Our study examined male and female animals. The data were analyzed without consideration of the sex of the animals. The control and experimental mice were age- and sex-matched whenever possible. The experiments were conducted in compliance with the German Animal Welfare Act and were authorized by the institutional committee on animal experimentation and the Government of Upper Bavaria.

To generate CD30stopfl mice, the cDNA of CD30 [19] was inserted into the Rosa26 locus via homologous recombination in embryonic stem (ES) cells. The CD30 transgene is under the control of the CAG promoter and a loxP flanked transcriptional/translational stop cassette, enabling cell type- and tissue-specific expression of the transgene via different Cre strains. At the 3′ priming end of the CD30 gene, a truncated form (without a signaling tail) of human hCD2 was inserted as a reporter gene, which is controlled by an internal ribosomal binding site (IRES), resulting in concomitant expression of CD30 and hCD2. To establish the mouse strain CD30stopfl, recombinant ES cells were injected into blastocytes. The resulting chimeras were backcrossed with BALB/c wild-type mice to establish the line. For expression of CD30 in all B cells, CD30stopfl mice were mated with CD19-Cre mice [47], and for germinal center-specific expression, Cγ1-Cre mice [22] were used, resulting in CD30stopfl/+//CD19-Cre+/– mice (CD30//CD19-Cre) and CD30stopfl/+//Cγ1-Cre+/– mice (CD30//Cγ1-Cre), respectively. As controls, we used CD19-Cre+/– (CD19-Cre) [47] for CD30//CD19-Cre mice and R26/CAG-CAR∆1StopF/+//Cγ1-Cre for CD30//Cγ1-Cre mice. The reporter strain R26/CAG-CAR∆1StopF was generated and characterized by the Marc Schmidt-Supprian laboratory [23]. The mouse strain contains the human coxsackie/adenovirus receptor CAR at the Rosa26 locus, preceded by a loxP-flanked STOP cassette (CAR//Cγ1-Cre). TD immunization leads to the simultaneous expression of CD30 and hCD2 in CD30//Cγ1-Cre mice and of CAR in CAR//Cγ1-Cre mice. CD19-Cre and Cγ1-Cre mice were kindly provided by Klaus Rajewsky, MDC Berlin, and R26/CAG-CAR∆1StopF mice were provided by Marc Schmidt-Supprian, TU Munich. All mouse lines were on a BALB/c background and housed in specific-pathogen-free environments.

All details about the different antibodies used in flow cytometry (FACS), enzyme-linked immunosorbent assay (ELISA), and histology are listed in Supplementary Table 1.

For surface staining, single-cell suspensions of the spleen were prepared, washed and adjusted to a concentration of 5 × 106 cells/ml in MACS buffer (Miltenyi Biotech) on ice. For each round of staining, 5 × 105 cells were used. To identify living cells, the cells were stained with a LIVE/DEAD Fixable Blue Dead Cell Stain Kit (Invitrogen) for 20 min in the dark on ice, followed by surface antibody staining for 20 min in the dark on ice. For intracellular FACS staining, the cells were first stained with the LIVE/DEAD Fixable Blue Dead Cell Stain Kit (Invitrogen) for 5 min on ice, fixed with 2% paraformaldehyde for 10 min at room temperature, and permeabilized in precooled methanol for 10 min on ice. The cells were then stained with antibodies for 1 h at room temperature. For staining with CD30-L and the respective isotype control, mouse FcR blocking reagent (Miltenyi) was added at the appropriate concentration according to the manufacturer´s protocol. Cytometry analysis was performed on an LSRII FACS Fortessa (BD Biosciences) coupled to BD FACS DIVA Software V8.0.1. The results were evaluated via FlowJo (v9 and v10).

MACS-purified splenic B cells were lysed in NP40 lysis buffer (150 mM NaCl; 50 mM Tris/HCl, pH 8; 1% Ipegal), separated on a 10% polyacrylamide SDS gel and transferred to a polyvinylidene fluoride (PVDF) membrane (ImmobilonTMP membrane). The membrane was stained with the anti-CD30 antibody sc-46683 (Santa Cruz).

To study T-cell-dependent immune responses, 8- to 16-week-old mice were injected intraperitoneally with 100 µg of alum-precipitated 4-hydroxy-3-nitrophenylacetyl-chicken-gamma-globulin (NP-CGG) (Biosearch Technologies, Novato, CA) in 200 µl of sterile PBS. The immune response of the mice was analyzed at the indicated time points postimmunization.

