In vitro Cytokine Secretion and Maturation Phenotype of lymphoma-Associated Splenic Macrophages

PA1HOLOGY RESEARCH AND PRACTICE © Gustav Fischer Verlag

In vitro Cytokine Secretion and Maturation Phenotype of lymphoma-Associated Splenic Macrophages Stefan KrOger, Kirsten Schuster 1, Alfred C. Feller, Hans-Jochen Stutte 1 and Hartmut MOiler Institut fUr Pathologie, Medizinische Universitat zu Lubeck und 'Senckenbergisches Zentrum der Pathologie, Johann-Wolfgang-Goethe-Universitat, Frankfurt am Main

Summary In this study, cytokine secretion capacity and maturation phenotype of human lymphoma-associated splenic macrophages (LASM) were evaluated in a long-term culture. Sixteen spleens from malignant lymphoma patients and five control spleens were investigated. Splenic macrophages (SM) were isolated by teflon adherence and cultured for 6-48 days. Secretion of IL-l a, IL-6 and TNFa was measured by ELISA following maximal stimulation with LPS and IFN -'¥, and cytokine mRNA expression was detected by in situ hybridization. Immunohistochemical expression of maturationassociated antigens was evaluated semiquantitatively. Cytokine secretion capacity was significantly altered in LASM which exhibited reduced TNFa and IL-6, but elevated IL-l a secretion when compared to control SM. Alterations of cytokine secretion capacity were associated with a modification of LASM maturation phenotype, showing impaired expression of early and chronic/late inflammatory markers. These findings obtained from a long-term culture model suggest that malignant lymphomas induce lasting modifications of cytokine secretion and maturation patterns in LASM. Key words: Long-term culture - Tumor-associated macrophages

Introduction The biological significance of the mononuclear phagocyte system in tumorigenesis and tumor progression has not yet been clarified. Tumor-associated Pathol. Res. Pract. 193: 511-518 (1997)

macrophages (TAM) are known to exert pleiotropic functions which influence tumor growth and progression. Although activated TAM are capable of selectively eliminating tumor cells in vitro, a direct antitumoral effect has never been proven in vivo [16, 18, 19]. On the other hand, TAM are potential promotors of tumor progression by inducing angiogenesis and stroma formation and by secreting cytokines which influence tumor cell growth in vitro [15]. It is assumed that the majority of TAM is sub-optimally activated and lacking tumoricidal activity in vitro [19J. This is probably induced by mediators derived from the tumor itself [5]. For example, lymphoma cells have been shown to produce a variety of cytokines [12, 24]. Additionally, it could be shown that even peripheral blood monocytes are functionally impaired in patients with malignant lymphomas and carcinomas in vitro [17]. The interactions between tumor cells and TAM are currently under investigation. Concerning malignant lymphomas, only few studies have elucidated functional aspects of lymphoma-associated splenic macrophages (LASM). Komanduri et al. [14] reported that macrophage-like stroma cells from murine spleens were able to regulate proliferation of murine RAWI7-HlO lymphoma cells by lysis following direct cell-to-cell contact. Schmidt et al. [22] showed that murine BW5147T lymphoma cells could be influenced by splenic macrophages (SM) and other cellular components of the spleen in their metastatic potential. Finally, nitric oxide production was found to be decreased in LASM during the expansion phase of metastatic tumor growth [26]. The present study was performed to evaluate cytokine secretion and maturation phenotype of macro0344-0338/97/0193-0511 $5.00/0

512 . S. Kruger et al.

phages from spleens with malignant lymphoma involvement and to compare it with that of healthy subjects. A long-term culture model was used in order to answer the question whether malignant lymphomas have a long-term influence on LASM. Morphologically uninvolved spleens of patients with malignant lymphomas were included in our study in order to examine a potential systemic influence of lymphoma cells on LASM. To summarize, the aims of the present study were (I) to evaluate cytokine secretion capacity and (2) to characterize maturation phenotype in SM from lymphoma patients and from healthy subjects in a long-term culture, and (3) to discuss relationships between maturation phenotype and cytokine secretion capacity with special consideration of possible biological implications.

