MSU

 

 

RETURNING MATERIALS:
P1ace in book drop to

 

 

LJBRARJES remove this checkout from
.—_ your record. FINE_S_ WIN
be charged if book is
returned after the date
stamped below.
‘32 2M ‘/
1:2?

 

 

THE RENAL ACTIONS OF ADENOSINE

BY

Lois Johanna Arend

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Physiology

1987

\49

ABSTRACT
THE RENAL ACTIONS OP ADENOSINE

BY

Lois Johanna Arend

Renal function is regulated by neural, humoral, and
physical factors. Intrinsic factors are also responsible
for the control of renal function. Many hypotheses have
been formulated for the intrinsic regulation of renal
function. One of these hypotheses involves adenosine,
acting as a retaliatory metabolite. That is, as renal
cellular metabolism is stimulated, and ATP hydrolysis is
increased, adenosine is formed and acts to return
metabolism to a normal state. Many previous studies have
demonstrated an effect of exogenous adenosine on renal
function consistent with the hypothesis that adenosine is
an intrinsic regulator of renal function.

The present studies were designed to test the effects
of endogenously-produced adenosine on renal function, and
to investigate the cellular mechanisms involved in some of
the renal actions of adenosine.

Increasing the intrarenal concentration of adenosine by

several pharmacological and pathophysiological manuevers

Lois Johanna Arend

led to pronounced changes in renal function. These effects
could be attenuated or inhibited by the competitive
adenosine receptor antagonist, theophylline. These studies
support the hypothesis that intrarenally-produced adenosine
can alter renal function.

To investigate the cellular mechanisms of the actions
of adenosine, cortical collecting tubule cells (RCCT) were
used as an epithelial model. The presence of adenosine
receptors in these cells and the second messenger systems
to which they are coupled were studied with the use of non-
transported, non-metabolized adenosine analogs. RCCT cells
were demonstrated to possess A1 and A2 adenosine receptors,
coupled to the production of cAMP, and a receptor of
unknown subtype coupled to the mobilization of
intracellular calcium. This latter receptor may be a
previously unidentified adenosine receptor. Alternatively,
because it has characteristics similar to the A1 receptor,
it may be a second signal transducing mechanism coupled to
the A1 receptor.

These studies support the hypothesis that adenosine is
involved in the regulation of renal function. This
conclusion is based on the ability of adenosine, produced
by the kidney, to alter renal function, and on the presence
of adenosine signal transducing mechanisms in one

population of renal cells.

To my grandmothers
Helen, Lois, and Marcella

ii

ACKNOWLEDGEMENTS

First and foremost, I would like to acknowledge the
love and understanding given to me by my parents. Without

their support this work would mean much less.

I would like to thank the members of my Guidance
Committee: Drs. Steven R. Heidemann, S. Richard Heisey,
William L. Smith, William S. Spielman, and Lynne C. Weaver,

for their advice and encouragement.

In particular, I thank Bill, my advisor, colleague,
and friend, for providing a challenging, feracious

environment in which to nurture my research skills.

iii

COPYRIGHT ACKNOWLEDGEMENTS

Sections III.A., III.B., and IV.A. include the text
and figures, in modified form, of the following
publications: L.J. Arend, C.I. Thompson, and W.S. Spielman,
Circ. Res. 56:242-251, 1985, American Heart Association;
L.J. Arend, C.I. Thompson, M.A. Brandt, and W.S. Spielman,
Kidney International 30:656-661, 1986, Springer-Verlag; and
L.J. Arend, W.K. Sonnenburg, W.L. Smith, and W.S. Spielman,
J. Clin. Invest. 79:710-714, 1987, Rockfeller University
Press.

This work was supported in part by a USPHS Predoctoral
Traineeship, NIH (HLO7404), and by a University Graduate
Fellowship awarded by the Michigan State University College

of Natural Science.

iv

TABLE OF CONTENTS

Page
List of Tables....... .................... .. ........... vii

List of Figures....................................... viii
I. IntrOductionOOOOOOOOO0.0.0000...00.0.00... 000000000 1

II. Literature Review.................... .............. 4
A. Adenosine Biochemistry.......................... 4
A.l. Adenosine metabolism..................... 4

A.2. Adenosine receptors...................... 6

8. Effect of Adenosine on Renal Function........... 10
8.1. Hemodynamics............................. 10

8.2. Renin release............................ 12

8.3. Neurotransmitter release................. 15

8.4. Sodium and water excretion............... 16

8.5. Endogenous adenosine..................... 18

C. Cellular Mediation of the Actions of Adenosine.. 21
C.1. Non-epithelial cells..................... 21

C.2. Epithelial cells......................... 24

III. The Renal Effects of Increasing Endogenous
Adenosine.......................................... 28
A. Dipyridamole decreases glomerular filtration

in the sodium-depleted dog...................... 28
A.1. Introduction............................. 28
A.2. Methods.................................. 30
A.3. Results.................................. 41
A.4. Discussion............................... 55
8. Increasing intrarenal adenosine by maleic
acid decreases GFR and renin release............ 65
8.1. Introduction............... ..... ......... 65
8.2. Methods.................................. 67
8.3. Results................ ....... . ...... .... 71
8.4. Discussion......... ........ .............. 78
C. A role for intrarenal adenosine in the renal
hemodynamic response to contrast media.......... 88
C.1. Introduction...................... ...... . 88
C.2. Methods..................... ..... ........ 9O

TABLE OF CONTENTS. CONT'D.

2892

C03. ReSUItSOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO. 93
CO4. DiSCUSSiOnOOOOOOO0.000000000000000000000 99

IV. The Renal Cellular Actions of Adenosine........... 106
A. A1 and A2 adenosine receptors in rabbit
cortical collecting tubule cells............... 106
A01. IntrwuctionOO0....OOOOOOOOIOOOOOIOOOOOO 106
A02. HethmsoOOOOIOIOOOOOOOOOOOOOOOI0.0.0.... 107
A03. ReSUJ-tBOOOO.00......OOOOOOCOOOOOOOOOOOOO 111
A04. Discussion.OOOOOOOOOOOOOOOOOOOO0.0.0.... 117
8. Adenosine signal transduction in the rabbit
cortical collecting tubule: receptor-mediated
calcium mobilization........................... 124
3.1. InttOductionOOOOOOOCOO-OIOOOOOOOOOOOI... 124
3.2. HethMSOOOOO0......OOOOOOOOOOCOOOOOOOOO. 126
8.3. ResaltscOOOOOOOOOOOOOOOOOOOOO0.000...... 131
30‘. Discussion.O00......OOOOOOOOOOOIOOOOOOOO 143

V. summary and ConCIUSionsO...OOOOOOOOIOOOOOOOOOOOOOO 148

Appendix A. Abbreviations and Definitions of
Compounds UsedOOOOOOIOOOOOIOOOOOOIIOOOOOO 154

Appendix 8. Derivation of Equation for Calculation
of Intracellular Calcium................. 156

Literature CitedOOOOOOOO0.0IOIOOOOOOOOOOOOO0.0.0.0.... 158

vi

LIST OF TABLES

Table Page
1. Classification of adenine receptors .............. 8
2. Effect of dipyridamole on renal function in

sodium-depleted and sodium-loaded dogs and
the effect of theophylline on the renal responses
to dipyridamole ....... ........... ........... ..... 47

Time course of effect of maleic acid in
sodium—depleted and sodium-loaded dogs............ 72

Effect of theophylline on the maleic acid-
induced changes in renal function of sodium—
depleted and SOdiUM'loaded dOgs. o o o o o o o o o o o o o o o o a o 73

Effects of contrast media on renal function, and
the effects of theophylline and dipyridamole on
the renal responses to contrast media ..... ... ..... 98

Effect of adenosine analogs on hormone-stimulated
cAMP production in RCCT cells.. ........ ........... 116

Effect of pertussis toxin on stimulation of cAMP
production by N‘-cyclohexyladenosine (CHA) in
RCCT 06118000.......0.0........OOOOOOOOOOOOOOOOOOO 119

Effect of 0-101 on stimulation and inhibition of

cAMP production by adenosine analogs in RCCT
cells......OOOOOOOOOO......OOOOOOOOOOOOOOO0.0.00.0 14o

vii

LIST OF FIGURES

figure

2.

3.

9.

Dilution curves for adenosine and ara-H in
plasma from a representative single-injection,
mu1t1p1e-tracer experiment.......................

Dilution curves for adenosine and ara-E in the
urinary effluentoo0.0.0.0...OOOOOOOIOOOOIIOOOOOOO

Comparison of glomerular filtration rate (GFR)
before, during, and after the intrarenal infusion
of dipyridamole in sodium-depleted dogs..........

Renal perfusion pressure (RPP). renal blood flow
(RBF), and glomerular filtration rate (GER) in
five sodium-depleted dogs before, during, and
after the intrarenal infusion of dipyridamole....

Mean arterial pressure (MAP), renal blood flow
(REF), and glomerular filtration rate (GFR), in
five sodium-depleted dogs before and during the
intrarenal infusion of dipyridamole, and during
the intrarenal infusion of theOphylline and
dipyridamole, simultaneously.....................

Effect of theophylline on the maleic acid—induced
changes in glomerular filtration rate (GER) and
renal blood flow (REP) in five sodium—depleted

dOQSOOOOOOOOOOOOOOOOOOO0.0.000...OOOOOOOOOOOOOOIO

Effect of theOphylline on the maleic acid-induced
changes in glomerular filtration rate (GER) and
renal blood flow (R8?) in five sodium-loaded

dOQBOOOOOOOOOOOOOOOOOOO00.0.0000...OOOOOOOIOOOOOO

Effect of theophylline on the maleic acid-induced
decrease in renin release presented as percent of

contIOIOOOOOOOOOOOOOOOOOOOOOOOOOOOIOOOOOOOOOOOOO.

Effect of maleic acid on arterial adenosine,
renal venous adenosine, and the urinary excretion
of adenosine in six sodium-depleted and ten
SOdium-loaded dogsooooooooooooooooooeoooeooaoocoo

viii

43

44

46

51

53

74

75

79

80

10.

11.

12.

13.

14A.

14B.

15.

16.

17.

18.

19.

20.

LIST OF FIGURES, CONT'D.

Representative renal blood flow tracing of

the responses to contrast media injection

(small arrows) in the control state, and in

the presence of dipyridamole....................

Effect of contrast media on glomerular
filtration rate (closed bars) and renal blood
flow (hatched bars) in the control state,

(CON), and during the infusion of theophylline
(T380) 0! dipyridamOIE (DP)oooooooeeoooooooooooo

Effect of adenosine analogs on total RCCT cell
cAMP production, expressed as percent of control

(basal)0.0.0....00......OOOOCOOOOOOOOOOOIOOOOOOO

Effect of pretreatment with pertussis toxin on
the modulation by CHA (50 nM) of 1 uM AVP-
stimulated cAMP production in RCCT cells........

Effect of adenosine analogs on intracellular
free calcium in primary RCCT cells, expressed
as percent change from control..................

Effect of adenosine analogs on intracellular
free calcium in first-passage RCCT cells,
expressed as percent change from control........

Representative fluorescence tracing of response
of intracellular calcium to adenosine analogs...

Effect of DPx on the NECA-induced increase in
intracellular calcium...........................

Effect of C-lOl on the CHA-induced increase in
intracellular calcium...........................

The response of intracellular calcium to NECA
in the presence and absence of extracellular

GaloiumOOOOOOOO0.0.0.0...OOOOOOOOOOOOOOOOOOOOOOO

Effect of pertussis toxin on the calcium
responses to the adenosine analogs, CHA and

NECAOO00......0.00......OOOOOOOOOOOOOOOOIOOO0.0.

Model cell of adenosine receptor systems........

ix

95

96

113

118

133

134
135
137

138

140

142
150

I . INTRODUCTION

Adenosine is a ubiquitous compound that has a variety
of actions in many tissues and is thought to be involved in
metabolic control of organ function. That is, adenosine
production results from the breakdown of ATP during periods
of increased cellular work and is thought to act in a
negative feedback manner to reduce energy consumption.
Specifically, in the kidney, it has been hypothesized that
adenosine acts as an intrinsic regulator of renal function
(Osswald 1980: Spielman and Thompson 1982). Because renal
energy utilization is almost entirely a result of active
sodium transport by the tubular epithelium and because the
rate of tubular transport of sodium is determined by the
rate of sodium filtered, a phenomenon known as
glomerulotubular balance, it was postulated that increased
ATP hydrolysis, secondary to stimulated transport, elevated
adenosine levels that decrease GFR, filtered load, and
thereby ATP hydrolysis toward normal.

In determining a role for adenosine in the metabolic
control of renal function it is necessary to determine if
the exogenous administration of adenosine produces the

proposed changes in renal function. Several previous

studies to characterize the effects of exogenous adenosine
support the hypothesis that adenosine may act as an
intrinsic regulator of renal function (reviewed in Spielman
and Thompson 1984, Osswald 1984). These studies are
discussed in the second section of the literature review
(II.8.l-4). It must also be proven that adenosine can be
produced in the kidney in sufficient amounts to elicit the
necessary changes in renal function, that there is a system
for the regulation of adenosine levels by degradation or
cellular removal, and that renal cells possess adenosine
receptors coupled to signal transduction mechanisms.

This dissertation extends what is known about the
functional effects of intrarenal adenosine by describing
three studies in which endogenous adenosine concentrations
were increased and the changes in renal function observed.
This dissertation further characterizes the renal actions
of adenosine by detailing two studies directed at
understanding the mechanism by which adenosine acts to
control renal cellular function. These studies identify the
presence of adenosine receptors in rabbit cortical
collecting tubule cells and the effects of adenosine on two
second messenger systems in those cells.

The literature review is composed of three sections,

the first being a general review of the adenosine molecule,

its biochemistry, and the present state of understanding of
the adenosine receptor systems. The other two sections
correspond generally to the two diverse areas of study.
Each of the five projects comprising this study is
represented by a chapter, consisting of a separate
introduction, methods, results, and discussion with
specific points relevant to each project. A general
discussion of the importance of adenosine in the regulation

of renal function completes the thesis.

II. LITERATURE REVIEW
A. Adenosine Biochemistry

A.1. Adenosine metabolism

Adenosine, a purine nucleoside that is found in all
cells, is an intermediate in the pathway of adenine
nucleotide synthesis and degradation. Adenosine consists
of a ribose moiety and an adenine purine ring, and is a
precursor to, and metabolite of, ATP and the other adenine
nucleotides. Adenosine is also a component of S-adenosyl
homocysteine (SAH).

Adenosine is produced by the action of 5' nucleotidase
to hydrolyze adenine nucleotides (Baer et al 1966) and by
the hydrolysis of SAH by SAH hydrolase (Schrader, Schutz,
and Bardenheuer 1981). Both of these enzymes are present
in the kidney (Eloranta 1977; Miller 1978; Schatz 1977).
Despite the favorable equilibrium of the hydrolase reaction
toward degradation of SAH and production of adenosine, SAH
is not likely to be a major source of adenosine since the
SAH hydrolase is inhibited by its products, adenosine and
inosine. The major source of adenosine is therefore

presumed to be the adenine nucleotides.

The extracellular concentration of adenosine depends
upon the balance between its rate of production and its
rate of removal. The removal of adenosine from the
extracellular space is primarily accomplished by cellular
uptake followed by deamination to inosine by adenosine
deaminase or phosphorylation to AMP by adenosine kinase.
Adenosine deaminase is also present in the interstitium
(Conway and Cooke 1939; Trams and Lauter 1974; Baer and
Drummond 1968). Movement of adenosine into and out of the
cell is governed by bi-directional facilitated diffusion.
The adenosine carrier has been extensively characterized in
red blood cells (Kolassa and Pfleger 1975; Kubler and
Bretschneider 1963), hepatoma cells (Plagemann 1971;
Plagemann and Wohlhueter 1980; Plagemann and Richey 1974),
aortic endothelial and smooth muscle cells (Pearson et al
.1978), and platelets (Sixma et al 1976). It has been
suggested that the nucleoside carrier is closely linked to
the membrane-bound 5'-nuc1eotidase, thereby producing and
transporting adenosine in one step (Fox and Kelley 1978).

Relatively little is known about the adenosine uptake
system in the kidney. Saturable, temperature-sensitive
adenosine uptake has been demonstrated in renal brush
border and antiluminal membrane vesicles (Trimble and

Coulson 1984). Thompson et al (1985) provided evidence

that the kidney possesses an adenosine uptake mechanism
that is sensitive to dipyridamole, a nucleoside uptake
blocker. As discussed in III.A., elevation of endogenous
adenosine levels in the kidney by blocking uptake results
in profound changes in renal function, suggesting that the
control of extracellular adenosine concentrations is an

important part of the renal actions of adenosine.

A.2. Adenosine receptors

 

Purine compounds have been recognized for over fifty
years to have effects on organ function. In 1929, Drury
and Szent-Gyorgyi first reported that adenosine produced
vasodilation of the coronary arteries, lowered blood
pressure, induced sleep, and reduced intestinal motility.
Much of the work with purines in the next 20-30 years
centered on adenine nucleotides since it had been shown
that tissue adenosine concentrations were very low,
suggesting that adenosine was not a physiologically
important compound. However, in 1963, Berne reported an
increase in the amounts of inosine and hypoxanthine,
degradation products of adenosine, in coronary sinus blood
during hypoxia. In addition, Haddy and coworkers
demonstrated that physiological concentrations of adenosine
produce vasodilation in a number of organs (Haddy and Scott

1968) and vasoconstriction in the kidney (Haddy et a1

1965). Subsequent studies have demonstrated actions of
adenine compounds on the liver, skeletal muscle, vascular
smooth muscle and many other tissues. It was suggested
that ATP functioned as a neurotransmitter to produce many
of the effects attributed to adenosine in these organs
(Burnstock 1972, 1976, 1978). Later, it was established
that many of these actions were due to the direct
stimulation of extracellular receptors by adenosine.
Purinergic (P) receptors activated preferentially by
adenosine over the adenine nucleotides were termed P1
receptors, while those activated to a greater degree by
adenine nucleotides than by adenosine were labelled P2
receptors (Table 1, Burnstock 1978). Purinergic receptors
have been further characterized by their responses to
xanthine derivatives and their effects on cAMP production.
The actions of adenine compounds at P1, but not P2
receptors are antagonized by xanthine derivatives.
Activation of P1 receptors evokes changes in cellular cAMP
production, while P2 receptors do not. Subsequently, it
has been demonstrated that P1 receptors can be further
subdivided into two extracellular receptors, A1 (or R1) and
A2 (or R‘) receptors (VanCalker 1979; Londos 1980), and an
intracellular receptor, the P-site, with a distinguishing

feature being the direction of the change in cAMP

Table 1. Classification of adenine receptors

P1 P2.

 

Exncifikflxy eaivAanunbATP ATENEHDAMPka>
_P_-81£_ _1_\1__ A.

Inuxndon :unzaoerhnar tadzaoerhnarcmdzaoerhner tadzacerhuar

Infindxes:un2ct

nflxaetmnety no YES YES

Ra;dxes:hfl2ct

lazinetmflety yes no no

Saxniive'u)

:mnhyhmmtmuue xx) yes yes no

Efflafizonmaiaunate

cmclmaaacthdxy innnnx: innnnt stUmuate xx>efifix¢

lg. 1-10 m 1-10 nM 1 m

Anakxypouxcy noxxnoad:

(:uoadoxuxm. lfiifloadOXJDI

production. Activation of the A1 receptor and the
intracellular P-site is associated with an inhibition of
cAMP production, presumably through a decrease in the
activity of the enzyme adenylate cyclase. Stimulation of
adenylate cyclase activity and subsequent increase in cAMP
production characterizes activation of the A2 receptor.
Both extracellular receptors (A1 and A2) are sensitive to
xanthine derivatives, while the intracellular P-site is
not. The extracellular inhibitory receptor (A1) is a high
affinity receptor with a Km of approximately 1-10 nM, while
the stimulatory (A2) receptor has lower affinity for
adenosine with an approximate Km of 1-10 uM (reviewed by
Daly 1982). The intracellular site is activated only at
concentrations of adenosine in the millimolar range, and is
possibly mediated by intracellular conversion of adenosine
to 5'AMP.

The study of adenosine's actions at the P1 receptors
is hindered because adenosine is rapidly removed from the
extracellular medium by cells and metabolized.
Furthermore, the study of adenosine receptors is
facilitated by compounds with greater specificity for one
receptor type. Consequently, adenosine derivatives have
been synthesized which are not substrates for the

nucleoside carrier and have more specificity of binding to

10

either the A1, A2, or P-site adenosine receptor. The most
commonly used of these compounds are: 5'—N-ethy1carboxa-
mideadenosine (NECA), N5 -cyclohexy1adenosine (CHA), N5-
phenylisopropyladenosine (PIA), and 2-chloroadenosine. The
rank order of potency for these analogs at the A2 receptor
is: NECA>2-chloro>CHA=PIA and at the A1 receptor is:
CHA=PIA> 2-chloro>NECA. 2'5' dideoxyadenosine (DDA) and 2‘
deoxyadenosine (DA) act at the P-site to inhibit cAMP

production (Londos and Wolff 1977).
B. The Effect of Adenosine on Renal Function

8.1. Hemodynamics

 

Adenosine is a smooth muscle relaxant in most organs
(Harvey 1960; Haddy and Scott 1968; Walter and Bassenge
1968; Kraupp 1969: Ally and Nakatsu 1976). Adenosine
causes vasodilation of intestinal smooth muscle (Ally and
Nakatsu 1976) and most vascular beds, including the portal
vein, coronary arteries, and renal interlobular artery
(Walter and Bassenge 1968).4 However, bolus injections of
adenosine into the renal artery result in vasoconstriction
(Hashimoto and Kumakura 1965: Osswald 1975: Osswald et a1
1975; Osswald et a1 1978; Sakai et a1 1981: Scott et a1
1965; Spielman and Osswald 1978; Thurau 1964). Furthermore,
intrarenal infusion of adenosine produces a biphasic renal

blood flow response (Osswald et a1 1978; Spielman et a1

11

1980; Tagawa and Vander 1970). That is, immediately
following initiation of an adenosine infusion, renal blood
flow is reduced, but with continued infusion, blood flow
returns to the pre-infusion level or above. This biphasic
renal blood flow response to an adenosine infusion is not
completely understood, but has been suggested to be the
result of differential effects of adenosine on afferent
and efferent arteriolar resistances (Thompson et a1 1985),
or a differential effect on nephrons in the outer and inner
cortex (Spielman et a1 1980). As demonstrated by
microsphere studies, after the initial vasoconstriction has
waned and blood flow is at or above control, inner cortical
blood flow is increased, while that to the outer cortical
nephrons is unchanged (Spielman et al 1980).

Adenosine infusion results in a decrease in glomerular
filtration rate (GFR) that is sustained throughout the
period of the infusion, despite the return of blood flow to
or above control. The decrease in GFR, as well as the
biphasic blood flow response, may be due to an action of
adenosine to differentially alter afferent and efferent
arteriolar resistances. Thompson et a1 (1985) demonstrated
that there is no change in filtration fraction as renal

blood flow declines at the beginning of the infusion, but

12

filtration fraction decreases as renal blood flow returns
to control. These results suggest that adenosine produces
an initial constriction of the afferent arteriole, leading
to both the decrease in renal blood flow and glomerular
filtration rate, followed by a dilation of the efferent
arteriole. Efferent arteriolar vasodilation results in a
subsequent return of renal blood flow but a maintained
decrease in glomerular filtration rate due to decreased
filtration pressure in the glomerulus. Whether adenosine
alone produces both the constriction of the afferent
arteriole and the dilation of the efferent arteriole, or
whether another hormone contributes to this response, is

yet to be determined.

8.2. Renin release

 

In addition to changes in renal hemodynamics,
adenosine infusion into the kidney results in a pronounced
decrease in the release of renin (Osswald et a1 1978, 1982;
Tagawa and Vander 1970; Spielman 1984). A decrease in
renin release has also been demonstrated in renal slices
(Murray and Churchill 1984, 1985; Churchill and Churchill
1985) and in isolated perfused afferent arterioles (Itoh et

a1 1985). Recently, Kurtz et al (1987) observed an

13

adenosine-induced decrease in renin production by isolated,
cultured juxtaglomerular cells.

The effect of adenosine on renin release, as well as
renal blood flow and glomerular filtration rate,
demonstrates a sodium-dependency. That is, the adenosine-
induced changes in renal hemodynamics and renin release
were found to be greatly enhanced by prior sodium-depletion
of the animal and could be abolished by feeding the animal
a diet high in sodium prior to experimentation (Thurau
1964; Osswald et a1 1975). This sodium-dependent response
is also observed in the renal responses to increased
endogenous adenosine concentrations (III.A-C). Because
renal renin concentrations, and therefore, angiotensin
concentrations, are altered by changing sodium intake, an
interaction of adenosine with the renin-angiotensin system
was proposed. Indeed, the adenosine-induced
vasoconstriction and decrease in GFR can be attenuated by
the administration of angiotensin antagonists (Spielman and
Osswald 1979) and converting enzyme inhibitors (Hall and
Granger 1986). Additionally, as recently observed by
Edwards (personal communication), in the presence of
adenosine, angiotensin results in afferent arteriolar

vasoconstriction, an effect of angiotensin that is not

14

normally observed. These results suggest that the
vasoconstrictive action of adenosine may be mediated by
angiotensin. However, Spielman (1984) reported that
angiotensin concentrations in renal lymphatic fluid of the
dog were not increased, but actually decreased, during an
adenosine infusion, suggesting that angiotensin does not
mediate the vasoconstriction. Therefore, the exact
relationship between adenosine and the renin-angiotensin
system has yet to be determined.

The mechanism of the adenosine-induced decrease in
renin release has been studied at several levels. In
whole-animal experiments, Arend et a1 (1984) reported that
adenosine inhibition of renin release is independent of the
hemodynamic actions of adenosine and is not affected by the
administration of the organic calcium influx blocker,
verapamil. These results suggest a direct action of
adenosine on the renin-producing juxtaglomerular cells. It
is generally believed that renin secretion from the
juxtaglomerular cell is inversely linked to changes in
intracellular calcium (reviewed in Fray 1980). That is, an
increase in intracellular calcium in the juxtaglomerular.
cell results in a decrease in renin release and vice versa.

This hypothesis extends from the results of several studies

15

in which the intracellular concentration of calcium was
changed with an accompanying inverse change in renin
release. Agents such as catecholamines, which are known to
act through calcium in other cells can inhibit renin
secretion (Fray and Park 1979). Changes in the calcium
concentration of an intrarenal infusate (Vandongen and
Peart 1974; Fray 1977: Baumbach and Leyssac 1977) leads to
an opposite change in renin release. An increase in the
intracellular calcium concentration by the addition of
calcium ionophores (Baumbach and Leyssac 1977: Fynn et a1
1977) or by membrane depolarization with potassium
(Ettienne and Fray 1979; Fray and Park 1979) lead to a
decrease in renin secretion. Very recently, Kurtz et a1
(1987) reported that in isolated juxtaglomerular cells,
adenosine increases the production of cyclic 3',5'
guanosine monophosphate (cGMP), but not cAMP or calcium, in
these cells. These results suggest that cGMP may be the
intracellular mediator of the adenosine-induced decrease in

renin release.

8.3. Neurotransmitter release
A further action of adenosine in the kidney is its
ability to block the release of neurotransmitter from

sympathetic neurons (Hedqvist and Fredholm 1978). This

16

effect of adenosine is not confined to renal sympathetic
neurons, but is a widely observed phenomenon at other
sympathetic nerve terminals (reviewed by Fox and Kelley
1978). It was observed by Phillis and Edstrom (1976) that
adenosine depresses electrical activity of cortical
neurons, resulting in speculation that adenosine interferes
with neurotransmission by an antagonism of calcium influx
(Silinsky 1981).

The sympathetic nerves in the kidney innervate the
afferent and efferent arterioles, the juxtaglomerular
cells, and the renal tubules. Thus, adenosine inhibition
of renal neurotransmitter release has been proposed as a
possible mechanism of action for the changes in
hemodynamics, renin release, and excretion that occur with
adenosine infusions. However, because adenosine decreases
GFR, renin release, and excretion in chronic denervated
kidneys (Cook and Churchill 1984), where renal nerves are
not functional, inhibition of neurotransmitter release by
adenosine may be involved in modulating, rather than

mediating, some of the actions of adenosine.

8.4. Sodium and water excretion
Adenosine infusion results in a fall in urine output

and a decrease in sodium excretion (Churchill 1982; Osswald

17

et a1 1978). This could occur by a direct action of
adenosine on transepithelial solute and water movement and
this action will be discussed in greater detail in the
section on the cellular actions of adenosine (II.C.2).
Alternatively, because excretion is tightly coupled to
filtration, the decrease in water and solute excretion may
be an indirect result of the action of adenosine to
decrease renal blood flow and glomerular filtration rate.
The question of how adenosine exerts its action on
excretion has not been adequately addressed, primarily due
to the difficulty of separating this effect from the other
actions of adenosine, thus providing a rationale for
investigating the direct action of adenosine on isolated,
segment-specific cells of the kidney. Further support for
a direct effect of adenosine on the tubular epithelium
comes from studies performed in renal cell lines (Lang et
a1 1985) and other epithelial tissue (Stoff et al 1979;
Forrest et a1 1980; Poeschla et a1 1982; Dobbins et a1
1984; Grasl and Turnheim 1984; Dillingham and Anderson

1985) as discussed in Sections II.C.2 and II.C.3.

8.5. Endogenous adenosine
A crucial point in defining a role for adenosine in

the regulation of renal function is to demonstrate that

18

adenosine is present in the kidney and that maneuvers
resulting in an increase in intrarenal adenosine produce
changes in renal function consistent with the hypothesis.
It was suggested by Scott et al in 1965 that adenosine
might be involved in the vascular changes following brief
ischemia of the kidney because the hemodynamic responses to
occlusion of the renal artery and an intrarenal injection
of adenosine were similar.

The presence of adenosine in the kidney was
established in the 1970's by several groups (Osswald et a1
1977; Miller et a1 1978). These studies demonstrated
adenosine in the normal kidney and increased concentrations
of adenosine following ischemia in the kidney. Ischemia
was induced by occluding the renal artery for varying times
(15 sec - 10 min), and adenosine and ATP concentrations
were measured in whole kidney tissue. Following ischemia,
ATP concentrations decline and adenosine concentrations
increase proportional to the length of the induced
ischemia. These increased concentrations of adenosine were
accompanied by a brief reduction in renal blood flow, which
was attenuated by the adenosine receptor antagonist,
theophylline. These results suggested that during

ischemia, ATP is broken down to adenosine. Adenosine

19

accumulates in the kidney and subsequently acts to reduce
blood flow when the occlusion is released, possibly to
allow ATP concentrations to be restored before energy-
requiring transport begins again. Similar results were
reported by Osswald et al, in 1980, following the
intrarenal infusion of hypertonic saline. Hypertonic
saline infusion caused a vasoconstriction and an inhibition
of renin release, both antagonized by theophylline.
Adenosine concentrations were increased and accompanied by
a decrease in the tissue content of ATP. This suggests
that as the transport load of the kidney increases and ATP
is hydrolyzed to adenosine, adenosine acts to return
filtered sodium toward normal by decreasing blood flow and
GFR.

These studies have, as a common goal, to investigate
the way in which the kidney can regulate its own blood flow
and glomerular filtration. One way the intrinsic
regulation of renal blood flow and GFR could occur is by
the process of tubuloglomerular feedback (TGF). In TGF, a
signal, probably a chemical mediator, transmits information
through the macula densa area of the distal tubule, to the
glomerulus and the afferent and efferent arterioles. In

this way, each nephron controls its own blood flow and

20

filtration rate. Micropuncture studies show that increases
in tubular fluid flow rate lead to a decrease in single
nephron filtration rate, suggesting changes in blood flow
and glomerular dynamics of that nephron (Schnermann et a1
1970). Therefore, a more direct assessment of the role of
adenosine in intrinsic regulation of the kidney is to
employ the method of micropuncture to study the TGF
response in regard to adenosine. Osswald et a1 (1980) and
Schnermann et a1 (1977) have shown that theophylline
interferes with the normal TGF response. That is, in the
presence of theophylline, an increase in tubular perfusion
rate did not result in a change in filtration in the
perfused nephron. In addition, in the latter study the TGF
response was enhanced by blocking adenosine uptake with
dipyridamole. These results suggest that adenosine may be
involved in regulation of renal function as a mediator of

the tubuloglomerular feedback response.

C. Cellular Mediation of the Actions of Adenosine.

C.1. Non-epithelial cells

 

The nature of the adenosine receptor and the post-
receptor mechanisms responsible for the actions of

adenosine have been studied extensively in non-renal

21

tissues and cells. Many reports have demonstrated an
action of adenosine on cellular adenylate cyclase activity
and/or cAMP production. Others have correlated these
biochemical effects with specific biological aspects of the
tissue or cells.

An effect of adenosine on the adenylate cyclase enzyme
was demonstrated independently by two groups (Sattin and
Rall 1970: Shimizu and Daly 1970). They reported that
adenosine and some adenine nucleotides increased cAMP
concentrations in brain slices. Since then, an effect of
adenosine on adenylate cyclase activity and cAMP production
has been documented in a wide variety of neural tissue,
including neuroblastoma (Blume et a1 1973), astrocytoma
(Clark et a1 1974), and glioma (Clark and Seney 1976)
cells. In addition to nervous tissue, adenosine-sensitive
cAMP production has been demonstrated in platelets (Haslam
and Lynham 1972), adipocytes (Fain 1973: Fain et a1 1972;
Schwabe et a1 1973; Hjemdahl and Fredholm 1976),
hepatocytes (Lund et al 1975), lung tissue (Palmer 1971),
fibroblasts (Bruns 1980), lymphocytes (Wolberg et a1 1975),
myocardium (Huang and Drummond 1976), and coronary arteries
(Kukovetz et al 1978). The majority of these studies

demonstrate stimulation of adenylate cyclase activity

22

and/or production of cAMP, suggesting the presence of A2
adenosine receptors. Inhibitory (A1 or P—site) adenosine
receptors have been demonstrated in cultured astroblasts
(VanCalker et a1 1979), myocardium (Dobson 1978), platelets
(Haslam and Lynham 1972: Londos and Woff 1977), hepatocytes
(Lund et a1 1975), and adipocytes (Fain 1973: Pain et al
1972; Schwabe et a1 1973; Hjemdahl and Fredholm 1976).
Adipocytes appear to have only inhibitory adenosine
receptors. However, neuroblastoma cells, hepatocytes,
myocardium, and platelets exhibit both stimulation and
inhibition of cAMP production in response to adenosine,
suggesting the presence of both A1 and A2 receptors.

Some functional correlates to the biochemical effects
of adenosine have been defined. The majority of this work
has focused on the action of adenosine to inhibit lipolysis
in adipose tissue. Dole (1961) reported that adenosine and
adenine nucleotides inhibited hormone-stimulated lipolysis
in rat adipose tissue. Fain and coworkers (1972, 1973)
demonstrated the inhibition of norepinephrine-stimulated
adenylate cyclase activity, cAMP accumulation, and
lipolysis in fat cell ghosts by adenosine, 2'deoxyadeno-

sine, and 2'5'dideoxy adenosine.

23

Adenosine has also been shown to alter the pancreatic
secretion of both insulin and glucagon. Adenosine has been
reported to inhibit (Ismail 1977) and stimulate (Feldman
and Jackson 1974; Capito and Hedeskov 1976) the secretion
of insulin and to stimulate the secretion of glucagon (Weir
et al 1975). Mast cell histamine release is also
stimulated by adenosine (Sullivan et a1 1976). Stimulation
of steroidogenesis in adrenal and Leydig tumor cells has
been demonstrated with adenosine (Wolff and Cook 1977:
Kowal and Fiedler 1969). In these studies, tumor cell
dihydroprogesterone production, adenylate cyclase activity,
and cAMP accumulation were stimulated by adenosine and
inhibited by theophylline. The adenosine concentration
range for this effect (E050 = 10 uM) suggests mediation by
the A2 receptor. An important metabolic action of
adenosine has been suggested to be linked to the
immunosuppression associated with adenosine deaminase
deficiency. Wolberg et a1 (1975) demonstrated that
micromolar concentrations of adenosine stimulated
lymphocyte cAMP production that was associated with a
decrease in the degree of lysis of tumor cells.

The inhibition of neurotransmitter release from

sympathetic nerve terminals (Hedqvist and Fredholm 1978)

24

and the stimulation of glucose oxidation in hepatocytes
(Souness and Chagoya de Sanchez 1981) by adenosine are not
associated with changes in cAMP production. These
observations have prompted the speculation that adenosine
may act to produce these and other changes through a system
other than the adenylate cyclase-cAMP pathway. A possible
alternative is the involvement of calcium: either
stimulation of calcium movement across the cell membrane,
or a mobilization of calcium from intracellular stores.
Whether these cAMP-independent actions of adenosine are
mediated through an extracellular receptor and whether this
receptor is one of the currently recognized adenosine

receptors is unknown and will be discussed further in IV.B.

C.2. Epithelial cells

a. non-renal epithelia. Adenosine and various
adenosine derivatives have been reported to alter the
function of renal and non-renal transporting epithelia. In
1979, Spinowitz and Zadunaisky demonstrated adenosine-
stimulated electrogenic chloride transport in amphibian
corneas as evidenced by an increase in chloride flux
accompanied by increases in short circuit current and
potential difference when micromolar concentrations of

adenosine were added. Concurrently, Stoff et a1 (1979)

25

reported adenosine regulation of chloride transport in the
shark rectal gland. Following these studies, Forrest and
coworkers (Forrest et a1 1980, 1982, 1984; Poeschla et a1
1982, 1985; Dobbins et al 1985; Kelley et al 1984, 1985)
have shown the presence of both A1 and A2 adenosine
receptors in the shark rectal gland, which inhibit and
stimulate chloride transport, respectively. The inhibition
of chloride secretion by adenosine can be blocked by
treatment with pertussis toxin (Kelley et a1 1985) and,
therefore, is very likely to be mediated by the inhibitory
guanine nucleotide binding protein, N1 . Accompanying the
alterations in chloride secretion were changes in cAMP
production by the shark rectal gland. Increases in cAMP
accumulation and adenylate cyclase activity were observed
with micromolar concentrations of adenosine and adenosine
analogs. Inhibition of cAMP accumulation was demonstrated
at low (nanomolar) concentrations of adenosine. Recently,
Kelley et a1 (1986) have suggested that the inhibition of
chloride secretion may also involve a cAMP-independent
mechanism because the 2-chloroadenosine-induced decrease in
cAMP production was reversed by addition of forskolin, a
non-receptor mediated activator of adenylate cyclase, but

the accompanying inhibition of chloride secretion was not

26

affected. These results suggest the involvement of an
additional second messenger system in the inhibition of
chloride secretion in the shark rectal gland.

In addition to actions of adenosine on non-mammalian
epithelia, studies that involve rabbit ileum and colon have
demonstrated similar effects of adenosine on ion movement.
In 1984, Dobbins et al reported adenosine regulation of
chloride secretion in the rabbit ileum accompanied by an
increase in cAMP production, and Grasl and Turnheim (1984)
reported stimulation of electrolyte secretion by adenosine
in the rabbit colon.

b. renal epithelia. The study of the regulation of
ion movement by adenosine in renal epithelium has been
limited primarily to cell lines, such as toad (A6) and
porcine (LLC-PK) kidney cells. Lang et al (1985) reported
that adenosine stimulated the short-circuit current related
to sodium ion transport in the A6 cell line. In LLC-PK
cells (Roy et a1 1984), the adenosine analog PIA stimulates
the activity of the adenylate cyclase enzyme at high
concentrations. At low concentrations, PIA inhibited basal
adenylate cyclase as well as arginine vasopressin-
stimulated adenylate cyclase activity. Dillingham and

Anderson (1985) recently reported that NECA stimulates

27

water movement in the isolated perfused rabbit cortical
collecting tubule and that the P-site agonist,
2'5'dideoxyadenosine inhibits vasopressin-stimulated water
movement, providing support for adenosine as a regulator of

salt and water transport in the kidney.

III. TEE RENAL EFFECTS OF INCREASED ENDOGENOUS ADENOSINE

A. Dipyridamole decreases glomerular filtration

in the sodium-depleted dog .
(reproduced with permission, Am. Heart Assoc.; Circ. Res.
56:242, 1985)

A.1. Introduction

 

Work in several laboratories has raised the
possibility that intrarenally produced adenosine plays a
role in the intrinsic control of glomerular filtration rate
(GFR) and renin release (Spielman and Thompson 1982). It
has been postulated that extracellular adenosine, produced
from cellular 5'-adenosine monophosphate (5'-AMP),
presumably, but not necessarily, as a product of the
adenosine triphosphate (ATP) hydrolysis associated with
active transepithelial transport, acts to alter renal
function. Stated simply, increased active solute
reabsorption, secondary to an increased GFR, would increase
adenosine production which would, in turn, act to constrict
the afferent arteriole and dilate the efferent arteriole
(Osswald et al 1978b) and thereby reduce glomerular
capillary hydrostatic pressure and GFR. Evidence in
support of this hypothesis has come from studies that have

evaluated the renal action of exogenously administered

28

29

adenosine (Osswald et a1 1975, Spielman et al 1980) or, in
a few cases, with the use of methylxanthines as adenosine
antagonists (Osswald et al 1980, Gerkens et al 1983). For
the most part, however, only circumstantial evidence has
been obtained that supports a role for endogenously
produced adenosine in the control of renal function. A
more direct assessment of the actions of intrarenal
adenosine would be to increase endogenous concentrations of
adenosine, through pharmacological or physiological
perturbations, while monitoring renal function.

In a variety of nonrenal tissues, extracellular
adenosine has been reported to enter cells by facilitated
diffusion through a nucleoside uptake mechanism, where it
is deaminated to inosine or phosphorylated to 5'-AMP (Arch
and Newsholme 1978). This nucleoside uptake mechanism is
inhibitable by a variety of pharmacological agents
(Paterson 1979), such as nitrobenzylthioinosine (NBTI) and
dipyridamole. The hypothesis tested in this study was that
inhibition of cellular uptake of adenosine increases the
extracellular concentrations of adenosine, which in turn
alters renal function. The present investigation was thus
undertaken (1) to identify the presence of a renal uptake
mechanism for adenosine, (2) to determine whether its

inhibition with dipyridamole leads to increased

30

extracellular adenosine, and (3) to determine whether
increasing endogenous concentrations of adenosine by
inhibition of cellular uptake results in changes in renal
function that mimic the effects of exogenously administered

adenosine.

A.2. Methods

Preparation: Experiments were performed on dogs of
either sex, weighing 12-20 kg. The animals were allowed
free access to water and were deprived of food 24 hours
before the experiment. On the day of the experiment, the
animals were anesthetized with sodium pentobarbital (30
mg/kg, iv) and maintained with periodic small doses as
needed. The trachea was exposed and cannulated, and each
animal was mechanically ventilated with a Harvard
respirator. Minute volumes were initially selected from
the nomogram of Kleinman and Radford (1964). Body
temperature was monitored with a rectal temperature probe
and maintained at 37 C by a. circulating water-heating pad
and a radiant heat lamp, when necessary. Catheters (PE
240) were placed in the femoral arteries for measurement of
blood pressure (Statham strain gauge, P23Db, Hewlett-
Packard recorder, 77548) and for blood sampling, and in the
femoral veins for systemic infusions and anesthetic

administration.

31

A retroperitoneal flank incision was made to expose
the left renal artery for placement of a noncannulating
electromagnetic flow probe (Zepeda Instruments). The flow
probe was calibrated by renal artery cannulation at the end
of each experiment. A small curved infusion needle (22-
gauge) was placed into the renal artery distal to the flow
probe. The patency of the needle was maintained by a
constant infusion at a rate of 1.38 ml/min. Another small
curved needle (20-gauge) was placed in the renal vein for
sampling of the renal effluent. The dogs were then placed
in a metal frame which held them in a position approxi-
mating their normal standing posture. After the completion
of all surgical procedures, at least 1 hour was allowed for
attainment of steady state, as determined by the stability
of blood pressure, renal blood flow, and blood gas
measurements.

Analytical: Inulin concentrations in plasma and urine
were determined by the anthrone method described by
Davidson and Sackner (1963). Plasma and urine sodium and
potassium concentrations were determined by flame
photometry.

One and one-half milliliter aliquots of plasma and
urine taken during each clearance period were processed and

analyzed for adenosine as follows. The blood samples (3-4

32

ml) were immediately diluted in 300 ul of a solution
containing 32 uM dipyridamole, to block adenosine uptake by
red blood cells, and 3 uM erythro-9-(2-hydroxy-3—
nony1)adenine (EHNA), an inhibitor of adenosine deaminase.
The blood samples were then centrifuged and a 1.5 ml plasma
aliquot was removed and placed in 60 ul of 70% perchloric
acid. Similarly, 1.5 m1 aliquots of urine were placed in
60 ul perchloric acid. All samples were then centrifuged
at 10,000 g for 30 minutes to remove the protein
precipitate. A 1.2 ml aliquot of the perchloric acid
extract was removed and neutralized with a concentrated
solution of potassium carbonate (7 M). The potassium
perchlorate precipitate was removed by centrifugation and a
200 ul aliquot of the neutralized acid extract was used for
high-pressure liquid chromatography (HPLC) analysis of
adenosine. Four control tubes were prepared from a single
arterial or venous blood sample and processed with each
sample set. Two tubes received known amounts of adenosine,
and the third was used to measure endogenous levels. In
the final tube, the collecting solution was replaced with
2-5 units of adenosine deaminase activity in saline. The
adenosine values of these tubes were used to ensure the
identity of the adenosine peak and to calculate the

recovery of adenosine for each experiment.

33

The neutralized acid extract was injected (WATERS-
WISP) onto a C18 reverse-phase column (Beckman Ultrasphere
ODS) and eluted using an isocratic (9-12% of a 70%
solution, balance was 4 mM KZHPO4) elution (1-l.5 ml/min).
A fraction (12.5 min) of the eluate containing adenosine,
as determined by the retention time of standards (and
verified by the recovery tubes), was collected using a
timer (ChronTrol), valve (The Lee Co. #LFAA1201418A), and
fraction collector (Eldex Universal) system. The adenosine
fraction was concentrated to dryness under reduced pressure
using a vortex evaporator (Buchler Instruments). The
samples were resuspended in 1 ml of water and the adenosine
converted to inosine by the addition of 0.5 unit of
adenosine deaminase and incubated for 15 min at room
temperature. The adenosine deaminase was inactivated by
the addition of 1 ml of HPLC grade methanol, and the
samples were once again concentrated to dryness under
reduced pressure, and a 200 ul aliquot was injected on a
018 reverse-phase column and eluted as described above. We
quantified the inosine peak by measuring the area under the
curve (WATERS data module) and relating this to standards
run under the same condition. Recoveries for all
experiments were 100178;, and individual values were not

corrected for recovery.

34

Indicator Dilution Curves: To determine whether renal
adenosine uptake is blocked by dipyridamole, we employed a
single-injection, multiple-tracer, indicator-dilution
technique. This technique involves the simultaneous
injection of radiolabeled adenosine, together with a
radiolabeled reference, in this case a nontransported,
nonmetabolized analog of adenosine, ara-H, and collection
of serial samples of the urinary and renal venous
effluents. Comparison of the tracer adenosine to the
reference in urine and renal venous blood yields important
information about the renal handling of adenosine (Thompson
et al 1985). Based on knowledge of adenosine transport in
other tissues, we hypothesized that injected adenosine can
leave the vascular compartment by simple diffusion into the
interstitial space, as well as by facilitated diffusion
into cells via the nucleoside carrier, whereas the
reference, ara-H, can leave the vascular compartment only
by simple diffusion. Thus, the difference in the recovery
of tracer adenosine and the recovery of the reference was
taken as the uptake and metabolism of adenosine by cells
(Thompson et a1 1985). The [U-“CJ-adenosine (278
mCi/mmol) was obtained from Amersham. The 9-s-D-arabino-
furanosyl-[adenine-2-3H] (ara-A) (32.6 Ci/mmol) was

obtained from ICN. The ara-A was converted to 9-B-D-

35

arabino-furanosyl hypoxanthine-[2-3H] (ara-H) by treatment
with adenosine deaminase (Sigma Chemical) and subsequent
purification on a HPLC reverse phase column. Injectate
containing [3H]ara-H (2 uCi/ml) and [“C]adenosine (0.34
uCi/ml) was made up in 1.5 ml of normal saline. The bolus
was quantified by diluting 100 ul of the injectate with 900
ul saline and counted for radioactivity. Bolus injections
(2-4 sec) of 1 ml were made directly into the renal artery
cannula while a constant infusion rate of 1.38 ml/min was
maintained to flush the cannula. Immediately before
initiation of the bolus injection, sampling was begun from
both ureters and the renal vein. Before each injection,
renal vein plasma and urine samples were obtained for
determination of blank activity. Urine samples were
collected for an average of 30 seconds (range 15-60 sec) to
Sobtain approximately 1 ml of urine per sample depending on
urine flow. Contiguous lO-second withdrawals via the renal
vein catheter (average 20 ml/min) were made starting with
the injection and lasting 2 minutes after completion of the
injection. All samples were corrected for the blank values
immediately before the injection, and results were
normalized as the plasma or urine concentration (% of

injectate/ml).

36

Urine samples (0.7-1.2 ml) for radioactivity
determinations were collected directly into preweighed
scintillation vials. After sample collection, the volume
was determined gravimetrically; :n: a few cases where the
sample volume was less than 0.7 ml, 0.5 m1 of saline was
added after volume determination.

Statistical: Statistical analysis was performed by

 

analysis of variance using the Student-Newman-Keuls' test
for multiple comparisons and by Student's paired t-test
when appropriate. A P value of <0.05 was considered
statistically significant. Data are presented as mean i 1
SEM.

Experimental: Group I - Effect of dipyridamole on the
renal uptake of adenosine and the concentration of
adenosine in plasma and urine (n=5).

To determine whether renal adenosine uptake is blocked
by dipyridamole, we employed a single-injection, multiple-
tracer, indicator-dilution technique. The uptake of
adenosine was compared with that of a nonmetabolized analog
of adenosine, are-H, used as an extracellular reference. A
group of dogs maintained on a normal laboratory diet were
prepared as described above, with the following additions.
A right flank incision was made to expose the right ureter,

which was cannulated with PE 240 tubing. The incision was

37

then closed to prevent excess fluid loss. After the
surgical preparation, the animals were given a saline
volume expansion of 5% of body weight to initiate a brisk
urine flow. An intravenous infusion of saline (4—5 ml/min)
was given to maintain urine flow rate at approximately 2
ml/min. After at least 60 minutes had been allowed for
stabilization of the animal preparation, renal adenosine
uptake studies were performed, and samples for adenosine
concentration were taken, as described above, during a
control period, and after 45 minutes of dipyridamole
infusion into the renal artery at 24 ug/kg/min.

Group II - Effects of an adenosine uptake
inhibitor, dipyridamole, on renal function in dogs
maintained on a low-sodium (Group IIa, n=9) or a high-
sodium intake (Group IIb, n25).

To determine whether increasing endogenous adenosine
by inhibition of cellular uptake produced renal effects
similar to those resulting from the intrarenal infusion of
exogenous adenosine, we studied renal function before,
during, and after the intrarenal infusion of dipyridamole.
Because the effects of exogenous adenosine on renal
function are enhanced by sodium depletion and attenuated or
blocked by sodium loading (Osswald et al 1978a), dogs were

divided into two groups and either sodium-depleted or

38

sodium-loaded to determine whether the renal effects of
increasing endogenous adenosine by inhibition of uptake are
also influenced by dietary sodium intake.

One group of dogs (Group Ila) was given 100 mg of the
diuretic, furosemide (iv) on the first day of the dietary
regimen to accelerate the volume-depletion desired, and
maintained on a low-sodium diet (<2 mEq/day) for 10-14 days
before the experiment. Dogs were allowed water ad libitum.
On the day of the experiment, animals were prepared as
described above. During the experiment, dextrose (5.5% in
water) was infused (1.38 ml/min) into the renal artery for
the control period. Urine was collected for two or three
15-minute clearance periods, and mid—point systemic
arterial and renal venous blood samples were taken. This
procedure was repeated during the infusion of dipyridamole
(24 ug/kg/min in 5.5% dextrose) and again during dextrose
infusion for the recovery period. A period of 30 minutes
was allowed after the onset and termination of the
dipyridamole infusion before beginning the dipyridamole and
recovery clearance periods, respectively.

Another group of dogs (Group IIb) was given 15 mg
deoxycorticosterone acetate each day to cause sodium
retention and fed the same low-sodium food as the previous

group to which had been added 200 mEq sodium (as NaCl) to

39

produce a high-sodium diet. This dietary regimen was
continued for 10-14 days. The experimental procedure for
Group IIb dogs was identical to the procedure outlined
above for Group IIa except that saline (0.9%) was used in
place of dextrose for the control and recovery periods and
was used as vehicle for the dipyridamole.

In both groups, bolus injections of adenosine (5 and
10 nmol) were given into the renal artery during the
control and dipyridamole periods to test for enhancement of
the renal vasoconstriction to exogenous adenosine by
dipyridamole.

Group III - Effect of dipyridamole on renal
function in the sodium-depleted dog, during constant renal
perfusion pressure (n=5).

To determine whether the renal actions of dipyridamole
in the low-sodium dog (Group IIa) were a consequence of the
decrease in arterial blood pressure, another group of dogs
was sodium-depleted and prepared as described above, except
that a clamp was placed around the abdominal aorta cephalad
to the left renal artery. Two clearance periods were
conducted during the infusion of 5.5% dextrose at the
animals' ambient blood pressure. The aortic clamp was then
tightened to lower the renal perfusion pressure (RPP) to

100 mm Hg, which is in the autoregulatory range and was

40

near the average mean arterial pressure (MAP) of the dogs
in Group IIa during steady-state dipyridamole infusion.
The dextrose infusion was continued, and clearance periods
were conducted. Maintaining RPP at 100 mm Hg, dipyridamole
was infused and clearance periods were conducted after 30
min of equilibration. The dipyridamole was replaced .by
dextrose, and RPP was maintained at 100 mm Hg for the
recovery clearance periods.

Group IV - Effect of an adenosine antagonist,
theophylline, on the dipyridamole-induced changes in renal
function in the sodium-depleted dog (n28).

To determine whether the renal effects of dipyridamole
could be inhibited or reversed by an adenosine antagonist,
a group of dogs (Group IV) was sodium-depleted and prepared
as described above. In five dogs (Group IVa) dextrose was
infused (2.76 umol/min) into the renal artery and control
urine and blood samples were collected. The dipyridamole
infusion was then started (24 ug/kg/min, 1.38 ml/min) while
the dextrose continued (1.38 ml/min). After a 30 min
equilibration period, blood and urine samples were
collected. The dextrose infusion was replaced by
theophylline (5 umol/min) while the dipyridamole continued,
and another equilibration period was allowed before samples

were collected again. In a group of three dogs (Group

41

IVb), the above procedure was followed, except that the
theophylline and dipyridamole periods were reversed.
Theophylline (5 umol/min) was infused during a dextrose
infusion for the second set of samples, and then the
dextrose was replaced by dipyridamole (24 ug/kg/min).

In both groups, bolus injections of adenosine (5 and
10 nmol) were given into the renal artery to test for
enhancement and attenuation of the vasoconstriction by

dipyridamole and theophylline, respectively.

A.3. Results

Group I - Effects of dipyridamole on the renal uptake
of adenosine and the concentration of adenosine in plasma
and urine.

The cumulative venous recovery of labeled adenosine
and ara-H was 15.6 1 2.7 and 50.5 1 7.6%, respectively,
during control. In the presence of dipyridamole, the
recovery of adenosine increased to 48.1 ~_t_ 8.9% (P<0.05
compared to control) and was not different from the venous
recovery of ara-H (56.4 i 1.5%, NS) during dipyridamole.
The ratio of adenosine to era-H in the first eluting
samples averaged 0.28 1 0.03 compared to 0.91 1 0.03 during
dipyridamole (P<0.05), indicating that the cellular uptake

of adenosine was inhibited in the presence of dipyridamole.

42

Representative dilution curves for the venous effluent
before and during dipyridamole are shown in Figure l.

The control urinary excretions following injection of
labeled adenosine and ara-H were 10.8 1 1.2 and 19.8 1 1.7%
of the amount injected. Excretion of the adenosine label
was increased in the presence of dipyridamole to 15.2 _+_
2.2% and was not different from ara-H excretion (20.1
2.5%, NS). The adenosine:ara-H ratio in the early eluting
samples was 0.57 i 0.03 in control and increased to 0.82 1
0.01 during dipyridamole (P<0.05). These results, similar
to those obtained with renal venous plasma, indicate that
this concentration of dipyridamole blocked the cellular
uptake of tracer adenosine. Representative dilution curves
for the urinary effluent before and during dipyridamole are
shown in Figure 2.

The renal venous concentration of endogenous adenosine
increased from 48 1 9 nM in control to 108 1 25 nM during
dipyridamole. The urinary excretion of endogenous
adenosine similarly increased from 595 1 77 pmol/min to
1695 i 207 pmol/min during dipyridamole (P<0.05). The
concentration of endogenous adenosine in arterial blood
also increased during dipyridamole, however, due to the

variability of the response (in two of the five dogs

'5 NJECTATE/ml

20

43

 

 

 

300

DIPYRIOAMOLE

 

 

O ARA-H
I ADENOSINE

 

 

 

 

 

 

 

 

 

 

3O 60 90 1 2O 0 5 O 100 ) 50 200

 

 

 

"I PLASMA ml PLASMA

Figure 1. Dilution curves for adenosine and ara-H in plasma

from a representative single-injection, multiple-
tracer experiment. Data in the left panel were
collected during saline infusion, and in the
right panel, during dipyridamole infusion. Each
data point represents the concentration of
tracer, normalized as percent of injected dose,
recovered in the venous effluent. The curves
were generated by fitting the data with a cubic
function. Total sampling time was 2 minutes.

44

 

 

CONTROL ‘ L onpvmomous
o ARA-H o ARA-H
6* I Aoenosme ‘ 3 I I ADENOSINE

 

 

 

 

 

 

 

% lNJECTATE/ml

 

 

 

 

 

 

 

I 1 22
rd URIPE ml URINE

Figure 2. Dilution curves for adenosine and ara-H in the
urinary effluent. See legend for Figure 1 for
explanation. Total sampling time was 10 minutes.

45

studied, arterial adenosine decreased) this increase was
not statistically significant (73 i 26 vs 165 i 59 nM, NS).

Group II - Effects of an adenosine uptake inhibitor,
dipyridamole, on renal function in dogs maintained on a
low-sodium or a high-sodium intake.

Sodium-depleted and sodium-loaded states were verified
by urine flow rate, sodium excretion, and fractional sodium
excretion. These indices of renal function were
significantly elevated in the animals receiving a high
sodium diet and were reduced in the sodium-depleted
animals. Urine flow rate was 0.55 i 0.25 ml/min in the
sodium-depleted group and 1.29 1 0.42 ml/min in the sodium-
1oaded group. The excretion of sodium was 4.4 i 3.6 and
77.8 i 25.1 mEq/min and the fractional excretion of sodium
0.296 1 0.158 and 2.32 i 0.61%, in the sodium-depleted and
sodium-loaded dogs, respectively. The attainment of a
desired sodium status was verified in all subsequent
animals by these measurements.

The intrarenal infusion of dipyridamole caused a
decrease in GFR in the dogs maintained on the low-sodium
diet but produced no significant change in GFR when infused
into the dogs that had been maintained on the high-sodium
diet (Figure 3; Table 2). GFR was decreased in the low-

sodium dogs (Group Ila) by an average of 58.6% (19.8 i 1.1

46

 

 

ear 1 j

rSodium-depleted, ms] Sodium-loaded,n:5 I

  

GFR

(ml/min)

 

 

 

 

 

CON DIPYR REC CON DIPYR REC

Figure 3. Comparison of glomerular filtration rate (GFR)
before, during, and after the intrarenal infusion
of dipyridamole (24 ug/kg/min) in sodium-depleted
and sodium-loaded dogs. Each bar represents the
mean 1 SEM for the group. Individual animal
values were taken as the mean of three clearances
in each period. Asterisk indicates P<0.05.

47

 

z .. . K n c a. w ...: ...... . z... ...... H 3. .. ...... « ...... ... a .... 2 5 K. ....a a m S ......<.§:.
2.2.... .25.: ......32. 2:22.. ......32 a3... :32 .33...” ...............:=
.22.; :43: 2:22.. 3:3... 2:282. on... .2... 3:...“ .25....
.n n ...
533...... 5...?! 52
....33 .... “.3: 2.2.2... 2....«2... ......“ ...... a“... ._ :2 ... :2 25.2.5:
: .2 ...... .22... 2:33.. 2....«2... 5:32... ...«2. .m. «.... ....N g... .......:....;...._
2...... 2:2: 2... .2. 2.38.. 2:32.... ..«z. : Hz. ... ...: .55....
.... u ...
3....._....—. 5:35.. :5.
.... .. “ mm c on H 2.: 2... « ...... z... a K... 2.... a .n .. ...: 2 H .... v p. a . z. 225......
: .. a 2 . o n a ....n. ...... . ... .. 2... « ...... .2... a 2 .. ...: 3 u 2.. ... n a .. m. .....5....5....:
52.2....“ 2:2. .252... .....«fi... £522. ...: .22. 3:: .....55
$.23... .23.: ......«R: 255...... 8.2.3.. :2... 2.3.: :53. .....5.
:m u ...
3:395 ...—395..
arse-...... =53...» ....
2.2%... 33...... 3..“ ...~ 2.12... 2:23.. ... «...... .... «a... 22.2 225......
..... .. « .2 ...n a 2.... ...... « .... N N. .. « ...... 2... « ..~ .. .. u .... ... a S. o n a . mm 51.55.25:
.22.: 32...: 8.3.... 2:32... 2:13.. :2: :2... :52... .....5...
.m u ...
3.4.5.. ...—5...... 5:
a... q 2. n ..n H I: 9.... a 3... 2... a 2.... ...... fl 2... m w ... 2 a m: ... u n ..~ 5.53....
N. .. fl 2.... .... a 2.: ...... u m. .. z .. a 2... .2... q n. .. ... u .... .... « 2. .... . a N... 35.53.25
2.. 3: .... «:2 2.32.. ......«z... .....IE. mu .2 ... um: ... «a... .....55
.93 u Ev
25...... 5.5.... 3..
......éaa. .8535. ...... ..5 .8. 55B: 55. u: a: 55. .5525. .5525. ......o
... «5...... ...z .53... ...... ...... ...... 2... ......

 

.mHOEmcwuamflo cu mmmcommmu ”anon on» :o ocwaamamowcu no uommwm on» can mmoc
poomoausafioom can cmumammcnaafipom :« newuocsu Hucmu co oaosmpwuamfic mo uommum .m manna

48

vs 8.2 1 1.5 ml/min, P<0.05), and returned to control in
the recovery period (20.3 1 1.1 ml/min). In the high-
sodium dogs (Group IIb), the average GFR during the
infusion of dipyridamole was not significantly different
from the control GFR (22.3 1 3.9 vs 25.1 1 3.4 ml/min,NS).
The intrarenal infusion of dipyridamole in the low-sodium
dogs consistently produced a small and transient increase
in renal blood flow at the onset of the infusion, which
waned, and during the steady state was lower than the
control blood flow (145 1 10 vs 117 1 10 ml/min, P<0.05).
Steady state blood flow during the infusion of dipyridamole
was also somewhat lower in the high-sodium dogs, albeit,
not significantly (178 1 24 vs 157 1 18 ml/min, NS).

The intrarenal infusion of dipyridamole had no effect
on the renal perfusion pressure in the dogs maintained on
the high-sodium diet. In contrast, the infusion of
dipyridamole resulted in a decrease of renal perfusion
pressure in the low-sodium dogs (121 1 5 to 101 1 6 mm Hg,
P<0.05). Renal vascular resistance (RVR) was not
significantly different in either group of dogs during the
infusion of dipyridamole. Filtration fraction (FF)
decreased significantly in the sodium-depleted dogs (0.28 1

0.03 vs 0.13 1 0.02, P<0.05) during the infusion of

49

dipyridamole, but was not significantly different in the
sodium-loaded dogs.

Fractional excretion of sodium and the plasma
concentrations of sodium and potassium were not
significantly different from control values during the
infusion of dipyridamole.

In Group IIa dogs (low-sodium), injections of 5 and 10
nmol of adenosine resulted in 45 1 8 and 56 1 10% decrease
in renal blood flow (RBF) during the infusion of dextrose;
during the infusion of dipyridamole, the decreases in REF
in response to 5 and 10 nmol adenosine were significantly
greater (66 1 11 and 74 1 ll%,respectively; P<0.05). In
the high-sodium group, RBF decreased by 14 1 3 and 25 1 8%
during the control saline infusion and by 28 1 5% (P<0.05)
in response to 5 nmol adenosine and 39 1 6% (NS) in
response to 10 nmol adenosine during the dipyridamole
infusion.

In some preliminary experiments, dipyridamole was
infused into the femoral vein. With systemic
administration, the effects of renal function were the same
as that seen in the intrarenal infusion.

Group III - Effect of dipyridamole on renal function

during constant renal perfusion pressure.

50

An initial clearance period was performed in each
animal at its ambient arterial pressure. The average
initial RPP for five dogs was 133 1 3 mm Hg. Initial GFR
was 25.7 1 3.4 ml/min, and RBF was 165 1‘ 27 ml/min.
Filtration fraction was 0.30 1 0.05 and fractional sodium
excretion was 0.77 1 0.54% (Figure 4).

There was no effect of decreasing renal perfusion
pressure (RPP) to 100 mm Hg on GFR (25.7 1 3.4 vs 24.4 1
4.1 ml/min, NS) or RBF (165 1 27 vs 163 1 34 ml/min, NS).
Filtration fraction and vascular resistance remained
constant as well (Table 2). With the decrease in RPP,
fractional excretion of sodium decreased significantly from
0.77 1 0.54 to 0.16 1 0.09% (P<0.05).

With RPP maintained at 100 mm Hg, dipyridamole
infusion resulted in a decrease in GFR from 24.4 1 4.1 to
15.0 1 3.6 ml/min (P<0.05). There was no change in RBF
with dipyridamole infusion (167 1 43 ml/min, NS);
consequently, filtration fraction decreased from 0.32 1
0.06 to 0.22 1 0.06 (P<0.05). Vascular resistance and
fractional sodium excretion did not change significantly
from the previous period.

During the recovery period, (i.e., 30 minutes after
stopping dipyridamole infusion), GFR and filtration

fraction both returned to their respective control values

51

 

 

 

RPP ‘50’
(ml-lg) 100- ‘
50.. d
0 1
w: 0

RBF 3°°’
("l/rm) 200' f ‘
100r' ‘
0P '1
4L 1‘

25r-
GFB 20.. -1
mm 15+ «
10" ‘
5- -

 

 

 

 

 

 

 

INTL CON WDIPYH” REC

Figure 4. Renal perfusion pressure (RPP), renal blood flow
(RBF), and glomerular filtration rate (GFR) in
five sodium-depleted dogs before, during, and
after the intrarenal infusion of dipyridamole (24
ug/kg/min) while renal perfusion pressure was
maintained at 100 mm Hg. Each value represents
the mean 1 SEM for the group. Individual animal
values were taken as the mean of three clearances
in each period. Asterisk indicates P<0.05.

52

(GFR: 24.1 1 3.4 ml/min; FF: 0.31 1 0.07). There was no
significant change in RBF, vascular resistance, or
fractional sodium excretion.

Group IV - The effect of theophylline on the renal
actions of dipyridamole.

The intrarenal infusion of dipyridamole in Group IV
resulted in changes similar to those described for Group 11
dogs (Table 2; Figure 5). GFR decreased from 21.9 1 1.0 to
7.9 1 2.5 ml/min (P<0.05); likewise RBF decreased from the
control value of 154 1 14 ml/min to 119 1 15 ml/min
(P<0.05). Filtration fraction also decreased from 0.29 1
0.04 to 0.12 1 0.03 (P<0.05). As with Group II, MAP
decreased during the dipyridamole infusion from 124 1 8 to
99 1 8 mm Hg (P<0.05). Injections of adenosine during the
control dextrose infusion resulted in decreases in RBF of
35 1 10 and 42 1 11% for 5 and 10 nmol adenosine,
respectively. In response to the same doses during the
dipyridamole infusion, the decreases in RBF were
significantly greater (51 1 14% and 60 1 12%; P<0.05). The
concomitant infusion of theophylline resulted in the
reversal of all of these effects (Figure 5). GFR increased
to 25.7 1 1.9 m1/min and RBF increased to 157 1 11 ml/min.
Neither increase was significantly different from the

respective control values. The filtration fraction

53

 

 

 

 

125 >
i- 0
MAP 0100 b
(mmHg)
9 - 150
. REF 0
« 100
_ (ml/min)
20 >
GFR A _
(ml/min) 1o _ '
o A l A
CON DIPYR DIPYR
8 THEO

Figure 5. Mean arterial pressure (MAP), renal blood flow
(RBF), and glomerular filtration rate (GFR), in
five sodium-depleted dogs before and during the
intrarenal infusion of dipyridamole (14 ug/kg/
min), and during the intrarenal infusion of
theophylline (5 umol/min) and dipyridamole,
simultaneously. Each value represents the mean 1
SEM for the group. Individual animal values were
taken as the mean of three clearances in each
period. Asterisk indicates P<0.05.

54

returned to the control value (0.31 1 0.03) during
theophylline infusion as well. Renal perfusion pressure
increased from 99 1 8 to 112 1 7 mm Hg, a value which was
not statistically different than the control value of 124 1
8 mm Hg. The decreases in RBF in response to adenosine
injections were 31 1 15 and 43 1 20% during the
theophylline infusion, and were not significantly different
than the control responses.

In an additional three dogs, theophylline was infused
prior to dipyridamole to determine if the effects of
dipyridamole on renal function could be inhibited by the
presence of theophylline. Theophylline infusion alone did
not significantly change GFR from the control value (31.3 1
5.2 vs 30.6 1 4.9 ml/min, NS). Likewise, RBF did not
change in the steady state (165 1 7 vs 175 1 15 ml/min,
NS), although there was an initial tendency of blood flow
to increase immediately after theophylline infusion was
begun. Renal perfusion pressure was also not affected by
theophylline (117 1 6 vs 107 1 3 mm Hg, NS). Fractional
sodium excretion increased from 0.28 1 0.08 to 0.74 1 0.10%
(P<0.05). Theophylline infusion did not affect renal
vascular resistance or filtration fraction (Table 2). The
infusion of theophylline attenuated the decreases in RBF to

adenosine injections as compared to control. Injections of

55

adenosine (5 and 10 nmol) during the dextrose infusion
resulted in 44 1 3 and 52 1 6% decreased in RBF. During
the theophylline infusion, the decreases in RBF in response
to adenosine were significantly less (19 1 5 and 33 1 3%,
P<0.05).

In the presence of theophylline, the infusion.of
dipyridamole did not significantly affect GFR (30.6 1 4.9
vs 25.5 1 3.5 ml/min, NS) or RBF (175 1 15 vs 174 1 23
ml/min, NS). Filtration fraction did not change from the
value during theophylline infusion alone (0.36 1 0.08 vs
0.29 1 0.05, NS). Likewise, dipyridamole infusion in the
presence of theophylline had no effect on renal perfusion
pressure (107 1 3 vs 96 1 6 mm Hg, NS); however, this was
significantly different from the control value of 117 1 6
mm Hg (P<0.05). Vascular resistance and fractional sodium
excretion were not changed with the infusion of
dipyridamole (Table 2). Adenosine injections (5 and 10
nmol) during the theophylline/dipyridamole infusion
resulted in 33 1 l7 and 47' 1 21% decreases of RBF which
were not significantly different than the control

responses.

A.4. Discussion

 

The reported actions of adenosine, namely, the

production of a marked decrease in GFR (Tagawa and Vander

56

1970), alteration of cortical distribution of blood flow
(Spielman et a1 1980), and a decrease in renin release
(Tagawa and Vander 1970), form the basis of the hypothesis
that intrarenally-produced adenosine, as'the active
extracellular metabolite of the ATP hydrolysis associated
with transepithelial transport, is the mediator of the
intrinsic regulation of renal function. However, the
majority of the previous studies have involved the use of
exogenous adenosine, and, whereas these studies provide
important support for the role of adenosine as a regulator
of renal function, alone, this evidence can be considered
as merely suggestive. The use of dipyridamole, a
nucleoside uptake blocker, in the present study more
directly allows for the investigation of the role that
endogenous adenosine may play in the regulation of renal
function. The hypothesis tested in the present study was
that inhibition of the cellular uptake of adenosine
increases the extracellular concentration of endogenous
adenosine, which, in turn, alters renal function. We
therefore sought to determine the effect of dipyridamole on
(1) the renal uptake of adenosine, (2) the plasma and
urinary concentrations of adenosine, and (3) renal

function.

57

For any substance to have a role as a regulatory
intercellular mediator, it must have a system for
inactivation, through uptake and/or metabolism. Nucleoside
uptake is known to occur by facilitated diffusion in a
variety of cell types (Berlin and Oliver 1975), and is
inhibitable pharmacologically (Paterson et a1 1975).
Dipyridamole has been shown to block the uptake of
adenosine by a number of cell types, including
erythrocytes, platelets, and myocardial cells (Feigl 1983).
In the present study, we used dipyridamole as a tool for
.determining whether an adenosine uptake mechanism exists in
renal cells and whether blocking this uptake system
increases extracellular adenosine concentrations.
Dipyridamole could then be used to probe the effects of
intrarenally-produced adenosine on kidney function. The
difference between the recovery of adenosine and that of
the reference is virtually abolished during the infusion of
dipyridamole (Figures 1 and 2), and indicates that the
cellular uptake of adenosine is blocked. This inhibition
'of the cellular uptake is apparent in both the renal venous
and urinary effluents, leading to the conclusion that the
kidney possesses a dipyridamole-sensitive cellular uptake
mechanism for adenosine in both the vascular and tubular

compartments.

58

Once in the cell, adenosine is metabolized to inosine
or 5'-AMP by adenosine deaminase or adenosine kinase,
respectively, Because adenosine metabolism is primarily
intracellular (Berne and Rubio 1974) and its physiological
actions are primarily extracellular (Arch and Newsholme
1978), exogenous administration of adenosine during
pharmacological inhibition of the cellular uptake is
generally expected to result in increased extracellular
concentrations and, hence, increase its biological
activity. In the heart, dipyridamole enhances the coronary
vasodilation produced by adenosine (Feigl 1983). Likewise,
the renal vasoconstriction produced by the intrarenal
injection of adenosine is augmented in the presence of
dipyridamole (Sakai et a1 1981). In the present study,
bolus injections of adenosine produced transient
vasoconstrictions that were enhanced during the infusion of
dipyridamole.

The knowledge that dipyridamole blocks the uptake and
potentiates the action of exogenously administered
adenosine has led investigators to use dipyridamole to
enhance events suspected to be mediated by increased
concentrations of endogenous adenosine (Feigl 1983).
However, because the mechanism by which intracellular

adenosine exits the cell remains to be defined, and because

59

the dipyridamole-inhibitable, facilitated diffusion is
presumably a bidirectional process (Plagemann and
wohlhueter 1980), release, as well as cellular uptake, may
be inhibited in the presence of dipyridamole. Thus, it is
important to ascertain whether blocking cellular uptake is
actually associated with increased concentrations of
adenosine.

In general, it can be assumed that the concentration
of adenosine in venous blood and urine can be used as an
indication of the concentration of adenosine in the
extracellular fluid. Measurements of adenosine in the
plasma and urine before and during the administration of
dipyridamole indicated that extracellular adenosine
concentrations are indeed increased by dipyridamole. The
concentration of adenosine in arterial blood also increased
during dipyridamole, and this is presumably due to the
effect of dipyridamole on adenosine uptake in other
tissues, a finding not surprising in light of the decrease
in arterial blood pressure. The increases in venous and
urinary adenosine concentration were not, however,
dependent on an increase in extrarenal adenosine levels,
since the increase in venous and urinary adenosine was also
seen in two of the five experiments in which the arterial

concentrations were decreased or unchanged. In addition,

60

the arterial concentration of adenosine was increased, on
the average, to slightly more than 100 nM, and the changes
in renal function that occurred would require, in an
infusion of exogenous adenosine, a concentration of 5 nM,
or 50 times the concentration observed with the
dipyridamole infusion, suggesting that the functional
changes were probably elicited by increases in intrarenal
adenosine levels.

Having established that (1) cellular uptake of
adenosine is blocked by dipyridamole in the kidney, and (2)
adenosine concentrations are increased during the
administration of dipyridamole, we then sought to determine
the effects of increased endogenous adenosine on renal
function. The renal actions of adenosine are influenced by
the dietary sodium intake of the animal (Osswald et al
1978a). That is, the intrarenal infusion of adenosine
results in a decrease in GFR in sodium-depleted rats and
dogs, but has relatively little, if any, effect when
infused into sodium-loaded animals. Likewise, the effect
of exogenous adenosine to produce a redistribution of renal
cortical blood flow is influenced by the dietary sodium
intake of the animal (Spielman et a1 1980). The mechanism
responsible for these differences in response to adenosine

is unknown at present, although the enhanced renin-

61

angiotensin system and increased renal nerve activity
present in the sodium-depleted animal are possibilities.
Several recent studies have been aimed at clarifying the
role of angiotensin in the renal actions of adenosine. The
reduction in RBF following a brief occlusion of the renal
artery - which is thought to be mediated by adenosine - as
well as the decrease in RBF in response to an injection of
adenosine, have been attenuated by high concentrations of
the angiotensin II antagonist [Sar1,Ile3]AII (Spielman and
Osswald 1979), suggesting an adenosine-All interaction.
However, measurements of renal lymph AII, used as an
indication of interstitial All levels, have shown that
intrarenal AII did not increase, but actually decreased,
during an adenosine infusion (Spielman 1984). Therefore,
whereas an adenosine-All interaction appears to exist, it
may be more complicated than has been previously
postulated.

To determine the renal effects of increased
endogenous adenosine and the influence of sodium intake on
these actions, we studied the effects of dipyridamole
administration in sodiumedepleted and sodiumrloaded dogs,
thereby comparing the renal effects of endogenous to
exogenous adenosine. Animals maintained on a diet low in

sodium were found to have a reversible decrease in GFR in

62

response to the dipyridamole infusion, whereas GFR did not
change upon infusion of dipyridamole in the sodium-loaded
dogs.

In addition to the decrease in GFR, renal blood flow
was also reduced during the dipyridamole infusion in the
sodium-depleted dog. There was, however, no calculated
change in the renal vascular resistance as mean arterial
blood pressure was concomitantly reduced. The filtration
fraction was markedly reduced during the dipyridamole
infusion, indicating that the decrease in GFR was
proportionally greater than the decrease in renal plasma
flow. No hemodynamic changes were observed during the
infusion of dipyridamole in the sodium-loaded dogs. These
effects are similar in many ways to the effects of the
intrarenal infusion of adenosine, but the decrease in the
arterial blood pressure, an effect not seen during the
infusion of adenosine, makes the comparison somewhat
complicated.

To compare the effects of dipyridamole to the effects
of exogenous adenosine more directly, and to determine what
contribution the decrease in arterial pressure had on the
decrease in GFR, we infused dipyridamole into another
series of sodium-depleted animals in which renal perfusion

pressure was held constant. When renal perfusion pressure

63

was held constant at 100 mm Hg, dipyridamole infusion still
produced a marked decrease in GFR (-38%), indicating that
the effect of dipyridamole is, for the_most part,
independent of the decrease in arterial pressure. It
should be noted here that, whereas mean arterial pressure
fell during the dipyridamole infusion and was a major
contributing factor to the decrease in RBF, systemic
pressure did not fall below the autoregulatory range of the
kidney and should not have been expected to result in a
decrease in RBF. These findings suggest that dipyridamole,
or endogenous adenosine, may cause an impairment of
autoregulation in sodium-depleted animals. The mechanism
by which this interference might occur is unknown, and may
be of sufficient importance to merit further investigation.

The ability of theophylline to antagonize the actions
of adenosine is well documented. In the kidney,
theophylline has been shown to act as a competitive
inhibitor of adenosine's actions to produce a
vasoconstriction (Osswald 1975), decrease GFR, and inhibit
renin secretion (Spielman 1984). To ascertain whether the
decrease in GFR resulting from dipyridamole infusion in the
sodium-depleted dog was mediated by increased adenosine
levels, theophylline was infused before and after the

dipyridamole infusion. Theophylline was able to reverse,

64

as well as inhibit, the dipyridamole-induced decrease in
GFR, suggesting that increased endogenous adenosine was
responsible for the decrease in GFR. Because in these
studies an infusion of theOphylline alone did not affect
renal function this may be taken as evidence that adenosine
is not involved in the regulation of renal function under
normal circumstances. Adenosine may, indeed, be involved
in the regulation of renal function only in circumstances
where there is increased metabolism and energy utilization.
However, because theophylline does increase urine volume
and sodium excretion, and in other experiments has been
shown to increase renin release, in animals without other
interventions, this may suggest that adenosine is important
normally as a local regulator of cellular function.

In summary, we found that, in the anesthetized dog,
dipyridamole blocked the cellular uptake of adenosine,
increased the extracellular concentration of adenosine,
and decreased GFR in the sodium-depleted but not the
sodium-loaded dog, and the effect on GFR was blocked by the
adenosine antagonist, theophylline. Increasing endogenous
adenosine by inhibition of cellular uptake resulted in a
decrease in GFR in the sodium-depleted animal, a condition
known to enhance the actions of adenosine. Although the

intent of this investigation was to evaluate the renal

65

effects of increased endogenous adenosine through
pharmacological inhibition of its cellular uptake, the
marked effect of dipyridamole to decrease GFR in the
sodium-depleted animal raises important issues concerning
the clinical use of dipyridamole in situations where
extracellular fluid volume is reduced. These findings
indicate that, given an appropriate physiological or
pathophysiological situation (e.g., sodium-depletion),
increased concentrations of endogenous adenosine have the
capability of altering GFR. Whereas the present findings
are consistent with a role for intrarenally produced
adenosine in the regulation of renal function, it remains
to be determined whether endogenous adenosine increased by
circumstances other than pharmacological ( physiological or
pathophysiological) results in similar changes in renal

function.

mmammmmmmmud

WW
(reproduced with permission, Springer Verlag; Kidney Int.
30:656, 1986)

B.1. Intrgdnsiign

Maleic acid administration produces a generalized
tubular transport defect that results in the increased
excretion of sodium, potassium, phosphate, glucose, and

amino acids, in addition to increasing urinary flow rate,

66

and thereby closely resembles the Fanconi syndrome (Al-
Bander et a1 1982; Berliner et al 1950; Brewer et al 1983;
Harrison and Harrison 1954; Kramer and Gonick 1973; Gmaj et
a1 1973). Several studies have demonstrated that,
associated with the defect in reabsorption, glomerular
filtration rate (GFR) is markedly reduced following maleic
acid administration (Al-Bender et a1 1982; Berliner et al
1950; Brewer et al 1983; Gmaj et al 1973; Gougoux et al
1976; Gunther et a1 1979; Maesaka and McCaffery 1980;
Osswald et a1 1980). Relatively little is known about the
mechanism of the decrease in GFR.

Previous work has demonstrated that maleic acid lowers
renal cortical adenosine triphosphate (ATP) content
(Osswald et al 1980; Kramer and Gonick 1970; Szczepanska
and Angielski 1980). Because of the marked alteration of
nucleotide metabolism, it has been proposed (Osswald et al
1980) that the decrease in GFR is the result of increased
intrarenal concentrations of adenosine. As a means of
testing this hypothesis, we sought to determine, during the
intrarenal infusion of maleic acid: 1) if the decrease in
GFR is greater in the sodium-depleted than in the sodium-
loaded dog, thereby mimicking the effect of dietary sodium
intake on the renal response to exogenous adenosine

infusion (Osswald et a1 1978); 2) if there is a suppression

67

of renin release, which is also seen with an adenosine
infusion (Osswald et a1 1978; Tagawa and Vander 1970;
Spielman 1984); 3) if the decrease in GFR is blocked by the
adenosine antagonist, theophylline; and 4) if renal venous
and urinary adenosine concentrations are increased. Data
are presented indicating that the decrease in GFR produced

by maleic acid is mediated by adenosine.

8.2. Methods

Experiments were performed on two groups of mongrel
dogs of either sex, weighing between 10 and 20 kg. One
group of 10 dogs (Group I) was maintained on a low-sodium
diet (Hill's Prescription Diet, h/d) of less than 2 mEq/day
for seven to 10 days, and was given 100 mg furosemide,
i.v., on the first day of the dietary regimen. A second
group of 10 dogs (Group II) was fed the same low-sodium
food as Group I, which had been supplemented with 200 mEq
NaCl. This diet was maintained for 10 to 14 days, and the
animals received 15 mg deoxycorticosterone pivalate (Ciba
Pharmaceutical Co.) each day of the dietary regimen. All
dogs were allowed free access to water. Because the
transport defect produced by maleic acid occurs primarily
in the proximal tubule (Al-Bander et a1 1982; Brewer et a1
1983; Kramer and Gonick 1973; Gmaj et al 1973; Gunther et

al 1979), and lithium is handled by the proximal tubule of

68

the kidney (Hayslett and Kashgarian 1979), all dogs
received 300 mg lithium carbonate 24 hours before the
experiment. Any change in lithium excretion observed can
be used as an index of proximal tubule function.

The dogs were surgically prepared as previously
described (Arend et a1 1984). Briefly, the dogs were
anesthetized with sodium pentobarbital (30 mg/kg, i.v.) and
a tracheostomy was performed for mechanical ventilation
with a Harvard respiration pump. Catheters (PE 240) were
placed in both femoral arteries for monitoring systemic
blood pressure and obtaining systemic blood samples, and in
both femoral veins for systemic infusions and supplemental
anesthetic administration . A left flank incision was made
to expose the left renal artery and vein and the ureter.
Small curved needles were placed in the renal artery for
intrarenal infusions and in the renal vein for sampling of
the renal venous effluent. The renal artery needle was
kept patent by a constant infusion of saline at 1.38
ml/min. The ureter was cannulated with polyethylene
tubing. An electromagnetic flow probe (Zepeda Instruments)
was placed around the renal artery for measurement of renal
blood flow. Systemic blood pressure and renal blood flow
were recorded on a Hewlett Packard recorder. A 1% solution

of inulin (Difco Laboratories) was infused at 2 ml/min into

69

a femoral vein throughout the experiment to achieve a
constant plasma concentration of approximately 25 mg/dl. A
60 min equilibration period was allowed after completion of
the surgical procedure.

Two control clearance periods were performed in each
experiment, which consisted of a fifteen minute urine
collection and the collection of systemic arterial and
renal venous blood samples at the mid-point of each period.
Saline (0.9%) was used as the intrarenal infusate for the
control periods as well as for the vehicle in subsequent
infusions. The saline was then replaced by maleic acid at
a dose of 100 mg/kg body wt/hr (neutralized with NaOH).
Ten-minute urine collections and blood samples were
obtained at 60 and 80 min during the maleic acid infusion.
To complete the time course of the effects of maleic acid
and to serve as a time control for the theophylline study,
urine and blood samples were collected at 100 min of maleic
acid in five dogs from each group. In the remaining five
sodium-depleted and five sodium—loaded dogs, an intrarenal
infusion of theophylline (5 umol/min) was begun at 90 min
of maleic acid. A five min equilibration period was
allowed after which three five-minute clearance periods
were performed. In all experiments, fluid losses due to

the polyuric effects of maleic acid and theophylline were

70

replaced by the intravenous infusion of saline at a rate
double to the urine flow rate of the experimental kidney.
The replacement rate was twice that of urine output to
account for urine formation in both kidneys, since the
contralateral kidney is also affected by the maleic acid
and theophylline infusions.

For these studies, adenosine and theophylline were
purchased from Sigma Chemical Co. and maleic acid was
purchased from J.T. Baker Chemical Co.

Inulin concentrations in plasma and urine were
determined by the anthrone method, as described by Davidson
and Sackner (1963). Sodium, potassium, and lithium
concentrations were determined by flame photometry.
Adenosine concentrations in plasma and urine during the
control and maleic acid periods were determined by HPLC as
previously described (Methods, III.A.). Plasma renin
activity (PRA) was measured as angiotensin I by
radioimmunoassay as described by Romero and Strong (1977).
Renin release was calculated as the difference between
arterial PRA and renal venous PRA multiplied by the renal
plasma flow. No samples for measurement of adenosine
concentrations were taken during the theophylline infusion.

Glomerular filtration rate was calculated as the

clearance of inulin. Changes in the fractional excretion

71

of lithium were used as an indication of changes in
proximal reabsorption.

Statistical significance was determined using two-way
analysis of variance, with Tukey's w-test for multiple
comparisons, and paired and unpaired t-tests where

appropriate.

8.3. Results

The intrarenal infusion of maleic acid resulted in a
decrease in GFR at 60, 80, and 100 min in the sodium-
depleted and the sodium-loaded animals (Table 3). GFR was
decreased significantly by 60 min of maleic acid and
continued to decrease for the remainder of the infusion in
the sodium-depleted group, however, in the sodium-loaded
group GFR was not significantly decreased until 100 min.
The infusion of theophylline (Table 4, Figures 6 and 7)
inhibited the further decline in GFR seen at 100 min in the
time course study; GFR increased to values that were not
significantly different from control in both the sodium-
depleted and sodium-loaded animals.

The response of RBF to maleic acid in the sodium-
depleted group is shown in a representative tracing in
Figure 6. There was an immediate increase in RBF at the
onset of the infusion which waned after approximately 60

min and fell below the control level for the remainder of

72

3:55 o. 12.2.88 2.... V a ..

55888 .3285... .m... woos—“Emu. 55:35 .58. .w.>~. Estes... sates...

 

...... ”8:32.. .58..." 53:. ...<2 ”3:: p83 Eco. in: ”2.2 :c..E=.. 5.50:5.» .KLC no... 323.53.? .23. H 5.5:. Ba «2...; .

 

 

 

 

...Nd H Show 2... H no... ...... H 3.: ... H .N. 3 H 3.. ..v.n H ...: 5:. 8.
:8... H 2...? a... H an. co... H .3... ... H .N. - H ... ...... H ..w. £5 8
...~.v H 2.3 R... H A.... .8... H an... ..N H mm. MN H .N. ...n H o.- £5 8
Bus 30...:
3... H no... 8.: H ca... 3.: H mm... . H c: n. H .... n... H aid .8369
a u 2. 822-533 a 955
3...». H 3.3 m... H 5.... 13.: H :2. ..c H 0.. cm H 9.. ......~ H 0.... 5:. cc.
...~.n H 8.3 m... H 9.... ax... H 2.: .... H 3. 2 H 3.. ..n.~ H 0.! ...... .3.
......N H ~v.~. B... H. 3.: ‘3... H mm... .... H .... .n H No. .56 H 0.3 5:. 3
Bus 03.5.2
2.: H 3... cc... H as... 3... H mm... v. H ..N. c. H ...: v... H «.2 3:55
.v. n 2. co.u...o.TE=.vcm H. 3:33
«o a: 5:. ...—5.... =.E\.E 2.8..
«2...... ~.>~. ...... ..<2 "...: «LO
.mmoo

popcoalegoom can cocoamoolegoom 5 ofios 0.5.3.: «o uooumo

mo onusoo we: .m manna.

2555 o. ease—coo 2.... V k ..
. 92.... c. v... 25.... u... 0:. 2.6.3.3212 .23 H :38 o... 32.; .

 

 

 

 

..~.n H 2.5.. 2.... H .... .. ...... H a... a H o... v. H e: ~.~ H ....N =.... 2
.3... H 3.2 c... .. H K. .. 1.... H a... o H N: .. H 3.. ~.~ H ..v~ ...E ...
12.... H :3... ...... H as .. 2.... H a... .... H .... .. H .n. n.~ H 0.2 a.... n
o:.........oo...=..oa 6.9.3..
......m H 3:... ..H 2. .. 3... H R... .... H .... v. H ~n. ..w. H ~...~ ...... ....
......n H 2...... ..H 3... ...... H mm... ..o H .N. .. H a... n.m H 93 ...... ....
3 Son 0.23..
a: vx_H:3 EéHnec 8¢H.2. oHnE SHo: anxfi .2ER.
a u 2.3.8.1568 a 22o
1...... H :61. ...... H on... ...... H g... .... H ...: o. H ..v. ad H m6. =.... n.
....N... H 2.3. 2... H 2.... ...... H R... m H n... ... H .... fi~ H N. a. ...... ...
.8... .+ ....cr. 2... H ...... 2.... H mm... a H v... ... H ..v. .... H ... ... ...... n
o:.........cu..<...oa 6.0.5..
1.3... H ......v I... H on... 2.... H ..N... o H o... o. H .2. .... . H ~.v. :.E ....
....m... H 32$ 2... H ...... 3.... H 2... a H .... X H on. ... n H W... ...E ....
v.2. 30.52
...... H 3... ...... H mm... 3.... H ..m... n H .... n. H n... .... H a.m~ .23....
.m n 2. ..u.o...o_TE=.—.om n. sac...
.8 u: .5: £35.: 5:5... 3...“...
Eu... :3. ...... ..<2 ...... ......

 

.mmoo ooomoalssfioom can oouoamoolszfioom
mo sowuocsm Hanan a. momcmso ooososfilowos ofioame as» so osuuamnmoosu mo uoouwm .v wanna

RBF

mllmin

GFR

ml/min

300

74

AA

 

SODIUM-DEPLETED '1

1

 

 

 

 

      

 

 

 

 

 

 

 

 

 

 

 

 

 

   

200 L- ; .

P _+__/ \+ + ' +_,+_+_ 1

O '— .

O _ _.
THEOPHYLLINE

4° F" MALEIC ACID ‘1

P H

30 - d

P' «1

20 - ‘ . 4

10 '- —1
o ..... 1+;- L L .31.; .5223; if 3.}.
CONTROL 20 40 95 100105

TIME

Figure 6. Effect of theophylline on the maleic acid-induced

changes in glomerular filtration rate (GFR) and
renal blood flow (RBF) in five sodium-depleted
dogs. Each value represents the mean _+_ SEM for
the group. Asterisk indicates P<0.05, compared
to control.

75

 

 

 

 

 

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

300 ,1;
.. SODIUM-LOADED ..
200 >- ..
RBF . 4
:- ¥ —.- —
ml/mm —.—"’. _§ +-"""" T
P d
0L -
THEOPHYLLINE
40 " MALEIC ACID ‘
30 '- ..
v- . _‘
GFR .
20 r- 1212‘: :‘3. .4
"‘"m'" 52222222222222
’ “"333332323? ‘
.0 - -.
. Eéiiééiéiéiéii ..
31333333335321 : :: ..... ::::
O n 1 ‘i'iiii'iiii . I: -:-::-: $7133: -:-:1-:-
CONTROL 20 40 80 95 100105
TIME
mm

Figure 7. Effect of theophylline on the maleic acid-induced
changes in glomerular filtration rate (GFR) and
renal blood flow (RBF) in five sodium-loaded
dogs. Each value represents the mean 3; SEM for
the group. Asterisk indicates P<0.05, compared
to control.

76

the infusion. With the infusion of theophylline, RBF
returned to the control level. In the sodium-loaded group,
shown in Figure 7, RBF increased only slightly after the
onset of the maleic acid infusion and was not different
than the control level during the remainder of the
infusion. Theophylline infusion in the sodium-loaded group
resulted in a small increase in RBF. Mean arterial
pressure (Table 3) decreased during maleic acid infusion in
the sodium-depleted group, but increased significantly in
the sodium-loaded group. Theophylline did not
significantly change mean arterial pressure (Table 4) in
either group.

The fractional excretion of sodium was significantly
greater in the sodium-loaded animals than in the sodium-
depleted animals in the control period and was
significantly increased to approximately 30% in both
groups, with maleic acid (Table 3). The infusion of
theophylline resulted in a further increase in fractional
sodium excretion. The fractional excretion of lithium also
increased in both groups with maleic acid. The average
plasma lithium concentration in the sodium-depleted group
(N =10) was 0.56 1 0.11 mEq/liter. There was a significant
increase in the fractional lithium excretion in this group

from a control value of 22.7 1 6.4% to 64.9 1 12.4%

77

(P<0.05) after 100 min of maleate. In many of the sodium-
1oaded dogs, the dose of 300 mg lithium carbonate was not
sufficient to produce a measurable plasma lithium
concentration. Therefore, for statistical comparisons,
lithium concentrations for all sodium-loaded dogs (time
course and theophylline) were pooled. In six sodium-loaded
dogs, the control plasma lithium concentration was 0.11 1
0.01 mEq/liter, and the fractional excretion of lithium
increased from 30.9 1 7.7% in control to 53.3 1 10.1%
(P<0.05) after 60 min and to 69.1 1 10.4% (P<0.05) after 80
min of maleic acid. At 100 min of maleic acid, as well as
during the infusion of theophylline, the urine lithium
concentration was undetectable in the sodium-loaded group.
During theophylline, in the sodium-depleted group,
fractional lithium excretion increased from 79.1 1 6.9% at
80 min of maleate to 84.6 1 9.9% (NS) after 25 min of
theophylline.

The effect of maleic acid on renin release is shown in
Figure 8. The data shown include all 10 of the sodium-
depleted animals during the maleic acid infusion. During
the control period, renin release averaged 608 1 253 ng
AI/min. Renin release decreased significantly (P<0.05) to

11.9 1 7.9% of control during maleic acid and returned to

78

60.8 1 9.9% of control (NS) during the theophylline
infusion (N = 5).

As shown in Figure 9, the arterial concentrations of
adenosine were not altered by the infusion of maleic acid
in either the sodium-depleted or the sodium-loaded animals.
However, the renal venous concentration of adenosine
increased significantly during maleic acid infusion, from
0.05 1 0.01 uM to 0.33 1 0.06 uM in the sodium-depleted
animals and from 0.13 1 0.02 uM to 0.51 1 0.10 uM in the
sodium-loaded animals. In addition, the urinary excretion
of adenosine increased significantly, from 0.96 1 0.23
umol/min to 7.22 1 2.26 umol/min in the sodium-depleted
animals and from l.2110.17 umol/min to 18.46 1 4.81

umol/min in the sodium-loaded animals.

3.4. Discussion

These studies demonstrate that the effects of maleic
acid on renal function are dependent on the state of sodium
balance of the animal and that the hemodynamic changes can
be reversed by the adenosine receptor antagonist,
theOphylline. In addition, these changes in renal function
were associated with an increase in the concentrations of
endogenous adenosine.

Several studies have demonstrated that the intravenous

or intrarenal administration of maleic acid results in a

79

 

100

I
O

RENIN 60 I I

RELEASE
% OF CONTROL

 

Y

40-

V

 

 

0 l 1 1
CONTROL MALEIC ACID MALEIC ACID/
THEOPHYLLINE

 

Figure 8. Effect of theophylline on the maleic acid-induced
decrease in renin release presented as percent of
control. Value during maleic acid represents the
mean 1 SEM for 10 sodium-depleted dogs, value
during maleic acid and theophylline represents
the mean 1 SEM for five sodium-depleted dogs.
Asterisk indicates P<0.05, compared to control.

ARTERIAL
ADENOSINE
g M

VENOUS
ADENOSNE
uM

EXCRE TED
ADENOSK
u "WWII

renal venous adenosine,

06

0.2 ’

00

06

02'

00

100

80

60

40*

20

00

80

 

 

I

 

 

 

 

 

 

 

 

/

4
1
1
4
4
4
¢
.4
1

 

 

W

W
In.”

MALEIC AGO

ARTERIAL
ADENOSIK
uM

VENOUS
AOENOSIW
p M

EXCRETED
ADENOSHE
.1 null!!!

06

04

0.2 I'

00

06

O2

00

160

80>

00

 

 

 

 

 

 

 

 

 

 

 

 

 

CWTROL
SWIM-LOAN!)
(n - ‘0)

MALEIC ACO

Figure 9. Effect of maleic acid on arterial adenosine,
and the urinary excretion

of adenosine in six sodium-depleted and ten
Values are the mean + SEM

sodium-loaded dogs.
for each group.
compared to control.

Asterisk indicates P<0 . 05 ,

81

profound decrease in1proximal and distal, tubular
reabsorption (Al-Bander et al 1982; Berliner et a1 1950;
Brewer et a1 1983; Harrison and Harrison 1954; Kramer and
Gonick 1973; Gmaj et al 1973). This defect in tubular
transport is believed to be due to the ability of maleate
to affect mitochondrial metabolic activity (Szczepanska and
Angielski 1980; Angielski and Rogulski 1975; Rogulski and
Pacanis 1978). The administration of maleic acid has also
been shown to be associated with a decrease in cortical ATP
concentrations(0sswald et a1 1980; Kramer and Gonick 1970;
Szczepanska and Angielski 1980). In addition, a decrease
in glomerular filtration rate (GFR) has been shown to occur
during the infusion of maleic acid (Al-Bander et al 1982;
Berliner et al 1950; Brewer et al 1983; Gmaj et al 1973:
Gougoux et al 1976; Gunther et a1 1979; Maesaka and
McCaffery 1980; Osswald et al 1980). While the mechanism
of action of maleic acid to produce defects in tubular
transport has been studied (Angielski and Rogulski 1975;
Rogulski and Pacanis 1978; Angielski 1963), the action of
maleic acid to cause a decrease in GFR has not been
extensively examined, though a recent micropuncture study
by Maesaka and McCaffery (1980) suggested that the measured
decline in GFR is due to a backleak of inulin in a distal

portion of the nephron. The ability of theophylline to

82

reverse the maleic acid-induced decrease in GFR in the
present study, as well as in a preliminary study by Osswald
(1980), indicates that in addition to, or alternative to a
backleak of inulin, the decrease in GFR may be due to an
adenosine-mediated mechanism.

The renal actions of exogenous adenosine have been
shown to be dependent on the state of sodium balance of the
animal. Osswald et a1 (1978) have shown that the effects
of adenosine on GFR, renal blood flow (RBF), and renin
release are enhanced by sodium restriction and attenuated
by sodium loading. In addition, the renal effects of
elevating endogenous adenosine have also been shown to have
a sodium-dependency (Results III.A.). Therefore, as a test
of the hypothesis that the decrease in GFR due to maleic
acid administration is the result of an increase in
endogenous adenosine, the animals in the present study were
either sodium-depleted or sodium-loaded. The magnitude of
the decrease in GFR was greater in the sodium-depleted than
in the sodium-loaded dogs. ‘In addition to the decrease in
GFR, the change in RBF was more pronounced in the sodium-
depleted animals, and there was a decrease in renin
release, both of which are known to occur with the
exogenous administration of adenosine (Osswald et a1 1978;

Tagawa and Vander 1970) as well as with increasing

83

endogenous adenosine concentrations (Results, III.A.).
These results were true for each time point during the
infusion of maleic acid and support the hypothesis that
intrarenal adenosine mediates these changes. The
possibility exists, however, that the metabolic effects of
maleic acid are altered by the presence of a high sodium
diet or volume expansion and that this alteration in the
cellular effect of maleic acid is responsible for the
difference in the functional responses of the two groups.
The fractional excretion of sodium at 100 min was
approximately 30% in both the sodium-depleted and the
sodium-loaded dogs, and was actually more pronounced in the
sodium-loaded animals at all time points, indicating that
the difference in sodium balance apparently was not the
cause of the difference in functional response between the
two groups.

The animals in this study were given lithium carbonate
prior to the experiments with the intention of using
fractional lithium excretion as an index of proximal
reabsorption, since it has been reported that lithium is
handled by the kidney at this site (Hayslett and Kashgarian
1979). Previous studies attempting to localize the tubular
action of maleic acid have shown that much of the transport

dysfunction occurs in the proximal tubule (Al-Bander et a1

84

1982; Brewer et a1 1983; Kramer and Gonick 1973; Gmaj et a1
1973; Gunther et a1 1979), with some reports of dysfunction
occurring distally (Al-Bander et a1 1982; Bergeron et a1
1976). The increased fractional excretion of lithium in
the present study served as evidence that the transport
defect had occurred. An interesting finding apart from the
increase in fractional lithium excretion was that, given
the same dose in both groups of animals, there was a
consistently lower plasma lithium concentration and higher
control fractional lithium excretion in the sodium-loaded
groups. This may reflect an overall decrease in proximal
reabsorption due to the volume-expanded state of these
animals.

Concomitant with the functional changes resulting from
the infusion of maleic acid, the intrarenal concentrations
of adenosine were increased as measured by an increase in
the level of adenosine in the renal venous and urinary
effluents. In view of work by other groups (Osswald et a1
1980; Kramer and Gonick 1970; Szczepanska and Angielski
1980) in which the infusion of maleic acid is shown to be
associated with a decrease in cortical ATP concentrations,
it was not surprising to find an increase in the
concentrations of adenosine in the present study. Indeed,

it was reported in a study by Osswald et a1 (1980) that

85

tissue adenosine concentration is increased following
maleic acid administration in the rat. However, given the
complexity of adenosine metabolism, its varied cellular
sources, and the possibility of intracellular binding
sites, the total tissue level of adenosine is not a
reliable index of extracellular adenosine, which is the
relevant pool of adenosine when proposing that adenosine
may be involved in the regulation of renal function. In a
recent report from this laboratory (Thompson et al 1985),
it was shown that, despite similar control arterial and
venous adenosine concentrations, very little of the venous
adenosine was derived from the arterial supply, therefore,
adenosine measured in the renal vein is primarily of renal
origin. Likewise, a small portion of the adenosine
measured in the urine is from renal production, the
majority of adenosine in the urine being from filtration of
the arterial plasma and incomplete reabsorption in the
tubules. Therefore, in the present study, adenosine was
measured in the renal vein and in the urine to provide an
indication of changes occurring in extracellular adenosine
levels. The level of adenosine in the kidney during maleic
acid infusion was increased to approximately the same
magnitude in both groups of dogs and, therefore, the

smaller changes in renal function that occurred in the

86

sodium-loaded animals were not due to less adenosine being
released in these dogs.

To further test the hypothesis that the increase in
intrarenal adenosine mediates the decrease in GFR, RBF, and
renin release seen with maleic acid, the adenosine receptor
antagonist, theophylline, was used. Theophylline, given at
a dose that is known to have no influence on cyclic
nucleotide phosphodiesterase activity (Spielman 1984),
reversed the decreases in GFR and RBF in both the sodium-
depleted and the sodium-loaded dogs, and reversed the
decrease in renin release in the sodium-depleted dogs. The
administration of theOphylline did not, however, attenuate
the increase in the fractional excretion of sodium,
suggesting that the defect in tubular transport caused by
maleic acid is an adenosine-independent phenomenon while
the hemodynamic and renin release effects may be mediated
by the increase in adenosine levels. The inability of
theophylline to completely reverse the decrease in GFR
could be due to other factors resulting from the
administration of maleic acid. As suggested above, maleic
acid may cause a back leak of inulin across the damaged
tubular epithelium. In addition, maleic acid has been
shown previously to produce a bicarbonaturia (Berliner et

a1 1950; Gmaj et al 1973; Gougoux et al 1976) and may lead

87

to acidosis, which can adversely affect GFR. Despite these
possible complications, theophylline largely returns GFR to
control, suggesting a major contribution by adenosine to
the decrease in GFR.

The decrease 1H1 GFR following maleic acid
administration may serve to limit the polyuria associated
with this nephrotoxic insult. It is tempting to speculate
that the increased delivery of fluid and solute from the
damaged proximal epithelia following maleic acid initiates
an adenosine-mediated tubuloglomerular feedback response to
reduce the filtered load in an attempt to limit the excess
fluid and solute loss. Previous data have suggested a role
for adenosine in mediation of the feedback response. Most
relevant to the present study are the results of Schnermann
and coworkers (1977) which showed that the feedback
response to an increase in tubular flow rate is abolished
by theophylline. Coupled with the present results, this is
consistent with adenosine acting as a feedback mediator in
certain circumstances.

The results of this study support the hypothesis that
adenosine mediates the renal hemodynamic and renin release
effects of maleic acid. However, while maleic acid is used
as an experimental model of the Fanconi syndrome, this

study does not necessarily imply that any changes in renal

88

function that occur in clinically relevant cases of the
Fanconi syndrome are mediated by adenosine. What can be
inferred from these results is that endogenous adenosine is
capable of altering renal function in some pathophysiologic
or nephrotoxic situations. It remains to be determined
what role adenosine plays in the normal regulation of renal
function.

C. A role for intrarenal adenosine in the renal
hemodynamic response to contrast media.

 

 

C.1. Introduction

 

The effects of hypertonic contrast media
administration on renal function have been well documented
(Ansari and Baldwin 1976; Caldicott et al 1970; Chou et al
1971; Gerber et a1 1982; Katzberg et al 1983; Larson et a1
1983; Reed et al 1983; Talner and Davidson 1968). The
injection of contrast media into the circulation, whether
intrarenally or intravenously, results in a pronounced,
though transient, decrease in renal blood flow and
glomerular filtration rate (Caldicott et al 1970; Chou et
a1 1971; Katzberg et al 1983; Larson et al 1983; Reed et al
1983; Talner and Davidson 1968; Katzberg et al 1977).
Under experimental conditions it has been shown that the
hemodynamic changes exhibit a sodium-dependency, that is,

the effects of contrast media on renal function are more

89

pronounced in animals maintained on a low-sodium diet than
in those maintained on a high-sodium diet (Caldicott et al
1970; Larson et al 1983; Katzberg et al 1977; McDonald et
al 1969). Several attempts have been made to elucidate the
mechanism by which contrast solutions produce these changes
1J1 renal function. Alpha-adrenoceptor antagonists
(Caldicott et a1 1970) and angiotensin II blockers (Larson
et a1 1983) have been used in attempts at altering the
hemodynamic effects of contrast media but did not
completely prevent the vasoconstrictive response. However,
in a more recent study, decreasing the availability of
calcium by the use of the calcium influx blockers,
verapamil and diltiazem, or by chelating extracellular
calcium with EGTA, was successful in attenuating the
decreases in renal blood flow (RBF) and glomerular
filtration rate (GFR) due to contrast media administration
in dogs (Bakris and Burnett 1985). These results suggest
that the mediator of contrast-induced changes in renal
hemodynamics acts through an increase in intracellular
calcium, presumably in vascular smooth muscle cells.

In a recent report from this laboratory, the decrease
in RBF and GFR resulting from the intrarenal administration
of adenosine could be abolished by verapamil (Arend et al

1984), suggesting that the renal hemodynamic effects of

9O

adenosine are mediated by increased calcium influx. In
addition, similar to the hemodynamic actions of contrast
media, the effects of adenosine on RBF and GFR are enhanced
by prior sodium depletion of the animal and attenuated by
sodium loading (Osswald et al 1978). Therefore, the
present study was undertaken to test the hypothesis that
the reduction in renal blood flow and GFR following the
administration of hypertonic contrast media is mediated by
an increase in endogenous adenosine levels. This
hypothesis was tested by employing the nucleoside uptake
blocker, dipyridamole, and the adenosine receptor
antagonist, theophylline, as well as by measurements of

arterial, renal venous, and urinary adenosine.

C.2. Methods

The experiments were performed on mongrel dogs of
either sex weighing 10-20 kg. All animals were maintained
on a low-sodium diet (Hill's Prescription Diet, h/d) for a
period of 10-14 days and were given an injection of Lasix
(100 mg, i.v.) on the first day of the dietary regimen. 0n
the day prior to the experiment, each dog was fasted and
allowed no water.

The animals were anesthetized with sodium
pentobarbital (30 mg/kg, i.v.) and a tracheostomy was

performed for mechanical ventilation with a Harvard

91

respirator. The dogs were then surgically prepared as
described previously (Arend et al 1984). Briefly,
catheters (PE 240) were introduced into both femoral veins
for systemic infusions and supplementalanesthetio
administration, into one femoral artery for the collection
of systemic arterial blood samples, and in one carotid
artery for monitoring of systemic blood pressure. A left
flank incision was made to expose the ureter and the left
renal artery and vein. An electromagnetic flow probe
(Zepeda Instruments) was placed around the renal artery,
and renal blood flow, as well as systemic blood pressure,
was recorded on a Hewlett-Packard recorder. Small curved
needles were placed in the renal artery for intrarenal
infusions, and in the renal vein for sampling of the renal
venous effluent. The arterial needle was kept patent by a
constant infusion of 0.9% saline at 1.38 ml/min. A
catheter was placed in the ureter and positioned near the
pelvis of the kidney for collection of urine samples. A 1%
solution of inulin was infused at 2 ml/min throughout the
experiments into a femoral vein.

All solutions were prepared on the day of the
experiment with 0.9% NaCl as the vehicle. The contrast
agent utilized in this study was Renografin-76

(meglumine/sodium diatrizoate-76%, E.R. Squibb). Inulin,

92

adenosine, and theophylline were obtained from Sigma.
Dipyridamole was a gift from Boehringer-Ingelheim.

Inulin concentrations in plasma and urine were
determined using the anthrone method of Davidson and
Sackner (1963). Sodium, potassium, and lithium
concentrations in plasma and urine were determined by flame
photometry. Samples were assayed for adenosine
concentration by HPLC using the method described previously
(Methods, III.A.). Glomerular filtration was calculated as
the extraction of inulin across the kidney. The data were
statistically analyzed by a two-way analysis of variance,
using the Student-Newman-Kuels test for multiple
comparisons, or by paired t-test where appropriate.
Results are reported as mean 1 SEM.

Group I: The effect of intrarenal injection of
contrast media on renal function (n=6). Two control
clearance periods were performed to determine initial rates
of glomerular filtration, renal blood flow, and other
indices of renal function. These clearance periods
consisted of two contiguous 15 min urine collections, with
two systemic arterial and renal venous blood samples taken
during each 15 min period at 5 and 10 min. Following these
control periods, the contrast media (4 ml) was injected

directly into the renal artery over a period of 15-30 sec.

93

Immediately following this injection, three 5 min clearance
periods were performed, with blood samples being drawn at
the midpoint of each period. A second injection of
contrast media and collection of samples was repeated
following approximately 30 min of recovery. Arterial and
renal venous blood and urine samples for adenosine
measurements were obtained during each period.

Group II: Effects of theophylline on the contrast
media-induced changes in renal function (n=5). Control and
contrast media clearance periods were performed as
described for Group I before and during theophylline.
Theophylline was infused at 5 umol/min for 30 min prior to
and during the second set of control and contrast media
clearance periods.

Group III: Effects of dipyridamole on the contrast
media-induced changes in renal function (n=6). Two control
clearance periods and one injection of contrast media,
followed immediately by three 5-min periods were performed
as described for Group I. The intrarenal infusion of
saline was then replaced by dipyridamole (24 ug/kg/min) and
after a 30 min equilibration period, two control clearance
periods and one injection of contrast media were performed

as before.

94

C.3. Results

Group I: Effect of contrast media injection on renal
function. The rapid (30 sec) injection of 4 ml Renografin—
76 directly into the renal artery resulted in an immediate
increase in renal blood flow (RBF), resulting from the
injection of a viscous solution, which lasted for the
duration of the injection, followed by a 17% decrease in
RBF, from 97 1 14 ml/min to 81 1 13 ml/min that lasted
approximately 5 min. A representative RBF tracing is shown
in Figure 10 (left half). An electronically—averaged RBF
tracing was used for computing changes in RBF. Blood flow
immediately prior to the contrast media injection was taken
as the control and blood flow at the nadir of the response
after the injection, was taken as the maximum decrease and
used for calculation of percent decrease. In the first five
minutes after contrast media injection, GFR decreased by
31% from 25 1 4 ml/min to 17 1 4 ml/min (P<0.05, Figure 11)
and returned to control by the second clearance period
following the injection. Data from the two control periods
were pooled and those from the second and third periods
following contrast injection were pooled as recovery. The
responses to the second injection of contrast media were no
different than the responses to the first injection and

were also pooled. Intrarenal injection of contrast media

95

 

[— DIPYRIDAMOLE 4.—

250 P
200 '-
150

RENAL
BLOOD
FLOW

(ml/min) so

0

 

 

h T 1 T

Figure 10. Representative renal blood flow tracing of the
responses to contrast media injection (small
arrows) in the control state, and in the
presence of dipyridamole.

96

20 1'

7% cm '
I m

0

Xflmmahmnwrmhflm
I
N
O

-40 4

 

 

 

CON THEO DP

Figure 11. Effect of contrast media on glomerular
filtration rate (closed bars) and renal blood
flow (hatched bars) in the control state (CON),
and during the infusion of theophylline (THEO)
or dipyridamole (DP). Results are expressed as
percent decrease from pre-contrast levels.

*, significantly different than response to contrast media

in the absence of theophylline or dipyridamole.

97

on mean arterial pressure (MAP) or filtration fraction (FF)
(Table 5). Renal vascular resistance (RVR) was increased
slightly. Urine flow increased by S-fold following
contrast injection, and the excretion and. fractional
excretion of sodium increased significantly. Systemic
arterial and renal venous adenosine concentrations were not
increased following contrast media injection (136 1 49 vs
162 1 86 nM, NS and 237 1 125 vs 144 1 55 nM, NS,
respectively).

The urinary concentration of adenosine was decreased
by contrast media (855 1 228 vs 357 1 94 nM, NS), however,
when factored by the large increase in urine volume and the
decreased GFR, excreted adenosine (388 1 80 vs 830 1 23
umol/min, P<0.05), and fractional excretion of adenosine
(15.9 1 4.1 vs 49.6 1 12.9%, P<0.05), were significantly
increased.

Group II: Effect of theophylline on the contrast
media-induced changes in renal function. As in Group I,
contrast media injection in the absence of theophylline
resulted in a significant decrease in RBF (138 1 11 vs 114
1 9 ml/min, P<0.05) and GFR (29 1 2 vs 16 1 1 ml/min,
P<0.05), with little or no change in filtration fraction,
mean arterial pressure, and renal vascular resistance

(Table 5). The fractional excretion of sodium was

0:...agdougu uaogu.) auouucoo on soundiou :93: no.6 v a .‘
ouccgo .oLucoo ou ovumalou cog: unease mus—cans w:.c v a .+
oa~o> «ouucoo o>.uoodaou on ouumdlou cog: mo.c v a .c

.co..uuuxo ls.ooa asco.uuauu .mzum “lagoon tauauuuu .ozu “coon rod. scan: .m: «oocsua.auu unusuaa> .acou
.x>= nuuzmmuua .m.uouua anal .m<x “co.uuauu :o.uauu..u .mh "nouns co.uauu..u umuauolodu .zmo m:o_u coo—n .mcou .mns

.....m: .2me .2233. arouse; ..HHZ ..86H3... a H: :2“: 2....- 88:8
few... 1 H: 2.99;... firms.— ..Hag 8.8.8.8 ~ H2 SH»: 3233.23
82...... 3&8: @2938 Erna... Aug: .8432. 1” H2 EH8. 3...... 2.3.28
.88... «HHS Séufié 2....+.8.. 8H2. Siva... n H2 2H2. .828
.Ao I c. a.eul unauucou on aoacoaaou .acou :o o.osoo«uaa.o no uuouum u.- cacao
Ru

9 I I II I I I I
3.2.... I- .1333.» 2.32.. «$2 8.9:... 3+2 +232 2...... 2.828
.ZHH... I- .36.}... 2.8.3.... HHS. 86$“... m: a .32 831.18.;
1...“... I- Stone... 36$... 35 86mg... ..H... 8 H3. 3...... 822.8
to“: I- Sang... 2.8.8.. «#2 SAWS... «H3 2H5: .828

.An I a. a.ool uoouucou o. acacodaou Anson co ocuduasaoogu no uua.um “—H cacao

1.93 .23.: «26?... 3.8.3... HHS. 3.:H2... in: 2H... 33- .2228
in. .. 3H8 2.8.“? .o 2 ...HS. . «we: Sans... “H3 . 1H3 3.38
oafl I CV GOdUUGfiu dflflflh BO fldfifl' UIIHUEOU HO HOUHHH nu QSOHQ

S. A: .3281. A511. ASIDIxuzl. 25'. €2.21. €325

fl .
2.: ..zm .5 a3. is. .... . Eu .2.

.a.aal uoouueou ou consensus
.ecuu oz. co o.osoo.uaa.v can o:..—agaoogu no auuouuo use can .couuocau deans co cues. unauucoo mo ouuuuum .n «dash

99

increased significantly (l.2 1 0.7 vs 4.6 1 1.8%, P<0.05)
by contrast media. The infusion of theophylline alone
resulted in a significant increase in urine flow rate and
the excretion of sodium (Table 5), but did not alter RBF or
GFR. In the presence of theophylline, there was no
significant change in RBF (131 1 9 vs 123 1 10 ml/min, NS)
or GFR (27 1 2 vs 23 1 2 ml/min, NS) with the injection of
contrast media (Figure 11). The urine flow rate and
excretion of sodium were significantly increased by
contrast media, despite the presence of theophylline.

Group III: Effect of dipyridamole on the contrast
media-induced changes in renal function. Contrast media
injection resulted in a decrease in RBF of 12% (119 1 12 vs
104 1 14 ml/min, NS), and in GFR of 30% (23 1 3 vs 16 1 3
ml/min, P<0.05). Filtration fraction also decreased
significantly in this group and vascular resistance, urine
flow, and the excretion of sodium increased (Table 5).
During the intrarenal infusion of dipyridamole, the
injection of contrast media resulted in a more pronounced
decrease in RBF and GFR (Figure 11). RBF decreased by 25 1
4% (118 1 14 vs 88 1 13 ml/min, P<0.05; Figure 10, right
half) and GFR decreased 44 1 7% (19 1 2 vs 11 1 2 ml/min,
P<0.05); these decreases are greater than those seen in the

absence of dipyridamole. The duration of the vasocon-

100

striction was not lengthened by the infusion‘of
dipyridamole; RBF returned to the control value by
approximately 5 min after the injection. As in Group I,
MAP and filtration fraction were essentially unchanged,
while urine flow and the excretion of sodium increased

after contrast media injection.

C.4. Discussion

 

The administration of hypertonic contrast solutions
has long been recognized to produce profound changes in
renal hemodynamics and in some cases to lead to the
initiation of acute renal failure (Ansari and Baldwin 1976;
McDonald et a1 1969; Byrd and Sherman 1979). The systemic
administration of contrast media, as well as injections
directly into the kidney, results in a transient decrease
in renal blood flow (RBF) and glomerular filtration rate
(GFR) (Caldicott et a1 1970; Chou et a1 1971; Katzberg et
a1 1983; Larson et al 1983; Reed et al 1983; Talner and
Davidson 1968; Katzberg et a1 1977). While many attempts
have been made to elucidate the mechanism of these actions,
the results, for the most part, have been negative or
incomplete. It is known that the large increase in urine
output following contrast media administration is due to an
osmotic diuretic effect in the proximal tubule (Cattell et

al 1982), and it has been proposed that this increase in

101

tubular fluid flow is responsible for the reductions in RBF
and GFR, that is, a tubuloglomerular feedback signal is
being generated to reduce filtration (Read at al 1983). In
support of this idea, it has been demonstrated that the
hemodynamic changes occurring following intravenous
contrast administration are reduced in streptozotocin-
induced diabetic rats with chronic glycosuric osmotic
diuresis (Reed et a1 1983), glycosuria having been
associated with suppression of the feedback response
(Blantz et a1 1975). However, the mediator of this
feedback response has not been determined. Attempts to
attenuate the renal hemodynamic changes produced by
contrast agents with alpha adrenoreceptor antagonists
(Caldicott et al 1970) as well as angiotensin II receptor
blockers and converting enzyme inhibitors (Larson et a1
1983) have provided only partial protection, therefore,
renal nerves and the renin-angiotensin system do not appear
to play a solitary role in mediating the contrast-induced
changes. An important advance in our understanding of the
mechanism by which contrast media produce a decrease in RBF
and GFR is the finding that the calcium-influx blockers,
verapamil and diltiazem, reduce the changes in renal
hemodynamics resulting from contrast media injections in

anesthetized dogs (Bakris and Burnett 1985). These results

102

suggest that the mediator of the contrast-induced changes
acts by increasing intracellular calcium, possibly by
stimulating an increased influx of calcium. In addition,
it was shown earlier by Larson et a1 (1983) that the
hemodynamic changes following contrast media administration
exhibit a sodium-dependency, that is, the decreases in RBF
and GFR are greater in animals that were fed a low-sodium
diet and are less pronounced in animals fed a high-sodium
diet.

The present study was undertaken to test the
hypothesis that an increase in intrarenal adenosine
concentrations mediates the changes in renal hemodynamics
following contrast media administration. The
administration of exogenous adenosine is known to cause a
brief vasoconstriction when injected into the renal artery
of dogs (Arend et al 1984; Thurau 1964) and results in a
biphasic change in RBF and a sustained decrease in GFR when
infused intrarenally (Osswald et a1 1978; Tagawa and Vander
1970). As a result of these, and other, observations,
adenosine has been proposed as an intrinsic regulator of
renal function (Spielman and Thompson 1982) and has been
suggested as the mediator of renal function changes in some
pathophysiological conditions. For example, the

administration of theophylline, the adenosine receptor

103

antagonist, prior to occlusion of the renal artery
minimized the post-occlusive reduction in RBF (Osswald
1975). In addition, the pronounced decrease in GFR that
occurs following the administration of maleic acid can be
attenuated by theophylline (Results III.B.; Osswald et al
1982). In both of these studies an increase in intrarenal
adenosine concentrations was measured.

The renal actions of adenosine have been shown to be
sodium-dependent (Osswald et a1 1978). That is, sodium-
depletion results in larger adenosine-induced decreases in
RBF and GFR as compared with the responses in animals
maintained on a normal sodium diet, while those animals
maintained on a high sodium diet show smaller responses to
intrarenal adenosine. Additionally, the renal hemodynamic
effects of adenosine can be blocked by the administration
of verapamil (Arend et a1 1984). The similarities between
the renal actions of contrast media and those of adenosine
led to the hypothesis that adenosine mediates the
hemodynamic changes due to contrast media administration.

In the present study, intrarenal injections of
Renografin-76 produced changes in renal function such as
those that have been reported previously (Ansari and
Baldwin 1976; Caldicott et a1 1970: Chou et a1 1971; Gerber

et a1 1982; Katzberg et a1 1983: Larson et a1 1983; Read at

104

a1 1983; Talner and Davidson 1968; Katzberg et a1 1977;
McDonald et a1 1969). In the presence of the adenosine
receptor antagonist, theophylline, the changes in renal
blood flow and glomerular filtration rate were
substantially reduced. The concentration of theOphylline
used in this study is well below that which has been shown
to affect cyclic nucleotide phosphodiesterase activity
(Spielman 1984), and therefore, these differences in the
hemodynamic response can not be attributed to theophylline-
induced changes in cAMP levels. These results suggest that
adenosine may partially mediate the hemodynamic effects of
contrast media.

Further evidence to support this hypothesis stems from
the results that the changes in renal function were
potentiated when the contrast media injections were
performed in the presence of an infusion of dipyridamole.
In an earlier study, we demonstrated that an intrarenal
infusion of dipyridamole significantly increases the
intrarenal concentrations of adenosine by interfering with
the cellular uptake and degradation of adenosine (Results,
III.A.; Thompson at al 1985). Increasing endogenous
adenosine by blocking its cellular uptake resulted in
changes in renal hemodynamics and renin release similar to

that seen with an infusion of exogenous adenosine. The

105

renal actions of exogenous adenosine are also potentiated
by the presence of dipyridamole (Results, III.A.). The
enhancement by dipyridamole of the contrast-induced
decreases in RBF and GFR suggests that these changes are
due to an increase in endogenous adenosine concentrations
above that which occurred without the dipyridamole.
Following the intrarenal injection of contrast media,
there was an increase in the excretion of adenosine in the
urine. In a recent study on the handling of adenosine by
the kidney it was shown that 50% of filtered adenosine is
excreted in the urine (Thompson et a1 1985). Therefore, in
the present study, since there was no increase in arterial
adenosine concentrations and there was a decrease in GFR,
hence a decrease in filtered adenosine, it appears that the
increased excretion of adenosine was due to an increase in
extracellular adenosine in the kidney. The decrease in
adenosine concentration of the urine likely reflects the
large increase in urine volume that occurred with contrast
media injection. There was, however, no increase in
adenosine leaving the kidney via the renal venous blood.
This does not support the hypothesis that increased renal
adenosine mediates the contrast media-induced hemodynamic
changes. However, the changes in RBF and GFR that occurred

after injection of contrast media were of short duration,

106

lasting at the most 5 minutes, and it is therefore possible
that there was an increase in intrarenal adenosine but it
could not be detected in the renal vein due to the short
time course of hemodynamic changes, or due to the rapid
cellular uptake and degradation of adenosine by cells,
particularly red blood cells. Alternatively, the increase
in adenosine may have been small or localized and was for
this reason not detectable in the renal vein.

In summary, the renal hemodynamic effects of contrast
media can be attenuated by an adenosine receptor
antagonist, and enhanced by an adenosine uptake inhibitor,
and the urinary excretion of adenosine is increased by
contrast media injection. Therefore, an increase in
intrarenal adenosine may be an important factor in the
evolution of the renal hemodynamic changes following the

administration of hypertonic contrast media.

IV. THE RENAL CELLULAR ACTIONS OF ADENOSINE

A. A1 and A2 adenosine receptors in rabbit cortical
collecting tubule cells: modulation of hormone-
stimulated cAMP

(reproduced with permission, Am. Soc. Clin. Invest.; J.

Clin.Invest. 79:710, 1987)

A.1. Introduction

 

Adenosine is an important intermediate in the pathway
of purine nucleotide degradation and has been shown to
modify a variety of renal functions (Osswald at al 1978;
Spielman et a1 1980; Tagawa and Vander 1970; Osswald et al
1978b: Spielman 1984; Hedqvist et al 1978; Churchill 1982;
Coulson and Scheinman 1985). Adenosine can alter renal
excretory function (Osswald et al 1978; Spielman 1984;
Churchill 1982; Coulson and Scheinman 1985; Osswald 1975),
but it is not clear to what extent these effects are the
result 'of a direct tubular action or are an indirect
result of altered renal hemodynamics. Recent studies by
Forrest and coworkers on cultured amphibian epithelial
(A6) cells (Lang et al 1985), the rectal gland of the shark
(Forrest at al 1982), and the rabbit colon (Dobbins et a1
1984) have demonstrated that adenosine can alter active
transepithelial transport. Dillingham and Anderson
(Dillingham and Anderson 1985) have shown that adenosine

analogs inhibit vasopressin-stimulated water movement in

107

108

rabbit collecting tubules. These studies provide important
precedents for a direct action of adenosine on mammalian
renal tubular epithelium, though relatively little is known
regarding the cellular mechanisms underlying the tubular
actions of adenosine. The present study was undertaken to
determine the effect of adenosine analogs on the
production of cAMP by cultured cortical collecting tubule
(RCCT) cells isolated by immunodissection from the rabbit
kidney, and to test the hypothesis that adenosine
modulates hormone-stimulated cAMP production. Adenosine
analogs were used in this study to investigate the actions
of adenosine on cAMP. The analogs used were 5'-N-
ethylcarboxamide adenosine (NECA), N6-cyclohexyl adenosine
(CI-IA), and R-N‘ -phenylisopropyl adenosine (PIA). These
analogs were chosen because they are not substrates for
adenosine deaminase or the nucleoside carrier, and because
they each have greater potency at either the A1 or the A2
receptor. Data are presented indicating the presence of
both inhibitory (A1) and stimulatory (A2) receptors for
adenosine in RCCT cells and that stimulation of the A1
receptor inhibits hormone-stimulated cAMP formation. The
inhibition is mediated by the inhibitory guanine

nucleotide-binding regulatory component of the adenylate

109

cyclase system and is not dependent on stimulation of

prostaglandin production.

A.2. Methods

Materials. Trypsin (1:250), Dulbecco's modified
Eagle's medium (DME), collagenase, fetal bovine serum
(FBS), and bovine serum albumin (BSA) were purchased from
Gibco Laboratories. The adenosine analogs, 5'-N-
ethylcarboxamideadenosine (NECA), N6 -cyclohexyladenosine
(CI-IA), and R-N‘-phenylisopropyladenosine (PIA), and
1,3-diethyl-8-phenylxanthine (DPX) were a gift from Warner
Lambert. The phosphodiesterase inhibitor, R0 20-1724, was
a gift from Hoffmann-LaRoche. Arginine vasopressin (AVP)
was from Peninsula Laboratories. Pertussis toxin was from
List Biochemicals. 1251 adenosine 3'5'cyclicmonophosphoric
acid and 3H prostaglandin E2 were from ICN Biochemicals.
Other chemicals were reagent grade or better obtained from
standard sources.

Isolation and culture of RCCT cells. Rabbit cortical
collecting tubule (RCCT) cells were isolated by
immunodissection with a cell-specific mouse monoclonal
antibody as described previously (Spielman et a1 1986).
Briefly, plastic culture dishes were coated with an IgG1
(Egg-30) that binds exclusively to collecting tubule cells

when assayed by indirect immunofluorescence on rabbit

110

kidney cryotome sections (Spielman et a1 1986). Both
kidneys were removed from 4-5 wk old rabbits and the
cortical tissue separated and minced into a brei. The
tissue was digested in 0.1% collagenase for 40 min with
agitation every 10 min. Following centrifugation to remove
the collagenase, red blood cells were lysed with
hypotonic saline and the suspension was filtered through
250 um Gelman filter holders. The cells were centrifuged
and washed in 10% BSA to remove cellular debris. The
resulting cell suspension was placed on the antibody-
treated dishes and after 2-3 min was washed off with 10-15
ml/plate of phosphate buffered saline (PBS, composition in
mM: 151 NaCl, 45 KH2P04, and 2.5 NaOH). Freshly isolated
RCCT cells were grown to confluency (4-5 days) in 100mm
Costar culture dishes in DME with 10% FBS, 2mM glutamine,
and 1 uM dexamethasone in a water-saturated, 7% C02
environment at 37°C. Confluent RCCT cells were detached
from the dishes by treatment with trypsin (0.1%, with 0.05%
EDTA in PBS, pH 7.4), transferred into 24-we11 culture
dishes (1x10‘ cells/well), and cultured under the
conditions described above.

Determination of total cAMP production. Treatments
were done in triplicate using first passage cells grown for

5 days in 24-well dishes. Culture medium was removed and

111

the cells were washed once with Krebs buffer (composition
in mM: 118 NaCl, 25 NaHC03, 14 glucose, 4.7 KCl, 2.5
CaClz, 1.8 M9804, and 1.8 KH2P0‘, pH 7.4). The cells were
pretreated for 1 hr with 0.1 mM R0 20-1724, a phospho-
diesterase inhibitor which is not an adenosine receptor
antagonist (Londos et a1 1978; Huang and Drummond 1976),
and 1.5 units adenosine deaminase/ml Krebs at 37°C.
Phosphodiesterase activity was inhibited so that increases
in cAMP production could be easily detected and quantified.
Adenosine deaminase was included to remove any pre-
existing endogenous adenosine that might compete with the
experimental agents at the receptor. After pretreatment,
the buffer was aspirated and the cells were treated with
the same buffer containing the various hormones and
effectors at 37°C. Dimethyl sulfoxide (DMSO) was used as
solvent for the adenosine analogs and antagonists and was
included in the control wells. Treatment was terminated by
adding an equal volume of 5% (w/v) trichloroacetic acid
(TCA). The cells were frozen, thawed, and allowed to stand
at 4°C for 60 min. The samples were extracted three times
with 5 volumes of water-saturated ether, and the residual
ether was evaporated and the samples dried under vacuum
with vortexing. The samples were reconstituted with

distilled water, and total cAMP was determined by

112

radioimmunoassay as described by Frandsen and Krishna
(1976).

Determination of PGE:2 production. Treatments were
done in tmiplicate using first passage cells grown for 5
days in 24-well dishes. The culture medium was removed and
the cells were washed with Krebs buffer. The cells were
incubated with adenosine analog in Krebs for 30 min at
37°C. At the end of the incubation period, the effector
solution was removed and assayed for PGE2 concentration by
radioimmunoassay as described previously (Grenier et al
1981).

Protein determination. Following removal of the media
for cAMP assay, the remaining liquid was aspirated and the
cells were solubilized with a solution of 2% Na2C03, 0.4%
NaOH, 0.16% sodium tartrate, and 1% SDS. After 24 hr at
24°C, (samples were assayed for protein using a
modification of the Lowry method (Markwell et al 1978).

Statistical methods. Data were analyzed for
statistical significance using two-way analysis of variance
with the multiple comparisons test of Student-Newman-

Kuels'. Significance is defined as P < 0.05.

A.3. Results
To determine the incubation times optimal for

measuring the effects of adenosine analogs, time courses of

113

cAMP accumulation by RCCT cells were performed. For all of
the adenosine analogs, maximal increases in total cAMP
production occurred between 10 and 30 min of incubation.
Inhibition of cAMP accumulation by the adenosine analogs
was maximal by 10 min and remained unchanged for up to 60
min. Therefore, a 30 min incubation period was used for
most experiments since both inhibition and stimulation
would be maximal at this time.

Each of the three adenosine analogs studied (NECA,
CHA, and PIA) produced both a dose-dependent inhibition and
a dose-dependent stimulation of cAMP accumulation (Figure
12) depending on the concentration of the analog. The
decreases in cAMP of 30-50% from basal accumulation were
consistently observed at CHA and PIA concentrations of
10-100nM. The rank order of potency for stimulation of
cAMP accumulation was: NECA>PIA>CHA. The opposite order
of potency was observed for the inhibition of cAMP
accumulation by adenosine analogs, ie., CHA>PIA>NECA.

Further evidence for the existence of specific
adenosine receptors in RCCT cell populations was obtained
using the adenosine receptor antagonists, isobutylmethyl-
xanthine (IBMX) and 1,3-diethyl-8-phenylxanthine (DPX).
Both IBMX and DPX produced dose-dependent attenuation of

the adenosine analog-induced increases and decreases in

114

 

350

ANALOG.
*
. NECA
300 4pm... . . ....................... ,. - CHA
*

zotmmmm.

100.1,. .......

 

 

 

50

Figure 12.

 

Effect of adenosine analogs on total RCCT cell
cAMP production, expressed as percent of
control (basal). Mean control values for each
analog were: NECA, 341; CHA, 367; PIA, 455 pmol
cAMP/mg prot/30 min. Each point is the mean of
three samples from at least three different
isolations.

115

cAMP accumulation. DPx was a more potent inhibitor of the
effects of the adenosine analogs than IBMX. As little as
0.1 uM DPx produced 50% inhibition of the analog-induced
stimulation of cAMP accumulation; the concentration of IBMx
required tx> inhibit stimulation was 100 uM. Neither IBMX
nor DPx affected basal cAMP concentrations, presumably
since these incubations already contained a phospho-
diesterase inhibitor (R0 20-1724). To investigate the
potential role of adenosine in modulating hormonal
regulation of tubular function, the effects of adenosine
analogs on AVP- and isoproterenol-induced stimulation of
cAMP accumulation by RCCT cells were studied. Treatment of
RCCT cells with an inhibitory concentration of CHA (50 nM)
and Inn AVP attenuated the AVP-induced stimulation of cAMP
accumulation (Table 6). Higher concentrations of CHA (100
uM), and PIA or NECA at low (50 nM) or high (100 uM)
concentrations (Table 6), were much less effective in
modulating AVP-induced cAMP accumulation although they
produced slight inhibition. The response of RCCT cells to
isoproterenol (lun) was also attenuated by 50 nM CHA (Table
6) and, to a lesser extent, by 100 uM CHA.

Products of the cyclooxygenase pathway can inhibit the
action of AVP to stimulate cAMP production in canine

cortical collecting tubule cells (Garcia-Perez and Smith

116

Table 6. Effect of adenosine analogs on hormone-stimulated
cAMP production in RCCT cells.

 

Humane-stimulated cAMP production
(pmol cAMP/m3 protein)

 

 

Adenosine analog None vasopressin Isopmmenol
l “M 1 uM
CHA 0 358366 2570:722‘I 3274:247°
50 nM 126215’ 651:77' 1295:353'
100 MW 675:45' 1650:60 25472493
PIA O 219: 13 10591-72‘
50 Md 1 13:14' 629245’
100 “M 9123295' 807:87
NECA 0 4411-31 2199:167'
50 nM 433222 19793160

100 uM 1525:76' 16072216

 

117

1984). Therefore, we tested the hypothesis that CHA
stimulated the formation of prostaglandins that, in turn,
mediate the attenuation of the AVP response. Treatment of
RCCT cells with can (100 uM) failed to cause asignificant
increase in PGE2 production (control, 115 1 14 vs treated,
150 1 ll fmol PGEz/ug protein, NS).

To test the possibility that CHA attenuates AVP-
induced increases in cAMP accumulation in RCCT cells via
the inhibitory guanine nucleotide-binding protein, N1: the
effect of pertussis toxin was studied. Pertussis toxin
catalyzes the transfer of the ADP-ribose moiety of NAD to
the N1 protein, thereby inactivating it and preventing the
actions of inhibitory agents on cAMP production.
Pretreatment of RCCT cells (12 hr) with 1 ug of pertussis
toxin per ml of culture medium completely prevented the
CHA-induced inhibition of cAMP accumulation, from both
basal and AVP-stimulated states (Figure 13); pertussis
toxin had no effect on the ability of AVP alone to increase
cAMP concentrations, however, despite an increase in basal
cAMP production seen in pertussis toxin-treated cells,
stimulation of cAMP accumulation by higher concentrations

of CHA (10-100uM) was enhanced (Table 7).

118

 

2400

20001»

1600*r

tom cAMP
(and/mg protein)

8001-

400 ‘

 

1200 "'

W

% CONTROL

- PERTUSSIS TOXIN

    
 
    

_*

   
   
    

“\\\\\\\\"

\\\\\\\

\\\\\\\\\\\

\\\\\\\\

  
   
 
 
 
  
 
 
 
  
 
     

\\

  

     

§\\\\\\\\\\\\\\

   

 

 

Figure 13. Effect of pretreatment with pertussis toxin on
the modulation by CHA (50 nM) of 1 uM AVP-
stimulated cAMP production in RCCT cells. The
cells were pretreated for 12 hr with l ug/ml
pertussis toxin. Each bar is the mean of three

samples
isolations.

from at least three different

119

Table 7. Effect of pertusiis toxin on stimulation of cAMP

production by N -cyclohexyladenosine (CHA) in
RCCT cells.

 

 

 

cAMP production
‘5 cm
pmol/m; protein from control

No treatment
CHA 0 «£258

10 11M 369247 -9

30 MW 412242 +1

100 and 686286‘ +70

Pertussis toxin
CHA 0 9742107

10 Md 9122129 -6

3O uM 12082104‘ +24

100 pl“ 16542162' +70

 

120

A.4. Discussion

 

Interest in the renal actions of adenosine stems from
its possible involvement in the regulation of renal
function (Spielman and Thompson 1982; Osswald et al 1980).
Previous work has defined the effects of adenosine on renal
hemodynamics (Osswald et al 1978b; Spielman et a1 1980;
Tagawa and Vander 1970), renin release (Tagawa and Vander
1970; Osswald et al 1978b; Spielman 1984), and salt and
water excretion (Osswald et al 1978; Spielman 1984:
Churchill 1982; Coulson and Scheinman 1985; Osswald 1975).
While these actions of adenosine have been well documented,
the mechanism(s) by which adenosine produces these effects
in the kidney remains unclear. A primary impediment to the
study of the cellular actions of adenosine in the kidney is
the heterogeneous nature of renal tissue. With the advent
of the technique of immunodissection, used originally by
Wysocki and Sato (1978) for selective isolation of
lymphocytes and modified by Garcia-Perez and Smith (1983)
for use with renal tissue, large populations of homogeneous
cells can be isolated and cultured for study. In the
present study, rabbit cortical collecting tubule (RCCT)
cells were isolated as described by Spielman et a1 (1986)
for use in investigating the cellular mechanism of action

of adenosine on renal collecting tubule epithelium.

121

Previous work on adenosine receptors and adenosine
post-receptor mechanisms in other tissues has indicated
that there are at least two types of adenosine receptors
(reviewed in Daly 1932), both of which interact with the
cAMP second messenger system. Several previous reports
have identified the presence of both inhibitory (A1) and
stimulatory (A2) receptors in the kidney. Stimulatory
receptors have been identified in bovine renal medulla
(Birnbaumer at al 1974), in cultures of rat (Newman and
Levitski 1983) and toad (Lang et al 1985) kidney cortex,
and in isolated rat glomeruli (Abboud and Dousa 1983),
while inhibitory receptors have been reported to be present
in rat renal cortex (Coulson and Harrington 1979; McKenzie
and Bar 1973).

The present study provides evidence that RCCT cell
populations contain both A1 and A2 adenosine receptors.
Accordingly CHA and PIA produce an inhibition of cAMP
production, while NECA produces only slight inhibition.
Moreover, the rank order of potency for stimulation
(NECA>PIA>CHA) and inhibition (CHA>PIA>NECA) of CAMP
production seen with RCCT cells is similar to those
reported for other cell types (Londos et a1 1978; Londos et
a1 1980; Daly et a1 1981: Fain and Malbon 1979; Bruns

1980). In addition, pretreatment of the cells with

122

pertussis toxin completely eliminated the inhibitory action
of the adenosine analogs on cAMP accumulation while
enhancing the stimulatory action, suggesting the presence
of both receptors. Particularly relevant is the
observation that 30 uM CHA, which did not increase the cAMP
content of untreated RCCT cells, caused a significant
increase in cAMP accumulation in cells pre-treated with
pertussis toxin. In this instance, pertussis toxin appears
to have prevented A1 receptor-mediated inhibition of cAMP
production, thereby unmasking the stimulatory effect of CHA
acting via the A2 receptor.

Inhibition by methylxanthines of the effects on cAMP
accumulation provides additional evidence that these
receptors are extracellular, as opposed to the intra-
cellular P site receptor for adenosine, which is not
affected by methylxanthines. The finding of both A1 and A2
receptors in cultured RCCT cells that contain both
principal and intercalated cell types (Spielman et al 1986)
raises the question of whether both receptor types coexist
on the same cell type or if the receptor types are
partitioned between subpopulations of the cultured cells.
Investigation of the distribution of adenosine receptors on
RCCT cells awaits separation of cultured RCCT cells into

homogeneous cultures of principal and intercalated cells.

123

RCCT cells responded to AVP treatment as reported
earlier (Spielman et a1 1986), with an increase in cAMP
production. CHA caused a significant reduction in the
stimulatory effect of AVP. This modulation by CHA of
AVP-stimulated cAMP accumulation is in accord with a
preliminary study by Berl and Teitlebaum (1985) in which
they report inhibition of AVP-mediated increases in
papillary cell cAMP concentrations by 0.5 uM adenosine. In
RCCT cells, inhibition of AVP-stimulated cAMP accumulation
was observed with all three adenosine analogs at
stimulatory, as well as inhibitory concentrations.
Attenuation by stimulatory concentrations of the analogs
was apparently mediated by activation of the inhibitory
adenosine receptor (1111 ), since activation of A2 receptors
would be expected to add to the increase in cAMP production
and the concentrations used (100 uM) were sufficient to
activate both A1 and A2 receptors. There does not appear
to be a specific interaction between adenosine and AVP
since CHA also inhibited isoproterenol-induced stimulation
of cAMP production. These results indicate that adenosine
can influence tubular transport processes by acting alone
or by modulating the actions of other hormones.

Prostaglandin E2 inhibits AVP-induced increases in

cAMP by canine collecting tubule cells (Garcia-Perez and

124

Smith 1984), and could therefore, provide a mechanism for
the CHA-induced inhibition of the AVP response in RCCT
cells. However, prostaglandin (PGEZ) production was
unaffected by 100 on CHA, suggesting that prostaglandins do
not mediate the adenosine inhibition of hormone-induced
cAMP accumulation.

Hormones that act through receptors to inhibit cAMP
accumulation, typically activate the inhibitory GTP-binding
protein of the adenylate cyclase system (Rodbell 1980).
This protein, Ni' is a substrate for ADP-ribosylation by
pertussis toxin (Katada et al 1984), and ADP-ribosylation
prevents Ni from exerting its.inhibitory effect on
adenylate cyclase. Pertussis toxin-pretreatment of RCCT
cells completely abolished the action of CHA to inhibit
basal as well as AVP-stimulated cAMP production. This
suggests that N1 is involved in the modulation of the AVP
response by CHA. However, it has been reported recently
that the increase in intracellular free calcium
concentrations produced by some hormones through the
breakdown of membrane phospholipids is also mediated
through a pertussis toxin-sensitive N protein (Verghese et
a1 1986). Therefore, adenosine analogs may cause an
increase in intracellular free calcium concentrations that

subsequently result in inhibition of the increased cAMP

125

produced by AVP. An interaction of calcium with
AVP-stimulated cAMP has previously been reported by
Teitelbaum and Berl (1986), where inhibition of AVP-induced
cAMP stimulation in rat papillary collecting tubule cells
was produced by treatment with the calcium ionophore,
A23187. That adenosine and adenosine analogs may act
through intracellular calcium is suggested by the actions
of adenosine to inhibit neurotransmitter release from nerve
terminals and glucose oxidation in adipose tissue, both of
which are thought to be mediated by increasing
intracellular calcium concen-trations (Hedgvist et a1 1978:
Souness and Chagoya de Sanchez 1981). A

The results of the present study provide evidence for
stimulatory (A2) and inhibitory (A1) adenosine receptors on
rabbit cortical collecting tubule cells and suggest a role
for adenosine in the regulation of tubular function by a
direct action on cAMP‘production, and by modulation of

hormone-induced cAMP production.

Adenosine is a ubiquitous compound that, among other

actions, causes vasodilatation and vasoconstriction,

126

inhibits neurotransmission, platelet aggregation, and
lipolysis, and stimulates glucose oxidation. These actions
have been reviewed extensively elsewhere (Arch and
Newsholme 1978; Feigl 1983; Fox and Kelley 1978). In the
kidney, elevation of adenosine levels results in changes in
renal blood flow, an inhibition of renin release, and
decreases in the glomerular filtration rate and the
excretion of sodium and water (Osswald et al 1978; Osswald
et al 1978a; Spielman 1984; Spielman and Thompson 1982;
Spielman et al 1980; Tagawa and Vander 1970). The cellular
mechanisms involved in the renal actions of adenosine have
not been well characterized. In a variety of other cell
types, extensive research has revealed the existence of two
extracellular receptors for adenosine through which changes
in adenylate cyclase activity, and cAMP production, are
mediated (Daly 1982; Fain and Malbon 1979; Londos et a1
1980). These receptors are denoted A1 (or R1) and A2 (or
R.), for inhibition and stimulation of cAMP, respectively
(Londos et al 1980), and are thought to be coupled to
guanine nucleotide binding proteins (N1 and N.,
respectively). Both of these receptor types have been
demonstrated in the renal cortex and medulla in a variety
of species (Birnbaumer et a1 1974; Coulson and Scheinman

1985; McKenzie and Bar 1973; Newman and Levitski 1983:

127

Woodcock et a1 1984). A recent report from this laboratory
has demonstrated the presence of Al- and Az-like effects of
adenosine analogs in cultures of rabbit cortical collecting
tubule (RCCT) cells (RESULTS, IV.A.). 1

Several of the actions of adenosine, namely inhibition
of neurotransmitter release (Hedqvist and Fredholm 1976)
and stimulation of glucose oxidation (Souness and Chagoya
de Sanchez 1981), do not appear to be mediated through the
adenylate cyclase-cAMP system. It has been suggested that
these actions involve calcium. Therefore, the present
study was designed to test the hypothesis that adenosine
can modulate tubular function through changes in
intracellular free calcium. Adenosine analogs were used to
determine the effect of adenosine on intracellular free
calcium. The sensitivity of this response to xanthines and
pertussis toxin was studied, and the source of the

elevation in intracellular free calcium was determined.

8.2. Methods

Immunodissection. RCCT cells were obtained as
described previously (Spielman et al 1986). Briefly,
polystyrene culture dishes were coated with 0.2 mg of a
cell-specific mouse monoclonal antibody, IgGa(£§-30),
which binds exclusively to collecting tubule cells when

assayed by indirect immunofluorescence on rabbit kidney

128

cryotome sections (Spielman et a1 1986). After 1-2 hours
at 24 C, unbound antibody was discarded and the plates
treated with a 1% bovine serum albumin solution to saturate
the exposed polystyrene sites. A

Collagenase-dispersed cells from rabbit kidney cortex
were overlaid on the antibody-coated plates for less than 3
min. The plates were then washed 3-5 times with buffered
saline (PBS, composition in mM: NaCl, 138; KHzPO‘, 1.5;
KCl, 3; Nazi-1P0“ 8.1; pH 7.4). To the remaining bound
cells was added 10 ml of a Dulbecco's Modified Eagle's
media containing 10% decomplemented fetal bovine serum, 2
mM glutamine, 1 uM dexamethasone, and 50 ug/ml each of
penicillin and streptomycin. Cells were incubated at 37 C
in 7% CO2 and grown to confluency in 3-5 days. When
first-passage cells were used, cells were detached from the
culture dishes by treatment with trypsin (0.1%, with 0.05%
EDTA in PBS, pH 7.4). For experiments involving subsequent
measurement of intracellular calcium, cells were
transferred, after detachment, to 100 mm culture dishes and
grown for 3-5 additional days; for experiments involving
measurement of cAMP accumulation, cells were transferred to
24-we11 dishes and grown for 3-5 days.

Measurement of intracellular free calcium. Intra—

cellular free calcium concentration in RCCT cells was

129

measured as described by Grynkiewicz et a1 (1985) using the
fluorescent calcium chelator, furs-2. Suspensions of RCCT
cells (ca. 107 cells/ml) were treated with the membrane
permeable acetoxymethyl ester of furs-2 (furaQZ/AM, final
concentration: 4 uM) for 30 min at 37 C in a shaking water
bath. This procedure allows endogenous esterases to cleave
the fura-2/AM molecuLe to produce the membrane impermeant
fura-2 molecule. The cells were centrifuged at 1500 rpm,
4°C for 6-8 min, and washed twice with simplified saline
solution (SSS, composition in mM: NaCl, 145; KCl, 5:
NaZHP04, l; CaClz, 1; MgClz, 0.5; glucose, 5; HEPES, 10; pH
7.4) to remove any extracellular fura-Z/AM. To measure
fluorescence of the trapped furs-2, cells were diluted
1:100 with SSS at 37 C. Fluorescence was measured with an
Aminco spectrofluorimeter. One milliliter samples (ca.
105 cells/ml) were contained in a temperature-controlled
1 cm square quartz cuvette continuously stirred.
Monochromator settings were: excitation, 342 and 385 nm
with a 5 nm slit; emission, 510 nm with an 11 nm slit.
Furs-2 and the fura-2/calcium complex have fluorescence
maximums at different excitation wavelengths, with the free
dye fluorescing at 385 nm and the calcium-bound form of the
dye fluorescing at 342 nm. Therefore, increases in

intracellular calcium are reflected as an increased signal

130

from the complexed form of the dye (342 nm) and a decreased
signal from the free dye (385 nm). To estimate
intracellular free calcium, the ratio (R) of fluorescence
at 342 and 385 nm was used in the following equation:
[Ca]1 = Kd x (R-Ro/Rs-R) x (Fo/Fs)385,

where R0 is the ratio with zero calcium (EGTA added) and_Rs
is the ratio with saturating calcium (digitonin added).
[Ca]1 is the cytosolic free calcium concentration in
nmol/1. K,I is the dissociation constant for the
fura-2-calcium complex, F0 is the fluorescence at 385 nm
with zero calcium and Fe is the fluorescence at 385 nm with
saturating calcium concentrations.

Measurement of total cellular cAMP accumulation.
Treatments were done in triplicate using first passage
cells grown for 4 days in 24-well dishes. Culture medium
was removed and the cells were washed once with Krebs
buffer (in mM: NaCl, 118; NaHCOa, 25; glucose, 14; KCl,
4.7; CaClz, 2.5; 14950,, 1.8; and KHZPOU 1.8; pH 7.4).
The cells were pretreated for 1 hr with 0.1 mM R0 20-1724,
3 phosphodiesterase inhibitor which is not an adenosine
receptor antagonist, and 1.5 units adenosine deaminase/ml
Krebs at 37 C. After pretreatment, the buffer was
aspirated and the cells were treated with the same buffer

containing the various hormones and effectors at 37 C.

131

Treatment was terminated by adding an equal volume of 8%
(w/v) trichloroacetic acid. The cells were frozen, thawed,
and allowed to stand at 4 C for 60 min. The samples were
extracted three times with 5 volumes of water-saturated
ether, and the residual ether was evaporated and the
samples dried under vacuum with vortexing. The samples
were reconstituted with distilled water and total cAMP was
determined by radioimmunoassay as described by Frandsen and
Krishna (1976).

Determination of cellular protein content. Following
removal of the media for cAMP assay, the remaining liquid
was aspirated and the cells were solubilized with a 1%
sodium dodecyl sulfate solution. After 24 hr at 24 C,
samples were assayed for protein using a modification of
the Lowry method (Markwell et a1 1978).

Statistical analysis. Data were analyzed for
statistical significance using two-way analysis of variance
with the multiple comparisons test of Student-Newman-
Kuels'. Significance is defined as a p value of less than
0.05.

Materials. Trypsin (1:250), Dulbecco's modified
Eagle's medium, collagenase, fetal bovine serum, and bovine
serum albumin were purchased from Gibco Laboratories, Grand

Island, NY. Fura-2 was purchased from Molecular Probes,

132

Junction City, OR. The adenosine analogs, 5'-N-ethylcar-
boxamideadenosine (NECA), N°-cyclohexyladenosine (CHA), and
R-N° -phenylisopropyladenosine (PIA), and 1,3-diethy1-8-
phenylxanthine (DPX) were a gift from Warner Lambert,
Milford, CN. 8-cyclopenty1-l,3-dipropy1 xanthine (C-101)
was purchased from Research Biochemicals, Inc., Wayland,
MA. The phosphodiesterase inhibitor, R0 20-1724, was a
gift from Hoffmann-LaRoche, Nutley, NJ. Arginine
vasopressin (AVP) was from Peninsula Laboratories, Inc.,
Belmont, CA. Pertussis and cholera toxins were from List
Biological Laboratories, Inc., Campbell, CA. 1251 adenosine
3'5'cyclicmonophosphoric acid was from ICN Biomedicals,
Inc., Irvine, CA. Other chemicals of reagent grade or

better were obtained from standard sources.

8.3. Results

The basal intracellular free calcium concentration in
primary cultures of rabbit cortical collecting tubule
(RCCT) cells was 165 1 8 nM (mean 1 SEM, n=67), and in
first-passage RCCT cells was 342 1 20 nM (n=64).
Subsequent experiments were performed on passaged RCCT
cells unless otherwise noted, to be comparable to the
experiments involving cAMP production, which were performed
on first-passage cells. These values for intracellular

calcium concentration are only estimates of the true

133

cytosolic free calcium concentration in these cells since
there are many shortcomings in any technique to measure
intracellular ion concentrations. Because we have not
corrected our results for dye leakage or sequestration, we
are presenting the majority of the data as percent change
from control.

The addition of the adenosine analogs NECA, CHA, and
PIA, as well as native adenosine resulted in a
dose-dependent, transient increase in the intracellular
free calcium concentration in both primary (Figure 14A) and
first-passage cells (Figure 148). The increase occurred
within 10 sec of the addition of analog and intracellular
calcium returned to the control level by 2 min after analog
addition (Figure 15). The ratio for the calculation of
calcium in all experiments was taken at the peak of the
response at 342 nm and immediately following this at 385
nm. The relative responses by the two groups of cells were
equal, despite the higher absolute basal calcium
concentration in the passaged cells. Reproducible changes
in intracellular free calcium were observed at adenosine
concentrations ranging from 10 nM to 100 nM, with an ECso
of approximately 0.5 uM. There was no difference in the
potency of the adenosine analogs to increase intracellular

calcium levels.

134

 

140

_ NECA

120‘l
.. .. CHA

‘
_.L
'

L
W

Increase in intracellular calcium
(percent change)

 

 

 

0 :— 2 ; 4 t
8 7 6 5 4
[ADENOSINE ANALOG]

(-|og 11)

 

Figure 14A. Effect of adenosine analogs on intracellular
free calcium in primary RCCT cells, expressed
as percent change from control.

135

 

140
._... NECA
120i"
__ CHA
‘§ 1001 /.u°
O -
0 ,' l
A .
L 0
2 g 1
3.2 w"
80
fit
.53
.5 8 6°
0.

0V
9‘)
3
b ‘00
.E

20"

 

 

 

- g n L
r v v r

8 7 6 5 4

 

[ADENOSINE ANALOG]
Hog M)

Figure 148. Effect of adenosine analogs on intracellular
free calcium in first-passage RCCT cells,
expressed as percent change from control.

Figure 15.

136

1001

PBSlOuM CHA [MG EGTA

 

50‘

 

96 FLUORESCENCE

 

—- aulofluorescence

 

 

Representative fluorescence tracing of response
of intracellular calcium to adenosine analogs.
Ordinate, relative fluorescence. Abcissa,
time; bar equals 1 min. Upper tracing is
fluorescence at excitation wavelength of 342
nm, lower tracing is at wavelength of 385 nm.
Emission wavelength was 510 nm. Digitonin
(DIG) was added to a final concentration of 50
uM to determine the fluorescence maximum (Fmax)
and EGTA was added to a final concentration of
3 mM to obtain the fluorescence minimum (Fmin).
Autofluoresence was the fluorescence of an
equal number of RCCT cells not treated with
fura-Z.

137

The effects of adenosine analogs on the intracellular
free calcium concentration were antagonized by the
adenosine receptor antagonists, 1,3-diethyl- 8-phenyl-
xanthine (DPX, Figure 16), and the Al-selective antagonist,
8-cyclopentyl—1,3-dipropylxanthine (C-101, Figure 17). The
ICso for DPX was approximately 5 uM at adenosine analog
concentrations of 1 uM, while C-101 was much more effective
at inhibiting the increase in calcium, with an IC5° of
approximately 10 nM for adenosine concentrations of 1 uM.
This xanthine compound (C-101) has previously been reported
in rat brain membranes to antagonize 3H CHA binding with
an ICso of 0.8 nM, and 3H NECA binding with an ICso of 500
nM, suggesting a much greater specificity of the antagonist
for the high affinity A1 adenosine receptor. Comparable to
its reported effects on binding of adenosine analogs in
brain membranes, in RCCT cells, C-101 antagonizes the
inhibition of basal cAMP production by CHA (Table 8), as
well as the inhibitory action of CHA on hormone-stimulated
cAMP production (Table 8), but does not affect the increase
in cAMP production by NECA (Table 8). The ICso for
blocking the CHA-induced inhibition of cAMP was
approximately 10 nM, similar to that for blocking the

calcium response.

138

 

 

 

7////////////l

     

     

         

 

    

gig/2%??? 4
1.7////1
! gig/3%.??? s >
- r/////lu 5mm.
Egg/2?, mm
,7/4 7 (

747/7?

 

 

100

0 no
no 4

.J

80 1r-
2C>v
O

Amococo E853
8228 co_:__ooo::_ E 8855

RCCT
(single

or with concomitant addition of NECA

and either 1 uM DPx (solid) or 5 uM DPX (double

Effect of DPX on the response to NECA.
hatch).

cells were treated with NECA alone

hatch),

Figure 16.

139

 

50

4o ..

Percent increase
(luMCMM

N
O

10*

H
O

 

 

L A

 

Figure 17.

o 0.1 1:0 16 160 1600
[C—101]
(0M)

Effect of C-101 on the response of
intracellular calcium to CHA. RCCT cells were
treated with 1 uM CHA in the presence of
varying concentrations of C-101. The increase
in intracellular calcium in response to 1 uM
CHA in the absence of C-101 was approximately
36%.

140

Table 8. Effect of C-lOl on stimulation and inhibition of
cAMP production by adenosine analogs in RCCT
cells.

 

MW—

 

;rgam C-lOl concentration (n11)
m. _l_ _J.9__ _1.0.0_ .1999.—
nme 631-10 62114 711-11 791-14 103212
CHA 50 till 231; 6* 141 4* 34:14 961; 7 144-:11
AVP 1 all 322221* 272-1141" 3471-39 3781;46* 397:, 5*
CHA + AVP 198125111; 1031-15! 2761;39* 36915? 4271; 6*
m 10 uM 122-1171" to 1401381” 154-521* 226118”

 

First passage RCCT cells were incubated for 1 h with R) 20-1724
and adenosine deaminase as described in Methods. Cells were
then treated for 20-30 min with either Krebs (none), adenosine
analog (CHA or NKA) , or vasopressin (AVP) , either alone or in
the presence of varying concentrations of the adenosine receptor
antagmist, C-lOl. All incubatiom were performed at 37 C under
a 7% C02 atmosphere. cAMP production was determined by RIA. ND,
not determined.

*, P< 0.05 when cmpared to respective sanple without treatment

#, P< 0.05 when compared to respective sanple with AVP

80-

60+

404

% FLUORESCENCE

201

Inn-*c-oelo“ .....- ...".-.-
ul...‘oe""...'

 

Figure 18.

141

luM SmM luM
P88 NECA EGTA NECA

—— aulolluorescence

M

‘0

The response of intracellular calcium to NECA
in the presence and absence of extracellular
calcium. See legend to Figure 2 for
explanation of tracing. Left half, response to
1 uM NECA in the presence of 1 mM extracellular
calcium. Right half, response to NECA in the
absence of extracellular calcium.

142

The source of the elevation of intracellular free
calcium in RCCT cells was determined by chelation of
extracellular calcium. EGTA was added to the cell
suspension at a final concentration of 3 mM (media calcium
concentration l-l.5 mM), and a control fluorescence ratio
was obtained immediately. Within 30 sec of the addition of
EGTA the effectors were added and the experimental
fluorescence ratio was obtained (Figure 18). Addition of
EGTA always resulted in a lowering of the measured control
calcium concentration by one-third to one-half. Despite a
change in baseline calcium concentration, there was no
effect of chelation of extracellular calcium on the
relative increase in intracellular calcium in response to
adenosine. When adenosine was added after 2-3 min of EGTA
treatment, the increase in calcium was reduced by 50% or
greater, suggesting an effect of EGTA on intracellular
calcium stores when the cells were exposed to EGTA for an
extended period of time.

It has been reported by Nakamura and Di (1985) that
treatment of mast cells with pertussis toxin inhibits 45Ca
influx and membrane phospholipid breakdown induced by the
mast cell receptor agonist, compound 48/80. They suggest
that a guanine nucleotide regulatory protein (N protein)

may be involved in calcium-mediated cellular responses. In

143

 

 

. CHA

. NECA

O CHA/PT
A NECA/PT

 

 

 

100
80-1»
E
.2
.‘3
o
0A
I— 0
;§ §‘ GOjr
§fi
33%
.5 i 401}
”V
a
o
8
o
E
201-
0
Figure 19.

 

Effect of pertussis toxin on the responses to
the adenosine analogs, CHA and NECA. Filled
symbols, no pertussis toxin; open symbols, RCCT
cells were pretreated for 12 h with 1 ug
pertussis toxin/ml before measuring calcium.

144

addition, the chemotaxis response of phagocytes, which is
believed to involve calcium, is inhibited by both pertussis
toxin and cholera toxin (Verghese et a1 1986). Therefore,
the effects of these toxins on the calcium response of RCCT
cells to adenosine were studied. Pretreatment for 12 hr
with 1 ug/ml pertussis toxin completely blocked the
increase in intracellular free calcium by the adenosine
analogs NECA and CHA (Figure 19). There was no effect of
12 hr pretreatment of RCCT cells with 5 ug/ml cholera toxin

on the increase in calcium in response to adenosine.

3.4. Discussion

Elevation of the intrarenal levels of adenosine
results in many changes in ~renal function, including
changes in hemodynamics, renin release, and a decrease in
the excretion of solute and water (Osswald et al 1978;
Osswald et al 1978a; Spielman 1984; Spielman and Thompson
1982; Spielman et a1 1980; Tagawa and Vander 1970).
Because of the cellular heterogeneity of the kidney and the
close interaction of these aspects of renal function, it
has been difficult to determine which cell types of the
kidney are directly affected by adenosine and what
intracellular events are responsible for transducing the

actions of adenosine.

145

It has been demonstrated in many cell types (Daly
1982; Fain and Malbon 1979; Londos et al 1980), including
renal cells (Birnbaumer et a1 1974; Coulson and Scheinman
1985; McKenzie and Bar 1973; Newman and Levitski 1983;
Woodcock et a1 1984) that adenosine can increase and
decrease the production of cAMP by altering the activity of
adenylate cyclase. These studies suggest a role for cAMP
as an intracellular mediator of the actions of adenosine.

A change in intracellular calcium has also been
suggested to be involved in some of the actions of
adenosine (Hedqvist and Fredholm 1976; Souness and Chagoya
de Sanchez 1981) but an increase in calcium by adenosine
has been directly demonstrated only in hepatocytes (Sistare
et a1 1985). A specific effect of adenosine to increase
intracellular calcium in renal cells has not been
demonstrated.

In the present study, we have used the
fluorescent calcium indicator, fura-2, to measure
intracellular free calcium concentrations in RCCT cells,
and to determine what effects, if any, occur with the
addition of adenosine to these cells. Adenosine, and
adenosine analogs that are relatively specific for A1 or A2

adenosine receptors, produced dose-dependent, transient

146

increases in intracellular free calcium. Primary and
first-passage RCCT cells responded equally to adenosine.

An increase in calcium may occur by mobilization of
calcium from intracellular stores, or by stimulation of an
influx of extracellular calcium. By adding EGTA to the
cell suspension, extracellular calcium is removed as a
possible source. Chelation of extracellular calcium did
not affect the adenosine-induced increase in intracellular
calcium, though the response could be significantly reduced
by prolonged exposure to an extracellular medium that was
calcium-free. These results suggests that the calcium is
mobilized from an intracellular source. Further studies
are necessary to determine the exact mechanism responsible.

In contrast to their effects on cAMP production, with
NECA being more potent at eliciting an increase and CHA
being more potent at the inhibitory receptor, the adenosine
analogs were all equally potent at inducing an increase in
intracellular calcium. This may represent an action of
adenosine which is not mediated through an extracellular
receptor, perhaps by changing metabolism, or by interacting
with the intracellular adenosine receptor (P site). That
the action of adenosine to increase intracellular calcium
is mediated via an extracellular receptor is supported by

the results that xanthines antagonize the effect. Xanthines

147

are adenosine receptor antagonists that do not alter the
inhibitory effect of adenosine on cAMP mediated through the
intracellular P site (Fain et al 1978).

To determine the involvement of guanine nucleotide
regulatory proteins (N proteins) in the increase in
calcium, RCCT cells were treated with pertussis toxin and
cholera toxin. Pertussis toxin ADP-ribosylates and
inactivates the inhibitory guanine nucleotide binding
protein, Ni (28). This protein is normally thought to be
associated with receptors that, when activated, result in
inhibition of cAMP production. Recently, a 40 to 41 kD
protein that is both pertussis toxin and cholera toxin
sensitive has been identified in leukocytes (Verghese et a1
1986) and appears to be involved in calcium-mediated
chemotactic responses. This protein appears to be distinct
from N1 and has been termed Nc (Verghese et a1 1986). In
the present study, the action of adenosine to mobilize
intracellular calcium was inhibited by pertussis toxin and
not by cholera toxin, indicating coupling to an Ni-like
protein.

The receptor-mediated mobilization of calcium showed
no preference for the various adenosine analogs, in
contrast 13) the analog selectivity for inhibition or

stimulation of cAMP production, and raises the possibility

148

of calcium-mobilization by a previously unidentified
adenosine receptor. On the other hand, our findings that a
highly specific A1 receptor antagonist blocks both the
inhibition of cAMP production and the increase in calcium,
and that both actions are pertussis toxin-sensitive seem to
indicate that the same receptor, A1, via coupling to Ni'
may be responsible for both inhibition of cAMP production
and calcium mobilization. The results of the present study
do not provide sufficient evidence for acceptance of either
of these receptor schemes for adenosine-mediated calcium
mobilization.

These studies demonstrate an adenosine receptor in
RCCT cells, which is responsible for mobilization of
calcium from intracellular stores and is coupled to a
guanine nucleotide regulatory protein. Further studies are
necessary to determine what extracellular adenosine
receptor mediates the changes in calcium and what role
this increase in intracellular calcium has in the renal

actions of adenosine.

V. SUMMARY AND CONCLUSIONS

The studies described in this dissertation provide

evidence that:

1.

the elevation of intrarenal adenosine levels, by
blocking its cellular uptake, or by
pharmacological or pathophysiological maneuvers,
leads to pronounced changes in renal function
(III.A., III.B., III.C.)

adenosine signal transduction mediating these
changes in renal function may occur through
changes in the intracellular levels of cyclic AMP
and/or calcium (IV.A., IV.B.)

an interaction between adenosine and other
hormones in the kidney may be responsible for

some of the renal actions of adenosine (IV.A.).

The elevation of intrarenal adenosine levels by

blocking its cellular uptake and degradation, as well as by

pathophysiologic manuevers can lead to changes in renal

function similar to that seen with the infusion of

exogenous adenosine. Elevation of endogenous adenosine

reduced renal blood flow, glomerular filtration rate, and

renin release. These changes mimicked those produced by

149

150

adenosine infusion by exhibiting a sodium-dependency. That
is, the changes in renal function are more pronounced in
animals that have been maintained on a low-sodium diet, and
are reduced in those fed normal or high-sodium diets.
While these studies have not proven that adenosine acts as
an intrinsic regulator of renal function under normal
circumstances, they have shown that endogenously-produced
adenosine is capable of altering renal function, and that
adenosine may be responsible for some of the changes in
renal function observed under certain pathophysiological
circumstances.

The hypothesis that adenosine is involved in the
regulation of renal function is supported by the results
reported in the second part of this dissertation which
provide evidence for adenosine receptors and post-receptor
systems ix: renal epithelial cells. Rabbit cortical
collecting tubule cells were shown to posess adenosine
receptors coupled to the cAMP and calcium second messenger
systems. A model for the cellular actions of adenosine via
these systems is shown in Figure 20.. Though the three
proposed receptors are depicted as being present in one
cell, the data presented in this dissertation do not allow
a conclusion to be drawn regarding the presence of these

receptors in specific cells of the collecting tubule.

151

. 1‘:

NECA)HA)CHA

(gs/l. .d..- a.

AVP GU’ GIP

 

 

CAMP 0?. Kinaae A

Ca"

/ P K'nnae C
GlP GDP /
/ \:\ DAG

K“ I)" ‘mmem J

“”m— __—.._

 

 

 

 

 

 

 

Figure 20. Model cell of adenosine receptor systems.
Hypothetical design of adenosine receptor
systems in RCCT cells. A,R and P: adenosine
receptors; N: guanine nucleotide regulatory
protein: AC: adenylate cyclase; PLC:

phospholipase C; i: inhibitory; a: stimulatory:
?: unknown.

152

Evidence was presented that RCCT cells possess A1 and A2
adenosine receptors, based on the observation that
adenosine analogs can inhibit and stimulate the
accumulation of cAMP. Adenosine analogs were capable of
increasing cAMP production at concentrations from 10'7 M to
10" M, and decreasing cAMP production at concentrations
from 10‘9 M to 10‘7 M. The action of adenosine analogs to
inhibit cAMP production was pertussis toxin-sensitive and
this suggests the involvement of the inhibitory guanine
nucleotide protein, N1 . A further action of adenosine
analogs in these cells was mobilization of calcium from
intracellular stores, at concentrations from 10'9 M to 10"
M. This is the first report of an effect of adenosine on
intracellular calcium in a renal cell, and one of the first
reports of this action of adenosine in any cell type. This
action of adenosine analogs was also found to be coupled to
a guanine nucleotide binding protein similar to N1 because
treatment with pertussis toxin abolished the response.
These results suggest that changes in intracellular
cAMP and/or calcium may mediate the renal actions of
adenosine, though these studies can not exclude the
involvement of other second messenger molecules such as
cGMP in the actions of adenosine. These studies also

demonstrate an interaction between the cellular actions of

153

adenosine and the actions of other hormones on the cAMP
system. That is, vasopressin- and isoproterenol-stimulated
cAMP production can be blunted by the presence of adenosine
analogs. Therefore, adenosine may produce some of its
actions on the kidney by modulating the effects of other
intrarenal hormones.

Because the cellular actions of adenosine were studied
in collecting tubule cells, a direct correlation cannot be
drawn between these results and the actions of adenosine on
other renal cell types such as the glomerular mesangial
cells, the endothelial cells of the afferent and efferent
arterioles, and the renin-secretory juxtaglomerular cells.
However, the collecting tubule is an extremely hormone-
sensitive segment of the nephron and can be used as a model
for the study of adenosine's actions in other cell types.
’The results of these investigations can be used as a data-
base for further studies in all of the different cells in
the kidney. Furthermore, adenosine does have effects on
excretion as shown in the first chapter of this
dissertation as well as in several previous reports. These
excretory changes may be mediated by the direct action of
adenosine on the collecting tubule cells through one or
more of the mechanisms described in this dissertation. The

retaliatory metabolite hypothesis for adenosine, though

154

based on organ-level actions of adenosine, may have an
analogy in the local actions of adenosine. That is,
adenosine may act as a paracrine or autocrine hormone to
regulate the energy utilization of neighboringcells. This
action of adenosine may be important for modulatory control
of transport processes initiated by other hormones acting
on the collecting tubule. The analogy may be extended to
include the regulation of energy-requiring processes in
other cells such as glomerular mesangial and endothelial

cells, or juxtaglomerular cells.

APPENDIX A

APPENDIX A

Abbreviations and Definitions of Compounds Used

Adenosine deaminase (ADA): enzyme that catalyzes
deamination of adenosne to inosine, used to remove
endogenous adenosine from receptors for studying
adenosine analogs.

Arginine vasopressin (AVP): antidiuretic hormone, known to
increase cAMP production in the collecting tubule.

Cholera toxin: ADP-ribosylates and permanently activates
stimulatory guanine nucleotide regulatory protein, N..

N6 -Cyclohexyladenosine (CHA): adenosine analog, greater
potency at A1 receptor.

8-Cyclopenty1-1,3-dipropylxanthine (C-101): adenosine
receptor antagonist, greater potency at A1 receptor.

Deoxycorticosterone pivalate: corticosteroid, promotes
sodium retention, enhancing effects of high-sodium
diet to produce sodium- and volume-expanded state.

1,3-Diethyl-8-pheny1xnathine (DPX): adenosine receptor
antagonist, similar potency for A1 and A2 receptors.

Dipyridamole: inhibitor of nucleoside carrier.

5'-N-Ethylcarboxamideadenosine (NECA): adenosine analog,
greater potency at A2 receptors.

Erytho-.9-(2-hydroxy-3-nonyl)adenine (El-INA): inhibitor of
adenosine deaminase, used in measurement of adenosine
in blood and urine.

Furosemide, Lasix: diuretic, used to supplement and enhance
low-sodium diet to obtain sodium- and volume-depleted
state.

Lithiwm carbonate: marker of proximal tubule reabsorptive
capacity.

Maleic acid: produces experimental model of Fanconi
syndrome, generalized tubular transport dysfunction.

155

155

Meglumine/sodium diatrizoate-76% (Renografin 76): contrast
agent used clinically for radiographic identification
of blood vessels.

Pertussis toxin: ADP-ribosylates and inactivates inhibitory
guanine nucleotide protein, N1.

R-NG-Phenylisopropyladenosine (PIA): adenosine analog,
greater potency at A1 receptor.

Theophylline: adenosine receptor antagonist,, similar
action at A1 and A2 receptors, less potent than DPX.

1“) 20-1724: cyclic nucleotide phosphodiesterase inhibitor
that lacks adenosine receptor antagonist properties,
used to allow increases in cAMP production to be
detected.

APPENDIX B

157

APPENDIX B
Derivation of the equation for calculation of intracellular

calcium

Basis: fluorescence of a compound is proportional to its
concentration (calcium-bound fura-Z fluorescence is a:
function of the concentration of the bound dye)

Components:

concentrations of free and bound dye:

Cr and ob
proportionality coefficients:
8,1, Sbl, 5,2, Sb2 where l = wavelength 1
(342)
2 = wavelength 2
(385)
f = free dye
b = bound dye

the proportionality coefficients are related to the
extinction coefficient, excitation intensity,
instrument efficiency, In 10, and quantum efficiency.
S is measured from fluorescence intensities of solutions
with known concentrations of free and calcium-bound dye,
and the fluorescence intensities at wavelengths 1 and 2 are
related to the proportionality coefficients and c, and cb
by:
F1 = 8:102 7’ Sblcb P2 = Srzcr 7' szcb
c, and ob are related to the calcium concentration by:
ab = c,[Ca]/Kd, where Kd = the dissociation
constant for
fura-Z/calcium
the fluorescence ratio, R equals Fl/Fz, or:
R 3 (521°: 7' Sblcb)/(szcf + szcb)
replacing cb (c, cancels out):

R = (s£1 + SMECaJ/xdvm£2 + Sb2[Ca]/Kd)

158

and solving for [Ca]:

[ca] = Kd x [R-(Sgl/ng)/(Sf2/Sb2)-R](Sf2/Sb2)
where St 1 /Sr 2 = minimum limiting value for R with zero
calcium, and SI) 1 /Sb 2 = maximum limiting value for R with

saturating calcium.

LITERATURE CITE)

LITERATURE CITED

Abboud, H.E., and T.P. Dousa (1983) Action of adenosine on
cyclic 3'5'- nucleotides in glomeruli. Am. J. Physiol.
244: F633-F638.

Al—Bander, R.A., R.A. Weiss, M.H. Humphreys, and R.C.
Morris, Jr. (1982) Dysfunction of the proximal tubule
underlies maleic acid—induced type II renal tubular
acidosis. Am. J. Physiol. 243: F604-F611.

Ally, A.1., and K. Nakatsu (1976) Adenosine inhibition of
isolated rat ileum and antagonism by theophylline. J.
Pharmacol. Exp. Ther. 199: 208-219.

Angielski, S. (1963) The effect and metabolism of maleic
acid in the kidney. Acta Biol. Med. Soc. Sci. Gedan.
7: 61-97.

Angielski, S., and J. Rogulski (1975) Metabolic studies in
experimental renal dysfunction resulting from maleate
administration. In S. Angielski and U.S. Dubach
(eds): Biochemical aspects of kidney function. Huber:
Bern, pp. 86-100.

Ansari, z., and S.D. Baldwin (1976) Acute renal failure due
to radio-contrast agents. Nephron 17: 28-40.

Arch, J.R.S., and E.A. Newsholme (1978) The control of the
metabolism and the hormonal role of adenosine. Essays
in Biochemistry 14: 82-123.

Arend, L.J., A. Haramati, C.I. Thompson, and W.S. Spielman
(1984) Adenosine-induced decrease in renin release:
dissociation from hemodynamic effects. Am. J. Physiol.
247: F447-F452.

Baer, H.P., G.I. Drummond, and E.L. Duncan (1966) Formation
and deamination of adenosine by cardiac muscle
enzymes. Mol. Pharmacol. 2: 67-76.

159

160

Bakris, G.L., and J.C. Burnett, Jr. (1985) A role for
calcium in radiocontrast-induced reductions in renal
hemodynamics. Kidney Int. 27: 465-468.

Baumbach, L., and P.P. Leyssac (1977) Studies on the
mechanism of renin release from isolated superfused
rat glomeruli: Effects of calcium, calcium ionophore,
and lanthanum. J. Physiol. (London) 272: 745-764.

Bergeron, M., L. Dubord, C. Hausser (1976) Membrane
permeability as a cause of transport defects in
experimental Fanconi syndrome. J. Clin. Invest. ’57:
1181-1189.

Berl, T., and I. Teitelbaum (1985) Control of prostaglandin
synthesis by rat inner medullary collecting tubule
cells: role of cAMP. Kidney Int. 27: 253. (Abstr.).

Berlin, R.D., and J.M. Oliver (1975) Membrane transport of
purine and pyrimidine bases and nucleosides in animal
cells. Int. Rev. Cytol. 42: 287-336.

Berliner, R.W., T.J. Kennedy, and J.C. Hilton (1950) Effect
of maleic acid on renal funcion. Proc. Soc. Exp. Biol.
Med. 75: 791-794.

Berne, R.M., and R. Rubio (1974) Adenine nucleotide
metabolism in the heart. Circ. Res. 34-35:
111109-111120.

Birnbaumer, L., T. Nakahara, and P.C. Yang (1974) Studies
on receptor-mediated activation of adenylyl cyclases.
J. Biol. Chem. 259: 7557-7566.

Blantz, R.C., D.W. Peterson, L.C. Gushwa, and B.J. Tucker
(1975) Effect of .modest hyperglycemia on
tubuloglomerular feedback activity. Kidney Int. 22:
$206-$212.

Blume, A.J., C. Dalton, and H. Sheppard (1973)
Adenosine-mediated elevation of cyclic 3',5' adenosine
monophosphate concentrations in cultured mouse
neuroblastoma cells. J. Biol. Chem. 70: 3099-3102.

Brewer, E.D., H.O. Senekjian, A. Ince, and E.J. Weinman
(1983) Maleic acid-induced reabsorptive dysfunction in
the proximal and distal nephron. Am J. Physiol. 245:
F339-F344.

161

Bruns, R.F. (1980) Adenosine receptor activation in human
fibroblasts: nucleoside agonists and antagonists. Can.
J. Physiol. Pharmacol. 58: 673-691.

Burnstock, G. (1972) Purinergic nerves. Pharm. Rev. 24:
509-581.

Burnstock, G. (1976) Do some nerve cells release more than
one neurotransmitter? Neurosci. 1: 239-248.

Burnstock, G. (1978) A basis for distinguishing two types
of purinergic receptor. In RuW. Straub and L. Bolis
(eds): Cell membrane receptors for drugs and hormones:
a multidisciplinary approach. New York: Raven Press,
pp. 107-118.

Byrd, L., and R.L. Sherman (1979) Radiocontrast-induced
acute renal failure: a clinical and pathophysiologic
review. Medicine 58: 270-279.

Caldicott, W.J.H., N.K. Hollenberg, and H.L. Abrams (1970)
Characteristics of response of renal vascular bed to
contrast media. Evidence of vasoconstriction induced
by the renin-angiotensin system. Invest. Radiol. 5:
539-547.

Capito, K., and C.J. Hedeskov (1976) Inosine-stimulated
insulin release and metabolism of inosine in isolated
mouse pancreatic islets. Biochem. J. 158: 335-340.

Cattell, W.R., I.K. Fry, A.G. Spencer, and P. Purkiss
(1982) Excretion urography.I. Factors determining the
excretion of Hypaque. British Journal of Radiology.
40: 561-571.

Chou, C.C., J.B. Hook, C.P. Hsieh, T.D. Burns, and J.M.
Dabney (1971) Effects of radiopaque dyes on renal
vascular resistance. J. Lab. Clin. Med. 78: 705-712.

Churchill, P.C. (1982) Renal effects of 2-chloroadenosine
and their antagonism by aminophylline in anesthetized
rats. J. Pharm. Exp. Ther. 222: 319-323.

Churchill, P.C., and M.C. Churchill (1985) A1 and A2
adenosine receptor activation inhibits and stimulates
renin secretion of rat renal cortical slices. J.
Pharm. Exp. Ther. 232: 589-594.

162

Churchill, P.C., and A.K. Bidani (1982) Hypothesis:
Adenosine mediates hemodynamic changes in renal
failure. Medical Hypotheses 8: 275-285.

Clark, R.B. and M.N. Seney (1976) Regulation of adenylate
cyclase from cultured human cell lines by adenosine.
J. Biol. Chem. 251:4239-4246.

Clark, R.B., R. Gross, J.F. Su, and J.F. Perkins (1974)
Regulation of adenosine 3',5' monophosphate content in
human astrocytoma cells by adenosine and the adenine
nucleotides. J. Biol. Chem. 249: 5296-5303.

Conway, E.J., and R. Cooke (1939) The deaminases of
adenosine and adenylic acid in blood and tissues.
Biochem. J. 33: 479-492.

Coulson, R., and W.W. Harrington (1979) Renal metabolism of

N°,OZ'-dibutyryl adenosine 3',5'-monophosphate. Am. J.
Physiol. 237: F75-F84.

Coulson, R., and S.J. Scheinman (1985) Evidence for renal
tubular adenosine receptors. Kidney Int. 27:
255.(Abstr.).

Daly, J.W. (1982) Adenosine receptors: targets for future
drugs. J. Med. Chem. 25: 197-207.

Daly, J.W., R.F. Bruns, and S.H. Snyder (1981) Adenosine
receptors in the central nervous system: relationship
to the central actions of methylxanthines. Life Sci.
28: 2083-2097.

Davidson, W.D., and M.A. Sackner (1963) Simplification of
the anthrone method for the determination of inulin in
clearance studies. J. Lab. Clin. Med. 62: 351-356.

Dillingham, M.A., and R.J. Anderson (1985) Purinergic
regulation of basal and arginine vasopressin-
stimulated hydraulic conductivity in rabbit cortical
collecting tubule. J. Membr. Biol. 88: 277-281.

Dobbins, J.W., J.F. Laurenson, and J.N. Forrest (1984)
Adenosine and adenosine analogs stimulate adenosine
cyclic 3',5'-monophosphate-dependent chloride
secretion in the mammalian ileum. J. Clin. Invest. 74:
929-935.

163

Dobson, J.G. (1978) Reduction by adenosine of the
isoproterenol-induced increase in cyclic adenosine
3',5' monophosphate formation and glycogen
phosphorylase activity in rat heart muscle. Circ. Res.
43: 785-792.

Dole, V.P. (1961) Effect of nucleic acid metabolites on
lipolysis in adipose tissue. J. Biol. Chem. 236: 3125-
3130.

Drury, A.N., and A. Szent-Gyorgyi (1929) The physiological
activity of adenine compounds with special reference
to their action upon mammalian heart. J. Physiol.,
Land. 68: 213-237.

Eloranta, T.O. (1977 ) Tissue distribution of S-adenosyl
methionine and S- adenosyl homocysteine in the rat.
Biochem. J. 166: 521-529.

Ettienne, E.M., and J.C.S. Fray (1979) Influence of
potassium, sodium, calcium, perfusion pressure, and
isoprenaline on renin release induced by high
concentrations of magnesium. J. Physiol. (London) 292:
373-380.

Fain, J.N. (1973) Inhibition of adenosine cyclic 3',5'
monophosphate accumulation in fat. cells by adenosine,
N‘ -(phenylisopropyl) adenosine, and related compounds.
Mol. Pharmacol. 9: 595-604.

Fain, J.N., and C.C. Malbon (1979) Regulation of adenylate
cyclase by adenosine. M01. and Cell. Biochem. 25:

Fain, J.N., R.H. Painter, and W.F. Ward (1972) Effects of
adenosine nucleosides on adenylate cyclase,
phosphodiesterase, cyclic adenosine monophosphate
accumulation, and lipolysis in fat cells. J. Biol.
Chem. 247: 6866-6872.

Fain, J.N., R.E. Shepherd, C.C. Malbon, and F.J. Moreno
(1978). In J.N. Dietschy (ed): Disturbances in lipid
and lipoprotein metabolism. Washington, DC: Am.
Physiol. Society, pp. 213-228.

Feigl, E.D. (1983) Coronary physiology. Physiol. Rev. 63:
1‘205.

164

Feldman, J.M., and T.B. Jackson (1974) Specificity of
nucleotide-induced insulin secretion. Endocrinology
94: 388-394.

Forrest, J.N., G. Kelly, D. Gifford, E. Poeschla, S.
Sander, and H. Osswald (1984) Characteristics of
adenosine receptors mediating chloride transport.
Clin. Res. 32: 532A (Abstr.).

Forrest, J.N., E. Poeschla, and D. Rieck (1982)
Ribose-specific adenosine receptors mediate cyclic AMP
accumulation and chloride transport in the shark
rectal gland. Kidney Int. 21: 253.(Abstr.).

Forrest, J.N., D. Rieck, and A. Murdaugh (1980) Evidence
for a ribose-specific adenosine receptor (Ra)
mediating stimulation of chloride secretion in the
rectal gland of Squalus acanthias. Bull. Mt. Desert
Isl. Biol. Lab. 20: 152-155.

Fox, I.H., and W.N. Kelley (1978) The role of adenosine and
2'-deoxyadenosine in mammalian cells. Ann. Rev.
Biochem. 47: 655-686.

Fox, I.H., and L. Kurpis (1983). Binding characteristics of
an adenosine receptor in human placenta. J. Biol.
Chem. 258: 6952-6955.

Frandsen, E.K., and G. Krishna (1976) A simple
ultrasensitive method for the assay of cyclic AMP and
cyclic GMP in tissues. Life Sci. 18: 529-541.

Fray, J.C.S. (1977) Stimulation of renin release in
perfused kidney by low calcium and high magnesium.
Am. J. Physiol. 232: F377-F382.

Fray, J.C.S. (1980) Stimulus-secretion coupling of renin.
Role of hemodynamic and other factors. Circ. Res. 47:
485-492.

Fray, J.C.S., and C.S. Park (1979) Influence of potassium,
sodium, perfusion pressure, and isoprenaline on renin
release induced by calcium deprivation. J. Physiol.
(London) 292: 363-372.

165

Fynn, M., N. Onomakpome, and W.S. Peart (1977) The effects
of ionophores (A23187 and R02-2985) on renin secretion
and renal vasoconstriction. Proc. R. Soc. Lond. [Biol]
199: 199-212.

Garcia-Perez, A., and W.L. Smith (1983) Use of monoclonal
antibodies to isolate cortical collecting tubule
cells: AVP induces PGE release. Am. J. Physiol. 244:
C211-C220.

Garcia-Perez, A., and W.L. Smith (1984) Apical-basolateral
membrane asymmetry in canine cortical collecting
tubule cells: bradykinin, AVP, PGE2
interrelationships. J. Clin. Invest. 74: 63-74.

Gerber, K.H., C.B. Higgins, Y.S. Yuh, and J.A. Koziol
(1982) Regional myocardial hemodynamics and metabolic
effects of ionic and nonionic contrast media in normal
and ischemic states. Circulation 65: 1307-1314.

Gerkens, J.F., H.T. Heidemann, E.K. Jackson, and R.A.
Branch (1983) Aminophylline inhibits renal
vasoconstriction produced by intrarenal hypertonic
saline. J. Pharm. Exper. Therap. 225: 611-615.

Gmaj, P., A. Hoppe, S. Angielski, and J. Rogulski (1973)
Effects of maleate and arsenite on renal reabsorption
of sodium and bicarbonate. Am. J. Physiol. 225: 90-04.

Gougoux, A., G. Lemieux, and N. Lavoie (1976)
Maleate-induced bicarbonaturia in the dog: a carbonic
anhydrase-independent effect. Am. J. Physiol. 231:
1010-1017.

Grasl, M., and K. Turnheim (1984) Stimulation of
electrolyte secretion in rabbit colon by adenosine. J.
Physiol. 346: 93-110.

Grenier, F.C., T.E. Rollins, and W.L. Smith (1981)
Kinin-induced prostaglandin synthesis by renal
papillary collecting tubule cells in culture. Am. J.
Physiol. 241: F94-F104.

Grynkiewicz, G., M. Poenie, and R.Y. Tsien (1985) A new
generation of calcium indicators with greatly improved
fluorescence properties. J. Biol. Chem. 260:
3440-3450.

166

Gunther, R., S. Silbernegl, and P. Deetjen (1979) Maleic
acid-induced aminoaciduria, studied by free flow
micropuncture and continuous microperfusion. Pfluegers
Arch. 382: 109-114.

Guyton, A.C., J.B. Langston, and L.G. Navar (1964) Theory
for renal autoregulation by feedback of
juxtaglomerular apparatus. Circ. Res. 15: 1187-1197.

Haddy, F.J., and J.B. Scott (1968) Metabolically linked
vasoactive chemicals in local regulation of blood
flow. Physiol. Rev. 48: 688-707.

Hall, J.E., and J.P. Granger (1986) Adenosine alters
glomerular filtration control by angiotensin II. Am.
J. Physiol. 250: F917-F923.

Hall, J.E., J.P. Granger, and R.L. Hester (1985)
Interactions between adenosine and angiotensin II in
controlling glomerular filtration. Am. J. Physiol.
248: F340-F346.

Harrison, H.E., and H.C. Harrison (1954) Experimental
production of renal glycosuria, phosphaturia, and
aminoaciduria by injection of maleic acid. Science
120: 606-608.

Harvey, P.B. (1960) Vascular resistance changes produced by
hypertonic solutions. Am. J. Physiol. 199: 31-34.

Hashimoto, K., and K. Kumakura (1965) The pharmacological
features of the coronary, renal, mesenteric, and
femoral arteries. Jpn. J. Physiol. 15: 540-551.

Haslam, R.J., and J.A. Lynham (1972) Activation and
inhibition of blood platelet adenylate cyclase by
adenosine or by 2-chloroadenosine. Life Sci. 11:
1143-1154.

Hayslett, J. P., M. Kashgarian (1979) A micropuncture study
of the renal handling of lithium. Pfluegers Arch. 380:
159-163.

Hedqvist, P., and B.B. Fredholm (1976) Effects of adenosine
on adrenergic neurotransmission: Prejunctional
inhibition and post-junctional enhancement.
Naunyn-Schmeiderbergs Arch. Pharmakol. 293: 217-223.

167

Hedqvist, P., B.B. Fredholm, and S. Olundh (1978)
Antagonistic effects of theophylline and adenosine on
adrenergic neuroeffector transmission in the rabbit
kidney. Circ. Res. 43: 592-598.

Hjemdahl, P., and B.B. Fredholm (1976) Cyclic AMP-dependent
and independent inhibition of lipolysis by adenosine
and decreased pH. Acta. Physiol. Scand. 96: 170-179.

Huang, M., and G.I. Drummond (1976) Effect of adenosine on
cyclic AMP accumulation in ventricular myocardium.
Biochemical Pharmacology 25: 2713-2719.

Ismail, N.A., E.S.M. ElDenshary, and W. Montague (1977)
Adenosine and the regulation of insulin secretion by
isolated rat islets of Langerhans. Biochem. J. 164:

Itoh, S., O.A. Carretero, and R.D. Murray (1985) Possible
role of adenosine in the macula densa mechanism of
renin release in rabbits. J. Clin. Invest. 76:
1412-1417.

Katada, T. , G.M. Bokoch, J.K. Northrup, M. U1, and A.G.
Gilman (1984) The inhibitory guanine
nucleotide-binding regulatory component of adenylate
cyclase . Properties and function of the purified
protein. J. Biol. Chem. 259: 3568-3577.

Katzberg, R.W., T.W. Morris, F.A. Burgener, D.E. Kamm, and
H.W. Fischer (1977) Renal renin and hemodynamic
responses to selctive renal artery catheterization and
angiography. Invest. Radiol. 12: 381-388.

Katzberg, R.W., G. Schulman, L.G. Meggs, W.J.H. Caldicott,
M.M. Damiano, and N.K. Hollenberg (1983) Mechanism of
renal response to contrast media in dogs: decrease in
renal function due to hypertonicity. Invest. Radiol.
18: 74-80.

Kelley, G.G., H. Barron, K. Andreoni, W.S. Curtis, and J.N.
Forrest, Jr. A1 adenosine receptors inhibit chloride
transport in the shark rectal gland by N1 -coupled
cyclic AMP dependent and cyclic AMP independent
mechanisms. Kidney Int. 31: 274 (Abstr.).

168

Kelley, G.G., W.S. Curtis, A.M. Nuland, and J.N. Forrest
(1985) Endogenous adenosine inhibits chloride
secretion in the shark rectal gland via an A1
adenosine receptor coupled to the inhibitory
nucleotide regulatory protein. Bull. Mt. Desert Isl.
Biol. Lab. 25: 108-110. .

Kelley, G.G., D.R. Gifford, F. Wang, and J.N. Forrest
(1984) Stimulation and inhibition of adenylate cyclase
in the rectal gland. Bull. Mt. Desert Isl. Biol. Lab.
24: 102-104.

Kleinman, L.I., and E.P. Radford (1964) Ventilation
standards for small mammals. J. Appl. Physiol. 19:
360-362.

Kolassa, N., and K. Pfleger (1975) Adenosine uptake by
erythrocytes of man, rat, and guinea pig and its
inhibition by hexobendine and dipyridamole. Biochem.
Pharmacol. 24: 154-156.

Kowal, J., and R.F. Fiedler (1969) Adrenal cells in tissue
culture.II: Steroidogenic responses to nucleosides and
nucleotides. Endocrinology 84: 1113-1117.

Kramer, H.J., and H.C. Gonick (1970) Experimental Fanconi
syndrome. I. Effect of maleic acid on renal cortical
Na-K-ATPase activity and ATP levels. J. Lab. Clin.
Med. 76: 799-808.

Kramer, H.J., and H.C. Gonick (1973) Effect of maleic acid
on sodium-linked tubular transport in experimental
Fanconi syndrome. Nephron 10: 306-319.

Kraupp, O. (1969) Zum wirkungsmechanismus der
coronardilatoren. Arch. Exp. Pathol. Pharmakol. 263:

Kubler, W., and H.J. Bretschneider (1963) Die permeation
von adenosin durch die erythrocyten membran des
hundes. Pflugers Arch. 277: 141-149.

Kurtz, A., R. Della Bruns, J. Pfeilschifter, and C. Bauer
(1987) Role of cGMP as second messenger of adenosine
in the inhibition of renin release from renal
juxtaglomerular cells. Kidney Int. In Press.

169

Lang, M.A., A.S. Preston, J.S. Handler, and J.N. Forrest
(1985) Adenosine stimulated sodium transport in kidney
A6 epithelia in culture. Am. J. Physiol. 249:
C330-C336.

Larson, T.S., K. Hudson, J.I. Mertz, J.C. Romero, and F.G.
Knox (1983) Renal vasoconstrictive response to
contrast medium: the role of sodium balance and the
renin-angiotensin system. J. Lab. Clin. Med. 101:
385-391.

Londos, C., D.M.F. Cooper, W. Schlegel, and M. Rodbell
(1978) Adenosine analogs inhibit adipocyte adenylate
cyclase by a GTP-dependent process: basis for actions
of adenosine and methylxanthines on cyclic AMP
production and lipolysis. Proc. Natl. Acad. Sci. USA
75: 5362-5366.

Londos, C., D.M.F. Cooper, and J. Wolff (1980) Subclasses
of external adenosine receptors. Proc. Natl. Acad.
Sci. USA 77: 2551-2554.

Londos, C., and J. Wolff (1977) Two distinct
adenosine-sensitive sites on adenylate cyclase. Proc.
Natl. Acad. Sci. [USA] 74: 5482-5486.

Lund, P., N.W. Cornell, and H.A. Krebs (1975) Effect of
adenosine on the adenine nucleotide content and
metabolism of hepatocytes. Biochem. J. 152: 593-599.

Maesaka, J.K., and M. McCaffery (1980) Evidence for renal
tubular leakage in maleic acid-induced Fanconi
syndrome. Am. J. Physiol. 239: F507-F513.

Markwell, M.A.K., S.M. Haas, L.L. Bieber, and N.E. Tolbert
(1978) A modification of the Lowry procedure to
simplify protein determination in membrane and
lipoprotein samples. Anal. Biochem. 87: 206-210.

McDonald, F.D., T. Thiel, D.R. Wilson, G.F. DiBona, and
D.E. Oken (1969) The prevention of acute renal failure
in the rat by long-term saline loading: a possible
role of the renin-angiotensin axis. Proc. Soc. Exp.
Biol. Med. 131: 610-614.

McKenzie, 8.6., and H.P. Bar (1973) On the mechanism of
adenyl cyclase inhibition by adenosine. Can. J.
Physiol. Pharmacol. 51: 190-196.

170

Miller, W., R.A. Thomas, R.M. Berne, and R. Rubio (1978)
Adenosine production in the ischemic kidney. Circ.
Res. 43: 390-397.

Molski, T.F.P., P.H. Naccache, M.L. Marsh, J. Kermode, E.L.
Becker, and R.I. Sha'afi (1984) Pertussis toxin
inhibits the rise in the intracellular concentration
of free calcium that is induced by chemotactic factors
in rabbit neutrophils: possible role of "G proteins"
in calcium mobilization. Biochem. Biophys. Res.
Commun. 124: 644-650.

Moss, J., S.J. Stanley, D.L. Burns, J.A. Hsia, D.A. Yost,
G.A. Myers, and E.L. Hewlett (1983) Activation by
thiol of the latent NAD glycohydrolase and
ADP-ribosyltransferase activities of Bordetella
pertussis toxin. J. Biol. Chem. 258: 11879-11882.

Murray, R.D., and P.C. Churchill (1984) The effects of
adenosine receptor agonists in the isolated perfused
rat kidney. Am. J. Physiol. 247: H343-H348.

Nakamura, T., and M. Ui (1985) Simultaneous inhibitions of
inositol phospholipid breakdown, arachidonic acid
release, and histamine secretion in mast cells by
islet-activating protein, pertussis toxin. J. Biol.
Chem. 260: 3584-3593.

Newman, M.E., and A. Levitski (1983) Desensitization of
normal rat kidney cells to adenosine. Biochemical
Pharmacology 32: 137-140.

Osswald, H. (1975) Renal effects of adenosine and their
inhibition by theophylline. Naunyn-Schmeidebergs Arch.
Pharmacol. 288: 79-86..

Osswald, H. (1984) The role of adenosine in the regulation
of glomerular filtration rate and renin secretion.
Trends Pharmacol. Sci. 5:94-97.

Osswald, H., B. Evers, R. Massmann, and G. Nabakowski
(1980) Tissue content of adenine nucleotides and
adenosine in the rat kidney after administration of
maleate. Kidney Int. 17: 413 (Abstract).

 

171

Osswald, H., H.H. Hermes, and G. Nabakowski (1982) Role of
adenosine in signal transmission of tubuloglomerular
feedback. Kidney Int. 22: $136-$142.

Osswald, H., G. Nabakowski, and H. Hermes (1980) Adenosine
as a possible mediator of metabolic control of
glomerular filtration rate. Int. J. Biochem. 12:
263-267.

Osswald, H., H.J. Schmitz, and O. Heidenreich (1975)
Adenosine response of the rat kidney after saline
loading, sodium restriction, and hemorrhagia.
Pfluegers Arch. 357: 323-333.

Osswald, H., H.J. Schmitz, and R. Kemper (1977) Tissue
content of adenosine, inosine, and hypoxanthine in the
rat kidney after ischemia and postischemic
recirculation. Pfluegers Arch. 371: 45-50.

Osswald, H., H.J. Schmitz, and R. Kemper (1978) Renal
action of adenosine: effect on renin secretion in the
rat. Naunyn-Schmiedebergs Arch. Exp. Pathol.
Pharmakol. 303: 95-99.

Osswald, H., W.S. Spielman, and F.G. Knox (1978) Mechanisms
of adenosine- mediated decreases in glomerular
filtration rate in dogs. Circ Res 43: 465-469.

Palmer, G.C. (1971) Characteristics of the hormonal induced
cyclic adenosine 3',5'-monophosphate response in the
rat and guinea pig lung in vitro. Biochem. Biophys.
Acta 252: 561-566.

Paterson, A.R.P. (1979) Adenosine transport. In H.P. Beer
and G.I. Drummond (eds): Physiological and regulatory
functions of adenosine and adenine nucleotides. New
York: Raven Press, pp. 305-313.

Paterson, A.R.P., S.C. Kim, 0. Bernard, and C.E. Cass
(1975) Transport of nucleotides. Ann. NY Acad. Sci.
255: 402-410.

Pearson, J.D., J.S. Carleton, A. Hutchings, and J.L. Gordon
(1978) Uptake and metabolism of adenosine by pig
aortic endothelial and smooth-muscle cells in culture.
Biochem. J. 170: 265-271.

 

172

Phillis, J.W., and J.P. Edstrom (1976) Effects of adenosine
analogs on rat cerebral cortical neurons. Life Sci.
19: 1041-1053.

Plagemann, P.G.W. (1971) Nucleoside transport by Novikoff
rat hepatoma cells growing in suspension culture.
Specificity and mechanism of transport reactions and
relationship to nucleoside incorporation into nucleic
acids. Biochem. Biophys. Acta. 233: 688-701.

Plagemann, P.G.W., and D.P. Richey (1974) Transport of
nucleosides, nucleic acid bases, choline and glucose
by animal cells in culture. Biochem. Biophys. Acta.
344: 263-305.

Plagemann, P.G.W., and R.M. Wohlhueter (1980) Permeation of
nucleosides, nucleic acid bases, and nucleotides in
animal cells. In F. Bronner and A. Kleinzeller (eds):
Current topics in membranes and transport. New York:
Academic Press, pp. 225-330.

Poeschla, E., G. Kelly, and J.N. Forrest (1982) Evidence
for an inhibitory adenosine receptor in the rectal
gland of Squalus acanthias. Bull. Mt. Desert Isl.
Biol. Lab. 22: $19-$23.

Poeschla, E., G.G. Kelly, D. Gifford, and J.N. Forrest
(1985) Identification of adenosine receptor subtypes
in the shark rectal gland. Clin. Res. 33: 587A
(Abstr.).

Reed, J.R., R.H. Williams, and R.G. Luke (1983) The renal
hemodynamic response to diatrizoate in normal and
diabetic rats. Invest. Radiol. 18: 536-540.

Rodbell, M. (1980) The role of hormone receptors and
GTP-regulatory proteins in membrane transduction.
Nature 284: 17-22.

Rogulski, J., and A. Pacanis (1978) Effects of maleate on
CoA metabolism in rat kidney. In W. Guder and U.
Schmidt (eds): Nephrological Biochemistry. Huber:
Bern, pp. 406-415.

Romero, J.C., and G.G. Strong (1977) The effect of
indomethacin blockade of prostaglandin synthesis on
blood pressure of normal rabbits and rabbits with
renovascular hypertension. Circ. Res. 40: 35-41.

173

Roy, C. (1984) Regulation by adenosine of the
vasopressin-sensitive adenylate cyclase in pig kidney
cells (LLC-PK ) grown in defined media. Eur. J.
Biochem. 143: 543-250.

Sakai, K., J. Aono, and K. Haruta (1981) Species
differences in renal vascular effects of dipyridamole
and in the potentials of adenosine action by
dipyridamole. J. Cardiovasc. Pharmacol. 3: 420-430.

Sattin, A., and T.W. Hall (1970) The effect of adenosine
and adenine nucleotides on the cyclic adenosine
3',5'-phosphate content of guinea pig cerebral cortex
slices. Mol. Pharmacol. 6: 13-23.

Schatz, R.A., C.R. Vannam, and o.z. Sellinger (1977)
Species and tissue differences in the catabolism of
S-adenosyl homocysteine: a quantitative
chromatographic study. Life Sci. 20: 375-383.

Schnermann, J., H. Osswald, and M. Hermle (1979) Inhibitory
effect of methylxanthines on feedback control of GFR
in the rat kidney. Pfluegers Arch. 369: 39-48.

Schnermann, J., F.S. Wright, J.M. Davis, W. VonStackelberg,
and G. Grill (1970) Regulation of superficial nephron
filtration rate by tubuloglomerular feedback.
Pfluegers Arch. 318: 147-175.

Schrader, J., W. Schutz, and H. Bardenheuer (1981) Role of
S-adenosylhomocysteine hydrolase in adenosine
metabolism in mammalian heart. Biochem. J. 198: 65-70.

Schwabe, U., R. Ebert, and H.C. Erbler (1973) Adenosine
release from isolated fat cells and its significance
for the effects of hormones on cyclic 3',5'-AMP levels
and lipolysis. Arch. Pharmacol. 276: 133-148.

Scott, J.B., R.M. Daugherty, Jr., and F.J. Haddy (1965)
Role of chemical factors in regulation of flow through
the kidney, hindlimb, and heart. Am. J. Physiol. 208:
813-824.

Shimizu, H., and J.W. Daly (1970) Formation of adenosine
3',5'- monophosphate from adenosine in brain slices.
Biochem. Biophys. Acta. 222: 465-473.

174

Silinsky, E.M. (1981) On the calcium receptor that mediates
depolarization-secretion coupling at cholinergic motor
nerve terminals. Br. J. Pharmacol. 73: 413-429.

Sistare, F.D., R.A. Picking, and R.C. Haynes (1985)
Sensitivity of the response of cytosolic calcium in
quin-2-loaded rat hepatocytes to glucagon, adenine
nucleosides, and adenine nucleotides. J. Biol. Chem.
260: 12744-12747.

Sixma, J.J., J.P.M. Lips, A.M.C. Trieschnigg, and H.
Holmsen (1976) Transport and metabolism of adenosine
in human blood platelets. Biochem. Biophys. Acta. 443:
33-48.

Smith, C.D., C.C. Cox, and R. Snyderman (1986) Receptor
coupled activation of phosphoinositide-specific
phosphlipase C by an N protein. Science 232: 97-100.

Smith, W.L., and A. Garcia-Perez (1985) Immunodissection:
use of monoclonal antibodies to isolate specific types
of renal cells. Am. J. Physiol. 248: Fl-F7.

Souness, J.E., and V. Chagoya de Sanchez (1981) The
stimulation of [1-1‘C] glucose oxidation in isolated
fat cells by N°-methyladenosine: An effect independent
of cAMP. FEBS Letts. 125: 249-252.

Spielman, W.S. (1984) Antagonistic effect of theophylline
on the adenosine-induced decrease in renin release.
Am. J. Physiol. 247: F246-F251.

Spielman, W.S., and H. Osswald (1978) Characterization of
the post-occlusive response of renal blood flow in the
cat. Am. J. Physiol. 235: F286-F290.

Spielman, W.S., and H. Osswald (1979) Blockade of
postocclusive renal vasoconstriction by an
angiotensin II antagonist: evidence for an
angiotensin-adenosine interaction. Am J. Physiol. 237:
F463-F467.

Spielman, W.S., and C.I. Thompson (1982) A proposed role
for adenosine in the regulation of renal hemodynamics
and renin release. Am. J. Physiol. 242: F423-F435.

175

Spielman, W.S., S.L. Britton, and M.J. Fiksen-Olsen (1980)
Effect of adenosine on the distribution of renal blood
flow in dogs. Circ Res 46: 449-456.

Spielman, W.S., W.K. Sonnenburg, M.L. Allen, L.J. Arend, K.
Gerozissis, and W.L. Smith (1986) Immunodissection and
culture of rabbit cortical collecting tubule cells.
Am.J. Physiol. 251: F348-F357.

Spinowitz, 8.8., and J.A. Zadunaisky (1979) Action of
adenosine on chloride active transport of isolated
frog cornea. Am. J. Physiol. 237: F121-F127.

Stoff, J.S., R. Rosa, R. Hallac, P. Silva, and P.H. Epstein
(1979) Hormonal regulation of active chloride
transport in the dogfish rectal gland. Am. J. Physiol.
237: F138-Fl44.

Sullivan, T.J., and C.W. Parker (1976) Pharmacological
modulation of inflammatory mediator release by rat
mast cells. Am. J. Pathol. 85: 437-464.

Szczepanska, M., and S. Angielski (1980) Prevention of
maleate-induced tubular dysfunction by acetoacetate.
Am. J. Phsyiol. 239: F50-F56.

Tagawa, H., and A.J. Vander (1970) Effects of adenosine
compounds on renal function and renin secretion in
dogs. Circ Res 26: 327-338.

Talner, L.B., and A.J. Davidson (1968) Renal hemodynamic
effects of contrast media. Invest. Radiol. 3: 310-317.

Teitelbaum, I., and T. Berl (1986) Effects of calcium on
vasopressin-mediated cyclic adenosine monophosphate
formation in cultured rat inner medullary collecting
tubule cells. Evidence for the role of intracellular
calcium. J. Clin. Invest. 77: 1574-1583.

Thompson, C.I., H.V. Sparks, and W.S. Spielman (1985) Renal
handling and production of plasma and urinary
adenosine. Am. J. Physiol. 248: F545-F551.

Thurau, K. (1964) Renal hemodynamics. Am. J. Med. 36:
698-719.

Trams, E.G., and C.J. Lauter (1974) On the sidedness of
plasma membrane enzymes. Biochem. Biophys. Acta. 345:

176

Trimble, M.E., and R. Coulson (1984) Adenine transport in
perfused rat kidney and renal cortical membrane
vesicles. Am. J. Physiol. 15: F794-F803.

VanCalker, D., M. Muller, and B. Hamprecht (1979) Adenosine
regulates via two different types of receptors, the
accumulation of cAMP in cultured brain cells. J.
Neurosci. 33: 999-1005.

Vandogen, R., and W.S. Peart (1979) Calcium dependence of
the inhibitory effect of angiotensin on renin
secretion in the isolated perfused kidney of the rat.
Br. J. Pharmacol. 50: 125-129.

Verghese, M., R.J. Uhing, and R. Snyderman (1986) A
pertussis/ choleratoxin-sensitive N protein may
mediate chemoattractant receptor signal transduction.
Biochem. Biophys. Res. Commun. 138: 887-894.

Walter, P., and E. Bassenge (1968) Wirkung von ATP,
A-3,5-MP, adenosine, und dipyridamol um
streifenpraparaten der a. coronari, a. renalis, und
der v. portal. Pfluegers Arch. 299: 52-65.

Weir, G.C., S.D. Knowlton, and D.B. Martin (1975)
Nucleotide and nucleoside stimulation of glucagon
secretion. Endocrinology 97: 932-936.

Wolberg, G., T.P. Zimmerman, K. Hiemstra, M. Winston, and
L.C. (Hut (1975) Adenosine inhibition of
lymphocyte-mediated cytolysis: possible role of cyclic
adenosine monophosphate. Science 187: 957-959.

Wolff, J., and G.H. Cook (1977) Activation of
steroidogenesis and adenylate cyclase by adenosine
adrenal and Leydig tumor cells. J. Biol. Chem. 252:
687-693.

Woodcock, E.A., R. Loxley, E. Leung, and C.I. Johnston
(1984) Demonstration of R. adenosine receptors in rat
renal papillae. Biochem. Biophys. Res. Commun. 121:
434-440.

Wysocki, L.J., and G.L. Sato (1978) "Panning” for
lymphoyctes: a method of cell selection. Proc. Natl.