Redefining the Role of Lymphotoxin Beta Receptor in the Maintenance of Lymphoid Organs and Immune Cell Homeostasis in Adulthood

Lymphotoxin beta receptor (LTβR) is a promising therapeutic target in autoimmune and infectious diseases as well as cancer. Mice with genetic inactivation of LTβR display multiple defects in development and organization of lymphoid organs, mucosal immune responses, IgA production and an autoimmune phenotype. As these defects are imprinted in embryogenesis and neonate stages, the impact of LTβR signaling in adulthood remains unclear. Here, to overcome developmental defects, we generated mice with inducible ubiquitous genetic inactivation of LTβR in adult mice (iLTβRΔ/Δ mice) and redefined the role of LTβR signaling in organization of lymphoid organs, immune response to mucosal bacterial pathogen, IgA production and autoimmunity. In spleen, postnatal LTβR signaling is required for development of B cell follicles, follicular dendritic cells (FDCs), recruitment of neutrophils and maintenance of the marginal zone. Lymph nodes of iLTβRΔ/Δ mice were reduced in size, lacked FDCs, and had disorganized subcapsular sinus macrophages. Peyer`s patches were smaller in size and numbers, and displayed reduced FDCs. The number of isolated lymphoid follicles in small intestine and colon were also reduced. In contrast to LTβR-/- mice, iLTβRΔ/Δ mice displayed normal thymus structure and did not develop signs of systemic inflammation and autoimmunity. Further, our results suggest that LTβR signaling in adulthood is required for homeostasis of neutrophils, NK, and iNKT cells, but is dispensable for the maintenance of polyclonal IgA production. However, iLTβRΔ/Δ mice exhibited an increased sensitivity to C. rodentium infection and failed to develop pathogen-specific IgA responses. Collectively, our study uncovers new insights of LTβR signaling in adulthood for the maintenance of lymphoid organs, neutrophils, NK and iNKT cells, and IgA production in response to mucosal bacterial pathogen.


INTRODUCTION
Lymphotoxin beta receptor (LTbR) belongs to the tumor necrosis factor receptor superfamily (TNFR) and is known as a key regulator of lymphoid organogenesis and inflammation (1)(2)(3)(4). Therapeutic strategies for inhibition or stimulation of LTbR signaling are currently in development for treatment of inflammatory and infectious diseases as well as cancer (5)(6)(7). However, the impact of LTbR inactivation in adulthood remains incompletely understood.
LTbR signaling can activate both canonical and alternative NF-kB signaling to induce various proinflammatory chemokines and cytokines (1,9,27). Paradoxically, LTbR -/mice exhibit an autoimmune phenotype which includes splenomegaly, autoantibody production and systemic inflammation with increased neutrophil numbers in spleen and multiple lymphocytic lymphoid infiltrates in non-lymphoid organs (28)(29)(30). Several explanations for an autoimmune phenotype in these mice have been suggested, including lack of lymph nodes and follicular dendritic cells (FDCs) (31,32), impaired thymus structure and central tolerance (29,(33)(34)(35), and altered microbiota composition (30,36). Since these defects are imprinted during embryogenesis or early neonatal stages, the role of LTbR in lymphoid tissue maintenance, autoimmunity, homeostasis of innate immune cells, IgA production and immune responses in adulthood remain unclear.
Biochemical inhibition of LTbR signaling with LTbR-Ig, a soluble decoy protein containing the extracellular portion of LTbR fused with the Fc portion of IgG, has proven to be a valuable tool for deciphering the role of LTbR signaling in adulthood (5). LTbR-Ig treatment partially impaired the structure of secondary lymphoid organs, and showed therapeutic efficacy in various models of inflammatory diseases (5). However, interpretation of LTbR-Ig effects in vivo is complicated, as it blocks both membrane LT and LIGHT ligands, which are known to play distinct and overlapping functions due to LIGHT binding to another receptor, HVEM (1,26,37,38). Additionally, repeated administration of LTbR-Ig may induce side effects due to anti-drug antibody formation and modulation of Fc-linked receptors (39,40).
In this study, to define the role of LTbR signaling in adulthood independent from LTbR signaling during embryogenesis and neonate stages, we generated mice with inducible ubiquitous genetic inactivation of LTbR in adult mice (iLTbR D/D mice). Our results demonstrate that LTbR signaling in adulthood is critical for the maintenance of spleen, lymph nodes, gut-associated lymphoid organs; homeostasis of neutrophils, NK and iNKT cells; and specific IgA antibody responses against mucosal pathogen Citrobacter rodentium; but is dispensable for the maintenance of polyclonal IgA production and autoimmunity.

