1 Fetal Medicine Research Unit, University of Bristol, St Michael's Hospital, Southwell Street, Bristol BS2 8EG and 2 Department of Surgery, University of Bristol, Bristol Royal Infirmary, Marlborough Street, Bristol BS2, UK
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: Doppler/fetal/growth/IGF/maternal
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Interpretation of the relationship between maternal serum concentrations of IGF-I and fetal growth is difficult without consideration of the binding proteins (IGFBP), especially IGFBP-3 which is the most abundant binding protein in adult serum and which has the highest affinity for IGF. It binds 7090% of adult human serum IGF-I and IGF-II in the form of a 150 kDa ternary complex which is unable to cross endothelium into the tissues and so provides a `reservoir' of circulating IGF which increases the half-life from 10 min to 1215 h (Guler et al., 1989). Circulating concentrations are regulated primarily by growth hormone although this may not be direct (Spagnoli and Rosenfeld, 1996
). IGFBP-3 thus plays a major role in determining the total concentration of IGF-I in the circulation.
A correlation between maternal serum IGF-I and IGFBP-3 measured in normal pregnancy by radioimmunoassay has been reported (Reece et al., 1994) but no changes in IGFBP-3 have been observed in FGR (Langford et al., 1995
; Larsen et al., 1996
) even though IGF-I is low (Holmes et al., 1997
). This apparent contradiction could be a consequence of a loss of the normal regulatory control of IGFBP-3 in FGR but could also be the result of population differences or methodological difficulties.
In late pregnancy, IGFBP-3 concentrations measured by radioimmunoassay are elevated (Baxter and Martin, 1986) but there is complete proteolytic modification by a specific protease by 8 weeks gestation (Giudice et al., 1990
; Davies et al., 1991
). Although proteolysed IGFBP-3 can bind IGFs in vivo it does not bind I125-labelled IGF so is not seen by Western ligand blotting (Suikkari and Baxter, 1991
). Furthermore, the proteolytic structural alterations result in reduced affinity for IGF-I and so increased bioavailability (Holly et al., 1993
). In six cases of FGR due to placental dysfunction at 27 weeks gestation no changes found were in the pattern of proteolysis suggesting that the effects of already low maternal IGF-I are not exacerbated by a reduction in IGFBP-3 proteolysis (Langford et al., 1995
).
IGFBP-2 has been less extensively studied than IGFBP-3. It shows some similarities to IGFBP-1, being in part regulated by metabolic factors, and it is thought to be able to move between the intravascular and interstitial spaces with or without bound IGF, suggesting a role in the distribution of IGF (Clemmons et al., 1991). It is the dominant IGFBP in the fetal circulation but there are no data on maternal serum IGFBP-2 in FGR and very few studies in normal human pregnancy. A gradual reduction of IGFBP-2 with advancing gestation has been reported (Giudice et al., 1990
) probably due to the action of a protease (Davies et al., 1991
).
In response to the similarities between IGFBP-2 and -3 with respect to the presence of specific proteases (unlike IGFBP-1) and the paucity of data in pregnancy, the mothers of fetuses with both normal and impaired growth were studied in order to explore the relationship between these IGFBPs and fetal growth. The aim was to confirm that in FGR pregnancies in the second half of pregnancy, there was no reduction in the degree of proteolysis of IGFBP-3 or alterations in the profile of proteolysed fragments compared to normal pregnancy. In parallel with these studies we wished to assess changes in maternal IGFBP-2 and its proteolysis both in normal pregnancy and in the presence of impaired fetal growth. Reduced growth was carefully assessed by ultrasound using umbilical artery Doppler which is the best non-invasive marker of placental function (Nicolaides et al., 1988; Karsdorp et al., 1994
). Therefore a distinction was made between pregnancies complicated by FGR (due to placental dysfunction) and normal pregnancies with constitutionally small for gestational age (SGA) fetuses.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Fetal AC, head circumference, biparietal diameter and femur length were measured by ultrasound (Ultramark 9, ATL, Bothell, WA, USA; or XP-10, Acuson, Mountain View, CA, USA) and placental function was assessed by umbilical Doppler velocimetry studies, growth velocity and measurement of the amniotic fluid index (AFI). A fetal anatomy survey was performed to identify structural anomalies or markers of other problems such as karyotypic abnormality or infection and invasive testing was undertaken as clinically indicated. The fetal assessment was repeated every 2 weeks until delivery and at each visit 8 ml of maternal blood was obtained between 1000 and 1600 h in a non-heparinized tube. This was allowed to clot, centrifuged at 3000 g for 5 min and the serum was stored at 24°C. When maternal dexamethasone was clinically indicated for fetal lung maturation, blood samples were taken before steroid administration although the IGF system may not be affected by exogenous steroids (Ogueh et al., 1998).
For cases, criteria for exclusion from analysis after birth were evidence of structural (n = 1), chromosomal (n = 2) or infectious congenital anomalies (n = 1) or a birthweight >5th centile for gestational age (1.645 SD) (n = 17). The remaining 54 cases were divided into two groups, FGR (due to placental dysfunction) and SGA (normal small), depending upon the umbilical artery pulsatility index (UA PI) at the last visit before delivery. UA PI falls with advancing gestational age, so it was expressed as multiples of SD from the mean for gestation to remove the effect of gestational age, and a UA PI of >2 SD above the mean was classified as FGR. Two potential SGA cases with a normal UA PI but with an AFI <5 cm were excluded from the analysis because a severe reduction in amniotic fluid volume in the absence of ruptured membranes or fetal renal abnormality may be associated with placental dysfunction. They were not, however, classified as FGR because AFI is more subjective and harder to reproduce than Doppler and it is a poor predictor of outcome compared with Doppler (Soothill et al., 1993). The final numbers of cases from which maternal blood samples were assayed were 25 FGR and 27 SGA cases.
For controls, criteria for exclusion from analysis after birth were a birthweight <5th centile for gestational age (1.645 SD) (n = 7) or >95th centile (+1.645 SD) (n = 3), the development of gestational hypertension (n = 3) or diabetes (n = 1), a reduced AFI at term (n = 4) or reduced growth velocity at term (n = 2). The final number of controls for analysis was 89. All exclusion criteria were determined before recruitment commenced with the aim of minimizing misclassification or overlap of groups.
The maternal serum sample obtained from each of the 52 cases (FGR and SGA) at the last assessment before delivery was selected for laboratory analysis. A cross-sectional normal range was constructed by selecting a single sample from each of the 89 controls, stratifying by gestational age (prior to any laboratory work) and so ensuring an even gestational age spread of control results from 24 to 42 weeks.
IGFBP-3 was assayed by a specific radioimmunoassay (Cheetham et al., 1994). The IGFBP-3 antibody was an in-house polyclonal antibody (SCH 2/5) raised in a rabbit against recombinant non-glycosylated IGFBP-3 and used at a dilution of 1:2000 in assay buffer (final dilution in assay was 1:2&0circ;000). Non-glycosylated IGFBP-3 donated by Dr C.Maack, Celtrix Pharmaceuticals Inc. (Santa Clara, CA, USA) was iodinated by the chloramine-T method (Blum and Breier, 1994
). A standard curve was constructed using normal human serum diluted with assay buffer and acetic acid solution to give a range of 6.6624 ng/ml. Normal human serum had a range of 4.26.8 µg/ml. At 5 µg/ml the intra- and inter-assay coefficients of variation were 5.14% and 4.28% respectively. Cross-reactivity with IGFBP-1 was <0.2%, there was no interference by either IGF-I or IGF-II and the limit of detection of the assay was 21.1 ng/ml.
To perform Western ligand blotting, samples were diluted 1:32 in 10% sodium dodecyl sulphate (SDS) assay buffer and heated at 100°C for 5 min. 80 µl was then loaded per lane of gel and separated on a 30% SDSpolyacrylamide 3 mm gel in a method essentially as described (Laemmli, 1970). Western ligand blotting to visualize the IGFBP profile was a modification of a method previously described (Coulson et al., 1991
). The proteins were electroblotted (0.8 A for 4 h) onto a nitrocellulose membrane (Hybond C-super, Amersham International plc, Amersham, Bucks, UK) and the membrane was blocked with bovine serum albumin and incubated with 125I-labelled IGF-I and IGF-II in equal proportions for 2 h. It was then dried and autoradiographed at 70°C.
For immunoblotting (Cwyfan-Hughes et al., 1995) the membrane was blocked with 5% milk and incubated overnight with an in-house polyclonal rabbit anti-human IGFBP-3 antibody (SCH 2/6) or monoclonal mouse anti-human IGFBP-2 antibody (Novartis Pharmaceuticals, Basle, Switzerland). The IGFBP-2 antisera showed no cross-reactivity with recombinant IGFBP-1, or IGFBP-3 to -6 (unpublished observations, S.C.Cwyfan-Hughes and J.M.P.Holly). The membrane was exposed to anti-rabbit or anti-mouse immunoglobulin G (IgG) (for IGFBP-3 and IGFBP-2 immunoblotting respectively) conjugated to peroxidase according to the manufacturer's instructions (SigmaAldrich, Gillingham, Dorset, UK). After washing, the bound peroxidase was visualised using the ECL Western blotting detection system (Amersham International plc) based on luminol oxidation. At a final dilution of 1 in 1000 there was no detectable cross-reactivity with IGFBP-1, IGFBP-2 (IGFBP-3 immunoblot) or IGFBP-3 (IGFBP-2 immunoblot).
Densitometric quantification of the intensity of IGFBP-2 immunoblot bands was made using a Biorad G5690 densitometer and Macintosh Molecular Analyst Version 2.1. Total amount of IGFBP-2 for each pregnant sample and each normal human serum non-pregnant control (NHS) was calculated by summation of densities of all bands obtained from a single well. Because densities of NHS samples varied between immunoblots despite being from the same pooled serum source (because of variations in the imaging processes), the densitometry result for each pregnancy in a single immunoblot experiment was expressed as a fraction of the NHS IGFBP-2 density on the same immunoblot. It was then possible to compare results of the pregnant groups from different experiments. The proportion of IGFBP-2 proteolysed in each sample was calculated from the immunoblots by the formula: density of lower mol. wt fragmented band/density of fragmented band + density of higher mol. wt intact band.
Control IGFBP-3 data were first tested for normal distribution by the ShapiroWilk test (no transformation was required). IGFBP-3, birthweight and UA PI all changed with gestational age in control pregnancies and were therefore expressed as multiples of SD from the normal mean for gestational age (z score). Correlations between IGFBP-3 SD and birthweight SD and between IGFBP-3 SD and UA PI SD were tested by regression analysis. For SGA and FGR groups, comparisons with controls of IGFBP-3 and of densitometric analysis of IGFBP-2 immunoblot bands were made using Student's unpaired t-test or the MannWhitney test depending upon the normality of the distribution.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Western ligand blotting of non-pregnant adult serum demonstrated a doublet at ~40 kDa which represented intact IGFBP-3 but this was absent in all control, SGA and FGR pregnancies at all gestations between 24 and 40 weeks (Figure 2). It should be noted that only intact IGFBP-3 can be identified by the Western ligand blot. A band at 34 kDa consistent with IGFBP-2 was present in 24 of the 25 FGR cases and the NHS but only faintly observed or absent in SGA and control pregnancies.
|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Factors involved in the regulation of IGFBP-3 proteases are poorly understood but it has been observed that the extent of proteolysis is inversely proportional to GH, being highest in GH-deficiency and lowest in acromegaly (Lassarre et al., 1994). Since IGF-I concentrations are controlled by GH, regulation of proteolytic activity may therefore be adapted to circulating concentrations of IGF-I, increasing bioavailability when its concentrations are low. This relationship was not found in this study in normal pregnancy because IGFBP-3 was entirely proteolysed despite the high IGF-I concentrations in pregnancy (Holmes et al., 1998
). The placenta produces several unique proteases (Gossrau et al., 1987
) and this may explain the increased proteolysis observed. However, a reduction in proteolysis was not found in the mothers of FGR pregnancies despite severe placental dysfunction and although this may be because IGFBP proteases are not placental in origin, alternative explanations are that the placenta has a large functional synthetic reserve for proteases or that there is a reduction of placentally synthesized protease inhibitors. The functional significance of proteolysis in pregnancy is unclear but there must be a marked increase in production of IGF-I for concentrations to rise in late pregnancy despite the half-life of IGF-I falling (Davenport et al., 1990
). If IGFBP-3 proteolysis increases bioavailability of IGF-I then any reduction in proteolysis would be expected to be detrimental to IGF-I action and this would be especially harmful in FGR because maternal concentrations of IGF-I are extremely low. Therefore, if IGFBP-3 protease is produced in the placenta, the maintenance of maximal proteolysis despite placental dysfunction might be an adaptive response to minimize the adverse consequences of low IGF-I.
It has been shown that in-vitro proteolysis, as measured by the capacity of maternal serum to fragment recombinant IGFBP-3, was increased in FGR compared with control pregnancies and suggested that this was a regulatory mechanism to increase the bioavailability of IGF-I (Langford et al., 1995). However, the results of this study suggest that this is unlikely to be of functional significance since there is complete proteolysis in vivo of the intact 3842 kDa doublet in all FGR, SGA and control pregnancies from 24 weeks gestation onwards. Further studies would be required to confirm that proteolysed 30 kDa IGFBP-3 in FGR is functionally the same as that in normal pregnancy but it is unlikely that a small increase in proteolytic activity is functionally significant. Similarly, in an earlier study of normal and diabetic pregnancies, marked variations in in-vitro protease activity were found within patient groups but without significant changes in the IGFBP-3 profile on Western ligand and immunoblotting (Cwyfan Hughes et al., 1995
).
This report documents changes in maternal IGFBP-2 in FGR for the first time. Low concentrations of the intact 34 kDa band were observed on Western ligand blotting in normal pregnancy and the presence of a 14 kDa fragment consistent with proteolytic activity was demonstrated. However, in contrast to IGFBP-3, proteolysis of IGFBP-2 is not complete; 10% of IGFBP-2 remains in a non-proteolysed form in control pregnancies and 24% in FGR. Since total concentrations of IGFBP-2 are increased by 60% in FGR the concentration of the intact 34 kDa form is markedly increased and can be readily visualized on the Western ligand blot. Although the significance of these changes is unclear, elevated concentrations of IGFBP-2 would be expected to further reduce bioavailability of IGF-I because IGFBP-2 is predominantly inhibitory to the actions of IGF-I.
The large changes in IGFBP-2 in FGR but not in SGA confirmed that in any study of fetal growth it is important to define aetiological groups. Small fetuses include those who fail to achieve their genetic potential for growth including fetal abnormalities (e.g. chromosomal or structural), infection or placental dysfunction as well as constitutionally SGA with an intrinsically low genetic potential for growth. It has been estimated that about 40% of small fetuses should not be considered growth-restricted (Weiner and Williamson, 1989). Careful recruitment and serial assessment of pregnancies using umbilical artery Doppler, which is the best non-invasive marker of placental function (Nicolaides et al., 1988
), ensured that the FGR pregnancies were all associated with placental dysfunction. Defining FGR solely in terms of reduced birthweight would have greatly reduced the power of the study to detect any changes in IGFBP-2 in the sub-group with impaired placental function.
This study has demonstrated that despite the rise in maternal IGFBP-3 with advancing gestational age there is no correlation with birthweight in normally grown or growth-restricted fetuses. The normal pregnancy adaptation of increased IGFBP-3 proteolysis, which may enhance bioavailability of IGF-I for the anabolic requirements of pregnancy, remains functional in FGR pregnancy and so is not a mechanism by which placental transfer is limited. It was possible that changes in concentrations of IGFBP-2 and IGFBP-3 or their proteolysis might have compensated for the low maternal IGF-I concentrations observed in FGR by increasing bioavailability compared with normal pregnancy. These results suggest that this is not the case.
![]() |
Notes |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Baxter, R.C. and Martin, J.L. (1986) Radioimmunoassay of growth hormone-dependent insulin like growth factor binding protein in human plasma. J. Clin. Invest., 78, 15041512.[ISI][Medline]
Caufriez, A., Frankenne, F., Englert, Y. et al. (1990) Placental growth hormone as a potential regulator of maternal IGF-I during human pregnancy. Am. J. Physiol., 258, E1014E1019.
Cheetham, T.D., Taylor, A., Holly, J.M.P. et al. (1994) The effects of recombinant human insulin-like growth factor-I (IGF-I) administration on the levels of IGF-I, IGF-II and IGF binding proteins in adolescents with insulin-dependent diabetes mellitus. J. Endocrinol., 142, 367374.[Abstract]
Clemmons, D.R., Snyder, D.K. and Busby, W.H. (1991) Variables controlling the secretion of insulin-like growth factor binding protein-2 in normal human subjects. J. Clin. Endocrinol. Metab., 73, 727733.[Abstract]
Coulson, V.J., Wass, J.A.H., Abdulla, A.F. et al. (1991) Insulin-like growth factor binding proteins (IGFBPs) in acromegaly. Growth Reg., 1, 119124.[ISI][Medline]
Cwyfan-Hughes, S.C., Johnson, M.R., Heinrich, G. et al. (1995) Could abnormalities in insulin-like growth factors and their binding proteins during pregnancy result in gestational diabetes. J. Endocrinol., 147, 517524.[Abstract]
Davenport, M.L., Clemmons, D.R., Miles, M.V. et al. (1990) Regulation of serum insulin-like growth factor-I (IGF-I) and IGF binding proteins during rat pregnancy. Endocrinology, 127, 12781286.[Abstract]
Davies, S.C., Holly, J.M.P., Coulson, V.J. et al. (1991) The presence of cation-dependent proteases for insulin-like growth factor binding proteins does not alter the size distribution of insulin-like growth factors in pregnancy. Clin. Endocrinol., 34, 501506.[ISI][Medline]
Giudice, L.C., Farrell, E.M., Pham, H. et al. (1990) Insulin-like growth factor binding proteins in maternal serum throughout gestation and in the puerperium: Effects of a pregnancy-associated serum protease activity. J. Clin. Endocrinol. Metab., 71, 806816.[Abstract]
Gossrau, R., Graf, R., Ruhnke, M. et al. (1987) Proteases in the human full-term placenta. Histochemistry, 86, 405413.[ISI][Medline]
Guler, H-P., Zapf, J., Schmid, C. et al. (1989) Insulin-like growth factors I and II in healthy man. Estimations of half-lives and production rates. Acta Endocrinol., 121, 753758.[ISI][Medline]
Hall, K., Enberg, G., Hellem, E. et al. (1984) Somatomedin levels in pregnancy: Longitudinal study in healthy subjects and patients with growth hormone deficiency. J. Clin. Endocrinol. Metab., 59, 587594.[Abstract]
Holly, J.M.P., Claffey, D.C.P., Cwyfan-Hughes, S.C. et al. (1993) Proteases acting on IGFBP's; their occurrence and physiological significance. Growth Reg., 3, 8891.[ISI][Medline]
Holmes, R.P., Holly, J.M.P. and Soothill, P.W. (1998) A prospective study of maternal serum insulin-like growth factor-I in pregnancies with appropriately grown or growth restricted fetuses. Br. J. Obstet. Gynaecol., 105, 12731278.[ISI][Medline]
Holmes, R., Montemagno, R., Jones, J. et al. (1997) Fetal and maternal plasma levels of insulin-like growth factors and binding proteins in pregnancies with appropriately grown or growth retarded fetuses. Early Hum. Dev., 49, 717.[ISI][Medline]
Karsdorp, V.H., van Vugt, J.M., Jakobs, C. et al. (1994) Amino acids, glucose and lactate concentrations in umbilical cord blood in relation to umbilical artery flow patterns. Eur. J. Obstet. Gynecol. Reprod. Biol., 57, 117122.[ISI][Medline]
Laemmli, U.K. (1970) Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature, 277, 680
Langford, K.S., Nicolaides, K.H., Jones, J. et al. (1995) Serum insulin-like growth factor binding protein-3 (IGFBP-3) levels and IGFBP-3 protease activity in normal, abnormal and multiple pregnancy. J. Clin. Endocrinol. Metab., 80, 2127.[Abstract]
Langford, K., Nicolaides, K. and Miell, J.P. (1998) Maternal and fetal insulin-like growth factors and their binding proteins in the second and third trimesters of human pregnancy. Hum. Reprod., 13, 138993.[Abstract]
Larsen, T., Main, K., Andersson, A.M. et al. (1996) Growth hormone, insulin-like growth factor1 and its binding proteins 1 and 3 in last trimester intrauterine growth retardation with increased pulsatility index in the umbilical artery. Clin. Endocrinol., 45, 315319.[ISI][Medline]
Lassarre, C., Lalou, C., Perin, L. et al. (1994) Protease-induced alteration of insulin-like growth factor binding protein-3 as detected by radioimmunoassay. Agreement with ligand blotting data. Growth Reg., 4, 4855.[ISI][Medline]
Liu, L., Harding, J.E., Evans, P.C. et al. (1994) Maternal insulin-like growth factor-I infusion alters feto-placental carbohydrate and protein metabolism in pregnant sheep. Endocrinology, 135, 895900.[Abstract]
Mirlesse, V., Frankenne, F., Alsat, E. et al. (1993) Placental growth hormone levels in normal pregnancy and in pregnancies with intrauterine growth retardation. Pediatr Res., 34, 439442.[Abstract]
Morel, P.C., Blair, H.T., Ormsby, J.E. et al. (1994) Influence of fetal and maternal genotype for circulating insulin-like growth factor-I on fetal growth in mice. J. Reprod. Fertil., 101, 914.[Abstract]
Nicolaides, K.H., Bilardo, C.M., Soothill, P.W. et al. (1988) Absence of end-diastolic frequencies in umbilical artery: A sign of fetal hypoxia and acidosis. Br. Med. J., 297, 10261027.[ISI][Medline]
Ogueh, O., Hills, F.A., Chard, T. et al. (1998) Antenatal dexamethasone therapy does not affect circulating concentrations of insulin-like growth factor binding protein-1. Hum. Reprod., 13, 17141716.[Abstract]
Owens, J.A. (1991) Endocrine and substrate control of fetal growth: Placental and maternal influences and insulin-like growth factors. Reprod. Fertil. Dev., 3, 501517.[ISI][Medline]
Reece, E.A., Wiznitzer, A., Le, E., Homko, C.J. et al. (1994) The relation between human fetal growth and fetal blood levels of insulin-like growth factors I and II, their binding proteins and receptors. Obstet. Gynecol., 84, 8895.[Abstract]
Soothill, P.W., Ajayi, R.A., Campbell, S. et al. (1993) Prediction of morbidity in small and normally grown fetuses by fetal heart rate variability, biophysical profile score and umbilical artery Doppler studies. Br. J. Obstet. Gynaecol., 100, 742745.[ISI][Medline]
Spagnoli, A. and Rosenfeld, R.G. (1996) The mechanisms by which growth hormone brings about growth. Endocrinol. Metab. Clin. North Am., 25, 615631.[ISI][Medline]
Suikkari, A-M. and Baxter, R.C. (1991) Insulin-like growth factor (IGF) binding protein-3 in pregnancy serum binds native IGF-1 but not iodo-IGF-1. J. Clin. Endocrinol. Metab., 73, 13771379.[Abstract]
Thissen, J-P., Ketelslegers, J-M. and Underwood, L.E. (1994) Nutritional regulation of the insulin-like growth factors. Endocr. Rev., 15, 80101.[Abstract]
Weiner, C.P. and Williamson, R.A. (1989) Evaluation of severe growth retardation using cordocentesis: haematologic and metabolic alterations by aetiology. Obstet. Gynecol., 73, 225229.[Abstract]
Yen, S.S.C., Vela, P. and Tsai, C.C. (1970) Impairment of growth hormone secretion in response to hypoglycaemia during early and late pregnancy. J. Clin. Endocrinol. Metab., 31, 2932.[ISI][Medline]
Submitted on November 4, 1998; accepted on March 30, 1999.