Many individuals develop a single or a few brief episodes of autoimmunity from which they recover. Mechanisms that quell pathologic autoimmunity following such a breakdown of self-tolerance are not clearly understood. In this study, we show that in nonautoimmune mice, dsDNA-specific autoreactive B cells exist but remain inactive. This state of inactivation in dsDNA-specific B cells could be disrupted by autoreactive Th cells; in this case T cells that react with peptides from the VH region of anti-DNA Abs (hereafter called anti-VH T cells). Immunization with anti-DNA mAb, its γ-chain or peptides derived from its VH region induced anti-VH Th cells, IgG anti-dsDNA Ab, and proteinuria. The breakdown of B cell tolerance in nonautoimmune mice, however, was short-lived: anti-DNA Ab and nephritis subsided despite subsequent immunizations. The recovery from autoimmunity temporally correlated with the appearance of T cells that inhibited anti-DNA Ab production. Such inhibitory T cells secreted TGFβ; the inhibition of anti-DNA Ab production by these cells was partly abolished by anti-TGFβ Ab. Even without immunization, nonautoimmune mice possess T cells that can inhibit autoantibody production. Thus, inhibitory T cells in nonautoimmune mice may normally inhibit T-dependent activation of autoreactive B cells and/or reverse such activation following stimulation by Th cells. The induction of such inhibitory T cells may play a role in protecting nonautoimmune mice from developing chronic autoimmunity.

Self-reactive B and Th cells exist in the normal immune repertoire, yet most individuals avoid pathologic autoantibody production. Some otherwise healthy individuals develop a single or a few brief episodes of autoimmunity from which they recover (1), whereas chronic autoantibody production is characteristic of individuals with lupus (2). Some autoantibodies, such as Ab against DNA, cause tissue damage (3); their production is largely dependent on Th cells (4, 5, 6, 7).

To understand the mechanisms that prevent autoreactive B cells from producing Ab, several laboratories have generated transgenic mice in which most B cells express Ig of a single specificity. These studies have identified several mechanisms, such as clonal deletion and anergy, that protect mice from breaking self-tolerance in B cells (8, 9, 10, 11, 12, 13). Not many studies, however, have addressed the protective mechanisms that come into play once self-tolerance has been broken in normal mice.

Here, we ask: 1) Can B cells in nonautoimmune mice be activated to secrete anti-DNA Ab by exposure to autoreactive Th cells? 2) Will such a breakdown in the state of inactivation in B cells cause persistent anti-DNA Ab production and disease in these mice? 3) Can these mice develop protective mechanisms that contain or reverse the loss of self-tolerance in B cells?

To address these questions, we used a nontransgenic model in which MHC class II-restricted Th cells help B cells produce IgG anti-DNA Ab (4); the (NZB × NZW)F1 (BWF1) strain of lupus-prone mice; and two nonautoimmune mice that have MHC class II molecules identical to those in BWF1 mice. Using this system, we show that nonautoimmune mice, upon persistent exposure to autoreactive Th cells, develop inhibitory T cells that can inhibit autoantibody production and renal disease.

NZB/B1nj, NZW/Lacj, BALB/c, and B10.PL mice were purchased from The Jackson Laboratory (Bar Harbor, ME) and intercrossed to generate BWF1 (NZB × NZW) F1, CWF1 (BALB/c × NZW) F1, and BPF1 (NZB × B10.Pl) F1 mice in the Cincinnati Veterans Affairs Medical Center or University of California, Los Angeles Vivarium. Female mice were used in all experiments. Experiments were performed in accordance with the institutional animal research committee guidelines.

15-mer overlapping peptides representing the entire VHD region sequence of an anti-dsDNA mAb, A6.1, were synthesized using a modified pin synthesis or a macrocrown method (14). Each 15-mer peptide overlapped its neighbor by 10 residues. Several 11- to 15-mer T cell epitopes and control peptides (Table I) were synthesized in large quantities at the Chiron Laboratories (Clayton, Australia), using F-moc chemistry. The synthetic peptides were analyzed for purity by HPLC and by mass spectrometry. Each peptide chromatographed essentially as a sharp single peak. All purified peptides had the expected molecular mass.

Table I.

Peptides used in this study: their sequences and origin

PeptidesSource (Ref. 3 )Sequence (Refs. 315 )
Stimulatory peptidesa   
A6H31–45 A6.1 VH GYFMNWVKQSHGKSL 
A6H34–45 A6.1 VH MNWVKQSHGKSL 
A6H41–55 A6.1 VH HGKSLEWIGRINPYN 
A6H58–69 A6.1 VH FYNQKFKGKATL 
A6H84–95 A6.1 VH SEDSALYYCARD 
Control peptidesb   
A6H93–107 A6.1 VH ARDSPYYYGSSYGFA 
A6H110–121 A6.1 VH YWGQGTSVTVSS 
HYHEL31–45 HYHEL-5 VH IEWVKQRPGHGLEWI 
HYHEL34–48 HYHEL-5 VH VKQRPGHGLEWIGEI 
HEL106–116 HEL NAWVAWRNRCK 
PeptidesSource (Ref. 3 )Sequence (Refs. 315 )
Stimulatory peptidesa   
A6H31–45 A6.1 VH GYFMNWVKQSHGKSL 
A6H34–45 A6.1 VH MNWVKQSHGKSL 
A6H41–55 A6.1 VH HGKSLEWIGRINPYN 
A6H58–69 A6.1 VH FYNQKFKGKATL 
A6H84–95 A6.1 VH SEDSALYYCARD 
Control peptidesb   
A6H93–107 A6.1 VH ARDSPYYYGSSYGFA 
A6H110–121 A6.1 VH YWGQGTSVTVSS 
HYHEL31–45 HYHEL-5 VH IEWVKQRPGHGLEWI 
HYHEL34–48 HYHEL-5 VH VKQRPGHGLEWIGEI 
HEL106–116 HEL NAWVAWRNRCK 
a

These peptides are derived from the VH region of a BWF1-derived anti-DNA mAb, A6.1 (3 ); they stimulate spontaneous T cell proliferation and help for anti-DNA Ab in unimmunized BWF1 mice (141520 ).

b

Controls used were: nonstimulatory peptides from the same VH (A6.1), such as A6H93–107; peptides, such as HYHEL31–45, from a non-autoreactive mAb, HYHEL-5 that uses the same VH gene family as does A6.1 (15 ); and a foreign peptide, HEL106–116. The control peptides do not induce spontaneous activation of BWF1 T cells but can elicit immune response when injected in vivo with adjuvant (4 ). Additionally, the control peptides do not increase anti-DNA Ab formation by BWF1 or CWF1 spleen cells.

Young CWF1 or BWF1 mice were immunized i.p. with 20 μg of A6.1 (purified from ascites fluid as IgG), emulsified in CFA, and boosted with 10–20 μg of A6.1 in IFA, as described in the figure legends.

Spleen cells were pooled from 3 to 12 mice and purified as mononuclear cells on Ficoll-Hypaque. Enrichment of CD4+, CD8+, and B cells was performed with the Vario MACS magnetic purification system using microbead-coated Ab (Miltenyi Biotec, Auburn, CA). Purity of the cell populations as determined by FACS varied from 92 to 99%.

