also required for vascular development (22). The Tie-2 ligand, angiopoietin-1 (Ang-1), induces tyrosine phosphorylation of the receptor (23), whereas angiopoietin-2 (Ang-2) acts as an antagonist to Tie-2 activation (24). Concurrent Ang-2 and VEGF signaling hinders blood vessel stabilization and promotes vessel branching (25), whereas Ang-2 signaling in the absence of VEGF results in vessel regression (24). Activators of Tie-1 have not yet been identified.
VEGF, Flk1, and Flt1 are all expressed during kidney organogenesis. VEGF production localizes primarily to podocytes, distal tubules, and collecting ducts (26,27,28 ), and Flk1 and Flt1 have been detected in endothelial cells of glomeruli and other renal blood vessels (26,27,29 ). By analyses of mice expressing -galactosidase under control of the Flk1 promoter, cells expressing Flk-1 can be seen migrating into the vascular cleft of comma and S-shaped glomeruli (30). Interference of VEGF activity in newborn mice by administration of anti-VEGF antibodies or soluble VEGF receptor, mFlt(1–3)-IgG, results in smaller kidneys and increased numbers of avascular glomeruli (27,31 ). The 5' flanking promoter region of the VEGF and Flt1 genes contain the consensus HRE (5'-CGTG-3'), and HIF have been shown to induce VEGF and Flt1 mRNA and protein in vivo and in vitro in hypoxic conditions (16,17,32 ). Flk1 expression has also been shown to be induced in hypoxia by HIF-2 (33). Furthermore, metanephric organ culture experiments show that hypoxic induction of VEGF activity is important for kidney endothelial cell differentiation. Embryonic rat kidney explants grown in hypoxic culture conditions (3% oxygen) develop endothelial cells, a process that is inhibited by anti-VEGF antibodies (34). Normoxic cultures do not develop endothelial cells unless grown in media supplemented with VEGF. Endothelial cells that contribute to glomerular capillaries also express Tie-1, and organ cultured kidneys from Tie1/LacZ mice sustain endothelial development if grown in hypoxic conditions but not in normoxia (35). Tie-2–expressing endothelial cells are also found in capillaries in developing kidney cortex, including glomeruli (36), and Ang-2 is expressed in large renal blood vessels, maturing tubules, and mesangial cells of developing kidney (37). Hypoxia stimulates production of Ang-2 mRNA and protein in cultured mesangial cells (38).
Taken together, these data suggest that hypoxia-driven gene expression plays a prominent role in development of the renal vasculature in organ-cultured metanephroi. Because the nephrogenic cortex of developing kidneys in situ is undergoing initial vascular assembly and cells are rapidly dividing (39), this area might also be experiencing transient hypoxia in vivo. We hypothesized that the compacted cells within the developing nephrons, including podocytes, are especially susceptible to microenvironmental hypoxia. Such oxygen gradients and localized regions of hypoxia might stabilize HIF, which then stimulate transcription of VEGF and other factors required for kidney development. To explore this possibility, we examined developing mouse kidneys for the expression of HIF-1, HIF-2, and HIF-1 and found HIF mRNA and protein distribution to coincide with sites of VEGF production. We also discovered that hypoxia induced HIF-1 and HIF-2 protein stabilization as well as VEGF mRNA expression by embryonic kidneys in vitro.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and HIF-1 RT-PCR were designed from the mouse mRNA sequences: HIF-1 (accession number, NM 010431) forward primer 5'-AGCCCTAGATGGCTTTGTGA-3' and reverse primer 5'-TATCGAGGCTGTGTCGACTG-3'; HIF-1 (accession number XM 131029) forward primer 5'-CAGAGGGCTATTAAGCGACG-3' and reverse primer 5'-ATACAGTGTGCTCCCGAACC-3'. Primers for HIF-2 RT-PCR were as described previously (12): HIF-2 forward primer 5'-AATGACAGCTGACAAGGAGAAAAA-3' and reverse primer 5'-GAGTGAAGTCAAAGATGCTGTGTC–3'. Total RNA was isolated from embryonic day 14, and newborn wild-type CD-1 mice with the RNAwiz reagent (Ambion, Inc., Austin, TX) according to the manufacturer’s protocol. One-microgram RNA aliquots underwent reverse transcription for 45 min at 48°C followed by 45 rounds of PCR under the following conditions: 30-s denaturing at 94°C, 1-min primer annealing at 60°C, and 2-min primer extension at 68°C with a final extension period of 7 min. RT-PCR products were electrophoresed through a 1% agarose gel and visualized with ethidium bromide. Bands containing the cDNA were excised from the gel, inserted into the TOPO-pCR2.1 vector (Invitrogen Corp., Carlsbad, CA), and cloned according to the TOPO cloning procedure. The cDNAs were sequenced by the University of Kansas Medical Center Biotechnology Core Facility.
