Selective Cyclooxygenase-2 Inhibition Impairs Glomerular Capillary Healing in Experimental Glomerulonephritis
Masashi Kitahara*,
Frank Eitner*,
Tammo Ostendorf*,
Uta Kunter*,
Ulf Janssen*,
Ralf Westenfeld*,
Katsuyuki Matsui,
Dontscho Kerjaschki and
Jürgen Floege*
*Division of Nephrology and Immunology, University of Aachen, Aachen, Germany, and Department of Pathology, University of Vienna, Vienna, Austria.
Correspondence to Dr. Jürgen Floege, Medizinische Klinik II, Klinikum der RWTH, Pauwelsstrasse 30, D-52074 Aachen, Germany. Phone: 01149-241-8089-530; Fax: 01149-241-8082-446; E-mail: Juergen.Floege{at}post.rwth-aachen.de
ABSTRACT. Selective cyclooxygenase-2 (COX-2) inhibitors haveanti-inflammatory activity and reduce proteinuria in experimentalmembranous glomerulonephritis. Antiangiogenic properties ofCOX-2 inhibitors were recently reported. Whether these propertiesare relevant to the glomerular healing process in inflammatoryglomerular diseases was investigated. For evaluation of theeffects of selective COX-2 inhibitors on the glomerular healingprocess in a rat model of mesangioproliferative glomerulonephritis(induced by anti-Thy 1.1 antibody), a selective COX-2 inhibitor(rofecoxib or celecoxib) or vehicle was administered daily fromday 1 after disease induction until euthanasia on day 6. Additionalnephritic rats were treated with rofecoxib or vehicle from day1 to day 10 and were monitored until day 28. Selective COX-2inhibition led to significant increases in mesangiolysis (upto +71%) on days 2 and 6 and in albuminuria (up to 3.1-fold)on day 6. This augmentation of glomerular capillary damage wasassociated with rarefaction of glomerular endothelial cells,whereas the proliferation and activation of mesangial cellswere not affected. No significant effects on the glomerularinflux of polymorphonuclear neutrophils or the infiltrationand proliferation of monocytes/macrophages at day 2 were noted.These effects were independent of systemic hemodynamic features,because rofecoxib did not affect systolic BP on day 2 or 5.Nephritic rats treated with rofecoxib for 10 d demonstratedpersistent glomerular injury at day 28, as indicated by increasedalbuminuria (10-fold) and mesangial type IV collagen deposition(+24%). In normal rats, 5-d administration of rofecoxib failedto induce albuminuria or morphologic renal damage. In conclusion,selective COX-2 inhibitors impair glomerular capillary repairafter mesangiolysis in rats with anti-Thy 1.1 glomerulonephritis.These data suggest that selective COX-2 inhibitors should beused with caution among patients with inflammatory endocapillaryglomerular disorders.
Cyclooxygenase (COX) is the key enzyme in the biosynthesis ofprostaglandins. In physiologic states, prostaglandins are criticallyinvolved in the maintenance of gastric mucosal integrity andthe regulation of renal vascular contractility, sodium and waterbalance, and BP (1). Therefore, the therapeutic use of nonselectiveCOX inhibitors as analgesics is limited by their adverse effects,particularly gastrointestinal ulcers. COX exists in two distinctisoforms, i.e., COX-1 and COX-2 (2,3). COX-1 is constitutivelyexpressed in many tissues and seems to be responsible for themajority of prostaglandin production, whereas COX-2 is normallyundetectable in most tissues but is highly overexpressed atsites of inflammation (4,5). Selective COX-2 inhibitors havenow reached the marketplace, and potent analgesic activity withoutserious gastrointestinal side effects has been reported (6).
Fetal and adult kidneys are among the few organs that constitutivelyexpress COX-2 (7–9), suggesting that expression of COX-2is necessary for the maintenance of normal renal architectureand function and for postnatal renal development (10–12).Not surprisingly, COX-2 inhibitor-induced acute renal failure,similar to that observed with nonselective COX inhibitors, isbeing increasingly recognized (13). However, renoprotectiveand antiproteinuric effects of selective COX-2 inhibitors wereobserved in rodent models of renal injury (14,15).
A hitherto unrecognized activity of selective COX-2 inhibitors,namely antiangiogenic properties, was recently reported. SelectiveCOX-2 inhibitors were demonstrated to suppress angiogenesisin tumors by inhibiting proangiogenic factor expression andendothelial cell growth (16,17). Another study demonstrateda delay in the healing of gastric ulcers in rats receiving aselective COX-2 inhibitor, which was linked to the antiangiogenicproperties of the compound (18).
On the basis of these observations, we asked whether selectiveCOX-2 inhibitors might also impair the capillary healing thatoccurs in the course of glomerulonephritis, such as anti-Thy1.1 nephritis in rats (19,20). In this model, antibody- andcomplement-mediated acute mesangiolysis is followed by capillaryrepair, involving mesangial and glomerular endothelial cellproliferation as well as increased matrix synthesis (20–22).A unique feature of this model is that glomeruli spontaneouslyrecover during a resolution phase (21).
Compounds
Rofecoxib (also known as MK-0966), a highly selective COX-2inhibitor, was obtained from Merck Sharp & Dohme GmbH (Haar,Germany). Another selective COX-2 inhibitor, celecoxib (SC-58635),was obtained from Pfizer GmbH (Karlsruhe, Germany).
Experimental Design
All animal experiments were approved by the local review boards.Anti-Thy 1.1 glomerulonephritis was induced in male Wistar rats(160 to 180 g at the start of the experiments; Charles River,Sulzfeld, Germany) as described (23). The rats were treatedby oral gavage, beginning 18 h after disease induction and continuingonce daily up to day 5, with rofecoxib (1 mg/kg, n = 11; 10mg/kg, n = 10), celecoxib (30 mg/kg, n = 9; 50 mg/kg, n = 8),or vehicle alone (n = 9). The choice of doses was based on variousstudies in rats that revealed significant anti-inflammatoryeffects in the dose ranges tested (24–29). Furthermore,COX-1 activity was not affected at the doses tested (30,31).The experimental doses reflected human clinical doses (up to1 mg/kg) for rofecoxib, whereas higher doses were used for celecoxib,in comparison with human doses (up to 12 mg/kg).
