Jul 20, 1987 - KAREN L. ELKINS,* PHILIP W. STASHAK, AND PHILLIP J. BAKER. Laboratory of MicrobialImmunity, National Institute ofAllergy and Infectious ...
INFECTION AND IMMUNITY, Dec. 1987, p. 3085-3092
Vol. 55, No. 12
0019-9567/87/123085-08$02.00/0 Copyright © 1987, American Society for Microbiology
Prior Exposure to Subimmunogenic Amounts of Some Bacterial Lipopolysaccharides Induces Specific Immunological Unresponsiveness KAREN L. ELKINS,* PHILIP W. STASHAK, AND PHILLIP J. BAKER Laboratory of Microbial Immunity, National Institute of Allergy and Infectious Diseases, Bethesda, Maryland 20892 Received 20 July 1987/Accepted 15 September 1987
Pretreatment (priming) of BALB/c mice with a low (subimmunogenic) dose of Escherichia coli 0113 lipopolysaccharide (LPS) generates immunological memory 7 to 30 days later; the direct (immunoglobulin M) plaque-forming cell (PFC) responses produced after subsequent immunization with an optimal dose are 4 to 20 times greater than those of unprimed mice. By contrast, priming with a low dose of E. coli 055 LPS, followed by immunization with an optimally immunogenic dose 2 to 30 days later, resulted in a significantly reduced antibody response. Similar results were obtained with Serratia marcescens LPS. Dose-response studies indicated that such unresponsiveness is antigen specific and could be induced with subimmunogenic amounts of LPS. Priming reduced the magnitude of the PFC response to all immunizing doses of LPS tested. Unresponsiveness is not due to (i) an alteration in the time course of the PFC response or to (ii) a change in the isotype of the anti-LPS antibody produced after priming and immunization.
Prior treatment (priming) of BALB/c mice with a subimmunogenic dose (5 ng) of lipopolysaccharide (LPS) derived from Escherichia coli 0113, followed by immunization 7 to 30 days later with an optimally immunogenic dose (10 ,ug) of E. coli 0113 LPS, regularly results in the development of direct (immunoglobulin M [IgM]) plaque-forming cell (PFC) responses that are 4 to 20 times greater than those of unprimed, immunized mice (8, 15, 35, 43, 44). Although such immunological memory can be generated in mice primed with either the complete E. coli 0113 LPS molecule or the nonmitogenic 0 polysaccharide portion of the LPS molecule (native protoplasmic polysaccharide or NPP), only the complete LPS molecule can be used for immunization to elicit the expression of memory (43). Since memory cannot be generated in LPS-defective C3H/HeJ mice (35), recognition of the lipid A portion of E. coli 0113 LPS appears to play an important role in the expression of memory. Memory can be generated in nulnu mice, suggesting that the development of memory is T cell independent (43). It develops in a cyclic fashion, the pattern of which is characteristic of the strain of mice used (8). Priming and subsequent immunization do not result in the generation of a large IgG or detectable IgA PFC or serum antibody response to E. coli 0113 LPS (8). LPS is a unique antigen in that most of the specific antibody response is directed against the 0-antigen polysaccharide determinants of the molecule, whereas the lipid A moiety can elicit a wide variety of nonspecific immunological and pharmacological effects (24, 39; reviewed in reference 32). Thus, it is possible that the generation and expression of memory to the 0 polysaccharides of LPS is a common feature of most preparations of LPS and involves recognition of the mitogenic lipid A. In the present work, we examined the ability of E. coli 055 LPS and Serratia marcescens LPS to generate memory. In contrast to the results obtained with E. coli 0113 LPS, priming with a single injection of a subimmunogenic dose of these preparations of LPS induced the development of antigen-specific immunological unresponsiveness, rather than memory. *