Northern Blot Analyses
Sense and antisense probes for Northern blots and in situ hybridization experiments were transcribed from the HIF-1 and HIF-2 cDNAs generated by RT-PCR. First, cDNAs were purified from 1% agarose gels and ligated into the pGEM-T Easy vector (Promega Corp., Madison, WI), which provides flanking SP6 and T7 RNA polymerase promoters. Riboprobes were then transcribed and biotinylated with reagents from the Maxiscript and BrightStar Psoralen-Biotin kits (Ambion). Briefly, plasmids for SP6 RNA polymerase transcription (sense probe) were linearized by digestion with NcoI and those for T7 transcription were linearized with SacI. The DNA templates were incubated with the appropriate RNA polymerase and NTP for 1 h at 37°C. After ethanol precipitation, purified RNA products were crosslinked to the Psoralen-Biotin reagent by UV irradiation. Poly A+ mRNA was isolated from newborn mouse kidney total RNA, and aliquots (1 µg) were resolved on a 1% agarose-formaldehyde gel, transferred to a nylon membrane by capillary action, and UV cross-linked. Nylon membranes were incubated in UltraHyb buffer (Ambion) containing sense or antisense probe (0.1 nM) for 16 h at 71°C. After hybridization, membranes were washed for 30 min in blocking buffer, followed by a 30-min wash in blocking buffer containing Streptavidin-alkaline phosphatase (0.1 µg/ml), three 15-min washes in wash buffer, and two 2-min washes in assay buffer. The CDP-Star chemiluminescence reagent was applied for 5 min, and the membranes were exposed to film. Blocking, wash, and assay buffers were purchased from Ambion.
In Situ Hybridization
Newborn kidney sections were fixed overnight in 10% formalin, dehydrated, and embedded in paraffin. All reagents and buffers for in situ hybridization were purchased from Ambion. Five-micron-thick sections were covered with proteinase K (20 µg/ml in 0.05 M Tris, pH 7.5) and incubated for 30 min at 37°C. After three 5-min washes in 0.05 M Tris, blocking buffer (100 mM Tris/100 mM NaCl, 1% BSA) was applied to sections for 30 min at 37°C. Sense and antisense probes were diluted in hybridization buffer (0.1 nM) and hybridized to sections for 3 h at 65°C. Two 10-min washes in 2x wash buffer and two 5-min washes in 1x wash buffer cleared the slides of unhybridized probe. Streptavidin-alkaline phosphatase conjugate, diluted 1:1000 in 0.05 M Tris, was incubated on slides at 37°C for 30 min, and after three 5-min washes in 1x Tris, was detected by addition of NBT/BCIP solution at 37°C. Color development was monitored closely and terminated by immersing slides in dH2O.
