Nanomaterials for Tumor Hypoxia Relief to Improve the Efficacy of ROS-Generated Cancer Therapy

Given the fact that excessive levels of reactive oxygen species (ROS) induce damage to proteins, lipids, and DNA, various ROS-generating agents and strategies have been explored to induce cell death and tumor destruction by generating ROS above toxic threshold. Unfortunately, hypoxia in tumor microenvironment (TME) not only promotes tumor metastasis but also enhances tumor resistance to the ROS-generated cancer therapies, thus leading to ineffective therapeutic outcomes. A variety of nanotechnology-based approaches that generate or release O2 continuously to overcome hypoxia in TME have showed promising results to improve the efficacy of ROS-generated cancer therapy. In this minireview, we present an overview of current nanomaterial-based strategies for advanced cancer therapy by modulating the hypoxia in the TME and promoting ROS generation. Particular emphasis is put on the O2 supply capability and mechanism of these nanoplatforms. Future challenges and opportunities of design consideration are also discussed. We believe that this review may provide some useful inspiration for the design and construction of other advanced nanomaterials with O2 supply ability for overcoming the tumor hypoxia-associated resistance of ROS-mediated cancer therapy and thus promoting ROS-generated cancer therapeutics.


ROS (including singlet oxygen ( 1 O 2 ), superoxide radicals (O 2
•− ), hydroxyl radicals (•OH), and peroxides (O 2 2− )) play a concentration-dependent role in physiological activity (Gorrini et al., 2013). Low to moderate levels of ROS regulate cell signaling and promote cell proliferation, and elevated levels of cellular ROS are one of the unique characteristics of cancer, whereas excessive ROS will induce nonspecific damage to proteins, lipids, and DNA. Because of the heightened basal level of ROS in cancer cells, cancer cells are more susceptible to exogenous ROS, compared to normal cells that maintain redox homeostasis . Therefore, modulation of the ROS level at cancer cells has been emerging as promising strategy for the tumor destruction by generating ROS above toxic threshold. Hypoxia, mild acid, and overexpressed H 2 O 2 are three characteristic features of tumor microenvironment (TME) (Dai et al., 2017;Kwon et al., 2019). Because of the aggressive proliferation of cancer cells and the insufficient blood supply in tumors, the O 2 supply in solid tumors was usually insufficient (partial pressure of O 2 < 2.5 mmHg). Hypoxia in TME not only promotes tumor metastasis but also enhances tumor resistance to the ROS-generated cancer therapies, such as photodynamic therapy (PDT), radiation therapy (RT), chemotherapy, chemodynamic therapy (CDT), and sonodynamic therapy (SDT), thus leading to ineffective therapeutic outcomes. Tumor oxygenation that aims at greatly increasing the oxygen concentrations in hypoxic tumors has been demonstrated to be an effective strategy to overcome tumor hypoxia and enhance the sensibility of hypoxic tumors toward the ROS-generated cancer therapy .
To relieve hypoxia, hyperbaric oxygen therapy, which involves the breath of pure O 2 in a pressurized chamber, has been developed. Unfortunately, its extensive application is limited by the intrinsic side effects including hyperoxic seizures and barotrauma as a result of the overproduced ROS in normal tissues (Kim et al., 2017). Also, angiogenesis inhibitors have been applied to transiently normalize the tumor vasculatures and suppress the consumption of O 2 . However, the oxygenation improvement resulting from the normalization of vessels only lasted for a few days (Liu J. N. et al., 2017). Promoted by recent advances in nanotechnology, a variety of nanotechnology-based approaches that generate or release O 2 continuously to overcome hypoxia in TME have showed promising results to improve the efficacy of ROS-generated cancer therapy. In this minireview, we present an overview of current nanotechnologybased strategies for advanced cancer therapy by modulating the hypoxia in TME and promoting the generation of ROS. To amplify the therapeutic outcomes, the approach of modulating tumor hypoxia was usually applied in combination with other therapeutic/theranostic modalities. This minireview mainly focused on the O 2 supply ability and mechanism of these nanoplatforms. Future challenges and opportunities of design consideration are also discussed and summarized.

Tumor Hypoxia-Regulating Approaches Based on Nanotechnology
Based on their different mechanisms and involved materials, nanotechnology-based tumor hypoxia-regulating approaches can be classified into the following categories: delivering O 2 by natural or artificial oxygen-carrying materials, the hydrolysis of exogenous peroxide, catalytic decomposition of intracellular H 2 O 2 by utilizing catalase or catalase-like nanozymes, and generating O 2 by water-splitting photocatalysts.

Delivering O 2 by Natural and Artificial Oxygen-Carrying Materials
Red blood cells (RBCs), the primary source of O 2 in mammals, contain 270 million hemoglobin (Hb) molecules per cell; each Hb molecule binds up to four O 2 . Hb allows efficient binding of O 2 under high O 2 pressure and rapid O 2 release under hypoxic environment. Because of the good biocompatibility and long circulation, RBCs have been widely investigated as biological drug carriers and O 2 shuttles for cancer therapy (Squires, 2002;Wang et al., 2013;Wang et al., 2014;Sun et al., 2015;Wang et al., 2017). Tang et al. (2016) demonstrated that RBCs tethered with photosensitizers (ZnF 16 Pc) onto the RBCs surface (P-FRT-RBCs) could realize the codelivery of O 2 and photosensitizers ( Figure 1A). The sustained O 2 supply adjacent to photosensitizers by RBCs enabled efficient PDT even under hypoxic conditions. However, the micrometer sizes of RBCs may limit their extravascular diffusion ability and reduce their chance to approach tumor cell. The oxygen-carrying ability of RBCs is limited by the inherent oxygen-binding ability of Hb. However, cell-free Hb suffers from severe problems, including short circulation time, potential side effect, and poor stability. Hb-based O 2 carriers via chemical modification or encapsulation with biodegradable materials could overcome the disadvantages of cell-free Hb and demonstrate the similar oxygen-carrying capability as that of natural RBCs (Gundersen and Palmer, 2008;Duan et al., 2012;Jia et al., 2012;Paciello et al., 2016;Zhou et al., 2016;Cao et al., 2018;Jansman and Hosta-Rigau, 2018;Yu et al., 2018;Hu et al., 2020). Compared to RBCs with micrometer sizes, nanodimensional Hb-based O 2 carriers can perfuse tumor tissues within the narrow vascular structure and thus can supply more O 2 in hypoxic tumor Luo et al., 2016;Zhao et al., 2016). Inspired by the biological nature of RBCs,  developed an aggressive man-made RBC (AmmRBC) as oxygen self-supplied PDT system to combat the hypoxia-mediated resistance of tumors to PDT ( Figure 1B). This biomimetic platform was prepared by encapsulating methylene blue (MB) adsorbed Hb-polydopamine complex into the biovesicle engineered from the recombined RBC membranes. Polydopamine played the role of the antioxidative enzymes to prevent Hb from the oxidation damage during the circulation.
In recent years, an artificial blood product, perfluorocarbon (PFC) compounds with good biocompatibility and high oxygen dissolving ability, has been extensively used as O 2 carriers to modulate the hypoxic TME (Squires, 2002;Lee et al., 2015;Que et al., 2016;Liang et al., 2020). By loading a near-infrared photosensitizer (IR780) into PFCs nanodroplets, Cheng et al. (2015) developed an oxygen self-enriching PDT (Oxy-PDT) nanoplatform ( Figure 1C). Owing to the higher oxygen capacity and longer 1 O 2 lifetime of PFCs, the PDT effect of the loaded photosensitizer was significantly enhanced. Gao et al. (2017) reported erythrocyte-membrane coated PFC nanoparticles as artificial RBCs to deliver O 2 and enhance radiation response.
Though having high oxygen solubility, PFC releases O 2 simply by diffusion through the O 2 concentration gradient, usually resulting in a low delivery efficiency. Using near-infrared (NIR) light or ultrasound (US) as trigger could accelerate the release of O 2 and promote the tumor oxygenation (Song G. S., Chen et al., 2017). Song et al. utilized the photothermal effect of Bi 2 Se 3 induced by NIR laser irradiation to trigger the burst release of O 2 from PFC loaded inside the hollow Bi 2 Se 3 nanoparticles, thereby greatly promoting the tumor oxygenation and overcoming the hypoxia-associated radioresistance of tumors  ( Figure 1D). Song X. J. et al. (2016) used an external lowfrequency/low-power US treatment to trigger the release of O 2 from nano-PFC to relief tumor hypoxia for enhanced PDT and RT ( Figure 1E). Given that several formulations of PFC emulsions have been either approved for clinical application or in late-phase clinical trials as blood substitutes, PFC-based nanomaterials may hold great potential in cancer treatment for future clinical translation. However, extensive exposure to PFCs may cause some side effects, including hypotension, cutaneous flushing, fever, pulmonary hypertension, chest tightness, and elevated central venous pressure (Zhou et al., 2016).

Hydrolysis of Exogenous Peroxide to Produce O 2
Because the hydrolysis of peroxide will generate O 2 , various peroxides (such as hydrogen peroxide, calcium peroxide, sodium percarbonate, and pyridine endoperoxides) have been utilized as O 2 -producing materials (Harrison et al., 2007;Oh et al., 2009;Wang et al., 2011;Li et al., 2012;Pedraza et al., 2012;Benz et al., 2013). However, the release of O 2 by the hydrolysis of exogenous peroxide in the absence of a catalyst or trigger was usually slow and limited. It will be more favorable if on-demand and uniform O 2 delivery to the cells for a sufficiently long time period can be achieved (Liu J. N. et al., 2017). Huang et al. (2016) reported an implantable oxygen-generating depot by coloading CaO 2 and catalase into the Ca 2+ -crosslinked microencapsulated alginate pellets. Catalase (CAT) in the alginate pellets could catalyze the breakdown of H 2 O 2 into O 2 , whereas the Ca 2+crosslinked alginate matrix could temper the hydrolytic reactivity of CaO 2 /catalase by limiting the infiltration of H 2 O into the pellets, thus prolonging the generation of O 2 . Upon implantation close to the tumor, this in situ oxygen-generating depot effectively alleviated the hypoxic regions in tumor and thus resulted in increased chemotherapeutic effect of DOX by promoting ROS production. Liu L. H. et al. (2017) encapsulated CaO 2 and methylene blue (MB) into liposome to fabricate an O 2 selfsufficient nanoplatform (LipoMB/CaO 2 ) to enhance PDT efficacy in hypoxic tumor. CaO 2 inside liposomes could react with H 2 O or weak acid to release O 2 slowly. Upon laser   (2018) reported an in situ free radical polymerization method by using a photosensitizer (mesotetra(p-hydroxyphenyl) porphine (THPP)) as the crosslinker to modify CAT for tumor hypoxia modulation and enhanced PDT.
In the obtained CAT-THPP-PEG nanocapsules, the PEG chains polymerized on the surface of CAT could prevent the direct contact between serum proteins and CAT and thus enhanced the enzyme stability, maintained its catalytic activity, and reduced its immunogenicity. Phua et al. (2019) reported that the integration of hyaluronic acid (HA) with CAT could not only improve the physiological stability of the system but also enable active targeting to tumors. The photosensitizer (Ce6)-loaded nanosystem (HA-CAT@aCe6) could target CD44overexpressed cancer cells, relieve hypoxia by converting endogenous H 2 O 2 to O 2 , and consequently improve PDT efficacy.
