ã

The Hematology Journal (2002) 3, 38 ± 42 2002 The European Haematology Association All rights reserved 1466 ± 4680/02 $25.00 www.nature.com/thj

Plasma thrombopoietin levels in patients with chronic renal failure Gabor E Linthorst*,1, Claudia C Folman1,2, Rudolf W van Olden3 and Albert EGKr von dem Borne1,2 1

Department of Hematology, Academic Medical Centre/Netherlands Cancer Institute, Amsterdam, The Netherlands; 2Central Laboratory of the Red Cross Blood Transfusion Service and Laboratory for Experimental and Clinical Immunology, University of Amsterdam, The Netherlands; 3Renal Unit, Academic Medical Centre, Amsterdam, The Netherlands

Introduction: Thrombopoietin (Tpo) is the most important regulator of thrombocytopoiesis. The main sites of Tpo production are the liver and the kidney produce Tpo. In the current study, the in¯uence of renal failure on overall Tpo production was evaluated. Materials and methods: Tpo levels were measured in 23 patients on hemodialysis (HD) and 16 patients on chronic ambulatory peritoneal dialysis (CAPD). Plasma glycocalicin (GC) levels and platelet counts were measured as parameters of platelet mass and platelet turnover. Results: Platelet counts were signi®cantly lower in the HD group, both before 207+986109/ l (P50.001) and after hemodialysis 202+1026109/l (P50.001) when compared to healthy controls, 293+796109/l. No signi®cant di€erence was found between platelet counts in patients on CAPD and healthy donors. Mean plasma Tpo levels of HD patients were higher both before 23+18 AU/ml (P50.0001) and after dialysis 25+26 AU/ml (P50.0001), as compared to Tpo levels in healthy controls (11+8 AU/ml). Patients on CAPD had signi®cantly higher Tpo concentrations, 29+25 AU/ml than healthy controls (P50.0001). There was no di€erence in Tpo level between the HD and CAPD group. No correlation was found between Tpo concentration and platelet count, hematocrit, creatinine or uremia levels. The GC concentration was signi®cantly higher in HD patients and CAPD patients when compared to healthy controls. There was no correlation between GC and Tpo level or platelet count. Conclusion: These results con®rm the increased platelet turnover in patients with chronic renal failure. Moreover this study shows that the kidney does not seem to play a major role in the overall Tpo production in the body. The Hematology Journal (2002) 3, 38 ± 42. DOI: 10.1038/sj/thj/6200153 Keywords:

thrombopoietin; Tpo; glycocalicin; renal failure; kidney

Introduction Thrombopoietin (Tpo) is the most important regulator of thrombocytopoiesis.1 ± 6 Tpo is produced constitutively in the liver and to a lesser extent in the kidney. In the kidney, Tpo mRNA is con®ned to the proximal tubule cells.7 Tpo is removed from the circulation by binding to the Tpo receptor c-Mpl that is present on platelets and its progenitors.8 The exact e€ect of renal failure on the production of Tpo and concurrently on the production of platelets is unknown. In contrast, the e€ect of renal failure on platelet homeostasis and platelet function has been *Correspondence: G Linthorst, Academic Medical Centre, Division of Internal Medicine F4-224, Meibergdreef 9, 1105 A2 Amsterdam, The Netherlands; Tel: +31 20 5665785; Fax: +31 20 6919743; E-mail:[email protected] Received 29 October 2001; accepted 6 December 2001

studied extensively. Circulating toxic substances (eg uremia) in patients with renal failure negatively a€ect platelet function.9 Platelet counts, although often within the normal range, are signi®cantly lower in patients on hemodialysis as compared to healthy controls.8,9 In addition, an increased platelet turnover has been reported.10 ± 12 A decreased number of circulating platelets may be caused by a decreased Tpo production. It is known that in renal failure the production of Epo is impaired, which can lead to a decreased production of red cells resulting in anemia. Epo is produced by cells of the distal tubules in the kidney.10,13 To investigate whether an impaired renal function a€ects Tpo production, two patient groups were analysed, patients on chronic intermittent hemodialysis (HD) and patients on chronic ambulatory peritoneal dialysis (CAPD). Tpo levels were measured both before and after hemodialysis. In addition, plasma glycocalicin

