Portal hypertension (PHT) is a common clinical syndrome

Transcription

1 GASTROENTEROLOGY 2003;124: The Role of Nitric Oxide Synthase Isoforms in Extrahepatic Portal Hypertension: Studies in Gene-Knockout Mice NICHOLAS G. THEODORAKIS,*, YI NING WANG,* NICHOLAS J. SKILL,* MATTHEW A. METZ,* PAUL A. CAHILL, EILEEN M. REDMOND,* and JAMES V. SITZMANN* Departments of *Surgery and Biochemistry and Biophysics, University of Rochester Medical Center, Rochester, New York; and Vascular Health Research Center, School of Biotechnology, Dublin City University, Dublin, Ireland Background & Aims: Considerable debate exists concerning which isoform of nitric oxide synthase (NOS) is responsible for the increased production of NO in PHT. We used the portal vein ligation model of PHT in wildtype and enos- or inos-knockout mice to definitively determine the contribution of these isoforms in the development of PHT. Methods: The portal vein of wildtype mice, or those with targeted mutations in the nos2 gene (inos) or the nos3 gene (enos), was ligated and portal venous pressure (Ppv), abdominal aortic blood flow (Qao), and portosystemic shunt determined 2 weeks later. Results: In wild-type mice, as compared with sham-operated controls, portal vein ligation (PVL) resulted in a time-dependent increase in Ppv ( vs cmh 2 O, at 14 days) concomitant with a significant increase in Qao ( vs ml/min/g) and portosystemic shunt (0.47% 0.01% vs 84.13% 0.09% shunt). Likewise, PVL in inos-deficient mice resulted in similar increases in Ppv, Qao, and shunt development. In contrast, after PVL in enos-deficient animals, there was no significant change in Ppv ( vs cmh 2 0) or Qao ( vs ml/min/g). However, enos ( / ) mice did develop a substantial portosystemic shunt (0.33% 0.005% vs 84.53% 0.19% shunt), comparable to that seen in wild-type animals after PVL. Conclusions: These data support a key role for enos, rather than inos, in the pathogenesis of PHT. Portal hypertension (PHT) is a common clinical syndrome associated with obstruction of portal blood flow to the liver, typically after chronic liver disease such as cirrhosis. The pathophysiological changes of PHT include arterial hypotension, a generalized reduced vascular resistance, hypovolemia, and high cardiac output. 1 4 Suggested etiology for the hyperdynamic circulatory state characteristic of PHT includes a decreased responsiveness to and/or production of endogenous vasoconstrictors 5,6 and exaggerated endothelial function with enhanced synthesis of vasoactive substances, especially nitric oxide (NO) 7 9 and prostacyclin (PGI 2 ). 10,11 Indeed, the role of NO in the pathogenesis of the reduced vascular resistance in PHT has attracted much interest since the initial studies of Vallance and Moncada. 12 Several reports, using NO synthesis inhibitors, have shown attenuation of the marked hyporeactivity of PHT vessels to vasoconstrictors in vitro and reversal of the arterial hypotension in vivo. 13,14 The whole body production of NO has been found to increase in PHT, and enhanced NO synthase (NOS) activity has been shown in PHT hyperemic vessels. 7,9,15 However, the source and etiology of the enhanced NO production in PHT remains controversial. Furthermore, there has been considerable debate concerning which isoform of NOS is responsible for the increased production of NO in PHT. Three NOS isoforms have been identified. The constitutive type I NOS isolated from the brain and type III NOS (enos) isolated from endothelial cells are both regulated by Ca 2 and calmodulin and require reduced NADPH and 5,6,7,8-tetrahydrobiopterin for optimal activity. In contrast, the inducible type II NOS (inos) found in macrophages and vascular smooth muscle cells after activation by various cytokines and endotoxin is Ca 2 independent. 16 Although some evidence from the portal vein ligated model of prehepatic PHT would support an exclusive increase in enos activity, 14,16 several studies have shown contradictory data concerning the role for induction of inos in vascular cells from PHT animals. 17,18 Using the portal vein ligation model of PHT in wildtype and enos- or inos-knockout mice, the aim of our study was to definitively determine the contribution of these NOS isoforms in the development of the hemodynamic derangements typical of PHT. Abbreviations used in this paper: NADPH, -nicotinamide adenine dinucleotide phosphate; NO, nitric oxide; NOS, nitric oxide synthase; Ppv, portal venous pressure; PHT, portal hypertension; PVL, portal vein ligation; Qao, abdominal aortic blood flow by the American Gastroenterological Association /03/$30.00 doi: /s (03)

