Interaction of arginine vasopressin and angiotensin II on Ca2+ in vascular smooth muscle cells

Kidney International, Vol. 38 (1990), pp. 47—54

Interaction of arginine vasopressin and angiotensin II on Ca in vascular smooth muscle cells CARLOS CARAMELO, Koi OKADA, PHOEBE TsAI, STUART L. LINAS, and ROBERT W. SCHRIER Department of Medicine, University of Colorado School of Medicine, Denver, Colorado, USA

Interaction of arginine vasopressin and angiotensin II on Ca in vascular smooth muscle cells. The non-osmotic release of arginine vasopressin (AVP) is associated with the concomitant activation of the renin-angiotensin and sympathetic nervous systems. In vivo studies

suggest that a positive interaction may occur between AVP and angiotensin II (Ang II), and other Ca2 mobilizing hormones. In the present study, the cellular mechanisms of this interaction between AVP and Ang II in vascular smooth muscle cell (VSMC) were examined. These results support the existence of a positive interaction between AVP and Ang II on Ca2 mobilization in VSMC. In fact, the challenge of VSMC with combined AVP and Ang II, in a range from 5 X lOIs to 10—8 M, enhanced cytosolic free Ca2 ([Ca2]1) and 43Ca2 efflux in a more than additive manner. This potentiation, which was not dependent of the presence of extracellular calcium, correlated with an increased

observed between a-l-adrenergic and histaminergic contraction

in isolated vessels [8]. Furthermore, the effect of AVP on vascular tone has also been shown to be increased by the influence of Ang II or the adrenergic system [9]. However, the subject of the pressor hormone interaction in in vivo circum-

stances is not yet completely elucidated, and controversial results have been published in different animal models [10, 11].

In recent years, studies on the post-receptor events of

hormone-mediated cellular activation have provided new insights on the mechanisms involved in smooth muscle contraction [12—16]. The contractile cell's response to pressor horVSMC shape change, Moreover, the combination of subthreshold mones seems to involve intracellular mechanisms which are, at doses of AVP and Ang II (5 x lO" M), which do not release Ca2 least in part, similar for AVP, Ang II, and NE [12—16]. The first, alone, evoked a Ca2 mobilizing response. A subthreshold dose of Ang II also shifted to the left the concentration-response curve of the rapid phase of cell contraction is mainly due to the transient AVP-mediated 45Ca2 efflux. Since there were no changes in receptor binding of either hormone by the other hormone and the interaction of

the two hormones on the production of inositol phosphatides was additive, the AVP and All positive interaction on Ca2 mobilization on

VSMC may occur at the level of the intracellular Ca2-releasing mechanism itself. Such an interaction can occur at hormone concentrations below the Ca2 release threshold and may explain an increased functional response to the combination of pressor hormones compared to that of each hormone alone.

A series of studies both in humans and experimental animals

have demonstrated that the non-osmotic release of AVP is accompanied in virtually every circumstance by a simultaneous

activation of the renin-angiotensin and sympathetic nervous systems [1]. Additionally, in various pathological circumstances accompanied by alterations of the vascular tone, such as DOC-salt hypertension [2], renovascular hypertension [3],

increase in [Ca2]1 by release of Ca2 from an intracellular store, namely the endoplasmic reticulum (ER) [12—16]. The elevated [Ca2]1 triggers the activation of the actin-myosin

system and rapidly returns to near baseline [12—16]. Also, the simultaneous activation of protein kinase C induces the phosphorylation of specific subclasses of proteins involved, among other functions, in the maintenance of sustained contraction [14]. Interactions between pressor hormones may therefore occur at any level of this process and may result in cooperative or competitive effects between the hormones. Since the only available information on this subject of pressor

hormone interaction comes from whole animal or isolated vessel studies [7—11], the present study was designed to exam-

ine the interactions between pressor hormones at the cellular level, using cultured VSMC. Such an approach allows the elimination of other potentially interacting mechanisms that are present in in vivo studies, such as endothelium- and nervous ganglionic blockade [4], adrenal insufficiency [5], and dehydra- system-mediated responses; the influence of circulating hortion [6], cooperation between the pressor systems is critical in mones other than those used in the experiments is also exmaintaining cardiovascular homeostasis. cluded. The present study involved the assessment of the In the conscious rat, simultaneous infusions of AVP, Ang II, interaction of two pressor hormones, AVP and Ang II, on and norepinephrine (NE) enhance the pressor response of one VSMC Ca2 mobilization. After a positive interaction of AVP another, as well as convert subpressor doses of these hormones and Ang II on [Ca211 and 45Ca2 effiux was found, the to pressor doses [7]. Similar positive interactions have been functional significance of such an interaction was explored by studying VSMC shape change. Finally, receptor binding and inositol phosphates production were studied to further clarify the mechanism of the positive AVP and Ang II interaction. Received for publication February 24, 1989 and in revised form January 29, 1990 Threshold and subthreshold doses were used in the experiAccepted for publication February 2, 1990

