Nephrology Dialysis Transplantation

Transcription

1 Nephrol Dial Transplant ( 1997) 12: Original Article Nephrology Dialysis Transplantation Free amino-acid levels simultaneously collected in plasma, muscle, and erythrocytes of uraemic patients J. C. Divino Filho1, P. Bárány1, P. Stehle3, P. Fürst3 and J. Bergström1,2 1Division of Renal Medicine and 2Baxter Novum, Department of Clinical Science, Karolinska Institutet, Huddinge University Hospital, Stockholm, Sweden; and 3Department of Nutritional Biochemistry, Hohenheim University, Stuttgart, Federal Republic of Germany Abstract AA in RBC should be considered when undertaking Background. Disturbances in amino acid (AA) metabolism in uraemia have mainly been reported to occur metabolic and clinical studies of AA disturbances. in plasma and muscle. The erythrocytes (RBC ) constiplasma; uraemia Key words: amino acid; dialysis; erythrocytes; muscle; tute a large proportion of the free AA in blood and may play an important role in the interorgan transport of AA. This report presents the first data on AA levels obtained simultaneously from three different compartments in uraemic patients. Methods. Muscle biopsy and blood samples were Introduction obtained from 38 haemodialysis (HD), 22 continuous peritoneal dialysis (CPD) and 10 end-stage renal failure The kidney plays a major role in the regulation of patients for determination of free amino acids by many body pools of amino acids (AA) through syn- reversed-phase HPLC. The results are compared to thesis, degradation and/or urinary excretion. In chronic data obtained from 27 healthy subjects under the same renal failure, a specific pattern with high concentrations conditions. of several non-essential amino acids (NEAA) and low Results. For a number of non-essential AA (alanine, concentrations of essential amino acids (EAA), includglycine, asparagine, arginine) and for lysine, elevated ing branched-chain amino acids (BCAA), has been concentrations were present simultaneously in RBC reported both in plasma and in muscle [1 6]. The and in muscle but not in plasma. On the other hand, distribution of AA to the different pools may be altered low concentrations of some essential AA ( leucine, by impairment of either the excretion (e.g. 3-methylvaline, phenylalanine, tyrosine) were observed in RBC histidine), the renal metabolism (e.g. citrulline and gly- and in plasma, while the concentrations in muscle were cine), or the synthesis (e.g. serine and tyrosine). Other normal. Most of the non-essential AA ( NEAA), abnormalities in AA metabolism may be related to especially taurine and glutamine, had much higher several features of chronic uraemia, such as disturb- muscle/plasma gradients than RBC/plasma gradients, ances in protein and energy metabolism, hormonal although an accumulation in RBC of glycine, serine, derangement, and alterations in the intermediary meta- arginine, asparagine, ornithine, glutamate and taurine bolism. When starting dialysis therapy the clinical was observed. Most of the essential AA ( EAA) showed symptoms of uraemia diminish or disappear, but sev- higher muscle/plasma gradients, whereas the RBC/ eral of the metabolic disturbances and catabolic factors plasma gradients were approximately 1.0. remain abnormal. Moreover, the different dialysis pro- Conclusion. Our findings are in agreement with studies cedures may induce catabolism as well as protein and that have shown that RBC and plasma play independmalnutrition in maintenance dialysis patients is AA losses into the dialysate [7,8]. Protein energy ent and opposing roles in AA interorgan transport. The results indicate that there are several AA abnor- common, with signs of malnutrition being observed in malities in all three compartments in uraemic patients. 10 to 70% of haemodialysis (HD) patients and in 18% They also suggest that there may be some specific to 51% of continuous ambulatory peritoneal dialysis common changes of selected transport systems for (CAPD) patients [8]. both RBC and muscle in uraemia. Determination of The plasma AA concentrations do not necessarily reflect the intracellular concentrations as the distribution of some AA between the extra- and intracellular Correspondence and offprint requests to: José Carolino Divino Filho MD, Department of Renal Medicine, Huddinge University Hospital compartments is altered [3 5]. One essential link K56, S Huddinge Sweden. between cell metabolism and the extracellular compart European Renal Association European Dialysis and Transplant Association

2 2340 J. C. Divino Filho et al. ment is the transport of AA across the cell membrane. fluid per night with standard PD solution). The patients used The intracellular free AA pool is one of the major the different glucose concentrations according to their need factors involved in the regulation of protein synthesis, to remove excess fluid. The CPD patients had not had any and by measuring the AA concentrations important episodes of peritonitis in the 30 days prior to the investi- gation. The predialysis patients had a median creatinine information on protein metabolism can be obtained. clearance of 9 ml/min ( 5 31 ml/min). Their daily protein The largest store of intracellular free AA is confined intake was estimated on the basis of their daily urinary urea to the skeletal muscle tissue [9], which also contains excretion using an equation presented by Borah et al. [18]. the largest pool of body protein, whereas the plasma The predialysis patients were not given AA supplementation. pool represents a small fraction of the total amount of The protein nitrogen appearance ( PNA) and the dialysis free AA in the body. The RBC have no nucleii, no index expressed as Kt/V of the HD patients was calculated urea mitochondria, no ribosomes or other organelles which based on urea kinetic modelling [19]. Four of the HD exist in muscle and other cells, and therefore they are patients had residual renal function (creatinine clearance incapable of protein synthesis. The RBC contain a between 1.0 and 1.5 ml/min). PNA and dialysis index large proportion of the free AA in blood, the intraer- in CPD patients were calculated from total daily ythrocyte pool of free AA being actively involved in (dialysate+urine) urea excretion and the total loss of protein in the dialysate was added, according to Bergström et al. the interorgan transport of AA [10 12]. Fervenza et al. [20]. Three of the CPD patients had residual renal function [13,14] have demonstrated specific changes in selected (creatinine clearance between 0.5 and 1.8 ml/min). Routine RBC membrane transport systems for AA in uraemia. medication included vitamin B and C supplementation, A few previous studies in patients with chronic renal sodium bicarbonate, phosphate binders, and diuretics. failure [15 17] indicate that the altered RBC AA Body mass index (BMI=BW in kg/(height in m)2) and a pattern is not identical to that in plasma (taken simul- relative weight index (weight index=bw 100/reference taneously) or in muscle (as compared with the literat- BW ) were calculated using actuarial tables from the ure). In the present study, free AA were measured in Metropolitan Life Insurance Company as reference [ 21]. In plasma, muscle, and RBC samples taken simultan- the HD patients, post-dialysis body weight was assessed. eously from patients with chronic renal failure who Arm muscle circumference was calculated as mid-arm circum- were not receiving dialysis (predialysis), as well as ference (cm) 0.1 (p triceps skinfold (mm)) [21]. Skinfold thickness was measured with a Harpenden skinfold caliper from those undergoing either HD or continuous peri- (British Indicators Ltd., St Albans, Herts, UK ). toneal dialysis (CPD). The nature, purpose, and potential risks of the study were carefully explained to all patients before they consented to Subjects and methods participate. The study protocol was approved by the Ethics Committee of the Karolinska Institutet. HD patients were studied, after an overnight fast, on the Subjects morning of a non-dialysis mid-week day. Venous blood (first) and muscle samples were obtained after the subject had The patients characteristics are shown in Table 1. The HD rested in a supine position for 30 min. Muscle samples were patients were dialysed thrice weekly with hollow-fibre diaquadriceps obtained by needle biopsy from the lateral portion of the lysers, using a glucose-free dialysate with acetate or bicarbonour femoris muscle [22]. The results are compared to ate as the buffer, blood flow between 200 and 350 ml/min previously reported data on AA levels in plasma, muscle and dialysate flow 500 ml/min. The CPD patients were and RBC samples obtained under the same conditions from treated either with CAPD (four to five daily 2-litre exchanges) 27 healthy individuals, 16 males and 11 females, with a mean or automated peritoneal dialysis (APD) (10 25 l of dialysis age of 38.5 years [23]. Table 1. Anthropometric, biochemical, and demographic characteristics of the uraemic patient groups and healthy control subjects CPD HD Predialysis Controls n=22 n=38 n=10 n=27 Age (years) 61±12 45±13 60±12 39±13 Time on dialysis (months) 22±24 ( 1 95) 36±51 ( 1 280) Sex 10 M, 12 F 19 M, 17 F 5 M, 5 F 16 M, 11 F Total protein (g/l) 67±3 78±9 72±7 74±4 Albumin (g/l) 29±4 37±4 33±2 41±1 Bicarbonate (mmol/l) 23±2 23±2 21±3 23±1 Insulin (mu/ml ) 18±9 13±4 16±4 8±3 Kt/V# 0.6± ±0.1 PNA# (g/kg/day) 1.1± ± ±0.1 Body mass index# (kg/m2) 22±4 22±3 24±3 23±4 Arm muscle circumference# (cm) 24±3 26±2 25±2 26±3 Weight index# (%) 100±17 101±12 110±14 105±17 Values are given as mean±sd (range). HD, haemodialysis; CPD, continuous peritoneal dialysis; PNA, protein nitrogen appearance. #See text for definitions and formulae.

3 Plasma, muscle and RBC amino acids in uraemia 2341 Analytical procedures (73%), and tyrosine (69%) were significantly reduced in the CAPD patients compared to the controls. The Serum electrolytes and selected routine biochemical factors concentrations of leucine (69%), valine (77%), and were evaluated by routine methods. A heparinized blood tyrosine (60%) were significantly reduced in the presample was centrifuged for 10 min at 4 C in order to obtain dialysis patients compared to the controls. The concenplasma, which was then deproteinized with sulphosalicylic trations of citrulline (275, 274%) were significantly acid (30 mg/ml plasma) and centrifuged. The supernatant was stored at 70 C until the analysis of AA was carried elevated and serine (72, 75%) reduced in both haemoout. For measurement of RBC AA, white cells and platelets dialysis and CAPD patients when compared to con- were carefully removed and 1 g of packed red cells was trols. The concentrations of ornithine ( 79%), glutamine rapidly haemolyzed by adding 1.0 ml of 1% Saponin (Sigma, (87%), and taurine (68%) were reduced in the haemo- St Louis, MO, USA). The sample was then extracted with dialysis patients compared to controls. Citrulline 0.3 ml 50% SSA, mixed and centrifuged at 1700 g for 20 min (252%) was also elevated in the predialysis patients at 4 C. The supernatant was filtered using a 0.45 mm HA compared to controls and taurine was increased comfilter ( Millipore) and frozen at 70 C until analysed [15]. pared to the haemodialysis patients. Total EAA and Norvaline was used as the internal standard. BCAA concentrations were significantly reduced in all Quadriceps femoris muscle samples, obtained by needle three groups of patients compared to controls. biopsy [22], were dissected free from blood and visible connective tissue, weighed repeatedly for extrapolation of the wet weight to time zero, frozen in liquid nitrogen, and Muscle amino acids (Table 3) freeze dried. The freeze-dried samples were weighed, fat was extracted in petroleum ether during 60 min, dried at room The concentrations of isoleucine (132%), leucine temperature, and reweighed. The weight is referred to as fat- (134%), lysine (138%), and phenylalanine (143%) were free solids ( FFS). The sample was powdered in an agat significantly increased in the haemodialysis patients mortar and rinsed from flakes of visible connective tissue. compared to controls. The concentrations of isoleucine The powder was divided into two portions: about 2.5 mg of (132%), phenylalanine (142%), and threonine ( 141%) it was used for the analyses of electrolytes and about 3 mg were significantly increased in the CAPD patients for the determination of muscle free AA. A comprehensive compared to controls. Phenylalanine (186%) was also description of the technique has been previously described [ 24]. Chloride was determined by electrometric titration, as elevated in the predialysis patients compared to condescribed earlier [ 22]. trols. The concentrations of alanine (148, 143%), aspar- Free AA were determined using an automated on-line agine (165, 172%), and glycine (152, 162%) were HPLC system with pre-column derivatization (orthophthaldialdehyde/3-mercaptopropionic significantly elevated and taurine (81, 66%), was acid, OPA/3-MPA) reduced in both haemodialysis and CAPD patients and with norvaline as the internal standard. The reproducibil- compared to controls. The concentration of arginine ity of the method, assessed on the basis of 25 standard (145%) was significantly elevated in haemodialysis analyses, yielded values between 0.4 and 2.2% (coefficient of patients compared to controls whereas CAPD patients variation, CV ). The error of the method was determined had also increased concentrations (130%) but the from 180 duplicate analyses of human plasma samples, increase was not of statistical significance. Glutamic ranging between 1.0 and 4.7% (CV ) [25]. For calculation of intracellular AA concentrations in RBC, acid was higher in CAPD patients than in haemodiathe water content was assumed to be 66% of RBC weight in lysis patients, and taurine was higher in predialysis all samples, as described by Flugel-Link et al. [15]. The patients than in haemodialysis patients. Asparagine calculations of extra- and intracellular