Fe‐containing metal–organic framework with D‐penicillamine for cancer‐specific hydrogen peroxide generation and enhanced chemodynamic therapy

Abstract Chemodynamic therapy (CDT) is based on the production of cytotoxic reactive oxygen species, such as hydroxyl radicals (•OH). Thus, CDT can be advantageous when it is cancer‐specific, in terms of efficacy and safety. Therefore, we propose NH2‐MIL‐101(Fe), a Fe‐containing metal–organic framework (MOF), as a carrier of Cu (copper)‐chelating agent, d‐penicillamine (d‐pen; i.e., the NH2‐MIL‐101(Fe)/d‐pen), as well as a catalyst with Fe‐metal clusters for Fenton reaction. NH2‐MIL‐101(Fe)/d‐pen in the form of nanoparticles was efficiently taken into cancer cells and released d‐pen in a sustained manner. The released d‐pen chelated Cu that is highly expressed in cancer environments and this produces extra H2O2, which is then decomposed by Fe in NH2‐MIL‐101(Fe) to generate •OH. Therefore, the cytotoxicity of NH2‐MIL‐101(Fe)/d‐pen was observed in cancer cells, not in normal cells. We also suggest a formulation of NH2‐MIL‐101(Fe)/d‐pen combined with NH2‐MIL‐101(Fe) loaded with the chemotherapeutic drug, irinotecan (CPT‐11; NH2‐MIL‐101(Fe)/CPT‐11). When intratumorally injected into tumor‐bearing mice in vivo, this combined formulation exhibited the most prominent anticancer effects among all tested formulations, owing to the synergistic effect of CDT and chemotherapy.


| INTRODUCTION
Reactive oxygen species (ROS) are small, highly reactive molecules formed due to incomplete oxygen reduction. 1 Recently, numerous studies have indicated that high levels of ROS can cause oxidative stress, thus damaging the cellular components of lipids, enzymes, proteins, and DNAs/RNAs, and inducing cell apoptosis. 2,3 Thus, this is being adopted as a potential approach in cancer therapy, as excessive ROS can increase oxidative damages, thus destroying cancer cells and inhibiting their proliferation. 4 Therefore, a ROS-mediated therapy called chemodynamic therapy (CDT) is being widely studied as a potent cancer treatment. 5,6 CDT is generally based on Fenton and Fenton-like reactions, in which highly cytotoxic hydroxyl radicals are generated from hydrogen peroxide in the presence of Fe ions as catalysts. 7,8 Due to their strong reactivity with biomolecules, hydroxyl radicals are known to be more damaging to cancer cells than any other type of ROS. 9 Therefore, there has been growing interest in the development of therapeutic formulations that can produce high levels of hydroxyl radicals via the Fenton reaction. Previous studies delivered Fe-containing particles, such as Fe 3 O 4 , α-Fe 2 O 3 , MnFe 2 O 4 , FePt, and FeO x -mesoporous silica nanoparticles (MSNs), to cancer environments, as key catalysts, which reacted with endogenous H 2 O 2 in cancer cells to generate hydroxyl radicals. 7,10 However, these particles were not stable in aqueous media, and their reaction was not efficient, most of which occurred only on the particle surfaces. [11][12][13] Moreover, even in the presence of Fe ions, endogenous H 2 O 2 , albeit higher in cancer environments, is not yet sufficient to generate therapeutic levels of ROS that effectively kill cancer cells, resulting in low treatment efficacy. 14,15 A variety of strategies have been proposed to provide more H 2 O 2 to cancer cells. For example, a bolus H 2 O 2 solution was directly injected into the target site; however, it was not retained but rapidly cleared. 16,17 As an alternative approach, biocatalysts were administered to catalyze the oxidation of endogenous molecules to induce exogenous H 2 O 2 . For example, glucose oxidase (GOx) was utilized to elevate H 2 O 2 concentration through a glucose metabolic reaction. 18 In another study, NADPH oxidase (NOX) and superoxide dismutase (SOD) were utilized together.
NOX catalyzed the conversion of endogenous oxygen into O 2 •À , which in turn is converted into H 2 O 2 via an SOD-catalyzed reaction. 19 Although showing the improved efficacy of CDT with extra H 2 O 2 production, the enzymes of natural proteins still pose challenges, such as immunogenicity, low stability, and high cost. 20 Furthermore, endogenous molecules, such as glucose and oxygen, are ubiquitous; thus, systemic generation of H 2 O 2 may still occur to induce side effects. 21,22 Cancer-specific H 2 O 2 generation followed by the Fenton reaction can be considered an advantageous strategy for safer and more efficient CDT. Therefore, we suggest a Fe-containing metal-organic framework (MOF), that is, the NH 2 -MIL-101(Fe), to be loaded with D-penicillamine (D-pen) for CDT (Scheme 1). In this study, we focused on Cu(II) in cancer cells as a cancer-specific endogenous component because its level is known to be significantly higher (>4 times) in cancer cells than that in normal cells. 23 and it possesses a highly porous structure with a large specific surface area (up to 2300 m 2 ). 26  Diamonsil C18 column (4.6 Â 150 mm, 5 μm pore, Dikma, Lake Forest, CA, USA). A mobile phase composed of acetonitrile and 0.1% formic acid (v/v = 60:40) was pumped at a rate of 0.7 ml/min. A peak was detected at a wavelength of 360 nm. 34 To measure the loading amount of calcein, the collected supernatant containing calcein was assayed spectrophotometrically at 495 nm using a UV-Vis spectrophotometer (UV-1800; Shimadzu). 35

