0021-972X/78/4705-1065$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1978 by The Endocrine Society
Vol. 47, No. 5 Printed in U.S.A.
Serum Neurophysins in Familial Central Diabetes Insipidus J. J. LEGROS AND J. CRABBE Neuroendocrine Unit, Radioimmunoassay Laboratory, Department of Internal Medicine, University of Liege (J.J.L.), and the Endocrinology Section, Departments of Internal Medicine and Physiology, University of Louvain (J.C.), Belgium ABSTRACT. Antibovine neurophysin antibodies (anti-bNpI and/or anti-bNpII) are present in certain patients with familial central diabetes insipidus; these are of exogenous origin, as they are not present in patients who have not received treatment with crude posterior pituitary extracts over the years preceding the analysis. Immunoreactive neurophysins were detectable in the blood of five patients with familial central diabetes insipidus, and in two of them, the levels increased after a short period of water restriction. There is marked polymorphism of these neurophysins from one serum to another: neurophysin I was consistently absent, while
T
HE NEUROPHYSINS are the hypothalamo neurohypophysial carriers of vasopressin (AVP) and oxytocin (OT) (1). Two principle neurophysins are present in the majority of animal species, characterized primarily by their electrophoretic properties. In the pig (2), man (3), and the rat (4), the major neurophysin which migrates most anodally is associated with AVP, while the less nodal species is associated with OT. In cattle, in contrast, the anodal bovine neurophysin (bNpI) is associated with OT and the bovine neurophysin demonstrating cathodal migration (bNpII) is associated with AVP (5, 6). It has now been clearly demonstrated that the release of active hormones is simultaneous with neurophysin release, both during basal secretion and after stimulation. Furthermore, various in vivo studies have shown that the specific release of one neurophysin can be stimulated without the release of the other in cattle, pigs, and in man (see review in Ref. 7). Quantitative clinical study of neurohypophysial function in man can be conducted using direct neurophysin assay (8-10). There
Received November 28, 1977. Address requests for reprints to: Dr. J. J. Legros, Secteur de Neuroendocrinologie, C. H. U. Local 4/12, University of Liege, Sart Tilman, 4000 Liege, Belgium.
neurophysin II, accessory neurophysins, and other immunoreactive substances not present in normal sera were sometimes present in variable amounts. Immunoreactive AVP was undetectable in the urine of all patients, while immunoreactive OT was found in three of them; the latter substance could, however, be arginine vasotocin. Data are presented suggesting that the association between the biosynthesis of neurophysin I and AVP on the one hand, and neurophysin II and OT on the other hand is maintained in patients with isolated AVP deficiency on the basis of a congenital defect. (J Clin Endocrinol Metab 47: 1065, 1978)
is, however, little data in patients with diabetes insipidus and some of it is conflicting, in part because of the presence of circulating antineurophysin antibodies (11), which could interfere in RIA methods, and in part because of the failure to distinguish between organic (posttraumatic, neoplastic, etc.) and idiopathic diabetes insipidus. In this work, we studied eight patients from one family with familial central diabetes insipidus to determine the levels of serum neurophysins during dehydration and characterize the circulating neurophysins by their electrophoretic characteristics. The measurement of AVP and OT in the urine allowed us to determine whether the specific association of one neurophysin and one hormone was maintained in this disorder. Materials and Methods Clinical material Eight patients (H, D, C, G, A, I, P, and S) were studied; the family tree is shown in Fig. 1. Table 1 shows the age, sex, and treatment given in the preceding years. Some clinical characteristics of this family were described previously (12). In all cases, treatment was withdrawn 48 h before the period of observation. A fasting blood sample (±10 cc) was taken at 0900 h without water restric-
1065
1066
LEGROS AND CRABBE
JCR&M
1978 No 5
cc/liter Tween 20; 50,000 IU/liter Trasylol; or a solution containing 100 ng bNpI in the incubations with [125I]bNpI or bNpII (incubation with [I25I]bNpII). After 4 days at 4 C, an aliquot of the incubation medium was spotted into a rectangle (1 X 0.5 cm) of Whatman 3 MM paper and separation of the free [I25I]bNp (migrating in its usual position) and of [125I]bNp bound to antibody (remaining at the origin) was effected by starch gel electrophoresis (13) over 4 h at 220 volts and 4 C. The gel was subsequently sliced at 0.5-cm intervals and each piece was counted in a y-counter. The ratio of the radioactivity remaining at the origin (B) to that migrating in the free neurophysin zone (F) was calculated for each serum. • Total immunoreactive neurophysins (IRN) were measured by the RIA technique described in detail FIG. 1. Family tree of family V.I., of which certain mempreviously (14). We have recently confirmed that bers have familial central diabetes insipidus. our RIA detects the neurophysins liberated during TABLE 1. Name,'birth date, and treatment given to eight nicotine infusion (15). bNpII was used as the labeled patients with familial central diabetes insipidus antigen and as the standard; the antiserum used (A5IV) reacts equally with porcine neurophysins I Treatment given in and II, bNpI, bNpII, and purified human neurophyPatient Sex Birth date the years before samsin I (hNpI) (16); furthermore, the immunological pling" reaction is completely inhibited by neurohypophyHA Nov. 1032 M PPE (im injection) sial extracts from either goat (17), rat (18), guinea DA F May 1931 PPE (im injection) PPE (Snuff) C F Aug. 1934 pig, dog, or monkey (unpublished observations). GA M Feb. 1937 PPE (im injection) Thus, it presumably detects an antigenic fragment A M Oct. 1961 PPE (Snuff) which is common to the different neurophysins of I F Nov. 1961 SV (LVP) these mammals. All measurements were done in P M Nov. 1963 SV (LVP) S F Aug. 1966 SV (LVP) triplicate in a single assay, the mean coefficient of The patients whose initial is followed by a triangle variation, calculated from 10 samples each meareceived only brief hormone therapy in the years before sured 10 times within a single assay, was 5.68 ± analysis. The others ceased their hormone therapy at 2.81% (SD). Each serum was assayed in triplicate least 2 days before sampling. " PPE, Posterior pituitary extract; SV, synthetic vaso- within the same run. Urinary AVP and OT were measured by specific pressin. antisera, as previously described (19), after the filtion; the patient was weighed and asked to empty tration of 2 ml urine on a column of Sephadex G-75 the bladder, the urine being discarded. Water was (30 x 1 cm) in a buffer of ammonium bicarbonate subsequently restricted for 3 h, at the end of which (0.05 M, pH 7.5; containing 0.9% NaCl and 0.1% egg time the weight was again recorded, blood was albumin). This technique enables one to discard taken, and urine was collected. An aliquot of urine immunoreactive substances of higher molecular was immediately acidified to pH 4 with HC1 (0.1 weight than the free nonapeptides (20). The mean N) and was kept at -20 C. After separation from coefficients of variation (CV), calculated from 10 samples each measured 10 times within a single the blood, the sera were stored at — 20 C. assay, were 8.63 ± 0.5% for AVP and 7.55 ± 3.2% for OT. Each sample was assayed in triplicate Analyses within a single assay. Anti bNpI and bNpII antibodies were sought The circulating neurophysins in the sera obusing starch gel electrophoresis. Tracer quantities tained after fluid restriction were separated by (±5000 cpm in 0.1 ml) of [125I]bNpI or [125I]bNpII starch gel electrophoresis and were subsequently were incubated with 0.1 ml serum from each patient extracted by the technique previously described and 0.1 ml either buffer (0.05 M ammonium bicar- (3). Four sera from individuals suffering from debonate, pH 7.5; 0.1% egg albumin; 0.02 M EDTA; 20 hydration as a result of hyperosmolar diabetic coma
I'd) 'A i '
I]bNpII previously incubated with the serum of eight patients with hereditary central diabetes insipidus
Results
Incubation was performed in the presence or absence of unlabeled neurophysin. Separation was by starch gel electrophoresis. Values given are percentages.
