|Home | About | Journals | Submit | Contact Us | Français|
The serious need for expanding the donor population has attracted attention to the use of steatotic donor livers in orthotopic liver transplantation (OLT). However, steatotic livers are highly susceptible to hepatic ischemia-reperfusion injury (IRI). Expression of fibronectin (FN) by endothelial cells is an important feature of hepatic response to injury. We report the effect of a cyclic RGD peptide with high affinity for the α5β1, the FN integrin receptor, in a rat model of steatotic liver cold ischemia, followed by transplantation. RGD peptide therapy ameliorated steatotic IRI and improved recipient survival rate. It significantly inhibited the recruitment of monocyte/macrophages and neutrophils, and depressed the expression of pro-inflammatory mediators, such as inducible nitric oxide synthase (iNOS) and interferon (IFN)-γ. Moreover, it resulted in profound inhibition of metalloproteinase-9 (MMP-9) expression, a gelatinase implied in leukocyte migration in damaged livers. Finally, we show that RGD peptide therapy reduced the expression of the 17-kDa active caspase-3 and the number of apoptotic cells in steatotic OLTs. The observed protection against steatotic liver IRI by the cyclic RGD peptides with high affinity for the α5β1 integrin suggests that this integrin is a potential therapeutic target to allow the successful utilization of marginal steatotic livers in transplantation.
Orthotopic liver transplantation (OLT) is an effective therapeutic modality for end-stage liver disease. However, due to the shortage of organ donors, many patients die every year while on the waiting list (1). The serious need in expanding the donor population has attracted attention to the possible use of steatotic donor livers, which are frequently discarded because of the fear of primary nonfunction, or dysfunction, after transplantation (2). Ischemia/reperfusion injury (IRI) is a multifactorial antigen-independent inflammatory process that can lead to graft loss, particularly with marginal donor organs (2,3). Indeed, a growing body of evidence shows that IRI is poorly tolerated in fatty livers (4–6).
The migration of leukocytes is a key event in acute inflammatory liver injury (7). The transmigration of these cells across endothelial and extracellular matrix (ECM) protein barriers is dependent on a cascade of adhesion and focal matrix degradation events (8). Fibronectin (FN) is a large glycoprotein with a central role in cellular adhesion and migration. The very early expression of the ‘so called’ cellular FN by sinusoidal endothelial cells is a prominent feature of hepatic response to injury (9), including IRI in steatotic livers (10). The role of FN in leukocyte adhesion, migration, and activation has been extensively reported (11). FN has been implicated in multiple pathological conditions, including tumor metastasis (12), rheumatoid arthritis (13), cardiac allograft rejection (14), liver fibrosis (9), and liver IRI (10). The effects of FN are primarily mediated by integrins, a superfamily of cell surface receptors (15). α4β1 (VLA-4) and α5β1 (VLA-5) integrins are the two major FN receptors expressed on leukocytes; of these, α5β1 integrin is highly selective for FN and requires the RGD sequence on the tenth type III repeats of fibronectin for ligand recognition (16). This integrin is expressed on T lymphocytes, monocyte/macrophages, and polymorphonuclear cells (17–19).
In the present study we have examined the effects in steatotic liver IRI of a cyclic RGD peptide, which avidly binds the α5β1 integrin and particularly inhibits cell attachment to fibronectin (20). We show that the cyclic RGD peptide therapy (1) ameliorated hepatocellular injury in steatotic OLTs and prolonged recipient survival, (2) disrupted monocyte/macrophage and neutrophil infiltration, (3) down-regulated metalloproteinase-9 (MMP-9) expression, (4) depressed proinflammatory mediators, and (5) resulted in Akt upregulation and decreased apoptosis in steatotic OLTs. These data support the concept of α5β1 integrin as a potential therapeutic target in steatotic liver I/R injury.
Genetically obese (fa−/fa−) male Zucker (230–275 g), and lean (fa/−) Zucker (260–300 g) rats were obtained from Harlan Sprague Dawley, Inc. (Indianapolis, IN). Syngenic OLTs were performed using fatty livers that were harvested from obese Zucker rats. Steatotic livers were stored at 4°C in University of Wisconsin (UW) solution for 4 hours before being transplanted into lean Zucker recipients. The standard techniques of liver harvesting and orthotopic transplantation without hepatic artery reconstruction were performed according to the previously described Kamada’s and Calne’s cuff technique (21), and an anhepatic phase of 16 to 20 minutes. Cyclic CRGDGWC (RGD) peptides (500 μg/rat) that avidly bind the α5β1 integrin (20) were administered ex vivo via portal vein to steatotic livers before cold storage, and immediately prior to reperfusion. In addition, OLT recipients received a 3-day course of cyclic RGD peptides (1 mg/rat per day, ip) post-transplantation. Cyclization of RGD peptides increases their affinity and inhibitory properties (22), and these cyclic RGD peptides are potent inhibitors of cell attachment to fibronectin (20). Control recipients received vehicle or a scrambled peptide in a similar fashion as in the RGD treated group. Rat recipients of steatotic OLTs were followed for survival. Separate groups of rats were sacrificed at 6h, 24h and day 7 after OLT, and liver samples were collected for further analysis. Animals were fed a standard rodent diet and water ad libitum and cared for according to guidelines approved by the American Association of Laboratory Animal Care. Oil Red-O staining confirmed the high content of fat in the steatotic donor livers. As shown in Figure 1, while naive livers harvested from normal Zucker rats (fa/−) (recipients) showed virtually no steatosis, naive livers harvested from fatty Zucker rat (fa/fa) (donors) showed over 30% steatosis. Indeed, Fatty Zucker rats of 230 to 275g body weight have > 30% liver steatosis, which sets them as marginal donors (10).
Serum glutamic-oxoaloacetic transaminase (sGOT), an indicator of hepatocellular injury, was measured in blood samples obtained at 6 and 24 hours, and day 7 after hepatic reperfusion. Measurements were made with an auto analyzer by ANTECH Diagnostics (Los Angeles, CA).
Liver specimens were fixed in 10% buffered formalin solution and embedded in paraffin. Sections were made at 4μm and stained with H&E. The histological severity of IRI in the liver was graded using modified Suzuki’s criteria (23). In this classification, sinusoidal congestion, hepatocyte necrosis and ballooning degeneration are graded from 0 to 4. The absence of necrosis, congestion or centrilobular ballooning is given a score of 0, while severe congestion and ballooning degeneration, as well as > 60% lobular necrosis is given a value of 4.
Steatotic OLTs were also examined for leukocyte infiltration and fibronectin deposition, as previously described (10). Briefly, cryostat sections were incubated with primary mouse antibody (Ab) against rat T cells (R73), monocytes/macrophages (ED1) (Abd Serotec, Indianapolis, IN), MMP-9 (gelatinase B) (NeoMarkers, Fremont, CA), and cellular FN (IST-9) (Accurate Chemical, Westbury, NY) at optimal dilutions. Bound primary antibody (Ab) was detected using biotinylated anti-mouse IgG and streptavidin peroxidase-conjugated complexes (Dako, Carpinteria, CA). Negative controls included sections in which the primary Ab was replaced with either dilution buffer or normal mouse serum. Control sections from inflammatory tissues known to be positive for each stain were included as positive controls. Sections were evaluated by counting the number of labeled cells within 20 high-power fields (HPF) per section. The relative abundance of some antigens was judged as (−) negative, (+) little, (++) moderately abundant, and (+++, >200 cells/20 HPF) highly abundant.
MPO is a naturally occurring constituent of neutrophils and is frequently used as a marker for neutrophil infiltration in rat livers (24). Frozen tissue was thawed and suspended in iced 0.5% hexadecyltrimethylammonium and 50 mmol potassium phosphate buffer solution (Sigma, St Louis, Mo, USA), of pH 5. After samples were homogenized and centrifuged, 0.1 mL of the supernatant was mixed in solution of hydrogen peroxide–sodium acetate and tetramethyl benzidine (Sigma). One unit of myeloperoxidase activity was defined as the quantity of enzyme that degraded 1 μmol peroxide per minute at 25°C per gram of tissue.
For evaluation of the gene expressions of the pro-inflammatory cytokines, iNOS, and MMP-9, RNA was extracted from livers with Trizol (Life Technologies, Inc., Grand Island, NY) using a Polytron RT-3000 (Kinematica AG, Cincinnati, Ohio), as described (25). Reverse transcription was performed using 4μg of total RNA in a first-strand cDNA synthesis reaction with SuperScript II RNaseH Reverse Transcriptase (Life Technologies, Inc.). One μl of the resulting reverse transcriptase product was used for polymerase chain reaction amplification.
Snap-frozen liver tissue was immediately homogenized as previously described (10). Protein content was determined by a colorimetric assay (Bio-Rad, Hercules, CA). Proteins (40 μg/sample) in sodium dodecyl sulfate (SDS)-loading buffer were electrophoresed through 12% SDS-polyacrylamide gel electrophoresis (PAGE) and transferred to nitrocellulose membranes (Bio-Rad). The gels were then stained with Coomassie blue to document equal protein loading. The membranes were blocked with 5% dry milk and 0.1% Tween 20 (USB, Cleveland, OH) in PBS and incubated with specific primary antibodies against Caspase-3 (Santa Cruz Biotechnology, Santa Cruz, CA), Akt and p-AKT Thr 308 (Cell Signaling Technology, Beverly, MA). The filters were washed and then incubated with horseradish peroxidase-conjugated anti-mouse or anti-rabbit antibodies (Amersham, Arlington Heights, IL). After development, membranes were stripped and reblotted with an antibody against actin (Santa Cruz Biotechnology). Relative quantities of protein were determined using a densitometer (Kodak Digital Science 1D Analysis Software, Rochester, NY).
The TUNEL assay was performed on 5-μm cryostat sections using the In Situ Cell Death detection kit (Roche) according to the manufacturer’s protocol. TUNEL-positive cells were detected under light microscopy. Terminal transferase was omitted as a negative control. Positive controls were generated by treatment with DNase 1 (30 U/ml in 40 mmol/L Tris-Cl (pH 7.6), 6 mmol/L MgCl2, and 2 mmol/L CaCl2 for 30 min).
All values are expressed as the mean ± the standard deviations. Differences between groups were compared using the Mann-Whitney test for continuous variables, and a two-tailed p value <0.05 was considered significant. Animal survival was analyzed according to the method of Kaplan-Meier, and the differences between the two groups were evaluated according to the log-rank test. Calculations were made using SPSS software (SPSS Inc., Chicago, IL, USA).
We examined the effects of a cyclic RGD peptide, which has high affinity for the fibronectin receptor α5β1 integrin (20), on the development of IRI in a well-established model of steatotic OLT. Steatotic OLTs treated with the cyclic RGD peptides had a significantly prolonged survival rate compared with respective controls at 14-days post-OLT (p< 0.02, n=10/gr.) (Fig. 2). The prolonged survival rate observed in the RGD treated steatotic OLTs correlated with improved liver function in these rat recipients, as shown by the decreased sGOT levels (U/L) at 6h (3470±400 vs. 6900±830; n=4/gr p<0.001), day 1 (720±80 vs. 2440±900; p<0.002, n=5/gr p<0.001), and day 7 (170±70 vs. 1330±250; p<0.006, n=5/gr) post-OLT, (Fig. 3A). RGD-treated steatotic OLTs showed mild signs of vascular congestion or necrosis, contrasting with severe hepatocyte necrosis and disruption of lobular architecture observed in respective control livers at day 1 post-OLT, (Fig 2B). Indeed, the modified Suzuki score was significantly decreased in the cyclic RGD-treated OLTs as compared with respective controls (0.5 ± 0.4 versus 2.5 ± 0.5 respectively, p<0.003; day 1 after ischemic insult, n=5/gr). Moreover, RGD therapy showed a long-lasting histological improvement; RGD-treated steatotic livers were characterized by a significant better histological preservation when compared with respective controls at day 7 post-OLT (Fig. 3B).
We evaluated the role of cyclic RGD peptide therapy on leukocyte infiltration in steatotic OLTs. T lymphocytes (17 ± 5) and ED1+ monocytes/macrophages (29 ± 9) were detected in modest numbers in naïve steatotic livers (n=3). Our earlier studies have shown that blockade of FN-α4β1 interactions significantly depressed T cell infiltration, while monocyte/macrophage infiltration was little affected in steatotic livers at day 1 post-OLT (10). Interestingly, in contrast with the blockade of α4β1 integrin in steatotic OLTs, which preferentially affected T-cell recruitment (10), the cyclic RGD peptide therapy was less effective in reducing T lymphocyte infiltration at 6h (66 ± 20 vs. 95 ± 19), and at day 1 post-OLT (70 ± 8 vs. 110 ± 28), (Table I and Fig. 4; n=4–5/gr); it significantly inhibited the intragraft infiltration of monocytes/macrophages at 6h (66 ± 18 vs. 105 ± 16; p<0.02) and at day 1 (74 ± 5 vs. 173 ± 34; p<0.001) post-OLT, (Table 1, and Fig. 4). On the other hand, both T cells (81 ± 12 vs. 192 ± 26; p<0.001), and monocytes/macrophages (47 ± 12 versus 129 ± 42; p<0.001) were detected in the RGD treated steatotic OLTs in significantly reduced numbers as compared to respective controls, at day 7 post-transplantation (Table I and Fig. 4). MPO activity (U), which is an index of neutrophil infiltration, was significantly reduced at day 1 (0.96 ± 0.12 versus 1.89 ± 0.35; p<0.03) and at day 7 (0.27±0.17 vs. 1.88±0.16; p<0.01) post-transplantation as compared with the respective controls, (Fig. 5). Moreover, cellular FN, which is virtually undetectable in steatotic naïve livers (10), was upregulated in both RGD treated and control OLTs at 6h and at day 1 post-transplantation. However, while controls OLTs still showed high levels of FN deposition at day 7 post-transplantation, RGD treated livers at day 7 were characterized by reduced deposition of this adhesion molecule in the vascular endothelium (Fig. 4). Therefore, our data suggest that cyclic RGD peptide therapy preferentially affected the initial monocyte/macrophage and neutrophil infiltration in steatotic OLTs, and protected against the ongoing inflammatory process observed in the control recipients.
Leukocyte migration requires a coordinated series of adhesion and focal matrix degradation events. We have recently shown that MMP-9, a gelatinase implied in FN breakdown, is a critical mediator of leukocyte migration in liver IRI (26,27), and that α4β1-FN interactions regulate MMP-9 expression by leukocytes in damaged livers (25). Others have shown that MMP-9 expression, which is associated with lung cancer invasion, is upregulated in lung cancer cells by the α5β1-FN interactions (28). We evaluated here whether the cyclic RGD peptide therapy affected MMP-9 expression in steatotic OLTs. Indeed, MMP-9 gene expression was profoundly depressed in RGD peptide treated livers at 6h (0.5 ± 0.1 versus 1.4 ± 0.5, p<0.003), day 1 (0.4 ± 0.1 versus 2 ± 0.2, p<0.01) and day 7 (0.35 ± 0.2 versus 2.1 ± 0.1, p<0.0001) after OLT, and contrasting with high MMP-9 expression levels detected in respective controls (Fig. 6A, and B). Moreover, the numbers of MMP-9+ leukocytes were also profoundly depressed in the RGD treated OLTs as compared to controls at 6h (34 ± 21 vs. 97 ± 19, p<0.01); day 1 (46 ± 12 vs. 167 ± 43, p<0.01) and at day 7 (54 ± 19 vs. 152 ± 32, p<0.01) post-OLT (Fig. 6C). Naïve steatotic livers were virtually negative for MMP-9+ leukocytes.
iNOS generates NO in a sustained manner for prolonged periods of time, leading to large amounts of NO (29) which have been associated with liver pathologic conditions (10,30). Therefore, we analyzed the expression of iNOS in steatotic OLTs treated with the cyclic RGD peptides. As shown in Fig. 7, cyclic RGD peptide therapy reduced the intragraft mRNA expression of iNOS at 6h (0.6 ± 0.5 versus 2.2 ± 0.9, p< 0.01), at day 1 (1 ± 0.5 versus 3.2 ± 0.5, p< 0.001) and at day 7 (0.1 ± 0.1 versus 2.7 ± 0.1, p< 0.0001) post-OLT. IFN-γ mRNA expression, which is an initiator of liver reperfusion injury (31), was also significantly depressed in the RGD treated steatotic livers at 6h (0.1 ± 0.1 versus 0.4 ± 0.1, p< 0.002), at day 1 (0.2 ± 0.1 versus 0.5 ± 0.2, p< 0.003) and at day 7 (0.1 ± 0.1 versus 1 ± 0.3, p< 0.0001) post-OLT (Fig. 7). Furthermore, other pro-inflammatory mediators such as IL1β (5.1 ± 0.05 versus 3.2 ± 0.2, p<0.001), IL-2 (3.5 ± 0.4 vs 2.1 ± 0.6, p< 0.05), and TNF-α (3.1 ± 0.2 vs 1.2 ± 0.2, p<0.0001) were markedly depressed in the RGD treated OLTs at day 7, as compared with respective control OLTs (Fig. 7).
Apoptosis is considered an important mechanism in liver IRI (32). To determine whether apoptosis played a role in our settings, we evaluated the expression of Caspase-3, a pro-apoptotic marker associated to liver IRI (27). Caspase 3 is expressed in tissues as an inactive 32-kDa precursor. During apoptosis the 32-kDa caspase-3 is cleaved and generates a 17-kDa mature active form, which is associated with caspase-3 activity (33). While 17-kDa caspase-3 was readily detected in control livers, it was virtually undetectable in RGD treated livers post-OLT, (Fig. 8A). Moreover, TUNEL positive cells were also less detected in the RGD treated livers when compared with controls, in particular at 24h post-OLT (2 ± 0.5 vs. 13 ± 10, p<0.04; n=4/gr.), (Fig. 8B, and C). Interestingly, Akt which is a 57-kD protein-serine/threonine kinase with functions mainly associated to pro-survival (anti-apoptotic) (34), was preferentially phosphorylated in the RGD treated livers (6h: 1.01 ± 0.26 vs. 0.03 ± 0.05, p<0.01; day 1: 0.25 ± 0.02 vs. 0.05 ± 0.02, p<0.004; n=4/gr.) (Fig. 9).
Ischemic damage in the liver associated with leukocyte recruitment, release of cytokines and free radicals, plays a major role in post-IRI, leading to a decline of liver function, and potentially resulting in graft loss (4). In general, leukocyte recruitment to target tissues is dependent on weak rolling adhesions, which are mostly mediated by selectins, and on firm integrin-mediated adhesions (35). However, we should have in consideration that liver is a venous driven vascular bed with slow flow rates and may require a distinct cascade of adhesive events as compared with other organs with higher flow rates. In this regard, it has been demonstrated that selectins are not essential for leukocyte recruitment into inflamed liver microvasculature (36). Lower shear rates may lead to selectin independent leukocyte weak rolling adhesion mechanisms (37). This concept is perhaps even more relevant for fatty livers in view of the observations that steatosis further decreases sinusoidal blood flow by approximately 50% in humans and rats (38,39). Together, these observations highlight the importance that firm adhesion mediated by integrins may have in steatotic OLTs.
We have shown that fibronectin deposition in the vascular endothelium is a very early feature in response to IRI in steatotic liver grafts (10). The α4β1 and α5β1 integrins are the two major FN receptors expressed on leukocytes. We have previously demonstrated that CS1 peptide facilitate blockade of α4β1-FN interactions ameliorated steatotic liver I/R injury (10). We report here the effect of cyclic RGD peptides, which are selective ligands for theα5β1 integrin and inhibit cell attachment to fibronectin (20). We found that administration of cyclic RGD peptides to steatotic OLT recipients significantly reduced liver damage and improved the recipient survival rate. Our observations, are in line with early reports showing that RGD peptides are capable of ameliorating ischemic acute renal failure (40) and that RGD analogues protect against concanavalin A-induced liver damage in mice and against the development of liver cirrhosis in rats (41,42).
One of the most striking effects observed in the cyclic RGD peptide treated steatotic OLT recipients was a marked decrease in monocyte/macrophage infiltration in the livers grafts. Monocyte/macrophages modulate inflammatory processes through the release of cytokines, growth factors, and oxygen radicals (43), and therefore, their migration across ECM proteins during inflammation is an important event. Interestingly, while blockade of α4β1-FN interactions was not very effective in decreasing the initial numbers of monocyte/macrophages infiltrating steatotic liver grafts (10), cyclic RGD peptide treated steatotic OLTs were characterized by a marked reduction of these cells, suggesting an important role for the α5β1 integrin in the recruitment of monocyte/macrophages in steatotic liver IRI. This is in agreement with previous observations indicating that α5β1 is the predominant receptor for fibronectin in monocytes (44), and that α5β1 is implicated in monocyte influx into inflamed sites (18). Interestingly, it has been shown that cyclic RGD peptides were capable of decreasing macrophage infiltration in kidneys and in carotid artery lesions of apo-E-deficient mice (45). Neutrophils, which are considered to be critical mediators in acute inflammatory liver injury (7), were also depressed in the cyclic RGD peptide treated steatotic OLTs. These observations are supported by others, showing that neutrophil adherence to fibronectin is mediated by the α5β1 integrin (46), and that the α5β1 integrin has a role on neutrophil recruitment in lung injury (47). Indeed, it was reported thatα4β1 and α5β1 integrins have a major function in mediating neutrophil recruitment into lung during acute LPS-induced inflammation (47).
Leukocyte transmigration across endothelial and ECM barriers is a complex process, which include cell activating chemokines, and adhesive interactions, as well as focal matrix degradation mechanisms (8). While adhesion molecules are critical to successfully promote leukocyte transmigration by providing leukocyte attachment to the vascular endothelium, matrix proteases are important to facilitate leukocyte movement across vascular barriers. We have recently shown that MMP-9 is a critical mediator of leukocyte migration in liver IRI (26,48), and that α4β1-FN interactions are capable of regulating MMP-9 expression by leukocytes in damaged livers (25). The present study provides evidence that in addition to the α4β1 integrin, the α5β1 integrin is also capable of regulating the expression of MMP-9 in steatotic OLTs. Indeed, FN has been shown to affect MMP-9 expression in several systems (25,49–52). Moreover, it has also been demonstrated that MMP-9 gene expression in human HL-60 myeloid leukemia cells, and in lung cancer cells is up-regulated by α5β1-FN interactions (28,49).
MMP-9, in addition to facilitate leukocytes infiltration in livers after I/R injury, may also cause parenchyma cell detachment from ECM and, consequently, promote adhesion-related apoptosis/anoikis of these cells (48). In these experimental settings, the 17-kDa caspase-3, which is associated with the caspase-3 pro-apoptotic activity (33), was readily detected in control livers and virtually undetectable in RGD treated livers post-OLT. Interestingly, the protein-serine/threonine kinase Akt, which functions have been mainly associated to pro-survival (anti-apoptotic) (34), was preferentially phosphorylated in the cyclic RGD peptide treated steatotic livers after IRI. While hepatocyte necrosis has been considered to be the predominant mode of cell death following IRI in steatotic livers (6), and in our settings control steatotic OLTs were characterized by extensive necrosis, it has been shown that hepatocyte apoptosis also occurs in damaged steatotic livers (53). Indeed, as suggested by Lemasters (54), apoptosis and necrosis are not distinct forms of cell death, and they often coexist in tissue injury due to ischemia-reperfusion. Thus, independently of the form of cell death, cyclic RGD peptide treated livers were characterized by much less injury as compared to respective controls after transplantation.
The expression of pro-inflammatory mediators, such as iNOS and IFN-γ, has been linked to tissue injury, including hepatic injury (10,31,55–57). We show here that both iNOS and IFN-γ were significantly depressed in the cyclic RGD peptide treated steatotic OLTs, which was correlated with the improved liver function observed in these treated animals. Moreover, TNF-α expression, which was initially upregulated in both cyclic RGD peptide treated and control OLTs, it was significantly depressed in well-preserved long-term cyclic RGD peptide treated liver grafts. However, it is important to note that the liver cytokine environment is complex and that cytokines, depending on the context, may be involved in both regenerative and injury processes. In this regard, TNF-α, which inhibition has some detrimental effects in liver regeneration after hepatectomy (58), it has been linked to liver injury in obesity (59).
In summary, this article is the first full report on the function of a cyclic RGD peptide, which avidly binds the α5β1 integrin and particularly inhibits cell attachment to fibronectin (20), in liver IRI. Our observations, which are outlined in Figure 10, show that cyclic RGD peptide therapy down-regulated MMP-9 expression, decreased leukocyte infiltration, and depressed the release of pro-inflammatory mediators, leading to protection against steatotic liver IRI and increased OLT recipient survival. Therefore, this work provides the rational to identify therapeutic approaches based in novel concepts that would allow the successful utilization of marginal steatotic livers in organ transplantation and, consequently, to expand the donor population.
This work was supported in part by the National Institutes of Health RO1 AIO57832 grant to AJC. CF was recipient of the 2008 American College of Surgeons International Scholar Award.