Cancer Research SABCS Jordan
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Cancer Research 68, 9212, November 15, 2008. doi: 10.1158/0008-5472.CAN-08-1135
© 2008 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Helczynska, K.
Right arrow Articles by Jirström, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Helczynska, K.
Right arrow Articles by Jirström, K.

Cell, Tumor, and Stem Cell Biology

Hypoxia-Inducible Factor-2{alpha} Correlates to Distant Recurrence and Poor Outcome in Invasive Breast Cancer

Karolina Helczynska1,2,3, Anna-Maria Larsson1,2, Linda Holmquist Mengelbier1,2, Esther Bridges1,2, Erik Fredlund1,2, Signe Borgquist1, Göran Landberg1, Sven Påhlman1,2 and Karin Jirström1

1 Center for Molecular Pathology, Department of Laboratory Medicine, 2 CREATE Health, Lund University and 3 Department of Surgery, University Hospital MAS, Malmö, Sweden

Requests for reprints: Sven Påhlman, Center for Molecular Pathology, Department of Laboratory Medicine, Lund University, University Hospital MAS, Entrance 78, SE-205 02 Malmö, Sweden. Phone: 46-40337427; Fax: 46-40336073; E-mail: sven.pahlman{at}med.lu.se.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
 References
 
Differential regulation as well as target gene specificity of the two hypoxia-inducible factor (HIF)-{alpha} subunits HIF-1{alpha} and HIF-2{alpha} in various tumors and cell lines have been suggested. In breast cancer, the prognostic significance of HIF-1{alpha} is not clear-cut and that of HIF-2{alpha} is largely unknown. Using IHC analyses of HIF-1{alpha}, HIF-2{alpha}, and vascular endothelial growth factor (VEGF) expression in a tissue microarray of invasive breast cancer specimens from 512 patients, we investigated the expression patterns of the 2 HIF-{alpha} subunits in relation to established clinicopathologic variables, VEGF expression, and survival. HIF-1{alpha} and HIF-2{alpha} protein levels and their effect on survival were additionally analyzed in a second cohort of 179 patients. To evaluate the individual role of each subunit in the hypoxic response and induction of VEGF, HIF-{alpha} protein and HIF-{alpha} and VEGF mRNA levels were further studied in cultured breast cancer cells after hypoxic induction and/or knockdown of HIF-{alpha} subunits by siRNA by Western blot and Quantitative Real-Time PCR techniques. We showed that although HIF-1{alpha} and HIF-2{alpha} protein levels in breast cancer specimens were not interrelated, high levels of both HIF-1{alpha} and HIF-2{alpha} associated to high VEGF expression. HIF-2{alpha} expression was an independent prognostic factor associated to reduced recurrence-free and breast cancer–specific survival, whereas HIF-1{alpha} did not exhibit these correlations. In cultured cells, acute hypoxia induced both HIF-proteins. At prolonged hypoxia, HIF-2{alpha} remained accumulated, whereas HIF-1{alpha} protein levels decreased, in agreement with the oxygen level and time-dependent induction of HIFs recently reported in neuroblastoma. [Cancer Res 2008;68(22):9212–20]


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
 References
 
Tissue hypoxia elicits adaptive responses in the affected cells largely governed by the hypoxia inducible factors (HIF). For example, transcription of genes regulating angiogenesis, metabolism, and survival is altered, resulting in a more resilient phenotype (1, 2). In malignant tumors, this phenotype has previously been assigned immature, stem cell–like properties (3, 4). As tumor hypoxia is generally linked to poor prognosis (5–7), we proposed that hypoxia-driven dedifferentiation is one factor behind the development of a more aggressive phenotype (8). In concert with this proposal, several studies have linked raised HIF-1{alpha} levels to poor outcome in various patient subgroups of invasive breast cancer (9–14). However, published results are not unequivocal, with reports proposing different HIF-1{alpha} regulatory pathways other than hypoxia, and opposing prognostic information linked to high HIF-1{alpha} expression (15, 16). Furthermore, a potential role of HIF-2{alpha} in adaptation to hypoxia in breast cancer has not been extensively studied, although the correlation of HIF-2{alpha}– and VEGF-expressing tumor-associated macrophages to poor prognosis has been reported (17).

Recently, we have shown differential temporal and oxygen-dependent stabilization of HIF-1{alpha} and HIF-2{alpha} in neuroblastoma, and we proposed that the acute versus chronic responses to hypoxia are mediated by HIF-1{alpha} and HIF-2{alpha}, respectively (18). The study also suggested an oncogenic, nonhypoxia-driven role of HIF-2{alpha}, showing a significantly impaired survival among patients with tumors containing tumor cells with strong HIF-2{alpha} expression. Given this observation, we have now investigated the potential effect of HIF-2{alpha} protein expression levels on survival in a cohort of patients with invasive breast cancer (n = 512) using tissue microarray (TMA)-based immunohistochemical (IHC) analyses. HIF-2{alpha} protein was evaluated in relation to HIF-1{alpha} and vascular endothelial growth factor (VEGF) expression. No significant correlation between the HIF-{alpha} protein levels was found. However, a significant correlation between HIF-1{alpha} and VEGF, and a less strong but significant correlation between HIF-2{alpha} and VEGF, were observed, suggesting different mechanisms of HIF function. Interestingly, HIF-2{alpha} positivity correlated to distant but not local or regional recurrence. In agreement with this observation HIF-2{alpha}, but not HIF-1{alpha}, was found to be a strong independent prognostic marker in breast cancer, predicting impaired recurrence-free (RFS) and breast cancer–specific (BCSS) survival.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosure of Potential...
 References
 
