Dual Contrast Molecular Imaging Allows Noninvasive Characterization of Myocardial Ischemia/Reperfusion Injury After Coronary Vessel Occlusion in Mice by mri running title
Download 1.2 Mb. Pdf ko'rish
|
- Bu sahifa navigatsiya:
- Figure 3D
- Figure 3F
- Figure 3J
- Molecular MRI noninvasively characterizes platelet accumulation and myocardial necrosis
- Figure 4A
- Figure 5A/B
- Figure 7A
- Figure 7E
- Figure 8A/B
|
Figure 3B, with platelets stained in red and Gr1 in black; corresponding negative controls omitting the primary antibody are shown in Figure 3C. Higher magnification (x100) shows an example of a platelet-neutrophil-conjugate, with platelet-bound MPIOs on the surface (Figure
shows equivalent numbers of platelet-neutrophil-conjugates in the two groups (Figure 3E). An exact quantification and characterization of platelet accumulation was performed in all animals. A representative myocardial section with platelet staining by anti-CD41 immunohisto- chemistry is depicted in Figure 3F, also showing MPIO binding (Figure 3G, arrows). In non- ischemic myocardium, such as the ventricular septum, no platelet staining can be observed, confirming that platelet accumulation is limited to ischemic/reperfused myocardial tissue (Figure 3H). Microthrombi are equally distributed in both groups (Figure 3I), whereas MPIO binding was found to be significantly increased in animals injected with LIBS-MPIO (p<0.005,
LIBS-MPIO and platelets in ischemic areas (Figure 3K), highlighting the stability of the chosen animal model and the specificity of LIBS-MPIO as an imaging tool that accurately reflects platelet involvement. Molecular MRI noninvasively characterizes platelet accumulation and myocardial necrosis The effect of MPIO-induced signal decrease was quantified as described in the “Methods”. Signal quantification was pooled for the two segments (anterior and anterolateral) representing the areas of ischemia (Figure 4A, red arrows). After LIBS-MPIO-injection, a significant signal decrease, as typical for a MPIO-induced effect, can be observed (Figure 4B, red line, LIBS- chemistry is depicted in Figure 3F, also showing MPIO binding (Figure 3G, ar r ro
ows w w ) ). . I I I n n n no no no n- n- schemic myocardium, such as the ventricular septum, no platelet staining can be observed, co o nf nf nf ir ir ir mi mi ming
ng ng t t t ha h h t pl pl pl a at atelet accumulation is limited to to to
i ischemic/reperf rf f
ed d m
m myocardial tissue F Fig gu gu re 3H). M M ic i i r r roth
th h ro ro mb mb mb i i i ar ar ar e e eq eq qu ua uall l y y d dist t r ri b b bute e d in b b bo ot oth h gr gr rou
ou p ps ps ( ( Fi Fi i gu g gu re re 3 3 3 I I I ) ), ), w w he he he re re r a a as M M M PI PI P O O O bi i nd nd ndin in ing g g wa wa a s s s f
fo foun
un nd d d to to b be e e si si gn n n if if i ic ic ic an an an tl tl tl y y i in incr
cr crea
ea ease
e ed d d in in n a a ani ni ma ma mals
ls ls i inj nj jec ec e t t ted
d d wi
wi w t t th L L IB IB BS- S- S-MP MP PI IO IO ( ( (p< < <0. 0. 0 0 005
5, 5,
Figure 3J). C C C or or or re re e la la la ti ti t on on on ana na na ly ly ly si i s s s sh sh s ow ow o s s s a a si si si gn gn gn if if if ic ic c a a nt nt nt c c or or r re re re la la la ti ti t on on on b b b et t t we we we en en en t t t he he he a a a mo mo mo un un un t t t of o bound
by guest on December 22, 2017 http://circ.ahajournals.org/ Downloaded from
DOI: 10.1161/CIRCULATIONAHA.113.008157 14
MPIO vs. control-MPIO: p<0.02), suggesting effective binding of LIBS-MPIO towards activated platelets. After gadolinium injection, the signal in LIBS-MPIO-injected animals increased above baseline, representing the gadolinium-induced signal increase in necrotic myocardium (p<0.003 for mean values post-CA vs. post Gd). Animals injected with control-MPIO showed no MPIO- induced signal decrease and no significant increase after injection of gadolinium when comparing the mean values of all time points before vs. after injection of Gd. P2Y 12 knockout mice demonstrate reduced platelet and neutrophil accumulation, which is reflected in molecular MRI Additional experiments were performed in P2Y 12 -/-
mice in order to study the extent of myocardial ischemia/reperfusion injury and platelet accumulation in a setting with reliable platelet inhibition. After injection of LIBS-MPIO, no signal increase was observed in these mice, and a similar effect was seen in animals injected with control-MPIO (Figure 5A/B). Furthermore, the presence of myocardial necrosis was confirmed after the injection of gadolinium in both groups (Figure 5A/B, right-hand side, yellow arrows). The extent of LGE was significantly lower in P2Y 12 -/- mice, as quantified by MRI (p<0.005, Figure 6A). As a consequence of the reduction of platelet accumulation in P2Y 12 -/-
mice, no significant difference in the MRI signal after MPIO injection was observed between control-MPIO (green line) and LIBS-MPIO (yellow line) in these animals (Figure 6B). When comparing the mean values of all time points before versus after injection of Gd, there was no significant difference for LIBS-MPIO-injected animals or control-MPIO-injected animals. To confirm the observed reduction in LGE signal, we also assessed infarct size in histology and echocardiography. Infarct size was significantly smaller in P2Y 12 -/-
mice in TTC staining (Figure 6C) and left ventricular ejection fraction demonstrated a trend towards better myocardial ischemia/reperfusion injury and platelet accumulation in a setting wit it t h h re
r r l li iab
ab ab le le le
platelet inhibition. After injection of LIBS-MPIO, no signal increase was observed in these mice an n d d d a a a si si si mi mi mi la la la r effe
fe fe c ct ct was seen in animals injected d d w
w i i ith control-MPI PI PI O ( (
Fi Fig g gure 5A/B). Furthermore h h he pr p esence o o f f f my my my oc oc a a ardi di di al al al n n n ec ec ecro
ro si si s s s s w wa as s s con
nf nf i irme e e d d af f t te te r r r th th h e e in inje je e ct t ti i f f f f on on o o o f f ga ga do do do li lini ni i um um um i i in n n bo
bo bo th h th
gr gr ou ou oups
ps ps (
( (
Fi gu gu gu re re 5 5 5A A/ A/B B , , , ri i igh gh g t- ha ha h nd nd nd s s s id id id e, e, y y el el el lo lo low w ar ar a ro ro ow w ws). ). ). T T Th h h e e e ex ex xte te t n nt nt o o o f f f LG LG LG E E w wa wa s s s si si gn gn gn i ific
c can
n ntl
tl y y y ower in P2Y Y 12
12 / / -/- m m m ic ic ic e, e, e, a a a s qu
qu qu an an a ti ti ti f f f ie ie i d d d by by y M M M RI RI RI ( ( ( p< p< p< 0. 0. 0. 00 00 0 5, 5,
f f f f f f
Fi Fi gu gu gu re re e 6 6 6 A A ). ). ). A A A s s s a a a co co co ns ns s eq eq eq ue ue ue nc nc nc e e of the by guest on December 22, 2017 http://circ.ahajournals.org/ Downloaded from DOI: 10.1161/CIRCULATIONAHA.113.008157 15
preservation in P2Y 12 -/- mice (Figure 6D) compared to WT. To support these findings, immunohistochemistry was performed in all P2Y 12 -/-
mice.
