monocytes - thorax.bmj.com

7Thorax 1996;51:143-149

Interleukin 1 0 (IL- 1 ) regulation of tumournecrosis factor cx (TNF-cx) from human alveolarmacrophages and peripheral blood monocytes

Lynne Armstrong, Nicola Jordan, Ann Millar

AbstractBackground - Regulation of the in-flanmmatory response within the humanlung is essential to prevent this importantpart of the normal host defence responsebecoming a pathological process. Tumournecrosis factor a (TNF-a) is a cytokineinvolved in the pathogenesis of shock andin granuloma formation, tissue necrosis,and fibrosis in many organ systems in-cluding the lung. Interleukin 10 (IL-10)has been proposed as having an inhibitoryeffect on the production of several in-flammatory cytokines including TNF-a.Methods - The effect of IL-10 ad-ministration on TNF-a production was ex-plored in human alveolar macrophagesand peripheral blood monocytes frommatched individuals. The effects of IL-10on TNF-a protein production were de-termined by sandwich enzyme linked im-munosorbant assay (ELISA), whereas theTNF-amRNA response was established byNorthern blotting using a TNF-a specificoligonucleotide probe. The protein syn-thesis inhibitors actinomycin D and cyclo-hexamide were utilised to monitor IL-10effects on mRNA degradation and de novoprotein synthesis, respectively.Results - The lipopolysaccharide-me-diated TNF-a production in alveolarmacrophages was reduced from 3 508(0.629) to 2 035 (0.385) ng/ml by 100 U/mlIL-10. Lipopolysaccharide-induced TNF-a production in peripheral blood mono-cytes was reduced from 2 035 (0.284) to0-698 (0.167) ng/ml. TNF-a gene ex-pression was also inhibited in both alveolarmacrophages and peripheral blood mono-cytes; lipopolysaccharide-induced TNF-amRNA was reduced by 47-8 (15.2)% and83-1 (4.2)%, respectively, by IL-1O. The IL-10 mediated suppression ofTNF-a mRNAwas unaffected by addition of cyclo-hexamide, suggesting that de novo proteinsynthesis was not required for TNF-ainhibition. mRNA stability experimentsindicated no acceleration in lipopoly-saccharide-induced TNF-a mRNA de-gradation in response to IL-10.Conclusions - These findings suggest thatIL-10 is a potent inhibitor of TNF-a ex-pression and release from alveolar macro-phages and peripheral blood monocytes,and thus it may have an important role in

the cytokine network of the pulmonaryimmune response.(Thorax 1996;51:143-149)

Keywords: interleukin 10, tumour necrosis factor re-gulation, alveolar macrophage.

The inflammatory response is a crucial part ofthe normal host defence mechanism and, assuch, has a protective role. However, when thisresponse is excessive or uncontrolled, patho-logical consequences may result. Inappropriateinflammatory responses have been implicatedas the causal mechanism in many pulmonarydiseases such as the adult respiratory distresssyndrome (ARDS),1-3 idiopathic pulmonaryfibrosis, and sarcoidosis.4-'6 Tumour necrosisfactor a (TNF-ot) is an important proximalmediator of the inflammatory response directlyinducing neutrophil influx and degranulation,7vascular endothelial cell permeability,89 andfibroblast proliferation.'0 TNF-cL also has anindirect role by eliciting the release of otherproinflammatory cytokines from human al-veolar macrophages and neutrophils includinginterleukin 1 (IL-1),1' interleukin 6 (IL-6),"2and interleukin 8 (IL-8) . The pivotal role ofTNF-cx in inflammation leads to interest in theregulatory mechanisms controlling its pro-duction and activity which may have a sig-nificant part to play in the modulation of theinflammatory response.

Interleukin 10 (IL- 10), formerly known ascytokine synthesis inhibitory factor (CSIF), wasoriginally described as a murine Th2 cellularclone product which inhibited gamma inter-feron production by Thi clones.'4 A humanform was soon identified'5 and has been shownto have a regulatory role on the inflammatoryresponse due to its pleiotropic effects on theleucocyte population.'6 These include in-hibition of antigen presenting cell function,downregulation of class II major histo-compatibility complex (MHC) and B7, andcytokine inhibition.'7-22 The current study wasundertaken to investigate the role of IL-10 inthe regulation ofTNF-ot production by alveolarmacrophages as a potential part of the cytokinenetwork regulating the inflammatory responsewithin the lung.

