Elevated Circulating Insulin-Like Growth Factor Binding Protein-1 Is Sufficient to Cause Fetal Growth Restriction
http://www.100md.com
《内分泌学杂志》
Canadian Institutes of Health Research Group in Fetal and Neonatal Health and Development (C.S.W., V.K.M.H.), Departments of Pediatrics, Obstetrics and Gynecology and Biochemistry, University of Western Ontario, Children’s Health Research Institute, London, Ontario, Canada, N6C 2V5
London Regional Cancer Program (P.B., S.-P.Y.), University of Western Ontario, London, Ontario, Canada, N6A 4L6
Division of Pediatric Endocrinology (M.A.), David Geffen School of Medicine at the University of California, Los Angeles, Los Angeles, California 90095
Diagnostic Systems Laboratories (Canada) Inc. (J.K.), Department of Laboratory Medicine and Pathobiology, Faculty of Medicine, University of Toronto and Mount Sinai Hospital, Toronto, Ontario, Canada, M5G 1X5
Abstract
IGF binding protein-1 (IGFBP-1) inhibits the mitogenic actions of the IGFs. Circulating IGFBP-1 is elevated in newborns and experimental animals with fetal growth restriction (FGR). To establish a causal relationship between high circulating IGFBP-1 and FGR, we have generated transgenic mice using the mouse -fetoprotein gene promoter to target overexpression of human IGFBP-1 (hIGFBP-1) in the fetal liver. These transgenic mice (AFP-BP1) expressed hIGFBP-1 mainly in the fetal hepatocytes, starting at embryonic d 14.5 (E14.5), with lower levels in the gut. The expression peaked at 1 wk postnatally (plasma concentration, 474 ± 34 ng/ml). At birth, AFP-BP1 pups were 18% smaller [weighed 1.34 ± 0.02 g compared with 1.62 ± 0.04 g for wild type (WT); P < 0.05], and they did not demonstrate any postnatal catch-up growth. The placentas of the AFP-BP1 mice were larger than WT from E16.5 onwards (150 ± 12 for AFP-BP1 vs. 100 ± 5 mg for WT at E16.5; P < 0.05). Thus, this model of FGR is associated with a larger placenta, but without postnatal catch-up growth. Overall, these data clearly demonstrate that high concentrations of circulating IGFBP-1 are sufficient to cause FGR.
Introduction
FETAL GROWTH IS a complex process controlled by many factors, including the genetics of mother and fetus as well as maternal and environmental factors. Clinically important impaired fetal growth resulting in fetal growth restriction (FGR) is often caused by a poor in utero environment resulting from maternal or placental pathology or abnormal physiology. FGR is associated with significant perinatal and neonatal mortality and morbidity, including hypoxic injury to vital organs and preterm delivery (1, 2, 3). FGR pregnancies result in small for gestational age newborns who are at high risk for negative effects on childhood growth and neurodevelopment. Only 80% of FGR newborns demonstrate catch-up growth during early infancy and achieve adult height within normal percentiles, whereas 20% remain small throughout life (4). Furthermore, adaptations to a hostile in utero environment resulting in small birth size increase the risk of cardiovascular and metabolic diseases as adults (5, 6), a process known as developmental programming.
The IGFs (IGF-I and -II) are essential for cell growth and differentiation during development, such that mice carrying null mutations of either the Igf1 or Igf2 genes weigh 60% of normal at birth (7, 8). IGF-I and -II are produced in most tissues such that they have both endocrine and paracrine actions. The interactions of the IGFs with their receptors are modulated by six species of high-affinity IGF binding proteins (IGFBP-1 to –6), which regulate the bioavailability of IGFs through protection of the IGFs from degradation in circulation and inhibition of IGF binding to cell surface receptors (9, 10).
There is mounting evidence that circulating IGFBP-1 plays a crucial role in the pathogenesis of FGR. In humans, birth weight is negatively correlated with IGFBP-1 concentrations in cord blood (11, 12, 13). In animal models of FGR, including uterine artery ligation, hypoxia, and maternal undernutrition, low birth weight is also associated with high circulating levels of IGFBP-1 as a result of increased production in the fetal liver (14, 15, 16, 17, 18, 19, 20). However, the elevated IGFBP-1 levels occur concurrently with reduced circulating IGF-I concentrations and a diminished supply of nutrients and substrates; thus the role of IGFBP-1 in FGR is unclear. Previous investigators have produced transgenic mice that overexpress IGFBP-1 under the control of various promoters to examine the role of IGFBP-1 in the mitogenic actions of the IGFs in vivo. Although some of these models demonstrated modest reductions in growth, these experiments were not necessarily designed to specifically examine the relationship between elevated circulating fetal IGFBP-1 and fetal growth. For example, in some models, overexpression was not targeted to the liver, whereas in others, levels of circulating IGFBP-1 were relatively low (21, 22, 23, 24). Therefore, the objective of this study was to determine the growth phenotype of transgenic mice (AFP-BP1) designed to overexpress IGFBP-1 specifically in the fetal liver with resultant elevated circulating IGFBP-1 concentrations, without directly altering levels of IGF-I or disturbing the nutrient and/or substrate supply to the fetus. To achieve this goal, we generated transgenic mice in which the regulatory region of the mouse -fetoprotein (AFP) gene was used to direct overexpression of a human IGFBP-1 cDNA in the fetal liver. Liver was chosen as the site of transgene expression because in the fetus, IGFBP-1 is expressed only in the liver, and it is the predominant source of circulating IGFBP-1 (25). Because AFP is also expressed mainly in the fetal liver and its expression declines after birth (26), we anticipated that the AFP-BP1 mice would have elevated levels of circulating human IGFBP-1 (hIGFBP-1) from mid-gestation to early postnatal life associated with a reduction in overall fetal growth.
Materials and Methods
Transgene construction
A hIGFBP-1 cDNA clone was generated by RT-PCR using human decidual RNA as the template. Primers were designed to amplify an 841-bp cDNA containing the coding sequence of hIGFBP-1 (Table 1). The resulting amplicon was inserted into the pGEM-T vector (Promega Corp, Madison, WI). The hIGFBP-1 cDNA was then subcloned into the NotI site of a pSP72 vector containing the 7.6-kb upstream regulatory region of the mouse AFP gene (a generous gift from S. Tilghman, Princeton University, Princeton, NJ), which is sufficient to target high levels of expression in the fetal liver (26, 27, 28, 29). Finally, the simian virus 40 (SV40) intronic-polyadenylation sequence obtained from the SV40neo plasmid (805 bp) was inserted into the XhoI site of the AFP-hIGFBP-1-pSP72 plasmid, downstream from the hIGFBP-1 cDNA, to increase the efficiency of in vivo transgene expression. The final size of the transgene (AFP-BP1) was 9.2 kb (Fig. 1).
Generation of transgenic mice
The transgene was released from the plasmid by EcoRI digestion, fractionated by agarose gel electrophoresis, and then electroeluted and purified by Elutip-d columns according to manufacturer instructions (Schleicher & Schuell, Dassel, Germany). The purified fragment was resuspended in 10 mM Tris (pH 7.5) and 0.25 mM EDTA and microinjected into the pronuclei of one-cell embryos obtained from mating between (C57BL6/CBA)F1 animals. Founder animals were bred to homozygosity and backcrossed into CBA, the wild-type (WT) strain, for at least four generations. For all experiments, CBA WT animals were used as control.
Animals and sample collection
The University of Western Ontario Committee on Animal Care, following the guidelines of the Canadian Council on Animal Care, approved all protocols. Mice were bred and maintained in the animal facility at the Lawson Health Research Institute. They were fed standard lab chow and maintained on a 12-h light, 12-h dark cycle. Postnatal animals were weighed daily at 0900 h from postnatal d 1–14 and then weekly. Fetuses were collected and weighed from embryonic d 12.5 (E12.5) to E18.5. Animals were killed with an ip overdose of sodium pentobarbital (Euthanyl; MTC Pharmaceuticals, Cambridge, Ontario, Canada) and blood samples collected by intracardiac puncture into heparinized capillary tubes (Chase Scientific Inc., Rockwood, TN). After centrifugation, plasma was removed and stored at –20 C. Organs were quickly dissected, weighed, and either snap-frozen in liquid nitrogen and stored at –80 C for RNA extraction or placed in 4% paraformaldehyde (BMD Science, Mississauga, Ontario, Canada) followed by blocking in paraffin for in situ hybridization analysis or immunohistochemistry.
Extraction of genomic DNA, Southern blot, and PCR analyses
Tail biopsies were collected postnatally at 2 wk and digested overnight with 800 U proteinase K (Invitrogen Life Technologies, Carlsbad, CA) in 20 mM Tris-HCl (pH 8.0), 10 mM EDTA, 10 mM NaCl, and 0.5% SDS at 55 C. Genomic DNA was then extracted using phenol-chloroform and precipitated in 100% ethanol (30).
Southern blot analysis was performed to identify founder mice, for the determination of copy number, and zygosity of the offspring. Briefly, 10 μg genomic DNA was digested for 6 h with 50 U PstI (Invitrogen) at 37 C. The DNA fragments were then fractionated by agarose gel electrophoresis and transferred onto nylon membrane (Hybond-N; Amersham-Pharmacia Biotech Canada, Baie d’Urfe, Quebec, Canada) in 10x standard saline citrate (SSC), and the membranes baked for 2 h at 80 C (31). The blots were prehybridized for at least 2 h at 65 C in 0.5 M phosphate buffer containing 7% SDS and 100 μg/ml sonicated salmon sperm DNA. Blots were then hybridized overnight at 65 C with a radiolabeled DNA fragment (2 x 106 cpm/ml) corresponding to the SV40 intronic-polyadenylation sequence in the transgene. The probe was labeled with [-32P]dCTP by random priming (30) (Amersham-Pharmacia). After hybridization, blots were washed twice for 15 min each in 2x SSC, 0.1% SDS at room temperature and twice for 15 min each in 0.1x SSC, 0.1% SDS at 65 C and then exposed to film (Biomax MR; Kodak Canada, Toronto, Ontario, Canada) at –80 C. Signal intensity of bands on the autoradiograms was analyzed with a densitometry program (Phoretix Software; Nonlinear Dynamics Ltd., Newcastle upon Tyne, UK).
After transgenic lines were verified by Southern blot analysis, subsequent transgenic offspring were confirmed by PCR analysis. Primers were designed to amplify the nonmurine intronic-SV40 polyadenylation sequence of the transgene (Table 1). Thirty cycles were performed in the PCR, with an annealing temperature of 60 C and 5 mM Mg2+.
RNA isolation and Northern blotting
Total RNA was extracted from up to 100 mg frozen tissue using the Trizol method (Invitrogen Life Technologies) with precipitation in isopropanol overnight at –80 C. Twenty micrograms of total RNA were loaded into 1% agarose/16% formaldehyde gels and electrophoresed overnight in 1x 3[N-morpholino]propanesulfonic acid. The RNA was then transferred to Zetaprobe membrane (Bio-Rad Laboratories Canada, Mississauga, Ontario, Canada) in 10x SSC and the membranes baked at 80 C for 2 h. Membranes were prehybridized for 2 h at 42 C in 50% deionized formamide, 1x saline sodium phosphate EDTA, 7% SDS, and 100 μg/ml sonicated salmon sperm DNA. The blots were then hybridized overnight at 42 C in the same buffer, with a [-32P]dCTP-labeled SV40 polyadenylation sequence probe (1 x 106 cpm/ml). After hybridization, the blots were washed twice for 15 min each in 1x SSC, 0.1% SDS at 42 C; once for 30 min in 1x SSC, 0.1% SDS at 65 C; and twice for 30 min each in 0. 1x SSC, 0.1% SDS at 65 C and then exposed to film at –80 C (Biomax MS; Kodak Canada). Consistency of total RNA loading and transfer was verified by probing the blots for 18S ribosomal RNA using the same conditions. Autoradiograms were quantified using Phoretix software (Nonlinear Dynamics, Ltd.).
In situ hybridization
In situ hybridization was performed as described previously (32) on 5-μm sections taken from AFP-BP1 and WT animals at various embryonic and neonatal time points, and individual organs collected from postnatal animals. Briefly, sections were deparaffinized, rehydrated, and incubated in 0.2% Triton for 1 h at room temperature followed by proteinase K (0.2 U/ml) at 37 C for 30 min and finally in 0.1 M triethanolamine for 10 min at room temperature. After subsequent dehydration, sections were prehybridized for 2 h at 45 C in prehybridization buffer [50% formamide, 0.3 M NaCl, 20 mM Tris (pH 8.0), 1 mM EDTA, 1x Denhardt’s, 500 μg/ml yeast tRNA, 0.1% SDS, 100 mM dithiothreitol, and 100 μg/ml salmon sperm DNA]. Sections were then hybridized with 35S-labeled antisense and sense cRNA probes at 55 C overnight in hybridization buffer (prehybridization buffer, plus 10% dextran sulfate). The probes were generated in our laboratory, using the SV40neo polyadenylation sequence of the transgene cloned into pGEM-T as a template so as to achieve hybridization only to the transgenic mRNA. After hybridization, slides were washed at a maximum stringency of 0.1x SSC at 65 C for 20 min. Sections collected from WT animals were analyzed as negative controls. In addition, sections were also probed with sense probes to verify the specificity of the antisense cRNA probes.
Immunohistochemistry
Immunohistochemistry was performed on 5-μm sections taken from whole AFP-BP1 and WT animals at embryonic and newborn time points, and individual organs collected from postnatal animals. After deparaffinization and rehydration, immunohistochemistry was performed using the Vectastain Elite Kit (Vector Laboratories, Burlington, Ontario, Canada). Briefly, sections were blocked with normal goat serum for 30 min at room temperature and then incubated overnight at 4 C with a polyclonal anti-hIGFBP-1 antibody (33) at a concentration of 1:5000. Protein was visualized using diaminobenzidine and counterstained with methyl green.
Western immunoblotting and ligand blotting
The presence of hIGFBP-1 in plasma and tissue extracts was determined by immunoblotting using a specific anti-hIGFBP-1 antibody (33), as previously described (34). The profile and levels of the other IGFBPs were determined by ligand blot analysis. Briefly, plasma samples (5 μl) or tissue extracts (60 μg total protein, as determined by BCA analysis) were boiled in Laemmli sample buffer [0.15 M Tris-HCl (pH 6.8), 4.8% SDS, 24% glycerol, and 0.024% bromophenol blue] for 20 min and separated in 12% SDS-PAGE under nonreducing conditions. Proteins were transferred electrophoretically to 0.2-μm nitrocellulose membranes (Schleicher & Schuell, Dassel, Germany) and then blocked in buffer BT [10 mM Tris-HCl (pH 7.4), 150 mM NaCl, 0.2% Tween 20] plus 1% BSA for 1 h at room temperature. The membranes were then washed three times for 10 min each in buffer BT and incubated with biotinylated IGF-I (10 ng/ml) for 3 h at room temperature. After washing, the membranes were incubated with streptavidin-horseradish peroxidase (Amersham-Pharmacia) (1:1500) for 45 min at room temperature. The proteins were visualized using chemiluminescence (Western Lightning; Perkin-Elmer Life Sciences, Inc., Boston, MA) and exposed to X-OMAT Blue autoradiography film (Kodak), and band intensity was analyzed using Phoretix Software (Nonlinear Dynamics Ltd.).
hIGFBP-1 ELISA
Concentrations of total hIGFBP-1 in plasma were measured using a total hIGFBP-1 ELISA (Diagnostic Systems Laboratories Inc., Webster, TX). Concentrations of nonphosphorylated hIGFBP-1 were measured as previously described (35). Samples were diluted 1:5 and assayed in duplicate. The coefficient of variation between duplicates was less than 5%, and all samples were analyzed in one assay. The antibody was specific for hIGFBP-1 and did not cross-react with murine IGFBP-1 in plasma collected from WT mice.
IGF-I ELISA
The 96-well microtiter plates were coated with capture antibody at 0.5 μg/well in 100 μl PBS (pH 7.2), incubated overnight at room temperature on a shaker, and then washed three times with 300 μl wash buffer (PBS and 0.05% Tween 20) followed by 1 h incubation with 300 μl blocking buffer (PBS, 5% Tween 20, 5% sucrose, and 0.05% sodium azide) and three final washes. Recombinant mouse IGF-I standard (R&D Systems, Minneapolis, MN) was diluted in assay buffer (50 mM sodium phosphate, 150 mM NaCl, 0.1% Tween 20, 0.25% BSA, pH 7.4) in concentrations ranging from 0–25 ng/ml. Before assay, serum samples were extracted with acid/ethanol reagent (12.5% 2 N HCl and 87.5% ethanol), neutralized with 1 M Tris base, and diluted with assay buffer. Standards, controls, or diluted samples (50 μl /well) and 50 ng biotinylated goat antimouse IGF-I antibody (R&D Systems) in assay buffer (50 μl/well) were incubated at room temperature for 2 h on a shaker. The wells were washed three times with wash buffer followed by the addition of streptavidin-horseradish peroxidase conjugate (100 μl/well) (Pierce, Rockford, IL) and additionally incubated for 20 min at room temperature. After four washes, 100 μl o-phenylenediamine dihydrochloride (1 mg/ml in hydrogen peroxide substrate) was added to each well and incubated for an additional 10–20 min. The reaction was stopped by the addition of 50 μl 2 N H2SO4, and the absorbance was determined at 490 nm in a plate reader (Molecular Design, Sunnyvale, CA). The sensitivity of this assay is 0.1 ng/ml. There is no cross-reactivity with mouse IGF-II; mouse IGFBP-1, -2, or -3; human IGF-I or -II; or hIGFBP-1, -2, -3, or -4 (all test substances at 200 ng/ml). The intra- and interassay coefficients of variation were less than 10% in the range from 1–10 ng/ml, and recovery was 91–101%.
Statistical analyses
Data were analyzed for differences between AFP-BP1 and WT animals using two-way ANOVA and unpaired Student’s t tests where applicable.
Results
Southern blot and PCR identification of founder mice
Four founder mice (designated as AFP-BP1 lines 2, 3, 4, and 16) were identified on the basis of Southern blot analysis of genomic DNA obtained from tail biopsies. With a restriction digest using PstI, which cleaved the transgene at a single site (Fig. 1A), Southern blotting detected a major single genomic fragment in these four founders, corresponding to the expected size of 9.2 kb (Fig. 1B). In addition, PstI cleavage produced only one other fragment (data not shown), corresponding to the DNA fragment at the 3' end of the inserted transgene, indicating that there is only a single site of transgene insertion in the genome of these founder mice.
Southern blot analysis was also performed to determine the copy number of the transgene. Lines 2, 3, 4, and 16 carried four, two, one, and eight copies of the transgene, respectively (Fig. 1, D and E). Founder mice were bred with WT littermates, and only lines 2, 3, and 16 produced transgenic offspring. Because lines 16 and 2 contained the highest copy number, they were further characterized to determine transgene expression patterns and any phenotypic alterations. Subsequent analysis of offspring genotype was completed by PCR. Primers (Table 1) were designed to amplify an 800-bp fragment specific to the SV40 intron-polyadenylation sequence of the transgene (Fig. 1C).
Ontogeny of transgene expression: in situ hybridization and Northern blot analysis
In situ hybridization was performed on sagittal sections of both AFP-BP1 and WT mice collected at various embryonic stages and at postnatal d 1 (newborn) and on postnatal liver, brain, and gut samples using an antisense riboprobe specific to the SV40 polyadenylation sequence. Overall, high levels of transgene expression were found in the fetal liver (Fig. 2) and yolk sac (data not shown) and at lower levels in the gut and brain in both lines 16 and 2 (Fig. 2).
Transgenic hIGFBP-1 mRNA was first detected in the liver at E14.5. Transgene expression in the fetal liver increased as gestation progressed, peaking on the day of birth. The transgenic mRNA was present at lower levels in the liver at 8 wk. At all ages, transgenic mRNA was expressed in the hepatocytes only. The expression levels of transgenic mRNA were much lower in the gut and brain compared with the liver at all ages. Expression was limited to epithelial cells in the gut and glial cells in the brain. Expression of the transgene in the yolk sac was also first evident at E14.5 and remained consistent until birth. There was no specific hybridization in any WT tissue examined (Fig. 2).
Northern blot analysis of total RNA obtained from E14.5 liver and gut, brain, and liver from E18.5 onward, showed two main transcripts of 1.6 and 2.1 kb and a low abundant transcript of 1.1 kb. The ontogeny of the transgenic mRNA expression pattern as revealed by Northern blotting was similar to that found with in situ hybridization. In the liver, transgene expression increased from E14.5 to 1 wk (Fig. 3A). Expression in the gut was highest at E18.5 and the newborn period, becoming almost undetectable by 1 wk (Fig. 3B). In contrast, expression in the brain increased from E18.5 to 4 wk (Fig. 3C), but overall the levels were much lower compared with the liver at all ages examined (Fig. 3, C–E).
hIGFBP-1 protein expression: immunohistochemical and Western blot analyses
Immunoreactive hIGFBP-1 was detected at high levels in the perinatal liver and at lower levels in the gut of transgenic mice. Despite the presence of transgenic mRNA in the brain, immunoreactive hIGFBP-1 could not be detected in the transgenic brain using this technique (Fig. 2). There was no antibody cross-reactivity with endogenous IGFBP-1 in sections taken from WT animals.
Plasma samples were obtained from the AFP-BP1 mice, and Western blot analysis using anti-hIGFBP-1 revealed an immunoreactive protein at 27 kDa in the AFP-BP1 plasma, with no cross-reactivity with any protein in WT samples. The level of the immunoreactive hIGFBP-1 decreased in plasma at 4 wk compared with the day of birth (Fig. 4A).
Western ligand blotting using IGF-1 as the probe detected a protein species of 27 kDa in the AFP-BP1 mice, but not WT. A protein species of 29 kDa, representing mouse IGFBP-1, was also visible. There was no difference in band intensity of mouse IGFBP-1 between transgenic and WT samples in newborn plasma. However, at 4 wk, there were significantly lower levels of mouse IGFBP-1 in plasma obtained from AFP-BP1 mice compared with WT animals (Fig. 4, B and C). Protein species of 38–44 kDa, representing IGFBP-3, were also visible. There was significantly less IGFBP-3 in the plasma of AFP-BP1 mice compared with WT animals at both d 1 and 4 wk. There was no difference in levels of circulating IGFBP-4 at either age (Fig. 4, B and C).
Western immunoblot analyses using anti-hIGFBP-1 antibody were also performed to examine protein extracts from brain and liver of AFP-BP1 and WT animals. Overall, these analyses demonstrated the presence of a 27-kDa immunoreactive protein in the AFP-BP1 mice that was not present in WT animals but was present in human amniotic fluid. Immunoreactive hIGFBP-1 was present in the liver of the AFP-BP1 mice at all ages examined. In the brain, immunoreactive IGFBP-1 was present on the day of birth but was not detectable at 4 wk in the AFP-BP1 mice (Fig. 4D).
Circulating hIGFBP-1 concentrations
hIGFBP-1 was detected in the plasma of AFP-BP1 animals at all ages examined. The circulating concentrations of total hIGFBP-1 in plasma were highest in the perinatal period, decreasing significantly by 4 wk of age. The percentage of total hIGFBP-1 that was nonphosphorylated was also highest in the perinatal period and declined postnatally. Levels of circulating hIGFBP-1 in line 2 transgenic mice were lower than that in line 16, but a similar percentage was phosphorylated at d 1 and 8 wk (Table 2). There was no hIGFBP-1 detected by ELISA in WT plasma.
Plasma IGF-I concentrations
Plasma IGF-I concentrations in AFP-BP1 mice were 21 ± 6 ng/ml in the newborn period and 325 ± 48 ng/ml and 336 ± 38 ng/ml at 4 and 8 wk, respectively. These concentrations were not different from those of the WT mice of the same ages (Table 3).
Fetoplacental growth
There were no reproductive abnormalities in the AFP-BP1 mice. Litter size was not altered (8.1 ± 0.5 pups per litter for AFP-BP1 vs. 7.9 ± 1.0 pups per litter for WT), and there was no effect of maternal genotypic background on survival rates.
At E12.5, there was no difference in total body weight between AFP-BP1 and WT fetuses. Starting at E14.5, the line 16 AFP-BP1 fetuses were significantly smaller than WT and remained so until birth (Fig. 5A). Starting at E16.5, the placentas of line 16 AFP-BP1 mice were significantly larger than WT (Fig. 5B), yielding a significantly larger placenta to body weight ratio in late gestation (Fig. 5C).
Postnatal growth
At birth, lines 16 and 2 AFP-BP1 mice weighed 1.33 ± 0.02 and 1.44 ± 0.05 g, respectively, compared with 1.62 ± 0.04 g for WT mice (P < 0.01). Postnatally, line 16 AFP-BP1 mice remained 15–20% smaller than WT at all ages examined. At weaning, line 2 AFP-BP1 mice continued to be small (12.5 ± 0.3 g) but were not significantly different from WT (13.7 ± 0.6 g) at this age. However, at 8 wk of age, line 2 AFP-BP1 mice (22.4 ± 0.5 g) were significantly smaller than WT (24.7 ± 0.8 g) but bigger than line 16 AFP-BP1 mice (20.4 ± 0.6 g) (Fig. 6A). The growth patterns were similar between males and females in both lines. At birth, the weights of male AFP-BP1 mice were 16.2% smaller than WT males, and female AFP-BP1 mice were 18.8% smaller than WT females. By 8 wk of age, male AFP-BP1 mice were 16.2% smaller, whereas females were 18.3% smaller than WT.
Discussion
IGFBP-1 is a potent inhibitor of IGF-I and IGF-II interactions with the IGF-I receptor. In addition, increased circulating IGFBP-1 is a consistent observation in human FGR fetuses and newborns and animal models of FGR (11, 12, 13, 14, 15, 16, 17, 18, 19, 20). However, whether impaired fetal growth is caused by elevated circulating IGFBP-1 has not been clearly established because of changes in IGF-I, nutrients, and substrates that accompany FGR, which could also explain the growth impairment. To establish a causal relationship between IGFBP-1 and FGR, we have generated a model of FGR by increasing circulating IGFBP-1 concentrations in AFP-BP1 mice that overexpress hIGFBP-1 in the fetal liver. AFP-BP1 mice differ from previous models of IGFBP-1 overexpression in the specificity of the transgene expression in the fetal liver, the sustained elevation in circulating IGFBP-1 in fetal life, and the consistent decrease in growth both pre- and postnatally. In addition, there is no concomitant change in circulating IGF-I concentrations in this model of IGFBP-1 overexpression. This study demonstrates clearly that an increase in circulating IGFBP-1 during fetal life, without a direct confounding reduction in circulating IGF-I concentrations and/or direct alteration in nutrients and substrates, is sufficient to cause FGR.
FGR was demonstrated in two AFP-BP1 lines, each carrying a different copy number of the transgene. Moreover, the growth restriction was less in line 2, which contained a fewer number of copies and had a lower level of circulating hIGFBP-1. These data indicate that the changes in growth in the AFP-BP1 mice were caused by the transgene itself and not to positional effects of the transgene integration site in the mouse genome.
Human intrauterine growth-restricted (IUGR) newborns have decreased IGF-I levels, in addition to the elevation in IGFBP-1, possibly contributing to the decreased growth in utero (36). A previous model of transgenic IGFBP-1 overexpression in the postnatal liver demonstrated a decrease in circulating IGF-I (23). The somatotrophic axis was extensively studied in those mice, and the decrease in IGF-I was attributed to a decrease in GH production by the pituitary gland, related to the impaired neurological development seen in these mice (37). In contrast, AFP-BP1 mice demonstrated no difference in circulating IGF-I levels from the newborn period to adulthood at 8 wk of age. Therefore, this would indicate that the growth restriction seen in this transgenic model is likely a result of a reduction in bioavailable IGF-I, not a decrease in total IGF-I, thereby providing strong evidence that the expression of Igfbp-1 in the fetal liver is a major factor in the control of fetal growth. Because oxygen- and insulin-response elements are present in the regulatory sequences of the endogenous Ifgbp-1 gene, hypoxia and low glucose levels would lead to high levels of IGFBP-1 (38). This relationship represents a link between poor placental transfer of nutrients and substrates and FGR. In the AFP-BP1 mice, we were able to separate the increase in circulating IGFBP-1 from a decrease in nutrients and IGF-I, and our results demonstrate that IGFBP-1 itself might be a pivotal control molecule in the pathogenesis of FGR.
A recent study demonstrated that a placental-specific Igf2 null mutation leads to impaired syncytiotrophoblast development and placental transport and subsequent FGR. This study provided the strongest evidence to date for the importance of placental development in fetal growth (39, 40). It is interesting that the growth-restricted AFP-BP1 mice described in our study exhibited an increase in placental weight. Although we have not examined the placental structure in detail, previous studies have demonstrated that an increased placental weight after elevated IGFBP-1 is caused by an increase in the size of the labyrinth zone, the area of nutrient exchange in the mouse placenta (21). These changes may be direct effects of IGFBP-1 on placental growth or may be compensatory changes designed to increase passage of nutrients to the growth-restricted fetus. This may explain, at least in part, the relatively small decrease in fetal size. A high placental to birth weight ratio is associated with increased adult blood pressure in children and adults as well as an increase in the prevalence of coronary heart disease and impaired glucose tolerance (41, 42, 43). The increase in circulating IGFBP-1 seen in growth-restricted fetuses may contribute to altered placental development and subsequent developmental programming for adult-onset diseases. Long-term studies of the AFP-BP1 mice will provide clues to some of the mechanisms underlying the relationship between placental growth and developmental programming.
Transgenic mice overexpressing IGFBP-1 in patterns that differ from the AFP-BP1 mice have been described previously, some of which also exhibited alterations in growth. However, none of these studies were designed to specifically examine the role of elevated IGFBP-1 in FGR. Phosphoglycerate kinase-IGFBP-1 mice express IGFBP-1 ubiquitously during development and adulthood (22). These mice are smaller than normal, despite low circulating levels of IGFBP-1, suggesting that the growth impairment was a result of paracrine actions of IGFBP-1. -1 antitrypsin-IGFBP-1 mice show late-gestation and postnatal liver-specific expression, but the growth of these mice is only mildly restricted, likely because levels of circulating IGFBP-1 reach a maximum of only 25 ng/ml in the perinatal period (23), which is significantly lower than the approximately 450 ng/ml in our AFP-BP1 mice. In addition, -1 antitrypsin-IGFBP-1 females have severe reproductive problems, including impaired fertilization and fecundity, despite a lack of transgene expression in the ovary or uterus. In contrast, the AFP-BP1 mice demonstrated no reproductive abnormalities. These findings indicate that high levels of circulating IGFBP-1 in themselves do not cause reproductive problems, as previously speculated (23).
In a more recent study in which the native hIGFBP-1 promoter was used to express hIGFBP-1 in the fetal liver and/or maternal decidua in transgenic mice (21), the animals exhibited a small transient decrease in fetal weight at mid-gestation. Fetal and neonatal serum levels of hIGFBP-1 were not measured, and the adult levels were lower than in the AFP-BP1 mice. By using the native hIGFBP-1 promoter, levels of hIGFBP-1 in the serum fluctuated with nutritional status (44), suggesting that the IGFBP-1 levels in the fetal circulation were likely lower and more variable, possibly accounting for the differences in growth phenotype. Despite the differences in these transgenic models, overall, these studies confirm the importance of appropriate levels of IGFBP-1 in fetal circulation and tissues for proper growth before birth.
The AFP-BP1 mice remained significantly small even as adults, which may be because of the initial FGR or the continuing postnatal transgene expression. The percentage decrease in weight in AFP-BP1 mice was similar at 8 wk of age to that in newborns, indicating no catch-up growth or alteration in the growth trajectory set at birth after FGR. FGR is one of the main causes of growth deficits in childhood with up to 20% of FGR children remaining pathologically short as adults (4). However, the clinical data associating IGFBP-1 levels and postnatal growth is not as definitive as in FGR. Children with idiopathic short stature have been shown to have elevated serum levels of IGFBP-1 and reduced levels of free IGF-I compared with appropriately grown age-matched controls (45, 46). However, in another study, there were no differences in IGFBP-1 in two groups of IUGR children, one that demonstrated catch-up growth and one that did not (47). Similarly, previous models of IGFBP-1 overexpression in transgenic mice have also demonstrated variable changes in postnatal growth, including decreased birth weight with and without catch-up growth and normal birth weight with decreased postnatal growth (21, 22, 23). Again, the differences in postnatal growth in these studies and our AFP-BP1 mice are likely a result of the differences in transgene expression patterns and circulating levels of IGFBP-1 as well as local concentrations in specific organs. These differences highlight the complexity of the role of IGFBP-1 in controlling the bioavailability of the IGFs in the circulation and at the cellular level, both pre- and postnatally. Because the mice in our study continue to produce elevated levels of hIGFBP-1 postnatally, as do some groups of IUGR children, it would appear that elevated postnatal IGFBP-1 may also be a potential causal factor in idiopathic short stature.
Liver was chosen as the site for the transgene expression in our study, because IGFBP-1 is expressed only in this organ in the developing fetus (25, 48). The transgene was also expressed in the yolk sac endoderm, and this likely contributed to some portion of the circulating hIGFBP-1 in the fetus, through the vitelline circulation. However, hIGFBP-1 was not expressed at high levels in other embryonic tissues, thereby removing possible paracrine effects of IGFBP-1 on organ development, as seen in previous models of IGFBP-1 overexpression (22). Although transgenic mRNA was unexpectedly expressed in the brain, no hIGFBP-1 protein was detected by immunohistochemistry. However, very low levels of hIGFBP-1 were detected in the neonatal brain by Western blotting. Because no cellular-specific hIGFBP-1 was detected by immunohistochemistry, it is likely that this low level detected by Western blotting originated from residual plasma present in the brain at the time of organ collection. Therefore, the levels of protein produced in the brain itself are extremely low, below the sensitivity of our immunohistochemical assay.
Transgenic mRNA expression in the liver peaked in the perinatal period, and ELISA measurements indicate that circulating hIGFBP-1 levels were highest at the same time, declining with postnatal age. This expression profile is consistent with the anticipated spatiotemporal activity of the AFP promoter in the transgene (26). However, based on the known expression profile of AFP (26), we expected the transgene expression to decline more rapidly postnatally. Why it did not is currently unclear, but because this occurred in both transgenic lines, it would appear that this expression profile is intrinsic to the promoter sequence itself or the expression in relation to the IGFBP-1 coding sequence. Studies are currently underway to determine whether expression continues past 8 wk.
Plasma levels of total hIGFBP-1 in these mice are in the pathological range detected in FGR humans and significantly greater than the concentrations seen in normally grown human term newborns of 40–90 ng/ml (11, 13, 36). Posttranslational phosphorylation increases the affinity of IGFBP-1 for the IGFs up to six times (49). The majority of IGFBP-1 in healthy adult human plasma is in a single phosphorylated isoform. In contrast, in fetal serum, amniotic fluid, and serum from pregnant women, the proportion of nonphosphorylated IGFBP-1 is much higher (50). It has been suggested that mouse kinases are unable to phosphorylate hIGFBP-1 (51). However, because the hIGFBP-1 in our study was partially phosphorylated, this speculation may not be correct. Interestingly, the percentage of phosphorylated hIGFBP-1 increased with advancing postnatal age, as seen in human neonates. Whether the presence of nonphosphorylated hIGFBP-1 in the fetal serum of these AFP-BP1 mice is a result of a conflict with mouse kinases or represents a phenomenon similar to that seen in human fetal serum is currently being studied. The binding capacity of phosphorylated vs. nonphosphorylated hIGFBP-1 in these transgenic mice would provide important information regarding the role of these isoforms during FGR.
Overall, our studies clearly demonstrate a causal relationship between elevated levels of circulating IGFBP-1 and FGR. IGFBP-1 likely causes restricted fetal growth by limiting the mitogenic actions of the IGFs during periods of hypoxia, undernutrition, or placental insufficiency. These AFP-BP1 mice will allow us to better define the factors contributing to impaired in utero growth as well as to determine the role that alterations in the fetal IGF system play in the fetal origins of adult disease.
Acknowledgments
We thank Dr. Robert Baxter, University of Sydney, Australia, for his generous gift of the polyclonal anti-hIGFBP-1 antibody and Dr. Shirley Tilghman, Princeton University, for her generation donation of the AFP promoter as well as Karen Nygard, Dawn Bryce, Michelle Peres, Morgan McCluskey, and Anthony Cheng for their technical assistance.
Footnotes
This work was supported by a grant to V.K.M.H. from the Canadian Institutes of Health Research (CIHR) and Canada Research Chairs Program and fellowship support to C.S.W. from the CIHR and Wyeth, Canada.
First Published Online November 17, 2005
Abbreviations: AFP, -Fetoprotein; E12.5, embryonic d 12.5; FGR, fetal growth restriction; IGFBP, IGF binding protein; IUGR, intrauterine growth restricted; SSC, standard saline citrate; WT, wild type.
Accepted for publication November 9, 2005.
References
Garite TJ, Clark R, Thorp JA 2004 Intrauterine growth restriction increases morbidity and mortality among premature neonates. Am J Obstet Gynecol 340:1234–1238
Lackman F, Capewell V, Richardson B, daSilva O, Gagnon R 2001 The risks of spontaneous preterm delivery and perinatal mortality in relation to size at birth according to fetal versus neonatal growth standards. Am J Obstet Gynecol 84:946–953
Bernstein I, Horbar J, Badger G, Ohlsson A, Golan A 2000 Morbidity and mortality among very-low-birth-weight neonates with intrauterine growth restriction. Am J Obstet Gynecol 182:198–206
Karlberg J, Albertsson-Wikland K 1995 Growth in full-term small-for-gestational-age infants: from birth to final height. Pediatr Res 38:733–739
Barker DJ, Winter PD, Osmond C, Margetts B, Simmonds SJ 1989 Weight in infancy and death from ischaemic heart disease. Lancet 2:577–580
Barker DJ, Hales CN, Fall CH, Osmond C, Phipps K, Clark PM 1993 Type 2 (non-insulin-dependent) diabetes mellitus, hypertension and hyperlipidaemia (syndrome X): relation to reduced fetal growth. Diabetologia 36:62–67
DeChiara TM, Efstratiadis A, Robertson EJ 1990 A growth-deficiency phenotype in heterozygous mice carrying an insulin-like growth factor II gene disrupted by targeting. Nature 345:78–80
Liu JP, Baker J, Perkins AS, Robertson EJ, Efstradiadis A 1993 Mice carrying null mutations of the genes encoding insulin-like growth factor I (Igf-1) and type 1 IGF receptor (Igf1r). Cell 75:59–72
Shimisaki S, Ling N 1991 Identification and molecular characterization of insulin-like growth factor binding proteins (IGFBP-1, -2, -3, -4, -5, and -6). Prog Growth Factor Res 3:243–266
Rajaram S, Baylink DJ, Mohan, S 1997 Insulin-like growth factor-binding proteins in serum and other biological fluids: regulation and functions. Endocr Rev 18:801–831
Wang HS, Lim J, English J, Irvine L, Chard T 1991 The concentration of insulin-like growth factor-1 and insulin-like growth factor-binding protein-1 in human umbilical cord serum at delivery: relation to fetal weight. J Endocrinol 129:459–464
Leger J, Oury JF, Noel M, Baron S, Benali P, Blot P, Czernichow P 1996 Growth factors and intrauterine growth retardation. I. Serum growth hormone, insulin-like growth factor (IGF)-I, IGF-II, and IGF binding protein 3 in normally grown and growth-retarded human fetuses during the second half of gestation. Pediatr Res 40:94–100
Verhaeghe J, Billen J, Giudice LC 2001 Insulin-like growth factor-binding protein-1 in umbilical artery and vein of term fetuses with signs suggestive of distress during labor. J Endocrinol 170:585–590
Straus DS, Ooi GT, Orlowski CC, Rechler MM 1991 Expression of the genes for insulin-like growth factor-I (IGF-I), IGF-II and IGF-binding proteins-1 and -2 in fetal rats under conditions of intrauterine growth restriction caused by maternal fasting. Endocrinology 128:518–525
Iwamoto HS, Murray MA, Chernausek SD 1992 Effects of acute hypoxemia on insulin-like growth factors and their binding proteins in fetal sheep. Am J Physiol 263:E1151–E1156
McLellan KC, Hooper SB, Bocking AD, Delhanty PJ, Phillips ID, Hill DJ, Han VK 1992 Prolonged hypoxia induced by the reduction of maternal uterine blood flow alters insulin-like growth factor-binding protein-1 (IGFBP-1) and IGFBP-2 gene expression in the ovine fetus. Endocrinology 131:1619–1628
Unterman TG, Simmons RA, Glick RF, Ogata RS 1993 Circulating levels of insulin, insulin-like growth factor-I (IGF-I), IGF-II, and IGF-binding proteins in the small for gestational age fetal rat. Endocrinology 132:327–336
Tapanainen PJ, Bang P, Wilson K, Unterman TG, Vreman HJ, Rosenfeld RG 1994 Maternal hypoxia as a model for intrauterine growth retardation: effects on insulin-like growth factors and their binding proteins. Pediatr Res 36:152–158
Cianfarani S, Germani D, Rossi P, Rossi L, Germani A, Ossicini C, Zuppa A, Argiro G, Holly JMP, Branca F 1998 Intrauterine growth retardation: evidence for the activation of the insulin-like growth factor (IGF)-related growth-promoting machinery and the presence of a cation-independent IGF binding protein-3 proteolytic activity by two months of life. Pediatr Res 44:374–380
Lo HC, Tsao LY, Hsu WY, Chen HN, Yu WK, Chi CY 2002 Relation of cord serum levels of growth hormone, insulin-like growth factors, insulin-like growth factor binding proteins, leptin, and interleukin-6 with birth weight, birth length, and head circumference in term and preterm neonates. Nutrition 18:604–608
Crossey PA, Claire CP, Miell JP 2002 Altered placental development and intrauterine growth restriction in IGF binding protein-1 transgenic mice. J Clin Invest 110:411–418
Rajkumar K, Barron D, Lewitt M, Murphy LJ 1995 Growth retardation and hyperglycemia in insulin-like growth factor binding protein-1 transgenic mice. Endocrinology 136:4029–4034
Gay E, Seurin D, Babajko S, Doublier S, Cazillis M, Binoux M 1997 Liver-specific expression of human insulin-like growth factor binding protein-1 in transgenic mice: repercussions on reproduction, ante- and perinatal mortality and postnatal growth. Endocrinology 138:2937–2947
Dai Z, Xing Y, Boney CM, Clemmons DR, D’Ercole AJ 1994 Human insulin-like growth factor-binding protein-1 (hIGFBP-1) in transgenic mice: characterization and insights into the regulation of IGFBP-1 expression. Endocrinology 135:1316–1327
van Kleffens M, Groffen CAH, Dits NFJ, Lindenbergh-Kortleve DJ, Schuller AGP, Bradshaw SL, Pintar JE, Zwarthoff EC, Drop SLS, van Neck JW 1999 Generation of antisera to mouse insulin-like growth factor binding proteins (IGFBP)-1 to -6: comparison of IGFBP protein and messenger ribonucleic acid localization in the mouse embryo. Endocrinology 140:5944–5952
Chen H, Egan JO, Chiu JF 1997 Regulation and activities of -fetoprotein. Crit Rev Eukaryot Gene Express 7:11–41
Godbout R, Ingram RS, Tilghman SM 1988 Fine-structure mapping of the three mouse -fetoprotein gene enhancers. Mol Cell Biol 8:1169–1178
Godbout R, Tilghman SM 1988 Configuration of the -fetoprotein regulatory domain during development. Genes Dev 2:949–956
Ramesh TM, Ellis AW, Spear BT 1995 Individual mouse -fetoprotein enhancer elements exhibit different patterns of tissue-specific and hepatic position-dependent activities. Mol Cell Biol 15:4947–4955
Sambrook J, Russell DW 2001 Molecular cloning: a laboratory manual. 3rd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press
Southern EM 1975 Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Biol 98:503–517
Han VKM, Basset N, Walton J, Challis JRG 1996 The expression of insulin-like growth factor (IGF) and IGF binding protein (IGFBP) genes in the human placenta and membranes: evidence for IGF-IGFBP interactions at the feto-maternal interface. J Clin Endocrinol Metab 81:2680–2693
Baxter RC, Martin JL, Wood MH 1987 Two immunoreactive binding proteins for insulin-like growth factors in human amniotic fluid: relationship to fetal maturity. J Clin Endocrinol Metab 65:423–431
Bradshaw SL, D’Ercole AJ, Han VKM 1999 Overexpression of insulin-like growth factor-binding protein-2 in C6 glioma cells results in conditional alteration of cellular growth. Endocrinology 140:575–584
Khosravi MJ, Diamandi A, Mistry J 1997 Immunoassay of insulin-like growth factor binding protein-1. Clin Chem 43:523–532
Giudice LC, de Zegher F, Gargosky SE, Dsupin BA, de las Fuentes L, Crystal RA, Hintz RL, Rosenfeld RG 1995 Insulin-like growth factors and their binding proteins in the term and preterm human fetus and neonate with normal and extremes of intrauterine growth. J Clin Endocrinol Metab 80: 1548–1555
Seurin D, Froment P, Bluet-Pajot MT, Epelbaum J, Monget P, Binoux M 2002 Functional alteration of the somatotrophic axis in transgenic mice with liver-specific expression of human insulin-like growth factor binding protein-1. Pediatr Res 52:168–174
Lee PD, Giudice LC, Conover CA, Powell DR 1997 Insulin-like growth factor binding protein-1: recent findings and new directions. Proc Soc Exp Biol Med 216:319–357
Constancia M, Hemberger M, Hughes J, Dean W, Ferguson-Smith A, Fundele R, Stewart F, Kelsey G, Fowden A, Sibley C, Reik W 2002 Placental-specific IGF-II is a major modulator of placental and fetal growth. Nature 417:945–948
Sibley CP, Coan PM, Ferguson-Smith AC, Dean W, Hughes J, Smith P, Reik W, Burton GJ, Fowden AL, Constancia M 2004 Placental-specific insulin-like growth factor 2 (Igf2) regulates the diffusional exchange characteristics of the mouse placenta. Proc Natl Acad Sci USA 101:8204–8208
Phipps K, Barker DJ, Hales CN, Fall CH, Osmond C, Clark PM 1993 Fetal growth and impaired glucose tolerance in men and women. Diabetologia 36:225–228
Martyn CN, Marker DJP, Osmond C 1996 Mothers’ pelvic size, fetal growth, and death from stroke and coronary heart disease in men in the UK. Lancet 348:1264–1268
Moore VM, Cockington RA, Ryan P, Robinson JS 1999 The relationship between birth weight and blood pressure amplifies from childhood to adulthood. J Hypertens 17:883–888
Crossey PA, Jones JS, Miell JP 2000 Dysregulation of the insulin/IGF binding protein-1 axis in transgenic mice is associated with hyperinsulinemia and glucose intolerance. Diabetes 49:457–465
Lindgren BF, Segovia B, Lassarre C, Binoux M, Gourmelen M 1996 Growth retardation in constitutionally short children is related both to low serum levels of insulin-like growth factor-I and to its reduced bioavailability. Growth Regul 6:158–164
Kamoda T, Saitoh H, Hirano T, Matsui A 2000 Serum levels of free insulin-like growth factor (IGF)-I and IGF-binding protein-1 in prepubertal children with idiopathic short stature. Clin Endocrinol 53:683–688
Cianfarani S, Geremia C, Scott CD, Germani D 2002 Growth, IGF system, and cortisol in children with intrauterine growth retardation: is catch-up growth affected by reprogramming of the hypothalamic-pituitary-adrenal axis Pediatr Res 51:94–99
Braulke T, Gotz W, Claussen M 1996 Immunohistochemical localization of insulin-like growth factor binding protein-1, -3 and -4 in human fetal tissues and their analysis in media from fetal tissue explants. Growth Regul 6:55–65
Jones JI, D’Ercole AJ, Camacho-Hubner C, Clemmons DR 1991 Phosphorylation of insulin-like growth factor (IGF)-binding protein-1 in cell culture and in vivo: effects on affinity for IGF-I. Proc Natl Acad Sci USA 88:7481–7485
Westwood M, Gibson JM, Davies AJ, Young RJ, White A 1994 The phosphorylation pattern of insulin-like growth factor binding protein-1 in normal plasma is different from that in amniotic fluid and changes during pregnancy. J Clin Endocrinol Metab 79:1735–1741
Sakai K, D’Ercole AJ, Murphy LJ, Clemmons DR 2001 Physiological differences in insulin-like growth factor binding protein-1 (IGFBP-1) phosphorylation in IGFBP-1 transgenic mice. Diabetes 50:32–38(Carole S. Watson, Peter Bialek, Makoto A)
London Regional Cancer Program (P.B., S.-P.Y.), University of Western Ontario, London, Ontario, Canada, N6A 4L6
Division of Pediatric Endocrinology (M.A.), David Geffen School of Medicine at the University of California, Los Angeles, Los Angeles, California 90095
Diagnostic Systems Laboratories (Canada) Inc. (J.K.), Department of Laboratory Medicine and Pathobiology, Faculty of Medicine, University of Toronto and Mount Sinai Hospital, Toronto, Ontario, Canada, M5G 1X5
Abstract
IGF binding protein-1 (IGFBP-1) inhibits the mitogenic actions of the IGFs. Circulating IGFBP-1 is elevated in newborns and experimental animals with fetal growth restriction (FGR). To establish a causal relationship between high circulating IGFBP-1 and FGR, we have generated transgenic mice using the mouse -fetoprotein gene promoter to target overexpression of human IGFBP-1 (hIGFBP-1) in the fetal liver. These transgenic mice (AFP-BP1) expressed hIGFBP-1 mainly in the fetal hepatocytes, starting at embryonic d 14.5 (E14.5), with lower levels in the gut. The expression peaked at 1 wk postnatally (plasma concentration, 474 ± 34 ng/ml). At birth, AFP-BP1 pups were 18% smaller [weighed 1.34 ± 0.02 g compared with 1.62 ± 0.04 g for wild type (WT); P < 0.05], and they did not demonstrate any postnatal catch-up growth. The placentas of the AFP-BP1 mice were larger than WT from E16.5 onwards (150 ± 12 for AFP-BP1 vs. 100 ± 5 mg for WT at E16.5; P < 0.05). Thus, this model of FGR is associated with a larger placenta, but without postnatal catch-up growth. Overall, these data clearly demonstrate that high concentrations of circulating IGFBP-1 are sufficient to cause FGR.
Introduction
FETAL GROWTH IS a complex process controlled by many factors, including the genetics of mother and fetus as well as maternal and environmental factors. Clinically important impaired fetal growth resulting in fetal growth restriction (FGR) is often caused by a poor in utero environment resulting from maternal or placental pathology or abnormal physiology. FGR is associated with significant perinatal and neonatal mortality and morbidity, including hypoxic injury to vital organs and preterm delivery (1, 2, 3). FGR pregnancies result in small for gestational age newborns who are at high risk for negative effects on childhood growth and neurodevelopment. Only 80% of FGR newborns demonstrate catch-up growth during early infancy and achieve adult height within normal percentiles, whereas 20% remain small throughout life (4). Furthermore, adaptations to a hostile in utero environment resulting in small birth size increase the risk of cardiovascular and metabolic diseases as adults (5, 6), a process known as developmental programming.
The IGFs (IGF-I and -II) are essential for cell growth and differentiation during development, such that mice carrying null mutations of either the Igf1 or Igf2 genes weigh 60% of normal at birth (7, 8). IGF-I and -II are produced in most tissues such that they have both endocrine and paracrine actions. The interactions of the IGFs with their receptors are modulated by six species of high-affinity IGF binding proteins (IGFBP-1 to –6), which regulate the bioavailability of IGFs through protection of the IGFs from degradation in circulation and inhibition of IGF binding to cell surface receptors (9, 10).
There is mounting evidence that circulating IGFBP-1 plays a crucial role in the pathogenesis of FGR. In humans, birth weight is negatively correlated with IGFBP-1 concentrations in cord blood (11, 12, 13). In animal models of FGR, including uterine artery ligation, hypoxia, and maternal undernutrition, low birth weight is also associated with high circulating levels of IGFBP-1 as a result of increased production in the fetal liver (14, 15, 16, 17, 18, 19, 20). However, the elevated IGFBP-1 levels occur concurrently with reduced circulating IGF-I concentrations and a diminished supply of nutrients and substrates; thus the role of IGFBP-1 in FGR is unclear. Previous investigators have produced transgenic mice that overexpress IGFBP-1 under the control of various promoters to examine the role of IGFBP-1 in the mitogenic actions of the IGFs in vivo. Although some of these models demonstrated modest reductions in growth, these experiments were not necessarily designed to specifically examine the relationship between elevated circulating fetal IGFBP-1 and fetal growth. For example, in some models, overexpression was not targeted to the liver, whereas in others, levels of circulating IGFBP-1 were relatively low (21, 22, 23, 24). Therefore, the objective of this study was to determine the growth phenotype of transgenic mice (AFP-BP1) designed to overexpress IGFBP-1 specifically in the fetal liver with resultant elevated circulating IGFBP-1 concentrations, without directly altering levels of IGF-I or disturbing the nutrient and/or substrate supply to the fetus. To achieve this goal, we generated transgenic mice in which the regulatory region of the mouse -fetoprotein (AFP) gene was used to direct overexpression of a human IGFBP-1 cDNA in the fetal liver. Liver was chosen as the site of transgene expression because in the fetus, IGFBP-1 is expressed only in the liver, and it is the predominant source of circulating IGFBP-1 (25). Because AFP is also expressed mainly in the fetal liver and its expression declines after birth (26), we anticipated that the AFP-BP1 mice would have elevated levels of circulating human IGFBP-1 (hIGFBP-1) from mid-gestation to early postnatal life associated with a reduction in overall fetal growth.
Materials and Methods
Transgene construction
A hIGFBP-1 cDNA clone was generated by RT-PCR using human decidual RNA as the template. Primers were designed to amplify an 841-bp cDNA containing the coding sequence of hIGFBP-1 (Table 1). The resulting amplicon was inserted into the pGEM-T vector (Promega Corp, Madison, WI). The hIGFBP-1 cDNA was then subcloned into the NotI site of a pSP72 vector containing the 7.6-kb upstream regulatory region of the mouse AFP gene (a generous gift from S. Tilghman, Princeton University, Princeton, NJ), which is sufficient to target high levels of expression in the fetal liver (26, 27, 28, 29). Finally, the simian virus 40 (SV40) intronic-polyadenylation sequence obtained from the SV40neo plasmid (805 bp) was inserted into the XhoI site of the AFP-hIGFBP-1-pSP72 plasmid, downstream from the hIGFBP-1 cDNA, to increase the efficiency of in vivo transgene expression. The final size of the transgene (AFP-BP1) was 9.2 kb (Fig. 1).
Generation of transgenic mice
The transgene was released from the plasmid by EcoRI digestion, fractionated by agarose gel electrophoresis, and then electroeluted and purified by Elutip-d columns according to manufacturer instructions (Schleicher & Schuell, Dassel, Germany). The purified fragment was resuspended in 10 mM Tris (pH 7.5) and 0.25 mM EDTA and microinjected into the pronuclei of one-cell embryos obtained from mating between (C57BL6/CBA)F1 animals. Founder animals were bred to homozygosity and backcrossed into CBA, the wild-type (WT) strain, for at least four generations. For all experiments, CBA WT animals were used as control.
Animals and sample collection
The University of Western Ontario Committee on Animal Care, following the guidelines of the Canadian Council on Animal Care, approved all protocols. Mice were bred and maintained in the animal facility at the Lawson Health Research Institute. They were fed standard lab chow and maintained on a 12-h light, 12-h dark cycle. Postnatal animals were weighed daily at 0900 h from postnatal d 1–14 and then weekly. Fetuses were collected and weighed from embryonic d 12.5 (E12.5) to E18.5. Animals were killed with an ip overdose of sodium pentobarbital (Euthanyl; MTC Pharmaceuticals, Cambridge, Ontario, Canada) and blood samples collected by intracardiac puncture into heparinized capillary tubes (Chase Scientific Inc., Rockwood, TN). After centrifugation, plasma was removed and stored at –20 C. Organs were quickly dissected, weighed, and either snap-frozen in liquid nitrogen and stored at –80 C for RNA extraction or placed in 4% paraformaldehyde (BMD Science, Mississauga, Ontario, Canada) followed by blocking in paraffin for in situ hybridization analysis or immunohistochemistry.
Extraction of genomic DNA, Southern blot, and PCR analyses
Tail biopsies were collected postnatally at 2 wk and digested overnight with 800 U proteinase K (Invitrogen Life Technologies, Carlsbad, CA) in 20 mM Tris-HCl (pH 8.0), 10 mM EDTA, 10 mM NaCl, and 0.5% SDS at 55 C. Genomic DNA was then extracted using phenol-chloroform and precipitated in 100% ethanol (30).
Southern blot analysis was performed to identify founder mice, for the determination of copy number, and zygosity of the offspring. Briefly, 10 μg genomic DNA was digested for 6 h with 50 U PstI (Invitrogen) at 37 C. The DNA fragments were then fractionated by agarose gel electrophoresis and transferred onto nylon membrane (Hybond-N; Amersham-Pharmacia Biotech Canada, Baie d’Urfe, Quebec, Canada) in 10x standard saline citrate (SSC), and the membranes baked for 2 h at 80 C (31). The blots were prehybridized for at least 2 h at 65 C in 0.5 M phosphate buffer containing 7% SDS and 100 μg/ml sonicated salmon sperm DNA. Blots were then hybridized overnight at 65 C with a radiolabeled DNA fragment (2 x 106 cpm/ml) corresponding to the SV40 intronic-polyadenylation sequence in the transgene. The probe was labeled with [-32P]dCTP by random priming (30) (Amersham-Pharmacia). After hybridization, blots were washed twice for 15 min each in 2x SSC, 0.1% SDS at room temperature and twice for 15 min each in 0.1x SSC, 0.1% SDS at 65 C and then exposed to film (Biomax MR; Kodak Canada, Toronto, Ontario, Canada) at –80 C. Signal intensity of bands on the autoradiograms was analyzed with a densitometry program (Phoretix Software; Nonlinear Dynamics Ltd., Newcastle upon Tyne, UK).
After transgenic lines were verified by Southern blot analysis, subsequent transgenic offspring were confirmed by PCR analysis. Primers were designed to amplify the nonmurine intronic-SV40 polyadenylation sequence of the transgene (Table 1). Thirty cycles were performed in the PCR, with an annealing temperature of 60 C and 5 mM Mg2+.
RNA isolation and Northern blotting
Total RNA was extracted from up to 100 mg frozen tissue using the Trizol method (Invitrogen Life Technologies) with precipitation in isopropanol overnight at –80 C. Twenty micrograms of total RNA were loaded into 1% agarose/16% formaldehyde gels and electrophoresed overnight in 1x 3[N-morpholino]propanesulfonic acid. The RNA was then transferred to Zetaprobe membrane (Bio-Rad Laboratories Canada, Mississauga, Ontario, Canada) in 10x SSC and the membranes baked at 80 C for 2 h. Membranes were prehybridized for 2 h at 42 C in 50% deionized formamide, 1x saline sodium phosphate EDTA, 7% SDS, and 100 μg/ml sonicated salmon sperm DNA. The blots were then hybridized overnight at 42 C in the same buffer, with a [-32P]dCTP-labeled SV40 polyadenylation sequence probe (1 x 106 cpm/ml). After hybridization, the blots were washed twice for 15 min each in 1x SSC, 0.1% SDS at 42 C; once for 30 min in 1x SSC, 0.1% SDS at 65 C; and twice for 30 min each in 0. 1x SSC, 0.1% SDS at 65 C and then exposed to film at –80 C (Biomax MS; Kodak Canada). Consistency of total RNA loading and transfer was verified by probing the blots for 18S ribosomal RNA using the same conditions. Autoradiograms were quantified using Phoretix software (Nonlinear Dynamics, Ltd.).
In situ hybridization
In situ hybridization was performed as described previously (32) on 5-μm sections taken from AFP-BP1 and WT animals at various embryonic and neonatal time points, and individual organs collected from postnatal animals. Briefly, sections were deparaffinized, rehydrated, and incubated in 0.2% Triton for 1 h at room temperature followed by proteinase K (0.2 U/ml) at 37 C for 30 min and finally in 0.1 M triethanolamine for 10 min at room temperature. After subsequent dehydration, sections were prehybridized for 2 h at 45 C in prehybridization buffer [50% formamide, 0.3 M NaCl, 20 mM Tris (pH 8.0), 1 mM EDTA, 1x Denhardt’s, 500 μg/ml yeast tRNA, 0.1% SDS, 100 mM dithiothreitol, and 100 μg/ml salmon sperm DNA]. Sections were then hybridized with 35S-labeled antisense and sense cRNA probes at 55 C overnight in hybridization buffer (prehybridization buffer, plus 10% dextran sulfate). The probes were generated in our laboratory, using the SV40neo polyadenylation sequence of the transgene cloned into pGEM-T as a template so as to achieve hybridization only to the transgenic mRNA. After hybridization, slides were washed at a maximum stringency of 0.1x SSC at 65 C for 20 min. Sections collected from WT animals were analyzed as negative controls. In addition, sections were also probed with sense probes to verify the specificity of the antisense cRNA probes.
Immunohistochemistry
Immunohistochemistry was performed on 5-μm sections taken from whole AFP-BP1 and WT animals at embryonic and newborn time points, and individual organs collected from postnatal animals. After deparaffinization and rehydration, immunohistochemistry was performed using the Vectastain Elite Kit (Vector Laboratories, Burlington, Ontario, Canada). Briefly, sections were blocked with normal goat serum for 30 min at room temperature and then incubated overnight at 4 C with a polyclonal anti-hIGFBP-1 antibody (33) at a concentration of 1:5000. Protein was visualized using diaminobenzidine and counterstained with methyl green.
Western immunoblotting and ligand blotting
The presence of hIGFBP-1 in plasma and tissue extracts was determined by immunoblotting using a specific anti-hIGFBP-1 antibody (33), as previously described (34). The profile and levels of the other IGFBPs were determined by ligand blot analysis. Briefly, plasma samples (5 μl) or tissue extracts (60 μg total protein, as determined by BCA analysis) were boiled in Laemmli sample buffer [0.15 M Tris-HCl (pH 6.8), 4.8% SDS, 24% glycerol, and 0.024% bromophenol blue] for 20 min and separated in 12% SDS-PAGE under nonreducing conditions. Proteins were transferred electrophoretically to 0.2-μm nitrocellulose membranes (Schleicher & Schuell, Dassel, Germany) and then blocked in buffer BT [10 mM Tris-HCl (pH 7.4), 150 mM NaCl, 0.2% Tween 20] plus 1% BSA for 1 h at room temperature. The membranes were then washed three times for 10 min each in buffer BT and incubated with biotinylated IGF-I (10 ng/ml) for 3 h at room temperature. After washing, the membranes were incubated with streptavidin-horseradish peroxidase (Amersham-Pharmacia) (1:1500) for 45 min at room temperature. The proteins were visualized using chemiluminescence (Western Lightning; Perkin-Elmer Life Sciences, Inc., Boston, MA) and exposed to X-OMAT Blue autoradiography film (Kodak), and band intensity was analyzed using Phoretix Software (Nonlinear Dynamics Ltd.).
hIGFBP-1 ELISA
Concentrations of total hIGFBP-1 in plasma were measured using a total hIGFBP-1 ELISA (Diagnostic Systems Laboratories Inc., Webster, TX). Concentrations of nonphosphorylated hIGFBP-1 were measured as previously described (35). Samples were diluted 1:5 and assayed in duplicate. The coefficient of variation between duplicates was less than 5%, and all samples were analyzed in one assay. The antibody was specific for hIGFBP-1 and did not cross-react with murine IGFBP-1 in plasma collected from WT mice.
IGF-I ELISA
The 96-well microtiter plates were coated with capture antibody at 0.5 μg/well in 100 μl PBS (pH 7.2), incubated overnight at room temperature on a shaker, and then washed three times with 300 μl wash buffer (PBS and 0.05% Tween 20) followed by 1 h incubation with 300 μl blocking buffer (PBS, 5% Tween 20, 5% sucrose, and 0.05% sodium azide) and three final washes. Recombinant mouse IGF-I standard (R&D Systems, Minneapolis, MN) was diluted in assay buffer (50 mM sodium phosphate, 150 mM NaCl, 0.1% Tween 20, 0.25% BSA, pH 7.4) in concentrations ranging from 0–25 ng/ml. Before assay, serum samples were extracted with acid/ethanol reagent (12.5% 2 N HCl and 87.5% ethanol), neutralized with 1 M Tris base, and diluted with assay buffer. Standards, controls, or diluted samples (50 μl /well) and 50 ng biotinylated goat antimouse IGF-I antibody (R&D Systems) in assay buffer (50 μl/well) were incubated at room temperature for 2 h on a shaker. The wells were washed three times with wash buffer followed by the addition of streptavidin-horseradish peroxidase conjugate (100 μl/well) (Pierce, Rockford, IL) and additionally incubated for 20 min at room temperature. After four washes, 100 μl o-phenylenediamine dihydrochloride (1 mg/ml in hydrogen peroxide substrate) was added to each well and incubated for an additional 10–20 min. The reaction was stopped by the addition of 50 μl 2 N H2SO4, and the absorbance was determined at 490 nm in a plate reader (Molecular Design, Sunnyvale, CA). The sensitivity of this assay is 0.1 ng/ml. There is no cross-reactivity with mouse IGF-II; mouse IGFBP-1, -2, or -3; human IGF-I or -II; or hIGFBP-1, -2, -3, or -4 (all test substances at 200 ng/ml). The intra- and interassay coefficients of variation were less than 10% in the range from 1–10 ng/ml, and recovery was 91–101%.
Statistical analyses
Data were analyzed for differences between AFP-BP1 and WT animals using two-way ANOVA and unpaired Student’s t tests where applicable.
Results
Southern blot and PCR identification of founder mice
Four founder mice (designated as AFP-BP1 lines 2, 3, 4, and 16) were identified on the basis of Southern blot analysis of genomic DNA obtained from tail biopsies. With a restriction digest using PstI, which cleaved the transgene at a single site (Fig. 1A), Southern blotting detected a major single genomic fragment in these four founders, corresponding to the expected size of 9.2 kb (Fig. 1B). In addition, PstI cleavage produced only one other fragment (data not shown), corresponding to the DNA fragment at the 3' end of the inserted transgene, indicating that there is only a single site of transgene insertion in the genome of these founder mice.
Southern blot analysis was also performed to determine the copy number of the transgene. Lines 2, 3, 4, and 16 carried four, two, one, and eight copies of the transgene, respectively (Fig. 1, D and E). Founder mice were bred with WT littermates, and only lines 2, 3, and 16 produced transgenic offspring. Because lines 16 and 2 contained the highest copy number, they were further characterized to determine transgene expression patterns and any phenotypic alterations. Subsequent analysis of offspring genotype was completed by PCR. Primers (Table 1) were designed to amplify an 800-bp fragment specific to the SV40 intron-polyadenylation sequence of the transgene (Fig. 1C).
Ontogeny of transgene expression: in situ hybridization and Northern blot analysis
In situ hybridization was performed on sagittal sections of both AFP-BP1 and WT mice collected at various embryonic stages and at postnatal d 1 (newborn) and on postnatal liver, brain, and gut samples using an antisense riboprobe specific to the SV40 polyadenylation sequence. Overall, high levels of transgene expression were found in the fetal liver (Fig. 2) and yolk sac (data not shown) and at lower levels in the gut and brain in both lines 16 and 2 (Fig. 2).
Transgenic hIGFBP-1 mRNA was first detected in the liver at E14.5. Transgene expression in the fetal liver increased as gestation progressed, peaking on the day of birth. The transgenic mRNA was present at lower levels in the liver at 8 wk. At all ages, transgenic mRNA was expressed in the hepatocytes only. The expression levels of transgenic mRNA were much lower in the gut and brain compared with the liver at all ages. Expression was limited to epithelial cells in the gut and glial cells in the brain. Expression of the transgene in the yolk sac was also first evident at E14.5 and remained consistent until birth. There was no specific hybridization in any WT tissue examined (Fig. 2).
Northern blot analysis of total RNA obtained from E14.5 liver and gut, brain, and liver from E18.5 onward, showed two main transcripts of 1.6 and 2.1 kb and a low abundant transcript of 1.1 kb. The ontogeny of the transgenic mRNA expression pattern as revealed by Northern blotting was similar to that found with in situ hybridization. In the liver, transgene expression increased from E14.5 to 1 wk (Fig. 3A). Expression in the gut was highest at E18.5 and the newborn period, becoming almost undetectable by 1 wk (Fig. 3B). In contrast, expression in the brain increased from E18.5 to 4 wk (Fig. 3C), but overall the levels were much lower compared with the liver at all ages examined (Fig. 3, C–E).
hIGFBP-1 protein expression: immunohistochemical and Western blot analyses
Immunoreactive hIGFBP-1 was detected at high levels in the perinatal liver and at lower levels in the gut of transgenic mice. Despite the presence of transgenic mRNA in the brain, immunoreactive hIGFBP-1 could not be detected in the transgenic brain using this technique (Fig. 2). There was no antibody cross-reactivity with endogenous IGFBP-1 in sections taken from WT animals.
Plasma samples were obtained from the AFP-BP1 mice, and Western blot analysis using anti-hIGFBP-1 revealed an immunoreactive protein at 27 kDa in the AFP-BP1 plasma, with no cross-reactivity with any protein in WT samples. The level of the immunoreactive hIGFBP-1 decreased in plasma at 4 wk compared with the day of birth (Fig. 4A).
Western ligand blotting using IGF-1 as the probe detected a protein species of 27 kDa in the AFP-BP1 mice, but not WT. A protein species of 29 kDa, representing mouse IGFBP-1, was also visible. There was no difference in band intensity of mouse IGFBP-1 between transgenic and WT samples in newborn plasma. However, at 4 wk, there were significantly lower levels of mouse IGFBP-1 in plasma obtained from AFP-BP1 mice compared with WT animals (Fig. 4, B and C). Protein species of 38–44 kDa, representing IGFBP-3, were also visible. There was significantly less IGFBP-3 in the plasma of AFP-BP1 mice compared with WT animals at both d 1 and 4 wk. There was no difference in levels of circulating IGFBP-4 at either age (Fig. 4, B and C).
Western immunoblot analyses using anti-hIGFBP-1 antibody were also performed to examine protein extracts from brain and liver of AFP-BP1 and WT animals. Overall, these analyses demonstrated the presence of a 27-kDa immunoreactive protein in the AFP-BP1 mice that was not present in WT animals but was present in human amniotic fluid. Immunoreactive hIGFBP-1 was present in the liver of the AFP-BP1 mice at all ages examined. In the brain, immunoreactive IGFBP-1 was present on the day of birth but was not detectable at 4 wk in the AFP-BP1 mice (Fig. 4D).
Circulating hIGFBP-1 concentrations
hIGFBP-1 was detected in the plasma of AFP-BP1 animals at all ages examined. The circulating concentrations of total hIGFBP-1 in plasma were highest in the perinatal period, decreasing significantly by 4 wk of age. The percentage of total hIGFBP-1 that was nonphosphorylated was also highest in the perinatal period and declined postnatally. Levels of circulating hIGFBP-1 in line 2 transgenic mice were lower than that in line 16, but a similar percentage was phosphorylated at d 1 and 8 wk (Table 2). There was no hIGFBP-1 detected by ELISA in WT plasma.
Plasma IGF-I concentrations
Plasma IGF-I concentrations in AFP-BP1 mice were 21 ± 6 ng/ml in the newborn period and 325 ± 48 ng/ml and 336 ± 38 ng/ml at 4 and 8 wk, respectively. These concentrations were not different from those of the WT mice of the same ages (Table 3).
Fetoplacental growth
There were no reproductive abnormalities in the AFP-BP1 mice. Litter size was not altered (8.1 ± 0.5 pups per litter for AFP-BP1 vs. 7.9 ± 1.0 pups per litter for WT), and there was no effect of maternal genotypic background on survival rates.
At E12.5, there was no difference in total body weight between AFP-BP1 and WT fetuses. Starting at E14.5, the line 16 AFP-BP1 fetuses were significantly smaller than WT and remained so until birth (Fig. 5A). Starting at E16.5, the placentas of line 16 AFP-BP1 mice were significantly larger than WT (Fig. 5B), yielding a significantly larger placenta to body weight ratio in late gestation (Fig. 5C).
Postnatal growth
At birth, lines 16 and 2 AFP-BP1 mice weighed 1.33 ± 0.02 and 1.44 ± 0.05 g, respectively, compared with 1.62 ± 0.04 g for WT mice (P < 0.01). Postnatally, line 16 AFP-BP1 mice remained 15–20% smaller than WT at all ages examined. At weaning, line 2 AFP-BP1 mice continued to be small (12.5 ± 0.3 g) but were not significantly different from WT (13.7 ± 0.6 g) at this age. However, at 8 wk of age, line 2 AFP-BP1 mice (22.4 ± 0.5 g) were significantly smaller than WT (24.7 ± 0.8 g) but bigger than line 16 AFP-BP1 mice (20.4 ± 0.6 g) (Fig. 6A). The growth patterns were similar between males and females in both lines. At birth, the weights of male AFP-BP1 mice were 16.2% smaller than WT males, and female AFP-BP1 mice were 18.8% smaller than WT females. By 8 wk of age, male AFP-BP1 mice were 16.2% smaller, whereas females were 18.3% smaller than WT.
Discussion
IGFBP-1 is a potent inhibitor of IGF-I and IGF-II interactions with the IGF-I receptor. In addition, increased circulating IGFBP-1 is a consistent observation in human FGR fetuses and newborns and animal models of FGR (11, 12, 13, 14, 15, 16, 17, 18, 19, 20). However, whether impaired fetal growth is caused by elevated circulating IGFBP-1 has not been clearly established because of changes in IGF-I, nutrients, and substrates that accompany FGR, which could also explain the growth impairment. To establish a causal relationship between IGFBP-1 and FGR, we have generated a model of FGR by increasing circulating IGFBP-1 concentrations in AFP-BP1 mice that overexpress hIGFBP-1 in the fetal liver. AFP-BP1 mice differ from previous models of IGFBP-1 overexpression in the specificity of the transgene expression in the fetal liver, the sustained elevation in circulating IGFBP-1 in fetal life, and the consistent decrease in growth both pre- and postnatally. In addition, there is no concomitant change in circulating IGF-I concentrations in this model of IGFBP-1 overexpression. This study demonstrates clearly that an increase in circulating IGFBP-1 during fetal life, without a direct confounding reduction in circulating IGF-I concentrations and/or direct alteration in nutrients and substrates, is sufficient to cause FGR.
FGR was demonstrated in two AFP-BP1 lines, each carrying a different copy number of the transgene. Moreover, the growth restriction was less in line 2, which contained a fewer number of copies and had a lower level of circulating hIGFBP-1. These data indicate that the changes in growth in the AFP-BP1 mice were caused by the transgene itself and not to positional effects of the transgene integration site in the mouse genome.
Human intrauterine growth-restricted (IUGR) newborns have decreased IGF-I levels, in addition to the elevation in IGFBP-1, possibly contributing to the decreased growth in utero (36). A previous model of transgenic IGFBP-1 overexpression in the postnatal liver demonstrated a decrease in circulating IGF-I (23). The somatotrophic axis was extensively studied in those mice, and the decrease in IGF-I was attributed to a decrease in GH production by the pituitary gland, related to the impaired neurological development seen in these mice (37). In contrast, AFP-BP1 mice demonstrated no difference in circulating IGF-I levels from the newborn period to adulthood at 8 wk of age. Therefore, this would indicate that the growth restriction seen in this transgenic model is likely a result of a reduction in bioavailable IGF-I, not a decrease in total IGF-I, thereby providing strong evidence that the expression of Igfbp-1 in the fetal liver is a major factor in the control of fetal growth. Because oxygen- and insulin-response elements are present in the regulatory sequences of the endogenous Ifgbp-1 gene, hypoxia and low glucose levels would lead to high levels of IGFBP-1 (38). This relationship represents a link between poor placental transfer of nutrients and substrates and FGR. In the AFP-BP1 mice, we were able to separate the increase in circulating IGFBP-1 from a decrease in nutrients and IGF-I, and our results demonstrate that IGFBP-1 itself might be a pivotal control molecule in the pathogenesis of FGR.
A recent study demonstrated that a placental-specific Igf2 null mutation leads to impaired syncytiotrophoblast development and placental transport and subsequent FGR. This study provided the strongest evidence to date for the importance of placental development in fetal growth (39, 40). It is interesting that the growth-restricted AFP-BP1 mice described in our study exhibited an increase in placental weight. Although we have not examined the placental structure in detail, previous studies have demonstrated that an increased placental weight after elevated IGFBP-1 is caused by an increase in the size of the labyrinth zone, the area of nutrient exchange in the mouse placenta (21). These changes may be direct effects of IGFBP-1 on placental growth or may be compensatory changes designed to increase passage of nutrients to the growth-restricted fetus. This may explain, at least in part, the relatively small decrease in fetal size. A high placental to birth weight ratio is associated with increased adult blood pressure in children and adults as well as an increase in the prevalence of coronary heart disease and impaired glucose tolerance (41, 42, 43). The increase in circulating IGFBP-1 seen in growth-restricted fetuses may contribute to altered placental development and subsequent developmental programming for adult-onset diseases. Long-term studies of the AFP-BP1 mice will provide clues to some of the mechanisms underlying the relationship between placental growth and developmental programming.
Transgenic mice overexpressing IGFBP-1 in patterns that differ from the AFP-BP1 mice have been described previously, some of which also exhibited alterations in growth. However, none of these studies were designed to specifically examine the role of elevated IGFBP-1 in FGR. Phosphoglycerate kinase-IGFBP-1 mice express IGFBP-1 ubiquitously during development and adulthood (22). These mice are smaller than normal, despite low circulating levels of IGFBP-1, suggesting that the growth impairment was a result of paracrine actions of IGFBP-1. -1 antitrypsin-IGFBP-1 mice show late-gestation and postnatal liver-specific expression, but the growth of these mice is only mildly restricted, likely because levels of circulating IGFBP-1 reach a maximum of only 25 ng/ml in the perinatal period (23), which is significantly lower than the approximately 450 ng/ml in our AFP-BP1 mice. In addition, -1 antitrypsin-IGFBP-1 females have severe reproductive problems, including impaired fertilization and fecundity, despite a lack of transgene expression in the ovary or uterus. In contrast, the AFP-BP1 mice demonstrated no reproductive abnormalities. These findings indicate that high levels of circulating IGFBP-1 in themselves do not cause reproductive problems, as previously speculated (23).
In a more recent study in which the native hIGFBP-1 promoter was used to express hIGFBP-1 in the fetal liver and/or maternal decidua in transgenic mice (21), the animals exhibited a small transient decrease in fetal weight at mid-gestation. Fetal and neonatal serum levels of hIGFBP-1 were not measured, and the adult levels were lower than in the AFP-BP1 mice. By using the native hIGFBP-1 promoter, levels of hIGFBP-1 in the serum fluctuated with nutritional status (44), suggesting that the IGFBP-1 levels in the fetal circulation were likely lower and more variable, possibly accounting for the differences in growth phenotype. Despite the differences in these transgenic models, overall, these studies confirm the importance of appropriate levels of IGFBP-1 in fetal circulation and tissues for proper growth before birth.
The AFP-BP1 mice remained significantly small even as adults, which may be because of the initial FGR or the continuing postnatal transgene expression. The percentage decrease in weight in AFP-BP1 mice was similar at 8 wk of age to that in newborns, indicating no catch-up growth or alteration in the growth trajectory set at birth after FGR. FGR is one of the main causes of growth deficits in childhood with up to 20% of FGR children remaining pathologically short as adults (4). However, the clinical data associating IGFBP-1 levels and postnatal growth is not as definitive as in FGR. Children with idiopathic short stature have been shown to have elevated serum levels of IGFBP-1 and reduced levels of free IGF-I compared with appropriately grown age-matched controls (45, 46). However, in another study, there were no differences in IGFBP-1 in two groups of IUGR children, one that demonstrated catch-up growth and one that did not (47). Similarly, previous models of IGFBP-1 overexpression in transgenic mice have also demonstrated variable changes in postnatal growth, including decreased birth weight with and without catch-up growth and normal birth weight with decreased postnatal growth (21, 22, 23). Again, the differences in postnatal growth in these studies and our AFP-BP1 mice are likely a result of the differences in transgene expression patterns and circulating levels of IGFBP-1 as well as local concentrations in specific organs. These differences highlight the complexity of the role of IGFBP-1 in controlling the bioavailability of the IGFs in the circulation and at the cellular level, both pre- and postnatally. Because the mice in our study continue to produce elevated levels of hIGFBP-1 postnatally, as do some groups of IUGR children, it would appear that elevated postnatal IGFBP-1 may also be a potential causal factor in idiopathic short stature.
Liver was chosen as the site for the transgene expression in our study, because IGFBP-1 is expressed only in this organ in the developing fetus (25, 48). The transgene was also expressed in the yolk sac endoderm, and this likely contributed to some portion of the circulating hIGFBP-1 in the fetus, through the vitelline circulation. However, hIGFBP-1 was not expressed at high levels in other embryonic tissues, thereby removing possible paracrine effects of IGFBP-1 on organ development, as seen in previous models of IGFBP-1 overexpression (22). Although transgenic mRNA was unexpectedly expressed in the brain, no hIGFBP-1 protein was detected by immunohistochemistry. However, very low levels of hIGFBP-1 were detected in the neonatal brain by Western blotting. Because no cellular-specific hIGFBP-1 was detected by immunohistochemistry, it is likely that this low level detected by Western blotting originated from residual plasma present in the brain at the time of organ collection. Therefore, the levels of protein produced in the brain itself are extremely low, below the sensitivity of our immunohistochemical assay.
Transgenic mRNA expression in the liver peaked in the perinatal period, and ELISA measurements indicate that circulating hIGFBP-1 levels were highest at the same time, declining with postnatal age. This expression profile is consistent with the anticipated spatiotemporal activity of the AFP promoter in the transgene (26). However, based on the known expression profile of AFP (26), we expected the transgene expression to decline more rapidly postnatally. Why it did not is currently unclear, but because this occurred in both transgenic lines, it would appear that this expression profile is intrinsic to the promoter sequence itself or the expression in relation to the IGFBP-1 coding sequence. Studies are currently underway to determine whether expression continues past 8 wk.
Plasma levels of total hIGFBP-1 in these mice are in the pathological range detected in FGR humans and significantly greater than the concentrations seen in normally grown human term newborns of 40–90 ng/ml (11, 13, 36). Posttranslational phosphorylation increases the affinity of IGFBP-1 for the IGFs up to six times (49). The majority of IGFBP-1 in healthy adult human plasma is in a single phosphorylated isoform. In contrast, in fetal serum, amniotic fluid, and serum from pregnant women, the proportion of nonphosphorylated IGFBP-1 is much higher (50). It has been suggested that mouse kinases are unable to phosphorylate hIGFBP-1 (51). However, because the hIGFBP-1 in our study was partially phosphorylated, this speculation may not be correct. Interestingly, the percentage of phosphorylated hIGFBP-1 increased with advancing postnatal age, as seen in human neonates. Whether the presence of nonphosphorylated hIGFBP-1 in the fetal serum of these AFP-BP1 mice is a result of a conflict with mouse kinases or represents a phenomenon similar to that seen in human fetal serum is currently being studied. The binding capacity of phosphorylated vs. nonphosphorylated hIGFBP-1 in these transgenic mice would provide important information regarding the role of these isoforms during FGR.
Overall, our studies clearly demonstrate a causal relationship between elevated levels of circulating IGFBP-1 and FGR. IGFBP-1 likely causes restricted fetal growth by limiting the mitogenic actions of the IGFs during periods of hypoxia, undernutrition, or placental insufficiency. These AFP-BP1 mice will allow us to better define the factors contributing to impaired in utero growth as well as to determine the role that alterations in the fetal IGF system play in the fetal origins of adult disease.
Acknowledgments
We thank Dr. Robert Baxter, University of Sydney, Australia, for his generous gift of the polyclonal anti-hIGFBP-1 antibody and Dr. Shirley Tilghman, Princeton University, for her generation donation of the AFP promoter as well as Karen Nygard, Dawn Bryce, Michelle Peres, Morgan McCluskey, and Anthony Cheng for their technical assistance.
Footnotes
This work was supported by a grant to V.K.M.H. from the Canadian Institutes of Health Research (CIHR) and Canada Research Chairs Program and fellowship support to C.S.W. from the CIHR and Wyeth, Canada.
First Published Online November 17, 2005
Abbreviations: AFP, -Fetoprotein; E12.5, embryonic d 12.5; FGR, fetal growth restriction; IGFBP, IGF binding protein; IUGR, intrauterine growth restricted; SSC, standard saline citrate; WT, wild type.
Accepted for publication November 9, 2005.
References
Garite TJ, Clark R, Thorp JA 2004 Intrauterine growth restriction increases morbidity and mortality among premature neonates. Am J Obstet Gynecol 340:1234–1238
Lackman F, Capewell V, Richardson B, daSilva O, Gagnon R 2001 The risks of spontaneous preterm delivery and perinatal mortality in relation to size at birth according to fetal versus neonatal growth standards. Am J Obstet Gynecol 84:946–953
Bernstein I, Horbar J, Badger G, Ohlsson A, Golan A 2000 Morbidity and mortality among very-low-birth-weight neonates with intrauterine growth restriction. Am J Obstet Gynecol 182:198–206
Karlberg J, Albertsson-Wikland K 1995 Growth in full-term small-for-gestational-age infants: from birth to final height. Pediatr Res 38:733–739
Barker DJ, Winter PD, Osmond C, Margetts B, Simmonds SJ 1989 Weight in infancy and death from ischaemic heart disease. Lancet 2:577–580
Barker DJ, Hales CN, Fall CH, Osmond C, Phipps K, Clark PM 1993 Type 2 (non-insulin-dependent) diabetes mellitus, hypertension and hyperlipidaemia (syndrome X): relation to reduced fetal growth. Diabetologia 36:62–67
DeChiara TM, Efstratiadis A, Robertson EJ 1990 A growth-deficiency phenotype in heterozygous mice carrying an insulin-like growth factor II gene disrupted by targeting. Nature 345:78–80
Liu JP, Baker J, Perkins AS, Robertson EJ, Efstradiadis A 1993 Mice carrying null mutations of the genes encoding insulin-like growth factor I (Igf-1) and type 1 IGF receptor (Igf1r). Cell 75:59–72
Shimisaki S, Ling N 1991 Identification and molecular characterization of insulin-like growth factor binding proteins (IGFBP-1, -2, -3, -4, -5, and -6). Prog Growth Factor Res 3:243–266
Rajaram S, Baylink DJ, Mohan, S 1997 Insulin-like growth factor-binding proteins in serum and other biological fluids: regulation and functions. Endocr Rev 18:801–831
Wang HS, Lim J, English J, Irvine L, Chard T 1991 The concentration of insulin-like growth factor-1 and insulin-like growth factor-binding protein-1 in human umbilical cord serum at delivery: relation to fetal weight. J Endocrinol 129:459–464
Leger J, Oury JF, Noel M, Baron S, Benali P, Blot P, Czernichow P 1996 Growth factors and intrauterine growth retardation. I. Serum growth hormone, insulin-like growth factor (IGF)-I, IGF-II, and IGF binding protein 3 in normally grown and growth-retarded human fetuses during the second half of gestation. Pediatr Res 40:94–100
Verhaeghe J, Billen J, Giudice LC 2001 Insulin-like growth factor-binding protein-1 in umbilical artery and vein of term fetuses with signs suggestive of distress during labor. J Endocrinol 170:585–590
Straus DS, Ooi GT, Orlowski CC, Rechler MM 1991 Expression of the genes for insulin-like growth factor-I (IGF-I), IGF-II and IGF-binding proteins-1 and -2 in fetal rats under conditions of intrauterine growth restriction caused by maternal fasting. Endocrinology 128:518–525
Iwamoto HS, Murray MA, Chernausek SD 1992 Effects of acute hypoxemia on insulin-like growth factors and their binding proteins in fetal sheep. Am J Physiol 263:E1151–E1156
McLellan KC, Hooper SB, Bocking AD, Delhanty PJ, Phillips ID, Hill DJ, Han VK 1992 Prolonged hypoxia induced by the reduction of maternal uterine blood flow alters insulin-like growth factor-binding protein-1 (IGFBP-1) and IGFBP-2 gene expression in the ovine fetus. Endocrinology 131:1619–1628
Unterman TG, Simmons RA, Glick RF, Ogata RS 1993 Circulating levels of insulin, insulin-like growth factor-I (IGF-I), IGF-II, and IGF-binding proteins in the small for gestational age fetal rat. Endocrinology 132:327–336
Tapanainen PJ, Bang P, Wilson K, Unterman TG, Vreman HJ, Rosenfeld RG 1994 Maternal hypoxia as a model for intrauterine growth retardation: effects on insulin-like growth factors and their binding proteins. Pediatr Res 36:152–158
Cianfarani S, Germani D, Rossi P, Rossi L, Germani A, Ossicini C, Zuppa A, Argiro G, Holly JMP, Branca F 1998 Intrauterine growth retardation: evidence for the activation of the insulin-like growth factor (IGF)-related growth-promoting machinery and the presence of a cation-independent IGF binding protein-3 proteolytic activity by two months of life. Pediatr Res 44:374–380
Lo HC, Tsao LY, Hsu WY, Chen HN, Yu WK, Chi CY 2002 Relation of cord serum levels of growth hormone, insulin-like growth factors, insulin-like growth factor binding proteins, leptin, and interleukin-6 with birth weight, birth length, and head circumference in term and preterm neonates. Nutrition 18:604–608
Crossey PA, Claire CP, Miell JP 2002 Altered placental development and intrauterine growth restriction in IGF binding protein-1 transgenic mice. J Clin Invest 110:411–418
Rajkumar K, Barron D, Lewitt M, Murphy LJ 1995 Growth retardation and hyperglycemia in insulin-like growth factor binding protein-1 transgenic mice. Endocrinology 136:4029–4034
Gay E, Seurin D, Babajko S, Doublier S, Cazillis M, Binoux M 1997 Liver-specific expression of human insulin-like growth factor binding protein-1 in transgenic mice: repercussions on reproduction, ante- and perinatal mortality and postnatal growth. Endocrinology 138:2937–2947
Dai Z, Xing Y, Boney CM, Clemmons DR, D’Ercole AJ 1994 Human insulin-like growth factor-binding protein-1 (hIGFBP-1) in transgenic mice: characterization and insights into the regulation of IGFBP-1 expression. Endocrinology 135:1316–1327
van Kleffens M, Groffen CAH, Dits NFJ, Lindenbergh-Kortleve DJ, Schuller AGP, Bradshaw SL, Pintar JE, Zwarthoff EC, Drop SLS, van Neck JW 1999 Generation of antisera to mouse insulin-like growth factor binding proteins (IGFBP)-1 to -6: comparison of IGFBP protein and messenger ribonucleic acid localization in the mouse embryo. Endocrinology 140:5944–5952
Chen H, Egan JO, Chiu JF 1997 Regulation and activities of -fetoprotein. Crit Rev Eukaryot Gene Express 7:11–41
Godbout R, Ingram RS, Tilghman SM 1988 Fine-structure mapping of the three mouse -fetoprotein gene enhancers. Mol Cell Biol 8:1169–1178
Godbout R, Tilghman SM 1988 Configuration of the -fetoprotein regulatory domain during development. Genes Dev 2:949–956
Ramesh TM, Ellis AW, Spear BT 1995 Individual mouse -fetoprotein enhancer elements exhibit different patterns of tissue-specific and hepatic position-dependent activities. Mol Cell Biol 15:4947–4955
Sambrook J, Russell DW 2001 Molecular cloning: a laboratory manual. 3rd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press
Southern EM 1975 Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Biol 98:503–517
Han VKM, Basset N, Walton J, Challis JRG 1996 The expression of insulin-like growth factor (IGF) and IGF binding protein (IGFBP) genes in the human placenta and membranes: evidence for IGF-IGFBP interactions at the feto-maternal interface. J Clin Endocrinol Metab 81:2680–2693
Baxter RC, Martin JL, Wood MH 1987 Two immunoreactive binding proteins for insulin-like growth factors in human amniotic fluid: relationship to fetal maturity. J Clin Endocrinol Metab 65:423–431
Bradshaw SL, D’Ercole AJ, Han VKM 1999 Overexpression of insulin-like growth factor-binding protein-2 in C6 glioma cells results in conditional alteration of cellular growth. Endocrinology 140:575–584
Khosravi MJ, Diamandi A, Mistry J 1997 Immunoassay of insulin-like growth factor binding protein-1. Clin Chem 43:523–532
Giudice LC, de Zegher F, Gargosky SE, Dsupin BA, de las Fuentes L, Crystal RA, Hintz RL, Rosenfeld RG 1995 Insulin-like growth factors and their binding proteins in the term and preterm human fetus and neonate with normal and extremes of intrauterine growth. J Clin Endocrinol Metab 80: 1548–1555
Seurin D, Froment P, Bluet-Pajot MT, Epelbaum J, Monget P, Binoux M 2002 Functional alteration of the somatotrophic axis in transgenic mice with liver-specific expression of human insulin-like growth factor binding protein-1. Pediatr Res 52:168–174
Lee PD, Giudice LC, Conover CA, Powell DR 1997 Insulin-like growth factor binding protein-1: recent findings and new directions. Proc Soc Exp Biol Med 216:319–357
Constancia M, Hemberger M, Hughes J, Dean W, Ferguson-Smith A, Fundele R, Stewart F, Kelsey G, Fowden A, Sibley C, Reik W 2002 Placental-specific IGF-II is a major modulator of placental and fetal growth. Nature 417:945–948
Sibley CP, Coan PM, Ferguson-Smith AC, Dean W, Hughes J, Smith P, Reik W, Burton GJ, Fowden AL, Constancia M 2004 Placental-specific insulin-like growth factor 2 (Igf2) regulates the diffusional exchange characteristics of the mouse placenta. Proc Natl Acad Sci USA 101:8204–8208
Phipps K, Barker DJ, Hales CN, Fall CH, Osmond C, Clark PM 1993 Fetal growth and impaired glucose tolerance in men and women. Diabetologia 36:225–228
Martyn CN, Marker DJP, Osmond C 1996 Mothers’ pelvic size, fetal growth, and death from stroke and coronary heart disease in men in the UK. Lancet 348:1264–1268
Moore VM, Cockington RA, Ryan P, Robinson JS 1999 The relationship between birth weight and blood pressure amplifies from childhood to adulthood. J Hypertens 17:883–888
Crossey PA, Jones JS, Miell JP 2000 Dysregulation of the insulin/IGF binding protein-1 axis in transgenic mice is associated with hyperinsulinemia and glucose intolerance. Diabetes 49:457–465
Lindgren BF, Segovia B, Lassarre C, Binoux M, Gourmelen M 1996 Growth retardation in constitutionally short children is related both to low serum levels of insulin-like growth factor-I and to its reduced bioavailability. Growth Regul 6:158–164
Kamoda T, Saitoh H, Hirano T, Matsui A 2000 Serum levels of free insulin-like growth factor (IGF)-I and IGF-binding protein-1 in prepubertal children with idiopathic short stature. Clin Endocrinol 53:683–688
Cianfarani S, Geremia C, Scott CD, Germani D 2002 Growth, IGF system, and cortisol in children with intrauterine growth retardation: is catch-up growth affected by reprogramming of the hypothalamic-pituitary-adrenal axis Pediatr Res 51:94–99
Braulke T, Gotz W, Claussen M 1996 Immunohistochemical localization of insulin-like growth factor binding protein-1, -3 and -4 in human fetal tissues and their analysis in media from fetal tissue explants. Growth Regul 6:55–65
Jones JI, D’Ercole AJ, Camacho-Hubner C, Clemmons DR 1991 Phosphorylation of insulin-like growth factor (IGF)-binding protein-1 in cell culture and in vivo: effects on affinity for IGF-I. Proc Natl Acad Sci USA 88:7481–7485
Westwood M, Gibson JM, Davies AJ, Young RJ, White A 1994 The phosphorylation pattern of insulin-like growth factor binding protein-1 in normal plasma is different from that in amniotic fluid and changes during pregnancy. J Clin Endocrinol Metab 79:1735–1741
Sakai K, D’Ercole AJ, Murphy LJ, Clemmons DR 2001 Physiological differences in insulin-like growth factor binding protein-1 (IGFBP-1) phosphorylation in IGFBP-1 transgenic mice. Diabetes 50:32–38(Carole S. Watson, Peter Bialek, Makoto A)