- Email: [email protected]
Pregnancy aggravates proteinuria in subclinical glomerulonephritis in the rat
Pregnancy aggravates proteinuria in subclinical glomerulonephritis in the rat MARIA M. FAAS, WINSTON W. BAKKER, REIN T. POELMAN, and GERARD A. SCHUILING GRONINGEN,THENETHERLANDS Because subclinical renal disease may be a g g r a v a t e d during p r e g n a n c y - - a s reflected in the occurrence of proteinuria, for e x a m p l e - - w e investigated whether a subclinical glomerulonephritis (SG) in the non-pregnant rat (passive Heymann nephritis), a condition without proteinuria, is aggravated when the animals b e c o m e pregnant and, if so, whether this is associated with a glomerular inflammatory reaction. SG was induced in non-pregnant rats 8 days before pregnancy. On day -I, part of the group of rats b e c a m e pregnant. Three experiments were performed. In experiment I, 4-hour urine albumin excretions and blood pressure (tail cuff) were measured. In experiment 2, the glomerular filtration rate (GFR) was measured with the chromium 51-1abeled ethylenediaminetetraacetic acid method, while in experiment 3, parameters characteristic of a glomerular inflammation were studied. Experiment I revealed that non-pregnant rats with SG did not exhibit proteinuria. However, after the rats b e c a m e pregnant, a significant proteinuria occurred, without an increase in systolic blood pressure. Experiment 2 revealed that the GFR did not increase in pregnant rats with SG, while experiment 3 showed that only pregnant animals exhibited a significant glomerular inflammation; this giomerular inflammation was characterized by intraglomerular influx of polymorphonuclear cells and monocytes. The results suggest that an SG in the rat may flare up during pregnancy. This exacerbation is characterized by proteinuria and coincides with a glomerular inflammatory reaction. It is suggested that proteinuria and the glomerular inflammatory reaction are causally related and promoted by the pregnant condition. (J Lab Clin Med 1999; 134:267-74)
Abbreviations: 51Cr-EDTA= chromium 51-labeled ethylenediaminetetraacetic acid; FITC =
fluorescein isothiocyanate; GFR = glomerular filtration rate; ICAM-1 = intracellular adhesion molecule-I; IgG = immunoglobulin G; LFA-1 = leukocyte functional antigen-I; MAC-1 : macrophage antigen-I; PMN = poiymorphonuclear neutrophil; SG = subclinical glomeruIonephritis; VCAM-1 = vascular cell adhesion molecule-I; VLA-4 : very late antigen-4
A
S early as the 1930s, A p i t z o b s e r v e d that in pregnant rabbits, a single injection o f the p r o i n f l a m m a t o r y bacterial agent e n d o t o x i n was sufficient to induce the so-called S h w a r t z m a n n reaction, whereas non-pregnant animals needed 2 such From the Division of Reproductive Biology, Department of Obstetrics and Gynecology and Department of Pathology, University of Groningen. Submitted for publication December 18, 1999; revision submitted April 2, 1999; accepted April 13, 1999. Reprint requests: M. M. Faas, PhD, Division of Reproductive Biology, Department of Obstetrics and Gynecology, University of Groningen, RO. Box 30.001, 9700 RB Groningen, The Netherlands. Copyright © 1999 by Mosby, Inc. 0022-2143/99 $8.00 + 0 5/1199474
injections given 24 hours after each other1; this observation was c o n f i r m e d and e x t e n d e d b y various authors. 2,3 Recently we demonstrated that in pregnant rats, low-dose endotoxin treatment elicits a much more intense and persistent g l o m e r u l a r i n f l a m m a t o r y response than in non-pregnant rats4; in addition, pregnant rats, but not non-pregnant rats, exhibited a significant proteinuria. 5 The fact that pregnant rats respond to endotoxin with a more intense glomerular inflammatory response than do non-pregnant rats and that they also show significant proteinuria agrees with the view that pregnancy m a y be considered a p r o i n f l a m m a t o r y condition. 6,7 In this p r o i n f l a m m a t o r y condition a m i l d p r o i n f l a m m a t o r y stimulus m a y evoke a (relatively) strong inflammatory response.
267
268
J Lab Clin Med September 1999
Faaset al
In pregnant rats both the endotoxin-induced glomerular i n f l a m m a t o r y response and the proteinuria can, at least partly, be suppressed by the a n t i - i n f l a m m a t o r y agent aspirin. 8 This suggests that, as in the non-pregnant condition, in the pregnant condition a causal relationship exists b e t w e e n the g l o m e r u l a r i n f l a m m a t o r y response and proteinuria. 9J° This putative causal relationship may well explain why during pregnancy a subclinical glomerulonephritis tends to be exacerbated, leading to proteinuria.11 Although the common view in this matter is that p r e g n a n c y - i n d u c e d proteinuria is caused by an enhanced G F R associated with pregnancy, u it can be argued that a subclinical glomerulonephritis might act as the eliciting stimulus, which in the p r o i n f l a m m a t o r y condition of pregnancy leads to glomerular infiltration and activation of inflammatory cells, d a m a g i n g the filtration barrier and u l t i m a t e l y leading to proteinuria. This hypothesis was tested in the present study. We thus investigated whether the presence of an SG in the rat (in the present study passive Heymann nephritis, which is a model for human membranous nephropathy12) is aggravated when the animals become pregnant and whether this flaring up is associated with a glomerular inflammatory reaction. To this end SG was induced in cyclic rats in such a way that proteinuria was not observed. Part of the group of rats were allowed to become pregnant. In pregnant and cyclic rats with SG, we studied urinary albumin excretion and g l o m e r u l a r i n f l a m m a t o r y parameters such as the influx of P M N s and m o n o c y t e s and the expression o f the g l o m e r u l a r adhesion m o l e c u l e s I C A M - 1 and V C A M - 1 and the leukocyte adhesion m o l e c u l e s L F A - 1 , M A C - l , and VLA-4. In addition, we measured physiologic parameters such as G F R and blood pressure. METHODS Experimental animals. Outbred Wistar rats (Harlan) at the age of 3 to 4 months were used. Rats were housed in a temperature- and light-controlled room (lights on 5:30 AMto 5:30 PM).Animals were given free access to food and water. Until animals were selected for experiments, daily vaginal smears were taken. Rats were allowed to become pregnant by housing them on the night of proestrus with fertile males. The next day, when spermatozoa were detected in the smear, was designated as day 0 of pregnancy. Induction of subclinical glomerulonephritis. Subclinical glomerulonephritis (passive Heymann nephritis) was induced by a single tail vein injection of rabbit-anti-FxlA IgG in a dose of 1 rag/200 g body weight. Anti-FxlA was prepared according to standard methods. 13 In pilot studies this antibody dose did not induce proteinuria or complement deposition in cyclic rats. Measurement of systolic blood pressure. Systolic blood pressure was measured with the tailcuff method as described
before. 5 In brief, after the rats were trained for 8 to 9 days to be accustomed to the procedure of measuring, blood pressure was measured at the tail with a 15-ram occlusion cuff and a pulse transducer, which was connected to an oscilloscope. Cuff pressure was increased until the pulse disappeared and then slowly released until the first pulse appeared. This was denoted as systolic blood pressure. Within 5 minutes, three measurements were made and systolic blood pressure was expressed as the mean of these three measurements. Measurement of proteinuria. For the determination of urinary albumin, rats were housed in metabolic cages on various experimental days from 4:30 PM until 8:30 PM as previously described.5 This relatively short period of urine collection was chosen to minimize stress, since in the same rats blood pressure was also measured. Albumin concentration was measured with Rocket electrophoresis as described previously.5 Results are expressed as total albumin over 4 hours. Measurement of GFR. GFR was measured with 51CREDTA and a single sample method as described before. 14 In brief, for these experiments rats received a permanent jugular vein cannula according to standard methods.15 A 10 mCi sample of 51Cr-EDTA (Radiochemical Centre, Amersham, UK) in 0.2 mL saline solution was injected via the permanent jugular vein cannula, after which the cannula was flushed with 0.2 mL of saline solution. Previous pilot experiments have shown that this flushing was sufficient to flush the 51CrEDTA from the cannula (unpublished results). Sixty minutes later, 0.5 mL of blood was withdrawn from the cannula for the measurement of radiolabel. GFR was calculated as described previously.14 The distribution volume for pregnant (day 14) and cyclic rats was measured according to standard procedures. 14 Pilot studies revealed no differences in plasma volume between rats with or without glomerulonephritis. Immunohistology procedures
Immunohistology.Cryostat sections (4 gin)were stained for the presence of activated neutrophils (PMNs) with a double-staining technique as described before. 4 PMNs were stained for their potential to produce O a- by the method of Briggs et al,16 and this staining was combined with standard immunostaining for PMNs with a mouse IgM monoclonal antibody against rat PMNs (HIS48, Pharmingen). Other sections (4 gm) were stained for the presence of monocytes and the adhesion molecules ICAM-1, VCAM-1, LFA-1, MAC-I, and VLA-4 according to standard procedures. 4 In brief, sections were fixed in acetone, and after preincubation with normal rabbit serum (5 %), they were incubated with the first antibody (see below) for 30 minutes. Peroxidase-conjugated second antibody (rabbit anti-mouse; Dako A/S, Glostrup, Denmark) was used. The second antibody was visualized with hydrogen peroxide and 3-amino-9-ethyl-carbazole (Sigma). Control sections, omitting primary or secondary antibody, were consistently negative. Antibodies. All antibodies used were described previously4 and are listed here: mouse IgG1 monoclonal antibody 1A29 (Genzyme, Cambridge, MA) against rat-ICAM-1; 5F10 (a gift from Dr Lobb, Biogen, Cambridge, MA), a mouse IgG2a monoclonal antibody against rat-VCAM-1; WT.1
J Lab Clin Med Volume 134,Number 3
(Genzyme), a mouse IgG2a monoclonal antibody against ratLFA-1; WT.5, a mouse IgA monoclonal antibody against ratMAC-lcq MRa4 (Pharmingen, San Diego, CA), a mouse monoclonal IgG2b antibody against rat-VLA-4; ED-1, I7 a mouse IgG1 monoclonal antibody against a cytoplasmic antigen of rat macrophages/monocytes. Evaluation of kidney sections. Kidney sections of each individual animal were scored by double-blind light microscopic examination by two independent observers. Sections were scored in the following ways: (1) sections stained with ED-1, LFA-1, MAC-l, and VLA-4 were quantitatively scored by counting the total number of positive ceils in 100 glomeruli in 1 section; (2) sections stained for the presence of PMNs and activated PMNs were quantitatively scored by counting both the total number of PMNs and the number of doublestained cells per 100 glomemli; (3) sections of individual animals, stained for ICAM-1 and VCAM-1, were semiquantitatively graded by scoring a total of 100 glomeruli per section with an arbitrary scale from 1 to 4 (1 = no staining, 2 = weak staining, 3 = moderate staining, 4 = bright staining). Experimental design. On day -8 (estrus), cyclic rats were injected with anti-FxlA IgG. On day -1 (proestrus), 50% of the rats were housed with a fertile male for 1 night. The next day was designated as day 0 of the experiment (which is also day 0 of pregnancy). When parturition did not take place between 4:00 PM on day 21 and 12:00 AN on day 22, rats were discarded from the experiments. Immediately after delivery, the offspring were taken away and killed. The day of delivery was day +1. The other 50% of the rats remained cyclic throughout the experiment. At the end of each experiment, the rats were killed and kidney slices were snap-frozen and routinely checked for the presence of rabbit IgG, rat IgG, and complement with FITClabeled goat-anti-rabbit IgG, FITC-labeled goat-anti-rat IgG (Nordic), and FITC-labeled goat-anti-rat C3c, respectively, according to standard methods. Three experiment.s were performed. In all three experiments, 1 group of control (not injected with the c~-FxlA antibody) cyclic rats (n = 5) and 1 group of control (day 14) pregnant rats (n = 5) were included for comparison. Experiment I: Albuminuria and blood pressure. On days -2 (before pregnancy), 14 (during pregnancy), and +6 (after pregnancy)(that is, 6, 22, and 35 days after antibody injection, respectively), blood pressure and albuminuria were measured in "pregnant" (n = 6) and cyclic rats (n = 5). Experiment 2: GFR. On days -2, 14, and +6 in "pregnant" (n = 6) and cyclic rats (n = 6), GFR was longitudinally measar'ed as described above. Experiment 3: Immunohistology. "Pregnant" and cyclic rats were killed on days -2 (n : 6), 14 (pregnant n = 6; cyclic n = 5), and +6 (after pregnancy n = 6; cyclic n -- 7), and kidney slices were snap-frozen and prepared for immunohistology. Statistics. Results are expressed as arithmetic mean _+SEM. The urinary albumin excretion, blood pressure, and GFR of pregnant and cyclic rats with SG were first analyzed with the Friedman nonparametric repeated measurements test followed by the Wilcoxon signed rank test. The Mann-Whitney U test was used to evaluate differences between pregnant and
Faasetal
269
500 Control rats rats with subclinical glomerulonephritis 400-
C~
.~ 300E
(~200{3)
I00O
,
~
-
day -2
-
day 14
day +6
experimental day
Fig 1. Total 4-hour urinary albuminexcretion (gg) of pregnant(solid bars) and cyclic (open bars) rats with SG (right panel) and of control day 14 pregnant and cyclic rats (left panel). For pregnant rats with SG, day -2 is before pregnancy,day 14 is day 14 of pregnancy, and day +6 is 6 days after parturition. *Significantly increased as compared with cyclic rats after identical treatment (Mann Whitney U test, P < .05). a, Significantlyincreased as compared with control pregnantrats (MannWhitneyU test, P < .05).
cyclic rats on the same experimental day or to compare rats with SG with control rats. The number of glomerular PMNs and monocytes and the number of LFA- 1-positive, MAC- 1-positive, and VLA-4-positive cells of all groups were first analyzed with analysis of variance (Kruskall Wallis) followed by Mann-Whitney U test to compare individual groups. Statistical significance was reached at P < .05. RESULTS
At the end of each experiment, cryostat kidney sections of each animal were stained for the presence of rabbit IgG and rat IgG in the glomeruli. In all rats injected with 0~-FxlA antibody, both rabbit IgG as well as rat IgG could be detected in a granular pattern along the capillary walls of the glomeruli, indicating that passive H e y m a n n nephritis had been induced in all animals. No effect of pregnancy on rabbit or rat IgG deposition was observed. Complement could not be detected in any of the rats. The presence of the subclinical glomerulonephritis did not prevent the rats from becoming pregnant. The maternal and fetal weights in pregnant rats with SG were not different from those in control pregnant rats (results not shown). Experiment I: Urinary albumin excretion and blood pressure. Fig 1 shows the urinary albumin excretion of pregnant and cyclic rats with SG (right panel) and for comparison, that of day 14 pregnant and control cyclic rats (left panel). It can be seen that cyclic rats with SG did not exhibit increased urinary albumin excretion as compared with control cyclic rats (Friedman). However, pregnant rats showed an increased urinary albumin
270
J Lab Clin Med September 1999
Faas et al
150
1.00 [control rats rats with subclinical glomerulonephritis
~
[ Control rats ratswith subclin~al glomerulonephritis
°fl111i1 1111 050
~, 120
"~
"~ 0.25
1
0
0
day -2
~
*
0 00"
day 14
day +6
'
experimental day Fig 2. Systolic blood pressure (ram Hg) of pregnant (solid bars) and cyclic (open bars) rats with SG before (day -2), during (day 14), and after pregnancy (day +6)(rightpanel) and that of control day 14 pregnant and cyclic rats (left panel). No significant differences were observed between the various experimental groups.
day -2 day 14 experimental day
day +6
Fig 4. Mean number of PMNs per glomerulus, as detected immunohistochemically in cryostat kidney sections of pregnant (solid bars) and cyclic rats (open bars) with SG (right panel) before (day -2), during (day 14), and after pregnancy (day +6) and in those of control pregnant and cyclic rats (left panel,). *Significantly increased as compared with cyclic rats after identical treatment (Mann Whitney U test, P < .05). c~, Significantly increased as compared with control pregnant rats (Mann Whitney U test, P < .05).
:toontro : r,,s=
rats with subclinical glomerulonephritis 1.75 / . .50J control rats rats with subclinical glomerulonephritis 1
3.0~-
E
2ii
(X 2.0 u. (.9 1.5
,00t o 0.75-1
000tl I
1.0
0.5 0.0
day -2 day 14 experimental day
day +6
Fig 3. GFR (mL/min) of pregnant (solid bars) and cyclic (open bars) rats with SG before (day -2), during (day 14), and after pregnancy (day +6)(right panel) and that of control day 14 pregnant and cyclic rats (left panel). No significant differences were observed between the various experimental groups.
e x c r e t i o n on day 14 o f p r e g n a n c y as c o m p a r e d w i t h both c y c l i c rats after i d e n t i c a l t r e a t m e n t ( M a n n - W h i t ney, P < .05) and w i t h c o n t r o l p r e g n a n t rats ( M a n n W h i t n e y , P < .05). A f t e r parturition, the u r i n a r y album i n e x c r e t i o n values of the e x p e r i m e n t a l rats r e t u r n e d to n o r m a l and w e r e not significantly different f r o m values in cyclic rats after identical treatment ( M a n n - W h i t ney). Fig 2 shows the b l o o d pressure of cyclic and pregnant rats w i t h or w i t h o u t SG. It can be seen that b l o o d p r e s s u r e was a f f e c t e d n e i t h e r b y the i n d u c t i o n o f S G nor by p r e g n a n c y (Friedman). Experiment 2: GFR. Fig 3 shows the G F R o f cyclic rats and pregnant rats (before, during, and after pregnancy) w i t h S G as w e l l as that o f c o n t r o l c y c l i c and p r e g n a n t
0.25 0.00
day -2
day 14
day +6
Fig 5. Mean number of monocytes per glomerulus, as detected immnnohistochemically in cryostat kidney sections of pregnant (solid bars) and cyclic rats (open bars) with SG (right panel) before (day -2), during (day 14) and after pregnancy (day +6) and in those of control pregnant and cyclic rats (left panel). *Significantly increased as compared with cyclic rats after identical treatment (Mann Whitney U test, P < .05). c~, Significantly increased as compared with control pregnant rats (Mann Whitney U test, P < .05). 13, Significantly increased as compared with control cyclic rats (Mann Whitney U test, P < .05).
rats. It can be s e e n that G F R was not a f f e c t e d by the i n d u c t i o n o f S G ( F r i e d m a n ) in c y c l i c rats. T h e r e was also no effect o f p r e g n a n c y on G F R in rats w i t h or without S G (Friedman). Experiment 3: Glomerular inflammation. F i g s 4 and 5 s h o w the n u m b e r o f i n f i l t r a t e d P M N s and m o n o c y t e s in the g l o m e r u l i o f p r e g n a n t and c y c l i c rats w i t h S G (right panel) and in c o n t r o l c y c l i c and c o n t r o l day 14 p r e g n a n t rats (left panel). A s can be s e e n f r o m F i g 4, in c y c l i c rats w i t h SG, the g l o m e r u l a r P M N n u m b e r
J Lab Clin M e d Volume 134, N u m b e r 3
Faas et al
~o 1.00-
"~ E o
.= 0.75 ,
I controlrats ratswithsubclinicalglomerulonephdtis
control rats rats with subclinical glomerulonephritis
0.75-
o
0.50 o
V
O~
T-.
