Non-invasive mapping of systemic neutrophil dynamics upon cardiovascular injury

Figures

• 

  Fig. 1. Mapping the trafficking of murine neutrophils after MI by mNPPFCs in vivo. a, Whole-body 3D 1H/19F MRI for systemic in vivo visualization of neutrophils before and after MI. Anatomical 1H data were rendered transparent in grayscale with 19F data overlaid in orange and red; for the sake of clarity, signals from the liver and spleen were faded out. Left, intravenous application of mNPPFCs before MI resulted in in situ labeling of neutrophils within their hematopoietic niches, showing the most prominent 19F signal in the femur and tibia. The yellow arrows indicate the areas of the local scans from the femur and heart in c,e, respectively. Right, re-investigation 24 h after MI revealed a pronounced reduction of 19F signals in the bone marrow of the femur and tibia with simultaneous appearance of 19F labeling in the infarcted heart. b, Post-mortem flow cytometry of the different bone marrow compartments confirmed the in vivo findings in that the femur was not only the bone containing the highest number of neutrophils before MI but also the compartment with the strongest release of neutrophils as compared to all other bones. c, Focal 1H/19F MR images of the bone marrow after in vivo…

• 

  Fig. 2. Correlation of the 19F signal with ischemic area, its temporal development and ex vivo validation. a, Linear regression between ischemic area (LGE) and fluorine signal within the infarcted myocardium. b, Time course of neutrophil infiltration into the injured heart. The fluorine signal was determined 1, 3, 6, 24, 48 and 72 h after induction of MI (examples are given at the top). c, To further corroborate the location of the 19F signal within the infarcted myocardium, hearts were excised, fixed with paraformaldehyde and analyzed by ex vivo high-resolution 1H/19F 3D MRI. In long-axis 1H MR images, the infarcted area can be unequivocally identified as a dark structure within the bright intact myocardium. Importantly, the corresponding 19F signal is restricted to the infarcted area, indicating infiltration of mNPPFC-labeled neutrophils only into the injured myocardium. Left, 3D volume rendering of the heart with superimposed 19F signal. LV, left ventricle; RV, right ventricle. Data are mean ± s.d. of n = 6 (a) and n = 5–6 (b) independent experiments. Source data

• 

  Fig. 3. Identification of mNPPFC-labeled neutrophils in the infarcted heart. a, For identification of mNPPFC-labeled neutrophils in the infarcted heart, mice received injections of rhodamine-tagged mNPPFCs or ConPFCs 2 h before induction of MI and hearts were excised 2 h after induction of MI. The infarct area was visualized by 2,3,5-triphenyltetrazolium chloride (TTC) staining, which was found to colocalize with strong rhodamine signals derived from mNPPFC-labeled cells (bottom). By contrast, ConPFCs led to much weaker and spread-out signals only (top). Magnifications and histograms (fourth and fifth columns) demonstrated distinct signals for mNPPFCs but only diffuse patterns for ConPFCs. b, Left, examination of infarct (MI) and remote (RM) regions revealed strong rhodamine labeling colocalized with Ly6G staining in the injured tissue. Right, quantification confirmed the selective uptake of mNPPFCs versus ConPFCs. c, Analysis of mNPPFC uptake by cardiac neutrophils (left, Ly6G staining), macrophages (middle, major histocompatibility complex (MHC) II staining) and T cells (right, CD3 staining). Rhodamine signals were colocalized with neutrophils, while macrophages and T cells showed…

• 

  Fig. 4. Limitations of mNP and identification of CD177 as a target for hNP on human neutrophils. a, Uptake of either mNPPFCs and ConPFCs (left) or the free peptides mNP and Con (right) by murine immune cells, quantified by flow cytometry. Note that mNPPFCs (red, left) were strongly taken up by neutrophils (Neu), whereas unconjugated free mNP (red, right) did not label neutrophils. Ly, lymphocytes; Mo; monocytes. b, Left, mNP (red) as well as the control peptide Con (gray) was conjugated to eight-arm PEG2000-maleimide, and the uptake of these conjugates by neutrophils was analyzed by flow cytometry. Right, impact of increasing mNP and Con concentrations (1, 2.5, 5 and 10 µg ml−1) on labeling of neutrophils. c, Representative flow cytometry histograms of human immune cells incubated with hNP (green) or its control peptide (Con, gray). d, Concept for identification of the binding target of hNP using TriCeps, which consists of an n-hydroxysuccinimide (NHS) group for conjugation to hNP, a hydrazide for binding to different sugar structures on the cell surface and biotin for purification. e, Isolated human neutrophils were incubated with the hNP–TriCeps conjugate, and, thereafter, cells were…

