Keywords

1 Hypoxic Response

Oxygen is directly involved in a wide range of physiological pathways essential for maintaining and promoting homeostasis such as injury response and blood pressure adaptation [2], as well as in pathological processes such as inflammation [3] and tumor formation. A key to understanding such regulations could be accomplished through elucidating the molecular mechanisms by which cells respond and adapt to insufficiency in oxygen supply, a phenomenon known as hypoxic response. Although hypoxia is characterized by suppression in both ATP and protein production as mechanisms to reserve energy, interestingly, there is an abundance of a wide spectrum of genes during the low oxygen status [4, 5]; these genes are referred to as hypoxia-responsive genes. It was reported that 2% of the entire human genome is involved in hypoxia response via interaction with what is known as hypoxia-inducible factors (HIFs), both in a direct and an indirect fashion [6]. Activation of the hypoxia-responsive genes serves to protect cells from the harmful ramifications of oxygen deficiency such as ischemia, particularly that many metabolic and energy-related pathways are controlled by these genes [7, 8]. In addition, hypoxia-responsive genes were found to be heavily involved in the embryonic development. For instance, the deletion of HIF-1α in a mouse embryo leads to death at day 10 [9, 10]. Altogether, hypoxic response is responsible for the activation of a global network of genes that through diverse mechanisms aim to maintain tissue integrity and promote cell survival [6, 11, 12]. Once activated, HIF transcription factors binds to specific DNA sequences unique to their target genes. These DNA regions are called hypoxia-response elements (HREs) [13,14,15,16]. Many factors could determine the potential binding between HIFs and HRE, including HIF-1 protein concentration, oxygen tension, availability of cofactors, and posttranslational protein modifications [17, 18].

2 Hypoxia-Inducible Factor

Hypoxia-inducible factors (HIFs) are members of a family of transcription factors that are involved in the adaptive responses to hypoxia. Structurally, all HIFs are composed of an alpha and a beta subunit with both subunits belong to the family of basic-helix-loop-helix PAS (Per-Ahr/ ARNT-Sim) family of transcription factors. The three HIF members are named HIF-1, HIF-2, and HIF-3. Both HIF-1 and HIF-2 are heavily involved in the response to hypoxia through their transcriptional activities, whereas the role of HIF-3 in this context remains ill-defined [19]. This family is characterized by its conserved domains for both DNA binding and target specificity [20]. Many characteristics such as heterodimerization, hypoxia-mediated stabilization, and transcriptional activity are shared among all three isoforms [21,22,23,24].

HIF-1 transcription factor is a heterodimer consisting of two subunits: an oxygen-labile alpha subunit (HIF-1α) and a stable, constitutively-expressed, oxygen-independent beta subunit (HIF-1β). In hypoxia, HIF-1α is the primary responder, and so, its stability is of greater impact when compared to ARNT [25,26,27,28,29]. Nevertheless, ARNT was shown to be required for the HIF1 overall activity such as binding to other bHLH proteins [30,31,32,33].

Structurally, HIF-1α contains two transactivational domains (TADs), the N-terminal (N-TAD) and the C-terminal (C-TAD), that drive the functional interaction with RNA polymerase. TADs also are the sites where the interaction with coactivators is established, a process that has an indispensable role in HIF-1α transcriptional activity. Furthermore, TADs are the domains where posttranslational modifications take place [17, 25, 34, 35]. Interestingly, upon hypoxia-mediated HIF1 activation, only its protein levels increase, while mRNA expression remains unchanged. This observation highlights the direct proportional relation between oxygen concentration and protein translation and stability [25, 36]. Lastly, the oxygen-dependent degradation domain (ODD domain) serves as an oxygen sensor site where oxygen-dependent interactions take place [37].

HIF-2 proteins have had many names, such as endothelial PAS protein 1 (EPAS1), HIF-related factor (HRF), HIF-1α-like factor (HLF), and member of PAS family 2 (MOP2) [21,22,23,24]. On the level of the protein’s primary structure, HIF-2α is very similar to HIF-1α with almost half the amino acids being identical between the two. In particular, the two proteins are sharing 70% and 83% homology in their PAS and bHLH domains, respectively. Moreover, both isoforms are subjected to the same regulatory mechanism owing to the presence of two critical proline residues in their ODD domains [38,39,40]. Moreover, both HIF-1α and HIF-2α contain N-TAD and C-TAD, while HIF-1β contains only C-TAD (Fig. 1).

Fig. 1
figure 1

Schematic of the structure of three HIFα and two HIFβ isoforms. NLS, nuclear localization signal; bHLH, basic helix-loop-helix domain; PAS, per-arnt-sim domain subdivided into PAS A and PAS B; ODD, oxygen-dependent degradation domain; TAD, transactivation domain. HIF-1α and HIF-2α have two distinct TADs, in the C- (C-TAD) and N- (N-TAD) terminal domains. The PAS and bHLH domains are dedicated to dimerization and recognition of target DNA sequences. (Reprinted with permission from [41])

Full size image

HIF-2 expression was thought initially to be exclusive to vascular endothelial cells, specifically in embryonic tissues [35, 42, 43]. Later, HIF-2 protein expression was confirmed in several adult hypoxic tissues [42]. In cancer tissues, HIF-2 expression was also reported to be upregulated, suggesting a potential role in cancer angiogenesis [44] especially with the protein’s preference for vascular and stromal tissues [43, 45]. As such, higher levels of HIF-1 are observed in tumor epithelial cells when compared with HIF-2, whereas the opposite is true in macrophages and endothelial cells [46]. In cancer, HIF-2 expression was shown to be directly proportional to the pathological staging of a number of solid cancers such as non-Hodgkin lymphoma [45, 47] and bladder cancer [43]. One explanation for such correlation is that the presence of HIF-2-positive macrophage populations in cancer tissues is beneficial to the tumor microenvironment (TME), therefore inversely affecting patient’s survival. Another explanation is related to the role played by HIF-2 in promoting cancer angiogenesis and vascularization [45], since VEGF, the master angiogenic protein, co-reside in stromal tissues as well [48]. Many studies have confirmed the correlation between the HIF-2 and VEGF proteins [43, 49,50,51,52].

Noteworthy, many studies reported a contradicting role of HIF-2α in cancer. For example, loss of HIF-2α in KRAS lung tumor increased tumor aggressive behavior [53], whereas overexpression and stabilization of HIF-2α protein in an identical tumor model promoted tumor angiogenesis and invasion by increasing the expression of VEGF and SNAIL [54], respectively. The observation that opposite HIF-2α expression profiles mediated tumor growth in the same tumor context, albeit by different mechanisms, suggests that effective targeting of HIF-2α subunit in cancer treatment may be complicated.

The third isoform is referred to as HIF-3 [55], and although it shares a significant structural similarity with the other two isoforms, it is reported that its main function is to inhibit HIF pathway [56]. On the other hand, other studies showed that HIF-3 has a dual action of both stimulating and suppressing other HIF members. Therefore, the role of HIF-3 in hypoxic and cancer tissues in particular is yet to be elucidated [57,58,59,60].

HIF β protein was first discovered in the neural tissues where involvement with neural development was assumed [61]. It is also known as the aryl hydrocarbon receptor nuclear translocator (ARNT) [16, 62, 63]. There are two forms of ARNT termed ARNT1 and ARNT2 [64, 65] with both forms are capable of forming a heterodimer with the HIF α isoforms, an interaction crucial for HRE binding and subsequent downstream effector gene activation [65]. Currently, HIF-1α, HIF-2α, and ARNT1 are viewed as the key molecules involved in HIF pathway in response to hypoxia, especially in tumor tissues, while the function of both HIF-3α and ARNT2 is still under investigation. In this review, we will focus on the functional role of HIF-1α in driving hypoxia response in human cancer.

2.1 Discovery of HIF-1

HIF-1 was initially viewed as an essential and exclusive key element in the human erythropoietin (EPO) gene in response to oxygen insufficiency in renal tissues [66]. At that time, the concept of direct oxygen sensing, which cells can independently and directly sense and respond to changes in oxygen level, was still developing. Subsequently, the novel discovery that the function of HIF-1 is not EPO gene-restricted and that it regulates other genes is considered a milestone in the field of direct oxygen sensing. HIF-1 was later purified [14], and shortly, the protein molecule was further characterized [62].

Afterward, a wide spectrum of HIF-1 target genes and microRNAs (miRNAs) that were involved in hypoxic response were identified on both genetic and protein levels [67]. Both HIF-1 target gene’s activation and suppression were characterized as being tissue-specific [68]. Moreover, HIF-1-mediated activation of gene transcription may be in both direct and indirect fashion [6]. For instance, by activating miRNAs [67] and gene-suppressors such as DEC1/Stra13 [69], HIF-1 is indirectly silencing certain downstream effector genes. Key biological pathways such as proliferation, energy metabolism, invasion, and metastasis were found to be driven by HIF-1 downstream effector genes [70, 71] (Fig. 2), suggesting the important role of HIF-1-mediated pathways in cancer development and progression.

Fig. 2
figure 2

Representative target genes of HIF-1α and their functions

Full size image

2.2 Regulation of HIF-1α

It was reported that many oncogenes activate HIF-1α pathway mainly via phosphorylation cascades through upregulating the transcription and translation of HIF-1α mRNA and protein, respectively, and independently of oxygen levels [72]. Similarly, growth factors and cytokines such as epidermal and fibroblast growth factors and insulin-like growth factor could activate HIF-1α through the same phosphorylation mechanism [73,74,75,76]. This phosphorylation cascade could promote HIF-1α expression via several pathways. One example is the PI3K/Akt/mTOR-mediated HIF-1α pathway activation as seen in many solid tumors such as in colon [77], prostate [78], and breast cancer [76]. Another mechanism is by enhancing the p300-HIF-1α-C-TAD activation complex [79] and favoring HIF-1α nuclear translocation as seen via MAPK-mediated phosphorylation [80]. MAPK-mediated phosphorylation also promotes HIF-1α transcriptional activity by blocking its nuclear export in an CRM1-dependent fashion [80]. HIF-1α pathway can also be activated by growth factors via ERK-dependent signaling [17]. In addition, vasoactive cytokines may promote HIF-1α transcriptional activity through diacylglycerol-sensitive protein kinase C [81]. Altogether, upregulation of HIF-1α exerted by growth factors and local hormones can overcome its oxygen-dependent degradation [76, 77].

Other kinases such as casein kinase 1 (CK1) were also reported to be involved in HIF-1α phosphorylation [82]. Moreover, it was reported that the phosphorylation status of HIF-1α is linked to the protein’s ability to repair DNA damage and reverse chromosomal instability, two characteristics that are extremely important in driving tumor progression and aggressiveness. For instance, dephosphorylation of HIF1α directly leads to repression of NBS1, a DNA mismatch repair gene [83].

In addition to the kinase signaling pathways, loss of function of the gene suppressor von Hippel-Lindau (VHL) results in activation of HIF-1α protein due to the associated lack of protein-degradation suppression [84]. A dysregulation of key carbohydrate metabolic intermediates was also shown to contribute to HIF-1α regulation independently to oxygen levels. For instance, α-ketoglutarate (α-KG) functions as a cofactor for PDH and FIH-1 (factor inhibiting HIF-1) hydroxylates [85] and is therefore directly involved in HIF-1α regulation.

Posttranslational modifications other than phosphorylation are also critical for HIF-1α protein activity. For instance, hydroxylation of HIF-1α protein by prolyl hydroxylase is viewed as the main regulatory mechanism that guards against HIF-1α protein activation in normoxic conditions. HIF hydroxylases exist in two forms: HIF-prolyl hydroxylase, also known as prolyl hydroxylase domain (PHD) proteins, and HIF-asparaginyl hydroxylase, also known as FIH-1 (factor inhibiting HIF-1) [86].

There are three closely-related isoforms of the PHD protein known as PHD1, PHD2, and PHD3, where PHD2 is considered the most critical under normoxic conditions [87]. PHD enzymes function by hydroxylation of two prolyl segments of HIF-1α protein, where oxygen concentration is an imperative determinant for the reaction initiation [34, 88,89,90]. This interaction takes place at the ODD domain where the two propyl residues reside. Once hydroxylated, HIF-1α develops a strong binding affinity for a part of an E3 ubiquitin ligase complex VHL protein leading to HIF-1α protein degradation by a proteasome. The reversal of this oxygen-dependent degradation process results in an observed increase in HIF-1α protein levels associated with hypoxia.

A second hydroxylation event targets the asparaginyl residue at the C-TAD of HIF-1α protein by FIH-1. FIH-1 reaction depends exclusively on oxygen availability in the ambient environment [86]. This reaction changes HIF-1α protein’s physical properties, such as its water affinity, hindering the interaction between the hydroxylated C-TAD and its coactivators p300/CREB binding protein (CBP) [91, 92]. This reaction will result in C-TAD domain blockage and ultimately HIF-1α transactivation activity inhibition, but not stability, in an oxygen-dependent reversible fashion (Fig. 3).

Fig. 3
figure 3

HIF-1 signaling cascade. Synthesis and constitutive expression of HIF-1α by a cascade involving a series of growth factors and signaling events are indicated. The major differences among the hypoxic and normoxic signaling and sequence of events are also depicted clearly in the flowchart. Normoxia leads to HIF-1α protein degradation whereas hypoxia leads to HIF-1α-regulated target gene expression. The downstream sequence of events leading to tumorigenesis is also portrayed (Modified with permission from [93])

Full size image

Redox sensors are equally important to oxygen sensors in the regulation of HIF-1α-mediated hypoxic response. An example is the SIRT1-mediated acetylation process, which is another critical posttranslational modification of HIF-1α. SIRT1 deacetylates HIF-1α by targeting the lysine amino acid leading to the blocking of p300-recruitment and eventually HIF-1α inactivation [94].

2.3 HIF-1α Stability

Although oxygen tension is considered the main factor governing HIF-1α protein stability during hypoxia through the hydroxylation events discussed earlier, mitochondria can also act as a stabilizer of HIF-1α proteins via increased production of reactive oxygen species (ROS) [95,96,97]. ROS might play a role in protein stabilization mainly through the inactivation of PHD leading to HIF-1α accumulation [98]. Lastly, reports on nitric oxide (NO) effect on HIF are contradictory, with some advocating for HIF-1α stabilization [99,100,101,102], whereas others demonstrating an opposite effect on HIF-1α activity [103,104,105]. Once the protein is stabilized, nuclear translocated, and dimerized with ARNT, hypoxia-responsive genes are activated through HIF-1α binding to a characteristic consensus sequence 5′-(A/G) CGTG-3 termed HRE [106] located in the upstream region of hypoxia-inducible genes [15, 107, 108].

3 HIF-1α and Metabolic Reprogramming

A shift from glucose metabolism coupled with mitochondrial oxidative to anabolic respiration, known as the Warburg effect, is a hallmark of hypoxia. This metabolic shift takes place through the upregulation of oxygen-independent metabolic pathways, such as glycolysis and downregulation of the oxygen-dependent pathways such as mitochondrial respiration [109]. For instance, overexpression of key glycolytic enzymes such as the rate-limiting enzyme phosphofructokinase [110] and the glycolytic flux regulatory enzymes, 6- phosphofructo-2-kinase and fructose-2,6-bisphosphate, is HIF-1α-mediated in hypoxia [37,38,39,40]. Other enzymes such as glucose transporter protein1 (GLUT1) and GLUT3 that are involved in glucose trafficking processes are also the targets of HIF-1α in hypoxia [111, 112]. Noteworthy, there is a positive correlation among cancer pathological staging, GLUT3 and HIF-1α expression and activity levels, a measure that might serve as a prognostic tool [112]. HIF-1α activation also upregulates key enzymes that inhibit acetyl-coenzyme A (acetyl-CoA) production from glucose, therefore inhibiting oxidative phosphorylation. There are two isoforms of the enzyme pyruvate dehydrogenase kinase (PDK) known as PDK1 and PDK3 that directly inhibit acetyl-CoA production and entering into the TCA cycle leading to the shutdown of the oxidative phosphorylation associated with hypoxia. Another approach for cells to shift away from oxidative metabolism is through the activation of mitochondrial autophagy by protein BCL2/adenovirus E1B 19-kDa interacting protein 3 (BNIP3) [113]. BNIP3 functions by activating lactose dehydrogenase A (LDHA) enzyme that converts pyruvate to lactate, therefore promoting the anaerobic respiration.

Similarly, PDK1 and PDK3 were also found to play a role in mitochondrial autophagy [114,115,116]. In addition, miRNA targets of HIF-1α were found to directly favor the metabolic shift through inhibiting genes that are critical for the mitochondria oxidation machinery and independently of the metabolic enzymes [117]. For instance, miR-210 activation [118] inhibits the iron-sulfur cluster assembly enzyme ISCU, which is essential for the mitochondrial electron transport complex I activity [119, 120]. Noteworthy, due to the scarcity of acetyl-CoA in response to hypoxia, tissues with a high proliferation rate such as cancer will utilize glutamine, instead of glucose, to generate α-ketoglutarate essential for fatty acid synthesis [121, 122]. Glutamine utilization as the main source of energy for such high-energy-demanding cells acts as another repelling force for the pyruvate away from the TCA cycle [121, 122]. HIF-1α-mediated fatty acid metabolism dysregulation in cancer hypoxic tissues was associated with poor survival in many solid tumors such as renal cancer [123]. The high-glycolytic-flux signature in hypoxia serves a unique benefit for cancer tissues, other than solely energy benefit, and that is providing precursors of the pyrimidine/purine pathway needed for DNA synthesis for cell proliferation [124]. For example, it was reported that glucose utilization is directly proportional to increased cancer tissue mass and invasion property [125], suggesting a critical role played by glucose metabolism in tumorigenesis.

HIF-1α significantly contributes to the acidic environment of cancer tissues through the activation of plasma membrane proteins [126] such as monocarboxylate transporter 4 (MCT4), encoded by SLC16A3 gene, through controlling lactic acid transport [127]. Another membranous protein named sodium-hydrogen exchanger 1 (NHE1) that is encoded by the SLC9A gene regulates the pH of the environment through protons pumping [128], and the same mechanism is adopted by carbonic anhydrase 9 (CA9) [129]. In fact, the intracellular alkalinization and extracellular acidification enhance cellular proliferation and invasion [130].

An interesting aspect of HIF-1α regulation lies in its activation loop with pyruvate kinase M2 (PKM2) enzyme. PKM2 is a glycolytic enzyme that may play an important role in cancer progression [131] by promoting glycolysis as well as acting as a coactivator for HIF-1α [132]. HIF-1α also activates the transcription of PKM2 leading to the activation of key oncogenes such as STAT3 and its downstream genes, which further enhances the progression of cancer [133].

4 Hypoxia and Cancer

Tumor hypoxia is an example of a chronic, pathophysiological condition, in which response is insufficient to completely reverse the hypoxic insult [134]. Hypoxia in cancer could be defined on the basis of oxygen and energy levels present in tumors. For instance, a concomitant decrease in both oxygen partial pressure and ATP level occurs in a fibrosarcoma model [135]. A key player in cancer hypoxia dynamics is the hypoxia-induced vascular endothelial growth factor (VEGF). Due to the constant hypoxic insult that cancer tissues are exposed to, VEGF-mediated new blood vessel formation to overcome the oxygen deficiency is slow and disordered. This continuous cycle of defective blood vessel architecture and the activation of hypoxia-mediated pathways is a hallmark for tumor microenvironment (TME) [136] as well as its aggressive phenotype [137]. Many solid tumors such as breast and lung, among others, respond to the decrease in oxygen tension by upregulation of HIF-1α [138,139,140]. Correlation between hypoxia and tumor malignant transformation had also been observed [141,142,143].

5 HIF-1α and Immune Cells in Cancer

Many solid tumors are characterized by hypoxia [144] and tumor-associated macrophage (TAM) infiltration [145]. For example, in breast cancer, HIF-1α knockout in TAM caused overstimulation of nitrous oxide (NO) [145, 146], a phenomenon that can put T lymphocytes into anergy status. Hypoxia can also increase the expression of the immune checkpoint programmed death-ligand 1 (PD-L1) on macrophages, dendritic cells, and tumor cells in an HIF-1α-dependent fashion leading to the suppression of effector T-lymphocyte recruitment and activation [147]. Another mechanism that HIF-1α can dampen the antitumor effect of immunity through is the upregulation of regulatory T lymphocytes [148].

6 Role of HIF-1α in Key Cancer Pathways

6.1 Cell Adhesion

Adhesion molecules play a significant role in cancer initiation and progression through promoting its interaction with both intracellular and extracellular environment [149]. HIF-1α is directly involved in the regulation of key adhesion molecules such as β1 integrins and α5β3 and α5β5 expression [150,151,152]. E-cadherin is also regulated by HIF-1α via its direct regulation of TCF3, ZFHX1A, and ZFHX1B [153] and up-regulation of Snail [154] which inhibits E-cadherin gene expression.

6.2 Cell Proliferation

One of the defining characteristics of cancer cells is their uncontrollable proliferation coupled with impairment of cell death pathways and signals awing to overexpression of survival and growth factors; such changes enable cancer cells to adapt to nutritional deprivation or to escape their unfavorable environment. For instance, hypoxia stimulation of VEGF transcription, via the HIF pathway, was shown to be strongly associated with cellular proliferation and metastasis in tumors [155]. Additionally, the expression of hypoxia-mediated telomerase reverse transcriptase (TERT) promotes tumor cells’ immortal phenotype [156]. Simultaneously, hypoxia-induced downregulation of membranous integrins was reported to facilitate tumor cell detachment and new tumor growth [157]. A recent study of ARK5 expression in colon cancer showed that it was upregulated in a HIF-1α-dependent manner and that ARK5 serves an important player in cancer proliferation and migration under hypoxic stress [158]; similar effects were also reported in other solid cancers [159,160,161].

6.3 Metastasis and Invasion

Hypoxia can activate epithelial-to-mesenchymal transition (EMT) via HIF-1α in various types of solid tumors [162,163,164]. HIF-1α can directly or indirectly regulate key EMT regulators, including TWIST, Snail, carbonic anhydrase IX (CAIX), and GLUT-1 [165,166,167,168]. These molecules then trans-activate EMT-related genes, including vimentin, E-cadherin and N-cadherin, to facilitate the progression of the EMT [169, 170]. Matrix metalloproteinase MMP2 and MMP9 have also been reported to be regulated by HIF-1α. The impact of hypoxia-induced MMP-9 expression is extremely central for cellular migration [171, 172]. Besides, two major components of the fibrinolysis system and thus metastasis, named urokinase-type plasminogen activator receptor (uPAR) and plasminogen activator inhibitor-1 (PAI-1), have also been shown to be targets of HIF-1α [173, 174]. TWIST, another essential transcription factor that is involved in hypoxia-mediated EMT and tumor metastasis, is directly regulated by HIF-1α [165]. Other significant HIF-1α target genes directly involved in cancer metastasis are CXC chemokine receptor-4 (CXCR4), c-Met and CC chemokine receptor 7 (CCR7) [175,176,177], lysyl oxidase (LOX) [6, 178], fibronectin, cathepsin D, and urokinase plasminogen activator [11]. HIF-1α also promotes cell invasion through the upregulation of key invasion-promoting genes such as the autocrine motility factor [179], vimentin, and the receptor tyrosine kinase c-Met [175]. Meanwhile, the stromal-derived factor-1, keratins 14, 18, and 19, the cytokine receptor CXCR4P [180, 181], caveolin-1 (CAV1) [182], uPAR, MMP2, cathepsin D, and fibronectin 1, among others, are transcriptionally upregulated by HIF-1α [183].

6.4 Angiogenesis

Angiogenesis plays an essential role in tumorigenesis. HIF-1α can stimulate an angiogenic response by activating a number of growth factor-encoding genes, including VEGF, angiopoietin1 (ANGPT1) and ANGPT2, placental growth factor (PGF), calcitonin receptor-like receptor (CRLR) [184], and platelet-derived growth factor B (PDGFB) [68, 185]. HIF-1α deletion was reported to be associated with abnormal vasculature [186]. On the other hand, a recent report on pancreatic tumor revealed an alternative mechanism by which cancer cells could maintain angiogenesis in an HIF1α-independent manner [187]. Noteworthy, SUMO-specific protease 1 (SENP1), a HIF-1α target enzyme, is of great importance for HIF-1α stabilization in hypoxia. This positive feedback loop is significant for VEGF activation and angiogenesis [188, 189].

6.5 Apoptosis

Although programmed cell death can be directly triggered by deficiency in oxygen levels in both normal and cancer tissues [190], with accompanying DNA damage [191], the direct effect of HIF pathway on apoptosis is reported to range from apposing cell death [192] to promoting apoptosis [193]. One explanation for this variation in HIF pathway response might be related to the degree for hypoxia and the variation in oxygen tension [194]. A second factor for such fluctuation is the presence of several apoptosis-related proteins, such as cyclin D1, p21, and p27 that are targeted by HIF-1α upon activation, and that apoptosis response depends on the expression profiles of these apoptotic molecules [195]. Lastly, the initial energy level of hypoxic tissues is inevitably a key factor in the apoptosis pathway [196].

Other factors such as the mitochondrial membrane integrity could also trigger an apoptotic response through activating key apoptotic mediators, such as caspase 9, independently of HIF-1α pathway [197]. Wild-type tumor-suppressor gene p53 also plays a critical role in hypoxia-induced apoptosis through caspase 9 and Apaf-1 downstream effector [198, 199]. Other key apoptotic molecules such as BNIP3, a member of the Bcl-2 family [200], and Noxa which is a p53-downstream protein that could sense ROS levels [201] have also been identified as targets of HIF-1α.

7 Clinical Significance and HIF-1α Inhibitors for Cancer Therapy

HIF-1α expression levels were positively correlated with tumor progression in a variety of solid tumors such as glioma and breast cancer, where HIF-1α correlates with tumor pathological grade and invasion in the former [202], and overall poor survival rate in the later [203, 204]. HIF-1α is used as a prognostic marker for different treatment modalities in a variety of solid tumors [205,206,207].

The combination of HIF-1α expression with oncogenes or tumor suppressor genes is viewed as another powerful prognostic factor. For instance, in ovarian cancer, the coexistence of mutant p53 expression and HIF-1 overexpression was associated with a poor survival rate [208] and resistance to chemotherapy mainly due to p53-mediated activation of RAS signaling that leads to apoptosis impairment [209]. Recently, a correlation between HIF-1α and the tumor suppressor NEDD4L levels in gastric cancer has been proposed as a prognostic marker [210]. In addition, HIF-1α upregulation combined with the antiapoptotic protein Bcl-2 downregulation in esophageal cancer is associated with treatment failure [211].

HIF-1α inhibition provides an innovative approach for modifying tumor niche with promising clinical results. Unfortunately, and awing to the complex network of genes that are regulated by HIF-1α as well as the multilayered HIF-1α regulation mechanisms, it is challenging to develop a specific HIF-1α inhibitor with a high specificity [212]. Another factor that might tremendously affect the drug discovery process is accuracy and sensitivity of the screening methods. Currently, there are several anti-HIF-1α molecules that are classified according to their target site into direct and indirect inhibitors. Direct inhibitors refer to molecules that target the transcriptional activity of HIF-1α, whereas indirect inhibitors are molecules that target HIF-1α transcription and translation on the mRNA and protein levels, respectively [213]. HIF-1α inhibitors are also classified according to the targeted stage of HIF-1α ranging from the mRNA transcription to protein degradation [214]. In conclusion, the continuous search for the specific HIF-1α inhibitor with fewer side effects and better patient tolerance and survival rate is still ongoing. Noteworthy, combination therapy with other target molecules such as antiangiogenic drugs is showing promising results in animal model studies [215, 216]. A comprehensive understanding of the structure, molecular biology, and regulatory machinery of HIF-1α domains will undoubtedly aid in the development of specific HIF-1α inhibitors.

8 Conclusion

It has been nearly three decades since the novel discovery of HIF-1α as a master regulator of hypoxic response as well as its implication in cancer progression and survival in many solid tumors. Since then, HIF-1α was regarded as a significant and promising target in anticancer therapy. A great deal of research in this area as well as the development of HIF-1α inhibitors have clearly translated such impact. Unfortunately, none of these therapies were proven to be precisely and exclusively targeting cancer, leading to undesirable side effects. Indeed, the involvement of HIF-1α in many aspects of physiological pathways seems to be the main obstacle for perfectly targeting it. Therefore, future research may emphasize more on unfolding all the genes and proteins involved in the HIF-1α pathway, elucidating the molecular mechanisms that regulate other HIF members, and finally aim to discover and target a novel cancer-specific molecule from the HIF-1α downstream effectors expanding pool.