Tumoricidal activation of murine alveolar macrophages by muramyldipeptide substituted mannosylated serum albumin

Vol. 121, No. 2, 1984 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

June 15, 1984 Pages 579-584

TUMORICIDAL ACTIVATION OF MURINE ALVEOLAR MACROPHAGES BY MURAMYLDIPEPTIDE SUBSTITUTED MANNOSYLATED

SERUM ALBUMIN

Michel Monsigny, Annie-Claude Roche and Pascal Bailly

Centre de Biophysique Moleculaire,, C. N. R.S. 45045 Orleans Cbdex. Universite d’OrlCns, 45046 Orleans Cddex (France)

Received April 24, 1984

Rat and mouse alveolar macrophages have almost no spontaneous tumoricidal activity and are only slightly activated by muramyldipeptide (MDP). When MDP was carried by serum albumin, the activation was higher than with free MDP but only at high concentration. When MDP was bound to a neoglycoprotein (mannosylated serum albumin) - which binds to the sugar binding receptor at the macrophage cell surface and is actively endocytosed - the activation of rat or mouse alveolar macrophages is dramatically enhanced even at very low concentration of neoglycoprotein-bound MDP. Furthermore, neoglycoprotein-bound MDP injected i.v. or i-p. was found to be able to activate alveolar macrophages, the activity of which was maximal after 48 hours in mice and 72 hours in rats. Such conjugates have so potential values as new immu- nostimulant agents in cancer and parasite therapy.

Macrophages which are associated within tumors are usually not

sufficient to inhibit tumor growth (1). Lymphokines (2) or bacterial products

such as muramyldipeptide (3) in in vitro experiments are able to activate ma-

crophage to a tumoricidal state. Unfortunately, muramyldipeptide is not efficient

in vivo because it is very rapidly cleared (4). Macrophages did not display

MDP receptors at their cell surface (5) but take up MDP by a fluid non speci-

fic pinocytic process. While free MDP is not an efficient macrophage activator,

it can become efficient by encapsulation within liposomes (6), because conve-

nient liposomes are selectively taken up by macrophages. Liposomes are not

yet suitable to be used as drug carriers (7). In order to make MDP efficient

in both in vitro and in vivo experiments, it should be possible to use bio-

logical macromolecules such as glycoproteins or neoglycoproteins which - having

receptors at the macrophage surface - are actively internalized and may deli-

ver carried MDP molecules. Rat alveolar macrophages being known to haveacell

surface mannose binding protein (8), MDP-mannosylated serum albumin conju-

gates could be efficient macrophage activators. In this paper, we describe the

Abbreviations : BSA : bovine serumalbumin ; Man-BSA : mannosylated bovine serumalbumin ; MDP : 2-acetamido-2-deoxy-3-~-(2’-P-lactyl-~-alany!-~- glutamylamine)-D-glucose ; MDP-BSA : bovine serum albumin substrtuted with MDP ; MEM : minimal essential medium.

0006-291X/84 $1.50

579 Copyright 0 1984 by Academic Press, Inc.

All rights of reproduction in any form resen,ed.

Vol. 121, No. 2, 1984 8lOCHEMlCAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

preparation of such conjugates and show that they are more active than free

MDP in inducing tumoricidal activation of rat and mouse alveolar macrophages

in in vitro and in vivo experiments.

MATERIALS AND METHODS

N-acetylmuramyl-L-alanyl-Q-isoglutamine (MDP) was obtained from lnstitut Pas?eur (Paris, France) or W?IS synthetized according to (9). Chemi- cals (if not otherwise stated) were purchased from Janssen Chimica (Pantin, France). Dimethylformamide (Merck, Darmstadt, F.R.G.) was extemporaneous- ly distilled on fluoro 2,4-dinitrobenzene (0,2 mglml) and HNaCO, (O-1 mglml; Merck, Darmstadt, F.R.C.). The purity of small molecular weight compounds was checked by thin layer chromatography on Silicagel G plate (Merck) with solvent A: CHCI,/CH, OH/H,D, 13:8:2 respectively per volume, or solvent B : CHCI,/CH,0H/CH3C0,H/H~0, 2:2:1:1 respectively per volume.

