Oxidized Activated Charcoal Nanozymes: Synthesis, and Optimization for In Vitro and In Vivo Bioactivity for Traumatic Brain Injury

Abstract

Carbon-based superoxide dismutase (SOD) mimetic nanozymes have recently been employed as promising antioxidant nanotherapeutics due to their distinct properties. The structural features responsible for the efficacy of these nanomaterials as antioxidants is, however, poorly understood. Here, we studied the process-structure-property-performance properties of coconut-derived oxidized activated charcoal (cOAC) nano-SOD mimetics by analyzing how modifications to the nanomaterial’s synthesis impact the size, as well as the elemental and electrochemical properties of the particles. We then correlate these properties to the in vitro antioxidant bioactivity of poly(ethylene glycol)-functionalized cOACs (PEG-cOAC). Chemical oxidative treatment methods that afford smaller, more homogeneous cOAC nanoparticles with higher levels of quinone functionalization show enhanced protection against oxidative damage in bEnd.3 murine endothelioma cells. In an in vivo rat model of mild traumatic brain injury (mTBI) and oxidative vascular injury, PEG-cOACs restored cerebral perfusion rapidly to the same extent as our former nanotube-derived PEG-hydrophilic carbon clusters (PEG-HCCs) with a single intravenous injection. These findings provide fa deeper understanding for how carbon nanozyme syntheses can be tailored for improved antioxidant bioactivity, and set the stage for translation for medical applications.

Graphical Abstract

Nanozymes synthesized via the oxidation of coconut-derived activated charcoal afford markable protection against oxidative injury in an in vivo rat model of mild traumatic brain injury. By tuning the oxidation process, smaller, more homogeneous nanomaterials containing free radical quenching quinone active sites are targeted for enhanced bioactivity. Such findings highlight the features that drive the oxidative power of these nanozymes.

1. Introduction

The development of nanoparticles with high surface area and enhanced reactivity after modification has attracted interest across many fields of research due to the distinct properties of these materials, specifically, their ability to catalyze chemical reactions with enzyme-like activity. Such nanomaterials are classified as nanozymes and have been shown to mimic the enzymatic activity of various naturally occurring enzymes (e.g. peroxidase, oxidase, catalase, and superoxide dismutase (SOD)).^[1–2](Nilewski et al. ACS Appl. Mater. Interfaces 2019, 11, 18, 16815–16821; Jalilov et al. ACS Appl. Mater. Interfaces 2016, 8, 24, 15086–15092; Derry et al. Nanoscale, 2019,11, 10791–10807; Wu et al. ACS Nano 2019, 13, 10, 11203–11213; Zhang et al, Advanced Materials. 2022, 2205324; Yang et al. Exploration, 2022, 20210267; Zhang, et al. Nano Today 49 (2023) 101768) The enzyme-like capabilities of SOD-mimetic nanozymes have been explored as a therapeutic tool for mitigating damage associated with the oxidative stress response of a perturbed biological system such as caused by ischemia/reperfusion injury (Bitner et al. ACS Nano 2012, 6, 9, 8007–8014; Mendoza et al. Journal of Neurotrauma 2019 36:13, 2139–2146; Marcano et al. Journal of Neurotrauma 2013 30:9, 789–796; Fabian et al. Front. Neurol. 9:199), and other acute and chronic injuries (Dharmalingam et al. ACS Nano 2020, 14, 3, 2827–2846; Yang et al. Exploration, 2022, 20210267.

These nano-SOD mimetics, which can be carbon,^[3–10] metal,^[11–13] metal oxide,^[14–21] or composite-based,^[22–24] quench superoxide (SO), a potent reactive oxygen species (ROS), generating one molecule each of O₂ and H₂O₂ for every two molecules of SO. By reducing the levels of ROS, the SOD-like nanozymes inhibit oxidative damage to cellular components (e.g. DNA, lipids, and proteins), limiting the cellular dysfunction and death typically imposed by the free radicals. Therefore, these nanozymes are interesting modalities for the treatment of conditions in which high oxidative stress is implicated.^[26–32]

We have previously reported on oxidized carbon-derived SOD-mimetic nanozymes that possess a high degree of sp²-hybridization: poly(ethylene glycolated) hydrophilic carbon clusters (PEG-HCCs),^{[3,6,32–37]} poly(ethylene glycolated) graphene quantum dots (PEG-GQDs),^[4,33] and oxidized activated charcoal (OACs).^[10] These carbon nanoparticles convert SO to O₂ and H₂O₂ with oxidation and reduction rate constants that vary depending on starting material but are within one order of magnitude of native CuZnSOD, two orders of magnitude for MnSOD, and within the same order of magnitude for FeSOD.(Wu et al. ACS Nano 2019, 13, 11203–11213) Moreover, with various oxygen-containing functionalities decorating the sp² carbon core of the particles, the PEG-HCCs, PEG-GQDs, and OACs have the potential to quench free radicals at multiple active sites, resulting in turnover numbers of ~10⁶ ROS per nanoparticle rather than just two for that of native SOD.^{[4,6,10,33,38]}

Analogous to oxidized carbon nanoparticles, but mechanistically distinct, are single-atom nanozymes which contain metal atoms (Cu, Fe, Pt) coordinated to the carbon structure of a carbon particle are a growing area of research with particles that exhibit catalase and SOD-like activity specifically. The nanozyme activity is facilitated by the metal atom much in the way native enzymes such as SOD rely on their metal centers to perform their catalytic action.(Zhang et al, Advanced Materials. 2022, 2205324; Yang et al. Exploration, 2022, 20210267; Muhammed et al. Nano Today 45 (2022) 101530; Zhang, et al. Nano Today 49 (2023) 101768)

Despite these promising advances, challenges remain before the carbon-based nanozymes can be used for clinical purposes as the process-structure-property-performance properties of these materials is not thoroughly mapped. Different nanoparticle properties, including size, morphology, and chemical composition, can strongly alter the activity of these compounds.^[39] It is, therefore, imperative to understand the structural and chemical factors that drive the oxidative power of these carbon nanozymes for quenching ROS, especially under biological conditions.

Herein, we synthesize ROS-quenching carbon nanozymes, OACs, derived from biologically applicable carbon sources, and study the effects of tailoring reaction conditions on the antioxidant bioactivity of the nanoparticles. The OACs are chosen for optimization as they are synthesized from coconut shell-derived activated charcoal (cAC). Activated charcoal (AC) is a desirable starting material for medical applications since it is already used by the medical community as a treatment for acute poisoning,^[40] potentially enhancing its acceptability as a therapeutic. Prior materials developed and studied by our laboratories, PEG-HCCs and PEG-GQDs, on the other hand, are synthesized from carbon nanotubes (CNTs) and coal, respectively.^[4,34] We have shown that these materials respond very differently than functionalized graphene oxide (GO) prepared under typical potassium permanganate (KmnO₄) treatment. This is reflected in the differences between the reduction potentials between the two classes of nanoparticles.^[4,10,37] While we have extensively studied the PEG-HCCs and demonstrated excellent in vivo protective abilities in models of traumatic brain injury (TBI) and stroke,^{[35,36,41,42]} with advantages over existing materials in that they were effective even at delayed time points relevant to their potential clinical utility, these starting materials raise concerns for biological applications as some unmodified CNTs are cytotoxic due to their shape anisotropy and propensity for aggregation. Coal contains heavy metal impurities (e.g. lead, mercury, cadmium, and arsenic) requiring an additional purification step.^[34–36] The coconut shell-derived oxidized activated charcoal (cOAC) nanoparticles were generated by subjecting cAC to harsh oxidative treatment with nitric acid. The cAC is available in a good manufacturing practice (GMP) grade, setting the stage for medical translation. The resulting size, chemical composition, and electrochemical activity were then studied. For analysis of the nanomaterial’s bioactivity, the cOAC was functionalized with poly(ethylene glycol) (PEG-cOAC) for in vitro assays modeling oxidative stress conditions. Our results indicate that longer oxidation durations of the carbon core etch the sp² domains of the cAC precursor, yielding nanoparticles with decreased size and varying distributions of oxygen-containing functionalities that affect the electrochemical profile and SOD-like bioactivity of the PEG-cOAC. We further discuss and demonstrate how these findings can drive new syntheses and processing strategies for improving the technical feasibility of scaling-up the manufacturing of PEG-cOAC nanozymes capable of addressing oxidative stress. Finally, we test PEG-cOAC in an in vivo model of ischemia reperfusion vascular injury in TBI accompanied by hemorrhagic hypotension, a common clinical scenario that worsens outcome due to oxidative injury to the vasculature,^[43] and compare them to PEG-HCCs, finding comparable efficacy.

