| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
University of Texas Southwestern Medical School (D.S.H., J.R.), Dallas, Texas 75390; Texas Childrens Hospital/Baylor College of Medicine (D.S.H., K.J.E., R.M., D.K.S.), Houston, Texas 77030; and Cook Childrens Hospital (M.D.), Fort Worth, Texas 76104
Address all correspondence and requests for reprints to: Dana S. Hardin, M.D., Associate Professor of Pediatrics, Pediatric Endocrinology, University of Texas Southwestern Medical School, 5323 Harry Hines Boulevard, G2.234, Dallas, Texas 75390-9063. E-mail: dana.hardin{at}utsouthwestern.edu
Abstract
Despite aggressive nutritional therapy, low body weight and protein
catabolism are common problems in children with cystic fibrosis.
Previous studies by our group and others have demonstrated improvement
in both height and weight in children with cystic fibrosis who were
treated with human recombinant GH, and our group has recently
documented improved clinical status and lean tissue mass as well. The
purpose of this report is to summarize our findings of the effect of GH
on whole body protein kinetics in cystic fibrosis and to relate these
findings to changes in TNF-
levels.
We conducted a 1-yr study of 19 prepubertal children with cystic fibrosis (age 712 yr, all <94% of ideal body weight). Ten children were randomly assigned to take daily injections of GH (0.3 mg/kg·wk), and nine were randomly assigned to be controls. Baseline results from the subjects with cystic fibrosis were compared with results obtained from nine age- and gender-matched healthy children. Whole body protein turnover was measured at baseline and every 6 months using the stable isotope [1-13C]leucine and mass spectrometric analysis.
Leucine rate of appearance, a measure of protein catabolism, was
similar in both cystic fibrosis subgroups at baseline and was
significantly higher than in the control children without cystic
fibrosis. Treatment with GH resulted in a significantly lower leucine
rate of appearance, as well as significantly lower leucine oxidation.
The rate of protein synthesis, as calculated from these numbers,
actually decreased in the cystic fibrosis subgroup. TNF-
levels were
higher in both cystic fibrosis subgroups than in controls and
correlated with leucine rate of appearance.
The results of this study suggest that one reason GH improves body weight and lean tissue mass is due to improved whole body protein catabolism and improved efficiency of whole body protein kinetics.
DESPITE RECENT EMPHASIS on adequate nutritional intake, many patients with cystic fibrosis (CF) have difficulty gaining weight and are less than ideal body weight for age (1, 2). Previous studies of CF patients have demonstrated (3, 4, 5, 6) overall protein catabolism, even in nonacutely ill subjects. Negative protein balance may contribute to morbidity and mortality in CF by decreasing body muscle mass, and possibly by contributing to worsened immune function (7). One study (8) has demonstrated that reversal of protein catabolism resulted in stabilization of pulmonary function and decreased number of hospitalizations.
Human recombinant GH is well known for its anabolic effects on whole body protein turnover (WBPT) and cell growth (9). Investigations of GH treatment in catabolic conditions such as burns (9, 10), sepsis (11), and acquired immunodeficiency syndrome (AIDS) (12) have demonstrated improved protein synthesis and subsequent increase in lean tissue mass (LTM). Use of GH in children with documented GH deficiency results in marked nitrogen retention (13), accelerated linear growth, and weight gain (14). Previous studies by our group (15, 16) and others (17, 18, 19) have documented improvement in weight and height in CF subjects treated with GH. Our group has most recently reported (20) improved clinical status and LTM in CF children randomly assigned to GH treatment. No study, to date, has evaluated GH-mediated changes in WBPT in CF subjects. Thus, the first purpose of our study was to determine the effect of long-term GH therapy on WBPT.
Recent evaluations in non-CF subjects have linked (21, 22)
protein catabolism to high levels of cytokines, particularly TNF-
.
Chronic inflammation is believed to contribute to increased levels of
circulating cytokines, and several groups (23, 24) have
reported high TNF-
levels in both bronchoalveolar lavage fluid and
serum from CF subjects. The second purpose of this study was to
evaluate the effect, if any, of GH on TNF-
levels in CF.
To confirm previous reports (3, 5) of protein catabolism in CF, and to validate our measures of WBPT, we compared our findings at baseline (BL) with those of age- and gender-matched healthy control children.
Experimental Design and Methods
Subjects
We recruited 19 prepubertal CF patients (age, 712 yr) from the Cystic Fibrosis Centers at Texas Childrens Hospital/Baylor College of Medicine in Houston and Cook Childrens Hospital in Fort Worth. Inclusion criteria included any child who was at or below the 10th percentile for both height and weight, despite adequate caloric intake on at least two evaluations. All children were considered pancreatic insufficient. Exclusion criterion included: current or previous glucose intolerance, infection with Burkholderia cepacia, pubertal status greater than Tanner stage II, weight loss greater than 3% in the 3 months before study, and either hospitalization or use of systemic steroids within 6 wk before the first study visit. Parents signed a written informed consent, and children signed a written assent (forms approved by the Institutional Review Boards at University of Texas Health Science CenterHouston and Baylor College of Medicine). Each child was randomly assigned to either the treatment group (GHTX) or to the control group (non-TX). Throughout the study, each child continued to receive standard CF care, including pancreatic enzyme replacement, vitamin supplementation, bronchodialators, and mucolytics. Antibiotic therapy was prescribed as indicated, and patients were hospitalized as needed. GH was continued during all hospitalizations, and none of the subjects received iv or oral corticosteroids during the study year.
We recruited nine healthy control children by advertisement within the
Texas Medical Center (Houston, TX). All were deemed to be healthy by
parental report and by physical examination. These children underwent
only BL measures of WBPT, LTM, and TNF-
, and none were treated with
GH. The parents of these children gave written informed consent, and
the children, written assent.
In vivo methods
Measurement of WBPT. At BL and every 6 months, WBPT was measured in all subjects. Each was admitted to the General Clinical Research Center at the University of TexasHouston on the morning of study, following a 12-h overnight fast. After height and weight were measured, the rate of appearance of carbon dioxide (RaCO2) was measured for 30 min using an indirect calorimeter (25). Afterward, a BL breath sample was collected (Meretek bags; Meretek Diagnostics, Houston, TX) in triplicate and placed into 20-ml evacuated tubes (Metabolic Solutions, Boston, MA). Next, the patients indwelling central venous catheter was accessed, or, in those subjects without an indwelling catheter, a peripheral iv catheter was inserted into an antecubital vein. BL blood samples were withdrawn in triplicate through the infusion line. Next, an iv bolus of NaH13CO3 (0.2 mg/kg) was given to saturate the bicarbonate pool and shorten the time required for achievement of isotopic equilibration (26). After the NaH13CO3, each subject received a 0.35 mg/kg bolus of the stable isotope [1-13C]leucine, followed by an infusion of the isotope at 0.7 mg/kg·h. After a 2.5-h equilibration period, breath samples were collected in triplicate, as described above, These were subsequently analyzed for the 13C enrichment of CO2 (Metabolic Solutions) by isotope ratio mass spectrometry. After the steady state breath samples were collected, blood was drawn from a peripheral vein (not the infusion line), in triplicate. Like the BL blood, the steady state blood was separated by centrifugation and the plasma was stored for future isotope analysis. The appropriate times for steady state measurement were determined by results from previous studies by our group and others.
GH administration. GH (Nutropin AQ; Genentech, Inc., South San Francisco, CA) was administered to the GHTX group by daily sc injection for 1 yr. The dose (0.3 mg/kg·wk) was adjusted for body weight every 3 months. Compliance was assessed by history, and by having subjects return empty vials to the study site. On the day of metabolic studies, GH was not given until after the study was completed.
