| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrinological Oncology |
Endocrinology Laboratory and Department of Medicine, Memorial Health Care (J.L.S., J.A.C., J.H.O.), and the Department of Medicine, University of Massachusetts Medical School (J.L.S., N.A.), Worcester, Massachusetts 01605; Faulkner Hospital and Tufts University School of Medicine (M.R.W.), Boston, Massachusetts 02130; the Department of Molecular Medicine, Karolinska Hospital (C.L., B.T.T.), Stockholm, Sweden; the Center for Endocrinology, Metabolism, and Molecular Medicine, Northwestern University Medical School (G.B.), Chicago, Illinois 60611; the Department of Pediatrics, Stanford University School of Medicine (R.L.H.), Stanford, California 94305; the Department of Medicine, University of Virginia Health Sciences Center (M.L.H.), Charlottesville, Virginia 22908; and the Molecular Neuro-Oncology Laboratory, Massachusetts General Hospital and Harvard Medical School (B.S.), Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: John L. Stock, M.D., Memorial Health Care, 119 Belmont Street, Worcester, Massachusetts 01605.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Localization of the gene(s) responsible for disease in such atypical families may aid in the understanding of the pathogenesis of MEN1. In addition, further study of the earliest changes in patterns of pulsatile GH release in familial acromegaly may allow more insight into the pathogenesis and natural history of this disease.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We report an MEN1 kindred in which four members have clinically
apparent disease: one patient with PHP in generation I, two patients
with acromegaly in generation II, and a patient with a pituitary tumor
and hyperprolactinemia in generation III (Fig. 1
). This
family is unusual in the preponderance of pituitary tumors and the low
penetrance of PHP. In addition, two of six clinically normal children
of a family member with acromegaly have biochemical and/or anatomical
evidence of early acromegaly. Linkage analysis suggests that the gene
responsible for the endocrine disorders in this unusual family is not
the same as the gene responsible for more typical MEN1 syndrome.
|
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
)Family member I-7 was an 84-yr-old woman with a 10-yr history of mild asymptomatic PHP. Her physical examination was unremarkable. Laboratory studies revealed elevated serum calcium (10.9 mg/dL), intact PTH (76 pg/mL; normal, 10–65), and fasting serum gastrin (472 pg/mL; normal, 0–200) concentrations, but she displayed no symptoms or signs of gastrointestinal disease. The serum concentrations of GH, insulin-like growth factor I (IGF-I), and PRL were normal.
Family member II-7 presented at age 60 yr with clinical signs of acromegaly during an evaluation for recurrent colon carcinoma. Laboratory studies revealed elevated serum GH (12.6 ng/mL) and IGF-I (14 U/mL; normal, <1.4) concentrations. The serum GH level increased further after TRH stimulation. The serum calcium concentration was 8.9 mg/dL, with an albumin level of 3.4 g/dL. The serum PRL concentration was 7.2 ng/mL. The patient died 10 months later of metastatic colon carcinoma.
Family member II-8 presented at age 58 yr with a 10-yr history of gradually increasing hand and foot size as well as increased sweating. The physical examination was significant for a prominent chin and large hands and feet. Laboratory studies revealed elevated serum GH (55.4 ng/mL; not suppressible with glucose), IGF-I (295 ng/mL; normal, 43–178), calcium (10.4 mg/dL), and C-terminal PTH (343 pg/mL; normal, 50–340) concentrations. A 1.8-cm pituitary tumor extending to the chiasm was treated with transsphenoidal resection. Pathological examination revealed a discrete adenoma with uniform cytomorphology. Half the cells contained acidophilic cytoplasm, and dense granules varying in size from 200–800 nm were seen by electron microscopy. The tumor showed diffusely uniform immunolabeling for GH, but not PRL.