Sonicated, nitrocellulose-filtered calf thymus dsDNA (100 μg/ml; Sigma-Aldrich, St. Louis, MO) was coated onto 96-well microtiter plates (catalogue no. 3590; Costar, Cambridge, MA) and incubated overnight at 4°C. After washing with PBS, DNA-coated plates were blocked for 1 h with 10% FCS in PBS at room temperature. Splenic T and B cells in complete medium (DMEM with 10% FBS) and varying concentrations of peptides were added to each well. After 8–12 h of incubation at 37°C, cells were poured off and plates were washed eight times. Plates were then incubated overnight at 4°C with alkaline phosphatase-conjugated antimouse IgG or IgA, diluted 1/500 in 10% FCS in PBS. Plates were washed again with Tris/Tween 20, and incubated with a mixture of 5-bromo-4-chloro-3-indolyl phosphate with agarose gel at a final concentration of 0.6%. Anti-dsDNA Ab-forming cells (AFC)3 were enumerated as blue spots using an inverted microscope (Leitz Canada, Midland, Ontario, Canada) by two of us (F.M.E. and R.R.S.) in a blinded fashion and recorded as number of AFC per 106 B cells. The specificity of anti-dsDNA ELISPOT was confirmed by competitive inhibition studies; the anti-dsDNA ELISPOT was inhibited by DNA but not with cardiolipin, keyhole limpet hemocyanin, or hen egg lysozyme (HEL) (15, 16, 17). The anti-dsDNA ELISPOT was inhibited by both dsDNA and ssDNA but it required two times more ssDNA than dsDNA for 50% inhibition (15, 17). This is consistent with observations that some populations of anti-DNA Abs share dsDNA and ssDNA binding specificities.

The results of ELISPOT assays for anti-dsDNA Ab were further confirmed in simultaneous ELISA. Briefly, spleen cells were cultured with or without peptides for 5 days; supernatants were tested for IgG anti-dsDNA Ab, as described previously (3, 4). Briefly, 96-well plates (Costar) were coated with dsDNA (100 μg/ml; Sigma-Aldrich) overnight at 4°C. After blocking with 10% FCS, samples were added. Bound Ab was detected with alkaline phosphatase-conjugated goat anti-mouse IgG (Fisher, Pittsburgh, PA). Reactions were developed with p-nitrophenyl phosphate (Sigma-Aldrich) and A405 was measured with an ELISA reader. The results of ELISA and ELISPOT assays correlated well with each other (15).

Splenic single-cell suspensions from the CWF1 or BWF1 mice immunized with A6H31–45 were fractionated into CD4+ or CD8+ T cells. The fractionated CD4+ and CD8+ T cells were cultured and periodically stimulated with syngeneic APCs (irradiated T cell-depleted spleen cells), A6H31–45 and Con A supernatant, as described previously (4).

Four B cell hybridomas, including a dsDNA-specific (A6.1) and three non-DNA binding (1GE6, 375-57, and 375-100), were established from BWF1 mice by fusion of their spleen cells with a nonsecreting murine myeloma cell, as described previously (3, 18).

Resting T cell lines were cultured with APCs and peptide; supernatants were tested for cytokines by ELISA, as previously described (19). TGFβ was assayed by sandwich ELISA using a TGFβ1-specific kit (G1230; Promega, Madison, WI) that is designed to measure biologically active TGFβ1. To assay for total TGFβ1, the samples were acid treated and neutralized according to the manufacturer’s instructions (www.promega.com). In brief, multiwell ELISA plates (Nunc MaxiSorp; Nunc, Naperville, IL) were coated with 100 μl of an anti-TGFβ1 mAb in carbonate buffer (pH 9.6) overnight at 4°C. Plates were washed three times with Tris-HCl (pH 7.6) containing 0.05% Tween 20, followed by blocking for 35 min at 37°C. Samples and standard were added, and plates were incubated for 90 min at room temperature. Plates were washed five times before the addition of 100 μl per well of a second polyclonal anti-TGFβ1 Ab. After 2 h at room temperature, plates were washed and 100 μl of a tertiary HRP-conjugated species-specific Ab was added. The plates were incubated for 2 more hours, then washed six times, and 100 μl of peroxidase substrate/tetramethylbenzidine solution was added. Fifteen minutes later, the reaction was stopped with 1 M phosphoric acid. A405 was measured with a Labsystem ELISA reader.

Proteinuria and blood urea nitrogen were tested using Albustix and Azostix (range, 0–4+), respectively, and creatinine was estimated by reagents from Stanbio Laboratories (San Antonio, TX), as described previously (4, 20).

Student’s t test was performed to compare the test and control groups. A Shapiro-Wilk test of normality was performed to determine whether the distribution of the effect of CWF1-derived T cell lines on anti-DNA Ab production (see Fig. 4 C) matches a normal distribution.

FIGURE 4.

Nonautoimmune mice develop T cells that down-regulate anti-DNA Ab production. A, BWF1 spleen cells were cultured with a T cell epitope (A6H31–45, peptide 7 in Fig. 2,B) or control (A6H110–121, peptide 23 in Fig. 2 B) peptide. To these cultures were added 105 freshly isolated T cells from CWF1 mice that were immunized five to eight times with A6.1 (▪) or from unimmunized BWF1 mice (control; ▧). Results are expressed as the mean triplicate IgG anti-DNA AFC per 106 B cells. The addition of CWF1 T cells to BWF1 spleen cell cultures decreased anti-DNA AFC in a peptide-specific manner (∗, p < 0.01). Similar results were obtained using fractionated CD4+ or CD8+ T cells and with another set of T cell determinant (A6H41–55) and control (A6H93–107) peptides. Data are the pool of three independent experiments. B, Short-term T cell lines from CWF1 mice (•) decrease anti-DNA Ab production by BWF1 B cells in a dose (T cell number)-dependent manner, while T cell lines from similarly immunized BWF1 mice (○) promote anti-DNA Ab production. Results are expressed as the mean triplicate IgG anti-DNA AFC per 106 B cells from one of two similar experiments; deviations from the mean were <10%. Similar results were obtained using another two sets of short-term T cell lines from CWF1 and BWF1 mice. C, Although most long-term T cell lines and clones from CWF1 mice inhibit anti-DNA Ab production, some promote autoantibody production. The resting T cell lines or clones that were established by subcloning short-term lines (shown in B) were cocultured with A6H31–45 and BWF1 splenic T plus B cells. Results are expressed as the mean IgG anti-DNA AFC per 106 B cells. The first bar on the left (BW B plus T cells without any added peptide) represents the background AFC number. The horizontal dashed line crossing this figure denotes one-half of the background AFC number. We arbitrarily used this cut off to label the T cell lines as inhibitory (▪), which included four CD4+ and nine CD8+ T cell lines. Two T cell lines, both CD4+, increased anti-DNA AFC 2-fold above the background (□, labeled arbitrarily as helper T cell lines). The remaining T cell lines that included 15 CD4+ and 12 CD8+ T cell lines had no or <2-fold effect on AFC (▨). Data are representative of two separate experiments.

FIGURE 4.