Apart from natural enzymes, various nanomaterial-based artificial enzymes show catalase-like activity; one of the typical representatives is MnO 2 . Various MnO 2 nanostructures have been designed and incorporated into multifunctional nanoplatforms to induce the decomposition of endogenous H 2 O 2 into O 2 , thus alleviating tumor hypoxia and improving therapeutic efficacy (Prasad et al., 2014;Fan et al., 2015;Abbasi et al., 2016;Yi et al., 2016;. Moreover, MnO 2 could be decomposed into soluble Mn 2+ in TME, thus reducing unwanted in vivo accumulation and long-term toxicity . The released Mn 2+ could mediate the Fenton-like reaction to convert H 2 O 2 into the highly reactive •OH, further enhancing the therapeutic potency by introducing extra CDT . Apart from the abovementioned benefits, MnO 2 could also be used for drug release, glutathione (GSH) depletion, the regulation of pH, and T1-weighted magnetic resonance (MR) imaging, consequently achieving multimodal theranostic effects and tumor-specific enhanced combination therapy Zhu H. et al., 2018;Yang G. et al., 2018;Zhang et al., 2019;Pu et al., 2020). For example, Yang et al. (2017) designed an intelligent theranostic platform based on hollow mesoporous MnO 2 (HMnO 2 ) nanoshells for tumor-targeted drug delivery, pH-triggered controllable release, and TME-responsive generation of O 2 to alleviate tumor hypoxia. Ce6 and DOX were coloaded into HMnO 2 to achieve combined chemo-photodynamic therapy (Figure 2A). Fluorescence signal of Ce6 and T1-weighted MR signals of the released Mn 2+ were applied to track the nanoparticles after the injection. Despite great progresses and promising results, the rapid consumption of MnO 2 during the reaction in TME may restrict its extensive application to a certain extent .
Differentiated from the aforementioned self-sacrificing MnO 2 , ferrite materials with catalase-like activity and enhanced stability could be served as a superior candidate for continuous O 2 supply. For example, Kim et al. (2017) Yin et al. (2019) reported that MnFe 2 O 4 @MOFs core-shell nanostructure exhibited dual catalytic ability in continuously triggering the decomposition of H 2 O 2 to release O 2 and persistently depleting endogenous GSH, resulting in improved PDT. Also, MnFe 2 O 4 nanoparticles were not consumed during the reaction.  developed CuFe 2 O 4 nanospheres that integrated PDT, PTT, photoenhanced CDT, and MR imaging functions along with TME-modulating capacity. The CuFe 2 O 4 nanospheres regulated the TME through the decomposition of H 2 O 2 to O 2 and the depletion of GSH, which relieved the tumor hypoxia and antioxidant capability, thus further improving the photoenhanced CDT and PDT efficiency ( Figure 2B).
Various Fe-doped nanoplatforms have been reported to catalyze the conversion of endogenous H 2 O 2 to O 2 and thus could enhance the therapeutic effects against hypoxic tumor, including Fe-doped polydiaminopyridine nanofusiforms (Fe-PDAP) (Bai et al., 2018), Fe III doped C 3 N 4 nanosheets , and Fe 3+ -driven assembly of fluorenylmethyloxycarbonyl (Fmoc) protected amino acids (Fmoc-Cys/Fe) . Lan et al. (2018) developed a nanoscale MOF (Fe-TBP, constructed from Fe 3 O clusters and 5,10,15,20-tetra(p-benzoato)porphyrin (TPB)) as a nanophotosensitizer to overcome tumor hypoxia for PDT-Frontiers in Chemistry | www.frontiersin.org April 2021 | Volume 9 | Article 649158 primed cancer immunotherapy. Intracellular H 2 O 2 could be decomposed by the Fe 3 O clusters to generate O 2 through a Fenton-like reaction, whereas the produced O 2 was converted to cytotoxic singlet oxygen ( 1 O 2 ) by photoexcited porphyrins. Prussian blue (PB), a clinical medicine approved by U.S. FDA for the treatment of radioactive exposure, has been proven with catalase-like activity (Cai et al., 2016;Zhou et al., 2018). Yang Z. L. et al. (2018) fabricated a PB-based integrated nanoplatform to elevate O 2 and ROS for highly efficient PDT. Other noble metals or metal oxide-based nanozymes with catalase-like activity have also been applied to overcome tumor hypoxia via H 2 O 2 -activated catalytic O 2 generation, thereby augmenting effect of ROS-generated cancer therapy, such as CeO 2 (Dong et al., 2020), RuO 2 (Huang et al., 2020;Xu et al., 2020), V 2 O 5 , mesoporous manganese cobalt oxide derived from MOFs , Pd@Pt nanoplates , gold nanoclusters (Liu, C. P., et al., 2017), MOF-Au nanohybrid (He et al., 2019), Pt nanoparticles decorated on MOFs , Pt-based core-shell nanoplatform (Wang X. S. et al., 2018), two-dimensional Pd@Au bimetallic core-shell nanostructure , etc. By taking the advantage of dual enzyme-mimic catalytic activity of ultrasmall CeO 2 , Dong et al. (2020) fabricated a nanocomposite with hyperthermiaenhanced peroxidase-like activity, catalase-mimic activity, and GSH depletion for efficient tumor therapy in the NIR-II window. Huang et al. (2020) reported that a multifunctional artificial metalloprotein nanoanalogue, RuO 2 -hybridized ovalbumin (OVA) nanoanalogues, not only exhibited photothermal/ photodynamic effect under NIR light irradiation but also effectively alleviated tumor hypoxia via catalysis of intracellular H 2 O 2 to produce O 2 , thereby concurrently enhancing PDT and reversing the immunosuppressive TME.  reported a two-dimensional Pd@Au core-shell nanostructure (TPAN) that could continuously catalyze endogenous H 2 O 2 to generate O 2 for relieving tumor hypoxia to overcome hypoxiainduced RT resistance. Moreover, the catalytic activity of TPAN toward H 2 O 2 could be enhanced via the surface plasmon resonance effect triggered by NIR-II laser irradiation ( Figure 2C). Wei et al. (2018) reported that Pd@Pt-PEG-Ce6 nanocomposite could not only deliver photosensitizers to tumor sites but also trigger the decomposition of endogenous H 2 O 2 to produce O 2 for a long period of time. Moreover, the moderate photothermal effect of Pd@Pt-PEG-Ce6 under 808 nm laser irradiation accelerated its catalytic decomposition of H 2 O 2 to O 2 . Liu C. P. et al. (2017) reported that the amine-terminated, PAMAM dendrimer-encapsulated gold nanoclusters (AuNCs-NH 2 ) can produce O 2 to improve PDT via the catalase-like activity. Importantly, AuNCs-NH 2 exhibited the catalase-like activity over a broad pH range (pH 4.8-7.4).

Generating O 2 by Water-Splitting Photocatalysts
Compared to the limited intracellular concentration of H 2 O 2 , H 2 O is the most abundant compound in living organisms. Consequently, using H 2 O as an alternative O 2 -generating reactant, the water-splitting strategy could provide unlimited raw materials for in vivo O 2 release. As a typical paradigm, Zheng et al. (2016) reported the use of carbon-dot-decorated C 3 N 4 nanocomposite as a water-splitting catalyst to produce O 2 to overcome tumor hypoxia and improve the PDT effect. The carbon dots were doped to decrease the band gap of C 3 N 4 , and a 630 nm laser was applied as the trigger to induce the water splitting. Chen et al. (2020) reported that in situ photocatalysis of TiO porphyrin encapsulated in folate liposome could not only conquer tumor hypoxia but also generate sufficient ROS to suppress the tumor growth. Analogous to the aforementioned photocatalysts, the photosensitizer nanoparticle-loaded photosynthetic bacteria were developed for tumor-targeted photosensitizer (indocyanine green, ICG) delivery and in situ photocatalyzed O 2 generation. This biomimetic system combined the photosynthetic capability of Synechococcus 7942 (a natural photosynthetic cyanobacterium) and the theranostic effect of ICG-encapsulated human serum albumin nanoparticles . Since hypoxic tumors are usually located in the deep tissues, the penetration depth of the laser is a limitation.

CONCLUSION AND CHALLENGES
We herein present an overview of current strategies to overcome the tumor hypoxia in ROS-generated cancer therapy. Despite great progresses and promising results, most attempts still remain at early stages of development. These strategies suffer from some disadvantages, for example, side effects after intravenous injection, H 2 O 2 dependence in H 2 O 2 -mediated O 2 production, rapid consumption or easy inactivation/instability of natural enzyme and nanozymes, and poor light penetration in photoactivated O 2 production. Moreover, to achieve enhanced therapeutic efficacy, integration of multiple therapeutic/ diagnostic capability and oxygen-supply ability into one nanosystem has become the most commonly used strategy to treat hypoxic tumors. Consequently, complicated and tedious preparation procedures are usually needed. To maximize their capabilities and minimize the side effects, toxicity and immunogenicity of all the involved components should be comprehensively evaluated before clinical trials. In addition, the degradability of the materials should be guaranteed, which will enable the body to clear them after performing the designated pharmacological functions.