2. Results and Discussion

2.1. Synthesis of cOAC

Antioxidant OAC nanozymes have been successfully produced from nitric acid treated AC sources, particularly cAC, and are shown to be capable of quenching SO in chemical assays, acting as catalytic SOD-mimetics.^[10] Herein, we expand on this work by extending the scope of the cOAC nanozymes to biological systems. In order to accomplish this, the design of the cOAC nanoparticles must be chemically tailored for optimal antioxidant bioactivity. Important features to consider are: (1) the size of the nanozymes for efficient cellular uptake; (2) the degree of oxidation, with a focus on quinoidal moieties for preferred electrochemical activity and SOD-like properties; and (3) the presence of a highly conjugated carbon core, as the sp²-hybridization of the cOACs facilitates the transfer of electrons by lowering the activation energy for the acquisition of an electron from SO.^[6]

To form a highly oxidized product with oxygen-containing functionalities, GMP grade cAC is subjected to oxidation with fuming nitric acid (Scheme 1). Since cAC is a porous material comprised of graphitic domains interlinked by hydrocarbons, the aromatic sp² pockets can be liberated by the harsh oxidizing conditions. Functionalization of the free edges of these regions with oxygen-containing moieties occurs concurrently.^[10] The size of the sp² carbon core and the distribution of the oxygen functionalities that comprise the nanozyme can be modified by altering the oxidation reaction conditions, such as the ratio of cAC:fuming nitric acid, reaction temperature, and the oxidation duration. Initially, we focused on the effect of oxidation duration by exploring reaction times ranging from 2 h to 6 h since the yield of the cOAC significantly decreased when exposed to the fuming nitric acid for >8 h. Additionally, the 2 h to 6 h range was chosen as a result of previous work we performed on bituminous coal-derived AC (bAC). It was deduced that increasing the oxidation duration to >6 h resulted in reduced antioxidant bioactivity of the PEGylated bituminous oxidized activated charcoal products (PEG-bOAC). Detailed characterization and biological H₂O₂ protection assays for bOAC and PEG-bOAC can be found in the Supplementary Information (Scheme S1, Figure S1–S6, Table S3–S4).

Scheme 1. Synthesis of cOAC and PEG-cOAC from GMP coconut-shell activated charcoal.

2.2. Characterization of cOAC

2.2.1. Morphological Characterization of cOAC.

The shape and size of the cOACs were determined by HR-TEM (Figure 1a–c). Analysis of the HR-TEM images of the cOACs indicate the nanoparticles are disc-like in shape, afforded by the harsh oxidative cutting process, which is in accord with our previous work.^[10] As the oxidation duration is lengthened, the diameter of the cOAC carbon core decreases while the homogeneity of the size distribution sharpens (Figure 1d). Longer reaction durations allow for more free edge etching of the liberated cAC sp² domains, resulting in smaller particles with an average diameter of 2.5 nm for the 6 h-cOAC compared to 7 nm and 13 nm for the 4 h-cOAC and 2 h-cOAC, respectively. During the course of this investigation we found that the PEG-cOACs were difficult to image at high resolution by TEM; therefore, we do not have high quality images to include in the manuscript at this time. The cOACs are small carbon particles, so the crystallinity and contrast are not as good as inorganic salts or heavy metals. When the cOACs are wrapped in PEG, the lattice lines of the cOAC could not be distinguished through the PEG layer. For our purposes, TEM was important for understanding the carbon core size of the cOAC particles. We were effectively able to analyze this via imaging of the non-PEGylated particles.

Figure 1. Characterization of cOAC and PEG-cOACs Synthesized for 2, 4, or 6-h from coconut activated charcoal.

Representative HR-TEM images of (a) 2 h-cOAC, (b) 4 h-cOAC, and (c) 6 h-cOAC. (a-c) Scale bar = 5 nm. The average carbon core diameter of the 2 h-cOAC, 4 h-cOAC, and 6 h-cOAC particles is 13 nm, 7 nm, and 2.5 nm, respectively. (d) Size distributions for the cOAC carbon core diameters are denoted by the histograms (2 h-cOAC, $n=116$ nanoparticles; 4 h-cOAC, $n=131$ nanoparticles; 6 h-cOAC, $n=200$ nanoparticles). E) Deconvoluted XPS high-resolution C 1s peak spectra of (a) 2 h-cOAC, (b) 4 h-cOAC, and (c) 6 h-cOAC compared to PEG-cOACs prepared the same. F) Thermogram of 2 h-, 4 h-, and 6 h-cOACs and PEG-cOACs. G) ATR-FTIR spectra of 2-, 4-, and 6-h cOACs. H) CV of 2 h-, 4 h-, and 6 h-cOAC. Reduction potentials were acquired in PBS (pH 7.4) with a scan rate of 200 mV s⁻¹. I) NTA of the 2 h-, 4 h-, and 6 h-PEG-cOAC.

Results acquired by Raman spectroscopy show the cOAC highly conjugated core is preserved throughout the oxidation process (Figure S7). Both the G and D peaks (~1610 cm⁻¹ and ~1390 cm⁻¹, respectively) commonly appear in the Raman spectra of graphene and amorphous carbon, whereas the shoulder peak in the upper region of the spectra (~3000 cm⁻¹) is reminiscent of amorphous carbon with high aromatic content. The presence of the D peak is also representative of induced defects by the oxidative etching, such as vacancies or interstitial and substitutional atoms in the carbon honeycomb lattice nanostructure.^[44] Furthermore, the cAC displays a sharper and more intense D peak, which is related to defects and disorder in graphitic materials, compared to that of the cOAC nanoparticles. This indicates the graphene content is possibly lower in the starting material and that the cAC is primarily comprised of amorphous carbon domains. Oxidatively etching the cAC removes these amorphous carbon regions, leaving the oxidized sp² domains characteristic of the cOAC.

2.2.2. Compositional Characterization of cOAC.

The cOACs were analyzed by XPS, ATR-FTIR, and TGA to support the successful oxidation of the cAC starting material. Quantification by XPS shows the 2 h-, 4 h-, and 6 h-cOAC nanozymes are primarily composed of carbon and oxygen with consistent C:O ratios around of 2.3 to 2.6 (Table 1). Deconvolution of the high-resolution C 1s XPS spectra yields the approximate distribution of oxygen-functionalities that comprise the cOACs (Figure 1e). The oxidation duration does not have a significant effect on the resulting C–O–C/C–O (ether/hydroxyl) content; however, it is observed that there is a moderate trend between the C=O (carbonyl) and O–C=O (carboxylic acid) content of the cOACs and the reaction duration (Table 1). Longer exposure to the strong oxidant results in a product with a higher C=O:O–C=O ratio. The harsher oxidizing conditions afforded by increasing reaction duration could lead to decarboxylation of the cOAC, forming CO₂, effectively reducing the carboxylic acid content of the nanozyme.

Table 1.

Elemental composition and oxygen-functionality distributions of cOAC approximated from the XPS survey scans and high-resolution C 1s peak spectra.

Sample	C:O	at%
		C=C/C–C	C–O–C/C–O	C=O	O–C=O
2 h-cOAC	2.3	64.7	11.0	5.1	19.2
4 h-cOAC	2.6	68.8	7.4	6.4	17.5
6 h-cOAC	2.6	65.7	11.1	7.1	16.1

The presence of C=O and O–C=O groups is further corroborated by ATR-FTIR (Figure 1g). There is a prominent C=O stretch (~1700 cm⁻¹) and a broad COO–H band (~3700–2000 cm⁻¹), which is representative of the carboxylic acid O–H stretch. The sp² lattice of the cOAC carbon core is noted by the C=C band (~1590 cm⁻¹). Additionally, the cOACs begin to experience losses in weight at ~190–210 °C when heated. This indicates the nanozymes are highly oxidized as oxygen functionalities begin decomposing within this temperature range.^[45] The total weight loss of the cOACs increases with oxidation duration from 54–59% (Figure 1f). It is possible that this technique is insufficient to most completely characterize the C:O content of the material a dried sample. X-Ray absorbance spectroscopy (XAS) is also another method which could allow for better characterization of the macroscopically large sample used for XPS analysis by sampling further into the material, however for our purposes, and the scarcity of synchrotron light sources, XPS is sufficient though may be explored in ongoing work.

2.2.3. Electrochemical Characterization of cOAC.

The 2 h-, 4 h-, and 6 h-cOAC underwent electrochemical analysis by CV to assess their oxidant strength and, therefore, their ability to function as antioxidants capable of quenching SO. From the acquired reduction potentials, it is noted that all cOAC nanozymes are electrochemically active (Figure 1h). This supports the compositional characterization of the cOACs in that the nanoparticles are highly oxidized since graphene alone is electrochemically inert in this potential range. Each nanozyme has a broad reduction potential, ranging up to −2.1 V, but the 2 h-cOAC onset and peak maxima differs from that of the 4 h- and 6 h-cOAC. The 4 h- and 6 h-cOAC onset is found at 0 V, whereas the 2 h-cOAC onset is depressed to −0.2 V. Moreover, the 4 h- and 6 h-cOAC have peak maxima at −1.9 V with shoulder peaks at −0.5 V and −1.7 V. The 2 h-cOAC shows a peak maximum at −1.8 V and a shoulder at −0.5 V. Variations in the reduction potentials are due to differences in the structural features and distribution of the oxygen-functionalities that comprise the nanomaterials. It is well-understood that oxygen-containing moieties possess different reduction potentials with quinone-based functionalities corresponding to reduction potentials higher than that of carboxylic acid- or hydroxyl-type groups.^[37,46] Quinone moieties easily undergo reversible redox chemistry, interconverting between the quinone and the semiquinone radical anion, and are implicated in the mechanisms of action for many classical antioxidants.^[46] The increase in C=O:O–C=O ratio with longer reaction durations indicates the 4 h- and 6 h-cOAC have a higher quantity of C=O functionalities, which could mean the nanoparticles synthesized under harsher oxidizing conditions contain a higher number of quinoidal moieties. The higher quinone content would then result in a more positive reduction potential, as is observed for the 4 h- and 6 h-cOAC. This, however, cannot be definitively deduced since quinones are not easily distinguished from other carbonyl functionalities, besides carboxyl groups, by XPS. Despite the varying reduction potential onsets, the 2 h-, 4 h-, and 6 h-cOAC nanozymes are predicted to have some antioxidant properties. The reduction potential of SO, $E_0\left({\mathrm{O}}_2/{\mathrm{O}}_2^{-\bullet }\right)=-0.35\text{V}$ (vs. Ag/AgCl), overlaps with that of the cOAC nanozymes; therefore, it is a thermodynamically favorable process for the cOAC to oxidize SO to O₂.^[47] Although the oxidized carbon nanomaterials all show reduction potential onsets higher than that of SO, the 4 h- and 6 h-cOAC are expected to be more effective antioxidants with greater bioactivity since their 0 V onsets suggest they are stronger oxidants than the 2 h-cOAC (−0.2 V onset). These results are substantially different than the onsets seen in GO.^[37] This underscores not just the size difference obtained from the oxidation of the cAC, but the inherent functionality differences imparted by the fuming HNO₃ conditions vs the more common KmnO₄ oxidation employed with the synthesis of GO.^[48]