Patients and tumor material. Cohort I consisted of 512 unselected patients, diagnosed with primary breast cancer between 1988 and 1992 at the Department of Pathology, University Hospital MAS. Median age at diagnosis was 64.2 y (range, 27–96 y) and median follow-up time from diagnosis until first breast cancer event was 106 mo (0–207). Follow-up data regarding RFS and death were available for 507 patients. All tumors were reclassified regarding histologic subtype and Nottingham Histological Grade (NHG) before TMA-construction. Histopathologically, 338 tumors (66%) were of ductal, 71 (14%) of lobular, and 36 (7%) of tubular subtype. Medullary, mucinous, and not otherwise specified cancers made up the remaining 13%. One hundred and twenty-eight tumors (25%) were graded as NHG I, 215 (42%) as NHG II, and 169 (33%) as NHG III. Lymph node status was negative in 291 (57%), positive in 166 (32%), and unknown in 55 (11%) cases. Estrogen receptor (ER)-{alpha} status was positive in 407 (80%), negative in 72 (14%), and unknown in 33 (6%) cases. Two hundred and eight patients had received radiotherapy and 190 had not. Twenty-three patients had received adjuvant chemotherapy and 369 had not. One hundred and sixty-one patients had received adjuvant endocrine therapy (tamoxifen) and 228 patients had not. Cohort II consisted of 179 women diagnosed with invasive breast cancer at Malmö University Hospital during 2001 to 2002. Median follow-up time was 69 mo, median age at diagnosis 65 y (range, 35–97). Histopathologically, 125 (70%) tumors were of ductal, 31 (17%) of lobular, 16 (9%) of tubular, and 7 (4%) of other type. Thirty-eight tumors (21%) were graded as NHG I, 79 (44%) as NHG II, and 62 (35%) as NHG III. Node status was negative in 89 (50%) cases, positive in 64 (36%), and missing in 26 (14%). ER status was negative in 22 (12%) cases and positive in 157 (88%). PgR status was negative in 52 (29%) cases and positive in 127 (71%). The association between NHG and distant metastasis was similar in both cohorts: 63% grade III tumors in cohort I versus 65% in cohort II, 30% grade II tumors in cohort I versus 35% in cohort II, and 7% grade I in cohort I versus 0% in cohort II (Supplementary Table S1). Ethical permissions were obtained from the Lund University Regional Ethics Board, ref. no. 613-02 (cohort I) and 445/2007 (cohort II).

TMA and IHC. Preceding TMA construction, H&E-stained tumor sections were reevaluated, and areas representative of invasive cancer were marked. Two 0.6-mm tissue cores were taken using an automated arraying device (ATA-27; Beecher, Inc.) and mounted in recipient blocks. IHC for HIF-1{alpha}, HIF-2{alpha}, and VEGF was performed in the DAKO Techmate 500 system (DAKO) using anti–HIF-1{alpha} (1:100; Upstate), anti–HIF-2{alpha} (1:1,000; clone ep190b; Abcam), and anti-VEGF (1:100; clone A-20). Heat-mediated antigen retrieval was obtained by boiling under pressure in DAKO Target Retrieval Buffer (pH 9.0) for HIF-1{alpha} and HIF-2{alpha} and by microwave treatment in DAKO Target Retrieval Buffer (pH 9.9) for VEGF. ER and progesterone receptor (PgR) were analyzed as described (19). Immunoreactivity was visualized using the DAKO Envision System kit. All sections (4 µm) were counterstained with hematoxylin.

For HIF-2{alpha}, cytoplasmic and nuclear protein levels were evaluated. The cytoplasmic staining intensity was scored on a five level scale with 0 for no staining, I for low, II for intermediate, III for high, and IV for very high staining intensity, whereas the nuclear staining intensity was evaluated as 0 for negative, I for intermediate, or II for high. The fraction of positively stained nuclei was scored as 0 for 0% to 1%, I for 2% to 10%, and II for >10% of the nuclei. The same scoring procedure was used for assessment of HIF-1{alpha} cytoplasmic and nuclear protein levels. For VEGF, cytoplasmic staining intensity was graded as 0 for no staining, 1 for low, 2 for intermediate, and 3 for high staining intensity. ER and PgR were evaluated according to the fraction of positively stained nuclei as 0 (0–1%), I (2–10%), II (11–75%), or III (76–100%). For statistical analyses, variables were dichotomized into "negative" or "positive" using the clinically established cutoff for ER and PgR at 10% positive nuclei. The analyses were performed by three independent observers (KH, AML, and KJ). In case of conflicting results (<10%), the biopsies were revised and consensus was reached. Images of IHC staining were obtained using Olympus BX51 microscope.

Cell culture. The human breast cancer cell lines MCF-7, T47D, and MDA-MB-468 were maintained under standard conditions before experimental procedures. To study HIF-protein levels at normoxia, near-physiologic oxygen levels, and hypoxia, cells were kept at 21%, 5%, and 1% O2, respectively, as described (20).

