The progress of incorporating deep learning in the field of medical image interpretation has been greatly hindered due to the tremendous cost and time associated with generating ground truth for supervised machine learning, alongside concerns about the inconsistent quality of images acquired. Active learning offers a potential solution to these problems of expanding dataset ground truth by algorithmically choosing the most informative samples for ground truth labeling. Still, this effort incurs the costs of human labeling, which needs minimization. Furthermore, automatic labeling approaches employing active learning often exhibit overfitting tendencies while selecting samples closely aligned with the training set distribution and excluding out-of-distribution samples, which could potentially improve the model’s effectiveness. We propose that the majority of out-of-distribution instances can be attributed to inconsistent cross images. Since the FDA approved the first whole-slide image system for medical diagnosis in 2017, whole-slide images have provided enriched critical information to advance the field of automated histopathology. Here, we exemplify the benefits of a novel deep learning strategy that utilizes high-resolution whole-slide microscopic images. We quantitatively assess and visually highlight the inconsistencies within the whole-slide image dataset employed in this study. Accordingly, we introduce a deep learning-based preprocessing algorithm designed to normalize unknown samples to the training set distribution, effectively mitigating the overfitting issue. Consequently, our approach significantly increases the amount of automatic region-of-interest ground truth labeling on high-resolution whole-slide images using active deep learning. We accept 92% of the automatic labels generated for our unlabeled data cohort, expanding the labeled dataset by 845%. Additionally, we demonstrate expert time savings of 96% relative to manual expert ground-truth labeling.

Myocardial ischemia reperfusion injury (IRI) in acute coronary syndromes is a condition in which ischemic/hypoxic injury to cells subtended by the occluded vessel continues despite successful resolution of the thrombotic obstruction. For decades, most efforts to attenuate IRI have focused on interdicting singular molecular targets or pathways, but none have successfully transitioned to clinical use. In this work, we investigate a nanoparticle-based therapeutic strategy for profound but local thrombin inhibition that may simultaneously mitigate both thrombosis and inflammatory signaling pathways to limit myocardial IRI. Perfluorocarbon nanoparticles (PFC NP) were covalently coupled with an irreversible thrombin inhibitor, PPACK (Phe[D]-Pro-Arg-Chloromethylketone), and delivered intravenously to animals in a single dose prior to ischemia reperfusion injury. Fluorescent microscopy of tissue sections and 19F magnetic resonance images of whole hearts ex vivo demonstrated abundant delivery of PFC NP to the area at risk. Echocardiography at 24 h after reperfusion demonstrated preserved ventricular structure and improved function. Treatment reduced thrombin deposition, suppressed endothelial activation, inhibited inflammasome signaling pathways, and limited microvascular injury and vascular pruning in infarct border zones. Accordingly, thrombin inhibition with an extraordinarily potent but locally acting agent suggested a critical role for thrombin and a promising therapeutic strategy in cardiac IRI.

Introduction: The role of thrombin in acute myocardial infarction (AMI) has been well described, with recent attention having turned to its effect on NLRP3 inflammasome activation and the role that plays in the pathogenesis of AMI. Perfluorocarbon nanoparticles conjugated to phenylalanine-proline-arginine-chloromethylketone (PPACK NPs) are extremely potent local inhibitors of thrombin, but their role in therapy for AMI remains unclear. We hypothesize that pre-treatment with sustained locally acting PPACK NPs preserves cardiac function by attenuating thrombin-induced inflammatory processes, including those driven by the NLRP3 inflammasome. Methods: Nine mice underwent left anterior descending (LAD) occlusion for 45 minutes. At 30 minutes before LAD occlusion, mice were injected with PPACK (n=5) or control NPs (n=4). 24 hours post-reperfusion, hearts were excised and perfused. LAD was re-occluded and Lectin-488 perfused through the excised heart. Echo was performed pre- and post-reperfusion to assess cardiac function. Hearts were sectioned for follow up histological and protein analysis. Results: PPACK NP treated mice exhibited evidence of preserved cardiac function 24 hours post-reperfusion (Figure 1, A-B), reduced nuclear translocation of p65 in infarcted myocardium (Figure 1, C-D), and reduced inflammatory markers with better vessel preservation in the area adjacent to uninjured myocardium (Figure 1, E-F). The PPACK NP group also demonstrated suppression of NLRP3 inflammasome activation (Figure 2). Conclusion: Local inhibition of thrombin with anti-inflammatory PPACK NPs preserves cardiac function when administered systemically as pre-treatment.

Cell-cell fusion or syncytialization is fundamental to the reproduction, development, and homeostasis of multicellular organisms. In addition to various cell type-specific fusogenic proteins, cell surface externalization of phosphatidylserine (PS), a universal eat-me signal in apoptotic cells, has been observed in different cell fusion events. Nevertheless, the molecular underpinnings of PS externalization and cellular mechanisms of PS-facilitated cell-cell fusion are unclear. Here, we report that TMEM16F, a Ca2+-activated phospholipid scramblase (CaPLSase), plays an essential role in placental trophoblast fusion by translocating PS to cell surface independent of apoptosis. The placentas from the TMEM16F knockout mice exhibit deficiency in trophoblast syncytialization and placental development, which lead to perinatal lethality. We thus identified a new biological function of TMEM16F CaPLSase in trophoblast fusion and placental development. Our findings provide insight into understanding cell-cell fusion mechanism of other cell types and on mitigating pregnancy complications such as miscarriage, intrauterine growth restriction, and preeclampsia.

Thrombin, a major protein involved in the clotting cascade by the conversion of inactive fibrinogen to fibrin, plays a crucial role in the development of thrombosis. Antithrombin nanoparticles enable site-specific anticoagulation without increasing bleeding risk. Here we outline the process of making and the characterization of bivalirudin and D-phenylalanyl-L-prolyl-L-arginyl-chloromethyl ketone (PPACK) nanoparticles. Additionally, the characterization of these nanoparticles, including particle size, zeta potential, and quantification of PPACK/bivalirudin loading, is also described.

Abstract Introduction We have shown previously that pretreatment of acute ischemic kidney injury (AKI) in mice prior to reperfusion with anti-thrombin perfluorocarbon nanoparticles (PFC NP) limits damage to endothelium and hastens functional recovery. However, whether such treatments are effective after AKI is established is not known. We hypothesized that thrombin would continue to exert deleterious clotting and molecular signaling effects in AKI well after reperfusion that would respond to sustained local inhibition with long acting anti-thrombin nanoparticles. Methods 23 C57Bl6 mice underwent bilateral kidney ischemia for 17 min, followed by 2 hours reperfusion and i.v. injection of anti-thrombin PPACK (D-phenylalanyl-L-prolyl-L-arginine chloromethyl ketone)-conjugated PFC NP (∼ 13,000 PPACK per PFC NP), or plain PFC NP (control: no drug). At 24 hours BUN was measured, and mice were euthanized for kidney histological assessment (H&E), protein expression (western blot) and eicosanoid mediators of inflammation (LC-MS/MS: AB SCIEX 5500 QTRAP). Results BUN at 24 hours after AKI was 63.29±9.09 vs 110.96±6.21 (P<0.002), for treated versus untreated mice, respectively, a 43% improvement. Western blots (Figure) indicated 40% reduction of canonical NF-kB signaling pathway protein p65 (p<0.01) and 2.2 fold increases in Bcl-xL: Bax ratio (P<0.01). Vascular damage, as indicated by glomerular and mesangial hemorrhage (Figure), was reduced, as was tubular cell swelling and edema. Levels of inflammatory procoagulant eicosanoids (e.g., PGE1, TBX2, PGA2, 15-HETE, 5-HETE, etc.) generally were higher in renal medulla than in cortex, and were suppressed by PPACK PFC NP. Discussion Continued inhibition of thrombin in AKI with locally-acting PPACK PFC NP preserved vascular integrity, limited renal hemorrhage, mitigated inflammation and tubular cell death, and accelerated functional recovery even when administered 2 hours after reperfusion. Because these PPACK PFC NP do not prolong bleeding times or coagulation parameters beyond ∼30–60 min after injection, yet maintain prolonged local surveillance against activated thrombin, they represent a potentially useful therapeutic strategy for established AKI after an ischemic insult. Acknowledgement/Funding DK102691

Lupus nephritis requires immunosuppressive therapy, which unfortunately, are associated with severe side effects. Thus, development of new treatment strategies which minimize the side effects while maintaining high therapeutic efficiency is essential. Perfluorocarbon (PFC) nanoparticles (NPs), with a nominal size of ~200 nm, are primarily limited to intravascular space. Collagen IV (col4) expresses in the glomerular basement membrane (GBM), where is the only place that col4 has direct contact with blood via fenestrated capillary endothelium. Accordingly, we designed a novel col4‐targeted NPs by coupling PFC NPs with amine‐carboxyl to a col4‐targeting ligand, which selectively target glomeruli by binding to col4 of GBM.The binding specificity and efficiency of the col4‐targeted NPs were evaluated on col4 surface, in vitro and in vivo. First, rhodamine labeled col4 targeted or non‐targeted PFC NPs were applied in col4 pre‐coated 96 wells plate at stepwise doses (1, 2, 5, 10 and 20 ul/ml). The binding affinity was determined by fluorescence intensity with IVIS. The wells incubated with col4 targeted PFC NPs showed a dose dependent fluorescence in radiant efficiency e10 (0.37 ± 0.03, 0.51 ± 0.05, 1.33 ± 0.07, 2.38 ± 0.19, 2.71 ± 0.28) (n=5, p<0.01 vs non‐targeted NP). Second, to evaluate the targeting properties in vitro, 1 μl/ml rhodamine labeled col4 targeted or non‐targeted NPs were applied on mouse primary glomerular endothelial cells and mouse primary mesangial cells for 2 h at 37°C. At the same confocal imaging settings, significant cellular uptake of the col4 targeted NPs was visualized, while it was not detectable in the cells incubated with non‐targeted NPs (n=3). Then, to evaluate the targeting properties in vivo, C57BL/6 mice were administered i.v. with 100 μl rhodamine labeled col4 targeted or non‐targeted NPs. At 24 hours after injections, col4 targeted NPs were selectively visualized at glomeruli in kidney sections, while non‐targeted NPs were barely detectable anywhere (n=3).Furthermore, the therapeutic effect of the col4 targeted NPs loaded with prednisone for lupus nephritis was evaluated in MRL‐lpr mice, a lupus‐prone animal model. The prednisone loaded col4 targeted NPs (1 ml/kg) were intravenously injected in MRL‐lpr mice (male, 8 weeks old) twice a week for 8 weeks. Compared with untreated animals, MRL‐lpr mice receiving therapeutic NPs exhibited less glomerular deposition of IgG (142±25 vs.193±18) and C3 (28±11 vs.41±13) evaluated by intensity of immunofluorescence (n=5, p<0.05 vs control), decreased proteinuria (54±17 vs.82±21 mg/dl) measured by with MultistixTM dipsticks (n=6, p<0.05 vs control), a better preservation of GFR (157±16 vs. 118±12 ml/min) assessed by clearance of plasma FITC‐inulin after a single intravenous bolus injection (n=6, p<0.01 vs control), and reduced glomerular pathology (1.3±0.8 vs. 2.4±0.6) evaluated by scoring 50 glomerular cross sections per kidney with periodic acid‐Schiff stain (n=5, p<0.05 vs control).In conclusion, we developed novel NPs that selectively target glomeruli by binding to col4 of GBM, which provide a novel tool to achieve precise therapeutics with minimum side effects for lupus nephritis.Support or Funding InformationNIH: DK099276 DK098582 HL137987 DK102691 AR067491 HL073646‐08ASN Fellowship Awards AHA: 18CDA34110441This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.

Acute myocardial infarction (AMI) is a life‐threatening medical condition. Based on the statistics updated in 2017, there is a person experiencing myocardial infarction every 40 seconds. For AMI patients, majority of them miss the treatment window for rescuing significant amounts of viable myocardium. However, infarction size can expand for weeks after the initial AMI, which provides an opportunity to prevent ventricular remodeling (progressive chamber enlargement) as well as improve healing process. AMI associated secondary ischemia further induces endothelial damage, elevated propagulation status, inflammation, and extracellular collagen degradation. Accordingly, the objective of this study is to reduce early ventricular remodeling by preventing microvascular thrombosis and reducing inflammation simultaneously with PPACK (phenylalanine‐proline‐arginine chloromethylketone) perfluorocarbon (PFC) nanoparticles (NP). Benefits of delivery PPACK by PFC NP are 1) enabling formation of anticoagulation surface specifically in the damaged vessels where thrombin is activated in order to preserve vascular integrity and limit inflammation induced by thrombin signaling and 2) minimum systemic exposure to avoid the risk of bleeding. Fluorescence microscopic images demonstrated the delivery of PFC NPs to the area at risk and remain there 24 hours after the i.v. injection (n=10) (Fig. A). For all 32 mice underwent closed‐chest ischemia (LAD occlusion) for 90 min followed by 24 hrs reperfusion. Before and 24 hrs after ischemic reperfusion injury, echocardiography was performed (VEVO 2200). The results (Fig. B) demonstrated that PPACK NP treated mice exhibited marked preservation of cardiac functional parameters as early as 24 hours after ischemia as compared to control mice according to changes in contractile metrics: 1) ΔEDV%: 40%±21% vs 77%±38% for treated vs control (p=0.02) and 2) ΔESV%: 118%±43% vs 216%±108%, for treated vs control (p=0.02) (n=16/group). To better understand the therapeutic mechanism, transcriptome profile has been evaluated by using RNASeq on normal zones and border zones of AMI mice with (n=5) or without (n=4) PPACK PFC NP treatment. The results demonstrated that in the border zone after 24 hours reperfusion, PPACK PFC NP treatment attenuated inflammatory NF‐kB signaling, and reduced NF‐kB dependent proteases (MMP 8, 9, 25), which were ~2 times higher in control group (Fig C). Moreover, numerous collagen transcripts, including Col1, 5, 6, 11, 16, and 24, were significantly elevated in the treatment group (Fig D), offering protective effects in AMI. These results demonstrate that PPACK PFC nanoparticle treatment may enhance healing in AMI and preserve cardiac microarchitecture. Support or Funding Information R01 DK102691R01 AR067491R01 HL073646‐08 This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal .

Abstract Introduction Persistent microvascular thrombosis and thrombin signaling in cardiac ischemia reperfusion injury (IRI) combine to promote infarct expansion and heart failure. We have reported previously that treatment of acute ischemic kidney injury (AKI) in rodents with anti-thrombin perfluorocarbon nanoparticles (PFC NP) limits damage to endothelium, reduces “no reflow”, and hastens recovery of kidney function. In cardiac IRI, we hypothesized that thrombin continues to exert deleterious effects well after reperfusion, but whether this strategy is effective in preserving vascular barrier function and preventing no reflow in cardiac IRI is not known. Accordingly, we postulated that sustained local thrombin inhibition with long acting anti-thrombin nanoparticles might limit vascular damage and barrier disruption that otherwise leads to hemorrhage and no reflow. Methods 14 Sprague Dawley rats undergoing LAD occlusion for 45 min were pretreated with i.v. injection of anti-thrombin PPACK (D-phenylalanyl-L-prolyl-L-arginine chloromethyl ketone)-conjugated PFC NP (∼13,000 PPACK per PFC NP), or saline (control) followed by 2 hours of reperfusion. Sham controls (no LAD ligation) were compared. Simultaneous to treatments, a different and unique nontargeted CE PFC NP (CE: Crown Ether core) was injected for subsequent tracking and spectroscopic quantification of the relative distributions and amounts of PFC NP trapped in non-reperfused myocardium. Hearts were perfused with saline in situ, then extracted for ex vivo 19F (fluorine) MRI and spectroscopy at 7T (Bruker). T2* images were acquired for hemorrhage/edema in the area at risk. 19F MRI was acquired for PFC NP trapping (spin echo: TR 2000ms, TE 4.51 ms, FOV 20x20, 64x64 slice 1mm), along with whole heart 19PFC CE spectroscopy. Results Fluorine MRI of an intact heart illustrates the trapping of PFC NP in areas of nonreperfusion (Figure) amidst infarct scar region. Quantification of total heart CE PFC NP signal emanating from trapped NP revealed reduced accumulation (−49%) in the hearts of treated rats (0.47±0.09 vs 0.92±0.19, treated vs control; p<0.05). For sham occlusions, minimal deposition of CE PFC NP was observed irrespective of PPACK PFC NP treatment (n=10 per group). Areas of T2* signal indicative of vascular leakage extended beyond that of the 19F signals (Figure T2*), indicating more severe damage and trapping in central infarct zones than in more peripheral border zones after 2 hours. Discussion Continued inhibition of thrombin in cardiac IRI with locally-acting PPACK PFC NP preserved vascular integrity, and limited hemorrhage and No Reflow (less trapping). Because these PPACK PFC NP do not prolong bleeding times or coagulation parameters beyond ∼30–60 min after injection, yet maintain prolonged local surveillance against activated thrombin, they represent a potentially useful therapeutic strategy for cardiac IRI.