FASEB J.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published as doi: 10.1096/fj.06-6280com.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
fj.06-6280comv1
21/3/700    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wotzlaw, C.
Right arrow Articles by Fandrey, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wotzlaw, C.
Right arrow Articles by Fandrey, J.
(The FASEB Journal. 2007;21:700-707.)
© 2007 FASEB

Optical analysis of the HIF-1 complex in living cells by FRET and FRAP

Christoph Wotzlaw*, Teresa Otto*, Utta Berchner-Pfannschmidt*, Eric Metzen{dagger}, Helmut Acker* and Joachim Fandrey*,1

* Institut für Physiologie, Universität Duisburg-Essen, Essen, Germany; and

{dagger} Institut für Physiologie, Universität zu Lübeck, Lübeck, Germany

1Correspondence: Institut für Physiologie, Universität Duisburg-Essen, Hufelandstrasse 55, D-45122 Essen, Germany. E-mail: joachim.fandrey{at}uni-due.de


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Hypoxia-inducible factor-1 (HIF-1) coordinates the cellular response to a lack of oxygen by controlling the expression of hypoxia-inducible genes that ensure an adequate energy supply. Assembly of the HIF-1 complex by its oxygen-regulated subunit HIF-1{alpha} and its constitutive ß subunit also known as ARNT is the key event of the cellular genetic response to hypoxia. By two-photon microscopy, we studied HIF-1 assembly in living cells and the mobility of fluorophore-labeled HIF-1 subunits by fluorescence recovery after photobleaching. We found a significantly slower nuclear migration of HIF-1{alpha} than of HIF-1ß, indicating that each subunit can move independently. We applied fluorescence resonance energy transfer to calculate the nanometer distance between {alpha} and ß subunits of the transcriptionally active HIF-1 complex bound to DNA. Both N termini of the fluorophore-labeled HIF-1 subunits were localized as close as 6.2 nm, but even the N and C terminus of the HIF-1 complex were not further apart than 7.4 nm. Our data suggest a more compact 3-dimensional organization of the HIF complex than described so far by 2-dimensional models.—Wotzlaw, C., Otto, T., Berchner-Pfannschmidt, U., Metzen, E., Acker, H., Fandrey, J. Optical analysis of the HIF-1 complex in living cells by FRET and FRAP.


Key Words: fluorescence resonance energy transfer • fluorescence recovery after photobleaching • hypoxia-inducible factor-1 • oxygen sensing


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
THE LACK OF OXYGEN, OR HYPOXIA, ENDANGERS the function and survival of mammalian cells by limiting mitochondrial oxidative phosphorylation to yield ATP. To ensure a sufficient supply of ATP under hypoxia, the transcription factor complex hypoxia-inducible factor-1 (HIF-1) coordinates the expression of genes in which protein products play key roles in vascular reactivity and remodeling, blood oxygen capacity, angiogenesis, glucose, and energy metabolism, as well as cell proliferation and survival (1 , 2) . Human HIF-1 is composed of an 826aa HIF-1{alpha} subunit and a 789aa HIF-1ß subunit, also known as ARNT, both of which are members of the basic helix-loop-helix (bHLH) and of the Per, ARNT, Sim (PAS) domain containing family of transcription factors (3) . Within this family, a protein that is closely related to HIF-1{alpha} in terms of structure and function has been termed HIF-2{alpha} (4) . In contrast to the constitutively expressed nuclear HIF-1ß, the oxygen-labile HIF-1{alpha} confers hypoxic inducibility to the HIF-1 complex. Post-translational hydroxylation of two proline residues by three oxygen-dependent prolyl hydroxylases (PHD) tags HIF-1{alpha} for recognition by the von Hippel-Lindau protein (pVHL), and initiates polyubiquitination and subsequent proteasomal degradation under high oxygen tension (PO2) (5 , 6) . Under hypoxia, PHD activity ceases, which allows HIF-1{alpha} to accumulate and translocate into the nucleus, where it dimerizes with HIF-1ß to bind to hypoxia response elements in the promoter and enhancer regions of hypoxia-inducible genes (7) .

While the bHLH domain is required for DNA binding of HIF-1, two PAS domains, termed PAS A and PAS B, are essential for dimerization of HIF-{alpha} subunits with HIF-1ß (8) . X-ray structure analysis of the complete HIF-{alpha} and HIF-1ß subunits has not yet been reported. Recently, however, some studies have revealed the structure of the PAS B domains of HIF-2{alpha} and the antiparallel orientation of both PAS B domains in the HIF-2{alpha}/HIF-1ß dimer (9 , 10) . These findings shed new light on the potential organization of the HIF complex and may well extend to HIF-1{alpha}/HIF-1ß. However, these authors also pointed out that their in vitro findings need to be corroborated in living cells.

