Journal of Molecular Medicine

, Volume 85, Issue 2, pp 139–148

Hypoxia-induced genetic instability—a calculated mechanism underlying tumor progression

Authors

    • Department of NeurosurgeryUniversity of Utah School of Medicine
  • Ranjit S. Bindra
    • Department of Therapeutic RadiologyYale University School of Medicine
  • Peter M. Glazer
    • Department of Therapeutic RadiologyYale University School of Medicine
  • Adrian L. Harris
    • Cancer Research UK Growth Factor Group, Weatherall Institute of Molecular MedicineJohn Radcliffe Hospital
Review

DOI: 10.1007/s00109-006-0133-6

Cite this article as:
Huang, L.E., Bindra, R.S., Glazer, P.M. et al. J Mol Med (2007) 85: 139. doi:10.1007/s00109-006-0133-6

Abstract

The cause of human cancers is imputed to the genetic alterations at nucleotide and chromosomal levels of ill-fated cells. It has long been recognized that genetic instability—the hallmark of human cancers—is responsible for the cellular changes that confer progressive transformation on cancerous cells. How cancer cells acquire genetic instability, however, is unclear. We propose that tumor development is a result of expansion and progression—two complementary aspects that collaborate with the tumor microenvironment—hypoxia in particular, on genetic alterations through the induction of genetic instability. In this article, we review the recent literature regarding how hypoxia functionally impairs various DNA repair pathways resulting in genetic instability and discuss the biomedical implications in cancer biology and treatment.

Keywords

DNA repairGenetic instabilityHIFHypoxiaTumor microenvironmentTumor progression

Abbreviations

DSB

double-strand break

HIF

hypoxia-inducible factor

HR

homologous recombination

HRE

hypoxia-responsive element

MMR

mismatch repair

NER

nucleotide excision repair

NHEJ

nonhomologous end-joining

Introduction

Decades of cancer research through arduous and circuitous paths have come to a conclusion that cancer is in essence a genetic disease acquiring dynamic changes in the genome [1, 2]. The development of cancer in humans requires a complex succession of genetic alterations that confer a selective growth advantage on cells undergoing progressive transformation into cancer cells over time. Such genetic changes, including chromosomal translocation, gene amplification, intragenic mutation, and gene silencing, are responsible for the activation of oncogenes and the inactivation of tumor-suppressor genes. The aberrant expressions of these genes collaboratively drive the neoplastic growth by following a set of emerging genetic and biochemical rules applicable to most and perhaps all types of human cancer [3]. What dictates malignant growth collectively has boiled down to six acquired capabilities: self-sufficiency in growth signals, insensitivity to growth inhibitory signals, evasion of apoptosis, limitless replicative potential, sustained angiogenesis, and tissue invasion and metastasis [1], all of which arise from defects in regulatory circuits that govern normal cell proliferation and homeostasis including p53 and RB pathways, receptor tyrosine kinase pathways, and apoptosis pathways [2, 3].

An alternative theory proposed here is that the tumor development comprises two complementary aspects: expansion and progression (Fig. 1). Whereas the former embodies quantitative change primarily through incessant tumor growth with sustainable angiogenesis and glycolysis, the latter signifies qualitative transformation involving the acquisition of malignant capabilities by means of genetic alteration. Furthermore, the two aspects cooperate by interacting with the tumor microenvironment, characterized by hypoxia, low pH, and nutrient deficiency. Thus, genetic mutations of oncogenes and tumor suppressor genes accelerate cell proliferation resulting in tumor hypoxia, whereas this seemingly adverse microenvironment may induce cell-cycle arrest, genetic instability, or even cell death, attempting to confer additional growth advantage for sustained tumor expansion. Such a “stop-and-go” mechanism [4] creates a vicious cycle that underlies the progressive nature of malignant tumors.
https://static-content.springer.com/image/art%3A10.1007%2Fs00109-006-0133-6/MediaObjects/109_2006_133_Fig1_HTML.gif
Fig. 1