Renal biopsies were obtained by intravital biopsy at day 2 andby post mortem biopsy at day 6 after disease induction. Twenty-four-hoururine collections were performed from day 5 to day 6. The ratsreceived intraperitoneal injections of the thymidine analogue5-bromo-2'-deoxyuridine (BrdU) (100 mg/kg; Sigma-Aldrich ChemieGmbH, Deisenhofen, Germany) 4 h before the day 2 intravitalbiopsy. BP was measured on day -1 before disease induction andon days 2 and 5 after disease induction.
Anti-Thy 1.1 glomerulonephritis was also induced in 10 ratsthat were treated with 10 mg/kg rofecoxib (n = 5) or vehiclealone (n = 5) from day 1 to day 10 and were then monitored untilday 28. The rats underwent an intravital renal biopsy at day6 after disease induction. BrdU was injected intraperitoneally4 h before the intravital biopsy. In addition, normal rats weretreated with rofecoxib at 1 or 10 mg/kg (n = 3 each) for 5 d,after which albuminuria was measured and renal tissue was obtained.
Renal Morphologic Assessments
Tissue for light microscopy and immunoperoxidase staining wasfixed in methyl Carnoy’s solution and embedded in paraffin.In periodic acid-Schiff-stained sections, the numbers of mitosesand polymorphonuclear neutrophils (PMN) in 50 to 100 glomerulartufts were determined as described (19,23). By using a 1000-foldmagnification, mitoses were differentiated into endothelialcells and nonendothelial cells, as described (19). Mesangiolysiswas graded on a semiquantitative scale (0, no mesangiolysis;1, segmental mesangiolysis; 2, global mesangiolysis; 3, microaneurysm),as described (19,22,32).
Immunoperoxidase Staining
Four-micrometer sections of methyl Carnoy’s solution-fixedbiopsy tissue were processed as described previously (22,32).To detect glomerular endothelial cells, we used JG-12, a monoclonalantibody to rat endothelial cells. The specificity of this antibodyfor rat renal microvascular endothelial cells was recently reported(33). To date, experimental studies have failed to demonstrateJG-12 antigen regulation at the single-cell level in rats withrenal disease (33). For detection of COX-2 immunoreactivity,tissue was stained with polyclonal rabbit anti-murine COX-2serum (Cayman Chemical, Ann Arbor, MI) (7). Other primary antibodiesincluded a murine monoclonal antibody to -smooth muscle actin(clone 1A4; Dako, Glostrup, Denmark), a murine monoclonal IgGantibody to a cytoplasmic antigen present in monocytes, macrophages,and dendritic cells (clone ED-1; Camon, Wiesbaden, Germany),and affinity-purified, polyclonal goat anti-human/bovine typeIV collagen IgG that had been preabsorbed with rat erythrocytes(Biozol, Birmingham, AL) (19,23,30). The sections were incubatedsequentially with biotinylated horse anti-mouse Ig antibody(Vector Laboratories, Burlingame, CA), goat anti-rabbit Ig antibody(Vector), or rabbit anti-goat Ig antibody (Zymed, San Francisco,CA). The ABC-Elite reagent (Vector) was used, with 3,3'-diaminobenzidine(with nickel chloride enhancement) as the chromogen. Sectionswere counterstained with methyl green.
For determination of mean numbers of proliferating cells, PMN,and infiltrating monocytes/macrophages, 30 to 100 consecutivecross-sections of glomeruli were evaluated and mean values perkidney were calculated. Proliferating cells (BrdU-positive cells)were again differentiated into endothelial and nonendothelialcells, as described above. The immunostaining for -smooth muscleactin, JG-12, and type IV collagen was evaluated by using apoint-counting method. For this, a grid composed of 100 dotswas superimposed on consecutive glomeruli (range, 25 to 30 glomeruli;magnification, x1000) and the numbers of dots overlying stainedareas were counted (33). Sections immunostained for COX-2 wereexamined by using the same grid but at 400-fold magnification;the numbers of dots overlying stained renal cortical areas wereevaluated in at least 100 consecutive fields for each kidney.
Immunohistochemical Double-Staining
Double-immunostaining for identification of the type of proliferatingcells was performed as reported previously (19,23). First, sectionswere stained for proliferating cells with a murine monoclonalantibody against BrdU-containing nuclease (clone BU-1; Amersham,Braunschweig, Germany), in Tris-buffered saline, using an immunoperoxidaseprocedure. 3,3'-diaminobenzidine was used as a substrate, resultingin a black product. Sections were then incubated with the IgG1monoclonal antibody ED-1 (directed against monocytes/macrophages)or the monoclonal antibody JG-12 (directed against rat endothelialcells). Aminoethylcarbazole was used as a substrate, resultingin a red product. Cells were identified as proliferating monocytes/macrophagesor proliferating glomerular endothelial cells if they demonstratedpositive nuclear staining for BrdU and the nucleus was completelysurrounded by cytoplasm positive for ED-1 or JG-12 antigen,respectively. Negative control assays included omission of eitherof the primary antibodies, in which cases no double-stainingwas noted.
Electron Microscopy
Kidneys were removed after in situ perfusion with Dulbecco’smodified Eagle’s medium/F-12 medium and 4% paraformaldehyde/phosphatebuffer. Blocks of renal tissue were immersed in 4% formaldehyde.The samples were then embedded in Epon and processed for transmissionelectron microscopy by using standard procedures.
Miscellaneous Measurements
Urinary albumin levels were determined on a 96-well enzyme-linkedimmunosorbent assay plate, using a peroxidase-conjugated anti-ratalbumin antibody (ICN Biomedical, Eschwege, Germany), as described(34). All measurements were performed in duplicate. BP measurementswere performed with the tail cuff method, using a programmedsphygmomanometer (BP-98A; Softron, Tokyo, Japan) (35).
