Rel-dependent induction of A1 transcription is required to protect B cells from antigen receptor ligation-induced apoptosis

Abstract

In response to different extracellular signals, Rel/NF-kappaB transcription factors are critical regulators of apoptosis in a variety of cell types. Here we show that in normal B and T cells, expression of the Bcl-2 prosurvival homolog, A1, is rapidly induced in a Rel-dependent manner by mitogens. In B-cell lines derived from c-rel-/- mice, which like primary cells lacking Rel undergo apoptosis in response to antigen receptor ligation, constitutive expression of an A1 transgene inhibits this pathway to cell death. These findings are the first to show that Rel/NF-kappaB regulates physiologically the expression of a Bcl-2-like protein that is critical for the control of cell survival during lymphocyte activation.

Figure 1

The mitogen-induced expression of A1 mRNA is markedly impaired in c-rel^−/− lymphocytes. Ten-microgram samples of total RNA isolated from untreated or mitogen stimulated wild-type, nfkb1^−/−, or c-rel^−/− splenocytes activated in culture for 2 hr were analyzed by Northern blot hybridization. Filters were sequentially hybridized with murine A1 and rat GAPDH cDNA probes and exposed to autoradiography for between 24 and 48 hr. The stimuli were untreated cells (lane 1); concanavalin A (lane 2); anti-CD3 plus anti-CD28 antibodies (lane 3); phorbol ester (lane 4); lipopolysaccharide (lane 5); anti-RP antibodies (lane 6); anti-IgM antibodies (lane 7).

Figure 2

Nucleotide sequence of the 5′ flanking region of the murine A1 gene. The DNA-binding motif for Rel/NF-κB located at residues −488 to −479 is boxed. The italicized sequence represents A1 transcript, with nucleotides +1 to +149 (regular type) corresponding to 5′ untranslated sequence and +150 to +155 (boldface type) representing coding sequence (GenBank accession no. 515003). The A1 transcription and translation initiation sites, indicated by the raised and inverted arrows, respectively, were determined by Lin et al. (1993).

Figure 3

Functional analysis of the murine A1 promoter. (A) Schematic diagram of the A1 5′ flanking region and the CAT reporter plasmids. The numbers in parenthesis indicate the positions of restriction enzyme sites introduced into the murine A1 5′ flanking region by in vitro mutagenesis according to the numbering of the sequence outlined in Fig. 2. The open box represents the putative NF-κB-binding site (5′-AGGGGATCCC-3′: −488 to −479), whereas the corresponding symbol with a cross represents the mutated motif (5′-AGGTCATAAC-3′). The wavy arrow denotes the transcription initiation site and the straight arrow denotes the murine A1 initiation codon. The CAT gene is depicted as a closed box. Plasmid nomencature is indicated to the right of each construct. (B) Mutation of the NF-κB-binding motif abolishes Rel-dependent transcription in T cells. Jurkat cells were transfected transiently with 2 μg of the reporter plasmids pCAT (lanes 1,2), A1 L-CAT (lanes 3,4), A1 S-CAT (lanes 5,6) or A1κB_m L-CAT (lanes 7,8) plus 10 μg of the expression plasmid DAMP56 containing no insert (lanes 1,3,5,7) or the c-rel expression plasmid pDAMP56c-rel (lanes 2,4,6,8). Chloramphenicol acetylation for transfections with the pDAMP56 or pDAMP56c-rel are indicated by open and closed bars, respectively. These results represent the mean percentage of chloramphenicol acetylation ± s.d. obtained from five separate sets of transient transfections.

Figure 3

Functional analysis of the murine A1 promoter. (A) Schematic diagram of the A1 5′ flanking region and the CAT reporter plasmids. The numbers in parenthesis indicate the positions of restriction enzyme sites introduced into the murine A1 5′ flanking region by in vitro mutagenesis according to the numbering of the sequence outlined in Fig. 2. The open box represents the putative NF-κB-binding site (5′-AGGGGATCCC-3′: −488 to −479), whereas the corresponding symbol with a cross represents the mutated motif (5′-AGGTCATAAC-3′). The wavy arrow denotes the transcription initiation site and the straight arrow denotes the murine A1 initiation codon. The CAT gene is depicted as a closed box. Plasmid nomencature is indicated to the right of each construct. (B) Mutation of the NF-κB-binding motif abolishes Rel-dependent transcription in T cells. Jurkat cells were transfected transiently with 2 μg of the reporter plasmids pCAT (lanes 1,2), A1 L-CAT (lanes 3,4), A1 S-CAT (lanes 5,6) or A1κB_m L-CAT (lanes 7,8) plus 10 μg of the expression plasmid DAMP56 containing no insert (lanes 1,3,5,7) or the c-rel expression plasmid pDAMP56c-rel (lanes 2,4,6,8). Chloramphenicol acetylation for transfections with the pDAMP56 or pDAMP56c-rel are indicated by open and closed bars, respectively. These results represent the mean percentage of chloramphenicol acetylation ± s.d. obtained from five separate sets of transient transfections.

Figure 4

Analysis of Rel/NF-κB complexes in resting and mitogen-stimulated B cells. Nuclear extracts (1–2 μg) isolated from purified normal and c-rel^−/− splenic B cells stimulated with anti-IgM for 2 hr, then incubated with ³²P-radiolabeled A1κB probe were resolved on 5% nondenaturing polyacrylamide gels and exposed to autoradiography for 6–24 hr at −70°C. (A) A κB-binding complex rapidly induced by mitogen is absent in c-rel^−/− B cells. Nuclear extracts from resting (lanes 1,3,5,7,9,11) and anti-IgM-stimulated (lanes 2,4,6,8,10,12) normal (lanes 1,2,5,6,9,10), and c-rel^−/− (lanes 3,4,7,8,11,12) B cells were preincubated in the absence (lanes 1–4) or presence of a 50-fold molar excess of unlabeled A1κB (lanes 5–8) or A1κB_m (lanes 9–12) probe before adding radiolabeled A1κB. The inducible slow mobility and constitutive fast mobility complexes are designated C1 and C2, respectively. (B) The inducible C1 complex is a Rel-containing heterodimer. Nuclear extracts from resting (lanes 1–4) and anti-IgM-stimulated (lanes 5–8) wild-type splenic B cells were incubated with preimmune (lanes 1,5) or NF-κB1 (lanes 2,6), Rel (lanes 3,7), and RelA (lanes 4,8) -specific sera before adding the radiolabeled A1κB probe.

Figure 4

Analysis of Rel/NF-κB complexes in resting and mitogen-stimulated B cells. Nuclear extracts (1–2 μg) isolated from purified normal and c-rel^−/− splenic B cells stimulated with anti-IgM for 2 hr, then incubated with ³²P-radiolabeled A1κB probe were resolved on 5% nondenaturing polyacrylamide gels and exposed to autoradiography for 6–24 hr at −70°C. (A) A κB-binding complex rapidly induced by mitogen is absent in c-rel^−/− B cells. Nuclear extracts from resting (lanes 1,3,5,7,9,11) and anti-IgM-stimulated (lanes 2,4,6,8,10,12) normal (lanes 1,2,5,6,9,10), and c-rel^−/− (lanes 3,4,7,8,11,12) B cells were preincubated in the absence (lanes 1–4) or presence of a 50-fold molar excess of unlabeled A1κB (lanes 5–8) or A1κB_m (lanes 9–12) probe before adding radiolabeled A1κB. The inducible slow mobility and constitutive fast mobility complexes are designated C1 and C2, respectively. (B) The inducible C1 complex is a Rel-containing heterodimer. Nuclear extracts from resting (lanes 1–4) and anti-IgM-stimulated (lanes 5–8) wild-type splenic B cells were incubated with preimmune (lanes 1,5) or NF-κB1 (lanes 2,6), Rel (lanes 3,7), and RelA (lanes 4,8) -specific sera before adding the radiolabeled A1κB probe.

Figure 5

Engaging the antigen receptor of immortalized surface IgM-positive c-rel^−/− B-cell lines induces apoptosis. Cell lines derived from tumors arising independently in c-rel^+/+ Eμ-myc (404.4, lanes 1–3; 417.1, lanes 4–6), and c-rel^−/− Eμ-myc (B1.1, lanes 7–9; B78, lanes 10–12) mice were either untreated (lanes 1,4,7,10) or stimulated in culture for 72 hr with mitogenic antibodies raised against IgM (lanes 2,5,8,11) or the B-cell-specific cell surface protein RP (lanes 3,6,9,12). The frequency of apoptotic cells in each culture is shown for the 72-hr time point and has been expressed as a proportion of the total cell number determined by flow cytometric analysis of fixed cells stained with propidium iodide. More than 99% of each cell line were viable at the start of the experiment. These results represent the mean ± s.d. from six independent experiments.

Figure 6

A hormone-regulated Rel protein rescues c-rel^−/− B cells from anti-IgM-induced apoptosis. (A) Expression of the Rel/ER protein. Whole cell (lanes 1–4) or nuclear (lanes 5–8) extracts from 417.1 (c-rel^+/+; lanes 1,3,5,7) and B1.1(c-rel^−/− : lanes 2,4,6,8) cells infected with a control retrovirus (lanes 1,2) or a retrovirus expressing c-rel/ER (lanes 3–8) that had been untreated (lanes 1–6) or stimulated with 50 nm 4-HT (lanes 7,8) were resolved by SDS-PAGE and subjected to Western blot analysis using antibodies raised against the human estrogen receptor. (B) The 4-HT inducible κB-binding nuclear complex in c-rel^−/− B cells expressing Rel/ER contains a Rel protein. Nuclear extracts isolated from B1.1 cells infected with the c-rel/ER retrovirus that were untreated (lane 1) or 4-HT activated (lanes 2,3) were preincubated on ice without (lanes 1,2) or with Rel-specific sera for 30 min before adding 1 μg of poly [d(I-C)] and ³²P-labeled A1 κB probe and incubating for a further 20 min. The resultant DNA-binding complexes were resolved on 5% nondenaturing polyacrylamide gels and exposed to autoradiography for 6 hr at −70°C. The 4-HT-inducible DNA-binding complex is indicated by an arrow. (C) Inducible expression of Rel/ER protects c-rel^−/− B cells from anti-IgM-induced apoptosis. The 417.1 (c-rel^+/+; lanes 1–4) and B1.1 (c-rel^−/−; lanes 5–8) cell lines infected with the HEDME2SV40puroc-rel/ER retrovirus were either untreated (lanes 1,2,5,6) or stimulated with anti-IgM (lanes 3,4,7,8) for 72 hr in the absence (lanes 1,3,5,7) or presence (lanes 2,4,6,8) of 4-HT. At 24-hr intervals, samples of cells from all cultures were fixed, stained with PI, and analyzed by flow cytometry. The frequency of apoptotic cells, shown for the 72-hr time point is expressed as a proportion of the total cell number. More than 99% of each cell line were viable at the start of the experiment. These results represent the mean ± s.d. from five independent experiments. (D) A1 transcript levels increase in 4-HT-treated c-rel^−/− B cells expressing Rel/ER. Ten-microgram samples of total RNA isolated from untreated (lanes 1,3) or 4-HT-stimulated (lanes 2,4) c-rel^−/− B1.1 cells infected with the control retrovirus (lanes 1,2) or the c-rel/ER virus (lanes 3,4) were analyzed by Northern blot hybridization. Filters were sequentially hybridized with ³²P-radiolabeled murine A1 and rat GAPDH cDNA probes, then exposed for autoradiography at −70°C for 24 hr.

Figure 6

A hormone-regulated Rel protein rescues c-rel^−/− B cells from anti-IgM-induced apoptosis. (A) Expression of the Rel/ER protein. Whole cell (lanes 1–4) or nuclear (lanes 5–8) extracts from 417.1 (c-rel^+/+; lanes 1,3,5,7) and B1.1(c-rel^−/− : lanes 2,4,6,8) cells infected with a control retrovirus (lanes 1,2) or a retrovirus expressing c-rel/ER (lanes 3–8) that had been untreated (lanes 1–6) or stimulated with 50 nm 4-HT (lanes 7,8) were resolved by SDS-PAGE and subjected to Western blot analysis using antibodies raised against the human estrogen receptor. (B) The 4-HT inducible κB-binding nuclear complex in c-rel^−/− B cells expressing Rel/ER contains a Rel protein. Nuclear extracts isolated from B1.1 cells infected with the c-rel/ER retrovirus that were untreated (lane 1) or 4-HT activated (lanes 2,3) were preincubated on ice without (lanes 1,2) or with Rel-specific sera for 30 min before adding 1 μg of poly [d(I-C)] and ³²P-labeled A1 κB probe and incubating for a further 20 min. The resultant DNA-binding complexes were resolved on 5% nondenaturing polyacrylamide gels and exposed to autoradiography for 6 hr at −70°C. The 4-HT-inducible DNA-binding complex is indicated by an arrow. (C) Inducible expression of Rel/ER protects c-rel^−/− B cells from anti-IgM-induced apoptosis. The 417.1 (c-rel^+/+; lanes 1–4) and B1.1 (c-rel^−/−; lanes 5–8) cell lines infected with the HEDME2SV40puroc-rel/ER retrovirus were either untreated (lanes 1,2,5,6) or stimulated with anti-IgM (lanes 3,4,7,8) for 72 hr in the absence (lanes 1,3,5,7) or presence (lanes 2,4,6,8) of 4-HT. At 24-hr intervals, samples of cells from all cultures were fixed, stained with PI, and analyzed by flow cytometry. The frequency of apoptotic cells, shown for the 72-hr time point is expressed as a proportion of the total cell number. More than 99% of each cell line were viable at the start of the experiment. These results represent the mean ± s.d. from five independent experiments. (D) A1 transcript levels increase in 4-HT-treated c-rel^−/− B cells expressing Rel/ER. Ten-microgram samples of total RNA isolated from untreated (lanes 1,3) or 4-HT-stimulated (lanes 2,4) c-rel^−/− B1.1 cells infected with the control retrovirus (lanes 1,2) or the c-rel/ER virus (lanes 3,4) were analyzed by Northern blot hybridization. Filters were sequentially hybridized with ³²P-radiolabeled murine A1 and rat GAPDH cDNA probes, then exposed for autoradiography at −70°C for 24 hr.

Figure 6

A hormone-regulated Rel protein rescues c-rel^−/− B cells from anti-IgM-induced apoptosis. (A) Expression of the Rel/ER protein. Whole cell (lanes 1–4) or nuclear (lanes 5–8) extracts from 417.1 (c-rel^+/+; lanes 1,3,5,7) and B1.1(c-rel^−/− : lanes 2,4,6,8) cells infected with a control retrovirus (lanes 1,2) or a retrovirus expressing c-rel/ER (lanes 3–8) that had been untreated (lanes 1–6) or stimulated with 50 nm 4-HT (lanes 7,8) were resolved by SDS-PAGE and subjected to Western blot analysis using antibodies raised against the human estrogen receptor. (B) The 4-HT inducible κB-binding nuclear complex in c-rel^−/− B cells expressing Rel/ER contains a Rel protein. Nuclear extracts isolated from B1.1 cells infected with the c-rel/ER retrovirus that were untreated (lane 1) or 4-HT activated (lanes 2,3) were preincubated on ice without (lanes 1,2) or with Rel-specific sera for 30 min before adding 1 μg of poly [d(I-C)] and ³²P-labeled A1 κB probe and incubating for a further 20 min. The resultant DNA-binding complexes were resolved on 5% nondenaturing polyacrylamide gels and exposed to autoradiography for 6 hr at −70°C. The 4-HT-inducible DNA-binding complex is indicated by an arrow. (C) Inducible expression of Rel/ER protects c-rel^−/− B cells from anti-IgM-induced apoptosis. The 417.1 (c-rel^+/+; lanes 1–4) and B1.1 (c-rel^−/−; lanes 5–8) cell lines infected with the HEDME2SV40puroc-rel/ER retrovirus were either untreated (lanes 1,2,5,6) or stimulated with anti-IgM (lanes 3,4,7,8) for 72 hr in the absence (lanes 1,3,5,7) or presence (lanes 2,4,6,8) of 4-HT. At 24-hr intervals, samples of cells from all cultures were fixed, stained with PI, and analyzed by flow cytometry. The frequency of apoptotic cells, shown for the 72-hr time point is expressed as a proportion of the total cell number. More than 99% of each cell line were viable at the start of the experiment. These results represent the mean ± s.d. from five independent experiments. (D) A1 transcript levels increase in 4-HT-treated c-rel^−/− B cells expressing Rel/ER. Ten-microgram samples of total RNA isolated from untreated (lanes 1,3) or 4-HT-stimulated (lanes 2,4) c-rel^−/− B1.1 cells infected with the control retrovirus (lanes 1,2) or the c-rel/ER virus (lanes 3,4) were analyzed by Northern blot hybridization. Filters were sequentially hybridized with ³²P-radiolabeled murine A1 and rat GAPDH cDNA probes, then exposed for autoradiography at −70°C for 24 hr.

Figure 6

A hormone-regulated Rel protein rescues c-rel^−/− B cells from anti-IgM-induced apoptosis. (A) Expression of the Rel/ER protein. Whole cell (lanes 1–4) or nuclear (lanes 5–8) extracts from 417.1 (c-rel^+/+; lanes 1,3,5,7) and B1.1(c-rel^−/− : lanes 2,4,6,8) cells infected with a control retrovirus (lanes 1,2) or a retrovirus expressing c-rel/ER (lanes 3–8) that had been untreated (lanes 1–6) or stimulated with 50 nm 4-HT (lanes 7,8) were resolved by SDS-PAGE and subjected to Western blot analysis using antibodies raised against the human estrogen receptor. (B) The 4-HT inducible κB-binding nuclear complex in c-rel^−/− B cells expressing Rel/ER contains a Rel protein. Nuclear extracts isolated from B1.1 cells infected with the c-rel/ER retrovirus that were untreated (lane 1) or 4-HT activated (lanes 2,3) were preincubated on ice without (lanes 1,2) or with Rel-specific sera for 30 min before adding 1 μg of poly [d(I-C)] and ³²P-labeled A1 κB probe and incubating for a further 20 min. The resultant DNA-binding complexes were resolved on 5% nondenaturing polyacrylamide gels and exposed to autoradiography for 6 hr at −70°C. The 4-HT-inducible DNA-binding complex is indicated by an arrow. (C) Inducible expression of Rel/ER protects c-rel^−/− B cells from anti-IgM-induced apoptosis. The 417.1 (c-rel^+/+; lanes 1–4) and B1.1 (c-rel^−/−; lanes 5–8) cell lines infected with the HEDME2SV40puroc-rel/ER retrovirus were either untreated (lanes 1,2,5,6) or stimulated with anti-IgM (lanes 3,4,7,8) for 72 hr in the absence (lanes 1,3,5,7) or presence (lanes 2,4,6,8) of 4-HT. At 24-hr intervals, samples of cells from all cultures were fixed, stained with PI, and analyzed by flow cytometry. The frequency of apoptotic cells, shown for the 72-hr time point is expressed as a proportion of the total cell number. More than 99% of each cell line were viable at the start of the experiment. These results represent the mean ± s.d. from five independent experiments. (D) A1 transcript levels increase in 4-HT-treated c-rel^−/− B cells expressing Rel/ER. Ten-microgram samples of total RNA isolated from untreated (lanes 1,3) or 4-HT-stimulated (lanes 2,4) c-rel^−/− B1.1 cells infected with the control retrovirus (lanes 1,2) or the c-rel/ER virus (lanes 3,4) were analyzed by Northern blot hybridization. Filters were sequentially hybridized with ³²P-radiolabeled murine A1 and rat GAPDH cDNA probes, then exposed for autoradiography at −70°C for 24 hr.

Figure 7

Constitutive expression of A1 protects c-rel^−/− B cells from anti-IgM-induced apoptosis. (A) Expression of wild-type and G87E mutant A1 proteins in Eμ-myc B cell lines. c-rel^+/+ (417.1) and c-rel^−/− (B1.1) Eμ-myc B cell lines stabily transfected with a vector control (solid line) Flag-A1 (amino-terminal Flag-tagged A1; open dark line) or Flag-A1G87E (amino-terminal Flag-tagged A1 mutant G87E; open light line) were fixed, permeabilized, and stained with an anti-Flag monoclonal antibody and analyzed by flow cytometry as described (Huang et al. 1997a). (B) Enforced expression of A1 protects c-rel^−/− B-cell lines from BCR ligation-induced apoptosis. Parental Eμ-myc cells (lanes 1,6) and stabily transfected clones expressing Flag-tagged A1 (lanes 2,3,7,8) or Flag-tagged A1 G87E (lanes 4,5,9,10) were either untreated (lanes 1,2,4,6,7,9) or stimulated with anti-IgM (lanes 3,5,8,10) for 72 hr. The frequency of apoptotic cells expressed as a proportion of the total cell number in each culture was determined by flow cytometric analysis of fixed cells stained with PI. More than 99% of each cell line were viable at the start of the experiment. These results represent the mean ± s.d. from six independent experiments.

Figure 7

Constitutive expression of A1 protects c-rel^−/− B cells from anti-IgM-induced apoptosis. (A) Expression of wild-type and G87E mutant A1 proteins in Eμ-myc B cell lines. c-rel^+/+ (417.1) and c-rel^−/− (B1.1) Eμ-myc B cell lines stabily transfected with a vector control (solid line) Flag-A1 (amino-terminal Flag-tagged A1; open dark line) or Flag-A1G87E (amino-terminal Flag-tagged A1 mutant G87E; open light line) were fixed, permeabilized, and stained with an anti-Flag monoclonal antibody and analyzed by flow cytometry as described (Huang et al. 1997a). (B) Enforced expression of A1 protects c-rel^−/− B-cell lines from BCR ligation-induced apoptosis. Parental Eμ-myc cells (lanes 1,6) and stabily transfected clones expressing Flag-tagged A1 (lanes 2,3,7,8) or Flag-tagged A1 G87E (lanes 4,5,9,10) were either untreated (lanes 1,2,4,6,7,9) or stimulated with anti-IgM (lanes 3,5,8,10) for 72 hr. The frequency of apoptotic cells expressed as a proportion of the total cell number in each culture was determined by flow cytometric analysis of fixed cells stained with PI. More than 99% of each cell line were viable at the start of the experiment. These results represent the mean ± s.d. from six independent experiments.