Two complementary aspects of tumor development—expansion and progression—are depicted respectively in green and yellow circles, which are surrounded by the tumor microenvironment (oval shaped) with O2 gradient decreasing toward right. Tumor expansion creates a hypoxic microenvironment that drives tumor progression through genetic alteration, and the genetically adapted tumor cells are selected for further expansion, thereby leading to a vicious cycle

Specific genetic changes that alter the foregoing biochemical pathways have been linked to a growing number of hereditary and sporadic cancers, yet the molecular mechanisms underlying these genetic alterations are just beginning to be identified. Numerous studies indicate that hypoxia, one of the key tumor microenvironmental factors, promotes genetic instability [57]. In this article, we review recent advancements in the understanding of molecular mechanisms for the induction of genetic instability by the tumor microenvironment and point out the biomedical implications of these studies in cancer biology and treatment.

Transcriptional response to hypoxia

Enormous progress has been made in the understanding of cellular adaptation to hypoxia at the molecular level since the hypoxia-inducible factor, HIF-1, came into the limelight [8]. This ubiquitously expressed transcription factor comprises two members of the basic helix-loop-helix transcription factor of the PAS superfamily: HIF-1α and ARNT (aka HIF-1β). Whereas both subunits are constitutively expressed at their mRNA and protein levels, HIF-1α is the one that undergoes O2-dependent proteolysis [911] via the VHL E3 ubiquitin-proteasome pathway [1214]. HIF-1α polyubiquitinylation is governed by three HIF prolyl hydroxylases EGLN1, EGLN2, and EGLN3 (aka PHD2, PHD1, and PHD3, respectively) that sense and transduce O2 signals via prolyl hydroxylation to HIF-1α [1519]. Among the three hydroxylases, EGLN1 apparently plays a key role in controlling HIF-1α levels [20]. Prolyl hydroxylation is inhibited under hypoxia, thereby blocking HIF-1α ubiquitinylation and proteolysis. In addition, genetic alterations leading to the aberrant activation of receptor tyrosine kinases such as HER2 also increase HIF-1α accumulation by stimulating protein synthesis [21]. Accumulated HIF-1α dimerizes with ARNT, recruits the transcription co-activator p300/CBP [22], and binds to the hypoxia-responsive element (HRE) [23] in the promoter for transcriptional activation of a battery of hypoxia-responsive genes [24].

Although this well-defined HRE-dependent mechanism accounts for many characteristics of tumor hypoxia, including angiogenesis, glycolysis, low pH, cell survival, and cell mobility, it fails, thus far, to explain genetic instability and cell-cycle arrest that are tightly associated with tumor progression. Recently, a novel mechanism of HIF-1α action has been identified; independent of its DNA-binding and transactivation domains, HIF-1α functionally counteracts oncoprotein Myc by displacing a repressive Myc from the target gene promoter for hypoxic up-regulation of cell-cycle genes [25, 26]. The identification of this HIF-1α–Myc pathway has facilitated the elucidation of how hypoxia affects genetic instability (Fig. 2). Furthermore, transcription factors such as E2Fs, although seemingly irresponsive to hypoxic signals, have also been shown to mediate hypoxic inhibition of DNA homologous recombination (HR) by altering their target gene promoter occupancy.
https://static-content.springer.com/image/art%3A10.1007%2Fs00109-006-0133-6/MediaObjects/109_2006_133_Fig2_HTML.gif
Fig. 2

A model depicts threonine (Thr) phosphorylation that differentiates the major role of HIF-2α from that of HIF-1α in the hypoxic response. Whereas non-phosphorylated (− encircled P) HIF-1α competes with Myc for Sp1 binding in the non-HRE promoter, resulting in the down-regulation of DNA repair genes NBS1 and MSH2 and consequently genetic instability, protein kinase D1 (PKD1)-phosphorylated (+ encircled P) HIF-2α primarily engages in the canonical hypoxia-responsive pathway via binding to the HRE

Tumor microenvironment and genetic changes

Several studies using both reporter genes and endogenous loci have demonstrated increased mutation frequencies in cells grown in tumors as compared to grown in culture [2729]. Studies have suggested that hypoxia is specifically associated with enhanced mutagenesis, oxidative DNA damage, DNA strand breaks, and numerous genetic aberrations.