Antigen-specific ELISA and ELISA for total immunoglobulin were performed as described previously [14]. In brief, 96-well plates (Nunc) were coated with NP3-BSA or NP14-BSA (10 mg/ml, LGC Biosearch Technologies) at 4 °C overnight, after which the process was carried out at room temperature. For the determination of total immunoglobulin titers, plates were coated with purified rat anti-mouse IgM (clone II/41, BD Biosciences, Cat# 553435) or purified rat anti-mouse IgG1 (clone A85-3, BD Biosciences, Cat# 553445) instead, both diluted 1:100 in carbonate-bicarbonate buffer (0.2 M, pH 9.5).

To determine the NP-specific IgM antibody titers, the respective plates were washed with PBST (PBS plus 0.05% Tween 20) and blocked with 5% milk in PBS for 2 h. To determine total immunoglobulin titers, plates coated with purified rat anti-mouse IgG1 were washed with PBS and blocked with 1% milk in PBS for 1 h, while plates coated with purified rat anti-mouse IgM were washed and blocked in the same way as described for the plates in which NP-specific IgM antibody titers were determined.

The sera were diluted in PBS with 1% milk (for IgM, 1:10; for IgG1, 1:100; for total immunoglobulin titers, 1:200) and were added to the plates at 1:2 serial dilutions across 8 wells. The plates were incubated for 1 h at room temperature. The plates were washed and incubated with anti-mouse IgM HRP (clone 1B4B1, Southern Biotech, Cat# 1140-05) or anti-mouse IgG1 biotin (clone A85-1, BD Biosciences, Cat# 553441) for 1 h to detect NP-specific and total IgM and IgG1- in the mouse sera, respectively.

NP-specific and total IgM antibodies were detected directly after this incubation. For NP-specific IgG1 antibodies, the plates were washed with PBS and incubated with streptavidin horseradish peroxidase (HRP) avidin D (Vector) diluted in blocking buffer. For total IgG1 antibodies, the plates were washed with PBS and incubated with alkaline phosphatase (AP) streptavidin (Vector, SA-5100) diluted in blocking buffer.

Finally, the plates were developed with substrate buffer (1 tablet of phenylenediamine dihydrochloride (Sigma, P-7288))+ 35 ml of substrate buffer (0.1 M citric acid, 0.1 M Tris supplemented with 21 µl of H2O2). The plates incubated with alkaline phosphatase streptavidin were developed with 1 tablet of nitrophenylphosphate (Thermo Fisher Scientific, Cat# 34045) per 5 ml of 1x Pierce™ diethanolamine substrate buffer (Thermo Fisher Scientific, Cat# 34064) for 30 min.

The absorbance was determined with an ELISA plate reader (Photometer Sunrise RC, Tecan) at a wavelength of 405 nm, and data were acquired with Infinite F200 PRO i-control software, version 3.37. To facilitate the comparison of independent assays for NP-specific antibody titers, internal standards derived from mouse sera obtained from mice recently immunized with NP-CGG were employed. To compare independent assays in which total immunoglobulin titers were determined, purified mouse IgM (clone G155--228, BD Biosciences, Cat# 553472) or purified mouse IgG1 (clone MOPC-31C, BD Biosciences, Cat# 557273) were used as standards. Calibration curves were constructed from the absorbance values obtained from the standards. These curves served as reference points to calculate the titers in the sera of the CAR//Cγ1-Cre and CD30//Cγ1-Cre animals via Microsoft Excel 365 and GraphPad Prism software. All the NP-specific titers are presented as relative units, and the total titers are presented as concentrations.