2.2.4. Synthesis and Characterization of PEG-cOAC.

To develop an understanding of how the structural and chemical features of the cOAC correlate to their antioxidant activity in biological systems, the nanozymes were prepared for in vitro analysis by appending PEG addends to the nanoparticle edges. PEG is important for enhancing the stability and the half-life of nanomaterials in biological conditions, especially in vivo.^[49] This was also observed in our former in vivo studies with HCCs and GQDs.^{[34–36,38,42,50]} Since the oxidized carbon nanoparticles are sufficiently carboxylated, carbodiimide coupling methods can be used to covalently attach amine-terminated PEG to the cOAC through amide bond formation (Scheme 1).

The success of PEGylating the nanozymes was determined by ATR-FTIR, XPS, and TGA. ATR-FTIR spectra for the PEG-cOAC (Figure S8) suggest the nanoconjugates are functionalized with the polymer due to the appearance of bands representative of the aliphatic C-H stretches (3000–2700 cm⁻¹) from the PEG ethyleneoxy repeat units. While the spectra primarily mirror that of PEG, there are still notable bands corresponding to the C=O stretch (~1720–1700 cm⁻¹) and C=C stretch (~1610–1590 cm⁻¹) of the cOAC. Confirmation of an amide stretch by FT-IR could not be made likely because it is relatively weak due to there being very few amide bonds compared to the entire rest of the mass thus making measurement difficult without oversaturating the detector. Deconvoluted C 1s XPS spectra further support the addition of PEG to the nanozymes (Figure 1e). There is an increase in the C–O–C/C–O content and a decrease in the O–C=O content of the PEG-cOAC relative to the unmodified carbon core (Table S5). The reduction in carboxylic acids implies the effective conversion of these moieties to amides upon reaction of the cOAC and the methoxy-PEG-amine under the carbodiimide coupling reaction conditions. There is also a distinguishable difference in the TGA of the PEG-cOAC compared to the cOAC. The PEG-cOACs show a significant weight loss starting ~350 °C (Figure 1f). This is the temperature at which PEG begins to decompose.^[4,34] After accounting for the weight loss of the cOAC under comparable heating conditions, it can be calculated that the PEG-cOACs are 80 to 84% PEG by weight of the nanoconjugates. The detailed PEGylation efficiency calculations can be found in the Supplementary Information. In accord with the former HR-TEM analyses of the cOAC carbon cores (Figure 1a–c), NTA shows the hydrodynamic diameter of the PEG-cOAC variants decreases, while becoming more homogeneous, with increasing reaction duration (2 h-PEG-cOAC: mode = 80 nm, mean = 90 nm; 4 h-PEG-cOAC: mode = 78 nm, mean = 83 nm; 6 h-PEG-cOAC: mode = 74 nm, mean = 78 nm) (Figure 1i).

2.2.5. Superoxide Dismutase Activity of PEG-cOAC.

The superoxide dismutase activity of PEG-cOACs was measured using a xanthine/xanthine oxidase superoxide dismutase inhibition assay kit. In this assay, cOAC and PEG-cOAC were added to solutions containing xanthine and xanthine oxidase, a combination which produces superoxide, and the chromogen WST-1 which becomes a colored formazan after being reduced by superoxide. We found that the cOAC and PEG-cOACs inhibited the rate that WST-1 was reduced to formazan only in the presence of the xanthine/xanthine oxidase mixture thus indicating that the cOAC and PEG-cOACs were superoxide dismutase mimetics (Figure 2a). We calculated that the IC₅₀ for the particles were 1.5 ± 0.19 μg mL⁻¹ (6630 U/mg) and 2.64 ± 0.22 μg mL⁻¹ (3750 U/mg) for cOAC and PEG-cOAC respectively and using a paired t-test showed that the two particle types were significantly different (n = 3, p < 0.01). These results indicate that PEGylation reduces modestly the activity of the particles. The non-PEGylated IC₅₀ of these particles is approximately 50% higher than thenon-pegylated oxidized carbon dots reported by Liu et al. (4,000 U/mg) (Liu et al. Adv. Funct. Mater. 2023, 2213856) while the PEGylated version is roughly equivalent.

Figure 2.

Antioxidant and cytoprotective properties of PEG-cOACs. A) Xanthine-Xanthine Oxidase Water Soluble Tetrazolium-1 (WST-1) Superoxide Dismutase (SOD) Inhibitory Assay using PEG-cOACs and cOACs as an SOD mimetic. cOAC and PEG-cOAC both have SOD-like activity with IC₅₀ values of 1.5 ± 0.19 μg mL⁻¹ and 2.64 ± 0.22 μg mL⁻¹ respectively. B) Viability of bEnd.3 cells following treatment with 50 μM Fe(NTA)₃ for 30 min followed by the addition of 6 h-PEG-cOACs and 24 hour incubation. Each result is an average of 8 samples within a single experiment. Statistical analysis was performed a one-way ANOVA with Dunnett’s Multiple Comparisons Test. Measurement significances are: Control: 100 ± 8.46% vs. Fe(NTA)₃: 70.30 ± 16.39%; ***p = 0.002) (Control: 100 ± 8.46%, n = 8 vs. PEG-cOAC + Fe(NTA)₃: 95.49 +/− 10.87%, n = 8, p = 0.69). C) Cell viability following treatment of bEnd.3 cells with the PEG-cOAC (4 mg L⁻¹), H₂O₂ (100 μM), or H₂O₂ (100 μM) + PEG-cOAC (4 mg L⁻¹). The cells were initially exposed to the H₂O₂ for 15 min prior to addition of the PEG-cOAC and then incubation for 24 h. The 2 h-, 4 h- and 6 h-PEG-cOAC nanoparticles are not inherently toxic to the cells when administered alone at a concentration of 4 mg L⁻¹. Number of live bEnd.3 cells after treatment with H₂O₂ and PEG-cOACs with different oxidation time relative to H₂O₂ alone (Mean ± SD). Significance between different oxidation times are: H₂O₂ + 4 h-PEG-cOAC vs. H₂O₂ + 2 h-PEG-cOAC: $p<0.05$; H₂O₂ + 6 h-PEG-cOAC vs. H₂O₂ + 4 h-PEG-cOAC: $p<0.01$; and H₂O₂ + 6 h-PEG-cOAC vs. H₂O₂ + 2 h-PEG-cOAC: $p<0.001$. All oxidation times are significantly different than H₂O₂ alone (H₂O₂ + 2 h-PEG-cOAC vs. H₂O₂: $p<0.05$; H₂O₂ + 4 h-PEG-cOAC vs. H₂O₂: $p<0.01$; and H₂O₂ + 6 h-PEG-cOAC vs. H₂O₂: $p<0.0001$). Each result is an average of 8 samples within a single experiment. Statistical analysis was performed using 2-sample t-tests.

A potential mechanism for reduction in activity by PEGylation is demonstrated by Gao et al. who recently showed that masking the -OH and -COOH groups of oxidized carbon dots with 1,3-propanesulfonate significantly reduced the SOD activity of the particles (Gao et al. Nature Communications, 2023, 14:160). In our present work, we found that PEGylation of the cOACs eliminates the XPS peak associated with -COO⁻ carbon (Figure 1e). This agrees with our finding that PEG-cOACs have a lower SOD activity than cOACs and concurs with existing literature (Gao et al. Nature Communications, 2023, 14:160). Masking quinone residues on PEG-HCCs with ethylenediamine eliminated their electrochemical activity and their cytoprotective capacity against H₂O₂ (Derry et al. Nanoscale, 2019,11, 10791–10807).

Another possible cause in the reduction perhaps due to steric hindrance created by the PEG molecules reducing access to the surface of the particle as a trade-off with improved bioavailability.(Veronese et al. BioDrugs, 2008;22(5):315–29; Hamidi et al. Drug Delivery, 13:6 399–409; Shi et al. Nanoscale, 2021,13, 10748–10764). Another interpretation is that the relatively small difference in IC₅₀ is due to a slight difference in concentration even though the solutions of nanoparticles were normalized by UV-vis spectroscopy $\left({\mathrm{\lambda }}_{700}=5.44\mathrm{m}\mathrm{g}{\mathrm{m}\mathrm{L}}^{-1}{\mathrm{c}\mathrm{m}}^{-1}\right)$ and that there is some influence on the optical properties of the nanoparticles by the PEG that has not been identified. Overall, the PEGylated version is able to efficiently dismutate superoxide compared to other nanozymes and naturally occurring enzymes (Table S7).

