0021-972X/98/$03.00/0
Journal of Clinical Endocrinology and Metabolism
Copyright © 1998 by The Endocrine Society
Vol. 83, No. 7
Printed in U.S.A.
Stature in Ecuadorians Heterozygous for Growth
Hormone Receptor Gene E180 Splice Mutation Does Not
Differ From That of Homozygous Normal Relatives*
ARLAN L. ROSENBLOOM, JAIME GUEVARA-AGUIRRE, MARY ANNE BERG†,
UTA FRANCKE
AND
Department of Pediatrics, University of Florida College of Medicine, Children’s Medical Services Center
(A.R.), 1701 SW 16th Avenue, Gainesville, Florida 32608; Institute of Endocrinology Metabolism and
Reproduction (J.G-A.), Quito, Ecuador; and Department of Genetics and Howard Hughes Medical Institute,
Stanford University Medical Center (M.S.B., U.F.), Stanford, California 94305-5323
ABSTRACT
Heterozygosity for certain mutations of the GH receptor (GHR)
gene has been proposed as the cause of partial resistance to GH, and
there has been a recent demonstration of a dominant-negative effect
of such a mutation in a mother and child. To examine the effect of
heterozygosity in a large genetically homogeneous population with
GHR deficiency, in which a substantial number of heterozygous (carrier) subjects and homozygous normal individuals can be compared,
we studied a population in Ecuador in which 70 individuals with GHR
deficiency were homozygous for the E180 splice mutation. We found
that 58 heterozygous relatives of probands were not significantly
shorter than 37 homozygous normal relatives [SD score (SDS) for
height 21.85 6 1.04 (SD) vs. 21.55 6 0.96, P . 0.10]. When only those
families with both homozygous normals and carriers were compared,
the 33 heterozygous and 29 normal relatives did not differ significantly in height SDS (21.98 6 1.07 vs. 21.77 6 0.91, P . 0.3).
O
VER 30 distinct mutations have been described since
characterization of the GH receptor (GHR) nearly a decade ago. These include a deletion and missense, nonsense,
frameshift, and splice site mutations that, in the homozygous
state, produce profound effects on stature and physical features
typical of severe GH/insulin-like growth factor-I (IGF-I) deficiency (1). The recent finding of heterozygosity for GHR mutations associated with partial resistance to GH in a group of
short statured patients partially responsive to GH and with low
GHBP levels, raised the question of whether certain mutations
might affect growth in the heterozygous state (2). It has been
recently demonstrated that alternatively spliced isoforms of the
GHR, expressed in transfected cells, modulate the function of
the normal full-length receptor (3), and that a single heterozygous mutation involving the cytoplasmic domain of the GHR
appears to have a dominant-negative effect in a mother and
daughter with short stature (4).
Received January 15, 1998. Revision received February 23, 1998. Accepted April 10, 1998.
Address all correspondence and requests for reprints to: Arlan L.
Rosenbloom, Department of Pediatrics, University of Florida College of
Medicine, Children’s Medical Services Center (A.R.), 1701 SW 16th Avenue, Gainesville, Florida 32608. E-mail: [email protected].
* This work was supported by NIH Grant DK-45830 and the Howard
Hughes Medical Institute.
† Current address: Department of Internal Medicine, University of
Iowa College of Medicine, Iowa City, Iowa 52242.
If heterozygosity for the E180 splice mutation were to influence stature, heights of heterozygous parents of probands would be expected to
correlate with those of probands and of carriers who are their offspring
and not with heights of their homozygous normal children. Parental
height SDS did not correlate with height SDS of affected offspring (r 5
0.24). For unaffected siblings as a group or analyzed separately as normals or carriers, there was a strong correlation between parental and
offspring SDS for height (P , 0.01 for all comparisons). Thus, the effect
of homozygosity for the GHR mutation was so profound as to abolish
parental influence on height, and there was no difference in the influence
of parental stature between carrier and noncarrier offspring. These findings demonstrate no meaningful effect on stature of heterozygosity for
the E180 splice mutation of the GHR, which is a functional null mutation
and, in the homozygous state, results in profound short stature from
severe insulin-like growth factor-I deficiency. (J Clin Endocrinol Metab
83: 2373–2375, 1998)
There has been little information about the influence on
stature of heterozygosity of those GHR mutations that
cause severe short stature and other clinical manifestations
of GHR deficiency (GHRD) in the homozygous state. Stature was thought to be below normal in unaffected siblings
and presumed heterozygous parents of Israeli patients (5,
6), but the reference standard for these observations was
not the immigrant middle eastern population of which
they were a part (7). To appropriately address this issue
requires an adequate number of individuals proven to be
homozygous normal or heterozygous for the GHR defect
causing GHRD in probands. The use of parents as presumed heterozygotes is problematic, because they would
have to be compared with their siblings, who may or may
not be heterozygotes, as well.
The Ecuadorian population with GHRD is the only large
genetically homogeneous population that has been reported
with GHRD, currently numbering 70 probands, all but one
homozygous for the E180 splice mutation (8). This made it
possible to apply contemporary molecular genetic techniques to identify the carrier state. In 1994, we reported an
insignificantly shorter mean stature among 41 individuals
who were carriers, compared with 24 homozygous normal
siblings (9). We have now expanded this study to a larger
number of relatives, and analyses indicate that heterozygosity for the E180 splice mutation does not affect stature.
2373
2374
JCE & M • 1998
Vol 83 • No 7
ROSENBLOOM ET AL.
Methods
Subjects
Parents or other relatives of 62 probands were measured for height
and studied for carrier status. Subjects and, if minors, their parents,
agreed to have height measurements with the understanding that these
were for study purposes and comparisons with affected relatives. Blood
spots were obtained for genetic analysis after explanation of the mechanism and inheritance of GHRD in this population and agreement by the
subject or parent with our policy for reporting results. This policy,
developed to avoid labeling or other social problems, is to report only
to the individual tested in a counseling session with J.G.-A.; in the case
of minors, results are retained at the Institute for Endocrinology Metabolism and Reproduction in Quito, Ecuador and are available when
the person reaches adulthood or marries. This approach was discussed
extensively within the community and universally approved.
There were 13 affected sibling pairs and one family with 3 siblings
affected; in these cases the average sd score (SDS) for height of the
affected siblings was used for the comparisons. Genetic and statural data
were available for 41 parent pairs and 6 single parents and 95 other
relatives. The latter included 74 first-degree relatives of probands (71
siblings and 3 offspring) and 21 second-degree relatives of probands (6
cousins, 8 aunts or uncles, 4 grandparents, and 3 nieces or nephews).
Non-first-degree relatives were added to expand the number of normal
relatives for comparison; this group provided 12 of the 37 normal relatives and 9 of the 58 heterozygous relatives. All subjects were between
the ages of 5 and 50 yr.
Height measurements
Stature was measured in centimeters, either with a fixed stadiometer
(Harpenden, Holtain Ltd., Crosswell, Crymych, Dyled, UK) at the Institute in Quito, or in the subjects’ homes with a Raven Minimetre (Raven
Equipment Ltd., Unit #4, Ford Farm Industrial Complex, Raintree Road,
Dunnow Essex CM6 1HU, UK) (10). Measurements were done three
times, and the average recorded. These height measurements were converted to SDS using U.S. population data (11). For affected children, the
last recorded height before starting replacement therapy with recombinant IGF-I was used (12).
Genetic analysis
All probands had been proven to be homozygous for the E180 splice mutation of the GHR gene responsible for GHRD in this population by restriction
analysis of a PCR product from exon 6 with the enzyme MnlI (13). This method
was also used to determine whether the unaffected relatives were heterozygous carriers of this mutation or homozygous normal.
Data analysis
The t test was used to compare mean SDS for height for homozygous
normal and heterozygous relatives. To explore the possible influence of
heterozygosity for the E180 splice mutation on stature, correlations for
height SDS were calculated between parents and their unaffected offspring, separately as carriers and normals, and combined. Within each
family, the mean SDS was used if more than one proband, parent, or
sibling of either type was studied.
Results
There was no significant difference between mean SDS in homozygous normal vs. heterozygous relatives (Table 1). Sibships of
probands with both normal and heterozygous (carrier) siblings
were separately analyzed to determine whether there was any bias
introduced by inclusion of families having only heterozygous and
no homozygous normal siblings. The mean SDS for normals and
heterozygotes in the subset of families with both kinds of siblings
also did not differ significantly (Table 1).
Heterozygous and homozygous normal relatives demonstrated a wide and overlapping range of SDS for height that
emphasized the lack of influence of heterozygosity for the
E180 splice mutation. Within the families in which both non-
TABLE 1. Comparison of SDS for height of normal and
heterozygous (for E180 splice mutation of GHR) relatives of
probands with GHRD (excluding parents)
Carrier
All relativesa
n 5 58
SDS 6 SD
21.85 6 1.04
Range
24.3 to 11.2
Siblings from families with
n 5 33
both noncarriers and
carriers in sibship
SDS 6 SD
21.98 6 1.07
Range
24.3 to 11.2
a
Noncarrier
n 5 37
21.55 6 0.96
23.9 to 10.4
n 5 29
21.77 6 0.91
23.9 to 10.4
P
.0.10
.0.30
Including families without noncarrier relatives.
carrier and carrier individuals were identified, the taller or
tallest individual was defined as one who was at least 0.5 sd
taller than the taller or tallest individual with a different
carrier status. In the 15 families with both carriers and noncarriers, carriers were the taller or tallest in 6, and noncarriers
in 3. In this normally short statured population (by North
American standards) only 6 individuals had heights at or
above zero SDS: 3 noncarriers at 0, 10.3, and 10.4 SDS and
3 carriers at 10.7, 11.0, and 11.2 SDS. Thus, in the population as a whole, the tallest individuals happened to be
heterozygous for the E180 splice mutation.
