Original Article

Expression of scavenger receptor class A and CD14 inlipopolysaccharide-induced lung injury

Takashi Yamamoto,1 Yusuke Ebe,1 Go Hasegawa,1 Masashi Kataoka,2 Shunsuke Yamamoto2 andMakoto Naito1

1Second Department of Pathology, Niigata University School of Medicine, Niigata and 2Department of Pathology, OitaMedical University, Oita, Japan

Macrophage scavenger receptor class A type I and II(MSR-A) was cloned as a receptor for the recognition anduptake of a wide range of negatively charged macromole-cules such as chemically modified low-density lipopro-teins.12–14 MSR-A has been demonstrated to be involved inthe pathogenesis of atherosclerosis.12–16 Furthermore, MSR-A mediates the recognition of lipid A domain of LPS17,18 andGram-positive bacteria19 and plays a protective role in hostdefense mechanism against Listeria monocytogenes andHerpes simplex virus.20 However, the role of MSR-A in LPS-induced tissue injury has not been well understood.

LPS administration causes necrosis and apoptosis inseveral organs.21–23 The LPS-induced tissue damage isadministration route and dose dependent. LPS is usuallyadministrated intraperitoneally or intravenously and CD14expression is induced in several organs. However, inhalationof pathogens is a probable route that causes pneumonia andalveolar damage. Therefore, we have developed a method togive LPS as an aerosol by using a nebulizer. Using thismethod our study was designed to clarify the expression ofCD14 and MSR-A in the LPS-exposed murine lung, in com-bination with the expression of several pro-inflammatorycytokines.

Macrophage colony stimulating factor (M-CSF) is animportant hematopoietic growth factor that stimulates devel-opment, differentiation, and proliferation of macrophages andtheir precursors.24 M-CSF regulates macrophage functionsby inducing the production and secretion of several bioactivesubstances. M-CSF is reported to be one of the importantregulators in LPS-induced inflammatory reactions.25 It hasalso been reported that M-CSF differentially sensitizemacrophage populations to LPS.26 However, the role of M-CSF for the expression of CD14 and MSR-A has not beenclarified in LPS-induced tissue injury. This study also aims toclarify the role of M-CSF in CD14 and MSR-A expressions invivo.

Pathology International 1999; 49: 983–992

CD14 and macrophage scavenger receptor class A type Iand II (MSR-A) are receptors for lipopolysaccharide (LPS). Inthis study, the expressions of both receptors in the lungafter administration of LPS in aerosol to mice with a nebu-lizer were observed. Bronchiolar epithelial cells and alveolarmacrophages immediately incorporated LPS and expressedCD14. CD14-positive neutrophils then appeared in the alveo-lar space followed by the appearance of MSR-A-expressingcells in the vascular lumen, pulmonary interstitium, andalveolar space. Numbers of apoptotic cells increased after 1day, and MSR-A-expressing macrophages actively incorpo-rated apoptotic bodies. Daily administration of macrophagecolony stimulating factor (M-CSF) to the mice resulted inincreased levels of MSR-A expression and reduced levels ofCD14 as well as several cytokine expressions, leading toshortening of the inflammatory process. The numbers ofapoptotic cells were reduced in M-CSF injected mice. Thesefindings imply that CD14 acts as an immediate expressingreceptor for LPS and MSR-A exerts a protective function byscavenging LPS and apoptotic cells in LPS-induced lunginjury.

Key words: CD14, lipopolysaccharide, lung, macrophagecolony stimulating factor, macrophages, scavenger receptor

Lipopolysaccharide (LPS) is a major component of the outersurface of Gram-negative bacteria.1 The lipid A domain ofLPS is a particularly potent activator for macrophages andleukocytes, and induces these cells to produce and releasevarious cytokines and oxygen radicals1,2 which contribute tothe pathophysiology of LPS-induced tissue damage. It hasbeen shown that CD14 is a receptor for LPS/LPS-bindingprotein complexes and is responsible for transmembrane signaling.2–11

Correspondence: Makoto Naito, MD, Second Department of Pathology, Niigata University, School of Medicine, Asahimachi-dori 1,Niigata 951-8510, Japan. Email: 2byori@med.niigata-u.ac.jp

Received 18 March 1999. Accepted for publication 6 July 1999.

MATERIALS AND METHODS

Animals

BALB/c mice were purchased (Charles River Inc. Japan,Tokyo, Japan) and maintained under routine conditions at theLaboratory Animal Center of Niigata University School ofMedicine. Eight-week-old female mice were kept in a woodenbox (30 � 20 � 20 cm) and exposed to an aerosol of LPSsolutions at various concentrations (10 µg/mL, 100 µg/mL, 1 mg/mL, 10 mg/mL) produced by a nebulizer (Pulmo-Aide,Devilbiss, Somerset, USA) for 30 min. Lipopolysaccharide(10 mL) solutions were evaporated in 30 min. At 2 and 8 hand 1, 2, 3, 5, 7 days after inhalation, they were killed byether anesthesia and lungs were obtained. LPS was used ata concentration of 1 mg/mL in the main experiment. A groupof mice were injected with 5 µg recombinant human M-CSF(Morinaga Milk Industry Co, Tokyo, Japan) daily from day 0.Three mice were used at each time point.

