| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
E47 Neurophysin Mutation
Divisions of Pediatric Endocrinology (C.P.M.) and Radiology (E.W., C.B.), Childrens Hospital and Regional Medical Center, Seattle, Washington 98105; and Division of Endocrinology, Metabolism and Molecular Medicine (Mi.I., J.L.J, Ma.I.), Northwestern University Medical School, Chicago, Illinois 60611
Address all correspondence and requests for reprints to: C. Patrick Mahoney, M.D., Pediatric Endocrine Division, CH-92, Childrens Hospital and Regional Medical Center, 4800 Sand Point Way NE, Seattle, Washington 98105-0371. E-mail: pmahon{at}chmc.org
Abstract
Postmortem examinations of the hypothalamus of patients with autosomal dominant neurohypophyseal diabetes insipidus (adNDI), which have been reported only on persons dying between the ages of 3787 yr, reveal the presence of the arginine vasopressin (AVP)-producing parvocellular neurons but the absence of 95% of the expected AVP-producing magnocellular neurons. To determine whether the clinical course of adNDI is compatible with the hypothesis that the neuropathologic findings are attributable to a progressive loss of magnocellular neurons beginning in early life, we performed posterior pituitary magnetic resonance imaging and water deprivation tests, including plasma ACTH measurements, on 17 affected members of a kindred with the
E47 neurophysin mutation whose ages ranged from 3 months to 54 yr. Nine adult nonaffected members (ages, 2056 yr) underwent these tests as controls.
All six children undergoing magnetic resonance imaging demonstrated a posterior pituitary hyperintense signal (PPHS). Eight of nine affected adults showed an absent or barely visible PPHS, whereas eight of nine age-matched nonaffected adults produced a normal size PPHS. During water deprivation tests, infants concentrated their urine normally, and a 3-month-old infant produced a high plasma AVP level of 15.7 pmol/liter. By school age, affected children were no longer able to concentrate their urine or prevent hypernatremia. Affected adults became dehydrated; their median plasma AVP level was less than 1.0 pmol/liter, but their median fasting plasma ACTH was 2-fold greater than the level of nonaffected adults (10.0 vs. 5.0 pmol/liter; P = 0.008).
These results suggest that adNDI is a progressive disease associated with chronic loss of the magnocellular neurons that supply AVP to the posterior pituitary but preservation of the parvocellular neurons that supply AVP and CRH to the median eminence and stimulate ACTH production during hypernatremia.
THE ARGININE VASOPRESSIN (AVP) gene, located on chromosome 20, encodes prepro-AVP (1, 2, 3). Prepro-AVP is translated in the hypothalamus on the ribosomes of the magnocellular and parvocellular neurons in the supraoptic nucleus (SON) and paraventricular nucleus (PVN). Prepro-AVP consists of its signal peptide, AVP, neurophysin II (NP), and copeptin domains. More than 30 mutations of the AVP gene leading to defects in the signal peptide, AVP and NP have been identified (4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27). All but one of the mutations are expressed clinically as autosomal dominant neurohypophyseal diabetes insipidus (adNDI) (25).
Postmortem neuropathologic examinations of patients with adNDI have been performed, including vasopressin immunohistochemical studies of the hypothalamus and pituitary gland (28, 29, 30, 31, 32). A marked reduction of AVP-containing magnocellular neurons was observed in the SON, an area normally occupied predominantly by these large neurosecretory cells. In the PVN, in which both AVP-containing magnocellular and parvocellular neurons are normally present, there was an even greater loss of the large neurons but a relative preservation of the small neurosecretory cells. There was also a dramatic reduction of AVP in the posterior pituitary and in the neurosecretory axons projecting from the SON and PVN to the posterior pituitary (32).
We recently performed DNA sequence analysis of members of a large kindred in which postmortem examination of a middle-aged woman with adNDI in 1967 revealed the presence of less than 5% of the expected magnocellular neurosecretory cells in the SON and PVN (31). To determine whether the clinical laboratory and radiologic manifestation of adNDI over a wide age range are compatible with the hypothesis that the neuropathic findings are due to a progressive loss of magnocellular neurons beginning in early life, we performed posterior pituitary magnetic resonance imaging (MRI) and water deprivation tests including plasma ACTH measurements on affected and nonaffected members of the kindred, whose ages ranged from infancy to over 50 yr.
Subjects and Methods
Subjects
The neuropathologic findings and methods of study of the kindred member who died have been previously reported (31). The current studies were performed on 17 affected members (8 males and 9 females; mean age, 23 yr; range, 3 months to 54 yr) from four generations of an American kindred of English-Irish ancestry. Nine healthy nonaffected members (ages, 2056 yr) of the kindred also participated in the project as adult controls. All studies performed were approved by the Childrens Hospital and Regional Medical Centers Institutional Review Board. Water deprivation tests were performed during infancy; but, as required by the Institutional Review Board, MRI scans were not done during infancy because the procedure would have required sedation. The only controls for children were MRI brain scans selected from the Childrens Hospital Radiology Departments files. The control brain scans had been performed on otherwise healthy, untreated children being evaluated for possible seizures. The children matched affected children in age and sex, and their scans, performed in the morning after an overnight fast, were reported as normal, including the presence of a posterior pituitary bright spot. Appropriate informed consents were obtained from adult subjects and from the parents of children participating in the study. The affected members or their parents completed a questionnaire form and were interviewed in person regarding the onset and clinical course of diabetes insipidus.
Nucleotide sequence analysis
Genomic DNA was extracted from whole blood using a DNA extraction kit (Puregene, Gentra Systems, Minneapolis, MN). As previously described (4), the AVP gene was amplified by PCR, and the PCR-amplified DNA was directly sequenced using a dRhodamine terminator cycle sequence kit and an ABI 377 automated DNA sequencer (PE Applied Biosystems, Foster City, CA).
Clinical procedures
Contiguous thin slice (3-mm thickness) T1-weighted sagittal and coronal images of the posterior pituitary, with and without fat saturation, were performed during MRI scanning (33). The MRIs were interpreted and scored by an experienced neuroradiologist (E.W.) who was blinded as to whether the MRI was from a control or affected subject. Laboratory tests included blood electrolytes, creatinine, osmolality, AVP, oxytocin, ACTH, and cortisol measurements, and urinary specific gravity and osmolality determinations. Laboratory tests and MRI examination were performed only after each subject or parent of a minor subject stated that neither food nor drink had been ingested during the previous 12 h, and affected members did not take desmopressin acetate (DDAVP) on the night before the tests. The three subjects who admitted to breaking their fasts came back on another day for testing.
Nine affected adult members of the kindred and nine age-matched nonaffected members had MRI scans. Eight affected and eight nonaffected adults had blood and urinary measurements. Eight affected infants and children had blood and urinary determinations, including one child who was tested both during infancy and childhood. The six affected children, but not the two affected infants, had MRI scans.
Equipment and methods
Pituitary MRI was performed using the Magnetom Symphony scanner (Siemens Co., Munich, Germany) employing Numaris 3.5, Va 13c version software. AVP and oxytocin were extracted and measured by RIAs (34). Osmolality of both urine and blood was determined by freezing point measurements using the Advanced Microosmometer Model 3300 (Advanced Instruments, Inc., Norwood, MA). Fasting AM serum cortisol and plasma ACTH were measured by fluorometric and immunoradiometric assays, respectively (35, 36). Statistical significance determinations of comparisons of the median fasting ACTH and cortisol levels of affected and nonaffected adult members of the kindred were performed using the two-sample Wilcoxon rank-sum (Mann-Whitney U) test to obtain two-sided P values.
Results
Detection of mutation
Direct sequence analysis revealed a 3-bp deletion (AGG) of one of two consecutive AGG sequences (nucleotide position 18241829) in exon 2 of the AVP gene (Fig. 1
). Sequence analysis of each allele after subcloning of PCR-amplified DNA confirmed the presence of this mutation in one allele (data not shown). This mutation, which has been previously described, eliminates glutamate at amino acid position 47 of the neurophysin moiety (7). The pedigree in Fig. 2
represents all of the affected and nonaffected members of the Washington State arm of a large U.S. kindred of adNDI (31). All individuals positive for the mutation were symptomatic except for two infants (Fig. 2
, IV-15 and V-4) who were less than 18 months old. All individuals negative for the mutation were over 3 yr of age and asymptomatic.
|
|
The parents noticed the onset of polyuria and polydipsia when their children were between 6 months and 3 yr of age. None of the children experienced an acute episode of dehydration or failure to thrive. Before the current study using genetic testing, the diagnosis of adNDI in this kindred was often delayed and was missed in two cases because the 8-h water deprivation tests, which had been performed during childhood, were interpreted as being equivocal or normal. All of the parents thought that the severity of polyuria and polydipsia increased during childhood, particularly during the first 10 yr of life. Affected mothers were able to breast-feed successfully. Plasma oxytocin determinations of fasting blood samples drawn from male subjects and from female subjects who at that time were not pregnant or lactating were in the normal range of 0.40.6 mU/liter in both affected and nonaffected members.
MRI findings
The PPHS of affected children was present, although smaller or slightly less intense than the PPHS of control children (Fig. 3
and Fig. 4, A and B
). In young adults, the PPHS in five of six affected members was absent or barely visible in contrast to the normal size of the PPHS in nonaffected members (Fig. 3
and Fig. 5, A and B
). By middle age, the bright spot was absent in three affected members, whereas it was absent in only one of three nonaffected members. The oldest nonaffected member being studied (age, 56 yr) failed to show a PPHS during two fasting MRIs even though he had a normal water deprivation test (Table 1
). After drinking 240 ml of water at bedtime and several times daily for 5 d, including on the day of his third MRI, he produced a 3+ PPHS (normal size, but slightly less intense). In contrast, the oldest affected member (age, 54 yr) still failed to exhibit a PPHS on her second MRI after following the same regimen of increased fluid intake plus discontinuing DDAVP the night before the MRI and matching her fluid intake to her urinary output during the night.
|
|
|
|
During infancy, urinary and plasma osmolality in affected individuals remained in the normal range of 2001,192 and 275295 mosmol/kg, respectively (Table 1
). There was a progressive decline, particularly in the first decade of life, in the ability of affected individuals to withstand a 12-h fast, as illustrated by patient V-1. At 3 months of age, patient V-1 was not only able to concentrate his urine and prevent hyperosmolality, but he also increased plasma AVP to 15.7 pmol/liter (normal range, 1.012). By 7 yr of age, his serum sodium and plasma osmolality rose to 152 and 310, respectively, but plasma AVP rose only to 1.0 pmol/liter (Table 1
).
Median plasma ACTH concentrations after 12 h of fasting in affected and nonaffected members were 10.0 (range, 7.720.5) and 5.0 (range, 2.97.4) pmol/liter (P = 0.008). Median plasma cortisol levels in affected and nonaffected members were 572.5 (range, 515.9670.4) and 355.9 (range, 182.0573.8) nmol/liter (P = 0.02). The reference ranges for morning plasma ACTH and serum cortisol were 1.511.2 and 124.0634.5 nmol/liter, respectively. In two affected adults tested in a nonfasting state and having normal serum sodium levels, morning plasma ACTH concentrations were 3.5 and 4.4 pmol/liter, and plasma cortisol levels were 253.3 and 275.9 nmol/liter.
Discussion
The origin of the PPHS has not been established, but the presence of some component of the AVP-NP-copeptin complex in the axons of the hypothalamohypophyseal tract is considered to be the most likely source (37, 38, 39, 40). The PPHS has been demonstrated in 10 of 14 children in whom the diagnosis of adNDI has been established, including all six of the children in our series (Figs. 3
and 4A
) (2, 41, 42, 43). In contrast, eight of nine affected adult members in our series, including a 31-yr-old woman who had never received antidiuretic therapy, showed an absent or barely visible pituitary bright spot, whereas eight of nine nonaffected members produced a normal size PPHS (Figs. 3
and 5A
). Of 29 previously reported adults with adNDI, only three exhibited a PPHS (2, 12, 18, 22, 41, 42, 43, 44). When combined with our study, the results suggest that about 90% (37 of 41) of adults with adNDI may be expected to show only a barely visible or absent PPHS.
Disappearance of the PPHS alone cannot be used as evidence of loss of vasopressinergic cells. The PPHS is frequently absent in conditions of increased AVP secretion such as nephrogenic diabetes insipidus and renal failure treated with hemodialysis (37). Excessive release of AVP and secondary loss of PPHS may also occur with old age, apparently in response to diminished renal function and chronically raised plasma osmolality (45, 46, 47). The absence of the PPHS during two fasting MRIs in the oldest nonaffected member (age, 56 yr) may have been due to aging because the PPHS was restored by 5 d of increased hydration, including on the day of the third MRI.
Measurement of plasma AVP as well as plasma and urinary osmolality in affected members after water deprivation provided confirmatory evidence that this form of adNDI is a progressive disease, especially during childhood. Infants were able to concentrate their urine and prevent plasma hyperosmolality. After infancy, affected children were no longer able to concentrate their urine; and by school age, they began developing hypernatremia and plasma hyperosmolality after fasting. The transition from infancy to school age is illustrated by the boy (Table 1
, V-1) who, after exhibiting a normal response to a water deprivation test including an elevated plasma AVP level at 3 months of age, showed the typical laboratory findings of diabetes insipidus at age 7 yr.
The number of neurons in the hypothalamus expressing AVP reaches the adult level as early as the second half of gestation (3). At birth, there still appears to be adequate remaining vasopressinergic cells in adNDI to maintain sufficient AVP production. The progressive decline between infancy and middle age of affected individuals to produce AVP and to withstand water deprivation suggests that the paucity of vasopressinergic neurons found at autopsy in later life stems from chronic loss (2, 13, 26). Consistent with the histologic changes of localized cellular destruction, postmortem examination of the brain of affected member II-4 in Fig. 3
in our series, as in other cases of adNDI, showed astrocytic proliferation restricted to the SON and PVN regions of the hypothalamus (28, 29, 30, 31, 32).
Previous studies suggest a potential cause for the loss of magnocellular neurons in this kindred of adNDI. DNA sequence analysis indicates that the gene mutation present in affected members produces an AVP precursor lacking GLU 47 in its NP moiety (7). GLU 47, as shown in the crystalline structure of bovine NP, is essential for this protein to form a salt bridge with AVP (48). Normal NP-AVP binding affinity is necessary for processing and trafficking the AVP precursors in neurons during axon transport from the SON and PVN to the posterior pituitary. Structural changes in NP have been associated with intracellular accumulation of mutant AVP precursors that have been postulated to be cytotoxic (2, 26, 49, 50, 51, 52, 53). In cell culture systems, structural changes in NP induced by the
E47 mutation have been associated with retention of AVP precursors in the endoplasmic reticulum, altered processing, and decreased cell viability (50).
The reason that there is a selective loss in the hypothalamus of magnocellular neurons but preservation of parvocellular neurons is unclear because both produce AVP. On the basis of vasopressin immunohistochemical studies, Bergeron et al. (32) suggest that adNDI represents a selective degeneration of the magnocellular system that projects to the posterior pituitary with sparing of the parvocellular component projecting to the median eminence and central brain regions. Parvocellular neurons project to the median eminence and coexpress CRH and AVP (3, 54). During hypertonic saline infusion, and presumably during water deprivation tests that induce hypernatremia, parvocellular neurons augment the release of ACTH by secreting AVP and CRH into the pituitary portal capillaries (55, 56, 57, 58). Consistent with preservation of this function of parvocellular neurons, the median fasting plasma ACTH level of the affected adults who developed hypernatremia in our study was 2-fold greater than that of nonaffected members of the kindred who did not develop hypernatremia.
Footnotes
Abbreviations: adNDI, Autosomal dominant neurohypophyseal diabetes insipidus; AVP, arginine vasopressin; MRI, magnetic resonance imaging or image; NP, neurophysin II; PPHS, posterior pituitary hyperintense signal; PVN, paraventricular nucleus; SON, supraoptic nucleus.
Received November 20, 2000.
Accepted November 8, 2001.
References
This article has been cited by other articles:
![]() |
L. Ye, X. Li, Y. Chen, H. Sun, W. Wang, T. Su, L. Jiang, B. Cui, and G. Ning Autosomal Dominant Neurohypophyseal Diabetes Insipidus with Linkage to Chromosome 20p13 but without Mutations in the AVP-NPII Gene J. Clin. Endocrinol. Metab., July 1, 2005; 90(7): 4388 - 4393. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. H. Christensen, C. Siggaard, T. J. Corydon, G. L. Robertson, N. Gregersen, L. Bolund, and S. Rittig Differential Cellular Handling of Defective Arginine Vasopressin (AVP) Prohormones in Cells Expressing Mutations of the AVP Gene Associated with Autosomal Dominant and Recessive Familial Neurohypophyseal Diabetes Insipidus J. Clin. Endocrinol. Metab., September 1, 2004; 89(9): 4521 - 4531. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. A. Friberg, M. Spiess, and J. Rutishauser Degradation of Wild-type Vasopressin Precursor and Pathogenic Mutants by the Proteasome J. Biol. Chem., May 7, 2004; 279(19): 19441 - 19447. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. T. Wahlstrom, M. J. Fowler, W. E. Nicholson, and W. J. Kovacs A Novel Mutation in the Preprovasopressin Gene Identified in a Kindred with Autosomal Dominant Neurohypophyseal Diabetes Insipidus J. Clin. Endocrinol. Metab., April 1, 2004; 89(4): 1963 - 1968. [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 |