Characterization of a murine ovarian reticulum cell sarcoma of histiocytic origin

[CANCER RESEARCH 41, 1271-1280, April 1981]0008-5472/81 /0041-OOOOS02.00

Characterization of a Murine Ovarian Reticulum Cell Sarcoma of HistiocyticOrigin1

James E. Talmadge,2 Marc E. Key, and Ian R. Hart

Cancer Mefastas/s and Treatment Laboratory, National Cancer Institute Frederick Cancer Research Center. Frederick, Maryland 21701

ABSTRACT

We have studied the M5076 tumor, a transplantable murinereticulum cell sarcoma that arose spontaneously in the ovaryof a C57BL/6 mouse. This tumor displays functional andultrastructural characteristics indicating that it is of macrophage origin. Cells from the M5076 tumor are phagocytic, formrosettes with sheep red blood cells, mediate antibody-dependent cellular cytotoxicity against 51Cr-labeled red blood cells,

and display macrophage-like cytotoxicity against syngeneic

tumor target cells but do not exhibit any natural killer cellactivity. The tumor cells possess lysozyme, nonspecific esterase, and phosphatase activities comparable to that seen inrodent macrophages. Ultrastructural examination revealedphagocytic vacuoles and a lack of tight junctions typical ofmacrophage morphology. Karyotype analysis showed thatM5076 tumor cells are hypodiploid with a high percentage(>80%) of metacentric chromosomes that serve as an excellentmarker for identification of these tumor cells.

INTRODUCTION

In the past, the terms reticulum cell sarcoma and histiocyticlymphoma have been applied to a variety of lymphomas believed to have originated from macrophages (10, 33). Classification of these neoplasms formerly was based on morphological criteria alone (33), although histological identification ofreticulum cell sarcomas is extremely difficult (19, 21). Recentstudies, in which immunological and functional markers wereused, have revealed that many of the histiocytic lymphomasoriginated from lymphocytes rather than monocytes (1, 7, 14,19, 21, 35). Relatively few murine or human tumors have beenshown definitively to have originated from monocyte macrophages (1, 7, 14, 19, 21, 35).

In this paper, we report on the in vitro characterization of aspontaneous ovarian tumor from a C57BL/6 mouse which, byfunctional, immunological, and ultrastructural criteria, appearsto have originated from histiocyte macrophages. The in vivobehavior and organ-specific metastatic properties of this tumor

line are described in the accompanying paper (12).

MATERIALS AND METHODS

Mice. Specific-pathogen-free adult mice (6 to 8 weeks old)of the inbred strain C57BL/6 were obtained from the FrederickCancer Research Center's Animal Production Area.

Media. Eagle's minimal essential medium (Auto-Pow) was

obtained from Flow Laboratories, Inc. (Rockville, Md.). This

medium was supplemented with 5% heat-inactivated, endo-toxin-free fetal bovine serum (Reheis Chemical Co., Kankakee,

III.), sodium pyruvate, sodium bicarbonate, nonessential aminoacids, L-glutamine, 2-fold vitamin solution, and 50 jug genta-mycin per ml and designated CMEM.3 The M5076 cells were

grown in RPMI Medium 1640 (Grand Island Biological Co.,Grand island, N. Y.), supplemented with 17% heat-inactivatedequine serum (Flow Laboratories), L-glutamine, sodium pyru

vate, and 50 fig gentamycin per ml. HBSS, pH 7.2, wasobtained from Grand Island Biological Co.

Cell Cultures. The murine tumor M5076 arose spontaneously in the ovary of a C57BL/6 mouse in the laboratory of Dr.W. F. Dunning of the Papanicolaou Research Institute in Miami,Fla. We obtained the tumor from Dr. D. P. Griswold of theSouthern Research Institute, Birmingham, Ala. The tumor wasfrozen and supplied to us in the 136th serial passage insyngeneic mice.

The B16 malignant melanoma variant cell line (F,) was derived from a spontaneous tumor of a C57BL/6 mouse in thislaboratory by Dr. I. J. Fidler (8). The cell line YAC-1 is a T-cell

lymphoma induced by Moloney leukemia virus in A/Sn mice(obtained from Dr. R. Burton, Massachusetts General Hospital,Boston, Mass.). These cell lines were propagated in CMEMand subcultured at a ratio of 1:10 weekly. The M5076 tumorcells were propagated in RPMI Medium 1640, supplementedas described above, and subcultured at a ratio of 1:2 eachweek. The cell lines were incubated at 37° in a humidified

atmosphere of air and 5% CO2.To establish primary cultures of M5076 tumor cells, we

removed ascites-passaged cells aseptically by lavaging theperitoneal cavity of tumor-bearing mice with HBSS. The cellswere washed twice in HBSS and cultured in 100- x 20-mm

tissue culture plates (Falcon, Oxnard, Calif.). Tumors used inthese experiments were examined for and found free of My-

coplasma and the following murine viruses: reovirus type 3,pneumonia virus of mice, K virus, Theiler's virus, Sendai virus,

minute virus, mouse adenovirus, mouse hepatitis virus, lym-

phocytic choriomeningitis virus, ectromelia virus, and lactatedehydrogenase virus (Microbiological Associates, Inc., Walk-

ersville, Md.).Enzyme Assay. Lysozyme activity of cells and supernates

was assayed by measuring the lysis of Micrococcus lysodeik-ticus (Sigma Chemical Co., St. Louis, Mo.) with a Zeiss spec-

trophotometer PM2 DL. Egg white lysozyme (Sigma) was the

1Research sponsored by the National Cancer Institute Contract N01-CO-

75380 with Litton Bionetics. Inc.2 To whom requests for reprints should be addressed.

