Ethics statement

All animal experiments were conducted in accordance with protocols that were approved by the University of Utah Institutional Animal Care and Use Committee.

Viral infections and protein immunizations

C57BL/6J mice (Jackson Laboratory, Bell Harbor, ME) were infected with either 30 μl of 500 TCID₅₀ mouse-adapted PR8-HA-GP_61-80 or 2x10⁵ PFU of LCMV Armstrong by intranasal inoculation or immunized by intramuscular (quadriceps) injection with 2 μg LCMV recombinant glycoprotein (rGP) with addition of Addavax (InvivoGen) adjuvant at a 1:1 ratio. PR8-HA-GP_61-80 recombinant virus strain is the H1N1 PR8 strain with the CD4-immunodominant LCMVgp61-80 epitope inserted into the HA region and was kindly provided by Dr. Florian Krammer (Icahn School of Medicine at Mount Sinai). 293A cells that express recombinant glycoprotein (from LCMV) were kindly provided by Dr. Carl Davis (Emory University), and recombinant glycoprotein was purified from supernatants as described previously [54]. For B cell reactivation experiments, mice were immunized by intraperitoneal injection with 10 μg recombinant HA (H1 subtype) protein from PR8 (H1N1) virus without adjuvant. For intranasal infections, mice were anesthetized with concurrent administration of aerosolized isoflurane and oxygen using a COMPAC⁵ Anesthesia Center (VetEquip). Prior to euthanasia, mice were intravenously injected by retro-orbital injection with 2 μg α-CD45-FITC (30-F11, Tonbo Biosciences) antibody to detect remaining circulating cells in lung samples [55].

Construction of the recombinant influenza virus PR8-HA-GP_61-80

To obtain a recombinant influenza virus containing the LCMV epitope GP_61-80 (GLKGPDIYKGVYQFKSVEFD) inserted in the hemagglutinin (HA) protein, the sequence encoding the GP _61–80 peptide was introduced in the rescue plasmid pDZ-HA, strain A/Puerto Rico/8/1934 (H1N1) (PR8). The epitope was inserted in-frame at the amino acid position 135, that is highly tolerant to small insertions [56]. Next, the recombinant virus was rescued by transfecting cells with 8 plasmids containing the sequences of the viral segments, as previously described [57].

Tissue processing

Single-cell suspensions of pooled mediastinal lymph nodes or pooled inguinal and lumbar lymph nodes were prepared using 70-μm cell strainers. Single-cell suspensions of spleens were prepared using 70-μm cell strainers and red blood cells lysed by incubation in Ammonium-Chloride-Potassium (ACK) Lysing Buffer (Life Technologies). Single-cell suspensions of lungs were prepared by digestion with 0.25mg/ml Collagenase IV and 15 μg/ml DNase for 1 hour at 37°C, then manually homogenized and red blood cells lysed by incubation in ACK Lysing Buffer and then cells were filtered using 70-μm cell strainers. Cell suspensions were resuspended in RPMI 1640 media supplemented with 5% fetal bovine serum (FBS) prior to FACS staining.

FACS analysis

Single-cell suspensions of spleens, lungs, and lymph nodes were prepared and up to 2x10⁶ cells were stained in 1X PBS supplemented with 2% fetal bovine serum (FACS buffer) for 15–30 minutes on ice with fluorochrome-conjugated antibodies. Antibodies for FACS included LIVE/DEAD Fixable Near-IR Dead Cell Stain, CD4 (RM4-5), CD8 (53–6.7), CD44 (IM7), IFNγ (XMG1.2), TNFα (MP6-XT22), IL-2 (JES6-5H4), PD-1 (29F.1A12), Ly6c (HK1.4), Bcl6 (K112-91), Tbet (4B10), CD19 (eBio1D3 (1D3)), B220 (RA3-6B2), Fas/CD95 (Jo2), GL7 (GL7), IgD (11-26c.2a), CD138 (281–2) (purchased from BD Biosciences, eBiosciences, BioLegend, Vector Laboratories Inc., and Invitrogen). For I-A^b:gp66-77 tetramer (provided by the National Institutes of Health Tetramer Core) staining, cells were incubated with tetramer in RPMI medium supplemented with 10% FBS for 2 h at 37°C with 5% CO₂. CXCR5 surface staining was performed using a three-step protocol described in Johnston et al. (2009) [29] using purified rat anti-mouse CXCR5 primary antibody (BD Biosciences, 2G8) in FACS buffer supplemented with 1% bovine serum albumin (Sigma, #A7284) and 2% normal mouse serum (Sigma, #M5905) (CXCR5 staining buffer), a secondary Biotin-SP-conjugated Affinipure F(Ab’)₂ Goat anti-Rat IgG (Jackson ImmunoResearch) in CXCR5 staining buffer and then with a fluorochrome-conjugated streptavidin in FACS buffer. For transcription factor staining, cells were first stained for surface antigens, followed by permeabilization, fixation and staining using the Foxp3 Permeabilization/Fixation kit and protocol (eBiosciences). Intracellular cytokine staining was done by standard techniques following 5-hour stimulation with Gp_61-80 peptide and Brefeldin A (GolgiPlug, BD Biosciences). No peptide controls were treated under the same conditions supplemented with Brefeldin A but without Gp_61-80 peptide. Cells were then stained for surface antigens, followed by permeabilization, fixation and staining using the Cytofix/Cytoperm kit and protocol (BD Biosciences). For influenza HA-specific B cell staining, recombinant HA protein from A/Puerto Rico/8/1934 (H1N1) virus strain (Immune Technology Corp., #IT-003-0010ΔTMp) was biotinylated with 80-fold molar excess of NHS-PEG4-Biotin solution from the EZ-Link NHS-PEG4-Biotin kit (ThermoFisher, #A39259). Excess biotin was removed by buffer exchange of protein into sterile 1X PBS using Zeba Spin Desalting Columns, 7K MWCO (ThermoFisher, #89882). Cells were stained on ice for 30min in FACS buffer with 1:100 dilutions of biotin-conjugated-HA and purified rat anti-mouse CD16/CD32 (Mouse BD Fc Block, Clone 2.4G2, BD Biosciences), then stained on ice for 30min in FACS buffer with 1:1000 dilution of allophycocyanin (APC)-conjugated streptavidin. Cells were analyzed on LSRFortessa X-20 and LSRFortessa (BD Biosciences) cytometers. FACS data were analyzed using FlowJo v10 software (Tree Star).

