Search

Progesterone-binding plasma proteins (PBPP, progesterone-binding globulin, PBG, and corticosteroid binding globulin, CBG) have been measured in plasma of guinea-pigs, casiragua, cuis, degu and plains viscacha. During pregnancy PBPP increased to reach peak values between Days 20-25 and Days 50-term in guinea-pigs, immediately before parturition in casiragua, and in mid-gestation in cuis, degu and viscacha. The pattern of PBPP concentration during pregnancy was similar to that of plasma progesterone concentration, and the ratio of molar concentration of PBPP and progesterone was greater than 1 in all species. Plasma proteins were separated by chromatography on SP-Sephadex C-50, and from its chromatographic behaviour and progesterone-binding property evidence was obtained for an anionic, high-affinity PBG in the plasma of all the animals studied during gestation: Electrophoretic mobility of SP-Sephadex-purified PBG applied to 7% polyacrylamide gels differed between species, being highest for casiragua PBG and lowest for guinea-pig PBG. Binding capacities ranged from 0.5 to 4.8 X 10(-6) M, and association constants from 7.6 to 17.4 X 10(8) M-1. Ligand specificities differed between species and marked contrasts were found for certain steroid isomers. The average molecular weights measured by gel electrophoresis of PBG under denaturing conditions were similar in cuis, degu and viscacha and close to PBG I of guinea-pigs, while coypu PBG had a molecular weight of about one half, approximately 75000. It is concluded that PBG has been adopted in pregnancy as a progesterone-conserving mechanism in this suborder of rodents with long gestation periods relative to maternal body weight. Restricted homologies in the physico-chemical characteristics of PBG from various hystricomorph rodents suggest that interspecific differences in structure are considerable.

Serum levels of alpha 1 microglobulin (s-alpha 1 m) in 92 recipients of renal transplants were elevated during pretransplant uremia (P less than 0.001), acute rejection (P less than 0.01), and cyclosporin-induced nephrotoxicity (P less than 0.01). In patients with stable renal function, those treated with cyclosporin had higher s-alpha 1 m than those receiving azathioprine: 81 +/- 4 and 64 +/- 3 mg/l (mean +/- SEM), respectively (P less than 0.05). The serum creatinine levels were 127 +/- 5 and 115 +/- 7 mumol/l (mean +/- SEM), respectively (N.S.). Two of the patients with normal serum creatinine had normal s-alpha 1m levels. There were positive linear correlations between s-alpha 1m and serum creatinine levels during stable renal function, rejection, cyclosporin-induced nephrotoxicity, and cytomegalovirus infections (r = 0.7-0.8, P less than 0.01-0.001) and between s-alpha 1m and beta 2 microglobulin (beta 2m) during the same conditions (r = 0.5-0.8, P less than 0.01-0.001). During infections, serum creatinine and beta 2m increased (P less than 0.001), but s-alpha 1m did not. S-alpha 1m values did not distinguish between rejection and cyclosporin-induced nephrotoxicity. It is concluded that s-alpha 1m might be a valuable complement to serum creatinine levels in the evaluation of renal function in renal transplant recipients.

This study was performed on patients (n = 18) suffering from strictly defined hyperdynamic septic shock. Plasma factors (C-reactive protein, acid alpha 1-glycoprotein, fibrinogen, fibrinopeptide A, fibrinogen-fibrin split products, factor XIII, antithrombin III, complement factors C3 and C4, inter-alpha-trypsin-inhibitor and alpha 2-macroglobulin) measured during hyperdynamic septic shock were highly abnormal. The activation and consumption of clotting, fibrinolytic and complement factors due to system-specific proteinases (such as thrombokinase or plasminogen activators) seemed to be intensified by the nonspecific proteolytic activity of granulocytic proteinases probably released by the action of endotoxins. Possible therapeutic measures to maintain the endogeneous defence mechanism against enhanced proteolysis during septic shock are discussed.

Evidence is presented that (transcortin and alpha 2u-globulin react as negative acute-phase proteins in the rat. Thirty-six hours after turpentine injection, the serum concentration of these proteins showed a two- to threefold decrease. Thereafter, transcortin rapidly returned to normal values, whereas alpha 2u-globulin remained low. This reaction pattern was still present after adrenalectomy, adrenalectomy and administration of glucocorticoids, and after treatment with bromocriptine, a suppressor of prolactin secretion. It is concluded that changes in the secretion of glucocorticoids and prolactin are not required for the observed turpentine-induced decrease of transcortin and alpha 2u-globulin.

Proximal renal tubular function was studied in 522 consecutive patients entered into the Medical Research Council's fourth myelomatosis trial. Assessment was made at presentation after a 48 h period of hydration but before administration of chemotherapy. The most common abnormalities in the urine other than light chain proteinuria were raised concentrations of the low molecular weight proteins alpha 1-microglobulin and alpha 1-acid glycoprotein. These were usually accompanied by increases in urinary beta-N-acetyl-D-glucosaminidase concentrations. The concentration of these substances in the urine directly correlated with urinary free light chain output. This tubular proteinuria was seen whether or not patients had impaired glomerular function, as assessed by a rise in serum creatinine concentration. Urinary concentrations of retinol binding protein, however, were generally increased only when serum creatinine concentrations were raised. This applied even when there were high concentrations of light chains, alpha 1-microglobulin, alpha 1-acid glycoprotein, and beta-N-acetyl-D-glucosaminidase in the urine. There is therefore a selective tubular proteinuria in myelomatosis which is seen in almost all patients with urinary light chain values greater than 1 u/l. This proteinuria is generally reversible, when light chains no longer appear in the urine. Patients whose serum creatinine was greater than 200 mumol/l, however, had increased urinary output of retinol binding protein in addition to increased excretion of alpha 1-microglobulin, alpha 1-acid glycoprotein, and beta-N-acetyl-D-glucosaminidase. Tubular proteinuria in many of these patients presenting in renal failure persisted even when light chain output was reduced after chemotherapy.

Antiserum was obtained from rabbits immunized with a highly purified preparation of urinary trypsin inhibitor (UTI) conjugated with rabbit serum albumin. Anti-UTI antibody did not cross-react with antibody against inter-alpha-trypsin inhibitor (I alpha I) as determined by immunodiffusion against human plasma. A highly sensitive enzyme immunoassay was developed for the determination of UTI-related antigens (UTIR) in plasma, urine and cancer tissues. The level of UTIR correlated with UTI activity in urine. UTIR levels in urine and plasma did not change with age, but UTIR levels were higher in the stomach cancer tissue than in the surrounding stomach mucosa. UTIR levels did not correlate with I alpha I levels in the plasma of patients with stomach cancer, thus the increase was not considered due to the contamination of circulating I alpha I in the cancer tissues.

alpha 1-Microglobulin (alpha 1m), a glycoprotein (Mr = 30,000) found in serum and urine, is also present in serum conjugated to monomeric IgA (alpha 1-m-IgA). We have developed a simultaneous enzyme-linked immunoenzyme/immunoradiometric assay that involves three different monoclonal antibodies. Assay of serial dilutions of serum and urine demonstrated parallelism. Normal mean concentrations in serum (n = 75) were: total alpha 1m, 2.33 mumol/L; alpha 1m-IgA, 1.24 mumol/L; unconjugated (free) alpha 1m, 1.09 mumol/L; molar ratio (alpha 1m-IgA/total alpha 1m), 0.53. The mean concentration of alpha 1m in eight urine specimens from normal individuals was 0.19 mumol/L, with no detectable alpha 1m-IgA. A low urinary pH does not significantly affect assay results, unlike assays of beta 2-microglobulin. In patients with myeloma-related renal disease, total and free alpha 1m values for serum correlated well with values for creatinine and beta 2-microglobulin in serum.

Serum low-molecular-weight proteins, beta-2-microglobulin and alpha-1-microglobulin, were measured in 34 patients undergoing maintenance haemodialysis. There was a strong positive correlation between beta 2M and alpha 1M and predialysis serum creatinine and strong negative correlations between these proteins and 24-h urine volume and creatinine clearance. Good correlation was also demonstrated with total duration of haemodialysis, which also correlated with residual renal function. It is suggested that the elevated beta 2M and alpha 1M in haemodialysis patients merely reflect the severity of renal failure and are not directly influenced by the duration of dialysis or type of dialysis membrane.

