bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Redefining CD4 T cell residency: Helper T cells orchestrate protective humoral immunity in the lung
Nivedya Swarnalekha1,2,*, David Schreiner1,2,*, Ludivine C Litzler1,2, Saadia Iftikhar3, Daniel Kirchmeier1,2, Marco Künzli1,2, Carolyn G King1,2,4
1 Immune Cell Biology Laboratory 2 Department of Biomedicine, University of Basel, University Hospital Basel CH-4031, Basel Switzerland 3 Personalised Health Oncology Basel, University of Basel 4 Lead contact
*These authors contributed equally to this work
To whom correspondence should be addressed:
Carolyn King Department of Biomedicine University of Basel Hebelstrasse 20 CH-4031 Basel, Switzerland Email: [email protected] Tel: +41 78 907 15 68
bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Abstract
Influenza is a severe and acute respiratory pathogen, and a significant cause for
morbidity, particularly in young children and the elderly. Following influenza infection,
clonally expanded T cells take up permanent residence in the lung where they are
poised to rapidly respond to challenge infection. The non-circulating status of these
tissue resident memory (TRM) cells makes them an attractive target for vaccination.
While many studies have characterized CD8 TRM cells, less is known about the
heterogeneity and protective capacity of CD4 TRM cells. Here we characterized the
dynamics and transcriptional regulation of lung resident CD4 T cells to define a non-
lymphoid signature that removes the bias created by the prevalence of Th1 helper cells
during viral infection. We identified a novel population of long-lived T resident helper
(TRH) cells that requires intrinsic Bcl6 expression for their differentiation. Although TRH
cells also depend on B cells, they are generated independently of T follicular helper
effector cells in the lymph node. In contrast to lung resident Th1 cells, TRH cells are
tightly co-localized with B cells in inducible Bronchus Associated Lymphoid Tissue
(iBALT). Deletion of Bcl6 in CD4 T cells prior to heterotypic challenge infection results
in redistribution of CD4 T cells outside of iBALT areas and impaired local antibody
production. These data highlight lung iBALT as a niche for the homeostasis and
survival of TRH cells, and further suggest that vaccination strategies to selectively
induce TRH cells can improve protective immunity in the tissue.
1 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Main Seasonal influenza epidemics are a major cause of global morbidity and
mortality. Although annually administered influenza vaccines are among the most
widely used in the world, vaccine-elicited neutralizing antibodies offer poor protection
against new influenza strains1. In contrast, there is evidence that prior influenza
infection can accelerate viral clearance after heterotypic infection in both mice and
humans2, 3, 4, 5, 6. Emerging data suggests that the targeted generation of CD4 memory
T cells recognizing conserved epitopes from internal viral proteins may form the basis of
a universal influenza virus vaccine7, 8. CD4 memory T cells are induced following
immunization or infection and can be recalled to generate secondary effectors during a
challenge infection. Several subsets of CD4 memory cells have been described,
including central memory (TCM) and effector memory (TEM) cells which circulate
through secondary lymphoid and non-lymphoid tissues9. More recently, tissue resident
memory (TRM) cells that persist in barrier tissues such as lung and skin have been
described10. Although CD4 T cells actually outnumber CD8 T cells in barrier tissues,
the majority of studies have focused on the requirements for CD8 TRM cell
differentiation. In addition, although CD4 T cells are renowned for their substantial
plasticity during immune responses, less is known about diversification within the CD4
TRM cell compartment11, 12, 13, 14, 15, 16.
Influenza infection induces the differentiation of CD4 TRM cells in the lung, where they
are maintained in an antigen and inflammation independent manner17. Following a
lethal re-challenge, influenza specific CD4 TRM cells rapidly produce effector cytokines
and promote both viral clearance and host survival7. Lung CD4 TRM cells can also be
2 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
induced by mucosal vaccination and were shown to mediate superior protection
following heterologous infection, highlighting their potential as a universal vaccine
target18. Influenza-specific CD4 TRM cells are generally characterized as Th1-like, with
the capacity to produce both IFNγ and IL-27, 19. However, IL-2 deficient memory CD4 T
cells were recently shown to provide superior protection compared to wild-type memory
cells, an outcome that correlated with decreased inflammation and host pathology
during re-challenge20. These data suggest that protection mediated by CD4 TRM cells
is not strictly dependent on their ability to produce effector cytokines, and that a
balanced secondary response is likely to involve the recruitment and coordination of
distinct and specialized CD4 TRM cell subsets21.
Heterogeneity within the non-antigen specific CD4 TRM cell compartment was recently
addressed in a study that reported enrichment of genes associated with the TNFRSF
and NFKB pathways in barrier T cells compared to T cells isolated from their respective
draining lymphoid compartments11. The residency signature derived from this dataset,
however, does not take into consideration tissue versus lymphoid organ differences
between distinct T cell subsets. Accordingly, this approach tends to over-represent
genes associated with type 1 helper T cells such as Klrg1, Itgae, Id2, and CXCR6,
which may comprise a poorer definition of residency for other Th cell subsets.
Consistent with this idea, the authors reported the presence of a stronger TRM
phenotype in lymphoid Th1 memory cells compared to lymphoid TFH cells which they
attributed to the relative ability of these cells to adapt to barrier tissues. On the other
hand, it is well appreciated that TFH cells share many surface markers and molecular
3 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
dependencies with TRM cells, including high expression of PD1, P2X7R, CD69 and
ICOS, and a requirement for S1PR1 and KLF2 downregulation to develop10, 22, 23, 24, 25.
In support of this idea, TFH memory cells isolated from the spleen after LCMV infection
were recently demonstrated to have a partially overlapping transcriptional signature with
TRM cells, consistent with the non-circulating nature of both subsets26. In contrast,
another study addressing the relationship between TRM cells present in secondary
lymphoid organs and TFH memory cells reported more distinct transcriptional profiles of
these subsets12. However, in this case the authors focused on gene expression in T
cell receptor (TCR) transgenic cells that have relatively impaired TFH memory cell
generation compared to polyclonal antigen specific cells26. In addition, although the
authors reported a strong overlap between CD4 and CD8 TRM signatures after LCMV
infection, these signatures are likely to be skewed by the induction of genes expressed
by cytotoxic and type 1 cytokine producing cells. Thus, an open question is whether
different tissue resident Th cell subsets express conserved or distinct residency
signatures compared to their lymphoid counterparts.
In this study we characterized the dynamics and diversification of polyclonal CD4 T cells
present in the lung and draining LN after influenza, using the resolution provided by
scRNAseq to decouple residency and functional signatures between these two
compartments. Our analyses of influenza-specific CD4 T cells reveal striking
heterogeneity within the lung compartment comprised of two broad subsets which we
designate tissue resident helper (TRH) and tissue resident memory 1 (TRM1) cells.
TRH cells persist stably and depend on expression of the transcription factor Bcl6 for
their differentiation. TRH are tightly localized together with B cells in the lung and
4 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
require ongoing antigen presentation for their maintenance. Importantly, CD4 T cell
intrinsic deletion of Bcl6 at late time points after primary infection impairs the local
humoral response upon reinfection. These data identify a novel TRH subset that may
be a rational target to drive potent and protective immunity in the lung mucosa11, 12.
Results
Residency signature in CD4 T cells is biased toward Th1 memory cells
To examine influenza-specific CD4 memory T cell populations, we performed
scRNA-seq on memory CD4 T cells from the lung and lung-draining mediastinal LN.
Virus specific T cells were detected by tetramer staining for IAb:NP311-325 and were
largely protected from intravascular antibody staining for CD45 (Fig. 1a, Supplementary
Fig. 1a). Treatment of mice with fingolimod (FTY720) to block the egress of memory
cells from lymphoid organs did not alter the total number of NP+ CD4 T cells isolated
from the lung, indicating that this population can be sustained without input from the
circulation at this time point (Supplementary Fig. 1b, c). Following quality control,
normalization and dimension reduction, the transcriptomes of lung and LN T cells were
clearly distinct, and an expected small population of putative circulating cells identified
by elevated expression of S1pr1, Il7r and Bcl2 was also detected in the lung (Fig. 1b, c,
Supplementary Table 1a). Differential expression analysis between lung and LN cells
confirmed enrichment of previously reported CD4 residency genes (e.g. Crem, Vps37b,
Rora, Ramp3, Tnfrsf18) in the lung (Fig. 1c)11, 12. To determine if the residency
signature was conserved in another viral infection model we analyzed GP66 specific
memory T cells from the liver and spleen of LCMV infected mice (Fig 1d). Lung specific
genes from influenza were broadly enriched in liver cells from LCMV and vice versa
5 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
(Supplementary Fig. 1d). To better define a conserved residency signature, we took the
intersection of genes differentially enriched in non-lymphoid tissues (NLT) from both
viral infections (Fig. 1e). This "pseudo-bulk", tissue-level analysis demonstrates an
apparent bias in NLT toward a Th1 phenotype, with Ccl5, Crip1, Vim, Nkg7, Id2 and
Cxcr6 among the top shared genes (Fig. 1e, f). Unsurprisingly, a multi-tissue signature
for CD8 TRM was also enriched in lung and liver, consistent with similarities between
the transcriptional regulation of cell identity among type I cytokine producing CD4 and
CD8 cells (Fig. 1g). Conversely, TFH signatures including genes such as Tcf7,
Izumo1r, Dennd2d, Shisa5, Rgs10 and Cxcr5 were enriched in lymphoid tissues (LT), in
line with the conception of TFH as an SLO-resident population (Fig. 1e, f)22. The
presence of Th subset-specific genes in both NLT and LT cells reported here and
elsewhere calls for a refinement of the residency signature that separates function from
niche11, 12.
Heterogeneity in the lung: pan residency versus Th-subset specific residency
We next investigated heterogeneity specifically within the NP-specific lung CD4 T cell
compartment. Unsupervised hierarchical clustering classified the cells into four
populations (Fig. 2a). Notably, we found clear evidence of a TFH-like phenotype in the
lung: cluster 3 showed increased expression of Sostdc1, Sh2d1a, Ppp1r14b, Rgs10, Id3
and Tcf7 (Fig 2b, Supplementary Table 2a). Although this TFH-like cluster was lower in
many of the genes characterizing residency compared to the other lung clusters,
expression was still clearly higher compared to cells in the LN (Supplementary Fig. 2a).
Cells in cluster 2 had a Th1 phenotype including enrichment for Selplg, Nkg7, Ccl5, Id2
6 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
and Cxcr6. The similarity of these clusters to known T-helper subsets was further
confirmed by scoring each cell according to published TFH and Th1 memory gene sets
(Fig. 2c). Elevated transcription of S1pr1, Il7r and Bcl2 suggested that cluster 4, distinct
in principal component 2 and further distinguished by high ribosomal protein content,
contained circulating cells recovered during lung processing. Strikingly, while cluster 1
was most similar to the TFH-like cluster, it had a distinct profile characterized by
enrichment for Hif1a, Areg, Tnfrsf4 and Tnfsf8 and also expressed more Nr4a3,
suggesting ongoing MHC-II engagement and activation of the calcineurin/NFAT
pathway (Fig. 2a-c, Supplementary Fig. 2b)27. Cells in cluster 1 were also enriched for
Rora, which is reportedly induced by extrinsic signals in the microenvironment and plays
a role in negatively regulating lung inflammation during infection28.
We next focused our analysis on NP-specific CD4 T cells from the draining LN, cutting
the resulting hierarchical clustering tree to four clusters (Fig. 2d). These clusters were
classified similarly to those found in the lung, with a notably small Th1 cluster 4
expressing Nkg7, Ccl5, Id2 and Cxcr6 and tracking with ribosomal protein-rich cluster 1
along PC1; strongly TFH cluster 3 with high expression of Pdcd1, Sh2d1a, Sostdc1; and
TFH-like cluster 2 enriched for Izumo1r, Tox, and Hif1a (Supplementary Fig. 2c,
Supplementary Table 2b). As with the lung data, we scored LN cells using TFH and
Th1 gene sets to further verify this classification (Supplementary Fig. 2d). To arrive at a
less biased signature for lung residency, we performed differential expression analysis
on each of the most phenotypically similar cluster pairs between lung and LN, for
example, comparing LN TFH with TFH-like lung cells and LN Th1 with Th1-like lung
cells. The final conserved residency signature consisted only of genes that were
7 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
differentially enriched in all of the non-circulating lung clusters over their LN
counterparts (Fig. 2e). This enrichment was confirmed in our LCMV data set as well as
other recently published TRM data (Supplementary Fig. 2e-g). Subsequent removal of
these conserved residency genes allowed us to identify Th-subset specific genes
differentially expressed between tissue and lymphoid compartments. Th1-like cells in
the lung were enriched over their LN counterparts for genes including Ctla2a, Crip1,
Lgals1, Tnfrsf18 and Infgr1 while TFH-like lung cells exhibited higher expression of Mt1,
Hspa1a, Klf6, Tnfsf8, and Areg (Fig. 2f, g, Supplementary Table 2c, d). In addition, we
compared the two TFH-like clusters in the lung and found a more lymphoid signature in
cluster 3 and enrichment of the residency signature in cluster 1, suggesting functional
diversification of TFH-like cells in the lung (Fig. 2h, Supplementary Table 2e).
To gain further insight into the transcriptional regulation of distinct resident CD4 T cell
subsets, we next assessed transcription factor (TF) activity using single-cell regulatory
network inference and clustering (SCENIC)29. This approach identified active regulatory
activity in TFs such as Hif1a, Crem, Fosl2, and known Th1-related factors Prdm1,
Runx2 and Runx3 (Supplementary Fig. 2h). Cluster-specific analysis revealed that the
activity of Prdm1, Runx2 and Runx3 was limited to Th1-like clusters while Hif1a activity
was focused primarily in lung cluster 1, mirroring gene expression of Hif1a. Bcl6
regulatory activity was detected in lung TFH-like cluster 3 and LN TFH cluster 3, along
with Cebpa, which is reported to inhibit both IFNγ production and TCR-driven
proliferation (Supplementary Fig. 2i)30, 31. Interestingly, Foxp3-associated regulatory
activity was enriched in lung over LN although there was only minimal expression of
8 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Foxp3 transcript in the data set and Foxp3+ cells were not detected by FACS in the
antigen-specific compartment (Supplementary Fig. 2j, k). In summary, the scRNA-seq
data gathered here constitute an extensive picture of CD4 T cell memory to influenza
and disentangle residency from helper function in order to highlight not only conserved
differences between NLT and LT, but also intra-tissue heterogeneity and functional
diversification of lung resident T cells.
Phenotypic characterization of resident CD4 T cell subsets
To determine if the heterogeneity detected by scRNAseq was also mirrored at
the protein level, we examined the phenotype of NP-specific CD4 T cells using flow
cytometry. At day 30 after infection, NP-specific CD4 T cells in the lung could be
divided into two major subsets with reciprocal expression of FR4 and PSGL1 (Fig. 3a).
Following viral clearance and T cell contraction, the number of cells falling within these
two T resident cell subsets remained stable to at least 120 days after infection (Fig. 3b).
Treatment with FTY720 at day 30 after infection did not change the number of cells in
either subset, indicating that both populations are resident (Supplementary Fig. 3a).
FR4hiPSGL1low (hereafter T resident helper, TRH) cells expressed many markers
associated with TFH effector cells including PD1, CXCR5, Nur77, Bcl6, P2X7R, CD73,
CXCR4 and ICOS (Fig. 3c, d). FR4loPSGL1hi (hereafter TRM1) cells expressed lower
levels of TFH-associated markers, but higher levels of markers associated with Th1
cells including CXCR6, T-bet and CD11a (Fig. 3e, f). Importantly, adoptively transferred
OT-II T cell receptor transgenic cells activated by infection with PR8-OVA2 generated
mostly TRM1 cells (Supplementary Fig. 3b, c). These data are consistent with earlier
9 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
reports using TCR transgenic strains that primarily identified Th1-like memory CD4 T
cells in the lung after influenza infection. Taken together, scRNAseq and flow cytometry
confirm the presence of TRM1 (Th1-like) and TRH (TFH-like) cells in the lung and
underscore the importance of studying polyclonal CD4 T cell responses to understand
T-dependent immunity.
Progressive differentiation of resident CD4 T cell subsets
To further investigate the kinetics of TRM1 and TRH cell differentiation in the
lung, we next characterized the phenotype of NP-specific CD4 T cells at multiple time
points after infection. At day 9, the majority of CD4 T cells in the lung had high
expression of PSGL1, bimodal expression of FR4 and were negative for TFH markers
CXCR5 and PD1 (Fig. 4a, b). By day 14 after infection, a small proportion of TRH cells
emerged in the lung, followed by a clear separation of TRM1 and TRH cell subsets by
day 30 (Fig. 4a). In contrast to the lung, fully differentiated NP-specific TFH effector
cells (CXCR5+PD1+) were clearly detected in the mediastinal LN at day 9 (Fig. 4b). To
understand if LN TFH effectors give rise to TRH cells, we began continuously treating
mice with FTY720 between days 9 and 30 to prevent the migration of LN-primed TFH
cells into the lung. At day 30 after infection, the number of TRH cells in FTY720 treated
mice was equivalent to the number of TRH cells in PBS treated mice, indicating that full
differentiation of lung TRH cells can be independently completed in the lung (Fig. 4c).
To investigate the transcriptional regulation of resident T cell generation, we expanded
the scRNA-seq analysis to include NP-specific CD4 T cells from days 9 and 14 after
infection. Cells were clearly discriminated by time point and tissue in reduced
10 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
dimensions, with day 9 cells appearing more distinct compared to days 14 and 30 (Fig.
3d, Supplementary Fig. 4a). Day 9 lung cells exhibited the lowest per cell gene counts
of any sample, which may reflect their susceptibility to apoptosis under inflammatory
conditions. Nevertheless, hierarchical clustering revealed a group of cells (cluster 1)
enriched for Stat4, Il7r, Icos, S1pr1, and Selplg, which may represent recent immigrants
from the lymph node (Fig. 4e). Consistent with the FACS analysis, we observed an
enrichment of Th1-like characteristics at day 9 while TRH cells were more prominent at
day 30 (Fig. 4f). To investigate the expression of Izumo1r and Selpg, encoding the
proteins FR4 and PSGL1, respectively, we used imputed values, replacing zeros with
values estimated from similar cells. In parallel with FACS expression of these markers,
day 9 lung cells were Selplghi and Izumo1rlo, while Selplglo and Izumo1rhi T cells
emerged at day 14 and were increased by day 30 (Supplementary Fig. 4b). In contrast,
the majority of cells in the LN were Izumo1rhi at all time points. Transcription factor
activity according to SCENIC analysis tracked the trend from Th1-like at day 9 (Runx2,
Runx3, Prdm1) to TFH-like at day 30 (Lef1, Pou2af1) (Fig. 4g)32, 33. In addition, a
consistent subset of genes associated with lung residency (Tnfrsf4, Vps37b, Fth1,
Crem) was detected by differential expression analysis between lung and LN at each
separate time point (Supplementary Fig. 4c). Other genes associated with residency at
day 30 (Gch1, Hif1a, Selenok, Prkca) appeared more gradually, potentially reflecting the
dynamics of signal access in the tissue microenvironment and the relatively late
development of TRH compared to TRM1 cells (Supplementary Fig. 4d). In summary,
high expression of PSGL1 by the majority of CD4 T cells at day 9 likely reflects their
11 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
recent migration into the inflammatory lung environment; this time point is dominated by
TRM1 cells while TRH cells emerge gradually at later time points.
TRH cell generation requires B cells and T cell intrinsic Bcl6
Given the phenotypic and transcriptional similarities between lung TRH cells and
lymphoid homing TFH cells, we hypothesized that TRH cells may depend on B cells and
intrinsic Bcl6 expression for their differentiation23, 34. To test a requirement for B cells,
we injected mice with anti-CD20 one week prior to influenza infection. Thirty days after
influenza infection, both the proportion and number of TRH cells were decreased in the
lungs of B cell depleted mice, correlating with a strong reduction in resident and
circulating B cell numbers in the lung (Fig. 5a-c, Supplementary Fig. 5a). To assess the
contribution of T cell intrinsic Bcl6 expression we generated radiation bone marrow
chimeras with a mixture of wild type and Bcl6flox/flox x CD4Cre (Bcl6ΔCD4) bone marrow.
Thirty days after influenza infection, Bcl6 deficient CD4 T cells were incapable of
differentiating into lung TRH cells, demonstrating their dependence on intrinsic Bcl6
expression (Fig. 5d). In the absence of Bcl6, the proportions of TRM1 cells in the lung
and non-TFH cells in the draining LN were also significantly reduced compared to
wildtype cells, suggesting that Bcl6 also plays a role in the expansion of these cells (Fig.
4e, Supplementary Fig. 5b, c)35. Taken together, these data indicate that similar to
lymphoid TFH cells, TRH cells in the lung require B cells and T cell intrinsic Bcl6
expression for their generation.
CD4 TRH cells localize in iBALT
12 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Influenza infection leads to the development of ectopic clusters of B cells and T cells in
the lung, known as inducible bronchus-associated lymphoid tissue (iBALT), that can
persist to at least 100 days after infection36, 37. The formation of iBALT supports primary
T and B cell proliferation and local antibody production and is thought to provide a
specialized niche for the reactivation of lung immunity during challenge infection. Given
their dependence on B cells, we hypothesized that TRH cells may localize together with
B cells in iBALT. To examine this, we assessed the spatial distribution of TRM1 and
TRH cells in the lungs of influenza infected T-bet and Bcl6 reporter mice, respectively.
Using flow cytometry, we first confirmed that infection-induced polyclonal CD4 T cells in
the lung largely phenocopy NP-specific TRM1 and TRH cells (Supplementary Fig. 6a).
Gating on the minimal markers CD4 and T-bet, or CD4 and Bcl6, revealed that over
85% of these cells fall within the TRM1 or TRH compartments, respectively
(Supplementary Fig. 6b, c). We next stained lung sections from T-bet or Bcl6 reporter
mice with CD4 and B220 to identify iBALT regions by spinning disk confocal microscopy
(Fig. 6a). T cell localization was quantified using two separate pipelines: one manually
guided and one leveraging machine learning (Fig. 6b-d). The number of Tbethi or Bcl6hi
CD4 cell objects inside and outside of the B cell clusters was calculated after
normalization by iBALT and tissue slice volumes. Bcl6 hi CD4 cells were enriched inside
of iBALT areas compared to Tbet hi CD4 cells, while Tbet hi CD4 cells were more
prevalent outside of iBALT (Fig 6c, d). The preferential co-localization of TRH cells and
B cells in the lung indicates that iBALT may provide a homeostatic niche for TRH cell
survival.
Maintenance of TRH cells requires antigen presentation
13 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Influenza infection has been shown to induce a depot of viral antigen in the
lung which can be transported to and presented in the draining LN to support CD4
memory T cell accumulation38. High expression of Nur77 by TRH cells (Fig. 3c, d)
suggested that they might be responding to ongoing antigen presentation in the lung.
To determine if TRH cells are actively proliferating at late time points, we administered
bromo-2-deoxyuridine (BrdU) in the drinking water of influenza infected mice at late time
points after influenza infection. Given the low number of NP-specific T cells that can be
recovered from the lung after intranuclear staining for BrdU, we analyzed both NP+ as
well as CD4+44+ T cells that were negative for intravascular staining with CD45.
Following BrdU treatment, CD4 TRM cells took up significantly less BrdU compared to
“circulating” lung CD4 T cells that stained positively for i.v. anti-CD45 (Fig. 7a, b).
Comparison of BrdU uptake by TRH and TRM1 cells, however, revealed similar levels
of proliferation between these two subsets, despite higher expression of Nur77 by lung
TRH cells (Fig. 7c). To further determine whether antigen presentation is required to
sustain resident CD4 T cells, we generated MHC-IIflox/flox x UBCCre-ERT2 mice (MHC-
IIΔUBC-ERT2) to induce the widespread deletion of MHC-II at late time points after
infection. MHC-IIΔUBC-ERT2 and control mice were infected with influenza followed by
tamoxifen administration beginning at least 40 days after infection (Fig. 7d). Seven
days after the last tamoxifen treatment, both resident and circulating B cells isolated
from the lungs of MHC-IIΔUBC-ERT2 mice expressed approximately 10-fold lower levels of
MHC-II compared to control antigen presenting cells (Supplementary Fig. 7a, b).
Although the number of B cells in the lung was not impacted by MHC-II deletion, the
proportion and number of TRH cells was significantly decreased (Fig. 7e-g,
14 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Supplementary Fig. 7c, d). These data indicate that while both TRH and TRM1 cells
are actively proliferating to a similar extent at late time points after infection, TRH are
uniquely dependent on sustained antigen presentation for their maintenance.
TRH cells contribute to protection during re-challenge
TFH memory cells express high levels of Tcf7, a transcription factor associated
with self-renewal and stem-like properties, and were recently shown to generate
multiple cell fates following recall infection26. The similarities between TRH cells and
TFH cells prompted us to investigate lung resident T cells for potential developmental
trajectories. We chose TRH as a starting cluster for the analysis because it most closely
resembles stem-like, Tcf7hiId3 hi TFH cells. Using PHATE for dimension reduction, we
observed a bifurcating trajectory starting with the TRH cluster, branching at the Hif1a hi
cluster 1 and terminating in the "circulating" cluster 4 and the TRM1 cluster 2 (Fig. 8a,
Supplementary Fig. 8a). Imputed expression of genes emblematic for each terminal cluster
could then be followed along developmental pseudotime (Fig. 8b). While we cannot
exclude the possibility of circulating cells entering the lung as a starting point for a
trajectory, enrichment of S1pr1 and Ccr7 in cluster 4 could also mark a "draining" population
of cells that are poised to exit to the LN. These data raise the possibility that TRH, like TFH,
may be multipotent following recall infection. To test this experimentally, we adoptively
transferred TRH and TRM1 cells from influenza infected mice into congenic recipients
followed by recall infection with PR8 influenza. At day 9 after challenge infection, both
transferred TRM1 and TRH cells maintained high expression of PSGL1, similar to
endogenous NP-specific cells at this time point (Fig. 8c, Supplementary Fig. 8a, b).
Although the majority of transferred TRM1 cells remained negative for FR4 expression,
15 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
transferred TRH cells gave rise to both FR4hi and FR4lo secondary effectors (Fig. 8c, d).
These data demonstrate that similar to TFH memory cells, TRH cells retain the capacity
to generate both FR4hi and FR4lo effectors while TRM1 cells appear to be more
terminally differentiated at this time point.
TFH memory cells are reported to provide accelerated help to memory B cells during a
challenge infection, in a T cell-intrinsic, Bcl6 dependent manner39, 40. As influenza
infection can induce the differentiation of resident memory B cells in the lung, we next
assessed the contribution of TRH cells to secondary humoral immunity in the tissue41.
To do this we generated Bcl6flox/flox x CD4Cre-ERT2 (Bcl6ΔCD4-ERT2) mice that allow for
tamoxifen inducible deletion of Bcl6 in CD4 T cells. Bcl6ΔCD4-ERT2 and Bcl6flox/flox control
mice were infected with influenza and treated with tamoxifen starting at 21 days. Prior
to secondary infection, late deletion of Bcl6 in CD4 T cells led to a mild decrease in the
number of NP-specific TRH cells while TRM1 cells numbers were not affected (Fig. 8e,
f). To understand if Bcl6 deletion had an impact on CD4 resident T cell maintenance
within iBALT, we examined the lungs of Bcl6ΔCD4-ERT2 and control mice by histology.
While CD4 cells were abundant in dense areas of strong B220 signal in control mice,
Bcl6 deletion led to significantly reduced colocalization of these cell types, despite the
maintenance of large clusters of B220+ cells (Fig. 8g, Supplementary Fig. 8b). In
contrast, normalized CD4 cell counts outside of iBALT areas were higher with Bcl6
deletion (Fig. 8h). As iBALT can act as a site for the local priming of pulmonary immune
responses we next examined whether Bcl6 deletion in CD4 T cells would lead to less
efficient secondary B cell responses. To test this, we further treated Bcl6ΔCD4-ERT2 and
16 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
control mice with FTY720 and a CD8 depleting antibody followed by challenge infection
with heterotypic X31 influenza (X31) (Fig. 8i). This approach allowed us to examine the
impact of Bcl6 deletion in lung CD4 T cells independently from circulating lymphoid T
and B cells, as well as memory CD8 T cells, which might otherwise lead to early viral
clearance. Importantly, antibodies generated after PR8 influenza infection do not
neutralize X31 infection, while CD4 T cells respond to the conserved NP epitope,
thereby enabling specific investigation of CD4 T cell mediated protection. Four days
after recall with X31 influenza, Bcl6 deletion in the CD4 compartment led to a four-fold
decrease in NP-specific antibody-secreting cells in the lung (Fig. 8j, Supplementary Fig.
8c). These data indicate that TRH cells contribute to heterologous protection by
promoting local antibody production during re-challenge.
Discussion
In this study we used scRNAseq to characterize the dynamics, heterogeneity and
transcriptional regulation of lung resident CD4 T cells after influenza infection. In
contrast to previous studies which mainly identified Th1 memory cells with
transcriptional resemblance to CD8 TRM cells, our data reveal the presence of long-
lived TRH cells with phenotypic and functional similarities to lymphoid TFH cells. In
agreement with these earlier studies, many of which tracked monoclonal T cell
responses with specificity to a single antigen, we did not observe TRH differentiation by
OT-II T cells responding to PR8-OVA, indicating that not all TCR lines recapitulate the
heterogeneity of polyclonal T cell populations. Importantly, our study identifies
phenotypically similar cell clusters in both LN and lung alongside a consistent signature
17 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
of residency in the lung. Given the predominance of type 1 responses to viral infection
in barrier tissues, the averaging effect of bulk studies, and the paucity of antigen-
specific polyclonal models to date, a general residency signature was previously
obscured.
