TROPICAL PEDIATRICS
Plasma somatomedin and growth hormone values in children with protein-calorie malnutrition This study was undertaken to define the interrelationships of somatomedin, growth hormone, and an inhibitor of S M in protein-calorie malnutrition. Twenty-seven patients, eight to 60 months o f age, were studied. Twenty-one well-nourished children acted as controls. S M was significantly depressed at Days 2 and 8 (p < 0.01) but was not different from controls at Day 29 and 50. In ten out o f 27 Day 2 samples and six out of 27 Day 8 samples an inhibitor was identified. When S M values were compared to simultaneous hGH levels, there was an inverse relationship. The low S M levels in the face o f markedly elevated hGH levels suggests a functional block in the synthesis and~or release of SM.
Raymond L. H i n t z , M.D.,* S t a n f o r d , Calif., R o b e r t Suskind, M . D . , C a m b r i d g e , M a s s . , Kosin Amatayakul, M.D., Ousa Thanangkul, M.D., C h i n g - M a i , T h a i l a n d , and Robert Olson, Ph.D., M . D . , St. L o u i s , M o .
SOMATOMEDIN is a peptide that appears to mediate the peripheral action of growth hormone on cartilage and perhaps other tissues?, 2 SM is under control of growth hormone and therefore, usually mimics the level of growth hormone secretion. Grant and associates3 have reported that SM levels are depressed, however, in malnourished children despite elevated plasma levels of hGH. This study was undertaken to better define the interrelationships of SM and hGH in children with both marasmus and kwashiorkor, and to determine whether an inhibitor of SM is present in these children.
PATIENTS AND M E T H O D S Malnourished patients. The patients, eight to 60 months of age, were admitted to the research ward of the Anemia and Malnutrition Research Center in Chiang Mai, Thai-
From the Department of Pediatrics, Stanford University School of Medicine; Department of Nutrition and Food Science, Clinical Research Center, Massachusetts Institute of Technology," Anemia and Malnutrition Research Center, Faculty o f Medicine, Ching-Mai," and Department of Biochemistry & Pediatrics, St. Louis University. Supported in part by AM 1104 and AM 19168from the National Institutes of Health. *Reprint address: Department of Pediatrics, Stanford University School of Medicine, Stanford, CA 94305.
land, where they were hospitalized for at least 50 days. Thirteen children with marasmus, 11 with marasmuskwashiorkor, and three children with kwashiorkor were admitted to the study? The children all had third-degree malnutrition. 5 Virtually all patients on admission had bacterial infections and received antibiotics until the infections cleared. All were treated for fluid and electrolyte imbalance during the first hospital days and received supplemental vitamins and mineralsr Children were fed gradually increasing amounts of a protein milkbase formula diet. A maximum intake of 175 calories and 4 gm protein/kg/day was reached by Hospital Day 16. Virtually all patients had recovered clinically and biochemically by Day 50. Twenty-one well-nourished Thai children of the same age served as controls. Abbreviations used SM: somatomedin hGH: human growth hormone Five milliliters of heparinized blood was drawn from fasting children in the morning, centrifuged, separated, and kept at - 7 0 ~ until shipped by dry ice to the senior author. Somatomedin was assayed by a modification of the method of Van den Brande and Du Caju. 7 Porcine costal cartilage is sliced into 1 mm thick transverse sections. Disks of cartilage 3/32 inch in diameter are punched a constant distance from the perichondrium. The The Journal o f P E D I A T R I C S Vol. 92, No. 1, pp. 153-156
153
154
Hintz et al.
The Journal of Pediatrics January 1978
100
0.7
0.6 150 0.5 E ,.~ E
100 0.4
'I"
(9 .I-
o.3 50
0.2
0
O. 1
0.5
0.8
>1
0.1 1"1
I
I
0 2
8
15
,I
29
50
HOSPITAL D A Y
Fig. 1. SM during treatment of malnutrition. SM (/x/ml) mean + SEM on hospital Days 2, 8, 29, and 50. Plasmas with inhibitor present have been excluded. The shaded area represents mean + SEM of 21 normal Thai controls. I
I
I
40
30
T'--.
