Non-toxic endotoxin polysaccharide induces soluble mediators which potentiate antibody production by murine retrovirus-suppressed splenocytes

Int. J. Immunopharmac., Vol. 10, No. 3, pp. 283-292, 1988. Printed in Great Britain.

0192 0561/88 $3.00+ .00 International Society for lmmunophat'macology.

NON-TOXIC ENDOTOXIN POLYSACCHARIDE INDUCES SOLUBLE MEDIATORS WHICH POTENTIATE ANTIBODY PRODUCTION BY MURINE RETROVIRUS-SUPPRESSED SPLENOCYTES HERMAN FRIEDMAN,* R. CHRISTOPHER BUTLER,t STEVEN SPECTER* and ALOIS N O W O T N Y ~: *Department of Medical Microbiology and Immunology, University of South Florida College of Medicine, Tampa, FL; tDepartment of Microbiology and Immunology, Arlington Hospital, Arlington, VA and *Center for Oral Health Research, University of Pennsylvania, Philadelphia, PA, U.S.A. (Received 4 August 1987 and in final form 21 October 1987)

Abstract -- Mice infected with Friend leukemia virus show marked acquired immunodeficiency characterized by the impairment of immune function of spleen cells to various antigens, both in vivo and in vitro. The large mol. wt. endotoxin derived from Serratia marcescens, as well as a smaller non-toxic polysaccharide derivative, were found to augment the antibody responsivenessof spleen cells from normal as well as FLV-infected mice. In addition, serum from normal donor mice pretreated with BCG and injected either with endotoxin or the polysaccharide derivative potentiated the antibody response of spleen cells from both normal and FLV-infected mice. Similar enhancement was induced by "antibody response helper factor(s)" present in 3 - 5 day spleen culture supernatants from endotoxin or polysaccharide-treated spleen cells from normal mice. Enhancement of the antibody response of spleen cells from FLVAnfectedmice by the antibody helper activity was due to stimulation of B-lymphocytesand reversal of a defect in antibody helper factor(s) formation by macrophages. Similar antibody response enhancing activity was induced by both endotoxin and the non-toxic polysaccharide derivative in cultures of normal spleen cells, adherent spleen cell populations, peritoneal cells and the P388D~ macrophage cell line.

Impairment of immunoresponsiveness, including that mediated by B-cells and T-cells, generally accompanies infection of susceptible mice with murine leukemia viruses such as Friend leukemia virus (FLV). Studies in this and other laboratories have shown that FLV, when injected into susceptible mice, depresses antibody responsiveness to a wide variety of antigens, as well as cell-mediated immune responses (Bendinelli & Friedman, 1980; Bendinelli, Matteucci & Friedman, 1985; Butler, Nowotny & Friedman, 1978; Friedman & Ceglowski, 1971a,b; Friedman & Specter, 1979; Hirano, Friedman & Ceglowski, 1971; Kateley, Kamo, Kaplan & Friedman, 1974). These effects appear to be associated mainly with direct interaction of the virus with immunocompetent cells and their precursors. However, it is also possible that FLV infection affects immune competence by influencing macrophages, which serve as accessory or helper cells for antibody formation and cellular immunity (Bendinelli et al., 1985; Notkins, Mergenhagen & Howard, 1970). In vitro studies have shown that spleen cells from FLV-infected mice, when co-cultured with normal mouse spleen ceils, diminish the expected

antibody responsiveness to sheep erythrocytes (SRBC) in the latter population (Kateley et al., 1974) Such suppression in vitro is induced also by serum or ascites fluid from FLV-infected mice, as well as cellfree extracts and supernatant fluids from virusinfected spleens. Purified virus preparations also suppress antibody formation. Previous studies in this laboratory have shown that impaired antibody responses of splenocytes from FLV-infected splenocytes may be restored under appropriate conditions (Butler & Friedman, 1980; Butler et al., 1978). Addition of peritoneal exudate macrophages to spleen cells from leukemic mice partially restored antibody-forming capacity, demonstrating that the possible mechanism for the unresponsiveness might be related to a defect in macrophage function. This supports the concept that suppression of antibody responses by FLV is not solely due to infection of lymphocytes or their precursors by the virus, but may also involve macrophages, either directly or indirectly. Recently studies in this laboratory have shown that the ability of spleen cell suspensions from FLVinfected mice to produce antibody response

283

284

H. FRIEDMANet al.

"helper" factor(s) in response to endotoxin as a macrophage stimulator was markedly depressed (Butler, Frier, Chapekar, Graham & Friedman, 1983). In addition, endotoxin and similar bacterial stimulators restored, at least partially, antibody responses to SRBC by spleen cells from immunosuppressed FLV-infected mice. It thus seemed plausible that antibody response helper factor(s), either similar to or different from well-described lymphokines or other mediators of immunity produced by normal spleen ceils treated with immunostimulators such as endotoxin, may be capable of restoring antibody responsiveness of FLV-suppressed spleen cells. In the present study it was found that such antibody response helper factor(s) present in serum from mice actively treated with endotoxin as well as a non-toxic polysaccharide (PS)-rich derivative, or cell-free culture supernatants from normal mouse spleen cells or macrophage cultures treated with these bacterial products, were capable of restoring antibody formation by FLVsuppressed spleen cells.

EXPERIMENTAL PROCEDURES

Experimental animals Inbred male Balb/c mice, 6 - 8 weeks of age, were obtained from Cumberland View Farms, Clinton, TN. The mice weighed 1 8 - 2 0 g at the start of an experiment. They were housed in groups of 6 - 10 in plastic mouse cages and fed Purina mouse pellets and water ad libitum. Leukemia virus and infection Mice were infected by intraperitoneal (i.p.) injection of 100 50070 infectious doses (IDs0) of FLV contained in 0.1 ml of a 10070 (w/v) clarified homogenate of infected spleen cells. The virus was maintained exactly as described previously by passage through adult Balb/c mice, and contained both the spleen focus forming and lymphatic leukemia virus components of the Friend complex (Friedman & Ceglowski, 1971a). The stock virus preparation did not have detectable lymphocytic choriomeningitis virus or lactic dehydrogenase virus. Antigen and immunization Sheep red blood cells (SRBC) in Alsevers' solution were obtained from BBL Microbiology System, Cockeysville, MD. The erythrocytes were washed several times in sterile saline and resuspended to a

concentration of 0.5°70 (v/v). For in vivo immunization, mice were injected i.p. with 0.5 ml saline containing 4 × 108 SRBC. In vitro immunization Covered plastic 24-well Linbro plates were used as culture chambers exactly as described previously (Hirano et al., 1971; Kateley et al., 1974). Spleen ceils from normal or FLV-infected mice were washed in RPMI 1640 medium containing 10°70 fetal calf serum and 100/ag each of penicillin G and streptomycin per ml. The numbers of viable nucleated ceils were determined by the trypan blue dye exclusion technique using a hemocytometer. A suspension of 8 × 106 viable spleen cells in 2.0 ml complete medium was added to each well of the Linbro plates and cultured for 5 days at 37°C in a humidified chamber containing 5070 CO2, 95070 air (Kamo, Pan & Friedman, 1976). For in vitro immunization 0.1 ml of a 0.1070 suspension of SRBC was added to each culture (i.e. 2 × 106 erythrocytes). Assay f o r antibody-forming cells The numbers of direct antibody plaque forming cells (PFC) to SRBC was determined by the standard micro-method exactly as described previously (Butler et al., 1983; Cunningham & Szenburg, 1968). The numbers of PFCs were enumerated for at least three to five cultures prepared from three to four spleen cell preparations and the average numbers of PFCs per 106 cells calculated. In all cases only direct nonfacilitated plaques were enumerated and these were considered due to 19S immunoglobulin M antibodyproducing cells. Lipopolysaccharide preparation Serratia marcescens LPS was prepared by the trichloroacetic acid extraction procedure as previously described (Nowotny, Behling & Chang, 1975). A non-toxic smaller mol. wt. (approximately 10,000 daltons) polysaccharide derivative, containing less than 0.1 070 lipid and less than 0.2°70 protein nitrogen or amino acids, was prepared by acid hydrolysis exactly as described (Nowotny et al., 1975). Post-treatment serum preparation Normal or FLV-infected mice were injected i.p. with a standard dose of Bacillus Calmette Guerin (BCG) (107 cfu/mouse) and then, 2 weeks later, injected i.p. with varying amounts of endotoxin or PS; the mice were exsanguinated 2 h later by aseptic

Enhancement of Antibody Response in FLV-infected Mice

285

Table 1. Effect of lipopolysaccharide or polysaccharide-rich derivative on antibody responses of normal and FLV-infected mice

In vivo

PFC/106 spleen cells*

treatment ~g/mouse)*

Normal mice

FLV-infected mice - 10 days - 2 0 days ~

None - LPS 10.0 50.0 100.0 PS 10.0 50.0 100.0

530±60 680±110 1350_+215* 1150± 175* 545±65 985_+130* 1040± 120*

185±43 210±62 580_+138* 650± 170* 160±38 465_+73* 510±62*

38±10 47+21 96± 19* 105±32' 61±18 138±32* 198_+42'

*Indicated dose of LPS or PS injected i.p. into groups of mice immunized at same time by i.p. injection with 4 x 10~ SRBC. tAverage PFC response _+ S.D., for three to four mice per group 5 days after challenge immunization with SRBC. *Statistically significant stimulation, calculated by Student's t-test, in comparison to control responses. ~Mice injected i.p. with 100 IDa0 FLV on day indicated before challenge immunization.

cardiac puncture. Serum was separated from the erythrocytes and kept at 4°C until used. In vitro factor production Suspensions o f 10 7 spleen cells/ml from normal donor mice were incubated in R P M I 1640 medium at 37°C in an atmosphere o f 5o7o CO2, 95o7o air, and either left untreated or treated with graded amounts of endotoxin or PS at the time of culture initiation. Supernatants were collected at various times thereafter by centrifugation o f the cultures and stored at 4°C or frozen at - 7 0 ° C until used.

Antibody response helper factor assay The presence of antibody response helper activity was determined by adding various amounts, usually 0.01 ml to 0.2 ml, of a post-treatment serum or 0.1 m of a cell-free culture supernatant to the in vitro antibody-producing cultures at the time of sensitization with SRBC. The helper factor activity of each preparation was considered proportional to the degree o f enhancement of the antibody response over that o f normal untreated control cultures (Butler et al., 1978, 1983).

RESULTS

Studies in this laboratory had shown that spleen cells from FLV-infected mice were strongly deficient in their antibody responsiveness to sheep

erythrocytes. As shown in Table 1, injection of normal or FLV-infected mice with graded amounts o f either endotoxin or its polysaccharide derivative markedly affected the subsequent immune response to SRBC. A concentration o f 5 0 - 100/ag of either the endotoxin or PS derivative enhanced the P F C response of normal mice upon challenge immunization with RBC. Mice infected with FLV 1 0 - 20 days earlier showed a progressive decrease of antibody responsiveness. However, when such mice were injected with endotoxin or PS, ranging from 10 to 100/ag per mouse at the time of challenge immunization with SRBC, there was a p r o f o u n d increase in antibody responses, especially with the higher doses of either the endotoxin or PS (Table 1). It is noteworthy, however, that mice infected with FLV for the shorter period o f time (i.e. 10 days earlier) and showing approximately a 6 0 - 7 0 % decrease in antibody responsiveness as compared to normal controls, had the greatest stimulation by LPS or PS. When mice were infected for 20 days with FLV, their antibody response was more diminished (generally 90-950/0) and endotoxin or PS had less restorative effect on their anti-SRBC responsiveness with regard to total number of PFC. Nevertheless, as compared to untreated mice, there was still a 3 - 4 fold increase in the number of P F C s after treatment. Pretreatment of normal mice with BCG, followed 2 weeks later by an injection with endotoxin or PS, resulted in post-treatment sera with strong antibody

286

H. FRIEDMANel al. Table 2. Effect of sera from BCG and/or endotoxin-treated donor mice on antibody response of spleen cells from normal or FLVinfected mice Donor mouse pretreatment*

Serum volume* (ml)

None (control) BCG only

0.2 0.01 0.1 0.2 0.01 0.1 0.01 0.1 0.2 0.2

BCG + LPS BCG + PS LPS only PS only

PFC/106 spleen cells* Normal mice FLV-infectedH 640±93 658±82 710±110 570+_72 1340±1958 1490± 1728 1165±1238 1290± 1208 735±112 682±96

143±62 198±71 168±92 190±62 425±608 530_+82~ 625±1158 650+_90~ 210±80 197±56

*Groups of donor mice injected i.p. with BCG 2 weeks earlier and then treated i.p. with 10.0/ag LPS or PS 2 h earlier. *Indicated volume of serum added to cultures of 8 × 106 spleen cells at time of immunization with 2 × 106 SRBC in vitro. *Average number of PFC ± S.D. for three to six cultures per group 5 days after in vitro immunization and culture initiation. 8Statistically significant stimulation, calculated by Student's t-test, in comparison to response of control cultures. ~Spleen ceils obtained from mice injected I 0 - 12 days earlier with I00 ID~0FLV. Table 3. Effect of LPS or PS post-treatment sera on antibody response in FLV-infected mice Donor mouse treatment* ~g/mouse) LPS PS