Notably, the TRH population we identified in the lung was itself heterogeneous, with
FR4hiPSGL1low cells clustering into two groups transcriptionally. Regulation of these
two subsets by Bcl6 and HIF-1α, respectively, is consistent with a previous study
showing that Bcl6 can repress gene programs activated by HIF-1α in effector CD4 T
cells42. While HIF-1α has been alternately reported to repress or promote TFH effector
cell differentiation43, 44, our data reveal a gradual upregulation of HIF-1α in CD4 TRM
cells, coincident with TRH cell differentiation, localization in iBALT and resolution of the
inflammatory/hypoxic phase of infection. HIF-1α expression might therefore reflect
differences in TRH cell localization and the dynamics of access to environmental
signals. In addition, expression of Areg, encoding the cytokine amphiregulin that
promotes epithelial cell proliferation, may also point to a role in tissue homeostasis45.
Amphiregulin produced by regulatory T cells was previously shown to prevent excessive
inflammation in the lung at very early time points after primary infection46. It will be
interesting to determine if amphiregulin produced by resident CD4 T cells plays a similar
role during recall infection, as a means to balance anti-pathogen immunity with tissue
integrity and niche preservation.
18 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Our data highlight the importance of B cell interactions for the generation of the TRH
compartment. B cells have been previously implicated in the differentiation of Th1 TRM
cells following intranasal LCMV infection, where they initially restrained Th1 cell
residency but later promoted their long-term persistence47. Although these results are
consistent with our observation that TRM1 cells differentiate earlier compared to TRH
cells, we did not observe any impact on TRM1 cell numbers in B cell depleted mice.
We additionally determined that TRH cells require intrinsic Bcl6 expression. The
absence of Bcl6 at the start of infection led to complete ablation of the TRH
compartment as well as a significant decrease in TRM1 cells. These data are
consistent with a previous study looking at CD4 T cell differentiation in the lung during
tuberculosis infection48. Here the authors reported that both Bcl6 and ICOS signaling
are required for the development of lung resident T cells with memory like properties
that mediate superior protection compared to more terminally differentiated lung Th1
cells. Taken together with our observations, these findings suggest that TRH cells
maintain the plasticity to differentiate into Th1 effectors following challenge infection, an
idea that is supported by our trajectory analysis and adoptive transfer experiments. The
ability of TRH cells to differentiate into multipotent effectors is similar to stem-like TFH
memory cells which were recently reported to generate diverse effectors following
secondary infection26.
Influenza infection results in the development of iBALT, a tertiary lymphoid structure
(TLS) in the lung which can act as an immunological hub, promoting rapid and localized
immune responses following secondary infection49. Compared to TRM1 cells, we
19 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
detected tight localization of TRH cells and B cells within iBALT cores. The importance
of T-B interactions in TLS has been highlighted in many immune contexts, including
tuberculosis infection, where the presence of iBALT correlates with bacterial control, as
well as the tumor microenvironment, where the presence of TLS predicts improved
patient outcome50, 51, 52, 53, 54. Our data provide additional insight to these studies,
demonstrating TRH cell dependency on both antigen presentation and Bcl6 expression
for their maintenance within iBALT. The decreased number of iBALT-localized CD4 T
cells observed after late Bcl6 deletion may be due to lower expression of ICOS which
normally promotes T cell interactions with ICOSL expressing B cells55. Further
experiments will help clarify if Bcl6 is also required for the long-term maintenance or
retention of CD4 memory cells in other TLS contexts such as the tumor
microenvironment or autoimmune diseases such as rheumatoid arthritis, where TFH
cells are implicated in health and pathology, respectively.
Finally, our data demonstrate an important role for resident CD4 T cells in orchestrating
local humoral responses during recall infection. This is reciprocal to the relationship
between memory TFH cells and memory B cells present in lymphoid organs where
antigen specific memory B cells induce Bcl6 expression in cognate TFH cells, leading to
more efficient T cell help and acceleration of humoral immune responses39, 40. Our
experiments suggest that TRH cells are poised to provide help due to their tight
localization with BRM cells. Thus, the targeted induction of lung TRH cells may serve
as the basis for rational design of a universal influenza vaccine. Taken together with
the observation that TRH cells maintain the capacity to generate Th1 effectors, these
20 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
findings have implications for reinvigorating or dampening T cell responses in tissues or
tumors where TLS are present.
21 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Methods
Mice
Male and female mice C57BL/6 J (CD45.2), B6.SJL-PtprcaPepcb/BoyJ (CD45.1),
C57BL/6-Tg(Nr4a1-EGFP/cre)820Khog (Nur77 GFP), Bcl6tm1.1Cdon (Bcl6 RFP),
Tg(Tbx21-ZsGreen)E3ZJfz (T-bet ZsGreen), B6.129X1-H2-Ab1tm1Koni/J B6.Cg-
Ndor1Tg(UBC-Cre/ERT2)1Ejb/1J (MHCIIΔUBC-ERT2), (UBC-ERT2 mice kindly provided by Prof.
Tobias Derfuss, University of Basel), B6.Bcl6tm1.1MttoTg(Cd4-cre)1Cwi (Bcl6ΔCD4),
B6.Bcl6 tm1.1Mtto Cd4tm1(cre/ERT2)Thbu (Bcl6ΔCD4-ERT2, Bcl6 fl/fl mice kindly provided by Dr.
Toshitada Takemori, RIKEN The Institute of Physical and Chemical Research), B6.Cg-
Tg(TcraTcrb)425Cbn/J (OT2) were used. Mice in each experiment were same sex
littermates, 6 to 12 weeks old at the start of experiment, maintained and bred in the
specific pathogen free animal facility at the University of Basel. All animal experiments
were performed in accordance with local and Swiss federal guidelines. Non-blinded
experimental groups were formed with random assignment of mice and no specific
methods applied to determine sample size.
Infections
Influenza A/PR8/34 OVAII (hereafter, PR8) and X31 starter material were kindly
provided by Dr. Paul G. Thomas, St. Jude Children’s Research Hospital and produced
at VIRAPUR. PR8 was intranasally administered at 500-2000 TCID50 for all primary
5 infections. Influenza X31 was used at 10 TCID50 for secondary infections. Mice were
anesthetized with vaporized isoflurane and infected intranasally with virus diluted in 30-
40μL volume of PBS.
22 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Mixed bone marrow chimera
Bone marrow cells from wild type (CD45.1) and Bcl6ΔCD4 (CD45.2) mice were collected,
mixed in 60:40 (CD45.1:CD45.2) ratio to compensate for reconstitution defects of Ly5.1
line and adoptively transferred into lethally irradiated (2x 500cGy) CD45.1 hosts56.
Reconstitution in blood was checked 6 weeks after irradiation.
Adoptive transfer
Inguinal, brachial and axillary lymph nodes from naive OT2 mice were harvested and
mashed through a 100μm Nylon cell strainer (Corning). Negative selection for CD8α,
CD19, B220, I-Ab, NK-1.1 using LS columns (Miltenyi Biotec) was performed to enrich
CD4 T cells. Cells were enumerated as previously described and 30,000 OT2 cells
were transferred into congenic hosts one day before infection with PR8-OVA257. For
the cell transfer experiment in Fig. 8c, non-circulating CD4 T cells from each subset
were sorted and transferred into naive congenic hosts that were infected with PR8 the
following day.
In vivo treatments
Tamoxifen (Sigma) was dissolved in corn oil (25mg/ml) and stored at 4°C during the
duration of treatment. Mice were administered 3.75mg in 150μl volume for 5-7 days by
oral gavage. Mice were administered BrdU (10280879001 Roche) in the drinking water
(1mg/ml with 1% glucose) for 13 days and sacrificed on the fourteenth day. 12.5μg
homemade ARTC2.2-blocking nanobody s+16 (NICD protector) was intravenously
injected into mice at least 15 min before sacrifice. Mice were injected intravenously with
23 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
3μg of fluorochrome conjugated anti-CD45 AF700 (clone A20, #110724, Biolegend) 3
min prior to sacrifice to label circulating cells and discriminate them from the non-
circulating compartment. B cell depletion was performed by weekly i.p. administrations
of 250μg of murine B cell depletion antibody (anti-mouse CD20, 18B12 obtained from
Biogen MA Inc.) starting from one week before infection until sacrifice. CD8 depletion
was performed by administering InVivoMAb anti-mouse CD8α (clone 2.43, BioXCell),
400μg i.p. and 100μg intranasal for 2 consecutive days before secondary infection.
The S1P1R agonist, FTY720 (AdipoGen Life Sciences) was dissolved in DMSO at a
concentration of 20mg/ml. 1mg/kg was administered i.p. for three consecutive days
before harvest for memory time point experiments (Supp. Fig 1), every 2 days starting
day 9 for migration experiment (Fig. 4) and every day starting 3 days before secondary
infection for recall experiments.
Tissue preparation
Lungs were harvested and diced into GentleMACS C Tubes (Miltenyi Biotec) and
washed down with 3ml media (RPMI, 10mM aminoguanidine hydrochloride (Sigma),
10mM HEPES, Penicillin-Streptomycin-Glutamine (100X, Gibco), 2-Mercaptoethanol
(50μM, Gibco)) devoid of FCS and EDTA pre-warmed in 37°C water bath. Digestion
mix containing 33.3μg/ml Liberase (Roche #05401020001) and 68μg/ml DNAse I
(Applichem) were added. Lungs were dissociated on gentleMACS Dissociator (Miltenyi
Biotec) and placed in a shaking incubator at 37°C for 30 min. Lungs were dissociated
again and mashed through 70μm MACS SmartStrainers (Miltenyi Biotec). Red blood
cells were lysed with 139.5mM NH4Cl and 17mM Tris HCl (Erylysis buffer). For further
24 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
processing, media containing 2% FCS and 1mM EDTA were used. Mouse mediastinal
lymph nodes were harvested and mashed through 100μm Corning Nylon cell strainer.
Tetramer, antibody staining, flow cytometry
I-Ab NP311-325 (QVYSLIRPNENPAHK) Allophycocyanin was obtained from the NIH
tetramer core facility. Single cell suspensions were tetramer stained, enriched and
counted as previously described57. Dasatinib (50nM) was added to the tetramer mix to
reduce pMHC tetramer binding affinity threshold and activation induced cell death58. To
analyze B cells, the flowthrough collected post tetramer enrichment was used to
enumerate and stain for B cell markers. For all fluorochrome-conjugated antibody
dilutions, FACS buffer (PBS, 2% FCS, 0.1% Sodium Azide) containing Fc block
(InVivoMAb anti-mouse CD16/CD32, Clone 2.4G2, BioXCell) was used. Live-dead
stain (Zombie Red Fixable Viability kit, Biolegend) was added to the antibody mix and
stained at 4°C for 30 min. FoxP3 fixation kit (eBioscience) was used for intracellular
staining at room temperature (RT) for 1 hour. For BrdU staining, BD Pharmingen BrdU
Flow Kit was used. Flow cytometric analysis was performed on BD LSR Fortessa.
Data were analyzed using FlowJo X software (TreeStar).
Gating strategy used in Flow cytometric analysis
To gate on resident CD4 T cells, lung cells were gated on Zombie red- (live) , dump-
(CD11b, CD11c, B220, F4/80) lymphocytes and then on CD4+, i.v- T cells (non-
circulating). For antigen-specific T cells, CD44+NP+ cells were gated on. Non-antigen
25 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
specific are considered as CD44+NP-. The same gating strategy was used for CD4 T
cells from mLN.
For B cell staining in the lung, cells were gated on Zombie red- (live) , dump- (CD3ε, Gr-
1, F4/80, CD11c) lymphocytes and then on B220+. Further gating on iv+ or iv- was done
to discriminate between circulating and non-circulating B cells.
scRNA sequencing
0.5-3 x 104 total NP-specific CD4 T cells were sorted from mLN and lung of PR8-OVA2
infected mice at indicated time points and were provided for library preparation using
the 10x Chromium platform. Each sample is pooled from 4-12 C57BL/6 J mice. Sorting
was performed using BDFACSAriaTM III and BDSorpAriaTM III. Single-cell capture and
cDNA library preparation were performed with a Single Cell 3’ v2 Reagent Kit (10x
Genomics) according to manufacturer’s instructions. Sequencing was performed on
one flow-cell of an Illumina NexSeq 500 at the Genomics Facility Basel of the ETH
Zurich. Paired-end reads were obtained and their quality was assessed with the
FastQC tool (version 0.11.5). The length of the first read was 26 nt, composed of
individual cells barcodes of 16 nt, and unique molecular identifiers (UMIs) of 10 nt. The
length of the second read, composed of the transcript sequence, was 58 nt. The
samples in the different wells were identified using sample barcodes of 8 nt.
scRNA sequencing analysis
Sequencing data was processed using 10X Genomics’ cellranger software version
2.1.0, modified to report only one alignment (randomly) for multi-mapped reads (keep
26 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
only those mapping up to 10 genomic locations). Raw molecule info from cellranger
was initially filtered leniently, discarding cells with fewer than 100 UMI counts, as well as
the highest 99.99% to account for likely doublets. The resulting UMI matrix was further
filtered to keep only cells with log library size > 2.9, log number of features > 2.6,
percent mitochondrial reads <= 6, and percent ribosomal protein reads >= 20. Genes
with average counts < .007 were removed. Normalization was done using the R
package scran’s deconvolution method. Technical noise within gene expression was
modeled using scran, and biologically relevant highly variable genes were calculated
after separating the technical from biological variance, using FDR < 0.05 and biological
variance > 0.1. PCA was run on the normalized data using the top 500 most variable
genes by biological variance, and the PCA was denoised to account for the modelled
technical variation. Cells were clustered hierarchically using Ward’s method on the
distance matrix of the PCA. Default dendrogram cut height using the R package
dynamicTreeCut resulted in clusterings that also mapped to the highest average
silhouette width. Data subsets, e.g. lung d30 cells only, were subjected to the same
pipeline after subsetting. The sex-linked gene Xist was removed from the signature
residency gene list. Subsequent visualization and analysis were performed using
version 3.1.1 of the Seurat R package; dropout imputation was performed using Seurat
version 2.3.4. Single-cell regulatory network inference and clustering (SCENIC) was
performed using version 0.9.7 and the published workflow on the pySCENIC github
repository (96). Pseudotime analysis was performed with slingshot version 1.5.0 and
PHATE dimension reduction29, 59. Secondary analysis of published data sets was taken
27 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
directly from published supplemental material or downloaded from GEO and analyzed
for differential expression between samples by the limma package60.
Histology
Lungs were inflated with 3% low melting agarose (LMA) (Sigma, #A9414), fixed in 1%
formaldehyde (Thermo scientific, #28908) for 24h at 4°C, washed with PBS and kept at
4°C in PBS + 0.01%NaN3. Lung tissues were embedded in 3% LMA and 100μm-thick
sections were cut at the vibratome (Leica VT 1200S). Sections were placed on
microscope slides, circumscribed with ImmEdge pen (Vector) and blocked in TBST
(100mL Tris-HCl, 150mM NaCl, 0.05% Tween-20) containing 5% Donkey serum
(Jackson ImmunoResearch #017-000-121) for 2 hours at RT. For lung staining from T-
bet ZsGreen mice, sections were incubated with anti-CD4 (1:50 in blocking buffer)
overnight at 4°C, then washed (4 times for 15 min with TBST) and incubated with anti-
rat-AF647 (1:160) for 4 hours at RT, followed by washes and subsequent incubation
with anti-B220-BV421 (1:50) overnight at 4°C. For Bcl6 RFP mice, sections were
processed similary by incubating sections with anti-CD4 (1:50) and anti-RFP (1:100)
overnight at 4°C, then with anti-rat-AF647 (1:160) and anti-rabbit-AF488 (1:200) for 4
hours at RT, and finally with anti-B220-BV421 (1:50) overnight at 4°C. For Bcl6fl/fl and
Bcl6ΔCD4-ERT2 mice, sections were incubated with anti-CD4 (1:50) overnight at 4°C, then
washed, incubated with anti-rat-AF647 (1:160) for 4 hours at RT. Then sections were
washed, incubated for 45 min with 5% rat serum in blocking buffer, washed, incubated
for 45 min with Fab anti-rat (100μg/ml), washed and incubated with anti-B220 (1:50)
overnight at 4°C. Sections were then washed and incubated with anti-rat-AF488 (1:300)
28 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
for 4 hours, RT. After the last wash, sections were mounted using ProLong Glass
antifade mountant (Invitrogen, P36984).
Microscopy
Images (16bts, 1000x1000 pixels images) were acquired using X20 or X40
magnification objectives at the spinning disc confocal microscope (Nikon CSU-W1) with
a photometrics 95B camera.
Image quantification
Confocal images were processed in Bitplane Imaris v9.5. Surfaces were created to
segment different features of interest: dense B cell areas (B220+) (automatic threshold,
surface detail 8 µm, seed point diameter >= 40, artefacts removed and final areas
unified manually), individual CD4+ T cells (automatic threshold, surface detail 0.8 µm,
seed point diameter 4 µm), and the total possible volume in which a CD4+ T cell could
be found (threshold ~150, surface detail 8 µm). Where necessary, surfaces for obvious
non-cell artefacts were created and used to mask the channels of interest before
creating target surfaces. The statistics including shortest distance between surfaces
was exported to CSV files. To confirm these results, images were also processed using
custom computer vision and machine learning algorithms mainly written in python. Two
separate image-based classifiers were trained at voxel and object level detection and
identification for 3D clusters (B220) and 3D single cells (CD4). Features such as
intensity, morphology, etc., were then measured per object at cluster and cellular level
and colocalized using a parent (cluster) and child (cell) relationship within clusters of B
29 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
cells. The distances between surfaces were calculated using centroids of cells and
clusters. Data from both segmentation pipelines were loaded into R and processed
using custom code to quantify and test distributions of CD4 T cells with respect to B cell
areas. CD4 objects less than approximately 2.5 µm in radius were discarded as
probable artefacts of surface creation. B cell clusters smaller than 20000 µm3 were also
discarded. Double positive CD4+Tbet+ or CD4+Bcl6+ cells were positively gated as if
for FACS based on a log distribution of mean intensity scores per CD4 object - by eye
with the manual pipeline and at a threshold of the top 10% of per image data for the
machine learning pipeline. Normalization: T cell counts within B cell areas were divided
by that B cell area's volume, and counts outside B cell areas were divided by the total
available T cell volume minus all B cell areas.
ELISpot
Multiscreen IP HTS filter plates (#MAIPS4W10, EMD Millipore) were coated with
2μg/mL NP (Sino biological #11675-V08B) and incubated overnight at 4°C. Plates were
blocked with cell culture medium. Cells were plated and incubated at 37°C for 5 hours.
Plates were washed in PBS+0.01% Tween-20 before goat anti-mouse IgG HRP (Fcγ
fragment, Jackson Immuno #115-035-008) was added and incubated overnight at 4°C.
Plates were washed and developed using AEC Substrate set (BDBiosciences
#551951). Spots were recorded with AID Elispot Reader and enumerated manually.
Statistical analyses
30 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Statistical analyses were performed with Prism Graphpad Software (Version 8.0).
Unpaired or paired t-tests (two-tailed) or Mann-Whitney-Wilcoxon tests were used
according to the type of experiments. P < 0.05 was considered significant.
31 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
References
1. Palache, A. et al. Survey of distribution of seasonal influenza vaccine doses in 201 countries (2004-2015): The 2003 World Health Assembly resolution on seasonal influenza vaccination coverage and the 2009 influenza pandemic have had very little impact on improving influenza control and pandemic preparedness. Vaccine 35, 4681-4686 (2017).
2. Sambhara, S. et al. Heterosubtypic immunity against human influenza A viruses, including recently emerged avian H5 and H9 viruses, induced by FLU-ISCOM vaccine in mice requires both cytotoxic T-lymphocyte and macrophage function. Cell Immunol 211, 143-153 (2001).
3. Tumpey, T.M., Renshaw, M., Clements, J.D. & Katz, J.M. Mucosal delivery of inactivated influenza vaccine induces B-cell-dependent heterosubtypic cross- protection against lethal influenza A H5N1 virus infection. J Virol 75, 5141-5150 (2001).
4. Rangel-Moreno, J. et al. B cells promote resistance to heterosubtypic strains of influenza via multiple mechanisms. J Immunol 180, 454-463 (2008).
5. Zens, K.D. et al. Reduced generation of lung tissue-resident memory T cells during infancy. J Exp Med 214, 2915-2932 (2017).
6. Zens, K.D. & Farber, D.L. Memory CD4 T cells in influenza. Curr Top Microbiol Immunol 386, 399-421 (2015).
7. Devarajan, P. et al. New Insights into the Generation of CD4 Memory May Shape Future Vaccine Strategies for Influenza. Front Immunol 7, 136 (2016).
8. Valkenburg, S.A. et al. The Hurdles From Bench to Bedside in the Realization and Implementation of a Universal Influenza Vaccine. Front Immunol 9, 1479 (2018).
9. Sallusto, F., Geginat, J. & Lanzavecchia, A. Central memory and effector memory T cell subsets: function, generation, and maintenance. Annu Rev Immunol 22, 745-763 (2004).
10. Schreiner D, K.C. CD4+ memory T cells at home in the tissue: mechanisms for health and disease. Frontiers in Immunology Available at: doi: 10.3389/fimmu.2018.02394 (2018).
11. Miragaia, R.J. et al. Single-Cell Transcriptomics of Regulatory T Cells Reveals Trajectories of Tissue Adaptation. Immunity 50, 493-504 e497 (2019).
32 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
12. Beura, L.K. et al. CD4(+) resident memory T cells dominate immunosurveillance and orchestrate local recall responses. J Exp Med 216, 1214-1229 (2019).
13. Amezcua Vesely, M.C. et al. Effector TH17 Cells Give Rise to Long-Lived TRM Cells that Are Essential for an Immediate Response against Bacterial Infection. Cell 178, 1176-1188 e1115 (2019).
14. Kumar, B.V. et al. Human Tissue-Resident Memory T Cells Are Defined by Core Transcriptional and Functional Signatures in Lymphoid and Mucosal Sites. Cell Rep 20, 2921-2934 (2017).
15. Oja, A.E. et al. Functional Heterogeneity of CD4(+) Tumor-Infiltrating Lymphocytes With a Resident Memory Phenotype in NSCLC. Front Immunol 9, 2654 (2018).
16. Oja, A.E. et al. Trigger-happy resident memory CD4(+) T cells inhabit the human lungs. Mucosal Immunol 11, 654-667 (2018).
17. Teijaro, J.R. et al. Cutting edge: Tissue-retentive lung memory CD4 T cells mediate optimal protection to respiratory virus infection. J Immunol 187, 5510- 5514 (2011).
18. Zens, K.D., Chen, J.K. & Farber, D.L. Vaccine-generated lung tissue-resident memory T cells provide heterosubtypic protection to influenza infection. JCI Insight 1 (2016).
19. Deng, N., Weaver, J.M. & Mosmann, T.R. Cytokine diversity in the Th1- dominated human anti-influenza response caused by variable cytokine expression by Th1 cells, and a minor population of uncommitted IL- 2+IFNgamma- Thpp cells. PLoS One 9, e95986 (2014).
20. McKinstry, K.K. et al. Memory CD4 T cell-derived IL-2 synergizes with viral infection to exacerbate lung inflammation. PLoS Pathog 15, e1007989 (2019).
21. Strutt, T.M., McKinstry, K.K., Kuang, Y., Bradley, L.M. & Swain, S.L. Memory CD4+ T-cell-mediated protection depends on secondary effectors that are distinct from and superior to primary effectors. Proc Natl Acad Sci U S A 109, E2551- 2560 (2012).
22. Asrir, A., Aloulou, M., Gador, M., Perals, C. & Fazilleau, N. Interconnected subsets of memory follicular helper T cells have different effector functions. Nat Commun 8, 847 (2017).
23. Pepper, M., Pagan, A.J., Igyarto, B.Z., Taylor, J.J. & Jenkins, M.K. Opposing signals from the Bcl6 transcription factor and the interleukin-2 receptor generate T helper 1 central and effector memory cells. Immunity 35, 583-595 (2011).
33 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
24. Skon, C.N. et al. Transcriptional downregulation of S1pr1 is required for the establishment of resident memory CD8+ T cells. Nat Immunol 14, 1285-1293 (2013).
25. Lee, J.Y. et al. The transcription factor KLF2 restrains CD4(+) T follicular helper cell differentiation. Immunity 42, 252-264 (2015).
26. Künzli M, S.D., Pereboom TC, Swarnalekha N, Litzler LC, Lötscher J, Ertuna YI, Roux J, Geier F, Jakob R, Maier T, Hess C, Taylor JJ, King CG. Long-lived T follicular helper cells retain plasticity and help sustain humoral immunity. Science Immunology in press (2020).
27. Jennings E, E.T., Thawait N, Kanabar S, Yam-Puc JC, Ono M, Toellner KM, Wraith DC, Anderson G, Bending D. Differential Nr4a1 and Nr4a3 expression discriminates tonic from activated TCR signalling events in vivo. bioRxiv (2020).
28. Vilmovsky LH, W.J., Henriksson J, Miao Z, Natan E, Kar G, Clare S, Barlow JL, Charidemou E, Mamanova L, Chen X, Proserpio V, Pramanik J, Woodhouse S, Protasio AV, Efremova M, Berriman M, Dougan G, Fisher J, Marioni J, McKenzie ANJ, Teichmann SA. Rora regulates activated T helper cells during inflammation. bioRxiv (2019).
29. Aibar, S. et al. SCENIC: single-cell regulatory network inference and clustering. Nat Methods 14, 1083-1086 (2017).
30. Tanaka, S. et al. CCAAT/enhancer-binding protein alpha negatively regulates IFN-gamma expression in T cells. J Immunol 193, 6152-6160 (2014).
31. Shimatani, K., Nakashima, Y., Hattori, M., Hamazaki, Y. & Minato, N. PD-1+ memory phenotype CD4+ T cells expressing C/EBPalpha underlie T cell immunodepression in senescence and leukemia. Proc Natl Acad Sci U S A 106, 15807-15812 (2009).
32. Choi, Y.S. et al. LEF-1 and TCF-1 orchestrate T(FH) differentiation by regulating differentiation circuits upstream of the transcriptional repressor Bcl6. Nat Immunol 16, 980-990 (2015).
33. Stauss, D. et al. The transcriptional coactivator Bob1 promotes the development of follicular T helper cells via Bcl6. Embo J 35, 881-898 (2016).
34. Johnston, R.J. et al. Bcl6 and Blimp-1 are reciprocal and antagonistic regulators of T follicular helper cell differentiation. Science 325, 1006-1010 (2009).
35. Keck, S. et al. Antigen affinity and antigen dose exert distinct influences on CD4 T-cell differentiation. Proc Natl Acad Sci U S A 111, 14852-14857 (2014).
34 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
36. Moyron-Quiroz, J.E. et al. Persistence and responsiveness of immunologic memory in the absence of secondary lymphoid organs. Immunity 25, 643-654 (2006).
37. Tan, H.X. et al. Inducible Bronchus-Associated Lymphoid Tissues (iBALT) Serve as Sites of B Cell Selection and Maturation Following Influenza Infection in Mice. Front Immunol 10, 611 (2019).
38. Kim, T.S., Hufford, M.M., Sun, J., Fu, Y.X. & Braciale, T.J. Antigen persistence and the control of local T cell memory by migrant respiratory dendritic cells after acute virus infection. J Exp Med 207, 1161-1172 (2010).
39. Ise, W. et al. Memory B cells contribute to rapid Bcl6 expression by memory follicular helper T cells. Proc Natl Acad Sci U S A 111, 11792-11797 (2014).
40. MacLeod, M.K. et al. Memory CD4 T cells that express CXCR5 provide accelerated help to B cells. J Immunol 186, 2889-2896 (2011).
41. Allie, S.R. et al. The establishment of resident memory B cells in the lung requires local antigen encounter. Nat Immunol 20, 97-108 (2019).
42. Oestreich, K.J. et al. Bcl-6 directly represses the gene program of the glycolysis pathway. Nat Immunol 15, 957-964 (2014).
43. Cho, S.H. et al. Hypoxia-inducible factors in CD4(+) T cells promote metabolism, switch cytokine secretion, and T cell help in humoral immunity. Proc Natl Acad Sci U S A 116, 8975-8984 (2019).
44. Zhu, Y. et al. The E3 ligase VHL promotes follicular helper T cell differentiation via glycolytic-epigenetic control. J Exp Med 216, 1664-1681 (2019).
45. Zaiss, D.M.W., Gause, W.C., Osborne, L.C. & Artis, D. Emerging functions of amphiregulin in orchestrating immunity, inflammation, and tissue repair. Immunity 42, 216-226 (2015).
46. Arpaia, N. et al. A Distinct Function of Regulatory T Cells in Tissue Protection. Cell 162, 1078-1089 (2015).
47. Hondowicz, B.D., Kim, K.S., Ruterbusch, M.J., Keitany, G.J. & Pepper, M. IL-2 is required for the generation of viral-specific CD4(+) Th1 tissue-resident memory cells and B cells are essential for maintenance in the lung. Eur J Immunol 48, 80- 86 (2018).
35 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
48. Moguche, A.O. et al. ICOS and Bcl6-dependent pathways maintain a CD4 T cell population with memory-like properties during tuberculosis. J Exp Med 212, 715- 728 (2015).
49. Silva-Sanchez, A. & Randall, T.D. Role of iBALT in Respiratory Immunity. Curr Top Microbiol Immunol (2020).
50. Khader, S.A. et al. In a murine tuberculosis model, the absence of homeostatic chemokines delays granuloma formation and protective immunity. J Immunol 183, 8004-8014 (2009).
51. Kahnert, A. et al. Mycobacterium tuberculosis triggers formation of lymphoid structure in murine lungs. J Infect Dis 195, 46-54 (2007).
52. Cabrita, R. et al. Tertiary lymphoid structures improve immunotherapy and survival in melanoma. Nature 577, 561-565 (2020).
53. Petitprez, F. et al. B cells are associated with survival and immunotherapy response in sarcoma. Nature 577, 556-560 (2020).
54. Helmink, B.A. et al. B cells and tertiary lymphoid structures promote immunotherapy response. Nature 577, 549-555 (2020).
55. Weber, J.P. et al. ICOS maintains the T follicular helper cell phenotype by down- regulating Kruppel-like factor 2. J Exp Med 212, 217-233 (2015).
56. Mercier, F.E., Sykes, D.B. & Scadden, D.T. Single Targeted Exon Mutation Creates a True Congenic Mouse for Competitive Hematopoietic Stem Cell Transplantation: The C57BL/6-CD45.1(STEM) Mouse. Stem Cell Reports 6, 985- 992 (2016).
57. Moon, J.J. et al. Tracking epitope-specific T cells. Nat Protoc 4, 565-581 (2009).
58. Lissina, A. et al. Protein kinase inhibitors substantially improve the physical detection of T-cells with peptide-MHC tetramers. J Immunol Methods 340, 11-24 (2009).
59. Moon, K.R. et al. Visualizing structure and transitions in high-dimensional biological data. Nat Biotechnol 37, 1482-1492 (2019).
60. Ritchie, M.E. et al. limma powers differential expression analyses for RNA- sequencing and microarray studies. Nucleic Acids Res 43, e47 (2015).
61. Mackay, L.K. et al. Hobit and Blimp1 instruct a universal transcriptional program of tissue residency in lymphocytes. Science 352, 459-463 (2016).
36 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
62. Mackay, L.K. et al. The developmental pathway for CD103(+)CD8+ tissue- resident memory T cells of skin. Nat Immunol 14, 1294-1301 (2013).
Acknowledgements: We thank D. Pinschewer and his lab for helpful discussions; J.
Roux and F. Geier for expertise and technical advice; R. Tussiwand and her lab for
feedback and support; the flow sorting facility and all the animal caretakers at the DBM
University of Basel. Single cell RNA sequencing was performed at the Genomics
Facility Basel, ETH Zurich. Calculations were performed at sciCORE
(http://scicore.unibas.ch/) scientific computing center at the University of Basel.
Funding: The work was supported by research grants to CGK (SNF PP00P3_157520,
Freiwillige Akademische Gesellschaft Basel, Forschungsfonds University of Basel and
Swiss Life Jubiläumsstiftung).
Competing interests: The authors declare that they have no competing interests.
Data and materials availability: scRNAseq data are deposited with the National
Center for Biotechnology Information Gene Expression Omnibus (accession nos:
PLACEHOLDER).
37 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Figure legends
Fig. 1. Residency signature in CD4 T cells is biased toward Th1 memory cells
a, NP-specific CD4 T cells in lung and mLN of naïve and PR8 memory mice. b, PCA
showing LN and lung samples with putative circulating cluster from lung26. c, Heatmap
showing scaled, centered single cell expression of top 20 genes sorted according to
cluster average log2FC, adjusted P value < 0.05. d, PCA showing spleen and liver
samples, LCMV infection. e, Differentially expressed genes discriminating lung from LN
(flu) or liver from spleen (LCMV). P values of 0 assigned the lowest non-zero P value. f,
Log-normalized average expression of TFH and Th1 memory signatures26. g, Log-
normalized average expression of combined CD8 residency signature from multiple
tissues61, 62.
Fig. 2. Heterogeneity in the lung: pan residency versus Th-subset specific
residency
a, Unsupervised hierarchical clustering using Ward's method of lung cells only: PCA
and tSNE dimension reductions. b, Heatmap showing scaled, centered single cell
expression of top 20 genes sorted according to lung cluster average log2FC, adjusted P
value < 0.05. c, Log-normalized average expression of TFH and Th1 memory
signatures26. d, Unsupervised hierarchical clustering using Ward's method of LN cells
only: PCA and tSNE dimension reductions. e, Heatmap of scaled, centered single cell
expression showing conserved genes for NLT and LT. Genes included are differentially
expressed in the tissue in all three of the non-circulating cluster pairs. Adjusted P value
< 0.05. f, Differentially expressed genes discriminating LN Th1 cells from lung Th1-like
38 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
cells with reference to conserved tissue-specific signature. Adjusted P value < 0.05. g,
Differentially expressed genes discriminating LN TFH cells from lung TFH-like cells with
reference to conserved tissue-specific signature. Adjusted P value < 0.05. h,
Differentially expressed genes discriminating lung TFH-like cluster 3 from lung TFH-like
cluster 1 with reference to conserved tissue-specific signature. Adjusted P value < 0.05.
Fig. 3. Phenotypic characterization of resident CD4 T cell subsets
a, Identification of TRM1 (red) and TRH (blue) subsets in NP-specific resident CD4 T
cells. b, Total numbers of FR4+ and FR4- NP-specific T cells over time. c-f, Histograms
c, e and MFI d, f of indicated phenotypic markers in TRH (blue), TRM1 (red) and naïve
CD4 T cells (gray). Data are represented as the mean ± s.e.m (n=5, representing 2
experiments) in b. Thin lines represent mean in d and f (n=4-5, representing 2
experiments). The significance was determined by unpaired Student’s t-test. P values:
*P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.
Fig. 4. Progressive differentiation of resident CD4 T cell subsets
a, FR4 and PSGL1 expression on NP-specific tissue resident CD4 T cells at indicated
time points. b, CXCR5 and PD1 expression on NP-specific CD4 T cells in lung and
mLN at indicated time points. c, NP-specific total CD4 T cells, TRM1 and TRH cell
numbers in control and FTY720 treated mice. Data in a, b, c represent 2 experiments
with n=4-6 mice. Statistical significance was determined by unpaired Student’s t-test.
Thin lines represent mean. p<0.05 is considered significant. d, PCA showing time point
by tissue. e, Scaled, centered average expression of select genes in lung at day 9. f,
39 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Log-normalized average expression of published TFH and Th1 signatures by time point
and tissue26. g, Scaled, centered average area under curve (AUC) calculated with
SCENIC showing top 5 differentially active transcription factors in lung by time point.
Adjusted P-value < 0.01.
Fig. 5. TRH cell generation requires B cells and T cell intrinsic Bcl6
a, b NP-specific T cells (a) and frequency of TRH (b) from control and B cell depleted
mice. c, Total number of NP-specific TRM1 and TRH from control and B cell depleted
mice. d, NP-specific T cells in control and Bcl6-deficient subsets. e, Bcl6ΔCD4:control
ratio in CD4 T cells after reconstitution from blood (pre-infection), TRM1 and TRH. Thin
lines represent mean in b, c (n=10, pooled from 2 experiments) and e (n=15, pooled
from 3 experiments). Significance was determined by unpaired Student’s t-test. P
values are as follows: *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.
Fig. 6. CD4 TRH cells localize in iBALT
a, Representative X40 immunofluorescence confocal images from Tbet ZsGreen and
Bcl6 RFP mice, 30-60 days post-infection, stained with the indicated markers. Scale
bars: 200um. b, Quantified representations of segmented CD4+ objects, showing
iBALT location, log mean transcription factor intensity per object, and location of Tbethi
(above) or Bcl6hi (below) objects. c, Count of Tbethi and Bcl6hi CD4 T cell objects inside
B220+ iBALT clusters, normalized by iBALT volume, analyzed by Mann-Whitney-
Wilcoxon test. Segmentation performed in Imaris (left) or with machine learning
algorithm (right). Each dot represents one iBALT. d, Count of Tbethi and Bcl6hi CD4 T
40 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
cell objects outside B220+ iBALT clusters, normalized by tissue volume less iBALT
volume, analyzed by Mann-Whitney-Wilcoxon test. Segmentation performed in Imaris.
Each dot represents one tissue slice.
Fig. 7. Maintenance of TRH cells requires antigen presentation
a, b Histograms and frequency of BrdU+ cells in NP-specific a and CD44+ b lung
resident and circulating memory cells. c, Frequency of BrdU+ cells in TRM1 and TRH.
d, Infected MHC-IIfl/fl and MHC-IIΔUBC-ERT2 mice were analyzed one week after inducible
deletion. e-f, Flow cytometry plots e and TRH frequency f in MHC-IIfl/fl and MHC-IIΔUBC-
ERT2 mice. g, Total numbers of NP-specific TRM1 and TRH. Data in a, b and c were
analyzed by paired t-tests (n=5, representing 3 experiments). Data in f and g were
analyzed by unpaired Student’s t-test (n=6, representing 2 experiments). Thin lines
represent mean. P values are as follows: *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.
Fig. 8. TRH cells contribute to protection during re-challenge
a, PHATE dimension reduction used as input to slingshot trajectory inference. PHATE
colored by cluster (left) and pseudotime (right). b, Imputed expression of selected
genes across pseudotime and two trajectories. c, Representative plots of FR4+ cells in
TRM1 and TRH recipients. d, Frequencies from c. e, Total numbers of NP-specific
TRM1. f, Total numbers of NP-specific TRH. g, Count of CD4 T cell objects inside
B220+ iBALT clusters, normalized by iBALT volume, analyzed by Mann-Whitney-
Wilcoxon test. Imaris segmentation. Each dot represents one iBALT. h, Count of
iBALT-external CD4 T cell objects across all images, normalized by total tissue volume
41 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
less iBALT volume. n = 2. i, PR8-infected, tamoxifen-treated Bcl6flox/flox and Bcl6ΔCD4-
ERT2 memory mice were treated with FTY720 and α-CD8 prior to X31 re-challenge and
analyzed 4 days later. j, Number of NP-specific IgG ASC per mouse lung. Data in c
(n=6, pooled from 2 experiments), e (n=11-12, pooled from 3 experiments) and h (n=4-
7, representing 2 experiments) were analyzed by unpaired Student’s t-test. Thin lines
represent mean. P values are as follows: *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.
Supplementary Fig. 1. Residency signature in CD4 T cells is biased toward Th1
memory cells
a, Identification and frequency of NP-specific resident T cells. b-c, Total numbers of
NP-specific T cells (b) and total CD4 T cell frequencies in blood (c) of control and
FTY720-treated mice. d, Tissue specific signatures derived from one infection model
plotted on data set from the other infection model. Heatmaps show cluster-averaged,
scaled centered expression.
Supplementary Fig. 2 Heterogeneity in the lung: pan residency versus Th-subset
specific residency
a, Imputed expression of residency genes. Dropout imputed using
Seurat::AddImputedScore. b, Imputed expression of genes typifying similar clusters in
lung and LN: blues = TFH-like, yellows = TFH-like, oranges = TCM-like, red = Th1-like.
c, Heatmap showing scaled, centered single cell expression of top 20 genes sorted
according to LN cluster average log2FC, adjusted P value < 0.05. d, Log-normalized
average expression of TFH and Th1 memory signatures26. e, Conserved tissue
42 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
residency signature genes on published single cell data set analyzing Treg adaptation
to tissue11. f, Conserved tissue residency signature genes on published data set
investigating CD8 TRM62. g, Imputed expression of conserved tissue residency
signature genes on LCMV data. h, Scaled, centered area under curve (AUC) calculated
with SCENIC showing top 20 differentially active transcription factors by tissue at day
30. Adjusted P-value < 0.01. i, Scaled, centered, cluster average SCENIC AUC
showing top 20 differentially active TFs by tissue and cluster. Adjusted P value < 0.01.
j, Log-normalized expression by cluster. k, Foxp3+ cells in the lung. Data in a (n=6,
representing 3 experiments), b (n=8, pooled from 2 experiments) and c (n=6, pooled
from 2 experiments) were analyzed by unpaired Student’s t-test. Thin lines represent
mean ± s.e.m in a and mean in b, c. Data in i represents 2 experiments, n=5. P values
are as follows: *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.
Supplementary Fig. 3. Phenotypic characterization of resident CD4 T cell subsets
a, Total numbers of NP-specific non-circulating TRM1 and TRH from PBS and FTY720-
treated mice. b, c Representative plot b of TRM1 and TRH among transferred OT2 and
frequency c of TRH in OT2 and endogenous NP-specific CD4 T cells. Data in a (n=4-5,
representing 2 experiments) and c (n=5, representing 2 experiments) were analyzed
using unpaired Student’s t-test. Thin lines represent mean± s.e.m in a and mean in c.
Significance was determined using unpaired Student’s t-test. P values are as follows:
*P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.
43 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Supplementary Fig. 4. Progressive differentiation of resident CD4 T cell subsets
a, tSNE of scRNA-seq data by tissue and time point. b, Imputed expression of Izumo1r
and Selplg by tissue and time point. c, Differentially expressed genes discriminating
lung (+) from LN (-) across time points. d, Imputed expression of selected genes in the
lung across time points.
Supplementary Fig. 5. TRH cell generation requires B cells and T cell intrinsic
Bcl6
a, Total numbers of iv+ and iv- B cells in control and B cell depleted mice. b, NP-specific
TFH and non-TFH cells of mLN from control and Bcl6-deficient subsets. c,
Bcl6ΔCD4:control ratio of CD4 T cells after reconstitution from blood (pre-infection), NP-
specific non-TFH and TFH. Thin lines represent mean in (a) (n=6, representing 2
experiments) and (c) (n=15, pooled from 3 experiments). Significance was determined
by unpaired Student’s t-test. P values are as follows: *P<0.05, **P<0.01, ***P<0.001,
****P<0.0001.
Supplementary Fig. 6. CD4 TRH cells localize in iBALT
a, Representative gating strategy for identification of non-antigen specific resident CD4
T cells (left) and phenotypic marker expression on gated TRM1 and TRH cells (right).
b-c, Flow cytometry plot representation of CD4+T-bet ZsGreen+ cells within TRM1 gate
(b) and CD4+Bcl6 RFP+ cells within TRH gate (c). NP-specific TRM1 and TRH are
shown on the left for comparison (n=4, representing 2 experiments).
44 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Supplementary Fig. 7. Maintenance of TRH cells requires antigen presentation
a-b, Histograms (left) and quantification (right) of MHC-II expression on iv- (a) and iv+ B
cells (b) upon inducible deletion of MHC-II. c-d, Total numbers of iv- (c) and iv+ (d) lung
B cells. Thin lines represent mean (n=5-6, representing 2 experiments). Significance
was determined using unpaired Student’s t-test. P values are as follows: *P<0.05,
**P<0.01, ***P<0.001, ****P<0.0001.
Supplementary Fig. 8. TRH cells contribute to protection during re-challenge
a, Flow cytometry plots representing proportion of resident CD4 T cells in donor vs host
in TRM1 and TRH recipients. b, Spatial distribution of CD4 cell objects (top) quantified
from immunofluorescence confocal images (bottom) from lungs of Bcl6flox/flox and
Bcl6ΔCD4-ERT2 mice, 30 days post-infection. Each dot represents a CD4+ T cell object,
located in a B220+ area (white), or outside B220+ areas (grey). c, Representative
pictures of ELISpot for NP-specific IgG ASC from lungs of Bcl6fl/fl and Bcl6ΔCD4-ERT2
mice. Data in (a) represents 2 experiments, n=3 and (b) represents 2 experiments,
n=4-7.
45 Figure 1
a bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280Lung ; this version postedLymph February node 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. Naive Day 30 Naive Day 30
5 5 10 10 0.0 0.4 0.0 20.5 4 4 10 10 CD44 3 CD44 3 10 10
0 0
3 4 5 3 4 5 0 10 10 10 0 10 10 10 IAbNP IAbNP
b Flu memory c Flu memory
Circulating LN Lung Circulating (n = 224) Crip1 LN (n = 1412) Il7r
PC 2 PC Bcl2 Lung (n = 2556) Rps19 Rps5 Rpl13a Rpl32 Ifngr1 Emb Rpl14 PC 1 Rplp0 Rps7 Rps16 Rpl35 Gramd3 d Rpl10a Rpl18a LCMV memory Rps4x Rpl8 Rps18 Ppp1r14b Ltb Rps29 Chd3 mt−Nd3 Limd2 Il16 Spleen (n = 3488) Ptprcap Rps28 PC 2 PC Liver (n = 3011) Apobec3 Dennd2d Gm10263 Tspan32 Tcf7 Rps27rt Gm42418 PC 1 Asap1 Eif3f Lrrc58 Rpl38 Tnfrsf4 Crem e Vps37b LT NLT Cstb Nkg7 Nkg7 ● Rgs1 ● ● Ccl5 Hspa1a Ccl5 300 Cxcr6 Areg ● ● Bhlhe40 H2afz Sub1 ● Ctla2a Id2 H3f3b ● Ramp3 Fth1 Tnfrsf18 Rgs2 Mt1 ● ● Gm10073 Sh3bgrl3 H3f3b ● ● ● Vps37b Hif1a Rps29 Tnfaip3 ● 200 ● Ifngr1 Mmd Fam107b Hilpda
Il16 ● Vps37b Izumo1r ● Dennd2d
● Rpl21 Sub1 Ctla2a ● ● Limd2 ● -2 -1 0 1 2 ● ● Gm42418 ● Lgals1 ● Tcf7 ● ● Dusp5 ● Tspan32 Hspa1a ● Sub1 Dennd2d ● Rps27rt ● ● ● ● ● ● Asap1 Crip1 Rgs10 Tcf7 ● Kcnn4 Cish ● ● Socs3 100 Tspan32 Lrrc58 Lrrc58 Rps27 ● f ● H3f3b ● Shisa5 ● ● ● Il6ra Ifngr1 ● ● Arhgef1 Nfkbia Gm42418 Cxcr5 ● ● ● ● ● Rps27rt ● Shisa5 ● ● Gramd3 -log10(adjusted p-value) ● Gm10073 ● ● Crip1 ● Rgs10 Cxcr5 ● ● ● S100a10 ● ● Tnfrsf18 Th1 TFH ● Izumo1r ● ● ● Rflnb Tnfrsf18 ● ● Sh3bgrl3 Ctla2a Il6ra ● Kcnn4 Rps27 ● ● ● ● ● Rps29 Vim ● ● Gramd3 Rpl21 ● Rflnb Asap1 ● Nkg7 ● ● Lck Bhlhe40 ● Socs3 Id2 Lgals1 Limd2 ● Rps12−ps3 Dusp5 ● Ccl5 Tnfaip3 ●● ● Nfkbia ● Rps12−ps3 Il16 ● ● ● ● ● Lck Cish ● ● ● 0 Arhgef1 Rgs1 Rgs2 Hspa1a S100a10 Cxcr6 Vim -1 0 1 2 LN Average log2FC NLT vs LT
Flu memory Flu Lung
LN flu Lung flu 0.5 1.0 1.5 2.0 0.8 1.2 1.6 Spleen LCMV Liver LCMV
g CD8 TRM signature Spleen
Liver LCMVmemory LN Spleen 0.5 1.0 1.5 2.0 0.6 0.9 1.2
Lung Liver Gene set average expression
0.2 0.4 0.6 0.8 0.0 0.2 0.4 0.6
Gene set average expression Figure 2 Lung day 30 a bioRxiv preprint doi:Lung https://doi.org/10.1101/2020.02.28.963280 day 30 ; thisb version posted February12 28, 2020. The copyright34 holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. Tnfsf8 Tnfrsf4 Cstb H2afz Eea1 Hif1a 1 (n = 1279) Ramp3 Areg 2 (n = 699) Prkca Ifi27l2a Ramp1 PC 2 PC 3 (n = 504) Eprs
tSNE 2 tSNE Tbc1d4 4 (n = 298) Nr4a3 Mt1 Wnk1 Mmd Nr4a2 Tnfaip8 PC 1 tSNE 1 Tspan3 Ccl5 Nkg7 Plac8 Crip1 Rgs1 c S100a6 Lung day 30 Ctla2a Lgals1 Lgals3 Ifngr1 AW112010 S100a4 Cxcr6 TFH memory Th1 memory Vim Rgs2 Id2 Ms4a4b Hilpda Ctsw 4 4 Ccr2 3 3 Hspa1a Dnajb1 2 2 Rgs10 1 1 Rnaset2b 0.4 0.6 0.8 1.0 1.2 1.4 0.5 1.0 1.5 2.0 Sostdc1 Rnaset2a Ubac2 Id3 Gene set average normalized expression 2310001H17Rik Ppp1r14b Sh2d1a Jun Ptprcap Zfp36 Ltb Actb d Tcf7 LN day 30 Cdk2ap2 Limd2 Pfn1 Rps19 Rplp0 Rps5 Rps7 Il7r Rpl35 1 (n = 531) Rps20 Rpl13a 2 (n = 405) Gm8730 Rpl32 PC 2 PC
tSNE 2 tSNE Rpl10a 3 (n = 388) Rps12 Gm2000 4 (n = 88) Gm10269 Rpl12 Gm9493 Rps8 Rpl14 Rps4x Rps16 PC 1 tSNE 1 -2 -1 0 1 2 e f LN Lung Lung 3 Lung 1 Lung 2 LN 3 LN 2 LN 4 LN 4 (Th1) Crem Tissue-specific Eif1 20 Fth1 Lung 2 (Th1-like) H3f3b Sh3bgrl3 Sub1 NLT Tmsb4x 15 Tnfrsf4 Ubb Vps37b Anp32b Apobec3 10 Chd3 Dennd2d Eif3f Gm42418 5
Il16 -log10(adjusted p-value) Limd2 LT mt-Nd3 Ptpn7 Ptprcap Rps29 −1 0 1 2 Sp100 Average log FC Tmem160 2 Tspan32
-2 -1 0 1 2 g h TFH-like cells LN Lung Lung 3 Lung 1
Tnfrsf4● LN 3 (TFH) Lung 3 50 Tissue-specific 50 Tissue-specific Vps37b Lung 3 (TFH-like) Lung 1 ● ● 40 Hspa1a 40 Ramp3 ●
●
H2afz● 30 Crem 30 ● Ostf1 Rgs10 Pfn1 ● ● Tnfsf8 ● ● Tnfrsf18 20 Ifngr1 Actb 20 ● Hspa1b Tmsb4x ● Got1 ● ● ● Rora ● Ppp1r14b Ptprcap Eif1 Emb D8Ertd738e Sub1 ● ● ● ● ● ● Rnaset2a Ltb ● Gch1 ● ● ● Il21r ● Rnaset2b Hsp90ab1 ●● ●● 10 ● Cdk2ap2 ● Cytip Jun ● ● Fam107b● -log10(adjusted p-value) ● ● 10 ● ● ●● ●● Dnajb1 ● ● ● ● ● ● Limd2 ● ● ● Nr4a2 Ubac2 ● Fth1 ●●● Crip1 ● ● ●● Isy1 Lrrc58 ● ● ●● ● ● ● ● ● Lmo4 ●●●● ● ● ● ● ●● Arpc3 ●●● ● ● ● ●●● ● Zfp36 ● ● ●●● ● ● Sdf4 Coro1a ●● Tubb2b ●● ● ● Eprs Cdk11b 0 0 Hsp90aa1 Sh2d1a Aldoa Itm2b −1 0 1 2 −1 0 1
Average log2FC Figure 3 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint a (which was not certifiedb by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 6 5 10 FR4+ 10 44 105 FR4- 4 10 104 3 10 103
PSGL1 0 51 102 3 4 5 0 10 10 10 101
# of NP+ CD4 T cells T CD4 NP+ of # 0 50 100 150 FR4 Days after infection c 100 100 100 100
80 80 80 80
60 60 60 60
40 40 40 40 20 20 20 20
Maximum(%) 0 0 0 0 3 4 5 4 5 3 4 3 4 5 0 10 10 10 0 10 10 0 10 10 0 10 10 10 PD1 CXCR5 Nur77 Bcl6-RFP
100 100 100 100
80 80 80 80 60 60 60 60
40 40 40 40
20 20 20 20
Maximum(%) 0 0 0 0 4 5 3 4 5 4 5 2 3 4 5 0 10 10 0 10 10 10 0 10 10 10 10 10 10 P2X7R CD73 CXCR4 ICOS d PD1 CXCR5 Nur77 Bcl6 TRH
15,000 2,500 2,500 15,000 TRM1 *** *** **** ** 2,000 2,000 10,000 10,000 1,500 1,500 MFI 1,000 1,000 5,000 5,000 500 500
0 0 0 0
P2X7R CD73 CXCR4 ICOS
6,000 **** 2,500 8,000 10,000 ** **** ** 8,000 2,000 6,000 4,000 1,500 6,000
MFI 4,000 2,000 1,000 4,000 2,000 500 2,000
0 0 0 0
e 100 100 100
80 80 80 60 60 60
40 40 40
20 20 20
Maximum(%) 0 0 0 4 5 3 4 5 4 5 0 10 10 0 10 10 10 0 10 10 CXCR6 T-bet-ZsGreen CD11a
f CXCR6 Tbet CD11a 10,000 18,000 *** ** **** 10,000 8,000 16,000 6,000
MFI 5,000 4,000 14,000 2,000
0 0 12,000 Figure 4 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint a (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. day 9 day 14 day 30 5 10 4 10
3 10 PSGL1 0
3 4 5 3 4 5 3 4 5 0 10 10 10 0 10 10 10 0 10 10 10 FR4 b c Lung LN 5 5 4 10 10 10 ns ns ns 4 4 10 10
3 3 3 10 10 10
day 9 day 0 0
3 4 5 3 4 5 0 10 10 10 0 10 10 10 102 5 5 10 10 # of NP+ CD4 T cells T CD4 NP+ of # 4 4 10 10 101
3 3 day 30 day 10 10 Total TRH TRM1 0 0
PD1 control FTY720 4 5 4 5 0 10 10 0 10 10 CXCR5 d e Lung day 9
3 1 24 Lung LN Rora Izumo1r ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 1.5 ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● Ccr7 ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● 10 ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ●● ●● ● ●● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● 1 ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● S1pr1 ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ●●● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ●●● ● ●● ●●●●● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●●● ● ● ●● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ●● ●●●● ● ● ● ● ● d9 d9 ● 5 ● ● ● ● ●●● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●●● ●● ● ● ●● ●● ● ●●● ● ● d9 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ●●● ● ●● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ●● ● ●● ●●● ● ●● ●● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ●●●● ● ●● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ●●●● ● ● ●● ● ● ●● ● ● ●● ● ● ● ●● ● ● ●● ● ● ● ●●● ● ●●● ●● ●● ●●●● ● ●●● ●● ●● ● ● ● ● ● ●● ● ● ●●●● ●●●●●●●● ●●●● ●●● ●● ● ● ● ● ● ● ●●●● ● ●●●● ●●●●●● ● ● ● ●● ●● 0.5 ● ● ● ● ●● ● ●● ● ●● ● ● ● ● ● ● ●● ● ● ●● ●● ● ● ●● ● ● ●●●● ● ●●●● ● ● ● ●● ●● ● ● ● ● Icos ● ● ●● ●●●● ● ● ● ●●●● ●●● ● ●● ● ●●●● ● ●●● ● ● ● ● ● ● ● ● ●● ●● ●●●● ●●●● ● ●● ● ● ● ● ● ● ●● ● ● ●● ●●● ●●● ●●● ●● ● ● ●●●●●●●●●●●● ●●● ● d14 ● ● ● ● ●● ●●●●●●● ●● ●● ● ● ● ● ● ● ●●● ● ●●● ● ● ●●● ● ●●●●●● ●●●●●●●●●●●●●●● ●● ● ● ● ● ● ● ●● ●● ● ● ●● ●●●●●● ●●● ● ●●●● ●●●●●● ●●●●● ●●●● ● ● ● ● ● ● ●● ● ● ● ● ●● ●●●●●●●●●●●●●●●●●●●● ●● ●●●● ● ● ● ● ● ● ● ●● ● ●●● ●● ● ● ● ●●●● ●● ●●●●●●●●●●● ● ● ● ●● ●● ● ●● ● ● ●●●●● ● ●● ● ● ●● ●●●●●● ●● ●●●●●● ●●●● ● ● ●●●●●● ● ● ● d14 ● ●● ● ● ●● ● ● ●●● ●● ●● ●●●●● ●●●●●●● ●●● ● ●●●● ● ● ●●● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ●●●● ●●● ● ● ● ●● ● ● ● ●● ●●● ●● ●●● ●●●●●● ● ●●●● ● ● ● ● pca_2 ● ● ● ● ● ●●● ●●●● ● ● ● ● ● d14 ● ● ● ● ●●●● ● ●●●●●● ● ●●● ●●●●●● ●● ●● ● ●● ●●●●●● ●● ● ● ● ● ● ●● ● ● ● ●●● ● ● ● ● ● ●●● ● ● ●● ●●●●●●● ● ●●● ● ●● ● ● PC2 ● ● ● ● ●● ●●● ● ● ●● ●● ●● ●● ● ● ● ●● ● ● ● ● ● ● ●● ●●● ●● ●●●●●●● ●●●● ● ● ●● ●● ● ●● ● ● ●● ● ●● ● ● ●● ● ● ● ●● ● ● ●●●●● ●●●●● ● ● ●● ● ●●● ● ●●● ●●●●●●●●● ●●●●●●● ●●●● ●●● ●●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●●● ●●●● ● ●● ● ●● ● ●● ●● ●● ● ●●●●●●●● ● ● ●●●● ●● ● ● ● ●●● ●●●●●●● ● ● ● ● ● ● PC2 ● ● ● ●●●● ● ● ●● ● ● ● ●● ● ● ● ● ●●●● ● ● ● d30 ● ●●● ●●● ●●●● ●● ●●●●●● ●● ● ● ●● ●●● ● ● ●●● ●●● ● ●●●●●●●● ●●● ●●● ● ●●● ● ● ●● ● ●● ●●● ● ● ●●●●●● ●●●●●●●●●● ●●● ●● ●● ● ● ● ●● ● ●● ●●●● ●●●●●●●●●●●●●●●●●●●●●●●● ●●● ●●●● ●● ●● ●● ● ●●●● ●●●●●●● ●● ● ● ● ● ● ●● ● ●●● ●● ● ●● ●●●● ●●●●●● ● ●●●●●●● ●● ● ● ● ● ● ● ●●● ● ●● ●●●● ●●● ●● ● ● ● ● ● ●●● ●● ● ●● ●●● ●●●●●●●●● ● ● ●● ● ●●●● ● ●● ● ●●●● ●●●●●●●●●●●●● ●●● ●● ●●● ●●●● ● ●● ●● ● ● ●●● ●●●●●●●●●●●●●●● ●● ●●●●●● ●●● ● ●●●●● ● 0 ●● ●●●●●●●●●●●● ●●●●●●●●●●● ● ● ● ● ●● ●●●●●●●●●●●●● ● ●●●●●●●●●●●●●●●●● ● ● ● ● ●●● ● ● ● ●● ●● ● ● ● ●●● ● ●● ● ● ●●● ●●●●●●●●●● ●●●●● ● ●●● ●● ● ●●●● ● ● ● ●●●● ● ●●●●● ●●●●●●●●●●●● ●● ●●●●●●● ●●●●●●●● ● ●● ●● ●●●●●● ●●● ●●●●●●●●●● ●●●● ●●● ● ●● ● ● ● ●●● ●● ●● ● ●●● ● ●●●●●●●●●●●●●● ●●●●●●●●● ●●●● ● ●●●● ●● ● ●● ●●●●●●●●●●●●● ●●● ● ● ● ●● ●● ● ●●●● ●● ● ● ● ●●●●●●●●●● ● ● ●●●●●●●● ●● ● ● Cxcr6 ● ● ●● ● ● ●● ●●●●● ●● ●●● ●● ● ● ● ● ● ● ●●● ●●● ●●●●●● ●●● ●●●● ●● ● ● ● 0 ● ● ● ● ●●● ● ● ● ●● ● ● ● ●● ● ● ● ● ●●● ●●●●● ●●●●●●●●●●●●●●●●●●●●● ●●● ●● ● ● ● ●● ●● ●●●● ●● ●●●●● ●●●●● ● ●●●●●● ●●●● ●● ● ● ● ●● ●●●●●●●●●●●●●●●●●● ●● ●● ● ●●● ● ●●● ● ●● ● ●●●●●●●●●●●● ●●●● ● d30 ●● ● ●● ● ●●●● ●●●●●●●●●●●●●● ●● ●●●●● ●● ● ● ● ● ●●●●●●● ●●● ●● ● ●● ●●●●●●● ●●●●● ●●● ●●● ● ● ●●● ●●●●●●●● ● ●●● ●● ● ● ● ● ● ●●● ● ●●●● ●●●● ● ●● ●● ●● ● ● ● ●● ● ●● ●● ●● ● ●●●●●●●●●●● ●●●●●●●●●● ●●● ● ●● ●●● ● ● ● ● ●●● ● ●● ●●●●● ● ● ●●●● ●●● d30 ● ● ● ● ● ●● ●● ●● ● ●● ● ●●●●● ● ●● ● ● ● ●● ●●●●●●●●● ●●● ● ●● ● ●● ●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●● ●●●● ● ●● ● ●●●●●● ●●●●●●●●●●●●●●●●● ● ● ●●●● ● ●●●●●●●● ●●●●●●●●● ●●●● ● ●● ● ● ●●● ●●●●●●● ●●● ●●●●●●● ● ● ● ● ● ● ● ● ●●●●●●●●●● ●●● ●●●●● ●● ●●●● ● ●● ●● ● ● ●● ●● ●●● ●● ● ●●●●● ● ● ● ●●● ● ●●● ●● ● ● ● ● ●●●●●●● ● ●●● ● ● ●● ●● ●●●● ● ●●● ●●● ●● ●●●●●● ●●●●●●●●●●●●●●●●●●●●●● ●● ● ● ●●● ● ● ● ● ● ● ●●●● ●● ●●●● ● ● ● ● ●● ●●●●●●●●●●●●● ●●● ●●●●●●●● ● ●●●● ●●●● ●●●●●●● ●● ● ●● ● ●●●●●● ●●● ●●●● ● ● ● ●● ● ●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●● ●●● ●● ● ●●●● ● ● ● ● ● ● ●● ● ● ●● ●●●●●●●●●●●●●●● ●● ● ● ● ●●●●●●●●●● ●● ● ● ● ● ● ●● ● ● ● ●●● ● ●●● ●●●●●●●●●●●●● ● ●●●●●● ●● ● ●● ●● ● ● ● ● ● ●●●● ●●● ●●●●● ●●●●●●●● ●●●●●●● ●●●●●● ●●● ● ● ● ● ●● ● ● ●●●●● ●●●●● ●●●●●●●●●●●●● ● ● ● ●●● ●● ● ● ● ● ● ● ● ● ●●●●● ●●●●●●●● ● ● ●●●● ●●●●● ●● ●● ●● ● ● ● ● ● ● ●● ● ● ● ● ●●●● ●●●● ●●●● ● ● ●● ● ● ● ● ● ● ● ●●●●●●●●●● ● ● ● ●●●●●● ● ● ●● ● ● ● ● ● ● ● ●● ● ●● ●● ●●●●● ●●● ● ●●●● ● ● ● ● ● ●● ●● ● ●●●●● ● ● ● ●● ● ●● ● ● ● ● ●●● ● ●●● ●● ●●●●● ●● ● ●● ●●● ●● ● ●●●● ● ● ● ● ● ● ● ●● ● ● ● ●●● Tcf7 ● ● ● ● ● ● ●● ●●● ● ● ● ● ● ● ●●● ● ●●● ● -0.5 ● ● ● ● ● ● −5 ● ● ● ● ● ● ● Il2rb −10 −5 0 5 10 -1 PC1 pca_1PC1 Cxcr5 Il7r Hif1a Stat4 Selplg
f g Lung Lung d9 d14 d30 Cluster Cluster 1 d30 Pou2af1 Pou2af1 Day 30 d9 0.5 Day 14 Npdc1 Npdc1 d14 Pole4 Pole4 Day 9 0
0.6 0.9 1.2 1.5 0.5 1.0 1.5 2.0 Nr4a3 Nr4a3 TFH Th1 Lef1 Lef1 −0.5 Smarcc2 Smarcc2 −1 LN Gtf2i Gtf2i Zfp729b Zfp729b Trp53 Trp53 Day 30 Gmeb1 Gmeb1 Day 14 Mlx Mlx Day 9 Las1l Las1l 0.5 1.0 1.5 2.0 12 Hsf2 Hsf2 TFH Th1 Prdm1 Prdm1 Runx3 Runx3
-1 -0.5 0 0.5 1 Figure 5 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. a b PBS anti-CD20 100 ****
5 50.7 5 77.9 80 10 10 4 4 60 RH
L1 10 10 T G
S 3 3 % 40
P 10 10 0 0 45.4 18.7 20
4 5 6 4 5 6 0 10 10 10 0 10 10 10 0 PBS anti-CD20 FR4 c PBS ns ** anti-CD20 104
103 # of NP+ CD4 T cells T CD4 NP+ of # 102 TRM1 TRH
d e 1.0 **** ΔCD4 wildtype (CD45.1) Bcl6 (CD45.2) **** 5 5 0.8 10 10 ype 54.8 98.9
4 4 0.6 wildt
10 10 :
CD4 L1
3 3 Δ 0.4
G 10 10
S
cl6 P B 0.2 0 0 42.6 0 0.0 4 5 4 5 n 0 10 10 0 10 10 io RH itut RM1 T FR4 T
reconst Figure 6 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint a (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. TbetB220CD4 B220CD4 TbetCD4
Bcl6B220CD4 B220CD4 Bcl6CD4 Bcl6 RFP Bcl6 ZsGreen Tbet b
iBALT location ZsGreen intensity CD4+ Tbethi Tbet ZsGreen Tbet
2.5 3.0 3.5 4.0 4.5
iBALT location RFP intensity CD4+ Bcl6hi Bcl6 RFP Bcl6
2.5 3.0 3.5 4.0
c d CD4 in iBALT CD4 in iBALT CD4 outside iBALT Machine learning Guided quantification Guided quantification quantification 5 **** 5 5 *** **** ) -6
) 4 ) 4 4 -4 -4 count 4 (x10
3 3 3 T CD L A
2 iB 2 2 alized hin outside iBALT (x10 iBALT outside NormalizedCD4 count within iBALT (x10 iBALT within 1 wit 1 1 NormalizedCD4 count Norm 0 0 0 Tbethi Bcl6hi Tbethi Bcl6hi Tbethi Bcl6hi Figure 7 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. a NP **** 100 resident circulating 100 100
80 80 (%) 50 60 60 % BrdU % um 40 40 27.0 81.7 20 20 Maxim 0 0 0 3 4 5 3 4 5 0 10 10 10 0 10 10 10
BrdU resident circulating b CD44+
circulating 80 resident **** 100 100
80 80 (%) 60 60 60 um 40 40 44.1 66.6 % BrdU % 20 20 40 Maxim 0 0 3 4 5 3 4 5 0 10 10 10 0 10 10 10 20 BrdU
resident circulating c d NP CD44+
50 80 Tamoxifen 0.1884 0.5148 fl/fl 40 MHC-II PR8 60 ΔUBC-ERT2 MHC-II 40-50d 54-64d 30
rdU 40 B 20 % 20 10
0 0
TRH TRH TRM1 TRM1 e f
MHC-IIfl/fl MHC-IIΔUBC-ERT2 50 * 5 5 10 69.3 10 93.5 40 4 4 10 10 30 L1 G 3 3 TRH % 20
S 10 10 P 0 0 10 28.6 5.2 0 4 5 4 5 0 10 10 0 10 10 MHC-IIfl/fl FR4 MHC-IIΔUBC-ERT2 g TRM1 TRH
104 ns 103 **
103 102
# of NP+ CD4 T cells T CD4 NP+ of # 102 101
MHC-IIfl/fl MHC-IIΔUBC-ERT2 Figure 8 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint a (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. Trajectory 1 Trajectory 2 2
2 1
E Early E
2 T
T 3 Mid HA
HA Late
4 P P
PHATE 1 PHATE 1 b Izumo1r IL7r Id2 Selplg ession
expr Hif1a S1pr1 Cxcr6
ed 1 2 3 mput I 4 Trajectory 1 Trajectory 2 Pseudotime c d 100 TRM1 TRH ** TRM1 5 5 10 TRH 10 10.1 53.1 80 4 endogenous
L1 4 10 10
G 60 S 3 3 R4+ 10 10 P F 0 0 40
3 % -10 3 4 5 3 4 5 20 0 10 10 10 0 10 10 10 FR4 0 e f g h Bcl6fl/fl Bcl6ΔCD4-ERT2 TRM1 TRH **** 4 ns 4 4 10 10 * 10 ) ) -3
-4 3 10 count count 10 (x 4 4
3 3 (x
10 10 T L T CD CD
L 2 cells A A
# 5 iB
iB
2 2 alized 10 10 alized
hin 1 side wit out Norm Norm 101 101 0 0 Bcl6fl/fl Bcl6ΔCD4-ERT2 i j NP+ ASCs Bcl6fl/fl ** ΔCD4-ERT2 104 Bcl6 Tamoxifen fl/fl αCD8 Bcl6 PR8 X31 ΔCD4-ERT2 30-35d d4 analysis Bcl6 20d FTY720 cells 3
# 10
102 Supplementary Figure 1 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint a (which was not certified by peer review) is the author/funder.b All rights reserved. No reuse allowed withoutc permission. Blood Lung NP+ CD4 T cells Lung 100 ns 105 40 ** 80 104 60 30 103 40 20 2
% of Maximum of % 10
% i.v. negative i.v. % 20 1 10 0 10 cells T CD4 %
i.v. CD45 cells T CD4 NP+ of # 100 0
DMSO DMSO FTY720 FTY720 d Liver-specific genes (LCMV) Spleen-specific genes (LCMV) on flu samples on flu samples
CirculatingLN Lung CirculatingLN Lung Pim1 Itgb1 Dusp1 Gm10073 Gimap7 Gm5093 Junb Gm10036 Ccnd2 Gm6133 Hist1h2ao Gm8730 Hist1h2ap Gm7808 Ly6c2 Rpl13a−ps1 Epsti1 Gm9493 Crip1 Gm26917 Vps37b mt−Atp6 Ifngr1 1 Ftl1−ps1 Lgals1 Malat1 Ifng Rnaset2a Rgs1 0.5 Sostdc1 Id2 Cd27 Gzmb 2310001H17Rik Bhlhe40 0 Rnaset2b Hspa1a e Izumo1r Ccl4 mt−Nd1 Xcl1 −0.5 Rgs10 Cxcr6 Id3 Tnfaip3 mt−Cytb Rgs2 −1 Lrrc58 Ccr2 Dennd2d Ccl5 Gm42418 Nfkbia Tcf7 Sh3bgrl3 Tspan32 Plac8 Arhgap45 Nkg7 Slamf6
Lung-specific genes (flu) LN-specific genes (flu) on LCMV samples on LCMV samples
Liver4 Spleen1Spleen4 Liver4 Spleen1Spleen4 Hif1a Gm10073 Areg Ptprcap Tsc22d3 Gm42418 Cstb Ltb Fth1 Rps27rt Ccl5 Rps29 Vps37b Ptpn7 Ifngr1 mt−Nd3 Nkg7 Art2b Lgals1 Asap1 Tnfrsf18 Izumo1r Hilpda Lrp10 Crip1 Lrrc58 S100a6 Shisa5 Tnfrsf4 Dennd2d Mt1 Cxcr5 Fam107b Pou2f2 Sub1 Tcf7 Rora Apobec3 Hspa1a Ppp1r14b Ostf1 Tspan32 H2afz Il16 Arpc3 Limd2 D8Ertd738e Coro1a H3f3b Eif3f Ctla2a Chd3 Crem Rpl38 Ramp3 Gm10263 Gch1 Rps28 Emb Supplementary Figure 2 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. a
Day 30
Rora Gch1 Fam107b Nop53 H2afz Emb 3.5 0.8 1.50 3.0 1.2 1.0 0.6 1.0 1.25 2.5 0.4 0.8 2.0 1.00 0.5 0.5 1.5 0.2 0.4 0.75 1.0 Lung LN Lung LN Lung LN Lung LN Lung LN Lung LN
Ramp3 Fth1 H3f3b Crem Vps37b Got1 2.0 4.5 1.5 4.4 0.75 1.5 2 1.0 4.0 4.0 1.0 0.50 0.5 3.5 1 3.6 0.5 0.25 0.0 3.0 Lung LN Lung LN Lung LN Lung LN Lung LN Lung LN
Lung 3 Lung 1 Lung 4 Lung 2 LN 3 LN 2 LN 1 LN 4
b Day 30
Sh2d1a Ppp1r14b Rgs10 Hif1a Areg Tnfrsf4 1.6 2.5 3.0 1.5 0.9 1.2 2 2.5 2.0 1.0 0.6 2.0 0.8 1.5 1 1.5 0.5 0.3 1.0 0.4 1.0 0.0 0.5 Lung LN Lung LN Lung LN Lung LN Lung LN Lung LN
Sostdc1 Cxcr5 Id3 Tnfsf8 Eea1 Ramp3 1.5 0.5 3 0.6 1.5 0.4 1.0 1.0 2 0.4 0.3 1.0
0.2 1 0.5 0.5 0.5 0.2 0.1 0 0.0 0.0 0.0 0.0 0.0 Lung LN Lung LN Lung LN Lung LN Lung LN Lung LN
Il7r S1pr1 Ccr7 Selplg Cxcr6 Tbx21 0.5 0.5 1.00 1.5 0.75 0.4 0.20 0.4 0.75 0.3 0.50 0.3 1.0 0.15 0.2 0.2 0.50 0.25 0.10 0.1 0.1 0.5 0.25 0.00 0.0 0.05 0.0 −0.1 0.0 Lung LN Lung LN Lung LN Lung LN Lung LN Lung LN
Bcl2 Eef1b2 Rps19 Nkg7 Ccl5 Id2 0.6 2.5 5 6 3 2.0 2.0 0.4 4 4 2 1.5 1.5
1 2 1.0 0.2 1.0 3
0 0 0.5 0.5 Lung LN Lung LN Lung LN Lung LN Lung LN Lung LN
Lung 3 Lung 1 Lung 4 Lung 2 LN 3 LN 2 LN 1 LN 4 Supplementary Figure 2 c bioRxiv preprint doi: LNhttps://doi.org/10.1101/2020.02.28.963280 day 30 ;d this version posted FebruaryLN 28, day 2020. 30 The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1234 TFH memory Th1 memory Rps19 Klf6 Rps18−ps3 Rpl12 Itgb1 4 Rps18 4 Gm10260 3 3 Rps12 2 2 Rps5 Rpl22l1 1 1 Rps4x 0.75 1.00 1.25 1.50 1.75 2.00 0.5 1.0 1.5 2.0 2.5 Ifi27l2a Vim Rps7 Gene set average normalized expression Rpl32 Il7r Gm9493 Rps12−ps3 Rps28 Rpl36a Marcksl1 Srgn e Rilpl2 Miragaia et. al. 2019 Angptl2 Tox Gem Cd200 Crem Eif1 Fth1 Nfatc1 2 Izumo1r Tespa1 Lmo4 Hif1a 1 Prkca 2 Tbc1d4 0 Ptprc 1 Egr2 Hsp90ab1 Apoe 0 −1 Ppp1r14b Tubb2a -1 Krt83 Pdcd1 H3f3b Sh3brgl3 Sub1 Tubb2b -2 2 Rnaset2a Sh2d1a Skin Rnaset2b 1 Rgs10 Colon Tmem
Art2b FC Sostdc1 2 0 Colon Treg Smco4 LT Tmem
Ascl2 log Matk Lgals1 −1 LT Treg Cxcr5 Nav2 Rgs16 Smox Tnfrsf4 Ubb Vps37b Ccl5 2 Nkg7 Rgs1 Id2 1 Cxcr6 AW112010 Ctsw 0 Gimap7 Epsti1 Ccl4 Serpina3g −1 Ccr2 Il10 Selplg Glrx Skin Skin Skin Gimap4 Ms4a4b LT Treg LT Treg LT Treg LT Tmem LT Tmem LT Tmem Dnajc15 Colon Treg Colon Treg Colon Treg Plac8 Colon Tmem Colon Tmem Colon Tmem Cish
g f Mackay et. al. 2013 Lung residency genes on LCMV data Lung Spleen Naive Crem Eif1 Fth1 Spleen TCM Spleen TEM Spleen Liver 2 10564809 Eif1 0.02 0.04 0.06 0.08 2.2 2.4 2.6 2.1 2.4 2.7 3.0
10597573 Eif1 1 H3f3b Sh3bgrl3 Sub1 10607865 Tmsb4x 10457205 Vps37b 0 Spleen 10533729 Crem Liver 10461402 Sh3bgrl3 −1 1.8 2.1 2.4 2.7 1.0 1.5 2.0 2.5 3.0 1.0 1.5 2.0 10517169 Fth1 −2 Tmsb4x Tnfrsf4 Ubb
Spleen Liver GSM1143718GSM1143717GSM1143719GSM1143720GSM1143727GSM1143728GSM1143721GSM1143722GSM1143726GSM1143725GSM1143723GSM1143724 4.5 5.0 5.5 0.0 0.2 0.4 0.6 0.8 2.6 2.8 3.0 3.2 Vps37b
Spleen Liver 0.2 0.4 0.6 Imputed expression Supplementary Figure 2 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. h i Regulons - day 30 Regulons - day 30 LN Lung
Cluster Cluster Hif1a LN1 Hif1a LN2 Zeb1 LN3 Foxp3 Foxp3 LN4 Lung1 Hmgb2 Fosl2 Fosl2 Lung2 Lung3 Klf4 Crem 1.5 Lung4 1.0 Hmgb2 Foxp3 0.5 E2f5 Fosl2 0.0 Klf10 Zeb1 -0.5 Nr3c1 -1.0 -1.5 Tgif1 Gmeb2 Runx3 Runx3 Atf3 Foxj2 Runx2 Klf4 Ppard Tgif1 Mef2d Mef2d Stat3 Crem Klf10 Mxi1 Ppard Stat3 Atf3 Hivep1 Prdm1 1.5 Runx2 Mxd1 Mxd1 Sp3 Foxp1 Irf2 0 Irf2Irf2 Mlx Irf9Irf9 Irf9 Nfkb2 -1.5 Stat6 Nfatc1 Stat6 Nfe2l1 Junb Junb Tfap4 Lef1 Foxp1 Max Stat6 E2f4 Stat1 Lef1 Stat1 Zbtb7b Stat1 Smarcb1 LN2 LN3 LN1 LN4 Mlx
Relb Lung3 Lung4 Lung2 Lung2 Mlx Arid5b Tfap4 Bcl3 Nr4a1 Ikzf1 Nfkb1 Batf Zbtb7b Xbp1 Max Bcl6 Cebpa Hsf4 Id1Id1 Relb Bcl3 Egr2 Nfatc1 j Arid5b Izumo1r Selplg Foxp3 LN2 LN3 LN1 LN4 Lung3 Lung4 Lung1 Lung2 4 2.5 4 2.0 3 3 1.5 2 2 1.0 1 1 0.5 0 0 0.0 Lung LN Lung LN Lung LN
k NP+ CD44+ 5 5 10 0.6 10 11.0
4 4 10 10
3 3 10 10 CD25
0 0
3 3 -10 -10 4 4 0 10 0 10 Foxp3 Supplementary Figure 3 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. a
5 PBS 10 ns ns FTY720 104
103
102
101
0 # of NP+ CD4 T cells T CD4 NP+ of # 10 TRM1 TRH
b c OT2 (PR8-OVA2) OT2 (PR8-OVA2) **** 50 4 10 40
L1 3 10
G 30 RH S 2 97.6 T
P 10 20 0 % 10 2.4 3 4 0 0 10 10 OT2 NP+ FR4 Supplementary Figure 4 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. a b Day 9 Day 14 Day 30 day 9 day 14 day 30
0.9 Lung
0.6
Lung 0.3 Selplg
0.9 LN tSNE2 0.6
0.3
0.5 1.0 1.5 2.0 0.5 1.0 1.5 2.0 0.5 1.0 1.5 2.0 Izumo1r Lymph nodeLymph d9 Lung d9 LN d14 Lung d14 LN tSNE1 d30 Lung d30 LN c
Day 9 Day 14 Day 30
Ccl5 Vps37b ● ● ● Fth1 600 ●
● 600 200 Ifngr1
● Vps37b Plac8 ● Rps27 ● ● ● ● Rps29 ● ● Crem 150 ● Rps27rt Crem ● ● ● 400 Chd3 Rps15a ● Tnfrsf4 400 Rps29 ● ● mt−Nd3 Rps28 ● ●● Gm9493 ● Lyz2 ● ● Actb Il16 Dennd2d Sub1 ●● Gm10260 ● Tcf7 Nkg7 ● ● ● ●● ● Dgat1 ● Rpl39 ● ● Limd2 Gm42418 Gm10073 Rpl35a Ctla2a Ramp3 ● ● ● ● Dennd2d Ltb ● ● ● Rora S100a4● ● Rps18−ps3● ● 100 ● ● ● ● ● ● ● ●● Apobec3 ● ●● ●● ● Nkg7 Tspan32 Limd2 ● Ptprcap H2afz ● Tnfrsf4 Ctla2a ● ● ● ● Lgals3 Fth1 ● ● ● ● ● ● ● ● ● ● ● ● Tspan32 ● ● ● ● ●● Ppp1r14b Ltb ● Rps28● ● ● ● Gm9844 ● ● ● ● ● Ramp3 ● ● ● Fcer1g ● ● ● ● Evl ● ● Ccl5 Tcf7 ● ● ● ●● ● ● ●● ● ●● Vps37b ● ● ● ● ● ● ● ● ● ● Sostdc1 Il16 ● 200 Ppp1r14b Tcf7 ● ● ● ● ●● 200 ● ● ● ● Cd74 ● ● ● ● ● ● Crip1 ●●●●● ● Ifngr1 ● ● ● ● ● ● ● ● ● Vim ● ● ●● ● ● Slamf6 ● Ptprcap ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● Slamf6● ● ● ● Gm10263● ● ● ● ● ● ● ●● Tyrobp ● ● ●● ● ● ● Rgs10Cdk2ap2● ● ● ● ● ● ●●● ● ● ●● ● ● ● 50 ● ● Plac8 ● Tnfrsf18 -log10(adjusted P-value) ● ● Tnfrsf4 ● ●● ● ●●● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● Apoe ● ●●● ● ● ●●●●● ● ●● ● ● ● ● ● ● ●● ● ●● ● ● Crip1 ● ● Izumo1r ● ● ● ● ● Ctla2a ● ● ● C1qa ●● ● ●● ● ● ● ● ●●● ● ●● ● ●● ● ●● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ●● Hspa1a ● Ostf1● ● ● ● ●● ● ●● Ubac2 ● ●● ● ● ●● ● ● ● ● ● ●●●●● ● ● ●●● ● ●●● ● ● ●●● ●● ● ● ● ● ● ● ●● ●● ● ● ● ● ●● ● ● ●● ● ● ● ●●●●●●● ● ● ● ● ●●● ● ● ● ●●●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ●●●●●● ● ● ● ●●● ● ● ●●● ● ● ● ●● ● ● ● ● ● ● ●●●● ● ● ● ●● ● ● ● ●●● ● ●● ● ● ● ● ● Nkg7 ● ●●●●●●● ● ● ● ●● ● ●● ● ●●● ●● ● ● ● ● ● ●●●●●● ● ● ●● ● ● ● ● ●●●●●● ●● ●●●●● ● ● ● Sostdc1 ● ● ●●●●●●●●●● ●● ● ● ● ●●●●●●●●● ● ●● ● ● ●●● ● ●● ● ● ● ●●●●● ●● ● ● ●●● ● ●● ●● ●● ● ●●●●●●●● ●● ● ● ●●● ●● ●●●●●●● ● ● ●●● ●● ●●●●● ● ● ●●● ● ●● ●● ● ● ● ● Ccl5 ● ●●●●●●● ●●● ● ● ● ●●●●●●●●●● ● ● ● ● ●●● ● ●● ● ● ●● ● ●●●●●●● ● ●●●●●●●●● ● ●● ●●●●● ●●●●●● ●●● ● ●● ● ● ●●● ●●●●●●●● ● ●● ● ●● ●●● ● ●●● ● ● ●● ● ●●●●●●● ● ●●● ●● ● ●●● ●● ● ● ●●●●●● ●●● ●● ●●●●●● ●●●●●● ● ●●● ●● ● ●●● ●●●●● ●● ●●●●● ●●● ● ●●● ●● ●●●●●●●●●●●●●●● ●●●●● ● ● ●●●● ● ● ● ● ●●●●● ●● ● ● ● 0 ● ● ● ●● ●● ● ●● ● ● 0 ● ● ● ● ● 0 Cstb −2 −10 123 −1 0 1 2 0 1 2 Average log2FC lung over lymph node d Lung
Gch1 Prkca
0.6 2.0 0.5 1.5 0.4
0.3 1.0 0.2 0.5 d9 Hif1a Selenok d14 d30 2.0 1.2
1.5 1.0
1.0 0.8
0.5 Supplementary Figure 5 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. a PBS **** ** 106 anti-CD20 105 104 103
102 # of B cells B of # 101 100 i.v. + B cells i.v. - B cells
b c Wildtype Bcl6ΔCD4 1.0 #### 5 61.6 5 0 10 10 **** 0.8 4 4 10 10 0.6 3 3 :wildtype PD1 10 10 CD4 0 35.2 0 100 Δ 0.4 3 3 -10 -10 3 4 3 4 Bcl6 0.2 0 10 10 0 10 10 CXCR5 0.0
TFH non TFH
reconstitution Supplementary Figure 6 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder.100 All rights reserved. No100 reuse allowed without permission. a 80 80 60 60
non antigen specific 40 40 5 5 10 10 20 20 Maximum(%) 4 4 0 0 10 10 4 5 4 5 0 10 10 0 10 10 3 CXCR6 CD11a
CD44 10 PSGL1 0 0 100 100
80 80 4 4 5 0 10 0 10 10 60 60
PD1 FR4 40 40 20 20
Maximum(%) 0 0 3 3 4 4 5 -10 0 10 10 0 10 10 P2X7R CXCR4
b NP-specific non antigen specific 5 5 5 10 60.1 10 6.8 10 87.0 4 4 4 10 10 10
3 3 3 10 CD4 10 10 PSGL1 PSGL1 0 0 0 33.3 3 4 5 3 4 5 3 4 5 0 10 10 10 0 10 10 10 0 10 10 10 FR4 T-bet ZsGreen FR4
NP-specific non antigen specific 5 5 5 10 49.3 10 4.5 10 4 4 4 c 10 10 10
3 CD4 3 3
PSGL1 10 10 PSGL1 10 0 0 0 47.4 85.2 3 4 5 4 5 3 4 5 0 10 10 10 0 10 10 0 10 10 10 FR4 Bcl6 RFP FR4 Supplementary Figure 7 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. a b CD45 i.v.- B cells (resident) CD45 i.v.+ lung B cells (circulating)
**** **** 4000 100 10000 100 (%) (%) 80 80 8000 3000 I I um um 60 6000 60 MF MF 2000 40 4000 40
Maxim 1000 Maxim 20 2000 20 0 0 4 5 0 4 5 0 0 10 10 0 10 10 MHC-II MHC-II c d MHC-IIfl/fl MHC-IIfl/fl MHC-IIΔUBC-ERT2 MHC-IIΔUBC-ERT2 ns 6 ns 106 10
5 105 10
4 104 10 # of circulating lung B cells circulatingB lung of # # of resident lungcells resident B of # 3 103 10
Supplementary Figure 8 bioRxiv preprint doi: https://doi.org/10.1101/2020.02.28.963280; this version posted February 28, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. a Transferred TRM1 Transferred TRH
5 5 10 10 99.1 99.0
4 4 10 10 0.17 0.19 3 3 10 10
0 0 3 CD45.1(host) 3 -10 -10 3 3 4 5 3 3 4 5 -10 0 10 10 10 -10 0 10 10 10 CD45.2 (donor)
b Bcl6fl/fl Bcl6ΔCD4-ERT2
B220 CD4
c Bcl6fl/fl Bcl6ΔCD4-ERT2
377778
125926
41975
13992
4664