Nonautoimmune mice develop T cells that down-regulate anti-DNA Ab production. A, BWF1 spleen cells were cultured with a T cell epitope (A6H31–45, peptide 7 in Fig. 2,B) or control (A6H110–121, peptide 23 in Fig. 2 B) peptide. To these cultures were added 105 freshly isolated T cells from CWF1 mice that were immunized five to eight times with A6.1 (▪) or from unimmunized BWF1 mice (control; ▧). Results are expressed as the mean triplicate IgG anti-DNA AFC per 106 B cells. The addition of CWF1 T cells to BWF1 spleen cell cultures decreased anti-DNA AFC in a peptide-specific manner (∗, p < 0.01). Similar results were obtained using fractionated CD4+ or CD8+ T cells and with another set of T cell determinant (A6H41–55) and control (A6H93–107) peptides. Data are the pool of three independent experiments. B, Short-term T cell lines from CWF1 mice (•) decrease anti-DNA Ab production by BWF1 B cells in a dose (T cell number)-dependent manner, while T cell lines from similarly immunized BWF1 mice (○) promote anti-DNA Ab production. Results are expressed as the mean triplicate IgG anti-DNA AFC per 106 B cells from one of two similar experiments; deviations from the mean were <10%. Similar results were obtained using another two sets of short-term T cell lines from CWF1 and BWF1 mice. C, Although most long-term T cell lines and clones from CWF1 mice inhibit anti-DNA Ab production, some promote autoantibody production. The resting T cell lines or clones that were established by subcloning short-term lines (shown in B) were cocultured with A6H31–45 and BWF1 splenic T plus B cells. Results are expressed as the mean IgG anti-DNA AFC per 106 B cells. The first bar on the left (BW B plus T cells without any added peptide) represents the background AFC number. The horizontal dashed line crossing this figure denotes one-half of the background AFC number. We arbitrarily used this cut off to label the T cell lines as inhibitory (▪), which included four CD4+ and nine CD8+ T cell lines. Two T cell lines, both CD4+, increased anti-DNA AFC 2-fold above the background (□, labeled arbitrarily as helper T cell lines). The remaining T cell lines that included 15 CD4+ and 12 CD8+ T cell lines had no or <2-fold effect on AFC (▨). Data are representative of two separate experiments.

Close modal

The BWF1 mice, which are the F1 intercross of NZB (H-2d) and NZW (H-2z) mice, develop an autoantibody-mediated disease that shares many features with human lupus (2, 4). The autoantibody production in these mice is mostly dependent on MHC class II-restricted TCRαβ+ cells (2, 3, 4). To determine whether F1 mouse strains that express MHC class II molecules (H-2dxz) identical to those of BWF1 mice develop a lupus-like disease, we crossed BALB/c (H-2d) with NZW mice to generate CWF1 mice. Since NZW class II molecules have an identical sequence to the H-2u haplotype (21), we crossed NZB with B10.Pl (H-2u = z) to generate another F1 strain, BPF1, that expresses H-2dxu = dxz. Twenty BWF1, 20 CWF1, and 32 BPF1 female mice were monitored for serum IgG anti-DNA Ab and proteinuria up to 1 year of age. All BWF1 mice had anti-DNA Ab by 15 wk of age, proteinuria by 28 wk, and died by 42 wk. In contrast, all CWF1 and BPF1 mice appeared to be healthy during the study period; <10% had anti-DNA Ab, but none had proteinuria. A similar observation in BPF1 mice has been reported by others (21).

To confirm the absence of IgG anti-DNA Ab-secreting B cells in naive nonautoimmune mice, we cultured spleen cells from CWF1, BPF1, or BWF1 mice for 8–12 h in microtiter plates coated with DNA and estimated the numbers of IgG anti-DNA AFC by ELISPOT. Less than 10 AFC per 106 B cells were detected in CWF1 or BPF1 mice, while BWF1 cultures had 180 AFC (Fig. 1, T plus B). Similar results were obtained when spleen cells from these mice were cultured for 5 days and supernatants were tested for anti-DNA Ab in an ELISA.

FIGURE 1.

B cells from nonautoimmune mice that express identical MHC molecules to those of lupus-prone BWF1 mice do not spontaneously secrete IgG anti-DNA Ab. Splenic B cells pooled from 15- to 20-wk-old, naive nonautoimmune (CWF1 and BPF1) or lupus-prone BWF1 mice (n = 3–5) were cultured with autologous T cells and without (T plus B) or with VH-derived T cell determinant (A6H34–45, A6H58–69, and A6H84–95) or control (A6H93–107) peptides (Table I). IgG anti-DNA AFC were enumerated by ELISPOT assay. Results are expressed as mean AFC number per 106 B cells in triplicate wells; deviations from the mean were <12%. Anti-DNA AFC in cultures containing B cells alone, with or without added peptides ranged from 20 to 50, 0, and <5 per 106 B cells in BWF1, CWF1, and BPF1, respectively (data not shown). Results from one of four similar experiments are shown here.

FIGURE 1.

B cells from nonautoimmune mice that express identical MHC molecules to those of lupus-prone BWF1 mice do not spontaneously secrete IgG anti-DNA Ab. Splenic B cells pooled from 15- to 20-wk-old, naive nonautoimmune (CWF1 and BPF1) or lupus-prone BWF1 mice (n = 3–5) were cultured with autologous T cells and without (T plus B) or with VH-derived T cell determinant (A6H34–45, A6H58–69, and A6H84–95) or control (A6H93–107) peptides (Table I). IgG anti-DNA AFC were enumerated by ELISPOT assay. Results are expressed as mean AFC number per 106 B cells in triplicate wells; deviations from the mean were <12%. Anti-DNA AFC in cultures containing B cells alone, with or without added peptides ranged from 20 to 50, 0, and <5 per 106 B cells in BWF1, CWF1, and BPF1, respectively (data not shown). Results from one of four similar experiments are shown here.

Close modal

Autoreactive, anti-VH, Th cells that recognize peptides processed from the VH of anti-DNA Ig molecules (4, 20, 22, 23, 24, 25, 26, 27) arise spontaneously in humans and mice with systemic lupus erythematosus (6, 20, 28) and modulate the development and progression of disease in lupus mice (4, 6, 20, 29, 30). These MHC class II-restricted, anti-VH Th cells increase anti-DNA Ab production by BWF1 B cells (15). To determine whether such anti-VH Th cells exist in nonautoimmune mice, splenic T and B cells from naive CWF1, BPF1, or BWF1 mice were cultured with VH peptides (Table I), and IgG anti-DNA Ab production was estimated by ELISPOT (Fig. 1). In the presence of T cell stimulatory peptides, the anti-DNA AFC significantly increased in BWF1, but not in CWF1 or BPF1 mice. Thus, in contrast to BWF1 mice, MHC class II-identical CWF1 and BPF1 mice do not exhibit spontaneous activation of DNA-specific B and anti-VH Th cells. These cells may either be absent, ignorant, or tolerant in these mice.

In four separate experiments, CWF1 or BPF1 B cells were cultured with BWF1 T cells and vice versa. Strikingly, CWF1 and BPF1 B cells were induced to produce IgG anti-DNA Ab in the presence of BWF1 T cells (Fig. 2,A, left panel). In contrast, T cells from the nonautoimmune mice did not promote anti-DNA Ab production by BWF1 B cells (Fig. 2 A, right panel). This indicates that DNA-specific B cells exist in nonautoimmune mice and can be activated to secrete IgG anti-DNA Ab by autoimmune Th cells in vitro.

FIGURE 2.

B cell self-tolerance can be disrupted in nonautoimmune mice, as demonstrated in vitro (A), ex vivo (B), and in vivo (C). A, Autoimmune Th cells activate nonautoimmune mouse B cells in vitro. T and B cells were separately pooled from 20-wk-old mice (n = 3–12) of different strains. BWF1 T cells were incubated with B cells from CWF1 or BPF1 mice (left panel) and vice versa (right panel). T cell determinant or control peptides (Table I) were added to the cultures. Results are expressed as the mean triplicate IgG anti-DNA AFC number per 106 B cells; deviations from the mean were <15%. The AFC detected in BWF1 T plus CWF1 or BPF1 B cell cultures were not due to contaminating BWF1 B cells in purified T cell populations, because anti-DNA AFC in cultures containing BWF1 T cell alone without or with added stimulatory peptides were <10 per 106 B cells (data not shown). Data are representative of four independent experiments. B, Th cells that help anti-DNA Ab production could be induced or recalled in nonautoimmune mice ex vivo. Splenic T cells from A6.1-immunized CWF1 mice (n = 12) were cultured with syngeneic B cells without (‘background’) or with 23 overlapping peptides from the A6.1 VH. Results are shown as the mean triplicate IgG anti-DNA AFC per 106 B cells from one of two similar experiments. A horizontal dashed line crossing the figure denotes AFC number 2 × background. C, Six CWF1 mice immunized as above were bled to determine serum anti-DNA Ab. Results are shown as the mean triplicate OD from individual mice. All immunized mice (•) had high levels of anti-DNA Ab, whereas none of seven age-matched unimmunized controls (○) had OD values above the negative controls (BALB/c, mean OD = 0.03). The mean positive control OD in 1/25 diluted BWF1 sera was 1.9. Data are representative of two independent experiments.