Family member III-16 was evaluated at age 22 yr for amenorrhea after discontinuation of an oral contraceptive. Laboratory studies revealed an elevated serum PRL concentration of 46.8 ng/mL, normal serum concentrations of GH (0.22 ng/mL) and calcium (9.4 mg/dL), and a 1-cm pituitary adenoma. The patient was treated with bromocriptine, with resumption of menses, normalization of the serum PRL concentration, and resolution of the pituitary tumor on magnetic resonance imaging (MRI) scan.
Family screening
Twenty-six members of the kindred were screened after informed consent was obtained. Fasting venous blood was obtained for determination of serum chemistries, PTH, gastrin, PRL, GH, and IGF-I concentrations. All members over age 14 yr were invited for endocrine evaluation, and only one did not participate. In addition, leukocytes for DNA isolation were obtained from the three living members with known disease as well as other family members and spouses. Subjects living in the local area were screened at Memorial Hospital and blood samples from the others were drawn by their local physicians and shipped either frozen or at room temperature by overnight mail.
GH studies in offspring of subject II-7 with acromegaly
Additional studies were performed in the six offspring of subject II-7 who were admitted to Memorial Hospital. After a complete history and physical examination and dinner at 1700 h, subjects remained in bed and fasted, but were allowed water ad libitum. An iv catheter was placed, and blood was sampled at 20-min intervals for GH determinations from 2000–0800 h. The lights were extinguished at 2300 h, and sleep records were obtained through the night by observation. At 0800 h, the subjects were given 500 µg TRH. iv, over 1 min, and blood samples were obtained at 0, 15, 30, 45, 60, and 90 min for determination of serum levels of GH and PRL. The iv catheter was removed, and the subjects were discharged and allowed to resume their normal diet. A MRI scan of the pituitary (Signa 1.5 tesla, General Electric, Milwaukee, WI) was obtained that afternoon, and a glucose tolerance test was performed the following morning after a 10-h fast. Samples for determination of GH levels were obtained at baseline and 30, 60, 120, and 180 min after the administration of 75 g oral glucose. The baseline sample was also analyzed for IGF-I, IGF-binding protein-3 (IGFBP-3), and GH-binding protein (GHBP).
Assays
Serum chemistries were determined by automatic analyzer. PTH was measured in the samples obtained during family screening by a midregion-specific RIA [midregion immunoreactive PTH (iPTH)] using an antiserum generated in a goat against crude native human PTH (generously supplied by Dr. Lawrence Mallette) and 125I-labeled Tyr43-(44–68)-human PTH as radioligand, as previously described (10). Subsequently, PTH was also measured in selected family members using an immunoradiometric assay for intact PTH (Nichols Institute Diagnostics, San Juan Capistrano, CA). Gastrin and PRL were measured by RIA. GH in the family screening samples was determined using a rabbit antiserum to human GH and 125I-labeled human GH (Smith Kline Bio-Science Laboratories, personal communication). GH in the frequent samples obtained from the offspring of subject II-7 was measured by immunoradiometric assay (Nichols Institute Diagnostics, San Juan Capistrano, CA). The sensitivity of this assay was 0.02 ng/mL. The interassay coefficient of variation ranged from 3.5–7.2% over a concentration range of 1.4–13.0 ng/mL. The intraassay coefficient of variation ranged from 2.8–4.2% over a concentration range of 1.4–12.2 ng/mL. IGF-I in the family screening samples was determined by RIA after acid extraction by the method of Furlanetto (11) (Smith Kline Bio-Science Laboratories, personal communication). IGF-I was also measured by similar methods in separate samples from the offspring of subject II-7 by two independent laboratories (Smith Kline Bio-Science Laboratories) (12). IGFBP-3 was determined in serum by Western ligand blotting (13). GHBP in plasma was measured in living member II-8 with acromegaly and in the offspring of family member II-7 by an assay previously described (14). GHRH in serum was measured by RIA in the same subjects, courtesy of Dr. Wylie Vale (15).
Pulse analysis of frequent GH sampling
An objective, statistically based pulse detection algorithm (Cluster) was used to define significant pulses in the serum GH concentrations obtained at 20-min intervals for 12 h (16, 17). The following parameters were used: 1 x 1 test sample nadir and peak, t statistic of 1, minimum peak height of 0.7 ng/mL. Specifically identified properties of pulsatile GH release included pulse frequency (number of significant GH peaks per 12 h), mean interpulse interval (time in minutes separating consecutive peak maxima), mean pulse duration in minutes, mean pulse height (maximal GH concentration in a peak), mean incremental pulse height (difference between peak maximum and preceding nadir), interpulse valley mean (a valley was defined as a region embracing nadirs without significant intervening peaks), integrated GH concentration (IGHC; area under curve), and mean pulse area (integrated concentration under a peak in excess of the mean pre- and postpeak nadirs).
Attributes of the pulsatile GH release in the family members were compared with results in a group of normal healthy volunteers studied at the University of Virginia. Twelve men (aged 27–28 yr) and eight women (aged 23–24 yr) were sampled every 5 min for 24 h as previously described (17). To create a time series of serum GH concentrations comparable to those of family members, only every 20 min samples between 2000–0800 h were included in the analysis. GH was measured in the control samples by the same assay used to analyze GH in family members, except for the use of equine serum as the matrix instead of human serum. To correct for the matrix effect caused by the use of equine serum, GH concentrations in control subjects were divided by 2, as previously described (17), and the assay sensitivity was set at 0.25 ng/mL for all samples.
Linkage analysis
Blood samples were obtained from consenting family members, and high mol wt DNA was isolated from Epstein-Barr virus-transformed lymphoblastoid cell lines. Southern blot hybridization using restriction fragment length polymorphism/variable number tandem repeat markers and the PCR-based detection of microsatellite repeat polymorphisms (i.e. CA repeats) were performed according to standard procedures (18). Nine polymorphic markers at seven loci close to and flanking the MEN1 locus at 11q13 were analyzed: D11S149/pTHH26, D11S288/p3C7, CD20/pB1-21, PYGM/(CA)(GA), PYGM/pMCMP1, PYGM/cCL15, D11S913/AFM164zf12, D11S97/pms51, and INT2/SS6.
Family and genotype data were typed into the linkage data managing computer database LINKSYS. Two-point linkage between "the disease" and different markers was calculated with the LIPED computer program. The scoring of affected status was conservative to avoid incorrect diagnosis. Individuals I-7, II-7, II-8, and III-16 were scored as affected, whereas all other individuals at risk were scored as unknown. The allele frequencies of the two microsatellite markers, (PYGM/(CA)(GA) and D11S913), were assumed to be equal.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The serum calcium and albumin concentrations were normal in all family members except subject II-1. This subject, a healthy asymptomatic 47-yr-old man, was found to have a mildly elevated serum calcium concentration of 10.4 mg/dL. The screening serum midregion iPTH was 900 pg/mL (normal, 160-1000). Repeat sampling in this subject 20 months later revealed a serum calcium concentration of 10.6 mg/dL and an intact PTH concentration of 45 pg/mL (normal, 10–65). The screening serum gastrin concentration was 156 pg/mL (normal, 0–100 pg/mL), and it was 111 pg/mL in the repeat sample.
Despite normal serum calcium concentrations, levels of serum midregion
iPTH were mildly elevated in 4 subjects screened: II-3, 1030 pg/mL;
III-7, 1300 pg/mL; III-9, 1070 pg/mL; and III-10, 1060 pg/mL (normal,
160-1000). Subject III-7 was nursing at the time of screening. The
serum gastrin concentration was mildly elevated in subject III-3 (117
pg/mL), and serum PRL levels were normal in all subjects screened. The
serum GH concentration in subject III-19 was 6.3 ng/mL, and the serum
IGF-I concentration was 230 ng/mL (normal, 116–270). The serum GH
concentrations were less than the detection limit of 1.5 ng/mL in 15
other family members and were detectable, but 3.6 ng/mL or less, in the
other members screened. Elevated levels of serum IGF-I were found in 7
subjects. Five of these were offspring of deceased subject II-7 with
acromegaly (III-9, III-10, III-12, III-13, and III-14; Table 1). The serum IGF-I concentration in subject III-15 was
290 ng/mL, and that in subject III-8 was 370 ng/mL, with a repeat test
level of 238 ng/mL (normal, 116–270).
|
)There were no symptoms or physical findings of acromegaly or any other endocrinopathy in any of these six family members. Of the five subjects with elevated levels of serum IGF-I at screening, serum IGF-I concentrations were still elevated in four upon repeat testing using the same assay, serum IGF-I concentrations were elevated in three using the other IGF-I assay, and serum IGFBP-3 concentrations were elevated in three. The family member with the initially normal serum IGF-I level was found to have elevated levels by both assays upon repeat testing. Serum GH concentrations 1 h after glucose administration were less than 1 ng/mL, and GHBP and plasma GHRH levels were normal in all subjects.
The serum GH concentrations obtained at 20-min intervals for 12 h
were analyzed by the Cluster algorithm; the results are shown in Fig. 2. Family member III-12 had a mean GH concentration of
5.6 µg/L, almost 2 SD above the mean value for normal
young women (3.7 ± 1.2 µg/L). She had mean absolute (13.3
µg/L) and incremental (12.1 µg/L) GH pulse heights that were more
than 2 SD above the mean values for normal young women
(6.9 ± 2.8 and 5.2 ± 2.9 µg/L, respectively). Family
member III-13 had a mean GH pulse height of 12.6 µg/L, which was also
2 SD above the mean for normal women; the valley mean,
total IGHC, and mean GH concentration in this subject were at the upper
limits of normal. The remaining family members had normal patterns of
pulsatile GH release (data not shown).
|
The pituitary MRI scan in subject III-14 revealed an enlarged pituitary measuring 11 mm in all dimensions, with a convex upper border and suprasellar extension. The scans were normal in the other five family members studied.
Linkage analysis
To ascertain whether the abnormalities in this family represent a
variant of MEN1, members were genotyped for 9 markers at 7 polymorphic
loci (D11S149, D11S288, CD20, PYGM, D11S913, D11S97, and INT2) within
the MEN1 region at 11q13. Figure 1 illustrates how linkage to MEN1 was
excluded based on haplotype analysis. The MEN1 gene was previously
assigned very close to the PYGM locus and flanked by CD20 and D11S913.
This MEN1 region of chromosome 11, which was inherited from the
affected mother (I-7) by the affected daughter (II-8), was not
transferred to the affected granddaughter (III-16). In addition, three
markers located in the MEN1 region resulted in lod scores
(log10 of the odds ratio favoring linkage) of less than -3
(Table 2). For 1 of these, PYGM, no meiotic recombinants have been
found to date in more than 80 MEN1 families studied. From these data it
is unlikely that the endocrine abnormalities in this family are linked
to the MEN1 region.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The predisposing genetic defect in typical MEN1 families has been mapped to chromosome region 11q13, close to the PYGM locus, with no evidence of heterogeneity between 87 families analyzed until this report (7). Genetic linkage analysis demonstrated that the putative disease-associated haplotype of the 11q13 region is present in 9 and absent in 14 family members at risk. In several of these cases, the clinical status is different from that expected from the genetic analysis. Four subjects (II-10, II-11, III-17, and III-18) carry the same haplotype as the affected cases but have no clinical evidence of being affected. The ages of subjects II-10 (60 yr) and II-11 (59 yr) are well above the usual age of onset for MEN1. Furthermore, family member III-16, who does not carry the disease-associated haplotype, is affected, and member II-1 also does not carry this haplotype and may be affected. Hence, for the family to be linked to the typical MEN1 locus at 11q13 would require that III-16 be a sporadic case of a pituitary tumor and hyperprolactinemia, and that subject II-1 either not be affected by PHP or be a sporadic case of male PHP. The possibility that the pituitary tumor in member III-16 is sporadic cannot be ruled out, but the probability is low considering its low incidence in the general population and its established occurrence in familial pituitary tumor syndromes. From our previous experience, we have encountered one case of sporadic PHP in the largest family known, which resides in Tasmania (19).
Clinical GH excess is unusual in MEN1 (20); it was noted in only one member of Marx’s kindred (9) and was not present in Ballard’s kindred (21). However, a number of patients with acromegaly are described in Ballard’s historical literature review of MEN1 (21), and in one series of pituitary adenomas resected in subjects with MEN1, 44% of the tumors were immunoreactive for GH (22). There have been several families reported with clustering of acromegaly, either with (23) or without (23, 24, 25, 26, 27, 28, 29, 30, 31) other associated endocrinopathies of MEN1. To date, no linkage data are available for these families.
Variants of typical MEN1 have been described. Four Canadian families (32, 33) and a family from the Pacific Northwest (34) have a high incidence of PHP and prolactinomas, and a lower incidence of carcinoid tumors and pancreatic islet cell disease (32, 33). Several members of these families did not have hypercalcemia at the time of pituitary diagnosis. Linkage analysis showed that this disorder mapped to the same region on chromosome 11 as typical MEN1 (35). The Tasman kindred has frequent PRL-secreting pituitary tumors and/or insulinomas, presenting in adolescence often before clinical evidence of PHP (36), but no cases of acromegaly (37). The 11q13 genetic defect has also been demonstrated in affected members of this kindred (38).
Screening of the family presented here revealed one member (II-1) with possible mild asymptomatic PHP. Several other members had mildly increased serum iPTH concentrations measured in a midregion-sensitive assay, but all had normal measurements of serum calcium and intact PTH, making PHP unlikely. The intact PTH assay is more sensitive and specific than the midregion assay (39), but was unavailable to us at the time of the initial screening. Fasting serum gastrin levels were mildly increased in two family members with no clinical evidence of ulcer disease. Other islet cell hormones were not measured, and it is possible that pancreatic involvement was present in some asymptomatic individuals (40).
Several asymptomatic family members were found to have laboratory
findings consistent with early acromegaly. Screening revealed increased
serum IGF-I levels in five of the six offspring of subject II-7 with
acromegaly. Two other independent measures of IGF-I revealed increased
levels as well (Table 1). The serum IGF-I concentrations
in subjects III-13 and III-14 were elevated in all three
determinations. Serum IGF-I concentrations peak during puberty and then
gradually decline throughout adulthood (41). Subjects III-13 and III-14
were the youngest family members screened, but their serum IGF-I
concentrations were above the normal range for normal 21- and 20-yr-old
subjects (41). Serum IGFBP-3 concentrations were also elevated in
these two family members. IGFBP-3 is the major GH-dependent IGFBP in
serum, where it acts as a carrier for IGF-I and IGF-II (42), and its
level is elevated in acromegaly (43) and decreased in GH deficiency
(44). The elevated serum levels of IGF-I and IGFBP-3 suggest that
these offspring, particularly III-13 and III-14, might be affected with
acromegaly in an early and asymptomatic phase. Further testing supports
this. The pituitary gland was abnormal in subject III-14, measuring 11
mm in all dimensions, with a convex upper border and suprasellar
extension. The GH response to TRH was abnormal in subject III-13,
although the rise in GH in this subject was later than that usually
seen in acromegaly (45) and may have represented a coincidental random
GH pulse.
In acromegaly, pulsatile GH release is disorderly, with higher rates of basal secretion than occur in normal subjects (46). In this study, blood samples were obtained at 20-min intervals for 12 h overnight in six offspring of family member II-7 with acromegaly to compare with data previously collected in normal men and women of similar ages (17). Although retrospective, the subjects had similar ages and were studied under similar conditions with the same GH assay (except for serum matrix differences). Two of the female offspring (subjects III-12 and III-13) had high normal mean GH concentrations and elevated mean GH pulse heights compared to normal young women. However, the indexes that correspond to elevated basal secretion (valley mean concentration, nonpulsatile IGHC) were not elevated in these subjects or in any of the other offspring, contrary to what has been described in acromegaly (17). The number of pulses and all other attributes of pulsatile release determined by Cluster were within the range of the normal subjects. Thus, the nocturnal patterns of GH release in these subjects did not resemble those previously described in patients with somatotroph adenomas (17, 46). However, the pattern of GH release in very early acromegaly, before clinical presentation, is not known. Alternatively, the high amplitude GH peaks and elevated IGF-I concentrations in subjects III-14 (aged 20 yr), III-13 (aged 21 yr), and III-12 (aged 23 yr), could reflect a delayed resolution of the changes in the GH-IGF axis during puberty (41, 47).
The high affinity GHBP is structurally related to the GH receptor, and
plasma levels of GHBP may reflect the GH receptor concentration in
tissue and possibly tissue responsiveness to GH (48). It is possible
that abnormalities in GHBP may relate to syndromes of GH excess, but it
is unlikely in this family given the normal plasma GHBP concentrations
in patient II-8 with acromegaly and the possibly affected offspring of
patient II-7 with acromegaly. In typical pituitary acromegaly, GHBP
concentrations are in the low normal range (49), and GHBP does not
significantly affect the GH immunoassay (50). Other possible
explanations for the discrepancy between IGF-I and GH levels in these
subjects include increased sensitivity to GH (51), abnormal GH
structure (52), anti-GH receptor antibodies (53), simultaneous
increased secretion of somatostatin (54), or abnormal clearance of
IGF-I. An ectopic GHRH-secreting tumor is an unusual cause of
acromegaly, may be clinically indistinguishable from pituitary
acromegaly (55), and may be associated with the MEN syndrome (22). The
lack of diffuse somatotroph hyperplasia in patient II-8 and the normal
serum levels of GHRH in patient II-8 with acromegaly and the possibly
affected offspring of patient II-7 with acromegaly suggest that ectopic
GHRH is an unlikely cause of acromegaly in this family. A subgroup of
GH-secreting pituitary tumors has been described (56) in which a point
mutation in the 
-subunit of the guanine nucleotide-binding protein
Gs leads to constitutive activation of adenylyl cyclase and
increased GH secretion (gsp mutation) similar to the
activating mutations described in tissues from patients with the
McCune-Albright syndrome (57, 58). This mutation has been described in
the pituitary from a patient with MEN1 and a mixed GH cell-PRL cell
adenoma (59) and in a non-MEN1 somatotropinoma that also revealed
allele loss of chromosome 11 (60). The recent demonstration of
McCune-Albright syndrome in a MEN1 kindred (61) suggests a link between
the MEN1 gene and the G protein-coupled transduction system in the
plasma membrane.
Genetic testing for MEN1 may be complicated by the heterogeneous nature of the germinal mutation, particularly in variant families (38). Localization of the gene(s) responsible for disease in such atypical families may aid in the understanding of the pathogenesis of MEN1. In addition, further study of the very earliest changes in the pulsatile GH secretion in possibly affected family members may allow more insight into the pathogenesis and natural history of acromegaly.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Present address: Department of Molecular Genetics and Oncology,
Bristol-Myers Squibb Pharmaceutical Research Institute, Princeton,
New Jersey 08543.
Received May 30, 1996.
Revised September 26, 1996.
Accepted October 10, 1996.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
gene in a pituitary adenoma of a patient with multiple endocrine
neoplasia (MEN) type 1. Endocrinol Jpn. 39:319–324.[Medline]
This article has been cited by other articles:
 |
|
 |
|
  B. T. Teh, S. Kytölä, F. Farnebo, L. Bergman, F. K. Wong, G. Weber, N. Hayward, C. Larsson, B. Skogseid, A. Beckers, et al. Mutation Analysis of the MEN1 Gene in Multiple Endocrine Neoplasia Type 1, Familial Acromegaly and Familial Isolated Hyperparathyroidism J. Clin. Endocrinol. Metab., August 1, 1998; 83(8): 2621 - 2626. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