ments, in an effort to approximate conditions in different

© 1990 by the International Society of Nephrology

physiological and pathological circumstances. 47

48

Caramelo et a!: Hormone interaction and calcium

Methods

Determination of 45Ca2 efflux 45Ca2 efflux was studied by a method similar to that previCell culture ously described [17, 181. The culture medium was removed by Rat aortic smooth muscle cells were isolated using a modifi- aspiration and cell monolayers were rinsed twice with 1 ml PSS cation [17, 181 of the technique described by Chamley, Camp- and loaded with 45Ca2 (8 Ci in I ml PSS, 37°C, 3 hr). Then the bell and McConnel [19]. Briefly, rat thoracic aortas were cultures were rapidly rinsed ten times (60 sec total) with 1 ml dissected and minced under sterile conditions and incubated in PSS and 1 ml of PSS was added. The medium was removed and Eagle's minimum essential medium (MEM) containing 2 mg/mI replaced at each minute with 1 ml of PSS, from one to six minutes. Samples from each interval were placed into vials for collagenase (Cooper Biomedical) (7.5 ml, 37°C, 2.5 to 3.0 hr). The freshly isolated cells were resuspended in Eagle's MEM liquid scintillation counting (Tricarb 460C, Packard Instrucontaining 1 M glutamine, 100 U/mI penicillin G, 100 /Lg/ml ments, Downers Grove, Illinois, USA). Hormones were added streptomycin, and 10% fetal bovine serum (FBS). Cells were at six minutes of the time course of 45Ca2 efflux, and one plated in 35 x 10mm culture dishes at a density of 2 x i05 cells minute samples collected for five additional minutes. At 10 per 1.5 ml of culture medium. For fura2 experiments, cells were minutes, ionomycin (106 M, 1 ml) was added and a one minute sample collected. Intracellular radioactivity was extracted with grown on round glass 13 mm diameter coverslips. Cultures were incubated at 37°C in a humidified atmosphere 1 ml sodium dodecyl sulfate (SDS) alkaline solution (0.1% SDS, of 95% air and 5% CO2. 45Ca2 efflux and fura2 experiments 2% Na2CO3, 0,1 N NaOH). Counts corresponding to basal were performed between days 14 and 21 when cells formed a 45Ca2 release were subtracted and results were calculated as confluent monolayer with a density of approximately 1.5 to 2 x percent of the total 45Ca2 counts released from the cells in the 106 cells per dish. All the experiments except receptor binding first 120 seconds after the addition of the hormone. 45Ca2 were done in primary cultured cells. For the receptor binding release is directly proportional to the [Ca2]1 transient during studies, cells were lifted from the culture dishes by Trypsin- the first two minutes after the challenge with the agonists (1O— AVP, r 0.938, iO M Ang II, r = 0.849, N = 10, P < EDTA (0.25 to 0.1%) treatment and subcultured. In order to obtain an adequate number of confluent cells, the cells used for 0.001). In both fura2 and 45Ca2 efflux experiments, doses of AVP AVP or Ang II binding were between third and fifth passages. and Ang II from 10" to l0_6 M were used for cell stimulation Shape change experiments were carried out at three to five days after cells plating for primary culture. Identity of cells was as detailed in the Results section. In previous experiments with confirmed by electron microscopy and viability was higher than both AVP and Ang II, 10'° and 10_6 M concentrations had been established as the threshold and maximal concentrations, 95%, as judged by Trypan blue exclusion. respectively [data not shown, Ref. 17]. Experiments in medium without Ca2 plus EGTA 0.5 m were also performed to study Measurement of intracellular free Ca2 the effect of extracellular Ca2 on the interaction of AVP and Ang II. In the latter experiments, the no-Ca2/EGTA medium As described [17, 18], the monolayers on coverslips were was added for no more than two minutes before the hormonal incubated with MEM containing no FBS and with 4 jM fura2 agonist, to avoid cellular Ca2 depletion. No alterations in cell AM for 60 minutes at 37°C. The monolayers were then washed viability or cell attachment to the culture surface were detected three times with MEM containing no fura2 and incubated for after two minutes in the presence of 0.5 m EGTA medium. another 15 minutes at 37°C to allow any non-hydrolized fura2 to diffuse from the cells. Before measurements were made, coyCell shape change studies erslips were rinsed again with physiological saline solution The functional response of VSMC was evaluated as the cell (PSS) (140 mrt NaCl, 4,6 m KCI, 1.0 mM MgCI2, 2.0 mM CaC12, 10 m glucose, 10 mM HEPES, pH 7.4). Fluorescence shape change by using a phase contrast microscope (Carl Zeiss, was measured at 37°C using a fluorescence spectrophotometer 1M35, Oberkochen, FRG) with a computerized scanning device equipped with a thermostatically controlled cuvette holder (Zidas, Carl Zeiss) and printer [17]. The VSMC shape change is (Perkin-Elmer 650-lOS, Norwalk, Connecticut, USA) at an mostly due to cell contraction, which causes a significant emission wavelength of 500 nm. Excitation wavelengths of 342 reduction in size. However, changes in the organization of the and 380 nm were chosen to monitor the Ca2-induced shift in cytoskeleton, as well as variations in cell volume secondary to fura2 fluorescence, as determined in previous calibrations. activation of ion transport mechanisms are probably also imDigitonin and EGTA were used for obtaining Rmax and Rmjn in portant factors of the shape changes. At the time of the each individual experiment [17, 18]. Autofluorescence was experiment, culture medium was aspirated and cells were measured in similar cells which had not been loaded with fura2, washed twice with PSS. The cells were then incubated at room and was below 10% of the total fluorescence of fura2 loaded temperature with PSS (37°C) in the presence of different concells in all of the experiments. No changes in autofluorescence centrations of the hormones, and the shape change was evaluwere detected after the addition of the hormones. Using a ated by measuring the area of the cells. The magnitude of the fluorescence microscope (Dialux 20, Leitz, FRG), more than shape change response was determined by comparing groups of 95% of the VSMC appeared loaded with fura2 in an homoge- 6 to 12 cells before and after 20 minutes of the addition of the neous pattern, suggesting a cytosolic distribution of the dye. hormone. Several aspects were considered to insure the reli[Ca2]1 was calculated by the formula used by Grynkiewicz, ability of the measurements: 1) the measurements were obPoenie and Tsien [20]. Peak [Ca2] was used for the purpose of tained by triplicate and the mean value of one cell was used for comparison; 2) all the groups of measurements with a standard calculations.

49

Caramelo et a!: Hormone interaction and calcium

deviation higher than 10%, respectively, to the mean value were discarded; and 3) since spontaneous change of the area of some cells of 5% or less were occasionally observed, only reductions

1200 1000

of 15% or more of the area of the cells were considered as significant.

800

Photographs of all the cellular groups before and after the hormone challenge were taken for planimetric measurement and the results compared with those of digital computerized planimetry. No significant discrepancies were found between the results obtained by the two methods. In the experiments

+

0(5

600 400 200

using combinations, both hormones were added simultaneously.

0

Receptor binding studies

10

The experiments examining AVP and Ang II binding to

9

8 M -Iog[AVP],

7

6

VSMC were carried out in confluent monolayer culture VSMC Fig. 1. Positive interaction between arginine vasopressin (A VP) and

on 35 x 10 mm plastic dishes. Cells were rinsed twice with angiotensin II (Ang H,) on cytosolic Ca2t at different hormone concenbinding buffer (25 mrvi HEPES, 119.2 mti NaCI, 3.0 ma'i KCI,

1.2 msi MgSO4, 1.0 mM CaCl2; 1.2 mM KH2PO4, 1.0 mM glucose, 0.1% BSA, pH 7.4). The dishes were incubated at 4°C for 60 minutes with buffer and effectors and 3H AVP (2 x i0 M). At the end of the incubation, all dishes were rinsed four times with ice-cold buffer. The cells were lifted up from dishes with 0.1% SDS + 0.1 N NaOH, to use for scintillation counting and protein determinations. Non-specific binding was determined by the same incubation procedure in the presence of 1 x iO M AVP. All of the determinations were done in triplicate samples. The Ang II binding experiments were performed in similar conditions as AVP binding, except that the composition

trations. Symbols are: (—•—) AVP + Ang II; (--0--) AVP; (—A—) Ang II. These results, and those in the following figures, except Figure 2, correspond to x SEM of a minimum of five experiments at each concentration. *D < 0.001 compared to the response to AVP or Ang II alone; p < 0.05 compared to the sum of the responses to AVP and Ang II.

LM). The extracts were applied to columns containing 1 ml of

Dowex-l (x8, Formate form) and serially eluted with 2 ml

aliquots of CH7O, Borax (5 ma disodium tetraborate, 60 mM sodium formate), 0.2, 0.5, and 1.0 M ammonium formate (in 0.1 of the binding buffer was 100 mrvi NaCI, 50 ma'i Tris-HCI, 10 mM M formic acid) separating, respectively, inositol, glycero-phosKC1, 5 mi MgCl2, supplemented with 0.5 mg/mI Bacitracin and phatidylinostol, inositol-l -phosphate, inositol bisphosphate, 0.25 BSA, pH 7.4. Fifty to 100 fmol of 125 Ang II were added to and inositol trisphosphate (1P3); this last fraction includes 1,3,4each well. Binding studies were performed at 4°C for 90 and l,4,5-1P3 and l,3,4,5-inositol phosphate. Samples were minutes. Specific binding was defined as total binding minus collected and counted in a scintillation counter. The extraction non-specific binding (in the presence of 1 tM Ang II). The procedure has been previously validated in this laboratory with experiments compared binding of the labeled hormone in the radioactive phosphoinositides. Results were expressed as numabsence or in the presence of the other unlabeled hormone, ber of 1P3 counts per milligram of protein. In all the experisuch as 3H AVP alone versus 3H AVP + unlabeled Ang II or ments, where needed, protein concentrations were measured by the method of Lowry et al [21]. '25Ang II alone versus 125Ang II + unlabeled AVP. Measurement of inositol phosphates Measurement of the intracellular levels of inositol phosphates were performed as described [17, 181. After cells were grown to confluence, the culture media of each dish was replaced with 1

Statistical analysis Two-way analysis of variance, unpaired Student's t-test and chi square on '2 x 2" tables were used for statistical comparisons. A P value of less than 0.05 was considered significant.

ml of inositol-free medium containing 10 pCi of myo-

Results

2[3Hjinositol (specific activity 16.3 Ci/mmol), and the cells were

Intracellular free Ca2 The basal value for [Ca2]1 measured by the fura2 method experiment, culture media were aspirated and replaced with 500 was 86.5 24.0 flM (N = 120). The concentration-response pildish of 37°C PSS containing 108 or l0— M AVP. The pattern of the VSMC [Ca2] levels in the presence of AVP and reaction was stopped after 30 seconds using 500 p1 of ice-cold Ang II alone or in the presence of an equimolar combination of trichloroacetic acid (20% solution). Cells were scraped using an both hormones is shown in Figure 1. At submaximal concen-

incubated for 24 hours. It has been previously shown that steady state labeling occurs by 24 hours. At the time of the

Eppendorf pipette tip and aliquots were taken for protein trations, that is, between 10_b and 10—8 M, the combination of determination.

AVP and Ang II increased [Ca2]1 at a level more than additive

The samples were centrifuged at 1,000 x g for 10 minutes. The supernatant, containing inositol phosphates, was washed three times with an equal volume of ether and stored at —20°C until analysis, which was performed during the same week of the extraction. At the time of the analysis, the water-soluble fraction was thawed and brought to pH 6.0 using Tris-base (50

compared to the separate effect of each hormone (Fig. 1). However, at iO— M the effect of the two hormones was additive

and at lO_6 M the effect of the two hormones added together was less than additive (Fig. 1). A similar potentiation, albeit at slightly reduced [Ca2] values, was observed when experiments were carried out in medium without Ca2 and supple-

50

Caramelo et a!: Hormone interaction and calcium 1000

Receptor binding No differences in the amount of specific receptor binding of AVP were found by incubating the cells in the presence of 10—8

800

M Ang II (Fig. 6) and 2 x 10-8 M (data not shown). The 600

concentrations of the hormones are detailed in the legend of the figure. Changes in Ang II binding were also not found when the

400

binding experiments were done in the presence of 2 x io (N = 4) and 10—8 M AVP (N = 4) (95.09 10.4 and 93.6 9.3% of the specific binding, respectively, NS).

+

0

200

Production of inositol phosphates

As shown in Figure 7, the values of 1P2 and 1P3 were 0 0

20

40

60

80

100

120

Time, seconds

Fig. 2. Time profile and shape of the cytosolic Ca2 curve with AVP

and Ang II added separately or together. Symbols are: (—•—) AVP iO M + Ang II iO M; (—0—) AVP i0 M; (—A—) Ang II iO M. This figure was constructed with the data from a representative experiment.

mented with EGTA ([Ca2]1, AVP 10—8 M, 160.7 44.6; Ang II 10—8 M, 149.6

10.1; AVP 10—8 M + Ang 1110—8 M 536.2

increased in an additive manner by the combination of the two hormones at l0— M. No changes in 1P1 levels were detected by

the action of the hormones at the time (15 sec) used for the incubations (10 M AVP, 24735 2557; i0 M Ang II, 27081

1332; i0 M AVP + i0 M Ang II, 27383

1970; basal 3950 cpm/mg prot', NS). A similar additive effect of the combination of the two hormones was found at a lower concentration of the hormones, since 108 M AVP produced an increase of 1P3 over the basal (cpm . mg prot') of 580 252, 108 M Ang II, 452 294 and both together 1173 634. 22463

Discussion

89.9, N = 4). The time profile and the shape of the [Ca2]1 transient produced by the association of AVP + Ang II were

The present study examined the interaction of two vasoactive hormones, AVP and Ang II, at the cellular level in VSMC. The

similar to that obtained with either hormone alone (Fig. 2). In addition, no Ca2 mobilization was detected at concentrations of 5 x 10—11 M of each hormone alone (subthreshold), whereas the combination of the same concentration of AVP (5 x

effect of such pressor agonists on vascular tissue could be

M) and All (5 x 10 ' M) produced a detectable increase of [Ca2]1 (50.3

5.1 M, N = 8). 45Ca2 + efflux

The effect of various concentrations of AVP and All alone and combined are depicted in Figure 3. A similar pattern of response as that observed in the fura2 experiments was also found in the 45Ca2 efflux experiments, in the range between

competitive, thus a diminution in their vascular effects would occur when the hormones are administered simultaneously. Alternatively, an additive or synergistic effect might occur in the same circumstance. The present results provide evidence that in cultured VSMC a potentiating effect between AVP and Ang II occurs on cellular Ca2 mobilization, as assessed by the

changes in [Ca2], levels and Ca2 efflux. Evidence for that

positive interaction between AVP and Ang II included: 1) the increase in the [Ca2]1 transient by the simultaneous administration of both hormones at submaximal concentrations was more than the sum of that of either hormone alone; 2) a Ca2 i0 to lO—'° M. The total amount of releasable 45Ca2 de- mobilizing response was observed with the simultaneous adcreased proportionally to the amount of 45Ca2 released by the ministration of subthreshold concentrations of the two horhormones. Subthreshold doses of the two hormones (5 x 101 mones; and 3) the enhanced effect on the AVP-mediated Ca2 M) administered simultaneously produced an increase in 45Ca2 efflux was observed with a subthreshold concentration of Ang efflux (12.4 0.9 counts/total counts 1) which was not present II. At maximal AVP concentrations (10—6 M), however, the when each hormone was administered separately. A subthresh- potentiating effect was no longer present; most likely, this old concentration of Ang II (5 x 10—11 M) also produced a observation may be due to the exhaustion of releasable intraleftward shift of the Ca2 efflux concentration-response curve cellular Ca2 stores (Fig. 3, inset). The results of the experiments performed in medium without extracellular Ca2 demonof AVP (Fig. 4). strate that the interaction between AVP and Ang II probably VSMC shape change involves Ca2 release from intracellular stores and suggests The combination of AVP + Ang II produced an increased that an increase in Ca2 uptake is not the primary mechanism of percent of shape change of VSMC as compared with the effect the increased rapid Ca2 mobilization. Moreover, Takeda et al of either hormone alone (Fig. 5), which was more marked at [22] have demonstrated that dantrolene, a blocker of Ca2 10b0 and 10—8 M. The combination of both hormones at a release from the endoplasmic reticulum, abolishes the Ca2 concentration of 5 x lO M elicited a shape change of 8.10 transient in response to pressor hormones such as AVP. How0.7% of the VSMC, whereas either AVP or Ang II alone at 5 x ever, an interaction between the hormones on Ca2 uptake at a 10h1 M did not provoke a cell shape change different from the later time cannot be excluded by the present experiments. In spontaneous (AVP 3.4 1.8 + Ang 113.2 1.6% of the VSMC, this regard, synergistic interactions between glucagon and Ca2 both P < 0.005, compared to the two hormones added togeth- mobilizing agents on a Ca2 influx pathway have been found in perfused rat liver [23—25]. er).

Caramelo et a!: Hormone interaction and calcium

51

A *1

60

E 0 C

C.,

50

0

0

B 60

>.

E a)

•0

a) U) a) a) a) a)

30

C

0C.,

* C.)

0,

.0 40

a)

20

40 20

9876

0

0 I-.

0a)

-IogIAVP], M

10

0-

9

8

7

concentrations of AVP and Ang I! on 4SCa2+ efflux. Symbols are: (--0--) AVP; (—A—) Ang 11; (—•—) AVP + Ang II. Inset. Amount of ionomycinreleasable 45Ca2 remaining in the cells at 4 mm

after the application of the hormones. *p < 0.001 compared to the response to AVP or Ang II alone; 0.05 compared to the sum of the responses to AVP and Ang II.

6

p<

-log(AVPI, M

chard analysis for each hormone in the presence of the other hormone, will be necessary to clarify whether there is "crosstalk" between the binding sites, perhaps as mediated by an effect of signal processing. A post-receptor interaction between AVP and All could involve inositol phosphates production. 1P3 would then act as a messenger to stimulate Ca2 release from the endoplasmic

U) C.)

0

Fig. 3. Effect of the combination of threshold

30

E

20 a) U)

reticulum [10—12]. Our results demonstrate that, at an early time C.)

'l

of activation (15 sec), the combination of the two pressor

10

hormones is additive for both 1P3 and 1P2. Nabika eta! [25] have also found an additive increase in inositol phosphatides production by VSMC simultaneously challenged with AVP and Ang II.

(-)

0

10

9

8

7

However, these authors did not find changes in 1P3 at one

minute after the administration of AVP alone. This difference in 1P3 results is probably due to the short duration of the IP3 peak; Fig. 4. Leftward shifting of the 45Ca2 efflux AVP concentration specifically, this 1P3 peak may no longer be detectable 30 to 45 -log, M

response curve in the presence of a subthreshold (5 X 10" M) seconds after the hormone stimulation. This time is simultaconcentration of Ang II. Symbols are: (.-O--) AVP; (—----) AVP + neous with the Ca2 release from the intracellular stores. Ang 11(5 x 10" M). * < 0.001. The fact that AVP and Ang II at subniaximal concentrations interact on [Ca2]1 mobilization and 1P3 production in a nonOn the basis of the current knowledge of the pathways competitive manner is therefore suggestive that the binding of leading to vascular smooth muscle contraction, several possible each hormone to its specific receptor stimulated production of cellular sites have to be considered in the analysis of the cellular IP3 from separate sources, for example, different subunits of interaction of AVP and Ang II. AVP and Ang II may interact at phospholipase C or different membrane phospholipid molea receptor level; specifically, one hormone might modify the cules. Hypothetically, this would provoke the breakdown of number of binding sites or the affinity of the receptors for the different PIP2 molecules and probably release Ca2 from indeother hormone. However, the positive interaction observed pendent subunits of the same ionomycin-releasable store. This between the two hormones at near maximal binding concentra- interpretation is consonant with evidence showing that the PIP2 tions, that is, 100 times or more higher than the Kd (nM) of their pool available for 1P3 formation is compartmentalized; thus, respective receptors, is difficult to explain by an effect at the only part of the total PIP2 pool may be available to each receptor level. Moreover, in the experiments assessing cellular separate receptor [15]. Ca2 changes, in which both hormones were added simultaSince at submaximal concentrations the two hormones added neously, a more rapid onset (15 sec) was observed than would together released more Ca2 than the sum of the Ca2 released be expected by interactions at the receptor level. The experi- by each hormone added alone, a second mechanism of hormone mental results of the binding experiments also provided evi- interaction may be operative. Specifically, an increased 1P3dence that the potentiated Ca2 peak was not due to increased mediated release of Ca2 from intracellular stores may occur. hormone binding. However, further experiments, that is, Scat- The experiments showing that subthreshold concentrations of

52

Caramelo et a!: Hormone interaction and calcium 50

40 8)

C CU

c 30 '1) CU

20 10_1010_10 10—10 + 10_b

io— io— io—

i0-

10_8 10_8 10—8 + 10—8

Hormone concentration, M

Ca2 transport [16]. The association of Ca2 containing endoplasmic reticulum vesicles with a membrane receptor-related domain may also be important in modulating the amount of Ca2 release [31] by increasing the availability or sensitivity of the intracellular Ca2 stores to the action of the Ca2 releasing mediators, namely 1P3 and GTP. An interaction between the two hormones on the production of prostaglandins, that is, PGE2, affecting Ca2 turnover is also theoretically possible and cannot be ruled out from the present results. Thus, the interaction between AVP and Ang II to provoke an increased Ca2 releasing response may involve one or more of the aforementioned mechanisms.

12000 8)

0.

10000

0) 8000

0) 6000 C

4000

< C

1O 10 +iO

Fig. 5. Vascular smooth muscle shape change in the presence of ddferent concentrations of separate or combined AVP and Ang II. Symbols are: (0) AVP; () Ang < 0.001 compared II; (U) AVP + Ang 11. to the response to AVP or Ang II alone. Each column corresponds to a minimum of 45 cells in three or more separate experiments.

2000 Control

Control

+

All (10-8 M)

6. Absence of changes in total and specUic receptor binding of AVF in the presence of Ang II. Symbols are: (0) specific binding; () non-specific binding. Labeled AVP concentration 2 x lO M; Ang II Fig.

concentration 10—8 M.

A functional correlation of the AVP and Ang II interaction on the Ca2 releasing mechanism was also provided by the results

of the VSMC shape change experiments. These results are consistent with results in in vivo experiments [7] and in isolated

vessel preparations [8]. However, the shape change experiments suggest that, especially at hormone concentrations higher than io' M, there may not be a full correspondence between the increase in Ca2 mobilization by the combined AVP and Ang II and the shape change response. Also, it should

one of the hormones provoked a leftward shift of the 45Ca2 release by the other hormone, and that the association of two subthreshold concentrations resulted in a threshold effect demonstrates that a measurable increase in [Ca2] by one of the hormones is not necessary for it to potentiate the Ca2 mobilizing effect of the other hormone. Ware, Smith and Salzman [261 have also found that a simultaneously elevated cytosolic Ca2 is not necessary for the synergistic interaction between platelet aggregating mediators to occur. Ca2 release from intracellular stores has been shown to be a complex phenomenon, probably involving interactions between 1P3 and GTP [27—29], cyclic nucleotides [14, 30], and phosphorylation of proteins of the endoplasmic reticulum which are involved in

be emphasized that the present results at the cellular level cannot be automatically generalized to the actual in vivo conditions in humans or experimental animals.

Lastly, it should be emphasized that there are important pathophysiological implications from this positive interaction between pressor hormones, since AVP, Ang II and the sympathetic nervous system are simultaneously activated in a number of clinical circumstances [1]. As recently found by Arnolda et a!

[32] in a rabbit model of congestive heart failure and by Zimmerman, Robertson and Jackson [33] in a model of renovascular hypertension, lower degrees of activation of combined pressor systems may be more efficacious in maintaining arterial pressure or eliciting a vasoconstrictive response than higher degrees of stimulation of any single pressor system. The present

53

Caramelo et al: Hormone interaction and calcium 25000

20000

q)

1'

15000

10000

Fig. 7. Effect of AVP and Ang ii on production of inositol phosphates, 15 sec after the hormones application. Symbols are: (D)

basal; () Ang II i0 M; () AVP i0 M; () AVP l0 M + Ang H i0 M. *D < 0.001

5000

compared to basal. P < 0.001 compared to AVP or Ang II alone. The P value was not

0

'I'2

1P3

1P2

P3

1P2

ip3

results thus emphasize the importance of considering not only the effect of each separate pressor agent, but also the potential interaction of such mediators on vascular smooth muscle contraction. One mechanism of this interaction appears to occur at a post-receptor site and involve increased Ca2 release from intracellular stores.

ip2

p3

significant between the sum of the 1P2 or 1P3 values in the presence of AVP alone and Ang II alone and the 1P2 or 1P3 values in the presence of both hormones added together.

pressin and other vasoactive hormones in the rat. Miner Electrol Metab 10:184—189, 1984 8. STUPECKY GL, MUNOZ DL, PURDY RE: Vasoconstrictor threshold

synergism and potentiation in the rabbit isolated thoracic aorta. J 9.

Pharmacol Exp Ther 238:802—808, 1986 TABRIZCHI R, KING KA, PANG CCY: Vascular role of vasopressin in

the presence and absence of influence of angiotensin II or

a-adrenergic system. Can J Physiol Pharmacol 64:1143—1148, 1986

Acknowledgments This work was supported by a grant from the NIH DK19928. The authors wish to thank Dr. Oswald Pfenninger for the fluorescence microscopy observations, Mrs. Rochelle Marzec-Calvert for the angiotensin H receptor binding, Mrs. Carolyn Burke for the illustrations, and Ms. Linda M. Benson for secretarial assistance.

Reprint requests to Dr. Robert W. Schrier, C-281, Department of Medicine, University of Colorado School of Medicine, 4200 E. 9th Ave., Denver, Colorado 80262, USA.

10. ELIJOVICH F, BARRY CR, KRAKOFF LR, KIRCHBERGER M: Differ-

ential effect of vasopressin on angiotensin and norepinephrine pressor action in rats. Am J Physiol 247:H973—H977, 1984 11. JOYNER WL, MOHAMA RE, GILMORE JE: Specificity of arginine

vasopressin and angiotensin II for microvessels in the hamster cheek pouch after the induction of renovascular hypertension. Micro vascular Res 35:8—20, 1988 12. RASMUSSEN M: The calcium messenger system. 314:1094—1101, 1986a 13. RASMUSSEN M: The calcium 314:1164—1170, 1986b 14.

RW: Pathogenesis of sodium and water retention in

high-output and low-output cardiac failure, nephrotic syndrome, cirrhosis, and pregnancy. NEnglfMed 319: 1065—1072, 1127—1134, 1988 2. MATSUGUCHI H, SCHMID P0: Acute interaction of vasopressin and

messenger system. N Engi J Med

ABDEL-LATIF A: Calcium mobilizing receptors, polyphospho-

inositides, and the generation of second messengers. Pharmacol

References 1. SCHRIER

N Engi J Med

Rev 38:227—272, 1986 15.

BERRIDGE MJ: Inositol trisphosphate and diacyiglycerol: Two

interacting second messengers. Ann Rev Biochem 58:159—193, 1987 16. CARAFOLI E: Intracellular calcium homeostasis. Ann Rev Biochem 56:395—433, 1987

17. MEYER-LEHNERT H, CARAMELO C, TSAJ P. SCHRIER RW: Interac-

neurogenic mechanisms in DOC-salt hypertension. Am J Physiol

tion of atriopeptin III and vasopressin on calcium kinetics and

242:H37—H43, 1982 3. ISHIKAWA S, JOHNSON JA, FOWLER WL, PAYNE CO, Kuiz K,

contraction of aortic smooth muscle cells. J Clin Invest 82:1407—

KEITZER WF: Pressor responses to norepinephrine in rabbits with 3-day and 30-day renal artery stenosis: The role of angiotensin II. Circ Res 43:437—446, 1978 4. MANCK PC, FIKSEN-OLSEN MJ, BiurroN SL, ROMERO JC: Role of

1414, 1989 18. CARAMELO C, T5AI P. SCHRIER RW: Mechanism of cellular effect of phorbol esters on action of arginine vasoressin and angiotensin II

on rat vascular smooth muscle cells in culture. Biochem J 254: 625—629,

1988

angiotensin and vasopressin on blood pressure of ganglionic 19. CHAMLEY JH, CAMPBELL OR, MCCONNEL JD: Comparison of blocked dogs. Am I Physiol 244:H115—H120, 1983 5. IsHIKAwA S, SCHRIER RW: Vascular effects of arginine vasopressin, angiotensin II and norepinephrine in adrenal insufficiency. Am JPhysiol 246:H104—H113, 1984 6. BURNIER M, BIOLLAZ J, BRUNNER DB, BRUNNER HR: Blood pressure maintenance in dehydrated awake rats: Renin-vasopressin and sympathetic activity. Am J Physiol 245:H203—H209, 1983 7. ISHIKAWA S, GOLDBERG JP, SCHRIER DM, AISENBREY GA, SCHRIER RW: Interrelationship between subpressor effects of vaso-

vascular smooth muscle cells from adult human, monkey and rabbit in primary culture and in subculture. Cell Tissue Res 177:503—522, 1977 20. GRYNKIEWICZ G, POENIE M, TSIEN RY: A new generation of Ca2

indicators with greatly improved fluorescence properties. J Biol Chem 260:3440—3450, 1985

21. LOWRY OH, ROSEBROUGH NJ, FARR AL, RANDALL RJ: Protein

measurement with the Folin phenol regent. J Biol Chem 193: 265—275, 1951

54

Caramelo et al: Hormone interaction and calcium

22. TAKEDA K, MEYER-LEHNERT H, KIM JK, SCHRIER RW: Effects of

angiotensin II on Ca2 kinetics and contraction in cultured rat glomerular mesangial cells. Am J Physiol 254:F254—F266, 1988 23. ALTIN JG, BYORAVE FL: Synergistic stimulation of Ca2' uptake by

glucagon and Ca2' mobilizing hormones in the perfused rat liver. Biochem J 238:661—663, 1986 24. ALTIN JG, BYORAVE FL: Ca2 uptake stimulated by the synergistic

action of glucagon and Ca2 mobilizing agents in the perfused rat liver occurs through the activation of a unidirectional Ca2 influx pathway. Biochem Biophys Res Comm 142:745—753, 1987 25. NABIKA T, VELLETRI P, LOVENBERG W, BEAVEN MA: Increase in

cytosolic calcium and phosphoinositide metabolism induced by angiotensin II and [Arg] vasopressin in vascular smooth muscle cells. J Biol Chem 260:4661—4670, 1985 26. WARE JA, SMITH M, SALZMAN EW: Synergism of platelet-aggregating agents. J Clin invest 80:267—271, 1987

27. CHUEH SH, GILL DL: Inositol 1,4,5-trisphosphate and guanine nucleotides activate calcium release from endoplasmic reticulum via distinct mechanisms. J Biol Chem 261:13883—13886, 1986

28. DAWSON AP, COMERFORD JG, FULTON DV: The effect of GTP on

inositol I ,4,5-trisphosphate-stimulated Ca2 efflux from a rat liver microsomal fraction. Biochem J 234:311—315, 1986 29. SAIDA K, VAN BREEMEN C: GTP requirement for inositol 1,4,5trisphosphate-induced Ca2 release from sarcoplasmic reticulum in smooth muscle. Biochem Biophys Res Commun 144:1313—1316, 1987

30. MURAD F, RAPOPORT RM, FIsoNs R: Role of cyclic GMP in relaxation of vascular smooth muscle. J Cardiovasc Pharmacol 7:5111—5118, 1985

31. DUNLOP ME, LARIUNS RG: GTP and inositol I ,4,5-trisphosphateinduced release of 45Ca2 from a membrane store co-localized with pancreatic-islet-cell plasma membrane. Biochem J 253:67—72, 1988 32, ARNOLDA L, MCGRATH BP, COCKS M, JOHNSTON CI: Vasocon-

strictor role for vasopressin in experimental heart failure in the rabbit. J Clin Invest 78:674—679, 1987 33. ZIMMERMAN JB, ROBERTSON D, JACKSON EK: Angiotensin IInoradrenergic interactions in renovascular hypertensive rats. J Gun Invest 80:554—457, 1987