Mannosylated serum albumin was obtained by allowing o-p-manno- pyranosylphenylisothiocyanate to react with serum albumin according to a method derived from (10). Briefly, p-nitrophenyl-a-Q-mannopyranoside (301.5 mg, 1 mmol, solvent A, Rf : a64 ; Sigma, Szint-Louis, Mo.,USA) was dissolved in 20 ml of a 4:1 (‘v/v) methanol-water mixture. To this solution, 30 mg of 10 % Palladium on charcoal (Merck) was added. The suspension was stirred. under hydrogen (1 atm) at room temperature for two hours. After filtration, the methanolic solution was evaporated under reduced pressure at 40°C. The Q-aminophenyl-crQ-mannopyranoside (solvent A, Rf : 0.44) was dissolved in 30 ml of 0-l M sodium carbonate, pH 8.6 ; to this solution was added 30 ml CHCI, containing 2 mmoles of thiophosgene, This mixture was shaked for 30 min, the o-Q-mannopyranosyl-phenylisothiocyanate (solvent A, Rf : 0.67) crystallized off: Crystals were harvested by filtration, washed with chloroform and cold distilled water. The o-Q-mannopyranosyl-phenylisothiocyanate was recrystallysed by slowly coo%g a water solution from 60°C to 4%.

To a solution of (268 mg, 4 pmoles) bovine serum albumin (BSA : IBF Reactifs, Pointet-Girard, Villeneuve-la-Garenne, France) in 25 ml 6.1 M sodium carbonate buffer (pH %5), crystalline crQ-mannopyranosyl- phenylisothiocyanate (31 mg, 0-l mmole) was added. Affer 20 hr stirring, mannopyranosyl-thiocarbamyl serum albumin was purified by gel filtration on anllltrogel ACA 202 (IBF Reactifs) column (3 x 56 cm) in distilled water. The mannose content determined by the resorcinol-sulfuric acid method (11) was found to be 20 mannose residues per molecule (Man-BSA).

MDP serum albumin conjugates were prepared by adding MDP- hydroxysuccinimide ester to a solution of serum albumin or Man-BSA in 1 M HNaCO, buffer, pH 8.5. MDP-hydroxysuccinimide ester was prepared by dissolving 1.00 ing (0.2 mmole). MDP,. 46 mg ‘f&24 mmole) dicyclohexylcarbo- diimide and 26.6 mg (6.24 mmole) ‘Nlhydroxysuccinimide in’ 2 ml, of freshly distilled dimethylformamide, After 24 hr at 25’%,’ MDP was quantitatively conver- ted into MDP-hydroxysuccinimide ester (sol<ent B, Rf : 0.57 and ‘O-67 respectively). Dicyclohexylurea was recovered by filtration. The solution of MDP- hydroxysuccinimidyl-ester in dimethylformamide was added to a 17 ml. solution of 170 mg of Man-BSA or 155 mg of BSA in 1 M HNaCO, buffer, pH 8.5, The reaction mixture was stirred for 24 hr at room temperature. The conjugate was purified by gel filtration on Ultrogel ACA 202 or Ultrogel GF 05 ‘(IBF RBactifs) column (3 x 50. cm) in phosphate buffered saline, pH 7.4. The MDP content of the conjugate, determined by the. method of Lewy and Mc Allan (12) was found to be 16 molecules per conjugate, The lipopolysaccharide content of MDP conjugates checked according to the previously described method (13) was found to be lower than 50 pg/mg of conjugate. Depending on the

580


activated MDP/Man-BSA ratio, it is easy to obtain MDP-Man-BSA conjugates containing between 5 and 30 MDP molecules per molecule.

Cells and animals

Lewis rats were obtained from CSEAL, CNRS, Orleans, France. C 571 BI 6 mice were purchased from IFFA CREDO, Lyon, France.