Immunoprecipitations and Western Blot Analyses
Newborn mouse kidneys were Dounce homogenized in RIPA buffer (50 mM Tris, pH 8.0, 150 mM NaCl, 1.0% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS; 3 µl/mg tissue), supplemented with protease inhibitors (10 µg/ml PMSF, 30 µl/ml aprotinin, 10 µ/ml sodium orthovanidate [1 M], 2 µg/ml pepstatin A, 2 µg/ml leupeptin, and 2 µg/ml antipain), and incubated on ice for 15 min. Cellular debris was pelleted by centrifugation at 15,000 x g for 25 min at 4°C, and the supernatant was cleared by incubation with protein A agarose (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) (1 µl/10µl lysate) for 15 min on ice and a final centrifugation at 15,000 x g for 5 min at 4°C. Monoclonal HIF-1, polyclonal HIF-2, or HIF-1 (all from Novus Biologicals, Littleton, CO) were added to the lysates (1 µg/100 µl lysate) and incubated on ice for 1 h. Protein A agarose (1 µg/10 µl lysate) was added to the reaction and incubated with shaking for 1 h at 4°C. The agarose was pelleted by centrifugation at 10,000 x g for 30 s at 4°C, washed four times with ice-cold RIPA buffer, and resuspended in 1x reducing SDS-sample buffer. The immunoprecipitated proteins were resolved by 7.5% SDS-PAGE, transferred to nitrocellulose, and blocked in 5% nonfat dried milk, 0.1% Tween-20 in PBS. Blots were incubated in primary antibody (1:100) overnight, diluted in blocking buffer, at 4°C. After washes in blocking buffer, blots were incubated in HRP-conjugated secondary antibodies (diluted 1:1000 in blocking buffer) for 1 h at room temperature. Membranes were washed in three changes of 0.1% Tween-20 in PBS, and the bands were visualized by ECL (Amersham Pharmacia Biotech, Piscataway, NJ). Biotinylated broad-range molecular weight markers (Bio-Rad Laboratories, Hercules, CA) were resolved in the gels to size the proteins.
Immunohistochemical Analyses
For HIF-1 immunohistochemistry, newborn kidneys were frozen immediately after dissection in OCT in isopentane chilled in a dry-ice/acetone bath. Sections (6-µm-thick) were fixed in 100% methanol on ice. For HIF-2, anesthetized newborn mice were perfused through the right ventricle with PBS followed by 4% paraformaldehyde. Kidneys were dissected immediately and fixed in 4% paraformaldehyde for 2 h on ice, embedded in paraffin, and sectioned at 6 µm. Sections were quenched for endogenous peroxidase activities by incubating slides in 3% H2O2 in methanol for 15 min. Nonspecific protein interactions were blocked by applying 10% normal goat serum in PBS. HIF-1 and HIF-2 rabbit polyclonal antibodies (Novus Biologicals) were diluted 1:100 in PBS and incubated on sections for 1 h at room temperature. After three washes in ice-cold PBS, sections were treated with secondary biotinylated anti-rabbit antibodies (15 µg/ml) for 30 min at room temperature and then with streptavidin-HRP conjugate for another 30 min at room temperature. Slides were color developed with DAB. Control experiments were performed with nonspecific rabbit IgG in place of the primary antibody.
Hypoxyprobe-1 Labeling
Pimonidazole hydrochloride (Hypoxyprobe-1; Chemicon International Inc., Temucula, CA) was injected intraperitoneally into newborn mice (200 mg/kg). One hour after injection, kidneys were dissected and immediately frozen in OCT in isopentane in an acetone/dry-ice bath. Cryostat sections were cut at a thickness of 6 µm and fixed in ice-cold methanol. Sections were blocked in 10% goat serum for 30 min at room temperature and then incubated in Hypoxyprobe-1 monoclonal antibody (1:50) for 1 h. After three washes in PBS, fluorescein-conjugated goat anti-mouse antibodies (1:1000) were applied for 1 h. Sections were then permanently mounted using the Prolong anti-fade reagent (Molecular Probes Inc., Eugene, OR). Control experiments were performed by injecting saline in place of Hypoxyprobe-1, and kidney sections were labeled exactly as described above. Results shown are representative from injections of Hypoxyprobe-1 or saline into multiple newborns from five separate litters born during a several-month period. Additionally, four mature mice received injections of Hypoxyprobe-1, two received saline, and kidney tissues were processed as described.
Embryonic Kidney Organ Culture
Embryos were extracted from timed-pregnant female mice at embryonic day 12 (E12). Metanephroi were dissected and transferred to transparent PET membrane cell culture inserts (0.4-µm pore size; BD Biosciences, San Jose, CA) in a six-well culture plate. Approximately 1 ml of organ culture media (Dulbecco modified Eagle medium supplemented with 10% calf serum, 50 µg/ml L-glutamine, 1% Matrigel [BD Biosciences], 50 U/ml penicillin, and 50 µg/ml streptomycin) was added to the lower chamber of each well. Cultures were maintained at 37°C in a humidified chamber containing 5% CO2 and 95% air (approximately 20% O2). For hypoxic cultures, nitrogen was used to displace air until O2 concentration was either 2% or 5%.
HIF Protein/VEGF mRNA Measurements
Two sets of experiments were performed: one in which E12 kidneys were maintained at either 20% or 5% O2 for 5 d, and another in which kidneys were cultured for 5 d at 20% O2 and then moved to an atmosphere of 2% O2 for 6 h. After incubations, separate kidneys (three per experimental condition) were vortexed for approximately 1 min in 1 ml of Trizol (Invitrogen Corp.). RNA and protein were isolated from each kidney as described in the manufacturer’s protocol. Briefly, after addition of chloroform to Trizol, RNA was precipitated from aqueous phase and protein from the organic phase with isopropanol. RNA and protein concentrations were determined spectrophotometrically and with a Modified Lowry Assay, respectively. Equal amounts of protein from each condition were resolved by 7.5% SDS-PAGE (20 to 50 µg, depending on total protein yield) and transferred to nitrocellulose. Western blots were performed as described previously, and total band densities were measured using an NIH Image (V1.61) program. VEGF mRNA concentrations were measured using the Quantikine mRNA colorimetric quantification kit (R&D Systems, Minneapolis, MN) according to manufacturer’s protocol. Total RNA samples (1 µg) were hybridized for 1 h in a 65°C water bath to two oligonucleotide probes: a biotinylated VEGF capture probe and a digoxigenin-labeled detection probe. This solution was then transferred to a streptavidin-coated microplate and incubated for 1 h at room temperature to immobilize the VEGF mRNA. After clearing the wells of unbound nonspecific RNAs by washing, polyclonal anti-digoxigenin alkaline phosphatase–conjugated antibodies were added to wells for 1 h. Substrate solution and amplifier were added for color development. Color intensity, which was proportional to the amount of VEGF mRNA captured, was by measured by reading the absorbance at 490 nm. A standard curve was generated for each experiment after manufacturer’s procedures to determine VEGF mRNA concentrations in organ culture samples.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, HIF-2, and HIF-1. In each case, we obtained a cDNA of the predicted size (467 bp, 407 bp, and 519 bp, respectively) (Figure 1a). The nucleotide sequences of the RT-PCR products were found to be identical to those reported for each transcript. By Northern blot analysis, newborn mouse kidneys contained two transcripts for HIF-1 (approximately 4.5 and 5.5 kb) and HIF-2 (approximately 5.0 and 6.2kb) (Figure 1b).
|
and HIF-2 mRNA were localized in newborn mouse kidneys by non-isotopic in situ hybridization. The tissue distribution patterns were similar for both transcripts, with concentrated expression in the nephrogenic zone of the extreme outer cortex and diminished expression in the more mature, inner cortex (Figures 2a and 3a). Intense signals for both HIF-1 and HIF-2 mRNAs were detected in early nephric figures (Figures 2a and 3b) and particularly in podocytes of immature glomeruli (Figures 2c and 3c). Proximal convoluted tubules in the inner cortex generally showed weak expression of each transcript (Figures 2a, 3a, and 3b). In the medulla, collecting duct epithelia and interstitial cells were positive for both HIF-1 and HIF-2 (Figures 2b and 3d). Experiments performed with sense probes as controls were completely negative for both HIF-1 (Figure 2d) and HIF-2 (Figure 3e).
|
|
antibodies were positive on Western blots for HIF-1 and HIF-1 (Figure 4), demonstrating the presence of HIF-1/1 heterodimers. Similarly, Western blots of HIF-2 immunoprecipitates were positive for HIF-1 and HIF-2 (Figure 4), indicating the presence of intact HIF-2 in newborn kidney lysates. Further evidence for the presence of HIF-1 and HIF-2 came from experiments showing that proteins immunoprecipitated with HIF-1 antibodies were positive on Western blots for HIF-1, HIF-1, and HIF-2 (Figure 4). For controls, proteins immunoprecipitated with nonspecific rabbit IgG were probed on Western blots with HIF-1, HIF-2, and HIF-1 antibodies and were negative in each case (Figure 4).
