Plasminogen activator inhibitor-1 regulates ... - Kidney International

2 downloads 0 Views 343KB Size Report
Sep 17, 2008 - Acute pyelonephritis, frequently caused by Escherichia coli, is a substantial health problem. Plasminogen activator inhibitor type-1 (PAI-1) not ...
original article

http://www.kidney-international.org & 2009 International Society of Nephrology

Plasminogen activator inhibitor-1 regulates neutrophil influx during acute pyelonephritis Joris J.T.H. Roelofs1, Gwendoline J.D. Teske1, Peter I. Bonta2, Carlie J.M. de Vries2, Joost C.M. Meijers3, Jan J. Weening1, Tom van der Poll4 and Sandrine Florquin1 1

Department of Pathology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; 2Department of Biochemistry, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; 3Department of Vascular Medicine, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands and 4Center for Experimental and Molecular Medicine, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands

Acute pyelonephritis, frequently caused by Escherichia coli, is a substantial health problem. Plasminogen activator inhibitor type-1 (PAI-1) not only inhibits plasminogen activation but is also involved in cell migration. To determine if it has a role in host defense, we induced pyelonephritis in PAI-1 gene knockout and wild-type mice by intravesical inoculation with uropathogenic E. coli 1677. Bacterial growth was determined on blood agar plates in portions of the kidneys homogenized in sterile saline. Kidney levels of PAI-1 were increased in infected compared to control mice, suggesting a physiological role for PAI-1 during pyelonephritis. The knockout mice had significantly more bacterial outgrowth in kidney homogenates compared to the wild-type mice. Strikingly, higher colony-forming units were accompanied by increased levels of the cytokines TNF-a, IL-1b, and IL-6 in the kidneys of knockout mice, but levels of the chemokines KC and MIP-2 were not different. Remarkably, plasma levels of KC were higher, but renal neutrophil influx was significantly lower, in the knockout than in the wild-type mice. Our study shows that PAI-1 is critically involved in host defense against E. coli-induced acute pyelonephritis, in part, by modulating neutrophil influx. Kidney International (2009) 75, 52–59; doi:10.1038/ki.2008.454; published online 17 September 2008 KEYWORDS: fibrinolytic system; immunology and pathology; pyelonephritis; plasminogen activator inhibitor type 1

Correspondence: Joris J.T.H. Roelofs, Department of Pathology, Room H2131, Academic Medical Center, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. E-mail: [email protected] Received 27 August 2007; revised 14 July 2008; accepted 15 July 2008; published online 17 September 2008 52

Urinary tract infections (UTI) belong to the most frequently occurring bacterial infections, affecting approximately 50% of women at one point in their lifetime.1 In up to 95% of cases, uropathogenic Escherichia coli is identified as the causative organism.2 Although UTI usually display an uncomplicated course, the upper urinary tract may become involved by ascending infection, giving rise to the clinical syndrome of acute pyelonephritis. When left untreated, pyelonephritis may lead to renal fibrosis and subsequent endstage renal failure.3 Adhesion of uropathogenic E. coli to the urothelium will trigger activation of several host defense mechanisms. After inoculation with uropathogenic E. coli, bladder and tubular epithelial cells have been shown to produce proinflammatory cytokines and chemokines, among which interleukin-1b (IL1b), interleukin-6 (IL-6), and interleukin-8 (IL-8).4–6 Next, the urothelium is invaded by neutrophils that phagocytose and subsequently kill E. coli organisms.7 The plasminogen activator inhibitor type-1 (PAI-1) belongs to the superfamily of serine protease inhibitors (serpins) and is the principal physiological inhibitor of tissue-type plasminogen activator and urokinase type plasminogen activator, thereby controlling plasmin generation and regulating fibrinolysis.8 Besides its role as a regulator of hemostasis by regulating fibrinolytic activity, PAI-1 plays a role in many other (patho)physiological processes, including wound healing, atherosclerosis, tumor angiogenesis, pulmonary fibrosis, and rheumatoid arthritis.9–13 In addition, PAI-1 acts as an acute phase protein during sepsis, with enhanced plasma levels of PAI-1, accounting for the so-called fibrinolytic shut down during severe sepsis.14,15 More recently, PAI-1 has been shown to be critically involved in the regulation of cell migration. PAI-1 can inhibit cell bound urokinase type plasminogen activator, resulting in reduced pericellular proteolysis and a subsequent decrease in cell migration. Furthermore, PAI-1 inhibits integrin- and vitronectin-mediated cell migration independently of its function as an inhibitor of plasminogen activation, by competing for vitronectin binding to integrins.16 On the contrary, the de-adhesive action of PAI-1 by inactivation of Kidney International (2009) 75, 52–59

original article

JJTH Roelofs et al.: PAI-1 is protective during acute pyelonephritis

a

PAI-1 protein (ng/g protein)

the cell-integrin-extra cellular matrix interaction, may result in an increase of cell mobility as well.17 In addition, it has been shown that PAI-1 in macrophages functions as a master switch between cell adhesion and detachment by inducing internalization of the aMb2 integrin/low density lipoprotein receptor like protein (Mac-1/LRP) complex, thus promoting macrophage migration.18 As a result, PAI-1 seems to exert both promoting and inhibitory effects on cell migration. With respect to neutrophil migration in reaction to infectious agents, different roles for PAI-1 have been described. PAI-1 genedeficient (PAI-1/) mice have an unaltered immune response to Streptococcus pneumoniae-induced pneumonia, with similar pulmonary leukocyte recruitment in PAI-1/ and wild-type (WT) mice.19 Recently however, our laboratory showed that PAI-1/ mice have an impaired host response during Klebsiella pneumoniae-induced pneumonia and sepsis, associated with decreased pulmonary neutrophil influx.20 Although upregulation of systemic PAI-1 levels during E. coli-induced sepsis has been reported extensively,21–24 the role of PAI-1 in host defense against E. coli infections remains unclear. To investigate the in vivo role of PAI-1 during UTI, we induced acute pyelonephritis in PAI-1-deficient mice by intravesical inoculation with uropathogenic E. coli.

25

*

20

15

10

5

0 Sham

24 h

b

c

d

e

4h

RESULTS PAI-1 expression is upregulated during acute pyelonephritis

To investigate local PAI-1 production on infection, renal PAI-1 levels were measured by enzyme-linked immunosorbent assay (ELISA) after induction of acute pyelonephritis in WT mice. Renal PAI-1 levels showed a twofold increase at 24 h after infection (Figure 1a). At 48 h, PAI-1 protein levels approached baseline levels again (Figure 1a). To reveal the cellular source of PAI-1, we performed in situ hybridization on infected and sham kidneys. In sham kidneys, no positive signal was detected (Figure 1b). The infected kidneys showed clear positive signal for PAI-1 mRNA in endothelial cells of arterioles (Figure 1c), veins (Figure 1d), and mononuclear cells within the inflammatory infiltrates (Figure 1e). Fibrinolytic parameters

As PAI-1 is one of the main inhibitors of fibrinolysis and infections generally induce a procoagulant state, we analyzed the amount of fibrin deposition by immunostainings and levels of D-dimer in kidney homogenates by ELISA. No fibrin deposits could be detected in WT and PAI-1/ kidneys at both time points after inoculation with E. coli (not shown). Furthermore, D-dimer levels were undetectable in kidney homogenates of WT and PAI-1/ mice after infection (not shown).

Figure 1 | Renal PAI-1 is upregulated during E. coli-induced acute pyelonephritis. (a) PAI-1 protein levels in kidney homogenates were measured by ELISA at 24 and 48 h after inoculation. Data are mean±s.e.m., *Po0.05 vs sham mice. In situ hybridization for murine PAI-1 mRNA was performed on paraffin slides of sham (b) and infected kidneys (c–e). Arrows indicate the urothelial cell layer of the renal pelvis. Twenty-four hours after infection, positive signal in black. Original magnifications:  40.

(CFU) in kidney homogenates than WT mice (Figure 2), indicating that PAI-1/ mice had a seriously impaired ability to clear E. coli-caused pyelonephritis. In addition, PAI-1/ mice experienced more severe systemic disease than WT mice. Blood cultures remained negative in WT mice at both time points. In contrast, at 24 h after infection, three out of nine PAI-1/ mice had positive blood cultures. At 48 h, two out of seven PAI-1/ mice had died (vs zero in WT mice) and one of the five remaining PAI-1/ mice was bacteremic.

Higher bacterial load in PAI-1/ mice

Lower neutrophil influx in bladders and kidneys from PAI-1/ mice

Renal bacterial outgrowth was determined at 24 and 48 h after infection. At both time points, PAI-1/ mice showed a significantly higher number of E. coli colony-forming units

As neutrophil recruitment plays a pivotal role in host defense against bacterial infections, histological sections of bladder and renal tissue were examined to investigate the extent of

Kidney International (2009) 75, 52–59

53

original article

*

JJTH Roelofs et al.: PAI-1 is protective during acute pyelonephritis

**

7

10

106

CFU/ml

105 104 103 102 101 100 24 h

48 h

Figure 2 | Higher bacterial outgrowth in PAI-1/ mice. At T ¼ 24 h and T ¼ 48 h after inoculation, PAI-1/ kidneys (K) contain higher E. coli CFU than WT kidneys (J). Horizontal lines represent medians, *Po0.05; **Po0.01.

inflammation. WT kidneys showed an extensive pyelonephritis, whereas PAI-1/ kidneys showed virtually no inflammation (Figure 3a). In both genotypes, neutrophils were located in the lamina propria and urothelial cell layer, with apparently fewer neutrophils in PAI-1/ kidneys (Figure 3a). To quantify neutrophil influx, myeloperoxidase (MPO) measurements on kidney homogenates were performed by ELISA. In accordance with the histological findings, neutrophil influx was significantly lower in PAI-1/ mice than WT mice at 24 h (Figure 3b). At 48 h, PAI-1/ kidneys tended to show lower MPO levels than WT kidneys, although this trend did not reach statistical significance (P ¼ 0.06). In accordance with these findings, at 24 h bladders of PAI-1/ mice showed lower numbers of infiltrating neutrophils than WT bladders (24 h: 76.3±13 (WT) vs 38.9±11 (PAI-1/) mean number of neutrophils/HPF (Po0.05); 48 h: 23.1±14 (WT) vs 18.8±11 (PAI-1/) mean number of neutrophils/HPF (P ¼ NS)). To monitor the systemic inflammatory response and to rule out a priori lower numbers of circulating neutrophils in PAI-1/ mice, leukocyte numbers in the circulation were counted. As shown in Figure 3c, sham WT and sham PAI-1/ mice had similar white blood cell counts. On infection, the number of peripheral leukocytes was upregulated to the same extent in both genotypes (Figure 3c). As intercellular adhesion molecule-1 (ICAM-1) and Eselectin belong to the most important adhesion molecules that enable neutrophil influx during pyelonephritis, and expressions of ICAM-1 and E-selectin can be modified by the fibrinolytic system,25 we analyzed immunostainings for ICAM-1 and E-selectin. Quantitative assessment of ICAM-1 and E-selectin stainings showed similar expression in WT and PAI-1/ mice (not shown). 54

