J Mol Cell Cardiol

The Role

Michael Dgartment

23, 1303-1312 (1991)

of Calcium

J. Daly, of Medicine,

in the Toxic E4kt.s Adult Rat Cardiac Richard

J. Young,

of Tert-Butyl Myocytes

Scott L. Britnell

Hydroperoxide

and Winifred

University of Melbourne, Austin Hospital, Heidelberg,

on

G. Nayler.

Victoria 3084, Australia

(Received 18 March 1991, accepted in revisedfonn 9 Ju& 1991) M. J. DALY, R. J. YOUNG, S. L. BRITNELL, AND W. G. NAYLER. The RoleofCalcium in theToxic EffectsofTertButyl Hydroperoxide on Adult Rat Cardiac Myocytes. Jouml of Molecular and Ctlluhr Cardiology (1991) 23, 1303-1312. Oxidant stress has been implicated in reoxygenation damage following hypoxia and can lead to loss of membrane integrity and cell death. In this study the effects of oxidant stress, induced by tert-butyl hydroperoxide myocytes isolated from rat (tBHP), on cell conformation and intracellular free calcium ([Ca’+ 1;) o f cardiac ventricles were examined. Incubation in the presence of 1 mM tBHP lead to a rise in [Ca’ +I;. hypercontracture and loss of membrane integrity (as judged by trypan blue stairling and loss of fluorescence of furaloaded cells). Incubation in calcium-free medium or medium containing 2,3 butanedione-monoxime (BDM), which decreases myofibrillar calcium sensitivity, delayed but did not prevent the cell shape changes and loss of membrane integrity. In the presence of BDM, hypercontracture occurred at a higher [Ca*+ 1; than in control cells, indicating a possible role for [Ca2+li in the generation of hypercontracture in this model. Treatment with calcium antagonists (10 - 6 or 10 - ‘M nisoldipine or 10 - 6M amlodipine) did not afford any protection against tBHP. ATP depletion did not accelerate loss of membrane integrity. Pretreatment of cells with the iron chelator, desferrioxamine mesylate greatly attenuated the effect of tBPH, delaying the rise in [Ca’ ‘Ii, cell shape changes and loss of membrane integrity. It appears, therefore, that tBHP-induced changes are mediated by the iron dependent generation of butyl alkoxyl radicals. The evidence suggests that tBHP-induced contractwe is [Ca’ + 1; dependent rather than ATP dependent. Calcium modifies, but is not essential for the action of tBHP on isolated myocytes. During reoxygenation of hypoxic hearts calcium overload and free radical generation may act synerincluding loss of sarcolemmal gistically resulting in the characteristic changes associated with this condition, integrity. KEY WORDS:

Rat

cardiac

myocytes;

Oxidant

stress;

Calcium;

Introduction Reperfusion, or reoxygenation, following extended periods of ischaemia or hypoxia exacerbates myocardial cell injury, hypercontracture and loss of membrane integrity [l-3]. Free radical generation has been demonstrated to occur during ischaemia and to be greatly enhanced upon reperfusion [4,5]. These oxyradicals, it has been suggested [4-61 may be responsible, in part at least, for reperfusion injury. Treatment of cardiac preparations with free radical generating systems results in loss of contractile function and structural changes similar to those seen during Abbreviations tBHP, tert-butyl monoxime; BSA, bovine N[2-hydroxyethllpiperazine-N’-2-ethanesulfonate. Please Hospital,

address all Heidelberg,

0022.2828/91/111303

hydroperoxide; serum albumin;

correspondence Victoria 3084, + 10$03.00/O

[Ca*

to: M. J, Australia.

hydroperoxide;

Contracture.

ischaemia and reperfusion [ 7-91. In isolated cardiac myocytes agents which induce oxidant stress have been shown to induce myocyte injury leading to altered calcium homeostasis, loss of ATP, glutathione depletion and eventually loss of viability [IO-131. The mechanism by which these agents exert their effects is not, however, fully understood. Recently Timerman et al., [12] demonstrated that glutathione depletion pm se does not result in loss of cellular ATP or membrane integrity. They suggest that external oxidant stress can induce cell damage by two mechanisms; a hypercontracture that correlates with ATP loss

‘I;, cytoplasmic CCCP, carbonyl

Daly,

Tert-butyl

Department

calcium concentration; BDM, cyanide m-chlorophenylhydrazone;

of Medicine.

University

2, 3 butanedioneHepes.

of Melbourne,

0 1991 Academic

Austin

Press

LImited

1304

M. J. Daly et al.

and lipid peroxidation which results in loss of sarcolemmal integrity. In the present study we have examined the possible role of calcium and calcium myofibrillar interactions in the loss of membrane integrity during oxidant stress induced by tert-butyl hydroperoxide (tBHP), a stable substrate of glutathione peroxidase. tBHP also serves as a substrate for the iron dependent formation of t-butyl alkoxyl radicals by the Fenton reaction. The formation of these radicals can be prevented by chelating a cellular pool of ferric iron with agents such as desferrioxamine (DFO).

Methods Ventricular myocytes were isolated from adult female Sprague-Dawley rats by the method of Piper et al., [II]. Briefly, the rats were anaesthetised with a diethylether/air mixture. Their hearts were rapidly excised, placed in cold Krebs solution containing 1lOmM NaCl, 2.6mM KCl, 1.2mM KH2PO+, 1.2mM MgS04, 25 mM NaHCOs, 5 mM Na pyruvate and 11 mM glucose. They were then perfused in the Langendorff mode for approximately 5mins with calcium free Krebs solution containing lmg/ml bovine serum albumin (BSA) (Sigma, St Louis MO, USA) gassed with 95 % 02-5s CO1 at 37%. Collagenase (Type 1, Worthington, Freehold, NJ, USA) and CaC12 were added to a final concentration of approximately 0.5-l mg/ml and 25 PM respectively. The perfusion was continued for 30 mins with recirculated perfusate. The hearts were then removed from the cannula. The aorta, atria and valves were discarded and the ventricles teased apart with scalpel blades. Ventricular tissue was then incubated for 20 min at 37% in collagenase-Krebs solution containing approximately 8 mg/ml BSA. The suspension was centrifuged gently for 2 min and the pellet washed twice with Krebs. The cells were suspended in Krebs solutions containing lOmg/ml BSA and the CaC12 concentration raised in four steps to 525 PM. The cell suspension was then layered over 4% BSA (Fraction V, St Louis MO, Sigma USA) in Krebs containing 1 mM CaCls and the loose pellet formed was suspended in Ml99 (Flow Laboratories, Irvine, Scotland, UK) plus 4% foetal calf serum (Commonwealth Serum

Laboratories, Melbourne, Australia). The suspension was then plated on 6 well tissue culture plates (Greiner, Labortecknik, Niirtingen, Germany) some of which contained 32 mm glass coverslips pretreated with laminin (Sigma, St Louis, MO, USA). They were then incubated in 5 % C02/air mixture at 37% overnight to allow the cell to attach to the surface.

Cell shape quantiJication Ml99 medium was removed from the culture plates which were washed with a Hepes buffered medium to remove unattached cells. The Hepes buffer contained 118mM NaCl, 4.8mM KCl, 1.2mM KHzPO,, 1.2mM MgS04, 1.25mM CaC12, 11 mM glucose, 5mM pyruvate and 25mM and 25mM NaN[2-hydroxyethyllpiperazine-N’-2-ethanesulfonate (Hepes). The pH was adjusted to 7.4 with HCl. The cells were incubated for 15-20mins in a 37’C air incubator in Hepes medium. Tert-butyl hydroperoxide (tBHP) (Sigma, St Louis MO, USA) in Hepes medium was added to the well to a final concentration of 1 mM. After O-40 min 1% glutaraldehyde and trypan blue were added to stop cell shape changes. This was carefully removed and replaced with normal Hepes. Under a microscope, cells were examined and categorized as follows; rod shaped (length to width ratio of 3:l or greater), square shaped (less than 3: 1 or greater than l:l), and round (either obviously round or with a ratio less than or equal to 1: 1). The cells were also categorized by their ability to exclude trypan blue.