MethodsHuman recombinant IL- 10 (specific activity1 5 x 107 U/mg protein) and JES3-19F1 ratanti-human IL-i 0 antibody were gifts from Dr

The Lung ResearchGroup,Department ofMedicine,Medical School Unit,Southmead Hospital,Westbury on Trym,Bristol BSlO 5NB, UKL ArmstrongA Millar

School of Pharmacyand Pharmacology,University of Bath,Bath, UKN Jordan

Correspondence to:Dr A Millar.Received 20 April 1995Retumed to authors26 July 1995Revised version received11 September 1995Accepted for publication22 September 1995

143

on April 10, 2022 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.51.2.143 on 1 F

ebruary 1996. Dow

nloaded from

http://thorax.bmj.com/

Armstrong, J'ordan, Millar

K W Moore, DNAX (Schering-Plough), PaloAlto, USA. Stock lipopolysaccharide (Es-cherichia coli 0111:B4; Sigma) was preparedat a concentration of 1 mg/ml in a completemedium consisting of sterile RPMI-1640(Gibco, UK) supplemented with 100 U/mlpenicillin, 100 mg/ml streptomycin, and0 5 mg/ml fungizone. Stock cyclohexamide andactinomycin D were prepared in completemedium and dimethyl sulphoxide (Sigma), re-spectively, and used at final concentrations of5 ,ug/ml. Tissue culture plastics (NUNC) wereobtained from Life Technologies (GIBCO),Paisley, UK. The TNF-oc mRNA probe waspurchased from R & D Systems, Abingdon,UK.

Alveolar macrophages for this study wereobtained from six non-smoking healthy vo-lunteers (mean age 25-3 (1 2) years) consentingto fibreoptic bronchoscopy and broncho-alveolar lavage (BAL) by standard techniques.Subjects were injected intramuscularly with 0-6mg atropine and sedated with 0-2 mg alfentaniland 0-10 mg midazolam intravenously. Topicallignocaine was administered to anaesthetise theupper airway. Aliquots of sterile 0 9% salinebuffered with 8A4% sodium bicarbonate(3 x 60 ml) were instilled into the right middlelobe in 60 ml aliquots and then aspirated intoa siliconised bottle kept on ice. The BAL fluidwas strained through a single layer of coarsegauze to remove mucus clumps and the filtratewashed twice at 500 g in complete media. Thecell pellet was resuspended in completemedium and viability assessed with 0 1% try-pan blue was >90%. The cells were re-suspended at a concentration of 3 x 1 06/ml andincubated in tissue culture petri dishes for twohours at 370C. The non-adherent cells werediscarded and the plate gently rinsed to removeresidual non-adherent cells. The adherentpopulation was scraped off gently with a sterilecell scraper (Costar, UK) and differential stain-ing with Diff-quick (Baxter, Thetford, UK)revealed >98% pure alveolar macrophages.Cells were resuspended in complete mediumat a concentration of 1 x 106 cells/ml.Venepuncture was performed on all bron-

choscopy subjects prior to the procedure. Bloodwas collected into heparinised tubes. Plasmawas removed by centrifugation at 1000 g forfive minutes, and the remaining blood was thendiluted with complete medium and carefullylayered on to a Ficoll-Hypaque density gradient(Pharmacia, St Albans, UK). After 30 minutescentrifugation at 400 g, the resultant interfaceof mononuclear cells was removed, washed,and resuspended in culture medium sup-plemented with 10% fetal calf serum (FCS).Following one hour of adherence in a tissueculture flask at 370C the adherent cells wereisolated by gentle scraping. Cells were >95%viable as assessed by trypan blue exclusionand >95% pure peripheral blood monocytes asdetermined by morphology and staining.

Alveolar macrophages and peripheral bloodmonocytes were incubated at 1 x 106 cells/mlin culture medium, 100 RI per well in 96 welltissue culture microtitre plates. Preliminarystudies had determined that optimum in-

hibition of TNF-oc protein occurred whenalveolar macrophages or peripheral bloodmonocytes were pretreated with IL-10 for twohours before addition of 10 jig/ml lipo-polysaccharide to some of the cultures. Cellswere then incubated for a further 24 hours.Supernatants were harvested and stored at- 70°C until TNF-oc determination by ELISA.Ninety six well microtitre plates (Nunc

Maxisorp) were coated with 8,ug/ml CB006monoclonal antibody (courtesy of Celltech,Slough, UK) and left overnight at 4°C. Samplesand the TNF-oc standard (NIBSC, Potters Bar,UK) were diluted in phosphate buffered saline(PBS) containing 2% normal mouse serum(Sigma). Following a one hour incubation,plates were washed three times with PBS con-taining 0-1% Tween (Sigma). Rabbit anti-human polyclonal antibody WBRA-pg2 wasadded at a dilution of 1:500 and incubated fora further hour. Following a further three washesthe polyclonal was captured by a goat anti-rabbit IgG peroxidase conjugate (Sigma) whichgave a colour reaction upon addition of tetra-methylbenzidine substrate. The reaction wasstopped with 1 M sulphuric acid and the prod-uct read at 405 nm on a plate reader (Titertek,Flow Laboratories, Herts, UK). This assay hasa detection limit of 30 pg/ml TNF-o.