In vitro methods
WBPT. Plasma was derivatized (27), and stable isotopic enrichment of [1-13C]leucine in serum was measured by gas chromatography/mass spectrometry (model 5970; Hewlett-Packard Co., Palo Alto, CA). Breath enrichment of 13CO2 was measured in the expired air samples using gas isotope-ratio mass spectrometry (Metabolic Solutions). Protein breakdown and synthesis were calculated from measurements of 13CO2 and 13C-leucine using the primary pool model (28) according to the following methods:
In the fasted state, the rate of appearance of leucine
(LeuRa) reflects the absolute rate of protein breakdown
(29, 30). The calculation for LeuRa was:
![]() |
The rate of total leucine oxidation (LeuOx) was assessed
using the equation:
![]() |
The rate of whole body protein synthesis (S) is equivalent
to the rate of nonoxidative leucine disappearance (NOLD)
(30) and was calculated:
![]() |
Measurement of TNF-
levels. TNF-
was measured by ELISA
(Mellennia; Diagnostic Products, Los Angeles, CA). Samples
were run in duplicate, and eight quality control samples were run with
each assay, as were standard curves.
Statistics
Subjects were assigned to the GHTX or non-TX groups by randomization, after agreeing to participate regardless of treatment group. Data from each group is reported as the mean ± SD. Statistical comparison of the CF groups was determined by the nonpaired t test. Comparison of the controls and the CF groups was done by ANOVA.
Results
Subjects
Children in both CF subgroups were similar at BL for age, gender,
height, weight, and percent ideal body weight. The control subjects
were matched to the CF subjects for age, gender, height, and for
pubertal status (all Tanner I). They were not matched for weight or
LTM. Subject characteristics are presented in Table 1
.
|
There was no difference between the CF children in either
subgroups at BL for LeuRa, LeuOX, or
NOLD (Table 2
). There was no
difference between RaCO2 at BL in any group (GHTX, 4.9 ± 0.9;
non-TX, 5.1 ± 1.0; Control, 4.7 ± 0.7). At BL there was
also no difference between the CF subjects and healthy controls for
LeuOx or NOLD. However, LeuRa was
significantly higher in the CF children than in controls (Fig. 1
).
|
|
) in LeuRa and
LeuOX in the GHTX group was significant at both 6 and 12
months (GHTX
LeuRa: 6 months = 38 ± 14, 12
months = 44 ± 15; GHTX
LeuOX: 6 months =
18 ± 8, 12 months = 19 ± 7). There was not a
significant difference between the 6- and 12-month
values in the
GHTX group. There was no significant change (
) from BL at either 6
or 12 months in the non-TX group.
Calculation of the ratio of leucine oxidation to nonoxidative leucine
disposal provides an estimate of whole body protein catabolism
(13, 31). Our results indicate that GH treatment resulted
in a significant decrease in catabolism at both 6 and 12 months (Table 2
).
Anthropometrics
As reported previously (20), the GHTX group demonstrated significant gain in body weight (change in weight, kilograms: 6 months, GHTX = 4.9 ± 1.8, non-TX = 2.2 ± 1.4, P < 0.03; 12 months, GHTX = 5.2 ± 1.7, non-TX = 2.4 ± 1.7, P < 0.03). The weight change was mostly from improved LTM (change in LTM, kilogram: 6 months, GHTX = 3.8 ± 1.6, non-TX = 1.2 ± 1.1; 12 months, GHTX = 4.9 ± 1.2, non-TX = 2.2 ± 1.4, P < 0.05).
TNF-
TNF-
did not differ between the CF groups at BL, but was
significantly lower in the GHTX group at 6 and 12 months. At BL both CF
subgroups were higher than controls (CF, 78 ± 10; control,
<5 ± 3, P = 0.001). TNF-
levels correlated
with leucine Ra (r = 0.72, P = 0.04). As reported
previously (20), IGF-I and insulin levels were
significantly higher at 6 and 12 months in the GHTX group. CF data are
shown in Table 3
.
|
Despite careful attention to nutritional management, poor weight gain remains a common problem in patients with CF (32, 33, 34). A likely contributor to this problem is protein catabolism, which has been previously described (8, 35, 36, 37) by multiple investigators. Human recombinant GH stimulates accrual of linear height (38) and is a potent anabolic agent used to improve nutrition and protein catabolism in many chronic illnesses (9, 39, 40). This study reports GH-mediated changes in protein kinetics in CF children. Using the stable isotope of leucine to obtain an estimate of protein breakdown and synthesis, we have demonstrated that GH treatment results in decreased protein catabolism (expressed as leucine Ra and as oxidation/NOLD). These results suggest that improvement in weight and LTM resulting from GH treatment of CF children is, in part, due to improved efficiency of whole body protein kinetics.
Our findings of decreased leucine oxidation with GH treatment agree with reports of GH effect by many others (12, 40, 41, 42). However, interestingly, we have also found that sustained GH treatment resulted in an overall decrease in calculated protein synthesis. Previous short-term studies in non-CF subjects have described (13, 40, 42) increased protein synthesis resulting from GH treatment. For example, Garibotto et al. (40) described a 25% increase in protein synthesis, yet no change in protein degradation, in malnourished hemodialysis patients treated with GH for 3 months. Horber and Haymond (13) described both decreased catabolism and increased synthesis in subjects given short-term glucocorticoids and GH. In agreement with our findings, Copeland and Nair (42) reported a 25% decrease in oxidation, and no increase in synthesis, in 15 young healthy men.
The timing of our evaluation of protein kinetics could be one explanation for the lack of increase in synthesis. We examined protein turnover at BL, 6 months, and 12 months. Perhaps we would have seen an increase in synthesis in the GHTX subjects if we had performed the measurements before 6 months. Another explanation may simply be due to the calculation. The LeuOx must be subtracted from the LeuRa to calculate S. Thus, a large decrease in leucine Ra results in decreased calculated S, unless leucine oxidation decreases to a greater extent then LeuRa. We may have seen an accentuated decrease in LeuRa due to the catabolic nature of CF, or due to enhanced insulin and IGF-I levels.
We have demonstrated a decrease in LeuRa with GH treatment, which has not been found in other studies (42, 43). However, LeuRa at BL was greater in the CF subjects compared with the healthy control children in our current study, and in studies of normal children described by other groups (44, 45). These findings suggest that CF is a catabolic condition (8, 26, 46). Similarly, Suchner et al.(46A ) studied malnourished adults with chronic obstructive pulmonary disease treated with both GH and parenteral nutrition and found significant decrease in protein catabolism. Several studies in non-CF subjects suggest that catabolism itself may be both a cause, and a result, of alterations in protein metabolism. Clemmons (41) found that administration of GH to calorically restricted patients resulted in decreased protein oxidation and hypothesized that amino acids were more efficiently reused for protein synthesis during short-term caloric restriction. Yarasheski et al. (47) studied protein metabolism in wasted AIDS men and reported that protein synthesis was inappropriately low and did not increase with feeding. They hypothesized that amino acid availability for protein synthesis is limited during long-term wasting. Thus, our findings may be associated with chronic illness and/or catabolism.
Another explanation for a decrease in LeuRa may have been the increase in insulin levels resulting from GH treatment. Previous studies (6, 48, 49) have documented the reduction of protein catabolism by insulin. However, these studies have used higher insulin levels, and a greater magnitude of change, than the insulin levels resulting from GH treatment of our subjects. The reduction in LeuRa may be also be secondary to increased IGF-I levels resulting from GH. In a study of 21 adults, Mauras and Beaufrere (50) noted a reduction in LeuRa and protein catabolism when recombinant human IGF-I was administered. We cannot rule out the effects of either elevated insulin or IGF-I levels on the reduction of LeuRa.
It is possible that previous studies evaluating short-term administration of GH may have detected an initial physiologic response for more efficient, synthesis-directed, reutilization of amino acids derived from breakdown. However, long-term, this level of "efficiency" may not be maintained due to depletion of amino acids. Another possibility in CF may be that the amino acids derived from breakdown are directed to other metabolic pathways. Our group has previously reported (51) increased hepatic glucose production and hepatic insulin resistance in adults with CF. If high hepatic glucose production is principally from gluconeogenesis, then it is possible that amino acids are channeled to this pathway instead of to protein synthesis.
Finally, several studies (52, 53, 54) have linked catabolism
to high TNF-
levels. Holecek et al. (21)
found an increase in WBPT when rats were given TNF-
. Studies in
humans with cancer (22), and with AIDS (52),
have linked TNF-
to protein wasting. GH treatment resulted in a
marked decrease in the TNF-
levels in our subjects. This may have
been due to improved clinical status of the GHTX subjects. In a study
of 15 malnourished children with CF, Holt et al.
(8) demonstrated that S decreased approximately
42% during pulmonary exacerbation, and that protein catabolism
decreased when the patients were clinically stable. We have previously
documented (20) improved clinical status in our CF
children treated with GH, and we have now demonstrated that TNF-
levels also decrease with GH. There is likely a relationship between
WBPT and clinical status in CF. This relationship underscores the
importance of GH effects on whole body protein catabolism in these
patients. Future studies evaluating the role of TNF-
and other
cytokines on protein metabolism in these patients will be
interesting.
Our study summarizes the protein kinetic findings of a 1-yr randomized control trial of GH use in children with CF. These results suggest that GH increases the efficiency of WBPT and decreases protein catabolism.
Acknowledgments
We gratefully acknowledge Drs. Dennis Bier and Ross Shepherd for review of the data.
Footnotes
These studies were supported by NIH Grants K08-DK02365-01 and M01-RR-02558 (to University of Texas Clinical Research Center) and by a grant from Genentech Foundation.
Abbreviations: AIDS, Acquired immunodeficiency syndrome; BL, baseline; CF, cystic fibrosis; GHTX, GH treatment group; LeuOx, rate of total leucine oxidation; LeuRa, rate of appearance of leucine; LTM, lean tissue mass; NOLD, rate of nonoxidative leucine disappearance; non-TX, control group; S, rate of whole body protein synthesis; WBPT, whole body protein turnover.
Received January 31, 2001.
Accepted May 7, 2001.
References
and endotoxin-treated rats:
contribution to hepatic tissue. Am J Physiol 273:E1052E1058
and interleukin 1 for experimental
tumor growth and the development of cancer cachexia. Cancer Res 51:880885
, resting energy expenditure and
cachexia in cystic fibrosis. Clin Sci 85:563568[Medline]
This article has been cited by other articles:
![]() |
D. Schnabel, C. Grasemann, D. Staab, H. Wollmann, F. Ratjen, and for the German Cystic Fibrosis Growth Hormone Stud A Multicenter, Randomized, Double-Blind, Placebo-Controlled Trial to Evaluate the Metabolic and Respiratory Effects of Growth Hormone in Children With Cystic Fibrosis Pediatrics, June 1, 2007; 119(6): e1230 - e1238. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. M. Aris, P. A. Merkel, L. K. Bachrach, D. S. Borowitz, M. P. Boyle, S. L. Elkin, T. A. Guise, D. S. Hardin, C. S. Haworth, M. F. Holick, et al. Guide to Bone Health and Disease in Cystic Fibrosis J. Clin. Endocrinol. Metab., March 1, 2005; 90(3): 1888 - 1896. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. B. Soeters, M. C. G. van de Poll, W. G. van Gemert, and C. H. C. Dejong Amino Acid Adequacy in Pathophysiological States J. Nutr., June 1, 2004; 134(6): 1575S - 1582S. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Darmaun, V. Hayes, D. Schaeffer, S. Welch, and N. Mauras Effects of Glutamine and Recombinant Human Growth Hormone on Protein Metabolism in Prepubertal Children with Cystic Fibrosis J. Clin. Endocrinol. Metab., March 1, 2004; 89(3): 1146 - 1152. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |