Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Oct 7.
Published in final edited form as: Neuroimage. 2026 Sep 10;340:122217. doi: 10.1016/j.neuroimage.2026.122217

Complementary PET and MRI of myeloperoxidase activity as a noninvasive surrogate for tissue acrolein in experimental neuroinflammation

Enrico Küllenberg a,b, Siyuan Sun d,e, Cuihua Wang a,b, Negin Jalali Motlagh a,b, Mariane Le Fur a,c, Gregory R Wojtkiewicz b, Seth A Herr d, Riyi Shi d,e,f,g, John W Chen a,b,*
PMCID: PMC13641933  NIHMSID: NIHMS2213707  PMID: 42716359

Abstract

Background:

Myeloperoxidase (MPO), a pro-inflammatory enzyme, and acrolein, a reactive aldehyde present in the environment and produced in the body, play pivotal roles in experimental autoimmune encephalomyelitis (EAE) and multiple sclerosis (MS). However, the relationship between MPO activity and acrolein levels is not well understood. The purpose of this study was to explore this relationship in different EAE mouse models utilizing molecular MRI (heMAMP) and PET ([18F]MAPP) imaging that report MPO activity.

Methods:

Chronic progressive C57BL/6-myelin oligodendrocyte glycoprotein (MOG)-EAE (n = 4–7/stage) and relapsing-remitting SJL-proteolipid protein (PLP)-EAE (n = 3–8/stage) mice were imaged using heMAMP-MRI and [18F]MAPP-PET across disease stages. 4-aminobenzoic acid hydrazide (ABAH, 40mg/kg IP qd) was used to inhibit MPO in SJL mice (n = 5). MPO protein and activity, and acrolein-adduct levels were assessed ex vivo. P < 0.05 was considered statistically significant.

Results:

heMAMP-MRI and [18F]MAPP-PET showed MPO+ lesions in the brains of SJL-PLP-EAE and spinal cords of C57BL/6-MOG-EAE mice during acute disease and a significant decrease at later stages. While heMAMP-MRI allowed detection of smaller MPO+ lesions (mean lesion volume MRI vs PET; SJL: 0.35 ± 0.037 mm3 vs 4.11 ± 0.56 mm3, p < 0.001; C57BL/6: 0.49 ± 0.069 mm3 vs 8.25 ± 1.37 mm3, p < 0.001), PET detected larger total MPO+ lesion volumes in both models. Ex vivo MPO protein/activity and tissue acrolein-adducts correlated in C57BL/6-MOG-EAE (r = 0.89, p < 0.001 and r = 0.82, p < 0.001, respectively) and acute SJL-PLP-EAE (r = 0.7, p = 0.05 and r = 0.76, p = 0.03, respectively). Serum acrolein-adduct levels showed inconsistent correlation with MPO activity imaging, ex vivo MPO assays, and tissue acrolein-adduct levels, likely due to acrolein-adducts lasting for days in tissues. Similarly, inhibiting MPO after symptom onset did not immediately decrease tissue levels of acrolein-adducts.

Conclusion:

These findings highlight the complex interplay between MPO and acrolein, with serum acrolein-adduct levels offering limited insight into CNS tissue pathophysiology. This underscores the importance of molecular imaging to dynamically detect active inflammatory lesions and monitor oxidative stress in vivo.

Keywords: Myeloperoxidase (MPO), Acrolein, Experimental autoimmune encephalomyelitis (EAE), Multiple sclerosis (MS), heMAMP-MRI, [18F]MAPP-PET

1. Background

Oxidative stress drives demyelination and axonal damage and initiates neurodegeneration in early multiple sclerosis (MS) (Adamczyk and Adamczyk-Sowa, 2016). Activated microglia, neutrophils, and macrophages release myeloperoxidase (MPO), which catalyzes the formation of highly reactive species, including hypochlorite, tyrosyl radicals, and reactive aldehydes such as acrolein (2-propenal) (Anderson et al., 1997; Nagra et al., 1997; Gray et al., 2008). One of the most abundant oxidizing enzymes secreted in active MS plaques and overexpressed in chronic active lesions of progressive MS patients, MPO creates a toxic environment that prevents remyelination and repair (Gray et al., 2008; Nagra et al., 1997). Inhibiting MPO during peak disease in experimental autoimmune encephalomyelitis (EAE) mice ameliorates disease severity, further highlighting its pathophysiological importance (Zhang et al., 2016).

Acrolein, a byproduct of combustion, can also be generated endogenously via lipid peroxidation and MPO-mediated degradation of threonine (Stevens and Maier, 2008). Acrolein has a half-life of several days and damages proteins, lipids, and DNA at micromolar concentrations, thereby activating harmful enzyme cascades and generating additional free radicals (Luo and Shi, 2005, 2004; Shi et al., 2011; Stevens and Maier, 2008). Systemic levels of 3-hydroxypropylmercapturic acid (3-HPMA), a metabolite of acrolein, are elevated in both EAE mice and MS patients (Tully et al., 2018). Chemical scavenging of acrolein improves clinical symptoms in EAE (Hamann et al., 2008; Park et al., 2014; Tang et al., 2024). Critically, acrolein stimulates production of macrophage chemotactic protein-1 (MCP-1), attracting macrophages that produce MPO (Ong et al., 2012). Thus, acrolein and MPO can create a vicious cycle perpetuating oxidative stress.

MPO-Gd (bis-5-HT-Gd-DTPA) is an activatable paramagnetic MRI agent that allows in vivo imaging of MPO activity (Chen et al., 2008). While MPO-Gd does not cross the intact blood-brain barrier (BBB), its activation results in signal amplification, enabling sensitive and specific detection of smaller active lesions at earlier disease stages compared to conventional MRI with nonspecific contrast agents, which often lag disease activity (Li et al., 2019; Pulli et al., 2017; Wang et al., 2021, 2019). A new generation of MPO activity-sensing MRI agent (heMAMP), based on a macrocyclic chelate, offers significantly greater Gd stability and up to 3X higher efficacy compared to MPO-Gd (Wang et al., 2021). Additionally, a PET MPO activity-sensing agent [18F]MAPP, due to its smaller molecular size, can cross the intact BBB and offers potentially higher sensitivity (Wang et al., 2019). This class of MPO imaging agents have been validated to reflect MPO activity (Chen et al., 2008; Pulli et al., 2013; Forghani et al., 2012; Wang et al., 2019, 2021).

The potential clinical impact extends beyond improved detection. Visualizing MPO activity could enable earlier therapeutic intervention, potentially preventing irreversible tissue damage, and serve as a valuable biomarker for monitoring treatment efficacy, particularly for novel therapeutics targeting oxidative stress as well as inflammatory and reparative pathways. Assessing the correlation between MPO activity and acrolein-adduct levels would provide a mechanistic framework to understand their relationship, rationalize the use of MPO imaging to better monitor oxidative damage in neuroinflammation, and set the stage for therapies targeting both molecules. Therefore, this study aims to characterize their relationships in two different EAE models across stages in tissues and serum.

2. Materials and methods

2.1. EAE induction, clinical scoring, and treatment protocol

All animal experiments were approved by the local institutional animal care and use committee and were conducted in accordance with the ARRIVE (Animal Research Reporting of In Vivo Experiments) Essential 10 guidelines for the reporting of animal experiments. Relapsing-remitting or chronic-progressive EAE was induced in 8–10-week-old female SJL (n = 65) and C57BL/6 (n = 46) mice using proteolipid protein peptide (PLP139–151, 100 μg/mouse) or myelin oligodendrocyte glycoprotein peptide (MOG35–55, 200 μg/mouse), respectively, as described previously (Li et al., 2019). Sham induction was performed identically without the MOG- or PLP-antigen. We performed clinical assessment as follows: 0 = normal; 1 = flaccid tail; 2 = unable to use either hind leg; 3 = complete paralysis of hindquarters; 4 = very weak, unresponsive, and rapid breathing; 5 = moribund (Fig. S1). Mice that remained asymptomatic were excluded. In a subset of SJL mice (n = 5), 4-aminobenzoic acid hydrazide (ABAH) was intraperitoneally injected (40 mg/kg BW/d) after symptom onset until sacrifice at the acute time point.

2.2. MR and PET-CT imaging

MR imaging was performed on a 4.7-T unit under respiration-monitored isoflurane anesthesia (Bruker). heMAMP was synthesized as described previously (18). The brains of SJL mice were imaged during the acute stage (1–2 days after symptom onset, ~d11), at remission (defined by improvement by at least 1 score, ~d18) and at relapse (defined as worsening by 1 point, ~d24) (Fig. S1A). Precontrast T1-weighted images were acquired, followed by postcontrast T1-weighted images sequentially every 15 min for 1 h after intravenous injection of 0.1 mmol/kg heMAMP. Imaging parameters, using a head coil, were as follows: T1: repetition time (TR) = 873 ms, echo time (TE) = 17.8289 ms, matrix size 256 × 256, and slice thickness 0.5 mm; Twelve axial sections were acquired. The spinal cords of C57BL/6 mice were imaged at an acute (1–2 days after symptom onset, ~d15) and a chronic time point (7–8 days after symptom onset, ~d21) (Fig. S1B). Twelve sagittal sections were acquired with the same parameters as above using a surface coil. Postcontrast images for heMAMP were subtracted from the precontrast images to identify the lesions. heMAMP images were quantified by lesion number, lesion volume and contrast-to-noise ratio (CNR) (Breckwoldt et al., 2008). ROIs were manually drawn around heMAMP enhancing lesions identified on subtraction images (60 min postcontrast – precontrast). Lesion volume was calculated by multiplying the lesion area with slice thickness (0.7 mm). CNR = (mean intensitylesion – mean intensitynormal appearing brain)/SDnoise. Corrected CNR: CNR60min = CNR60min– CNRprecontrast.

PET imaging was performed on a Siemens Inveon PET-CT scanner. Mice were intravenously injected with 11.1 MBq (300μCi) of [18F] MAPP. PET-CT imaging was performed under respiratory-monitored isoflurane anesthesia 2 h after radioprobe injection for 30 min. Reconstruction was performed using 3D ordered-subsets expectation-maximization with maximum a posteriori, with a spatial resolution of approximately 1.5 mm. Image visualization was performed using 3DSlicer. To calculate the standard uptake value ratio (SUVR), the mean voxel intensity of a manually drawn volume of interest (VOI) around the lesion was divided by the mean voxel intensity of cervical musculature. The SUV of cervical musculature remained stable across disease stages (Fig. S2). SUVR maps were calculated by dividing the mean intensity of each voxel by that of the cervical musculature. Areas with uptake exceeding the background (SUVR > 1.6) were manually segmented. Image analyses were performed by a neuroradiologist with > 20 years of experience and a neuropathologist with 5 years of experience, blinded to the identity of the animals.

2.3. Tissue harvest and biochemical assays

For ex vivo analysis SJL mice were sacrificed after ~11 (acute and ABAH treated), ~18 (remission), and ~24 (relapse) days, and C57BL/6 mice after ~15 (acute) and 21 (chronic) days (n = 8–10 per group) after induction, corresponding to the imaging time points. Under deep isoflurane anesthesia, blood was drawn via cardiac puncture. Afterward, brains and spinal cords were harvested into ice-cold phosphate-buffered saline (PBS). Blood was kept at room temperature (15 min), centrifuged (15,000 rpm, 4 °C, 15 min) to collect serum. Tissue was mechanically homogenized in 500 μl cetyltrimethylammonium bromide (CTAB) buffer (30 s) followed by pulsed sonification (30 s). After centrifugation (15,000 rpm, 4 °C, 15 min), protein concentration was determined using a bicinchoninic acid (BCA) kit per manufacturer’s instructions (Pierce BCA, Thermo Scientific, Waltham, MA). MPO Enzyme-linked Immunosorbent Assay (ELISA) and activity assay were performed as previously described using a commercially available kit (Mouse Myeloperoxidase DuoSet Elisa, R&D Systems) (Pulli et al., 2013). MPO ELISA and MPO activity assays were not conducted under blinded conditions since these assays are based on objective readouts that are quantified according to standardized protocols and are therefore not considered susceptible to observer bias.

2.4. Acrolein western-immunoblot

30 μg of protein were mixed with β-mercaptoethanol and Laemmli buffer and loaded onto 15% Tris-HCL gels. Duplicate gels were stained with Coomassie blue (ThermoScientific, 20278) for normalization. Proteins were transferred to nitrocellulose membrane using Power Blotter-Semi-dry Transfer System (Thermo Fisher Scientific) which was blocked with 1x casein (Vector, #SP-5020) for 30 min, incubated overnight (4 °C) with anti-acrolein-lysine antibody (StressMarq, SMC-505, 1:1000), washed with 0.1% TBST, and incubated with secondary antibodies (Vector, #BA-2000, 1:2000) for 45 min. Following TBST washing, the membrane was treated with VECTASTAIN ABC-AmP reagent (Vector Laboratories) for 10 min. DuoLuX substrate (Vector, #SK-6605) was used for chemiluminescent detection on an Azure C300 Western blot imaging system (Azure Biosystems). Band intensities were quantified using ImageJ software (NIH). Acrolein-adduct levels were normalized to total protein loading control assessed by Coomassie Brilliant Blue staining. The acrolein-lysine antibody is specific for acrolein-modified proteins and does not detect free acrolein, does not cross-react with crotonaldehyde, hexanoyl lysine, 4-hydroxy-2-hexenal, 4-hydroxy nonenal, malondialdehyde, or methylglyoxal modified proteins. The sham cohort served as a negative control to verify that at the recommended dilution of the antibody was able to detect differences in acrolein-adduct concentrations with adequate resolution. Acrolein-adduct quantification was performed with investigators blinded to group allocation.

2.5. Statistical analysis

We conducted a prospective power analysis for the comparison of acute and ABAH treated mice based on prior studies (Forghani et al., 2012), which yielded a minimum sample size of n = 3 per group to achieve 90% power to detect a difference in means of 50% with a standard deviation of 20% at the 0.05 significance level using the two-sided t-test. Data were analyzed using GraphPad Prism (La Jolla, Calif). Results were expressed as mean ± standard error of the mean (SEM). Student’s t-test, simple or Brown-Forsythe and Welch’s one-way ANOVA was used for normally distributed data, and the Mann-Whitney U test for non-normally distributed data. Bonferroni or Dunnett’s correction was used to adjust p-values for multiple comparisons. The ROUT test was used to identify outliers, and two outliers were identified when the acute MOG-EAE cohort was analyzed separately (Fig. S3A, B). These data points were retained in the pooled analysis; no other outliers were found. Correlations were assessed with Pearson’s r and R2. A p-value < 0.05 was considered to indicate statistical significance.

3. Results

3.1. heMAMP- MRI vs. [18F]MAPP PET in the brains of SJL-PLP-EAE mice

MPO-targeted MRI and PET revealed numerous MPO+ parenchymal lesions as well as compartmentalized leptomeningeal enhancement (Fig. 1A). In acute SJL-PLP-EAE, heMAMP+ lesion volume was significantly larger (4.09 ± 0.79 mm3 vs. 0.41 ± 0.14 mm3, p < 0.001) and mean CNR60min significantly increased (6.68 ± 0.69 vs 3.04 ± 0.23, p = 0.004) compared to remission, indicating higher MPO activity (Fig. 1A–B). Interestingly, relapsing mice did not show significant differences in lesion volume, number, or CNR compared to remission, suggesting that there is active disease even during clinical “remission”.

Fig. 1.

Fig. 1.

Open in a new tab

MPO-targeted MRI and PET of SJL PLP-EAE mice at three stages of EAE. (A) 60 min after i.v. administration, acutely affected mice showed heMAMP enhancement in parenchymal lesions in the cerebrum and the cerebellum (red arrows), as well as compartmentalized leptomeningeal enhancement (white arrows). [18F]-F-MAPP-PET of the same mice, performed immediately after MRI, showed increased uptake in the corresponding regions. (B) CNR analysis of heMAMP+ lesions. (C) SUVR analysis of [18F]MAPP+ lesions. (D) Lesion volumes of heMAMP+ MRI and [18F]MAPP-PET+ lesions across the different stages of EAE. [18F]MAPP-PET detected significantly larger lesion volumes in all stages of EAE compared to heMAMP-enhanced MRI (unpaired comparison). PET Scale given as SUVR. Acute n = 7 (n = 5 heMAMP and n = 6 [18F]MAPP), remission n = 10 (n = 10 heMAMP and n = 4 [18F]MAPP), relapse n = 5, sham n = 3. Data was given as mean ± SEM, ** p < 0.01, **** p < 0.0001. Animals received both imaging agents unless otherwise indicated.

[18F]MAPP is a PET imaging agent that reports MPO activity, but unlike heMAMP, can cross the BBB (Wang et al., 2019). Compared to heMAMP-enhanced MRI, PET imaging revealed [18F]MAPP uptake extending beyond the regions identified by MRI (Fig. 1A). Similarly, [18F]MAPP+ lesion volume during acute disease exceeded that during remission (21.72 ± 6.2 mm3 vs. 4.74 ± 2.3 mm3, p = 0.03) with a significantly higher mean lesion SUVR (7.9 ± 0.52 vs. 1.9 ± 0.26, p < 0.001) (Fig. 1C). In relapsing mice [18F]MAPP uptake was less pronounced, although focal increases were detectable. In comparison, heMAMP-MRI was able to detect smaller individual lesions (mean individual lesion volume MRI vs PET; acute SJL: 0.35 ± 0.037 mm3 vs 4.11 ± 0.56 mm3, p < 0.001), whereas [18F]MAPP-PET identified significantly larger total lesion volumes per mouse across all disease stages: acute (21.72 ± 6.27 vs 3.10 ± 0.39 mm3, p = 0.004), remission (4.74 ± 2.29 vs 0.85 ± 0.27 mm3, p = 0.01) and relapse (10.29 ± 4.17 vs 1.9 ± 0.32 mm3, p = 0.035) (Fig. 1D).

3.2. heMAMP- MRI vs. [18F]MAPP PET in the spinal cords of C57BL/6-MOG-EAE mice

During acute disease, total heMAMP+ lesion volume was tenfold larger compared to the chronic stage (1.17 ± 0.11 mm3 vs. 0.11 ± 0.11 mm3, p = 0.01) (Fig. 2A) with a more than twofold higher mean lesion CNR60min (3.4 ± 0.6 vs 1.4 ± 0.6, p = 0.02) (Fig. 2B). On [18F]MAPP-PET, the average lesion SUVR was also significantly higher during acute relative to chronic stages (2.82 ± 0.7 vs. 1.84 ± 0.45, p = 0.002) (Fig. 2C). In unpaired comparison to heMAMP MRI, [18F] MAPP PET detected approximately twice as many lesions in acute EAE (2.16 vs 4.16, p = 0.11) and much larger total lesion volume at both acute (PET: 42.44 ± 6.55 mm3 vs MRI: 1.17 ± 0.11 mm3, p = 0.001) and chronic stages (PET: 8.37 ± 1.9 mm3 vs MRI: 0.11 ± 0.11 mm3, p = 0.001) (Fig. 2D).