Total cellular RNA was obtained from peri-pheral blood monocytes or alveolar macro-phages by the methods described by Jonas etal 23 and Chomczynski and Sacchi24 and asmodified by Strieter et al.25 In brief, the cellswere lysed into a guanidinium thiocyanatebuffer and frozen at -70°C overnight. TotalRNA was then extracted using phenol chlo-roform and chloroform isoamyl alcohol. RNAwas ethanol precipitated and dissolved in water,and the concentration was determined by meas-uring the absorbance at 260 nm. RNA (10 [tgper lane) was separated by electrophoresisthrough a 1% agarose/formaldehyde gel. Ethi-dium bromide was included in each sample,enabling equal loading to be assessed by moni-toring the 18S and 28S ribosomal RNA. RNAwas transblotted overnight to nylon membrane(Boehringer Mannheim) and fixed by bakingat 1 20°C for 20 minutes. The membraneswere prehybridised at 420C, then hybridisedovernight with digoxygenin-labelled oligo-nucleotide probe for TNF-oc at 10 ng/ml(Boehringer Mannheim). Membranes werewashed, blocked with blocking buffer (Boehr-inger Mannheim), and incubated for 30 min-utes with alkaline phosphatase-conjugated anti-digoxygenin antibody diluted 1: 10000. Lu-migen PPD (Boehringer Mannheim) diluted1:100 was added as a chemiluminescent sub-strate. Blots were exposed to x ray film (KodakOmat AR5) for 1-2 hours at room temperature.The autoradiographs were quantitated by ascanning laser densitometer (GS-670, Biorad,Hemel Hempstead, UK). Equivalent amountsof total RNA were assessed by monitoring 28Sand 18S ribosomal RNA.

DATA ANALYSISData were analysed using the Minitab packagefor Windows. Supernatant data are expressed as

144



j.com/


ebruary 1996. Dow

nloaded from


145Interleukin 10 (IL-10) regulation of tumour necrosis factor a (TNF-i)

CD

U-

z

-

z

1 2 3 4 5 6 7

5B

4

3

2

1 1_ I

100°

80 H-

E

x

E4

0

?tc

0

0

60k

40

20k

0L

1 2 3 4 5 6 7

Figure 1 Suppression of TNF-a protein release in (A) inalveolar macrophage supernatants and (B) fromperipheral blood monocytes. These ELISA datademonstrate the effect of a dose range of IL-10 on

lipopolysaccharide induced TNF-a release by alveolarmacrophages and peripheral blood monocytes after 24hours in culture. Lane 1: medium control; lane 2: 10 /iglml lipopolysaccharide; lanes 3-7: 1, 10, 50, 100 and 200unitslml IL-10+ 10 pglml lipopolysaccharide.Representative of six experiments.

mean (SE) and the significance was determinedusing a two-way analysis of variance with Dun-net's correction. p values of <0 05 were con-

sidered statistically significant.

ResultsAdherence purified alveolar macrophages andperipheral blood monocytes (n = 6) were platedon to 96 well microtitre plates and incubatedfor 24 hours with 10l,g/ml lipopolysaccharidealone and with the addition of recombinant IL-10 (1-200 U/ml). Extracellular TNF-cx in thesupernatants was measured with a specific en-

zyme linked immunosorbant assay (ELISA).Previous work had determined that re-

combinant IL-10 had no effect on basal TNF-cx production in normal alveolar macrophagesand peripheral blood monocytes (data notshown). Constitutively, alveolar macrophagesand peripheral blood monocytes produced lowlevels of TNF-oa protein (0-641 (0 160) ng/mland 0-058 (0018) ng/ml, respectively) whichwas increased to 3-508 (0 629) ng/ml and 2035(0-284) ng/ml, respectively, when maximallystimulated with lipopolysaccharide (fig 1A).The lipopolysaccharide-mediated increase inTNF-ot protein was significantly abrogated byIL-10 at doses of 50 U/ml and above; 50 U/mlIL- 10 reduced alveolar macrophage TNF-cxprotein by 37% to give 2-035 (0-284) ng/ml(p<0 01), whereas peripheral blood monocytesgave a more than 60% reduction which trans-lated to 0-698 (0K167) ng/ml (p<001) (fig 1 B).

2 3

Figure 2 Suppression of TNF-cx mRNA in alveolarmacrophages by IL-10. Representative Northern analysisshowing IL-10 mediated inhibition of TNF-a mRNA.Lane 1: medium control; lane 2: 10 iglmllipopolysaccharide; lane 3: 100 Ulml IL-10+lipopolysaccharide; lane 4: IL-10+ lipopolysaccharide+10 pg/ml JES3-19F1 antibody. Representative of threeexperiments. The top panel is an autoradiograph of theNorthern blot, the middle section is the densitometry, andthe lower panel is the 18S and 28S mRNA.Representative of six experiments.

These data demonstrate that IL- 10 significantlysuppresses lipopolysaccharide-induced extra-cellular TNF-ot production from alveolarmacrophages and peripheral blood monocytes.Having established a decrease in TNF-a ex-

tracellular protein secretion by alveolar macro-phages, we wanted to assess the effect of IL-10 administration on TNF-oa mRNA levels.Alveolar macrophages (1 x 1 07 per point) werepretreated for two hours with 100 U/ml IL-10followed by incubation for a further one hourwith 10 jtg/ml lipopolysaccharide. Northernanalysis of alveolar macrophage extractedmRNA revealed a reduction of 47-8 (15-2)%(p<0 05) in the level of TNF-oa message inthree cultures containing 100 U/ml IL-10 (fig2).To establish the temporal window of IL-

10 mediated suppression of TNF-ox mRNA,peripheral blood monocytes were treated with100 U/ml IL-10 either one or two hours beforelipopolysaccharide challenge, simultaneously,and 30 minutes after lipopolysaccharide. TotalRNA was extracted one hour after lipo-polysaccharide administration. Pretreatmentfor two hours and one hour with IL- 10 ensuredsubstantial suppression of lipopolysaccharide-induced TNF-ot message (91-3 (7-3)% and

[. I

4

nI



j.com/


ebruary 1996. Dow

nloaded from


Armstrong, J7ordan, Millar

m i -_ .... ...........................I''.'-l

=. , .1.w ..

oI

25

20

cn_C_

M 15

U)

E0

5 - * *

2 65

E

x

mE

0

4,E0

.t-cnCa

0

100 -

go .-

80 H

70

60 r-

50 L

40

30 i-

20 -

10 _

02 3 4 5 a

Figure 3 Effect ofpretreatment and post-treatment withIL-10 on TNF-c mRNA expression from peripheral bloodmonocytes. Lane 1: medium control; lane 2: 10 gg/mllipopolysaccharide, lanes 3 and 4: preincubation for twoand one hour, respectively, with 100 Ulml IL-10; lane 5:simultaneous addition of IL-10 and lipopolysaccharide;lane 6: addition of IL-10 30 minutes after addition oflipopolysaccharide. Representative offour experiments.

84A4 (6&6)%, respectively) as shown in fig 3(representative of three experiments). A re-

duction in the inhibitory effect of IL-i 0 of 27 6(11 4)% was shown at an interval as short as 30minutes after lipopolysaccharide stimulation.Having established a decrease in TNF-oa ex-

tracellular protein secretion by peripheral bloodmonocytes, we wanted to assess the effect ofIL- 10 administration on TNF-ot mRNA levels.Cells were pretreated for two hours with IL-10 (dose range 0 -1O00 U/ml) before additionof 10 ztg/ml lipopolysaccharide. Cells were

lysed after a further hour of incubation andNorthern analysis was performed to determinemRNA levels. The data show that 100 U/mlIL-10 is the optimum dose for TNF-a mRNAsuppression, giving a reduction of 83 1 (4 2)%(p<OOl) (see fig 4 which represents six ex-

periments). No loss of cell viability was ob-served at any of the doses of IL-10 used up to500 U/ml.To determine if de novo protein synthesis

was involved in IL-10 induced suppressionof TNF-oa, peripheral blood monocytes were

treated with and without IL-10 plus 5,tg/mlcyclohexamide two hours before addition oflipopolysaccharide. Cyclohexamide at this dosehas been shown to have an inhibitory effect ontranslation.26 Total RNA was extracted onehour later. TNF-oa mRNA with and withoutlipopolysaccharide treatment was increased bycyclohexamide treatment (fig 5), but additionin the presence of 100 U/ml IL-10 led to a

decrease in lipopolysaccharide and cyclo-hexamide-induced TNF-a message by 83A4(11 7)%. These data, which are representativeof three experiments, indicate that IL-10 can

directly suppress the transcription of TNF-cz

.. I

.:......s..... .. :.:r

Figure 4 Suppression of TNF-oa mRNA in peripheralblood monocytes by IL-10. Representative Northernanalysis showing the effect of a dose range of IL-10 onTNF-a mRNA expression. Lane 1: medium control; lane2: 10 iglml lipopolysaccharide alone; lanes 3-7: 100, 50,10, 1, and 01 Ulml IL-10, respectively, co-incubatedwith 10 pglml lipopolysaccharide.

mRNA without the requirement for de novoprotein synthesis.The effect of IL-10 on mRNA half life was

assessed by mRNA stability experiments. Peri-pheral blood monocytes were cultured withand without pretreatment with 100 U/ml IL-

100 --

E__

Ex

co

E

0

L,

.4J

0

cnc

80 -

60

2040 i...

1 2 3 4 5 6 7 8

.:....; .:.;; .. ~~~~~. ....

.i..

Figure 5 Effect of cyclohexamide on IL-10 inducedsuppression of TNF-a mRNA from peripheral bloodmonocytes. Lane 1: medium control; lane 2: 100 U/mlIL-10 alone; lane 3: 5 yg/ml cyclohexamide alone;lane 4: 101ig/ml lipopolysaccharide; lane 5: IL-10+cyclohexamide; lane 6: lipopolysaccharide + cyclohexamide;lane 7: lipopolysaccharide+IL-10; lane 8:lipopolysaccharide + cyclohexamide. Representative ofthree experiments.

146



j.com/


ebruary 1996. Dow

nloaded from


Interleukin 10 (IL-10) regulation of tumour necrosis factor oc (TNF-x)

1001

< 80

zF- 60

0X

0

m 10,ug/mI LPS* 100 U/ml IL-10 + LPS

_

_

20

02 4 6 8

Hours after actinomycin D administration

Figure 6 Effect of IL-10 on TNF-cx mRNA stability.TNF-a mRNA was quantified in lipopolysaccharide (LPS)-treated peripheral blood monocytes in the presence andabsence of 100 U/ml IL-10 following treatment with 5 /iglmlactinomycin D. Representative offour experiments.

10 before the addition of 10 gg/ml lipo-polysaccharide. Following incubation withlipopolysaccharide for one hour, 5 gg/ml ac-

tinomycin D was added and cells were lysedat one, two, four, and eight hours following theaddition of actinomycin D. Actinomycin D atthis dose has been shown to inhibit trans-lation.27 No differences were seen in the rateof TNF-ot mRNA decay between the IL-10treated group and the group treated with lipo-polysaccharide alone. Both groups exhibited100% mRNA degradation by eight hours (fig6, representative of four experiments).

DiscussionThe inflammatory response is a crucial part ofthe normal host defences within the lungs as

well as other organ sites. The exposure of thelungs to the environment makes it vital to havean effective host defence system. However, itmust be closely regulated so that this protectivemechanism does not become pathological. Todate, the homeostatic mechanisms by whichthe inflammatory response is regulated withinthe lung are poorly understood. The alveolarmacrophage is regarded as a pivotal influencein orchestrating the pulmonary inflammatoryresponse. Many studies have focused on theincreased production of potentially inflam-matory mediators such as TNF-cx by alveolarmacrophages in lung disease.' 6 Furthermore,several studies of differing inflammatory mod-els have shown abolition or amelioration by theuse ofTNF-ot neutralising antibodies includinganimal models of acute and chronic lunginjury.28 29The production of TNF-ct is also part of the

protective response, and similar experimentshave shown that the timing of TNF-ct pro-

duction in relation to the initiating event iscrucial in determining the outcome.30 In ad-dition, the effect of TNF-ct on a target cell mayalter in the presence of other cytokines.3" Thishas led to considerable interest in agents whichmight provide checks and balances within thiscomplex network. Indeed, a recent paper hasdescribed a TNF-ot mediated induction of IL-10 mRNA (and to a lesser extent protein) inhuman peripheral blood mononuclear cells.32This demonstrates that TNF-a is capable ofinducing at least one cytokine which may beinvolved in its regulation.

IL-10 has previously been shown to inhibitthe release of IL-1 and IL-6 and IL-8 pro-duction from human blood monocytes.2' Spe-cific studies on the regulatory role of IL-10in TNF-ot synthesis have shown an inhibitoryeffect in macrophage cell lines'8 and humanperipheral blood monocytes.2' This study wasconducted in order to determine the relevanceof IL-i 0 to TNF-ot production in the arena ofpulmonary inflammation, specifically its effectson the alveolar macrophage. Our findings sug-gest that both peripheral blood monocytes andalveolar macrophages are susceptible to in-hibition of lipopolysaccharide-induced TNF-ctproduction by IL-10, with peripheral bloodmonocytes perhaps showing the greater sensi-tivity as previously reported.3334 It has beensuggested that the more terminally differ-entiated alveolar macrophages are ap-propriately less sensitive to immunoregulationbecause of their continuous exposure to po-tential allergen and antigen. Both cell typesdemonstrate a reduction in lipopolysaccharide-induced TNF-cL production by IL-10 in termsof released protein and at the mRNA level.Our results indicate that the inhibitory effects

of IL-10 on peripheral blood monocytes andalveolar macrophages are seen at the level ofTNF-ot mRNA expression when stimulated bylipopolysaccharide. IL- 10 had no demonstrableeffect on basal levels of TNF-ct production.Previous studies have examined in some detailthe regulation ofTNF-ot gene transcription andtranslation33"- and the specific role of IL- 10inhibition.'72"22 Data on the mechanisms bywhich lipopolysaccharide increases TNF pro-duction (and hence the possible site of IL-10 activity) are conflicting in terms of thesemechanisms. Two in vitro studies have re-ported an increase in the transcription rate ofthe TNF-a gene in response to lipopoly-saccharide,3839 whereas Bogdan et al40 foundthat transcription rates were unaffected inmouse peritoneal macrophages. By contrast,Han et al39 reported that lipopolysaccharidemodified TNF-ot at both the transcriptionaland translational level by regulating the TNF-ot gene promoter and the 3' untranslated region,respectively, but that the increase in TNF-otprotein was largely due to translational de-pression.34 The current study has not addressedthe possibility that IL-10 may inhibit TNF bydecreasing gene transcription, nor by increasedtranslation, but we have established that no denovo protein synthesis is required.The inhibitory effect of IL- 10 has been