|
and HIF-1 antibodies. HIF-1 protein was widely expressed in newborn kidneys, and nuclear staining was seen in glomeruli and all tubular segments (Figure 5). On the other hand, HIF-2 was localized selectively to the nephrogenic zone of the outer cortex, where intense immunolabeling was detected primarily in nuclei of early nephric figures (Figure 6a). Many tubular segments of the cortex displayed a finely granular, cytoplasmic pattern for HIF-2, in addition to nuclear staining (Figure 6a). Higher magnification views of developing nephrons showed strong nuclear localization of HIF-2 primarily in developing podocytes of comma-shaped and S-shaped figures (Figure 6b) as well as capillary loop stage glomeruli (Figure 6a). In maturing glomeruli located deeper in the cortex, podocyte nuclei were also positive for HIF-2 (Figure 6a). No staining was observed in control tissue labeled with nonspecific IgG (Figure 6c). We were not able to immunolocalize HIF-1 with the monoclonal antibody and histochemical techniques used here.
|
|
|
Protein and VEGF mRNA Expression in Response to Hypoxia In Vitro protein stability and VEGF mRNA transcription in metanephric organ culture. After 5 d in culture, E12 kidneys expressed HIF-1 and HIF-2 proteins and VEGF mRNA in the presence of 20% oxygen, but expression of each was significantly higher in cultures maintained at 5% oxygen (Figure 8, a and b). Similarly, organ-cultured kidneys contained higher levels of HIF-1 and HIF-2 proteins and VEGF mRNA when shifted from a normoxic to hypoxic environment (2% oxygen) for 6 h, when compared with those maintained at normoxia (Figure 9, a and b).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and HIF-2 mRNAs are expressed in mature human and mouse kidneys (40) and that 4 to 6-wk-old mouse kidneys express HIF-1 protein in normoxia that elevates in hypoxia (41). The expression patterns of HIF during kidney development, however, had not been examined in detail previously. The data presented here reveal that high levels of HIF mRNA and protein expression occur in developing cortex and medulla, especially in podocytes and collecting ducts, which are sites of robust VEGF production. Of these regions of HIF expression, however, only the medulla and inner cortex reacted strongly with Hypoxyprobe-1 and can therefore be considered to be severely hypoxic. Furthermore, our results show that HIF-1 and HIF-2 proteins are expressed in embryonic kidney organ cultures in normoxia, are further stimulated by hypoxia, and that VEGF mRNA levels also increase during hypoxic conditions. These findings suggest that HIF activation may stimulate production of VEGF in developing kidneys and that both hypoxia-dependent and -independent modes of HIF activation are probably occurring.
Northern blot analyses of human tissues have revealed a message size of approximately 4.4 kb for HIF-1 and approximately 5.8 kb for HIF-2 mRNA (40). Our analysis of newborn mouse kidney RNA is in rough agreement with these data, but our blots also indicate that two transcripts were produced from both the HIF-1 and HIF-2 genes. The murine HIF-1 gene has previously been shown to be under the control of two distinct promoters, resulting in two mRNA products (42). One transcript is nearly ubiquitously expressed, but the other is tissue-specific and found primarily in kidney, tongue, stomach, and intestine (42). Here we report by Northern analyses the size of the two HIF-1 mRNA products in mouse kidneys to be approximately 4.5 and 5.5 kb. We also discovered two mRNA products for HIF-2, approximately 5.0 and 6.2 kb, suggesting that the HIF-2 gene might be structurally similar to HIF-1. The functions of the alternate HIF- transcripts and factors regulating their expression have yet to be determined.
To date, few studies have demonstrated HIF- protein expression in normoxic or nondiseased tissues (41,43 ). The mice used in our experiments were not undergoing experimentally imposed hypoxic stress, yet HIF-1 and HIF-2 proteins were readily detectable during normal kidney development. As determined by Western blotting, HIF-1 was co-immunoprecipitated from newborn kidney lysates with anti-HIF-1 and HIF-2 antibodies, demonstrating that both HIF-1 and HIF-2 heterodimers were present. HIF-1 antibodies also pulled down both subunits from kidney lysates, providing further evidence for the formation of stable HIF heterodimers.
We localized HIF-1 and HIF-2 mRNA expression by in situ hybridization and found a similar pattern for each transcript. Previously, in situ hybridization of adult mouse kidney showed glomerular expression of HIF-2, but cellular definition was not resolvable with the 33P-labeled riboprobe technique used (40). HIF-2 was originally described as an endothelial-specific transcription factor (40) but subsequent studies have shown expression in non-endothelial cells in culture (44). Our results further confirm that HIF-2 is expressed by non-endothelial cell types in vivo, including glomerular and tubular epithelium and mesenchymal cells of developing kidneys. Although posttranslational modifications are largely responsible for regulating HIF- availability and function in hypoxic cells (6), strong evidence exists for HIF-1 mRNA induction in response to hypoxia (45–47 ). We observed intense antisense hybridization signals for HIF-1 and HIF-2 in the medulla, particularly in the collecting duct epithelium, which is a site of VEGF secretion. Oxygen tensions are exceedingly low in the mature renal medulla (10 to 20 mmHg) (48–50 ), which is also consistent with our data showing prominent Hypoxyprobe-1 labeling in this region of the kidney (discussed further below). These observations suggest that hypoxia in the developing medulla is associated with HIF-1 and HIF-2 transcription. In addition to the medulla, the developing renal cortex showed intense antisense hybridization signals for HIF-1 and HIF-2, and, in developing glomeruli, podocytes showed particularly high levels of these mRNAs.
Immunohistochemical analysis of HIF-1 protein in mature mouse kidney has shown normoxic expression in distal convoluted tubules (41), whereas another study has demonstrated HIF-1 and HIF-2 expression in mature rat kidney only after hypoxic stimulus (51). To determine which cell types in developing mouse kidney contained HIF-2, we looked for overlapping protein expression patterns for HIF-1 and HIF-2 by immunohistochemistry. HIF-1 protein stability is not dependent on oxygen concentration, and this HIF subunit is constitutively found in the nucleus of most cell types, which is consistent with the widespread nuclear staining observed in our experiments. All tubular segments and glomerular podocytes were positive for HIF-1. HIF-2 protein, however, displayed a much more restricted expression pattern. Nuclear HIF-2 staining was observed in developing nephrons of the outer cortex, a distribution that is consistent with that for HIF-2 mRNA. Nuclei in developing podocytes of immature nephrons were intensely positive for HIF-2, making HIF-2 a candidate regulator of VEGF transcription and glomerular capillary development.
Whether hypoxia or other conditions are chiefly responsible for HIF stabilization in developing glomeruli is unclear, however. Our Hypoxyprobe-1 immunofluorescence experiments showed only weak labeling in regions of the outer cortex that contained nuclear HIF-2 protein, suggesting that factors other than hypoxia may promote HIF-2 stabilization in these cells. On the other hand, Hypoxyprobe-1 forms protein adducts only in severely hypoxic cells (with half-maximal labeling of cells at O2 concentrations as low as 0.1 to 0.3%) (52), whereas HIF- protein stabilization occurs at oxygen concentrations well above the Hypoxyprobe-1 binding range (53). Therefore, relatively mild hypoxia might contribute to sufficient HIF heterodimer formation that could be physiologically important for activation of VEGF and other hypoxia-responsive genes. Alternatively, cortical HIF induction might be regulated by factors unrelated to hypoxia. Candidates for hypoxia-independent activation of HIF include insulin, IGF-1, nitric oxide, IL-1, and TNF-. Insulin has recently been shown to induce HIF-1 protein accumulation, binding of HIF-1 to DNA, and reporter gene expression in cultured human retinal epithelial cells through a PI3K/TOR–dependent pathway (54). Insulin-induced HIF-1 activity in these cells results in increased VEGF transcription. After cardiac arrest and resuscitation in rats, HIF-1 expression in cerebral cortex persists beyond reoxygenation and is accompanied by an increase in insulin-like growth factor (IGF-1) (55). IGF-1 induces HIF-1 expression in neurons in vitro and in vivo in rats receiving IGF-1 infusion. Additionally, nitric oxide has been shown to stimulate HIF-1 accumulation (56) and the cytokines interleukin-1 (IL-1) and tumor necrosis factor (TNF-) also induce HIF activity (57). Any or all of these factors may operate synergistically with mild hypoxia to activate HIF in the developing kidney cortex, but these possibilities require further investigation.
