Plasma Selenium in Patients With Cirrhosis

Transcription

1 Plasma Selenium in Patients With Cirrhosis RAYMOND F. B URK, 1 DAYNA S. EARLY, 1 KRISTINA E. HILL, 1 IVAN S. PALMER, 2 AND MARTHA E. BOEGLIN 1 Plasma selenium concentration is decreased in patients with cirrhosis and, based on this finding, it has been suggested that patients with cirrhosis are selenium deficient. We measured plasma selenium concentration and the two plasma selenoproteins, glutathione peroxidase (GSHPx-3) and selenoprotein P, in the plasma of patients with cirrhosis of Child classes A, B, and C and in control subjects. Plasma selenium declined in proportion to the severity of the cirrhotic condition, as indicated by the Child class. Selenoprotein P, which originates largely in the liver, declined in a similar manner. Plasma glutathione peroxidase activity increased, and GSHPx-3 originates in the kidney. Selenium in the non-selenoprotein pool, shown by others to be largely selenomethionine in albumin, declined. Thus, although plasma selenium is decreased in patients with cirrhosis, the increase in plasma glutathione peroxidase activity, which occurs in them, suggests that patients with cirrhosis do not have selenium deficiency. (HEPATOL- OGY 1998;27: ) Selenium exerts biological effects as an essential constituent of selenoproteins. 1 Pathological conditions, such as selenium deficiency and organ damage, which are severe enough to alter selenium metabolism, can be expected to affect the concentrations of selenoproteins and to have biochemical and pathological consequences. Several groups have reported that patients with cirrhosis have plasma selenium concentrations lower than those of healthy controls. 2-6 The pathogenesis of those depressed selenium levels is unknown; therefore it is not known whether a remediable state of selenium deficiency exists in cirrhotics. Virtually all the selenium in plasma is present in the form of seleno-amino acids in the primary structure of proteins. One of these amino acids is selenocysteine, the physiologically active form of the element, which is synthesized in animal cells and is present in stoichiometric amounts in selenoproteins. Glutathione peroxidase (isoform GSHPx-3) and selenoprotein P are the only selenoproteins that have been identified in plasma. 7,8 The other amino acid form of selenium in plasma is selenomethionine, 9 an amino acid that is synthesized by plants and is incorporated randomly into animal proteins as a constituent of the methionine pool. From the 1 Division of Gastroenterology, Department of Medicine and Clinical Nutrition Research Unit, Vanderbilt University School of Medicine, Nashville, TN; and 2 Department of Biochemistry, South Dakota State University, Brookings, SD. Received July 31, 1997; accepted October 31, Supported by grants from the National Institutes of Health ES02497 and ES Address reprint requests to: Raymond F. Burk, M.D., C2104, Medical Center North, Vanderbilt Medical Center, Nashville, TN Fax: (615) Copyright 1998 by the American Association for the Study of Liver Diseases /98/ $3.00/0 794 Proteins that contain selenomethionine do not contain the element in stoichiometric amounts and are often referred to as selenium-containing proteins to distinguish them from true selenoproteins. Small molecule forms of uncharacterized selenium are involved in homeostasis of the element. Those forms account for approximately 3% of plasma selenium under steady state conditions. 10 The purpose of this study is to determine the distribution of plasma selenium among the three protein compartments in patients with cirrhosis and, thereby, to gain greater insight into the cause of their depressed plasma selenium levels. The two plasma selenoproteins were measured directly, as was plasma selenium. From these results the three protein compartments of plasma selenium were estimated. Whereas cirrhosis was associated with a decrease in total plasma selenium, as reported previously, only two of the protein compartments of selenium decreased. The other one, glutathione peroxidase, increased. PATIENTS AND METHODS Patients Patients under the care of gastroenterologists at Vanderbilt University Hospital and the Nashville Veterans Affairs Medical Center were recruited for study from April 19, 1994, to June 30, All patients were at least 18 years old and had been diagnosed as having cirrhosis (diagnoses listed in Table 1) by their physicians. Several patients awaiting liver transplant were studied. Five patients who were studied before transplant and who received a liver transplant were re-studied 6 or more weeks after the transplant when their clinical condition was stable. No patients with a hematocrit below 25% or who were taking selenium supplements were studied. Normal subjects were recruited from members of the Division of Gastroenterology. All were healthy and none was taking selenium supplements. The study was approved by the Vanderbilt University Institutional Review Board. Methods Protocol. A 20-mL blood sample was taken from an antecubital vein after an overnight fast. The blood was treated with ethylenediaminetetraacetic acid (1 mg/ml) to prevent coagulation. The hematocrit was determined and plasma was separated by centrifugation. Selenium concentration in plasma was determined using a fluorometric method. 11 Glutathione peroxidase activity with hydrogen peroxide as the substrate was determined using a method optimized for plasma. 12 Selenoprotein P concentration was determined by radioimmunoassay using a stored plasma as a reference. 13 Laboratory values pertaining to liver disease were obtained from the medical records of the patients. The values were usually determined the day the experimental blood sample was taken. The Child class was determined from clinical and laboratory data in the medical record according to the Child-Pugh classification using assessments of ascites, encephalopathy, bilirubin, albumin, and prothrombin time

