Urinary tract infection in iNOS-deficient mice with focus onbacterial sensitivi - 《American Journal of Physiology》美国生理学杂志干细胞之家



免疫细胞治疗专区	欢迎关注干细胞微信公众号

12 3 4 5 6 7 8 9 10 ... 26 下一页

返回列表

查看: 593537\|回复: 256	go Urinary tract infection in iNOS-deficient mice with focus onbacterial sensitivi [复制链接]

轻羽

新手上路

Rank: 1

积分: 0
威望: 0
包包: 0

楼主

发表于 2009-4-21 13:24 |只看该作者 |倒序浏览 |打印

作者：MirjanaPoljakovic and KatarinaPersson,作者单位：1 Department of Clinical Pharmacology, LundUniversity Hospital, SE-221 85 Lund; and Department of Chemistry and Biomedical Sciences,University of Kalmar, SE-391 82 Kalmar, Sweden
   ' ^% ~9 h- i$ k3 I; w0 A. x
   & m/ L9 x  l' Z( {2 B) p# D  n



         + ]' e9 Q! `. W- t
         , D* Z6 C& Y2 u

                  * e2 ?" n" d5 i
      6 S9 J5 s9 O! d. r


      【摘要】
   Inducible nitric oxidesynthase (iNOS)-deficient mice were used to examine the role ofiNOS in Escherichia coli -induced urinary tract infection(UTI). The toxicity of nitric oxide (NO)/peroxynitrite to bacteria andhost was also investigated. The nitrite levels in urine ofiNOS  /  but not iNOS / mice increased afterinfection. No differences in bacterial clearance or persistence werenoted between the genotypes. In vitro, the uropathogenic E. coli 1177 was sensitive to 3-morpholinosydnonimine, whereas theavirulent E. coli HB101 was sensitive to both NO and 3-morpholinosydnonimine. E. coli HB101 was statistically( P 0.05) more sensitive to peroxynitrite than E. coli 1177. Nitrotyrosine immunoreactivity was observed ininfected bladders of both genotypes and in infected kidneys ofiNOS  /  mice. Myeloperoxidase, neuronal (n)NOS, andendothelial (e)NOS immunoreactivity was observed in inflammatory cellsof both genotypes. Our results indicate that iNOS / andiNOS  /  mice are equally susceptible to E. coli -induced UTI and that the toxicity of NO to E. coli depends on bacterial virulence. Furthermore, myeloperoxidase andnNOS/eNOS may contribute to nitrotyrosine formation in the absence of iNOS.
      【关键词】 inducible nitric oxide synthase myeloperoxidase Escherichia coli transgenic nitrotyrosine
   INTRODUCTION
! _0 S- l1 s4 E0 w
URINARY TRACT INFECTIONS (UTIs), including cystitis and pyelonephritis, are among the mostcommon bacterial infections in man and Escherichia coli isthe most causative agent of the infection. The bacteria are clearedfrom the urinary tract through the action of inflammatory cells,particularly by polymorphonuclear (PMN) cells ( 23 ). Theproduction of antimicrobial factors such as nitric oxide (NO) maycontribute to control UTIs. It has been demonstrated that patients withUTI have an elevated nitrite concentration in the urine compared withhealthy controls ( 42 ). Increased gaseous NO concentrationsin the urinary bladder in patients with lower UTI have also beenreported ( 30 ). Previously, we have demonstrated increasedurinary nitrite production in E. coli -infected mice( 41 ). However, the actual role of NO in UTI has not been elucidated, and it is not clear whether NO has a bactericidal effect on UTI.

In inflammatory responses, NO is produced by an enzyme known asinducible nitric oxide synthase (iNOS). Genetically altered mice with amutation of the iNOS gene have been used to define the role of NOproduction and iNOS expression in infection. Mice lacking iNOS weremore susceptible to herpes simplex virus infection than theircorresponding wild-type controls ( 31 ), and NO production seemed to be important in the host response to extracellulargram-positive bacteria ( 33 ). However, disruption of iNOSimproved the clearance of Mycobacterium avium, and increasedNO production seemed to exacerbate this infection rather than clear itup ( 19 ).  N4 X, ^3 T" }1 W6 n4 y4 [- G

NO itself is neither highly reactive nor particularly toxic but formsoxidants that are responsible for its toxicity. The chemical reactivityand toxicity of NO can be increased by its diffusion-limited reactionwith superoxide (O 2 − ) to form peroxynitrite( 4 ). Peroxynitrite is a major damaging oxidant that islikely to account for most of the cytotoxicity commonly attributed toNO. Once formed, peroxynitrite reacts with most biological molecules,and modification of structural proteins may be an importantpathological target ( 2 ). The nitration of protein tyrosineresidues by peroxynitrite produces 3-nitrotyrosine, an index ofperoxynitrite formation in vivo ( 4 ). In inflammatory processes, tissue nitration is preferentially located around areas inwhich peroxynitrite-producing cells are most abundant, and strongnitrotyrosine immunoreactivity is usually observed in macrophage- orneutrophil-rich areas ( 7, 14, 28 ). Macrophages,neutrophils, and other phagocytic cells are known to generate largeamounts of toxic molecules, including peroxynitrite, which has beenfound to be bactericidal ( 49 ). Neutrophils in the urinarytract have been shown to express iNOS after a bacterial infection( 41, 46 ), but it remains to be established whether urinaryneutrophils also produce peroxynitrite. Extensive nitrotyrosineexpression has been observed in proximal tubular epithelial cells inkidneys of LPS-treated rats ( 6 ), but nitrotyrosineexpression has not been investigated in bladder tissue., P4 Z" _$ F0 q$ _9 M

This study examined the significance of iNOS induction and NOproduction for the clearance of E. coli from the urinarytract of wild-type and iNOS-deficient mice. Furthermore, the toxicity of NO and peroxynitrite to different E. coli strains wasexamined in vitro and the toxicity of NO to the host tissue wasassessed by investigation of nitrotyrosine formation.0 o5 |6 d0 t4 z. _* B
; c/ i* D  V1 c
METHODS6 t5 r" H) w# E8 q4 ^$ C
- X4 W0 U7 p4 ]
Mice

C57BL/6J and C57BL/6-NOS2 / mice were obtainedfrom The Jackson Laboratory (Bar Harbor, ME). The mice were constructedas previously described ( 29 ). Female mice were used at 9 wk of age. The mice were kept on a nitrate/nitrite-free pellet diet(Altromin 1324N, Petersen, Ringsted, Denmark), starting 1 wk before theexperiment. A few days before infection, groups of five mice wereseparated into individual cages, and urine samples were collected andexamined for bacterial growth and the leukocyte cell content wasdetermined microscopically using a hemocytometer chamber. Mice withmore than 5 × 10 5 leukocytes/ml in preinoculationurine samples or with a positive bacterial culture were excluded fromthe experiment. The experimental protocol was approved by the AnimalEthics Committee, Lund University, Lund, Sweden.
' Y7 m( ?" ~4 \$ }
Bacteria/ Z; U" ?& @+ u- v( P& J

