Neutrophil Influx in Response to a Peritoneal Infection with Salmonella Is Delayed in Lipopolysaccharide-Binding Protein or CD14-Deficient Mice This information is current as of September 26, 2021. Kang K. Yang, Brigitte G. Dorner, Ulrike Merkel, Bernard Ryffel, Christine Schütt, Douglas Golenbock, Mason W. Freeman and Robert S. Jack J Immunol 2002; 169:4475-4480; ;

doi: 10.4049/jimmunol.169.8.4475 Downloaded from http://www.jimmunol.org/content/169/8/4475

References This article cites 26 articles, 15 of which you can access for free at: http://www.jimmunol.org/content/169/8/4475.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 26, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2002 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Neutrophil Influx in Response to a Peritoneal Infection with Salmonella Is Delayed in Lipopolysaccharide-Binding Protein or CD14-Deficient Mice1

Kang K. Yang,2* Brigitte G. Dorner,2† Ulrike Merkel,* Bernard Ryffel,‡ Christine Schu¬tt,* Douglas Golenbock,¤ Mason W. Freeman,¶ and Robert S. Jack3*

The induction of an adaptive immune response to a previously unencountered pathogen is a time-consuming process and initially the infection must be held in check by the innate immune system. In the case of an i.p. infection with Salmonella typhimurium, survival requires both CD14 and LPS-binding protein (LBP) which, together with Toll-like receptor 4 and myeloid differentiation protein 2, provide a sensitive means to detect bacterial LPS. In this study, we show that in the first hours after i.p. infection with Salmonella a local inflammatory response is evident and that concomitantly neutrophils flood into the peritoneum. This rapid Downloaded from neutrophil influx is dependent on TNF since it is 1) abolished in TNF KO mice and 2) can be induced by i.p. injection of TNF in uninfected animals. Neutrophil influx is not strictly dependent on the presence of either LBP or CD14. However, in their absence, no local inflammatory response is evident, neutrophil migration is delayed, and the mice succumb to the infection. Using confocal microscopy, we show that the neutrophils which accumulate in CD14 and LBP null mice, albeit with delayed kinetics, are nevertheless fully capable of ingesting the bacteria. We suggest that the short delay in neutrophil influx gives the pathogen a decisive advantage in this infection model. The Journal of Immunology, 2002, 169: 4475Ð4480. http://www.jimmunol.org/

he innate immune system must detect an incipient infec- transferring LPS released from the outer membrane of Gram-neg- tion and hold it in check until adaptive immunity can ative bacteria to CD14 and by so doing to initiate the signal cas- T mount an effective response. Innate immunity is thought cade involving TLR4, myeloid differentiation protein 2, myeloid to detect bacterial infections by use of receptors directed against differentiation factor 88, and NF-␬B, which results in the synthesis structures commonly expressed on microbes but not on our own and release of proinflammatory mediators from macrophages (8). cells (1, 2). One such structure is LPS which is an essential com- The observation that both CD14- and TLR4-deficient mice are ponent of the outer membrane of Gram-negative bacteria and is much more susceptible to a Salmonella infection is in line with the

