BALB/c mice were purchased from Daehan Biolink Co. (Seoul, Korea). They were maintained on an 8:16-h light: dark cycle in an animal environmental control chamber (Daehan Biolink. Co., Korea). Animal care was in accordance with the institutional guidelines of Kangwon National University. Eight- to twelve-week-old female mice were used in this study.

Imject Alum (hereafter simply called alum) was from Pierce Biotechnology (Rockford, IL, USA). IL-4 and TGF-β1 were purchased from R&D Systems (Minneapolis, MN, USA). LPS (Escherichia coli O111:B4) was from Sigma Chemical Co. (St. Louis, MO, USA). TRIZOL reagent was purchased from Invitrogen Life Technologies (Carlsbad, CA, USA). The antibodies used in ELISA were purchased from Southern Biotechnology (Birmingham, AL, USA). Anti-mouse CD3ε Ab and anti-mouse CD28 Ab were purchased from BD Biosciences (San Jose, CA, USA).

The murine B cell lymphoma line, CH12F3-2A (surface µ+) (14) was provided by Dr. T. Honjo (Kyoto University, Japan). The cells were cultured in RPMI-1640 (100 U/ml penicillin; 100µg/ml streptomycin) plus 10% fetal bovine serum (HyClone Labs, Logan, UT, USA) in a humidified CO2 incubator. Mouse spleen B cell population was prepared as described before (15). A total of 2×10⁶ cells/well were cultured in flat-bottomed, 24-well tissue culture plates (SPL, Korea) in a volume of 2 ml complete medium or a total of 2×10⁵ cells/well were cultured in flat-bottomed, 96-well tissue culture plates in a volume of 200µl complete medium with added stimulants.

AID promoter reporter was given by Dr. M. Sugai (Center for Molecular Biology and Genetics, Kyoto University) (16). Transfection was performed by electroporation with a Gene Pulser II (Bio-Rad, USA) as described (15). Reporter plasmids were cotransfected with expression plasmids and pCMV-βgal (Stratagene), and luciferase and β-gal assays were performed as described (15).

Enzyme-linked immunosorbent assays (ELISAs) were performed as described previously (15). Ig-secreting cells were enumerated by ELISPOT assay as described (17).

RNA preparation, reverse transcription, and PCR were performed as described before (15). PCR primers were synthesized by Bioneer Corp. (Seoul, Korea): GLTγ1 sense, 5'-CAG CCT GGT GTC AAC TAG-3' and antisense, 5'-CTG TAC ATA TGC AAG GCT-3' (product size : 532 bp); GLTγ2a sense, 5'-GCT GAT GTA CCT ACC TGA GAG A-3' and antisense, 5'-GCT GGG CCA GGT GCT CGA GGT T-3' (product size : 394 bp); GLTγ2b sense, 5'-GGG AGA GCA CTG GGC CTT-3' and antisense, 5'-AGT CAC TGA CTC AGG GAA-3' (product size : 318 bp); GLTγ3 sense, 5'- CAA GTG GAT CTG AAC ACA-3' and antisense, 5'-GGC TCC ATA GTT CCA TT-3' (product size : 349 bp); GLTα sense, 5'-CAA GAA GGA GAA GGT GAT TCA G-3' and antisense, 5'-GAG CTG GTG GGA GTG TCA GTG-3' (product size : 206 bp); CTγ1 sense, 5'-CCA AAA CAG GAA CAG AGA CG-3' and antisense, 5'-GTT CCA GGT CAC TGT CA-3' (product size : 434 bp); AID sense, 5'-TGC TAC GTG GTG AAG AGG AG-3' and antisense, 5'-TCC CAG TCT GAG ATG TAG CG-3' (product size : 119 bp); IFN-γ sense, 5'-ACA CTG CAT CTT GGC TTT GC-3' and antisense, 5'-TGG ACC TGT GGG TTG TTG AC-3' (product size : 369 bp); T-bet sense, 5'-ACC AAC AAC AAG GGG GCT TC-3' and antisense, 5'-CTC TGG CTC TCC ATC ATT CAC C-3' (product size : 111 bp); IL-4 sense, 5'-ATA TCC ACG GAT GCG ACA AA-3' and antisense, 5'-AAG CCC GAA AGA GTC TCT GC-3' (product size : 252 bp); GATA-3 sense, 5'-ACA GAA GGC AGG GAG TGT GTG AAC-3' and antisense, 5'-TTT TAT GGT AGA GTC CGC AGG C-3' (product size : 118 bp); CXCR5 sense, 5'-GAC CTT CAA CCG TGC CTT TCT C-3' and antisense, 5'-GAA CTT GCC CTC AGT CTG TAA TCC-3' (product size : 139 bp); Foxp3 sense, 5'-CTT CAT GCA TCA GCT CTC CA-3' and antisense, 5'-AGA CTC CAT TTG CCA GCA GT-3' (product size : 293 bp); β-actin sense, 5'-CAT GTT TGA GAC CTT CAA CAC CCC-3' and antisense, 5'-GCC ATC TCC TGC TCG AAG TCT AG-3' (product size : 320 bp). PCR reactions for β-actin were performed in parallel in order to normalize cDNA concentrations within each set of samples. PCR products were separated on a 2% agarose gel and photographed.

Statistical differences between experimental groups were determined by analyses of variance, and values of p<0.05 by unpaired two-tailed Student's t-test were considered significant.

We first explore whether alum affects Ig production by LPS-stimulated mouse B cells. As shown in Fig. 1, alum consistently increased IgG1 isotype secretion in a dose-dependent manner. In this, though less potent, alum also increased secretion of IgG2b and IgG3 isotypes. Thus, potency of alum on IgG isotypes appeared to be IgG1>IgG3>IgG2b. In contrast, it little affected secretion of IgM, IgG2a and IgA isotypes. In the absence of LPS, alum by itself did not result in secretion of any isotypes (data not shown). These results clearly show that alum directly modulates B cells to produce IgG1 isotype under the influence of LPS, a mouse B cell polyclonal activator. This is a somewhat surprising finding since no any reports have demonstrated the direct effect of alum on Ig synthesis by B cells ever since its adjuvanticity was discovered (1).

Since alum markedly influenced B cells in the synthesis of Ig isotypes, it was necessary to determine its effect on B cell growth. In the absence of LPS, alum did not increase cell viability but alum at the high dose (120µg/ml) delayed cell death (Fig. 2). In contrast, its effect was marginal in the presence of LPS. These results indicate that the modulatory effect of alum on Ig synthesis is not associated with its potential effect on B cell growth.

IL-4 and TGF-β1 are well established switching factors for IgG1 and IgA isotypes, respectively (17,18). It was curious how alum affects production of IL-4-induced IgG1 and TGF-β1-induced IgA isotypes. As shown in Fig. 3, IL-4 increased IgG1 production more strongly than alum. However, IL-4 and alum in combination did not result in any additive effect on IgG1 production. Similary, TGF-β1-induced IgA production was not further augmented by treatment of alum. These results suggest that signalling pathway of IL-4 and alum toward IgG1 gene expression is identical or at least largely overlapping, of which mechanisms remains to be determined.

Because alum directly modulates B cells to differentiate mainly into IgG1 B cells, it was necessary to investigate if alum increases IgG1 production by increasing the total number of cells that secrete IgG1, or simply by increasing the amount of IgG1 secreted per cell. As shown in Fig. 4, there was an increase in the number of IgG1-secreting cells in alum-stimulated cultures. This was paralleled by an increase in the total amount of IgG1 secreted (Fig. 1). Here, IgG1-spot sizes between cultures of LPS alone and LPS plus alum were not different, indicating that alum does not increase the amount of IgG1 secreted per cell.

Increases in the number of IgG1-secreting cells in the above experiments is: 1) due to an increase in the frequency of B cells that switch to express the IgG1 isotype after stimulation with alum, or 2) due to increased proliferation of B cells that are already committed to express the IgG1 isotype. To distinguish between these two possibilities, we explored whether alum affects IgG1 class switch recombination (CSR). Transcription of unrearranged C_γ1 gene to produce germ-line transcript γ1 (GLT_γ1) precedes CSR to IgG1 (Fig. 5A). Once IgG1 CSR occurs, post-switch γ1 transcripts (PST_γ1) and γ1 circle transcript (CT_γ1) are also expressed. Therefore, expression of GLT_γ1, CT_γ1 and PST_γ1, can be used to indicate active IgG1 CSR (19-21). As shown in Fig. 5B, alum increased the expression of GLT_γ1 and CT_γ1 but not other GLTs. Since alum did not stimulate cell proliferation (Fig. 2), it is regarded that alum induces IgG1 production mainly through IgG1 CSR.

It is well established that AID is prerequisite for Ig CSR (19,22), it was important to ask if alum has effect on AID expression. As shown in Fig. 5C, alum did not increase both the expression of endogenous AID transcripts and the exogenous AID promoter activity, suggesting that alum does not cause general Ig CSR. Finally, we investigated whether alum directly regulate CD4 T cell differentiation. To assess this issue, we measured the transcriptional levels of canonical markers for each T cell subsets: IFN-γ and T-bet for Th1, IL-4 and GATA-3 for Th2, CXCR5 for follicular helper T cell, Foxp3 for regulatory T cell. We found no significant change in CD4 T cell subset population under the influence of alum (Fig. 6). Again, these results support our finding that alum by itself can act on B cells resulting in IgG1 CSR without any helper T cell subset including Th2. Then, what would be the mechanisms by which alum directs B cells for Ig regulation? In this regard, one needs to consider at least three possible ways which were the case in DCs. That is, alum may direct B cells 1) through release of DNA (23), 2) through forming of inflammasome (6,7), 3) through lipid rafts (24).

The present study demonstrates that alum can directly modulate B cells to commit IgG1 isotype switching. This implies that other immune cells such as dendritic cells, macrophages, and T helper cells are dispensable for the enhancement of IgG1 production by alum, though their involvement can not be formally excluded and to be elucidated in the future. Nevertheless, our findings of alum's direct effect on B cells would be useful to improve its vaccine adjuvanticity.