MATERIALS AND METHODS Mice. Female BALB/cByJ mice were purchased from Jackson Laboratories (Bar Harbor, Maine); they were 8 to 12 weeks of age at the time of use. Antigens. All preparations of LPS used were extracted from cell walls of bacteria by the phenol-water procedure (28, 34). E. coli 055 LPS (lots RIR-326, RIR-342, and 026101) and S. marcescens LPS (lots RIR-322 and RIR-332) were purchased from Ribi ImmunoChem (Hamilton, Mont.). These preparations of LPS appeared to lack lipid Aassociated protein, since they were not mitogenic in vitro for C3H/HeJ lymphocytes and were not immunogenic in vivo for C3H/HeJ mice; no protein could be detected by UV spectroscopy or by chemical (Lowry) analysis (data not shown). All solutions of LPS and other reagents used in vivo in this study were prepared in sterile, pyrogen-free saline (Abbott Laboratories, Chicago, Ill.). Mice were immunized intraperitoneally (i.p.) with doses of LPS indicated in the text. Detection of antibody-producing cells. Antibody-producing PFC secreting IgM antibody specific for LPS were detected by a slide version of the technique of localized hemolysisin-gel (9, 10). Sheep erythrocytes were sensitized with LPS by the chromic chloride coupling method (10), using 1 mg of LPS per ml to sensitize 0.5 ml of packed cells. The results obtained for individual mice (PFC per spleen) were corrected by subtraction for small numbers of sheep erythrocyte-specific PFC detected, so that only values for LPSspecific PFC are considered in this report. The specificity of PFC detected was affirmed by plaque inhibition tests in which 10 ,g of an appropriate or inappropriate (serologically different) preparation of LPS was added to each slide of the agarose reaction mixture before incubation and the subsequent development of PFC. Since values for PFC per spleen are log normally distributed (18), geometric means of the values for groups of similarly treated mice were calculated. The data are expressed as the mean log1o LPS-specific PFC per spleen + standard error of the mean for groups of 5 to 10 mice, or as the antilog (actual numbers of PFC) of the mean.
Corresponding author. 3085
3086
INFECT. IMMUN.
ELKINS ET AL.
a
b (167.163)
(2,1251
ICJ)
c 0
0
C.)
0.
CL
U-
3
C7,
40
0
.0
0.01
0.1
1.0
10.0
Ag/mouse E. coli 055 LPS (i.p.)
-0
0.01
jg/mouse S.
0.1
1.0
marcescens
10.0 LPS (i.p.)
FIG. 1. Dose-response relationships for primary PFC responses to E. coli 055 LPS or S. marcescens LPS. BALB/cByJ female mice were immunized (i.p.) with the indicated amount of E. coli 055 LPS or S. marcescens LPS. E. coli 055 LPS-specific PFC (a) and S. marcescens LPS-specific PFC (b) were determined 4 days after immunization. Mean log1o LPS-specific PFC per spleen standard error of the mean (error bars) is shown for groups of five mice; geometric means are shown in parentheses. ±
Student's t test was used to evaluate the significance of the differences observed (38). Differences were considered to be significant when P values of 0.05) the magnitude of the PFC response to 1 ,ig of S. marcescens LPS (data not shown), priming with higher doses (500 pg, 5 ng, and 50 ng) significantly reduced the magnitude of the PFC response upon subsequent immunization with 1 jig of S. marcescens LPS. Therefore, 5 ng of
VOL. 55, 1987
INDUCTION OF SPECIFIC UNRESPONSIVENESS TO LPS
3087
TABLE 1. Priming with a subimmunogenic dose of E. coli 055 or S. marcescens LPS decreases the magnitude of PFC response to an optimally immunogenic dose of LPS' Mean
Dose
Prepn E. coli 055 LPS
S. marcescens LPS
logl0
LPS-
specific PFC/spleenb
Priming
Immunization
1 ng
Saline
2.890 ± 0.138
Saline
1 ,ug
1 ng
1 ,ug
5.099 ± 0.070 (125,523) 4.774 ± 0.092
±SEM (geometric mean)
Decrease in DeCreas response"
(777) 0
53
