[CANCER RESEARCH 45,1320-1327, March 1985]

Direct and Indirect Effects of Interferon on in Vivo Murine Tumor Cell Growth1

Kazuko Uno, Satoshi Shimizu, Motoharu Ido, Koji Naito, Kayo Inaba, Tohru Oku,2 Tsunatarou Kishida,3 andShigeru Muramatsu4

Department of Zoology, Faculty ol Science, Kyoto University, Kitashirakawa-Oiwakecho, Sakyo-ku, Kyoto 606 [K. U., S. S., M. I., K. N., K. I., S. M.¡,and Department of

Microbiology, Kyoto Prefectura! University of Medicine, Kyoto 602 [T. O., T. K.], Japan

ABSTRACT

We cloned two sublines (S1 and R1) of murine Meth A fibrosarcoma cells with respect to their sensitivity to a murine a/ß-

interferon (IFN) preparation. The growth of S1 cells was suppressed and that of R1 cells was hardly affected by IFN in vitro.This was also the case with cells enclosed in cell-impermeable

diffusion chambers in peritoneal cavities. Nevertheless, IFN suppressed the growth of not only S1 cells but also R1 cells in miceinoculated i.p. with these cells, and the survival rates of both S1cell recipients and R1 cell recipients were markedly improved.S1 cells were observed microscopically to be injured by thedirect effect of IFN in vitro and in vivo, but R1 cells in in vitroculture with IFN and those surviving in vivo in the presence ofIFN appeared to proliferate well. In the peritoneal cavity of R1recipients treated daily with IFN, the recruitment of macrophageswas enhanced in comparison with untreated R1 recipients. Adherent peritoneal exúdate cells obtained from IFN-treated, R1-

bearing mice were highly suppressive for the in vitro growth ofnot only R1 cells but also allogeneic and human cells. The roleof macrophages in the indirect effect of IFN on tumor cell growthis discussed.

INTRODUCTION

Investigations in vitro have demonstrated that IFN5 sup

presses, although not always, tumor cell growth (1, 15, 28).Thus, it seems reasonable to expect that IFN also exerts asuppressive effect on the in vivo tumor cell growth. This ideahas been supported by some workers through investigations onthe antitumor effect of IFN in mice (10-12), although the in vivo

efficacy of IFN has not yet been unequivocally admitted (8, 25).The in vivo antitumor activity of IFN may be manifested not

only by the direct action of IFN on tumor cells but also by theindirect effect mediated by some host defense mechanism, sinceIFN is known to be a potent activator for macrophages (7, 26)and for NK (9, 20, 21). Gresser ef a/, isolated IFN-sensitive and-resistant sublines in vitro from murine leukemia L1210 cell (14,

16) and Friend leukemia cell (2) populations and demonstratedthat daily administrations of IFN preparations suppressed the invivo growth of these subline cells, irrespective of the differencein IFN sensitivity, to improve the survival rate of tumor recipients.

1This work was supported by a Grant-in-Akj for Cancer Research from the

Ministry of Education, Science, and Culture of Japan.2 Present address: Department of Biochemistry, Kyoto Prefectural University of

Medicine, Kyoto 602, Japan.3 Present address: Department of Microbiology, Biochemical Technology Insti

tute, Kobe Tokiwa Junior College, Kobe 650. Japan.4To whom requests for reprints should be addressed.5The abbreviations used are: IFN, interferoni FBS, fetal bovine serum; MST,

median survival time; PEC, peritoneal exúdate cell; NK, natural killer cell.Received June 22, 1984; accepted November 12, 1984.

Thus, the IFN-resistant tumor cells seem to have been indirectly

affected by IFN.This series of investigations was undertaken in order to eluci

date the mechanism of the indirect effect of IFN on the in vivogrowth of tumor cells. We isolated an IFN-sensitive subline (S1)

and a resistant subline (R1) from a murine fibrosarcoma Meth Acell population. Mice were inoculated with either of these sublinecells and received daily administrations of murine IFN preparation. The data to be presented in this paper, which forms thefirst part of our serial reports, indicate that IFN exerts a directeffect on S1 cells and an indirect effect on R1 cells to suppressthe growth of these cells. Circumstantial but strong evidence forthe participation of macrophages in the indirect effect of IFN isalso given.

MATERIALS AND METHODS

Mice. Male and female BALB/cCrSIc mice (Shizuoka Agricultural,Cooperative Association for Laboratory Animals, Shizuoka, Japan) 2 to

4 months old, were used. No sexual difference was observed in the invivo Meth A cell growth.

Tumor Cells. Meth A cells, from a methylcholanthrene-induced fibro

sarcoma of BALB/c mice, maintained by i.p. passages served as theoriginal tumor cell source from which S1 and R1 cell sublines wereisolated as described later.

IFN. Partially purified murine «//3-IFNwas prepared from the culture

supernatant of L929 cells infected with Newcastle disease virus, asdescribed previously (30). The IFN was titrated on L929 cells withvesicular stomatitis virus against the NIH mouse IFN standard (G-0020904-511). Several lots of IFN were used in this study, and each lot

was titrated in triplicate cultures. The range of specific activity amongthe IFN preparations was 3 to 9 x 106 lU/mg protein. In the in vivo

experiment, IFN was diluted with 0.15 M NaCI and injected i.p. in avolume of 0.5 ml. Control mice received 0.15 M NaCI alone.

Anti-IFN Serum. Anti-mouse IFN sheep serum was provided by Dr.

N. Minato, Jichi Medical School (23). The antiserum was preabsorbedwith BALB/c thymus and spleen cells for 30 min at 4°.Normal sheep

serum, treated similarly, served as a negative control. The antiserum ata final dilution of 1:1200 perfectly neutralized IFN at 105 Ill/ml.

