Lrba participates in the differentiation of IgA+ B lymphocytes through TGFβR signaling

Introduction Lrba is a cytoplasmic protein involved in vesicular trafficking. Lrba-deficient (Lrba-/-) mice exhibit substantially higher levels of IgA in both serum and feces than wild-type (WT) mice. Transforming growth factor β1 (TGFβ1) and its receptors (TGFβR I and II) is essential for differentiating IgA+ B cells. Furthermore, increased IgA production suggests a potential connection between Lrba and the TGFβR signaling pathway in IgA production. However, the specific function of Lrba in B cell biology remains unknown. Aim Given the increased IgA levels in Lrba-/- mice, the goal in this work was to explore the lymph organs where the switch to IgA occurs, and if TGFβR function is affected. Methods Non-immunized Lrba-/- mice were compared with Lrba+/+ mice. IgA levels in the serum and feces, as well as during peripheral B cell development, were determined. IgA+ B cells and plasma cells were assessed in the small intestine and secondary lymphoid organs, such as the spleen, mesenteric lymph nodes, and Peyer’s patches. The TGFβR signaling pathway was evaluated by determining the expression of TGFβR on B cells. Additionally, SMAD2 phosphorylation was measured under basal conditions and in response to recombinant TGFβ. Finally, confocal microscopy was performed to investigate a possible interaction between Lrba and TGFβR in B cells. Results Lrba-/- mice exhibited significantly higher levels of circulating IgA, IgA+ B, and plasma cells than in peripheral lymphoid organs those in WT mice. TGFβR expression on the membrane of B cells was similar in both Lrba-/- and Lrba+/+ mice. However, intracellular TGFβR expression was reduced in Lrba-/- mice. SMAD2 phosphorylation showed increased levels under basal conditions; stimulation with recombinant TGFβ elicited a poorer response than in that in Lrba+/+ B cells. Finally, we found that Lrba colocalizes with TGFβR in B cells. Conclusion Lrba is essential in controlling TGFβR signaling, subsequently regulating SMAD2 phosphorylation on B cells. This mechanism may explain the increased differentiation of IgA+ B cells and production of IgA-producing plasma cells.

Introduction: Lrba is a cytoplasmic protein involved in vesicular trafficking.Lrbadeficient (Lrba-/-) mice exhibit substantially higher levels of IgA in both serum and feces than wild-type (WT) mice.Transforming growth factor b1 (TGFb1) and its receptors (TGFbR I and II) is essential for differentiating IgA+ B cells.Furthermore, increased IgA production suggests a potential connection between Lrba and the TGFbR signaling pathway in IgA production.However, the specific function of Lrba in B cell biology remains unknown.
Aim: Given the increased IgA levels in Lrba-/mice, the goal in this work was to explore the lymph organs where the switch to IgA occurs, and if TGFbR function is affected.
Methods: Non-immunized Lrba-/mice were compared with Lrba+/+ mice.IgA levels in the serum and feces, as well as during peripheral B cell development, were determined.IgA+ B cells and plasma cells were assessed in the small intestine and secondary lymphoid organs, such as the spleen, mesenteric lymph nodes, and Peyer's patches.The TGFbR signaling pathway was evaluated by determining the expression of TGFbR on B cells.Additionally, SMAD2 phosphorylation was measured under basal conditions and in response to recombinant TGFb.Finally, confocal microscopy was performed to investigate a possible interaction between Lrba and TGFbR in B cells.
Results: Lrba-/mice exhibited significantly higher levels of circulating IgA, IgA+ B, and plasma cells than in peripheral lymphoid organs those in WT mice.TGFbR expression on the membrane of B cells was similar in both Lrba-/and Lrba+/+