In an additional approach, we used stopped-flow kinetic spectroscopy to determine the dismutation rate enhancement by our carbon nanoparticles and compared them to naturally occurring enzymes such as MnSOD and CuZnSOD from previous work (Wu et al. 2019). We found that in this case, the PEG-cOAC and cOAC nanoparticles behaved similarly in rate (Table S7) but the PEG-cOAC particles were slightly faster than the cOAC nanoparticles which differs from the SOD inhibition assay. Regardless, PEG is an important element for in-vivo use as it enhances the circulation time of many drugs and likely also does the same here. Even if the IC50 is slightly higher, the prolonged circulation time may substantially outweigh the small, but statistically significant, difference. Veronese et al. reported on the notable PEG-Bovine Superoxide Dismutase (PEG-SOD) which has a longer circulation time (Intravenously: ${\mathrm{t}}_{1/2}=25$ h vs 6 min, PEG-SOD vs. SOD) (Veronese et al. Journal of Controlled Release. 1989;10(1):145–54.)

2.2.7. Antioxidant Efficacy of PEG-cOAC Against iron(III) trinitrilotriacetate.

Another source of oxidative injury is extracellular iron, we tested the efficacy of PEG-cOAC against Fe(NTA)₃, a potent hydroxyl radical generator. We treated bEnd.3 cells for 30 min with 50 uM Fe(NTA)₃ dissolved in complete media followed by 6-h fuming HNO₃-oxidized PEG-cOACs for 24 h. After treatment, the remaining cells were detached with trypsin and counted. We used a one-way ANOVA analysis with Dunnett’s Multiple Comparison Test and found that Fe(NTA)₃ produced a significant reduction in viability (Control: 100 ± 8.46% vs. Fe(NTA)₃: 70.30 ± 16.39%; ***p = 0.002) while there was not a significant reduction in viability when PEG-cOACs were included (Control: 100 ± 8.46%, n = 8 vs. PEG-cOAC + Fe(NTA)₃: 95.49 ±10.87%, n = 8, p = 0.69). Because the control and the PEG-cOAC + Fe(NTA)₃ samples were not significantly different, it can be interpreted that the particles effectively reduced the toxicity associated with Fe(NTA)₃ in this setting.

2.2.8. Antioxidant Efficacy of PEG-cOAC Against Hydrogen Peroxide In Vitro.

The antioxidant efficacy of the PEG-cOAC variants was evaluated using an in vitro assay modeling oxidative stress. In this assay, bEnd.3 murine endothelioma cells are treated with H₂O₂, which induces oxidative stress through the generation of intracellular radicals (SO and HO^•). The ability of the PEG-cOAC to mitigate oxidative damage, and eventual cellular death, was assessed by administering the carbon nanomaterials (4 mg L⁻¹) 15 min after treatment of the bEnd.3 cells with 100 μM H₂O₂. After incubating the cells for 24 h, the number of living cells were counted. Treatment with the PEG-cOAC showed all nanoparticle variants are protective against 100 μM H₂O₂, even with delayed administration (Figure 2c). Additionally, all PEG-cOAC variants are nontoxic, at a concentration of 4 mg L⁻¹, to bEnd.3 cells after incubation for 24 h (Figure 2c). Treatment with the PEG-cOAC nanomaterials against H₂O₂ presented statistically significant improvements in cell viability compared to cells treated with H₂O₂ alone, albeit the significance is greater for the cOACs synthesized using longer reaction durations (H₂O₂ + 2 h-PEG-cOAC vs. H₂O₂: $p<0.05,p=0.031$; H₂O₂ + 4 h-PEG-cOAC vs. H₂O₂: $p<0.01,p=0.0015$; and H₂O₂ + 6 h-PEG-cOAC vs. H₂O₂: $p<0.0001$). These results are consistent with the hypothesized antioxidant capabilities of the PEG-cOAC based on the size and elemental characterization of the nanomaterials. With increasing reaction duration, the carbonyl to carboxylic acid ratio of the cOAC carbon core becomes more prominent. This suggests the 6 h-cOAC-PEG would have the most active sites for thermodynamically favorable interactions with ROS, which is further corroborated by the higher reduction potential of the material. Additionally, the smaller and more homogeneous size distribution afforded by longer reaction durations is predicted to be preferable for cellular uptake of the nanoparticles, allowing the 6 h-PEG-cOAC to show greater intracellular bioactivity. In vitro assays are currently underway to study the effect of PEG-cOAC size and distribution on cell internalization.

2.3. Improving Technical Feasibility of cOAC Production.

Once a viable synthetic route for producing bioactive cOAC nanoparticles was identified, we optimized the synthesis further for improved technical feasibility (Scheme 2). One limitation of using fuming nitric acid for the synthesis of these nanozymes is the safety concerns, especially in regard to scaled-up manufacturing of the nanodrug in pharmaceutical industry settings. To address this issue, we increased the concentration of the reaction by 10x (from 10 mg mL⁻¹ to 100 mg mL⁻¹ of cAC in fuming nitric acid). We then used a higher reaction temperature, by heating the oil bath to 140 °C, so we could oxidize at the shorter reaction duration of 4 h while still attaining cOAC products with comparable size and degree of oxidation to the 6 h-cOAC. Furthermore, we collected the <20 μm size fraction of cAC by sieving the bulk material prior to oxidation in order to evaluate whether fractionating the starting material results in a more homogenous product. This is appealing from a pharmaceutical and quality control standpoint as uniformity and reproducibility of nano-based drugs is important for acquiring FDA approval. Additionally, to reduce the costs associated with purifying the nanoparticles by bath dialysis after oxidation, we subjected the acidic reaction mixture to a chemical quench using aqueous 2 M glycine. This decreased the bath dialysis purification process from 7 to 1 d as glycine reacts with the nanomaterial to more rapidly remove the nitrate ester functionalities on the nanoparticles.

Scheme 2. Synthesis of cOAC and PEG-cOAC by the oxidation method for improved technical feasibility.

Another important aspect of synthesis is purification of byproducts from the reaction mixture. Here we used exhaustive dialysis into ddH₂O to remove excess PEG and residual reaction components. We performed 8 rounds of dialysis into 4 L of ddH₂O from the nanoparticle synthesis solution which itself is transferred to water by two rounds of centrifugal filtration to reduce the DMF concentration to below 10%. After the 8^th round of centrifugation we found that the amount of PEG in the final product was below the detection limit of our assay kit (<1 ng mL⁻¹) (Figure S10). Alternative methods such as anion exchange chromatography exist, however we considered dialysis to be the better option because it likely the most gentle and least likely to alter the nanoparticles through aggregation or other means.

Characterization of the improved cOAC nanomaterials show they are similar to the 4 h- and 6 h-cOAC. Both the bulk- and sieved-derived optimized cOAC are smaller than the 4 h-cOAC, but larger than the 6 h-cOAC. On average, the bulk-cOAC are 4.2 nm in diameter (Figure 3a–c), whereas the sieved-cOAC are 3.2 nm in diameter with a narrower size distribution, as expected (Figure 3c). The preservation of the cOAC sp²-hybridized carbon core is observed by the graphitic D and G peaks (~1620 cm⁻¹ and ~1390 cm⁻¹, respectively) in the Raman spectra (Figure S9) and the C=C stretch (~1590 cm⁻¹) in the ATR-FTIR spectra (Figure S11a). Additionally, the optimized nanoparticles are highly oxidized. The C:O ratio for the bulk- and sieved-cOAC is 2.1 with the C=O:O–C=O ratio being higher for the sieved-cOAC than the bulk-cOAC, as determined by XPS (Figure 3d, Table 2). There are also evident C=O (~1700 cm⁻¹) and COO-H stretches (~3700–2000 cm⁻¹) in the ATR-FTIR spectra (Figure S11a). Both optimized nanozymes show weight losses of 54% by TGA (Figure S11b). The sufficient functionalization of the bulk- and sieved-cOAC with oxygen-containing moieties yields electrochemically active nanoparticles. Cyclic voltammograms of the bulk- and sieved-cOAC predict the nanozymes can effectively oxidize SO. The reduction potentials of the two optimized nanomaterials are comparable with an onset ~0 V and shoulder peaks at −0.5 V, −0.7 V, −1.2 V, −1.5 V, and −2.0 V (Figure 3e).

Figure 3.

Representative HR-TEM images of (a) bulk-cOAC and (b) sieved-cOAC. (a-b) Scale bar = 5 nm. The average carbon core diameter of the bulk-cOAC and sieved-cOAC particles is 4.2 nm and 3.2 nm, respectively. (c) Representative size distributions for the cOAC carbon core diameters are denoted by the histograms (bulk-cOAC, $n=201$ nanoparticles; sieved-cOAC, $n=360$ nanoparticles). D) Deconvoluted XPS high-resolution C 1s peak spectra of bulk-cOAC, sieved-cOAC, their PEGylated derivatives. E) CV of the optimized cOAC nanoparticles. Reduction potentials were acquired in PBS (pH 7.4) with a scan rate of 200 mV s⁻¹. F) Thermograms of bulk and sieved-PEG-cOACs, and G) NTA of the PEG-cOAC synthesized from the bulk and sieved-cOAC nanoparticles.

Table 2.

Elemental composition and oxygen-functionality distributions of bulk- and sieved-cOAC approximated from the XPS survey scans and high-resolution C 1s peak spectra.

Sample	C:O	at%
		C=C/C-C	C-O-C/C-O	C=O	O-C=O
bulk-cOAC	2.1	64.5	11.8	5.9	17.9
sieved-cOAC	2.1	68.5	9.1	7.6	14.9

For in vitro testing, the bulk- and sieved-cOAC were functionalized with methoxy-PEG-amine using the carbodiimide coupling method (Scheme 2). Again, ATR-FTIR, XPS, and TGA show the oxidized nanoparticles are effectively PEGylated. The observation of the aliphatic C-H stretches (3000–2700 cm⁻¹) in the ATR-FTIR (Figure S12) and the increase in the C–O–C/C–O content determined by XPS (Figure 3d, Table S6) denote the polymer is present in the nanoconjugate. There is also a notable loss in weight ~350–450 °C in the collected thermograms (Figure 3f). The bulk- and sieved-cOAC were calculated to be 91% and 89% PEG by weight of the nanoconjugates, respectively. Analysis by NTA shows the hydrodynamic sizes of the bulk- and sieved-PEG-cOAC are within the range expected (bulk-PEG-cOAC: mode = 74 nm, mean = 75 nm; sieved-PEG-cOAC: mode = 69 nm, mean = 76 nm) with the sieved nanoparticles being more uniform in hydrodynamic diameter (Figure 3g).

Antioxidant Efficacy of the Improved PEG-cOAC In Vitro.

Both the bulk- and sieved-PEG-cOAC are nontoxic to bEnd.3 cells after incubation for 24 h (Figure 4), rendering them viable for in vitro use at a concentration of 4 mg L⁻¹. To determine the antioxidant capabilities of the improved PEG-cOAC variants, the bulk- and sieved-PEG-cOAC (4 mg L⁻¹) were tested in the H₂O₂ challenge assay. Results indicate both the bulk- and sieved-PEG-cOAC nanoparticles are protective against 100 μM H₂O₂ after delayed administration of 15 min (Figure 4). Treatment with either of the improved PEG-cOAC nanomaterials against H₂O₂ presented statistically significant improvements in cell viability compared to H₂O₂ alone (H₂O₂ + bulk-cOAC-PEG vs. H₂O₂: $p<0.01,p=0.0038$; and H₂O₂ + sieved-PEG-cOAC vs. H₂O₂: $p<0.0001$); however, the sieved-PEG-cOAC were found to be more protective than the bulk-PEG-cOAC (H₂O₂ + bulk-PEG-cOAC vs. H₂O₂ + sieved-PEG-cOAC: 84.9 ± 9.1 vs. 98.3 ± 5.3, $p=0.0042$). This suggests the sieved-PEG-cOAC are more effective at mitigating oxidative stress, which aligns with the in vitro results acquired for the 6 h-PEG-cOAC vs. the 2 h- and 4 h-PEG-cOAC. Analogous to the 6 h-PEG-cOAC, the sieved-derived nanozyme shows a higher C=O:O–C=O ratio, predicting a greater number of ROS quenching active sites. Furthermore, we hypothesize the sieved-PEG-cOAC are more effectively taken up by cells owing to the smaller and narrower size distribution of the nanoparticle’s carbon core, however, cellular uptake studies must be completed to validate this suggestion.

Figure 4.

Example of cell viability following treatment of bEnd.3 cells with the PEG-cOAC (4 mg L⁻¹), H₂O₂ (100 μM), or H₂O₂ (100 μM) + PEG-cOAC (4 mg L⁻¹). The cells were initially exposed to the H₂O₂ for 15 min prior to addition of the PEG-cOAC and then incubation for 24 h. The PEG-cOAC nanoparticles are not inherently toxic to the cells when administered alone at a concentration of 4 mg L⁻¹. Treatment with the PEG-cOAC after H₂O₂ exposure is statistically significant compared to H₂O₂ alone ($p<0.01$ for H₂O₂ + bulk-PEG-cOAC vs. H₂O₂ and $p<0.0001$ for H₂O₂ + sieved-PEG-cOAC vs. H₂O₂). Each result is an average of 8 samples within a single experiment. Statistical analysis was performed using 2-sample t-tests.

Cellular Uptake by bEnd.3 Cells of PEG-cOACs

We performed a particle uptake experiment using immunofluorescent labeling with an anti-PEG antibody to label the position of the PEG-cOACs (red), an anti-Mitochondrial Complex I antibody to show the mitochondria which take up a large portion of the cytosolic volume (green), and DAPI to label the nucleus (blue). Fluorescent micrographs taken at 5 and 60 min following administration of the sieved 4-h PEG-cOAC and the bulk 6-h PEG-cOAC and washing off of free particles, show a more robust cellular association (red label) for the sieved 4 hour material, although there is particle association with both (Figure 5a–d).

Figure 5. Uptake of PEG-OACs into cultured b.End 3 cells.

Fluorescent microscopy (40X) utilizing mitochondrial (green), nuclear (blue) and anti-PEG PEG-OAC (red) staining to indicate cellular association following addition of bulk 6-h PEG-cOACs and sieved 4-h PEG-cOACs and washing to eliminate any free PEG-OACs: (a,c) indicates some visible red at 5 min, (b,d) show more widespread red visible at 60 min. By comparison, there appears to be already red cellular association of the sieved 4-h PEG-cOAC at 5 min and robust cell association at 60 min. e) Oblique view of bEnd.3 cell treated with sieved 4-h PEG-cOACs for 5 min. Note there are yellow/orange pixels suggesting co-localization of the PEG-OAC with mitochondria label. F) Cross-section of cell, PEG-cOAC signal is shown throughout the cell, on the cell membrane (b.1), surrounding the nucleus (b.2), and within the cytosolic compartment (b.3), near to the mitochondrial label.

To more clearly demonstrate intracellular uptake, we collected a series of z-stack images using a GE Deltavision wide-field fluorescence microscope equipped with a 100x objective, deconvolved them, and generated three dimensional projections using Imaris Viewer for the 5 minute time point of the sieved 4-h PEG-cOACs (Figure 5e). The nanoparticles were localized throughout the cell without significant discrimination by region. Using an orthogonal section (Figure 5f) Anti-PEG signal was observed on the cell membrane (Figure 5f.1), and especially in the cytosolic compartments and associating with mitochondria (Figure 5f.3; note yellow/orange color indicating co-localization with the green mitochondria antibody stain), and surrounding the nucleus (Fig. 16.2; note red/pink surrounding the blue DAPI nuclear stain).

In Vivo Effectiveness in a Clinically Relevant Model of Traumatic Brain Injury.

Prior work by Berlin et al. (Berlin et al. ACS Nano. 2010;4(8):4621–36.) showed that poly(ethylene glycol)-functionized hydrophilic carbon clusters (PEG-HCCs) distributed throughout the body in rats following intravenous administration with uptake that varied by organ system. The same phenomenon was expected with PEG-cOACs: we performed a preliminary assessment of the biodistribution of PEG-cOACs and found by 24 hours they were in major organs including the brain (Figure S13). We proceeded to then study PEG-cOACs to compare their effectiveness to PEG-HCCs in a mild traumatic brain injury model (Figure 6a) with hypotensive shock and resuscitation. PBS-treatment was a vehicle control. The rats were treated intravenously with 2 mg kg⁻¹ of PEG-cOACs, PEG-HCCs, or an equivalent volume of PBS 80 min after the pneumatic ram-induced traumatic brain injury and reinfusion of blood (Figure 6b). Later, a second injection of the PEG-HCCs was given at 200 min. We found there was no significant difference in perfusion between PBS or the nanoparticles immediately after reinfusion (PBS: 79.6 ± 4.2%, PEG-HCC: 76.3 ± 6.2%, PEG-cOAC: 80.5 ± 6.5%; ns) in the central region (Figure 6c). After the second injection of PEG-HCCs and PBS at 200 min, perfusion remained higher than PBS for both PEG-HCCs and PEG-cOACs as measured at 290 min but PEG-HCCs and PEG-cOACs were not significantly different from each other. After 120 min, perfusion in rats treated with PEG-HCCs and PEG-cOACs remained higher than PBS (PBS: 39.8 ± 2.6%; PEG-HCC: * 78.2 ± 4.9%; PEG-cOAC: * 83.5 ± 1.5%; $p<0.0012$ *; Figure 6d) for the duration, but no significant difference between the PEG-HCC and PEG-cOACs was again observed. Comparable effects were also observed in the penumbra (PBS: 45.7 ± 2.2%; PEG-HCC: * 80.7 ± 1.7%; PEG-cOAC: * 80.2 ± 1.2%; $p<0.00014$ *). No significant difference between PEG-cOACs and PEG-HCCs was observed at the 290 min time point (Figure 6e, f, g). Because only a single dose of PEG-cOACs was necessary to produce an effect that was not significantly different than the PEG-HCC treatment, these results indicate that the PEG-cOACs may be more protective on the basis of mass concentration than PEG-HCCs.

Figure 6.

Traumatic brain injury (TBI) model mimicking hypotension and resuscitation was performed in rats using 2 doses of 0.35 mL PBS (control; $n=10$), two doses of 2 mg kg⁻¹ PEG-HCCs ($n=10$), and one dose of 2 mg kg⁻¹ PEG-OACs ($n=8$). To simulate a drop in blood pressure, blood is withdrawn at $t=0$ min (labeled as hypotension). Blood is reinfused at $t=50$ min (labeled as Pre-Hosp, or pre-hospital), and finally, the nanoparticles or PBS are given at $t=80$ min (labeled as Hospital). a) Injury of the right hemisphere by a pneumatic ram produces two distinct regions: a center surrounded by a penumbra. Perfusion measurements are taken from both sites using laser doppler speckle imaging. b) Perfusion of the injury center following TBI plotted over time in rats treated with 2 mg kg⁻¹ PEG-cOACs or PEG-HCCs or an equivalent volume of PBS at $t=80$- or 200-min. c) As in (b), time dependent perfusion plot of the injury penumbra. d) Average of the perfusion in all rats in each treatment group (PBS, $n=10$; PEG-HCC, $n=10$; PEG-cOAC, $n=8$) between ($t=90-110$ min) following the initial dose ($t=80$), none are significant compared to PBS. e) Average of the final perfusion measurements in all rats in each treatment group (PBS, $n=10$; PEG-HCC, $n=10$; PEG-cOAC, $n=8$) between ($t=260-320$ min). Nanoparticle treatments significantly increase perfusion compared to PBS control (PEG-HCCs, $p=1.4\times {10}^{-3}$,**; PEG-cOACs, $p=1.2\times {10}^{-3}$_, **). Statistical analysis was performed by one-way ANOVA with Dunnett correction. (f) and (g) are identical measurement types to (c) and (d) Statistical analysis was performed by one-way ANOVA with Dunnett correction.