Diffusion Chambers. Cell-impermeable diffusion chambers were pre

pared according to the method of Groves ef a/. (17). The chamber wasassembled with an acryl ring (20 mm outer diameter, 10 mm inner

diameter, 6 mm thick) and membrane filters (0.22 Mm porosity; MilliporeCo., Bedford, MA) for covering both sides of the ring. Each chamberwas filled with 104 S1 or R1 cells and was implanted into the peritoneal

cavity of BALB/c mice. Cells to be enclosed in the diffusion chamber hadbeen cultured in vitro, washed, and resuspended in RPMI 1640 (NissuiSeiyaku Co., Ltd., Tokyo, Japan). One day after the last injection of IFNor 0.15 M NaCI, the chamber was removed, wiped with cotton wool, andsoaked in 5 ml of 0.5% Pronase in a small beaker. The fibrin clot in thechamber was dissolved by shaking and pipeting at room temperatureafter the membranes were broken. The number of viable cells recovered

CANCER RESEARCH VOL. 45 MARCH 1985

1320

on April 14, 2017. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ANTITUMOR EFFECT OF IFN IN VIVO

from each chamber was counted by trypan blue dye exclusion test.Culture Conditions. RPM11640 supplemented with 10% FBS (Grand

Island Biological Co., Grand Island, NY), 5 HIM L-glutamine, and penicillin(100 units/ml) and buffered by 5 rriM 4-(2-hydroxyethyl)-1-piperazineetha-

nesulfonic acid and NaHCO3 at pH 7.0, was used as the culture medium.Tumor cells were maintained in 10-cm dishes (Falcon No. 3001 ; Becton-

Dickinson, Oxnard, CA) with passages at 3-day intervals. In experimentalpractice, 2.5 x 10* tumor cells each in 1 ml of the medium were seeded

on 16-mm dishes in 24-well multidish plates (A/S Nunc, Kamstrup,

Roskilde, Denmark). IFN was added to the culture in a volume of 50 //I.In Vivo Experiments. In order to acclimatize in wfro-cultured cells to

the in vivo conditions and to remove the components of culture medium,1 to 2 x 106 S1 or R1 cells cultured in vitro were injected i.p. into BALB/

c mice and were allowed to expand. Tumor cells were harvested 10 to14 days later, washed with Hanks' solution, and used for in vivo exper

iments.The in vivo growth of tumor cells were determined by counting the

number of S1 or R1 cells recovered from IFN-treated or untreated mice.

Mice were exsanguinated by severing carotid arteries, and cells in theperitoneal cavities were harvested by lavage with Hanks' solution. Viable

cells were counted by the trypan blue dye exclusion test with a Thomahemocytometer. Meth A cells inoculated i.p. did not form solid tumors,and the ascitic Meth A cells were easily distinguishable from host PECby their large size. Aliquots of cell suspensions were smeared onto slideglasses by means of a cytocentrifugation device (Model BEST SC-2;Tomy Works, Ltd., Tokyo, Japan), fixed, and stained with May-Griinwald-

Giemsa stain for microscopic observation.Preparation of Crude and Purified Macrophage Populations from

PEC of R1 Cell Recipients. Cells in the peritoneal cavity harvested frommice in each experimental group by lavage with Hanks' balanced salt

solution were pooled, washed 3 times, and resuspended in RPM11640supplemented with 10% FBS, 5 mM L-glutamine, and penicillin (100 units/ml) and buffered with 5 mM 4-(2-hydroxyethyl)-1-piperazineethanesul-

fonic acid and NaHC03 at pH 7.O. Routinely, crude macrophage populations were prepared according to the method of Kumagai ef al. (19)with modifications. Thus, appropriate numbers of peritoneal cells wereplated on 10-cm plastic Petri dishes (Falcon No. 3003; Becton-Dickinson

and Co., Oxnard, CA) which had been coated with FBS. After incubationfor 2 hr at 37°,R1 cells and nonadherent PEC were removed thoroughly

by 3 washes. Adherent cells were removed twice after incubations with1.2 mw EDTA in 0.01 M phosphate-buffered 0.15 M NaCI for 5 and 10

min each at room temperature. The cells recovered were washed 3 timesby centrifugation, resuspended in the culture medium, and used as crudemacrophage populations. These populations contained neutrophils anda very low percentage of lymphoid cells in addition to macrophages, andthe macrophage content varied in the range from about 50 to 90%depending on the neutrophil content of PEC populations.

A purified macrophage population was prepared by centrifuging thecrude macrophage population on Percoli colloid (density, 1.085; Pharmacia Fine Chemicals, Uppsala, Sweden) for 10 min at 1800 x g. Thecells floating on the Percoli were collected, washed, and resuspended inthe culture medium. The nonspecific esterase-positive and phagocytic

macrophage content of this population was more than 99%.Evaluation of the Effect of Crude and Purified Macrophage Popu

lations on Tumor Cell Growth in Vitro. Various numbers of cells ofmacrophage populations and 2.5 x 104 R1 cells were cultured in 16-mmdishes at 37°in a 5% CO2-95% air atmosphere, and viable R1 cells were

counted 48 hr later. The antitumor activity of macrophage preparations

was evaluated as the percentage growth of R1 cells, calculated as

RESULTS

Isolation of IFN-sensitive and IFN-resistant Sublines. AsciticMeth A cells were cultured in 10-cm dishes for 1 month withpassages at 3-day intervals in order to be adapted to the in vitro

condition. Single cells picked up with a fine Pasteur pipet underan inverted microscope were transferred each into a 16-mm dishof 24-well plates and cultured for clonal expansion. Each clone

thus obtained was divided into control and experimental groupsto check the sensitivity to IFN (10,000 Ill/ml). A clone proliferatingwell in the absence of IFN but very badly in the presence of IFNwas chosen as an IFN-sensitive subline, termed S1, and has

been maintained by in vitro passages without alteration of theoriginal properties. Chart ~\A represents the in vitro growth

kinetics of S1 cells in the presence of varying concentrations ofIFN; the growth rate of S1 cells decreased as the IFN doseincreased, although heat-inactivated IFN equivalent to 10,000 IU

did not affect the S1 growth. Chart 1ßshows that the effect ofIFN on S1 cells was completely abolished in the presence ofanti-IFN serum.