Introduction
Lipopolysaccharide (LPS)-responsive beige-like anchor (LRBA) is a cytoplasmic protein belonging to the beige and Chediak-Higashi syndrome (BEACH) family of proteins associated with vesicular trafficking processes (1).This deficiency affects regulatory T cells (Treg) function.The human phenotype was first described in 2012; biallelic mutations in LRBA are associated with immunodeficiency characterized by autoimmunity, recurrent infections, defects in B cell activation, decreased class-switched memory B cells, and low IgG and IgA levels (2,3).LRBA is associated with the endomembrane system, including the endoplasmic reticulum, endocytic vesicles, lysosomes, and Golgi apparatus, suggesting its participation in vesicular trafficking (4).Additionally, LRBA has been shown to interact with cytotoxic T-lymphocyte-associated protein 4 (CTLA4) in endosomes; such interactions are essential for recycling this protein, and therefore, LRBA-deficient Treg cells show reduced CTLA4 expression and increased degradation (5).
The LRBA protein is 90% homologous to the murine sequence.Lrba is ubiquitously expressed and is induced two-to four-fold in immune cells after stimulation with LPS (4).Lrba-/-murine deficiency also exhibited reduced CTLA4 expression in Tregs.In B cells, Lrba-/mice showed increased IgA levels in the serum and small intestine (6).
The class switch recombination to IgA can be carried out by two pathways: T-cell dependent or independent.In the T cell-dependent pathway, activated T cells interact with B cells through the CD40 ligand, (CD40L), and CD40 receptors, inducing the expression of activation-induced cytidine deaminase (AID), leading to class switch recombination (7).Cytokine secretion by activated T-cells guides the switch to specific immunoglobulins.In the case of IgA induction, B lymphocytes require transforming growth factor b1 (TGFb1), a cytokine secreted by various cell types, including several subsets of CD4+ T cells (8).
In the case of T-independent IgA induction, antigens interact with B cells and multiple innate immune cells through either the Bcell receptor (BCR) or Toll-like receptors (9,10).Additionally, dendritic cells release B-cell-activating factor (BAFF; also known as BLyS) and proliferation-inducing ligand (APRIL).APRIL induces AID expression, while TGFb1 is required to direct the isotype switch to IgA (11)(12)(13)(14).
TGFb1 is a pleiotropic cytokine belonging to the TGFb superfamily.It is derived from the proteolytic cleavage of latencyassociated peptide (LAP) with subsequent dimerization (15).TGFb receptor (TGFbR), in B cells is composed of two of each TGFbRII and TGFbRI subunit, both with kinase activity.Upon TGFb1 interaction with TGFbRII, phosphorylation of TGFbRI occurs, leading to its activation and subsequent phosphorylation of receptor-regulated SMAD proteins, including SMAD2 and SMAD3.The phosphorylation of these proteins causes their association with SMAD4 in the cytoplasm.These complexes are then translocated to the nucleus where they bind to SMAD-binding elements.One target of these SMAD heterodimers is the constant a heavy chain (IgHa) (7,8).
The relevance of TGFbR in the class switch to IgA has been demonstrated in conditional murine-deficient models.Specifically, B cells deficient in TGFbRII showed reduced IgA+ B cells in both spleen and Peyer patches (PP), along with low levels of IgA in the serum (16).
Given the increased IgA levels in Lrba-/-mice, we aim to explore the lymph organs where the switch to IgA occurs, and if TGFbR function is affected.Here, we explored the presence of IgA+ B and plasma cells in secondary lymphoid organs, as well as TGFbR expression and SMAD2 phosphorylation under both basal and activation conditions; finally, the possible interaction between Lrba and TGFbR was explored by co-localization experiments, the data obtained suggest a direct role for Lrba in controlling TGFbR signaling.

Experimental animals
Lrba-deficient mice were (6,17) engineered at Taconic Artemis (Köln, Germany) (official allele designation: Lrbatm1.1KiliMGI: 5796558; laboratory designation: Lrba2) by constitutive deletion of the coding exon 3.This deletion introduced a frameshift mutation that was predicted to produce a truncated protein that was prone to degradation.Consequently, a shift in the reading frame occurred after amino acid 149 (5% of the coding sequence), rendering the expression of Lrba undetectable by Western Blot analysis.
The mice were bred and maintained in a C57BL/6 N gene pool under pathogen-free conditions at the animal facilities of the Research Centre and Advanced Studies (CINVESTAV, Mexico City, Mexico).Genotyping was performed using PCR amplification of genomic DNA extracted from splenocytes or tail samples using the following primer combinations: a sense primer for exon 3 (Ex3F: 5′-GAAAGTTGACAGTATGATTGCAGG-3′) paired with a wild-type-specific reverse primer in exon 4 (Ex4R: 5′-CATTGTCCTTTATCTCCTTGAA-3′), or a combination of Ex3F and a reverse primer for intron 4 (Int4R: 5′-CTAAGGAGGATG GCTCTAACC-3').
This study was reviewed and approved by the CINVESTAV Ethics Committee and all the mice were maintained according to the Institutional Animal Guidelines for Animal Care and Experimentation (protocol number: 145-15, UPEAL-CINVESTAV-IPN).

