TSG-14 transgenic mice have improved survival to endotoxemia and to CLP-induced sepsis

TSG-14 transgenic mice have improved survival toendotoxemia and to CLP-induced sepsis

Adriana A. M. Dias,*,† Adam R. Goodman,‡ Jane Lima Dos Santos,§ Rachel Novaes Gomes,i

Anne Altmeyer,‡ Patrıcia T. Bozza,i Maria de Fatima Horta,§ Jan Vilcek,‡ and Luiz F. L. Reis**Ludwig Institute for Cancer Research, Sao Paulo, Brazil; Departments of †Microbiology, and §Biochemistry andImmunology, UFMG, Minas Gerais, Brazil; ‡Department of Microbiology, New York University School of Medicine,New York, New York; and iDepartment of Physiology and Pharmacodynamics, Instituto Oswaldo Cruz, Rio deJaneiro, Brazil

Abstract: Tumor necrosis factor-stimulated gene14 (TSG-14)/PTX3 was identified originally as aTNF-a and IL-1b-stimulated gene from normal,human foreskin fibroblasts and vascular endothe-lial cells, respectively. TSG-14 gene encodes a 42-kDa-secreted glycoprotein with a carboxy-terminalhalf that shares homology with the entire sequenceof C-reactive protein (CRP) and serum amyloid Pcomponent (SAP), acute-phase proteins of the pen-traxin family. Some experimental evidence sug-gests that TSG-14 plays a role in inflammation, yetits function and mechanism of action remain un-clear. We have generated transgenic mice thatoverexpress the murine TSG-14 gene under thecontrol of its own promoter. From eight transgenicfounders, two lineages were derived and bettercharacterized: Tg2 and Tg4, carrying two and fourcopies of the transgene, respectively. TSG-14transgenic mice were found to be more resistant tothe endotoxic shock induced by LPS and to thepolymicrobial sepsis caused by cecal ligation andpuncture (CLP). Moreover, macrophages derivedfrom the transgenic mice produced higher amountsof nitric oxide in response to IFN-g, TNF-a, andLPS as compared with macrophages from wild-typeanimals, and the augmented response appears to bethe consequence of a higher responsiveness oftransgenic macrophages to IFN-g. The data shownhere are the first in vivo evidence of the involve-ment of TSG-14 in the inflammatory process andsuggest a role for TSG-14 in the defense againstbacterial infections. J. Leukoc. Biol. 69:928–936; 2001.

Key Words: TNF z lipopolysaccharide z endotoxin z inflamma-tion z nitric oxide

INTRODUCTION

Tumor necrosis factor (TNF)-stimulated gene 14 (TSG-14), alsocalled PTX3, was isolated by differential hybridization from acDNA library derived from TNF-a-treated, human FS-4, dip-loid foreskin fibroblasts [1] and from human umbilical vein

endothelia cells exposed to interleukin (IL)b-1 [2]. A murinehomologue of TSG-14 gene was also identified, and the pro-moter structure of the gene and its complete cDNA sequencewere determined [3, 4]. TSG-14 is a single-copy gene, orga-nized into three exons and two introns on murine chromosome3 and on human chromosome 3 band q25 [2, 4]. The geneencodes a 42-kDa-secreted glycoprotein with its carboxy-ter-minus displaying up to 27% amino acid sequence identity tohuman C-reactive protein (CRP) and serum amyloid protein(SAP) [2, 5]. CRP and SAP are acute-phase proteins that aremembers of the pentraxin family, characterized by a discoidarrangement of five noncovalently bound subunits and pres-ence of an eight amino acid domain (“the pentraxin signature”)[6]. TSG-14 protein contains the pentraxin signature, but it isapproximately twice as large as a CRP or SAP subunit. Theamino-terminal portion of TSG-14 protein does not show sig-nificant sequence homology with the pentraxins or other knownproteins [2, 5].

The recent discovery of several new proteins that, likeTSG-14, contain carboxy-termini homologous to CRP and SAPbut have divergent amino-terminal sequences suggests that anew group of proteins emerged from fusions of novel, amino-terminal domains to an ancestral, pentraxin domain [7]. Thenew family of pentraxins is known as the “long pentraxins” andcomprises the Xenopus laevis pentraxin (XL-PXN1) [8], theguinea pig apexin [9, 10], the rat neuronal pentraxin (NPI) [11]and its receptor (NPR) [12], the human neuronal pentraxin(NPTX2) [13], the human neuronal activity-related pentraxin(NARP) [14], and TSG-14. Unlike TSG-14, the other longpentraxins have relatively restricted patterns of expression andare not regulated by cytokines.

TSG-14 is produced by fibroblasts [1], endothelial cells [2],chondrocytes and synoviocytes [15], and cells of the monocyte/macrophage lineage [16] stimulated with TNF-a [1] or IL-1b

Correspondence: Luiz F. L. Reis, Ludwig Institute for Cancer Research, RuaProf. Antonio Prudente 109, 4th Floor, CEP 01509-010, Sao Paulo, Brazil.E-mail: [email protected]

Current address of Adam R. Goodman: Howard Hughes Medical Institute,Waksman Institute of Microbiology, Rutgers University, 190 FrelinghuysenRoad, Piscataway, NJ 08854.

Current address of Anne Altmeyer: Merck & Co., 126 E. Lincoln Avenue,Rahway, NJ 07065.

Received December 4, 2000; revised February 2, 2001; accepted February7, 2001.

928 Journal of Leukocyte Biology Volume 69, June 2001 http://www.jleukbio.org

[2] and also with lipopolysaccharides (LPS) [17] or componentsof the mycobacterial cell wall [18]. TSG-14 expression inmonocytes and macrophages is regulated negatively by inter-feron-g (IFN-g) [19, 20]. It was shown that, like CRP, TSG-14protein forms multimers [7, 21]. The capacity of TSG-14 tobind to the C1q complement component is suggestive of a rolefor TSG-14 as a local regulator of innate immunity [21]. UponLPS injection, TSG-14 protein level in the serum rises withkinetics similar to other acute-phase proteins [4, 17]. However,in contrast to classic acute-phase proteins, TSG-14 is notexpressed in the liver, and its expression is not stimulated byIL-6 [4, 17]. The major sites of TSG-14 expression in vivo areskeletal muscle and heart [4].

