Previously, we showed that a hydroxyethyl starch colloid in a balanced solution, but not in normal saline, reduced hepatic leukocyte recruitment in a mouse model of early sepsis [29]. Recent clinical trials have raised concerns about the safety of starch products [8], whereas albumin and saline appear equivalent [9]. Accordingly, in this study, our objective was to compare AGP to albumin and normal saline as resuscitation fluids, with respect to the ability of these fluids to dampen the inflammatory response in the liver in

murine models of early endotoxemia and sepsis. All in vivo experiments followed protocols approved by the Animal Research Ethics Board of Health Sciences, McMaster University. Male C57Bl/6 mice (20–25 g) from Taconic

CP-690550 clinical trial (Germantown, NY, USA) were used in all of the BGB324 mw experiments. Human AGP was purified from human plasma either prepared from citrated blood drawn from volunteers by trained phlebotomists under the terms of a protocol approved by the Research Ethics Board, Hamilton Health Sciences Corporation, or from units of transfusable plasma obtained at outdate from Canadian Blood Services. AGP purification from plasma was performed as described [23]; briefly, this entailed sulphosalicylic acid precipitation, neutralization of the supernatant, hydroxyapatite and Cibacron blue chromatography. AGP preparations were tested for endotoxin contamination and depyrogenated, as described, until endotoxin levels fell below 5 endotoxin units/kg body weight for all mice treated with this purified plasma protein. Clinically outdated, apyrogenic HAS (Plasmalbulin 5; Bayer Healthcare, Toronto, ON, USA) was the generous gift of Dr. John Kelton, Department of Medicine, McMaster University. Mice were warmed with an infrared heat lamp for 10 minutes and anesthetized with isofluorane. LPS from Escherichia coli type 0127:B8 (Sigma-Aldrich, St. MYO10 Louis, MO, USA) in normal

saline, or saline alone (for shams), was injected intraperitoneally at 5 or 100 mg/kg body weight. Statistical review of the responses (leukocyte count and recruitment) of both doses was indistinguishable; therefore, data for both doses were combined in the final analysis. One milliliter of normal saline was injected subcutaneously following LPS administration to ensure adequate hydration of the animals. In some experiments, LPS was injected intravenously at a dose of 0.08 mg/kg body weight. The CLP procedure followed the original report by Baker et al. [1], as modified by us [29]. Briefly, mice were anesthetized with isofluorane and the right jugular vein was cannulated to deliver the fluids. The abdomen was opened and the cecum delivered, ligated, and perforated with an 18-gauge needle.