Insulin-like growth factors (IGFs) and IGF binding proteins, serum acid-labile subunit and growth hormone binding protein in

Kidney International, Vol. 52 (1997), pp. 82 81 Insulin-like growth factors (IGFs) and IGF binding proteins, serum acid-labile subunit and growth hormone binding protein in nephrotic children DIETER HAFFNER, BURKIIARD TöNSHOFF, WERNER F. BLUM, MARC VIcIRs, THOMAS SIEBLER, MIcHL J. Cioir, ROBERT C. BAXTER, and OTTO MEHLS Division of Pediatric Nephrology, University Children's Hospital Heidelberg, University Children's Hospital GieJ3en, Germany; Genentech Inc., South San Francisco, USA; and Kolling Institute of Medical Research, Royal North Shore Hospital, St. Leonards, Australia Insulin-like growth factors (IGFs) and IGF binding proteins, serum acid-labile subunit and growth hormone binding protein in nephrotic children. We hypothesized that the increased glomerular permeability to serum proteins in the nephrotic syndrome might lead to alterations of the somatotropic hormone axis, thereby contributing to growth failure and catabolism in the nephrotic state. The insulin-like growth factors (IGF)-I and -II and the IGF binding proteins (IGFBP)-1, -2 and -3 were analyzed in serum and urine of 21 children with the nephrotic syndrome and normal glomerular filtration rate. Mean age-related serum IGF-I levels by RIA (.53.34 SD) were slightly, but significantly (P <.5) decreased compared with the reference population, whereas mean age-related serum IGF-II levels (.68.21 SD) were slightly, but significantly (P <.5) increased. The urinary excretion rate of both peptides was enhanced fivefold. By RIA, mean age-related serum IGFBP-1 (2.5.19 SD) and, even more pronounced, IGFBP-2 (5.97.65 SD) were clearly elevated despite a 12-fold and 2-fold increase of the respective urinary excretion rate. There was a tight and specific correlation between age-related serum IGFBP-2 levels and the degree of the nephrotic syndrome, as estimated by serum albumin levels (r =.78, P <.1). Serum immunoreactive IGFBP-3 levels were also elevated (1.79.33 SD) in nephrotic serum, due to an increase of low-molecular weight IGFBP-3 fragments. By FPLC analysis, there was a decrease of the 15 kda IGFBP ternary complex in nephrotic serum, which in the presence of normal concentrations of the acid-labile subunit by RIA appears to be due to a reduction of intact IGFBP-3. Serum levels of the high-affinity GH binding protein that presumably reflects GH receptor status in tissues were normal. In summary, total serum IGFs in children with the nephrotic syndrome are normal, but the binding of IGFs to IGFBPs in the circulation is altered with a shift from the 15 kda IGFBP complex to an excess of low molecular weight IGFBPs. Because increased unsaturated high-affinity IGFBPs in ncphrotic serum have the ability to inhibit IOF action on target tissues by competing with the type 1 IOF receptor for TGF binding, this alteration is likely to contribute to growth failure and tissue catabolism in the nephrotic state. Prior to the introduction of glucocorticoid therapy, growth retardation and reduced tissue anabolism was frequently observed Key words: insulin-like growth factor, growth hormone, protein, children and growth, serum proteins. Received for publication February 19, 1997 and in revised form April 16, 1997 Accepted for publication April 18, 1997 1997 by the International Society of Nephrology in children with the nephrotic syndrome, and this remains a problem in glucocorticoid-resistant nephrotic syndrome [1]. Multiple pathogenic mechanisms are involved such as protein-calorie malnutrition secondary to poor appetite, malabsorption due to edema of the gastrointestinal tract and, in particular, protein losses in the urine due to increased glomerular permeability to plasma proteins. This alteration is likely to lead to secondary alterations of the somatotropic hormone axis that might contribute to growth failure and reduced tissue anabolism in the nephrotic state. The insulin-like growth factors (IGFs) are a unique family of polypeptide hormones with strong mitogenic and anabolic properties. IGFs in biological fluids are complexed to high-affinity IGF binding proteins (IGFBPs), six of which have been identified. The IGFBPs regulate the action of IGFs by prolonging their half-lives, directing their distribution, and modulating the interaction with their respective receptors [21. Each of the IGFBPs exhibits differential regulatory mechanisms involving developmental, hormonal, and nutritional status. In the circulation, over 99% of IGFs are tightly bound to IGFBPs [3]. IGFBP-3 is the predominant circulating IGFBP in postnatal life. In contrast to the other IGFBPs, it has the unique property to associate with an acid-labile subunit (ALS) (Mr 85 kda) after binding either IGF-I or IGF-II. The subsequent constituted ternary complex is a highmolecular weight complex of approximately 15 kda [4]. Its function is thought to act as a reservoir for the IGFs, preventing rapid changes of free IGF levels. IGFBP-3 serum levels are positively regulated by growth hormone (GH) secretory status [5]. It inhibits IOF action under some [61, but not all experimental conditions [7 91. IGFBP-2 (32 kda) is the second most abundant TGFBP in the circulation and it binds to IGF-1I with a greater affinity than IGF-I. IGFBP-i and IGFBP-2 inhibit IGF action under many [1, Ii], but not all experimental conditions, apparently by competing with IGF receptors for IGF binding. IGFBP-1 is a 28 kda nonglycosylated protein that is primarily produced by liver and female reproductive tissue [121. Insulin is the principal suppressive regulator of hepatic IGFBP-1 production [12]. In the present study, we analyzed the IGF/IGFBP system in serum and urine of children with active nephrotic syndrome. Serum levels of ALS, the third component of the 15 kda IGFBP complex, were also examined. Moreover, we measured the levels 82

Diagnosis Age years Height SDS BMI SDS HaJJiwr et al: IGFs and IGFBPs in the nephrotic syndrome 83 Table 1. Patient characteristics Serum albumin giliter Proteinuria g/m2/day Proteinuria Creatinine Duration of selective! clearance proteinuria nonselective mi/mini days 1.73 m2 MCNS 7.6 1.1.3.6.4.6 21.2 3.2 5.7.7 1 3 7/ 135 8.7 (N =7) FSGS 8.1 1.7 1.1.4.3.4 21.9 1.9 1.8 1.4 959 176 4/3 123 13.5 (N =7) MPGN 11.3 1.7.7.4.6.7 29.5 1.9 4.1 1. 969 463 /4 114 9. (N =4) HSP 1.1 2.4.2.3.6.7 26.6 3.4 3.3.5 63 37 /3 13 11.3 (N =3) Total 9..8.6.3.4.3 23.7 1.5 6.9.9 592 135 11/1 126 5.7 (N = 21) Data are mean SCM. Abbreviations are: MCNS, minimal change nephrotic syndrome; FSGS, focal segmental glomerulosclerosis; MPGN, membranoproliferative glomerulonephritis; HSP, Henoch-Schonlein purpura glomerulonephritis; SDS, standard deviation score; BMI, body mass index. a Significant difference between patients with the nephrotic syndrome and the reference population of the high-affinity GH binding protein (GHBP) in nephrotic serum. GHBP (Mr 6 kda) is thought to reflect GH receptor status in GH target tissues, because it is derived from the extracellular domain of the GH receptor by proteolytic cleavage [131. Our study is the first extensive analysis of the distal components of the somatotropic hormone axis in children with the nephrotic syndrome. METHODS Patients Twenty-one children (11 boys/lo girls) with active nephrotic syndrome were analyzed. Patient characteristics are listed in Table 1. Height and weight were measured in all subjects with standardized equipment and techniques. To estimate the nutritional status of the patients, body mass index (BMI) was calculated using the formula: weight (kg)/height2 (m2) (Quetelet index). To obtain age-independent estimates of body size and mass, height and BMI (after logarithmic transformation to obtain normally distributed data) were converted to SD score (SDS) values related to age- and gender-specific means and SDS of European reference populations [14, 151. The mean relative height was slightly below the normal mean and the mean relative BMI as an index of the nutritional status was normal (Table 1). Seven children were studied during multiple relapses of the nephrotic syndrome, resulting in a total number of 3 observations. All children had gross proteinuria and low serum albumin levels (Table 1). Exclusion criteria were increased serum creatinine concentration for age, a decreased creatinine clearance below < 8 ml/min/l.73 m2, or treatment with glucocorticoids, angiotensin converting enzyme inhibitors or clonidin up to one month prior to this study. Some patients were treated with frusemide (N = 3) or hydrochlorothiazide (N = 4). The time of onset of proteinuria was defined as the first day of documented proteinuria by a dip stick test (Albustif ). Response to glucocorticoid therapy was assessed according to the criteria of the International Study Group of Kidney Disease in Children [16]. Blood samples were obtained in the morning in the fasting state. Serum and 24-hour urine samples were stored at - 2 C until analysis. For comparison, urine samples of 12 age- and gendermatched healthy children (age 9.9.7 years) were analyzed. Assays IGF-I was measured by a novel IGFBP-blocked RIA utilizing a specific high-affinity antiserum (gift of Drs. Breier and Gluckman, Auckland, New Zealand) [17]. Details of the assay have been published [18]. In brief, serum was diluted 1:15 with an acidic phosphate buffer, to dissociate IGFs from IGFBPs. Concomitantly with re-neutralization by adding the first antibody solution, a large excess of IGF-II (25 ngltube) was added to block IGFBPs. After addition of IGF-I tracer and incubation at 4 C for two days, bound and free '251-IGF-I were separated by a second antibody method [19]. There was an excellent parallelism between serial dilutions of serum and recombinant IGF-I standard (Pharmacia Upjohn, Stockholm, Sweden). Half-maximal displacement of tracer occurred at.1 ngltube and sensitivity was.81 ng. The intra- and interassay coefficients of variation were 3.1 and 8.1%, respectively. For measurement of urinary IGF-I, 3 il untreated urine samples were acidified by adding 3.tl.5 mol/liter phosphoric acid. One hundred microliters of these acidified samples were then directly assayed for IGF-I by the IGFBP-blocked RIA. Serum IGF-II was measured after acid-ethanol extraction using a highly specific polyclonal antibody and blocking residual IGFBP in the extract by an excess of IGF-I as described previously [19]. Urinary IGF-II was directly measured in acidified urine as described for IGF-I. IGFBP was blocked by 25 ng IGF-I per tube. IGFBP-1, IGFBP-2 and IGFBP-3 were measured by specific RIAs that have been described in detail previously [2 22]. The IGFBP-3 RIA recognizes both intact IGFBP-3 (Mr 45 kda) as well as low-molecular weight IGFBP-3 fragments [23]. Serum samples were diluted 1:6, and unprocessed urine samples were diluted 1:5 with an assay buffer before measurement. All samples were assayed in duplicate. Both urinary IGFs and IGFBPs were expressed as excretion rates normalized by body surface area. For determination of the molecular size of IGFBP-3 and the binding fraction of IGF-I in serum, 5.tl of serum or concentrated urine samples (about 1:1 using an Amicon ultrafiltration cell at a molecular weight cut-off of 2) were chromatographed at room temperature over a Superose 12 column using a fast protein liquid chromatography (FPLC) system (Pharmacia, Freiburg, Germany). Fractions of.5 ml were collected at a rate of 1 mi/mm. The elution was performed with.5 mol/liter sodium phosphate

84 Haffner et a!: JGFs and IGFBPs in the nephrotic syndrome 6 A 12 B 5 1 4 8 3 2 1 2 4 6 81121416182 Age, years 6 4 2 2 4 6 81121416182 Age, years Fig. 1. Serum IGF-I (A) and IGF-II (B) levels in children with active nephrotic syndrome related to the age-dependent normal range. For IGF-I the normal range is given by the.1, 5th, 5th, and 95th percentiles; for IGF-II it is given by the 5th, 5th, and 95th percentiles. Closed symbols indicate patients with serum albumin levels below 25 g/liter, and open symbols above 25 g/liter. buffer, ph 7.4, containing.1 mol/liter NaCI,.1% BSA (wt:vol),.2% NaN5 (wt:vol). The fractions were assayed for IGFBP-3 by RIA after 1:1 dilution with assay buffer. For determination of serum protease activity towards recombinant human (rh) IGFBP-3, sera of patients and controls were diluted 1:1 with phosphate-buffered saline, ph 7.4. Five microliters of resuspended 1251 [IGFBP-3] tracer (Diagnostic System Laboratories, Houston, TX, USA), corresponding to 4 cpm/ sl, were mixed with 5 jsl diluted serum from patients or controls and incubated at 37 C for six hours, according to the recommendations of the manufacturer. Ten microliters of Laemmli sample buffer [24] were added to each sample. The samples were heated at 94 C for five minutes and analyzed by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE, 12.5%) under non-reducing conditions. The Kaleidoscope Prestained Standard markers (Bio-Rad Lab., Hercules, CA, USA) were used as markers. Gels were vacuum-dried immediately at 8 C and exposed to X-ray films (Kodak, Tokyo, Japan) at 2 C. ALS in serum was determined by a specific RIA, as described previously [25]. The intra- and interassay coefficients of variation were 3.4% and 6.5%, respectively. Serum GHBP was measured by a ligand-mediated immunofunctional assay, as described previously [26]. The intra- and interassay coefficients of variation were 7% and 11%, respectively. Serum albumin was analyzed by an autoanalyzer method. Urinary protein concentration was determined by the Coomassie method [27]. The selectivity of proteinuria was assessed by analysis of the molecular mass of urinary proteins by SDS-PAGE [28]. Selective proteinuria was defined by a Mr of urinary proteins 6 kda. Statistical analysis Reference values of serum levels for IGF-1, IGF-I1, IGFBP-3 [29], IGFBP-1, IGFBP-2 [3], ALS [25] and GI-IBP [31] in healthy children of various ages have been published previously. The distribution of these reference data are log normal. Measured values were, therefore, transformed to their logarithms before calculating SD scores (SDS) to obtain age-independent values for comparison. Data are given as mean SEM, if not indicated otherwise. Data were examined for normal and non-gaussian distribution by the Shapiro-Wilk test. For comparison between two normally distributed groups the unpaired Student's t-test (two-tailed) was used. For comparison of more than two normally distributed groups, one-way analysis of variance followed by pairwise multiple comparisons (Student-Newman-Keuls method) was taken. For nonnormally distributed data, the nonparametric Wilcoxon signed rank test or the Kruskal-WaIlis test were used. Correlations between variables were assessed by univariate linear regression analysis. P <.5 was accepted as statistically significant. RESULTS IGF and IGFBP serum levels Individual serum IGF-l levels in patients with the nephrotic syndrome related to the age-dependent normal range are depicted in Figure 1 A. IGF-I levels clustered in the low-normal range; mean age-related serum IGF-I levels (.53.34 SD) were slightly, but significantly reduced compared with the reference population (Table 2). In contrast, individual serum IGF-II levels clustered in the high-normal range (Fig. 1B); mean agerelated serum IGF-II levels (.68.21 SD) were slightly, but significantly elevated compared with the reference population (Table 2). Individual serum IGFBP-1, -2, and -3 levels by RIA in children with the nephrotic syndrome related to the age-dependent normal range are shown in Figure 2. In the majority of patients, serum IGFBP levels were clearly elevated above the upper normal range. The increase of serum IGFBP-2 levels (5.97.65 SD) was particularly striking, while IGFBP-1 and IGFBP-3 levels were increased to a comparable extent (Table 2). Analysis of the molecular size of serum IGFBP-3 by FPLC revealed an increase of low-molecular weight IGFBP-3 fragments in nephrotic patients and a decrease of the 15 kda ternary complex (Fig. 3). We sought to determine whether the increase of low molecular weight IGFBP-3 fragments in nephrotic serum is due to increased proteolytic activity towards IGFBP-3. Incubation of nephrotic serum with rhigfbp-3 did not result in a rapid degradation of the binding protein (Fig. 4). In contrast, serum from late pregnancy that was analyzed concomitantly as a positive control degraded rhigfbp-3 in our assay (Fig. 4), consistent with previous observations [32]. In controls, most of the immunoreactive serum IGF-I was bound to the high-molecular weight IGFBP ternary complex (Fig. 5). In patients with the nephrotic syndrome, however, IGF-I was preferably bound to low molecular weight IGFBPs.

Haffner et al: IGFs and IGFBPs in the nephrotic syndrome 85 Table 2. Serum levels of IGFs, IGFBPs, ALS and GHBP in children with active nephrotic syndrome IGF-1 IGF-1l IGFBP 1 IGFBP-2 IGFBP-3 Parameter (N = 3) (N = 27) (AT = 25) (N 25) (N = 3) (N 22) Absolute serum 169 19 672 37.17.3 1.92.33 4.44.24 26.6 1.8 levels jrg/liter.tg/liter mg/liter mg/liter mg/liter mg/liter Age-related.5.34.68.21 2.5.19 5.97.65 1.79.33.11.4 serum levels So SD SD SD SD SD (SD score) Statistics P <.5 P <.5 P <.1 P <.1 P <.1 NS Significant difference between patients with the nephrotie syndrome and the reference population. NS = not significant. GHBP (N = 21) 344 62 pmol/liter.41.37 SD NS.7 A B C 4. - 8. E.4 S.. 'S 2.5 ci 2. 1.5 U- o 1..5...s S. - I I I I I 24 6 81121416182 24 6 81121416182 24 6 81121416182 ' 7 3 U) U- Age, years Age, years Age, years Fig. 2. Serum IGFBP-1 (A), IGFBP-2 (B), and IGFBP-3 (C) levels in children with active nephrotic syndrome related to the respective age-dependent normal range. The normal range is given by the 5th, 5th, and 95th percentiles. Closed symbols indicate patients with serum albumin levels below 25 g/liter and open symbols above 25 g/liter. 16 14 ci) 12 C,) U) U- E ci) (I) 1 8 6 4 2 2 4 6 8 1 12 14 16 18 2 22 24 26 28 3 Fraction number Fig. 3. Serum IGFBP-3 mol wt distribution by FPLC analysis in normal serum (N = 5) (straight line) and in three individual children with active nephrotic syndrome. Nearly identical results were obtained for the elution profiles of serum IGFBP-3 in the five controls tested (r =.998) and, therefore, only mean values are shown. In normal serum, immunoreactive IGFBP-3 eluted mainly at a mol wt of 15 kda, corresponding to the ternary IGF-IGFBP-3-ALS complex. In nephrotic serum, there was a decrease of the IGFBP ternary complex and an increase of the binary IGF-1GFBP-3 complex (4 to 5 kda) and of low molecular weight IGFBP-3 fragments. Next, we sought to determine whether these alterations of serum IGFs/IGFBPs are related to the degree of the nephrotic syndrome. There was a tight inverse correlation of age-related IGFBP-2 levels with serum albumin levels (r =.78, P <.1; Fig. 6) and a positive correlation of serum IGFBP-2 with the degree of proteinuria (r =.43, P <.5). This correlation was specific, because there was no relationship between serum albumin levels or the degree of proteinuria and the other components

86 Haffner et a!: IGFs and IGFBPs in the nephrotic syndrome 44 32 rr' 4rc 17.4 Ctr LP NS NS NS NS NS Fig. 4. A representative autoradiogram of serum IGFBP-3 protease activity by 1251 IGFBP-3 degradation assay. By incubation of rhigfbp-3 with serum of late pregnancy (LP), three different rhigfbp-3 fragments were observed ( 32, 21, and 17 kda). In nephrotic serum (NS), there was no clear-cut increase of protease activity towards rhigfbp-3 compared with control serum (ctr). 1 9 8 7 5 U) (I) 3 2 2 4 6 8 1 12 14 16 18 2 22 24 26 28 3 Fraction number Fig. 5. Elution profiles of immunoreactive IGF- I in normal serum (N = 5) (straight line) and in three individual children with active nephrotic syndrome. Nearly identical results were obtained for the elution profiles of serum IGF-I in the five controls tested (r =.98), and therefore, only mean values are shown. In control serum, most of the immunoreactive serum IGF-I was bound to the 15 kda IGFBP ternary complex. In nephrotic serum, however, IGF-I was preferably bound to low molecular weight IGFBPs. of the IGF/IGFBP-axis. Neither the duration of the nephrotic syndrome nor the selectivity of proteinuria, as estimated by SDS PAGE, correlated with IGF or IGFBP serum levels. However, patients with FSGS tended to have lower age-related 1GF-I levels ( 1.47 SDS) than patients with other types of nephrotic syndrome. Apart from that, IGF and JGFBP serum levels were similar in the different histologic types of nephrotic syndrome. Urinary excretion of IGF and IGFBP Urinary excretion rates of IGF-I and IGF-II were increased fivefold (Table 3). The ratio of IGF-1I over IGF-I in control urine (14:1) was comparable to the ratio in patients with the nephrotic syndrome (12:1). Also the urinary excretion rates of IGFBP-1 (12-fold), IGFBP-2 (2-fold), and IGFBP-3 (2-fold) were significantly increased in nephrotic patients (Table 3). In urine of patients and controls, only immunoreactive IGFBP-3 of less than 6 kda were detected by FPLC analysis with a peak at 2 kda (Fig. 7). Urinary 1GF-1I (r =.55, P <.5) and IGFBP-3 excretion (r =.4, P <.5) correlated with the degree of proteinuria. There was no correlation between serum levels of IGFs or IGFBPs and their respective urinary excretion rates. ALS and GHBP in serum Age-related serum levels of ALS in children with active nephrotic syndrome were normal (Fig. 8 and Table 2). There was a tight correlation between serum ALS and IGFBP-3 levels (r =.76, P <.1), as observed under normal conditions [25], despite the fact that some of the IGFBP-3 in nephrotic serum was present as low molecular weight IGFBP-3 fragments. The concentration of the high affinity GHBP in nephrotic serum was within the normal range (Table 2).

Hafjher et al: IGFs and IGFBPs in the nephrotic syndrome 87 (1) C (1) cj U- (!3 E ci) Cl) 14 12 1 8 4 2- - 2 5 1 15 2 25 3 35 Serum albumin, g/iiter Fig. 6. Age-related IGFBP-2 levels (SDS) as a function of serum albumin levels (N = 23). There was a tight inverse correlation (r =.78, P <. 1). DISCUSSION The present study is the first extensive analysis of the distal components of the somatotropic hormone axis in patients with the nephrotic syndrome. We observed that mean serum IGF-I levels in nephrotic serum were slightly decreased. However, there was a considerable variability from clearly decreased to increased serum levels. Mean serum IGF-II levels were slightly increased. Two previous studies have examined serum IGF levels in children with the nephrotic syndrome. Garin, Grant and Silverstein [331 reported a 5% reduction of serum IGF-I and a 6% of serum IGF-1I in children with the nephrotic syndrome compared with matched controls. Lee, Park and Kim [34] observed a 6% reduction of serum IGF-I, but did not measure serum IGF-II. The reason for the somewhat different results in our study may involve differences in the patient population regarding the degree or the duration of the nephrotic syndrome, the selectivity of the proteinuria or a combination of these factors. The main stimulatory regulators of serum IGF-I levels in humans are the GH secretory status and the nutritional supply. Both calorie and protein malnutrition reduce circulating IGF-I levels [351 A better nutritional status of the patients in our study who were closely supervised by a dietitian might explain the higher circulating IGF-l levels than in the two previous reports. Serum GH levels are less likely to play a role, because children with the nephrotic syndrome have normal basal and peak GH levels to insulin provocation [36] and a normal spontaneous GH secretory pattern [37]. Also, differences in the assay technique may play a role. In particular in conditions where serum IGFBP concentrations are high, such as chronic renal failure, it is important to remove IGFBPs from the sample prior to assay, because residual IGFBP in the assay mixture can interfere with antibody binding [18, 38]. We have previously shown that the most reliable estimates of IGF levels in biological fluids can be obtained by blocking residual IGFBPs in the assay mixture with excess of IGF-I or IGF-II, respectively [18]. In the setting of experimental nephrotic syndrome, similarly inconsistent results about serum IGF-I levels were reported. In one study in puromycin-induced nephrosis in the rat [39], serum IGF-I levels were normal compared to pair-fed controls, but hepatic IGF-I mrna levels were decreased by 5%. In contrast, a second study in adriamycin-induced nephrosis in rats [4] demonstrated a reduction of serum IGF-I levels by 4%, but hepatic IGF-I mrna was unchanged. Hence, there is clearly a considerable biological variability in serum IGF-I levels in the setting of clinical and experimental nephrotic syndrome that is most likely due to differences in the production rate, transcapillaiy movement and urinary losses of the peptide. Our study is the first that also quantified the urinary losses of IGFs in patients with the nephrotic syndrome. We report a fivefold increase of urinary IGF-I and IGF-Il excretion compared with matched controls that is most likely due to increased glomerular permeability to these peptides. The comparable ratio of IGF-II over IGF-I in nephrotic and control urine might indicate that a potentially altered renal production of IGFs does not contribute significantly to the urinary losses of IGFs in the nephrotic syndrome. The urinary losses of IGFs occur in the presence of a 12-fold increase of IGFBP-1 and a twofold increase of IGFBP-2 and -3 excretion in the urine. Whether the increased glomerular ultrafiltration of IGFs in the nephrotic syndrome exerts any biological effect in the kidney is not known. Feld and Hirschberg [41] have suggested that the exposure of apical tubule receptors to supraphysiological IGF-I in nephrotic urine may contribute to increased phosphate retention and increase the expression of extracellular matrix genes, thereby contributing to the interstitial fibrosis that is sometimes observed in chronic nephrotic glomerulopathies. IGF serum levels must be interpreted in relation to their respective serum IGFBP levels. We report for the first time a marked increase of serum IGFBP-1 levels in children with the nephrotic syndrome despite a 12-fold increase of urinary IGFBP-1 excretion. Because circulating IGFBP-l is mainly produced in liver tissue, the constellation of increased IGFBP-1 serum levels despite enhanced urinary losses of the protein indicates increased hepatic production. This hypothesis remains to be tested in experimental studies. We also report for the first time by RIA a pronounced increase of IGFBP-2 in nephrotic serum despite a twofold increased urinary excretion rate of the protein. Lee et al [34] had previously assumed increased IGFBP-2 levels in nephrotic children by use of Western ligand blots. The mechanism for increased IGFBP-2 levels in the nephrotic syndrome is most likely increased hcpatic production, because there was a fourfold increase of liver IGFBP-2 mrna in experimental nephrosis [4]. The specific stimulus for increased hepatic production of IGFBP-2 and, most likely, also of IGFBP-1 in the nephrotic syndrome remains to be elucidated. Calorie and protein malnutrition is not likely to play a role in our study, as indicated above. It is noteworthy that in humans prolonged fasting over nine days is needed to cause an increase in serum IGFBP-2 levels [42]. In our study, IGFBP-2 levels were closely related to the degree of the nephrotic syndrome, as estimated by serum albumin levels. However, the specific metabolic link between these two parameters remains to be elucidated. We report an increase of serum IGFBP-3 by RIA in the

88 Hafflieret al: IGEs and IGFBPs in the nephrotic syndrome Table 3. Urinary excretion rates (U) of IGFs and IGFBPs in children with active ncphrotic syndrome and age-matched controls U-IGF-I ng/m2/day (N = 16) U-IGF-11 ng/m2/day (N = 16) U-IGFBP-1 pg/m2/day (N = 1) U-IGFBP-2 pglm2lday (N = 1) U-IGFBP-2 pg/m2/day (N = 25) Nephrotic syndrome 497 (84 37) 64 (177 13,5) 44.1 (21.9 69.4) 3.7 (8.1-182) 68.8 (13. 282) Controls (N = 12) 83 (25 124) 1157 (612 1662) 3.6 (.3 11.8) 15.4 (6.6 36.5) 28.2 (16. 1 43.8) Statistics P <.1 P <.1 P <.1 P <.1 P <.5 Data are median (range). Significant difference between patients with active nephrotic syndrome and controls 8 7 i5 C') (!3 > C 24681121416182222426283 Fraction number Fig. 7. Representative elution profiles of immunoreactive IGFBP.3 in normal urine (N = 5) (straight line) and in two children with active nephrotic syndrome by FPLC. Nearly identical results were obtained for the elution profiles of urinary IGFBP-3 in the five controls tested (r =.97), and therefore, only mean values are shown. Both in control and nephrotic urine, immunoreactive IGFBP-3 eluted at a mol wt below 6 kda with a peak at 2 kda, representing intact IGFBP-3 (4 to 5 kda) and low molecular weight IGFBP-3 fragments. nephrotic syndrome in the presence of a twofold increased urinary excretion rate. FPLC analysis showed that the increase of immunoreactive IGFBP-3 in nephrotie syndrome is due to increased levels of low molecular weight IGFBP-3 fragments, whereas the intact 15 kda IGFBP-3 ternary complex was clearly reduced and the binding of IGF-I to low molecular weight IGFBPs was clearly increased. Lee et a! [34] reported a decrease of intact IGFBP-3 by Western ligand blotting in children with the nephrotic syndrome. The reason for decreased intact IGFBP-3 in nephrotie serum is not entirely clear. We did not detect increased protease activity towards IGFBP-3 in nephrotie serum; however, we cannot exclude that the proteolytic activity was too low in serum to be detected by our assay. In the experimental nephrotic syndrome, Hirsehberg and Kaysen [4] reported decreased intact IGFBP-3 by Western ligand blot and Western immunoblot and an increased protease activity towards rhigfbp-3 in nephrotie urine, but not in serum. We demonstrated by FPLC analysis that nephrotic urine contains IGFBP-3 with a molecular mass between 45 kda (intact IGFBP-3) and 15 kda, most likely representing low molecular weight IGFBP-3 fragments. The difference between this and the observation of Garin, Grant and Silverstein [33], who showed in gel filtration and binding studies with iodinated IGFs the presence of the 15 kda complex in nephrotie urine, might be due to differences in the permseleetivity of the proteinuria. ALS is a glycoprotein of 85 kda that combines with the GH-dependent IGFBP-3 and IGF-I or IGF-Il to form the 15 kda IGFBP complex [4]. Two thirds of the total ALS in serum is present in the -15 kda complex, and one third is uncomplexed [25]. We observed normal serum levels of ALS in the nephrotic syndrome, indicating that the decrease of the 15 kda IGFBP complex in nephrotic serum is not due to a deficiency of ALS. We report normal levels of the high-affinity GHBP in nephrotic serum. Because GHBP in humans is thought to reflect GH receptor status in GH target tissues [13], this finding indicates normal tissue density of GH receptors in patients with the nephrotic syndrome. In contrast, Thabet et a! [39] suggested by semiquantitative estimations of hepatic GH receptor mrna levels that hepatic GH receptor density is low in experimental purimycin-induced nephrosis. Our data can be summarized as follows: Serum IGF-l levels in the setting of clinical nephrotic syndrome are slightly decreased, whereas IGF-II levels are slightly increased, resulting in normal total serum IGF levels. However, the binding of IGFs to IGFBPs in the circulation is altered. There is a clear-cut increase of serum IGFBP-1 and a pronounced increase of serum IGFBP-2, whereas the high molecular weight IGFBP ternary complex is decreased in nephrotic serum. In the presence of normal ALS concentrations, this alteration is most likely due to an increased proteolytic degradation of intact IGFBP-3. The shift of bound IGFs from the

HafJner et at: IGFs and IGFBPs in the nephrotic syndrome 89 55 5 45 4 35 3 E 25 -J <2 15 1 5 Age, years Fig. 8. Serum ALS levels in children with active nephrotic syndrome related to the age-dependent normal range (mean 2 su). Closed symbols indicate patients with serum albumin levels below 25 g/liter, open symbols above 25 g/liter. 15 kda complex that is restricted to the circulation to low molecular weight IGFBPs should on the one hand facilitate the transepithelial transport of IGFs to IGF target tissues. On the other hand, the excess of unsaturated high-affinity IGFBPs in nephrotic serum has the ability to inhibit IGF action on target tissues by competing with the type I IGF receptor for IOF binding. Hence, it is likely that the altered IGF/IGFBP system in nephrotic serum results in decreased IGF action in target tissues, thereby contributing to catabolism and growth failure in the nephrotic state. Reprint requests to Burkhard TonshoJf M.D., Division of Pediatric Nephrology, University Children s Hospital, Im Neuenheimer Feld 15, 6912 Heidelberg, Germany. E-mail: BurkhardToenshoff(.ájkrzmail.kn.uni-heidelherg.de REFERENCES 2 4 6 8 1 12 14 16 18 2 1. PADILLA R, BREM AS: Linear growth of children with nephrotic serum: Effect of alkylating agents. Pediatrics 84:495 499, 1989 2. CLEMMONS DR: Insulin-like growth factor binding proteins: Roles in regulating IGF physiology. J Dcv Physiol 15:15 I 1, 1991 3. FRYSTYK J, SKJAERBAEK C, DINESEN B, ORSKOV H: Free insulin-like growth factors (IGF-I and TOF-lI) in human serum. FEBS Lett 348:185 191, 1994 4. BAXTER RC, MARTIN JL, BENIAC VA: High molecular weight insulinlike growth factor binding protein complex. J Biol Chem 264:11843 11848, 1989 5. BLUM WF, AIBER]ssoN-WIKI AND K, ROSBERG 5, RANKE MB: Serum levels of insulin-like growth factor I (IGF-E) and IGF binding protein 3 reflect spontaneous growth hormone secretion. J Clin Endocrinol Metab 76:161-1616, 1993 6. BICSAK TA, SHIMONAKA M, MALKOWSKI M, LING M: Insulin-like growth factor-binding protein (IGF-BP) inhibition of granulosa cell function: Effect on cyclic adenosine 3',5'-monophosphate, deoxyribonucleic acid synthesis, and comparison with the effect of an IGF-I antibody. Endocrinology 126:2184 2189, 199 7. Dc MELLOW JS, BAXTER RC: Growth hormone-dependent insulinlike growth factor (IGF) binding protein both inhibits and potentiates IGF-I-stimulated DNA synthesis in human skin fibroblasts. Biochem Biophys Res Commun 156:199 24, 1988 8. CONOVER CA, RNK M, LOMBANA F, POWELL DR: Structural and biological characterization of bovine insulin-like growth factor binding protein-3. Endocrinology 127:2795 83, 199 9. BLUM WF, JENNE EW, REPPIN F, KIErZMANN K, RANKE MB, BIERICI-I JR: Insulin-like growth factor I (IGF-I)-binding protein complex is a better mitogen than free IGF-l. Endocrinology 125:766 772, 1989 It). R/TVOS, RANTA T, JALKANEN J, SUIKKARI AM, VOUTII.AINEN R, BHN H, RUTANEN EM: Insulin-like growth factor (IGF) binding protein from human decidua inhibits the binding and biological action of IGF-l in cultured ehoriocarcinoma cells. Endocrinology 122:215 2157, 1988 11. BURCH WM, CORREA J, SHIVELY JE, POWELL DR: The 25-kilodalton insulin-like growth factor (IGF)-binding protein inhibits both basal and IGF-I-mediated growth of chick embryo pelvic cartilages in vitro. J Clin Endocrinol Metab 7:173 18, 199 12. LEE PDK, CONOVER CA, POWELL DR: Regulation and function of insulin-like growth factor-binding protein-i. Proc Soc Exp Biol Med 24:4 29, 1993 13. LEUNG DW, SPENCER SA, CACHIANES G, HAMMONDS RG, COLLINs C, HENTZEL WJ, BARNARD R, WATERS MJ, WOOD WI: Growth hormone receptor and serum binding protein: Purification, cloning and expression. Nature 33:537 543, 1987 14. PRADER A, LARGO RH, MOLINARI L, ISSLER C: Physical growth of Swiss children from birth to 2 years of age. Helv Paediatr Ac/a 52(Suppl):I 125, 1988 15. ROLLAND-CACHERA MF, CIE TJ, SEMPE M, TICHET J, ROSSIGNOL C, CHARRAUD A: Body mass index variations: Centiles from birth to 87 years. EurJ Clin Nutr 45:13 21, 1991 16. INTERNATIONAL STUDY OF KIDNEY DISEASE IN CHILDREN: The primary nephrotic syndrome in children. Identification of patients with minimal change nephrotic syndrome from initial response to prednisone. J Pediatr 98:56 1 564, 1981 17. BREIER BH, GALLAFIER BW, GI.UCKMAN PD: Radioimmunoassay for insulin-like growth factor-i: Solutions to some potential problems and pitfalls. J Endocrinol 128:347 357, 1991 18. BLUM WF, BREIER BH: Radioimmunoassay for IGFs and IGFBPs. Grow/h Reg 4:11 3, 1994 19. BLUM WF, RANKE MB, BIERICH JR: A specific radioimniunoassay for insulin-like growth factor II: The interference of IGF binding proteins can be blocked by excess IGF-I. Acta Endoc,-inol (Copenh) 118:374 38, 1988 2. BREIER BH, MILSOM SR, BLUM WF, SCIIwANDER J, GALLAHER BW, GLUCKMANN PD: Insulin-like growth factors and their binding proteins in plasma and milk after growth hormone-stimulated galactopoiesis in normally lactating women. Acta Endocrinol (Copenh) 129:427 435, 1993 21. BLUM WF, HORN N, KRATZSCH J, JORGENSEN JO, JUUL A, TEALE D, MOIINIKE K, RANKE MB: Clinical studies of IGFBP-2 by radioimmunoassay. Growth Reg 3:1 14, 1993 22. BLUM WF, RANKE MB, KIETZMANN K, GAUGGEL E, ZEISEL HJ, BIERICH JR: A specific radioimmunoassay for the growth hormone (GH)-dependent somatomedin-hinding protein: Its use for diagnosis of GH deficiency.] Clin Endocrinol Metab 7:1292 1298, 199 23. BLUM WF, RANKE MB, KIETZMANN K, TONSHOFF B, MEHI.S : Growth hormone resistance and inhibition of somatomedin activity by excess of insulin-like growth factor binding protein in uremia. Pediatr Nephrol 5:539 544, 1991 24. LAEMMLI UK: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:68 685, 1971) 25. BAXTER RC: Circulating levels and molecular distribution of the acid-labile (a) subunit of the high molecular weight insulin-like growth factor-binding protein complex. J Clin Endocrinol Metab 7:1347 1353, 199 26. CARI.SSON LMS, ROWlAND AM, CLARK RG, GESUNDHEIT N, WONG WLT: Ligand mediated immunofunctional assay (I.IFA) for quantification of growth hormone binding protein in human blood. J Clio Endocrinol Metab 73:1216 1223, 1991 27. BRADFORD MM: A rapid and sensitive method for the quantitation of

81 Haffner Ct a!: IGFs and IGFBPs in the nephrotic syndrome microgram quantities of protein utilizing the principle of protein dye binding. Anal Biochem 72:248 254, 1976 28. BoEsKlN WH, KOPF K, SCIIOLLMEYER P: Differentiation of proteinuric diseases by discelectrophoretic molecular weight analysis of urinary proteins. Clin Nephrol 4:311 318, 1973 29. BLUM WF: Insulin-like growth factors and their binding proteins, in Functional Endocrinologic Diagnostics in Children and Adolescence, edited by RANKE MB, Mannheim, J & J Verlag, 1992, pp 12 117 3. TONSHOFF B, BLUM WF, WINGEN AM, MEHLs, EUROPEAN STUDY GROUP FOR NUTRITIONAL TREATMENT OF CHRONIC RENAL FAILURE IN CHILDFIOOD: Serum insulin-like growth factors (IGFs) and IGF binding proteins 1, 2, and 3 in children with chronic renal failure: Relationship to height and glomerular filtration rate. J C/in Endocrinol Metab 8:2684 2691, 1995 31. CARLSSON LMS, ACnE KM, COMPTON PG, VITANGCOL RV, MERIMEE TJ, NATIONAL COOPERATIVE GROWTH STUDY: Reduced concentration of serum growth hormone-binding protein in children with idiopathic short stature. J C/in Endocrinol Metab 78:1325 133, 1994 32. HOSSENLOPP P, SERGOVIA B, LASSARRE C, ROGHANI M, BREDON M, BINUx M: Evidence of enzymatic degradation of insulin-like growth factor binding proteins (IGFBPs) in the 15 K complex during pregnancy. J Cl/n Endocrinol Metab 7 1:797 85, 199 33. GARIN EH, GRANT MB, SILVERSTEIN JH: Insulin-like growth factors in patients with active nephrotic syndrome. AJDC 143:865 867, 1989 34. LEE D-Y, PARK SK, KIM J-S: Insulin-like growth factor-i (IGF-l) and IGF-binding proteins in children with nephrotic syndrome. J C/in Endocrinol Metab 81:1856 186, 1996 35. THISSEN JP, KETELSLEGERS JM, UNDERWOOD LE: Nutritional regulation of the insulin-like growth factors. Endocrine Rev 15:8 11, 1994 36. SADEGHI-NEJAD A, SENIOR B: Adrenal function, growth and insulin in patients treated with corticoids on alternate days. Pediatrics 43:277 283, 1969 37. REES L, GREENE SA, ADLARD P, JONES J, HAYCOCK GB, RIGDEN SPA, PREECE M, CHANTLER C: Growth and endocrine function in steroid sensitive nephrotic syndrome. Arch Dis Child 63:484 49, 1988 38. POWELL DR, ROSENFELD R, BAKER B, LIE F, HINTZ R: Serum somatomedin levels in adults with chronic renal failure: The importance of measuring insulin-like growth factor I (IGF 1) and IGF 2 in acid-chromatographed uremic serum. J Clin Endocrinol Metab 63: 1186 1192, 1986 39. THABET MA, CHALLA A, CHAN W, Lw F, HINTZ RL, CHAN JCM: Insulin-like growth factor and growth hormone receptor in nephrotic rats. Am J Physiol 2 66:E12 E16, 1994 4. HIRSCHBERG R, KAYSEN GA: Insulin-like growth factor I and its binding proteins in the experimental nephrotic syndrome. Endocrinology 136:1565 1571, 1995 41. FELD SM, HIRSCHBERG R: Insulin-like growth factor-i and insulin-like growth factor-binding proteins in the nephrotic syndrome. Pediatr Nephro/ 1:355 358, 1996 42. CLEMMONS DR, SNYDER DK, BUSBY WH: Variables controlling the secretion of insulin-like growth factor binding protein-2 in normal human subjects. J C/in Endocrinol Metab 73:727 733, 1991