3. Conclusion

In conclusion, SOD-mimetic carbon nanozymes with high degrees of sp²-hybridization and oxygen-containing functionalities have been synthesized from GMP medical-grade cAC. Nanoparticles with enhanced antioxidant capacity and cellular protection against H₂O₂ and excellent in vivo protection in a mTBI model in which oxidative vascular injury causes poor outcome, can be synthesized through the modification of the cOAC oxidation protocol. The effects of such optimization on particle size, elemental composition, and electrochemical activity were analyzed. Specifically, it was noted that increasing the reaction duration renders smaller and more homogeneous nanoparticles, whereas the oxidant strength of the nanozymes increases with longer exposure times to the nitric acid. After functionalization of the cOAC variants with PEG for improved stability in biological conditions, it was demonstrated that all PEG-cOAC nanoparticles are protective against H₂O₂; however, nanozymes with narrower size distributions and greater carbonyl to carboxylic acid ratios afforded the highest level of protection. Based on an understanding of the nanozyme characteristics that relate to better antioxidant bioactivity, the synthesis of the cOAC was then further optimized for improved technical feasibility. Importantly in terms of potential clinical translation, oxidized nanoparticles derived from a medicinal grade source, cAC, was comparable to our highly effective nanotube derived PEG-HCCs. While PEG-OACs will need to go through a full safety and effectiveness pre-clinical assessment prior to clinical testing, it is encouraging that a source material familiar to the medical community was successfully synthesized to an active in vivo material. Elucidating a method for safely synthesizing the carbon nanozymes under harsh oxidizing conditions increases the practicality for scaled-up manufacturing of such nanomaterials, ultimately facilitating the potential translation of the antioxidant nanotherapeutic to industry for pharmaceutical applications.

4. Materials and Methods

4.1. Materials

Powdered medical grade activated charcoal was purchased from EnviroSupply & Service. Fuming HNO₃ (90%, ACS grade) was purchased from Alfa Aesar. Iron(III) nitrate nonahydrate was purchased from Fisher Scientific. N,N-dimethylformamide (DMF, ≥99.9%, HPLC grade), nitrilotriacetic acid disodium salt, and phosphate buffered saline (PBS, pH 7.4) were purchased from MilliporeSigma. Methoxy-poly(ethylene glycol)-amine, with an average molecular weight of 5,000 Da, was purchased from Laysan Bio, Inc. N,N’-diisopropylcarbodiimide (DIC, >98.0%) was purchased from Tokyo Chemical Industry (TCI). Cell Culture: bEnd.3 murine endothelioma cell lines (CRL-2299) were purchased from American Type Culture Collection (ATCC, Manassas, VA). bEnd.3 cells were cultured in Dulbecco’s Modified Eagle’s Media (DMEM) with high glucose (Gibco), 10% FBS (Atlanta Biologicals), and 1% (10,000 U/mL) penicillin-streptomycin (Lonza). Biochemical Assays: A PEGylated protein ELISA kit (ADI-900-213-0001) was purchased from Enzo Life Sciences. A SOD Inhibitory Assay Activity kit (S311) was purchased from Dojindo Molecular Technologies. Rabbit recombinant Anti-poly(ethylene glycol) antibody (PEG-B-47), and anti-Mouse monoclonal Anti-Complex I antibody (ab109798) was purchased from Abcam. Goat Anti-mouse IgG (H+L) AlexaFluor488, and Goat Anti-rabbit IgG AlexaFluor 594 were purchased from ThermoFisher. 4’-6-diaminidino-2-phenylindole (DAPI) was purchased from Sigma-Aldrich.

4.2. Oxidation of cAC

The synthesis of cOAC by nitric acid is similar to that which was previously reported by our group.^[10] Powdered medical grade cAC (0.500 g)was added to a 250 mL round-bottom flask, followed by the addition of fuming (90%) HNO₃ (50 mL, 1.08 mol). The reaction mixture was placed in an oil bath pre-heated to 100 °C and was allowed to stir under reflux for the specified reaction duration (2 h, 4 h, or 6 h). After heating, the reaction mixture was removed from the heat source and allowed to cool to room temperature. The reaction was quenched by pouring the acidic solution over 70 mL of deionized (DI) ice in a 2 L beaker. The quenched solution was left to reach room temperature before pouring into one Spectra/Por 7 dialysis membrane (regenerated cellulose, 1 kDa MWCO, 40 cm × 45 mm dimensions). The cOAC solution was allowed to purify by bath dialysis, with continuous flow of DI endotoxin free water (Milli-Q; Millipore), for a total of 7 d. The purified reaction mixture was then removed from the dialysis bath and filtered through a 0.22 μm poly(ethersulfone) (PES) membrane. UV-vis spectrophotometry was used to determine the concentration of the aqueous cOAC product. Mass extinction coefficients of the cOAC variants at a wavelength of 700 nm are denoted in Table S1.

4.3. Functionalization of cOAC with PEG

Aqueous cOAC was dried by lyophilization (Labconco Freeze Dry System/Freezone 4.5). The lyophilized 2 h-, 4 h-, or 6 h-cOAC (17.0 mg) was suspended in DMF (17 mL) and cup horn sonicated for 60 min at 50% amplitude (Cole-Parmer Ultrasonic Processor CP 750). Then 5 kDa methoxy-PEG-amine (0.3408 g, 0.06816 mmol) was added to the reaction vessel. We had found that for the 4 h and 6 h oxidized cOAC, the amount of PEG could be reduced by half (0.1704 g, 0.03408 mmol). Reducing the amount of PEG used to functionalize the particles increased the antioxidant bioactivity of the 4 h- and 6 h-cOAC (data not shown). Such an improvement in the bioactivity is hypothesized to be a result of less PEG packing onto the particles, thereby, improving accessibility to the active sites of the cOAC. This effect is not as notable for the 2 h-cOAC due to the much higher surface area of the carbon core. The mixture was bath-sonicated for 20 min (Cole-Parmer 08849–00 ultrasonic cleaner). Next, the reaction vessel was removed from the sonicator and DIC (0.17 mL, 1.1 mmol) was added. The solution was stirred for 48 h at rt. Prior to purification by bath dialysis, excess DMF was removed from the reaction mixture. The sample was diluted in Milli-Q water and then concentrated by centrifugation (NuAire NU-C200R, parameters: 4500 × g, 30 min, room temperature) using Amicon Ultra-15 Centrifugal Filters (regenerated cellulose, 10 kDa MWCO). The resulting retentate was diluted with Milli-Q water and transferred to one Float-A-Lyzer G2 dialysis device (regenerated cellulose, MWCO 50 kDa, 10 mL volume capacity). To dialyze the material, the filled device was placed in a 2 L beaker containing Milli-Q water. During a period of 48 h, the water bath was slowly and continuously stirred by a magnetic stir bar/stir plate. The water was exchanged 6–8 times over the course of this time. After dialysis, the PEG-cOAC particles were sterile filtered by passing through a 0.22 μm PES membrane. UV-vis spectrophotometry was used to determine the cOAC carbon core concentration of the aqueous PEG-cOAC product.

4.4. Improved Oxidation of cAC

Powdered medical grade cAC (2.500 g, 208.3 mmol carbon) was added to a 250 mL round-bottom flask, followed by the addition of fuming (90%) HNO₃ (25 mL, 0.6 mol). The reaction mixture was placed in an oil bath pre-heated to 140 °C and was allowed to stir under reflux for 4 h. After heating, the reaction mixture was removed from the heat source and allowed to cool to room temperature. The round-bottom flask containing the reaction mixture was then placed in an ice bath to chill before pouring into a 2 M aqueous glycine solution (25 mL, 50 mmol). The resulting quenched solution was transferred to one Spectra/Por 7 dialysis membrane (regenerated cellulose, 1 kDa MWCO, 5 cm × 45 mm dimensions) for purification by dynamic bath dialysis with DI water. The OAC solution was purified via the bath dialysis for 24 h. After purification, the reaction mixture was removed from the dialysis bath and filtered through a 0.22 μm PES membrane. UV-vis spectrophotometry was used to determine the concentration of the aqueous cOAC product. Mass extinction coefficients of the cOAC variants at a wavelength of 700 nm are denoted in Table S2.

For the sieved cAC derived cOAC, <20 μm powdered medical grade cAC was collected by sieving the bulk cAC (3.0 g, 0.25 mol carbon) through a U.S.A. standard test sieve with No. 635 (20 μm) openings prior to oxidation. To aid the sieving process, a Gilson vibratory sieve shaker was used (parameters: mode = manual, time = 2 h, amplitude = 100%). The collected <20 μm cAC was subsequently oxidized by the aforementioned optimized method.

