Introduction

Neutrophils play a complex role during onset of tissue injury as well as subsequent resolution and healing. To more precisely assess neutrophil dynamics upon cardiovascular injury, this study aimed at developing a noninvasive, background-free approach for specific mapping of murine and human neutrophils dynamics by whole-body MRI. For this, targeted multimodal fluorine-loaded nanotracers were engineered with binding peptides specifically directed against murine and human neutrophils, respectively.

Methods

Peptides against murine (mNP)^[1] and human (hNP)^[2] neutrophils as well as corresponding control peptides (Con) were equipped with n-terminal carboxyfluorescein and c-terminal GGG-cysteine spacer, with the latter used for coupling of the peptides to preformed maleimide-PFCs. Binding studies were performed by flow cytometry and ¹⁹F MRI using human neutrophils from healthy volunteers and patients with ST-elevated myocardial infarction. Murine neutrophils were obtained from blood of C57BL/6 mice, isolated from the bone marrow, or LPS-doped matrigel plugs^[3]. Murine myocardial infarction was induced by 50 min of transient ligation of the left anterior descending artery. MRI experiments were performed at a 9.4T Bruker AVANCEIII Wide Bore NMR spectrometer and datasets were acquired using a 25 mm birdcage resonator tuneable to ¹H and ¹⁹F.

Results and Discussion

Binding studies of ^mNPPFCs and ^hNPPFCs revealed the specific labeling of neutrophils from both mice and humans, exhibiting much lower affinity for monocytes or lymphocytes (Figs. 1+2). Moreover, PFCs modified with control peptides (Con) did not show any labelling of neutrophils (Figs. 1+2). These findings were further confirmed by ¹⁹F MRI which revealed a stronger ¹⁹F signal in neutrophils incubated with ^NPPFCs as compared to ^ConPFCs (Fig. 2).
To assess neutrophil dynamics upon cardiovascular injury, we used a model of cardiac ischemia/reperfusion, well known to be associated with a massive neutrophil recruitment into the injured myocardium^[4]. For monitoring the fate of neutrophils upon myocardial infarction (MI), mice received daily intravenous injections of ^mNPPFCs over 3 d prior surgery. Whole-body ¹H/¹⁹F MRI before induction of MI corroborated that this labeling protocol resulted in a strong ¹⁹F uptake by bone marrow neutrophils, particularly in femur and tibia (Fig. 3A left). Re-investigation 24 hrs after MI revealed a substantial reduction of ¹⁹F signals in these compartments (Fig. 3A, right) with concomitant appearance of ¹⁹F label in the infarcted heart. In line with this, flow cytometry identified the femur as main neutrophil reservoir and also as the bone marrow compartment with the largest decrease in neutrophils 24 hrs after MI (Fig. 3B). Focal scanning of the thighs before and 24 hrs after MI confirmed the strong decrease of ¹⁹F signals in the femur after MI (Fig. 4A+B). In parallel, images of the thorax unequivocally corroborated the simultaneous appearance of ¹⁹F label in the heart (Fig. 4C+D). Cine MRI in combination with late gadolinium enhancement (LGE) demonstrated that the detected ¹⁹F pattern perfectly matched the LGE-delineated myocardium (Fig. 4C, top). Importantly, animals which received ^ConPFCs exhibited significantly less labeling of the bone marrow before MI, which was unchanged after MI and led to only minor amounts of ¹⁹F label in the infarcted region (Fig. 4A-D). Furthermore, application of neutralizing antibodies (NAbs) to inhibit the egress of ^mNPPFC-loaded neutrophils from the femur into the blood blunted the MI-induced effects in heart and femur (Fig. 4C+D).
Next, we verified whether (i) this approach might be transferred to the human setting and (ii) ^NPPFC uptake by neutrophils is altered under pathological challenges. To mirror the conditions in the murine trafficking experiments after MI, we utilized isolated neutrophils from blood of STEMI patients (i.e. ST elevation MI = STEMI) obtained within the first 24 hrs after MI and found a substantially stronger cellular uptake of ^hNPPFCs compared to healthy volunteers (Fig. 5A+B). To evaluate whether this observation is associated with various stimuli, we further induced murine inflammation by subcutaneous implantation of a matrigel plug doped with LPS^[3]. Both ex vivo incubation of isolated neutrophils and in vivo application of ^mNPPFCs revealed a more rapid and potent uptake of ^mNPPFCs by murine neutrophils under stimulated conditions (Fig. 5C+D). Importantly, this effect was restricted to neutrophils while lymphocytes and monocytes showed only minor mNPPFC uptake under inflammatory conditions.
Finally, we investigated whether the enhanced incorporation of ^mNPPFCs can be exploited to assess in vivo the inflammatory state of neutrophils. To this end, we monitored in situ ¹⁹F incorporation into bone marrow neutrophils under stimulated conditions employing again the matrigel/LPS-based inflammation model. Twenty-four hrs after plug implantation, ^mNPPFCs were applied and further 24 hrs later mice were subjected to ¹H/¹⁹F MRI. As shown in Fig. 5E-G, we observed a substantially stronger ¹⁹F signal in the bone marrow upon LPS pre-activation. This effect became even more evident when relating the detected ¹⁹F signal to the number of neutrophils present in the bone marrow (Fig. 5G).

Conclusions

In summary, the present study demonstrates that both human and murine neutrophils can be specifically targeted to track their dynamic trafficking by non-invasive MRI in vivo. In clinical translation, this approach will allow not only to identify hidden origins of bacterial or sterile inflammation in patients but also to unravel cardiovascular disease states that are on the verge of severe aggravation due to enhanced neutrophil infiltration or activation.

Acknowledgements

No acknowledgement found.