The DBA/Z lymphoma L 1210 cells (kindly given by Dr. I. Cresser, Villejuif, France) were adapted to grow in suspension in MEM with Earle’s salts supplemented with 10 % heat inactivated horse serum (IBF Reactifs) and were collected from exponential growth culture.

In v&eo activation of murine alveolar macrophaqes

Mouse and rat alveolar macrophages were obtained from anesthe- tized animals, by tracheobronchial lavages with sterile phosphate buffered saline containing 0.02 % disodium ethylenetetraacetate at 37OC according to (14). Macrophages were suspended in RPMI. medium supplemented with 100 units/ml penicillin and 100 ug/ml streptomycin, spun down by centrifugation at 500 g and resuspended in the same medium. Two hundred ul of macrophages suspension (5 x 105 cells/ml) identified by neutral red uptake were allowed to adhere for 1 hr at 37OC in wells of Microtest III plates (Falcon, Beckton- Dickinson, But, France) in a humidified atmosphere of 5 % CO,/95 % air. Non adherent cells were then washed out. Macrophages were incubated for 24 hr with free or bound MDP, washed and then 104 tumor target cells (L 1210 cells) were added and cocultivated with macrophages for 48 hr (macrophages : target cells ratio, 1O:l) ; culture medium was MEM with Hank’s salt, supplemented with 2 mM L-glutamine, antibiotics and 10 % heat inactivated fetal bovine serum (Seromed, Biopro, Mulhouse, France). To determine growth inhibition, 3H-thymidine (0.1 uCi, 3.7 KBq ; specific activity 0.74-l -1 TBqlmmol, CEA, Saclay, France) was added 4 hr before harvesting. Cells were collected and washed on Whatman glass fiber filters and radioactivity measured in aqueous counting scintillant (ACS, Amersham, U.K.) in a 6-scintillator counter (Beckman, Geneve, CH). The percentage of cytostatic activity was expressed as growth inhibition : Cl % = (R-S/R) 100, where R is the radioactivity incorporated in tumor cells cultivated on unstimulated macrophages and S is the radioactivity incorporated in tumor cells cultivated on stimulated macrophages.

In v<vo activation of alveolar macrophages

Male Lewis rats (200 g) received 0.5 ml of sterile sodium chloride containing 100 pg of MDP bound to Man-BSA by either intraperitoneal injection or intravenous injection. Alveolar macrophages were harvested 24, 48, 72 or 96 hours later, washed and allowed to adhere for 1 hr at 37OC as described above, in Microtest III wells (105 macrophages per well). L 1210 (104 cells) were added and cells were cocultivated and labeled as described above. Male C 57/Bl 6 mice (20 g)received either 30 pg, 11 pg or 1.1 pg of MDP bound to Man-BSA by intravenous injection. Mouse alveolar macrophages were harvested and used as described for rat macrophages.

RESULTS AND DISCUSSION

Mannosylated-BSA is very easily obtained by reaction of crys-

talline cc-D-mannopyranosyl-phenylisothiocyanate on bovine serum albumin. - The coupling reaction is almost quantitative (yield higher than 90 8). There-

fore it is possible to reproductibily prepare neoglycoproteins with a given

number of sugars. As it is known that neoglycoproteins containing mannose

bind and are endocytosed by rat alveolar macrophages (8) and mouse perito-

581


neal macrophages (5,15), neoglycoproteins containing 20 residues of mannose

wereselected in order to be further substituted with 10 to 30 MDP molecules.

The coupling efficiency of MDP-hydroxy-succinimidyl ester to Man-BSA was

usually close to 30 %, because of the relatively fast hydrolysis of the acti-

vated carboxylic group in slightly basic solution, However, under the used

conditions, the N-acetylmuramyl moiety was not degraded and so it was pos-

sible to recover the unbound MDP after gel filtration and to recycle it to

obtain authentic MDP-hydroxy-succinimidyl ester.

In vitroactivation of rat alveolar macrophages

Rat alveolar macrophages did not significantly inhibit the

incorporation of 3H-thymidine into L 1210 leukemic cells. Macrophages pre-

incubated for 24 hours in the presence of Man-BSA (lo-100 pg/ml range)

were not more cytostatic. Free MDP had little effect, even when used at

high concentration such as between 10 to 30 pg/mI (Figure 1). High concen-

trations of MDP-BSA conjugate (25 to 100 ug/ml) containing 5 to 20 pg bound

MDP were able to activate macrophages to some extent (25 %). In contrast,

MDP-Man-BSA conjugate was found to be quite efficient in rendering cyto-

toxic rat alveolar macrophages ; at concentration as low as 1 pglml of carrier

bound MDP, macrophages were still activated enough to inhibit 3H-thymidine

incorporation into tumor cells (Figure 1). It is worthwile to notice that MDP-

BSA conjugate is more efficient than free MDP. These results show that

the activity of MDP directly bound to serum albumin is not impaired and

that MDP-BSA which is internalized. by pinocytosis is much less active than

25 50 100 Protein lJ9/ml

-PC--- In vitro cytostatic activity of rat alveolar macrophages activated by ree MDP. orotein-bound MDP and mannosvlatec&BSA. Mkrophages-( iO5) were activated for 24 hr by free MDP (x-x), Man-BSA (o-o), MDP-BSA ( A-0-A ) and MDP-Man-BSA (o-o), and then cocultivated for 48 hr with 104 L 1210 cells. Percentage of cytostatic activity was determined by measuring 3H-thymidine incorporation. Control experiments conducted with alveolar macrophages in the absence of stimulant gave 10 _+ 5 % growth inhibition. Experiments were done in triplicate.

582


MDP-Man-BSA which is internalized by adsorptive endocytosis via the mem- -

brahe lectin (8). Neither free MDP, Man-BSA, MDP-Man-BSA or a mixture of

free MDP and Man-BSA were found to inhibit ‘L 1210 ceils growth, when these

compounds were used in the concentration ra’nge tested to activate macropha-

ges. Experiments along this line with mouse peritoneal macrophages and human

monocytes (unpublished results) led to a similar efficient activation of macro-

phages. On the basis of the above findings, it was decided to test the abi-

lity of MDP-neoglycoprotein conjugate to activate macrophages in vivo.

In ~$0 activation of murine alveolar macrophages

Rat alveolar macrophages were not made cytostatic upon i-v.

injection of 1 mg of MDP per animal. In contrast, after i.v, injection of about

100 pg of MDP bound to Man-BSA, alveolar macrophages were activated and

the maximal activation was evidenced 72 hours after injection (Figure 2), Simi-

lar results were obtained when MDP bound to Man-BSA was injected intra-

peritoneally. The activation returned to a very low level 24 hours later (96 hr

after injection of MDP-neoglycoprotein conjugate). It is interesting to point

out that i.v, and i.p. injections have similar efficiency and a similar time

course effect. The maximal activation in both cases is limited at 20 % growth

inhibition. In similar experiments conducted in mice, it was found that 10 ug

of MDP bound to mannosylated serum albumin induced the highest alveolar ma-

crophage activation (up to 50 %) 48 hours after i.v. injection (Figure 3). So

25.

0 0 2 0 24 48 72 96 Hours ofkr injcckx, 0 3

Hours after injection

+--- Fi ure 2 : In vivo activation of rat alveolar macrophages. Cytostatic activity o Lewis ratmar macrophages activated by intravenous or intraperitoneal injection of 100 ug MDP bound to Man-BSA and harvested 24, 48, 72 or 96 hr later. Macrophages (105) and 10 4 L 1210 cells were cocultivated for 48 hr. Empty bars : intraperitoneal injection ; dashed bars, intravenous injection. See details under figure 1 and Materials and Methods.

Figure 3 : In vivo activation of mouse alveolar macrophages, Cytostatic activity of C 57 Bl/6 mouse alveolar macrophages activated by intravenous injection of MDP bound to Man-BSA and harvested 24. 48 or 72 hours later. Macrophages (105) and 104 L 1210 cells were cocultivated for 48 hr. Empty bars : 1.1 Dg of bound MDP ; dashed bars ; 11 ug of bound MDP ; full bars ; 30 ug of bound MDP. See details under figure 1 and Materials and Methods.

583


the activation of mouse alveolar macrophages was higher than that of rat ma-

crophages. In control experiments using Man-BSA (100 uglmouse) or free MDP

+ Man-BSA (10 ug and 100 ug respectively per mouse), alveolar macrophages

were not activated. It appears therefore that MDP carried by neoglycoproteins

is at least as efficient as MDP carried in liposomes (16) in in situ activation

of alveolar macrophages.

Concluding remarks

Muramyldipeptide bound to a neoglycoprotein (mannosylated BSA)

is much more efficient than free MDP in activating macrophages to a cyto-

static state both in in vitro and in vivo experiments. Furthermore such

activators were found to activate alveolar macrophages not only upon intra-

venous injection but also upon intraperitoneal injection. Such macrophage

activators which can be easily obtained in large amounts at a very low

expense seems to be potential therapeutic agehts particularly in treatment

of tumor metastasis and of parasite invasion such as schistosomiasis. Experi-

ments along these lines are currently under investigation in our laboratory.

Acknowledgements : We thank Ph. BOUCHARD and M.T. BERNEDES for

their skillful technical assistance. This work was partly supported by grants

from lnstitut National de la Sante et de la Recherche Mbdicale,(INSERM,

CRL 811014) and from Association pour le Developpement de la Recherche

sur le Cancer (Aide 415-14). A.C.R. is Chargee de Recherche INSERM,

P.B. received a fellowship of the Ligue Nationale Francaise contre le Cancer.

Mrs. J. Florian is warmly acknowledged for her valuable help in preparing the manuscript.

REFERENCES

1. 2.

Key, M.E. (1983) Cancer Metastasis Reviews, 1, 75-88 Pi'essens, W.F., Churchill, W.H.Jr and David, J.R. (1975)

J. Immunol., 114, 293-299 3. Adam, A., Petit,.F., Le Francier, P. and Lederer, E. (1981)

Mol. Cell. Biochem., 41, 27-47 4. Parant, M., Parant, F.,Xhedid, L., Yapo, A., Petit, J.F.

5. and Lederer, E. (1979) Int. J. Pharmacol., 1, 35-41

Tenu, J.P., Roche, A.C., Yapo, A., Kieda, C., Monsigny,

6. M. and Petit, J.F. (1982) Biol. Cell., 44, 157-164

l3:

Sone, S. and Fidler, I.J. (1981) Cell. Immunol., 57, 42-50 Poste, G. (1983) Biol. Cell., 47, 19-38 Stahl, P.D., Rodman, J.S., MilXr, M.J. and Schlesinger,

9. P.H. (1978) Proc. Natl. Acad. Sci. USA, c, 1399-1403

Merser, C., Sinaji, P. and Adam, A. (1975) Blochem. Biophys. Res. Comn., 66, 1316-1322

10. MC Broom, C.R., Samanen, C.H. and Goldstein,I.J.(1972)Met.Enzymol.,28,212-219 11. Roche, A.C., Barzilay, M., Midoux, P., Junqua, S., Sharon,N.

and Monsigny, M. (1983) J. Cell. Biochem., 22, 131-140 12. Levvy, G.A. and MC Allan, A. (1959) Biochem. X, 73, 127-132 13. Monsigny, M., Kieda, C. and Maillet, T. (1982) EMm J., 73, 127-132 14. Holt, P.G. (1979) J. Immunol. Methods, 27, 189-198 15. Stahl, P. and Gordon, S. (1982) Cell, 93; 43-56 16. Schroit, A.J., Galligioni, E. and Fidler, I.J. (1983) Biol.Cell.,g, 87-94

Documents

Tumoricidal activation of murine alveolar macrophages by muramyldipeptide substituted mannosylated serum albumin