Liver International ISSN 1478-3223
CIRRHOSIS AND ITS COMPLICATIONS
Expression of vasoactive proteins in gastric antral mucosa reflects vascular dysfunction in patients with cirrhosis and portal hypertension Jonel Trebicka1,2, Cyrus Wix1, Matthias von Heydebrand1, Kanishka Hittatiya4, Thomas Reiberger5, Sabine Klein1, Robert Schierwagen1, Glen Kristiansen4, Markus Peck-Radosavljevic5, Hans-Peter Fischer4, Søren Møller6, Flemming Bendtsen2, Aleksander Krag3 and Tilman Sauerbruch1 1 2 3 4 5
Department of Internal Medicine I, University of Bonn, Bonn, Germany Gastrounit, Medical Division, Hvidovre Hospital, University of Copenhagen, Copenhagen, Denmark Department of Gastroenterology, Odense University Hospital, University of Southern Denmark, Odense, Denmark Institute for Pathology, University of Bonn, Bonn, Germany Department of Internal Medicine III, Division of Gastroenterology & Hepatology, Vienna Hepatic Hemodynamic Lab, Medical University of Vienna, Vienna, Austria 6 Centre of Functional Imaging and Research, Department of Clinical Physiology and Nuclear Medicine, Hvidovre Hospital, University of Copenhagen, Copenhagen, Denmark
Keywords cirrhosis – portal hypertension – transjugular intrahepatic portosystemic shunt – vascular dysfunction – vasoactive pathways Abbreviations BUN, blood urea nitrogen; cGMP, cyclic guanosine monophosphate; CTP, ChildTurcotte-Pugh; GI, gastrointestinal; HVPG, hepatic-venous pressure gradient; INR, international normalized ratio; MELD, model for end-stage liver disease; MLCK, myosin light chain kinase; MLCP, myosin light chain phosphatase; PCPG, porto-caval pressure gradient; PKG, protein kinase G; ROCK, rho-kinase; SD, standard deviation; TIPS, transjugular intrahepatic portosystemic shunt; VASP, vasodilator-associated stimulated phosphoprotein. Correspondence Priv.-Doz. Dr. med. Jonel Trebicka, Department of Internal Medicine I, University of Bonn, Sigmund-Freud Str. 25, D-53105 Bonn, Germany Tel: +49 228 287 15507 Fax: +49 228 287 19718 e-mail: [email protected]
Abstract Background & Aims: Patients with cirrhosis display hypocontractility of splanchnic vessels because of dysregulation of vasoactive proteins, such as decreased effect of RhoA/ROCK and increased activity of b-Arrestin-2 and eNOS. However, it is unknown whether the dysregulation of vasoactive proteins is displayed in other vessels. We investigated whether expression of vasoactive proteins can be evaluated in gastric mucosa vessels. Methods: Biopsies from the gastric mucosa of 111 patients with cirrhosis were collected at three different centres and from 13 controls. Forty-nine patients had received TIPS. Portal pressure gradient was measured in 49 patients with TIPS and in 16 patients without TIPS. Biopsies from the antrum were conserved in formaldehyde for immunohistochemistry or shock-frozen for PCR and Western blot. Results: The mucosal transcription of vascular markers (aSMA, CD31) was higher in cirrhotic patients than controls, which was confirmed by immunohistochemistry. On average, relative mucosal levels of RhoA and ROCK were lower, while b-Arrestin-2 levels were higher in cirrhotic patients compared to controls. Transcriptional levels of eNOS increased with presence of ascites and grade of oesophageal varices. Patients with TIPS showed less pronounced markers of vascular dysfunction in gastric mucosa. Conclusion: This is the first evidence that the expression of vasoactive proteins in mucosa from the gastric antrum of patients with cirrhosis reflects their vascular dysfunction and possibly changes after therapeutic interventions.
Received 21 March 2014 Accepted 24 May 2014 DOI:10.1111/liv.12613 Liver Int. 2015; 35: 1393–1402
Cirrhosis is associated with splanchnic vascular dysfunction, as shown by an inadequate response to vasoconstrictors (1–3). This decreased vascular tone leads to an Liver International (2015) © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
increased portal venous inflow, which augments portal pressure (1–3). Furthermore, because of splanchnic vasodilatation, the effective arterial blood volume
1393
Trebicka et al.
Vasoactive pathways in gastric mucosa in cirrhosis
decreases. This leads to the compensatory activation of several vasoactive hormones, which may aggravate the hyperdynamic circulation in cirrhosis (4) and which is frequently associated with ascites formation and renal failure (5). Previous studies from our group have shown that vascular dysfunction in cirrhosis is, at least partly, caused by a defective RhoA/ROCK pathway (Fig. 1A) (1, 3, 6). Furthermore, b-Arrestin-2 expression is increased in splanchnic vessels of animals and humans with cirrhosis (2, 3). This possibly uncouples the G-protein-related pathways from the AT1 receptor and thereby blunts contraction (Fig. 1A) (2, 3). An additional factor is formation of nitric oxide (NO), a major vasodilator, increased in cirrhosis with portal hypertension because of upregulated eNOS and iNOS in the splanchnic vasculature (Fig. 1A) (7–9). To date, it is completely unknown whether the dysregulation of vasoactive protein in cirrhosis is displayed in other than mesenteric vessels. Patients with cirrhosis and portal hypertension frequently present characteristic gastric endoscopic signs of portal hypertension, the so-called portal hypertensive gastropathy (10–12). We hypothesized that this is associated with dysfunction of the splanchnic mucosal vessels and a related dysregulation of vasoactive proteins (RhoA, ROCK, b-Arrestin-2, eNOS). Thus, we investigated for the first time the expression of these vasoactive proteins in gastric mucosal biopsies obtained during upper gastrointestinal (GI) endoscopy, a routine investigation in patients with cirrhosis to assess the variceal status. Furthermore, we assessed possible changes induced by TIPS. Patients and methods Patients and data collection
We enrolled 124 patients in this study in whom upper GI endoscopy was performed for assessment of oesophageal varices in cirrhotic patients or abdominal discomfort in controls (Fig. 1B). All patients received a standardized upper GI endoscopy with biopsy of gastric antrum mucosa. The probes were conserved in formaldehyde or shock-frozen for PCR and Western blot. Biochemical parameters were analysed using standard methods. Patients with known malignancy including hepatocellular carcinoma, pregnancy, and those who refused upper GI endoscopy or had contraindications for biopsy because of severely impaired coagulation were excluded. Patients with alcohol abuse within 6 weeks prior to the study and patients with gastric antral vascular ectasia were not included in this study. The local ethics committees approved the protocols (University of Bonn 236/07, Capital Region of Copenhagen H-1-2011-151, University of Vienna EK 813/ 2009). All patients signed an informed consent in accor-
1394
dance with the Helsinki declaration before being enrolled in the study. Thirteen patients served as controls (Table 1). Of the 111 patients with cirrhosis, TIPS had been placed in 49 patients at least 6 months prior to the study, with a mean time of 12 months, in 10 of whom the biopsy was taken before and 6 months after TIPS. The gradient between the pressure in the portal vein and in the hepatic vein at the ostium to vena cava was determined in all TIPS patients (PCPG) and the gradient between wedged and free hepatic venous pressure (HVPG) was determined in additional 16 patients, who had not received a TIPS (eight from Vienna and eight from Copenhagen), as previously described (13–15). The clinical data (Table 1) were collected at patient enrolment, while presence of gastritis, helicobacter pylori infection and portal hypertensive gastropathy were diagnosed by histology. The oesophageal varices and portal hypertensive gastropathy were determined in addition macroscopically by endoscopy. Western blotting and antibodies
Samples of shock-frozen mucosal biopsies were homogenized in a buffer containing 25 mM Tris/HCl, 5 mM ethylenediamine tetraacetic acid, 10 lM phenylmethanesulphonyl fluoride, 1 mM benzamidine and 10 lg/ml leupeptin. Samples were diluted with sample buffer. Protein determination of the homogenates was performed with the Dc-Assay kit (Biorad, Munich, Germany). Samples (20 lg of protein/lane) were subjected to SDS-PAGE (15% gels for RhoA; 8% gels for ROCK, b-Arrestin-2, eNOS and p-eNOS; 10% gels for moesin, p-moesin, VASP and p-VASP), and proteins were blotted on nitrocellulose membranes. The membranes were blocked and incubated with the following primary antibodies: RhoA 119, Rock-2 H-85, NOS3, moesin, p-moesin (at threonine 558), VASP, bactin and GAPDH from Santa Cruz Biotechnology (Santa Cruz, CA, USA); peNOS (Ser 1177) from Cell Signalling (Boston, MA, USA); and p-VASP (at serine 239) clone 16C2 from Calbiochem (San Diego, CA, USA). Thereafter, the membranes were incubated with a corresponding secondary peroxidase-coupled antibody (Calbiochem, San Diego, CA, USA). Blots were developed with enhanced chemiluminescence (ECL, Amersham, UK). Intensities of the digitally detected bands were evaluated densitometrically using Chemi-Smart (PeqLab-Biotechnologies, Erlangen, Germany) as previously described (6, 16–18). Quantitative RT-PCR
RNA isolation, reverse transcription with 0.5 lg total RNA and detection by RT-PCR were performed as described previously (6, 16–18). Primers and probes for RT-PCR were obtained as a ready to use mix (RhoA, ROCK, b-Arrestin-2, eNOS, aSMA, CD31) from
Liver International (2015) © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Trebicka et al.
Vasoactive pathways in gastric mucosa in cirrhosis
(A)
(B)
Fig. 1. The schematic overview of vasoactive pathways dysregulated in extrahepatic vessels and leading to vascular hypocontractility in cirrhosis with portal hypertension (A). Vasoconstrictors fail to elicit adequate contraction in cirrhotic splanchnic vessels because of decreased activity of RhoA and ROCK (read out: phosphorylation of moesin at threonine 558) and thereby impaired inhibition of myosin light chain phosphatase (MLCP), which leads to relaxation of vessels. Furthermore, the upregulation of b-Arrestin-2 uncouples the G-protein-mediated RhoA/ROCK-pathway and additionally worsens the contractility. Because of several factors, eNOS expression and activity in endothelial cells are enhanced. Nitric oxide (NO) induces cyclic guanosine monophosphate (cGMP) formation and activates protein kinase G (PKG) as shown by the phosphorylation of its substrate vasodilator-associated stimulated phosphoprotein at serine 239. Activated PKG in turn further enhances the activity of MLCP and aggravates vasodilation. The recruitment overview is presented in Fig. 1B. One hundred and twenty-four patients were recruited. Thirteen patients were noncirrhotic controls recruited in Bonn. Sixty-two of the 111 cirrhotic patients did not receive TIPS insertion. Sixteen of these patients were enrolled at the Hvidovre Hospital, University of Copenhagen (n = 8) and at the Medical University of Vienna (n = 8) and had their hepatic-venous pressure gradient (HVPG) measured. Forty-nine patients were enrolled at the University of Bonn at least 6 months after receiving TIPS and measurement of porto-caval pressure gradient (PCPG).
Applied Biosystems (Foster City, CA, USA). 18S rRNA served as the endogenous control (ready to use mix of primers and probes from Applied Biosystems). RT-PCR (ABI 7300 sequence detector) and PCR reaction (2 9 TaqMan-PCR-mastermix, Applied Biosystems) were performed as described previously (16). For each of the genes, a validation experiment was performed. Efficiencies of RT-PCR for the target gene and the endogenous Liver International (2015) © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
control were approximately equal. DCT expresses the difference between the number of cycles (CT) of the target genes and the endogenous control. For better understanding, we normalized all DCT values to those of aSMA for RhoA, ROCK and b-Arrestin-2 and to CD31 for eNOS. The results of liver samples were expressed as 2 DDCt, and express the x-fold increase of gene expression compared to aSMA or CD31.
1395
Trebicka et al.
Vasoactive pathways in gastric mucosa in cirrhosis
Table 1. Demographic and clinical characteristics of all patients (n = 124)
Parameters Gender (female/male) Age (in years) Median (range) Gastritis n (%) Helicobacter pylori infection n (%) Proton-pump inhibitors n (%) Portal hypertensive gastropathy none/mild/severe Child-Score median (range) Child category (A/B/C) MELD-score Median (range) Aetiology (alcoholic/chronic viral hepatitis/other) Ascites (absent/present) Oesophageal varices (no or small/ large) History of bleeding n (%)
Noncirrhotic controls n = 13
Cirrhotic patients n = 111
6/7 52 (24–78)
33/78 57 (28–79)
11 (85) 2 (15) 11 (85) 13/0/0
63 (57) 3 (3) 85 (77) 14/92/5
–
7 (5–11)
– –
44/64/3 11 (6–22)
–
80/18/13
– –
67/44 79/32
–
40 (32)
aSMA and CD34 immunohistochemistry
Cryosections from mucosa biopsy (20 lm) were fixed in 2% formaldehyde und 0.2% glutaraldehyde. For immunohistochemical staining of a-smooth muscle actin (a-SMA), slides with sections were incubated with a mouse anti-SMA antibody (clone 1A4; Sigma-Aldrich GmbH, Munich, Germany) diluted 1:600 in Tris-buffered saline for 60 min. For immunohistochemical staining of CD34, sections were incubated with a monoclonal mouse CD34 antibody (Clone QBEnd-10; DakoCytomation) diluted 1:200 in Tris-buffered saline for 30 min. A biotinylated rabbit anti-mouse secondary antibody absorbed with rat serum (Dako, Glostrup, Denmark) was applied (1:200, 45 min) and complexed with strepdavidin-conjugated alkaline phosphatase (1:200, 45 min; Dako). The stain was developed in fresh fuchsin naphtol AS-BI solution (Sigma-Aldrich). For the morphometric analysis of stained specimens, at least 10 mm² of liver tissue were analysed by means of computational analysis (Histoquant, 3DHistech, Budapest, Hungary) as described (18–20). Statistical analysis
Data are presented as mean ± standard deviation (SD) or median (range). The Mann–Whitney-U test was used for statistical comparisons. Correlations were analysed by the Spearman correlation coefficient. Two-sided P-values
CIRRHOSIS AND ITS COMPLICATIONS
Expression of vasoactive proteins in gastric antral mucosa reflects vascular dysfunction in patients with cirrhosis and portal hypertension Jonel Trebicka1,2, Cyrus Wix1, Matthias von Heydebrand1, Kanishka Hittatiya4, Thomas Reiberger5, Sabine Klein1, Robert Schierwagen1, Glen Kristiansen4, Markus Peck-Radosavljevic5, Hans-Peter Fischer4, Søren Møller6, Flemming Bendtsen2, Aleksander Krag3 and Tilman Sauerbruch1 1 2 3 4 5
Department of Internal Medicine I, University of Bonn, Bonn, Germany Gastrounit, Medical Division, Hvidovre Hospital, University of Copenhagen, Copenhagen, Denmark Department of Gastroenterology, Odense University Hospital, University of Southern Denmark, Odense, Denmark Institute for Pathology, University of Bonn, Bonn, Germany Department of Internal Medicine III, Division of Gastroenterology & Hepatology, Vienna Hepatic Hemodynamic Lab, Medical University of Vienna, Vienna, Austria 6 Centre of Functional Imaging and Research, Department of Clinical Physiology and Nuclear Medicine, Hvidovre Hospital, University of Copenhagen, Copenhagen, Denmark
Keywords cirrhosis – portal hypertension – transjugular intrahepatic portosystemic shunt – vascular dysfunction – vasoactive pathways Abbreviations BUN, blood urea nitrogen; cGMP, cyclic guanosine monophosphate; CTP, ChildTurcotte-Pugh; GI, gastrointestinal; HVPG, hepatic-venous pressure gradient; INR, international normalized ratio; MELD, model for end-stage liver disease; MLCK, myosin light chain kinase; MLCP, myosin light chain phosphatase; PCPG, porto-caval pressure gradient; PKG, protein kinase G; ROCK, rho-kinase; SD, standard deviation; TIPS, transjugular intrahepatic portosystemic shunt; VASP, vasodilator-associated stimulated phosphoprotein. Correspondence Priv.-Doz. Dr. med. Jonel Trebicka, Department of Internal Medicine I, University of Bonn, Sigmund-Freud Str. 25, D-53105 Bonn, Germany Tel: +49 228 287 15507 Fax: +49 228 287 19718 e-mail: [email protected]
Abstract Background & Aims: Patients with cirrhosis display hypocontractility of splanchnic vessels because of dysregulation of vasoactive proteins, such as decreased effect of RhoA/ROCK and increased activity of b-Arrestin-2 and eNOS. However, it is unknown whether the dysregulation of vasoactive proteins is displayed in other vessels. We investigated whether expression of vasoactive proteins can be evaluated in gastric mucosa vessels. Methods: Biopsies from the gastric mucosa of 111 patients with cirrhosis were collected at three different centres and from 13 controls. Forty-nine patients had received TIPS. Portal pressure gradient was measured in 49 patients with TIPS and in 16 patients without TIPS. Biopsies from the antrum were conserved in formaldehyde for immunohistochemistry or shock-frozen for PCR and Western blot. Results: The mucosal transcription of vascular markers (aSMA, CD31) was higher in cirrhotic patients than controls, which was confirmed by immunohistochemistry. On average, relative mucosal levels of RhoA and ROCK were lower, while b-Arrestin-2 levels were higher in cirrhotic patients compared to controls. Transcriptional levels of eNOS increased with presence of ascites and grade of oesophageal varices. Patients with TIPS showed less pronounced markers of vascular dysfunction in gastric mucosa. Conclusion: This is the first evidence that the expression of vasoactive proteins in mucosa from the gastric antrum of patients with cirrhosis reflects their vascular dysfunction and possibly changes after therapeutic interventions.
Received 21 March 2014 Accepted 24 May 2014 DOI:10.1111/liv.12613 Liver Int. 2015; 35: 1393–1402
Cirrhosis is associated with splanchnic vascular dysfunction, as shown by an inadequate response to vasoconstrictors (1–3). This decreased vascular tone leads to an Liver International (2015) © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
increased portal venous inflow, which augments portal pressure (1–3). Furthermore, because of splanchnic vasodilatation, the effective arterial blood volume
1393
Trebicka et al.
Vasoactive pathways in gastric mucosa in cirrhosis
decreases. This leads to the compensatory activation of several vasoactive hormones, which may aggravate the hyperdynamic circulation in cirrhosis (4) and which is frequently associated with ascites formation and renal failure (5). Previous studies from our group have shown that vascular dysfunction in cirrhosis is, at least partly, caused by a defective RhoA/ROCK pathway (Fig. 1A) (1, 3, 6). Furthermore, b-Arrestin-2 expression is increased in splanchnic vessels of animals and humans with cirrhosis (2, 3). This possibly uncouples the G-protein-related pathways from the AT1 receptor and thereby blunts contraction (Fig. 1A) (2, 3). An additional factor is formation of nitric oxide (NO), a major vasodilator, increased in cirrhosis with portal hypertension because of upregulated eNOS and iNOS in the splanchnic vasculature (Fig. 1A) (7–9). To date, it is completely unknown whether the dysregulation of vasoactive protein in cirrhosis is displayed in other than mesenteric vessels. Patients with cirrhosis and portal hypertension frequently present characteristic gastric endoscopic signs of portal hypertension, the so-called portal hypertensive gastropathy (10–12). We hypothesized that this is associated with dysfunction of the splanchnic mucosal vessels and a related dysregulation of vasoactive proteins (RhoA, ROCK, b-Arrestin-2, eNOS). Thus, we investigated for the first time the expression of these vasoactive proteins in gastric mucosal biopsies obtained during upper gastrointestinal (GI) endoscopy, a routine investigation in patients with cirrhosis to assess the variceal status. Furthermore, we assessed possible changes induced by TIPS. Patients and methods Patients and data collection
We enrolled 124 patients in this study in whom upper GI endoscopy was performed for assessment of oesophageal varices in cirrhotic patients or abdominal discomfort in controls (Fig. 1B). All patients received a standardized upper GI endoscopy with biopsy of gastric antrum mucosa. The probes were conserved in formaldehyde or shock-frozen for PCR and Western blot. Biochemical parameters were analysed using standard methods. Patients with known malignancy including hepatocellular carcinoma, pregnancy, and those who refused upper GI endoscopy or had contraindications for biopsy because of severely impaired coagulation were excluded. Patients with alcohol abuse within 6 weeks prior to the study and patients with gastric antral vascular ectasia were not included in this study. The local ethics committees approved the protocols (University of Bonn 236/07, Capital Region of Copenhagen H-1-2011-151, University of Vienna EK 813/ 2009). All patients signed an informed consent in accor-
1394
dance with the Helsinki declaration before being enrolled in the study. Thirteen patients served as controls (Table 1). Of the 111 patients with cirrhosis, TIPS had been placed in 49 patients at least 6 months prior to the study, with a mean time of 12 months, in 10 of whom the biopsy was taken before and 6 months after TIPS. The gradient between the pressure in the portal vein and in the hepatic vein at the ostium to vena cava was determined in all TIPS patients (PCPG) and the gradient between wedged and free hepatic venous pressure (HVPG) was determined in additional 16 patients, who had not received a TIPS (eight from Vienna and eight from Copenhagen), as previously described (13–15). The clinical data (Table 1) were collected at patient enrolment, while presence of gastritis, helicobacter pylori infection and portal hypertensive gastropathy were diagnosed by histology. The oesophageal varices and portal hypertensive gastropathy were determined in addition macroscopically by endoscopy. Western blotting and antibodies
Samples of shock-frozen mucosal biopsies were homogenized in a buffer containing 25 mM Tris/HCl, 5 mM ethylenediamine tetraacetic acid, 10 lM phenylmethanesulphonyl fluoride, 1 mM benzamidine and 10 lg/ml leupeptin. Samples were diluted with sample buffer. Protein determination of the homogenates was performed with the Dc-Assay kit (Biorad, Munich, Germany). Samples (20 lg of protein/lane) were subjected to SDS-PAGE (15% gels for RhoA; 8% gels for ROCK, b-Arrestin-2, eNOS and p-eNOS; 10% gels for moesin, p-moesin, VASP and p-VASP), and proteins were blotted on nitrocellulose membranes. The membranes were blocked and incubated with the following primary antibodies: RhoA 119, Rock-2 H-85, NOS3, moesin, p-moesin (at threonine 558), VASP, bactin and GAPDH from Santa Cruz Biotechnology (Santa Cruz, CA, USA); peNOS (Ser 1177) from Cell Signalling (Boston, MA, USA); and p-VASP (at serine 239) clone 16C2 from Calbiochem (San Diego, CA, USA). Thereafter, the membranes were incubated with a corresponding secondary peroxidase-coupled antibody (Calbiochem, San Diego, CA, USA). Blots were developed with enhanced chemiluminescence (ECL, Amersham, UK). Intensities of the digitally detected bands were evaluated densitometrically using Chemi-Smart (PeqLab-Biotechnologies, Erlangen, Germany) as previously described (6, 16–18). Quantitative RT-PCR
RNA isolation, reverse transcription with 0.5 lg total RNA and detection by RT-PCR were performed as described previously (6, 16–18). Primers and probes for RT-PCR were obtained as a ready to use mix (RhoA, ROCK, b-Arrestin-2, eNOS, aSMA, CD31) from
Liver International (2015) © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Trebicka et al.
Vasoactive pathways in gastric mucosa in cirrhosis
(A)
(B)
Fig. 1. The schematic overview of vasoactive pathways dysregulated in extrahepatic vessels and leading to vascular hypocontractility in cirrhosis with portal hypertension (A). Vasoconstrictors fail to elicit adequate contraction in cirrhotic splanchnic vessels because of decreased activity of RhoA and ROCK (read out: phosphorylation of moesin at threonine 558) and thereby impaired inhibition of myosin light chain phosphatase (MLCP), which leads to relaxation of vessels. Furthermore, the upregulation of b-Arrestin-2 uncouples the G-protein-mediated RhoA/ROCK-pathway and additionally worsens the contractility. Because of several factors, eNOS expression and activity in endothelial cells are enhanced. Nitric oxide (NO) induces cyclic guanosine monophosphate (cGMP) formation and activates protein kinase G (PKG) as shown by the phosphorylation of its substrate vasodilator-associated stimulated phosphoprotein at serine 239. Activated PKG in turn further enhances the activity of MLCP and aggravates vasodilation. The recruitment overview is presented in Fig. 1B. One hundred and twenty-four patients were recruited. Thirteen patients were noncirrhotic controls recruited in Bonn. Sixty-two of the 111 cirrhotic patients did not receive TIPS insertion. Sixteen of these patients were enrolled at the Hvidovre Hospital, University of Copenhagen (n = 8) and at the Medical University of Vienna (n = 8) and had their hepatic-venous pressure gradient (HVPG) measured. Forty-nine patients were enrolled at the University of Bonn at least 6 months after receiving TIPS and measurement of porto-caval pressure gradient (PCPG).
Applied Biosystems (Foster City, CA, USA). 18S rRNA served as the endogenous control (ready to use mix of primers and probes from Applied Biosystems). RT-PCR (ABI 7300 sequence detector) and PCR reaction (2 9 TaqMan-PCR-mastermix, Applied Biosystems) were performed as described previously (16). For each of the genes, a validation experiment was performed. Efficiencies of RT-PCR for the target gene and the endogenous Liver International (2015) © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
control were approximately equal. DCT expresses the difference between the number of cycles (CT) of the target genes and the endogenous control. For better understanding, we normalized all DCT values to those of aSMA for RhoA, ROCK and b-Arrestin-2 and to CD31 for eNOS. The results of liver samples were expressed as 2 DDCt, and express the x-fold increase of gene expression compared to aSMA or CD31.
1395
Trebicka et al.
Vasoactive pathways in gastric mucosa in cirrhosis
Table 1. Demographic and clinical characteristics of all patients (n = 124)
Parameters Gender (female/male) Age (in years) Median (range) Gastritis n (%) Helicobacter pylori infection n (%) Proton-pump inhibitors n (%) Portal hypertensive gastropathy none/mild/severe Child-Score median (range) Child category (A/B/C) MELD-score Median (range) Aetiology (alcoholic/chronic viral hepatitis/other) Ascites (absent/present) Oesophageal varices (no or small/ large) History of bleeding n (%)
Noncirrhotic controls n = 13
Cirrhotic patients n = 111
6/7 52 (24–78)
33/78 57 (28–79)
11 (85) 2 (15) 11 (85) 13/0/0
63 (57) 3 (3) 85 (77) 14/92/5
–
7 (5–11)
– –
44/64/3 11 (6–22)
–
80/18/13
– –
67/44 79/32
–
40 (32)
aSMA and CD34 immunohistochemistry
Cryosections from mucosa biopsy (20 lm) were fixed in 2% formaldehyde und 0.2% glutaraldehyde. For immunohistochemical staining of a-smooth muscle actin (a-SMA), slides with sections were incubated with a mouse anti-SMA antibody (clone 1A4; Sigma-Aldrich GmbH, Munich, Germany) diluted 1:600 in Tris-buffered saline for 60 min. For immunohistochemical staining of CD34, sections were incubated with a monoclonal mouse CD34 antibody (Clone QBEnd-10; DakoCytomation) diluted 1:200 in Tris-buffered saline for 30 min. A biotinylated rabbit anti-mouse secondary antibody absorbed with rat serum (Dako, Glostrup, Denmark) was applied (1:200, 45 min) and complexed with strepdavidin-conjugated alkaline phosphatase (1:200, 45 min; Dako). The stain was developed in fresh fuchsin naphtol AS-BI solution (Sigma-Aldrich). For the morphometric analysis of stained specimens, at least 10 mm² of liver tissue were analysed by means of computational analysis (Histoquant, 3DHistech, Budapest, Hungary) as described (18–20). Statistical analysis
Data are presented as mean ± standard deviation (SD) or median (range). The Mann–Whitney-U test was used for statistical comparisons. Correlations were analysed by the Spearman correlation coefficient. Two-sided P-values