shown to be greatly reduced if added more

147



j.com/


ebruary 1996. Dow

nloaded from


Armstrong, _rordan, Millar

than two hours after lipopolysaccharide ad-ministration, as demonstrated here and else-where,40 by which time TNF-ct mRNA levelshave peaked.39 This suggests that IL-10 is im-portant at the early stage of TNF-ct induction.One possible mechanism of IL-10 activitywould be an increase in the rate of transcriptdegradation. The TNF-cx gene has, in commonwith other cytokines, a 3' untranslated regionwith a UA rich motif. Such a sequence hasbeen shown to confer instability on the resultantmRNA transcript.4' It has been reported thatlipopolysaccharide-treated murine macro-phages have increased ribonuclease activitywhich degrades mRNA containing the UA richmotif,42 and it is possible that IL-10 enhancesproduction of this or another ribonuclease.However, our mRNA stability experimentsshowed no increase in mRNA degradation inresponse to IL-10, which has been confirmedin murine peritoneal macrophages.40 This sug-gests that IL-1 0 must be acting (at least inpart) to inhibit gene transcription of themRNA, since this is the only other mechanismby which the steady state mRNA level of TNF-may be reduced. This contrasts with the

results of a study of the inhibition of lipo-polysaccharide-induced TNF-ct production inhuman neutrophils by IL-1 0,22 which suggestthat enhanced degradation ofmRNA did occur.Conflicting results from different cell systemsand species suggest that there are cell andspecies specific mechanisms of TNF-oc in-hibition by IL-10, thus underlying the import-ance of examining mechanisms in specific celltypes such as the alveolar macrophages.

Investigation into the role of IL-1 0 in in-flammatory disease is only just beginning.4344The data we have presented have shown an invitro mechanism of TNF-o regulation by IL-10 following lipopolysaccharide stimulation. Inmany inflammatory lung diseases such as fi-brosing alveolitis it has been proposed thatalveolar macrophages are activated in vivo byhitherto unidentified antigens where changesin this mechanism might be observed withoutextraneous lipopolysaccharide stimulation. Arole of IL-10 has been suggested both in sep-ticaemia and rheumatoid and osteoarthritis.4344Although, in the latter case, IL-10 productionwas found to occur, the use of anti-IL-10 dem-onstrated an additional increase in pro-inflammatory cytokines (including TNF-otlevels). These data suggest that IL-10 is at-tempting to reduce the inflammatory responsein this situation and leads to the speculationthat increased levels of IL-10 would be be-neficial. Indeed, one recent study has indicatedthat IL-10 may be important in the auto-regulation ofTNF-c in human peripheral bloodmonocytes, and that cells of the monocyte/macrophage phenotype may be a major sourceof IL-10 in human systems.32 We also havepreliminary data to suggest that IL-i1 0 may beconstitutively produced by peripheral bloodmonocytes and alveolar macrophages. We havenot addressed the issue of specificity in thisstudy, and it would be surprising if IL-i0 alonehad an inhibitory effect on TNF-oc in viewof the many examples of redundancy in the

cytokine network. This is most elegantly shownby studies of genetically engineered mice withthe absence of single cytokines ("knockoutmice") in which, although disease processesmay be more common, fatality rarely results.Studies on the autocrine role of IL-1 0 andits relationship with other anti-inflammatorycytokines such as IL-445 and IL-1 346 may givefurther insight into the complex regulation ofthe cytokine network.

The authors would like to thank the Wessex Research HealthAuthority for partly funding this study, the nursing staff andpatients of the Royal United Hospital, Bath and SouthmeadHospital, Bristol for their cooperation, and Mrs Sharon Standenfor typing the manuscript.

1 Jacobs RF, Tabor DR, Burks AW, Campbell GD. ElevatedIL-1 release by human alveolar macrophages duringARDS.Am Rev RespirDis 1989;140:1686-92.

2 Miller EJ, Cohen AB, Nagao, S, Griffith D, Maunder RJ,Martin TR,et al. Elevated levels ofNAP-1/IL-8 are presentin the airspaces of patients with ARDS and are associatedwith increased mortality. Am Rev Respir Dis 1992;146:427-32.

3 Tran Van Nhieu J,Misset B, Lebargy F, Carlet J, BemaudinJF. Expression of TNF-a gene in alveolar macrophagesfrom patients with ARDS. Am Rev Respir Dis 1993;147:1585-89.

4 Carre PC, Mortenson RL, King TE, Noble PW, Sable CL,Riches DWH. Increased expression of the IL-8 gene byalveolar macrophages in idiopathic pulmonary fibrosis. JClin Invest 1991;88:1802-10.

5 Zhang Y, Lee TC, Guillemin B, Yu M-C, Rom WN. En-hanced IL-1 j and TNF-a release and mRNA expressionin macrophages from idiopathic pulmonary fibrosis orafter asbestos exposure. J Immunol 1993;150:4188-96.