10.0 50.0 10.0 50.0

1h 210±43' 280±97 196±40 230±72

PFC/106 spleen cells* 2- 4 h 8 - 13 h 580±130 650±110 628±90 590±115

320±60 350±95 180±40 220±68

24 - 48 h 158±73 210±80 240±76 --

*Groups of mice injected i.p. with BCG 2 weeks earlier and then pretreated with either 10 or 50/ag LPS or PS i.p. at indicated times from 1 - 4 8 h before collection; 0.1 ml volume of serum added to cultures of 8 × 106 spleen cells from mice infected 12 days earlier with 100 IDa0 FLV. *Average PFC response ± S.D. for l07 spleen cells from three to five mice per group 5 days after in vitro immunization with 2 × 106 SRBC. *Control mice (FLV-infected but normal mouse serum added) had less than 100 PFC/106 cells. response helper activity (Butler et al., 1983). As is evident f r o m Table 2, the p o s t - t r e a t m e n t sera h a d an enhancing effect on the a n t i b o d y responsiveness o f spleen ceils f r o m FLV-infected mice. C o n t r o l sera f r o m mice given only BCG 2 weeks earlier, or only L P S or PS 2 h earlier had little effect o n the a n t i b o d y responsiveness o f spleen cells f r o m either n o r m a l or

FLV-infected mice. In contrast, sera o b t a i n e d 2 h after administration o f e n d o t o x i n or the PS to BCG pretreated mice h a d m a r k e d i m m u n o e n h a n c i n g effects on the in vitro a n t i b o d y response o f spleen cells f r o m b o t h n o r m a l and FLV-infected mice. A l t h o u g h the total n u m b e r o f P F C s in the cultures o f spleen cells f r o m infected mice did not reach the

287

Enhancement of Antibody Response in FLV-infected Mice Serum Specimen

c---,--o a,...ma &.....~

500

Culture Supernst ant

Control Post-Endotoxin PolII-PS

600

I s.o.

]

Endotoxin Stimulated

]

,s S,,mu,.,.~

~/~. 500

400 .J J tU 0

Z uJ uJ Q. u)

--J u,J 0 Z uJ uJ

I" I:" F I" L: /:

300

400

300 ,-xx~

200

,.o 200

,'%.~

Q.

100 100

2 DAY

4 OF IN V/TRO

6 CULTURE

Fig. 1. Effect of serum from BCG and post-LPS-or PStreated mice on cytokinetics of antibody responses of spleen cells from mice infected i.p. 12 days earlier with I00 IDs0 FLV. Each point represents average PFC response _ S.D. for at least five mice per group, tested on indicated day after in vitro immunization of 8 × 106 spleen cells with 2 x 106 SRBC. levels attained by spleen cells from normal animals, enhancement was still evident and approached levels of responsiveness of normal untreated cells with the PS (Table 2). It is evident, however, that sera from both endotoxin- and PS-treated donor mice were generally similar in their immunoenhancing activity. Optimum antibody response helper activity was present in sera collected 2 - 4 h after treatment of the mice, regardless of the bacterial product or concentration used (Table 3). There was much less antibody enhancing activity in 1 h post-treatment sera, suggesting that possible endotoxin or PS carry over in the sera was not responsible for the enhanced activity. Antibody response helper activity decreased markedly by 1 2 - 18 h after injection of donors with the bacterial products and was essentially nonexistent by 2 4 - 48 h. The post-endotoxin and post-PS treatment sera from BCG-infected mice enhanced the antibody response of FLV infected spleen cells throughout the entire culture period in vitro, indicating that an increase in peak antibody response was not due to a shift of the peak day of the response (Fig. 1). Not only was the total number of PFCs on the peak day of the response markedly increased, but the number

+2

+3

+4

+5

+6

D A Y OF C U L T U R E

Fig. 2. Enhancement of antibody response to SRBC by spleen cells from FLV injected mice incubated in vitro with cell-free culture supernatants from P388D~ cell cultures treated in vitro with 50/~gLPS or PS. Each point represents PFC response _+ S.D. for three to six cultures of 8 × 106 spleen cells for 20 day infected mice on indicated day after in vitro immunization with 2 × 106 SRBC and treated with indicated supernatant. Untreated spleen cells from FLV infected mice as controls had fewer than 100 PFC per 106 spleen cells tested. of PFCs on days 3 through 6 of assay were generally increased also as compared to untreated cultures. Similar enhancement was observed in regards to the cytokinetics of the PFC response potentiated in vitro by the cell-free culture supernatants from the macrophage cell line P388D1, stimulated with LPS or PS (Fig. 2). It is evident from Table 4 that both endotoxin and the PS, when added directly to spleen cells from either normal or FLV-infected mice, resulted in a dose-dependent enhancement of the antibody response, indicating that these materials were directly stimulatory not only for normal mouse spleen cells, but also for spleen cells derived from the leukemic mice. Optimal enhancement occurred with the I0/ag dose of either bacterial product, regardless of whether the responding spleen cells were derived from normal or virus-infected animals. Less effect occurred with 1.0/~g or even with a 20/~g dose of the bacterial product. Previous studies had shown that treatment of FLV-infected spleen cells with LPS in vitro failed to induce antibody helper activity (Butler

288

H . FRIEDMAN el al.

