Systems analysis reveals high genetic and antigen-driven predetermination of antibody repertoires throughout B-cell development

We performed high throughput antibody repertoire sequencing (HTRS) of key B-cell developmental stages of untreated and immunized C57BL/6 cohorts to allow analysis of genetic and antigen-driven predetermination and stochastic variation along the development of murine antibody repertoires.
Conclusion:
We isolated bone marrow pre B-cells (preBC), splenic naive follicular B-cells (nBC) and bone marrow long-lived plasma cells (BMPC) from untreated and prime-boost immunized (NP-HEL, Ova, HBsAg) cohorts (n=5) of C57BL/6 mice. Cells were directly sorted in Trizol and stored at -80°C for later RNA purification and antibody library preparation. Age matched 6-8 week old female C57BL/6 mice were randomly stratified into four cohorts of ten mice. Three cohorts followed a prime-boost immunization scheme. Primary immunizations consisted of intraperitoneal injections of alum-precipitated antigens. Antigen dosage per mouse was: 100 µg ovalbumin (OVA, Invivogen), 100 µg 4-hydroxy-3-nitrophenylacetyl-conjugated hen egg lysozyme (NP-HEL, Biosearch Technologies), 4 µg Hepatitis B virus surface antigen (HBsAg, Cell Sciences). Mice were boosted intraperitoneally after three weeks with identical amounts of antigen as used in primary in PBS. Immunized mice were sacrificed 14 days post-secondary immunization; untreated control mice were sacrificed at corresponding age. Spleen, bone marrow (from femur, tibia, pelvis and sternum) and blood were collected from all mice. Of 8 sorted mice per cohort, those 4-5 were chosen based on highest antigen-specific titers. Single-cell suspensions were prepared from harvested bone marrow and spleen. Red blood cells were lysed (RBC lysis buffer, eBioscience) and Fc receptor block was performed using rat serum to inhibit non-specific binding of rat-derived labeling antibodies. Bone marrow cells were labeled with a panel of fluorescence-conjugated antibodies consisting of CD19-FITC (eBioscience, clone 1D3), CD138-PE (BD, clone 281-2), CD22-APC (Biolegend, clone OX-97), MHCII-APC-Cy7 (Biolegend, clone M5/114.15.2), IgM-BV510 (BD, clone R6-60.2), CD25-PE-Cy7 (Biolegend, clone PC61), c-kit-BV421 (Biolegend, clone 2B8). Splenocytes were stained with an antibody panel consisting of CD19-FITC, CD138-PE, CD22-APC, MHCII-APC-Cy7, IgM-BV510, IgD-PerCP-Cy5.5 (Biolegend, clone 11-26c.2a), CD21-BV421 (Biolegend, clone 7E9) and CD23-PE-Cy7 (Biolegend, clone B3B4). Cells were sorted on a BD FACS Aria II and analyzed using FACS Diva software. Compensation was performed with single stain controls (UltraComp eBeads, eBioscience) and fluorescence minus one (FMO) controls were used to set the gates. The following cellular populations were isolated: pre-B cells from bone marrow (c-kit-CD19+IgM-CD25+), long-lived memory plasma cells from bone marrow (CD138+CD22-MHCII-CD19-IgM-) and naïve follicular B cells from spleen (CD138-CD19+IgD++IgM+CD23++CD21+). All cells were sorted directly into TRIzol (Life Technologies) and stored at -80°C.