key: cord-0021437-oeq1t0l7
authors: Chen, Yu-Chieh; Chen, Yi-Hong; Chiu, Han; Ko, Yi-Hsuan; Wang, Ruei-Ting; Wang, Wei-Ping; Chuang, Yung-Jen; Huang, Chieh-Cheng; Lu, Tsai-Te
title: Cell-Penetrating Delivery of Nitric Oxide by Biocompatible Dinitrosyl Iron Complex and Its Dermato-Physiological Implications
date: 2021-09-18
journal: Int J Mol Sci
DOI: 10.3390/ijms221810101
sha: 9d30310651b2668b754b207ea660337e4dce74e1
doc_id: 21437
cord_uid: oeq1t0l7

After the discovery of endogenous dinitrosyl iron complexes (DNICs) as a potential biological equivalent of nitric oxide (NO), bioinorganic engineering of [Fe(NO)(2)] unit has emerged to develop biomimetic DNICs [(NO)(2)Fe(L)(2)] as a chemical biology tool for controlled delivery of NO. For example, water-soluble DNIC [Fe(2)(μ-SCH(2)CH(2)OH)(2)(NO)(4)] (DNIC-1) was explored for oral delivery of NO to the brain and for the activation of hippocampal neurogenesis. However, the kinetics and mechanism for cellular uptake and intracellular release of NO, as well as the biocompatibility of synthetic DNICs, remain elusive. Prompted by the potential application of NO to dermato-physiological regulations, in this study, cellular uptake and intracellular delivery of DNIC [Fe(2)(μ-SCH(2)CH(2)COOH)(2)(NO)(4)] (DNIC-2) and its regulatory effect/biocompatibility toward epidermal cells were investigated. Upon the treatment of DNIC-2 to human fibroblast cells, cellular uptake of DNIC-2 followed by transformation into protein-bound DNICs occur to trigger the intracellular release of NO with a half-life of 1.8 ± 0.2 h. As opposed to the burst release of extracellular NO from diethylamine NONOate (DEANO), the cell-penetrating nature of DNIC-2 rationalizes its overwhelming efficacy for intracellular delivery of NO. Moreover, NO-delivery DNIC-2 can regulate cell proliferation, accelerate wound healing, and enhance the deposition of collagen in human fibroblast cells. Based on the in vitro and in vivo biocompatibility evaluation, biocompatible DNIC-2 holds the potential to be a novel active ingredient for skincare products.

Based on the distinctive electron paramagnetic resonance (EPR) signal at g ≈ 2.03 observed in the baker's yeast cells (S. cerevisiae), biosynthesis of dinitrosyl iron complexes (DNICs) was first discovered by Prof. Vanin in 1964 [1] [2] [3] [4] . Later, continued investigations explored tetrahedral DNICs [(NO) 2 Fe(L) 2 ] as a natural and ubiquitous cofactor generated upon the interaction of nitric oxide (NO) with non-heme [Fe-S] proteins and cellular labile iron pools [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] . Despite the identification of NO as an endothelium-derived relaxing factor (EDRF) [19, 20] , DNICs in a low-molecular-weight or protein-bound form were proposed as a "working form" of NO under physiological conditions [21, 22] . As a potential biological equivalent of NO, bioinorganic engineering of [Fe(NO) 2 ] unit has emerged to develop biomimetic DNICs as a chemical biology tool for controlled delivery of NO and translation of NO-related functions into biomedical applications [23, 24] . In particular, biological equivalent of NO, bioinorganic engineering of [Fe(NO)2] unit has emerged to develop biomimetic DNICs as a chemical biology tool for controlled delivery of NO and translation of NO-related functions into biomedical applications [23, 24] . In particular, biomimetic DNICs and their conjugates with drug delivery systems were explored for antiaging/anti-inflammatory/anti-viral effect, anti-cancer/anti-hypertensive therapy, diabetic angiogenesis/wound healing, penile erection, and treatment of mild cognitive impairment and neurodegenerative diseases [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] . In particular, phase I and phase II clinical trials of glutathione-bound DNICs [Fe2(μ-glutathione)2(NO)4], a commercial medical product with the pharmacological name Oxacom ® , was reported for application as a clinical medicine against hypertension [30, 31] .

Inspired by the potential of biomimetic DNICs for biomedical applications, synthetic advances in the development of structure-characterized and water-soluble DNICs enabled the study of trafficking and NO-delivery reactivity of DNICs under simulated physiological conditions, in vitro, and in vivo [26, 28, [36] [37] [38] [39] . Water-soluble DNIC [Fe2(μ-SCH2CH2OH)2(NO)4] (DNIC-1) displays an O2-triggered degradation mechanism for the release of nitric oxide, ferric ion, and disulfide with a half-life of 27.4 h and 15.5 h in an aqueous buffer solution (pH = 7.0 or 7.4) at 25 °C and 37 °C, respectively [25, 32, 34] . In simulated gastric fluid (SGFsp, pH 1.2), a shortened half-life of 0.4 h indicates the acidsensitive nature of DNIC-1. To mimic the systemic circulation system, the addition of serum albumin initiated the coordination of Cys-34 toward DNIC-1 leading to the partial formation of albumin-bound DNIC, which features an accelerated release of NO with a half-life of 3.5 h in phosphate-buffered saline (PBS) [34, [39] [40] [41] [42] . Regarding the NO-release reactivity of synthetic DNICs, their capability for the elevation of intracellular NO levels, activation of the NO-soluble guanylyl cyclase (sGC)-cyclic guanosine monophosphate (cGMP) pathway (NO-sGC-cGMP pathway), and regulation of cGMP-dependent cell proliferation was also reported [32, 34, 39, 43] . Moreover, the discovery of multidrug-resistance-associated protein 1 as a transporter for the efflux of glutathione-bound DNICs from the cells may establish a mechanism for transcellular trafficking of DNIC after its uptake by cells [44] [45] [46] . Of importance, the endogenous conjugation of DNIC-1 with gastrointestinal mucin and serum albumin in a reversible manner was reported to facilitate oral delivery of the [Fe(NO)2] unit to the brain across the intestinal epithelial and bloodbrain barriers, respectively [34] . Despite the investigations put forward, kinetics for in vitro NO-release reactivity, the mechanism for cellular uptake and intracellular release of NO, and the biocompatibility of synthetic DNICs remain elusive. Prompted by the potential application of NO to dermato-physiological regulations [47] [48] [49] [50] , herein, the mechanism for intracellular delivery of DNIC [Fe2(μ-SCH2CH2COOH)2(NO)4] (DNIC-2) and its regulatory effect/biocompatibility toward epidermal cells were investigated (Chart 1). Upon the treatment of DNIC-2 to human fibro-Chart 1. Structure of DNICs [Fe 2 (µ-SCH 2 CH 2 OH) 2 (NO) 4 ] (DNIC-1) and [Fe 2 (µ-SCH 2 CH 2 COOH) 2 (NO) 4 ] (DNIC-2). Kinetics and efficacy for the release of NO from DNIC-2 in a cell culturing medium, the minimum essential medium (MEM), were investigated to project on the in vitro study discussed below. Similar to the reported DNIC [Fe 2 (µ-SCH 2 CH 2 OH) 2 (NO) 4 ] (DNIC-1) [25, 34] , DNIC-2 exhibits a steady and aerobic decomposition with a half-life of 4.2 ± 0.6/3.5 ± 0.3 h in MEM with/without the presence of 10% fetal bovine serum (FBS) (Figure 1a-d) . Moreover, this aerobic degradation of DNIC-2 results in a concomitant release of~4 equiv. of NO based on the total nitrite/nitrate assay (Figure 1e ). In addition, the aerobic decomposition of DNIC-2 and release of NO was also validated using an NO-specific fluorescence probe, 4-amino-5-methylamino-2 ,7 -difluorofluorescein diacetate (DAF-FM), whereas the NO donor diethylamine NONOate (DEANO) was used for comparison. Regarding the stoichiometry of released NO (≈4 equiv. for DNIC-2 and 1.5 equiv. for DEANO), reactions of DAF-FM with 50 µM of DNIC-2 and 133 µM of DEANO, respectively, were further performed. In comparison with DAF-FM only, both DNIC-2 and DEANO trigger a time-dependent and significant increase of fluorescence intensity (Figure 1f ). Relevant to the reported DNIC-1 [34] , presumably, O 2 -induced oxidation of the [Fe(µ-SR) 2 Fe] core in DNIC-2 weakens the Fe-to-NO π-backbonding interaction and initiates the release of NO as probed by DAF-FM under aerobic conditions [51] . In the absence of an NO-responsive target (i.e., DAF-FM), the subsequent oxidation of released NO results in the formation of nitrite/nitrate. biocompatibility evaluation using a 2-D culture of fibroblast/melanocyte/keratinocyte cells, 3-D culture of reconstructed human epidermis and human cornea-like epithelium models, and zebrafish embryo, DNIC-2 holds the potential for further development as a novel active ingredient for skincare products. Kinetics and efficacy for the release of NO from DNIC-2 in a cell culturing medium, the minimum essential medium (MEM), were investigated to project on the in vitro study discussed below. Similar to the reported DNIC [Fe2(μ-SCH2CH2OH)2(NO)4] (DNIC-1) [25, 34] , DNIC-2 exhibits a steady and aerobic decomposition with a half-life of 4.2 ± 0.6/3.5 ± 0.3 h in MEM with/without the presence of 10% fetal bovine serum (FBS) (Figure 1a-d) . Moreover, this aerobic degradation of DNIC-2 results in a concomitant release of ~4 equiv. of NO based on the total nitrite/nitrate assay (Figure 1e ). In addition, the aerobic decomposition of DNIC-2 and release of NO was also validated using an NO-specific fluorescence probe, 4-amino-5-methylamino-2′,7′-difluorofluorescein diacetate (DAF-FM), whereas the NO donor diethylamine NONOate (DEANO) was used for comparison. Regarding the stoichiometry of released NO (≈4 equiv. for DNIC-2 and 1.5 equiv. for DEANO), reactions of DAF-FM with 50 μM of DNIC-2 and 133 μM of DEANO, respectively, were further performed. In comparison with DAF-FM only, both DNIC-2 and DEANO trigger a time-dependent and significant increase of fluorescence intensity (Figure 1f ). Relevant to the reported DNIC-1 [34] , presumably, O2-induced oxidation of the [Fe(μ-SR)2Fe] core in DNIC-2 weakens the Fe-to-NO π-backbonding interaction and initiates the release of NO as probed by DAF-FM under aerobic conditions [51] . In the absence of an NO-responsive target (i.e., DAF-FM), the subsequent oxidation of released NO results in the formation of nitrite/nitrate. The data are the mean values ± SD pooled from six independent experiments. (f) Time-dependent change of relative fluorescence intensity of NO-specific probe DAF-FM (excitation at 495 nm and emission at 535 nm) with the treatment of 50 µM of DNIC-2 (black) or 133 µM of NONOate (red) in MEM with 10% FBS, which were fitted to pseudo-first-order kinetics. Data show the mean ± SD from three independent experiments. (g) EPR spectra for DNIC-2 in MEM with (black) and without (blue) the presence of 10% FBS, respectively.

With regard to the reported interconversion between low-molecular-weight and proteinbound DNICs [7, 34, 52, 53] , the formation of protein-bound DNIC derived from DNIC-2 in MEM with/without the presence of 10% FBS was further explored. As opposed to the EPR silence displayed by DNIC-2 in MEM, the addition of 10% FBS results in the assembly of protein-bound DNIC based on the formation of a rhombic EPR signal at g = 2.043, 2.036, and 2.015 ( Figure 1g ). Using DNIC [PPN][(NO) 2 Fe(S 5 )] (PPN = bis(triphenylphosphine)iminium, S 5 = pentasulfide) as a standard, spin quantitation of this protein-bound DNIC determines a 1.5%-conversion of low-molecular-weight DNIC-2 into the protein-bound form ( Figure S1 ). In comparison with other protein-bound and biomimetic DNICs [(NO) 2 Fe(SR)(L)] − [54] , serum albumin is proposed to facilitate this minor formation of protein-bound DNIC upon incubation of DNIC-2 in MEM with the presence of 10% FBS, in which the concentration of serum albumin is 2.1 mg/mL (Tables S1 and S2). This minor formation of protein-bound DNIC, moreover, rationalizes the comparable half-life and efficacy for the release of NO from DNIC-2 in MEM with or without the presence of 10% FBS discussed above. Considering 10% FBS as an essential ingredient, the in vitro study described in the following sections was performed in MEM with the presence of 10% FBS. Presumably, cellular uptake of DNIC-2 and its transformation into protein-bound DNICs followed by the intracellular release of NO may rationalize its overwhelming efficacy for the cell-penetrating delivery of NO. Considering that the biological targets responsive for NO-related physiology (i.e., soluble guanylyl cyclase (sGC)) are located in the intracellular compartment, the potential of effective NO-delivery DNIC-2 prompts our in vitro investigations on its dermato-physiological applications.

To evaluate the potential of the developed NO donor for dermatological application, we first investigated the effects of low-dose DNIC-2 on the behaviors of CCD-966SK cells. After a 48-h incubation, a higher cell density was observed in the groups that received the treatment of DNIC-2 than that of the untreated control ( Figure 3a) . Furthermore, the fibroblasts that were treated with 100 pM, 10 nM, and 1 µM DNIC-2 showed 18.1 ± 4.1%, 13.5 ± 4.0%, and 12.2 ± 2.4% increases in cell density, respectively (p < 0.001; Figure 3b ), suggesting that DNIC-2 could efficiently promote fibroblast proliferation. Although the effects of NO on modulating skin fibroblast proliferation were reportedly controversial [47] [48] [49] 55, 56] , most studies employed short-term NO donors that evolve NO for only several minutes. In the present study, however, the developed DNIC-2 with a half-life of~4 h for the steady release of NO, thus, features a distinct impact on fibroblast proliferation. 

To evaluate the potential of the developed NO donor for dermatological application, we first investigated the effects of low-dose DNIC-2 on the behaviors of CCD-966SK cells. After a 48-h incubation, a higher cell density was observed in the groups that received the treatment of DNIC-2 than that of the untreated control ( Figure 3a) . Furthermore, the fibroblasts that were treated with 100 pM, 10 nM, and 1 μM DNIC-2 showed 18.1 ± 4.1%, 13.5 ± 4.0%, and 12.2 ± 2.4% increases in cell density, respectively (p < 0.001; Figure 3b ), suggesting that DNIC-2 could efficiently promote fibroblast proliferation. Although the effects of NO on modulating skin fibroblast proliferation were reportedly controversial The potential of exogenously administered NO in accelerating fibroblast migration, a key step involved in skin wound healing [57] , has been well-documented [50, 55] . Herein, an in vitro scratch wound healing assay was conducted to analyze the capacity of DNIC-2 in promoting fibroblast migration. After creating a wound gap on a confluent culture using a sterilized P200 tip, the CCD-966SK cells were treated with the media that contained DNIC-2 before being incubated for 24 h. As revealed in Figure 4a ,b, more cells were observed in the gap area in DNIC-2-treated groups than that of the untreated control (p < 0.05), indicating the capacity of DNIC-2 in promoting fibroblast migration. [47] [48] [49] 55, 56] , most studies employed short-term NO donors that evolve NO for only several minutes. In the present study, however, the developed DNIC-2 with a half-life of ~4 h for the steady release of NO, thus, features a distinct impact on fibroblast proliferation. The potential of exogenously administered NO in accelerating fibroblast migration, a key step involved in skin wound healing [57] , has been well-documented [50, 55] . Herein, an in vitro scratch wound healing assay was conducted to analyze the capacity of DNIC-2 in promoting fibroblast migration. After creating a wound gap on a confluent culture using a sterilized P200 tip, the CCD-966SK cells were treated with the media that contained DNIC-2 before being incubated for 24 h. As revealed in Figure 4a ,b, more cells were observed in the gap area in DNIC-2-treated groups than that of the untreated control (p < 0.05), indicating the capacity of DNIC-2 in promoting fibroblast migration. The potential of exogenously administered NO in accelerating fibroblast migration, a key step involved in skin wound healing [57] , has been well-documented [50, 55] . Herein, an in vitro scratch wound healing assay was conducted to analyze the capacity of DNIC-2 in promoting fibroblast migration. After creating a wound gap on a confluent culture using a sterilized P200 tip, the CCD-966SK cells were treated with the media that contained DNIC-2 before being incubated for 24 h. As revealed in Figure 4a ,b, more cells were observed in the gap area in DNIC-2-treated groups than that of the untreated control (p < 0.05), indicating the capacity of DNIC-2 in promoting fibroblast migration. Another important physiological function of NO is to modulate the collagen deposition behavior of skin fibroblasts [47] [48] [49] . Therefore, NO donors hold great promise for various dermatological applications or to act as cosmeceuticals. Herein, to evaluate its capacity to promote collagen deposition, the administration of DNIC-2 to CCD-966SK cells was performed daily for seven consecutive days. The secreted collagen was detected by staining with Sirius Red that can bind the fibrillar collagens. According to the results in Figure 5a , the fibroblast cultures that received 10 nM or 1 µM DNIC-2 daily exhibited much denser staining than that of the untreated control. The Sirius Red dye was then extracted followed by quantification using a spectrophotometer. The obtained data further confirmed the increased collagen content in DNIC-2-treated groups (p < 0.05; Figure 5b) , a clear indication of DNIC-2 in promoting fibroblast collagen deposition. As NO has been identified as a critical mediator for controlling skin collagen accumulation [58] , it is no surprise that direct administration of DNIC-2, which can release NO for a prolonged duration, can achieve enhanced collagen deposition. Collectively, our results clearly demonstrated that DNIC-2 could efficiently promote the proliferation, migration, and collagen deposition of skin fibroblast, thus being a promising agent for dermatological applications. much denser staining than that of the untreated control. The Sirius Red dye was then extracted followed by quantification using a spectrophotometer. The obtained data further confirmed the increased collagen content in DNIC-2-treated groups (p < 0.05; Figure 5b) , a clear indication of DNIC-2 in promoting fibroblast collagen deposition. As NO has been identified as a critical mediator for controlling skin collagen accumulation [58] , it is no surprise that direct administration of DNIC-2, which can release NO for a prolonged duration, can achieve enhanced collagen deposition. Collectively, our results clearly demonstrated that DNIC-2 could efficiently promote the proliferation, migration, and collagen deposition of skin fibroblast, thus being a promising agent for dermatological applications. The data are presented as mean ± SD. * p < 0.05; *** p < 0.005; ns, not significant.

By upregulating the expression level and activity of the key enzyme tyrosinase, NO has been reported as an essential signaling molecule involved in melanogenesis, a process that results in melanin biosynthesis and skin pigmentation [59, 60] . To assess the effect of DNIC-2 on pigment formation, murine B16-F10 melanoma cells that received α-melanocyte-stimulating hormone (α-MSH) induction were used as a model and treated with the developed NO-donor DNIC-2. Arbutin (the common whitening agent) was used as a control. After a 48-h incubation, the cell lysate and culture medium were collected for analyzing the intracellular and extracellular melanin content, respectively.

As revealed in Figure 6a ,b, the melanin content in B16-F10 cells and the grown medium was increased after stimulating with α-MSH, while arbutin treatment effectively The data are presented as mean ± SD. * p < 0.05; *** p < 0.005; ns, not significant.

By upregulating the expression level and activity of the key enzyme tyrosinase, NO has been reported as an essential signaling molecule involved in melanogenesis, a process that results in melanin biosynthesis and skin pigmentation [59, 60] . To assess the effect of DNIC-2 on pigment formation, murine B16-F10 melanoma cells that received α-melanocyte-stimulating hormone (α-MSH) induction were used as a model and treated with the developed NO-donor DNIC-2. Arbutin (the common whitening agent) was used as a control. After a 48-h incubation, the cell lysate and culture medium were collected for analyzing the intracellular and extracellular melanin content, respectively.

As revealed in Figure 6a ,b, the melanin content in B16-F10 cells and the grown medium was increased after stimulating with α-MSH, while arbutin treatment effectively inhibited the α-MSH-induced melanin synthesis. In all the DNIC-2-treated groups, however, changes in the level of melanin content were insignificant. We next analyzed the tyrosinase activity of the cell lysate. As shown in Figure 6c , cell lysate harvested from the DNIC-2-treated groups exhibited a similar tyrosinase activity compared to that of α-MSH-stimulated cells. The aforementioned data demonstrated that the developed NO-delivery DNIC-2 could neither promote nor inhibit the melanogenesis of B16-F10 cells. (Figure 7a,b) . Based on the OECD TG 439 Skin Irritation Test and OECD TG 492 Eye Irritation Test, absent toxicity of 50-µM DNIC-2 toward 3-D cell culture of the reconstructed human epidermis (RhE) and reconstructed human cornea-like epithelium models supports DNIC-2 as a biocompatible and effective NO-delivery reagent (Figure 7c,d) . inhibited the α-MSH-induced melanin synthesis. In all the DNIC-2-treated groups, however, changes in the level of melanin content were insignificant. We next analyzed the tyrosinase activity of the cell lysate. As shown in Figure 6c , cell lysate harvested from the DNIC-2-treated groups exhibited a similar tyrosinase activity compared to that of α-MSH-stimulated cells. The aforementioned data demonstrated that the developed NOdelivery DNIC-2 could neither promote nor inhibit the melanogenesis of B16-F10 cells. 

To evaluate the biocompatibility of DNIC-2, zebrafish embryos were exposed to different concentrations of DNIC-2 until 96 h post-fertilization (hpf), whereas a parallel study of FDA-approved sodium nitroprusside (SNP) was also performed (Figure 8a) . The survival rate for the control, DNIC-2-treated, and SNP-treated groups showed no significant difference (Figure 8b ). At 96 hpf, the body length was also comparable among all 

To evaluate the biocompatibility of DNIC-2, zebrafish embryos were exposed to different concentrations of DNIC-2 until 96 h post-fertilization (hpf), whereas a parallel study of FDA-approved sodium nitroprusside (SNP) was also performed (Figure 8a ). The survival rate for the control, DNIC-2-treated, and SNP-treated groups showed no significant difference (Figure 8b ). At 96 hpf, the body length was also comparable among all groups (Figure 8c) . To further investigate the toxicity of DNIC-2 on embryonic development, a general morphology scoring system was adopted to evaluate developmental timing and the occurrence of teratogenicity [61] . As shown in Figure 8d , no delay or hastening of embryonic development was observed in DNIC-2-treated and SNP-treated groups in comparison with the control group at 96 hpf. For teratogenic assessment, the number of larvae with one or more abnormal anatomical phenotypes at 96 hpf was recorded, and the fraction of larvae with malformation was calculated to infer teratogenicity. The analysis revealed teratogenic tendency seemed to increase in a dose-dependent manner in both DNIC-2treated and SNP-treated groups (Figure 8e ). Collectively, these data suggested that the effects of DNIC-2 on zebrafish developmental toxicity were equivalent to that of SNP.

Upon treatment of zebrafish embryos with different concentrations of DNIC-2 (or SNP), the heart rate, the distance between sinus venosus (SV) and bulbus arteriosus (BA), the incidence of pericardial edema, and the edema index were further investigated in order to assess the potential cardiac toxicity of DNIC-2 and SNP. At 96 hpf, no significant difference in heart rate and SV-BA distance was observed among all groups (Figure 9ac) , which suggests the absent toxicity of DNIC-2 and SNP on embryonic heart development. Based on the value of SV-BA distance, the severity of pericardial edema is classified into three categories: normal, mild, and severe groups as shown in Figure 9d -f. As shown in Figure 9g , a treatment of 0.1 μM of DNIC-2 to zebrafish embryos induced significantly less incidence of pericardial edema than 0.1 μM of SNP treatment. Of interest, DNIC-2 features a linear trend of dose-dependent induction of cardiac edema, while an exponential trend of dose-dependent induction of cardiac edema was observed when the zebrafish embryo was treated with SNP ( Figure 9h ). On the other hand, the pericardial edema index indicates similar dose-dependent toxicity of both DNIC-2 and SNP (Figure 9i ). Accordingly, DNIC-2 demonstrates better biocompatibility than SNP regarding the potential zebrafish embryo cardiac toxicity. To further investigate the toxicity of DNIC-2 on embryonic development, a general morphology scoring system was adopted to evaluate developmental timing and the occurrence of teratogenicity [61] . As shown in Figure 8d , no delay or hastening of embryonic development was observed in DNIC-2-treated and SNP-treated groups in comparison with the control group at 96 hpf. For teratogenic assessment, the number of larvae with one or more abnormal anatomical phenotypes at 96 hpf was recorded, and the fraction of larvae with malformation was calculated to infer teratogenicity. The analysis revealed teratogenic tendency seemed to increase in a dose-dependent manner in both DNIC-2-treated and SNP-treated groups (Figure 8e ). Collectively, these data suggested that the effects of DNIC-2 on zebrafish developmental toxicity were equivalent to that of SNP.

Upon treatment of zebrafish embryos with different concentrations of DNIC-2 (or SNP), the heart rate, the distance between sinus venosus (SV) and bulbus arteriosus (BA), the incidence of pericardial edema, and the edema index were further investigated in order to assess the potential cardiac toxicity of DNIC-2 and SNP. At 96 hpf, no significant difference in heart rate and SV-BA distance was observed among all groups (Figure 9a-c) , which suggests the absent toxicity of DNIC-2 and SNP on embryonic heart development. Based on the value of SV-BA distance, the severity of pericardial edema is classified into three categories: normal, mild, and severe groups as shown in Figure 9d -f. As shown in Figure 9g , a treatment of 0.1 µM of DNIC-2 to zebrafish embryos induced significantly less incidence of pericardial edema than 0.1 µM of SNP treatment. Of interest, DNIC-2 features a linear trend of dose-dependent induction of cardiac edema, while an exponential trend of dose-dependent induction of cardiac edema was observed when the zebrafish embryo was treated with SNP ( Figure 9h ). On the other hand, the pericardial edema index indicates similar dose-dependent toxicity of both DNIC-2 and SNP (Figure 9i) . Accordingly, DNIC-2 demonstrates better biocompatibility than SNP regarding the potential zebrafish embryo cardiac toxicity. According to the zebrafish embryotoxicity test (ZET) results described above, pericardial edema induced by NO-delivery DNIC-2 and SNP, presumably, can be ascribed to the excessive nitrite derived from the oxidation of released nitric oxide [62] . Excessive nitrite was reported to trigger abnormal heart development of zebrafish embryos and to According to the zebrafish embryotoxicity test (ZET) results described above, pericardial edema induced by NO-delivery DNIC-2 and SNP, presumably, can be ascribed to the excessive nitrite derived from the oxidation of released nitric oxide [62] . Excessive nitrite was reported to trigger abnormal heart development of zebrafish embryos and to affect the development of the nervous system or muscles [62] . Another potential mechanism for dose-dependent pericardial edema induced by DNIC-2 (or SNP) in zebrafish embryos is excessive NO-sGC-cGMP signaling, which awaits further verification by exploring the histological and molecular changes of DNIC-2-treated embryos with cGMP signaling antagonist [62] .

Complex [Fe 2 (µ-SCH 2 CH 2 COOH) 2 (NO) 4 ] (DNIC-2) was synthesized based on published procedures [36] . UV-vis spectra were recorded on a Perkin-Elmer Lambda 365 spectrometer. EPR measurements were performed at X-band using a Bruker EMXmicro-6/1/S/L spectrometer equipped with a Bruker E4119001 super high sensitivity cavity. X-band EPR spectra were obtained with a microwave power of 20.27-20.54 mW, the frequency at 9. with the sequential treatment of 10 µM of DAF-FM and 133 µM of DEANO, respectively, were also taken using a fluorescence microscope (Olympus, CKX53, Tokyo, Japan).

Biological activities of DNIC-2 toward skin fibroblasts. To evaluate the effects of DNIC-2 on fibroblast viability, CCD-966SK cells grown in a 96-well plate (2500 cells/well) were treated with various dosages of DNIC-2. After incubation for 24 h, cell viability was determined using a CCK-8 assay (IMT Formosa New Materials, Kaohsiung, Taiwan). Alternatively, the grown cells were stained with 0.5 µg/mL DAPI (ThermoFisher Scientific, Waltham, MA, USA) for 20 min, their nuclei were visualized and counted under a fluorescence microscope (Axio Observer 7; Carl Zeiss, Oberkochen, Germany). For each well, six visual fields were selected randomly for nuclei counting. Moreover, live/dead staining was performed using a ViaQuant Viability/Cytotoxicity Kit (GeneCopoeia, Rockville, MD, USA) according to the manufacturer's instruction.

For the scratch wound healing assay, CCD-966SK cells were seeded into a 24-well plate (4 × 10 4 cells/well). After a 24-h culture, a P200 tip was utilized to scratch the cellular monolayer to create a cell-free wound in each well. The culture medium was then substituted with a fresh medium with or without DNIC-2. The migration of cells into the gap region was photographed under a microscope, and the remaining wound area was determined using ImageJ software [63, 64] .

To assess the effects of DNIC-2 on the collagen deposition behavior of skin fibroblasts, CCD-966SK cells were cultured in a 48-well plate (7500 cells/well) for 7 days. The culture medium was replaced with fresh medium with DNIC-2 every day. To evaluate the collagen content, cells were fixed with 4% paraformaldehyde (Sigma-Aldrich, St. Louis, MO, USA) for 15 min and stained with a Sirius Red/Fast Green Collagen Staining Kit (Chondrex, Woodinville, WA, USA) according to the manufacturer's protocol [65] . The stained cells were photographed under a microscope before subjecting to dye extraction. The optical density of the extracted dye at 540 nm was determined using a microplate reader to quantify the collagen content of test samples. To determine the amount of the melanin secreted by the grown cells, the absorbance of the culture medium at 405 nm was measured using a microplate reader. To quantify the intracellular melanin content, the cells were washed with PBS, trypsinized, pelleted by centrifugation, lysed using 1 N NaOH at 70 • C for 1 h, and analyzed using a microplate reader to measure the optical density at 405 nm [66] [67] [68] .

To determine the activity of tyrosinase, the cells prepared according to the aforementioned methods were lysed with RIPA buffer (20 mM Tris-HCl, 1 mM EGTA, 150 mM NaCl, and 1% Triton X-100) [69] at 4 • C for 20 min followed by centrifugation at 14,000 rpm for 15 min. A bicinchoninic acid assay (G-Biosciences, St. Louis, MO, USA) was used to quantify the total protein of the supernatant. The supernatant that contained 40 µg protein (adjusted to 100 µL with 0.1 mM sodium phosphate buffer; pH 6.8) was mixed with 100 µL of freshly prepared L-DOPA (5 mM; Sigma-Aldrich, St. Louis, MO, USA) and incubated at 37 • C for 1 h [66, 70] . Finally, the relative tyrosinase activity of samples was determined by measuring the optical density of the reaction mixture at 475 nm using a microplate reader and normalized to that of the untreated control group.

Cell viability assay of DNIC-2 toward keratinocyte. To evaluate the effects of DNIC-2 on keratinocyte viability, HaCaT cells grown in a 96-well plate (2 × 10 5 cells/well) were treated with various dosages of DNIC-2. After incubation for 24 h, cell viability was determined using an MTT assay (Sigma Aldrich, St. Louis, MO, USA).

In Vitro Skin Irritation Test by Reconstructed human Epidermis Model. The validated EPI-200-SIT protocol and the Reconstructed human Epidermis (RhE) model EpiDerm TM provided by MatTek Life Sciences (Ashland, MA, USA) were used for the in vitro skin irritation test (SIT). EpiDerm TM SIT was performed to fulfill the criteria set forth in OECD TG 439 [71] . The EpiDerm TM skin inserts were transferred to 6-well plates containing 0.9 mL EPI-100-NMM-SIT/Assay Medium (MatTek Life Sciences) per well and pre-incubated in the humidified incubator with 5% CO 2 at 37 • C overnight. Before conducting the SIT test, the assay medium was replaced by the fresh medium and 35 µL of DNIC-2 (50 µM) was applied topically on top of the EpiDerm TM skin surface for 60 min. Then, DPBS was used to rinse the EpiDerm TM skin surface gently several times, and finally, a sterile cotton swab was gently used to remove excess water and avoid touching the tissue. After that, the EpiDerm TM skin insert was placed in an incubator with 5% CO 2 at 37 • C for 24 h. Then, the tissue was transferred to a 6-well plate that had been added with a new proprietary medium and incubated again for 18 ± 2 h. After the EpiDerm TM skin inserts were rinsed with PBS twice, 2 mL of MTT reagent (MatTek Life Sciences) was added to the EpiDerm TM skin inserts, which were further incubated in a 5% CO 2 environment for 3 h before the MTT assay. 2 mL of isopropanol was used to extract formazan. The optical density (OD) of 200 µL of the isopropanol extract was measured at 570 nm using a UV-Vis Spectrophotometer (BioTeK, Winooski, VT, USA). 50 µM of DNIC-2 was considered to be an irritant to the skin if the tissue viability after exposure was ≤50%. In addition, 5% SDS (Sodium dodecyl sulfate, MatTek Life Sciences, Ashland, MA, USA) was used as a positive control, while DPBS (MatTek Life Sciences, Ashland, MA, USA) was used as a negative control. Three replicates of EpiDerm TM skin inserts were used for each experiment.

In Vitro Eye Irritation Test by Reconstructed human Cornea-like Epithelium Model. The validated EPI-200-EIT protocol and the Reconstructed human Cornea-like Epithelium (RhCE) model EpiOcular TM provided by MatTek Life Sciences (Ashland, MA, USA) were used for the in vitro eye irritation test (EIT). EpiOcular TM EIT was performed to fulfill the criteria set forth in OECD TG 492 [72] . The EpiOcular TM cornea inserts were transferred to 6-well plates containing 0.9 mL OCL-200-ASY-Assay Medium (MatTek Life Sciences, Ashland, MA, USA) per well and pre-incubated in the humidified incubator with 5% CO 2 at 37 • C overnight. Before conducting the SIT test, the assay medium was replaced by the fresh medium and 50 µL of DNIC-2 (50 µM) was applied topically on top of the EpiOcular TM cornea surface for 30 min. Then, DPBS was used to rinse the EpiOcular TM cornea surface gently several times, and finally, a sterile cotton swab was used to remove excess water and avoid touching the tissue. After that, the EpiOcular TM cornea tissue was soaked in a 5 mL culture medium for 12 min, then transferred to a 12-well plate that had been added with a new proprietary medium, and incubated again for 2 h. Afterward, 2 mL of MTT reagent (MatTek Life Sciences, Ashland, MA 01721) and 2 mL of the medium were added to the EpiDerm TM skin inserts, which were further incubated in a 5% CO 2 environment for 3 h before the MTT assay. Furthermore, 2 mL of isopropanol was added to the EpiOcular TM cornea inserts and incubated overnight to extract formazan. The optical density (OD) of 200 µL of the isopropanol extract was measured at 570 nm using a UV-Vis Spectrophotometer (BioTeK, Winooski, VT, USA). 50 µM of DNIC-2 was considered to be an irritant to the eyes if the tissue viability after exposure was ≤60%. In this study, 100% Methyl Acetate (MatTek Life Sciences, Ashland, MA, USA) was used as a positive control, while DPBS (MatTek Life Sciences, Ashland, MA, USA) was used as a negative control. Three replicates of EpiOcular TM cornea inserts were used for each experiment.

Zebrafish embryotoxicity test (ZET). All procedures involving animals were approved by the institutional animal experiments committees and performed in compliance with animal welfare laws, guidelines, and policies. (IACUC No. 109022). ZET was carried out following the reported protocol with modifications [61] . Zebrafish embryos were prepared by natural spawning from adult zebrafish and were exposed to DNIC-2 (or SNP) at different concentrations (between 0.1 and 100 µM) in 2 mL of fresh water from 4 hpf to 96 hpf using 24-well culture plates (4 embryos in 2 mL solution/well). The aqueous solution of DNIC-2 (or SNP) was renewed every 24 h and the viability of embryos was recorded at the same time. At 48 hpf, the chorion membrane of zebrafish embryos was removed carefully to ensure same exposure condition among all embryos.

To explore the effect of DNIC-2 (or SNP) on zebrafish embryo development, zebrafish embryos were anesthetized with a drop of 2000-ppm tricaine (Sigma-Aldrich, St. Louis, MO, USA), and then mounted with 2% agarose (Lonza, Basel, Switzerland). Morphological evaluation of zebrafish embryos was performed at 96 hpf using a general morphology scoring system, [61] which comprises the developmental phase up to 96 hpf.

After treatment of different concentrations of DNIC-2 (or SNP), the heart rate, presence of pericardial edema, and the SV-BA distance in zebrafish embryos were examined to assess drug-induced cardiotoxicity. The evaluation protocol of pericardial edema was modified from previous studies [62, 73] . The SV-BA distance normalized to the body length was used to monitor the formation of embryonic heart tubes [74] . The pericardial edema index was defined as the ratio of the semi-diameter of the pericardial cavity to that of the ventricle, which reflects the severity of pericardial edema [62] . Body length, SV-BA distance, and pericardial edema index were measured from a lateral view and processed with Image J software after capturing images using a stereoscopic zoom microscope (SMZ1500; Nikon, Tokyo, Japan) with a microscope camera (DS-Ri2; Nikon, Tokyo, Japan).

A mechanistic and efficacy study of the cell-penetrating delivery of NO by DNIC-2 and its regulatory effect on dermato-physiology has led to the following results. DNIC-2 displays a release of~4 equiv. of NO with a comparable half-life of 4.2 ± 0.6/3.5 ± 0.3 h in MEM with and without the presence of 10% FBS, respectively. Upon treatment of DNIC-2 to CCD-966SK human skin fibroblast cells, cellular uptake of DNIC-2 followed by its intracellular transformation into protein-bound DNICs rationalizes the steady and effective release of NO with an intracellular half-life of 1.8 ± 0.2 h, which overwhelms the NO donor diethylamine NONOate (DEANO). The treatment of skin fibroblasts with low-dose DNIC-2 resulted in enhanced cell proliferation, migration, and collagen deposition, thus holding substantial potential for dermatological applications. On the other hand, the application of DNIC-2 to melanoma cells did not lead to significant alternation in melanogenesis. Based on the biocompatibility evaluation using a 2-D culture of fibroblast/melanocyte/keratinocyte cells, the 3-D culture of reconstructed human epidermis and human cornea-like epithelium models, and zebrafish embryo toxicity assay, DNIC-2 holds the potential for further development as a novel active ingredient for skincare products. Institutional Review Board Statement: The study was conducted according to the guidelines of the animal protection standards, and the animal use and experimental protocols had been approved by the Experimental Animal Care and Use Committee of National Tsing Hua University, Hsinchu, Taiwan.

Free radicals in yeast cells

Free Radicals of a New Type in Yeast Cells

Free radical species with unpaired electron localization on sulfur atom in yeast cells

Changes in electron spin resonance signals of rat liver during chemical carcinogenesis

Factors influencing formation of dinitrosyl complexes of non-heme iron in vitro preparations of mouse liver and yeasts

Factors influencing formation of dinitrosyl complexes of non-heme iron in the organs of animals in vivo

L-cysteine-mediated destabilization of dinitrosyl iron complexes in proteins

Repair of nitric oxide-modified ferredoxin [2Fe-2S] cluster by cysteine desulfurase (IscS)

Characterization of iron dinitrosyl species formed in the reaction of nitric oxide with a biological Rieske center

Nitric Oxide Modulates Endonuclease III Redox Activity by a 800 mV Negative Shift upon

Dinitrosyliron complexes and the mechanism(s) of cellular protein nitrosothiol formation from nitric oxide

) 2 FeS 5 ] 2-: Spectroscopic characterization of species relevant to the nitric oxide modification and repair of

Interaction of nitric oxide with tetrathiolato iron(II) complexes: Relevance to the reaction pathways of iron nitrosyls in sulfur-rich biological coordination environments

Dinitrosyl iron complexes (DNICs): From biomimetic synthesis and spectroscopic characterization toward unveiling the biological and catalytic roles of DNICs

Synthetic modeling chemistry of iron-sulfur clusters in nitric oxide signaling

Dinitrosyl iron complexes with cysteine. Kinetics studies of the formation and reactions of DNICs in aqueous solution

Unusual Synthetic Pathway for an {Fe(NO) 2 } 9 Dinitrosyl Iron Complex (DNIC) and Insight into DNIC Electronic Structure via Nuclear Resonance Vibrational Spectroscopy

From Spontaneous Assembly to Biological Roles

Nitric oxide activates guanylate cyclase and increases guanosine 3 :5 -cyclic monophosphate levels in various tissue preparations

Endothelium-dependent modulation of cGMP levels and intrinsic smooth muscle tone in isolated bovine intrapulmonary artery and vein

Dinitrosyl iron complexes with thiol-containing ligands as a "working form" of endogenous nitric oxide

Dinitrosyl Iron Complexes As a "Working Form" of Nitric Oxide in Living Organisms

Bioinorganic Chemistry of the Natural [Fe(NO) 2 ] Motif: Evolution of a Functional Model for NO-Related Biomedical Application and Revolutionary Development of a Translational Model

Fe in biosynthesis, translocation, and signal transduction of NO: Toward bioinorganic engineering of dinitrosyl iron complexes into NO-delivery scaffolds for tissue engineering