water contents and of (149%) was also elevated in the predialysis group when the intracellular AA concentrations in muscle based on the compared to controls. Total AA ( 124%) and BCAA chloride method have been described previously [9]. (121%) concentrations were significantly elevated in the haemodialysis patients compared to controls and Statistical analysis total AA (127%) was also elevated in CAPD patients compared to controls. Taurine was the only AA which Data are reported as means and SD if not stated otherwise. showed the same pattern simultaneously in muscle and For statistical analysis of the data, analysis of variance in plasma, whereas the most striking differences in AA followed by Student s t test with the Bonferroni procedure patterns between muscle and plasma were that of were used; P<0.05 was considered significant. leucine, isoleucine, and lysine. Results RBC amino acids (Table 4) Plasma amino acids (Table 2) The concentrations of histidine (87%, in the following the percentage denotes the mean of the patients in relation to the mean of the healthy control subjects), isoleucine (80%), leucine (71%), valine (79%), tyrosine ( 60%), and lysine (84%) were significantly reduced in the haemodialysis patients compared to controls. The concentrations of histidine (78%), leucine (68%), valine The concentrations of leucine (77, 68, 68%), valine (78, 70, 74%), and tyrosine (62, 60, 55%) were significantly reduced, and histidine ( 119, 120, 134%) was elevated in all three groups of patients when compared to controls, while the concentration of lysine ( 116%) was increased in the haemodialysis patients compared to controls. The concentration of citrulline ( 214, 339, 252%) was significantly increased in all three groups compared to controls and the concentration of arginine

4 2342 J. C. Divino Filho et al. Table 2. Plasma free amino acid concentrations (mmol/l ) in uraemic patients and healthy control subjects CPD HD Predialysis Controls P versus controls Mean SD n Mean SD n Mean SD n Mean SD n CPD HD Predialysis Essential Histidine *** * Isoleucine ** Leucine *** *** ** Lysine * Phenylalanine Threonine Tyrosine * *** * Valine *** *** * Non-essential Alanine Arginine Asparagine Citrulline *** *** *** Glutamic acid * * * Glutamine * Glycine Ornithine Serine *** *** Taurine c ** S EAA *** *** ** S NEAA S BCAA *** *** ** S AA Glycine/serine ratio *** *** ** Valine/glycine ratio * *** Tyrosine/phenylalanine ratio *** *** *** EAA/NEAA ratio ** *** ** *Patients versus controls, P<0.05; **patients versus control, P<0.01; ***patients versus controls, P< chd versus predialysis; P<0.01.

5 Plasma, muscle and RBC amino acids in uraemia 2343 Table 3. Muscle free amino-acid concentrations (mmol/l) in uraemic patients and healthy control subjects CPD HD Predialysis Controls P versus controls Mean SD n Mean SD n Mean SD n Mean SD n CPD HD Predialysis Essential Histidine Isoleucine ** ** Leucine ** Lysine ** Phenylalanine * *** ** Threonine * Tyrosine Valine Non-essential Alanine *** *** Arginine ** Asparagine *** *** *** Citrulline Glutamic acid 6208b ** Glutamine Glycine *** *** Ornithine Serine Taurine c * *** S EAA S NEAA S BCAA * S AA * ** Glycine/serine ratio ** Valine/glycine ratio ** *** Tyrosine/phenylalanine ratio *** *** *** EAA/NEAA ratio *Patients versus controls, P<0.05; **patients versus controls, P<0.01; ***patients versus controls, P< bhd versus CPD; P<0.01; chd versus predialysis; P<0.01.

6 2344 J. C. Divino Filho et al. Table 4. RBC free amino-acid concentrations (mmol/l ) in uraemic patients and healthy control subjects CPD HD Predialysis Controls P versus controls Mean SD n Mean SD n Mean SD n Mean SD n CPD HD Predialysis Essential Histidine ** * ** Isoleucine Leucine *** ** ** Lysine * Phenylalanine Threonine Tyrosine *** *** *** Valine *** *** ** Non-essential Alanine *** Arginine 493a *** *** Asparagine ** Citrulline 161b *** *** *** Glutamic acid Glutamine Glycine ** Ornithine Serine Taurine * * S EAA S NEAA * *** * S BCAA *** ** ** S AA *** Glycine/serine ratio ** ** Valine/glycine ratio *** *** *** Tyrosine/phenylalanine ratio *** *** *** EAA/NEAA ratio *** *** *** *Patients versus controls, P<0.05; **patients versus control, P<0.01; ***patients versus controls, P< ahd versus CPD; P<0.05; bhd versus CPD; P<0.01.

7 Plasma, muscle and RBC amino acids in uraemia 2345 ( 121, 191, 179%) was also elevated in all three groups The muscle/rbc concentration gradient was significantly but did not reach statistical significance for the haemodialysis elevated for isoleucine (136, 156, 137%), leucbut group (121%). Asparagine (129%) and glycine ine ( 146, 188, 154%), valine (142, 145, 141%), ( 124%) were also significantly elevated in the haemodialysis phenylalanine (138, 133, 227%), and tyrosine (153, patients compared to controls, and alanine 172, 149%) in all three groups when compared to (134%) was elevated in the CAPD group compared to controls. The asparagine (136, 161%), glutamine (117, controls. Taurine was elevated in the haemodialysis 132%), and glycine (130, 156%) gradients were also ( 188%) and CAPD (197%) patients compared to controls. elevated significantly, while the taurine gradient was The concentrations of citrulline and arginine decreased in both the haemodialysis (31%) and CAPD were significantly higher in the CAPD patients when (41%) groups compared to controls. The predialysis compared to the haemodialysis patients. The concen- group had increased asparagine (140%) and decreased tration of total NEAA (135, 115, 115%) was significantly citrulline (36%) muscle/rbc gradients and the CAPD elevated, and total BCAA (80, 75, 74%) was group had an increased serine (171%) gradient com- reduced in all three groups compared to controls; total pared to controls. The total BCAA (142, 149, 139%) AA ( 123%) was elevated in the haemodialysis patients muscle/rbc gradient was elevated in all three groups, compared to controls. Alanine, asparagine, serine, and the total NEAA gradient was diminished in glycine, lysine, and arginine levels in RBC and in haemodialysis (72%) and predialysis (76%) groups muscle as well as citrulline, valine, leucine, phenylalanine, when compared to controls. and tyrosine levels in RBC and plasma, showed similar patterns. The most striking differences in AA Amino-acid ratios (Tables 2, 3 and 4) patterns between RBC and muscle were that of taurine and leucine, whereas the main differences between The plasma glycine/serine ratio (160, 149, 141%) was RBC and plasma were observed in the taurine, histidglycine/serine significantly increased in all three groups, the muscle ine, and lysine levels. ratio was elevated in the haemodialysis group (141%) and the RBC glycine/serine ratio was Amino-acids concentration gradients elevated in the haemodialysis (128%) and CAPD (124%) groups, when compared to controls. The muscle/plasma concentration gradient was signiwere The plasma, muscle, and RBC valine/glycine ratio ficantly elevated for all EAA except for histidine in significantly reduced in the haemodialysis (68, both haemodialysis and CAPD patients compared to 66, 63%) and CAPD (73, 61, 64%) groups compared controls; while only leucine ( 154%), valine (137%), to controls. In the predialysis group only the reduced and phenylalanine (233%) gradients were significantly RBC valine/glycine ratio (63%) attained statistical increased in the predialysis patients compared to connificantly significance. The tyrosine/phenylalanine ratio was sig- trols. The muscle/plasma gradients for total EAA decreased for all three compartments in all three groups when compared to controls. (158%), BCAA (155%) and AA (140%) were significantly increased in the haemodialysis group compared The plasma and RBC EAA/NEAA ratio were signito controls. The total BCAA gradient was also elevated ficantly decreased in all three groups of patients com- in the CAPD ( 135%) and predialysis (141%) patients pared to controls, while the muscle EAA/NEAA ratio when compared to controls. The RBC/plasma concen- tration gradient was significantly increased for histidine ( 140, 171%) and lysine (135, 147%) in both haemodialysis and CAPD patients when compared to controls. The histidine gradient (150%) was also elevated in the predialysis patients compared to controls. The gradients for serine (145, 142%) and taurine (241, 315%) was exactly the same for all three patient groups as well as for the controls. Discussion This report presents the first data on AA concentrations were also significantly elevated in both haemodialysis obtained simultaneously from three different and CAPD patients when compared to controls. The compartments (plasma, muscle, and RBC) in patients RBC/plasma serine concentration gradient (121%) was with end-stage renal failure. The results are compared also elevated in predialysis patients compared to con- to our previously reported data on AA concentrations trols. Alanine ( 128, 109%) and arginine (113, 185%) obtained under the same conditions from 27 healthy gradients, in haemodialysis and CAPD patients individuals [23]. respectively, were also elevated compared to controls. In the present study most of the NEAA, especially The RBC/plasma concentration gradient for citrulline taurine and glutamine, had much higher muscle/plasma was significantly higher in the CAPD (147%) and lower gradients than RBC/plasma gradients, although RBC in the haemodialysis (81%) patients when compared accumulation of glycine, serine, arginine, asparagine, to the controls. The total EAA (114, 127%), NEAA ornithine, glutamate, and taurine was observed. Most ( 142, 126%), and BCAA (103, 102%) RBC/plasma of the EAA showed higher muscle/plasma gradients, gradients were significantly higher in the haemodialysis whereas the RBC/plasma gradients were approximately and CAPD groups compared to controls. Moreover 1.0, which indicates a difference between muscle and the RBC/plasma NEAA gradient was higher in the RBC regarding the maintenance of intra/extracellular haemodialysis than in the CAPD patients. gradients for these AA. Studies by Elwyn et al. [11]

8 2346 J. C. Divino Filho et al. Predialysis patients tended to have increased RBC taurine levels too, while the muscle and plasma levels were normal. Flügel-Link et al. [15] have previously reported elevated RBC levels of the sulphur-containing AA in uraemic patients. The increased RBC taurine levels in uraemia may indicate an active accumulation due to altered membrane transport. In this study valine and leucine concentrations were reduced in plasma and RBC, while the muscle concentrations were normal. On the other hand, isoleucine was normal in plasma and RBC, but increased in muscle. Earlier studies have shown low muscle valine levels in the presence of normal or elevated intracellular concentrations of leucine and isoleucine in untreated uraemic patients and in patients treated with HD or intermittent peritoneal dialysis [35,36 ], but not in CAPD patients, who had normal intracellular concentrations of all three BCAA [4]. Reduced muscle valine concentrations in HD patients has been shown to correlate with the degree of metabolic acidosis [5]. Moreover, it has been demonstrated that acidosis appears to enhance muscle protein catabolism in rats with chronic renal failure [37]. This effect seems to be mediated by stimulation of the ATP dependent ubiquit- in proteasome catabolic pathway [38] and by activation of skeletal muscle branched-chain keto-acid dehydrogenase, which increases the catabolism of the BCAA that are mainly metabolized in muscle tissue [39]. Hence, more efficient dialysis treatment with full correction of acidosis and better nutrition may presum- ably explain the differences in the results between the present study and the earlier ones. The elevated RBC histidine and low plasma histidine levels confirm earlier findings in uraemic patients receiving or not receiving histidine supplementation [15 17]. Schmid et al. [40] have found raised histidine levels in the brains of uraemic rats in spite of low plasma levels. The role of anaemia on histidine meta- bolism in uraemia is still unclear but studies on anaemia correction by treatment with erythropoietin may be helpful. None of our patients were being treated with erythropoietin at the time of the study. The results of the only study that has been published so far, on RBC AA concentrations in renal failure (predialysis and HD patients) are to a large extent in agreement with the data presented here [15]. An important question is to what extent changes in RBC AA reflect changes in the other compartments, and especially if there is a covariance between RBC and muscle AA. Elevated concentrations of a number of NEAA (alanine, glycine, asparagine, arginine) and lysine were present simultaneously in RBC and muscle, but not in plasma. On the other hand, low concentrations of some EAA ( leucine, valine, phenylalanine, tyrosine) were observed in RBC and in plasma in the presence of normal concentrations in muscle. The results may suggest that the determination of AA in RBC is a more sensitive method for detecting a defi- ciency in one of these AA than AA analysis in muscle tissue, an exception being taurine. Our findings that the concentration of some AA in suggested that an exchange of AA between plasma and tissue cells and an exchange between RBC and tissue cells may take place independently. Furthermore, data published by Felig et al. [12] implicate the blood cellular elements as important carriers in the net flux of various AA between peripheral tissues, gut, and liver in normal humans. The patients in our study showed increased lysine and glycine levels both in RBC and muscle compared to controls, while plasma levels were decreased and normal respectively. Muscle and RBC serine levels in the patients were not different from controls, while plasma levels were decreased. Several AA transport systems have been described in human RBC and these appear similar to those found in other tissues [26 29]. Earlier studies have shown an increased lysine transport capacity in RBC in patients with chronic renal failure [13] as well as abnormal RBC membrane transport of serine and glycine in HD patients [14]. The aromatic AAs tyrosine and phenylalanine show a significant transport affinity for system T, which seems to be confined to human RBCs [29]. In plasma and RBC the tyrosine levels were decreased and the phenylalanine levels were the same as in the controls, whereas in muscle phenylalanine level was increased and the tyrosine level was not different from the controls, in all three groups. Low tyrosine/phenylalanine ratio has been described earlier in the plasma, muscle and RBC of dialysis patients and it is attributed to a reduced synthesis of tyrosine from phenylalanine due to the inhibition of phenylalanine hydroxylase [30] and/or to an increased activity of cytosol tyrosine aminotransferase [31]. In our patients with chronic renal failure, although plasma and RBC citrulline levels were increased when compared to controls, arginine levels were not reduced, suggesting an alternative pathway for either arginine synthesis or metabolism. Citrulline is normally taken up by the kidneys and converted to arginine and it has been suggested that the high citrulline levels in uraemic patients may be the result of reduced conversion. The high citrulline concentration in plasma and RBC in presence of normal or high arginine levels may also suggest an increased activity of the nitric oxide (NO) pathway in uraemia, resulting in increased citrulline production. On the other hand there is some evidence that NO production may be impaired by accumulation of endogenous inhibitors of NO synthase; a mechanism proposed to be involved in the development of hypertension and vascular disease [32]. The CAPD patients showed significantly increased RBC levels of citrulline and arginine when compared to the HD patients, which may denote differences in the way that dialysis modalities influence the citrulline arginine pathway. The HD and CPD patients showed evidence of taurine depletion with significantly reduced muscle levels and in the HD patients the plasma levels were also reduced, when compared to the controls. These findings are in keeping with earlier results from our group [33,34]. Interestingly, RBC taurine concentrations were elevated in both HD and CPD patients.

9 Plasma, muscle and RBC amino acids in uraemia 2347 RBC, but not in plasma, reflects the concentration in plasma and erythrocytes in interorgan transport of amino acids in dogs. Am J Physiol 1972; 222: muscle are in agreement with studies that have shown 12. Felig P, Wahren J, Räf L. Evidence of inter-organ amino acid that RBC and plasma play independent and frequently transport by blood cells in humans. Proc Natl Acad Sci USA opposing roles in AA interorgan transport in several 1973; 70: mammalian species [11,12]. Moreover, the gradient 13. Fervenza FC, Harvey CM, Hendry BM, Ellory JC. Increased data suggest that although there are differences in the lysine transport capacity in erythrocytes from patients with chronic renal failure. Clin Sci 1989; 76: membrane transport characteristics of several AA 14. Fervenza FC, Meredith D, Ellory JC, Hendry BM. A study of between RBC and muscle, there are some specific the membrane transport of amino acids in erythrocytes from common changes of selected membrane transport sys- patients on haemodialysis. Nephrol Dial Transplant 1990; 5: tems for both RBC and muscle in uraemia We conclude that measurement of AA in RBC can 15. Flugel-Link RM, Jones M, Kopple JD. Red cell and plasma amino acid concentrations in renal failure. J Parent Ent Nutr give important additional information to that obtained 1983; 7: from plasma aminograms, especially as analysis of 16. Jontofsohn R, Trivisas G, Katz N et al. Amino acid content of RBC AA does not require either larger blood samples erythrocytes in uremia. Am J Clin Nutr 1978; 31: or laborious procedures for preparing the samples. 17. Ganda OP, Aoki TT, Soeldner JS et al. Hormone-fuel concentra- tions in anephric subjects: effect of hemodialysis. J Clin Invest Muscle biopsy, on the other hand, is an invasive and 1976; 57: sometimes uncomfortable procedure, which precludes 18. Borah M, Schoenfeld P, Gotch F, Sargent J, Wolfson M, its utilization in large groups of patients. The deter- Humphreys M. Nitrogen balance during intermittent dialysis mination of free AA in RBC should be seen as a therapy of uremia. Kidney Int 1978; 14: complementary tool in order to add further insight 19. Farrel PC, Gotch FA. Dialysis therapy guided by kinetic model- ling: applications of a variable-volume single-pool model of urea into the nature of the AA and protein metabolism kinetics. Second Australasian Conference on Heat and Mass disturbances occurring in uraemia and other disease Transfer. Australia, The University of Sydney, 1977; conditions. We therefore recommend the determination 20. Bergström J, Fürst P, Alvestrand A, Lindholm B. Protein and of free AA in simultaneously collected samples of energy intake, nitrogen balance and nitrogen losses in patients plasma and RBC, when undertaking metabolic and treated with continuous ambulatory peritoneal dialysis. Kidney Int 1993; 44: clinical studies of AA abnormalities, although this 21. Wright RA, Heymsfield SB. Nutritional Assessment. Blackwell should be not seen as a substitute for muscle AA Scientific Publications, Boston; 1984 determination. 22. Bergström J. Muscle electrolytes in man. Scand J Clin Lab Invest 1962; 14: Suppl 68 Acknowledgements. These studies were supported by a grant from 23. Divino Filho JC, Stehle P, Fürst P, Bergström J. Plasma, muscle the Swedish Medical Research Council, Project no The and erythrocyte free amino acids and nutritional status in authors wish to thank Ann Helström for her secretarial assistance healthy subjects. Clin Nutr (Submitted) as well as Ms Elsy Digreus, Ms Ulla Petersson, Ms Amy Forsberg, 24. Forsberg AM, Nilsson E, Wennerman J, Bergström J, and Ms Eva Nilsson for their technical assistance. Hultman E. Muscle composition in relation to age and sex. Clin Sci 1991; 81: Fürst P, Pollack L, Graser TA, Godel H, Stehle P. Appraisal of References four pre-column derivatization methods for the highperformance liquid chromatographic determination of free 1. Kopple JD. Amino acid metabolism in chronic renal failure. In: amino acids in biological materials. J Chromatogr 1990; 499: Blackburn GL, Grant JP, Young VR, (eds.), Amino Acids, Metabolism and Medical Applications. John Wright, Boston 26. Young JD, Ellory JC. Red cell amino acid transport. In: Ellory 1983; JC, Lew VL, (eds.), Membrane Transport in Red Cells. Academic 2. Gulyassy PF, Peters JH, Lin SC, Ryan PM. Hemodialysis and Press, New York, 1977; plasma amino acid composition in chronic renal failure. Am 27. Ellory JC. Amino acid transport systems in mammalian red J Clin Nutr 1968; 21: cells. In: Yudilevich DL, Boyd CAR, (eds.), Amino Acid 3. Alvestrand A, Fürst P, Bergström J. Intracellular amino acids Transport in Animal Cells. University of Manchester Press, in uremia. Kidney Int 1983; 24: 9 16 Manchester, 1987; Lindholm B, Alvestrand A, Fürst P, Bergström J. Plasma and 28. Harvey CM, Ellory JC. Identification of amino acid transporters muscle free amino acids during continuous ambulatory periton- in red blood cells. Methods Enzymol 1989; 173: eal dialysis. Kidney Int 1989; 35: Rosenberg R. A kinetic analysis of L-tryptophan transport in 5. Bergström J, Alvestrand A, Fürst P. Plasma and muscle free human red blood cells. Biochim Biophys Acta 1981; 649: amino acids in maintenance hemodialysis patients without prohydroxylation in chronic renal insufficiency. Clin Sci Mol Med 30. Young G, Parsons F. Impairment of phenylalanine tein malnutrition. Kidney Int 1990; 38: Canepa A, Divino Filho JC, Forsberg AM et al. Nutritional 1973; 48: 88 status and muscle amino acids in children with end-stage renal 31. Sapico V, Shear L, Litwack G. Translocation of inducible failure. Kidney Int 1992; 41: tyrosine aminotransferase to the mitochondrial fraction. J Biol 7. Gutierrez A, Alvestrand A, Wahren J, Bergström J. Effect of in Chem 1974; 249: 2122 vivo contact between blood and dialysis membranes on protein 32. Ritz E, Vallance P, Nowicki M. The effect of malnutrition on catabolism in humans. Kidney Int 1990; 38: cardiovascular mortality in dialysis patients: is L-arginine the 8. Bergström J, Lindholm B. Nutrition and adequacy of dialysis: answer? Nephrol Dial Transplant 1994; 9: how do hemodialysis and CAPD compare? Kidney Int 1993; 33. Bergström J, Alvestrand A, Fürst P, Lindholm B. Sulphur 43: amino acids in plasma and muscle in patients with chronic renal 9. BergströmJ, Fürst P, Noree LO, Vinnars E. Intracellular free failure: evidence for taurine depletion. J Intern Med 1989; amino acid concentration in human muscle tissue. J Appl Physiol 226: ; 36: Suliman M, Anderstam B, Bergström J. Evidence of taurine 10. Christensen HN. Interorgan amino acid nutrition. Physiol Rev depletion and accumulation of cysteinesulfinic acid in chronic 1982; 62: dialysis patients. Kidney Int 1996; 50: Elwyn DH, Launder WJ, Parikh HC, Wise EM Jr. Roles of 35. Alvestrand A, Fürst P, Bergström J. Plasma and muscle free

10 2348 J. C. Divino Filho et al. amino acids in uremia: influence of nutrition with amino acids. stimulates muscle protein degradation by activating the ATPdependent Clin Nephrol 1982; 18: pathway involving ubiquitin and proteasomes. J Clin 36. BergströmJ, Fürst P, Noree LO, Vinnars E. Intracellular free Invest 1994; 93: amino acid concentration in human muscle tissue of patients 39. May R, Hara Y, Kelly R, Block K, Buse M, Mitch W. Branched- with chronic uremia: effect of peritoneal dialysis and infusion of chain amino acid metabolism in rat muscle: abnormal regulation amino acids. Clin Sci Mol Med 1978; 54: in acidosis. Am J Physiol 1987; 252: E712 E Hara Y, May R, Kelly R, Mitch W. Acidosis, not azotemia, 40. Schmid G, Fricke LA, Lange HW, Heidland R. Intracellular stimulates branched-chain amino acid catabolism in uremic rats. histidine content of various tissues (brain, striated muscle and Kidney Int 1987; 32: liver) in experimental chronic renal failure. Klin Wochenschr 38. Mitch W, Medina R, Greiber S et al. Metabolic acidosis 1977; 55: 583 Received for publication: Accepted: