- Email: [email protected]
Low somatomedin activity in cord serum from infants with intrauterine growth retardation
March 1980 TheJournalofPEDIATRICS
605
Low somatomedin activity in cord serum from infants with intrauterine growth retardation Somatomedin activity was determined by the simultaneous incorporation o f 3aS-sulflte and all-methyl thymidine into costal cartilage from hypophysectomized rats in cord sera from term and preterm infants and infants with intrauterine growth retardation. Mean Sm activity by sulfate incorporation was 0.49 +_ 0.04, 0.35 +_ 0.05, and 0.09 +_ 0.03 units/ml (+_ SE) in the term, preterm, and IGR cord sera, respectively. The levels for each group were significantly different from each of the other groups. There was no significant difference between the mean Sm activit/ by thymidine incorporation in cord sera from term (0.92 • 0.09 units/ml) and preterm (0.87 + 0.08 units/ml) infants. These levels were significantly higher, however, than the Sm activity by sulfate incorporation for the respective groups, P < 0.001 for both groups. The mean Sm activity by thymidine incorporation in cord sera for IGR infants was 0.36 • 0.13 units/ml, and significantly lower than the levels" in cord sera o f term and preterm infants (P < 0.01). Inhibition of Sm activity by mixing cord serum and pooled adult serum was found in one of the two cord specimens tested from IGR infants. The low levels" of Sm activity in cord sera from IGR infants may reflect altered intrauterine nutrition. The discrepancy in the thymidine and sulfate incorporation by the costal cartilage bioassay for term and preterm cord sera might result from Sin-like factors in human fetal serum with greater mitogenic or thymidine transport activity compared to the activity for proteoglycan synthesis in cartilage.
Thomas P. Foley, Jr., M.D.,* Robert DePhilip, Ph.D., Anita Perricelli, B.S., and Arthur Miller, B.S., Pittsburgh, Pa.
THE MECHANISMS that regulate fetal and perinatal growth in the human are poorly understood. As early as 8.5 weeks of gestation human growth hormone has been demonstrated in the sella turcica by immunoelectrophoretic techniques? By ten weeks of gestation, immunoreactive hGH in fetal serum was reported at a concentration of 14.5 ng/ml.'-' Although pituitary content of hGH increases progressively during gestation, immunoreactive hGH in fetal serum increases to peak concentrations between 20 and 24 weeks Of gestation, then decreases to term. ~ In umbilical cord sera, hGH concentrations are considerably From the Department o f Pediatrics, School o f Medicine, University o f Pittsburgh, and Children's Hospital of Pittsburgh. Supported in part by the Renziehausen Fund, the Press Old Newsboys Fund, the Health Research Services Foundation and United States Public Health Service Grant RR-05507. *Reprint address: Children's Hospital of Pittsburgh, 125 DeSoto St., Pittsburgh, PA 15213.
0022-3476/80/030605 + 06500.60/0 9 1980 The C. V. Mosby Co.
higher than the levels reported in paired maternal sera or in sera from infants after the neonatal period. ~-~ Clinical and experimental evidence indicates that normal fetal growth is not dependent upon either maternal or fetal growth hormone. Birth length is ~usually normal after fetal hypophysectomy in monkeys, sheep, rats, and rabbits, ~-~ and in congenital growth hormone Abbreviations used Sm: somatomedin IGR: intrauterine growth retardation hGH: human growth hormone MSA: multiplication stimulating activity deficiency in mice. 1~ Body proportions including birth length are normal in the human fetus with absence of the pituitary or with anencephaly.-~11 During postnatal life, the hGH-dependent peptide, somatomedin, is essential for normal linear growth and stimulates several metabolic processes in cartilage in vitro. 12 Since hGH is not essential for fetal growth, other hormones or factors may stimulate
VoL 9~ No. 3, part Z pp. 605-610
606
Foley et al.
The Journal of Pediatrics March 1980
.[
C)
-
~-
I
I
,.oo .90
I
o
O..
I
I
I
I
I
.
.
9
.frO
.40 hul
I
9
9
~l
0
.50
z
I
I
I
!
80 .70
O
I
o GA UNCERTAIN IGR 9 GA KNOWN 0 GA UNCERTAIN 9 ANENCEPHALY
o 1.10 oc
I
9 NORMAL TERM PREMATURE 9 GA KNOWN
-
o
.30
-o .20 .10
9
~ "
Oo
~
o
~
"
9
I man Inn I
9
o
I
9
?
mg
9
t
"
26 2[7 28 2~9 3~0 31 32 33 34 35 26 37 38 39 40 41 42 GESTATION
Fig. 1. Somatomedin potency ratio of 3~S-sulfateincorporation for individual term, preterm, and IGR infants is plotted against gestational age (GA). When the GA by the last normal menstrual cycle and physical examination of the infant did not agree within two weeks, the GA from the physical examination was selected and designated as "GA uncertain." Sm production in the fetus. The induction of Sm in hypophysectomized rats by ovine placental lactogen suggests a role of the latter in the control of fetal growth. 13 Hereditary dwarfism with elevated levels of hGH but decreased Sm activity in serum is associated with retarded birth length. 1~ Compared to normal pooled adult sera, however, the levels of Sm are low in cord sera from term infants when determined by bioassay, 1~-1~ placental radioreceptor assayr 9-2~and radioimmunoassay.~ In one study the Sm levels by radioreceptor assay in term and preterm infants were not significantly different; however, the Sm levels in term infants who were small for gestational age were significantly lower than the Sm levels in term infants whose weight was appropriate for gestational age? ~ Low levels of Sm have been reported during starvation in the experimental animal and man, and increase promptly on refeeding? 2..... ~ Therefore, Sm or other related peptides with Smqike activity may be the regulators of fetal and neonatal growth. In this study we describe the Sm biologic activity in umbilical cord sera from normal term and preterm infants and infants with intrauterine growth retardation to further elucidate the role of gestation and fetal nutrition on Sm activity at birth in the human. METHODS The method for collection of umbilical venous blood, the assessment of gestational age, and the criteria for the
diagnosis of IGR or fetal growth retardation have been reported previously.'-'7~-~Gestational age was determined by the last normal menstrual period and the physical examination of the infant at birth. If these two determinants of gestational age did not agree within two weeks, the gestational age by physical examination alone was selected. Biologic activity of Sm was determined by the simultaneous incorporation of :~S-sulfate (Cambridge Nuclear) and :~H-methyl thymidine (New England Nuclear) into costal cartilage segments from 21- to 25-day-old male Sprague Dawley rats at least ten days following hypophysectomy.29-a~ Standard and unknown samples were determined in triplicate at dilutions of 9, 3, and 1%, respectively, with a balanced amino acid-sucrose-electrolyte medium containing 100 units/ml of penicillin and 100 ~g/ml of streptomycin and kanamycin after passage through a 0.45 micron Millipore filter. After complete dissection of the cartilagenous portion of the ribs, each cartilage at least 3 mm from the osseous portion of the rib was divided into 2 to 3 mm segments. Using tables of random distribution of numbers, three cartilage segments were randomly added 'to each sterile 12 x 75 mm tube containing a total volume of 1 ml of medium plus sample. After 12 to 18 hours of incubation at 37 ~ C in a shaking metabolic incubator, 5/~1 of medium containing from 2.0 to 2.4 x 106 DPM each of 3~S-sulfate and 3H-methyl thymidine were added, and the incubation continued for
Volume 96 Number 3, part 2
Somatomedin in intrauterine growth retardation
[
r
I
1
I
I
I
1
I
607
F
1.60 9 NORMAL TERM PREMATURE 9 6A KNOWN o GA UNCERTAIN IGR OGA KNOWN OGA UNCERTAIN
1.50 1.40 z
O
].30
1,20 o
Q-
l. IO
0
1 1
0
he-
S
~.oo
-
190
la.I Z
0
80 >-
o
~e
1
!
.70
Il
O 1
60 .50
0
et...
.40 z LIA I"--
o
.30
9
0
9
40
.20
o_
,10
.
o** I
1
1
I
i
I
I
I
I
1
I
l
9I
I
I
I
I
26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 GESTATION Fig. 2. Somatomedin potency ratio of 3H-methyl thymidine incorporation for individual term, preterm, and IGR infants is plotted against gestational age (GA). When the GA by the last normal menstrual cycle and physical examination of the infant did not agree within two weeks, the GA from the physical examination was selected and designated as "GA uncertain."
an additional 24 hours. Cartilagenous segments were removed from the incubation medium, heated in a boiling water bath for 20 minutes, washed in running tap water for 6 to 8 hours, and dried overnight. After weighing the cartilagenous segments for each sample to the nearest microgram (Cahn G-2 Electrobalance, Ventron Instruments Corp., Paramount, Calif.), the segments were placed into scintillation bottles to which 0.4 ml of 91% formic acid was added. The mouth of each bottle was covered with a marble and the bottles were heated in a boiling water bath for 45 minutes. After cooling, 0.4 ml of hyamine of hydroxide (Research Products) and scintillation cocktail were added and the bottles were counted in a Packard Scintillation counter which computes the DPM for each isotope. The data were analyzed by the DEC-10 computer of the University of Pittsburgh utilizing the program of Van Wyk et aP ~ to determine potency ratios between unknown and standard and the 95% fiducial fimits from symmetric four- and six-point bioassay design.31 Results were not acceptable if there were significant deviations from parallelism, opposite curvature, or linearity, or if the lambda value of assay precision exceeded 0.40.
RESULTS The mean Sm activity in the cord sera as determined by sulfate incorporation was 0.49 +_ 0.04 (_+ SE) units per ml for 38 normal term infants. The Sm activity for 20 preterm infants was 0.35 _+ 0.05 units/ml, significantly less than the activity in the cord sera from term infants, P < 0.05 (Table). There was considerable overlap, however, of individual levels of Sm activity for term and preterm infants (Fig. 1). For nine infants with IGR, the mean Sm activity of 0.09 _+ 0.03 units/ml was significantly less than both term (P < 0.001) and preterm (P < 0.001) infants with minimal overlap of individual values for the IGR infants as compared to the term and preterm infants (Fig. 1). The Sm activity in the cord serum from an anencephatic preterm infant was 0.59 units/ml, and the laGH concentration in the same specimen was only 0.4 ng/ml, The Sm activity in cord sera as determined by thymidine incorporation was significantly higher for term and preterm infants when compared to the Sm activity as determined by sulfate incorporation, P < 0.001 for both groups (Table). The mean Sm activity of 0.92 _+ 0.09 and 0.87 _+ 0.08 units per ml for term and preterm infants, respectively, was not significantly different, P > 0.1. The
608
Foley et al.
The Journal of Pediatrics March 1980
Table. Somatomedin in cord serum
Normal Premature IGR Normal vs premature Normal vs IGR Premature vs IGR
No.
I Mean
l + SEM
38 20 9
0.49 0.35 0.09
_+0.04 _+0.05 -+0.03
Sulfate ps
Thymidine
Sulfate p
9 19 12 <0.05 <0.001 <0.001
0,92 0.87 0.36
• 0,09 _+0.08 __0.13
thymidine P value
r
<0.001 <0.00l >0.05 >0.1 <0.01 <0.01
Abbreviation used: IGR = Intrauterine growth retardation.
mean Sm activity of 0.36 _+ 0.13 units per ml for 12 infants with IGR, however, was significantly lower than values for both the term (P < 0.01) and preterm (P < 0.01) infants although the mean Sm ~activity as determined by sulfate and thymidine incorporation did not significantly differ, P > 0.05. Although there was more scatter in the levels of Sm activity by thymidine incorporation for term and preterm infants, there was very little overlap with the levels of Sm activity for infants with IGR (Fig. 2). To further investigate the differences in Sm activities for the three groups of infants, we studied the inhibitory effect of the addition of cord serum from term, preterm, and IGR infants to an adult serum pool as reported by Van den Brande and DuCaju'-':' in the study of malnourished children. To a 10% concentration of an adult serum pool we added either medium or cord serum to concentrations of 2.5, 5.0, and 10%, respectively. Inhibition of the incorporation of sulfate and thymidine was found at the 10% concentration in one of the two cord sera specimens from infants with IGR (Figs. 3 and 4). One cord serum from a preterm infant was inhibitory for thymidine incorporation (Fig. 3), but not for sulfate incorporation (Fig. 4), DISCUSSION Since the most rapid rate of linear growth in man occurs during fetal life, one would expect to find levels of Sm activity to be the highest at this age. On the contrary, however, there are several reports of low Sm levels in cord serum when determined by bioassay, competitive membrane receptor assay, and radioimmunoassay.1~-~1 These observations raise important questions regarding the role of Sm in the growth of the fetus and the possibility that other factors might regulate growth during intrauterine life. The levels of Sm activity that we are reporting in cord sera from term infants are very similar to previous studies using costal cartilage bioassays2 ~..... ~ In addition, our observation of the significantly lower levels of Sm activity in preterm infants are in agreement with the report of
Gtuckman and Brinsmead ~6 in which the Sm activity in cord serum as measured by the porcine cartilage bioassaY correlated with weight, length, and head circumference at birth. However, D'Ercole et al TM did not find a difference between Sm levels in term and preterm infants as determined by the competitive membrane-binding assay. This discrepancy with our results using the cartilage bioassay might be explained by the differences in the specificity of the two assays for Sm-like activity i n addition to the differences in the patient population studied. The Sm levels in small for gestational age infants by the competitive membrane-binding assay were significantly lower than the levels for term infants; the differences were not nearly as striking, however, as those between IGR infants and term or preterm infants that we are reporting for Sm activity by sulfate incorporation in the costal cartilage bioassay. The very low levels of bioassayable Sm in the cord serum of IGR infants might be explained, in part, by impaired intrauterine nutrition. In the serum from children with malnutrition, low levels of Sm activity have been reported. 27-~4 The studies of Van den Brande and DuCaju 23 suggest that some of the decrement in Sm biologic activity in malnutrition may be caused by the presence in serum of inhibitors of the biologic action of Sm as measured in vitro. '-'3 Our studies suggest that this mechanism might contribute to the very low levels of Sm activity in one infant with IGR. The Sm activity in a 10% concentration of pooled adult serum decreased upon the addition of a 10% concentration of cord serum from an infant with IGR, and both thymidine and sulfate incorporation were similarly decreased. The addition o}" serum from another infant with IGR, however, had no effect. Unfortunately, an adequate volume of serum from additional infants with IGR was not available for study. The inhibitory effect on thymidine, but not sulfate incorporation by cord serum from a preterm infant, is difficult to explain without evidence of some process at birth that might selectively inhibit thymidine transport or DNA synthesis. Studies of the simultaneous incorporation of sulfate
Volume 96 Number 3, part 2
Sornatomedi~ in intrauterine growth retardation
609
175 125
150 1.1.1 uJ Z
o_
100 125
E)
100
>f_) z w t-o
>.'-'r>.-
75
13-J
O IX.
_< 75
Z
t--Z
5O
u_
o 50
25
25 0 0
2.
1
TERM 1
2
TERM
1
2
PRETERM
1
IGR
Fig. 3. The effect Of the addition to 10% pooled adult serum of 2.5, 5.0, and 10% of cord serum from term, preterm, and IGR infants upon ~H-methyl thymidine incorporation as a percentage of the thymidine incorporation of 10% pooled adult serum.
and thymidine into costal cartilage segments have shown similar effects upon these metabolic processes by stimulatory or inhibitory activity in vitro. 29 Our report that the potency ratio for thymidine incorporation is significantly greater than that for simultaneous sulfate incorporation by term and preterm cord sera suggests that there may exist in fetal serum mitogenic factors that have minimal effect upon the stimulation of sulfate incorporation into proteoglycans of cartilage. An analogous condition may exist in the rat. Stuart and co-workers ~'-' have reported levels of Sm activity in the sera from 21-day-old fetuses that are measurable only at the sensitivity of their assay (0.12 units/ml) in the costal cartilage bi0assay. These low levels begin to increase by day 4 after birth to reach the level found in hypophysectomized rat serum, and finally increase to the level of an adult rat by day 11. 32 Very recently, however, Moses and coworkers :':' have reported eIevated levels of the Sm, multipIication stimulating activity, in fetal rat serum by radioimmunoassay, rat liver m e m b r a n e radioreceptor assay, and competitive binding protein assay using rat serum Sm-binding protein. Immuno-reactive MSA levels were 10 to 60-fold higher in fetal rat serum (1 to 3/~g/ml at 20 and 21 days ' gestation)
2
1
2
PRETERM
1
2
IGR
Fig. 4. The effect of the addition to 10% pooled adult serum of 2.5, 5.0, and 10% of cord serum from term, preterm, and IGR infants upon 3~S-sulfate incorporation as a percentage of the Sulfate incorporation of 10% pooled adult serum.
compared to maternal levels, and did not begin to decrease to adult levels until days 10 to 20 of extrauterine life. They have also reported the synthesis of MSA by 19 tO 20-day-old fetal rat liver explants in organ culture? ~ MSA stimulates both thymidine incorporation into D N A in chick embry o fibroblasts and sulfate incorporation into costal cartilage from hypophysectomized rats. 3~'The lesser potency of MSA compared to the other somatomedins for sulfate incorporation in the costal cartilage bioassay would be in agreement with the low Sm levels in fetal rat serum reported by Stuart and co-workers. 3~ MSA may play an important role in the regulation of fetal and early extrauterine growth in the rat. The discrepancy in thymidine and sulfate incorporation by costal cartilage for term and preterm cord sera that we are reporting might result from a MSA-like factor or factors in h u m a n fetal serum which has greater mitogenic or thymidine transport activity compared to the activity for prote0glycan synthesis in cartilage. We are grateful to Dr. Judson J. Van Wyk tbr the use of the computer program for the computation and statistical analysis of the double isotope somatomedin bioassay, and appreciate the assistance of Dr. Alexander C, Allen in the use of the computer program for the identity of the patients whose cord serum specimens were analyzed in this study.
6 10
Foley et al.
REFERENCES
1. Gitlin D, and Biasucci A: Ontogenesis of immunoreactive growth hormone, follicle-stimulating hormone, thyroidstimulating hormone, luteinizing hormone, chorionic prolactin, and chorionic gonadotropin in the human conceptus. J Clin Endocrinol Metab 29:926, 1969. 2. Kaplan SL, Grumbach MM, and Shepard TH: The ontogenesis of human fetal hormones. I. Growth hormone and insulin, J Clin Invest 51:3080, 1972. 3. Glick SM, Roth J, Yalow RS, et al: Immunoassay of human growth hormone in plasma, Nature (London) 199:784, 1963. 4. Greenwood FC, Hunter WM, and Klopper A: Assay of human growth hormone in pregnancy, at parturition and in lactation. Detection of a growth hormone-like substance from the placenta, Br Med J 1:22, 1964. 5. Cornblath M, Parker ML, Reisner SH, et al: Secretion and metabolism of growth hormone in premature and full-term infants, J Chn Endocrinol Metab 25:209, 19651 6. Chez RH, Hutchinson DL, Salazar H, et al: Some effects of fetal and maternal hypophysectomy in pregnancy, Am J Obstet Gynec01 108:643, 1970. 7. Liggins GC, and Kennedy PC: Effects of electrocoagulation of the foetal lamb hypophysis on growth and development, J Endocrinol 40:371, 1968. 8. Lanman JT, and Schaffer A: Gestational effects of fetal decapitation in sheep, Fertil Steril 19:598, 1968. 9. Jost A: Hormonal factors in the development of the fetus, Cold Spring Harbor Symp Quant Biol 19:167, 1954. 10. deBeer GR, and Gruneberg H: A note on pituitary dwarfism in the mouse, J Genet 39:297, 1940. 11. Hoet JJ: Normal and abnormal foetal weight gain, in Wolstenholme GEW, and O'Connor M, editors: Foetal autonomy, London, 1969, J & A Churchill. Ltd, p 186. 12. Van Wyk JJ, and Underwood LE: The somatomedjns and their actions, in Litwack G, editor: Biochemical actions of hormones, vol V, New York 1978, Academic Press, Inc., pp 101-148. 13. Hurley TW, D'Erco|e AJ, Handwerger S, et al: Ovine placental lactogen induces somatomedin: a possible role in fetal growth, Endocrinology 101:1635, 1977. 14. Laron Z, Pertzelan A, and Karp M: Pituitary dwarfism with high serum levels of growth hormone, Isr J Med Sci 4:883, 1968. 15. Giordano G, Foppiani E, Minuto F, et al: Growth hormone and somatomedin behavior in the newborn, Acta Endocrinol (Kbh) 81:449, 1976. 16. Gluckman PD, and Brinsmead MW: Somatomedin in cord blood: relationship to gestational age and birth size, J Clin Endocrinol Metab 43:1378, 1976. 17. Hintz RL, Seeds JM, and Johnsonbaugh RE: Somatomedin and growth hormone in the newborn, Am J Dis Child 131:1249, 1977. 18. Bala RM, Wright C, Bardai A, et al: Somatomedin bioactivity in serum and amnionic fluid during pregnancy, J Clin Endocrinol Metab 46:649, 1978.
The Journal o f Pediatrics March 1980
19. D'Ercole AJ, Foushee DB, and Underwood LE: Somatomedin-C receptor ontogeny and levels in porcine fetal and human cord serum, J Clin Endocrinol Metab 43:1069, 1976." 20. Svan H, Hall K, Ritzien M, et al: Somatomedin A and B in serum from neonates, their mothers and cord blood, Acta Endocrinoi (Kbh) 85:636, 1977. 21. Furlanetto RW, Underwood LE, Van Wyk JJ, et al: Estimation of somatomedin-C leyels in normals and patients with pituitary disease by radioimmunoassay, J Clin Invest 60:648, 1977. 22. Grant DB, Hambley J, Becker D, et al: Reduced sulfation factor in undernourished children, Arch Dis Child 48:596, 1973. 23. Van den Brande JL, and DuCaju MVL: Plasma somatomedin activity in children with growth disturbances, in Raiti S, editor: Advances in human growth hormone research, DHEW Publ No NIH 74-612, Washington, DC, 1974, US Govt Printing Office,. pp 98. 24. Hintz RL, Suskind R, Amatayakul K, Thanangkul O, and Olson R: Plasma somatomedin and growth hormone values in children with protein-calorie malnutrition. J PEDIAWR 92:153. 1978. 25. Phillips LS. and Young HS: Nutrition and somatomedin. I. Effects of fasting and refeeding on serum somatomedin in rats. Endocrinol 99:304. 1976. 26. Phillips LS. Orawski AT. and Belosky DC: Somatomedin and nutrition. IV. Regulation of somatomedin activity and growth cartilage activity by quantity and composition of diet in rats. Endocrinology 103:121. 1978. 27. Klein AH. Agustin AV. and Foley TP Jr: Successful laboratory screening for congenital hypothyroidism. Lancet 2:77. 1974. 28. Alien AC: Intrauterine failure to thrive, in Brennemann's practice of pediatrics, vol 1. New York. 1973. Harper & Row. Publishers. Inc. chapter 87. 29. Van den Brande JL. Van Wyk JJ. Weaver RP. et al: Partial characterization of sulfation and thymidine factors in acromegalic plasma. Acta End0crinol 66:65. 1971. 30. Van Wyk JJ. Hall K. and Van den Brande JL: Further purification and characterization of sulfation factor and thymidine factor from acromegalic plasma, J Clin Endocrinol Metab 32:389. 1971. 31. Finney DJ: Statistical method in biological assay, ed 2. London. 1964. Charles Griffin & Co Ltd. 32 Stuart MC. Lazarus L. Moore SS. el al: Somatomedin production in the neonatal rat. Horm Metab Res 8:442. 1976. 33. Moses AC. Nissley SP. Rechler MM. et al: Elevated levels{ of the insulin-like growth factor, multiplication stimulatin~ activity, in fetal rat serum. Clin Res 27:488A. 1979. 34. Rechler MM. Eisen HJ. Higa OZ. et al: Characterization of a somatomedin (insulin-like growth factor) synthesized by fetal ral liver organ cultures, J Biol Chem (in pressl.
605
Low somatomedin activity in cord serum from infants with intrauterine growth retardation Somatomedin activity was determined by the simultaneous incorporation o f 3aS-sulflte and all-methyl thymidine into costal cartilage from hypophysectomized rats in cord sera from term and preterm infants and infants with intrauterine growth retardation. Mean Sm activity by sulfate incorporation was 0.49 +_ 0.04, 0.35 +_ 0.05, and 0.09 +_ 0.03 units/ml (+_ SE) in the term, preterm, and IGR cord sera, respectively. The levels for each group were significantly different from each of the other groups. There was no significant difference between the mean Sm activit/ by thymidine incorporation in cord sera from term (0.92 • 0.09 units/ml) and preterm (0.87 + 0.08 units/ml) infants. These levels were significantly higher, however, than the Sm activity by sulfate incorporation for the respective groups, P < 0.001 for both groups. The mean Sm activity by thymidine incorporation in cord sera for IGR infants was 0.36 • 0.13 units/ml, and significantly lower than the levels" in cord sera o f term and preterm infants (P < 0.01). Inhibition of Sm activity by mixing cord serum and pooled adult serum was found in one of the two cord specimens tested from IGR infants. The low levels" of Sm activity in cord sera from IGR infants may reflect altered intrauterine nutrition. The discrepancy in the thymidine and sulfate incorporation by the costal cartilage bioassay for term and preterm cord sera might result from Sin-like factors in human fetal serum with greater mitogenic or thymidine transport activity compared to the activity for proteoglycan synthesis in cartilage.
Thomas P. Foley, Jr., M.D.,* Robert DePhilip, Ph.D., Anita Perricelli, B.S., and Arthur Miller, B.S., Pittsburgh, Pa.
THE MECHANISMS that regulate fetal and perinatal growth in the human are poorly understood. As early as 8.5 weeks of gestation human growth hormone has been demonstrated in the sella turcica by immunoelectrophoretic techniques? By ten weeks of gestation, immunoreactive hGH in fetal serum was reported at a concentration of 14.5 ng/ml.'-' Although pituitary content of hGH increases progressively during gestation, immunoreactive hGH in fetal serum increases to peak concentrations between 20 and 24 weeks Of gestation, then decreases to term. ~ In umbilical cord sera, hGH concentrations are considerably From the Department o f Pediatrics, School o f Medicine, University o f Pittsburgh, and Children's Hospital of Pittsburgh. Supported in part by the Renziehausen Fund, the Press Old Newsboys Fund, the Health Research Services Foundation and United States Public Health Service Grant RR-05507. *Reprint address: Children's Hospital of Pittsburgh, 125 DeSoto St., Pittsburgh, PA 15213.
0022-3476/80/030605 + 06500.60/0 9 1980 The C. V. Mosby Co.
higher than the levels reported in paired maternal sera or in sera from infants after the neonatal period. ~-~ Clinical and experimental evidence indicates that normal fetal growth is not dependent upon either maternal or fetal growth hormone. Birth length is ~usually normal after fetal hypophysectomy in monkeys, sheep, rats, and rabbits, ~-~ and in congenital growth hormone Abbreviations used Sm: somatomedin IGR: intrauterine growth retardation hGH: human growth hormone MSA: multiplication stimulating activity deficiency in mice. 1~ Body proportions including birth length are normal in the human fetus with absence of the pituitary or with anencephaly.-~11 During postnatal life, the hGH-dependent peptide, somatomedin, is essential for normal linear growth and stimulates several metabolic processes in cartilage in vitro. 12 Since hGH is not essential for fetal growth, other hormones or factors may stimulate
VoL 9~ No. 3, part Z pp. 605-610
606
Foley et al.
The Journal of Pediatrics March 1980
.[
C)
-
~-
I
I
,.oo .90
I
o
O..
I
I
I
I
I
.
.
9
.frO
.40 hul
I
9
9
~l
0
.50
z
I
I
I
!
80 .70
O
I
o GA UNCERTAIN IGR 9 GA KNOWN 0 GA UNCERTAIN 9 ANENCEPHALY
o 1.10 oc
I
9 NORMAL TERM PREMATURE 9 GA KNOWN
-
o
.30
-o .20 .10
9
~ "
Oo
~
o
~
"
9
I man Inn I
9
o
I
9
?
mg
9
t
"
26 2[7 28 2~9 3~0 31 32 33 34 35 26 37 38 39 40 41 42 GESTATION
Fig. 1. Somatomedin potency ratio of 3~S-sulfateincorporation for individual term, preterm, and IGR infants is plotted against gestational age (GA). When the GA by the last normal menstrual cycle and physical examination of the infant did not agree within two weeks, the GA from the physical examination was selected and designated as "GA uncertain." Sm production in the fetus. The induction of Sm in hypophysectomized rats by ovine placental lactogen suggests a role of the latter in the control of fetal growth. 13 Hereditary dwarfism with elevated levels of hGH but decreased Sm activity in serum is associated with retarded birth length. 1~ Compared to normal pooled adult sera, however, the levels of Sm are low in cord sera from term infants when determined by bioassay, 1~-1~ placental radioreceptor assayr 9-2~and radioimmunoassay.~ In one study the Sm levels by radioreceptor assay in term and preterm infants were not significantly different; however, the Sm levels in term infants who were small for gestational age were significantly lower than the Sm levels in term infants whose weight was appropriate for gestational age? ~ Low levels of Sm have been reported during starvation in the experimental animal and man, and increase promptly on refeeding? 2..... ~ Therefore, Sm or other related peptides with Smqike activity may be the regulators of fetal and neonatal growth. In this study we describe the Sm biologic activity in umbilical cord sera from normal term and preterm infants and infants with intrauterine growth retardation to further elucidate the role of gestation and fetal nutrition on Sm activity at birth in the human. METHODS The method for collection of umbilical venous blood, the assessment of gestational age, and the criteria for the
diagnosis of IGR or fetal growth retardation have been reported previously.'-'7~-~Gestational age was determined by the last normal menstrual period and the physical examination of the infant at birth. If these two determinants of gestational age did not agree within two weeks, the gestational age by physical examination alone was selected. Biologic activity of Sm was determined by the simultaneous incorporation of :~S-sulfate (Cambridge Nuclear) and :~H-methyl thymidine (New England Nuclear) into costal cartilage segments from 21- to 25-day-old male Sprague Dawley rats at least ten days following hypophysectomy.29-a~ Standard and unknown samples were determined in triplicate at dilutions of 9, 3, and 1%, respectively, with a balanced amino acid-sucrose-electrolyte medium containing 100 units/ml of penicillin and 100 ~g/ml of streptomycin and kanamycin after passage through a 0.45 micron Millipore filter. After complete dissection of the cartilagenous portion of the ribs, each cartilage at least 3 mm from the osseous portion of the rib was divided into 2 to 3 mm segments. Using tables of random distribution of numbers, three cartilage segments were randomly added 'to each sterile 12 x 75 mm tube containing a total volume of 1 ml of medium plus sample. After 12 to 18 hours of incubation at 37 ~ C in a shaking metabolic incubator, 5/~1 of medium containing from 2.0 to 2.4 x 106 DPM each of 3~S-sulfate and 3H-methyl thymidine were added, and the incubation continued for
Volume 96 Number 3, part 2
Somatomedin in intrauterine growth retardation
[
r
I
1
I
I
I
1
I
607
F
1.60 9 NORMAL TERM PREMATURE 9 6A KNOWN o GA UNCERTAIN IGR OGA KNOWN OGA UNCERTAIN
1.50 1.40 z
O
].30
1,20 o
Q-
l. IO
0
1 1
0
he-
S
~.oo
-
190
la.I Z
0
80 >-
o
~e
1
!
.70
Il
O 1
60 .50
0
et...
.40 z LIA I"--
o
.30
9
0
9
40
.20
o_
,10
.
o** I
1
1
I
i
I
I
I
I
1
I
l
9I
I
I
I
I
26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 GESTATION Fig. 2. Somatomedin potency ratio of 3H-methyl thymidine incorporation for individual term, preterm, and IGR infants is plotted against gestational age (GA). When the GA by the last normal menstrual cycle and physical examination of the infant did not agree within two weeks, the GA from the physical examination was selected and designated as "GA uncertain."
an additional 24 hours. Cartilagenous segments were removed from the incubation medium, heated in a boiling water bath for 20 minutes, washed in running tap water for 6 to 8 hours, and dried overnight. After weighing the cartilagenous segments for each sample to the nearest microgram (Cahn G-2 Electrobalance, Ventron Instruments Corp., Paramount, Calif.), the segments were placed into scintillation bottles to which 0.4 ml of 91% formic acid was added. The mouth of each bottle was covered with a marble and the bottles were heated in a boiling water bath for 45 minutes. After cooling, 0.4 ml of hyamine of hydroxide (Research Products) and scintillation cocktail were added and the bottles were counted in a Packard Scintillation counter which computes the DPM for each isotope. The data were analyzed by the DEC-10 computer of the University of Pittsburgh utilizing the program of Van Wyk et aP ~ to determine potency ratios between unknown and standard and the 95% fiducial fimits from symmetric four- and six-point bioassay design.31 Results were not acceptable if there were significant deviations from parallelism, opposite curvature, or linearity, or if the lambda value of assay precision exceeded 0.40.
RESULTS The mean Sm activity in the cord sera as determined by sulfate incorporation was 0.49 +_ 0.04 (_+ SE) units per ml for 38 normal term infants. The Sm activity for 20 preterm infants was 0.35 _+ 0.05 units/ml, significantly less than the activity in the cord sera from term infants, P < 0.05 (Table). There was considerable overlap, however, of individual levels of Sm activity for term and preterm infants (Fig. 1). For nine infants with IGR, the mean Sm activity of 0.09 _+ 0.03 units/ml was significantly less than both term (P < 0.001) and preterm (P < 0.001) infants with minimal overlap of individual values for the IGR infants as compared to the term and preterm infants (Fig. 1). The Sm activity in the cord serum from an anencephatic preterm infant was 0.59 units/ml, and the laGH concentration in the same specimen was only 0.4 ng/ml, The Sm activity in cord sera as determined by thymidine incorporation was significantly higher for term and preterm infants when compared to the Sm activity as determined by sulfate incorporation, P < 0.001 for both groups (Table). The mean Sm activity of 0.92 _+ 0.09 and 0.87 _+ 0.08 units per ml for term and preterm infants, respectively, was not significantly different, P > 0.1. The
608
Foley et al.
The Journal of Pediatrics March 1980
Table. Somatomedin in cord serum
Normal Premature IGR Normal vs premature Normal vs IGR Premature vs IGR
No.
I Mean
l + SEM
38 20 9
0.49 0.35 0.09
_+0.04 _+0.05 -+0.03
Sulfate ps
Thymidine
Sulfate p
9 19 12 <0.05 <0.001 <0.001
0,92 0.87 0.36
• 0,09 _+0.08 __0.13
thymidine P value
r
<0.001 <0.00l >0.05 >0.1 <0.01 <0.01
Abbreviation used: IGR = Intrauterine growth retardation.
mean Sm activity of 0.36 _+ 0.13 units per ml for 12 infants with IGR, however, was significantly lower than values for both the term (P < 0.01) and preterm (P < 0.01) infants although the mean Sm ~activity as determined by sulfate and thymidine incorporation did not significantly differ, P > 0.05. Although there was more scatter in the levels of Sm activity by thymidine incorporation for term and preterm infants, there was very little overlap with the levels of Sm activity for infants with IGR (Fig. 2). To further investigate the differences in Sm activities for the three groups of infants, we studied the inhibitory effect of the addition of cord serum from term, preterm, and IGR infants to an adult serum pool as reported by Van den Brande and DuCaju'-':' in the study of malnourished children. To a 10% concentration of an adult serum pool we added either medium or cord serum to concentrations of 2.5, 5.0, and 10%, respectively. Inhibition of the incorporation of sulfate and thymidine was found at the 10% concentration in one of the two cord sera specimens from infants with IGR (Figs. 3 and 4). One cord serum from a preterm infant was inhibitory for thymidine incorporation (Fig. 3), but not for sulfate incorporation (Fig. 4), DISCUSSION Since the most rapid rate of linear growth in man occurs during fetal life, one would expect to find levels of Sm activity to be the highest at this age. On the contrary, however, there are several reports of low Sm levels in cord serum when determined by bioassay, competitive membrane receptor assay, and radioimmunoassay.1~-~1 These observations raise important questions regarding the role of Sm in the growth of the fetus and the possibility that other factors might regulate growth during intrauterine life. The levels of Sm activity that we are reporting in cord sera from term infants are very similar to previous studies using costal cartilage bioassays2 ~..... ~ In addition, our observation of the significantly lower levels of Sm activity in preterm infants are in agreement with the report of
Gtuckman and Brinsmead ~6 in which the Sm activity in cord serum as measured by the porcine cartilage bioassaY correlated with weight, length, and head circumference at birth. However, D'Ercole et al TM did not find a difference between Sm levels in term and preterm infants as determined by the competitive membrane-binding assay. This discrepancy with our results using the cartilage bioassay might be explained by the differences in the specificity of the two assays for Sm-like activity i n addition to the differences in the patient population studied. The Sm levels in small for gestational age infants by the competitive membrane-binding assay were significantly lower than the levels for term infants; the differences were not nearly as striking, however, as those between IGR infants and term or preterm infants that we are reporting for Sm activity by sulfate incorporation in the costal cartilage bioassay. The very low levels of bioassayable Sm in the cord serum of IGR infants might be explained, in part, by impaired intrauterine nutrition. In the serum from children with malnutrition, low levels of Sm activity have been reported. 27-~4 The studies of Van den Brande and DuCaju 23 suggest that some of the decrement in Sm biologic activity in malnutrition may be caused by the presence in serum of inhibitors of the biologic action of Sm as measured in vitro. '-'3 Our studies suggest that this mechanism might contribute to the very low levels of Sm activity in one infant with IGR. The Sm activity in a 10% concentration of pooled adult serum decreased upon the addition of a 10% concentration of cord serum from an infant with IGR, and both thymidine and sulfate incorporation were similarly decreased. The addition o}" serum from another infant with IGR, however, had no effect. Unfortunately, an adequate volume of serum from additional infants with IGR was not available for study. The inhibitory effect on thymidine, but not sulfate incorporation by cord serum from a preterm infant, is difficult to explain without evidence of some process at birth that might selectively inhibit thymidine transport or DNA synthesis. Studies of the simultaneous incorporation of sulfate
Volume 96 Number 3, part 2
Sornatomedi~ in intrauterine growth retardation
609
175 125
150 1.1.1 uJ Z
o_
100 125
E)
100
>f_) z w t-o
>.'-'r>.-
75
13-J
O IX.
_< 75
Z
t--Z
5O
u_
o 50
25
25 0 0
2.
1
TERM 1
2
TERM
1
2
PRETERM
1
IGR
Fig. 3. The effect Of the addition to 10% pooled adult serum of 2.5, 5.0, and 10% of cord serum from term, preterm, and IGR infants upon ~H-methyl thymidine incorporation as a percentage of the thymidine incorporation of 10% pooled adult serum.
and thymidine into costal cartilage segments have shown similar effects upon these metabolic processes by stimulatory or inhibitory activity in vitro. 29 Our report that the potency ratio for thymidine incorporation is significantly greater than that for simultaneous sulfate incorporation by term and preterm cord sera suggests that there may exist in fetal serum mitogenic factors that have minimal effect upon the stimulation of sulfate incorporation into proteoglycans of cartilage. An analogous condition may exist in the rat. Stuart and co-workers ~'-' have reported levels of Sm activity in the sera from 21-day-old fetuses that are measurable only at the sensitivity of their assay (0.12 units/ml) in the costal cartilage bi0assay. These low levels begin to increase by day 4 after birth to reach the level found in hypophysectomized rat serum, and finally increase to the level of an adult rat by day 11. 32 Very recently, however, Moses and coworkers :':' have reported eIevated levels of the Sm, multipIication stimulating activity, in fetal rat serum by radioimmunoassay, rat liver m e m b r a n e radioreceptor assay, and competitive binding protein assay using rat serum Sm-binding protein. Immuno-reactive MSA levels were 10 to 60-fold higher in fetal rat serum (1 to 3/~g/ml at 20 and 21 days ' gestation)
2
1
2
PRETERM
1
2
IGR
Fig. 4. The effect of the addition to 10% pooled adult serum of 2.5, 5.0, and 10% of cord serum from term, preterm, and IGR infants upon 3~S-sulfate incorporation as a percentage of the Sulfate incorporation of 10% pooled adult serum.
compared to maternal levels, and did not begin to decrease to adult levels until days 10 to 20 of extrauterine life. They have also reported the synthesis of MSA by 19 tO 20-day-old fetal rat liver explants in organ culture? ~ MSA stimulates both thymidine incorporation into D N A in chick embry o fibroblasts and sulfate incorporation into costal cartilage from hypophysectomized rats. 3~'The lesser potency of MSA compared to the other somatomedins for sulfate incorporation in the costal cartilage bioassay would be in agreement with the low Sm levels in fetal rat serum reported by Stuart and co-workers. 3~ MSA may play an important role in the regulation of fetal and early extrauterine growth in the rat. The discrepancy in thymidine and sulfate incorporation by costal cartilage for term and preterm cord sera that we are reporting might result from a MSA-like factor or factors in h u m a n fetal serum which has greater mitogenic or thymidine transport activity compared to the activity for prote0glycan synthesis in cartilage. We are grateful to Dr. Judson J. Van Wyk tbr the use of the computer program for the computation and statistical analysis of the double isotope somatomedin bioassay, and appreciate the assistance of Dr. Alexander C, Allen in the use of the computer program for the identity of the patients whose cord serum specimens were analyzed in this study.
6 10
Foley et al.
REFERENCES
1. Gitlin D, and Biasucci A: Ontogenesis of immunoreactive growth hormone, follicle-stimulating hormone, thyroidstimulating hormone, luteinizing hormone, chorionic prolactin, and chorionic gonadotropin in the human conceptus. J Clin Endocrinol Metab 29:926, 1969. 2. Kaplan SL, Grumbach MM, and Shepard TH: The ontogenesis of human fetal hormones. I. Growth hormone and insulin, J Clin Invest 51:3080, 1972. 3. Glick SM, Roth J, Yalow RS, et al: Immunoassay of human growth hormone in plasma, Nature (London) 199:784, 1963. 4. Greenwood FC, Hunter WM, and Klopper A: Assay of human growth hormone in pregnancy, at parturition and in lactation. Detection of a growth hormone-like substance from the placenta, Br Med J 1:22, 1964. 5. Cornblath M, Parker ML, Reisner SH, et al: Secretion and metabolism of growth hormone in premature and full-term infants, J Chn Endocrinol Metab 25:209, 19651 6. Chez RH, Hutchinson DL, Salazar H, et al: Some effects of fetal and maternal hypophysectomy in pregnancy, Am J Obstet Gynec01 108:643, 1970. 7. Liggins GC, and Kennedy PC: Effects of electrocoagulation of the foetal lamb hypophysis on growth and development, J Endocrinol 40:371, 1968. 8. Lanman JT, and Schaffer A: Gestational effects of fetal decapitation in sheep, Fertil Steril 19:598, 1968. 9. Jost A: Hormonal factors in the development of the fetus, Cold Spring Harbor Symp Quant Biol 19:167, 1954. 10. deBeer GR, and Gruneberg H: A note on pituitary dwarfism in the mouse, J Genet 39:297, 1940. 11. Hoet JJ: Normal and abnormal foetal weight gain, in Wolstenholme GEW, and O'Connor M, editors: Foetal autonomy, London, 1969, J & A Churchill. Ltd, p 186. 12. Van Wyk JJ, and Underwood LE: The somatomedjns and their actions, in Litwack G, editor: Biochemical actions of hormones, vol V, New York 1978, Academic Press, Inc., pp 101-148. 13. Hurley TW, D'Erco|e AJ, Handwerger S, et al: Ovine placental lactogen induces somatomedin: a possible role in fetal growth, Endocrinology 101:1635, 1977. 14. Laron Z, Pertzelan A, and Karp M: Pituitary dwarfism with high serum levels of growth hormone, Isr J Med Sci 4:883, 1968. 15. Giordano G, Foppiani E, Minuto F, et al: Growth hormone and somatomedin behavior in the newborn, Acta Endocrinol (Kbh) 81:449, 1976. 16. Gluckman PD, and Brinsmead MW: Somatomedin in cord blood: relationship to gestational age and birth size, J Clin Endocrinol Metab 43:1378, 1976. 17. Hintz RL, Seeds JM, and Johnsonbaugh RE: Somatomedin and growth hormone in the newborn, Am J Dis Child 131:1249, 1977. 18. Bala RM, Wright C, Bardai A, et al: Somatomedin bioactivity in serum and amnionic fluid during pregnancy, J Clin Endocrinol Metab 46:649, 1978.
The Journal o f Pediatrics March 1980
19. D'Ercole AJ, Foushee DB, and Underwood LE: Somatomedin-C receptor ontogeny and levels in porcine fetal and human cord serum, J Clin Endocrinol Metab 43:1069, 1976." 20. Svan H, Hall K, Ritzien M, et al: Somatomedin A and B in serum from neonates, their mothers and cord blood, Acta Endocrinoi (Kbh) 85:636, 1977. 21. Furlanetto RW, Underwood LE, Van Wyk JJ, et al: Estimation of somatomedin-C leyels in normals and patients with pituitary disease by radioimmunoassay, J Clin Invest 60:648, 1977. 22. Grant DB, Hambley J, Becker D, et al: Reduced sulfation factor in undernourished children, Arch Dis Child 48:596, 1973. 23. Van den Brande JL, and DuCaju MVL: Plasma somatomedin activity in children with growth disturbances, in Raiti S, editor: Advances in human growth hormone research, DHEW Publ No NIH 74-612, Washington, DC, 1974, US Govt Printing Office,. pp 98. 24. Hintz RL, Suskind R, Amatayakul K, Thanangkul O, and Olson R: Plasma somatomedin and growth hormone values in children with protein-calorie malnutrition. J PEDIAWR 92:153. 1978. 25. Phillips LS. and Young HS: Nutrition and somatomedin. I. Effects of fasting and refeeding on serum somatomedin in rats. Endocrinol 99:304. 1976. 26. Phillips LS. Orawski AT. and Belosky DC: Somatomedin and nutrition. IV. Regulation of somatomedin activity and growth cartilage activity by quantity and composition of diet in rats. Endocrinology 103:121. 1978. 27. Klein AH. Agustin AV. and Foley TP Jr: Successful laboratory screening for congenital hypothyroidism. Lancet 2:77. 1974. 28. Alien AC: Intrauterine failure to thrive, in Brennemann's practice of pediatrics, vol 1. New York. 1973. Harper & Row. Publishers. Inc. chapter 87. 29. Van den Brande JL. Van Wyk JJ. Weaver RP. et al: Partial characterization of sulfation and thymidine factors in acromegalic plasma. Acta End0crinol 66:65. 1971. 30. Van Wyk JJ. Hall K. and Van den Brande JL: Further purification and characterization of sulfation factor and thymidine factor from acromegalic plasma, J Clin Endocrinol Metab 32:389. 1971. 31. Finney DJ: Statistical method in biological assay, ed 2. London. 1964. Charles Griffin & Co Ltd. 32 Stuart MC. Lazarus L. Moore SS. el al: Somatomedin production in the neonatal rat. Horm Metab Res 8:442. 1976. 33. Moses AC. Nissley SP. Rechler MM. et al: Elevated levels{ of the insulin-like growth factor, multiplication stimulatin~ activity, in fetal rat serum. Clin Res 27:488A. 1979. 34. Rechler MM. Eisen HJ. Higa OZ. et al: Characterization of a somatomedin (insulin-like growth factor) synthesized by fetal ral liver organ cultures, J Biol Chem (in pressl.