- Email: [email protected]
Intestinal expression of genes involved in iron absorption in humans: effects of hemochromatosis and anemia
HEPATOLOGYVol. 34, No. 4, Pt. 2, 2001
AASLD ABSTRACTS
207A
133
134
AGE OF LIVER DONOR AND NORMAL HISTOLOGY 10 YEARS AFTER LIVER TRANSPLANTATION. K Rifai, M Sebagh, M Reynes, F Saliba, F Saliba, V Karam, H Bismuth, D Samuel, C Feray, Paul Brousse Hospital, Viflejuif France
MELD AND CTP SCORES ARE EQUIVALENT PREDICTORS OF MORTALITY IN CIRRHOTIC VETERANS REFERRED FOR ORTFIOTOPIC LIVER TRANSPLANTATION(OLT). Souheil G Abouassi, Anastasios A Mihas, Leslie M Williams, Hochong S Gflles, Douglas M Heuman, VCU Health System and MeGuire DVAMC, Richmond, VA
While long-term survival after liver transplantation is now well-defined, the long-term histological prognosis after liver transplantation remained unknown in the absence of routine biopsies. Numerous factors such as rejection and recurrent liver diseases likely influence the long term histological prognosis. We herein report the factors influencing the liver histology 10-year after a first transplantation. Patients and methods: All the120 available biopsies were reviewed by two pathologists. They were performed during the 10 th year post-transplantation in 58 men and 62 women who had been transplanted at a mean age of 48 -+ 13 years for cirrhosis due to HDV (23), HBV (14) and HCV (11), primary biliary cirrhosis (22), acute hepatic failure (21), autoimmune cirrhosis (10), hepatoceflular carcinoma (10) or other diagnoses (9). Transplantation was non ABO-identical in 12 cases. Results. 31/120 (25%) available biopsies were normal. Fibrosis and ductopenia were noticed in 69 (57%) and 41 (34 %)/120 patients respectivelly. In univariate analysis, normal histology was positively correlated to normal liver tests, absence of infection by HCV(recurrent in 9 or acquired in 44) and young age of donor. Young age of the donor was the only independent factor in multivariate analysis (p =0.03). Ductopenia was positively correlated to PBC as indication (p=0.01), to CMV mismatching (positive donor in negative recipient: p=0.02) and non ABO-identical transplantation (p=0.01). In multivariate analysis, non ABO-identical transplantation was the only independent factor (p=0.01) of ductopenia, Fibrosis score was correlated in multivariate analysis to HCV (p<0.001) and to non ABOidentical transplantation (p = 0.01 ). Conclusion. Long-term histology of liver graft is frequently abnormal due to recurrent or acquired liver diseases and is stongly i n f u e n c e d by the age of donor and ABO-matching. Since ABO matching is applied worldwide, the possibility to transplant graft from young donors to young recipients should be considered.
Background: Indices for predicting survival in patients with cirrhosis are critically importam in assigning priority for OLT. We examined the accuracy of three commonly used indices (the originalMELD, MELD as modified by UNOS, and CTP scores) as predictors of survival in 140 patients referred for OLT from 56 centers through the US Department of Veterans Affairs between 1997 and 2000. Methods: Scores were calculated based on most recent laboratory and clinical data included with the initial referral. Survival was determined from the date of the most recent liver function tests included in the referral package. Receiver operating curves (ROC's) for MELD and CTP scores as predictors of survival were calculated for survival at 45, 90, 120, 180, 365 and 730 days with censoring at transplantation; alternatively ROC's were calculated for death or transplantation at 90 or 180 days. We identified cutoffvalues of each score providing equivalent sensitivity and specificity for prediction of death within 180 days, and we determined OLT-free 2 year survival as a function of MELD or CTP score. Results: 46 patients died without OLT. Time between referral and death ranged from 8 days to 2 years. Both MELD and CTP scores were good predictors of mortality and/or need for early liver transplantation, with accuracy best within the first 120 days. Areas under the ROC (+SEM) for MELD and CTP scores in predicting mortality are shown in the table. Although a trend to greater accuracy was observed with the MELD scores through 180 days, differences between indices did not achieve statistical significance at any time point. CTP>= 12, MELD (original) > 18, or MELD (UNOS) > 13 defne a population representing approximately 20% of VA transplant referrals that have a poor 2-year prognosis (75% mortality without liver transplantation) and a high early mortality. CTP < or = 8, MELD (original) <10, or MELD (UNOS) <7 identify a group representing about 40% of our OLT referrals with 2 year mortality of <20%.Similar results were obtained when patients with referred diagnosis of hepatocellular carcinoma (n= 15) were excluded from the analysis.Conclusions: For the first 120 days, MELD (original or UNOS-modified) may be slightly more accurate than CTP score in predicting mortality; after 120 days, MELD and CTP scores are equally good predictors of survival in veterans referred for OLT.MELD UNOS values > 7 and >13 are equivalent to CTP scores of > = 9 and > = 12.
Areas under the ROC(±SEM)for MELD and CTP scores in predictingmortality Mortality at N(# of deaths) MELD(UNOS) MELD(original) CTP 45 days 90 days 120 days 180 days 365 days 730 days
9 13 14 22 32 46
.887 ± .045 .847 ± ,070 .855 ± ,085 ,791 _+.058 .744 ± .049 ,703 _+.045
,874 ± ,838 + ,847 ± .799 ± ,737 ±
,055 .062 ,058 .055 .050 .726 ± ,044
181i _+,061 .805 ± ,058 .813 + ,055 ,769 ± ,049 ,719 ± ,046 .702 _+.044
135
136
INTESTINAL EXPRESSION OF GENES INVOLVED IN IRON ABSORPTION IN HUMANS: EFFECTS OF HEMOCHROMATOSIS AND ANEMIA. Herbert Bonkovsky, Dept of Medicine and Liver-Biliary-Pancreatie Ctr, Univ of Mass Medical Sch, Worcester, MA; Andreas Rolls, Membrane Biology Program, Harvard Medical Sch, Boston, MA; James Kohlroser, Kristina McNeal, Ashish Sharma, Dept of Medicine and Liver-Biliary-Pancreatic Ctr, Univ of Mass Medical Sch, Worcester, MA; Urs Berger, Mathias Hediger, Membrane Biology Program, Harvard Medical Sch, Boston, MA
HAPTOGLOBIN PFIENOTYPE 2-2 OVERREPRESENTATION IN CYS282TYR ItEMOCHROMATOTIC PATIENTS. Hans Van Vlierberghe, Michel Langlois, Joris Delanghe, Geert Leroux-Roels, Martine De Vos, Ghent University Hospital, Gent Belgium
Background: Hereditary hemochromatosis (HHC) is one of the most frequent genetic disorders in humans. It is primarily caused by a missense mutation (C282Y) in the HFE gene, which triggers excessive intestinal iron absorption, followed by toxic accumulation of iron in the liver and other organs. In the normal intestine, the absorption of iron is tightly regulated by the tissues with highest iron demand, in particular bone marrow. In HHC, this regulation is disturbed, leading to excessive iron absorption despite iron overload. Recent cloning and study of the intestinal iron transporter proteins, DCT1 (apical) and IREG1/hephaestin (basolateral), provided new insight into the mechanisms and regulation of intestinal iron absorption. In rodents, both DCT1 and IREG1 are regulated by iron, probably in a similar fashion as for the transferrin receptor and ferritin, i.e., through iron regulatory elements and proteins. Aim: To assess whether in man the two transporters are also regulated in an iron dependent manner, and whether this regulation is disturbed in HHC. Methods: Using quantitative RT-PCR, we measured mRNA expression of DCT1 (both the IRE-containing and non-IRE forms), IREG, and hephaestin in duodenal biopsy samples from patients with HHC (C282Y + / + , n = 8 ) , iron-lack anemia ( n = 7 ) and subjects with neither (n=32). Results: In the latter group, we found inverse relationships between DCTI-IRE and blood hemoglobin (r2=0.42, p=0.0003), serum iron saturation (r2=0.21, p=0.03), or ferritin (r2=0.51, p=0.0001). Only the DCT1 splice form containing an iron-responsive element (IRE) showed iron-dependent regulation. Subjects with iron deficiency exhibited up to 10-fold increases in expression of DCTI-IRE mRNA expression. In C282Y + / + subjects the inverse relationship was shifted to higher DCTI-IRE levels, despite high serum iron saturation. The basolateral iron transporter IREG1, but not hephaestin, exhibited a similar inverse correlation with serum ferritin (r2=0.22, p=0.0001). Conclusions: DCTI-IRE and IREG1 mRNA expression are regulated in an iron-dependent manner, in humans as in other mammals, while the mRNAs of the non-IRE containing DCT1 and hephaestin are not affected by iron status. In HHC patients, the mRNA expression level of DCT1 remains high despite elevated serum transferrin saturation and ferritin levels. The lack of appropriate down regulation of this apical iron transporter in the duodenum leads to excessive iron absorption in persons with HHC.
Introduction. Patients with genotypic Cys282Tyr homozygous hemochromatosis differ largerly in phenotypic presentation. The HFE mutation on itself does not explain the different manifestations of hemochromatosis. We hypothesized that the genetic haptoglobin ( H p ) polymorphism, because of its effect on iron metabolism, could be a modifying factor that influences the clinical presentation of hereditary hemochromatosis. Materials and methods. In 167 Cys282Tyr homozygous hemochromatotic patients, the frequencies of Hp types ( 1-1, 2-1 and 2-2) and alleles (Hp 1, Hp2) were compared with those in 918 healthy objects. Clinical and laboratory indices of iron overload, occurrence of liver fibrosis and age at diagnosis were incorporated in the analysis. Results. The H p l alllele frequency was low among hemochromatotic patients compared to the healthy reference population, due to an overrepresentation of the Hp 2-2 type (p<0.01). Male patients carrying Hp 2-2 had higher serum iron (p=0.003) and ferritin levels (p=0.03) than those with a Hp 1-1 or 2-1 type. The amount of iron removed with phlebotomy was also higher in Hp 2-2 patients ( p = 0 . 0 3 ) . No haptoglobin type-related differences were found when considering age at presentation, hepatic iron index, occurrence of clinical symptoms or liver fibrosis. Conclusion. The Hp 2-2 type is overrepresented among Cys282Tyr homozygous hemochromatotic patients. At diagnosis, iron overload was more pronounced in male patients carrying Hp 2-2. Our data suggest that Hp polymorphism is one of the modifier factors in the phenotypic expression of hereditary hemochromatosis.
AASLD ABSTRACTS
207A
133
134
AGE OF LIVER DONOR AND NORMAL HISTOLOGY 10 YEARS AFTER LIVER TRANSPLANTATION. K Rifai, M Sebagh, M Reynes, F Saliba, F Saliba, V Karam, H Bismuth, D Samuel, C Feray, Paul Brousse Hospital, Viflejuif France
MELD AND CTP SCORES ARE EQUIVALENT PREDICTORS OF MORTALITY IN CIRRHOTIC VETERANS REFERRED FOR ORTFIOTOPIC LIVER TRANSPLANTATION(OLT). Souheil G Abouassi, Anastasios A Mihas, Leslie M Williams, Hochong S Gflles, Douglas M Heuman, VCU Health System and MeGuire DVAMC, Richmond, VA
While long-term survival after liver transplantation is now well-defined, the long-term histological prognosis after liver transplantation remained unknown in the absence of routine biopsies. Numerous factors such as rejection and recurrent liver diseases likely influence the long term histological prognosis. We herein report the factors influencing the liver histology 10-year after a first transplantation. Patients and methods: All the120 available biopsies were reviewed by two pathologists. They were performed during the 10 th year post-transplantation in 58 men and 62 women who had been transplanted at a mean age of 48 -+ 13 years for cirrhosis due to HDV (23), HBV (14) and HCV (11), primary biliary cirrhosis (22), acute hepatic failure (21), autoimmune cirrhosis (10), hepatoceflular carcinoma (10) or other diagnoses (9). Transplantation was non ABO-identical in 12 cases. Results. 31/120 (25%) available biopsies were normal. Fibrosis and ductopenia were noticed in 69 (57%) and 41 (34 %)/120 patients respectivelly. In univariate analysis, normal histology was positively correlated to normal liver tests, absence of infection by HCV(recurrent in 9 or acquired in 44) and young age of donor. Young age of the donor was the only independent factor in multivariate analysis (p =0.03). Ductopenia was positively correlated to PBC as indication (p=0.01), to CMV mismatching (positive donor in negative recipient: p=0.02) and non ABO-identical transplantation (p=0.01). In multivariate analysis, non ABO-identical transplantation was the only independent factor (p=0.01) of ductopenia, Fibrosis score was correlated in multivariate analysis to HCV (p<0.001) and to non ABOidentical transplantation (p = 0.01 ). Conclusion. Long-term histology of liver graft is frequently abnormal due to recurrent or acquired liver diseases and is stongly i n f u e n c e d by the age of donor and ABO-matching. Since ABO matching is applied worldwide, the possibility to transplant graft from young donors to young recipients should be considered.
Background: Indices for predicting survival in patients with cirrhosis are critically importam in assigning priority for OLT. We examined the accuracy of three commonly used indices (the originalMELD, MELD as modified by UNOS, and CTP scores) as predictors of survival in 140 patients referred for OLT from 56 centers through the US Department of Veterans Affairs between 1997 and 2000. Methods: Scores were calculated based on most recent laboratory and clinical data included with the initial referral. Survival was determined from the date of the most recent liver function tests included in the referral package. Receiver operating curves (ROC's) for MELD and CTP scores as predictors of survival were calculated for survival at 45, 90, 120, 180, 365 and 730 days with censoring at transplantation; alternatively ROC's were calculated for death or transplantation at 90 or 180 days. We identified cutoffvalues of each score providing equivalent sensitivity and specificity for prediction of death within 180 days, and we determined OLT-free 2 year survival as a function of MELD or CTP score. Results: 46 patients died without OLT. Time between referral and death ranged from 8 days to 2 years. Both MELD and CTP scores were good predictors of mortality and/or need for early liver transplantation, with accuracy best within the first 120 days. Areas under the ROC (+SEM) for MELD and CTP scores in predicting mortality are shown in the table. Although a trend to greater accuracy was observed with the MELD scores through 180 days, differences between indices did not achieve statistical significance at any time point. CTP>= 12, MELD (original) > 18, or MELD (UNOS) > 13 defne a population representing approximately 20% of VA transplant referrals that have a poor 2-year prognosis (75% mortality without liver transplantation) and a high early mortality. CTP < or = 8, MELD (original) <10, or MELD (UNOS) <7 identify a group representing about 40% of our OLT referrals with 2 year mortality of <20%.Similar results were obtained when patients with referred diagnosis of hepatocellular carcinoma (n= 15) were excluded from the analysis.Conclusions: For the first 120 days, MELD (original or UNOS-modified) may be slightly more accurate than CTP score in predicting mortality; after 120 days, MELD and CTP scores are equally good predictors of survival in veterans referred for OLT.MELD UNOS values > 7 and >13 are equivalent to CTP scores of > = 9 and > = 12.
Areas under the ROC(±SEM)for MELD and CTP scores in predictingmortality Mortality at N(# of deaths) MELD(UNOS) MELD(original) CTP 45 days 90 days 120 days 180 days 365 days 730 days
9 13 14 22 32 46
.887 ± .045 .847 ± ,070 .855 ± ,085 ,791 _+.058 .744 ± .049 ,703 _+.045
,874 ± ,838 + ,847 ± .799 ± ,737 ±
,055 .062 ,058 .055 .050 .726 ± ,044
181i _+,061 .805 ± ,058 .813 + ,055 ,769 ± ,049 ,719 ± ,046 .702 _+.044
135
136
INTESTINAL EXPRESSION OF GENES INVOLVED IN IRON ABSORPTION IN HUMANS: EFFECTS OF HEMOCHROMATOSIS AND ANEMIA. Herbert Bonkovsky, Dept of Medicine and Liver-Biliary-Pancreatie Ctr, Univ of Mass Medical Sch, Worcester, MA; Andreas Rolls, Membrane Biology Program, Harvard Medical Sch, Boston, MA; James Kohlroser, Kristina McNeal, Ashish Sharma, Dept of Medicine and Liver-Biliary-Pancreatic Ctr, Univ of Mass Medical Sch, Worcester, MA; Urs Berger, Mathias Hediger, Membrane Biology Program, Harvard Medical Sch, Boston, MA
HAPTOGLOBIN PFIENOTYPE 2-2 OVERREPRESENTATION IN CYS282TYR ItEMOCHROMATOTIC PATIENTS. Hans Van Vlierberghe, Michel Langlois, Joris Delanghe, Geert Leroux-Roels, Martine De Vos, Ghent University Hospital, Gent Belgium
Background: Hereditary hemochromatosis (HHC) is one of the most frequent genetic disorders in humans. It is primarily caused by a missense mutation (C282Y) in the HFE gene, which triggers excessive intestinal iron absorption, followed by toxic accumulation of iron in the liver and other organs. In the normal intestine, the absorption of iron is tightly regulated by the tissues with highest iron demand, in particular bone marrow. In HHC, this regulation is disturbed, leading to excessive iron absorption despite iron overload. Recent cloning and study of the intestinal iron transporter proteins, DCT1 (apical) and IREG1/hephaestin (basolateral), provided new insight into the mechanisms and regulation of intestinal iron absorption. In rodents, both DCT1 and IREG1 are regulated by iron, probably in a similar fashion as for the transferrin receptor and ferritin, i.e., through iron regulatory elements and proteins. Aim: To assess whether in man the two transporters are also regulated in an iron dependent manner, and whether this regulation is disturbed in HHC. Methods: Using quantitative RT-PCR, we measured mRNA expression of DCT1 (both the IRE-containing and non-IRE forms), IREG, and hephaestin in duodenal biopsy samples from patients with HHC (C282Y + / + , n = 8 ) , iron-lack anemia ( n = 7 ) and subjects with neither (n=32). Results: In the latter group, we found inverse relationships between DCTI-IRE and blood hemoglobin (r2=0.42, p=0.0003), serum iron saturation (r2=0.21, p=0.03), or ferritin (r2=0.51, p=0.0001). Only the DCT1 splice form containing an iron-responsive element (IRE) showed iron-dependent regulation. Subjects with iron deficiency exhibited up to 10-fold increases in expression of DCTI-IRE mRNA expression. In C282Y + / + subjects the inverse relationship was shifted to higher DCTI-IRE levels, despite high serum iron saturation. The basolateral iron transporter IREG1, but not hephaestin, exhibited a similar inverse correlation with serum ferritin (r2=0.22, p=0.0001). Conclusions: DCTI-IRE and IREG1 mRNA expression are regulated in an iron-dependent manner, in humans as in other mammals, while the mRNAs of the non-IRE containing DCT1 and hephaestin are not affected by iron status. In HHC patients, the mRNA expression level of DCT1 remains high despite elevated serum transferrin saturation and ferritin levels. The lack of appropriate down regulation of this apical iron transporter in the duodenum leads to excessive iron absorption in persons with HHC.
Introduction. Patients with genotypic Cys282Tyr homozygous hemochromatosis differ largerly in phenotypic presentation. The HFE mutation on itself does not explain the different manifestations of hemochromatosis. We hypothesized that the genetic haptoglobin ( H p ) polymorphism, because of its effect on iron metabolism, could be a modifying factor that influences the clinical presentation of hereditary hemochromatosis. Materials and methods. In 167 Cys282Tyr homozygous hemochromatotic patients, the frequencies of Hp types ( 1-1, 2-1 and 2-2) and alleles (Hp 1, Hp2) were compared with those in 918 healthy objects. Clinical and laboratory indices of iron overload, occurrence of liver fibrosis and age at diagnosis were incorporated in the analysis. Results. The H p l alllele frequency was low among hemochromatotic patients compared to the healthy reference population, due to an overrepresentation of the Hp 2-2 type (p<0.01). Male patients carrying Hp 2-2 had higher serum iron (p=0.003) and ferritin levels (p=0.03) than those with a Hp 1-1 or 2-1 type. The amount of iron removed with phlebotomy was also higher in Hp 2-2 patients ( p = 0 . 0 3 ) . No haptoglobin type-related differences were found when considering age at presentation, hepatic iron index, occurrence of clinical symptoms or liver fibrosis. Conclusion. The Hp 2-2 type is overrepresented among Cys282Tyr homozygous hemochromatotic patients. At diagnosis, iron overload was more pronounced in male patients carrying Hp 2-2. Our data suggest that Hp polymorphism is one of the modifier factors in the phenotypic expression of hereditary hemochromatosis.