Alveolar lavage

Cells in the alveolar space and airways were obtained bybronchoalveolar lavage. In brief, after taking both lungs andbronchotracheal trees together, a silicon tube with a diameterof 0.5 mm connected to a needle (27 G) was introduced to the trachea of the mouse. About 0.5 mL phosphate-buffered saline (PBS) was injected and aspirated. Numbersof cells were calculated and cytospin specimens were made.

Monoclonal antibodies

The monoclonal antibody, F4/80, recognizes an antigenpresent on the surface membrane of macrophages and somemonocytic cells.27 A monoclonal antibody designated rmC5–3was used as antimurine CD14 antibody.10 The monoclonalantibody, 2F8, recognizes murine MSR-A.28

Immunohistochemistry and immunohistochemicaldouble staining

The lungs were fixed in periodate-lysine-paraformaldehyde(PLP) at 4°C for 4 h, washed with phosphate buffer solutioncontaining 10, 15, 20% sucrose for 4 h, embedded in OCTcompound (Miles, Elkhart, USA), frozen in dry ice-acetoneand cut by a cryostat (Bright, Huntington, UK) into 6 µm-thicksections. Cytospin specimens of cells obtained by bron-choalveolar lavage were fixed in acetone for 5 min. After inhi-bition of endogenous peroxidase activity by the method of Isobe et al.,29 we performed immunohistochemistry using

the antimouse monoclonal antibodies as described above. As a secondary antibody, we used antirat Ig-horseradish peroxidase-linked F(ab)2 fragment (Amersham, Poole, UK).After visualization with 3,3�-diaminobenzidine (Dojin Chemi-cal Co., Kumamoto, Japan), nuclear staining with methylenegreen or hematoxylin and mounting with resin, the positivecells with nuclei per 1 mm2 were counted using a light micro-scope. Apoptotic cells were detected in sections by the directimmunoperoxidase method using the Apop Tag in situ apop-tosis detection kit (Oncor, Gaithersburg, USA).

Immunohistochemical double staining with CD14 and 2F8,F4/80 and CD14, and F4/80 and 2F8 was performed asdescribed with a minor modification.30 In brief, after inhibitingendogenous peroxidase activity,29 cryostat sections wereincubated with the first primary monoclonal antibody. Afterincubation with antirat Ig-horseradish peroxidase-linkedF(ab)2 fragment, the reaction was stained brown with 3,3�-diaminobenzidine. The sections were washed twice withglycine-HCl buffer for 1 h, then incubated with the secondprimary monoclonal antibody. After incubation with antirat Ig-horseradish peroxidase-linked F(ab)2 fragment, they were incubated with a nickel chloride solution in 3,3�-diaminobenzidine substrate kit (Vector Lab, Inc., Burlingame,USA) and processed as above to stain positive cells blueblack.

Histochemistry

Tissues were fixed in 10% phosphate-buffered formalin andprocessed routinely for paraffin sections. Sections were pre-pared, 5 µm-thick, deparaffinized, and hydrated prior to incu-bation with the staining solution. Sections were stained withhematoxylin and eosin. Neutrophils were stained using amodified AS-D chloroacetate method. The slides were incu-bated in the staining solution and counterstained with hema-toxylin. Neutrophils and mast cells were stained red by thismethod and the discrimination of neutrophils from mast cellswas based both on the cell and nuclear morphology.

Ribonucleic acid (RNA) isolation and messenger RNAanalysis by reverse transcriptase-polymerase chainreaction

At 2, 8 h, 1, 2, 3, 5, and 7 days after inhalation of LPS, thelung was removed and homogenized. Total cellular ribonu-cleic acid (RNA) was isolated by phenol-chloroform extrac-tion.30 Polymerase chain reaction (PCR) amplification wasperformed using a TP cycler-100 (Toyobo, Osaka, Japan). Allthe PCR primers were made to order by Kurabo Biomedicals(Osaka, Japan). The oligonucleotides used are shown inTable 1. The samples were separated on a 1.5% low melting

984 T. Yamamoto et al.

agarose gel containing 0.3 mg/mL (0.003%) of ethidiumbromide and bands were visualized and photographed usingultraviolet transillumination.

Northern blot analysis

Northern blot hybridization was performed as previouslydescribed.10 Briefly, total RNA prepared from tissues waselectrophoresed through a 1.5% agarose-6% (v/v) formalde-hyde gel and blotted onto a nylon membrane. The mem-branes were exposed to ultraviolet for 7 min and thenprehybridized and hybridized with 3–5X 106 c.p.m./mL of 32P-labeled RNA probes prepared from mCD14 cDNA MS7X.3,4

Relative expression of mCD14 message measured usingBAS1000 bioimaging analyzer (Fuji Film, Tokyo, Japan) wasdetermined after normalization to levels of �-actin mRNA.

Statistics

The significance of the data was evaluated by Student’s t-test.

RESULTS

Changes in the numbers of cells obtained bybronchoalveolar lavage

Eight hours after inhalation of aerosol of LPS solutions atvarious concentrations, a great number of leukocytes werefound in the bronchoalveolar lavage fluid (BALF) of micewhich had inhaled LPS at the concentration of more than 100 µg/mL (Fig. 1a), and more than 90% of the cells wereneutrophils. A time course of BALF analysis after inhalation of

Scavenger receptor and CD14 in the lung 985

Table 1 Oligonucleotides used

Primer (5�–3�) Products (b.p.)

c-fms Sense GGA TAG AGA CCC ACC ATG AA 247Antisense CAG TAG CAC CAG CAG AGA CA

MSR-A Sense CCAAGT CCT TGC AGA TGC TG 236Antisense GCTC TGA GGT CGT TGG TGA TG

M-CSF Sense ACT GTA GCC ACA TGA TTG G 410Antisense GCT GTT GTT GCA GTT CTT G

GM-CSF Sense GAA ACG GAC TGT GAA ACA CA 230Antisense CTG TGG CTG TGC CAC ATC TC

TNF-α Sense GGC AGG TCT ACT TTG GAG TCA TTG C 309Antisense ACA TTC GAG GCT CCT GTG AAT TCG G

IL-1α Sense CTC TAG AGC ACC ATG CTA CAG AC 309Antisense TGG ATT CCA GGG GAA ACA CTG

�-actin Sense TGG ATT CCT GTG GCA TCC ATG AAA C 348Antisense TAA AAC GCA GCT CAG TAA CAG TCC G

MSR-A, macrophage scavenger receptor class A type I and II; M-CSF, macrophage colony stimulating factor; GM-CSF, granulocyte macrophagecolony stimulating factor; IL, interleukin.

LPS at the concentration of 1 mg/mL is shown in Fig. 1(b).The leukocyte numbers increased at 4 h and reached a peakat 48 h. There was only a slight increase of macrophages upto 24 h, but their number increased at 48 h. Apoptotic cellswere frequently encountered in lavaged cells between days 2 and 5.

Changes in the numbers of cells in the lung

The changes in the numbers of cells in the lung after inhala-tion of LPS aerosol are shown in Fig. 2. The leukocytenumbers increased at 8 h and reached a peak at 48 h. Thenumber of neutrophils increased rapidly and reached a peakat day 2. There was a gradual increase of macrophages up to3 days, but their number decreased at 5 days. The numbersof T and B lymphocytes were small in the lung as well as inBALF throughout the experimental period. There was a dis-crepancy between BALF and lung tissues in the ratio ofmacrophages and neutrophils.

CD14- and MSR-A-expressing cells in the lung

There were only a few macrophages in the lungs of normaluntreated mice. Antibodies against murine macrophages andMSR-A, F4/80 and 2F8, reacted with macrophages in thealveolar space and peribronchiolar connective tissue in thelungs. Leukocytes including monocytes within the lumen ofblood vessels showed no positive reaction against both anti-bodies. By immunohistochemical double staining, alveolarmacrophages were double positive for 2F8 and F4/80. CD14-positive cells were absent.

Two hours after LPS (1 mg/mL) inhalation, bronchiolarepithelial cells expressed CD14 (Fig. 3a) and there was a

rapid increase of CD14-positive cells in the alveolar space.Immunohistochemical staining of cells in BALF showed thatmost of the CD14-positive cells were neutrophils at 2 h (Fig. 3a, inset), but CD-14-positive macrophages increasedprogressively with time. The number of MSR-A-expressingcells was not increased at 2 h, but increased in alveolarspace, peribronchiolar tissue, pericapillary or perivenousareas (Fig. 3b), and the BALF (Fig. 3b, inset) at 8 h. MSR-A-expressing cells were also found in the blood vessels, oftenattaching to the endothelial cells and migrating through thevascular wall (Fig. 3c). At this stage, CD14 expression inbronchiolar epithelial cells was prominent. At days 1 and 2,the number of CD14-positive cells gradually decreased andthe expression of CD14 by bronchiolar epithelial cells dimin-

ished. In contrast, the number of MSR-A-expressingmacrophages increased up to day 3 (Fig. 3d) and ingestedapoptotic bodies (Fig. 3e). Aggregates of MSR-A-expressingcells were often found in the bronchiolar lumen from day 3 to5 (Fig. 3f). After 5 days, MSR-A-expressing cells decreasedin number and staining intensity.

By immunohistochemical double staining, the number ofCD14 single positive cells increased abruptly and peaked at2–8 h. CD14 and F4/80 double positive cells graduallyincreased up to day 2. Almost all CD14 single positive cellswere morphologically determined to be neutrophils. Thedouble positive populations were regarded as macrophagesand their number peaked at day 2 (Fig. 4a).

Comparing the expression of MSR-A and CD14, thenumber of CD14 single positive cells first increased, followedby an increase of MSR-A single positive cells. CD14 and MSR-A double positive cells were a minor populationthroughout the experimental period (Fig. 4b).

Apoptosis induced by LPS

In situ labeling of fragmented DNA was observed in somecells in the lungs from 1 day after LPS inhalation. Most of the apoptotic cells did not show any immunophenotypes as macrophages, but some of them were CD14 positive.Nuclear fragments of apoptotic cells were often incorporatedby MSR-A-expressing macrophages (Fig. 3e, inset). Thenumber of apoptotic cells increased up to day 3 anddecreased afterwards. Apoptotic cells were often incorpo-rated by macrophages in the bronchiolar lumen (Fig. 3f). Thenumbers of MSR-A-expressing macrophages phagocytizingapoptotic cells were increased in parallel with those of

986 T. Yamamoto et al.

Figure 1 (a) Numbers of cells in bronchoalveolar lavage fluid at 8 h after inhalation of various concentrations of lipopolysaccharide(LPS). Cell numbers increase at a concentration of more than 100 µg/mL. (a): ( ), total cell count; (�), polymorphonuclear leuko-cytes; (�), macrophage. (b) Changes in the numbers of cells in bron-choalveolar lavage fluid after inhalation of LPS at a concentration of1 mg/mL. (b): (�), total cell count; (�), polymorphonuclear leuko-cytes; (�), macrophage; (�), lymphocytes. Data are expressed asmean � SD.

Figure 2 Changes in the numbers of cells in the lung after inhala-tion of lipopolysaccharide at a concentration of 1 mg/mL. (�), F4/80,total cell count; (�), polymorphonuclear leukocytes; (�), lympho-cytes. Data are expressed as mean � SD.

Scavenger receptor and CD14 in the lung 987

Figure 3 (a–d) Immunohistochemical double staining using anti-CD14 monoclonal antibody and anti-scavenger receptor monoclonal anti-body 2F8. (a) Expression of CD14 at 2 h after the administration of lipopolysaccharide (LPS). Bronchiolar epithelial cells express CD14 (brown).CD14-positive cells (brown) are more abundant than macrophage scavenger receptor class A type I and II (MSR-A)-positive cells (black), �100.Inset: Expression of CD14 among cells in bronchoalveolar lavage fluid. CD14 is expressed in most neutrophils and a macrophage (arrow).Immunohistochemical staining using anti-CD14 antibody, �800. (b) Expression of MSR-A at 8 h after LPS inhalation. The number of MSR-A-expressing cells (black) was increased in the lung, especially in the pericapillary and peribronchiolar areas. CD14 expression (brown) is promi-nent in bronchiolar epithelial cells, �100. Inset: MSR-A is expressed only in macrophages in bronchoalveolar lavage fluid.Immunohistochemical staining using 2F8, �800. (c) Expression of MSR-A 8 h after LPS inhalation. The MSR-A-expressing cells (black) arefound in the vascular lumen, perivascular area, alveolar space as well as in the vascular wall. CD14-expressing cells (brown) are abundant inthe alveolar space, �200. Inset: MSR-A-expressing (black arrow), CD14-expressing (black arrowhead), and double positive (white arrowhead)cells are present, �400. (d) Expression of MSR-A at day 3 after LPS inhalation. The lung is filled with MSR-A-expressing cells (black), �200.(e,f). Combined immunohistochemical staining using 2F8 and TUNEL staining. (e) In the bronchiolar lumen, aggregates of MSR-A-expressingmacrophages (brown) are often found at day 3. Some macrophages (arrows) contain apoptotic bodies (black), �100. (f) There are severalMSR-A-expressing macrophages ingesting apoptotic bodies (arrows) in the lung at day 3, �200. Inset: Apoptotic bodies (black) are phagocy-tized by a macrophage-expressing MSR-A (brown), �1000.

apoptotic bodies (Fig. 5). Incorporation of apoptotic bodiesby CD14-positive cells was found only occasionally.

Changes in the number of macrophages after dailyadministration of M-CSF

After LPS inhalation, the number of total cells in BALF of M-CSF-treated mice was smaller than that of non-treated mice(Fig. 6a). The number of macrophages in the lungs of M-CSF-treated mice increased more rapidly than that of non-treated mice (Fig. 6b). The number of MSR-A-positivemacrophages in M-CSF-treated mice increased anddecreased more rapidly than in non-treated mice. Thenumber of CD14-positive cells was larger in M-CSF-treated

mice than non-treated mice during this stage (Fig. 6c). Incontrast, the number of apoptotic bodies was smaller in M-CSF-treated mice than in untreated mice (Fig. 5).