The bound to free (B:F) ratios in each of the eight sera studied are shown in Table 2. In four of these (H, C, D, and A), more than 10% of the [125I]bN,,I remained at the origin; similar results were obtained in three sera (D, C, and A) using [12r>I]bNpII. In all of these cases, addition of 100 ng cold neurophysin caused a clear-cut decrease in the amount of radioactivity remaining at the origin. In three patients, the presence of antibodies directed against bN,,II interfered with the RIA measurement. Therefore, only the five patients whose sera showed less than 10% adsorption of [12T]bN,,II at the origin in the absence of unlabeled bN,,II were studied subsequently. Figure 2 shows the osmolalities and blood IRN levels as well as the weights of the five patients before and after fluid restriction; the volume, osmolality, and AVP and OT levels in the urine are given in Table 3. The effectiveness of the fluid restriction was shown by the degree of weight loss and the increase in serum osmolality. Furthermore, there was an elevation (small in three cases, large in two cases) in serum IRN levels. Urinary osmolality was always less than serum osmolality, while serum osmolality had increased from a mean of 297.4 to a mean of 302.6 mosm/kg (2 P < 0.01). lmmunoreactive AVP was undetectable in all urine specimens, while OT could be detected in significant quantities in the urine of patients G, I, and P. The identification of the circulating neurophysins is shown in Fig. 3 in patients with
Patient
[iar'I]bNpI
[12nI]bN,,I + 100 ng bNpI
H D C G A I P S
71.9 89.0 89.7 6.0 93.9 3.6 3.9 3.5
14.9 19.7 60.2 1.8 30.8 3.5 4.1 2.3
bNnll
bN,,II + 100 ng bNpII
5.9 94.1 13.6 9.4 47.8 2.2 2.2 2.5
4.5 20.0 5.7 2.1 7.7 2.2 3.0 2.4
idiopathic diabetes insipidus. The results were expressed, according to the method of Cheng and Friesen (9), as a percentage of the total of immunoreactive materials found in each gel taken as 100%. The migration profile differed greatly from one serum to another; only one characteristic was common, the absence of an immunoreactive material in the usual zone of migration of hNpI which migrates slightly less anodally than bN,,I (21). The other IRNs showed marked polymorphism. Substances migrating more anodally than bN,,I were found in five sera, while substances showing migration similar to that of bN,,II [corresponding to the migration of human neurophysin II (hNpII)] (21) were present in patients G, H, and I. Neurophysins migrating less rapidly than bN,,II were found in patients P, G, and H. Finally, immunoreactive materials migrating close to the origin or even clearly cathodally, were found in patients G, H, I, and S. The identification of circulating neurophysins in four patients with hyperosmolar coma is shown in Fig. 4. In contrast to the situation in patients with central diabetes insipidus, the electrophoretic patterns appear relatively homogeneous. The major species are hNpI and hN,,II as compared with minimal quantities of immunoreactivity in the more anodic and cathodic zones than the two principle neurophysins.
LEGROS AND CRABBE
1068
JCK&M • 1978 Vol47 • No S
90 -
FIG. 2. Body weight, serum osmolality, and IRN levels in five patients with familial central diabetes insipidus before and after fluid restriction for 3 h.
Before
TABLE 3. Volume, osmolality, and immunoreactive AVP and OT in the urine during a period of fluid restriction of 3 h in five patients with hereditary central diabetes insipidus Patient
G Urinary volume 840 (ml/3 h) Osmolality 256 (mosm/kg) AVP (ng/ml)
Vol. 47, No. 5 Printed in U.S.A.
Serum Neurophysins in Familial Central Diabetes Insipidus J. J. LEGROS AND J. CRABBE Neuroendocrine Unit, Radioimmunoassay Laboratory, Department of Internal Medicine, University of Liege (J.J.L.), and the Endocrinology Section, Departments of Internal Medicine and Physiology, University of Louvain (J.C.), Belgium ABSTRACT. Antibovine neurophysin antibodies (anti-bNpI and/or anti-bNpII) are present in certain patients with familial central diabetes insipidus; these are of exogenous origin, as they are not present in patients who have not received treatment with crude posterior pituitary extracts over the years preceding the analysis. Immunoreactive neurophysins were detectable in the blood of five patients with familial central diabetes insipidus, and in two of them, the levels increased after a short period of water restriction. There is marked polymorphism of these neurophysins from one serum to another: neurophysin I was consistently absent, while
T
HE NEUROPHYSINS are the hypothalamo neurohypophysial carriers of vasopressin (AVP) and oxytocin (OT) (1). Two principle neurophysins are present in the majority of animal species, characterized primarily by their electrophoretic properties. In the pig (2), man (3), and the rat (4), the major neurophysin which migrates most anodally is associated with AVP, while the less nodal species is associated with OT. In cattle, in contrast, the anodal bovine neurophysin (bNpI) is associated with OT and the bovine neurophysin demonstrating cathodal migration (bNpII) is associated with AVP (5, 6). It has now been clearly demonstrated that the release of active hormones is simultaneous with neurophysin release, both during basal secretion and after stimulation. Furthermore, various in vivo studies have shown that the specific release of one neurophysin can be stimulated without the release of the other in cattle, pigs, and in man (see review in Ref. 7). Quantitative clinical study of neurohypophysial function in man can be conducted using direct neurophysin assay (8-10). There
Received November 28, 1977. Address requests for reprints to: Dr. J. J. Legros, Secteur de Neuroendocrinologie, C. H. U. Local 4/12, University of Liege, Sart Tilman, 4000 Liege, Belgium.
neurophysin II, accessory neurophysins, and other immunoreactive substances not present in normal sera were sometimes present in variable amounts. Immunoreactive AVP was undetectable in the urine of all patients, while immunoreactive OT was found in three of them; the latter substance could, however, be arginine vasotocin. Data are presented suggesting that the association between the biosynthesis of neurophysin I and AVP on the one hand, and neurophysin II and OT on the other hand is maintained in patients with isolated AVP deficiency on the basis of a congenital defect. (J Clin Endocrinol Metab 47: 1065, 1978)
is, however, little data in patients with diabetes insipidus and some of it is conflicting, in part because of the presence of circulating antineurophysin antibodies (11), which could interfere in RIA methods, and in part because of the failure to distinguish between organic (posttraumatic, neoplastic, etc.) and idiopathic diabetes insipidus. In this work, we studied eight patients from one family with familial central diabetes insipidus to determine the levels of serum neurophysins during dehydration and characterize the circulating neurophysins by their electrophoretic characteristics. The measurement of AVP and OT in the urine allowed us to determine whether the specific association of one neurophysin and one hormone was maintained in this disorder. Materials and Methods Clinical material Eight patients (H, D, C, G, A, I, P, and S) were studied; the family tree is shown in Fig. 1. Table 1 shows the age, sex, and treatment given in the preceding years. Some clinical characteristics of this family were described previously (12). In all cases, treatment was withdrawn 48 h before the period of observation. A fasting blood sample (±10 cc) was taken at 0900 h without water restric-
1065
1066
LEGROS AND CRABBE
JCR&M
1978 No 5
cc/liter Tween 20; 50,000 IU/liter Trasylol; or a solution containing 100 ng bNpI in the incubations with [125I]bNpI or bNpII (incubation with [I25I]bNpII). After 4 days at 4 C, an aliquot of the incubation medium was spotted into a rectangle (1 X 0.5 cm) of Whatman 3 MM paper and separation of the free [I25I]bNp (migrating in its usual position) and of [125I]bNp bound to antibody (remaining at the origin) was effected by starch gel electrophoresis (13) over 4 h at 220 volts and 4 C. The gel was subsequently sliced at 0.5-cm intervals and each piece was counted in a y-counter. The ratio of the radioactivity remaining at the origin (B) to that migrating in the free neurophysin zone (F) was calculated for each serum. • Total immunoreactive neurophysins (IRN) were measured by the RIA technique described in detail FIG. 1. Family tree of family V.I., of which certain mempreviously (14). We have recently confirmed that bers have familial central diabetes insipidus. our RIA detects the neurophysins liberated during TABLE 1. Name,'birth date, and treatment given to eight nicotine infusion (15). bNpII was used as the labeled patients with familial central diabetes insipidus antigen and as the standard; the antiserum used (A5IV) reacts equally with porcine neurophysins I Treatment given in and II, bNpI, bNpII, and purified human neurophyPatient Sex Birth date the years before samsin I (hNpI) (16); furthermore, the immunological pling" reaction is completely inhibited by neurohypophyHA Nov. 1032 M PPE (im injection) sial extracts from either goat (17), rat (18), guinea DA F May 1931 PPE (im injection) PPE (Snuff) C F Aug. 1934 pig, dog, or monkey (unpublished observations). GA M Feb. 1937 PPE (im injection) Thus, it presumably detects an antigenic fragment A M Oct. 1961 PPE (Snuff) which is common to the different neurophysins of I F Nov. 1961 SV (LVP) these mammals. All measurements were done in P M Nov. 1963 SV (LVP) S F Aug. 1966 SV (LVP) triplicate in a single assay, the mean coefficient of The patients whose initial is followed by a triangle variation, calculated from 10 samples each meareceived only brief hormone therapy in the years before sured 10 times within a single assay, was 5.68 ± analysis. The others ceased their hormone therapy at 2.81% (SD). Each serum was assayed in triplicate least 2 days before sampling. " PPE, Posterior pituitary extract; SV, synthetic vaso- within the same run. Urinary AVP and OT were measured by specific pressin. antisera, as previously described (19), after the filtion; the patient was weighed and asked to empty tration of 2 ml urine on a column of Sephadex G-75 the bladder, the urine being discarded. Water was (30 x 1 cm) in a buffer of ammonium bicarbonate subsequently restricted for 3 h, at the end of which (0.05 M, pH 7.5; containing 0.9% NaCl and 0.1% egg time the weight was again recorded, blood was albumin). This technique enables one to discard taken, and urine was collected. An aliquot of urine immunoreactive substances of higher molecular was immediately acidified to pH 4 with HC1 (0.1 weight than the free nonapeptides (20). The mean N) and was kept at -20 C. After separation from coefficients of variation (CV), calculated from 10 samples each measured 10 times within a single the blood, the sera were stored at — 20 C. assay, were 8.63 ± 0.5% for AVP and 7.55 ± 3.2% for OT. Each sample was assayed in triplicate Analyses within a single assay. Anti bNpI and bNpII antibodies were sought The circulating neurophysins in the sera obusing starch gel electrophoresis. Tracer quantities tained after fluid restriction were separated by (±5000 cpm in 0.1 ml) of [125I]bNpI or [125I]bNpII starch gel electrophoresis and were subsequently were incubated with 0.1 ml serum from each patient extracted by the technique previously described and 0.1 ml either buffer (0.05 M ammonium bicar- (3). Four sera from individuals suffering from debonate, pH 7.5; 0.1% egg albumin; 0.02 M EDTA; 20 hydration as a result of hyperosmolar diabetic coma
I'd) 'A i '
I]bNpII previously incubated with the serum of eight patients with hereditary central diabetes insipidus
Results
Incubation was performed in the presence or absence of unlabeled neurophysin. Separation was by starch gel electrophoresis. Values given are percentages.
The bound to free (B:F) ratios in each of the eight sera studied are shown in Table 2. In four of these (H, C, D, and A), more than 10% of the [125I]bN,,I remained at the origin; similar results were obtained in three sera (D, C, and A) using [12r>I]bNpII. In all of these cases, addition of 100 ng cold neurophysin caused a clear-cut decrease in the amount of radioactivity remaining at the origin. In three patients, the presence of antibodies directed against bN,,II interfered with the RIA measurement. Therefore, only the five patients whose sera showed less than 10% adsorption of [12T]bN,,II at the origin in the absence of unlabeled bN,,II were studied subsequently. Figure 2 shows the osmolalities and blood IRN levels as well as the weights of the five patients before and after fluid restriction; the volume, osmolality, and AVP and OT levels in the urine are given in Table 3. The effectiveness of the fluid restriction was shown by the degree of weight loss and the increase in serum osmolality. Furthermore, there was an elevation (small in three cases, large in two cases) in serum IRN levels. Urinary osmolality was always less than serum osmolality, while serum osmolality had increased from a mean of 297.4 to a mean of 302.6 mosm/kg (2 P < 0.01). lmmunoreactive AVP was undetectable in all urine specimens, while OT could be detected in significant quantities in the urine of patients G, I, and P. The identification of the circulating neurophysins is shown in Fig. 3 in patients with
Patient
[iar'I]bNpI
[12nI]bN,,I + 100 ng bNpI
H D C G A I P S
71.9 89.0 89.7 6.0 93.9 3.6 3.9 3.5
14.9 19.7 60.2 1.8 30.8 3.5 4.1 2.3
bNnll
bN,,II + 100 ng bNpII
5.9 94.1 13.6 9.4 47.8 2.2 2.2 2.5
4.5 20.0 5.7 2.1 7.7 2.2 3.0 2.4
idiopathic diabetes insipidus. The results were expressed, according to the method of Cheng and Friesen (9), as a percentage of the total of immunoreactive materials found in each gel taken as 100%. The migration profile differed greatly from one serum to another; only one characteristic was common, the absence of an immunoreactive material in the usual zone of migration of hNpI which migrates slightly less anodally than bN,,I (21). The other IRNs showed marked polymorphism. Substances migrating more anodally than bN,,I were found in five sera, while substances showing migration similar to that of bN,,II [corresponding to the migration of human neurophysin II (hNpII)] (21) were present in patients G, H, and I. Neurophysins migrating less rapidly than bN,,II were found in patients P, G, and H. Finally, immunoreactive materials migrating close to the origin or even clearly cathodally, were found in patients G, H, I, and S. The identification of circulating neurophysins in four patients with hyperosmolar coma is shown in Fig. 4. In contrast to the situation in patients with central diabetes insipidus, the electrophoretic patterns appear relatively homogeneous. The major species are hNpI and hN,,II as compared with minimal quantities of immunoreactivity in the more anodic and cathodic zones than the two principle neurophysins.
LEGROS AND CRABBE
1068
JCK&M • 1978 Vol47 • No S
90 -
FIG. 2. Body weight, serum osmolality, and IRN levels in five patients with familial central diabetes insipidus before and after fluid restriction for 3 h.
Before
TABLE 3. Volume, osmolality, and immunoreactive AVP and OT in the urine during a period of fluid restriction of 3 h in five patients with hereditary central diabetes insipidus Patient
G Urinary volume 840 (ml/3 h) Osmolality 256 (mosm/kg) AVP (ng/ml)
