Bradley S. Miller, MD, PhD, Division of Pediatric Endocrinology, University of Minnesota, MMC 404, 420 Delaware St., SE, Minneapolis, MN 55455. mille685@umn.edu Article appeared in CDG Newsletter Summer 2004.
Growth in CDG
Background:
Children with CDG exhibit growth failure that may be a result of abnormal hormone signaling. Growth in children is regulated by growth hormone (GH) that is made in the pituitary and secreted into the blood. GH exerts its effects on growth of the whole body by activation of the Insulin-like Growth Factor (IGF) cascade. IGFs are small proteins produced primarily in the liver in response to GH. The levels of free IGF in tissues and serum are regulated by the formation of complexes with IGF binding proteins (IGFBPs). In the blood, more than 90% of IGF is found in a complex comprised of IGF, IGFBP-3 and an acid-labile subunit (ALS). The formation of this ternary complex is important in determining the stability of IGFs in blood and their delivery to the target tissues. In the target tissues such as the growth plate, binding proteins deliver free IGF to cell-surface receptors to stimulate growth.
The degree of ALS glycosylation has been shown to affect its ability to form a complex with IGFBP-3 when mixed together in a test-tube (in vitro). Glycosylation of IGFBP-3 does not affect its ability to bind IGF in vitro, but may increase its susceptibility to proteolysis (degradation). Studies in rats have shown that non-glycosylated IGFBP-3 is able to form ternary complexes but has a shorter half-life.
Previous studies of the GH/IGF cascade in children with CDG have shown that random levels of GH are low in males and high to very high in females. Serum IGF-1 concentrations were low in infants and children and low to normal in adolescents independent of gender. There has been one report of a young lady with normal levels of IGF-1& IGFBP-3. However, except for patients with CDG 1b and CDG 1f, this young woman is the only reported patient with CDG 1 to have experienced normal puberty. One patient with CDG 1f was investigated because of growth retardation at 2 years and found to have growth hormone deficiency and dwarfism as an adult. IGFBP-3 and ALS values have not been reported in children with both CDG and failure to thrive. To date, there have been no human or animal studies of the impact of incomplete ALS glycosylation on ternary complex formation.
Methods:
Stored sera were analyzed from children participating in a study of mannose therapy in CDG. ALS levels were determined by ELISA and immunoblot. ALS was also analyzed by 2-D gel electrophoresis. IGF-1 and -2 were measured by IRMA. IGFBP-3 levels were determined using immunoblot, Western ligand blot and IRMA. Proteolysis of IGFBP-3 was measured by ELISA. Ternary complex formation was assayed using Superose 200 size-exclusion chromatography. An additional patient with CDG 1b was recruited separately due to her recent diagnosis. All patients or their parents consented to take part in the study. The studies were approved by the Mayo Clinic Institutional Review Board.
Results:
The components of the IGF cascade were analyzed in serum from children with CDG 1a (n = 12) and compared to a group of age-matched controls (n = 11). Analysis of ALS by ELISA yielded reduced levels of total ALS in patients with CDG 1A (CDG 1A 9.74 +/- 1.09, Control 19.63 +/- 1.65 g/mL p<0.05). Immunoblot analysis of ALS showed multiple bands suggestive of underglycosylation in children with CDG 1A. This was supported by immunoblot of ALS following 2-D gel electrophoresis. IGFBP-3 levels measured by immunoradiometric assay were significantly lower in CDG 1a (CDG 1a 2069 +/- 192, Control 3563 +/- 306 ng/mL; p<0.05). Immunoblot analysis of IGFBP-3 showed a reduction in the ~50 kDa form with an increase in a ~30 kDa form. Western ligand blot analysis of IGFBPs from serum of children with CDG 1a also indicated a reduction of bound ligand at ~50 kDa with an increase in ligand binding at 30 kDa. Unglycosylated IGFBP-3 and the major proteolytic fragment of glycosylated IGFBP-3 both migrate at ~30 kDa. However, the IGFBP-3 proteolytic fragment does not bind IGFs on Western ligand blot. These data suggest that IGFBP-3 is underglycosylated in children with CDG 1A. Further analysis of IGFBP-3 and its proteolytic fragments, using a set of two-site ELISAs to decipher the mixture, showed that proteolysis is not the major cause of low IGFBP-3 levels in children with CDG 1A. Serum levels of IGF-1 (CDG 1A 102 28, Control 220 +/- 47 ng/mL; p<0.05) and IGF-2 (CDG 1A 404 +/- 35, Control 677 +/- 63 ng/mL; p<0.05) were lower in CDG 1A. However, free IGF-1 was not significantly different (CDG 1A 1.40 +/- 0.32, Control 2.69 +/- 0.59 ng/mL; p=0.09). Size exclusion chromatography showed a reduction in the amount of ternary complex formation in CDG 1a compared to age-matched controls.
In two patients with CDG 1b, ALS and IGFBP-3 were analyzed before and during mannose therapy. A rapid improvement in the degree of ALS and IGFBP-3 glycosylation was seen following the initiation of mannose therapy. The improvement in ALS glycosylation was accompanied by an increase in ternary complex formation. One child with CDG 1b had a remarkable growth response to mannose therapy. At 30 months of age, she was <3 percentile for height and >97 percentile for weight. At 36 months, following 6 months of mannose therapy, her height was at the 43rd percentile and weight was unchanged but now at the 89th percentile. This represents an extraordinary annualized height velocity of 12.6 cm/yr with normal for her age being 8 cm/yr. Subsequently, she continued to grow along the 40th percentile.
Conclusions:
These studies represent the first analyses of the effects of N-glycosylation on the components of the IGF ternary complex in humans. Multiple components of the IGF ternary complex are altered in the serum of children with CDG 1a with a resulting decrease in ternary complex formation. Improvement in ALS/IGFBP-3 glycosylation in response to mannose in CDG 1b coincided with an extraordinary period of catch-up growth. This extraordinary growth response to mannose in a child who was either normal or overweight suggests that glycosylation of growth factors, carrier proteins and receptors played an important role in the growth. In addition, this finding downplays the role of poor nutrition as a factor in the short stature of children with CDG. It is likely that the growth of children with CDG is affected both by nutritional factors and by the effects of hypoglycosylation on the function and stability of the components of the GH/IGF cascade. Further studies are necessary to determine the extent to which protein hypoglycosylation negatively impacts growth in children with CDG.
Although children with CDG do not appear to be growth hormone deficient, growth hormone therapy may be a reasonable option to improve growth in children with CDG. GH treatment increases all three components of the ternary complex and may overcome their rapid degradation due to incomplete glycosylation. GH therapy has also been seen to improve muscle tone in children with short stature and hypotonia as a result of other conditions such as Prader-Willi Syndrome. However, it is difficult to predict if GH treatment would improve muscle tone in children with CDG.
Puberty in females with CDG
Background:
Puberty and sexual function in men and women is regulated by pituitary hormones called gonadotropins. Follicle stimulating hormone (FSH) and leutinizing hormone (LH) regulate the production of sex steroids (testosterone and estrogen) in the gonads (ovaries and testes). The gonadotropins also regulate egg and sperm production. The original twin girls with CDG 1a described by Dr. Jaeken had fluctuating levels FSH and failed to have pubertal development. Most other females with CDG of pubertal age have been described as having an absence of secondary sexual characteristics. Studies of these children have demonstrated hypergonadotropic hypogonadism with elevated FSH and LH and low estradiol (one form of estrogen) which suggests failure of the gonad to make the appropriate level of sex hormone. Normal pubertal development has been described in four women, one with CDG 1b, one with CDG 1f and two with untyped CDG. The young woman with CDG1b delivered three healthy children after uncomplicated pregnancies. There are no prospective studies of puberty or menstrual function in woman CDG 1b. Males with CDG have been described to have normal or delayed puberty with normal virilization, small or normal testes, and testosterone values in the low to normal range.
It is unclear whether incomplete glycosylation of pituitary gonadotropins or alterations in ovarian function are the cause of pubertal failure in females with CDG. Studies demonstrating normal estrogen production after short term therapy with human menopausal gonadotropin (mostly FSH) have suggested intact ovarian function. However, the biological activity of FSH in CDG women was in the low normal range using a rat granulosa aromatase bioassay. In addition, the half-life of FSH was low to normal in CDG consistent with previous data showing increased clearance of desialylated FSH. The electrophoretic mobility of FSH in females with CDG was more consistent with that obtained from menopausal women and did not resemble the FSH pattern characteristic of the follicular phase of the menstrual cycle. These findings suggest that reduced FSH bioactivity and bioavailability, due to inappropriate or insufficient glycosylation, is responsible for delayed or absent sexual maturation in girls with CDG.
Methods:
One young lady with CDG 1c has been found to have normal pubertal development and menstrual function. I am proposing to study circulating and urinary gonadotropins during the menstrual cycle in pubertal females with CDG. Evaluation of these hormones in pubertal aged girls with CDG who are not experiencing puberty would also help us understand key differences in the glycosylation that may be necessary for normal sexual development. This study is currently being prepared for review by the Institutional Review Board of the University of Minnesota.
Conclusions:
The proposed future studies will help determine the importance of gonadotropin glycosylation on female pubertal development and menstrual function.
Estrogen Therapy in Young Women with CDG
Background:
Estrogen replacement therapy in women after menopause has been the topic of much recent interest. Multiple studies suggest that women who receive estrogen therapy after menopause have an increase in the risk of breast cancer. The benefits of estrogen therapy in these women include improvement in post-menopausal symptoms and an increase in bone strength. There was no evidence of improvement in heart health or cognition in post-menopausal women taking estrogen supplements.
These studies are important in the care of women with CDG when they are older, but not when they are young. As mentioned earlier, girls with CDG tend not to have much pubertal development because of low estrogen production. Treatment of these young ladies should be considered as a replacement of the estrogen that their bodies would normally have made. The estrogen produced during puberty is important in the development of secondary sexual characteristics including breasts and widening of the hips. Estrogen is also very important in the skeletal growth including height and bone mineral density. Our skeleton increases in density until about 25 years of age. After that time, our body leaches the mineral out of the bones over time. In both men and women, the sex steroids reduce this loss. After menopause in women, the rate of mineral loss increases. This mineral loss leads to osteoporosis, which is more common in women. Men and women with osteoporosis are at an increased risk of fractures, particularly hip and vertebral fractures.
Low or absent sex steroids in childhood can lead to low peak bone density and the early development of osteoporosis. If the osteoporosis is severe, fractures can develop during childhood. Other factors can lead to low bone density in children. Osteopenia, a milder form of thinning of the bones, has been described in children with CDG at an early age. This suggests that bone density formation is impaired in CDG even before the sex steroids would be expected to contribute.
Estrogen replacement therapy is not without side effects and risks. Estrogen can cause an increase in the formation of blood clots (hypercoaguability). Children with CDG can have problems with hypercoagulation or inability to form blood clots. Because of this potentially serious side effect the use of estrogen in young women with CDG needs to be carefully considered. If there is no previous history of coagulation problems or strokes, I would consider beginning low dose estrogen in a young woman with CDG between the ages of 13 and 15. Such women should be monitored closely for side effects including deep vein thrombosis, headaches, increased seizure frequency and stroke-like symptoms. I am aware of one young lady from New Zealand who is receiving estrogen therapy and has tolerated it well.
I would be interested to hear from women with CDG, their families, and their physicians about their experiences with estrogen treatment. I would also be happy to answer any questions that arise about this treatment.
Conclusions:
Estrogen replacement therapy should be considered carefully in young women with CDG and pubertal failure for development of secondary sexual characteristics and improvement of bone mineral density. There is little data regarding this treatment, so sharing our experiences will help improve our knowledge in this area. Estrogen treatment has serious potential side effects that need to be considered before initiating treatment and monitored while on therapy.
Dr. Bradley Miller, M.D., PhD. is an Assistant Professor at the University of Minnesota with an interest in the insulin-like growth factor (IGF) system. He is also interested in the endocrine aspects of abnormal glycosylation in children with CDG. Dr. Miller grew up in Illinois and the Upper Peninsula of Michigan. He attended Nothern Michigan University where he received a Bachelor of Science in Biochemistry. While at NMU, he became interested in basic science research. He pursued a combined MD/PhD as part of the Medical Scientist Training Program at the Medical University of South Carolina in Charleston, South Carolina. His research at MUSC involved the cellular signaling of the insulin and IGF-1 receptors. He returned to the Midwest for training in Pediatrics and Pediatric Endocrinology at the Mayo Clinic in Rochester, MN. While at the Mayo Clinic, he first became involved in caring for children with CDG. Dr. Miller was involved in studies at Mayo Clinic to help understand how protein glycosylation impacts the growth cascade.