Expression of MSR-A, CD14, and cytokine messengerRNA in the lungs of LPS-treated mice

The expression of CD14 messenger RNA as detected bynorthern blotting peaked in the lung of M-CSF-untreated miceat 8 h (Fig. 7a). In M-CSF-treated mice, the expressionpattern was similar, but the expression of CD14 messengerRNA was more prominent in control than in M-CSF-treatedmice. CD14 messenger RNA expression diminished morerapidly in M-CSF-treated mice than in control mice (Fig. 7b).

By RT-PCR expressions of CD14, MSR-A, interleukin (IL)-1 and tumor necrosis factor (TNF)-α messenger RNAwere detected from 4 h. M-CSF, c-fms, and granulocyte–macrophage colony stimulating factor (GM-CSF) messengerRNA were detected in the lungs in control mice at low levelsand were enhanced at 4 h. The expression of MSR-A mRNAwas prominent from 4 h to 2 days (Fig. 8a). Daily administra-tion of M-CSF enhanced MSR-A expression and shortenedthe period of cytokine expression. Most cytokine messengerRNA expression was diminished at day 2 in M-CSF-treatedmice (Fig. 8b).

DISCUSSION

This study demonstrated that bronchiolar epithelial cellsimmediately expressed CD14 followed by the appearance ofCD14-positive neutrophils and macrophages after inhalationof LPS aerosol. The MSR-A-expressing cells then appeared

988 T. Yamamoto et al.

Figure 4 (a) Relationship between CD14 expression and F4/80-positive macrophages after lipopolysaccharide challenge. (a): (�),CD14-positive and F4/80-positive; (�), CD14-positive; (�), F4/80-positive. (b) Relationship between macrophage scavenger receptorclass A type I and II (MSR-A)-expressing cells and CD14-expressingcells. (b): (�), CD14-positive and 2F8-positive; (�), CD14-positive;(�), 2F8-positive. (a,b) Immunohistochemical double staining. Dataare expressed as mean � SD.

Figure 5 The number of macrophages ingesting apoptotic bodiesafter lipopolysaccharide challenge. (�), 2F8� apoptosis: M-CSF(–);(�), 2F8� apoptosis: M-CSF(�). Macrophage colony stimulatingfactor (M-CSF)-positive mice daily injected with M-CSF. M-CSF-negative, mice without injection of M-CSF.

in the blood and migrated through the pulmonary interstitium to alveolar space. CD14- and MSR-A-positive macrophageswere regarded as separate populations based on theimmunohistochemical double staining. MSR-A-expressingmacrophages actively incorporated cell fragments of apop-totic cells. Expressions of CD14 and MSR-A as well as ofseveral cytokine messenger RNA were markedly enhanced.Daily administration of M-CSF to this mouse model resultedin increased levels of MSR-A expression and shortening ofthe period of inflammatory process.

Scavenger receptor and CD14 in the lung 989

Figure 6 (a) Changes in the number of total cells in bronchoalveo-lar lavage fluid (BALF) of macrophage colony stimulating factor (M-CSF)-treated mice and non-treated mice. (�) total cell count:M-CSF(–); (�) total cell count:M-CSF(�). (b) The number of F4/80-positive macrophages in the lung sections of M-CSF-treated mice and non-treated mice. (�) F4/80:M-CSF(�); (�)F4/80:M-CSF(–). (c) The number of CD14-positive cells and MSR-A-bearing cells in the lung of M-CSF-treated mice and non-treatedmice. Determined by the immunohistochemical double staining. (�)2F8:M-CSF(�), (�) CD14:M-CSF(�), (�) 2F8:M-CSF(–), (�)CD14:M-CSF(–). Data are expressed as mean � SD.

Figure 7 Expression of CD14 messenger RNA (mRNA) in the lungdetected by northern blotting. (a) CD14 messenger RNA (mRNA)expression is enhanced at 2, 12, and 24 h after the administration oflipopolysaccharide. (b) Expression of CD14 mRNA is less prominentin mice after daily injection of macrophage colony stimulating factor.

Migration of neutrophils into inflamed tissue is part of theearly phase of an acute inflammatory response. LPS inducesa variety of inflammatory responses in CD14-bearing cells,such as macrophages and neutrophils. CD14-mediated sig-naling in neutrophils induces adhesion molecule expressionsand cytokine production which cause tissue injury duringLPS-induced inflammatory processes. CD14 exists in twoforms: (i) a glycosyl phosphatidylinositol-anchored mem-brane glycoprotein; and (ii) a soluble form of CD14 present

in serum. The membrane-bound form participates in themacrophage and neutrophil response to LPS and solubleCD14 is involved in LPS-mediated activation of non-myeloidcells such as endothelial and epithelial cells. In agreementwith the previous observations,9–11 we have observed CD14expression in epithelial cells. Diamond et al. demonstratedthat CD14 of epithelial cell origin mediated the induction of apeptide (a defensins) with microbicidal activity.31 Thus theepithelial cells actively participate in the local host defensesystem mediated by several molecules including CD14.

