Myeloperoxidase-targeted imaging of active
inflammatory lesions in murine experimental
John W.Chen,1,2Michael O. Breckwoldt,2Elena Aikawa,2Gloria Chiang2and Ralph Weissleder1,2
1Center for Systems Biology and
Harvard Medical School, 5404 Building149,13th Street,Charlestown, MA 02129,USA
2Center for Molecular Imaging Research, Massachusetts General Hospital,
Correspondence to: John W.Chen, 5404 Building149,13th Street,Charlestown, MA 02129,USA
Inflammatory demyelinating plaques are the pathologic hallmark of active multiple sclerosis and often precede
clinical manifestations. Non-invasive early detection of active plaques would thus be crucial in establishing pre-
symptomatic diagnosis and could lead to early preventive treatment strategies. Using murine experimental
autoimmune encephalomyelitis as a modelof multiple sclerosis, we demonstrate that a prototype paramagnetic
myeloperoxidase (MPO) sensor can detect and confirm more, smaller, and earlier active inflammatory lesions
in living mice by in vivo MRI. We show that MPO expression corresponded with areas of inflammatory cell
infiltration and demyelination, and higher MPO activity as detected by MPO imaging, biochemical assays, and
histopathological analyses correlated with increased clinical disease severity. Our findings present a potential
new translational approach for specific non-invasive inflammatory plaque imaging.This approach could be used
in longitudinalstudiestoidentify active demyelinating plaques aswellas to more accurately trackdisease course
following treatment in clinical trials.
Keywords: myeloperoxidase; neuroinflammation; demyelination; targeted imaging; MRI
Abbreviations: MPO=myeloperoxidase; DTPA(Gd)=diethylenetriamine-pentaacetate gadolinium;
5-HT-DTPA(Gd)=bis-5-hydroxytryptamide-diethylenetriamine-pentaacetate gadolinium; EAE=experimental
autoimmune encephalomyelitis; BBB=blood^brain barrier; ROI=region of interest; PLP=proteolipid protein;
MS=multiple sclerosis; R1=longitudinal relaxivity; PBS=phosphate-buffered saline; FITC=fluorescein isothiocyanate;
CNR=contrast-to-noise ratio; SD=standard deviation; SEM=standard error of measurement; PET=positron emission
tomography; SPECT=single photon emission computed tomography
Received September 3, 2007 . Revised January 4, 2008. Accepted January 7 , 2008
Multiple sclerosis (MS), the leading cause of non-traumatic
neurological disability in young adults (Noseworthy et al.,
2000), is characterized by the formation of demyelinating
plaques in the CNS from an immune-mediated inflamma-
tory response induced by lymphocytes, macrophages and
microglia (Noseworthy et al., 2000; Bruck, 2005; Frohman
et al., 2006; Imitola et al., 2006). In the active stage of the
disease, macrophages/microglia are abundant in perivascu-
lar locations (Noseworthy et al., 2000). Inflammatory
plaques form early in MS and often precede clinical symp-
toms (Frohman et al., 2003; Miller et al., 2005). Early
diagnosis with prompt treatment has been found to delay
relapse (Jacobs et al., 2000) and decrease axonal loss
from inflammation (Bruck, 2005; Kutzelnigg et al., 2005;
Frohman et al., 2006). Currently, active disease is inferred
from contrast enhancement identified on MRI. However,
it is widely recognized that these MRI techniques have
limitations because contrast enhancement reflects break-
down in the blood–brain barrier (BBB) with leakage of
paramagnetic chelates rather than active inflammation,
and the two may not always correspond. In particular,
MS lesions at all stages demonstrate some BBB breakdown
(Cotton et al., 2003), and lesions can remain enhanced 1 to
13 weeks after the onset of clinical symptoms (Bruck et al.,
1997; Cotton et al., 2003). Furthermore, to be detectable
by contrast enhanced MRI, a lesion needs to have signifi-
cant breakdown of the BBB to allow micromolar accumula-
tion of conventional gadolinium (Gd) agents. However,
early demyelinating lesions may only have subtle BBB
doi:10.1093/brain/awn004Brain (2008) Page1of11
? The Author (2008).Publishedby Oxford University Pressonbehalfofthe Guarantorsof Brain. Allrightsreserved.For Permissions, please email: email@example.com
Brain Advance Access published January 29, 2008
impairment, insufficient for a large amount of Gd agents to
extravasate. As such, contrast-enhanced MRI has been
found to underreport active MS lesions (Filippi et al., 1996;
Kidd et al., 1999) and correlate poorly with immunological
markers (Giovannoni et al., 2000). Therefore, a non-
invasive method to detect and confirm acute inflammatory
plaques in patients suspected of MS would allow prompt
diagnosis and preventive treatment before irreversible
Myeloperoxidase (MPO), one of the most abundant
enzymes secreted by inflammatory cells (Bradley et al.,
1982) including neutrophils, macrophages and microglia, is
found in active MS plaques (Nagra et al., 1997). Individuals
with higher MPO expression have increased susceptibility
to MS (Chataway et al., 1999; Zakrzewska-Pniewska et al.,
2004). MPO generates highly reactive molecular moieties,
such as hypochlorite, tyrosyl radicals and aldehydes and can
cause local damage and further activate the inflammatory
cascade (Heinecke, 1997). MPO would thus be a good
imaging target for the detection and confirmation of active
inflammation in MS.
In the present study, we use a prototype MPO-activatable
paramagnetic sensor (Chen et al., 2006) to show that MPO-
targeted MR imaging can increase our ability to detect more
and confirm smaller and earlier active demyelinating lesions.
When converted by MPO in the presence of hydrogen per-
oxide (e.g. from NADPH oxidase), the sensor is radicalized
and forms oligomers of higher longitudinal relaxivity (R1),
and in addition can covalently bind to proteins, again
accompanied by R1increases (Fig. 1). These MPO-induced
chemical changes result in markedly increased MR signal on
T1-weighted MRI sequence and improved pharmacokinetics
because converted products are locally retained.
The protocol for animal experiments was approved by the
institutional animal care committee. SJL female mice 6–10 weeks
of age were obtained from Jackson Laboratories (Bar Harbor,
ME). A total of 40 SJL mice were used for this study. Five mice
were used as controls with sham induction by injecting saline
instead of proteolipid protein. EAE in the remaining 35 mice was
induced with synthetic proteolipid protein (PLP139–151, Axxora,
CA) according to Greer et al. (1996). Briefly, 2mg of PLP and
8mg M. tuberculosis H37Ra (Difco, MI) were dissolved in 1ml
H2O, and then combined with 1ml of complete Freud’s adjuvant
(Sigma-Aldrich, MO), and transferred to a 3ml glass syringe
connected to another 3ml glass syringe via a 20-G needle
connector. The content was emulsified by transferring between
the two syringes while on ice. Each mouse received 100ml of the
PLP emulsion (25ml each in the bilateral inguinal and axillary
regions). On days 0 and 2, 0.1mg of pertussinogen dissolved
in 200ml of PBS was injected intravenously via the tail vein.
The animals were monitored at least daily with the following
clinical grading: 0=normal, 1=complete tail limpness with no
limb weakness, 2=hind limb weakness but no obvious paralysis
on ambulation, 3=partial hind limb paralysis, 4=complete hind
limb paralysis, 5=moribund.
The chemicals were purchased from Sigma-Aldrich (St Louis,
MO). The MPO-sensitive imaging agent bis-5-hydroxytryptamide-
diethylenetriamine-pentaacetate gadolinium (bis-5HT-DTPA(Gd))
was synthesized according to Querol et al. (2005) Briefly, DTPA-
bisanhydride was reacted with serotonin in dimethylformamide in
the presence of an excess of triethylamine. The product bis-5HT-
DTPA was isolated by recrystallization from methanol and
acetone. Complexation with gadolinium was performed in the
presence of 1% citric acid (w/w), and purified by high perfor-
purchased from Berlex Laboratories (Berlex, NJ).
Six C57BL/6 mice were injected with 100mCi of
DTPA. Blood was extracted at 1, 10, 30, 60, 240, 480 and
1400min after injection and radioactivity measured on a 1480
Wizard gamma counter (Perkin-Elmer, MA).
Fig.1 Structure of the MPO agent and mechanism of action. In the presence of MPO, the agent is radicalized and forms higher
relaxivity oligomers, which can also bind to proteins.
Page 2 of11Brain (2008)J.W.Chen et al.
Fresh-frozen sections of brains were prepared by embedding the
tissue in OCT and snap-frozen with isopentane on dry ice.
Demyelination of axons was detected with the luxol fast blue stain,
which binds to phospholipids of myelin to result in a blue
staining. Five micrometres fresh-frozen sections were incubated
with 0.1% of luxol fast blue solution at room temperature
overnight and differentiated in lithium carbonate and 70% ethyl
alcohol. Quantification of demyelination was performed by tracing
the areas resistant to the luxol fast blue stain and dividing the
resultant area by the total white matter area. The tracing was
performed using the software OsiriX (version 2.7.5, www.osirix-
Five micrometres sections of fresh-frozen tissues were examined
for the presence of myeloperoxidase (rabbit polyclonal antibody;
AbCam, MA) and macrophages/microglia (mac-3, BD Biosciences,
CA). The avidin–biotin peroxidase method was employed. The
reaction was visualized with 3-amino-9-athyl-carbazol substrate
(AEC, Sigma Chemical, Mo). Tissue sections from healthy animals
were used as controls. Haematoxylin–eosin staining was also
performed to study the overall morphology. All sections were also
counterstained with haematoxylin. Images were captured with
a digital camera (Nikon DXM 1200-F, Nikon Inc., NY).
Double immunofluorescence confocalmicroscopy
To show co-localization of MPO and macrophages/microglia,
we performed dual channel fluorescence confocal microscopy.
The same antibodies were used as for immunohistochemistry.
Secondary antibodies were detected with streptavidin conjugated
with Texas Red (MPO) and streptavidin coupled to FITC (Mac-3)
(both 1:100, Amersham, NJ) and an avidin/biotin blocking kit
(Vector Laboratories, CA) to prevent cross-reaction of the anti-
bodies. A Nikon 80i microscope and an Axiovert 200M inverted
confocal microscope (Carl Zeiss, NY) equipped with an LSM
Pascal Vario RGB Laser (Arg 458/488/514nm, HeNE 543nm,
HeNe 633nm) were used. Summation of projection of all
background-corrected slices was produced using the LSM 5
Pascal Software (v 3.2WS). Final images were colour-coded
green for FITC and red for Texas Red.
Western blot analysis
To confirm the presence of MPO in the brains, mice were
sacrificed, brains homogenized and proteins extracted in 1%
cetyltrimethylammonium bromide (Sigma-Aldrich, MO) in PBS
w/v. The resultant suspension was sonicated for 30s and then
underwentthreecycles of freeze-thaw
Subsequently, the suspensions were centrifuged at 14000rpm
for 15min, and the supernatant used for protein analysis
(Bicinchoninic acid kit, Sigma-Aldrich, MO). The blots were
performed using a monoclonal rabbit anti-MPO (Upstate, CA),
1:1000 dilution, and a rabbit polyclonal ß-Actin antibody (Abcam,
MA), 1:5,000 dilution using chemiluminescence detection. Thirty
micrograms of protein from the samples were loaded and ß-Actin
was used as a loading control.
in liquid nitrogen.
MPO activity assay
To quantify MPO activity and to correlate the activity to clinical
disease severity, we performed MPO activity assays according
to the method established by Klebanoff et al. (1984) against
guaiacol using a UV/vis spectrometer (Varian Cary 50 Bio
UV-Vis spectrometer, CA) at 470nm. Mouse brains were prepared
as described earlier for Western blot analyses. Forty micro-
grams of protein were used for each assay. The units of
activity were computed according to the following formula:
in absorbance; Vt=total volume; Vs=sample volume; E (extinc-
tion coefficient)=26.6mM?1; ?t=change in time.
MR imaging was performed using a 4.7T Bruker Pharmascan MRI
scanner with a mouse brain coil under respiration-monitored
isoflurane gas anaesthesia (Bruker Biosciences, Billerica, MA).
Some animals were also imaged on a 7T Bruker Pharmascan MRI
scanner (Bruker Biosciences, Billerica, MA) to demonstrate that
the MPO sensor can report MPO activity at higher field strengths
due to improved pharmacokinetics after MPO activation, which is
independent of field strength. Comparison between MPO imaging
and DTPA(Gd) imaging was also performed at the same field
strength for each animal. Pre- and post-contrast spin-echo
T1-weighted images (TR=800, TE=13, four signal acquired,
acquisition time of 6min 57s, matrix size 192?192, field of view
2.5?2.5cm2, slice thickness 0.7mm and 16 sections were
acquired) were obtained after the administration of 0.3mmol/kg
of either agent. Post-contrast imaging was obtained sequentially
for at least 60min after contrast administration. To minimize
differences resulting from lesion progression between imaging
sessions, we randomized the order of the agent administration,
with half of the animals administered with the MPO sensor first,
and the other half of the animals administered with the conven-
tional agent first. We also only imaged those animals that did
not change clinical staging between the imaging periods. Both
MPO-imaging and DTPA(Gd) imaging were performed within
24h of each other, with a minimum of 6h between agent
injections to ensure clearance of the previously injected agent,
resulting in an average time between different agent administra-
tion of 15.1?8.5h.
Contrast-to-noise ratios (CNR) were computed for each region
of interest (ROI) according to the formula: CNR=(ROIlesion–
ROInormal brain)/SDnoise, where ROIlesionis the ROI of an enhanc-
ing lesion, ROInormal brainindicates the ROI of an unaffected area
of the brain, and SDnoiseis the standard deviation of noise from
an ROI measuring empty space. The resultant curves were
compared using the paired permutative Kolmogorov–Smirnov
test, which does not assume normal distributions. A P-value50.05
was considered to be statistically significant.
Comparisons of lesion detection and lesion burden were
performed by visually counting the number of and area of
enhanced lesions over the entire brain for each mouse by two
independent observers blinded to the injected agent (JWC and
MOB), and the results were averaged. Only parenchymal lesions
were included in the analyses. For volumetric analysis, the area
of the lesion was multiplied by the slice thickness to arrive at the
MPO imaging in murine EAE Brain (2008) Page 3 of11
lesion volume. The resultant data were analysed with the student
t-test. MPO activity assay was also analysed with the two-tailed
student t-test. The number of animals used for each part of
the study was chosen to achieve 90% power. A P-value 50.05
was considered to indicate a statistically significant difference.
All results were reported with standard deviation, except where
indicated. All statistical computations were performed using a
statistical software package (R, version 2.4.1, R Foundation for
Statistical Computing, Vienna, Austria).
MPOis present in the brains of MS model
mice and correlates with disease severity
Experimental autoimmune encephalomyelitis (EAE) is an
animal model of MS most often used in experimental
and clinical trials. To determine baseline levels of MPO we
performed Western blotting on whole brain specimen of
mice with and without EAE. Control mice have negligible
MPO heavy chain levels while MPO was markedly
up-regulated in EAE mice (Fig. 2a). We next assessed
MPO enzymatic activity in EAE mice at different stages of
disease progression. In control mice, no significant MPO
activity was detected in the brain. In EAE mice, MPO
activity in the brain was substantially higher (from 4.8 to
55 U/mg of protein, with an average of 26 (SEM 8.2) U/mg
of protein). When stratified by clinical staging, there was a
statistically significant difference between the MPO activity
of stages 0–1 (no limb weakness) and stages 2–3 (with limb
weakness) (Fig. 2b, P=0.012). There was increased percent
demyelination at increasing clinical disease severity (Fig. 2c,
MPO activity was plotted against its corresponding clinical
disease stage (Fig. 2c), we found a positive correlation
Areas of increased MPO activity co-localized
with demyelination and tissue destruction
In EAE mice sacrificed at clinical stages 2 and 3, there
were multiple areas of macrophage/microglia accumulation,
a finding most pronounced in the cerebellum. Macrophage/
microglia accumulated primarily around microvascular
structures and contained MPO (Fig. 3). In one of the
mice a lesion was detected by MPO imaging as early as day
5 after induction in the cerebellum (Fig. 3a). The areas of
demyelination in the white matter corresponded well to
areas identified by MPO and the MR images obtained with
Fig. 2 MPO in EAE. (a) Western blotting confirmed the presence of MPO heavy chain in the brain specimen of mice induced with EAE.
(b) Biochemical analysis of the MPO activity as a function of clinical staging demonstrated that higher clinical staging is associated
with increased MPO activity. (c) There was increased demyelination at increasing clinical disease severity (R2=0.94, P=0.0021).
(d) MPO activity of each animal versus its clinical disease stage. A positive correlation was found with R2=0.73, P=0.014.
Page 4 of11Brain (2008) J.W.Chen et al.
the MPO sensor (Fig. 3b). Double immunofluorescence
labelling (Fig. 4a) showed co-localization of MPO positive
cells to macrophage/microglial cells, confirming macro-
phage/microglial cells as the source of MPO. Confocal
microscopy (Fig. 4b) further demonstrated widespread
architectural distortion and loss of normal white matter
in areas of macrophage/microglia accumulation and MPO
expression (Fig. 4b). MPO and inflammatory cells infil-
trated not only white matter but also affected the cortex
(Fig. 3). Similar findings have been reported in mouse, rat
and marmoset monkey EAE models (Pomeroy et al., 2005;
MacKenzie-Graham et al., 2006; Merkler et al., 2006;
Storch et al., 2006), and in human MS (Kidd et al., 1999;
Kutzelnigg et al., 2005).
Active lesions show increased enhancement
and retention of the MPO sensor
To determine whether MPO imaging would allow in vivo
sensing of active inflammation we performed serial compar-
ative MRI studies using the MPO sensor and the conven-
tional agent DTPA(Gd). Similar to previous studies,
Fig. 3 MPO imaging and histopathological correlation. (a) A lesion in the cerebellum positive for the MPO sensor, and corresponded
to MPO and macrophage/microglia positive areas on histopathology. Note the increased enhancement at 90min compared to at 6min.
This lesion was detected by MPO imaging on day 5 after induction. Imaging was performed at 4.7T . (b) A large area in the cerebellum
with marked delayed enhancement that corresponded to areas positive for MPO and macrophages/microglia.The areas of
demyelination corresponded closely to MPO positive areas and areas exhibiting delayed enhancement on MPO imaging. Imaging
was performed at 7T .
MPO imaging in murine EAEBrain (2008) Page 5 of11
there was diffuse breakdown in the BBB seen by both
DTPA(Gd) and MPO imaging, particularly in the cerebel-
lum and the brainstem (Seeldrayers et al., 1993; Floris et al.,
2004). With the conventional contrast agent DTPA(Gd)
there was an initial increase and subsequent rapid loss
of enhancement over time (Fig. 5c), similar as reported
clinically. In contradistinction, because of activation by
MPO, imaging with bis-5HT-DTPA(Gd) showed higher
increase in contrast at the same dosage as DTPA(Gd) and
in addition demonstrated prolonged tissue enhancement
persisting over at least 60min (Fig. 5c). Some representative
examples are shown in Fig. 5a and b where many lesions
were detected by MPO imaging but not by conventional
contrast enhanced MRI. Delayed MPO images confirmed
MPO activation by demonstrating persistent enhancement
(examples given by arrows in Fig. 5a). Comparing the
immediate post contrast images, conventional MRI showed
a more diffuse, less well-defined pattern of enhancement
(Fig. 5b), while MPO imaging demonstrated a more
discrete enhancement pattern (Fig. 5a) that reflected focal
Fig. 4 Macrophages/microglia are the primary source of MPO. (a) Double immunofluorescence microscopy of a perivascular
lesion shows that MPO co-localized with macrophages/microglia. (b) Confocal microscopy confirmed MPO correlated
with macrophages/microglia. In addition, there was widespread architectural distortion in MPO positive areas.
Page 6 of11 Brain (2008)J.W.Chen et al.
areas of inflammation corresponding to inflamed perivas-
cular lesions histopathologically (Fig. 4a). In addition,
delayed images demonstrated widespread retained enhance-
ment (Fig. 5), indicating more diffuse MPO infiltration,
similar to histopathological results (Fig. 3). Furthermore,
larger lesion volume determined on delayed MPO images
correlated with higher MPO activity (R2=0.72, Fig. 5d) and
corresponded to higher clinical disease severity (P=0.0052,
Fig. 5e). Furthermore, enhancing areas measured from
Fig. 5f) with percent demyelination than those measured
from conventional imaging (R2=0.65, P=0.098, Fig. 5f).
Areas of persistent enhancement on MPO imaging thus
represented areas with elevated MPO expression and
activity, and co-localized to regions of increased inflamma-
tion with macrophage/microglia infiltration, demyelination,
and disruption of the normal brain structure. Given the
short blood half-life of the non-activated MPO-sensitive
Fig. 5 Representative MPO versus DTPA(Gd) imaging. Both MPO imaging and conventional imaging (top versus bottom) represented
the same animal at the same level.Three representative animals are shown. In the first two animals (left to right), conventional gadolinium
was administered first followed by MPO sensor. In the last animal, the MPO sensor was administered first, followed by conventional
gadolinium. (a) MPO imaging demonstrated more lesions on early post contrast images compared to conventional imaging, shown in
(b).In addition, delayed enhancementconfirmed MPO-mediated activation with the resultant prolonged pharmacokinetics. Arrows identify
several focal active inflammatory lesions that are confirmed on the delayed MPO images. Note that some of the lesions detected by
MPO imaging were either absent or barely perceptible, even retrospectively, on conventional DTPA(Gd) imaging. (b) Conventional
DTPA(Gd) imaging for comparison. Imaging was performed at 4.7T . (c) Time course evaluations of normalized CNR (nCNR) revealed
increased contrast enhancement and prolonged pharmacokinetics in mice EAE lesions when imaged with the MPO sensor compared to
the nonspecific DTPA(Gd). (d) Lesion volume measured from delayed MPO imaging correlated with MPO activity (R2=0.72, P=0.034).
(e) Larger lesion volume measured on delayed MPO images corresponded to worse clinical disease. (f) Enhancing area from MPO
imaging correlated significantly better with percent demyelination (solid black line, R2=0.96, P=0.002) than did conventional imaging
(dashed gray line, R2=0.65, P=0.098).
MPO imaging in murine EAE Brain (2008) Page 7 of11
agent bis-5HT-DTPA(Gd) monomer [5.4?0.9min, similar
as DTPA(Gd)], the prolonged retention is consistent with
MPO-agent activation and protein binding. While there is
protein binding once the MPO sensor is activated, resulting
in prolonged enhancement, we found that within 6h after
injection, the MRI signal changes had reverted to back-
ground (Fig. 5a/b, third animal from the left) potentially
allowing for serial/longitudinal imaging.
MPOimaging improves sensitivity
of lesion detection
There was consistently improved lesion detection by
MPO imaging compared to conventional MRI (Fig. 6a).
MPO imaging was able to detect more lesions (Fig. 6a,
P=0.0013) and much smaller lesions (about 40% smaller,
P=0.031, Fig. 6b). Furthermore, MPO imaging detected
lesions at an earlier time point (e.g. Fig. 5, third animal
from left) where more lesions were detected and confirmed
by MPO imaging but not by conventional DTPA(Gd)
imaging performed 6h later where some of the lesions
detected by MPO imaging were either absent or barely
perceptible, even retrospectively. Correlative MPO histo-
pathology confirmed the improved imaging sensitivity
by MPO imaging (Fig. 6c).
In the present study, we demonstrate that a small molecule
myeloperoxidase substrate can be used to non-invasively
image active inflammatory, demyelinating lesions in the
CNS. Similar to previous studies (Nagra et al., 1997;
Chataway et al., 1999; Brennan et al., 2001), we have found
that MPO, being a key modulator of inflammation
Fig. 6 MPO imaging increases lesions detection sensitivity. (a) MPO imaging detected more lesions, and (b) MPO imaging detected
smaller lesions. (c) Correlative MPO histopathology to imaging for the first and third mice shown in Fig. 5 illustrates the increased
lesion sensitivity of MPO imaging compared to conventional imaging.
Page 8 of11Brain (2008) J.W.Chen et al.
(Klebanoff, 1967, 1970; Heinecke, 1999), is closely related to
active demyelination. Additional biochemical assays indicate
that MPO enzyme activity is markedly increased in active
inflammatory lesions. Our histopathological results reveal
extensive MPO infiltration, consistent with previous find-
ings that MPO can transcytose after secretion (Baldus et al.,
2001; Tiruppathi et al., 2004). These findings validate the
approach of targeting MPO to monitor active inflammation
and demyelination in vivo.
Both the DTPA(Gd) and the inactive MPO sensor have
similar R1 relaxivity (both about 4.5mM?1s?1at 1.5T).
Both conventional DTPA(Gd) and the MPO sensor reach
the CNS through breakdown in the BBB. However,
DTPA(Gd) relies on relatively large local concentrations
to be detectable by MRI and also does not possess any
tissue or molecular specificity (Bruck et al., 1997; Cotton
et al., 2003). On the other hand, in a separate study, we
have found that in mouse model of myocardial infarction,
the MPO sensor was able to distinguish between wild-
type mice with full MPO expression, heterozygous MPO-
deficient mice with intermediate MPO expression and
MPO-knockout mice with no MPO expression, thereby
demonstrating high specificity to the enzyme myeloperox-
idase (unpublished data). The increased lesion detection
sensitivity and specificity seen on MPO imaging result from
MPO-mediated activation of the MPO sensor that leads
to signal amplification and persistent enhancement in
actively inflamed areas. The increased sensitivity of MPO
imaging allowed the detection and confirmation of earlier
and less severe lesions with more subtle breakdown in
[Figs 3a (early lesions) and 6c (small lesions)]. Therefore,
unlike conventional, non-specific imaging agents, MPO
imaging not only indicates BBB breakdown, but also the
presence of actual inflammation. Specifically, MPO oxidizes
the 5-hydroxytryptamide (5-HT) moieties on the MPO
sensor into free radicals, which then combine to form
oligomers up to 5 unites in length (Chen et al., 2006) and
bind to proteins (Querol Sans et al., 2006). The resultant
increase in molecular size and molecular dynamics generate
a large increase in R1 relaxivity (shortened T1 relaxation
time) and increase in MR signal intensity on T1-weighted
MRI sequence (Chen et al., 2006). Furthermore, the larger
molecular size and protein binding of the activated sensors
cause retention of these products at sites of higher MPO
activity, which has been confirmed by previous isotope
studies (Querol Sans et al., 2005; Chen et al., 2006).
However, because we found that the activated MPO sensor
was cleared from the brain within 6h after administration,
the protein-bound, activated MPO agents are likely digested
and released by proteases that are present at sites of
inflammation. While increasing the dose of DTPA(Gd) may
increase lesion detection sensitivity, DTPA(Gd) cannot
definitively confirm if the detected lesions represent active
inflammation. Collectively, these properties of the MPO
sensor allow earlier detection of active inflammatory,
demyelinating lesions, and confirm areas of MPO activity
non-invasively. The power of MPO imaging lies not only
in its increased sensitivity from MPO-mediated signal
amplification, but more importantly, in its ability to non-
invasively confirm with high specificity the presence of
pathology—that of elevated MPO activity and the presence
of active inflammation.
We have shown that MPO imaging corresponds to areas
of MPO expression and secretion, and represents regions
of inflammation with macrophage/microglia infiltration,
demyelination, and tissue disorganization. Interestingly, not
all mac-3 positive cells (microglia/macrophages) contained
MPO (Fig. 4b). These cells may represent a sub-population
of macrophage/microglia that does not participate directly
in inflammation (e.g. the M2 sub-population) (Stein et al.,
1992; Gordon, 2003; Mosser, 2003). Recently, several
groups have demonstrated that ultrasmall superparamag-
netic iron oxide (USPIO) nanoparticles can be used to
non-invasively image macrophages in rodent EAE and
human MS brains (Xu et al., 1998; Dousset et al., 1999;
Rausch et al., 2003; Floris et al., 2004; Berger et al., 2006;
Brochet et al., 2006; Dousset et al., 2006). It is likely that
a combination of MPO imaging and macrophage imaging
would allow the non-invasive identification and tracking of
these different sub-populations of macrophages/microglia
to further our understanding of their different roles in
demyelinating diseases. While the current study used para-
magnetic Gd in the MPO-sensitive chelator, it is possible
to exchange the metallic cation for PET (64Cu) or SPECT
(111In) detectable tracers. These comparative studies are
ongoing and may offer additional insight into disease
processes and further improve sensitivity. The MRI studies
performed here had high spatial resolutions but were
essentially identical to clinical pulse sequences. Thus, MPO
imaging should be easily deployed in the community since
no specialized pulse sequence or new equipment is needed.
Our results could have several implications in the clinical
setting. MPO imaging could be useful to screen susceptible
individuals in the presymptomatic stage, leading to earlier
treatment to decrease neurodegeneration and consequent
morbidity. In addition, in established MS patients, MPO
imaging could be used to better match clinical symptoms to
improve relapse detection as well as more accurate temporal
monitoring of active disease and therapeutic response.
While this study used a higher dosage compared to the
clinical human dose [0.3mmol/kg versus 0.1mmol/kg for
both DTPA(Gd) and MPO sensor], because mouse MPO
is only about 10–20% as active as that of human MPO
(Rausch and Moore, 1975), we expect that in humans the
increase in sensitivity would be even more pronounced
than demonstrated in our study. While formal toxicity
and stability studies are on-going, preliminary data thus
far revealed no adverse events in mice serially injected
with 5-fold the normal dose, and we found that the
gadolinium chelation in the MPO sensor is more stable
than that of DTPA(Gd) in zinc transmetallation assays
MPO imaging in murine EAEBrain (2008) Page 9 of11
(unpublished data). Therefore, enzymatic imaging targeting
MPO points to a promising new technology for non-
invasive confirmation of active inflammatory lesions in MS,
potentially not only improve disease diagnosis and treat-
ment assessment in the clinical setting, but may also lead to
better evaluation of drug development and clinical trials
of new therapies.
National Institute of Health (R24-CA92782, R01-HL078641
to R.W.); the National Multiple Sclerosis Society (J.W.C.);
the Dana Foundation (J.W.C.); the German National
the Boehringer Ingelheim
We thank the CMIR Chemistry Core (F. Reynolds and
L. Josephson) for performing the chemical synthesis. We
also thank C. Rangel, C. Kaufman, T. Sponholtz, V. Lok
and L. Stangenberg for experimental assistance.
Baldus S, Eiserich JP, Mani A, Castro L, Figueroa M, Chumley P, et al.
Endothelial transcytosis of myeloperoxidase confers specificity to
vascular ECM proteins as targets of tyrosine nitration. J Clin Invest
2001; 108: 1759–70.
Berger C, Hiestand P, Kindler-Baumann D, Rudin M, Rausch M. Analysis
of lesion development during acute inflammation and remission in a rat
model of experimental autoimmune encephalomyelitis by visualization
of macrophage infiltration, demyelination and blood-brain barrier
damage. NMR Biomed 2006; 19: 101–7.
Bradley PP, Christensen RD, Rothstein G. Cellular and extracellular
myeloperoxidase in pyogenic inflammation. Blood 1982; 60: 618–22.
Brennan M, Gaur A, Pahuja A, Lusis AJ, Reynolds WF. Mice lacking
myeloperoxidase are more susceptible to experimental autoimmune
encephalomyelitis. J Neuroimmunol 2001; 112: 97–105.
Brochet B, Deloire MS, Touil T, Anne O, Caille JM, Dousset V, et al. Early
macrophage MRI of inflammatory lesions predicts lesion severity and
disease development in relapsing EAE. Neuroimage 2006; 32: 266–74.
Bruck W. The pathology of multiple sclerosis is the result of focal
inflammatory demyelination with axonal damage. J Neurol 2005; 252
(Suppl 5): v3–9.
Bruck W, Bitsch A, Kolenda H, Bruck Y, Stiefel M, Lassmann H.
Inflammatory central nervous system demyelination: correlation of
magnetic resonance imaging findings with lesion pathology. Ann Neurol
1997; 42: 783–93.
Chataway J, Sawcer S, Feakes R, Coraddu F, Broadley S, Jones HB, et al.
A screen of candidates from peaks of linkage: evidence for the involve-
ment of myeloperoxidase in multiple sclerosis. J Neuroimmunol 1999;
Chen JW, Querol Sans M, Bogdanov AA Jr, Weissleder R. Imaging
myeloperoxidase in mice using novel amplifiable paramagnetic sub-
strates. Radiology 2006; 240: 473–81.
Cotton F, Weiner HL, Jolesz FA, Guttmann CR. MRI contrast uptake in
new lesions in relapsing-remitting MS followed at weekly intervals.
Neurology 2003; 60: 640–6.
Dousset V, Ballarino L, Delalande C, Coussemacq M, Canioni P, Petry KG,
et al. Comparison of ultrasmall particles of iron oxide (USPIO)-
enhanced T2-weighted, conventional T2-weighted, and gadolinium-
enhanced T1-weightedMR imagesin rats with experimental
autoimmune encephalomyelitis. AJNR Am J Neuroradiol 1999; 20:
Dousset V, Brochet B, Deloire MS, Lagoarde L, Barroso B, Caille JM, et al.
MR imaging of relapsing multiple sclerosis patients using ultra-small-
particle iron oxide and compared with gadolinium. AJNR Am J
Neuroradiol 2006; 27: 1000–5.
Filippi M, Yousry T, Campi A, Kandziora C, Colombo B, Voltz R, et al.
Comparison of triple dose versus standard dose gadolinium-DTPA for
detection of MRI enhancing lesions in patients with MS. Neurology
1996; 46: 379–84.
Floris S, Blezer EL, Schreibelt G, Dopp E, van der Pol SM, Schadee-
Eestermans IL, et al. Blood-brain barrier permeability and monocyte
infiltration in experimental allergic encephalomyelitis: a quantitative
MRI study. Brain 2004; 127: 616–27.
Frohman EM, Goodin DS, Calabresi PA, Corboy JR, Coyle PK, Filippi M,
et al. The utility of MRI in suspected MS: report of the Therapeutics
and Technology Assessment Subcommittee of the American Academy of
Neurology. Neurology 2003; 61: 602–11.
Frohman EM, Racke MK, Raine CS. Multiple sclerosis–the plaque and its
pathogenesis. N Engl J Med 2006; 354: 942–55.
Giovannoni G, Silver NC, Good CD, Miller DH, Thompson EJ.
Immunological time-course of gadolinium-enhancing MRI lesions in
patients with multiple sclerosis. Eur Neurol 2000; 44: 222–8.
Gordon S. Alternative activation of macrophages. Nat Rev Immunol 2003;
Greer JM, Sobel RA, Sette A, Southwood S, Lees MB, Kuchroo VK.
Immunogenic and encephalitogenic epitope clusters of myelin proteo-
lipid protein. J Immunol 1996; 156: 371–9.
Heinecke JW. Pathways for oxidation of low density lipoprotein by
myeloperoxidase: tyrosyl radical, reactive aldehydes, hypochlorous acid
and molecular chlorine. Biofactors 1997; 6: 145–55.
Heinecke JW. Mechanisms of oxidative damage by myeloperoxidase in
atherosclerosis and other inflammatory disorders. J Lab Clin Med 1999;
Imitola J, Chitnis T, Khoury SJ. Insights into the molecular pathogenesis
of progression in multiple sclerosis: potential implications for future
therapies. Arch Neurol 2006; 63: 25–33.
Jacobs LD, Beck RW, Simon JH, Kinkel RP, Brownscheidle CM,
Murray TJ, et al. Intramuscular interferon beta-1a therapy initiated
during a first demyelinating event in multiple sclerosis. CHAMPS Study
Group. N Engl J Med 2000; 343: 898–904.
Kidd D, Barkhof F, McConnell R, Algra PR, Allen IV, Revesz T. Cortical
lesions in multiple sclerosis. Brain 1999; 122 (Pt 1): 17–26.
Klebanoff SJ. A peroxidase-mediated antimicrobial system in leukocytes.
J Clin Invest 1967; 46: 1078–85.
Klebanoff SJ. Myeloperoxidase: contribution to the microbicidal activity of
intact leukocytes. Science 1970; 169: 1095–7.
Klebanoff SJ, Waltersdorph AM, Rosen H. Antimicrobial activity of
myeloperoxidase. Methods Enzymol 1984; 105: 399–403.
Kutzelnigg A, Lucchinetti CF, Stadelmann C, Bruck W, Rauschka H,
Bergmann M, et al. Cortical demyelination and diffuse white matter
injury in multiple sclerosis. Brain 2005; 128: 2705–12.
MacKenzie-Graham A, Tinsley MR, Shah KP, Aguilar C, Strickland LV,
Boline J, et al. Cerebellar cortical atrophy in experimental autoimmune
encephalomyelitis. Neuroimage 2006; 32: 1016–23.
Merkler D, Boscke R, Schmelting B, Czeh B, Fuchs E, Bruck W, et al.
Differential macrophage/microglia activation in neocortical EAE lesions
in the marmoset monkey. Brain Pathol 2006; 16: 117–23.
Miller D, Barkhof F, Montalban X, Thompson A, Filippi M. Clinically
isolated syndromes suggestive of multiple sclerosis, part I: natural
history, pathogenesis, diagnosis, and prognosis. Lancet Neurol 2005; 4:
Mosser DM. The many faces of macrophage activation. J Leukoc Biol
2003; 73: 209–12.
Nagra RM, Becher B, Tourtellotte WW, Antel JP, Gold D, Paladino T,
et al. Immunohistochemical and genetic evidence of myeloperoxidase
involvement in multiple sclerosis. J Neuroimmunol 1997; 78: 97–107.
Page10 of11Brain (2008)J.W.Chen et al.
Noseworthy JH, Lucchinetti C, Rodriguez M, Weinshenker BG. Multiple Download full-text
sclerosis. N Engl J Med 2000; 343: 938–52.
PomeroyIM, MatthewsPM, Frank
Demyelinated neocortical lesions in marmoset autoimmune encephalo-
myelitis mimic those in multiple sclerosis. Brain 2005; 128: 2713–21.
Querol M, Chen JW, Weissleder R, Bogdanov A Jr. DTPA-bisamide-
based MR sensor agents for peroxidase imaging. Org Lett 2005; 7:
Querol Sans M, Chen JW, Bogdanov AA Jr. A paramagnetic contrast agent
with myeloperoxidase-sensing properties. Org Biomol Chem 2006; 4:
QuerolSansM, ChenJW, Weissleder
Myeloperoxidase activity imaging using (67)Ga labeled substrate. Mol
Imaging Biol 2005; 7: 403–10.
Rausch M, Hiestand P, Baumann D, Cannet C, Rudin M. MRI-based
monitoring of inflammation and tissue damage in acute and chronic
relapsing EAE. Magn Reson Med 2003; 50: 309–14.
RauschPG, Moore TG.Granule
neutrophils: a phylogenetic comparison. Blood 1975; 46: 913–9.
Seeldrayers PA, Syha J, Morrissey SP, Stodal H, Vass K, Jung S, et al.
Magnetic resonance imaging investigation of blood-brain barrier
JA, JordanEK, EsiriMM.
R, BogdanovAA Jr.
damage in adoptive transfer experimental autoimmune encephalomy-
elitis. J Neuroimmunol 1993; 46: 199–206.
Stein M, Keshav S, Harris N, Gordon S. Interleukin 4 potently enhances
murine macrophage mannose receptor activity: a marker of alternative
immunologic macrophage activation. J Exp Med 1992; 176: 287–92.
Storch MK, Bauer J, Linington C, Olsson T, Weissert R, Lassmann H.
Cortical demyelination can be modeled in specific rat models of
autoimmune encephalomyelitis and is major histocompatability complex
(MHC) haplotype-related. J Neuropathol Exp Neurol 2006; 65: 1137–42.
Tiruppathi C, Naqvi T, Wu Y, Vogel SM, Minshall RD, Malik AB.
Albumin mediates the transcytosis of myeloperoxidase by means of
caveolae in endothelial cells. Proc Natl Acad Sci U S A 2004; 101:
Xu S, Jordan EK, Brocke S, Bulte JWM, Quigley L, Tresser N, et al. Study
of relapsing remitting experimental allergic encephalomyelitis SJL Mouse
Model using MION-46L enhanced in vivo MRI: early histopathological
correlation. J Neurosci Res 1998; 52: 549–58.
Zakrzewska-Pniewska B, Styczynska
Peplonska B, Barcikowska M, et al. Association of apolipoprotein E
and myeloperoxidase genotypes to clinical course of familial and
sporadic multiple sclerosis. Mult Scler 2004; 10: 266–71.
M, PodleckaA, SamockaR,
MPO imaging in murine EAEBrain (2008) Page11of11