As we noted in the early description of a segment of the
Ecuadorian population with GHRD, there was no correlation
between mean parental height SDS and proband height SDS
(14) (Table 2). Correlations between stature of unaffected
siblings and parents were strong, and not different when
parents were compared with all their unaffected offspring, or
separately to their offspring who were homozygous normal
or carriers (Table 2).
Discussion
Despite extensive studies of the various mutations responsible for GHRD, there has been no systematic attempt to characterize the effects of the carrier state for these mutations on
stature. This is because small numbers of patients are affected
by any single mutation, making the accumulation of a statistically relevant population of carrier and noncarrier relatives
difficult. The Ecuadorian GHRD population, comprised of 70
patients, all but one homozygous for the same mutation, offers
a unique opportunity to determine the influence of heterozygosity for this mutation on stature. Although the analyses consistently show lower mean SDS for height for the heterozygous
group compared with the noncarrier relatives, the difference is
not significant, the ranges overlap completely, and the tallest
individuals in the population are carriers.
As we have previously noted, the effect of homozygosity
for this mutation is so profound that there is not a significant
influence of parental heights on proband heights, nor do the
proband heights correlate with unaffected sibling heights
(14). This lack of correlation for stature between the heterozygous carrier parents and affected offspring argues
against an effect on stature of the carrier state. This contrasts
with conditions such as Turner syndrome in which, despite
substantial growth retardation, the probands’ heights are
still strongly influenced by parental endowment and correlate with those of normal siblings (15). If heterozygosity for
STATURE WITH GHR MUTATION HETEROZYGOSITY
2375
TABLE 2. Correlations of height SDS among probands with GHRD, their parents, and unaffected siblings of the probands
Groups
Probands/parents
Parents/all siblings
n
Pairs compared
53/82
44a
42/67
Means 6
SD
r
P
28.39 (1.21)/22.27 (0.79)
0.244
NS
22b
210.5 to 26.3/24.5 to 20.8
22.29 (0.76)/21.61 (1.25)
0.774
,0.01
0.707
,0.01
0.809
,0.01
Ranges
Parents/normal siblings
30/22
15c
23.45 to 20.8/24.3 to 11.2
22.24 (0.73)/21.43 6 0.92
Parents/carrier siblings
34/39
18d
23.25 to 20.95/23.0 to 10.4
22.48 (0.70)/21.93 (1.50)
23.45 to 20.95/23.7 to 11.2
a
Within families means were used for affected siblings (2 in 14 families, 3 in 1), and 38 parent pairs; there were 6 single parents.
See above; represents 20 parent pairs, 2 single parents, and means for siblings within each family.
c
As above, represents 14 parent pairs and 1 single parent.
d
As above, represents 16 parent pairs and 2 single parents.
b
the GHR mutation in this population were to be phenotypically influential, one would also expect that the correlation
between stature of offspring who do not have GHRD and
that of their heterozygous parents would be influenced by
whether the offspring are homozygous normal or carriers. In
contrast to the lack of significant correlation of parental with
proband height SDS, there was a highly significant correlation between parental and unaffected offspring SDS, whether
the offspring were carriers or not.
The distinct mutations that have been identified in the
GHR gene include a number resulting in GHRD in the
homozygous or compound heterozygous state, several
purported to result in partial GH insensitivity in the heterozygous state (2), and one proven to do so in a dominantnegative fashion (4). In addition, a number of polymorphisms have been described that do not appear to influence
the GH/IGF-I axis (16, 17). The findings reported here indicate that there is no effect on stature of heterozygosity for
the E180 splice mutation that causes GHRD in the homozygous state. This mutation creates a new splice site within
exon 6 that is exclusively used and causes a predicted deletion of eight amino acids. The predicted mutant receptor
molecule has the potential to act in a dominant-negative
fashion by heterodimerization with the products of the normal allele. The lack of clinical evidence for this occurring
supports our previous hypothesis that the eight-amino acid
deletion causes protein misfolding and degradation (18).
Therefore, it is likely that the E180 splice mutation is a functional null mutation. In contrast, missense mutations and
structural mutations that lead to a stable mutant protein
might have effects on growth in heterozygotes.
Acknowledgments
We are grateful to Linda Allen of the Department of Health Policy and
Epidemiology of the University of Florida College of Medicine for assistance
with data analysis, Drs. Victor Martinez and Oswaldo Vasconez of the Institute
for Endocrinology Metabolism and Reproduction in Quito for assistance with
data collection, and Margaret Stanley for manuscript preparation.
Note Added in Proof
Since submission of this report, another family has been reported with
a heterozygous mutation resulting in moderate growth failure through
a presumed dominant negative effect on GH signaling (Iida K, Takahashi Y, Kaji H, et al. 1998 Growth hormone (GH) insensitivity syndrome
with high serum GH-binding protein levels caused by a heterozygous
splice site mutation of the GH receptor gene producing a lack of intracellular domain. J Clin Endocrinol Metab. 83:531–537).
References
1. Rosenbloom AL, Rosenfeld RG, Guevara-Aguirre J. 1997 Growth hormone
insensitivity. Ped Clin North Am. 44:423– 442.
2. Goddard AD, Covello R, Luoh S-M, et al. 1995 Mutations of the growth
hormone receptor in children with idiopathic short stature. N Engl J Med.
333:1093–1098.
3. Ross RJM, Esposito N, Shen XY, et al. 1997 A short isoform of the human
growth hormone receptor functions as a dominant negative inhibitor of the full
length receptor and generates large amounts of binding protein. Mol Endocrinol. 11:265–273.
4. Ayling RM, Ross R, Towner P, et al. 1997 A dominant-negative mutation of
the growth hormone receptor causes familial short stature. Nature Genet.
16:13–14.
5. Pertzelan A, Adam A, Laron Z. 1968 Genetic aspects of pituitary dwarfism due
to absence or biological inactivity of growth hormone. Isr J Med Sci. 4:895–900.
6. Laron Z. 1993 Laron syndrome: from description to therapy. Endocrinologist.
3:21–28.
7. Weingarten MA. 1992 Changing health and changing culture: the Yemenite
Jews in Israel. Westport, CT: Praeger.
8. Berg MA, Guevara-Aguirre JG, Rosenbloom AL, Rosenfeld RG, Francke U.
1992 Mutation creating a new donor splice site in the growth hormone receptor
genes of 37 Ecuadorean patients with Laron syndrome. Hum Mut. 1:24 –34.
9. Rosenbloom AL, Guevara-Aguirre J, Rosenfeld RG, Fielder PJ. 1994 Is there
heterozygote expression for growth hormone receptor deficiency? Acta Paediatr Suppl. 399:125–127.
10. Diamond FB Jr, Guevara-Aguirre J, Rosenbloom AL. 1994 Accurate determination of height using an inexpensive measuring device. Clin Pediatr. 33:172–174.
11. National Center for Health Statistics. 1997 NCHS growth curves for children
birth to 18 years: United States. In: Vital and health statistics. Series 11, number
165. (DHEW Publication No. (PHS) 78 –1650). Washington, DC: Government
Printing Office.
12. Guevara-Aguirre J, Rosenbloom AL, Vasconez O, Martinez V, Gargosky SE,
Rosenfeld RG. 1997 Two year treatment of growth hormone receptor deficiency (GHRD) with recombinant insulin-like growth factor-I in 22 children:
comparison of two dosage levels and to GH treated GH deficiency. J Clin
Endocrinol Metab. 82:629 – 633.
13. Berg MA, Peoples R, Perez-Jurado L, et al. 1994 Receptor mutations and
haplotypes in growth hormone receptor deficiency: a global survey and identification of the Ecuadorian E-180 splice mutation in an oriental Jewish patient.
Acta Paediatr Suppl. 399:112–114.
14. Guevara-Aguirre J, Rosenbloom AL, Vaccarello MA, Fielder PJ, Diamond FB,
Rosenfeld RG. 1991 Growth hormone receptor deficiency (Laron syndrome):
clinical and genetic characteristics. Acta Paediatr Scand [Suppl] 377:96–103.
15. Lemli L, Smith DW. 1963 The XO syndrome: a study of the differentiated
phenotype in 25 patients. J Pediatr. 63:577–588.
16. Kou K, Lajara R, Rotwein P. 1993 Amino acid substitutions in the intracellular
part of the growth hormone receptor in a patient with the Laron syndrome.
J Clin Endocrinol Metab. 76:54 –59.
17. Chujo S, Kaji H, Takahashi Y, Okimura Y, Abe H, Chihara K. 1996 No
correlation of growth hormone receptor gene mutation P561T with body
height. Eur J Endocrinol. 134:560 –562.
18. Francke U, Berg MA. 1993 Genetic heterogeneity in Laron syndrome. Acta
Paediatr Suppl. 391:3–7.