In Situ Non‐Invasive Imaging of Neutrophil Myeloperoxidase and Skin Reactive Oxygen Species in Experimental Murine Atopic Dermatitis

Neutrophils play a key role in the innate immune inflammatory response, notably through the release of the myeloperoxidase enzyme from azurophilic granules, locally generating reactive oxygen species. Although short‐lived, these reactive oxygen species are directly involved in local tissue damage in response to microbial intrusion. Neutrophil‐derived myeloperoxidase has been reported as an important factor in the elicitation of atopic dermatitis and is considered a potential target and biomarker. This study describes the use of in situ imaging techniques comprising both chemiluminescent resonance energy‐transfer and ratiometric fluorescent microtattoos to locally and non‐invasively image myeloperoxidase activity and skin reactive oxygen species in a murine model of calcipotriol‐induced atopic dermatitis. Using neutrophil depletion to assess granulocyte contribution to the observed imaging signals, the non‐invasive longitudinal data are found to correlate with endpoint biochemical activity assays for both myeloperoxidase and reactive oxygen species by‐products, as well as with immunohistochemical analysis.


Introduction
The skin is a complex organ with a multifaceted biological role, functioning as a barrier against a broad range of environmental intrusions, while maintaining appropriate adaptive immune tolerance to the naturally occurring skin microbiome. Maintenance of this balance is critical to the skin's healthy function, as dysregulation or defects in either of these areas can result in pathological skin conditions, such as atopic dermatitis (AD). AD, also commonly known as eczema, is a chronic inflammatory disorder associated with immunological disturbance of the skin, leading to dermal sensitization and epithelial barrier dysfunction. The condition results in skin dryness, redness, and irritation, as well as increased vascular permeability, leading to the swelling of tissues. AD induction is known to be mediated by the presence of haptens-small molecules which activate skin keratinocytes and mast cells, which in turn recruit neutrophils to the skin during the elicitation phase of this disorder. These neutrophils secrete factors that create a local environment key to the elicitation phase of AD, resulting in the dermal symptoms associated with the condition.
In AD, oxidative stress has been highlighted as an important feature, not only locally, [1,2] but also systemically. [3] Neutrophils play an important role in mediating oxidative stress, notably through the generation of hypochlorous acid (HOCl)-among the most powerful oxidizing species-by the enzymatic activity of myeloperoxidase (MPO). MPO is an oxidoreductase enzyme, abundantly stored in the azurophilic granules of neutrophils, that is released upon granulocyte activation, resulting in the local generation of reactive oxygen species (ROS) from hydrogen peroxide for microbicidal purposes. Recently, neutrophils-and more specifically neutrophil-derived MPO-have been identified as playing a central role in the pathogenesis of AD, potentially serving as novel therapeutic targets for the management of this condition. [4][5][6] MPO and other ROS-generating enzymes act to induce inflammatory responses, DNA damage, and cellular stress. Because the impact of neutrophil degranulation on tissue environment is highly localized and dependent on ROS, these highly www.advancedsciencenews.com www.advsensorres.com reactive species are considered key mediators of inflammationrelated damage. However, ROS are molecules that are challenging to track in vivo as they are very labile and short-lived by nature. In both preclinical models and clinical samples, MPO evaluation often requires the ex vivo collection of tissue to evaluate enzymatic activity, though this parameter can vary significantly as a function of time, or the specific section of tissue sampled, and may necessitate subsequent histological staining analysis. Regarding ROS themselves, their concentrations are often measured indirectly through the quantification of lipid peroxidation products, such as malondialdehyde (MDA), or by measuring the relative levels of oxidizing and/or reducing enzymes. [6] It should be noted that such measurements typically require either an invasive biopsy or sacrifice of the animal being studied, rendering this method poorly suited for longitudinal studies. Accordingly, the use of non-invasive imaging modalities represents an interesting opportunity to directly evaluate and quantify either ROS themselves, or the biological activity of enzymes such as MPO in living tissues.