FIGURE 2.

B cell self-tolerance can be disrupted in nonautoimmune mice, as demonstrated in vitro (A), ex vivo (B), and in vivo (C). A, Autoimmune Th cells activate nonautoimmune mouse B cells in vitro. T and B cells were separately pooled from 20-wk-old mice (n = 3–12) of different strains. BWF1 T cells were incubated with B cells from CWF1 or BPF1 mice (left panel) and vice versa (right panel). T cell determinant or control peptides (Table I) were added to the cultures. Results are expressed as the mean triplicate IgG anti-DNA AFC number per 106 B cells; deviations from the mean were <15%. The AFC detected in BWF1 T plus CWF1 or BPF1 B cell cultures were not due to contaminating BWF1 B cells in purified T cell populations, because anti-DNA AFC in cultures containing BWF1 T cell alone without or with added stimulatory peptides were <10 per 106 B cells (data not shown). Data are representative of four independent experiments. B, Th cells that help anti-DNA Ab production could be induced or recalled in nonautoimmune mice ex vivo. Splenic T cells from A6.1-immunized CWF1 mice (n = 12) were cultured with syngeneic B cells without (‘background’) or with 23 overlapping peptides from the A6.1 VH. Results are shown as the mean triplicate IgG anti-DNA AFC per 106 B cells from one of two similar experiments. A horizontal dashed line crossing the figure denotes AFC number 2 × background. C, Six CWF1 mice immunized as above were bled to determine serum anti-DNA Ab. Results are shown as the mean triplicate OD from individual mice. All immunized mice (•) had high levels of anti-DNA Ab, whereas none of seven age-matched unimmunized controls (○) had OD values above the negative controls (BALB/c, mean OD = 0.03). The mean positive control OD in 1/25 diluted BWF1 sera was 1.9. Data are representative of two independent experiments.

Close modal

Next, we determined whether autoreactive Th cells break tolerance in DNA-specific B cells in vivo and induce lupus-like disease in nonautoimmune mice. First, immunization with a mAb anti-DNA, A6.1, or its γ-chain or VH epitopes (listed in Table I) in adjuvant induced VH peptide-specific T cell proliferative or Ab responses in CWF1 mice (data not shown), suggesting that anti-VH Th cell repertoire exists in nonautoimmune mice. To determine whether these in vivo-primed anti-VH Th cells can promote anti-DNA Ab production, 20-wk-old CWF1 mice (n = 12) were immunized with A6.1 in CFA and boosted twice at 2-wk intervals with A6.1/IFA. Splenic T cells from these mice were cultured with syngeneic B cells without or with 23 overlapping 15-mer peptides representing the entire VH sequence of the A6.1 (Fig. 2,B). Results show that three peptides (peptides 7–9) increased anti-DNA AFC 4- to 5-fold above the background. Importantly, all immunized CWF1 mice developed high levels of serum anti-DNA Ab (Fig. 2 C) and most developed renal involvement. Three of 20 mice had a severe nephritis (proteinuria, 300 mg/dl with increased blood urea nitrogen (80 mg/dl) and creatinine (1.8 mg/dl)); 14 developed a mild disease (proteinuria, 100 mg/dl with normal blood urea nitrogen (<20 mg/dl) and creatinine (<0.6 mg/dl); and 3 remained normal. Injection of a control, nonautoreactive mAb, MOPC-21, did not induce IgG anti-DNA Ab or proteinuria (Ref. 3 and data not shown). This suggests that potentially autoreactive B cells in nonautoimmune mice can be activated to secrete pathogenic IgG anti-DNA Ab by appropriate Th cells.

To exclude the possibility that serum anti-DNA titers in A6.1-immunized mice (Fig. 2 C) were simply coming from the injected IgG2a mAb, we showed that the IgG anti-DNA Abs of all IgG isotypes were detected in the immunized CWF1 mice (data not shown).

The induction of serum IgG anti-DNA Ab and mild proteinuria upon immunizations with A6.1 prompted us to test whether persistent anti-DNA response and severe nephritis would be induced following chronic immunization of CWF1 mice. Serum anti-DNA Ab levels did increase between immunizations two and four, and 75% (8 of 12) of the mice developed proteinuria (100–300 mg/dl) with normal to mildly elevated plasma creatinine (mean ± SD, 0.9 ± 0.3). Thereafter, serum anti-DNA levels decreased (Fig. 3,A). All immunized mice slowly recovered from disease by the sixth immunization, at which time all had ≤30 mg/dl proteinuria and blood urea nitrogen and creatinine in the normal range. Subsequent immunizations did not induce a vigorous anti-DNA Ab response (Fig. 3 A) or proteinuria. This suggests that, although CWF1 mice are susceptible to breakdown of B cell tolerance, this loss of tolerance is self-limited.

FIGURE 3.

The breakdown of B cell self-tolerance in nonautoimmune mice is transitory: Nonautoimmune mice lose ability to make IgG anti-DNA Ab despite multiple immunizations. A, Ten 10-wk-old CWF1 mice were immunized with A6.1 in CFA and boosted every 2–4 wk with A6.1/IFA. These (▪) and 10 age-matched unimmunized controls (□) were monitored for serum anti-DNA Ab. Results are shown as the mean ± SD of OD at 1/25 serum dilution. After the second and fourth immunizations, the immunized mice had increased anti-DNA Ab compared with the controls (@, p < 0.05; $, p < 0.001). However, anti-DNA Ab in 8 of 10 mice returned to the background level when tested after the sixth immunization (∗, p < 0.01 compared with anti-DNA Ab level after the fourth immunization) and remained low despite subsequent immunizations (∗∗, p < 0.05). B, Fifteen young (12-wk-old) CWF1 mice were immunized with A6.1/CFA and boosted every 2–4 wk with A6.1/IFA. Five mice each were sacrificed after the second, third, and fifth immunizations. Their pooled splenic T and B cells were cultured with 23 overlapping peptides from the VH of A6.1 mAb for 5 days. Supernatants were tested for IgG anti-DNA Ab. Results are expressed as the mean triplicate OD from one of two similar experiments.

FIGURE 3.

The breakdown of B cell self-tolerance in nonautoimmune mice is transitory: Nonautoimmune mice lose ability to make IgG anti-DNA Ab despite multiple immunizations. A, Ten 10-wk-old CWF1 mice were immunized with A6.1 in CFA and boosted every 2–4 wk with A6.1/IFA. These (▪) and 10 age-matched unimmunized controls (□) were monitored for serum anti-DNA Ab. Results are shown as the mean ± SD of OD at 1/25 serum dilution. After the second and fourth immunizations, the immunized mice had increased anti-DNA Ab compared with the controls (@, p < 0.05; $, p < 0.001). However, anti-DNA Ab in 8 of 10 mice returned to the background level when tested after the sixth immunization (∗, p < 0.01 compared with anti-DNA Ab level after the fourth immunization) and remained low despite subsequent immunizations (∗∗, p < 0.05). B, Fifteen young (12-wk-old) CWF1 mice were immunized with A6.1/CFA and boosted every 2–4 wk with A6.1/IFA. Five mice each were sacrificed after the second, third, and fifth immunizations. Their pooled splenic T and B cells were cultured with 23 overlapping peptides from the VH of A6.1 mAb for 5 days. Supernatants were tested for IgG anti-DNA Ab. Results are expressed as the mean triplicate OD from one of two similar experiments.