6 Foley NM, Millar AB, Meager A, Johnson McI N, RookGAW. TNF production by alveolar macrophages in pul-monary sarcoidosis and tuberculosis. Sarcoidosis 1992;9:29-34.

7 Shalaby MR, Aggarwal BB, Ainderknecht E, SvenderskyLP, Finkle BS, Palladino Jr A. Activation of human poly-morphonuclear functions by interferon-gamma andtumour necrosis factors. J Immunol 1985;135:2069-73.

8 Nawroth PP, Stern DM. Modulation of endothelial cellhaemostatic properties by TNF. J Exp Med 1986;163:740-5.

9 Horvath CJ, Ferro TJ, Jesmok G, Malik AB. RecombinantTNF increases pulmonary vascular permeability in-dependent of neutrophils. Proc Natl Acad Sci USA 1988;85:9219-23.

10 Vilcek J, Palombella VJ, Henriksen-DeStefano D, SwensonC, Feinman R, Hirai M, Tsuyimoto M. Fibroblast growth-enhancing activity of TNF and its relationship to otherpolypeptide growth factors. JExp Med 1986;163:632-43.

11 Mantovani A, Dejana E. Cytokines as communication sig-nals between leukocytes and endothelial cells. ImmunolToday 1989;10:370-5.

12 Jablons DM, Mule JJ, McIntosh JK, Sehgal PB, May LT,Huang CM, et al. IL-6/IFN-P-2 as a circulating hormone.Immunol 1989;142:1542-7.

13 Matsushima K, Morishita K, Yoshimura T, LavuS, Ko-bayashi Y, Lew W, et al. Molecular cloning of a humanmonocyte-derived neutrophil chemotactic factor(MDCNF) and the induction of MDCNF mRNA by IL-1 and TNF. JExp Med 1988;167:1883-9.

14 Fiorentino DF, Bond MW, Mosmann TR. Two types ofmouse helper T cell. IV. Th2 clones secrete a factor thatinhibits cytokine production by Thl clones. J Exp Med1989;170:2081-95.

15 Viera P, de Waal Malefyt R, DangMN, Johnson KE,Kastelein R, Fiorentino DF, et al. Isolation and expressionof human cytokine synthesis inhibitor factor (CSIF/IL-10) cDNA clones: homology to Epstein-Barr virus openreading frame BCRFI. Proc Natl Acad Sci USA 1991;88:1172-8.

16 Yssel H, De Waal Malefyt R, Roncarolo M-G, Abrams JS,Lahesmaa R, Spits H, et al. IL-10 is produced by subsetsof human CD4 + T cell clones and peripheral blood T-cells. J Immunol 1992;149:2378-84.

17 De Waal Malefyt R, Abrams J, Bennet B, Figdor CG, DeVries JE. IL-10 (IL-10) inhibits cytokine synthesis byhuman monocytes- an autoregulatory role for IL-10produced by monocytes. J Exp Med 1991;174:1209-20.

18 Ding L, Shevach EM. IL-10 inhibits mitogen induced T-cell proliferation by selectively inhibiting macrophage co-stimulatoryfunction. J Immunol 1992;148:3133-9.

19 Ding L, Linsley DS, Huang LY, Germain RN, ShervachEM. IL-10 inhibits macrophage co-stimulatory activity byselectively inhibiting the up-regulation of B7 expression.J3Immunol 1993;151:1224-34.

20 De Waal Malefyt, R, HaanenJ, Spits H, Roncarolo M-G,te Velde A, Figdor C, Johnson K, et al. IL-10 and viralIL-10 strongly reduced antigen specific human T-cellproliferation by diminishing the antigen presenting cap-acity of monocytes via down-regulation of class II majorhistocompatibility complex expression. J Exp Med 1991;174:915-24.

148



j.com/


ebruary 1996. Dow

nloaded from


Interleukin 10 (IL-10) regulation of tumour necrosis factor a (TNF-a)

21 Ralph P, Nakoinz I, Sampson-Johannes A, Fong S, LoweD, Min H, Lin L. IL-10, T lymphocyte inhibitor ofhumanblood cell production of Il-I and TNF. J Immunol 1992;148:808-14.

22 Cassatella MA, Meda L, Bonora S. Ceska M, ConstantinG. IL- 0 inhibits the release ofpro-inflammatory cytokinesfrom human polymorphonuclear leukocytes. Evidence foran autocrine role of TNF and IL-1pI in mediating theproduction of IL-8 triggered by LPS. J Exp Med 1993;178:2207-11.

23 Jonas E, Sargent T, Dawid I. Epidermal keratin gene ex-pressed in embryos of Xenopus laevis. Proc Natl Acad SciUSA 1985;82:5413-7.

24 Chomczynski P, Sacchi N. Single-step method of RNAisolation by acid guanidinium thiocyanate-phenol-chlo-roform extraction. Anal Biochem 1987;162: 156-9.

25 Strieter RM, Phan SH, Showell HJ, Remick DG, Lynch JP,Genord M, et al. Monokine-induced neutrophil chemo-tactic factor gene expression in human fibroblasts. J BiolChem 1989;264: 10621-6.

26 Obrig TG, Gulp WJ, Mckeehan WL, Hordesty B. Themechanisms by which cyclohexamide and related glu-tamide antibiotics inhibit peptide synthesis on reticulocyteribosomes. J Biol Chem 197 1;246:174-8.

27 Sobell HM. Actinomycin D and DNA transcription. ProcNat Acad Sci USA 1985;82:5328-31.

28 Piguet PF, Collart MA, Grau GE, Kapanci Y, Vassalli P.TNF/cachectin plays a key role in bleomycin-inducedpneumopathy and fibrosis. J Exp Med 1989;170:655-63.

29 Tracey KJ, Fong Y, Hesse DG, Manogue KR, Lee AT, KuoGC, et al. Anti-cachectinrTNF monoclonal antibodiesprevent septic shock during lethel bacteraemia. Nature1987;330:662-4.

30 Nakane A, Minagawa T, Kohanawa M, Chen Y, Sato H,Moriyama M, et al. Interactions between endogenousgamma interferon and TNF in host resistance againstprimary and secondary Listeria monocytogenes infections.Infect Immun 1989;57:3331-7.

31 Elia J. Tumour necrosis factor interacts with IL-1 andinterferon to inhibit fibroblast proliferation via fibroblastpostoglandin-dependent and -independent mechanisms.Am Rev Respir Dis 1988;138:652.

32 Wanidworanun C, Strober W. Predonimant role of tumornecrosis factor-a in human monocyte IL-10 synthesis. 7Immunol 1993;151:6853-61.

33 Beutler B, Krochin N, Milsark IW, Leudke C, Cerami A.Control of cachectin (tumor necrosis factor) synthesis:mechanisms of endotoxin resistance. Science 1986;232:977-80.

34 Han J, Beutler B, Huez G. Complex regulation of tumornecrosis factor mRNA turnover in lipopolysaccharide-activated macrophages. Biochim Biophys Acta 1991;1090:22-8.

35 Strieter RM, Remick DG, Lynch III JP, Genord M, RaifordC, Spengler R, et al. Differential regulation of TNF-a inhuman AMs and PBMs: a cellular and molecular analysis.Am J Respir Cell Mol Biol 1989;1:57-63.

36 Martinet Y, Yamauchi K, Crystal RG. Differential expressionof the tumor necrosis factor/cachectin gene by blood andlung mononuclear phagocytes. Am Rev Respir Dis 1988;138:659-65.

37 Beutler B, Cerami A. The biology of cachectin/lNF -primary mediator of the host response. Annu Rev Immunol1989;7:625-55.

38 Han J, Brown T, Beutler B. Endotoxin-responsive sequencescontrol cachectin tumour necrosis. Biosynthesis at thetranslational level. J Exp Med 1990;171:465-75.

39 Han J, Huez G, Beutler B. Interactive effects of the tumornecrosis factor promoter and 3' untranslated regions. JImmunol 1991;146:1843-50.

40 Bogdan C, Paik J, Vodovotz Y, Nathan C. Contrastingmechanisms for the suppression of macrophage cytokinerelease by transforming growth factor-beta and inter-leukin- 10. J Biol Chem 1992;267:23301-8.

41 Beutler B, Thompson P, Keyes J, Hagerty K, CrawfordD. Assay of a ribonuclease that preferentially hydrolysesmRNAs containing cytokine-derived UA-rich instabilitysequences. Biochem Biophys Res Commun 1988;152:973-80.

42 Shaw G, Kamen R. A conserved AU sequence from the 3'untranslated region ofGM-CSF mRNA mediates selectivemRNA degradation. Cell 1986;46:659-67.

43 Marchant A, Deviere J, Byl B, De Groote D, Vincent J-L,Goldman M. IL- 10 production during septicaemia. Lancet1994;343:707-8.

44 Katsikis PD, Chu C-Q, Brennan FM, Maini RM, FeldmannM. Immunoregulatory role ofinterleukin 10 in rheumatoidarthritis. J Exp Med 1994;179:1517-27.

45 Hart PH, Vitti GF, Burgess DR, Whitty GA, Piccoli DS,Hamilton JA. Potential anti-inflammatory effects of in-terleukin 4: Suppression of human monocyte tumor nec-rosis factor-a, interleukin 1, and prostaglandin E2. ProcNad Acad Sci USA 1989;86:3803-7.

46 De Waal Malefyt R, Figdor CG, Huijbens R, Mohan-Peterson S, Bennett B, Culpepper J, et al. Effects ofIL-13 on phenotype, cytokine production, and cytotoxicfunction ofhuman monocytes: comparison with IL-4 andmodulation by IFN-y or IL-10. J Immunol 1993;151:6370-81.

149



j.com/


ebruary 1996. Dow

nloaded from


monocytes - thorax.bmj.com

Documents