4.5. Functionalization of Improved cOAC with PEG

Aqueous cOAC was dried by lyophilization (Labconco Freeze Dry System/Freezone 4.5). The lyophilized bulk- or sieved-cOAC (16.6 mg, 1.38 mmol) was suspended in DMF (10 mL) and cup horn sonicated for 60 min at 50% amplitude (Cole-Parmer Ultrasonic Processor CP 750). Then 5 kDa methoxy-PEG-amine (0.3327 g, 0.06654 mmol) was added to the reaction vessel. The mixture was bath-sonicated for 20 min (Cole-Parmer 08849–00 ultrasonic cleaner). Next, the reaction vessel was removed from the sonicator and DIC (0.17 mL, 1.1 mmol) was added. The solution was stirred for 48 h at rt. Prior to purification by bath dialysis, excess DMF was removed by centrifugation (NuAire NU-C200R, parameters: 4500 × g, 30 min, room temperature) using Amicon Ultra-15 Centrifugal Filters (regenerated cellulose, 10 kDa MWCO). The resulting retentate was diluted in Milli-Q water and then transferred to one Float-A-Lyzer G2 dialysis device (regenerated cellulose, MWCO 50 kDa, 10 mL volume capacity). To dialyze the material, the filled device was placed in a 2 L beaker containing Milli-Q water. During a period of 48 h, the water bath was slowly and continuously stirred by a magnetic stir bar/stir plate. The water was exchanged 6–8 times over the course of this time. After dialysis, the PEG-cOAC particles were sterile filtered by passing through a 0.22 μm PES membrane. UV-vis spectrophotometry was used to determine the cOAC carbon core concentration of the aqueous PEG-cOAC product.

4.6. Characterization

4.6.1. High-Resolution Transmission Electron Microscopy (HR-TEM).

HR-TEM was performed on a JEOL JEM-2100F field emission electron microscope. Samples were prepared for analysis by bath sonicating the aqueous cOAC solutions prior to drop-casting onto 300-mesh holey lacey carbon TEM grids supported on copper. The grids with cOAC were dried in a vacuum desiccator.

4.6.2. X-Ray Photoelectron Spectroscopy (XPS).

Characterization by XPS was performed using a PHI Quantera SXM scanning X-ray microprobe with a 200 μm X-ray beam under a base pressure of 5 × 10^–9 Torr. Survey spectra were acquired using a pass energy of 140 eV with step sizes of 0.5 eV. Elemental spectra were acquired using a pass energy of 26 eV with step sizes of 0.1 eV. The carbon 1s (C 1s) peaks (284.8 eV) were used as reference for all XPS spectra corrections. cOAC samples were dried and powderized for data collection.

4.6.3. Attenuated Total Reflectance Fourier-Transform Infrared (ATR-FTIR) Spectroscopy.

ATR-FTIR spectra were acquired on a Thermo Scientific Nicolet 6700 attenuated total reflectance Fourier-transform infrared spectrometer. cOAC samples were dried and powderized for data collection.

4.6.4. Thermogravimetric Analysis (TGA).

Thermogravimetric analysis was performed on a Mettler Toledo TGA/DSC 3+ system. Alumina crucibles were used for the sample and reference pans. Prior to data collection, dried and powderized cOAC was pre-treated by ramping to 120 °C at a rate of 10 °C min⁻¹ under nitrogen. The temperature was held at 120 °C for 10 min to remove any excess moisture in the cOAC. Thermograms were acquired from 120 °C to 800 °C at a ramp rate of 10 °C min⁻¹ under nitrogen.

4.6.5. Cyclic Voltammetry (CV).

Cyclic voltammograms were acquired with a Bioanalytical Systems CV-50W voltammetric analyzer equipped with a three-electrode cell: glassy carbon working electrode, Ag/AgCl reference electrode, and platinum wire counter electrode. The aqueous cOAC was drop-cast air dried onto the surface of the glassy carbon working electrode. All measurements were run in 15 mL of 0.01 M PBS (pH 7.4) at a scan rate of 200 mV s⁻¹.

4.6.6. Nanoparticle Tracking Analysis (NTA).

NTA was performed on a Malvern Panalytical NanoSight NS300 system equipped with a continuous flow syringe pump, a blue 488 nm laser, and a 500 nm long pass filter. Aqueous cOAC samples were used for data collection.

4.6.7. Cell Culture.

bEnd.3 cells purchased from American Type Culture Collection (ATCC) were cultured in Dulbecco’s Minimal Essential Media (DMEM) (High Glucose) with 10% FBS and 1% penicillin-streptomycin (10,000 U mL⁻¹) in T-75 filter vented flasks in a 5% CO₂ incubator at 37 °C. Every three days the cells were passed 1:3 with the remainder discarded. After the 30^th pass the cells are discarded and a new line is retrieved from liquid nitrogen storage.

4.6.8. Toxicity Assessment of PEG-cOACs with bEnd.3 Cells.

To determine the toxicity of the nanoparticles alone, bEnd.3 cells were seeded at an initial concentration of 50,000 cells mL⁻¹ in 6 well plates and incubated at 37 °C in 5% CO₂ for 48 h. Afterwards, the cells were treated with 4 mg L⁻¹ of the PEG-cOACs. Live cell counts were performed after incubating the cells for 24 h. The bEnd.3 cells were detached with trypsin and stained with trypan blue before counting with a hemacytometer.

4.6.9. Protection of bEnd.3 Cells from H₂O₂ by PEG-cOACs.

In 6-well plates, bEnd.3 cells were seeded at an initial concentration of 50,000 cells mL⁻¹ and incubated at 37 °C in 5% CO₂ for 48 h. The following day, the H₂O₂ challenge assay was completed by treating the bEnd.3 cells with PBS (control), H₂O₂ (100 μM) alone, PEG-cOAC (4 mg L⁻¹) alone, or H₂O₂ (100 μM) + PEG-cOAC (4 mg L⁻¹). For the latter, the PEG-cOAC was administered to the cells 15 min after exposure to the H₂O₂. The cells were incubated for 24 h and then live cell counts were performed, detaching the bEnd.3 cells with trypsin and staining them with trypan blue before counting with a hemacytometer.

4.6.10. Protection of bEnd.3 Cells from Fe(NTA)₃ by PEG-cOACs.

In 6-well plates, bEnd.3 cells were seeded at an initial concentration of 50,000 cells mL⁻¹ and incubated at 37 °C in 5% CO₂ for 48 h. The following day, a solution of 0.1 M Fe(NO₃)₃ was prepared in 1 N HCl. Afterwards, 162 mL of the Fe(NO₃)_x(Cl)_y 1 N HCl solution was added to 100 mL of 0.08 M nitrilotriacetic acid disodium salt, Na₂H(NTA), in ddH₂O to produce a dark green solution. To the solution containing Fe(NTA)₃, NaHCO₃ was added to raise the pH to 7.4 under magnetic stirring and controlled addition to reduce foaming. The neutralized solution was filtered using a 0.22 μm PVDF syringe filter and yields an approximately 30.5 mM solution for immediate use. The Fe(NTA)₃ challenge assay was completed by treating the bEnd.3 cells with PBS (control), Fe(NTA)₃ (50 μM) alone, PEG-cOAC (4 mg L⁻¹) alone, or Fe(NTA)₃ (50 μM) + PEG-cOAC (4 mg L⁻¹). For the latter, the PEG-cOAC was administered to the cells 30 min after exposure to the Fe(NTA)₃. The cells were incubated for 24 h and then live cell counts were performed, detaching the bEnd.3 cells with trypsin and staining them with trypan blue before counting with a hemacytometer.

4.6.12. Superoxide Dismutase Activity in a Xanthine/Xanthine Oxidase (X/XO) Assay.

We used a Superoxide Dismutase Inhibition Assay purchased from Dojindo and calculated the IC₅₀ of the (4-h, nitric acid oxidized) cOAC and PEG-cOACs using Graphpad Prism 9.5. Following the procedure outlined by the manufacturer, we prepared 6 1:5 dilutions from a 740 ug mL-1 stock solution of nanoparticles as the sample and used the same solutions as a negative control (NP) for each concentration due to the optical darkening of the solutions and its potential interference with the assay. To measure the rate of $\mathrm{W}\mathrm{S}\mathrm{T}-1$ conversion to formazan, we used a BMG Clariostar operating in absorbance mode $(\lambda =440\mathrm{n}\mathrm{m})$ and measured the absorbance of each well containing xanthine/xanthine oxidase, $\mathrm{W}\mathrm{S}\mathrm{T}-1$, and the nanoparticles every 1 min for 12 min ($\mathrm{N}\mathrm{P}+\mathrm{W}\mathrm{S}\mathrm{T}1$). A blank containing $\mathrm{W}\mathrm{S}\mathrm{T}-1$ and the X/XO solution ($\mathrm{B}1$), and one containing $\mathrm{W}\mathrm{S}\mathrm{T}-1$ and water ($\mathrm{B}3$) were used to determine the maximum and minimum rates in this assay to calculate the rate of the reaction with the nanoparticles. The following equation with rates (Absorbance/min) was used, per the manufacturer:

$$\left[\right(\mathrm{B}1-\mathrm{B}3)-(\mathrm{N}\mathrm{P}+\mathrm{W}\mathrm{S}\mathrm{T}1-\mathrm{N}\mathrm{P}\left)\right]/(\mathrm{B}1-\mathrm{B}3)\times 100$$

4.6.13. Uptake of PEG-cOACs into bEnd.3 Cells.

To a #1.5-glass bottomed 24-well cell culture plate, bEnd.3 cells were seeded at a concentration of 25,000 cells mL⁻¹ and incubated for 48 hours at 37 C in 5% CO₂. Afterwards, the cells were treated with 4-h 90% HNO₃-oxidized PEG-cOACs for 5 min before washing with phosphate buffered saline (PBS). The primary antibody to PEG (Abcam, rabbit monoclonal, PEG-B-47) was diluted 1:250, and the primary antibody to Complex I was (Abcam, mouse monoclonal, ab109798) diluted 1:1000 in PBST (PBS with tween 20, 0.1%). Each well on the plate was washed with PBS, permeabilized with PBST, and blocked with 0.1% milk fat. The plates were incubated overnight in a 4 C cold room with orbital shaker at low speed. Afterwards, the cells were washed 3x with PBST 5 min at room temperature. The secondary antibodies (Anti-rabbit, AlexaFluor 594; Anti-mouse AlexaFluor 488) were added to the plate on an Orbital shaker at low speed in one addition. When staining was complete samples were fixed and nuclear stained with Hoechst 33342. Visualization and imaging analysis with fluorescence microscope Leica DMi8.