On the other hand, a part of the mass-cultured Meth A cells

mentioned above were further cultured to grow in the presenceof IFN at 3,000 lU/ml for 2 weeks and 10,000 lU/ml for another2 weeks with passages at 3-day intervals. The cells still growing

were cloned as described above, and a clone proliferating wellin the presence of IFN (10,000 ID/ml) was obtained. This clonewas termed R1. The resistance of R1 cells to IFN (10,000 ID/ml)is shown in Chart 2.

Effect of IFN on S1 and R1 Cells in Diffusion Chambers.Mice were inoculated i.p. with diffusion chambers containingeither 104 S1 or 10" R1 cells at Day 0 and received daily i.p.

injections of IFN (50,000 IU) or 0.15 M NaCI from Day 3 to Day5. The number of cells in diffusion chambers was counted atDay 6. Results are shown in Chart 3. The effect of IFN wasessentially similar to that in the in vitro culture. S1 and R1 cellsin control groups grew well, and IFN suppressed the proliferationof S1 cells (p < 0.05) but not R1 cells.

Effect of IFN on the Survival Rate of S1 and R1 CellRecipients. Mice inoculated i.p. with 10" S1 or R1 cells at Day

0 received daily i.p. injections of 50,000 IU IFN or 0.15 M NaCIfrom Day 3 to Day 13. These mice were left untreated thereafter,

Days Days

x 100No. of tumor cells cultured with macrophage population

No. of tumor cells cultured without macrophage population

Statistics. The data were treated statistically with Student's t test.

Chart 1. Sensitivity of S1 cells to IFN. In A, S1 cells were cultured with IFN atconcentrations of 100 lU/ml (x), 1,000 lU/ml (A), 3,000 lU/ml (A), or 10,000 lU/ml(D) or without IFN(O).9, lack of suppressiveeffect of neat-treated IFN (80°for 30min) at a concentration equivalent to 10,000 ID/ml In 8, S1 cells were culturedwith (D) or without (O) 10,000 IU IFN/ml in the presence of anti-IFN serum (•)ornormalsheep serum (•).Points, geometric means of triplicate cultures; bars, S.E.

CANCER RESEARCH VOL. 45 MARCH 1985

1321

on April 14, 2017. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ANTITUMOR EFFECT OF IFN IN VIVO

Days

Chart 2. Resistance of R1 cells to IFN. R1 cells were cultured with IFN (10,000Ill/ml; D) or without IFN (O). For other details, see the legend to Chart 1.

1C*g*u•61,103-

IFN(lUtíay)SI•Af

*Ai

0 5x10*.-RI*

A

A|7AÓ

5x10*

Chart 3. Sensitivity of S1 cells and resistance of R1 cells to IFN in vivo withoutcontact with host cells. S1 or R1 cells (104) were enclosed in diffusion chambers

and implanted i.p. on Day 0. The recipient mice received daily injections of 50,000IU IFN (A) or 0.15 M NaCI (•)from Day 3 to Day 5. The number of cells wascounted at Day 6. Points, number of cells in each chamber; bars, mean.

Table 1

Improvement of survival rate of mice inoculated with S1 or R1 cells by theadministration of IFN

Mice inoculatedon Day 0with10"

S1 cells

10"R1 cellsDaily

injectionsfrom Day 3 to

Day130.1

5 M NaCI50,000 IUIFN0.1

5 M NaCI50,000 IU IFN60-day

sur

vivors/totalmice0/10

2/100/14

2/9MST

(days)18.4

44.114.0

34.8Confidence

interval16.9-20.2

37.3-53.813.3-14.8

28.5-44.8P<0.01<0.01

and the date of death was recorded. MST after tumor inoculationwas calculated according to the method of Grosser ef al. (10).Results are shown in Table 1.

IFN was found to be effective in improving not only the survivalrate of S1 recipients but also that of R1 recipients. All mice incontrol groups given 0.15 M NaCI without IFN died in the firstmonth of the experiment with MSTs of 18.4 days in S1 recipientsand 14.0 days in R1 recipients, and the administration of IFNsignificantly increased the MST to 44.1 and 34.8 days, respec

tively. It may be surmised from the data in Chart 5, that S1 andR1 cells at a quantity around 108/mouse induced the death of

hosts. Among 8 mice receiving S1 cells and IFN, 4 mice whichdied at Day 38 or later did not bear a fatal number of ascitictumor cells. Thus, their death seemed to be caused not by ascitictumor cells but by s.c. solid tumors which developed from asmall number of S1 cells leaking from peritoneal cavities throughholes made by syringe needles.

Effect of IFN on in Vivo Growth of S1 and R1 Cells. Micewere inoculated i.p. with S1 or R1 cells, and IFN or 0.15 M NaCIwas administered i.p. from Day 3 to Day 5, 9, or 13. One dayafter the last injections, cells in the peritoneal cavities wereharvested, and viable cells were counted. In this experiment, thedose of tumor cells inoculated initially and that of IFN per injectionwere fixed at 106 and 50,000 IU, since we observed in a prelim

inary experiment that the combination of these doses was suitable for obtaining the obvious effect of IFN.

S1 cells in control mice grew almost exponentially, while thenumber of those in IFN-treated mice was found to be significantly

smaller than that in control mice at Day 6 and scarcely increasedthereafter (p at Days 6, 10, and 14 were <0.02, <0.01, <0.01,respectively.) (Chart 44). On the other hand, the effect of IFN onR1 cells was not as acute as that on S1 cells. As shown in Chart4B, daily administrations of IFN for 3 days did not affect the R1cell growth, and an apparent difference in the mean R1 numberbetween control and IFN groups at Day 10 was still statisticallyinsignificant (p = 0.23). The number of R1 cells in IFN-treated

mice, however, did not increase after Day 6 and became significantly smaller than that in control mice at Day 14 (p < 0.01).