In this study we determine the assembly of the HIF-1 complex in living cells under well-defined O2 conditions by a microscopic approach. For that purpose, a special cell chamber was designed that was mounted on the microscope stage and continuously equilibrated with the desired PO2 (Fig. 1 ). Distribution and mobility of fluorophore-labeled HIF-1 subunits within the nucleus were determined by two-photon laser scanning microscopy (2P-LSM), where excitation of the fluorophore is limited to the focal plane and allows optical sectioning of cells. In addition, compared with standard fluorescence microscopy systems, near-infrared 2P excitation reduces the generation of reactive oxygen species (ROS) during scanning by minimizing photoreduction of flavin-containing oxidases (11) . ROS have been found to affect HIF-1 activation, and light-induced ROS generation is likely to interfere with HIF-1 assembly (12 13 14) . Fluorescence recovery after photobleaching (FRAP) allowed us to determine differences in mobility of HIF-1{alpha}/ß, whereas fluorescence resonance energy transfer (FRET) in living cells under hypoxic conditions was used for HIF-1{alpha}/ß protein-protein interaction analysis. Variation of the position of the fluorophores for both subunits allowed us to calculate the distances between the fluorophore-labeled HIF-1 subunits, suggesting a close neighborhood of the DNA-bound N terminus and the C terminus of the HIF-1 complex. These in vivo findings are fully compatible with the above-mentioned organization of the HIF-1 complex (10) , and suggest a more compact assembly than previous 2-dimensional models derived from in vitro interaction studies.


Figure 1
View larger version (96K):
[in this window]
[in a new window]

  		Figure 1. A) Cell culture chamber mounted on the microscopy stage. B) Removed top cover enables a view into the chamber with a cell culture dish in position.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Cell culture and transient transfections
Human osteosarcoma cells (U2OS) were grown in Dulbecco’s modified Eagle medium (DMEM; Life Technologies, Inc., Karlsruhe, Germany) supplemented with 10% fetal calf serum (Life Technologies), 100 U/ml penicillin, 100 µg/ml streptomycin sulfate (Sigma, Munich, Germany), and 2 mM L-glutamine. Cells were grown in 24-well dishes for luciferase reporter gene assays, in 6-well dishes for Western blot, and in 35 mm-diameter dishes (Wellco, Amsterdam, Netherlands) for FRET and FRAP studies. For transient transfections, plasmid DNA was mixed with Fugene6 (Roche, Mannheim, Germany) and suspended in appropriate volumes of DMEM for each culture dish. Cells were transfected with the vectors for 24 h and medium was renewed for another 24 h, then again 1 h before starting the experiments. Hypoxic incubations were done in an atmosphere of 1% O2/5% CO2/94% N2 at 37°C in a chamber on the microscope stage (Fig. 1 ; Luigs&Neumann, Ratingen, Germany). All other hypoxic incubations were done in a hypoxia workstation in 1% O2/5% CO2/94% N2 (Ruskinn Technology, Leeds, UK).

Plasmids
HIF-1{alpha} full-length and HIF-1{alpha} (1–683aa) coding sequences were amplified by polymerase chain reaction (PCR) using pcDNA3-HIF-1{alpha} as matrix. Appropriate restriction endonuclease sites were attached to the oligonucleotide primers. The products were ligated into pECFP-C1 and pECFP-N1 (Clontech, Heidelberg, Germany), respectively. Likewise, human HIF-1ß coding sequence from pcDNA3-HIF-1ß was introduced into pEYFP-N1 (Clontech, Palo Alto, CA, USA). Mouse EYFP-HIF-1ß, a kind gift from L. Poellinger (Stockholm, Sweden), was described earlier (15) . The designation of the expressed fusion proteins reflects the color and position of the fluorescent protein relative to the HIF fusion partner (e.g., transfection with the plasmid pECFP-C1-HIF-1{alpha} leads to the expression of the protein product ECFP-HIF-1{alpha}). Oligonucleotide sequences are available upon request. All expression vectors were sequence verified.

Western blot analysis, luciferase assay, and immunofluorescence
Total protein (70 µg) was used for Western blot analysis, which was performed exactly as described in ref. 16 . For luciferase assays, plasmid pH3-SVL containing three HREs was used (17) , and immunofluorescence was performed as described recently (16) .

Microscopy system
For all microscopic scans, a confocal laser scanning microscopy system (2P-LSM) with a two-photon (2P) laser (Coherent Mira 900 F Ti: Sapphire laser, 30W, 760–920 nm), a one-photon (1P) laser (DPSS, 35 mW, 532 nm), and a monochrome CCD camera (Orca ER 1394, Hamamatsu, Japan) was used as described (12 , 18 19 20) . Images of cells for FRAP and FRET experiments were taken with a resolution of 256 x 256 or 1024 x 1024 pixel. The excitation wavelength for ECFP signal recording was 800 nm; the emission filter was 480/30. Unless otherwise noted, EYFP was excited/bleached with 532 nm and its fluorescence was detected using a 590/60 band-pass filter. For fluorescence data analysis, ImageProPlus V4.5.1 software (Media Cybernetics, Silver Spring, MD, USA) was used.

Analysis of characteristics of chimeric HIF-1{alpha} and HIF-1ß proteins
Transiently transfected U2OS cells were placed into the microscopic chamber and preincubated under normoxic conditions (21% O2/5% CO2) for 1 h. Analysis of intracellular ECFP-HIF-1{alpha}, HIF-1ß-EYFP, and ECFP-HIF-1ß constructs was performed at normoxic or hypoxic conditions after 4 h (1% O2). To analyze the nuclear amount of fusion proteins, 3-dimensional scans with seven optical slices at distances of 700 nm were performed with the two-photon laser. The cellular plane with the highest mean fluorescence was selected for subsequent analysis. EYFP scans were done at 850 nm excitation wavelength and with a 535/30 band-pass filter for fluorescence detection. Oxygen concentration in cell culture medium was measured using a polarographic O2 catheter electrode (Licox, Kiel, Germany).