Transfection with siRNA. Transfections were performed essentially as described using 50 nmol siRNA (18). To increase knockdown efficiency, cells were incubated with siRNA for 4 h at 21% O2 and thereafter washed with medium. The transfection procedure was repeated 24 h later. As control, Silencer Negative Control siRNA (Ambion) was used. After transfection, cells were grown at 21% or 1% O2 for either 4 or 24 h. Cells were then harvested for Quantitative Real-Time PCR analyses (Q-PCR) or fixed and embedded in paraffin for IHC as described above for HIF-1{alpha} and HIF-2{alpha}.

Q-PCR, Western blot, and ELISA analyses. RNA extraction, cDNA synthesis, and Q-PCR reactions with SYBR Green PCR master-mix (Applied Biosystems) were performed as described (21). Expression levels of the analyzed genes were normalized to the expression of three nonhypoxia-induced housekeeping genes (SDHA, YWHAZ, and UBC). Primers were designed as described previously, using Primer-express (Applied Biosystems; ref. 18). Protein extraction and Western blotting were performed as described (22). In short, equal amount (80 µg) of total protein was loaded on a 6% SDS-PAGE gel and blotted to Hybond C nitrocellulose filters (Amersham). Equal loading was checked by Ponceau S staining. Primary antibodies used were as follows: anti-Actin monoclonal antibody (mAb; ICN Biomedicals), anti–HIF-1{alpha}, and anti–HIF-2{alpha} mAb (Novus Biologicals). To measure VEGF protein levels in cell supernatants, culture medium was collected, centrifuged to remove cellular debris, and stored at –80°C until assayed for VEGF using a commercially available ELISA kit (R&D systems).

Statistics. Spearman's correlation test and the {chi}2 test were used for comparison between variables. Kaplan-Meier analysis and the log-rank test were used to illustrate differences between RFS and BCSS according to HIF expression. RFS was defined as time from diagnosis until local, regional, or distant recurrence or breast cancer–related death. BCSS was defined as time from diagnosis until breast cancer–related death and overall survival (OS) as time from diagnosis until death of any cause. Multivariate Cox proportional hazards regression model was used to estimate the relative risk of recurrence and breast cancer–specific death in HIF-defined breast cancer subgroups. All statistical tests were two sided. P values of <0.05 were considered significant. Calculations were performed using SPSS version 12 (SPSS, Inc.).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosure of Potential...
 References
 
HIF-1{alpha}, HIF-2{alpha}, and VEGF protein expression in invasive breast cancer. IHC evaluation of HIF expression based on commercially available antibodies/sera has turned out to be demanding due to irregular variations in antibody preparations, antibody characteristics, etc. Specificity of the HIF antibodies was therefore confirmed using cultured hypoxic tumor cells pretreated with either HIF-1{alpha} or HIF-2{alpha} siRNA, which were fixed and embedded according to clinical practice. The IHC staining signals were virtually abolished in MCF-7 cells transfected with either HIF-1{alpha} or HIF-2{alpha} siRNA when stained with the respective antibody (Fig. 1A ). Control of knockdown efficiencies were performed by Q-PCR (Fig. 1B). Testing our anti-HIF IHC protocols on breast tumor sections revealed specific nuclear and cytoplasmic staining with varying intensities of tumor but not stromal cells, whereas blood vessel endothelium frequently stained weakly for HIF-2{alpha} (Fig. 1C). In cohort I, 316 cases were available for IHC evaluation of HIF-1{alpha} expression, 317 cases for HIF-2{alpha}, and 320 cases for VEGF. The remaining cases were either lost during technical preparation/staining or excluded due to lack of invasive cancer in the biopsy. These tumors had similar NHG-distribution as the investigated cases (Supplementary Fig. S1).


Figure 1
View larger version (79K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. A, HIF-1{alpha} and HIF-2{alpha} immunohistochemistry on MCF-7 cells grown at 1% O2 after transfection with corresponding HIF siRNA. B, relative HIF-1{alpha} and HIF-2{alpha} mRNA levels after transfection in A. C, nuclear HIF-1{alpha} and HIF-2{alpha} IHC staining intensities 0, I, and II in breast cancer specimens. Arrows, HIF-2{alpha}–positive blood vessel endothelial cells. Scale bar, 10 µm. siC, control siRNA; siH1, HIF-1{alpha} siRNA; siH2, HIF-2{alpha} siRNA.

 
We have previously shown that both nuclear and cytoplasmic HIF stainings represent actual HIF protein deposits (18). In the present study, nuclear and cytoplasmic stainings were strongly interrelated for both HIF-2{alpha} and HIF-1{alpha} (P < 0.001). However, there was no significant correlation between HIF-1{alpha} and HIF-2{alpha} protein levels (P = 0.570), suggesting different mechanisms underlying the presence of HIF-1{alpha} and HIF-2{alpha} protein in invasive breast cancer. HIF-1{alpha} showed a significant positive correlation to VEGF expression (P ≤ 0.001; {rho} = 0.215). There was also a significant positive, although not as strong, correlation between HIF-2{alpha} and VEGF expression (P = 0.017; {rho} = 0.138).