Microscopic Observation of IFN Effect. S1 and R1 cells fromthe peritoneal cavity of mice checked on Day 14 in the experimentof Chart 4 were cytocentrifuged on slides to smear, fixed,stained, and observed microscopically. Several mice carryingaverage numbers of tumor cells in each group were examined.S1 and R1 cells cultured in vitro with or without IFN were alsoobserved.

S1 cells in mice receiving IFN were observed to be swollenand vacuolated in comparison with those in control mice (Fig. 1,

(A)K*iI10"in

"ofio6-zIFN

(ID/day)6dt•

A•

*

SAAA0

5x10*.--Wd1*A

iA'

ÕÓ

5x10*.•-•

14d!t••

AAA

AAA0

5xtf

»«A.ir/6d•

^•1A•

*!A-•

,-rodt»

AT

tAAA•

A,

Wd•Ì;ÕAi

IFN 0 5x10* 0 5x)0* 0 5x10*

Chart 4. Time course of S1 (A) and R1 (B) cell growth in control and IFN-treatedmice. S1 or R1 cells (108) were inoculated i.p. on Day (d) 0, and recipient mice

received daily injections of 50,000 IU IFN (A) or 0.15 M NaCI (•)from Day 3 to Day5, Day 9, or Day 13. Viable 81 or R1 cells in the peritoneal cavity were enumeratedon Day 6, Day 10, or Day 14. Points, number of cells in each mouse; oars, mean.

CANCER RESEARCH VOL. 45 MARCH 1985

1322

on April 14, 2017. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ANTITUMOR EFFECT OF IFN IN VIVO

Chart 5. In vitro growth of S1 (XI)and R1 (8)cells recovered from mice receiving daily injections of 50,000 IU IFN (•,A, •)or 0.15 M NaCI(O, A, D). S1 and R1 cells recovered during theexperiment in Chart 4 were used. S1 and R1cells were cultured for 3 days (d) with 1,000 IUIFN/ml (A, A) or 10,000 IU IFN/ml (D, •)orwithout IFN (O, •).Cell populations from 3 to 5mice carrying S1 or R1 cells around the meannumber of Chart 4 were not pooled and culturedin triplicate each in each group of the experiment. Points, geometric means of triplicate cultures; bars, S.E.

Kd

Days

IO4

A and ß).This seemed to imply that S1 cells were injured in vivoby the direct effect of IFN as in the case of in vitro experiments(Fig. 1, C and D). In contrast with S1 cells, R1 cells per seappeared to be uninjured by the injection of IFN, and mitoticfigures were often found in similar frequencies between IFN-

treated and control groups (Fig. 1, E and F). This may reflectthat R1 cells are resistant to the direct action of IFN, as in thecase of in vitro culture (Fig. 1, G and H).

Growth and IFN Sensitivity of S1 and R1 Cells Recoveredfrom IFN-treated and Control Mice. Cells recovered from peritoneal cavities of IFN-treated or untreated S1 or R1 recipients at

Day 6,10, or 14 were incubated in Petri dishes for 2 hr to removeadherent PEC of hosts. Then, 2.5 x 10* nonadherent S1 or R1

cells, which were readily identifiable by their large size, werecultured in 16-mm dishes for 3 days in the presence or absence

of IFN, and the time course of cell growth was determined.S1 and R1 cells recovered from untreated mice were unaltered

from the original populations with respect to growth capacityand sensitivity to IFN. S1 cells from IFN-treated mice seemed to

be impaired, and the injury became more severe as the numberof IFN injections increased. This was clearly realized in theprogressive decrease of their in vitro growth rates in the absenceof IFN with the length of the in vivo experiment period (Chart5A). On the other hand, surviving R1 cells from IFN-treated mice

were healthy enough to proliferate in vitro and were still resistantto IFN irrespective of the time of cell recovery (Chart 5B).

As described above, S1 cells recovered from IFN-treated mice

seemed to become effete as the number of IFN injections increased, so that it was difficult to estimate the conservation ofthe original sensitivity to IFN during the course of experiment.Thus, we kept 3 mice left untreated after the last injection of IFNat Day 13. These mice became moribund 5 to 7 weeks after S1inoculation because of the growth of S1 cells surviving the IFNeffect. S1 cells from such moribund mice grew well in vitro in theabsence of IFN and were still sensitive to IFN; the growth kineticswas similar to that shown in Chart 1. These results indicatedthat the original nature of S1 cells was not altered by theexposure to IFN.

Cells Infiltrating the Peritoneal Cavity of R1 Recipients.Cells in peritoneal cavities of mice inoculated with 106 R1 cells

and treated with 50,000 IU IFN or 0.15 M NaCI from Day 3 wereharvested at Day 6,10, or 14, and the viable cell number of R1cells and that of host PEC were scored. The contents of PECwere determined in smeared, Giemsa-stained preparations, and

the number of each cell type was calculated.As shown in Chart 6, the number of PECwas smaller in control

Number of cells

IFN(IU)

50000

50000

0R1UPEC

50000ree

Chart 6. Number of R1 cells, number of PEC, and PEC contents of R1-bearingmice receiving daily injections of 50,000 IU IFN or 0.15 M NaCI. The experimentalschedule was the same as in the experiment of Chart 4B. M<t>,macrophages; PMN,polymorphonuclear leukocytes; L, lymphocytes; d, days. Columns, mean cell numbers of 3 to 5 mice; bars, S.E.

mice and slightly larger in IFN-treated mice than that of R1 cells

at both Day 10 and Day 14, although no significant differencewas seen at Day 6 between PEC and R1 numbers. At Day 14,the difference in R1 cell number between control and IFN-treated

groups was statistically significant (p < 0.01), similar to the resultin Chart 4. This seemed to parallel the eminent increase ofmacrophage infiltration in the IFN-treated group. Thus, the mac

rophage percentages in PEC at Day 14 were 42.4% in the controlgroup and 78.6% in the IFN-treated group (p < 0.01).