Table 4. Dose-response for stimulation of the in vitro antibody response of normal or FLVinfected spleen cells to sheep RBCs by lipopolysaccharide or polysaccharide derivative

In vitro

addition* 0ag/culture) None LPS

PS

(control) 0.1 1.0 10.0 20.0 0.1 1.0 10.0 20.0

PFC response per 106 spleen cells* Normal mice FLV-infected 070 of 070 of Response control P Response control 685_+47 710__.53

930+_72 1750+_130 1590+_145 695_+72 810_+98 1530+_140 1450_+135

-104 136 256 232 102 118 224 112

-NS 0.05 0.01 0.01 NS NS 0.01 0.01

146_+40 158_+53 320_+97 565_+150 480+_92 195_+43 275_+130 620_+165 510_+94

-108 219 387 329 134 188 425 349

-NS 0.01 0.01 0.01 NS 0.01 0.01 0.01

*Indicated amount of Serratia LPS or PS added to cultures of 8 x 106 spleen ceils at time of culture initiation and in vitro immunization with 2 x 10 6 SRBC. *Average number of PFC _+ S.D., for three to five cultures per group 5 days after in vitro immunization. *Donor mice infected by i.p. injection of 100 IDs0 FLV 12 days earlier.

Table 5. Antibody response by spleen cells from FLV-infected mice incubated in vitro with culture supernatants from LPS or PS pre-treated donor mice

Culture treatment* None (control) Non-treated LPS-treated PS-treated

Day of culture supernatant +

PFC/106 spleen ceils

-+ 3 days + 5 days + 3 days + 5 days + 3 days + 5 days

180_+32 190_+65 215_+43 365_+75 595___110 298_+42 520_+95

Antibody response* 070 of control __ 106 119 203 331 166 289

P __ NS NS 0.01 0.01 0.05 0.01

*Cultures of 1 z 107 normal mouse spleen cells incubated in vitro with 50 lag LPS or PS for 3 - 5 days. *0.1 ml of supernatants obtained 3 or 5 days after in vitro stimulation added to cultures of 8 x 106 spleen cells from FLV-infected (100 IDs0 12 days earlier) mice immunized in vitro with 2 x 106 SRBC. *Average PFC response __. S.D., for three to six cultures per group 5 days after in vitro immunization.

& F r i e d m a n , 1980). Thus, it was o f interest to determine whether FLV-infected splenocytes could r e s p o n d to a n t i b o d y response helper activity f r o m n o r m a l m o u s e spleen cell cultures treated in vitro with either L P S or the non-toxic PS. As is evident in Table 5, n o r m a l spleen cell cultures, when incubated in vitro with e n d o t o x i n or PS for 3 - 5 days, resulted in d e v e l o p m e n t o f a n t i b o d y response helper activity in the s u p e r n a t a n t s as s h o w n by the subsequent e n h a n c e m e n t o f the P F C response o f spleen cells f r o m FLV i m m u n o s u p p r e s s e d animals (Table 5).

Greater activity was n o t e d for 5 day culture supernatants as c o m p a r e d to 3 day supernatants, indicating again that it is unlikely that carryover o f the bacterial p r o d u c t in the supernatants was affecting the response o f the treated cell cultures. It would be expected that greater quantities o f endotoxin or PS would be present in the 3 day culture supernatants as c o m p a r e d to 5 day supernatants. Previous studies had suggested that endotoxin induces antibody helper factor(s) mainly by activating macrophages. As shown in Table 6, glass-

E n h a n c e m e n t of Antibody Response in FLV-infected Mice

289

Table 6. Effects of cell-free supernatants from spleen or PE cell cultures treated in vitro with LPS or PS on antibody response of FLV-infected spleen cells to sheep RBCs Cell culture supernatants* None Spleen cells Unfractionated

Adherent

Non-adherent

PE cells ~

PFC/106 spleen cells*

In vitro treatment*

°7o of control response

P

114 212 257 121 247 236 98 113 124 118 279 269

NS 0.01 0.001 NS 0.001 0.001 NS NS NS NS 0.001 0.001

174_+43 None LPS PS None LPS PS None LPS PS None LPS PS

198_+62 385_+56 447__.94 210_+37 430-+20 410_+72 170_+72 196_+74 215_+85 205_+72 485_+ 110 465_+73

*Cell-free culture supernatants from 2 × 1 0 7 spleen cells, unfractionated or separated into glass-adherent vs nonadherent cells, or PE cells, after in vitro incubation with LPS or PS, added in 0.1 ml volumes to cultures of 8 x 106 spleen cells from mice injected 12 days earlier with 100 IDs0 FLV. tCultures incubated with 50 Mg LPS or PS for 5 days before obtaining cell-free supernatants. *Average number of PFC _+ S.D. for three to six cultures per group 5 days after in vitro i m m m u n i z a t i o n with 2 x l 0 6 SRBC. ~Peritoneal exudate cells from mice injected 2 - 3 days earlier with 1.0 ml 5°70 dextran. Table 7. Effect of absorption of post-LPS or PS treatment culture supernatants on antibody formation by spleen cell cultures from FLV-infected mice

Culture treatment None (control) Normal supernatant Unabsorbed Absorbed with thymocytes Absorbed with B.M. cells Post-LPS supernatant Unabsorbed Absorbed with thymocytes Absorbed with B.M. cells Post-PS supernatant Unabsorbed Absorbed with thymocytes Absorbed with B.M. cells

PFC/106 spleen cells*

07o of control response

230_+48

__

__

246+63 197_+43 210_+38

107 86 92

NS NS NS

560_+92 483_+72 215_+62

244 210 94

0.001 0.001 NS

493-+60 516_+42 240_+68

214 224 105

0,001 0.001 NS

*Indicated supernatants, obtained 5 days after in vitro treatment of 2 × 107 normal spleen cells in vitro with 50 Mg LPS or PS, with or without absorption, added in 0.1 ml volumes to 8 x 106 spleen cells from mice infected 12 days earlier with 100 IDs0 FLV; supernatants absorbed for 30 rain at 25°C with 107 thymocytes or bone marrow cells from normal syngeneic mice. tAverage P F C response _+ S.D. for three to six cultures per group 5 days after in vitro immunization with 2 × 106 SRBC.

adherent spleen cell preparations, rich in macrophages, were a much better source for the antibody response helper activity for FLV

s u p p r e s s e d s p l e e n cells t h a n c o m p a r a b l e n o n adherent leukocyte populations which consisted mainly of lymphocytes. For these experiments,

290

H. FRIEDMANel al.

supernatants from unfractionated spleen cells, as controls, induced marked enhancement when the cells were stimulated in vitro for 5 days with endotoxin or PS. The supernatants from the cell fractions obtained by adherence to glass dishes resulted in similar enhanced antibody helper activity after the cells were incubated with endotoxin or PS. Similarly, dextran elicited peritoneal exudate (PE) cell suspensions from normal mice, when incubated with endotoxin or PS, resulted in similar antibody response helper activity in cell-free culture supernatants for FLV-infected spleen cells (Table 6). Since the PE cell population consisted mainly of macrophages, it seemed likely that these cells were the main, if not exclusive source of the antibody response helper activity stimulated by the endotoxin or PS. To further substantiate that macrophages were the source of the immunoenhancing factor, the continuous macrophage cell line P388D~ was cultured in vitro for up to 6 days with 10/ag LPS or PS. The cell-free supernatants obtained by centrifugation of the cultures were found to have marked immunoenhancing activity for spleen ceils from FLV-infected mice (Fig. 2). The greatest enhancement occurred with cell culture supernatants obtained 4 - 6 days after culture initiation, again suggesting that possible carryover of bacterial product into the culture supernatant was not the cause of enhancement, since it would be expected that similar or larger amounts of LPS or PS would be present in the culture supernatants early after culture initiation. The target of the antibody helper activity in the culture supernatants appeared to be bone marrow cells, probably B-cells or their direct precursors, since antibody helper activity could be absorbed out of the active supernatants only with suspensions of bone marrow cells. As is evident in Table 7, supernatants from endotoxin- or PS-treated cultures lost their antibody enhancing activity when absorbed with suspensions of bone marrow cells derived from normal syngeneic mice but not with similar numbers of thymocytes from the same animals. DISCUSSION The data presented clearly demonstrate that a nontoxic polysaccharide derivative of Serratia marcescens is capable of stimulating macrophages to produce soluble factor(s) which enhance antibody formation to SRBC by splenic lymphocytes from immunosuppressed FLV-infected mice. This extends earlier studies which demonstrated that PSstimulated antibody response helper factor(s) boost

PFC responses in normal mice and that LPS-induced factors from normal mice could boost PFC responses in FLV-infected mice (Butler & Friedman, 1980; Butler et al., 1978). That a non-toxic derivative of endotoxin retains the immunopotentiating activity of LPS has significant implications for the therapeutic potential of this PS in immunodeficiency or immune suppressed individuals. The PS induction of antibody response helper activity was achieved both in vivo and in vitro. The observation that cells cultured with PS for 3 - 5 days yielded supernatants with maximum activity, argues against the likelihood that the fluid is contaminated with the PS. Certainly earlier times would yield fluids with as much or more PS. In addition, PS added to culture supernatants from unstimulated mouse spleen cell culture does not display antibody helper activity. Thus, contamination of fluids by PS does not seem to be a significant source of antibody response helper activity. It seems plausible that suppression of antibody response helper factor(s) production in vitro may be due to a decrease in the number or activity level of macrophages in the leukemic spleens. Macrophages from leukemic mice appear to be deficient in some activities but they may show enhanced suppressor cell function. Macrophages, after FLV infection, lose their ability to migrate normally in vivo and thus may not be present in sufficient quantities in vitro to respond to an antigen (Bendinelli et al., 1985). Therefore, it seems likely that FLV infection could result in alteration of macrophage populations normally capable of producing antibody helper factors in response to antigen or immunologic stimulation and thus cannot serve as functional accessory ceils. It is apparent from the present studies that the factor(s) is produced by macrophages. Splenic adherent cells and peritoneal exudate cells isolated from mice, as well as the P388D~ cell line, all of which are comprised mainly of macrophages, produce the antibody response helper activity. Lymphocyte-enriched populations did not produce this activity. It is of interest that, similar to intact endotoxins, the non-toxic polysaccharide was shown in this laboratory to induce murine spleen cells from normal mice to produce interleukin 1 (IL-1), interferons, tumor necrotizing factor (TNF), as well as colonystimulating factor (CSF) (in preparation). The nature of the antibody response helper factor(s) presently is unknown. However, other studies in these laboratories suggest that the factor(s) are not likely to be IL-I or IL-2 (Lopez-Cepero, Specter, Friedman &

Enhancement of Antibody Response in FLV-infected Mice Bendinelli, 1987). In these studies, FLV mice were noted to produce normal or slightly increased levels of IL-I, and thymocytes from FLV-infected mice responded normally to exogenous IL-I. Conversely, FLV-infected mice lose the ability to produce IL-2 and lymphocytes from these mice do not respond to exogenous IL-2. Since PS does stimulate antibody response helper factor(s) in normal but not FLV-infected mice, it is unlikely to be IL-1, Also, FLV-infected mice respond to antibody response helper factor(s) but not to exogenous IL-2 so the helper activity is unlikely to be IL-2. This poses the question of what does comprise the antibody response helper factor(s)? It is possible that the post-endotoxin serum, as well as the cell-free supernatants of spleen ceils from normal mice treated with endotoxin or the polysaccharide, contains interferon, TNF, CSF and other unidentified factors. There have been no reported studies to date indicating that any of these factors, including purified as well as recombinant DNAproduced IL-1, interferons, TNF or CSF, have the ability to increase antibody formation by spleen cells from leukemia virus immmunosuppressed mice. However, Prat and co-workers have reported that I F N - y is important in antitumor immunity in routine retrovirus-induced sarcomas (Prat, Bretti, Amedeo, Landolfo & Conoglio, 1987). Furthermore,

291

I F N - y has been reported to stimulate antibody formation in normal mice (Sidman, Marshall, Schultz, Grey & Johnson, 1984). More interestingly, Johnson, Somers & Hamilton (1987) have reported that LPS-stimulated mouse macrophages in vitro produce at least seven new polypeptides ranging from 38,000 to 85,000 daltons, and that I F N - y induces three polypeptides, ranging from 20,000 to 90,000 daltons. In addition, when LPS and I F N - y are used together, another polypeptide of 120,000 daltons is produced. This latter polypeptide is associated with antitumor activity. Furthermore, these authors indicate the BCG can induce macrophages to produce the 120,000 dalton polypeptide in vivo. Thus, the antibody response helper factor(s) stimulated in this study may involve one or several of these polypeptides described by Johnson and co-workers. Nevertheless, the present study shows that not only endotoxin, but also the non-toxic PS has direct immunostimulatory activities for cultures of spleen cells from virus immunosuppressed mice. Studies are in progress to characterize this factor(s) present in sera and culture supernatants which enhances antibody production by leukemic spleen cells and to determine whether such activity can be related to known mediators induced by endotoxins or the polypeptides described by Johnston et al., (1987).

REFERENCES

BENDINELLI,M. & FRIEDMAN,H. (1980). B and T lymphocyte activation by murine leukemia virus infection. Adv. exp. Biol. Med., 121B, 91- 97. BENDINELLI,M., MATTEUCCI,P. & FRIEDMAN,H. (1986). Retrovirus-induced acquired immunodeficiencies. Adv. Cancer Res., 45, 125 - 181. BUTLER,R. C. ~,~FRIEDMAN,H. (1979). Leukemia virus-induced immunosuppression reversed by subcellular factors. Ann. N. Y, Acad. Sci., 352, 445-450. BUTLER, R. C. & FRIEDMAN,H. (1980). Restoration of leukemia cell immune responses by bacterial products. In Current Chemotherapy and Infectious Disease (eds Nelson, J. D. and Grassi, C.) pp. 1719- 1920. American Society of Microbiology, Washington, DC. BUTLER, R. C., FRIER, J. M., CHAPEKAR,M. S., GRAHAM,M. O. & FRIEDMAN,H. (1983). Role of antibody response helper factors in immunosuppressive effects of Friend leukemia virus. Infec. Immun., 39, 1260- 1264. BUTLER, R. C., NOWOTNY,A. & FRIEDMAN,H. 0978). Macrophage factors that enhance the antibody response. Ann. N. Y. Acad. Sci., 333, 564 - 578. CUNNINGIqAM,A. J. & SZENBURG,A. (1968). Further improvements in the plaque technique for detecting single antibodyforming cells. Immunology, 14, 599-601. FRIEDMAN, H. & CEGLOWSKI,W. S. (1971a). Leukemia virus-induced immunosuppression VII1. Rapid depression of in vitro leukocyte migration after infection of mice with Friend leukemia virus. J. Immun., 107, 1673- 1681. FRIEDMAN,H. t~ CEGLOWSKI,W. S. (1971b). Leukemia virus-induced alteration of leukocyte migration in vitro. Nature, Lond., 233, 415-416. FRIEDMAN,I-I. & SPECTER,S. (1979). Interaction of leukemia virus with cells of the immune response system. Transplantn Procs, 11, 1060-1065.

292

H. FRIEDMANet al.

H IRANO,S., FRIEDMAN,H. • CEGLOWSKI,W. S. (1971). Immunosuppression by leukemia viruses. VII. Stimulatory effects of Friend leukemia virus on pre-existing antibody-forming cells to sheep erythrocytes and Escherichia coli in nonimmunized mice. J. l m m u n . , 107, 1400- 1409. JOHNSTON, P, A., SOMERS, S. D. t~ HAMILTON, T. A. (1987). Expression of a 120,000 dalton protein during tumoricidal activation in murine peritoneal macrophages. J. I m m u n . , 138, 2739-2744. KAMO, I., PAN, S. H. & FRIEDMAN, H. (1976). A simplified procedure for in vitro immunization of dispersed spleen cell cultures. J. I m m u n . Meths, 11, 55-62. KATELEY, J. R., KAMO, I., KAPLAN,G. & FRIEDMAN, H. (1974). Suppressive effect of leukemia virus-infected lymphoid cells on in vitro immunization of normal splenocytes. J. natn Cancer Inst., 53, 1371 - 1378. LOPEZ-CEPERO, M., SPECTER, S., FRIEDMAN, H. & BENDINELLI, M. (1987). Suppression of IL-2 production during retrovirus infection. Abstr, A m . Soc. Microbiol., 87, ll0. NOTK1NS, A. L., MERGENHAGEN,S. E. & HOWARD, R. J. (1970). Effect of virus infections on the function of the immune system. Ann. Rev. Microbiol., 24, 325- 338. NOWOTNY, A., BEHLING, U. H. & CHANG, H. L. (1975). Relation of structure to function in bacterial endotoxins. VIII. Biological activities in a polysaccharide-rich fraction. J. I m m u n . , 115, 197- 211. PRAT, M., BRETTI, S., AMEDEO, M., LANDOLEO, S. & CONOGLIO, P.M. (1987). Monoclonal antibodies against murine IFN gamma abrogate in vivo tumor immunity against RSV-induced murine sarcomas. J. I m m u n . , 138, 4530 - 4533. S1DMAN, C. L., MARSHALL,J. D., SCHULTZ, L. D., GREY, P. W. & JOHNSON, H. M. (1984). Gamma-interferon is one of the several direct B-cell maturing lymphokines. Nature, 309, 801 -804. SPECTER, S . C . , BENDINEELI, M., CEGLOWSKI, W . S . & FRIEDMAN, H. (1978). Macrophage induced reversal of immunosuppression by leukemia viruses. Fedn Proc. Fedn A m . Socs exp. Biol., 37, 9 7 - 101. SPECTER, S., PATEL, N. & FRIEDMAN, H. (1976). Restoration of leukemia virus-suppressed immunocytes in vitro by peritoneal exudate cells. Proc. Soc. exp. Biol. Med., 151, 163 - 167. -