Adoptive transfer of autoimmune splenic dendritic cells to lupus-prone mice triggers a B lymphocyte humoral response

Female lupus-prone [NZB×NZW]F1 (BWF1) mice were purchased from the Jackson Laboratory (Bar Harbor, ME, USA). All mice used in this study were housed in the animal facility of Fundación Ciencia & Vida. Animal work was carried out under the institutional regulations of the Fundación Ciencia & Vida and was approved locally by the ethical review committee of the Facultad de Ciencias, Universidad de Chile.

BWF1 female mice aged 2 months old represented young mice, while 8 -month-old mice with severe proteinuria (i.e., ≥500 mg/dl protein) and high antibody titers against double-stranded DNA (dsDNA) represented aged autoimmune mice. Age-matched [NZW×BALB/c]F1 female mice were used as controls.

Proteinuria was measured on a monthly basis during the first 6 months of age by a standard semi-quantitative test using a Combur Test N (Roche Diagnostics, Germany). After 6 months of age, proteinuria was measured every week to detect premature lupus. Autoantibodies against dsDNA were measured in serum samples by a standard ELISA using calf thymus DNA. Briefly, 650 ng/ml dsDNA was used to coat ELISA plates (Nalge Nunc International, USA) in an overnight incubation. Antigen-coated plates were subsequently blocked for 1 h with phosphate-buffered saline (PBS) containing 1.5% bovine serum albumin (BSA) and then incubated for 1 h at room temperature with sample sera (1:250 dilution). The plates were then washed with PBS-0.05% Tween 20 and incubated for 1 h with a peroxidase-labeled goat anti-mouse IgG antibody (Dako, USA). The color was developed by adding the TMB substrate kit (BD Bioscience, USA), and the absorbance at 450 nm (OD 450 nm) was measured using a plate reader (Jenway, UK).

Monoclonal antibodies (mAbs) against mouse CD86 FITC (GL1), CD138 PE (281-2), CD45R/B220 PE-Cy7 (RA3-6B2), CD4 PE (RM4-5), CD19 FITC or eFluor 780 (6D5), IL-10 PE (JES5-16E3), CD1d APC (1B1), CD69 (H1.2F3), IgM PE-Cy7 (RMM-1), purified CD16/32 (93), NK1.1 Alexa Fluor 488 (PK136), CD49b PE (DX5), CD11b APC (N1/70), and PDCA-1 APC (927) were purchased from BioLegend (San Diego, CA, USA). mAbs against mouse CD5 PE-Cy7 (53-7.3), CD11c PE (N418), IFN-γ FITC (XMG1.2), CD62L PE (MEL-14), CD25 APC (PC61.5), CD273 PE (PD-L2) (TY25), CD3 FITC (17A2), purified CD3 (145-2C11), and CD279 FITC (PD-1) (J43) were purchased from eBioscience (San Diego, CA, USA). mAbs against mouse IgD FITC (11-26c.2a), I-Ad FITC or APC (MHC-II) (AMS-32.1), CD79b FITC (HM79-12), and mouse anti-Armenian hamster IgG2/3 FITC (G70-204) were purchased from BD Pharmingen (San Diego, CA, USA). Peroxidase-labeled goat anti-mouse IgG antibody was purchased from Dako (USA).

Surface staining was performed in ice-cold PBS with 2% fetal calf serum (FCS) for 30 min in the presence of FcγR blocking antibody (CD16/32). 1.5 ng/μl propidium iodide (PI) (Sigma-Aldrich) was used for live-dead cell discrimination.

Intracellular staining was performed with the BD Cytofix/Cytoperm and Perm/Wash buffers. For intracellular IFN-γ staining, 1 × 10⁶ cells were cultured for 4 h at 37 °C in RPMI 1640 medium with 10% FCS containing 1 μg/ml ionomycin, 0.25 μM phorbol myristate acetate (PMA), and 10 μg/ml brefeldin A. For intracellular IL-10 staining, 1 × 10⁶ cells were cultured for 5 h at 37 °C in RPMI 1640 medium with 10% FCS containing 2 μg/ml lipopolysaccharide (LPS) (Sigma), 1 μg/ml ionomycin (Sigma), PMA (Sigma), and 10 μg/ml GolgiStop (BD Biosciences, USA). Viability dye eFluor 780 reagent (eBioscience) was used for live-dead cell discrimination.

Flow cytometry was conducted on a FACSCanto II flow cytometer (BD Biosciences), and data analysis was performed using the FlowJo software (Tree Star, Inc., Ashland, OR, USA).