In Vitro Antitumor Activity of Crude and Purified Macrophage Populations from IFN-treated or Untreated R1-bearingMice. PEC were prepared from R1-bearing mice treated with

IFN or 0.15 M NaCI, according to the schedule of the aboveexperiment. The number of R1 cells and that of PEC wereessentially similar to those shown in Chart 6. Crude macrophagepopulations and purified macrophage populations were preparedfrom the PEC, and various numbers of cells of these populationswere cultured with 2.5 x 104 R1 cells in the presence or absence

of IFN (10,000 Ill/ml). The number of viable R1 cells was counted48 hr after the initiation of culture. Results are shown in Chart7.

Both macrophage populations of IFN-treated mice and thoseof untreated mice were effective in suppressing the/'n vitro tumor

CANCER RESEARCH VOL. 45 MARCH 1985

1323

on April 14, 2017. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ANTITUMOR EFFECT OF IFN IN VIVO

(A)

BO-

(B)

100-

01-50-

(C)

100-

£

1a»

£

O A 8 16Macrophage :R1ratio

0 L. 8 16Macrophage : RI ratio

0.25 05 1 2 A 8 16 32Macrophage :R1 ratio

Chart 7. In vitro antitumor activity of crude and purified macrophage populations prepared from IFN-treatedor untreated R1-bearing mice. Mice were inoculated i.p.with 10* R1 cells at Day 0 and receiveddaily injections of 50,000 IU IFN (A, A) or 0.15 MNaCI(O, •)from Day 3 to Day 5 (A),Day 9 (B),or Day 13 (C).Crude peritonealmacrophage populations were prepared at Day 6 (A),Day 10 (S),or Day 14 (C)from 5 to 10 mice and were pooled. Varying doses of the crude macrophagepopulationswere cultured with 2.5 x 104R1 cells in the presence (•,A) or absence (O, A) of 10,000 IU IFN/ml. The number of viable R1 cells was counted 48 hr after the initiationof culture. In C, a purified macrophage population prepared from IFN-treated, R1-bearing mice (V) and that from IFN-untreated R1-bearing mice (O) were also used.Abscissa, ratio of macrophage numbers in macrophage populations to 2.5 x 104R1 cells. Points, geometric means of triplicate cultures; bars, S.E. Points without bars,

range of S.E. is small and included in the vertical length of the point.

(A)

100-

50'

r

(B)

100-

c•»u

I

N,

\0248

Macrophage :RI ratio0248

Macrophage : R1 ratio

Chart 8. Effect of in wvo-activated macrophages in R1-bearing mice on the invitro growth of R1 cells and other tumor cells. Mice were inoculated i.p. with 10*R1 cells at Day 0 and received daily injections of 50,000 IU IFN (A, •,•.T) or0.15 M NaCI (A, O, G, V) from Day 3 to Day 13. Peritoneal macrophages wereprepared on Day 14 (A) and Day 10 (B). In A and 8, macrophages were culturedwith 2.5 x 10* YAC-1 cells (•,O) or L1210 cells (A, A) and with R1 cells (•,D) orDG75cells (T, V), respectively. For other details, see legend to Chart 7.

cell growth, irrespective of the time of PEC harvest. In theabsence of IFN in the culture, the suppressive effect was muchhigher in IFN-treated mouse macrophage populations than in

untreated mouse macrophage populations; the macrophage:R1cell ratio for 50% suppression of R1 cell growth was about 4 inthe former. Antitumor activity of macrophage populations wasremarkably enhanced in the presence of IFN in the culture, andno significant difference in the activity was apparent betweenmacrophage populations of IFN-treated mice and those of un

treated mice. As shown in Chart 7C, the macrophage:R1 cellratio for 50% suppression was as low as 0.5.

In the experiments of Chart 7, A and B, we prepared onlycrude macrophage populations, but Chart 7C shows that theeffect of purified macrophage populations was almost comparable with that of the original crude preparations. There was asimilar case of another purified macrophage population (morethan 99%, nonspecific esterase-positive macrophages which

was prepared in a preliminary experiment by culturing the crudemacrophage populations for 24 hr in order to remove neutrophilsby their death during the culture period) (data not shown).Moreover, we also examined the effect of nonadherent cell

populations, separated from adherent cells in the experiment ofChart 7, on the in vitro growth of R1 cells and observed nosignificant effect on R1 cell growth (data not shown).

Effect on Allogeneic and Human Tumor Cells. These experiments were conducted to investigate whether the antitumoractivity of macrophage populations from R1-bearing mice was

restricted to R1 cells or not. Thus, the effect of macrophagepopulations prepared as in the preceding experiment on the invitro growth of allogeneic tumor cells, YAC-1 and L1210 (ChartSA), and of human DG75 cells (Chart 85) was assessed. Macrophage populations of IFN-treated mice were effective in sup

pressing the growth of any tumor cells as well as R1 cells. Onthe other hand, macrophage populations of IFN-untreated mice

were virtually ineffective for these allogeneic and xenogeneiccells, although the growth of R1 cells was moderately suppressed, similar to the result in Chart 7.

DISCUSSION

In this study, we isolated 81 and R1 cell sublines differingfrom each other in the sensitivity to IFN under in vitro culturecondition (Charts 1 and 2). This indicates that the Meth A cellpopulation which has been maintained in our laboratory by invivo passages consists of a mixture of cells heterogeneous withrespect to the sensitivity to IFN. The spectrum of heterogeneityseems to be continuous, since we observed various degrees ofIFN sensitivity among many clones in the process of selectionfor S1 and R1 cells (data not shown).