Material and Methods Splenectomy samples

Macrophages were isolated from 16 spleens freshly resected from lymphoma patients. Lymphoma entities are given in Table 1. Patients' age ranged from 28-79 years (mean: 58 years). In all lymphoma patients, clinical stages were III or IV, and none of the patients had received chemotherapy at the time of splenic resection. Five spleens proved to be histologically uninvolved by lymphoma. In these cases malignant lymphoma was diagnosed from lymph node biopsies. SM from five trauma patients without tumor burden served as a control group. Isolation, culture and in vitro activation ofSM

Tissue samples were dissociated mechanically after onehour incubation in sterile phosphate-buffered saline (PBS) containing 10 V/ml penicillin and 100 mg/ml streptomycin followed by mechanical disaggregation with 1% collagenase D. Cell suspension was washed twice and resuspended in RPMI 1640 medium containing 5% human AB serum, 50 mM

Table 1. Histopathological diagnosis of spleens resected from lymphoma patients (n = 16) Histopathological diagnosis spleens involved by lymphoma

=

Hodgkin's disease low-grade NHL high-grade NHL

(n 3) (n =4) (n =4)

spleens morphologically Hodgkin's disease uninvolved by lymphoma low-grade NHL

(n =4) (n = 1)

2-mercaptoethanol, 10 mM sodium pyruvate, 20 mM L-glutamine and antibiotics. SM were separated by seeding the cell suspension into bags made of hydrophobic teflon (Biofolie 25; Heraeus, Hanau, Germany) as described by Andreesen et al. fl]. Adherent cells were at least 95% macrophages as assessed by morphology and immunohistochemical examination with the antobodies MAC387, EBMll, PG-Ml (all Dako, Hamburg) and Ki-M7 (Behring, Marburg). Contaminating cells were unidentified small mononuclear cells. Cells were grown on 8-well tissue culture slide chambers (10 6 cells/well chamber) at 37°C and 5% CO2 for 6, 12, 24 or 48 days. One day prior to examination, SM were stimulated with 1 ng/ml IFN-y and 40 ng/ml LPS. These concentrations yielded a maximal stimulatory effect on cytokine secretion as assessed in preceding experiments. On the other hand, stimulating by IFN-yand LPS had no effect on maturation phenotype as determined by immunohistochemical staining with maturation-associated antibodies. Reagents

Enzyme-linked immunosorbent assay (ELISA) kits for IL-l a, IL-6 and TNFa were from R&D System (Wiesbaden, Germany). In situ hybridisation (ISH) detection kit and digoxigenin-labeled cytokine probes for IL-la, IL-6 and TNFa were purchased from Biermann (Bad Nauheim, Germany). Primary monoclonal antibodies were from Dianova (Hamburg, Germany). The specifities of these antibodies are given in Table 2. Secondary rabbit/anti-mouse antibody and APAAP complex were from Dako (Hamburg, Germany).

Table 2. Primary antibodies used in the study Antibody clone

specifity

source

working dilution

reference

27ElO

monocytes and macrophages of early inflammatory stage

Dianova, Hamburg

1:50

3

RM3/l

mature macrophages of intermediate inflammatory stage

Dianova, Hamburg

1:50

33

25F9

mature macrophages of late inflammatory stage

Dianova, Hamburg

1:50

32

G16/1

mature macrophages of chronic inflammatory stage

Dianova, Hamburg

1:50

20

Lymphoma-Associated Macrophages . 513 Cytokine measurements by ELISA

Supernatants were harvested for measurements of monokines using ELISA assays for IL-I a, IL-6 and TNFa. Briefly, supernatants were pipetted into microtiter plates coated with a monoclonal antibody specific for the cytokine to be measured. After 2-hour incubation at room temperature (RT) , unbound proteins were washed away, and an enzyme-linked polyclonal antibody specific for the cytokine was added to sandwich the cytokine molecules that were immobilized during the first incubation. Following one-hour incubation, any unbound antibody-enzyme reagent was removed by washing, and a substrate solution was added to the microtiter plates to develop color in proportion to the amount of cytokine molecules bound in the initial step. After 20 min incubation, color development was stopped with a stopping reagent, and the intensity of the color was measured on a spectrophotometer (Ultraspec III, Pharmacia LKB Biochrom, Cambridge, England) at a wave length of 450 nm. Cytokine content was calculated using a standard curve. Detection of cytokine mRNA by in situ hybridisation

Cytospin preparations were made from cultured cells, airdried and fixed in buffered parafomaldehyde (4%) for 20 min. After several washing steps in saline sodium citrate (SSC), tris-buffered saline (TBS) and Triton X-I 00, cells were dehydrated in an ascending ethanol series and washed in 0.1 % diethyl pyrocarbonate. Following incubation in SSC for 10 min at 60°C, cytospin preparations were washed in diethyl pyrocarbonate and incubated again in 50 mM Tris buffer (pH 7.6) for 5 min at RT. Negative controls were then incubated in RNase (0.1 mg/ml in SSC) for 60 min at 37°C, while test slides were incubated in RNase-free SSC only, followed by post-fixation in 0.4% buffered paraformaldehyd for 20 min at 4 0c. Next, cytospin preparations were covered with prehy-

Fig. 1. In situ hybridisation for IL-la mRNA in lymphoma-associated splenic macrophages (LASM) after 6 days of culture. Reacitivity is indicated by numerous black dots in the cells' cytoplasm. Slight counterstaining was performed immunohistochemically with a macrophage marker (MAC387, supplied by Dako, Hamburg, Germany).

bridisation solution and incubated for 60 min at 37°C, followed by hybridisation with digoxigenin-labeled cytokine probes (control slides) for 16 hours at 37°C. After several washing steps in SSC, TBS and Triton X-lOO, detection of digoxigenin-Iabeled RNA hybrids was performed using sheep anti-digoxigenin/alkine phosphatase labelled antibody (supplied in the ISH kit; working dilution 1:600; incubation for 30 min at RT). Subsequently, a revealing reaction was introduced by incubating slides for 120 min (37°C) with revealing reagent containing 0.325 mg/ml NBT, 0.166 mg/ml BCIP and 2% levamisole. Slides were then counterstained with hematoxylin for 3 min and mounted with glycerine gelantine. Detected cytokine RNA was visualised as black precipitates within the cytoplasm (Fig. 1). As no quantification was possible, cytokine RNA detection was given as "positive" (+) or "negative" (-). Immunohistochemical staining

For immunohistochemical detection of maturation-associated antigens, a modified three-stage alkaline phosphatase anti-alkaline phosphatase (APAAP) method was used [9]. Briefly, cytospin preparations were fixed in acetone for 10 min and incubated with primary monoclonal antibody (I :50, diluted in TBS containing I% bovine serum albumin) for 60 min at RT. Next, slides were incubated with a secondary rabbit/antimouse antibody (I: 100) for 30 min at RT, followed by incubation with the APAAP complex (I: 100) for 60 min at RT. The alkaline phosphatase activity was visualised using naphthol-AS-Bi-phosphoric acid (0.5 mg/ml) and Fast Red TR (0.01 %). Endogenous alkaline phosphatase was blocked with levamisole (0.5 mg/ml). Immunoreactivity was indicated by a red staining of the cells' cytoplasm. Only cases with a least moderately stained cells were classified as "positive." Cells with no or only weak staining were registered as "negative."

514 . S. Kriiger et al. IL-la: content

A

(lLg/mIJ

600

C

(lL9/ml)

600

•

500

500

•

•

400

t\

300

200

c?

'",

200

,~

...... 6

400

300

100

12

100

~~"'if 24

48

48

culture duration

culture duration

(daysJ

(days)

B IL-6 content (lL9/ml)

Statistical analysis

Statistical analysis was performed using the STATIST program (Maxon Computers, Eschbom, Germany) on an ATARI Mega ST computer. Cytokine levels of different groups were compared using the Mann-Whitney non-parametric U test. Immunohistochemical reactivity of different groups were analysed by chi-square analysis. Level of statistical significance was <0.05 unless stated otherwise.