2 HEPATOLOGY Vol. 27, No. 3, 1998 BURK ET AL. 795 TABLE 1. Cause of Cirrhosis Child Classes A B C Alcohol Cryptogenic Autoimmune 3 1 Hepatitis C Hepatitis B 1 Hepatitis B and C 1 Primary biliary cirrhosis 2 Primary sclerosing cholangitis 1 Hemochromatosis 1 Non-alcoholic steatohepatitis 1 Total Estimation of Plasma Selenium Compartments. Reference plasma pools were characterized with respect to distribution of selenium among protein compartments. In one reference plasma, glutathione peroxidase was reported to contain 12% of plasma selenium 16 and, in another, selenoprotein P was reported to contain 44% of plasma selenium. 13 Because these are the only two selenoproteins known to be in plasma, the remaining 44% of the plasma selenium was considered to be present in the non-specific fraction, most of which has been shown to be selenomethionine. 9 The small amount of selenium in the small molecule fraction was ignored in these estimations. The control plasma samples in this study contained µg of selenium/dl. We assumed that they contained the same concentrations of selenium in the two selenoproteins as was estimated in the previous reference plasmas: 1.7 µg of selenium/dl plasma in glutathione peroxidase and 6.4 µg of selenium/dl plasma in selenoprotein P. 13,16 Selenium in the other protein compartment was estimated to be 5.4 µg/dl of plasma by subtracting the selenium content of the two selenoproteins from the plasma selenium concentration. Selenium compartments in plasma from patients with cirrhosis were estimated from the selenoprotein determinations and the plasma selenium concentration. The selenium in each selenoprotein was estimated by comparing the selenoprotein value in each patient with the corresponding value in the control subjects. For example, if the selenoprotein P concentration in plasma from a patient with cirrhosis was 50% of the selenoprotein P concentration in the controls, the plasma selenium accounted for by selenoprotein P in that patient was estimated to be 50% of the selenium in selenoprotein P in control subjects or 3.2 µg of selenium/dl plasma. The calculation used to arrive at this value would be as follows: 6.4 µg of selenium/dl plasma µg of selenium/dl plasma. The other compartment was estimated by subtracting the selenium content of the two selenoproteins from the plasma selenium content. Statistical comparisons were made using the program Statview [Abacus Concepts, Berkeley, CA] on a Macintosh IIci. Statistical differences were determined using Fisher s protected least significance difference test and the paired Student s t test. RESULTS Patients Studied. Table 2 shows some characteristics of patients with cirrhosis and of controls at Vanderbilt Hospital and the Veterans Affairs Medical Center. There was a preponderance of female patients in the Child A group because they came largely from the Vanderbilt population. The Child B and C groups were largely from the VA Medical Center and were mostly males. The reason for this distribution of patients is the patterns of practice at the two hospitals. Both an active liver transplant program and the referral nature of the patient population at Vanderbilt limit care for the number of patients with severe and decompensated liver disease. The VA Medical Center provides primary care to eligible patients with severe liver disease who are not transplant candidates. Table 1 shows the causes of the cirrhosis. The more severely ill groups contained a greater proportion of patients with alcoholic liver disease. Plasma Selenium Compartments. Plasma selenium concentration was depressed in patients with cirrhosis (Fig. 1). Moreover, as liver disease worsened, according to the Child classification, plasma selenium concentration decreased progressively. The two selenoproteins in plasma were affected differently by cirrhosis (Fig. 2). Selenoprotein P concentration decreased as liver disease worsened; glutathione peroxidase activity increased as liver disease worsened. In Child group C patients, selenoprotein P concentration was 49% of control and glutathione peroxidase activity was 150% of control. In addition to the specific presence of selenium as selenocysteine in the two plasma selenoproteins, a third protein compartment of the element is present in plasma, corresponding to the selenomethionine incorporated randomly at methionine positions in proteins. It is possible to estimate the amount of plasma selenium in this third pool by calculating the selenium content of each of the two selenoproteins and subtracting those values from total plasma selenium concentration. Figure 3 shows our estimates of each of these three compartments in control subjects and in patients with cirrhosis. The selenium in selenoprotein P and in the other compartment both declined in liver disease. The selenium in the glutathione peroxidase compartment rose and its fraction of the total plasma selenium increased from 13% to 31%. Liver transplant caused a rise in selenoprotein P and plasma selenium concentration without affecting glutathione peroxidase activity (Table 3). This indicates that the decline in selenoprotein P in liver disease is reversed by improvement in hepatic function. DISCUSSION This study shows that the changes in plasma selenium associated with cirrhosis are complex. While confirming the observation that patients with cirrhosis have a depressed plasma selenium concentration, the study reports that selenoprotein P, largely synthesized by the liver, 17 is depressed whereas glutathione peroxidase, largely synthesized by the kidneys, 18 is increased. The third selenium pool, generally TABLE 2. Characteristics of Study Subjects* Patients With Cirrhosis (Child Groups) Healthy Controls A B C n Age Male/female 6/7 3/12 11/5 10/0 Hematocrit Albumin (g/dl) Bilirubin (mg/dl) Prothrombin time (sec) *Values are means SD. Normal values 10 to 13 seconds.

3 796 BURK ET AL. HEPATOLOGY March 1998 FIG. 1. Plasma selenium concentration in patients with cirrhosis and in controls. Cirrhotics are grouped by Child class of severity (Table 2). Values shown are means with the error bar indicating one SD. *Significant differences (P.05) between patient groups and controls by using Fisher s protected least-significance difference test. considered to represent selenomethionine in proteins, is depressed. Thus, the net drop in plasma selenium concentration was caused by decreases in selenoprotein P and the selenomethionine pool that, taken together, were greater than the increase in the selenoprotein glutathione peroxidase. To interpret these findings in patients with cirrhosis, it is necessary to consider how selenium is metabolized. Much, probably most, of the selenium ingested by human beings is in the form of selenomethionine. Plants take up soil selenium and, because of its chemical similarity to sulfur, incorporate it into selenomethionine. Animals cannot distinguish selenomethionine from methionine and incorporate it randomly into proteins at methionine positions The amount of selenium (selenomethionine) in the methionine pool is not homeostatically regulated and depends on the entry of selenomethionine into the pool, which is selenomethionine ingestion, and on the turnover of the pool. Turnover of the pool is dependent on methionine intake and on other factors that are not influenced by selenium. 20 It should be noted that the amount of selenium present in the FIG. 3. Estimate of selenium compartments in plasma protein of patients with cirrhosis and their comparison with control values. Percentages of plasma selenium in each compartment were calculated as described in Materials and Methods and are shown beside their respective compartments. Each small open rectangle within the selenium compartments of the patients with cirrhosis 1 SD. No statistical comparisons could be made between control subjects and patients with cirrhosis because the control values are assigned as described in Materials and Methods. Selenoprotein P and glutathione peroxidase were each significantly different (P.05) between Child groups A and C by Fisher s protected least-significance difference test. No other differences within the patient groups were significant. Se-P, selenoprotein P; GSH-Px, glutathione peroxidase; other, remainder of the plasma selenium that is largely selenomethionine. selenomethionine compartment in control subjects [5.4 µg/dl of plasma] can be accounted for by the replacement of one methionine residue in 10,000 by selenomethionine. Thus, plasma selenium concentration in the form of selenomethionine depends on a number of factors unrelated to selenium and is not a true measure of physiological selenium status. Therefore, the shrinkage of this pool of selenium in patients with cirrhosis cannot be indicative of the presence of selenium deficiency. When selenomethionine is catabolized, selenium is released as selenide 21 and enters the metabolic pathway specific for the element (Fig. 4). Ingested selenocysteine, selenite, and selenate are rapidly converted to selenide and enter the specific selenium metabolic pathway without delay. 22 These latter forms of selenium constitute a variable, but probably minor, fraction of selenium intake in free-living human beings. Unlike selenium in the methionine pool, selenium in the selenium metabolic pathway is under homeostatic control. It is metabolized through the intermediates selenide and selenophosphate to selenocysteine, which is incorporated into the active sites of selenoproteins. 1 Excess selenium entering the pathway is methylated and excreted. 23 Thus, this pathway TABLE 3. Liver Transplant and Plasma Selenium* Before Transplant After Transplant FIG. 2. Plasma selenoprotein concentrations in patients with cirrhosis and in controls. Cirrhotics are grouped by Child class of severity (Table 2). ( ) Selenoprotein P concentrations; and ( ) glutathione peroxidase activities. *Significant differences (P.05) between patient groups and controls by Fisher s protected least-significance difference test. Error bar 1 SD. Plasma selenium (µg/dl) Selenoprotein P (mu/ml) Glutathione peroxidase (mu/ml) *Values are means SD (n 5). Values are greater than pre-transplant values by paired Student t test (P.05).

4 HEPATOLOGY Vol. 27, No. 3, 1998 BURK ET AL. 797 FIG. 4. Scheme depicting plasma protein compartments of selenium and their origins. The liver is the major source of selenomethionine-containing plasma protein because it synthesizes albumin, although proteins derived from other tissues, such as immunoglobulins (not shown in scheme), may also contain selenomethionine. Selenomethionine is catabolized in the liver [l] by the transsulfuration pathway to selenide that enters the selenium pool and serves as a source of selenium for selenoproteins. 22 Selenoprotein P is largely secreted into plasma by the liver. 17 The selenium pool in the kidney serves as a source of selenium for plasma glutathione peroxidase. 18 It is not known how or to what extent selenium is transported between organs as indicated by [2]. can conserve selenium by curtailing excretion when the element is in short supply. This decreased excretion conserves selenoproteins even when selenium intake is low and when the selenomethionine pool of selenium is very small. However, when selenium intake is truly deficient, selenoprotein concentrations decline and serve as indices of selenium nutritional status. Healthy people living in low-selenium areas of China have biochemical selenium deficiency. Their plasma selenium pools have been studied before and after selenium supplementation Both selenoprotein P concentration and glutathione peroxidase activity were 10% to 20% of the corresponding values in control subjects before supplementation. Administration of selenium increased both selenoproteins toward the control subject levels over a period of two weeks. Do patients with cirrhosis have selenium deficiency? This is an important question because selenium deficiency might contribute to the morbidity of cirrhosis and can easily be eliminated by administration of selenium. On the other hand, selenium can have toxic effects when present in excess, 24 thus, administration of the element to patients with adequate stores might be associated with some risk. Patients with cirrhosis have a pattern of plasma selenoproteins different from that of healthy subjects with selenium deficiency. Selenoprotein P concentration is depressed in both groups, but glutathione peroxidase activity is decreased in selenium deficiency while being increased in cirrhosis. There is no precedent of an increase in selenoprotein during selenium deficiency. Therefore, the increase in plasma glutathione peroxidase activity in cirrhotics is evidence against these patients being selenium deficient. The possibility exists, however, that the increase in glutathione peroxidase activity could be caused by release of intracellular isoforms of the enzyme from the diseased liver or other tissues. Identification of the isoform(s) causing the increase will be necessary to distinguish between these two alternatives. The most direct explanation of changes found in patients with cirrhosis is that cirrhosis impairs the hepatic synthesis of selenoprotein P, resulting in the low plasma selenoprotein P concentration. Plasma glutathione peroxidase (GSHPx-3) originates in the kidney, and cirrhosis would not be expected to impair its synthesis. The increase in plasma glutathione peroxidase activity in cirrhosis is a consistent finding and correlates with severity of disease (Fig. 2). If the increased activity is caused by the plasma isoform GSHPx-3, it can be speculated that this increase is a secondary response to cirrhosis. However, the function of plasma glutathione peroxidase has not been established with certainty, so the reason for the increase is not known. 