 

...
22‘)... 3‘
5:1...20‘33523 322......
290.3:ulu‘3uiiu3i 7
33:... I112 I259... x...
‘2: $2.}... it}... ‘fﬁnr «12.5 v
2 ! V323... ...;Atl
31.3.1.2...411 5.1.0)... 1‘ {III
)kﬂ‘fﬁ. “Nu.“‘s‘slt I H ...) ....-
4 v 9 hi.- ‘2
.9312: .. . .x i \‘Ozg'izz 1‘ ‘3 I...
r .. ... . r . . .. .‘Mﬁ.§hnﬂ. u;
. a!!! .2- .
a . n ....... 21.4.Qs.n.sﬁit.lwll$\uuduob
. K5§it§otra lii‘h-) i
2h-l.l$.l.|l}l. I.
£

  

 

                      
                                 

                    

                  
           

 

 

   

   

  

. .. 1.2.1.2.}:
. . .. .. 1.313.: ....Q. t?..:.t‘.5 I; 213921
2 v .g'."’§-"¢ql ‘1. ; ‘\ 5“},I-I"! It. ... ni’i
.90». Via: .. fa: utthv‘aﬁto {Susi-3.9. 3":5\!.ulatbs§ﬁhﬁl.ﬁsi)§sn|||2«§nt l!!!
.:~Lutl . Inna-.:x>&!5=1klthsi .1ittl95‘.1.2|5$~vi gqigtti.i§r.l|$l t5.
...-€3.71 , - ~u. C‘ag‘lxkii ﬁ.$~.\)xy\il€l2i2$§l\i|31vl 1.00%..1: {.3}: nit!
t .“’:-¢F‘ ~ . 3”,,ﬂlll‘.¢: ﬁzli z)¥’§§‘ I‘v‘]5i"§ I... ‘3‘ ‘zdo‘..“ui (‘3
. 35.1.). ‘31-‘33..in .12....‘315.«...-33:642....h?v 29581:. ’2. 0:...
.5...” . 1r... :13. 33.”.IE2 ‘51,}.2242. ;K,\.)..._ iii“:l\n.u . it! h)
21.? 2.5%. .532; .. .nthﬁ:vtzv. 52.72.222.221v25u22. , 3.5.12 . 1.1.12.2; xﬁ: :3...“
. .7. .. u . . . 2. .
hhuﬁxﬁ 41... z . . a . :51 “Jifaéiﬂtvtano 92$” aiiahunuiifgsﬁsnnwWﬂan.irﬂslﬂﬂl.
.li’lgtw I. 5‘. . I E'i:-g'\r\v I ' v‘ﬁ .‘ QA‘hvh..¢ ’JQ.O“I“ ’1. Q." 13-5 4 3,} . .‘n‘A IE {‘1'
. .59. ‘5! 71 )t: Gauzil‘z. ...-ﬁxliu . .1. 3 lit... 5.93) ..n:: «51!? .2a)\n§.\. zillisttil D {tail 2 z altar-kl»...
:7..N,..ﬁa.?€l.ﬁla§. eeeeee FRI-.... .11? \ ; ...-.252. .52.: assayéakca;t¥¢lxlaki§ai:iti$ﬁt . t
.01., iqli‘fqak: 90:1.‘331‘ P. . x§ﬁhM9wlt o \ Av? . :15: :03 I54! 5?. it)? hibil I I!) u ..
3‘52; 7 3......Pwrﬁ‘ln23k‘332 ‘3'» x . Elia... a}: .59 )9 “up: n .. a: ....nthis {:19 .
22...“: ..v.5llllua.uhua¢ti 42:“... : .vzts mitigﬁﬂuwvﬂﬁﬁunviaﬁlaihlt it. it} .nit‘ai‘ih
1%”. :Ihiﬂ.u.l|k.ﬂaunt1§év§\vt.!avx.7;:- 393613 3|31:1xivatduiqlthuﬂziﬁran‘tlviiai‘i,mltinvsﬁh.|2h¥iil.iliiixl $2...»-
r..iO.I ~4th aka .: Rid-£1,345.03: I:-:$a).¢vcllu I: v» .t:}n.§.lu1:31:315ﬂv‘tétlzaixrs. .iiit’bh
3 In... 3. 3:28?! in 811:5..iiuiilkuuaﬁau; ‘3‘ .... .€~!(ﬂ§ 35’si752l S‘ﬁlﬂeSAogth-s.qii§i \l..3\
, 515;.«3‘iliacl :n. J 2.3.5:. a:‘t..xia(1x:)kuilx.,?$\4>ti utuellta..\xin 1.3%. ‘2 3...?) 99.1.3
511:5. .115?! ot a a 1:13;: 35;): a: . 2. ....(Ci is {that}. it}... ti..ti‘§t ..n .50...) xtsuhsliahkr
2,5‘.’:Fr~w°l!ri.a .avrlvaauzln 22! a . ﬁg... . ; I! ii {i 3.134. t ‘ixib! I‘ll! Clitlék“ b
it) «WEI.1.1.3.5...z.l‘....‘ﬁt;l.\.hu.lh. MN»? 3 . p aziH-g-mrliw. . dhuwnuiw.(.é§r gé’t: \.L.¢xt§2.:{:.lbtﬁ§$i K {I
bariuzriasa~urx.t.s 7.53.21}. sank I ﬁt; u ..... 1!!) :3: w . .1. at. {l‘giiz ’81 .. ..iﬁblx‘.
.zas..5.lr~v!!l 53"“, a k. x I 5.l“:v:.€1.i:|a!l ; 7!|Ut)|t93.l?. 3.4123705 .i..{vll‘\il in ...-(I- n.
J .v . ii‘iVlga!ialilﬁvt?htagi {151;};5. bl’l‘fiii a H. ... . . ("implant-I _ I 0.3
l!.¥i’zlnﬂ"vI.Ht.:.9_..i:1m.s2;L-¢£t\:hﬂ.a $3.2m.) . 5 . n IISlQI‘A. 35‘}? .352... an"?
Van"... . %!§ 1“}...-
tgg‘ lg I iii.“ ! i‘l'l
. o

.22... z. .
22.22». .2 :21}: 2.2. .732. . .
11.3.1. #3!!!~f93halia}?¢€z.f 10?)“.3szt‘
31:75.1...) _'a.1¢233~vr.:l;pk t... .iviliiil.i‘§i¥ill.uulb
..‘lluﬂ‘lfﬁl \ 11.2522 («‘9‘ I th!‘§i3:«:§it
.35.!- ..i..~x.h ... I:\.‘.‘.Ilu.¢lk9 it)!!! :{vtft‘ftlav‘n x. I )3!-
,» ‘1.‘..€..I.uul'.l\zti~ .Pala‘nilt. :3. iri.§.'lll .Jﬁhh‘iz:f~i.
I ski-XI {Vililia‘gé‘ :57 ‘30.;23I-z1vlﬂ1l I 131273 .12
. t it“s. 3 iv .11)": iv-4in..ix§ala§lisi3:ﬂwh Iii-19.! ($.31)....‘.
I it . . ; «...... :2.va ; 75.14.22..- 2
._ 1.. i~|rtlfﬂ\ Isis ‘3 n5... :.Q3§ n“
2. ....zﬁiunmuﬁ
2..

 

t .9 113.5...2:

      

     

. .

39.. {Sui-2.1.3.8“ .9 2... 11.21!

11..) ).£...?5v.‘2§)1 0 31:10:52? ... 18 luv}!

.élvzl.5v’.ttioggalitvll $§0§ul§lﬁ1ny3$ lﬂrl.h\o.l.\!\slvtl’ 9019:.“

In...)- 5....:iliv‘>:ltws\ .t.‘ttﬂr§t1¢.?:ti‘3.é!i8¢t§. 3!: h4g5 vii...

1.13:1 32.1. lvvkiv.9.§1>zl.;t.2:|¥.\vllaﬁ.!. .....rav;a.r)‘... it»; . .i)..«.\ :1 ’l ... it 5.5:

.5271?! 30....I-.;|12\L.\I\X.)\Bl: :E‘bashé. éﬁai’i ...-{ll- L .3: I‘Iéﬁatw. I
yin-wittvvnll, niggg-gtbi inxuhx7 :5. 11...».- y‘Qi‘l . i... ig‘fl‘ ‘16.}! E
1.2.2.521... I15 .33;- V 2 t ... it at}? tii“\§:~§ III. I! 22:93 p

11%;!!! 71...}. . 30:11:23.3”: 1! 3! rub}! .2 31.19))..‘2 s. t 31,3... ill-"é; \
r2...l\? .Zvni - 21$: 3. \ writ. . l! 1£5§3Q§I§IL t!!‘t§t!§l!> ‘ {r
12y|iz5vrb “1&2! a at it)?! . ‘Iti‘niEI; tlkhu‘l‘isa-i! t:-
170 . .... ‘ n I..L§lel¢4\§I\-I§!zi.l’tigl.
.n 221: .51 ‘2: xi
A: i
v

   

 

.....
tail.
:2. ~
..

..I. a
v. Q6~Ié|l " Is: ‘1 Y!-
»A...) aklrvatf 0.355.379.7132“ 2.!!!
aE.w€1.V\?~‘l§§-Vc 3. ..lJIi r3333: hr
... 3:. 92230.: {#31:}..th1b3tsi 1.115221 Elixiir‘b v
Evazilux-izinbuthzrlle INK-Jun) ;.Mvhvv~nu)~§~kw‘>.llv€~«; It 3% V . 1
... ’l‘v. . . 1.... I: ‘5 I. I: YIDI II! 0..
‘lit..7,x’liuuv}%aiuivsnixha .14. E. .. ngLilaalvlgé-l- '91}: it i
:K.«>.,!..53L $2.31.: {.3393}. . .. ivy: a y «at... 4.: 5 ...-272.8123)... .4.
s. 1332!: x.‘l.21:7::a:7‘$21§..| a) . ‘a\zv~n.\~l§$s I! . 3’39) 01.3.13...”
..175?:D)Jl.v¢ri|\:1l.33?.l3l031i‘v1359731i5.vl: .‘bxltgfvwl 42:)...13‘3 I)
$3....1215‘ irax~1i§2l~rsezioaiugl§v girtl.!‘\ {1. 10.90.“!
c nvx¥gv~vg TA‘aﬁz 3‘ 3"!)g it BIKE.‘ ’2 :!\§,- i
r V 3 7.91.: 7'22}: y§e.§i O A Elia .
2 _ glaiﬁvviﬁnrzszsz
s.
31:00.:-

     
          

  

  

~v9|3~21 1
0.;T‘IB

         

 

 

vi...» I ..
. illisahuug‘;
\ﬂ.

. . . _ .

                 

$212228!“ 2..
.
. 22.1%....

1552i)!
2302‘
2:

~...ié...:22§7 2.2
ltgﬂgﬁﬂiii‘i 7
31-.‘l’§§?1§; 1.1;.UE‘3E;
igzg§g|i ggiégg.5§§
. EQSIEEE it)
Eliiéi'VE to, it.
\EI'SIL“« Egg? itgi, ’
1.2.2.2 2222:. 22§3§2.522222222222.§§22222
;. 2.1213221?! 22223222223222.2222?!
2.32252 23222212232 2 22227722922.
22:22:22.2»;222223 2.32222222225
. i.l§s&s¥222252.21$&22s2y32222222,
22.9.2. {22222222237
. (ruin. i§\tllt
(5222221322, ...
.zzt~tlzﬁuvtx..2..2!sifi
5.2.2- 22

 

2g..5\l9)>l

nit: 2.252).! ts‘ié V

E‘t2ﬁzi‘ziii 5| 7 A 7.
aggriig’ gille..§i:§ .
ii£n$.§2. c221 . x. 2 (r . .

"21.2.1.6; ‘ i. .iiz.iv>t!i8

{I {gait}. sufrlv‘ . 2
7 .. 2.2).. it: (2.0. .
éﬁt\2t{r >21: 2I\2\\

      
            

         

   

          
    

 

. .
2 E 1 . q\ 7
2 . 0‘3" tit-It- Eti’ a
(‘1'..ﬁtlgt‘litunlsilzs ~is.§ﬁt 0592’?! « V
.23.?!‘2'. 5.1:}...‘3‘: ’9’, II E EA
7 . ..;.2. !.- .i‘ﬁsti erg; 2212222522! - 2 2.x
. 2. luau... 32:2... in}... his..2§ull%vl 2272...”:92293 2222.22. §E%§§.,§§iglti72§.§
u y ‘5 VI..." E 7 2 ill-...? Illz$7 "E u
. 1i”? .7 . in;:iiuk.:82."luupﬁ .. 57%)!“ﬂﬁl2g , 2 iéitiit E‘LEE‘ktKEE .
, . 1:25;.72724. . ..k;...§....2;.2..:s.....}w.i..22 .Fix.>!..::.. 1.22271. IuhtEEimyJw2§iicsttﬁvﬂ2§ 22v1222272>2s2§shntlﬂﬁ222222222
. . , . . .. ... . u (52‘), £§ilii\i\ﬁ|i§ﬁ\iii\\t2tE315!
.2352)... . a .z ;22.223w..2311.i En‘ai‘ilivztiti:223~>252222s222222222222
. .... is: 1. i§2n2§t§2223§22222222;...2il22222222222
I 7 . , 1 A . . . . . .~ ..x...\...r.\k it 'iii‘ii‘z§£~2\!k£i§iqu
. . . , . . 7 . . 212.3; :2EIEIE.2.2232.2.23222§.: . 2.2.22. 7 \ -
.1 2 . V. .7 7 .. : i .. ,2 2.522222222222222322221222222229U222222222
v . . .7 5.271. 7... .3 .2 :12.L , x222... ,7 {hr ‘ P .7 2 v..E:~xI}P§ﬂltE|izilbﬁ\§
22 r 2 2| .. . .. Shark. ..ux2.1..; 7~\\1...H\l.)......7. vlhr2l2llmlt1ﬁ . 2.757 . .22. . 7 . . . . ; 7 y|uL|$h\L«IO)\A|.\U[L£i\2I.
73222.2: 2 1, (1.222222222222222) .2. 2.. t 22. .22 72.222142: {.2 4 7 : 7 .; 7
|. Y» . 4.; ‘9 x§;’§ltiig 2" V 1‘ In |\I\2I 7 .

7 172,22...‘ £22.22... 72 2. ... 2t .2. x..-
: _. ; 7:2 2 ...... 371222222222§2us 2222222.:
V .: . 2- 2 211.2. 2.72.2221..272222322 ... 2 222.2222}.
L : 17:22 2. 3:121:55? .2 . 2 2:22;?!
. . . .. . ,2 ...; 22.2....

«\ii)......2t

i2: . n 7 .

 

THESIS

LIBRARY

Michigan State
University

 

This is to certify that the

dissertation entitled

Mechanism of enhancement of the arrhythmogenic
effects of digitalis by ischemia

presented by

Donghee Kim

has been accepted towards fulﬁllment
of the requirements for

Ph. 0 degree in PMJIIUIOIR’

Pharmacology and Toxicology

UR ﬁlm

Major professor

 

Date fold? 7. /¢fz

MS U is an Afﬁrmative ActiOn/Equai Opportunity Institution 0-12771

 

 

 

IV1ESI_] RETURNING MATERIALS:
Place in book drop to
LJBRARJES remove this checkout from
w your record. FINES W‘iH

 

 

 

be charged if book is
returned after the date
stamped below.

 

 

 

MECHANISM OF ENHANCEMENT OF THE ARRHYTHMOGENIC EFFECTS OF
DIGITALIS BY ISCHEMIA

By

Donghee Kim

AN ABSTRACT OF A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Pharmacology and Toxicology

I982

 

ABSTRACT

MECHANISM OF ENHANCEMENT OF THE ARRHYTHMOGENIC EFFECTS OF
DIGITALIS BY ISCHEMIA

By

Donghee Kim

Myocardial ischemia sensitizes the heart to the arrhythmogenic
effects of digitalis thereby reducing the therapeutic usefulness of
digitalis for the treatment of heart failure in the presence of myo-
cardial ischemia. Therefore, mechanisms by which this phenomenon
occurs were studied. Ligation of the left anterior descending (LAD)
coronary artery in isolated guinea-pig hearts reduced the time to
onset of arrhythmias during perfusion with a toxic concentration of
digoxin. The period of occlusion was less than 80 min in all pre-
parations. Kinetic parameters of (3H)ouabain binding to Na,K-ATPase,
a putative receptor for the glycoside, and the enzyme activity itself,
were unchanged by 2-hr of partial or complete ischemia. Thus, the
enhanced sensitivity of the LAD artery ligated guinea-pig heart to
digitalis appears to be due to mechanisms other than an altered
glycoside binding to the enzyme. LAD artery ligation in anesthetized
cats also reduced the dose of digoxin required to produce arrhythmias.
Removal of the autonomic nervous system input to the heart by bila-

teral vagotomy and spinal section or by propranolol pretreatment

Donghee Kim
failed to influence the enhanced sensitivity of LAD artery ligated
animals to digitalis.

The direct effects of ischemia on the heart that might enhance
the arrhythmogenic actions of digitalis were further examined using
Langendorff preparations of guinea-pig heart. Reserpine or cimetidine
pretreatment failed to abolish the increased digitalis sensitivity
caused by LAD artery ligation. Infusion of a high (10.6 mM) K+
solution, but not a 5.8 mM K+ solution into the LAD coronary artery
and perfusion of the rest of the heart with digoxin significantly
enhanced the arrhythmogenic action of the glycoside. These results
suggest that the ischemia-induced increase in digitalis sensitivity
may be due to an interaction between K+ ion and digitalis in certain

areas of the heart.

to my lovely wife, Sachiko

11'

ACKNOWLEDGEMENTS

I sincerely thank Dr. Tai Akera for his invaluable guidance
during the graduate training. I thank him for his kindness, support,
constructive criticism, advice, patience and concern.

I sincerely appreciate the advice and criticisms offered by Drs.
Theodore M. Brody, Kyosuke Temma, and Richard H, Kennedy. I also
would like to thank the other members of my graduate committee, Drs.
Gerald L. Gebber, Gregory D. Fink and Lynne C. Weaver for their advice
and friendship. I deeply value the friendships of Josh Berlin, Eddie
Ng, Richard Kennedy, Sue Glashower, Dr. Yumi Katano, Paul Stemmer,
Annie Han, and Cathy Berney.

I thank Katie Hilliker and Bonnie Baranyi for their help during
the years of my graduate training, and Diane Hummel for her excellent
secretarial help to complete my thesis dissertation.

I sincerely thank my parents, brother and sister for their love

and constant emotional support.

TABLE OF CONTENTS

Page
DEDICATION ------------------------------------------------------- ii
ACKNOWLEDGEMENTS ------------------------------------------------- iii
LIST OF TABLES --------------------------------------------------- iv
LIST OF FIGURES -------------------------------------------------- vii
INTRODUCTION ----------------------------------------------------- l
A. General Background ------------------------------------- l
B. Effects of Ischemia on the Direct Effects of Digitalis
on the Heart ------------------------------------------- 2
l. Mechanisms of Inotropic and Toxic Actions of
Digitalis ----------------------------------------- 2
2. Factors involving Na,K-ATPase that may enhance
digitalis toxicity -------------------------------- 5
C. Effects of Ischemia on the Indirect Action of Digitalis
on the Heart ------------------------------------------- 8
l. Role of Autonomic Nervous System in Toxic Actions
of Digitalis -------------------------------------- 8
2. Factors that may enhance Digitalis Toxicity ------- 10
D. Release of Endogenous Substances that may enhance Digi-
talis Toxicity ----------------------------------------- 12
l. Catecholamines ------------------------------------ l2
2. Histamine ----------------------------------------- l4
3. Potassium ion ------------------------------------- l5
E. Objectives -------------------------------------------- - 18
MATERIALS AND METHODS -------------------------------------------- l8
A. Materials ---------------------------------------------- l8
B. Isolated Heart Studies --------------------------------- l9
C. Whole Animal Studies ----------------------------------- 22
D. Plasma and Myocardial Digoxin Concentration Studies---- 24
E. (3H)Ouabain Binding Studies ---------------------------- 25
F. Na, ,K- ATPase Studies ------------------------------------ 28
G. K+ or8 6Rb+ Uptake Studies --------------------------- 29
H. Nerve Activity Studies --------------------------------- 30
I. Miscellaneous Methods ---------------------------------- 32
J. Statistical Analysis ----------------------------------- 32

iv

TABLE OF CONTENTS (continued)

Page
RESULTS ---------------------------------------------------------- 33
A. Ischemia-Induced Enhancement of the Arrhythmogenic Ac-
tions of Digitalis: Effects of Ischemia, and Reperfu-
sion on Cardiac Sarcolemmal Na,K-ATPase Activity, on
Kinetic Parameters of Na,K—ATPase for (3H)Ouabain
Binding and on Sodium Pump Activity -------------------- 33
1 Isolated Heart Perfusion Studies ------------------ 33
2. Na,K—ATPase Studies ------------------------------- 37
3. gngOuabggn Binding Studies ----------------------- 45
4. K or Rb+ Uptake Studies ---------------------- 51
B. Ischemia-Induced Enhancement of Arrhythmogenic Actions
of Digitalis: Involvement of the Sympathetic Nervous
System-----------------4 ------------------------------- 55
1. Whole Animal Studies ------------------------------ 55
2. Plasma and Myocardial Digoxin Concentration
Studies ------------------------------------------- 63
3. Nerve Activity Studies ---------------------------- 66
C. Ischemia-Induced Enhancement of the Arrhythmogenic Ac-
tions of Digitalis: Involvement of Endogenously Re-
leased Substances and Non-uniform Digitalis Effect ----- 7l
l. Isolated Heart Studies: Involvement of Catechol-
amine and Histamine ------------------------------- 72
2. Isolated Heart Studies: Non-uniform Digitalis
J Effect and K+ ------------------------------------- 75
3. (3H)0uabain Binding ------------------------------- 78
DISCUSSION ------------------------------------------------------- 81

A. Mechanisms for Toxic Actions of Digitalis on the Heart- 8l
B. Effects of Ischemia on Na,K-ATPase Activity, Sensiti-
vity of the Enzyme to the Inhibitory Effects of Digi-

talis and Glycoside Binding to Na,K-ATPase ------------- 85
C. Role of Sympathetic Nervous System in Ischemia-Induced
Enhancement of Arrhythmogenic Effect of Digitalis ------ 96
D. Role of Ischemia-Induced Release of Catecholamine,
Histamine and K+ in Digitalis Toxicity ----------------- lOO
SUMMARY AND CONCLUSIONS ------------------------------------------ l08
BIBLIOGRAPHY ----------------------------------------------------- l09

LIST OF TABLES

Table Page
1 Effects of global ischemia on kinetic parameters of
Na,K-ATPase for (3H)ouabain binding reaction ----------- 47
2 Concentrations of K+ and Na+ affecting specific binding
of (3H)ouabain to Na,K—ATPase during ischemia ---------- 50
3 Plasma digoxin concentrations at the onset of digoxin-
induced ventricular arrhythmias in coronary artery
occluded and non-occluded anesthetized cats ------------ 65

vi

Figure

l

LIST OF FIGURES
Page
Effects of digoxin on the isometric twitch tension

curves observed in Langendorff preparations of guinea—
pig hearts 35

 

Effects of LAD artery ligation on the time of digoxin
perfusion required to produce arrhythmias -------------- 36

Effects of 2-hr global ischemia on Na,K-ATPase activity
and its inhibition by dihydrodigoxin -------------------

Effects of 6-hr global ischemia on Na,K—ATPase activity
and its inhibition by dihydrodigoxin -------------------

Effects of reperfusion on Na,K—ATPase activity and its
inhibition by dihydrodigoxin

 

A cross-section of the heart showing stained and non-
stained ventricular muscle following infusion with a

dye at the onset of digoxin-induced arrhythmias in

LAD artery ligated cats 43

 

Effects of ischemia on Na,K-ATPase activity in ventri-
cular muscle homogenates obtained from anesthetized
cat hearts 44

 

Effects of ischemia on the ATP-dependent (3H)ouabain
binding to ventricular muscle homogenates obtained from
anesthetized cat hearts 48

 

Effects of ischemia and reperfusion on 86Rb+ uptake in
ventricular slices of guinea—pig heart ----------------- 52

Effects of ischemia on the specific 42K+ uptake by
ventricular strips of guinea-pig heart ----------------- 53

Electrocardiogram and femoral arterial blood pressure
during digoxin infusion in coronary artery occluded and
non-occluded anesthetized cats 56

 

Effects of LAD artery ligation on the dose of digoxin
required to produce arrhythmias in neurally intact or
bilaterally vagotomized anesthetized cats -------------- 60

LIST OF FIGURES (continued)

Figure

13

20

 

 

 

Page
Effects of isoproterenol-HCl on the heart rate in the
absence and presence of dl-propranolol-HCI in anesthe—
tized cats 6l
Effects of LAD artery ligation on the dose of digoxin
required to produce arrhythmias in C—l sectioned or dl-
propranolol-treated anesthetized cats ------------------ 62
Effects of LAD artery ligation on plasma digoxin con-
centration in anesthetized cats 64
Effects of LAD artery ligation on myocardial digoxin
uptake 67
Effects of LAD artery ligation on inferior cardiac
sympathetic nerve activity in anesthetized cats -------- 68
Effects of LAD artery ligation on the time of digoxin
perfusion required to produce arrhythmias in Langen-
dorff preparations of guinea-pig hearts ---------------- 74

Effects of non-uniform digoxin perfusion with or with-
out local high K+ infusion on the time to the onset of
arrhythmias 77

 

Effects of non—uniform digoxin perfusion on the speci-
fic ( H)ouabain binding to ventricular muscle homogen—
ates 79

 

viii

 

INTRODUCTION

A. General Background

A group of cardiac glycosides and aglycones that are found in the
digitalis leaves, strophanthus seeds and squill bulbs are collectively
referred to as digitalis. Although digitalis has been in use for many
hundreds of years, the first written account of its beneficial effects
in cardiac insufficiency was that by William Nithering in 1785. Later
in 1912, Herrick advocated digitalis therapy in acute myocardial
infarction as a means of improving cardiac function. Early clinical
observations, however, indicated that treatment of myocardial infarc—
tion with digitalis was largely associated with the development of
ventricular arrhythmias. These clinical observations prompted several
investigators to examine the possibility that myocardial infarction
may sensitize the heart to digitalis, resulting in toxicity.

The earliest of such studies in experimental animals demonstrated
that the arrhythmogenic doses of digitalis preparations were signifi-
cantly reduced in coronary artery occluded animals compared to that in
non-occluded animals (Bellet §t_al,, 1973; Travell §t_al,, 1938). In
another series of studies, experimental animals with acute myocardial
ischemia (Kumar gt 11., 1970; Ku and Lucchesi, 1979) as well as
chronic myocardial infarction (Hood §t_al,, 1967; Morris gt al.,

1969; Moss gt_al., 1981) showed greater sensitivity to the arrhythmo-

genic effects of digitalis.

2

In addition, pathologic conditions producing hypoxemia, an
important component of ischemia, were also observed to enhance digi-
talis-induced rhythm disturbances (Baum et 31,, 1959; Hargreave, 1965;
Harrison, 1968; Williams gt_al., 1968; Beller gt_al., 1971; Mason gt
al,, 1971; Friedman gt_al,, 1972; Beller and Smith, 1975). Thus, in
patients with acute or chronic pulmonary insufficiency with or without
cor pulmonale, or in experimental animals with artificially—induced
acute hypoxemia, the sensitivity to the arrhythmogenic actions of
cardiac glycosides were significantly increased.

It appears then that ischemic conditions enhance the arrhythmo—
genic potential of digitalis. The mechanism of these phenomenon is
not completely understood at present. It is of clinical significance
to understand how ischemia sensitizes the heart to the arrhythmogenic
activity of digitalis since these drugs are of immense value in the
treatment of certain types of heart failure and arrhythmias, such as
congestive heart failure, in acute pulmonary disease, and in supra-
ventricular tachycardia with or without myocardial infarction (Selzer,
1965; Karliner and Braunwald, 1972). In the present study, attempts
were made to elucidate the mechanism by which ischemia enhances the
toxic (arrhythmogenic) actions of digitalis.

B. Effects of Ischemia on the Direct Effects of Digitalis on the
Heart

 

1. Mechanism of inotropic and toxic actions of digitalis

 

In 1938, Cattell and Gold demonstrated that digitalis in-
creased the force of contraction of isolated, electrically-driven cat

papillary muscle. Since then, the direct positive inotropic action of

3
digitalis on the heart has been firmly established. It was further
observed that cardiac glycosides specifically inhibit the active
transport of sodium and potassium ions across the cell membrane
(Schatzmann, 1953; Glynn, 1957). The ATP-dependent sodium-potassium
ATPase, also known as the sodium pump was highly sensitive to the
inhibitory effects of the cardiac glycosides (Skou, 1957; Glynn, 1964;
Skou, 1965). The amount of cardiac glycosides bound to Na,K—ATPase
was demonstrated to strongly and inversely correlate with Na,K-ATPase
activity (Matsui and Schwartz, 1968; Hansen gt_al,, 1971). These two
different effects by digitalis, sodium pump inhibition and positive
inotropy, have therefore resulted in speculation that they may be
causally related.

Further studies on the relationship between Na,K-ATPase
inhibition and positive inotropy produced by digitalis have shown that
the two effects are positively and temporally correlated (Repke, 1964,
1965; Tobin and Brody, 1972; Akera et_al,, 1969; Allen and Schwartz,
1969; Hougen and Smith, 1978). Therefore, it is generally accepted
that inhibition of Na+,K+-ATPase is the mechanism by which cardiac
glycosides produce their positive inotropic effects, and that Na,K-
ATPase is the putative pharmacological receptor for the inotropic
action of digitalis (Repke, 1964; Schwartz §t_al,, 1975; Akera and
Brody, 1977). Other mechanisms for the positive inotropic actions of
digitalis have been proposed (Dutta gt al., 1968; Schwartz, 1976).
Schwartz (1976) has proposed that digitalis binds to Na,K-ATPase and
alters the affinity of enzyme-associated lipids for Ca++. Dutta gt

31. (1968) proposed that Na,K-ATPase transports digitalis to

4
intracellular space. These mechanisms for the positive inotropic
effect of digitalis, however, have received little support.

Further investigations of the relationship between digi-
talis-induced Na+,K+-ATPase inhibition and subsequent positive ino-
tropic effect have revealed that a transient increase in intracellular
sodium concentration may be responsible for the observed positive
inotropic effect following enzyme inhibition. For example, increasing
the intracellular sodium concentration by sodium ionophores, veratrum
alkaloids, or grayanotoxins, similar to the outcome of sodium pump
inhibition, was followed by a positive inotropic effect (Horackova and
Vassort, 1974; Akera et_al., 1976; Howard et_§1,, 1976; Ku et_al,,
1977).

The relationship between digitalis cardiotoxicity and Na,K-
ATPase inhibition has not been as well defined as that of the positive
inotropic effect and enzyme inhibition. For example, the arrhythmo—
genic actions of cardioactive glycosides could be dissociated from
their inotropic actions. The percent increases in contractility by
various glycosides did not correlate with the percent development of
arrhythmias (Haustein and Hauptmann, 1974). It is generally accepted,
however, that an excessive sodium pump inhibition is responsible for
the toxic effects of cardiac glycosides. Toxicity occurs when the
reserve capacity of the sodium pump is exhausted and further inhi—
bition of the sodium pump results in accumulation of Na+ and loss of
K+ (Akera and Brody, 1977; Akera and Brody, 1982). Akera et_a1.
(1976) have observed that an approximately 60% inhibition of sodium

pump occurred with a toxic dose of digitalis whereas a lesser

5
inhibition (40%) occurred with a positive inotropic dose. The in-
volvement of Na,K-ATPase in digitalis toxicity is further supported by
the findings that K+ was lost to the extracellular space with toxic
doses of digitalis (Langer and Serena, 1970; Lee and Klaus, 1971).
Such large increases in extracellular K+ and the resulting electro—
physiological changes were therefore proposed to be responsible for
digitalis toxicity (Langer, 1972). Thus, a K+-induced reduction in
the membrane potential, which may reach the threshold level for spon—
taneous firing has been proposed to be responsible for the arrhythmo-
genic actions of digitalis. More recent studies have demonstrated
that the digitalis—induced arrhythmias can be developed from delayed
afterdepolarizations which are observed with a toxic concentration of
digitalis (Ferrier gt_al:, 1973; Rosen et_al,, 1975; Ferrier, 1977).
Although the exact ionic mechanism by which delayed afterdepolari-
zations occur is not known, it is well established that digitalis-
induced inhibition of the sodium pump is the first step in the pro-
cess. Thus, the toxic or arrhythmogenic effects of digitalis reSult
from sodium pump inhibition; however, the steps following the sodium
pump inhibition appear to be very complex.

2. Factors Involving Na,K-ATPase that May Enhance Digitalis
Tbxicity

 

Ischemia may affect glycoside binding to the sarcolemmal
Na,K—ATPase in several ways. It may alter the affinity of the binding
sites on the enzyme for the glycoside, the number of functioning
enzyme units, or both. It has been shown that ischemia or hypoxia

. . . . . +
1ncreases the membrane permeab111ty to certa1n 1ons such as K and

6
Ca++ (Nayler gt al,, 1979; Nayler, 1981; Shine, 1981). Such effects
will alter the ionic environment immediately surrounding the Na,K—
ATPase. Since various ions determine the activity of the enzyme and
therefore the turnover rate, the affinity for the glycoside will
concomitantly change (Clausen and Hansen, 1977). It has been well
demonstrated that stimulation of the sodium pump by increasing the
intracellular sodium concentration either by high frequency stimu-
lation of cardiac muscle (Yamamoto §t_alf, 1979) or using pharmaco-
logical agents that specifically increase Na-K-pump activity (Clausen
and Hansen, 1977) enhances glycoside binding to Na,K-ATPase. By this
mechanism ischemia may enhance digitalis toxicity.

Alternatively, ischemia itself may cause inhibition of the
cardiac Na,K-ATPase activity and contribute to digitalis toxicity.
Beller gt_al, (1976) reported that following 2-hr of coronary artery
ligation in dogs, Na,K-ATPase activity in the partially purified
enzyme obtained from ischemic tissues was significantly reduced. Ku
and Lucchesi (1979) observed, however, that the sodium pump activity
in ventricular slices obtained from partially ischemic and completely
ischemic tissues in coronary artery ligated dog heart was increased
and was also more sensitive to the inhibitory effects of digitalis.
In other studies, no changes in Na,K-ATPase activity following a 24-hr
occlusion in dogs were observed (Schwartz gt 11., 1973).

Studies with hypoxia, an important component of ischemia,
also showed contrasting results on Na,K-ATPase activity. Hypoxic
perfusion for 15 min caused significant reduction in Na,K-ATPase

activity in the rat heart (Balasubramanian et a1., 1973) whereas a

7
much longer period (60 min) of anoxia failed to affect Na pump acti-
vity in rabbit heart (Rau gt_a1,, 1977). McDonald and MacLeod (1971,
1973) reported that in electrically stimulated guinea-pig papillary
muscle maintained at 35°C, 8 hr of anoxic perfusion had a stimulatory
effect on sodium pump activity that was responsible for the main—
tenance of the resting membrane potential in spite of a large K+
leakage to the extracellular space. These investigators further
reported that sensitivity of the sodium pump to the inhibitory effect
of ouabain was markedly greater in 8 hr anoxic when compared with 10
min anoxic tissues. This effect was attributed to the increased
sodium pump activity in 8 hr anoxic tissues. This latter finding is
supported by the results that increased sodium pump activity enhanced
the inhibitory effects of digitalis by causing greater glycoside
binding to Na,K-ATPase (Yamamoto §t_al., 1979).

Thus, the above contrasting results suggest that there may
be species-specificity with respect to the effects of ischemia on
Na,K-ATPase or sodium pump activity. If so, ischemia may or may not
alter the glycoside effect on Na,K-ATPase, depending on the animal
species. Ischemia may also reduce the reserve capacity of the sodium
pump and enhance the toxic effects of digitalis. It has been proposed
that digitalis toxicity occurs when glycoside-induced inhibition of
the sodium pump exceeds its reserve capacity, and the remaining sodium
pump activity becomes inadequate for maintaining a low intracellular
sodium ion concentration (Akera and Brody, 1977). Membrane depolari—

zation may subsequently occur and give rise to an action potential

 

8

when the membrane potential reaches the threshold potential for spon-
taneous activity.

Therefore, an increase in the affinity of the Na,K-ATPase
for glycoside, a decrease in the number of functioning enzyme mole-
cules or a reduction in the reserve capacity of the sodium pump may

enhance digitalis-induced toxicity in the heart.

C. Effects of Ischemia on the Indirect Action of Digitalis on the

 

 

Heart
1. Role of Autonomic Nervous System in Toxic Actions of
Digitalis

In addition to the direct action of digitalis on cardiac
tissues, the extracardiac effects have been shown to significantly
contribute to the toxic effects of digitalis, i.e. to the development
of ventricular arrhythmias. Although neuroexcitatory effects of
digitalis have been known for a long time, it was not until 1937 that
Korth et_al. emphasized a specific effect of digitalis to increase
central sympathetic outflow to the heart. Since then a vast amount of
research has accumulated implicating the sympathetic nervous system as
one of the causes for digitalis-induced arrhythmias.

A strong correlation between digitalis-induced increase in
sympathetic activity and the occurrence of ventricular arrhythmias was
observed by several investigators (Gillis, 1969; Gillis §t_al,, 1972;
McLain, 1969; Pace and Gillis, 1976). Procedures that interfered with
the central sympathetic outflow to the heart, such as intraventricular
administration of propranolol, abolished neurally associated cardiac
arrhythmias (Lewis and Haeusler, 1975). Using hypothalamic stimu—

lation to augment sympathetic outflow, Evans and Gillis (1975) were

 

9
able to produce arrhythmias with a dose of ouabain that would not
normally be arrhythmogenic.

Studies utilizing spinal cord transection techniques also
demonstrated the involvement of the sympathetic nervous system in
digitalis-induced arrhythmias. Raines gt_gl: (1967) observed that C-l
section increased the dose of ouabain to produce ventricular tachy-
cardia and fibrillation. Similar results were obtained by many other
investigators (Gillis gt_gl,, 1972; Cagin gt_gl,, 1974; Levitt gt_gl:,
1973; Somberg gt gt., 1978). Acute or chronic cardiac denervation in
dogs decreased the sensitivity of the heart to the arrhythmogenic
actions of digitalis (Mendez gt_gl,, 1961; Cooper gt_gl,, 1961; Solti
gt_gl,, 1965).

In addition to surgical methods, drugs that modify sympa—
thetic nervous system function have been extensively studied. Drugs
that depress central sympathetic outflow such as clonidine (Gillis gt
gl,, 1972); drugs that block cholinergic transmission, both nicotinic
and muscarinic, such as hexamethonium and atropine (Gillis gt_gl.,
1975); beta-adrenergic receptor blocking agents, such as dl-proprano—
lol (Evans gt_gl,, 1976), or sotalol (Kelliher and Roberts, 1974); or
drugs that interfere with storage and release of norepinephrine at the
postganglionic sympathetic nerve endings, such as reserpine (Boyajy
and Nash, 1965; Ciofalo gt_gl,, 1967; Nadeau and Champlaine, 1973), 6—
hydroxydopamine (Saito gt_gl., 1974), guanethidine (Raines gt gl.,
1968), or bretylium (Papp and Vaughn Williams, 1969; Gillis gt_gl,,
1973), all increased the dose of digitalis required to produce ven—

tricular arrhythmias either in dogs, cats, or guinea—pigs. In

 

10
conjunction with these results was the finding that exogenously ad-
ministered catecholamines enhanced the arrhythmogenic effect of digi-
talis (Morrow, 1967; Raper and Wale, 1969; Lum gt_gl:, 1977). These
observations strongly suggest the importance of the sympathetic
nervous system in digitalis-induced ventricular arrhythmias.

The role of the parasympathetic nervous system in digitalis-
induced arrhythmias, however, remains controversial. The vagus is
reported to have a protective effect (LoSasso and Paradise, 1969;
Levitt gt_gl,, 1970), an enhancing action (Robinson and Wilson, 1918;
Kreuger and Unna, 1942), or no effect (McLain gt gl., 1958; Boyazy and
Nash, 1966; Lechat and Schmitt, 1982) on digitalis-induced arrhyth—
mias.

2. Factors that May Enhance Digitalis Toxicity

From above studies, it may be concluded that if ischemia
affects autonomic nerve activity, the toxic actions of digitalis on
the heart may be altered. In 1967, Brown noted that coronary artery
occlusion activated sympathetic cardiac afferent nerves in cats.
Since then, a series of studies by other investigators have confirmed
this observation and further demonstrated coronary artery occlusion-
induced increases in sympathetic cardiac efferent nerve activity
(Malliani gt_gt., 1969; Uchida and Murao, 1974; Felder and Thames,
1979; Bosnjak gt_gl,, 1979). Such a sympathetic reflex was also
present in C—l spinal-sectioned animals (Malliani gt_gl,, 1969), an
observation that led to the proposal that a cardiocardiac sympathetic
reflex may be present during myocardial ischemia. In addition to

changes in cardiac efferent nerve activity, coronary artery occlusion

11

caused changes in renal nerve activity (Weaver gt_gl,, 1981). There-
fore, coronary artery occlusion, by activating afferent cardiac
sympathetic nerves, influenced efferent sympathetic nerve activity not
only to the heart, but also to other organs. On the other hand, many
investigators observed a coronary artery occlusion-induced inhibitory
influence on cardiac sympathetic nerve activity in similar animal
models (Constantin, 1963; Thoren, 1973; Kedzi gt_gl,, 1974; Felder and
Thames, 1979). Kedzi gt_gl, (1974) recorded postganglionic sympa-
thetic nerve activity and blood pressure following occlusion of
circumflex coronary artery of anesthetized dogs. Although blood
pressure decreased following occlusion, postganglionic sympathetic
nerve activity was also reduced. Cutting the vagus nerves resulted in
an immediate increase in sympathetic nerve activity and blood pres-
sure. Thus, in these experiments, coronary artery occlusion, by
activating vagal afferents, reflexly inhibited sympathetic outflow.

The reason for these disparate observations on sympathetic
cardiac nerve activity induced by coronary artery occlusion is not
clear. The area of myocardial ischemia may be an important determi-
nant of the direction of change in sympathetic nerve activity, since
sympathetic or vagal afferents in the heart may be unequally distri-
buted (Oberg and Thoren, 1973; Thoren, 1973; Uchida gt_gl,, 1974; Corr
gt_gl,, 1976). Of particular interest are the findings of Lathers gt_
gt, (1978) that coronary artery occlusion produced increases, de-
creases or no change in discharge within a bundle of postganglionic
cardiac sympathetic nerves. They therefore postulated that such non-

uniform changes in nerve activity may cause arrhythmias to develop.

12

In both human patients and experimental animals with ische—
mic heart and coronary artery ligation respectively, the catecholamine
concentration in the blood and urine were significantly elevated
(Forssman gt_gl,, 1952; Valorie gt_gl,, 1967; Klein gt_gl,, 1968;
Staszewska-Barczak and Ceremuzynski, 1968; Hayashi gt_gl,, 1969;
Siggers gt gt., 1971; Staszewska-Barczak, 1971). In many cases, the
amount of catecholamine released into the blood was closely correlated
with morphological changes in the ischemic heart. Conversely, admi-
nistration of exogenous catecholamines produced features similar to
myocardial ischemia (Raab gt_gl,, 1962; Waldenstrom gt_gl,, 1978).
These findings suggest that during the ischemic process there is
increased sympathetic activity and release of catecholamines. Such

effects may predispose the heart to digitalis-induced toxicity.

D. Release of Endogenous Substances that may Enhance Digitalis
Toxicity

It has been demonstrated that ischemia can release endogenous

 

substances such as catecholamine and histamine into the circulating
blood. These substances have been shown to produce changes in bio-
chemical and physiological functions of the heart. Ischemia also
impairs ionic movements across the cell membrane and within the cell
(Nayler gt_gl,, 1979). Therefore, electrical events during exci-
tation-contraction coupling may be depressed. These effects may alter
the toxic actions of digitalis on the heart.

1. Catecholamines

 

It is well known that catecholamines potentiate the arrhyth-

mogenic effects of digitalis. As described in section C of the

13

Introduction, beta—adrenergic blocking agents, catecholamine depleting
drugs, and sympathectomy all reduced digitalis toxicity. Consistent
with these findings are the results that exogenously administered
catecholamines potentiated digitalis toxicity. Thus, if ischemia
increases the level of catecholamines in the heart, the arrhythmogenic
actions of digitalis will be enhanced.

In addition to increases in catecholamine release by sympa—
thetic stimulation, ischemia (Lammerant gt a1., 1966; Ceremuzynski,

1969; Shahab gt gl,, 1969; Abrahamsson gt. 1., 1981) or hypoxia

(Wollenberger and Shahab, 1965; Penna gt_gl,, 1965) has also been
shown to cause local release of catecholamines from the nerve termi-
nals in isolated heart muscle.

Digitalis has also been shown to release catecholamines from
nerve terminals in isolated cardiac tissues. In isolated perfused rat
or guinea pig hearts, ouabain increased spontaneously occurring
norepinephrine release (Linmar and Loffelholz, 1974; Harvey, 1975). A
large number of experiments have also been performed to study the
effects of digitalis on the uptake of norepinephrine. These studies
have indicated that digitalis inhibited the uptake of norepinephrine
by neuronal membranes of heart tissues (Dengler gt_gl,, 1962; Berti
and Shore, 1967; Stickney, 1976; Sharma and Banerjee, 1977). Ischemia
may augment these effects and elevate digitalis toxicity. Thus,
ischemia—induced release of catecholamines in the heart or its en-
hancing effects on the release and inhibition of uptake of norepin-
ephrine by digitalis may potentiate the arrhythmogenic actions of

digitalis in the ischemic heart.

 

2. Histamine

Histamine is found in cardiac tissues of many mammalian
species including humans. Using H1 and H2 antagonists, both H1 and
H2 receptors and responses of cardiac tissues to histamine were
identified (Powell and Brody, 1976; Levi gt gl,, 1976). Although the
physiological role of endogenous cardiac histamine is uncertain, when
released from the cardiac tissues during immediate hypersensitivity
reactions (Levi, 1972), histamine produced marked changes in cardiac
function by altering contractility, conduction, automaticity and
coronary flow (Rocha and Silva, 1966; Levi, 1972). The arrhythmogenic
actions of histamine at high doses were demonstrated to be by an
action on HZ-receptors of cardiac tissues.

The interaction between digitalis and histamine to poten-
tiate their effects on atrioventricular conduction block and ventri-
cular automaticity was observed by Levi and Capurro (1974). In their
study, ouabain enhanced histamine-induced changes in the above para-
meters in isolated guinea-pig hearts. In a related study, Somberg gt
gt. (1980) found that both H1 and H2 antagonists were able to protect
the cat heart from ouabain-induced cardiotoxicity. Histamine ad—
ministration abolished the increase in lethal doses of ouabain caused
by H2 antagonists, but not that caused by H1 antagonists. Thus,
although the mechanism of the antiarrhythmogenic action of histamine
antagonists is still not understood, the above studies indicate that
histamine, when released, may augment the toxic actions of digitalis.
Thus, ischemia may sensitize the heart to digitalis—induced toxicity

by causing histamine release.

3. Potassium Ion

It has been clearly demonstrated that ischemia causes in-
creases in extracellular K+ concentration as a result of leakage from
the intracellular space (Harris gt_gl,, 1954; Hill and Gettes, 1980;
Hirche gt_gl,, 1980). Harris (1954) has proposed that such an effect
is a major cause for the ventricular arrhythmias observed early in
acute myocardial ischemia. The exact cause for the genesis of ven-
tricular ectopic activity was attributed to a local increase in K+
concentration within a normal region such that a K+ gradient was
present and a current of injury was flowing at the boundary between
ischemic and normal tissue (Harris, 1954, 1966; Hoffman, 1966; Hill
and Gettes, 1978; Hirche gt_gl,, 1980; Weiss and Shine, 1981). How-
ever, the precise mechanism by which coronary artery occlusion pro-
duced such ectopic activities has not been identified.

Several mechanisms have been proposed for the genesis of
ventricular arrhythmias in early myocardial ischemia. In parallel
with the loss of intracellular K+, the resting membrane potential of
ischemic cells was decreased (Kleber gt_gl,, 1978). It is well
established that a reduction of the resting membrane potential to
values lower than approximately -65 mV inactivates the fast inward
sodium current without affecting the slow inward current carried
mainly by calcium. It also was discovered that the slow inward
current can under certain conditions generate propagated and automatic
action potentials, called "slow responses" (Carmeliet and Vereecke,
1969; Pappano, 1970; Cranefield gt_g13, 1971; Aronson and Cranefield,

1973). In the presence of high concentrations of beta-adrenergic

 

l6

agonists which enhance the inward calcium movement and therefore the
slow inward current, slow responses were observed (Wit gt_gl., 1972;
Sherlag gt_gl,, 1974). Thus, K+-induced depolarizations may generate
such propagated ”slow” action potentials and contribute to ectopic
activities observed in acute myocardial ischemia, and catecholamines
may augment these effects of K+ (Wit and Bigger, 1975; Zipes gt gig,
1975).

In ischemic tissues, an elevation of extracellular K+
concentrations was also closely related to the shortening of the
action potential duration and lengthening of conduction time (Weiss
and Shine, 1981). Such localized changes in these parameters during
regional ischemia have been hypothesized to be an important mechanism
that sets a condition for re-entrant arrhythmias (Wit and Bigger,
1975; El—Sherif gt_gl,, 1977). In this mechanism, slow responses need
not be present for arrhythmias to occur, but the presence of non—
uniform changes in action potential duration and conduction time is
sufficient for development of arrhythmias. Thus, in ischemic hearts
with no arrhythmic activities, the addition of digitalis may help to
bring the above processes into play and cause ventricular ectopic
activities to appear.

In summary, digitalis alters ion movements across the cell
membrane and also alters ion concentrations inside the cell to cause
inotropic and toxic effects. Several possibilities exist that may
enhance digitalis toxicity in the ischemic heart. Any one or a

combination of these may be responsible for this phenomenon.

 

E. Objectives

The objective of the present study was to elucidate the mecha-

nisms responsible for the enhanced arrhythmogenic effects of digitalis

in ischemic hearts. The possible mechanisms were divided into three

major categories and each was examined.

1.

Ischemia may enhance the direct effects of digitalis on the
heart. Since digitalis has been proposed to bind to cardiac
sarcolemmal Na,K-ATPase to produce its pharmacological and
toxic effects, the possibility that ischemia alters Na,K—
ATPase activity, glycoside binding to the enzyme, or reserve
capacity of the sodium pump was studied.

The toxic (arrhythmic) effect of digitalis is enhanced by
the sympathetic nervous system. Therefore, ischemia may
alter sympathetic activity and potentiate digitalis toxi—
city. Thus, the role of the sympathetic nervous system in
ischemia-induced sensitization to arrhythmogenic actions of
digitalis was examined.

Endogenous substances such as cardiac histamine and cate-
cholamines are known to potentiate digitalis toxicity.
Enhanced digitalis toxicity observed in ischemic hearts may
be due to release of such substances. Ischemia also causes
an increase in membrane permeability to K+ and possibly to
other ions. The possibility that such changes elevate

digitalis-induced toxicity were studied.

MATERIALS AND METHODS

A. Materials
Tritium—labeled ouabain (generally labeled, specific radioacti-
vity, 14 Ci/mmol) and 3H-labeled digoxin (12a-1abeled, specific
radioactivity, 14 Ci/mmol) were purchased from New England Nuclear,
(42

Boston, MA. Potassium hydroxide K—labeled, specific radioactivity,

0.6 mCi/mmol) was prepared at the Nuclear Reactor Laboratory of

42KOH was neutralized with HCl before use.

Michigan State University.

Ouabain octahydrate (strophanthin-G), digoxin, Tris-ATP, rubidium
chloride, reserpine, a—chloralose, dl-propranolol—HCl, isoproterenol-
HCl, Patent Blue Violet dye, gallamine triethiodide, bovine serum
albumin, tyramine-HCl were all purchased from Sigma Chemical Company,
St. Louis, MO. Dihydrodigoxin was purchased from Boehringer Mannheim
Biochemicals, Indianapolis, IN. Rubidium chloride (ultrapure grade)
was obtained from Alfa Division, Danvers, MA. Other chemicals were of
analytical reagent grade.

Radioimmunoassay kits for digoxin were obtained from Clinical
Assays, Cambridge, MA. Biofluor (liquid scintillation counting solu—
tion) was purchased from New England Nuclear, Boston, MA. Nitro-
cellulose filters were obtained from Millipore Filter Corporation,
Bedford, MA type AA, pore size 0.8 um). Gas mixtures were obtained

from AIRCO, Inc., Montvale, NJ.

 

B. Isolated Heart Studies

 

An isolated perfused heart is a good model to study the direct
effects of pathophysiological or pharmacological interventions on
cardiac function. Various physiological parameters such as heart
rate, temperature or perfusion rate may be precisely controlled,
making the interpretation of data easier and accurate. In the present
study, isolated heart preparations of guinea pigs were used primarily
because preliminary results have indicated that this animal model is
suitable for this study, and for other practical reasons such as size
of the heart and availability.

Guinea pigs of either sex weighing 300-400 g were stunned by a
sharp blow to the head, and the hearts were quickly removed and
immersed in a bath containing Krebs—Henseleit bicarbonate buffer (pH
7.5) solution containing 118 mM NaCl, 27.2 mM NaHC03, 4.8 mM KCl, 1.0
mM KH2P04, 1.2 mM MgC12, 1.2 mM CaCl2 and 11.1 mM glucose, and satur-
ated with a 95% 02—5% C02 gas mixture at room temperature. The aortic
root was then cannulated and the heart was allowed to hang from a
Langendorff perfusion apparatus. The hearts were perfused at a
constant flow rate of approximately 2.5 ml/g tissue/min or at a con-
stant pressure of 20 mmHg with the oxygenated Krebs-Henseleit buffer
(pH 7.5) solution maintained at either 32°C or 36°C.

When the visible blood in the heart was washed away, both left
and right atrial muscles were excised. The hearts were electrically
stimulated at 1.5 Hz with square wave pulses of 5 msec duration at a
voltage 30% above threshold by a pair of platinum electrodes placed
near the atrioventricular node. One end of a thin silk thread was

tied to the apex of the heart and the other end was connected to a

 

20
force-displacement transducer via a pulley. Resting tension was
adjusted to 1.0 g and the force of contraction and twitch tension
curves were recorded on a polygraph recorder using a force—displace—
ment transducer (Grass Instrument Co., model 70 and FT-03C, respec—
tively).

For studies of ischemia-induced changes in digitalis sensitivity,
the hearts were equilibrated for 30 min and when no arrhythmic beats
were observed, the left anterior descending coronary artery, approxi—
mately 1 cm above the apex of the heart, was completely occluded by
passing a silk thread through the muscle surrounding the artery with a
needle, and tying the ends of the threads tightly around the artery.
The control preparations were subjected to the same procedure, but
the thread was not tied. Following coronary artery occlusion, the
hearts were perfused via the aorta for an additional 40 min. When no
arrhythmic contractions were present during this period, digoxin was
added to the perfusing solution (final digoxin concentration, 1.8 or
2.5 uM) and twitch tension curves were monitored to determine glyco—
side-induced arrhythmias. Control preparations were similarly per-
fused with digoxin solution. Approximately 5% of all heart prepara-
tions produced arrhythmic beats before or after coronary occlusion,
but not after digoxin perfusion and therefore were discarded.

In another set of experiments, guinea pigs were intraperitoneally
injected with 5 mg/kg reserpine solution 24 hr prior to sacrifice. To
test catecholamine depleting effects of reserpine, a dose-response

6

curve for the positive inotropic action of tyramine-HCl (10' - 10'4M)

was obtained in reserpine-treated and control heart preparations. In

21

histamine studies, cimetidine (10'5M), an HZ-receptor blocker, was
added to the perfusing solution 60 min prior to addition of digoxin to
block effects of endogenously released histamine. The HZ-receptor
blocking effects of cimetidine were tested by obtaining a dose-re-
sponse curve for the positive inotropic effect of histamine in the
presence and absence of cimetidine-pretreatment. The time to the
onset of digoxin-induced arrhythmias was similarly monitored from
twitch-tension recordings.

To study the non-uniform effects of digitalis on the heart, the
left anterior descending coronary artery was carefully cannulated with
a 27 gauge needle and normal Krebs-Henseleit bicarbonate buffer (pH
7.5) solution (5.8 mM K+) was infused through the needle at a constant
rate of 1.0 ml/min. After a 40—min equilibration period, digoxin was
added to the solution that perfused the rest of the coronary arteries
via the aorta. In another group of hearts, a modified Krebs-Henseleit
bicarbonate buffer containing 10.6 mM K+ instead of 5.8 mM K+, was
infused through the cannulated LAD artery. Digoxin was similarly
added to the solution perfusing the rest of the heart via the aorta
and the time to onset of arrhythmias was monitored. Heart prepara-
tions with the cannulated artery infused with a solution containing
either 5.8 mM K+ or 10.6 mM K+ without digoxin perfusion were also
monitored for comparable time periods for development of arrhythmias.
In order to confirm the non-uniform distribution of digoxin in the
heart by color staining, Patent Blue Violet dye was added to either
solution perfusing the heart via the aorta or the solution perfusing

the cannulated artery. At the end of the experiment, the heart was

22
cut in cross-section to visually identify areas of stained and non-
stained tissues.

For global ischemia studies, Langendorff preparations of guinea-
pig heart were maintained at 36°C in a humidified chamber and perfused
at control flow rate of 2.5 m1/g tissue/min. Following a 20-min
period, the flow rate was either maintained at control rate, or re-
duced to 5% of the control flow rate or the flow was completely
stopped for 2 or 6 hr. For reperfusion studies, perfusion of the
hearts were completely stopped for 2 or 5 hr and reperfused at control
flow rate for 20 min or 1 hr. For the transition from no-flow to
reperfusion, the rate of increase of flow rate was 0.5 ml/g tissue/
min. At the end of global ischemia or reperfusion, ventricular
muscles were dissected and homogenized. The homogenates were imme-
diately assayed for Na,K—ATPase activity and (3H)ouabain binding (see

below).

C. Whole Animal Studies

 

Cats of either sex weighing 1.9-3.2 kg were anesthetized by
intravenous injection of 60 mg/kg a-chloralose. A tracheostomy was
performed and the animals were artificially respired with room air
supplemented with approximately 30% oxygen. Femoral artery and vein
were cannulated with polyethylene tubing for blood pressure and drug
administration, respectively. Blood pressure and electrocardiogram
from lead II were recorded using a pressure transducer and EKG-ampli-
fier on a polygraph recorder (Statham-Gould, Oxnard, CA, model P23-ID

and Grass Instruments Co., model 70). Animal body temperature was

23
maintained at 36-38°C using a heating pad. A left thoracotomy was
then performed and the heart was exposed by cutting open the peri-
cardium. The left anterior descending coronary artery was identified
and a silk ligature was passed around the artery approximately 1 cm
below the lower tip of the left atria and left untied. The animals
were then either left neurally intact, bilaterally vagotomized, or
bilaterally vagotomized and spinal-cord sectioned at C1 or pretreated
with dl—propranolol. After 60 min, when the blood pressure became
stable, the artery was either untied (control) or ligated (LAD liga-
tion) by tying the thread around the artery.

Digoxin was first dissolved in 50% alcohol and further diluted
with 0.9% saline solution. The final alcohol content was approxi-
mately 0.5%. This digoxin solution was continuously infusedat a rate
of 60 pg/kg/hr after 40 min of LAD artery ligation. In some cats with
LAD artery ligation, 0.9% saline solution containing similar amounts
of alcohol with no digoxin was infused. For digoxin infusion in the
presence of dl-propranolol, isoproterenol (0.1, 0.3 or 1.0 pg/kg) was
infused intravenously and the changes in heart rate were noted for
each dose of isoproterenol. A loading dose of dl-propranolol (1
mg/kg, i.v.) followed by a continuous infusion of l mg/kg/hr was
started and the effect of isoproterenol on the heart rate were ex-
amined again after 30 min. This dose of dl-propranolol completely
inhibited isoproterenol-induced changes in heart rate.

At the onset of digoxin-induced arrhythmias in both coronary
artery occluded and non-occluded animals, 5 ml blood samples were

collected and then 3 ml of Patent Blue Violet dye (100 mg dissolved in

 

 

24

3 m1 of 0.9% saline solution) was infused via a femoral vein. When
the heart turned blue, it was cut in cross-section to visualize dye—
stained and non-stained areas. Darkly stained, lightly stained and
non-stained tissues were visually delineated and small pieces of each
type were cut out and immediately homogenized for biochemical assays
and myocardial digoxin uptake determinations (see below). Darkly-
stained, lightly-stained or non-stained tissues were designated as
non-ischemic (NI), partially ischemic (PI) or completely ischemic

(CI), respectively.

D. Plasma and Myocardial Digoxin Concentration Studies

 

Digoxin concentrations in plasma were determined by either radio-
immunoassay for digoxin or by assaying for radioactivity of the plasma
following an intravenous infusion or 3H-digoxin. At appropriate times
before and after digoxin infusion to anesthetized cats, blood samples
were obtained from the femoral artery and placed in heparinized tubes.
Blood samples were then centrifuged for 10 min at 2000 rpm to obtain
plasma samples which were quickly frozen at ~20°C. The plasma samples
were subsequently analyzed for digoxin using the radioimmunoassay
method (Clinical assays, Cambridge, MA; kit for digoxin). In one
series of experiments, digoxin solution containing tracer amounts of
126-3H-digoxin was infused into the animals. Plasma digoxin concen-
trations of these cats were estimated by dissolving the plasma samples
in 20 m1 of Biofluor (New England Nuclear, Boston, MA) and assaying
for radioactivity using a liquid scintillation spectrometer. At the

onset of digoxin-induced ventricular arrhythmias, Patent Blue Violet

25
dye was infused into the animals via a femoral vein. Hearts were
removed several minutes later and ischemic, partially ischemic and
non-ischemic areas of the heart were dissected after they were visu-
ally identified from the intensity of dye staining. The respective
tissues were homogenized in distilled water (1 g/ml) and a 200 pl
aliquot of each homogenate was dissolved in 1.0 ml of NCS tissue
solubilizer (Amersham/Searle). To each sample, 15 m1 of Dimilume-30
solution (Packard Instrument Company, Inc., Downers Grove, IL) was
added and radioactivity was assayed. Counting efficiency (approxi-
mately 30%) was monitored by the external standard channel ratio
method. Background radioactivity was assayed using homogenates or

plasma samples obtained from saline—infused animals.

E. (3H)0uabain Binding Studies

 

Affinity of glycoside binding sites on Na,K-ATPase for (3H)-
ouabain and the number of glycoside binding sites were estimated by
the method developed by Akera and Cheng (1977). Ventricular muscles
obtained in ischemic studies were homogenized at 0°C in 10 mM Tris-HCl
buffer (pH 7.5) containing 1 mM EDTA using a Potter-Elvehjem glass
homogenizer with a motor driven Teflon pestle to the final concen-
tration of approximately 40 mg of tissue per 1 m1 of buffer solution.
The ventricular muscle homogenates (final concentration in incubation
medium, 0.3-0.4 mg protein/ml) were added to a pre-warmed (37°C for 5
min) mixture containing final concentrations of 1 mM MgC12, 1 mM Tris-
phosphate, 10 mM Tris-HCl buffer (pH 7.5), 10 nM (3H)ouabain and
various concentrations of non-labeled ouabain (0, 20, 50, 100, 200,

500 or 1000 nM). The mixture was incubated at 37°C for 90 min to

 

26
attain equilibrium of the binding reaction. After incubation, the
(3H)ouabain binding reaction was terminated by the addition of non-
labeled ouabain (final concentration, 0.1 mM). Bound and free (3H)-
ouabain were separated by filtering the aliquot through nitrocellulose
Millipore filters. The filters were washed twice with 5 ml each of an
ice-cold solution containing 0.1 mM nonradioactive ouabain, 15 mM KCl
and 50 mM Tris-HCl buffer (pH 7.5) and dissolved in 1 m1 of ethylene-
glycol monomethyl ether. The radioactivity on the filter (bound
ouabain) was assayed using a liquid scintillation spectrometer.
Counting efficiency (approximately 30%) was monitored by the external
standard channel ratio method. Saturable ouabain binding was calcu-
lated by subtracting the non-saturable binding observed in the pre-
sence of 0.1 mM non-labelled ouabain from the total binding. From
these values, the number of specific (3H)ouabain binding sites and the
affinity of these sites for (3H)ouabain were calculated by the method
of Akera and Cheng (1977).

Fractional occupancy of the glycoside binding sites on Na,K-
ATPase in non-ischemic, partially ichemic and ischemic tissues by
digoxin during digoxin or saline infusion was estimated from the
reduction in ATP-dependent (3H)ouabain binding to the respective
tissue homogenates. Tissues were homogenized in a 10 mM Tris-HCl
buffer (pH 7.5) containing 1 mM EDTA. The homogenates were added to a
prewarmed (37°C) incubation mixture yielding final concentrations of

0.5-0.8 mg of protein per ml, 200 mM NaCl, 5 mM MgCl 50 mM Tris-HCl

2,
buffer (pH 7.5) and 20 nM (3H)ouabain with or without 5 mM Tris-ATP.

After a 2-min incubation at 37°C, the binding reaction was terminated

 

 

 

Hm— hr”, ‘_ ‘___

27
by adding a cold solution containing excess non-labeled ouabain which
inhibits further binding of (3H)ouabain. The mixture was immediately
filtered through a nitrocellulose filter (Millipore Filter Corpora-
tion, Bedford, MA; type AA, pore size, 0.8 pm) to separate bound and
free (3H)ouabain. The filter was washed with two 5-m1 aliquots of an
ice-cold solution containing 0.1 mM nonradioactive ouabain, 15 mM KCl
and 50 mM Tris-HCl buffer (PH 7.5) within 10 sec after the termination
of the binding reaction. The radioactivity trapped on the filter
(bound ouabain) was assayed using a liquid scintillation spectrometer
after dissolving the filter in ethylene glycol monomethyl ether. The
ATP-dependent (3H)ouabain binding is the difference in values observed
in the presence and absence of ATP. This value represents the binding
of (3H)ouabain to the glycoside binding sites on Na,K-ATPase in tissue
homogenates (Allen gt_gl,, 1971) and its reduction indicates the
previous occupancy of these sites by a glycoside (Ku gt_gl,, 1974).
Since the digoxin-Na,K-ATPase complex has a relatively long half-life
(Akera gt_gl,, 1974), the observed values are reasonable estimates of
fractional occupancy of the glycoside binding sites by digoxin which
takes place during digoxin infusion.

To estimate ischemic-induced changes in the effects of K+ or Na+
on the glycoside binding to Na,K-ATPase, ventricular muscle homogen-
ates were incubated in the presence of 200 mM NaCl, 5 mM M9012, 5 mM
Tris-ATP, 50 mM Tris-HCl buffer (pH 7.5) and 0-150 mM KCl (medium A),
or in the presence of 5 mM MgC12, 5 mM Tris-ATP, 50 mM Tris-HCl buffer
(pH 7.5) and 0-300 mM NaCl (medium B). (3H)ouabain binding was
assayed as above and the specific binding was calculated as the differ-

ence in values observed in the presence and absence of ATP. The

28
affinity (K0.5 value) of Na,K—ATPase for K+ or Na+ was estimated from
the concentration of K+ to cause a half-maximal inhibition of the
specific (3H)ouabain binding observed in Medium A, or from the con—
centration of Na+ to cause a half-maximal stimulation of the specific

(3H)0uabain binding in Medium B.

F. Na,K—ATPase Studies

ATPase activity of the ventricular muscle homogenates was assayed
by measuring the amount of inorganic phosphate (P1) released from ATP
(Bonting gt_gl,, 1961). Ventricular muscle homogenates were prepared
as described above. Homogenates were added to a prewarmed (37°C)
incubation mixture (final concentration of protein, 0.3—0.4 mg/ml)
containing 5 mM MgCl2 5 mM sodium azide, 50 mM Tris-HCl buffer (pH
7.5), 5 mM Tris-ATP, 100 mM NaCl, 15 mM KCl and an indicated concen—
tration of dihydrodigoxin in the presence or absence of 0.1 mM vana-
date. After a lO—min incubation, the reaction was terminated by
adding 1 m1 of an ice-cold 0.8 M perchloric acid solution. The amount
of inorganic phosphate (P1) released from ATP was assayed by adding a
color reagent and reading the light absorption of the mixture at 700
nM using a spectrophotometer (Gilford, Oberlin, 0H). Mg—ATPase acti-
vity, assayed in the presence of 0.1 mM vanadate was subtracted from
the value observed in its absence to calculate the Na,K-ATPase acti-
vity. Sodium azide was used to inhibit Mg—ATPase activity and the
possible regeneration of ATP by oxidative phosphorylation (Ismail-
Beigi and Edelman, 1971). Since Ca—ATPase activity was negligible

under the present assay conditions, the vanadate-sensitive ATPase

 

29
activity could be taken as the representative of Na,K-ATPase activity.

Essentially, the same results were obtained by using 1 mM ouabain.

42K+ or 86Rb+ Uptake Binding

 

Sodium pump activity was estimated from the ouabain-sensitive
42K+ or 86Rb+ uptake by strips of right ventricular muscle or slices
of left ventricular muscle, respectively. Langendorff preparations of
guinea-pig heart were perfused with Krebs-Henseleit bicarbonate buffer
solution (pH 7.5) at a control flow rate of 2.5 ml/g tissue/min, or at
5% of the control flow rate, or not perfused at all for 2 hr. Several
preparations of the last group were reperfused for an additional 20-
min period at the control flow rate. From the first two groups of
preparations, thin ventricular strips were obtained and suspended in a
tissue bath containing the above Krebs-Henseleit bicarbonate buffer

solution maintained at 36°C and saturated with a 5% 02-90% N2-5% C02

gas mixture. The strips were electrically stimulated at either 1.5 or

7 Hz. Tracer amounts of 42K+ were then added to the incubation medium
and the 42K+ uptake during a 10-min incubation period was assayed.
42

The tissues were rinsed in a K+—free solution, blotted on filter
paper, and the radioactivity in the tissue was estimated using a gamma
scintillation spectrometer (Searle Analytic Inc., Des Plaines, IL;

42K+ uptake is the difference in

model 1185). Ouabain-sensitive

values observed in the presence and absence of 0.3 mM ouabain.
Ventricular strips obtained from the reperfused hearts did not

respond to electrical stimulation, making it impossible to evaluate

the viability of preparations on the effect of electrical stimulation.

30
Furthermore, in noncontracting tissues, the exchange of substances
between interstitial fluid and the medium may be limited (Lullmann gt_
gl,, 1979). Therefore, sodium pump activity was estimated from the

86Rb+ uptake using thin slices of

more conventional method involving
the ventricular muscle. Approximately 0.5 mm thick slices were pre-
pared by using a Stadie-Riggs tissue slicer (A.H. Thomas, Philadel-
phia, PA). The slices were incubated in a prewarmed (37°C) Krebs-
Henseleit bicarbonate buffer solution containing 2 mM RbCl with tracer

86Rb+ (specific activity, 0.16 Ci/mmol) and no K+. After a

amounts of
10-min incubation at 37°C, the slices were rinsed twice by immersing
in separate K+-free solutions for 20 sec each and blotting each time
with filter paper. The radioactivity remaining in the slices was

assayed using a gamma scintillation spectrometer. Ouabain-sensitive
86Rb+ uptake is the difference in values observed in the absence and

presence of 0.3 mM ouabain.

H. Nerve Activity Studies

 

Cats weighing 1.8-3.6 kg were anesthetized and prepared as de-
scribed above. During surgical preparation, cats were immobilized
with gallamine triethiodide (4 mg/kg) which caused muscle relaxation.
In all animals, the first and second ribs on the left were removed
extrapleurally and a thin postganglionic segment, the inferior cardiac
efferent nerve, arising from the left stellate ganglion was carefully
dissected out and mounted on a paired platinum electrode for monitor—
ing electrical activity. A left thoracotomy was then performed and

the pericardial sac opened to expose the left anterior descending

31
coronary artery. A silk thread was then passed through the myocardium
surrounding the artery just below the left atrium. The ends of the
silk sutures were fitted through a short piece of polyethylene tubing
and the artery was occluded, when needed, by pulling on the suture
through the tubing.

Neural discharges from the cardiac nerve were monitored with an
oscilloscope, and the mean discharge rates were determined using a
window discriminator and rate analyzer (Frederick Haer, Brunswick, ME)
and displayed on a polygraph recorder. The femoral arterial blood
pressure and lead II electrocardiogram were also recorded on the
polygraph. After an equilibration period of 30 min, when the nerve
activity was stable, the LAD was occluded for 60 sec and released.
Following recovery of the occlusion-induced changes in nerve activity,
digoxin infusion (60 ug/kg/hr) was started. When approximately 80-85%
of the arrhythmogenic dose of digoxin had been administered, the LAD
was again occluded for another 60 sec period and the electrical
activity of the nerve was monitored. When ventricular arrhythmias
finally developed, the 60 sec occlusion was again performed. When the
blood pressure was reduced by LAD artery occlusion, animals were
hemorrhaged, following release of occlusion and recovery of nerve
activity, by collecting blood from the femoral artery until the blood
pressure decreased to the same level as observed during occlusion.

The blood was infused back into the animal after the hemorrhage

experiment.

32

I. Miscellaneous Methods

 

Protein concentrations were determined by the method of Lowry gt_

gl,, (1951), using bovine serum albumin as the standard.

J. Statistical Analysis

 

Statistical evaluations of the data were performed by student's
t-test, paired t test, linear regression and two-way analysis of
variance with or without block design. Criterion for significance was

a probability value of less than 0.05.

RESULTS

A. Ischemia-induced Enhancement of the Arrhythmogenic Actions of
Digitalis: Effects of Ischemia, and Reperfusion on Cardiac
Sarcolemmal Na,K—ATPase Activity, on the Kinetic Parameters of
Na,K-ATPase for (5H)0uabain Binding and on Sodium Pump Activity

 

 

 

 

The arrhythmogenic effects of digitalis glycosides have been
shown to result from both a direct action on the heart and an indirect
action involving the autonomic nervous system. A direct action of the
glycoside on the heart has been shown to result from its binding to
sarcolemmal Na,K-ATPase, a putative receptor for the glycoside (Akera
and Brody, 1977). Since ischemia is well known to produce membrane
damage, electrolyte imbalance and abnormal metabolism (Jennings gt
gl,, 1981), such effects may augment the direct action of the glyco-
side on the heart, by altering Na,K-ATPase activity, the sensitivity
of Na,K-ATPase to the inhibitory effects of digitalis, or by reducing
the reserve capacity of the sodium pump. Digitalis toxicity is pro-
posed to occur when glycoside-induced inhibition of the sodium pump
exceeds its reserve capacity, and the remaining pump becomes inade-
quate for maintaining a low intracellular sodium ion concentration.
Therefore, these possibilities were studied using Langendorff pre-
parations of guinea-pig hearts.

1. Isolated-Heart Perfusion Studies

 

In order to determine whether ischemia augments the arrhyth-

mogenic effects of digitalis in the isolated guinea-pig heart, the

33

34

left anterior descending (LAD) coronary artery was completely occluded
for 40 min before perfusion with an arrhythmogenic concentration of
digoxin.

, The perfusion of the heart at a constant flow rate of 2.5
ml/g tissue/min did not produce arrhythmias and the force of contrac-
tion remained stable for at least three hours. Coronary artery occlu-
sion by itself also did not produce arrhythmias during the perfusion
under the experimental condition (Figure 1a). The perfusion at a
constant flow rate with either 1.8 or 2.5 uM digoxin produced a posi-
tive inotropic effect and then arrhythmias. Although the rate of
onset of the positive inotropic effect of digoxin was similar in both
control and LAD artery ligated animals, the onset of arrhythmias was
significantly earlier in the LAD artery ligated animals for both
concentrations of digoxin (Figure 2). The types of digoxin-induced
arrhythmic contractions observed in control and the LAD artery ligated
hearts are shown in Figure 1 (b-f). Extra beats following the normal
stimulation-induced beat were present in all digoxin—induced arrhyth-
mias. During the extra beat, the muscle failed to respond to elec-
trical stimulation, probably due to refractoriness during repolari-
zation. When the extra beat was complete, stimulation was again
followed by contractions. All types of arrhythmic contractions shown
were observed in both control and LAD-ligated hearts.

In order to examine the possbility that changes in the time

to onset of digoxin-induced arrhythmias are due to alterations in the
flow distribution caused by LAD artery ligation, several heart pre-

parations were perfused under a constant pressure of 20 mmHg. In this

 

35

a b

C d

e f
Figure 1

Figure 1. Effects of digoxin on the isometric twitch tension curves
observed in Langendorf preparations of guinea-pig heart. The heart
preparations were electrically stimulated at 1.5 Hz and perfused with
Krebs-Henseleit bicarbonate buffer (pH 7.5) solution at 32°C. Follow—
ing an equilibration period, the left anterior descending coronary
artery was occluded, and digoxin (final concentration, 1.8 or 2.5 pM)
was added to the perfusing solution 40 min later. Development of
digoxin-induced arrhythmias was monitored by the twitch tension
curves. a: Non-arrythmic beats observed before the onset of arryth-
mias in all preparations perfused with or without digoxin. b—f:
Typical arrhythmic beats observed at the onset of digoxin-induced
arrhythmias in both control (b,d,e) and occluded (c,f) preparations.
Typical tracings are shown from several experiments.

 

36

 

 

 

 

 

 

 

      

35
30 t ‘
.S 20 -
E
Q’ l-
E
t.—
10 ’
0
Digoxin 1.8 2.5
Constant Constant Constant
flow flow pressure
Figure 2

Figure 2. Effects of LAD artery ligation on the time of digoxin
perfusion required to produce arrhythmias. See legend to Figure 1.
Open bars: Control hearts without occlusion. Shaded bars: LAD
artery occluded hearts. The rate of perfusion of the hearts were
either at 2.5 ml/g tissue/min or at constant pressure of 20 mmHg.
Each bar represents the mean of 6 experiments. Vertical lines indi-
cate S.E. *Significantly different from the corresponding control
value (p<0.05).

 

 

37
model too, 2.5 pM digoxin produced arrhythmias earlier in the LAD
artery ligated than in control preparations (Figure 2). In these
hearts, there were no clear differences between the two groups with
respect to the types of contractions. These results demonstrate that
LAD artery ligation sensitizes the isolated heart preparations to the
arrhythmogenic actions of digoxin and that Langendorff preparations of
guinea-pig hearts may be used to study the mechanisms of increased
sensitivity of ischemic heart to digitalis.

2. Na,K—ATPase Studies

 

An enhanced digitalis sensitivity of the ischemic heart may
be due to a change in Na,K-ATPase activity or its sensitivity to the
inhibitory actions of the glycoside. Thus, enzyme activity of homo-
genates obtained from ischemic hearts and the sensitivity of the
enzyme to the inhibitory actions of dihydrodigoxin were examined using
globally ischemic guinea-pig heart preparations. Perfusion of Langen-
dorff preparations was completely stopped or they were perfused at 5%
of the control flow rate for 2 or 6 hr. Since digitalis action on the
ischemic border zone may be important, heart preparations were per-
fused with severely-reduced flow rates (5% of the control flow rate).
Subsequently, ventricular muscle homogenates were prepared and assayed
for Na,K-ATPase in the absence and presence of dihydrodigoxin. This
glycoside was selected because of its rapid rate of binding to the
glycoside binding sites on Na,K-ATPase.

Perfusion of Langendorff preparations for 2 hr with a re-
duced flow rate (5% of the control rate) or no perfusion (0% of the

control rate) failed to significantly alter Na,K-ATPase activity in

38
homogenates obtained from these preparations (Figure 3). Perfusion of
Langendorff preparations for 6 hr with a reduced flow rate failed to
alter enzyme activity whereas 6 hr of zero-perfusion caused a signi-
ficant decrease in Na,K-ATPase activity (Figure 4). Dihydrodigoxin
inhibited the enzyme in concentrations ranging from 30 to 300 pM.
Sensitivity of Na,K-ATPase to the inhibitory action of dihydrodigoxin
was not altered by ischemia, as indicated by the similar dihydro-
digoxin concentrations necessary to cause a 50% inhibition (approxi-
mately 100 uM) of Na,K-ATPase in homogenates obtained from control or
ischemic tissues (Figures 3 and 4).

The effect of reperfusion of ischemic tissues on Na,K-ATPase
activity was examined since ischemia-induced myocardial damage may
recover or may be further enhanced during reperfusion. In these
studies, perfusion was completely stopped for 2 or 5 hr, and subse—
quently, the preparations were reperfused at a control flow rate (2.5
ml/g tissue/min) for an additional 1 hr period. After 2 or 5 hr of
complete ischemia and l-hr reperfusion, Na,K-ATPase activity of ven-
tricular muscle homogenates, assayed as above, was significantly
reduced (Figure 5). It should be noted that 2-hr ischemia failed to
cause significant decreases in Na,K-ATPase activity. Therefore,
reperfusion of ischemic tissues apparently enhanced the reduction in
Na,K-ATPase activity caused by ischemia. Sensitivity of the remaining
active enzymes to the inhibitory actions of dihydrodigoxin was un-
affected by ischemia and reperfusion.

It is possible that ischemic tissues obtained from globally

ischemic hearts may not be similar to the ischemic tissues formed in

39

 

O
4:.

I I T I I
E
SE -
'f- - a-
>.c, 0 3 1......
'r-H
H\
U:
«3'!-
OJ
“"8
(I) 1—
cos. 02 T
CLO.
'—
<05
IE
+ \
)1 H'-
On- F— q
+ 001
(Or--
20
E
3
0

 

 

 

_.L__./r£ 1 L 1 L
O 10 30 100 300
Dihydrodigoxin guM)

Figure 3

Figure 3. Effects of global ischemia on Na,K-ATPase activity and its
inhibition by dihydrodigoxin. Langendorff preparations of guinea-pig
hearts were perfused at control flow rate of 2.5 ml/g tissue/min (O),
or 5% (I) or 0% (A) of the control flow rate for 2 hr. Subsequent-
ly, ventricular muscle was homogenized and assayed for Na,K-ATPase
activity. Na,K—ATPase activity was calculated as the difference in
values observed in the absence and presence of 0.1 mM vanadate. Each
point represents mean of 6 experiments and vertical lines indicate
S.E. None of the values were significantly different from the corre-
sponding control values (p<0.05).

40

O
p

 

O
1.0

O
N

O
p—n

 

Na+,K+—ATPase act1v1ty
(umol Pi/mg protein/10 min)

 

 

 

“(t-#1.; ‘ ‘

30 100 306
Dihydrodigoxin (AM )

Figure 4

Figure 4. Effects of global ischemia on Na,K-ATPase activity and its
inhibition by dihydrodigoxin. Langendorff preparations of guinea-pi
hearts were perfused at control flow rate of 2.5 ml/g tissue/min ((3%,
or 5% (O) or 0% (A) of the control flow rate for 6 hr. Subsequent-
ly, ventricular muscle was homogenized and assayed for Na,K-ATPase
activity. Na,K—ATPase activity was calculated as the difference in
values observed in the absence and presence of 0.1 mM vanadate. Each
point represents mean of six experiments and vertical lines indicate
%.E. *Significantly different from the corresponding control value
p<0.05 .

 

41

 

0.4, 4 , r 1

 

 

 

'2
5“? p
to 0'3 . l
".4...
r
(“‘8 * *
8‘0) 0.2 '- * . H
on
g... {a
< 00
NE 01 * *
“in,“ . P 1
+23 * .
V O_J_¥rl l l l 1
0 10 30 100 300

Dihydrodigoxin (,uM )

Figure 5

Figure 5. Effects of reperfusion on Na,K-ATPase activity and its
inhibition by dihydrodigoxin. Langendorff preparations of guinea-pig
heart were perfused at 0% of the control flow rate for 2 hr (II) or 5
hr (IL) and subsequently reperfused at control flow rates for 1 hr.
Control preparations were perfused for 3 hr (CD) or 6 hr (ID). Na,K-
ATPase activity of the ventricular homogenates were assayed, Na,K-
ATPase activity was calculated as the difference in values observed in
the absence and presence of 0.1 mM vanadate. Each point represents
the mean of six experiments and vertical lines indicate S.E. *Signi-
ficantly different from the corresponding control values (p<0.05).

 

42
coronary artery occluded hearts with respect to uniformity of ischemia
and metabolic events. In particular, the ischemic border zone, pro-
posed to exist in ischemic hearts, may not be mimicked simply by
reducing the perfusion flow rate. Therefore, the effect of coronary
artery occlusion-induced ischemia on Na,K-ATPase activity was examined
using ischemic tissues obtained from LAD artery occluded hearts of
anesthetized cats. This animal species was used because preliminary
studies (see Part B) have indicate that LAD artery ligation enhanced
the arrhythmogenic effects of digitalis, and the size of hearts is
relatively large, allowing sufficient quantities of ischemic tissues
to be obtained for analysis.

Following LAD artery ligation, either digoxin was infused
intravenously at a rate of 60 ug/kg/hr until the onset of arrhythmias
or 0.9% saline solution was infused for a similar period of time. A
dye solution was infused and non-ischemic (NI), partially ischemic
(PI) and completely ischemic (CI) ventricular tissues were obtained
(see Figure 6), and the homogenates of these tissues were assayed for
Na,K-ATPase activity. Animal preparations without LAD artery occlu-
sion were also infused with digoxin and heart tissues obtained at the
onset of arrhythmias.

Na,K-ATPase activities of the homogenates obtained from
perfused, partially perfused and non-perfused tissues in saline
infused animals were all similar (Figure 7). Na,K—ATPase activities
of the homogenates obtained from partially ischemic and non-ischemic
tissues in digoxin-infused animals were significantly reduced compared
to the corresponding control values in saline-infused animals. Na,K—

ATPase activity of the homogenates obtained from completely ischemic

 

43

 

Figure 6

Figure 6. A cross-section of the heart showing stained and non-
stained ventricular muscle following infusion with a dye at the onset
of digoxin-induced arrhythmias in LAD artery ligated cats.

44

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E; 0.20 T T
is 015* -l l/ ’
'55 1/ .. ../
g? 0.10 r 1% I E It 72% 1

 

 

 

 

NI PI CI C NI
Saline Digoxin Digoxin

CI

Figure 7

Figure 7. Effects of ischemia on Na,K-ATPase activity in ventricular
muscle homogenates obtained from anesthetized cat hearts. LAD artery
was completely ligated and either digoxin or similar volume of 0.9%
saline was infused 40 min later. At the onset of arrhythmias in
digoxin-infused animals or 3 hr later in saline-infused animals, dye
was infused and completely stained (CI), partially stained (P1) or
nonstained (NI) ventricular tissues were prepared. Homogenates ob—
tained from these tissues were assayed for Na,K-ATPase activity.
Na,K-ATPase activity is the difference in values observed in the
presence and absence of 0.3 mM ouabain. Open bars: Na,K-ATPase
activity in non-ischemic (NI); dotted bars: in partially ischemic
(PI); and shaded bars: in completely ischemic (CI) tissue homogen-
ates. The smallest open bar (C) represents Na,K—ATPase activity in
tissue homogenates obtained from control (no coronary artery occlu-
sion), digoxin infused animals. Each bar represents the mean of five
experiments. Vertical lines indicate the S.E. *Significantly differ—
ent from the corresponding control value (p<0.05).

 

45
tissues was, however, similar to the control value suggesting that
ischemia itself did not inhibit the enzyme activity. In digoxin-
infused animals with no LAD-ligation, Na,K-ATPase activity of the
ventricular homogenates was lower than that from LAD artery ligated.
hearts. This is probably due to a longer infusion period in control
animals, and therefore a larger dose of digoxin infused for the
development of arrhythmias.

Thus, ischemia produced globally, or by coronary artery
occlusion, failed to alter Na,K-ATPase when the period of ischemia was
2 or 3 hr, respectively. A longer period of ischemia (6 hr) or re-
perfusion following 2 hr of ischemia reduced Na,K-ATPase activity by
decreasing the number of active Na,K-ATPase molecules, but the sensi-
tivity of the residual enzyme to the inhibitory effects of digitalis
remained unaltered. In animals in which LAD coronary artery was
ligated and infused with digoxin, the completely ischemic tissue were
not reduced in enzyme activity. This is probably due to absence of
digoxin in this tissue.

3. (3H)Ouabain Binding Studies

 

The elevated sensitivity of the ischemic heart to the toxic
actions of the cardiac glycosides may result from a change in glyco-
side binding to sarcolemmal Na,K-ATPase. Therefore, possible changes
in the number of glycoside binding sites and their affinity for (3H)—
ouabain binding were determined. Initial velocity of (3H)ouabain
binding reaction was also estimated to determine whether ischemia

alters the rate of glycoside binding to Na,K-ATPase.

46

Ventricular muscle homogenates obtained from globally is-
chemic Langendorff preparations were incubated with (3H)ouabain for 90
min, allowing the binding reaction to reach equilibrium. The number
of (3H)ouabain binding sites (Bmax) and the affinity of the sites for
ouabain (Kd value) were calculated as described previously (Akera and
Cheng, 1977). A reduced perfusion rate (5% of the control flow rate)
for 2 or 6 hr, or a zero-perfusion for 2 hr, failed to significantly
alter either Bmax or Kd value in ventricular muscle homogenates (Table
1); however, zero-perfusion for 6 hr, or 2 or 5 hr of zero-perfusion
followed by 1 hr of reperfusion, significantly reduced the number of
binding sites, without affecting the affinity of the remaining binding
sites for ouabain.

Ventricular muscle homogenates obtained from hearts of
anesthetized cats in which the LAD artery was ligated and then infused
with either saline or digoxin were prepared as described above, and
assayed for initial velocity of ATP-dependent (3H)ouabain binding.
(3H)0uabain binding was similar in ventricular muscle homogenates
obtained from non—ischemic (NI), partially ichemic (PI) and completely
ischemic (CI) tissues (Figure 8). In digoxin-infused animals, ATP-
dependent (3H)ouabain binding to the non-ischemic and partially
ischemic ventricular muscle homogenates was significantly less than
the values observed in similar tissues obtained from saline-infused
animals (control). (3H)0uabain binding to completely ischemic muscle
homogenates was slightly, but not significantly, reduced from the
control value. This is probably because of the absence of digoxin in

the ischemic tissues as a result of lack of perfusion. Therefore, the

47

TABLE 1

Effect of Global Ischemia and Reperfusion 0n Kinetic Parameters
of Na,K-ATPase for (3H)0uabain Binding Reaction

Langendorff preparations of guinea-pig heart were perfused at
control flow rate of 2.5 ml/g/min 0r 5% 0r 0% 0f the control flow rate
for either 2 0r 6 hr. In some preparations, hearts were perfused at
0% for 2 or 5 hr and reperfused at control rate or 1 hr. Ventricular
muscle was then homogenized and incubated with ( )ouabain in the
presence of 1 mM MgClg, 1 mM Tris-P04, 10 mM Tris-HCl buffer (pH 7.5)
and various concentrations of non-labeled ouabain (0-1000 nM). After
a 90-min incubation, specific (3H)0uabain binding was calculated by
subtracting non-specific binding observed in the presence of 0.1 mM
ouabain from the total binding observed in its absence. The affinity
of the binding sites for ouabain (Kd value) and the method of Akera
and Cheng (1977). Each value represents mean :_S.E. of six experi-
ments.

 

 

Perfusion (time) Bmax1 Kd

Pmol/mg protein nM
Controlz (2 hr) 6.30:0.38 110.9:5.5
5% (2 hr) 5.85i0.45 95.1:7.9
0% (2 hr) 5.30:0.34 91.1i7.0
Controlz (6 hr) 5.54:0.43 90.2:8.3
5% (6 hr) 4.41:0.273 80.8:3.4
0% (6 hr) 3.80:0.21 103.0:6.6
Contr012 (3 hr) 6.05:0.51 105.8:4.2
0% (2 hr) p1us controi2 (1 hr) 4.20:0.363 107.5:7.5
Controlz (6 hr) 6.18:0.40 102.0:3.7
0% (5 hr) p1us controlz (1 hr) 3.50:0.413 101.2:7.4

 

1Bmax (binding site concentration) and Kd (apparent dissociation

constant).

22.5 ml/g tissue/min.

3Significantly different from control values (p<0.50).

 

48

 

O
0‘

 

0
0‘1
T

 

 

O
4:.
T

 

.....
oooooo
oooooo
0000000
......
ooooooo
......
ooooooo
------
ooooooo
OOOOOO
......
oooooo
------
......
------
............
ooooo
.0 no
oooooo
ooooo

 

......
......
oooooo
nnnnnn
......
''''''''''''
oooooo
......
......
......
......
oooooooooooo
......

o o
t—I N
I

GGGGG
IIIIII
lllll
......
.....
000000
00000
000000
OOOOO
......
.....

 

O—Jl—i

(umol/mg protein/2 min)
0
(A)

(3H)ouabain binding

 

 

 

 

 

 

.. ’1,
\\\\\\\\~

 

 

 

 

 

 

ooooo
ccccc
..........
ooooooo

------
oooooo
......
......
......
------
oooooo
nnnnnn
ooooo
.....

CD

 

NI PI CI C NI PI CI
Saline Digoxin Digoxin

Figure 8

Figure 8. Effects of ischemia on the ATP-dependent (3H)ouabain
b1nd1ng to ventricular muscle homogenates obtained from anesthetized
cat hearts. See legend to Figure 6. Homogenates obtained from com-
pletely 1schemic (CI, shaded bars), partially ischemic (PI, dotted
bars) and non—ischemic (NI, open bars) ventricular muscle were assayed
for (§H)ouabain binding. Specific (3H)ouabain binding is the differ-
ence 1n value observed in the presence and absence of ATP. The
smallest open bar (C) represents (3H)ouabain binding to homogenates
obta1ned from control, digoxin-infused animals. Each bar represents
the mean of five experiments. Vertical lines indicate the S.E.
*S1gn1f1cantly different from the corresponding control value (p<0.05).

 

49
glycoside binding sites were not occupied by digoxin. In digoxin-
infused animals in which the LAD artery was not occluded, the initial
velocity of (3H)ouabain binding to the ventricular muscle homogenates
prepared at the onset of glycoside-induced arrhythmias was slightly
lower than the value observed in homogenates obtained from occluded
animals. Thus, fractional occupancy of the glycoside binding sites on
Na,K-ATPase by digoxin was non-uniform in LAD artery ligated hearts
and was less in ischemic tissues than in non-ischemic tissues.

The binding of the cardiac glycoside to Na,K-ATPase is
elevated by increased intracellular Na+ or decreased extracellular K+
(Akera and Brody, 1977). Therefore, possible changes in the effects
of these cations to stimulate or inhibit the specific (3H)ouabain
binding were examined in ventricular muscle homogenates obtained from
globally ischemic heart preparations. After a reduced perfusion (5%
of control flow rate) or zero-perfusion for 6 hr, or after 2 or 5 hr
of zero-perfusion followed by 1 hr of reperfusion, the concentration
of Na+ to cause a half-maximal stimulation of the glycoside binding or
the concentration of K+ to cause a half-maximal inhibition of the
binding was unchanged (Table 2). Thus, a 2-hr ischemia, whether
produced globally by reducing or stopping the perfusion of the whole
heart, or produced locally by coronary artery occlusion, failed to
affect the number of glycoside binding sites or their affinity for
ouabain. A much longer ischemia (6 hr) or reperfusion following 2- or
5-hr ischemia markedly reduced the number of binding sites without

affecting their affinity for ouabain, Na+ and K+.

 

50

TABLE 2

Concentrations of K+ and Na+ Affecting Specific Binding
of (3H)0uabain to Na,K-ATPase During Ischemia

Langendorff preparations of guinea-pig heart were per-
fused as described in the legend to Table l. The ventricu-
lar muscle homogenates were incubated in a medium containing
200 mM NaCl, 5 mM MgClz, 5 mM Tris-ATP, 50 mM Tris-HCl
buffer (pH 7.5), 10 nM (3H)ouabain and KCl (0-150 mM), or in
a similar medium without KCl containing 0-300 mM NaCl. Each
value represents the mean :_S.E. of six experiments.

 

 

Perfusion (time) ?;M; ?;;)Z
Control (3 hr) 2.7510.10 32.821.0
Control (6 hr) 2.90:0.20 30.5:l.5
5% (6 hr) 2.80iO.1O 30.6i1.7
0% (6 hr) 3.00:0.10 31.6i1.3
0% (2 hr) + control (1 hr) 3.10iO.26 30.2i2.9

0% (5 hr) + control (1 hr) 2.60:0.15 32.0i4.0

 

None of the values were significantly different from the
corresponding control values (p<0.05).

1The concentration of K+ to cause a 50% inhibition of the
specific ouabain binding.

2The concentration of Na+ to cause a half-maximal stimula—
tion of the specific ouabain binding.

51

42 + or 86

4. K Rb+ Uptake Studies

 

Myocardial ischemia may elevate the digitalis sensitivity of
the heart by decreasing either the sodium pump activity or the reserve
capacity of the sodium pump. _In order to examine these possibilities,
ventricular slices were prepared from isolated guinea—pig hearts which
were perfused at 5% of the control flow rate or not perfused for 2 hr,
or subjected to zero-perfusion for 2 hr and subsequently reperfused at

the control flow rate for 20 min. Activity of the sodium pump was

86

estimated from the specific (ouabain-sensitive) Rb+ uptake. In

86

these preparations, specific Rb+ uptake accounted for approximately

50% of the total uptake, the remaining 50% representing nonspecific

86

uptake (Figure 9). The specific Rb+ uptake was unaffected by either

the reduced perfusion or zero-perfusion for 2 hr; however, 2 hr of

zero-perfusion followed by 20 min reperfusion significantly reduced

86Rb+ uptake by ventricular slices obtained from these pre-

86

specific

parations. It should be noted that nonspecific Rb+ uptake was also

reduced in these preparations.

The reserve capacity of the sodium pump may be estimated

42

from the difference in the specific K+ uptake by the tissues when

stimulated at 7 and 1.5 Hz. Since thin slices of ventricular muscle
do not respond to electrical stimulation by visible contraction, right
ventricular strips were used in these experiments. In ventricular
strips obtained from non-ischemic preparations, the specific 42K+
uptake observed at 7 Hz stimulation was significantly greater than

42 +

that observed at 1.5 Hz stimulation (Figure 10). The specific K

uptake by ventricular muscle strips obtained from Langendorff

 

11

 

 

 

52

 

 

 

 

 

 

 

 

 

 

2.5
EzofE—ﬁﬁ .
E1.5~.1
131.0”; ?? * 1
ﬁjééég
NI PI CI R
Figure 9

Figure 9. Effects of ischemia, and reperfusion on 86Rb+ uptake in
ventricular slices of guinea-pig heart. Langendorff preparations of
guinea-pig heart were perfused at control flow rate of 2.5 ml/g
tissue/min, 5% (PI) or 0% (CI) of the control rate for 2 hr, or 0% of
the control rate for 2 hr and reperfused at control rate for 20 min
(R). Ventricular slices were prepared and incubated at 37°C in modi—
fied Krebs-Henseleit bicarbonate buffer (pH 7.5) solution containing 2
mM RbCl and no Ki, an 86Rb+ uptake by the tissues was‘estimated 10
min later. Specific 6Rb+ uptake (open bars) is the difference in
values observed in the absence (whole bars) and presence (shaded bars)
of 0.3 mM ouabain. Each bar represents the mean of six experiments.
Vertical lines indicate S.E. (Significantly different from the
control value (p<0.05).

 

53

 

H
UT

.—o
N

  
 

ﬂ/

/

*
/ (3

/
:

\D

   
    
    
       

42K+ uptake
0‘

(nmol/mg protein/30 min)

U)

If

I

w 4 if ,
742/77.

 

 

 

 

O

1.5 7.0 1.5 7.0 Hz
Control Ischemic
Figure 10

Figure 10. Effects of ischemia on the specific 42K+ uptake by ven-
tricular strips of guinea-pig heart. Langendorff preparations of
guinea-pig heart were perfused at the control flow rate of 2.5 ml/g
tissue/min or 5% of the control rate for 2 hr. Subsequently, right
ventricular strips were prepared and incubated in Krebs-Henseleit
bicarbonate buffer solution (pH 7.5) containing 4 K+. Preparations
were electrically stimulated at either 1.5 (open bars) or 7 Hz (shaded
bars) and 42K+ uptake was estimated 10 min later. Specific 42K+
uptake is the difference in uptake observed in the absence and pre-
sence of 0.3 mM ouabain. Each bar represents the mean of seven
experiments. Vertical lines indicate S.E. *Significantly different
from the control value (p<0.05).

54
preparations perfused for 2 hr at a reduced flow rate (5% of the
control flow rate) was slightly higher than corresponding values
observed with control (non-ischemic) preparations. The difference in

42K+ uptake observed at 1.5 and 7 Hz stimulation, how-

the specific
ever, was not altered by partial ischemia. These results indicate
that sodium pump activity was not significantly affected by the par-
tial reduction or complete stopping of the perfusion for 2 hr. The
reserve capacity of the sodium pump was also unchanged by 2 hr of
reduced perufsion.

In summary, coronary artery occlusion significantly enhanced
the arrhythmogenic actions of digitalis in isolated perfused guinea-
pig heart. This effect occurred within 80 min following coronary
occlusion. However, Na,K—ATPase activity, its sensitivity to the
inhibitory effects of digitalis, number of glycoside binding sites and
their affinity for ouabain, K+ and Na+, sodium pump activity, and the
reserve capacity of the sodium pump were not affected by 120 min of
ischemia. Therefore, although the primary mechanism of digitalis
sensitization appears to be by a direct action on the heart, it is not
by an enhanced glycoside effect on the cardiac sarcolemmal Na,K-
ATPase. It is possible, however, that the lack of effect of digoxin
in completely ischemic tissues may contribute to the reduced tolerance
of the heart to digitalis by causing a non-uniform digitalis effect on
the heart. This problem was further examined and is described in Part
C.

Although the direct action of digitalis is sufficient to
account for the enhanced toxicity in the ischemic heart, one cannot

rule out the additional influence of indirect effects of digitalis via

 

55
the autonomic nervous system. Therefore, the indirect effects of
digitalis on the sensitivity of the ischemic hearts to the glycosides
were examined next.

8. Ischemia-induced Enhancement of Arrhythmogenic Actions of Digi-
talis: Involvement of the Sympathetic Nervous System

 

 

Results of studies in Part A have demonstrated that ischemia
enhances the arrhythmogenic effects of digitalis by a direct action on
the heart. It has been shown that in patients with myocardial is-
chemia, as well as in experimental animals with coronary artery occlu-
sion, the catecholamine concentrations in the blood are significantly
elevated (Valorie gt_gl,, 1967; Staszewska-Barczak, 1971). In ani-
mals, coronary artery occlusion has been shown to produce alterations
in sympathetic discharge (Brown and Malliani, 1971) and vagal afferent
nerve activities (Kedzi gt_gl,, 1974). Since the effect of digitalis
on the heart has been demonstrated to be affected by the autonomic
nervous system (Gillis and Quest, 1979), the possibility that the
nervous system might influence the arrhythmogenic effects of digitalis
on the ischemic heart strongly exists. Therefore, intact animals were
employed to examine this possibility.

1. Whole Animal Studies

 

In order to study the role of the sympathetic nervous system
in the reduced tolerance of the ischemic heart to the toxic effects of
digitalis, sympathetic discharge to the heart was interrupted surgi-
cally or pharmacologically. Anesthetized cats were either left
neurally intact, bilaterally vagotomized, or bilaterally vagotomized
and spinal cord sectioned at C-1 or pretreated with dl-propranolol.

After a 40-min occlusion of the left anterior descending coronary

 

56

artery, digoxin (60 ug/kg/hr) was infused until the onset of ventri-
cular arrhythmias. In four animals in which the coronary artery was
occluded, the intravenous infusion of 0.9% saline solution did not
produce arrhythmias for the entire experimental period of 4 hr.
Figure 11 shows typical electrocardiograms and blood pressure in
neurally-intact (control) and LAD artery ligated animals infused with
digoxin. In all animal preparations, digoxin infusion caused a
gradual rise in blood pressure, and a gradual fall in heart rate.
Blood pressure was not significantly altered by occlusion itself
although in some animals, a slight and transient fall was observed.
The mean blood pressures observed before and at various time points
after digoxin infusion were similar in both coronary artery occluded
and non-occluded hearts. Coronary artery occlusion caused S—T segment
elevation which became more pronounced with time. The onset of
digoxin-induced arrhythmias was accompanied by QRS complexes with
markedly greater amplitudes and disappearance of P waves in both LAD
artery occluded and control animals. However, in either neurally
intact or bilaterally vagotomized animal preparations, the onset of
arrhythmias was earlier in LAD artery occluded than in control, non-
occluded preparations (Figure 12). The differences in the dose of
digoxin required to produce arrhythmias in control and LAD artery
ligated animals were similar in intact and vagotomized animals.

Bilateral vagotomy caused an initial rise in blood pressure
that returned to control levels within 30 min. In these animals,
spinal cord section at C-1 produced a dramatic and immediate increase

in blood pressure and heart rate followed by a gradual decline to a

 

57

 

.mpcmswcmaxm cho>om cote exegm mew

mmcvomep Pmowaxe .mw25pxscem pm use .cowm:_ooo sue; cormsecw :wxomvu to cps cop empem
.20wszooo a<4 Loewe cps om um>cmmno mezmmwca voo_a new Emtmowucmoocpomrm “m _w:ea
.m8252pxseem mo pmmco we“ on use .:o_m:ecv :wxomwc to avg 00— Lopes .cowmzccm :wxomwc
mcommn um>cmmno meammwca voo_n new EnemowucmoocpomFm "Apemwc op pem_ Eoemv < Focma
.mpmo em~_pm;umw:m umuz_ooo-:o: ecu emuzpooo acmpcm xemcoeoo cw cowmzmce :rxomve
mcwczu ﬂamv mezmmoea coo—a _mwcmpem FmeoEmm use wamv EmemorcemoocpomFM .PP mczmvm

58

 

evacpxgcc<

__ etsmze

:wE ooH

 

 

 

 

oo—

om—

 

59
level approximately 60 mmHg below the pre-surgical blood pressure.
The mean blood pressure remained at approximately.80—100 mmHg. Such
effects indicated success of C-1 spinal section. In these C-l sec-
tioned animal preparations, the dose of digoxin required to produce
arrhythmias was significantly greater than that in neurally-intact
animals in both coronary artery occluded and non-occluded animals
(Figure 14). The differences in the dose of digoxin required to
produce arrhythmias in control and LAD artery ligated animals were
similar in neurally-intact and C-1 sectioned animals. These results
indicated that C-l spinal section did not influence the ischemia-
induced sensitivity to the arrhythmogenic actions of digitalis.

In animal preparations pretreated with dl-propranolol, the
effect of isoproterenol-HCl on the heart rate was examined to test the
completeness of beta-adrenergic receptor blockade (Figure 13). Con-
centrations of isoproterenol-HCl (0.1-1.0 pg/kg) produced marked dose-
dependent increases in heart rate which were completely blocked by the
dl-propranolol dose-regimen used in this study. In these animals
without LAD artery ligation the dose of digoxin required to produce
arrhythmias was greater than that in control, untreated animals. The
effect of dl-propranolol pretreatment on the dose of digoxin required
to produce arrhythmias was slightly greater than the effect of C-1
section, but the difference in dose was not statistically significant.
These results demonstrate that bilateral vagotomy and/or sympathectomy
produced by either spinal transection or beta-adrenergic receptor
blockade failed to influence the increase in digitalis sensitivity of

the ischemic heart.

 

60

 

180

 

 

p—a
03
O
T
h—«—4

O—I
«b
O

 

H
N
O
l

Digoxin lug/kg)
8

 

 

 

 

 

 

 

 

 

 

LAD intact intact
occluded occluded

sham operated vagotomized

Figure 12

Figure 12. Effects of LAD artery ligation on the dose of digoxin
required to produce arrhythmias in neurally intact or bilaterally
vagotomized anesthetized cats. Sixty min following bilateral vagotomy
or sham operation, the left anterior descending coronary artery was
occluded. Forty min later, digoxin infusion (60 ug/kg/ hr) was '
started in LAD artery occluded (shaded bars) and non-occluded animals
(open bars) and the time to the onset of glycoside-induced arrhythmias
was monitored. Each bar represents the mean of six experiments.
Vertical lines indicate the S.E. *Significantly different from con-
trol values (p<0.05). '

 

61

 

300 l " * ‘

250 r

  

200

l
,L
0

Isoproterenol (ug/kg)

 

Heart Rate/min

 

l I

0.1 0.3 1.0

Figure 13

Figure 13. Effects of isoproterenol-HCl on the heart rate in the
absence and presence of dl-propranolol-HC1 in anesthetized cats. An
intravenous loading dose of dl-propranolol (1 mg/kg) was followed by a
slow continuous infusion (l ug/kg/min). Thirty min later, an intra-
venous bolus injection of 0.1, 0.3 and 1.0 ug/kg isoproterenol-HCl
were given at 20 min intervals and the changes in the heart rate was
monitored. Control animals were similarly tested with isoproterenol-
HCl. Peak chronotropic effects were noted and plotted against the
isoproterenol concentrations. Each point represents the mean of six
experiments. Vertical lines indicate S.E. *Significantly different
from the corresponding control values (p<0.05).

62

 

 

 

 

 

 

 

 

 

 

 

250 ' l
';;200 r -
i 1
, 3?
~« 5 .
.5 1 0 '7
x
8
-- 100 L -
C3
0 _
LAD intact intact

occluded occluded
C-l sectioned propranolol

Figure 14

Figure 14. Effects of LAD artery ligation on the dose of digoxin
required to produce arrhythmias in C-l sectioned or dl-propranolol-
treated anesthetized cats. Sixty min following bilateral vagotomy and
C-1 section or propranolol treatment, LAD was occluded. Digoxin
infusion (60 pg/kg/hr) was started in both LAD-occluded (shaded bars)
and control (open bars) preparations 40 min later and the time to the
onset of glycoside-induced arrythmias was monitored. Each bar repre-
sents the mean of five experiments. Vertical lines indicate S.E.
*Significantly different from the control values (p<0.05).

 

63

2. Plasma and Myocardial Digoxin Concentration Studies

 

In order to examine whether the animals in which the coro-
nary artery was occluded have altered digoxin pharmacokinetics, the
plasma digoxin concentration was assayed at various time points
following digoxin infusion. Either specifically labeled 126-3H-
digoxin or unlabeled digoxin was infused and the plasma digoxin was
estimated by assaying for radioactivity or by radioimmunoassay, re-
spectively. An intravenous infusion of digoxin (60 pg/kg/hr) to
control animals caused a linear increase in plasma glycoside concen-
tration (Figure 15). Similar digoxin infusion to coronary-occluded
animals changed neither the rate of increase nor the concentration of
digoxin in plasma at any time period from the corresponding control
values. The plasma digoxin concentration at the onset of arrhythmia
was, however, significantly less in LAD artery occluded than in con-
trol animals. Since the method of assay for digoxin was the same for
any paired group of animals (Table 3), these results indicate that a
lower plasma glycoside concentration was present at the onset of
arrhythmias in LAD artery occluded animals than control animals and
that this was not due to altered digoxin pharmacokinetics. It is
possible that other factors such as heart rate may affect digoxin
binding to cardiac tissues since the rate of glycoside binding has
been shown to be frequency-dependent (Yamamoto gt gl,, 1979). This
possibility was examined by determining myocardial digoxin uptake in
LAD artery occluded and control animals infused with radioactive
digoxin. Cardiac tissues were prepared from both groups of animals

and assayed for radioactivity to estimate tissue-bound digoxin.

 

64

 

 

 

 

150 l l l l l l
r -1
E 100 1- .4
ii
0: .- -
Q
8
-~ 50 r i
C:
0 1 l 1 J a J
0 60 120 150
Time (min)
Figure 15

Figure 15. Effects of LAD artery ligation on plasma digoxin con—
centration on anesthetized cats. Digoxin solution was infused at a
rate of 60 ug/kg/hr, and blood samples were obtained at various time
points and digoxin concentration in the plasma was determined by
radioimmunoassay. Open circles: control, unoccluded animals.

Filled circles: the left anterior descending coronary artery was
occluded 40 min prior to digoxin infusion. Each point represents the
mean of six experiments. Vertical lines indicate S.E. None of the
values were significantly different from the corresponding control

values (p<0.05

65

TABLE 3

Plasma Digoxin Concentrations at the Onset of
Digoxin-induced Ventricular Arrhythmias in
Coronary-occluded and Non-occluded
Anesthetized Cats

See legends for Figures 11 and 13. Blood samples
obtained from the femoral artery at the onset of ventri-
cular arrhythmias were centrifuged at 1000 rpm for 10
min. The plasma samples thus obtained were assayed for
digoxin by radioimmunoassay or by assaying for the
radioactivity in the plasma samples obtained from animals
infused with (3 H)digoxin (the last pair of groups in the
table). Each value represents the mean :_S. E. of five
experiments.

 

 

Animal Preparations Digoxin (nM)
Neurally intact 127.9: 4.6
Intact with occlusion 108.8: 2.9*
Bilateral vagotomy 116.4: 5.6
Vagotomy with occlusion 94.4: 4.8*
C-l section + vagotomy 151.3: 9.9
C-l section with occlusion 132.8: 3.8*

+ vagotomy
dl-propranolol + vagotomy 128.8: 9.7
dl-propranolol with occlusion 93.9:ll.2*
+ vagotomy

 

*Significantly different from the corresponding
control values (p<0. 05).

66
Figure 16 shows myocardial digoxin bound at the onset of arrhythmia in
both LAD artery occluded and control animals. Since the time to onset
of arrhythmias was longer in control than LAD artery occluded animals,
the total amount of digoxin infused to the LAD artery occluded animals
was less than that of the control animals. Digoxin concentrations in
hearts of control animals were significantly greater than that in LAD
artery occluded animals. Furthermore, ischemic areas of the heart
contained much less digoxin than non-ischemic areas, probably due to
reduced blood flow to these tissues following coronary occlusion.
These data indicate that the amount of digoxin bound to the LAD artery
occluded heart at the time of arrhythmias is less than that bound to
non-occluded hearts and therefore, the increased sensitivity of the
ischemic heart to the arrhythmogenic effects of digitalis is not by an
enhanced uptake of the glycoside by the ischemic myocardium.

3. Nerve Activity Studies

 

In order to further examine the role of sympathetic outflow
on the increased digitalis sensitivity of ischemic heart, the effects
of digitalis and coronary artery occlusion on the cardiac efferent
sympathetic nerve activity were studied. The inferior cardiac effer-
ent sympathetic nerve arising from the left stellate ganglion was
isolated and its electrical activity measured during a 60-sec occlu-
.sion of the left anterior descending coronary artery, in the absence
23nd presence of subtoxic and toxic doses of digoxin. LAD artery
cxsclusion before digoxin infusion to anesthetized cats caused variable
(affects on nerve activity (Figure 17), i.e., increases, decreases or

1K) change were observed. The mean changes in nerve activity from the

67

 

10.0 - . 2. e

 

\1
Ln
T
L

 

Digoxin uptake
(pmol/mg protein)
9‘
O .

. N
01
T

 

 

 

 

 

 

 

 

NI NI PI CI
Control Occluded

Figure 16

Figure 16. Effects of LAD artery ligation on myocardial digoxin
uptake. Anesthetized cats were infused with digoxin solution (60
pg/kg/hr) containing (3H)digoxin, and at the onset of arrhythmia, a
dye was further infused. Non-ischemic (NI) tissues from LAD artery
occluded and control animals, and partially (PI) and completely (CI)
ischemic tissues from occluded animals were obtained from stained,
Inedium-stained and non-stained tissues, respectively. Tissue digoxin
content was determined by assaying for the radioactivity of tissue
homogenates. Each bar represents the mean of five experiments.
Vertical lines indicate S.E. *Significantly different from the
«control value observed in unoccluded animals (p<0.05). **Signifi-
c:antly different from the control values observed in both occluded and
tion-occluded animals (p<0.05).

68

. me.ove
nopcme peep :p apw>ppue pocpcoo esp soap pcmcmeppe xppeeopmpcmpm mew: mmmpp>ppow
o>cmc wcp peep mpeopecp mzocc< .emEcowcme me: copmerooo ommuoo ecpep e .epsepxecce
mo pmmco asp p< .Apmcee mpeuesv emscoecma we: copm=_ooo umm om cospoce .mmoe
opscpzecce ceme esp po xm.¢Ho.mw eoeoemc :pxompe pepop oep cog; use Aee\mx\m:

oov empcepm we; :opmemcp :pxompo .eopcwe Amv >cm>oomc ommtom e mcpcze ece A v
:opm=_ooo mo mFe>cmpcp ommtom msozeppcoo emcep mercee .on newton Focpzoo ommuoo
e mcpcse xpp>ppoe mmecw>e opecpmspPp mpcpoa epea .Feepce wco cp em>comno mmmceco
app>ppoe m>cmc mpcmmmcemc mcpp :oem .empepsopeo mew; copm=~ooo mep ecoemn no>cmm
-no mmzpe> pocpcoo Eoce emceeo pcwocmq esp wee mpec emceeompe m>cme oepeceo :emz
.emmempmc xpxopze use owm on com eweepooo we: o<p ece emecoomc we: app>ppoe o>cmc
pocpcoo .Aenmp .cm>em3v emnpcomoe Apmsop>mca eczemooce mcp an emcoppcoe >_m:o::pp:oo
we: m>cmc oepeceo mo App>ppoe m>cwc peopcpompm .mpeo emNppmspmmce cw App>ppue
m>cmc oppmspeaezm oepeceo copcmecp co coppemp_ Acmpce o<p mo mpommmm .mp mczmpm

69

a

mo No

Ho

pp eeeepn

a

mo No

Ho

 

u

 

(abueuo auaoaad) KltAtlov enaaN oetpaeg

70
control during the three successive 20 sec periods were 109 + 8%, 91 +
11%, and 109 + 12%, respectively. Release of the occlusion returned
the nerve activity to control levels observed before the occlusion.
When digoxin infusion was started and the amount of infused digoxin
reached 83.0+4.3% of the arrhythmogenic dose, a similar 60-sec LAD
artery ligation caused either reduction or elevation of nerve acti—
vity; however, nerve activity tended to be lower than the response
observed prior to digoxin infusion. During the second 20-sec occlu-
sion period, reduction of nerve activity caused by LAD artery ligation
was statistically significant (66+12% 3g, 9l+ll%). LAD artery liga—
tion during digoxin-induced ventricular arrhythmias caused a further
decrease in the activity of the nerve. Thus, temporary LAD artery
ligation reduced the mean cardiac sympathetic nerve activity when
near-toxic or toxic concentrations of digoxin were present in the
animal.

LAD artery ligation was generally followed by either a
moderate decrease or no change in blood pressure. The blood pressure
decrease, when present, was transient. To determine whether such
changes in blood pressure might alter the response of the electrical
activity of the nerve caused by coronary artery occlusion, the blood
pressure was reduced to a similar extent by hemorrhage. As expected
in a baroreceptor-intact animal, reduction in blood pressure by
hemorrahge increased the firing of the nerve (data not shown). These
results demonstrate that reduction in the nerve activity observed

during LAD artery ligation was not due to changes in blood pressure.

71
It appears from above studies that ischemia-induced changes
in sympathetic input to the heart or vagal activity does not play a

significant role in the enhancement of arrhythmogenic effects of

 

digitalis on the ischemic heart. These results further support the
observation that a direct action of digitalis on the ischemic heart is
of primary importance in the increased sensitivity to digitalis.

C. Ischemia-induced Enhancement of the Arrhythmggenic Actions of

Digitalis: Involvement of EndogenousLy-Released Substances and
non-Uniform Digitalis Effect

 

 

 

'It has been shown that coronary artery occlusion causes release
of endogenous catecholamines from the nerve terminals in the early
stages of myocardial ischemia (Lammerant gt_gl,, 1966; Shahab gt_g13,
1969; Staszewska-Barczak, 1971). Reperfusion of the ischemic hearts
has been also shown to cause a marked increase in norepinephrine
concentration in the perfusate (Abrahamsson gt_gl,, 1981) suggesting
that the catecholamine concentration in the extracellular space was
elevated during ischemia. Coronary artery occlusion may also release
endogenous cardiac histamine. Histamine has been shown to be released
from the heart during immediate hypersensitivity reactions (Levi and
Burke, 1980). Since catecholamine and histamine at high concentra-
tions are arrhythmogenic in themselves and are known to potentiate
toxic digitalis actions on the heart, ischemia-induced release of
these substances may enhance the arrhythmogenic actions of digitalis.

Results from studies in sections A and B demonstrate that digi-
talis is unevenly distributed within the ischemic heart. Addition-

ally, extracellular K+ concentrations have been shown to be elevated

72
in ischemic areas of the heart, as a result of increased K+ permea-
bility. Such effects have been proposed to be responsible for is-
chemia-induced arrhythmias (Harris, 1966), since automatic activity
may be elicited by high K+. Therefore, the possibility that such
events cause elevation of toxic effects of digitalis on the ischemic
heart was examined.

1. Isolated Heart Studies: Involvement of Catecholamine and
Histamine

To examine the possibility that ischemia-induced release of
endogenous catecholamines or histamine may enhance the arrhythmogenic
effects of digitalis, guinea-pigs were pretreated with reserpine 24 hr
prior to sacrifice, or Langendorff preparations were pretreated with

10"5

M cimetidine, a H2 receptor blocker, respectively. In guinea-pig
hearts, the HZ-receptors have been demonstrated to mediate histamine-
induced idioventricular tachyarrhythmias (Levi 33.213’ 1976).
Addition of cimetidine to the perfusing solution caused an
approximately 15% reduction in developed tension of the heart. The
catecholamine depleting effects of reserpine or the HZ—receptor
blocking actions of cimetidine was tested by challenging with tyra-
mine-HCl or histamine, respectively, and obtaining a dose-response
curve for the positive inotropic effect of tyramine or histamine. The
dose of reserpine or cimetidine employed in the present study com-
pletely inhibited the positive inotropic effect of 1074M tyramine or

10"6

M histamine in the guinea-pig heart preparation (data not shown).
Following an equilibration period, the left anterior de—

scending coronary artery of the Langendorff preparation of guinea-pig

73
heart was completely occluded and 40 min later, the heart was perfused
at a constant pressure with 2.5 uM digoxin. Digoxin produced a
positive inotropic effect followed by arrhythmias in all preparations.
The onset of arrhythmias was earlier in LAD artery occluded than in
control preparations (Figure 18). Reserpine or cimetidine pretreat-
ment did not significantly alter the time to onset of arrhythmias in
either control or LAD artery ligated hearts. The types of digoxin-
induced arrhythmic contractions observed in both LAD artery occluded
and control preparations were similar to those shown in Figure 1.
There were no clear differences between the two groups with respect to
the types of arrhythmic contractions.

Although a separate effect of catecholamine depletion or
histamine antagonism did not affect the sensitivity of the LAD artery
occluded heart to the arrhythmogenic effect of digitalis, it is
possible than an additive effect may abolish the increased digitalis
sensitivity. Therefore, isolated heart preparations from guinea pigs
pretreated with reserpine were perfused with a solution containing
cimetidine. In these preparations also, the onset of digoxin-induced
arrhythmias was earlier in LAD artery occluded preparations (Figure
18). These results indicate that the increased arrhythmogenic effects
of digitalis in ischemic hearts can be observed in the absence of
endogenously released cardiac catecholamine, histamine or both. These
results partly support the earlier finding in intact cats that d1-
propranolol treatment also failed to influence the ischemia-induced

enhancement of digitalis toxicity.

74

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.. 1 .“l”
E t * j *
:20 ,1/ / /

Figure 18

Figure 18. Effects of LAD artery ligation on the time to the onset
of digoxin-induced arrhythmias in the isolated perfused guinea-pig
heart. Following an equilibration period, LAD artery was ligated and
digoxin added to the perfusing solution 40 min later (final concen-
tration, 2.5 uM) (shaded bars). Control preparations were also per-
fused with a similar concentration of digoxin (open bars). Reserpine,
when used, was injected 24 hr prior to sacrifice of the animal (R).
Cimetidine (final concentration, 10 uM) was added 40 min prior to
digoxin perfusion to either control (C) or reserpine-treated (R + C)
hearts. Each bar represents the mean of six experiments. Vertical
lines indicate S.E. *Significantly different from the corresponding
control value (p<0.05).

75

2. Isolated Heart Studies: Non-uniform Digitalis Effect
and KT

In hearts in which the LAD coronary artery is ligated and
then perfused with digoxin, an uneven glycoside distribution occurs
due to areas that are only partially perfused or not perfused at all
(Thompson gt_gl,, 1974). Therefore, in coronary artery-occluded
hearts, non-homogeneous glycoside effects will occur and non-uniform
electrophysiological characteristics such as an altered action poten—
tial duration or conduction velocity may result. It is believed by
some investigators (Weiss and Shine, 1981; Wit and Bigger, 1975) that
non-uniform alterations in these electrical events may cause the
development of arrhythmias. Therefore, possible contributions of such
non-uniform digitalis effects on the heart to the earlier generation
of arrhythmias were examined using Langendorff preparations of guinea-
pig heart.

To mimic a non-uniform digitalis distribution, the left
anterior descending coronary artery of the isolated guinea-pig heart
was cannulated. Thus, the heart was perfused via the coronary ar-
teries by way of the aorta and also separately through the cannulated
coronary artery. Following an equilibration period, an arrhythmogenic
concentration of digoxin was added only to the solution perfusing the
heart via the aorta such that the tissue areas perfused by the cannu-
lated artery were digoxin-free. The non-homogeneity of glycoside
distribution was visually identified by adding a dye only to the
solution perfusing the heart via the aorta. Control heart prepara-

tions were similarly cannulated but the entire heart was perfused with

76
digoxin by adding the glycoside to the solution perfusing the heart
via both the aorta and the cannulated artery. The time to the onset
of digoxin-induced arrhythmias was similar in the two groups of
preparations (Figure 198). Thus, a simple, non—uniform digitalis
distribution within the heart did not increase the sensitivity of the
heart to digitalis-induced arrhythmias.

The extracellular K+ concentration has been shown to be
elevated in ischemic tissues (Hill and Gettes, 1980; Hirche gt_gl,,
1980). Although the estimated values of extracellular K+ concentra-
tions were varied, they were generally between 8 mM and 14 mM. The
extracellular K+ concentration remained elevated when it was measured
30 min following ischemia. Therefore, the effects of such an elevated
extracellular K+ concentration on the arrhythmogenic actions of digi-
talis were further examined using isolated perfused guinea-pig heart
in which the left anterior descending coronary artery was cannulated
and separately perfused. Normal Krebs-Henseleit bicarbonate buffer
solution (pH 7.5) containing 5.8 mM KI was perfused through the aorta
whereas a modified solution containing 10.6 mM K+ was infused through
the cannulated artery. Such perfusion of the heart with two different
concentrations of K+ solution to different coronary arteries did not
cause arrhythmias to develop during the 90-min period under the
experimental conditions employed. An arrhythmogenic concentration of
digoxin (2.5 uM) was then added to the solution perfusing the heart
via the aorta but not to the solution perfusing the cannulated coro-

nary artery. Digoxin-induced arrhythmias appeared significantly

77

 

 

 

 

 

 

 

 

 

 

 

 

 

3n
1 1

TE 20 ” f -1

E

" I

0)

.E 10 r- -

....

0

LAD 1<+ 5.8 5.8 10.6 mM

A B c
Figure 19

Figure 19. Effects of non— —uniform digitalis perfusion with or
without locally elevated K+ concentration on the time to onset of
arrhythmias in the isolated, perfused guinea- pig heart. Left anterior
descending coronary artery was cannulated and Krebs- Henseleit bicar-
bonate buffer (pH 7. 5) solution containing either 5. 8 mM K+ and 2. 5 uM
digoxin (control, A), 5. 8 mM K+ with no digoxin (B) or 10. 6 mM K+ with
no digoxin (C) was infused while the rest of the heart was perfused
with 2. 5 uM digoxin solution through the coronary arteries via the
aorta. Development of arrhythmias was monitored from twitch-tension
curves. Each bar represents the mean of six experiments. Vertical
lines indicate S.E. *Significantly different from the values of other
bars (p<0.05).

78
earlier in these hearts than in the hearts entirely perfused with
solution containing 5.8 mM K+ and perfused with digoxin solution via
the aorta (Figure 19C). When the whole heart was perfused with a
solution containing 10.6 mM Kl, a similar concentration of digoxin
added to the solution perfusing the aorta produced a positive ino-
tropic effect with no arrhythmias occurring up to 90 min.

3. (3H)0uabain Binding

 

Non-uniform glycoside distribution and effects in hearts
perfused with digoxin through the aorta but not through the cannulated
LAD were further studied. Dye was perfused through the aorta at the
onset of arrhythmias. The fractional occupancy of glycoside binding
sites on Na,K-ATPase from the reduction in the initial velocity of
ATP-dependent (3H)ouabain binding to ventricular muscle homogenates
obtained from dye-stained (S) and non-stained (NS) myocardial tissues
were estimated. Control heart preparations were perfused for a simi-
lar period of time. Figure 20 shows the (3H)ouabain binding to dye-
stained and nonstained tissues. Since dye-stained tissues were per-
fused with digoxin, the glycoside binding sites were bound with I'cold"
digoxin and therefore the (3H)ouabain binding was low. In the non-
stained tissue, which was not perfused with digoxin, the (3H)ouabain
binding was high, indicating that glycoside binding sites were not
bound, as expected. To test whether infusion of a solution via the
cannulated artery in fact perfused the tissues, digoxin was infused to
the cannulated artery. The tissue was then assayed for (3H)ouabain

binding. The (3H)ouabain binding in this tissue was low, indicating

79

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

”£1.00

BE 1

15.00.75,. 1 T .
SE 7

.Q'r-

“80.50. * / * * .
SE Tr/ r‘f‘ T
05%;;ég () ﬁéij /12;

3 C 3 NS 5 NS

A B
Figure 20

Figure 20. Effects of non-uniform digoxin perfusion on specific
(3H)ouabain binding to ventricular muscle homogenates. LAD artery
of Langendorff preparations of guinea-pig heart was cannulated and
infused with solution containing digoxin and 5.8 mM K+ (B) or 5.8 mM
Ki (A) while the heart was perfused with a similar concentration of
digoxin and 5.8 mM K+. At the onset of arrhythmias, a dye was added
to the solution perfusing via aorta and stained (S), and non-stained
(NS) portions of ventricular homogenates, including that obtained from
control hearts perfused with no digoxin (C) were assayed for (3H)-
ouabain binding. (3H)ouabain binding shown is the difference in
values observed in the presence and absence of ATP. *Significantly
different from the control values (p<0.05).

80

that digoxin bound to the tissue normally perfused by the cannulated
artery (Figure 20). The dye itself did not affect the (3H)ouabain
binding reaction. These results show that separate perfusion to the
cannulated artery indeed produces a non-homogeneous glycoside effect
on the heart.

In summary, the presence of an area of the heart with an
elevated K+ concentration, adjacent to areas of the heart exposed to
digoxin appears to enhance the arrhythmogenic effects of digitalis.
It is tempting to speculate that similar conditions may be present in

acute myocardial ischemia and predispose the heart to digitalis-

induced arrhythmias.

DISCUSSION

A. Mechanisms of Toxic Actions of Digitalis in the Heart

 

Cardiac glycosides produce arrhythmias by a direct action on the
heart. This is readily shown by perfusion of isolated heart pre-
parations or superfusion of atrial, ventricular or papillary muscle,
or Purkinje fibers with toxic concentrations of a glycoside which
produce an initial positive inotropic effect followed by arrhythmias.
Digitalis also acts on the nervous system to enhance its arrhythmo-
genic actions. In particular, the potentiating action of the sympa—
thetic nervous system on digitalis-induced arrhythmias has been well
established.

It is generally accepted that the initial event in the toxic
action of cardiac glycosides is the binding to and inhibition of
sarcolemmal Na,K—ATPase. As described in the Introduction section,
the positive inotropic action of digitalis is closely related to
enzyme or sodium pump inhibition; however, the mechanism by which
toxic (arrhythmogenic) effects of digitalis are produced has not been
elucidated although sodium pump inhibition is thought to be the pri-
mary event. Additional changes in ion movement are proposed to
explain the toxicity.

The sodium pump actively counter-transports Na+ and K+ across the

cell membrane. Therefore, inhibition of the sodium pump causes

81

82
accumulation of extracellular K+ which in turn reduces the trans-
membrane K+ gradient and partially depolarizes the cells. Digitalis-
induced arrhythmias have been proposed to develop from the increase in
the inhibition of the sodium pump and progressive reduction in the
resting membrane potential that reach the threshold level for spon-
taneous depolarization and automatic activities. However, K+ outside
the cell may not accumulate enough to reduce the resting membrane
potential to the thresold potential since excess K+ may be washed out
by coronary blood flow.

Recently, several groups of investigators have observed that
digitalis causes the appearance of transient depolarizations (also
known as oscillatory afterpotentials, or delayed afterdepolarizations)
in Purkinje fibers treated with toxic concentrations of digitalis.
When the transient depolarizations reached the threshold potential,
action potentials were elicited (Ferrier gt_gl,, 1973; Rosen gt_gl,,
1975; Ferrier, 1977). Ferrier Si;§l: (1977) noted while recording
transmembrane potential simultaneously from canine Purkinje fibers and
attached papillary muscle that the triggered action potentials arose
from the myocardial conducting system (Purkinje fibers), conducted to
the adjacent papillary muscle and produced contractions in that
muscle. Papillary muscle, however, did not generate oscillatory
afterpotentials. These observations lead to the concept that digi-
talis produces arrhythmias by causing development of oscillatory
afterpotentials which are capable of producing extrasystoles or

arrhythmias.

83

Even when transient depolarizations do not attain the threshold
potential, they were shown to modify conduction (Rosen gt gt., 1975).
Since transient depolarizations can vary in magnitude from one cycle
to the next, they can modify the amplitude and upstroke velocity of
successive action potentials and conduction velocity. Thus, bigeminy
and "concealed" conduction were observed in myocardial tissues ex-
hibiting oscillatory afterpotentials.

Several studies using voltage clamp techniques have suggested
that the ionic mechanism responsible for the oscillatory afterpoten-
tials is a transient inward current (Lederer and Tsien, 1976; Aronson
and Gelles, 1977). The specific ion involved in the transient de-
polarization was further proposed to be Ca++ since calcium blockers
such as verapamil depressed the depolarization. Other investigators
have suggested that Na+ may also be involved since lowering of the
extracellular sodium concentration or TTX treatment reduced the
magnitude of the transient depolarizations (Tsien and Carpenter, 1978;
Vassalle and Scida, 1979). These results suggest that although the
mechanism is uncertain, the inward currents carried by Ca++ and/or Na+
are important for the development of digitalis-induced oscillatory
afterpotentials.

It is known that toxic concentration of cardiac glycosides cause
accumulation of intracellular calcium. It is possible that increases
in calcium in the cell may contribute to the appearance of oscillatory
afterpotentials. This view is supported by the observations that
raising extracellular calcium or reducing extracellular sodium, both

of which cause elevation of intracellular calcium, the latter probably

 

 

84
by the sodium/calcium exchange mechanism, were capable of producing
oscillatory afterpotentials (Kass gt_gl,, 1978). Digitalis-induced
increases in intracellular calcium may alter calcium fluxes within the
cell, i.e., cause calcium redistribution between intracellular stores,
and cause transient changes in membrane conductance to cations such as
sodium.

Thus, the putative scheme for the development of digitalis—
induced arrhythmias begins by the binding of a glycoside to sarco-
lemmal Na,K—ATPase and the inhibition of sodium pump which results in
an elevation of intracellular sodium. Via the sodium/calcium exchange
mechanism, sodium efflux is coupled to calcium influx and the intra-
cellular calcium level rises. Calcium redistribution in the cell and
possibly oscillatory movements of calcium between intracellular stores
may cause changes in membrane conductance in a phasic manner. Tran-
sient inward movements of sodium or calcium then produce transient
depolarizations. Toxic concentrations of digitalis inhibit the sodium
pump to an extent such that the above events are enhanced and tran-
sient depolarizations succeed in reaching the threshold potential
necessary to produce triggered action potentials or arrhythmias.

Another possible mechanism by which transient depolarization may
occur is by accumulation of K+ just outside the cell membrane, i.e.,
at the glycocalyx surface coat. The transient accumulation of the
potassium ions released during an action potential may partially
depolarize the cells. Digitalis, by inhibiting the sodium pump, may
cause a greater transient elevation of K+ outside the cell and cause

transient depolarizations to occur.

85

Ischemia may alter the effect of digitalis on the deve10pment of
transient depolarizations by interfering at any of the steps described
above. The effect of reduced or no coronary blood flow to the cardiac
tissue, or ischemia, has been intensively studied using the isolated
perfused heart model. These studies show that ischemia impairs all
aspects of cellular function, i.e., metabolic, electrophysiologic,
mechanical and structural. In cardiac tissues that are partially
perfused, however, cellular function may be depressed but not de-
stroyed. In these cells, the effects of digitalis may be altered at a
specific site if ischemia-induced damages are confined to a particular
function. For example, an enzyme that has a low Km rather than a high
Km for ATP may survive the partial ischemia when ATP synthesis is
reduced. Therefore, depending upon the severity of ischemia, the
sensitivity of the myocardial cells to ischemic insult may change.
Thus, the sensitivity of the cell to the effects of digitalis may be
changed by the severity of ischemia. In the present studies, the
effect of ischemia on the effect of digitalis was examined in hearts
perfused at a reduced rate of flow or zero flow. Some of the steps in
the mechanism of action of digitalis were studied to determine whether
ischemia caused any changes at glycoside binding sites.
B. Effects of Ischemia on Na,K-ATPase Activity, Sensitivity of the

Enzyme to the Inhibitory Effects of Digitalis, and’Glycoside
Binding to Na,K-ATPase

 

 

 

Effects of ischemia on the sensitivity to the arrhythmogenic
actions of digitalis was first examined using Langendorff preparations
of guinea-pig heart to determine whether a direct action of the glyco-

side on the ischemic heart is an important mechanism for the

86
development of the early onset of arrhythmias. When the left anterior
descending coronary artery of the isolated perfused heart was com-
pletely ligated and then the heart was challenged with a toxic con-
centration of digoxin, the onset of glycoside-induced arrhythmias was
significantly earlier in the ligated preparation than in the control,
non-ligated hearts. These results confirm the observations made by
other investigators that coronary artery occlusion reduced the dose of
digitalis required to develop ventricular arrhythmias in cats, dogs,
and pigs (Hood gt_gl,, 1967; Balcon gt gl,, 1968; Morris gt_gl,,
1969; Kumar gt_gl,, 1970; Ku and Lucchesi, 1979) and further demon-
strate that altered sensitivity to digitalis can be observed in the
isolated ischemic heart.

In myocardial ischemia, only a small portion of the heart becomes
ischemic due to lack of blood flow to this area. Therefore, both
ischemic and non-ischemic zones exist in the coronary artery-occluded
heart. The existence of a "border zone" which lies between the
ischemic and non-ischemic tissues has also been considered and studied
using various techniques. Surface fluorescence studies, histological
and ultrastructural studies, and histochemical and electrophysiologic
studies have generally failed to prove or disprove the concept of a
"border zone". Evidence for and against the existence of a border
zone was obtained in the studies of many investigators (Sayen gt_gl,,
1952; Reimer gt_gl,, 1977; Hearse gt_gl,, 1977; Harken gt_gl,, 1978;
Vokonas gt_gl,, 1978). However, flow studies (Lubbe gt_gl,, 1974;
Malsky gt_gl,, 1977) and electrocardiographic studies (Braunwald gt
31,, 1974; Janse gt_gl,, 1979) have indicated that zones with inter-

mediate flow or S—T segment elevations clearly exist. These studies

87
all suggest that although the size of the border zone may be small,
there probably exists a narrow zone of partially ischemic tissues
during acute myocardial ischemia. This assumption is reasonable and
experimentally sound since the infarct size in ischemic heart can be
reduced by cardiac drugs such as propranolol, indicating that there
exist border ischemic tissues that can be salvaged (Puri, 1974;
Braunwald gt gl,, 1974; Hearse gt_gl,, 1977).

In a coronary artery occluded heart, the completely ischemic
tissues would not be perfused whereas the partially ischemic tissues
present in the border zone would be perfused at a reduced rate. When
digitalis is perfused to the heart, the partially ischemic tissues,
however, may be exposed to a similar concentration of glycoside as the
non-ischemic tissues. Therefore, if a border zone has an increased
sensitivity to digitalis, toxic effects of the glycoside may appear
earlier at this site.

The direct action of digitalis on the heart may be on cardiac
sarcolemmal Na,K-ATPase, a putative receptor for the pharmacological
and toxic actions of digitalis (Repke, 1964; Schwartz gt_gl,, 1975;
Akera and Brody, 1977). Therefore, alterations in Na,K-ATPase acti-
vity, or sensitivity of the enzyme to digitalis may augment the
cardiac response to the toxic effects of digitalis. One suggestion is
that the cardiac electrogenic sodium pump is stimulated under anoxic
conditions (McDonald and MacLeod, 1971). Such an increase in sodium
pump activity has been shown to enhance the inhibitory effects of
digitalis (Clausen and Hansen, 1977; Yamamoto gt_gl,, 1979). On the
other hand, various results have been observed on the effects of

ischemia on myocardial Na,K-ATPase activity. In failing rabbit hearts

88

produced by constriction of the aorta, Na,K-ATPase activity of ventri-
cular muscle homogenates were significantly reduced (Yazaki and Fjuii,
1972). Coronary artery occlusion of the canine heart failed to alter
Na,K-ATPase activity when assayed 24 hr later but did depress the
enzyme activity after 7 days (Schwartz gt_gl,, 1973). In similar
studies, Beller gt_gl, (1976) noted that Na,K-ATPase activity of
partially purified enzyme preparation from ventricular tissue obtained
from ischemic areas of 2-hr coronary artery occluded canine heart was
markedly lower than that in non-ischemic tissue. Hypoxia, an import-
ant component of ischemia, also decreased Na,K-ATPase activity in both
isolated rat heart tissue (Balasubramanian gt gt., 1973) and dog heart
muscle (Friedman gt gl,, 1972). Rau gt_gl, (1977), however, did not
observe any alteration in enzyme activity following 60 min of anoxic
perfusion of rabbit heart muscle. The reasons for these different
results are not clear. It may be due to differences in the procedure
of enzyme preparation since slightly different methods were employed
by these investigators, the duration of the ischemia or other factors.

In the present study, the effects of ischemia on Na,K-ATPase
activity and glycoside binding to the enzyme were examined. Globally
ischemic guinea-pig hearts were utilized as the experimental model,
since large quantities of uniformly ischemic tissues could be obtained
and pathways of energy production and depletion were shown to be
similar to those produced jt_tjtg_by ischemia (Jennings gt_gl,, 1981).
As discussed above, the presence of a marginally ischemic zone may be
important in ischemia-induced augmentation of glycoside toxicity.

Therefore, cardiac tissues were made either partially or completely

89
ischemic by reducing the perfusion flow to the heart to either 5% of
the control flow rate by completely stopping the flow, to represent
marginally and totally ischemic zone, respectively. Results of the
present study show that following a 2-hr of partial or complete
ischemia, or 6-hr of partial ischemia, the Na,K-ATPase activity of the
ventricular muscle homogenates remains unchanged. In hearts that were
made completely ischemic for 6 hr, Na,K-ATPase activity and the number
of (3H)ouabain binding sites were significantly reduced. The affinity
of the binding sites for (3H)ouabain, however, was not affected. The
sensitivity of the remaining and active Na,K-ATPase to the inhibitory
effects of dihydrodigoxin was also unchanged. The affinity of Na,K-
ATPase for Na+ or K+, i.e., the sensitivity of the enzyme for K+-
induced inhibition or Na+-induced stimulation of glycoside binding,
was also unchanged. Thus, digitalis sensitivity of the ischemic
tissues was not altered by changes in the intrinsic properties of the
enzyme or changes in glycoside binding to the enzyme.

It is possible that since Na,K-ATPase activity or digitalis
binding was performed with muscle homogenates, ischemia-induced
changes in the ionic milieu to which the enzyme is exposed may have
been altered during homogenization. This is indeed likely since it
has been demonstrated that early ischemia causes K+ leakage into the
extracellular space (Harris, 1954; Hill and Gettes, 1980; Weiss and
Shine, 1981) and intracellular calcium accumulation (Shen gt gt.,
1972; Tomlinson and Dhalla, 1973; Nayler, 1981; Farber gt_g1,, 1981).
Such changes will alter Na,K-ATPase activity since a rise in extra-

cellular K+ stimulates, and a rise in intracellular Ca++ inhibits

 

90
sodium pump activity (,Dhalla gtgt., 1974). Thus, the lit vivo enzyme

sensitivity to the inhibitory effects of digitalis may be masked. In
addition, ischemia may cause an elevertion of the intracellular sodium
concentration by increasing membrane permeability or by non-specific

membrane leakage. Since the sodium cxoncentration in the extracellular
space is almost ten-fold higher than that in the cytosol, the gradient

itself may cause all influx of SOdlUHL. Such increases in the intra-

cellular sodium concentration have been shown to enhance glycoside
binding to Na,K-ATPase and to augment the toxic actions of digitans
(Yamamoto gt gt., 1979). Thus, increases in the intracellular Na+
concentration enhance and increases in the extracellular K+ concen-

tration reduce glycosidelbinding to Na,KeATPase. Ischemia-induced

changes in glycoside binding to the enzyme may or may not change,
depending upon the extent of redistribution of the ions.

In intact tissues, the changes in ionic gradients may not be as
quickly altered as in the homogenized preparations, since cellular
integrity is maintained. Therefore, monovalent cation transport
studies using intact ventricular strips and slices were performed. In
these studies, the effect of partial and complete ischemia on the
sodium pump activity and the reserve capacity of the sodium pump were
examined. A moderate inhibition of the sodium pump by the cardiac
glycosides or increasing the sodium influx by modest high frequency
electrical stimulation does not cause a corresponding increase in the
extracellular sodium concentration (Langer, 1972), indicating that the

sodium pump has a reserve capacity. However, inhibition of the sodium

pump exceeding the reserve capacity causes a progressive increase in

 

91
the intracellular sodium concentration and produces electrophysiolo-
gical 'hnbalance and finally toxicity «or arrhythmias. Thus, if is-

chemia reduces the reserve capacity of the sodium pump, digitalis-
induced toxicity will be facilitated. Results of the present study
show that ouabain-sensitive (Specifi<:) 86Rb and 42K uptakes, estimates
of sodium pump activity, to ventricular slices or strips prepared from
hearts made partially or completely ischemic for 2 hr are similar to
values observed in non-ischemic tissues. These studies further show
that the reserve capacity of the sodium pump, reduction of which en-
hances digitalis toxicity, is not altered by 2 hr of partial ischemia,
Thus, even in intact tissue preparations, ischemia failed to alter the
sodium pump activity, suggesting that glycoside binding to Na,K-ATPase
may be unchanged.

Although these data.indicate that 2 hr of ischemia does not
affect Na,K-ATPase or sodium pump activity, or the sensitivity of the
enzyme to digitalis, it appeared necessary to determine whether
ischemia produced by coronary artery occlusion, a situation closer to
myocardial ischemic disease in man, affects glycoside binding to Na,K-
ATPase or influences Na,K-ATPase activity itself. Since anesthetized

cats with LAD artery ligation required less digoxin to produce arrhyth-

mias. possible alterations in glycoside sensitivity of ischemic

tissues were examined in this animal species. Partially and complete-

ly ischemic tissues from LAD-ligated animals were visually identified
by dye staining after a 3 hr period or at the onset of digoxin-induced
arrhythmias. It is uncertain whether ischemic tissues obtained by dye

staining are in fact ischemic or represent tissues that have an

 

92
altered affinity for the dye. However, since the dye was infused
intravenously to the LAD artery ligated animal, tissues that are
normally not perfused will not be expected to be stained with the dye.
Flow distribution studies using radioactive microspheres have indi-
cated that ischemic tissues are considerably less well perfused than
non-ischemic tissues (Lubbe gt gt., 1974). Therefore, it is reason-
able to assume that ischemic tissues prepared under the present ex-
perimental conditions are indeed obtained from cardiac tissues that
have either partially or completely reduced perfusion.

When ischemic tissues, both partial and complete, were prepared
and assayed for Na,K-ATPase activity and (3H)ouabain binding, the
values were similar to those in non-ischemic tissues. In coronary
artery occluded animals infused with digoxin, higher enzyme activity
and (3H)ouabain binding were observed in ischemic tissues compared to
non-ischemic tissues at the onset of arrhythmias. Therefore, in
ischemic hearts, arrhythmias were produced with a lower fractional
occupancy of the glycoside binding sites on Na,K-ATPase. It is clear
from these studies that at a time when coronary artery occlusion
significantly enhanced the arrhythmogenic effects of digitalis, Na,K—
ATPase activity, and the number of glycoside binding sites are higher
in the ischemic tissues than in non-ischemic tissues. Thus, the
(mechanism by which digitalis produces enhanced the arrhythmic response
in the ischemic heart does not appear to be due to lower Na,K-ATPase
activity or greater digoxin binding to the enzyme. It is possible,
however, that non-uniform digitalis effects on the heart, observed in

coronary artery-occluded cats infused with digoxin, may contribute to

 

93

digitalis-induced arrhythmias. Myocardial digoxin uptake was lower in
coronary artery-occluded than in non-occluded hearts. Furthermore, in
coronary artery-occluded hearts, the myocardial digoxin uptake in
ischemic tissues was less than 10% of that in non-ischemic tissues.
Myocardial digoxin uptake in partially ischemic tissues was also
significantly less than non-ischemic tissue. These results are in
agreement with those of other investigators (Thompson gt_gl,, 1974;
Beller gt_gl,, 1976). The possibility that non-uniform effects of
digitalis influences arrhythmogenic actions of digoxin is discussed
below.

Reperfusion following a 2 hr ischemia, on the other hand, caused
a significant reduction in Na,K-ATPase activity without altering the
sensitivity of the residual enzyme to digitalis inhibition. This
finding is not surprising in view of the fact that reperfusion can
cause deleterious effects due to hemorrhage (Bresnahan gt_gl,, 1974)
and also can alter metabolic function due to intracellular accumula-
tion of calcium early following reperfusion (Kane gt_gl,, 1975;
Nayler, 1981). Consistent with these findings are the results that
sodium pump activity in ventricular slices following 20 min reperfu-
sion after 2 hr of ischemia was significantly reduced. These results
are in contrast to those obtained by Ku and Lucchesi (1979). In their
study, sodium pump activity in sodium-loaded ventricular slices ob-
tained from border ischemic and completely ischemic tissues in coro-
nary artery-occluded and then reperfused canine heart was signifi-
cantly increased compared to control tissues. Furthermore, the

sensitivity of the sodium pump to the inhibitory effects of ouabain

 

94

was increased. However, the jg_vivo significance of these results are

 

not clear since sodium pump activity estimated in sodium-loaded
tissues does not reflect the actual sodium pump activity in the
tissues. In the present studies, it was not experimentally feasible
to examine whether reperfused hearts were more sensitive to the
arrhythmogenic effects of digitalis since reperfusion itself caused
the development of arrhythmias. However, reperfusion may potentiate
digitalis toxicity by reducing the number of functional Na,K-ATPase
molecules, and thus the reserve capcity of the sodium pump.

It is possible that myocardial ischemia may alter factors other
than the glycoside-Na,K-ATPase interaction to sensitize the heart to
the arrhythmogenic action of digitalis. Such possibilities include
changes in glycoside pharmacokinetics such that the glycoside effect
on the heart may be altered in a non-uniform manner, or by changes in
myocardial glycoside uptake. Though unlikely, changes in glycoside
binding to sites other than Na,K-ATPase may affect the cellular re-
sponse to the toxic effects of digitalis.

Changes in glycoside pharmacokinetics, such as an altered body
distribution, may have reduced the dose of digitalis required to
produce arrhythmias in intact coronary artery-occluded animals. In
these animals, occlusion of the left anterior descending coronary
artery reduced cardiac output and left ventricular pressure (Hood gt
g1,, 1967, 1969; Banka gt gl,, 1975). Such effects may reduce blood
perfusion to body tissues and organs, and elevate the plasma glycoside

concentration. Digitalis absorption following oral administration

 

95
was further shown to be slower in animals in which the coronary artery
was occluded (Korhonen gt_gl,, 1979). However, in the present experi-
ments digoxin was infused intravenously. Digoxin infusion at a con-
stant rate caused a linear increase in the plasma glycoside concen-
tration; however, the rate of increase was similar in both occluded
and non-occluded animals. Therefore, the plasma digoxin concentra-
tions at any given time were similar in the two groups. The onset of
glycoside-induced arrhythmias was earlier in LAD artery occluded than
in control animals and therefore, the plasma digoxin concentrations at
the onset of arrhythmias were lower in the former than in the latter
group of animals. Table 3 shows that the plasma digoxin concentration
in dl-propranolol pretreated animals are significantly lower than in
the C-1 sectioned animals despite the longer infusion time in the
former. This is apparently due to differences in the assay procedure.
In the C-1 sectioned animals, a radioimmunoassay was used, whereas in
propranolol-pretreated animals plasma samples were assayed for radio-
activity following 3H-labeled digoxin infusion.

It can be concluded from above results that a direct action of
digitalis on the ischemic heart is an important mechanism for the
enhanced arrhythmic response to digitalis. However, it does not
appear to be due to changes in Na,K-ATPase activity, the number of
glycoside binding sites or in their affinity for the glycosides or
changes in sensitivity of the enzyme to the inhibitory effects of

digitalis.

96

C. Role of Sympathetic Nervous System in Ischemia-Induced Enhance-
ment of Arrhythmogenic Effects of Digitalis

 

 

The myocardial ischemia produced in this study by coronary artery
occlusion in anesthetized cats reduced the dose of digoxin to produce
glycoside-induced ventricular arrhythmias. These results are similar
to those reported by other investigators who used either cats, dogs or
pigs (Bellet gt_gl,, 1934; Hood gt gl,, 1967; Balcon gt_gl,, 1968;
Morris EE;El:’ 1969; Kumar gt_gl,, 1970; Ku and Lucchesi, 1979).

Using the anesthetized cat as a model, therefore, the role of auto-
nomic nervous system on ischemia-induced digitalis-sensitization of
the heart was examined.

Several groups of investigators have observed marked elevation of
plasma and urine norepinephrine and epinephrine concentrations in
patients with myocardial ischemia and in experimental animals with
coronary artery occlusion (Forssman gt_gl,, 1952; Raab and Gigee,
1954; Valorie gt_gl,, 1967; Klein gt_gl,, 1968; Staszewska-Barczak and
Ceremuzynski, 1968; Siggers gt gl., 1971). These findings suggest
that ischemia increases sympathetic activity either centrally or
peripherally and causes catecholamine release from nerve terminals
and/or the adrenal medulla.

It is well known that increased sympathetic nerve activity,
by either centrally or peripherally-induced, is closely related to the
production of digitalis-induced arrhythmias (Gillis and Quest, 1979).
Exogenous catecholamines have also been demonstrated to enhance
digitalis toxicity (Angelucci, 1966; Morrow, 1967; Raper and Wale,
1969; Tse and Han, 1974; Gillis and Quest, 1979). Studies employing

97
spinal cord section (Raines gt_gl,, 1967; Gills gt_gl,, 1972; Somberg
gt_gl,, 1978), cardiac denervation (Mendez gt_gl,, 1961), agents that
depress central sympathetic outflow (Gillis gt gt., 1972), drugs that
block either ganglionic transmission (Gillis gthl,, 1975) or beta-
adrenergic receptors (Kelliher and Roberts, 1974; Evans gt_gl,, 1976),
all clearly demonstrate that sympathetic discharge to the heart can
potentiate digitalis toxicity. Therefore, if ischemia enhances
sympathetic outflow to the heart or to the adrenal glands to release
catecholamines, digitalis toxicity may be potentiated. In the present
study, when sympathectomy was performed by sectioning the spinal cord
at the C-1 level to abolish the central sympathetic outflow, the dose
of digoxin required to produce ventricular arrhythmias was signifi-
cantly increased in these animals. These results give support to the
important role of the sympathetic nervous system in digitalis—induced
arrhythmias. S pinal section at C-1, however, failed to abolish the
effect of ischemia to enhance the sensitivity of the heart to the
arrhythmogenic action of digitalis. These results suggest that the
mechanism of increased digitalis sensitivity is by an effect other
than a centrally—mediated change in sympathetic activity produced by
ischemia.

Several investigators have suggested the importance of the role
of a cardio-cardiac sympathetic reflex in the mediation of ischemia-
induced elevation of sympathetic discharge to the heart (Malliani gt
gl,, 1969; Uchida and Murao, 1974; Bosnjak gt_gl,, 1979). Other

investigators have disputed these findings based on their observations

98
that this reflex was not present in neurally-intact animals (Felder
and Thames, 1981) or that sympathoinhibition occurs in animals in
which coronary artery was occluded (Thoren, 1973; Kedzi gt_gl,, 1974).
The cardio-cardiac sympathetic reflex system, whether it reduces or
elevates cardiac efferent sympathetic nerve activities, would not be
removed by C-l section since brain regions above C-l level may not be
involved in this reflex. Therefore, in order to eliminate the effect
of this reflex on the heart, in addition to any central sympathetic
effects, dl-propranolol was administered to the anesthetized cats
prior to coronary occlusion or digoxin infusion. dl-Propranolol
significantly prolonged the time to glycoside-induced arrhythmias in
both LAD artery ligated and control animals when digoxin was infused
at a constant rate of 60 ug/kg/hr. It failed, however, to abolish the
enhanced arrhythmic response observed in LAD artery ligated animals.

These results demonstrate the lack of significant involvement of the

 

sympathetic nervous system in the elevated digitalis sensitivity of
the ischemic heart. Since the beta-adrenergic blocking agent, d1-
propranolol, was used, one can additionally rule out the possibility
that ischemia-induced sympathoadrenal activation that elevates plasma
catecholamine concentration (Staszewska-Barczak, 1971; Karlsberg gt
gl,, 1981) or that locally-mediated increase in norepinephrine release
from the nerve terminals in ischemic tissues of the heart (Lammerant
gt gl,, 1966; Staszewska-Barczak and Ceremuzynski, 1968; Shahab gt
gl,, 1969; Abrahamsson gt_gl,, 1981) play an important role.

Studies using nerve activity recordings have shown that coronary

artery occlusion or digitalis can elicit either a decrease, increase

 

99
or no change in cardiac efferent nerve activity (Lathers gt_gl,,
1977, 1978). It was therefore postulated that such non-uniform neural
discharges observed in individual nerve filaments within the same
nerve bundle may alter myocardial excitability and conduction in a
non-uniform manner and cause development of arrhythmias. Weaver and
co-workers have additionally demonstrated that coronary artery occlu-
sion alters renal nerve activity (Weaver gt_gl,, 1981). Thus, coro-
nary artery occlusion influences nerve activities not only of those
related to the heart but also to other organs. It is possible that
coronary artery occlusion may affect the activity of the afferent
nerves arising from parts of the body other than the heart such as
kidney, baroreceptors and carotid sinus and influence the efferent
cardiac sympathetic nerve activity. Digitalis action on the heart may
thus be altered by such reflex mechanisms. Furthermore, if the above
hypothesis is correct, the effects of both ischemia and digitalis to
cause non-uniformity in the nerve activities may potentiate each other
and enhance digitalis toxicity.

In the present study with anesthetized cats, inferior cardiac
efferent nerve activity was monitored during one minute coronary
artery occlusions in the presence and absence of digoxin. LAD artery
ligation itself caused either a decrease, increase or no change in
nerve activity; results that support the findings of Lathers gt g1,
(1978) and others. However, the mean electrical activity of the nerve
was not different from the control activity. In the presence of near-
toxic or toxic doses of digoxin, however, the mean nerve activity was
significantly reduced during part of the occlusion, although in-

creases, decreases and no changes in nerve activity were still

 

100
observed. Thus, coronary artery occlusion-induced sympathoinhibition
in the presence of digoxin may have a slight protective effect against
digitalis toxicity, similar to the effects of spinal-section or
propranolol treatment.

The cause of sympathoinhibition during coronary artery occlusion
has been identified to be due to activation of vagal afferent nerves
which reflexly inhibit sympathetic outflow (Recordati gt_gl,, 1971;
Thoren, 1973; Kedzi gt_gl,, 1974; Felder and Thames, 1979; Thames,
1979). It is possible that a reduction in sympathetic activity in the
presence of digitalis-induced vagal activation may cause A-V block and
produce ventricular arrhythmias. Bilateral vagotomy, however, failed
to affect either the dose of digoxin to produce arrhythmias, or the
increased sensitivity of the ischemic heart to the glycoside. Thus,
any changes in sympathetic nerve activity caused by coronary artery
occlusion do not appear to play a significant role in the increased
digitalis sensitivity of the ischemic heart.

From these studies, it appears that removal of autonomic input to
the heart does not affect the ischemia-induced enhancement of arrhyth-
mogenic effects of digitalis. Therefore, effects of digitalis on the
autonomic nervous system may not be Significantly affected by is—
chemia. These results confirm the previous conclusion that ischemia
enhances a direct action of digitalis on the heart.

0. Role of Ischemia-induced Release of Catecholamines, Histamine and
KT in Digitalis Toxicity

 

 

The toxic actions of digitalis on the heart may be enhanced by

many factors. Of these factors, effects of norepinephrine, histamine

 

101
and K+ were examined since these substances have been shown, either
directly or indirectly, to be released from ischemic hearts, and to
produce arrhythmias at high concentrations. The potentiating effects
of catecholamines in the arrhythmogenic actions of digitalis have been
well demonstrated. Pretreatment of animals with reserpine (Ciafalo gt_
gl,, 1967), beta-adrenergic receptor blocking agents (Vaughan Wil-
liams, 1963; Evans gt_gl,, 1976) increased the dose of digitalis
required to produce cardiac arrhythmias. Procedures that blocked
central sympathetic outflow such as by spinal section (Gillis gt gl,,
1972; Somberg gt_gl,, 1978) or central aZ-adrenoceptor stimulation
with clonidine (Lechat and Schmitt, 1982) also desensitized the
animals to the arrhythmogenic effects of digitalis. In a related
study, animals infused with norepinephrine or epinephrine were also
sensitized to the toxic effects of digitalis (Morrow, 1967; Lum gt_
gl,, 1977). Thus, if ischemia increases catecholamine levels in the
heart, digitalis toxicity would be expected to be enhanced.

Studies to determine whether ischemia causes release of endo—
genous catecholamines have indicated that a sustained norepinephrine
secretion from the heart occurs following coronary artery occlusion in
dogs (Staszewska-Barczak, 1971). Similar studies have also demon-
strated that coronary artery occlusion increased the norepinephrine
concentration in the coronary effluents (Richardson, 1963; Lammerant
gt_gl,, 1966; Ceremuzynski, 1969; Shahab gt_gl,, 1969). In isolated
guinea-pig heart made ischemic and reperfused, the norepinephrine
concentration in the effluent was significantly increased (Abrahamsson

gt_gl,, 1981). These results suggest that ischemia—induced release of

 

102
catecholamine in the isolated heart may enhance the arrhythmogenic
effects of digitalis.

To study such a possibility, the endogenous norepinephrine from
the nerve terminals were depleted by pretreatment of guinea-pigs with
reserpine 24 hr prior to sacrifice. Langendorff preparations obtained
from these guinea pigs did not respond to a dose of tyramine (10'4M)
that caused a marked positive inotropic effect in control prepara-
tions. The result that coronary artery occlusion of the heart still
reduced the time to the onset of arrhythmias following perfusion with
a toxic dose of digoxin signifies that either ischemia did not cause
enough catecholamine release to potentiate the digitalis effect or
that locally-released catecholamines had no significant effect on the
ischemia-induced enhancement of the toxic actions of digitalis. Since
norepinephrine release by ischemia was not determined in the present
study, neither the former nor the latter possibility can be ruled out.
It is clear, however, that ischemia potentiates the arrhythmogenic
potential of digitalis even in the absence of catecholamines. There-
fore, a factor or factors other than ischemia-induced release of
catecholamines seem to be important.

Although not experimentally proven, results from several studies
suggest that ischemia may cause the release of endogenous cardiac
histamine. During immediate hypersensitivity reactions, histamine was
released together with other mediators such as prostaglandins (Levi
and Burke, 1980). Since prostaglandins are released during acute
myocardial ischemia (Alexander gt_gl,, 1975; Kraemer gt_gl,, 1976;

Berger gt_gl,, 1976; Ogletree gt gt., 1977), histamine may also be

103
released simultaneously. Histamine was also suggested to be one of
the agents responsible for coronary arterial spasms (Ginsburg gt_gl,,
1981).

Both H1 and H2 antagonists were shown to increase the dose of
digitalis required to produce cardiac arrhythmias (Levi and Capurro,
1975; Somberg gt_gl,, 1980). In isolated hearts, however, the posi-
tive inotropic, chronotropic and arrhythmogenic effects of histamine
were demonstrated to be mediated exclusively by HZ-receptors, since
the effects were antagonized by HZ-blockers and mimicked by H2-
agonists, whereas H1-antagonists were without effect (Levi gt_gl,,
1976). Therefore, if ischemia causes histamine release, the arrhyth-
mogenic effects of digitalis may be enhanced. Pretreatment with cime-
tidine, however, failed to abolish the LAD artery ligation-induced
reduction in time to onset of arrhythmias during perfusion of isolated
hearts with a toxic concentration of digoxin. In hearts obtained from
reserpine-treated guinea pigs, cimetidine pretreatment produced
similar results. Thus, neither catecholamines or histamine nor the
combination both are responsible for the early arrhythmias observed in
LAD artery ligated hearts perfused with digitalis.

Studies with coronary artery-occluded isolated guinea-pig hearts
and intact, anesthetized cats have indicated that digitalis is not
uniformly distributed. Similar unequal digitalis binding in ischemic
and non-ischemic tissues were also observed by other investigators
(Thompson gt_gl,, 1974; Beller gt_gl,, 1976). Therefore, such non-
uniform glycoside distribution resulting in non-homogeneous glycoside

effect may predispose the heart to the development of arrhythmias by

104
non-uniformly altering the action potential duration and conduction in
the myocardium. Such a system was simulated by cannulation of the
left anterior descending coronary artery in isolated guinea-pig heart,
and by differential perfusion of the heart with digitalis. In this
model, the time to the glycoside-induced arrhythmias was, however, not
different from that observed in uniformly perfused hearts. Thus, a
non-uniform glycoside effect produced under the present experimental
conditions did not reduce the tolerance of the heart to the arrhyth-
mogenic effects of digitalis.

On the basis of these data and results that LAD artery ligation
enhances the toxic actions of digitalis, the increased sensitivity of
an ischemic heart to digitalis appears to be the presence of the
glycoside in normal (non-ischemic) tissue adjacent to an ischemic
zone. Since the concentration of digitalis in ischemic tissues is
low, it may be speculated that electrophysiological changes in is-
chemic tissues play an important role. Thus, non-uniform alterations
in electrical events occurring in border zones may produce arrhyth-
mias.

Ischemic cells have been shown to release K+ to the extracellular
space by increasing the membrane permeability to this ion (Harris gt
gl,, 1954; Hill and Gettes, 1980). Since elevation of extracellular
K+ partially depolarizes myocardial cells, it was hypothesized that
the local liberation of K+ is an important factor in ectopic excita-
tion during early ischemia and necrosis (Harris, 1966). It was

further speculated that the elevated concentration of K+ within a

105
region and the resulting K+ gradient at the boundary produce an in-
creased excitability to electrical stimulation and injury current
within the ischemic region, both of which summate and generate ectopic
impulses (Hoffman, 1966; Hirshe gt_g1,, 1980; Janse gt_g1,, 1980).
Therefore, in the ischemic myocardium with no arrhythmic beats,
digitalis may enhance these electrical events and produce arrhythmias.
The hypothesis that an elevated extracellular K+ concentration in the
ischemic zone augments the arrhythmogenic action of digitalis was
tested using a similar separate-perfusion guinea-pig heart model.
Instead of a normal solution containing 5.8 mM K+, a solution con-
taining 10.6 mM K+ was perfused via the cannulated LAD artery, while
the rest of the heart was perfused with a solution containing normal
K+ and digoxin. Since the extracellular K+ concentration in acutely
ischemic tissues was found to be between 8 and 12 mM, 10.6 mM K+ was
chosen as the value to be used for the present experiments. The
rationale behind this experimental procedure was to produce a model as
close to an intact ischemic heart as possible. In this model, the
onset of glycoside-induced arrhythmias was significantly earlier than
that in control hearts perfused with normal K+ and digoxin to the
whole heart. Since the differential perfusion by itself did not
produce arrhythmias, these results clearly demonstrate that the
presence of an area with elevated extracellular K+ concentrations
within the heart, even in the absence of ischemia, is sufficient to
augment the toxic action of digitalis. Whether this is due to the
proposed mechanism described above is still a question to be answered.

However, the present results, despite the well known observation that

106
K+ antagonizes glycoside binding to Na,K—ATPase and thus reduces the
digitalis action upon the heart, suggest that the mechanism involved
is a complex interaction of ionic events occurring in tissue areas in
contact with both non-ischemic and ischemic tissues.

Several mechanisms by which the above phenomenon could occur may
be proposed. It has been shown that ischemia causes a reduction in
the resting membrane potential of myocardial cells (Carmeliet gt_gl,,
1969; Wit and Bigger, 1975). It has also been reported that under
certain conditions ischemic cardiac cells can generate propagated and
automatic action potentials (Carmeliet gt gl,, 1969; Cranefield gt_
gl,, 1971). These action potentials were due to slow inward currents
mainly carried by calcium ions and therefore were named "slow re-
sponses". Slow responses were elicited in the ischemic myocardium
when a high concentration of catecholamines or K+ was administered to
the ischemic area (Scherlag gt_gl,, 1974). This is probably due to
catecholamine- or K+-induced enhancement of slow inward current.
Digitalis may cause the appearance of automatic activity by contri-
buting to the depolarization process such that the slow inward current
is augmented and slow responses result. Since the completely ischemic
areas are not exposed to digitalis, only the border ischemic areas may
undergo such electrophysiological events. This concept supports the
importance of border ischemic tissues in the genesis of arrhythmias.

Elevation of extracellular K+ causes a shortening of the action
potential duration and a lengthening of conduction time (Weiss and
Shine, 1981). Local changes in these parameters have been proposed to

be responsible for arrhythmias observed in myocardial ischemia (Han

107
and Moe, 1964; Wit and Bigger, 1975; El-Sherif gt_gl., 1977). There-
fore, in ischemic conditions without arrhythmias, digitalis may
enhance the non-uniform changes in refractory period and conduction,
and elicit arrhythmias early.
In conclusion, the enhanced arrhythmogenic effects of digitalis
in acute myocardial ischemia may be due to regional extracellular K+

accumulation.

SUMMARY AND CONCLUSIONS

Myocardial ischemia produced by LAD artery ligation enhanced the
arrhythmogenic effects of digitalis in both intact animals and in
(isolated perfused heart preparations. Removal of the autonomic input
to the heart failed to modify this effect. The direct action of
digitalis on the heart appears to be the primary mechanism for the
phenomenon.

Ischemia produced by LAD artery ligation or by reduced perfusion
of the entire heart failed to alter Na,K-ATPase activity, glycoside
binding to the enzyme or sodium pump activity. Partial ischemia also
did not affect the reserve capacity of the sodium pump. Infusion of a
high K+, but not normal K+, solution to the LAD artery caused an
earlier onset of digoxin-induced arrhythmias. It may be speculated
that the mechanism by which acute myocardial ischemia predisposes the
heart to the toxic effects of digitalis is due to an ischemia-induced

eleVation of extracellular K+.

108

BIBLIOGRAPHY

BIBLIOGRAPHY

Abrahamsson, T., Almgren, 0. and Carlsson, L.: Ischemia-induced
release of noradrenaline from the isolated perfused rat heart.
Influence of perfusion substrate and duration of ischemia. J.
Mol. Cell. Cardiol. 1;; 1, 1981.

Akera, T. and Brody, T.M.: The role of Na+,K+-ATPase in the inotro-
pic action of digitalis. Pharmacol. Rev. 29; 187-220, 1977.

Akera, T. and Brody, T.M.: Myocardial membranes: Regulation and
function of the sodium pump. Ann. Rev. Physiol. 45; 375-388,
1982.

Akera, T. and Cheng, V-J.K.: A simple method for the determination
of affinity and binding site concentration in receptor binding
studies. Biochim. Biophys. Acta 319; 412-423, 1977.

Akera, 1., Bennett, R.T., Olgaard , M.K. and Brody, T.M.: Cardiac
Na ,K+-adenosine triphosphatase inhibition by ouabain and myo-
cardial sodium: A computer simulation. J. Pharmacol. Exp. Ther.
199; 287-297, 1976.

Akera, T., Brody, T.M., So, R.H.-M., Tobin, T. and Baskin, S.I.:
Factors and agents that influence cardiac glycoside-Na,K-ATPase
interaction. Ann. N.Y. Acad. Sci. 242; 617-634, 1974.

Akera, T., Larsen, F.S. and Brody, T.M.: The effect of ouabain on
sodium-and potassium-activated adenosine triphosphatase from the
hearts of several mammalian species. J. Pharmacol. Exp. Ther.
119; 17-26, 1969.

Akera, T., Olgaard, M.K., Temma, K. and Brody, T.M.: Development of
the positive inotropic action of ouabain: Effects of transmem-
brane sodium movement. J. Pharmacol. Exp. Ther. ggg; 675-684,
1977.

Alexander, R.W., Kent, K.M., Pisano, J.J., Keiser, H.R. and Cooper,
T.: Regulation of postocclusive hyperaemia by endogenously
synthesized prostaglandins in the dog heart. J. Clin. Invest.
_55: 1174-1181, 1975.

109

110

Allen, J.C. and Schwartz, A.: A possible biochemical explanation for
the insensitivity of the rat to cardiac glycosides. J. Phar-
macol. Exp. Ther. 1§§: 42-46, 1969.

Allen, J.C., Martinez-Maldonado, M., Eknoyan, G., Suki, W.N. and
Schwartz, A.: Relation between digitalis binding in vivo and
inhibition of sodium, potassium-adenosine triphosphatase in
canine kidney. Biochem. Pharmacol. 29; 73-80, 1971.

Angelucci, L., Lorentz, G. and Baldieri, M.: The relation between
noradrenaline content of rabbit heart muscle and the amount of K-
strophanthin needed to produce arrhythmias. J. Pharm. Pharmacol.
18; 775-782, 1966.

Aronson, R.S. and Cranefield, P.F.: Electrical activity of canine
cardiac Purkinje fibers in sodium-free, calcium-rich solution.
J. Gen. Physiol. t1; 786-808, 1973.

Aronson, R.S. and Gelles, J.M.: The effect of ouabain, dinitrophenol
and lithium on the pacemaker current in sheep cardiac Purkinje
fibers. Circ. Res. 39; 5174524, 1977.

Balasubramanian, V., McNamara, D.B., Singh, J.N. and Dhalla, N.S.:+ +
Biochemical basis of heart function. X. Reduction in the Na ,K -
stimulated ATPase activity in failing rat heart due to hypoxia.

Can. J. Physiol. Pharmacol. 51; 504-510, 1973.

Balcon, R., Hay, J. and Sowton, E.: Hemodynamic effects of rapid
digitalization following acute myocardial infarction. Brit.
Heart J. g9; 373-376, 1968.

Banka, V.S., Chadda, K.D., Bodenheimer, M.M. and Helfant, R.H.:
Digitalis in experimental acute myocardial infarction: Differ-
ential effects on contractile performance of ischemic, border and
nonischemic ventricular zones in the dog. Am. J. Cardiol. §§;
801-808, 1975.

Baum, G.L., Dick, M.M., Blum, A., Kaupe, A. and Carballo, J.: Fac-
tors involved in digitalis sensitivity in chronic pulmonary
insufficiency. Am. Heart J. t1; 460-642, 1959.

Beller, G.A. and Smith, T.W.: Digitalis toxicity during acute
hypoxia in intact conscious dogs. J. Pharmacol. Exp. Ther. 12g;
963-968, 1975.

Beller, G.A., Conroy, J. and Smith, T.W.: Ischemia-induced altera-
tions in myocardial Na+,K+-ATPase and cardiac glycoside binding.
J. Clin. Invest. 52; 341-350, 1976.

Beller, G.A., Smith, T.W., Abelmann, W.H., Haber, E. and Hood, W.B.
Jr.: Digitalis intoxication: A prospective clinical study with
serum level correlations. N. Engl. J. Med. 281; 989-997, 1971.

111

Bellet, S., Johnston, C.G. and Schecter, A.: Effect of cardiac

infarction on the tolerance of dogs to digitalis. Arch. Int.
Med. t3; 509-516, 1934.

Berger, H.J., Zarbet, B.L., Speroff, L., Cohen, L.S. and Wolfson, 5.:
Regional cardiac prostaglandin release during myocardial ischemia
in anesthetized dogs. Circ. Res. gg; 566-571, 1976.

Berti, F. and Shore, P.A.: Interaction of reserpine and ouabain on
amine concentrating mechanisms in the adrenergic neurone.
Biochem. Pharmacol. 1t; 2271-2274, 1967.

Bonting, S.L., Simon, K.A. and Hawkins, N.M.: Studies on sodium-
potassium-activated adenosine triphosphatase. I. Quantitative
distribution in several tissues of the cat. Arch. Biochem.
Biophys. 9t; 416-423, 1961.

Bosnjak, Z.J., Zuperku, E.J., Coon, R.L. and Kampine, J.P.: Acute
coronary artery occlusion and cardiac sympathetic afferent nerve
activity. Proc. Soc. Exp. Biol. Med. 1§15 142-148, 1979.

Boyajy, L.D. and Nash, C.B.: Influence of reserpine on arrhythmias,
inotropic effects and myocardial potassium balance induced by
digitalis materials. J. Pharmacol. Exp. Ther. 1113 193-201,
1965.

Boyajy, L.D. and Nash, C.B.: Alteration of ouabain toxicity by
cardiac denervation. Toxicol. Appl. Pharmacol. 9; 199-208, 1966.

Braunwald, E., Maroko, P.R. and Libby, P.: Reduction of infarct size
following coronary occlusion. Circ. Res. §§; 192-201, 1974.

Bresnahan, G.F., Roberts, R. and Shell, W.E.: Deleterious effects
due to hemorrhage after myocardial reperfusion. Am. J. Cardiol.
§§3 82-86, 1974.

Brown, A.M.: Excitation of afferent cardiac sympathetic nerve fibers
during myocardial ischemia. J. Physiol. London 199; 35-53, 1967.

Brown, A.M. and Malliani, A.: Spinal sympathetic reflexes initiated
by coronary receptors. J. Physiol. London 212: 685-705, 1971.

Cagin, N.A., Somberg, J., Bounous, H., Mittag, T., Raines, A. and
Levitt, B.: The influence of spinal cord transection on the
capacity of digitoxin to induce cardiotoxicity. Arch. Int.
Pharmacodyn. Ther. 291: 340-347, 1974.

Carmeliet, E. and Vereecke, J.: Adrenalin and the plateau phase of
the cardiac action potential. Importance of Ca++, Na+, and K+
conductance. Pflugers Arch. Ges. Physiol. 31g; 300-315, 1969.

112

Cattell, M. and Gold, H.: The influence of digitalis glycosides on
the force of contraction of mammalian cardiac muscle. J. Phar-
macol. Exp. Ther. gg; 116-125, 1938.

Ceremuzynski, L., Staszewska-Barczak, J. and Herbaczynaska-Cedro, K.:
Cardiac rhythm disturbances and the release of catecholamines
after acute coronary occlusion in dogs. Cardiovasc. Res. 3;
190-197, 1969.

Ciofalo, F., Levitt, B. and Roberts, J.: Some factors affecting
ouabain-induced arrhythmias in the reserpine-treated cat. Brit.
J. Pharmacol. t9: 143-154, 1967.

Clausen, T. and Hansen, 0.: Active Na-K transport and the rate of
ouabain binding. The effect of insulin and other stimuli on
skeletal muscle and adipocytes. J. Physiol. 219; 415-430, 1977.

Cooper, Y., Gilbert, J.W., Bloodwell, R.D. and Crout, R.J.: Chronic
extrinsic cardiac denervation by regional neural ablation. Circ.
Res._9: 275-281, 1961.

Corr, P.B., Pearle, D.L., Hinton, J.R., Roberts, W.C. and Gillis,
R.A.: Site of myocardial infarction. A determinant of the
cardiovascular changes induced in the cat by coronary occlusion.
Circ. Res. g2; 840-847, 1976.

Costantin, L.: Extracardiac factors contributing to hypotension
during coronary occlusion. Am. J. Cardiol. 11; 205-217, 1963.

Cranefield, P., Klein, H.0. and Hoffman, B.F.: Conduction of the
cardiac impulse: I. Delay, block, and one-way block in de-
pressed Purkinje fibers. Circ. Res. gg; 199-208, 1971.

Dengler, H.J., Spiegel, H.E. and Titus, E.0.: Uptake of tritium-
labelled norepinephrine in brain and other tissues of cat 1g
vitro. Science 1§§; 1072-1073, 1962.

Dhalla, N.S., Singh, J.N., Fedelesova, M., Balasubramanian, V. and
McNamara, D.B.: Biochemical basis of heart function. XII.
Sodium-potassium stimulated adenosine triphosphatase activity in
the perfused rat heart made to fail by substrate-lack. Cardio-
vasc. Res. 8; 227-236, 1974.

Dutta, 3., Goswami, S., Datta, D.K., Lindower, J.0. and Marks, B.H.:
The uptake and binding of six radiolabeled cardiac glycosides by
guinea-pig hearts and by isolated sarcoplasmic reticulum. J.
Pharmacol. Exp. Ther. 1§§; 10-21, 1968.

El-Sherif, N., Scherlag, B.J., Lazzara, R. and Hope, R.R.: Reentrant
ventricular arrhythmias in the late myocardial infarction period.
II. Patterns of initiation and termination of reentry. Circu-
lation §§; 702-719, 1977.

113

Evans, D.B., Peschka, M.T., Lee, R.J. and Laffan, R.J.: Anti-arrhyth-
mic action of nadolol, a beta-adrenergic receptor blocking agent.
Eur. J. Pharmacol. §§; 17-27, 1976.

Evans, D.E. and Gillis, R.A.: Effect of ouabain and its interaction
with diphenylhydantoin on cardiac arrhythmias induced by hypo-
thalamic stimulation. J. Pharmacol. Exp. Ther. 19§; 577-586,
1975.

Farber, J.L., Chien, K.R. and Mittnacht, S.: The pathogenesis of
irreversible cell injury in ischemia. Am. J. Pathol. 192; 271-
281, 1981.

Felder, R.B. and Thames, M.D.: Interaction between cardiac receptors
and sinoaortic baroreceptors in the control of efferent cardiac
sympathetic nerve activity during myocardial ischemia in dogs.
Circ. Res. §§; 728-736, 1979.

Felder, R.B. and Thames, M.D.: The cardiocardiac sympathetic reflex
during coronary occlusion in anesthetized dogs. Circ. Res. §§;
685-692, 1981.

Ferrier, G.R.: Digitalis arrhythmias: Role of oscillatory after-
potentials. Progr. Cardiovasc. Dis. 12; 459-474, 1977.

Ferrier, G.R., Saunders, J.H. and Mendez, C.: Cellular mechanism for
the generation of ventricular arrhythmias by acetylstrophanthi-
din. C irc. Res. gg; 600-609, 1973.

Forssman, D., Hansson, G. and Jansen, C.G.: Adrenal function in
coronary thrombosis. Acta Medica Scandinavica 152; 441-449,
1952.

Friedman, J.P., Harris, C.N. and Goldman, R.H.: Digitalis toxicity
in normal and hypoxemic dogs: Correlation with myocardial Na+,K -
activated ATPase activity. Clin. Res. 29; 205, 1972.

Gillis, R.A. and Quest, J.A.: The role of the nervous system in the
cardiovascular effects of digitalis. Pharmacol. Rev. g1; 19-97,
1979.

Gillis, R.A., Clancy, M.M. and Anderson, R.J.: Deleterious effects
of bretylium in cats with digitalis-induced ventricular tachy-
cardia. Circulation 32; 974-983, 1973.

Gillis, R.A., Jolson, H., Thibodeaux, H. and Levitt, B.: Antagonism
of deslanoside-induced cardiotoxicity by combined nicotinic and
muscarinic blockade of autonomic ganglia. J. Pharmacol. Exp.
Ther. lg§; 126-132, 1975.

Gillis, R.A., Raines, A., Sohn, Y.J., Levitt, B. and Standaert, F.G.:
Neuroexcitatory effects of digitalis and their role in the de-
velopment of cardiac arrhythmias. J. Pharmacol. Exp. Ther. l§§;
154-168, 1972.

 

114

Ginsburg, R., Bristow, M.R., Kantrowitz, N., Baim, D.S. and Harrison,
D.C.: Histamine provocation of clinical coronary artery spasm:
Implications concerning pathogenesis of variant angina pectoris.
Am. Heart J. 192; 819-822, 1981.

Glynn, I.M.: The action of cardiac glycosides on sodium and potas-
sium movements in human red cells. J. Physiol. London 126; 148-
173, 1957.

Glynn, I.M.: The action of cardiac glycosides on ion movements.
Pharmacol. Rev. 16; 381-407, 1964.

Goldman, R.H., Coltart, D.J. and Friedman, J.P.: The inotropic
effects of digitalis in hyperkalemia: Relation to Na,K-ATPase.
Circulation 12; 830-839, 1973.

Han, J. and Moe, G.K.: Nonuniform recovery of excitability in ven-
tricular muscle. Circ. Res. 11; 44-60, 1964.

Hansen, 0., Jensen, J. and Norby, J.G.: + Mutual exclusion of ATP, ADP
and g-strophanthin binding to Na+,K -ATPase. Nature 221; 122-
123, 1971.

Hargreave, F.E.: Digitalis and cor pulmonale. Brit. Med. J. 2; 943,
1965.

Harken, A.H., Barlow, C.H., Harden, H.R. and Chance, B.: Two and
three dimensional display of myocardial ischemic "border zone" in
dogs. Am. J. Cardiol. 12; 954-959, 1978.

Harris, A.S.z Potassium and experimental coronary artery occlusion.
Am. Heart J. 21; 797-802, 1966.

Harris, A.S., Bisteni, A., Russell, R.A., Brigham, J.C. and Firestone,
J.E.: Excitatory factors in ventricular tachycardia resulting
from myocardial ischemia: Potassium a major excitant. Science
199; 200-203, 1954.

Harrison, D.C., Robinson, M.D. and Kleiger, R.E.: Role of hypoxia in
digitalis toxicity. Am. J. Med. Sci. 256; 352-359, 1968.

Harvey, S.C.: The effects of ouabain and phenytoin on myocardial
noradrenaline. Arch. Int., Pharmacodyn. Ther. 212; 222-234,
1975.

Hayashi, K.D., Moss, A.J. and Yu, P.N.: Urinary catecholamine secre-
tion in myocardial infarction. Circulation 19: 473~481, 1969.

Hearse, D.J., Opie, L.H., Katzeff, I.E., Lubbe, w.F., Vanderwerff,
T.J., Peisach, M. and Boulle, 6.: Characterization of the

border zone in acute regional ischemia in the dog. Am. J.
Cardiol. 19; 716-726, 1977.

115

Herrick, J.B.: Clinical features of sudden obstruction of the coro-
nary arteries. J. Am. Med. Assoc. 99; 2015-2020, 1912.

Hill, J.L. and Gettes, L.S.: Effect of acute coronary artery occlu-
sion on local myocardial extracellular K+ activity in swine.
Circulation 91; 768—778, 1980.

Hill, J.L., Gettes, L.S., Lynch, M.R. and Hebert, N.C.: Flexible
valinomycin electrodes for on-line determination of intravascular
and myocardial K+. Am. J. Physiol. g§_5_: H455-H459, 1978.

Hirche, H.J., Franz, C., Bos, L., Bissig, R., Lang, R. and Schramm,
M.: Myocardial extracellular K+ and H increase and noradren-
aline release as possible cause of early arrhythmias following
acute coronary artery occlusion in pigs. J. M01. Cell. Cardiol.
12; 579-593, 1980.

Hoffman, B.F.: The genesis of cardiac arrhythmia. Prog. Cardiovasc.
Dis. 9; 319-329, 1966.

Hood, W.B. Jr., Covelli, V.H. and Norman, J.C.: Acute coronary
occlusion in pigs: Effects of acetylstrophanthidin. Cardiovasc.
Res. 9; 441-446, 1969.

Hood, W.B. Jr., McCarthy, B. and Lown, B.: Myocardial effects of
isoproterenol and acetylstrophanthidin. Circ. Res. 21; 195-199,
1967.

Horackova, M. and Vassort, G.: Excitation-contraction coupling in
frog heart: Effect of veratrine. Pfluegers Arch. Eur. J.
Physiol. 992; 291302, 1974.

Hougen, T.J. and Smith, T.W.: Inhibition of myocardial monovalent
cation active transport by subtoxic doses of ouabain in the dog.
Circ. Res. 92; 856-863, 1978.

Howard, L.L., Brunschwig, J.P. and Pressman, B.C.: Affect of the
ionophore monensin on cultured beating heart cells. Pharma-
cologist 19; 122, 1976.

Ismail-Beigi, F. and Edelman, 1.5.: The mechanism of the calorigenic
action of thyroid hormone. Stimulation of Na+,K -activated
adenosine triphosphatase activity. J. Gen. Physiol. 91; 710-
722, 1971.

Janse, M.J., Cinca, J. and Morena, H.: The border zone in myocardial
ischemia. An electrophysiological, metabolic and histochemical
correlation in the pig heart. Circ. Res. 99; 576-588, 1979.

Jennings, R.B., Reimer, K.A., Hill, M.L. and Mayer, S.E.: Total
ischemia in dog hearts in vitro. 1. Comparison of high energy
phosphate production, utilization and depletion, and of adenosine
nucleotide catabolism in total ischemia 13_vitro vs. severe
ischemia jg_vivo. Circ. Res. 99; 892-900, 198l.

116

Kane, J.J., Murphy, M.L., Bissett, J.K., De Soyza, N., Doherty, J.E.
and Straub, K.D.: Mitochondrial function, oxygen extraction,
epicardial S-T segment changes in tritiated digoxin distribution
after reperfusion of ischemic myocardium. Am. J. Cardiol. 99;
218-224, 1975.

Karliner, J.S. and Braunwald, E.: Present status of digitalis treat-
ment of acute myocardial infarction. Circulation 99; 891-902,
1972.

Karlsberg, R.P., Cryer, P.E. and Roberts, R.: Serial plasma cate~
cholamine responses early in the course of clinical acute myo-
cardial infarction: Relationship to infarct extent and mortali-
ty. Am. Heart J. 192; 24-29, 1981.

Kass, R.S., Tsien, R.w. and Heingart, M.: Ionic basis of transient
inward current induced by strophanthidln in cardiac Purkinje
fibers. J. Physiol. London 291; 209-226, 1978.

Kelliher, G.J. and Roberts, J.: A study of the anti-arrhythmic
action of certain beta-blocking agents. Am. Heart J. 91; 458-
467, 1974.

Kedzi, P., Kordenat, R. and Misra, S.: Reflex inhibitory effects of
vagal afferents in experimental myocardial infarction. Am. J.
Cardiol. 99; 853-860, 1974.

Kleber, A.G., Janse, M.J., Capelle, F.J.L. and Durrer, D.: Mechanism
and time course of S-T and T-Q segment changes during acute
regional myocardial ischemia in the pig heart determined by
extracellular and intracellular recordings. Circ. Res. 92; 603-
613, 1978.

Klein, R.F., Troyer, N.G. and Thompson, H.K.: Catecholamine excre-
tion in myocardial infarction. Arch. Intern. Med. 122; 476-482,
1968.

Korhonen, U.R., Jounela, A.J., Pakarinen, A.J., Pentikainen, P.J. and
Takkunen, J.T.: Pharmacokinetics of digoxin in patients with
acute myocardial infarction. Am. J. Cardiol. 99; 1190-1194,
1979.

Korth, C., Marx, H. and Weinberg, 5.: Uber die Mirkung des Strophan-
thins auf des zentralnervensystem. Naunyn-Schmiedeberg's Arch.
Pharmakol. Exp. Pathol. 199; 42-56, 1937.

Kraemer, R.J., Phernetton, T.M. and Folts, J.D.: Prostaglandin-like
substances in coronary venous blood following myocardial ische-
mia. J. Pharmacol. Exp. Ther. 199; 611-619, 1976.

Kreuger, E. and Unna, K.: Comparative studies on the toxic effects
of digitoxin and ouabain in cats. J. Pharmacol. Exp. Ther. 29;
282-294, 1942.

117

Ku, D. and Lucchesi, B.R.: Ischemic—induced alterations in cardiac
sensitivity to digitalis. Europ. J. Pharmacol. 91; 135-147,
1979.

Ku, D., Akera, T., Frank, M., Brody, T.M. and Iwasa, J.: The effects
of grayanotoxin I and a-dihydrograyanotoxin II on guinea-pig
myocardium. J. Pharmacol. Exp. Ther. 299: 363-372, 1977.

Ku, D., Akera, T., Pew, C.L. and Brody, T.M.: Cardiac glycosides:
Correlations among Na+,K+—ATPase, sodium pump and contractility
in the guinea-pig heart. Naunyn-Schmiedeberg's Arch. Pharmacol.
299; 185-200, 1974.

Kumar, R., Hood, W.B. Jr., Joison, J., Gilmour, D.P., Norman, J.C. and
Abelmann, W.H.z Experimental myocardial infarction. VI.
Efficacy and toxicity of digitalis in acute and healing phase in
intact conscious dogs. J. Clin. Invest. 99: 358—364, 1970.

Lammerant, J., De Herdt, P. and De Schryver, C.: Direct release of
myocardial catecholamine into the left heart chambers: The
enhancing effect of acute coronary occlusion. Arch. Int. Phar—
macodyn. Ther. 199: 219-226, 1966.

Langer, G.A.: Effects of digitalis on myocardial ionic exchange.
Circulation 99; 180-187, 1972.

Langer, G.A. and Serena, S.D.: Effects of strophanthidin upon con-
traction and ionic exchange in rabbit ventricular myocardium:
Relation to control of active state. J. M01. Cell. Cardiol. 65-
90, 1970.

Lathers, C.M., Kelliher, G.J., Roberts, J. and Beasley, A.B.: Non-
uniform cardiac sympathetic nerve discharge. Mechanism for
coronary occlusion and digitalis-induced arrhythmia. Circulation
91; 1058—1065, 1978.

Lechat, P. and Schmitt, H.: Interactions between the autonomic
nervous system and the cardiovascular effects of ouabain in
guinea—pig. Europ. J. Pharmacol. 19: 21-32, 1982.

Lederer, w.J. and Tsien, R.H.: Transient inward current underlying
arrhythmogenic effects of cardiotonic steroids in Purkinje
fibers. J. Physiol. London 299; 73-100, 1976.

Lee, K.S. and Klaus, M.: The subcellular basis for the mechanism of
inotropic action of cardiac glycosides. Pharmacol. Rev. 29; 193-
261, 1971.

Levi, R.: Effects of exogenous and immunologically released hista-
mine on the isolated heart: A quantitative comparison. J.
Pharmacol. Exp. Ther. 192; 227-238, 1972.

118

Levi, R., Allan, G. and Zavecz, J.H.: Cardiac histamine receptors.
Fed. Proc. 99; 1942-1947, 1976.

Levi, R. and Burke, J.A.: Cardiac anaphylaxis: SRS-A potentiates
and extends the effects of released histamine. Europ. J. Phar-
macol. 92; 41-49, 1980.

Levi, R. and Capurro, N.: Cardiac histamine-ouabain interaction:
Potentiation by ouabain of the arrhythmogenic effects of hista-
mine. J. Pharmacol. Exp. Ther._192: 113-119, 1974.

Levitt, B., Cagin, N.A., Somberg, J., Bounous, H., Mittag, T. and
Raines, A.: Alteration of the effects and distribution of
ouabain by spinal cord transection in the cat. J. Pharmacol.
Exp. Ther. 199; 24—28, 1973.

Levitt, B., Gillis, R.A., Roberts, J. and Raines, A.: Influence of
the cardiac vagus nerves on the cardiotoxicity of acetylstrophan-
thidin (AcS), ouabain (0), and digitoxin (D). Pharmacologist_12:
304, 1970.

Lewis, P.J. and Haeusler, G.: Reduction of sympathetic nervous
system activity as a mechanism for the hypotensive effect of
propranolol. Nature New Biol. 299; 440-441, 1975.

Lindmar, R. and Loffelholz, K.: The neuronal efflux of noradren-
aline: Dependency on sodium and facilitation by ouabain.
Naunyn-Schmiedeberg's Arch. Pharmakol. Exp. Pathol. 299; 93-100,
1974.

Lo Sasso, A.M. and Paradise, R.R.: Influence of the vagus on develop-
ment of ventricular arrhythmias induced by acetylstrophanthidin
in dogs anesthetized with pentobarbital or halothane. Anesth.
Analg. Curr. Res. 99; 317-327, 1969.

Lowry, 0.H., Rosebrough, M.J., Farr, A.L. and Randall, R.J.: Protein
measurements with the Folin phenol reagent. J. Biol. Chem. 199;
265-275, 1951.

Lubbe, N.F., Peisach, M., Pretorius, R., Bryuneel, K.J.J. and Opie,
L.H.: Distribution of myocardial blood flow before and after
coronary artery ligation in the baboon. Relation to early
ventricular fibrillation. Cardiovasc. Res. 9; 478-487, 1974.

Lullmann, H., Timmermans, P.B.M.W.M. and Ziegler, A.: Accumulation
of drugs by resting or beating cardiac tissue. Europ. J. Phar-
macol. 99; 277-285, 1979.

Lum, B.K.B., Yano, 3.5. and Yatsushiro, G.N.: Cardiotoxic interac-
tions between sympathomimetic amines and ouabain. Proc. West
Pharmacol. Soc._29: 275-280, 1977.

 

119,

Malliani, A., Schwartz, P.J. and Zanchetti, A.: A sympathetic reflex
elicited by experimental coronary occlusion. Am. J. Physiol.
211; 703-709, 1969.

Malsky, P.M., Vokonas, F.S., Paul, S.J., Robbins, S.L. and Hood, N.S.z
Autoradiographic measurement of regional blood flow in normal and
ischemic myocardium. Am. J. Physiol. 292; 576-583, 1977.

Mason, D.T., Zelis, R., Lee, G., Hughes, J.L., Spann, J.F. Jr. and
Amsterdam, E.A.: Current concepts and treatment of digitalis
toxicity. Am. J. Cardiol. 21; 546-559, 1971.

Matsui, H. and Schwartz$ A.: Mechanism of cardiac glycoside inhibi-
tion of the (Na+,K )-dependent ATPase from cardiac tissue.
Biochem. Biophys. Acta 191; 655-663, 1968.

McDonald, T.F. and MacLeod, D.P.: Maintenance of resting potential
in anoxic guinea-pig ventricular muscle: Electrogenic sodium
pumping. Science 112; 570-572, 1971.

McDonald, T.F. and MacLeod, D.P.: Anoxic atrial and ventricular
muscle electrical activity, cell potassium and metabolism. A
comparative study. J. Mol. Cell. Cardiol. 9; 149-159, 1973.

McDonald, T.F. and MacLeod, D.P.: Metabolism and the electrical
activity of anoxic ventricular muscle. J. Physiol. London 229;
559-582, 1973.

McLain, P.L.: Effects of cardiac glycosides on spontaneous efferent
activity in vagus and sympathetic nerves of cats. Int. J.
Neuropharmacol. 9; 379-387, 1969.

McLain, P.L., Kruse, T.K. and Redick, T.F.: The effect of atropine
on digitoxin bradycardia in cats. J. Pharmacol. Exp. Ther. 129;
76-81, 1958.

Mendez, C., Aceves, J. and Mendez, R.: Inhibition of adrenergic
cardiac acceleration by cardiac glycosides. J. Pharmacol. Exp.
Ther. 191; 191-198, 1961.

Morris, J.J., Taft, C.V., Whalen, R.E. and McIntosh, H.D.: Digitalis
and experimental myocardial infarction. Am. Heart J. 11; 342-
355, 1969.

Moss, A.J., Davis, H.T., Conard, D.L., De Camilla, J.J. and Odoroff,
C.L.: Digitalis-associated cardiac mortality after myocardial
infarction. Circulation 99; 1150-1155, 1981.

Nadeau, R. and De Champlain, J.: Comparative effects of 6-hydroxy-
dopamine and reserpine on ouabain toxicity. Life Sci. 19; 1753-
1761, 1973.

120

Nayler, W.G.: The role of calcium in the ischemic myocardium. Am.
J. Pathol. 192; 262-270.

Nayler, W.G., Poole-Wilson, P.A. and Williams, A.: Hypoxia and
calcium. J. Mol. Cell. Cardiol. 11; 683-706, 1979.

Oberg, B. and Thoren, P.: Circulatory responses to stimulation of
medullated and non—medullated afferents in the cardiac nerve in
the cat. Acta Physiol. Scand. 91; 121-132, 1973.

Ogletree, M.L., Flynn, J.T., Feola, M. and Lefer, A.M.: Early
prostaglandin release from the ischemic myocardium in the dog.
Surgery, Gynecol. Obstet. 199; 743-740, 1977.

Pace, D.G. and Gillis, R.A.: Neuroexcitatory effects of digoxin in
the cat. J. Pharmacol. Exp. Ther. 199; 583-600, 1976.

Papp, J.G. and Vaughan Williams, E.M.: The effect of bretylium on
intracellular cardiac action potentials in relation to its anti-
arrhythmic and local anesthetic activity. Brit. J. Pharmacol.
91; 380—390, 1969.

Pappano, A.J.: Calcium-dependent action potentials produced by
catecholamines in guinea pig atrial muscle fibers depolarized by
potassium. Circ. Res. 21; 379-390, 1970.

Penna, M., Linares, F. and Caceres, L.: Mechanism for cardiac stimu-
lation during hypoxia. Am. J. Physiol. 299; 1237-1242, 1965.

Powell, J.R. and Brody, M.J.: Identification and specific blockage
of two receptors for histamine in the cardiovascular system. J.
Pharmacol. Exp. Ther. 199: 1-14, 1976.

Puri, P.S.: Modification of experimental infarct size by cardiac
drugs. Am. J. Cardiol. 99; 521-528, 1974.

Raab, W. and Gigee, W.: Total urinary catechol excretion in cardio-
vascular and other clinical conditions. Circulation 9; 592-599,
1954.

Raab, W., Van Lith, P., Lepeschkin, E. and Herrlich, H.C.: Catechol-
amine-induced myocardial hypoxia in the presence of impaired
coronary dilatability independent of external cardiac work. Am.
J. Cardiol. 9; 455-470, 1962.

Raines, A., Levitt, B. and Standaert, F.G.: The effect of spinal
section on ventricular rhythm disorders induced by ouabain.
Arch. Int. Pharmacodyn. Ther. 119; 485-490, 1967.

Raines, A., Moros, D. and Levitt, B.: The effect of guanethidine on
ouabain-induced ventricular arrhythmia in the cat. Arch. Int.
Pharmacodyn. Ther. 119; 373-377, 1968.

 

121

Raper, C. and Wale, J.: Cardiac arrhythmias produced by interaction
of ouabain and beta-receptor stimulation. Eur. J. Pharmacol. 9;
223-234, 1969.

Rau, E.E., Shine, K.I. and Langer, G.A.: Potassium exchange and
mechanical performance in anoxic mammalian myocardium. Am. J.
Physiol. 292; H85-H94, 1977.

Recordati, G., Schwartz, P.J., Pagani, M., Malliani, A. and Brown,
A.M.: Activation of actiac vagal receptors during myocardial
ischemia. Experientia 21; 1423-1424, 1971.

Reimer, K.A., Lowe, J.E., Rasmussen, M.M. and Jennings, R.B.: Wave-
front phenomenon of ischemic cell death. 1. Myocardial infarct
size vs. duration of coronary occlusion in dogs. Circulation 99;
786-798, 1977.

Repke, K.R.H.: Uber den biochemischen wirkungsmodus von digitalis.
Klin. Wochenschr. 92; 157-165, 1964.

Repke, K.R.H., Est, M. and Portius, H.J.: Uber die ursache der
spicies-untershiede in der digitalisempfindlichkeit. Biochem.
Pharmacol. 19; 1785-1802, 1965.

Richardson, J.A.: Plasma catecholamine concentration in acute infarc-
tion. Coronary Heart Diseases (Likoff, W. and Moyer, J.H.,
eds.), Grune and Strutton, New York, pp. 273-277, 1963.

Robinson, G.C. and Wilson, F.N.: A quantitative study of the effect
of digitalis on the heart of the cat. J. Pharmacol. Exp. Ther.
19; 491-507, 1918.

Rocha E Silva, M.: Action of histamine upon the circulatory appara-
tus. In Handbook of Experimental Pharmacology (Rocha E Silva,
M., edTY, 18: part 1, Histamine. Springer-Verlag, New York,
1966.

Rosen, M.R., Wit, A.L. and Hoffmen, B.F.: Electrophysiology and
pharmacology of cardiac arrhythmias. IV. Cardiac antiarrhythmic
and toxic effects of digitalis. Am. Heart J. 99; 391-399, 1975.

Saito, H., Otani, T., Shudo, I. and Tanabe, T.: Effect of 6-hydroxy-
dopamine on cardiotoxicity of ouabain in guinea pigs. Jap. J.
Pharmacol. 29; 923-925, 1974.

Sayen, J.J., Sheldon, W.F. and Horwitz, 0.: Studies of coronary
disease in the experimental animal. II. Polarographic determi-
nations of local oxygen availability in the dogs left ventricle
during coronary occlusion and pure oxygen breathing. J. Clin.
Invest. 99; 932-940, 1952.

 

122

Schatzmann, H.J.: Herzyglykoside als hemmstoffe fur den aktiven
kalium- und natriumtransport durch die erythrocytenmembran.
Helv. Physiol. Pharmacol. Acta 11; 346-354, 1953.

Scherlag, B.J., El-Sherif, N., Hope, R.R. and Lazzara, R.: Charac-
terization and localization of ventricular arrhythmias resulting
from myocardial ischemia and infarction. Circ. Res. 99; 372—383,

1974.

Schwartz, A.: Is the cell membrane Na+,K+—ATPase enzyme system the
pharmacological receptor for digitalis? Circ. Res. 99; 2-7,
1976.

Schwartz, A., Lindenmayer, G.E. and Allen, J.C.: The sodium-potas-
sium adenosine triphosphatase: Pharmacological, physiological
and biochemical aspects. Pharmacol. Rev. 21; 3—134, 1975.

Schwartz, A., Wood, J.M., Allen, J.C., Bornet, E.P., Entman, M.L.,
Goldstein, M.A., Sordahl, L.A. and Suzuki, M.: Biochemical and
morphologic correlates of cardiac ischemia. Am. J. Cardiol. 92-
46-61, 1973

Selzer, A. The use of digitalis in acute myocardial infarction.
Prog. Cardiovasc. Dis. 10: 518— 528, 1965.

Shahab, L., Wollenberger, A., Haase, M. and Schiller, U.: Noradren-
1inabgabe aus dem hundeherzen nach vorubergehender okklusion
einer korronararterie. Acta Biol. Med. Germ. 22; 135—143, 1969.

Sharma, V. K. angH Banerjee, S.P.: The effect of 6-hydroxydopamine on
specific( H)ouabain binding to some sympathetically innervated
organs of the cat. Mol. Pharmacol. 19; 796-804, 1977.

Shen, A.C. and Jennings, R.B.: Myocardial calcium and magnesium in
acute ischemic injury. Am. J. Pathol. 91: 417-440, 1972.

Shine, K.I.: Ionic events in ischemia and anoxia. Am. J. Pathol.
192; 256-261, 1981.

Siggers, D.C., Salter, C. and Fluck, D.C.: Serial plasma adrenaline
and noradrenaline levels in myocardial infarction using a new
double isotope technique. Brit. Heart J. 99: 878-883, 1971.

Skou, J.C.: The influence of some cations on an adenosine triphos-
phatase frOm peripheral nerves. Biochem. Biophys. Acta 29: 394-
401, 1957.

Skou, J.C.: Enzymatic basis for active transport of Na+ and K+
across cell membranes. Physiol. Rev. 99; 596-617, 1965.

123

Solti, F., Iskum, M. and Nagy, J.: Studies on the acute cardiac
action of strophanthin in the dog by means of cardiac denerva-
tion. Acta Physiol. Acad. Sci. Hung. 29; 377-385, 1965.

Somberg, J.C., Bounous, H., Cagin, N. and Levitt, B.: Histamine
antagonists as antiarrhythmic agents in ouabain cardiotoxicity in
the cat. J. Pharmacol. Exp. Ther. 219: 375-380, 1980.

Somberg, J.C., Risler, T. and Smith, T.W.: Neural factors in digi-
talis toxicity: Protective effect of C-1 spinal cord transec—
tion. Am. J. Physiol. 299: H53l-H536, 1978.

Staszewska-Barczak, J.: The reflex stimulation of catecholamine
section during the acute stages of myocardial infarction in the
dog. Clin. Sci. 91; 419-439, 1971.

Staszewska-Barczak, J. and Ceremuzynski, L.: The continuous estima-
tion of catecholamine release in the early stages of myocardial
infarction in the dog. Clin. Sci. 99: 531-539, 1968.

Stickney, J.L.: Inhibition of 3H-l-norepinephrine uptake by ouabain
is species dependent. Res. Commun. Chem. Pathol. Pharmacol. 19;
227-236, 1976.

Thames, M.D.: Acetylstrophanthidin-induced reflex inhibition of
canine renal sympathetic nerve activity mediated by cardiac
receptors with vagal afferents. Circ. Res. 99; 8—15, 1979.

Thompson, A.J., Hargis, J., Murphy, M.L. and Doherty, J.E.: Tri—
tiated digoxin. XX. Tissue distribution in experimental myo-
cardial infarction. Am. Heart J. 99; 319-324, 1974.

Thoren, P.: Evidence for a depressor reflex elicited from left
ventricular receptors during occlusion of one coronary artery in
the cat. Acta Physiol. Scand. 99: 23-34, 1973.

Tobin, T. and Brody, T.M.: Rates of dissociation of enzyme—ouabain
complexes and K0.5 values in Na+,K+-ATPase from different
species. Biochem. Pharmacol. 21; 1553-1560, 1972.

Tomlinson, C.W. and Dhalla, N.S.: Excitation-contraction coupling in
heart. IX. Changes in the intracellular stores of calcium in
failing hearts due to lack of substrate and oxygen. Cardiovasc.
Res. 1; 470-476, 1973.

Travell, J., Gold, H. and Modell, W.: Effect of experimental cardiac
infarction of response to digitalis. Arch. Intern. Med. 91: 184-
197, 1938.

Tse, W.W. and Han, J.: Interaction of epinephrine and ouabain on
automaticity of canine Purkinje fibers. Circ. Res. 99; 777-782,
1974.

124

Tsien, R.W. and Carpenter, D.0.: Ionic mechanisms of pacemaker
activity in cardiac Purkinje fibers. Fed. Proc. 91; 2127-2131,
1978.

Uchida, Y. and Murao, 8.: Excitation of afferent cardiac sympathetic
nerve fibers during coronary occlusion. Am. J. Physiol. 229;
1094-1099, 1974.

Uchida, Y., Kamisaka, K., Murao, S. and Ueda, H.: Mechanosensitivity
of afferent cardiac sympathetic nerve fibers. Am. J. Physiol.
229; 1088-1093, 1974.

Valorie, C., Thomas, M. and Shillingford, J.P.: Urinary excretion of
free noradrenaline and adrenaline following acute myocardial
infarction. Lancet 1; 127-130, 1967.

Vassalle, M. and Scida, E.E.: The role of sodium in spontaneous
discharge in the absence and in the presence of strophanthidin.
Fed. Proc. 99; 880, 1979.

Vaughan Williams, E.M.: Prevention of arrhythmias due to cardiac
glycosides by block of sympathetic beta-receptors. Lancet 1;
420-421, 1963.

Vokonas, P.S., Malsky, P.M., Paul, S.J., Robbins, S.L. and Hook, W.B.:
Radioautographic studies in experimental myocardial infarction:
Profiles of ischemic blood flow and quantification of infarct
size in relation to magnitude of ischemic zone. Am. J. Cardiol.
92; 67-75, 1978.

Waldenstrom, A.P., Hjalmarson, A.C. and Thornell, L.T.: A possible
role of noradrenaline in the development of myocardial infarc-
tion: An experimental study in the isolated rat heart. Am.
Heart J._99: 43-51, 1978.

Weaver, L.C.: Cardiopulmonary sympathetic afferent influences on
renal nerve activity. Am. J. Physiol. 299; H592-H599, 1977.

Weaver, L.C., Danos, L.M., 0ehl, R.S. and Meckler, R.L.: Contrasting
reflex influences of cardiac afferent nerves during coronary
occlusion. Am. J. Physiol. 299; H620-H629, 1981.

Weiss, J. and Shine, K.I.: Extracellular potassium accumulation
during myocardial ischemia: Implications for arrhythmogenesis.
J. Mol. Cell. Cardiol. 19: 699-704, 1981.

Williams, J.F., Boyd, D.L. Jr. and Border, J.F.: Effect of acute
hypoxia and hypercapneic acidosis on the development of acetyl-
strophanthidin-induced arrhythmias. J. Clin. Invest. 91; 1885-
1894, 1968.

125

Wit, A.L. and Bigger, T.J.: Possible electrophysiological mechanisms
for lethal arrhythmias accompanying myocardial ischemia and
infarction. Circulation 51/52(Suppl. III): 96-115, 1975.

 

Wit, A.L., Hoffman, B.F. and Cranefield, P.F.: Slow conduction and
reentry in the ventricular conduction system. I. Return extra-
systole in canine Purkinje fibers. Circ. Res. 99: l-lO, 1972.

Withering, W.: An account of the foxglove, and some of its medical
uses: With practical remarks on dropsy, and other diseases.
Birmingham. 1785.

Wollenberger, A. and Sahab, L.: Anoxic—induced release of noradren-
aline from the isolated perfused heart. Nature 291: 88-89, 1965.

Yamamoto, S., Akera, T. and Brody, T.M.: Sodium influx rate and
ouabain-sensitive rubidium uptake in isolated guinea—pig atria.
Biochem. Biophys. Acta 999; 270—284, 1979.

Yazaki, Y. and Fujii, J.: Depressed Na-K-ATPase activity in the
failing rabbit heart. Jap. Heart J. 19; 73-83, 1972.

Zipes, D.P., Besch, H.R. and Watanabe, A.M.: Role of the slow
current in cardiac electrophysiology. Circulation 91; 761-766,
1975.

 

     

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

1|)1111111111111111111111111111111111