2 May 2003 NOS ISOFORMS AND PORTAL HYPERTENSION 1501 Materials and Methods Prehepatic PHT Model: Partial Portal Vein Ligation All studies were approved by the University of Rochester committee for animal research and adhered to Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) and federal guidelines for the humane care and treatment of animals. Mice were maintained in sterilized isolette cages on a 12-hour light/dark cycle and were allowed access to food and water ad libitum. Mice were anesthetized using halothane inhalation. A midline laparotomy was performed, and the portal vein was exposed. A blunt-ended 27 gauge needle was placed alongside the portal vein, and a 4-0 silk suture was tied around the vein and needle, after which the needle was withdrawn, producing a standardized stenosis. In sham-operated animals, (used as controls), the procedure consisted of dissection and visual inspection of the portal vein without ligature. The abdomen was closed, and the animals were allowed to recover under a heat lamp. Gene-Deficient Mice Mice containing targeted mutations in the nos2 gene (inos; strain B6,129P-Nos2 tm1lau, and the nos3 gene (enos, strain C57BL/6J-Nos3 tm1unc, were purchased from The Jackson Laboratory (Bar Harbor, ME). Age-matched mice from congeneic strains (B6, 129P, or C57BL/6J) were used as wild-type controls. Mice genotypes were confirmed by using polymerase chain reaction (PCR) on DNA isolated from tail samples using Qiagen DNeasy kit (Qiagen Inc, Stamford, CA) as per manufacturer s instructions. Gene-specific primers were used to detect the targeted mutations in the nos3 gene 19 (gtgtgaaggcaaccattctg and actcatccatgcacaggacc) and the nos2 gene 20 (ggcttcacgggtcagagcca and tgcccattgctgggacagtc). PCR conditions were 25 cycles of 1 minute each of 94 C, 60 C, and 74 C. Physiologic Measurements At the indicated times after sham operation or portal vein ligation (PVL), animals were anesthetized and subjected to laparotomy to allow physiologic measurements to be taken. Splenic pulp pressure was measured as an index of portal venous pressure (Ppv). To measure splenic pulp pressure, a cannula made from a 25-gauge needle connected to a salinefilled manometer was inserted into the spleen pulp, as previously described. 21 To confirm the legitimacy of using the splenic pulp pressure as an indicator of portal vein pressure, the pressures in the portal vein (measured via a PE50 catheter) and the splenic pulp were simultaneously analyzed in the same animal (Table 1). We observed that there was a good agreement between the 2 measurements, with the portal venous pressure slightly lower than the pressure within the splenic pulp (Figure 1).This result is in good agreement with clinical observations: in humans, the portal venous pressure is 2 to 3 mm Hg lower than the splenic pressure. 22 Table 1. Comparison of Splenic Pulp Pressure and Portal Venous Pressure Psp Ppv nd NOTE. Pressures are given in cm H 2 O. Psp, splenic pulp pressure; Ppv, portal venous pressure; nd, not determined; this animal died before the pressure could be recorded. Abdominal aortic flow (Qao) was measured by placing an ultrasonic Doppler flow probe (Transonic #11RB) around the abdominal aorta between the diaphragm and celiac artery. Flow rates were obtained with a Transonic T206 Blood Flow Meter (Transonic Instruments, Ithaca, NY). Aortic blood flows were standardized per gram of body weight. Portosystemic Shunt Fluorescent microspheres were used to assess the degree of portosystemic shunting. The feasibility, accuracy, and reliability of the fluorescent microsphere technique in analyzing blood flow have been shown in other studies Two weeks after sham operation or PVL, mice were anesthetized and a laparotomy was performed as described earlier. Approximately m red or blue polystyrene fluorescent microspheres (Molecular Probes, Eugene, OR) were injected into the spleen (red spheres) or femoral vein (blue spheres). The liver and lungs were collected and placed in 20 ml of 2% sodium dodecyl sulfate, 0.1 mol/l EDTA, 10 mmol/l Tris, ph 8.0, and the tissue was homogenized. Proteinase K was added to 0.1 mg/ml, and the proteins were digested overnight at 45 C. Ten milliliters of ethanol was added to reduce the density of the fluid, and the homogenate was briefly sonicated. Microspheres were collected by centrifugation at 1000 g, washed in 0.2% Tween-80, centrifuged again, and resuspended in 0.1 ml of 0.2% Tween-80. Microspheres were counted using a hemacytometer and a Nikon TE300 inverted microscope equipped for epifluorescence. The degree of shunting was calculated as the number of microspheres in the lungs divided by the total number of microspheres in the lungs and liver times 100. Plasma NO Levels Products of NO breakdown (nitrate plus nitrite) were measured using a nonenzymatic colorimetric nitrite assay (Oxford Biomedical, Oxford, MI). Blood was collected by cardiac puncture and was injected into heparinized tubes, centrifuged, and deproteinized using zinc sulfate. Nitrates were reduced to nitrite using cadmium, and total nitrites were determined using the Griess reagent and measurement of absorbance at 545 nm.

