CLINICAL STUDY

Renal loss of leptin in patients with nephrotic syndrome

Michael Schroth, Michael Groschl, Helmut G Dorr, Werner F Blum1, Wolfgang Rascher and Jorg Dotsch

Klinik mit Poliklinik fur Kinder und Jugendliche, Friedrich-Alexander-Universitat Erlangen-Nurnberg, Loschgestrae 15, D-91054 Erlangen, Germanyand 1Lilly Deutschland GmbH, Saalburgstrae 153, D-61350 Bad Homburg, Germany

(Correspondence should be addressed to Michael Schroth; Email: michael_schroth@yahoo.de)

Abstract

Objective: In humans, short term changes of serum leptin lead to alterations in food intake and energyexpenditure. The objective of the present study was to relate urine leptin concentrations with theextent of proteinuria in patients with nephrotic syndrome (NS). A second goal was to investigate theimpact of potential urinary leptin losses on serum leptin concentrations and body composition.Design and Methods: Seventeen patients with proteinuria were compared with twenty patients withremission of NS and ten healthy children. Leptin was measured by radioimmunoassay.Results: Urinary leptin excretion in proteinuric patients was significantly higher than in non-proteinuric patients with and without NS and in healthy controls 2:64^0:034mg=g creatinine,0:026^0:05mg=g creatinine, and 0:073^0:11mg=g creatinine respectively; P , 0:001 andP , 0:01 respectively compared with controls). Urine leptin positively correlated with urine IgGconcentration P 0:013; r2 0:36 in the proteinuric group. No difference in serum leptinvalues could be demonstrated between the three groups.Conclusions: In summary, our data demonstrate a significant leptin excretion in children with severeproteinuria. Proteinuria, however, does not lead to changes in serum leptin, suggesting that thesignificant loss of leptin is compensated for by sustained up-regulation of leptin production.

European Journal of Endocrinology 145 463468

Introduction

Leptin, a 167 amino acid polypeptide with a molecularsize of 16 kDa, is synthesized in adipose tissue (1, 2).Leptin, the product of the obese (ob) gene, plays animportant role in the regulation of appetite and foodintake in mice and humans (35). Mutations of theleptin gene or its receptor gene lead to obesity in miceand humans (1, 6). There is a positive correlationbetween serum leptin concentrations and body massindex (711). In adipose patients chronically raisedlevels of leptin do not lead to weight reduction possiblydue to the set-point changes in the hypothalamic leptinsensor. Long-term leptin treatment, however, causesmoderate weight reduction. Short-term changes inleptin serum concentrations may modulate appetiteand energy expenditure in humans during fasting andafter major surgery (12). Leptin is known to play animportant role in the regulation of weight gain in earlyinfancy and shows a relationship between its concen-tration and fetal growth (1315).

The idiopathic nephrotic syndrome (INS) is analbumin-losing nephropathy in childhood often dueto minimal change lesions of the kidneys. Severeproteinuria (urinary protein level exceeding 1 g/m2 ofbody surface area per day) leads to hypoproteinemia(hypalbuminemia, albumin in serum ,25 g/l) and

chronically to a catabolic nutritional state (16).Investigations revealed that the defect in minimalchange glomerulonephritis results mainly from a lossof charge selectivity, whereas the defect in membranousglomerulonephritis results from a loss of size selectivity(17, 18). As a consequence, the protein loss affectsbody composition. Children are often seen in an edemicstate. Blood coagulation as well as the immunologicsystem can be affected due to the loss of components ofthe coagulation system or the loss of immunoglobulins.

Since proteinuria may be associated with urinaryleptin wasting in renal protein-losing diseases, thepresent study was designed to examine the relationshipof serum and urine leptin concentration and protein-uria in patients with nephrotic syndrome (NS). Asecond goal of the study was to examine whether or notan increased urinary leptin excretion may influenceserum leptin concentrations and body composition.

Experimental subjectsThirty-seven pediatric patients with NS were studied.Patients characteristics are shown in Table 1. In 17patients (group I) proteinuria (urinary protein levelexceeding 1 g/m2 of body surface area per day) waspresent. Twenty patients were in remission (group II,

ISSN 0804-4643European Journal of Endocrinology (2001) 145 463468

q 2001 Society of the European Journal of Endocrinology Online version via http://www.eje.org

no significant proteinuria). In this study groupproteinuria did not exceed 100 mg/m2 of body surfacearea per day. The period of remission was at least 6months and was equal in all patients included. Thepatients groups were compared with 10 healthychildren (group III).

Leptin concentrations in serum and urine weremeasured by radioimmunoassay (as shown in detailbelow). After centrifugation of blood samples, whichwere all drawn between 1000 and 1200 h, serum andurine samples were kept frozen for up to 8 weeks at220 8C and were analyzed when all specimens hadbeen obtained.

In all patients, body weight and body height weremeasured to obtain the body mass index (BMI).Subscapular and triceps skinfold thickness were deter-mined with a caliper by the same trained investigator.No patient was fasting. All clinical and auxological datawere obtained during routine visits and were recordedusing standard data sheets. Details are shown inTable 1. The study was approved by the local ethicscommittee of the University of Erlangen. Informed,written consent was obtained from all parents and theolder patients.

Materials and methodsLeptin was measured by a specific radioimmunoassay.Recombinant human leptin was used for raising apolyclonal antibody in rabbits, for the production oftracer by the Chloramine T method (Lilly, BadHomburg, Germany), and for the preparation ofstandards. Tracer and samples were appropriatelydiluted in assay buffer containing 0.05 mol/lNaH2PO4, 0.1 mol/l NaCl, 0.05% NaN3 (Merck,

Darmstadt, Germany), 0.1% Triton X-100 (Roth,Heidelberg, Germany), and 0.2% (v/v) bovine serumalbumin (Serva, Heidelberg, Germany). For the dilutionof the specific antibody, rabbit immunoglobulin G (IgG)(Sigma, Deisenhofen, Germany) was added to thebuffer. For the measurement of serum samples astandard curve from 18.75 to 0.049 ng/ml wasprepared by geometrical dilution. The tracer activitywas ,20 000 c.p.m./100 ml, the final dilution of thespecific antibody was 1:50 000. In the assay, 100 ml ofeach of tracer, sample and antibody were added. Incuba-tion lasted for 24 h at room temperature. Bound and non-bound fractions were separated by a goat-anti-rabbit IgG(DSL, Sinsheim, Germany) in 4% polyethylene glycol(PEG) at a final dilution of ,1:100. After 2-hincubation at room temperature the tubes werecentrifuged (15 min, 3000 r.p.m., 4 8C). Activity ofthe pellet was counted for one min in a gamma counter.Fifty percent binding occurred at 0.9 ng/ml. Thesensitivity of undiluted samples was 0.01 ng/ml.Inter- and intra-assay coefficients of variation (%)were 8.5 and 0.8 respectively.

Leptin in urine was measured by a highly sensitivevariation of the assay. Tracer activity was 8000 c.p.m./25 ml. The antibody had a final dilution of 1:8000 in25 ml. The standard curve expanded from 1250 to9.8 pg/ml. The buffer and 2nd antibody dilutions werethe same as described above. Urine leptin concentra-tions were related to the respective urine creatinineconcentration (U leptin/U creatinine (mg/g creatinine))and are also given as absolute values (mg/l). Leptinstandard deviation scores (SDS) were calculatedaccording to Blum et al. by the following equation:leptin SDS lnleptin2 lna2 b*BMI=day where aand b are constants and BMI, the body mass index (19).

Table 1 Patient characteristics of 37 patients suffering from nephrotic syndrome and of 10 healthy children in the control group.Patient number, age, gender, BMI and pubertal stage (Tanner) as well as medication (prednisone, cyclosporine, tacrolimus, b-blockingagents) and diagnoses of all patients are listed. Age and BMI are expressed as means^standard deviation and median.

Group I Group II Group III

Number (n) 17 20 10Age (years) 8.8^3.8; 8 9.9^3.6; 10 10.5^4.2; 9.5Gender (m/f) 8/9 17/3 5/5BMI (kg/m2) 18.1^2.9; 18.1 18.3^3.6; 17.45 17.2^2.4; 17.15Tanner 1/2 10/7 11/9 4/6Patients receiving prednisone 9 1 0Patients ever (actually) treated

with cyclosporine A or tacrolimus6 (2) 10 (7) 0

Patients ever treated withcyclophosphamid

10 10 0

Patients receiving b-blockingagents

12 4 0

Diagnosis INS (8), FGGS* (1), FSGS* (3),MCGN (3), MPGN* (1),congenital nephroticsyndrome* (1)

INS (15), MCGN* (5), Healthy

*Diagnosis confirmed by histological examination.FGGS, focal global glomerulosclerosis; FSGS, focal segmental glomerulosclerosis; MCGN, minimal change glomerulonephritis; MPGN, membranoproliferative glomerulonephritis.

464 M Schroth and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 145

www.eje.org

To exclude the interference of massive proteinuriawith leptin measurements we have demonstrated thatin leptin-spiked urine samples even high IgG andalbumin levels do not create non-specific effects leadingto falsely high leptin levels in the urine. The amount ofleptin spiked into the urine was 3 ng/ml. Albuminconcentrations ranged from 3.13 to 100.0 mg/ml anddid not influence leptin concentrations; IgG was addedin concentrations from 31.25 up to 1000.0 mg/ml anddid not affect leptin concentrations. There was an 8.4%variation in the urinary leptin value over the range ofconcentrations of added albumin, and an 8.8%variation for added IgG. The measured value of leptinin the absence of albumin was 3.13 ng/ml, and in theabsence of IgG it was 3.06 ng/ml. There was nostatistical difference in measured leptin concentrationsafter adding albumin or IgG. In a further experiment,leptin at a starting concentration of 3.18 ng/ml, wasdiluted 1:2, 1:4, 1:8, 1:16 up to 1:32 (0.53 ng/ml).Leptin dilution in urine showed linearity betweencalculated leptin and measured leptin concentrationsr2 0:98; P , 0:001:

Statistical analysis

Data with Gaussian distribution were correlated bylinear regression. Parametric data were compared bytwo-tailed t-test. In the case of multiple tests, data werecompared using one-way ANOVA, and, in the case ofsignificance, post-hoc t-test. P values were correctedaccording to Bonferroni. A P value ,0.05 wasconsidered statistically significant.

ResultsUrinary excretion of leptin in proteinuric patientswith nephrotic syndrome was significantly higher2:64^0:034mg=g creatinine, absolute value 100:3^21:23mg=l than in patients without proteinuria0:026^0:047mg=g creatinine, absolute value 2:37^1:15mg=l; P , 0:001 or healthy controls 0:073^0:11mg=g creatinine, absolute value 3:83^2:06mg=l;P , 0:01 (P-values for all data related to creatinine aswell as for absolute values) (Fig. 1). Urine leptin valueswere also related to the different pubertal stages andsex (data shown in correlation to individual creatininevalues): there was no difference in urinary leptinexcretion between boys and girls, nor between Tannerstages 1 and 2 (Table 2).

Urine leptin positively correlated with urine IgGconcentration P 0:01; r2 0:36; y 47:8x44:0mg=g creatinine) and with urine albumin con-centration P 0:04; r2 0:24; y 3220x4119mg=g creatinine) in group I. When patientsshowed signs of remission (proteinuria did not exceed100 mg/m2 of body surface area for a period of at least6 months, group II), a relationship of urine leptin wasonly seen with the urine albumin fraction P 0:014;r2 0:31; y 27:89x 12:9mg=g creatinine). Therewas no correlation found between urine leptin excre-tion and urine albumin or urine IgG in controls.

Serum leptin concentrations were similar in all threegroups (Fig. 2). There was no correlation betweenserum leptin and excreted urine leptin concentrations,which were all creatinine-corrected, in group I, groupII or in the controls. In addition, there was nosignificant difference in serum leptin concentrations

Figure 1 Urine leptin concentrations in group I (patients with NSwith severe proteinuria, urinary protein level exceeding 1 g/m2 ofbody surface area per day), group II (patients with NS withoutproteinuria) and group III (control group, healthy children). Data areshown as scattergrams and as boxes and whiskers. Significantlyhigher levels in group I ***P 0:0005 were evaluated using one-way ANOVA, P values were corrected according to Bonferoni.There is no difference in urinary leptin between boys and girls andTanner stages 1 and 2, as shown in Table 2.

Table 2 Urinary leptin excretion in boys and girls, Tanner stages 1 and 2.

Urine leptin concentrationsin patients with

nephrotic syndromeand with proteinuria

(.1 g/m2/day)(mg/g creatinine)

Urine leptin concentrationsin patients with

nephrotic syndromewithout proteinuria

(mg/g creatinine)

Boys, Tanner 1 2.41^0.02 0.024^0.02Boys, Tanner 2 2.62^0.03 0.029^0.01Girls, Tanner 1 2.68^0.02 0.032^0.03Girls, Tanner 2 2.64^0.04 0.026^0.02

There is no difference between boys and girls, nor between Tanner stage 1 and 2.

Renal loss of leptin 465EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 145

www.eje.org

between male and female patients. Serum leptin SDSwere calculated in all three groups according to Blumet al. and showed no significant difference betweenmales and females (all Tanner stage 1 or 2) in any ofthe groups nor between the three groups (Fig. 3) (19).The patients medication (cyclosporine, tacrolimus,cyclophosphamid, b-blocking agents, prednisone) didnot significantly influence serum or urine leptinconcentrations when groups receiving one of themedications were compared with non-treated patients.

A positive relationship was obtained between BMIand serum leptin in nephrotic patients with proteinuriaP 0:016; r2 0:36; y 0:43x 2 5:92; in patientswith remission P , 0:0001; r2 0:82; y 0:74x 211:2 and in controls P 0:012; r2 0:41: Nosignificant correlations were found between BMI andurine leptin concentrations in any group. BMI corre-lated positively with triceps skinfold thickness in

groups I and II P 0:006; r2 0:52 and P 0:0025; r2 0:49 respectively) as well as with subscap-ular skinfold thickness in both groups P , 0:0001; r2 0:84 and P 0:001; r2 0:66 respectively). There wasa positive correlation between triceps and subscapularskinfold thickness and serum leptin in nephroticpatients with proteinuria P 0:032; r2 0:27; P 0:027; r2 0:31; in patients with remission of nephroticsyndrome P 0:021; r2 0:33; P , 0:0001; r2 0:79 and in healthy controls P , 0:0001; r2 0:41;P 0:028; r2 0:29: No significant correlationscould be demonstrated between the skinfold thicknessesand urine leptin concentrations in any group.

DiscussionPatients in groups I and II show similar auxologicaldata including BMI (Table 1). The distribution of girlsand boys and their respective pubertal stage (Tannerstage 1 and 2) was comparable. Classification ofpatients by pubertal stage revealed no significantdifference with respect to Tanner stage or sex. Thesefinding are in line with data presented by Blum et al.,who found that for Tanner stage 1 or 2 the samenormal ranges for leptin in serum can be used for boysand girls (19).

These data demonstrate a renal urinary leptin loss inprepubertal and early pubertal children suffering fromactive nephrotic syndrome with proteinuria .1 g/m2.Urinary leptin loss disappeared after remission and wasminimal in the control group. However, despite urinaryleptin loss, serum leptin levels were similar in all threegroups. Several studies have shown a relationshipbetween body mass index, methods for body fatdetermination such as subscapular and triceps skinfoldthickness measurements, and serum concentrations ofleptin (811). These observations could be confirmedin the present study. However, our data did not revealany relationship between urine leptin concentrationsand BMI or triceps and subscapular skinfold thick-nesses. Unexpectedly, serum leptin levels did not differbetween groups despite a 100-fold increment in leptinexcretion in group I. This finding strongly suggests thatthe renal loss of leptin is compensated for by asubstantial increase in leptin production. The mechan-isms that cause the up-regulation of leptin synthesisremain unclear (2025).

To answer the question as to what degree ofglomerular protein leakage allows for the loss of leptin,excretion of leptin was related to albumin and IgG inurine. Our data show that there is no significantdifference in urinary leptin excretion in proteinuricpatients with an IgG/albumin ratio lower or higherthan 1. Consequently, glomerular albumin loss isassociated with the loss of leptin. According to thesedata, we propose a glomerular loss of the 16 kDapeptide leptin in patients with proteinuria that isindependent of the molecular size.

Figure 3 Serum leptin SDS in male and female children in group I(patients with NS with severe proteinuria, urinary protein levelexceeding 1 g/m2 of body surface area per day), group II (patientswith NS without proteinuria), and group III (control group, healthychildren). Data are shown as scattergrams and as boxes andwhiskers including minimum and maximum data.

Figure 2 Serum leptin concentrations in group I (patients with NSwith severe proteinuria, urinary protein level exceeding 1 g/m2 ofbody surface area per day), group II (patients with NS withoutproteinuria) and group III (control group, healthy children). Data areshown as scattergrams and as boxes and whiskers includingminimum and maximum data.

466 M Schroth and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 145

www.eje.org

It would be of interest to establish data about leptinbinding proteins, or data about free leptin concentra-tions in urine and serum in proteinuric patients inorder to gain information about possible leptin bindingprotein excretion in urine. The clearance of leptin couldhelp to estimate the extent of binding protein loss.However, determination of leptin clearance is limited byseveral factors: 24-h urine sampling is necessary, whichrequires a urinary catheter especially in very youngchildren. Furthermore, from the literature we do nothave any data about tubular reabsorption of leptin andits binding protein.

Recently published studies revealed that leptin iscleared by the kidney, and patients with end-stage renaldisease show increased plasma leptin concentrations(2628). Leptin promotes renal growth and fibro-genesis in vivo and in vitro (29). Proliferation ofendothelial cells stimulated by leptin may contributeto a progression of renal damage and, in consequence,may promote glomerulosclerosis. Leptin was shown tostimulate the expression of transforming growth factor-b1 (TGF-b1), a major modulator of renal fibrosis, invitro and in vivo. This may activate the transcription ofextracellular matrix proteins such as collagen type IV,which may eventually lead to damage of renalglomerular endothelial cells (29). It is well knownthat renal injury leads to an increased generation ofangiotensin II, which induces an up-regulation of theTGF-b1 gene. As a consequence, tubular cells showhypertropy which, together with increased synthesis oftype IV collagen, leads ultimately to fibrogenesis andrenal scarring (30).

Thus, grossly elevated glomerular and urinary leptinlevels might accelerate the proliferation of endothelialand possibly tubular cells in renal glomerula, leading toan aggravation of (predominant) glomerulosclerosisand tubular damage. In consequence, continuousleptin excretion might be an additional mechanism inthe progression of renal failure.

This hypothesis may be tested by long-term observa-tion of children with severe proteinuria and leptinuriain relation to the progression of renal failure. Further-more, the effect of leptin on renal endothelial andtubular cell proliferation should be examined in vivoand in vitro.

In summary, our data demonstrate significant leptinexcretion in children with severe proteinuria. Elevatedurinary leptin concentrations may contribute to theproliferation of glomerular endothelial cells andglomerulosclerosis.

References1 Halaas JL, Gajiwala KS, Maffei M, Cohen SL, Chait BT,

Rabinowitz D, Lallone RL, Burley SK & Friedman JM. Weight-reducing effects of the plasma protein encoded by the obese gene.Science 1995 269 543546.

2 Kiess W, Anil M, Blum WF, Englaro P, Juul A, Attanasio A,Dotsch J & Rascher W. Serum leptin levels in children andadolescents with insulin-dependent diabetes mellitus in relationto metabolic control and body mass index. European Journal ofEndocrinology 1998 138 501509.

3 Zhang Y, Proenca R, Maffei M, Barone M, Leopold L &Friedman JM. Positional cloning of the mouse obese gene andits human homologue. Nature 1994 372 425432.

4 Bergendahl M, Evans WS, Pastor C, Patel A, Iranmanesh A &Veldhuis JD. Short-term fasting suppresses leptin and (conversely)activates disorderly growth hormone secretion in midlutealphase women a clinical research center study. Journal ofClinical Endocrinology and Metabolism 1999 84 883894.

5 Dallongeville J, Hecquet B, Lebel P, Edme JL, Le Fur C, Fruchart JC,Auwerx J & Romon M. Short term response of circulating leptinto feeding and fasting in man: influence of circadian cycle.International Journal of Obesity-Related Metabolic Disorders 199822 728733.

6 Pelleymounter MA, Cullen MJ, Baker MB, Hecht R, Winters D,Boone T & Collins F. Effects of the obese gene product on bodyweight regulation in ob/ob mice. Science 1995 269 540543.

7 Blum WF, Englaro P, Hanitsch S, Juul A, Hertel NT, Muller J et al.Plasma leptin levels in healthy children and adolescentsdependence on body mass index, body fat mass, gender, pubertalstage and testosterone. Journal of Clinical Endocrinology andMetabolism 1997 82 29042910.

8 Considine RV, Sinha MK & Heiman ML. Serum immunoreactiveleptin concentrations in normal-weight and obese humans. NewEngland Journal of Medicine 1996 334 292295.

9 Murakami T & Shima K. Cloning of rat obese cDNA and itsexpression in obese rats. Biochemical and Biophysical ResearchCommunications 1995 209 944952.

10 Funahashi T, Shimomura I & Hiraoka H. Enhanced expression ofrat obese (ob) gene in adipose tissues of ventromedial hypotha-lamus (VMH)-lesioned rats. Biochemical and Biophysical ResearchCommunications 1995 211 469475.

11 Hirose H, Saito I, Kuwai T, Nakamura K, Maruyama H &Saruta T. Serum leptin level: possible association with haema-topoiesis in adolescents, independent of body mass index andserum insulin. Clinical Science 1998 94 633636.

12 Stratton RJ, Dewit O, Crowe E, Jennings G, Villar RN & Elia M.Plasma leptin, energy intake and hunger following total hipreplacement surgery. Clinical Science 1997 93 113117.

13 Harigaya A, Onigata K, Nako Y, Nagashima K & Morikawa A.Role of serum leptin in the regulation of weight gain in earlyinfancy. Biology of the Neonate 1999 75 234238.

14 Harigaya A, Nagashima K, Nako Y & Morikawa A. Relationshipbetween concentration of serum leptin level and fetal growth.Journal of Clinical Endocrinology and Metabolism 1997 82 32813284.

15 Hytinantti T, Koistinen HA, Koivisto VA, Karonen SL &Andersson S. Changes in leptin concentration during the earlypostnatal period: adjustment to extrauterine life? PediatricResearch 1999 45 197201.

16 Orth SR & Ritz E. The nephrotic syndrome. New England Journal ofMedicine 1998 338 12021211.

17 Bridges CR, Myers BD, Brenner BM & Deen WM. Glomerularcharge alterations in human minimal change nephropathy.Kidney International 1982 22 677684.

18 Shemesh O, Ross JC, Deen WM, Grant GW & Myers BD. Nature ofthe glomerular capillary injury in human membranousglomerulopathy. Journal of Clinical Investigation 1986 77 868877.

19 Blum WF & Juul A. Reference ranges of serum leptin levelsaccording to body mass index, gender, and developmental stage.In Leptin The Voice of Adipose Tissue, edn 1, pp 318326. EdsWF Blum, W Kiess & W Rascher. Heidelberg, Leipzig: Johann ABarth, 1997.

20 Kiess W, Englaro P, Hanitsch S, Rascher W, Attanasio A &Blum WF. High leptin concentrations in serum of very obese

Renal loss of leptin 467EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 145

www.eje.org

children are further stimulated by dexamethasone. Hormone andMetabolic Research 1996 28 708710.

21 Wabitsch M, Blum WF, Muche R, Braun M, Hube F, Rascher Wet al. Contribution of androgens to the gender difference in leptinproduction in obese children and adolescents. Journal of ClinicalInvestigation 1997 100 808813.

22 Wabitsch M, Jensen PB, Blum WF, Christoffersen CT, Englaro P,Heinze E et al. Insulin and cortisol promote leptin production incultured human fat cells. Diabetes 1996 45 14351438.

23 Jockenhovel F, Blum WF, Vogel E, Englaro P, Muller-Wieland D,Reinwein D et al. Testosterone substitution normalizes elevatedserum leptin levels in hypogonadal men. Journal of ClinicalEndocrinology and Metabolism 1997 82 25102513.

24 Fritsche A, Wahl HG, Metzinger E, Renn W, Kellerer H, Haring Het al. Evidence for inhibition of leptin secretion by catecholaminesin man. Experimental and Clinical Endocrinology and Diabetes 1998106 415418.

25 Mantzoros CS, Qu D, Frederich RC, Susulic VS, Lowell BB,Maratos-Flier E et al. Activation of beta(3)adrenergic recep-tors suppresses leptin expression and mediates a leptin-independent inhibition of food intake in mice. Diabetes 199645 909914.

26 Merabet E, Dagogo-Jack S, Coyne DW, Klein S, Santiago JV,Hmiel SP et al. Increased plasma leptin concentration in

end-stage renal disease. Journal of Clinical Endocrinology andMetabolism 1997 82 847850.

27 Cumin F, Baum HP & Levens N. Leptin is cleared from thecirculation primarily by the kidney. International Journal of Obesity1996 20 11201126.

28 Daschner M, Tonshoff B, Blum WF, Englaro P, Winge AM,Schaefer F et al. Inappropriate elevation of serum leptin levels inchildren with chronic renal failure. European Study Group forNutritional Treatment of Chronic Renal Failure in Childhood.Journal of the American Society for Nephrology 1998 9 10741079.

29 Wolf G, Hamann A, Han DC, Helmchen U, Thaiss F, Ziyandeh FNet al. Leptin stimulates proliferation and TGF-beta expression inrenal glomerular endothelial cells: potential role in glomerulo-sclerosis. Kidney International 1999 56 860872.

30 Remuzzi G & Bertani T. Pathophysiology of progressive nephro-pathies. New England Journal of Medicine 1998 339 14481456.

Received 19 March 2001Accepted 5 July 2001

468 M Schroth and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 145

www.eje.org