Generation of antigen-specific memory CD4 T cells by heterologous immunization enhances the magnitude of the germinal center response upon influenza infection

Current influenza vaccine strategies have yet to overcome significant obstacles, including rapid antigenic drift of seasonal influenza viruses, in generating efficacious long-term humoral immunity. Due to the necessity of germinal center formation in generating long-lived high affinity antibodies, the germinal center has increasingly become a target for the development of novel or improvement of less-efficacious vaccines. However, there remains a major gap in current influenza research to effectively target T follicular helper cells during vaccination to alter the germinal center reaction. In this study, we used a heterologous infection or immunization priming strategy to seed an antigen-specific memory CD4+ T cell pool prior to influenza infection in mice to evaluate the effect of recalled memory T follicular helper cells in increased help to influenza-specific primary B cells and enhanced generation of neutralizing antibodies. We found that heterologous priming with intranasal infection with acute lymphocytic choriomeningitis virus (LCMV) or intramuscular immunization with adjuvanted recombinant LCMV glycoprotein induced increased antigen-specific effector CD4+ T and B cellular responses following infection with a recombinant influenza strain that expresses LCMV glycoprotein. Heterologously primed mice had increased expansion of secondary Th1 and Tfh cell subsets, including increased CD4+ TRM cells in the lung. However, the early enhancement of the germinal center cellular response following influenza infection did not impact influenza-specific antibody generation or B cell repertoires compared to primary influenza infection. Overall, our study suggests that while heterologous infection/immunization priming of CD4+ T cells is able to enhance the early germinal center reaction, further studies to understand how to target the germinal center and CD4+ T cells specifically to increase long-lived antiviral humoral immunity are needed.

148 The epitope was inserted in-frame at the amino acid position 135, that is highly tolerant to small insertions 149 (53). Next, the recombinant virus was rescued by transfecting cells with 8 plasmids containing the 150 sequences of the viral segments, as previously described (54). (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted August 31, 2023. ; https://doi.org/10.1101/2023.08.29.555253 doi: bioRxiv preprint . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made   (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted August 31, 2023. ; https://doi.org/10.1101/2023.08.29.555253 doi: bioRxiv preprint 266 by i.n. inoculation with 2x10 5 PFU acute LCMV-Armstrong (LCMV(1 o ) group) (Fig 1A).  . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted August 31, 2023. ; https://doi.org/10.1101/2023.08.29.555253 doi: bioRxiv preprint 289 We next evaluated if using heterologous priming with adjuvanted rGP immunization or i.n. LCMV 290 infection to generate antigen-specific memory CD4+ T cells would enhance the early effector germinal 291 center response to influenza infection. Using the heterologous priming strategies detailed in Figure 1A, 42 292 days after priming infection or immunization we infected i.n. with 500 TCID 50 PR8-HA-GP 61-80 293 recombinant influenza virus (GP(1 o )PR8(2 o ) and LCMV(1 o )PR8(2 o ) groups) (Fig 2A). For control groups, 294 we used age-and sex-matched naïve mice infected i.n. with PR8-HA-GP 61-80 to evaluate the primary 295 influenza response (PR8(1 o ) group) and a separate group was homologously primed with PR8-HA-GP 61-80 296 i.n. infection (PR8(1 o )PR8(2 o ) group) to evaluate recalled cellular and antibody responses (Fig 2A).   (Fig S2C). We also evaluated 313 differences in tetramer+ Th1 and Tfh cells in the spleen and found that Th1 cells (by both TBET and IFN . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted August 31, 2023.

333
To determine if there was a correlation between the numbers of tetramer+ GC Tfh cells and HA-334 specific GC B cells in medLN, we performed Spearman correlation analysis and found that in all groups 335 there was a statistically significant positive correlation (Fig 3I). In addition, linear regression analysis of 336 GC Tfh and HA-specific GC B cell numbers showed statistically significant positive associations in all 337 groups (Fig 3I), consistent with previous studies describing the critical interaction between Tfh and B cells 338 in the germinal center (26, 48, 61-63). We further compared the numbers of HA-specific GC B cells to the . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted August 31, 2023. ; https://doi.org/10.1101/2023.08.29.555253 doi: bioRxiv preprint 339 number of GC Tfh cells and found that in the GP(1 o )PR8(2 o ) group, there was a significantly higher number 340 of HA-specific GC B cells to every GC Tfh cell (Fig 3J). These data suggest that while the number of GC 341 Tfh cells in the GP(1 o )PR8(2 o ) group were similar to the PR8(1 o ) group, the GC Tfh cells may be of a higher 342 quality as to sustain support for higher numbers of HA-specific GC B cells, as has been previously 343 suggested (41, 44). Together, these data suggest that heterologous priming with adjuvanted rGP 344 immunization or LCMV infection significantly enhanced the early anti-influenza germinal center B cell 345 response following influenza infection compared to primary influenza infection.

361
As the CD4+ T cell-specific immunodominant LCMVgp61-80 epitope contains a cryptic epitope 362 recognized by CD8+ T cells (64), we analyzed CD8+CD44+ T cells at 8 and 42 days after influenza 363 challenge for IFN expression following LCMVgp61-80 peptide restimulation. Our data show that most . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted August 31, 2023. ; https://doi.org/10.1101/2023.08.29.555253 doi: bioRxiv preprint 364 mice had frequencies of CD44+IFN+ cells of CD8+ T cells below background levels as normalized to a 365 no peptide control, and all other mice had less than 1% of CD8+ T cells expressing CD44 and IFN (Fig   366 S4E). These data suggest that while non-specific secretion of IFN by CD8+ T cells was detected, we do 367 not expect these cells significantly influenced the increases in Th1 cells and IFN-secreting CD4+ T cells.   Fig 4G). When we analyzed HA-specific GC B cell kinetics at 8, 15, and 382 42 days after influenza infection, HA-specific GC B cells underwent contraction after peak expansion 383 around 8 dpi in heterologously primed mice (Fig 4H). However, in the PR8(1 o ) group HA-specific GC B 384 cells increased in number after 8 dpi, though numbers were not significantly different from heterologously 385 primed at 15 or 42 dpi in this experiment (Fig 4H).

386
As heterologous infection/immunization priming significantly increased HA-specific GC B cells 387 and plasmablasts 8 days after influenza infection, we analyzed the sera of influenza infected mice to 388 determine HA-specific neutralizing antibody and IgG antibody titers. We found that LCMV infection had . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted August 31, 2023. ; https://doi.org/10.1101/2023.08.29.555253 doi: bioRxiv preprint 389 a slight but statistically significant adverse impact on HA-specific IgG antibody titers compared to influenza 390 infection alone (Fig 4I). In addition, despite heterologous infection/immunization priming inducing 391 increased antigen-specific GC Tfh and GC B cells 8 days after influenza infection, all groups had similar 392 HA-specific neutralizing antibody titers at all timepoints (Fig S4F). To determine if the enhanced early 393 germinal center cellular response in heterologously primed mice corresponded to an increase in HA-specific 394 long-lived plasma cells, we analyzed enriched B cells from bone marrow of infected mice for IgG secretion 395 by ELISpot 42 and 105 days after influenza infection. As with our serology data, we found no differences 396 in HA-specific IgG-secreting B cells from infected mice regardless of priming strategy (Figs S4G-H).

440
We next sought to determine the impact of heterologous infection/immunization priming of CD4+ 441 T cells on the establishment of lung-infiltrating memory CD4+ T cells following influenza infection. As 442 was the case for the secondary effector response in the lung (Fig 5)  To determine if heterologous infection/immunization priming of CD4+ T cells markedly impacted 459 the B cell clonal repertoire selection compared to mice infected with only influenza, we used the priming 460 and influenza challenge experimental setups as previously described in Figure 2A, then 100 days after 461 influenza challenge, mice were immunized i.p. with 10 g recombinant HA (rHA) from PR8 influenza . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted August 31, 2023. ; https://doi.org/10.1101/2023.08.29.555253 doi: bioRxiv preprint 462 without adjuvant (Fig 7A) to preferentially engage HA-specific memory B cells to analyze secondary 463 plasmablasts derived from the recalled HA-specific B cells. Five days after rHA immunization, IgD-464 CD19+B220 high/low Fas+CD138+ plasmablasts were sorted from spleens (Figs 7A and S6A). Genomic DNA 465 was isolated from sorted plasmablasts and Igh amplification and sequencing were performed using the 466 immunoSEQ platform from Adaptive Biotechnologies.

467
To investigate the overlap of individual mice repertoires, we performed multidimensional scaling 468 (MDS) analysis on CDR3 amino acid sequences using the overlap coefficient of the repOverlap function 469 of the Immunarch (58) package. There was no discernible clustering by priming strategy by MDS analysis 470 (Fig 7B), indicating there was no significant impact on the repertoires of mice primed by the same strategy. 471 We performed Chao1 estimation (76) and found no differences in clonal repertoire diversity richness by 472 priming strategy (Fig 7C). We next analyzed the diversity of the productive rearrangements 473 (rearrangements that produce functional B cell receptors) in individual mice by Simpson clonality measure, 474 which is calculated as the square root of Simpson's Index (77), which suggested that all repertoires skewed 475 more polyclonal than mono-or oligoclonal (Fig 7D). When we compared the number of unique CDR3 476 amino acid sequence clonotypes to total clonotypes in individual mice, we found unique clones accounted 477 for 85-95% of every individual repertoire (Fig S6B). In addition, we analyzed CDR3 clonotypes for 478 differences in amino acid sequence length and number of somatic hypermutations (SHM) within nucleotide 479 sequences and found no differences when compared by priming strategy (Figs S6C-D). We next performed 480 the Morisita overlap index test (78-81) on CDR3 amino acid sequences pooled for all 5 mice in each group 481 to evaluate the repertoire overlap by priming strategy. Our data suggest the PR8(1 o ) group had the most 482 unique repertoire, while the GP(1 o )PR8(2 o ) and LCMV(1 o )PR8(2 o ) groups had more similar repertoires to 483 one another (Fig 7E). When we analyzed the amino acid CDR3 sequences of individual mice with the 484 Morisita overlap index test, we found that most of the GP(1 o )PR8(2 o ) and LCMV(1 o )PR8(2 o ) mice were 485 more similar to each other than mice only infected with influenza (Fig S6E).

486
To evaluate shared CDR3 sequences and investigate proportional differences in mice by priming 487 strategy, we tracked the largest 10 clonotypes by total proportion and shared in at least 5 of 20 total infected . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted August 31, 2023. ; https://doi.org/10.1101/2023.08.29.555253 doi: bioRxiv preprint 488 mice ("public" clonotypes) using the trackClonotypes feature of the Immunarch (58) package (Fig 7F). We 489 found trending differences in clonotype proportions, including increased proportions of the CARGGYW 490 and CARGTYW clones and a lack of the CARGGYDGYYGAMDYW clone in the GP(1 o )PR8(2 o ) group 491 (Fig 7F). In addition, the CARHEVSYWYFDVW clone was found in mice only infected with PR8-HA-492 GP 61-80 (3 of 5 PR8(1 o ) mice and 4 of 5 PR8(1 o )PR8(2 o ) mice) (Fig 7F). Only the CARGAYW clone was 493 shared in all 20 infected mice, and thus had the largest representation by proportion (Fig 7F). Additionally, 494 this clone was not contained in our control naïve CD19+Fas-IgD+ B cells (data not shown). We then 495 analyzed the 10 clonotypes largest by proportion in each priming strategy group shared in at least 2 of 5 496 mice (Figs 7G-J). Our data showed that the largest shared clone, CARGAYW, was less represented 497 proportionally in the PR8(1 o ) group while unique clones, including CVQMEERPPLFTYW, were more 498 largely represented (Fig 7G). In addition, our data show the 10 proportionally largest clones in the PR8(1 o ) 499 group comprised over 2-fold more of the total pooled repertoire proportion (>4% total) compared to the 500 other three groups (all <2% total) (Fig 7G), and with the Morisita overlap data (Fig 7E) suggests a more 501 unique repertoire for the PR8(1 o ) group. Together, these data show that while the plasmablast repertoires 502 of individual mice were dominated by unique clones, analyses of the shared clones among individual mice 503 were able to characterize differences in the representation of specific clonal sequences by priming strategy.

504
. CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted August 31, 2023. ; https://doi.org/10.1101/2023.08.29.555253 doi: bioRxiv preprint 505 Discussion 506 Current vaccine strategies, including seasonal influenza vaccines, are not specifically designed to 507 engage CD4+ T cells, despite their necessity in germinal center formation and long-lived humoral 508 immunity, as well as their contribution to cellular immunity in infected tissues. In this study, we used 509 heterologous priming with adjuvanted rGP immunization or LCMV intranasal infection to generate 510 memory CD4+ T cells and investigate the effects of recalled memory CD4+ Tfh cells and established T RM 511 cells on the response to influenza challenge. Our findings demonstrated that heterologous 512 infection/immunization priming induced a population of antigen-specific memory CD4+CXCR5+ Tfh cells 513 that were successfully recalled to secondary effector GC Tfh cells and induced an increased magnitude of 514 HA-specific GC B cells compared to primary influenza infection. Furthermore, while LCMV-primed mice 515 had significantly higher GC Tfh cells 8 dpi, our data suggested rGP-immunization priming produced higher 516 quality GC Tfh cells as these mice had a significantly higher ratio of HA-specific GC B cells to GC Tfh 517 cells. Heterologous infection/immunization priming also induced increased secondary effector CXCR5-518 Th1 cells that expressed both TBET and IFN, which were maintained at a higher magnitude even at 519 memory. In addition, heterologous infection/immunization priming generated an increased long-lived 520 CD4+ T RM pool and induced increased expansion of recalled antigen-specific CD4+ T cells in the lung after 521 influenza challenge. Interestingly, the skewing of lung-infiltrating CD4+ T cells was dependent on priming 522 activation, as rGP immunization-primed mice preferentially recalled Tfh-like cells compared to LCMV-523 primed mice that preferentially recalled Th1-like cells. However, despite the early enhancement of the 524 germinal center cellular response after influenza challenge, heterologous infection/immunization priming 525 of CD4+ T cells did not enhance HA-specific antibody titers. Overall, our findings suggest that heterologous 526 infection/immunization priming of CD4+ T cells can be used to enhance both the early GC response, 527 including the GC Tfh and GC B cell magnitude, and establishment of CD4+ T RM cells that respond to 528 influenza challenge.
. CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made

567
Prior studies investigating the recall of memory CD8+ T cells in heterosubtypic influenza infection 568 have shown that protective CD8+ T RM cells were found to undergo robust clonal expansion after secondary 569 infection and express large amounts IFN, though the secondary effectors were dominated by recognition 570 of a single immunodominant epitope (124-128). As one study found neither infection of the lung nor antigen 571 persistence was required for establishment in the lung of antigen-specific CD8+ T cells (126), we found 572 similar results in our study investigating CD4+ T cells as adjuvanted rGP immunization showed minimal 573 lung-resident memory CD4+ T cells prior to influenza challenge but had significantly expanded secondary 574 effector CD4+ T cells and CD4+ T RM in the lung compared to primary influenza infection, suggesting either 575 increased trafficking to the lung or a larger antigen-specific memory T RM pool compared to naïve mice.

576
In agreement with the idea that Tfh cells are the limiting cell subset in the GC reaction and the 577 generation of GC-derived products, following heterologous influenza rechallenge of memory CD4+ T cells 578 we saw an early increased magnitude of antigen-specific GC Tfh and GC B cells. However, additional 579 studies are needed to assess the direct impact of heterologous infection/immunization priming of CD4+ T . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made  . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted August 31, 2023.

614
. CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted August 31, 2023. ; https://doi.org/10.1101/2023.08.29.555253 doi: bioRxiv preprint . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted August 31, 2023. ; https://doi.org/10.1101/2023.08.29.555253 doi: bioRxiv preprint . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted August 31, 2023. ; https://doi.org/10.1101/2023.08.29.555253 doi: bioRxiv preprint . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted August 31, 2023. ; https://doi.org/10.1101/2023.08.29.555253 doi: bioRxiv preprint