Kinderklinik, University of Heidelberg1), Heidelberg, Germany, University of Colorado Medical Center2). Denver, Colorado, USA
and
NSILA AND FOETAL GROWTH
By U. E. Heinrich, D. S. Schalch, M. H. and C. J. Johnson
Jawadi
ABSTRACT
Employing a sensitive competitive protein binding assay for NSILA (non-suppressible insulin-like activity), circulating levels of this somatomedin (SM) have been measured throughout pregnancy, at parturition, and in foetal and newborn sera. Acid-dissociable serum NSILA (mean \m=+-\sem) in 57 women was significantly higher during pregnancy (1106 \m=+-\46 \g=m\U/ml), than in 11 adult non-pregnant control subjects (844 \m=+-\ 22 \g=m\U/ml), but not correlated with week of gestation or with serum growth hormone (GH) or cortisol levels. At parturition, the NSILA concentration in 28 cord sera (598 \m=+-\38 \g=m\U/ml) was significantly less than in the corresponding maternal sera (1039 \m=+-\ 63 \g=m\U/ml). The NSILA levels in 23 premature newborns (370 \m=+-\20 \g=m\U/ml) and 8 smallfor-gestational-age newborns (310 \m=+-\46 \g=m\U/ml) were significantly less than in 33 term newborns (494 \m=+-\18 \g=m\U/ml). Serum NSILA in 56 term and premature newborns exhibited a significant positive correlation both with gestational age and birth weight but not with serum GH or cortisol levels. These data suggest that the maternal-foetal growth-promoting system is a highly complex one in which NSILA levels both in maternal and foetal circulations appear to be\ under multifactorial control.
There
multiple genetic and environmental influences on intra-uterine growth (Abdul-Karim 8. Beydoun 1974; Warshaw 1974). Genetic differences exist between individuals which depend upon familial growth patterns and potential racial, and to some extent ethnic, characteristics. Environmental in¬ fluences superimposed include maternal age, parity, nutrition and illness, exposure to drugs (e.g. heroin, nicotine, alcohol) and transplacentally acquired are
To Prof. Dr. H. Bickel
on
occasion of his 60th
birthday.
intra-uterine infections (e.g. rubella, cytomegaly, toxoplasmosis). The extent of hormonal influence on foetal growth compared with the predominant role of maternal environment is still uncertain. As important anabolic hormones, insulin and growth hormone (GH) must be considered prime candidates, but their physiological significance for the growing foetus is controversial (Jost 1966; Dourov 8c Buyl-Strouvens 1969; Naeye 8c Blanc 1971; Kaplan et al. 1972; Laron 1972; Laron et al. 1972; Honnebier 8c Swaab 1973; Turner 8c Cohen 1974; Hill 1976). Moreover, the somatomedins (SMs), mediating the action of GH, and possessing both growth-promoting and insulin-like activities (Daughaday et al. 1972), have been reported by several investigators to be low in newborns ( Tato et al. 1975; D'Ercole et al. 1976; Giordano et al. 1976; Gluckman 8c Brinsmead 1976; Furlanetto et al. 1977; Hintz 8c Seeds 1977). On the other hand, there is evidence to suggest that prolactin (Young et al. 1973; Francis 8c Hill 1975), human chorionic somatomammotrophin (Kaplan 1974; Hurley et al. 1977) and oestrogens (Abdul-Karim et al. 1971; Wiedemann 8c Schwartz 1972), play a role in foetal growth, either directly or through the SM pathway. Non-suppressible insulin-like activity (NSILA) is recognized as one of the SMs or "growth promoting peptides" whose chemical structure has been recently clarified by Rinderknecht 8c Hiimbel (1978). Actually, this SM consists of two closely-related peptides, recently designated as IGE (insulin-like growth factor) I and II, whose potencies have been shown to be much less than that of insulin on a molar basis with respect to their insulin-like activity on fat tissue, but much more with respect to their growth-promoting activity in fibroblasts (Froesch et al. 1975). A number of studies have been concerned with the role of the SMs in foetal growth, but since the bioassays employed are very sensitive to non-specific stimulatory and inhibitory activities in serum, their results, sometimes contradictory, are difficult to interpret (Daughaday et al. 1959; Chesley 1962; Tato el al. 1975; D'Ercole et al. 1976; Giordano et al. 1976; Gluckman 8c Brinsmead 1976; Hintz 8c Seeds 1977; Svan et al. 1977). The recent purification of several somatomedins (A,B,C, NSILA and multi¬ plication stimulating activity (MSA)), has promoted the development of more specific and precise radioreceptor assays (Hall et al. 1974; Marshall et al. 1974; Megyesi et al. 1974), radioimmunoassays (Yalow et al. 1975; Furlanetto et al. 1977; Hall et al., in press; Moses et al. 1978), and competitive protein binding assays (Zapf et al. 1977; Scheuch et al. 1978) for the determination of the various peptides. Simultaneously, it has become apparent that NSILA and the other SMs circulate in serum largely bound to high molecular weight carrier protein(s) (Zapf et al. 1975; Moses et al. 1976; Hintz 8c Liu 1977; Heinrich et al. 1978). Utilizing this observation, sensitive competitive protein binding assays for the determination of NSILA in serum extracts have recently been developed (Zapf el al. 1977; Schalch et al. 1978). At present, a role for NSILA and the other SMs in foetal growth is mainly speculative. No information on serum
NSILA levels in the foetus has been available so far. This data and attempts to answer the following basic questions. 1. Are there
and
study reports
systematic changes of NSILA levels in women during how do they relate to changes of other hormones?
such
pregnancy
2. Is NSILA
present in cord serum and that of newborns and how do levels relate to maternal concentrations, birth weight, gestational age and to growth hormone and cortisol levels?
MATERIALS AND METHODS
Subjects Blood samples were obtained from several groups of subjects which are briefly described below. A. Healthy pregnant women (n 57).- A venous sample was obtained from each of 57 normal women, 14 to 39 years of age, on a single occasion between the 7th and 40th week of pregnancy at the time of a routine visit to the outpatient clinic, Department of Obstetrics and Gynecology, University of Colorado Medical Center, Denver, Colorado, USA. =
B. Healthy pregnant women at from 28 women, ages 20-32 years, normal vaginal delivery.
parturition (n 28).- Blood samples were during the early contraction phase shortly =
drawn before
newborns (n 28). Mixed cord blood was collected from the cor¬ newborn babies of the mothers (group B) sampled at the time of parturi¬ tion. APGAR scores were 7-10 at 1 min, and no major or minor constitutional ab¬ normalities were observed. Birth weights ranged between 2040 and 3640 g.
C.
Healthy
responding
=
-
D. Neonates, ages 1-10 days (n 64). This group included 33 normal term, 23 pre¬ (32-36 weeks gestation) and 8 ''small-for-gestational-age" (term/SGA) babies. Taken as a group, birth weights ranged from 1650 to 3890 g, and gestational ages (Finnstrom 1972) from 33 to 40 weeks. Cases of neonatal hypoglycaemia, hypocalcaemia, hyperbilirubinaemia (> 14 mg/100 ml) and respiratory distress syndrome were excluded from this study. =
mature
-
E. Normal adult control subjects (n=ll). Blood samples were obtained from 7 normal non-pregnant women and 4 normal men ranging in age from 18 to 46. It has been previously shown in adults that there are no sex- or age-related differences in serum NSILA levels (Schalch et al. 1978). -
Assay procedures All blood samples were centrifuged as quickly as possible, and the sera separated and stored at -20°C until assayed. Most values are expressed as the mean ± sem. Statistical differences between two groups were calculated using Student's t-test, while either coefficients of correlation or multiple regression analyses were used to com¬ pare multiple points within or between various groups. A brief description of the assay techniques used follows.
A. NSILA. Acid soluble serum NSILA was measured by a competitive protein binding assay, the details of which have been published elsewhere (Schalch et al. 1978). Sera were first chromatographed on Sephadex G-50 (fine) in 1 m acetic acid (pH 2.3), and those fractions eluting between 45 and 70% of total column volume were pooled, lyophilized and reconstituted in 0.15 M HOAc/0.I5 M NaCl prior to assay. The NSILA preparation used for iodination had a purity of ^200 mU/mg, while that used for standards was - ,3.8 mU/ml 1) Both preparations contained appro¬ ximately equimolar amounts of IGF I and II. Partially purified NSILA-free carrier protein employed in the assay was derived from Cohn fraction IV of human plasma (kindly supplied by Armour Pharmaceutical Co., Kankakee, 111.) by chromatography on Sephadex G-200 in 0.15 M HOAc/0.15 m NaCl and concentrated by ultrafiltration (Amicon CH 3 Hollow fiber concentrator, membrane cut-off 10 000). Two limitations inherent to the design of this competitive protein binding assay must be considered in interpreting its results: 1) since both tracer and standard NSILA preparations contain (IGF) peptides I and II, this assay cannot distinguish between these two moieties in serum eluates, but simply measures "total" acid-dissociabe NSILA activity, and 2) since it has been shown recently that MSA cross-reacts with NSILA in a similar competitive protein binding system (Recider et al. 1977), this procedure is not specific for NSILA per se but rather for NSILA "activity". It should be pointed out that for reasons that are obscure, serum NSILA levels measured by our assay system are approximately twice those measured by a similar procedure (Zapi et al. 1977), though values in subjects with GH excess or deficiency are quite com¬ parable relative to normals. B. Hormones. Serum cortisol was assayed by the method of Murphy (1967), and human GH by the double antibody method of Schalch Se Parker (1964). Human GH antiserum with low cross-reactivity to human chorionic somatomammotrophin was chosen for assays of maternal sera with high concentrations of this hormone. -
-
RESULTS
NSILA, GH and cortisol in maternal
sera. No significant differences NSILA levels between stages of pregnancy (Fig. 1). appeared to exist in total There was neither a significant correlation between serum NSILA concentra¬ tion and week of pregnancy, nor a significant difference between the mean (± sem) NSILA level in the third trimester (1207 1 106 µ /mì) and that in the first trimester (959 1 70 ¿/U/ml, P>0.05). However, in the majority of subjects the NSILA was above the range established for adult control subjects, and the mean value throughout pregnancy (1106 1 46 ¿iU/ml) was significantly higher (P
and
NSILA AND FOETAL GROWTH
By U. E. Heinrich, D. S. Schalch, M. H. and C. J. Johnson
Jawadi
ABSTRACT
Employing a sensitive competitive protein binding assay for NSILA (non-suppressible insulin-like activity), circulating levels of this somatomedin (SM) have been measured throughout pregnancy, at parturition, and in foetal and newborn sera. Acid-dissociable serum NSILA (mean \m=+-\sem) in 57 women was significantly higher during pregnancy (1106 \m=+-\46 \g=m\U/ml), than in 11 adult non-pregnant control subjects (844 \m=+-\ 22 \g=m\U/ml), but not correlated with week of gestation or with serum growth hormone (GH) or cortisol levels. At parturition, the NSILA concentration in 28 cord sera (598 \m=+-\38 \g=m\U/ml) was significantly less than in the corresponding maternal sera (1039 \m=+-\ 63 \g=m\U/ml). The NSILA levels in 23 premature newborns (370 \m=+-\20 \g=m\U/ml) and 8 smallfor-gestational-age newborns (310 \m=+-\46 \g=m\U/ml) were significantly less than in 33 term newborns (494 \m=+-\18 \g=m\U/ml). Serum NSILA in 56 term and premature newborns exhibited a significant positive correlation both with gestational age and birth weight but not with serum GH or cortisol levels. These data suggest that the maternal-foetal growth-promoting system is a highly complex one in which NSILA levels both in maternal and foetal circulations appear to be\ under multifactorial control.
There
multiple genetic and environmental influences on intra-uterine growth (Abdul-Karim 8. Beydoun 1974; Warshaw 1974). Genetic differences exist between individuals which depend upon familial growth patterns and potential racial, and to some extent ethnic, characteristics. Environmental in¬ fluences superimposed include maternal age, parity, nutrition and illness, exposure to drugs (e.g. heroin, nicotine, alcohol) and transplacentally acquired are
To Prof. Dr. H. Bickel
on
occasion of his 60th
birthday.
intra-uterine infections (e.g. rubella, cytomegaly, toxoplasmosis). The extent of hormonal influence on foetal growth compared with the predominant role of maternal environment is still uncertain. As important anabolic hormones, insulin and growth hormone (GH) must be considered prime candidates, but their physiological significance for the growing foetus is controversial (Jost 1966; Dourov 8c Buyl-Strouvens 1969; Naeye 8c Blanc 1971; Kaplan et al. 1972; Laron 1972; Laron et al. 1972; Honnebier 8c Swaab 1973; Turner 8c Cohen 1974; Hill 1976). Moreover, the somatomedins (SMs), mediating the action of GH, and possessing both growth-promoting and insulin-like activities (Daughaday et al. 1972), have been reported by several investigators to be low in newborns ( Tato et al. 1975; D'Ercole et al. 1976; Giordano et al. 1976; Gluckman 8c Brinsmead 1976; Furlanetto et al. 1977; Hintz 8c Seeds 1977). On the other hand, there is evidence to suggest that prolactin (Young et al. 1973; Francis 8c Hill 1975), human chorionic somatomammotrophin (Kaplan 1974; Hurley et al. 1977) and oestrogens (Abdul-Karim et al. 1971; Wiedemann 8c Schwartz 1972), play a role in foetal growth, either directly or through the SM pathway. Non-suppressible insulin-like activity (NSILA) is recognized as one of the SMs or "growth promoting peptides" whose chemical structure has been recently clarified by Rinderknecht 8c Hiimbel (1978). Actually, this SM consists of two closely-related peptides, recently designated as IGE (insulin-like growth factor) I and II, whose potencies have been shown to be much less than that of insulin on a molar basis with respect to their insulin-like activity on fat tissue, but much more with respect to their growth-promoting activity in fibroblasts (Froesch et al. 1975). A number of studies have been concerned with the role of the SMs in foetal growth, but since the bioassays employed are very sensitive to non-specific stimulatory and inhibitory activities in serum, their results, sometimes contradictory, are difficult to interpret (Daughaday et al. 1959; Chesley 1962; Tato el al. 1975; D'Ercole et al. 1976; Giordano et al. 1976; Gluckman 8c Brinsmead 1976; Hintz 8c Seeds 1977; Svan et al. 1977). The recent purification of several somatomedins (A,B,C, NSILA and multi¬ plication stimulating activity (MSA)), has promoted the development of more specific and precise radioreceptor assays (Hall et al. 1974; Marshall et al. 1974; Megyesi et al. 1974), radioimmunoassays (Yalow et al. 1975; Furlanetto et al. 1977; Hall et al., in press; Moses et al. 1978), and competitive protein binding assays (Zapf et al. 1977; Scheuch et al. 1978) for the determination of the various peptides. Simultaneously, it has become apparent that NSILA and the other SMs circulate in serum largely bound to high molecular weight carrier protein(s) (Zapf et al. 1975; Moses et al. 1976; Hintz 8c Liu 1977; Heinrich et al. 1978). Utilizing this observation, sensitive competitive protein binding assays for the determination of NSILA in serum extracts have recently been developed (Zapf el al. 1977; Schalch et al. 1978). At present, a role for NSILA and the other SMs in foetal growth is mainly speculative. No information on serum
NSILA levels in the foetus has been available so far. This data and attempts to answer the following basic questions. 1. Are there
and
study reports
systematic changes of NSILA levels in women during how do they relate to changes of other hormones?
such
pregnancy
2. Is NSILA
present in cord serum and that of newborns and how do levels relate to maternal concentrations, birth weight, gestational age and to growth hormone and cortisol levels?
MATERIALS AND METHODS
Subjects Blood samples were obtained from several groups of subjects which are briefly described below. A. Healthy pregnant women (n 57).- A venous sample was obtained from each of 57 normal women, 14 to 39 years of age, on a single occasion between the 7th and 40th week of pregnancy at the time of a routine visit to the outpatient clinic, Department of Obstetrics and Gynecology, University of Colorado Medical Center, Denver, Colorado, USA. =
B. Healthy pregnant women at from 28 women, ages 20-32 years, normal vaginal delivery.
parturition (n 28).- Blood samples were during the early contraction phase shortly =
drawn before
newborns (n 28). Mixed cord blood was collected from the cor¬ newborn babies of the mothers (group B) sampled at the time of parturi¬ tion. APGAR scores were 7-10 at 1 min, and no major or minor constitutional ab¬ normalities were observed. Birth weights ranged between 2040 and 3640 g.
C.
Healthy
responding
=
-
D. Neonates, ages 1-10 days (n 64). This group included 33 normal term, 23 pre¬ (32-36 weeks gestation) and 8 ''small-for-gestational-age" (term/SGA) babies. Taken as a group, birth weights ranged from 1650 to 3890 g, and gestational ages (Finnstrom 1972) from 33 to 40 weeks. Cases of neonatal hypoglycaemia, hypocalcaemia, hyperbilirubinaemia (> 14 mg/100 ml) and respiratory distress syndrome were excluded from this study. =
mature
-
E. Normal adult control subjects (n=ll). Blood samples were obtained from 7 normal non-pregnant women and 4 normal men ranging in age from 18 to 46. It has been previously shown in adults that there are no sex- or age-related differences in serum NSILA levels (Schalch et al. 1978). -
Assay procedures All blood samples were centrifuged as quickly as possible, and the sera separated and stored at -20°C until assayed. Most values are expressed as the mean ± sem. Statistical differences between two groups were calculated using Student's t-test, while either coefficients of correlation or multiple regression analyses were used to com¬ pare multiple points within or between various groups. A brief description of the assay techniques used follows.
A. NSILA. Acid soluble serum NSILA was measured by a competitive protein binding assay, the details of which have been published elsewhere (Schalch et al. 1978). Sera were first chromatographed on Sephadex G-50 (fine) in 1 m acetic acid (pH 2.3), and those fractions eluting between 45 and 70% of total column volume were pooled, lyophilized and reconstituted in 0.15 M HOAc/0.I5 M NaCl prior to assay. The NSILA preparation used for iodination had a purity of ^200 mU/mg, while that used for standards was - ,3.8 mU/ml 1) Both preparations contained appro¬ ximately equimolar amounts of IGF I and II. Partially purified NSILA-free carrier protein employed in the assay was derived from Cohn fraction IV of human plasma (kindly supplied by Armour Pharmaceutical Co., Kankakee, 111.) by chromatography on Sephadex G-200 in 0.15 M HOAc/0.15 m NaCl and concentrated by ultrafiltration (Amicon CH 3 Hollow fiber concentrator, membrane cut-off 10 000). Two limitations inherent to the design of this competitive protein binding assay must be considered in interpreting its results: 1) since both tracer and standard NSILA preparations contain (IGF) peptides I and II, this assay cannot distinguish between these two moieties in serum eluates, but simply measures "total" acid-dissociabe NSILA activity, and 2) since it has been shown recently that MSA cross-reacts with NSILA in a similar competitive protein binding system (Recider et al. 1977), this procedure is not specific for NSILA per se but rather for NSILA "activity". It should be pointed out that for reasons that are obscure, serum NSILA levels measured by our assay system are approximately twice those measured by a similar procedure (Zapi et al. 1977), though values in subjects with GH excess or deficiency are quite com¬ parable relative to normals. B. Hormones. Serum cortisol was assayed by the method of Murphy (1967), and human GH by the double antibody method of Schalch Se Parker (1964). Human GH antiserum with low cross-reactivity to human chorionic somatomammotrophin was chosen for assays of maternal sera with high concentrations of this hormone. -
-
RESULTS
NSILA, GH and cortisol in maternal
sera. No significant differences NSILA levels between stages of pregnancy (Fig. 1). appeared to exist in total There was neither a significant correlation between serum NSILA concentra¬ tion and week of pregnancy, nor a significant difference between the mean (± sem) NSILA level in the third trimester (1207 1 106 µ /mì) and that in the first trimester (959 1 70 ¿/U/ml, P>0.05). However, in the majority of subjects the NSILA was above the range established for adult control subjects, and the mean value throughout pregnancy (1106 1 46 ¿iU/ml) was significantly higher (P