The stability of human chorionic gonadotropin (hCG) and its alpha-subunit in whole blood, plasma, and serum under a variety of sample handling conditions commonly encountered in clinics, hospital wards, physician's offices, and clinical service laboratories was investigated with the use of radioreceptor assay, radioimmunoassays, as well as hormone integrity determinations. The results clearly demonstrate that hCG and its alpha-subunit are stable in unfrozen whole blood, plasma, and serum for at least 6 days and in frozen plasma and serum samples for at least 6 months. Repeated freezing and thawing of the samples during this period had no effect. Separation of plasma or serum from erythrocytes is not needed for at least 12 hours. Hemolysis in samples resulted in a 20% to 30% decrease in hCG and its alpha-subunit levels, which may be attributable to sample dilution.

Tumour tissues obtained from 35 patients with malignancies of the female genital tract were investigated for production of pregnancy specific beta 1-glycoprotein (PS beta 1G) and alpha 2-pregnancy associated globulin (alpha 2-PAG). PS beta 1G was found in all five trophoblastic tumours studied and in 10 of the 30 nontrophoblastic cancers. alpha 2-PAG was not detected in any of the neoplastic tissues. In 18 of the patients with non-trophoblastic tumours PS beta 1G and alpha 2-PAG serum concentrations were also determined. No correlation between serum and tissue findings were noted. Thus, PS beta 1G does not seem to be a valuable serum indicator for monitoring the extent or variation of tumour mass in nontrophoblastic gynecological malignancies. Likewise, alpha 2-PAG is unlikely to serve as a clinical useful tumour marker in various gynecological cancers.

A sensitive and rapid enzyme-linked immunosorbent assay for simultaneous quantitation of free and IgA-complexed human protein HC was developed with monoclonal murine antibodies. The total amount of protein HC (free plus IgA-complexed) was measured by a competitive procedure while the protein HC-IgA complex was quantitated by a sandwich enzyme immunoassay. The amount of free protein HC was then obtained as the difference between the 2 measured values. The sensitivity of the procedure was 70 micrograms/1 for the total amount of protein HC and 80 micrograms/1 for the protein HC-IgA complex. At ordinary human serum levels of free protein HC and protein HC-IgA complexes the coefficient of variation for the procedure was about 5%. One worker could determine the concentrations of free and IgA-complexed protein HC in up to 400 serum samples in a working day.

The plasma clearance of neutrophil elastase, plasmin, and their complexes with human inter-alpha-trypsin inhibitor (I alpha I) was examined in mice, and the distribution of the proteinases among the plasma proteinase inhibitors was quantified in mixtures of purified inhibitors, in human or murine plasma, and in murine plasma following injection of purified proteins. The results demonstrate that I alpha I acts as a shuttle by transferring proteinases to other plasma proteinase inhibitors for clearance, and that I alpha I modulates the distribution of proteinase among inhibitors. The clearance of I alpha I-elastase involved transfer of proteinase to alpha 2-macroglobulin and alpha 1-proteinase inhibitor. The partition of elastase between these inhibitors was altered by I alpha I to favor formation of alpha 2-macroglobulin-elastase complexes. The clearance of I alpha I-plasmin involved transfer of plasmin to alpha 2-macroglobulin and alpha 2-plasmin inhibitor. Results of distribution studies suggest that plasmin binds to endothelium in vivo and reacts with I alpha I before transfer to alpha 2-macroglobulin and alpha 2-plasmin inhibitor. Evidence for this sequence of events includes observations that plasmin in complex with I alpha I cleared faster than free plasmin, that plasma obtained after injection of plasmin contained a complex identified as I alpha I-plasmin, and that a murine I alpha I-plasmin complex remained intact following injection into mice. Plasmin initially in complex with I alpha I more readily associated with alpha 2-plasmin inhibitor than did free plasmin.

A single-point progesterone tracer binding assay (TBA) was developed to measure the concentration of progesterone-binding globulin (PBG) in the serum of pregnant guinea-pigs. Tritiated progesterone was added to dilutions of a reference serum, the percentage bound calculated and a standard curve constructed. The amount of progesterone tracer bound to serum samples was determined, and the concentration of binding protein in the sample relative to the reference was calculated. The molar concentration in the reference serum was determined by Scatchard analysis. The TBA could readily process 30 samples per day. There was a rapid rise in PBG concentrations between days 15 and 20 of gestation, in parallel with progesterone concentrations. The molar ratio of PBG: progesterone was approximately 10:1 after this time, until about day 45 when there was an increase in ratio.

The amino-acid sequences of the acid-resistant inhibitors released from horse and pig inter-alpha-trypsin inhibitor (ITI) by tryptic proteolysis were determined. They are composed of two covalently linked Kunitz-type domains. In both cases the reactive site of their C-terminal antitryptic domains is occupied by arginine as in the homologous human and bovine inhibitors. The reactive site of their N-terminal domain exhibits only a weak interaction with polymorphonuclear granulocytic elastase and is occupied by leucine as in the strong elastase inhibitor released from bovine ITI. The differences between inhibitory activities of the ITI-derived inhibitors from horse, pig, and cattle are discussed on the basis of sequence differences in position P'2.

The binding properties towards vitamin D metabolites of plasma from individuals with the three common Gc-globulin phenotypes, Gc-1, Gc-2 and Gc-2-1, have been found to be identical. In patients with liver disease there is a good correlation between the levels of Gc-globulin andalbumin in plasma. In addition the Gc-globulin levels correlate well with the ability of plasma to bind 25-hydroxycholecalciferol. Patients with the Gc-2-1 phenotype showed a significantly smaller depression in plasma Gc-globulin than those with the Gc-2 and Gc-1 phenotypes. The relation of these findings to the pathogenesis of disorders of calcium metabolism in liver disease is discussed.

A 66-year-old female patient complained of loss of body weight and fatigue. The clinical examination revealed a thrombocytosis with a maximum count of 3.200 . 10(9) platelets and a leukocytosis with maximally 25 . 10(9) white cells in the peripheral blood. The bone marrow showed a large increase of megakaryocytes. Under the diagnosis of megakaryocytic myelosis a chemotherapy with 186 mg busulfan was performed. In the course of this treatment the clinical picture of a sepsis occurred which could not be controlled by antibiotics. The patient died four months after her admission to the clinic. The essential findings in autopsy were a caseous tuberculosis of the lymph nodes with haematogenic generalization which appeared as a septic tuberculosa gravissima ("typhobacillosis" Landouzy). The bone marrow was atrophic. Spleen liver and lymph nodes were without evidence for a myeloproliferative disorder. Thus, the initial diagnosis had to be changed to a megakaryocytic pseudomyelosis with massive thrombocytosis as a reaction to the tuberculous infection. The differential diagnosis of megakaryocytic myelosis, other disorders of the myeloproliferative syndrome, and the reactive thrombocytosis are discussed.

The levels of serum alpha 1-microglobulin in 60 normal persons and in 191 patients suffering from a variety of benign and malignant disorders were determined by an enzyme immunoassay, and these values were compared with the levels of beta 2-microglobulin. A discrepancy between the serum levels of these proteins was found in hepatobiliary disorders; that is, an increased serum level of beta 2-microglobulin was observed in 73.9%, while in only 4.3% was there an elevation of alpha 1-microglobulin. In particular, alpha 1-microglobulin levels in patients with liver cirrhosis were well below the normal range, while beta 2-microglobulin levels were elevated. Elevated levels of both proteins were noted in patients with some impairments of renal function, particularly in chronic renal failure, and in immunological diseases. In 81 patients with neoplastic diseases, a high alpha 1-microglobulin value was found in only 15 patients (16.4%), while a high beta 2-microglobulin value in 62 patients (76.5%). The serum levels of both alpha 1-microglobulin and beta 2-microglobulin were especially high in plasma cell dyscrasia with Bence Jones protein, but other neoplastic diseases were mostly associated with beta 2-microglobulin elevation alone.

Two solid phase enzyme immunoassays for the measurement of alpha1-microglobulin (alpha1m) in biological fluids are reported. These assays are complementary and allow the separate determination of free alpha1m and alpha1m bound to IgA without fractionation of the fluids. Specificity, reproducibility and sensitivity of the assays are reported. Differential measurements on normal or pathological sera were compared to the measurements obtained by chromatographic separation of free and bound alpha1m before assay. Results obtained with the enzyme immunoassays were significantly correlated with those obtained by radioimmunoassay.

The relationship between the histological and clinical findings of 128 cases of renal cell carcinoma operated on during the past 17 years were analyzed. With elevation of tumor grading, the elevated ESR and CRP-positive cases were increased with a significant difference. The serum alpha 2-globulin was elevated in the grade 3 group significantly, but not in the grade 1 or 2 group. Seven of the eight spindle cell type cases showed an elevation of ESR, CRP-positive and an elevation of alpha 2-globulin. There was no difference in clinical findings between well circumscribed border and poorly demarcated margin. A high percentage of quick type cases of renal cell carcinoma were found in high grade cases with a significant difference.