E 20
{9
10
2 0 0 2
I 8
I 15
i 29
SM (u/ml)
J
I 50
HOSPITAL DAY
Fig. 2. hGH during treatment of malnutrition, hGH (ng/ml) mean +_ SEM on hospital Days 2, 8, 29, and 50. cartilage disks are preincubated 24 hours at 37~ in Kreb phosphate saline buffer enriched with amino acids. 8 The cartilage is then distributed into tubes with dilutions of standard and unknown plasmas. After a further 24-hour incubation, 0.5 /~Ci 3~SO4 is added to each tube and the
Fig. 3. SM and hGH in each patient sample, hGH (ng/ml) and SM (/Jml) values on each patient sample irrespective of hospital day. incubation continued overnight. The cartilages are immersed in boiling water and washed in running water for six hours. They are then hydrolyzed in 24N formic acid, and :~SO4 determined by liquid scintillation counting. Estimates of relative potency are made using.parallel line bioassay statistics. A unit of SM is defined as the amount of activity present in 1 ml of pooled plasma from 14 normal adults. When the dose response cmwe was significantly nonparallel to the standard plasma, the presence of an inhibitor of the bioassay was confirmed using a modification of the method of SalmonY To a final concentration of 24% normal plasma, 5% and 10% of the test plasma was added. A depression of ~5SO4 uptake greater than 2 SEM below the mean of 24% normal plasma alone was taken as evidence of an inhibitor. The plasmas with inhibitor present were excluded from the statistical analysis of somatomedin values. No attempt was made in this study to further characterize the inhibitor. Growth hormone was determined by a double antibody radioimmunoassay method? ~ RESULTS The SM levels of each patient group taken at Days 2, 8, 29, and 50, respectively, and the hGH values at these days plus Day 15, are presented in Figs. 1 and 2. The S M levels at Days 2 and 8 are significantly depressed (p < 0.01) when compared to either the control Thai children, or the malnourished children on Days 29 or 50. It is important to
Volume 92 Number 1
note that the Day 2 and Day 8 SM values are depressed even with the exclusion of the inhibitory plasmas. When the somatomedin data are compared to simultaneous hGH values, there is a generally inverse relationship between the two hormones in these patients. Irrespective of when in the course of hospitalization the plasma was obtained, the patients with normal SM levels (__ 0.35 /B/ml) had relatively low hGH levels. On the other hand, many of the patients with SM levels < 0.35 /~/ml had elevated hGH levels (Fig. 3). The patients with marasmus-kwashiorkor and kwashiorkor had .slightly higher levels (0.26 _+ 0.10 t~/ml) than those with marasmus (0.18 _+ 0.07 /~/ml). This difference was not statistically significant (p < 0.1-- > 0.05). Ten out of the 27 Day 2 samples and six out of the 27 Day 8 samples showed the presence of an inhibitor of the somatomedin bioassay, and no valid estimate of SM could be determined. The samples with inhibitor had mean hGH levels of 37.'25 _+ 5.30 (mean _+ SD), which was indistinguishable from the group as a whole (34.92 _+ 6.01). The serum albumin levels of those with inhibitor (2.20 _+ 0.76 SD) were not different from those with no inhibitor on Day 2 (2.32 _+ 0.80). Although no quantification of the inhibitor was possible, it was clearly more prevalent in the early samples and disappeared during treatment (Table I). There was no correlation observed between initial SM and severity of illness as judged by either serum albumin or weight percent of expected weight for height. DISCUSSION The above data indicate that serum somatomedin is significantly reduced in children with marasmus as well as those with marasmus-kwashiorkor and kwashiorkor. The presence of an inhibitor in severe malnutrition was first described by Van den Brande and Du Caju. 11 In contrast to the rise soon after admission described by Grant and associates, :~ the children we studied had significantly depressed SM levels on Day 8: The values after 29 days of hospitalization were normal. The low SM levels in the face of markedly elevated hGH levels seen in many of our patients suggest that these patients have a functional block in the synthesis and/or release of SM. There is evidence that at least one site of SM synthesis is the liver?.... The low SM levels could be due to a general impairment of protein synthesis, b y the liver at least, in the case of children with kwashiorkor. The lack of correlation between initial SM and serum albumin levels makes this explanation seem too simplistic. More information will be needed before the mechanism of this block of hGH stimulation of somatomedin in these patients can be defined. The inverse relationship between
S M and hGH in protein-calorie malnutrition
15 5
Table I. Numbers of patients who had inhibitor to SM
bioassay
Hospital day 2 8 29 50
[
Inhibitor 10/27 6/27 0/26 0/25
SM and hGH in these patients fits the hypothesis that SM is involved in the feedback inhibition of bGH release. Those patients with low SM and high hGH levels may have impairment in SM synthesis due to their malnutrition. The low SM levels may actually serve as a stimulus to compensatory hGH release. Some patients had an unexplained combination o f low hGH and SM. In each group (i.e., high and low hGH), feeding produced a rise of SM levels as recovery progressed. This study documents the presence of an inhibitor of somatomedin bioactivity in 10 out of 27 of our initial samples. Grant and associates ~ were unable in their study to demonstrate inhibition of somatomedin. The nature of inhibitors of the SM bioassay is not yet defined? It may be that they play a physiologic role in preventing SM action on tissues. However, the plasma may contain numerous cytotoxic factors. These factors may be having an in vitro effect on the cartilage explants. It remains to be seen whether the action of the inhibitor is specific for SM activity. Until specificity is shown and there is a quantitative way to measure levels of SM bioassay inhibitors, their physiologic role will remain undefined. However, they are an important pitfall in SM bioassay studies. The data presented here further sharpen the paradox of the elevated hGH levels in malnutrition. It is clear that the hGH is functionally inactive in at least one of its major actions, the release of SM. In addition, there may also be an inhibitor of SM end organ action. The future unravelling of the mechanisms of these effects in malnutrition will provide a clearer view or= the action of hGH and SM in the normal child.
REFERENCES
1. Daughaday WH, Hall K, Raben MS, Salmon WD Jr, Van den Brande JL, and Van Wyk JJ: Somatomedin: Proposed designation for sulphation factor, Nature 235:107, 1972. 2. Van Wyk JJ, Underwood LE, Hintz RL, Voina SJ, and Weaver RW: The somatomedins: A family of insulin-like hormones under growth hormone control, Recent Prog Horm Res 30:259, 1974. 3. Grant DB, Hambley J, Becker D, and Pimstone BL: Reduced sulfation factor in undernourished children, Arch Dis Child 48:596, 1973. 4. McLaren DS, Pellet PL, and Read WWC: A simple scoring
15 6
5.
6.
7.
8.
9.
Hintz et al.
system for classifying the severe forms of protein-calorie malnutrition of early childhood, Lancet 1:533, 1967. Gomez F, Galvan RR, Cravioto J, and Frank S: Malnutrition in infancy and childhood, with special reference to kwashiorkor, Adv Pediatr 7:131, 1955. Olson RE: The effect of variations in protein and calorie intake on the rate of recovery and selected physiological responses in Thai children with protein-calorie malnutrition, in Olson RE, editor: Protein-calorie malnutrition, New York, 1975, Academic Press, Inc., pp 275-295. Van den Brande JL, and Du Caju MVL: An ameliorated somatomedin assay using porcine rib cartilage, Excerpta Med Int Cong Ser 256:41, 1972. Daughaday WH, and Reeder CJ: Synchronous activation of DNA synthesis in hypophysectomized rat cartilage by growth hormone, J Lab Clin Med 68:357, 1966. Salmon WD Jr: Effects of somatomedin on cartilage metab-
The Journal of Pediatrics January 1978
10.
11.
12. 13.
olism: Further observations on an inhibitory serum factor, in Raiti S, editor: Recent progress in human growth hormone res, 1974, U.S. Government Printing Office, DHEW (NIH) 74-612, pp 76-97. Schalch DS, and Parker ML: A double-antibody radioimmunoassay for human growth hormone, Nature 203:1141, 1964. Van den Brande JL, and Du Caju MVL: Plasma somatomedin activity in children with growth disturbances, in Raiti S, editor: Advances in human growth hormone research, 1974, DHEW Publ (NIH) 74-612, pp 74-612. McConaghey P, and Sledge CB: Production of sulfation factor by the perfused liver, Nature 225:1249, 1970. Hintz RL, Clemmons DR, and Van Wyk JJ: Growth hormone induced somatomedin-like activity from liver, Pediatr Res 6:353, 1972.
Plasma somatomedin and growth hormone values in children with protein-calorie malnutrition This study was undertaken to define the interrelationships of somatomedin, growth hormone, and an inhibitor of S M in protein-calorie malnutrition. Twenty-seven patients, eight to 60 months o f age, were studied. Twenty-one well-nourished children acted as controls. S M was significantly depressed at Days 2 and 8 (p < 0.01) but was not different from controls at Day 29 and 50. In ten out o f 27 Day 2 samples and six out of 27 Day 8 samples an inhibitor was identified. When S M values were compared to simultaneous hGH levels, there was an inverse relationship. The low S M levels in the face o f markedly elevated hGH levels suggests a functional block in the synthesis and~or release of SM.
Raymond L. H i n t z , M.D.,* S t a n f o r d , Calif., R o b e r t Suskind, M . D . , C a m b r i d g e , M a s s . , Kosin Amatayakul, M.D., Ousa Thanangkul, M.D., C h i n g - M a i , T h a i l a n d , and Robert Olson, Ph.D., M . D . , St. L o u i s , M o .
SOMATOMEDIN is a peptide that appears to mediate the peripheral action of growth hormone on cartilage and perhaps other tissues?, 2 SM is under control of growth hormone and therefore, usually mimics the level of growth hormone secretion. Grant and associates3 have reported that SM levels are depressed, however, in malnourished children despite elevated plasma levels of hGH. This study was undertaken to better define the interrelationships of SM and hGH in children with both marasmus and kwashiorkor, and to determine whether an inhibitor of SM is present in these children.
PATIENTS AND M E T H O D S Malnourished patients. The patients, eight to 60 months of age, were admitted to the research ward of the Anemia and Malnutrition Research Center in Chiang Mai, Thai-
From the Department of Pediatrics, Stanford University School of Medicine; Department of Nutrition and Food Science, Clinical Research Center, Massachusetts Institute of Technology," Anemia and Malnutrition Research Center, Faculty o f Medicine, Ching-Mai," and Department of Biochemistry & Pediatrics, St. Louis University. Supported in part by AM 1104 and AM 19168from the National Institutes of Health. *Reprint address: Department of Pediatrics, Stanford University School of Medicine, Stanford, CA 94305.
land, where they were hospitalized for at least 50 days. Thirteen children with marasmus, 11 with marasmuskwashiorkor, and three children with kwashiorkor were admitted to the study? The children all had third-degree malnutrition. 5 Virtually all patients on admission had bacterial infections and received antibiotics until the infections cleared. All were treated for fluid and electrolyte imbalance during the first hospital days and received supplemental vitamins and mineralsr Children were fed gradually increasing amounts of a protein milkbase formula diet. A maximum intake of 175 calories and 4 gm protein/kg/day was reached by Hospital Day 16. Virtually all patients had recovered clinically and biochemically by Day 50. Twenty-one well-nourished Thai children of the same age served as controls. Abbreviations used SM: somatomedin hGH: human growth hormone Five milliliters of heparinized blood was drawn from fasting children in the morning, centrifuged, separated, and kept at - 7 0 ~ until shipped by dry ice to the senior author. Somatomedin was assayed by a modification of the method of Van den Brande and Du Caju. 7 Porcine costal cartilage is sliced into 1 mm thick transverse sections. Disks of cartilage 3/32 inch in diameter are punched a constant distance from the perichondrium. The The Journal o f P E D I A T R I C S Vol. 92, No. 1, pp. 153-156
153
154
Hintz et al.
The Journal of Pediatrics January 1978
100
0.7
0.6 150 0.5 E ,.~ E
100 0.4
'I"
(9 .I-
o.3 50
0.2
0
O. 1
0.5
0.8
>1
0.1 1"1
I
I
0 2
8
15
,I
29
50
HOSPITAL D A Y
Fig. 1. SM during treatment of malnutrition. SM (/x/ml) mean + SEM on hospital Days 2, 8, 29, and 50. Plasmas with inhibitor present have been excluded. The shaded area represents mean + SEM of 21 normal Thai controls. I
I
I
40
30
T'--.
E 20
{9
10
2 0 0 2
I 8
I 15
i 29
SM (u/ml)
J
I 50
HOSPITAL DAY
Fig. 2. hGH during treatment of malnutrition, hGH (ng/ml) mean +_ SEM on hospital Days 2, 8, 29, and 50. cartilage disks are preincubated 24 hours at 37~ in Kreb phosphate saline buffer enriched with amino acids. 8 The cartilage is then distributed into tubes with dilutions of standard and unknown plasmas. After a further 24-hour incubation, 0.5 /~Ci 3~SO4 is added to each tube and the
Fig. 3. SM and hGH in each patient sample, hGH (ng/ml) and SM (/Jml) values on each patient sample irrespective of hospital day. incubation continued overnight. The cartilages are immersed in boiling water and washed in running water for six hours. They are then hydrolyzed in 24N formic acid, and :~SO4 determined by liquid scintillation counting. Estimates of relative potency are made using.parallel line bioassay statistics. A unit of SM is defined as the amount of activity present in 1 ml of pooled plasma from 14 normal adults. When the dose response cmwe was significantly nonparallel to the standard plasma, the presence of an inhibitor of the bioassay was confirmed using a modification of the method of SalmonY To a final concentration of 24% normal plasma, 5% and 10% of the test plasma was added. A depression of ~5SO4 uptake greater than 2 SEM below the mean of 24% normal plasma alone was taken as evidence of an inhibitor. The plasmas with inhibitor present were excluded from the statistical analysis of somatomedin values. No attempt was made in this study to further characterize the inhibitor. Growth hormone was determined by a double antibody radioimmunoassay method? ~ RESULTS The SM levels of each patient group taken at Days 2, 8, 29, and 50, respectively, and the hGH values at these days plus Day 15, are presented in Figs. 1 and 2. The S M levels at Days 2 and 8 are significantly depressed (p < 0.01) when compared to either the control Thai children, or the malnourished children on Days 29 or 50. It is important to
Volume 92 Number 1
note that the Day 2 and Day 8 SM values are depressed even with the exclusion of the inhibitory plasmas. When the somatomedin data are compared to simultaneous hGH values, there is a generally inverse relationship between the two hormones in these patients. Irrespective of when in the course of hospitalization the plasma was obtained, the patients with normal SM levels (__ 0.35 /B/ml) had relatively low hGH levels. On the other hand, many of the patients with SM levels < 0.35 /~/ml had elevated hGH levels (Fig. 3). The patients with marasmus-kwashiorkor and kwashiorkor had .slightly higher levels (0.26 _+ 0.10 t~/ml) than those with marasmus (0.18 _+ 0.07 /~/ml). This difference was not statistically significant (p < 0.1-- > 0.05). Ten out of the 27 Day 2 samples and six out of the 27 Day 8 samples showed the presence of an inhibitor of the somatomedin bioassay, and no valid estimate of SM could be determined. The samples with inhibitor had mean hGH levels of 37.'25 _+ 5.30 (mean _+ SD), which was indistinguishable from the group as a whole (34.92 _+ 6.01). The serum albumin levels of those with inhibitor (2.20 _+ 0.76 SD) were not different from those with no inhibitor on Day 2 (2.32 _+ 0.80). Although no quantification of the inhibitor was possible, it was clearly more prevalent in the early samples and disappeared during treatment (Table I). There was no correlation observed between initial SM and severity of illness as judged by either serum albumin or weight percent of expected weight for height. DISCUSSION The above data indicate that serum somatomedin is significantly reduced in children with marasmus as well as those with marasmus-kwashiorkor and kwashiorkor. The presence of an inhibitor in severe malnutrition was first described by Van den Brande and Du Caju. 11 In contrast to the rise soon after admission described by Grant and associates, :~ the children we studied had significantly depressed SM levels on Day 8: The values after 29 days of hospitalization were normal. The low SM levels in the face of markedly elevated hGH levels seen in many of our patients suggest that these patients have a functional block in the synthesis and/or release of SM. There is evidence that at least one site of SM synthesis is the liver?.... The low SM levels could be due to a general impairment of protein synthesis, b y the liver at least, in the case of children with kwashiorkor. The lack of correlation between initial SM and serum albumin levels makes this explanation seem too simplistic. More information will be needed before the mechanism of this block of hGH stimulation of somatomedin in these patients can be defined. The inverse relationship between
S M and hGH in protein-calorie malnutrition
15 5
Table I. Numbers of patients who had inhibitor to SM
bioassay
Hospital day 2 8 29 50
[
Inhibitor 10/27 6/27 0/26 0/25
SM and hGH in these patients fits the hypothesis that SM is involved in the feedback inhibition of bGH release. Those patients with low SM and high hGH levels may have impairment in SM synthesis due to their malnutrition. The low SM levels may actually serve as a stimulus to compensatory hGH release. Some patients had an unexplained combination o f low hGH and SM. In each group (i.e., high and low hGH), feeding produced a rise of SM levels as recovery progressed. This study documents the presence of an inhibitor of somatomedin bioactivity in 10 out of 27 of our initial samples. Grant and associates ~ were unable in their study to demonstrate inhibition of somatomedin. The nature of inhibitors of the SM bioassay is not yet defined? It may be that they play a physiologic role in preventing SM action on tissues. However, the plasma may contain numerous cytotoxic factors. These factors may be having an in vitro effect on the cartilage explants. It remains to be seen whether the action of the inhibitor is specific for SM activity. Until specificity is shown and there is a quantitative way to measure levels of SM bioassay inhibitors, their physiologic role will remain undefined. However, they are an important pitfall in SM bioassay studies. The data presented here further sharpen the paradox of the elevated hGH levels in malnutrition. It is clear that the hGH is functionally inactive in at least one of its major actions, the release of SM. In addition, there may also be an inhibitor of SM end organ action. The future unravelling of the mechanisms of these effects in malnutrition will provide a clearer view or= the action of hGH and SM in the normal child.
REFERENCES
1. Daughaday WH, Hall K, Raben MS, Salmon WD Jr, Van den Brande JL, and Van Wyk JJ: Somatomedin: Proposed designation for sulphation factor, Nature 235:107, 1972. 2. Van Wyk JJ, Underwood LE, Hintz RL, Voina SJ, and Weaver RW: The somatomedins: A family of insulin-like hormones under growth hormone control, Recent Prog Horm Res 30:259, 1974. 3. Grant DB, Hambley J, Becker D, and Pimstone BL: Reduced sulfation factor in undernourished children, Arch Dis Child 48:596, 1973. 4. McLaren DS, Pellet PL, and Read WWC: A simple scoring
15 6
5.
6.
7.
8.
9.
Hintz et al.
system for classifying the severe forms of protein-calorie malnutrition of early childhood, Lancet 1:533, 1967. Gomez F, Galvan RR, Cravioto J, and Frank S: Malnutrition in infancy and childhood, with special reference to kwashiorkor, Adv Pediatr 7:131, 1955. Olson RE: The effect of variations in protein and calorie intake on the rate of recovery and selected physiological responses in Thai children with protein-calorie malnutrition, in Olson RE, editor: Protein-calorie malnutrition, New York, 1975, Academic Press, Inc., pp 275-295. Van den Brande JL, and Du Caju MVL: An ameliorated somatomedin assay using porcine rib cartilage, Excerpta Med Int Cong Ser 256:41, 1972. Daughaday WH, and Reeder CJ: Synchronous activation of DNA synthesis in hypophysectomized rat cartilage by growth hormone, J Lab Clin Med 68:357, 1966. Salmon WD Jr: Effects of somatomedin on cartilage metab-
The Journal of Pediatrics January 1978
10.
11.
12. 13.
olism: Further observations on an inhibitory serum factor, in Raiti S, editor: Recent progress in human growth hormone res, 1974, U.S. Government Printing Office, DHEW (NIH) 74-612, pp 76-97. Schalch DS, and Parker ML: A double-antibody radioimmunoassay for human growth hormone, Nature 203:1141, 1964. Van den Brande JL, and Du Caju MVL: Plasma somatomedin activity in children with growth disturbances, in Raiti S, editor: Advances in human growth hormone research, 1974, DHEW Publ (NIH) 74-612, pp 74-612. McConaghey P, and Sledge CB: Production of sulfation factor by the perfused liver, Nature 225:1249, 1970. Hintz RL, Clemmons DR, and Van Wyk JJ: Growth hormone induced somatomedin-like activity from liver, Pediatr Res 6:353, 1972.