Received September 29. 1980; accepted December 30. 1980.

3 The abbreviations used are: CMEM. complete Eagle s minimal essential

medium; RPMI Medium 1640. Roswell Park Memorial Institute Tissue CultureMedium 1640; HBSS, Hanks' balanced salt solution; PEC, peritoneal exúdate

cells; LPS. lipopolysaccharide; SRBC. sheep red blood cells; mouse «SRBC.mouse anti-sheep red blood cells (sheep red blood cells coated with appropriateantiserum); rat nSRBC. rat anti-sheep red blood cells (sheep red blood cellscoated with appropriate antiserum); ADCC. antibody-dependent cell-mediatedcytotoxicity; NK. natural killer.

APRIL 1981 1271

on March 31, 2016. © 1981 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

J. E. Talmadge et al.

standard used. The technique of Gordon et al. (12) was utilizedwith only minor modifications.

Nonspecific esterase (39) reactions were determined onslides prepared by standard techniques on a cytocentrifuge(Shandon Elliott, England).

In Vivo Growth Rate. Mice given s.c. injections in the ear orin a caudal mammary fat pad were examined 3 times a weekfor tumor growth. Tumor diameters were measured in 2 dimensions at right angles to one another using a vernier caliper.Tumor volumes were calculated using the following formula:Volume = 0.4 x a x b2, where a = the larger axis and b = the

smaller axis (27).In Vitro Growth Curves. Cells established in tissue culture

flasks were harvested by tapping the flasks by hand, and thedislodged cells were washed in serum-free medium and platedat a density of 105 cells/60-mm tissue culture dish (Falcon) in

5 ml complete RPMI Medium 1640. Cultures were incubated at37°in a humidified atmosphere containing 5% CO2. Triplicate

cultures were counted every 48 hr following culture initiationusing a hemocytometer and trypan blue exclusion.

Karyotyping and Banding. Cells for karyotyping were arrested in mitosis with 10 /tg Colcemid (Sigma) per ml for 1 hrat 37°. The cells were pelleted and washed once in HBSS.

Following hypotonie treatment with 0.5% aqueous KCI for 10min at room temperature, these cells were pelleted gently andfixed in 3 changes of ice-cold methanol:acetic acid (3:1). Thecells were stored at 20°for several days before spreads wereprepared. Spreads were G-banded in 0.001 % trypsin in Ca2+-Mg2+-free HBSS for 12 min at 37°, rinsed in ice-cold distilled

water, and stained in a 5% solution of Giemsa blood stain type620 (Harleco, Gibbstown, N. J.) in Gurr's pH 6.8 buffer for 4

min. Nucleolus organizer regions (11) were stained by incubating slides in a 50% aqueous silver nitrate solution for 18 hrat 56°in a humidified atmosphere. The silver stain was devel

oped in a 3% v/v pH 4.5 aqueous formalin solution for 15 min.Collection and Culture of Macrophages. Resident perito

neal macrophages were collected by lavage from untreatedC57BL/6 mice. Peritoneal macrophages were collected frommice injected with 2 ml of thioglycollate (Becton Dickinson,Cockeysville, Md.) 5 days before cell harvest.

PEC were washed and suspended in HBSS and plated into96-well Micro Test II plates (Falcon). After cell attachment (60min at 37°), the wells were rinsed thoroughly with HBSS to

remove nonadherent cells and then refed with the appropriatemedium.

Collection and Culture of Ascitic M5076 Cells. AsciticM5076 cells were obtained by repeated peritoneal lavage. Thecells so obtained contained less than 5% macrophages andless than 1% polymorphonuclear leukocytes as determined byusing combined morphological and histocytochemical criteriaon cytocentrifuge preparations. These ascitic M5076 cellswere washed twice in HBSS and processed in the same wayas PEC for cytotoxicity assays.

In Vitro Cytotoxicity Assay. Thioglycollate-induced PEC,ascites-grown M5076 cells, or tissue culture-propagated

M5076 cells were assayed for their cytotoxic activity againstB16-F1 melanoma target cells following 24-hr prior incubation

in CMEM or CMEM containing 5 /ig LPS per ml. Cytotoxicitywas assayed using a radiorelease assay as described elsewhere (37). Briefly, B16-F1 melanoma target cells were labeled

by incubation for 24 hr in CMEM supplemented with 0.3 jtCi of

5-[125l]iodo-2'-deoxyuridine (specific activity, 200 ftCi/mol;

New England Nuclear, Boston, Mass.). Following harvestingwith a brief trypsinization (0.25% trypsin and 0.02% EDTA for1 min at 37°), the melanoma cells were washed several timeswith CMEM to remove unbound label and then plated (103

cells/well) into 96-well Micro Test II dishes containing effectorcells (104 cells/well) in the appropriate medium to give a 10:1

ratio of effector to target cells.These cultures were incubated for 3 days at 37°.The assays

were terminated by washing the wells with warm HBSS toremove all nonadherent cells. Adherent cells were removed bylysis with 0.1 ml 0.1 N NaOH and then wiping the wells withcotton-tipped swabs. The radioactivity of these swabs wasmeasured in a Nal crystal well-type y counter.

The percentage of cytotoxicity, mediated by PEC or M5076tumor cells, was calculated as follows:

% of cytotoxicity = 100

cpm in target cells cultured without effector cells—cpm in target cells cultured with effector cells

cpm in target cells cultured without effector cells

Chromium Labeling and Sensitization of SRBC. SRBC thatwere to be used as target cells were washed 3 times with HBSSand counted. These cells (2 x 107) were incubated in CMEMcontaining 200 ¡id 51Cr for 45 min at 37°. The labeled cellswere washed 3 times with CMEM and divided into equal ali-

quots which were pelleted, and the radiolabeled SRBC wereresuspended into 200 /il of heat-inactivated (56° for 30 min)

mouse aSRBC, rat aSRBC, or normal murine serum. Thesecells were incubated for 30 min at 37°and then diluted with

CMEM by vigorous agitation to obtain a single-cell suspension.

These cells were then pelleted and washed again with agitation.ADCC. The desired numbers of macrophages or M5076

tumor cells were added to a final volume of 200 ¡i\of CMEMcontaining 1 x 105 51Cr-labeled SRBC. Cultures were estab

lished in U-bottom Micro Test plates and incubated for 4 hr at37°. Following incubation, the plates were centrifuged at 600rpm for 4 min in a Beckman TJ-6 centrifuge; an aliquot (100/il) of each supernate was removed carefully, and the radioactivity was measured. Total cpm was determined by countingaliquots of 1 x 105 51Cr-labeled SRBC. The percentage of lysis

was calculated as follows:

% of lysis

= 100 x(2 x experimental release - 2 x spontaneous release)

(Total —2 x spontaneous release)

Assessment of Phagocytosis and Fc Receptors. Macrophages (thioglycollate stimulated or resident peritoneal) orascitic M5076 tumor cells were added to the wells of a FalconMicro Test II plate. These cells (1 x 105 in 100 n\ of HBSS)were incubated at 37°for 60 min and vigorously washed with

HBSS to remove nonadherent cells. Radiolabeled SRBC (5 x106), sensitized or not sensitized in antiserum, were added to

each well in 100 fil of CMEM. The cultures were incubated at37°for 60 min and washed 3 times with HBSS. Nonphagocy-

tosed SRBC were removed from some of the wells using a brief(10-sec) hypotonie lysis with distilled water. The number ofadherent erythrocytes was taken as the difference between thecpm obtained from the wells and the wells with hypotonie lysis.

1272 CANCER RESEARCH VOL. 41

on March 31, 2016. © 1981 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

Characterization of a Macrophage Tumor

NK Activity Assay. Target cells for this assay were 51Cr-labeled YAC-1 cells obtained by incubating 2 x 106 cells for40 min at 37°in RPMI Medium 1640 containing 200 ftCi 51Cr.

Labeled cells were washed 3 times with RPMI Medium 1640with 30-min incubation between washes to remove unbound51Cr and to allow for the release of 51Cr from lysed cells.

The cells were then washed and diluted in medium to a finalconcentration of 105 cells/ml. Aliquots (50 jul of cells) were

added to the wells of Micro Test II dishes containing 200 ¡i\ofRPMI Medium 1640. The cells were incubated at 37°for 4 hr

and washed once with CMEM, and effector cells were added.Control effector cells were spleen cells, with the adherent cellsremoved by preincubation for 1 hr at 37°in serum-free medium

on 10- x 60-mm tissue culture dishes. The nonadherent, nu

cleated cells were recovered, counted, and diluted in RPMIMedium 1640, and 200-/J aliquots were added to the targetcells. Test effector cells were ascites-passaged or tissue culture-propagated M5076 cells. Either test or control effector

cells were added at a ratio of 200:1, 100:1, and 50:1/targetcell, and the cell mixtures were incubated for 4 hr. Aliquots of100 fil of supernatant fluid were removed and counted in aBeckman 300 y counter. Spontaneous release (background)was computed from wells containing target cells with RPMIMedium 1640 alone and was always less than 10% of the totalcounts obtained from wells in which the target cells were lysedwith 2% sodium dodecyl sulfate. A total of 4 wells was used foreach data point, and the percentage of release was computedusing the following formula:

% of release

100 x(2 x experimental cpm —2 x spontaneous cpm)

(Total cpm - 2 x spontaneous cpm)

RESULTS

Growth Characteristics in Vitro. Single-cell suspensions of

the M5076 tumor were obtained by mincing a primary tumorand forcing it through a tissue sieve (E-C Apparatus). The

single cells attached very rapidly (<15 min) to tissue cultureplastic or glass in serum-free medium (Fig. 1). These cells

remained attached in the presence of 0.25% trypsin:0.02%EDTA, even after prolonged incubation at 37°.After the primary

culture was incubated in RPMI Medium 1640 with 17% equineserum for 48 hr, the tumor cells became nonadherent and grewas a suspension culture (Fig. IB). These M5076 suspensioncultures contained cells that attached lightly to the substratum,generally without extensive spreading, and could be releasedusing gentle agitation.

The M5076 tumor cells grown in RPMI Medium 1640 containing 17% equine serum have a doubling time of 58 hr (Chart1). Equine serum supported the proliferation of M5076 cells,whereas an equal concentration of fetal bovine serum resultedin reduced growth (Chart 1).

Growth Characteristics in Vivo. The s.c. injection of 25,000M5076 cells into a dorsal-caudal site produced a tumor in 13days and a tumor-doubling time of 2.7 days. The same numberof tumor cells injected into the ventral pinna of the ear produceda tumor in 19 days and had a tumor-doubling time of 4 days.

Tumor cells (25,000) injected i.m. into a caudal footpad hadan induction time of 19 days and achieved an average tumor

50

20

f¿10

Oz

4 6 8 10 12 14

No. of days after culture Initiation

16

Chart 1. In vitro growth of M5076 tumor cells in RPMI Medium 1640 containing either 17% equine serum (•}or 17% fetal bovine serum (A). Cells grown inthe presence of 17% equine serum had a doubling time of 58 hr.

diameter of 0.15 ± 0.04 (S.E.) cu cm 42 days after tumorinjection.

Morphology and Ultrastructure. M5076 cells were highlyanaplastic, whereas the pleomorphic nuclei displayed multiplenucleoli and chromatin margination; the eosinophilic cytoplasmcontained a number of granules. Histological sections of primary tumor masses revealed numerous mitotic figures, andsilver staining demonstrated intracytoplasmic reticulin-like ma

terial, particularly in cells located at the periphery of the tumormass or in well-vascularized areas (Fig. 2).

Transmission electron microscopy failed to reveal desmo-

somes or tight junctions but did show a number of cytoplasmicelectron-dense granules and phagocytic vacuoles (Fig. 3).

Fc Receptors and Phagocytosis. The ascitic and tissueculture-propagated M5076 tumor cells phagocytosed latex

beads and carbon particles in vitro. Tumor cell RBC rosetteswere formed using SRBC coated with subagglutinating levels(1:2048) of rat aSRBC (Fig. 2). Such rosettes were not foundusing unsensitized SRBC. During the 60-min incubation period,

SRBC were phagocytosed also (Fig. 2).M5076 tumor cells were more phagocytic than unsensitized

resident peritoneal macrophages (Table 1), as assayed using51Cr-labeled SRBC coated with mouse or rat aSRBC (aggluti

nating activity, 1:512 and 1:1024, respectively). M5076 tumorcells phagocytized SRBC in 60 min, whereas resident PEC didnot. However, antibody coating to produce opsonized SRBCenhanced phagocytosis by the M5076 tumor cells (Table 1).

Enzymatic Activities. Nonspecific esterase enzyme activities were studied using histochemical stains of cellular cyto-

preparations. The ascitic and tissue culture M5076 cells exhibited enzyme activities, although the tissue culture cellsexhibited a somewhat more intense staining. However, prepa-

APRIL 1981 1273

on March 31, 2016. © 1981 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

J. E. Talmadge et al.

Table 1Adherence and phagocytosis of antibody-sensitized "Cr-labeled SRBC by M5076 tumor cells and

peritoneal macrophages

The number of adherent SRBC was calculated from the difference between total (after washing) andphagocytosed SRBC per culture. Method as in text.

Adherence of SRBC (Fcreceptors)Treatment

ofSRBCSRBC

Mouse uSRBC-sensitizedSRBC

Rat «SRBC-sensitized

SRBCM5076

cells0.07±0.1a

7.0 ±2.15.4

±3.2Peritoneal

macrophages0

1.8 ±0.42.2

±0.7Phagocytosis

ofSRBCM5076

cells2.5

±1.720.4 ±9.818.3

±7.8Peritoneal

macrophages0.05

±0.026.2 ±2.36.4

±3.1

Mean ±S.D.

Table 2

Lysozyme activity of M5076 tumor cells

Enzyme activity

M5076 tissue culture cellsM5076 ascites cells (5% macrophages)0EL4 ascites cells (<1% macrophages)0

Thioglycollate-stimulated macrophages

Triton X lysis fluid controlM5076-induced ascites fluid (cell free)EL4-induced ascites fluid (cell free)M5076 tissue culture supernatant (4 days)d

Control medium

0.17 ±0.05 H9/106 cells0.23 ±0.05/ig/106 cells

<0.01 jig/106 cells0.48 ±0.05/ig/106 cells

0 /ig/ml57.8 ±5.9 Mg/106 cells5.0 ±0.06 jig/106 cells7.8 ±0.09 Mg/106 cells

0 ng/ml

Equivalents of egg white lysozyme. Standard, 40,000 units/mg. Mean ±S.D. of 4 assays.

0 6.5 x 107 cells/ml ascites.c 9.0 x 10a cells/ml ascites.d 1.2 x 10' cells incubated for 96 hr in 20 ml medium.

29

I 20

! 15

O

S 10

I

rffl rn m n n20 30 40 50 60

Chromosome Number

Chart 2. Histogram of chromosome number obtained from 200 metaphasespreads of M5076 tumor cells. Median number of chromosomes, 34.

rations of M5076 tumor cells always exhibited less intensestaining than thioglycollate-stimulated PEC.

Lysozyme activity was seen intracellularly in both ascitic andtissue culture-grown M5076 tumor cells, but lysozyme activityin thioglycollate-stimulated PEC was 2-fold higher than that

contained in M5076 tumor cells. Tissue culture supernatesobtained from suspension cultures of M5076 cells, free ofmacrophages, contained lysozyme levels high enough to suggest that most of the activity was secreted. Cell-free ascitesfluid induced by M5076 tumor cells had very high levels oflysozyme that were not found in control ascites fluid inducedby the T-cell lymphoma EL4 (Table 2).