Mice
LTbR -/and LTbR fl/fl mice were described previously (41). R26-CreERT2 mice (stock #008463) (42) and MRL/MpJ-Fas lpr /J (MRL mice) (stock #000485) were purchased from the Jackson Laboratory. Mice with inducible LTbR deletion (iLTbR D/D mice) were generated by crossing LTbR fl/fl mice with R26-CreERT2 mice. 6-8 week old adult mice were treated with 5mg tamoxifen (TAM) in 100ml of corn oil containing 10% ethanol by oral gavage for 4 consecutive days to generate stable deletion of Ltbr gene (iLTbR D/D mice). Depending for experiment, two different types of experimental controls were used: Cre-positive LTbR fl/fl littermates treated with corn oil containing 10% ethanol or Crenegative LTbR fl/fl littermates treated with TAM. We did not find any difference between these two control groups (data not shown). Both sexes were used for experiments. Mice were used for experiments one month after TAM treatment, unless specified. Animals were housed under specific-pathogen-free conditions in accordance with National Institutes of Health guidelines. All experimental animal procedures were approved by the Institutional Animal Care and Use Committee of University of Texas Health Science Center San Antonio. rinsed twice in PBS to remove fecal material. Colon and cecum pieces were incubated in complete medium (RPMI-40 supplemented with 3% FBS, 1 mM penicillin-streptomycin) containing 2 mM EDTA for 20 min at 37°C with slow rotation (100 rpm) to remove epithelial cells. Remaining tissue pieces were digested in serum-free RPMI-40 containing 200 mg/ml Liberase TM (Roche) and 0.05% DNase I (Sigma) for 40 min at 37°C with shaking at 100 rpm. The digested tissue was passed through a mesh screen, washed with PBS containing 3% FBS and separated by 80/40% Percoll gradient (GE Healthcare).

Real-Time RT-PCR Analysis
RNA from murine tissue was isolated using the E.Z.N.A. Total RNA kit I (Omega Bio-tek). cDNA synthesis and real-time RT-PCR was performed as described previously (41) using Power SYBR Green master mix (Applied Biosystems). Relative mRNA expression of target genes was determined using the comparative 2 −DDCt method. Primers used are listed in Supplementary  Table 1.

Analysis of LTbR Gene Deletion
To assess LTbR deletion efficiency, tissue was homogenized in 800ml lysis buffer (100mM Tris-HCL, 5mM EDTA, 0.2% SDS, 200mM NaCl). 4ml of 20mg/ml proteinase K was then added to each samples and samples were incubated at 55°C on a shaker overnight. Tubes were vortexed and centrifuged for 15 minutes at full speed in a microcentrifuge. Supernatants were transferred to new 1.5mL tubes and DNA precipitated with 500ml of isopropanol. The DNA pellets were washed with 70% ethanol and resuspended in 70ml TE buffer (10mM Tris, 1mM EDTA, pH 7.5). DNA deletion of LTbR was analyzed by PCR using primers: 351-CAGTGGCTCCAAGTGGCTTG, 352-GCAAACCG T G T C T T G G C T G C , a n d 4 4 1 -A C A G G G C A G A C A TTAGGGTTCC as described (41

C. rodentium Infection and Assessment of Bacterial Burden
C. rodentium infection and analysis of bacterial burdens was done as described (44). Briefly, mice were infected with C. rodentium strain DBS100 (ATCC 51459) by oral gavage with 2*10 9 colony-forming units (CFUs) in 0.2mL of PBS. Tissue samples were homogenized in PBS, serially diluted, and plated on MacConkey agar plates. CFUs were counted after incubation at 37°C for 18-24 hours.

ELISA
For the analysis of total IgA, IgG, and IgM in sera and feces, 96 well plates (Immulon 4 HBX) were first coated with anti-mouse IgA (1:500), IgG (1:1000), or IgM (1:1000) diluted in 1x carbonatebicarbonate buffer at room temperature overnight. Plates were then washed 3 times with PBS containing 0.1% tween 20 (PBST). Serially diluted serum or fecal extracts were added and incubated at 25°C for 1 hour. Plates were washed, and HRP-conjugated goat anti-mouse IgG, IgA, or IgM antibodies (Southern Biotechnology Associates, Inc.) were added and incubated for 1h at 37°C. Plates were developed using ABTS (1-Step ™ ABTS Substrate Solution (ThermoFisher) and OD410 values obtained on a BioTek SynergyHT plate reader. Analysis of C. rodentium-specific immunoglobulin (IgA, IgG and IgM) levels in serum or fecal samples was performed as described (44). Anti-dsDNA antibody detection was done on serum samples from naïve mice 2-4 months after TAM treatment as previously described (45). Briefly, plates were precoated with methylated BSA overnight (5mg/mL) at 4°C overnight. Plates were then washed with PBST and coated with 50mg/mL of calf thymus DNA (Sigma D-4522) at 4°C overnight, blocked with BSA (5mg/mL) and diluted serum added for 2h at room temperature followed by secondary HRP-conjugated antibody and TMB substrate detection.