detected by the high-affinity receptor CD14 (3). Together with the notion that this pathway is a necessary component of innate de- by guest on September 26, 2021 soluble LPS-binding protein (LBP),4 which accelerates the inter- fense against the pathogen (7). Indeed, the requirement for LBP action of LPS with CD14 (4), and the signal-transducing element can be entirely replaced by exogenous application of the proin- Toll-like receptor (TLR) 4 (5), both of which are expressed on the flammatory mediator TNF, suggesting that the essential nonredun- surface of macrophages, this provides a sensitive in vivo mecha- dant function of the LBP in these mice is to initiate an inflamma- nism for the recognition and reaction to LPS. tory response (9). To investigate the involvement of LBP and CD14 in innate de- An inflammatory response involves a complex set of events fense, we have generated LBP- and CD14-deficient mouse lines which include changes in the structure of the cytokine network, and used them to show that both genes are essential for survival in rearrangement of innate immune cell populations, and changes in an i.p. Salmonella infection (6, 7). LBP is thought to function by the activation status of these cells. In many infection systems, neu- trophil activation plays a central role in innate defense. This is certainly the case with an i.p. Salmonella infection since it has *Department of Immunology, Klinikum der Universita¬t Greifswald, Sauerbruch- recently been shown that neutrophil depletion abrogates the strasse, Germany; †Molecular Immunology, Robert Koch-Institute, Berlin, Germany; ‡Centre National de la Recherche Scientifique Institut Transgenose, Orleans, France; mouse’s capacity to mount an adequate defense (10). In this study, ¤Infectious Disease Division, University of Massachusetts, Worcester, MA 01655; we have examined the role of LBP and CD14 in inducing neutro- ¶ and Lipid Metabolism Unit, Massachusetts General Hospital, Harvard Medical phil influx in response to an infection with Salmonella. School, Boston, MA 02114 Received for publication February 25, 2002. Accepted for publication July 31, 2002. Materials and Methods The costs of publication of this article were defrayed in part by the payment of page Salmonella typhimurium growth charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Salmonella enterica var typhimurium (S. typhimurium) (ATCC 14028s) 1 This work was supported by Grant DFG-Ja 729-3 from the Deutsche Forschungs- transfected with green fluorescent protein (GFP) was a gift from Dr. H. gemeinschaft (to R.S.J.), Deutsche Forschungemeinschaft Graduierten Kolleg stipen- Hensel (Max von Pettenkofer Institut, Munich, Germany). Mid-log phase dium DFG-GH 212/3 (to K.K.Y.), and National Institutes of Health Grants DK 50305 cells were collected by centrifugation and washed with PBS. Numbers of and RR14466 (to M.W.F. and D.T.G.). CFU injected were estimated from the OD at 550 nm and checked by 2 K.K.Y. and B.G.D. contributed equally to this work. plating aliquots of the inoculum. Mice were injected i.p. 3 Address correspondence and reprint requests to Dr. Robert S. Jack, Institut fu¬r Mice Immunologie und Transfusionsmedizin, Klinikum der Universita¬t Greifswald, Sauerbruchstra␤e, D-17489 Greifswald, Germany. E-mail address: [email protected] LBP-deficient mice carrying the NRAMP-1R allele derived from strain greifswald.de CBA were as previously described (9). LBP-deficient animals were back- ϩ/Ϫ 4 Abbreviations used in this paper: LBP, LPS-binding protein; TLR, Toll-like recep- crossed to strain CBA for four backcross generations and LBP het- Ϫ Ϫ tor; GFP, green fluorescent protein. erozygotes were then crossed inter se. LBP / male progeny were then

Copyright © 2002 by The American Association of Immunologists, Inc. 0022-1767/02/$02.00 4476 PERITONEAL NEUTROPHIL INFLUX AFTER INFECTION WITH Salmonella

crossed to their LBPϩ/ϩ littermates to generate heterozygotes or to their iodide (Molecular Probes, Eugene, OR) at a final concentration of 0.2 ␮M, LBPϪ/Ϫ littermates to expand the number of LBP-deficient animals. and incubated at room temperature for 15 min. Cells were then washed CD14Ϫ/Ϫ (11) and CD14ϩ/Ϫ animals carrying the CBA-derived once with FACS-PBS, resuspended in 10 ␮l of Citifluor AF1 (PLANO, NRAMP-1R allele were generated in the same way. Genotyping of these available at www.plano-em.com), and stored at 4¡C before analysis. The animals by PCR was as previously described (7). Mice deficient for the labeled cells were transferred to indented glass slides and examined in a TNF gene (12) on a C57BL/6 background were reared in the animal facility Zeiss model LSM510 confocal microscope (Zeiss, Oberkochen, Germany) in Orleans, France. Animals of strain BALB/c, C57BL/6, and CBA were using a ϫ63 water immersion lens. Single cells were analyzed in three purchased from Charles River Breeding Laboratories (Sulzfeld, Germany). dimensions by creating a stack of images at overlapping Z levels. Orthog- Mice were used at 8Ð12 wk of age. onal sections were cut through the Z-stack with the help of the LSM510 image analysis software. Bacterial load in the peritoneum Mice were infected with Salmonella and 24 h later sacrificed by cervical Results dislocation and the peritoneal cavity was washed with 10 ml of cold sterile Changes in the peritoneal neutrophil population after infection ϳ PBS. The preparation of the mice and the peritoneal wash require 5 min. with Salmonella Because of this, control animals infected and immediately processed are equivalent to an i.p. infection of 5Ð10 min. Appropriate dilutions of the We have previously shown that LBP and CD14 are required for peritoneal wash were plated in duplicate. Plates were incubated for 16 h at mice to survive an i.p. infection with Salmonella (6, 7). Since it is 37¡C and colonies were counted. well established that in this infection model neutrophils play an Assay of IL-6 and TNF in the peritoneum of infected animals essential role in innate defense (10), we have asked whether neu- Groups of animals were infected with 5000 CFU of Salmonella and 0.5, trophil influx into the peritoneum following infection is altered in