Using calcipotriol (MC903)-induced AD in mice as a preclinical model, this study describes the use of non-invasive in situ imaging techniques, including chemiluminescent resonance energy-transfer (CRET) for the imaging of MPO activity as well as recently described ratiometric cyanine-based fluorescent microtattoos for the measurement of ROS in the skin. [7] CRET was conducted via combined luminescence imaging following the systemic administration of luminol-which is oxidized by MPO, generating UV light-and near-infrared (NIR) quantum dot (QD) nanoparticles-which convert this emitted UV light into NIR light-to image neutrophil MPO activity in situ. [8,9] This approach was combined with the non-invasive percutaneous monitoring of ROS using a ratiometric ROS-responsive microtattoo coupled with a portable, hand-held fluorescence reader to image dermal ROS production. [10,11] The microtattoo consists of a PEGylated ROS-sensitive fluorescent sensor, delivered using microneedles (MNs) composed of a dissolving polymer. When applied to the skin, the tips of these MNs dissolve, releasing the sensor into the skin, along with an inert reference dye to allow for ratiometric monitoring. The fluorescence of these dyes can be measured to allow real-time detection of skin ROS. [11] To confirm the neutrophil MPO-specificity of the two modalities, neutrophils were depleted using anti-Ly6G/Ly6C ( Gr-1) antibodies in MC903-treated mice. These methods were compared to a standard o-dianisidine (ODA) biochemical MPO activity assay in tissue homogenates, as well as to immunohistochemical tissue analyses.

MC903 Induces Neutrophil Recruitment into Dermatitis Lesions Leading to Increased MPO Activity
Experimental dermatitis was induced in C57BL/6 mice by chronic daily exposure of depilated back skin and dorsal ear skin to either MC903 (50 μm in ethanol (EtOH)), or EtOH alone as a vehicle control. A neutrophil-depleted group was investigated alongside this, through the administration of Gr-1 antibodies by intraperitoneal injection prior to and during MC903 induction. Depletion resulted in a strong (76%) reduction in neutrophil counts in the depleted group compared to the reference group ( Figure 1A,B). Within three days of MC903 exposure, signs of dermatitis (skin thickening, erythema, and desquamation) became observable and continued to increase in severity until day 10 ( Figure 1C). Gr-1 antibody-receiving mice displayed attenuated onset of this phenotype, with ear skin PASI-scores appearing statistically lower on the 10th day of induction (average scores of 3.4 vs 4.6 for Gr-1+MC903 and MC903 alone respectively). Ear swelling was also delayed by neutrophil depletion ( Figure 1D) with ear thickness of 241 μm vs 263 μm observed on day 7 (p ≤ 0.01), and 296 μm vs 322 μm observed on day 10 (p ≤ 0.001) for Gr-1+MC903 and MC903 alone respectively. The same trend was observed for the back skin of mice induced with MC903, though the difference was less striking, as PASI scores plateaued after day 7 ( Figure 1E).
To confirm that antibody-mediated depletion ( Gr-1 group) truly resulted in decreased neutrophil counts and increased MPO activity in MC903-induced AD skin lesions, tissues were excised and analyzed for MPO activity. As indicated in Figure 2A, MC903 treatment resulted in a substantial elevation of detectable MPO activity in tissue lysate using a colorimetric ODA oxidation assay, reflecting the quantity of active neutrophil granulocytes within the tissue. While this assay is inherently variable, it served to clearly demonstrate that skin lesions from Gr-1-treated mice showed a marked reduction in neutrophil-derived MPO. When examined by histology, it was possible to distinguish this difference by observing the extent of MPO-positive cell infiltration within both the ear and back skin of induced mice, as compared to the reference group ( Figure 2B-D). When examining hematoxylin and eosin (H&E) stained sections, it was possible to clearly distinguish the reduced epithelial thickening and dermal edema in Gr-1-treated mice.
By using toluidine blue (TB), a metachromatic dye which stains acidophilic mast cell granules in purple, it was observed that the infiltration of mastocytes upon MC903 induction was not meaningfully affected by neutrophil depletion.

MC903-Induced Atopic Dermatitis and Human Atopic Dermatitis Show Imbalanced ROS-Response Gene Expression with Common and Distinctive Features
Although MC903-induced AD is well known to recapitulate some clinical and molecular features of human AD, the ROS balance associated with this model has not been exhaustively characterized. To investigate whether dermal symptoms were indicative of changes to ROS metabolism, RNA from the skin collected upon completion of the study was analyzed by RT-qPCR against a panel of ROS-scavenging and ROS-producing enzymes to determine the degree of response compared to vehicle-treated animals ( Figure 3A). The same genes were retrieved from human microarray data collected by De Benedetto et al. comparing non-lesional skin to skin from patients with AD and psoriasis ( Figure 3B). [12] Upon examination of these data, it appeared clear that MC903-induced dermatitis resulted in changes in gene expression for many of these genes. Of particular interest was the Gr-1+MC903 group, where important changes in gene expression levels were observed compared to mice treated with MC903 alone, in particular for the ROS-clearing enzymes PRDX5, SOD2, and GPX2, which were more strongly expressed in antibody-depleted mice compared to those treated with MC903 without antibody depletion. This same observation was noted for genes found to be downregulated by MC903-induction, where the extent of this downregulation was attenuated in the Gr-1+MC903 group ( Figure 3A). When examining human skin microarray data, the changes observed were more modest than those observed by RT-qPCR, but nonetheless showed differences in expression levels for several ROS-associated enzymes. However, it should be noted that the murine models and human samples did not show the same degree of alteration, nor necessarily alterations to the same genes, likely reflecting their distinct etiology and spanning period (days vs months/years for mice and humans, respectively).

