Abstract. In an attempt to clarify the mechanism responsible for the prolonged effect of

Transcription

1 Endocrinol. Japon. 1980, 27 (3), Prolonged. Antidiuresis by 1-Desamino-8-D-Arginine Vasopressin (DDAVP) : Correlation to Its Plasma Levels and Nephrogenous Cyclic AMP Production SHIGEHIRO KATAYAMA, AKIRA ITABASHI AND TOHRU YAMAJI The Third Department of Internal Medicine, Faculty of Medicine, University of Tokyo, Hongo, Tokyo, 113, Japan Abstract In an attempt to clarify the mechanism responsible for the prolonged effect of DDAVP (1-desamino-8-D-arginine vasopressin), plasma levels of DDAVP and nephrogenous cyclic AMP were determined in patients with diabetes insipidus after a single intranasal administration of 10 Đg of DDAVP. Plasma DDAVP levels were uniformly elevated within 30 min, and showed a peak ranging from 5.6 to 25.0 pg/ml between 30 and 120 min. The subsequent time-course of plasma DDAVP differed however, from patient to patient, and was irregular in most of them. In all of the patients whose plasma DDAVP dropped below 1.0 pg/m/, antidiuresis was still observed. Although the mean basal level of nephrogenous cyclic AMP in patients with diabetes insipidus was not significantly different from that in control subjects, the administration of DDAVP resulted in a 2-fold increase. A negative correlation between nephrogenous cyclic AMP and free water clearance was obtained. These results suggest that the longacting nature of DDAVP may be attributed, in addition to its gradual absorption from nasal mucosa and slow metabolic clearance, to a higher or persistent biological activity at the receptor site in the kidney and that a nearly physiological level of antidiuretic hormone may cause de novo synthesis of cyclic AMP in the kidney and exert its biological action. A vasopressin analogue, 1-desamino-8- D-arginine vasopression (DDAVP) has been available for the treatment of diabetes insipidus for several years. Previous studies have shown that intranasal administration of this analogue possesses several advantages over other types of therapies (Vavra et al., 1969 ; Anderson and Arner, 1972 ; Edwards et al., 1973 ; Robinson, 1976 ; Seif et al., 1978). These include its convenience, a longer duration of action and less pressor activity (Anderson and Arner, 1972) com- Received December 3, Reprint requests to : Shigehiro Katayama, M. D., The Third Department of Internal Medicine, Faculty of Medicine, University of Tokyo, T-3-1, Hongo, Bunkyo-ku, Tokyo, Japan. pared with any other commercial preparation of vasopressin. Of particular interest is its prolonged effect. A single intranasal administration of DDAVP has been reported to produce an antidiuresis lasting up to 20 hr in patients with diabetes insipidus (Anderson and Arner, 1972 ; Robinson, 1976 ; Seif et al., 1978). The present study was undertaken to clarify possible factors responsible for the prologed effect of DDAVP by relating the plasma levels of DDAVP to antidiuresis, and by determining the changes in urinary as well as plasma cyclic AMP after DDAVP administration.

2 364 KATAYAMA et al. Table 1. Clinical data on patients with diabetes insipidus. Endocrinol. June 1980 Japon. D, PT and C indicate hydrochlorothiazide, Pitressin tannate in oil and Chlorpropamide, respectively. The test was performed by restricting fluid intake until the patient lost 3% of body weight. These two patients were diagnosed as partial diabetes insipidus by the criteria of Miller et al. Materials and Methods Protocol Ten patients with central diabetes insipidus, aged between 26 and 67 years, were studied. The diagnosis was established by water deprivation and/or a hypertonic saline infusion test with the determination of plasma immunoreactive arginine vasopressin (AVP) levels. Their clinical data are summarized in Table 1. In patients with diabetes insipidus due to ectopic pinealoma or after hypophysectomy for pituitary tumor, an adequate dose of 1-thyroxine as well as adrenal glucocorticoid was being replaced. To institute therapy with DDAVP, five patients were withdrawn from previous medications and allowed to establish polyuria and polydipsia. The other patients had never been treated for diabetes insipidus. Informed consent was obtained from all participants. The experiment was conducted in the morning and the stable polyuria was used as a base-line to determine the response to medication. The patients were given oral fluids in a volume equivalent to that of the urine passed throughout the study. After base-line blood and urine samples were collected, 10 ƒêg of DDAVP was administered intranasally with a rhinyle catheter. Blood samples were collected into heparinized syringes at 30, 45 and 60 min, then every half-hour up to 4 hr and after that every hr until 6 hr. When plasma cyclic AMP was assayed, two ml of heparinized blood was transferred to a chilled glass tube containing 10 mmol of EDTA. All blood samples were immediately centrifuged at 4 Ž. Plasma was separated and stored at -20 Ž until analyzed. Urine samples were collected every half-hour for 4 hr and then every hr until 6 or 8 hr. For urinary cyclic AMP determination, two ml of Fig. 1. A dose-response curve of DDAVP in the present radioimmunoassay. urine was collected in a chilled glass tube with 10 mmol of EDTA and frozen for later assay. In the evaluation of plasma and urinary levels of cyclic AMP, ten age-matched subjects served as controls. Their one-hour urine samples and midpoint blood specimens were collected as described above. Osmolality was measured by freezing-point depression on an Advanced Instruments Osmometer. Free water clearance (CH2O, ml/min) was calculated according to the conventional equation. Plasma and urinary creatinine was determined by a modification (Borsnes and Taussky, 1945) of the method of Folin and Wu. Serum calcium and inorganic phosphorus were measured with an Technicon Smac autoanalyzer. DDAVP assay Plasma DDAVP levels were determined by a radioimmunoassay using an antiserum raised against lysine vasopressin and 125I-labelled AVP. Detail of the procedure have been reported previously (Shima-

3 Vol. 27, No. 3 DDAVP, ANTIDIURESIS AND CYCLIC AMP 365 where UcA, V and Ccr represent urinary cyclic AMP concentration (nmol/ml), urine volume (ml/min) and creatinine clearance (ml/min), respectively. Nephrogenous cyclic AMP (Broadus et al., 1970) was expressed as a function of GFR, as given by the formula where the symbols except PcA which represents plasma cyclic AMP concentration (pmol/ml) correspond to those employed above. Unpaired Student's t test was used for statistical analysis and I near regression lines were obtained by the method of least squares. Results Fig. 2. Changes of urine volume and urine osmolality after an intranasal administration of 10 Đg of DDAVP in 10 patients with diabetes insipidus. An arrow indicates the time of administration. moto et al., 1976). A dose-response curve was obtained between 0.2 and 50 pg of DDAVP (Fig. 1). The minimum detectable concentration was 1.0 pg/ ml, since the labelled AVP at this level was significantly different from zero in each assay. Recovery of the acetone extraction method was 70.5 }8.4% (mean }SD, n=7). The coefficient of variation averaged 13.7% for interassay and 15.2% for interassay. All samples from a single patient were run in one assay to avoid any interassay variation. Plasma and urinary cyclic AMP were determined by a specific radioimmunoassay without extraction (Honma et al., 1977). Urinary cyclic AMP was expressed as: (a) a rate (nmol/min), (b) a function of urinary creatinine (nmol/mg creatinine) and (c) a function of glomerular filtration rate (GFR), as given by All patients responded to DDAVP administration and no major side effect was observed except urticaria in one patient. In Fig. 2 are shown the changes in urine volume and osmolality. Urine volume decreased within the first hr of administration and this effect of DDAVP lasted for at least 6 hr. In all patients, urine osmolality exceeded that of plasma by 60 min. Concentrated urine was excreted. in 8 patients during the period of observation, while urine osmolality decreased after 5 or 6 hr in the remaining two patients. As shown in Table 2, free water clearance, which was determined in five patients, decreased significantly at 60 min and remained negative until 6 hr. Plasma DDAVP levels were determined in eight of ten patients and the results are shown in Fig. 3. In the control samples, no immunoreactive AVP was detectable in plasma, in any of the patients. Thirty min after an intranasal spray, plasma concentrations of DDAVP were uniformly elevated. Its peak ranging from 5.6 to 25.0 pg/ml appeared at 30 to 120 min. However, both the temporal pattern of circulating DDAVP and the levels showed great individual variations. In some patients, several peaks. of plasma DDAVP were observed. A semilogarithmic plot of plasma DDAVP levels against time revealed a biexponential curve

4 in only two patients. The half-lives of plasma DDAVP were 51 and 74 min for the first, and 102 and 86 min for the second exponential in these patients. In other subjects, the calculation of half-lives of plasma DDAVP was impossible, because of its irregular time-course. In one patient, plasma DDAVP levels decreased to less than 1.0 pg/ml at 3.5 hr, however, the effect persisted for another 90 min. Six hr after the administration, five of the patients showed plasma DDAVP levels below 1.0 pg/ml, although the effect was still observed in all but one patient described above. In Table 2 are shown the change in plasma and urinary cyclic AMP after the administration of DDAVP. The basal levels of urinary as well as plasma cyclic AMP were not significantly different from those in normal subjects. After DDAVP administration, urinary cyclic AMP excretion tended to rise and reached significantly higher levels at 150 min (p<0.01). The Fig. 3. Plasma DDAVP levels after an intranasal administration of 10 Đg of DDAVP in patients with diabetes insipidus. An arrow indicates the time of administration. peak levels in individual patients ranged from 110 to 290% (164 }19, mean }SEM) of the basal level. Similarly, the mean Table 2. Changes of urine volume, urine osmolality, free water AMP and creatinine clearance (mean }SEM) after an

5 Vol. 27, No. 3 DDAVP, ANTIDIURESIS AND CYCLIC AMP 367 urinary cyclic AMP corrected by creatinine excretion was significantly elevated (p< 0.05). The peak varied from 120 to 350% (174+24, mean+sem) of the baseline. Plasma cyclic AMP levels, on the other hand, did not show any significant change during the study. As a result, nephrogenous cyclic AMP gradually increased and reached significantly higher levels between 90 and 180 min. When the relationship between nephrogenous cyclic AMP and urine osmolality was examined, a significant positive correlation (r=0.349, n=55, p< 0.01) was obtained. As shown in Fig. 4, a significant negative correlation was also obtained between nephrogenous cyclic AMP and free water clearance. Serum calcium and inorganic phosphorous concentrations were within the normal range in all of the patients as well as control subjects. Fig. 4. Relationship between nephrogenous cyclic AMP and free water clearance. clearance, urinary and plasma cyclic AMP, nephrogenous cyclic intranasal administration of 10 Đg of DDAVP.

6 368 KATAYAMA et al. Endocrinol. June 1980 Japon. Discussion The results of the present study are consistent with earlier observations that a single intranasal administration of DDAVP resulted in a prompt and persistent antidiuresis in patients with diabetes insipidus (Vavra et al., 1969; Anderson and Amer, 1972; Edwards et al., 1973; Robinson, 1976; Seif et al., 1978). The effect appeared within the first hr and lasted for at least 6 hr in all of the patients studied. Plasma DDAVP levels were uniformly elevated within 30 min after the administration, which is in accordance with its rapid action. The subsequent timecourse of plasma DDAVP levels was, however, different in each patient. In most subjects, semi-logarithmic plots of plasma DDAVP against time had no rectilinear portion. Moreover, multiple peaks of plasma DDAVP were observed in some of them. Along with the finding that peak plasma DDAVP concentration showed wide individual variation, these results suggest that one of the major factors responsible for a longer action of DDAVP is its gradual absorption from the nasal mucosa. This assumption was also proposed by Seif et al based on the positive correlation in individual patients between the time to reach the peak of plasma DDAVP and its apparent half-life, although the full data on changes in plasma DDAVP were not shown in their report. In two patients, plasma DDAVP was cleared from the circulation showing a biexponential curve with a mean half-life of 63 min for the first and 94 min for the second exponential. The biological meaning of the biexponential curve of plasma DDAVP remains unknown. Since the recognition site of the antiserum used in the present radioimmunoassay resides in the ring-portion side of vasopression molecule (Shimamoto et al., 1976), it may be unlikely that C-terminal fragments (Walter and Shlank, 1975) in the circulation were measured. The second exponential curve may result from N-terminal fragments released during the inactivation process, or alternatively, a slow rate of absorption from nasal mucosa contributing to the second exponential. Whatever it is, these values are close to the 51 to 117 min obtained by Edwards et al after an intravenous injection of DDAVP to patients with diabetes insipidus and much longer than those reported for lysine vasopressin (Johnston, 1972) or AVP (Robertson et al., 1973; Beardwell, 1971; Skowsky et al., 1974). The apparent plasma half-life of DDAVP ranging from 0.4 to 4 hr reported by Seif et al may represent a combination of absorption and degradation in plasma. In five of eight patients whose plasma DDAVP concentrations were estimated, the plasma DDAVP levels were less than 1.0 pg/ml at 6 hr after administration. Antidiuresis, however, was still observed. In both normal subjects and patients with diabetes insipidus, maximum diuresis usually occurs when their plasma AVP concentrations are around 1.0 pg/ml (Shimamoto et al., 1976; Robertson et al., 1973; Husain et al., 1973). The long-acting nature of DDAVP may be derived from higher or persistent biological activity at the receptor level in the kidney as well as its gradual absorption from nasal mucosa and its slow metabolic clearance. This concept is supported by an in vitro study (Seif et al., 1978) showing a greater cyclic AMP generation with DDAVP in the renal outer medullary tissue, although different results were obtained by other investigators (Roy et al., 1975; Dousa et al., 1971). Of interest is one patient whose plasma DDAVP levels fell below the limit of detection after 3.5 hr. The possibility that was a simply the result of experimental error was excluded by repeating the study. Because the peak level of plasma DDAVP in the patient was similar as in other patients, the rapid

7 Vol.27,No.3 DDAVP, ANTIDIURESIS AND CYCLIC AMP 369 clearance of DDAVP might be due to an accelerated inactivation process. A fairly wide range of DDAVP doses required for the treatment of diabetes insipidus regardless of its severity (Anderson and Arner, 1972 ; Robinson, 1976) may support the view that there exists a considerable individual variation in absorption as well as in inactivation of DDAVP. Although it is now well established that the action of antidiuretic hormone is mediated by the activation of adenylate cyclase system in the renal medulla (Roy et al., 1975 ; Dousa et al., 1971), the effect of vasopression on urinary cyclic AMP excretion has been not consistently demonstrated. Infusion of pharmacological doses of Pitressin (10 mu/min for an hr or two to three hr) resulted in an increase in urinary cyclic AMP in both water-loaded healthy subjects and in patients with diabetes insipidus (Taylor et al., 1970 ; Fichman and Brooker, 1972). An intramuscular injection of synthetic lysine vasopressin at a dose of 0.1 pressor units/kg was also effective in raising the urinary cyclic AMP levels (Takahashi et al., 1966). Similar maneuvers, however, failed to result in an increase in cyclic AMP excretion in other studies (Williams et al., 1972 ; Kaminsky et al., 1970). Various or even conflicting results were obtained also in rats (Bulten and Jard, 1972 ; Chase and Aurbach, 1967). The discrepancy between these studies is hardly explained. The present study clearly demonstrated that a vasopressin analogue at nearly physiological levels monitored by the plasma concentration produces a significant rise in urinary cyclic AMP excretion rate. The increase was significant when the cyclic AMP excretion was corrected by urinary creatinine concentrations. Moreover, nephrogenous cyclic AMP, an index which is assumed to reflect parathyroid activity (Broadus et al., 1977), was significantly elevated with a two-fold increase over the control levels. That nephrogenow cyclic AMP is closely related to the biological action of DDAVP was supported by a negative correlation between nephrogenous cyclic AMP and free water clearance in each subject. These results indicate that a nearly physiological level of vasopressin may cause de novo synthesis of cyclic AMP in the kidney and exert its biological activity. If urinary cyclic AMP is under the influence of antidiuretic hormone, a difference between nephrogenous cyclic AMP in patients with diabetes insipidus and that in normal control subjects would be expected. This was not the case in the present study. The mean base-line nephrogenous cyclic AMP in patients was lower than in control subjects but the difference was not significant. This may be explained by the fact that several factors except vasopressin, a major one of which is parathyroid hormone (Broadus et al., 1970), are involved in the regulation of urinary cyclic AMP excretion and that vasopressin-induced elevation of nephrogenous cyclic AMP is modest compared with that induced by parathyroid hormone (Williams et al., 1972 ; Kaminsky et al., 1970 ; Bulten and Jard, 1972 ; Chase and Aurbach, 1967 ; Broadus et al., 1977) In any event, a small portion of nephrogenous cyclic AMP which was not suppressed by calcium infusion in normal subjects (Broadus et al., 1978) might be explained by vasopressin-dependent nephrogenous cyclic AMP. Acknowledgement The authors are indebted to Dr. K. Kosaka for his encouragement throughout this study.

8 370 KATAYAMA et al. Endocrinol. June 1980 Japon. References Anderson, K. E. and B. Arner (1972). Acta Med. Scand. 192, 21. Beardwell, C. G.(1971). J. Clin. Endocrinol. Metab. 33, 254. Borsnes, R. W. and H. H. Taussky (1945). J. Biol. Chem. 158, 581. Broadus, A. E., N. I. Kaminsky, J. G. Hardman, E. A. Sutherland and G. W. Liddle (1970). J. Clin. Invest. 49, Broadus, A. E., J. E. Mahaffey, F. C. Bartter and R. M. Neer (1977). Ibid. 60, 771. Broadus, A. E., L. J. Deftos and F. C. Bartter (1978). J. Clin. Endocrinol. Metab. 46, 477. Bulten, D. and S. Jard (1972). Pflugers Arch. Eur. J. Physiol. 331, 172. Chase, L. R. and G. D. Aurbach (1967). Proc. Natl. Acad. Sci. USA 58, 518. Dousa, T., 0. Hechter, I. L. Schwartz and R. Walter (1971). Ibid. 68, Edwards, C. R. W., M. J. Kitau, T. Chard and G. M. Besser (1973). Br. Med. J. 3, 375. Fichman, M. P. and G. Brooker (1972). J. Clin. Endocrinol. Metab. 35, 35. Honma, M., T. Satoh, J. Takezawa and M. Ui (1977). Biochem. Med. 18, 257. Husain, M. K., N. Fernando, M. Shapiro, A. Kagen and S. M. Glick (1973). J. Clin. Endocrinol. Metab. 37, 616. Johnston, C. I. (1972). J. Endocrinol. 52, 69. Kaminsky, N. I., J. H. Ball, A. E. Broadus, J. G. Hardman, E. W. Sutherland and G. W. Liddle (1970). Trans. Assoc. Am. Physicians 83, 235. Miller, M., T. Daakos, A. M. Moses, H. Fellerman and D. H. T. Streeten (1970). Ann. Intern. Med. 73, 721. Robertson, G. L., E. M. Mahr, S. Athar and T. Sinha (1973). J. Clin. Invest. 52, Robinson, A. G. (1976). N. Engl. J. Med. 294, 507. Roy, C., T. Barth and S. Jard (1975). J. Biol. Chem. 250, Seif, S. M., T. V. Zenser, F. F. Ciarochi, B. B. Davis and A. G. Robinson (1978). J. Clin. Endocrinol. Metab. 46, 338. Shimamoto, K., T. Murase and T. Yamaji (1976). J. Lab. Clin. Med. 87, 338. Skowsky, W. R., A. A. Rosenbloom and D. A. Fisher (1974). J. Clin. Endocrinol. Med. 38, 278. Takahashi, K., M. Kamimura, T. Shinko and S. Tsuji (1966). Lancet 2, 967. Taylor, A. L., B. B. Davis, L. G. Pawlson, J. B. Josimovich and D. H. Mintz (1970). J. Clin. Endocrinol. Metab. 30, 316. Vavra, I., A. Machova, V. Holeeek, J. H. Cort, M. Zaoral and F. Sorm (1969). Lancet 1, 948. Walter, R. and H. Sclank (1975). J. Clin. Endocrinol. Metab. 96, 811. Williams, R. H., J. Barish and J. W. Ensinck (1972). Proc. Soc. Exp. Biol. Med. 139, 447.