Extension of C. elegans lifespan using the center dot NO-delivery dinitrosyl iron complexes

Coordination-triggered NO release from a dinitrosyl iron complex leads to anti-inflammatory activity

Dinitrosyl iron complexes (DNICs) as inhibitors of the SARS-CoV-2 main protease

Water-Soluble Dinitrosyl Iron Complex (DNIC): A Nitric Oxide Vehicle Triggering Cancer Cell Death via Apoptosis

Delivery of nitric oxide with a nanocarrier promotes tumour vessel normalization and potentiates anti-cancer therapies

Hypotensive effect of Oxacom®containing a dinitrosyl iron complex with glutathione: Animal studies and clinical trials on healthy volunteers

Hemodynamic effects of the synthetic analogue of endogenous nitric oxide (II) donors a dinitrosyl iron complex in hypertensive patients with uncomplicated hypertensive crisis

Activation of Angiogenesis and Wound Healing in Diabetic Mice Using NO-Delivery Dinitrosyl Iron Complexes

Penile erectile activity of dinitrosyl iron complexes with thiol-containing ligands

Endogenous Conjugation of Biomimetic Dinitrosyl Iron Complex with Protein Vehicles for Oral Delivery of Nitric Oxide to Brain and Activation of Hippocampal Neurogenesis

Assessing the cognitive status of Drosophila by the value-based feeding decision

Synthetic methodology for preparation of dinitrosyl iron complexes

trans-Bis(µ-2-hydroxy-ethanethiol-ato-κ 2 S:S)bis-[dinitro-syliron(II)](Fe-Fe)

Molecular and Crystal Structure of a Cationic Dinitrosyl Iron Complex with 1,3-Dimethylthiourea

Toward biocompatible dinitrosyl iron complexes: Sugar-appended thiolates

Effect of albumin on the transformation of dinitrosyl iron complexes with thiourea ligands

Features of the decomposition of cationic nitrosyl iron complexes with N-ethylthiourea and penicillamine ligands in the presence of albumin

Effects of Glutathione and Histidine on NO Release from a Dimeric Dinitrosyl Iron Complex (DNIC)

Toward the Optimization of Dinitrosyl Iron Complexes as Therapeutics for Smooth Muscle Cells

Nitrogen monoxide (no) and glucose: Unexpected links between energy metabolism and nomediated iron mobilization from cells

Nitrogen monoxide (NO)-mediated iron release from cells is linked to NO-induced glutathione efflux via multidrug resistance-associated protein 1

Nitric oxide storage and transport in cells are mediated by glutathione S-transferase P1-1 and multidrug resistance protein 1 via dinitrosyl iron complexes

Nitric oxide: A newly discovered function on wound healing

Enhancement of fibroblast collagen synthesis by nitric oxide

Inducible nitric oxide synthase: From cloning to therapeutic applications

Copper-Based Metal-Organic Framework as a Controllable Nitric Oxide-Releasing Vehicle for Enhanced Diabetic Wound Healing

Insight into the Electronic Structure of Biomimetic Dinitrosyliron Complexes (DNICs): Toward the Syntheses of Amido-Bridging Dinuclear DNICs

Nitrosylation of human glutathione transferase P1-1 with dinitrosyl diglutathionyl iron complex in vitro and in vivo

Fe(µ-S t Bu) 2 Fe(NO) 2 ] -: Relevance to the products of nitrosylation of cytosolic and mitochondrial aconitases, and highpotential iron proteins

S-nitrosation of serum albumin by dinitrosyl-iron complex

Nitric oxide-releasing nanoparticles accelerate wound healing by promoting fibroblast migration and collagen deposition

Biphasic effect of exogenous nitric oxide on proliferation and differentiation in skin derived keratinocytes but not fibroblasts

Wound repair and regeneration

Nitric oxide drives skin repair: Novel functions of an established mediator

Suppression of melanogenesis by a newly synthesized compound, MHY966 via the nitric oxide/protein kinase G signaling pathway in murine skin

Involvement of nitric oxide in UVB-induced pigmentation in guinea pig skin

From cutting edge to guideline: A first step in harmonization of the zebrafish embryotoxicity test (ZET) by describing the most optimal test conditions and morphology scoring system

Excessive nitrite affects zebrafish valvulogenesis through yielding too much NO signaling

Transplantation of 3D MSC/HUVEC spheroids with neuroprotective and proangiogenic potentials ameliorates ischemic stroke brain injury

Gelatin scaffold with multifunctional curcumin-loaded lipid-PLGA hybrid microparticles for regenerating corneal endothelium

Adipose-derived mesenchymal stromal cells promote corneal wound healing by accelerating the clearance of neutrophils in cornea

Leathesia difformis Extract Inhibits alpha-MSH-Induced Melanogenesis in B16F10 Cells via Down-Regulation of CREB Signaling Pathway

Tropical Ginger Sesquiterpene of Zingiber officinale Roscoe, Attenuates alpha-MSH-Induced Melanogenesis in B16F10 Cells

Antimelanogenic Effects of Polygonum tinctorium Flower Extract from Traditional Jeju Fermentation via Upregulation of Extracellular Signal-Regulated Kinase and Protein Kinase B Activation

Bioactive Decellularized Extracellular Matrix Derived from 3D Stem Cell Spheroids under Macromolecular Crowding Serves as a Scaffold for Tissue Engineering

Anti melanogenic effect of Croton roxburghii and Croton sublyratus leaves in alpha-MSH stimulated B16F10cells

Reconstructed Human Cornea-Like Epithelium (RhCE) Test Method for Identifying Chemicals Not Requiring Classification and Labeling for Eye Irritation or Serious Eye Damage

A dominant negative zebrafish Ahr2 partially protects developing zebrafish from dioxin toxicity

Herbul black henna (hair dye) causes cardiovascular defects in zebrafish (Danio rerio) embryo model

We thank ImDerma Laboratories Co., Ltd. for the support on OECD TG 439 Skin Irritation Test and OECD TG 492 Eye Irritation Test.

The authors declare no conflict of interest.