Augmentation of Murine Natural Killer Cell Activity by ... · biological response modifiers such as interferon and interleukin 2 is lower than that which exhibits maximum biological

(CANCER RESEARCH 48, 1410-1415. March 15, 1988)

Augmentation of Murine Natural Killer Cell Activity by Swainsonine, a NewAntimetastatic Immunomodulator1

Martin J. Humphries,2 Kazue Matsumoto, Sandra L. White, Russell J. Molyneux, and Kenneth Olden

Departments of Oncology [M. J. //., K. M., S. L. W., K. O.J and Microbiology fS. L. W.], Howard University Cancer Center, Washington, DC 20060, and WesternRegional Research Center, Agricultural Research Service, USDA, Albany, California 94710 [R. J. MJ

ABSTRACT

Swainsonine, an indolizidine alkaloid, has been found to inhibit theexperimental metastasis of B16-F10 melanoma cells when administeredsystemic-ally to syngeneic C57BL/6 mice. The inhibition was both potent

and dose dependent with Â£80%reduction in pulmonary colonizationbeing observed after only 24-h exposure to 3 Â«ig/mlof swainsonine indrinking water. In contrast, the inhibitory activity of swainsonine wascompletely abrogated when assays were performed in mice depleted oftheir natural killer (NK) cell activity either experimentally (anti-asialo-GMi antibody- or cyclophosphamide-treated C57BL/6 mice) or as aresult of genetic mutation (homozygous C57BL/64*'**beige mice). Swain

sonine elicited a 32.0% increase in spleen cell number 2 days afteradministration and induced a concomitant 2- to 3-fold increase in splenicNK cell activity. These results indicate (a) an absolute requirement for afunctional NK cell population in order for swainsonine to exert itsinhibitory effects on experimental metastasis, and (b) that the antimeta-static activity of swainsonine is mediated primarily through the ability ofthe drug to augment NK cell reactivity. On the basis of these findings,swainsonine can be classified as a new immunomodulator that has theability, at least in a prophylactic setting, to block tumor metastasis.

INTRODUCTION

The development of therapeutic modalities which lead toeither prevention or elimination of tumor mÃ©tastasesis currently one of the principal challenges facing oncologists. Onepromising approach is that of immunotherapy. The administration of various naturally occurring or synthetic biological response modifiers has been reported to enhance host immunereactivity and, in combination with systemic transfer of immunecell populations, has been shown to elicit antitumor activity (1-6). In patients, however, the maximum tolerated dosage ofbiological response modifiers such as interferon and interleukin2 is lower than that which exhibits maximum biological activity(7-9). Consequently, the full therapeutic potential of theseagents has yet to be realized. Development of strategies toovercome the toxicity of biological response modifiers or, alternatively, identification and characterization of new agents thatdo not exhibit such side effects would greatly enhance theprospects for immunotherapy of human patients.

In previous studies from this and other laboratories, it hasbeen reported that pretreatment of highly metastatic B16-1 10murine melanoma cells with inhibitors of protein glycosylationor processing [the nucleoside analogue tunicamycin (10, 11)and the plant alkaloids swainsonine (12, 13) and castanosper-mine (14, IS)] causes a substantial impairment of their pulmonary colonization potential following i.v. injection into mice(16-20). Furthermore, the inhibition arising from pretreatmentof cells with one of these agents (swainsonine) has been foundto be enhanced by simultaneous systemic administration of free

Received 8/12/87; revised 12/9/87; accepted 12/11/87.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1This work was supported in part by NIH Grants CA-14718 and CA-45290,and ACS Grant PDT-312 to K. O., and the Howard University Faculty ResearchSupport Grant.'' Supported by a Cancer Research Institute fellowship.

drug to mice (21). In this report we have further investigatedthe antimetastatic activity of swainsonine and now report that,even when administered alone without pretreatment of tumorcells, the drug possesses potent inhibitory activity in experimental metastasis assays. Furthermore, by testing the effect ofswainsonine on metastasis of B16-F10 cells in NK3-deficient

mice, we present evidence that a principal mode of action ofthe drug is to enhance NK cell function in vivo.

MATERIALS AND METHODS

Materials. The swainsonine used in these studies was isolated fromdiablo locoweed (Astragalus oxyphysus) using the procedure describedpreviously (22) or was purchased from Calbiochem (San Diego, CA).Similar results to those presented here on immunomodulation in micewere obtained using chemically synthesized swainsonine, thereby discounting a role for endotoxin in the biological effects of the drug.Furthermore, the swainsonine isolated from plant sources was foundto contain undetectable levels of endotoxin based on the LimulusAmebocyte Lysate assay.

Pulmonary Colonization. B16-F10 murine melanoma cells (from Dr.I. J. Fidler, M. D. Anderson Hospital, University of Texas, Houston,TX) were cultured and tested for their pulmonary colonization capacityin C57BL/6 mice (Charles River Breeding Laboratories, Wilmington,MA) exactly as described previously (19, 23). Cells were found to befree of both Mycoplasma and mouse antibody production activity byroutine testing. The NK-deficient mice used in these studies were: (a)C57BL/6 mice given injections i.v. of 0.2 ml of anti-asialo-GMi antibody (Wako Fine Chemicals, Dallas, TX; 1:50 dilution of IgG withphosphate-buffered saline) 2 days before tumor cell injection; (b)C57BL/6 mice given injections i.p. of 0.2 ml of 25 mg/ml of cyclo-phosphamide (Sigma, St. Louis, MO) in Hanks' balanced salt solution4 days before tumor cell injection; and (c) homozygous C57BL/6'*/**

(beige) mice (Jackson Laboratory, Bar Harbor, ME).Pulmonary Retention. The arrest and subsequent clearance of

[125I]-iododeoxyuridine-labeled B16-F10 cells from the lungs of C57BL/

6 mice were examined exactly as described previously (19, 23).Cytotoxicity Assay. Spleen cells were prepared from either control,

NK-deficient, or swainsonine-treated mice and tested in a standard 4-hcytotoxicity assay using "Cr-labeled B16-F10 cells or YAC-1 cells as

the target (24).

