Severe sepsis and septic shock remain a leading cause for mortality in the setting of noncardiac intensive care units. In Germany, the incidence of severe sepsis was estimated to be 76 to 110 cases per year per 100,000 population. Mortality rates have been reported to be between 48% and 55% (1). The economic burden of severe sepsis is immense, with reported daily hospital care costs of 1,100 € (2). In 2010, the latest Sepsis Guidelines were released by the German Sepsis Society (3), but even in the latest guideline, only two causal therapeutic approaches such as antibiotic and surgical treatment of the underlying infection were itemized.
The pathophysiology of sepsis can be described as a complex interaction between the infecting microorganism and the host immune and inflammatory responses (4). Despite intensive research over decades focusing on leukocytes with increasing knowledge about the complex interaction between innate, adaptive immunity, only few new therapies have been developed and used in the clinical setting. Unfortunately, most of the drugs that appeared promising on the basis of animal models were ineffective in clinical trials (5).
Patients with severe sepsis develop progressive subcutaneous and body cavity edema due to a progressive loss of the vascular integrity, leading to an increase in vascular permeability or leakage. The mechanism for the increase in vascular permeability is subject of matter in actual research and seems to represent a potential therapeutic target in the context of severe sepsis. To date, only little is known about the mechanisms leading to the disruption of the interendothelial junctions. London et al. (6) were able to show that VE-cadherin plays an important role in the pathophysiology of the vascular leakage. During sepsis, endothelial cells alter their structure and thus expose the basement membrane. This subcellular structure strongly activates the coagulation cascade (7), which in turn can contribute to the pathogenesis of inflammatory conditions via fibrin deposition and microvascular failure and furthermore enhancing the inflammatory response (8, 9).
The peptide Bβ15-42, a 28-amino-acid peptide corresponding to the N-terminal sequence of the β-chain of fibrin, represents a natural cleavage product of fibrin. It is released from fibrin E1 fragments via plasmin and represents a strong indicator of fibrinolytic activity (8) and targets endothelial adherens junction protein VE-cadherin.
It has been shown that Bβ15-42 prevents myocardial reperfusion injury and reduces infarct size in animal models (10), and the results have been transferred successfully into a phase II clinical trial (11). Gröger et al. (12) were the first to show that, in lipopolysaccharide (LPS)–induced shock, Bβ15-42 was able to preserve the endothelial barrier function by modulating VE-cadherin.
The aim of the present study was to test the hypothesis whether Bβ15-42 is able to attenuate a systemic inflammatory reaction in a murine model of polymicrobial sepsis. In addition, we compare an intermittent application versus a continuous application of Bβ15-42 via osmotic pumps implanted in the peritoneal cavity.
MATERIALS AND METHODS
Bβ15-42 (GHRPLDKKREEAPSLRPAPPPISGGGYR) was kindly provided by Peter Petzelbauer, University Wien, Austria.
Animals and cecal ligation and dissection model
Male 6-week-old C57Bl/6 mice were purchased from Janvier (Le Genest-Saint-Isle, France). All experiments received institutional approval by the Regierungspräsidium Darmstadt. Animals (n = 8 per group) were randomly assigned to four groups: (i) sham, (ii) cecal ligation and dissection (CLD), (iii) CLD + bolus injection Bβ15-42, and (iv) CLD + Bβ15-42 administration via implantable micro-osmotic pumps (see below).
General anesthesia was induced with ketamine (75 mg/kg) and xylazine (10 mg/kg) before a median laparotomy was performed. The distal two thirds of the cecum were marked with a bulldog clamp (Aesculap, Diefenbach, Germany), and two ligations were placed proximal and distal of the clamp. The cecum was then dissected in between the two ligations. The truncated two thirds of the cecum were placed back in the abdominal cavity, and the abdominal cavity was closed. The remaining one third of the cecum was still attached to the colon ascendens, and a normal gut passage was ensured. After 6 h, relaparotomy and lavage were performed, and the truncated two thirds of the cecum was removed. All animals received a single shot of cefuroxime (30 mg/kg) and metronidazole (7.5 mg/kg) intraperitoneally after the second intervention. Animals were killed 6 and 12 h after relaparotomy.