Inducible Genetic Inactivation of LTbR in Adulthood
To clearly define the role of LTbR signaling in the adulthood, independent from the role of LTbR during embryogenesis and neonate stages, we utilized a well-established genetic approach to induce ubiquitous gene deletion. To do this, we crossed LTbR fl/fl mice (41) with transgenic mice expressing Cre recombinase linked to estrogen receptor-T2 (Cre-ERT2) (42), generating mice with inducible LTbR gene inactivation (iLTbR D/D mice). Tamoxifen (TAM) administration permits Cre release to the nucleus and induces recombination of target LoxP sites. Adult iLTbR D/D mice were treated with 5 mg of tamoxifen (TAM) in 100ml of corn oil containing 10% ethanol by oral gavage for 4 consecutive days. To define the efficacy of LTbR inactivation in mice following TAM treatment, we analyzed LTbR expression by qPCR from tissues collected from iLTbR D/D and control LTbR fl/fl mice one month after TAM treatment ( Figure S1A). In tissues with high epithelial/mesenchymal cell populations (colon, liver, and kidney), the LTbR mRNA downregulation was over 90% ( Figure S1A). Notably, LTbR expression in the colon and liver was decreased 100-fold in iLTbR D/D mice following oral or intraperitoneal TAM treatment ( Figure S1A, B). High efficacy of LTbR mRNA inhibition was also detected in lymphoid organs: mesenteric LN (MLN), inguinal LN (ILN), PPs and spleen ( Figure S1A). Overall, we found the highest reduction of LTbR expression in organs, which have large populations of LTbR expressing cells (epithelial cells, stromal cells, DCs and Mph). To confirm LTbR gene deletion, we assessed genomic LTbR DNA levels in colons from iLTbR D/D mice after TAM treatment ( Figure S1C). PCR analysis with primers flanking loxP sites showed effective deletion of LTbR promoter and the first two exons ( Figure S1C, D). Combined, the mRNA and genomic DNA analyses demonstrate an effective inactivation of LTbR in adult mice following TAM treatment.

LTbR Signaling in Adulthood Is Required for the Maintenance of LNs, PP and ILFs
LTbR -/mice fail to develop LNs and PPs due to lack of LTbR signaling in embryogenesis (4,28). We found that despite normal development of LNs, iLTbR D/D mice had smaller LNs compared to controls ( Figure 1A). Consistent with the smaller LN size, total cell numbers in LNs of iLTbR D/D mice were reduced ( Figure 1B).
These results suggest that LTbR signaling in adulthood contributes to the migration of naïve lymphocytes to the lymph nodes. Flow cytometry analysis confirmed reduction of B, CD4 + T and CD8 + T cells in the LNs of iLTbR D/D mice compared to WT mice ( Figures S2A, S3A, B). In MLNs, B and CD4 + T cell proportions were not changed while CD8 + T cells were reduced in iLTbR D/D mice ( Figure S3A). ILN cell proportions in iLTbR D/D mice were differentially affected as B cells were decreased, CD4 + T cells were increased and CD8 + T cells populations were comparable to WT mice ( Figure S3B). To define the impact of genetic LTbR inactivation on LN microarchitecture in adulthood, we next analyzed B cell follicles, follicular dendritic cells (FDCs) and sinus macrophages in the MLN of iLTbR D/D mice ( Figures  1C-E). FDCs play an important role in germinal center development by presenting antigen to B cells (46,47). In iLTbR D/D mice, B cell areas were disorganized and FDCs were significantly diminished ( Figures 1C, E). Within LNs, sinus macrophages are divided into two subgroups, CD169 + SIGNR1subcapsular sinus (SCS) macrophages and CD169 lo SIGNR1 + medullary sinus macrophages (48). Although both populations of sinus macrophages can capture antigen, their functions differ and are dependent on their location within the LN (48,49). We found a reduced number of CD169 + SCS macrophages in MLN of iLTbR D/D mice. Interestingly, these macrophages co-expressed SIGNR1, which is normally expressed by medullary sinus macrophages ( Figures 1D, E).
LTbR signaling controls the migration of B and T cells to lymphoid tissues via production of homeostatic CXCL13, CCL19, CCL21 chemokines and adhesion molecules (31,50). Consistently, the expression of CCL21 was reduced in the MLN of iLTbR D/D mice ( Figure 1F). Although not significant, CCL19 demonstrated a declining trend ( Figure 1F). CXCL13 expression was slightly reduced, although not significantly ( Figure 1F). Together, these results demonstrate that LTbR signaling in adulthood is required for LN cellularity, production of homeostatic chemokines, as well as maintenance of FDCs and SCS macrophages.
We next sought to characterize the effects of inducible genetic LTbR inactivation in adulthood on the microarchitecture of PPs. Compared to WT mice, iLTbR D/D mice displayed fewer and smaller PPs ( Figure 1G and Figure S4A). The smaller number and size of PPs in iLTbR D/D mice corresponded to a reduction in the number of cells isolated from PPs though the proportion of B, CD4 + T and CD8 + T cells was not changed compared to controls ( Figure 1H and Figure S3C). Immunohistochemical analysis of PP structure revealed fewer FDCs and poorly defined B and T cell areas ( Figures 1I, J) suggesting that active LTbR signaling in adulthood is required for the maintenance of PPs.
In contrast to PPs which develop during embryogenesis and require LTbR signaling in stromal cells (