The tissues were embedded in O.C.T. compound (VWR Chemicals, USA), snap frozen and stored at –20 °C. The samples were sliced to a thickness of 7 µm with a cryostat, mounted on glass slides and dried for 20 min at room temperature before fixation. For immunofluorescence staining, the slides were fixed with 3% Histofix for 10 min, rinsed with PBS, and rehydrated for 5 min in PBS (supplemented with 50 mM NH4Cl) for 5 min. Subsequently, blocking was performed in PBS containing 1% BSA and 5% rat serum for 20 min. For detection of GC B cells, the slides were stained with the following antibodies: anti-GL7-FITC (anti-mouse T- and B-cell activation antigen (GL7)) FITC (clone GL7, BD Biosciences, Cat# 553666) diluted in 1% BSA/PBS at 4 °C overnight. The secondary stains used were as follows: anti-Thy1.2-Biotin (Anti-Mouse CD90.2 clone 30-H12 Biotin, BD Biosciences, Cat# 553011). The third stains used were as follows: SA-594 (Thermo Fisher) and anti-B220-APC (Anti-Mouse CD45R/B220 APC, clone RA3-6B2, BD Biosciences, Cat# 553092) in 1% BSA/PBS for 1 h at room temperature. The following stains were performed for the detection of plasmablasts: first, Rat-anti-mouse Irf4 (Anti-Mouse Irf4 purified antibody, clone 3E4, eBioscience, Cat# 14-9858-80) and rabbit anti-mouse Laminin (Anti-Laminin, clone L9393, Sigma‒Aldrich, Cat# L9393) were diluted in 1% BSA/PBS and incubated for 1 h at room temperature. The secondary stains used were as follows: goat anti-rat Alexa Fluor 488 (anti-rat IgG Alexa Fluor 488, Jackson ImmunoResearch, Cat# 112--545--003) and goat anti-rabbit Cy3 (goat anti-rabbit IgG Cyanine Cy3, Jackson ImmunoResearch, Cat# 111--165--003) diluted in 1% BSA/PBS and incubated for 1 h at room temperature. Third, B220 APCs (Anti-Mouse CD45R/B220 APCs (clone RA3-6B2), BD Biosciences, Cat# 553092) were diluted in 1% BSA/PBS and incubated for 2 h at room temperature. Slides were embedded in SlowFade Glass Antifade (Invitrogen) and imaged on a TCS SP5 II confocal microscope (Leica). For chromogenic immunohistochemistry, the sections were incubated for 10 min in ice-cold acetone (–20 °C) and subsequently air dried for 10 min. The sections were rehydrated for 5 min in PBS at room temperature. The tissue sections were blocked for 20 min with 5% goat serum diluted in PBS and 1% BSA. After being washed with PBS, the slides were incubated with PNA-Biotin (B-1075, Vector) and Anti-Moma-1-Biotin (Anti-Metallophilic Macrophages (MOMA-1), clone MOMA-1, Abcam, Cat# ab51814) for 1 h at room temperature.

After a washing step, the slides were incubated with streptavidin-alkaline phosphatase (S-2890; Sigma‒Aldrich) for 1 h at room temperature. Stained samples were developed with an alkaline phosphatase kit (Vector Red™ Alkaline Phosphatase (Red AP) Substrate Kit - LS-J1086 (lsbio.com)) and embedded in Kaiser´s gelatin.

To isolate naïve B cells from total splenocytes, a CD43-microbead B-cell isolation kit (Miltenyi, Cat. No. 130-049-801) was used according to the manufacturer´s protocol. A total of 1–4 × 107 enriched naïve B cells were stained with a LIVE/DEAD Fixable nearIR Dead Cell Stain Kit (Thermo Fisher Scientific, Cat. No. L34975) and subsequently with the surface markers B220 PerCP, CD95 BV421, CD38 PeVio77, and hCD2 Fitc/CAR Fitc. The cells were subsequently resuspended in MACS buffer (Miltenyi, Cat. No. 130--091--221) and passed through a cell strainer with a 35 µm mesh size (Falcon Corning, Cat. No. 352235) immediately before cell sorting. After dead cell and doublet exclusion, three populations of B220-positive cells were sorted: reporter‒, reporter+ GC B cells and reporter+ non-GC B cells. After sorting, the cells were resuspended in 1 ml of TRI Reagent (Sigma‒Aldrich, T9424). FACS was performed on a BD FACSAria III via FACSDiva V8.0.1 software.

For qRT‒PCR, RNA was isolated from >4 × 105 FACS-sorted cells, and cDNA synthesis and quantitative PCR were performed as previously described [48]. Target-specific primers were designed via Primer-BLAST. For improved specificity, matching UPL probes were identified, and real-time PCR was performed as described for UPL assays (Roche). Supplementary Table 2 lists all primer and probe combinations. Throughout, primer efficiencies were >1.9.

All the statistical analyses, including tests for distributions and equal variance and calculations of medians, means and standard deviations (SDs), were performed with GraphPad Prism (versions 9.5.1--10.2.3). The data were subjected to normality and lognormality tests. In the case of a lognormal distribution, the data sets were logtransformed prior to statistical analyses. Sample sizes were determined on the basis of previous experiments and preliminary data. No statistical methods were used to predetermine sample sizes. Control and mutant mice were analyzed in parallel and predominantly used in an age-matched manner whenever possible. The experiments were repeated with different individual animals in at least two independent immunization or cell preparation rounds. Sample sizes and the chosen statistical tests are indicated in each figure legend.