| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Endocrinology (R.M.C.P., M.H.A.-O., C.R.P.O., F.T.O., V.C.C., C.T.F., T.A.R.V., M.B.G., J.L.M.O., C.M.-S., I.E.S.R, J.A.S.B.-F.), Federal University of Sergipe, Aracaju, SE Brazil 49060-100; and Division of Endocrinology (A.S., R.S.), The Johns Hopkins University School of Medicine, Baltimore, Maryland 21287
Address all correspondence and requests for reprints to: Roberto Salvatori, M.D., Division of Endocrinology, Johns Hopkins University, 1830 East Monument Street, Suite 333, Baltimore, Maryland 21287. E-mail: salvator{at}jhmi.edu.
| Abstract |
|---|
|
|
|---|
Methods: We studied members of a large Brazilian kindred with IGHD (Itabaianinha cohort) caused by a homozygous null GHRHR mutation. We compared 76 adult subjects (age, 2575 yr) heterozygous for the mutation (WT/MT) with 77 sex-matched controls from the same population who are homozygous for the wild-type GHRHR allele (WT/WT).
Results: We found no difference in adult height and SD score for serum IGF-I between the two groups. Body weight, body mass index, skin folds, waist and hip circumferences, and lean mass were all reduced in WT/MT subjects. Percentage fat mass and waist/hip ratio were similar in the two groups. Fasting insulin and homeostasis model assessment of insulin resistance were lower in WT/MT. The other biochemical parameters [total and fractionated cholesterol, triglycerides, lipoprotein (a), and C-reactive protein] were not different between the two groups.
Conclusions: Heterozygosity for a null GHRHR mutation is not associated with reduction in adult stature or in serum IGF-I but is associated with changes in body composition and possibly an increase in insulin sensitivity. These effects do not seem to be modulated by changes in circulating IGF-I.
| Introduction |
|---|
|
|
|---|
In several families reported to date, the affected subjects are compound HTZ for two distinct GHRHR mutations, suggesting that faulty GHRHR alleles may not be very rare in the general population (2, 8, 9, 10). Data obtained from a large GH-deficient (GHD) kindred from Pakistan (also due to a HMZ null GHRHR mutation) (11) suggested a "slight growth delay and, perhaps, adult height deficit in the HTZ state" (12), but the strength of the conclusions was reduced by the relatively small size of the sample. We have recently identified a large kindred of patients with familial IGHD, with 105 affected individuals over seven generations, residing in Itabaianinha County in the northeastern Brazilian state of Sergipe. In this population, IGHD is caused by a HMZ mutation in the splice donor site of intron 1 (IVS1 + 1G
A) of the GHRHR (13). GH-naive adults and children from this kindred have very low serum GH and IGF-I, reduced fat free mass, an increase in percentage of fat mass (FM), increased waist/hip ratio, increased total and LDL-cholesterol, and C-reactive protein (CRP) levels, and increased systolic blood pressure, but no insulin resistance and no evidence of premature atherosclerosis (14, 15, 16, 17). In a preliminary study, HTZ subjects have been reported to have lower serum IGF-I than HMZ normal (18), but the small number of subjects and their heterogeneous age compared with controls prevented the drawing of firm conclusions. Here we have studied a significantly larger number of subjects HTZ for the mutation and compared them with age- and sex-matched controls.
| Subjects and Methods |
|---|
|
|
|---|
Normal stature adult subjects aged 2575 yr from the Itabaianinha community who have first-, second-, or third-degree relatives with a history of GHD were recruited by advertisement on a local bulletin board and by word of mouth. The protocol was approved by ethics committes of both the Johns Hopkins University and the Federal University of Sergipe. All subjects signed written informed consent. Subjects with a history of childhood diseases known to influence adult stature such as malabsorption, renal failure, and exposure to chronic steroids were not included. Assuming that the SD of adult height in the Brazilian population is similar to that in the U.S. population (19), we have calculated that to detect a height difference of 5 cm between HTZ [wild-type (WT)/mutant (MT)] and normal (WT/WT) family members with a power of 80%, we needed 75 subjects per group. Subjects were initially asked to undergo collection of peripheral DNA by buccal swabs. Buccal swabs from a total of 240 adult individuals were collected, DNA was extracted by alkaline lysis, and genotyping for the IVS1 + 1 G
A mutation was carried out by denaturing gradient gel electrophoresis analysis of the amplified exon 1-intron 1 junction of the GHRHR, as previously described (13). In the group of 240 volunteers, we found 76 WT/MT subjects. Among the individuals who had enrolled in the study, we then identified 77 WT/WT subjects who were sex-matched and as age-matched as possible to the WT/MT group.
We asked them to report after an overnight fast to the local Little People Association building in daily groups of 20 during 8 consecutive days. Three WT/MT subjects were subsequently diagnosed with primary hypothyroidism (by moderate increase in TSH) and were only included in the height analysis. Due to logistical problems, not every patient had all the variables assessed.
Anthropometric measurement
The subjects height and body weight (BW) were measured with portable stadiometer and portable scale; body mass index (BMI) was calculated as weight in kilograms/height in meters squared. Body surface area (BSA) in square meters was calculated using the formula: w0.425 x h0.725 x 71.84 x 10 4, where w is the weight in kilograms, and h is the height in centimeters (20). Triceps, biceps, subscapular, and suprailiac skin folds were measured (in millimeters) with the subjects standing erect using the Lange skinfold caliper (Cambridge Scientific Industries, Cambridge, MD) on the right side of the body. Hip and waist circumferences were measured (in centimeters) as described previously (20), and waist/hip ratio was calculated (21).
FM and fat free mass (FFM) contents were measured using the infrared interactance method (22). This procedure involves placing a fiber optic probe tangentially to the belly of the subjects biceps and measuring the reflected energy in the near-infrared region at two different frequencies. A multiple regression equation is employed to use the interactance information in combination with other anthropometric data to predict body composition (22).
Laboratory assessment
Total cholesterol, triglycerides, and glucose were measured by the enzymatic Trinder colorimetric test (23). The HDL-C was separated using the phosphotungstic acid/magnesium chloride method, and the LDL-C concentration was calculated indirectly (Friedwald formula). Insulin was measured by a solid-phase, two-site chemiluminescent immunometric assay, with the sensitivity of 2 µIU/ml (Diagnostic Products Corporation, Los Angeles, CA) and the intra- and interassay variability were 4.26 and 5.43%, respectively; IGF-I was measured by the immunoradiometric assay, with double extraction and an assay sensitivity of 0.8 ng/ml (Diagnostic Systems Laboratories, Inc., Webster, TX). The intra- and interassay variabilities were 2.25 and 2.6%, respectively. The SD score (SDS) for serum IGF-I was calculated by subtracting the mean of IGF-I level of the age from the individual value and dividing this value by the SD of the respective mean age given by the manufacturer.
Insulin resistance was estimated using the homeostasis model assessment of insulin resistance (HOMAIR) with the formula: fasting serum insulin (microunits per milliliter) x fasting plasma glucose (millimoles per liter)/22.5 (24). CRP and lipoprotein (a) [Lp(a)] were measured by nephelometry (Dade Behring Marburg GmbH, Marburg, Germany; and Image Immunochemistry System, Beckman Coulter, Inc., Brea, CA, respectively). CRP and Lp(a) sensitivities were 0.175 mg/liter and 2 mg/dl, respectively. The intra- and interassay variabilities were 3.5 and 3.4% and 5.0 and 6.5%, respectively.
Insulin, CRP, and IGF-I were measured in serum after storage at 20 C. Lp(a) was measured in serum after storage at 4 C. All these samples were assayed together after the end of all the collections. Lipid profiles and serum glucose levels were measured daily in the clinical laboratory of the hospital of the Federal University of Sergipe.
Statistical analysis
Values for continuous variables are expressed as mean ± SD. Statistical analysis was performed by independent samples t test by using the statistical software SPSS/PC 11.5 (Statistical Packet for Social Science, Inc., Chicago, IL). Insulin, CRP, Lp(a), and HOMAIR data were transformed into decimal logarithm before analysis. To assess a possible effect of age on the variables, one-way ANOVA with Bonferroni post hoc test was used.
P values of 0.05 or less were considered statistically significant.
| Results |
|---|
|
|
|---|
Demographics of the subjects HTZ for the GHRHR mutation (WT/MT) and of the HMZ normal (WT/WT) subjects are shown in Table 1
. Sixty-six percent were under 50 yr of age. As expected, there was an overrepresentation of subjects who were first-degree relatives of GHD subjects in the WT/MT group (55.2 vs. 7.8%; P < 0.001). No difference in the percentage of second-degree relatives was observed (19.7 vs. 27.2%). The two groups were similar in sex distribution in all age categories. The percentage of each age category was similar in both groups, with the exception of an overrepresentation of younger subjects (25, 26, 27, 28, 29, 30) in WT/MT compared with WT/WT (25 vs. 14%; P = 0.009). We kept this group because age would not influence adult stature or SDS IGF-I. The significant differences between the two groups reported below were maintained when this age group was removed from the analysis.
|
Anthropometric data (mean ± SD) are shown in Table 2
. There was no difference in height and SDS IGF-I between the two groups. The minimal and maximal heights were 1.35 and 1.82 m in WT/WT and 1.36 and 1.84 m in WT/MT. BW, BMI, and BSA of WT/MT were lower than WT/WT. If one analyzes the data dividing subjects by sex, such differences remain for BW (males, 62.6 ± 11.5 vs. 71.3 ± 13.6 kg, P = 0.018; females, 57.9 ± 10.8 vs. 63.1 ± 13.6 kg, P = 0.036), and for BMI (males, 22.8 ± 2.5 vs. 25.6 ± 4.3 kg/m2, P = 0.06; females, 24.5 ± 4.9 vs. 26.7 ± 5.1 kg/m2; P = 0.032).
|
Thickness of all the skin folds was reduced in WT/MT in comparison to WT/WT. Waist and hip circumferences and FFM were reduced in WT/MT in comparison to WT/WT. FM percentage and waist/hip ratio were similar in the two groups (Table 3
).
|
Insulin and HOMAIR were lower in WT/MT than in WT/WT (Table 4
). The other parameters [total and fractionated cholesterol, triglycerides, Lp(a), CRP] were not different between the two groups.
|
In the WT/WT group, we did not observe any significant difference in any of the variables across age groups. In the WT/MT group, absolute and percent FM were lower in the 2530 than in the 4050 group (9.47 ± 6.67 vs. 16.76 ± 7.76 kg, P = 0.047; and 14.46 ± 10.13 vs. 27.67 ± 8.3%, P = 0.01).
| Discussion |
|---|
|
|
|---|
The IVS1 + 1G
A GHRHR mutation is predicted to cause retention of part or all of the very large intron 1, with introduction of a premature stop codon and likely complete lack of mature GHRHR protein. Therefore, the mutated allele would not produce any functional receptor (null allele). Because the GHRHR protein is expressed only in the pituitary and in the renal medulla (25), we cannot prove that in our patients the lack of one allele causes the receptor level to be 50% of that in normal subjects. However, the lack of effect of a null mutation on adult height can be applied to subjects carrying less disruptive mutations (such as missense mutations) that may allow expression of the receptor protein, albeit dysfunctional (26).
The lack of effect of heterozygosity for a GHRHR mutation on adult stature does not necessarily apply to mutations in other genes involved in controlling GH secretion or action. Although it has been suggested that a subgroup of children with idiopathic short stature have HTZ mutations in the GH receptor gene causing a partial resistance to GH (27, 28), no difference in height was found between subjects HTZ for the GH receptor gene E180 splice mutation and normal relatives in a large kindred of patients with GH resistance from Ecuador (29).
Despite the lack of difference in adult stature, we found some surprising effects of GHRHR mutation heterozygosity on BW and body composition. WT/MT individuals weighed less than WT/WT subjects, with reduction in BMI and BSA. To determine which compartment was mainly responsible for such weight reduction, we studied body composition by infrared interactance. Although we are aware that dual energy x-ray is the gold standard for these measurements, our technique allowed us to study the subjects locally. Although FFM was lower in WT/MT, there was no significant difference in absolute or percentage FM. There was a nonsignificant trend of FM to be lower in WT/MT and no significant difference in percentage FM. Similarly, both waist and hip circumferences were lower in WT/MT, with no difference in waist/hip ratio. These data seem to indicate that heterozygosity for the Itabaianinha GHRHR mutation causes a decline both in FM and FFM, albeit more marked for FFM. Reduction in FFM is commonly found in adults with acquired GHD (30), and we have found it in GH-naive GHD patients from Itabaianinha throughout life span (15, 16, 17). However, the reduction in FFM caused by GHD is usually associated with an increase in absolute and percentage FM (31). The WT/MT subjects have, on the contrary, a trend toward reduction in FFM. It is possible that individual effects of GH have different thresholds, although we cannot explain the fact that FM in WT/MT would not be expected in a status of mild GH deficiency. Our results, however, are reminiscent of the finding that adults with partial GHD have a decrease in FFM, but no significant reduction in absolute and percentage FM (32). Whatever the mechanism, the lack of any difference in serum IGF-I between WT/MT and WT/WT suggests that these effects are not modulated by serum IGF-I.
The only biochemical differences between the two groups were in insulin level and in HOMAIR, suggesting that WT/MT subjects have increased insulin sensitivity, similar to GHD subjects (16). In liver-specific IGF-I knockout animals (which have low serum IGF-I and high GH levels), blocking the effect of GH enhances insulin sensitivity, pointing to a direct effect of GH on insulin action (33). Although we did not directly evaluate insulin sensitivity by clamp method, it is conceivable that a mild reduction in GH secretion in the WT/MT group could cause an increase in insulin sensitivity despite any difference in serum IGF-I. Because increased insulin sensitivity has been associated with longevity (34), future studies will concentrate on determining the life span of WT/MT subjects.
Taken together, these data suggest that individual effects of GH may have different thresholds. It is conceivable that a modest reduction in GH secretion (as one may hypothesize occurs in WT/MT) might cause changes in insulin sensitivity and FFM, whereas reduction in somatic growth and liver IGF-I secretion, increase in FM, and changes in lipid metabolism may need more marked GH deficit. To test the hypothesis of reduced GH secretion, we are planning studies to compare the 24-h GH secretion profile of WT/MT and WT/WT subjects.
It is important to point out that the homogeneity of the study subjects makes it possible that the changes in body composition we observed are not due to the presence of the GHRHR mutation, but of some other genetic trait that may be associated with this mutation in the Itabaianinha cohort. Studies on other large kindreds with different GHRHR mutations, such as the Pakistani one (12), may help to resolve this question. In addition, first-degree relatives of GHD subjects were significantly overrepresented in the WT/MT group, adding a degree of genetic variability between the two groups. We do not believe that this alters our results, because all the subjects belong to the same population. In addition, they all live in the same community, with similar environmental exposures.
In conclusion, our data show that heterozygosity for the IVS1 + 1G
A GHRHR mutation does not reduce adult height but is associated with changes in body composition and indirect evidence of increased insulin sensitivity. Future studies will aim to determine whether the WT/MT status is associated with detectable reduction of GH secretion.
| Acknowledgments |
|---|
| Footnotes |
|---|
Disclosure Statement: R.M.C.P., M.H.A.-O., A.S., C.R.P.O., F.T.O., V.C.C., C.T.F., T.A.R.V., M.B.G., J.L.M.O., C.M.-S., I.E.S.R, and J.A.S.B.-F. have nothing to declare. R.S. receives grant support from Genentech (8/1/047/31/08).
First Published Online March 13, 2007
Abbreviations: BMI, Body mass index; BSA, body surface area; BW, body weight; CRP, C-reactive protein; FFM, fat free mass; FM, fat mass; GHD, GH-deficient or GH deficiency; GHRHR, GHRH receptor; GHRHR, GHRHR gene; HMZ, homozygous; HOMAIR, homeostasis model assessment of insulin resistance; HTZ, heterozygous; IGHD, isolated GH deficiency; LDL, low-density lipoprotein; Lp(a), lipoprotein (a); MT, mutant; SDS, SD score; WT, wild type.
Received January 16, 2007.
Accepted March 2, 2007.
| References |
|---|
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |