Top Banner
J Clin Pathol 1983;36:723-733 Review article Granulomatous inflammation - a review GERAINT T WILLIAMS, W JONES WILLIAMS From the Department of Pathology, Welsh National School of Medicine, Cardiff SUMMARY The granulomatous inflammatory response is a special type of chronic inflammation characterised by often focal collections of macrophages, epithelioid cells and multinucleated giant cells. In this review the characteristics of these cells of the mononuclear phagocyte series are considered, with particular reference to the properties of epithelioid cells and the formation of multinucleated giant cells. The initiation and development of granulomatous inflammation is discussed, stressing the importance of persistence of the inciting agent and the complex role of the immune system, not only in the perpetuation of the granulomatous response but also in the development of necrosis and fibrosis. The granulomatous inflammatory response is ubiquitous in pathology, being a manifestation of many infective, toxic, allergic, autoimmune and neoplastic diseases and also conditions of unknown aetiology. Schistosomiasis, tuberculosis and leprosy, all infective granulomatous diseases, together affect more than 200 million people worldwide, and granulomatous reactions to other irritants are a regular occurrence in everyday clinical histopathology. A knowledge of the basic pathophysiology of this distinctive tissue reaction is therefore of fundamental importance in the under- standing of many disease processes. Granulomatous inflammation is best defined as a special variety of chronic inflammation in which cells of the mononuclear phagocyte system are pre- dominant and take the form of macrophages, epithelioid cells and multinucleated giant cells. In most instances these cells are aggregated into well demarcated focal lesions called granulomas, although a looser, more diffuse arrangement may be found. In addition there is usually an admixture of other cells, especially lymphocytes, plasma cells and fibroblasts. Before considering the pathogenesis of granulomatous inflammation it is essential to review our knowledge of the three fundamental cells involved, namely the macrophage, the epithelioid cell and the multinucleated giant cell. Accepted for publication 14 March 1983 Macrophages and the mononuclear phagocyte system The name "mononuclear phagocyte system" was proposed in 1969 to describe the group of highly phagocytic mononuclear cells and their precursors which are widely distributed in the body, related by morphology and function, and which originate from the bone marrow.' Macrophages, monocytes, pro- monocytes and their precursor monoblasts are included, as are Kupffer cells and microglia. Label- ling studies with tritiated thymidine have shown that granuloma cells, including both epithelioid cells and multinucleated giant cells, are also of the same lineage2-5 and it is claimed that monocytes in tissue culture may develop into epithelioid cells and giant cells.6 The origin of tissue macrophages (histiocytes) from bone marrow precursors via circulating mono- cytes is now well established,7 the maturation pro- cess being accompanied by progressive morphologi- cal and functional changes which continue even when macrophages enter the tissues.8 The produc- tion of monocytes is under positive and negative feedback control, with peripheral macrophages and lymphocytes secreting factors that are both stimulatory and inhibitory to stem cell proliferation in the marrow.9 Recruitment and localisation of monocytes into inflammatory lesions is aided by two groups of substances. The emigration of monocytes from the circulation is promoted by chemotactic agents including microbial products, complement 723 on August 23, 2020 by guest. Protected by copyright. http://jcp.bmj.com/ J Clin Pathol: first published as 10.1136/jcp.36.7.723 on 1 July 1983. Downloaded from
11

Review article Granulomatous inflammation- areview · The granulomatous inflammatory response is ubiquitous in pathology, being a manifestation of many infective, toxic, allergic,

Jul 13, 2020

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Review article Granulomatous inflammation- areview · The granulomatous inflammatory response is ubiquitous in pathology, being a manifestation of many infective, toxic, allergic,

J Clin Pathol 1983;36:723-733

Review article

Granulomatous inflammation- a reviewGERAINT T WILLIAMS, W JONES WILLIAMS

From the Department ofPathology, Welsh National School ofMedicine, Cardiff

SUMMARY The granulomatous inflammatory response is a special type of chronic inflammationcharacterised by often focal collections of macrophages, epithelioid cells and multinucleated giantcells. In this review the characteristics of these cells of the mononuclear phagocyte series areconsidered, with particular reference to the properties of epithelioid cells and the formation ofmultinucleated giant cells. The initiation and development of granulomatous inflammation isdiscussed, stressing the importance of persistence of the inciting agent and the complex role of theimmune system, not only in the perpetuation of the granulomatous response but also in thedevelopment of necrosis and fibrosis.

The granulomatous inflammatory response isubiquitous in pathology, being a manifestation ofmany infective, toxic, allergic, autoimmune andneoplastic diseases and also conditions of unknownaetiology. Schistosomiasis, tuberculosis and leprosy,all infective granulomatous diseases, together affectmore than 200 million people worldwide, andgranulomatous reactions to other irritants are aregular occurrence in everyday clinicalhistopathology. A knowledge of the basicpathophysiology of this distinctive tissue reaction istherefore of fundamental importance in the under-standing of many disease processes.Granulomatous inflammation is best defined as a

special variety of chronic inflammation in which cellsof the mononuclear phagocyte system are pre-dominant and take the form of macrophages,epithelioid cells and multinucleated giant cells. Inmost instances these cells are aggregated into welldemarcated focal lesions called granulomas,although a looser, more diffuse arrangement may befound. In addition there is usually an admixture ofother cells, especially lymphocytes, plasma cells andfibroblasts.

Before considering the pathogenesis ofgranulomatous inflammation it is essential to reviewour knowledge of the three fundamental cellsinvolved, namely the macrophage, the epithelioidcell and the multinucleated giant cell.

Accepted for publication 14 March 1983

Macrophages and the mononuclear phagocytesystem

The name "mononuclear phagocyte system" wasproposed in 1969 to describe the group of highlyphagocytic mononuclear cells and their precursorswhich are widely distributed in the body, related bymorphology and function, and which originate fromthe bone marrow.' Macrophages, monocytes, pro-monocytes and their precursor monoblasts areincluded, as are Kupffer cells and microglia. Label-ling studies with tritiated thymidine have shown thatgranuloma cells, including both epithelioid cells andmultinucleated giant cells, are also of the samelineage2-5 and it is claimed that monocytes in tissueculture may develop into epithelioid cells and giantcells.6The origin of tissue macrophages (histiocytes)

from bone marrow precursors via circulating mono-cytes is now well established,7 the maturation pro-cess being accompanied by progressive morphologi-cal and functional changes which continue evenwhen macrophages enter the tissues.8 The produc-tion of monocytes is under positive and negativefeedback control, with peripheral macrophages andlymphocytes secreting factors that are bothstimulatory and inhibitory to stem cell proliferationin the marrow.9 Recruitment and localisation ofmonocytes into inflammatory lesions is aided by twogroups of substances. The emigration of monocytesfrom the circulation is promoted by chemotacticagents including microbial products, complement

723

on August 23, 2020 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.36.7.723 on 1 July 1983. Dow

nloaded from

Page 2: Review article Granulomatous inflammation- areview · The granulomatous inflammatory response is ubiquitous in pathology, being a manifestation of many infective, toxic, allergic,

724

components, fibrin degradation products andlymphokines while the immobilisation ofmacrophages within a lesion is aided by otherlymphokines including migration inhibition andmacrophage adhesion factors.'0 " Althoughimmigration from the circulation seems to be by farthe most important source of macrophages in theinflammatory reaction, local macrophage mitosisdoes occur.'2 However, perhaps due tochromosomal instability, this seems to be limited tovery few divisions."3

Light microscopy of routine haematoxylin andeosin stained sections does not allow macrophagesto be distinguished from other mononuclear cells inan inflammatory infiltrate unless they containrecognisable ingested material. Macrophages maybe round, oval or spindle shaped in outline with acytoplasm which varies from eosinophilic and finelygranular to clear and vesicular. The nucleus has asmooth or sometimes indented membrane withmarginated heterochromatin and usually a singlenucleolus. On ultrastructural examination the cellmembrane is irregular, being thrown into folds andprocesses. Cytoplasmic organelles vary greatlyaccording to the functional activity of the cell8 andinclude endoplasmic reticulum (rough and smooth),mitochondria, Golgi complexes, microtubules,microfilaments and membrane bound vesicles, thelatter including primary and secondary lysosomesand residual bodies. However, in the absence ofobviously phagocytosed material there is no ultra-structural feature that is absolutely diagnostic forthe macrophage.Enzyme histochemistry is more valuable in the

tissue identification of cells of the mononuclearphagocyte series, but suffers from the disadvantageof requiring specially prepared material.Macrophages typically contain non-specific ester-ases diffusely in the cytoplasm, and the presence ofmembrane bound lysosomal enzymes, especiallyacid phosphatase and lysozyme is also useful in theirrecognition. Furthermore, variations in thecytoplasmic distribution of peroxidase correspond todifferent stages in macrophage differentiation.'45-nucleotidase, leucine aminopeptidase and alkalinephosphodiesterase I are cell membrane-associatedmacrophage enzymes which have also been used asmacrophage markers.'5 Intracellular localisation ofmacrophage products, notably alpha-i -antitrypsin,'6also has a role in cell identification. However, it isthe development of monoclonal antibodies tospecific cell types that holds the most promise for theidentification of mononuclear phagocyte cells.7 18 Ifthe various stages of functional differentiation ofthese cells are to be recognised morphologicallythen immunohistology with specific monoclonal

Williams, Williams

antibodies, similar to those presently available forlymphocyte subsets'9 is the most hopeful approach.The ability to ingest a wide variety of substances

into membrane bound vacuoles (endocytosis) is adistinctive but not unique property of mononuclearphagocytes. Two mechanisms are involved,pinocytosis and phagocytosis. Macrophages take upfluids and soluble proteins, immune complexes,hormones, lectins and other macromolecules bypinocytosis20 21 whereas larger particles are engulfedby phagocytosis. The process is initiated by interac-tion between a particle and a surface receptor whichthen triggers intracytoplasmic contractile proteins,including actin and myosin, to create membranemovement and pseudopodial ingestion.20 22-24Macrophage surface receptors are of many typesand specificity. Particles opsonised byimmunoglobulin G and E interact with Fcreceptors25 26 while those attached to the third com-ponent of complement bind to C3b receptors.27Both Fc and C3b receptors may also react withimmune complexes while non-immunologicalreceptors exist for lectins,28 alternative pathwaycomplement activators29 and other particles.

Particulate ingestion is followed by phagosome-lysosome fusion allowing intracellular degradationand microbial killing. Digestion of particulate mater-ial and dead organisms is accomplished by lysosomalenzymes'3 but killing of micro-organisms dependson other methods, in particular the production ofsuperoxides, hydrogen peroxide and hydroxyl radi-cals30 and other microbicidal substances. The pro-cess is aided by the presence of antibody, IgG forbacteria, IgE for metazoa."

Intracellular killing of micro-organisms is greatlyenhanced by the phenomenon of macrophageactivation,32 a change which is accompanied bymorphological8 and many other functional altera-tions including the secretion of a range of differentsubstances, enhanced phagocytic capacity, and anability to recognise and kill tumour cells."' Activa-tion can be accomplished by immunoglobulins,immune complexes, activated complement compo-nents, lymphokines, and by non-immunologicalagents such as bacterial endotoxin."

Extracellular secretion by activated macrophagesis now recognised as a most important function34 35and some idea of the myriad of secretory productscan be obtained from the Table. It can be seen thatamong the factors produced are those with essen-tially opposing effects, for example collagenase andfibrogenic substances, suggesting that there are verysubtle controls over secretion which at present arealmost a complete mystery. Nevertheless it is knownthat activation is not an "all or none" phenomenon,and that the function of activated macrophages can

on August 23, 2020 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.36.7.723 on 1 July 1983. Dow

nloaded from

Page 3: Review article Granulomatous inflammation- areview · The granulomatous inflammatory response is ubiquitous in pathology, being a manifestation of many infective, toxic, allergic,

Granulomatous inflammation-a review

Secretory products ofmacrophages

EnzymesNeutral proteases-for example, collagenase, elastase,plasnunogen activator, angiotensin convertingenzymne, enzymes denatunng proteoglycans and myelin.

Acid hydrolases-for example, phosphatases, sulphatases,proteases, ribonucleases.

LysozymeEsterasesEnzyme inhibitorsAlpha-1-antitrypsinPlasmin inhibitorsComplement componentsCl, C2, C3, C4, C5Properdin, Factors B, DOxygen metabolitesSuperoxides, hydrogen peroxide, hydroxyl radicalEndogenous pyrogens

be varied according to the nature of the activatingstimulus.36 Furthermore there is increasing evidencethat macrophages do not form a homogeneousgroup of cells, but that they are markedlyheterogeneous, different populations having differ-ent characteristics and functions.37 38 The tumorici-dal effect of activated macrophages is only poorlyunderstood. It is likely that macrophage-tumour cellcontact is necessary in some instances313539 whilemacrophage secretions, especially peroxides, areimportant in others.40Macrophages have complex interactions with

lymphoid cells at different phases of the immuneresponse.4'43 Induction of immunity probablyrequires the initial presentation of an antigen to Tlymphocytes which can then initiate cell-mediatedimmunity or generate "helper" functions forantibody-producing B lymphocytes. Macrophagesare responsible for this presentation, but only afterthey have endocytosed the antigen and "processed"it. Macrophage-T cell interaction is restricted tocells bearing surface HLA-DR (or Ia) moleculesand hence is under the control of immune responsegenes. Cell-to-cell contact is usually necessary at theonset, but growth and differentiation factors whichcause T cell proliferation and differentiation andwhich are secreted by macrophages ("monokines")are also involved. Mention has already been made ofthe role of macrophages as effectors of the immuneresponse. Lymphokines, antibodies and immunecomplexes cause macrophage chemotaxis andactivation, allowing ingestion and final eliminationof the antigen, a process that is enhanced by opson-isation with antibody and complement. It is clear,therefore, that macrophages have properties whichmake them highly suited to their central role ingranulomatous inflammation which is the defence ofthe host from exogenous or endogenous irritants.

Bioactive lipidsProstaglandins, thromboxanes, leukotrienesPlatelet activatorsBinding proteinsTransferrin, B12 binding protein, fibronectinCyclic AMPFactors stimulating proliferation of:Lymphocytes (T and B)Myeloid precursors (colony stimulating factors)Erythroid precursorsFibroblastsSmall blood vesselsFactors inhibiting proliferation of:LymphocytesTumour cellsViruses (interferon)

EpithelHoid cellsEpithelioid cells are mononuclear cells with finelygranular eosinophilic cytoplasm, vesicular nuclei,and indistinct cell boundaries which are usuallyfound aggregated into clusters within certaingranulomas. Their mononuclear phagocyte origin isnot in doubt,2-5"4 but there remains controversyover the mechanisms by which epithelioid cells areformed, and in particular the role of cell-mediatedimmunity. Epithelioid cells have been considered tobe a hallmark of delayed hypersensitivitygranulomas,45 a fact well illustrated in the pathologyof leprosy, where epithelioid cells only occur withthe appearance of cell-mediated immunity to thecausative organism.46 However there are nowreports of epithelioid granuloma formation in con-genitally athymic "nude" animals47 48 suggesting thatT cell function is not essential. Furthermore,although lymphokines induce dramatic changes inmacrophages in vitro, the changes are not quitethose of epithelioid transformation.4950

Ultrastructural examination of epithelioidcells5l-55 reveals closely applied and ofteninterdigitating cell membranes which, in theexperience of most workers, lack any junctionalspecialisation. Nevertheless some authors56 haveillustrated desmosome and hemidesmosome-likestructures. The nuclei are regular and ovoid withmarginated heterochromatin, and the complex cyto-plasm contains numerous mitochondria and anactive Golgi apparatus. Some epithelioid cells con-tain rough surfaced endoplasmic reticulum resembl-ing that of plasma cells or fibroblasts52-54 (type A orplasmacytoid epithelioid cells) while others havenumerous cytoplasmic single membrane-boundvesicles containing electron-lucent or weaklyosmiophilic material (type B or vesicular epithelioidcells). There is now good evidence that these two

725

on August 23, 2020 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.36.7.723 on 1 July 1983. Dow

nloaded from

Page 4: Review article Granulomatous inflammation- areview · The granulomatous inflammatory response is ubiquitous in pathology, being a manifestation of many infective, toxic, allergic,

Williams, Williams

cytoplasmic appearances represent the two ends of amorphological spectrum, intermediate forms havingrough endoplasmic reticulum and vesicles in varyingproportions.5357 The proportions of the different celltypes vary in different granulomas according to theaetiology,53 but generally speaking plasmacytoidepithelioid cells are predominant in the early phaseof granulomatous inflammation while the vesicularcells become numerous at a later stage, suggestingthat plasmacytoid cells mature with time into vesicu-lar cells, presumably with a progressive modificationof function."One of the consistent features of the epithelioid

cell is the virtual absence of recognisable endocyt-osed material, either on light or electronmicroscopy, suggesting that the cell is not activelyphagocytic. Nevertheless, a common finding is theintracellular Schaumann body, a complex ofcrystalline calcium salts and conchoidal bodies pro-bably derived from autophagocytic residuallysosomal bodies.58 59 Functional studies onepithelioid cells have, until recently, only been poss-ible on "facsimile" epithelioid cells, that is, cellsproduced artificially from macrophages byexperimental manipulation.6'62 Although havingmany of the morphological features, these facsimilesare not identical to epithelioid cells and theirauthenticity has been questioned.63 However, viableepithelioid cells have recently been isolated fromgranulomas produced in vivo, allowing functionalstudies to be made.' Both facsimile and isolated"true" epithelioid cells are, as would be expectedfrom t8heir morphology, poorly phagocytic, andthere is a blanket inhibition of endocytosis by bothimmunological and non-immunological receptors.Moreover, there is evidence that the expression ofsurface immune receptors (Fc and C3b) is reducedin epithelioid cells compared with macrophages,although there is some dispute over the detailedchanges.6466 It appears therefore that theepithelioid cell is not specialised to interact withextracellular particulate matter. Nevertheless recentstudies in sarcoidosis indicate that epithelioid cellsexpress surface HLA-DR (Ta) antigens67 and con-sequently have the potential to interact immunolog-ically with activated T lymphocytes in their vicin-ity.42

Electron microscopy suggests that epithelioid cellshave important biosynthetic properties. Enzymehistochemistry has shown the presence of acidphosphatases, ,3-galactosidase, lysozyme and non-specific esterase, while ultrastructural cytochemistryhas revealed a mucoglycoprotein, not yet furthercharacterised, within the vesicles of type B cells."Some of these vesicles have been seen to fuse withthe plasmalemma, presumably discharging their

contents into the extracellular space.56 Considerableattention has been given to the secretion ofangiotensin-converting enzyme by epithelioid cells,not only to the immunocytochemical localisation ofthe enzyme69 but also to the use of serumangiotensin-converting enzyme activities in the clin-ical diagnosis of granulomatous diseases.70 The roleof angiotensin-converting enzyme in the granulomais highly speculative but there is preliminary evi-dence to suggest that it inhibits the migration ofmacrophages and leucocytes.7' Furthermore its sec-retion by the epithelioid cell can be controlled by Tlymphocytes.72 It is highly likely that epithelioid cellssecrete many more substances, similar to and inkeeping with, macrophages. However, details arenot known at present.As secretory cells, epithelioid cells share features

in common with activated macrophages. Indeed,some workers consider the two terms to besynonymous, but this is surely an oversimplification.Most reports of activated macrophages describeincreasing phagocytic capacity and expression ofsurface receptors, but epithelioid cells exhibit theopposite. Nevertheless it is possible that epithelioidtransformation could be a specialised type ofmacrophage activation, perhaps by a distinct sub-population of mononuclear phagocytes. The lack ofany evidence of phagocytosis, recent or past, inepithelioid cells raises the possibility that these cellsdo not arise from phagocytic tissue macrophages4"but develop directly from monocytes entering thelesion which are already destined to mature intoepithelioid cells.63 " The factors which initiate thisprocess, however, remain a complete mystery.The epithelioid cell is, therefore, best regarded as

a very specialised type of mononuclear phagocyte,immobilised in the granuloma, whose function hasbeen diverted away from phagocytosis to extracellu-lar secretion.

Multinucleated giant cells

Multinucleated giant cells are a regular feature ofgranulomatous inflammation. There is now over-whelming evidence that they are macrophagepolykaryons, produced by the fusion ofmacrophages, rather than by nuclear mitosis withoutcytoplasmic division.25 6 73 74 Traditionallyinflammatory giant cells have been divided into theLanghans (tuberculous) type, in which up to 20nuclei are distributed centrally or around theperiphery of the cell, and the foreign-body type withoften very numerous haphazardly arranged nucleithroughout the cytoplasm. However it is now clearthat there is no fundamental difference betweenthese two cell types, and there is no diagnostic

726

on August 23, 2020 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.36.7.723 on 1 July 1983. Dow

nloaded from

Page 5: Review article Granulomatous inflammation- areview · The granulomatous inflammatory response is ubiquitous in pathology, being a manifestation of many infective, toxic, allergic,

Granulomatous inflammation-a review

significance. Both types are commonly found tocoexist in the same lesion, transitional forms havebeen described, and studies in tissue culture haveshown that foreign body type giant cells "mature"into Langhans type cells, probably by movements ofthe intracellular cytoskeleton.5 '3Mechanisms whereby macrophages fuse to pro-

duce giant cells have been widely studied. Fusioninduced by viruses occurs in many cell typesthroughout the body, including macrophages,2' butis of limited significance in the granulomatousenvironment. Three main ideas have been suggestedfor inflammatory giant cell formation. First, it wasproposed that fusion may be an immune-mediatedphenomenon, giant cell production being stimulatedby lymphokines.7576 However, the evidence for thishas been questioned," macrophage fusion occurs invitro in the absence of immune factors and giant cellformation occurs equally well in normal and athymicmice." The second suggestion is that fusion occursbetween "young" macrophages and "older" cells,the latter having existed for some time in thegranulomatous environment acquiring chromosomalabnormalities and changes in the macrophagesurface.5 The recognition of the altered and abnor-mal cell surface by young macrophages is thestimulus for cell fusion, and the process is regardedas a means whereby altered, effete and senescentcells can be removed. Hovever, other studies havefailed to produce giant cells by altering macrophagesurfaces in vitro, or by coculturing macrophages ofdifferent genetic makeup.78 The third proposal isthat giant cells form as a result of simultaneousattempted phagocytosis,' during which twomacrophages attempt to ingest the same particle.The endosome margins of one macrophage, insteadof fusing together around the particle, fuse with theendosome margins of a second macrophage, result-ing in fusion of the two cells. There is considerablecircumstantial and experimental evidence to supportthis theory, including the relatively poor phagocyticcapacity of giant cells79 which can be explained atleast partly by the interiorisation of surfacemembrane receptors during the original fusion pro-cess.80 Although ingested foreign material canfrequently be found within giant cells this is notalways the case, and this is one of the possibleanomalies of the simultaneous endocytosis theory.The proponents of the idea suggest that this can beexplained by the fact that the granulomatousenvironment itself produces endocytogenic materialfrom endogenous macromolecules, independently ofthe initial irritant.367380Ultrastructural examination of multinucleated

giant cells from granulomas produced by foreignmaterial provides some evidence to support the

727

simultaneous endocytosis theory, in that the cellsoften contain ingested material or the products of itsdegradation, residual bodies with myelin figures.8'Prominent microfilaments are also present, espe-cially in the periphery of the cytoplasm and thesesometimes fuse together to produce the star-shapedasteroid bodies seen on light microscopy.82 Morecentrally there is an active Golgi apparatus andnumerous mitochondria, lysosomal bodies and somemembrane-bound vesicles. The giant cells ofepithelioid cell granulomas are different, however.They rarely contain microfilaments or any recognis-able ingested material. Instead they have an ultra-structure similar to that of epithelioid cellss15383 andsome authors have described giant cells with thecytoplasmic features of both type A and type Bepithelioid cells.55 The way by which these cellscould be formed is a mystery-the poor phagocyticcapacity of epithelioid cells" would make theirfusion by simultaneous endocytosis very unlikelyunless the cytoplasm did not develop its epithelioidappearance until after cell fusion had occurred.There is now some electron microscopical evidenceto suggest that fusion of epithelioid-like cells mayfollow the development of specialised desmosome-like intercellular junctions between adjacentceUs.56 84

The functions of multinucleated giant cells areonly speculative. While their formation in foreignbody granulomas may be only an accident ofsimultaneous endocytosis the process does have theadvantage of successfully interiorising particleswhich would otherwise be too large for endocytosisby a single cell. Moreover there is no reason tosuspect that intracellular digestion by giant cells isinferior to that by mononuclear macrophages. Theultrastructure of giant cells, especially those inepithelioid granulomas, suggests that they too couldhave important biosynthetic and secretory functionssimilar to mononuclear epithelioid cells.