Fluorescence recovery after photobleaching
For FRAP analysis, cells transiently expressing ECFP-labeled HIF-1{alpha}- and EYFP-labeled HIF-1ß were scanned ("nucleus before bleaching"). An area of 3.6 µm x 3.6 µm was bleached in the center of the nucleus by exposing this area to laser light 10 times with equal laser power. Migration of the fusion proteins back into the bleached area was followed during 22 scans at intervals of 16 s. To correct for the amount of chromophore bleaching by the laser during measurements, cells were scanned in the same manner without previous bleaching of the small area.

Fluorescence resonance energy transfer
To prove protein-protein interaction by FRET, it is mandatory to record fluorescence of the acceptor, EYFP fused to HIF-1ß, after excitation of the donor, ECFP fused toHIF-1{alpha}; after bleaching of the acceptor, there is an increase in fluorescence of the donor when energy can no longer be transferred. From the efficiency of the energy transfer, the distance between the ECFP-labeled HIF-1{alpha} terminus and the EYFP-labeled HIF-1ß terminus was calculated according to ref. 21 . FRET was recorded after an incubation period of 4 h under hypoxic conditions (1% O2) in the chamber on the microscopic stage. After 20 scans, acceptor fluorescence was reduced to ~4–10% of the original value. A second donor scan followed. Analogous scans with HIF-1{alpha}-ECFP- or HIF-1ß-EYFP-transfected cells allowed determination of the influence of the 2P and 1P laser on the fluorescence characteristics of the fusion proteins. Those results were integrated into the calculation of the FRET efficiencies. Background was subtracted from fluorescence intensity values. FRET efficiency E, at which the energy is transferred from the donor to the acceptor molecule, is described by the equation E = 1 – (IDA/ID), where IDA means donor fluorescence intensity before, ID after acceptor bleaching. Calculation of the distance r between donor and acceptor was done using r = Ro [(1/E) – 1]1/6. The Förster distance Ro is the distance between the donor and acceptor, at which half the excitation energy is transferred. Ro was set to 4.92 nm (22) . Regression analysis by a sigmoid function (3 parameters) was performed with SigmaPlot 2001 (SYSTAT Software Inc., Richmond, CA, USA).

Statistical analysis
Statistical analysis was performed using the statistical package of SigmaPlot 2001 applying Student’s t test after ANOVA (SYSTAT Software).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Characteristics of chimeric HIF-1{alpha} and HIF-1ß proteins
U2OS cells transiently transfected with pECFP-C1-HIF-1{alpha} were placed in a purpose-built chamber on the microscope stage (Fig. 1) , and the oxygen fraction was continuously decreased (Fig. 2 A, dash-dotted trace). ECFP fluorescence recorded from single cells by 2P-LSM significantly increased by 77% within 4 h, with decreasing O2 fractions. Fluorescence from HIF-1ß-EYFP (Fig. 2A ) and from cells transfected with pECFP-C1-HIF-1ß (n=9; data not shown) showed no increase after 4 h under hypoxic conditions, excluding the possibility that the ECFP molecule itself was responsible for the hypoxic response. Likewise, no increase in fluorescence of ECFP-HIF-1{alpha} was observed under normoxic conditions (21%O2; n=10; data not shown). Western blot of cell extracts from parallel cultures confirmed the hypoxic accumulation of the ECFP-HIF-1{alpha} fusion protein, indicating physiological O2-dependent regulation of our constructs (Fig. 2B ).


Figure 2
View larger version (18K):
[in this window]
[in a new window]

  		Figure 2. A) Change in fluorescence intensities of fusion proteins accumulated in the nucleus upon reduction of oxygen fraction in the gas phase to 1% O2. The dotted line shows a typical course of the oxygen concentration in the medium of the culture dish within the chamber on the microscope stage at a constant temperature of 37°C. The Clark-type O2 electrode was placed at the bottom of the cell culture dish, close to the cell layer. Solid symbols represent ECFP-HIF-1{alpha} (n=11) and open symbols represent HIF-1ß-EYFP (n=6) fluorescence intensities 24 h after transfection. *ECFP-HIF-1{alpha} fluorescence values 4 h after switching to 1% O2 are significantly different from the values under normoxic conditions (Student’s t test, with P<0.01). B) Western blot analysis of pECFP-C1-HIF-1{alpha}-transfected cells incubated under normoxic (21%) and hypoxic (1%) conditions for 4 h; 70 µg lysate of total cell lysate was loaded per lane. The anti-HIF-1{alpha} antibody (BD Transduction, Lexington, KY, USA) detected both endogenous HIF-1{alpha} and the 27 kDa larger fusion protein ECFP-HIF-1{alpha}. {alpha}-Tubulin (Santa Cruz Biotechnology, Santa Cruz, CA, USA) was used as loading control. Results are representative of three independent experiments.