0.50-
i
O
<
LL .,J
..Q
271
v,_ O
0.25-
0
.
2
5
~
c
0.00
day -2
day 14
0,00
day +6
day -2 day 14 experimentalday
experimental day
day +6
Fig 6.
Mean number of LFA-l-positive cells per glomerulus, as detected immunohistochemicallyin cryostat kidney sections of pregnant rats (solid bars) and cyclic rats (open bars) with SG (rightpaneI) before (day -2), during (day 14), and after pregnancy (day +6) and in those of control pregnant and cyclic rats (left panel). *Significantly increased as compared with cyclic rats after identical treatment (Mann Whitney U test, P < .05). c~, Significantly increased as compared with control pregnant rats (Mann Whitney U test, P < .05).
Fig 7. Mean number of MAC-l-positive cells per glomerulus,
was not significantly different from that in control cyclic rats (Kruskall Wallis) on either experimental day. However, in pregnant rats with S G (day 14 of pregnancy), the g l o m e r u l a r P M N n u m b e r was significantly increased as c o m p a r e d with both cyclic rats with S G on day 14 (Mann W h i t n e y U test, P < .05) and with control pregnant rats (Mann W h i t n e y U test, P < .05). The mean percentage o f these cells that were activated (ie, that p r o d u c e d o x y g e n free radicals) did not differ among the groups and a m o u n t e d to about 55% ( K r u s k a l l Wallis). The n u m b e r o f g l o m e r u l a r m o n o cytes (Fig 5) was significantly increased on day - 2 (that is, 6 days after the induction of SG) as c o m p a r e d with control cyclic rats (Kruskall Wallis followed by M a n n W h i t n e y ; P < .05). In the rats that r e m a i n e d cyclic throughout the experiment, the g l o m e r u l a r m o n o c y t e n u m b e r returned to control values on days 14 and +6 and was on these days not significantly different from values in control cyclic rats (Mann Whitney). In contrast, in rats that b e c a m e pregnant, the g l o m e r u l a r m o n o c y t e number increased even further as c o m p a r e d with day - 2 and was significantly i n c r e a s e d as compared with cyclic rats with SG on day 14 (Mann Whitney, P < .05) and as c o m p a r e d with control pregnant rats (Mann Whitney, P < .05). After pregnancy the numbers of m o n o c y t e s p e r glomerulus returned to control values and were not different from values in control cyclic rats. Figs 6, 7, and 8 show the n u m b e r o f cells per glomerulus expressing the adhesion molecules LFA-1, M A C - I , and VLA-4, respectively.
.=
as detected immunohistochemicallyin cryostat kidney sections of pregnant (solid bars) and cyclic rats (open bars) with SG (right panel), before (day -2), during (day 14), and after pregnancy (day +6) and in those of control pregnant and cyclic rats (left panel). *Significantly increased as compared with cyclic rats after identical treatment (Mann Whitney U test, P < .05). a, Significantly increased as compared with control pregnant rats (Mann Whitney U test, P < .05).
0.75Control rats
rats with subclinical glomerulonephritis
E _o "~ 0.50-
5
-k
0.25-
== 0.00
n
day -2 day 14 experimentalday
day +6
Fig 8. Mean number of VLA-4-positive
cells per glomerulus, as detected immunohistochemicallyin cryostat kidney sections of pregnant rats (solid bars) and cyclic rats (open bars) with SG (right panel) before (day -2), during (day 14), and after pregnancy (day +6) and in those of control pregnant and cyclic rats (left panel). *Significantly increased as compared with cyclic rats after identical treatment (Mann Whitney U test, P < .05). a, Significantly increased as compared with control pregnant rats (Mann Whitney U test, P < .05).
In cyclic rats the glomerular numbers of LFA- 1-positive cells (Fig 6), M A C - 1 - p o s i t i v e cells (Fig 7), and V L A - 4 - p o s i t i v e cells were not affected by SG. In day 14 pregnant rats with SG, the numbers o f L F A - l - p o s i tive, M A C - l - p o s i t i v e , and V L A - 4 - p o s i t i v e ceils per glomerulus were significantly increased as c o m p a r e d with those o f cyclic rats with S G on day 14 (Mann Whitney, P < .05) and with those o f control pregnant rats (Mann Whitney, P < .05). The numbers o f LFA-1-
272
Faaset al
positive, MAC-l-positive, and VLA-4-positive cells per glomerulus returned to control values after pregnancy and were not significantly different from control cyclic rats (Kruskall Wallis). The ligands for the leukocyte adhesion molecules LFA-1/MAC-1 and VLA-4, ICAM-1, and VCAM-1, respectively, could not be detected in either group of rats in the present study (results not shown). DISCUSSION
The present protocol for the induction of an SG (1.0 mg e~-FxlA/200 g body weight to induce passive Heymann nephritis) did not induce albuminuria in cyclic rats; this agrees with the results of previous studies. 18 Also, the animals exhibited only a very mild glomerular inflammatory reaction: the induction of SG per se had no effect on the number of glomerular PMNs and a small but significant effect on the number of glomerular monocyctes on day - 2 only (ie, 6 days after the induction of SG). This mild form of glomerulonephritis can therefore be considered subclinical. Moreover, the present subclinical glomerulonephritis did not prevent the animals from becoming pregnant and did not affect maternal weight, litter size, or litter weight (data not shown). Once the rats were pregnant, however, the urinary albumin excretion of rats with SG increased significantly. This was associated with a more intense glomerular inflammatory response, as apparent from the increased numbers of PMNs and monocytes in the glomeruli on day 14 of pregnancy. Because after parturition both the albuminuria and the inflammatory response returned to the control situation, it appears that there is a specific influence of pregnancy on these parameters in rats with SG. The exacerbating effect of pregnancy on subclinical SG reported here that leads to a glomerular inflammatory reaction and albuminuria is undocumented and may support the notion that pregnancy is a proinflammatory condition. 6 At first glance, the present results seem to contrast with those of Baylis et al, who reported the absence of an effect of pregnancy on urinary protein excretion in both a model of experimentally induced anti-glomerular basement membrane antibody-mediated glomerulonephritis 19 and in the F x l A model of accelerated, passive Heymann nephritis, s0 The experiments of these authors, however, differed from the present ones in that massive proteinuria was already observed in the nonpregnant condition; hence the experiments of Baylis et al and the present ones are not entirely comparable, since the condition of their experimental animals was not "subclinical." With respect to the effect of pregnancy on the GFR, there is also a discrepancy between the present data and data in the literature. In human patients the aggravation of subclinical renal disease during preg-
J Lab Clin Med September 1999
nancy is well documented and is usually attributed to an increased GFR. u In the present study, however, the increased albuminuria of pregnant rats with SG cannot be attributed to an increased GFR, nor can it be attributed to hypertension, since neither pregnancy nor the induction of SG affected these two parameters. This lack of increase in GFR during pregnancy is in line with results of a previous study of our group in which we compared various methods of GFR measurement and found that the (apparent) increase in the GFR in pregnant rats observed with the inulin clearance method is probably due to the (relatively) large saline load that is inherent to this method, not to an effect of pregnancy per se. 14 Glomerular complement deposition has been suggested to be one of the factors responsible for proteinuria in (nonpregnant) rats with passive Heymann nephritis. 21 Indeed, the absence of complement deposition in cyclic rats with passive Heymann nephritis agrees with the absence of albuminuria in these rats. However, there was no complement deposition observed in pregnant SG rats either, despite the significant albuminuria of these animals. Complement deposition, therefore, probably plays no role in the pregnancy-associated albuminuria in rats with SG. Rather, the albuminuria in the present study appears to coincide with an inflammatory response observed in the glomeruli of the pregnant animals with SG. Indeed, there are several lines of evidence, both experimental22 and clinical, 23 indicating that the influx of (activated) monocytes or PMNs may injure the glomerular filtration barrier, leading to proteinuria (for a review see reference 24). Although the numbers of inflammatory cells in the glomeruli of pregnant rats with SG are relatively low, they may still inflict significant damage on the glomerular filtration barrier in the pregnant condition, because during pregnancy, glomerular tissue is much more sensitive to products of activated inflammatory cells.25 The question remains how the influx of inflammatory cells is brought about once the animals become pregnant. The present study does not give an unequivocal answer to this question. The infiltration of inflammatory cells is usually regulated by the expression of endothelial 26 and leukocytic27 adhesion molecules. In the case of pregnant SG rats, the expression of adhesion molecules was seen only on the leukocytes (ie, LFA-1, MAC-l, and VLA-4); the expression of the respective ligands--that is, ICAM-1 and V C A M - 1 - on the endothelial cells could not be demonstrated. This may suggest that other factors within the glomeruli, capable of binding to the leukocytic adhesion molecules - - for example, IgG (in the present study causing the SG) --which can bind to monocytes (via MAC-128) and
J Lab Clin Med Volume 134,Number 3
P M N s (although i n d e p e n d e n t from MAC-129), are involved in the infiltration of i n f l a m m a t o r y cells into the glomeruli of pregnant rats. Also, the local production of chemoattractants 3°,3t m a y p l a y a role, since it has been shown that the glomeruli of rats with passive H e y m a n n nephritis p r o d u c e increased amounts of leukotriene B4, a potent chemoattractant for inflammatory cells 31 when compared with the amounts produced in control rats. 32 One can speculate that the glomeruli o f pregnant rats with passive H e y m a n n nephritis produce increased amounts of leukotriene B4 as compared with cyclic rats with passive Heymann nephritis, but so far this has not been measured. Also, it is possible that i n f l a m m a t o r y cells in pregnant rats are more responsive to leukotriene B4. This notion is supported b y data showing that progesterone, levels of which are increased during pregnancy, 33 increases the responsiveness of inflammatory cells to chemoattractants. 34 SPECULATIONS
This study shows that experimental S G may become aggravated during pregnancy, as reflected by increased urinary albumin excretion. The increased urinary albumin excretion is not associated with an increased G F R or with an increased b l o o d pressure. The present data agree with the notion that in fact the aggravation of SG in the rat develops when the animals become pregnant, b e c a u s e in that condition a g l o m e r u l a r i n f l a m m a t o r y response develops. Although further studies to evaluate the functional role of the leukocytes in the present pregnancy-associated proteinuria are needed, it is tempting to extrapolate the present data to the human situation and speculate that in human patients too, the flaring up of subclinical renal disease results from a g l o m e r u l a r inflammatory reaction that develops during pregnancy. This inflammatory reaction, then, causes damage to the glomerular filtration barrier, which ultimately leads to proteinuria. REFERENCES
l. Apitz K. A study on the generalized Shwartzman phenomenon. J Immunol 1935;29:255-66. 2. Beller FK, Schrdidt EH, Holzgreve W, Hanss J. Septicemia during pregnancy: a study in different species of experimental animals. Am J Obstet Gynecol 1985; 151:967-75. 3. Muller-Berghans G, Schmidt-Ehry B. The role of pregnancy in the induction of the generalized Shwartzman reaction. Am J Obstet Gynecol 1972;114:847-9. 4. Faas MM, Schuiling GA, Baller JFW, Bakker WW. Glomerular inflammation in pregnant rats after infusion of low dose endotoxin: an immunohistological study in experimental pre-eclampsia. Am J Pathol 1995;147:1510-8. 5. Faas MM, Schuiling GA, Bailer JFW, Visscher CA, Bakker WW. A new animal model for human pre-eclampsia: ultralow dose endotoxin infusion in pregnant rats. Am J Obstet Gynecol 1994; 171:158-64.
Faaset al
273
6. Sacks GP, Studena K, Sargent IL, Redman CWG. CDllb expression on circulating neutrophils in pre-eclampsia. Clin Sci 1997;93:187-9. 7. Schuiling GA, Koiter TR, Faas MM. Why pre-eclampsia? Hum Reprod 1997;12:2087-92. 8. Faas MM, Schuiling GA, Baller JFW, Valkhof N, Bakker WW. Aspirin treatment of the low-dose-endotoxin-treated pregnant rat: pathophysiologic and immunohistologic aspects. J Lab Clin Med 1997;130:496-508. 9, Faruqi RM, DiCorleto PE. Mechanism of monocyte recruitments and accumulation. Br Heart J 1993;69(suppl):S19-29. 10. Schreiner GE The role of the macrophage in glomerular injury. Semin Nephrol 1991;11:268-75. 11. Katz AI, Davison JM, Hayslett JR Singsong E, Lindheimer MD. Pregnancy in women with kidney diseases. Kidney Int 1980;18:192-206. 12. Van Leel EHG, Ronco R Verroust P, de Roo GM, Hoedemaeker PJ, De Heer E. Heymann nephritis: a model of human membranous glomerulopathy. A study of the role of additional antigens. Nephrol Dial Transplant 1992;7(suppl 1):1-8. 13. Bakker WW, Mulder I, Lee RLvd, Fleuren GJ, Hoedemaeker PJ. Experimental glomerulonephritis in the rat induced by antibodies directed against tubular antigens. III. In vitro evaluation of cell-mediated immune responses against immune complexes influenced by irmnunosuppressive therapy. Int Arch Allergy Appl Immunol 1977;54:405-13. 14. Faas MM, Schuiling GA, Klok PA, Valkhof N, Bakker WW. The glomerular filtration rate during pregnancy: saline infusion enhances the glomerular filtration rate in the pregnant rat. Kidney Blood Press Res 1996;19: 121-7. 15. Steffens AB. A method for frequent sampling of blood and continuous infusion of fluids in the rat without disturbing the animal. Physiol Behav 1969;4:833-6. 16. Briggs RT, Robinson JM, Karnovski ML, Karnovski MJ. Superoxide production by polymorphonuclear leukocytes: a cytochemical approach. Histochem 1986;84:371-8. 17. Dijkstra CD, Dopp EA, Joling R Kraal G. The heterogeneity of mononuclear phagocytes in lymphoid organs: distinct macrophage subpopulation in the rat recognized by monoclonal antibodies ED1, ED2 and ED3. Immunology 1985;54:589. 18. Iversen BM, Matre R, Ofstad J. The heterologous immune complex glomerulonephritis: a dose dependent glomerulonephritis with acute, latent and chronic phase in long term study. Acta Path Microbiol Immunol Scandinav 1982;90:241-50. 19. Baylis C, Reese K, Wilson CB. Glomerular effects of pregnancy in a model of glomerulonephritis in the rat. Am J Kidney Dis 1989;14:452-60, 20. Baylis C, Deng A, Couser WG. Glomerular hemodynamic effects of late pregnancy in rats with experimental membranous glomerulopathy. J Am Soc Nephrol 1995;6:1197-201. 21. Salant DJ, Quigg RJ, Cybulsky AV. Heymann nephritis: mechanism of renal injury. Kidney Int 1989;35:976-84. 22. Lefkowith JB, Nagamatsu T, Pippin J, Schreiner GF. Role of leukocytes in metabolic and functional derangements of experimental glomerulonephritis. Am J Physiol 1991;261:F213-20. 23. Monga G, Mazzucco G, Barbiano di Belgiojoso G, Busnach G. The presence and possible role of monocyte infiltration in human chronic proliferative glomerulonephritis. Am J Pathol 1979;94:271-84.
274
Faas et al
24. van Goor H, Ding G, Kees-Folts D, Grond J, Schreiner GF, Diamond JR. Macrophages and renal disease. Lab Invest 1994;71:456-64. 25. FaasMM, BakkerWW, Baller JFW, SchuilingGA. Pregnancy enhances the sensitivity of glomerular ecto-ATP-dipbosphohydrolase to products of activated polymorphonuclear leukocytes (PMN). Am J Obstet Gynecol 1999;180:112-3. 26. Munro JM, Pober JS, Cotran RS. Recruitment of neu, trophils in the local endotoxin response: association with de novo endothelial expression of endothelial leukocyte adhesion molecule-1. Lab Invest 1991;64:295-9. 27. Cronstein BN, Kimmel SC, Levin RI, Martiniuk F, Weissmann G. A mechanism for the antiinflammatory effects of corticosteroids: the glucocorticoid receptor regulates leukocyte adhesion to endothelial cells and expression of endothelial-leukocyte adhesion molecule 1 and intercellular adhesion molecule 1. Proc Natl Acad Sci USA 1992;89:9991-5. 28. Brown EJ, Bohnsack JF, Gresham HD. Mechanism of inhibition of immunoglobullinG-mediated phagocytosis by monoclonal antibodies that recognize the Mac-1 antigen.
J Lab Clin Med September 1999
J Clin Invest 1988;81:365-75. 29. Arnaout MA, Pitt J, Cohen HJ, Melamed J, Rosen FS, Colten HR. Deficiency of a granulocyte-membrane glycoprotein (gpl50) in a boy with recurrent bacterial infections. N Engl J Med 1982;306:693-9. 30. Cattell V, Cook HT, Smith J, Salmon JA, Moncada S. Leukotriene B4 production in normal rat glomeruli. Nephrol Dial Transplant 1987;2:154-7. 31. Samuelsson B, Dahlen SE, Lindgren JA, Rouzer CA, Serhan CN. Leukotrienes and lipoxins: structures, biosynthesis, and biological effects. Science 1987;237:1171-6. 32. Lianos EA, Noble B. Glomerular leukotriene synthesis in Heymann nephritis. Kidney Int 1989;36:998-1002. 33. Bast JD, Melampy RM. Luteinizing hormone, prolactin and ovarian 20c~-hydroxysteroiddehydrogenase levels during pregnancy and pseudopregnancy in the rat. Endocrinology 1972;91:1499-505. 34. Miyagi M, Aoyama H, Morishita M, Iwamoto Y. Effects of sex hon~aoneson chemotaxis of human peripheral polymorphonuclear leukocytes and monocytes. J Periodontol 1992;63:28-32.
Abbreviations: 51Cr-EDTA= chromium 51-labeled ethylenediaminetetraacetic acid; FITC =
fluorescein isothiocyanate; GFR = glomerular filtration rate; ICAM-1 = intracellular adhesion molecule-I; IgG = immunoglobulin G; LFA-1 = leukocyte functional antigen-I; MAC-1 : macrophage antigen-I; PMN = poiymorphonuclear neutrophil; SG = subclinical glomeruIonephritis; VCAM-1 = vascular cell adhesion molecule-I; VLA-4 : very late antigen-4
A
S early as the 1930s, A p i t z o b s e r v e d that in pregnant rabbits, a single injection o f the p r o i n f l a m m a t o r y bacterial agent e n d o t o x i n was sufficient to induce the so-called S h w a r t z m a n n reaction, whereas non-pregnant animals needed 2 such From the Division of Reproductive Biology, Department of Obstetrics and Gynecology and Department of Pathology, University of Groningen. Submitted for publication December 18, 1999; revision submitted April 2, 1999; accepted April 13, 1999. Reprint requests: M. M. Faas, PhD, Division of Reproductive Biology, Department of Obstetrics and Gynecology, University of Groningen, RO. Box 30.001, 9700 RB Groningen, The Netherlands. Copyright © 1999 by Mosby, Inc. 0022-2143/99 $8.00 + 0 5/1199474
injections given 24 hours after each other1; this observation was c o n f i r m e d and e x t e n d e d b y various authors. 2,3 Recently we demonstrated that in pregnant rats, low-dose endotoxin treatment elicits a much more intense and persistent g l o m e r u l a r i n f l a m m a t o r y response than in non-pregnant rats4; in addition, pregnant rats, but not non-pregnant rats, exhibited a significant proteinuria. 5 The fact that pregnant rats respond to endotoxin with a more intense glomerular inflammatory response than do non-pregnant rats and that they also show significant proteinuria agrees with the view that pregnancy m a y be considered a p r o i n f l a m m a t o r y condition. 6,7 In this p r o i n f l a m m a t o r y condition a m i l d p r o i n f l a m m a t o r y stimulus m a y evoke a (relatively) strong inflammatory response.
267
268
J Lab Clin Med September 1999
Faaset al
In pregnant rats both the endotoxin-induced glomerular i n f l a m m a t o r y response and the proteinuria can, at least partly, be suppressed by the a n t i - i n f l a m m a t o r y agent aspirin. 8 This suggests that, as in the non-pregnant condition, in the pregnant condition a causal relationship exists b e t w e e n the g l o m e r u l a r i n f l a m m a t o r y response and proteinuria. 9J° This putative causal relationship may well explain why during pregnancy a subclinical glomerulonephritis tends to be exacerbated, leading to proteinuria.11 Although the common view in this matter is that p r e g n a n c y - i n d u c e d proteinuria is caused by an enhanced G F R associated with pregnancy, u it can be argued that a subclinical glomerulonephritis might act as the eliciting stimulus, which in the p r o i n f l a m m a t o r y condition of pregnancy leads to glomerular infiltration and activation of inflammatory cells, d a m a g i n g the filtration barrier and u l t i m a t e l y leading to proteinuria. This hypothesis was tested in the present study. We thus investigated whether the presence of an SG in the rat (in the present study passive Heymann nephritis, which is a model for human membranous nephropathy12) is aggravated when the animals become pregnant and whether this flaring up is associated with a glomerular inflammatory reaction. To this end SG was induced in cyclic rats in such a way that proteinuria was not observed. Part of the group of rats were allowed to become pregnant. In pregnant and cyclic rats with SG, we studied urinary albumin excretion and g l o m e r u l a r i n f l a m m a t o r y parameters such as the influx of P M N s and m o n o c y t e s and the expression o f the g l o m e r u l a r adhesion m o l e c u l e s I C A M - 1 and V C A M - 1 and the leukocyte adhesion m o l e c u l e s L F A - 1 , M A C - l , and VLA-4. In addition, we measured physiologic parameters such as G F R and blood pressure. METHODS Experimental animals. Outbred Wistar rats (Harlan) at the age of 3 to 4 months were used. Rats were housed in a temperature- and light-controlled room (lights on 5:30 AMto 5:30 PM).Animals were given free access to food and water. Until animals were selected for experiments, daily vaginal smears were taken. Rats were allowed to become pregnant by housing them on the night of proestrus with fertile males. The next day, when spermatozoa were detected in the smear, was designated as day 0 of pregnancy. Induction of subclinical glomerulonephritis. Subclinical glomerulonephritis (passive Heymann nephritis) was induced by a single tail vein injection of rabbit-anti-FxlA IgG in a dose of 1 rag/200 g body weight. Anti-FxlA was prepared according to standard methods. 13 In pilot studies this antibody dose did not induce proteinuria or complement deposition in cyclic rats. Measurement of systolic blood pressure. Systolic blood pressure was measured with the tailcuff method as described
before. 5 In brief, after the rats were trained for 8 to 9 days to be accustomed to the procedure of measuring, blood pressure was measured at the tail with a 15-ram occlusion cuff and a pulse transducer, which was connected to an oscilloscope. Cuff pressure was increased until the pulse disappeared and then slowly released until the first pulse appeared. This was denoted as systolic blood pressure. Within 5 minutes, three measurements were made and systolic blood pressure was expressed as the mean of these three measurements. Measurement of proteinuria. For the determination of urinary albumin, rats were housed in metabolic cages on various experimental days from 4:30 PM until 8:30 PM as previously described.5 This relatively short period of urine collection was chosen to minimize stress, since in the same rats blood pressure was also measured. Albumin concentration was measured with Rocket electrophoresis as described previously.5 Results are expressed as total albumin over 4 hours. Measurement of GFR. GFR was measured with 51CREDTA and a single sample method as described before. 14 In brief, for these experiments rats received a permanent jugular vein cannula according to standard methods.15 A 10 mCi sample of 51Cr-EDTA (Radiochemical Centre, Amersham, UK) in 0.2 mL saline solution was injected via the permanent jugular vein cannula, after which the cannula was flushed with 0.2 mL of saline solution. Previous pilot experiments have shown that this flushing was sufficient to flush the 51CrEDTA from the cannula (unpublished results). Sixty minutes later, 0.5 mL of blood was withdrawn from the cannula for the measurement of radiolabel. GFR was calculated as described previously.14 The distribution volume for pregnant (day 14) and cyclic rats was measured according to standard procedures. 14 Pilot studies revealed no differences in plasma volume between rats with or without glomerulonephritis. Immunohistology procedures
Immunohistology.Cryostat sections (4 gin)were stained for the presence of activated neutrophils (PMNs) with a double-staining technique as described before. 4 PMNs were stained for their potential to produce O a- by the method of Briggs et al,16 and this staining was combined with standard immunostaining for PMNs with a mouse IgM monoclonal antibody against rat PMNs (HIS48, Pharmingen). Other sections (4 gm) were stained for the presence of monocytes and the adhesion molecules ICAM-1, VCAM-1, LFA-1, MAC-I, and VLA-4 according to standard procedures. 4 In brief, sections were fixed in acetone, and after preincubation with normal rabbit serum (5 %), they were incubated with the first antibody (see below) for 30 minutes. Peroxidase-conjugated second antibody (rabbit anti-mouse; Dako A/S, Glostrup, Denmark) was used. The second antibody was visualized with hydrogen peroxide and 3-amino-9-ethyl-carbazole (Sigma). Control sections, omitting primary or secondary antibody, were consistently negative. Antibodies. All antibodies used were described previously4 and are listed here: mouse IgG1 monoclonal antibody 1A29 (Genzyme, Cambridge, MA) against rat-ICAM-1; 5F10 (a gift from Dr Lobb, Biogen, Cambridge, MA), a mouse IgG2a monoclonal antibody against rat-VCAM-1; WT.1
J Lab Clin Med Volume 134,Number 3
