Macrophages in Acute and Chronic Inflammation

Macrophages in Acute and Chronic Inflammation* Clemens Sorg, Ph.D.

M

acrophages constitute the mononuclear phagocytic system ·and are present under normal conditions in virtually every tissue. 1 Their most prominent propertiesphagocytosis and cell toxicity- have been extensively studied in recent years. In addition, a large variety of secretory products for macrophages have been characterized and the interaction of macrophages with various cellular and humoral systems has been described. 1 Most notable is the interaction of macrophages with T- and B-Iymphocytes as accessory cells in the induction and regulation of immune responses, 3 with the vascular endothelium (diapedesis, angiogenesis, atberoslerosis), and with other cells in tissue repair (wound healing) and tissue growth (tumor, placenta, bone formation, granuloma formation). 4 Experimental studies on macrophages, however, are hampered by the fact that their behavior in vitro, eg, in plastic dish ~tures, bears little resemblance to their behavior in situ. Further, the procedure of isolating macrophages from tissues usually requires harsh measures that cause damage and induce the macrophages to respond to these insults. 5 With the advent of the monoclonal antibody technique, however, tools became available to investigate macrophages in situ. All antibodies so far described react to a greater or lesser extent with the fresh blood monocytes. Thus, the antibodies detect antigens expressed during the early stages of monon~clear phagocyte differentiation. Our goal, therefore, was to develop monoclonal antibodies for identification ofsubsets of mononuclear pbagocytes in the tissues and for their evaluation in acute and chronic inflammatory reactions. MONOCLONAL ANTIBODIES TO HUMAN MACROPHAGES

Three monoclonal antibodies have been developed and their specificity studied in great detail: 27EIO,8 RM3Il,7 and 25F9. 8 Their reactivity with normal blood cells is shown in Table 1. None of these antibodies reacts with platelets or lymphocytes. Only the antibody 27EIO is positive on granulocytes and monocytes (up to 20%). The antibody RM3Il reacts with 5%-30% of blood monocytes, whereas the antibody 25¥9 shows no reaction at all. The marker 27EIO is maximally expressed by cultured monocytes on day 2 and

declines from day 4 on. The marker RM3Il is also expressed maximally on days 2 to 3 of culture and the~ decreases. On the other band, the marker 2514'9 is found on cultured monocytes only from day 4 on and reaches JDaximal levels by days 8 to 9. This indicates that the markers 27EIO and RM3Il are expressed maximally after the blood monocyte *From the lnstitute of Experimental Dermatology, University of MUnster, Milnster, German~ Reprint ~8t8: Prof Sorg, Institute of Experimental Dermatology, University Of Munster; IJ..44OO Munster; Germany

Table I-Reactivity ojMonoclontJl Antibodia on Peripheral Blood Cella % Positive Cells

Cells

27EI0

RM3Il

25F9

Platelets Lymphocytes Granulocytes Monocytes Monocytes, d 1-4* Monocytes, d 4-13*

0 0

0 0 0 5-30

o o

5-20 5-20

10-50 0-50

o o

30-80 50-80

0-20

20-60

*Monocyte cultures.

state, that is, early during macrophage differentiation. In contrast, the marker 25F9 seems to be expressed only by resting mature macrophages. This conclusion is supported by the tissue distribution of the three different macrophage subsets (Table 2). It also indicates that the antibodies are strictly macrophage speci6c. PHENOTYPIC DYNAMICS OF MACROPHAGES IN

INFLAMMATORY nSSUES

Contact DermatitiB

Table 3 shows the kinetics of appearance of the different macrophage phenotypes in the course of a contact eczema. While some RM3Il-positive macrophages and few scattered 25F9-positive macrophages are always found in human skin, no 27EIo-positive cells are seen. This situation does not change 6 h after elicitation with the contact allergen (nickel sulfate). By 24 h later, however, an influx of 27EIo-positive macrophages is observed and the number ofRM3Il-positive cells is increased. After 72 h, 25F'9-positive cells have accumulated. Gingivitis

These phenotypic dynamics of macrophage subpopulations could be con6nned during human experimental ginTable I-Antigm &preaion ofMonoclonal Antibodia in Normal Human Tiuua 27EI0

RM3I1

25F9

Thymus

0 0 Few monocytes Cells in red pulp 0

Placenta

Few monocytes

Histiocytes Few macrophages Kupffer cells All macrophages in red pulp Macrophages in cortex and medulla

Histiocytes All macrophages Kupffer cells Macrophages in white pulp Macrophages in cortex and medulla 0

TIssues Skin Lung Liver Spleen

in vessels

Many

macrophages

CHEST I 100 I 3 I SEPTEMBER, 1991 I Supplement

1738

Table 3-Reaction ftJttem of Monoclonal Antibodies in

Contact Eczemaa of Human Skin

TIme after Application of Test Allergen (Nickel Sulfate)

27EIO RM3Il

6h

24h

72 h

o

+ ++

+ ++ +

+ ±

25F9

±

givitis. 9 Eight male probands were introduced to an oral hygiene program until all reached mean plaque and gingival index scores approaching zero. During the following 19 days all oral hygiene was abandoned. At various times gingival biopsies were performed. Cryostat sections were incubated with monoclonal antibodies against mature macrophages (25F9), inflammatory macrophages (27EI0), and anti-inftammatory macrophages (RM3Il). Positive cells were counted in the inflammatory infiltrate and the connective tissue. At day -14, elevated numbers of 27EIO-positive cells were observed, which decreased significantly by day 0 and increased again by day 19. Significant differences in the number of RM3Il-positive cells were found between day - 0 and days - 14, - 2, 4 and 7, while no differences in the number of 25F9-positive cells were observed throughout this study. It was concluded that experimental gingival inflammation is characterized by the appearance and disappearance of functionally different macrophage subpopulations. Table 4 summarizes the data obtained by phenotyping the macrophages in various inflammatory tissues. Quite clearly, the 27EIO phenotype is found predominantly in acute inflammatory lesions and is absent from chronic inflammatory tissues, with the exception of lepromatous leprosy tissue. MfPCtIrditis

The detection and phenotyping of macrophages in the cardiac muscle was particularly helpful in the diagnosis of

Table 4-DiBtribution of Macrophaga in Acute and Chronic Inflammatory Taauea

Inftammatory Condition Acute Contact dermatitis Erythrodermia Neurodermitis Psoriasis Gingivitis Chronic Sarcoidosis Erythema nodosum Rheumatoid arthritis Osteoarthritis Granuloma annulare Syphilis II BCG granulonla Tuberculoid leprosy Lepromatous leprosy

1748

27EIO

RM3II

25F9

+

±

±

+

± ± ±

± ± ±

+ + +

myocarditis. IO The diagnosis of myocarditis is difficult to establish. Clinical symptoms of the disease, such as chest pain, fever, palpitations, pulmonary edema, lymphadenopathy, different forms of arrhythmia~ and pathologic electrocardiographic changes, are nonspecific, and histomorphologic definitions are controversial. Moreover, it is still unknown whether the microscopically identifiable inflammatory infiltrate is a destructive process, a repair process, or irrelevant to the myocardial disorder. Cryostat sections of endomyocardial biopsy specimens from 53 patients with the clinical diagnosis of myocarditis were therefore stained using monoclonal antibodies against subpopulations of T-Iymphocytes and macrophages. No inflammatory infiltrate cells were found in 13 cases. Mononuclear cell infiltrates were present in 40 cases. In 10 cases biopsy specimens showed a predominance of macrophages bearing the marker 27EI0, and 18 biopsy specimens contained 25F9-positive macrophages. Patients with immunohistologically confirmed myocarditis had atrial, ventricular, or combined forms of arrhythmias (78%), scars on the vectorcardiogram (100%), and radiologic evidence of cardiomegaly (36%). In comparison with conventional histopathologic techniques, typing ofendomyocardial biopsies from macrophage subpopulations thus proved to be a sensitive new approach to assess the diagnosis of myocarditis. Allotransplants

Patterns of macrophage immigration and differentiation were also investigated in human liver,l1 heart (B. Mues et ai, unpublished findings), and kidney graftS. 12 In all 3 transplantation models a strong infiltration of27E IO-positive macrophages was associated with acute rejection, whereas infiltration and predominance of the RM3Il phenotype was associated with an uncomplicated clinical course. From these studies it appears that the rejection or utake" of an allograft is critically dependent on the type and amount of infiltrating macrophages. Tumors

Most malignant tumors are accompanied by inflammatory infiltrates which, in the human solid tumors so far investigated, are mainly composed of T-Iymphocytes and macrophages. It is still not known, however, whether this inflammatory infiltrate is favorable or detrimental to tumor growth. In a study of primary and metastatic melanomas, accumulation of25F9-positive macrophages in the tumor was clearly associated with a poor prognosis of the patient. 13 Similar results were obtained in gastric carcinoma: the presence of 25F9-positive macrophages was associated with progression of the tumor. I" PHENOTYPIC MODULATIONS IN VITRO AND IN VIVO

+ + +

+

+ + + + + +

+ + + + + + + + ±

From the studies cited above it became clear that the various macrophage phenotypes are associated with different stages of the inHammatory process, but their specific role remained unclear. The 27EIO-positive macrophage appeared to exert aggressive mechanisms against cells and tissues, whereas the 25F9-positive macrophage appeared to support tissue growth and repair. To further elucidate the role of the macrophage phenotypes, in vitro studies using various biologic response modifiers like the chemotactic Macrophages in Acute and Chronic Inflammation (Clemens Borg)

Table 5- Effect of Di.f1ererat Biological CorrapourIIl. on Antigen &preaion of MOROCfIIa*

Antibody 25F9 RM3Il

27EI0

fMLP

Oeu.

LPS

TPA

o

n.d.

o o

o

o o

++

o

+

++

CONCLUSION

IFN--y

o o

+(+)

*fMLP = formyl-methionyl-leucyl-phenylalanine, dexa. = dexamethasone, LPS =lipopoly-saccbaride, TPA = tissue plasminogen activator, n.d. = not done.

,

induces the RM3Il phenotype or acts via the release of corticoids remains to be investigated.

peptide fML~ dexamethasone, lipopolysaccharide (LPS), phorbol myristine acetate, and interferon-gamma (IFN1) were performed (I'able 5). No modulation ofthe 25F9 antigen was observed, whereas LPS, tissue plasminogen activator, and IFN1 upregulated the expression of the 27EI0 antigen. In contrast, the RM3I 1 antigen was strongly upreguJated by dexamethasone and downregulated by phorbol myristine acetate. These findings, together with the data shown in Tables 3 and 4, indicate that the 27EI0 marker is expressed by an inflammatory macrophage type, whereas the RM3Il marker seems to be associated with a late-stage, possibly anti-inflammatory macrophage type. Whether these in vitro effects of g1ucocorticoids have any relevance for the in vioo situation was studied with the intravenous administration of the glucocorticoid prednylidene, followed by assessment of the macrophage subtype RM~l in the peripheral blood of volunteers. 15 Injection of 60 mg of steroid resulted in an approximately 80% increase (from baseline) in the proportion of RM3Il-positive monocytes 12 h after administration. After 24 h the number of RM3Il-positive cells decreased but remained elevated over baseline for at least 15 d. Similar results were obtained after administration of30 mg or 6 mg prednylidene, even though the peak value occurred with a delay of 12 h after a 6-mg dose. A dose- and time-dependent induction of the RM3Ilpositive subtype could also be demonstrated in vitro by the addition of prednylidene or dexamethasone to cultured purified human monocyte-macrophages. Deoxycortone or indomethacin had no effect. These results suggest that glucocorticoids exert their influence on cells of the mononuclear phagocytic lineage by inducing a distinct monocytemacrophage subpopulation that seems to be associated with anti-inflammatory functions. An interesting phenomenon was observed in the following study: Patients who underwent operation for spondylodesis received either thiopental sodium (TrapanaI, Byck Gulden) or etomidate (Hypnomidate, Janssen) as anesthetics (U. Sperl, M. M611mano, E. Michels, C. Sorg, unpublished observations). To investigate the influence of anesthetics on the mononuclear phagocytic system, monocytes in the peripheral blood were analyzed before operation and on days 1 and 8 after operation. Whereas patients who had received Trapanal exhibited 110 changes in the RM~I­ positive monocyte population that could be attributed to Trapanal, patients given Hypnomidate consistently showed a sharp increase from preoperative values (10%-20%) to 809()lf, RM3Il-positive monocytes. On the other hand, no significant changes in the absolute numbers or the ratio of CD4 (helper) to CD8 (suppressor) T-Iymphocytes could be observed in the 2 groups. Whether Hypnomidate directly

Acute inflammatory reactions are dynamic processes that involve the sequential appearance of different macrophage populations. The association of the 27EIO phenotype with acute inflammatory reactions and the RM:lII phenotype with the downregulatory stage of the inflammatory reaction opens new perspectives with regard to diagnosis and treatment of diseases. The dominating role of macrophages in most inflammatory lesions also suggests that the macrophage system could play a central role in a variety of chronic inflammatory diseases. Thus. it would be logical to focus on this cellular system not only in the treatment ofinflammatory diseases but also in their prevention. REFERENCES

1 Furth van R. Current view of the mononuclear phagocyte system. Immunobiology 1982; 161:178

2 Nathan C. Secretory products of macropbages. J Coo Invest. 1987; 79:319 3 Sorg C, Bowers WE. Accessory cells in immune responses.

Immunobiology 1984; 168:133 4 van Furth R, ed. Mononuclear pbagocytes: characteristics, physiology and function. Boston: Martinus Nijhofr, 1985 5 Malorny U, Neumann C, Sorg C. Influence of various detachment procedures on the functional state of cultured murine macrophages. Immunobiology 1981; 159:327 6 Zwadlo G, Schlegel R, Sorg C. A monoclonal antibody to a subset of human monocytes found only in the peripheral blood and in8ammatory tissues. J Immunoll986; 137:512 7 Zwadlo G, \begeli R, Schulze Osthoft'IC, Sorg C. A monoclonal antibody to a novel differentiation antigen on human macrophages associated with the down-regulatory phase of the in8ammatory process. Exp Cell Bioi 1987; 55:295 8 Zwadlo G, BriScker E-B, von Bassewitz DB, Feige U, Sorg C. A monoclonal antibody to a difterentiation antigen present on mature macrophages and absent from monocytes. J Immunol 1985; 134:1487 9 Topoll H, Zwadlo G, Lange DE, Sorg C. Phenotypic dynamics fA macrophage subpopulations during human experimental gingivitis. J Periodont Res 1989; 24:106 10 Mues B, Brisse B, ZwadIo G, Themann H, Bender F, Sorg C. PhenotypiDg of macropbages in endomyocardial biopsies as a new approach to diagnosis of myocarditis. Eur Heart J 1990;

11:1

11 SteinhofF G, \\bnigeit Ie, Sorg C, Behrend M, Mues B, Pichlmayr R. Patterns of macrophage immigration and differentiation in human liver grafts. Transplant Proc 1989; 21:398400 12 Burkhardt K, Sorg C, Andreesen R, Mues B, Hammer C, Land ~ Gurland HJ. Analysis of macrophage subsets in renal allograft biopsies with new moDOC1onal antibodies. Transplant Proc 1990; 22:1844

13 Brticlcer E-B, Zwadlo G, Suter L, Brune M, Sorg C. In<ration of primary and metastatic melanomas with macrophages fA the 25F9-positive phenotype. Cancer Immunol Immunother 1987; 25:81 14 Heidi G, Davaris ~ Zwadlo G, Jagoda MS, D6chtiDg S, BierbofF E, et aI. Association of macrophages detected with monoclonal antibody 25F9 with progression and pathobiological classification of gastric carcinoma. J Cancer Res Coo 0nc0l1987; 113:567 15 Zwadlo-IClarwasser G, Bent S, Haubeck H-D, Sorg C, Schmutzler W Glutoocorticoid-induced appearance of the macrophage subtype RM~1 in peripheral blood of man. Int Arch Allergy Appl Immunoll990; 91:175 CHEST /100 /3/ SEPTEMBER, 1981 / ~

1758

.