Immunoglobulin levels quantification by ELISA
Ninety-six-well plates were coated with 5 ng/well of the capture antibody, anti-mouse IgG, IgA, or anti-IgM (SouthernBiotech, Birmingham, AL, USA) and incubated overnight at 4°C.The plates were blocked with 1% milk in 1X PBS containing 0.05% Tween for 2 h.Samples, including sera or fecal supernatants diluted at a 1:100 ratio in PBS, were added to the wells and incubated for 1 h at 37°C.Subsequently, biotinylated antibodies specific for either IgM (Southern Biotech, 102008), IgG (Southern Biotech, 103005), or IgA (Southern Biotech, 104008) were diluted at 1:1000 and added to the plates, followed by incubation for 1 h at 37°C.Finally, streptavidin was added and coupled with horseradish peroxidase (HRP; Abcam, Cambridge, UK, ab7403) diluted at 1:5000.Tetramethylbenzidine was added, and the reaction was stopped with 0.2 M phosphoric.The absorbance was measured at 450 nm using an ELISA reader (Microplate, Sunrise ™ ).

Splenic B cell subpopulations and B1 immunophenotyping
Splenocytes were stained with anti-CD19 BV421, anti-CD23 PerCPCy5.5,anti-CD21 PE, and anti-IgM PECy7 (all from Biolegend, San Diego, CA, USA) to detect Transitional 1, Transitional 2, Follicular, and Marginal Zone B cells.Briefly, B cells were incubated with a mix of antibodies for 30 minutes, washed, and fixed with 1% paraformaldehyde.For B1 cells, a peritoneal exudate was obtained, and cells were stained with anti-CD19 BV421, anti-CD21 PE, and anti-CD5 PerCPCy5.5 (all from Biolegend).After incubation, cells were washed and fixed with 1% paraformaldehyde.Data was acquired using a FACS LSRFortessa ™ (Beckton Dickinson, Franklin Lakes, NJ, USA).Data analysis was performed using FlowJo v10.10 (Beckton Dickinson).Percentages for each subpopulation were determined, and the total number was calculated using the total cell counts obtained from either the spleens or the peritoneal exudates.

Histological sections
Secondary organs, including the spleen, mesenteric lymph nodes, PP, and small intestine were obtained and placed in a tissue preservation medium (Leica, Wetzlar, Germany).They were then frozen in liquid nitrogen and stored at -70°C until use.Histological sections of 5-6 mm were mounted on slides coated with poly-L-lysine, fixed in cold acetone for 15 min, and stored at -20°C after drying at room temperature.

In situ immunofluorescent staining
Histological sections were rehydrated with 0.2% BSA in 1X PBS, blocked with Power Block Universal Blocking Reagent (BioGenex, CA, USA), and incubated for 1 h with anti-IgD (Becton Dickinson, 553438), anti-CD138 (Becton Dickinson, 553712), and biotinylated anti-IgA antibodies (eBioscience, San Diego, CA, USA, 13-5994-84).Subsequently, three washes with 0.2% BSA were performed, and anti-rat Alexa Fluor 594 secondary antibodies (Life Technologies, Carlsband, CA, USA, A21209) were added to detect anti-IgD and anti-CD138 antibodies, followed by incubation for 1 h at room temperature.Streptavidin coupled with Alexa Fluor 488 (Invitrogen, Waltham, MA, USA, 511223) was incubated with biotinylated primary antibodies for 30 min at room temperature.After staining, glass slides with histological sections were washed three times with 1× PBS, covered with coverslips, and mounted with 90% glycerol to keep the histological sections hydrated.Images were acquired using an Olympus BX51 microscope equipped with an Olympus U-CMAD3 camera and Olympus RFL-T epifluorescence lamp (Olympus, Tokyo, Japan) with 10× lenses.The analysis was performed using Image-Pro-Plus 7.0 (Media Cybernetics, Rockville, MD, USA) and Fiji ImageJ v2.7.0 software (18).
For intracellular detection, after fixation with 4% PFA for 10 min, the cells were washed with 1× PBS.Blocking with 10% goat serum was performed for 30 min, followed by washing with 1× PBS.The cells were then permeabilized with a BD Perm/Wash ™ (Becton Dickinson) for 10 min, followed by a final wash.Lymphocytes were then incubated with anti-TGFbRI PE and anti-TGFbRII for 1 h at 37°C.Subsequently, cells were washed with PBS and incubated with secondary antibodies against rabbit Alexa Fluor 488 in the case of the cell suspension with anti-TGFbRII.After the designated time, the cells were wash with 1X PBS was performed.

B-cell enrichment
Spleens were obtained from Lrba-/-and wild-type (WT) mice, disaggregated, and resuspended in 1x PBS.Subsequently, the MojoSort ™ Mouse CD19 Nanobeads kit (BioLegend, San Diego, CA, USA) was used to enrich B cells, following the manufacturer's instructions.

B-cell stimulation with TGFb1
Purified B cells were adjusted to 3x10 6 cells, resuspended in PBS, and placed in 1.5-mL tubes.Subsequently, 10 ng/mL of recombinant TGFb1 (BioLegend, 763102) was added, and the cells were incubated at 37°C for 5, 15, and 30 min.Unstimulated cells served as controls.After stimulation, the cells were processed for immunoblotting.
The proteins were transferred onto nitrocellulose membranes and blocked with 5% milk in TBS-Tween for 1 h at room temperature.Subsequently, the membranes were washed thrice for 10 min each with TBS-Tween, after which primary antibodies were added according to the specific assay requirements.

Immunoblots
Lrba and TGFbRI subunits were detected either in the immunoprecipitates, or total protein extracts were determined through immunoblotting, following standard procedures.Briefly, 20 mg of protein extracts from B cells were loaded onto a gradient gel 8-15% polyacrylamide (Bio-Rad, Hercules, CA, USA).A 10% polyacrylamide gel was used to detect the phosphorylated SMADs.The proteins were transferred to a nitrocellulose membrane for 1.5h at 100 V in standard Tris-glycine buffer with 20% methanol.The membranes were incubated overnight at 4°C with agitation using anti-Lrba, anti-TGFbRI (GeneTex, Zeeland, MI, USA, GTX134290), anti-TGFbRII, anti-phospho-SMAD2 (Cell Signaling, 138D4), antitotal SMAD 2/3 (Abcam, ab202445), and anti-Actin (Santa Cruz Biotechnology, Dallas, TX, USA).Subsequently, the membrane was washed, and a secondary antibody coupled to HRP was added and incubated for 1 h with agitation at room temperature.Finally, chemiluminescence detection was performed using Super Signal West Femto Maximum Sensitivity substrate on a ChemiDoc XRS (Bio-Rad), and images were acquired with a ChemiDoc ™ (Bio-Rad).

Immunofluorescence staining for confocal microscopy
The previously purified B cells were adjusted to 1x10 6 cells in 50 mL of 1× PBS and fixed with 4% PFA for 15 min.They were then washed with 1× PBS, followed by a 30-min blocking step with 10% goat serum.After another wash with 1× PBS, the fixed cells were permeabilized with BD Perm/Wash ™ (Becton Dickinson, 51-2091 KZ) for 10 min.Following a second wash, the lymphocytes were incubated with goat anti-LRBA (Santa Cruz Biotechnology, sc-164907), rabbit anti-TGFbRI, or rabbit anti-TGFbRII antibodies for 1 h at 37°C.Subsequently, the cells were washed with PBS and incubated with secondary antibodies: anti-goat-Alexa Fluor 488 (Invitrogen, A11055), anti-rabbit-Alexa Fluor 594 (Jackson ImmunoResearch, West Grove, PA, USA, St Louis, MO, USA, 711-585-152), and DAPI (Sigma-Aldrich, 10236276001) for 1 h.Finally, the cells were washed and adhered to glass coverslips treated with poly-L-lysine (Sigma-Aldrich, P8920) for 1 h at 37°C.The preparations were mounted on slides using the VECTASHIELD mounting medium (Vector Laboratories, Newark, CA, USA, H-1000).

Confocal microscopy acquisition
Images were captured using a TCS SP-8 microscope (Leica Microsystems) at 63× magnification.Images were acquired using Leica LAS X software (Leica Microsystems).Co-localization analysis of samples with dual staining was conducted using Mander's correlation coefficient for at least 50 cells from three independent experiments.Data analysis and Mander's correlations were calculated using Fiji Image J software v2.7.0.

Statistical analysis
Results are presented as means ± standard deviation.Unless otherwise specified, non-parametric tests were used for statistical analysis, with p values <0.05 considered significant.