HIF-{alpha} subtype protein levels are regulated in a time- and oxygen-dependent manner. Given the observed lack of correlation between HIF-1{alpha} and HIF-2{alpha} protein levels, as well as the subtype-specific correlation to VEGF expression, a hypoxia-driven gene, we further investigated hypoxia-related HIF subtype accumulation under defined in vitro conditions. Cultured cells are routinely kept at 21% O2, usually termed "normoxia," but in vivo, the end capillary O2 concentration approaches 5% (23, 24). Therefore, protein levels of HIF-1{alpha} and HIF-2{alpha} in MCF-7, T47D, and MDA-MB-468 cell lines grown at 21%, 5%, and 1% O2 for 4 and 72 hours were analyzed, the time points chosen to represent acute and prolonged hypoxia. Analyses revealed an overall similar oxygen-dependent HIF protein pattern in the 3 cell lines, with induction at 1% O2 (Fig. 2 ). Specifically, at 21% O2, HIF-1{alpha} was undetectable, whereas HIF-2{alpha} was barely detected at 4 hours in the MDA-MB-468 and T47D cells. At 5% O2, only HIF-2{alpha} was detected at 4 hours in MCF-7. In the other cell lines, both HIF proteins were detected also at 5% O2 at both time points. HIF-1{alpha} levels decreased after 72 hours at hypoxia compared with the 4-hour levels, whereas HIF-2{alpha} protein levels stayed comparatively high during prolonged growth at 1% O2.


Figure 2
View larger version (44K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Time- and oxygen level–dependent, selective stabilization of HIF-1{alpha} and/or HIF-2{alpha} in MCF-7, T47D, and MDA-MB-468 breast cancer cells grown at 21%, 5%, and 1% O2 for 4 and 72 h determined by Western blotting. Hypoxic cells were harvested under hypoxia and 80 µg of total protein of each sample were separated by SDS-PAGE. Equal loading was checked by Ponceau S staining.

 
Having observed the HIF subtype–specific correlation to VEGF in vivo as well as the time- and oxygen level–dependent regulation of HIF-{alpha} subunits in vitro, we next analyzed the specific contribution of HIF-1 and HIF-2, respectively, to VEGF transcription. MCF-7 and MDA-MB-468 cell lines were transfected with siRNA against HIF-1{alpha} or HIF-2{alpha} or the combination of both, then grown for 4 and 24 hours at 21% or 1% O2. Overall, the HIF siRNAs efficiently knocked down respective HIF mRNA (Fig. 3A and B ) and protein (Fig. 1A). Regarding hypoxia-induced VEGF expression, primarily HIF-1 and not HIF-2 seemed to be involved at 4 hours, although the effect of HIF-1{alpha} siRNA was small in MDA-MB-468 cells. Knocking down both HIF-{alpha} isoforms revealed that the contribution of HIF-2 to VEGF expression at acute hypoxia was limited (Fig. 3C). At prolonged hypoxia (24 hours), VEGF transcription was slightly affected by both HIF-1{alpha} and HIF-2{alpha} siRNA, whereas combined siRNA treatment reduced the VEGF expression to normoxic levels in both cell lines (Fig. 3C). The changes in VEGF mRNA levels were largely mirrored in VEGF protein in matched cell supernatants (Fig. 3D).


Figure 3
View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Effects of combined and selective siRNA knockdown of HIF-1{alpha} and/or HIF-2{alpha} on hypoxia-induced VEGF transcription in MCF-7 and MDA-MB-468 cells at acute (4 h) or prolonged (24 h) hypoxia showing relative HIF-1{alpha} (A), HIF-2{alpha} (B), and VEGF (C) mRNA levels. A representative experiment analyzed in triplicate is shown. D, VEGF levels in matched cell supernatants collected at time of harvest and quantified by ELISA. Columns, mean; bars, SD. C, Lipofectamine control.

 
HIF-2{alpha} predicts impaired outcome in invasive breast cancer. We next analyzed HIF protein levels in relation to clinicopathologic variables. Both cytoplasmic (data not shown) and nuclear HIF-1{alpha} protein expression exhibited a significant but weak, negative correlation to NHG and a borderline-significant, negative correlation to node status (Table 1 ). HIF-2{alpha} was not significantly associated with conventional clinicopathologic variables but did show a significant correlation to incidence of distant but not local or regional recurrence (Table 1). The increased risk of distant relapse remained significant in multivariate analyses adjusted for NHG, node status, tumor size, and age [hazard ratio (HR), 2.5; 95% confidence interval (95% CI), 1.5–4.1; P < 0.001]. Neither HIF-1{alpha} nor HIF-2{alpha} nuclear protein levels correlated significantly to ER or PgR expression. For survival analyses, both nuclear and cytoplasmic HIF-1{alpha} and HIF-2{alpha} expression scores were used, showing a similar effect on RFS and BCSS. As active HIF-1{alpha} is commonly assumed to be located to the nucleus and as scoring the fraction of positively stained nuclei might be considered more objective than intensity scoring, only results with nuclear fraction scores of HIF-1{alpha} (Fig. 4A ) and HIF-2{alpha} (Fig. 4B–D) expression are shown. Survival analyses showed that HIF-2{alpha} nuclear fraction scores I and II exhibited a similar effect on survival (Fig. 4B), and the data were therefore dichotomized into "negative" for 0 and "positive" for fraction scores I and II for all survival analyses (Fig. 4C and D). Interestingly, although HIF-1{alpha} did not have any significant effect on either RFS or BCSS, HIF-2{alpha} expression was associated with significantly shorter RFS and BCSS (Fig. 4A–D). The association remained significant after adjusting for NHG, node status, tumor size, and age (Table 2 ).