Statistical Analyses
Values are expressed as mean ± SD unless otherwise noted.Statistical significance (defined as P < 0.05) was evaluatedwith Mann-Whitney U tests or paired t tests, where appropriate.
Effects of Selective COX-2 Inhibitors on Mesangiolytic Injury and Albuminuria
Mesangiolytic changes were more pronounced at day 2 (i.e., 48h) after disease induction in nephritic rats treated with rofecoxibor celecoxib, compared with rats treated with vehicle alone(Figure 1A). Furthermore, in rats treated with rofecoxib, adose of 1 mg/kg led to a marked increase in the frequency ofglomerular microaneurysms at day 2 (Figure 1B). The mesangiolyticchanges and the frequency of glomerular microaneurysms werestill more pronounced at day 6 for most groups that receivedselective COX-2 inhibitors, compared with control rats treatedwith vehicle alone. Apparent dose dependence in the degree ofmesangiolysis and the frequency of glomerular microaneurysmsat day 6 was noted for the celecoxib-treated animals (Figure 1, A and B).Rats treated with COX-2 inhibitors not only developedmore severe glomerular pathologic conditions but also exhibitedfunctional deterioration, as evidenced by an increase in urinaryalbumin excretion on day 6 (Figure 1C).
Figure 1. Effects of selective cyclooxygenase-2 (COX-2) inhibitors on mesangiolytic injury and albuminuria. (A) Mesangiolysis on days 2 and 6 after disease induction. (B) Frequency of microaneurysms on days 2 and 6 after disease induction. (C) Albuminuria on day 6 after disease induction (n = 8 to 11 for each group).
Effects of Selective COX-2 Inhibitors on Glomerular Endothelial Cell Behavior
Endothelial cell injury and repair are of importance in anti-Thy1.1 glomerulonephritis. Selective COX-2 inhibition led to asignificant rarefaction of glomerular endothelial cells at day2, compared with rats treated with vehicle alone, as indicatedby a rarefaction of glomerular JG-12 antigen expression (Figure 2).The proliferation of endothelial cells and nonendothelialcells was quantified by BrdU labeling and counting of mitoticfigures. There was no significant difference in BrdU incorporationat day 2 between rats receiving selective COX-2 inhibitors andthose receiving vehicle alone (data not shown). Similarly, thenumbers of mitotic figures in glomerular endothelial cells andnonendothelial cells were not different between the groups.In an additional experiment, proliferating glomerular endothelialcells were also identified by double-immunostaining for BrdUand JG-12. This experiment also demonstrated no significantdifference between rats treated with selective COX-2 inhibitorsand those treated with vehicle alone (data not shown). Electron-microscopicevidence of glomerular endothelial damage was detected in nephriticrats at day 2 (Figure 3A) and seemed to be increased in ratsreceiving rofecoxib (Figure 3B).
Figure 2. Effects of selective COX-2 inhibitors on glomerular endothelial cells. (A to C) Renal glomerular endothelial cell staining with JG-12 in normal rats (A), on day 2 in nephritic rats receiving vehicle alone (B), and on day 2 in nephritic rats receiving rofecoxib at a dose of 1 mg/kg (C). Magnification, x600. A decrease in glomerular endothelial cell staining was noted on day 2 in rats treated with selective COX-2 inhibitors, compared with vehicle alone. (D) Quantitative evaluation, by grid counting, of glomerular JG-12 immunostaining on days 2 and 6 after disease induction (n = 5 to 9 for each group).
Figure 3. Effects of selective COX-2 inhibitors on glomerular endothelial morphologic features. (A) Nephritic rats on day 2 exhibit evidence of glomerular endothelial damage, such as edema of endothelial cell bodies, whereas the morphologic features of the fenestrated lining remain intact. Magnification, x4000. (B) Endothelial damage is aggravated in rats treated with 10 mg/kg rofecoxib, as evidenced by extensive cytoplasmic blebbing of endothelial and visceral epithelial cells. In addition, the endothelial lining is partially affected by intracellular edema. Magnification, x2500.
Effects of Selective COX-2 Inhibitors on Glomerular Mesangial Cell Activation
We also investigated the effects of selective COX-2 inhibitionon mesangial cell activation in anti-Thy 1.1 nephritis. Glomerular-smooth muscle actin expression (a marker of mesangial cellactivation) did not differ between rats receiving selectiveCOX-2 inhibitors and those receiving vehicle alone at days 2and 6 after disease induction (Figure 4).
Figure 4. Effects of selective COX-2 inhibitors on glomerular mesangial cell activation. The quantitative evaluation, by grid counting, of the de novo glomerular expression of -smooth muscle actin on days 2 and 6 after disease induction is presented (n = 5 to 9 for each group).
Effects on Glomerular Leukocyte Influx and Activation
One of the early events in anti-Thy 1.1 nephritis is an influxof PMN into glomeruli, followed by monocytes/macrophages. At48 h after disease induction, glomerular accumulation of PMN(Figure 5A) and the numbers of monocytes/macrophages in glomeruli(Figure 5B) were not significantly affected in rats receivingCOX-2 inhibitors. Selective COX-2 inhibitors also did not affectthe intraglomerular proliferation of monocytes/macrophages,as determined by double-immunostaining for the ED-1 antigenand BrdU, at day 2 (Figure 5C). The glomerular influx of monocytes/macrophagesat day 6 was significantly increased in rats treated with thelow dose of rofecoxib (1 mg/kg) or the high dose of celecoxib(50 mg/kg) (Figure 5B).
Figure 5. Effects of selective COX-2 inhibitors on glomerular leukocyte accumulation and activation. (A) Numbers of infiltrating glomerular polymorphonuclear neutrophils (PMN) on days 2 and 6 after disease induction (n = 8 to 11 for each group). (B) Numbers of infiltrating ED-1-positive monocytes/macrophages on days 2 and 6 after disease induction (n = 8 to 11 for each group). (C) Numbers of proliferating glomerular ED-1-positive monocytes/macrophages on days 2 and 6 after disease induction (n = 5 to 9 for each group). BrdU, 5-bromo-2'-deoxyuridine.
Effects of Rofecoxib on Systemic Hemodynamics
To exclude the possibility that differences between COX-2 inhibitor-and vehicle-treated animals were attributable to effects onBP, we determined the effects of rofecoxib on systemic hemodynamics.No significant effect of rofecoxib on systolic BP was notedon days 2 and 5 after disease induction, compared with ratstreated with vehicle alone (Table 1). COX-2 inhibitor treatmentalso did not affect the rate of body weight gain or food orwater intake (data not shown).
Chronic Effects of Rofecoxib on the Resolution of Renal Functional and Pathologic Damage
The anti-Thy 1.1 glomerulonephritis model used is characterizedby complete resolution of the injury within 4 to 8 wk afterantibody injection. When nephritic rats were monitored for 28d after disease induction, marked glomerular accumulation oftype IV collagen persisted only in rats treated transientlywith rofecoxib and not in the vehicle-treated control group(Table 2). This increased type IV collagen deposition was associatedwith an increase in albuminuria for the rats treated with rofecoxib(Table 2).
Table 2. Effects of rofecoxib on the resolution of renal functional and pathologic changes on day 28a
Expression of COX-2 Immunoreactive Protein
COX-2 expression in the renal cortex of normal and vehicle-treatednephritic rats was restricted to epithelial cells in a smallnumber of tubular segments. According to anatomic criteria,these segments were the macula densa and cortical thick ascendinglimb of Henle’s loop, in agreement with previous descriptionsof COX-2-immunoreactive cells within renal tissues (7). Tubularexpression of COX-2 was dose-dependently increased on days 2and 6 in the nephritic rats treated with rofecoxib. Treatmentwith celecoxib also increased COX-2 expression, but the effectsdid not demonstrate dose dependence at the doses tested (Figure 6).In glomeruli, there were few cells with COX-2 immunoreactivityat day 6 in nephritic rats treated with COX-2 inhibitors, whereaswe failed to observe COX-2-positive cells at day 6 in glomeruliof nephritic rats treated with vehicle alone or at day 2 inany group (data not shown).
Figure 6. Renal cortical expression of COX-2. (A) COX-2 expression in the renal cortex of a vehicle-treated nephritic rat on day 6 after disease induction. Magnification, x600. (B) COX-2 expression in the renal cortex of a rat treated with rofecoxib at a dose of 10 mg/kg, on day 6 after disease induction. Magnification, x600. COX-2 expression was increased in tubular cells in the nephritic rat treated with rofecoxib. (C) Semiquantitative evaluation, by grid counting, of COX-2 expression in the renal cortex on days 2 and 6 (n = 5 to 9 for each group).
Effects of Selective COX-2 Inhibition in Normal Rats
For normal rats, 5 d of treatment with 1 or 10 mg/kg rofecoxibfailed to induce albuminuria (Table 3). Furthermore, glomerularendothelial expression of the JG-12 antigen did not change.As expected, rofecoxib dose-dependently increased the intrarenalexpression of COX-2 (Table 3).
In this study, we investigated the effects of specific COX-2inhibition in the anti-Thy 1.1 model of mesangioproliferativeglomerulonephritis. In this model, glomerular disease is initiatedby selective immune-mediated mesangial cell damage, resultingin mesangiolysis, but secondary damage to the glomerular capillaries,with the formation of microaneurysms and endothelial damage,is also well recognized (20). This initial phase of the diseaseis quickly followed by overshooting mesangial cell proliferation,which leads to a histologic pattern that resembles human mesangioproliferativeglomerulonephritis. Finally, at least for anti-Thy 1.1 nephritisinduced by the monoclonal antibody OX-7, spontaneous restorationof the normal glomerular architecture ensues.
The major finding of this study was that the administrationof different COX-2 inhibitors significantly aggravated earlyglomerular capillary injury in anti-Thy 1.1 nephritis, as evidencedby increased mesangiolysis, microaneurysm formation, and albuminuria.The pathophysiologic relevance of this early augmentation ofglomerular damage by COX-2 inhibitors was emphasized by ourfindings in the long-term study, in which transient COX-2 inhibitionresulted in persistent glomerular damage at 4 wk after diseaseinduction, i.e., at a time when most glomerular changes haveusually resolved.
The aggravation of early glomerular injury in anti-Thy 1.1 nephritisby COX-2 inhibitors is in accordance with our underlying hypothesis,namely that COX-2 inhibitors may affect glomerular capillaryhealing via their antiangiogenic properties (16,17). As describedearlier, glomerular capillary healing in anti-Thy 1.1 nephritisexhibits features of angiogenesis, such as glomerular endothelialcell proliferation and elongation and mesangial cell proliferation(20). In this study, increased mesangiolysis and microaneurysmformation in rats treated with COX-2 inhibitors represent indirectevidence of augmented intracapillary damage. More importantis the rarefaction of staining with the JG-12 antibody. Thismonoclonal antibody, which recognizes a currently unknown antigen,was recently demonstrated to allow the sensitive detection ofendothelial cells in normal and diseased rat glomeruli (33).Rarefaction of the JG-12 staining pattern thus provides more-directevidence for increased damage to endothelial cells in the COX-2inhibitor-treated rats. Of note, we were unable to documentreduced proliferation of glomerular endothelial cells in theseexperimental groups, suggesting that mechanisms other than antimitogenicactions of the COX-2 inhibitors were underlying the increasedendothelial damage. In this respect, Iruela-Arispe et al. (20)observed that, in anti-Thy 1.1 nephritis, glomerular endothelialcells often appeared elongated, with extended processes, duringthe course of capillary (in particular, microaneurysm) repairand endothelial cells encircled areas of mesangial cellularity.Therefore, one mechanism by which COX-2 inhibitors might havereduced the JG-12 staining pattern could be related to inhibitionof endothelial cell migration and shape changes. This idea issupported by data obtained with cultured umbilical cord venousendothelial cells, as well as human renal microvascular endothelialcells; COX-2 inhibition reduced endothelial cell migration (36,37),possibly by preventing the integrin-dependent activation ofsmall GTPases (38). This may also explain why inhibition ofCOX-2 from day 1 to day 10 in our long-term study resulted inpersistent glomerular damage, because increased proliferationand migration of glomerular endothelial cells have been documenteduntil day 14 of anti-Thy 1.1 nephritis (20).
