Differential Involvement of the Transcription Factor Blimp-1 in T Cell-Independent and -Dependent B Cell Differentiation to Plasma Cells

Mice, cell cultures, and proliferation assays

BALB/c and C3H/HeJ mice, 8–12 wk old, were maintained in the specific pathogen-free animal facilities of the Instituto de Salud Carlos III. Total splenocytes and purified B cells (see below) were cultured at 10⁶ cells/ml in RPMI 1640 medium (Life Technologies, Paisley, U.K.) supplemented with 10% FCS (Labtech, Uckfield, U.K.), 2 mM l-glutamine, 1 mM pyruvate, 50 mM 2-ME, 10 mM HEPES, and antibiotics. Cultures were maintained at 37°C and 5% CO₂ in a humidified atmosphere. Escherichia coli LPS (Difco, Detroit, MI) was added to 25 μg/ml. The SR.D10 cell line is an autoreactive variant of the CD4⁺ Th2 cell line D10.G4.1, known to secrete IL-4, IL-6, and IL-10 (9, 10), and it was maintained in vitro by coculture with mitomycin C-treated, syngeneic C3H/HeJ feeder splenocytes in Click’s medium supplemented with 10% FCS and 100 μg/ml conalbumin (Sigma Chemicals, St. Louis, MO). For Th cell assays, SR.D10 cells were treated with mitomycin C and cocultured with C3H/HeJ-purified spleen B cells (cell ratio of 1 SR.D10 vs 100 splenic B cells). Cultures with anti-CD40 FGK45.5 mAb (10 μg/ml) and mouse rIL-4 (100 U/ml) were set up as described before (11). Cell proliferation was measured after a 12-h pulse of 1 μCi [³H]thymidine, and incorporated radioactivity was quantified by scintillation counting (Wallac LKB, Turku, Finland).

Flow cytometry analysis and sorting

Fluoresceinated anti-leukosialin S7 (anti-CD43) (12) and biotinylated anti-Synd-1 (13) mAb were purchased from PharMingen (San Diego, CA). Anti-IgM (331.12) (14) mAb was purified from culture supernatants by affinity chromatography on protein G columns (Pharmacia, Uppsala, Sweden) and fluoresceinated by standard methods. Phycoerythrin-conjugated Streptavidin was from Southern Biotechnology Associates (Birmingham, AL). Two-color stainings were performed as previously described (15). Dead cells were excluded on the basis of forward- and side-light scatter parameters, and propidium iodide staining. Specific mAb signals were defined against the fluorescence provided by isotype-matched, irrelevant mAb. Flow cytometry analyses were performed on an EPICS-XL flow cytometer (Coulter Electronics, Hialeah, FL). Cell sorting of splenic IgM⁺ B cells and cultured IgM⁺Synd⁺ cells were undertaken in a FACStar (Becton Dickinson, Mountain View, CA), and reanalyses of the purified cell samples were done in the EPICS-XL analyzer, to control the degree of purity. The frequencies of contaminating cells in purified samples were <5% after flow cytometry reanalysis.

ELISPOT assays

Ninety-six-well ELISA plates (Nunc, Kamstrup, Denmark) were precoated with either goat anti-mouse total Ig or rat anti-mouse IgG1 mAb (Southern Biotechnology Associates) for detection of IgM- and IgG1-secreting cells, respectively (3 μg/ml), and blocked with 1% gelatin/PBS. Cells were removed from cultures, washed three times, and counted. Serially diluted cells were seeded in triplicate in 1% FCS/RPMI 1640 and incubated at 37°C overnight in the coated plates. After washing, IgM- and IgG1-secreting cells were detected by incubation with biotinylated goat anti-mouse IgM (Southern Biotechnology Associates) or rat anti-mouse κ-chain mAb (187.1) (16), respectively. The plates were washed again and incubated with Streptavidin-conjugated alkaline phosphatase (Southern Biotechnology Associates). Plates were developed by adding 5-bromo-4-chloro-3-indolyl-phosphate (Sigma) in alkaline buffer containing 1% low melt agarose (Sigma) and incubated at 37°C. Developed spots were counted under a dissecting microscope.

RT-PCR analyses

Total RNA was isolated and cDNA was prepared as described (17), using 1 μg oligo(dT) as primer and avian myeloblastosis virus reverse transcriptase (Promega, Madison, WI). Equal amounts of cDNA were used to amplify Blimp-1, BSAP, and β-actin transcripts. Enzymatic amplifications were performed as follows: initial denaturation (95°C, 5 min), incubation (80°C, 5 min), addition of 2.5 U of Taq DNA polymerase (DNAzyme, Espoo, Finland), and then 30 cycles of the following steps: denaturation (95°C, 1 min), annealing (60°C, 1 min), and elongation (72°C, 1 min). The final polymerization step was extended for 10 min more. Primers for β-actin are described (15); for Blimp-1 and BSAP amplifications, the following primers were designed: Blimp-1 sense: 5′-TCAGCCTCTTCCCTAGGTTGTATC-3′; Blimp-1 reverse: 5′-AATCTTAAGGATCCATCGGTTCAAC-3′; BSAP sense: 5′-CAGCATAGTGTCTACAGGCT-3′; BSAP reverse: 5′-GTCTCGGCCTGTGAAATAGG-3′.

The amplification product (one-fifth of each) was separated electrophoretically on 1.5% agarose gels and then transferred to Zeta-probe membranes (Bio-Rad, Hercules, CA) after treatment with 0.4 M NaOH. Hybridization was performed with β-actin and Blimp-1 probes, ³²P-labeled by random priming (18). The β-actin insert was the PstI fragment of the plasmid p41AL (19). The Blimp-1 insert was obtained after cloning the 1178-bp fragment of the PCR-amplified Blimp-1 from LPS-activated B cells into the Bluescript vector (Stratagene, La Jolla, CA). The following internal ON was used as a ³²P-labeled BSAP probe: 5′-AAGCCATGGCTGAATACTCTG-3′. PCR signal intensities were quantified by densitometry (Fujibas-1000 detector, Fuji, Tokyo, Japan). Control experiments of titration were performed in order to ensure the quantification of the cDNA samples: aliquots of the amplifications were removed at 20, 25, 30, or 35 cycles in preliminary PCRs, and serial dilutions of the positive cDNA samples were PCR amplified for 35 cycles.

In vitro protocols with phosphorothioate-modified ON

Phosphorothioate ON were prepared by Isogen Bioscience BV (Maarssen, The Netherlands). These consisted of three 18-mer that were antisense (AS), sense (S), and non-sense (NS) for the DNA sequence spanning the fragment of the murine Blimp-1 cDNA located −4 to +14 from the first ATG: AS: 5′-TAAGCCTCTCTCATGTCC-3′; S: 5′-GGACATGAGAGAGGCTTA-3′; NS: 5′-ARSRWWRWSRWRRWRWWA-3′ (W: A + T; R: A + G; S: G + C).

Spleen B cells were purified by cell sorting, washed twice with BSS, and then cultured in serum-free Opti-MEM (Life Technologies) supplemented with 5% NCTC 109 medium (BioWhittaker, Walkersville, MD) in basal conditions, under stimulation with LPS, anti-CD40 mAb + IL-4, or cocultured with SR.D10 cells, as described above. Different concentrations of ON were added at the start of cultures and for various lengths of time.

Western blot analysis

Cell extracts (10⁷ cells/sample) were obtained after lysis in: 20 mM Tris-HCl (pH 7.5), 20% glycerol, 100 mM KCl, 2 mM DTT, and 1 mM PMSF. The resulting samples were boiled for 5 min in 2× SDS Laemmli sample buffer (0.04 M Tris-HCl (pH 6.8), 2% SDS, 5% β-mercaptoethanol, 10% glycerol, 0.002% bromphenol blue), electrophoresed in SDS-PAGE (6%), and then transferred to Hybond ECL nitrocellulose membranes (Amersham Life Science, Little Chalfont, U.K.). The total amount of protein/sample was determined before loading, using a protein determination reaction (Bio-Rad). As a further loading control, the lower part of the gel was sliced and stained with Coomasie blue, and two selected bands/lane were quantified (Gelworks 1-D intermediate program, UVP, Cambridge, U.K.). Filters were blocked with 5% fat-free dry milk in PBS/0.05% Tween-20 and incubated overnight with anti-Blimp-1 rabbit polyclonal antiserum (generously provided by Dr. M. M. Davis, see 6), washed, and further incubated with HRP-conjugated anti-rabbit Ig (Southern Biotechnology Associates; 2 h at room temperature). Filters were developed using ECL Western blotting detection reagent (Amersham).

The Blimp-1 gene is selectively up-regulated in IgM⁺Synd⁺CD43⁺ preplasma cells

Although it is absent in pre-B/B cell lines, Blimp-1 expression is detected in tumors with plasma cell phenotype and in plasma cell-containing cultures (6). We have dissected the kinetics of Blimp-1 transcription in normal splenocytes activated with LPS mitogen. The appearance of plasma cell markers in flow cytometry, Ig-secreting cells in ELISPOT assays, and Blimp-1 transcripts in specific RT-PCR were analyzed in parallel. Blimp-1 was well expressed after a few hours of LPS culture, long before significant numbers of IgM-secreting cells were detected in the same cultures (24 to 48 h) (Fig. 1⇓A, white triangles). Blimp-1 specific transcripts reached half of their maximum values (obtained in 72-h cell cultures) at 18–24 h, when only small numbers of IgM-plasma cells (relative to those of IgM-secreting cells obtained at 72 h) had developed. The mean ratio between specific Blimp-1 gene expression and IgM plasma cells subsequently peaked at early time points (18 h) and later decreased, although IgM plasma cells were still expanding in the cultures (Fig. 1⇓A, black circles). These data show that the Blimp-1 gene is up-regulated very early after LPS B cell stimulation, and probably at cellular stages previous to fully mature plasma cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Kinetics of Blimp-1 expression and terminal B cell differentiation of LPS-activated spleen B cells. A, BALB/c splenocytes were stimulated in vitro with LPS, and both Blimp-1 specific transcripts (relative to those of β-actin gene; ▵) and IgM-secreting cells (□) were calculated in simultaneous RT-PCR and ELISPOT assays, respectively (left scale of ordinates). The data are expressed as relative values of the maximum obtained at 72 h of culture. RT-PCR signals obtained for Blimp-1 and β-actin genes were quantified by densitometric analyses. The Blimp-1 message levels were defined for each time point as follows: 100 × (OD Blimp-1/OD β-actin)/(maximal OD of Blimp-1/maximal OD of β-actin). A total of 107,000 ± 16,000 IgM-secreting cells/10⁶ cultured cells were detected in 72-h LPS cultures. The ratios between Blimp-1 transcription levels and IgM-secreting cells were calculated (•; right scale of ordinates). The data obtained in five independent experiments are shown in the figure (mean ± SD). B, Two-color stainings of LPS-stimulated BALB/c splenocytes were performed at the starting point and in 72-h cultures by using fluoresceinated anti-CD43 and anti-IgM mAb, and biotinylated anti-Synd mAb, as described in Materials and Methods. Cell size (as defined by forward scatter) is shown in histograms for normal splenocyte (left histogram), and for 72 h LPS IgM⁺Synd⁻ and IgM⁺Synd⁺ cell populations (right histograms: shaded and empty areas, respectively). All data are displayed in log₁₀ fluorescence scales, except for forward light scatter (FSC), which is linear. In the bottom, RT-PCR assays for Blimp-1, BSAP, and β-actin are represented. They were performed on normal splenocytes, 72-h LPS spleen cell cultures, and 72-h LPS IgM⁺Synd⁺ cells purified by FACS sorting. The Blimp-1/β-actin message ratios obtained in the representative experiment shown were 0.31 for normal splenocytes, 26 for 72-h LPS spleen cultures, and 337 for 72-h IgM⁺Synd⁺ cells.

Mitogenic stimulation of spleen B cells induced their entry into G1, an increase in cell size (as shown by large forward scatter in FACS; histograms of Fig. 1⇑B) and progression through the rest of the cell cycle. A fraction of these activated B cells progressively acquired features of differentiating plasma cells. We analyzed in the LPS stimulated cultures the expression of CD43 and Synd-1 Ags, which are normally present in bone marrow pre-B cells, down-regulated in mature B cells, and reexpressed again in IgM-secreting plasma cells (12, 20). A novel population of blastic IgM⁺Synd⁺CD43⁺ cells emerged very early in the cultures (2% in 24-h cultures) and expanded during the following days (5% and 10% at 48 and 72 h, respectively) (Fig. 1⇑B, and data not shown). The IgM⁺Synd⁺ cell population contained most of the IgM-secreting cells of the cultures (as observed after its purification with FACS sorting and IgM-specific ELISPOT analyses; data not shown), although it also included other preplasma cell stages. The frequencies of IgM⁺Synd⁺CD43⁺ cells in the LPS cultures were higher (10-fold) than the relative number of IgM-secreting cells present in the same cultures. The early Blimp-1 transcription shown above (Fig. 1⇑A) suggested that this gene might be selectively up-regulated at the level of the IgM⁺Synd⁺CD43⁺ B cell/plasma cell transitional stage. We tested this possibility directly by analyzing the Blimp-1 transcripts present in IgM⁺Synd⁺ cells purified by FACS sorting, in quantitative RT-PCR analyses. The relative expression of Blimp-1 (compared with the one of β-actin) was 10–20 times higher in the purified IgM⁺Synd⁺ cells than in total LPS cultures (Fig. 1⇑B). Blimp-1 transcripts were barely detected in the sorted Synd⁻ population of the same cultures (data not shown). In contrast, transcription of the BSAP gene, expressed in mature B cells (5, 21), was strongly down-regulated in the LPS cultures and in the purified IgM⁺Synd⁺ cells (Fig. 1⇑B).

Inhibition of Blimp-1 protein expression suppresses the generation of IgM⁺Synd⁺ preplasma and IgM-secreting plasma cells in LPS-activated spleen B lymphocytes

Blimp-1 expression was previously shown to be sufficient for plasma cell maturation, as demonstrated by Blimp-1 transfection into either nonsecreting B cell lines or anti-IgM-blocked B cells (6, 8). We asked whether Blimp-1 activity was also necessarily required for plasma cell differentiation. Phosphorothioate-modified, Blimp-1-specific S, AS, and NS ON were generated and added to LPS-stimulated cultures of FACS-sorted, C3H/HeJ spleen B cells. To reduce sequence-independent effects of ON, we included no CpG motifs (B cell stimulatory) in their sequences, and we worked with low ON concentrations (1–10 μM) (22, 23). These experiments revealed a specific, dose-dependent reduction of the IgM⁺Synd⁺ preplasma cell population in cultures treated with AS Blimp-1 ON (Fig. 2⇓A). More important, the emergence of IgM-secreting plasma cells was progressively reduced in a time- and dose-dependent manner with the AS Blimp-1 therapy. Thus, IgM-secreting cells of AS Blimp-1 ON-treated cultures (3 μM) reached only one-third to one-fourth of those present in 3-day basal LPS cultures. Some nonspecific reductions in plasma cells were observed with both S and NS Blimp-1 ON, but they were significantly different from those induced by Blimp-1 AS ON (p < 0.05, for all the ON doses in 72-h cell cultures, after applying the two-tailed unpaired Student t test) (Fig. 2⇓B). Cell cultures extended until 96 h with 3 and 10 μM doses of ON provided similar results, although nonspecific cell death increased (data not shown). In order to check for the cellular level of action and the specificity of the anti-Blimp-1 ON treatment, we analyzed both Blimp-1 gene transcripts by RT-PCR and Blimp-1 protein levels by means of Western blots with a Blimp-1 antiserum (6), in selected samples of the cell cultures. While Blimp-1 gene transcription remained unmodified with/without ON treatments (Fig. 2⇓C), the small Blimp-1 protein levels detected in ex vivo B cells were augmented upon LPS stimulation (3-fold) and this increase was suppressed with AS ON, but not with S ON (Fig. 2⇓D). Blimp-1 AS ON thus acted at the posttranscriptional level of expression. We here conclude that Blimp-1 gene activity is necessary for normal spleen B lymphocyte differentiation to IgM-secreting plasma cells in T cell-independent conditions. The Blimp-1 role is first observed at the level of the generation of IgM⁺Synd⁺ preplasma cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Effects of the incubation with Blimp-1 S, AS, or NS phosphorothioate-modified ON on LPS-activated spleen B cells. A, Two-color flow cytometry analyses were undertaken as in Fig. 1⇑. C3H/HeJ spleen B cells were purified by FACS sorting. The changes produced by S and AS ON (1 μM and 3 μM) on the IgM⁺Synd⁺ cell populations present in 72-h LPS cell cultures are displayed in the figure. B, IgM-secreting cell numbers were determined in the LPS-stimulated B cell cultures by ELISPOT assays, as described in Materials and Methods. The kinetics of changes in the presence of 3 μM ON at different time points (left) and a dose-response curve at 72 h of culture (right) of ON treatments are displayed. The asterisk denotes the IgM-secreting cells in the absence of ON. The mean ± SEM of three independent experiments is shown in the figure. C, Detection of Blimp-1 and β-actin gene transcripts in 72-h splenic B cell cultures in the presence and absence of ON treatments, undertaken as in Fig. 1⇑. The Blimp-1 transcription levels were calculated by densitometry, and were 0.5 ± 0.1, 54 ± 3.8, 62 ± 3.5, and 57 ± 17 for basal, LPS, LPS + AS ON, and LPS + S ON, respectively (mean ± SD). D, Western blot analyses of cell extracts from LPS-activated B cells cultured for 72 h with or without ON treatments were performed with anti-Blimp-1-specific Abs, as described in Material and Methods. The amount of protein loaded per sample was 4, 4.1, 4.7, and 5.2 μg for basal conditions, LPS cultures, LPS + AS ON, and LPS + S ON, respectively.

Th-induced B cell differentiation to IgG1-switched plasma cells is independent of Blimp-1 activity

T-B cell collaboration is crucial to efficient Ab responses to Ags, in order to trigger the processes of Ig switching and somatic hypermutation in the germinal center reaction. This intercellular cooperation is sustained by defined receptor-ligand pairs (e.g., CD40-CD40L, OX40-OX40L) and T cell-derived soluble factors (24, 25). Since LPS mitogen directly activates B cells in the absence of T cell help, we decided to study whether Blimp-1 transcription factor was also required in bona fide T cell-driven plasma cell formation. We cocultured spleen B cells from C3H/HeJ mice (H-2^k) with syngeneic autoreactive SR.D10 Th2 cells (9, 10). We then analyzed B cell proliferation and the production of IgG1-secreting cells in different basal conditions (medium alone, LPS, SR.D10 T/B cell coculture) and following the addition of S and AS Blimp-1 ON. B cell proliferation (as defined by [³H]thymidine incorporation) was unmodified by ON treatments under any condition tested (Fig. 3⇓; top columns). The production of IgG1-secreting cells was significantly diminished upon the addition of AS Blimp-1 ON to the LPS cultures (Fig. 3⇓; bottom left columns), as it also happened for IgM-secreting cells (see Fig. 2⇑). In contrast, no change in IgG1 plasma cell numbers was observed in the T/B cell cocultures, in the presence of either S or AS Blimp-1 ON (Fig. 3⇓; bottom right columns). Similar negative results were observed for the IgM-secreting cells (not shown). Blimp-1 transcripts were not induced in these T/B cultures (2.8-fold vs 146-fold of day 0 specific Blimp-1 signals detected in 96-h SR.D10/C3H/HeJ B cell and LPS cultures, respectively; data not shown from three independent experiments). The findings described here suggest 1) that B cell proliferation rates are independent of Blimp-1 activity, and 2) that Blimp-1 may also be dispensable for T cell-driven plasma cell differentiation and Ig switching.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Effects of Blimp-1 ON treatment on B cell proliferation and induction of IgG1-secreting cells in Th2/B cell cocultures. Purified spleen B cells from C3H/HeJ mice were activated with LPS as above, or with mitomycin C-treated SR.D10 cells, as described in Materials and Methods. Proliferation rates (top columns) were calculated by [³H]thymidine incorporation and are displayed in logarithmic scales. IgG1-secreting cell numbers (bottom columns) were detected in specific ELISPOT assays. Only the data obtained with the representative 3 μM dose of ON are shown in the figure. The results were obtained from 72-h cultures set up in triplicate with values expressed as mean ± SEM (n = 4 experiments). *, p < 0.001, compared with ON-untreated and with S ON-treated LPS cultures. The significance was analyzed using the two-tailed unpaired Student t test.

Anti-CD40 + IL-4 stimulation of spleen B cells gives rise to IgG1-secreting cells in the absence of Blimp-1

Anti-CD40 mAb and IL-4 cooperate in the differentiation of B cells to IgG1 plasma cells (25, 26, 27), and partially mimic conventional Th activities. Both stimuli together drive the B cells away from early IgM secretion and commit them to isotype switching and, thus, presumably to enter the long-lived plasma and memory cell pools (28). We set up cultures of spleen B cells stimulated with anti-CD40 mAb and IL-4 as a surrogate of T cell help (11), in the presence or not of Blimp-1 ON treatments. A population of IgG1-secreting cells newly appeared in the anti-CD40 + IL-4 cell cultures. Neither the IgM⁺Synd⁺CD43⁺ cell subset nor a significant up-regulation of Blimp-1 gene transcription was observed, when compared with parallel LPS cell cultures (Fig. 4⇓, A and B). While the numbers of IgG1-secreting cells did not change upon Blimp-1 AS ON therapy in anti-CD40 + IL-4 cell cultures, they were selectively decreased in LPS cell cultures submitted to the same Blimp-1 AS ON (Fig. 4⇓C). Cell cultures stimulated with either anti-CD40 mAb alone or with IL-4 alone did not show either the IgM⁺Synd⁺CD43⁺ cells, the Blimp-1 up-regulation, or IgG1-secreting plasma cells. A consistent decrease of basal Blimp-1 transcription levels was observed in the IL-4 enriched cell cultures, as has been described (29). BSAP transcripts of mature B cells were decreased in all the stimulated B cell cultures (IL-4, anti-CD40 mAb, anti-CD40 mAb + IL-4, and LPS), even in those cases in which Blimp-1 was not up-regulated (Fig. 4⇓). We conclude that B cell differentiation to IgG1-secreting plasma cells secondary to CD40/CD40L + IL-4 signals follows a distinctive pathway in the absence of IgM⁺Synd⁺CD43⁺ preplasma cells, and without the requirement of Blimp-1 gene expression and activity.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Absence of Blimp-1 up-regulation and activity in spleen B cell differentiation driven by anti-CD40 + IL-4. The cell cultures of C3H/HeJ splenocytes were established in basal conditions and stimulated with anti-CD40 mAb (10 μg/ml), IL-4 (100 U/ml), anti-CD40 mAb + IL-4 and LPS (25 μg/ml), and analyzed after 72 h. A, Flow cytometry detection of the IgM⁺Synd⁺CD43⁺ cell population in the above-mentioned samples. B, Levels of transcription of Blimp-1, BSAP, and β-actin genes by specific RT-PCR analyses as described in Figure 1⇑. C, Quantitative changes of IgG1-secreting cells newly appearing in the stimulated B cell cultures, submitted to anti-Blimp-1 ON treatment. (Mean ± SEM; n = 4 experiments). St^† = stimulus.