Vasopressin by lmmunoassay - Europe PMC

1 downloads 0 Views 444KB Size Report
of chlorpropamide on the renal threshold of responsiveness to minimal doseinfusions of vasopressin in patients with diabetes insipidus, and have shown that ...
842 Proc. roy. Soc. Med. Volume 64 August 1971

32

1 mg) were given at 08.00 hours on the morning of Discussion recording on nine occasions. These data support the observation of a morning peak of plasma testosterone and suggest an Results association of individual fluctuations in testo(1) There was considerable variation of plasma sterone level with periods of REM sleep. The testosterone levels during sleep (Fig 1). A decline function of testosterone fluctuations is not yet was usual with sleep onset, and subsequently known. It is possible that this change will relate fluctuations in testosterone level produced an to the penile erections and sexual fantasy of overall increase and culminated in the morning dreams during REM sleep. Alternatively it is peak. suggested that testosterone has an anabolic role in the physiological changes of REM sleep which (2) Fluctuations were found to relate to sleep in involves brain protein synthesis (Oswald 1969). three different ways: (a) A rise occurred at the Further studies have been designed to test these onset of REM sleep periods. (b) Frequently the hypotheses. rise preceded the periods of REM sleep. (c) OccaREFERENCES sionally periods of arousal during the night were Dray F, Reinberg A& Sebacum J associated with a rise in testosterone level. (1965) C.R. Acad. Sci. (Paris) 261, 573

(3) Sleep reversal: In one subject, fluctuations in plasma testosterone (Fig 2) accompanied sleep in the day on three occasions. The usual diurnal trend was therefore reversed. (4) While soluble ACTH had no effect on sleep or testosterone profiles when given at 08.00 hours, Synacthen 1 mg severely depressed REM sleep and promoted wakefulness during the night. Fluctuations in plasma testosterone were minimal and there was no overall rise.

(5) At a late point in the experiment, periods of intensive sampling (interval less than 5 min) showed marked fluctuations in plasma testosterone forming a series of peaks. Thirty-minute sampling was found to exert an overall 'smoothing' function on the testosterone profile. 1000 I

1100 1200 I

EATNEIER HI STARE AWAKE REM SLEEP

1300

1100 1500 1600 1700 1000 I

FINISH AA

iii] 3G1001l 400z

PLASMA 301 17 ONCS 20 A /000m1 101

700000

RONE TESTOSTE"OE

Fig 2 Sleep during the day. The increase in plasma testosterone reverses the normal diurnal trend

by C R W Edwards MB MRCP (The Medical Unit, St Bartholomew's Hospital, London ECI)

substances producing nonspecific inhibition of antigen-antibody binding (Robertson et al. 1970), these immunoassays have not yet been generally applied to the measurement of plasma and urinary vasopressin in man. The development of a satisfactory extraction procedure has overcome some of these problems (Edwards et al. 1970, Ratcliffe & Edwards 1970).

000

ug,uomi 100: 300200' U 100A-AAA A'AAA IA I 0AA I 1 1000 1100 1200 1300 1100 1500 100 1700 TIME

Measurement of Plasma and Urinary Vasopressin by lmmunoassay

Four other groups have described the development of radioimmunoassays for vasopressin (Klein et al. 1966, Permutt et al. 1966, Beardwell & Wright 1968, Miller & Moses 1969). However, because of the low circulating levels of this hormone, and the presence in plasma and urine of

'I-I

PLASMA

Evans J I, Lewis S A, Gibb I A M & Cheetham M (1968) Brit. med. J. iv, 291 Ismail A A A & Love D (1970) Res. Steroids 4 Nieschlag E & Ismail A A A (1970) Klin. Wschr. 48, 53 Oswald I (1962) Proc. roy. Soc. Med. 55, 910 (1969) Nature (Lond.) 223, 893 Resko J A & Eik-nes K B (1966) J. cdin. Endocr. 26, 573 Van Kirk K & Sassin J F (1969) Amer. J. EEG Technol. 9, 143

1"0

Radioimmunoassay of Vasopressin Three antisera specific to different parts of the vasopressin molecule have been used in these

1800xi

minimal detectable level of the immunoassay is 10 pg/ml plasma and it is not yet possible to measure levels of circulating vasopressin in basal or dehydrated subjects.

700 600 B IOSSAY

PU/ml

843

Section ofEndocrinology

33

500

400 300

200

100_./ 100 200 300 400 500 600 700 IMMUNOASSAY pU/ ml

1800

Fig 1 The correlation between bioactive and immunoreactive plasma vasopressin levels in 16 blood samples obtained during hamorrhage experiments in sheep; r =0-98, PK< 001; 1 iu =2-5 pg argininevasopressin

studies. One antiserum was produced by immunization with a mixture of arginine-vasopressin (AVP) and lysine-vasopressin (LVP) conjugated to protein. This antiserum reacts identically with AVP and LVP, but shows virtually no crossreactivity with oxytocin. Two antisera were produced by immunization with non-conjugated AVP. These antisera react specifically with AVP. Lysine-vasopressin has only 0 05 % and 0-06 % of the activity of AVP using these two antisera. Advantage can be taken of these differences in specificity in measuring one peptide analogue in the presence of another, and also in the identification of unknown analogues. We have, for example, measured the AVP and LVP content of two batches of Pitressin, the commercial preparation used for the treatment of diabetes inisipidus. This is prepared from bovine posterior pituitaries which contain AVP and porcine posterior pituitaries containing LVP. Of the two batches analysed one contained 55 % AVP and 45 % LVP, and the other 20 % AVP and 80 % LVP. The iodination of vasopressin and the extraction procedure have been described elsewhere (Edwards et al. 1970, Ratcliffe & Edwards 1970). Blood samples were taken into cooled heparinized syringes, and the plasma separated immediately in a refrigerated centrifuge. IN HCI 1 ml was added to each 5 ml plasma sample and the samples stored at -20°C before assay. Urine samples are acidified and stored in the same way.

(1) Physiological stimuli for the release of vasopressin: Serial samples from a feetal sheep subjected to hiemorrhage four days before term were assayed for vasopressin by immunoassay, and by bioassay by Dr Mary Forsling using the ethanol anesthetized rat (Forsling et al. 1968). Fig 1 shows the correlation between the results of the two assays. The calculated correlation coefficient was 0-98 on sixteen samples. The regression line intercepted the immunoassay axis at 16 ,uu/ml (40 pg/ml). The nature of this immunoreactive but non-biologically active material is under investigation.

Another physiological situation in which we have studied vasopressin release is parturition. It has been shown by Hoppenstein et al. (1968) using bioassay that cord venous blood contains a high level of vasopressin. However, these results did not indicate whether the source was the placenta or the foetus itself. In 15 out of 17 cases in which we obtained paired arterial and venous cord blood samples, the arterial vasopressin level was higher than the venous, indicating that the vasopressin had been released by the feetal pituitary. The cord blood vasopressin levels were higher following spontaneous vaginal delivery than during elective Ciesarean section or Cesarean section in labour. The highest levels of vasopressin were found in cord blood samples taken after forceps delivery for foetal distress or delay in the second stage of labour. In this group the mean arterial level was 2,600 pg/ml. This extremely high level may well have a significant effect on placental blood flow. 100 90 80 \

70

% MAX. VALUE 60 50 40 30 20 _ 10

MVeasurement of Vasopressin in Plasma To validate the immunoassay and extraction procedure, the physiological and pharmacological release of vasopressin has been studied. Bioassay results suggest that the basal circulating level of AVP is 1-5 pg/ml plasma. The present

- On Chlorpropamide >< On Placebo 0

" -

.

0

TIME (MINUTES ) Fig 2 Measurement ofplasma immunoreactive vasopressin following infusion of lysine-vasopressin in a patient with diabetes insipidus on chlorpropamide andplacebo

.a

844 Proc. roy. Soc. Med. Volhme 64 August 1971

(2) Pharmacological release of vasopressin: Blood samples were taken following the intravenous administration of 0-25 mg and 1-0 mg nicotine base. The subject was hydrated and the samples taken from a cannula in the right internal jugular vein, in which the concentration of pituitary peptides would be expected to be considerably higher than in the general circulation. After the lower dose of nicotine the jugular vein level of vasopressin rose from 46 pg/ml to 158 pg/ml but there was no antidiuresis. Following the higher dose the jugular level of vasopressin rose to 238 pg/ml and there was a prolonged antidiuresis.

(3) Measurement of plasma vasopressin levels following vasopressin infusion: Chlorpropamide is an effective form of therapy in many cases of diabetes insipidus (Arduino et al. 1966). It has been shown in rats with congenital diabetes insipidus that the effect of chlorpropamide is to potentiate the action of small doses of vasopressin (Miller & Moses 1970, Berndt et al. 1970). It seemed possible that chlorpropamide might act by prolonging the half life of circulating vasopressin. This possibility was investigated by the measurement of plasma vasopressin levels following an infusion of LVP in patients with diabetes insipidus pretreated with either chlorpropamide (350 mg daily) or placebo tablets. Lysine-vasopressin was infused at a rate of 6 ,ig/hour for one hour and samples taken at intervals after stopping the infusion. Fig 2 shows the results of one of these studies, and indicates that chlorpropamide does not alter the half life of infused vasopressin. Other studies in progress have measured the effect of chlorpropamide on the renal threshold of responsiveness to minimal dose infusions of vasopressin in patients with diabetes insipidus, and have shown that chlorpropamide lowers the renal threshold (Edwards & Besser 1971). Measurement of Urinary Vasopressin The extraction can be used for the measurement of vasopressin in urine, and has also been applied to the measurement of urinary oxytocin (Boyd & Chard 1971). An extraction procedure is still necessary because high levels of salt and urea can produce nonspecific inhibition in the assay (Beardwell & Wright 1968). We have found that the measurement of urinary vasopressin during a water load test is of value in confirming a diagnosis of the syndrome of inappropriate antidiuretic hormone production (SIADH). Normally during a water load test urinary vasopressin is suppressed to undetectable levels. Fig 3 shows the measurement of urinary

34

URINE OSMOLALITY

300

600

1 Litre

500

-250

400

(mOsmlkg 300

100

: : : :.

1

-200 ::::: :.:::::, :..:

URINE ARG-

VASOPRESSIN

..50 .pgmlI)

_

, .. ,, ,., , '. , , ', ,. _ 50

:.:x No Urine ;:::. ::.: *::x.... _. ._._ Passed __.. 5 4 3 1 2 COLLECTION PERIODS/HOURS ) EJOSMOLALITY 11J ARG-VASOPRESSIN

Fig 3 The cxci etion ofurinary immunoreactive vasopressin following a one-litre oral wdter load int patient with oat-cell carcinoma ofbronchus and hyponatrwgmia

vasopressin in a patient with an oat-cell carcinoma of the bronchus and hyponatrsmia. There was no suppression of urinary vasopressin following the water load, thus confirming the diagnosis of SIADH. Conclusion

The use of radioimmunoassay for vasopressin, coupled with a simple extraction procedure for urine and plasma, provides a valuable tool for the investigation of physiological and pathological variations of this hormone. The results correlate well with those obtained by classical bioassays, while the procedure has a number of advantages over these in convenience and specificity. REFERENCES Arduino F, Ferraz F P J & Rodrigues J (1966) J. clini. Endocr. 26, 1325 Beardwell C G & Wright A D (1968) Excerpta med. (Amst.) lint. Congr. Ser. 157, p 48 Berndt W 0, Miller M, Kettyle W M & Valtin H (1970) Endocrinology 86, 1028 Boyd N R H & Chard T (1971) In: Radioinmrunoassay Methods. Europeans workshoj). Ed. K E Kirkham & W M Hunter. Edinburgh; p 512 Edwards CR W & Besser GM (1971) (in preparation) Edwards C R W, Chard T, Kitau M J & Forsling M L (1970) J. Endocr. 48, xi Forsling M L, Jones J J & Lee J (1 96&)J. Physiol. (Lond.) 196, 495 Hoppenstein J M, Miltenherger F W & Moran W\ H (1968) Surg. Gynec. Obstet. 127, 966 Klein L A, Roth J & Petersen M J (1966) Surg. Forum 17, 240 Miller M & Moses AM (1969) Endocrinology 84,'557 (1970) Endocrinology 86,"1024 Permutt M A, Parker C W & Utiger RD (1966) Endocrinology 78,U809 Ratcliffe J G & Edwards C R W (1971) In: Radioimmunoassay Mdethods. European Workshop. Ed. K E Kirkham & W M Hunter. Edinburgh.; p 502 Robertson G L, Klein L A, Roth J & Gorden P (1970) Proc. not. Acad. Sci. (Wash.) 66, 1293