(Genzyme), a mouse IgG2a monoclonal antibody against ratLFA-1; WT.5, a mouse IgA monoclonal antibody against ratMAC-lcq MRa4 (Pharmingen, San Diego, CA), a mouse monoclonal IgG2b antibody against rat-VLA-4; ED-1, I7 a mouse IgG1 monoclonal antibody against a cytoplasmic antigen of rat macrophages/monocytes. Evaluation of kidney sections. Kidney sections of each individual animal were scored by double-blind light microscopic examination by two independent observers. Sections were scored in the following ways: (1) sections stained with ED-1, LFA-1, MAC-l, and VLA-4 were quantitatively scored by counting the total number of positive ceils in 100 glomeruli in 1 section; (2) sections stained for the presence of PMNs and activated PMNs were quantitatively scored by counting both the total number of PMNs and the number of doublestained cells per 100 glomemli; (3) sections of individual animals, stained for ICAM-1 and VCAM-1, were semiquantitatively graded by scoring a total of 100 glomeruli per section with an arbitrary scale from 1 to 4 (1 = no staining, 2 = weak staining, 3 = moderate staining, 4 = bright staining). Experimental design. On day -8 (estrus), cyclic rats were injected with anti-FxlA IgG. On day -1 (proestrus), 50% of the rats were housed with a fertile male for 1 night. The next day was designated as day 0 of the experiment (which is also day 0 of pregnancy). When parturition did not take place between 4:00 PM on day 21 and 12:00 AN on day 22, rats were discarded from the experiments. Immediately after delivery, the offspring were taken away and killed. The day of delivery was day +1. The other 50% of the rats remained cyclic throughout the experiment. At the end of each experiment, the rats were killed and kidney slices were snap-frozen and routinely checked for the presence of rabbit IgG, rat IgG, and complement with FITClabeled goat-anti-rabbit IgG, FITC-labeled goat-anti-rat IgG (Nordic), and FITC-labeled goat-anti-rat C3c, respectively, according to standard methods. Three experiment.s were performed. In all three experiments, 1 group of control (not injected with the c~-FxlA antibody) cyclic rats (n = 5) and 1 group of control (day 14) pregnant rats (n = 5) were included for comparison. Experiment I: Albuminuria and blood pressure. On days -2 (before pregnancy), 14 (during pregnancy), and +6 (after pregnancy)(that is, 6, 22, and 35 days after antibody injection, respectively), blood pressure and albuminuria were measured in "pregnant" (n = 6) and cyclic rats (n = 5). Experiment 2: GFR. On days -2, 14, and +6 in "pregnant" (n = 6) and cyclic rats (n = 6), GFR was longitudinally measar'ed as described above. Experiment 3: Immunohistology. "Pregnant" and cyclic rats were killed on days -2 (n : 6), 14 (pregnant n = 6; cyclic n = 5), and +6 (after pregnancy n = 6; cyclic n -- 7), and kidney slices were snap-frozen and prepared for immunohistology. Statistics. Results are expressed as arithmetic mean _+SEM. The urinary albumin excretion, blood pressure, and GFR of pregnant and cyclic rats with SG were first analyzed with the Friedman nonparametric repeated measurements test followed by the Wilcoxon signed rank test. The Mann-Whitney U test was used to evaluate differences between pregnant and
Faasetal
269
500 Control rats rats with subclinical glomerulonephritis 400-
C~
.~ 300E
(~200{3)
I00O
,
~
-
day -2
-
day 14
day +6
experimental day
Fig 1. Total 4-hour urinary albuminexcretion (gg) of pregnant(solid bars) and cyclic (open bars) rats with SG (right panel) and of control day 14 pregnant and cyclic rats (left panel). For pregnant rats with SG, day -2 is before pregnancy,day 14 is day 14 of pregnancy, and day +6 is 6 days after parturition. *Significantly increased as compared with cyclic rats after identical treatment (Mann Whitney U test, P < .05). a, Significantlyincreased as compared with control pregnantrats (MannWhitneyU test, P < .05).
cyclic rats on the same experimental day or to compare rats with SG with control rats. The number of glomerular PMNs and monocytes and the number of LFA- 1-positive, MAC- 1-positive, and VLA-4-positive cells of all groups were first analyzed with analysis of variance (Kruskall Wallis) followed by Mann-Whitney U test to compare individual groups. Statistical significance was reached at P < .05. RESULTS
At the end of each experiment, cryostat kidney sections of each animal were stained for the presence of rabbit IgG and rat IgG in the glomeruli. In all rats injected with 0~-FxlA antibody, both rabbit IgG as well as rat IgG could be detected in a granular pattern along the capillary walls of the glomeruli, indicating that passive H e y m a n n nephritis had been induced in all animals. No effect of pregnancy on rabbit or rat IgG deposition was observed. Complement could not be detected in any of the rats. The presence of the subclinical glomerulonephritis did not prevent the rats from becoming pregnant. The maternal and fetal weights in pregnant rats with SG were not different from those in control pregnant rats (results not shown). Experiment I: Urinary albumin excretion and blood pressure. Fig 1 shows the urinary albumin excretion of pregnant and cyclic rats with SG (right panel) and for comparison, that of day 14 pregnant and control cyclic rats (left panel). It can be seen that cyclic rats with SG did not exhibit increased urinary albumin excretion as compared with control cyclic rats (Friedman). However, pregnant rats showed an increased urinary albumin
270
J Lab Clin Med September 1999
Faas et al
150
1.00 [control rats rats with subclinical glomerulonephritis
~
[ Control rats ratswith subclin~al glomerulonephritis
°fl111i1 1111 050
~, 120
"~
"~ 0.25
1
0
0
day -2
~
*
0 00"
day 14
day +6
'
experimental day Fig 2. Systolic blood pressure (ram Hg) of pregnant (solid bars) and cyclic (open bars) rats with SG before (day -2), during (day 14), and after pregnancy (day +6)(rightpanel) and that of control day 14 pregnant and cyclic rats (left panel). No significant differences were observed between the various experimental groups.
day -2 day 14 experimental day
day +6
Fig 4. Mean number of PMNs per glomerulus, as detected immunohistochemically in cryostat kidney sections of pregnant (solid bars) and cyclic rats (open bars) with SG (right panel) before (day -2), during (day 14), and after pregnancy (day +6) and in those of control pregnant and cyclic rats (left panel,). *Significantly increased as compared with cyclic rats after identical treatment (Mann Whitney U test, P < .05). c~, Significantly increased as compared with control pregnant rats (Mann Whitney U test, P < .05).
:toontro : r,,s=
rats with subclinical glomerulonephritis 1.75 / . .50J control rats rats with subclinical glomerulonephritis 1
3.0~-
E
2ii
(X 2.0 u. (.9 1.5
,00t o 0.75-1
000tl I
1.0
0.5 0.0
day -2 day 14 experimental day
day +6
Fig 3. GFR (mL/min) of pregnant (solid bars) and cyclic (open bars) rats with SG before (day -2), during (day 14), and after pregnancy (day +6)(right panel) and that of control day 14 pregnant and cyclic rats (left panel). No significant differences were observed between the various experimental groups.
e x c r e t i o n on day 14 o f p r e g n a n c y as c o m p a r e d w i t h both c y c l i c rats after i d e n t i c a l t r e a t m e n t ( M a n n - W h i t ney, P < .05) and w i t h c o n t r o l p r e g n a n t rats ( M a n n W h i t n e y , P < .05). A f t e r parturition, the u r i n a r y album i n e x c r e t i o n values of the e x p e r i m e n t a l rats r e t u r n e d to n o r m a l and w e r e not significantly different f r o m values in cyclic rats after identical treatment ( M a n n - W h i t ney). Fig 2 shows the b l o o d pressure of cyclic and pregnant rats w i t h or w i t h o u t SG. It can be seen that b l o o d p r e s s u r e was a f f e c t e d n e i t h e r b y the i n d u c t i o n o f S G nor by p r e g n a n c y (Friedman). Experiment 2: GFR. Fig 3 shows the G F R o f cyclic rats and pregnant rats (before, during, and after pregnancy) w i t h S G as w e l l as that o f c o n t r o l c y c l i c and p r e g n a n t
0.25 0.00
day -2
day 14
day +6
Fig 5. Mean number of monocytes per glomerulus, as detected immnnohistochemically in cryostat kidney sections of pregnant (solid bars) and cyclic rats (open bars) with SG (right panel) before (day -2), during (day 14) and after pregnancy (day +6) and in those of control pregnant and cyclic rats (left panel). *Significantly increased as compared with cyclic rats after identical treatment (Mann Whitney U test, P < .05). c~, Significantly increased as compared with control pregnant rats (Mann Whitney U test, P < .05). 13, Significantly increased as compared with control cyclic rats (Mann Whitney U test, P < .05).
rats. It can be s e e n that G F R was not a f f e c t e d by the i n d u c t i o n o f S G ( F r i e d m a n ) in c y c l i c rats. T h e r e was also no effect o f p r e g n a n c y on G F R in rats w i t h or without S G (Friedman). Experiment 3: Glomerular inflammation. F i g s 4 and 5 s h o w the n u m b e r o f i n f i l t r a t e d P M N s and m o n o c y t e s in the g l o m e r u l i o f p r e g n a n t and c y c l i c rats w i t h S G (right panel) and in c o n t r o l c y c l i c and c o n t r o l day 14 p r e g n a n t rats (left panel). A s can be s e e n f r o m F i g 4, in c y c l i c rats w i t h SG, the g l o m e r u l a r P M N n u m b e r
J Lab Clin M e d Volume 134, N u m b e r 3
Faas et al
~o 1.00-
"~ E o
.= 0.75 ,
I controlrats ratswithsubclinicalglomerulonephdtis
control rats rats with subclinical glomerulonephritis
0.75-
o
0.50 o
V
O~
T-.
0.50-
i
O
<
LL .,J
..Q
271
v,_ O
0.25-
0
.
2
5
~
c
0.00
day -2
day 14
0,00
day +6
day -2 day 14 experimentalday
experimental day
day +6
Fig 6.
Mean number of LFA-l-positive cells per glomerulus, as detected immunohistochemicallyin cryostat kidney sections of pregnant rats (solid bars) and cyclic rats (open bars) with SG (rightpaneI) before (day -2), during (day 14), and after pregnancy (day +6) and in those of control pregnant and cyclic rats (left panel). *Significantly increased as compared with cyclic rats after identical treatment (Mann Whitney U test, P < .05). c~, Significantly increased as compared with control pregnant rats (Mann Whitney U test, P < .05).
Fig 7. Mean number of MAC-l-positive cells per glomerulus,
was not significantly different from that in control cyclic rats (Kruskall Wallis) on either experimental day. However, in pregnant rats with S G (day 14 of pregnancy), the g l o m e r u l a r P M N n u m b e r was significantly increased as c o m p a r e d with both cyclic rats with S G on day 14 (Mann W h i t n e y U test, P < .05) and with control pregnant rats (Mann W h i t n e y U test, P < .05). The mean percentage o f these cells that were activated (ie, that p r o d u c e d o x y g e n free radicals) did not differ among the groups and a m o u n t e d to about 55% ( K r u s k a l l Wallis). The n u m b e r o f g l o m e r u l a r m o n o cytes (Fig 5) was significantly increased on day - 2 (that is, 6 days after the induction of SG) as c o m p a r e d with control cyclic rats (Kruskall Wallis followed by M a n n W h i t n e y ; P < .05). In the rats that r e m a i n e d cyclic throughout the experiment, the g l o m e r u l a r m o n o c y t e n u m b e r returned to control values on days 14 and +6 and was on these days not significantly different from values in control cyclic rats (Mann Whitney). In contrast, in rats that b e c a m e pregnant, the g l o m e r u l a r m o n o c y t e number increased even further as c o m p a r e d with day - 2 and was significantly i n c r e a s e d as compared with cyclic rats with SG on day 14 (Mann Whitney, P < .05) and as c o m p a r e d with control pregnant rats (Mann Whitney, P < .05). After pregnancy the numbers of m o n o c y t e s p e r glomerulus returned to control values and were not different from values in control cyclic rats. Figs 6, 7, and 8 show the n u m b e r o f cells per glomerulus expressing the adhesion molecules LFA-1, M A C - I , and VLA-4, respectively.
.=
as detected immunohistochemicallyin cryostat kidney sections of pregnant (solid bars) and cyclic rats (open bars) with SG (right panel), before (day -2), during (day 14), and after pregnancy (day +6) and in those of control pregnant and cyclic rats (left panel). *Significantly increased as compared with cyclic rats after identical treatment (Mann Whitney U test, P < .05). a, Significantly increased as compared with control pregnant rats (Mann Whitney U test, P < .05).
0.75Control rats
rats with subclinical glomerulonephritis
E _o "~ 0.50-
5
-k
0.25-
== 0.00
n
day -2 day 14 experimentalday
day +6
Fig 8. Mean number of VLA-4-positive
cells per glomerulus, as detected immunohistochemicallyin cryostat kidney sections of pregnant rats (solid bars) and cyclic rats (open bars) with SG (right panel) before (day -2), during (day 14), and after pregnancy (day +6) and in those of control pregnant and cyclic rats (left panel). *Significantly increased as compared with cyclic rats after identical treatment (Mann Whitney U test, P < .05). a, Significantly increased as compared with control pregnant rats (Mann Whitney U test, P < .05).
In cyclic rats the glomerular numbers of LFA- 1-positive cells (Fig 6), M A C - 1 - p o s i t i v e cells (Fig 7), and V L A - 4 - p o s i t i v e cells were not affected by SG. In day 14 pregnant rats with SG, the numbers o f L F A - l - p o s i tive, M A C - l - p o s i t i v e , and V L A - 4 - p o s i t i v e ceils per glomerulus were significantly increased as c o m p a r e d with those o f cyclic rats with S G on day 14 (Mann Whitney, P < .05) and with those o f control pregnant rats (Mann Whitney, P < .05). The numbers o f LFA-1-
272
Faaset al
positive, MAC-l-positive, and VLA-4-positive cells per glomerulus returned to control values after pregnancy and were not significantly different from control cyclic rats (Kruskall Wallis). The ligands for the leukocyte adhesion molecules LFA-1/MAC-1 and VLA-4, ICAM-1, and VCAM-1, respectively, could not be detected in either group of rats in the present study (results not shown). DISCUSSION
The present protocol for the induction of an SG (1.0 mg e~-FxlA/200 g body weight to induce passive Heymann nephritis) did not induce albuminuria in cyclic rats; this agrees with the results of previous studies. 18 Also, the animals exhibited only a very mild glomerular inflammatory reaction: the induction of SG per se had no effect on the number of glomerular PMNs and a small but significant effect on the number of glomerular monocyctes on day - 2 only (ie, 6 days after the induction of SG). This mild form of glomerulonephritis can therefore be considered subclinical. Moreover, the present subclinical glomerulonephritis did not prevent the animals from becoming pregnant and did not affect maternal weight, litter size, or litter weight (data not shown). Once the rats were pregnant, however, the urinary albumin excretion of rats with SG increased significantly. This was associated with a more intense glomerular inflammatory response, as apparent from the increased numbers of PMNs and monocytes in the glomeruli on day 14 of pregnancy. Because after parturition both the albuminuria and the inflammatory response returned to the control situation, it appears that there is a specific influence of pregnancy on these parameters in rats with SG. The exacerbating effect of pregnancy on subclinical SG reported here that leads to a glomerular inflammatory reaction and albuminuria is undocumented and may support the notion that pregnancy is a proinflammatory condition. 6 At first glance, the present results seem to contrast with those of Baylis et al, who reported the absence of an effect of pregnancy on urinary protein excretion in both a model of experimentally induced anti-glomerular basement membrane antibody-mediated glomerulonephritis 19 and in the F x l A model of accelerated, passive Heymann nephritis, s0 The experiments of these authors, however, differed from the present ones in that massive proteinuria was already observed in the nonpregnant condition; hence the experiments of Baylis et al and the present ones are not entirely comparable, since the condition of their experimental animals was not "subclinical." With respect to the effect of pregnancy on the GFR, there is also a discrepancy between the present data and data in the literature. In human patients the aggravation of subclinical renal disease during preg-
J Lab Clin Med September 1999
nancy is well documented and is usually attributed to an increased GFR. u In the present study, however, the increased albuminuria of pregnant rats with SG cannot be attributed to an increased GFR, nor can it be attributed to hypertension, since neither pregnancy nor the induction of SG affected these two parameters. This lack of increase in GFR during pregnancy is in line with results of a previous study of our group in which we compared various methods of GFR measurement and found that the (apparent) increase in the GFR in pregnant rats observed with the inulin clearance method is probably due to the (relatively) large saline load that is inherent to this method, not to an effect of pregnancy per se. 14 Glomerular complement deposition has been suggested to be one of the factors responsible for proteinuria in (nonpregnant) rats with passive Heymann nephritis. 21 Indeed, the absence of complement deposition in cyclic rats with passive Heymann nephritis agrees with the absence of albuminuria in these rats. However, there was no complement deposition observed in pregnant SG rats either, despite the significant albuminuria of these animals. Complement deposition, therefore, probably plays no role in the pregnancy-associated albuminuria in rats with SG. Rather, the albuminuria in the present study appears to coincide with an inflammatory response observed in the glomeruli of the pregnant animals with SG. Indeed, there are several lines of evidence, both experimental22 and clinical, 23 indicating that the influx of (activated) monocytes or PMNs may injure the glomerular filtration barrier, leading to proteinuria (for a review see reference 24). Although the numbers of inflammatory cells in the glomeruli of pregnant rats with SG are relatively low, they may still inflict significant damage on the glomerular filtration barrier in the pregnant condition, because during pregnancy, glomerular tissue is much more sensitive to products of activated inflammatory cells.25 The question remains how the influx of inflammatory cells is brought about once the animals become pregnant. The present study does not give an unequivocal answer to this question. The infiltration of inflammatory cells is usually regulated by the expression of endothelial 26 and leukocytic27 adhesion molecules. In the case of pregnant SG rats, the expression of adhesion molecules was seen only on the leukocytes (ie, LFA-1, MAC-l, and VLA-4); the expression of the respective ligands--that is, ICAM-1 and V C A M - 1 - on the endothelial cells could not be demonstrated. This may suggest that other factors within the glomeruli, capable of binding to the leukocytic adhesion molecules - - for example, IgG (in the present study causing the SG) --which can bind to monocytes (via MAC-128) and
J Lab Clin Med Volume 134,Number 3
P M N s (although i n d e p e n d e n t from MAC-129), are involved in the infiltration of i n f l a m m a t o r y cells into the glomeruli of pregnant rats. Also, the local production of chemoattractants 3°,3t m a y p l a y a role, since it has been shown that the glomeruli of rats with passive H e y m a n n nephritis p r o d u c e increased amounts of leukotriene B4, a potent chemoattractant for inflammatory cells 31 when compared with the amounts produced in control rats. 32 One can speculate that the glomeruli o f pregnant rats with passive H e y m a n n nephritis produce increased amounts of leukotriene B4 as compared with cyclic rats with passive Heymann nephritis, but so far this has not been measured. Also, it is possible that i n f l a m m a t o r y cells in pregnant rats are more responsive to leukotriene B4. This notion is supported b y data showing that progesterone, levels of which are increased during pregnancy, 33 increases the responsiveness of inflammatory cells to chemoattractants. 34 SPECULATIONS
This study shows that experimental S G may become aggravated during pregnancy, as reflected by increased urinary albumin excretion. The increased urinary albumin excretion is not associated with an increased G F R or with an increased b l o o d pressure. The present data agree with the notion that in fact the aggravation of SG in the rat develops when the animals become pregnant, b e c a u s e in that condition a g l o m e r u l a r i n f l a m m a t o r y response develops. Although further studies to evaluate the functional role of the leukocytes in the present pregnancy-associated proteinuria are needed, it is tempting to extrapolate the present data to the human situation and speculate that in human patients too, the flaring up of subclinical renal disease results from a g l o m e r u l a r inflammatory reaction that develops during pregnancy. This inflammatory reaction, then, causes damage to the glomerular filtration barrier, which ultimately leads to proteinuria. REFERENCES
l. Apitz K. A study on the generalized Shwartzman phenomenon. J Immunol 1935;29:255-66. 2. Beller FK, Schrdidt EH, Holzgreve W, Hanss J. Septicemia during pregnancy: a study in different species of experimental animals. Am J Obstet Gynecol 1985; 151:967-75. 3. Muller-Berghans G, Schmidt-Ehry B. The role of pregnancy in the induction of the generalized Shwartzman reaction. Am J Obstet Gynecol 1972;114:847-9. 4. Faas MM, Schuiling GA, Baller JFW, Bakker WW. Glomerular inflammation in pregnant rats after infusion of low dose endotoxin: an immunohistological study in experimental pre-eclampsia. Am J Pathol 1995;147:1510-8. 5. Faas MM, Schuiling GA, Bailer JFW, Visscher CA, Bakker WW. A new animal model for human pre-eclampsia: ultralow dose endotoxin infusion in pregnant rats. Am J Obstet Gynecol 1994; 171:158-64.
Faaset al
273
6. Sacks GP, Studena K, Sargent IL, Redman CWG. CDllb expression on circulating neutrophils in pre-eclampsia. Clin Sci 1997;93:187-9. 7. Schuiling GA, Koiter TR, Faas MM. Why pre-eclampsia? Hum Reprod 1997;12:2087-92. 8. Faas MM, Schuiling GA, Baller JFW, Valkhof N, Bakker WW. Aspirin treatment of the low-dose-endotoxin-treated pregnant rat: pathophysiologic and immunohistologic aspects. J Lab Clin Med 1997;130:496-508. 9, Faruqi RM, DiCorleto PE. Mechanism of monocyte recruitments and accumulation. Br Heart J 1993;69(suppl):S19-29. 10. Schreiner GE The role of the macrophage in glomerular injury. Semin Nephrol 1991;11:268-75. 11. Katz AI, Davison JM, Hayslett JR Singsong E, Lindheimer MD. Pregnancy in women with kidney diseases. Kidney Int 1980;18:192-206. 12. Van Leel EHG, Ronco R Verroust P, de Roo GM, Hoedemaeker PJ, De Heer E. Heymann nephritis: a model of human membranous glomerulopathy. A study of the role of additional antigens. Nephrol Dial Transplant 1992;7(suppl 1):1-8. 13. Bakker WW, Mulder I, Lee RLvd, Fleuren GJ, Hoedemaeker PJ. Experimental glomerulonephritis in the rat induced by antibodies directed against tubular antigens. III. In vitro evaluation of cell-mediated immune responses against immune complexes influenced by irmnunosuppressive therapy. Int Arch Allergy Appl Immunol 1977;54:405-13. 14. Faas MM, Schuiling GA, Klok PA, Valkhof N, Bakker WW. The glomerular filtration rate during pregnancy: saline infusion enhances the glomerular filtration rate in the pregnant rat. Kidney Blood Press Res 1996;19: 121-7. 15. Steffens AB. A method for frequent sampling of blood and continuous infusion of fluids in the rat without disturbing the animal. Physiol Behav 1969;4:833-6. 16. Briggs RT, Robinson JM, Karnovski ML, Karnovski MJ. Superoxide production by polymorphonuclear leukocytes: a cytochemical approach. Histochem 1986;84:371-8. 17. Dijkstra CD, Dopp EA, Joling R Kraal G. The heterogeneity of mononuclear phagocytes in lymphoid organs: distinct macrophage subpopulation in the rat recognized by monoclonal antibodies ED1, ED2 and ED3. Immunology 1985;54:589. 18. Iversen BM, Matre R, Ofstad J. The heterologous immune complex glomerulonephritis: a dose dependent glomerulonephritis with acute, latent and chronic phase in long term study. Acta Path Microbiol Immunol Scandinav 1982;90:241-50. 19. Baylis C, Reese K, Wilson CB. Glomerular effects of pregnancy in a model of glomerulonephritis in the rat. Am J Kidney Dis 1989;14:452-60, 20. Baylis C, Deng A, Couser WG. Glomerular hemodynamic effects of late pregnancy in rats with experimental membranous glomerulopathy. J Am Soc Nephrol 1995;6:1197-201. 21. Salant DJ, Quigg RJ, Cybulsky AV. Heymann nephritis: mechanism of renal injury. Kidney Int 1989;35:976-84. 22. Lefkowith JB, Nagamatsu T, Pippin J, Schreiner GF. Role of leukocytes in metabolic and functional derangements of experimental glomerulonephritis. Am J Physiol 1991;261:F213-20. 23. Monga G, Mazzucco G, Barbiano di Belgiojoso G, Busnach G. The presence and possible role of monocyte infiltration in human chronic proliferative glomerulonephritis. Am J Pathol 1979;94:271-84.
274
Faas et al
24. van Goor H, Ding G, Kees-Folts D, Grond J, Schreiner GF, Diamond JR. Macrophages and renal disease. Lab Invest 1994;71:456-64. 25. FaasMM, BakkerWW, Baller JFW, SchuilingGA. Pregnancy enhances the sensitivity of glomerular ecto-ATP-dipbosphohydrolase to products of activated polymorphonuclear leukocytes (PMN). Am J Obstet Gynecol 1999;180:112-3. 26. Munro JM, Pober JS, Cotran RS. Recruitment of neu, trophils in the local endotoxin response: association with de novo endothelial expression of endothelial leukocyte adhesion molecule-1. Lab Invest 1991;64:295-9. 27. Cronstein BN, Kimmel SC, Levin RI, Martiniuk F, Weissmann G. A mechanism for the antiinflammatory effects of corticosteroids: the glucocorticoid receptor regulates leukocyte adhesion to endothelial cells and expression of endothelial-leukocyte adhesion molecule 1 and intercellular adhesion molecule 1. Proc Natl Acad Sci USA 1992;89:9991-5. 28. Brown EJ, Bohnsack JF, Gresham HD. Mechanism of inhibition of immunoglobullinG-mediated phagocytosis by monoclonal antibodies that recognize the Mac-1 antigen.
J Lab Clin Med September 1999
J Clin Invest 1988;81:365-75. 29. Arnaout MA, Pitt J, Cohen HJ, Melamed J, Rosen FS, Colten HR. Deficiency of a granulocyte-membrane glycoprotein (gpl50) in a boy with recurrent bacterial infections. N Engl J Med 1982;306:693-9. 30. Cattell V, Cook HT, Smith J, Salmon JA, Moncada S. Leukotriene B4 production in normal rat glomeruli. Nephrol Dial Transplant 1987;2:154-7. 31. Samuelsson B, Dahlen SE, Lindgren JA, Rouzer CA, Serhan CN. Leukotrienes and lipoxins: structures, biosynthesis, and biological effects. Science 1987;237:1171-6. 32. Lianos EA, Noble B. Glomerular leukotriene synthesis in Heymann nephritis. Kidney Int 1989;36:998-1002. 33. Bast JD, Melampy RM. Luteinizing hormone, prolactin and ovarian 20c~-hydroxysteroiddehydrogenase levels during pregnancy and pseudopregnancy in the rat. Endocrinology 1972;91:1499-505. 34. Miyagi M, Aoyama H, Morishita M, Iwamoto Y. Effects of sex hon~aoneson chemotaxis of human peripheral polymorphonuclear leukocytes and monocytes. J Periodontol 1992;63:28-32.