Nuclear fluorescence distribution of endogenous HIF-1{alpha}/ß immunofluorescence (Fig. 3 A, B) was compared with fluorescence of ECFP-HIF-1{alpha} and HIF-1ß-EYFP fusion proteins transfected separately (Fig. 3C, D ). The more homogeneous fluorescence distribution of single expressed fusion proteins contrasted with the speckle-like nuclear distribution of endogenous HIF-1{alpha}/ß (Fig. 3A, B ). However, cotransfection of both fusion proteins revealed that the speckle-like structure was comparable to the pattern of endogenous HIF-1 subunits. This indicates normal nuclear distribution of the fusion proteins (Fig. 3E, F ). Coexpression of ECFP-labeled HIF-1{alpha} and EYFP-labeled HIF-1ß into cells transfected with a luciferase reporter gene driven by three hypoxia response elements revealed hypoxia-inducible activity and indicated unimpaired DNA binding of the fusion proteins (Table 1 ).


Figure 3
View larger version (55K):
[in this window]
[in a new window]

  		Figure 3. Localization of endogenous and chimeric HIF-1 subunits in U2OS cells under hypoxic conditions (1% O2, 4 h). Images (small inserts) were taken with a fluorescence microscope and visualized using the surface plot technique. Heterogeneous speckle-like distribution was noticed for the endogenous HIF-1{alpha} (A) and HIF-1ß (B) subunits by immunohistochemistry. Fluorescence of separately transfected fusion proteins was homogenously distributed throughout the nucleus (ECFP-HIF-1{alpha}, C; HIF-1ß-EYFP, D). A speckle-like distribution as for endogenous proteins was observed upon cotransfection of both fusion proteins (ECFP-HIF-1{alpha} fluorescence, E; HIF-1ß-EYFP fluorescence, F). Images of fusion protein distribution were taken 48 h after transient transfection. Bar represents 5 µm.


View this table:
[in this window]
[in a new window]

  		Table 1. Transcriptional activity of HIF-1 fusion proteinsa

Analysis of the mobility of HIF-1 subunits
FRAP allows one to determine diffusion capabilities of proteins (23) . U2OS cells were transfected to coexpress EYFP-labeled HIF-1ß- and ECFP-labeled full-length HIF-1{alpha} or an ECFP-labeled HIF-1{alpha} (1–683aa) deletion mutant lacking the C-TAD. After bleaching of a 3.6 µm x 3.6 µm nuclear area to destroy the chromophores of ECFP-HIF-1{alpha} or HIF-1ß-EYFP, respectively, fluorescence recovered, as shown in a representative trace in Fig. 4 A for HIF-1{alpha} and in Fig. 4B for HIF-1ß. Within 51 s the exact heterogeneous distribution pattern was recovered, indicating a targeted redistribution of HIF-1{alpha}/ß fusion proteins (Fig. 4 ; compare before bleaching with t=51 s). In addition, cells were treated under normoxic conditions with dimethyl oxalyl glycine (DMOG), an inhibitor of PHDs and FIH-1 (24) . Fluorescence recovery for both HIF-1 subunits was evaluated for >300 s (Fig. 5 ). A high percentage of both subunits was mobile within the nucleus as indicated by the 100% recovery rate of fluorescence. Calculation of the t50 value (indicating 50% recovery of the fluorescence) resulted in a much higher value of ~65 s for both HIF-1{alpha} constructs than for HIF-1ß (~31 s). Treatment with DMOG did not affect the mobility of either HIF-1{alpha} or HIF-1ß. This indicates that full-length as well as shortened HIF-1{alpha} (1–683aa) fusion proteins move much slower than the HIF-1ß fusion proteins and that inhibition of hydroxylation of HIF-1{alpha} does not affect mobility. Each subunit can move independently of its dimerizing partner.


Figure 4
View larger version (66K):
[in this window]
[in a new window]

  		Figure 4. FRAP analysis of ECFP-labeled HIF-1{alpha}- (A) and EYFP-labeled HIF-1ß (B) to determine their mobility in living U2OS cells under hypoxic conditions (1% O2, 4 h). Nuclear distribution of the fusion proteins was imaged 48 h after transfection ("nucleus before bleaching"). Subsequently, a nuclear area of 3.6 µm x 3.6 µm was bleached and fluorescence recovery was analyzed for each fusion protein independently. Fusion proteins were redirected to exactly the same positions (compare "nucleus before bleaching" with "51 s"). Arrows indicate identical fluorescent structures before and after bleaching. The bar in the left photographs represents 3.6 µm.


Figure 5
View larger version (33K):
[in this window]
[in a new window]

  		Figure 5. Analysis of the mobility of chimeric full-length HIF-1{alpha} and HIF-1ß subunits and of a HIF-1{alpha} (1–683aa) deletion mutant by FRAP. U2OS cells were transiently cotransfected with pECFP-C1-HIF-1{alpha} and pEYFP-N1-HIF-1ß or pECFP-C1-HIF-1{alpha} (1–683aa) and pEYFP-N1-HIF-1ß, respectively. Fluorescence recovery within a bleached 3.6 µm x 3.6 µm area was recorded for 341 s. The fluorescence intensity values within this area directly after bleaching are considered to be 0%, the last value after 341 s or recovery 100%. For t50 value (indicating 50% recovery of fluorescence; horizontal line), a statistically significant difference between the HIF-1{alpha} and HIF-1ß constructs with P < 0.01 was calculated. Values are means ± SD; for each point 5 to 9 cells were analyzed.

Protein-protein interaction analysis
We applied FRET to study the interaction of HIF-1{alpha} and HIF-1ß, which assemble the HIF-1 complex in living cells under hypoxia. According to the theory of FRET, energy from the donor molecule (ECFP-labeled HIF-1{alpha}) is transferred to the acceptor (EYFP-labeled HIF-1ß) if the two molecules come closer than 10 nm. In this case, fluorescence of the acceptor is recorded, although only the donor is excited with the 2P laser (Fig. 6 ). To further prove FRET, the EYFP label of HIF-1ß was bleached, which reduced EYFP fluorescence only, and resulted in an increase in ECFP fluorescence because energy could no longer be transferred. To control for random collisions and dimerization of fluorophores at high concentrations causing false positive FRET signals (25) , cells were transiently cotransfected with pECFP and pEYFP only. All FRET efficiencies determined with fusion proteins of HIF-1 subunits and fluorescent protein were statistically different from this random FRET control (Student’s t test; P<0.001, 15–30 cells randomly selected).


Figure 6
View larger version (25K):
[in this window]
[in a new window]

  		Figure 6. FRET analysis of HIF-1{alpha}/ß dimerization. Either the N or C terminus of the HIF-1 subunits was labeled. The HIF-1{alpha} or HIF-1ß fusion protein are indicated for each subset. Microscopic photographs demonstrate the heterogeneous distribution of the expressed protein. All measurements were done under hypoxia (1% O2, 4 h) in a chamber on the microscope stage. U2OS cells were transiently cotransfected with respective plasmids for 48 h before measurements were taken. FRET efficiency was found to be dependent on the acceptor/donor ratio and reaches a plateau, as expected for clustered donors and acceptors (25) . Therefore, FRET efficiency was determined over a whole range of acceptor/donor ratios and distances were determined when the plateau was reached. The ratio between acceptor fluorescence and donor fluorescence is a microscope system-specific ratio depending on laser characteristics, fluorescence detector properties, and scanning parameters.

Highest FRET efficiency, and thus the closest association of the HIF-1{alpha}/ß fusion protein, was observed when ECFP and EYFP were located at either the N terminus (Fig. 6A ) or the C terminus (Fig. 6B ) of the HIF-1 subunits. FRET efficiency was reproducibly higher for fusion proteins with fluorophores attached to the N terminus (FRET efficiency 20.5% in Fig. 6A vs. 13.4% in Fig. 6B ). Unexpectedly, however, considerable FRET was observed for pairs of fusion proteins, where the fluorophores were placed at opposite ends of the HIF-1 subunits (Fig. 6C, D ). These data indicate that N and C termini of both subunits come close to each other in the DNA-bound HIF-1 complex. Replacing the full-length HIF-1{alpha} coupled to ECFP with its 1–683aa deletion mutant did not significantly change maximum FRET efficiency (Fig. 6C, D ). However, according to the transcriptional activity of the chimeric HIF-1 proteins (Table 1) , the fusion proteins analyzed by FRET are bound to DNA, which will in part determine the organization of the HIF-1 complex. Compared with the specific FRET, the scan results of cells containing ECFP and EYFP only in order to control for random FRET were below 6% FRET efficiency.

Efficiency of HIF-{alpha}/ß FRET was found to depend strongly on the donor/acceptor ratio (Fig. 6) . Therefore, a range of acceptor-to-donor fluorescence ratios was recorded during scanning. and FRET efficiency was analyzed by regression analysis and mathematically fitted to calculate the distance between HIF-1{alpha} and HIF-1ß for each organization of the dimer (Table 2 ). Both N termini of the fluorophore-labeled HIF-1 subunits were colocalized as close as 6.2 nm, but even the N and C termini of the HIF-1 complex were not further apart than 7.4 nm. Using the HIF-1{alpha} (1–683aa) and full-length HIF-1ß labeled in the same way revealed a distance between the N and C termini of the HIF-1 complex of only 6.9 nm. When HIF-1 subunits were labeled at both C termini, their distance was calculated as 6.7 nm.


View this table:
[in this window]
[in a new window]

  		Table 2. Distances between HIF-1{alpha} and HIF-1ß fluorescently labeled at different terminia


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Understanding hypoxic gene activation depends on insights into the assembly and activation process of the HIF-1 complex. Here we have visualized nuclear distribution and mobility of the HIF-1{alpha} and -ß subunits by 2P-LSM, and applied FRET to analyze the protein-protein interaction of both HIF-1 subunits. Complete analysis was done in living cells at well-defined pO2 values in an especially designed temperature controlled chamber on the microscopic stage. The use of two-photon laser excitation of all fluorescently labeled fusion proteins confers an immediate advantage for such in vivo studies. Two-photon laser-induced fluorescence is inherently confocal because excitation is achieved only in a single optical plane (26) . Thus, photobleaching and cellular damage in focal planes above or below the observed molecules is reduced to a minimum. More important with respect to studying HIF activation, however, is the fact that long-wave two-photon laser light avoids phototoxicity, which may result in the generation of ROS, which has been described to interfere with hypoxia-induced HIF activation (12 13 14 , 27) .

2P-LSM has been successfully applied to localize HIF-1{alpha} and -ß as well as cofactors, and to visualize their 3-dimensional distribution in the nuclear compartment (20) . Here fusion constructs of ECFP with HIF-1{alpha} and EYFP with HIF-1ß were used. Both constructs resembled physiological regulation of HIF-1 subunits. First, whereas HIF-1ß-EYFP was constitutively localized in the nucleus, ECFP-HIF-1{alpha} accumulation was hypoxia inducible in our chamber on the microscope stage. Second, transfected fusion proteins bound to DNA irrespective of the fluorophores added to their N or C terminus and also stimulated HIF-1-dependent activity of a luciferase reporter gene controlled by three hypoxia response elements at a magnitude comparable to what has been observed in other studies with fluorescently labeled HIF-1 proteins (15) . Third, heterogeneous nuclear distribution of both subunits, observed on HIF-1{alpha} and HIF-1ß immunofluorescence, could be achieved by cotransfecting both chimeric proteins. At the limit of optical resolution, this appears to be important in order to ensure that the obtained data are physiologically relevant with respect to nuclear mobility of both subunits. Our FRAP experiments revealed that HIF-1ß moves much faster than HIF-1{alpha} within the nuclear compartment. This clearly indicates that both HIF-1 subunits can move separately and therefore do not always form a HIF-1 complex inside the nucleus. Since HIF-1ß has other dimerization partners (28) , those proteins could affect HIF-1ß mobility. We also speculated that a shortened fusion protein, HIF-1{alpha} (1–683aa), which lacks the C-TAD to which p300/CBP and potentially other coactivators like SRC-1 bind (29) , might have a different mobility. However, HIF-1{alpha} (1–683aa) was found to move as fast as the full-length HIF-1{alpha}, disproving the hypothesis that proteins associated with the C-TAD of a HIF-1{alpha} monomer might be responsible for the slower mobility. However, our data provide evidence that upon fluorescence recovery of bleached regions, HIF-1 subunits are redirected to specific locations within the nucleus, indicating that the heterogeneous distribution is not a random process. As shown in Fig. 4 , it was striking how subnuclear fluorescence structures were recovered by HIF-1{alpha}/ß dimers. In fact, these structures were observed only when both fusion proteins were coexpressed (Fig. 3) and both fluorescent subunits were colocalized to the same area with a high Pearson’s correlation coefficient of >0.8 (data not shown).

Our constructs showed physiological regulation and localization within the nucleus when analyzed in vivo. Since optical resolution, however, is at best ~200 nm, light microscopy does not allow one to draw conclusions with respect to protein-protein interaction. To analyze the assembly of the HIF-1 complex, therefore, we applied FRET in combination with two-photon laser excitation. FRET could be successfully established with HIF-1 subunits labeled at the N terminus. Since these constructs were transcriptionally active, we are confident that our analysis provides data of DNA-bound HIF-1 complexes. FRET efficiencies were determined over an entire range of fluorescent acceptor-to-donor ratios. This is important since the ratio affects energy transfer efficiency, as can be seen from the data in Fig. 6 (30) . Mean distances between HIF-1{alpha}/ß were calculated from efficiency values at the plateau of the FRET efficiency curve, where all donor molecules in the analyzed area are saturated with acceptor molecules.

An inherent problem of FRET studies is the overload of cells transfected with fusion proteins, which might cause unspecific FRET due to random collisions and dimerization of the molecules. We used only the minimum amount of DNA for a transfection that generated sufficient fluorescence for FRET. The fact that we had retained, at least in part, physiological, hypoxia-inducible regulation of our fusion proteins (Fig. 2) supports this notion and indicates that we have not overrun endogenous regulation. Moreover, our FRAP data indicate that our fusion proteins were targeted to the exactly the same localization after bleaching. This strongly argues against significant amounts of HIF-1{alpha}/ß-chimera floating through the nucleus and being potentially prone to random collisions, and therefore unspecific FRET.

FRET has been used as a molecular ruler and allowed us to calculate distances between two fluorophores. In the case of N-terminal-labeled HIF-1 subunits, the distance was only 6.2 nm. With labeling of both subunits at their C termini, HIF-1{alpha}/ß were 7.2 nm apart, which suggests that binding to DNA may support the proximity of both N termini. The current model of the HIF complex is based on in vitro data, where interaction between the two subunits was deduced from co-immune precipitation studies. From this model, one would expect that the N and C termini of the HIF-1 subunits are substantially further apart than the two N termini are from each other (9) . Surprisingly, however, based on an unexpected high FRET efficiency between HIF-1{alpha} labeled at the N terminus and HIF-1ß labeled at the C terminus, the distance of these fluorophores was calculated to be only 7.4 nm. Experimental data on the 3-dimensional structure of the HIF-1 complex are not yet available due to the low solubility of HIF-1 monomers (9) . X-ray crystallography has been used to study the interaction of HIF-1{alpha} with pVHL and CBP/p300, but interaction of HIF-1 subunits has not yet been reported (31 32 33) .

However, recent work by Card et al. has provided evidence that the PAS B domains of HIF-{alpha} and HIF-ß that contribute to the dimerization are orientated in an antiparallel way (10) . Using various methods, the authors showed the orientation of the PAS B domains for an {alpha}/ß dimer in vitro, thus providing a new model of the HIF complex assembly. This model fits with our FRET measurements, supporting the notion that N and C termini of HIF-1{alpha} and HIF-1ß are much closer than previously thought. A smaller distance of 7.4 nm was measured with the shortened HIF-1{alpha} (1–683aa), which does not contain the C-TAD for binding coactivator proteins. Obviously, deletion of the C terminus could affect the 3-dimensional structure of HIF-1{alpha} (1–683aa), causing a different FRET efficiency, but it is tempting to speculate that the lack of binding of coactivators to HIF-1{alpha} (1–683aa) may allow the C terminus to come even closer to the N terminus of the HIF-1 complex.

In conclusion, optical analysis of HIF-1 assembly provided important new information about the 3-dimensional organization of the complex. Data from living cells under hypoxic conditions are now available. Our system may serve as a valuable tool for future experiments to manipulate the assembly and activity of HIF-1.


   ACKNOWLEDGMENTS
 
HIF-1{alpha}-pcDNA3 and HIF-1ß-pcDNA3 were generously provided by J. M. Gleadle and P. J. Ratcliffe (University Oxford, Oxford, UK). PH3SVL was a gift from R. Wenger (Zurich, Switzerland) and EYFP-mHIF-1ß was from L. Poellinger (Stockholm, Sweden). This work was supported by a grant from the Deutsche Forschungsgemeinschaft (FA 225/19–2).

Received for publication April 21, 2006. Accepted for publication October 11, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Semenza, G. L. (1998) Hypoxia-inducible factor 1: master regulator of O2 homeostasis. Curr. Opin. Genet. Dev. 8,588-594[CrossRef][Medline]
  2. Höpfl, G., Ogunshola, O., Gassmann, M. (2004) HIFs and tumors—causes and consequences. Am. J. Physiol. 286,R608-R623
  3. Wang, G. L., Jiang, B. H., Rue, E. A., Semenza, G. L. (1995) Hypoxia-inducible factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2 tension. Proc. Natl. Acad. Sci. U. S. A. 92,5510-5514[Abstract/Free Full Text]
  4. Tian, H., McKnight, S. L., Russell, D. W. (1997) Endothelial PAS domain protein 1 (EPAS1), a transcription factor selectively expressed in endothelial cells. Genes Dev. 11,72-82[Abstract/Free Full Text]
  5. Schofield, C. J., Ratcliffe, P. J. (2005) Signalling hypoxia by HIF hydroxylases. Biochem. Biophys. Res. Commun. 338,617-626[CrossRef][Medline]
  6. Fandrey, J., Gorr, T. A., Gassmann, M. (2006) Regulating cellular oxygen sensing by hydroxylation. Cardiovasc. Res. 71,642-651[Abstract/Free Full Text]
  7. Berra, E., Ginouves, A., Pouyssegur, J. (2006) The hypoxia-inducible-factor hydroxylases bring fresh air into hypoxia signalling. EMBO Rep. 7,41-45[CrossRef][Medline]
  8. Semenza, G. L., Agani, F., Booth, G., Forsythe, J., Iyer, N., Jiang, B. H., Leung, S., Roe, R., Wiener, C., Yu, A. (1997) Structural and functional analysis of hypoxia-inducible factor 1. Kidney Int. 51,553-555[Medline]
  9. Erbel, P. J. A., Card, P. B., Karakuzu, O., Bruick, R. K., Gardner, K. H. (2003) Structural basis for PAS domain heterodimerization in the basic helix-loop-helix-PAS transcription factor hypoxia-inducible factor. Proc. Natl. Acad. Sci. U. S. A. 100,15504-15509[Abstract/Free Full Text]
  10. Card, P. B., Erbel, P. J. A., Gardner, K. H. (2005) Structural Basis of ARNT PAS-B dimerization: use of a common beta-sheet interface for hetero- and homodimerization. J. Mol. Biol. 353,664-677[CrossRef][Medline]
  11. Hockberger, P. E., Skimina, T. A., Centonze, V. E., Lavin, C., Chu, S., Dadras, S., Reddy, J. K., White, J. G. (1999) Activation of flavin-containing oxidases underlies light-induced production of H2O2 in mammalian cells. Proc. Natl. Acad. Sci. U. S. A. 96,6255-6260[Abstract/Free Full Text]
  12. Liu, Q., Berchner-Pfannschmidt, U., Moller, U., Brecht, M., Wotzlaw, C., Acker, H., Jungermann, K., Kietzmann, T. (2004) A Fenton reaction at the endoplasmic reticulum is involved in the redox control of hypoxia-inducible gene expression. Proc. Natl. Acad. Sci. U. S. A. 101,4302-4307[Abstract/Free Full Text]
  13. Guzy, R. D., Hoyos, B., Robin, E., Chen, H., Liu, L., Mansfield, K. D., Simon, M. C., Hammerling, U., Schumacker, P. T. (2005) Mitochondrial complex III is required for hypoxia-induced ROS production and cellular oxygen sensing. Cell. Metab. 1,401-408[CrossRef][Medline]
  14. Mansfield, K. D., Guzy, R. D., Pan, Y., Young, R. M., Cash, T. P., Schumacker, P. T., Simon, M. C. (2005) Mitochondrial dysfunction resulting from loss of cytochrome c impairs cellular oxygen sensing and hypoxic HIF-alpha activation. Cell. Metab. 1,393-399[CrossRef][Medline]
  15. Ruas, J. L., Poellinger, L., Pereira, T. (2005) Role of CBP in regulating HIF-1-mediated activation of transcription. J. Cell Sci. 118,301-311[Abstract/Free Full Text]
  16. Berchner-Pfannschmidt, U., Petrat, F., Doege, K., Trinidad, B., Freitag, P., Metzen, E., de Groot, H., Fandrey, J. (2004) Chelation of cellular calcium modulates hypoxia-inducible gene expression through activation of hypoxia-inducible factor-1alpha. J. Biol. Chem. 279,44976-44986[Abstract/Free Full Text]
  17. Rolfs, A., Kvietikova, I., Gassmann, M., Wenger, R. H. (1997) Oxygen-regulated transferrin expression is mediated by hypoxia-inducible factor-1. J. Biol. Chem. 272,20055-20062[Abstract/Free Full Text]
  18. Bestvater, F., Spiess, E., Stobrawa, G., Hacker, M., Feurer, T., Porwol, T., Berchner-Pfannschmidt, U., Wotzlaw, C., Acker, H. (2002) Two-photon fluorescence absorption and emission spectra of dyes relevant for cell imaging. J. Microsc. 208,108-115[Medline]
  19. Metzen, E., Berchner-Pfannschmidt, U., Stengel, P., Marxsen, J. H., Stolze, I., Klinger, M., Huang, W. Q., Wotzlaw, C., Hellwig-Burgel, T., Jelkmann, W., et al (2003) Intracellular localisation of human HIF-1 alpha hydroxylases: implications for oxygen sensing. J. Cell Sci. 116,1319-1326[Abstract/Free Full Text]
  20. Berchner-Pfannschmidt, U., Wotzlaw, C., Merten, E., Acker, H., Fandrey, J. (2004) Visualization of the three-dimensional organization of hypoxia-inducible factor-1alpha and interacting cofactors in subnuclear structures. Biol. Chem. 385,231-237[CrossRef][Medline]
  21. Tramier, M., Gautier, I., Piolot, T., Ravalet, S., Kemnitz, K., Coppey, J., Durieux, C., Mignotte, V., Coppey-Moisan, M. (2002) Picosecond-hetero-FRET microscopy to probe protein-protein interactions in live Cells. Biophys. J. 83,3570-3577
  22. Patterson, G. H., Piston, D. W., Barisas, B. G. (2000) Forster distances between green fluorescent protein pairs. Anal. Biochem. 284,438-440[CrossRef][Medline]
  23. Houtsmuller, A. B., Vermeulen, W. (2001) Macromolecular dynamics in living cell nuclei revealed by fluorescence redistribution after photobleaching. Histochem. Cell Biol. 115,13-21[Medline]
  24. Epstein, A. C., Gleadle, J. M., McNeill, L. A., Hewitson, K. S., O’Rourke, J., Mole, D. R., Mukherji, M., Metzen, E., Wilson, M. I., Dhanda, A., et al (2001) C. elegans EGL-9 and mammalian homologs define a family of dioxygenases that regulate HIF by prolyl hydroxylation. Cell 107,43-54[CrossRef][Medline]
  25. Zacharias, D. A., Violin, J. D., Newton, A. C., Tsien, R. Y. (2002) Partitioning of lipid-modified monomeric GFPs into membrane microdomains of live cells. Science 296,913-916[Abstract/Free Full Text]
  26. Rubart, M. (2004) Two-photon microscopy of cells and tissue. Circ. Res. 95,1154-1166[Abstract/Free Full Text]
  27. Kietzmann, T., Gorlach, A. (2005) Reactive oxygen species in the control of hypoxia-inducible factor-mediated gene expression. Semin. Cell. Dev. Biol. 16,474-486[CrossRef][Medline]
  28. Gu, Y. Z., Hogenesch, J. B., Bradfield, C. A. (2000) The PAS superfamily: sensors of environmental and developmental signals. Annu. Rev. Pharmacol. Toxicol. 40,519-561
  29. Carrero, P., Okamoto, K., Coumailleau, P., O’Brien, S., Tanaka, H., Poellinger, L. (2000) Redox-regulated recruitment of the transcriptional coactivators CREB-binding protein and SRC-1 to hypoxia-inducible factor 1alpha. Mol. Cell. Biol. 20,402-415[Abstract/Free Full Text]
  30. Berney, C., Danuser, G. (2003) FRET or No FRET: a quantitative comparison. Biophys. J. 84,3992-4010
  31. Hon, W. C., Wilson, M. I., Harlos, K., Claridge, T. D., Schofield, C. J., Pugh, C. W., Maxwell, P. H., Ratcliffe, P. J., Stuart, D. I., Jones, E. Y. (2002) Structural basis for the recognition of hydroxyproline in HIF-1 alpha by pVHL. Nature 417,975-978[CrossRef][Medline]
  32. Freedman, S. J., Sun, Z. Y., Poy, F., Kung, A. L., Livingston, D. M., Wagner, G., Eck, M. J. (2002) Structural basis for recruitment of CBP/p300 by hypoxia-inducible factor-1alpha. Proc. Natl. Acad. Sci. U. S. A. 99,5367-5372[Abstract/Free Full Text]
  33. Lee, C., Kim, S. J., Jeong, D. G., Lee, S. M., Ryu, S. E. (2003) Structure of human FIH-1 reveals a unique active site pocket and interaction sites for HIF-1 and von Hippel-Lindau. J. Biol. Chem. 278,7558-7563[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
fj.06-6280comv1
21/3/700    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wotzlaw, C.
Right arrow Articles by Fandrey, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wotzlaw, C.
Right arrow Articles by Fandrey, J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS