| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Laboratory of Pediatrics (M.v.K., C.A.H.G., N.F.J.D., D.J.L.-K., S.L.S.D., J.W.v.N.), Subdivision Molecular Endocrinology, and Department of Pathology (E.C.Z.), Erasmus University Rotterdam, Postbus 1738, 3000 DR Rotterdam, The Netherlands; and Department of Neuroscience and Cell Biology (A.G.P.S., S.L.B., J.E.P.), University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School, Piscataway, New Jersey 08854
Address all correspondence and requests for reprints to: Dr. J. W. van Neck, Erasmus University Rotterdam, Laboratory of Pediatrics, Room Ee15.00, Postbus 1738, 3000 DR Rotterdam, The Netherlands. E-mail: vanneck{at}kgk.fgg.eur.nl
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Immunohistochemistry and messenger RNA (mRNA) in situ hybridization on sections of mouse embryos of 13.5 days post coitum revealed tissue-specific expression patterns for the six IGFBPs. The only site of IGFBP-1 protein and mRNA production was the liver. IGFBP-2, -4, and -5 protein and mRNA were detected in various organs and tissues. IGFBP-3 and -6 protein and mRNA levels were low. In several tissues, such as lung, liver, kidney, and tongue, more than one IGFBP (protein and mRNA) could be detected. Differences between mRNA and protein localization were extensive for IGFBP-3, -5, and -6, suggesting that these IGFBPs are secreted and transported.
These results confirm the different spatial localization of the IGFBPs, on the mRNA and protein level. The overlapping mRNA and protein localization for IGFBP-2 and -4, on the other hand, may indicate that these IGFBPs also function in an auto- or paracrine manner.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mouse models are widely used to obtain insight into the actions of the IGF system members during development (10, 11, 12, 13, 14). Gene expression studies have shown that the IGF system components have specific spatial and temporal messenger RNA (mRNA) expression patterns during development (15, 16, 17, 18, 19, 20). Because the IGFs and IGFBPs are secreted proteins, it is relevant to extend analysis of gene transcription to the localization of the proteins at their site of action.
Several studies have confirmed the tissue-specific protein localization of the components of the IGF system in the human (7, 8, 9). The existing antibodies to rat IGFBPs, however, are not applicable in immunohistochemical detection in mice (21).
Therefore, the aim of this study was to generate specific antisera against the mouse IGFBPs (mIGFBPs) and to compare IGFBP protein localization patterns with IGFBP mRNA patterns during mouse development. Specific mIGFBP antisera were raised using synthetic peptides specific for each of the IGFBPs and were characterized by enzyme-linked immunosorbent assay (ELISA), Western blotting, and immunohistochemistry. Immunohistochemistry and in situ hybridization were performed on sections of 13.5-dpc (days post coitum) mouse embryos to analyze protein and mRNA localization of the six IGFBPs.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). These peptides were chosen
from the unique middle part of the IGFBP amino acid sequences, to limit
cross-reactivity between the various IGFBPs. The choice of the peptide
fragments was guided by hydrophobicity plots (22), surface probability
predictions according to Emini, Chou-Fasman secondary structure
predictions (23), and antigenicity index (24). A terminal cystein was
added to the peptide for sulfhydryl coupling. Conjugation to the
carrier keyhole limpet hemocyanin (KLH) was performed, following the
manufacturer’s prescription (Pierce Chemical Co.,
Rockford, IL). KLH-coupled peptides were injected sc into rabbits (0.25
mg), using specol (Central Veterinary Institute, Lelystad, the
Netherlands) as adjuvant. Three weeks after primary immunization, the
rabbits were boosted (protocol similar to primary immunization). After
a final boost, 3 weeks later (similar to previous boost), sera were
collected and used in subsequent experiments.
|
Negative controls were included, consisting of incubation of the wells plate, coated with peptides, with preimmune sera of the respective antisera, in the same dilutions. Cross-reactivity was checked by incubating each peptide with all IGFBP antisera (dilutions 1:500 and 1:1,000).
Western blotting
Murine erythroleukemia (MEL) cells were transfected with
IGFBP-1, 2, 4, and 6 complementary DNAs (17), according to Needham and
co-workers (25). In the log phase, these cells were induced for 48
h with dimethylsulfoxide. Subsequently, cell culture medium was
trichloric acid precipitated (50 vol/vol 20%trichloric acid,
overnight at 4 C) (Merck KGaA, Darmstadt, Germany).
IGFBP production was tested by IGF-ligand blotting, as described by
Hossenlopp et al. (26).
For IGFBP-3, normal mouse serum was used; whereas for IGFBP-5, total protein was extracted from fetal mouse lungs (13.5 dpc) by the TRIzol method, following the manufacturer’s prescription (Life Technologies, Rockville, MD).
Proteins were separated on a reducing 8% or 10% SDS:PAGE and transferred either to polyvinylidene difluoride (PVDF) (Millipore Corp., Bedford, MA) or nitrocellulose (Schleicher & Schuell, Inc., Dassel, Germany) membranes. The PVDF membranes were blocked with 5% nonfat dry milk (Profilar, Nutricia, Zoetermeer, the Netherlands), the nitrocellulose membranes with 3% BSA (Roche Molecular Biochemicals GmbH). Membranes were incubated overnight at 4 C with the primary antibody (dilutions varying from 1:200 to 1:20,000). As a secondary step, a peroxidase-conjugated swine antirabbit antibody was used (DAKO Corp., A/S; 1:2,000) on the PVDF membranes, 1 h at room temperature. The nitrocellulose membranes were incubated with a secondary alkaline phosphatase antibody (DAKO Corp., A/S). PVDF membranes were incubated according to the manufacturer’s procedure (Pierce Chemical Co.) and subsequently covered with ECL hyperfilms (Amersham Pharmacia Biotech, Buckinghamshire, UK). Films were illuminated varying from 1 min to 2 h, depending on the intensity of the signal. Detection on nitrocellulose membranes took place with NBT (4-nitroblue tetrazolium chloride) and BCIP (5-bromo-4-chloro-3-indolyl-phosphate).
Immunohistochemistry on sections
Balb/c mice were mated, and the morning of appearance of a
vaginal plug was assigned 0.5 dpc. Pregnant females were killed by
cervical dislocation, and embryos were collected at 13.5 dpc and fixed
overnight in 4% paraformaldehyde. Similarly, liver and kidneys from
adult gene deletion (knock-out) IGFBP-2 through -6 mutants (27, 28, 29, 30) and
their wild-type litter mates were fixed overnight in 4%
paraformaldehyde. After embedding of the fixed tissues in paraffin,
5-µm sections were cut and mounted onto aminopropyl
trioxysilane-coated slides. Immunohistochemistry was performed, making
use of a peroxidase-antiperoxidase (PAP) method or an avidine-biotine
complex method. The anti-mIGFBP antisera were used in a dilution of
1:250. Unlabeled goat antirabbit Igs (dilution 1:50) (DAKO Corp.) were used as second antibody and rabbit PAP (dilution
1:100) (DAKO Corp.) as a linker in the PAP method.
Incubation with a biotin-conjugated goat antirabbit (DAKO Corp.), completed with a streptavidin horseradish peroxidase
complex (BioGenex Laboratories, Inc., San Ramon, CA)
incubation, was used for the avidine-biotine complex method. Staining
was performed with diaminobenzidine (0.75 mg/ml) (DAB, Fluka Chemical Co., Buchs, Switzerland). Nuclei were visualized
with Mayer’s hematoxilin. The tissues were analyzed under light
microscopy. Controls were performed on sections with preimmune sera of
each rabbit, diluted 1:10 to 1:50. As a second control, the immune sera
were mixed in a dilution range with the corresponding peptides (1, 0.5,
and 0.17 µg/µl) and incubated overnight at 4 C. With these
mixtures, immunohistochemistry on sections of mouse embryos was
performed.
Probe preparation (in situ hybridization)
Specific mIGFBP complementary RNA probes were transcribed from
IGFBP-2, -3, -4, -5, and -6, as described by Schuller et al.
(17). The mouse complementary DNA SphI-SacI
fragment was used as template for the IGFBP-1 complementary RNA probe
and was cloned into pTZ18R or pTZ19R (Amersham Pharmacia Biotech, Uppsala, Sweden) for the antisense and sense
probes, respectively. Digoxigenin-11-UTP-labeled RNA probes were
prepared according to the manufacturer’s prescription (Roche Molecular Biochemicals GmbH) using T7 or SP6 RNA polymerase.
In situ hybridization
A nonradioactive in situ hybridization was performed,
essentially as described before (16). Hybridization was performed
overnight at 55 C in a humid chamber in a hybridization solution
containing 50% deionized formamide, 10% dextran sulfate, 2 x
SSC, 1 x Denhardt’s solution, 1 µg/ml transfer RNA, 250
µg/ml herring sperm DNA, and the respective probes at a concentration
of 100 ng/ml. After various washes (50% formamide/2 x SSC, 50%
formamide/1 x SSC, and 0.1 x SSC) sections were incubated
with ribonuclease T1 (2 U/ml in 1 mM EDTA/2 x SSC)
for 15 min at 37 C. The digoxigenin-labeled hybrids were detected by
antibody incubation performed according to the manufacturer’s
recommendations (Roche Molecular Biochemicals GmbH). PVA
(Polyvinylalcohol; molecular mass, 31–50 kDa; Aldrich,
Milwaukee, WI)-enhanced staining with BCIP, NBT, and levamisol was
stopped when the desired intensity of the blue precipitate was reached.
This was dependent on the relative abundance of the respective mRNAs.
Sections were counterstained with Nuclear red solution and mounted with
Euparal (ChromaGesellschaft, Stuttgart, Germany). Negative controls for
in situ hybridization were performed using sense probes,
which never gave any significant staining.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Detection of mIGFBPs on Western blot was possible for mIGFBP-1, -3, and
-5 (Fig. 1). The patterns on Western blot
corresponded with the IGF-ligand blot results (data not shown). A
distinct band of approximately 30 kDa was visible for mIGFBP-1, after
dilution of the antiserum to 1:20,000. When 1 µl mouse serum was
loaded, the IGFBP-3 43–45 kDa doublet was visible after incubation
with the anti-IGFBP-3 antiserum (diluted 1:10,000). Loading of total
protein of 13.5-dpc mouse lungs gave a band of approximately 29 kDa
when incubated with the IGFBP-5 antiserum (1:1,000). Neither of the
antisera demonstrated cross-reactivity to the other IGFBPs.
|
Furthermore, for the antisera directed against mIGFBP-2 through -6, specificity was determined, using immunohistochemistry on sections of liver and kidney from wild-type animals, using preimmune serum as a control for background staining (see panels a in Figs. 3–7), on liver and kidney from adult wild-type animals using the antiserum (see panels b in Figs. 3–7), and on livers and kidneys from adult gene deletion (knock-out) mutants with their respective antisera as a negative control (see panels c in Figs. 3–7).
IGFBP-2 preimmune serum resulted in a uniform background staining of renal structures including the tubules (Fig. 3a). The mIGFBP-2 antiserum revealed a prominent staining of the renal proximal tubules (Fig. 3b) that was absent in IGFBP-2 gene deletion mutants (Fig. 3c).
IGFBP-3 preimmune serum resulted in a weak, uniform hepatic staining
(Fig. 4a). The mIGFBP-3 antiserum
prominently stained the endothelium of the veins and arteries
(Fig. 4b) that was absent in IGFBP-3 gene deletion mutants (Fig. 4c).
|
IGFBP-5 preimmune serum resulted in a moderate staining of all renal tubules (Fig. 6a), whereas the mIGFBP-5 antiserum demonstrated a strong staining of the renal proximal tubules superimposed on the background staining (Fig. 6b). This was in contrast to the IGFBP-5 gene deletion mutants that only demonstrated a weak general renal staining (Fig. 6c).
IGFBP-6 preimmune serum resulted in a moderate staining of all renal structures including the tubules (Fig. 7a). The mIGFBP-6 antiserum demonstrated a strong staining of those renal proximal tubules that were located in the proximity of a blood vessel (Fig. 7b). This staining was absent in kidney obtained from IGFBP-6 gene deletion mutants (Fig. 7c).
Comparison of protein and mRNA localization of the six IGFBPs in
the 13.5-dpc mouse embryo
The antisera against mIGFBP-1 through -6 were applied to paraffin
sections of mouse embryos (13.5 dpc) to localize mIGFBP proteins.
Similarly, mRNA expression patterns in 13.5-dpc mouse embryos were
determined by in situ hybridization. In situ
hybridization and immunohistochemistry staining patterns are summarized
in Table 2.
|
Like IGFBP-2 mRNA expression, IGFBP-2 protein occurrence was diverse. Major sites of IGFBP-2 protein localization were liver, kidney, choroid plexus, lung, and floor plate. This is in accordance with mRNA expression. Both IGFBP-2 mRNA and protein were located in the epithelial cells of the kidney and choroid plexus (Fig. 3, d–f). In lung, IGFBP-2 mRNA and protein were present in the tubule, but also scattered in underlying mesenchymal cells (Fig. 3, g and h). However, there was a discrepancy between the tubular localization of IGFBP-2 mRNA and protein. The mRNA was located in the epithelial cells, whereas IGFBP-2 protein was detected in the underlying tubular cells.
IGFBP-3 mRNA expression was weak and colocalized with IGFBP-3 protein
in liver, tongue, lung and in nose epithelium (very weak signal) (Fig. 4
, d–g). In the tongue, IGFBP-3 mRNA and protein were not colocalizing
but expressed in neighboring cell types in muscle (Fig. 4, f and g). In
contrast to mRNA expression, IGFBP-3 protein also could be detected in
the choroid plexus and kidney, whereas mRNA expression could also be
detected in the heart.
In accordance with its mRNA expression, staining of IGFBP-4 protein was detected widespread. Overlapping mRNA and protein expression was visualized in mesencephalon, telencephalon (Fig. 5, d and e), heart, liver, lung, tongue mesenchyme, and blood vessels. In addition to the mRNA expression profiles, IGFBP-4 protein could be demonstrated in the kidney (results not shown).
IGFBP-5 mRNA was expressed abundantly and was largely overlapping
protein occurrence. Identical to mRNA expression, IGFBP-5 protein was
localized in the liver, endothelium of the gut, meninges, tubules and
mesenchyme of the lung (very weak protein staining) (Fig. 6, d–g), and
tongue. In addition, protein was detected in the notochord, the floor
plate, and muscle.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All antisera demonstrated specific staining patterns that were absent in preimmune controls and in tissues obtained from the corresponding gene deletion mutant. The fact that the antisera against mIGFBP-2, -4, and -6 were not able to detect the corresponding IGFBP on Western blot may be attributable to conformational changes of the protein induced by the Western-blot procedure.
Furthermore, immunohistochemical localization of mIGFBP proteins (mIGFBP-1 to -6) was compared with the mRNA patterns obtained with in situ hybridization of sections of the midgestational mouse embryo. These data extend our former gene expression studies that describe the tissue-specific mRNA expression of IGFBP-1 through -6 in the mouse embryo (15, 16, 17, 31, 32) and will be discussed.
Our results show that, similar to mRNA expression patterns, localization of IGFBP proteins in the mouse midgestational embryo is tissue-specific. IGFBP-1 mRNA and protein were detected primarily in the liver, whereas IGFBP-2, -4, and -5 were distributed among various tissues. Conform the low IGFBP-1, -3, and -6 mRNA levels, these IGFBPs were more difficult to detect than the other IGFBPs.
The limited data available on IGF system mRNA and protein expression during human development confirm our data (7, 8, 9, 33). IGFBP-2 expression is prominent during development of the nervous system, and the abundant IGFBP-2 protein and mRNA expression we observed in the choroid plexus confirmed previous observations (34, 35). In this respect, it is striking to note that we often observed a nuclear IGFBP-2 immunoreactivity in the embryological tissues. Although it is tempting to speculate about an intranuclear IGFBP-2 localization, such as described for IGFBP-3 (36), this cannot be concluded from these experiments.
Although protein localization of the IGFBPs seemed to be mostly similar to mRNA expression, some differences in localization in the 13.5-dpc mouse embryo were dem-onstrated.
Differences between IGFBP mRNA and protein localization may reveal detection thresholds for either mRNA or protein, or secretion of the protein without binding to the target organ (and protein is then probably washed out during the immunohistochemical procedure). Discrepancy between protein and mRNA expression was seen within tissues. In lung, IGFBP-2 mRNA was detected in epithelial cells of the tubules. However, IGFBP-2 protein was localized in the neighboring tubular cells. Similarly, IGFBP-3 mRNA and protein were located in neighboring cells in muscles of the tongue. Apparently, in these cases, mRNA expression in a given cell-type gives rise to protein localization in a cell-type that is in close contact. This may point to a mechanism of action where a certain cell type makes and secretes an IGFBP that acts at another (neighboring) cell-type.
Furthermore, substantial differences between mRNA and protein localization were found for IGFBP-3, -5, and -6. IGFBP-3 is the major circulatory IGFBP (1); and, after its secretion in liver and the cardiovasculair system, IGFBP-3 can easily move throughout the embryo. The IGFBPs may be transported to specific sites, but the differences also may indicate that the mRNA is very labile in these specific tissues and, hence, present in concentrations below detection limits. In that case, translation should be efficient, and the protein must be very stable to enable protein detection.
Whereas mRNA and protein expression of the different IGFBPs was clearly distinct, most tissues were found to express more than one IGFBP. Examples are liver, lung, heart, and tongue. Two possible explanations for this phenomenon exist: all IGFBPs have a specific function and each contribution is necessary for a functional organ; or all IGFBPs have similar functions and show redundancy.
The fact that IGFBP knock-outs do not show dramatic phenotypes (14, 37) may suggest overlapping function or redundancy. The IGFBP-2 knock-outs are characterized by an decreased spleen size and changed IGFBP serum levels, probably to compensate for the lack of IGFBP-2 (37). The body weight and body length of the IGFBP-4 knock-outs are slightly smaller than of wild-type mice (14). However, these minor changes seen cannot be explained by changed patterns of IGFBP expression in tissues of interest.
The tissue-specificity of the IGFBPs and the consistence between IGFBP mRNA and protein localization patterns may suggest that the IGFBPs (except IGFBP-3) function in an autocrine or paracrine manner. The fact that several tissues express more than one IGFBP can also be explained as an indication for the complex regulation of the IGF system.
Summarizing, the generated antisera against mIGFBP-1 through -6 demonstrated specificity in ELISA, Western-blot, and immunohistochemistry. With these antisera, IGFBP protein localization could be compared with IGFBP mRNA expression patterns in 13.5-dpc mouse embryos. This revealed tissue-specific and consistent mRNA and protein localization, in confirmation with the suggested para- and autocrine functions of the IGFBPs on IGF action.
We anticipate that this set of specific antisera may become an important tool for future studies of IGFBP studies in the mouse.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received March 1, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. A Meester-Smoor, A. C Molijn, Y. Zhao, N. A Groen, C. A H Groffen, M. Boogaard, D. van Dalsum-Verbiest, G. C Grosveld, and E. C Zwarthoff The MN1 oncoprotein activates transcription of the IGFBP5 promoter through a CACCC-rich consensus sequence J. Mol. Endocrinol., January 1, 2007; 38(1): 113 - 125. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y W Wong, J Chew, H Yang, D T H Tan, and R Beuerman Expression of insulin-like growth factor binding protein-3 in pterygium tissue Br. J. Ophthalmol., June 1, 2006; 90(6): 769 - 772. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. S. Watson, P. Bialek, M. Anzo, J. Khosravi, S.-P. Yee, and V. K. M. Han Elevated Circulating Insulin-Like Growth Factor Binding Protein-1 Is Sufficient to Cause Fetal Growth Restriction Endocrinology, March 1, 2006; 147(3): 1175 - 1186. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. W. Wood, P. J. Schlueter, and C. Duan Targeted Knockdown of Insulin-Like Growth Factor Binding Protein-2 Disrupts Cardiovascular Development in Zebrafish Embryos Mol. Endocrinol., April 1, 2005; 19(4): 1024 - 1034. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Li, J. Xiang, and C. Duan Insulin-like Growth Factor-binding Protein-3 Plays an Important Role in Regulating Pharyngeal Skeleton and Inner Ear Formation and Differentiation J. Biol. Chem., February 4, 2005; 280(5): 3613 - 3620. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