25 Another, although less likely, explanation of the selenoprotein changes in cirrhosis is possible. Selenomethionine must be catabolized via the transsulfuration pathway for its selenium to be made available for selenoprotein synthesis (Fig. 4). It is known that cirrhosis impairs this pathway 26 and, thus, might render selenomethionine a poor source of selenium; this would lead to a functional selenium deficiency. The transsulfuration pathway can be bypassed by administering inorganic selenium. Thus, to evaluate this possibility a study of the effect of inorganic selenium supplementation on plasma selenoprotein P concentration in patients with cirrhosis would be needed. In summary, this study confirms that plasma selenium is depressed in patients with cirrhosis. However, the changes noted in plasma selenoproteins of cirrhotics are not the same as those found in selenium-deficient subjects without liver disease. Thus, the fall in plasma selenium associated with cirrhosis appears more likely to be caused by hepatic dysfunction than by selenium deficiency. REFERENCES 1. Stadtman TC. Biosynthesis and function of selenocysteine-containing enzymes. J Biol Chem 1991;266: Chin SF, Shepherd RW, Thomas BJ, et al. The nature of malnutrition in children with end-state liver disease awaiting orthotopic liver transplantation. Am J Clin Nutr 1992;56: Aaseth J, Alexander J, Thomassen Y, Blomhoff JP, Skrede S. Serum selenium levels in liver diseases. Clin Biochem 1982;15: Dworkin B, Rosenthal WS, Jankowski RH, Gordon GG, Haldea D. Low blood selenium levels in alcoholics with and without advanced liver disease. Dig Dis Sci 1985;30: Thuluvath PJ, Triger DR. Selenium in chronic liver disease. J Hepatol 1992;14: Valimaki M, Alfthan G, Pikkarainen J, Ylikahri R, Salaspuro M. Blood and liver selenium concentrations in patients with liver diseases. Clin Chim Acta 1987;166: Takahashi K, Cohen HJ, Selenium-dependent glutathione peroxidase protein and activity: immunological investigations on cellular and plasma enzymes. Blood 1986;68: Åkesson B, Bellew T, Burk RF. Purification of selenoprotein P from human plasma. Biochim Biophys Acta 1994;1204: Deagen JT, Beilstein MA, Whanger PD. Chemical forms of selenium in selenium containing proteins from human plasma. J Inorg Biochem 1991;41: Burk RF. Binding of 75 Se to human plasma proteins. Biochim Biophys Acta 1974;372: Koh TS, Benson TH. Critical re-appraisal of fluorometric method for determination of selenium in biological materials. J Assoc Off Anal Chem 1983;66: Xia Y, Hill KE, Burk RF. Biochemical studies of a selenium-deficient population in China: measurement of selenium, glutathione peroxidase and oxidant defenses in blood. J Nutr 1989;119: Hill KE, Xia Y, Åkesson B, Boeglin ME, Burk RF. Selenoprotein P concentration in plasma is an index of selenium status in seleniumdeficient and selenium-supplemented Chinese subjects. J Nutr 1996;126: Child CG, Turcotte JG. Surgery and portal hypertension. In: Child CG, ed. The liver and portal hypertension. Philadelphia, PA: Saunders, 1964: Pugh RNH, Murray-Lyon IM, Dawson JL, Pietroni MC, Williams R.

5 798 BURK ET AL. HEPATOLOGY March 1998 Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973;60: Avissar N, Whitin JC, Allen PZ, Palmer IS, Cohen HJ. Antihuman plasma glutathione peroxidase antibodies: immunologic investigations to determine plasma glutathione peroxidase. Blood 1989;73: Kato T, Read R, Rozga J, Burk RF. Evidence for intestinal release of absorbed selenium in a form with high hepatic extraction. Am J Physiol 1992;262:G854-G Avissar N, Ornt DB, Yagil Y, et al. Human kidney proximal tubules are the main source of plasma glutathione peroxidase. Am J Physiol 1994;266:C367-C Whanger PD, Xia Y, Thomson CD. Protein technics for selenium speciation in human body fluids. J Trace Elem Electrolytes Health Dis 1994;8: Waschulewski IH, Sunde RA, Effect of dietary methionine on tissue selenium and glutathione peroxidase activity in rats given selenomethionine. Br J Nutr 1988;60: Esaki N, Nakamura T, Tanaka H, Suzuki T, Morino Y, Soda K. Enzymatic synthesis of selenocysteine in rat liver. Biochemistry 1981;20: Levander OA, Burk RF, Selenium. In: Ziegler EE, Filer LJ, eds. Present Knowledge in Nutrition. Washington: ILSI Press, 1996: Bopp BA, Sonders RC, Kesterson JW. Metabolic fate of selected selenium compounds in laboratory animals and man. Drug Metab Rev 1982;13: Yang GQ, Wang S, Zhou R, Sun S. Endemic selenium intoxication of humans in China. Am J Clin Nutr 1983;37: Burk RF. Selenium-dependent glutathione peroxidases. In: Guengerich FP, ed. Vol. 3. Biotransformation. Oxford: Pergamon, 1997: Horowitz JH, Rypins EB, Henderson JM, et al. Evidence for impairment of transsulfuration pathway in cirrhosis. Gastroenterology 1981;81: