Nephrol Dial Transplant (1998) 13: 875–885
Nephrology
Dialysis
Transplantation
Original Article
Effects of vascular endothelial growth factor ( VEGF )/vascular
permeability factor ( VPF ) on haemodynamics and permselectivity of the
isolated perfused rat kidney
Bernd Klanke1, Matthias Simon2, Wolfgang Röckl3, Herbert A. Weich3, Hilmar Stolte1 and
Hermann-Josef Gröne2
1Exp. Nephrology, Department of Nephrology, Medical School Hannover, Hannover, 2Department of Pathology, University
of Marburg, Klinikum Lahnberge, Marburg, 3Department of Gene Expression, Gesellschaft für Biotechnologische
Forschung (GBF ), Braunschweig, Germany
Abstract
Background. Vascular endothelial growth factor
( VEGF ) or vascular permeability factor ( VPF ) is a
selective mitogen for endothelial cells; it increases
microvascular permeability and has been shown to
relax isolated canine coronary arteries by an endothelium-dependent mechanism. In many tissues
VEGF/VPF is expressed after an appropriate stimulus,
mostly hypoxia. In the kidney VEGF/VPF is constitutively expressed in glomerular podocytes and epithelia
of collecting duct. Glomerular and peritubular capillary endothelia also constitutively express specific
VEGF receptors. The in vivo function of renal
VEGF/VPF is unknown.
Method. In the present study the effects of human
recombinant VEGF
on renal haemodynamics and
165
glomerular permselectivity was investigated in the isolated perfused kidney of the rat.
Results. In kidneys preconstricted by noradrenaline
(NA 1.5×10−7 mol/l ) VEGF/VPF (155 pmol/l )
caused an almost complete return of renal perfusion
flow rate to pre-NA values (before NA 113±4%, after
NA 100%, 15 min with VEGF/VPF 111±4%). Shortly
after VEGF/VPF administration VEGF/VPF-induced
relaxation commenced, and became significant after
2 min (15 min with VEGF/VPF vs without VEGF/VPF
111±4% vs 103±2%; P<0.05). In the presence of the
NO-synthase inhibitor NW-nitro--arginine (-NNA;
5×10−5 mol/l ) VEGF/VPF caused only small, transient relaxations (before NNA 109±5%, after NNA
100%, 15 min with VEGF 95±2%). The cyclooxygenase inhibitor diclofenac failed to inhibit the
relaxing activity of VEGF/VPF (before NA 119±4%,
after NA+diclofenac 100%, 15 min with VEGF/VPF
123±5%).
VEGF demonstrated no significant increase in
Correspondence and offprint requests to: H.-J. Gröne MD,
Department of Pathology, Philipps University of Marburg, Klinikum
Lahnberge, D-35043 Marburg, Germany.
renal protein excretion rate (after NA pretreatment
(=100%): 12.5 min with VEGF/VPF vs without
VEGF/VPF: 119±10% vs 132±11%, n.s.) (after NNA
pretreatment (=100%) 12.5 min with VEGF/VPF vs
without VEGF/VPF 94±5% vs 96±4%; n.s.) or clearance quotient of albumin.
Glomerular filtration rate was not influenced by
VEGF/VPF in kidneys pretreated with NA (before
NA 105±5%, after NA 100%, 12.5 min with
VEGF/VPF 94±2%) or with NNA (before NNA
107±6%, after NNA 100%, 12.5 min with VEGF/VPF
96±2%). Fractional glucose and fractional sodium
excretion showed flow-dependent changes.
Conclusion. VEGF/VPF can contribute to the relaxing
capacity of the renal vasculature. This relaxation is
partly mediated by the NO/endothelium-derived
relaxing factor ( EDRF ) pathway. In the isolated
perfused rat kidney the glomerular permeability for
albumin is not affected by VEGF/VPF.
Key words: vascular endothelial growth factor; vascular
permeability factor; endothelium-dependent vasodilatation; endothelium-derived relaxing factor; glomerular
permeability; NW-nitro--arginine
Introduction
Vascular endothelial growth factor ( VEGF ), a selective
endothelial mitogen, was first described by Ferrara and
Henzel [1]. cDNA cloning revealed that VEGF is
encoded by the same gene as the vascular permeability
factor ( VPF ) [2–4]. As a result of alternative splicing
at least four transcripts have been detected, which
encode VEGF isoforms containing 121, 165, 189 or
206 amino-acid residues in man [5]. In addition to its
endothelial proliferative activity VEGF/VPF increases
microvascular permeability, protease activity, and
chemotaxis of monocytes. VEGF/VPF is expressed in
© 1998 European Renal Association–European Dialysis and Transplant Association
876
numerous rodent and human tumour cells [6,7]. With
regard to kidney and renal disease VEGF expression
has been found in activated macrophages [8,9], renal
glomerular visceral epithelial and epithelia of collecting
duct [10,11]. Mesangial cells [9], glomerular endothelial cells [12], and vascular smooth-muscle cells in
culture [13] also have been shown to synthesize
VEGF/VPF.
Receptors for VEGF appear to be expressed specifically by endothelial [11,14,15] and haematopoietic
cells [16,17]. In human kidneys VEGF/VPF receptors
are detectable by radiolabelled VEGF on pre-and
postglomerular vessels and more distinctly on glomerular capillaries. Postglomerular capillaries and venous
vessels have receptors with an affinity for VEGF/VPF
higher than for other renal endothelial cells (Simon
et al., unpublished observations).
VEGF/VPF expression persists in glomerular visceral epithelia and in the epithelia of the collecting
duct from the fetal to the adult kidney, although the
adult kidney exhibits a very low endothelial proliferative activity [10,11]. The expression of VEGF receptors
on quiescent endothelium of the adult kidney [11]
suggests that VEGF/VPF might have functions other
than mediating endothelial growth.
The functions of VEGF/VPF in normal renal physiology and in renal diseases are not known. In other
organs VEGF is active in increasing microvascular
permeability presumably by affecting intercellular junctions between endothelial cells [18], the fenestration of
these cells [19], or the expression of proteases [20]. It
has been postulated that VEGF/VPF may also influence the permselective properties of the glomerular
filtration barrier, which can be altered by structural or
compositional changes in the capillary wall but also
on the basis of glomerular haemodynamic alterations
[21]. Ku et al. [22] have reported that VEGF/VPF
can relax coronary arteries and thereby influence coronary haemodynamics.
The present study was designed to investigate both
the effects of VEGF/VPF on renal haemodynamics
and on glomerular permselectivity in the isolated perfused rat kidney. The results described in this article
demonstrate that VEGF/VPF increased the renal perfusion flow but did not influence the glomerular filtration rate or the permselectivity of the glomerular
barrier wall in the isolated perfused rat kidney.
Vascular dilatation by VEGF/VPF was dependent on
the vascular nitric oxide (NO) system but not on
prostaglandin synthesis.
Subjects and methods
B. Klanke et al.
100 mg/kg body-weight i.p., Byk Gulden, Konstanz,
Germany).
Isolated perfused kidney
The right kidney was isolated and perfused as described in
detail by Radermacher et al. [23]. Briefly, the right suprarenal
artery was ligated and the right ureter was cannulated with
a short polypropylene catheter. Anticoagulation was achieved
by injection of 100 U/100 g body-weight of heparin. The left
renal artery and the superior mesenteric artery were ligated.
The aorta was cannulated caudal to the right renal artery
and the perfusion was started with constant flow
(8–10 ml/min). The aorta was ligated cranial to the right
renal artery, which interrupted the blood supply to the
kidney. Three to five minutes after starting the perfusion,
recirculation was established and the effective perfusion
pressure was held constant at 100 mmHg [24].
Perfusion flow rate was measured with an electromagnetic
flowmeter (Statham, SP-2202, Spectramed Inc., Oxnard, CA,
USA) at the cannulated venous outflow. Oxygen partial
pressure was measured via an arterial and a venous bypass
near the kidney, alternatively drawn through a Clark type
electrode. All perfusion pathways and the kidney chamber
were kept at a constant temperature of 37.5°C. The basic
perfusion medium consisted of a substrate enriched
Krebs–Henseleit buffer containing 0.01 U/l arginine–vasopressin and 5% (w/v) of predialysed bovine serum albumin
(for details see [23]). Polyfructosan (1 g/l ) was added for
determination of the glomerular filtration rate [24,25]. The
perfusate (150 ml )—containing albumin—was dialysed
throughout the experiment against a 33-fold larger volume
of dialysate (Ca2+ 1.2 mM, Mg2+ 0.9 mM in basic perfusion
medium devoid of albumin). The commercial dialyser
(LunDiaA Alpha 400 Cuprophan membrane, cut-off approximately 5000, Gambro, Lund, Sweden) was handled according
to the instructions of the manufacturer. The perfusate volume
losses—caused by ultrafiltration during dialysis—were automatically balanced by the addition of dialysate. The dialysate
was oxygenated with a prewarmed and moistened gas mixture
(95% O /5% CO ]. The pH was 7.37–7.43 in the prewarmed
2
2
solution. All solutions were filter sterilized on the day of
experiment, all glass and metalware utensils were heat sterilized (130°C, 4 h) and all tubes were gas sterilized (ethylene
oxide).
After starting the perfusion, an equilibrium period of
30 min was allowed before the first urine was collected. Urine
was continuously collected in 5-min intervals until minute
120. Vasoconstrictors were freshly dissolved in buffer and
added to the dialysate. Vascular endothelial growth factor
(recombinant VEGF (Bacculo virus system) a 43-kDa homodimeric protein consisting of two 165 amino-acid polypeptide
chains, lyophilized with no additives (Department of Gene
Expression, GBF, Braunschweig, Germany, [26 ]) was dissolved in 500 ml perfusion medium and added to the perfusate. The cut-off of the dialyser restricted the recombinant
VEGF to the perfusate volume.
165
Animals
Experimental groups
Male Sprague–Dawley rats (SPRD/Ztm strain; breed of
Central Animal Facilities, Medical School Hannover) were
used. Animals had free access to food (AltrominA standard
pellet chow 1314, Altromin, Lage, Germany) and tap
water. The rats (n=26, weighing 218–385 g, median 286 g),
were anaesthetized with thiopental-sodium (TrapanalA,
Group Ia (noradrenaline, n=5). After 40 min of perfusion,
the kidneys were preconstricted with 1.5×10−7 mol/l noradrenaline (NA). The NA concentration was comparable to
that used in experiments by Radermacher et al. [27]. The
administration of noradrenaline partly substituted for the
VEGF/VPF and renal haemodynamics
lost vasotonus after acute interruption of sympathetic
innervation of the isolated kidney.
Group Ib (noradrenaline plus VEGF/VPF, n=5). As in group
Ia 40 min after starting the perfusion, kidneys were
preconstricted with 1.5×10−7 mol/l NA, 20 min later 1 mg
VEGF (155 pmol/l ) was added. The VEGF concentration
was chosen according to data on the effects of VEGF on
isolated PGF preconstricted coronary arteries showing a
2a
half-maximal dilatation of the artery with this concentration
[22]. A dose–response curve was not established for the
isolated kidney because of the limited supply of recombinant
VEGF available.
Group IIa (NW-nitro--arginine, n=3). To study the possibility that VEGF/VPF-induced effects were mediated by the
endothelium-derived relaxing factor, kidneys were preconstricted by the NO-synthase inhibitor NW-nitro--arginine
(-NNA, 5×10−5 mol/l ), added at 40 min of perfusion. This
group IIa served as a control for group IIb. The NNA
concentration was chosen after results obtained previously
[23], demonstrating effective inhibition of endothelial constitutive NO synthesis.
Group IIb (-NNA plus VEGF/VPF, n=6) Twenty minutes
after administration of -NNA (5×10−5 mol/l at minute
40), 1 mg VEGF/VPF (155 pmol/l ) was added.
Group IIIa (noradrenaline plus diclofenac plus VEGF, n=3).
Because prostaglandins are thought to be involved in the
regulation of renal haemodynamics we tested the hypothesis
that the effects of VEGF/VPF may be mediated by prostaglandins. The same design as in group IIb was chosen. Ten
minutes after noradrenaline and 10 min before VEGF/VPF
the cyclo-oxygenase inhibitor sodium-diclofenac (10 mmol/l )
was added to the perfusion system. For this dose of diclofenac
Radermacher et al. [27] demonstrated suppression of renal
prostaglandin synthesis by about 90%.
Group IIIb (noradrenaline plus -NNA plus VEGF/VPF, n=
4). Additionally to group IIb this set of experiments was
carried out to test whether VEGF/VPF may relax renal
vasculature by a second mechanism besides the NO/EDRF
pathway. After NA preconstriction (1.5×10−7 mol/l, minute
40) the NO/EDRF synthesis was inhibited by -NNA
(5×10−5 mol/l, minute 50). VEGF (155 pmol/l ) was given
at minute 60.
Analytical methods
Glucose was measured enzymatically by the hexokinase/
glucose-6-phosphate dehydrogenase method, and polyfructosan was measured after acid hydrolysis by including glucose6-phosphate isomerase reaction in the assay [25]. The polyfructosan clearance was used as an estimate of the glomerular
filtration rate [24]. Sodium and potassium were analysed by
System E2A electrolyte analyser (Beckman Instruments
GmbH, München, Germany). Fractional sodium and glucose
reabsorption was calculated as the fractional relationship of
excreted to filtered quantities.
Albumin was measured using a protein-dye binding
method with Coomassie-blue and bovine serum albumin as
a standard [28]. In this model of isolated perfused kidney
albumin was the predominant protein in perfusate and urine
(>95%, determined by microdisc electrophoresis and a
double-beam microdensitometer (3CS, Joyce–Loebl Ltd.,
Gateshead, UK )). Clearance quotient of albumin was calculated as the relationship of albumin filtration rate and
glomerular filtration rate.
877
Oxygen supply was calculated from arterial oxygen pressure, renal perfusion rate, and kidney weight. Oxygen consumption was calculated from the difference of arteriovenous
oxygen pressure, renal perfusion rate, and kidney weight.
Urine flow rate was determined gravimetrically.
Materials
The enzymes used for glucose analysis were supplied by von
Minden (Moers, Germany). Glucose-6-phosphate isomerase
and sodium pyruvate was purchased from Boehringer
(Mannheim, Germany). Urea, -glutamine, d-ketoglutaric
acid, malic acid and creatinine were obtained from Merck
(Darmstadt, Germany), sodium lactate from Serva
(Heidelberg, Germany), neomycin sulphate from Byk Gulden
( Konstanz, Germany), and glucose (GlucosterilA 50%) from
Fresenius (Bad Homburg, Germany). The mixture of
-amino acids (AminoplasmalA 5%E ) was purchased from
Braun (Melsungen, Germany). Polyfructosan (InutestA) was
obtained from Laevosan (Linz, Austria), bovine serum albumin (fraction V powder, A-2153), NW-nitro--arginine
(N-5501), noradrenaline (A-7256), and diclofenac (D-6899)
from Sigma Chemie (Deisenhofen, Germany), and arginine–
vasopressin (PitressinA) from Parke–Davis (Berlin,
Germany).
Calculations and statistical analysis
All results are expressed for 1 g wet kidney weight of the
remaining left decapsulated kidney. Determination of unperfused left and unperfused right kidney weights had shown
that kidney weights did not differ significantly ( left kidney
1.16±0.04 g vs 1.1 3±0.03 g right kidney, n=20).
All data are reported as means±SEM. Levels of significance were calculated using the t test (two-tailed, unpaired)
of the computer program SPSS/PC+ Advanced Statistics
4.0 (SPSS Inc., Chicago, IL, USA). A value of P<0.05 was
considered significant.
Results
Renal haemodynamics
Renal perfusion flow rate. To test possible vasodilating
effects of VEGF/VPF the nearly completely dilated
isolated perfused kidney was preconstricted with NA
(1.5×10−7 mol/l ). After addition of NA the renal perfusion flow rate (RPF ) decreased rapidly by about 10%.
VEGF/VPF (155 pmol/l=6.67 ng/ml ) caused a significant increase in the RPF compared to kidneys without
VEGF (Figure 1A). The RPF of NA-preconstricted
kidneys was restored by VEGF to flow rates comparable
to those in the unconstricted situation.
RPF decreased slowly by about 5% following administration of NW-nitro-L-arginine (-NNA, 5×10−5
mol/l ) ( Figure 1B). Only a small and transient increase
of RPF (not shown in Figure 1B) could be obtained
by the addition of VEGF/VPF (155 pmol/l ). The same
result was obtained when the kidneys were preconstricted with a combination of NA and -NNA (group
IIIb, NA+-NNA+VEGF/VPF; Figure 1C ).
NA plus diclofenac (group IIIa, NA+Diclo+
VEGF/VPF; Figure 1C ) did not inhibit the VEGF/
878
B. Klanke et al.
VPF-induced vasodilatation as seen after NA plus NNA ( Figure 1C ). The glomerular filtration rate was
not influenced by VEGF/VPF either after NA or after
NNA preconstriction (Figure 2). As a result of the
increase of RPF and no change of GFR, the filtration
fraction decreased after VEGF when preconstricted
with NA ( Table 1). VEGF/VPF had no influence on
the filtration fraction in the presence of -NNA
( Tables 2,3).
Effects of VEGF on renal excretion
Urine flow rate ( UFR) ( Tables 1–3). VEGF/VPF had
no significant influence on the UFR after preconstriction with NA ( Table 1) or after the addition of
-NNA or diclofenac ( Tables 2,3).
Fractional sodium excretion (%NaEx) ( Tables 1–3).
The fractional sodium excretion showed no significant
difference between the groups. VEGF/VPF led to a
decline of the fractional sodium excretion while the
kidneys demonstrated an elevation of %NaEx during
NA administration. This effect was independent of NO
synthase or cyclo-oxygenase inhibition ( Table 3).
Fractional glucose excretion (%GluEx) (Tables 1–3).
For the first 75 min of perfusion the fractional glucose
excretion was stable. Thereafter the %GluEx increased
without relation to VEGF/VPF administration.
Effects of VEGF/VPF on protein handling
Protein excretion rate and clearance quotient of albumin ( Tables 1–3). The addition of NA decreased
protein excretion rate and the clearance quotient of
albumin. Thereafter, both parameters increased with
perfusion time, but independently from VEGF/VPF
administration ( Table 1). In the presence of -NNA
protein excretion rate and albumin clearance demonstrated a nearly unchanged course without any influence of VEGF on these parameters ( Table 2). When
NE preconstriction was combined with an inhibition
of cyclo-oxygenase or an inhibition of NO synthase an
influence of VEGF/VPF was not detectable with reference to protein excretion and clearance quotient of
albumin ( Table 3).
Effects of VEGF/VPF on oxygen consumption
Fig. 1. A. Renal perfusion rate after NE preconstriction
(1.5×10−7 M ) with or without VEGF/VPF (155 pM ). VEGF/VPF
induces vasodilation of the renal vasculature. Dilation started almost
immediately on VEGF/VPF administration and reached a maximum
at 9–10 min in the presence of VEGF/VPF. Values (means±SEM )
are expressed in % of absolute RPF measured 1 min prior to
VEGF/VPF administration (=minute−1). B. Renal perfusion rate
under NOS inhibition (-NNA, 5×10−5 M ) with or without
VEGF/VPF (155 pM ). VEGF/VPF induces only a small and transient vasodilation of the renal vasculature. Values (means±SEM ) are
expressed in per cent of absolute RPF measured 1 min prior to
VEGF/VPF administration (=minute−1). C. Renal perfusion rate
after VEGF/VPF (155 pM ) under NOS (-NNA, 5×10−5 M ) or
Oxygen consumption (Tables 1–3). In all groups a
time-dependent slow decline was measured for this
parameter which became clearer when VEGF/VPF
was given.
Oxygen supply/oxygen consumption. The relationship
of oxygen supply to oxygen consumption was
cyclooxygenase (sodium-diclofenac, 10 mM ) inhibition combined
with NE preconstriction (1.5×10−7 M ). Under NOS inhibition
VEGF/VPF induced only a small and transient vasodilation of the
renal vasculature. Cyclooxygenase inhibition did not inhibit the
VEGF/VPF-induced relaxations of the renal vasculature. Values
(means±SEM ) are expressed in per cent of absolute RPF measured
1 min prior to VEGF/VPF administration (=minute−1).
VEGF/VPF and renal haemodynamics
879
Fig. 2. Changes of renal functional parameters with or without administration of VEGF. Presented bars depict the differences between
data of post-VEGF period (minute 65–70) and of pre-VEGF period (minute 55–60) or eight selected renal functional parameters under
NE preconstriction (NE; left) or NOS inhibition (NNA; right). VEGF/VPF (+VEGF ) or no VEGF/VPF (−VEGF ) was given at minute
60. The asterisks indicate significant differences (P<0.05 in the Mann–Whitney U test).
decreased following NA treatment because of the
reduced delivery of oxygen by the decreased flow rate.
This relation went up when VEGF/VPF increased the
flow rate. VEGF/VPF had no influence on this
parameter when -NNA inhibited the renal dilatation.
Sodium reabsorption/oxygen consumption (Tables 1–3).
The relationship of sodium reabsorption to oxygen
consumption was unaffected by any pharmacological
intervention. A time-dependent decline was noted, as
described for the isolated perfused kidney.
Discussion
The functions that constitutively expressed VEGF may
have in renal physiology and in renal diseases have
not been investigated. VEGF/VPF can influence vascular permeability and endothelial cell function [6 ].
It is likely that as in other organs VEGF acts on
renal endothelial cells and monocytes, both of which
exhibit specific receptors for VEGF [11,17]. This action
will probably occur by a paracrine loop as vascular
smooth muscle cells and glomerular and tubular epithelia have been demonstrated to synthesize VEGF
[10,11,13]. The infusion of recombinant VEGF into
the isolated kidney was taken as a model system to
analyse these in vivo paracrine effects of VEGF. In this
study we chose the isolated perfused rat kidney model
instead of in vivo VEGF/VPF infusions to prevent
extrarenal counter-regulation and interactions of
VEGF/VPF with blood cells or plasma components.
Stable renal function was obtained by dialysis of the
perfusion medium.
Groups
NA
NA+VEGF
NA
NA+VEGF
NA
NA+VEGF
NA
NA+VEGF
NA
NA+VEGF
NA
NA+VEGF
92.5±6.7
81.6±5.4
91.4±4.3
93.3±5.7
85.1±7.1
72.8±8.2
86.1±4.2
89.3±8.2
100.0±2.1
94.1±1.0
201.4±23.8
158.8±14.3
94.6±4.8
89.8±3.8
98.7±2.2
99.7±2.6
90.9±4.2
85.7±4.5
86.7±1.8
90.8±4.4
100.0±0.9
99.0±1.3
109.0±17.0
139.2±42.4
45–50
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
55–60
±VEGF (60)
96.0±4.7
102.1±5.4
94.5±1.4
84.5±2.4
101.2±3.4
113.2±6.7
140.8±10.2
162.3±14.5
96.5±0.6
98.3±1.9
137.7±18.2
130.5±1.1
75–80
88.1±6.8
95.1±9.2
86.1±2.9
81.7±3.0
102.7±5.4
114.2±9.2
326.3±66.5
419.4±76.4
94.0±0.9
94.5±1.7
178.4±31.1
201.6±49.3
95–100
72.1±7.8
74.5±14.1
78.7±3.5
70.9±9.6
92.2±6.9
104.2±10.3
510.6±114.5
691.7±171.7
91.9±1.3
90.5±2.2
224.4±50.5
352.1±166.0
115–120
158.9±22.4 (ml/min/g kidney)
109.5±10.8 (ml/min/g kidney)
4.1±0.3 (%)
2.9±0.3 (%)
14.4±1.8 (%)
14.3±2.2 (%)
1.3±0.4 (%)
1.8±0.9 (%)
6.3±0.7 (mmol/min/g kidney)
6.0±0.4 (mmol/min/g kidney)
0.7±0.1×10−3
0.9±0.5×10−3
Absolute value of period 55–60
Values are means±SE in percent of period 55–60; n=5 rats for NA group, n=5 for NA+VEGF group.
NA, noradrenaline; VEGF, vascular endothelial growth factor.
%UFR, % urine flow rate; %FF, % filtration fraction; %Fract.NaEx, % fractional sodium excretion; %Fract.GluEx, % fractional glucose excretion; %O use, % oxygen consumption; %AlbClearQuot,
2
% albumin clearance quotient.
%AlbClearQuot
%O use
2
%Fract.GluEx
%Fract.NaEx
%FF
Parameters
%UFR
35–40
NA (40)
Perfusion period and administration time of drugs (min)
Table 1. Effect of VEGF after NA preconstriction on the functional parameters of the isolated perfused rat kidney
880
B. Klanke et al.
Groups
NNA
NNA+VEGF
NNA
NNA+VEGF
NNA
NNA+VEGF
NNA
NNA+VEGF
NNA
NNA+VEGF
NNA
NNA+VEGF
82.6±7.8
90.7±9.3
98.4±1.7
99.4±2.5
78.7±7.8
75.9±6.7
110.9±11.5
91.7±7.3
97.4±1.6
103.0±3.1
107.4±9.0
104.2±7.9
94.0±4.3
99.9±5.9
96.2±1.9
96.8±1.2
93.2±5.8
92.2±3.9
99.9±3.9
94.2±3.7
100.4±0.6
103.6±2.0
100.8±6.9
115.7±9.1
45–50
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
55–60
±VEGF (60)
105.6±7.0
95.9±2.2
101.7±1.8
102.4±2.9
113.1±8.0
107.7±4.0
114.9±6.1
167.2±16.8
95.9±1.8
95.9±1.5
105.4±7.5
94.8±5.5
75–80
107.8±11.5
86.5±5.2
101.3±1.6
104.6±3.6
122.0±12.9
109.3±6.3
178.3±16.2
286.4±46.1
95.9±2.0
91.7±3.1
97.8±17.8
106.3±13.2
95–100
102.1±16.9
71.8±7.2
99.2±1.4
100.1±4.6
122.3±17.6
90.5±18.7
289.9±43.9
444.1±80.6
93.2±2.1
86.2±4.0
98.8±12.8
127.7±27.2
115–120
126.1±15.5 (ml/min/g kidney)
105.5±13.3 (ml/min/g kidney)
2.8±0.2 (%)
3.2±0.3 (%)
18.0±3.8 (%)
13.1±2.4 (%)
2.3±0.4 (%)
2.1±0.7 (%)
5.6±0.6 (mmol/min/g kidney)
5.5±0.3 (mmol/min/g kidney)
3.9±0.9×10−3
4.7±2.0×10−3
Absolute value of period 55–60
Values are means±SE in percent of period 55–60; n=6 rats for NNA group, n=6 for NNA+VEGF group.
NNA, N-nitro--arginine; VEGF, vascular endothelial growth factor.
%UFR, % urine flow rate; %FF, % filtration fraction; %Fract.NaEx, % fractional sodium excretion; %Fract.GluEx, % fractional glucose excretion; %O use, % oxygen consumption; %AlbClearQuot,
2
% albumin clearance quotient.
%AlbClearQuot
%O use
2
%Fract.GluEx
%Fract.NaEx
%FF
Parameters
%UFR
35–40
NNA (40)
Perfusion period and administration time of drugs (min)
Table 2. Effect of VEGF under NOS inhibition on the functional parameters of the isolated perfused rat kidney
VEGF/VPF and renal haemodynamics
881
Groups
NA+Diclo+VEGF
NA+NNA+VEGF
NA+Diclo+VEGF
NA+NNA+VEGF
NA+NNA+VEGF
NA+Diclo+VEGF
NA+NNA+VEGF
NA+Diclo+VEGF
NA+Diclo+VEGF
NA+NNA+VEGF
NA+Diclo+VEGF
NA+NNA+VEGF
57.3±8.1
62.7±13.8
66.4±4.9
68.1±2.8
79.9±15.0
78.3±15.2
80.2±2.9
89.2±6.6
78.2±2.2
81.2±1.6
130.7±11.2
143.3±36.5
87.0±15.9
83.7±3.1
112.4±5.6
98.7±1.8
90.9±2.5
89.7±12.0
93.0±2.2
102.9±8.7
92.8±1.6
95.0±0.6
85.2±12.8
97.1±7.9
45–50
Diclo or
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
100.0±0.0
55–60
±VEGF (60)
75.0±12.8
84.9±9.3
72.3±6.9
99.2±10.3
96.6±6.0
81.9±1.2
177.8±28.3
148.1±14.1
93.7±5.4
91.2±0.9
112.6±9.7
130.9±16.1
75–80
73.5±15.1
79.5±14.1
61.4±4.3
94.4±12.4
109.6±12.4
103.0±9.9
287.0±54.0
257.3±11.0
88.0±5.5
80.6±1.9
198.3±23.9
172.2±30.3
95–100
72.2±15.1
71.7±13.8
59.7±7.9
92.0±12.1
114.4±13.4
112.5±9.1
365.8±53.1
388.0±67.3
84.2±5.3
76.0±1.6
180.8±93.5
191.6±32.2
115–120
224.9±41.1 (ml/min/g kidney)
217.4±18.9 (ml/min/g kidney)
6.0±1.5 (%)
5.4±0.7 (%)
14.4±1.1 (%)
15.8±2.3 (%)
2.0±0.3 (%)
2.3±0.3 (%)
6.7±0.7 (mmol/min/g kidney)
7.0±0.5 (mmol/min/g kidney)
1.4±0.2×10−3
1.5±0.3×10−3
Absolute value of period 55–60
Values are means±SE in percent of period 55–60; n=3 rats for NA+Diclo+VEGF group, n=4 for NA+NNA+VEGF group.
NA, noradrenaline; Diclo, diclofenac; NNA, N-nitro--arginine; VEGF, vascular endothelial growth factor.
%UFR, % urine flow rate; %FF, % filtration fraction; %Fract.NaEx, % fractional sodium excretion; %Fract.GluEx, % fractional glucose excretion; %O use, % oxygen consumption; %AlbClearQuot,
2
% albumin clearance quotient.
%AlbClearQuot
%O use
2
%Fract.GluEx
%Fract.NaEx
%FF
Parameters
%UFR
35–40
NA (40)
NNA (50)
Perfusion period and administration time of drugs (min)
Table 3. Effect of VEGF after NA preconstriction combined with NNA or diclofenac administration on the functional parameters of the isolated perfused rat kidney
882
B. Klanke et al.
VEGF/VPF and renal haemodynamics
Previous studies have shown an endothelium
dependent relaxing effect of VEGF/VPF on isolated
canine coronary arteries [22]. In the present study we
could demonstrate a rapid dilatation of the preconstricted vasculature of the isolated perfused rat kidney
shortly after VEGF/VPF administration.
It can be assumed that the vasodilatory effect is
mediated by constitutively expressed endothelial receptors for VEGF/VPF [11]. The interaction between
VEGF/VPF and its receptors leads to a cytosolic
increase of calcium in endothelial cells [29]. As a
cytosolic increase of calcium can release endotheliumderived relaxing factor ( EDRF ) [30] we investigated
the relationship between renal VEGF/VPF relaxation
and the endothelial dependent vasodilatation of
the EDRF/NO pathway. After administration of
NW--nitro-arginine (-NNA), a NO-synthase inhibitor for constitutive and inducible NO synthase,
VEGF/VPF lost most of its ability to increase the
renal flow, demonstrating the dependence of the
VEGF/VPF-related vasodilatation on the EDRF/NO
pathway. The remaining small transient dilatation may
be explained by an incomplete NO-synthase inhibition
with -NNA or by another relaxing pathway induced
by VEGF/VPF. C-type natriuretic peptide is released
from vascular endothelium and is discussed as an
additional endothelial regulator of local vascular tone
[31]. The experiments with the cyclo-oxygenase inhibitor diclofenac showing no effect on VEGF/VPFinduced dilatation exclude prostaglandins as major
mediators of the relaxation induced by VEGF/VPF. It
should be stressed that the vasodilatory effects of
VEGF were determined in a cell-free perfusion system.
As the VEGF-induced vasodilatation seemed to
depend mainly on the action of endothelial NO on
vascular smooth muscle without blood constituents in
between, a cell (erythrocytes)-containing perfusate
would probably give the same effects.
Another known in vivo action of VEGF/VPF is an
increase in microvascular fluid and protein extravasation [3,32,33]. On a molar basis, VEGF is 50 000 times
more potent than histamine [32]. Three mechanism
have been implied: (1) functional activation of vesicular-vacuolar organelles in the cytoplasm of endothelia
lining venules, (2) induction of interendothelial gaps,
and (3) induction of endothelial fenestrations [13,19].
In the kidney the VEGF/VPF-induced microvascular
hyperpermeability may have special relevance because
the glomerular capillary wall has to allow the crossing
of a high volume flow while at the same time maintaining a low escape of plasma proteins. It is tempting
to speculate that VEGF/VPF may influence the glomerular protein permeability.
The intimate anatomical relationship between the
glomerular visceral epithelial cells which synthesize
VEGF and the glomerular endothelium may promote
an effect of VEGF on glomerular endothelium and
glomerular permselectivity; secreted VEGF only needs
cross the intervening glomerular basement membrane
to reach its receptors on the glomerular endothelial
cells.
883
In the kidney the major filtration barrier for macromolecules is thought to be the glomerular basement
membrane [34] and the negatively charged glycocalyx
of the endothelial and epithelial cells. One may speculate that VEGF/VPF may influence the composition
of the glomerular basement membrane by its ability to
induce protease activity [20].
Farquhar et al. [35] suggested a role for all three
major components of the glomerular capillary wall in
the filtration process: the basement membrane as a
filter, the epithelium as its monitor, and the fenestrated
endothelium as a valve. In this concept VEGF/VPF,
produced in the epithelium and affecting the endothelium, may serve as a mediator. Our data demonstrate
that contrary to findings in capillaries and venules of
other organs such as skin or muscle the capillary
permeability of the glomerulus was not influenced by
VEGF/VPF in the isolated perfused rat kidney model.
Although it may be pointed out that the effects of
VEGF on the glomerular basement membrane may
require more time than was allowed in the current
experiment, VEGF, injected into subcutaneous tissue,
increased vascular permeability and altered endothelial
morphology within minutes [19].
The intact morphology of the glomerular cells and
an unaltered distribution of glomerular polyanions,
documented in the isolated perfused rat kidney [36 ],
indicates that the size-selective and charge-selective
properties of the glomerular capillary wall are
unchanged in this experimental model. Nevertheless,
there are some deviations from the in vivo situation:
the experimental protein excretion rate is 10 times
higher than that of a kidney in situ. This is explained
by the loss of trapped macromolecules inside the
glomerular filtration barrier during the starting phase
of kidney perfusion [37]. These components, normally
obtained from blood, may function as repair material
for morphological defects of the glomerular filtration
barrier. It is speculative that this lack of proteins in
the glomerular capillary wall may have been the reason
that VEGF/VPF failed to enhance the protein excretion rate above the level of proteinuria seen in this
model. Angiotensin II is known to induce proteinuria
in vivo but it also has no further effect on proteinuria
in a cell-free perfusion system [38]. On the other hand
the protein excretion rate of the isolated perfused
kidney was not maximal, as demonstrated by the timedependent increased protein loss with or without
VEGF/VPF. Moreover, in vivo infusions of VEGF
into the renal artery of rats revealed no increase in
protein excretion rate, while mean arterial pressure
decreased following administration of VEGF (B.
Klanke, H. J. Gröne, unpublished observations).
To explain the unchanged protein excretion rate
during VEGF/VPF administration, it probably also has
to be considered that the glomerular endothelium per se
is fenestrated in contrast to the endothelium in skin and
muscle. Roberts and Palade [19] observed that
VEGF/VPF could induce fenestration in continuous
endothelia of the dermis and thereby could modulate
the endothelial barrier function. In the glomerular endo-
884
thelia the expansion of fenestrae seems maximal. These
fenestrae expose 30–40% of the glomerular basement
membrane [35]. Moreover only few diaphragms can be
observed within the fenestrae of the glomerular endothelium [39]. It is conceivable that endogenous VEGF/VPF
exerted a maximal effect on the glomerular endothelium
(maintenance of fenestration) which was not enhanced
by exogenous VEGF/VPF.
In summary, our data demonstrated the capability
of VEGF/VPF to relax the renal vascular bed. This
dilating effect was mainly mediated by the EDRF/NO
pathway. Because hypoxia is a known potent stimulus
for VEGF/VPF gene expression, VEGF/VPF-mediated acute relaxation may contribute to a better supply
of oxygen and substrates to hypoxic tissue areas and
enhance hypoxic tolerance. Contrary to capillaries and
venules of skin, muscle, and mesentery, the glomerular
capillaries showed no increased permeability to
VEGF/VPF.
Acknowledgements. We thank Dr H.-J. Schurek for careful reading
and helpful discussion of the manuscript. The study was supported
by a grant from the Deutsche Forschungsgemeinschaft (DFG): Gr
728/5–1 to H.-J. Gröne.
B. Klanke et al.
14.
15.
16.
17.
18.
19.
20.
21.
References
22.
1. Ferrara N, Henzel WJ. Pituitary follicular cells secrete a novel
heparin-binding growth factor specific for vascular endothelial cells.
Biochem Biophys Res Commun 1989; 161: 851–858
2. Conn G, Bayne ML, Soderman DD et al. Amino acid and cDNA
sequences of a vascular endothelial cell mitogen that is homologous
to platelet-derived growth factor. Proc Natl Acad Sci USA 1990;
87: 2628–2632
3. Keck PJ, Hauser SD, Krivi G et al. Vascular permeability factor,
an endothelial cell mitogen related to PDGF. Science 1989; 246:
1309–1312
4. Leung DW, Cachianes G, Kuang WJ, Goeddel DV, Ferrara N.
Vascular endothelial growth factor is a secreted angiogenic mitogen.
Science 1989; 246: 1306–1309
5. Tischer E, Mitchell R, Hartman T et al. The human gene for
vascular endothelial growth factor. Multiple protein forms are
encoded through alternative exon splicing. J Biol Chem 1991; 266:
11947–11954
6. Ferrara N, Houck K, Jakeman L, Leung DW. Molecular and
biological properties of the vascular endothelial growth factor
family of proteins. Endocrinol Rev 1992; 13:18–32
7. Sugihara T, Kaul SC, Mitsui Y, Wadhwa R. Enhanced expression
of multiple forms of VEGF is associated with spontaneous immortalization of murine fibroblasts. Biochim Biophys Acta 1994; 1224:
365–370
8. Fava RA, Olsen NJ, Spencer-Green G et al. Vascular permeability
factor/endothelial growth factor ( VPF/VEGF ): accumulation and
expression in human synovial fluids and rheumatoid synovial tissue
J Exp Med 1994; 180: 341–346
9. Iijima K, Yoshikawa N, Connolly DT, Nakamura H. Human
mesangial cells and peripheral blood mononuclear cells produce
vascular permeability factor. Kidney Int 1993; 44: 959–966
10. Brown LF, Berse B, Tognazzi K et al. Vascular permeability factor
mRNA and protein expression in human kidney. Kidney Int 1992;
42: 1457–1461
11. Simon M, Gröne H-J, Jöhren O et al. Expression of vascular
endothelial growth factor and its recept in human renal ontogenesis
and in adult kidney. Am J Physiol 1995; 268: F240–F250
12. chida K, Uchida S, Nitta K, Yumura W, Marumo F, Nihei H.
Glomerular endothelial cells in culture express and secrete vascular
endothelial growth factor. Am J Physiol 1994; 266: F81–F88
13. Senger DR. Molecular framework for angiogenesis. A complex
web of interactions between extravasated plasma proteins and
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
endothelial cell proteins induced by angiogenic cytokines. Am
J Pathol 1996; 149: 1–7
Peters KG, De Vries C, Williams LT. Vascular endothelial growth
factor receptor expression during embryogenesis and tissue repair
suggests a role in endothelial differentiation and blood vessel
growth. Proc Natl Acad Sci USA 1993; 90: 8915–8919
Yamane A, Seetharam L, Yamaguchi S et al. A new communication
system between hepatocytes and sinusoidal endothelial cells in liver
through vascular endothelial growth factor and Flt tyrosine kinase
receptor family (Flt-1 and KDR/Flk-1). Oncogene 1994; 9:
2683–2690
Matthews W, Jordan CT, Gavin M, Jenkins NA, Copeland NG,
Lemischka IR. A receptor tyrosine kinase cDNA isolated from a
population of enriched primitive hematopoietic cells and exhibiting
close genetic linkage to c-kit. Proc Natl Acad Sci USA 1991;
88: 9026–9030
Shen H, Clauss M, Ryan J et al. Characterization of vascular
permeability factor/vascular endothelial growth factor receptors on
mononuclear phagocytes. Blood 1993; 81: 2767–2773
Connolly DT. Vascular permeability factor: a unique regulator of
blood vessel function. J Cell Biochem 1991; 47: 219–223
Roberts WG, Palade GE. Increased microvascular permeability an
endothelial fenestration induced by vascular endothelial growth
factor. J Cell Sci 1995; 108: 2369–2379
Pepper MS, Ferrara N, Orci L, Montesano R. Vascular endothelial
growth factor (VEGF ) induces plasminogen activators and
plasminogen activator inhibitor-1 in microvascular endothelial cells.
Biochem Biophys Res Commun 1991; 181: 902–906
Ryan GB, Karnovsky MJ. Distribution of endogenous albumin in
the rat glomerulus. Role of hemodynamic factors in glomerular
barrier function Kidney Int 1976; 9: 36–45
Ku DD, Zaleski JK, Liu S, Brock TA. Vascular endothelial growth
factor induces EDRF-dependent relaxation in coronary arteries.
Am J Physiol 1993; 265: H586–H592
Radermacher J, Klanke B, Schurek HJ, Stolte H, Frölich JC.
Importance of NO/EDRF for glomerular and tubular function:
studies in the isolated perfused rat kidney. Kidney Int 1992;
41: 1549–1559
Schurek HJ, Alt JM. Effect of albumin on the function of perfused
rat kidney. Am J Physiol 1981; 240: F569–F576
Schmidt FH. Die enzymatische Bestimmung von Glucose und
Fructose nebeneinander. Klin Wochenschr 1961; 39: 1244–1247
Fiebich BL, Jager B, Schöllmann C et al. Synthesis and assembly
of functional active human vascular endothelial growth factor
homodimers in insect cells. Eur J Biochem 1993; 211: 19–26
Radermacher J, Förstermann U, Frölich JC. Endothelium-derived
relaxing factor influences renal vascular resistance. Am J Physiol
1990; 259: F9–F17
Bradford MM. A rapid and sensitive method for the quantification
of microgram quantities of protein utilizing the principle of proteindye binding. Anal Biochem 1976; 72: 248–254
Brock TA, Dvorak HF, Senger DR. Tumor-secreted vascular
permeability factor increases cytosolic Ca2+ and von Willebrand
factor release in human endothelial cells. Am J Pathol 1991;
138: 213–221
Vanhoutte PM, Miller VM. Alpha 2-adrenoceptors and endothelium-derived relaxing factor. Am J Med 1989; 87 [Suppl. 3C ]: 1S–5S
Suga S, Nakao K, Itoh H et al. Endothelial production of C-type
natriuretic peptide and its marked augmentation by transforming
growth factor-b. Possible existence of ‘VascuIar natriuretic peptide
system’. J Clin Invest 1992; 90: 1145–1149
Senger DR, Galli SJ, Dvorak AM, Perruzzi CA, Harvey VS,
Dvorak HF. Tumor cells secrete a vascular permeability factor
that promotes accumulation of ascites fluid. Science 1983; 219:
983–985
Connolly DT, Olander JV, Heuvelman D et al. Human vascular
permeability factor. Isolation from U937 cells. J Biol Chem 1989;
264: 20017–20024
Kanwar YS. Biology of disease: biophysiology of glomerular
filtration and proteinuria. Lab Invest 1984; 51: 7–21
Farquhar MG, Wissig SL, Palade GE. Glomerular permeability.
I. Ferritin transfer across the normal glomerular capillary wall.
J Exp Med 1961; 113: 47–66
VEGF/VPF and renal haemodynamics
36. Sonnenburg-Hatzopoulos C, Assel E, Schurek HJ, Stolte H.
Glomerular albumin leakage and morphology after neutralization
of polyanions. II. Discrepancy of protamine induced albuminuria
and fine structure of the glomerular filtration barrier. J Submicrosc
Cytol 1984; 16: 741–751
37. Schurek HJ. Mechanisms of glomerular proteinuria and hematuria.
Kidney Int. 1994; [Suppl. 47] 46: S12–S16
885
38. Schurek HJ, Neumann KH, Flohr H, Zeh M, Stolte H. The
physiological and pathophysiological basis of glomerular permeability for plasma proteins and erythrocytes. Eur J Clin Chem Clin
Biochem 1992; 30: 627–633
39. Larsson L, Maunsbach AB. The ultrastructural development of
the glomerular filtration barrier in the rat kidney: a morphometric
analysis. J Ultrastruct Res 1980; 72: 392–406
Received for publication: 10.6.97
Accepted in revised form: 12.11.97