Fig. 2.

Fig. 2.

Open in a new tab

MPO-targeted MRI and PET-CT of C57BL/6 mice with acute and chronic EAE. (A) In acute EAE, T1-weighted MRI after i.v. injection of heMAMP revealed focal enhancement in the thoracic and faint enhancement of the lumbar spinal cord of C57BL/6 mice. During chronic disease, markedly fewer lesions were detectable with MPO-targeted MRI. [18F]MAPP-PET revealed enhancement of the spinal cord with accentuation of the thoracic and lumbar portion. (B) 60-minute delayed CNR and (C) SUVR of representative lesions decreased significantly between acute and chronic EAE. (D) Significantly more lesions were identified on [18F]MAPP-PET (unpaired comparison). PET Scale given as SUVR. Acute n = 8 (n = 8 heMAMP and n = 6 [18F]MAPP), chronic n = 6 (n = 5 heMAMP and n = 5 [18F]MAPP), sham n = 3. Data shown as mean ± SEM, * p < 0.05, ** p < 0.01. Animals received both imaging agents unless otherwise indicated.

3.3. Ex vivo analyses confirmed dynamic changes in MPO and acrolein-adducts in the CNS of EAE mice

In SJL-PLP-EAE mice, brain MPO protein concentration was 16-fold higher during acute disease than sham (8.34 ± 0.95 ng/ml vs. 0.55 ± 0.046 ng/ml, p < 0.001), dropped during remission (1.89 ± 0.14 ng/ml vs. 8.34 ± 0.95 ng/ml, p < 0.001), and increased at relapse (3.26 ± 0.31 ng/ml vs. 1.89 ± 0.14, p = 0.007) (Fig. 3A). Brain MPO activity followed a similar pattern, decreasing sharply in remission (22.91 ± 3.9 RFU/s vs 1.59 ± 0.1 RFU/s, p < 0.001) and did not increase significantly during relapse (0.63 ± 0.19 RFU/s vs. 0.22 ± 0.09 RFU/s, p = 0.32) (Fig. 3A), similarly to the in vivo imaging results (Fig. 1). Acrolein-adduct levels were approximately threefold higher in the brains of acute SJL-PLP-EAE mice compared to sham-induced mice (431 ± 62.28 vs. 145.7 ± 9.77, p < 0.001). Interestingly, acrolein-adduct levels remained significantly elevated during both remission (381.7 ± 26.73 vs. 145.7 ± 9.77, p = 0.006) and relapse (497.9 ± 44.03 vs. 145.7 ± 9.77, p < 0.001) (Fig. 3B).

Fig. 3.

Fig. 3.

Open in a new tab

Ex vivo MPO and acrolein-adducts of EAE-mice at different disease stages. (A) MPO protein was significantly increased in all disease stages in SJL mice, most pronounced during the acute disease stage when compared to sham induced mice, as was MPO activity, which decreased to levels not significantly different from sham during remission. Acrolein-adducts in the tissue were elevated above sham level in all disease stages, as detected by western blot (B). (C) In C57BL/6 EAE spinal cords, MPO protein, MPO activity, and acrolein-adduct levels detected by western blot (D) decreased sharply after the acute stage and were not significantly elevated above sham in chronic mice. Representative acrolein-adduct bands are shown. Representative Coomassie bands were converted to 8-bit black-and-white images for better visualization. Acrolein-adduct levels were normalized to total protein loading control assessed by Coomassie Brilliant Blue staining. Data shown as mean ± SEM. SJL acute n = 8, relapse and sham n = 10, remission n = 7; C57BL/6 sham and acute n = 10, chronic n = 7. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

In C57BL/6-MOG-EAE spinal cords, MPO protein was over tenfold higher in acute versus both chronic and sham conditions (5.7 ± 0.49 ng/ml vs 0.68 ± 0.1 ng/ml, p < 0.001 and 0.43 ± 0.04 ng/ml, p < 0.001, respectively) (Fig. 3C). MPO activity likewise decreased significantly from the acute to the chronic stage (5.42 ± 0.92 RFU/s vs 0.27 ± 0.17 RFU/s, p = 0.033) (Fig. 3C). Acrolein-adducts decreased more than twofold (1.070 ± 65.30 vs. 365.6 ± 49.29, p < 0.001) post-acute stage and were not significantly elevated during chronic disease (Fig. 3D).

3.4. Ex vivo correlation of MPO and acrolein-adducts in EAE mice

In SJL-PLP-EAE brains, acrolein-adduct levels correlated moderately with MPO protein (r = 0.70, p = 0.053) (Fig. 4A) and strongly with MPO activity when inspecting the acute stage, during which ex vivo MPO activity was significantly increased (r = 0.76, p = 0.025) (Fig. 4B). In C57BL/6-MOG-EAE spinal cords, acrolein-adduct levels correlated strongly with MPO protein (r = 0.90, p < 0.001) and activity (r = 0.82, p < 0.001) (Fig. 4C–D).

Fig. 4.

Fig. 4.

Open in a new tab

Correlation of MPO concentration, activity, and relative acrolein-adduct abundance. (A) In the brains of acute SJL mice, there is a trend of positive correlation between MPO protein and acrolein. (B) MPO activity and acrolein-adducts show a positive correlation. In the spinal cords of C57BL/6 mice, MPO protein (C) and MPO activity (D) showed positive correlations with acrolein-adduct levels across chronic and active disease stages. SJL acute n = 8, C57BL/6 sham and acute n = 10, chronic n = 7.

Within-group analyses confirmed strong positive correlation of acrolein-adducts with MPO concentration in acute EAE (r = 0.77, p = 0.026) (Fig. S3A), while the correlation with MPO activity just missed significance (r = 0.69, p = 0.056) (Fig. S3B). Inspecting chronic EAE, MPO concentration (r = 0.88, p = 0.0082) and activity (r = 0.96, p = < 0.001) showed a strong positive correlation with tissue acrolein-adducts (Fig. S3C, D).

Serum acrolein-adduct abundance correlated moderately with ex vivo brain MPO protein and activity in SJL-PLP-EAE mice (r = 0.52, p = 0.005 and r = 0.47, p = 0.01, respectively) (Fig. S4A). However, serum and brain acrolein-adduct levels did not correlate (Fig. S4B), suggesting that the apparent correlation between ex vivo brain MPO and serum acrolein-adducts was likely spurious. Similarly, no such correlation was observed between serum acrolein-adduct abundance and spinal cord MPO protein or activity in C57BL/6-MOG-EAE (Fig S4C–D). Thus, it was not surprising that no consistent trend between imaging metrics and serum acrolein-adducts was observed in SJL-PLP-EAE, and [18F] MAPP+ lesion volume showed a non-significant correlation with serum acrolein-adducts (r = 0.52, p = 0.36) in C57BL/6-MOG-EAE mice. Collectively, these findings suggest that while there is a robust local MPO - acrolein-adduct correlation at the tissue level, serum acrolein-adduct levels do not reflect local tissue acrolein-adduct levels and MPO activity.

3.5. Inhibition of MPO activity after symptom onset does not affect tissue acrolein-adduct formation

Acute SJL-EAE mice treated with the MPO inhibitor ABAH showed unchanged brain acrolein-adduct levels relative to sham-treated mice (431.0 ± 62.28 vs 433.7 ± 75.65, p > 0.99), while serum acrolein-adduct decreased threefold (195.0 ± 11.56 vs 61.33 ± 6.71, p < 0.001). ABAH treatment furthermore led to a twofold decrease in brain MPO protein (8.34 ± 0.95 ng/ml vs 3.87 ± 0.19 ng/ml, p < 0.001) and 20-fold decrease in MPO activity (12.58 ± 2.14 RFU/s vs 0.63 ± 0.04 RFU/s, p = 0.002), but neither correlated with serum acrolein-adducts (r = 0.60, p = 0.28 and r = 0.25, p = 0.67, respectively) (Fig. S5).

4. Discussion

In this study, we demonstrated that MPO and acrolein-adduct levels are similarly elevated during acute disease in two EAE models. MPO activity was reported by both heMAMP-MRI and [18F]MAPP-PET, visualizing disease activity in vivo. Tissue MPO protein and activity correlated with acrolein-adducts ex vivo across different models and stages, supporting MPO as a main source of acrolein in neuroinflammation. However, this relationship was inconsistently reflected by serum acrolein-adduct levels, revealing that serum acrolein-adducts may not be an accurate biomarker for acrolein or MPO at the tissue level. Instead, MPO imaging can provide a noninvasive means to report and track not only MPO activity but also acrolein-adduct levels in the CNS, potentially serving as an imaging biomarker for oxidative stress and neuroinflammation.

In chronic C57BL/6-MOG-EAE, molecular imaging showed a decline in MPO+ lesion volume and MPO activation ratio after the acute stage, and tissue assays confirmed that MPO and acrolein-adduct levels decreased to near-sham levels during the chronic stage. Previous work found that the composition of the cellular infiltrate shifts after the acute stage from predominantly neutrophils, which contain abundant MPO in their azurophilic granules, to mainly activated macrophages and microglia (Caravagna et al., 2018). These highly versatile cells can also secrete MPO when polarized to a proinflammatory state, but during chronic disease, likely shift to an anti-inflammatory phenotype to promote repair and remyelination (Mikita et al., 2011; Miron et al., 2013; Rawji and Yong, 2013). The low MPO activity observed likely reflects this transition toward a reparative phenotype.

In relapsing-remitting SJL-PLP-EAE, ex vivo MPO protein and activity decreased after the acute stage, whereas acrolein-adducts remained elevated. Indeed, acrolein-adducts are known to persist for days, outlasting the brief interval between acute disease and remission in SJL-PLP-EAE. Previous reports have shown that MPO activity and the formation of acrolein-adducts precede clinical manifestation in mouse and rat EAE, respectively (Chen et al., 2008; Tully et al., 2018). Our results indicate that MPO inhibition effectively reduces systemic acrolein-adducts but, once symptoms are established, does not alter CNS acrolein accumulation. It is therefore likely that the bulk of acrolein-adducts have formed by that time and inhibition of MPO only prevents further acrolein formation. The discrepant decrease in serum acrolein-adducts after MPO inhibition suggests that while modified proteins in circulation are still metabolized, clearance is impaired in the diseased CNS (Smerjac and Bizzozero, 2008). Of note, ABAH nearly abolished MPO activity in the brain, but the inhibited MPO remained detectable by immunostaining. In addition, MPO can act as a signaling molecule, and its inhibition leads to decreased recruitment of inflammatory cells, which, in turn, may account for decreased MPO protein (Forghani et al., 2012; Klinke et al., 2011). This highlights the importance of monitoring MPO activity rather than MPO protein when evaluating therapeutic efficacy, as enzymatic activity may be inhibited by endogenous inhibitors such as ceruloplasmin (Chapman et al., 2013).

In MS patients, meningeal inflammation has emerged as a key driver of MS progression (Magliozzi et al., 2023). Furthermore, the degree of meningeal inflammation corresponds to clinical severity in relapsing-remitting EAE (Walker-Caulfield et al., 2015). In this study, MPO-targeted imaging revealed significant MPO activity along the cerebellar meninges of SJL mice in vivo. The meninges harbor a population of CNS macrophages that surveil the underlying subarachnoid space. These cells have been reported to be the earliest stimulator of autoreactive T-cells during EAE initiation (Kivisakk et al., 2009).

The differential sensitivities and resolutions of heMAMP-MRI and [18F]MAPP-PET suggest potentially complementary roles for these imaging modalities in clinical monitoring of multiple sclerosis. [18F]MAPP-PET detected active spinal cord lesions in C57BL/6-MOG-EAE with higher sensitivity than heMAMP-MRI, likely because PET requires only nanomolar probe accumulation and [18F]MAPP can cross the intact BBB (Wang et al., 2019). Together with the wider field of view, [18F]MAPP PET may enable more sensitive detection of the overall burden of active CNS disease. However, the larger total lesion volumes detected by PET should be interpreted with caution, as they may partly reflect method-intrinsic limitations, including lower spatial resolution and spillover effects. Consistent with this, MRI identified a greater number of individual brain lesions in SJL-PLP-EAE, which is diagnostically valuable to pinpoint the locations of active disease. Beyond active lesion detection, both types of MPO imaging may serve as tools for treatment selection in MS patients. Early aggressive therapy has shown promise in preventing long-term disability (He et al., 2020), but identifying patients who benefit most from this approach remains challenging. Serum biomarkers, such as acrolein-adduct abundance, may obscure more dynamic aspects of tissue inflammation that are significant for making sound treatment decisions. Translation of [18F]MAPP is underway, with an exploratory investigational new drug application recently approved by the FDA.

This study has several limitations. While EAE models reproduce key features of MS, they cannot recapitulate the heterogeneous course of human disease. We mitigated this by using two different models across different mouse strains and disease stages. The short relapse-remission cycle of SJL mice may exaggerate the persistence of tissue acrolein-adducts compared to MS patients, in whom disease evolution is slower. However, while tissue acrolein-adduct levels lagged MPO activity changes in the post-acute SJL mouse model due to the fast time course of this model, in MS this is unlikely to be an issue given the longer time course of the human disease. Although MPO correlated with acrolein-adducts, supporting an MPO-dependent generation, the data do not establish MPO as the principal source of acrolein in EAE. Apart from MPO-derived oxidants, the relative contributions of other sources of oxidative stress and the resulting peroxidation of polyunsaturated fatty acids to acrolein-adduct formation cannot be determined from the present study (Stevens and Maier, 2008; Uchida et al., 1998). ABAH-treated animals were used exclusively for ex vivo analyses and did not undergo MPO-targeted PET or MRI. Consequently, pharmacologic responsiveness of the imaging signals was not independently demonstrated in the present experimental cohort. However, previous studies have demonstrated reduced MPO-targeted MRI signal following ABAH-mediated MPO inhibition in EAE (Forghani et al., 2012; Li et al., 2019) and reduced [18F] MAPP PET uptake following pharmacologic MPO inhibition in experimental myocardial infarction (Zeller et al., 2023). While sex-differences in EAE induced in the C57BL/6-strain are less pronounced, female SJL/J mice develop relapsing–remitting EAE course more consistently than age-matched males, enabling investigation of the acute, remission, and relapse phases within a reproducible model (Papenfuss et al., 2004). However, as EAE susceptibility, immunopathology, and disease course are influenced by biological sex, the present findings cannot be directly generalized to male mice, and future studies incorporating both sexes will be required to determine whether the observed imaging and biochemical relationships are sex-dependent.

5. Conclusion

In conclusion, our results demonstrate that while ex vivo tissue concentrations of MPO and acrolein-adducts, indicative of oxidative stress, correlate, serum acrolein-adduct levels did not. Therefore, assessing MPO activity in vivo might more adequately reflect inflammatory activity, providing a rationale for future studies to use MPO imaging to understand how anti-oxidative stress/inflammation therapies, such as scavenging acrolein (Leung et al., 2011; Spaas et al., 2021; Tang et al., 2024), can affect the tissue and inflammatory environment and to use MPO imaging as an imaging biomarker to study and monitor oxidative damage and treatment changes.

Supplementary Material

1

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.neuroimage.2026.122217.

Acknowledgements

The work presented in this manuscript was supported by the National Multiple Sclerosis Society (RG-1507–05486) and the National Institutes of Health (R01AG075055). EK received a scholarship by the German Academic Scholarship Foundation.

Footnotes

Declaration of competing interest

The work presented in this manuscript was supported by the National Multiple Sclerosis Society (RG-1507–05486). Dr. Chen and Dr. Wang are inventors of a patent application based on the technology reported in this manuscript. Dr. Chen has a financial interest in Silverier, a parent company for biotech-focused start-up companies that is currently focused on translating PET and MRI imaging agents that target myeloperoxidase. Dr. Chen and Dr. Wang have a financial interest in Einsenca, a company focused on translating PET and MRI imaging agents targeting myeloperoxidase. Dr. Chen’s and Dr. Wang’s interests were reviewed and are managed by MGH and Mass General Brigham in accordance with their conflict-of-interest policies. Dr. Shi is a co-founder of Neuro Vigor, a startup company specializing in therapeutic developments for central nervous system trauma and neurodegenerative disorders. Other authors have reported that they have no relationship relevant to the contents of this paper to disclose.

CRediT authorship contribution statement

Enrico Küllenberg: Writing – review & editing, Writing – original draft, Visualization, Validation, Project administration, Investigation, Formal analysis, Data curation. Siyuan Sun: Writing – review & editing, Investigation, Formal analysis. Cuihua Wang: Writing – review & editing, Resources. Negin Jalali Motlagh: Writing – review & editing, Investigation. Mariane Le Fur: Writing – review & editing, Resources. Gregory R. Wojtkiewicz: Writing – review & editing, Investigation. Seth A. Herr: Writing – review & editing, Investigation, Formal analysis. Riyi Shi: Writing – review & editing, Validation, Supervision, Project administration, Funding acquisition, Conceptualization. John W. Chen: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Project administration, Funding acquisition, Formal analysis, Data curation, Conceptualization.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

References

  1. Adamczyk B, Adamczyk-Sowa M, 2016. New insights into the role of oxidative stress mechanisms in the pathophysiology and treatment of multiple sclerosis. Oxid. Med. Cell. Longev. 2016, 1973834. 10.1155/2016/1973834. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Anderson MM, Hazen SL, Hsu FF, Heinecke JW, 1997. Human neutrophils employ the myeloperoxidase-hydrogen peroxide-chloride system to convert hydroxy-amino acids into glycolaldehyde, 2-hydroxypropanal, and acrolein. A mechanism for the generation of highly reactive alpha-hydroxy and alpha,beta-unsaturated aldehydes by phagocytes at sites of inflammation. J. Clin. Invest. 99, 424–432. 10.1172/JCI119176. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Breckwoldt MO, Chen JW, Stangenberg L, Aikawa E, Rodriguez E, Qiu S, Moskowitz MA, Weissleder R, 2008. Tracking the inflammatory response in stroke in vivo by sensing the enzyme myeloperoxidase. Proc. Natl. Acad. Sci. 105, 18584–18589. 10.1073/pnas.0803945105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Caravagna C, Jaouën A, Desplat-Jégo S, Fenrich KK, Bergot E, Luche H, Grenot P, Rougon G, Malissen M, Debarbieux F, 2018. Diversity of innate immune cell subsets across spatial and temporal scales in an EAE mouse model. Sci. Rep. 8, 5146. 10.1038/s41598-018-22872-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Chapman AP, Mocatta TJ, Shiva S, Seidel A, Chen B, Khalilova I, Paumann-Page ME, Jameson GNL, Winterbourn CC, Kettle AJ, 2013. Ceruloplasmin is an endogenous inhibitor of myeloperoxidase. J. Biol. Chem. 288, 6465–6477. 10.1074/jbc.M112.418970. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chen JW, Breckwoldt MO, Aikawa E, Chiang G, Weissleder R, 2008. Myeloperoxidase-targeted imaging of active inflammatory lesions in murine experimental autoimmune encephalomyelitis. Brain 131, 1123–1133. 10.1093/brain/awn004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Forghani R, Wojtkiewicz GR, Zhang Y, Seeburg D, Bautz BRM, Pulli B, Milewski AR, Atkinson WL, Iwamoto Y, Zhang ER, Etzrodt M, Rodriguez E, Robbins CS, Swirski FK, Weissleder R, Chen JW, 2012. Demyelinating diseases: myeloperoxidase as an imaging biomarker and therapeutic target. Radiology 263, 451–460. 10.1148/radiol.12111593. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Gray E, Thomas TL, Betmouni S, Scolding N, Love S, 2008. Elevated activity and microglial expression of myeloperoxidase in demyelinated cerebral cortex in multiple sclerosis. Brain Pathol. 18, 86–95. 10.1111/j.1750-3639.2007.00110.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Hamann K, Nehrt G, Ouyang H, Duerstock B, Shi R, 2008. Hydralazine inhibits compression and acrolein-mediated injuries in ex vivo spinal cord. J. Neurochem. 104, 708–718. 10.1111/j.1471-4159.2007.05002.x. [DOI] [PubMed] [Google Scholar]
  10. He A, Merkel B, Brown JWL, Zhovits Ryerson L, Kister I, Malpas CB, Sharmin S, Horakova D, Kubala Havrdova E, Spelman T, Izquierdo G, Eichau S, Trojano M, Lugaresi A, Hupperts R, Sola P, Ferraro D, Lycke J, Grand’Maison F, Prat A, Girard M, Duquette P, Larochelle C, Svenningsson A, Petersen T, Grammond P, Granella F, Van Pesch V, Bergamaschi R, McGuigan C, Coles A, Hillert J, Piehl F, Butzkueven H, Kalincik T, 2020. Timing of high-efficacy therapy for multiple sclerosis: a retrospective observational cohort study. Lancet Neurol. 19, 307–316. 10.1016/S1474-4422(20)30067-3. [DOI] [PubMed] [Google Scholar]
  11. Kivisakk P, Imitola J, Rasmussen S, Elyaman W, Zhu B, Ransohoff RM, Khoury SJ, 2009. Localizing central nervous system immune surveillance: meningeal antigen-presenting cells activate T cells during experimental autoimmune encephalomyelitis. Ann. Neurol. 65, 457–469. 10.1002/ana.21379. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Klinke A, Nussbaum C, Kubala L, Friedrichs K, Rudolph TK, Rudolph V, Paust H-J, Schröder C, Benten D, Lau D, Szocs K, Furtmüller PG, Heeringa P, Sydow K, Duchstein H-J, Ehmke H, Schumacher U, Meinertz T, Sperandio M, Baldus S, 2011. Myeloperoxidase attracts neutrophils by physical forces. Blood 117, 1350–1358. 10.1182/blood-2010-05-284513. [DOI] [PubMed] [Google Scholar]
  13. Leung G, Sun W, Zheng L, Brookes S, Tully M, Shi R, 2011. Anti-acrolein treatment improves behavioral outcome and alleviates myelin damage in experimental autoimmune enchephalomyelitis mouse. Neuroscience 173, 150–155. 10.1016/j.neuroscience.2010.11.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Li A, Wu Y, Pulli B, Wojtkiewicz GR, Iwamoto Y, Wang C, Li J-H, Ali M, Feng X, Yao Z, Chen JW, 2019. Myeloperoxidase molecular MRI reveals synergistic combination therapy in murine experimental autoimmune neuroinflammation. Radiology 293, 158–165. 10.1148/radiol.2019182492. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Luo J, Shi R, 2004. Acrolein induces axolemmal disruption, oxidative stress, and mitochondrial impairment in spinal cord tissue. Neurochem. Int. 44, 475–486. 10.1016/j.neuint.2003.09.006. [DOI] [PubMed] [Google Scholar]
  16. Luo J, Shi R, 2005. Acrolein induces oxidative stress in brain mitochondria. Neurochem. Int. 46, 243–252. 10.1016/j.neuint.2004.09.001. [DOI] [PubMed] [Google Scholar]
  17. Magliozzi R, Howell OW, Calabrese M, Reynolds R, 2023. Meningeal inflammation as a driver of cortical grey matter pathology and clinical progression in multiple sclerosis. Nat. Rev. Neurol. 19, 461–476. 10.1038/s41582-023-00838-7. [DOI] [PubMed] [Google Scholar]
  18. Mikita J, Dubourdieu-Cassagno N, Deloire MS, Vekris A, Biran M, Raffard G, Brochet B, Canron M-H, Franconi J-M, Boiziau C, Petry KG, 2011. Altered M1/M2 activation patterns of monocytes in severe relapsing experimental rat model of multiple sclerosis. Amelioration of clinical status by M2 activated monocyte administration. Mult. Scler. J. 17, 2–15. 10.1177/1352458510379243. [DOI] [PubMed] [Google Scholar]
  19. Miron VE, Boyd A, Zhao J-W, Yuen TJ, Ruckh JM, Shadrach JL, Van Wijngaarden P, Wagers AJ, Williams A, Franklin RJM, ffrench-Constant C, 2013. M2 microglia and macrophages drive oligodendrocyte differentiation during CNS remyelination. Nat. Neurosci. 16, 1211–1218. 10.1038/nn.3469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Nagra RM, Becher B, Tourtellotte WW, Antel JP, Gold D, Paladino T, Smith RA, Nelson JR, Reynolds WF, 1997. Immunohistochemical and genetic evidence of myeloperoxidase involvement in multiple sclerosis. J. Neuroimmunol. 78, 97–107. 10.1016/S0165-5728(97)00089-1. [DOI] [PubMed] [Google Scholar]
  21. Ong FHC, Henry PJ, Burcham PC, 2012. Prior exposure to acrolein accelerates pulmonary inflammation in influenza a-infected mice. Toxicol. Lett. 212, 241–251. 10.1016/j.toxlet.2012.06.003. [DOI] [PubMed] [Google Scholar]
  22. Papenfuss TL, Rogers CJ, Gienapp I, Yurrita M, McClain M, Damico N, Valo J, Song F, Whitacre CC, 2004. Sex differences in experimental autoimmune encephalomyelitis in multiple murine strains. J. Neuroimmunol. 150, 59–69. 10.1016/j.jneuroim.2004.01.018. [DOI] [PubMed] [Google Scholar]
  23. Park J, Zheng L, Marquis A, Walls M, Duerstock B, Pond A, Vega-Alvarez S, Wang H, Ouyang Z, Shi R, 2014. Neuroprotective role of hydralazine in rat spinal cord injury-attenuation of acrolein-mediated damage. J. Neurochem. 129, 339–349. 10.1111/jnc.12628. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Pulli B, Ali M, Forghani R, Schob S, Hsieh KLC, Wojtkiewicz G, 2013. Measuring myeloperoxidase activity in biological samples. Johnson R. PLoS One. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Pulli B, Wojtkiewicz G, Iwamoto Y, Ali M, Zeller MW, Bure L, Wang C, Choi Y, Masia R, Guimaraes AR, Corey KE, Chen JW, 2017. Molecular MR imaging of myeloperoxidase distinguishes steatosis from steatohepatitis in nonalcoholic fatty liver disease. Radiology 284, 390–400. 10.1148/radiol.2017160588. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Rawji KS, Yong VW, 2013. The benefits and detriments of macrophages/microglia in models of multiple sclerosis. Clin. Dev. Immunol. 2013, 1–13. 10.1155/2013/948976. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Shi Y, Sun W, McBride JJ, Cheng J-X, Shi R, 2011. Acrolein induces myelin damage in mammalian spinal cord: acrolein-mediated myelin damage. J. Neurochem. 117, 554–564. 10.1111/j.1471-4159.2011.07226.x. [DOI] [PubMed] [Google Scholar]
  28. Smerjac SM, Bizzozero OA, 2008. Cytoskeletal protein carbonylation and degradation in experimental autoimmune encephalomyelitis. J. Neurochem. 105, 763–772. 10.1111/j.1471-4159.2007.05178.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Spaas J, Franssen WMA, Keytsman C, Blancquaert L, Vanmierlo T, Bogie J, Broux B, Hellings N, Van Horssen J, Posa DK, Hoetker D, Baba SP, Derave W, Eijnde BO, 2021. Carnosine quenches the reactive carbonyl acrolein in the central nervous system and attenuates autoimmune neuroinflammation. J. Neuroinflammation 18, 255. 10.1186/s12974-021-02306-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Stevens JF, Maier CS, 2008. Acrolein: sources, metabolism, and biomolecular interactions relevant to human health and disease. Mol. Nutr. Food Res. 52, 7–25. 10.1002/mnfr.200700412. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Tang J, Alford A, Leung G, Tully M, Shi R, 2024. Neuroprotection by acrolein sequestration through exogenously applied scavengers and endogenous enzymatic enabling strategies in mouse EAE model. Sci. Rep. 14, 1–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Tully M, Tang J, Zheng L, Acosta G, Tian R, Hayward L, Race N, Mattson D, Shi R, 2018. Systemic acrolein elevations in mice with experimental autoimmune encephalomyelitis and patients with multiple sclerosis. Front. Neurol. 9, 420. 10.3389/fneur.2018.00420. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Uchida K, Kanematsu M, Morimitsu Y, Osawa T, Noguchi N, Niki E, 1998. Acrolein is a product of lipid peroxidation reaction. J. Biol. Chem. 273, 16058–16066. 10.1074/jbc.273.26.16058. [DOI] [PubMed] [Google Scholar]
  34. Walker-Caulfield Margaret E., Hatfield Julianne K., Brown Melissa A., 2015. Dynamic changes in meningeal inflammation correspond to clinical exacerbations in a murine model of relapsing–remitting multiple sclerosis. Journal of Neuroimmunology 278, 112–122. 10.1016/j.jneuroim.2014.12.009. https://linkinghub.elsevier.com/retrieve/pii/S016557281400993X. [DOI] [PubMed] [Google Scholar]
  35. Wang C, Cheng D, Jalali Motlagh N, Kuellenberg EG, Wojtkiewicz GR, Schmidt SP, Stocker R, Chen JW, 2021. Highly efficient activatable MRI probe to sense myeloperoxidase activity. J. Med. Chem. 64, 5874–5885. 10.1021/acs.jmedchem.1c00038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Wang C, Keliher E, Zeller MWG, Wojtkiewicz GR, Aguirre AD, Buckbinder L, Kim H-Y, Chen J, Maresca K, Ahmed MS, Motlagh NJ, Nahrendorf M, Chen JW, 2019. An activatable PET imaging radioprobe is a dynamic reporter of myeloperoxidase activity in vivo. Proc. Natl. Acad. Sci. 116, 11966–11971. 10.1073/pnas.1818434116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Zeller MWG, Wang C, Keliher EJ, Wojtkiewicz GR, Aguirre A, Maresca K, Su C, Buckbinder L, Wang J, Nahrendorf M, Chen JW, 2023. Myeloperoxidase PET imaging tracks intracellular and extracellular treatment changes in experimental myocardial infarction. Int. J. Mol. Sci. 24, 5704. 10.3390/ijms24065704. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Zhang H, Ray A, Miller NM, Hartwig D, Pritchard KA, Dittel BN, 2016. Inhibition of myeloperoxidase at the peak of experimental autoimmune encephalomyelitis restores blood–brain barrier integrity and ameliorates disease severity. J. Neurochem. 136, 826–836. 10.1111/jnc.13426. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2213707-supplement-1.docx (10.3MB, docx)

Data Availability Statement

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

RESOURCES