The animals assigned to the Bβ15-42 bolus group received 1 mg/kg Bβ15-42 intravenously into the tail vein immediately after relaparotomy. The group of mice assigned to receive an osmotic pump (Alzet D 1000; Alzet, Cupertino, Calif) for continuous application of Bβ15-42 8 mg/kg per 12 h underwent implantation at the second laparotomy directly before closing of the abdominal cavity. In addition, they received a loading dose of 8 mg/kg Bβ15-42 intraperitoneally in the same procedure. At the end of the observation period, animals were killed by exsanguination under deep pentobarbital/buprenorphine anesthesia. Venous blood samples were collected via the inferior caval vein. Plasma was obtained by centrifugation (13,000g for 7 min at 4°C) and stored at −80°C until assayed. Organs were harvested and snap-frozen in liquid nitrogen or fixed in formalin, respectively.
The human microvascular endothelial cells (HMEC-1) were cultured as described previously (13). Primary human macrophages were isolated from buffy coats (purchased from (DRK-Blutspendedienst Baden-Württemberg-Hessen, Institut für Transfusionsmedizin und Immunhämatologie, Frankfurt am Main, Germany) using Ficoll-Hypaque gradients. Monocytes were differentiated into macrophages with RPMI 1640 containing 10% AB-positive human serum for 7 days as described elsewhere (14). To analyze a potential impact of Bβ15-42 on inflammatory signaling, cells were prestimulated with 3 μM Bβ15-42 for 1 or 16 h followed by exposure to 100 ng/mL LPS or 3 μg/mL LTA (both from Sigma-Aldrich, St Louis, Mo).
Quantitative real-time polymerase chain reaction
RNA was extracted from cells or tissue using Trizol (Sigma-Aldrich), and cDNA was synthesized using iScript (BioRad Laboratories Inc, Hercules, Calif). Gene expression was analyzed by quantitative polymerase chain reaction (PCR) using Power SYBR Green PCR Master Mix, the StepOnePlus (Life Technologies, Heidelberg, Germany) and the following oligonucleotides (Sigma-Aldrich GmbH, Hamburg, Germany): mTNFα sense: 5′-TCT ACTGAACTTCGGGGTGA-3′, antisense: 5′-CACTTGGTGGTTTGCTACGA-3′; mIL-6 sense: 5′-CCGGAGAGGAGACTTCACAG-3′, antisense: 5′- TTCTGCAAG TGCATCATCGT -3′; mIL-10 sense: 5′- TCCTAGAGCTGCGGACTGCC-3′, antisense: 5′-TGGGCCATGCTTCTCTGCCT-3′; hICAM-1 sense: 5′-CCCCCCGG TATGAGATTGT-3′, antisense: 5′-GCCTGCAGTGCCCATTATG-3′; hMCP-1 sense: 5′-GACCATTGTGGCCAAGGAGAT-3′, antisense: 5′-TGTCCAGGTGGTCCATGGA-3′; hIL-6: sense: 5′-ACCCCCAGGAGAAGAT TCCA-3′, antisense: 5′-TCAATTGCTTCTGAAGAGGTGAGT-3′; hTNFα: sense: 5′-TGGGCTACAGGCTTGTCACT-3′, antisense: 5-TGCTTGTTCC TCAGCCTCTT-3′; 18S sense: 5′-GTAACCCGTTGAACCCCATT-3′, antisense: 5′-CCATCCAATCGGTAGTAGCG-3′. Expression was normalized to ribosomal 18S RNA as housekeeping gene. In the analysis, expression of sham-operated animals was set to 1.
Analysis of plasma cytokine levels
Cytokine/chemokine levels in plasma were analyzed using the Luminex technology and the Inflammatory Cytokine Mouse 4-Plex Panel (Life Technologies) or enzyme-linked immunosorbent assay (ELISA) Duo Sets (R&D Systems, Abingdon, UK) according to the manufacturer’s protocols. Plasma was analyzed in duplicate.
Lungs and liver were embedded in paraffin, and 3-μm sections were prepared using a microtome. Neutrophils were stained using the Ly-6B.2 antibody (1:1,000; AbD Serotec, Kidlington, England) and the Vectastain ABC Rat IgG Kit (Vector Labs, Burlingame, Calif). The surrounding tissue was counterstained with hematoxylin. Images were taken with a Leica DM5000B microscope (Leica Microsystems, Wetzlar, Germany) and analyzed using an automatized MatLab algorithm (kindly provided by Bertram Scheller). The software determines the number of pixels in each RGB channel, thus measuring the relative area occupied by neutrophils per image. Quantification was done setting brown pixels proportioned to the total number of pixels.
Alanine aminotransferase/aspartate aminotransferase measurement
Alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were measurements within plasma using ALT and AST Reagent Set Colorimetric method (Teco Diagnostics, Anaheim, Calif) according to the manufacturer’s protocol. Plasma was analyzed in duplicates.
Data are expressed as mean ± SEM and were analyzed using one-way analysis of variance (ANOVA) and Bonferroni post hoc analysis for multiple comparisons. P < 0.05 was considered statistically significant. Alanine aminotransferase and AST data were analyzed using Kruskal-Wallis test modified with Dunn multiple-comparisons test because of nonparametric distribution. Statistical analysis was performed with GraphPad Prism 5.02 (GraphPad Software Inc, La Jolla, Calif).
Bβ15-42 reduces proinflammatory cytokine expression in the lung and liver
The inflammatory response following CLD was analyzed by measuring mRNA expression of interleukin 6 (IL-6), tumor necrosis factor α (TNF-α), IL-10, IL-1β, and monocyte chemoattractant protein 1 (MCP-1) in the lung and liver by quantitative PCR. Cecal ligation and dissection–treated animals showed significantly increased IL-6, TNF-α, IL-10, IL-1β, and MCP-1 mRNA levels after 6 h in the lung (Fig. 1, A–E) and liver (Fig. 2, A–E) when compared with the sham group. At 6 h after treatment with Bβ15-42 (bolus and osmotic pump group), IL-6, TNF-α, IL-10, and MCP-1 mRNA expressions in the lung (Fig. 1, A–C, E) and liver (Fig. 2, A–C, E) were significantly decreased, whereas reduction of IL-1β mRNA was not significant (Fig. 1D and Fig. 2D). After 12 h, a significant induction was observed only for TNF-α in the lung (Fig. 1B) and IL-6 expression in the liver (Fig. 2B) compared with sham-operated mice.
The Bβ15-42–mediated reduction of cytokine expression in the lung was still present but not significant in direct comparison to the 12-h CLD group (Fig. 1). In the liver, we observed a significant decrease in IL-6 mRNA levels at 12 h after bolus or continuous administration of Bβ15-42 (Fig. 2A).
Bβ15-42 reduces plasma proinflammatory cytokine levels
Interleukin 6 plasma levels were decreased about 84% in the Bβ15-42 bolus group and about 72% in the osmotic pump group at 6 h after intervention. After 12 h, only the Bβ15-42 bolus group showed a significant decrease in IL-6 release; however, no significance was observed in the osmotic pump. Similar effects were observed for IL-1β, TNF-α, IL-10, and MCP-1 levels. After 6 h, a significant decrease in plasma cytokine levels was detected in all interventions groups, whereas after 12 h only the bolus group remained significantly reduced. The animals treated with Bβ15-42 via osmotic pumps showed a decrease without significance (Fig. 3).
Bβ15-42 reduces neutrophil invasion in the lungs
To investigate the effects on pulmonary inflammation and neutrophil invasion, immunohistochemical staining of neutrophils with Ly6B.2 antibody was performed after 6 h and compared with the sham group. Lungs from sham-operated mice showed only few resident neutrophils. However, a significant increase was observed following CLD. Treatment with Bβ15-42—bolus as well as continuous administration—significantly reduced the number of infiltrated neutrophils (Fig. 4A). Unstained controls did not show any brown staining (Fig. 4B). The analysis of images clearly reveals an increase in neutrophils in CLD animals compared with shams. Infiltration of neutrophils into the lungs was significantly reduced only in animals treated with a bolus injection of Bβ15-42 (Fig. 4C).
Effects of Bβ15-42 on liver injury following polymicrobial sepsis
To estimate potential effects of Bβ15-42 on organ injury following CLD we analyzed aspartate aminotransferase/alanine aminotransferase (ALT and AST) levels within plasma samples, both markers for liver injury. ALT and AST levels were increased by CLD in comparison to shams. Administration of Bβ15-42 significantly reduced AST levels in comparison to the CLD group, while ALT levels were only significantly decreased in animals treated with a bolus injection of Bβ15-42 (Fig. 5, A and B).
Effects of Bβ15-42 on inflammation in endothelial cells and macrophages
To investigate if Bβ15-42 affects the innate immune system, we analyzed its impact on Toll-like receptor 2 (TLR2) or TLR4 signaling in vitro. Endothelial cells (HMEC-1) and primary human macrophages were prestimulated with 3 μM Bβ15-42 for 1 or 16 h followed by LPS (TLR4 agonist) or LTA (TLR2 agonist) exposure. The inflammatory response was determined by measuring mRNA expression of endothelial MCP-1 and intracellular adhesion molecule 1 or TNF-α and IL-6 in macrophages, respectively. Lipopolysaccharide significantly increased MCP-1 and IL-6 expression in HMEC-1, whereas neither 1 nor 16 h of prestimulation with Bβ15-42 affected LPS-induced mRNA expression. LTA only slightly induced gene expression, which was also not modified by Bβ15-42 (Figs. 6, A and B; http://links.lww.com/SHK/A139). To investigate if Bβ15-42 has any posttranscriptional effects, we analyzed MCP-1 protein level within supernatants of treated HMEC-1. Prestimulation of HMEC-1 with Bβ15-42 for 1 h did neither affect LPS- nor LTA-induced secretion of MCP-1 (Fig. 6C).
Even though the effects of Bβ15-42 were so far mostly referred to endothelial cells, we also treated primary human macrophages with Bβ15-42 followed by LPS and LTA exposure, respectively. Also, in macrophages, Bβ15-42 did not reduce LPS- or LTA-mediated expression of the proinflammatory cytokines IL-6 and TNF-α (http://links.lww.com/SHK/A139).
In the present study, we demonstrated that the fibrin fragment Bβ15-42, administered in a mouse model of polymicrobial sepsis, is able to attenuate systemic inflammation. Either bolus injection or continuous administration of Bβ15-42 via osmotic pumps reduced cytokine levels in the lung, liver, and plasma after 6 h and in the bolus group also after 12 h. Lower efficiency of the pump group after 12 h might be due to underdosing or reduced peritoneal adsorption of Bβ15-42. To our knowledge, continuous application of Bβ15-42 has not been performed before.
However, protective effects are in line with various studies that addressed the protective potential of Bβ15-42 during inflammation-associated disease conditions. Thus, Gröger et al. (12) showed that, during endotoxemia, Bβ15-42 was able to preserve the endothelial barrier function by modulating VE-cadherin via dissociation of the Src kinase Fyn from VE-cadherin.
Roesner et al. (15) showed that pigs that underwent a treatment with Bβ15-42 parallel to fluid resuscitation following hemorrhagic shock have significantly lower levels of proinflammatory plasma cytokines than did the corresponding controls that did not receive Bβ15-42.
To investigate whether Bβ15-42 exhibits its anti-inflammatory properties directly or indirectly by preserving the adherens junctions, we performed in vitro investigation. Prestimulation with Bβ15-42 did not affect the LPS- and LTA-induced inflammatory response in endothelial cells and macrophages. Using both ligands, we could at least cover the two most prominent receptors recognizing components of gram-negative and gram-positive bacteria, namely, TLR2 and TLR4. In a previous study, we could demonstrate that Bβ15-42 reduced the VE-cadherin–dependent migration of monocytes, neutrophils, and lymphocytes across human umbilical vein endothelial cell monolayers (10). In vivo studies further demonstrated that inhibition of VE-cadherin resulted in increased vascular permeability and subsequent neutrophil infiltration in vivo (16). Thus, we assume that attenuated inflammation is rather a result of the reduced capillary leakage. Subsequently, leukocyte infiltration, which is frequently occurring during severe sepsis, is also reduced.
We think that our modified CLD model is closer to the clinical settings and therefore highly valid. In our model, animals underwent a peritoneal lavage and treatment with antibiotics, similar to how patients with peritonitis would be treated. We suggest that this explains why there is nearly no inflammation at 12 h after relaparotomy.
However, in clinical trials, the so-called “immunomodulators” have been tested extensively in the setting of sepsis, but most of the initially promising drugs were ineffective in the clinical setting (17–19). During sepsis, many intracellular pathways are activated (4, 5). This explains why targeting only one pathway in sepsis treatment is an ineffective strategy. It is well known that a certain degree of inflammation is necessary to overcome sepsis. Sepsis shows a dynamic inflammatory process including a hyperinflammatory phase followed by an immune paralysis (20). Hyperinflammation is known to be excessively harmful and should therefore be targeted. Recent research is focusing on the mechanisms responsible for vascular permeability increase that accompanies sepsis. This strategy may have substantial potential for a new therapeutic target in the context of severe sepsis (21). To date, only little is known about the mechanisms leading to the disruption of the endothelial integrity. VE-cadherin has been identified as an important component of endothelial adherens junctions, and recently it has been shown that displacement of VE-cadherin is sufficient to induce gaps between endothelial cells (6), leading to the capillary leakage. Thus, targeting VE-cadherin by Bβ15-42 might represent a novel therapeutic target for the treatment of severe sepsis, by preventing organ edema and consequently organ inflammation and damage.
The authors thank the members of the Central Research Facility of the University Hospital Frankfurt, in particular, Dr. A. Theisen, Dr. C. Tandi, and Dr. M. Wagenblast for outstanding animal care and continuing support. Moreover, the authors are grateful to Madlen Dildey and Christin Reiβig for their technical support.
1. Engel C, Brunkhorst FM, Bone HG, Brunkhorst R, Gerlach H, Grond S, Gruendling M, Huhle G, Jaschinski U, John S, et al.: Epidemiology of sepsis in Germany: results from a national prospective multicenter study. Intensive Care Med 33 (4): 606–618, 2007.
2. Moerer O, Plock E, Mgbor U, Schmid A, Schneider H, Wischnewsky MB, Burchardi H: A German national prevalence study on the cost of intensive care: an evaluation from 51 intensive care units. Crit Care 11 (3): R69, 2007.
3. Reinhart K, Brunkhorst FM, Bone HG, Bardutzky J, Dempfle CE, Forst H, Gastmeier P, Gerlach H, Grundling M, John S, et al.: Prevention, diagnosis, treatment, and follow-up care of sepsis. First revision of the S2k Guidelines of the German Sepsis Society (DSG) and the German Interdisciplinary Association for Intensive and Emergency Care Medicine (DIVI) (German). Anaesthesist 59 (4): 347–370, 2010.
4. Hotchkiss RS, Karl IE: The pathophysiology and treatment of sepsis. N Engl J Med 348 (2): 138–150, 2003.
5. Russell JA: Management of sepsis. N Engl J Med 355 (16): 1699–1713, 2006.
6. London NR, Zhu W, Bozza FA, Smith MC, Greif DM, Sorensen LK, Chen L, Kaminoh Y, Chan AC, Passi SF, et al.: Targeting Robo4-dependent Slit signaling to survive the cytokine storm in sepsis and influenza. Sci Transl Med 2 (23): 23ra19, 2010.
7. Levi M: The coagulant response in sepsis and inflammation
. Hamostaseologie 30 (1): 10–12, 14–16, 2010.
8. Jennewein C, Tran N, Paulus P, Ellinghaus P, Eble JA, Zacharowski K: Novel aspects of fibrin(ogen) fragments during inflammation
. Mol Med 17 (5–6): 568–573, 2011.
9. Levi M, van der Poll T, Buller HR: Bidirectional relation between inflammation
and coagulation. Circulation 109 (22): 2698–2704, 2004.
10. Petzelbauer P, Zacharowski PA, Miyazaki Y, Friedl P, Wickenhauser G, Castellino FJ, Groger M, Wolff K, Zacharowski K: The fibrin-derived peptide Bbeta15-42 protects the myocardium against ischemia-reperfusion injury. Nat Med 11 (3): 298–304, 2005.
11. Atar D, Petzelbauer P, Schwitter J, Huber K, Rensing B, Kasprzak JD, Butter C, Grip L, Hansen PR, Suselbeck T, et al.: Effect of intravenous FX06 as an adjunct to primary percutaneous coronary intervention for acute ST-segment elevation myocardial infarction results of the F.I.R.E. (Efficacy of FX06 in the Prevention of Myocardial Reperfusion Injury) trial. J Am Coll Cardiol 53 (8): 720–729, 2009.
12. Gröger M, Pasteiner W, Ignatyev G, Matt U, Knapp S, Atrasheuskaya A, Bukin E, Friedl P, Zinkl D, Hofer-Warbinek R, et al.: Peptide Bbeta(15-42) preserves endothelial barrier function in shock. PLoS One 4 (4): e5391, 2009.
13. Ades EW, Candal FJ, Swerlick RA, George VG, Summers S, Bosse DC, Lawley TJ: HMEC-1: establishment of an immortalized human microvascular endothelial cell line. J Invest Dermatol 99 (6): 683–690, 1992.
14. Weigert A, Johann AM, von Knethen A, Schmidt H, Geisslinger G, Brune B: Apoptotic cells promote macrophage survival by releasing the antiapoptotic mediator sphingosine-1-phosphate. Blood 108 (5): 1635–1642, 2006.
15. Roesner JP, Petzelbauer P, Koch A, Tran N, Iber T, Vagts DA, Scheeren TW, Vollmar B, Noldge-Schomburg GE, Zacharowski K: Bbeta15-42 (FX06) reduces pulmonary, myocardial, liver, and small intestine damage in a pig model of hemorrhagic shock and reperfusion. Crit Care Med 37 (2): 598–605, 2009.
16. Gotsch U, Borges E, Bosse R, Boggemeyer E, Simon M, Mossmann H, Vestweber D: VE-cadherin antibody accelerates neutrophil recruitment in vivo
. J Cell Sci 110 (Pt 5): 583–588, 1997.
17. Sharma VK, Dellinger RP. Treatment options for severe sepsis and septic shock. Expert Rev Anti Infect Ther 4 (3): 395–403, 2006.
18. Skrupky LP, Kerby PW, Hotchkiss RS. Advances in the management of sepsis and the understanding of key immunologic defects. Anesthesiology 115 (6): 1349–1362, 2011.
19. Webster NR, Galley HF: Immunomodulation in the critically ill. Br J Anaesth 103 (1): 70–81, 2009.
20. Schmidt MV, Paulus P, Kuhn AM, Weigert A, Morbitzer V, Zacharowski K, Kempf VA, Brune B, von Knethen A: Peroxisome proliferator-activated receptor gamma-induced T cell apoptosis reduces survival during polymicrobial sepsis. Am J Respir Crit Care Med 184 (1): 64–74, 2011.
21. Lee WL, Slutsky AS: Sepsis and endothelial permeability. N Engl J Med 363 (7): 689–691, 2010.