Indeed, only minor platelet infiltration was detected in ischemic areas (Figure 7A, CD41- staining), and infiltration with platelet-neutrophil-conjugates decreased (Figure 7B, CD41/Gr1- staining). Quantification of platelets in WT vs. P2Y 12 -/- mice confirmed that there is significantly less platelet accumulation in P2Y 12 -/- mice
(p<0.0005, Figure 7C), and the amount of bound MPIOs was significantly reduced towards the level of WT-mice in all P2Y 12 -/- mice (p<0.01 vs. WT with LIBS-MPIO, Figure 7D). When adding the results of the P2Y 12
mice to WT-animals, correlation between bound MPIOs and platelet presence remains highly significant (p=0.0001, Figure 7E) and the amount of platelet-neutrophil-conjugates was reduced in P2Y 12 -/- animals
(p<0.005, Figure 7F). In summary, the results from immunohistochemistry confirm a decrease in platelet and neutrophil accumulation in P2Y 12 -/-
mice, and this corresponds well to the MRI findings in these animals. As expected, the extent of myocardial necrosis was not significantly different between LIBS-MPIO and control-MPIO injected mice, whereas the extent of myocardial necrosis was less in P2Y 12 -/- mice (Figure 8A/B). The size of the area with a LIBS-MPIO-induced effect in MRI correlated well with the size of the LGE-area of the left ventricle (p<0.01, Figure 8C), as well as with the area of necrosis in histology (p<0.05, Figure 8D). Together with the data obtained with the P2Y 12 -/-
mice, the area of necrosis correlated well with the LGE-area (p<0.0001, Figure 8E).
In this study, we were able to noninvasively characterize ischemia/reperfusion injury after temporary coronary artery ligation in mice. We established a unique dual contrast magnetic
Y Y 12 2 -/- -/- -/-
an an an im im m al al als s s p<0.005, Figure 7F). In summary, the results from immunohistochemistry confirm a decrease n n p p p la la la te te te l le le t t t an an a d d d ne
ne e ut ut utro
r phil accumulation in P2Y 12 2
-/- -/-
m m mice, and this co c c rr r es es sp po ponds well to the MRI fi fi ind d din
i gs in thes es e
a an nim im mal
al s s. s. As As e e e xp x xpec ec te te d, d, d t t h he he e e xt x en en nt t of of of m m m yo yo o ca ca ard rd rd ia ia ia l ne ne n c cr cro os sis s s w w was
as as n
n ot ot ot s s ig ig i ni ni ni fi fi f c c ca n n ntl
tl y y di di iff ff f er er r en en ent b
b be t tw twee
ee en
n LIBS-MPIO O an
an d d d co co co nt nt n ro ro ro l- l MP MP MP IO IO I i i i nj nj n ec ec ec te e e d d d mi mi mi ce ce ce, , wh wh wh er er e ea ea e s s th th th e e e ex ex e te te te nt nt n o o o f f f my my my oc oc oc ar ar a di di i al al al n n n ec ec ec ro ro r sis was by guest on December 22, 2017 http://circ.ahajournals.org/ Downloaded from DOI: 10.1161/CIRCULATIONAHA.113.008157 16
resonance imaging approach, which allows on the one hand the detection of necrosis in myocardium subjected to ischemia/reperfusion, employing the late enhancement effect of gadolinium-enhanced MRI. On the other hand, using a unique molecular contrast agent, which is specifically targeted towards activated platelets, we were able to detect platelet accumulation as a marker of microvascular obstruction as well as inflammation in mouse myocardium. Immunohistochemical analysis demonstrated an excellent correlation of infarct size and platelet accumulation with the MRI findings, showing the potential of this noninvasive approach. Furthermore, using P2Y 12 -/-
mice as a control providing reliable platelet inhibition and reflecting therapeutic intervention, we were able to confirm the feasibility of activated platelets as targets for imaging, the central role of platelets in myocardial ischemia/reperfusion injury, and the potential benefits of imaging the effects of therapeutic inhibition of platelets, in particular of the P2Y
12 receptor. With increasing reduction of mortality of acute MI, the assessment of the risk to develop adverse cardiovascular events and the decision making between revascularization and medical treatment has become a major challenge 28, 29
. Left ventricular ejection fraction (e.g. LVEF 30%) is currently one of the major means for risk stratification e.g. to predict the development of sudden cardiac death 30 . However, a considerable number of patients with LVEF > 30% still suffer from sudden cardiac death and on the financial side, the consequence of classifying a high number of patients falsely as being at high risk is exerting pressure on our health care systems 30 .
Various methods have been assessed towards delivering a patient-specific risk stratifica- tion and thus ultimately to provide personalized medicine. Amongst these are advanced technologies such as CT and MRI: Dual-energy and multidetector CT, especially in combination with functionalized contrast reagents, have the potential to improve the contrast between for imaging, the central role of platelets in myocardial ischemia/reperfusion inju u r ry ry , an an d d d th th the
e e potential benefits of imaging the effects of therapeutic inhibition of platelets, in particular of the P2 2 Y Y Y 12 12 2 r r r ec ec ep ep ep o to to r. With inc
nc r r rea a asin in ing
g r re re du du duct ct cti io ion
n of of of m m m o or rtali ity ty of ac ac ac u ute e e M MI MI, , th h he e a a as s se ess
s m me me nt nt o o o f f f th the e ri ri r sk sk sk t t t o o de de dev ve elo lo op ad d d v ve vers
rs rse
e e ca ca rd rd rd io i iova va asc c cul
ul a ar ar e e eve ve nt nt t s s s an an and
d d th th th e e de de de ci ci ci si si si on n n
m ma mak
ki king
ng g b b b et et et we we we en en e rev ev evas
as a c c cu l la lari
ri i za za z ti ti ti on on n a
a a n nd nd m m ed edic ic cal
l l
reatment ha a s s be be be co co o me me m a a maj aj aj or or or c c c ha ha ha ll ll en en e ge ge ge
28 28 28, 29 29 29 . . Le Le Le ft ft ft v v v en en e tr tr r ic ic ic ul ul ul ar ar ar e e e je je je ct t t io io io n n n fr fr fr ac ac ac ti ti t on on n ( ( ( e. e. e g. g. g L L L VEF by guest on December 22, 2017 http://circ.ahajournals.org/ Downloaded from
DOI: 10.1161/CIRCULATIONAHA.113.008157 17
diseased and normal myocardium. However, the lower contrast sensitivity and the required radiation are challenging limitations in comparison to MRI 31 .
MRI after MI is increasingly seen as a leading imaging modality for risk assessment and personalized therapeutic decision making. For example, so far LGE in MRI is the strongest predictor of mortality and major cardiac adverse events as compared with clinical characteristics, coronary angiographic assessment and LV echocardiographic parameters 31, 32 . Recently, another new MRI method allowing detection of fibrosis via T1-weighted imaging has been shown to be a promising risk classifier 30, 33 . The molecular MR imaging of activated platelets may represent an additional method that warrants further testing in regards to its potential use for risk prediction in patients after MI. In contrast to LGE and T1-weighted MRI, imaging of activated platelets detects a process that directly drives inflammation and may thus be a better reflection of ischemia/reperfusion injury. MRI offers the advantage to compare these methods head to head in the same patient. Positron emission tomography (PET) has superior sensitivity, and especially in combination with the capability of the CT and MRI to exactly localize PET signals it is highly attractive for the assessment of cardiac ischemia and viability 28, 34 .
F-fluoro-2-deoxy-D- glucose ( 18 F-FDG) can be used as a marker of inflammation in MI. However, ischemia induces a shift towards glycolysis in cardiac cells, which can result in an 18 F-FDG signal that is not inflammation-specific 35 . Other molecular markers in PET, which are increasingly available, could be used in comparison to LIBS-MPIO to determine the functional role of platelets in cardiac ischemia/reperfusion. Nevertheless, if a functionally predictive, molecular imaging approach directed against activated platelets can be developed towards application in humans, MRI would be preferable compared to PET as it is non-radioactive, independent of cyclotron patients after MI. In contrast to LGE and T1-weighted MRI, imaging of activated ed d p p la la l t te e le le le ts ts ts
detects a process that directly drives inflammation and may thus be a better reflection of s s ch ch h em em emi ia ia /r /r /r ep ep ep e e erfu
u u si si si on on injury. MRI offers the adva va a nt nt a age to compare re e the e se se se m methods head to head in h h he sa s me patie e nt nt . . Po Po si si si tr tr tr on on e em mi mi ss ss s io io n n n to t m mo mo gr gr gr ap ap ap h hy hy ( ( PE PE PET
T) T) h h has
as a s s up up upe e ri rior or or s s s en en nsi
si s ti ti i vi vi ity
ty ty , , a a an d d d es es spe
pe peci
ci c al al al ly ly ly in
n n combination n wi wi wi th th t t t t he he h c c c ap ap a ab ab b il il l it i i y y y of of o t t t he h h C C C T T T an an an d d d MR MR MR I I to to t e e e xa xa xa ct ct ct ly ly ly l l l oc oc o al al l iz iz iz e e e PE PE PE T T T si si gn gn gn al al al s s s it it it is highly by guest on December 22, 2017 http://circ.ahajournals.org/ Downloaded from
DOI: 10.1161/CIRCULATIONAHA.113.008157 18
access, cheaper and more broadly available. Interestingly, although there is a strong correlation between the overall area of LIBS- MPIO signal and LGE-signal, the two imaging techniques do not represent the same ventricular area. LIBS-MPIO signals and histological localization are partially found to be outside of the necrotic myocardium. This may reflect the inflammatory reaction of ischemia/reperfusion injury, which is in accordance with recent findings that platelets play a pathogenic role in this pathological process, are a therapeutic target, and that their involvement determines the rate of complications in mice such as ventricular rupture 6, 36 . Overall, the combined LIBS-MPIO/LGE imaging could provide important clinical information delineating the area at risk and inflammation, in addition to the area of necrosis as determined by LGE-imaging. Whether the combination of these two imaging methods allows a clinically relevant risk prediction of adverse cardiovascular events in patients after MI remains to be determined. The area of LIBS-MPIO signal correlated well with the area of histological platelet accumulation. Also the area of infarct as measured in histology and by LGE correlated well with the overall LIBS-MPIO signal. However, the area of platelet accumulation does not have to be restricted to ischemic myocardium, as was similarly shown for matrix metalloproteinase activity as a marker of inflammation in the non-ischemic myocardium 35 . The area of post-ischemic inflammation detectable by platelet targeted imaging might be larger and might have a prognostic value on its own. However, the clinical value of the LIPS-MPIO signal remains to be determined. The increasingly available combination of MRI and PET will provide the opportunity to compare MR platelet accumulation data with PET inflammation data such as obtained by 18 F-FDG 28 .
P2Y 12 receptor blockers are known to prevent atherothrombosis. However, recent data nflammation, in addition to the area of necrosis as determined by LGE-imaging. g. W W W he he eth
th th er er er t t t he h he
combination of these two imaging methods allows a clinically relevant risk prediction of adverse k ca a rd rd d io io iov va vasc sc c ul ul ul a a ar e
e ve ve ve n nts in patients after MI remai i ns ns n t t o o be determine ne e
d
The ar r ea ea o f f f LI LI BS BS S M -M -M PI PI PIO
O O si
si i gn gn gnal
al l c
c orre
e e la a ated
w w well w wi with th h t h he he a a a re e ea of
of of h
i ist
t ol ol ol o og ogic ic al al a p p p la la l te tele le let t ac c cu cu cumu mu mula
la ti ti i on on on . . A Al lso o o t
t h he he a a a re re a of of of i i i nf nf nf ar ar ar ct ct a a s s s me me meas s sur
ur u e ed ed i in hi hi st st st ol ol olog g gy
y y a an an d d d by by by L L LGE
GE E c
c or or o re re e la la a t ted d d w
we well
ll w w it it h he overall LI LI BS BS BS -M -M M PI PI PI O O O sign gn gn al a a . Ho Ho Ho we we w ve ve ve r, r, r, t t t he he h a a a re re r a a a of of f p p la la a te te te le le le t t t ac ac ac cu cu c mu mu mu la la la ti ti ti on on on d d oe oe e s s s no no no t t t ha h h ve to be by guest on December 22, 2017 http://circ.ahajournals.org/ Downloaded from DOI: 10.1161/CIRCULATIONAHA.113.008157 19
indicates that they can also reduce ischemia/reperfusion injury in myocardial infarction, which is suggested to be mediated by reduced platelet and neutrophil accumulation 2, 6, 37 . Molecular MRI accurately reflects reduced platelet accumulation and platelet-neutrophil complex deposition as well as a reduction of myocardial damage in the histological assessment of P2Y 12 -/-
mice. Overall, these data suggest that molecular magnetic resonance imaging of activated platelets may represent a novel method to assess the extent of functional inhibition, achieved by various anti- platelet regimens. Platelets also play a pivotal role in the context of microvascular obstruction after MI 7 . The angiographically observed “no-reflow”-phenomenon after opening of occluded vessels by percutaneous coronary intervention is a critical phenomenon, with such patients having an increased risk for congestive heart failure, rhythm disturbances, or death 38-40
. So far, there is no consensus for the detection or characterization of a microvascular obstruction, e.g. by SPECT or contrast-enhanced echocardiography. The so far available techniques have limitations in sensitivity and specificity, which includes LGE in MRI that tends to underestimate the subsequent scar formation 41 . Therapeutic approaches to reduce microvascular obstruction by using GPIIb/IIIa inhibitors positively influenced myocardial flow and infarct size in a dog model 42 , thereby indicating that platelets play a causative role in microvascular obstruction. Platelets may either accumulate intravascularly via adhesion to inflamed endothelium, in the form of occluding microthrombi or extravascularly through the leaking of ruptured microvessels 7 .
as such potentially represents a tool for the prognosis of outcome in MI patients. In addition, MR imaging of activated platelets is attractive to be used as a direct parameter of successful (or unsuccessful) reperfusion of the microcirculation. percutaneous coronary intervention is a critical phenomenon, with such patients h h hav
v v in ing g g an an an ncreased risk for congestive heart failure, rhythm disturbances, or death 38-40 . So far, there is no co o ns ns ns en en n su su s s s fo fo fo r r r th h e e e d de detection or characterization of of of a a microvascular ar r
s tr tr tru uc uction, e.g. by SPECT or r r co on nt ntra r st-enhan
n ce ce e d d ec c h ho h c ca ca rd rd rd io io iog
gr grap
ap p hy hy hy.
Th Th he so o o f f far a
av av a aila a a b bl le e t t te c ch chni ni i qu qu ues e h h h av v ve e li li li mi m mita
ta ti ti ti on on ns s in in
e e ns ns nsit it itiv iv iv it it y y y an an an d d sp sp spe
ec cif
if i ic cit it ty, y, y w w hi hi h ch ch ch i i i nc n nclu lu u de de es s s L LG LG E E E in in n M M MR RI RI t t t ha ha ha t t te te end
nd nds
to to to u und
nd nd er eres es sti ti i ma ma m t te te t the e e ubsequent s s ca ca ca r r r fo fo o rm rm rm at at a io io i n n 41 41 41 . . . Th Th h er er e ap ap ap eu u u ti ti t c c c ap ap ap pr pr p oa oa oa ch ch h es es e t t o o o re re re du du du ce ce ce m m m ic ic c ro ro ro va va va sc sc sc ul ul u ar ar r o o o bs bs b tr tr t uc uc u tion by by guest on December 22, 2017 http://circ.ahajournals.org/ Downloaded from DOI: 10.1161/CIRCULATIONAHA.113.008157 20
A major strength of our imaging approach is the platelet-targeted contrast agent itself. The LIBS-MPIO contrast agent allows the detection of activated platelets with a unique level of sensitivity and specificity. LIBS-MPIO has already been used in a number of studies by our group, and has allowed for the detection of coronary and carotid thrombosis or cerebrovascular inflammation in mice 16, 22
. The LIBS-antibody also binds to human platelets, also in an activation-specific manner, which is an important step towards the translation of this promising technology to application in humans 17 . Presence of LIBS-MPIO in ischemic/reperfused heart correlates well with the histological presence of platelets in the heart - this is an important prerequisite for noninvasive characterization of pathologies by MRI. The hypo-intense contrast effect generated by MPIOs necessitated pre- and post-contrast agent imaging, and therefore a single MRI sequence throughout the experiment. Such an imaging protocol has to provide good T 2 * as well as T 1 contrast in one scan. In addition, it has to allow a reasonable time resolution at sufficient signal to noise rates. The applied ECG-triggered FLASH sequence provides this balance. Moreover, during the protocol setup, it proved to be superior to the retrospectively triggered INTRAGATE method. A translation of the recently described cardiac mouse imaging approach using INTRAGATE 43 was not transferable to our scientific question, due to irregular heart rates resulting from the induced myocardial injury. Download 1.2 Mb. Do'stlaringiz bilan baham: 
|