CRET Full Body Imaging Monitors Neutrophil MPO to Efficiently Track Neutrophil MPO Activity in MC903-Induced AD
Although neutrophil MPO activity can be measured using tissue lysates (Figure 2A), such assays require invasive-and typically terminal-sampling of tissues. Moreover, the enzymatic activity is short-lived upon tissue freezing and must be evaluated quickly following sample collection. Additionally, even after taking such considerations, the data are often heterogenous between animals and between biological replicates, and this methodology also does not allow for longitudinal monitoring.
MPO activity was detected in situ using CRET full-body imaging following intravenous injection of luminol along with NIR QD nanoparticles, generating MPO-induced UV luminescence which is converted by the co-injected QDs to generate an IRshifted bioluminescent signal to image neutrophil activity. This technique has been tested on MPO-knockout mice and was found to be specific to the activity of this enzyme. [8,9] As seen in the images shown in Figure 4A, luminescence was significantly elevated above background levels on the right ear and back skin of MC903-induced mice. In contrast, neutrophildepleted mice displayed reduced MPO activity which was coherent with delayed phenotype onset. By day 8, however, neutrophils had initiated a rebound response to Gr-1 antibody treatment, as classically observed, with neutrophil counts increasing to 2.53 × 10 9 per mL and 0.84 × 10 9 per mL for Gr-1-treated and MC903-induced mice respectively, see Figure S1 (Supporting Information). [13][14][15] To properly adjust for the basal levels of activity for each animal, as well as to compensate for any deviation in intravenous dosage, quantification was corrected using the left ear as a reference value for both the induced (right) ear ( Figure 4B) and back skin regions ( Figure 4C). Imaging sessions were conducted on the 3rd, 8th, and 10th day following initial MC903 application and data were used to draw the MPO activity onset graphs seen in Figure 4. These luminescence data were coherent with the observed MPO activity in lysates at the end of the study (Figure 2A) and also corroborated the IHC data ( Figure 2B,C).

Cyanine-Based Ratiometric Microtattoos Efficiently Track Skin ROS Production In Situ and Correlate with MPO Activity and ROS-Peroxidation Products
Prior to this study, cyanine-based fluorescent microtattoos had never been applied to mouse skin. Since AD skin lesions are typically dry due to desquamation, 500 and 800 μm MNs were tested on both vehicle-and MC903-induced mice to determine which length was best suited to deliver the microtattoo into the skin.
MNs of both lengths were cast and prepared as previously described (Figure 5A,B). [7] Following application to the skin, mice (uninduced or MC903-induced) were imaged using an IVIS Imager to measure Cy7.5 fluorescence as well as using a portable Heatmap representation of relative gene expression levels of MC903-induced mouse skin compared to vehicle-treated skin for the listed genes determined by RT-qPCR using RPLP0 as a reference housekeeping gene (n = 9, 5, and 4 respectively for the EtOH, MC903, and Gr-1+MC903-induced groups). B) Expression levels retrieved from transcription profiling by array dataset E-GEOD-26952 expressed as fold-changes compared to the mean of all nonlesional samples of the dataset to produce a heat-map. [12] (n = 7, 4, and 5 respectively for Normal, Psoriasis, and AD samples). Data are expressed using the mean fold-changes, with the scale as a color-reference.