RESULTS AND DISCUSSION

Antimetastatic Effect of Systemic Swainsonine Administration. As an initial step towards examining the effects of systemicadministration of swainsonine on experimental metastasis ofB16-1"10 cells, the drug was sol uhi 1Â¡zed,diluted into drinking

water, and provided to mice ad libitum for varying periods oftime immediately prior to tumor cell injection. As shown inFig. 1, swainsonine was highly inhibitory; maximal activity ofthe drug was observed when mice were on swainsonine-supple-mented drinking water for only 24 to 48 h. In other experiments, a similar degree of inhibition was found when swainsonine was present not only for 24 h prior to injection of tumorcells but also for the entire postinjection incubation period(data not shown). Half-maximal inhibition of colonization was

3The abbreviation used is: NK, natural killer.

1410

Research. on August 12, 2020. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

ENHANCEMENT OF NK CELL ACTIVITY BY SWAINSONINE

ce

iz>â€¢o

6 24 48 72

TIME OF SW TREATMENT (h)Fig. 1. Effect of swainsonine on experimental metastasis of B16-F10 cells.

Swainsonine (-VIÃ•) was diluted to 3 Mg/ml in water and supplied ad libitum forthe indicated times to groups of 8 C57BL/6 mice. Aliquots of 9 x 10* B16-F10

cells were then injected into the lateral tail vein, and animals were returned tounsupplemented drinking water. Two wk later, surface melanotic colonies werecounted at necropsy. Bars, SE.

300 -

250

SDO

8 100

50

0 L1 2

SWAINSONINE (ng ml)

Fig. 2. Dose dependence of swainsonine-mediated inhibition of B16-F10pulmonary colonization. Swainsonine, diluted into drinking water at the indicatedconcentrations, was supplied to mice for 24 h prior to injection with 9 x l O4B16-

F10 cells. Bars, SE. The inhibition obtained by treatment with 3 Â«ig/mlofswainsonine was significant at P < 0.001 using the Mann-Whitney U test.

observed after only 2 to 6 h of drug treatment (Fig. 1). Thus,only a brief exposure to swainsonine is sufficient for it to exhibitantimetastatic activity. This finding, coupled with preliminarydata suggesting that ingested swainsonine may be cleared rapidly from the body (25), implies that the drug may initiate abiological cascade which, in turn, leads to an increase in anti-tumor cell reactivity.

For all future experiments, a single 24-h administration ofswainsonine-supplemented drinking water was used immediately preceding tumor cell injection. Under these conditions,swainsonine-mediated inhibition of pulmonary colonizationwas dose dependent; half-maximal inhibition was observed at a

100

50

-? 20

V)O 103

0.5

0246 24

TIME AFTER INJECTION (H)

15

u> 10o

6 8 24

TIME OF SW TREATMENT (h)

Fig. 3. Effect of swainsonine on pulmonary retention of B16-F10 cells. In . I.mice were provided with unsupplemented drinking water ( . â€¢)or 3 Mg/ml ofswainsonine in drinking water l .s II. O) for 24 h prior to injection of 8 x IO4|'"I]-iododeoxyuridine-labeled tumor cells. At the indicated times, mice were

sacrificed by cervical dislocation, and their lungs were excised, exsanguinated,and counted for radioactivity. In B, drinking water containing 3 pg/ml of swainsonine was administered to mice for various times before injection of 5 x IO4radiolabeled tumor cells. Seventeen h later, pulmonary radioactivity was determined as above. Bars, SE.

swainsonine concentration of 100 ng/ml, and approximately80% inhibition was obtained at 3 /ug/ml (Fig. 2). In otherexperiments using lower tumor cell inocula, as much as 96%inhibition of colonization was obtained. A similar dose-response curve was found for longer periods of drug administration to mice, demonstrating that prolonged exposure at lowconcentration had no further beneficial effect (data not shown).For the experiments reported in this study, at least sevendifferent batches of swainsonine from four independent sourceshave been tested with virtually identical results.

The effect of swainsonine on the time course of pulmonaryretention of B16-F10 cells is shown in Fig. 3a. Cells wereradiolabeled in culture with ['"Ijiododeoxyuridine and then

injected i.v. as for experimental metastasis assays. At various1411




times, mice were sacrificed, and the number of cells remainingin the target organ was determined by 7-counting. For controlmice on unsupplemented drinking water, there was almostquantitative recovery of injected cells in the lungs 2 min afterinjection. After a characteristic negative exponential loss ofcells over the immediate postinjection period, 2.0% of injectedcells remained 24 h later (Fig. 3a). For mice receiving swain-sonine in their drinking water, the initial level of retention wassimilar to that of control mice (>95%). Furthermore, from 0to 6 h postinjection, the profile of cell loss was also similar(e.g., at 6 h, 9.3% of control cells remained in the lungscompared to 6.3% in mice treated with swainsonine; Fig. 3a).By 24 h postinjection, however, the percentage of cells retainedin the lungs of swainsonine-treated mice was only 0.46% compared to 2.0% for mice on unsupplemented drinking water (Fig.3a). Thus, swainsonine appears to elicit accelerated removal oftumor cells that is manifested primarily >24 h postinjection.