To investigate the physiological function of TSG-14 and itsrole in the inflammatory response, we generated transgenic (Tg)mice overexpressing the murine gene under the control of itsown promoter. Two lineages of Tg mice containing two and fourextra copies of the gene were characterized. We show thatTSG-14 Tg mice are more resistant to the systemic adminis-tration of LPS, compared with macrophages from wild-type(Wt) animals and to sepsis caused by cecal ligation and punc-ture (CLP). We also show that, compared with macrophagesfrom Wt animals, peritoneal macrophages from TSG-14 Tganimals produce larger amounts of nitric oxide (NO) in re-sponse to IFN-g or IFN-g in combination with LPS or TNF-a.

MATERIALS AND METHODS

Generation of Tg mice

Two murine TSG-14 genomic fragments were cloned originally into the BamHIsite of the plasmid pGEM7Zf(1) (Promega, Madison, WI): pGEM(2.7)TSG-14and pGEM(5.6)TSG-14. The pGEM(2.7)TSG-14 contains the TSG-14 genomicfragment that comprises the first and second exons and the flanking regions aswell as 1.5 kb of 59 up-stream sequence that contains the promoter of the gene[3]. The pGEM(5.6)TSG-14 construct contains a 5.6-kb TSG-14 genomic

fragment that includes the remains of the second intron, the entire third exon,and 700 bp of the flanking 39 sequence of the gene. The transgene comprisingthe whole TSG-14 genomic sequence was obtained by the ligation of the 2.7-and 5.6-kb genomic fragments. The TSG-14 5.6-kb sequence was liberatedfrom the plasmid by a BamHI digestion and joined to a pGEM(2.7)TSG-14,partially digested with BamHI. The junctions were sequenced to determine theorientation, and the pGEM7Zf(1) backbone sequence was later removed by aApaI digestion to create the 8.3-kb genomic fragment, identical to the residentgene, that was used for microinjection. Microinjections of DNA and embryoimplantation were performed at the Transgenic Animal Facility of the SkirballInstitute, New York University School of Medicine (New York, NY; directed byAnna Auerbach). Tg animals were produced in an outbred CD1 geneticbackground, and the same strain was used for the propagation of the Tglineages.

Southern blot analysis of genomic DNA

Genomic DNA was obtained from a 1-cm-long fragment of the tail tip aspreviously described [22]. Purified, genomic DNA was digested with therestriction enyme NcoI, electrophoresed through a 0.9% agarose gel, trans-ferred to Hybond nylon membrane (Amersham, Arlington Heights, IL), andhybridized [23] with a [32P]dCTP-labeled BamHI/NsiI fragment (374 bp)derived from the pGEM(2.7)TSG-14 genomic sequence (Fig. 1B, hatchedbars). To estimate the number of integrated transgenes, densitometric analysesof the radioactive signals in the filters were performed by using the Gel-ProAnalyzer™ software (Media Cybernetics, L.P., Baltimore, MD).

Northern blot analysis

Mice (6- to 8-weeks old) were injected intraperitoneally (i.p.) with 30 mgEscherichia coli endotoxin (LPS serotype 0111:B4; Sigma Chemical Co., St.Louis, MO). Organs were removed 4 h after the LPS injection, and total RNAwas isolated by using the TRIZOL Reagent in accordance with the instructionsprovided by the manufacturer (Gibco BRL, Grand Island, NY). For Northernblot analysis, 15–20 mg total RNA derived from the indicated organs waselectrophoresed through a 1% denaturing, agarose gel, transferred to Hybondnylon membrane (Amersham), and hybridized with random-primed [32P]dCTP-labeled probes according to Church and Gilbert [23]. The following probeswere used: for glyceraldehyde-3-phosphate dehydrogenase (GAPDH), theAvaI/PstI fragment from the murine cDNA [24]; for TSG-14, a StyI/StyIfragment, corresponding to 53 bp of the first intron and 311 bp of the secondexon, isolated from the plasmid pGEM(2.7)TSG-14 [3]; for the N51/KC [amurine homologue of macrophage-inflammatory protein (MIP)-2] probe, a

Fig. 1. Southern blot analysis of genomic DNA from Wt and TSG-14 Tg animals. Genomic DNA was digested with the restriction enzyme NcoI, electrophoresedthrough a 0.9% agarose gel, blotted onto nylon membrane, and hybridized with a probe derived from the second exon of the TSG-14 gene (A). In each lane, thetop band corresponds to the Wt alleles and the additional, faster-moving bands, to the transgene copies. The intensity of the bands relative to the two Wt alleles,determined by scanning densitometry, was used to estimate the copy number in each Tg lineage. (B) A schematic representation of the number and orientationof integrated copies is shown. Arrowheads represent the NcoI restriction sites, and the hatched regions in the transgenes are complementary to the cDNA probeused for Southern blot analysis.

Dias et al. Improved survival of TSG-14 transgenic mice 929

fragment corresponding to the entire coding region of the gene obtained byreverse transcriptase-polymerase chain reaction (RT-PCR) using primersbased on the sequence deposited in the GenBank (accession number J04596)subcloned into the BamHI site of pGEM7Zf(1).

Induction of endotoxic shock

To determine the best dose of LPS for the induction of endotoxemia, 6- to8-week-old, CD1 Wt mice were injected i.p. with five different doses of LPS (E.coli endotoxin serotype 0111:B4; Sigma): 1.5 mg/kg, 2.7 mg/kg, 5.5 mg/kg,13.5 mg/kg, 27.0 mg/kg. Because 100% of animals died upon injection with13.5 mg/kg LPS, and 90% of the mice injected with 5.5 mg/kg survived, wedecided to use 11 mg/kg to compare the survival of Wt and Tg lineages. For theexperiments, the LPS was dissolved in 0.2 ml phosphate-buffered saline (PBS)and injected i.p., and lethality was monitored every 12 h for 7 days afterinjection. The survival curve was calculated with the aid of the Kaplan-Meierlife-table method [25]. All statistic tests were performed with 95% confidenceinterval (CI; 50.05) using the Stata program (Stata Corp., College Station, TX).

CLP

Sepsis was induced through CLP as previously described [26]. Briefly, 6- to8-week-old, CD1 Wt and TSG-14 Tg mice were anesthetized with an i.p.injection of 0.2 ml of a mixture of 100 mg/ml tyazine (Vetanarcol-Konig,Avellaneda, Argentina) and 2% ketamin (Rompun-Bayer, Merriam, KS) insaline (0.9% NaCl). A laparotomy was performed, and the cecum was exposedand ligated below the ileocecal junction, without causing bowel obstruction.The cecum was punctured once with an 18-gauge needle and then gentlysqueezed to ensure the leakage of the cecum contents through the puncture.The cecum was returned to the peritoneal cavity, and the body wall and skinincision were closed with a 6-0 silk suture. The animals were analyzed for thesurvival rate, assessed every 12 h for 6 days.

Isolation of peritoneal macrophages and nitritedetermination

Peritoneal macrophages were isolated from mice 4 days after an i.p. injectionof thioglycollate broth (2 ml/mice). The cells were then recovered by lavage ofthe peritoneal cavity with ice-cold PBS. The resulting cell suspension waswashed twice with PBS containing 50 mg/ml gentamicin sulfate by centrifu-gation for 10 min at 200 g. Cells (50,000) were plated in each well of a 96-wellplate in 200 ml RPMI 1640 (Sigma) supplemented with 10% of heat-inacti-vated fetal bovine serum and 50 mg/ml gentamycin sulfate and incubated for18 h. The adherent macrophages were then treated with the indicated doses ofIFN-g (R&D Systems, Minneapolis, MN), TNF-a (R&D Systems), or LPS (E.coli serotype 0127; Sigma) or with a combination of stimuli. Nitrite (NO2

2)concentration in the culture supernatants was determined as a measure thatreflects NO production, because the released NO reacts rapidly with water andoxygen to produce NO2

2, which can be measured as described by Hibbs andco-workers [27]. Supernatants (100 ml) were collected 24, 48, 72, and 96 hafter each treatment, mixed with an equal volume of Griess reagent [1%sulfanilamide (Sigma), 0.1% naphthylethylenediamine hydrochloride (Sigma),2.5% orthophosphoric acid (Riedel-de Haen, Hannover, Germany)], and leftfor 10 min at room temperature. Standards were prepared by using serialdilutions of sodium nitrite (2–200 mM). Absorbance was measured at 540 nmin a Spectramax™ 340 UV-Vis spectrophotometer. Macrophage viability wasevaluated using the MTT assay [28]. Briefly, after the removal of the culturesupernatant for the NO2

2 assay, fresh culture medium containing MTT (finalconcentration, 500 mg/ml) was added, and the cells were incubated for 4 h. Themedium was then removed, and MTT-formazan was solubilized by adding 10%sodium dodecyl sulfate (SDS), prepared in dymethylformamide/H20 (1:1 v/v),and measured spectrophotometrically at 550 nm.

Cytokine levels in LPS-injected mice

Wt and Tg2 mice were injected i.p. with LPS (11 mg/kg), and plasma wascollected 1.5 or 6 h after injection (n55 for each time-point). TNF, IL-6, andIL-10 were measured by enzyme-linked immunosorbent assay (ELISA) using akit from R&D Systems (for TNF) or reagents from PharMingen (San Diego, CA;for IL-6 and IL-10), following instructions of the manufacturers.

RESULTS

Generation of TSG-14 Tg mice

Murine TSG-14 is expressed in vivo mainly in the vascularendothelium of skeletal muscle and heart [4]. Our rationale wasthat mice overexpressing TSG-14 at the sites of its normalproduction would be a useful model for the investigation ofTSG-14 function in vivo. For this purpose, we used a 8.3-kbmurine, TSG-14 genomic fragment, with 100% identity to theresident gene, as a transgene for the pronuclear microinjectionand production of Tg mice. This transgene contained 1.5 kb ofthe up-stream sequence flanking the TSG-14 transcriptionalstart site, which has been shown previously to be sufficient forTSG-14 inducibility by TNF-a and IL-1 [3]. Eight TSG-14 Tgfounders were obtained. The number of the transgenes incor-porated into these founders and transmission to progeny weredetermined by Southern blot hybridization and found to varybetween one and four copies. All of the founders containingextra copies of the TSG-14 gene appeared normal phenotypi-cally and transmitted all copies of the transgenes to theirprogeny with the expected frequency. From these eightfounders, two Tg lines were derived. We refer to these lines asTg2 and Tg4 because the number of extra copies of the TSG-14genomic locus integrated in each Tg line is 2 and 4, respec-tively. For the analysis of the genotypes, the genomic DNA wasdigested with the restriction enzyme NcoI and hybridized witha probe corresponding to the 59 end of the promoter region ofTSG-14 (Fig. 1B, hatched portion). As indicated in Figure 1B,one NcoI restriction site is located at position 2045 of theTSG-14 gene, and the other NcoI site is approximately 8000 bpup-stream of position 1 of the 8.3-kb fragment. Digestion ofgenomic DNA from Wt and TSG-14 Tg mice with the restric-tion enzyme NcoI generates fragments of approximately 10 kb(Fig. 1A), corresponding to the endogenous alleles that hybrid-ize with the probe. In the lines Tg2 and Tg4, digestion withNcoI revealed additional copies, and the number of integratedtransgenes and their orientation is depicted in Figure 1B. Thenumber of the additional copies was estimated from the inten-sity of hybridization signals as measured by scanning densi-tometry.

LPS-induced, TSG-14 gene expression inTSG-14 Tg mice

To evaluate the level of TSG-14 expression, mice wereinjected i.p. with 30 mg LPS. Northern blot analysis of totalRNA isolated from the heart (Fig. 2A), skeletal muscle (Fig.2B), and liver (Fig. 2C) of these animals and of control,noninjected mice was performed. No constitutive TSG-14expression was observed in the tested organs of Wt- orTg-control mice. LPS-induced expression of TSG-14 wasreadily detectable in the heart and skeletal muscle of LPS-treated mice with higher mRNA levels present in the organsof Tg animals (Fig. 2 and unpublished results). Differentfrom other acute-phase response proteins, TSG-14 is notexpressed in the liver [4]. In agreement with this observa-tion, no TSG-14 mRNA was detected in the liver of Wt or Tgmice (Fig. 2C). The same blot was rehybridized with amurine probe specific for the chemokine N51/KC mRNA


(Fig. 2C) to demonstrate that LPS administration was induc-ing gene expression in the liver, indicative of an acute-phase response. Furthermore, there was no significant dif-ference in the levels of IL-6 mRNA in liver, heart, andmuscle of LPS-injected Wt or Tg mice (unpublished results).

Resistance of TSG-14 Tg mice to septic shockinduced by LPS injection or CLP

To evaluate the impact of higher TSG-14 expression on theresistance to endotoxic shock, we compared the survival of Wt

and TSG-14 Tg mice injected with LPS. Animals of the threegenotypes were injected i.p. with 11 mg/kg LPS, and theirsurvival was evaluated by the Kaplan-Meier life-table method[25] (Fig. 3). After the injection of LPS, Tg and Wt micedisplayed clear symptoms of murine endotoxic shock: de-creased motor activities, ruffled fur, and diarrhea. However, asignificant difference (p50.003) was detected in the death rate,with 15.4% (2/13) of death observed in Tg4 mice, 25% (2/8) inTg2, and 71.4% (15/21) in Wt mice. No difference was ob-served among the groups in the average survival time of themice that succumbed to the LPS injection.

The murine CLP model is a clinically relevant model ofsepsis that results in a polymicrobial bacteremia associatedwith low levels of circulating endotoxin and high concentra-tions of proinflammatory cytokines [29]. We evaluated the roleof TSG-4 in sepsis by comparing mortality in TSG-14 Tg andWt mice subjected to CLP. After surgery, Wt and Tg miceexhibited signs of sepsis, including piloerection, periorbitalexudates, diarrhea, and lethargy, but the manifestations weremore pronounced in Wt mice. As shown in Figure 4, a greaterproportion of Tg mice survived CLP, with 69% and 64%survival between Tg2 and Tg4 mice, respectively, and 28% inWt mice (p,0.01).

Increased NO production by macrophages fromTSG-14 Tg mice

Macrophages are important in the inflammatory response, andtheir function can be modulated by the production of severalsoluble mediators of inflammation [30]. To determine if en-hanced TSG-14 expression affects macrophage function, weevaluated the capacity of thioglycollate-elicited, peritonealmacrophages from the Tg animals to produce NO in response toIFN-g, TNF-a, and LPS. NO2

2 concentration was measured inthe culture supernatants collected at 24, 48, 72, and 96 h afterstimulation. Four independent experiments were performed inwhich we examined macrophages from Tg2, Tg4, and Wt micestimulated with IFN-g (40 U/ml), LPS (10 ng/ml), and TNF-a(500 U/ml), individually or with the combinations of IFN-g 1LPS and IFN-g 1 TNF-a. In all four independent experiments,macrophages from Tg mice released more NO than macro-

Fig. 2. Northern blot analysis of the TSG-14 mRNA in the heart (A), muscle(B), and liver (C) from CD1 Wt and TSG-14 Tg mice. Total RNA was isolatedfrom the organs of control (uninjected) mice and from mice sacrificed 4 h afteri.p. injection with LPS (30 mg/animal). Blots (20 mg RNA per lane) werehybridized with murine TSG-14 and N51/KC probes as described in Materialsand Methods. The same blots were rehybridized with a murine GAPDH probefor control of RNA loading. (C) Total RNA from TNF-treated, mouse embryonicfibroblasts (MEFs) was loaded as a positive control for the hybridization withthe TSG-14 probe.

Fig. 3. Survival of Wt and TSG-14 Tg mice followingLPS injection. Mice (6- to 8-weeks old; n513, 8, and21 for Tg4, Tg2, and Wt, respectively) were injectedi.p. with 11 mg/kg LPS, and lethality was monitoredevery 12 h for 7 days after injection. The survivalcurve for each group was determined by the Kaplan-Meier life-table method [25]. Survival in the Tg groupwas different significantly from that of Wt mice(P50.0032log-rank test).


phages from Wt mice at all time-points analyzed, i.e., 24, 48,72, and 96 h (unpublished results).

To further dissect the increased NO production by Tg mac-rophages, we performed a dose-response analysis using mac-rophages from Wt or Tg4 mice treated with different doses ofeach agent separately or with a combination of IFN-g 1 LPS orIFN-g 1 TNF-a, as illustrated in Figure 5. The different dosesof LPS and TNF-a tested alone were not sufficient to inducethe production of NO by the peritoneal macrophages from theTg or Wt mice at any of the time-points examined (Fig. 5A andunpublished results). It is interesting that at 24-h post-induc-tion, IFN-g alone stimulated NO production in Tg4 macro-phages but not in Wt macrophages (Fig. 5A). At later time-points, it was possible to detect IFN-g-stimulated NO produc-tion by Wt macrophages, but the levels were much less than inTg-derived macrophages. TNF and LPS dose-dependently syn-ergized with IFN-g in NO induction in Wt and Tg4 macro-phages (Fig. 5, B and C). However, under all conditions tested,NO yields were about threefold greater in Tg4 macrophagesthan in Wt macrophages. Although Figure 5, B and C, repre-sents the levels of NO production at 48 h after stimulation,similar differences between NO production in Wt and Tg4macrophages were observed at 24, 72, and 96 h (unpublishedresults). To confirm that equal amounts of macrophages wereseen in every well (53104 cells/well), we performed the MTTassay, which confirmed a comparable density of viable cells inall groups (unpublished results).

Cytokine expression in LPS-injected mice

To analyze the role of TSG-14 in endotoxemia, we measuredcytokine levels in the plasma of Wt and Tg mice after LPSchallenge. In untreated Wt and Tg mice, there was no differ-ence in the levels of circulating TNF and IL-6. In contrast, Tgmice showed a higher basal level for IL-10 (7.16 ng/ml) thanWt mice (0.18 ng/ml; Fig. 6). In LPS-treated mice, the levelsof IL-10 start to increase in Wt mice at 1.5-h post-injection,but, nevertheless, it remained higher in Tg mice. However,after 6 h of LPS injection, the levels of IL-10 were higher in Wtthan in Tg mice (Fig. 6, top panel). For TNF, there was a sharpincrease in the levels of circulating cytokine after 1.5 h of LPSinjection, and again, Tg mice had about twice as much TNF

than Wt mice (Fig. 6, middle panel). Finally, no differenceswere observed in the levels of IL-6 at 1.5 h and 6 h after LPSinjection (Fig, 6, bottom panel).

DISCUSSION

Our results suggest a role for TSG-14 in the inflammatoryresponse in vivo. TSG-14 was the first long pentraxin to beidentified, and its structure and induction by several cytokineshave been well-characterized [7]. Earlier findings suggested arole for TSG-14 in the inflammatory process, such as thebinding of TSG-14 to the C1q component of the complementcascade [21], the high levels of the protein detected in theserum of LPS-injected humans and mice [4, 17], and in thejoint fluid from rheumatoid arthritis patients [15], as well as thefact that the expression of TSG-14 is regulated by the inflam-matory cytokines TNF-a and IL-1b [1, 4, 17]. However, theactual function of TSG-14 and its mode of action are stillunknown.

We have generated TSG-14 Tg mice as a model for thefunctional characterization of TSG-14 in vivo and for the in-vestigation of its role in inflammation. From eight Tg foundersobtained, two lineages were derived and characterized. Noconstitutive expression of the transgene was detected in theliver, heart, and skeletal muscle. TSG-14 induction by LPS wasobserved at the sites where the resident gene was also induced,i.e., heart and skeletal muscle, but not in the liver (Fig. 2). Thisobservation suggests that the Tg TSG-14 promoter activity wasnot affected by the neighboring sequences at the sites ofintegration. The two Tg lineages, containing either two or fourextra copies of the gene, showed an average increase of four- tosixfold as compared with Wt in the levels of induced, TSG-14mRNA expression (Fig. 2 and unpublished results). It is alsonoteworthy that no single base pair was changed within theentire 8.3 kb of the TSG-14 transgene, making the mRNAderived from the transgene identical to that derived from theresident gene. It appears that in the Tg mice, TSG-14 isexpressed at higher levels but in a physiologically coordinatedmanner. Therefore, the phenotypic changes observed in the Tg

Fig. 4. Survival of Wt and TSG-14 Tg mice afterCLP. Mice (6- to 8-weeks old; n519, 14, and 18 forTg4, Tg2, and Wt, respectively) were subjected toCLP, and survival was determined every 12 h for 7days after surgery. Survival in Wt (27.8%) and in Tg4and Tg2 (64.3% and 69.1%, respectively) was differ-ent significantly (P,0.012log-rank test).


mice can be attributed to the higher abundance of TSG-14mRNA and, consequently, to TSG-14 protein.

We demonstrate that overexpression of TSG-14 confers re-sistance to the lethal effects of LPS. LPS is used commonly forinducing a sepsis-like state in experimental animals. LPSinjection elicits the secretion of several cytokines, includingTNF-a and IL-1b, and these cytokines orchestrate the re-sponses observed in endotoxemia by affecting changes in ther-moregulation, vascular permeability and resistance, cardiacfunction, bone marrow function, and activity of key enzymes[31, 32]. TNF-a and IL-1b also stimulate the production ofadditional cytokines along with other inflammatory mediators,playing a role in the activation of the complement as well as the

coagulation and kinin cascades [33]. TSG-14 Tg mice injectedi.p. with endotoxin had a significantly higher survival rate (Fig.3). It has been shown previously that mice overexpressing theacute-phase, response-protein CRP have a higher survival ratethan control mice when injected with gram-positive bacteria orwith LPS [34, 35]. In the latter model, the protective mecha-nism was attributed, at least in part, to the augmentation of

Fig. 6. Cytokine levels in plasma of LPS-injected mice. Wt (black bars) andTSG-14 Tg mice (Tg2 line, gray bars) were left untreated or were injected i.p.with 11 mg/kg LPS, and plasma was collected at the indicated time-points (fiveanimals for each time-point). The levels of IL-10, TNF, and IL-6 were mea-sured by ELISA using a kit from R&D Systems (for TNF) or reagents fromPharMingen (IL-10 and IL-6), following instructions of the manufacturers. Datarepresent the average for each group.

Fig. 5. Production of NO by macrophages from Wt and TSG-14 Tg mice.Macrophages (53104 cells/well) from Wt and Tg4 mice were treated withIFN-g, LPS, and TNF-a alone or with IFN-g 1 LPS and IFN-g 1 TNF-a atthe doses shown in the figure. The bars represent the amounts (mM) of NO2

2

detected by the Griess assay in the supernatant of the cultures collected 48 hafter treatment.


complement-mediated phagocytosis or to a reduced, LPS-in-duced, cytokine-gene expression by macrophages [34]. Ourobservation that TSG-14 Tg mice have a higher survival ratethan Wt mice when challenged with LPS is in agreement witha role for TSG-14 as an acute-phase protein that, like CRP, hasa potential, protective role in LPS-induced shock.TSG-14 Tgmice also showed an improved survival after CLP-induced,polymicrobial infection (Fig. 4).

It remains to be shown how TSG-14 affects the mechanismsof the inflammatory response and exerts its protective effect inLPS and CLP models. The higher levels of IL-10 observed inuntreated or in the very early stage of the LPS-induced re-sponse (Fig. 6) could be part of this mechanism. The beneficialrole of IL-10 during sepsis has been documented widely. Forexample, it has been shown that injection of IL-10 can de-crease lethality in endotoxemia [36] or in sepsis [37]. Theprotective mechanism of IL-10 could be related to its ability toinhibit activation of coagulatin and fibrinolysis during endo-toxemia [38] and reduced production of pro-inflammatory cy-tokines (reviewed in [39]). Thus, the fact that TSG-14 Tg micehave augmented levels of IL-10 even prior to LPS injectionmight be related to its higher resistance to LPS and CLP.

Also, TSG-14 Tg mice have augmented levels of TNF, atleast during the early phase of the LPS-induced response. It isthus possible that, during this early phase, higher production ofTNF confers a beneficial advantage for survival by activatingpro-inflammatory activities. It has been shown that injection ofTNF can reduce lethality in the CLP model [40], and it wassuggested by Echtenacher and co-workers [41] that during theearly phase of peritonitis, endogenous TNF may stimulatenonlymphoid cells such as granulocytes, macrophages, plate-lets, and fibroblasts to ingest bacteria and localize inflamma-tion.

Available data suggest several other possibilities includingthe demonstrated capacity of TSG-14 protein to bind the C1qcomplement component [21] and as we show here, augmentedproduction of NO by macrophages derived from TSG-14 Tgmice (Fig. 5). Activation of complement is essential for theinitial containment of systemic, bacterial infection, as demon-strated by the susceptibility of complement-deficient mice tobacterial infection [42, 43]. Macrophages play a central role inthe inflammatory response and are targets for several pro-inflammatory cytokines and acute-phase proteins. The pen-traxin CRP, for example, can modulate gene expression inmacrophages, leading to an increased production of pro-in-flammatory cytokines [44, 45] or their antagonists [46]. Pro-duction of NO by activated macrophages affects a multitude ofbiological functions [47, 48].