IFN preparations used in the present experiment were partiallypurified ones having relatively low titers on the order of 106 IU/

mg protein. One might not exclude, therefore, a possibility thatit was not the IFN but some other substances which suppressed81 cell proliferation and to which R1 cells are resistant. Thisseems unlikely, since a high-liter anti-IFN perfectly neutralized

the inhibitory activity of our IFN preparation for 81 cell growthand since, if not a definite evidence, heat-inactivated IFN prepa

ration did not affect 81 cells (Chart 18). In addition, we haveconfirmed that 81 cells are sensitive and R1 cells resistant alsoto rIFN-aA/D (a recombinant human leukocyte IFN, provided by

Nippon Roche Research Center, Kanagawa, Japan) which iseffective for both human and mouse cells (22) (our data will bepublished elsewhere).

CANCER RESEARCH VOL. 45 MARCH 1985

1324

on April 14, 2017. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ANTITUMOR EFFECT OF IFN IN VIVO

The susceptibility of S1 cells and the resistance of R1 cells invitro to the direct action of IFN seem to be unaltered under thein vivo environmental conditions, since S1 and R1 cells includedin cell-impermeable diffusion chambers in the peritoneal cavitywere still sensitive and resistant, respectively, to i.p.-injected IFN

(Chart 3). Thus, the improvement of survival rate in S1 recipientsafter the administration of IFN is understandable as a naturalconsequence. In spite of the IFN resistance of R1 cells, themortality of R1 cell recipients was reduced paradoxically by dailyadministrations of IFN (Table 1), and the effect of IFN on in vivoR1 cell growth was not as acute as that on S1 cells as observedin the experiment of Chart 4. This encouraged us to investigatethe indirect effect of IFN through a host defense mechanism onthe IFN-resistant R1 cell.

Microscopic observation of S1 and R1 cells recovered fromthe peritoneal cavity of mice receiving IFN revealed that themajority of S1 cells, but not R1 cells, appeared to be injured,with swelling and vacuolation (Fig. 1). Such an appearance of 81cells resembled that of those cultured in vitro with IFN, suggesting that S1 cells were impaired in vivo by the direct action ofIFN. On the other hand, R1 cells surviving IFN administrationsappeared to be uninjured like those cultured in vitro with IFN.This and the suppression of in vivo R1 cell growth by IFN (Chart4) seem to become coincident with each other, if we assumethat R1 cells are resistant to the direct effect of IFN but sensitiveto the indirect effect of IFN mediated by some host cellularmechanism, which may act very locally by the contact of hostcells with R1 cells.

The wasting of S1 cells on the whole by the direct action ofIFN and the selective damage of R1 cells by the indirect effectof IFN are clearly shown in Chart 5. In that experiment, S1 andR1 cells were recovered from control and IFN-treated mice on

different days, and a fixed number of viable cells excluding trypanblue dye were cultured with or without IFN for further 3 days.S1 cells recovered after 3 daily administrations of IFN were stillable to proliferate as well as those recovered from control mice,but the vitality of S1 cells recovered from IFN-treated mice

thereafter tended to be reduced with the multiplication of IFNadministration. On the contrary, viable R1 cells from IFN-treated

mice always proliferated well irrespective of the time of cellrecovery. Chart 5 also indicates that the sensitivity of S1 cellsand the resistance of R1 cells to the direct action of IFN wereunaltered after in vivo culture in the presence or absence of IFN.This was conspicuously shown in the cases of R1 cells and ofS1 cells from control mice but was not as clear in the case ofS1 cells from IFN-treated mice. The reason for the latter seems

to be ascribable to the general wasting of S1 cells caused bythe direct effect of IFN. As described before, we recovered S1cells after keeping IFN-treated S1 recipients for 3 to 5 weeks

afterwards without IFN administration and confirmed that theywere still sensitive to IFN. This strongly suggests that the sensitivity of S1 cells can be conserved even in the presence of IFNunder in vivo condition.

Chart 6 shows that the number of host cells infiltrating theperitoneal cavity of R1 cell recipients was slightly but significantlylarger in IFN-treated mice than in IFN-untreated mice and that a

significant difference in the R1 cell number between these 2experimental groups was seen in the late period of experiment,in parallel with the increase of the recruitment of macrophagesin IFN-treated mice. This seems to imply that the indirect effect

of IFN on R1 cells may be mediated by macrophages. Theparticipation of macrophages in the indirect effect may also besuggested by the experiment of Chart 7. In this experiment, itwas found that the macrophage populations from IFN-treatedR1 recipients were much more effective than those from IFN-

untreated ones in suppressing the in vitro growth of R1 cells.The suppressive effect was also seen in purified macrophagepopulations used in an experiment (Chart 7C).

The suppressive activity of macrophage populations preparedfrom mice receiving R1 cells and IFN was not restricted to R1cells. Thus, macrophage populations were suppressive for allo-

genic and also for human tumor cells (Chart B). Nevertheless,macrophage populations from IFN-untreated R1-bearing mice

appeared to be effective only for R1 cells. Although the reasonfor this is still obscure, it is surmisable that the allogeneic andhuman cells that we used may be relatively insensitive to theweakly activated effector cells in macrophage populations incomparison with R1 cells.