• 

  Fig. 5. Specific targeting of human neutrophils by hNPPFCs. a, Uptake of hNPPFCs (red) or ConPFCs (gray) by human lymphocytes, monocytes or neutrophils over time as determined by flow cytometry. **P < 0.01. b, For MRI analysis, human immune cells were incubated with hNPPFCs (top row) or ConPFCs (bottom row). After several washing steps, cells were purified by density gradient centrifugation and analyzed by MRI. First column, 1H MR image of the centrifugation tube with the cell layer (arrow) on top of the dark Percoll layer; second column, 19F MR image of the same area; third column, merge of both datasets. Quantification of the 19F data is shown on the right. c, hNP was conjugated to the pH-sensitive dye pHrodo, incubated with neutrophils at 4 °C (blue) and 37 °C (red) and analyzed by flow cytometry. d, Confocal microscopy of neutrophils incubated with hNPPFCs (top) or ConPFCs (bottom). Fluorescence signals of PFCs (rhodamine; Rho) as well as ligands (carboxyfluorescein; CF) were recorded, and nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI) (blue). Accumulation of hNPPFCs within the endosomal–lysosomal system is highlighted by yellow arrows. e, Neutrophils were…

• 

  Fig. 6. Targeting with hNPPFCs does not impact human neutrophil properties. a, Differentially expressed genes identified by bulk RNA sequencing of human blood neutrophils after incubation with hNPPFCs, ConPFCs or NaCl as the control. Volcano plots of differentially expressed genes for human neutrophils treated with saline compared to hNPPFCs (left) or ConPFCs compared to hNPPFCs (right). Genes marked in red are significantly upregulated with a log2 (fold change) greater than 1.5. BF, Bonferroni-corrected P values of the false discovery rate. In total, 45,413 RNA transcripts were analyzed. b, Migration of neutrophils treated with hNPPFCs (red) or left untreated (gray) toward IL-8. c, Neutrophils were incubated with hNPPFC or ConPFC or the hNP or Con peptides, and ROS generation was determined by enzyme-linked immunosorbent assay. wo, without any PFC. d, Cell surface expression of neutrophil activation markers: neutrophils were left untreated (gray, control), incubated with hNPPFCs (red) or stimulated with LPS (blue) as the positive control, followed by flow cytometry for CD11b (top), CD63 (middle) or CD66b (bottom). Mean fluorescence intensities were normalized to those of untreated…

• 

  Fig. 7. Inflammatory stimuli increase uptake of NPPFCs by neutrophils. a, Isolated neutrophils obtained from patients with STEMI 24 h after MI (red) and from healthy controls (gray) were exposed to hNPPFCs, and their uptake was determined over time by flow cytometry. b, Human neutrophils from patients with STEMI (top) or healthy controls (bottom) were incubated with hNPPFCs, purified by density gradient centrifugation and analyzed by 1H/19F MRI. Cells are visible in 1H MR images as a small layer (arrow) in the bright aqueous phase on top of the black Percoll layer (left). Fluorine-19 MRI (middle) and subsequent fusion of both datasets (right) confirmed localization of the 19F label within the cellular layer. Quantification of 19F signals showed enhanced labeling of neutrophils from patients with STEMI (red) in comparison to controls (gray). c, Murine neutrophils isolated from mice with an inflammatory hot spot (Matrigel–LPS plug; red) or Matrigel–PBS as a control (gray) were incubated ex vivo with mNPPFCs and analyzed over time by flow cytometry. Histograms (left) display representative data after 40 min, and the time course of mean fluorescence values is illustrated on the right. d,…

• 

  Fig. 8. NPPFC loading as a readout for neutrophil-activation state. a, Cell surface expression of CD177 on murine neutrophils isolated from healthy controls (Ctrl, gray) after implantation of Matrigel–LPS (left) or after experimental MI (right) as determined by flow cytometry. Data were separated for circulating neutrophils isolated from the blood (BN, blood neutrophils, light red) and neutrophils obtained from the inflamed Matrigel plug or the infarcted heart (TN, tissue neutrophils, dark red). b, CD177 expression of human neutrophils isolated from blood of healthy controls (gray), patients with STEMI (blood neutrophils, light red) and neutrophils obtained from pericoronitis tissue specimens (tissue neutrophils). Similar to mice, CD177 expression was significantly increased in blood neutrophils compared to control values and again substantially higher in tissue neutrophils than in blood neutrophils. c, In vivo mNPPFC labeling of bone marrow neutrophils dependent on their activation state: mNPPFCs were intravenously injected into mice 24 h after implantation of Matrigel doped with LPS (bottom) or PBS as a control (top). Another 24 h later, the bone marrow was analyzed by 1H/19F MRI. d,…