Hemagglutination inhibition assay (HAI)

Serum was separated from whole blood by centrifugation at 10,000xg for 30min at 4°C. HAI to determine neutralizing antibody titers was performed by incubating 25 μL of two-fold serially diluted serum with 25 μL of 4 agglutinating doses (4AD) of WT PR8 (H1N1) virus strain for 30min at room temperature (RT) prior to addition of 50μL of 1% chicken red blood cells (cRBCs) (Lampire Biological Laboratories) in 1X PBS. Plates were gently agitated to mix and then incubated for 30min at RT. HAI titers were determined as the reciprocal dilution of the final well which contained non-agglutinated cRBCs. Naïve mouse serum was used as a negative control. Mice with titers of <1:10 were not included in final analyses.

Enzyme-linked immunosorbent assay (ELISA)

ELISA to determine HA-specific IgG antibody titers was performed by coating MaxiSorp Clear Flat-Bottom Immuno Nonsterile 96-Well Plates (ThermoFisher) with 1 μg/mL of recombinant HA protein from A/Puerto Rico/8/1934 (H1N1) virus strain (Immune Technology Corp., #IT-003-0010ΔTMp) overnight at 4°C. Plates were blocked for 90min at RT with a solution of 1X PBS with 0.05% Tween 20 and 10% fetal bovine serum (blocking solution). Plates were incubated with three-fold serially diluted serum in technical duplicates for 90min at RT. Plates were then incubated for 90min at RT with goat anti-mouse IgG conjugated to horseradish peroxidase (HRP) (Southern Biotech, #1030–05) at 1:5000 dilution in blocking solution. Plates were washed with 1X PBS with 0.05% Tween 20 (PBST) after each blocking and incubation step. Plates were then incubated with 100 μL of substrate solution consisting of 4 mg o-Phenylenediamine dihydrochloride (OPD, Sigma, #P8787) dissolved in 10 mL filter sterilized citrate buffer (0.05M citric acid anhydrous, 0.1M sodium phosphate dibasic anhydrous (Na2HPO4)) and 33 μL of 3% H₂O₂. The reaction was stopped after 10 min with 100 μL of 1M hydrochloric acid and plates were scanned at 490nm using a Biotek Synergy H1 microplate reader. Naïve mouse serum was used as negative controls. OD readings were averaged between technical duplicates for all samples. Titer cutoff value was determined using the OD values of negative controls as described in Frey et al. (1998) [58] using a 95% confidence level. Relative endpoint titers were calculated by nonlinear regression interpolation of a standard curve (Sigmoidal, 4PL, X is concentration) of individual samples using GraphPad Prism version 9.4.1 for macOS and calculating the titer at which each curve crosses the background cutoff value.

Enzyme-linked immunosorbent spot assay (ELISpot)

Bone marrow was collected from femur and tibia bones and red blood cells lysed by incubation in ACK Lysing Buffer. B cell enrichment of bone marrow cells was performed using the Pan B Cell Isolation Kit (Miltenyi Biotec, #130-095-813). MultiScreen-IP Filter Plates (Sigma, #MAIPS4510) were pre-wet with 15 μL 35% ethanol for 30 sec and washed with 1X PBS. Plates were coated with 2 μg/mL of recombinant HA protein from A/Puerto Rico/8/1934 (H1N1) virus strain (Immune Technology Corp.) overnight at 4°C. Plates were washed with 1X PBS and then blocked for 2hr at RT with RPMI 1640 medium supplemented with 10% fetal bovine serum, 1% Penicillin-Streptomycin, 2 mM L-glutamine (complete culture medium). Plates were washed with 1X PBS and then enriched B cells in complete culture medium were added to plates at two-fold serial dilutions and in technical duplicates for each sample at maximum 4x10⁶ cells/well and incubated overnight at 37°C at 5% CO₂. Plates were washed with 1X PBS, then washed with 1X PBS with 0.05% Tween 20, then incubated with Goat anti-Mouse IgG (H+L) Cross-Adsorbed Secondary Antibody conjugated to HRP (ThermoFisher, #G-21040) at a 1:350 dilution in 1X PBS with 0.05% Tween 20 and 1% fetal bovine serum overnight at 37°C at 5% CO₂. Plates were washed with 1X PBS with 0.05% Tween 20, then washed with 1X PBS, and plates were developed using the AEC Staining Kit (Sigma, #AEC101-1KT). After spot development, plates were washed with water and allowed to dry before counting.

Histology and immunofluorescence

To analyze influenza infected lungs by histology, mice were humanely euthanized by CO₂ and perfused with PBS (Thermo Fisher Scientific) through the left ventricle of the heart. Lungs were then inflated with 4% paraformaldehyde, removed, and placed in 4% paraformaldehyde at 4°C overnight. The next day, the left lung was carefully dissected and stored in 4% paraformaldehyde at 4°C for 24 hours, while the right lung was carefully dissected and stored in 70% ethanol. After 24hrs, the right lung was sent to the University of Utah Biorepository and Molecular Pathology (BMP) Shared Resource for paraffin embedding, sectioning, and staining with hematoxylin and eosin. The left lung was placed into a 30% sucrose solution in PBS at 4°C for 2 days before being placed in a 10% sucrose solution for 4 hours. Next, lungs were washed with PBS at RT with agitation for 1 hour before being embedded in OCT (Tissue-Tek) and flash frozen. Twenty-micron cryosections were sliced from each sample and stored at 80°C until staining. Before staining, sections were fixed using a mixture or ethanol and acetone, rehydrated with PBS, and blocked using 5% rat and 5%rabbit serum in PBS. After blocking for an hour at 4°C, sections were stained with the following antibodies from Invitrogen overnight at 4°C: CD4-FITC (RM4-5), CD31-PE (390), B220-AF561(RA3-6B2). Slides were then washed 5x with PBS and agitation before mounting with Fluoromount-G (Thermo Fisher Scientific). Samples were imaged using a Leica SP8 DIVE. Tile scans were obtained at 1048x1048 resolution and z-stacks are represented by maximum projection.

B cell repertoire sequencing

Single-cell suspensions of splenocytes from individual mice were prepared and B cell enrichment was performed using the Pan B Cell Isolation Kit (Miltenyi Biotec, #130-095-813). Plasmablast cell sorting was performed using a FACSAria (BD Biosciences). Genomic DNA was isolated from sorted plasmablasts using QIAamp DNA Mini Kit (Qiagen), and amplification and sequencing of the Igh locus were performed using the immunoSEQ platform (Adaptive Biotechnologies). Data analyses were conducted in the immunoSEQ Analyzer (Adaptive Biotechnologies) and R [59], RStudio [60], and the Immunarch package [61]. Data was exported from RStudio using the writexl package [62]. Figures were created with the Immunarch package or GraphPad Prism version 9.4.1 for macOS. The full sequencing dataset is available in S1 Table.