Tpo levels in renal failure GE Linthorst et al

39

concentrations were measured as an indicator of platelet turnover.14

Materials and methods Patients and blood collection A group of 23 patients on chronic hemodialysis participated in the study. Their mean age was 56 years (range 27 ± 79), with 13 males and 10 females. The CAPD-group consisted of 16 patients with a mean age of 53 years (range 28 ± 70 ), with 13 being male. All patients were in a stable condition and normally hydrated on clinical grounds. Of these patients, 22 required rHuEPO for correction of their anemia (15 HD patients and seven CAPD patients). None of the patients su€ered from polycystic kidney disease. All patients gave informed consent. Samples were collected in EDTA-containing tubes and were processed 2 ± 6 h after collection. In the HD group, samples were taken both before and after dialysis. The samples were centrifuged and plasma was frozen at 7208C until use. Platelet count, hemoglobin, hematocrit, creatinine and urea levels of the patients were routinely assessed. Hemodialysate was obtained from ®ve patients. The dialysate was concentrated with a Centricon-3 (Amicon, USA) concentrator, according to the manufacturors' recommendations. The concentrated dialysate was stored at 7208C until before use. This procedure was also applied to saline spiked with known concentrations of Tpo, which served as controls.

Enzyme-linked immunosorbent assay for thrombopoietin A Tpo assay that has been described in detail elsewhere was used for determination of the Tpo levels in all blood samples.15 The intra- and inter-assay variance of this test

is 5 ± 7% and 7 ± 13%, respectively. Blood samples of one patient were tested in the same plate. A pool of plasma was used as standard, with the ®rst standard dilution arbitrarily set at 100 arbitrary units (AU). Normal values as determined in a group of 193 healthy individuals (mean age 39+11, range 17 ± 69), ranged from 4 ± 32 AU/ml (2.5th ± 97.5th percentile).

Glycocalicin assay Glycocalicin (GC) concentrations were measured with a sandwich ELISA as previously described.16 The assay is based on two non cross-reactive MoAbs, one of which is used to capture glycocalicin. A biotinylated one is used for detection. The supernatant of a platelet concentrate was used as a standard and arbitrarily set at 100 AU. Normal GC plasma levels as determined in 95 healthy individuals were between 144 and 444 AU/ ml.16 To determine whether the time span between blood collection and plasma separation in¯uenced GC levels, EDTA-anticoagulated blood of four di€erent donors was processed at di€erent time points.

Data analysis Results were statistically analysed with the software application SPSS for Windows version 8.0 (SPSS INC.). The Mann-Whitney U-test was used to show di€erences between groups. Wilcoxon test was used to compare data before and after dialysis. Spearman's test was applied to show possible correlations. All levels are expressed as mean+s.d.

Results Platelet counts and thrombopoietin levels Figure 1A depicts the platelet counts and Figure 1B the plasma Tpo levels in patients with renal failure and

Figure 1 Platelet counts and plasma Tpo levels in patients with renal failure. Box plots are shown that represent the platelet counts (A) and the plasma Tpo level (B) in controls and the di€erent subgroups of patients with renal failure. Boxes represent the interquartile range containing 50% of all values. The whiskers extend to the highest and lowest value and the line across the box indicates the median. Outliers and extremes are not shown. *: Signi®cantly di€erent from control values (P50.001). The Hematology Journal

Tpo levels in renal failure GE Linthorst et al

40

control donors. Platelet counts in the HD group were signi®cantly lower both before, 207+986109/l (range 73 ± 403) and after hemodialysis 202+1026109/l (range 73 ± 411) when compared to the platelet counts of normal controls 270+626109/l (range 146 ± 510) (P50.001). No statistical di€erence was found between platelet counts of CAPD patients 293+796109/l (range 139 ± 452) and controls. The Tpo concentration before hemodialysis was 23+18 AU/ml (range 8 ± 78) and rose after hemodialysis to 25+26 AU/ml (range 10 ± 105), a small but signi®cant di€erence (P50.05). These levels were signi®cantly higher when compared to Tpo levels in healthy donors, 11+8 AU/ml (range 3 ± 60; P50.0001). Patients on CAPD also had signi®cantly higher Tpo levels 29+25 AU/ml (range 14 ± 148) than healthy donors. There was no signi®cant di€erence between plasma Tpo levels in the CAPD group and the hemodialysis patients. No correlation was present between the plasma Tpo concentration and platelet count or plasma Tpo concentration and creatinine or urea level in any group. The change in Tpo level after hemodialysis did not correlate with the change in hematocrit. The change in platelet count as a result of hemodialysis did correlate (P50.01, rs=0.75) with the rise in hematocrit in the HD group, as did the increment in hemoglobin level (P50,001, rs=0.95), which served as control (data not shown). Patients on rHuEpo treatment had no signi®cant di€erence in platelet count or Tpo level when compared to patients without rHuEpo treatment.

Figure 2 Plasma GC levels in patients with renal failure. Box plots representing plasma GC levels in controls and in the di€erent subgroups are shown. With regard to the box plots, the same legend as in Figure 1 applies.

Glycocalicin data Results are shown in Figure 2. GC levels were 1088+670 AU/ml before and 1020+543 AU/ml after hemodialysis, and 972+197 AU/ml in the CAPD group, which is signi®cantly higher in all cases compared to healthy controls (293+75 AU/ml, P50.001). There was no correlation between Tpo and GC level. GC and platelet count showed a signi®cant correlation in the HD group after dialysis (P50.001, rs=0.79) but in none of the other groups. Plasma GC concentrations did not correlate with creatinine or urea level in the dialysis groups, nor was there a correlation between the change in GC level and hematocrit due to the hemodialysis. To exclude the increment in plasma GC concentration in the patients was due to delayed processing of samples, control samples were processed at di€erent time points. As shown in Figure 3, no signi®cant di€erences were present up to 7 h of storage at room temperature.

Dialysate data No Tpo was detected in non-concentrated or concentrated dialysate samples. Tpo added to these samples as a control was recovered completely (data not shown). The Hematology Journal

Figure 3 GC levels in plasma samples that were prepared at di€erent time points. After blood collection, plasma was isolated every hour for up to 7 h. Mean and s.d. are shown for four di€erent donors.

Discussion Tpo is constitutively produced by the liver and the kidney. Circulating platelets and bone-marrow megakaryocytes are responsible for removal of Tpo from the circulation, thus regulating the Tpo concentration in blood.17 Although the kidney is one of the Tpo-producing organs, the current study shows that loss of renal function did not result in a decreased circulating concentration of Tpo. On the contrary, in patients on HD and CAPD, even those with normal platelet counts, a small but signi®cant increase in plasma Tpo concentration was found. Recently, similar results were found by Ando et al.8 who also found slightly elevated serum Tpo levels in patients on hemodialysis and/or CAPD and by Stockelberg et al.18 who showed normal to increased Tpo levels in patients on HD for end-stage renal failure. In contrast, Altun et al.19 reported a signi®cant reduction of the Tpo-level in 31 HD-patients compared to 12 healthy controls. However, when

Tpo levels in renal failure GE Linthorst et al

41

analysing the individual data, the Tpo levels of the patients seemed to be within the Tpo range of the healthy controls. In our study, most patients (22/39) required rHuEpo to maintain normal red cell levels. Thus, loss of parenchym was to such an extent that the kidney could produce no sucient amounts of Epo. It is likely that the loss of kidney parenchym also a€ected Tpo production. An indication for the role of the kidney in Tpo production was recently obtained from a study in which wild-type mice were transplanted with a liver from Tpo knockout mice. These animals showed a 60% decrease in circulating platelets.20 Measured with a less sensitive assay, Tpo concentrations in these mice were below the detection limit of the assay. This suggests that the liver is responsible for most of the Tpo production. Apparently in mice, the loss of Tpo production by the liver can not be overcome by other Tpo-producing organs. The normal or increased plasma Tpo levels in our study con®rms the small contribution of the kidneys towards the total Tpo production. Platelet turnover seems increased in the studied patients, since plasma GC concentrations were found to be increased. GC has previously been shown to be a marker for platelet turnover.14 An arti®cial increment of GC concentration due to delayed processing of samples as reported earlier14 is unlikely because control samples were stable for up to 24 h. The ®nding of elevated GC levels as a re¯ection of increased platelet turnover is in accordance with a study from Himmelfarb et al.,21 who showed that the percentage of reticulated platelets is increased in dialysis patients. Reticulated platelets are young

platelets, and an increment in the percentage of reticulated platelets re¯ects enhanced thrombopoiesis. In contrast, Ando et al. reported a decrease in the absolute amount of reticulated platelets in patients with renal failure.8 In the current study no di€erence in GC and Tpo level in patients on HD as compared to patients on CAPD was observed. The increased Tpo concentration might be responsible for the enhanced platelet production. The source of the slight increase in Tpo concentration is unknown. A decreased removal by platelets is unlikely since no correlation between Tpo levels and platelet counts was present. Since the kidney is not involved in the clearance of Tpo, which is supported by the ®nding that Tpo levels did not correlate with creatinine or urea levels, the loss of renal function is not responsible for the elevated Tpo levels. Platelet activation occurs in patients on hemodialysis.21 We previously showed that platelet activation during coagulation results in Tpo release.15,22 Indeed, Tpo levels were slightly but signi®cantly increased after hemodialysis, indicating that platelet activation might occur, leading to Tpo release and a subsequent increase in circulating Tpo. However, we were unable to show a di€erence in GC level before and after dialysis. Also, Tpo and GC levels did not di€er between HD and CAPD group. Apparently, GC levels re¯ect platelet turnover rather than platelet activation. In conclusion, in contrast to the e€ect renal failure has on the circulating amount of Epo, the Tpo concentration does not seem to be a€ected in a major way, i.e. there is no Tpo de®ciency. Thus, compared to the liver, the kidney does not seem to play a major role in the constitutive production of Tpo.

References 1 Bartley TD, Bogenberger J, Hunt P, Li YS, Lu HS, Martin F et al. Identi®cation and cloning of a megakaryocyte growth and development factor that is a ligand for the cytokine receptor Mpl. Cell 1994; 77: 1117. 2 de Sauvage FJ, Hass PE, Spencer SD, Malloy BE, Gurney AL, Spencer SA et al. Stimulation of megakaryocytopoiesis and thrombopoiesis by the c-mpl ligand. Nature 1994; 369: 533. 3 Kaushansky K. Thrombopoietin. The primary regulator of platelet production. Blood 1995; 86: 419. 4 Kaushansky K, Lok S, Holly RD, Broudy VC, Lin N, Bailey MC et al. Promotion of megakaryocyte progenitor expansion and di€erentiation by the c-Mpl ligand thrombopoietin. Nature 1994; 369: 568. 5 Lok S, Kaushansky K, Holly RD, Kuijper JL, LoftonDay CE, Oort PJ et al. Cloning and expression of murine thrombopoietin cDNA and stimulation of platelet production in vivo. Nature 1994; 369: 565. 6 Wendling F, Maraskovsky E, Debili N, Florindo C, Teepe M, Titeux M et al. c-Mpl ligand is a humoral regulator of megakaryocytopoiesis. Nature 1994; 369: 571.

7 Sungaran R, Markovic B, Chong BH. Localization and regulation of thrombopoietin mRNA expression in human kidney, liver, bone marrow and spleen using in situ hybridisation. Blood 1997; 89: 101. 8 Ando M, Iwamoto Y, Suda A, Tsuchiya K, Nihei H. New insights into the thrombopoietic status of patients on dialysis through the evaluation of megakaryocytopoiesis in bone marrow and of endogenous thrombopoietin levels. Blood 2001; 97: 915. 9 Panicucci F, Sagripanti A, Vispi M, Pinori E, Lecchini L, Barsotti G, Giovannetti S. Comprehensive study of haemostasis in nephrotic syndrome. Nephron 1983; 33: 9. 10 George CR, Slichter SJ, Quadracci LJ, Striker GE, Harker LA. A kinetic evaluation of hemostasis in renal disease. N Engl J Med 1974; 291: 1111. 11 Sagripanti A, Barsotti G. Bleeding and thrombosis in chronic uremia [editorial]. Nephron 1997; 75: 125. 12 Tanaka H, Umimoto K, Izumi N, Nishimoto K, Maekawa T, Kishimoto T, Maekawa M. Platelet life span in uraemia. Proceedings of the European Dialysis and Transplant Associaton-European Renal Association 1985; 21: 306.

The Hematology Journal

Tpo levels in renal failure GE Linthorst et al

42

13 Shanks JH, Hill CM, Lappin TR, Maxwell AP. Localization of erythropoietin gene expression in proximal renal tubular cells detected by digoxigenin-labelled oligonucleotide probes. J Pathol 1996; 179: 283. 14 Beer JH, Buchi L, Steiner B. Glycocalicin:a new assay-the normal plasma levels and its potential usefulness in selected diseases. Blood 1994; 3: 691. 15 Folman CC, von dem Borne AEGKr, Rensink IHJAM, Gerritsen W, van der Schoot CE, de Haas M, Aarden L. Sensitive measurements of thrombopoietin by a monoclonal antibody based sandwich enzyme-linked immunosorbent assay. Thrombosis Haemostasis 1997; 78: 1262. 16 Porcelijn L, Folman CC, Bossers B, Huiskes E, Overbeeke MAM, van der Schoot CE et al. The diagnostic value of thrombopoietin levels measurements in thrombocytopenia. Thrombosis Haemostasis 1998; 79: 1101. 17 Fielder PJ, Hass P, Nagel M, Stefanich E, Widmer R, Bennett GL et al. Human platelets as a model for the binding and degradation of thrombopoietin. Blood 1997; 89: 2782.

The Hematology Journal

18 Stockelberg D, Andersson P, Bjornsson E, Bjork S, Wadenvik H. Plasma thrombopoietin levels in liver cirrhosis and kidney failure. J Intern Med 1999; 246: 471. 19 Altun, B, Arici M, Haznedaroglu IC, Usalan C, Erdem Y, Yasavul U et al. Serum thrombopoietin levels in haemodialysis patients: involvement of arteriovenous ®stula. Nephrol Dialysis Transplant 1999; 14: 2173. 20 Qian S, Fu F, Li W. Primary role of the liver in thrombopoietin production shown by tissue-speci®c knockout. Blood 1998; 92: 2189. 21 Himmelfarb J, Holbrook D, McMonagle E, Ault K. Increased reticulated platelets in dialysis patients. Kidney Int 1997; 51: 834. 22 Folman CC, Linthorst G, van Mourik JA, van Willigen G, de Jonge E, Levi M et al. Platelets release thrombopoietin (Tpo) upon activation: another regulatory loop in thrombocytopoiesis. Thrombosis Haemostasis 2000; 83: 923.