• 

  Extended Data Fig. 1. Synthesis scheme for PFC-based nanotracers to target neutrophils. a, Basic structure of the used peptides for targeting neutrophils. The N-terminal carboxyfluorescein is incorporated for fluorescence-based analysis. For coupling the peptides to preformed MalPFCs, a cysteine with a free sulfhydryl group is located at the C-terminus, next to the GGG spacer. The red dashed circle indicates the neutrophil binding sequence. b, For generating NPPFCs, preformed MalPFCs were incubated with the peptides over night to form NP/ConPFCs. MalPFCs were equipped with a separate label (rhodamine) to control for a potential dissociation of binding peptide and PFC. To impair the passive uptake by monocytes/macrophages poly-ethylene-glycol (PEG) coupled lipids (DSPE-PEG2000 = 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[amino(poly-ethylene glycol)-2000]) were inserted in a 5 mol% ratio into the phospholipid shell of PFCs.

• 

  Extended Data Fig. 2. Specific targeting of neutrophils by mNPPFCs. a, Flow cytometric analysis of murine blood leukocytes incubated with mNPPFCs over 80 min illustrating their predominant uptake by neutrophils (red) as compared to monocytes (blue) and lymphocytes (gray). b, To further validate specificity, binding of CD177 mAb as well as incorporation of mNPPFCs under basal and LPS-stimulated conditions were determined. Almost no binding/uptake into classical or non-classical monocytes as well as macrophages or eosinophils could be detected. In contrast to all other immune cell subsets, solely neutrophils showed a significant binding/uptake. c, Labelling of bone marrow neutrophils isolated 2 h upon intravenous injection of mNPPFCs or ConPFCs as determined via flow cytometry indicating a strong uptake of mNPPFCs and almost no incorporation of ConPFCs. d, To verify that cellular uptake of mNPPFCs can be visualized by 19F MRI, we isolated murine neutrophils from the bone marrow, incubated the cells with mNPPFCs or ConPFCs for 3 h at 37 °C, washed intensively and pelleted the cells by centrifugation. Subsequently, cells were analyzed by 1H/19F MRI. Left: 1H/19F MRI of centrifugation tubes with…

• 

  Extended Data Fig. 3. Assessment of mNPPFC/ConPFC biodistribution, liver serum markers and histology of liver and spleen. a, Biodistribution of intravenously injected mNPPFCs was determined via 19F MRI measurements at distinct time points after injection (1, 3, 6 and 24 h after injection). Shown are quantification of fluorine signal intensities in the blood, liver and spleen indicating a fast wash-out from the blood accompanied by a strong influx into the liver while signal intensities in the spleen were constant over the entire observation period; a.u. = arbitrary units. b, Serum levels of GLDH (glutamate dehydrogenase), AST (aspartate aminotransferase), ALP (alkaline phosphatase), ALT (alanine aminotransferase) and bilirubin after intravenous application of NaCl as control, ConPFCs or mNPPFCs. c + d, 4 µm sections of liver (C) and spleen (D) stained with hematoxylin and eosin. Organs were excised and processed for histology 24 h post injection of NaCl (upper row), ConPFCs (middle), mNPPFCs (lower). Representative liver/spleen images of organs from three different animals are shown. Scale bars indicate 100 µm. Data are mean values ± SD of n = 6 (a), n = 4-6 (b) and n = 6 (c + d) independent…

• 

  Extended Data Fig. 4. Experimental timeline and whole body 1H/19F MRI of a mNPPFC-treated mouse. a, To label neutrophils within the bone marrow, mNPPFCs (or ConPFCs as control) were intravenously injected on three consecutive days (day −3, −2, −1) followed by 1H/19F MRI of the bone marrow on day 0. Subsequently, MI was induced and bone marrow and the infarcted heart were analyzed by combined 1H/19F MRI 24 h later. b, In separate experiments, neutralizing antibodies against CXCL1, CXCL2, G-CSF and GM-CSF were injected intraperitoneally (i. p.; 50 µg) 1 h before and 4 h post induction of MI to inhibit the MI-stimulated egress of neutrophils from the bone marrow and their infiltration into the heart. c, To determine the whole body biodistribution of mNPPFCs, mNPPFCs were intravenously injected daily over three consecutive days. Afterwards, whole body 1H/19F MRI measurements were performed to visualize the labelled neutrophils in the different bone marrow compartments. The bright 19F spot in the middle represents the liver – an organ known to be as major site of (PFC) nanoparticle deposition.