Using a chromosome-based λ shuttle system for mutation detection, Reynolds et al. [29] found that exposure of cells in culture to hypoxic conditions yielded increased frequencies of point mutations at reporter gene loci. Furthermore, multiple hypoxia–reoxygenation cycles resulted in a greater increase in mutagenesis in these studies. More recently, Papp-Szabo et al. [30] reported similar hypoxia-induced increases in mutation frequency using mammary epithelial cells derived from the BigBlue rat, which contains a transgene for mutation detection.

With regard to DNA damage, hypoxia and subsequent reoxygenation can induce oxidative base damage such as 8-oxoguanine and thymine glycols [31, 32]. Oxidative base damage under these conditions may occur in a manner similar to that observed in the phenomenon of reperfusion injury that has been attributed to the increased production of reactive O2 species [33], which can induce numerous types of base damage in DNA including 8-oxoguanine and thymine glycols [34]. Hypoxia appears to occur transiently and heterogeneously in the tumor microenvironment, suggesting that hypoxia–reoxygenation cycles occur frequently within tumor tissue, potentially leading to significant oxidative base damage.

Hypoxia–reoxygenation cycles are also associated with other genetic aberrations, including the induction of fragile site regions in chromosomes susceptible to breaks [35] and gene amplification, as well as DNA over-replication [36, 37]. Intriguingly, research suggests that these processes are tightly linked, as both fragile site induction and DNA over-replication appear to be associated with substantial increases in gene amplification. For example, Rice et al. [38] demonstrated that exposure of Chinese hamster ovary cells to cycles of hypoxia–reoxygenation resulted in substantial DNA over-replication in a fraction of cells, and that these cells had become resistant to methotrexate as a result of dihydrofolate reductase gene amplification.

Coquelle et al. [39] elegantly demonstrated that hypoxia activates fragile sites that trigger breakage-fusion-bridge cycles for gene amplification. These cycles lead to the formation of double minutes and, with targeted reintegration into chromosomal fragile sites, the subsequent generation of homogeneously staining regions in chromosomes containing amplified genes. Interestingly, approximately 90 common fragile site regions have been identified in the human genome [40]. It is thought that genes located within these unstable genomic regions are highly susceptible to deletions and rearrangements (and possible gene amplification as described above), and they may play a role in tumor initiation and progression [41, 42]. Thus, further studies are necessary to assess whether genes within the common fragile site regions are specifically modified in response to hypoxia–reoxygenation cycles. This will enhance characterization of the spectrum of gene amplifications and genomic alterations under these conditions. Collectively, these phenomena constitute a source of genetic instability induced by hypoxia, thus potentially accelerating the multi-step process of tumor progression.

Mechanisms of DNA repair gene regulation by hypoxia

Mismatch repair

Mutation in the human DNA mismatch repair (MMR) genes such as MLH1, MSH2, and MSH6 are associated with the development of both hereditary and sporadic cancers [43]. Several independent studies have shown that hypoxia/ischemia functionally impairs the MMR system by inhibiting gene expression [4446].

Mihaylova et al. [44] first reported that the MLH1 gene was down-regulated specifically by hypoxia. The MLH1 down-regulation was accompanied by destabilization of its binding partner PMS2, thereby giving rise to a reduction in MutLα activity for mismatch recognition. Consistently, functional MMR deficiency under hypoxia was correlated with increased mutagenesis of episomal and chromosomal reporter genes. The histone deacetylase inhibitor trichostatin A was found to block MHL1 inhibition, suggesting the involvement of histone deacetylation in the hypoxic repression. However, how histone deacetylation selectively contributes the hypoxic repression of MLH1 gene remains to be determined. Later, Koshiji et al. [45] reported that the MutSα genes, MSH2 and MSH6 responsible for the initiation of MMR, were specifically down-regulated by hypoxia via the HIF-1α-Myc pathway in the human colon cancer cell line HCT116. The MSH2 and MSH6 down-regulation was strictly HIF-1α and p53 dependent. This finding was corroborated by the observation that MSH2 protein levels were inversely associated with HIF-1α over-expression in human sporadic colon cancers in a p53-dependent way [47]. Mechanistically, HIF-1α displaces an activating Myc from binding to Sp1 on the MSH2 promoter, resulting in gene repression (Fig. 2). As expected, no Myc displacement was observed in p53-null HCT116 cells. Although the MSH2 and MSH6 down-regulation was observed in wild-type p53 cells, including normal human primary cultures, enhanced microsatellite instability by hypoxia was detectable only in HCT116 cells deficient in MLH1. In p53-inactivated human cervical cancer HeLa cells, MLH1, but not MSH2 and MSH6, was modestly inhibited, consistent with the finding by Mihaylova et al. [44]. However, it remains to be demonstrated that the impaired MMR activity by hypoxia directly contributes to gene mutation critical for tumor development. Interestingly, Shahrzad et al. [46] addressed this question by focusing on the accelerating mutation rate of oncogene KRAS in MMR-deficient cells experienced with combined hypoxia and hypoglycemia or derived from xenografted tumors. The greater incidence of K-rasG13D mutation was paralleled by lowered MSH2 levels in these cells. Likewise, the MSH2 protein level was inversely correlated with the hypoxic areas in the HCT116 xenografted tumors. However, the authors found that hypoglycemia rather than hypoxia was critical in vitro for the reduction of MSH2 protein levels in MMR-deficient TP53+/+ HCT116 and TP53−/− DLD-1 cells, but not in MMR-proficient, TP53 mutated Caco-2 cells. Moreover, the increased mutagenesis was unrelated to oxidative damage. Although further investigations are warranted to resolve the differences among these studies, it is evident that the tumor microenvironment inhibits MMR activities by down-regulating MMR gene expression and promotes genetic instability through gene mutations.

Nucleotide excision repair

It has been shown that exposure to hypoxic stress can suppress the nucleotide excision repair (NER) pathway [48]. Specifically, Yuan et al. demonstrated that hypoxia inhibited the reactivation of an ultraviolet (UV)-damaged plasmid, which was directly correlated with an increase in UV-induced mutagenesis in cells. Interestingly, the decreased expression of NER-related proteins under hypoxia was not detected in these studies. More recently, Bindra et al. [49, 50] again found no significant changes in NER gene expression patterns after exposure to hypoxic stress in a microarray-based analysis of DNA repair gene expression in cancer cells. Based on these findings, it was hypothesized that the unique metabolic conditions induced by hypoxia (e.g., decreased adenosine triphosphate [ATP] production [47]) may impair the enzymatic activity of specific NER proteins, leading to a functional decrease in NER repair under these conditions.

Double-strand break repair

In addition to genetic changes at the nucleotide levels, chromosomal instability is more commonly observed in cancers [2]. Earlier studies showed that transient hypoxia induces DNA over-replication and amplification of drug-resistant genes encoding dihydrofolate reductase and P-glycoprotein [38, 51]. Furthermore, as demonstrated by the causal relationship between hypoxia and induction of fragile sites and gene amplification, hypoxia drives fusion and reintegration of double minutes into fragile sites, thereby generating homogeneously staining regions [39]. Specifically, Coquelle et al. observed that a 5-h severe hypoxia treatment gives rise to a few DNA breaks per mitosis. These reports underscore the role of hypoxia in inducing DNA double-strand breaks (DSBs).

However, when the comet assays were used to detect DNA damage by hypoxia, no discernible damages were detected [52], whereas reoxygenation after hypoxia induced significant levels DNA damage [53]. It was proposed therefore that hypoxia activates ATR through DNA replication arrest, whereas reoxygenation increases ATM activity. Yet, Hammond et al. [54] later detected DNA damage directly in hypoxic S-phase cells with the inactivation of ATR activity, suggesting the presence of a cellular protection mechanism in response to DNA damage. Similarly, Nelson et al. [55] demonstrated cogently that even when p53 function was compromised, carcinomas derived from apoptosis-defective cells displayed aberrant metaphases and progressive polyploidy changes resulting from the hypoxic tumor microenvironment. Presumably, apoptosis is an essential part of the hypoxic response that eliminates cells suffering from DNA damages beyond repair. Conversely, abrogation of this process facilitates the accumulation of aberrant chromosomes and leads to genomic instability and tumorigenesis. Hence, these studies indicate that cells not only experience DNA damage under hypoxia but also actively respond to it.