In addition to developing glomeruli, many developing tubules were also immunoreactive for HIF-2, with staining observed in nuclei as well as cytoplasm. We believe that the cytoplasmic distribution might represent HIF-2 protein degradation by proteasomes. Unlike glomeruli, several developing and mature tubular segments were strongly positive for Hypoxyprobe-1, suggesting that hypoxia may indeed induce HIF-2 stabilization in these cells. Angiopoietin-2 is expressed in developing tubules (37) and might therefore be regulated by the extreme hypoxia and HIF localization observed here.
Our metanephric organ culture experiments showed that HIF-1 and HIF-2 proteins and VEGF mRNA were all expressed in cultures maintained at 20% oxygen. However, exposing kidney explants to 5% oxygen for 5 d or 2% oxygen for 6 h caused increases in HIF protein and VEGF mRNA expression. Nevertheless, the presence of HIF- proteins in metanephroi maintained at 20% oxygen again strongly suggests that hypoxia-independent mechanisms can induce stabilization of these proteins in kidneys. Previous studies have shown that vascular smooth muscle cells contain stabilized HIF-1 protein when cultured in serum at normoxia, but stabilized proteins are degraded upon serum removal (58). Because kidneys in our experiments were also cultured in serum, perhaps serum stimulated HIF- protein stabilization through a hypoxia-independent mechanism. Regardless, our organ culture results also showed that hypoxia strongly induced HIF- protein and VEGF expression in kidneys in vitro. Together with our HIF immunolocalization findings on kidney sections, we believe that both hypoxia-dependent and hypoxia-independent mechanisms are probably operative in vivo.
On the basis of data presented here, we propose that metanephric vasculogenesis may be initiated, in part, by HIF-1 and HIF-2 induced gene expression. Physiologically important amounts of HIF protein may be stabilized by severe or mild hypoxia and/or by other mechanisms. By whatever means, upregulated HIF-1 and HIF-2 protein stability could stimulate production of VEGF and other HIF-inducible gene products. HIF stabilization specifically in immature podocytes might therefore represent a critical initial signal for the inception of glomerular capillary morphogenesis, and we are currently testing this hypothesis.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and developmentally expressed in blood vessels. Mech Dev 63: 51–60, 1997[CrossRef][Medline]
to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation. Science 292: 468–472, 2001 by the ubiquitin-proteasome pathway. J Biol Chem 274: 6519–6525, 1999. Genes Dev 12: 149–162, 1998 is required for solid tumor formation and embryonic vascularization. EMBO J 17: 3003–3015, 1998
. Blood 92: 2260–2268, 1998 and -2 in Hypoxic and Ischemic Rat Kidneys. J Am Soc Nephrol 13: 1721–1732, 2002 under the influence of nitric oxide. Blood 97: 1009–1015, 2001 and tumor necrosis factor- stimulate DNA binding of hypoxia-inducible factor-1. Blood 94: 1561–1567, 1999 in vascular smooth muscle cells. J Biol Chem 275: 26765–26771, 2000This article has been cited by other articles:
 |
|
 |
|
  K. Brukamp, B. Jim, M. J. Moeller, and V. H. Haase Hypoxia and podocyte-specific Vhlh deletion confer risk of glomerular disease Am J Physiol Renal Physiol, October 1, 2007; 293(4): F1397 - F1407. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Makino, Y. Miyamoto, K. Sawai, K. Mori, M. Mukoyama, K. Nakao, Y. Yoshimasa, and S.-i. Suga Altered Gene Expression Related to Glomerulogenesis and Podocyte Structure in Early Diabetic Nephropathy of db/db Mice and Its Restoration by Pioglitazone. Diabetes, October 1, 2006; 55(10): 2747 - 2756. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. H. Haase Hypoxia-inducible factors in the kidney Am J Physiol Renal Physiol, August 1, 2006; 291(2): F271 - F281. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
10 mmHg). (a) Overview of Nb kidney immunolabeled for Hypoxyprobe-1. Cells in the nephrogenic zone (*) appear diffusely but only weakly positive for Hypoxyprobe-1. Many developing and maturing tubules (arrows) show strong fluorescence. (b) High-power view of medulla shows Hypoxyprobe-1 immunofluorescence in collecting ducts. (c) Mice injected with saline as controls and processed for Hypoxyprobe-1 immunofluorescence are negative. (d) Section from cortex of adult mouse that received Hypoxyprobe-1 injection. No specific binding is observed. Fields shown in panels a, c, and d were all taken at the same magnification.
