HIF and Tumorigenesis

The importance of HIF in the regulation of angiogenesis, together with the striking upregulation of this system following inactivation of the VHL tumor suppressor, raises the question as to the role played by HIF activation in the onco-genic process. Currently this important question is not fully resolved. However, recent advances in the characterization of the HIF/VHL system have enabled the question to be considered in VHL-associated malignancy and in cancer in general. In the following, we have summarized current knowledge as to the role of HIF in both non-VHL and VHL-associated malignancy.

Non-VHL Associated Cancer

Upregulation of the HIF system is observed in many common cancers and occurs by a multiplicity of genetic and environmental factors. Microenvironmental activation of HIF in cancer occurs at the simplest level by physiological activation of the oxygen-sensitive pathways by hypoxia within the growing mass of cells. However, in addition to VHL inactivation, a wide range of tumor suppressor and oncogenic mutations have been reported to activate the HIF system by a variety of mechanisms. These include the association of p53 loss-of-function with decreased ubiquityla-tion of HIF-1a; PTEN loss-of-function, PI3K/AKT/FRAP signaling, and SRC gain-of-function with increased HIF-1a synthesis; and p14ARF loss-of-function with decreased nucleolar sequestration of HIF-1a.

In addition to promoting angiogenesis, activation of the HIF system may also mediate metabolic abnormalities associated with cancer. Tumor cells are characterized by a marked increase in glycolytic metabolism, even when cultured in the presence of high oxygen concentrations. First described by Otto Warburg more than 70 years ago, these abnormalities are, like angiogenesis, particularly associated with aggressive rapid growth tumors. HIF transcriptional targets include the glucose transporter isoforms, glycolytic isoenzymes and regulatory enzymes [e.g., 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase-3 (PFKFB3)] whose upregulation is classically associated with the Warburg effect. Thus HIF activation mediates coordinate transcrip-tional activation of the entire pathway, from glucose uptake to lactate production, providing a molecular explanation for this "metabolic switch." HIF transcriptional targets include other molecules that have potential protumorigenic actions, such as transforming growth factor-a and the c-Met protooncogene that mediates hepatocyte growth factor signaling.

These observations, together with clinico-pathological studies that have associated high levels of HIF-a subunits in human tumors with an aggressive phenotype, have led to the proposal that HIF activation contributes causally to the oncogenic processes. However, not all data support this conclusion. First, the correlation with tumor aggression is not universal. Second, some HIF-inducible genes are apparently anti-tumorigenic. For instance, in mouse embryonic stem cells HIF-1a can have proapoptotic effects, possibly through induction of target genes such as NIP3, which is involved in apoptotic pathways and binds Bcl-2. Third, though most studies of experimental tumor growth by cells bearing genetic modifications of the HIF system have demonstrated pro-tumorigenic effects, this finding has again not been universal, with some studies reporting the opposite finding. One straightforward explanation for these apparently conflicting results is that the protumorigenic effects of HIF activation are cell type specific—such a phenomenon being commonly observed with tumor suppressor inactiva-tion. Potentially in keeping with this, VHL-associated tumor formation is highly tissue specific. Nevertheless even in

VHL disease the causative role of HIF activation in the oncogenic process as opposed to the development of benign angiomata remains unclear.

VHL-Associated Cancer

VHL disease shows clear genotype-phenotype associations so that effects of particular mutations on HIF activation can be correlated with particular patterns of tumor predisposition. Families have been divided into type 1 or type 2A, 2B, or 2C, according to the risk of renal cell carcinoma, pheochromocytoma, and hemangioblastoma (Table I). The tight correlation between hemangioblastoma susceptibility and HIF dysregulation strongly supports a causal role of HIF activation in these benign vascular tumors consistent with the mechanistic findings linking the VHL/HIF pathway to the regulation of angiogenic growth factors. The associations are also compatible with a role for HIF activation in the development of renal cell carcinoma, but they indicate that HIF activation is unlikely to underlie pheochro-mocytoma development, strongly suggesting the existence of other VHL tumor suppressor mechanisms.

In kidneys removed from patients with VHL disease a very large number of foci of HIF activation can be observed and enhanced vascular density can be observed adjacent to these microscopic lesions. This suggests that HIF-mediated enhancement of angiogenesis is an early event following VHL inactivation and that other events (most probably several other events) are required for tumor formation. Though indices of cell proliferation are low in these early lesions, they must in some way provide an advantageous background on which subsequent oncogenic mutations can progress cancer development. Whether this is related to HIF activation and the early enhancement of angiogenesis is unclear.

Some direct support for a role of HIF in renal cell carcinoma development has been provided by studies of the effect of transfected HIF-a genes on experimental tumors derived from renal cell carcinoma cell lines and grown in

Table I Genotype-Phenotype Correlations in VHL Disease.

