25.01 Notch1—WISP-1 axis determines the regulatory role of MSC-derived fibroblasts in melanoma metastasis

Posted on January 25, 2017

P. P. Parikh1, H. Shao1, L. Cai1, M. Moller1, B. Issac2, R. Vazquez-Padron1, M. Owyong1, Z. Liu1,2  1University Of Miami,Surgery,Miami, FL, USA 2University Of Miami,Sylvester Comprehensive Cancer Center,Miami, FL, USA

Introduction:  Mesenchymal stem cells-derived fibroblasts (MSC-DF) constitute a significant portion of stromal fibroblasts in the tumor microenvironment (TME) and are key modulators of tumor progression. However, the molecular mechanisms that determine their tumor regulatory function are poorly understood. Here, we uncover the Notch1 pathway as a molecular determinant that selectively controls the regulatory role of MSC-DF in melanoma metastasis.

Methods:  MSC-DF were generated from bone marrow (BM)-derived MSC obtained from Notch1Flox/Flox and ROSAN1IC mice and transduced with Cre-GFP/Lentivirus or GFP/Lentivirus (control).  Various MSC-DF were co-cultured or co-grafted with human melanoma cells (labeled with Luc2) in vitro or in vivo. Melanoma growth and metastasis in vivo were examined by measuring tumor size and IVIS scanning. cDNA microarray was employed to identify downstream targets of Notch1 signaling.

Results: We demonstrated that the Notch1 pathway’s activity is inversely correlated with the metastasis-regulating function of fibroblasts and can determine the metastasis-promoting or -suppressing phenotype of MSC-DF. When co-grafted with melanoma cells, MSC-DFNotch1-/- selectively promoted, while MSC-DFN1IC+/+ preferentially suppressed melanoma metastasis, but not growth, in mouse models. When co-cultured with metastatic melanoma cells in vitro, MSC-DFNotch1-/- supported, but MSC-DFN1IC+/+ inhibited melanoma cells to form spheroids. We also identified Wnt-induced secreted protein-1 (WISP-1) from MSC-DF as a key downstream secretory mediator of Notch1 signaling on melanoma metastasis. 

Conclusion: These findings expand the repertoire of Notch1 signaling as a molecular switch in determining the tumor metastasis-regulating function of MSC-DF. We demonstrate that the Notch1—WISP-1 axis functions as a crucial molecular determinant in governing stromal regulation of melanoma metastasis; thus, establishing this axis as a potential therapeutic target for melanoma metastasis.

 

24.10 ECM and CEACAM Proteins Characterize Pancreatic Duct Exosomes in Patients with Pancreatic Cancer

Posted on January 25, 2017

J. Zheng1, J. Hernandez1, A. Doussot1, L. Bojmar2, B. Costa-Silva2, E. VanBeek2, C. Zambirinis1, M. Tesic Mark3, H. Molina3, M. Gonen1, T. Kingham1, P. Allen1, M. D’Angelica1, R. DeMatteo1, D. Lyden1,2, W. Jarnagin1  1Memorial Sloan-Kettering Cancer Center,New York, NY, USA 2Weill Cornell Medical College,New York, NY, USA 3Rockefeller University,New York, NY, USA

Introduction: Exosomes have emerged as important vehicles for intercellular communication between cancer cells and their microenvironment and may facilitate tumor progression. Exosomes in pancreatic duct fluid have not been previously studied. We hypothesize that exosomes can be detected in the pancreatic duct fluid and that exosomal proteins may be useful as diagnostic and/or prognostic biomarkers in patients with pancreatic ductal adenocarcinoma (PDAC).

Methods: Pancreatic duct fluid was collected from 22 patients at the time of partial pancreatectomy for PDAC (n=13), intraductal papillary mucinous neoplasm (IPMN) (n=4) and other benign pancreatic diseases (n=5). Serial ultra centrifugation was used to isolate exosomes. Exosomes were quantified using the DS500 nanoparticle characterization system (NanoSight), and confirmed by transmission electron microscopy (TEM). Exosomal proteins were identified and analyzed using liquid chromatography-mass spectrometry (LC-MS). Expression levels of exosomal proteins were compared between patients with PDAC and benign diseases using Wilcoxon rank-sum test. Survival analyses among patients with PDAC were performed using Kaplan-Meier with log rank test.

Results: Exosomes were successfully isolated from the pancreatic duct fluid of all patients. The mean concentration of exosomes isolated was 5.2 +/- 1 x 108 particles/mL, with a mean vesicle size of 186 +/- 4nm. The presence of exosomes was confirmed for each patient using TEM. Characteristic exosome surface markers, including tetraspanins CD9 and CD63, were identified. No differences in either size or concentration of EVs were detected when patients were stratified by diagnosis, but significant differences in protein composition were found. Laminin β3, Laminin C2, Tenascin C, and matrix metalloproteinases 1 and 7 were significantly overexpressed among patients with PDAC as compared to patients with IPMN and benign diseases (Table). Similarly, CEACAM1, CEACAM5, and CEACAM6 were significantly overexpressed among patients with PDAC (p=<0.001, 0.025, and 0.030, respectively). Among patients with PDAC, patients having expression of all three CEACAMs (n=6) had significantly worse overall and recurrence-free survival compared to patients lacking any of the CEACAMs (n=7) (p=0.01 and 0.005, respectively).

Conclusion: Exosomes can be isolated from pancreatic duct fluid. Exosomal ECM-associated proteins and CEACAMs appear to possess diagnostic and prognostic utility for patients with PDAC. Further evaluation of the mechanistic significance of these proteins in exosomes is warranted.

 

24.09 Tumor Homing Peptide Display on AAV2 Surface Enhances Viral Targeting of Pancreatic Cancer

Posted on January 25, 2017

S. Liu1, J. Lee3, J. Yu1, R. Sanchez1, M. Van Dam3, E. Rozengurt2, F. Brunicardi1  1David Geffen School Of Medicine, University Of California At Los Angeles,Department Of Surgery,Los Angeles, CA, USA 2David Geffen School Of Medicine, University Of California At Los Angeles,Division Of Digestive Disease,Los Angeles, CA, USA 3David Geffen School Of Medicine, University Of California At Los Angeles,Crump Institute For Molecular Imaging,Los Angeles, CA, USA

Background: Recombinant adeno-associated virus type 2 (AAV2) vectors transduce a wide variety of tissues in vivo and provide long-term gene expression with minimal immune responses and no pathological responses to the host, thus making AAV2 an attractive vector for gene delivery in vivo. However, its broad host range limits its usefulness in which transgene expression needs to be limited to a specific organ or cell type, such as cancer. Furthermore, low efficiency of wild type AAV2 (AAV2wt) transfection of pancreatic cancer (PDAC) cells further limits its effectiveness. In this study, we explored the possibility of directing rAAV2 transduction by incorporating a panel of tumor homing peptides to target human PDAC cells and tumors in mice.

Methods: 18 tumor homing peptides (TumorHoPe) were selected from literature and TumorHoPe database for AAV2 capsid display. Synthetic oligos were inserted into the rAAV2 capsid at R588. scAAV-eGFP and scAAV2-Gaussia Luciferase (GLuc) reporters were used. Real time-PCR was used to quantify the AAV2 titers. AAV infection of human PDAC cell lines (PANC-1, Mia PaCa2, Capan-2 and AsPC-1) and benign HPDE cells were performed, followed by peptide competition assays. GLuc secretory activity was determined by bioluminescence assay. In vivo imaging was performed in PANC-1 subQ xenograft using systemic delivery of AAV2TumorHoPe-GLuc imaging vector via tail vein.

Results: 18 TumorHoPe displayed AAV2  and one mutated AAV2 were created (AAV2TumorHoPe). The greatest viral infection was obtained in PANC-1 by AAV2RGD (CDCRGDCFC), Mia PaCa2 by AAV2RGR (CRGRRST), Capan-2 by AAV2RGD and AsPC-1 by AAV2LGL (RGDLGLS), with no infection of benign HPDE cells (Fig. Upper). There was no significant GFP expression in AAV2WT transfections of PDAC cells. To further identify whether the TumorHoPe display virus targeted PDAC cells and enhancing the infection efficiency, competition assays were performed on each cells infected by virus in the presence of excess exogenous TumorHoPe peptides; infection efficiency dropped to 42.5%, 34.3%, 25.6% and 36.9% in PANC-1, Mia PaCa2, Capan-2 and AsPC-1 cells, respectively. Systemic delivery of AAV2TumorHoPe-GLuc imaging vector in PANC-1 tumor model resulted in highly specific imaging of tumors without toxicity (Fig. Bottom A), whereas control AAV2 vector showed only non-specific transfection with none seen in the PANC1 tumor (Fig. Bottom B).

Conclusion: This study demonstrates that TumorHoPe display on AAV2 surface enhances viral targeting of PDAC cells and tumor, but not benign HPDE cells. The unique type of TumorHoPe on the AAV2 surface contributes to specific efficiency in PDAC cells. These preclinical data suggest that TumorHoPe AAV2 gene delivery could be used for targeted imaging of PDAC.

24.08 Cancer Stemness In Bone Marrow Micrometastases Of Human Breast Cancer

Posted on January 25, 2017

M. C. Kuo1, A. N. Kothari1, Z. Mi1  1Loyola University Chicago Stritch School Of Medicine,Surgery,Maywood, IL, USA

Introduction: Cancer cells metastasize to the bone marrow (BM) to create the pre-metastatic niche. Cancer stemness (or expression of stem cell characteristics) is regulated by the tumor microenvironment (TME) and associated with self renewal, clonal maintenance, and poor clinical outcomes. Osteopontin (OPN) induces mesenchymal stem cells (MSC) in the TME adopt a cancer-associated fibroblast (CAF) phenotype to potentiate cancer growth and metastasis. The mechanisms by which cancer cells and TME regulate stemness in the BM pre-metastatic niche are unknown.  

Methods: Human breast cancer cell lines, MB231-Luc (OPN+) and MCF7-Luc (OPN-), were used in an orthotopic murine xenograft model. 6-week-old female NOD scid mice were implanted with 2×106 tumor cells in the presence and absence of human MSC-GFP cells in the R4 position of the mammary fat pad. In selected instances,  MCF7-Luc transfected to express OPN (MCF7[lvOPN]) were co-implanted instead of MCF7-Luc. OPN aptamer (APT), which blocks and inactivates extracellular OPN, and/or inactive mutant APT (muAPT) were utilized. After 8 weeks, animals were sacrificed and femoral BM isolated. Stem cell markers, Sox2, Oct4 and Nanog, and CAF markers, SMA and vimentin, were measured by RT-PCR using human specific primers and normalized to Luc and GFP mRNA, respectively. Relative cell number was determined by titrating cell number to Ct value of GFP or Luc in vitro. Each treatment had 3-5 mice. Statistical analysis was performed using Student’s t-test; p-values < 0.05 were considered significant.

Results: 
In BM from MB231+MSC, expression of Sox2, Oct4 and Nanog was 4x greater than MB231 alone.(p<0.05) Administration of APT to block OPN decreased Sox2, Oct4 and Nanog to levels equivalent to MB231 alone and MB231+MuAPT. Total MB231 cell numbers were not different. In contrast, although CAF markers were not different among all treatments, CAF numbers in BM were 15-fold greater in MB231+MSC vs MB231 alone and MB231+MSC+APT.(p<0.05) In parallel gain-of-function studies, MCF7-Luc (which do not express OPN) co-implanted with MSC did not express increased stem cell markers in BM when compared to MCF-Luc alone. In contrast, using MCF7 expressing OPN, MCF7[lvOPN]+MSC had 5x greater Sox2, Oct4 and Nanog expression.(p<0.05) APT with MCF7[vOPN]+MSC ablated the increase in stem cell markers. Again, CAF markers were unchanged among treatments, but CAF numbers were 70x higher in MCF7[lvOPN] + MSC vs MCF7+MSC and MCF7[lvOPN]+MSC+APT.(p<0.05)

Conclusion:   

In this xenograft model of human breast cancer, our results indicate: 1) both breast cancer cells and accompanying CAF from the primary site metastasize to the BM to form the pre-metastatic niche, 2) tumor-derived OPN mediates CAF migration to the BM, and 3) cancer cells exhibit significantly increased stemness in the presence of CAF in the BM. We conclude that OPN-dependent migration of CAF is required for increased cancer cell stemness in the BM pre-metastatic niche.   

24.07 Nucleotide Depletion by Autophagy Inhibition Sensitizes Kras-driven PDAC to Replication Stress

Posted on January 25, 2017

I. A. Elliott1, J. L. Williams2, R. Ghukasyan1, C. C. Matsumura1, N. Wu1, L. Li1, W. Kim3, S. Poddar3, E. R. Abt3, A. M. Dann1, H. DeRubertis1, D. Braas4, T. M. Le3, C. G. Radu3, T. R. Donahue1,3  1University Of California – Los Angeles,Department Of Surgery,Los Angeles, CA, USA 2Harbor-UCLA Medical Center,Department of Surgery,Torrance, CA, USA 3University Of California – Los Angeles,Department Of Molecular And Medical Pharmacology,Los Angeles, CA, USA 4University Of California – Los Angeles,Metabolomics Center,Los Angeles, CA, USA

Introduction:
Autophagy is a critical source of nucleotides, which are rate-limiting for Ras-driven cancer cell proliferation. Genetically disabling autophagy impairs energy and redox homeostasis by depletion of nucleotide pools, and sensitizes cancer cells to radiation. We hypothesized that the autophagy inhibitor chloroquine (CQ) would deplete deoxyribonucleotide triphosphate (dNTP) availability for incorporation into DNA, thus sensitizing cancer cells to replication stress and revealing a targetable liability in Ras-driven pancreatic ductal adenocarcinoma. 

Methods:
Human (MiaPaCa-2) and murine (KPC) PDAC cells were treated with CQ +/- the replication stress response (RSR) inhibitor VE822. S-phase cells were labeled by pulsing with 5’-ethynyl-2’-deoxyuridine (EdU) and cell cycle progression measured by flow cytometry. Newly synthesized dNTP incorporation into DNA was measured via [13C6]glucose labeling and our novel liquid chromatography mass spectrometry (LC-MS) platform. Global metabolomics analyses were performed by detection of relative amounts of metabolites using LC-MS. Western blots (WB) were done on cell lysates. Cell viability was measured by Cell-Titer-Glo assay. MiaPaCa-2 or KPC cells were injected s.q. in the flanks of NSG or C57BL/6 mice. Mice were treated 5x/week with CQ+/-VE822 (60mg/kg p.o.). Immunohistochemistry (IHC) was performed on explanted KPC tumors after 3 days of CQ+VE822.

Results:
We found that CQ caused S-phase arrest in MiaPaCa-2, and impaired incorporation of newly synthesized dNTPs into replicating DNA(Fig.1a), indicating nucleotide pool insufficiency. Global metabolomic profiling of MiaPaCa-2 revealed that this was due to depletion of purine and pyrimidine substrates for dN synthesis under CQ, and this was exacerbated by addition of the RSR inhibitor VE822(Fig.1b). We then tested the impact of this nucleotide depletion on cell fate; CQ+VE822 treatment led to synergistic induction of DNA damage (reflected by pH2A.X WB), and decreased viability of MiaPaCa-2 and KPC in vitro (Fig.1c). We also observed increased IHC staining for pH2A.X in KPC tumors and impaired growth of MiaPaCa-2 tumors after CQ+VE822 treatment in vivo. Finally, we confirmed the ability of CQ to induce oxidative stress as indicated by HO-1 levels on WB. Management of redox balance and DNA damage are critical to recovery from radiation; accordingly, we found that CQ+VE822 profoundly impaired survival after radiation of MiaPaCa-2 and KPC cells.

Conclusion:
Pharmacologic inhibition of autophagy by CQ impairs incorporation of dNTPs into DNA by depleting substrates for nucleotide biosynthesis. When combined with an RSR inhibitor, this leads to induction of DNA damage and synthetic lethality in PDAC cells in vitro and in vivo.

24.06 Functional Screens to Identify Gene Drivers of Pancreatic Cancer

Posted on January 25, 2017

N. Villafane1,2, Y. Tsang2, K. Scott2  1Baylor College Of Medicine,Michael E. DeBakey Department Of Surgery,Houston, TX, USA 2Baylor College Of Medicine,Department Of Molecular And Human Genetics,Houston, TX, USA

Introduction:  Pancreatic ductal adenocarcinoma (PDAC) is not one of the most prevalent cancers but is by far one of the deadliest given its 5-year survival rate of <5%. The majority of PDAC tumors harbor activating mutations in the KRAS oncogene, a therapeutically undruggable and initiating “driver” event found in >90% of early-stage lesions. Following activation of KRAS, PDAC acquires other driver mutations that include inactivating events in tumor suppressor genes and numerous low frequency gene mutations, amplifications, and deletions likely required for KRAS-mediated tumor progression. Our objective is to determine which of these events are functional drivers or “driver effectors” required for KRAS biology, as targeting these events or their activated pathways offers hope of improving patient outcomes.

Methods:  We used a highly-vetted list of 250 amplified gene candidates (copy number > 5; GISTIC q-value < 0.075) that was previously defined at our lab by analyzing The Cancer Genome Atlas (TCGA) and the International Cancer Genome Consortium (ICGC) PDAC datasets (derived from 218 tumors). We engineered pooled gene libraries consisting of DNA-barcoded open reading frame (ORF) clones encoding the amplified driver candidates. We used a lentiviral vector to deliver these ORFs to human pancreatic ductal epithelial (HPDE) cells engineered with a doxycycline-inducible KRASG12D allele in order to identify ORFs that drive tumor growth and metastasis in mice maintained on or off doxycycline diet to control for KRASG12D activation. 

Results: Our screens identified SEMA6A, a transmembrane protein member of the axon guidance pathway, as a high priority PDAC target. Our data indicate that SEMA6A expression is increased through KRASG12D– mediated down-regulation of miR145. This suggests that the SEMA6A/plexinA2/A4 signaling pathway may be hyper-activated in the presence of oncogenic KRAS, resulting in cell cycle progression and tumor growth. 

Conclusion: Screening of candidate driver pools in our laboratory has revealed potent drivers of PDAC, such as SEMA6A. High priority PDAC drivers resulting from these functional screens will lead to the discovery of detection biomarkers and potential targets for drug development efforts needed for patients suffering from this disease. 

 

24.03 Minnelide Decreases Tumor Burden in Murine Models of Malignant Melanoma

Posted on January 25, 2017

V. Sethi1, B. Giri1, B. Garg1, S. Modi1, S. Banerjee1, A. Saluja1, V. Dudeja1  1University Of Miami,Surgical Oncology, Department Of Surgery,Miami, FL, USA

Introduction:  Malignant melanoma is the most lethal of all skin cancers. Modern immune checkpoint inhibitors used to treat melanoma show response in < 50% of patients and are associated with various immune-related adverse events. Therefore, more effective and safer therapeutic strategies are urgently needed. Triptolide, a diterpenoid tri-epoxide and its water soluble pro-drug Minnelide, have been previously shown by our group to be highly effective in treating various solid tumors. Minnelide is currently undergoing phase I clinical trial for treating advanced GI malignancies. This study aims to evaluate the effect of Triptolide in vitro and of Minnelide in vivo against melanoma.

Methods:

In-vitro, B16-F10 cells, a highly metastatic murine melanoma cell line, were treated with various concentrations of Triptolide (0-75 nM). Cancer cell viability at 24 and 48 hours was measured using a formazan dye based colorimetric assay. Apoptosis at 24 and 48 hours was measured by quantifying Caspase 3 and 7 activities in a luminescence based assay. Effect of Triptolide on invasion and migration of melanoma cells across a chemotactic gradient was measured by using Transwells (having PET membrane with 8.0 μm  pore size) coated with and without Matrigel respectively.

 

In vivo, we treated mice with 0.21 mg/kg/day of Minnelide, a clinically relevant, well tolerated dose derived from phase I trial data. In the first model, 7.5 x104 B16-F10 cells were injected subcutaneously in mice and these were randomized into Minnelide and saline arms on day 3. Tumor volume was measured at endpoint (Day 19). In the second model, 1.5 x 105  cells were injected through an intrasplenic injection in a hepatic metastases model of melanoma. Randomization was done and treatment was started on day 5. Liver metastases were measured by quantifying liver volume and liver weight/body weight ratio at endpoint (Day 21).

Results: In vitro, Triptolide caused a significant decrease in viability and increase in apoptosis of B16-F10 cells in a dose dependent manner. Invasion and migration of melanoma cells was also significantly inhibited. In vivo, Minnelide treatment caused a significant decrease in volumes of subcutaneous tumors (mean tumor volume ± SEM was 814 ± 97.5mm3 in saline arm vs 467 ± 79.3mm3 in Minnelide arm; p < 0.05) and a drastic decrease in hepatic metastases burden as measured by liver volumes and liver weight/body weight ratio (mean liver volume ± SEM was 3.8 ± 0.4 ml in saline arm vs 2.2 ± 0.4 ml in Minnelide arm; p < 0.05)

Conclusion: Phase II clinical trial of Minnelide for treating GI cancers is starting soon. Minnelide has a potential to become an effective therapeutic option for treating advanced melanoma too.
 

24.04 High Depth Hybrid Capture Exome Sequencing Of Somatic Mutations For Personalized Cancer Therapy

Posted on January 25, 2017

N. Paez Arango1, L. Brusco2, K. R. Shaw2, K. Chen3, K. Eterovic4, V. Holla2, A. Johnson2, B. Litzenburger2, Y. B. Khotskaya2, N. Sanchez2, A. Bailey2, X. Zheng4, C. Horombe2, S. Kopetz5, C. Farhangfar6, M. Routbort7, R. Broaddus8, E. S. Bernstam9, J. Mendelsohn2, G. B. Mills2,4, F. Meric-Bernstam1,2,10  1University Of Texas MD Anderson Cancer Center,Department Of Surgical Oncology,Houston, TX, USA 2Univerisity Of Texas MD Anderson Cancer Center,Khalifa Institute For Personalized Cancer Therapy,Houston, TX, USA 3University Of Texas MD Anderson Cancer Center,Department Of Bioinformatics And Computational Biology,Houston, TX, USA 4University Of Texas MD Anderson Cancer Center,Department Of Systems Biology,Houston, TX, USA 5University Of Texas MD Anderson Cancer Center,Gastrointestinal Medical Oncology,Houston, TX, USA 6Carolinas Healthcare System,Levine Cancer Institute,Charlotte, NC, USA 7University Of Texas MD Anderson Cancer Center,Department Of Hematopathology,Houston, TX, USA 8University Of Texas MD Anderson Cancer Center,Department Of Pathology,Houston, TX, USA 9University Of Texas Health Science Center At Houston,School Of Biomedical Informatics,Houston, TX, USA 10University Of Texas MD Anderson Cancer Center,Department Of Investigational Cancer Therapeutics,Houston, TX, USA

Introduction: The increasing availability of molecular therapeutics targeting specific genomic alterations has led to an increased interest in the use of next-generation sequencing to help identify patients that could potentially benefit from targeted clinical trials and personalized cancer therapies.  Expanding the number of genes and incorporating copy number testing could potentially aid in identifying a higher number of genomic alterations in clinically actionable genes, and thus identify a greater number of patients that could likely benefit from targeted therapy. The purpose of this study was to look at patients who underwent routine genomic profiling under standard hotspot analysis and then using a deep targeted sequencing platform in the research setting, assess the added value to having a more comprehensive genomic test.

Methods: 1200 patients with advanced cancer underwent routine tumor profiling for characterization of common mutations on an 11, 46, or 50 gene hotspot sequencing platform in the Clinical Laboratory Improvement Amendments (CLIA) certified environment, as well as on 201 gene deep targeted sequencing platform in the research setting. 120 genes were considered actionable based on their potential to be targeted with available therapies.

Results:Five hundred and twenty eight patients (44%) had at least one mutation detected in a potentially actionable gene using standard hotspot genomic testing under a limited gene panel, with a total of 644 mutations detected. Under deep targeted sequencing in the research setting, we identified 686 patients (57%) that had at least one somatic mutation in a potentially actionable gene that was previously undetected in the standard testing, with a total of 2448 newly identified mutations. Additionally, we detected 654 (55%) patients with at least one potential copy number variation, with a total of 2784 potential copy number variations identified. A grand total of 5232 alterations in potentially actionable genes were found that would have been otherwise undetected (see table).

Conclusion:Expanded cancer gene sequencing for profiling somatic mutations identifies a greater number of alterations in potentially actionable genes that could have a clinical significance, allowing for the identification of an increased number of targets for personalized cancer therapies.

 

24.02 Targeting Neutrophil Extracellular Traps (NETs) prevents HCC in a mouse model of NASH

Posted on January 25, 2017

D. J. Van Der Windt1, V. Sud1, P. Varley1, J. Goswami1, H. Yazdani1, P. Loughran3, M. I. Minervini4, H. Huang1,2, R. L. Simmons1, A. Tsung1  1University Of Pittsburgh Medical Center,Surgery,Pittsburgh, PA, USA 2Huazhong University Of Science And Technology,Surgery,Wuhan, HUBEI, China 3Center For Biologic Imaging,Cell Biology,Pittsburgh, PA, USA 4University Of Pittsburgh Medical Center,Pathology,Pittsburgh, PA, USA

Introduction:  Non-alcoholic steatohepatitis (NASH) is a progressive, inflammatory form of non-alcoholic fatty liver disease. Its prevalence is rapidly increasing, resulting in more patients with NASH-related hepatocellular carcinoma (HCC). Surgical resection and liver transplantation are the only curative options for HCC. As a consequence, there is an increased number of patients with NASH undergoing liver resection, and those patients have higher risks of post-operative complications. The inflammatory pathogenesis of NASH progression to HCC remains incompletely understood. Neutrophils are increasingly recognized as cells that regulate the inflammatory environment and can affect the progression of inflammatory cancers. The formation of neutrophil extracellular traps (NETs) has been discovered as a novel neutrophil function, during which chromatin structures with pro-inflammatory proteins are exposed to the extracellular environment. Here we sought to evaluate the role of NETs in the inflammatory development of NASH and the progression to HCC.

Methods:  Wild type (WT) and peptidylarginine deiminase 4 knockout (PAD4 KO) mice (that are unable to form NETs) were exposed to a model of NASH-related HCC, and compared to healthy control (HC) mice. NASH was established by injection of streptozotocin to induce diabetes (200ug I.P., 1 dose within 5d from birth), and feeding of a high fat diet starting at age 3 weeks. WT and PAD4KO mice were sacrificed at various time points for analysis of liver inflammation and the formation of HCC. To test the efficacy of a clinically available treatment strategy for eliminating NETs, WT mice were treated with DNase1 (100U I.P. 3x/wk).

Results: WT male mice developed NASH as evidenced by progressive increase in NASH activity score (NAS), and developed multiple liver tumors by age 20 weeks. Neutrophil infiltration was seen as early as age 6 weeks as evidenced by flow cytometry (12.6±0.5% in WT vs. 4.0±0.2% of CD45+ liver non-parenchymal cells in HC, p<0.001). NETs were observed on liver immunofluorescence. At age 8 weeks, WT mice had elevated serum ALT and IL-6 levels, indicating liver injury and inflammation. ALT and IL-6 levels were reduced to baseline in PAD4 KO mice and by DNase1 treatment, respectively (ALT [U/L]: WT 166±35, PAD4 KO 102±4, DNase1 73±16, p<0.05. IL-6 [pg/mL]: WT 23±7.9, PAD4 KO 3.9±0.7, DNase1 5.8±1.2, p<0.05). Most importantly, PAD4KO mice developed significantly fewer tumors at age 20wks than WT mice (1.7±0.6 vs. 6.3±1.8, p<0.05).

Conclusion: Elimination of NETs reduces inflammation and prevents progression to HCC in murine NASH. Clinically, inhibition of NETs in patients with NASH may reduce their risk of developing HCC.

 

24.01 Acceleration of Pancreatic Cancer Development in KrasG12D Mice Deficient in Hormone Sensitive Lipase

Posted on January 25, 2017

M. Xu1, X. Jung1, A. Moro1, C. Chou1, H. Chang1, A. P. Stark1, A. Schmidt1, Y. Chen1, J. King1, O. J. Hines1, G. Eibl1  1University Of California – Los Angeles,Surgery,Los Angeles, CA, USA

Introduction: Obesity has been associated with multiple types of cancer. Our previous work has shown that a high-fat, high-calorie diet (HFCD) leads to obesity in conditional KrasG12D (KC) mice with inflammation of the visceral adipose tissue (VAT) and acceleration of pancreatic intraepithelial neoplasia (PanIN) development. In this study, we investigated whether diet-independent inflammation in the visceral adipose tissue is sufficient to accelerate pancreatic cancer (PDAC) development using hormone-sensitive lipase (hsl) deficient mice. Hsl is one of the enzymes involved in the hydrolysis of stored triglycerides. Non-obese hsl deficient mice are characterized by a significant inflammation in visceral adipose tissue.  

Methods: KC mice were crossed into an hsl deficient background to generate KC;hsl+/+ and KC;hsl-/- mice. Mice were fed regular chow for six months and body weight recorded weekly. At sacrifice tissues were harvested for histological and immunohistochemical analysis.

Results: Over the study period of six months KC;hsl+/+ and KC;hsl-/- mice displayed similar weight gain. KC;hsl-/- males had an average weight of 29.5±4.5g at sacrifice, whereas KC;hsl+/+ males weighed 29.7±3.2g. KC;hsl-/- and KC;hsl+/+ females had an average weight of 25.2±1.8g and 24.3±1.8g, respectively. Twenty percent of the KC;hsl-/- mice developed invasive PDAC (n=20), whereas none of the KC;hsl+/+ mice developed PDAC (n=23; p<0.05). Interestingly, all KC;hsl-/- mice with PDAC were male. Histological and immune-histochemical analyses revealed that compared to KC;hsl+/+ mice KC;hsl-/- mice had significantly increased VAT inflammation as measured by the number of crown-like structures per high power field (hpf) (4.6±3.0 in KC;hsl-/- vs. 0.2±0.6 in KC;hsl+/+; p<0.01), enhanced activation of pancreatic stellate cells assessed by counting the cells per hpf positive for alpha-smooth muscle actin (44.5±9.0 in KC;hsl-/- vs. 15.8±5.2 in KC;hsl+/+; p<0.01) and more F4/80 positive macrophages per hpf in the pancreas (18.8±9.1 in KC;hsl-/- vs. 11.5±6.7 in KC;hsl+/+; p<0.01). PanIN lesions in KC;hsl-/- mice showed a higher rate of proliferation (per hpf) as measured by Ki67 staining (14.9±6.4 in KC;hsl-/- vs. 5.5±3.6 in KC;hsl+/+; p<0.01).

Conclusions: KC;hsl-/- mice had a significantly more robust VAT and pancreatic inflammation and accelerated PDAC development compared to KC;hsl+/+ mice. Our results suggest that VAT inflammation plays an important role in PDAC development. 

23.10 Topically Applied Stem Cell Conditioned Media Accelerates Donor Site Healing and Vascular Response

Posted on January 25, 2017

J. H. Chen1,2, M. M. Kebede2, S. E. Epstein3, T. E. Travis1,2, L. T. Moffatt2, J. W. Shupp1,2  1Washington Hospital Center,Department Of Surgery, The Burn Center,Washington, DC, USA 2MedStar Health Research Institute,Firefighters’ Burn And Surgical Research Laboratory,Washington, DC, USA 3MedStar Health Research Institute,Cardiovascular Research Institute,Washington, DC, USA

Introduction: Severe burn injuries often require grafting, resulting in a second wound, the donor site. Patients with large burns may have limited options for harvesting. Decreasing time to re-epithelialization would allow for more frequent re-harvesting and therefore earlier wound closure. The regenerative properties of stem cells are thought to be secondary to secreted products including various cytokines and growth factors. We have developed a gel-matrix conditioned-media (CM)-based product from stem cells.  Previous studies using a mouse model of limb ischemia-reperfusion demonstrated potent angiogenic responses, including greater total vessels present in CM-treated vs. VEGF-treated animals. We hypothesized that similar effects may be seen in donor sites treated with CM. In this study, we use an established model of donor site healing to investigate these effects.   

Methods: Stem cell-derived molecules, obtained by concentrating stem cell growth media supernatants, were embedded in a biodegradable polymer and formulated to be gradually released. Wounds measuring 3" x 3" were created on Duroc pigs with a dermatome (0.06” depth) in triplicate on the flank of five animals. These wounds were treated with conventional dressings to protect from infection. Digital photos, laser Doppler imaging (LDI), and biopsies were taken pre- and post- wounding and every 48 hours until re-epithelialized (16 days). Subsequently, a set of identical wounds were created on the contralateral flank of each animal, treated with CM dressing, and assessed using the same metrics. Formalin-fixed biopsies were paraffin-embedded and stained for VEGF. RNA was isolated and qRT-PCR was used to examine differential gene expression of VEGF and other markers of wound healing over time. 

Results:Using paired comparisons of sham and treated wounds on each animal (n = 5 pigs and 15 lesion pairs) CM treatment significantly accelerated healing. In the treatment arm, time to complete healing of 50% of the wound surface area was decreased by 12% (from 8.3 to 7.1 days; p = 0.04) and final degree of healing was increased by 14% (from 77 to 88%; p = 0.002). Three days after treatment with CM, IHC revealed elevated VEGF levels that exceeded those of sham wounds. By day seven, treated wounds continued to have higher VEGF levels compared to sham. Wound perfusion as measured by LDI was significantly higher in treated wounds than in control wounds by day four (784 ± 62 vs 680 ± 46 PU; p = 0.04). Differential expression of growth factors was assessed using qRT-PCR.

Conclusion:We conclude that topical application of conditioned media, the secreted products of stem cells, is a novel and potentially potent means for accelerating wound healing of burn-related donor sites by promoting VEGF-induced angiogenesis. Further study is warranted to elucidate the key components of conditioned media responsible for this effect and to assess other metrics of neovascularization and angiogenesis.

 

23.09 Stored Erythrocytes Alter Endothelial Cell Tissue Factor Expression in Mice

Posted on January 25, 2017

A. D. Jung1, Y. Kim1, W. Abplanalp1, A. T. Makley1, M. D. Goodman1, T. A. Pritts1  1University Of Cincinnati,Surgery,Cincinnati, OH, USA

Introduction:  Transfusion with packed red blood cells (pRBCs) remains essential for resuscitation after hemorrhage. During storage, pRBCs undergo a series of physical and biochemical changes that are collectively known as the red blood cell storage lesion. Previous work from our laboratory shows that this is associated with harmful effects in the transfusion recipient, including alterations in the coagulation cascade. Tissue factor, a constitutively expressed cell-surface glycoprotein, plays a key role in initiating the coagulation cascade, but the effect of transfusion of stored pRBCs on endothelial tissue factor expression is unknown. We hypothesized that aged pRBCs alter endothelial cell tissue factor expression.

Methods:  Blood was harvested from murine donors, then separated into components. Leukoreduced and platelet-depleted pRBC units were stored in additive solution-3 at 4°C for 14 days. Microparticles were isolated from aged pRBC units by ultracentrifugation. In a simulated in vitro transfusion model, murine lung endothelial cells were incubated with fresh or 14-day-old erythrocytes suspended in media at 37°C for 1 h and endothelial cell tissue factor expression was analyzed via immunofluorescence. In a separate experiment, endothelial cells were exposed to aged vs fresh erythrocytes, then tissue factor expression was analyzed by Western blot. The third experiment involved exposing endothelial cells to media or isolated microparticles from aged pRBC units, and soluble tissue factor was determined using ELISA at 0.5, 6, 12, and 24h. 

Results: Whole cell immunofluorescent analysis of endothelial cells exposed to aged pRBCs demonstrated significant decreases in tissue factor expression vs baseline (11.7±0.7 vs 16.9±0.6 RFI, p<0.05). Western Blot analysis also demonstrated a significant decrease in tissue factor expression as compared to baseline. Treatment of endothelial cells with microparticles isolated from stored PRBC units resulted in a significant attenuation in tissue factor concentration at 30 minutes (4.3±0.4 vs 6.2±0.8, p<0.05), 6 hours (3.63±0.3 vs 5.5±0.6, p<0.05), and 24 hours (2.8±0.6 vs 4.7±0.5, p<0.05) as determined by ELISA.

Conclusion: Treatment of endothelial cells with aged pRBCs attenuates expression of tissue factor. Additional experiments indicate that this process is driven by red blood cell microparticles. Together, these data provide further evidence that transfusion with stored pRBC units causes endothelial cell dysfunction.  

 

23.08 CitH3: a Long-Lasting Blood Biomarker for Diagnosis and Treatment of Endotoxic Shock in Mice

Posted on January 25, 2017

B. Pan1, B. Liu1, E. Chen1, Y. Wang2, Y. Liang1, Y. Li1, H. B. Alam1  1University Of Michigan Hospital,General Surgery/Surgery/Medical School,Ann Arbor, MI, USA 2Penn State University,Department Of Biochemistry And Molecular Biology,University Park, PA, USA

Introduction: Usefulness of many biomarkers is limited by their short half-life in the circulation. We previously showed that citrullinated histone H3 (CitH3) is a potential diagnostic biomarker and mediator in sepsis, but it is unclear if it persists in blood long enough to be clinically useful. The present study was designed to test the hypothesis that circulating CitH3 is a long-lasting biomarker, and that treatment with YW3-56, a peptidylarginine deiminase-4 (PAD4) inhibitor, can improve survival in a lethal lipopolysaccharide (LPS) induced shock model while attenuating blood levels of CitH3.

Methods: Three experiments were carried out. In experiment I, CitH3 specific enzyme-linked immunosorbent assay (ELISA) was established by our laboratory. A 96-well plate was coated with CitH3 monoclonal antibody overnight, incubated with mouse serum for 2 hours (h) followed by incubation with CitH3 polyclonal antibody and HRP-linked anti-rabbit antibody.  Synthesized CitH3 was used to generate a standard curve (0–20 ng/ml). In experiment II, C57BL/6J mice were randomized to the following 3 groups with intraperitoneal injection of different reagents (n=7/group): (1) Dimethyl sulfoxide (DMSO), (2) LPS (35 mg/kg) +DMSO, and (3) LPS+YW3-56 (5 mg/kg) dissolved in DMSO (post LPS). Survival was monitored for 10 days. In experiment III, mice were subjected to the same treatment as the experiment II, and blood samples were collected at 0, 0.5, 3, 12, and 24 h after treatment. Plasma CitH3 were measured by both Western blot and ELISA. ANOVA was used for multiple comparisons and Kaplan-Meier curves for survival.

Results: All mice in the DMSO group survived. The LPS injection was universally lethal, with the majority of deaths within 24 h. YW3-56 treatment significantly improved survival compared to the LPS alone group (p<0.0001). No circulating CitH3 was found in DMSO group. LPS injection was associated with elevated plasma CitH3 at 0.5, 3, 12, and 24 h (87pg/ml, 245 pg/ml, 1180 pg/ml, and 318 pg/ml, respectively) with peak level at the 12-h time point. YW3-56 treatment significantly attenuated LPS-induced CitH3 levels in blood (13pg/ml, 73pg/ml, 183pg/ml, and 3pg/ml, respectively), compared to the untreated LPS group.  The elevated CitH3 was detectible by Western blot and ELISA, but only ELISA could accurately quantify the concentration of blood CitH3 (Figure 1).  

Conclusion: Our study demonstrated for the first time that (1) the CitH3 specific ELISA developed by us is a more reliable and accurate quantitative method than Western blot; (2) CitH3 protein in the peripheral blood is an ideal biomarker for sepsis because it responds to an insult early (0.5 h), lasts long, and is responsive to therapeutic interventions.

 

23.07 Improved Sepsis Survival in JAM-A-/- Mice is Associated with Enhanced Intestinal Mucosal Immunity

Posted on January 25, 2017

N. J. Klingensmith1, K. Fay1, C. Chen1, J. Lyons1, C. M. Coopersmith1  1Emory University School Of Medicine,Surgery,Atlanta, GA, USA

Background – Sepsis is associated with intestinal hyperpermeability which carries significant mortality. This permeability is augmented in mice lacking Junctional Adhesion Molecule A (JAM-A), an enterocyte tight junction protein. Our unpublished data reveal JAM-A-/- mice paradoxically have decreased systemic bacterial burden and improved sepsis survival following cecal ligation and puncture (CLP). At baseline, the intestinal epithelium, local immune system, and gut microbiome all interact to maintain homeostasis. How the immune system and microbiome respond to intestinal hyperpermeability in septic JAM-A-/- animals may provide a mechanism for their improved sepsis survival.

 

Methods – 6-12 week old male and female C57BL/6 (WT) and JAM-A-/- mice underwent CLP to induce sepsis and were sacrificed 24 hours later for functional studies. Mesenteric lymph nodes (MLN) and small bowel Peyer’s Patches (PP) were collected and B220+ cells were stained and counted by flow cytometry. Serum and small bowel homogenate IgA were measured by ELISA. Intestinal microbiome taxonomy were identified from fecal samples by 16S rRNA amplicon pyrosequencing followed by QIIME pipeline analysis. Statistical analysis was performed using a Student’s t-test or Mann-Whittney test depending on Gaussian distribution and a p-value of <0.05 was considered to be statistically significant.

 

Results-  Septic JAM-A-/- mice had a greater frequency of B cells compared to septic WT controls in their MLN (mean frequency, 25.2±2.3% vs 17.2±1.4%, p=0.009, n=10) and PP (mean frequency, 19.0±1.7% vs 12.8±2.2%, p=0.04, n=7). IgA within the intestine, a product of B cells, was elevated in septic JAM-A-/- animals compared to septic WT controls (mean, 241.7±37.9 vs 131.8±19.8 ng/mg of gut tissue, p=0.02, n=7-10) suggesting prior antigenic exposure. Furthermore, serum IgA concentration in septic JAM-A-/- mice was increased compared to septic WT mice (mean, 178.1±43.9 vs 39.5±6.9 ng/mL, p=0.005, n=11) providing a potential mechanism for decreased systemic bacterial burden in septic JAM-A-/- mice. Principle coordinate analysis of  weighted and unweighted UniFrac distances of 16S rRNA sequencing revealed separation between septic JAM-A-/- mice and septic WT animals (p<0.001, n=5-7). 

 

Conclusion – JAM-A-/- mice have greater populations of B cells in their MLN and small bowel PP that produce more IgA. The increased gut IgA may result in a shift toward less pathogenic microflora which are potentially cleared more easily in the event of systemic contamination, such as in the setting of sepsis. Greater antigen presentation, local immune priming, and alteration of the microflora may be the source of improved sepsis survival in JAM-A-/- mice.

23.06 Effects of Valproic Acid and Molecular Hydrogen on Phenotypes of Microglia Treated with Hypoxia

Posted on January 25, 2017

X. Wu1, P. Chang1, R. Zeng1, Y. Deng1, Y. Zhao1, W. Chong1  1The First Affiliated Hospital Of China Medical University,Emergency,Shenyang, LIAONING, China

Introduction: Secondary hit is regarded as a lethal complication of traumatic brain injury (TBI), due to severe inflammation response induced by ischemia and hypoxia. Recently, it has been demonstrated that macrophages are mainly involved in inflammation with two different functional phenotype (or polarization): ones that encourage inflammation are called M1 macrophages, whereas those that decrease inflammation are called M2 macrophages. In TBI, microglia have the similar phenotype conversion as macrophages. Researches show that valproic acid (VPA) and molecular hydrogen (H2) attenuate cerebral swelling and manifest nervous protection in animal model of TBI in uncertain ways. The hypothesis in this study was that hypoxia enhances microglia polarization toward the pro-inflammatory phenotype. Both VPA and H2 inhibit hypoxia-induced inflammatory effects on microglia.

Methods: Mouse BV2 microglia were divided into 4 groups: Sham (no hypoxia, no VPA), hypoxia group, VPA group and H2 group. Hypoxia and VPA groups were exposed to hypoxic conditions (0.5% O2, 10% CO2, and 89.5% N2) at 37°C in the absence or presence of VPA (3mM) for 2 hours. H2 group were treated with H2 (0.5%O2, 2%H2, 10%CO2, 87.5%N2) for 1 hour under same hypoxic condition. Enzyme-Linked Immunosorbent Assay (ELISA), flow cytometry and Real-Time PCR were performed to assess protein and mRNA levels of phenotypic markers in microglia respectively.

Results: Hypoxia increased mRNA levels of M1 markers and reduced mRNA levels of M2 markers in microglia (P<0.05). In addition, the ratio between the mRNA levels of M1 markers and M2 markers were also increased by hypoxia (P<0.05). VPA reduced protein and mRNA levels of M1 markers (P<0.05), decreased the ratio between the mRNA levels of M1 markers and M2 markers of microglia (P<0.05) compared with hypoxia group. H2 not only decreased protein and mRNA levels of M1 markers (P<0.05) but also increased cytokine secretion and mRNA levels of M2 markers in microglia (P<0.05), and lowered the ratio between the mRNA levels of M1 markers and M2 markers of microglia as well in comparison with hypoxia group (P<0.05).

Conclusion: Hypoxia enhances microglia polarization toward the pro-inflammatory phenotype. Both VPA and H2 attenuate hypoxia-induced pro-inflammatory effects on microglia.

23.05 IL-33 DRIVES TYPE 2 CYTOKINES THROUGH ILC2 IN THE LUNGS AFTER HEMORRHAGIC SHOCK/TRAUMA (HS/T)

Posted on January 25, 2017

J. Xu1, J. Guardado1, R. Hoffman1, H. R. Turnquist1, T. R. Billiar1  1UPMC,Surgery Department,Pittsburgh, PA, USA

Introduction: The immunosuppression/immune dysregulation that follows severe injury includes a strong bias towards type 2 immune responses. These responses, which are characterized by the production of the type 2 cytokines (IL-4, IL-5 and IL-13), are typically thought to be the result of polarization of CD4+ T cells towards a Th2 phenotype. However, polarization of naïve T cells can take several days while elevations of type 2 cytokine production is often detected early after injury. This lead us to hypothesize that IL-33, an alarmin released early after injury, and type II innate lymphoid cells (ILC2) , a recently described population of T cell receptor negative innate lymphocyte cells (most abundant in the lungs), contribute to the early type 2 immune responses after injury.

Methods:  IL-33, and the type 2 cytokines (IL-4, IL-5 and IL-13) levels were detected in the plasma of blunt trauma patients. C57BL/6, IL-33-/- and ILC2 deficient mice were subjected to resuscitated hemorrhagic shock + bilateral lower extremity injury (HS/T), ILC2 percentage and activation were assessed by flow cytometry of lung leukocytes using IL-5 as a representative type 2 cytokine. Mice deficient in ILC2 were generated by reconstituting irradiated wild type (WT) with bone marrow from staggerer (Rorasg/sg) mice.

Results: Severely injured human blunt trauma patients (n=493, average ISS=20.2) exhibited early elevations in plasma IL-33 which correlated positively with increases in IL-4, IL-5 and IL-13. In C57BL/6 mice, HS/T led to a significant increase in lung IL-5 (measured by flow), IL-33 (measured by ELISA) and ILC2 percentage (CD45+Lineage–CD25+ CD127+CD90.2+ ST2+Sca-1+ CD117int) in the lungs at 6 hrs (p<0.05, n=4 for each time point). IL-5+ ILC2 were easily identified in the lungs by 6hrs. However, the increases in both ILC2 percentage and ILC2 IL-5 expression were absent in IL-33-/- mice subjected to HS/T (n=4/group). IL-5 could also be detected in PMN (Ly6G+CD11b+) in the lungs of injured mice but this was suppressed by IL-33-/- or ILC2 deletion. Using cultured PMN, we confirmed that exogenous IL-5 could lead to the increase of PMN IL-5 expression, measured by both flow and PCR (n=4/group).

Conclusion: These data show that early IL-33 elevations correlate with type 2 cytokines levels in blunt trauma patients. Reverse translation experiments in mice establishes that IL-33 drives ILC2 activation and type 2 cytokine production in the lungs of injured mice within 6 hrs. Furthermore, ILC2 derived IL-5 appears to upregulate IL-5 in PMNs thus amplifying the local type 2 response.  Combined these data show that the shift towards type 2 responses occurs rapidly after injury involve innate lymphoid cells responding to IL-33.
 

23.04 Post-Shock Mesenteric Lymph Exosomes are Key Mediators of Acute Lung Injury

Posted on January 25, 2017

M. Kojima1, T. Chan1, J. Gimenes1, B. Eliceiri1, A. Baird1, T. Costantini1, R. Coimbra1  1University Of California – San Diego,Division Of Trauma, Surgical Critical Care, Burns And Acute Care Surgery/Department Of Surgery,San Diego, CA, USA

Introduction: Acute lung injury (ALI) is a common cause of morbidity in patients following severe injury. Studies have shown that mesenteric lymph (ML) carries gut-derived inflammatory mediators to the lung and serves as the inciting event in ALI. Alveolar macrophages (AM), a lung resident macrophage, play a critical role in the development of ALI. We have recently demonstrated that exosomes, nano-sized extracellular vesicles, are present in ML and have the ability to trigger NF-κB activation in human monocytic cells in vitro. We hypothesized that exosomes in post-injury ML induce pro-inflammatory cytokine production in AM and contributes to post-injury ALI. 

Methods: Male rats underwent cannulation of the femoral artery, jugular vein and ML duct prior to trauma/hemorrhagic shock (T/HS; mean arterial pressure 35 mmHg for 60 min), followed by resuscitation with shed blood and two times normal saline. The ML was collected before hemorrhagic shock (pre-shock) and after T/HS (post-T/HS) for isolation of exosomes by differential centrifugation. In vitro AM were stimulated with exosomes harvested from pre-shock or post-T/HS ML for 6 or 12 hours for measurement of cytokine production by quantitative reverse transcription PCR (qRT-PCR). ML Exosomes from each experimental group (2 x 109 exosomes/g) were also intravenously injected into male naïve C57BL/6 mice to assess in vivo biologic activity. Lung injury was evaluated by measuring histologic lung injury, vascular permeability and immune cell recruitment in bronchoalveolar lavage (BAL) and lung tissue. 

Results:  Exosomes released into post-T/HS ML increased the gene expression of pro-inflammatory cytokines (TNF-a and CINC-1; see figure), NF-kB (5-fold increase at 6h vs. pre-shock; p<0.001) and iNOS (4-fold increase at 12h vs. pre-shock; p<0.001) in AM in vitro. Compared to pre-shock ML exosomes, the in vivo injection of post-T/HS ML exosomes resulted in increased histological lung injury score, a 2-fold increase in Evan’s blue dye lung permeability (0.058 ± 0.003 to 0.117 ± 0.011 mg/g tissue; p<0.05) as well as an increase of wet-to-dry ratio (4.608 ± 0.071 to 5.202 ± 0.149; p<0.05). Exosomes in post-T/HS ML also induced increased recruitment of neutrophils (CD45+Ly-6G+) and macrophages (CD45+CD11c+) in BALF and lung parenchyma determined by flow cytometry (p<0.05 vs. pre-shock).   

Conclusion:  Exosomes released into post-T/HS ML caused an inflammatory response from AM in vitro and ALI in vivo. Our findings define the critical role of ML exosomes as a biologically active mediator of ALI after severe injury.

 

23.03 FER Gene Delivery Improves Survival by Enhancing Lung Immune Response Against Lethal Pneumonia

Posted on January 25, 2017

L. K. McCandless1, Y. S. Yin1, V. Dolgachev1, S. Panicker1, M. Suresh1, M. Hemmila1, K. Raghavendran1, D. Machado-Aranda1  1University Of Michigan,Division Of Acute Care Surgery/Department Of Surgery,Ann Arbor, MI, USA

Introduction: Alternative treatments against antibiotic-resistant bacteria are being strongly investigated. Potentially, transient manipulations of the genome could induce a more efficient immune response. A recent Genome-wide Association Study demonstrated correlation of FER gene expression with survival among pneumonia (PNA) patients. This was confirmed by electroporation-mediated delivery of FER gene in a lethal murine model of combined lung contusion (LC) and PNA. Neither study was conceived to determine the mechanisms for these favorable outcomes. We propose that FER improves survival by recruitment of activated immune cells primed to remove bacteria from the lung.

Methods: C57/BL6 mice were inoculated with 500 CFU of Klebsiella pneumoniae. At 1-hr they received plasmid DNA encoding human (pFER) gene via pharyngeal drop followed by 8 electroporation pulses (EP)  inducing its expression in the lung. We recorded survival data for pFER-EP and control groups (PNA-only; PNA/EP-empty vector and PNA/EP-Na+/K+-ATPase gene). In parallel experiments, animals were sacrificed at specific time points (24, 48, 72 hr), bronchial alveolar lavage (BAL) fluid and lung tissues were harvested; cellular subpopulations counted by flow cytometry; specific genes and signaling pathways were assessed by TaqMan/Western Blot and finally cytokines by ELISA.

Results: After pFER-EP; 5-day survival was markedly improved compared to PNA-only control (80% vs 20%; p < 0.05). Early significant numbers of inflammatory monocytes were only detected in BAL fluid from pFER-EP animals exhibiting known antibacterial markers Toll Receptor- 2 and 4 (TLR2 and TLR4) respectively; being the later more predominant. Both BAL cells and lung tissues had higher protein expression of phosphorylated STAT3 (p-STAT3), a transcription factor critical in bacterial removal. The increased levels of p-STAT3 correlated with the decreased expression of its chaperone Heat-Shock Protein 90 (HSP90). However at 72 hr, HSP90 and other proteosome genes (NRLP2, NRF2) dramatically increased to counter-regulate p-STAT3, inducing monocyte apoptosis and avoiding damage to surrounding tissues.  BAL TLR4+ monocytes were found to highly express Nitric oxide synthase-2 (Nos-2), Resistin-like molecule α1 (Fizz1), Tumor Necrosis Factor-α (TNFα) and Interferon-γ  (IFNγ) all important against bacterial infection. Additionally, pFER-EP was able to rescue neutrophilic response in TLR4-/- mice, via increased production of strong chemoattractant KC and counterbalanced by decoy receptor for advance glycosylation (sRAGE), and independent from TNFα  and IFNγ levels.

Conclusion: Lung gene delivery of the FER improved pneumonia survival by early STAT3 phosphorylation and suppression of HSP90, in turn enhancing antibacterial TLR4+ monocytes. Additionally, FER expression can modulate neutrophilic response via KC and sRAGE cytokines independently from TLR4, TNFα and IFNγ, thus constituting a novel therapeutic strategy.  

23.02 Storage of Thawed Plasma Over 14-Days Does Not Affect Ability to Inhibit tPA-induced Fibrinolysis

Posted on January 25, 2017

B. R. Huebner1, C. C. Silliman1,3, R. Shepherd-Singh2, G. Stettler1, H. B. Moore1, G. Nunns1, A. Banerjee1, E. E. Moore1,2  1University Of Colorado Denver,Trauma Surgery,Aurora, CO, USA 2Denver Health Medical Center,Trauma Surgery,Aurora, CO, USA 3Bonfils Blood Center,Denver, CO, USA

Introduction:
Emerging evidence supports plasma-first resuscitation in the treatment of trauma-induced-coagulopathy (TIC). The logistics of plasma-first resuscitation require thawed plasma (TP) be readily available due to the thawing time of fresh frozen plasma (FFP) (30-minutes at 37°C). The current standard is storage of TP for up to 5 days at 4°C for use in major hemorrhage scenarios, but there remains a 2.2% outdated wastage rate accounting for 129,000 discarded units each year. This 5-day lifespan of TP prevents smaller hospital from having it readily available, and is based on factors V, VII, and VIII levels at outdate. However, the key plasma proteins in attenuating fibrinolysis and correction of TIC remain unknown as does their change in activity level over the thawed storage time. We hypothesize that TP retains the ability to inhibit tPA-induced fibrinolysis at 14 days storage at 1-4°C.
 

Methods:
Pooled FFP was thawed and stored at 1-4°C at the following intervals: 14, 10, 7, 5, 3, and 1-day prior to the experiment. Healthy volunteers participated: 80% men, not pregnant or taking any medications at the time of blood draw (n=5). Blood was drawn into citrated tubes followed by a 50% dilution of whole blood (WB) with the above TP intervals as well as FFP, normal saline (NS), albumin, and WB control. Samples underwent citrated tPA-modified (75ng/ml) thromboelastography (TEG). TEG parameters including R-time, angle, maximum amplitude (MA), and LY30 were examined.
 

Results:

14-day TP retained the ability to inhibit tPA-induced hyperfibrinolysis (median LY30% 8.4) similar to FFP (6.4%), WB (14.6%) and superior to albumin (59.3%, p<0.00001) and NS (58.1%, p<0.00001) (figure 1). All TP intervals demonstrated similar clot characteristics to FFP including rate of clot initiation (R-time), rate of clot formation (angle), maximum clot strength (MA), and resistance to clot breakdown (LY30, figure 1) on tPA-challenge TEG. 14-day TP retained superior clot strength (median MA, 61.5mm) to albumin (21.6, p=0.001) and normal saline (32.2, p=0.0003). 14-day TP also had a faster rate of clot formation (median anlge, 66.2°) compared to albumin (34.8°, p=0.002) and normal saline (41.6°, p=0.006).

 

Conclusion:
TP plasma retained the ability to inhibit tPA-induced fibrinolysis as well as speed of clot formation (angle) and strength of clot formation (MA) over 14-day storage at 1-4°C. A clinical trial is needed to validate these in vitro results but consideration should be made to increasing the storage life of TP.

 

23.01 Microparticles from Aged pRBC Units Stimulate Thrombin Generation and Cause Pulmonary Vascular Congestion

Posted on January 25, 2017

Y. Kim1, B. T. Xia1, A. D. Jung1, A. L. Chang1, R. M. Schuster1, R. A. Veile1, M. D. Goodman1, T. A. Pritts1  1University Of Cincinnati,Surgery,Cincinnati, OH, USA

Introduction:  During storage, packed red blood cells undergo a series of physical, metabolic, and chemical changes collectively known as the red blood cell storage lesion. One key component of the red blood cell storage lesion is the accumulation of microparticles, which are submicron vesicles shed from erythrocytes as part of the aging process. Previous studies from our laboratory indicate that transfusion of these microparticles leads to lung injury, but the mechanism underlying this process is unknown. In the present study, we hypothesized that microparticles from aged packed red blood cell units induce procoagulant activity via accelerated thrombin generation. 

Methods:  Microparticles were isolated from leukoreduced, platelet-depleted, stored murine erythrocytes via ultracentrifugation. These microparticles were added to microplate wells supplemented with Factor Xa-Va mixture, calcium, thrombin inhibitors, and purified prothrombin and the activity of the generated thrombin substrate was measured. We also investigated the effects of microparticle transfusion on the murine pulmonary vasculature.  In separate experiments, mice were transfused with erythrocyte-derived microparticles or an equivalent volume of vehicle. After 1, 4, 8, or 24 hours, lungs were harvested and analyzed under light microscopy or Martius Scarlet Blue (MSB) staining. Blood was harvested from a separate set of mice following injection with either microparticles or saline vehicle and tested for fibrinogen levels.

Results: Erythrocyte-derived microparticles demonstrated a nearly hundredfold greater conversion of prothrombin to thrombin than controls (66.60±0.03 vs 0.70±0.01 peak OD; p<0.0001). Fibrinogen levels were significantly lower in mice injected with microparticles compared with saline vehicle (14.0 vs 16.5 μg/mL, p<0.05), suggesting conversion of fibrinogen to fibrin. Mice injected with erythrocyte-derived microparticles exhibited a significantly greater number of congested pulmonary vessels at 1 hour (10.0 vs 6.5 vessels/HPF) and 4 hours (10.3 vs 6.4 vessels/HPF) as compared to control mice (p<0.05 each) and elevated pulmonary fibrin deposition at 24 hours as determined by MSB staining (3.5±1.5 vs zero deposits, p<0.0001).

Conclusion: Erythrocyte-derived microparticles directly propagate the conversion of prothrombin to thrombin in our murine model. Microparticle treatment in vivo led to decreased fibrinogen and increased pulmonary vessel congestion and fibrin deposition. Microparticles from aged packed red blood cell units may damage injure recipient lungs through accelerated thrombin conversion. 

 

« Previous Page
Next Page »