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

  		Table 1. HIF-1{alpha} and HIF-2{alpha} expression in breast cancer in relation to clinicopathologic variables

 

Figure 4
View larger version (17K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Survival among breast cancer patients by HIF-1{alpha} or HIF-2{alpha} expression analyzed as fraction of positively stained nuclei. HIF-1{alpha}/HIF-2{alpha}–negative, nuclear fraction score 0; HIF-1{alpha}/HIF-2{alpha}–positive, nuclear fraction scores I and II. A, RFS according to HIF-1{alpha}. B and C, RFS according to HIF-2{alpha}. D, BCSS according to HIF-2{alpha}.

 

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

  		Table 2. Cox multivariate analysis of RFS and BCSS in breast cancer, adjusted for NHG, node status, tumor size, and HIF-2{alpha} expression

 
HIF-{alpha} protein levels in a second breast cancer material (cohort II). The negative correlation between HIF-1{alpha} and NHG, and the tendency of a positive relationship of nuclear HIF-1{alpha} to survival were unexpected findings. As the HIF-1{alpha} antibody used had been rigorously tested for specific IHC staining, arguing that our data were trustworthy, we extended our study to an additional breast cancer material (cohort II) to further test the effect of HIF-1{alpha} on survival. In this cohort, as opposed to cohort I, HIF-1{alpha} nuclear staining intensity had a positive, although nonsignificant association, to NHG (P = 0.563; {rho} = 0.051) and thereby a negative association to OS (Supplementary Fig. S2A and B). However, in agreement with cohort I data, there was no correlation between nuclear HIF-2{alpha} and HIF-1{alpha} levels (P = 0.650; {rho} = –0.042) but a trend toward shorter OS for patients with HIF-2{alpha} positive tumors (HR, 1.58; 95% CI, 0.83–3.02; P = 0.159; Supplementary Fig. S3).


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosure of Potential...
 References
 
We have tested the association between nuclear HIF-1{alpha} and HIF-2{alpha} protein levels and RFS and BCSS within the context of established prognostic markers in a large breast cancer material. Although we found a significant association between HIF-2{alpha} protein and adverse outcome, no such association was found for HIF-1{alpha}. Accordingly, HIF-1{alpha} and HIF-2{alpha} protein levels did not correlate, strongly suggesting that HIF-1{alpha} and HIF-2{alpha} are differently regulated in breast cancer tissue and that hypoxia is only one of the mechanisms. Of potential clinical importance, we further show that patients with HIF-2{alpha}–positive cancers have increased likelihood of distant recurrence. With regards to hypoxia, in vitro data suggest that HIF-1 is primarily accumulated during acute hypoxia, whereas accumulation of HIF-2 persists during prolonged response to hypoxia, similar to our previous findings in neuroblastoma (18).

Several clinical studies have used HIF-1{alpha} and VEGF as surrogate markers of tumor hypoxia in breast cancer; however, their correlations to poor prognosis and to clinicopathologic variables have not been clear-cut (9, 11–14, 16, 25–30). In this report, we show that in a large tumor material (cohort I), HIF-1{alpha} correlated negatively to NHG and lymph node status and lacked correlation to patient age, tumor size, and ER and PgR status. HIF-1{alpha} did not have a significant effect on survival, although surprisingly, the HIF-1{alpha} protein expressing tumors did show a tendency toward better prognosis than their HIF-1{alpha}–negative counterparts, which is in line with the negative correlation to NHG. Our in vitro data further suggest that hypoxia-induced VEGF transcription in breast cancer cells at the acute phase is primarily supported by HIF-1{alpha}, a conclusion in agreement with published observations (31, 32). Our data further suggest that HIF-2{alpha} gradually takes over the role as VEGF inducer at prolonged hypoxia, an observation in concert with our previous findings in neuroblastoma and in von Hippel-Lindau–associated vascular tumors (33). Other laboratories report cell line–specific levels and ratios of HIF-{alpha} proteins as well as their nonredundant hypoxia-induced roles as transcription factors, adding to the complexity of HIF-regulated VEGF expression in breast cancer (18, 34–38). Lately, nonhypoxia-induced regulatory pathways of HIF-1{alpha} and HIF-2{alpha} have come into focus (16, 39–42). Our in vivo results are in line with such mechanisms, as there was lack of correlation between HIF-1{alpha} and HIF-2{alpha} proteins as well as HIF-{alpha} subunit–specific association to survival. Our in vitro data further show that HIF-1{alpha} and HIF-2{alpha} protein levels are differently regulated in an oxygen level– and time-dependent manner. Taken together, our findings suggest independent and specific roles of the two HIF-{alpha} subunits and an oncogenic behavior of HIF-2{alpha} in breast cancer.