It is well known that representative cells involved in the hostdefense mechanism against tumor cells are macrophages, T-

lymphocytes, and NK. Among them, macrophages seem to playthe major role in our experimental system, since, as mentionedabove, the suppressive activity for tumor cell growth resided inthe macrophage populations of PEC and also in the purifiedmacrophage population. In addition, the suppressive effect ofmacrophage populations from R1-bearing mice was not restricted to R1 cells; this seems to argue the implausibility of T-

lymphocytes as direct effectors in macrophage populations. Thepossibility of NKs as direct effectors seems also implausible,because macrophage populations from IFN-treated, R1-bearingmice suppressed the growth of not only YAC-1 cells, which are

NK sensitive, but also L1210. Meth A (R1), and DG75 cells,which are NK resistant (18, 24) (Chart 8), and the suppressiveactivity of a purified macrophage population containing less than1% non-macrophage cells was not lower than that of the original

macrophage population, both the populations being effectual for50% suppression of in vitro R1 cell growth in an effector: R1 cellratio as low as 4 (Chart 1C). Assuming that the non-macrophage

cells in purified macrophage population were direct effectors,they would have had to affect the R1 cell growth in an effector: R1cell ratio of 0.04 or lower. This seems not to have occurred,since even pure NK line cells were reported to require theeffectontarget ratio of 1 or more (29).

Definite information is still scanty on the antitumor activity ofmacrophage activated in vivo by the administration of IFN. Chi-rigos ef al. (6) reported that both IFN and pyran copolymer, anIFN inducer, were effective in suppressing the in vivo growth ofMBL-2 leukemia cells and in reducing the mortality of mice andthat macrophages obtained from pyran-treated mice, but notthose from IFN-treated mice, damaged MBL-2 cells in the in vitro

culture. Gresser and Bourali (10) suggested a possibility ofmacrophage activation by IFN based on the observation thatEhrlich ascites tumor cells in the peritoneal cavity were phago-cytosed by macrophages of IFN-treated mice. On the other hand,they also reported that an i.p. inoculation of silica as a macro-phage-blocking agent did not alter the effect of IFN on the invivo growth of IFN-resistant L1210 or Friend leukemia cells (3,

13). It seems problematical, however, whether a single injectionof silica on the day of tumor inoculation, as Gresser ef a/, did,was effective in completely abrogating the function and the

CANCER RESEARCH VOL. 45 MARCH 1985

1325

on April 14, 2017. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ANTITUMOR EFFECT OF IFN IN VIVO

recruitment of macrophages thereafter, and we believe that theirresults do not give a convincing evidence against the positivecontribution of macrophages to the antitumor activity in IFN-

treated animals. Indeed, we conducted an experiment in whichDBA/2 mice were inoculated i.p. with IFN-resistant L1210 cells

and received daily injections of IFN, and we observed thatmacrophage recruitment was enhanced by IFN administrationsand that such macrophage populations exerted a suppressiveeffect on the in vitro growth of L1210 cells and also of othertumor cells (data not shown).

IFN seems to activate macrophages under in vitro conditionsto suppress the tumor cell growth, as described by severalworkers (4, 5, 26, 27). This may be connected to some resultsshown in Chart 7. Thus, not only macrophage populations fromIFN-treated R1-bearing mice but also those from untreated ones

manifested very high antitumor activity in the presence of IFN inthe in vitro test for the effect of macrophages on R1 cell growth.We have investigated the role of IFN and also that of tumor cellsin the activation of macrophages in vitro; results will be reportedin our succeeding papers.

REFERENCES

1. Balkwill, F. R., and Oliver, R. T. Growth inhibitory effects of interferon onnormal and malignant human haemopoieticcells. Int. J. Cancer,20: 500-505,1977.

2. Belardelli,F., Grosser, I., Maury, C., and Maunoury, M. T. Antitumor effects ofinterferon in mice injected with interferon-sensitive and interferon-resistantFriend leukemiacells. I. Int. J. Cancer,30: 813-820, 1982.

3. Belardelli,F., Grosser, I., Maury, C., and Maunoury, M. T. Antitumor effects ofinterferon in mice injected with interteron-sensitive and interferon-resistantFriend leukemia cells. II. Role of host mechanisms. Int. J. Cancer, 30: 821-825,1982.

4. Boraschi, D., Soldateschi, D., and Tagliabile. A. Macrophage activation byinterferon: dissociation between tumoricidal capacity and suppressiveactivity.Eur. J. Immunol., 12: 320-326,1982.

5. Boraschi, D., and Tagliabue, A. Interferon-induced enhancement of macro-phage-mediatedtumor cytolysis and its difference from activation by lympho-lines. Eur. J. Immunol., 11: 110-114, 1981.

6. Chingos, M. A., Schultz, R. M., and Stylos, W. A. Interaction of interferon,macrophage and lymphocyte tumoricidal activity with prostaglandin effect.Ann. NY Acad. Sci., 350: 91-101, 1980.

7. Dean, R. T., and Virelizier,J. L. Interferon as a macrophage activating factor.I. Enhancementof cytotoxicity by fresh and matured human monocytes in theabsence of other soluble signals. Clin. Exp. Immunol.,51: 501-510, 1983.

8. Gazdar, A. F., Sims, H., Spahn, G. J., and Baron, S. Interferon mediatesenhancementof tumor growth and virus-inducedsarcomas in mice. Nat. NewBiol., 245: 77-78, 1973.

9. Gidlund, M., Om, A., Wigzell, H., Senik, A., and Gresser, I. Enhanced NK cellactivity in mice injected with interferon and interferon inducers. Nature (Lond.),273:759-761, 1978.

10. Gresser, I., and Bourali, C. Antitumor effects of interferon preparations in mice.J. Nati. Cancer Inst., 45: 365-376, 1970.

11. Gresser, I., Bourali, C., Chouroulinkov, I., Fontaine-Brouty-Boyé,D., and

Thomas, M. T. Treatment of neoplasia in mice with interferon preparations.Ann. NY Acad. Sci., 773: 694-707,1970.

12. Gresser, I., Bourali, C., Levy, J. P., Fontaine-Brouty-Boyé,D., and Thomas,M. T. Increasedsurvival in mice inoculated with tumor cells and treated withinterferon preparations. Proc. Nati. Acad. Sci. USA,63: 51-57,1969.