1.0, and 2.0 h later they were sacrificed. The peritoneal cavity was washed the absence of LBP or CD14. To follow changes in the neutrophil Downloaded from with 1 ml of PBS and the recovered peritoneal wash (ϳ0.5 ml) was cen- population, we have used the mAb RB6-8C5 (13). In an uninfected trifuged at 500 ϫ g to remove cells. The cell-free supernatant was assayed LBPϩ/Ϫ heterozygous mouse, granulocytes (Ly6Ghigh cells) make for IL-6 and TNF by ELISA using kits from BD PharMingen (Heidelberg, up 1% or less of the peritoneal wash cell population. Injection of Germany). The values for the amount of detected cytokines are given as picograms per peritoneum. PBS alone does not cause neutrophil influx (Fig. 1B). However, injection of 5000 CFU of Salmonella results in a rapid influx of Preparation of peritoneal wash cells

cells so that within2hasubstantial fraction of the population of http://www.jimmunol.org/ Groups of six animals were injected i.p. with 100 ␮lofSalmonella sus- peritoneal wash cells is composed of neutrophils and by 24 h up to pended in sterile pyrogen-free saline. Since injection of PBS or saline at 50% of the peritoneal cell population consists of neutrophils. A 4¡C also results in a substantial neutrophil influx into the peritoneum, all typical result is shown in Fig. 1. This is not due to some peculiarity solutions were prewarmed to 37¡C before injection. At the appropriate time ϩ/Ϫ points, animals were sacrificed by cervical dislocation and the peritoneum of our LBP animals since the same result is reproducibly seen was washed with 10 ml of sterile PBS. Recovered cells were pelleted by with BALB/c mice (Fig. 2) as well as for CBA and C57BL/6 centrifugation at 300 ϫ g for 10 min at 4¡C and resuspended in 500 ␮lof animals (data not shown). PBS. Cell number was determined using a counting chamber. Although in The size of the peritoneal wash cell population increases slightly general more cells are recovered from larger animals than from smaller over the course of the first 24 h of an infection initiated with 3000 ones, there is no significant strain-dependent difference between the total number of peritoneal cells recovered from the strains used in this study CFU injected i.p., although the increase is only by about a factor by guest on September 26, 2021 (C57BL/6, BALB/c, and CBA). of 2 (Fig. 2). This relatively small increase in the size of the total peritoneal population is matched by a dramatic increase in the FACS analysis number of neutrophils. In uninfected animals neutrophils make up The fraction of neutrophils in the population of peritoneal wash cells was Ͻ1% of the total peritoneal cell population. Twenty-four hours determined by FACS analysis using FITC-labeled mAb RB6-8C5 (BD after i.p. infection with 3000 CFU of Salmonella, they are ϳ50% PharMingen, Heidelberg, Germany). This Ab detects an epitope on the Ly6G molecule which is highly expressed on neutrophil lineage cells and of the peritoneal cells. Thus, a 2-fold increase in the total perito- to a lower extent on a population of CD8ϩ T cells (13). 7-azo-actinomycin neal cell population reflects a Ͼ50-fold increase in the neutrophil D was used as a nuclear stain to define the living cells and within this gate subpopulation. When the infection is initiated with as few as 30 the Ly6Ghigh cells were counted as neutrophils. In preliminary experi- ments, the identity of the cells flooding into the peritoneum after infection was checked by examining smears of the isolated peritoneal cells. There was no significant difference between the numbers of neutrophils estimated by FACS analysis using mAb RB6-8C5 and by morphological criteria. The fraction of macrophages was determined using FITC-labeled anti-F4/80 (Caltag Laboratories, Burlingame, CA). Statistical comparisons of the re- sponses of the animals were conducted using the Mann-Whitney U test. Confocal microscopy Initial attempts to study phagocytosis of Salmonella by peritoneal wash cells using confocal microscopy were frustrated by the fact that the cyto- spin procedure which we used resulted in a flattening of the neutrophils and macrophages on the slides. We therefore adopted a different approach de- signed to maintain the three-dimensional structure of the cells. Peritoneal wash cells were surface stained in suspension, fixed, and dropped onto indented slides. The staining procedure was as follows: 2 ϫ 105 peritoneal wash cells in 50 ␮l of FACS-PBS (PBS supplemented with 2.5% FCS and ␮ 0.01% NaN3) were incubated on ice with 4 l of unlabeled RB6-8C5 (20 ␮ ␮ ϩ Ϫ g/ml) for 30 min. The cells were then washed twice in 200 l of FACS- FIGURE 1. Influx of LY6Ghigh cells into the peritoneum of LBP / PBS. The cell pellet was resuspended in 50 ␮l of FACS-PBS containing ␮ ␮ mice 2 h after injection of 100 l of saline i.p. (A and B) or with 5000 CFU 7.5 g/ml anti-rat Ig-Cy3 (Dianova, Hamburg, Germany). After incubation ␮ on ice for 30 min, the labeled cells were washed twice with FACS-PBS, of S. typhimurium (S.typh)in100 l of saline (C and D). Cells were stained resuspended in 50 ␮l of PBS, and treated with 50 ␮l of 4% formaldehyde. with isotype control (A and C) or with the mAb RB6-8C5 (B and D). After high Fixation was for 20 min at room temperature. The fixed cells were washed staining the cells were examined in the FACScan. The influx of Ly6G twice with FACS-PBS, resuspended in 50 ␮l of FACS-PBS supplemented cells is highly reproducible and the result shown is typical of Ͼ200 animals with 0.2% saponin (Sigma-Aldrich, St. Louis, MO) containing TOTO-3 examined. The Journal of Immunology 4477