Spleens of aged BWF1 and control [NZW×BALB/c]F1 mice were mechanically disaggregated. The cells were incubated for 45 min at 37 °C in a solution containing 1 mg/ml collagenase D (Roche) and 20 U/ml DNase I (Roche) dissolved in PBS supplemented with 2% FCS. Single cell suspensions were washed in RPMI 1640 medium and depleted of erythrocytes by incubation for 5 min with red blood cell (RBC) lysis buffer (BioLegend, USA) at 4 °C. Total CD11c⁺ cells were purified by cell sorting on a FACSAria II (BD Biosciences). Before cell sorting, T and B cells were eliminated by labeling the cells with a mixture of rat anti-mouse CD3 FITC plus Armenian hamster anti-mouse CD79b FITC. Then, cells were incubated with mouse anti-Armenian hamster IgG2/3 FITC, followed by incubation with Dynabeads coupled with anti-rat IgG and anti-mouse IgG (Invitrogen). Enriched cells were further stained with anti-CD3 and anti-CD79b antibodies to eliminate residual T or B cells and with an anti-CD11c antibody to select pure CD11c⁺ cells by cell sorting. The purity of cells was >98%, as determined by flow cytometry.

Two doses of 4 × 10⁶ splenic DCs from aged BWF1 (autoimmune DCs, H-2^dxz haplotype) or [NZW×BALB/c]F1 (control DCs, H-2^zxd haplotype) mice were injected intravenously (i.v.) into young healthy BWF1 mice or [NZW×BALB/c]F1 control mice within an interval of 20 days apart. Every 5 days, the mice were tested for proteinuria, and blood samples were taken to measure the anti-dsDNA autoantibody titers in the serum by ELISA. Flow cytometric analysis of the blood samples was conducted every 15 days to evaluate T and B cell phenotypes. Finally, at the end of 2 months, the mice were sacrificed, and DCs and T and B cells from the lymphoid organs were harvested and analyzed by flow cytometry. To determine which population of DCs is responsible for the induction of autoantibodies in young healthy BWF1 mice, we injected purified 0.6 × 10⁶ autoimmune splenic conventional or plasmacytoid DCs (cDCs or pDCs, respectively) in a single dose into young BWF1 mice and blood samples were taken to measure anti-dsDNA autoantibody titers in the serum by ELISA.

The cytokine levels were measured from the serum of mice using the cytometric bead array (CBA) assay. To detect inflammatory and T helper cell cytokines, we used the BD CBA Mouse Inflammation Kit and the CBA Mouse Th1/Th2/Th17 Kit according to the manufacturer’s instructions (BD Biosciences, USA).

Total splenic B cells from young BWF1 mice were isolated by negative selection using the B cell isolation kit from Miltenyi (Miltenyi Biotec, USA) following the manufacturer’s instructions. The B cell purity was always ≥95%, as determined by flow cytometry. Splenic CD4⁺CD25⁻ T cells from young BWF1 mice were sorted on a FACSAria II sorter (BD Biosciences). The purity of the cells was always ≥90%, as monitored by flow cytometry. Total splenic CD11c⁺ cells from aged BWF1 or control mice were sorted on a FACSAria II sorter (BD Biosciences) as described in section “Isolation of splenic DCs.” The purity of the DCs was >98%, as determined by flow cytometry.

DCs from aged BWF1 or control mice were co-cultured with young BWF1 B cells at a 1:5 ratio in 96-well U-bottomed plates in RPMI 1640 medium supplemented with 10% FCS and 0.5 μg/ml β-mercaptoethanol (Gibco, Life Technologies). The cells were cultured at 37 °C in a humidified 5% CO₂ incubator for 24 h or 3 days before flow cytometric analysis. Alternatively, DCs from aged BWF1 or control mice were co-cultured with T cells from a young BWF1 mice at a 1:2 ratio in 96-well U-bottomed plates in IMDM medium supplemented with 10% FCS, 0.5 μg/ml amphotericin B (Fungizone) (Gibco, Life Technologies), 0.5 μg/ml β-mercaptoethanol, and 50 μg/ml gentamicin (Gibco, Life Technologies) in the presence of 1 μg/ml of purified soluble anti-CD3 antibody (clone 2C11). The cells were cultured at 37 °C in a humidified 5% CO₂ incubator for 5 days to achieve T cell differentiation before flow cytometric analysis.

In light of published results [15, 17] demonstrating the abnormal function and phenotype of DCs in aged autoimmune BWF1 mice, we sought to determine the role of DCs in the development of lupus in the BWF1 murine model in vivo. For this, we adoptively transferred total splenic DCs (consisting of cDCs and pDCs) obtained from aged autoimmune BWF1 or control [NZW×BALB/c]F1 mice to young healthy BWF1 or control [NZW×BALB/c]F1 mice. The phenotype of transferred cells is shown in Suppl. Fig. 1a and b. When sorting spleen cells based on CD11c expression, the resulting population (over 98% of CD11c⁺ cells) contained a fraction of natural killer (NK) CD11c-expressing cells (CD11c⁺NK1.1⁺). On the other hand, the CD11c⁺NK1.1⁻ fraction contained pDCs (CD11c^intB220⁺) and cDCs (CD11c^hiB220⁻) in similar percentages when comparing control and autoimmune cells. Thus, transferred cells consisted of a known mixture of pDCs and cDCs. To assess the effect of adoptively transferred DCs on young healthy BWF1 mice, we evaluated the production of IgG autoantibodies against dsDNA in serum samples of these mice. Following the transfer of aged autoimmune BWF1 DCs at days 0 and 20, we observed high autoantibody production compared to the production induced by the transfer of control DCs (Fig. 1a). The maximum effect was observed 10 days after the second administration of DCs, showing that DCs from autoimmune BWF1 mice induce autoantibody secretion in young BWF1 mice a short period after adoptive transfer (Fig. 1a). Interestingly, the transfer of autoimmune DCs to control mice did not induce autoantibody production (Suppl. Fig. 2a), indicating that the genetic background is important for activating autoreactive B cells. Moreover, both pDCs and cDCs from autoimmune mice are sufficient to trigger the production of autoantibodies since the transfer of 0.6 × 10⁶ pDCs or cDCs to young healthy BWF1 mice induced autoantibody production (Suppl. Fig. 2b). Autoantibody production was not followed by renal damage, as proteinuria levels did not increase in either group of treated mice (data not shown).

In agreement with the autoantibody production, the percentage of CD19⁺CD138⁺ plasmablasts and CD19⁻CD138⁺ plasma cells increased in the peripheral blood of young, healthy BWF1 mice injected with autoimmune DCs (Fig. 1b). Together, these findings demonstrate that autoimmune DCs substantially promote the generation of autoantibodies, thereby triggering autoimmunity in young, previously healthy BWF1 mice.

To understand how autoimmune DCs drive humoral autoimmunity, we analyzed the percentage of naïve and long-lived memory B cells, defined as IgM⁺IgD⁻ cells by Pape and collaborators [18], in the spleens of mice sacrificed 60 days post injection of control or autoimmune DCs. As shown in Fig. 2a, the transfer of autoimmune DCs significantly reduces the percentage of IgM⁺IgD⁺ naïve B cells and increases the percentage of IgM⁺IgD⁻ long-lived memory B cells in the spleens of the recipient mice. Although not statistically significant, the absolute number of naïve or long-lived memory B cells follows the same trend (Suppl. Fig. 3a and b, respectively). Similar to what we found in the blood, the absolute cell numbers of both CD138⁺CD19⁺ plasmablasts and CD138⁺CD19⁻ plasma cells in the spleen increase following transfer of autoimmune DCs (Suppl. Fig. 4). In concordance with the loss of naïve B cells, the total B cells from the autoimmune DC-treated mice presented a higher activation state, as demonstrated by an increase in the expression of the activation markers CD86 and PD-L2 and a lower MHC-II expression, compared to control mice (Fig. 2b). Overall, these results indicate that autoimmune DCs alone are capable of triggering B cell differentiation into memory B cells in young, healthy BWF1 mice.

Although short-lived plasmablasts are a primary source of autoantibodies in various lupus mouse models, CD5⁺ B cells or B1-like cells have also been characterized as possible producers of autoantibodies [17, 19, 20]. CD5⁺ B cells are known for their capacity to produce polyreactive natural antibodies, which recognize autoantigens with low affinity. CD5⁺ B cells, which are found preferentially in the peritoneal cavity, are increased in the peripheral blood of SLE patients, and they are positively correlated with some autoantibodies detected in the serum [21]. To investigate whether autoimmune DCs promote the expansion of CD5⁺ B cells, we analyzed the presence of CD5⁺ B cells in the blood and spleens of BWF1 mice treated with control or autoimmune DCs and found that the frequency of CD5+ B cells in the blood increase 30 days after the transfer of autoimmune DCs (Fig. 3a). We also observed an increase in the percentage of CD5⁺ B cells in the spleens 60 days post transfer of autoimmune DCs compared to the transfer of control DCs (Fig. 3b). Besides, transfer of autoimmune DCs had a major effect on the expansion of CD5⁺ B cells, even higher to the one observed in aged BWF1 mice with lupus (Suppl. Fig. 5).

Patients with SLE have higher levels of serum IL-10 than healthy subjects [22]. Increased IL-10 may be attributed to the expansion of monocytes, B10 cells, and regulatory B cells and possibly to a subpopulation of memory T lymphocytes [23, 24]. To study whether the transfer of autoimmune DCs promotes the expansion of B10 cells, we evaluated IL-10 production using intracellular staining of splenic B cells from young BWF1 mice sacrificed 60 days post injection of control or autoimmune DCs. As shown in Suppl. Fig. 6a, the administration of autoimmune DCs had a potent effect on the expansion of splenic B10 cells compared to control DCs. Also, regulatory B cells, identified as CD1d^hiCD5⁺, were increased in autoimmune DC-treated mice (Suppl. Fig. 6b). Collectively, these findings demonstrate that adoptively transferred autoimmune DCs are involved in the expansion of CD5⁺ B cells and regulatory B cells in young, healthy BWF1 mice.

It has been reported that in aged autoimmune BWF1 mice, DCs accumulate in lymphoid organs, such as the spleen, mesenteric lymph nodes, and peripheral lymph nodes [15]. Interestingly, the adoptive transfer of autoimmune DCs to young healthy BWF1 mice generated a significant expansion of DCs in the blood and spleen (Fig. 4). Adoptive transfer of autoimmune DCs in young BWF1 mice mainly expanded cDCs in the blood while in the spleen, we observed an increase in both cDCs and pDCs. The expansion of DCs in the peripheral blood was observed starting on day 30 after the first injection, and the percentage of these cells doubled by the time the mice were sacrificed. Moreover, in Fig. 4, we can observe directly on the contour graphs that autoimmune DC-treated mice produce a larger subpopulation of cells that are negative for CD11c and express an intermediate level of B220 that correspond to CD5⁺ B cells present in the blood and spleen.

Next, we investigated the role of autoimmune DCs on the T cell response. BWF1 mice develop pathological proteinuria levels starting at 6 to 7 months of age, severe involution of the thymus and splenomegaly involving T cells, B cells, and DCs [25, 26]. We studied these characteristic phenomena of lupus in young, healthy BWF1 mice treated with control or autoimmune DCs. Involution of the thymus was not observed in autoimmune DC-treated mice, indicating that this may be a late symptom of lupus (Fig. 5a). On the other hand, mice treated with autoimmune DCs showed modest splenomegaly, as evaluated by the absolute cell count of total splenocytes (Fig. 5a). Although we found higher absolute CD4⁺ T cell numbers in the spleens of mice treated with autoimmune DCs than in those of control DC-treated mice (Fig. 5b), no differences were observed on the activation status of these cells, as assessed by CD25, CD62L, CD69, and PD-1 expression (Fig. 5c). Because peripheral blood mononuclear cells (PBMCs) from SLE patients produce large amounts of IFN-γ [22, 27], we evaluated whether autoimmune DCs could be involved in Th1 cell differentiation. Consistent with this, the frequency of IFN-γ-producing cells among CD4⁺ T cells derived from BWF1 mice treated with autoimmune DCs was higher than that in control DC-treated mice (Fig. 5d). Next, we analyzed serum obtained from the mice 60 days after the transfer of DCs. In agreement with the induction of IFN-γ-producing T cells, we observed that autoimmune DC transfer to young BWF1 mice induced a striking increase in IL-12, IL-6, and IL-10 compared to mice injected with control DCs (Fig. 5e). Thus, it can be concluded that autoimmune DCs contribute to the hyperproduction of IFN-γ by T cells and to the induction of a potent inflammatory cytokine production in diseased BWF1 mice.

To determine whether B cell activation and differentiation following the transfer of autoimmune DCs was due to direct interactions between DCs and B cells, we carried out in vitro experiments with purified subpopulations. For this, we co-cultured control or autoimmune DCs with purified splenic B cells from young BWF1 mice. After 24 h of co-culture, we observed that the autoimmune DCs were more efficient at inducing B cell maturation, as evidenced by a decreased percentage of IgM⁺IgD⁺ naïve B cells and an increase in IgM⁺IgD⁻ persistent memory B cells (Fig. 6a). Furthermore, although not statistically significant, the autoimmune DCs had a direct positive impact on B cell viability, which was noticeable after 3 days of co-culture (Fig. 6b). On the other hand, after 3 days of co-culture, we also observed a notable decrease in MHC-II expression on B cells activated by autoimmune DCs (Fig. 6c), replicating our observations in vivo (Fig. 2b). However, there were no differences in the expression levels of the activation markers CD86 and PD-L2 on B cells after co-culture with control or autoimmune DCs (Fig. 6c).

To determine whether autoimmune DCs affect T cell differentiation directly, we co-cultured autoimmune DCs with purified splenic CD4⁺ T cells from young BWF1 mice. After 5 days of polyclonal activation in the presence of anti-CD3, the autoimmune DCs were more efficient in differentiating splenic CD4⁺ T cells into IFN-γ-producing cells compared to control DCs (Fig. 6d). Thus, DCs from an autoimmune context contribute importantly to the hyperproduction of IFN-γ by CD4⁺ T cells.