multi-wavelength fluorescence reader to measure both Cy5 and Cy7.5 fluorescence intensity. As visible in the Cy7.5 images in Figure 5C, the 800 μm MNs embedded deeper into the skin, delivering more fluorescent dye than the 500 μm MNs. The 800 μm MNs yielded a microtattoo with an average of threefold greater reference dye fluorescence intensity than the 500 μm MNs ( Figure 5D). Ratiometric measurements between Cy5 and Cy7.5 demonstrated a strong difference (i.e., a > 13-fold increase (p ≤ 0.05)) between vehicle-and MC903-induced skin, compared to an ≈ twofold difference for the 500 μm MNs ( Figure 5E). Upon excising the skin from these mice immediately following application and cryosectioning it for fluorescence microscopy, the amount of Cy7.5 fluorophore delivered to the skin was found to be visibly higher for the 800 μm MNs. The signal from the 500 μm MN-treated skin was barely observable and was mostly superficial, as opposed to the 800 μm MNs which yielded strong signals 100 μm deep in the skin, with diffuse fluorescence observed as deep as ≈250 μm ( Figure 5F).
The 800 μm cyanine-based fluorescent microtattoos were thus used to investigate ROS present in the skin of the MC903induced and vehicle-treated mice on the 7th day following initiation of AD induction alongside Gr-1-treated mice also induced with MC903. Microtattoos were applied and their fluorescence intensities were measured over 96 h to establish the kinetics of the response (Figure 6E). Within the first 24 h, the ratiometric fluorescent signal from both dyes was found to be over 100-fold greater in the MC903-induced group when compared to the vehicle-treated group and this difference persisted over the following 72 h (up to the 10th day of induction). Consistent with the observations from CRET imaging, ratiometric fluorescence did not increase in the tattoo applied to the Gr-1treated MC903-induced mice. To complement these MN-derived data, skin from the vehicle-treated or MC903-induced zones on these mice was collected upon sacrifice of the mice and was evaluated for MDA content. MDA is a well-established marker of oxidative stress as it is an end-product of polyunsaturated fatty acid peroxidation within cells that is often used to indirectly measure the presence of ROS in biological matrices. Whereas MC903 ex-posure induced a threefold increase in MDA content within the skin compared to the EtOH control, skin from the neutrophildepleted mice induced with MC903 displayed MDA levels in the same range as the skin of the vehicle-treated mice ( Figure 6B). This observation appears to confirm that ROS production is truly impaired by neutrophil depletion, a notion directly reflected in the CRET and microtattoo fluorescence measurements (Figures 4 and 6A).

Discussion and Conclusion
ROS are dually involved in oxidative damage as well redox signaling, and accordingly play physiological as well as pathophysiological roles in biological processes. This is notably the case for MPO-derived ROS, which are generated following the stimulation of neutrophils, leading to the production of several potent oxidants including H 2 O 2 , O 2 − , and HOCl. Although these reactive compounds are short-lived, their chemical reactions result in oxidative damage to local tissues.
Neutrophils are also known to be required for the development of itch-evoked scratching in AD. [5] In oxazolone-induced AD, neutrophils, and more precisely neutrophil-derived MPO itself-as demonstrated through the use of neutrophil-specific knockout models-have been documented to play key roles in the initiation phase of the disease through the promotion of dermal production of IL-1 , which drives the activation of migratory dendritic cells, thus promoting the priming of T cells. [4] This is, to our knowledge, the first study to illustrate the effect of neutrophils on MC903-induced experimental AD. MC903 is known to induce neutrophil recruitment in the skin, with an apex occurring around day 9 following induction, based on the presence of MPO, as detected by ELISA in skin homogenates. [16] These data are coherent with our observation that MPO activity peaks between the 8th and 10th days of induction ( Figure 4B,C). In a recent study by Li et al., the authors identified-using knockout mice-that the IL-33-ST2/MyD88 axis is pivotal to the inflammatory response, particularly in dendritic cells, likely by limiting the elicitation phase of the induced condition. [16] In the absence of ST2 (the receptor for IL-33), MC903 elicited an attenuated AD-like phenotype with reduced neutrophil infiltration, inflammation, and MPO release. [17] In early phenotype onset-prior to day 3 following induction-thymic stromal lymphopoietin (TSLP) is reported to be the main driver. [18] Our study also demonstrated the feasibility of efficiently tracking ROS generation within skin lesions, which represents an important asset for non-invasively tracking the biological efficacy of therapeutic agents. Measurement of MPO activity in a longitudinal and non-invasive manner by imaging has been found to provide i) less variability than conventional and terminal biochemical assays (Figure 2A), and ii) a better definition of the time-frame corresponding to elevated MPO production ( Figure 4B,C).