13. Gresser, I., and Bourali-Maury,C. Antitumor effect of interferon in lymphocyte-and macrophage-depressedmice. Proc. Soc. Exp. Biol. Med., 744: 896-900,1973.

14. Gresser, I., Brouty-Boyé,D., Thomas, M. T., and Macieira-Coelho,A. Interferonand cell division. I. Inhibition of the multiplication of mouse leukemia L1210cells in vitro by interferon preparations. Proc. Nati. Acad. Sci. USA,66:1052-1058, 1970.

15. Gresser, I., De Maeyer-Guignard,J., Tovey, M. G., and De Maeyer, E. Etectro-phoreticallypure mouse interferon exerts multiple biologic effects. Proc. Nati.Acad. Sci. USA, 76: 5308-5312,1979.

16. Gresser, I., Maury, C., and Brouty-Boyè,D. Mechanisms of the antitumoreffect of interferon in mice. Nature (Lond.),239:167-168,1972.

17. Groves, D. L., Lever, W. E., and Makinodan,T. A model for the interaction ofcell types in the generation of hemolytic plaque forming cells. J. Immunol.,704:148-165,1970.

18. Haller,0., Kiessling,R., Ören,A., Karre,K., Nilsson,K., and Wigzell,H. Naturalcytotoxicity to human leukemiamediatedby mouse non-Tcells. Int. J. Cancer,20: 93-103,1977.

19. Kumagai, K., Itoh, K., Hinuma, S., and Tada, M. Pretreatment of plastic Petridishes with fetal calf serum. A simple method for macrophage isolation. J.Immunol.Methods, 29: 17-25,1979.

20. Minato, N„Reid, L., Cantor, H., Lengyel, P., and Bloom, B. R. Mode ofregulation of natural killer activity by interferon. J. Exp. Med., T52: 124-137,1980.

21. Oehler, J. R., Lindsay, L. R., Nunn, M. E., Holden, H. T., and Herberman, R.B. Natural cell-mediatedcytotoxicity in rats. II. in vivo augmentationof NK cellactivity. Int. J. Cancer,21: 210-220,1978.

22. Pestka, S., and Maeda, S. The human interferons: their purification andsequence,cloning and expression in bacteria, and biological properties. In: Y.Yamamuraet al. (eds.), Humoral Factors in Host Defense, pp. 191-243. NewYork: Academic Press, Inc., 1982.

23. Reid, L. M., Minato, N., Gresser, I., Holland,J., Kadish, A., and Bloom, B. R.Influence of anti-mouse ¡nterferonserum on the growth and metastasis oftumor cells persistently infected with virus and human prostatic tumors inathymic nude mice. Proc. Nati. Acad. Sci. USA, 78: 1171-1175,1981.

24. Roder, J. C., Lohmann-Matthes,M.-L, Domzig, W., Kiessling,R., and Haller,O. A functional comparisonof tumor cell killing by activated macrophagesandnatural killer cells. Eur. J. Immunol.,9: 283-288,1979.

25. Ryd, W., Hagmar,B., Lundgren,E., and Strannegard,0. Discrepanteffects ofinterferon on murine syngeneic ascites tumors and their solid metastasizingcounterparts. Int. J. Cancer, 23: 397-401,1979.

26. Schultz, R. M., Chirigos, M. A., and Heine, U. I. Functional and morphologiccharacteristics of interferon-treated macrophages.Cell. Immunol.,35: 84-91,1978.

27. Schultz, R. M., Papamatheakis,J. D., and Chirigos,M. A. Interferon:an inducerof macrophageactivation by polyanions.Science(Wash. DC). 797: 674-676,1977.

28. Stewart, W. E., Gresser, I., Tovey, M. G., Bandu, M. T., and Goff, S. LIdentification of the cell multiplication inhibitory factors in interferon preparations as interferons. Nature (Lond.),262: 300-302,1976.

29. Suzuki, R., Handa, K., Itoh, K., and Kumagai, K. Natural killer (NK) cells as aresponder to interteukin2 (IL 2). I. Proliferativeresponseand establishmentofcloned cells. J. Immunol., 730: 981-987,1983.

30. Yoshida, R., Imanishi,J., Oku, T., Kishida, T., and Hayaishi, O. Induction ofpulmonary indoteamine2,3-dtoxygenase by interferon. Proc. Nati. Acad. Sci.USA, 78:129-132,1981.

CANCER RESEARCH VOL. 45 MARCH 1985

1326

on April 14, 2017. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Fig. 1. In vivo and in vitro effects of IFN on the morphology of S1 and R1 cells, x 600. Mice were inoculated ¡.p.with 10* S1 (A, 8) or R1 (£,F) cells on Day 0 andreceived daily injections of 50,000 ID IFN (B, F) or 0.15 M NaCI(A, E) from Day 3 to Day 13. Cells in peritonealcavities were harvested on Day 14. In vitro culture of S1cells (C, D) and R1 cells (G, H) was performed for 3 days in the presence (D,H) or absence (C, G) of 10,000 IU IFN/ml. Cells were fixed and stainedwith May-Grunwald-Giemsa stain. Cells smaller than S1 or R1 cells in A, B, E, and F are macrophages, neutrophils, or erythrocytes. The absolute number of macrophagesshould not becompared between B and F, since much smaller numbers of S1 cells were recovered than those for R1 cells at Day 14.

CANCER RESEARCH VOL. 45 MARCH 1985

1327

on April 14, 2017. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1985;45:1320-1327. Cancer Res   Kazuko Uno, Satoshi Shimizu, Motoharu Ido, et al.   Cell Growth

Murine Tumorin VivoDirect and Indirect Effects of Interferon on

  Updated version

  http://cancerres.aacrjournals.org/content/45/3/1320

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

  .pubs@aacr.orgDepartment at

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

  Permissions

  .permissions@aacr.orgDepartment at

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

on April 14, 2017. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from