Could mechanisms other than the actions of COX-2 inhibitorson glomerular endothelial cells underlie our observations? Itseems unlikely that COX-2 inhibitors altered the extent of immunologicmesangial damage, because oral treatment was initiated 18 hafter disease induction (a time at which anti-Thy 1.1 antibodybinding in the glomeruli has already reached its maximum). Asecond possibility is that COX-2 inhibition altered systemicBP and thus hypertension was superimposed on immunologic injury.We failed to demonstrate any effect of COX-2 inhibition on systemicBP, either in the very early phase or during the peak of mesangialcell proliferation on day 6. This finding is in agreement withdata obtained by Wang et al. (14) in rats with renal ablationand by Komers et al. (39) in streptozotocin-diabetic rats. Importantly,in the latter study, renal plasma flow was also not affectedby acute COX-2 inhibition (39). In rats with renovascular hypertension,COX-2 inhibition even decreased BP and plasma renin activity(40). However, at least some instances in which selective COX-2inhibitors elevated BP have been documented (41). A third possibilityis related to the effects of COX-2 inhibition on intrinsic glomerularcells other than endothelial cells, particularly mesangial cells.This possibility also is unlikely, because the glomerular expressionof -smooth muscle actin, which is a very sensitive marker ofmesangial damage (42), was not affected in the studies presentedhere. Finally, COX-2 inhibitors might have affected glomerularleukocyte influx and thus influenced the course of the disease.Indeed, we noted a nonsignificant trend toward increases inglomerular PMN counts on day 2 for the low-dose rofecoxib-treatedgroup. Although we cannot formally exclude the possibility thatthis was a direct effect of COX-2 inhibition, a more likelyexplanation for this observation is that increased PMN numbersin this group were the result, rather than the cause, of increasedcapillary damage. In contrast to PMN, early monocyte/macrophagecounts were not affected by COX-2 inhibition. The enhancementof glomerular monocyte/macrophage counts for the two COX-2 inhibitor-treatedgroups on day 6 after disease induction may be accounted forby the finding that selective COX-2 inhibitors can increaseglomerular monocyte chemoattractant protein-1 mRNA expressionin anti-Thy 1.1 nephritis on day 5 (43).
To demonstrate that COX-2 inhibition indeed affected COX-2 invivo, we also investigated the renal expression of COX-2 inthe various groups. The COX-2 expression pattern in our untreatednephritic rats corresponded well to that observed in normaland diabetic rats (7,39), as well as that observed in anti-Thy1.1 nephritis (44). Our data are at variance with those of Hiroseet al. (45), who, using a different antibody and technique,demonstrated strong COX-2 expression in glomerular epithelialcells of rats with anti-Thy 1.1 nephritis. Our data also indicatethe existence of a feedback mechanism in vivo, because COX-2inhibition uniformly led to upregulated (but not redistributed)expression of the enzyme within the kidney.
The findings of this study are at variance with conclusionsderived from other rodent models of renal disease. In the renalablation model, COX-2 inhibition reduced proteinuria and glomerulosclerosis(14). In the passive Heymann nephritis model of human membranousnephropathy, COX-2 inhibition with the relatively selectiveinhibitor flosulide also resulted in a decrease in proteinuria(15). The major difference between those two models and theanti-Thy 1.1 nephritis model is the extent of early intracapillarydamage, particularly the extent of damage to glomerular endothelialand mesangial cells. Although some endothelial cell damage andearly mesangial cell activation have been documented (particularlyin the renal ablation model) (46,47), both are minor in comparisonwith those in the anti-Thy 1.1 model. These considerations leadus to conclude that COX-2 inhibition may be harmful in instancesof pronounced glomerular capillary damage, such as endocapillaryor mesangiocapillary glomerulonephritis, transplant glomerulopathy,or thrombotic microangiopathies.
Acknowledgments
The technical assistance of Gabi Dietzel, Gerti Minartz, KerstinSchenk, and Andrea Cosler is gratefully acknowledged. This studywas supported by the German Research Foundation (SFB 542/C7).We thank Karin Kluge for support in measuring BP in rats.
Needleman P, Turk J, Jakschik BA, Morrison AR, Lefkowith JB: Arachidonic acid metabolism. Annu Rev Biochem 55: 69–102, 1986[CrossRef][Medline]
Kujubu DA, Fletcher BS, Varnum BC, Lim RW, Herschman HR: TIS10, a phorbol ester tumor promoter-inducible mRNA from Swiss 3T3 cells, encodes a novel prostaglandin synthase/cyclooxygenase homologue. J Biol Chem 266: 12866–12872, 1991[Abstract/Free Full Text]
Xie W, Chipman JG, Robertson DL, Erikson RL, Simmons DL: Expression of a mitogen-responsive gene encoding prostaglandin synthase is regulated by mRNA splicing. Proc Natl Acad Sci USA 88: 2692–2696, 1991[Abstract/Free Full Text]
Vane JR, Mitchell JA, Appleton I, Tomlinson A, Bishop-Bailey D, Croxtall J, Willoughby DA: Inducible isoforms of cyclooxygenase and nitric-oxide synthase in inflammation. Proc Natl Acad Sci USA 91: 2046–2050, 1994[Abstract/Free Full Text]
Seibert K, Zhang Y, Leahy K, Hauser S, Masferrer J, Isakson P: Distribution of COX-1 and COX-2 in normal and inflamed tissues. Adv Exp Med Biol 400A: 167–170, 1997
Simon LS, Lanza FL, Lipsky PE, Hubbard RC, Talwalker S, Schwartz BD, Isakson PC, Geis GS: Preliminary study of the safety and efficacy of SC-58635, a novel cyclooxygenase 2 inhibitor: Efficacy and safety in two placebo-controlled trials in osteoarthritis and rheumatoid arthritis, and studies of gastrointestinal and platelet effects. Arthritis Rheum 41: 1591–1602, 1998[CrossRef][Medline]
Harris RC, McKanna JA, Akai Y, Jacobson HR, Dubois RN, Breyer MD: Cyclooxygenase-2 is associated with the macula densa of rat kidney and increases with salt restriction. J Clin Invest 94: 2504–2510, 1994
Zhang MZ, Wang JL, Cheng HF, Harris RC, McKanna JA: Cyclooxygenase-2 in rat nephron development. Am J Physiol 273: F994–F1002, 1997[Abstract/Free Full Text]
Kömhoff M, Grone HJ, Klein T, Seyberth HW, Nusing RM: Localization of cyclooxygenase-1 and -2 in adult and fetal human kidney: Implication for renal function. Am J Physiol 272: F460–F468, 1997[Abstract/Free Full Text]
Morham SG, Langenbach R, Loftin CD, Tiano HF, Vouloumanos N, Jennette JC, Mahler JF, Kluckman KD, Ledford A, Lee CA, Smithies O: Prostaglandin synthase 2 gene disruption causes severe renal pathology in the mouse. Cell 83: 473–482, 1995[CrossRef][Medline]
Dinchuk JE, Car BD, Focht RJ, Johnston JJ, Jaffee BD, Covington MB, Contel NR, Eng VM, Collins RJ, Czerniak PM, Gorry SA, Trzaskos JM: Renal abnormalities and an altered inflammatory response in mice lacking cyclooxygenase II. Nature (Lond) 378: 406–409, 1995[CrossRef][Medline]
Norwood VF, Morham SG, Smithies O: Postnatal development and progression of renal dysplasia in cyclooxygenase-2 null mice. Kidney Int 58: 2291–2300, 2000[CrossRef][Medline]
Perazella MA, Eras J: Are selective COX-2 inhibitors nephrotoxic? Am J Kidney Dis 35: 937–940, 2000[Medline]
Wang JL, Cheng HF, Shappell S, Harris RC: A selective cyclooxygenase-2 inhibitor decreases proteinuria and retards progressive renal injury in rats. Kidney Int 57: 2334–2342, 2000[CrossRef][Medline]
Blume C, Heise G, Mühlfeld A, Bach D, Schror K, Gerhardz CD, Grabensee B, Heering P: Effect of flosulide, a selective cyclooxygenase 2 inhibitor, on passive Heymann nephritis in the rat. Kidney Int 56: 1770–1778, 1999[CrossRef][Medline]
Sawaoka H, Tsuji S, Tsujii M, Gunawan ES, Sasaki Y, Kawano S, Hori M: Cyclooxygenase inhibitors suppress angiogenesis and reduce tumor growth in vivo. Lab Invest 79: 1469–1477, 1999[Medline]
Jones MK, Wang H, Peskar BM, Levin E, Itani RM, Sarfeh IJ, Tarnawski AS: Inhibition of angiogenesis by nonsteroidal anti-inflammatory drugs: Insight into mechanisms and implications for cancer and ulcer healing. Nat Med 5: 1418–1423, 1999[CrossRef][Medline]
Shigeta J, Takahashi S, Okabe S: Role of cyclooxygenase-2 in the healing of gastric ulcers in rats. J Pharmacol Exp Ther 286: 1383–1390, 1998[Abstract/Free Full Text]
Ostendorf T, Kunter U, Eitner F, Loos A, Regele H, Kerjaschki D, Henninger DD, Janjic N, Floege J: VEGF(165) mediates glomerular endothelial repair. J Clin Invest 104: 913–923, 1999[Medline]
Iruela-Arispe L, Gordon K, Hugo C, Duijvestijn AM, Claffey KP, Reilly M, Couser WG, Alpers CE, Johnson RJ: Participation of glomerular endothelial cells in the capillary repair of glomerulonephritis. Am J Pathol 147: 1715–1727, 1995[Abstract]
Floege J, Johnson RJ, Gordon K, Iida H, Pritzl P, Yoshimura A, Campbell C, Alpers CE, Couser WG: Increased synthesis of extracellular matrix in mesangial proliferative nephritis. Kidney Int 40: 477–488, 1991[Medline]
Johnson RJ, Garcia RL, Pritzl P, Alpers CE: Platelets mediate glomerular cell proliferation in immune complex nephritis induced by anti-mesangial cell antibodies in the rat. Am J Pathol 136: 369–374, 1990[Abstract]
Floege J, Ostendorf T, Janssen U, Burg M, Radeke HH, Vargeese C, Gill SC, Green LS, Janjic N: Novel approach to specific growth factor inhibition in vivo: Antagonism of platelet-derived growth factor in glomerulonephritis by aptamers. Am J Pathol 154: 169–179, 1999[Abstract/Free Full Text]
Chan CC, Boyce S, Brideau C, Charleson S, Cromlish W, Ethier D, Evans J, Ford-Hutchinson AW, Forrest MJ, Gauthier JY, Gordon R, Gresser M, Guay J, Kargman S, Kennedy B, Leblanc Y, Leger S, Mancini J, O’Neill GP, Ouellet M, Patrick D, Percival MD, Perrier H, Prasit P, Rodger I, Tagari P, Therien M, Vickers P, Visco D, Wang Z, Webb J, Wong E, Xu LJ, Young RN, Zamboni R, Riendeau D: Rofecoxib [Vioxx, MK-0966; 4-(4'-methylsulfonylphenyl)-3-phenyl-2-(5H)-furanone], a potent and orally active cyclooxygenase-2 inhibitor: Pharmacological and biochemical profiles. J Pharmacol Exp Ther 290: 551–560, 1999[Abstract/Free Full Text]
Brzozowski T, Konturek PC, Konturek SJ, Sliwowski Z, Drozdowicz D, Stachura J, Pajdo R, Hahn EG: Role of prostaglandins generated by cyclooxygenase-1 and cyclooxygenase-2 in healing of ischemia-reperfusion-induced gastric lesions. Eur J Pharmacol 385: 47–61, 1999[CrossRef][Medline]
Zhang Y, Shaffer A, Portanova J, Seibert K, Isakson PC: Inhibition of cyclooxygenase-2 rapidly reverses inflammatory hyperalgesia and prostaglandin E2 production. J Pharmacol Exp Ther 283: 1069–1075, 1997[Abstract/Free Full Text]
Smith CJ, Zhang Y, Koboldt CM, Muhammad J, Zweifel BS, Shaffer A, Talley JJ, Masferrer JL, Seibert K, Isakson PC: Pharmacological analysis of cyclooxygenase-1 in inflammation. Proc Natl Acad Sci USA 95: 13313–13318, 1998[Abstract/Free Full Text]
Jett MF, Ramesha CS, Brown CD, Chiu S, Emmett C, Voronin T, Sun T, O’Yang C, Hunter JC, Eglen RM, Johnson RM: Characterization of the analgesic and anti-inflammatory activities of ketorolac and its enantiomers in the rat. J Pharmacol Exp Ther 288: 1288–1297, 1999[Abstract/Free Full Text]
Tibble JA, Sigthorsson G, Foster R, Bjarnason I: Comparison of the intestinal toxicity of celecoxib, a selective COX-2 inhibitor, and indomethacin in the experimental rat. Scand J Gastroenterol 35: 802–807, 2000[CrossRef][Medline]
Matheson AJ, Figgitt DP: Rofecoxib: A review of its use in the management of osteoarthritis, acute pain and rheumatoid arthritis. Drugs 61: 833–865, 2001[CrossRef][Medline]
Jackson LM, Hawkey CJ: COX-2 selective nonsteroidal anti-inflammatory drugs: Do they really offer any advantages? Drugs 59: 1207–1216, 2000[CrossRef][Medline]
Burg M, Ostendorf T, Mooney A, Koch KM, Floege J: Treatment of experimental mesangioproliferative glomerulonephritis with non-anticoagulant heparin: Therapeutic efficacy and safety. Lab Invest 76: 505–516, 1997[Medline]
Kim YG, Suga SI, Kang DH, Jefferson JA, Mazzali M, Gordon KL, Matsui K, Breiteneder-Geleff S, Shankland SJ, Hughes J, Kerjaschki D, Schreiner GF, Johnson RJ: Vascular endothelial growth factor accelerates renal recovery in experimental thrombotic microangiopathy. Kidney Int 58: 2390–2399, 2000[CrossRef][Medline]
Magnotti RA Jr, Stephens GW, Rogers RK, Pesce AJ: Microplate measurement of urinary albumin and creatinine. Clin Chem 35: 1371–1375, 1989[Abstract/Free Full Text]
Sutoo D, Akiyama K: Effect of dopamine receptor antagonists on the calcium-dependent central function that reduces blood pressure in spontaneously hypertensive rats. Neurosci Lett 269: 133–136, 1999[CrossRef][Medline]
Tsujii M, Kawano S, Tsuji S, Sawaoka H, Hori M, DuBois RN: Cyclooxygenase regulates angiogenesis induced by colon cancer cells. Cell 93: 705–716, 1998[CrossRef][Medline]
Daniel TO, Liu H, Morrow JD, Crews BC, Marnett LJ: Thromboxane A2 is a mediator of cyclooxygenase-2-dependent endothelial migration and angiogenesis. Cancer Res 59: 4574–4577, 1999[Abstract/Free Full Text]
Dormond O, Foletti A, Paroz C, Ruegg C: NSAIDs inhibit V3 integrin-mediated and Cdc42/Rac-dependent endothelial-cell spreading, migration and angiogenesis. Nat Med 7: 1041–1047, 2001[CrossRef][Medline]
Komers R, Lindsley JN, Oyama TT, Schutzer WE, Reed JF, Mader SL, Anderson S: Immunohistochemical and functional correlations of renal cyclooxygenase-2 in experimental diabetes. J Clin Invest 107: 889–898, 2001[Medline]
Wang JL, Cheng HF, Harris RC: Cyclooxygenase-2 inhibition decreases renin content and lowers blood pressure in a model of renovascular hypertension. Hypertension 34: 96–101, 1999[Abstract/Free Full Text]
Johnson RJ, Iida H, Alpers CE, Majesky MW, Schwartz SM, Pritzl P, Gordon K, Gown AM: Expression of smooth muscle cell phenotype by rat mesangial cells in immune complex nephritis: -Smooth muscle actin is a marker of mesangial cell proliferation. J Clin Invest 87: 847–858, 1991
Schneider A, Harendza S, Zahner G, Jocks T, Wenzel U, Wolf G, Thaiss F, Helmchen U, Stahl RA: Cyclooxygenase metabolites mediate glomerular monocyte chemoattractant protein-1 formation and monocyte recruitment in experimental glomerulonephritis. Kidney Int 55: 430–441, 1999[CrossRef][Medline]
Hartner A, Pahl A, Brune K, Goppelt-Struebe M: Upregulation of cyclooxygenase-1 and the PGE2 receptor EP2 in rat and human mesangioproliferative glomerulonephritis. Inflamm Res 49: 345–354, 2000[CrossRef][Medline]
Hirose S, Yamamoto T, Feng L, Yaoita E, Kawasaki K, Goto S, Fujinaka H, Wilson CB, Arakawa M, Kihara I: Expression and localization of cyclooxygenase isoforms and cytosolic phospholipase A2 in anti-Thy-1 glomerulonephritis. J Am Soc Nephrol 9: 408–416, 1998[Abstract]
Floege J, Burns MW, Alpers CE, Yoshimura A, Pritzl P, Gordon K, Seifert RA, Bowen-Pope DF, Couser WG, Johnson RJ: Glomerular cell proliferation and PDGF expression precede glomerulosclerosis in the remnant kidney model. Kidney Int 41: 297–309, 1992[Medline]
Lee LK, Meyer TW, Pollock AS, Lovett DH: Endothelial cell injury initiates glomerular sclerosis in the rat remnant kidney. J Clin Invest 96: 953–964, 1995
Received for publication July 6, 2001.
Accepted for publication December 22, 2001.
This article has been cited by other articles:
P. Michielsen, L. Heinemann, M. Mihatsch, P. Schnulle, H. Graf, and K.-M. Koch Non-phenacetin analgesics and analgesic nephropathy: Clinical assessment of high users from a case-control study [
Nephrol. Dial. Transplant.,
November 26, 2008;
(2008)
gfn643v1.
[Abstract][Full Text][PDF]
H. Li, M. Hortmann, A. Daiber, M. Oelze, M. A. Ostad, P. M. Schwarz, H. Xu, N. Xia, A. L. Kleschyov, C. Mang, et al. Cyclooxygenase 2-Selective and Nonselective Nonsteroidal Anti-Inflammatory Drugs Induce Oxidative Stress by Up-Regulating Vascular NADPH Oxidases
J. Pharmacol. Exp. Ther.,
September 1, 2008;
326(3):
745 - 753.
[Abstract][Full Text][PDF]
D. Sankaran, N. Bankovic-Calic, M. R. Ogborn, G. Crow, and H. M. Aukema Selective COX-2 inhibition markedly slows disease progression and attenuates altered prostanoid production in Han:SPRD-cy rats with inherited kidney disease
Am J Physiol Renal Physiol,
September 1, 2007;
293(3):
F821 - F830.
[Abstract][Full Text][PDF]
L. Warford-Woolgar, C. Y.-C. Peng, J. Shuhyta, A. Wakefield, D. Sankaran, M. Ogborn, and H. M. Aukema Selectivity of cyclooxygenase isoform activity and prostanoid production in normal and diseased Han:SPRD-cy rat kidneys
Am J Physiol Renal Physiol,
April 1, 2006;
290(4):
F897 - F904.
[Abstract][Full Text][PDF]
T. Ostendorf, C. van Roeyen, R. Westenfeld, A. Gawlik, M. Kitahara, E. de Heer, D. Kerjaschki, J. Floege, and M. Ketteler Inducible Nitric Oxide Synthase-Derived Nitric Oxide Promotes Glomerular Angiogenesis via Upregulation of Vascular Endothelial Growth Factor Receptors
J. Am. Soc. Nephrol.,
September 1, 2004;
15(9):
2307 - 2319.
[Abstract][Full Text][PDF]
T. D. WARNER and J. A. MITCHELL Cyclooxygenases: new forms, new inhibitors, and lessons from the clinic
FASEB J,
May 1, 2004;
18(7):
790 - 804.
[Abstract][Full Text][PDF]
C Waldner, K Schror, and P Heering COX-2 dependent PGE2 downregulates {alpha}v integrin expression via the EP3 receptor in cultured mesangial cells
J. Clin. Pathol.,
May 1, 2004;
57(5):
553 - 555.
[Abstract][Full Text][PDF]
M. Hermann, G. Camici, A. Fratton, D. Hurlimann, F. C. Tanner, J. P. Hellermann, M. Fiedler, J. Thiery, M. Neidhart, R. E. Gay, et al. Differential Effects of Selective Cyclooxygenase-2 Inhibitors on Endothelial Function in Salt-Induced Hypertension
Circulation,
November 11, 2003;
108(19):
2308 - 2311.
[Abstract][Full Text][PDF]
T. Ostendorf, C. R.C. van Roeyen, J. D. Peterson, U. Kunter, F. Eitner, A. J. Hamad, G. Chan, X.-C. Jia, J. Macaluso, G. Gazit-Bornstein, et al. A Fully Human Monoclonal Antibody (CR002) Identifies PDGF-D as a Novel Mediator of Mesangioproliferative Glomerulonephritis
J. Am. Soc. Nephrol.,
September 1, 2003;
14(9):
2237 - 2247.
[Abstract][Full Text][PDF]
V. Campean, F. Theilig, A. Paliege, M. Breyer, and S. Bachmann Key enzymes for renal prostaglandin synthesis: site-specific expression in rodent kidney (rat, mouse)
Am J Physiol Renal Physiol,
July 1, 2003;
285(1):
F19 - F32.
[Abstract][Full Text][PDF]
Top
Abstract
Introduction
-smooth muscle actin (clone 1A4; Dako, Glostrup, Denmark), a murine monoclonal IgG antibody to a cytoplasmic antigen present in monocytes, macrophages, and dendritic cells (clone ED-1; Camon, Wiesbaden, Germany), and affinity-purified, polyclonal goat anti-human/bovine type IV collagen IgG that had been preabsorbed with rat erythrocytes (Biozol, Birmingham, AL) (19,23,30). The sections were incubated sequentially with biotinylated horse anti-mouse Ig antibody (Vector Laboratories, Burlingame, CA), goat anti-rabbit Ig antibody (Vector), or rabbit anti-goat Ig antibody (Zymed, San Francisco, CA). The ABC-Elite reagent (Vector) was used, with 3,3'-diaminobenzidine (with nickel chloride enhancement) as the chromogen. Sections were counterstained with methyl green. 
3 integrin-mediated and Cdc42/Rac-dependent endothelial-cell spreading, migration and angiogenesis. Nat Med 7: 1041–1047, 2001[CrossRef][Medline]