FIGURE 2. Influx of cells into the peritoneum following Salmonella infection. Total peritoneal cells (T) were determined in a counting cham-

ber. Neutrophils (N) were determined by FACS analysis as shown in Fig. Downloaded from 1. These cell populations were evaluated in uninfected BALB/c animals, in mice infected for 24 h with 3000 CFU of Salmonella i.p. or with 30 CFU. One of two duplicate experiments is shown.

CFU, the effects are smaller but even here a substantial neutrophil influx takes place (Fig. 2). http://www.jimmunol.org/ Peritoneal neutrophils in LBPϪ/Ϫ and CD14Ϫ/Ϫ mice To determine what role LBP and CD14 may play in the host re- sponse to the infection, we investigated the kinetics of neutrophil influx into the peritoneum in LBPϩ/Ϫ, LBPϪ/Ϫ, and CD14Ϫ/Ϫ mice. Animals were infected i.p. with 3000 CFU of Salmonella and after 24 h the influx of neutrophils was determined by FAC- Scan analysis. In all three lines, a considerable neutrophil influx was evident following infection, showing that neither LBP nor CD14 is essential for this to occur (Fig. 3A). Nevertheless, when by guest on September 26, 2021 earlier stages of the infection were examined, we found that the absence of either CD14 or LBP delays the neutrophil influx. Two hours after infection, many neutrophils have entered the perito- neum of the control animals (Fig. 3B) but no such influx is appar- ent in the CD14Ϫ/Ϫ or LBPϪ/Ϫ animals (Fig. 3B). By 4 h after infection, however, neutrophil influx into the peritoneum is appar- ent even in the LBPϪ/Ϫ and the CD14Ϫ/Ϫ animals (Fig. 3C). In contrast, the absolute number of F4/80ϩ macrophages, which in normal mice make up 20Ð50% of the peritoneal cell population, FIGURE 3. Neutrophil influx in LBP- and CD14-deficient animals. A, Influx of neutrophils into the peritoneum measured as percentage of total remains unchanged and this result is unaffected by the presence or high ϩ/Ϫ Ϫ/Ϫ Ϫ/Ϫ absence of either LBP or CD14 (data not shown). peritoneal cells which are Ly6G in LBP , CD14 , and LBP animals 24 h after injection of 3000 CFU of Salmonella. B, Influx of neu- Delayed influx of neutrophils is accompanied by expansion of trophils 2 h after infection. C, Influx of neutrophils 4 h after infection. One Salmonella representative experiment of two is shown. We next asked whether differences in the number of bacteria are detectable 2Ð4 h after infection. As shown in Fig. 4, LBPϪ/Ϫ an- imals, both at 2 and 4 h after infection, have a small but significant IL-6. The serum IL-6 increase is already evident 2 h after infec- ( p Ͻ 0.05, Mann-Whitney U test) deficiency in their capacity to tion. This inflammatory response, which is essential for survival, is clear the infection and hold the multiplication of the pathogen in not detected in LBPϪ/Ϫ animals (9). Since neutrophil influx into check. Indeed, a substantially increased number of Salmonella are detected in the peritoneum and in peripheral organs of LBP-defi- the peritoneum following infection is by this time already appar- cient animals 24Ð72 h after infection (6, 9). Thus, a delayed influx ent, we asked whether an inflammatory reaction could be detected Ϫ/Ϫ of neutrophils into the peritoneum might provide an opportunity in the peritoneum at early time points. Groups of three LBP , Ϫ/Ϫ for the Salmonella to expand and overwhelm the innate defense CD14 , and control mice were infected with 5000 CFU of Sal- system. monella and after 0.5, 1.0, and 2.0 h the titer of the proinflam- matory cytokines IL-6 and TNF in the peritoneum was mea- Both TNF and IL-6 can be detected in the peritoneum early in sured by ELISA. As shown in Fig. 5, both IL-6 and TNF are infection readily detectable in the peritoneum 1 h after infection in the In wild-type mice, a peritoneal Salmonella infection is followed by control animals but are not detectable in the CD14Ϫ/Ϫ or in the a rapid inflammatory response measured as an increase in serum LBPϪ/Ϫ mice. 4478 PERITONEAL NEUTROPHIL INFLUX AFTER INFECTION WITH Salmonella

FIGURE 4. Bacterial load in the peritoneum after infection of LBPϪ/Ϫ or LBPϩ/Ϫ animals with 3000 CFU of Salmonella. Data are shown for immediately (5Ð10 min), 2 h, or 4 h after infection. The number of CFU in the peritoneal wash was determined. One of two duplicate experiments is shown Downloaded from FIGURE 6. TNF induces neutrophil influx into the peritoneum. LBPϩ/Ϫ ␮ ␮ Neutrophil influx can be induced by injection of TNF animals were injected i.p. either with 200 l of saline or with 200 lof saline containing 800 ng of TNF. Two and 4 h later peritoneal wash cells were Since an early proinflammatory response involving TNF is of cen- recovered and the neutrophils were enumerated by FACScan analysis. A sim- tral importance in surviving a peritoneal Salmonella infection, we ilar result was obtained in a repeat experiment using mice of strain CBA. asked whether injection of TNF alone would stimulate neutrophil

influx in uninfected mice. As shown in Fig. 6 injection of 800 ng http://www.jimmunol.org/ of recombinant human TNF (9) into LBPϩ/Ϫ animals indeed re- in organizing neutrophil influx early in the infection. To test this sulted in a substantial influx of neutrophils measured 2 and 4 h hypothesis, we have made use of TNF-deficient animals. As shown later. in Fig. 7, the animals carrying a targeted deletion of the TNF gene do mount a neutrophil influx in response to Salmonella infection Neutrophil influx in animals lacking TNF but the response is significantly less than that seen in the controls The fact that exogenously applied TNF is able to induce neutrophil injected with the same number (800 CFU) of Salmonella. We have influx raises the possibility that TNF may play a major role in vivo tried as an alternative approach to reduce the available TNF in the mice by injecting 1.25 mg (63 mg/kg) of recombinant human TNFR2-IgG1-Fc H chain (Enbrel; Immunex, Seattle, WA) at the by guest on September 26, 2021 same time as the Salmonella. In contrast to the result with the TNF KO mice, there is no reduction in the neutrophil influx in Enbrel- treated animals.

Late influxing neutrophils in CD14Ϫ/Ϫ and LBPϪ/Ϫ mice are able to phagocytose Salmonella To determine whether the neutrophils which move into the peri- toneum late are competent to phagocytose the Salmonella, we have conducted confocal microscopy of peritoneal cells recovered after peritoneal infection with Salmonella expressing GFP. To be able to

FIGURE 5. IL-6 (A) and TNF (B) in the peritoneum of CD14Ϫ/Ϫ, LBPϪ/Ϫ, and LBPϩ/Ϫ animals 30 min, 1 h, or 2 h after infection with 5000 FIGURE 7. Neutrophil influx in TNF-deficient animals and in controls CFU of Salmonella. ELISA determinations were conducted in triplicate. 2 h after infection with 800 CFU of Salmonella. Results shown are pooled Values shown are picograms cytokine per peritoneum. data from two experiments. The Journal of Immunology 4479

FIGURE 8. Delayed entry neutrophils which enter the peritoneum late are able to phagocytose Salmonella. LBPϩ/Ϫ, LBPϪ/Ϫ, and CD14Ϫ/Ϫ mice were infected with 1 ϫ 107 CFU of Salmonella i.p.;2h(LBPϩ/Ϫ) or 4 h (LBPϪ/Ϫ and CD14Ϫ/Ϫ) later the peritoneal cells were recovered from the mice, stained for neutrophil Ag RB6-8C5/anti-rat Ig-Cy3 (red), fixed, and analyzed in three dimensions by confocal microscopy. Salmonella expressing GFP are depicted in green. The image is a projection of a 20-␮m Z-stack collected through a ϫ63 objective on a Zeiss LSM510 microscope. One xy plane out of the whole Z-stack is shown. The red line indicates an orthogonal section along the xz plane, the green line along the yz plane, and the blue line along the xy plane. Together, the three views show the intracellular localization of the Salmonella inside neutrophils. The LBPϪ/Ϫ neutrophil is a representative example of Ϫ Ϫ ϩ Ϫ seven optically sectioned cells. The CD14 / neutrophil is representative of five serially sectioned cells. The LBP / neutrophil is representative of seven Downloaded from optically sectioned cells. detect sufficient phagocytosed bacteria, we injected ϳ1 ϫ 107 with the release of proinflammatory mediators (18). Whether they CFU of Salmonella and examined the peritoneal wash 2Ð4 h after require soluble CD14 to do this has not yet been established. An infection. After recovery from the peritoneum, the neutrophils early consequence of the proinflammatory response involves a http://www.jimmunol.org/ were surface stained using the mAb RB6-8C5. The staining pro- rapid and massive influx of neutrophils into the peritoneum. cedure was optimized to maintain the three-dimensional structure Within a few hours sufficient neutrophils flow in to double the size of the cells for confocal microscopy (see Materials and Methods). of the peritoneal population. One of the most astonishing aspects A preliminary examination in the confocal microscope at low mag- of this response is that it can be triggered by as few as 30 CFU nification showed that ϳ20% of neutrophils in LBPϪ/Ϫ, CD14Ϫ/Ϫ, (Fig. 2). and LBPϩ/Ϫ controls were associated with Salmonella. This may It has previously been shown in a peritoneal Klebsiella pneu- well be an underestimate since the GFP fluorescence seems to be monia infection in WBB6F1 mice that removal of TNF reduces but rather rapidly lost after phagocytosis. Neutrophils staining positive does not abolish neutrophil influx. It was concluded that TNF plays for the GFP label of the bacteria were subjected to confocal serial a major role in stimulating neutrophil influx but that other medi- by guest on September 26, 2021 optical sectioning to determine whether the associated Salmonella ators may also be involved (15). The results we obtained in the were inside the cell or merely bound on the surface. In three in- Salmonella model are strikingly similar. The neutrophil influx at dependent experiments, a total of seven LBPϪ/Ϫ neutrophils, five 2 h was substantially reduced but not to the same extent as was CD14Ϫ/Ϫ neutrophils, and seven neutrophils from the LBPϩ/Ϫ seen in the LBPϪ/Ϫ or CD14Ϫ/Ϫanimals. This suggests that in this controls were examined in this way. Typical results shown in Fig. 8 infection model the proinflammatory response initiated via LBP demonstrate that not only neutrophils from LBPϩ/Ϫ mice (Fig. 8A) and CD14 generates mediators other than TNF which can also, but also those from both CD14Ϫ/Ϫ (Fig. 7B) and LBPϪ/Ϫ (Fig. 8C) albeit less efficiently, organize the influx of neutrophils into the animals were able to ingest the bacteria in vivo. Thus, these neutro- peritoneum. In this respect, it is interesting that the attempt to phils, although delayed in their influx kinetics, are phagocytosis com- abrogate neutrophil influx using Enbrel was not successful even petent once they reach the peritoneal cavity. though the human TNFR2 is known to bind mouse TNF and the 62.5-mg/kg dose used is substantial in comparison to the 0.4 Discussion mg/kg routinely used in the treatment of human patients with poly- Since peritonitis is a life-threatening situation, the host’s capacity articular-course juvenile rheumatoid arthritis. One possible expla- to monitor potential danger signals within the peritoneum may be nation might be that in this infection model the level of TNF re- of considerable survival value. Peritoneal infection models in the quired to induce neutrophil influx is very low, lower than can be mouse using Escherichia coli (14), Klebsiella pneumoniae (15), reached by entrapment with the soluble TNFR. Were this to be the Staphylococcus aureus (16), or polymicrobial sepsis induced by case it might help explain why there is only a moderate increase of cecal ligation and puncture (17) all involve large numbers of bac- the risk of serious infection in patients treated with this drug (19). teria (Ͼ107) to initiate the infection. For the analysis of host bac- The low but significant level of neutrophil influx discernible in terial detection systems, it may be helpful to examine an infection the TNF KO mice 2 h after infection indicates that TNF-indepen- system in which the load of infecting agent is much smaller. For dent processes are able to initiate neutrophil influx. In line with this this reason, we use a peritoneal infection with S. typhimurium is the observation that loss of either LBP or CD14 does not lead to which can be initiated with Ͻ1000 CFU (6). a complete failure to recruit neutrophils into the peritoneal cavity. One of the earliest responses to a peritoneal infection with Sal- Indeed, 24 h after infection there is no discernible difference be- monella which we have so far been able to detect involves the tween the KO animals and the controls in the number of peritoneal release of proinflammatory mediators, including IL-6 and TNF. neutrophils. Clearly, systems other than LBP-CD14-TLR4 are also The rapid release of TNF is a hallmark of mast cell activation since able to induce an influx of phagocytosis competent neutrophils in these are the only cells in the body that store preformed TNF (15, response to a Salmonella infection. However, in the absence of the 17). Although bone marrow-derived mast cells do not express LPS detection system, there is a short but crucial delay in the CD14, they do express TLR4 and are competent to respond to LPS recruitment of neutrophils. 4480 PERITONEAL NEUTROPHIL INFLUX AFTER INFECTION WITH Salmonella

Since any single recognition system may eventually be evaded protein is required to combat a murine Gram-negative bacterial infection. Nature by the pathogen, it is likely that the host relies on a spectrum of 389:742. 7. Bernheiden, M., J. M. Heinrich, G. Minigo, C. Schutt, F. Stelter, M. Freeman, detection systems for its defense. Indeed, recent evidence indicates D. Golenbock, and R. S. Jack. 2001. LBP, CD14, TLR4 and the murine innate that there are multiple possibilities for the mouse to detect an in- immune response to a peritoneal Salmonella infection. J. Endotoxin Res. 7:447. cipient peritoneal infection with Salmonella. Like all other bacte- 8. Takeuchi, O., K. Hoshino, T. Kawai, H. Sanjo, H. Takada, T. Ogawa, K. Takeda, and S. Akira. 1999. Differential roles of TLR2 and TLR4 in recognition of Gram- ria, Salmonella produces lipoproteins capped with the N-acyl-S- negative and Gram-positive bacterial cell wall components. Immunity 11:443. diacylglyceryl cysteine element necessary for interaction with 9. Heinrich, J. M., M. Bernheiden, G. Minigo, K. K. Yang, C. Schutt, D. N. Mannel, TLR2 (20, 21). The LPS detection system, consisting of LBP, and R. S. Jack. 2001. The essential role of lipopolysaccharide-binding protein in protection of mice against a peritoneal Salmonella infection involves the rapid CD14, and TLR4, provides a second possibility (6, 22). A third induction of an inflammatory response. J. Immunol. 167:1624. possibility may be offered by the fact that Salmonella is a flagel- 10. Seiler, P., P. Aichele, B. Raupach, B. Odermatt, U. Steinhoff, and S. H. Kaufmann. lated bacterium whose flagellin is a potent activator of TLR5 (23). 2000. Rapid neutrophil response controls fast-replicating intracellular bacteria but not slow-replicating Mycobacterium tuberculosis. J. Infect. Dis. 181:671. Beyond these TLR-based systems, complement activation pro- 11. Moore, K. J., L. P. Andersson, R. R. Ingalls, B. G. Monks, R. Li, M. A. Arnaout, vides a further innate means of sensing a peritoneal infection (24). D. T. Golenbock, and M. W. Freeman. 2000. Divergent response to LPS and It is likely that many, perhaps all, of these systems play a role in bacteria in CD14-deficient murine macrophages. J. Immunol. 165:4272. 12. Marino, M. W., A. Dunn, D. Grail, M. Inglese, Y. Noguchi, E. Richards, defense against a peritoneal Salmonella infection. In the race be- A. Jungbluth, H. Wada, M. Moore, B. Williamson, et al. 1997. Characterization tween the host and pathogen, survival of the host depends on of tumor necrosis factor-deficient mice. Proc. Natl. Acad. Sci. USA 94:8093. achieving maximal advantage from an integrated set of defense 13. Goossens, P. L., H. Jouin, and G. Milon. 1991. Dynamics of lymphocytes and inflammatory cells recruited in liver during murine listeriosis: a cytofluorimetric strategies. Incapacitation of any of the host’s defenses may tip the study. J. Immunol. 147:3514. balance in favor of the pathogen. Loss of the LPS detection system 14. Haziot, A., N. Hijiya, S. C. Gangloff, J. Silver, and S. M. Goyert. 2001. Induction Downloaded from through ablation of LBP (6) or of CD14 (11) results in a failure to of a novel mechanism of accelerated bacterial clearance by lipopolysaccharide in CD14-deficient and Toll-like receptor 4-deficient mice. J. Immunol. 166:1075. induce an early inflammatory response, leading to a delayed re- 15. Malaviya, R., T. Ikeda, E. Ross, and S. N. Abraham. 1996. Mast cell modulation cruitment of neutrophils into the peritoneum. It would seem un- of neutrophil influx and bacterial clearance at sites of infection through TNF-␣. likely that the modest increase in bacterial numbers which this Nature 381:77. 16. Thomas, C. A., Y. Li, T. Kodama, H. Suzuki, S. C. Silverstein, and J. El Khoury. delay permits (Fig. 4) would be sufficient to explain the central 2000. Protection from lethal Gram-positive infection by macrophage scavenger importance of LBP and CD14 in defense against Salmonella. How- receptor-dependent phagocytosis. J. Exp. Med. 191:147. http://www.jimmunol.org/ ever, this pathogen is known to adapt to life within the host by a 17. Echtenacher, B., D. N. Ma¬nnel, and L. Hultner. 1996. Critical protective role of mast cells in a model of acute septic peritonitis. Nature 381:75. reorganization of its gene expression pattern (25, 26). The delayed 18. Supajatura, V., H. Ushio, A. Nakao, K. Okumura, C. Ra, and H. Ogawa. 2001. response may give the pathogen a short but crucial window of Protective roles of mast cells against enterobacterial infection are mediated by opportunity in which to complete this adaptation process, thus en- Toll-like receptor 4. J. Immunol. 167:2250. 19. Phillips, K., M. E. Husni, E. W. Karlson, and J. S. Coblyn. 2002. Experience with hancing its capacity to overrun the remaining defense systems. In etanercept in an academic medical center: are infection rates increased? Arthritis this way an apparently minor delay in neutrophil recruitment in the Rheum. 47:17. initial hours of the infection may make the difference between life 20. Brightbill, H. D., D. H. Libraty, S. R. Krutzik, R. B. Yang, J. T. Belisle, J. R. Bleharski, M. Maitland, M. V. Norgard, S. E. Plevy, S. T. Smale, et al. 1999. and death a week later. We are currently attempting to clarify this Host defense mechanisms triggered by microbial lipoproteins through Toll-like issue by extending our analysis of the development of the infection receptors. Science 285:732. by guest on September 26, 2021 beyond the first 24 h. 21. Aliprantis, A. O., R. B. Yang, M. R. Mark, S. Suggett, B. Devaux, J. D. Radolf, G. R. Klimpel, P. Godowski, and A. Zychlinsky. 1999. Cell activation and apo- ptosis by bacterial lipoproteins through Toll-like receptor-2. Science 285:736. References 22. Poltorak, A., X. He, I. Smirnova, M. Y. Liu, C. V. Huffel, X. Du, D. Birdwell, 1. Medzhitov, R., and C. A. Janeway.1997. Innate immunity: the virtues of a non- E. Alejos, M. Silva, C. Galanos, et al. 1998. Defective LPS signaling in C3H/HeJ clonal system of recognition. Cell 91:295. and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 282:2085. 2. Hoffmann, J. A., F. C. Kafatos, C. A. Janeway, and R. A. Ezekowitz. 1999. 23. Hayashi, F., K. D. Smith, A. Ozinsky, T. R. Hawn, E. C. Yi, D. R. Goodlett, Phylogenetic perspectives in innate immunity. Science 284:1313. J. K. Eng, S. Akira, D. M. Underhill, and A. Aderem. 2001. The innate immune 3. Wright, S. D., R. A. Ramos, P. S. Tobias, R. J. Ulevitch, and J. C. Mathison. response to bacterial flagellin is mediated by Toll-like receptor 5. Nature 410: 1990. CD14, a receptor for complexes of lipopolysaccharide (LPS) and LPS 1099. binding protein. Science 249:1431. 24. Gommerman, J. L., D. Y. Oh, X. Zhou, T. F. Tedder, M. Maurer, S. J. Galli, and 4. Tobias, P. S., K. Soldau, and R. J. Ulevitch. 1986. Isolation of a lipopolysaccha- M. C. Carroll. 2000. A role for CD21/CD35 and CD19 in responses to acute ride binding acute phase reactant from rabbit serum. J. Exp. Med. 164:777. septic peritonitis: a potential mechanism for mast cell activation. J. Immunol. 5. Medzhitov, R., P. Preston-Hurlburt, and C. A. Janeway, Jr. 1997. A human ho- 165:6915. ϩ mologue of the Drosophila Toll protein signals activation of adaptive immunity. 25. Vescovi, G. E., F. C. Soncini, and E. A. Groisman. 1996. Mg2 as an extracel- Nature 388:394. lular signal: environmental regulation of Salmonella virulence. Cell 84:165. 6. Jack, R. S., X. Fan, M. Bernheiden, G. Rune, M. Ehlers, A. Weber, G. Kirsch, 26. Burns-Keliher, L., C. A. Nickerson, B. J. Morrow, and R. Curtiss. 1998. Cell- R. Mentel, B. Furll, M. Freudenberg, et al. 1997. Lipopolysaccharide-binding specific proteins synthesized by Salmonella typhimurium. Infect. Immun. 66:856.