How does hypoxia inflict damages on DNA? Several recent studies indicate that hypoxia down-regulates genes involved in DNA DSB repair. Kim et al. [56] showed that hypoxia down-regulates the expression of RAD21 (human homolog of rad21 S. pombe) in various tumor cell lines, although the biological consequence of such down-regulation is unknown. Recently, Meng et al. [57] surveyed the expression of a group of HR and nonhomologous end-joining (NHEJ) genes after severe hypoxic (0.2% O2) treatment of normal diploid fibroblasts and premalignant and malignant prostate cell lines. In both normal and malignant cultures, many of the HR-related (RAD51, BRCA1, BRCA2) and NHEJ-related (XRCC3, XRCC4, XRCC6 aka Ku70, and LIG4) genes were down-regulated at the mRNA levels. However, at the protein level, only few (e.g., RAD51 and XRCC3) were inhibited. It was unclear whether HR and NHEJ were functionally impaired (see below). More recently, To et al. [58] reported that HIF-1α, specifically its PAS-B domain, mediates hypoxic repression of NBS1, a component of the MRE11A–RAD50–NBS1 complex critical for the DNA damage response [59, 60]. The NBS1 down-regulation was p53-independent and at least in part responsible for hypoxia-induced DSBs, as indicated by marked increase in γ-H2AX foci, which were well co-localized with another DSB repair protein 53BP1. More importantly, they demonstrated that differential phosphorylation distinguishes HIF-1α from its isoform HIF-2α (aka EPAS1) in NBS1 repression. Of note, HIF-2α resembles HIF-1α biochemically but differs biologically [61]. This functional distinction stems from phosphorylation of HIF-2α by protein kinase D1, thereby precluding NBS1 repression. The study not only delineates a molecular pathway that functionally distinguishes HIF-1α from HIF-2α, but also argues a unique role for HIF-1α in tumor progression by promoting genomic instability (Fig. 2).

Homologous recombination

Two of the recombinational repair-associated genes, RAD51 and BRCA1, have been shown to be specifically down-regulated by hypoxia [49, 50, 57, 62]. These genes were identified in the microarray-based analysis of DNA repair gene expression described earlier [49]. In the case of RAD51, analyses of RAD51 promoter activity, as well as of RAD51 mRNA and protein stability, indicated that the hypoxia-mediated regulation of this gene occurs via transcriptional repression. Importantly, the decreased levels of Rad51 under hypoxia were found to be independent of cell-cycle profile or HIF expression. RAD51 mRNA was down-regulated under hypoxia in both G1- and S-phase cells. Hypoxia-mediated Rad51 down-regulation in vivo was also confirmed via immunofluorescent image analysis of experimental tumors in mice. Functionally, it was demonstrated that hypoxia was associated with a reduction in the capacity of cells to carry out HR, as detected using an episomally based DSB repair assay. The region in the RAD51 promoter that mediates repression by hypoxia was narrowed down to an approximately 300-base-pair fragment within the core promoter by deletion studies with a promoter-luciferase construct (Bindra et al., unpublished studies), and studies are currently underway to determine the role of specific regulatory elements within this region.

With regard to the regulation of the BRCA1 gene, two E2F sites were identified within the proximal promoter of the gene, which together mediate repression of BRCA1 promoter activity after exposure to hypoxia [50]. The E2F family of transcription factors can be subdivided into activating (E2F1-3a) and repressive (E2F3b-5; 6–8) proteins, and the transcriptional activities of these factors are dependent on their interactions with members of the Dp family (Dp1 and Dp2) and the repressive pocket proteins (Rb, p130 and p107) [6366]. Rb associates with E2F1 in cells, whereas p130 and p107 preferentially associate with E2F4 and E2F5, and these complexes are thought to recruit other cofactors that together repress gene expression via binding to E2F sites at target gene promoters [6366]. Using the technique of quantitative chromatin immunoprecipitation, Bindra et al. found that hypoxia induced a dynamic shift in BRCA1 promoter occupancy from activating E2F1 to repressive E2F4/p130 complexes in cancer cells, resulting in BRCA1 transcriptional repression. Inactivation of pocket protein function (and consequently E2F4 repression) using the human papillomavirus E7 oncoprotein [67] completely abrogated the down-regulation of BRCA1 expression by hypoxia, which further confirmed the role of repressive E2F4/p130 complexes in BRCA1 repression under these conditions. As observed in the case of RAD51, it was also found that hypoxia repressed BRCA1 expression via a HIF- and cell cycle-independent manner. Interestingly, Koshiji et al. [25] previously reported down-regulation of BRCA1 expression by HIF-1α, although these decreases were observed in different cell lines from those tested in the above analyses.

Using a chromosome-based DSB repair assay, it was found that hypoxia significantly down-regulates HR activity, which is consistent with previous studies of HR activity using an episomal reporter system in hypoxic versus normoxic cells as described above [49]. Intriguingly, hypoxia did not significantly affect the expression or activity of two key NHEJ proteins, XRCC5 aka Ku80 and XRCC6, suggesting that this pathway remains intact in hypoxic cells. Along these lines, Um et al. [68] demonstrated substantial increases in the kinase activity of another NHEJ-related protein, PRKDC aka DNA-PK, under hypoxia.

Collectively, the findings discussed above indicate that specific genes within the HR repair pathway are down-regulated in hypoxia, which is associated with substantial suppression of recombinational repair activity in cells. It has been proposed that the BRCA1 protein functions as a caretaker of genomic integrity through its role in repairing DNA DSBs instead of by directly inhibiting cell growth [69, 70]. Specifically, recent studies have suggested that BRCA1 functions as a tumor suppressor primarily through its role in promoting high-fidelity HR while simultaneously suppressing the error-prone NHEJ pathway [69, 71]. Whereas NHEJ appears to be unaffected under these conditions, a novel mechanism of hypoxia-induced genetic instability based on these findings that hypoxia suppresses HR activity has been proposed involving the inappropriate shunting of DSBs from the HR repair pathway to the NHEJ pathway under these conditions. Importantly, the down-regulation of BRCA1 and RAD51 expression likely underlies this process in hypoxia.

Biomedical implications

Genetic instability and tumor progression

So why does hypoxia repress DNA repair, thereby inducing genetic instability? The answer may lie in O2 homeostasis through conservation of the intracellular ATP [45, 47]. Accordingly, hypoxia not only stimulates angiogenesis and glycolysis to maintain intracellular ATP production, but also curtails ATP consumption by inhibiting cell proliferation and DNA repair. Likewise, energy conservation under hypoxia is also achieved independent of HIF by inhibiting mRNA translation and cell growth resulting from the suppression of multiple key regulators, including eIF2α, eEF2, and mTOR effectors [72]. Furthermore, HIF-1 actively represses mitochondrial function and O2 consumption by inducing pyruvate dehydrogenase kinase 1, which inactivates the tricarboxylic acid cycle for cell survival [73, 74]. Therefore, energy conservation, in addition to energy generation, is an integral part of the hypoxic response.

As mutations are fundamental to life and evolution [75], in the name of cell survival, hypoxia inhibits DNA repair at the expense of potential mutations attempting to drive clonal selection of tumor cells for autonomous growth advantage. However, when and what type of genetic changes take place during the tumor development remains unanswered. Given the essential requirement of p53 in the hypoxic repression of MutSα genes, it was conjectured that the functional impairment of MMR by hypoxia is more likely responsible for incipient tumorigenicity because the majority of full-blown human cancers harbor TP53 mutation [45]. In contrast, hypoxia-induced DSBs may contribute to a later stage of tumor progression by inducing chromosomal instability. Although hypoxic induction of DSB may occur irrespective of the p53 status, whether cells can tolerate the resulting chromosomal aberrations depend presumably on the biological integrity of the cells. Hypoxic cells defective in p53 and apoptosis have a greater propensity to acquire genomic instability during tumor progression [55]. Conversely, p53 mutated cells with diminished apoptotic potential are selected by tumor hypoxia [76]. However, drastic genetic alterations arising from extreme hypoxia may account for necrosis, a salient feature unique to malignant but not benign tumors. Due to its stochastic nature and ineffective O2 delivery, the hypoxic response in malignant tumors may be more relevant to clonal selection than to amelioration of the hypoxic stress [77]. On the contrary, growth of benign tumors is apparently accompanied by adequate vascularity, thereby less likely to bear severe hypoxia necessary for tumor progression.

Chemotherapy

The majority of anticancer chemotherapy drugs interact with DNA directly or indirectly, e.g., cross-linking with cisplatin and induction of DSBs with topoisomerase 2 inhibitors such as epirubicin and etoposide. There is a very extensive literature on the ability of hypoxia to induce drug resistance, particularly to anthracyclines such as epirubicin, but also to other drugs including platinum. The mechanisms have not been well defined apart from anthracyclines where an acidic extracellular pH blocks uptake into the cells. This is certainly one of the major explanations, but recently Unruh et al. [78] showed in transformed mouse embryonic fibroblasts deficient in HIF-1α, there was a greater sensitivity to carboplatin, etoposide, and radiation. This was independent of p53, and in vivo experiments confirmed the in vitro results. Agents that did not cause DNA DSBs, such as DNA synthesis inhibitors, had equal effects on HIF-1α positive and HIF-1α negative cells. There was a decreased repair of a fragmented reporter gene in normoxic HIF-1α deficient cells, suggesting that basal HIF-1α expression is required for DSB repair gene expression, although the detailed mechanism was not understood in these mouse embryonic fibroblasts.

More recently, silencing HIF-1α expression by RNAi in human non-small cell lung cancer cell lines was reported to decrease resistance to cisplatin and doxorubicin by modest degrees of hypoxia of 0.5% O2 [79]. This confirms the previous study although, again, the mechanism was not described. It is important to also evaluate the P-glycoprotein in these studies of doxorubicin because P-glycoprotein is known to be a HIF target and affects transport of drugs such as epirubicin and doxorubicin. In sum, this area of research needs more intensive investigation with detailed biochemical pathways linked to drug resistance and sensitivity studies.

Radiation therapy

Hypoxia is one of the major mechanisms causing radio resistance and it is mainly due to lack of O2 radicals generated by radiation to create DNA damage. However, there are many physiological processes regulated by HIF that could also contribute to radiation resistance. Moeller et al. [80, 81] have partly elucidated the complex role of radiation interaction with HIF. They showed that with conventional doses of radiation used therapeutically, HIF-1 was induced within 24 h in tumors of animal models, along with vascular endothelial growth factor (VEGF) induction. One explanation is that oxygenated cells were killed by radiation therapy, and this allowed vessels to expand and reoxygenate the hypoxic areas. Although this would have been expected to switch off HIF, in fact, free radicals produced by this pathway induced HIF. The source of the free radicals is not yet clear but may well be inflammatory cells or macrophages induced by the inflammatory response to radiation. The effects of HIF induction were complex in that HIF induction induced angiogenic factors that helped endothelial cells survive radiation damage, and therefore HIF appeared to cause radiation resistance from the point of view of allowing survival of the vasculature.

On the other hand, HIF-1α produces cell-cycle blockade by modifying p21 or p27 expression. In addition, it may interact with p53 so in the few tumor types that have wild-type p53 apoptosis may be enhanced, and this was shown also in appropriate p53 plus or minus control cell lines. Thus, from the tumor cell point of view, HIF-1α expression may enhance cell death through p53-dependent pathways and cell cycle regulation but enhance survival of vasculature, and the final effect will be complex depending on individual genetic factors in the tumor and in the interaction with the endothelial cells [82]. Those studies also showed that HIF-1α on balance radiosensitizes tumors by keeping them proliferating when glucose levels are low because of the HIF effect on promoting glycolysis and maintaining ATP levels.

However, using another model cell line of the mouse hepatoma HEPA1 with a HIF deficient mutation (in ARNT), it was found that in vivo HIF-1 defective tumors were more readily responsive than the parental ones. The mutant cells induced less angiogenesis, but they did not have any difference in oxygenation status and in this case radiosensitizers did not further sensitize these radio-responsive xenografts [83]. In contrast, HIF-1α proficient xenografts were sensitized by misonidazole. In this model, the lack of HIF was associated with radiosensitivity, but this did not appear to be due to O2 or vascular effects.

Despite being compatible with the previous reports in terms of the in vivo effect, this was a cell autonomous effect as shown by mixing experiments whereby the overall sensitivity depend on the proportion of HIF deficient cells, and this implies that the mechanism differs and is not due to a paracrine vascular effect. Further work needs to be done to characterize the vascular response and free radical response in this model for comparative purposes.

Cautionary remarks

The clear biochemical data that DNA repair is suppressed in hypoxia and multiple pathways are modulated by HIF-1 may appear to contradict with direct in vivo experimentation, which often suggests that HIF-1α mediate resistance. These observations are not necessarily contradictory but emphasize how complex the in vivo microenvironment really is and that hypoxia is one aspect, but there are also problems in low glucose and nutrition, as well as acid pH and the interactions of angiogenesis and inflammatory response from the stroma. The DNA defects might be expected to give hypersensitivity to radiation and anthracyclines and DNA binding agents, but if the factors that induce hypoxia, i.e., poor blood supply, contribute to poor drug delivery and reduce oxygenation and reduce free radical production then, in vivo, resistance might be the outcome. Furthermore, impaired DNA repair may confer resistance to chemotherapeutic drugs and more importantly massively increasing mutation frequency in a time- and pathway-dependent way. Defective repair may allow cell death and replication on a damaged template during early stage of drug treatment, but later clonogenic survival of mutants repopulate with resistant cells. Therefore, in the short term, defective DNA repair may be a disadvantage, but in the longer term, it is a survival advantage depending on the balance of mutation frequency versus death, which will vary from drug and pathway.

That resistance might be even worse if HIF was inhibited because then the interaction of poor delivery plus enhanced repair might be detrimental. On the other hand, it does appear that the HIF mediates angiogenic responses, and other adaptive responses to hypoxia might be proportionately more important than the change in DNA repair and therefore blocking HIF-1 function should be a positive therapeutic outcome. There are no drugs yet that specifically inhibit HIF, but drugs that block downstream pathways such as VEGF, nitric oxide synthase, targeted kinase receptor inhibitors affecting vessels, and proteasome inhibitors are all producing major substantial clinical benefit. Thus, although caution is necessary in considering the blockade of HIF, on balance the multiple pathways mediated in vivo seem to suggest that HIF is mainly protumourigenic and blockade of this pathway is likely to be of substantial clinical benefit [24]. However, it is clear that there are different DNA repair pathways involved and better understanding of how they interact with specific chemotherapy in vivo and in vitro, with down-regulation of HIF-1 needs to be undertaken. The possibility of selective repair of genes under hypoxia needs to be investigated. Finally, of course there are also HIF-2 and HIF-3 that need to be analyzed and could be relevant in different tumor types, particularly, for example, renal cancer [84, 85].

Tumor expansion, as measured by size, weight, and cell proliferation and death, etc., has been the gold standard in cancer research to determine the efficacy of experimental therapeutics. Yet, any genetic alterations critical for tumor progression are often overlooked, chiefly due to a dearth of user-friendly assays. Considering that the tumor size does not necessarily correlate with the degree of its malignancy, a smaller cancer may be more malignant than a larger one. Furthermore, as cancer cells experience genetic instability under various types of stress [27], it is conceivable that an “anti-cancer” agent may in the beginning markedly inhibit tumor expansion by blocking cell proliferation and enhancing cell killing and yet concomitantly exacerbate tumor progression through the induction of genetic instability. Such “win the battle but lose the war” therapies are still a harsh reality because cancer cells constantly evolve to adapt to changes in their microenvironment for survival. Therefore, it is imperative that methodologies for detecting genetic changes be developed to better determine tumor progression apart from tumor expansion, so that the efficacy of potential anti-cancer drugs can be accurately assessed. After all, cancer is not merely a proliferative disease but a genetic one enabling adaptation and evasion.

Copyright information

© Springer-Verlag 2006