MNs are particularly well-suited for the transdermal evaluation of biologically relevant processes taking place within the skin, such as ROS production. Although MC903 induction resulted in skin thickening and scaling, tattoos from 800 μm MNs were successfully administered to the skin ( Figure 5) and generated fluorescent signals correlating with the degree of MDA formation in the skin ( Figure 6A,B). Such sensitive devices may  www.advancedsciencenews.com www.advsensorres.com benefit the study of other skin conditions such as wounds or psoriasis, where neutrophils and their associated molecules are known to be present at greater levels than in AD. [19] Although neutrophil MPO is known to generate specific, highly active ROS, the microtattoos serve to measure the total oxidizing potential of the skin and interstitial fluid at the site of application. It is thus clear that other forms of ROS, such as MPO-derived HOCl, likely contribute to the observed fluorescent signal, as substantial changes to the gene expression of many ROS-generating and -scavenging enzymes in the skin resulting from MC903-induced AD were observed ( Figure 3A). For a disease like AD, oxidative stress has long been proposed as an important feature, since ROS are notably increased during AD exacerbation, along with decreased antioxidant capabilities, [3] both of which are known to occur during dermal inflammation in AD skin lesions. Accordingly, such ROS-sensing/imaging modalities represent an interesting advance for the longitudinal monitoring of these processes in vivo to better track the skin's response to a wide range of test articles.

Experimental Section
Experimental AD Mice: Female C57BL/6 mice 6-8 weeks of age (Charles Rivers Laboratories, St-Constant, QC) were used for this study. This study was approved by the Cégep de Lévis Animal Care Committee (#002-22) and complied with CACC standards and regulations governing the use of animals for research. Following arrival, animals were subjected to an acclimation period of 7 days before the beginning of the study. Tap water and standard certified commercial rodent diet (Envigo 2018) was provided ad libitum except during designated procedures requiring the handling of animals outside of their housing cages (such as blood sampling or dosing).
Mice were shaved with an electric razor under isoflurane anesthesia and depilated using a depilating cream (Nair Sensitive Skin, Church & Dwight) for 3 min, after which cream was removed and the skin was further rinsed with tap water. The following day, induction of AD was initiated using a 50 μm solution of MC903 (MedChemExpress, Monmouth Junction, NJ, USA) solubilized in 100% EtOH. Right ears were treated with 20 μL of MC903 solution (equivalent to 1 nmol of MC903) on the dorsal side, while left ears were used as a reference (20 μL EtOH). The upper half of the back skin (≈ 3 × 3 cm) was treated with 50 μL MC903 under anesthesia and allowed to dry for 5 min.
For neutrophil depletion, mice were administered via intraperitoneal injection 100 μg anti-Ly6G/Ly6C (RB6-8C5; BioXCell, Lebanon, NH, USA) on study days −1, 2, 5, and 8. For neutrophil counts, blood samples (≈50 μL) were collected and mixed with 4 mL of red blood cell lysis buffer (ThermoFisher Scientific) for 5 min at room temperature, then centrifuged 5 min at 500 g and treated again with lysis buffer. After the second centrifugation, the cell pellet was adjusted to 10 6 cells mL −1 , using a MOXI Z Cell Counter (ORFLO Technologies, Ketchum, ID, USA) to place 750 000 cells on a glass slide using a Cytospin 4 Cytocentrifuge (Epredia, Waltham, MA, USA). Slides were air-dried at least overnight. Slides were then fixed in methanol for 5 s and Wright-Giemsa stained (Hemacolor Rapid staining, Sigma) prior to mounting. Slides were visualized using a digital slide scanner (PANNORAMIC MIDI II, 3DHistech, Budapest, Hungary). The sections were then visualized using CaseViewer software. Differential leukocyte counts were determined using an Element HT5 Veterinary Hematology Analyzer (Heska) using blood collected on K 2 EDTA-coated tube.
Mice were weighed and evaluated for ear thickness and PASI scores 3× per week. PASI scores (see Table S1, Supporting Information) were determined by visual evaluation of both erythema and desquamation/scaling and by measurement of ear thickness using a micrometric caliper. For back skin, a manual palpation was performed to score the skin thickening. Animal nails were cut every week to prevent scratching at the induced sites.
Chemiluminescence Resonance Energy Transfer (CRET) Imaging: For in vivo imaging of MPO activity, an adaptation of the Zhang et al. procedure was utilized. [8] Each mouse was injected by the tail vein with 125 μL of 100 mm luminol sodium salt (Sigma) and 0.2 μm Qtracker 800 Vascular Labels (Invitrogen, Ex/Em (405-760/800 nm) in PBS. Two minutes after injection, animals were imaged in an IVIS Lumina XR in vivo imager (PerkinElmer) in bioluminescence configuration (blocked excitation, fully open emission, f number: 1, binning factor: 8, exposure: 60 s, field of view: 12.5 cm) and imaged for 1 min. Photon flux (radiance photon s −1 cm −2 ) was calculated in area of interest by manually circling each ear and the corresponding back skin section.