As expected from the data presented in Fig. 1, the swainson-ine-induced decrease in pulmonary retention of tumor cells wasalso dependent on the time of exposure of mice to drug-supplemented drinking water (Fig. 30). In this experiment, after24-1] administration of swainsonine, 2.6% of radiolabeled B16-F10 cells were retained in the lungs 17 h after injection compared to 14.1% for the same cells injected into mice receivingunsupplemented drinking water; longer periods of administration had no greater effect (data not shown). An approximatehalf-maximal effect was obtained by administration of swainsonine for 2 h, a result consistent with the data presented inFig. 1. Thus, although swainsonine prevents pulmonary colonization after only a brief exposure to mice, the inhibition isnot mediated primarily through early effects on injected tumorcells (0 to 10 h postinjection). Instead, swainsonine appears toexert most of its biological activity at least 24 h after administration to mice.

Considering the time course over which swainsonine acts,one potential mechanism of action of the drug could be stimulation of immune system function. Interestingly, there is nowsubstantial experimental evidence implicating carbohydrate residues in immune cell recognition processes (26, 27). In particular, the relationship of cell surface oligosaccharides to NK cellfunction has been investigated in some detail (28-31). In arelated approach to our own, depression of murine lymphocyteproliferation caused by exposure to an endogenous Â¡nummosuppressive factor has been reported to be restored by swainsonine treatment (32). Furthermore, the time course over whichswainsonine inhibits experimental metastasis |>24 h after addition, with a peak of activity 2 to 3 days after administration(see Fig. 5 below)] parallels previous reports of the kineticprofile of NK cell activation (33-36). It should also be noted,however, that an equally plausible explanation for the antimet-astatic activity of swainsonine is that the drug causes structuralalterations in the tumor cell surface oligosaccharides in vivowhich in turn lead to an increase in their susceptibility to NKcell lysis.

We have used two approaches to directly test the hypothesisthat swainsonine-mediated inhibition of B16-F10 experimentalmetastasis might be mediated at least partly through its abilityto enhance NK cell reactivity, (a) The pulmonary colonizationpotential of B16-F10 cells was measured in the presence orabsence of swainsonine in mice with compromised NK cellfunction, (b) The effect of systemic swainsonine administrationon NK cell activity was examined in a standard 4-h cytotoxicity

assay.Antimetastatic Effect of Swainsonine in NK-deficient Mice.

For the first approach, C57BL/6 mice treated with either anti-asialo-GM, antibody or cyclophosphamide and C57BL/6 micehomozygous for the beige mutation were used as models forNK deficiency. Asialo-GMi is a marker for murine NK cells(37, 38), and anti-asialo-GMi antibody treatment currentlyrepresents one of the most specific methods available for removal of NK cells from a mixed population, since immature T-cells are the only other cell type known to express asialo-GM|,and this is at a much lower level than NK cells (37-42).Cyclophosphamide treatment has been reported to reduce NKcell activity (43, 44), but it is also known to suppress otherimmune cell populations (45-47). Beige mice possess a mutation in an autosomal recessive gene that results in a relativelyselective decrease in their NK cell activity (48-51). This defectappears to be linked to an inability of the NK cells in beigemice to secrete cytotoxic factors required for target cell lysis(48). To demonstrate that each of these types of mice wereindeed defective in NK function, we measured their endogenouslevels of NK cell activity in a 4-h cytotoxicity assay against51Cr-labeled YAC-1 cells. Consistent with previously publisheddata (37-39, 48-55), both anti-asialo-GMi antibody and cyclophosphamide treatment almost completely abolished NK activity, and the cytolytic activity of spleen cells from beige micewas approximately 30% that of spleen cells from controlC57BL/6 mice (data not shown).

Fig. 4 shows data combined from three sets of independentexperiments comparing the metastatic activity of B16-F10 cellsin NK-deficient and normal mice receiving either unsupplemented or swainsonine-supplemented drinking water. In allexperiments, swainsonine was administered at 3 /Â¿g/mlfor 24h. For anti-asialo-GMi antibody-treated mice on control drinking water, the number of melanotic colonies in the lungs 14

350300250tc111

g200z

>â€¢Â§150O

U100

M

nAvs/A rTJn.1PB-TÃ•mm)PICj.A,-

SW-+-+ -+-+ -+- +

Fig. 4. Effect of swainsonine on experimental metastasis of Hid 1 1(1cells inNK-deficient mice. The NK-deficient animals used were: A, anti-asialo-GM!antibody-treated C57BL/6 mice; B, cyclophosphamide-treated mice; and C,homozygous beige mice (C57BL/6**"*) (see "Materials and Methods"). Each

panel represents an independent experiment, but in each case untreated C57BL/6 mice were used as the control. All mice received either unsupplemented drinkingwater SW~) or drinking water containing 3 Â»ig/mlof swainsonine (SW*) for 24h prior to injection of 8 x 10' (anti-asialo-GMÂ¡-treated mice) or 9 x IO4 B16-F10 cells (cyclophosphamide-treated and beige mice). The left two columns ineach panel represent control mice, and the right two columns, NK-deficient mice.The number of pulmonary colonies was determined after 2 wk. Bars, SE. Theinhibition obtained with swainsonine treatment in control mice was significant atP < 0.001 in all three panels using the Mann-Whitney I test. No significantinhibition was observed for anti-asialo-GMi- or cyclophosphamide-treated mice.P = 0.08 for swainsonine-treated beige mice.

