| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Articles |
Department of Pediatrics and Physiology, State University of New York at Stony Brook (B.T., D.Z., L.C.M., F.J.K.), Stony Brook, New York 11794; Department of Pediatrics, Baylor College of Medicine Houston (D.R.P., S.K.D., M.E.C.), Texas 77030; Diabetes Branch, National Institute of Diabetes and Digestive and Kidney Disease, National Institutes of Health (H.M.D.), Bethesda, Maryland 20892; University Children’s Hospital, Gießen 35385 (W.F.B.), Germany; and Kolling Institute of Medical Research (R.C.B.), St. Leonards, NSW 2065, Australia
Address all correspondence and requests for reprints to: Burkhard Tönshoff, Division of Pediatric Nephrology, University Children’s Hospital, Im Neuenheimer Feld 150, 69120 Heidelberg, Germany. E-mail: Burkhard-Toenshoff{at}krzmail.krz.uni-heidelberg.de
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
48 kDa) and low molecular IGFBP-3 fragments were not
significantly different among the three groups. By Northern blot
analysis, hepatic IGFBP-1 mRNA levels in uremia were 2-fold higher than
in controls. IGFBP-2 mRNA abundance in liver tissue was increased
4-fold, whereas in kidney there was a significant reduction of IGFBP-2
mRNA (30% of control). IGFBP-4 mRNA was increased by 50% in
kidney but not in liver. Plasma insulin and corticosterone levels were
not different among the groups. Our study shows that hepatic IGF-I gene
expression was specifically reduced in uremia, partially as the
consequence of a reduced hepatic GH receptor gene expression. One of
the mechanisms contributing to increased IGFBP levels in uremia is
increased hepatic gene expression of IGFBP-1 and IGFBP-2. The imbalance
between reduced hepatic IGF-I production and increased hepatic IGFBP-1
and -2 production is likely to play a pathogenic role for catabolism
and growth failure in CRF. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
IGF-I partially mediates the effect of GH on longitudinal growth by both an endocrine and autocrine/paracrine mode of action (4). IGF-I in biological fluids is complexed to high-affinity IGF binding proteins (IGFBPs), six of which have been identified. The IGFBPs regulate the actions of IGF-I by prolonging its half-life, directing its distribution, and modulating the interaction of IGF-I with its receptor (5). Each of the IGFBPs exhibits differential regulatory mechanisms involving developmental, hormonal, and nutritional status.
Under clinical conditions, circulating IGF-I levels are normal in preterminal CRF and slightly decreased in end-stage renal disease (6, 7, 8, 9). In contrast, IGFBP-1, IGFBP-2, and low molecular weight IGFBP-3 fragments are increased in CRF serum in relation to the degree of renal dysfunction (6, 7, 8, 9, 10, 11, 12). The biological action of IGF-I is mediated via the type I IGF receptor. Because IGFBPs bind IGFs with affinities similar or higher to those of the type 1 IGF receptor, the excess of unsaturated high-affinity IGFBPs in CRF serum has the ability to inhibit IGF action on target tissues by competing with the type 1 IGF receptor for IGF binding (6, 7, 13). Indeed, increased IGFBP levels in CRF have been identified as inhibitors of IGF bioactivity both in vitro (7) and in vivo (8).
Little is known about the production rates of IGF-I and IGFBPs in CRF. In a previous investigation (14) we suggested that the constellation of increased IGFBP over normal IGFs indicates a reduced IGF-I secretion rate in CRF, because under normal conditions an increased IGF-binding capacity would be expected to be immediately saturated by IGFs produced in the liver. This hypothesis was therefore tested in the present study by analysis of hepatic IGF-I gene expression in a rat model of experimental uremia. Liver tissue was examined because this tissue is the major source of circulating IGF-I (15). For comparison, IGF-I gene expression was examined in other tissues where it exerts mainly an autocrine/paracrine mode of action. Because hepatic IGF-I gene expression is partially regulated by the GH receptor status, the abundance of GH receptor messenger RNA (mRNA) levels in liver tissue was measured concomitantly.
Our previous analysis of plasma IGFBP levels in the setting of clinical CRF had also suggested that an increased IGFBP-2 production rate might contribute to elevated IGFBP levels in CRF plasma (8). To test this hypothesis we analyzed IGFBP-1, -2, -3, and -4 plasma levels in experimental uremia and the gene expression of IGFBP-1, -2, and -4 in liver and kidney because plasma levels of these IGFBPs were found to be elevated. We studied IGFBP gene expression in liver because this organ is a major site of production of these peptides (16, 17). Kidney was examined to see whether partial nephrectomy affected IGFBP gene expression in the same way in this tissue as in liver. To determine the relation of any changes to nutritional status, which is known to affect plasma levels and hepatic gene expression of IGF-I and IGFBPs (18), uremic animals were compared both with pair-fed and ad libitum-fed controls. Our study represents the first detailed analysis of hepatic IGF-I and IGFBP-1, -2, and -4 gene expression under the controlled conditions of experimental uremia.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). The animals were subjected to two-stage subtotal
nephrectomy as described previously (19). Seven days after 2/3
nephrectomy of the left kidney, the contralateral kidney was removed
resulting in a 5/6 nephrectomy. Control animals were sham operated
(renal decapsulation). Uremic animals and ad libitum-fed
controls had free access to food and water, whereas one of the control
groups was pair-fed to the uremic group as described previously (19).
Pair-feeding started after the first stage operation. The duration of
uremia was 14 days. These studies were conducted in accord with the NIH
guidelines for the care and use of laboratory animals.
|
Biochemical measurements
Plasma creatinine, urea, electrolytes, phosphate, glucose, uric
acid, and total protein levels were measured by an autoanalyzer,
adapted for rat plasma (Anilytics, Gaithersburg, MD). Plasma insulin
(DPC, Los Angeles, CA) and corticosterone (ICN, Costa Mesa, CA) were
measured by specific RIAs. Plasma IGF-I was measured after acid-ethanol
extraction as described previously (20), utilizing the high-affinity
antibody of Breier, Gallaher, and Gluckman (21). Residual IGFBP in the
extract was blocked by adding 25 ng recombinant IGF-II per tube. In
addition, plasma IGF-I levels were measured by a novel RIA specific for
rat IGF-I, following the recommendations of the manufacturer
(Diagnostic Systems Laboratories, Webster, TX). This assay uses a goat
antirat IGF-I antibody that does not cross-react with human IGF-I (22).
Plasma IGFBP-1 was measured with a RIA specific for rat IGFBP-1
(23).
RNA preparation
Total RNA was prepared from tissues of individual rats using a
guanidine isothiocyante-LiCl precipitation technique (24). RNA
concentrations were calculated on the basis of absorbance at 260 nm.
The integrity of the RNA and the accuracy of the quantitation were
confirmed by comparing the ethidium bromide stain of the 28S and 18S
ribosomal RNA bands after size-separating 15 µg of total RNA by
agarose gel electrophoresis (1.25% agarose-2.2 M
formaldehyde gels).
Antisense RNA probes
IGF-I. As a hybridization probe for the solution
hybridization/RNase protection assays to determine IGF-I mRNA levels, a
riboprobe vector was used, which was kindly provided by D. LeRoith
(NIH, Bethesda, MD). Briefly, this construct contains a 376-bp IGF-I
complementary DNA (cDNA) containing sequences encoding part of the A
domain, the entire D and E domains (including the 52-bp insert present
in some IGF-I mRNAs), and part of the 3'-untranslated region (25).
Antisense RNAs generated from this linearized construction permit the
simultaneous detection of IGF-I mRNAs without (IGF-Ia) and with
(IGF-Ib) the 52-base insert (25). After ligation into the riboprobe
vector pGEM-3 cut with BamHI and EcoRI this
construct was linearized with HindIII and gel purified
before synthesis of antisense RNA with T7 polymerase. Hybridization of
IGF-I mRNAs containing the insert to the antisense RNA generates a band
376 bases in length (IGF-Ib transcript), whereas hybridization of IGF-I
mRNAs lacking the insert to the antisense RNA results in bands 224
bases and 100 bases in length (IGF-Ia transcripts) (25).
GH receptor/GHBP. The rat GH receptor probe was transcribed from a 900-bp BglII fragment of a rat GH receptor cDNA corresponding to the region encoding the signal peptide, the extracellular domain, and the transmembrane domain, and a portion of the intracellular domain (26). This template was linearized with BamHI and transcribed with T7 RNA polymerase to generate a 445-base antisense RNA probe. This probe produces two protected bands when hybridized to total liver RNA, a 439-base band corresponding to the GH receptor mRNA and a 298-base band corresponding to the alternatively spliced mRNA which, in the rat, encodes the GH-binding protein (GHBP) (27).
Solution hybridization RNase protection assay
Antisense RNA probes were labeled with
[32P]uridine triphosphate, using the Riboprobe Gemini
System (Promega, Madison, WI). Hybridizations were performed as
described previously (28). Briefly, 20-µg aliquots of total RNA were
hybridized with 200,000 cpm probe at 45 C for 16 h. As an internal
control for potential losses during the processing of the samples,
equal amounts (8000 cpm) of a 32P end-labeled
AviII-digested pGEM-4Z DNA (1.7 and 1.0 kilobase in size)
were added to each RNA sample before ethanol precipitation. As a
reference RNA for accurate quantitation of RNA levels, a commercially
available pT7 RNA 18S antisense control template was used, which
contains a 82-bp fragment of a highly conserved region of the 18S
ribosomal gene (Ambion, Austin, TX). After RNase digestion, protected
probe bands were denatured and resolved on 8% polyacrylamide-8
M urea gels. Autoradiography was performed at -80 C using
Kodak X-AR film (Eastman Kodak, Rochester, NY) and two intensifying
screens. Autoradiographic densities of these protected bands were
quantified using a LKB laser densitometer in the two-dimensional scan
mode to obtain a densitometric value for the entire autoradiographic
band. The arbitrary densitometric value obtained for each IGF-Ia
transcript (224-bp band) was corrected with the respective intensity of
the pGEM-4Z and 18S RNA band as internal standards and expressed as
percentage of ad libitum-fed controls. For GH receptor/GHBP
mRNA levels, the protected bands were quantified with a PhosphorImager
apparatus using ImageQuant software (Molecular Dynamics, Sunnyvale, CA)
as previously described (29).
IGFBP Northern blot analysis
Individual samples of 20 µg total RNA per lane were
size-fractionated on a 1% agarose gel as described previously (30).
The RNA was transferred to a GeneScreen nylon membrane (New England
Nuclear, Boston, MA), cross-linked to the membrane by exposure to
120,000 µJ UV irradiation (254 nM), and incubated at 80 C
for 2 h Prehybridization and hybridization of the RNA to
32P-labeled cDNAs were performed according to standard
procedures (31). The cDNA probes used included full-length probes for
rat IGFBP-1 (32) and rat IGFBP-2 (33) kindly provided by Matthew
Rechler and Guck Ooi (NIH, Bethesda, MD), a 444-bp probe from the
coding region of rat IGFBP-4 (34), kindly provided by Shunichi
Shimasaki (Whittier Institute, La Jolla, CA), and a 324-bp probe from
murine glyceraldelyde phosphate dehydrogenase (GAPDH) (Ambion, Austin,
TX) as an internal standard. These probes were labeled with
[32P]uridine triphosphate by random priming using the
Pharmacia Ready to Go kit (Pharmacia, Uppsala, Sweden). Filters were
subsequently exposed to x-ray films at -70 C (Kodak X-omat AR) with
intensifying screens. Signal intensities were quantified using a
Betascope model 603 blot analyzer (Betagen, Waltham, MA). Values were
expressed as percentage of GAPDH expression.
Western immunoblot analysis
Western immunoblot analysis was performed after electrophoresis
of 2 µl serum on 12.5% SDS-PAGE and transferred overnight onto
nitrocellulose. Nitrocellulose sheets were blocked with 5% nonfat dry
milk and incubated for 1 h with rabbit antibovine IGFBP-2
antiserum and rabbit antihuman IGFBP-4 antibody, which also detect rat
IGFBP-2 and -4 specifically (Upstate Biotechnology, Lake Placid, NY;
1:2000 dilution). Subsequently, nitrocellulose sheets were washed
vigorously in 0.5% Tween 20, incubated with goat antirabbit
immunoglobulin G conjugated with horseradish peroxidase (Sigma Chemical
Co., St. Louis, MO) for 1 h, washed again, and exposed to enhanced
chemiluminescence reagents (Amersham Corp., Arlington Heights, IL) for
1 min and to x-ray film for 10 sec to 5 min. All procedures were
carried out at room temperature. Specific bands were quantified by
laser densitometry (Hoefer Scientific Instruments, San Francisco, CA).
To obtain semiquantitative data for statistical analysis, the density
of the IGFBP-2 and IGFBP-4 band in each lane was expressed as
percentage of the density of the respective band of ad
libitum-fed control plasma processed in parallel on each gel. Each
gel was run with three plasma samples from uremic animals, three
pair-fed control samples, and three ad libitum-fed
control samples.
The Western immunoblot for determination of plasma IGFBP-3 was performed as follows. Three microliters rat plasma were separated under reducing conditions (100 mM dithiothreitol) by SDS-PAGE using a 7.5–15% linear gradient gel. Proteins were electroblotted onto nitrocellulose filters (Schleicher & Schuell, Keene, NH) using a Trans-Blot SD apparatus (Bio-Rad Labs., Richmond, CA) and then blocked for 3 h at 4 C in 3% BSA. Filters were incubated at 4 C overnight in a 1:250 final dilution of rat IGFBP-3 antibody (kindly provided by Shunichi Shimazaki, Whittier Institute,) and then incubated for 2 h with antirabbit biotinylated IgG (Vector Labs., Burlingame, CA). Blots were incubated with Vectastain ABC-AP Reagent, and antigen-antibody interactions were then detected by the alkaline phosphatase method using Alkaline Phosphatase Substrate Kit IV reagents (BCIP/NBT kit) following the instructions of the manufacturer (Vector Labs.). The sizes of the IGFBP-3 forms were different from usual because dithiothreitol was used to enhance the ability of the proteins to be recognized by this antibody, which was originally raised using a synthetic IGFBP-3 peptide.
[125I]IGF ligand blotting
[125I]IGF ligand blotting was performed as
described previously (12). Briefly, plasma was electrophoresed on a
12% SDS polyacrylamide gel under nonreducing conditions. Separated
proteins were transferred to nitrocellulose (0.45-µm pore size) using
a Biotrans semidry electrophoretic transfer unit (Gelman Sciences, Ann
Arbor, MI) and Towbin buffer (25 mM Tris/192 mM
glycine/20% methanol). Nitrocellulose membranes were blocked in
succession with 3% Nonidet P-40 (Sigma), 1% BSA (Sigma), and 0.1%
Tween-20 (Sigma) in Tris-saline, pH 7.4, and then incubated overnight
with approximately 2 x 106 cpm
[125I]IGF-II. After washing and drying, filters were
exposed to film for 48 h.
Statistical analysis
Values are given as mean ± SEM. Data were
examined for normal and non-Gaussian distribution by the Shapiro-Wilk
test. For comparison among normally distributed groups, one-way ANOVA
followed by pairwise multiple comparisons (Student-Newman-Keuls method)
was used. For nonnormally distributed data, the nonparametric
Kruskal-Wallis test followed by an all pairwise multiple comparison
(Dunn’s method) was used. Correlations between variables were assessed
using univariate linear regression analysis. P < 0.05
was accepted as statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Uremia in this experiment was relatively mild and was not
associated with metabolic acidosis, as estimated by whole blood pH and
bicarbonate levels (Table 1). Uremic animals showed a significant
growth retardation because the overall weight gain in uremic animals
during the observation period of 14 days was significantly lower than
in ad libitum-fed controls (Table 1). Spontaneous food
intake in uremic animals was approximately 76% of that of ad
libitum-fed controls. Accordingly, the food conversion ratio
(i.e. weight gain per food intake) was
significantly lower in uremia than in ad libitum-fed
controls (Table 1). Plasma potassium levels were slightly higher in
uremia and plasma insulin and corticosterone levels were not different
among the groups (Table 1). Also plasma sodium, chloride, calcium,
phosphate, glucose, uric acid, and total protein levels were not
different (results not shown).
Figure 1 shows a representative solution
hybridization/RNase protection assay of rat liver RNA from uremic rats
and pair-fed and ad libitum-fed controls.
Hybridization of total liver RNA with the 376-base antisense RNA probe
generated bands of 100 bases and 224 bases in length corresponding to
the IGF-Ia mRNA and low abundance bands of 376 bases in length
corresponding to IGF-Ib mRNA. These results are in accordance with the
first description of this RNA probe (25). In uremic animals, there was
a marked reduction of the IGF-Ia and IGF-Ib transcripts compared with
pair-fed controls, and to an even higher extent compared with ad
libitum-fed controls. The mean hepatic IGF-I mRNA abundance
(IGF-Ia transcript) was reduced by 70% compared with ad
libitum-fed controls and by 50% compared with pair-fed
controls (Fig. 2). This difference indicates that the
reduction of hepatic IGF-I expression in uremia is only partially due
to the slightly reduced food intake, but also is caused by alterations
that are specific for the uremic milieu. Reduced IGF-I expression in
uremia was clearly organ-specific and mainly found in liver
(significant vs. pair-fed controls) and lung and muscle
tissue (significant vs. ad libitum-fed
controls); there was no change in kidney and heart (Fig. 2). Plasma
IGF-I levels in CRF were not different from ad
libitum-fed controls despite the reduction of hepatic IGF-I
mRNA abundance. In pair-fed rats, however, the reduction of hepatic
IGF-I mRNA abundance led to a corresponding decrease of plasma IGF-I
peptide levels (Fig. 3A). The use of a RIA specific for
rat IGF-I gave a similar pattern of plasma IGF-I levels; however,
absolute levels were approximately 40% higher than measured by the
human-specific RIA (Fig. 3A). When individual plasma levels were
compared, there was a tight linear correlation between the two assay
results (Fig. 3B).
|
|
|
). The
reduced food intake of pair-fed animals did not significantly influence
hepatic GH receptor mRNA abundance. Compared with the relative
reduction of hepatic IGF-I gene expression in uremia, the decrease of
hepatic GH receptor gene expression in uremic animals was less
pronounced.
|
) and IGFBP-4
levels (Fig. 5B) were clearly elevated compared with both control
groups. Quantification of the autoradiographic bands with laser
densitometry showed a 7-fold increase of plasma IGFBP-2 (Fig. 6B) and a 2-fold increase of IGFBP-4 levels in uremic
animals (Fig. 6C). Mean plasma IGFBP-1 levels by RIA were increased
4-fold in uremia compared with both control groups (Fig. 6A). The
slightly reduced food intake in pair-fed animals did not significantly
influence plasma IGFBP-1, -2, and -4 levels. Figure 7
shows an immunoblot of IGFBP-3 in uremic plasma. Intact IGFBP-3
(Mr, 48 kDa) and low molecular IGFBP-3 fragments were
comparable among the three groups. We also examined intact IGFBP-3 by
[125I] IGF-II ligand blotting. Intact IGFBP-3 was
comparable between uremic animals and pair-fed controls and slightly
higher in ad libitum-fed controls (Fig. 8), which can be attributed most likely to the higher
food intake of ad libitum-fed controls (18).
|
|
|
|
and 10A) and a 4-fold increase of
hepatic IGFBP-2 mRNA abundance (Figs. 9 and 10B). The mild malnutrition
in pair-fed animals did not have a significant influence on the
expression of these two IGFBPs. In contrast to liver tissue, there was
a significant reduction (30% of ad libitum-fed
control) of IGFBP-2 expression in kidney in CRF animals (Figs. 11 and 10B). Hence, the increase of IGFBP-2 mRNA in
uremia is organ-specific. The signal for IGFBP-1 in kidney was too low
to be detected by Northern analysis. IGFBP-4 mRNA in uremia was
increased 1.5-fold in kidney, but not in liver tissue (Figs. 12 and 13A). The reduced food intake in
pair-fed animals was associated with a significant decrease of IGFBP-4
in liver but not in kidney (Figs. 12 and 13B).
|
|
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We also attempted to determine the mechanism of decreased steady state IGF-I mRNA abundance in liver tissue in experimental CRF. Circulating GH levels in this animal model are not decreased but rather increased compared with pair-fed controls and are normal compared with ad libitum-fed controls (35). Hence, reduced hepatic IGF-I gene expression in experimental CRF cannot be attributed to GH deficiency. More likely, a reduced hepatic GH receptor gene expression in experimental uremia appears to play a role. In the present study, we analyzed hepatic GH receptor gene expression in the same experimental animals that were used for analysis of IGF-I gene expression. There was a significant reduction of hepatic GH receptor mRNA at steady state in experimental CRF compared with pair-fed controls, confirming our previous observation (3). However, this decline was less pronounced than the respective decrease in hepatic IGF-I gene expression. This finding may suggest that a postreceptor defect that either impedes transmission of the GH signal or synthesis of IGF-I may also contribute to reduced hepatic IGF-I gene expression in experimental CRF. However, altered GH receptor mRNA levels may not accurately reflect membrane GH receptor binding capacity under all experimental conditions. GH binding studies using hepatic membranes from the different experimental groups will be necessary to more definitively state that GH receptor expression does not fully account for the decrease in IGF-I expression. Our observation adds further evidence to the concept of GH insensitivity in the uremic state. It is noteworthy that hepatic IGF-I gene expression is also down-regulated in a rat model of chronic metabolic acidosis (36). However, this alteration is not likely to play a pathogenic role in the experimental model of relatively mild CRF used in this study, because blood pH and bicarbonate levels were unchanged.
In a previous report, we had observed increased circulating GHBP levels in experimental uremia (3). In an attempt to determine a potential mechanism of this alteration, we analyzed hepatic GHBP mRNA expression in the present study with a specific solution hybridization/RNase protection assay that could differentiate between the GH receptor mRNA and the alternatively spliced mRNA that, in the rat, encodes the GHBP (26). Liver was examined, because this tissue is thought to be the major source of circulating GHBP (37). Hepatic GHBP mRNA was reduced to a comparable extent in experimental CRF as the respective GH receptor mRNA. Thus these data indicate that increased circulating GHBP levels in experimental CRF are not the result of increased hepatic production, but rather a consequence of reduced clearance.
As expected, the reduction in hepatic IGF-I gene expression in pair-fed
animals in our study led to a similar decline in circulating IGF-I
peptide levels. In experimental uremia, however, the even more
pronounced decline of hepatic IGF-I mRNA abundance was not converted
into a reduction of circulating IGF-I levels (Fig. 3). The pattern of
reduced hepatic IGF-I gene expression and normal total IGF-I peptide
levels is unique for CRF. This discrepancy can be explained by the
marked increase of IGFBP levels in CRF plasma that most likely increase
IGF-I half-life, and thereby lead to normal total IGF-I levels despite
a reduced hepatic IGF-I production rate. The finding that circulating
immunoreactive IGF-I levels were not reduced in experimental CRF is in
agreement with studies in clinical CRF (6, 7, 8). However, IGF bioactivity
in CRF plasma is impaired as a consequence of increased IGF binding
capacity (7).
The present study also provides the first detailed analysis of plasma
IGFBPs and respective hepatic IGFBP expression in experimental uremia.
As observed in the setting of clinical CRF, IGFBP-1, -2, and -4 plasma
levels were elevated in uremic animals. Intact IGFBP-3 (Mr,
48K) and low molecular mass IGFBP-3 fragments were comparable among
the three groups. The difference to the findings in clinical CRF, in
which plasma levels of low molecular IGFBP-3 fragments are elevated (7, 12), may be due to the mild degree of CRF in our experimental model.
Theoretically, increased IGFBP levels in CRF plasma could result from
reduced elimination by the diseased kidneys, reduced transcapillary
movement, an increased production rate, or a combination of these
factors. In our study, hepatic gene expression of IGFBP-1 and IGFBP-2
was clearly elevated in experimental uremia compared with both control
groups. This alteration was tissue-specific, because IGFBP-2 gene
expression in kidney tissue was reduced in CRF. Provided that
stimulated gene expression reflects increased protein production, this
novel finding indicates that increased hepatic production contributes
to elevated IGFBP-1 and -2 serum levels in CRF plasma. In addition, an
increase of renal IGFBP-4 mRNA abundance in uremia might contribute to
elevated IGFBP-4 plasma levels.
The mechanism for increased IGFBP-1 mRNA in liver in experimental
uremia is not known. Studies in hypophysectomized rats (38) or primary
cultures of hepatocytes (39) indicate that GH is a potent inhibitor of
IGFBP-1 synthesis. Furthermore, insulin is a negative regulator of
IGFBP-1 in humans (40) and rats (41). Although GH and insulin plasma
levels are not decreased in experimental uremia (Table 1) (35),
evidence exists for insensitivity to their action in CRF (1, 42). GH
and insulin regulate IGFBP-1 production at the levels of mRNA abundance
(43, 44), and resistance to their action should increase IGFBP-1 mRNA
levels, as occurs in liver of uremic rats. Treatment of CRF children
with pharmacological doses of exogenous GH is associated with an
increase of GH and insulin levels (45) and a decrease of IGFBP-1 levels
(46). Thus, abnormalities in GH and insulin action may participate in
IGFBP-1 regulation in CRF. However, glucocorticoids also regulate
IGFBP-1 levels at the level of transcription and, thus, mRNA abundance
(44). Our data showing that corticosterone levels are not different
among the groups suggest that glucocorticoids are not the mechanism for
high IGFBP-1 levels in experimental uremia.
IGFBP-2 mRNA levels in liver in experimental uremia were even more stimulated in the present study. In contrast to liver, kidney IGFBP-2 mRNA decreased slightly. Similar discrepancies between hepatic and renal IGFBP-2 gene expression have been observed in hypophysectomized (47), fasted (48), and diabetic rats (49). These findings support the concept of differential organ-specific regulation of IGFBP-2 in response to hormonal and nutritional changes and in chronic disease states. The specific metabolic signal responsible for the marked increase of hepatic IGFBP-2 gene expression remains to be elucidated. Because IGFBP-2 mRNA levels in liver are elevated in diabetic (49) and hypophysectomized rats (47), it has been suggested that insulin and GH are involved in the long-term regulation of hepatic IGFBP-2 gene expression. An insensitivity to the action of GH and insulin in uremic rats might therefore contribute to increased hepatic IGFBP-2 gene expression.
We report a moderate (50%) increase of renal IGFBP-4 mRNA in
experimental CRF, whereas hepatic IGFBP-4 gene expression was not
clearly altered. The mechanism for this change is not known. Hepatic
IGFBP-4 mRNA was also not modified in spontaneously diabetic (50) or
hypophysectomized rats (51), indicating that insulin and GH are not
predominant regulators of IGFBP-4 gene expression in liver.
In summary, our data demonstrate pronounced alterations of the IGF/IGFBP-axis in an experimental model of uremia. Hepatic IGF-I gene expression was specifically reduced in uremia, partially as the consequence of a reduced hepatic GH receptor gene expression. In view of normal or elevated circulating GH levels in experimental uremia, these data add further evidence to the concept of GH insensitivity in the uremic state. Despite the reduced hepatic IGF-I production rate, the IGF-I peptide levels are normal in uremic plasma as a consequence of a marked increase of IGFBP levels. One of the mechanisms contributing to increased IGFBP levels in uremia is increased hepatic gene expression of IGFBP-1 and IGFBP-2. The imbalance between reduced hepatic IGF-I production and increased hepatic IGFBP-1 and -2 production is likely to play a pathogenic role for catabolism and growth failure in CRF.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Supported by NIH Grant DK-38773.
Received August 13, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  B. Gigante, S. Rubattu, R. Stanzione, A. Lombardi, A. Baldi, F. Baldi, and M. Volpe Contribution of Genetic Factors to Renal Lesions in the Stroke-Prone Spontaneously Hypertensive Rat Hypertension, October 1, 2003; 42(4): 702 - 706. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C. M. Chan, D. M. Williams, and K. S. Roth Kidney Failure in Infants and Children Pediatr. Rev., February 1, 2002; 23(2): 47 - 60. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. ROELFSEMA and R. G. CLARK The Growth Hormone and Insulin-Like Growth Factor Axis: Its Manipulation for the Benefit of Growth Disorders in Renal Failure J. Am. Soc. Nephrol., June 1, 2001; 12(6): 1297 - 1306. [Abstract] [Full Text]
|
|
|
|
 |
|
 Y. Dubaquie, D. L. Mortensen, A. Intintoli, D. A. Hogue, G. Nakamura, P. Rancatore, P. Lester, M. D. Sadick, E. Filvaroff, P. J. Fielder, et al. Binding Protein-3-Selective Insulin-Like Growth Factor I Variants: Engineering, Biodistributions, and Clearance Endocrinology, January 1, 2001; 142(1): 165 - 173. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Guerra, A. Leri, X. Wang, N. Finato, C. Di Loreto, C. A. Beltrami, J. Kajstura, and P. Anversa Myocyte Death in the Failing Human Heart Is Gender Dependent Circ. Res., October 29, 1999; 85(9): 856 - 866. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Jain, D. W. Golde, R. Bailey, and M. E. Geffner Insulin-Like Growth Factor-I Resistance Endocr. Rev., October 1, 1998; 19(5): 625 - 646. [Abstract] [Full Text]
|
|
|
|
 |
|
 D. Haffner, W. F. Blum, U. Heinrich, O. Mehls, and B. Tonshoff Impaired Postprandial Regulation of Insulin-Like Growth Factor Binding Protein-1 in Children with Chronic Renal Failure J. Clin. Endocrinol. Metab., September 1, 1997; 82(9): 2832 - 2835. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. R. Powell, F. Liu, B. K. Baker, R. L. Hintz, S. K. Durham, E. D. Brewer, J. W. Frane, B. Tonshoff, O. Mehls, A.-M. Wingen, et al. Insulin-Like Growth Factor-Binding Protein-6 Levels Are Elevated in Serum of Children with Chronic Renal Failure: A Report of the Southwest Pediatric Nephrology Study Group J. Clin. Endocrinol. Metab., September 1, 1997; 82(9): 2978 - 2984. [Abstract] [Full Text] [PDF]
|
|
| |||||||||
X174 DNA was
used to indicate size of protected bands. Native IGF-I antisense RNA
probe was diluted 1:100 (native probe) and as a control undiluted probe
was digested with RNases.
