Increased motor disturbances in response to arginine vasopressin following hemorrhage or hypertonic saline: Evidence for central AVP release in rats

Brain Research, 273 (1983) 59--65 Elsevier

59

Increased Motor Disturbances in Response to Arginine Vasopressin Following Hemorrhage or Hypertonic Saline: Evidence for Central AVP Release in Rats D. M. BURNARD l, Q. J. PITTMAN2 and W. L. VEALE 1

Departments of Medical Physiology 1, and Pharmacology and Therapeutics2, Faculty of Medicine, The University of Calgary, Calgary, Alberta T2N 4N1 (Canada) (Accepted January 4th, 1983)

Key words: arginine vasopressin - - convulsions - - hemorrhage - - neuropeptides

The effects of hemorrhage and parenteral hypertonic saline on the behavioural responses to centrally-administered arginine vasopressin (AVP) were examined in rats. Both hemorrhage and hypertonic saline act as potent stimuli for neurohypophysial vasopressin release, and may serve as potential stimuli for cerebral AVP release. When administered into a lateral cerebral ventricle of the rat brain, AVP has a potent convulsant action; this effect increases in severity upon subsequent administration. Removal of 15% of the estimated blood volume from the conscious rat or infusion of 1.0 ml of 1.5 M sodium chloride solution into the peritoneal cavity can mimic the effect of a central injection of AVP in 'sensitizing' the brain to the behavioural effects of subsequent injections of AVP. This suggests that these stimuli which are known to activate posterior pituitary secretion of AVP also induce the release of AVP (or a closely related molecule), from neuronal fibres within the brain. INTRODUCTION Arginine vasopressin (AVP) has a potent convulsive action when injected into a lateral cerebral ventricle of the ratl,12a 4. This p h e n o m e n o n displays a type of sensitization, whereby the first exposure to 1.0/zg of A V P increases the likelihood of a convulsion on subsequent treatment and decreases the threshold dose necessary to cause a convulsion12. This convulsive activity in the brain, along with the many other reported behavioural and physiological effects 8,16, supports the suggestion that vasopressin may play a role as a neurotransmitter or neuromodulator within the central nervous system (CNS). In further support of such a role are immunohistochemical, radioimmunoassay, and axonal transport studies localizing A V P in extensive fibre systems throughout much of the brain, extending well beyond the hypothalamo-neurohypophysial system (cf. refs. 4, 19). A V P has been well studied as a h o r m o n e involved in water balance and blood pressure regulation18, whereby cells situated in the paraventricular, supraoptic and other hypothalamic nuclei respond to 0006-8993/83/$03.00 (~) 1983 Elsevier Science Publishers B.V.

osmotic, chemoreceptor and baroreceptor inputs and release A V P into the pituitary circulation 17. In view of the proximity (and possible coexistence)20 of the cells giving rise to the neurohypophysial projections and those forming the basis for the cerebral networks, the idea has been raised that the two systems may function in a related manner16. The experiments reported here were designed to investigate the possibility that potent stimuli for neurohypophysial A V P release, such as hemorrhage and hypetonic saline 18, may also serve as stimuli for the release of cerebral AVP. Such a stimulus-induced release of endogenous A V P could then mimic the effect of an intracerebroventricular (i.c.v.) injection of this peptide in sensitizing the rat brain to the convulsant action of AVP. Since the release of A V P is thought to be controlled by anatomically separate pathways responsive to either osmotic or non-osmotic stimuli~8, the experiments reported here examined the effects of both types of stimuli. Preliminary results from this work have been reported elsewhereS,6.

6(I MATERIALS AND METHODS Male Long Evans rats (250-350 g) (n = 31) were implanted stereotaxically with bilateral stainless steel guide cannulae directed towards the lateral cerebral ventricles. Each of the 16 rats to be used for the hemorrhage study was also implanted with a chronic jugular cannula (Silastic tubing; Dow Corning) positioned directly above the right atrium of the heart and exteriorized to the top of the skull. When not in use, the jugular cannula was filled with heparinized saline and occluded with a metal pin. Bipolar depth electrodes were also implanted stereotaxically into the cortex and both sides of the dorsal hippocampus of 2 of the 16 rats undergoing hemorrhage. E E G was recorded and displayed on a Beckman polygraph R611. Rats were kept on a 12 h light-12 h dark cycle and housed in separate cages with free access to food and water for the duration of the experiments. During the 7-10 days allowed for recovery from surgery, the chronic jugular cannulae were flushed daily with sterile physiological saline and refilled with sterile heparinized saline. Experiments were conducted between 13.00 and 18.00 h. All i.c.v, injections were given into awake, behaving animals and were accomplished by gravitational flow of 5.0/~1 of hormone solution or vehicle (sterile, pyrogen-free, physiological saline) through a 27-gauge injection cannula. Synthetic AVP (Bachem) was made up fresh for each day's experiment. During the experiments, rats were placed in a plexiglass observation chamber and observed continuously for a 10 min period before and after AVP administration. Behavioural responses were recorded and scored according to a predetermined behavioural code, similar to that used by other authorsT, 9. In our laboratory we have confirmed by a double-blind study that this code consistently defines the behavioural responses to i.c.v, AVP, and that the responses to the vehicle can be distinguished readily from those to the peptide (unpublished observations). Since E E G does not always correlate well with behavioural events (unpublished results) this behavioural code represents a more reliable, repeatable and quantifiable means of representing the experimental results. The code was as follows: 0 = no effect; 1 = pauses of 5-10 s duration or more, asso-

crated with absence of activity and staring; 2 -sprawled out posture with hind leg extension and locomotor difficulties; 3 = discrete myocionic jerks, often followed by exaggerated scratching; and 4 = motor disturbances, including barrel rotations (spinning along the long axis of the body) and/or full myoclonic-myotonic convulsions. Results from the present work are expressed as means of these behavioural scores, although there is no evidence for a truly linear progression through the symptoms as described above. The individual behaviourai scores from the various experimental groups were statistically analyzed by the non-parametric Mann-Whitney U-test, and Wilcoxon matched pairs signed-ranks test. On day 1 naive rats were subjected to one of the following treatments: (1) a 4.0 ml hemorrhage (approximately 15% of the blood volume) over a 20 min period, after which red blood cells were resuspended in sterile physiological saline to the original volume and reinfused (n = 16); (2) an intraperitoneal injection of 1.0 ml of 1.5 M hypertonic saline (n = 5); or (3) 1.0#g of AVP i.c.v, in 5.0/A of saline (n = 10). Two days later rats received either: 1.0/~g of AVP i.c.v, in 5.0/~1 saline, 5.0/A sterile physiological saline i.c.v. (vehicle) followed 2 h later by 1.0/~g of AVP i.c.v, in 5.0 ~! saline, or another hemorrhage over a 20 min period (n = 3), as described above. The latter group was again hemorrhaged on the fifth day, followed by 1.0~g AVP i.c.v. 3 days later. Injection sites were verified upon completion of these experiments by the injection of 5.0/~1 of Evans Blue dye solution into the lateral ventricle, after which the brain was examined for the presence of dye within the ventricular system. In all animals the dye was present within the cerebral ventricles and no dye was present along the guide cannula or on the cortical surface of the brain. RESULTS Following a hemorrhage of 15% of the blood volume, rats exhibited minor behavioural disturbances rating a score of 1.0 (Fig. 1). That is, all rats exhibited a behavioural pause associated with immobility and staring. This behaviour is also seen following a first injection of 1.0/~g AVP i.c.v. (Fig. 1). The behavioural effects were no longer observed once the original blood volume had been re-established. When rats

61 HEMORRHAGE COMPARED WITH CENTRAL ADMINISTRATION OF AVP

i:!:!:~:!:!:!:!:~:~:~:!:~:i:i:~¢ ::::::::::::::::::::::::::::::::::: i:i!!ili~i!i!i~ili!i!i!i~i~i~i~i~il

DAY 1 1.0pg AVP

DAY 1 Hemorrhage

N=IO DAY 3 1.0IJg AVP

iiiiiiiiiiiiiiiiiiiiiiiiiiiiiii DAY 3 1.0Bg AVP

Fig. 1. The bars represent the mean behavioural scores (+ S.E.M.) of 2 groups of rats in response to either AVP i.c.v, on day 1 and 3 (open bars) or a 4.0 ml hemorrhage on day 1 followed by i.c.v. AVP on day 3 (stippled bars). A low score represents a mild effect and a high score indicates convulsive behaviour (explained in text). No S.E.M. is shown on day 1 in response to hemorrhage since all rats received the same score. Both groups demonstrate the increased severity of convulsive behaviour to AVP on day 3 when preceded by AVP i.c.v, or a hemorrhage of 15% of the blood volume 2 days earlier.

received 1.0 pg AVP i.c.v. 2 days following hemorrhage the behavioural effects were striking. The mean behavioural score was 2.6, indicating that rats displayed locomotor difficulties (n = 5), discrete myoclonic jerks (n = 5), barrel rotations and full myoclonic-myotonic convulsions (n = 2) or long periods of immobility and staring (n = 1) following AVP administration. Convulsive behaviour of this magnitude is not normally seen during the 10 min observation period immediately following a first i.c.v, injection of 1.0 pg AVP, as shown in Fig. 1 and previously reported12. Fig. 1 also illustrates results from the group of rats receiving 1.0pg AVP on days 1 and 3, replicating the previously reported sensitization to i.c.v. AVP 12, and demonstrating that hemorrhage on day 1 mimics the effect of i.c.v. AVP on day i in causing sensitization and hence relatively severe behavioural disturbances on day 3 in response to an i.c.v. injection of AVP. E E G recordings from a rat that has been sensitized to the behavioural and convulsive effects of AVP by a hemorrhage 2 days earlier are shown in Fig. 2. Prior to the injection of AVP (A), the recordings show fast, low-amplitude E E G activity. This is similar to recordings seen during and after hemorrhage on day 1. Following the injection of 1.0/~g of AVP i.c.v. (B) there was an immediate increase in the amplitude

and decrease in the frequency of hippocampal electrical activity followed by cortical spiking activity (Fig. 2). Abood et al.l also observed marked changes in the frequency and amplitude of the dorsal hippocampal electrical recordings following i.c.v, lysine vasopressin administration. Two-and-a-half minutes after the injection of AVP the rat began to display circling behaviour and locomotor difficulties but there were no marked qualitative changes in the E E G at this time compared to that of naive animals. Rats (n --- 3) which were repetitively hemorrhaged on days 1, 3 and 5 displayed behavioural anomalies which did not increase in severity. Furthermore, injection of 1.0pg of AVP i.c.v. 3 days following the final hemorrhage resulted in mild behavioural disturbances of considerably less severity than that seen when AVP was given after only one hemorrhage (data not shown). Rats which received 5.0 pl of saline 2 h prior to AVP injection, either 2 days following hemorrhage or 3 days following repetitive hemorrhage on earlier days, did not exhibit convulsive behaviour. This injection of vehicle had no effect on the behavioural response compared to those animals which received AVP with no prior vehicle injection. In response to pre-treatment with intraperitoneal (i.p.) hypertonic saline, rats again displayed increased sensitivity to i.c.v. AVP. After i.p. injection of hypertonic saline solution on day 1, rats often displayed slight locomotor abnormalities, possibly due to the irritating effect of the saline. On day 3 in response to the i.c.v, injection of AVP, the mean behavioural score was 2.8; that is, rats exhibited locomotor difficulties (n = 2), discrete myoclonic jerks (n = 2) and a myotonic convulsion with a slow myoclonic component (n = 1). Convulsive behaviour of this magnitude in naive rats is normally seen following a second injection of AVP, as shown in Fig. 1 and previously reported~2. Fig. 3 summarizes the responses of 3 different groups of rats to a first injection of 1.0pg AVP into a lateral cerebral ventricle. The response of naive rats to a first i.c.v, injection of 1.0/~g AVP is significantly different (Mann-Whitney U-test) from the response of rats pre-treated with either hemorrhage or hypertonic saline (P < 0.05 and 0.025, respectively). The responses on day 3 of rats from the latter 2 groups to a first injection of AVP are not significantly different.

62

A.

LC

I.!t

1

~ I

t

IIH

BI

'

i vtt

y j-!

I

Fig. 2. EEG recordings from a motionless rat demonstrating sensitization of the brain to 1.0 #g of AVP i.c.v. 2 days following hemorrhage. A: immediately prior to the injection of AVP. B: 70 s following the start of the first injection of AVP. LC, left cortex; LH, left hippocampus; RH, right hippocampus. Bar in lower left represents 1 s and bars at the end of the traces represent 100 gV.

DISCUSSION The results of the experiments reported here indicate that both hemorrhage and parenteral hypertonic saline can mimic the effect of a central injection of A V P in 'sensitizing' a rat to the behavioural effects of AVP. The mechanism by which the brain becomes sensitized to this action of A V P is unknown, thus it is uncertain as to whether the sequelae of an i.c.v, injection of A V P and that of hemorrhage or parenteral hypertonic saline are similar. However, we interpret these data to suggest that both these stimuli caused the release of central A V P , or that of a closely related peptide, so as to sensitize the brain to the be-

havioural anomalies induced by AVP. It is unlikely that circulating A V P is responsible for the sensitizing phenomenon, since other researchers have demonstrated that very little or no vasopressin administered peripherally penetrates the blood-CSF barrier25, even at blood A V P concentrations observed in hemorrhaged animals23. The present data suggest that stimuli which activate pituitary secretion of A V P also stimulate release of A V P into the brain. In support of these findings, it appears that there is a positive correlation between the concentrations of A V P in the plasma and CSF, and that they may be influenced to some extent by the same stimuli23. Other researchers have demon-

63 Elevated Responses to icv AVP 4-

Following Hemorrhage or Hypertonic Saline

3, LU

0 o t/) < tr

o_

2¸

> ,< 'I" LU m

'N =10' '

First Exposure to 1.0 #g AVP icy

First Exposure to First Exposure to 1.0 IJg AVP Icy 1.0 ;Jg AVP Isv following hemorrhage following hypettoni¢ (1,5M) saline

Fig. 3. Mean behavioural responses (+S.E.M.) in groups of rats receiving only AVP i.c.v. (stippled), AVP i.c.v. 2 days after a hemorrhage of 15% of the blood volume (vertical slashes) or AVP i.c.v. 2 days after receiving 1.0 ml of 1.5 M hypertonic

saline intraperitoneally (hatched). Both hemorrhage and hypertonic saline are effective,in increasing the sensitivityof the brain to the convulsiveaction of AVP i.c.v..

strated that AVP is released simultaneously into the blood and cerebrospinal fluid (CSF) of rabbits in response to bleeding, therefore increasing AVP levels in the brain itself22, Wang and co-workers 23 have also shown that although hemorrhage increases the concentration of AVP in both the plasma and CSF of dogs, the threshold for the increase in plasma AVP is much lower and the magnitude of increase in plasma AVP concentration is considerably greater than that for AVP in the CSF. This suggests that AVP released into the CSF during hemorrhage may have a different origin than that released into the blood 23. The site(s) of AVP release into the CSF is unknown. Immunocytochemical studies have demonstrated the presence of AVP-containing neurons ending at the median eminence, choroid plexus, lateral ventricle and third ventricle, in particular at its infundibular recess4,19. Such pathways may represent possible sites for AVP release directly into the CSF. It is also possible that

AVP simply diffuses from the neuropil of circumventricular organs. More direct measurements of central AVP release come from push-pull perfusion studies of various brain structures and third ventricular CSF in the conscious rabbit 15 and rat a. Results from this work indicate that AVP levels in the third ventricular CSF samples 3 and push-pull perfusates of limbic structures 15 are elevated in response to hypertonic saline, a potent stimulus for pituitary AVP secretion. Further evidence for endogenous brain release of vasopressin or a closely related peptide comes from a recent study by Kasting et al. 13, whereby hemorrhage in the sheep mimicked the antipyretic action of centrally injected AVP. The sensitizing action of this centrally released AVP may exert its effect by a number of mechanisms. It may be released into the ventricles and then be transported by the CSF to distant target cells in the CNS, or alternatively, it may be released by AVP-containing neurons at the appropriate target sites in the CNS. The observations made in the present experiments do not permit us to distinguish among these possibilities, but if the action of brain AVP is by the second mechanism, the relative inaccessability of the central AVP receptors to i.c.v, administered peptide would explain why large doses of AVP are needed to produce sensitization and convulsive effects. The reason for the lack of effect of i.c.v. AVP following 3 successive intermittent hemorrhages is unknown. It is clear that repetitive hemorrhage alone does not result in convulsive behaviour, and it is feasible that rather than sensitizing the receptor, successive exposure to AVP may result in relative desensitization or blocking of the receptor. Further studies are required to clarify the mechanism for this effect. These studies do not exclude the possibility that other endogenous substances which may be released during hemorrhage may also have effects on the sensitization process. Oxytocin levels have recently been reported to rise in response to both hemorrhage and hypertonic saline in dogs 24 and rats 3. It seems unlikely that oxytocin is responsible for the sensitization observed, however, since the dose required to produce convulsive effects when given i.c.v, is over 100 times that for vasopressin 14, and pretreatment of rats with i.c.v, oxytocin has been shown to prevent vasopressin-induced seizures 1.

64 The intraperitoneal injection of hypertonic saline also serves as a stimulus for neurohypophysial A V P release, possibly by acting at either an intracranial o s m o r e c e p t o r 21 or a sodium r e c e p t o r nearby the walls of the third ventricle 2. Small changes in plasma osmolality cause a rapid and considerable change in plasma A V P levels TM, corresponding with phasic bursts of activity in vasopressin magnocellular cells 10. Kannan and Yagi n have d e m o n s t r a t e d that a single cell in the supraoptic nucleus of the hypothalamus can respond to both osmotic (i.e. hypertonic saline) and non-osmotic (e.g. carotid occlusion) stimuli, suggesting that osmotic and non-osmotic pathways m a y affect the magnocellular neurosecretory cells. This may explain why both h e m o r r h a g e and hypertonic saline were equally effective in sensitizing the rat to the behavioural anomalies p r o d u c e d by an i.c.v, injection of A V P .

The correlative data presented in this p a p e r are consistent with the supposition that A V P is released in a coordinated fashion from both the neurohypophysis and from neurons within the CNS in response to osmotic (hypertonic saline) and non-osmotic (hemorrhage) stimuli. This stimulus-induced release of A V P mimicks the effect on convulsive and m o t o r behaviour of administering A V P into the lateral cerebral ventricle.

REFERENCES

12 Kasting, N. W., Veale, W. L. and Cooper, K. E., Convulsive and hypothermic effects of vasopressin in the brain of the rat, Canad. J. Physiol. Pharmacol., 58 (1980) 316--319. 13 Kasting, N. W., Veale, W. L. and Cooper, K. E., Effect of hemorrhage on fever: the putative role of vasopressin, Canad. J. Physiol. Pharmacol., 59 (1981) 324--328. 14 Kruse, H., Van Wimersma Greidanus, T. B. and DeWied, D., Barrel rotation induced by vasopressin and related peptides in rats, Pharmacol. Biochem. Behav., 7 (1977) 3tl-313. 15 Pittman, Q. J., Malkinson, T. J., Veale, W. L. and Lederis, K., Central release of arginine vasopressin (AVP) in rabbit brain, Soc. Neurosci. Abstr., 7 (1981) 506. 16 Pittman, Q. J., Veale, W. L. and Lederis, K., Central neurohypophysial peptide pathways - - interactions with endocrine and other autonomic functions, Peptides, 3 (1982) 515-520. 17 Renaud, L. P., Pittman, Q. J., Blume, H. W., Neurophysiology of hypothalamic peptidergic neurons. In K. Fuxe, T. Hokfelt and R. Luft (Eds.), Central Regulation of the Endocrine System, Plenum Press, NY, 1979, pp. 119-135. 18 Schrier, R. W., Beri, T. and Anderson, R. J., Osmotic and non-osmotic control of vasopressin release, Amer. J. Physiol., 236 (1979) F321-F332. 19 Sofroniew, M. V. and Weindl, A., Central nervous system distribution of vasopressin, oxytoein and neurophysin, tn J. Martinez, R. Jensen, R. Messing, H. Rigter and J. McGaugb (Eds.), Endogenous Peptides and Learning and Memory Processes, Academic Press, NY, 1981, pp. 327-369. 20 Swanson, L. W. and Kuypers, H. G. J. M., The paraventricular nucleus of the hypothalamus: cytoarchitectonic subdivisions and organization of projections to the pituitary, dorsal vagal complex, and spinal cord as demonstrated by retrograde fluorescence double-labelling methods, J. comp. Neurol., 194 (1980) 555-570. 21 Thrasher, T. N., Brown, C. J., Keil, L. C. and Ramsay, D.

1 Abood, L. G., Knapp, R., Mitchell, T., Booth, H. and Schwab, L., Chemical requirements of vasopressins for barrel rotation convulsions and reversal by oxytocin, J. Neurosci. Res., 5 (1980) 191-199. 2 Andersson, B., Thirst and brain control of water balance, Amer. Sci., 59 (1971) 408--415. 3 Barnard, R. R. and Morris, M., Cerebrospinal fluid vasopressin and oxytocin: evidence for an osmotic response, Neurosci. Lett., 29 (1982) 275-279. 4 Buijs, R. M., Swaab, D. F., Dogterom, J. and Van Leeuwen, F. W., Intra- and extrahypothalamic vasopressin and oxytocin pathways in the rat, Cell Tiss. Res., 186 (1978) 423--433. 5 Burnard, D. M., Pittman, Q. J. and Veale, W. L., Vasopressin-induced convulsions: sensitization by hemorrhage, Proc. Canad. Fed. Biol. Soc., 24 (1981) 171. 6 Burnard, D. M., Pinman, Q. J. and Veale, W. L., Increased convulsive effects of arginine vasopressin following hypertonic saline: evidence for central AVP release, Fed. Proc., 41 (1982) 1508. 7 Cain, D. P., Effects of kindling or brain stimulation on pentylenetetrazol-induced convulsion susceptibility, Epilepsia, 21 (1980) 243-249. 8 DeWied, D. and Gispen, W. H., Behavioural effects of peptides. In H. Gainer (Ed.), Peptides in Neurobiology, Plenum Press, NY, 1977, pp. 397--448. 9 Goddard, G. V., Mcfntyre, D. C. and Leech, C. K., A permanent change in brain function resulting from daily electrical stimulation, Exp. Neurol., 25 (1969) 295-330. 10 Hayward, J. N., Functional and morphological aspects of hypothalamic neurons, Physiol. Rev., 57 (1977) 574-658. 11 Kannan, H. and Yagi, R., Supraoptic neurosecretory neurons: evidence for the existence of converging inputs from carotid baroreceptors and osmoreceptors, Brain Research, 145 (1978) 385-390.

ACKNOWLEDGEMENTS This research was s u p p o r t e d by the Medical Research Council of C a n a d a ( M R C C ) , D . M . B . is an M R C Student, and Q.J.P. is an M R C Scholar.

65 J., Thirst and vasopressin release in the dog: an osmoreceptor or sodium receptor mechanism? Amer. J. Physiol., 238 (1980) R333-R339. 22 Vorheer, H., Bradbury, M. W. B., Hoghoughi, M. and Kleeman, C. R., Antidiuretic hormone in cerebrospinal fluid during endogenous and exogenous changes in its blood level, Endocrinology, 83 (1968) 246-250. 23 Wang, B. C., Share, L., Crofton, J. T. and Kimura, T., Changes in vasopressin concentration in plasma and cere-

brospinal fluid in response to hemorrhage in anesthetized dogs, Neuroendocrinology, 33 (1981) 61-66. 24 Weitzman, R. E., Glatz, T. H. and Fisher, D. A., The effect of hemorrhage and hypertonic saline upon plasma oxytocin and arginine vasopressin in conscious dogs, Endocrinology, 103 (1978) 2154-2160. 25 Zaidi, S. M. A. and Heller, H., Can neurohypophysial hormones cross the blood-cerebrospinal fluid barrier? J. Endocri., 60 (1974) 195--196.