600

500

•

400

300

TNFa: content

Results

·f

Cytokine secretion capacity ofSM

200

100

6

12 culture duration (days)

Fig. 2 (A-C). IL-la. (A), IL-6 (B) and TNFa. secretion (C) of splenic macrophages (SM) from lymphoma patients (solid lines) and from healthy subjects (dotted lines). For reasons of clearness, data from Hodgkin's and Non-Hodgkin's lymphoma are grouped together. Data from lymphoma patients were analysed separately according to splenic involvement by lymphoma (black squares: spleens involved, grey squares: spleens uninvolved histologically). SM were kept in a long-term culture for 6-48 days and stimulated with LPS (40 ng/ml) and IFN-y (l ng/ml) one day prior to examination. • p < 0.05, # P < 0.01, § P < 0.001 vs. control SM.

In the inital phase (6 to 12 days of culture), activated SM from lymphoma patients secreted significantly more IL-la and significantly less TNFa than control SM (Fig. 2A, 2C). IL-6 secretion was reduced in LASM only on day 6 of culture (Fig. 2B). In all experiments cytokine secretion capacity declined with increasing culture periods and was nearly extinct on day 48 of culture. A remarkable finding is that the secretion capacity of activated SM is similar in involved and histologically uninvolved spleens from lymphoma patients (Fig. 2A-C). Furthermore, cytokine secretion capacity was found to be similar in SM from cases of Hodgkin's disease, low-grade and high-grade non-Hodgkin's lymphoma (NHL) and is therefore independent of the lymphopma entity (data not shown). As indicated in Table 3, RNA of IL-la, IL-6 and TNFa could be detected in the cultivated SM from day 6 to day 24 of culture by ISH (Fig. 1). After 48 days of culture, there was no more cytokine RNA detectable.

Lymphoma-Associated Macrophages . 515 Table 3. Detection of cytokine RNA (lL-la, IL-6 and TNFa) in cultured SM' Cytokine

IL-la mRNA IL-6mRNA TNFamRNA

Culture duration 6 days

12 days

24 days

+ + +

+

+ +

+ +

48 days

+

, results identical in control SM and LASM; results given in "+" (detectable) and "-" (not detectable)

Maturation phenotype ofSM

In contrast to LASM, control SM showed higher ratios of immunohistochemical positivity for the early inflammatory marker 27ElO from day 6 to day 24 of culture (Table 4). A significant difference (p < 0.05) was found On day 24 of culture. 27ElO reactivity was negative On day 48 in all experiments. Expression of late (25F9) and chronic (G 16/1) inflammatory markers was tendentiously enhanced with increasing culture duration (Table 4). Although not statistically significant, a general trend was observed in that the ratios of 25F9 and G16/l positivity were lower in LASM compared to control SM (exception: G 16/1 ratio On day 12). Regarding the intermediate inflammatory marker RM3/1, a peak of reactivity was found On day 12 of culture both in control SM and in LASM.

Discussion In the present study, a long-term culture model for the in vitro examination of TAM is introduced. Although macrophages are long-living cells with a livespan of several months [2], nO studies dealing with the long-term influence of malignant tumors On TAM exist up to now. Additionally, only a few studies have been performed On in vitro cytokine secretion of TAM at all. In most of these studies, peripheral blood monocytes or peritoneal macrophages from "malignant" ascites were examined. Investigation of resident TAM from tumors appears more difficult because problems arise from their separation. According to other studies [7, 28] and own experiments (unpublished), cytokine secretion and LPS responsiveness of TAM are altered significantly following enzymatic tissue disaggregation and macrophage separation by adherence or density gradient centrifugation. Additionally, our Own preceding experiments have shown that stimulation of freshly separated macrophages by IFN-yand LPS yields highly fluctuating values of cytokine secretion during the first days of culture. Thus, the experimental design of our culture model was standardized by introducing a six-day delay before examination of TAM in order to exclude unCOntrollable influences of the isolation procedure On macrophage activation. For the same reasons, unstimulated controls were not included in our study because cytokine levels had been shown to be either near the level of detectability or not reproducible in parallel performed cell separations from the same tumor (data

Table 4. Immunohistochemical reactivity for maturation-associated antibodies 27E 10, RM3/1, 25F9 and G 16/1 in SM from lymphoma patients day 6

day 12

day 24

day 48

27EIO LASM control SM

7/16=44% 4/5 = 80%

7/16= 44% 3/5 = 60%

* 2/16 =

0/16 = 0/16 =

RM3/1 LASM control SM

6/16 = 38% 1/5 = 20%

12/16 = 75% 4/5 = 80%

11/16= 69% 3/5 = 60%

8/16 = 50% 3/5 = 60%

25F9 LASM control SM

4/16 = 25% 3/5 = 60%

12/16 = 75% 5/5 = 100%

11/16 = 69% 5/5 = 100%

14/16 = 88% 5/5 = 100%

G16/1 LASM control SM

1/16 = 6% 1/5 = 20%

1/16 = 0/5 =

$ 3/16 = 19% 3/5 = 60%

$ 6/16 = 38% 4/5 = 80%

6% 0%

13% 3/5 = 60%

0% 0%

27EIO is specific for early, RM3/1 for intermediate, 25F9 for late and G 16/1 for chronic inflammatory macrophages. Reactivity is given as the proportion of positive to negative cases. For reasons of clearness, data from Hodgkin's and Non-Hodgkin's lymphoma were grouped together. , p < 0.05 vs. control SM; $ P < 0.10 vs. control SM

516 . S. Kruger et al. not shown). Apparently, standardized separation and cell culture conditions can not be achieved avoiding non-specific, unpredictable reactions in macrophages. On the other hand, maximal stimulation of cytokine secretion was shown to be reproducibly attained in TAM after 6 days of culture following stimulation with 40 ng/ml LPS and I ng/ml IFN-y. Thus, our model is suitable for studying long-term effects on macrophages, but does not reflect their in vivo status. Nevertheless, the fact that most of the alterations of macrophage function and phenotype were found more pronounced in the early stages of long-term culture may possibly point to alterations of the original (pre-) stimulation and differentiation status in TAM. In our long-term in vitro culture model, significant abnormalities in the cytokine secretion capacity of LASM could be demonstrated. To begin with, activated LASM secreted significantly more IL-I a than control SM (Fig. 2A). IL-l is known to stimulate cellular and humoral immune response and to act as a hematopoietic growth factor [6J. Discrepant impacts of IL-I on the proliferative activity of malignant lymphoma cells have been reported. While an antiproliferative effect on Daudi lymphoma cells was described [21], a stimulatory effect on the proliferation of Epstein-Barr virustransformed Burkitt lymphoma cells was observed [4J. Thus, the biological implication of an elevated IL-l secretion by LASM demonstrated in our study remains speculative. LASM secreted significantly less IL-6 on day 6 of culture (Fig. 2B). IL-6 was described as the central growth factor in multiple myeloma in vivo and in vitro [13J. Additionally, it is supposed to serve as an autocrine growth factor in some Epstein-Barr virus-transformed B cell lymphomas [10, 11, 30J. Since no EBVtransformed B cell lymphomas were included in our study, the biological significance of an initially reduced IL-6 secretions capacity in LASM is equivocal. Considering the fact that IL-6 production is known to be stimulated by IL-l in monocytes and macrophages, it is striking that the increased extracellular IL-l a content observed in our study did not induce elevated IL-6 secretion in LASM. This finding possibly indicates a signal defect in LASM. Our observation that TNFa secretion is reduced in LASM (Fig. 2C) is confirmed by Yamamoto et al. [29J who examined splenic stroma cells from patients with advanced malignant lymphoma. These authors conclude that the tumor-bearing state induces an abnormal cytokine network under which the production of antitumoral cytokines like TNFa is negatively regulated. Cytokine secretion capacity of LASM was shown to be modulated even in spleens where no lymphoma involvement could be histologically detected (Fig. 2A-C). A possible explanation for this finding may be that splenic involvement by lymphoma cells had oc-

cured at a level not detectable by histological examination. According to Veelken et al. [27J, lymphoma cells can be detected by polymerase chain reaction and RNase protection analysis techniques even in morphologically uninvolved tissues from patients with apparently localized malignant lymphomas. Regarding the expression of cytokine mRNA within cultured SM, it is remarkable that IL-l a, IL-6 and TNFa RNA was detectable until day 24 of culture (Table 3). This observation does not correspond to cytokine secretion measured by ELISA since extracellular IL-la and TNFa levels were found nearly extinct on day 24 of culture. A lack of correlation between cytokine secretion and cytokine mRNA expression in activated monocytes and macrophages was also observed by other authors [25, 31]. This phenomenon may be due to a defect in cytokine release or cytokine processing occuring in long-term cultured macrophages. According to Carinci et al. [8J, only young macrophages are capable of synthesizing and processing the IL-l B molecule. In the present study, not only cytokine secretion, but also maturation of LASM was shown to be abnormal in comparison to control SM. LASM had a lower reactivity for early inflammatory antigen (27ElO antibody) in the inital culture period and lower reactivity for late/chronic inflammatory antigen (25F9 and G16/1 antibody) in the late culture period (Table 4). The antibodies 27ElO [3J, RM3/1 [33J, 25F9 [32J and G16/1 [20J are described as maturation-associated markers binding with antigens that are expressed in specific inflammatory stages only. Our data suggest that the early inflammatory phenotype defined by the 27ElO antibody is primarily downregulated in LASM. The fact that upregulation of late/chronic inflammatory markers (25F9 and GI6/1) was performed more slowly in LASM may be interpreted as a delay of maturation in LASM. Functional and phenotypical modifications in monocytes and macrophages are not restricted to the tumor's close vicinity, but do occur also in peripheral blood monocytes from lymphoma patients [17J which are precursors of LASM. Thus, the tumor's influence on cytokine secretion and maturation of the mononuclear phagocyte system is a systemic phenomenon. Regarding different lymphoma entities, similar alterations of cytokine secretion and maturation were found in Hodgkin's and Non-Hodgkin's lymphoma. Thus, one may ask whether even TAM from other tumors than lymphomas would show a similar modification. Indeed, TAM separated from breast carcinomas were found to be modified in their cytokine secretion capacity, but modifications differed from those observed in LASM inasmuch as IL-l a secretion was reduced and IL-6 as well as TNFa secretion was enhanced, while an identical modification of maturation phenotype occurded [23J. Thus, the impacts of malignant tumors of different

Lymphoma-Associated Macrophages . 517

entities seem to be similar on maturation and tumorspecific on cytokine secretion of TAM. Modifications of cytokine secretion may additionally be caused by microenvironmental or organ-specific factors. In conclusion, our findings obtained from a longterm culture model suggest that LASM are subject to lasting modifications in cytokine secretion and maturation which are hallmarked by reduced TNFa and IL-6 secretion, elevated IL-l a secretion and altered expression of maturation-associated antigens. Our data indicate a systemic, long-lasting influence of malignant lymphomas on cells of the mononuclear phagocyte system. The molecular mechanism underlying the complex interactions between malignant lymphoma cells and LASM remain to be identified.

11.

12.

13.

14.

15.

References 1. Andreesen R, Picht J, Ltihr GW (1983) Primary cultures of human blood-born macrophages grown on hydrophobic teflon membranes. Immunol Methods 56: 295-304 2. Auger MJ, Ross JA (1992) The biology of the macrophage. In: Lewis CEO, McGee Jod (Eds) The macrophage, pp 2-74. ILR Press, Oxford 3. Bhardwaj RS, Zotz C, Zwadlo-Klarwasser G, Roth J, Gtibeler M, Mahnke K, Falk M, Meinhardus-Hager G, Sorg C (1992) The calcium-binding proteins MRP8 and MRP 14 form a membrane-associated heterodimer in a subset of monocytes/macrophages present in acute but absent in chronic inflammatory lesions. Em J Immunol 22: 1891-1897 4. B1azar BA, Murphy AM (1990) Induction of B cell responsiveness to growth factors by Epstein-Barr virus conversion: Comparison of endogenous factors and interleukin-I. Clin Exp Immunol80: 62-68 5. Bottazzi B, Erba E, Nobili N, Fazioli F, Rambaldi A, Mantovani A (1990) A paracrine circuit in th regulation of the proliferation of macrophages infiltrating murine sarcomas. J Immunol144: 2409-2412 6. Brugger W, Mocklin W, Heimfeld S, Berenson RJ, Mertelsmann R, Kanz L (1993) Ex vivo expansion of enriched peripheral blood CD34+ progenitor cells by stem cell factor, interleukin-Ip (IL-IP), 1L-6, 1L-3, interferonyand erythropoietin. Blood 81: 2579-2584 7. Brunda MJ, Salich V, Wright RB, Palleroni AV (1991) Tumoricidal activity and cytokine secretion by tumor-infiltrating macrophages. Int J Cancer 48: 704-708 8. Carinci V, Guida S, Fontana MR, Palla E, Rossini M, Malli M (1992) Processing interleukin-l in cells of monocytic lineage is differentiation-dependent. Eur J Biochem 205: 295-301 9. Cordell JL, Falini B, Erker WN, Gosh AK, Abdulaziz Z, MacDonald S, Pulford KAF, Stein H, Mason DY (1984) Immunoenzymatic labelling of monoclonal antibodies using immune complexes of alkaline phosphatase and anti-alkaline phosphatase (APAAP). J Histochem Cytochem 32: 219-229 10. Durandy A, Emilie 0, Peuchmaur M, Forveille M, Clement C, Wijdenes J, Fischer A (1994) Role of IL"6 in

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

promoting growth of human EBV-induced B-cell tumors in severe combined immunodeficient mice. J Immunol 152:5361-5367 Hsu SM, Xie SS, Waldron JA (1992) Functional heterogeneity and pathogenic significance of interleukin-6 in Bcell lymphomas. Am J Pathol141: 915-923 Hsu SM, Waldron JW, Hsu PL, Hough AJ (1993) Cytokines in malignant lymphomas: Review and prospective evaluation. Hum Pathol24: 1040--1057 Klein B, Zahng XG, Jourdan M, Boiron JM, Portier M, Lu ZY, Wijdenes J, Brochier J, Bataille R (1990) Interleukin-6 is the central tumor growth factor in vitro and in vivo in multiple myeloma. Eur Cytokine Netw 1: 193-201 Komanduri KC, Tomes OJ, Gabius HJ, Joshi SS (1995) Macrophage-like-cell-induced growth regulation of lymphoma cells. Oncology 52: 196-201 Mantovani A, Ming JM, Balotta C, Abdeljalil B, Bottazzi B (1986) Origin and regulation of tumor-associated macrophages: The role of tumor-derived chemotactic factor. Biochem Biophys Acta 865: 59-65 Mantovani A, Bottazzi B, Colotta F, Sozzani S, Ruco L (1992) The origin and function of tumor-associated macrophages. Immunol Today 13: 265-270 MUller H, Schuster K, Kruger S, Glienke W, RUbsamenWaigmann H, Stutte HJ (1994) In vitro analysis of HIVand non-HIV-infected monocytes/macrophages from healthy subjects and patients with malignant tumours. Res Vir 145: 183-192 O'Sullivan C, Lewis CE (1994) Tumour-associated leucocytes: Friends or foes in breast carcinoma. J Pathol 172:229-235 Rees RC, Parry H (1992) Macrophages in tumour immunity. In: Lewis CEO, McGee Jod (Eds) The macrophage, pp 317-357. ILR Press, Oxford Ringler OJ, Walsh DG, MacKey 11, Hunt RD, King NW (1988) Immunophenotypic characterization of mononuclear phagocytes and dendritic cells in lymphoid organs of the rhesus monkey. Clin Immunol Immunopathol 49:349-364 Santavenere E, Di Pietro R, Centurione MA, Trubiani 0, Zamai L, Rana R (1994) lL-la antiproliferative and differentiative effects on Daudi lymphoma cells: Multiparametric analysis. Cell Biolint 18: 777-782 Schmidt C, Verschueren H, Toussaint-Demylle 0, van den Berg T, Kraal G, De Baetselier P (1994) The role of the spleen in the organ-specific matastasis of murine BW 5147 T lymphomas. Clin Exp Metastasis 12: 164-174 Schuster K, Kruger S, Stutte HJ, MUller H (1995) Zytokinsekretion und Maturationsverhalten von tumorassoziierten Makrophagen im invasiven duktalen Mammakarzinom. Verh Dtsch Ges Path 79: 486 Takeshita M, Sumiyoshi Y, Masuda Y, Ohshima K, Yoshida T, Kikuchi M, MUller H (1993) Cytokine (interleukinI a, interleukin-I p, tumor necrosis factor ex and interleukin-6) - possessing in lymph nodes of malignant lymphoma. Pathol Res Pract 189: 18-25 Thomassen MJ, Ahmad M, Barna BP, Antal J, Wiedemann JP, Meeker DP, Klein J, Bauer L, Gibson V, Andresen S, Bukowski RM (1991) Induction of cytokine messenger RNA and secretion in alveolar macrophages and blood monocytes from patients with lung cancer re-

518 . S. Kruger et al.

26.

27.

28.

29.

30.

ceiving granulocyte-macrophage colony-stimulating factor therapy. Cancer Res 51: 857-862 Umansky V, Rocha M, Kruger A, von Hoegen P, Schirrmacher V (995) In situ activated macrophages are involved in host resistance to lymphoma metastasis by production of nitric oxide. Int J Oncology 7: 33-40 Veelken H, Wood GS, Sklar J (995) Molecular staging of cutaneous T-cell lymphoma: Evidence for systemic involvement in early disease. J Invest Dermatol 104: 889-894 Verstovsek S, Zaleskis G, Maccubbin DL, Mihich E, Ehrke MJ (1994) Lipopolysaccaride and splenic t~mori­ cidal macrophage activation. J Leucocyte BioI 56: 714-722 Yamamoto N, Zou JP, Li XF, Takenaka H, Noda S, Fujii T, Ono S, Koayashi Y, Mukaida N, Matsushima K, Fujiwara H, Hamaoka T (1995) Regulatory mechanisms for production of IFN-y and TNF by antitumor T cells or macrophages in the tumor-bearing state. J Immunol154: 2281-2290 Yokoi T, Miyawaki T, Yachie A, Kato K, Kasahara Y, Taniguchi N (990) Epstein-Barr virus-immortalized B cells produce IL-6 as an autocrine growth factor. Immunology 70: 100-105

31. Zhong WW, Burke A, Hand AT, Walsk MJ, Hughes LA, Forse RA (993) Regulation of cytokine mRNA expression in LPS-stimulated human macrophages. Arch Surg 128: 158-164 32. Zwadlo G, Brocker EB, von Bassewitz DB, Feige U, Sorg C (1985) A monoclonal antibody to a differentiation antigen present on mature human macrophages and absent from monocytes. J Immunol134: 1487-1492 33. Zwadlo G, Voegli R, Schulze-Osthoff K, Sorg C (987) A monoclonal antibody to a novel differentiation antigen on human macrophages associated with the down-regulatory phase of the inflammatory process. Exp Cell BioI 55: 295-304

Received: September 23, 1996 Accepted in revised form: Juni 23, 1997 Address for correspondence: Prof. Dr. Hartmut Muller, Institute of Pathology, Medical University of Lubeck, Ratzeburger Allee 160, D-23538 Lubeck, Germany. Phone: (0049)451-5002714, Fax: (0049)-451-5004851