Karyotypic Analysis. Chromosomal analysis of M5076 cellsindicated a modal number of 34 chromosomes (normal 2N =

Table 3In vitro ADCC by thioglycollate-stimulated peritoneal macrophages and ascitic

or tissue culture-grown M5076 tumor cells against sensitized SRBC5lCr-labeled sensitized (mouse aSRBC) or unsensitized SRBC (105) were

incubated with effector cells at various effector:target cell ratios for 4 hr: 100-/ilaliquots of supernate were removed and counted. Total cpm for sensitized SRBCwere 4306 ± 541 (S.D.; n = 4) and for unsensitized SRBC. 9451 ± 217.Spontaneous release (background) for sensitized SRBC was 604 ±51 cpm andfor sensitized SRBC, 200 ±17 cpm.

% of lysis abovebackground13Effector

cellsPeritoneal

macrophagesAscites

M5076Tissue

culture M5076E:T

ratio310:1

5:12.5:110:1

5:12.5:110:1

5:12.5:1Sensitized

SRBC16.9

16.414.229.1

30.324.118.6

18.314.2Unsensitized

SRBC3.6

2.12.61.3

1.400.6

0.70

Effector: target cell ratio." Percentage of 5'Cr-labeled SRBC lysed in 4 hr.

Table 4

Effect of iodoacetamide on phagocytosis and ADCC cellular cytotoxicity byM5076 tumor cells

Phagocytosis of 5'Cr-labeledSRBC"Coating

antibodyMouse

oSRBCNormal

serumRPMI

Medium16401762

±63372

±64RPMI

Medium1640 + 0.15 X103 M iodoaceta

mide760±13(57)c313

±31ADCC

activity0RPMI

Medium164033.52.7RPMI

Medium 1640 +0.15 X 103M

iodoacetamide32.7ND"

Results are expressed as mean cpm ± S.D. of triplicate cultures; inputvalues of 100-fil samples (6 x 106 RBC) were 11,438 ±331 cpm for mouseaSRBC and 11,829 ±287 cpm for normal serum SRBC; 1 x 105 effector cells;

method as in text.6 Results (from triplicate cultures) are expressed as the percentage of lysis

above background (S.D. < 5%). Effectortarget ratio of 2:1; 4-hr incubationperiod.

0 Number in parentheses, the percentage of reduction in phagocytosis.d ND, not done.

40) with a range from 22 to 56 (Chart 2), indicating that thecells were hypoploid in their chromosome number.

Most of these chromosomes are metacentric (median number of 25), occasionally with 1 to 3 dot-like chromosomefragments (Fig. 4). Using the silver staining method of Good-

pasture and Bloom (11 ), we demonstrated that each metacentric chromosome contained 2 centromeric regions, suggesting

1274 CANCER RESEARCH VOL. 41

on March 31, 2016. © 1981 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

Characterization of a Macrophage Tumor

Table 5In vitro cytotoxicity of thioglycollate-stimulated peritoneal macrophages and ascitic or tissue culture-grown

M5076 tumor cells against syngeneic B i 6 target cells

Culture conditions described in text.

Radioactivity in surviving target cells(cpm)

Calculated kill (%)

EffectorcellsTarget

(B1 6) cells aloneAscites M5076 tumorcellsTissue

culture M5076 tumorcellsPeritonealmacrophages+

LPS2206±243°

248 ±641037±1401079±248-

LPS2286

±294465±1931126±2611920±347+

LPS8965447- LPS794914

Mean ±S.D. of quadruplicate cultures.' Percentage of cytotoxicity compared with target cells cultured alone.

that the metacentrics arose from the fusion of 2 telocentricchromosomes (Fig. 5).

ADCC Activity. M5076 tumor cells and thioglycollate-stimulated peritoneal macrophages were compared for their abilityto participate in ADCC of mouse aSRBC-coated erythrocytes.

A representative experiment (Table 3) demonstrates that thelysis of 51Cr-labeled SRBC is dependent on antibody sensiti-

zation. The ascites-propagated tumor cells were more active in

the lysis of the sensitized SRBC than either the macrophagesor the tumor cells grown in vitro. These results could indeed beattributed to ADCC activity and not to phagocytic activity,inasmuch as the inclusion of the phagocytic inhibitor iodoacet-amide(0.15 x 103w) in the incubating medium failed to reduce

the ADCC activity (Table 4).Cytotoxicity Assay. In vitro cytotoxicity by M5076 tumor

cells and thioglycollate-stimulated peritoneal macrophages wasstudied using the 72-hr assay as described previously (38).The effector cells were stimulated with endotoxin (LPS), and 5-[125l]iodo-2'-deoxyuridine-labeled B16 melanoma cells were

used as targets. Macrophages, even those stimulated by thio-glycollate, are not very cytotoxic unless activated by an agentsuch as LPS (Table 5). The M5076 tumor cells, however, werecytotoxic without activation by LPS, although prior incubationwith this agent did slightly increase their cytotoxicity (p <0.05). The ascitic tumor cells were about 2-fold more active as

effector cells than either the tissue culture cells or macrophages.

NK Cell Activity. M5076 tumor cells and spleen cells fromnude mice were compared for their ability to lyse the NK cell-sensitive target YAC-1. Although in every experiment thespleen cells had NK cell activity against YAC-1 target cells,neither the in vitro- nor the in wVo-propagated cells werecapable of causing 6'Cr release from YAC-1 cells at any effec

tor: target cell ratio (Table 6).