| In vitro release study
To examine the release profile of D-pen, CPT-11, or calcein, 5 mg of

| In vitro study on • OH generation
To assess the capacity of • OH generation, the formulations were each tested in a medium of pH 5.5 PBS containing 5 μM cupric sulfate, that is, a slightly acidic and Cu-containing medium, mimicking the cancer environment. 36   BALB/c nude mice were housed in a pathogen-free facility with controlled environments: temperature, 21 ± 1 C; humidity, 55% ± 1%; light/dark cycle, 12 h/12 h.
In this study, we utilized 7-week-old tumor-bearing BALB/c nude mice to examine the in vivo anticancer efficacy. 44  confirming that the microporous structures were retained before and after compound loading. As shown in Table 1, the surface area and total pore volume of NH 2 -MIL-101(Fe) decreased after D-pen or CPT-11 loading because of the molecules being encapsulated in the pores of NH 2 -MIL-101(Fe). These results were consistent with those reported previously. 46 The loading amounts of D-pen and CPT-11 were 101 ± 3 and 254 ± 5 μg/mg, respectively.
The SEM and TEM images in Figure 1d,f show that most of synthesized NH 2 -MIL-101(Fe) (>90%) is a nano-sized particle with an octahedral shape, as previously reported, 46 due to its lower water solubility than that of D-pen. 57,58 In this study, we hypothesized that H 2 O 2 could be generated   Figure S4). 61 The D-pen is biocompatible, 30 and it did not produce   and Figure S4). 23,68 The anticancer efficacy was further improved when the formu-  (Figure 4c), which could be further supported by cancer-specific • OH generation with our current strategy (Figure 3b and Figure S4). After the completion of the reaction and drug release, almost all NH 2 -MIL-101(Fe) was expected to degrade into soluble small molecules of biocompatible ligands and metal ions ( Figure S5). 69 For this reason, the regimen of the current strategy could be repeated for a prolonged period in case a further treatment was needed. With the designated sizes of the NH 2 -MIL-101

| In vitro cell tests
(Fe) herein, 70 the formulation could be injected intravenously with a possible targeted delivery of therapeutic compounds. 71,72 In previous studies, ZIF-8 and MIL-100 MOFs were employed for CDT purposes, where the generation of exogenous H 2 O 2 was based on glucose oxidase (GOx). Thus, ZIF-8 was utilized as a carrier for the delivery of GOx and to produce • OH, which needed to be loaded together with catalytic compounds for the Fenton reaction. [73][74][75][76] However, ZIF-8 was degraded rapidly in hours in acidic cancer environments, 77 and thus, the bioavailability of the encapsulated compounds was low, leading to a limited production of • OH. 78 MIL-100 also worked as a carrier for the delivery of GOx, which, on the other hand, could also serve as a catalyst for the Fenton reaction by itself owing to the constituent metal clusters of Fe. 79 However, the use of GOx is limited by its immunogenicity, low stability, and high cost. 80 Furthermore, the substrate of GOx, for example, glucose, is ubiquitous in the body; thus, the Fenton reaction is not cancer specific, making the formulation susceptible to systemic side effects.

CONFLICT OF INTEREST
Han Bi Ji and Young Bin Choy are listed as inventors on the pending patents (KR 10-2022-0159628) filed by SNU R&DB for the formulation to reactive oxygen species described in this article.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.