Cell calcium measurement Coverslips with cells attached were placed in Hepes containing medium. Fura-P/AM (Molecular Probes, Eugene, Oregon, USA) was added to final concentration of 2pM and the cells incubated for 30 min at room temperature. Excess fura-2lAM was removed and the coverslips washed three times with Hepes containing medium and allowed to stand at room temperature for 2 h before measurements were commenced. Coverslips with cells attached formed the base of a tissue chamber through which warm medium (36-37%) was continuously superfused. After a period of equili-

Calcium

and Oxidant

bration the cells were superfused with medium containing 1 mM tBHP. The fluorescence was measured from a circular area (62.5 p in diameter) centred on the cell (i.e. 40-50s of total cell area). Fluorescent measurements were made at 340nm and 380nm every 30 s with a Zeiss epifluorescent microscope with photometry attachment (Zeiss, Oberkochen, Germany). Cytoplasmic free calcium concentration ([Ca* +I;) was estimated from the fluorescence at these two wavelengths.

Statisticul analysis Results were by Student’s t-test or one-way analysis of variance followed by Fisher’s protected least significant difference test. Results are expressed as mean f S.E. of n experiments, with P= 0.05 as the limit of significance.

Results Efect of t-B& Hydroperoxide (tBHP) Calcium tolerant adult rat ventricular myocytes attached to culture plates were predominately rod shaped (72.3 f 1.6%) and excluded trypan blue (less than 10% of the total were stained). Addition of 1 mM tBHP led to a time dependent loss of rod shaped cells

H 5 i-B a

1305

Stress

beginning at 10 min and being almost complete by 30 min (Fig. 1). This was accompanied by a time dependent increase in round cells (Fig. 1). The percentage of square-shaped cells was not greatly increased at any time but fell to low levels at 30min. Trypan blue staining increased dramatically after 25min of incubation (Fig. 1). Fluorescent measurements were made from rod-shaped, fura 2 loaded cells. Stable fluorescent measurements could be made from control cells for at least 60min. Continuous superfusion with 1 m tBHP led to a slowly developing increase in resting [Ca’ + 1, [Fig. 2(a)]. Mean time to an irreversible increase of 10-15s in [Ca2+li was 618 f. 20.6 s. [Ca*+]; continued to increase until total fluorescence (both wavelengths) suddenly decreased by 40-80s indicating a leakage of furaand a loss of sarcolemmal integrity. In some cells before this dramatic fall in total fluorescence, an increase in total fluorescence could be seen. Observation of the cell revealed that this increase in fluorescence at both wavelengths coincided with hypercontracture of the cell so that all of the cell or at least a greater proportion of the cell was within the measurement area. This hypercontracture was observed in 75% of control cells. Plotting [Ca*+]i against the percentage of cells hyper-

__O-

Rods

-

Rounds

*,

Stained

40

20

Time

FIGURE 1. Changes in cardiac morphology tBHP. The percentage of rod-shaped, round-shaped for clarity. Values are mean f S.E n = 6-15.

and

(min)

trypan blue staining during and trypan blue stained cells

incubation are shown.

in the presence of lm~ Square cells are excluded

1306

M. J. Daly et al.

Time (sl

TT

2500

I 500

1 1000 Time

I!

($1

FIGURE 2. The effect of incubation in the presence of 1 mM tBHP on myocyte [Ca* * li in 3 media; 6% Ca * + -free medium ( A ), in the presence of 30 mM BDM ( q ). (a) typical recordings; (b) mean values f S.E. of time to initial rise in [Ca2+ji (open symbols), hypercontracture (shaded symbols) and toss of total fluorescence @lied symbols). @ Si~ificautly different from control with respect to time, PKO.01. l * Significantly different from control with respect to [Ca2’li, PCO.01.

occurred within a relatively narrow band of [Ca2*li (Fig. 3). In the remaining cells, no increase was seen before the loss of fluorescence although the signal at 340nm rose. Presumably, hypercontracture followed by loss of fluorescence occurred within the 30 s interval between measurements in these cells. After the loss of fluorescence all cells were hypercontracted. In 90% of cells in which increased total fluorescence (hypercontracture) was recorded, decreased fluorescence {loss of membrane integrity) was recorded at

control

medium

development

the next measurement (30 s later). [Ca* +]i recorded immediately before fluorescence was 780 + 64nkt, {Fig. 2(b)].

of

Mean loss of ?t=21,

Incubation of cells in calcium-free medium delayed the effect of 1 mM tBHP on cell shape changes (Fig. 4). After 40 min incubation there was still a significant number of rod- and squared-shaped cells in the calcium-free

Calcium

100 v 2:: L E 8 2 s $ 0 E

and

Oxidant

1307

Stress

-

00-

60P 40-

tBHP

------o-

tBHP+BDM

a -07.5

-7.0

-6.5

-5.5

-6.0 LOG

FIGURE hypercontracture

3. Percentage of hypercontracted during incubation in 1mM

tBHP

-5.0

-4.5

[Co*+],

cells plotted against alone or 1 mm tBHP

r*

log [Ca’+]; recorded + 30 mM BDM.

immediately

d-

tBHP

-

tBHP,

__t_

tBHP+BDM

prior

to

Co’+-fref

60

**

60 Time

FIGURE 4. The effect on cell morphology ofincubation medium and in the presence of 30 mM BDM. * Significantly different from control, P-=0.05. * * Significantly different from control, PCO.01. Values are mean f s E. a = 6-15 (control), 4(Ca* ‘-free).

(min)

in the presence

of 1mM

tBHP

in control

medium.

Ca*

+ -free

4 (BDM).

group. In furaloaded cells no increase in [Ca* ‘Ii was observed in cells superfused in calcium-free buffer with 1 mM. Indeed the [Ca*+]i initially decreased slightly and remained stable at this level until total fluorescence fell dramatically (Fig. 2). This event was delayed in cells superfused with calcium-free medium (1275 f 85 s, n = 12

versus 978 f 44 s n = 21, in calcium-free and controlled cells respectively, WO.01) (Fig. 2). Effect of 2,3-butanedione-monoxime (BDM) Treatment with 30 mM BDM, which decreases the calcium sensitivity of the myofibrils, also delayed cell shaped changes (Fig. 4). Most cells

i308

M. J. Daly et al.

remained rod shaped for 25 min (61.9 and 18.6% for BDM treated and control cells respectively, WO.05). After this point no protection was seen with virtually 100% round cells in both groups. No protection against trypan blue staining was seen after this time. {Ca2+ji also rose in BDM-treated cells. Time before [Ca2+Ji began to rise irreversibly was shortened slightly by BDM treatment. However the time between the beginning of the rise and the loss of fluorescence was increased (504 +- 65.6 s, ra = 11 versus 360 f 38,5s, n = 21 for 3DM-treated and control groups respectively, PcO.05) (Fig. 2). The relationship between hypercontracture and [Ca* + ]i measured immediately before hypercontracture was shifted to the right by BDM (Fig. 3). The [Ca” + ]i (measured immediately before contracture) at which 50 % of cells hypercontracted was increased from 450 nM to 1612 nM in the presence of BDM. The mean [Ca* + 1; reached immediately before loss of fluorescence was increased (780 + 64, tt = 21 and 2220 + 436nM, n = 10 for control and BDM-treated cells respectively, P