Close modal

We then demonstrate that the in vivo decrease in anti-DNA Ab temporally correlates with the disappearance of anti-VH Th responses ex vivo. At different time points after immunizations, CWF1 spleen cells were cultured with 23 overlapping A6.1 VH peptides. After the second and third immunizations, culture with several peptides increased anti-DNA Ab ≥2-fold of the background. Thereafter, none of the peptides elicited increases in anti-DNA production (Fig. 3 B).

Multiple mechanisms may contribute to the decrease in anti-DNA Ab production after repeated immunizations: 1) induction of inhibitory or regulatory T cells or cytokines; 2) induction of anergy or deletion in anti-VH Th cells; 3) induction of regulatory anti-idiotypic Ab; and 4) development of Ab against A6.1 and formation of A6.1-anti-A6.1 immune complexes that could alter Ag processing and presentation. Although we have not excluded the relative contribution of each of these possibilities, we demonstrate several lines of evidence supporting the activation of inhibitory T cells that down-regulate anti-DNA Ab production.

First, we demonstrate that the freshly isolated T cells from the “recovered” CWF1 mice decrease anti-DNA Ab production by BWF1 spleen cells in a peptide-specific manner (Fig. 4,A). This decrease in AFC was not due to a changed proportion of T plus B cells or crowding, since addition of the same numbers of BWF1 T cells or CWF1-derived, HEL106–116-reactive T cells to BWF1 T plus B cell cultures did not affect anti-DNA AFC numbers (Fig. 4 A and data not shown).

Second, we asked whether a similar immunization regimen would activate inhibitory T cells in lupus mice (Fig. 4,B). In two independent experiments, 8–10 short-term T cell lines were established from pooled spleen cells of 5 each of CWF1 or BWF1 mice that were immunized with A6H31–45 (stimulatory peptide, no. 7 in Fig. 2,B) five times at 2-wk intervals. Three sets of independent, resting short-term T cell lines (in varying numbers from 102 to 105) were cocultured with 105–106 B cells from 26- to 32-wk-old unimmunized BWF1 mice. Results show that while CWF1-derived T cell lines decreased anti-DNA production by BWF1 B cells in a dose (T cell number)-dependent manner, BWF1-derived T cell lines increased anti-DNA Ab production (Fig. 4 B). This suggests that the development of inhibitory T cells may be inadequate or defective in BWF1 mice.

Third, 42 long-term lines or clones were established by subcloning CWF1-derived short-term T cell lines and tested for the ability to influence anti-DNA Ab production by BWF1 B cells (Fig. 4,C). Although only 2 (5%) T cell lines increased anti-DNA AFC numbers by 2-fold, 13 (31%) T cell lines decreased the AFC numbers to <50% of the background. The remaining T cell lines had no or <2-fold effect on anti-DNA Ab production (Fig. 4 C). This distribution of the effect of T cell lines on anti-DNA Ab production did not match a normal distribution (p = 0.003, Shapiro-Wilk test). Thus, both inhibitory T and Th cells coexist in CWF1 mice.

Although most (16 of 20 tested (80%)) inhibitory cell T lines suppressed anti-DNA AFC only in the presence of A6H31–45, some (20%) inhibitory T cell lines were activated in the presence of control peptides (HYHEL31–45 or A6H110–121) or even without any added peptide. To test the specificity of inhibition of anti-DNA Ab production by inhibitory T cells, an inhibitory T cell line (CWF1 no. 3) or purified T cells from CWF1 mice were cultured for 5 days with three BWF1-derived B cell hybridomas—A6.1 that expresses anti-DNA Ig that contains the A6H31–45 sequence and 375–57 and 375–100 that express non-DNA binding Ig that do not contain the A6H31–45 sequence. Results show that inhibitory T cells decreased total Ig production by A6.1 by 40–60%, but not by 375–57 and 375–100.

TGFβ is an immunoregulatory cytokine that can inhibit autoantibody production (31, 32). To determine whether inhibitory T cells selectively secrete this cytokine, we cultured APCs with peptide and eight T cell lines each from CWF1 and BWF1 mice and measured TGFβ in culture supernatants (Fig. 5 A). Collectively, CWF1-derived T cell lines secreted higher levels of TGFβ than BWF1-derived T cell lines (p < 0.05, Student’s t test). Essentially similar results were obtained with another set of T cell lines. Interestingly, all CD8+ inhibitory T cell lines tested secreted 2- to10-fold higher amounts of TGFβ when compared with all CD4+ T cell lines from CWF1 or BWF1 mice.

FIGURE 5.

TGFβ1 may partly mediate the inhibition of anti-DNA Ab production by inhibitory T cells. A, Some inhibitory T cells secrete TGFβ1. Resting T cell lines (n = 8 each) from CWF1 or BWF1 mice were cultured with APC and A6H31–45. Results are expressed as the mean total TGFβ1 (picograms per milliliter) in culture supernatants. CWF1-derived CD8+ inhibitory T cell lines (CWF1#1 to CWF1#4) secrete 2- to10-fold more TGFβ than syngeneic CD4+ inhibitory T cell lines (CWF1#5–CWF1#8) or CD4+ and CD8+ T cell lines (BWF1#9–16) from BWF1 mice. B, Inhibition of anti-DNA Ab production by CWF1-derived T cells is abolished by anti-TGFβ Ab. Splenic B plus T cells from six BWF1 mice were cultured in microtiter plates coated with DNA (□). Results are shown as the mean ± SE of IgG anti-DNA AFC per 106 B cells from representative one of two independent experiments. Addition of peptide A6H31–45 increased anti-DNA AFC (▤, p < 0.001), but the AFC numbers decreased below the background level when T cells from eight times immunized CWF1 mice (n = 5) were added to these cultures (▪; ∗, p < 0.0001). This inhibition of anti-DNA Ab production by CWF1 T cells was abolished by the addition of an anti-TGFβ mAb (1 μg/ml 1.D.11.16, ▨; ∗∗, p < 0.001 as compared with the ▪ or ▥ (an isotype-matched control rat IgG1)).

FIGURE 5.

TGFβ1 may partly mediate the inhibition of anti-DNA Ab production by inhibitory T cells. A, Some inhibitory T cells secrete TGFβ1. Resting T cell lines (n = 8 each) from CWF1 or BWF1 mice were cultured with APC and A6H31–45. Results are expressed as the mean total TGFβ1 (picograms per milliliter) in culture supernatants. CWF1-derived CD8+ inhibitory T cell lines (CWF1#1 to CWF1#4) secrete 2- to10-fold more TGFβ than syngeneic CD4+ inhibitory T cell lines (CWF1#5–CWF1#8) or CD4+ and CD8+ T cell lines (BWF1#9–16) from BWF1 mice. B, Inhibition of anti-DNA Ab production by CWF1-derived T cells is abolished by anti-TGFβ Ab. Splenic B plus T cells from six BWF1 mice were cultured in microtiter plates coated with DNA (□). Results are shown as the mean ± SE of IgG anti-DNA AFC per 106 B cells from representative one of two independent experiments. Addition of peptide A6H31–45 increased anti-DNA AFC (▤, p < 0.001), but the AFC numbers decreased below the background level when T cells from eight times immunized CWF1 mice (n = 5) were added to these cultures (▪; ∗, p < 0.0001). This inhibition of anti-DNA Ab production by CWF1 T cells was abolished by the addition of an anti-TGFβ mAb (1 μg/ml 1.D.11.16, ▨; ∗∗, p < 0.001 as compared with the ▪ or ▥ (an isotype-matched control rat IgG1)).