Macrophages from Tg2 and Tg4 mice produced higheramounts of NO than control, Wt mice, when stimulated withIFN-g alone or with IFN-g in combination with LPS or TNF-a(Fig. 5). It appears that the major difference between Wt andTg macrophages is a significantly greater responsiveness of thelatter cells to IFN-g. It was demonstrated recently that IFN-gcan inhibit the expression of TSG-14 in monocytes and mac-rophages [19, 20]. Thus, it is possible that TSG-14 leads to ahigher responsiveness of cells to IFN-g, which, in turn, couldinhibit TSG-14 expression, leading to a suppression of theinflammatory response.

In infection as well as inflammation, NO generated bymacrophages appears to act as a direct effector and also asa regulator of other effector molecules [49]. Experimentsusing NO inhibitors implicated NO in anti-microbial, in-flammatory responses. However, whether NO is beneficial ordetrimental in endotoxin-induced shock is still controver-sial. Mice lacking inducible NO synthase (iNOS) are moresusceptible to infection by Listeria [50] and Leishmania [51]as well as to sepsis induced by the CLP procedure [52].However, it has been shown that, in spite of being a medi-ator of LPS-induced, endotoxic shock, NO can also protectmice against LPS because mice deficient in iNOS wereshown to be more resistant to endotoxin-induced shock [50,51, 53]. In addition, mice overexpressing the endothelialNO synthase (eNOS) as a transgene show a higher resistanceto the lethal effects of LPS than control mice [54]; theprotective effect of NO was ascribed to a reduced, vascularreactivity to NO. It was also shown that the deletion of eNOSresults in increased myocardial dysfunction following isch-emia-reperfusion [55] and that NO synthesis during endo-toxemia is associated with prevention of hepatic damage andintravascular thrombosis [56]. Other evidence of an anti-inflammatory effect of NO rests on the ability of NO to limitendothelial activation and inhibit leukocyte adhesion [57].In addition, it was demonstrated recently that NO canincrease the shedding of a soluble form of TNF receptor 1[58] as well as other cytokines, cytokine receptors, andadhesion molecules by the activation of TNF-convertingenzyme mediated ectodomain shedding [59]; that the iNOS-derived NO mediates protective activities indispensable tosurvive a TNF challenge [60]; and that iNOS gene functionprovides a survival benefit in septic mice [52].

Thus, although it may be premature to conclude that thegreater resistance to endotoxin and to sepsis of the TSG-14 Tgmice is caused by an increased production of NO by peritonealmacrophages, these two features are not paradoxical necessar-ily. It has been pointed out that NO can deliver death- andlife-promoting messages [49, 60]. It is well possible that in theTSG-14 Tg mice, the protective role prevails.

Although the precise, molecular aspects of TSG-14 functionremain to be elucidated, our data represent the first in vivoevidence for a role of TSG-14 as an important component of theinnate, immune response and as part of the host mechanismsthat control endotoxemia and bacterial infection.

ACKNOWLEDGMENTS

This work was supported financially by grants from PADCT/CNPq, FAPEMIG, FAPESP PRONEX, and CA75071 from theNational Institutes of Health. We are grateful to Stavros Gian-nakopoulos for helping with the murine TSG-14 gene cloning.We thank Ines Nobuko Nishimoto and Paulo Cesar Maciag forhelping with the statistical analysis, Domingos Savio for animalcare, and all members of our laboratories for helpful discus-sions. We also thank Ms. Ilene M. Totillo for assistance withthe preparation of the manuscript.


REFERENCES

1. Lee, T. H., Lee, G. W., Ziff, E. B., Vilcek, J. (1990) Isolation andcharacterization of eight tumor necrosis factor-induced gene sequencesfrom human fibroblasts. Mol. Cell. Biol. 10, 1982–1988.

2. Breviario, F., d’Aniello, E. M., Golay, J., Peri, G., Bottazzi, B., Bairoch, A.,Saccone, S., Marzella, R., Predazzi, V., Rocchi, M. (1992) Interleukin-1-inducible genes in endothelial cells. Cloning of a new gene related toC-reactive protein and serum amyloid P component. J. Biol. Chem. 267,22190–22197.

3. Altmeyer, A., Klampfer, L., Goodman, A. R., Vilcek, J. (1995) Promoterstructure and transcriptional activation of the murine TSG-14 gene en-coding a tumor necrosis factor/interleukin-1-inducible pentraxin protein.J. Biol. Chem. 270, 25584–25590.

4. Introna, M., Alles, V. V., Castellano, M., Picardi, G., De Gioia, L.,Bottazzai, B., Peri, G., Breviario, F., Salmona, M., De Gregorio, L.,Dragani, T. A., Srinivasan, N., Blundell, T. L., Hamilton, T. A., Man-tovani, A. (1996) Cloning of mouse ptx3, a new member of the pentraxingene family expressed at extrahepatic sites. Blood 87, 1862–1872.

5. Lee, G. W., Lee, T. H., Vilcek, J. (1993) TSG-14, a tumor necrosis factor-and IL-1-inducible protein, is a novel member of the pentaxin family ofacute phase proteins. J. Immunol. 150, 1804–1812.

6. Osmand, A. P., Friedenson, B., Gewurz, H., Painter, R. H., Hofmann, T.,Shelton, E. (1977) Characterization of C-reactive protein and the comple-ment subcomponent C1t as homologous proteins displaying cyclic pen-tameric symmetry (pentraxins). Proc. Natl. Acad. Sci. USA 74, 739–743.

7. Goodman, A. R., Cardozo, T., Abagyan, R., Altmeyer, A., Wisniewski,H. G., Vilcek, J. (1996) Long pentraxins: an emerging group of proteinswith diverse functions. Cytokine Growth Factor Rev. 7, 191–202.

8. Seery, L. T., Schoenberg, D. R., Barbaux, S., Sharp, P. M., Whitehead,A. S. (1993) Identification of a novel member of the pentraxin family inXenopus laevis. Proc. R. Soc. Lond. B Biol. Sci. 253, 263–270.

9. Reid, M. S., Blobel, C. P. (1994) Apexin, an acrosomal pentaxin. J. Biol.Chem. 269, 32615–32620.

10. Noland, T. D., Friday, B. B., Maulit, M. T., Gerton, G. L. (1994) The spermacrosomal matrix contains a novel member of the pentaxin family ofcalcium-dependent binding proteins. J. Biol. Chem. 269, 32607–32614.

11. Schlimgen, A. K., Helms, J. A., Vogel, H., Perin, M. S. (1995) Neuronalpentraxin, a secreted protein with homology to acute phase proteins of theimmune system. Neuron 14, 519–526.