3 1502 THEODORAKIS ET AL. GASTROENTEROLOGY Vol. 124, No. 5 Figure 1. Comparison of splenic pulp pressure and portal venous pressure in mice. Wild-type mice (n 6) were anesthetized and laparotomized. The splenic pulp pressure and the Ppv were determined using a saline-filled manometer. *P Statistics The data shown are means standard error of the mean, with 3 to 11 animals per experimental group. Statistical significance was estimated using unpaired Student t and Mann Whitney tests. A value of P 0.05 was considered significant. Results Murine PVL Model of PHT The salient feature of portal hypertension is an increase in the portal venous pressure initiated by an obstruction in the portal venous flow. In other animal models such as rabbits and rats, because there is a direct connection between the spleen and portal venous system, the splenic pulp pressure is used as an indicator of portal pressure. To confirm that this was also the case in mice, we examined the route of blood flow from the spleen by injection of 15- m red fluorescent microspheres. Microspheres that were injected into the spleen were observed to accumulate almost exclusively in the liver (Table 2, Figure 2), confirming that in mice the spleen is directly connected to the portal system. We also compared the pressure in the splenic pulp to that of the portal vein in the same animal, and observed that they were similar (Figure 1), with the splenic pulp pressure being slightly higher than the portal venous pressure, as described in humans. 22 Accordingly, we used splenic pulp pressure as an index of portal hypertension in this study. After PVL in mice, splenic pulp pressure was elevated as compared with sham-operated controls by day 2 and remained significantly elevated up to day 14 ( vs cmh 2 O[n 7]; P 0.001, for PVL vs sham, respectively) (Figure 3). There was no significant difference in body weights between PVL and sham mice after 14 days (data not shown). Consistent with the splanchnic hyperemia characteristic of PHT, abdominal aortic flow (Qao) was significantly increased in 14 day post-pvl animals compared with sham-operated controls ( [n 7] vs ml/min [n 3], respectively; P 0.05). To determine the degree of portosystemic shunt in PVL animals, 15- m red fluorescent microspheres were injected into the spleens of both sham-operated or PVL mice at 14 days after the procedure. In sham-operated mice, microspheres that were introduced into the portal circulation by splenic injection were found exclusively in the liver, as expected (Figure 2A). However, in PVL animals, a significant number of microspheres that were injected into the spleen were found in the lungs, indicating that they were shunted into the systemic venous circulation (84.13% 0.09% [n 7] vs 0.47% 0.01% shunt [n 3] for PVL vs sham, respectively, P ) (Figure 2B). In some experiments, 15- m blue fluorescent microspheres were also introduced into the systemic venous circulation by injection into the femoral vein. These spheres were subsequently found in the lungs (Figure 2B and D). To further examine the ability of 15- m fluorescent microspheres to circulate and accumulate in the capillary beds of various target organs, we examined the distribution of microspheres in various organs after injection into the systemic arterial circulation as described. 25 Mice were sham operated or portal vein ligated; 14 days later, 15- m blue fluorescent microspheres were injected into the splenic pulp, and 15- m red fluorescent microspheres were injected into the left ventricle of the same animal. The kidneys, brains, gut (stomach, small and Table 2. Organ Distribution of Fluorescent Microspheres After Injection in Sham-Operated or Portal-Vein Ligated Mice Sham PVL K B G Li Lu K B G Li LU LV SP NOTE. Mice were sham-operated or portal-vein ligated; 14 days later 15- m blue fluorescent microspheres were injected into the splenic pulp (SP) and 15- m red fluorescent microspheres were injected into the left ventricle (LV) of the same animal. Organs were harvested and the microspheres were purified and examined by fluorescence microscopy. The numbers are the sum of the spheres seen in 6 randomly chosen microscopic fields. K, kidney; B, brain; G, gut; Li, liver; Lu, lung.

4 May 2003 NOS ISOFORMS AND PORTAL HYPERTENSION 1503 Figure 2. Portosystemic shunt in portal vein-ligated mice. Wild-type mice were subjected to partial PVL or sham-operated as described in Materials and Methods. Fourteen days later, the mice were anesthetized and laparotomized. Fifteen micron red or blue fluorescent microspheres (approximately 10 5 each) were injected in the spleen (red spheres) or the femoral vein (blue spheres). The livers and lungs were harvested and the spheres extracted and visualized by fluorescence microscopy. Representative microscopic fields of microspheres isolated from the livers (A and C) or the lungs (B and D) of sham (A and B) or portal vein-ligated (C and D) mice are shown. large bowel), liver, and lungs were harvested, and the microspheres were purified and examined by fluorescence microscopy (Table 2). Red microspheres that were injected into the left ventricle were able to circulate to various target organs, where they were trapped, as previously shown. 25 However, blue microspheres that were Figure 3. Elevation of splenic pulp pressure in portal vein-ligated mice. Wild-type mice were subjected to partial portal vein ligation (7 11 animals per time point) or sham operated (3 5 animals per time point) as described in Materials and Methods. At the indicated times, the mice were anesthetized, laparotomized, and the splenic pulp pressure (an index of Ppv) determined using a saline-filled manometer. *P vs. unligated (sham) mice. introduced into the portal circulation by injection into the splenic pulp were found in no organ other than the liver or lungs. Thus, as in previously characterized rat and rabbit models of PHT, the mouse (after PVL) developed the characteristic hemodynamic derangements of PHT namely, elevated portal pressure, splanchnic hyperemia, and portosystemic shunt. Portal Venous Pressure and Serum Nitrite Levels After PVL in inos ( / ) and enos ( / ) Mice To investigate the role of NOS isoforms in the development of PHT after PVL, we examined the temporal effects of the ligation on hemodynamics in both inos and enos gene deficient mice and their wildtype congeneic strain counterparts as controls. We used PCR to confirm that the inos- and enos-knockout mice lacked the inos and enos genes, respectively, whereas in wild-type animals, both inos and enos genes were present (Figure 4C). After PVL, inos ( / ) animals had significantly elevated splenic pulp pressures ( cmh 2 O, n 5) similar to their PVL wild-type controls ( cmh 2 O, n 3), when compared with sham animals ( cm H 2 O, n 5 for wild type; 8.78

5 1504 THEODORAKIS ET AL. GASTROENTEROLOGY Vol. 124, No. 5 Figure 4. Splenic pulp pressure in wild-type and gene-deficient mice after PVL. Wild-type or inos- or enos-deficient mice were sham operated or portal vein-ligated as described. Splenic pulp pressures (an index of Ppv) were measured 14 days later. (A) Splenic pulp pressure in wild-type (B6, 129P) or inos ( / ) knockout mice after sham operation or after partial PVL. (B) Splenic pulp pressure in wild-type (C57BL.6J) or enos ( / ) knockout mice after sham operation or partial PVL. *P 0.05 versus unligated mice. (C) PCR products from wild-type (lanes 1, 3, 5, 7), inos knockout (lanes 2, 6), enos knockout (lanes 4, 8) mouse tail DNA after amplification using primers specific for the inos (lanes 1 4) or enos (lanes 5 8) gene. Lane 9 is a 100 base pair DNA ladder. The positions of the specific amplification products are indicated on the left with arrows cmh 2 O, n 5 for inos [ / ]; Figure 4A). In contrast, enos ( / ) mice failed to develop elevated splenic pulp pressure after PVL (Figure 4B). Whereas wild-type mice showed an elevation of splenic pulp pressure after PVL ( cmh 2 O, n 4) as compared with sham animals ( cmh 2 O, n 5) in enos ( / ) mice, splenic pulp pressure after PVL ( cmh 2 O, n 6) was not significantly different from sham animals ( cmh 2 O, n 5). In wild-type and inos ( / ) animals serum nitrite levels were elevated at 2 days after ligation (Figure 4) (an increase from mol to mol for wild type and mol to mol for inos knockouts, n 7) before returning to baseline levels at 14 days. In contrast, there was no