Pathogenesis of granulomas

Granuloma formation is usually regarded as a meansof defending the host from persistent irritants ofeither exogenous or endogenous origin. The causa-tive agent is walled off and sequestered by cells ofmacrophage lineage allowing it to be contained, ifnot destroyed altogether. Experimental models ofgranulomatous inflammation have provided much ofour present knowledge of the pathogenesis ofgranulomas.85 Such studies have shown that both thenature of the irritant and host factors are importantin governing the type of reaction that is produced.All injected substances cause an initial influx ofmononuclear cells by the phenomenon of

on August 23, 2020 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.36.7.723 on 1 July 1983. Dow

nloaded from

Page 6: Review article Granulomatous inflammation- areview · The granulomatous inflammatory response is ubiquitous in pathology, being a manifestation of many infective, toxic, allergic,

728

chemotaxis. However, what happens next dependson the resistance of the irritant to degradation bymacrophages. If it is a soluble substance that is easilydigested then the macrophages move away oncedegradation is complete.'3 However if it is poorlysoluble, persistent and undegradable a granuloma isformed. The exception to this rule is that solublematerials can produce granulomas if they combinewith endogenous macromolecules to form insoluble,undegradable compounds, a mechanism consideredimportant in granuloma formation by certain solublemetal salts such as beryllium.85 Experimentalgranulomas can also be produced by soluble irritantscomplexed either with insoluble inert materials86 orwith antibodies to form insoluble immune com-plexes.87Why poorly soluble, undegradable material

causes immobilisation of macrophages and theirorganisation into a granuloma is unknown, althoughin some instances, where there is involvement of theimmune system, lymphocyte-produced MIFsundoubtedly have a role. The macrophages in thelesions are often activated, making them particularlysuited to a degradative function, but in spite of this,granuloma-producing agents often persist withincells for a long time. Their resistance to degradationis quite unexplained in many cases, although some-times there is evidence that the macrophage' sarmamentarium of lysosomal enzymes isinappropriate to denature the chemical structure ofthe irritant, such as the cell walls of certainbacteria.88 Other irritants, such as some parasites,escape destruction by acquiring a coat of host anti-gens89 or becoming sequestered within themacrophage cytoplasm, apparently safe fromattack.90 The latter sequestration is in part due to thefailure of phagosomes to fuse with lysosomes, a fea-ture of some mycobacterial infections.9'

Observations that granulomas are of differingmorphology and caused by a wide variety of irritantshave led to numerous attempts to classifygranulomatous inflammation, either to help in thediagnosis of granulomatous disease or to further theunderstanding of the granulomatous process. How-ever, none has been very successful. On a puremorphological level, histopathologists have dividedgranulomas into "foreign body" and "epithelioid"types, depending primarily on the absence or pre-sence of epithelioid cells. An inducing agent is oftenrecognisable in foreign body granulomas while it isdifficult or impossible to find in epithelioid lesions.Moreover the two lesions are said to contain eitherforeign body or Langhans type giant cells respec-tively. This classification is very unsatisfactory, how-ever; the two types of giant cell are not distinct, ashas been alluded to earlier, and in practice it is often

Williams, Williams

difficult to achieve agreement among differentobservers as to the presence or absence ofepithelioid cells.A second classification of granulomas is based on

cell kinetics.'392 Within any lesion there is a con-tinuous turnover of macrophages, dying cells beingreplaced either by new recruits from the circulationor by local mitosis. However, there are strikingvariations in the turnover rate between differentgranulomas. "Low turnover" granulomas are thosewith little macrophage death, immigration ormitosis. They are typically produced by agentswhich, although poorly degradable, are relativelyinert and non-toxic to the cells-for example,carrageenan, barium sulphate. The macrophagespresent are long lived and contain large amounts ofthe irritant.93 Epithelioid cells are not found, andlymphoid cells are unusual, suggesting that immunemechanisms are of minor importance in theirpathogenesis. These low turnover lesionscorrespond to foreign body granulomas. "Highturnover" granulomas, on the other hand, are pro-duced by irritants which are toxic to macrophagessuch as mycobacteria or silica. They are character-ised by a high rate of recruitment and local divisionof macrophages to compensate for their relativelyshort life span and high death rate within thelesion.94 The causative agent is present in only asmall proportion of the cells and the lesions thushave some features in common with epithelioidgranulomas. However, not all high turnovergranulomas have epithelioid cells. Although aclassification of granulomas by their cell kinetics isof considerable theoretical importance it is unfortu-nate that in clinical practice most lesions are of thehigh turnover type and contain macrophages in vary-ing degrees of activation. There is some evidence offunctional heterogeneity among the infiltratingmacrophage population, not only betweengranulomas of different aetiology,63 but also indifferent zones of the same granuloma.38The characteristics of a granuloma are not only

dependent upon the properties of the causative irrit-ant. Host factors are also of great importance, a factwell illustrated in the pathology of leprosy, wheredifferent individuals produce very differentgranulomatous reactions to the causative bacillus,with a spectrum of appearances between twoextremes. At one end of the spectrum is lepromat-ous leprosy in which there are ill-defined collectionsof foamy macrophages containing large numbers ofbacilli-features associated with foreign bodygranulomas. At the other end is tuberculoid leprosywith organised epithelioid granulomas in whichbacilli are difficult to find. It is now clear that themost important factor governing the type of reaction

on August 23, 2020 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.36.7.723 on 1 July 1983. Dow

nloaded from

Page 7: Review article Granulomatous inflammation- areview · The granulomatous inflammatory response is ubiquitous in pathology, being a manifestation of many infective, toxic, allergic,

Granulomatous inflammation-a review

is the degree of immunological resistance to theorganism developed by the host. Patients with highimmunity develop tuberculoid reactions while thosewith low immunity have the lepromatous form."Similar findings have been found in experimentalzirconium granulomas in guinea pigs.9596 A thirdclassification of granulomas, therefore, is based onthe immunological dependence of the lesions,97 withgranulomas being divided into immunological andnon-immunological types. In view of the closeinteractions of lymphocytes and macrophagesdescribed earlier it is not surprising that both cell-mediated and antibody-mediated immunity havetheir role in the accumulation and differentiation ofmononuclear phagocytes that is granuloma forma-tion. Nowhere has this been studied more than inexperimental schistosomiasis, where it is interestingthat different species of the parasite apparently pro-duce immunological granulomas by different means.The reaction to Schistosoma mansoni is largely acell-mediated, T lymphocyte dependent reac-tion,85 98 whose progression is controlled by the bal-ance of T cell subsets (helper and suppressor cells)and possibly by the antibody response to the para-site.99- 101 Schistosoma japonicum, on the otherhand, produces immunological granulomas by amethod that appears to be independent of cell-mediated immunity, but whose initiation and controlrequires the presence of antibody.'02 103 Cell-mediated immunity is also involved in thegranulomatous reaction of tuberculosis andberylliosis,85 while antibodies, especially when com-bined with antigen in the form of immune com-plexes, not only produce granulomas in theexperimental model,87 but are also probably impli-cated in the granulomatous reaction of extrinsicallergic alveolitis,'0 primary biliary cirrhosis,'05 andeven in mycobacterial infections.'06 Recent studiesusing monoclonal antibodies to identify lymphocytesubtypes have highlighted the importance of cellmediated immunity in the genesis of epithelioidgranulomas. In the granulomas of sarcoidosis67 andtuberculoid leprosy,'0' the great majority of thelymphocytes present are T cells. Furthermore,helper T cells greatly outnumber suppressor/cytotoxic subsets, the ratio of helper to suppressorcells in sarcoidosis increasing with the clinical activ-ity of the disease. Helper T cells in sarcoidosis aredistributed uniformly throughout the granuloma67while in tuberculoid leprosy they are concentrated ina cuff around the central epithelioid cell zone.'07 Thenumber of T lymphocytes in the non-epithelioidgranulomas of lepromatous leprosy, on the otherhand are small, the cells present being predomin-antly of the suppressor/cytotoxic subset.

In considering immune mechanisms in granuloma

729

formation it is also essential to take account ofchanges in the circulation. These have been studiedrecently in sarcoidosis where the high density ofhelper T cells at the site of granuloma formation isaccompanied by a reduction in the proportion ofhelper cells in the circulation and an increase in theproportion of suppressor T cells.'08 Moreover thereis evidence that the serum of many patients withsarcoidosis contains T cell suppressant factors, someof which are probably immune complexes.'09The idea that immune mechanisms can initiate

granulomatous inflammation has not gained univer-sal acceptance. Epstein,63 11 while stressing the roleof immune mechanisms in amplifying thegranulomatous response, has cast doubt on the soleimportance of immunity in the initiation of the reac-tion and has suggested the existence of a specifictype of hypersensitivity leading to epithelioidgranuloma formation which he terms granulomatoushypersensitivity. He states that the nature of thisreaction is a mystery but suggests that it might be aspecific function of the mononuclear phagocytesystem. Other workers"' have extended this idea bypostulating that a specialised subgroup of mono-nuclear phagocyte cells can act as memory cells,being primed on first exposure to a granuloma-producing agent. Subsequent exposure is thenfollowed by proliferation of these memory cells toproduce epithelioid cells and giant cells. There is nogood evidence to support this hypothesis, but in factthe concept of granulomatous hypersensitivity isvery difficult to prove or disprove. However somesupport for its existence comes from the productionof epithelioid granulomas in immune deficient ani-mals and from the development of zirconiumgranulomas in man in the absence of demonstrablecell mediated immunity to this metal."2 Ifgranulomatous hypersensitivity occurs at all as anentity, its effects are greatly amplified by the con-ventional immune response.

Complications of granulomatous inflammadon:necrosis and fibrosis

Granulomatous inflammation, like any inflammat-ory reaction, frequently results in tissue damage dur-ing the active phase and fibrosis during the healingprocess. It is not surprising, in view of the nature ofsome macrophage secretions (Table), that tissuenecrosis is a frequent complication of somegranulomas, especially towards the centre of lesionscontaining highly activated cells which are continu-ally dying and releasing their toxic contents. Suchnecrosis may take the form of caseation and cavita-tion, as occurs classically in tuberculosis, or it mayappear as a microabscess containing polymorphs. In

on August 23, 2020 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.36.7.723 on 1 July 1983. Dow

nloaded from

Page 8: Review article Granulomatous inflammation- areview · The granulomatous inflammatory response is ubiquitous in pathology, being a manifestation of many infective, toxic, allergic,

730

addition to autodigestion by macrophage enzymes,tissue necrosis may also be produced by the directtoxic action of a causative agent, especially in thecase of infectious micro-organisms. There is alsoevidence that the process is augmented by theimmune response, both by its cellular and humoralarms. Cavitation in tuberculous granulomas hasbeen associated with strong delayed hypersensitivityto the tubercle bacillus."3 However recent studieswith experimental mycobacterial infections in ratshave shown that necrosis in granulomas is undermuch more subtle immunological influence."4 It hasbeen suggested that the main stimulus to necrosis inthis model is the formation of immune complexesbetween antibodies and excess antigen(mycobacteria) in the centre of the lesion. Thissituation develops when cell-mediated immunity,initially strong, begins to decline for unknownreasons, allowing the mycobacteria to proliferate,out of macrophage control. If cell-mediated immun-ity then improves again the number of bacillidiminishes and immune complexes will form in anti-body excess causing epithelioid granuloma forma-tion instead of necrosis.87 Support for this theory hasbeen obtained from experimental studies withmycobacteria coated with antibodies in differingproportions.'06 It may have important clinical rele-vance in the understanding of infective granulomat-ous diseases, particularly the "reactivation" oftuberculosis and the timing of BCG vaccination."4

Fibrosis is a common and important complicationof granulomatous inflammation because it is oftenresponsible for permanent tissue damage even afterthe causative agent has been eliminated. Thus hepa-tic and pulmonary fibrosis are important long-termcomplications of schistosomiasis and sarcoidosisrespectively. However, until very recently,knowledge of how granulomas lead to fibrosis wasvery scanty. It is clear that permanent fibrosis is notinevitable-most pathologists have seen lungbiopsies with florid granulomatous inflammationbeing followed by apparent complete resolution andsome granulomas, such as those of lepromatousleprosy or carrageenan 115 116 are associated withlittle fibrosis. Generally speaking, non-immunological, low tumover, foreign body typegranulomas appear to stimulate the least amount ofcollagen production. Nevertheless, the granulomat-ous process can also be damped down byimmunological mediators, particularly by suppressorT cells.99-'0'

Experimental studies have illuminated manymechanisms whereby fibrosis within granulomas canbe controlled by the secretions of endogenous cells.The degree of collagenisation is govemed by thebalance between collagen synthesis by activated

Williams, Williams

fibroblasts and collagen degradation, chiefly byneutral proteases. Macrophages have the potentialto affect both sides of this balance. Their presence ishighly desirable for successful wound healing"'7 and,when cultured under appropriate conditions theysecrete substances which increase hydroxyprolineproduction'8 or stimulate proliferation in fibro-blasts. 119 Interleukin- 1, a macrophage productwhich is closely related to endogenous pyrogen, isprobably one such substance,'20 while fibronectin, aglycoprotein secreted by macrophages withimportant roles in cellular adhesion, is a chemotacticagent for fibroblasts.'2' On the other hand,macrophage supernatants which inhibit collagensynthesis have been described,'22 and collagenasesecretion by activated macrophages is wellestablished.34 Lymphocytes also have the potentialfor affecting fibrosis by their secretion oflymphokines which can induce fibroblast migration,proliferation and collagen synthesis.'23 124 Studies onwhole granulomas have also found them to containsubstances that induce fibroblast proliferation.'25 126They probably originate from macrophages orlymphoid cells. Moreover, a study of explantedgranulomas of different types has found that col-lagen synthesis is most active in immunologicalgranulomas and lowest in foreign body lesions,'27corresponding with findings in vivo described above.This suggests that cell-mediated immunity is of con-siderable importance in controlling fibrogenesis butwhether this is achieved by a direct action oflymphoid cell products, or by an indirect effect ofmacrophage activation is uncertain. Epithelioid cellshave been suggested as having a role in fibrosis,45and while to date there is no direct evidence for this,the degree of fibrosis in mycobacterial granulomasdoes correlate with their content of epithelioidcells.128

Conclusion

Granulomatous inflammation represents a distinc-tive tissue reaction to an irritant in which the centralcell is the mononuclear phagocyte cell, but whichcan be modified by other phenomena, especiallyhypersensitivity. The last 25 years have seentremendous improvement in our knowledge of cellbiology, immunology and macrophage function butin spite of this many mysteries continue to surroundthe pathogenesis of organised granulomas and thefunction and significance of their two distinctive celltypes, epithelioid cells and giant cells. Continuedresearch into granulomatous inflammation is essen-tial, not only for its theoretical value, but also for itsimportant potential clinical implications. Betterknowledge of the granulomatous process will both

on August 23, 2020 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.36.7.723 on 1 July 1983. Dow

nloaded from

Page 9: Review article Granulomatous inflammation- areview · The granulomatous inflammatory response is ubiquitous in pathology, being a manifestation of many infective, toxic, allergic,

Granulomatous inflammation-a review

help to elucidate the causes of so-called idiopathicgranulomatous diseases, such as sarcoidosis, Crohn'sdisease or primary biliary cirrhosis, and alsoimprove opportunities for therapeutic intervention.If the destructive properties of granulomas can bereduced while the beneficial functions are amplified,there will be immense scope for preventing thelong-term complications, especially fibrosis, of manyinfective granulomatous diseases and for improvinghost defence, especially against neoplasia. Manipu-lations of this latter kind have already met withlimited success with BCG immunotherapy formalignant diseases, and there is now early experi-mental evidence that cyclosporin A, a modulator ofT lymphocyte function, could have a therapeuticrole in the suppression of epithelioid granulomaformation. 129

We wish to thank Miss Jayne Stitfall for her valuablehelp in preparing the manuscript.

References

lVan Furth R, Cohn ZA, Hirsch JG, Humphrey JH, Spector WG,Langevoort HL. The mononuclear phagocyte system: a newclassification of macrophages, monocytes, and their precursorcells. Bull WHO 1972;46:845-52.

2 Spector WG, Lykke AWJ. The cellular evolution of inflammat-ory granulomata. J Pathol Bacteriol 1966;92:163-77.

3Spector WG, Willoughby DA. The origin of mononuclear cells inchronic inflammation and tuberculin reactions in the rat. JPathol Bacteriol 1968;96:389-99.

4Dannenberg AM Jr, Ando M, Shima K, Tsuda T. Macrophageturnover and activation in tuberculous granulomata. In: vanFurth R, ed. Mononuclear phagocytes in immunity, infectionand pathology. Oxford: Blackwell Scientific Publications,1975:959-80.

sMariano M, Spector WG. The formation and properties of mac-rophage polykaryons (inflammatory giant cells). J Pathol1974;113:1-19.

6Sutton JS, Weiss L. Transformation of monocytes in tissue cul-ture into macrophages, epithelioid cells and multinucleatedgiant cells: an electron microscope study. J Cell Biol1966;28:303-32.

Van Furth R, Diesselhoff-den Dulk MMC, Raeburn JA, vanZwet TL, Crofton R, Blusse van Oud Alblas A. Characteris-tics, origin and kinetics of human and murine mononuclearphagocytes. In: van Furth R, ed. Mononuclear phagocytes,functional aspects. The Hague: Martinus Nijhoff, 1980:279-98.

'Nichols BA, Bainton DF. Ultrastructure and cytochemistry ofmononuclear phagocytes. In: van Furth R, ed. Mononuclearphagocytes in immunity, infection and pathology. Oxford:Blackwell Scientific Publications, 1975:17-55.

9 Quesenberry P, Levitt L. Haematopoietic stem cells. N Engl JMed 1979;301:819-23.

'0 Ward PA. Leukotaxis and leukotactic disorders: a review. Am JPathol 1974;77:520-38.

" Allison AC. Macrophage activation and non-specific immunity.Int Rev Exp Pathol 1978;18:303-46.

12 Volkman A. Disparity in origin of mononuclear phagocyte popu-lations. J Reticuloendothel Soc 1976;19:249-68.

13 Spector WG. The macrophage: its origin and role in pathology.Pathobiol Ann 1974:4:33-64.

41 Van Der Meer JWM, Beelen RHJ, Fluitsma DM, van Furth R.Ultrastructure of mononuclear phagocytes developing in liquidbone marrow cultures: a study on peroxidatic activity. J ExpMed 1979;149:17-26.

,s Van Furth R. Cells of the mononuclear phagocyte system:nomenclature in terms of sites and conditions. In: van Furth R,ed. Mononuclear phagocytes, functional aspects. The Hague:Martinus Nijhoff, 1980:1-30.

16 Isaacson P, Jones DB, Millward-Sadler GH, Judd MA, Payne S.Alpha-1-antitrypsin in human macrophages. J Clin Pathol1981;34:982-90.

71 Breard J, Reinherz EL, Kung PC, Goldstein G, Schlossman SF.A monoclonal antibody reactive with human peripheral bloodmonocytes. J Immunol 1980;124:1943-8.

18 Springer T, Galfre G, Secher DS, Milstein C. MAC-1: a mac-rophage differentiation antigen identified by monoclonal anti-body. Eur J Immunol 1979;9:301-6.

9 Sikora K. Monoclonal antibodies in oncology. J Clin Pathol1982;35:369-75.

20 Michl J. Receptor mediated endocytosis. Am J Clin Nutr1980;33:2462-71.

21 Gordon S, Cohn ZA. The macrophage. Int Rev Cytol1973;36:171-214.

22 Stossel TP. Phagocytosis: recognition and ingestion. SeminHematol 1975;12:83-116.

23 Silverstein SC, Loike JD. Phagocytosis. In: van Furth R, ed.Mononuclear phagocytes, functional aspects. The Hague: Mar-tinus Nijhoff, 1980:895-917.

24 Muiler WA, Steinman RM, Cohn ZA. Membrane flow duringendocytosis. In: van Furth R, ed. Mononuclear phagocytes,functional aspects. The Hague: Martinus Nijhoff, 1980:595-609.

25 LoBuglio AF, Cotran RS, Jandl JH. Red cells coated withimmunoglobulin G: binding and sphering by mononuclearcells in man. Science 1967;158:1582-5.

26 Melewicz FM, Boltz-Nitulescu G, Spiegelberg HL. Fc receptorsfor IgE on human monocytes and rat macrophages. In: Forster0, Landy M, eds. Heterogeneity of mononuclear phagocytes.London: Academic Press, 1981:102-8.

27 Griffin FM Jr, Bianco C, Silverstein SC. Characterization of themacrophage receptor for complement and demonstration of itsfunctional independence from the receptor for the Fc portionof immunoglobulin G. J Exp Med 1975;141:1269-77.

21 Weir DM, Ogmundsdottir HM. Cellular recognition by phago-cytes: role of lectin-like receptor(s). In: van Furth R, ed.Mononuclear phagocytes, functional aspects. The Hague: Mar-tinus Nijhoff, 1980:865-81.

29 Czop JK, Fearon DT, Austen KF. Opsonin-independentphagocytosis of activators of the alternative complementpathway by human monocytes. J Immunol 1978;120:1132-8.

30 Babior EM. Oxygen-dependent microbial killing by phagocytes.N Engl J Med 1978;298:659-68;721-5.

31 Nelson DS. Macrophages: progress and problems. Clin ExpImmunol 1981;45:225-33.

32 Mackaness GB. Cellular resistance to infection. J Exp Med1962;116:381-406.

33 Ogmundsdottir HM, Weir DM. Mechanisms of macrophage acti-vation. Clin Exp Immunol 1980;40:223-34.

34Unanue ER. Secretory function of mononuclear phagocytes. AmJ Pathol 1976;83:396-4.

35 Nathan CF, Murray HW, Cohn ZA. The macrophage as an effec-tor cell. N Engl J Med 1980;303:622-6.

36 Wing EJ, Gardner ID, Ryning FW, Remington JS. Dissociationof effector functions in populations of activated macrophages.Nature 1977;268:642-4.

3" Forster 0, Landy M, eds. Heterogeneity ofmononuclear phago-cytes. London: Academic Press, 1981.

31 Suga M, Dannenberg AM Jr, Higuchi S. Macrophage functionalheterogeneity in vivo. Am J Pathol 1980;99:305-20.

3 Fink MA, ed. The macrophage in neoplasia. New York:

731

on August 23, 2020 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.36.7.723 on 1 July 1983. Dow

nloaded from

Page 10: Review article Granulomatous inflammation- areview · The granulomatous inflammatory response is ubiquitous in pathology, being a manifestation of many infective, toxic, allergic,

732

Academic Press, 1976.40 Weiss SJ, LoBuglio AF. Phagocyte-generated oxygen metabo-

lites and cellular injury. Lab Invest 1982;47:5-18.41 Rosenthal AS. Regulation of the immune response-role of the

macrophage. N Engl J Med 1980;303:1153-6.42 Unanue ER. Cooperation between mononuclear phagocytes and

lymphocytes in immunity. N Engl J Med 1980;303:977-85.43 Pierce CW. Macrophages: modulators of immunity. Am J Pathol

1980i98:10-28." Adams DO. The structure of mononuclear phagocytes differen-

tiating in vivo. Am J Pathol 1974;76:17-40.45 Turk JL. Immunologic and non-immunologic activation of mac-

rophages. J Invest Dermatol 1980;74:301-6.4Ridley DS. Histological classification and the immunological

spectrum of leprosy. Bull WHO 1974;51:451-64.41 Epstein WL, Fukuyama K, Danno K, Kwan-Wong E.

Granulomatous inflammation in normal and athymic miceinfected with Schistosoma mansoni: an ultrastructural study. JPathol 1979;127:207-15.

Tanaka A, Emori K, Nagao S, et al. Epithelioid granuloma for-mation requiring no T cell function. Am J Pathol1982;106:165-70.

"Nath I, Poulter LW, Turk JL. Effect of lymphocyte mediators onmacrophages in vitro. A correlation of morphological andcytochemical changes. Clin Exp Immunol 1973;13:455-66.

50 Poulter LW, Turk JL. Studies on the effect of soluble lymphocyteproducts (lymphokines) on macrophage physiology. CellImmunol 1975;20:12-24, 25-32.

s, Wanstrup J, Christensen HE. Sarcoidosis. I. Ultrastructuralinvestigations on epithelioid cell granulomas. Acta Pathol Mic-robiol Scand 1966;66:169-85.

52 Elias PM, Epstein WL. Ultrastructural observations on experi-mentally induced foreign body and organised epithelioid cellgranulomas in man. Am J Pathol 1968;52:1207-33.

Jones Williams W, Erasmus DA, James EMV, Davies T. Thefine structure of sarcoid and tuberculous granulomas. PostgradMed J 1970;46:496-500.

S4 Turk JL, Badenoch-Jones P, Parker D. Ultrastructural observa-tions on epithelioid cell granulomas induced by Zirconium inthe guinea pig. J Pathol 1978;124:45-9.

11 Ridley MJ. The mononuclear cell series in leprosy: an ultrastruc-tural report. Lepr Rev 1981;52:35-50.

S6 Cain H, Kraus B. Cellular aspects of granulomas. Pathol ResPract 1982;175:13-37.

S7 Williams GT. Isolated epithelioid cells from disaggregated BCGgranulomas - an ultrastructural study. J Pathol 1982;136:1-13.

SWJones Wiiams W. The nature and origin of Schaumann bodies.J Pathol Bacteriol 1960;79:193-201.

Jones Williams W and Williams D. The properties and develop-ment of conchoidal bodies in sarcoid and sarcoid-likegranulomas. J Pathol Bacteriol 1968;96:491-6.

60 Papadimitriou JM, Spector WG. The origin, properties and fateof epithelioid celis. J Pathol 1971;105:187-203.

61 Mariano M, Nikitin T, Malucelli BE. Immunological and non-immunological phagocytosis by inflammatory macrophages,epithelioid cells and macrophage polykaryons from foreignbody granulomata. J Pathol 1976;120:151-9.

62 Mokhtar NM, Spector WG. Facsimile epithelioid cells obtainedfrom stimulated peritoneal macrophages and their secretoryactivity in vitro. J Pathol 1970;128: 117-26.

63 Epstein WL. Granuloma formation in man. Pathobiol Ann1977;7:1-30.

"Williams GT. Isolated epithelioid cells from disaggregated BCGgranulomas - some functional studies. J Pathol 1982;136:15-25.

65 Tannenbaum H, Pinkus GS, Schur PH. Immunological charac-teristics of subpopulations of mononuclear cells in tissue andperipheral blood from patients with sarcoidosis. Clin ImmunolImmunopathol 1976;5:133-41.

Williams, Williams

66Ridley MJ, Ridley DS, Turk JL. Surface markers on lympho-cytes and cells of the mononuclear phagocyte series in skinsections in leprosy. J Pathol 1978;125:91-8.

67 Mishra B, James DG, Poulter LW, Janossy G. Personal com-munication, 1982.

6 James EMV, Jones Williams W. Fine structure and histochemis-try of epithelioid cells in sarcoidosis. Thorax 1974;29:115-20.

69 Pertschuk LP, Silverstein E, Friedland J. Immunohistologicdiagnosis of sarcoidosis. Detection of angiotensin-convertingenzyme in sarcoid granulomas. Am J Clin Pathol1981 ;75:350-4.

70 Studdy P, Bird R, James DG, Sherlock S. Serum angiotensin-converting enzyme (SACE) in sarcoidosis and othergranulomatous disease. Lancet 1978;Hi:1331-4.

71 Gronhagen-Riska C, Fyhrquist F, Pettersson T, Welin M4G,Weber TH. Leukocyte migration inhibiting effect of purifiedhuman lung angiotensin 1-converting enzyme. In: Chretien J,Marsac J, Saltiel JC, eds. Sarcoidosis and other granulomatousdisorders. Paris: Pergamon, 1983:326-31.

72 Weinstock JV, Boros DL. Alteration of granulomaangiotensin-l-converting enzyme activity by regulatory Tlymphocytes in murine schistosomiasis. Infect Immun1982;35:465-70.

73 Chambers TJ. Multinucleate giant cells. J Pathol1978;126:125-48.

74 Murch AR, Grounds MD, Marshall CA, Papadimitriou JM.Direct evidence that inflammatory multinucleate giant cellsform by fusion. J Pathol 1982;137:177-80.

7' Galindo B, Lazdins J, Castillo R. Fusion of normal rabbit alveo-lar macrophages induced by supernatant fluids from BCG-sensitised lymph node cells after elicitation by antigen. InfectImmun 1974;9:212-6.

76 Parks DE, Weiser RW. The role of phagocytosis and naturallymphokines in the fusion of alveolar macrophages to formLanghans giant cells. J Reticuloendothel Soc 1975;17:219-28.

77 Papadimitriou JM. The influence of the thymus on multinucleategiant cell formation. J Pathol 1976;118:153-6.

70 Chambers TJ. Failure of altered macrophage surface to lead tothe formation of polykaryons. J Pathol 1977;122:185-9.

79 Papadimitriou JM, Robertson TA, Walters MN. An analysis ofthe phagocytic potential of multinucleate foreign body giantcells. Am J Pathol 1975;78:343-58.

8 Chambers TJ. Studies on the phagocytic capacity of macrophagepolykaryons. J Pathol 1977;123:65-77.

81 Papadimitriou JM, Archer M. The morphology of murine fore-ign body multinucleate giant cells. J Ultrastruct Res1974;49:372-86.

02 Cain H, Kraus B. The ultrastructure and morphogenesis ofasteroid bodies in sarcoidosis and other granulomatous disor-ders. In: Jones Williams W, Davies BH, eds. Eighth Interna-tonal Conference on Sarcoidosis and Other GranulomatousDiseases. Cardiff: Alpha Omega, 1980:30-7.

03 Black MM, Epstein WL. Formation of multinucleate giant cellsin organised epithelioid cell granulomas. Am J Pathol1974;74:263-74.

8 Knecht H, Hedinger CE. Ultrastructural findings in Hashimoto'sthyroiditis and focal lymphocytic thyroiditis with reference togiant cell formation. Histopathology 1982;6:511-38.

85Boros DL. Granulomatous inflammation. Prog Allergy1978;24: 183-267.

86Boros DL, Warren KS. The bentonite granuloma. Immunology1973;24:511-29.

07 Spector WG, Heesom N. The production of granulomata byantigen-antibody complexes. J Pathol 1969;98:31-9.

"Ginsburg I, Mitrani S, Ne'eman N, Lahav M. Granulomata instreptococcal inflammation: mechanisms of localisation, trans-port and degradation of streptococci in inflammatory sites. In:van Furth R, ed. Mononuclar phagocytes in immunity, infec-don and pathology. Oxford: Blackwell Scientific Publications,1975:981-1010.

on August 23, 2020 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.36.7.723 on 1 July 1983. Dow

nloaded from

Page 11: Review article Granulomatous inflammation- areview · The granulomatous inflammatory response is ubiquitous in pathology, being a manifestation of many infective, toxic, allergic,

Granulomatous inflammation-a review

Smithers SR, Terry BJ, Hockley DJ. Host antigens in schis-tosomiasis. Proc R Soc Lond [Bioll 1969;171:483-94.

Spector WG, Reichold N, Ryan GB. Degradation ofgranuloma-inducing micro-organisms by macrophages. JPathol 1970;101:339-54.

Draper P, D'Arcy Hart P. Phagosomes, lysosomes and mycobac-teria: cellular and microbial aspects. In: van Furth R, ed.Mononuclear phagocytes in immunity, infection and pathology.Oxford: Blackwell Scientific Publications, 1975:575-89.

Spector WG. The granulomatous inflammatory exudate. Int RevExp Pathol 1969;8:1-55.

93 Spector WG, Ryan GB. New evidence for the existence of longlived macrophages. Nature 1969;221:860.

9 Ryan GB, Spector WG. Macrophage turnover in inflamed con-nective tissue. Proc R Soc Lond [Bioll 1970;175:269-92.

g- Turk JL, Parker D. Granuloma formation in normal guinea pigsinjected intradermally with aluminium and zirconium com-pounds. J Invest Dermatol 1977;68:336-40.

46 Turk JL, Parker D. Sensitisation with Cr, Ni and Zr salts andallergic type granuloma formation in the guinea pig. J InvestDermatol 1977;88:341-5.

97 Warren KS. A functional classification of granulomatousinflammation. Ann NY Acad Sci 1976;278:7-18.

98 Warren KS, Domingo EO, Cowan RBT. Granuloma formationaround schistosome eggs as a manifestation of delayedhypersensitivity. Am J Pathol 1967;51:735-56.

99 Pelley RP, Warren KS. Immunoregulation in chronic infectiousdisease: schistosomiasis as a model. J Invest Dermatol1978;71:49-55.

'° Chensue SW, Boros DL. Modulation of granulomatoushypersensitivity. I. Characterization of T lymphocytes involvedin the adoptive suppression of granuloma formation in Schis-tosoma mansoni-infected mice. J Immunol 1979;123:1409-14.

'' Chensue SW, Wellhausen SR, Boros DL. Modulation ofgranulomatous hypersensitivity. II. Participation of Ly 1 + andLy 2+ T lymphocytes in the suppression of granuloma forma-tion and lymphokine production in Schistosoma mansoni-infected mice. J Immunol 1981;127:363-7.

102 Warren KS, Boros DL, Hang LM, Mahmood AAF. TheSchistosoma japonicum egg granuloma. Am J Pathol1976;80:279-93.

103 Olds GR, Olveda R, Tracy JW, Mahmood AAF. Adoptivetransfer of modulation of granuloma formation and hepatosp-lenic disease in murine schistosomiasis japonica by serum fromchronically infected animals. J Immunol 1982;128:1391-3.

104Seal RME, Edwards JH, Hayes M. The response of the lung toinhaled antigens. Prog Respir Res 1975;8:125-9.

05 Sherlock S, Thomas HC, Potter BJ, Jain S, Epstein 0. Primarybiliary cirrhosis as a granulomatous disease. In: Jones WilliamsW, Davies BH, eds. Eighth International Conference on Sar-coidosis and Other Granulomatous Disease. Cardiff: AlphaOmega, 1980:746-9.

106Ridley MJ, Marianayagam Y, Spector WG. Experimentalgranulomas induced by mycobacterial immune complexes inrats. J Pathol 1982;136:59-72.

07 Van Voorhis WC, Kaplan G, Sarno EN, et al. The cutaneousinfiltrates of leprosy. Cellular characteristics and the predo-minant T cell phenotypes. N Engl J Med 1982;307:1593-7.

108 Hunninghake GW, Crystal RG. Pulmonary sarcoidosis: a disor-der mediated by excess helper T-lymphocyte activity at sites ofdisease activity. N Engl J Med 1981;305:429-34.

109 Davies BH, Williams JD, Smith MD, Jones Williams W, JonesK. Peripheral blood lymphocytes in sarcoidosis. Pathol ResPract 1982;175:97-109.

110 Epstein WL. Granulomatous hypersensitivity. Prog Allergy1967;11:36-88.

Simpson LO, Blennerhassett JB, Browett PJ, Newhook CJ.Pseudo-lymphocyte monocytes- the memory cells responsible

for the development of epithelioid cell granulomata. A newhypothesis. Pathology 1981;13:557-67.

112 Epstein WL. Metal-induced granulomatous hypersensitivity inman. Adv Biol Skin 1969;11:313-35.

113 Yamamura T, Ogawa Y, Maeda H, Yaipamura Y. Prevention oftuberculous cavity formation by desensitisation withtuberculin-active peptide. Am Rev Respir Disl974;109:594-601.

114 Spector WG, Marianayagam Y, Ridley MJ. The role of antibodyin primary and reinfection BCG granulomas of rat skin. JPathol 1982;136:41-57.

115 Narayanan RB, Badenoch-Jones P, Turk JL. Experimentalmycobacterial granulomas in guinea pig lymph nodes: ultras-tructural observations. J Pathol 1981;134:253-65.

116 Bowers RR, Houston F, Clinton R, Lewis M, Ballard R. Ahistological study of the carrageenan-induced granuloma inthe rat lung. J Pathol 1980;132:243-53.

117 Leibovich SJ, Ross R. A role of the macrophage in woundrepair. A study of hydrocortisone and antimacrophage serum.Am J Pathol 1975;78:71-100.

118 Heppleston AG, Styles JA. Activity of a macrophage factor incollagen formation by silica. Nature 1967;214:521-2.

19 Leibovich SJ, Ross R. A macrophage-dependant factor thatstimulates the proliferation of fibroblasts in vitro. Am J Pathol1976;84:501-13.

120 Schmidt JA, Mizel SB, Cohen D, Green I. Interleukin 1, apotential regulator of fibroblast proliferation. J Immunol1982;128:2177-82.

121 Tsukamoto Y, Helsel WE, Wahl SM. Macrophage production offibronectin, a chemoattractant for fibroblasts. J Immunol1981;127:673-8.

J22Jimenez SA, McArthur W, Rosenbloom J. Inhibition of collagensynthesis by mononuclear cell supernates. J Exp Med1979;150:1421-31.

123 Postlethwaite AE, Snyderman R, Kang AH. The chemotacticattraction of human fibroblasts to a lymphocyte-derived factor.J Exp Med 1976;144:1188-1203.

124 Wahl SM, Wahl LM, McCarthy JB. Lymphocyte-mediated acti-vation of fibroblast proliferation and collagen production. JImmunol 1978;121:942-6.

125 Wyler DJ, Wahl SM, Wahl LM. Hepatic fibrosis in schis-tosomiasis: egg granulomas secrete fibroblast stimulating fac-tor in vitro. Science 1978;202:438-40.

126 Wyler DJ, Wahl SM, Cheever AW, Wahl LM. Fibroblast stimu-lation in schistosomiasis. I. Stimulation in vitro of fibroblastsby soluble products of egg granulomas. J Infect Dis1981;144:254-62.

27 Boros DL, Lande MA, Carrick L. The artificial granuloma. III.Collagen synthesis during the cell-mediated granulomatousresponse as determined in explanted granulomas. ClinImmunol Immunopathol 1981 ;18:276-86

128 Narayanan RB, Badenoch-Jones P, Curtis J, Turk JL. Compari-son of mycobaterial granulomas in guinea pig lymph nodes. JPathol 1982;138:219-33.

129 Muller-Hermelink HK, Kaiserling E, Sonntag HG. Modulationof epithelioid cell granuloma formation to apathogenicmycobacteria by cyclosporin A. Pathol Res Pract1982;175:80-96.

Requests for reprints to: Dr GT Williams, Department ofPathology, Welsh National School of Medicine, HeathPark, Cardiff CF4 4XN, Wales.

733

on August 23, 2020 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.36.7.723 on 1 July 1983. Dow

nloaded from