LTbR Signaling Is Required to Maintain Spleen Microarchitecture in Adulthood
The spleen is characterized by its distinctive white pulp (WP) and red pulp (RP) areas (46,47). Both WP and RP areas are permeated by cells which play specialized roles in splenic homeostasis, architecture, and in immune protection (46,47). In the marginal zone (MZ), which separates the WP and RP, antigens are captured for presentation to maturing B and T cells within the WP (46,47). LTbR -/mice are known to have disorganized spleen structure evidenced by lack of MZ and clearly defined RP and WP areas [ Figure S4A, and (28)]. Unlike LTbR -/mice, iLTbR D/D mice had less disorganized spleens though their structure was notably impaired when compared to WT mice ( Figure S4A). Although the WP size was similar in iLTbR D/D mice and control mice, the MZ was less defined ( Figure S4A). LTbR -/mice are known to develop splenomegaly due to increased numbers of splenocytes and neutrophils (30,54). The weight of spleens from iLTbR D/D mice was comparable to WT mice ( Figure S4C) and the proportion of B and CD4 + T cells was not changed in comparison to WT or LTbR -/mice ( Figure S3D). CD8 + T cells were reduced in LTbR -/but not in iLTbR D/D mice when compared to WT mice. These results suggest that genetic inactivation of LTbR in adulthood does not induce the development of splenomegaly.
To assess the spleen microarchitecture of iLTbR D/D mice, we stained frozen spleen sections with CD21/CD35, CD3, B220, and ER-TR-7 antibodies (Figures 2A, S4D). Notably, organized B cell follicles were absent, while B cells formed a ring-like structures around T cells areas (Figure 2A). Further, FDC numbers were strongly reduced in spleen of iLTbR D/D mice (Figures 2A, B, S4D). ER-TR7 + fibroblastic reticular networks were less organized in iLTbR D/D mice compared to WT mice ( Figure 2A). However, the degree of spleen disruption in iLTbR D/D mice was not as strong as in LTbR -/mice which displayed reduced WP area, mixed T and B cell areas, complete loss of FDCs and the marginal zone ( Figure S4A and data not shown). Consistent with reduced FDCs and less organized stromal components, iLTbR D/D mice had reduced expression of homeostatic chemokines CXCL13, CCL21, and CCL19 ( Figure 2C).
To define the role of LTbR signaling for the maintenance of the marginal zone in adulthood, we stained spleens of iLTbR D/D mice with CD169, MAdCAM-1 + and SIGNR1 antibodies ( Figure 2D). The number of MAdCAM-1 + and CD169 + cells was dramatically reduced, whereas the number of SIGNR1 + macrophages was less affected in spleen of iLTbR D/D mice ( Figures 2D, E and Figure S4D). These results suggest that LTbR signaling in adulthood is required for the maintenance of spleen microarchitecture.

iLTbR D/D Mice Do Not Develop Autoimmunity and Systemic Inflammation
Splenomegaly is often associated with systemic inflammation and autoimmunity (30,55). To define the role of LTbR signaling in adulthood on autoimmunity and systemic inflammation, we analyzed the liver, lung, and thymus of iLTbR D/D mice for tissue abnormalities or key autoimmune indicators ( Figure 3). Histological analysis of livers showed considerable perivascular lymphocytic infiltration in LTbR -/but not in iLTbR D/D mice relative to that of age-matched WT controls ( Figure 3A). The lungs of LTbR -/mice also displayed an enhanced pattern of perivascular inflammation that was not detected in iLTbR D/D mice ( Figure 3A). These results suggest that formation of lymphocytic infiltrates observed in adult LTbR -/mice depends on LTbR inactivation during embryogenesis or neonate stages.
We next analyzed thymus organization, as impaired negative selection of lymphocytes due to defects in thymus structure was attributed to autoimmunity in LTbR -/mice (29,56,57). Several studies demonstrated the role of LTbR signaling in controlling thymic epithelial cells, endothelial cells and stromal cells (29,33,34,(56)(57)(58). Our analysis revealed that thymus structure of iLTbR D/D mice was not impaired ( Figure 3A). In contrast, thymic medulla of LTbR -/mice was smaller and disorganized ( Figure 3A). Thymus weight of iLTbR D/D and LTbR -/mice were normal ( Figure 3B). Consistent with the lack of thymus impairment in iLTbR D/D mice, expression levels of Aire, a transcriptional regulator of autoimmunity, were not reduced ( Figure 3C). The proportion of B, CD8 + T, and CD4 + CD8 + T cells were also not affected ( Figures S2C, S3F). However, an increase in the proportion of thymic CD4 + T cells compared to WT mice was observed in iLTbR D/D and LTbR -/mice was observed compared to WT mice ( Figure S3F). These results are consistent with previous studies conducted in LTbR -/mice (33,58). Collectively, this data suggest that inducible inactivation of LTbR in adult mice does not impair thymus organization. Furthermore, serological analysis of dsDNA autoantibody levels, a hallmark of autoimmune disease, demonstrated a lack of autoantibodies in iLTbR D/D mice compared to control MRL mice ( Figure 3D). Additionally, we did not find changes in the proportions of B, CD8 + T, and CD4 + T cells in the blood of iLTbR D/D and LTbR -/mice compared to WT mice ( Figure S3E). Thus, our results suggest that inactivation of LTbR signaling in adulthood does not result in autoimmunity and systemic inflammation.

LTbR Signaling Controls Homeostasis of Neutrophils, NK, and iNKT Cells in Adult Mice
Defects in homeostasis of neutrophils, NK, and NKT cells have been described in LTbR -/mice (13-15, 30, 59-61). However, the role of LTbR signaling in homeostasis of these cells in adult mice is not clear, as defects in development of lymphoid organs and systemic inflammation in LTbR -/mice can affect these cells. To clarify this, we analyzed neutrophils, NK, and iNKT cells in iLTbR D/D mice by flow cytometry (Figures 4 and S2). A previous study detected increased numbers of neutrophils in the spleens of adult LTbR -/mice which was dependent on microbiota (30). Increased neutrophil numbers were also detected in the blood of adult BALB/C mice treated weekly with LTbR-Ig for four weeks (31). Surprisingly, we found reduced frequencies of neutrophils in the spleens of iLTbR D/D mice, but not in blood ( Figure 4A), but not in blood ( Figure 4A) compared to LTbR fl/fl and LTbR -/mice. As recruitment of neutrophils is mainly dependent on CXCL2 and CXCL1 chemokines (62, 63), we measured expression of these chemokines in the spleens of iLTbR D/D mice. Expression of CXCL2 was reduced in iLTbR D/D mice compared to control or LTbR -/mice ( Figure 4B). However, expression of CXCL1 was not changed in iLTbR D/D mice compared to controls though it was significantly lower than levels observed in LTbR -/mice ( Figure 4B). These results suggest that active LTbR signaling in adulthood is required for recruitment of neutrophils to the spleen.
Consistent with previous studies in LTbR -/mice and mice treated with LTbR-Ig (13, 61), our analysis revealed reduced frequencies of NK cells in the spleens and livers of iLTbR D/D and LTbR -/mice compared to LTbR fl/fl mice ( Figure 4C). No difference was observed in spleen NK cell frequencies between LTbR -/and iLTbR D/D mice, although there was an increase in liver NK cell frequencies from iLTbR D/D mice compared to LTbR -/mice ( Figure 4C). NK cell frequencies were not significantly reduced in the thymuses of iLTbR D/D mice ( Figure 4C). Additional analysis of iNKT cells (CD1dtet + TCRb + NK1.1 + ) revealed no difference in iNKT cell proportions in the thymuses of iLTbR D/D mice compared to LTbR fl/fl control mice though iNKT cell frequencies were reduced in LTbR -/mice compared to controls ( Figure 4D). In contrast, iNKT cell frequencies were reduced in the livers and spleens of LTbR -/and iLTbR D/D mice compared to WT mice ( Figure 4D). These results suggest that LTbR signaling supports the migration of iNKT cells to the liver and spleen in adulthood but is dispensable for iNKT cell development in the thymus.

LTbR Signaling Is Not Required for the Maintenance of IgA in Adulthood
LTbR -/mice display severely reduced levels of IgA in the gut and blood (16). Since LTbR -/mice have developmental defects in the formation of gut-associated lymphoid organs, it is not clear whether LTbR signaling is required for IgA production in adulthood. To define the impact of LTbR inactivation in adulthood for IgA production, we measured fecal and serum IgA levels in adult iLTbR D/D mice four months after TAM treatment ( Figure 5). We chose a four month time period for the analysis of non-specific antibody responses to allow sufficient time for de novo generation of IgA producing plasma cells after inactivation of LTbR. Surprisingly, IgA levels in both serum and feces of iLTbR D/D mice were comparable to those of WT mice, in contrast to reduced IgA levels in LTbR -/mice ( Figure 5A). IgG and IgM levels were not affected in both iLTbR D/D and LTbR -/mice as expected ( Figure 5A). Accordingly, we did not find difference in IgA producing cells between iLTbR D/D mice and LTbR fl/fl control mice ( Figure 5B and Figure S5). Examination of IgG producing cells revealed a trend toward decreased frequency of IgG + plasmablasts (PB) as well IgG + GC B cells ( Figure 5B center panel) in iLTbR D/D mice. IgM + plasma cells (PC), PB, and B cells were slightly reduced in iLTbR D/D mice, albeit not significantly ( Figure 5B). Thus, these results suggest that inactivation of LTbR signaling in adulthood does not result in impaired maintenance of IgA production.

LTbR Signaling Is Required for De Novo IgA Production and Protection Against Mucosal Bacterial Pathogen
To further define the impact of LTbR signaling in mucosal immune responses, we orally infected iLTbR D/D mice with C. rodentium ( Figure 6A). C. rodentium is a murine pathogen Significance was determined using one-way ANOVA with Tukey's correction. Not significant (ns, p>0.05).

Shou et al. Inactivation of LTbR in Adulthood
Frontiers in Immunology | www.frontiersin.org July 2021 | Volume 12 | Article 712632 which mimics the diarrheagenic disease caused by the human pathogens enteropathogenic Escherichia coli (EPEC) and enterohemorrhagic E coli (EHEC) (44,64). LTbR -/mice are highly sensitive to C. rodentium infection, which results in 100% mortality in these mice (65,66). iLTbR D/D mice showed an increased susceptibility to C. rodentium infection, as they exhibited increased body weight loss ( Figure 6B), colon shortening ( Figure 6G), increased spleen weight ( Figure 6G), and increased bacterial titers in their blood and colons ( Figures  6D-F) compared to LTbR fl/fl control mice. However, 90% of iLTbR D/D mice survived infection, compared to 100% mortality in LTbR -/mice ( Figure 6C). Multiple immune abnormalities and lack of gut-associated lymphoid tissues in LTbR -/mice could be responsible for their increased susceptibility, compared to iLTbR D/D mice. Rapid production of IL-22 after C. rodentium infection is one of the major mechanisms of protection against this pathogen (64). Impaired IL-22 production in the colon of LTbR -/mice leads to increased susceptibility of these mice to C.
rodentium (67,68). Expression of IL-22 and IL-22-dependent antibacterial protein RegIIIb were significantly reduced in colon of iLTbR D/D mice compared to LTbR fl/fl control mice, whereas RegIIIg was not notably affected ( Figure 6H). We next thought to define the impact of LTbR inactivation on generation of C. rodentium-specific antibody responses ( Figures  6I-K). We found reduced fecal levels of C. rodentium-specific IgA in iLTbR D/D mice, whereas serum IgG and IgM were not changed compared to LTbR fl/fl mice ( Figures 6I-K). These results suggest that disruption of LTbR signaling in adulthood increases susceptibility to C. rodentium infection and blocks generation of pathogen-specific IgA responses.

DISCUSSION
LTbR is known as key regulator of lymphoid organogenesis. Nearly three decades of work has expanded the role of LTbR Representative flow cytometry plots of CD1d tet + TCRb + iNKT cells in the liver. Graphs show % of iNKT cell frequencies in the CD45 + populations of liver, thymus, and spleen. For panels (C, D), data shown is from two of three independent experiments. Significance was determined using one-way ANOVA with Tukey's correction for multiple comparisons. All gating strategies are defined in Figure S2. Data shown are means ± SEM. Bars show the mean, symbols represent individual mice. Not significant (ns, p > 0.05), *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Shou et al. Inactivation of LTbR in Adulthood
Frontiers in Immunology | www.frontiersin.org July 2021 | Volume 12 | Article 712632 beyond the scope of lymphoid organ development and maintenance to include roles the in regulation of mucosal repair, cancer, inflammation and autoimmunity (6,25,41,(66)(67)(68)(69). LTbR -/mice provided a useful animal model to address the role of LTbR signaling in vivo. However, to better understand the role of LTbR signaling in the maintenance of lymphoid organs and immune homeostasis in adulthood, we need to uncouple these functions from the developmental defects and systemic inflammation observed in LTbR -/mice. In this study we redefined the role of LTbR signaling in adulthood using mice with inducible genetic inactivation of LTbR. Though a potential issue with all conditional gene targeted strategies is efficacy of gene deletion, we demonstrated strong gene deletion in several tissues including colon, liver, kidney, and LNs. Our results suggest that LTbR signaling in adulthood is critical for the maintenance of spleen, lymph nodes, and gut-associated lymphoid organs; homeostasis of neutrophils, NK and iNKT cells; and specific IgA antibody responses against mucosal pathogen Citrobacter rodentium. Furthermore, our results indicate that adult stage LTbR signaling is dispensable for the maintenance of polyclonal IgA responses and autoimmunity. Analysis of LNs in iLTbR D/D mice helped us clarify the role of LTbR signaling during adulthood in regulating the cellularity and microarchitecture of LNs. Our results demonstrate that active LTbR signaling is required for the migration of B, CD4 + T and CD8 + T cells to the LNs. Reduced cellularity of LNs was also observed in a previous study using long-term 28 day biochemical inhibition of LTb and LIGHT signaling with LTbR-Ig fusion protein in BALB/C mice (31). In contrast, other studies showed that short-term LTbR-Ig administration in adult mice or in neonate mice did not affect LN size (18,70), suggesting that long-term and short-term inhibition of LTbR signaling have different impacts on LN cellularity. Reduced LN cellularity was consistent with reduced expression of CXCL13, CCL19, and CCL21 chemokines. In fact, ablation of LTbR specifically in CXCL13 expressing cells (using CXCL13-Cre) reduced cellularity of LNs and ablated formation of B cell follicles without affecting the development of LNs during embryogenesis (71). Ablation of LTbR in endothelial cells resulted in impaired formation of LNs during embryogenesis and reduced cellularity in adult mice (71,72). In contrast, targeting LTbR in CCL19-expressing cells (CCL19-Cre) did not affect LN development and structure, but impaired resistance to viral infection (73). Thus, LTbR signaling in distinct cell populations is responsible for the development, cellularity, and microarchitecture of LNs. It remains to be determined whether impaired microarchitecture of LNs in iLTbR D/D mice would result in increased susceptibility to pathogens.
Further analysis of LN microarchitecture in iLTbR D/D mice revealed reduction of FDCs, disruption of B cell follicles and disorganization of marginal sinus macrophages. FDC survival is known to be highly dependent on LTbR signaling, as LTbR-Ig treatment resulted in rapid elimination of FDCs (within 24h) in the LNs and spleen (74). In contrast, prolonged LTbR inhibition (for several weeks) is required for disruption of B cell follicles, sinus macrophages in lymph nodes and the marginal zone of the spleen (49,70,75). Consistent with these previous observations, we found a reduction of CD169 + SCS macrophages in the MLN of iLTbR D/D mice. However, CD169 + SCS macrophages coexpressed SIGNR1, which is normally expressed by medulla macrophages (48). Similar co-expression of CD169 and SIGNR1 by SCS macrophages has been observed in mice treated with LTbR-Ig (75). It remains to be determined CD138 -CD19 + GL7 -B cells, CD138 -CD19 + GL-7 + germinal center (GC) B cells, CD138 + CD19 + plasmablasts (PB), CD138 + CD19plasma cells (PC). Graphs depict % of cells in CD45 + gate. N=3 for all groups. Significance was determined by unpaired t-test. Data shown is from one representative experiment out of two. Gating strategy shown in Figure S5. Data shown are means ± SEM. Not significant (ns, p > 0.05), *p < 0.05, **p < 0.01.
whether CD169 + SCS macrophages were replaced with CD169 +/ lo SIGNR1 + medullary sinus macrophages or converted their phenotype in iLTbR D/D mice. LT provided by B cells was shown to be responsible for the maintenance of SCS macrophages (22,75). Conversely, the cellular source for the maintenance of FDCs in LNs remains unclear, as elimination of LT from B cells only resulted in minimal disruption of LN FDCs (22). It is possible that cooperation of LT signaling by B cells, T cells, and ILCs is necessary to support the maintenance of FDCs in LNs.
Our analysis of PPs and ILFs in iLTbR D/D mice is in agreement with previous studies which used long-term biochemical inhibition of LTbR (LTbR-Ig) in adult mice (53). Specifically, LTbR blockade was associated with flattened PP appearance and reduction in the number of macroscopically visible PPs (76) and ILFs (53). Additionally, we found that active LTbR signaling in adulthood is required for the formation of FDCs in PPs. The function of FDCs in PPs was previously associated with generation of IgA in the gut (77). However, our study did not find defects in polyclonal IgA production in iLTbR D/D mice. In contrast, specific IgA responses to enteric pathogen C. rodentium were reduced in iLTbR D/D mice.
We next analyzed the impact of inducible LTbR inactivation on spleen microarchitecture in adult mice. Our analysis revealed impaired formation of B cell follicles and reduced FDCs in the spleen. These defects were consistent with biochemical inhibition of LTbR using LTbR-Ig reagent (70,74), supporting the role of LTbR signaling in the maintenance of B cell follicles and FDCs in the spleen. The integrity of the marginal zone was also impaired in iLTbR D/D mice as evidenced by reduced MAdCAM-1 + sinus endothelial cells, CD169 + metallophilic macrophages, and SIGNR1 + macrophages. In contrast, MZ structure is completely ablated in LTbR -/mice (28) as formation of MAdCAM-1 + marginal sinus is fixed in the early postnatal period and cannot be restored in adult mice by transplantation of bone marrow from WT mice (19). Our results are consistent with biochemical inhibition of LTbR signaling in adult mice, which demonstrated various degrees of disruption of MAdCAM-1 + sinus lining cells, marginal zone CD169 + metallophilic macrophages and SIGNR1 + marginal zone macrophages depending on the dose and duration of treatment (19,70). B cells are the primary source of LT required for the maintenance of FDCs and support of marginal zone integrity (22), with additional help from LT-expressing T cells (78). Although LT expressed by ILC3 contributes to early postnatal development of white pulp (79,80), its role for the maintenance of spleen microarchitecture in adult mice is less clear. Inactivation of LT on ILC3s did not result in defects in the formation of FDCs and B cell follicles in adult mice, whereas differentiation of cDC2 in spleen was impaired (81). In contrast to homeostatic conditions, LT from ILC3s can contribute to the restoration of spleen microarchitecture after viral-induced tissue damage (82). Likewise, LIGHT -/mice do not display defects in organization of spleen and lymph nodes in homeostatic conditions (83). However, LIGHT signaling via LTbR can restore FDCs and B cell follicles in LT-deficient mice (23) as well as support LN remodeling during inflammatory response (24). Thus, we propose that the requirements for LTbR signaling in the maintenance of lymphoid organs during homeostatic conditions and during inflammation can be different and provided by distinct LT and LIGHT producing cells. iLTbR D/D mice can be used as a robust model to test this hypothesis in follow up studies. LTbR -/mice develop an autoimmune phenotype such as splenomegaly, production of autoantibodies and lymphocytic infiltrations to non-lymphoid organs (28-30, 57, 84, 85). The mechanistic explanation behind an autoimmune phenotype in LTbR-deficient mice remains a contentious topic. Since an autoimmune phenotype in LTbR -/mice can be due to developmental defects in lymphoid organs and intestinal microbiota composition (28,30,32,36,56,57), the role of LTbR signaling in autoimmunity and systemic inflammation in adulthood remains controversial. We assessed key parameters which previously have been associated with systemic inflammation and autoimmunity in LTbR -/mice. Our results demonstrate that inactivation of LTbR signaling during adulthood in iLTbR D/D mice did not result in splenomegaly, infiltration of immune cells to non-lymphoid organs, impairment of thymus microarchitecture, or reduction of thymic Aire expression. We also did not find abnormal production of dsDNA autoantibodies in iLTbR D/D mice. Thus, we conclude that inhibition of LTbR signaling in adulthood does not lead to systemic inflammation and autoimmunity. These results are important to consider in translational studies, as agonists and antagonists of LTbR signaling are currently being tested as potential therapies in autoimmune diseases and cancer (5,6,69,86,87).
Our analysis of innate immune cell populations revealed a reduced number of neutrophils in the spleens of iLTbR D/D mice. In contrast, LTbR -/mice were reported to have increased neutrophil numbers in the spleen which was dependent on microbiota as antibiotic treatment of LTbR -/mice reduced neutrophil numbers (30). Reduced frequencies of neutrophils in iLTbR D/D mice was not due to differences in microbiota, as iLTbR D/D mice and littermate LTbR fl/fl control mice were cohoused in our experiments. Reduced neutrophil frequencies in the spleen of iLTbR D/D mice was associated with impaired expression of key neutrophil-recruiting chemokine, CXCL2. In line with these results, previous studies suggested a role for LTbR signaling in control of CXCL1 and CXCL2 expression in response to mucosal bacterial pathogens (66) and suppression of metabolic activation via neutrophil-intrinsic LTbR signaling during colitis (59).
Previous studies using LTbR -/and LTa -/mice or WT mice treated with LTbR-Ig demonstrated loss of NK cell populations in the spleen, lung, blood and bone marrow as well as impairment of NK cell anti-tumor activities (13,61). LT expression on RORgt + ILC3s was suggested to be critical for NK cell development via interaction with bone marrow stromal cells (60). Our results in iLTbR D/D mice are consistent with these studies and support the role of LTbR signaling in adulthood for the homeostasis of NK cells.
LTbR -/mice have impaired NKT development due to developmental defects in thymic stroma (14,15). However, the role of adult LTbR signaling in NKT cell development is unclear as administration of LTbR-Ig in adult WT mice did not result in impaired development of NKT cells which is in contrast to LTbR-Ig blockade during embryogenesis and neonate phases (15,88). Consistent with these studies, we did not find defects in thymus CD1d-tet + TCRb + NK1.1 + iNKT cell populations in iLTbR D/D mice indicating that LTbR signaling is dispensable for development of iNKT cells in adulthood. Interestingly, iNKT cell frequencies were reduced in the livers and spleens of iLTbR D/D mice. The mechanism of LTbR-dependent iNKT cell recruitment to the liver remains to be determined. Impaired IgA production in LTbR -/mice was initially attributed to stromal cell populations that provide signals to B cells for homing and antibody class switching to IgA (16). However, due to lack of gut-associated lymphoid tissues, such as PPs, MLN, ILFs and cryptopatches in LTbR -/mice, the contribution of LTbR signaling in adulthood remained undefined. Follow up studies implicated the role of ILFs in gut IgA production that were dependent on the interaction of LT expressing ILC3s with LTbR on stromal cells (89,90). However, genetic inactivation of LTb on ILC3s did not result in impaired IgA production in spite of impaired formation of ILFs and reduced INOS production by DCs in MLN (17). FDCs have been suggested to contribute to IgA generation in PPs (77). Additionally, a recent study indicated the role of intrinsic LTbR signaling in PP DCs for IgA generation in response to the microbiota (91). Our results suggest that LTbR signaling in adulthood is dispensable for the maintenance of fecal or serum IgA despite reduced ILF numbers and impaired FDCs in PPs. Consistently, we found very efficient LTbR deletion in the colon. These results suggest that other compensatory pathways contribute to IgA production when LTbR signaling is inhibited in adult mice. We also do not exclude the possibility of long-lived plasma cells that could survive for several months after LTbR inactivation and contribute to IgA production in iLTbR D/D mice. The non-specific IgA antibodies we detected in iLTbR D/D mice could be derived from such long-lived cells that underwent antibody class switching prior to inactivation of LTbR. The impact of LTbR signaling on IgA production in humans remains to be determined, as human and mouse IgA systems have distinct anatomical and functional differences, including presence of two IgA1 and IgA2 isotypes, lack of cryptopatches and TLR4 expression by B cells in humans (92).
To define the role of LTbR signaling in generation of specific IgA in response to mucosal bacterial infection, we infected iLTbR D/D mice with C. rodentium. While iLTbR D/D mice showed an increased susceptibility to C. rodentium infection, demonstrated by increased weight loss and increased bacterial load, this phenotype was less pronounced compared to LTbR -/mice. Importantly, we found reduced C. rodentium-specific IgA levels in feces of iLTbR D/D mice, whereas serum IgG and IgM were not changed. These results suggest that LTbR signaling in adulthood contributes to generation of pathogen-specific IgA. Impaired structure of PPs and MLN as well as reduced numbers of ILFs could contribute to the defect in C. rodentium specific IgA production, consistent with the role of these tissues in response to bacterial antigens (77,91). Interestingly, a recent study demonstrated that treatment of adult mice with LTbR-Ig did not result in impaired rotavirus-specific IgA production, whereas LTbR blockade during embryogenesis and neonate period impaired rotavirus-specific IgA production (18). These and our own results suggest distinct LTbR requirements in adulthood for production of IgA in response to bacterial and viral pathogens.
In summary, our study redefined the role of LTbR signaling in adulthood for organization of lymphoid organs, autoimmunity, homeostasis of innate immune cells and IgA production. Our results suggest that inducible genetic LTbR inactivation during adulthood results in impaired organization of LNs and spleen; homeostasis of neutrophils, NK, and iNKT cells; and generation of mucosal pathogen-specific IgA responses but does not result in autoimmunity. Mice with inducible genetic inactivation of LTbR provide a robust preclinical model to evaluate the impact of LTbR agonists and inhibitors in disease treatment.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The animal study was reviewed and approved by Institutional Animal Care and Use Committee of University of Texas Health Science Center San Antonio.