12. Dodds, D. C., Omeis, I. A., Cushman, S. J., Helms, J. A., Perin, M. S.(1997) Neuronal pentraxin receptor, a novel putative integral membranepentraxin that interacts with neuronal pentraxin 1 and 2 and taipoxin-associated calcium-binding protein 49. J. Biol. Chem. 272, 21488–21494.

13. Hsu, Y. C., Perin, M. S. (1995) Human neuronal pentraxin II (NPTX2):conservation, genomic structure, and chromosomal localization. Genomics28, 220–227.

14. Tsui, C. C., Copeland, N. G., Gilbert, D. J., Jenkins, N. A., Barnes, C.,Worley, P. F. (1996) Narp, a novel member of the pentraxin family,promotes neurite outgrowth and is dynamically regulated by neuronalactivity. J. Neurosci. 16, 2463–2478.

15. Luchetti, M. M., Piccinini, G., Mantovani, A., Peri, G., Matteucci, C.,Pomponio, G., Fratini, M., Fraticelli, P., Sambo, P., Di Loreto, C., Doni,A., Introna, M., Gabrielli, A. (2000) Expression and production of the longpentraxin PTX3 in rheumatoid arthritis (RA). Clin. Exp. Immunol. 119,196–202.

16. Alles, V. V., Bottazzi, B., Peri, G., Golay, J., Introna, M., Mantovani, A.(1994) Inducible expression of PTX3, a new member of the pentraxinfamily, in human mononuclear phagocytes. Blood 84, 3483–3493.

17. Lee, G. W., Goodman, A. R., Lee, T. H., Vilcek, J. (1994) Relationship ofTSG-14 protein to the pentraxin family of major acute phase proteins.J. Immunol. 153, 3700–3707.

18. Vouret-Craviari, V., Matteucci, C., Peri, G., Poli, G., Introna, M., Man-tovani, A. (1997) Expression of a long pentraxin, PTX3, by monocytesexposed to the mycobacterial cell wall component lipoarabinomannan.Infect. Immun. 65, 1345–1350.

19. Polentarutti, N., Picardi, G., Basile, A., Cenzuales, S., Rivolta, A., Mat-teucci, C., Peri, G., Mantovani, A., Introna, M. (1998) Interferon-gammainhibits expression of the long pentraxin PTX3 in human monocytes. Eur.J. Immunol. 28, 496–501.

20. Goodman, A. R., Levy, D. E., Reis, L. F., Vilcek, J. (2000) Differentialregulation of TSG-14 expression in murine fibroblasts and peritonealmacrophages. J. Leukoc. Biol. 67, 387–395.

21. Bottazzi, B., Vouret-Craviari, V., Bastone, A., De Gioia, L., Matteucci, C.,Peri, G., Spreafico, F., Pausa, M., D’Ettorre, C., Gianazza, E., Tagliabue,A., Salmona, M., Tedesco, F., Introna, M., Mantovani, A. (1997) Multimerformation and ligand recognition by the long pentraxin PTX3. Similarities

and differences with the short pentraxins C-reactive protein and serumamyloid P component. J. Biol. Chem. 272, 32817–32823.

22. Laird, P. W., Zijderveld, A., Linders, K., Rudnicki, M. A., Jaenisch, R.,Berns, A. (1991) Simplified mammalian DNA isolation procedure. NucleicAcids Res. 19, 4293.

23. Church, G. M., Gilbert, W. (1984) Genomic sequencing. Proc. Natl. Acad.Sci. USA 81, 1991–1995.

24. Fort, P., Marty, L., Piechaczyk, M., el Sabrouty, S., Dani, C., Jeanteur, P.,Blanchard, J. M. (1985) Various rat adult tissues express only one majormRNA species from the glyceraldehyde-3-phosphate-dehydrogenase mul-tigenic family. Nucleic Acids Res. 13, 1431–1442.

25. Kaplan, E. L., Meier, P. (1958) Nonparametric estimation from incompleteobservations. J. Am. Stat. Assoc. 53, 457–481.

26. Wichterman, K. A., Chaudry, I. H., Baue, A. E. (1979) Studies of periph-eral glucose uptake during sepsis. Arch. Surg. 114, 740–745.

27. Hibbs Jr., J. B., Taintor, R. R., Vavrin, Z., Rachlin, E. M. (1988) Nitricoxide: a cytotoxic activated macrophage effector molecule. Biochem. Bio-phys. Res. Commun. 157, 87–94.

28. Mosmann, T. (1983) Rapid colorimetric assay for cellular growth andsurvival: application to proliferation and cytotoxicity assays. J. Immunol.Methods 65, 55–63.

29. Wichterman, K. A., Baue, A. E., Chaudry, I. H. (1980) Sepsis and septicshock—a review of laboratory models and a proposal. J. Surg. Res. 29,189–201.

30. Unanue, E. R. (1984) Antigen-presenting function of the macrophage.Annu. Rev. Immunol. 2, 395–428.

31. Worrall, N. K., Chang, K., LeJeune, W. S., Misko, T. P., Sullivan, P. M.,Ferguson Jr., T. B., Williamson, J. R. (1997) TNF-alpha causes reversiblein vivo systemic vascular barrier dysfunction via NO-dependent and-independent mechanisms. Am. J. Physiol. 273, H2565–H2574.

32. Dinarello, C. A. (1996) Cytokines as mediators in the pathogenesis ofseptic shock. Curr. Top. Microbiol. Immunol. 216, 133–165.

33. Lowry, S. F. (1994) Sepsis and its complications: clinical definitions andtherapeutic prospects. Crit. Care Med. 22, S1–S2.

34. Xia, D., Samols, D. (1997) Transgenic mice expressing rabbit C-reactiveprotein are resistant to endotoxemia. Proc. Natl. Acad. Sci. USA 94,2575–2580.

35. Szalai, A. J., Agrawal, A., Greenhough, T. J., Volanakis, J. E. (1999)C-reactive protein: structural biology and host defense function. Clin.Chem. Lab. Med. 37, 265–270.

36. Howard, M., Muchamuel, T., Andrade, S., Menon, S. (1993) Interleukin 10protects mice from lethal endotoxemia. J. Exp. Med. 177, 1205–1208.

37. Kato, T., Murata, A., Ishida, H., Toda, H., Tanaka, N., Hayashida, H.,Monden, M., Matsuura, N. (1995) Interleukin 10 reduces mortality fromsevere peritonitis in mice. Antimicrob. Agents Chemother. 39, 1336–1340.

38. Pajkrt, D., Van der, P. T., Levi, M., Cutler, D. L., Affrime, M. B., van den,E. A., ten Cate, J. W., van Deventer, S. J. (1997) Interleukin-10 inhibitsactivation of coagulation and fibrinolysis during human endotoxemia.Blood 89, 2701–2705.

39. Malefyt, R. (2001) IL-10. In Cytokine Reference, J. J. Oppenheim and M.Feldmann, eds. Academic Press: San Diego, CA; pp. 165–185.

40. Sheppard, B. C., Fraker, D. L., Norton, J. A. (1989) Prevention andtreatment of endotoxin and sepsis lethality with recombinant human tumornecrosis factor. Surgery 106, 156–161.

41. Echtenacher, B., Falk, W., Mannel, D. N., Krammer, P. H. (1990) Re-quirement of endogenous tumor necrosis factor/cachectin for recoveryfrom experimental peritonitis. J. Immunol. 145, 3762–3766.

42. Prodeus, A. P., Zhou, X., Maurer, M., Galli, S. J., Carroll, M. C. (1997)Impaired mast cell-dependent natural immunity in complement C3-defi-cient mice. Nature 390, 172–175.

43. Wessels, M. R., Butko, P., Ma, M., Warren, H. B., Lage, A. L., Carroll,M. C. (1995) Studies of group B streptococcal infection in mice deficientin complement component C3 or C4 demonstrate an essential role forcomplement in both innate and acquired immunity. Proc. Natl. Acad. Sci.USA 92, 11490–11494.

44. Ballou, S. P., Lozanski, G. (1992) Induction of inflammatory cytokinerelease from cultured human monocytes by C-reactive protein. Cytokine 4,361–368.

45. Galve-de Rochemonteix, B., Wiktorowicz, K., Kushner, I., Dayer, J. M.(1993) C-reactive protein increases production of IL-1 alpha, IL-1 beta,and TNF-alpha, and expression of mRNA by human alveolar macro-phages. J. Leukoc. Biol. 53, 439–445.

46. Tilg, H., Vannier, E., Vachino, G., Dinarello, C. A., Mier, J. W. (1993)Antiinflammatory properties of hepatic acute phase proteins: preferentialinduction of interleukin 1 (IL-1) receptor antagonist over IL-1 beta synthesisby human peripheral blood mononuclear cells. J. Exp. Med. 178, 1629–1636.


47. Nathan, C., Shiloh, M. U. (2000) Reactive oxygen and nitrogen interme-diates in the relationship between mammalian hosts and microbial patho-gens. Proc. Natl. Acad. Sci. USA 97, 8841–8848.

48. MacMicking, J., Xie, Q. W., Nathan, C. (1997) Nitric oxide and macro-phage function. Annu. Rev. Immunol. 15, 323–350.

49. Nathan, C. (1997) Inducible nitric oxide synthase: what difference does itmake? J. Clin. Invest 100, 2417–2423.

50. MacMicking, J. D., Nathan, C., Hom, G., Chartrain, N., Fletcher, D. S.,Trumbauer, M., Stevens, K., Xie, Q. W., Sokol, K., Hutchinson, N. (1995)Altered responses to bacterial infection and endotoxic shock in micelacking inducible nitric oxide synthase. Cell 81, 641–650.

51. Wei, X. Q., Charles, I. G., Smith, A., Ure, J., Feng, G. J., Huang, F. P.,Xu, D., Muller, W., Moncada, S., Liew, F. Y. (1995) Altered immuneresponses in mice lacking inducible nitric oxide synthase. Nature 375,408–411.

52. Cobb, J. P., Hotchkiss, R. S., Swanson, P. E., Chang, K., Qiu, Y., Laubach,V. E., Karl, I. E., Buchman, T. G. (1999) Inducible nitric oxide synthase(iNOS) gene deficiency increases the mortality of sepsis in mice. Surgery126, 438–442.

53. Huang, P. L. (2000) Lessons learned from nitric oxide synthase knockoutanimals. Semin. Perinatol. 24, 87–90.

54. Yamashita, T., Kawashima, S., Ohashi, Y., Ozaki, M., Ueyama, T., Ishida,T., Inoue, N., Hirata, K., Akita, H., Yokoyama, M. (2000) Resistance to

endotoxin shock in transgenic mice overexpressing endothelial nitricoxide synthase. Circulation 101, 931–937.

55. Hannan, R. L., Hack, B. D., Matherne, G. P., Laubach, V. E. (2000)Deletion of endothelial nitric oxide synthase exacerbates myocardial stun-ning in an isolated mouse heart model. J. Surg. Res. 93, 127–132.

56. Harbrecht, B. G., Billiar, T. R., Stadler, J., Demetris, A. J., Ochoa, J.,Curran, R. D., Simmons, R. L. (1992) Inhibition of nitric oxide synthesisduring endotoxemia promotes intrahepatic thrombosis and an oxygenradical-mediated hepatic injury. J. Leukoc. Biol. 52, 390–394.

57. Masamune, A., Shimosegawa, T., Satoh, A., Fujita, M., Sakai, Y., Toyota,T. (2000) Nitric oxide decreases endothelial activation by rat experimentalsevere pancreatitis-associated ascitic fluids. Pancreas 20, 297–304.

58. Okuyama, M., Yamaguchi, S., Yamaoka, M., Nitobe, J., Fujii, S., Yo-shimura, T., Tomoike, H. (2000) Nitric oxide enhances expression andshedding of tumor necrosis factor receptor I (p55) in endothelial cells.Arterioscler. Thromb. Vasc. Biol. 20, 1506–1511.

59. Zhang, Z., Kolls, J. K., Oliver, P., Good, D., Schwarzenberger, P. O., Joshi,M. S., Ponthier, J. L., Lancaster Jr., J. R. (2000) Activation of tumornecrosis factor-alpha-converting enzyme-mediated ectodomain sheddingby nitric oxide. J. Biol. Chem. 275, 15839–15844.

60. Cauwels, A., Van Molle, W., Janssen, B., Everaerdt, B., Huang, P., Fiers,W., Brouckaert, P. (2000) Protection against TNF-induced lethal shock bysoluble guanylate cyclase inhibition requires functional inducible nitricoxide synthase. Immunity 13, 223–231.


TSG-14 transgenic mice have improved survival to endotoxemia and to CLP-induced sepsis

Documents