DISCUSSION

The data presented in this paper characterize the M5076tumor as being of histiocyte (macrophage) lineage. The M5076tumor, which arose spontaneously in a C57BL/6 mouse, wasdescribed originally as an ovarian carcinoma (36) and latertentatively identified as a granulosa cell tumor (22) on the basisof light-microscope morphology alone.4 Dunn (5) has com

mented upon the extreme difficulty of differentiating morphologically between granulosa cell tumors and reticulum cell

Table 6NK cell activity by nude mouse spleen cells and ascitic or tissue culture-grown

M5076 tumor cells against YAC-1 target cellsA short-term (4 hr) assay using 51Cr-labeled YAC-1 cells as target cells was

performed.

EffectorcellsNude

spleencellsAscites

M5076cellsTissue

culture M5076 cellsE:T

Ratio"10:15:12.5:110:15:12.5:110:15:12.5:1hcpm16,924

±664C1

1.990 ±2249,450±1685.934

±3085.690±6745.935±2386,108

±4065.930±3606,164±452Cytotoxicity

(%)34.719.912.31.71.01.72.31.72.3

a Effectortarget cell ratio." YAC-1 total cpm were 38,677 ±4,077 (S.D.; n

was 5,350 ±228 cpm.0 Mean ±S.D. of quadruplicate cultures.

•4). Spontaneous release

sarcomas when the primary tumor originates in the ovary.Although the in vivo behavior of the M5076 (13) is indicative ofa reticulum cell sarcoma (3, 28, 29, 32), this study demonstrates the value of using functional assays to help determinethe neoplasm classification.

Reticulum cell sarcomas or histiocytic lymphomas are characterized morphologically by their large cell size and cellularpleomorphism (18, 19, 21). However, functionally and immu-

nologically, many of these tumors have been shown to becomposed of malignant lymphocytes (2, 9). A few human (1, 7,14, 15, 35, 38) and murine (4, 21, 30, 32, 34) tumors havebeen cultured in vitro and shown to be of macrophage originby functional, immunological, histochemical, and enzymaticcriteria.

M5076 tumor cells propagated either in vivo or in vitroexpress membrane Fc receptors by both resetting5 and radio-

label binding studies (Table 1). These cells are also capable ofphagocytosing carbon particles, latex beads, and SRBC. Thephagocytosis of SRBC does not require antibody coating,although opsonization through antibody coating does result inincreased phagocytosis. The demonstration of Fc receptors isconsidered important in the classification of macrophage tumors (21). Both phagocytic activity and Fc receptors havebeen reported for human (17) and murine (16, 31 ) macrophagetumors.

' D. R. Coman (Jackson Laboratories, Bar Harbor, Maine), personal commu

nication.

5 Dr. B. Lane (University of Southern California, Los Angeles, Calif.) has shown

that M5076 tumor cells have Fc receptors for lgG1, lgG2a, and lgG2b by usingmonoclonal antibodies and an erythrocyte resetting assay. Only the Fc receptorfor lgG2a is trypsin sensitive. Personal communication.

APRIL 1981 1275

on March 31, 2016. © 1981 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

J. E. Talmadge et al.

The M5076 tumor cells exhibit antibody-dependent cellularcytolysis of antibody-coated SRBC (Table 3). Further, whenassayed using a 72-hr macrophage cytotoxicity assay, theM5076 tumor cells expressed a macrophage-like effector cell

activity against target B16 melanoma tumor cells. However,unlike macrophages, the M5076 cells do not require activationto be cytotoxic, although preincubation with LPS did increasecytotoxicity slightly (Table 4). In both assays of effector cellactivity (ADCC and macrophage cytotoxicity), the ascitic tumorcells were about 2-fold more active than the tissue culture-

propagated cells. Because host cell contamination of the ascitictumors was low (<10%), it is unlikely that the observed highercytotoxicity was caused by these contaminant host cells.Rather, it appears that in vitro cultivation diminishes the cytotoxic capacity of the M5076 cells. The M5076 tumor cells didnot express any NK-like activity when assayed against YAC-1cells in a 4-hr 51Cr release assay (Table 6). M5076 tumor cells

express many of the enzyme activities, such as lysozyme andnonspecific esterase or phosphatase, commonly associatedwith macrophages (25). Lysozyme activity was demonstratedintracellularly, and lysozyme was secreted in vitro into theculture fluid or in vivo into ascitic fluid. Again, this characteristicwas unlikely to be caused by contaminating host cells sinceboth intracellular and secreted activity could be detected incells passaged in tissue culture for 3 to 4 months which werefree of contaminant host cells.

Like macrophages, primary M5076 cell cultures adhere veryrapidly to glass or plastic without a serum requirement (6).However, on further culture, the cells are released and groweither as a suspension culture or lightly attached to the substrate.

Ultrastructural studies of the M5076 cells failed to reveal anytight junctions, a finding that agrees with the proposed macrophage origin of this tumor line and provides another distinction from granulosa cell tumors which are characterized by thepresence of many of these structures (21 ).

We have included some data on the karyotype of the M5076tumor line because of the relative rarity of the displayed chromosome pattern. Most murine tumors are aneuploid or hyper-diploid (26), whereas relatively few hypoploid neoplasms havebeen described (24). Also, the high frequency of metacentricchromosomes, compared with the more characteristic telocen-tric forms, is unusual (20). Based on the centromeric stainingcharacteristics of these cells, it seems probable that the metacentric chromosomes have arisen as the result of Robertson-ian translocations (23). This distinctive chromosomal patternprovides a good marker for the unequivocal identification ofM5076 cells.