Scavenger receptors are classified into three classes: (i)class A (type I and type II scavenger receptor, and MARCO);(ii) class B (CD36 and SR-BI); (iii) and class C (dSR-CI,FcγRII-B2, and CD68/macrosialin).32 MSR-A is a trimeric glycoprotein which mediates the recognition and uptake of a wide range of negatively charged macromolecules and isimplicated in the deposition of cholesterol in arterial wallsduring atherogenesis through receptor-mediated endo-cytosis of chemically modified low-density lipoproteins.12–14

Because of a wide range of ligand binding capacity of thesereceptors, a broad spectrum of their biological role has beensuggested. MSR-A has been shown to bind to a precursor ofGramnegative bacterial lipid A.17,18 In the present study, up-regulation of MSR-A in the lung was remarkable after LPSinhalation, indicating that MSR-A is a receptor for LPS.However, the exact role of MSR-A in macrophage activationand LPS-induced tissue injury remains to be solved.Recently, MSR-A knockout mice were generated.20 It is sug-gested that MSR-A might play a protective role against LPSchallenge, because protective effects of MSR-A have beenreported in infection models to Listeria monocytogenes andHerpes simplex virus.20

Several agents and growth factors regulate macrophageproliferation, differentiation, and maturation. Among them, M-CSF effectively controls the development, differentiation, andfunction of monocyte/macrophages. M-CSF stimulates MSR-A expression in macrophages in vitro.33 In the present study,administration of M-CSF to mice increased both expres-sion of MSR-A and the number of MSR-A-expressingmacrophages. It is probable that MSR-A expression isinvolved in scavenging LPS and degradation of enzymatic orchemical agents released from various cells in inflammatoryfoci. Because MSR-A-mediated uptake of LPS does not acti-vate macrophages in vitro,18 it appears likely that MSR-Aplays a protective role against LPS-induced tissue injury. Theshortening of the time course of inflammation in M-CSF-treated mice may also be attributed to the protective functionof MSR-A. In contrast, M-CSF appeared to suppress CD14expression. It is assumed that CD14 down-regulation isattributed to the suppressive effect of some cytokine on CD14expression and GM-CSF suppresses CD14 expression onmonocytes.34 However, the present study demonstrated that

990 T. Yamamoto et al.

Figure 8 Expression of macrophage scavenger receptor class Atype I and II (MSR-A) and cytokine messenger RNA (mRNA) in thelung detected by reverse transcription–polymerase chain reaction.(a) MSR-A and c-fms mRNA expressions are enhanced from 4 hafter the administration of lipopolysaccharide (LPS). Macrophagecolony stimulating factor (M-CSF), granulocyte–macrophage colonystimulating factor (GM-CSF), tumor necrosis factor (TNF)-α, andinterleukin (IL)-1 mRNA expressions are also enhanced from 4 h to 2days. (b) MSR-A mRNA expression is enhanced from 4 h to 1 dayafter lipopolysaccharide (LPS) inhalation in mice that received dailyadministration of M-CSF. The cytokine mRNA expression is en-hanced, but shortened in M-CSF-treated mice.

GM-CSF expression was not accelerated in M-CSF-treatedmice. Taken together, M-CSF might suppress CD14 expres-sion via unknown mechanisms.

In inflammatory foci, leukocytes undergo apoptosis underthe influence of pathogens and cytokines. In LPS models,TNF-α is shown to induce apoptosis in the liver and lungs.22,23

In our model, the expression of TNF-α messenger RNA wasenhanced soon after LPS inhalation and abundant apoptoticcells appeared. MSR-A-expressing macrophages preferen-tially phagocytized apoptotic bodies in the lungs. Scavengerreceptors are known to play an important role in the recogni-tion of apoptotic or senescent cells during the course ofdevelopment and normal cell turnover.35–38 In M-CSF-treatedmice, the number of apoptotic cells was reduced and thenumber of macrophages containing apoptotic cells was alsodecreased. From this fact, it is postulated that M-CSF pro-moted the survival of inflammatory cells and/or that M-CSF-mediated MSR-A up-regulation may have contributed to therapid clearance of the cell fragments. Macrophages thus playan important role in resorption of inflammation by scavengingapoptotic cells and possibly released substances using MSR-A.

In summary, MSR-A as well as CD14 play pivotal roles inLPS-induced acute respiratory disorders. Exogenouslyadministrated M-CSF exerts beneficial effects by enhancingMSR-A expression.

ACKNOWLEDGMENTS

The authors thank Professor Kiyoshi Takahashi, KumamotoUniversity School of Medicine and Professor SiamonGordon, Sir William Dunn School of Pathology, University ofOxford, UK, for their critical review of the manuscript. Wethank Dr H. Nakamura, Mr K. Sato, Mr S. Momozaki, Mr K.Ohyachi, and Mr H. Sano for their excellent technical assis-tance. We also thank Morinaga Milk Industry Co. Ltd, (Tokyo,Japan) for providing human rM-CSF and Dr I. Fraser and DrD.A. Hughes for the 2F8 mAb. This study was supported inpart by Grants-in-Aid for Scientific Research from the Ministryof Education, Science, and Culture of Japan and JapanFoundation Grant for Ageing and Health.