Type of

Most prevalent type

Effect on pVHL function

Tumor risk

VHL disease

of VHL mutation

in the HIF pathway



Renal cell carcinoma

Type 1

Loss or protein

Impaired binding to HIF-a.





Upregulation of HIF and

its target genes

Type 2a


Impaired binding to HIF-a.




Upregulation of HIF and its

target genes

Type 2b


Impaired binding to HIF-a.




Upregulation of HIF and its

target genes

Type 2c


Retains ability to bind and




degrade HIF-a

nude mice. Transfection of VHL-defective renal cell carcinoma lines with a wild-type VHL gene suppresses tumor formation, but this tumor suppression can be relieved by supertransfection with a HIF-a gene that has been mutated so as to retain transcriptional activity but escape proteolytic destruction by VHL. Interestingly reversal of VHL tumor suppression was observed for HIF-2a but not HIF-1a, suggesting that HIF-2a but not HIF-1a upregulation contributes to the oncogenic process in these tumors. Despite these findings it is notable that direct (as opposed to indirect by VHL) mutational activation of HIF has not yet been observed in this tumor type—suggesting that mutational selection driving the tumor development may be affecting other VHL-dependent pathways.

A variety of VHL interacting proteins and VHL functions have been described that are apparently unrelated to effects on the HIF system. These include effects on other transcription factors such as Sp1, effects on the general transcrip-tional apparatus, effects on signaling molecules such as protein kinase C, and effects on matrix metabolism. Perhaps of greatest interest is evidence for a role of VHL in fibronectin assembly, given associations between abnormal fibronectin function and the cancer phenotype in other settings.

Interaction assays have demonstrated that pVHL can bind to fibronectin, and it is notable that type 2C, pheochro-mocytoma-causing pVHL mutants, which retain normal HIF-a regulation, are defective in fibronectin binding, suggesting a possible causative role in formation of these tumors. How intracellular pVHL interacts with the secreted fibronectin protein is uncertain. There is no evidence that pVHL is secreted into the lumen of the endoplasmic reticu-lum, although some fibronectin (possibly malfolded or misprocessed) may undergo retrograde transport to the cytosolic surface and be accessible to pVHL. What is clear, however, is that although cells lacking pVHL secrete fibronectin, they are defective with respect to fibronectin-matrix assembly. This phenotype is potentially complicated by the HIF dependence of several other genes involved in extracellular matrix turnover, such as tissue inhibitors of metalloproteinases (TIMPs), matrix metalloproteinases (MMPs), and collagen prolyl hydroxylase. However, VHL-deficient cells are unable to process exogenous fibronectin, suggesting a problem with pVHL handling rather than synthesis, and have been shown to have a defect in pi-integrin function, which is involved in assembly of fibrillary adhesions.

Thus in summary HIF-mediated angiogenesis most likely accounts for VHL-associated hemangioblastoma, but its relation to other aspects of the cancer predisposition is less clear. If HIF activation is not directly implicated in causing tumor development, it is necessary to consider other explanations for the common association of HIF activation with aggressive malignancy, and the paradox that although the HIF pathway is commonly activated by oncogenic and tumor suppressor mutations, it is not itself subject to onco-genic mutation.

A potential explanation exists in the function of HIF as a coordinator of extensive physiological pathways that link the metabolic demands of a proliferating cell mass to the need for development of an adequate oxygen supply. The multiple links between the stimuli that promote cell proliferation and the activation of HIF suggest that such pathways are "hard-wired" into development as part of the fundamental physiological need to preserve oxygen homeostasis. In this case, the classical genetic model of cancer involving the progressive selection of mutant cells that manifest a cell autonomous survival advantage would be expected to evolve other new properties, such as activation of HIF and associated angiogenic or metabolic phenotypes through the coselection of these intrinsically linked pathways. Such a coselection process would be compatible with the observation that hypoxic and angiogenic pathways are indirectly rather than directly activated by oncogenic mutations, and could also explain how properties such as angiogenesis, which confer a common advantage to a mass of cells rather than a cell autonomous advantage, can be selected. Since the concept of clonal selection in cancer requires that a cell gain predominance over its neighbors by individual rather than group advantage, it is otherwise difficult to understand how selection of an angiogenic phenotype occurs. The operation of coselection also predicts that angiogenesis might be activated in a disorganized way since activation of the relevant pathways reflects links to stochastic events driving clonal selection rather than a coordinated physiological response. If the links between a tumor suppressor gene or oncogene and proangiogenic pathways are strong, then early and excessive angiogenesis might be expected, as is the case for VHL disease. If the links were less powerful, then angio-genesis might lag behind tumor growth, and then develop more rapidly when further mutations are selected with more robust links to proangiogenic pathways. Given the enormous complexity of the pathways that are being revealed by molecular analysis of cell physiology, it is likely that the role played by coselection in the generation of cancer phe-notypes has been significantly underestimated. Further analysis of the HIF system in cancer biology should be illuminating in this respect.


Angiogenic switch: A discrete step in the development of tumors in which the balance between pro- and anti-angiogenic factors changes to favor constant growth of new tumor blood vessels. It may occur at different stages of the tumor progression pathway, depending on the type of tumor and its microenvironment, but generally occurs early. Prior to the switch tumor growth is limited to a few cubic millimeters.

Hypoxia-inducible factor: A heterodimeric transcription factor that senses and responds to ambient oxygen tension, through oxygen-dependent regulation of its abundance and transactivating ability. This in turn controls the transcription of whole families of hypoxia-inducible genes, including many involved in the regulation of angiogenesis.

Ubiquitin ligase: A protein complex that recognizes its substrate and covalently attaches a series of ubiquitin molecules. This polyubiquitin chain then tags the protein for destruction by a cellular structure known as the proteasome. pVHL acts as the recognition component for a ubiquitin ligase complex that targets HIF-a.


Bergers, G., and Benjamin, L. E. (2003). Tumorigenesis and the angiogenic switch. Nat. Rev. Cancer 3(6), 401-410. Using a multistep model of tumorigenesis, this review examines the impact of angiogenesis on tumor development and progression.

Clifford, S. C., Cockman, M. E., Smallwood, A. C., Mole, D. R., Woodward, E. R., Maxwell, P. H., Ratcliffe, P. J., and Maher, E. R. (2001). Contrasting effects on HIF-1 alpha regulation by disease-causing pVHL mutations correlate with patterns of tumourigenesis in von Hippel-Lindau disease. Hum. Mol. Genet. 10, 1029-1038. This paper characterizes the phenotype—genotype correlations in VHL syndrome.

Hanahan, D., and Folkman J. (1996). Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis. Cell 86(3), 353-364.

Kaelin, W. G., Jr. (2002). Molecular basis of the VHL hereditary cancer syndrome. Nat. Rev. Cancer 2(9), 673-682. This is a clear and concise review of the VHL syndrome, and the known functions of the VHL protein, with respect to both HIF and other interacting proteins.

Maxwell, P. H., Pugh, C. W., and Ratcliffe, P. J. (2001). Activation of the HIF pathway in cancer. Curr. Opin. Genet. Dev. 11(3), 293-299.

Maxwell, P. H., Wiesener, M. S., Chang, G. W., Clifford, S. C., Vaux, E. C., Cockman, M. E., Wykoff, C. C., Pugh, C. W., Maher, E. R., and Ratcliffe, P. J. (1999). The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 399(6733), 271-275. This is the first demonstration that pVHL is involved in the proteolytic regulation of HIF-a subunits.

Pugh, C. W., and Ratcliffe, P. J. (2003). Regulation of angiogenesis by hypoxia: Role of the HIF system. Nat. Med. 9(6), 677-684. This is a clear review of the regulation ofHIF by oxygen, and a summary of the action of hypoxia on molecular signals involved in different stages of angiogenesis. It forms part of a special issue of reviews on angiogenesis.

Semenza, G. L. (2000). Hypoxia, clonal selection, and the role of HIF-1 in tumor progression. Crit. Rev. Biochem. Mol. Biol. 35(2), 71-103.

Capsule Biographies

Dr. David R. Mole is Clinical Lecturer in Nephrology and General Medicine at Oxford University. He has worked on the oxygen-dependent interaction between pVHL and HIF-a, defining post-translational modification by prolyl hydroxylation as a key regulatory event leading to recognition of HIF-a by the VHL ligase complex in normoxia. By demonstrating a requirement for iron, oxygen, and 2-oxoglutarate in this hydroxylation of prolyl residues, he has helped to define the involvement of a new family of 2-oxoglutarate-dependent dioxygenases (PHD1, 2, and 3) in the regulation of HIF-a abundance.

Peter J. Ratcliffe is Professor of Renal Medicine and Head of the Nephrology and Cell Physiology Group at the Henry Wellcome Building of Genomic Medicine, University of Oxford. Initially studying the regulation of erythropoietin by the kidney, his group was the first to recognize the general role of this pathway in directing cellular responses to hypoxia. He has published more than 100 papers in the field of cellular oxygen sensing and the HIF/VHL/hydroxylase pathway, and in 2002 was elected to the Fellowship of the Royal Society.

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook

Post a comment