In vivo experiments. E. coli 1177, of serotype O1:K1:H7, was isolated from achild with acute pyelonephritis ( 32 ). The strain isvirulent in the mouse UTI model and evokes a strong inflammatory hostresponse ( 10 ). It expresses type 1 and P fimbrial adhesinsbut is hemolysin negative. E. coli 1177 was maintained indeep agar stabs, passaged on tryptic soy agar (TSA; Difco Laboratories,Detroit, MI), grown overnight at 37°C in static Luria broth, andharvested by centrifugation at 3,200 rpm for 10 min. The pellet wasresuspended in sterile PBS, pH 7.2, to a concentration of10 9 colony-forming units (CFU)/ml.* ]7 i# d( J& g
  f, O, q- k& f2 w8 Z! z- J1 v
In vitro experiments. When used for the in vitro experiments, E. coli 1177 wasgrown overnight at 37°C on TSA and harvested in sterile PBS bycentrifugation at 3,200 rpm for 10 min. As an avirulent strain, anonfimbriated K12 derivative, E. coli HB101, was used. HB101was grown overnight at 37°C on TSA and harvested in sterile PBS bycentrifugation at 3,200 rpm for 10 min. The pellets of both bacterialstrains were resuspended and then diluted in sterile PBS to a finalconcentration of 10 9 CFU/ml.4 U0 \! P% c- @

Infection Procedure

The mouse bladder was emptied by gentle compression on the lowerabdomen, and urine was saved for preinfection measurements of nitrite(see below). Experimental UTI was established in the mice byintravesical injection of E. coli 1177 as previouslydescribed ( 22 ). After anesthesia, 0.1 ml of bacterialsuspension was slowly instilled into the bladder transurethrally, usinga soft polyethylene catheter (outer diameter 0.61 mm; Kebolab,Malmö, Sweden). The catheter was immediately withdrawn afterinoculation, and no further manipulations were carried out. The animalswere placed in the cages after instillation and allowed food and waterad libitum. Infection was monitored at 6, 24, and 72 h and 7 days.Urine samples were stored on ice or at 4°C until further analysis.$ Z) Y& q! e1 u- \% A$ H

Infection was quantified by viable bacterial counts on tissuehomogenates from mice killed by CO 2 asphyxia at differenttimes after inoculation. One-half of the bladder and one of the kidneys were aseptically harvested and homogenized in 5 ml of sterile PBS insterile disposable plastic bags using a LAB Stomacher 80 Homogenizer(Seward Medical UAC House, London, UK). Serial dilutions of the tissuehomogenates were plated on TSA. After overnight culture, plates withbacterial colonies were scored and bacterial numbers (CFU/ml of tissuehomogenate) were determined after adjustment for the dilution factor.The other half of the bladder and the second kidney were processed forimmunohistochemistry as described below.

Immunohistochemistry

Bladders and kidneys were immersion-fixed for 4 h in cold4% formaldehyde in PBS (pH 7.4) and then rinsed for 3 days in PBS containing 15% sucrose. Both fixation and rinsing were performed at4°C, after which the specimens were frozen in isopentane at 40°Cand stored at 70°C until sectioning. Sections were cut (10 µm) ina cryostat (Leica CM3050; Leica Microsystems, Askik, Sweden) andpreincubated with PBS containing 0.2% Triton X-100 and 0.1% BSA for2 h at room temperature. Sections were incubated with thefollowing primary antisera (diluted with PBS containing 0.2% TritonX-100 and 0.1% BSA): a rabbit polyclonal antibody raised to murineiNOS (1:1,000; Santa Cruz Biotechnology, Santa Cruz, CA); a rabbitpolyclonal antibody raised to nitrotyrosine (1:470; UpstateBiotechnology, Lake Placid, NY); a sheep polyclonal antibody raised toneuronal (n)NOS (1:1,000; a gift from Drs. I. Charles and P. C. Emson, Cambridge Univ., Cambridge, UK); or a rabbit polyclonal antibodyraised to endothelial (e)NOS (1:250; Santa Cruz Biotechnology)overnight in a moisture chamber at room temperature. Sections wereincubated with a rabbit polyclonal antibody raised to MPO (1:200;NeoMarkers, Fremont, CA) for 2 h at room temperature. The sectionswere rinsed in PBS and incubated for 90 min with FITC-conjugated donkeyanti-rabbit IgG (1:80), Texas red (TR)-conjugated F(ab') 2 fragment donkey anti-rabbit IgG (1:160; both from JacksonImmunoresearch Laboratories, West Grove, PA), or FITC-conjugated donkeyanti-sheep IgG (1:80; Sigma) diluted in PBS. The sections were rinsedin PBS and mounted in glycerol with p -phenylenediamine toprevent fading.: }) i& R8 ~6 U: H) I

To identify the inflammatory cells, a double-label immunofluorescencemethod was used. Sections were first incubated overnight withRB6-8C5, a rat IgG2b monoclonal antibody specific for murine neutrophils (a gift from Dr. A. Sjöstedt, Umeå University,Umeå, Sweden). After being rinsed in PBS, the iNOS antibody was added and the sections were incubated again overnight. RB6-8C5 staining was visualized by incubating the sections for 90 min with TR-conjugated F(ab') 2 fragment donkey anti-rat IgG (1:160). After beingrinsed, the sections were incubated for 90 min with FITC-conjugateddonkey anti-rabbit IgG (1:80) to visualize iNOS. The sections weremounted as described above.$ Z! C. h, r, f( s
: b; a) _3 x- g/ L
In control experiments, no immunoreactivity was detected in sectionsincubated with only the secondary antibodies. The specificity of theiNOS antibody was confirmed by incubating the antibody with excessiveantigen. No iNOS immunoreactivity was detected in sections incubatedwith absorbed antibody. All micrographs of the immunolabeled sectionswere obtained using a digital camera system (Olympus BX60F-3 microscopeand Olympus Digital camera DP-50; Olympus Optical, Tokyo, Japan), andthe pictures were captured using appropriate filter settings for FITCand TR. Adobe Photoshop was used for image handling, and thethree-color channels were handled separately. Only the backgroundlevel, contrast, and brightness of the entire image were changed in thefinal picture.( E5 g5 O* d& s) d" j) \

Hematoxylin and Eosin Staining' V1 x# a+ Y/ C4 U7 H

The morphology of the inflammatory cells was examined afterstaining of the tissue sections with hematoxylin and eosin(Apoteksbolaget, Malmö, Sweden).
  @. f1 F  }; M, N8 `& N; }. p
DNA Isolation and Genotyping with PCR: o6 i' K3 e$ l" R; q5 E" w
6 \; e* Y. p" B7 K; j
PCR analyses for the intact and disrupted iNOS gene wereperformed to genotype the mice. Briefly, DNA was extracted from the tipof the tail by incubating the tail in lysis buffer (50 mM Tris · Cl, 5 mM EDTA, 100 mM NaCl, and 0.5% SDS;all from Sigma, St. Louis, MO) containing proteinase K (10 mg/ml;Sigma) overnight at 56°C. PCR was performed according to the SigmaPCR-Core kit with Taq DNA polymerase (Sigma), using 2 µggenomic DNA. The primers for mouse iNOS were obtained from MWG-Biotech(Ebersberg, Germany) and were as follows: sense, 5'-CAG ATC GAG CCC TGGAAG ACC-3', and antisense, 5'-CCT TGG TGT TGA AGG CGT AGC-3',amplifying a 533-bp product. PCR was performed in an automated thermalcycler (GeneAmp PCR System 2400, PerkinElmer, Foster City, CA), with one cycle at 94°C for 5 min, followed by 30 cycles at 94°C for 60 s, at 58°C for 60 s, at 72°C for 60 s, and afinal extension at 72°C for 7 min. PCR products were separated by 2%agarose gel electrophoresis, and bands were visualized by ethidiumbromide staining." |" y) H0 F- u/ c0 x; I1 h/ e1 p. Z
: R7 Z" @+ ~0 j1 r4 S0 `+ x5 F3 f
Nitrite Assay
% C8 y, m# e7 J
NO production was determined as urinary nitrite levels by theGriess reaction assay. Briefly, 10 µl of centrifuged urine were transferred to 96-well plates and mixed with 20 µl of water and 100 µl of Griess reagent [1 part 0.1% N -(1-naphtyl)ethylene-diamine dihydrochloride (Sigma) in water and 1 part 1%sulfanilamide (Sigma) in 5% concentratedH 3 PO 4 ]. The mixture was incubated for 5 min atroom temperature and read at 540 nm by spectrophotometry (Labsystems Multiscan PLUS; Labsystems, Lund, Sweden). Concentrations were determined compared with a standard curve of sodium nitrite. The detection limit of the assay was 1 µM nitrite.* ^, i2 z5 C* m! _4 Z. @* p6 k5 B

Bacterial Viability Experiments6 q8 V+ O- [4 {9 z% [; r# _6 ?8 B8 S
, E; I: g- q' j1 i9 ^5 F: M. T( Q
Bacterial viability in response to exogenously applied NO and3-morpholinosydnonimine (SIN-1), a generator of peroxynitrite insolution, was examined. One milliliter (10 9 CFU/ml) of E. coli 1177 or E. coli HB101 in PBS wastransferred to a sterile test tube. The bacteria were exposed to eitherNO [2,2-(hydroxynitrosohydrazino)bis-ethanamine (DETA/NO); 500 µM, Alexis Biochemicals, Lausen, Switzerland] or SIN-1 (500 µM, Casella, Frankfurt am Main, Germany) for 24 h. Untreated bacteria were usedas control. Serial dilutions were plated on TSA and, after overnightculture, plates with bacterial colonies were scored and bacterialnumbers (CFU/ml) were determined after adjustment for the dilution factor.
. K. Y4 D; C0 P) W1 S- i
Analysis of Data' F8 O: `+ W& m/ J& ^

Data are expressed as means ± SE. Student's paired orunpaired t -test was used to compare two means, and ANOVAfollowed by the Bonferroni-Dunn post hoc test was used for multiplecomparisons (GraphPad Prism 3.0). Differences were consideredsignificant at P# N. w' x9 ]+ R, t4 m) q9 `3 A

RESULTS

The genotype of the mice was confirmed by PCR using primersspecific for the calmodulin-binding domain of the iNOS gene. A 533-bpproduct, corresponding to the calmodulin-binding domain of iNOS, wasdetected in all the investigated C57BL/6J mice (iNOS  /  )but not in the investigated transgenic iNOS / mice (Fig. 1 ).

Fig. 1. PCR genotypic analysis of wild-type and inducible nitricoxide synthase (iNOS)-deficient mice. Lanes designated  /  show thepresence of the calcium-calmodulin binding domain of iNOS DNA inwild-type mice. Lanes designated / show the absence of thecalcium-calmodulin binding domain of iNOS DNA in iNOS-deficient mice.% \5 h$ r# G% m. \& e, h6 @8 L

Nitrite Levels in Urine

The production of NO into the urine in response to infection wasquantified as nitrite, a stable end product of NO ( 20 ). Most uropathogens are members of the family Enterobacteriaceae, known to reduce nitrate to nitrite.Urine contains nitrates from dietary sources, but the mice in our studywere fed a nitrate/nitrite-free diet to ensure that we measured nitriteformed from endogenous NO and not from bacterially converted nitrate.The urinary nitrite levels increased in wild-type mice after infectionwith E. coli 1177 at 6 h postinfection (42 ± 12 µM, n = 12, P withat 0 h (12 ± 1 µM, n = 12) (Fig. 2 A ). The urinary nitrite levels were low at 24 and 72 h and 7 days after bacterialinstillation. iNOS / mice showed no urinary nitriteresponse to infection (Fig. 2 B ).

Fig. 2. Nitrite levels in urine samples obtained afterinstillation of Escherichia coli strain 1177 in wild-type( A ) and iNOS-deficient mice ( B ). Thepreinoculation nitrite values are given at time 0. Valuesare means ± SE ( n = 3-24).* P  {" A1 r9 W- z6 A

Bacterial Counts of Infected Tissue

Bladders and kidneys were harvested from the E. coli 1177-infected iNOS  /  and iNOS / mice, andbacterial persistence was determined by viable counts on tissuehomogenates. There was no significant difference in bacterialpersistence or clearance in bladders obtained from iNOS wild-type oriNOS-deficient mice (Fig. 3 A ).The highest bacterial numbers were found 24 h after infection inwild-type mice and 6 h after infection in iNOS-deficient mice.Between 72 h and 7 days after infection, the bacterial numberswere fairly constant and none of the mice strains managed to clear theinfection.$ `7 x, I+ ?( o  S# q

Fig. 3. Wild-type ( ) and iNOS-deficient( ) mice were inoculated with E. coli 1177, and the infection was monitored by viable bacterial counts of bladder( A ) and kidney homogenates ( B ) after 6, 24, and72 h or 7 days. There was no significant difference in bacterialclearance or persistence between the genotypes. Values are means ± SE expressed as colony-forming units (CFU)/ml ( n = 3-6).

There was no difference in bacterial numbers in kidneys fromiNOS / and wild-type mice (Fig. 3 B ). After areduction between 24 and 72 h postinfection, the bacterial numbersof wild-type mice remained stationary for up to 7 days. The bacterialnumbers in kidneys of iNOS / mice increased up to72 h and then decreased slightly between 72 h and 7 days toreach the same numbers as observed in kidneys of iNOS  /  mice.
, N; X! W! a( X5 v1 M
Morphological Examination
4 {6 l9 {% d- |7 A0 z3 U( O
Bladders and kidneys were examined for iNOS, nNOS, eNOS,nitrotyrosine, and MPO expression by immunohistochemistry at 6, 24, and72 h and 7 days postinstillation.& I: B8 Z$ F! O

Identification of inflammatory cells. Hematoxylin and eosin staining showed that the majority of theinflammatory cells in bladders and kidneys were PMN cells. Bydouble-label immunofluorescence using the neutrophil marker RB6-8C5, the majority of the iNOS-positive cells were identified as neutrophils (data not shown).# C3 U/ C0 B: X7 d# h
2 \) j1 l% s3 V+ u- }0 M7 |0 b
iNOS immunoreactivity. Low levels of two abnormal iNOS transcripts have been detected inmacrophages isolated from iNOS / mice ( 29 ).These transcripts may produce low levels of immunoreactive peptidesthat are enzymatically inactive but recognized by some iNOS antibodies(The Jackson Laboratory, personal communication). The iNOS antibodyused in our study detected iNOS in inflammatory cells in bladders ofboth wild-type and iNOS-deficient mice 6 h after infection (datanot shown). iNOS-positive uroepithelial cells were found in themajority of bladders from iNOS  /  mice from 6 h and upto 7 days after infection (Fig. 4 A ). iNOS immunoreactivitywas, however, not observed in the bladder urothelium ofiNOS / mice (Fig. 4 B ). Kidneys of infectediNOS  /  and iNOS / mice were devoid ofiNOS-expressing inflammatory cells at all times. iNOS immunoreactivityin transitional and columnar epithelial cells lining therenal pelvis was found in the majority of wild-type mice at 6 and24 h postinstillation, and this immunoreactivity was furtherincreased at 72 h and 7 days after infection (Fig. 4 C ).The iNOS-deficient mice did not express iNOS immunoreactivity inepithelial cells lining the renal pelvis (Fig. 4 D ).
8 ~8 \8 x' k3 ^/ W8 b7 j; ]3 ^
Fig. 4. Immunohistochemistry of iNOS in bladder and kidneysections from E. coli 1177-infected mice. A : iNOSimmunoreactivity in the bladder urothelium of iNOS  /  mice. B : iNOS immunoreactivity was not observed in the bladderurothelium of iNOS-deficient mice. C : iNOS immunoreactivityin epithelial cells lining the renal pelvis ofiNOS  /  mice. D : iNOS immunoreactivity was notobserved in epithelial cells lining the renal pelvis of iNOS-deficientmice. u, Urothelium; l, lumen of the bladder/renal pelvis. Scalebars = 30 µm.' L' X8 V: g) \/ r( S6 ?' \+ n
: O  n7 \$ c: h5 M
nNOS and eNOS immunoreactivity. We next examined whether a compensatory increase in nNOS and eNOSexpression had occurred in iNOS-deficient mice. In the bladder, nNOS-immunoreactive neuronal structures were observed within the smoothmuscle and submucosa of E. coli -infected bladders of both wild-type and iNOS-deficient mice. In the kidney, nNOS immunoreactivity was sparse and only observed in the macula densa and in a few nervefibers associated with blood vessels. eNOS-positive immunoreactivity was observed in vascular endothelium. The inflammatory cells of bothgenotypes were found to express nNOS and eNOS (Fig. 5, A and B ). As judged by visualobservation, no difference in nNOS or eNOS expression was observed wheninfected iNOS-deficient mice were compared with infected wild-typemice.

Fig. 5. Immunohistochemistry of neuronal (n)NOS and endothelial (e)NOS inbladder sections from E. coli 1177-infected mice. A : nNOS immunoreactivity in nerve fibers (arrows) and ininflammatory cells (arrowheads) in bladders of iNOS-deficient mice. B : eNOS immunoreactivity in vascular endothelium (arrow) andin inflammatory cells (arrowheads) in bladders of iNOS-deficient mice.Scale bars = 60 µm.
. w6 v5 |7 q# N8 x$ H' v2 W
Nitrotyrosine immunoreactivity. Immunoreactivity to nitrotyrosine was not found in uninfected bladdersand kidneys. Nitrotyrosine immunoreactivity was, however, found inbladders of infected iNOS  /  and iNOS / miceat all times (Fig. 6 A ). Threetypes of cells/structures were found to express nitrotyrosine. PMNcells and large round cells, resembling inflammatory cells, showednitrotyrosine staining (Fig. 6 B ). In addition, smallnitrotyrosine-positive structures, with no visible nucleus, were foundclose to the inflammatory cells in the submucosa (Fig. 6 B ).In kidneys of wild-type mice, nitrotyrosine was observed ininflammatory cells and in small structures in the renal pelvisprimarily at 72 h and 7 days postinfection (Fig. 6 C ).Nitrotyrosine immunoreactivity was found in the glomeruli of oneiNOS  /  mouse 72 h after bacterial infection (Fig. 6 D ). Nitrotyrosine was not detected in kidneys ofiNOS / mice (Fig. 6 E ).9 |# k  M: a' Y" R; p% r: a3 _
* `4 U! q) G$ P, t( l
Fig. 6. Immunohistochemistry of nitrotyrosine in bladder and kidneysections from E. coli 1177-infected mice. A :immunoreactivity to nitrotyrosine in inflammatory cells in bladders ofwild-type mice. B : detailed demonstration ofnitrotyrosine-positive cells/structures. Small nitrotyrosine-positivestructures (arrows) were observed in the bladder submucosa and close tothe urothelium. Small arrowheads, polymorphonuclear cells; largearrowheads, round cells with intense nitrotyrosine staining. C : immunoreactivity to nitrotyrosine in inflammatory cellsin kidneys of wild-type mice. D : nitrotyrosineimmunoreactivity in the glomeruli of a wild-type mouse. E :immunoreactivity to nitrotyrosine was not observed in kidneys ofiNOS-deficient mice. p, Renal papilla. Scale bars: 120 µm( A and C ), 60 µm ( B and E ), 30 µm ( D )./ A3 j/ K' }4 q* P; u+ e6 p3 Y! k

MPO immunoreactivity. Because nitrotyrosine immunoreactivity was observed in iNOS-deficientmice, possible sources for protein nitration other than peroxynitritewere investigated. MPO is an enzyme used by granulocytes duringphagocytosis. In the presence of nitrite, MPO can nitrate proteintyrosines ( 13 ). No immunoreactivity to MPO was observed inuninfected bladders and kidneys. Numerous MPO-immunoreactive inflammatory cells were observed in infected bladders of both iNOS  /  (Fig. 7 A )and iNOS / (Fig. 7 B ) mice at all times. TheMPO-immunoreactive cells were located within the bladder smooth musclelayer, lamina propria, and close to or within the uroepithelium. In thekidney, inflammatory cells expressing MPO were observed close to theepithelium lining the renal pelvis mainly at 6 h after infection.Some MPO-immunoreactive inflammatory cells were also observed in theglomeruli (Fig. 7 C ) at all times. There was no difference inthe number or distribution of MPO-expressing cells between thegenotypes as judged by visual observation. However, MPOimmunoreactivity was by far more pronounced in the bladder than in thekidney. The uroepithelial cells were not stained for MPO.
: V; o8 A7 T; y: R
Fig. 7. Immunohistochemistry of MPO in bladder and kidney sections from E. coli 1177-infected mice. A : MPOimmunoreactivity in inflammatory cells in bladders of wild-type mice. B : MPO immunoreactivity in inflammatory cells in bladders ofiNOS-deficient mice. C : immunoreactivity to MPO ininflammatory cells (arrowheads) in kidney glomeruli of iNOS-deficientmice. g, Glomeruli. Scale bars = 60 µm.

Bacterial Viability

The effect of exogenously added NO and SIN-1, a peroxynitritedonor in solution, on bacterial viability was investigated. When E. coli 1177 was exposed to DETA/NO (500 µM), whichspontaneously decomposes and provides a constant NO supply over hours( 3 ), the number of colonies formed was only slightly(14 ± 8.4%, n = 3) decreased compared withuntreated controls. However, when exposed to SIN-1 (500 µM), thenumber of viable colonies of E. coli 1177 was significantlydecreased by 53 ± 12% ( n = 3, P 0.05) compared with untreated bacteria (Fig. 8 A). HB101, the nonfimbriated E. coli strain, was sensitive to both NO and SIN-1. DETA/NO (500 µM) and SIN-1 (500 µM) caused a significant decrease in E. coli HB101 viability by 25 ± 7.8 ( n = 3, P n = 3, P 8 B ). E. coli HB101 was statistically ( P E. coli 1177.

Fig. 8. Bacterial viability in vitro in response to nitric oxide(NO) and 3-morpholinosydnonimine (SIN-1), a generator of peroxynitritein solution. A : E. coli 1177 was sensitive toSIN-1 (500 µM) but not to NO[2,2-(hydroxynitrosohydrazino)bis-ethanamine (DETA/NO); 500 µM]. B : E. coli HB101 was sensitive to both SIN-1 (500 µM) and NO (DETA/NO; 500 µM). Values are means ± SE, and theviability is expressed as percentage of untreated controls (set at100%; n = 3). * P P
! A6 Q# _: }) Q5 U& Z
DISCUSSION) d$ [# p6 A) z! @& s
. P% O! N) v, @1 s! r. e' ^8 S8 D- g/ k
In this study, iNOS-deficient mice were used to examine the roleof iNOS expression and NO production in bacterial clearance after an E. coli -induced UTI. Uropathogenic E. coli wasfound to persist in bladders and kidneys of iNOS  /  andiNOS / mice for up to 7 days after infection, and noneof the mice strains managed to clear the infection. Bacteria havepreviously been reported to persist within the bladder of C57BL/6 micefor days and weeks ( 26, 37 ), with the persistence beingattributed to the type 1 pili ( 10, 36 ). There was nosignificant difference in bacterial clearance or persistence betweenwild-type and iNOS / mice, suggesting that the bacteriapersisted in bladders and kidneys irrespective of iNOS induction. Thesefindings are consistent with an earlier study, showing thatpharmacological inhibition of iNOS did not alter the sensitivity ofC3H/HeN and C3H/HeJ mice to renal infection caused by two different E. coli uropathogens ( 39 ). Furthermore, inexperimental glomerulonephritis, no difference in the disease wasobserved when iNOS-deficient mice were compared with wild-type mice( 9 ).1 V; ?6 W' m. ?( \  H+ Z/ b
2 }) G9 c& h5 z1 {
Immunoreactivity to iNOS was observed in bladder inflammatory cells ofiNOS  /  mice and, surprisingly, also in theiNOS / mice. Low expression of two abnormal iNOStranscripts, which may produce immunoreactive peptides, has beendescribed in macrophages isolated from the iNOS / miceused in this study ( 29 ). Some iNOS antibodies mayrecognize the produced peptides and, indeed, several investigators have noticed iNOS-positive labeling in the iNOS / mice (TheJackson Laboratory, personal communication). Nevertheless, thesetranscripts are enzymatically defective and do not produce NO. In ourstudy, no increase in urinary nitrite levels of infected iNOS / mice was detected, confirming the absence of iNOSactivity in these animals. Genotypic studies further confirmed thedisruption of the iNOS gene in transgenic mice. The nitrite levels inurine from E. coli -infected iNOS  /  mice reacheda peak 6 h after instillation, which coincided in time with thepresence of numerous iNOS-positive inflammatory cells in the bladder.iNOS immunoreactivity in uroepithelial cells was found up to 7 daysafter infection, but urinary nitrite levels were not increased insamples analyzed 1, 3, or 7 days postinfection. This suggests thatinflammatory cells, and not uroepithelial cells, are the maincontributors of NO/nitrite in urine samples of infected mice.

We have previously demonstrated iNOS immunoreactivity in uroepithelialcells at 24 and 72 h after infection in C3H/HeN mice infected with E. coli ( 41 ). In the present study, usingC57BL/6 wild-type mice and a different E. coli strain, iNOSimmunoreactivity in uroepithelial cells was seen as early as 6 hpostinfection, and the expression increased with time. Thus the geneticbackground of the mice as well as the bacterial strain seem to affectthe rate and degree of the iNOS response in host uroepithelial cells. iNOS immunoreactivity was not found in uroepithelial cells in bladdersand kidneys of iNOS-deficient mice. This may suggest that uroepithelialcells in iNOS / mice, unlike inflammatory cells, do notproduce the abnormal iNOS transcripts.

It is generally believed that NO in inflammatory cells is antimicrobialand that it participates in the host defense against invading pathogens( 15 ). However, the role of iNOS induction in uroepithelialcells is not clear. Induction of iNOS in uroepithelial cells may beinvolved in uroepithelial cell shedding by promoting deletion ofinfected and damaged cells ( 16, 34 ). Inhibition of iNOSwas found to reduce intestinal permeability and to decrease bacterialtranslocation by limiting the damage to the gut mucosa ( 12, 43 ). This suggests that NO may favor bacterial translocation through the epithelium. Massive bladder uroepithelial cell shedding hasbeen reported within 6 h after infection with type 1 piliated E. coli in the same mice strain, C57BL/6, used in our study( 38 ). The degree of uroepithelial cell shedding was notspecifically investigated in the present study, and it is unclearwhether iNOS-deficient bladders and kidneys showed less shedding thantissue from wild-type mice.

Because the uropathogenic E. coli strain 1177 persisted inboth genotypes, we examined the bactericidal effect of NO to E. coli in vitro. E. coli 1177 was not sensitive to NOexposure. The compound SIN-1, which in solution generatesperoxynitrite, caused a significant decrease in bacterial viability.Consistent with our results, other studies have shown that exposure of E. coli to NO did not decrease the viability of the bacteria( 8, 39 ), whereas exposure to peroxynitrite did ( 8, 49 ). However, when a nonfimbriated avirulent E. coli strain, HB101, was exposed to NO and SIN-1, a significant decrease inviability was observed with both agents. It may be speculated whetherthe effect of NO on E. coli HB101 is a direct effect of theexogenously applied NO or whether NO has reacted with bacteriallyderived oxygen and formed peroxynitrite. E. coli containsrelatively high concentrations of catalases that could provide a sourceof oxygen ( 25 ). Added NO and bacterially derived oxygenmay thus form low concentrations of peroxynitrite that is bactericidalto E. coli HB101 but not to E. coli 1177.9 o$ v7 F& O( G9 {) z

Microbial pathogens have multiple means of defending themselves againstthe cytotoxic effects of NO. Recent evidence shows that most organisms,including E. coli, are able to metabolize and detoxify NO.Studies have shown that aerobic E. coli strains areprotected against NO toxicity by expressing an NO-inducible NOdeoxygenase (NOD) ( 17 ). NOD is a flavohemoglobin, whichoxidizes NO to NO 3 − ( 18 ), therebyprotecting E. coli against the toxic effects of NO.Additional protective mechanisms against NO include expression ofstress regulons at the DNA level. Studies on E. coli havedemonstrated the presence of specific antioxidant regulons, such as the soxRS regulon, which after induction may protect bacteriaagainst NO ( 44 ). E. coli carrying a deletion ofthe soxRS locus is hypersusceptible to NO-dependent killing( 40 ). It is likely that pathogenic bacteria benefit mostby developing NO/peroxynitrite resistance. In the present study, thepathogenic E. coli strain 1177 was significantly moreresistant to peroxynitrite than the nonpathogenic E. coli strain HB101. This suggests that the outcome of NO toxicity to E. coli in UTI may depend on bacterial virulence. Moreover, the relative resistance of E. coli 1177 to NO/peroxynitrite, asfound in vitro, may explain that no differences in bacterial clearance were noted between infected iNOS-deficient and wild-type mice.1 V$ J& {7 Q. b  Y/ m$ v8 C( @
! t* \1 v3 L2 G) _" E: i: m% ^
The cytotoxicity of NO/peroxynitrite is not only directed to invadingpathogens but may also affect NO-producing cells and surrounding tissue( 35 ). Increased iNOS expression and NO production havebeen shown to coincide in time with maximal kidney damage in acutepyelonephritis ( 27 ). Nitrotyrosine has become a useful marker of peroxynitrite formation in vivo ( 5 ).Immunoreactivity to nitrotyrosine was found in bladder inflammatorycells at all time points, demonstrating that urinary neutrophilsproduce peroxynitrite. No difference in nitrotyrosine staining in thebladder was detected between the genotypes, suggesting thatperoxynitrite formed in iNOS / mice must depend on NOsources other than iNOS. It has previously been demonstrated that thelack of iNOS does not fully abolish tyrosine nitration( 50 ). Production of NO by other isoforms of NOS is apossible source of NO in iNOS-deficient mice. In our study, nNOS andeNOS immunoreactivity was observed in inflammatory cells in bladders of E. coli -infected iNOS-deficient mice. It was recentlydemonstrated that human neutrophils express eNOS ( 11 ) andthat human and rat neutrophils express nNOS mRNA ( 21 ). Also, rat neutrophils have been shown to express nNOS protein andspontaneously release nitrite and nitrate anions ( 21 ).Purified nNOS was found to produce both O 2 − and NO andto form peroxynitrite ( 48 ). Thus evidence exists tosupport the notion that nitrotyrosine can be formed by NO derived from nNOS and/or eNOS in mice lacking iNOS. A compensatory increase in NO production from nNOS and/or eNOS may have evolved iniNOS-deficient mice. However, the expression of nNOS and eNOS proteindid not appear to differ in wild-type and iNOS / mice,at least not after immunohistochemical evaluation.
! E6 ?% T: H0 I& R5 g
MPO, an enzyme implicated in various inflammatory diseases, has beensuggested as a potential pathway for nitrotyrosine formation. MPO is amajor neutrophil protein and is stored in granules and released duringphagocytosis ( 47 ). MPO may use either hydrogen peroxide(H 2 O 2 ) or hypochlorous acid (HOCl) to oxidizenitrite and form reactive nitrogen intermediates that may result intissue nitration ( 24, 45 ). In our study, numerousMPO-stained inflammatory cells were observed in infected bladders ofboth wild-type and iNOS-deficient mice. This suggests that MPO maycontribute to the detected nitrotyrosine formation in bladders ofiNOS / mice. Nitrotyrosine expression, but no iNOSexpression, was observed in inflammatory cells in kidneys ofiNOS  /  mice. We have previously detected iNOS-positiveinflammatory cells in the kidney of infected C3H/HeN mice( 41 ). It is likely that iNOS expression in kidneyinflammatory cells in the present study, which used a different mousestrain and a more virulent bacterial strain, peaked earlier than 6 h. Furthermore, MPO-positive cells were observed in the kidney and, asdiscussed above, MPO may contribute to nitrotyrosine formation. Unlikein the bladder, nitrotyrosine was not observed in kidneys ofiNOS / mice, suggesting that no peroxynitrite- orMPO-derived nitrotyrosine formation occurred in the kidney ofiNOS-deficient mice. Indeed, MPO and nNOS/eNOS expression, the possiblealternative sources for nitrotyrosine formation in the absence of iNOS,was not as pronounced in the kidney as in the bladder.

Nitrotyrosine was also observed in the vicinity of inflammatory cellsin small submucosal structures. The inflammatory cells may secretperoxynitrite and cause nitration in membrane or intracellular compartments of target cells ( 1 ). In our study, the smallstructures observed could be nitrated structural tissue proteins ornitrotyrosine formed on the bacteria ( 14 ). Themodification of structural proteins may be a particularly importantpathological target of nitration by causing disruption of the normalfunction of cellular structures ( 2 ).) P8 d6 p1 U4 x

In conclusion, our results indicate that wild-type and iNOS-deficientmice are equally susceptible to E. coli -induced UTI and thatthe outcome of NO toxicity to E. coli may depend onbacterial virulence. Furthermore, the lack of iNOS did not abolishnitrotyrosine formation. Myeloperoxidase and nNOS/eNOS may contributeto nitrotyrosine formation in the absence of iNOS.
: ~4 b0 j" ]+ o1 K
ACKNOWLEDGEMENTS
- `6 B* X# B4 }9 r) A1 ]  `
We thank Prof. Catharina Svanborg for generously providing thebacteria and valuable comments. Dr. Kristian Moller is acknowledged forhelp with the iNOS primers.# b5 \: l1 a& `3 h, K4 |
      【参考文献】
1. Babior, BM. Phagocytes and oxidative stress. Am J Med 109:33-44,2000  .

2. Beckman, JS. Oxidative damage and tyrosine nitration from peroxynitrite. Chem Res Toxicol 9:836-844,1996  .
) }' B6 Q& J5 h6 S& D

0 I, K1 m& r& F& y" V& D) m
3. Beckman, JS. Parsing the effects of nitric oxide, S -nitrosothiols, and peroxynitrite on inducible nitric oxide synthase-dependent cardiac myocyte apoptosis. Circ Res 85:870-871,1999 .* K  s4 |' b9 x, o8 k% v

4. Beckman, JS,andKoppenol WH. Nitric oxide, superoxide, and peroxynitrite: the good, the bad, and ugly. Am J Physiol Cell Physiol 271:C1424-C1437,1996 .

# p3 C9 q/ a. s

5. Beckman, JS,Ye YZ,Anderson PG,Chen J,Accavitti MA,Tarpey MM,andWhite CR. Extensive nitration of protein tyrosines in human atherosclerosis detected by immunohistochemistry. Biol Chem Hoppe Seyler 375:81-88,1994  .

: V  \: E2 h, C; r9 h" q/ y6 g- }

6. Bian, K,Davis K,Kuret J,Binder L,andMurad F. Nitrotyrosine formation with endotoxin-induced kidney injury detected by immunohistochemistry. Am J Physiol Renal Physiol 277:F33-F40,1999 .

, s9 F4 _- X" }  r4 V0 `
7. Brito, C,Naviliat M,Tiscornia AC,Vuillier F,Gualco G,Dighiero G,Radi R,andCayota AM. Peroxynitrite inhibits T lymphocyte activation and proliferation by promoting impairment of tyrosine phosphorylation and peroxynitrite-driven apoptotic death. J Immunol 162:3356-3366,1999 .
: r; A+ V7 G& b) u+ H/ y
& B7 o+ E) z1 }

8. Brunelli, L,Crow JP,andBeckman JS. The comparative toxicity of nitric oxide and peroxynitrite to Escherichia coli. Arch Biochem Biophys 316:327-334,1995  .# C9 c9 y" F4 d, u7 p% e

8 j6 X7 K' B1 p+ O6 K
; H, b+ K/ x! I: L
9. Cattell, V,Cook HT,Ebrahim H,Waddington SN,Wei XQ,Assmann KJ,andLiew FY. Anti-GBM glomerulonephritis in mice lacking nitric oxide synthase type 2. Kidney Int 53:932-936,1998  ., F' E/ k4 x7 S! c4 {7 H) P  l: ^: m

10. Connell, I,Agace W,Klemm P,Schembri M,Marild S,andSvanborg C. Type 1 fimbrial expression enhances Escherichia coli virulence for the urinary tract. Proc Natl Acad Sci USA 93:9827-9832,1996 .3 P' z/ V, t1 W: u* S% h' p! y
4 Y5 N! s8 x$ I+ s& R! p) E
5 a3 D* w1 P& J+ r; _4 M" @2 d# Z

11. De Frutos Sanchez de Miguel, L,Farre J,Gomez J,Romero J,Marcos-Alberca P,Nunez A,Rico L,andLopez-Farre A. Expression of an endothelial-type nitric oxide synthase isoform in human neutrophils: modification by tumor necrosis factor-alpha and during acute myocardial infarction. J Am Coll Cardiol 37:800-807,2001 .$ g% [  V9 G# i  u8 J

3 Q, v# B# U% o1 j, s
. q8 {5 g/ [5 S
12. Dickinson, E,Tuncer R,Nadler E,Boyle P,Alber S,Watkins S,andFord H. NOX, a novel nitric oxide scavenger, reduces bacterial translocation in rats after endotoxin challenge. Am J Physiol Gastrointest Liver Physiol 277:G1281-G1287,1999 .3 ^% i! H7 o9 W: i; t

. o$ S, t4 \1 M$ ^  g- g; |3 g9 S

13. Eiserich, JP,Hristova M,Cross CE,Jones AD,Freeman BA,Halliwell B,andvan der Vliet A. Formation of nitric oxide-derived inflammatory oxidants by myeloperoxidase in neutrophils. Nature 391:393-397,1998 .) ]3 c, e  ^7 }; S
* e" V+ {& G9 g
, f, x$ g% T3 [7 C2 M
0 ]9 F2 g) H7 F
14. Evans, TJ,Buttery LD,Carpenter A,Springall DR,Polak JM,andCohen J. Cytokine-treated human neutrophils contain inducible nitric oxide synthase that produces nitration of ingested bacteria. Proc Natl Acad Sci USA 93:9553-9558,1996 .
/ U2 ~" }! }: f
" S$ W- p( F+ s5 H! }  T# J7 g

15. Fang, FC. Perspectives series: host/pathogen interactions. Mechanisms of nitric oxide-related antimicrobial activity. J Clin Invest 99:2818-2825,1997  .: J6 s0 _/ S3 g/ }! ^
4 g# u& k' t5 y: T; ~/ d& R
( i; ?6 m$ U# Y2 _" D

16. Fukushi, Y,Orikasa S,andKagayama M. An electron microscopic study of the interaction between vesical epithelium and E. coli. Invest Urol 17:61-68,1979  .( b6 E" K; ~: S" k& {3 C9 m$ q& t' S  c
% E4 D/ y. a, y

1 Y' l! |% ~( s0 R9 I
17. Gardner, PR,Costantino G,andSalzman AL. Constitutive and adaptive detoxification of nitric oxide in Escherichia coli. Role of nitric-oxide dioxygenase in the protection of aconitase. J Biol Chem 273:26528-26533,1998 .
0 o) v' Q0 q$ X( g6 a9 F
5 X$ R9 e% g( @! ^3 _
6 k( t. V- H) o0 j9 N
18. Gardner, PR,Gardner AM,Martin LA,andSalzman AL. Nitric oxide dioxygenase: an enzymic function for flavohemoglobin. Proc Natl Acad Sci USA 95:10378-10383,1998 .
7 Q* G5 N% p% B
0 {& q! z) [9 T/ a: |
6 ?) \7 M* i$ Z: T/ M# v. ]! H
19. Gomes, MS,Florido M,Pais TF,andAppelberg R. Improved clearance of Mycobacterium avium upon disruption of the inducible nitric oxide synthase gene. J Immunol 162:6734-6739,1999 .

8 n5 _. C9 c0 V+ E  Z. q/ L
& U/ ?1 {5 G$ C/ C, T# T! E$ {( B) a
20. Green, LC,Wagner DA,Glogowski J,Skipper PL,Wishnok JS,andTannenbaum SR. Analysis of nitrate, nitrite, and [ 15 N]nitrate in biological fluids. Anal Biochem 126:131-138,1982  .

! P2 D1 ^4 D; e

21. Greenberg, SS,Ouyang J,Zhao X,andGiles TD. Human and rat neutrophils constitutively express neural nitric oxide synthase mRNA. Nitric Oxide 2:203-212,1998  .
; {7 B5 x: G( e4 I+ u

22. Hagberg, L,Engberg I,Freter R,Lam J,Olling S,andSvanborg Eden C. Ascending, unobstructed urinary tract infection in mice caused by pyelonephritogenic Escherichia coli of human origin. Infect Immun 40:273-283,1983 .
6 O5 C6 r- w4 F' {, k2 y! e* x5 {3 \

23. Haraoka, M,Hang L,Frendeus B,Godaly G,Burdick M,Strieter R,andSvanborg C. Neutrophil recruitment and resistance to urinary tract infection. J Infect Dis 180:1220-1229,1999  .

5 S( h$ D: M/ ]9 J+ O6 U7 F% @
24. Hazen, S.L,Zhang R,Shen Z,Wu W,Podrez EA,MacPherson JC,Schmitt D,Mitra SN,Mukhopadhyay C,Chen Y,Cohen PA,Hoff HF,andAbu-Soud HM. Formation of nitric oxide-derived oxidants by myeloperoxidase in monocytes: pathways for monocyte-mediated protein nitration and lipid peroxidation in vivo. Circ Res 85:950-958,1999 .9 n5 H$ L* R! ?3 U$ W; i
9 D6 d. V! d; _( ?+ h! h& q* g7 n

25. Heimberger, A,andEisenstark A. Compartmentalization of catalases in Escherichia coli. Biochem Biophys Res Commun 154:392-397,1988  .

6 _# n9 H4 M+ u" X  r+ P: |2 K

26. Hopkins, WJ,Gendron-Fitzpatrick A,Balish E,andUehling DT. Time course and host responses to Escherichia coli urinary tract infection in genetically distinct mouse strains. Infect Immun 66:2798-2802,1998 .% G& g9 l, W- V/ E, j
; C5 n) W* ^6 X# i: M' V

. Z7 K; O. ^1 k6 d
27. Kabore, AF,Simard M,andBergeron MG. Local production of inflammatory mediators in an experimental model of acute obstructive pyelonephritis. J Infect Dis 179:1162-1172,1999  ." I& R; T* W' j' @- _3 Z0 l

; ^( L9 n' \6 h& \2 L) N: k, S

28. Kooy, NW,Royall JA,Ye YZ,Kelly DR,andBeckman JS. Evidence for in vivo peroxynitrite production in human acute lung injury. Am J Respir Crit Care Med 151:1250-1254,1995 .

' g; z) R9 }0 E/ \$ `* k/ o

29. Laubach, VE,Shesely EG,Smithies O,andSherman PA. Mice lacking inducible nitric oxide synthase are not resistant to lipopolysaccharide-induced death. Proc Natl Acad Sci USA 92:10688-10692,1995 .

8 M2 e& y2 l# c% B+ C1 S3 B9 p; g$ S
9 w7 J) I4 U2 C
30. Lundberg, JO,Ehren I,Jansson O,Adolfsson J,Lundberg JM,Weitzberg E,Alving K,andWiklund NP. Elevated nitric oxide in the urinary bladder in infectious and noninfectious cystitis. Urology 48:700-702,1996  .

31. MacLean, A,Wei XQ,Huang FP,Al-Alem UA,Chan WL,andLiew FY. Mice lacking inducible nitric-oxide synthase are more susceptible to herpes simplex virus infection despite enhanced Th1 cell responses. J Gen Virol 79:825-830,1998 .

0 P# _4 f% y: h8 a$ F3 x
; Z( r# O% M% V+ T
32. Marild, S,Jodal U,Orskov I,Orskov F,andSvanborg Eden C. Special virulence of the Escherichia coli O1:K1:H7 clone in acute pyelonephritis. J Pediatr 115:40-45,1989  .
5 Q/ S, S0 ?1 q: @# T

$ p9 v# A1 c4 ~: N, c  u1 H5 ^
33. McInnes, IB,Leung B,Wei XQ,Gemmell CC,andLiew FY. Septic arthritis following Staphylococcus aureus infection in mice lacking inducible nitric oxide synthase. J Immunol 160:308-315,1998 ., ?2 @5 h" o8 v+ H: b
' e% C8 W5 Q6 h

34. McTaggart, LA,Rigby RC,andElliott TS. The pathogenesis of urinary tract infections associated with Escherichia coli, Staphylococcus saprophyticus and S. epidermidis. J Med Microbiol 32:135-141,1990 .
3 m) H) J! C% k" W; y; ~5 @- z6 R
4 U7 d! S- C! Z' Q+ m& L6 G8 h6 T
6 L7 F/ \. \, V
35. Messmer, UK,Ankarcrona M,Nicotera P,andBrune B. p53 expression in nitric oxide-induced apoptosis. FEBS Lett 355:23-26,1994  .
; i* H1 ^# e  w1 `! L! @
- d( ]; }! I" {: k' z/ U, s

36. Mulvey, MA,Lopez-Boado YS,Wilson CL,Roth R,Parks WC,Heuser J,andHultgren SJ. Induction and evasion of host defenses by type 1-piliated uropathogenic Escherichia coli. Science 282:1494-1497,1998 .. d- j& E! g7 H' b- b/ P
7 l" \3 P% @  F/ D. K( @

5 Y! ~& w# m. d
37. Mulvey, MA,Schilling JD,andHultgren SJ. Establishment of a persistent Escherichia coli reservoir during the acute phase of a bladder infection. Infect Immun 69:4572-4579,2001 .
( W  H9 f3 q) G  _& J  y7 m

38. Mulvey, MA,Schilling JD,Martinez JJ,andHultgren SJ. Bad bugs and beleaguered bladders: interplay between uropathogenic Escherichia coli and innate host defenses. Proc Natl Acad Sci USA 97:8829-8835,2000 .
9 c; K8 A: L. j1 a$ r! r5 j4 ^' j" d
2 M0 H. {  c, p

39. Nowicki, B,Singhal J,Fang L,Nowicki S,andYallampalli C. Inverse relationship between severity of experimental pyelonephritis and nitric oxide production in C3H/HeJ mice. Infect Immun 67:2421-2427,1999 .

6 v4 g7 ?* C, d: k9 f  L
40. Nunoshiba, T,DeRojas-Walker T,Tannenbaum SR,andDemple B. Roles of nitric oxide in inducible resistance of Escherichia coli to activated murine macrophages. Infect Immun 63:794-798,1995 .3 ^7 j3 `. }* K# p  C

0 h! ^/ J; \% Y
; l$ w/ `, t* K0 H
41. Poljakovic, M,Svensson ML,Svanborg C,Johansson K,Larsson B,andPersson K. Escherichia coli -induced inducible nitric oxide synthase and cyclooxygenase expression in the mouse bladder and kidney. Kidney Int 59:893-904,2001  .& k) c9 {1 l; @6 q
% r. Y: n% u  U) i' a
5 m% P) g) F9 @2 A

42. Smith, SD,Wheeler MA,andWeiss RM. Nitric oxide synthase: an endogenous source of elevated nitrite in infected urine. Kidney Int 45:586-591,1994  ." ~- K0 s1 i! k! j" A0 W
$ F0 {8 g! {$ O# ]

. d: Q% K3 W4 l( n# h
43. Sorrells, DL,Friend C,Koltuksuz U,Courcoulas A,Boyle P,Garrett M,Watkins S,Rowe MI,andFord HR. Inhibition of nitric oxide with aminoguanidine reduces bacterial translocation after endotoxin challenge in vivo. Arch Surg 131:1155-1163,1996 .( d6 p/ F5 `$ ]2 B' k8 J' J

1 E8 C  K0 T- p* X
8 y. U0 Z3 `% _# N6 A. U: Z& D
44. Tsaneva, IR,andWeiss B. soxR, a locus governing a superoxide response regulon in Escherichia coli K-12. J Bacteriol 172:4197-4205,1990 .0 j+ w: W  Y, Y( K  p
* K" T& E/ z# ^" j! J
+ l9 y, ]* Z/ n" {* [
8 v6 J+ Z* C0 I& f: k
45. Van der Vliet, A,Eiserich JP,Halliwell B,andCross CE. Formation of reactive nitrogen species during peroxidase-catalyzed oxidation of nitrite. A potential additional mechanism of nitric oxide-dependent toxicity. J Biol Chem 272:7617-7625,1997 .5 I: `0 Y; T$ x5 a3 N3 C0 u

% z6 `" F; U- r. K8 f+ m! j! M: T
46. Wheeler, MA,Smith SD,Garcia-Cardena G,Nathan CF,Weiss RM,andSessa WC. Bacterial infection induces nitric oxide synthase in human neutrophils. J Clin Invest 99:110-116,1997  .

2 T* s9 T$ ?- q6 [# U
2 R! c2 C' K/ N$ ~  J$ l1 p, J* S
47. Winterbourn, CC,Vissers MC,andKettle AJ. Myeloperoxidase. Curr Opin Hematol 7:53-58,2000  .
% X6 M0 O5 D* d" o; S" }, c7 l2 T

48. Xia, Y,Dawson VL,Dawson TM,Snyder SH,andZweier JL. Nitric oxide synthase generates superoxide and nitric oxide in arginine-depleted cells leading to peroxynitrite-mediated cellular injury. Proc Natl Acad Sci USA 93:6770-6774,1996 .

4 R( X# N* n. s* U7 A$ p
9 m5 O; M& Z1 M' b. o
49. Zhu, L,Gunn C,andBeckman JS. Bactericidal activity of peroxynitrite. Arch Biochem Biophys 298:452-457,1992  .7 \0 ]' c) A0 _! o5 H

; a1 e7 g  |. r0 |3 ^' y
! Y2 p+ A+ h1 P
50. Zingarelli, B,Virag L,Szabo A,Cuzzocrea S,Salzman AL,andSzabo C. Oxidation, tyrosine nitration and cytostasis induction in the absence of inducible nitric oxide synthase. Int J Mol Med 1:787-795,1998  .

收藏0 分享0 顶0 踩0