REFERENCES

1 Raetz CRH. Biochemistry of endotoxin. Ann. Rev. Biochem.1990; 59: 129–170.

2 Sweet MJ, Hume DA. Endotoxin signal transduction inmacrophages. J. Leukoc. Biol. 1996; 60: 8–26.

3 Setoguchi M, Nasu N, Yoshida S, Higuchi Y, Akizuki S,Yamamoto S. Mouse and human CD14 (myeloid cell-specificleucine-rich glycoprotein). Primary structure deduced fromcDNA clones. Biochim. Biophys. Acta 1989; 1008: 213–222.

4 Nasu N, Yoshida S, Akizuki S, Higuchi Y, Setoguchi M,Yamamoto S. Molecular and physiological properties of murineCD14. Int. Immunol. 1990; 3: 205–213.

5 Wright SD, Ramos RA, Tobias PS, Ulevitch RJ, Mathison JC.CD14, a receptor for complexes of lipopolysaccharide (LPS)and LPS binding protein [see comments]. Science 1990; 249:1431–1433.

6 Lee JD, Kato K, Tobias PS, Kirkland TN, Ulevitch RJ. Transfec-tion of CD14 into 70Z/3 cells dramatically enhances the sensi-tivity to complexes of lipopolysaccharide (LPS) and LPS bindingprotein. J. Exp. Med. 1992; 175: 1697–1705.

7 Tobias PS, Soldau K, Kline L et al. Crosslinking of lipopolysac-charide to CD14 on THP-1 cells mediated by lipopolysac-charide binding protein. J. Immunol. 1993; 150: 3011–3021.

8 Frey EA, Miller DS, Jahr TG et al. Soluble CD14 participates inthe response of cells to lipopolysaccharide. J. Exp. Med. 1992;176: 1665–1671.

9 Pugin J, Schurer-Maly CC, Leturcq D, Moriarty A, Ulevitch RJ,Tobias PS. Lipopolysaccharide (LPS) activation of humanendothelial and epithelial cells is mediated by LPS bindingprotein and soluble CD14. Proc. Natl Acad. Sci. USA 1993; 90:2744–2748.

10 Matsuura K, Ishida T, Setoguchi M, Higuchi Y, Akizuki S,Yamamoto S. Upregulation of mouse CD14 expression inKupffer cells by lipopolysaccharide. J. Exp. Med. 1994; 179:1671–1676.

11 Fearns C, Kravchenko VV, Ulevitch RJ, Loskutoff DJ. MurineCD14 gene expression in vivo: Extra myeloid synthesis and regulation by lipopolysaccharide. J. Exp. Med. 1995; 181:857–866.

12 Kodama T, Freeman M, Rohrer L, Zabrewcky J, Matsudaira P,Krieger M. Type I macrophage scavenger receptor contains α-helical and collagen-like coiled coils. Nature 1990; 343:531–535.

13 Rohrer L, Freeman MT, Kodama T, Penman M, Krieger M.Coiled-coil fibrous domains mediate ligand binding bymacrophage scavenger receptor type II. Nature 1990; 343:570–572.

14 Freeman M, Ashkenas J, Rees KJG et al. An ancient, highlyconserved family of cysteine-rich protein domains revealed by cloning type I and type II murine macrophage scavengerreceptors. Proc. Natl Acad. Sci. USA 1990; 87: 8810–8814.

15 Matsumoto A, Naito M, Itakura H et al. Human macrophagescavenger receptors: Primary structure, expression, and local-ization in atherosclerotic lesions. Proc. Natl Acad. Sci. USA1990; 87: 9133–9137.

16 Naito M, Kodama T, Matsumoto A, Doi T, Takahashi K. Tissuedistribution, intracellular localization, and in vitro expression of bovine scavenger receptors. Am. J. Pathol. 1991; 139:1411–1423.

17 Hampton RY, Golenbock DT, Penman M, Krieger M, Raetz CR.Recognition and plasma clearance of endotoxin by scavengerreceptors. Nature 1991; 352: 342–344.

18 Ashkenas J, Penman M, Vasile E, Freeman M, Krieger M.Structures and high and low affinity ligand binding properties of murine type I and type II macrophage scavenger receptors.J. Lipid Res. 1993; 34: 983–1000.

19 Dunne DW, Resnick K, Greenberg J, Krieger M, Joiner KA. Thetype I macrophage scavenger receptor binds to Gram-positivebacteria and recognizes lipoteichoic acid. Proc. Natl Acad. Sci.USA 1994; 91: 1863–1867.

20 Suzuki H, Kurihara Y, Takeya M et al. A role for macrophagescavenger receptors in atherosclerosis and susceptibility toinfection. Nature 1977; 386: 292–294.

Scavenger receptor and CD14 in the lung 991

21 Bohlinger I, Leist M, Gantner F, Angermuller S, Tiegs G, Wendel A. DNA fragmentation in mouse organs during endo-toxic shock. Am. J. Pathol. 1996; 149: 1381–1393.

22 Leist M, Gantner F, Bohlinger I, Tiegs G, Germann PG, WendelA. Tumor necrosis factor-α induced hepatocyte apoptosis pre-cedes liver failure in experimental murine shock models. Am. J.Pathol. 1995; 146: 1220–1234.

23 Tsuchida H, Takeda Y, Takei H, Shinzawa H, TakahashT, SendoF. In vivo regulation of rat neutrophil apoptosis occurring spon-taneously or induced with TNF-α or cycloheximide. J. Immunol.1995; 154: 2403– 2412.

24 Metcalf D, Nicola NA. The biology of colony-stimulating factorproduction, degradation, and clearance. In: Metcalf D, NicolaNA, eds. The Hemopoietic Colony-stimulating Factors. Univer-sity Press, New York, 1995; 166–187.

25 Evans R, Kamdar SJ, Duffy TM, Fuller J. Synergistic interactionof bacterial lipopolysaccharide and the monocyte-macrophagecolony-stimulating factor: Potential quantitative and qualitativechanges in macrophage-produced cytokine bioactivity. J.Leukoc. Biol. 1992; 51: 93–96.

26 Chapoval AI, Kamdar SJ, Kremlev SG, Evans R. CSF-1 differ-entially sensitizes mononuclear phagocyte subpopulations toendotoxin in vivo: a potential pathway that regulates the sever-ity of Gram-negative infections. J. Leukoc. Biol. 1998; 63: 245–252.

27 Austyn JM, Gordon S. F4/80, a monoclonal antibody directedspecifically against the mouse macrophage. Eur. J. Immunol.1981; 11: 805–815.

28 Hughes DA, Fraser IP, Gordon S. Murine macrophage scav-enger receptor: in vivo expression and function as receptor formacrophage adhesion in lymphoid and non-lymphoid organs.Eur. J. Immunol. 1995; 25: 466–473.

29 Isobe Y, Chen ST, Nakane PK, Brown WR. Studies on translo-cation of immunoglobulins across intestinal epithelium. I.Improvements in the peroxidase-labeled antibody method forapplication to study of human intestinal mucosa. Acta His-tochem. Cytochem. 1977; 10: 161– 171.

30 Yamamoto T, Naito M, Moriyama H et al. Repopulation of murine Kupffer cells after intravenous administration of liposome-encapsulated dichloromethylene diphosphonate. Am.J. Pathol. 1966; 149: 1271–1286.

31 Diamond G, Russell JP, Bevins CL. Inducible expression of anantibiotic peptide gene in lipopolysaccharide-challenged tra-cheal epithelial cells. Proc. Natl Acad. Sci. USA 1996; 93:5156–5160.

32 Kodama T, Doi T, Suzuki H, Takahashi K, Wada Y, Gordon S.Collagenous macrophage scavenger receptors. Curr. Opin.Lipidol. 1996; 7: 287–291.

33 de Villiers WJS, Fraser IP, Hughes DA, Doyle AG, Gordon S.Macrophage-colony-stimulating factor selectively enhancesmacrophage scavenger receptor expression and function.J. Exp. Med. 1994; 180: 705–709.

34 Kruger M, Van De Winkel JGJ, De Wit TPM, Coorevits L, Ceuppens JL. Granulocyte- macrophage colony-stimulatingfactor down-regulates CD14 expression on monocytes.Immunology 1996; 89: 89–95.

35 Ren Y, Silverstein RL, Allen J, Savill J. CD36 gene transferconfers capacity for phagocytosis of cells undergoing apopto-sis. J. Exp. Med. 1995; 181: 1857–1862.

36 Platt N, Suzuki H, Kurihara Y, Kodama T, Gordon S. Role for theclass A macrophage scavenger receptor in the phagocytosis ofapoptotic thymocytes in vitro. Proc. Natl Acad. Sci. USA 1996;93: 12 456–12 460.

37 Fukasawa M, Adachi H, Hirota K, Tsujimoto M, Akai H, Inoue K.SR-BI, a class B scavenger receptor, recognizes both nega-tively charged liposomes and apoptotic cells. Exp. Cell Res.1996; 222: 246–250.

38 Murao K, Terpstra V, Green SR, Kondratenko N, Steinberg D,Quehenberger O. Characterization of CLA-1, a human homo-logue of rodent scavenger receptor BI, as a receptor for highdensity lipoprotein and apoptotic thymocytes. J. Biol. Chem.1997; 272: 17 551–17 557.

992 T. Yamamoto et al.