1412




days after tumor cell injection was enhanced approximately 4-fold over control mice receiving the same tumor cell inoculum(Fig. 4Â«).Injection of normal rabbit serum as a control had noeffect on colony number (data not shown). In untreated micereceiving an inoculum of 8 x IO4 B16-F10 cells, swainsonine

treatment inhibited the formation of mÃ©tastasesby 87.8%. Inanti-asialo-GM, antibody-treated mice, the drug treatment wascompletely ineffective (0.87% stimulation of colonization; Fig.4<z).The possibility that the decreased antimetastatic activity ofswainsonine might be an artifact of the increased colony numberin antibody-treated animals was tested by examining the effectof swainsonine on pulmonary colonization obtained with lowernumbers of injected tumor cells. These inocula were designedto yield similar numbers of colonies to that produced by 8 xIO4cells in non-antibody-treated mice. From the data presented

in Table 1, a reduction in the size of the tumor cell inoculumfrom 8 x 10" to 4 x 10" or 2 x 10" had only a small effect on

the degree of inhibition of colonization obtained with swainsonine in anti-asialo-GM i antibody-treated mice. For example,87.8% inhibition was obtained in non-antibody-treated micereceiving 8 x IO4 cells (40.0 colonies in the control), but only4.1% inhibition was observed when 4 x IO4cells were injected

into antibody-treated mice (65.1 colonies in the control). Thus,for control mice bearing an approximately equal number ofcolonies, the inhibitory activity of swainsonine was markedlydifferent between the two sets of animals. This finding demonstrates that the lack of activity of swainsonine in anti-asialo-GM i antibody-treated mice reported in Fig. 4a was not due tothe high colony number in the control but rather to a specificeffect of the antibody.

Very similar results to those observed with anti-asialo-GM,antibody were obtained by treatment of mice with cyclophos-phamide (Fig. 4b). Administration of cyclophosphamide i.p.elicited a 5-fold enhancement of B16-F10 colony formation,but at the same time it completely abrogated the antimetastaticactivity of swainsonine (70.7% inhibition in untreated micecompared to 10.8% stimulation in cyclophosphamide-treatedanimals; Fig. 4/>).

When mice homozygous for the beige mutation were usedfor metastasis studies, the inhibition obtained by swainsoninetreatment was reduced approximately 3-fold (65.8% inhibitionin untreated animals compared to 21.1% in beige mice; Fig.4c). This significant yet incomplete effect of swainsonine isconsistent with the intermediate NK activity of beige mice whenassayed in the standard cytotoxicity assay (51) and with previous reports that the NK cells in beige mice can be activatedwith interferon inducers (52). Since swainsonine administrationin Fig. 4 was the same for all animals, yet each of the NK-deficient groups of mice was defective in its response to theantimetastatic effects of the drug, these findings are consistentwith the hypothesis that swainsonine inhibits experimentaltumor metastasis by enhancing NK cell function.

Effect of Swainsonine on NK Cell Activity. Our second approach to examining the relationship between swainsonine andNK cell activity was to test the effect of drug administration ina standard NK cytotoxicity assay. Initially, two different strategies were used, (a) Groups of mice were fed swainsonine for24 h, and then spleen cells were removed and tested for theirability to lyse 5lCr-labeled B16-F10 cells or YAC-1 cells in

comparison to spleen cell preparations from mice receivingunsupplemented drinking water. A representative experimentfor YAC-1 cells is shown in Table 2. In this assay, swainsoninetreatment had no effect on YAC-1 cell lysis, demonstrating thatthe drug was not functioning as an activator of NK cells. Similar

Table 1 Examination of the effect ofBl6-F10 cell inoculum size on theantimetastatic activity of swainsonine in anti-asialo-GM, antibody-treated mice

C57BL/6 mice were depleted of NK activity by i.v. injection of anti-asialo-GM i antibody 2 days before tumor cell injection. Swainsonine (3 Â¿tg/mlindrinking water) was administered for the 24-h period immediately precedingtumor cell injection. Statistical significance was estimated using the Mann-Whitney V test.

TreatmentNone

SW*Anti-aGM,

Anti-aGM,/SWAnti-aGM,

Anti-aGM,/SWAnti-aGMi

Anti-aGM./SWNo.

of cellsinjected8x

IO48x10*8x

10"8x10*4x

IO44xIO42x

10*2x10"Colony

no.40.0Â±15.3Â°

4.9 Â±1.7150.0

Â±11.7151.3 Â±19.665.1

Â±6.062.4 Â±3.325.9

Â±4.619.7 Â±4.3Inhibition(%)87.7-0.94.123.9P<0.001NSNSNS

" Mean Â±SD.* SW, swainsonine; anti-aGMi, anti-asialo-GM,; NS, not significant.

Table 2 Effect of swainsonine on activation ofNK cellsC57BL/6 mice were provided with either unsupplemented drinking water or

drinking water containing 3 Mg/ml of swainsonine for 24 h prior to sacrifice.Spleen cells were then prepared from each group of mice and tested for theirability to lyse "Cr-labeled YAC-1 cells at the indicated effectortarget cell ratios.