Close modal

We then determined whether TGFβ secreted by CWF1-derived T cells contributes to the inhibition of autoantibody production. First, we show that splenic T cells from the hyperimmunized CWF1 mice when cocultured with BWF1 T plus B cells decreased the IgG anti-DNA AFC number. The addition of an anti-TGFβ mAb (1.D.11.16; Celltrix, Santa Clara, CA), but not an isotype-matched control Ig, to these cultures abolished the inhibition of anti-DNA Ab (Fig. 5 B). The TGFβ effect on the inhibition of IgG anti-DNA Ab production was not associated with the isotype skewing of anti-DNA Ab toward IgA (data not shown).

So far, we have shown that primed T cells from CWF1 mice suppress anti-DNA Ab production. To determine whether such inhibitory T cells exist in naive mice, we cultured BWF1 spleen cells for 5 days and tested supernatant for anti-DNA Ab by ELISA. To these cultures, addition of T cells from 10-wk-old, naive CWF1 mice significantly decreased anti-DNA Ab; this inhibition by naive T cells was also partially abrogated in the presence of an anti-TGFβ mAb (Fig. 6).

FIGURE 6.

T cells from naive CWF1 mice inhibit IgG anti-DNA Ab production, which is also abrogated by anti-TGFβ Ab. BWF1 spleen cells (5 × 106/ml from six 35-wk-old mice) were cultured without or with naive CWF1 T cells (105/ml from six 10-wk-old mice) and an isotype-matched control rat IgG1 or an anti-TGFβ mAb for 5 days. IgG anti-DNA Ab levels in culture supernatants are expressed as the mean ± SE of OD from one representative of six independent experiments. Addition of naive CWF1 T cells to BWF1 spleen cell cultures decreased anti-DNA Ab levels (∗, p < 0.01). The inhibition of anti-DNA Ab production was partially abrogated in the presence of an anti-TGFβ mAb (∗∗, p < 0.05 ▨ vs □ or ▪).

FIGURE 6.

T cells from naive CWF1 mice inhibit IgG anti-DNA Ab production, which is also abrogated by anti-TGFβ Ab. BWF1 spleen cells (5 × 106/ml from six 35-wk-old mice) were cultured without or with naive CWF1 T cells (105/ml from six 10-wk-old mice) and an isotype-matched control rat IgG1 or an anti-TGFβ mAb for 5 days. IgG anti-DNA Ab levels in culture supernatants are expressed as the mean ± SE of OD from one representative of six independent experiments. Addition of naive CWF1 T cells to BWF1 spleen cell cultures decreased anti-DNA Ab levels (∗, p < 0.01). The inhibition of anti-DNA Ab production was partially abrogated in the presence of an anti-TGFβ mAb (∗∗, p < 0.05 ▨ vs □ or ▪).

Close modal

Similar findings were observed in a separate experiment where T cells from 35- to 45-wk-old naive or 8 to 12 times immunized CWF1 mice were cocultured with T plus B cells from 30- to 40-wk-old BWF1 mice (n = 12 each). BWF1 T plus B cell cultures had 240 ± 117 (mean ± SD) anti-DNA AFC per 106 B cells; this number was decreased in the presence of naive (98 ± 34, p < 0.01) or primed (24 ± 11, p < 0.0001) CWF1 T cells. Thus, naive CWF1 T cells also inhibit, albeit to a smaller extent than primed T cells (p < 0.01, naive vs primed), anti-DNA Ab production.

We hypothesized that the tendency to develop TGFβ-secreting inhibitory T cells is due to the intrinsic ability of nonautoimmune mice to make more active TGFβ in vivo. To test this, we determined serum levels of active and total TGFβ. Serum-active TGFβ1 was significantly more in young CWF1 mice than in BWF1 mice (Fig. 7). Total TGFβ1 levels, however, were similar in the two strains; thus, the mean ratio of active:total TGFβ was significantly higher in CWF1 (0.11 ± 0.016) than in BWF1 (0.05 ± 0.009) mice (p = 0.02, n = 6 each).

FIGURE 7.

In vivo TGFβ levels in nonautoimmune mice. Serum-active TGFβ levels were measured in young (8- to 14-wk-old) CWF1 and age- and sex-matched BWF1 mice (n = 6 each). Results are expressed as the mean ± SE (picograms per milliliter; ∗, p < 0.05). Data are from one representative of three independent experiments.

FIGURE 7.

In vivo TGFβ levels in nonautoimmune mice. Serum-active TGFβ levels were measured in young (8- to 14-wk-old) CWF1 and age- and sex-matched BWF1 mice (n = 6 each). Results are expressed as the mean ± SE (picograms per milliliter; ∗, p < 0.05). Data are from one representative of three independent experiments.

Close modal

In this article, we show that in nonautoimmune mice, DNA-specific B cells can be activated in vitro by coculture with anti-VH Th cells or in vivo by immunization with an anti-DNA mAb. Such in vivo breakdown of tolerance caused a self-limited autoimmune disease in these mice. T cells from the recovered mice inhibited anti-DNA Ab production and secreted TGFβ. Addition of anti-TGFβ Ab to the cultures partly abrogated the ability of regulatory T cells to inhibit anti-DNA Ab production.

In nonautoimmune mice bearing transgenes that encode anti-DNA Ig, DNA-specific B cells accumulate at the T-B interface of the splenic follicle, whereas in lupus-prone MRL/lpr mice the anti-DNA B cells are present in the follicle, thus allowing T-B cell interaction (10, 11, 12). Anti-DNA B cells from nonautoimmune transgenic mice proliferate in response to CD40 ligand and IL-4 (13), suggesting that B cell self-tolerance is due primarily to lack of availability of T cell help. Indeed, that is the case in our nontransgenic system: B cells from nonautoimmune mice could readily make IgG anti-DNA Ab when exposed to autoreactive Th cells from lupus mice or syngeneic Th cells that were activated following immunization with an anti-DNA mAb (Fig. 2). Such B cell activation induced a strong anti-DNA Ab response in vivo (Fig. 2 C) and caused renal disease. Thus, potentially pathogenic autoreactive B cells that may have escaped clonal deletion exist in nonautoimmune mice.

Although anti-DNA Ab production by CWF1 mice was readily induced by immunizations that activated Th cells (Fig. 2,B), this effect was short-lived, as anti-DNA Ab decreased despite further immunizations (Fig. 3). It is possible to explain this decrease in anti-DNA Ab response in nonautoimmune mice by the emergence of T cells that decreased anti-DNA Ab in these mice. Remarkably, adoptive transfer of such inhibitory T cells into BWF1 mice that are genetically prone to develop lupus resulted in decreased serum anti-DNA Ab levels, delayed onset of nephritis, and improved survival (our unpublished data).

Although the mechanisms by which inhibitory T cells arise and inhibit autoimmunity are not well defined, our studies offer some interesting clues. Most Th and inhibitory T cells appear to recognize the same peptides (Figs. 2,B and 4C). Analogous findings have been reported in other models (33, 34, 35). For instance, the same self-MHC-reactive T cells helped as well as suppressed B cell Ig production under different conditions (33). The different conditions, in our case, could be created by differences in the in vivo TGFβ levels between nonautoimmune CWF1 and lupus-prone BWF1 mice. Naive, young CWF1 mice that exhibit a dominant inhibitory T cell response had more active TGFβ than BWF1 mice that develop a mostly Th response (Figs. 4,B and 7). Such a cytokine milieu might contribute to the development of inhibitory T cells in vivo. An exposure of CD8+ T cells to TGFβ can condition these cells to become suppressors of Ab production (36). TGFβ selectively activates CD8+ T cells to proliferate and augments maturation of naive to memory T cells (37). The regulatory role of TGFβ (33, 36, 38) was confirmed in our experiments in which anti-TGFβ Ab abrogated, at least in part, the T cell-mediated inhibition of anti-DNA Ab production (Fig. 5 B).

Although, TGFβ-mediated suppression appears to be a major mechanism in our model (Fig. 5), it does not represent the sole mechanism of regulation of autoantibody production. Although the inhibition of anti-DNA Ab production by CD8+ inhibitory T cell lines was markedly inhibited by an anti-TGFβ mAb (Fig. 5 B), there was no or minimal effect of anti-TGFβ mAb on the inhibition of autoantibody production by CD4+ inhibitory T cell lines (data not shown).

Most Th and inhibitory T cells recognized a peptide, A6H31–45 that contains at least two epitopes: one epitope binds an MHC class II molecule, I-Ed, and activates CD4+ Th cells that promote anti-DNA Ab production in BWF1 mice (4), and the other epitope binds an MHC class I molecule, Kd, and activates CD8+ T cells that inhibit anti-DNA Ab production.4 The 15-mer peptide, A6H31–45, activated Th cells in both BWF1 (Fig. 4,B) and CWF1 mice (Figs. 2,B, 3B, and 4C), but activated inhibitory T cells only in CWF1 mice (Figs. 4, A–C, and 5B). A shorter version of this peptide, a 12-mer A6H34–45 that was used in initial experiments (Figs. 1 and 2,A), also contains the same helper and inhibitory epitopes. It remains to be determined whether a differential processing of the VH molecule into the helper vs inhibitory epitopes are responsible for the predominant helper response in BWF1 mice vs prominent inhibitory response in CWF1 mice, respectively, shown in Fig. 4 B.

Numerous examples illustrate that inhibitory T cells recognize molecules involved in Ag recognition (33, 39, 40, 41). TCR-reactive T cells arise during recovery from experimental autoimmune encephalomyelitis and suppress it (39); MHC-reactive T cells protect rats from autoantibody-mediated disease (40); and Ig-reactive T cells confer resistance to induction of autoimmune keratitis (41). In our experiments, Ig peptide-reactive inhibitory T cells protect mice from developing persistent autoantibody-mediated disease (Fig. 3 A and data not shown).

Finally, the development of inhibitory T cells did not appear to be an artifact of multiple immunizations, as even without immunization splenic T cells from CWF1 mice inhibit autoantibody production, albeit to a much smaller extent than T cells from immunized CWF1 mice (Fig. 6). Thus, it appears that the normal immune repertoire is composed of potentially autoreactive Th as well as inhibitory T cells (Fig. 4 C) (35, 42, 43) and inhibitory T cells expand whenever there is an impending danger of pathologic autoimmunity.

In conclusion, inhibitory T cells may normally prevent activation of autoreactive B cells and/or reverse their activation following stimulation by Th cells. We suggest that the ability to develop inhibitory T cells may be one of many checkpoints that protect nonautoimmune mice from developing clinical autoimmunity. Such inhibitory T cells may be responsible for spontaneous regression or waxing and waning course of autoimmune diseases, while defective regulation may contribute to sustained autoimmunity and its consequences such as systemic lupus erythematosus.

We thank Jonathan Katz for critical reading of this manuscript, Suzanne Morris for helpful suggestions, and Jonathan Jacinto and Vivaik Tyagi for technical assistance.

1

This work was supported by grants from the National Institutes of Health (AR47322 and AR47363P&F to R.R.S., AI46776 and AR36834 to B.H.H., and AR47784 to E.H.G.), National Arthritis Foundation (to R.R.S.), Southern California Chapter of the Arthritis Foundation (to B.H.H.), and the Durough and Paxson Family Foundation (to B.H.H.).

3

Abbreviations used in this paper: AFC, Ab-forming cell; HEL, hen egg lysozyme.

4

G. C. Fan and R. R. Singh. Vaccination with minigenes encoding VH-derived MHC class I-binding epitopes activates cytotoxic T cells that ablate autoantibody-producing B cells and inhibit lupus. Submitted for publication.