Preparation of MNs Containing Sensor and Reference Dyes H-Cy5-PEG and Cy7.5-PEG: MNs were manufactured by adapting a previously described solvent casting method. [7,14] Briefly, to 2 mL of dH 2 O, ULMW HA (1.917 g) and dextran (0.234 g) were added and mixed thoroughly. The mixture was heated for 30 min in a 75°C oven and centrifuged (Sorvall ST 16R, Ther-moFisher Scientific, Waltham, MA) for 5 min at 4700 g. H-Cy5-PEG and Cy7.5-PEG were added, resulting in concentrations of 150 μm (H-Cy5-PEG) and 100 μm (Cy7.5-PEG) respectively. Using a 1 mL syringe, roughly 100 μL of this solution was cast into PDMS molds (Micropoint Technologies Pte. Ltd., Singapore) and these molds were secured with tape in 6-well cell culture plates (Sarstedt AG & Co., Nümbrecht, Germany). The plates were covered, secured with parafilm, and centrifuged for 5 min at 2300 g. After centrifugation, polymer solution was re-applied, and the plates were rotated 180°and centrifuged again. This process was repeated a total of four times. After the final centrifugation, any excess polymer solution was removed from the molds using a spatula, and the molds were placed in a vacuum chamber at 100 mbar for 30 min. Roughly 100 μL of dye-free polymer solution was added to each mold to form a backing layer, and the MNs were allowed to dry for 18-24 h at 25°C and 60% humidity, after which they were removed from the molds ( Figure S7, Supporting Information).
Detection of Skin ROS Using MN Tattoos: MNs were applied under isoflurane anesthesia using a spring-loaded applicator (Micropoint Technologies Pte LTD, Singapore) for 2 min for each application. MNs were removed and the application sites were scanned for fluorescence using a portable fluorescence reader. Ratiometric fluorescence measurements (Cy5/Cy7.5) were used as an indication of ROS levels. Animals were then immediately imaged using an IVIS (Lumina XR, PerkinElmer) equipped with ICG filter sets. After the final fluorescence readings, animals were sacrificed, and the MN application sites were dissected and directly embedded in OCT-sucrose (20%) in isopentane chilled by dry ice and immediately sectioned with a cryostat (50 μm sections). Sections were covered and histologically examined using a Cy7 filter set.
Malondialdehyde and MPO Biochemical Assay: Malondialdehyde (MDA) concentrations were determined using a Thiobarbituric Acid Reactive Substances (TBARS) Assay (Cayman Chemical #700870). Briefly, skin samples (25 mg) frozen directly into Precellys CK-14 tubes were homogenized in 10 μL mg −1 RIPA buffer (50 mm Tris-HCl, pH 8.0, 150 mm NaCl, 1% NP-40, 1% SDS, and 0.5% Na-deoxycholate) freshly supplemented with protease inhibitor (Roche Complete Mini). Samples were homogenized using Precellys for 2 rounds at 6500 rpm (30 s) between which samples were held for 5 min on ice. Samples were then clarified by centrifugation at 1600 g for 10 min at 4°C. 100 μL of sample lysate was then incubated with 100 μL of 10% trichloroacetic acid (TCA) and 800 μL of thiobarbituric acid reagent (5.3 mg mL −1 ). Samples were incubated along with MDA standards at 95°C for 1 h, then centrifuged at 1600 g to allow fluorescence measurement ( ex = 530 nm/ em = 550 nm) of the sample supernatant in a plate reader. MDA concentrations were interpolated from the standard curve and reported as ng MDA/mg tissue.
For MPO activity measurements, frozen tissues (≈20 mg) were homogenized in CK28 Precellys tubes with 10 μL mg −1 of 50 mm KH 2 PO 4 0.5% hexadecyltrimethyl ammonium bromide (HTABr) for 2 rounds at 2500 g (30 s) between which samples were held for 5 min on ice. Samples were then subjected to three freeze-thaw cycles using dry ice and clarified by centrifugation at 13 000 g for 10 min at 4°C. The supernatant was then transferred and added to 96-well plates containing 0.262 mg mL −1