Effectortarget cell ratio

Treatment 10:1 25:1 50:1 75:1None 9.0 Â±2.4Â° 17.1 Â±4.9 24.1 Â±1.8 31.7 Â±2.1 52.4 Â±2.8

Swainsonine 8.1 Â±1.5 25.3 Â±2.4 26.1 Â±1.6 31.4 Â±1.5 46.0 Â±1.5Â°Mean Â±SD.

results to those in Table 2 were obtained by exogenous additionof swainsonine to spleen cells from untreated mice prior to andduring the cytotoxicity assay. In addition, no enhancement ofB16-F10 cell killing was observed in the same assays (data notshown), (b) We examined the possibility that swainsonine mightstimulate spleen cell proliferation rather than activate existingNK cells. In this scenario, no difference would be expected inthe cytotoxicity assay performed in Table 2 because all resultsare normalized to standard effectontarget cell ratios and therefore do not take into account possible differences in total spleencell numbers in the original cell preparations. To correct forthis, we examined the NK activity of spleen cells prepared fromgroups of mice either treated or untreated with swainsonine asbefore, but rather than diluting each preparation to yield thesame effectontarget ratios, both sets of cells were resuspendedto the same volume and then tested in the cytotoxicity assay.In this way, the NK activity was determined per animal ratherthan per spleen cell. In these experiments, swainsonine wasfound to cause a reproducible increase in spleen cell number[32.0 Â±22.2% increase in 2 days in 8 independent experiments(P < 0.05); see legend to Fig. 5]. Therefore, this approachactually resulted in a higher number of spleen cells being addedto the incubation mixture for swainsonine-treated mice. Fromthe data presented in Fig. 5a, it is clear that spleen cell preparations obtained from swainsonine-treated mice 2 days afteradministration of the drug possessed 2 to 3 times the NKactivity of spleen cell preparations from mice receiving unsupplemented drinking water (26.7 LU20/105 for control mice and6.19 LU2o/105 for swainsonine-treated mice). This result was

consistently obtained in 6 independent experiments. Interestingly, the swainsonine-induced increase in NK cell activitypeaked after 2 days and had declined by 3 days. This result isconsistent with the time course profile of NK cell stimulationachieved by administration of previously characterized biological response modifiers (33-36). The increased killing observed

1413




40 -

20 40 60 80 100

20 -

30 45 60

E:T RATIO

Fig. 5. Effect of swainsonine on NK cell activity. In A, C57BL/6 mice wereprovided with either unsupplemented drinking water (O) or drinking water containing 3 jig/ml of swainsonine for a 24-h period commencing 1 (A), 2 (â€¢),or 3days (A) before sacrifice. Spleen cells from untreated mice were resuspended toan appropriate density for mixing with target cells at the indicated effectortargetcell ratios (E:T). After correcting for small differences in the average body weightof the groups of mice, the final spleen cell pellets from swainsonine-treatedanimals were resuspended to the same final volume as untreated animals irrespective of cell number. Each cell suspension was then tested for specific lysis of"Cr-labeled YAC-1 cells. When the results of 8 independent experiments werecompared, the average number of spleen cells/mouse was found to be 6.24 Â±0.16x Id' for untreated animals and 8.08 Â±0.17 x Id" for animals receiving

swainsonine. B, effect of anti-asialo-GM, antibody on swainsonine-mediatedstimulation of NK activity. C57BL/6 mice were provided with unsupplementeddrinking water (A, O) or with drinking water containing 3 Mg/ml of swainsonine(A, â€¢)for 24 h commencing 2 days before sacrifice as above. In addition, half ofeach group received an i.v. injection of anti-asialo-GMi antibody (0.2 ml of 1:50dilution with phosphate-buffered saline; â€¢,O) 2 days before sacrifice. Spleen cellswere prepared from each group of mice as above and tested for cytolytic activity.

in Fig. 5a was accounted for by enhanced NK cell reactivity,since preincubation of mice with anti-asialo-GM] antibodycompletely eliminated the swainsonine effect (Fig. 5Â¿>).Theseresults imply that swainsonine acts as a mitogenic stimulus forspleen cell proliferation in vivo and that through its ability toincrease the number of active NK cells, swainsonine is able toinduce an antimetastatic response. Recent studies performed inour laboratory have shown that swainsonine is able to enhancethymidine incorporation into spleen cell preparations by afactor of 1.6, thereby confirming the ability of the drug to actas a spleen cell mitogen.4 It is notable that the high increase in

spleen cell number observed after 2 days of treatment withswainsonine (32.0%) suggests that cell populations other thanNK cells are also stimulated to proliferate as a result of drugtreatment. It will be interesting in the future to examine exactlywhich cell types are able to respond to swainsonine.

In summary, we have extended our original findings on theantimetastatic effects of swainsonine and now show that: (a)

4S. I White ci al., manuscript submitted for publication.

swainsonine is able to interfere with the pulmonary colonizationof B16-F10 cells after only a short, systemic administration tomice; (Â¿>)the kinetics of inhibition of both experimental metastasis and pulmonary retention of injected tumor cells is consistent with swainsonine mediating its biological effects 1 to 3days after administration; (c) NK cells appear to be requiredfor swainsonine function in vivo, since mice with depleted NKcell activity are unable to respond to swainsonine; and (d) directevidence is presented to show that swainsonine is able tostimulate spleen cell proliferation and that this activity may beone mechanism through which the antimetastatic effects of thedrug are manifested.