1
Singh, R. R., A. N. Malaviya.
1988
. Dermato-polymyositis: a recent outbreak.
J. Assoc. Physicians India
36
:
678
2
Hahn, B. H..
1998
. Antibodies to DNA.
N. Engl. J. Med.
338
:
1359
3
Ohnishi, K., F. M. Ebling, B. Mitchell, R. R. Singh, B. H. Hahn, B. P. Tsao.
1994
. Comparison of pathogenic and nonpathogenic murine antibodies to DNA: antigen binding and structural characteristics.
Int. Immunol.
6
:
817
4
Singh, R. R., V. Kumar, F. M. Ebling, S. Southwood, A. Sette, E. E. Sercarz, B. H. Hahn.
1995
. T-cell determinants from autoantibodies to DNA can upregulate autoimmunity in murine lupus.
J. Exp. Med.
181
:
2017
5
Kaliyaperumal, A., C. Mohan, W. Wu, S. K. Datta.
1996
. Nucleosomal peptide epitopes for nephritis-inducing helper T cells of murine lupus.
J. Exp. Med.
183
:
2459
6
Singh, R. R..
2000
. Potential use of peptides and vaccination to treat systemic lupus erythematosus.
Curr. Opin. Rheumatol.
12
:
399
7
Craft, J., S. Peng, T. Fujii, M. Okada, S. Fatenejad.
1999
. Autoreactive T cells in murine lupus: origins and roles in autoantibody production.
Immunol. Res.
19
:
245
8
Erikson, J., M. Z. Radic, S. A. Camper, R. R. Hardy, C. Carmack, M. Weigert.
1991
. Expression of anti-DNA immunoglobulin transgenes in non-autoimmune mice.
Nature
349
:
331
9
Hartley, S. B., M. P. Cooke, D. A. Fulcher, A. W. Harris, S. Cory, A. Basten, C. C. Goodnow.
1993
. Elimination of self-reactive B lymphocytes proceeds in two stages: arrested development and cell death.
Cell
72
:
325
10
Mandik-Nayak, L., A. Bui, H. Noorchashm, A. Eaton, J. Erikson.
1997
. Regulation of anti-double stranded DNA B cells in nonautoimmune mice: localization to the T-B interface of the splenic follicle.
J. Exp. Med.
186
:
1257
11
Jacobson, B. A., D. J. Panka, K. A. Nguyen, J. Erikson, A. K. Abbas, A. Marshak-Rothstein.
1995
. Anatomy of autoantibody production: dominant localization of antibody-producing cells to T cell zones in Fas-deficient mice.
Immunity
3
:
509
12
Mandik-Nayak, L., S. J. Seo, C. Sokol, K. M. Potts, A. Bui, J. Erikson.
1999
. MRL-lpr/lpr mice exhibit a defect in maintaining developmental arrest and follicular exclusion of anti-double-stranded DNA B cells.
J. Exp. Med.
189
:
1799
13
Mandik-Nayak, L., S. J. Seo, A. Eaton-Bassiri, D. Allman, R. R. Hardy, J. Erikson.
2000
. Functional consequences of the developmental arrest and follicular exclusion of anti-double-stranded DNA B cells.
J. Immunol.
164
:
1161
14
Ebling, F. M., B. P. Tsao, R. R. Singh, E. Sercarz, B. H. Hahn.
1993
. A peptide derived from an autoantibody can stimulate T cells in the (NZB × NZW)F1 mouse model of systemic lupus erythematosus.
Arthritis Rheum.
36
:
355
15
Singh, R. R., B. H. Hahn, B. P. Tsao, F. M. Ebling.
1998
. Evidence for multiple mechanisms of polyclonal T cell activation in murine lupus.
J. Clin. Invest.
102
:
1841
16
Ando, D. G., E. E. Sercarz, B. H. Hahn.
1987
. Mechanisms of T and B cell collaboration in the in vitro production of anti-DNA antibodies in the NZB/NZW F1 murine SLE model.
J. Immunol.
138
:
3185
17
Ando, D. G., F. M. Ebling, B. H. Hahn.
1986
. Detection of native and denatured DNA antibody forming cells by the enzyme-linked immunospot assay.
Arthritis Rheum.
29
:
1139
18
Krishnan, M. R., T. N. Marion.
1998
. Comparison of the frequencies of arginines in heavy chain CDR3 of antibodies expressed in the primary B-cell repertoires of autoimmune-prone and normal mice.
Scand. J. Immunol.
48
:
223
19
Singh, R. R., B. H. Hahn, E. E. Sercarz.
1996
. Neonatal peptide exposure can prime T cells, and upon subsequent immunization induce their immune deviation: implications for antibody vs. T cell-mediated autoimmunity.
J. Exp. Med.
183
:
1613
20
Singh, R. R., F. M. Ebling, E. E. Sercarz, B. H. Hahn.
1995
. Immune tolerance to autoantibody-derived peptides delays development of autoimmunity in murine lupus.
J. Clin. Invest.
96
:
2990
21
Schiffenbauer, J., L. Wegrzyn, B. P. Croker.
1992
. Background genes mediate the development of autoimmunity in (NZB × PL/J)F1 or (NZB × B10.PL)F1 mice.
Clin. Immunol. Immunopathol.
62
:
227
22
Weiss, S., B. Bogen.
1991
. MHC class II-restricted presentation of intracellular antigen.
Cell
64
:
767
23
Singh, R. R., B. H. Hahn.
1998
. Reciprocal T-B determinant spreading develops spontaneously in murine lupus: implications for pathogenesis.
Immunol. Rev.
164
:
201
24
Wysocki, L. J., X. Zhang, D. S. Smith, C. M. Snyder, C. Bonorino.
1998
. Somatic origin of T-cell epitopes within antibody variable regions: significance to monoclonal therapy and genesis of systemic autoimmune diseases.
Immunol. Rev.
162
:
233
25
Brosh, N., E. Eilat, H. Zinger, E. Mozes.
2000
. Characterization and role in experimental systemic lupus erythematosus of T cell lines specific to peptides based on CDR-1 and CDR-3 of a pathogenic anti-DNA monoclonal antibody.
Immunology
99
:
257
26
Yurin, V. L., A. Y. Rudensky, S. M. Mazel, J. M. Blechman.
1989
. Immunoglobulin-specific T-B cell interaction. II. T cell clones recognize the processed form of B cells’ own surface immunoglobulin in the context of the major histocompatibility complex class II molecule.
Eur. J. Immunol.
19
:
1685
27
Zhang, X., D. S. Smith, A. Guth, L. J. Wysocki.
2001
. A receptor presentation hypothesis for T cell help that recruits autoreactive B cells.
J. Immunol.
166
:
1562
28
Williams, W. M., N. A. Staines, S. Muller, D. A. Isenberg.
1995
. Human T cell responses to autoantibody variable region peptides.
Lupus
4
:
464
29
Jouanne, C., S. Avrameas, B. Payelle-Brogard.
1999
. A peptide derived from a polyreactive monoclonal anti-DNA natural antibody can modulate lupus development in (NZB × NZW)F1 mice.
Immunology
96
:
333
30
Eilat, E., H. Zinger, A. Nyska, E. Mozes.
2000
. Prevention of systemic lupus erythematosus-like disease (NZB × NZW)F1 mice by treatment with CDR1 and CDR3-based peptides of pathogenic autoantibody.
J. Clin. Immunol.
20
:
268
31
Raz, E., J. Dudler, M. Lotz, S. M. Baird, C. C. Berry, R. A. Eisesenberg, D. A. Carson.
1995
. Modulation of the disease activity in murine systemic lupus erythematosus by cytokine gene delivery.
Lupus
4
:
286
32
Ohtsuka, K., J. D. Gray, M. M. Stimmler, B. Toro, D. A. Horwitz.
1998
. Decreased production of TGF-beta by lymphocytes from patients with systemic lupus erythematosus.
J. Immunol.
160
:
2539
33
Kitani, A., K. Chua, K. Nakamura, W. Strober.
2000
. Activated self-MHC-reactive T cells have the cytokine phenotype of Th3/T regulatory cell 1 T cells.
J. Immunol.
165
:
691
34
Lider, O., A. Miller, S. Miron, R. Hershkoviz, H. L. Weiner, X. M. Zhang, E. Heber-Katz.
1991
. Nonencephalitogenic CD4CD8 Vα2Vβ8.2+ anti-myelin basic protein rat T lymphocytes inhibit disease induction.
J. Immunol.
147
:
1208
35
Shevach, E. M..
2000
. Regulatory T cells in autoimmunity.
Annu. Rev. Immunol.
18
:
423
36
Gray, J. D., M. Hirokawa, D. A. Horwitz.
1994
. The role of transforming growth factor beta in the generation of suppression: an interaction between CD8+ T and NK cells.
J. Exp. Med.
180
:
1937
37
Rich, S., M. Seelig, H. M. Lee, J. Lin.
1995
. Transforming growth factor β1 costimulated growth and regulatory function of staphylococcal enterotoxin B-responsive CD8+ T cells.
J. Immunol.
155
:
609
38
Seddon, B., D. Mason.
1999
. Regulatory T cells in the control of autoimmunity: the essential role of transforming growth factor β and interleukin 4 in the prevention of autoimmune thyroiditis in rats by peripheral CD4+CD45RC cells and CD4+CD8 thymocytes.
J. Exp. Med.
189
:
279
39
Kumar, V., F. Aziz, E. Sercarz, A. Miller.
1997
. Regulatory T cells specific for the same framework 3 region of the Vβ8.2 chain are involved in the control of collagen II-induced arthritis and experimental autoimmune encephalomyelitis.
J. Exp. Med.
185
:
1725
40
Bridoux, F., A. Badou, A. Saoudi, I. Bernard, E. Druet, R. Pasquier, P. Druet, L. Pelletier.
1997
. Transforming growth factor β (TGF-β)-dependent inhibition of T helper cell 2 (Th2)-induced autoimmunity by self-major histocompatibility complex (MHC) class II-specific, regulatory CD4+ T cell lines.
J. Exp. Med.
185
:
1769
41
Avery, A. C., Z.-S. Zhao, A. Rodriguez, E. K. Bikoff, M. Sohellian, C. S. Foster, H. Cantor.
1995
. Resistance to herpes stromal keratitis conferred by an IgG2a-derived peptide.
Nature
376
:
431
42
Olivares-Villagomez, D., A. K. Wensky, Y. Wang, J. J. Lafaille.
2000
. Repertoire requirements of CD4+ T cells that prevent spontaneous autoimmune encephalomyelitis.
J. Immunol.
164
:
5499
43
Seddon, B., D. Mason.
1999
. Peripheral autoantigen induces regulatory T cells that prevent autoimmunity.
J. Exp. Med.
189
:
877