- Email: [email protected]
CYTOKINE GENE EXPRESSION DURING EXPERIMENTAL ESCHERICHIA COLI PYELONEPHRITIS IN MICE
0022-5347/97/15&1-1576$03.00/0
Vol. 158, 1576-1580, October 1997 Printed in U.S.A.
THEJOURNAL OF UROLOGY Copyright Q 1997 by AMERIC~V UROUXXCAL ASS(X'L\TIOS, Iw.
CYTOKINE GENE EXPRESSION DURING EXPERIMENTAL ESCHERICHIA COLI PYELONEPHRITIS IN MICE ADLI KHALIL, ANNELIE BRAUNER, MOIZ BAKHIET, LARS G. BURMAN, GEORG JAREMKO, BENGT WRETLIND AND W E L L TULLUS* From the Department of Surgery, Huddinge Hospital, Departments of Bacteriology, Karolinska Hospital, Swedish Institute for I n fectious Disease Control, Danderyd Hospital. Diixsion of Neurology, Huddinge Hospital, Department of Pathology. Karolinska Hospltal and Department of Pediatrrcs, Karolinska Hospital 1st Gorans Hospital, the Karolinska Instrtute, Stockholm, Sweden
ABSTRACT
Purpose: We studied nine inflammatory and immunoregulatory cytokines in acute pyelonephritis and urethral obstruction in mice to better understand the processes underlying kidney inflammation and scarring. Materials and Methods: Experimental acute pyelonephritis was established in Bki NMRI outbred mice by bladder inoculation of Escherichia coli, followed by 6 h urethral obstruction. The numbers of cytokine mRNA expressing cells for interleukin-1 (IL-11, IL-4, IL-6, IL-10, IL-12, tumor necrosis factor a (TNF-a), TNF-P, transforming growth factor /3 (TGF-P) and interferon y (IFN-y) were determined in the kidneys and spleens from the infected, non-infected but obstructed and untouched mice using in situ hybridization with radio-labelled oligonucleotide probes a t 12 h, 48 h and 6 d after release of the urethral obstruction. Results: Kidney cell expression of IL-1, IL-6 and TNF-a mRNA was observed already at 12 h and persisted on day 6 in the infected animals. A significant proinflammatory cytokine response occurred also in the non-infected obstructed animals, albeit later and at lower levels. A marked increase of IL-4, IL-10, TGF-P and IFN-y mRNA producing cells was also found in the kidneys of these two groups again with higher levels in the infected animals. Very high numbers of splenocytes expressing mRNA for IL-1 were observed especially in the infected animals. A high proportion of splenocytes further expressed mRNA for IL-6, TNF-a, IL-4, IL-10, IFN-y and TGF-0, again with highest numbers in the infected group of animals. Conclusions: The present study extends previous knowledge about the local and systemic cytokine expression profiles during acute pyelonephritis and after urethral obstruction. Of particular interest was the marked kidney cell expression of mRNA for TGF-P, presumed to be important both for obstructive and post-infectious renal scarring. K E Y WORDS:experimental acute pyelonephritis, obstruction, cytokines, interleukins, TGF-P
Acute pyelonephritis is one of the most important bacterial infections in childhood, which in 10-40% of cases leads t o renal scarring.' Post-infectious renal damage is still today in many countries one of the major causes of chronic renal failure in children and early adulthood.2 The risk for kidney damage is greatly increased in children with urinary obstruction. The major hypothesis of the events leading to renal scarring has been that bacterial products e.g. endotoxin (lipopolysaccharide, LPS) stimulate the release of proinflammatory cytokines, which initiates a n inflammatory response including chemotaxis leading to the extravasation of polymorphonuclear neutrophil leukocytes.3 These cells release toxic products, e.g. free oxygen radicals and lysozymes that seem to be responsible for the tissue damage.4 We have indirectly studied the inflammation in the kidneys by following the excretion of cytokines into the urine of infants and children with acute pyelonephritis.5-6 Only children who had detectable urinary levels of interleukin-6 (IL-6) during the acute infection were at a risk of developing renal scarring at the one year f ~ l l o w - u pUsing .~ ELISA technique we have also found high kidney tissue levels of IL-1 and IL-6 in mice with experimental ascending E. coli pyelonephritis already at 30 min. after bladder inocu1ation.H Accepted for publication May 12, 1997. * Requests for reprints: Department of Pediatrics, St. Gorans Children's Hospital, S-112 81 Stockholm, Sweden. Supported by grant B96-16X-08302-09Bfrom the Swedish Medical Research Council and from Stiftelsen Svenska Frimurarebarnhuset.
Detailed knowledge of the cytokine profiles within the kidneys is important for a further understanding of renal scarring following acute pyelonephritis and urinary obstruction. Studies of the local and systemic mRNA expression for interferon y (IFN-?), IL-1, IL-4, IL-6, IL-10, IL-12, transforming growth factor p (TGF-P), tumor necrosis factor a (TNF-a), and tumor necrosis factor p (TNF-p), in mice with experimental acute pyelonephritis and in mice subjected to 6 h urethral obstruction without infection are presented. MATERIALS A N D METHODS
Animals and experimental infection. Female Bki NMRI outbred mice (20-30 gm.) were anaesthetized with diazepam 10 pl./gm. body weight and hypnorm 30 pl./gm. body weight. A sterile 25 mm. polyethene catheter (diameter inside, 0.28 mm.; outside, 0.61 mm. Clay Adams Parsippany, N.J.) was connected to a 30-gauge needle attached to a tuberculin syringe containing the bacterial suspension stained with india ink, and was gently inserted into the bladder. 50 p1. (10" CFU/ml., 5 x lo7 bacteria) was injected per mouse and the catheter was then removed. Immediately after bacterial challenge, temporary urethral obstruction was accomplished by coating the urethral meatus with collodium (sves 46, Swedish Pharmacopeia)." During urethral obstruction, mice were caged individually on absorbent paper t o document the integrity of the collodium seal by the absence of urination. After 6 h the collodium film was softened with acetone and
1576
1577
CYTOKINES IN ACUTE PYELONEPHRITIS IN MICE
removed from the urethral meatus. Another group of mice received 50 p1. of sterile normal saline and was obstructed in the, >;inie way for 6 h. The mice were then caged in groups an(] ,sacrificed at 12 h , 48 h, and 6 d after release of obstruction. Each kidney was removed aseptically, frozen immediatcly in isopentone and stored at -7OC. The spleens were di<:.;fscted and stored in tissue culture medium at 8C. The stu(]>- was approved by the ethical committee for animal r e w a r c h in Stockholm. &c.tt>ria. E. coli CFT 073 was cultured from the blood of a patit,nt with acute pyelonephritis. It expresses P fimbriae and hemolysin and induces pyelonephritis in experimentally challenged normal mice (kindly provided by David E. Johnson, VA Medical Center, Baltimore, Maryland, USA). Prc,puration of rnononuclear cell suspension. Each spleen WLIS gently crushed through a stainless steel meshwork. The cells released were washed once in Iscove's modified Dultiecco's medium (Flow Lab, Irvine U.K.) Erythrocytes in thr cell pellets were lysed by adding 2 ml. cold sterile water for : j O sec., followed by addition of 2 ml. 2.7% saline. The splrnocytes were then washed in medium twice and rediluted to oI)tain a cell concentration of 5 x 1O"/ml. l'rc*pciration of tissue sections. The kidneys were cut along thr longitudinal axis in 7 pm. thick sections in the cryotome u n t i l the middle of the kidney where 10 sections were anal y c d The sections were placed on probeon, microscope slide (Fisher,Biotech), and stored at -2OC until hybridization. I)vtc,c,tion of cytokine mRNA expression by in situ hybrid~ z c i t i o n .In situ hybridization was performed as described by Dngtarlind et al.1" for tissue sections and modified for cell suspensions. Synthetic oligonucleotide probes (Scandinavian ( ; t w Synthesis AB,Koping, Sweden) were labelled using 35S adenosine-5,-a-(thio)-triphosphate(New England NuCambridge, MA) with terminal deoxynucleotidyl transferasc (Amersham, Little Chalfont, U.K.). To increase the scnsitivity of the method, a mixture of four different approximately 48 base pairs long oligonucleotide probes were used f h cach cytokine. The oligonucleotide sequences were obtained from GenBank and probes were designed using Mac vector software (table 1). Emulsion autoradiography was pr,rformed and slides were coded. Cells expressing more than 15 gpxins over their cytoplasm were counted in dark field Iiiicroscopy a t loox magnification. Accuracy was checked at l%her magnification and light microscopy as described pre\ . ~ ~ ~ ~For d y .each y cytokine the results were expressed as mean number of labelled cells per 10" spleen cells or per 100 mn.') tissue section. The tissue section areas were measured b image analysis (Seescan-Image Analysis System, Cnl11bridge, U.K.). kitis is tical methods. The Mann-Whitney U-test and Spear111;111 Rank Correlation test were used. ~
RESULTS
(:\.tokine ni RNA expression in kidneys. A statistically sig11ificclntincrease ofthe number of cells expressing mRNA for proinflammatory cytokines IL-1 and TNF-a was found ;11r~adyat 12 h in the kidneys of the infected animals both cwpared to the obstructed non-infected animals and the ~llltouchedanimals ( p 10.05, respectively; fig. 1). The inC ~ ( ' ; ~ S C of IL-6 and TNF-P reached statistical significance at i s h and 6 d, respectively. The non-infected but obstructed kl(l11eys showed increased numbers of cells expressing IL-1, IIA and TNF-(UmRNA compared to the untouched controls h t at significantly lower levels than in the infected kidneys IfiK. 1). &O TGF-P, IL-4 and IL-10 mRNA expression was signifIcmtly increased in the infected kidneys compared to the other two groups of animals (p c0.05). High levels of IL-10 \\.tire recorded at 48 h while IL-4 and TGF-P peaked at 6 d If%. 1). Again, the non-infected obstructed animals had in-
Probes used for detection ofcytokine mRNA expression by in situ hybridization Probe
GenBank acc #
IFN-y
M29315 M29316 M29317
IL-4
X16058
TGF-p
X02812
IL-10
M37897
IL-12
M86771 M86672
TNF-a
DO0478
TNF-/3
YO0137
IL-lP
M98820
IL-6
M26744
Bases 298-345 (exon 1) 80-125 (exon 2) 303-350 (exon 3) 18C227 iexon 4 ) 83-130 (exon 1) 209-256 (exon 2 ) 270-317 (exon 3) 331-378 (exon 3) 1363-1410 lexon 1) 1457-1504 (exon 2 ) 1'7661813 (exon 3) 1953-2000 lexon 4 ) 79-126 (exon 1) 134-181 (exon 2) 184-231 (exon 3) 4 0 2 4 9 (exon 4) 147-194 (exon 1) 595-642 iexon2) 190-238 (exon 3) 7 0 6 7 5 3 (exon 4) 913-960 lexon 1) 2059-2106 (exon 2) 2152-2199 (exon 3) 23162363 (exon 4 ) 118-165 iexon 1) 202-249 (exon 2) 342-389 (exon 31 502-549 (exon 4 ) 6 3 9 4 8 6 lexon 1) 569-616 (exon 2) 278-325 (exon 3) 295-342 (exon 4 ) 139-187 (exon 1) 180-223 (exon 2)
Reference (11)
(121
(13)
I141
(15)
i16)
(17.18)
Feeser, W. et a1 1992 Unpublished (19)
creased numbers of kidney tissue cells expressing mRNA for TGF-P, IL-4 and IL-10 compared to untouched controls but at lower levels than infected animals (p <0.05, respectively). IFN-y showed a biphasic response in the infected kidneys with significantly higher number of cells at 12 h and 6 d but not at 48 h compared to the untouched animals (p c0.05, respectively; fig. 1).The IFN-y response was not statistically different between the infected and the non-infected kidneys. The non-infected animals showed a successive increase in IFN--y mRNA expressing cells peaking at 6 d. IL-12 showed significantly increased numbers of mRNA expressing cells at all time points after obstruction in the infected but not in the non-infected kidneys (p c0.05, respectively). Significant correlations were found between the number of cells expressing IL-1 and TGF-p, IL-4 and IL-10 mRNA in the infected kidneys (R, = 0.54,0.54 and 0.55 respectively; p <0.001 for all). These three cytokines also showed significant correlations between one another, R, = 0.47-0.77, p c0.05, p <0.001 and p <0.001). Systemic cytokine mRNA expression in splenocytes. The proportion of splenocytes expressing IL-1 mRNA showed a very rapid increase in the infected and at significantly lower levels also in the non-infected but obstructed mice, while only a few cells expressed IL-1 mRNA in the untouched kidneys ( p <0.05 for all tests; fig. 2). The increase in IL-1 persisted over the whole study period of 6 days. Also IL-6 and TNF-a increased rapidly in both groups of obstructed mice. In the infected mice TNF-a peaked already at 12 h and persisted until 6 d while IL-6 showed somewhat lower levels at all three time-points (p <0.05 compared to the untouched controls except for IL-6 at 48 h; fig. 2). A similar pattern was found also in the non-infected but obstructed group albeit at significantly lower levels (p c0.05 for IL-6 at 48 h and for TNF-a at 12 and 48 h). No splenocytes were found to express mRNA for TNF-p. Significantly increased numbers of cells producing TGF-p, IL-4, IL-10, IFN-y and IL-12 were found in the spleens of the infected mice at all timepoints except at 12 h for TGF-B and
CYTOKINES IN ACUTE PYELONEPHRITIS IN MICE
1578
1'L-4
8o
z
.IL-6
1
A0
80/TNF-a
ITNF-$
I
60
40
2o 0
1
1
ITOF+
1
uu Oh
1zk
48h
6 drro
Oh
12h
48h
8 dare
Oh
12h
48h
6 days
FIG.1. Cytokine mRNA expression in kidney sections. Filled bars represent animals infected with E. coli and subjected to urethral obstruction for 6 h (means and standard deviations for six to ten mice at each time point), hatched and unfilled bars non-infected animals (means and standard deviations of four mice at each time point) subjected to urethral obstruction for 6 h.
6 d for IL-12. Surprisingly, similarly increased levels were found also in the non-infected group compared to the untouched animals until 6 d after the release of the obstruction (fig. 2). The number of splenocytes expressing mRNA to TGF-p, IL-4, IL-10, IFN-y was significantly higher in the infected animals compared to the non-infected. DISCUSSION
In this study we extend previous observati~ns~.~O and describe the cytokine profile during experimental pyelonephritis in mice. We found rapid expression of mRNA for almost all cytokines, both the proinflammatory cytokines IL-1, IL-6 and TNF-a, and the immunomodulating and often down regulating cytokines (IL-4, IL-10, TGF-p, IFN-y and IL-12). Interestingly, also non-infected animals subjected to 6 h urethral obstruction showed increased mRNA expression for these cytokines except IL-12 albeit mostly at significantly lower levels. IL-la, IL-6 and IL-8 have previously been found in markedly increased concentrations in the urine of children and adult patients with acute p y e l ~ n e p h r i t i s .These ~ . ~ cytokines are thought t o be the prime initiators of the immune response inducing the further cascade of cytokines, which results in the events that later will cause kidney scarring. In our clinical studies we found that renal scarring one year post-infection occurred only in children with increased IL-6 in their acute urine samples." A most interesting new finding in our present study was the high number of renal cells expressing TGFp mRNA. This is potentially of great clinical interest. TGF-p is known to be an immunosuppressive agent, but also seems to be a key cytokine that initiates and terminates tissue repair."' Sus-
tained production of TGF-P in the kidneys has been associated with both glomerulosclerosis and interstitial fibrosis.22 Previously, longstanding, ureteral obstruction has been shown to increase the expression of TGF-0 and cause increased tubular apoptosis.23 In obstructive nephropathy TGF-p seems to be induced by the reninfangiotensin axis and to be downregulated by ACE inhibitors and losartan.24 The experiments presented here show that also E. coli pyelonephritis contributes to TGFp expression. Thus, it is possible that e.g. losartan could reduce the risk of kidney scarring both in obstruction and in acute pyelonephritis. The marked and sustained increase in the number of kidney cells expressing IL-10 mRNA is also of interest. IL-10 is a potent downregulator of the T h l cytokines and of the proinflammatory cytokines and has been shown to protect from lethality in experimental mice endotoxemia.25 IL-10 has also been found in high concentrations in plasma during human septicaemia, but without being able to control the massive production of proinflammatory mediators during septic shock.26 IFN-y is a pleiotropic TH1 cytokine with important biological effects. High levels of cells expressing mRNA to this cytokine were recorded in this study. Interesting in the present context is the reduced mortality in Gram-negative septic shock in mice by antibodies blocking IFN-7." We found even higher cytokine levels in the splenocytes than in the kidneys, especially for IL-1. This systemic response could possibly be due to a transient bacteremia in the infected animals. Furthermore, peripheral lymphocytes can be antigenically activated by circulation through gut associated lymphoid tissue e.g. in patients with shigellosis.28 Thus passage to the spleens of lymphocytes stimulated in the kid-
CYTOKINES IN ACUTE PYELONEPHRITIS IN MICE
80 -
IL- 10
60
-
8o
ITNF-u
1579
-1L-12
i
ITNF-
] TGF-B
IFN-1
J
20
-
Oh
12h
48h
6 drys
Oh
12h
48h
6 drys
Oh
1Zh
48h
6 days
Flc:. 2. Cytokine mRNA expression in spleen mononuclear cells. Filled bars represent animals infected with E. coli and subjected to urethral obstruction for 6 h, hatched and unfilled bars non-infected animals subjected to urethral obstruction for 6 h.
neys could also contribute to the high cytokine mRNA levels found in t h e spleen. None of these hypotheses would, however, not explain the increased cytokine levels found also in the obstructed but non-infected animals. Whether the spleen responded to the urinary tract obstruction per se (via circulating cytokines?) or to t h e general stress caused by inability to urinate for 6 h and our manipulations of the animals remains obscure. Our present study contributes to the increased knowledge about cytokine induction within the kidneys during acute PYehephritis and after urethral obstruction. The potentially most interesting finding from a clinical point of view is the high levels of TGF+ niRNA expressing cells. TGF-p in other studies appears to be closely associated with renal scarring and potentially amenable to therapeutic interven!on. Increased knowledge of the complex cytokine response 1s important to define targets for future potential immunomodulatory therapies aiming a t reducting and even prevent1% renal scarring. Such efforts have until now often been crude and without detailed knowledge of the biological events in the process of t h e scarring.”“ K I.:F E K I.:N (‘ ES
1 . Rushton, H. G.. Majd, M.,Jantausch, B.. Wiederniann. B. L. and
Behian, A. B.: Renal scarring following reflux and nonrellux Pyclonephritis in children: evaluation with w”’Technetium dimrrcaptosuccinic acid (DMSA) scintigmphy. J. Urol.. 147: 1327, 1992. 2. Sirin, A , , Enire, S.. Alpay, H., Nayir, A,, Bilge, I. and Tanman. F.: Etiology of chronic renal failure in Turkish children. Pediatr. Nephrol., 9: 549, 1995. 3. Glauser, M. p., Meylan, p. and Bilk, J.: The inflammatory response a n d tissue damage. The example of renal scars following acute renal infection. Pediatr. Nephrol., 1: 615, 1987.
J . K.. Domingue. G . , Lewis, R. W.. Kaack, B. and Baskin, G.: Immunology of pyelonephritis in the primate model. V. Effect of superoxide dismutase. J. Urol.. 128 1394. 1982. 5. Tullus, K., Fituri, 0..Burman, L. G., Wretlind. B. and Brauner. A,: Interleukin-6 and interleukin-8 in the urine of children with acute pyelonephritis. Pediatr. Nephrol.. 8 280. 1994. 6. Tullus, K., Escobar-Billing, R., Fituri, O., Burman. I,. G., Karlsson, A., Wikstad, I., Wretlind. B. and Braunder. A.: Interleukin-lu and interleukin-1 receptor antagonist in t h e urine of children with acute pyelonephritis and relation to renal scarring. Acta Paediatr.. 85: 158. 1996. 7. Tullus, K., Fituri, 0..Linne. T.. Escobar-Billing, R.. Wikstad. 1.. Karlsson. A.. Burman. L. G.. Wretlind. B. and Briiuner. A.: Urine interleukin-6 and interleukin-8 in children with acute pyelonephritis in relation to DMSA-scintigraphy in t h e acute phase and a t one year follow-up. Pediatr. Radiol.. 24: 513. 1994. 8. Tullus. K., Wang, J., IAI*Y.. Burman, I,. G . a n d Brauner. A,: Interleukin-lo and interleukin 6 in the urine. kidney and bladder of mice inoculated with E. coli. Pediatr. Nrphrol.. 10: 453, 1996. 9. Johnson, D. E.. Russel. R. G.. Lockatell. C. V., Zulty. .I, C. and Warren, J. W.: Urethral obstruction of 6 hours or less causw bacteriuria, bacteremia. and pyelonephritis in mice challenged with “nonuropathogenic” Escherichia coli. Infect. Ininiun.. 61: 3422. 1993. 10. Dagerlind, A.. Bean, K. A. J. a n d Hiikfelt. T.: Sensitive mRNA detection using unfixed tissue: Combined radiactive and nonradioactive in situ hybridization histochemistry. Ilistochemistry. 9 8 39, 1992. 11. Dijkema. R., Van d e r Meide. H. P.. Dubbeld. M.,Wubben. J. and Schellekens, H.: Cloning. expression and purification of rat IFN-y. Meth. Enzymo., 119: 441, 1986. 12. McKnight, A. J.. Barclay, A. N. and Mason, D. W.: Molecular cloning of r a t interleukin 4 cDNA and analysis of the cytokine 4. Roberts, J. A,, Roth, J r .
1580
CYTOKINES IN ACUTE PYELONEPHRITIS IN MICE
repertoire of subsets of CD4+ T cells. Eur. J . Immunol., 21: 22. Yamamoto, T., Noble, N. A., Miller, D. E. and Border, W. A.: Sustained expression of TGF-Pl underlies development of pro1187, 1991. gressive kidney fibrosis. Kidney Int., 4 5 916, 1994. 13. Derynck, R., Jarret, J. A., Chen, E . Y., Eaton, D. H., Bell, J . R., Assoian, R. K., Roberts, A. B.. Sporn, M. B. and Goeddel, D. V.: 23. Kennedy, W. A. 11, Stenberg, A., Lackgren, G., Hensle, T. W. and Sawczuk, I. S.: Renal tubular apoptosis after partial ureteral Human transforming growth factor-p complementary DNA obstruction. J . Urol., 152 658, 1994. sequence and expression in normal and transformed cells. 24. Pimentel, J. L. J r , Sundell, C. L., Wang, S., Kopp, J. B., Monterq Nature, 316 701, 1985. A. and Martinez-Maldonado. M.: Role of angotensin I1 in the 14. Moore, K. W., Vieira, P., Fiorentino, D. F., Trounsteine, M. L., expression and regulation of transforming growth factor p in Khan, T. A. and Mosmann, T. R.: Homology of cytokine synobstructive nephropathy. Kidney Int., 48 1233, 1995. thesis inhibitory factor ( I L 1 0 )to the Epstein Barr Virus gene 25. G r a r d , C., Bruyns, C., Marchant, A.. Abramowicz, D., BCRFI. Science, 248: 1230, 1990. Vandenabeele, P., Delvaux, A., Fiers, W., Goldman. M. and 15. Schoenhaut, D. S.. Chua, A. O., Wolitzdy, A. G., Quinn, P. M., Velu, T.: Interleukin 10 reduces the release of tumor necrosis Dwyer, C. M., McComas, W., Familletti. P. C., Gately, M. K. factor and prevents lethality in experimental endotoxemia. J. and Gubler, U.: Cloning and expression of murine IL-12. J . ImExp. Med., 177:547, 1993. mun., 148 3433, 1992. 16. Shirai, T.. Shimizu, N., Shiojiri, S., Horiguchi, S. and Ito, H.: 26. Gomez Jimenez, J., Martin, M. C., Sauri, R., Segura, R. M., Esteban, F., Ruiz, J. C., Nuvials, X., Boveda, J. L., Peracaula, Cloning and expression in Echerichia coli of the gene for R. and Salgado, A.: Interleukin-10 and the monocyte/ mouse tumor necrosis factor. DNA, 7: 193, 1989. macrophage-induced inflammatory response in septic shock. 17. Gray, P. W., Chen, E., Tang, W. L. and Ruddle, N.: The murine J . Infect. Dis., 171:472, 1995. tumour necrosis factor-beta (lymphotoxin) gene sequence. Nucleic. Acid. Res., 15 3937, 1987. 27. Kohler, J., Heumann, D., Garotta, G., Leroy, D., Bailat, S., Barras, C., Baumgartner, J . D. and Glauser, M. P.: IFN-y 18. Fashena, S. J.. Tang, W. L., Sarr, T. and Ruddle, N. H.: The involvement in the severity of Gram-negative infections in murine lymphotoxin gene promoter. Characterization and mice. J . Immunol., 151: 916, 1993. negative regulation. J . Immunol., 145 177, 1980. 19. Northemann, W., Braciak, T. A., Hattori, M., Lee, F. and Fey, 28. Islam, D., Kumar Bardhan, P., Lindberg, A. A. and Christensson, B.: Shigella infection induces cellular activation G. H.: Structure of the rat interleukin 6 and its expression in of T and B cells and distinct species-related changes in periphmacrophage-derived cells. J . Biol. Chem., 264: 16072, 1989. eral blood lymphocyte subsets during the course of the disease. 20. Rugo, H. S., OHanley, P., Bishop, A. G., Pearce, M. K., Infect. Immun., 63:2941, 1995. Abrahams, J. S., Howard, M. and OGarra, A,: Local cytokine production in a murine model of Escherichia coli pyelonephri- 29. Haraoka, M., Matsumoto, T., Takahashi, K., Kubo, S., Tanaka, M. and Kumazawa, J.: Suppression of renal scarring by predtis. J. Clin. Invest., 8 9 1032, 1992. nisolone combined with ciprofloxacin in ascending pyelone21. Border, W. A. and Noble, N. A,: Transforming growth factor p in phritis in rats. J . Urol., 151: 1078, 1994. tissue fibrosis. N. Eng. J. Med., 331: 1286, 1994.
Vol. 158, 1576-1580, October 1997 Printed in U.S.A.
THEJOURNAL OF UROLOGY Copyright Q 1997 by AMERIC~V UROUXXCAL ASS(X'L\TIOS, Iw.
CYTOKINE GENE EXPRESSION DURING EXPERIMENTAL ESCHERICHIA COLI PYELONEPHRITIS IN MICE ADLI KHALIL, ANNELIE BRAUNER, MOIZ BAKHIET, LARS G. BURMAN, GEORG JAREMKO, BENGT WRETLIND AND W E L L TULLUS* From the Department of Surgery, Huddinge Hospital, Departments of Bacteriology, Karolinska Hospital, Swedish Institute for I n fectious Disease Control, Danderyd Hospital. Diixsion of Neurology, Huddinge Hospital, Department of Pathology. Karolinska Hospltal and Department of Pediatrrcs, Karolinska Hospital 1st Gorans Hospital, the Karolinska Instrtute, Stockholm, Sweden
ABSTRACT
Purpose: We studied nine inflammatory and immunoregulatory cytokines in acute pyelonephritis and urethral obstruction in mice to better understand the processes underlying kidney inflammation and scarring. Materials and Methods: Experimental acute pyelonephritis was established in Bki NMRI outbred mice by bladder inoculation of Escherichia coli, followed by 6 h urethral obstruction. The numbers of cytokine mRNA expressing cells for interleukin-1 (IL-11, IL-4, IL-6, IL-10, IL-12, tumor necrosis factor a (TNF-a), TNF-P, transforming growth factor /3 (TGF-P) and interferon y (IFN-y) were determined in the kidneys and spleens from the infected, non-infected but obstructed and untouched mice using in situ hybridization with radio-labelled oligonucleotide probes a t 12 h, 48 h and 6 d after release of the urethral obstruction. Results: Kidney cell expression of IL-1, IL-6 and TNF-a mRNA was observed already at 12 h and persisted on day 6 in the infected animals. A significant proinflammatory cytokine response occurred also in the non-infected obstructed animals, albeit later and at lower levels. A marked increase of IL-4, IL-10, TGF-P and IFN-y mRNA producing cells was also found in the kidneys of these two groups again with higher levels in the infected animals. Very high numbers of splenocytes expressing mRNA for IL-1 were observed especially in the infected animals. A high proportion of splenocytes further expressed mRNA for IL-6, TNF-a, IL-4, IL-10, IFN-y and TGF-0, again with highest numbers in the infected group of animals. Conclusions: The present study extends previous knowledge about the local and systemic cytokine expression profiles during acute pyelonephritis and after urethral obstruction. Of particular interest was the marked kidney cell expression of mRNA for TGF-P, presumed to be important both for obstructive and post-infectious renal scarring. K E Y WORDS:experimental acute pyelonephritis, obstruction, cytokines, interleukins, TGF-P
Acute pyelonephritis is one of the most important bacterial infections in childhood, which in 10-40% of cases leads t o renal scarring.' Post-infectious renal damage is still today in many countries one of the major causes of chronic renal failure in children and early adulthood.2 The risk for kidney damage is greatly increased in children with urinary obstruction. The major hypothesis of the events leading to renal scarring has been that bacterial products e.g. endotoxin (lipopolysaccharide, LPS) stimulate the release of proinflammatory cytokines, which initiates a n inflammatory response including chemotaxis leading to the extravasation of polymorphonuclear neutrophil leukocytes.3 These cells release toxic products, e.g. free oxygen radicals and lysozymes that seem to be responsible for the tissue damage.4 We have indirectly studied the inflammation in the kidneys by following the excretion of cytokines into the urine of infants and children with acute pyelonephritis.5-6 Only children who had detectable urinary levels of interleukin-6 (IL-6) during the acute infection were at a risk of developing renal scarring at the one year f ~ l l o w - u pUsing .~ ELISA technique we have also found high kidney tissue levels of IL-1 and IL-6 in mice with experimental ascending E. coli pyelonephritis already at 30 min. after bladder inocu1ation.H Accepted for publication May 12, 1997. * Requests for reprints: Department of Pediatrics, St. Gorans Children's Hospital, S-112 81 Stockholm, Sweden. Supported by grant B96-16X-08302-09Bfrom the Swedish Medical Research Council and from Stiftelsen Svenska Frimurarebarnhuset.
Detailed knowledge of the cytokine profiles within the kidneys is important for a further understanding of renal scarring following acute pyelonephritis and urinary obstruction. Studies of the local and systemic mRNA expression for interferon y (IFN-?), IL-1, IL-4, IL-6, IL-10, IL-12, transforming growth factor p (TGF-P), tumor necrosis factor a (TNF-a), and tumor necrosis factor p (TNF-p), in mice with experimental acute pyelonephritis and in mice subjected to 6 h urethral obstruction without infection are presented. MATERIALS A N D METHODS
Animals and experimental infection. Female Bki NMRI outbred mice (20-30 gm.) were anaesthetized with diazepam 10 pl./gm. body weight and hypnorm 30 pl./gm. body weight. A sterile 25 mm. polyethene catheter (diameter inside, 0.28 mm.; outside, 0.61 mm. Clay Adams Parsippany, N.J.) was connected to a 30-gauge needle attached to a tuberculin syringe containing the bacterial suspension stained with india ink, and was gently inserted into the bladder. 50 p1. (10" CFU/ml., 5 x lo7 bacteria) was injected per mouse and the catheter was then removed. Immediately after bacterial challenge, temporary urethral obstruction was accomplished by coating the urethral meatus with collodium (sves 46, Swedish Pharmacopeia)." During urethral obstruction, mice were caged individually on absorbent paper t o document the integrity of the collodium seal by the absence of urination. After 6 h the collodium film was softened with acetone and
1576
1577
CYTOKINES IN ACUTE PYELONEPHRITIS IN MICE
removed from the urethral meatus. Another group of mice received 50 p1. of sterile normal saline and was obstructed in the, >;inie way for 6 h. The mice were then caged in groups an(] ,sacrificed at 12 h , 48 h, and 6 d after release of obstruction. Each kidney was removed aseptically, frozen immediatcly in isopentone and stored at -7OC. The spleens were di<:.;fscted and stored in tissue culture medium at 8C. The stu(]>- was approved by the ethical committee for animal r e w a r c h in Stockholm. &c.tt>ria. E. coli CFT 073 was cultured from the blood of a patit,nt with acute pyelonephritis. It expresses P fimbriae and hemolysin and induces pyelonephritis in experimentally challenged normal mice (kindly provided by David E. Johnson, VA Medical Center, Baltimore, Maryland, USA). Prc,puration of rnononuclear cell suspension. Each spleen WLIS gently crushed through a stainless steel meshwork. The cells released were washed once in Iscove's modified Dultiecco's medium (Flow Lab, Irvine U.K.) Erythrocytes in thr cell pellets were lysed by adding 2 ml. cold sterile water for : j O sec., followed by addition of 2 ml. 2.7% saline. The splrnocytes were then washed in medium twice and rediluted to oI)tain a cell concentration of 5 x 1O"/ml. l'rc*pciration of tissue sections. The kidneys were cut along thr longitudinal axis in 7 pm. thick sections in the cryotome u n t i l the middle of the kidney where 10 sections were anal y c d The sections were placed on probeon, microscope slide (Fisher,Biotech), and stored at -2OC until hybridization. I)vtc,c,tion of cytokine mRNA expression by in situ hybrid~ z c i t i o n .In situ hybridization was performed as described by Dngtarlind et al.1" for tissue sections and modified for cell suspensions. Synthetic oligonucleotide probes (Scandinavian ( ; t w Synthesis AB,Koping, Sweden) were labelled using 35S adenosine-5,-a-(thio)-triphosphate(New England NuCambridge, MA) with terminal deoxynucleotidyl transferasc (Amersham, Little Chalfont, U.K.). To increase the scnsitivity of the method, a mixture of four different approximately 48 base pairs long oligonucleotide probes were used f h cach cytokine. The oligonucleotide sequences were obtained from GenBank and probes were designed using Mac vector software (table 1). Emulsion autoradiography was pr,rformed and slides were coded. Cells expressing more than 15 gpxins over their cytoplasm were counted in dark field Iiiicroscopy a t loox magnification. Accuracy was checked at l%her magnification and light microscopy as described pre\ . ~ ~ ~ ~For d y .each y cytokine the results were expressed as mean number of labelled cells per 10" spleen cells or per 100 mn.') tissue section. The tissue section areas were measured b image analysis (Seescan-Image Analysis System, Cnl11bridge, U.K.). kitis is tical methods. The Mann-Whitney U-test and Spear111;111 Rank Correlation test were used. ~
RESULTS
(:\.tokine ni RNA expression in kidneys. A statistically sig11ificclntincrease ofthe number of cells expressing mRNA for proinflammatory cytokines IL-1 and TNF-a was found ;11r~adyat 12 h in the kidneys of the infected animals both cwpared to the obstructed non-infected animals and the ~llltouchedanimals ( p 10.05, respectively; fig. 1). The inC ~ ( ' ; ~ S C of IL-6 and TNF-P reached statistical significance at i s h and 6 d, respectively. The non-infected but obstructed kl(l11eys showed increased numbers of cells expressing IL-1, IIA and TNF-(UmRNA compared to the untouched controls h t at significantly lower levels than in the infected kidneys IfiK. 1). &O TGF-P, IL-4 and IL-10 mRNA expression was signifIcmtly increased in the infected kidneys compared to the other two groups of animals (p c0.05). High levels of IL-10 \\.tire recorded at 48 h while IL-4 and TGF-P peaked at 6 d If%. 1). Again, the non-infected obstructed animals had in-
Probes used for detection ofcytokine mRNA expression by in situ hybridization Probe
GenBank acc #
IFN-y
M29315 M29316 M29317
IL-4
X16058
TGF-p
X02812
IL-10
M37897
IL-12
M86771 M86672
TNF-a
DO0478
TNF-/3
YO0137
IL-lP
M98820
IL-6
M26744
Bases 298-345 (exon 1) 80-125 (exon 2) 303-350 (exon 3) 18C227 iexon 4 ) 83-130 (exon 1) 209-256 (exon 2 ) 270-317 (exon 3) 331-378 (exon 3) 1363-1410 lexon 1) 1457-1504 (exon 2 ) 1'7661813 (exon 3) 1953-2000 lexon 4 ) 79-126 (exon 1) 134-181 (exon 2) 184-231 (exon 3) 4 0 2 4 9 (exon 4) 147-194 (exon 1) 595-642 iexon2) 190-238 (exon 3) 7 0 6 7 5 3 (exon 4) 913-960 lexon 1) 2059-2106 (exon 2) 2152-2199 (exon 3) 23162363 (exon 4 ) 118-165 iexon 1) 202-249 (exon 2) 342-389 (exon 31 502-549 (exon 4 ) 6 3 9 4 8 6 lexon 1) 569-616 (exon 2) 278-325 (exon 3) 295-342 (exon 4 ) 139-187 (exon 1) 180-223 (exon 2)
Reference (11)
(121
(13)
I141
(15)
i16)
(17.18)
Feeser, W. et a1 1992 Unpublished (19)
creased numbers of kidney tissue cells expressing mRNA for TGF-P, IL-4 and IL-10 compared to untouched controls but at lower levels than infected animals (p <0.05, respectively). IFN-y showed a biphasic response in the infected kidneys with significantly higher number of cells at 12 h and 6 d but not at 48 h compared to the untouched animals (p c0.05, respectively; fig. 1).The IFN-y response was not statistically different between the infected and the non-infected kidneys. The non-infected animals showed a successive increase in IFN--y mRNA expressing cells peaking at 6 d. IL-12 showed significantly increased numbers of mRNA expressing cells at all time points after obstruction in the infected but not in the non-infected kidneys (p c0.05, respectively). Significant correlations were found between the number of cells expressing IL-1 and TGF-p, IL-4 and IL-10 mRNA in the infected kidneys (R, = 0.54,0.54 and 0.55 respectively; p <0.001 for all). These three cytokines also showed significant correlations between one another, R, = 0.47-0.77, p c0.05, p <0.001 and p <0.001). Systemic cytokine mRNA expression in splenocytes. The proportion of splenocytes expressing IL-1 mRNA showed a very rapid increase in the infected and at significantly lower levels also in the non-infected but obstructed mice, while only a few cells expressed IL-1 mRNA in the untouched kidneys ( p <0.05 for all tests; fig. 2). The increase in IL-1 persisted over the whole study period of 6 days. Also IL-6 and TNF-a increased rapidly in both groups of obstructed mice. In the infected mice TNF-a peaked already at 12 h and persisted until 6 d while IL-6 showed somewhat lower levels at all three time-points (p <0.05 compared to the untouched controls except for IL-6 at 48 h; fig. 2). A similar pattern was found also in the non-infected but obstructed group albeit at significantly lower levels (p c0.05 for IL-6 at 48 h and for TNF-a at 12 and 48 h). No splenocytes were found to express mRNA for TNF-p. Significantly increased numbers of cells producing TGF-p, IL-4, IL-10, IFN-y and IL-12 were found in the spleens of the infected mice at all timepoints except at 12 h for TGF-B and
CYTOKINES IN ACUTE PYELONEPHRITIS IN MICE
1578
1'L-4
8o
z
.IL-6
1
A0
80/TNF-a
ITNF-$
I
60
40
2o 0
1
1
ITOF+
1
uu Oh
1zk
48h
6 drro
Oh
12h
48h
8 dare
Oh
12h
48h
6 days
FIG.1. Cytokine mRNA expression in kidney sections. Filled bars represent animals infected with E. coli and subjected to urethral obstruction for 6 h (means and standard deviations for six to ten mice at each time point), hatched and unfilled bars non-infected animals (means and standard deviations of four mice at each time point) subjected to urethral obstruction for 6 h.
6 d for IL-12. Surprisingly, similarly increased levels were found also in the non-infected group compared to the untouched animals until 6 d after the release of the obstruction (fig. 2). The number of splenocytes expressing mRNA to TGF-p, IL-4, IL-10, IFN-y was significantly higher in the infected animals compared to the non-infected. DISCUSSION
In this study we extend previous observati~ns~.~O and describe the cytokine profile during experimental pyelonephritis in mice. We found rapid expression of mRNA for almost all cytokines, both the proinflammatory cytokines IL-1, IL-6 and TNF-a, and the immunomodulating and often down regulating cytokines (IL-4, IL-10, TGF-p, IFN-y and IL-12). Interestingly, also non-infected animals subjected to 6 h urethral obstruction showed increased mRNA expression for these cytokines except IL-12 albeit mostly at significantly lower levels. IL-la, IL-6 and IL-8 have previously been found in markedly increased concentrations in the urine of children and adult patients with acute p y e l ~ n e p h r i t i s .These ~ . ~ cytokines are thought t o be the prime initiators of the immune response inducing the further cascade of cytokines, which results in the events that later will cause kidney scarring. In our clinical studies we found that renal scarring one year post-infection occurred only in children with increased IL-6 in their acute urine samples." A most interesting new finding in our present study was the high number of renal cells expressing TGFp mRNA. This is potentially of great clinical interest. TGF-p is known to be an immunosuppressive agent, but also seems to be a key cytokine that initiates and terminates tissue repair."' Sus-
tained production of TGF-P in the kidneys has been associated with both glomerulosclerosis and interstitial fibrosis.22 Previously, longstanding, ureteral obstruction has been shown to increase the expression of TGF-0 and cause increased tubular apoptosis.23 In obstructive nephropathy TGF-p seems to be induced by the reninfangiotensin axis and to be downregulated by ACE inhibitors and losartan.24 The experiments presented here show that also E. coli pyelonephritis contributes to TGFp expression. Thus, it is possible that e.g. losartan could reduce the risk of kidney scarring both in obstruction and in acute pyelonephritis. The marked and sustained increase in the number of kidney cells expressing IL-10 mRNA is also of interest. IL-10 is a potent downregulator of the T h l cytokines and of the proinflammatory cytokines and has been shown to protect from lethality in experimental mice endotoxemia.25 IL-10 has also been found in high concentrations in plasma during human septicaemia, but without being able to control the massive production of proinflammatory mediators during septic shock.26 IFN-y is a pleiotropic TH1 cytokine with important biological effects. High levels of cells expressing mRNA to this cytokine were recorded in this study. Interesting in the present context is the reduced mortality in Gram-negative septic shock in mice by antibodies blocking IFN-7." We found even higher cytokine levels in the splenocytes than in the kidneys, especially for IL-1. This systemic response could possibly be due to a transient bacteremia in the infected animals. Furthermore, peripheral lymphocytes can be antigenically activated by circulation through gut associated lymphoid tissue e.g. in patients with shigellosis.28 Thus passage to the spleens of lymphocytes stimulated in the kid-
CYTOKINES IN ACUTE PYELONEPHRITIS IN MICE
80 -
IL- 10
60
-
8o
ITNF-u
1579
-1L-12
i
ITNF-
] TGF-B
IFN-1
J
20
-
Oh
12h
48h
6 drys
Oh
12h
48h
6 drys
Oh
1Zh
48h
6 days
Flc:. 2. Cytokine mRNA expression in spleen mononuclear cells. Filled bars represent animals infected with E. coli and subjected to urethral obstruction for 6 h, hatched and unfilled bars non-infected animals subjected to urethral obstruction for 6 h.
neys could also contribute to the high cytokine mRNA levels found in t h e spleen. None of these hypotheses would, however, not explain the increased cytokine levels found also in the obstructed but non-infected animals. Whether the spleen responded to the urinary tract obstruction per se (via circulating cytokines?) or to t h e general stress caused by inability to urinate for 6 h and our manipulations of the animals remains obscure. Our present study contributes to the increased knowledge about cytokine induction within the kidneys during acute PYehephritis and after urethral obstruction. The potentially most interesting finding from a clinical point of view is the high levels of TGF+ niRNA expressing cells. TGF-p in other studies appears to be closely associated with renal scarring and potentially amenable to therapeutic interven!on. Increased knowledge of the complex cytokine response 1s important to define targets for future potential immunomodulatory therapies aiming a t reducting and even prevent1% renal scarring. Such efforts have until now often been crude and without detailed knowledge of the biological events in the process of t h e scarring.”“ K I.:F E K I.:N (‘ ES
1 . Rushton, H. G.. Majd, M.,Jantausch, B.. Wiederniann. B. L. and
Behian, A. B.: Renal scarring following reflux and nonrellux Pyclonephritis in children: evaluation with w”’Technetium dimrrcaptosuccinic acid (DMSA) scintigmphy. J. Urol.. 147: 1327, 1992. 2. Sirin, A , , Enire, S.. Alpay, H., Nayir, A,, Bilge, I. and Tanman. F.: Etiology of chronic renal failure in Turkish children. Pediatr. Nephrol., 9: 549, 1995. 3. Glauser, M. p., Meylan, p. and Bilk, J.: The inflammatory response a n d tissue damage. The example of renal scars following acute renal infection. Pediatr. Nephrol., 1: 615, 1987.
J . K.. Domingue. G . , Lewis, R. W.. Kaack, B. and Baskin, G.: Immunology of pyelonephritis in the primate model. V. Effect of superoxide dismutase. J. Urol.. 128 1394. 1982. 5. Tullus, K., Fituri, 0..Burman, L. G., Wretlind. B. and Brauner. A,: Interleukin-6 and interleukin-8 in the urine of children with acute pyelonephritis. Pediatr. Nephrol.. 8 280. 1994. 6. Tullus, K., Escobar-Billing, R., Fituri, O., Burman. I,. G., Karlsson, A., Wikstad, I., Wretlind. B. and Braunder. A.: Interleukin-lu and interleukin-1 receptor antagonist in t h e urine of children with acute pyelonephritis and relation to renal scarring. Acta Paediatr.. 85: 158. 1996. 7. Tullus, K., Fituri, 0..Linne. T.. Escobar-Billing, R.. Wikstad. 1.. Karlsson. A.. Burman. L. G.. Wretlind. B. and Briiuner. A.: Urine interleukin-6 and interleukin-8 in children with acute pyelonephritis in relation to DMSA-scintigraphy in t h e acute phase and a t one year follow-up. Pediatr. Radiol.. 24: 513. 1994. 8. Tullus. K., Wang, J., IAI*Y.. Burman, I,. G . a n d Brauner. A,: Interleukin-lo and interleukin 6 in the urine. kidney and bladder of mice inoculated with E. coli. Pediatr. Nrphrol.. 10: 453, 1996. 9. Johnson, D. E.. Russel. R. G.. Lockatell. C. V., Zulty. .I, C. and Warren, J. W.: Urethral obstruction of 6 hours or less causw bacteriuria, bacteremia. and pyelonephritis in mice challenged with “nonuropathogenic” Escherichia coli. Infect. Ininiun.. 61: 3422. 1993. 10. Dagerlind, A.. Bean, K. A. J. a n d Hiikfelt. T.: Sensitive mRNA detection using unfixed tissue: Combined radiactive and nonradioactive in situ hybridization histochemistry. Ilistochemistry. 9 8 39, 1992. 11. Dijkema. R., Van d e r Meide. H. P.. Dubbeld. M.,Wubben. J. and Schellekens, H.: Cloning. expression and purification of rat IFN-y. Meth. Enzymo., 119: 441, 1986. 12. McKnight, A. J.. Barclay, A. N. and Mason, D. W.: Molecular cloning of r a t interleukin 4 cDNA and analysis of the cytokine 4. Roberts, J. A,, Roth, J r .
1580
CYTOKINES IN ACUTE PYELONEPHRITIS IN MICE
repertoire of subsets of CD4+ T cells. Eur. J . Immunol., 21: 22. Yamamoto, T., Noble, N. A., Miller, D. E. and Border, W. A.: Sustained expression of TGF-Pl underlies development of pro1187, 1991. gressive kidney fibrosis. Kidney Int., 4 5 916, 1994. 13. Derynck, R., Jarret, J. A., Chen, E . Y., Eaton, D. H., Bell, J . R., Assoian, R. K., Roberts, A. B.. Sporn, M. B. and Goeddel, D. V.: 23. Kennedy, W. A. 11, Stenberg, A., Lackgren, G., Hensle, T. W. and Sawczuk, I. S.: Renal tubular apoptosis after partial ureteral Human transforming growth factor-p complementary DNA obstruction. J . Urol., 152 658, 1994. sequence and expression in normal and transformed cells. 24. Pimentel, J. L. J r , Sundell, C. L., Wang, S., Kopp, J. B., Monterq Nature, 316 701, 1985. A. and Martinez-Maldonado. M.: Role of angotensin I1 in the 14. Moore, K. W., Vieira, P., Fiorentino, D. F., Trounsteine, M. L., expression and regulation of transforming growth factor p in Khan, T. A. and Mosmann, T. R.: Homology of cytokine synobstructive nephropathy. Kidney Int., 48 1233, 1995. thesis inhibitory factor ( I L 1 0 )to the Epstein Barr Virus gene 25. G r a r d , C., Bruyns, C., Marchant, A.. Abramowicz, D., BCRFI. Science, 248: 1230, 1990. Vandenabeele, P., Delvaux, A., Fiers, W., Goldman. M. and 15. Schoenhaut, D. S.. Chua, A. O., Wolitzdy, A. G., Quinn, P. M., Velu, T.: Interleukin 10 reduces the release of tumor necrosis Dwyer, C. M., McComas, W., Familletti. P. C., Gately, M. K. factor and prevents lethality in experimental endotoxemia. J. and Gubler, U.: Cloning and expression of murine IL-12. J . ImExp. Med., 177:547, 1993. mun., 148 3433, 1992. 16. Shirai, T.. Shimizu, N., Shiojiri, S., Horiguchi, S. and Ito, H.: 26. Gomez Jimenez, J., Martin, M. C., Sauri, R., Segura, R. M., Esteban, F., Ruiz, J. C., Nuvials, X., Boveda, J. L., Peracaula, Cloning and expression in Echerichia coli of the gene for R. and Salgado, A.: Interleukin-10 and the monocyte/ mouse tumor necrosis factor. DNA, 7: 193, 1989. macrophage-induced inflammatory response in septic shock. 17. Gray, P. W., Chen, E., Tang, W. L. and Ruddle, N.: The murine J . Infect. Dis., 171:472, 1995. tumour necrosis factor-beta (lymphotoxin) gene sequence. Nucleic. Acid. Res., 15 3937, 1987. 27. Kohler, J., Heumann, D., Garotta, G., Leroy, D., Bailat, S., Barras, C., Baumgartner, J . D. and Glauser, M. P.: IFN-y 18. Fashena, S. J.. Tang, W. L., Sarr, T. and Ruddle, N. H.: The involvement in the severity of Gram-negative infections in murine lymphotoxin gene promoter. Characterization and mice. J . Immunol., 151: 916, 1993. negative regulation. J . Immunol., 145 177, 1980. 19. Northemann, W., Braciak, T. A., Hattori, M., Lee, F. and Fey, 28. Islam, D., Kumar Bardhan, P., Lindberg, A. A. and Christensson, B.: Shigella infection induces cellular activation G. H.: Structure of the rat interleukin 6 and its expression in of T and B cells and distinct species-related changes in periphmacrophage-derived cells. J . Biol. Chem., 264: 16072, 1989. eral blood lymphocyte subsets during the course of the disease. 20. Rugo, H. S., OHanley, P., Bishop, A. G., Pearce, M. K., Infect. Immun., 63:2941, 1995. Abrahams, J. S., Howard, M. and OGarra, A,: Local cytokine production in a murine model of Escherichia coli pyelonephri- 29. Haraoka, M., Matsumoto, T., Takahashi, K., Kubo, S., Tanaka, M. and Kumazawa, J.: Suppression of renal scarring by predtis. J. Clin. Invest., 8 9 1032, 1992. nisolone combined with ciprofloxacin in ascending pyelone21. Border, W. A. and Noble, N. A,: Transforming growth factor p in phritis in rats. J . Urol., 151: 1078, 1994. tissue fibrosis. N. Eng. J. Med., 331: 1286, 1994.