Currently, the biochemical mechanisms of swainsonine-mediated enhancement of NK cell activity and inhibition of experimental metastasis are unknown. In the future, it will be particularly instructive to examine whether these effects are relatedto the well-characterized ability of swainsonine to function asan inhibitor of glycoprotein processing (12, 13, 22, 56-59), orwhether the drug has other as yet undiscovered activities. It willalso be important to test the antimetastatic and antitumoractivity of swainsonine in combination with other chemother-apeutic and/or immunotherapeutic agents. Since swainsonineexhibits activity after only transient exposure of animals tomicrogram levels of the drug, a dosage unlikely to produce sideeffects, it is conceivable that combination therapy with swainsonine may help alleviate some of the toxicity obtained withother biological response modifiers.

ACKNOWLEDGMENTS

We gratefully acknowledge Dr. I. J. Fidler for B16-F10 cells and Dr.Kenneth M. Yamada for critical reviewing of the manuscript.

REFERENCES

1. Terry, W. D., and Rosenberg, S. A. (eds.). Immunotherapy of Human Cancer.New York: Elsevier North-Holland, 1982.

2. Herberman, R. B. Biological response modifiers for the therapy of cancer.In: E. Tsubura, I. Urushizaki, T. Aoki, S. SonÃ©,and Y. Kaneko (eds.),Rationale of Biological Response Modifiers in Cancer Treatment, pp. 240-255. Princeton: Excerpta Medica, 1985.

3. Rosenberg, S. A., and Lotze, M. T. Cancer immunotherapy using interleukin-2 and interleukin-2-activated lymphocytes. Annu. Rev. Irumunni.. 4: 681-709, 1986.

4. Forni, G., Giovarelli, M., Santoni, A., Modesti, A., and Forni, M. Tumourinhibition by interleukin-2 at the tumour/host interface. Biochim. Biophys.Acta, Â«65:307-327, 1986.

5. Talmadge, J. E., and Herberman, R. B. The preclinical screening laboratory:evaluation of immunomodulatory and therapeutic properties of biologicalresponse modifiers. Cancer Treat Rep., 70: 171-182, 1986.

6. Hollinshead, A. Immunotherapy trials: current status and future directionswith special emphasis on biologic drugs. Springer Semin. Immunopathol., 9:85-103, 1986.

7. Lotze, M. T., Frana, L. W., Sharrow, S. O., Robb, R. J., and Rosenberg, S.A. In vivo administration of purified human interleukin 2. I. Half-life andimmunologie effects of the Jurkat cell line-derived interleukin 2. J. Immunol.,Â¡34:157-166, 1985.

8. Lotze, M. T., Matory, Y. L., Ettinghausen, S. E., Raynor, A. A., Sharrow,S. O., Seipp, C. A., Custer, M. C., and Rosenberg, S. A. In vivoadministrationof purified human interleukin 2. II. Half life, immunologie effects, andexpansion of peripheral lymphoid cells in vivo with recombinant IL 2. J.Immunol., 135: 2865-2875, 1985.

9. Urba, W. J., and Longo, D. L. a-Interferon in the treatment of nodularlymphomas. Semin. Oncol., 13 (Supplement 5): 40-47, 1986.

10. Takatsuki, A., Arinia. K., and Tamura, G. Tunicamycin, a new antibiotic. I.Isolation and characterization of tunicamycin. J. Antibiot., 24: 215-223,1971.

11. Takatsuki, A., and Tamura, G. Effect of tunicamycin on the synthesis ofmacromolecules in cultures of chick embryo fibroblasts infected with Newcastle disease virus. J. Antibiot., 24: 785-794, 1971.

12. Dorling, P. R., Huxtable, C. R., and Colegate, S. M. Lysosomal storage inSwainsona spp. toxicosis: an induced mannosidosis. Neuropathol. Appi.Neurobiol., 4: 285-295, 1978.

13. Colegate, S. M., Dorling, P. R., and Huxtable, C. R. A spectroscopicinvestigation of swainsonine: an a-mannosidase inhibitor isolated from

1414




Swainsona canescens. Aust. J. Chem., 32: 2257-2264, 1979. 37.14. Hohenschutz, L. D., Bell, E. A., Jewess, P. J., Leworthy, D. P., Pryce, R. J.,

Arnold, E., and ('lardv, J. Castanospermine, a 1,6,7,8-tetrahydroxyoctahy-dro-indolizine alkaloid, from seeds of Castanospermum australe. Phytochem- 38.istry, 20:811-814, 1981.

15. Saul, K . Chambers, J. P., Molyneux, R. J., and Elbein, A. D. Castanosper-mine, a tetrahydroxylated alkaloid that inhibits ^-glucosidase and 0-gluco- 39.cerebrosidase. Arch. Biochem. Biophys., 221: 593-597, 1983.

16. Ir imuni. T., Gonzalez, R., and Nicolson, G. L. Effect of tunicamycin on B16metastatic melanoma cell surface glycoproteins and blood-borne arrest and 40.survival properties. Cancer Res., 41: 3411-3418, 1981.

17. IriunirÃ .T., and Nicolson, G. L. The role of glycoconjugates in metastaticmelanoma blood-borne arrest and cell surface properties. J. Supramol. Struct. 41.Cell. Biochem., 17: 325-336, 1981.

18. Olden, K., Humphries, M. J., and White, S. L. Biochemical effects andcancer therapeutic potential of tunicamycin. In: R. Reisfeld and S. Sell (eds.), 42.Monoclonal Antibodies and Cancer Therapy, pp. 443-472. New York: AlanR. Liss, Inc., 1985.

19. Humphries, M. J., Matsumoto, K., White, S. L., and Olden, K. Oligosaccha-ride modification by swainsonine treatment inhibits pulmonary colonization 43.by B16-F10 murine melanoma cells. Proc. Nati. Acad. Sci. USA, 83: 1752-1756, 1986.

20. Humphries, M. J., Matsumoto, K., White, S. L., and Olden, K. Inhibition of 44.experimental metastasis by castanospermine in mice: blockage of two distinctstages of tumor colonization by oligosaccharide processing inhibitors. CancerRes., 46: 5215-5222, 1986. 45.

21. Dennis, J. W. Effects of swainsonine and polyinosinic:polycytidylic acid onmurine tumor cell growth and metastasis. Cancer Res., 46:5131-5136,1986.

22. Molyneux, R. J., and James, L. F. Loco intoxication: indolizidine alkaloids 46.of spotted lociiweed (Astragalus lenliginosus). Science (Wash. DC), 216:190-191, 1982.

23. Humphries, M. J., Olden, K., and Yamada, K. M. A synthetic peptide from 47.fibronectin inhibits experimental metastasis of murine melanoma cells. Science (Wash. DC), 233:467-470, 1986.

24. Talmadge, J. E., Fidler, I. J., and Oldham, R. K. (eds.). Screening for 48.Biological Response Modifiers: Methods and Rationale. Boston: MartinusNijhoff, 1985. 49.

25. Broquist, H. P., Mason, P. S., Hagler, W. M., and Croom, W. J. Transmission of swainsonine into milk. Fed. Proc., 44:1860, 1985.

26. Reading, C. L., and Hutchins, J. T. Carbohydrate structure in tumor Â¡ninni 50.nity. Cancer Metastasis Rev., 4: 221-260, 1985.

27. White, S. L. Role of lectins in immune recognition. In: K. Olden, and J. B.Parent (eds.). Vertebrate Lectins, pp. 182-194. New York: Van Nostrand 51.Reinhold, 1987.

28. Stutman, O., Dien, P., Wisun, R. E., and Lattime, E. C. Natural cytotoxiccells against solid tumors in mice: blocking of cytotoxicity by D-mannose.Proc. Nati. Acad. Sci. USA, 77: 2895-2898, 1980. 52.

29. Forbes, J. T., Bretthauer, R. K., and Oeltmann, T. N. Mannose 6-, fructosel-, and fructose 6-phosphates inhibit human natural cell-mediated cytotoxicity. Proc. Nati. Acad. Sci. USA, 78: 5797-5801, 1981.

30. Werkmeister, J. A., Pross, H. F., and Roder, J. C. Modulation of K562 cells 53.with sodium butyrate. Association of impaired NK susceptibility with sialicacid and analysis of other parameters. Int. J. Cancer, 32: 71-78, 1983.

31. Dennis, J. W., and Laferte, S. Recognition of asparagine-linked oligosaccha-rides on murine tumor cells by natural killer cells. Cancer Res., 45: 6034- 54.6040, 1985.

32. HiÃ±o,M., Nakayama, O., Tsurumi, Y., Adachi, K., Shibata, T., Terano, H.,Kohsaka, M., Aoki, H., and Imanaka, H. Studies of an immunomodulator, 55.swainsonine. I. Enhancement of immune response by swainsonine in vitro.J. Antibiot. (Tokyo), 38:926-935, 1985.

33. Herberman, R. B., Nunn, M. E., Holden, H. T., Staal, S., and Djeu, J. Y.Augmentation of natural cytotoxic reactivity of mouse lymphoid cells against 56.syngeneic and allogeneic target cells. Int. J. Cancer, 19: 555-564, 1977.

34. Welsh, R. M., and Zinkemagel, R. M. HeterospecifÃ¯ccytotoxic cell activityinduced during the first three days of acute lymphocytic choriomeningitis 57.virus infection in mice. Nature (Lond.), 268: 646-648, 1977.

35. Herberman, R. B., Djeu, J. Y., Ortaldo, J. R., Holden, H. T., West, W. H.,and Bonnard, G. D. Role of interferon in augmentation of natural and 58.antibody-dependent cell-mediated cytotoxicity. Cancer Treat. Rep., 62:1893-1896, 1978.

36. Djeu, J. Y., Heinbaugh, J. A., Holden, H. T., and Herberman, R. B. 59.Augmentation of mouse natural killer cell activity by interferon and interferon inducers. J. Immunol., 122:175-181, 1979.

Kasai, M., Iwamori, M., Nagai, Y., Okumura, K., and Tada, T. A. glycolipidon the surface of mouse natural killer cells. Eur. J. Immunol., 10: 175-180,1980.Young, W. W.. Hakomori, S. I., Durdik, J. M., and Henney, C. S. Identification of ganglio-AMetraosylceramide as a new cell surface marker for murinenatural killer (NK) cells. J. Immunol., 124: 199-201, 1980.Kasai, M., Yoneda, T., Habu, S., Maruyama, Y., Okumura, K., and Tokun-aga, T. In vivo effect of anti-asialo GM1 antibody on natural killer activity.Nature (Lond.), 291: 334-335, 1981.Beck, B. N., Gillis, S., and Henney, C. S. Display of the neutral glycolipidganglio-n-tetraosylceramide (asialo GM1) on the cells of the natural killerand T lineages. Transplantation (Baltimore), 33: 118-122, 1982.Kawase, I., Urdal, D. L., Brooks, C. G., and Henney, C. S. Selective depletionof NK cell activity in vivo and its effect on the growth of NK-sensitive andNK-resistant tumor cell variants. Int. J. Cancer, 29: 567-574, 1982.Wiltrout, R. H., Santoni, A., Peterson, E. S., Knot!, D. C., Overton, W. R.,Herberman, R. B., and Holden, H. T. Reactivity of anti-asialo (.M, serumwith tumoricidal and non-tumoricidal mouse macrophages. J. Leuk. Biol.,57:597-614, 1985.van Putten, L. M., Kram, L. K., van Dierendonck, H. H., Smink, T., andFuzy, M. Enhancement by drugs of metastatic lung nodule formation afterintravenous tumour cell injection. Int. J. Cancer, 15: 588-595, 1975.Carmel, R. J., and Brown, J. M. The effect of cyclophosphamide and otherdrugs on the incidence of pulmonary mÃ©tastasesin mice. Cancer Res., 37:145-151, 1977.van Putten. L. M., and Lelieveld, P. Factors determining cell killing bychemotherapeutic agents in vivo. I. Cyclophosphamide. Eur. J. Cancer, 6:313-321, 1970.Stockman, G. D., Heim, L. R., South, M. A., and Trentin, J. J. Differentialeffects of cyclophosphamide on the B cell and T cell compartments of adultmice. J. Immunol., Â¡10:277-282, 1973.Lagrange, P. H., Mackaness, G. B. and Miller, T. E. Potentiation of T-cell-mediated immunity by selective suppression of antibody formation withcyclophosphamide. J. Exp. Med., 139: 1529-1539, 1974.Roder, J., and Duwe, A. The beige mutation in the mouse selectively impairsnatural killer cell function. Nature (Lond.), 27Â«:451-453, 1979.Roder, J. C. The beige mutation in the mouse. I. A stem cell predeterminedimpairment in natural killer cell function. J. Immunol., 123: 2168-2173,1979.Roder, J. C., Lohmann-Mathes, M-L., Domzig, W., and Wigzell, H. Thebeige mutation in the mouse. II. Selectivity of the natural killer (NK) celldefect. J. Immunol., 123: 2174-2181, 1979.Roder, J. C. Different genes regulate tumor cell recognition and cytolysis byNK cells in the mouse. In: E. Skamene, P. A. L. Kongshavn, and M. Landy(eds.), Genetic Control of Natural Resistance to Infection and Malignancy,pp. 405-412. New York: Academic Press, Inc., 1980.Brunda, M. J., Holden, H. T., and Herberman, R. B. Augmentation ofnatural killer cell activity of beige mice by interferon and interferon inducers.In: R. B. Herberman (ed.), Natural Cell-mediated Immunity against Tumors,pp. 411-415. New York: Academic Press. Inc., 1980.Oehler, J. R., and Herberman. R. B. Natural cell-mediated cytotoxicity inrats. III. Effects of immunopharmacologic treatments on natural reactivityand on reactivity augmented by polyinosinic-polycytidylic acid. Int. J. Cancer,21: 221-229, 1978.Mantovani, A., Luini, W.. Peri, G., Vecchi, A., and Spreafico. F. Effect ofchemotherapeutic agents on natural cell-mediated cytotoxicity in mice. J.Nati. Cancer Inst., 61: 1255-1261, 1978.Djeu. J. Y., Heinbaugh, J. A., Vieira, W. D.. Holden, H. T., and Herberman,R. B. The effect of immunopharmacological agents on mouse natural cell-mediated cytotoxicity and on its augmentation by poly I:C. Immunophar-macology, 1: 231-244, 1979.Dorling, P. R., Huxtable, C. R., and Colegate, S. M. Inhibition of lysosomala-mannosidase (EC 3.2.1.24) by swainsonine, an indolizidine alkaloid isolated from Swainsona canescens. Biochem. J., 191:649-651, 1980.Elbein, A. D., Solf. R., Dorling, P. R., and Vosbeck, K. Swainsonine: aninhibitor of glycoprotein processing. Proc. Nati. Acad. Sci. USA, 78: 7393-7397, 1981.Elbein, A. D., Dorling, P. R., Vosbeck, K., and Horisberger, M. Swainsonineprevents the processing of the oligosaccharide chains of influenza virushemagglutinin. J. Biol. Chem., 257: 1573-1576, 1982.Tulsiani, D. R. P., Harris, T. M., and Touster, O. Swainsonine inhibits thebiosynthesis of complex glycoproteins by inhibition of Golgi mannosidase H.J. Biol. Chem., 257: 7936-7939, 1982.

1415



1988;48:1410-1415. Cancer Res Martin J. Humphries, Kazue Matsumoto, Sandra L. White, et al. Swainsonine, a New Antimetastatic ImmunomodulatorAugmentation of Murine Natural Killer Cell Activity by

Updated version

http://cancerres.aacrjournals.org/content/48/6/1410

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

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/48/6/1410To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

Augmentation of Murine Natural Killer Cell Activity by ... · biological response modifiers such as interferon and interleukin 2 is lower than that which exhibits maximum biological