Statistical analysis

All experiments were analyzed using GraphPad Prism version 9.4.1 for macOS. Statistically significant p values of <0.05 are indicated and were determined using either a two-tailed unpaired Student’s t test with Welch’s correction or Mann-Whitney U test. Error bars represent Mean±SEM, *p≤0.05, **p≤0.01, ***p≤0.001, ****p≤0.0001.

Generation of antigen-specific memory Tfh cells in rGP-immunized and LCMV-infected mice prior to influenza challenge

Our goal was to generate antigen-specific memory CD4+ T cells by heterologous priming with protein immunization or viral infection that could provide help during a primary influenza response. We first evaluated the kinetics and differentiation of polyclonal antigen-specific CD4+ T cells following adjuvanted protein immunization or acute viral infection. C57BL/6J mice were primed by i.m. immunization with 2 μg recombinant LCMV glycoprotein (rGP) in AddaVax adjuvant (GP(1°) group) or by i.n. inoculation with 2x10⁵ PFU acute LCMV-Armstrong (LCMV(1°) group) (Fig 1A). At 8, 15, and 39 days post-infection or -immunization (dpi) we analyzed CD4+ T cells in draining lymph nodes–pooled inguinal and lumbar (dLN) following i.m. rGP immunization or mediastinal (medLN) following i.n. LCMV infection–and spleens by staining with the LCMV I-A^b:gp66-77 MHC class II tetramer. At day 39 post-infection or -immunization, memory I-A^b:gp66-77 tetramer+ CD4+ T cells were detected in draining lymph nodes and spleens in both the GP(1°) and LCMV(1°) groups (Figs 1B and S1A). In addition, the longitudinal kinetics of I-A^b:gp66-77 tetramer+ CD4+ T cells analyzed were similar in the draining lymph nodes and spleens of the GP(1°) and LCMV(1°) groups, with peak clonal expansion at 8 dpi and maintenance of the post-contraction memory tetramer+ CD4+ T cell pool detectable at 39 dpi (S1B–S1C Fig). Tetramer+ memory CXCR5+ Tfh cells were similar by frequency and number in the draining lymph nodes of the GP(1°) and LCMV(1°) groups (Figs 1C and S1A–S1C), though the LCMV(1°) group had significantly higher numbers of these cells in the spleen (Fig 1D). While both Tfh (CXCR5+Ly6C-) and Th1 (CXCR5-Ly6C+) cells were readily detectable in the draining lymph nodes following the establishment of memory (day 39 post-infection) after both LCMV infection and GP immunization, some germinal center Tfh (CXCR5+PD-1+) persisted (S1D Fig).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Polyclonal memory CD4+ T follicular helper cell formation following recombinant protein immunization and acute viral infection.

C57BL/6J mice were immunized i.m. with 2 μg rGP in AddaVax adjuvant (GP(1°), filled triangle) or infected i.n. with 2x10⁵ PFU of LCMV-Armstrong (LCMV(1°), filled diamond). 8-, 15-, and 39-days postinfection or -immunization, lymphocytes from pooled lumbar and inguinal draining lymph nodes (dLN) (rGP immunization), mediastinal lymph nodes (medLN) (LCMV infection), or spleens were stained with I-A^b:gp66-77 tetramer to analyze antigen-specific CD4+ T cell responses. (A) Schematic of experimental design. (B) Representative flow plots of CD44 and tetramer analysis of total CD4+ T cells in dLN or medLN 39 days postinfection or -immunization. (C) Frequency and number of memory CXCR5+ Tfh cells of total tetramer+CD4+ T cells in dLN or medLN at 39 days postinfection or -immunization. (D) Frequency and number of memory CXCR5+ Tfh cells of total tetramer+CD4+ T cells in spleen at 39 days postinfection or -immunization. n ≥ 3 per group per experiment at each timepoint. Data shown are from one independent experiment. Statistically significant p values of <0.05 are indicated and were determined using a two-tailed unpaired Student’s t test with Welch’s correction. Error bars represent Mean±SEM, *p≤0.05, **p≤0.01, ***p≤0.001, ****p≤0.0001.

https://doi.org/10.1371/journal.ppat.1011639.g001

When we analyzed for IFNγ-expressing memory CD4+ T cells in spleen following gp61-80 peptide restimulation and normalized to IFNγ background expression in naïve CD4+ T cells, there was a significantly higher frequency and number of IFNγ-expressing cells in the LCMV(1°) group (S1E–S1F Fig). The lack of IFNγ-expressing cells in the GP(1°) group was expected, as we previously reported that adjuvanted rGP immunization induces expansion of a CXCR5–IFNγ–nonpolarized T helper cell population in lieu of highly IFNγ-expressing Th1 cells as seen following LCMV infection [63]. Together, these data show that both adjuvanted rGP immunization and LCMV infection induced I-A^b:gp66-77 tetramer+ memory CD4+ Tfh and non-Tfh cells.

In addition, we wanted to confirm that our heterologous priming did not induce antibody production to HA protein. We analyzed the serum from GP(1°), LCMV(1°), and naïve mice, along with mice infected intranasally with PR8-HA-GP_61-80 recombinant influenza virus at 39–42 dpi for antibodies generated against PR8-HA protein. Antibodies were found that bound to PR8-HA-GP_61-80 recombinant protein in the GP(1°) and the PR8-HA-GP_61-80 primary influenza response (PR8(1°)) group, but not the naïve or LCMV(1°) groups (S1G Fig). Importantly, antibodies that bound to PR8-HA recombinant protein were only found in the PR8(1°) serum, suggesting that adjuvanted rGP immunization and LCMV infection did not induce primary B cell responses specific to PR8-HA (S1G Fig).

Previously generated antigen-specific memory CD4+ T cells induced increased effector Th1 and GC Tfh cells upon influenza infection

We next evaluated if using heterologous priming with adjuvanted rGP immunization or i.n. LCMV infection to generate antigen-specific memory CD4+ T cells would enhance the early effector germinal center response to influenza infection. Using the heterologous priming strategies detailed in Fig 1A, 42 days after priming infection or immunization we infected i.n. with 500 TCID₅₀ PR8-HA-GP_61-80 recombinant influenza virus (GP(1°)PR8(2°) and LCMV(1°)PR8(2°) groups) (Fig 2A). For control groups, we used age- and sex-matched naïve mice infected i.n. with PR8-HA-GP_61-80 to evaluate the primary influenza response (PR8(1°) group) and a separate group was homologously primed with PR8-HA-GP_61-80 i.n. infection (PR8(1°)PR8(2°) group) to evaluate recalled cellular and antibody responses (Fig 2A).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Generation of memory CD4+ T cells by heterologous immunization induced increased effector antigen-specific Th1 and GC Tfh cells following influenza infection.

C57BL/6J mice were primed by i.m. immunization with 2 μg rGP in AddaVax (GP(1°)PR8(2°), filled triangle), by i.n. infection with 2x10⁵ PFU of LCMV-Armstrong (LCMV(1°)PR8(2°), filled diamond), or by i.n. infection with 500 TCID₅₀ of PR8-HA-GP_61-80 (PR8(1°)PR8(2°), filled hexagon). 42 days postinfection or -immunization, primed mice and unprimed age-matched naïve mice (PR8(1°), unfilled circle) were infected i.n. with 500 TCID₅₀ of PR8-HA-GP_61-80 influenza virus. 8 days after influenza infection, lymphocytes from medLN were stained with I-A^b:gp66-77 tetramer to analyze antigen-specific CD4+ T cell responses or stained with I-A^b human CLIP87-101 as a control. (A) Schematic of experimental design. (B) Representative FACS plots of CD44 and tetramer analysis of total CD4+ T cells. (C) Frequency and number of effector tetramer+CD44+ of total CD4+ T cells. (D) Representative flow plots of CXCR5, TBET, PD-1, and BCL6 analysis of tetramer+CD44+ CD4+ T cells. (E) Frequency and number of effector CXCR5–TBET+ T helper 1 cells of total tetramer+CD4+ T cells. (F) Frequency and number of effector CXCR5+PD-1+ Tfh cells of total tetramer+CD4+ T cells. (G) Frequency and number of effector CXCR5+BCL6+ GC Tfh cells of total tetramer+CD4+ T cells. (H) BCL6 geometric mean fluorescence intensity (gMFI) of tetramer+ CXCR5+ cells. n ≥ 3 per group per experiment. Data shown are from three independent experiments. Statistically significant p values of <0.05 are indicated and were determined using a two-tailed unpaired Student’s t test with Welch’s correction. Error bars represent Mean±SEM, *p≤0.05, **p≤0.01, ***p≤0.001, ****p≤0.0001.

https://doi.org/10.1371/journal.ppat.1011639.g002

First, we used weight loss and histology to examine the impact of our heterologous priming on PR8-HA-GP_61-80 infection. We found that the PR8(1°), GP(1°)PR8(2°), and LCMV(1°)PR8(2°) groups lost similar amounts of weight after infection and recovered after two weeks while PR8(1°)PR8(2°) mice were protected from weight loss (S2A Fig). PR8(1°), GP(1°)PR8(2°) and LCMV(1°)PR8(2°) groups also showed perivascular and peribronchial leukocytic infiltrates in the lungs by histology, while PR8(1°)PR8(2°) mice had reduced lung inflammation (S2B Fig). Together, these data suggested that homologous primed, but not heterologous primed mice, had reduced pathology due to secondary influenza infection.

8 days after influenza infection, both GP(1°)PR8(2°) and LCMV(1°)PR8(2°) groups had significantly higher frequencies and numbers of effector tetramer+ CD4+ T cells in medLN than the PR8(1°) and PR8(1°)PR8(2°) groups, with LCMV-primed mice having the highest overall (Fig 2B–2C); while total lymphocytes were mostly comparable across the four groups (S3A Fig). Heterologous infection or immunization priming prior to influenza infection induced increased frequencies and numbers of effector tetramer+ CXCR5–TBET+ Th1 cells compared to both primary influenza infection alone and influenza-immune mice at 8 dpi (Fig 2D–2E). While PR8(1°) mice had a significantly higher frequency of CXCR5+PD-1+ Tfh and CXCR5+BCL6+ GC Tfh cells in medLN, the numbers of GC Tfh cells were significantly higher in LCMV-primed mice (Fig 2D and 2F–2G). Although the numbers of BCL6-expressing CXCR5+ GC Tfh cells were highest in LCMV-primed mice, the amount of BCL6 expression in tetramer+ CXCR5+ cells was significantly lower than in both GP-primed mice and after primary influenza infection (Fig 2G–2H). In addition, our data showed distinct populations in medLN of CXCR5+LY6C^low Tfh cells and CXCR5–LY6C^high Th1 cells, although the Th1 cells in the GP(1°)PR8(2°) group were mostly LY6C^low (S3B Fig).

Cytokine analysis of CD4+ T cells in medLN following gp61-80 restimulation revealed that LCMV-primed mice had the highest frequency and number of both IFNγ+ and polyfunctional IFNγ+TNFα+IL-2+ expressing cells (S3C–S3E Fig). The GP(1°)PR8(2°) group also had a significantly higher number of IFNγ+ expressing cells than the PR8(1°) and PR8(1°)PR8(2°) groups, further confirming that memory nonpolarized T helper cells can form Th1 cells after secondary heterologous activation (S3D Fig). We also evaluated differences in tetramer+ Th1 and Tfh cells in the spleen and found that similar to medLN, while total lymphocytes were mostly comparable in number (S3A Fig), Th1 cells (by both TBET and IFNγ expression) in LCMV-primed mice were significantly higher overall (S3F–S3H Fig). However, our data showed no differences in Tfh cells in the spleen by PD-1 or BCL6 expression in the heterologously-primed or primary influenza infection groups (S3I–S3J Fig). Together, our data show that heterologous priming with rGP immunization or LCMV infection had induced enhanced expansion of CD4+ Tfh and Th1 cells compared to both primary influenza and secondary homologous influenza infection.

Prior generation of memory Tfh cells promoted increased influenza-specific GC B cells and plasmablasts upon influenza infection

To determine if recalled antigen-specific memory CD4+ T cells in heterologously primed mice could enhance the primary anti-influenza B cell response, we next analyzed GC B cell and plasmablast populations in medLN and spleen. Total CD19+B220+/low B cells in medLN were mostly comparable by frequency and number, with the lowest B cell numbers in the PR8(1°)PR8(2°) group (S4A Fig). However, heterologous infection or immunization priming of CD4+ T cells drove a significant increase in the number of total Fas+GL7+ GC B cells and influenza HA-specific GC B cells (Fig 3A–3D) at 8 dpi. In addition, while the number of total IgD–CD138+ plasmablasts in medLN were similar between the unprimed and heterologously-primed groups, there was a significant increase of HA-specific plasmablasts in both rGP- and LCMV-primed mice (Fig 3E–3H). Analysis of the splenic B cell response revealed significant increases in numbers of total GC B cells and plasmablasts, as well as HA-specific GC B cells and plasmablasts in the GP(1°)PR8(2°) group compared to the PR8(1°) group (S4B–S4F Fig), despite no differences in splenic tetramer+ GC Tfh cells (S3I–S3J Fig).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Generation of memory CD4+ T cells by heterologous immunization induced increased influenza-specific B cells following influenza infection.

Flow cytometry analysis of B cells from medLN 8 days after PR8-HA-GP_61-80 influenza virus infection in rGP immunization-primed (GP(1°)PR8(2°), filled triangle), LCMV-primed (LCMV(1°)PR8(2°), filled diamond), homologously primed mice (PR8(1°)PR8(2°), filled hexagon), or unprimed naïve mice (PR8(1°), unfilled circle). (A) Representative flow plots of Fas and GL7 analysis gated on total CD19+B220+/low cells. (B) Frequency and number of Fas+GL7+ GC B cells of total CD19+B220+/low B cells. (C) Representative flow plots of influenza HA-specific GC B cells gated on total Fas+GL7+ GC B cells. (D) Frequency and number of HA-specific GC B cells of total Fas+GL7+ GC B cells. (E) Representative flow plots of IgD and CD138 analysis gated on total CD19+B220+/low cells. (F) Frequency and number of IgD–CD138+ plasmablasts of total CD19+B220+/low cells. (G) Representative flow plots of influenza HA-specific plasmablasts gated on total IgD–CD138+ plasmablasts. (H) Frequency and number of influenza HA-specific plasmablasts of total IgD–CD138+ plasmablasts. (I) Correlation analysis of number of tetramer+ CXCR5+BCL6+ GC Tfh cells to number of HA-specific Fas+GL7+ GC B cells. Spearman rank-order correlation values (r) and statistically significant p values are shown with linear regression curve fit line slopes and statistically significant p values. (J) Ratio of number of HA-specific GC B cells to number of tetramer+ CXCR5+BCL6+ GC Tfh cells. n ≥ 3 per group per experiment. Data shown are from three independent experiments. Statistically significant p values of <0.05 are indicated and were determined using a two-tailed unpaired Student’s t test with Welch’s correction. Error bars represent Mean±SEM, *p≤0.05, **p≤0.01, ***p≤0.001, ****p≤0.0001.

https://doi.org/10.1371/journal.ppat.1011639.g003

To determine if there was a correlation between the numbers of tetramer+ GC Tfh cells and HA-specific GC B cells in medLN, we performed Spearman correlation analysis and found that in all groups there was a statistically significant positive correlation (Fig 3I). In addition, linear regression analysis of GC Tfh and HA-specific GC B cell numbers showed statistically significant positive associations in the unprimed and heterologously-primed groups (Fig 3I), consistent with previous studies describing the critical interaction between Tfh and B cells in the germinal center [29,51,64–66]. We further compared the numbers of HA-specific GC B cells to the number of GC Tfh cells and found that in the GP(1°)PR8(2°) and PR8(1°)PR8(2°) groups, there were significantly higher numbers of HA-specific GC B cells to every GC Tfh cell (Fig 3J). These data suggest that while the number of GC Tfh cells in the GP(1°)PR8(2°) group were similar to the PR8(1°) group, the GC Tfh cells may be of a higher quality as to sustain support for higher numbers of HA-specific GC B cells, as has been previously suggested [44,47]. For the PR8(1°)PR8(2°) group, prior studies have shown poor recall of memory CD4+ cells during homologous secondary infection [67], likely causing the higher ratio of HA-specific GC B cells to antigen-specific GC Tfh cells. Together, these data suggest that heterologous priming with adjuvanted rGP immunization or LCMV infection significantly enhanced the early anti-influenza germinal center B cell response following influenza infection compared to primary influenza infection.

Previously generated memory CD4+ T cells by heterologous immunization did not impact anti-influenza antibody titers

To determine if the early increases in tetramer+ CD4+ T cells and influenza-specific B cells in heterologously primed mice were maintained at memory, we analyzed the cellular response longitudinally in medLN and spleen 42 days after influenza challenge as detailed in Fig 2A. Similar to 8 days after influenza challenge, heterologously primed mice maintained significantly higher frequencies and numbers of I-A^b:gp66-77 tetramer+ CD4+ T cells in medLN 42 days after influenza challenge compared to the unprimed PR8(1°) and homologously primed PR8(1°)PR8(2°) groups (Figs 4A and S5A). Our data also showed significantly greater frequencies and numbers of tetramer+ CXCR5–TBET+ Th1 cells and significantly higher numbers of antigen-specific polyfunctional cytokine-secreting (IFNγ+TNFα+IL-2+) CD4+ T cells in heterologously primed mice in medLN (Fig 4B–4C). In addition, memory I-A^b:gp66-77 tetramer+ CD4+ T cells, tetramer+ CXCR5–TBET+ Th1 cells and polyfunctional cytokine secreting CD4+ T cells were detectable in the spleen at significantly higher numbers in heterologously primed mice (S5B–S5E Fig).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Generation of memory CD4+ T cells by heterologous immunization induced increased memory Th1 cells remaining after influenza infection but did not enhance influenza-specific antibody titers.

Flow cytometry analysis of CD4+ T cells and B cells from medLN 42 days after PR8-HA-GP_61-80 influenza virus infection in heterologously primed mice (GP(1°)PR8(2°), filled triangle, or LCMV(1°)PR8(2°), filled diamond), homologously primed mice (PR8(1°)PR8(2°), filled hexagon), or unprimed naïve mice (PR8(1°), unfilled circle). Antigen-specific CD4+ T cell responses were analyzed either by staining with I-A^b:gp66-77 tetramer or cytokine expression following restimulation with LCMV gp61-80 peptide. Serum was isolated from whole blood collected from influenza infected mice at 15–16, 42–50, and 100+ days postinfection and analyzed by ELISA. (A) Frequency and number of tetramer+CD44+ of total CD4+ T cells in medLN at 42 days postinfection. (B) Frequency and number of CXCR5–TBET+ T helper 1 cells of total tetramer+CD4+ T cells. (C) Number of antigen-specific IFNγ+TNFα+IL-2+ cells of total CD4+ T cells. (D) Frequency and number of CXCR5+PD-1+ Tfh cells of total tetramer+CD4+ T cells. (E) Frequency and number of CXCR5+BCL6+ GC Tfh cells of total tetramer+CD4+ T cells. (F) Kinetics of CXCR5+BCL6+ GC Tfh cells of total tetramer+CD4+ T cells in medLN at 8, 15, and 42 days postinfection. (G) Number of HA-specific GC B cells of total Fas+GL7+ GC B cells in medLN at 42 days postinfection. (H) Kinetics of HA-specific GC B cells in medLN at 8, 15, and 42 days postinfection. (I) Anti-influenza H1 HA-specific IgG antibody titers from serum at 42–50 and 100+ days postinfection by ELISA. n ≥ 3 per group per experiment. Kinetics data (panels F and H) shown are from one independent experiment. FACS and serology data shown are from two to three independent experiments. Statistically significant p values of <0.05 are indicated and were determined using a two-tailed unpaired Student’s t test with Welch’s correction. Error bars represent Mean±SEM, *p≤0.05, **p≤0.01, ***p≤0.001, ****p≤0.0001. NS = not significant.

https://doi.org/10.1371/journal.ppat.1011639.g004

As the CD4+ T cell-specific immunodominant LCMVgp61-80 epitope contains a cryptic epitope recognized by CD8+ T cells [68], we analyzed CD8+CD44+ T cells at 8 and 42 days after influenza challenge for IFNγ expression following LCMVgp61-80 peptide restimulation. Our data show that most mice had frequencies of CD44+IFNγ+ cells of CD8+ T cells below background levels as normalized to a no peptide control, and all other mice had less than 1% of CD8+ T cells expressing CD44 and IFNγ (S5F Fig). These data suggest that while non-specific secretion of IFNγ by CD8+ T cells was detected, we do not expect these cells significantly influenced the increases in Th1 cells and IFNγ-secreting CD4+ T cells.

We next analyzed I-A^b:gp66-77 tetramer+ CD4+ T cells for CXCR5, PD-1, and BCL6 expression following influenza infection in medLN to determine the kinetics of memory antigen-specific Tfh cells. At 42 days after influenza infection, the PR8(1°) group had maintained a significantly higher frequency of both PD-1- and BCL6-expressing Tfh cells (Fig 4D–4E) similar to the effector timepoint. While BCL6 expression in memory CXCR5+ Tfh cells is significantly reduced following acute viral clearance compared to effector CXCR5+ GC Tfh cells [69], LCMV(1°)PR8(2°) mice still had significantly higher numbers of BCL6-expressing Tfh cells maintained at memory compared to the other three groups (Fig 4E). When we analyzed lymphocytes at 8, 15 and 42 days after influenza infection to evaluate differences in proliferation or contraction kinetics of CXCR5+BCL6+ GC Tfh cells, our data showed no differences in longitudinal kinetics of GC Tfh cells in this experiment and only significantly higher numbers of GC Tfh cells in heterologously primed mice at 8 dpi (Fig 4F).

We then analyzed the B cell populations in medLN at 42 dpi and found only the LCMV(1°)PR8(2°) group had significantly higher numbers of HA-specific GC B cells compared to GP(1°)PR8(2°) and PR8(1°)PR8(2°) mice (Fig 4G). When we analyzed HA-specific GC B cell kinetics at 8, 15, and 42 days after influenza infection, HA-specific GC B cells underwent contraction after peak expansion around 8 dpi in heterologously primed mice (Fig 4H). However, in the PR8(1°) group HA-specific GC B cells increased in number after 8 dpi, though numbers were not significantly different from heterologously primed at 15 or 42 dpi in this experiment (Fig 4H).

As heterologous infection or immunization priming significantly increased HA-specific GC B cells and plasmablasts 8 days after influenza infection, we analyzed the sera of influenza infected mice to determine HA-specific neutralizing antibody and IgG antibody titers. We found that LCMV infection had a slight but statistically significant adverse impact on HA-specific IgG antibody titers compared to influenza infection alone (Fig 4I). In addition, despite heterologous infection or immunization priming inducing increased antigen-specific GC Tfh and GC B cells 8 days after influenza infection, all groups had similar HA-specific neutralizing antibody titers at all timepoints (S5G Fig). To determine if the enhanced early germinal center cellular response in heterologously primed mice corresponded to an increase in HA-specific long-lived plasma cells, we analyzed enriched B cells from bone marrow of infected mice for IgG secretion by ELISpot 42 and 105 days after influenza infection. As with our serology data, we found no differences in HA-specific IgG-secreting B cells from infected mice regardless of priming strategy (S5H–S5I Fig). Together these data suggest that while heterologous infection or immunization priming of CD4+ T cells did significantly enhance germinal center CD4+ T and B cell responses early after influenza infection, those effects did not significantly impact long-term germinal center-driven humoral responses compared to primary influenza infection. In addition, our data show that both adjuvanted rGP immunization and LCMV infection significantly enhanced the memory antigen-specific Th1 cell pool after influenza infection, despite differences in the CXCR5– non-Tfh cell populations prior to influenza infection.

Prior generation of antigen-specific memory CD4+ T cells enhanced early GC responses and long-term lung-resident Th1 cells upon influenza infection

Recent evidence has indicated an important role for CD4+ T resident memory (T_RM) cells in mediating protection from influenza infection in the lung [70–74]. Lung-resident CD4+ T cell responses result in the formation of T_RM with either Th1 or Tfh properties that can coordinate localized immune responses [75,76]. Furthermore, lung-specific immune responses are characterized by the induction of localized B cell responses and the formation of long-lived tissue-resident memory B cells that primarily home to the bronchoalveolar lymphoid tissue (BALT), and it is likely that localized antibody responses comprise a key line of defense against influenza infection in the lung [77–79]. To determine the impact of heterologous infection or immunization priming of CD4+ T cells on the establishment and boosting of secondary T_RM we assessed CD4+ T cell responses in the lung following influenza infection of primed and unprimed mice as previously described in Fig 2A. We employed intravascular anti-CD45 staining to distinguish lung-infiltrating leukocytes from those in circulation [55], combined with CD69 staining to identify T_RM [71]. Both primary adjuvanted rGP immunization and LCMV i.n. infection induced small numbers of lung-infiltrating memory I-A^b:gp66-77 tetramer+ CD4+ T cells detected at 39 dpi (S6A–S6C Fig) that were dramatically boosted in frequency and number at 8 days after influenza infection and were significantly higher compared to the PR8(1°) and PR8(1°)PR8(2°) groups (Fig 5A–5B). The priming strategy utilized also impacted the resulting secondary effector CD4+ T cell subsets. After PR8-HA-GP_61-80 infection, rGP immunization-induced memory CD4+ T cells preferentially gave rise to FR4+LY6C– Tfh-like secondary effector cells, whereas LCMV-induced memory T cells gave rise to FR4–LY6C+ Th1-like secondary effector cells (Fig 5C–5E).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Generation of memory CD4+ T cells by heterologous immunization induced increased GC B cells and Th1 cells in lung early following influenza infection.

Flow cytometry analysis of CD45-i.v.^negative CD4+ T cells and B cells from lung 8 days after PR8-HA-GP_61-80 influenza virus infection in heterologously primed mice (GP(1°)PR8(2°), filled triangle, or LCMV(1°)PR8(2°), filled diamond), homologously primed mice (PR8(1°)PR8(2°), filled hexagon) or unprimed naïve mice (PR8(1°), unfilled circle). Antigen-specific CD4+ T cell responses were analyzed either by staining with I-A^b:gp66-77 tetramer or cytokine expression following restimulation with LCMV gp61-80 peptide. (A) Representative flow plots of tetramer analysis of total CD4+ T cells. (B) Number of tetramer+ cells of total CD4+ T cells. (C) Representative flow plots of FR4 and LY6C analysis gated on tetramer+CD4+ T cells. (D) Number of LY6C+FR4– (Th1) cells of total tetramer+CD4+ T cells. (E) Ratio of the number of tetramer+ LY6C+FR4– (Th1) cells to number of tetramer+ LY6C–FR4+ (Tfh) cells, as determined using the gating strategy in panels A and C. (F) Representative flow plots of TNFα and IFNγ analysis gated on total CD4+ T cells. (G) Number of IFNγ+TNFα+ cells of total CD4+ T cells. (H) Representative flow plots of Fas and GL7 gated on CD19+ B cells. (I) Number of Fas+GL7+ GC B cells of total B cells. n ≥ 3 per group per experiment at each timepoint. Data shown are from one experiment and are representative of two to three independent experiments. Statistically significant p values of <0.05 are indicated and were determined using Mann-Whitney U test. Error bars represent Mean±SEM, *p≤0.05, **p≤0.01, ***p≤0.001, ****p≤0.0001.

https://doi.org/10.1371/journal.ppat.1011639.g005

Prior to PR8-HA-GP_61-80 infection, we analyzed lung-infiltrating CD4+ T cells for cytokine expression following ex vivo gp61-80 peptide restimulation and found that primary LCMV i.n. infection induced significantly more IFNγ- and TNFα-producing T cells in the lung compared to adjuvanted rGP immunization at 39 dpi (S6D–S6E Fig) similar to our data of CD4+ T cells in the lymph nodes and spleen. When we analyzed effector CD4+ T cells in the lung for cytokine expression 8 days after PR8-HA-GP_61-80 challenge, the LCMV(1°)PR8(2°) group had the highest expansion of CD4+ T cells producing IFNγ and TNFα (Fig 5F–5G), despite the presence of similar numbers of total tetramer+ CD4+ T cells to the GP(1°)PR8(2°) group (Fig 5B).

We then investigated the impact of heterologous infection or immunization priming of CD4+ T cells on GC B cells in the lung. We found that primary adjuvated rGP immunization and LCMV i.n. infection induced similar numbers of total B cells and GC B cells in the lung prior to PR8-HA-GP_61-80 challenge detected at 39 dpi (S6F–S6I Fig). However, 8 days after PR8-HA-GP_61-80 challenge, our data showed that the GP(1°)PR8(2°) and PR8(1°)PR8(2°) groups had significantly more GC B cells (CD19+GL7+Fas+) in the lung, as compared to the PR8(1°) group (Fig 5H–5I). To determine if these CD4+ T cells and B cells were co-localizing in the lung, we analyzed lung tissue by immunofluorescence. We found co-localization of CD4+ T cells and B220+ B cells in the GP(1°)PR8(2°), LCMV(1°)PR8(2°), and PR8(1°)PR8(2°) groups, but less so in the PR8(1°) group, suggesting the development of BALT in the primed mice (S7 Fig). Furthermore, GP(1°)PR8(2°) and LCMV(1°)PR8(2°) mice showed substantial interstitial T cell infiltration, PR8(1°) had mild interstitial T cell infiltration, and PR8(1°)PR8(2°) mice did not show significant interstitial T cell infiltration (S7 Fig).

We next sought to determine the impact of heterologous infection or immunization priming of CD4+ T cells on the establishment of lung-infiltrating memory CD4+ T cells following influenza infection. As was the case for the secondary effector response in the lung (Fig 5), PR8-HA-GP_61-80 rechallenge of rGP immunization- or LCMV infection-derived memory CD4+ T cells resulted in a large population of tetramer+ secondary memory T cells in the lung at 42 dpi as compared to the PR8(1°) and PR8(1°)PR8(2°) groups (Fig 6A–6B). Most of these cells expressed CD69, a marker of lung CD4+ T_RM following influenza infection [71,72], resulting in a 50-100-fold increase in CD4+ T_RM following heterologous rechallenge of rGP- or LCMV-derived CD4+ memory T cells (Fig 6C–6D). In addition, the memory CD4+ T cells maintained their primary activation-dependent Th1 and Tfh bias, as the LCMV(1°)PR8(2°) mice had significantly more LY6C+ Th1-like secondary memory T cells 42 days after influenza infection and the GP(1°)PR8(2°) mice had significantly more FR4+ Tfh-like secondary memory T cells (Fig 6E–6G). Overall, our findings showed that heterologous infection or immunization priming of CD4+ T cells induced large numbers of lung T_RM following PR8-HA-GP_61-80 rechallenge, with a Th1-like or Tfh-like subset distribution that was dependent on the primary immunization or infection challenge.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. Generation of memory CD4+ T cells by heterologous immunization enhanced long-term memory Th1 and CD4+ T_RM cells following influenza infection.

Flow cytometry analysis of CD45-i.v.^negative CD4+ T cells from lung 42 days after PR8-HA-GP_61-80 influenza virus infection in heterologously primed mice (GP(1°)PR8(2°), filled triangle, or LCMV(1°)PR8(2°), filled diamond), homologously primed mice (PR8(1°)PR8(2°), filled hexagon) or unprimed naïve mice (PR8(1°), unfilled circle). CD4+ T cells were analyzed by staining with I-A^b:gp66-77 tetramer. (A) Representative flow plots of CD4 and tetramer analysis of total CD4+ T cells. (B) Number of tetramer+CD44+ of total CD4+ T cells. (C) Representative flow plots of CD69 analysis gated on tetramer+CD4+ T cells. (D) Number of CD69+ T_RM cells of total tetramer+CD4+ T cells. (E) Representative flow plots of FR4 and LY6C analysis gated on tetramer+CD4+ T cells. (F) Number of LY6C+FR4– (Th1) cells of total tetramer+CD4+ T cells. (G) Ratio of the number of tetramer+ LY6C+FR4– (Th1) cells to number of tetramer+ LY6C–FR4+ (Tfh) cells. n ≥ 3 per group per experiment at each timepoint. Data shown are from one experiment and are representative of two to three independent experiments. Statistically significant p values of <0.05 are indicated and were determined using Mann-Whitney U test. Error bars represent Mean±SEM, *p≤0.05, **p≤0.01, ***p≤0.001, ****p≤0.0001.

https://doi.org/10.1371/journal.ppat.1011639.g006

Igh sequencing of reactivated plasmablasts suggests that heterologous priming did not significantly impact the repertoire diversity or shared clones compared to influenza infection alone

To determine if heterologous infection or immunization priming of CD4+ T cells markedly impacted the B cell clonal repertoire selection compared to mice infected with only influenza, we used the priming and influenza challenge experimental setups as previously described in Fig 2A, then 100 days after influenza challenge, mice were immunized i.p. with 10 μg recombinant HA (rHA) from PR8 influenza without adjuvant (Fig 7A) to preferentially engage HA-specific memory B cells to analyze secondary plasmablasts derived from the recalled HA-specific B cells. Five days after rHA immunization, IgD–CD19+B220^high/lowFas+CD138+ plasmablasts were sorted from spleens. Gating was determined using naïve control mice that lacked plasmablast formation (Figs 7A and S8A). Genomic DNA was isolated from sorted plasmablasts and Igh amplification and sequencing were performed using the immunoSEQ platform from Adaptive Biotechnologies.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. Generation of memory CD4+ T cells by heterologous immunization did not significantly impact long-lived plasmablast repertoire diversity compared to influenza infection alone.

105 days after influenza infection, unprimed mice (PR8(1°), unfilled circle), heterologously primed mice (GP(1°)PR8(2°), filled triangle, or LCMV(1°)PR8(2°), filled diamond), and homologously primed mice (PR8(1°)PR8(2°), filled hexagon) were immunized i.p. with 10 μg rHA to reactivate influenza-specific plasmablasts. 5 days postimmunization with rHA, IgD–CD19+B220+/low Fas+CD138+ plasmablasts were sorted from spleens and genomic DNA was isolated for Igh sequencing. (A) Schematic of experimental design. (B) Multidimensional scaling plot of CDR3 amino acid sequence repertoire overlap of individual mice. (C) Chao1 estimation of Igh repertoire diversity of plasmablasts from individual mice. (D) Simpson clonality diversity measure for all productive rearrangements of individual mice. (E) Heatmap of repertoire overlap analysis by Morisita overlap index of all mice pooled for each infection group. (F) Clonotype tracking analysis across priming groups of the ten largest CDR3 (amino acid sequence) clones by proportion of productive frequency shared in ≥5 infected mice (“public” clones). The productive frequency proportion value is the sum of a clone’s productive frequency in all individual mice. (G-J) Clonotype tracking analysis across priming groups of the ten largest CDR3 (amino acid sequence) clones by proportion of productive frequency shared in ≥2 mice in one priming group. (G) Ten largest CDR3 clones by proportion of productive frequency shared in ≥2 mice from the PR8(1°) (Unprimed) group. (H) Ten largest CDR3 clones by proportion of productive frequency shared in ≥2 mice from the GP(1°)PR8(2°) group. (I) Ten largest CDR3 clones by proportion of productive frequency shared in ≥2 mice from the LCMV(1°)PR8(2°) group. (J) Ten largest CDR3 clones by proportion of productive frequency shared in ≥2 mice from the PR8(1°)PR8(2°) group. n = 5 per group. Data shown are from one independent experiment. Error bars (panels C and D) are Mean±SD.

https://doi.org/10.1371/journal.ppat.1011639.g007

To investigate the overlap of individual mice repertoires, we performed multidimensional scaling (MDS) analysis on CDR3 amino acid sequences using the overlap coefficient of the repOverlap function of the Immunarch [61] package. There was no discernible clustering by priming strategy by MDS analysis (Fig 7B), indicating there was no significant impact on the repertoires of mice primed by the same strategy. We performed Chao1 estimation [80] and found no differences in clonal repertoire diversity richness by priming strategy (Fig 7C). We next analyzed the diversity of the productive rearrangements (rearrangements that produce functional B cell receptors) in individual mice by Simpson clonality measure, which is calculated as the square root of Simpson’s Index [81], which suggested that all repertoires skewed more polyclonal than mono- or oligoclonal (Fig 7D). When we compared the number of unique CDR3 amino acid sequence clonotypes to total clonotypes in individual mice, we found unique clones accounted for 85–95% of every individual repertoire (S8B Fig). In addition, we analyzed CDR3 clonotypes for differences in amino acid sequence length and number of somatic hypermutations (SHM) within nucleotide sequences and found no differences when compared by priming strategy (S8C–S8D Fig). We next performed the Morisita overlap index test [82–85] on CDR3 amino acid sequences pooled for all 5 mice in each group to evaluate the repertoire overlap by priming strategy. Our data suggest the PR8(1°) group had the most unique repertoire, while the GP(1°)PR8(2°) and LCMV(1°)PR8(2°) groups had more similar repertoires to one another (Fig 7E). When we analyzed the amino acid CDR3 sequences of individual mice with the Morisita overlap index test, we found that most of the GP(1°)PR8(2°) and LCMV(1°)PR8(2°) mice were more similar to each other than mice only infected with influenza (S8E Fig).

To evaluate shared CDR3 sequences and investigate proportional differences in mice by priming strategy, we tracked the largest 10 clonotypes by total proportion and shared in at least 5 of 20 total infected mice (partially shared clonotypes) using the trackClonotypes feature of the Immunarch [61] package (Fig 7F). We found trending differences in clonotype proportions, including increased proportions of the CARGGYW and CARGTYW clones and a lack of the CARGGYDGYYGAMDYW clone in the GP(1°)PR8(2°) group (Fig 7F). In addition, the CARHEVSYWYFDVW clone was found in mice only infected with PR8-HA-GP_61-80 (3 of 5 PR8(1°) mice and 4 of 5 PR8(1°)PR8(2°) mice) (Fig 7F). Only the CARGAYW clone was shared in all 20 infected mice, and thus had the largest representation by proportion (Fig 7F). Additionally, this clone was not contained in our control naïve CD19+Fas–IgD+ B cells. We then analyzed the 10 clonotypes largest by proportion in each priming strategy group shared in at least 2 of 5 mice (Fig 7G–7J). Our data showed that the largest shared clone, CARGAYW, was less represented proportionally in the PR8(1°) group while unique clones, including CVQMEERPPLFTYW, were more largely represented (Fig 7G). In addition, our data show the 10 proportionally largest clones in the PR8(1°) group comprised over 2-fold more of the total pooled repertoire proportion (>4% total) compared to the other three groups (all <2% total) (Fig 7G), and with the Morisita overlap data (Fig 7E) suggests a more unique repertoire for the PR8(1°) group. Together, these data show that while the plasmablast repertoires of individual mice were dominated by unique clones, analyses of the partially shared clones among individual mice were able to characterize differences in the representation of specific clonal sequences by priming strategy.