Higher cytokine levels in PAI-1/ mice

As proinflammatory cytokines and chemokines play a crucial role in the generation of an inflammatory response during acute pyelonephritis,6,26 levels of IL-1b, IL-6, keratinocytederived chemokine (KC), macrophage inflammatory protein2 (MIP-2), and tumor-necrosis factor-a (TNF-a) were measured in kidneys at mRNA and protein level. Both WT and PAI-1/ kidneys showed upregulation of IL-1b, IL-6, MIP-2, and TNF-a mRNA at 24 h. As shown in Table 1, PAI-1/ kidneys contained higher amounts of MIP-2 and TNF-a mRNA than WT kidneys. Protein levels of these cytokines were determined by ELISA. Measurements are shown in Table 2. At 24 h, PAI-1/ kidneys contained higher levels of IL-1b than WT kidneys. At 48 h, renal levels of IL-1b and IL-6 protein were significantly higher in PAI-1/ mice than in WT mice (Table 2). At both time points, PAI-1/ kidneys contained higher levels of TNF-a protein than WT kidneys. The same cytokines were measured in plasma. Plasma levels of IL-1b, IL-6, MIP-2, and TNF-a were too low to be detected by ELISA (not shown). However, as shown in Figure 4, plasma KC levels were significantly higher in PAI-1/ mice in comparison with WT mice at both time points. DISCUSSION

This study shows that endogenous PAI-1 is involved in host defense during E. coli-induced acute pyelonephritis. PAI-1/ mice had a seriously impaired bacterial clearance when compared to WT mice, at least, in part, caused by lower influx of neutrophils into the infected renal tissue. Elevated plasma levels of PAI-1 have been reported in various clinical conditions, including pancreatitis, liver disease, and various malignancies.27,28 In addition, PAI-1 has proven to act as an acute phase protein, with marked elevation of plasma PAI-1 levels on surgery, myocardial infarction, and during sepsis.14,15 Within the kidney, PAI-1 is normally produced in barely measurable amounts.24 In this study, WT mice showed increased renal PAI-1 mRNA and protein levels on inoculation with E. coli, indicating a physiological role for PAI-1 in host defense against acute pyelonephritis. Several kidney diseases are associated with an upregulation of local PAI-1 production.29 Most of these concern conditions that are associated with matrix accumulation, fibrin deposits, or fibrosis, including diabetic nephropathy,30,31 membranous nephropathy,32 thrombotic microangiopathy,33,34 and chronic allograft nephropathy.35,36 Studies with genetically engineered mice have demonstrated that PAI-1 exerts profibrotic effects, as a result of its function as a proteolysis inhibitor. Indeed, in experimental diabetic nephropathy and obstructive nephropathy, PAI-1/ mice display significantly reduced fibrosis in comparison with WT mice.37,38 PAI-1/ mice display a mild hyperfibrinolytic state and have a quicker whole-blood clot lysis than WT mice.39 Fibrin deposition has been implicated in the pathogenesis of inflammatory glomerular diseases.40 In crescentic antiKidney International (2009) 75, 52–59

original article

JJTH Roelofs et al.: PAI-1 is protective during acute pyelonephritis

a

WT

PAI-1 –/–

WT

PAI-1 –/–

c 15,000

WT PAI-1 –/–

WBC (x106 cells/ml)

MPO (pg/mg protein)

b

10,000

* 5000 0 Sham

24 h

48 h

8 7 6 5 4 3 2 1 0

WT PAI-1 –/–

Sham

24 h

48 h

Figure 3 | PAI-1/ mice show lower neutrophil influx during acute pyelonephritis. (a) Photomicrographs of WT and PAI-1/ renal tissue, 24 h after infection. The upper panels show an overview of the pelvis region. The lower panels show a detail of the urothelial cell layer of the renal pelvis. H&E staining, original magnifications:  4 (upper) and  40 (lower). (b) Myeloperoxidase (MPO) levels were measured in kidney homogenates by ELISA. Data are mean±s.e.m., *Po0.05. (c) Peripheral blood leukocytes (WBC) were counted using a hemocytometer. Data are mean±s.e.m., P ¼ NS.

Table 1 | Gene expression of pro-inflammatory cytokines by QPCR 24 h Gene expression/TBP IL-1b IL-6 KC MIP-2 TNF-a

WT

PAI-1 / 

0.79±0.58 1.50±1.10 0.10±0.07 1.0±1.0 1.97±0.10

2.4±1.3a 1.68±0.16 0.09±0.07 5.9±5.2* 14.95±1.14*

KC, keratinocyte-derived chemokine; IL, interleukin; MIP-2, macrophage inflammatory protein-2; PAI-1, plasminogen activator inhibitor type-1; TBP, TATA-binding protein; TNF-a, tumor-necrosis factor-a; WT, wild type. Data are expressed as amount of mRNA relative to housekeeping gene TBP; mean±s.e.m., *Po0.05. a P=0.05.

glomerular basement membrane glomerulonephritis, tissuetype plasminogen activator proved to be protective against glomerular fibrin deposition and inflammation.41 In the same disease model, PAI-1/ mice developed fewer crescents, less glomerular fibrin deposition, and fewer Kidney International (2009) 75, 52–59

infiltrating leukocytes than WT mice.42 In accordance, mice that overexpress PAI-1 show increased glomerular crescent formation, more glomerular fibrin deposition, and more glomerular inflammation than WT mice.42 Although fibrin deposits have been described in severe acute pyelonephritis at electron microscopic level,43 fibrin is not a regular feature of this condition and probably not of pathogenetic significance. In our model, no appreciable fibrin deposits were encountered and D-dimer levels were too low to be determined effectively. This is in line with our previous study concerning the role of tissue-type plasminogen activator in acute pyelonephritis, where no fibrin deposits could be detected using a highly specific western blotting technique (JJ Roelofs, submitted for publication). Altogether, it seems unlikely that differences in fibrin accumulation accounts for the more severe UTI in PAI-1/ mice compared to WT mice. The recruitment of neutrophils is a crucial part of host defense against acute pyelonephritis.44 Our study shows that PAI-1/ mice have a decreased ability to clear infection with E. coli, at least partly due to diminished neutrophil influx. Indeed, despite higher cytokine levels in PAI-1/ 55

original article

JJTH Roelofs et al.: PAI-1 is protective during acute pyelonephritis

Table 2 | Proinflammatory cytokines and chemokines in kidney homogenates 24 h pg/mg Protein IL-1b IL-6 KC MIP-2 TNF-a

48 h

WT

PAI-1 / 

WT

PAI-1 / 

28.4±5.4 6.7±1.2 234.3±29.4 41.0±4.7 8.8±3.4

45.1±4.3* 7.3±1.4 249.7±37.1 40.8±7.8 16.5±2.9*

25.6±5.5 5.2±0.8 217.8±22.1 47.0±9.7 9.4±2.2

51.8±11.2* 14.1±4.2* 213.3±14.8 45.5±5.3 21.1±4.2*

IL, interleukin; KC, keratinocyte-derived chemokine; MIP-2, macrophage inflammatory protein-2; PAI-1, plasminogen activator inhibitor type-1; TNF-a, tumor-necrosis factor-a; WT, wild type. Data are expressed as pg/mg protein; mean±s.e.m., *Po0.05.

KC (pg/ml plasma)

25,000

WT PAI-1–/–

*

20,000 *

15,000 10,000 5000 0 Sham

24 h

48 h

Figure 4 | Higher plasma KC levels in PAI-1/ mice. Plasma KC levels were determined at 24 and 48 h after inoculation by ELISA. Data are mean±s.e.m., *Po0.05.

kidneys compared to WT kidneys, less inflammation was observed. The higher amount of CFU at 24 and 48 h after inoculation were accompanied by higher levels of TNF-a, IL1b, and (at 48 h) IL-6 in the PAI-1/ kidneys. This is in line with one of our previous studies, that showed that renal cytokine levels are directly proportional to the amount of CFU in this model.7 Indeed, After inoculation with uropathogenic E. coli, bladder and tubular epithelial cells have been shown to produce IL-1b, IL-6, IL-8, and TNF-a.4–6 Recent studies have shown that tubular epithelial cells produce CXCL2 cytokines on inoculation with uropathogenic E. coli in a Toll-like receptor-4 (TLR-4)-mediated fashion.45 In an elegant study involving a model of ascending UTI in mice, Chassin et al.46 described that E. coli-infected TLR-4-deficient mice fail to clear renal bacteria and exhibit dramatically lower production of proinflammatory mediators in comparison to WT mice. In the same study it was shown that uropathogenic E. coli stimulate primary cultures of collecting duct intercalated cells both in a TLR4-mediated, MyD88-dependent pathway and in a TLR4-independent pathway, leading to bipolarized secretion of MIP-2 by renal epithelial cells.46 Only one study describes an interaction between the fibrinolytic system and TLR activity.47 This study describes that plasmin potentiates TLR2 and TLR4 signaling in a 56

macrophage cell line in vitro. Plasmin enhances endogenous production of TNF-a and activation of an nuclear factor-kB reporter plasmid, in a proteolysis-dependent fashion. As we measured higher TNF-a levels in our PAI-1/ mice, this mechanism might be at play in our study as well. We measured TLR-4 expression at 24 h. at the mRNA level by quantitative PCR (QPCR). There were no significant differences in TLR-4 expression levels between WT and PAI-1/ mice (data not shown). Our current observation that PAI-1 can exert proinflammatory protective effects is in accordance with a recent study by Kwak et al.,48 that demonstrated that PAI-1 potentiates lipopolysaccharide-induced neutrophil activation in vitro through a c-Jun N-terminal kinase-mediated pathway. Furthermore, these results are largely in accordance with a previous report, which demonstrated that PAI-1 plays a protective role during K. pneumoniae-induced pneumonia.20 In this study, PAI-1/ mice had higher bacterial outgrowth than WT mice, associated with a reduction in pulmonary neutrophil influx.20 Conversely, transgenic overexpression of PAI-1 in the lung by means of a replication-defective adenoviral vector resulted in reduction of bacterial outgrowth. Furthermore, PAI-1 overexpression in healthy (that is, noninfected) lungs provoked neutrophil influx.20 A growing number of studies confirm a regulatory role for PAI-1 in mobilization of various cell types. Apart from regulating cell movement through plasmin-dependent extra cellular matrix degradation,49,50 PAI-1 orchestrates cell migration through modulation of urokinase plasminogon activator receptor (uPAR) and integrin-mediated cell adhesion.17,51 In general, PAI-1 exerts ‘de-adhesive’ actions, resulting in either increased or decreased cell migration.16,17 As cell migration is the result of a dynamical balance between adhesion, cytoskeleton modulation, and detachment, the net effect of PAI-1 on cell movement depends strongly on the cellular milieu and the applied experimental set up. Another mechanism via which PAI-1 can influence inflammatory cell migration, is the modification of chemotactic gradients. Endothelial heparan sulfate (HS)-syndecan-1 ectodomains have the ability to bind and stabilize chemokines, such as IL-8, thereby promoting neutrophil extravasation. IL-8–HS-syndecan-1 complexes are constitutively shed by plasmin, which provides a negative regulation of IL-8mediated neutrophil migration. Marshall et al.52 have demonstrated that PAI-1 negatively regulates the constitutive shedding of endothelial IL-8–HS-syndecan-1 complexes. Inhibition of PAI-1 activity through incubation with neutralizing antibody MAI-12 induced the shedding of IL8–HS-syndecan-1 from endothelial cells in a concentration dependent manner, resulting in diminished IL-8-mediated transendothelial neutrophil migration.52 In addition, elevated levels of soluble IL-8 have been shown to downregulate functional IL-8 receptor on neutrophils53 and impede neutrophil recruitment to inflammatory foci in vivo.54 Furthermore, transgenic mice that permanently overexpress IL-8 exhibit a marked intravascular neutrophilia and Kidney International (2009) 75, 52–59

JJTH Roelofs et al.: PAI-1 is protective during acute pyelonephritis

decreased neutrophil extravasation in response to inflammatory stimulants.55 In a study on the role of PAI-1 in Gramnegative pneumonia, it was demonstrated that PAI-1/ mice have elevated plasma levels of KC and impaired pulmonary neutrophil influx.20 In this study, PAI-1/ mice displayed significantly elevated plasma levels of the murine IL-8 homologue KC during pyelonephritis, whereas KC levels in renal tissue were similar in WT and PAI-1/ mice. Most probably, PAI-1 deficiency caused increased shedding of KC via the above-mentioned mechanism, resulting in neutrophil ‘entrapment’ in the vascular compartment. In particular with respect to acute pyelonephritis, IL-8 has been shown to be a key chemokine to promote neutrophil recruitment to the urinary compartment.56 It has been shown that IL-8 receptor is of critical importance for neutrophils to cross the urothelial basement membrane and launch an adequate host response on infection with uropathogenic E. coli.57,58 Furthermore, very recently IL-8 gene polymorphisms have been implicated in susceptibility to acute pyelonephritis in a pediatric population.59 Taken together, this study provides new evidence for a protective role for PAI-1 during E. coli-induced acute pyelonephritis through the promotion of neutrophil migration. MATERIALS AND METHODS Mice and experimental protocol Female PAI-1/ mice on a C57BL/6J background60 were purchased from The Jackson Laboratory (Bar Harbor, ME, USA) and bred in the institutional animal facilities. Mice were backcrossed more than 12 times. Age- and weight-matched C57Bl/6J mice (Charles River, Maastricht, The Netherlands) served as a WT control. UTI was induced as described previously.7,61 E. coli 1677, isolated from a uroseptic patient, was donated by Dr. W.J. Hopkins (University of Wisconsin Medical School, Madison). This strain has virulence factors that include type 1 and P fimbriae, hemolysin, and aerobactin.62 This specific serotype has proven uropathogenicity in mice.61,62 Bacteria were cultured for 16 h at 37 1C in trypticase soy broth. After dilution 1:100 in fresh trypticase soy broth, the suspension was grown for 3 h to midlogarithmic phase. E. coli were washed three times in sterile phosphate-buffered saline. After resuspension in phosphate-buffered saline, the final concentration was determined by plating 10-fold serial dilutions of the suspension on blood agar plates. Acute pyelonephritis was induced under general anesthesia (0.07 ml/10 g mouse of fentanyl–fluanisone–midazolam mixture, containing 1.25 mg/ml midazolam (Roche, Mijdrecht, The Netherlands), 0.08 mg/ml fentanyl citrate and 2.5 mg/ml fluanisone (Janssen Pharmaceutica, Beerse, Belgium)) in 8- to 10-week old female mice. Bacterial suspension (100 ml) was administered transurethrally through a 0.55 mm catheter (Abbott, Zwolle, The Netherlands). Sham control mice (n ¼ 5 per group) underwent the same procedure with administration of 100 ml of sterile phosphate-buffered saline instead of bacterial suspension. Mice killed at 24 h after inoculation (n ¼ 9 per group) received 2.7  108 CFU per mouse. In a separate experiment, mice killed at 48 h after inoculation (n ¼ 7 per group) received 1.0  109 CFU per mouse. The choice of these time points was based on the literature Kidney International (2009) 75, 52–59

original article

and on our own broad experience with this model.7,61–63 White blood cell counts in peripheral blood were determined using a hemocytometer (Beckman Coulter, Fullerton, CA, USA). The Committee on Use and Care of Animals of the Academic Medical Center at the University of Amsterdam approved all experiments. Animal experimentation guidelines were followed in all experiments. Determination of bacterial outgrowth Bacterial load was determined in renal tissue and blood as described earlier.7,61 One half of the left kidney from each mouse was homogenized in 4 volumes of sterile saline with a tissue homogenizer which was cleaned with 70% ethanol after each homogenization. Serial 10-fold dilutions were made in sterile saline, and 50 ml volumes of kidney homogenate and blood were plated onto blood agar plates, which were incubated at 37 1C for 16 h, after which E. coli CFU were counted. Histological examination The bladder and the remaining part of the left kidney were fixed in 10% formalin and embedded in paraffin. Sections were stained with hematoxylin and eosin in a standard fashion. Neutrophils in the bladder were counted in four randomly chosen fields (  400 magnification). Immunostaining for fibrin, ICAM-1, and E-selectin were performed as described previously,64 using anti-mouse fibrin(ogen) (Ixell; Accurate Chemical & Scientific, Westbury, NY, USA), antimouse ICAM-1 (R&D Systems, Minneapolis, MN, USA), and antimouse E-selectin (R&D Systems) as primary antibodies. The amount of ICAM-1 and E-selectin staining was measured as described previously64 in eight randomly chosen fields (  200 magnification) using digital image analysis (Image pro-plus; Mediacybernetics, Dortmund, Germany). Murine PAI-1 mRNA in situ hybridization was performed using radiolabeled [35S]-UTP (Amersham, Arlington Heights, IL, USA) mouse (NM_008871, bp 680–823) PAI-1-specific riboprobes, as described previously.20,65 In short, 7-mm-thick paraffin sections of renal tissue were pretreated with proteinase K (20 mg/ml) for 5 min, refixed in 4% paraformaldehyde, and treated for 10 min with 0.25% acetic anhydride in 0.1 mol/l triethanolamine. Hybridizations were performed overnight at 50 1C (0.5 mCi probe) per section under a cover slip in a moist chamber. After hybridization, slides were washed in 50% formamide, 2  saline-sodium citrate, and 10 mmol/l dithiothreitol. After RNase A digestion, washing and dehydration, autoradiography emulsion was applied (Ilford G5 emulsion 1:1 diluted with 2% glycerol). Slides were developed in Kodak D19 after an exposure of 3–6 weeks, fixed in Kodak UNIFIX, and counterstained with hematoxylin. Real-time quantitative PCR analysis Real-time QPCR analysis was performed using RNA obtained from kidneys from WT and PAI-1/ mice, at T ¼ 24 h. Total RNA was isolated using TRIzol LS Reagent (Invitrogen Life Technologies, Breda, The Netherlands), following the manufacturer’s instructions. The concentration and purity of RNA were determined spectrophotometrically at 260/280 nm. Five micrograms of total RNA was reverse-transcribed into cDNA using oligo dT primers (Roche). Primer sets used were IL-1b: forward CTGCAGCTGGAGAGTGTG GAT, reverse GCTTGTGCTCTGCTTGTGAG; IL-6: forward CAAA GCCAGAGTCCTTCAGAG, reverse TGGTCCTTAGCCACTCCTTC; KC: forward ATAATGGGCTTTTACATTCTTTAACC, reverse 57

original article

AGTCCTTTGAACGTCTCTGTCC; MIP-2: forward CCCTGGTTCA GAAAATCATCC, reverse CTTCCGTTGAGGGACAGC; TNF-a: forward TCGTAGCAAACCACCAAGTG, reverse CCTTGTCCCTTG AAGAGAACC; TLR-4: forward GGACTCTGATCATGGCACTG, reverse CTGATCCATGCATTGGTAGGT. TATA-binding protein was used as housekeeping gene: forward GGAGAATCATGGACCA GAACA, reverse GATGGGAATTCCAGGAGTCA. QPCR was performed with a LightCycler LC480 (Roche). For the PCR, a master mixture was prepared on ice, containing per sample: 1 ml of cDNA, 1 ml of FastStart Reaction Mix SYBR Green I (Roche Applied Science, Indianapolis, IN, USA), 0.5 ml of 10 mM primers, and 1.6 ml of 25 mM MgCl2. After the reaction mixture was loaded into a glass capillary tube, the cycling conditions were carried out as follows: initial denaturation at 95 1C for 6 min, followed by 45 cycles of denaturation at 95 1C for 15 s, annealing at 55–60 1C for 5–10 s, and extension at 72 1C for 8–12 s. The temperature transition rate was set at 20 1C/s. Fluorescent product was measured by a single acquisition mode at 72 1C after each cycle. Quantification of data was performed using the standard Light Cycler analysis software. The amount of amplified cytokine products was divided by the amount of TATAbinding protein for each sample. PAI-1, cytokine, MPO, and D-dimer measurements The right kidney of each mouse was homogenized in phosphatebuffered saline containing 1% Triton X-100, 1 mM EDTA (Merck, Schiphol, The Netherlands), and 1% protease inhibitor cocktail (P8340, Sigma, Zwijndrecht, The Netherlands). Levels of PAI-1, IL1b, IL-6, KC, MIP-2, TNF-a, mouse MPO, and D-dimer in kidney homogenates and plasma were measured by specific ELISA according to the instructions of the manufacturer (PAI-1: Molecular Innovations Inc., Southfield, MI, USA; IL-1b, IL-6, KC, MIP-2, and TNF-a: R&D systems; MPO: HyCult Biotechnologies, Uden, The Netherlands; D-dimer: Diagnostica Stago, Asnie`res sur Seine, France). Total protein content of the samples was determined according to the method of Bradford,66 using a commercially available protein assay kit (Bio-Rad, Hercules, CA, USA).

JJTH Roelofs et al.: PAI-1 is protective during acute pyelonephritis

7.

8. 9.

10.

11.

12.

13.

14.

15. 16.

17.

18.

19.

20.

21.

Statistical analysis All data are presented as mean±standard error of the mean, unless stated otherwise. Data were analyzed by Mann–Whitney U-test or unpaired Student’s t-test when appropriate. Po0.05 was considered to represent a statistically significant difference.

22.

23. 24.

DISCLOSURE

The authors declared no conflicting interests. REFERENCES 1. Barnett BJ, Stephens DS. Urinary tract infection: an overview. Am J Med Sci 1997; 314: 245–249. 2. Mulvey MA, Schilling JD, Martinez JJ et al. Bad bugs and beleaguered bladders: interplay between uropathogenic Escherichia coli and innate

25.

26.

host defenses. Proc Natl Acad Sci USA 2000; 97: 8829–8835. 3. Cotton SA, Gbadegesin RA, Williams S et al. Role of TGF-beta1 in renal parenchymal scarring following childhood urinary tract infection. Kidney

27.

Int 2002; 61: 61–67. 4. Hedges S, Agace W, Svensson M et al. Uroepithelial cells are part of a mucosal cytokine network. Infect Immun 1994; 62: 2315–2321. 5. Hedges S, Svensson M, Svanborg C. Interleukin-6 response of epithelial cell lines to bacterial stimulation in vitro. Infect Immun 1992; 60:

28.

1295–1301. 6. Khalil A, Brauner A, Bakhiet M et al. Cytokine gene expression during experimental Escherichia coli pyelonephritis in mice. J Urol 1997; 158: 1576–1580.

58

29.

30.

Roelofs JJ, Rouschop KM, Teske GJ et al. The urokinase plasminogen activator receptor is crucially involved in host defense during acute pyelonephritis. Kidney Int 2006; 70: 1942–1947. Collen D. The plasminogen (fibrinolytic) system. Thromb Haemost 1999; 82: 259–270. Bajou K, Noel A, Gerard RD et al. Absence of host plasminogen activator inhibitor 1 prevents cancer invasion and vascularization. Nat Med 1998; 4: 923–928. Carmeliet P, Moons L, Lijnen R et al. Inhibitory role of plasminogen activator inhibitor-1 in arterial wound healing and neointima formation: a gene targeting and gene transfer study in mice. Circulation 1997; 96: 3180–3191. Eitzman DT, McCoy RD, Zheng X et al. Bleomycin-induced pulmonary fibrosis in transgenic mice that either lack or overexpress the murine plasminogen activator inhibitor-1 gene. J Clin Invest 1996; 97: 232–237. Luttun A, Lupu F, Storkebaum E et al. Lack of plasminogen activator inhibitor-1 promotes growth and abnormal matrix remodeling of advanced atherosclerotic plaques in apolipoprotein E-deficient mice. Arterioscler Thromb Vasc Biol 2002; 22: 499–505. Van Ness K, Chobaz-Peclat V, Castellucci M et al. Plasminogen activator inhibitor type-1 deficiency attenuates murine antigen-induced arthritis. Rheumatology (Oxford) 2002; 41: 136–141. Kluft C, Verheijen JH, Jie AF et al. The postoperative fibrinolytic shutdown: a rapidly reverting acute phase pattern for the fast-acting inhibitor of tissue-type plasminogen activator after trauma. Scand J Clin Lab Invest 1985; 45: 605–610. Levi M, de Jonge E, van der Poll T. Sepsis and disseminated intravascular coagulation. J Thromb Thrombolysis 2003; 16: 43–47. Kjoller L, Kanse SM, Kirkegaard T et al. Plasminogen activator inhibitor-1 represses integrin- and vitronectin-mediated cell migration independently of its function as an inhibitor of plasminogen activation. Exp Cell Res 1997; 232: 420–429. Czekay RP, Aertgeerts K, Curriden SA et al. Plasminogen activator inhibitor-1 detaches cells from extracellular matrices by inactivating integrins. J Cell Biol 2003; 160: 781–791. Cao C, Lawrence DA, Li Y et al. Endocytic receptor LRP together with tPA and PAI-1 coordinates Mac-1-dependent macrophage migration. EMBO J 2006; 25: 1860–1870. Rijneveld AW, Florquin S, Bresser P et al. Plasminogen activator inhibitor type-1 deficiency does not influence the outcome of murine pneumococcal pneumonia. Blood 2003; 102: 934–939. Renckens R, Roelofs JJ, Bonta PI et al. Plasminogen activator inhibitor type 1 is protective during severe Gram-negative pneumonia. Blood 2007; 109: 1593–1601. de Boer JP, Creasy AA, Chang A et al. Activation patterns of coagulation and fibrinolysis in baboons following infusion with lethal or sublethal dose of Escherichia coli. Circ Shock 1993; 39: 59–67. Paramo JA, Perez JL, Serrano M et al. Types 1 and 2 plasminogen activator inhibitor and tumor necrosis factor alpha in patients with sepsis. Thromb Haemost 1990; 64: 3–6. Zeerleder S, Schroeder V, Hack CE et al. TAFI and PAI-1 levels in human sepsis. Thromb Res 2006; 118: 205–212. Sawdey MS, Loskutoff DJ. Regulation of murine type 1 plasminogen activator inhibitor gene expression in vivo. Tissue specificity and induction by lipopolysaccharide, tumor necrosis factor-alpha, and transforming growth factor-beta. J Clin Invest 1991; 88: 1346–1353. Zhang RL, Zhang ZG, Chopp M et al. Thrombolysis with tissue plasminogen activator alters adhesion molecule expression in the ischemic rat brain. Stroke 1999; 30: 624–629. Webb NJ, Brenchley PE. Cytokines and cell adhesion molecules in the inflammatory response during acute pyelonephritis. Nephron Exp Nephrol 2004; 96: e1–e6. Juhan-Vague I, Moerman B, De Cock F et al. Plasma levels of a specific inhibitor of tissue-type plasminogen activator (and urokinase) in normal and pathological conditions. Thromb Res 1984; 33: 523–530. Harbeck N, Kates RE, Gauger K et al. Urokinase-type plasminogen activator (uPA) and its inhibitor PAI-I: novel tumor-derived factors with a high prognostic and predictive impact in breast cancer. Thromb Haemost 2004; 91: 450–456. Eddy AA, Fogo AB. Plasminogen activator inhibitor-1 in chronic kidney disease: evidence and mechanisms of action. J Am Soc Nephrol 2006; 17: 2999–3012. Paueksakon P, Revelo MP, Ma LJ et al. Microangiopathic injury and augmented PAI-1 in human diabetic nephropathy. Kidney Int 2002; 61: 2142–2148. Kidney International (2009) 75, 52–59

original article

JJTH Roelofs et al.: PAI-1 is protective during acute pyelonephritis

31.

32.

33.

34.

35.

36.

37.

38. 39.

40. 41.

42.

43.

44. 45.

46.

47. 48.

49.

Nicholas SB, Aguiniga E, Ren Y et al. Plasminogen activator inhibitor-1 deficiency retards diabetic nephropathy. Kidney Int 2005; 67: 1297–1307. Hamano K, Iwano M, Akai Y et al. Expression of glomerular plasminogen activator inhibitor type 1 in glomerulonephritis. Am J Kidney Dis 2002; 39: 695–705. Xu Y, Hagege J, Mougenot B et al. Different expression of the plasminogen activation system in renal thrombotic microangiopathy and the normal human kidney. Kidney Int 1996; 50: 2011–2019. Rondeau E, Mougenot B, Lacave R et al. Plasminogen activator inhibitor 1 in renal fibrin deposits of human nephropathies. Clin Nephrol 1990; 33: 55–60. Shihab FS, Yamamoto T, Nast CC et al. Transforming growth factor-beta and matrix protein expression in acute and chronic rejection of human renal allografts. J Am Soc Nephrol 1995; 6: 286–294. Wang Y, Pratt JR, Hartley B et al. Expression of tissue type plasminogen activator and type 1 plasminogen activator inhibitor, and persistent fibrin deposition in chronic renal allograft failure. Kidney Int 1997; 52: 371–377. Collins SJ, Alexander SL, Lopez-Guisa JM et al. Plasminogen activator inhibitor-1 deficiency has renal benefits but some adverse systemic consequences in diabetic mice. Nephron Exp Nephrol 2006; 104: e23–e34. Oda T, Jung YO, Kim HS et al. PAI-1 deficiency attenuates the fibrogenic response to ureteral obstruction. Kidney Int 2001; 60: 587–596. Carmeliet P, Stassen JM, Schoonjans L et al. Plasminogen activator inhibitor-1 gene-deficient mice. II. Effects on hemostasis, thrombosis, and thrombolysis. J Clin Invest 1993; 92: 2756–2760. Hertig A, Rondeau E. Role of the coagulation/fibrinolysis system in fibrinassociated glomerular injury. J Am Soc Nephrol 2004; 15: 844–853. Kitching AR, Holdsworth SR, Ploplis VA et al. Plasminogen and plasminogen activators protect against renal injury in crescentic glomerulonephritis. J Exp Med 1997; 185: 963–968. Kitching AR, Kong YZ, Huang XR et al. Plasminogen activator inhibitor-1 is a significant determinant of renal injury in experimental crescentic glomerulonephritis. J Am Soc Nephrol 2003; 14: 1487–1495. Ivanyi B, Thoenes W. Microvascular injury and repair in acute human bacterial pyelonephritis. Virchows Arch A Pathol Anat Histopathol 1987; 411: 257–265. Haraoka M, Hang L, Frendeus B et al. Neutrophil recruitment and resistance to urinary tract infection. J Infect Dis 1999; 180: 1220–1229. Patole PS, Schubert S, Hildinger K et al. Toll-like receptor-4: renal cells and bone marrow cells signal for neutrophil recruitment during pyelonephritis. Kidney Int 2005; 68: 2582–2587. Chassin C, Goujon JM, Darche S et al. Renal collecting duct epithelial cells react to pyelonephritis-associated Escherichia coli by activating distinct TLR4-dependent and -independent inflammatory pathways. J Immunol 2006; 177: 4773–4784. Ward JR, Dower SK, Whyte MK et al. Potentiation of TLR4 signalling by plasmin activity. Biochem Biophys Res Commun 2006; 341: 299–303. Kwak SH, Wang XQ, He Q et al. Plasminogen activator inhibitor-1 potentiates LPS-induced neutrophil activation through a JNK-mediated pathway. Thromb Haemost 2006; 95: 829–835. Werb Z. ECM and cell surface proteolysis: regulating cellular ecology. Cell 1997; 91: 439–442.

Kidney International (2009) 75, 52–59

50.

51. 52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

Chavakis T, Kanse SM, Yutzy B et al. Vitronectin concentrates proteolytic activity on the cell surface and extracellular matrix by trapping soluble urokinase receptor-urokinase complexes. Blood 1998; 91: 2305–2312. Lijnen HR. Pleiotropic functions of plasminogen activator inhibitor-1. J Thromb Haemost 2005; 3: 35–45. Marshall LJ, Ramdin LS, Brooks T et al. Plasminogen activator inhibitor-1 supports IL-8-mediated neutrophil transendothelial migration by inhibition of the constitutive shedding of endothelial IL-8/heparan sulfate/syndecan-1 complexes. J Immunol 2003; 171: 2057–2065. Samanta AK, Oppenheim JJ, Matsushima K. Interleukin 8 (monocytederived neutrophil chemotactic factor) dynamically regulates its own receptor expression on human neutrophils. J Biol Chem 1990; 265: 183–189. Hechtman DH, Cybulsky MI, Fuchs HJ et al. Intravascular IL-8. Inhibitor of polymorphonuclear leukocyte accumulation at sites of acute inflammation. J Immunol 1991; 147: 883–892. Simonet WS, Hughes TM, Nguyen HQ et al. Long-term impaired neutrophil migration in mice overexpressing human interleukin-8. J Clin Invest 1994; 94: 1310–1319. Olszyna DP, Opal SM, Prins JM et al. Chemotactic activity of CXC chemokines interleukin-8, growth-related oncogene-alpha, and epithelial cell-derived neutrophil-activating protein-78 in urine of patients with urosepsis. J Infect Dis 2000; 182: 1731–1737. Frendeus B, Godaly G, Hang L et al. Interleukin 8 receptor deficiency confers susceptibility to acute experimental pyelonephritis and may have a human counterpart. J Exp Med 2000; 192: 881–890. Hang L, Frendeus B, Godaly G et al. Interleukin-8 receptor knockout mice have subepithelial neutrophil entrapment and renal scarring following acute pyelonephritis. J Infect Dis 2000; 182: 1738–1748. Artifoni L, Negrisolo S, Montini G et al. Interleukin-8 and CXCR1 receptor functional polymorphisms and susceptibility to acute pyelonephritis. J Urol 2007; 177: 1102–1106. Carmeliet P, Kieckens L, Schoonjans L et al. Plasminogen activator inhibitor-1 gene-deficient mice. I. Generation by homologous recombination and characterization. J Clin Invest 1993; 92: 2746–2755. Olszyna DP, Florquin S, Sewnath M et al. CXC chemokine receptor 2 contributes to host defense in murine urinary tract infection. J Infect Dis 2001; 184: 301–307. Hopkins WJ, Gendron-Fitzpatrick A, Balish E et al. Time course and host responses to Escherichia coli urinary tract infection in genetically distinct mouse strains. Infect Immun 1998; 66: 2798–2802. Rouschop KM, Sylva M, Teske GJD et al. Urothelial CD44 facilitates Escherichia coli infection of the murine urinary tract. J Immunol 2006; 177: 7225–7232. Roelofs JJ, Rouschop KM, Leemans JC et al. Tissue-type plasminogen activator modulates inflammatory responses and renal function in ischemia reperfusion injury. J Am Soc Nephrol 2006; 17: 131–140. Boot RG, van Achterberg TA, van Aken BE et al. Strong induction of members of the chitinase family of proteins in atherosclerosis: chitotriosidase and human cartilage gp-39 expressed in lesion macrophages. Arterioscler Thromb Vasc Biol 1999; 19: 687–694. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976; 72: 248–254.

59