- Email: [email protected]
PiMduarte
GASTROENTEROLOGY 1987;93242-4
Fatal Liver Disease Associated With arl-Antitrypsin Deficiency PiMl/PiMduarte JOSEPH J, CROWLEY, HARVEY L. SHARP, ESTHER FREIER, KAMAL G. ISHAK, and PATRICK SCHOW Department of Medicine, University of Washington, Seattle, Washington, and Medical Service, Boise Veterans Administration Medical Center, Boise, Idaho; Departments of Pediatrics and
Laboratory Medicine and Pathology, University of Minnesota, Minneapolis, Minnesota; Department of Hepatic Pathology, Armed Forces Institute of Pathology, Washington, DC.; Medical Service, Saint Luke’s Hospital, Boise, Idaho
A 53-yr-old man with a rare form of partial alantitrypsin deficiency, PiM,/PiMd,,ti,, died of endstage cirrhosis. Typical cytoplasmic al-antitrypsin globules were present in the hepatocyte cytoplasm. Initial protease inhibitor phenotyping on the patient was reported as normal PiMl in more than one laboratory. This case emphasizes the diagnostic importance of a,-antitrypsin and illustrates the point that protease inhibitor phenotyping without family genotyping may be misleading in heterozygous patients with liver disease. The association between al-antitrypsin (AAT) deficiency and liver disease was described in 1969 (1) with subsequent recognition that individuals with protease inhibitor (Pi)ZZ and PiMZ are at increased risk for developing cirrhosis (2,3). It is less clear whether other rarer deficiency variants of AAT are associated with the development of liver disease. We report a patient with cirrhosis associated with the genotype PiM1/PiMd,,ti2, and emphasize the difficulty in distinguishing this variant from the normal PiMIMl phenotype. Case Report A 53-yr-old white man presented with right upper quadrant abdominal pain and cholelithiasis requiring cholecystectomy. He had no history of jaundice, ethanol Received August 4, 1986. Accepted December 22, 1986. Address requests for reprints to: Dr. Joseph Crowley, Research Institute of Scripps Clinic, BCR-4, 10668 North Torrey Pines Road, La Jolla, California 92037. This work was supported in part by the Research Core Center for the Study of Advanced Liver Disease (lP30AM34931-01). The authors thank Dr. Barbara Cox for help with phenotyping in this family and with manuscript preparation. 0 1987 by the American Gastroenterological Association 0016-5085/87/$3.50
abuse, exposure to blood products, or hepatic toxins. Family history was negative for liver or pulmonary disease. Examination revealed signs consistent with hepatic cirrhosis. Liver tests were abnqrmal (albumin 2.8 g/dl, aspartate aminotransferase 87 IU/L, alkaline phosphatase 129 IU/L, and total bilirubin 2.1 mg/dl) and the hepatitis B surface antigen and hepatitis B core antigen were negative. Prothrombin clearance time was 14.2 s with a control of 10.5 s. Blood gases and pulmonary function tests were normal. An enlarged liver was noted at surgery and wedge biopsy sections showed a mixed micronodular and macronodular cirrhosis with proliferation of small cholangioles along the septa. Iron stains were negative. Most hepatocytes near the septa revealed eosinophilic globules that were strongly periodic acid-Schiff-positive and resisted diastase digestion (Figure 1). Immunoperoxidase staining showed that these globules contained AAT. The diagnosis of AAT deficiency-associated liver disease was made as a result of this biopsy. Over the ensuing z yr before death from liver failure, the patient’s clinical course was punctuated by recurrent gastrointestinal bleeding requiring a total of 28 U of blood. Serum iron level was 136 mg/dl with total saturation of iron binding capacity 1 yr before death. Postmortem examination confirmed the observations noted on the initial wedge biopsy sections except for the accumulation of iron in the cytoplasm of hepatocytes. Tissue assay of the liver revealed a total of 8.5 g of elemental iron. The spleen, heart, pancreas, and bone marrow also stained positive for iron. The proband and several family members exhibited intermediate deficiency levels of AAT (Figure 2). Protease inhibitor phenotyping was done using a modified version of Buffone’s method (4) for isoelectric focusing in agarose. A mixture of two ampholytes having pH ranges of 3.5-5.0 and 4.2-4.9 was used to establish the gradient across the gel. Those with the intermediate deficiency exhibited one Abbreviations used in this paper: AAT, a,-antitrypsin; protease inhibitor.
Pi,
August
1987
CIRRHOSIS
Figure
AND AAT DEFICIENCY
1. Numerous a,-antitrypsin ules (block) are present
243
globin the
cytoplasm of liver cells adjacent to a fibrous septum (periodic acid-Schiff after diastase digestion, X160). a,-Antitrypsin is identified immunohistochemitally in globules in inset (peroxidase-antiperoxidase technique, X 6301.
our isoelectric focus ing system, the bands were identical with the minor bal nds observed in the members with genotype was intermediate deficit mcy. The propositus
or two “extra” faint bands just cathodal to the major M bands when phenotyped in agarose. Protease inhibitor M 1naltonand PiZ were ruled out by immunofixation. When a serum specimen of PiMd,,,JPiZ was applied 40 times in
PiMI/PiMd,,,,t,.
DIED CA 8tOYACn
I
1 1
-7
64Y
L ---
I
I -1
1
/*-’ \ 5lY
:
\
\_.’ DIED
0 CB
DIED MI
I
/
1 /‘-’ \ 591
’ ._
/ \ 0-N
I
\
/’
DIED
TIC : 1.03
TIC: 0.59
Pi: kt+t,
” ’ +Ydu.rt.
Pi:Y,Y,
Pi : Y,Y,
t”,Yd”a,t,
HETEROZYGOTE
FYduart4
I
” : Y1”du.rt4
1
TIC: 0.50 pi I FYduart.
Figure
\ _# /
\
I TIC: 1.12
NORMAL PHENOTYPE
WTEROZYGOTE
5GY
2. Pedigree of family with Mduartc variant of u,-antitrypsin. Pi phenotype are listed (TIC normal range: 0.83-1.31
For each individual, mg trypsin inhibited
TIC: 0.67 Pi:M Y 1 dulrte
both the trypsin-inhibitory capacity (TIC) and the per milliliter serum). Ages are listed in years (y).
244
GASTROENTEROLOGY Vol. 93, No. 2
CROWLEY ET AL.
Discussion Periodic acid-Schiff-positive, diastase-resistant globules in periportal hepatocytes are intracellular AAT retained in the liver of individuals carrying the Z allele (2) or individuals with other phenotypes including PiMduarte and PiMmaiton (5). As the isoelectric point of these latter two rare phenotypes is in the same range as the most frequent phenotype, PiM, they are difficult to identify, particularly in heterozygotes. Intrahepatic globules have been reported, often without clear explanation for their accumulation, in individuals with the normal phenotypes PiM (6,7), PiMS (8), and PiS (9) and in hepatic tumor patients without unusual serum phenotypes (10). It is possible that some of these patients represented heterozygotes of either the Duarte or Malton variants. The predisposition for liver disease is difficult to understand in AAT deficiency. Homozygotes typically develop progressive emphysema rather than liver disease (11). The slightly increased incidence of cirrhosis in heterozygotes (3) suggests that another inciting agent or other genetic factors, or a combination thereof, contribute to more permanent liver injury. In our patient, accumulation of iron may have hastened the progression of cirrhosis. Most forms of chronic liver disease are characterized by elevated serum AAT levels whereas, in hemochromatosis, the lowest serum protein concentration is al-globulin, suggesting either an abnormal allele or suppression of AAT product by iron (12). The iron-binding glycoprotein transferrin may require proteolytic processing by cleavage of a 20 amino acid signal before hepatocyte secretion (13). Retention of a protease inhibitor in the endoplasmic reticulum of deficient AAT phenotypes could result in an inability to normally secrete iron-bound transferrin from the hepatocyte, leading to secondary hemochromatosis, particularly in those patients receiving increased iron. Hepatocyte iron overload has been demonstrated in patients with PiMZ (14) and in 3 male patients with liver disease and AAT deficiency (2). Increased iron stores have also been documented morphologically in AAT-deficient livers from infants (15). The prognosis of men with significant liver dysfunction and AAT deficiency is worse than that of women (16), and increased serum iron concentrations in these men may be a contributing factor toward progression of their liver disease. We suggest that the PiM,Md,,,t, phenotype may place an individual at increased risk for developing
hepatic disease. The difficulty in recognizing rare AAT phenotype variants makes us suspect that some prior reports of patients with typical hepatic inclusions and normal phenotypes were, instead, descriptions of individuals heterozygous for either the PiM dU&? or PiMmaiton alleles.
References 1.
Sharp HL, Bridges RA, Krivit W, Freier EF. Cirrhosis associated with alpha-l-antitrypsin deficiency: a previously unrecognized inherited disorder. J Lab Clin Med 1969;73:934-9. 2. Berg NO, Eriksson S. Liver disease in adults with alpha-lantitrypsin deficiency. N Engl J Med 1972;287:1264-7. 3. Hodges JR, Millward-Sadler GH, Barbatis C, Wright R. Heterozygous MZ alpha,-antitrypsin deficiency in adults with chronic active hepatitis and cryptogenic cirrhosis. N EnglJ Med 1981;304:557-60. 4. Buffone GJ, Stennis BJ, Schimbor CM. Isoelectric focusing in agarose: classification of genetic variants of alpha,-antitrypsin. Clin Chem 1983;29:328-31. 5. Cox DW, Billingsley GD, Smyth S. Rare types of a,antitrypsin associated with deficiency. In: Arnaud A, ed. Electrophoresis. Berlin: Walter de Gruyter, 1981:507-10. 6. Bradfield JWB, Blenkinsopp WK. Alpha-I-antitrypsin globules in the liver and PiM phenotype. J Clin Path01 1977;30: 464-6. 7. Pariente EA, Degott C, Martin JP, Feldmann G, Potet F, Benhamou JP. Hepatocytic PAS-positive diastase-resistant inclusions in the absence of alpha-l-antitrypsin deficiencyhigh prevalence in alcoholic cirrhosis. Am J Clin Path01 1981;76:299-302. 8. Fisher RL, Taylor L, Sherlock S. Alpha-l-antitrypsin deficiency in liver disease: the extent of the problem. Gastroenterology 1976;71:646-51. 9. Kelly JK, Taylor TV, Milford-Ward A. Alpha-l-antitrypsin PiS phenotype and liver cell inclusion bodies in alcoholic hepatitis. J Clin Path01 1979;32:706-9. 10. Palmer PE, Christopherson WM, Wolfe HJ. Alpha,-antitrypsin, protein marker in oral contraceptive-associated hepatic tumors. Am J Clin Path01 1977;68:736-9. 11. Larsson C. Natural history and life expectancy in severe alpha-l-antitrypsin deficiency, PiZ. Acta Med Stand 1978; 204:345-51. 12. Rigas D, Finch S. Electrophoretic studies of serum proteins in hemochromatosis. Am J Med Sci 1959;273:565-74. 13. Shreiver G, Drybursk H, Millership A, et al. The synthesis and secretion of rat transferrin. J Biol Chem 1979;244:12013-9. 14. Janus ED, Carrel1 RW. Alpha,-antitrypsin deficiency and adult liver disease. In: Peeters H, ed. Proteins and related subjects. Volume 23. Oxford: Pergamon, 1975:383-5. 15. Sharp HL. Relationship between alpha-l-antitrypsin deficiency and liver disease. In: Berenberg SH, ed. Liver disease in infancy and childhood. The Hague: Martinus Nijhoff Medical Division, 1976:52-71. 16. Carlson J, Eriksson S. Chronic “cryptogenic” liver disease and malignant hepatoma in intermediate alpha,-antitrypsin deficiency identified by a PiZ-specific monoclonal antibody. Stand J Gastroenterol 1985;20:834-42.
Fatal Liver Disease Associated With arl-Antitrypsin Deficiency PiMl/PiMduarte JOSEPH J, CROWLEY, HARVEY L. SHARP, ESTHER FREIER, KAMAL G. ISHAK, and PATRICK SCHOW Department of Medicine, University of Washington, Seattle, Washington, and Medical Service, Boise Veterans Administration Medical Center, Boise, Idaho; Departments of Pediatrics and
Laboratory Medicine and Pathology, University of Minnesota, Minneapolis, Minnesota; Department of Hepatic Pathology, Armed Forces Institute of Pathology, Washington, DC.; Medical Service, Saint Luke’s Hospital, Boise, Idaho
A 53-yr-old man with a rare form of partial alantitrypsin deficiency, PiM,/PiMd,,ti,, died of endstage cirrhosis. Typical cytoplasmic al-antitrypsin globules were present in the hepatocyte cytoplasm. Initial protease inhibitor phenotyping on the patient was reported as normal PiMl in more than one laboratory. This case emphasizes the diagnostic importance of a,-antitrypsin and illustrates the point that protease inhibitor phenotyping without family genotyping may be misleading in heterozygous patients with liver disease. The association between al-antitrypsin (AAT) deficiency and liver disease was described in 1969 (1) with subsequent recognition that individuals with protease inhibitor (Pi)ZZ and PiMZ are at increased risk for developing cirrhosis (2,3). It is less clear whether other rarer deficiency variants of AAT are associated with the development of liver disease. We report a patient with cirrhosis associated with the genotype PiM1/PiMd,,ti2, and emphasize the difficulty in distinguishing this variant from the normal PiMIMl phenotype. Case Report A 53-yr-old white man presented with right upper quadrant abdominal pain and cholelithiasis requiring cholecystectomy. He had no history of jaundice, ethanol Received August 4, 1986. Accepted December 22, 1986. Address requests for reprints to: Dr. Joseph Crowley, Research Institute of Scripps Clinic, BCR-4, 10668 North Torrey Pines Road, La Jolla, California 92037. This work was supported in part by the Research Core Center for the Study of Advanced Liver Disease (lP30AM34931-01). The authors thank Dr. Barbara Cox for help with phenotyping in this family and with manuscript preparation. 0 1987 by the American Gastroenterological Association 0016-5085/87/$3.50
abuse, exposure to blood products, or hepatic toxins. Family history was negative for liver or pulmonary disease. Examination revealed signs consistent with hepatic cirrhosis. Liver tests were abnqrmal (albumin 2.8 g/dl, aspartate aminotransferase 87 IU/L, alkaline phosphatase 129 IU/L, and total bilirubin 2.1 mg/dl) and the hepatitis B surface antigen and hepatitis B core antigen were negative. Prothrombin clearance time was 14.2 s with a control of 10.5 s. Blood gases and pulmonary function tests were normal. An enlarged liver was noted at surgery and wedge biopsy sections showed a mixed micronodular and macronodular cirrhosis with proliferation of small cholangioles along the septa. Iron stains were negative. Most hepatocytes near the septa revealed eosinophilic globules that were strongly periodic acid-Schiff-positive and resisted diastase digestion (Figure 1). Immunoperoxidase staining showed that these globules contained AAT. The diagnosis of AAT deficiency-associated liver disease was made as a result of this biopsy. Over the ensuing z yr before death from liver failure, the patient’s clinical course was punctuated by recurrent gastrointestinal bleeding requiring a total of 28 U of blood. Serum iron level was 136 mg/dl with total saturation of iron binding capacity 1 yr before death. Postmortem examination confirmed the observations noted on the initial wedge biopsy sections except for the accumulation of iron in the cytoplasm of hepatocytes. Tissue assay of the liver revealed a total of 8.5 g of elemental iron. The spleen, heart, pancreas, and bone marrow also stained positive for iron. The proband and several family members exhibited intermediate deficiency levels of AAT (Figure 2). Protease inhibitor phenotyping was done using a modified version of Buffone’s method (4) for isoelectric focusing in agarose. A mixture of two ampholytes having pH ranges of 3.5-5.0 and 4.2-4.9 was used to establish the gradient across the gel. Those with the intermediate deficiency exhibited one Abbreviations used in this paper: AAT, a,-antitrypsin; protease inhibitor.
Pi,
August
1987
CIRRHOSIS
Figure
AND AAT DEFICIENCY
1. Numerous a,-antitrypsin ules (block) are present
243
globin the
cytoplasm of liver cells adjacent to a fibrous septum (periodic acid-Schiff after diastase digestion, X160). a,-Antitrypsin is identified immunohistochemitally in globules in inset (peroxidase-antiperoxidase technique, X 6301.
our isoelectric focus ing system, the bands were identical with the minor bal nds observed in the members with genotype was intermediate deficit mcy. The propositus
or two “extra” faint bands just cathodal to the major M bands when phenotyped in agarose. Protease inhibitor M 1naltonand PiZ were ruled out by immunofixation. When a serum specimen of PiMd,,,JPiZ was applied 40 times in
PiMI/PiMd,,,,t,.
DIED CA 8tOYACn
I
1 1
-7
64Y
L ---
I
I -1
1
/*-’ \ 5lY
:
\
\_.’ DIED
0 CB
DIED MI
I
/
1 /‘-’ \ 591
’ ._
/ \ 0-N
I
\
/’
DIED
TIC : 1.03
TIC: 0.59
Pi: kt+t,
” ’ +Ydu.rt.
Pi:Y,Y,
Pi : Y,Y,
t”,Yd”a,t,
HETEROZYGOTE
FYduart4
I
” : Y1”du.rt4
1
TIC: 0.50 pi I FYduart.
Figure
\ _# /
\
I TIC: 1.12
NORMAL PHENOTYPE
WTEROZYGOTE
5GY
2. Pedigree of family with Mduartc variant of u,-antitrypsin. Pi phenotype are listed (TIC normal range: 0.83-1.31
For each individual, mg trypsin inhibited
TIC: 0.67 Pi:M Y 1 dulrte
both the trypsin-inhibitory capacity (TIC) and the per milliliter serum). Ages are listed in years (y).
244
GASTROENTEROLOGY Vol. 93, No. 2
CROWLEY ET AL.
Discussion Periodic acid-Schiff-positive, diastase-resistant globules in periportal hepatocytes are intracellular AAT retained in the liver of individuals carrying the Z allele (2) or individuals with other phenotypes including PiMduarte and PiMmaiton (5). As the isoelectric point of these latter two rare phenotypes is in the same range as the most frequent phenotype, PiM, they are difficult to identify, particularly in heterozygotes. Intrahepatic globules have been reported, often without clear explanation for their accumulation, in individuals with the normal phenotypes PiM (6,7), PiMS (8), and PiS (9) and in hepatic tumor patients without unusual serum phenotypes (10). It is possible that some of these patients represented heterozygotes of either the Duarte or Malton variants. The predisposition for liver disease is difficult to understand in AAT deficiency. Homozygotes typically develop progressive emphysema rather than liver disease (11). The slightly increased incidence of cirrhosis in heterozygotes (3) suggests that another inciting agent or other genetic factors, or a combination thereof, contribute to more permanent liver injury. In our patient, accumulation of iron may have hastened the progression of cirrhosis. Most forms of chronic liver disease are characterized by elevated serum AAT levels whereas, in hemochromatosis, the lowest serum protein concentration is al-globulin, suggesting either an abnormal allele or suppression of AAT product by iron (12). The iron-binding glycoprotein transferrin may require proteolytic processing by cleavage of a 20 amino acid signal before hepatocyte secretion (13). Retention of a protease inhibitor in the endoplasmic reticulum of deficient AAT phenotypes could result in an inability to normally secrete iron-bound transferrin from the hepatocyte, leading to secondary hemochromatosis, particularly in those patients receiving increased iron. Hepatocyte iron overload has been demonstrated in patients with PiMZ (14) and in 3 male patients with liver disease and AAT deficiency (2). Increased iron stores have also been documented morphologically in AAT-deficient livers from infants (15). The prognosis of men with significant liver dysfunction and AAT deficiency is worse than that of women (16), and increased serum iron concentrations in these men may be a contributing factor toward progression of their liver disease. We suggest that the PiM,Md,,,t, phenotype may place an individual at increased risk for developing
hepatic disease. The difficulty in recognizing rare AAT phenotype variants makes us suspect that some prior reports of patients with typical hepatic inclusions and normal phenotypes were, instead, descriptions of individuals heterozygous for either the PiM dU&? or PiMmaiton alleles.
References 1.
Sharp HL, Bridges RA, Krivit W, Freier EF. Cirrhosis associated with alpha-l-antitrypsin deficiency: a previously unrecognized inherited disorder. J Lab Clin Med 1969;73:934-9. 2. Berg NO, Eriksson S. Liver disease in adults with alpha-lantitrypsin deficiency. N Engl J Med 1972;287:1264-7. 3. Hodges JR, Millward-Sadler GH, Barbatis C, Wright R. Heterozygous MZ alpha,-antitrypsin deficiency in adults with chronic active hepatitis and cryptogenic cirrhosis. N EnglJ Med 1981;304:557-60. 4. Buffone GJ, Stennis BJ, Schimbor CM. Isoelectric focusing in agarose: classification of genetic variants of alpha,-antitrypsin. Clin Chem 1983;29:328-31. 5. Cox DW, Billingsley GD, Smyth S. Rare types of a,antitrypsin associated with deficiency. In: Arnaud A, ed. Electrophoresis. Berlin: Walter de Gruyter, 1981:507-10. 6. Bradfield JWB, Blenkinsopp WK. Alpha-I-antitrypsin globules in the liver and PiM phenotype. J Clin Path01 1977;30: 464-6. 7. Pariente EA, Degott C, Martin JP, Feldmann G, Potet F, Benhamou JP. Hepatocytic PAS-positive diastase-resistant inclusions in the absence of alpha-l-antitrypsin deficiencyhigh prevalence in alcoholic cirrhosis. Am J Clin Path01 1981;76:299-302. 8. Fisher RL, Taylor L, Sherlock S. Alpha-l-antitrypsin deficiency in liver disease: the extent of the problem. Gastroenterology 1976;71:646-51. 9. Kelly JK, Taylor TV, Milford-Ward A. Alpha-l-antitrypsin PiS phenotype and liver cell inclusion bodies in alcoholic hepatitis. J Clin Path01 1979;32:706-9. 10. Palmer PE, Christopherson WM, Wolfe HJ. Alpha,-antitrypsin, protein marker in oral contraceptive-associated hepatic tumors. Am J Clin Path01 1977;68:736-9. 11. Larsson C. Natural history and life expectancy in severe alpha-l-antitrypsin deficiency, PiZ. Acta Med Stand 1978; 204:345-51. 12. Rigas D, Finch S. Electrophoretic studies of serum proteins in hemochromatosis. Am J Med Sci 1959;273:565-74. 13. Shreiver G, Drybursk H, Millership A, et al. The synthesis and secretion of rat transferrin. J Biol Chem 1979;244:12013-9. 14. Janus ED, Carrel1 RW. Alpha,-antitrypsin deficiency and adult liver disease. In: Peeters H, ed. Proteins and related subjects. Volume 23. Oxford: Pergamon, 1975:383-5. 15. Sharp HL. Relationship between alpha-l-antitrypsin deficiency and liver disease. In: Berenberg SH, ed. Liver disease in infancy and childhood. The Hague: Martinus Nijhoff Medical Division, 1976:52-71. 16. Carlson J, Eriksson S. Chronic “cryptogenic” liver disease and malignant hepatoma in intermediate alpha,-antitrypsin deficiency identified by a PiZ-specific monoclonal antibody. Stand J Gastroenterol 1985;20:834-42.