6 May 2003 NOS ISOFORMS AND PORTAL HYPERTENSION 1505 Figure 5. Serum NO x levels in wild-type and gene-deficient mice after PVL. Blood was collected by cardiac puncture from unligated (day 0) wild-type, inos ( / ), and enos ( / ) mice and 2, 4, and 14 days after PVL. Each group had 6 to 8 animals per time point. Total serum nitrate/nitrite was measured by the Greiss reaction after reduction of nitrate to nitrite. The results are expressed as the percent of the serum nitrate/nitrite levels in untreated mice. *P 0.05 versus unligated (sham) mice. significant change in serum nitrite levels in enos ( / ) mice at 2, 4, or 14 days after ligation (Figure 5). Qao and Portosystemic Shunt in Sham and PVL inos ( / ) and enos ( / ) Mice After PVL, there was a significant increase in Qao in both wild-type (1.84-fold increase) and inos ( / ) (2.2-fold increase) animals (Figure 6), as compared with sham-operated controls. However, in the enos ( / ) Figure 6. Abdominal aortic flow (Qao) in wild-type (WT) and genedeficient mice after PVL. Wild-type or inos- or enos-deficient mice were sham operated (3 animals per time point) or portal vein ligated (6 animals per time point) as described. Fourteen days later the mice were anesthetized, laparotomized, and the abdominal aortic flow was measured using an ultrasonic flow probe. *P 0.05 versus unligated (sham) mice. Figure 7. Portosystemic shunt in wild-type and gene-deficient mice after PVL. Wild-type or inos- or enos-deficient mice were sham operated or portal vein ligated as described (3 to 6 animals per time point). Fourteen days later the fluorescent microspheres were injected in the spleen as described. The number of spheres present in the lungs and livers was determined by fluorescence microscopy. The level of portosystemic shunt is calculated by the number of spheres present in the lungs divided by the number in the liver plus lungs, 100. *P versus unligated (sham) mice. animals, there was no significant increase in Qao after PVL, as compared with sham animals (Figure 6). After PVL, all groups of animals (ie, wild-type, inos [ / ], and enos [ / ] animals) developed portosystemic shunt to a similar degree ( 80%) (Figure 7). Discussion PHT is a common clinical entity associated with chronic liver disease and is characterized by both a pathological increase in portal pressure and a hyperdynamic circulatory state. 1,2,26 In the current study, we report the development and utilization of a murine model of PVL that exhibits all the major clinical hallmarks of PHT. The altered vascular response of wild-type mice after partial PVL closely reflects previous studies performed in rats and rabbits. 4,7,15,27 29 In particular, after PVL, the mice developed higher Ppv, elevated abdominal aortic blood flow (representative of splanchnic hyperemia), and a substantial portosystemic shunt, confirming that these animals were indeed portal hypertensive and therefore suitable for studies using gene-knockout animals. In addition, we show by using gene-knockout animals that enos (but not inos) is a key mediator of the altered vascular response in this model of PHT. Nevertheless, shunting occurs after PVL regardless of the presence of functional inos or enos genes. These data suggest that the maintenance of portosystemic shunts that develop after PVL allows for the normalization of portal pressure,

7 1506 THEODORAKIS ET AL. GASTROENTEROLOGY Vol. 124, No. 5 as long as enos-mediated splanchnic vasodilation does not occur. An increased basal release of NO is thought to play a major role in the pathogenesis of the vasodilation and vascular hypocontractility associated with PHT. 7,15,17,30 In agreement with this hypothesis, in portal vein ligated models of prehepatic PHT, the whole-body production of NO was increased and enos activity was significantly enhanced within the hyperemic vasculature, suggesting that the splanchnic vascular tree is a major source of NO release in PHT. 7,8,15,30 35 More recent studies suggest that upregulation of enos activity occurs within the splanchnic vascular tree. This activity precedes any increase in splanchnic blood flow leading to splanchnic hyperemia, 9,11,35,36 and thereby suggests that shear stress (because of the increase in splanchnic blood flow) is not solely responsible for the enhanced NO production. Similarly, in our murine PHT model, serum nitrite levels were significantly increased in wild-type animals 2 days after ligation, before an increase in abdominal aortic blood flow might be expected in these animals, and then returned to baseline levels by 14 days. The early increase in serum nitrite levels was associated with a significant increase in abdominal aortic blood flow that persisted for 14 days after ligation, concomitant with an increase in portal pressure (splenic pulp pressure) and the development of a substantial portosystemic shunt, which persisted despite the return of nitrate levels to normal. Similar increases in serum nitrite levels were seen in previous PVL studies using rats. 37 However, in enosdeficient animals after PVL, there was no increase in serum nitrite levels at 2, 4, or 14 days, concomitant with a lack of an increase in portal pressure and abdominal aortic blood flow at 2 weeks. These data suggest that an increase in enos activity early after PVL (within 2 days) can cause changes in systemic hemodynamics leading to portal hypertension that persist even after NO levels have returned to normal. The data presented strongly suggest that enos was responsible for the enhanced portal pressure and splanchnic hyperemia observed in these portal hypertensive animals. However, despite no elevation in portal venous pressure or abdominal aortic blood flow in enos deficient animals following PVL, the increased vascular resistance provided by portal vein stenosis caused the portosystemic shunt to fully develop. These data suggest that the mechanisms for the development of shunting and PHT are independent, and that enhanced NO production may not be solely responsible for the development or maintenance of the collateral circulation in this model of PHT. However, alternative models of shunting have shown that treatment with enos antagonists reduced hyporesponsiveness to vasoconstrictors, 38,39 suggesting a contributory role for NO in the maintenance of collateral tone in PHT rat models. Moreover, in shortterm treatment studies, NO blockade failed to significantly reduce portal pressure despite reducing splanchnic blood flow and collateral resistance. In addition, there is evidence that NO may act as an angiogenic factor in promoting collateral vessel formation in PHT. Inhibition of NO formation by N-nitro-L-arginine diminished basal and basic fibroblast growth factor (bfgf)-stimulated mesenteric angiogenesis in PVL rats. 40 The reasons for these apparent differences are unclear but may include different models and species of animals used, and, that the method of shunting used in this study does not address the rate at which shunting develops. Despite these discrepancies, we conclude from the current study that enos plays a critical role in the development and maintenance of the elevated portal pressure and splanchnic hyperemia independent of any significant effect that NO may have on the collateral circulation. Although some evidence from PVL models of prehepatic PHT would support an exclusive increase in enos, 14,15 several studies have shown contradictory data concerning the role for induction of inos in vascular cells from PHT animals. 17,18,29,41,42 Endotoxin, released from the intestinal flora and detoxified by the liver, could stimulate inos directly or indirectly via the cytokine cascade. 43,44 However, failure to detect inos messenger RNA or protein expression within the hyperemic vasculature of PHT and cirrhotic animals, coupled with the failure of polymyxin B (an endotoxin antagonist) to attenuate the hemodynamic abnormalities of PHT rats 43,44 despite reversing endotoxin-induced changes, suggests that inos may not play a major role in PHT hemodynamic abnormalities. In agreement with these reports, our study shows that there was no significant difference in the hemodynamic characteristics of inos gene-deficient mice after portal vein ligation, as compared with their wild-type controls. Despite the observations in enos ( / ) mice the hemodynamic alterations of PHT are not exclusively enos dependent, and other mediators have been implicated. Inhibition of cyclooxygenase has been shown previously to both decrease hyperemia within the splanchnic vasculature and reduce portosystemic shunt in prehepatic animal models of PHT. 8 11,28,45 47 However, although we and others have shown previously that Cox products may play a contributory role in the pathology of PHT, enos-deficient animals failed to develop any significant elevation in portal pressure or

8 May 2003 NOS ISOFORMS AND PORTAL HYPERTENSION 1507 abdominal aortic blood flow as compared with their wild-type counterparts. This does not exclude a role for Cox products, in particular prostacyclin, in controlling collateral blood flow and hence portal pressure in these animals. Given the fact that the NO and Cox pathways interact, 48 the precise role of prostacyclin or other vasoactive mediators in this animal model of PHT should be investigated. Although our results are in agreement with our previous studies using pharmacological inhibition of NOS to ameliorate an increase in mesenteric blood flow after PVL of rats, 11 it has recently been shown by other investigators that alterations in hemodynamics associated with PVL of mice did not depend on enos or inos activity. 49 It is difficult to determine why enos is required in one model but not in another, especially because both models use a similar PVL to induce PHT. Nevertheless, this indicates that at least under some conditions, molecules other than NO may be able to mediate the hemodynamic derangements associated with PHT. In conclusion, we report a murine portal vein ligation model that exhibits the hallmarks of chronic PHT. Moreover, deletion of the enos gene preferentially protects these animals from the hemodynamic abnormalities typical of PHT. Furthermore, inhibition of the increase in portal pressure and abdominal aortic blood flow in these animals is independent of any changes in portosystemic shunting. In contrast, deletion of the inos gene failed to reverse the hemodynamic profile of these animals as compared with their wild-type counterparts. These studies suggest, at least in a prehepatic animal model of PHT, that enos rather than inos is the major contributor to the increased portal pressure and splanchnic hyperemia of PHT. References 1. Lebrec D, Bataille C, Bercoff E, Valla D. Hemodynamic changes in patients with venous obstruction. Hepatology 1983;3: Reichen J. Liver function and pharmacological considerations in the pathogenesis and treatment of portal hypertension. Hepatology 1990;11: Blendis L, Wong F. The hyperdynamic circulation in cirrhosis: an overview. Pharmacol Ther 2001;89: Benoit JN, Womack WA, Hernandez L, Granger DN. Forward and backward flow mechanisms of portal hypertension: relative contributions in the rat model of portal vein stenosis. Gastroenterology 1985;89: Kiel JW, Pitts L, Benoit JN, Granger DN, Shepherd AP. Reduced vascular sensitivity to norepinephrine in portal hypertensive rats. Am J Physiol 1985;248:G192 G Castro A, Jimenez W, Claria J, Ros J, Martinez JM, Bosch M, Arroyo V, Piulats J, Rivera F, Jodes J. Impaired responsiveness to angiotensin II experimental cirrhosis: role of nitric oxide. Hepatology 1993;18: Cahill PA, Foster CL, Redmond EM, Gingalewski C, Wu Y, Sitzmann JV. Enhanced nitric oxide synthase activity in portal hypertensive rabbits. Hepatology 1995;22: Cahill PA, Redmond EM, Sitzmann JV. Endothelial dysfunction in cirrhosis and portal hypertension. Pharmacol Ther 2001;89: Hori N, Wiest R, Groszmann RJ. Enhanced release of nitric oxide in response to changes in flow and shear stress in the superior mesenteric arteries of portal hypertensive rats. Hepatology 1998;28: Bruix J, Bosch J, Kravetz D, Mastai R, Rodes J. Effects of prostaglandin inhibition on systemic and hepatic hemodynamics in patients with cirrhosis of the liver. Gastroenterology 1985;88: Hou MC, Cahill PA, Zhang S, Wang YN, Hendrickson RJ, Redmond EM, Sitzmann JV. Enhanced cyclooxygenase-1 expression within the superior mesenteric artery of portal hypertensive rats: Role in the hyperdynamic circulation. Hepatology 1998;27: Vallance P, Moncada S. Hyperdynamic circulation in cirrhosis: a role for nitric oxide. Lancet 1991;337: Lee FY, Colombato LA, Albillos A, Groszmann RJ. N omega-nitro- L-arginine administration corrects peripheral vasodilation and systemic capillary hypotension and ameliorates plasma volume expansion and sodium retention in portal hypertensive rats. Hepatology 1993;17: Sieber CC, Groszmann RJ. Nitric oxide mediates hyporeactivity to vasopressors in mesenteric vessels of portal hypertensive rats. Gastroenterology 1992;103: Cahill PA, Redmond EM, Hodges R, Zhang S, Sitzmann JV. Increased endothelial nitric oxide synthase activity in the hyperemic vessels of portal hypertensive rats. J Hepatol 1996;25: Moncada S, Palmer PM, Higgs EA. Nitric oxide: physiology, pathophysiology and pharmacology. Pharmacol Rev 1991;43: Gadano AC, Sogni P, Heller J, Moreau R, Bories PN, Lebrec D. Vascular nitric oxide production during the development of two experimental models of portal hypertension. J Hepatol 1999;30: Kaviani A, Ohta M, Itani R, Sander F, Tarnawski AS, Sarfeh IJ. Tumor necrosis factor-alpha regulates inducible nitric oxide synthase expression in the portal hypertensive gastric mucosa of the rat. J Gastrointest Surg 1997;1: Shesely EG, Maeda N, Kim HS, Desai KM, Krege JH, Laubach VE, Sherman PA, Sessa WC, Smithies O. Elevated blood pressures in mice lacking endothelial nitric oxide synthase. Proc Natl Acad Sci U S A 1996;93: Laubach VE, Shesely EG, Smithies O, Sherman PA. Mice lacking inducible nitric oxide synthase are not resistant to lipopolysaccharide-induced death. Proc Natl Acad Sci U S A 1995;92: Ackerman Z, Karmeli F, Amir G, Rachmilewitz D. Gastric and colonic inflammatory and vasoactive mediators in experimental portal hypertension. Liver 1996;16: Schwarz SI. Liver. In: Schwarz SI, Shires GT, Spencer FC, Daly JM, Fisher JE, Galloway AC, eds. Principles of surgery. 7th ed. New York: MacGraw-Hill, 2000; Fluorescence Microsphere Resource Center (1999). Manual for Using Fluorescent Microspheres to Measure Organ Perfusion ( Accessed October 1, Glenny RW, Bernard S, Brinkley M. Validation of fluorescentlabeled microspheres for measurement of regional organ perfusion. J Appl Physiol 1993;74: Richer, CR, Domergue V, Gervais M, Bruneval P, Giudicelli J-F. Fluorospheres for cardiovascular phenotyping genetically modified mice. J. Cardiovascular Pharmacology 2000;36: Geraghty JG, Anderson WJ, Carter DC. Portal venous pressure,

9 1508 THEODORAKIS ET AL. GASTROENTEROLOGY Vol. 124, No. 5 and portosystemic shunt in experimental portal hypertension. Am J Physiol 1989;257:G52 G Sikuler E, Groszmann RJ. Hemodynamic studies in long- and short-term portal hypertensive rats: the relationship to systemic glucagon levels. Hepatology 1986;6: Campbell KA, Wu YP, Sitzmann JV. Modulation of portosystemic shunting by prostacyclin in portal hypertension. Curr Surg 1990; 47: Cahill PA, Wu Y, Sitzmann JV. Altered adenylyl cyclase activities and G-protein abnormalities in portal hypertensive rabbits. J Clin Invest 1994;93: Kanwar S, Kubes P, Tepperman BL, Lee SS. Nitric oxide synthase activity in portal-hypertensive and cirrhotic rats. J Hepatol 1996; 25: Martin PY, Gines P, Schrier RW. Nitric oxide as a mediator of hemodynamic abnormalities and sodium and water retention in cirrhosis. N Engl J Med 1998;339: Hartleb M, Michielsen PP, Dziurkowska-Marek A. The role of nitric oxide in portal hypertensive systemic and portal vascular pathology. Acta Gastroenterol Belg 1997;60: Ramsay MA, Spikes C, East CA, Lynch K, Hein HA, Ramsay KJ, Klintmalm GB. The perioperative management of portopulmonary hypertension with nitric oxide and epoprostenol. Anesthesiology 1999;90: Woitas RP, Heller J, Stoffel-Wagner B, Spengler U, Sauerbruch T. Renal functional reserve and nitric oxide in patients with compensated liver cirrhosis. Hepatology 1997;26: Shah V, Wiest R, Garcia-Cardena G, Cadelina G, Groszmann RJ, Sessa WC. Hsp90 regulation of endothelial nitric oxide synthase contributes to vascular control in portal hypertension. Am J Physiol 1999;277:G463 G Wiest R, Shah V, Sessa WC, Groszmann RJ. NO overproduction by enos precedes hyperdynamic splanchnic circulation in portal hypertensive rats. Am J Physiol 1999;276:G1043 G Tsugawa K, Hashizume M, Migou S, Kishihara F, Kawanaka H, Tomikawa M, Tanoue K, Sugimachi K. Role of Nitric oxide and endothelin-1 in a portal hypertensive rat model. Scand J Gastroenterol 2000;35: Bandi JC, Fernandez M, Bernadich C, De Lacy AM, Garcia-Pagan JC, Bosch J, Rodes J. Hyperkinetic circulation and decreased sensitivity to vasoconstrictors following portacaval shunt in the rat. Effects of chronic nitric oxide inhibition. J Hepatol 1999;31: Mosca P, Lee FY, Kaumann AJ, Groszmann RJ. Pharmacology of portal-systemic collaterals in portal hypertensive rats: role of endothelium. Am J Physiol 1992;263:G544 G Sieber CC, Sumanovski LT, Stumm M, van der Kooij M, Battegay E. In vivo angiogenesis in normal and portal hypertensive rats: role of basic fibroblast growth factor and nitric oxide. J Hepatol 2001;34: Mizumoto M, Arii S, Furutani M, Nakamura T, Ishigami S, Monden K, Ishiguro S, Fujita S, Imamura M. NO as an indicator of portal hemodynamics and the role of inos in increased NO production in CCl4-induced liver cirrhosis. J Surg Res 1997;70: Genesca J, Gonzalez A, Segura R, Catalan R, Marti R, Varela E, Cadelina G, Martinez M, Lopez-Talavera JC, Esteban R, Groszmann RJ, Guardia J. Interleukin-6, nitric oxide, and the clinical and hemodynamic alterations of patients with liver cirrhosis. Am J Gastroenterol 1999;94: Chu CJ, Lee FY, Wang SS, Lu RH, Tsai YT, Lin HC, Hou MC, Chan CC, Lee SD. Hyperdynamic circulation of cirrhotic rats with ascites: role of endotoxin, tumour necrosis factor-alpha and nitric oxide. Clin Sci (Lond) 1997;93: Chu CJ, Lee FY, Wang SS, Chang FY, Lin HC, Lu RH, Wu SL, Chan CC, Tai CC, Lai IN, Lee SD. Evidence against a role for endotoxin in the hyperdynamic circulation of rats with prehepatic portal hypertension. J Hepatol 1999;30: Sitzmann JV, Li SS, Adkinson NF. Evidence for role of prostacyclin as a systemic hormone in portal hypertension. Surgery 1991; 109: Wu Y, Burns RC, Sitzmann JV. Effects of nitric oxide and cyclooxygenase inhibition on splanchnic hemodynamics in portal hypertension. Hepatology 1993;18: Fernandez M, Garcia-Pagan JC, Casadevall M, Mourelle MI, Pique JM, Bosch J, Rodes J. Acute and chronic cyclooxygenase blockage in portal-hypertensive rats; influence in nitric oxide biosynthesis. Gastroenterology 1996;110: Salvemini D, Misko TP, Masferrer JL, Seibert K, Currie MG, Needleman P. Nitric oxide activates cyclooxygenase enzymes. Proc Natl Acad Sci U S A 1993;90: Iwakiri Y, Cadelina G, Sessa WC, Groszmann RJ. Mice with targeted deletion of enos develop hyperdynamic circulation associated with portal hypertension. Am J Physiol 2002;283:G1074 G1081. Received December 11, Accepted January 27, Address requests for reprints to: James V. Sitzmann, M.D., University of Rochester Medical Center, Department of Surgery, Box SURG, 601 Elmwood Avenue, Rochester, New York James_Sitzmann@urmc.rochester.edu; fax: (716)