4.6.15. Stopped Flow Superoxide Dismutation Kinetic Analysis

The SOD-like activity of OACs were measured using stopped-flow method with an Applied Photophysics DX-17MV stopped-flow machine (Leatherhead, UK). The superoxide dismutation reactions in the absence and presence of OACs (or PEG-OACs) were conducted by unequally mixing DMSO/crown ether solution of KO₂ with OACs (or PEG-OACs) in 50 mM KPi, pH 8.5. The volume ratio between DMSO and KPi buffer was 1:25, and the final [OACs] (or [PEG-OACs]) = 60 nM. The decay of superoxide anion was followed with the time courses of A₂₄₄, which were adjusted for absorbance of OACs (or PEG-OACs). The data was analyzed using the program included with the stopped-flow instrument. For SO self-dismutation, the data were analyzed using a second-order rate function:

$$\mathrm{A}=\left\{{\mathrm{A}}_0/\left({\mathrm{A}}_0\times k\times \mathrm{t}+1\right)\right\}+{\mathrm{A}}_{\mathrm{f}}$$

where A is absorbance at 244 nm, ${\mathrm{A}}_0$ the initial absorbance, and ${\mathrm{A}}_{\mathrm{f}}$ the final absorbance. The second-order rate constant, $k$, is in units of OD⁻¹s⁻¹ and was converted to units of M⁻¹s⁻¹ using the molar absorbance coefficient of superoxide. For the reactions of OACs (PEG-OACs), the time courses were analyzed using a biphasic exponential decay function:

$$\mathrm{A}={\mathrm{A}}_{\mathrm{f}}+{\mathrm{A}}_1{\mathrm{e}}^{-k1\times \mathrm{t}}+{\mathrm{A}}_2{\mathrm{e}}^{-k2\times \mathrm{t}}$$

where $k_1$ and $k_2$ are the rate constants of the two phases in units of s⁻¹ and ${\mathrm{A}}_1$ and ${\mathrm{A}}_2$ are the amplitudes of the two phases. The second-order rate constants were obtained by dividing [OACs] (or [PEG-OACs]).

4.7. Traumatic Brain Injury Rodent Model.

We employed a rat model of TBI that involves a mild injury exacerbated with hemorrhagic shock and resuscitation. This model mimics the clinical condition in which systemic hypotension worsens the injury primarily through injury to the vasculature and poor reperfusion. Rapid restoration of cerebral blood flow correlates with improved structural and functional outcome.^[36,51,52] All animal procedures were approved by the IACUC of the Baylor College of Medicine (C. Robertson, PI). General anesthesia was induced utilizing 5% isoflurane in 100% oxygen and rats were placed in a vented anesthesia chamber for approximately 3–5 min. The animals were intubated and mechanically ventilated with a volume-controlled ventilator.

4.7.1. Surgical Preparation.

Under aseptic conditions, intravascular catheters were placed in the tail artery and femoral vein. The tail artery was cannulated with a 22-gauge catheter to allow for blood pressure monitoring. The femoral vein was cannulated with a 22-gauge catheter to allow for controlled hemorrhagic shock and resuscitation with Lactated Ringer solution and shed blood. Afterwards, the animals were placed on a stereotactic frame in the prone position. The head is secured by an incisor bar and ear bars. The animals were placed on a heating pad controlled by a rectal probe and kept between 36–37 °C to allow for controlled body temperature monitoring. Arterial blood gas values including pH, pCO₂, and pO₂ were obtained from the tail artery catheter on blood draw utilizing an iStat blood gas analysis system.

4.7.2. Craniectomy and Controlled Cortical Impact.

The scalp of the animal was shaved and then cleaned using an iodine-based solution. A medial sagittal skin incision was performed and the scalp was reflected. A 10 mm diameter craniectomy was performed over the right parietal cortex between the bregma and lambda using a dental drill. Care was taken to not injure the dural surface. The controlled cortical impact device was adjusted to give an impact velocity of 3 m s⁻¹ and a brain deformation of 2.5 mm. Using a heating lamp aiming at the head of the animal, the brain temperature was kept between 36 and 37 °C using a temperature probe placed into the temporalis muscle. After the cortical injury, the skull defect was closed by using an artificial bone flap made of dental acrylic to avoid extrusion of brain tissue.

4.7.3. Hemorrhagic Shock.^[36]

To induce hemorrhagic shock, blood was withdrawn with a mechanical infusion/withdrawal pump (Harvard Pump Dual RS-232) to reduce mean arterial blood pressure (MAP) to approximately 40 mmHg for 50 min. Withdrawal of blood volume at approximately 100 mL g⁻¹ of weight was required to decrease the MAP to 40 mmHg. To mimic a clinical situation of traumatic blood loss, after impact injury, blood was withdrawn at a decelerating rate with half withdrawn within the first 5 min and 25% over the next 5 min and the final 25% 10 min after injury. Animals were kept in a hypotensive phase throughout the entire hypotensive period and blood was withdrawn, if necessary, to maintain continued hemorrhage. The shed blood withdrawn was collected in citrate phosphate dextrose and stored at 4 °C until the resuscitation phase. Just prior to resuscitation, the shed blood was rewarmed to body temperature (36–37 °C).

To mimic an ambulatory transport phase, Phase II or pre-hospital, lasting 30 min utilizes Lactate Ringer’s solution to increase $\mathrm{M}\mathrm{A}\mathrm{P}\ge 50\mathrm{m}\mathrm{m}\mathrm{H}\mathrm{g}$. The animals receive a constant infusion rate of 1 mL min⁻¹ until this MAP is achieved and maintained. Following initial resuscitation, Phase III or the definitive hospital period, 80 min after impact injury, involves administration of the antioxidant compound or PBS solution. This time point would be similar to that following an accident and ambulatory transfer. This final resuscitative phase involves ventilating animals with 100% oxygen and shed blood is infused to increase MAP $\ge 60\mathrm{m}\mathrm{m}\mathrm{H}\mathrm{g}$.

4.7.4. Operative Management.

Animals were monitored throughout the entire surgical procedure taking note of blood pressure and intracranial pressure. Arterial blood gas measurements occur at the following times: end of hypotension (50 min post mTBI), end of pre-hospital period (80 min post mTBI) and finally at the end of the procedure (320 min post mTBI).

4.7.5. Brain Perfusion Measurement.

The primary outcome for this study was regional cerebral perfusion which is diminished following hemorrhagic shock in TBI and shows further decline without intervention. Blood flow was measured using Perimed laser speckle. Two distinct areas are evaluated via laser doppler: center (site of injury) and penumbra (area surrounding ischemic tissue).

4.7.5. Euthanasia.

Animals are humanely euthanized at the end of the experiment, under deep anesthesia via bilateral thoracotomy and decapitation. Necropsy is performed to visually conduct gross examination of all organs and to note any other pathological findings that may be present.

4.7.6. Treatment Groups.

In this study, a total of 43 Long Evans rats, weighing 300–350 g, were used. The TBI model used was a mild cortical impact injury (3 m s⁻¹, 2.5 mm deformation) followed by 50 min of hemorrhagic hypotension. PEG-cOAC (2 mg kg⁻¹, $n=12$; 2 died pre-operation), PEG-HCC (2 mg kg⁻¹ × 2, $n=15$; 5 died perioperative) or PBS (2 mg kg⁻¹ × 2, $n=16$; 6 died perioperative) as placebo. The assigned study drug was given intravenously at the beginning of resuscitation for all study drugs, and again 2 h after the first dose for only PBS and PEG-HCCs. Dosing was derived from equivalent in vivo concentrations to that obtained in vitro using published rat blood volume estimates and the vascular half-life of the PEG-HCCs of 1.5–2 h.^[36,53] We do not yet have an estimated half-life for the PEG-cOACs, and so here, we tested whether a single injection would be sufficient which would be advantageous in terms of a lower total dose requirement for clinical translation. Rats that died during or shortly after the procedure were excluded from the CBF analysis.

4.7.7. Statistical Methods.

One-way ANOVA with Dunnett’s comparison test was used to compare the effect of PEG-cOAC and PEG-HCCs to PBS on the difference of the mean perfusion at $t=80-100$ min and $t=290-320$ min. An $\mathrm{\alpha }$ of 0.001 was used as the significance level. Significance was established by comparing the mean $t=80-100$ and $t=200-320$ min measurements of each group. All error and graphed error bars were calculated as standard deviation.

Data Availability Statement

Research data not available.