Based on the foregoing criteria, we suggest that the M5076is a reticulum cell sarcoma of macrophage origin. In the following paper, we describe the in vivo behavior of this cell line,which provides a good model for site-specific metastasis (13).

ACKNOWLEDGMENT

We are grateful to Dr. Cora Bucana for the transmission electron photomicrographs.

REFERENCES1. Brouet, J. C., Preud'Homme, J. L., Flandrin, G.. Chelloul, N., and Seligmann,

M Membrane markers in "histiocyte" lymphomas (reticulum cell sarcomas).

J. Nati. Cancer Inst.. 56. 631-633, 1976.

2. Chang, K. S. S., and Loy, T. Natural killer cell activity associated withreticulum cell neoplasms. Int. J. Cancer, 25. 405-416, 1980.

3. Cloudman, A. M. Organophilic tendencies of two transplantable tumors ofthe mouse. Cancer Res., 7. 585-591, 1947.

4. Dawe, G. J., and Potter, M. Morphologic and biologic progression of alymphoid neoplasm of the mouse in vivo and in vitro. Am. J. Pathol., 33:603-611, 1957.

5. Dunn. T. B. Normal and pathologic anatomy of the reticular tissue inlaboratory mice with a classification and discussion of neoplasms. J. Nati.Cancer Inst., 4: 1281-1390, 1954.

6. Edelson, P. J., and Cohn, Z. A. Purification and cultivation of monocytesand macrophages. In: B. R. Bloom and J. R. David (eds.). In Vitro Methodsin Cell-mediated and Tumor Immunity, pp. 333-340. New York: AcademicPress. Inc.. 1976.

7. Epstein, A. L.. and Kaplan, H. S. Biology of the human malignant lymphomas.Cancer (Phila.), 34: 1851-1872, 1974.

8. Fidler, I. J. Selection of successive tumor lines for metastasis. Nat. NewBiol., 242. 148-149, 1973.

9. Fitzgerald, K. L., and Ponzio, N. M. Natural killer activity in reticulum cellsarcomas (RCS) of SJL/J mice. Cell. Immunol., 43: 185-191, 1979.

10. Gall, E. A. The cytological identity and interrelation of mesenchymal cells oflymphoid tissue. Ann. N. Y. Acad. Sci., 73. 120-130, 1958.

11. Goodpasture, C., and Bloom. S. E. Visualization of nucleolar organizerregions in mammalian chromosomes using silver staining. Chromosoma(Beri.), 53. 37-50, 1975.

12. Gordon, S., Todd, J., and Cohn, Z. A. In vitro synthesis and secretion oflysozyme by mononuclear phagocytes. J. Exp. Med., (39. 1228-1248,1974.

13. Hart, I. R., Talmadge, J. E., and Fidler, I. J. Metastatic behavior of a murinereticulum cell sarcoma exhibiting organ-specific growth. Cancer Res., 41:1281-1287, 1981.

14. Henry, K. Tumors of the mononuclear phagocytic system. An ultrastructuralstudy of histiocytic lymphomas. IRCS Med. Sci. Cell Membr. Biol. Immunol.Allergy Pathol., 5: 326-327. 1977.

15. Kaplan, H., Goodenow. R. S.. Gartner, S.. and Bieber, M. M. Biology andvirology of the human malignant lymphomas. Cancer (Phila.), 43: 1-24,1979.

16. Koren, H. S., Handwerger, B. S., and Wunderlich, J. R. Identification ofmacrophage-like characteristics in a cultured murine tumor line. J. Immunol.,114: 894-897, 1975.

17. Larrick, J. W., Fischer, D. G., Anderson, S. J.. and Koren, H. S. Characterization of a human macrophage-like cell line stimulated in vitro: a model ofmacrophage functions. J. Immunol., 725. 6-12, 1980.

18. Leder, L. D. On the terms "reticulosis" and "reticulum cell sarcoma" with

regard to the modern concept of the monocyte macrophage system. Klin.Wochenschr., 56. 1091-1096, 1978.

19. Lennert, K., Stein, H., and Kaiserling, E. Cytological and functional criteriafor the classification of malignant lymphoma. Br. J. Cancer, 3) (Suppl. II).29-43, 1975.

20. Levan, A., Bregula. U., and Klein, G. The stemline idiogram of the MSWBStumor of the mouse and the problem of centric fusion. Hereditas, 70: 283-294. 1972.

21. Lukes, R. J., and Collins, R. D. New approaches to the classification of thelymphomata. Br. J. Cancer, 31 (Suppl. II). 1-28, 1975.

22. MacAulay. M. A., Weliky, I., and Schulz, R. A. Ultrastructure of a biosyn-thetically active granulosa cell tumor. Lab. Invest.. / 7. 562-570, 1967.

23. Miller, O. J., Miller, D. A., Tantravahi, R., and Dev, V. G. Nucleolus organizeractivity and the origin of Robertsonian translocations. Cytogenet. Cell Genet., 20. 40-50, 1978.

24. Natarajan, A. T., Ahnstrom, G., and Raposa, T. Distribution of constitutiveheterochromatin in théchromosomes of MSWBS ascites tumor cells. J. Nati.Cancer Inst., 50. 1721-1726, 1973.

25. Nichols, B. A., Bainton, D. F., and Farquhar, M. G. Differentiation of macrophages. J. Cell Biol., 50. 498-504, 1971.

26. Ohnuki, Y., Mameli, M. M., Babcock, M. S.. Lechner, J. F., and Kaighn, M.E. Chromosomal analysis of human prostatic adenocarcinoma cell lines.Cancer Res., 40. 524-534, 1980.

27. Ovadia, H., Hanna, N., and Nelken, D. Effect of normal immunosuppressiveprotein (NIP) on tumor growth in mice. Eur. J. Cancer. 11: 413-417, 1975.

28. Parks, R. C. Organ-specific metastasis of a transplantable reticulum cellsarcoma. J. Nati. Cancer Inst., 52. 971-973. 1974.

29. Pilgrim, H.I. The kinetics of the organ-specific metastasis of a transplantablereticuloendothelial tumor. Cancer Res., 29 1200-1205, 1969.

30. Ralph, P., Moore, A. S., and Nilsson, K. Lysozyme synthesis by establishedhuman and murine histiocytic lymphoma cell lines. J. Exp. Med.. ¡43:1528-

1533, 1976.31. Ralph, P., and Nakoinz, I. Phagocytosis and cytolysis by a macrophage

tumor and its cloned cell line. Nature (Lond.), 257. 393-394, 1975.32. Ralph, P., Prichard, J., and Cohn. M. Reticulum cell sarcoma: an effector

cell in antibody-dependent cell-mediated immunity. J. Immunol., /14: 898-

905, 1975.33. Rappaport, H. Tumors of the hematopoietic systems. In: Atlas of Tumor

Pathology, pp. 37-48. Washington, D. C.: Armed Forces Institute of Pathology, 1966.

1276 CANCER RESEARCH VOL. 41

on March 31, 2016. © 1981 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

Characterization of a Macrophage Tumor

34. Raschke, W. C., Baird, S., Ralph, P., and Nakoinz, I. Functional macrophagecell lines transformed by Abelson leukemia virus. Cell, i 5: 261 -267. 1978.

35. Said, J. W., Hargreaves, H. K., and Pinkus, G. S. Non-Hodgkins lymphomas.

An ultrastructural study correlating morphology with immunologie cell type.Cancer (Phila.), 44: 504-528, 1979.

36. Simpson-Herren, L., Griswold, D. P., and Dykes, D. J. Population kineticsand chemotherapeutic response of transplantable ovarian carcinomaM5076. Proc. Am. Assoc. Cancer Res., 20: 80, 1979.

37. Soné,S., and Fidler, I. J. Tumor cytotoxicity of rat alveolar macrophagesactivated in vitro by endotoxin. RES J. Reticuloendothel. Soc., 27. 269-279, 1980.

38. Sundstrom, C., and Nilsson, K. Establishment and characterization of ahuman histiocytic lymphoma cell line (U-937). Int. J. Cancer, 77. 565-571.1976.

39. Yam, L. T., Lee, G. Y., and Crosby, W. H. Cytochemical identification ofmonocytes and granulocytes. Am. J. Clin. Pathol., 55. 283-290, 1970.

Fig. 1. Plating behavior of M5076 tumor cells examined by phase microscopy. In A. ascites-derivedM5076 cells incubated in serum-free medium for 15 minshow initial spreading and attachment to plastic substratum, x 200. In B. incubation of cells for 48 hr in RPMIMedium 1640 supplemented with 17% equine serumresults in a general rounding up of the tumor cells andrelease from attachment. Viability >98% as assessedby trypan blue exclusion, x 40.

•Vii.

*•.'.-;v» •%>¿s*%.**<?'•

•o* f o."»A «•»•

APRIL 1981 1277

on March 31, 2016. © 1981 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

J. E. Talmadge et al.

I

-

r- 2BFig. 2. A silver-stained section of primary M5076 tumor growing in a s.c. site; arrows, intracellular reticulin-like material, x 315; B, tumor cell-SRBC rosettes

formed after incubation of M5076 tumor cells with SRBC coated with subagglutinating levels of rat aSRBC antiserum; arrows, phagocytosed SRBC incubation periodof 60 min. x 500.

Fig. 3. Electron micrograph of M5076 metastatic deposit in the ovary of C57BL/6 mouse showing presence of what appear to be phagocytic vacuoles in tumorcells, x 7000.

Fig. 4. Two karyotypes derived from metaphase spreads of M5076 cells stained with Giemsa as described in the text. The majority of chromosomes aremetacentric. X1000

1278 CANCER RESEARCH VOL. 41

on March 31, 2016. © 1981 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

Characterization of a Macrophage Tumor

.^Jr

»»

Ifl'lf Õ> i Õ( •l i * »

^•^ w^

I »APRIL 1981 1279

on March 31, 2016. © 1981 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

J. E. Talmadge et al.

i ; n«e**«*

î

; i ! u v »- i , :

^ ~ 4B-w*^

Fig. 5. Metaphase spread from M5076 cell stained with silver nitrate solution as described in the text. Arrows. 2 darkly stained centromeric regions, suggestingthat metacentric chromosomes arose by fusion of 2 telocentric chromosomes, x 1260.

1280 CANCER RESEARCH VOL. 41

on March 31, 2016. © 1981 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

1981;41:1271-1280. Cancer Res   James E. Talmadge, Marc E. Key and Ian R. Hart  Histiocytic OriginCharacterization of a Murine Ovarian Reticulum Cell Sarcoma of

  Updated version

  http://cancerres.aacrjournals.org/content/41/4/1271

Access the most recent version of this article at:

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  [email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  [email protected] at

To request permission to re-use all or part of this article, contact the AACR Publications

on March 31, 2016. © 1981 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from