The N-terminal amino-acid sequence of human ITI has been found to be identical with that of the acid-stable human 30-kDa inhibitors (HI-30) from urine, serum, and those released from inter-alpha-trypsin inhibitor by trypsin or chymotrypsin. Serum HI-30 and HI-30 released by trypsin differ from the urinary inhibitor by an additional C-terminal arginine residue. Compared to these two inhibitors the inhibitor released by chymotryptic proteolysis is elongated C-terminally by an additional phenylalanine residue. These results strongly favour HI-30 as the N-terminus of the inter-alpha-trypsin inhibitor and its release from this inhibitor in vivo by cleavage of the Arg123-Phe124 peptide bond by trypsin-like proteinases.

Alpha-M-fetoprotein (AMFP) was detected in the serum of rats subjected to intraoral manipulations, i.e., instillations into the parotid gland and injections into the gingival mucosa of saline or foreign proteins. The identity of AMFP with acute-phase protein is pointed out.

The level of progesterone in the serum of pregnant guinea-pigs which continued their pregnancies after ovariectomy on Day 30 fell and then rose again. In contrast, progesterone-binding globulin (PBG) was apparently not affected. PBG did however fall in guinea-pigs which aborted and also after parturition. It is apparent therefore that high PBG levels were associated with continued pregnancy. The half life of PBG was found to be approximately 2 days.

Serum total proteins as well as its electrophoretic separated fractions were determined in alloxan-diabetic rats. The total and individual serum amino acids were estimated. Certain abnormalities, with no specific pattern, for total and individual fractions were observed in alloxan-diabetic rats. Most of the free serum amono acids showed low values which was attributed to direct reaction of alloxan with amino acids, urinary loss of amino and/or diverse hormonal and metabolic changes.

The sensitivity of crossed immunoelectrophoresis (CIE) and rocket immunoelectrophoresis (RIE) does not usually permit the study of diluted protein solutions or the detection of trace components. Antigen-antibody precipitates obtained after immunoelectrophoresis were therefore dissociated and transferred from the agarose gel onto nitrocellulose filter by electrophoresis at pH 11. Immunochemical detection of proteins was then performed with the peroxidase-anti-peroxidase method. Diluted (1/100) solutions gave patterns similar to those obtained with the starting biological fluid. As an example, this technique was applied to human inter-alpha-trypsin inhibitor present at concentrations in the nanogram range. Other proteins, differing in charge and/or molecular weight, were also identified.

This rapid screening procedure for detection of low but functional elastase-inhibitory activity in human plasma is based on the fact that incubation of excess porcine pancreatic elastase (EC 3.4.21.36) with plasma results in formation of a complex with active alpha 1-proteinase inhibitor (alpha 1PI, also called alpha 1-antitrypsin). In normal individuals all of the elastase is complexed, leaving no free enzyme to hydrolyze the elastase substrate, and the reaction mixture remains clear. Because individuals homozygous for the Z allele have relatively low concentrations of alpha 1PI, their plasma cannot complex all of the elastase in the assay. The uncomplexed enzyme hydrolyzes the elastase-specific p-nitroanilide substrate, producing a yellow reaction mixture. Use of this simple assay for early screening of individuals for alpha 1PI deficiency may substantially decrease the number of untreated cases of familial emphysema, a disorder that develops as a result of a genetically derived proteinase-proteinase inhibitor imbalance.

Plasma levels of two oral contraceptive gestagens were studied during 24 hour oral challenges in seven morbidly obese women after jejunoileal bypass comparing the results to unoperated normal controls. To study binding and transport of the gestagens, sex hormone binding globulin levels were determined in these patients, in normal-weight controls, and in unoperated morbidly obese patients. The unoperated morbidly obese patients had significantly decreased levels of sex hormone binding globulin, compared to normal-weight controls and to the jejunoileal bypass patients, whose levels were similar to the controls. The jejunoileal bypass patients had a reduced capacity to absorb oral contraceptive gestagens, although the plasma levels were of the same magnitude found in normal subjects using lower doses of gestagens. However, the gestagen to sex hormone binding globulin ratio was lower in jejunoileal bypass patients, implying reduced biologic activity and thus reduced contraceptive efficiency of the gestagen. There were no enzymatic or morphologic signs of liver dysfunction. No correlations were found between plasma gestagen levels and number or volume of stools, fecal fat excretion, or intravenous C-cholic acid load. It is concluded that caution must be exercised in prescribing oral contraceptives to jejunoileal bypass patients.

An apparently novel human plasma and urinary protein of low molecular weight was isolated from several highly purified preparations of protein HC by gel chromatography and high voltage electrophoresis with a yield of about 8 mg/g. The protein has a molecular weight of about 20,000, neutral electrophoretic mobility at pH 6.5 and a high content of half-cystine. It is associated with a yellow-brown chromophore like protein HC and could be demonstrated in all investigated preparations of isolated human, rabbit and guinea-pig protein HC and alpha 1-microglobulin.

The concentration in serum of the acid-stable Mr 30.000 fragment of the Inter-alpha-Trypsin Inhibitor (SPI), and the acid-stable Mr 14.000 fragment of the inhibitor prepared by trypsin by digestion of the acid-precipitated serum proteins (API) was determined. A considerable increase in concentration of SPI and a decrease in concentration of API were observed in inflammatory conditions. In uraemic conditions the concentration of SPI was increased while that of API was unchanged.

Circulating concentrations of progesterone, progesterone-binding plasma proteins (PBPP) and oestradiol-17 beta in pregnant porcupines remained relatively low until Days 25-30 post coitum. Progesterone values peaked (102-180 ng/ml; N = 3) 42-60 days post coitum and the rapid increase in oestradiol-17 beta concentrations approximated that of progesterone with peak values (170-210 pg/ml) being attained 60-85 days post coitum. The pattern of PBPP synthesis, as suggested by circulating concentrations, was closely related to that of plasma progesterone, with values remaining low (less than 20 pmol/ml) until Day 31 post coitum, reaching peak levels at Days 50-56 and Days 73-77 post coitum. The production of PBPP during pregnancy is, as in related New World hystricomorph species, considered to be a mechanism which facilitates a reduction in the rate of progesterone metabolism during pregnancy.

Inter-alpha-trypsin inhibitor (I alpha I) is a unique proteinase inhibitor that can be proteolyzed by the same enzymes that are inhibited, to generate smaller inhibitors. This study examines the reactions of I alpha I with trypsin, chymotrypsin, plasmin, and leukocyte elastase. Complexes of I alpha I and proteinase were demonstrated by gel filtration chromatography. Complete digestion of I alpha I by each proteinase was not accompanied by a comparable loss of inhibition of that enzyme or a different enzyme. Following proteolysis, inhibitory activity was identified in I alpha I fragments of molecular weight 50,000-100,000 and less than 40,000. Addition of a second proteinase inhibitor prevented proteolysis. Both I alpha I and its complex with proteinase were susceptible to degradation. Kinetic parameters for both the inhibition and proteolysis reactions of I alpha I with four proteinases were measured under physiological conditions. On the basis of these results, a model for the mechanism of action of I alpha I is proposed: Proteinase can react with either of two independent sites on I alpha I to form an inhibitory complex or a complex that leads to proteolysis. Both reactions occur simultaneously, but the inhibitory capacity of I alpha I is not significantly affected by proteolysis since the product of proteolysis is also an inhibitor. For a given proteinase, the inhibition equilibrium constant and the Michaelis constant for proteolysis describe the relative stability of the inhibition and proteolysis complexes; the second-order rate constants for inhibition and proteolysis indicate the likelihood of either reaction. The incidence of inhibition or proteolysis reactions involving I alpha I in vivo cannot be assessed without knowledge of the exact concentrations of inhibitor and proteinases; however, analysis of inhibition rate constants suggests that I alpha I might be involved in plasmin inhibition.

The chronic retention of proteins of low molecular weight (LMWP) seems to be connected with some unsolved problems of end-stage renal disease (e.g. reduced immune function, AB amyloidosis, hormonal imbalances), irrespective of the type of renal replacement therapy. Therefore, several recently discovered LMWPs such as alpha 1-microglobulin have to be evaluated concerning their contribution to the uremic syndrome. Although we observed a considerable daily loss of total protein into the CAPD dialysate, no differences were found in the removal of small proteins (MW less than 68 kilodaltons) when compared to conventional hemodialysis. Maintenance of renal function (residual diuresis greater than 100 ml/day) seems to be more potent keeping the serum concentrations of LMWP low than CAPD or hemodialysis.