The majority of studies on HIF-1{alpha} protein levels in breast cancer report a negative association to outcome (9–11, 13, 14, 27). In two different breast cancer cohorts, we found tendencies toward two opposite results. As these results were obtained by the same staining procedure, we can in our studies rule out technical differences and shortcomings, an otherwise highly plausible explanation to discordant HIF staining data. However, based on our own and reported data, we conclude that HIF-1{alpha} correlations to patient outcome, either positive or negative, are rarely strong, implying that whatever biological effect HIF-1{alpha} has on breast tumor growth and behavior, the net effect of high HIF-1{alpha} activity is comparatively small. HIF-2{alpha} expression on the other hand, significantly increased the risk of distant recurrence and breast cancer–specific death, and retained its value as an independent prognostic marker after adjustment for established prognostic factors. These nonrandomized cohorts do not allow for proper analyses of a potential treatment-predictive value of HIF-2{alpha}. The reason for the reduced and nonsignificant association of HIF-2{alpha} to survival in cohort II may have several explanations. The follow-up was much shorter in cohort II (median 69 versus 106 months in cohort I), whereas the median time to distant recurrence was 61 months in cohort II versus 41 months in cohort I. Evidently, tumors in cohort II were diagnosed at an earlier stage, probably due to the fact that they belong to an era of routine mammography screening. Thus, as distant recurrences most likely reflect late events, they will occur later in cohort II compared with cohort I, and the association of HIF-2{alpha} to distant recurrence in cohort II might be expected to reach significance with a longer follow-up (see Supplementary Fig. S3).

At this stage, one can only speculate why HIF-2{alpha} and not HIF-1{alpha} is associated with aggressive breast tumor growth. In a number of different tumors, HIF-2{alpha} has recently been associated with aggressive growth, VEGF expression, and tumor angiogenesis (17, 18, 43–45). It has further been suggested that HIF-2a is highly expressed and active in stem cells and tumor stem cells or initiating cells of various origins (46–48). The high rate of distant but not local or regional recurrence implicates HIF-2{alpha} in hematogenous spreading. Future efforts should be directed toward the understanding of the role of HIF-2a in metastasizing breast cancer.


   Disclosure of Potential Conflicts of Interest
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
References
 
No potential conflicts of interest were disclosed.


   Acknowledgments
 
Grant support: Swedish Cancer Society, the Children's Cancer Foundation of Sweden, the Swedish Research Council, the SSF Strategic Center for Translational Cancer Research-CREATE Health, the Knut and Alice Wallenberg Foundation, Research Program in Medical Bioinformatics of the Swedish Knowledge Foundation, Hans von Kantzows Stiftelse, Gunnar Nilsson's Cancer Foundation, and the research funds of Malmö University Hospital.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Siv Beckman, Elisabet Johansson, and Elise Nilsson for expert technical assistance.


   Footnotes
 
Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

Received 3/27/08. Revised 8/11/08. Accepted 9/11/08.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
 References
 

  1. Harris AL. Hypoxia-a key regulatory factor in tumour growth. Nat Rev Cancer 2002;2:38–47.[CrossRef][Medline]
  2. Hockel M, Vaupel P. Tumor hypoxia: definitions and current clinical, biologic, and molecular aspects. J Natl Cancer Inst 2001;93:266–76.[Abstract/Free Full Text]
  3. Jögi A, Öra I, Nilsson H, et al. Hypoxia alters gene expression in human neuroblastoma cells toward an immature and neural crest-like phenotype. Proc Natl Acad Sci U S A 2002;99:7021–6.[Abstract/Free Full Text]
  4. Helczynska K, Kronblad A, Jögi A, et al. Hypoxia promotes a dedifferentiated phenotype in ductal breast carcinoma in situ. Cancer Res 2003;63:1441–4.[Abstract/Free Full Text]
  5. Hockel M, Schlenger K, Aral B, Mitze M, Schaffer U, Vaupel P. Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. Cancer Res 1996;56:4509–15.[Abstract/Free Full Text]
  6. Brizel DM, Sibley GS, Prosnitz LR, Scher RL, Dewhirst MW. Tumor hypoxia adversely affects the prognosis of carcinoma of the head and neck. Int J Radiat Oncol Biol Phys 1997;38:285–9.[CrossRef][Medline]
  7. Brizel DM, Scully SP, Harrelson JM, et al. Tumor oxygenation predicts for the likelihood of distant metastases in human soft tissue sarcoma. Cancer Res 1996;56:941–3.[Abstract/Free Full Text]
  8. Axelson H, Fredlund E, Ovenberger M, Landberg G, Påhlman S. Hypoxia-induced dedifferentiation of tumor cells - a mechanism behind heterogeneity and aggressiveness of solid tumors. Semin Cell Dev Biol 2005;16:554–63.[CrossRef][Medline]
  9. Schindl M, Schoppmann SF, Samonigg H, et al. Overexpression of hypoxia-inducible factor 1{alpha} is associated with an unfavorable prognosis in lymph node-positive breast cancer. Clin Cancer Res 2002;8:1831–7.[Abstract/Free Full Text]
  10. Bos R, van der Groep P, Greijer AE, et al. Levels of hypoxia-inducible factor-1{alpha} independently predict prognosis in patients with lymph node negative breast carcinoma. Cancer 2003;97:1573–81.[CrossRef][Medline]
  11. Dales JP, Garcia S, Meunier-Carpentier S, et al. Overexpression of hypoxia-inducible factor HIF-1{alpha} predicts early relapse in breast cancer: retrospective study in a series of 745 patients. Int J Cancer 2005;116:734–9.[CrossRef][Medline]
  12. Kronblad A, Jirström K, Ryden L, Nordenskjold B, Landberg G. Hypoxia inducible factor-1{alpha} is a prognostic marker in premenopausal patients with intermediate to highly differentiated breast cancer but not a predictive marker for tamoxifen response. Int J Cancer 2006;118:2609–16.[CrossRef][Medline]
  13. Generali D, Berruti A, Brizzi MP, et al. Hypoxia-inducible factor-1{alpha} expression predicts a poor response to primary chemoendocrine therapy and disease-free survival in primary human breast cancer. Clin Cancer Res 2006;12:4562–8.[Abstract/Free Full Text]
  14. Trastour C, Benizri E, Ettore F, et al. HIF-1{alpha} and CA IX staining in invasive breast carcinomas: prognosis and treatment outcome. Int J Cancer 2007;120:1451–8.[CrossRef][Medline]
  15. Tan EY, Campo L, Han C, et al. BNIP3 as a progression marker in primary human breast cancer; opposing functions in in situ versus invasive cancer. Clin Cancer Res 2007;13:467–74.[Abstract/Free Full Text]
  16. Vleugel MM, Greijer AE, Shvarts A, et al. Differential prognostic impact of hypoxia induced and diffuse HIF-1{alpha} expression in invasive breast cancer. J Clin Pathol 2005;58:172–7.[Abstract/Free Full Text]
  17. Leek RD, Talks KL, Pezzella F, et al. Relation of hypoxia-inducible factor-2{alpha} (HIF-2{alpha}) expression in tumor-infiltrative macrophages to tumor angiogenesis and the oxidative thymidine phosphorylase pathway in human breast cancer. Cancer Res 2002;62:1326–9.[Abstract/Free Full Text]
  18. Holmquist-Mengelbier L, Fredlund E, Löfstedt T, et al. Recruitment of HIF-1{alpha} and HIF-2{alpha} to common target genes is differentially regulated in neuroblastoma: HIF-2{alpha} promotes an aggressive phenotype. Cancer Cell 2006;10:413–23.[CrossRef][Medline]
  19. Borgquist S, Holm C, Stendahl M, Anagnostaki L, Landberg G, Jirström K. Oestrogen receptors {alpha} and β show different associations to clinicopathological parameters and their co-expression might predict a better response to endocrine treatment in breast cancer. J Clin Pathol 2008;61:197–203.[Abstract/Free Full Text]
  20. Holmquist L, Jögi A, Påhlman S. Phenotypic persistence after reoxygenation of hypoxic neuroblastoma cells. Int J Cancer 2005;116:218–25.[CrossRef][Medline]
  21. Löfstedt T, Jögi A, Sigvardsson M, et al. Induction of ID2 expression by hypoxia-inducible factor-1: a role in dedifferentiation of hypoxic neuroblastoma cells. J Biol Chem 2004;279:39223–31.[Abstract/Free Full Text]
  22. Nilsson H, Jögi A, Beckman S, Harris AL, Poellinger L, Påhlman S. HIF-2{alpha} expression in human fetal paraganglia and neuroblastoma: relation to sympathetic differentiation, glucose deficiency, and hypoxia. Exp Cell Res 2005;303:447–56.[CrossRef][Medline]
  23. Brown JM. The Hypoxic Cell: A Target for Selective Cancer Therapy-Eighteenth Bruce F. Cain Memorial Award Lecture. Cancer Res 1999;59:5863–70.[Abstract/Free Full Text]
  24. Vaupel P, Schlenger K, Knoop C, Hockel M. oxygenation of human tumors: evaluation of tissue oxygen distribution in breast cancers by computerized O2 tension measurements. Cancer Res 1991;51:3316–22.[Abstract/Free Full Text]
  25. Lundgren K, Holm C, Landberg G. Hypoxia and breast cancer: prognostic and therapeutic implications. Cell Mol Life Sci 2007;64:3233–47.[CrossRef][Medline]
  26. Byrne GJ, McDowell G, Agarawal R, Sinha G, Kumar S, Bundred NJ. Serum vascular endothelial growth factor in breast cancer. Anticancer Res 2007;27:3481–7.[Abstract/Free Full Text]
  27. Gruber G, Greiner RH, Hlushchuk R, et al. Hypoxia-inducible factor 1 {alpha} in high-risk breast cancer: an independent prognostic parameter? Breast Cancer Res 2004;6:R191–8.[CrossRef][Medline]
  28. Linderholm B, Lindh B, Tavelin B, Grankvist K, Henriksson R. p53 and vascular-endothelial-growth-factor (VEGF) expression predicts outcome in 833 patients with primary breast carcinoma. Int J Cancer 2000;89:51–62.[CrossRef][Medline]
  29. Linderholm B, Grankvist K, Wilking N, Johansson M, Tavelin B, Henriksson R. Correlation of vascular endothelial growth factor content with recurrences, survival, and first relapse site in primary node-positive breast carcinoma after adjuvant treatment. J Clin Oncol 2000;18:1423–31.[Abstract/Free Full Text]
  30. Gasparini G, Toi M, Gion M, et al. Prognostic significance of vascular endothelial growth factor protein in node-negative breast carcinoma. J Natl Cancer Inst 1997;89:139–47.[Abstract/Free Full Text]
  31. Elvidge GP, Glenny L, Appelhoff RJ, Ratcliffe PJ, Ragoussis J, Gleadle JM. Concordant regulation of gene expression by hypoxia and 2-oxoglutarate-dependent dioxygenase inhibition: the role of HIF-1{alpha}, HIF-2{alpha}, and other pathways. J Biol Chem 2006;281:15215–26.[Abstract/Free Full Text]
  32. Aprelikova O, Wood M, Tackett S, Chandramouli GV, Barrett JC. Role of ETS transcription factors in the hypoxia-inducible factor-2 target gene selection. Cancer Res 2006;66:5641–7.[Abstract/Free Full Text]
  33. Rankin EB, Rha J, Unger TL, et al. Hypoxia-inducible factor-2 regulates vascular tumorigenesis in mice. Oncogene 2008;27:5354–8. Epub 2008 May 19.[CrossRef][Medline]
  34. Blancher C, Moore JW, Talks KL, Houlbrook S, Harris AL. Relationship of hypoxia-inducible factor (HIF)-1{alpha} and HIF-2{alpha} expression to vascular endothelial growth factor induction and hypoxia survival in human breast cancer cell lines. Cancer Res 2000;60:7106–13.[Abstract/Free Full Text]
  35. Hu CJ, Wang LY, Chodosh LA, Keith B, Simon MC. Differential roles of hypoxia-inducible factor 1{alpha} (HIF-1{alpha}) and HIF-2{alpha} in hypoxic gene regulation. Mol Cell Biol 2003;23:9361–74.[Abstract/Free Full Text]
  36. Carroll VA, Ashcroft M. Role of hypoxia-inducible factor (HIF)-1{alpha} versus HIF-2{alpha} in the regulation of HIF Target genes in response to hypoxia, insulin-like growth factor-I, or loss of von Hippel-Lindau function: implications for targeting the HIF pathway. Cancer Res 2006;66:6264–70.[Abstract/Free Full Text]
  37. Park SK, Dadak AM, Haase VH, Fontana L, Giaccia AJ, Johnson RS. Hypoxia-induced gene expression occurs solely through the action of hypoxia-inducible factor 1{alpha} (HIF-1{alpha}): role of cytoplasmic trapping of HIF-2{alpha}. Mol Cell Biol 2003;23:4959–71.[Abstract/Free Full Text]
  38. Sowter HM, Raval RR, Moore JW, Ratcliffe PJ, Harris AL. Predominant role of hypoxia-inducible transcription factor (HIF)-1{alpha} versus HIF-2{alpha} in regulation of the transcriptional response to hypoxia. Cancer Res 2003;63:6130–4.[Abstract/Free Full Text]
  39. Bardos JI, Ashcroft M. Hypoxia-inducible factor-1 and oncogenic signalling. Bioessays 2004;26:262–9.[CrossRef][Medline]
  40. Akeno N, Robins J, Zhang M, Czyzyk-Krzeska MF, Clemens TL. Induction of vascular endothelial growth factor by IGF-I in osteoblast-like cells is mediated by the PI3K signaling pathway through the hypoxia-inducible factor-2{alpha}. Endocrinology 2002;143:420–5.[Abstract/Free Full Text]
  41. Fukuda R, Hirota K, Fan F, Jung YD, Ellis LM, Semenza GL. Insulin-like growth factor 1 induces hypoxia-inducible factor 1-mediated vascular endothelial growth factor expression, which is dependent on MAP kinase and phosphatidylinositol 3-kinase signaling in colon cancer cells. J Biol Chem 2002;277:38205–11.[Abstract/Free Full Text]
  42. Frede S, Berchner Pfannschmidt U, Fandrey J, Sies H, Bernhard B. Regulation of hypoxia-inducible factors during inflammation. Methods Enzymol 2007;435:40319.[CrossRef]
  43. Giatromanolaki A, Sivridis E, Fiska A, Koukourakis MI. Hypoxia-inducible factor-2 {alpha} (HIF-2{alpha}) induces angiogenesis in breast carcinomas. Appl Immunohistochem Mol Morphol 2006;14:78–82.[CrossRef][Medline]
  44. Giatromanolaki A, Sivridis E, Simopoulos C, et al. Hypoxia inducible factors 1{alpha} and 2{alpha} are associated with VEGF expression and angiogenesis in gallbladder carcinomas. J Surg Oncol 2006;94:242–7.[CrossRef][Medline]
  45. Koukourakis MI, Giatromanolaki A, Sivridis E, et al. Hypoxia-inducible factor (HIF1A and HIF2A), angiogenesis, and chemoradiotherapy outcome of squamous cell head-and-neck cancer. Int J Radiat Oncol Biol Phys 2002;53:1192–202.[CrossRef][Medline]
  46. Covello KL, Kehler J, Yu H, et al. HIF-2{alpha} regulates Oct-4: effects of hypoxia on stem cell function, embryonic development, and tumor growth. Genes Dev 2006;20:557–70.[Abstract/Free Full Text]
  47. Pietras A, Gisselsson D, Öra I, et al. High levels of HIF-2{alpha} highlight an immature neural crest-like neuroblastoma cell cohort located in a perivascular niche. J Pathol 2008;214:482–8.[CrossRef][Medline]
  48. Keith B, Simon MC. Hypoxia-inducible factors, stem cells, and cancer. Cell 2007;129:465–72.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Proc. Natl. Acad. Sci. USAHome page
A. Pietras, L. M. Hansford, A. S. Johnsson, E. Bridges, J. Sjolund, D. Gisselsson, M. Rehn, S. Beckman, R. Noguera, S. Navarro, et al.
HIF-2{alpha} maintains an undifferentiated state in neural crest-like human neuroblastoma tumor-initiating cells
PNAS, September 29, 2009; 106(39): 16805 - 16810.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Helczynska, K.
Right arrow Articles by Jirström, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Helczynska, K.
Right arrow Articles by Jirström, K.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online