Lrba-/-mice show high levels of IgA in both the serum and feces
The IgA levels in Lrba-/-were explored in serum and feces, we found elevated IgA levels in Lrba-/-mice.We observed higher levels of IgA in both the serum and feces of Lrba-/-mice at baseline, regardless of whether the mice were young (10 weeks) or old (12 months), compared with those in WT mice.We also observed higher levels of circulating IgG in young Lrba-deficient mice than that in WT mice (Figure 1).As Burnett et al. previously reported increased IgG2b levels, we determined this isotype in the sera from these mice; however, as shown in Supplementary Figure 1, similar IgG2b levels were observed in both mice.These results suggest potential alterations in B cell pathways that are essential for IgA production in Lrba-deficient mice.In addition, the mechanisms underlying the induction of the other isotypes were unaffected.

Altered B-cell subpopulations in
Lrba-/-mice After observing elevated IgA levels in Lrba-deficient mice, we assessed whether there was an alteration in the development of the spleen and B1 cell subpopulations in the peritoneal cavity.The spleen populations were not affected by the absence of Lrba.However, upon analyzing B1 lymphocyte subpopulations in the peritoneal cavity, we observed a decrease in both B1a and B1b subpopulations (Figures 2A, B).Normal B cell differentiation in the spleen and altered levels of B1a and B1b cells in the peritoneal cavity have led to an unclear source of IgA.To determine the lymphoid organs that predominantly produce IgA, we analyzed the mesenteric lymph nodes (MLN), PPs, and small intestine to detect the presence of IgA+ B cells and IgA+ plasma cells.
In the histological sections corresponding to the spleen and MLN, we observed a trend toward a higher number of IgA+ B cells and a more substantial presence of IgA+ plasma cells in both organs of Lrba-deficient mice (Figures 3A, B).
Smaller germinal centers were observed in the PP of Lrba-mice than that in WT, but with an evident size increase for these organs (Figures 4A, B).Additionally, a lower number of IgA+ B cells was counted in these organs, which contrasts with the results obtained in the spleen and MLN; however, upon determining the number of IgA+ plasma cells, we observed similar numbers in both mouse groups (Figure 4C).Analysis of the number of PPs in the small bowel of mice and measurement of the width and height of these organs confirmed a significantly higher number and size in Lrba-/-mice (Figure 4D).Considering the predominance of PPs in Lrba-/-mice, these lymphoid organs can be considered important sources of IgA.
The small intestine was analyzed, as it is one of the sites where IgA+ plasma cells reside, produce, and secrete IgA.Notably, significantly higher number of IgA+ plasma cells were observed in Lrba-/-mice (Figures 5A, B).
Considering these results, the high levels of IgA in the serum and feces of Lrba-deficient mice could be explained by the increased number of IgA-producing plasma cells observed in all anatomical sites evaluated.

B cells from Lrba-deficient mice show reduced TGFbR expression
We analyzed aspects of the signaling pathway, including TGFbRI and TGFbRII expression in B cells.As shown in Figure 6A, Lrba-/-B cells have lower total TGFbRI and TGFbRII expression.As TGFbR is a membrane receptor that undergoes continuous recycling, we sought to determine if there was a difference in expression on the cell surface or intracellularly in B cells.
Surface and intracellular expression of TGFbRI is depicted in Figure 6B.There were no differences in the surface expression of this protein in B cells between the WT and Lrba-deficient mice.However, when analyzing the intracellular expression, we detected significantly lower expression of TGFbRI within the B cells of Lrba-deficient mice.Similarly, there were no differences in surface expression of TGFbRII, and a significant reduction in intracellular expression was also detected.

Lrba-deficient B cells showed increased SMAD2 phosphorylation in unstimulated conditions
After observing reduced intracellular expression of TGFbR in B cells from Lrba-deficient mice, SMAD2 phosphorylation was analyzed.Protein extracts were obtained from unstimulated Notably, SMAD2 phosphorylation was significantly increased in unstimulated Lrba-/-B cells (Figure 7), whereas stimulation with the recombinant cytokine slightly induced SMAD2 phosphorylation, which remained similar to the basal levels throughout the entire kinetics (Figures 7A, B).This result contrasts with B cells from Lrba +/+ mice, where initial phosphorylation was low, and it gradually increased with TGFb1 stimulation.These data suggest that Lrba participates in controlling the activation of TGFbR.

Lrba co-localizes with both TGFbRI and TGFbRII in B cells
We analyzed whether TGFbR colocalizes with Lrba.Figures 8A  and B show that Lrba co-localizes with both TGFbRI and II, respectively, with an approximate Mander's correlation coefficient of around 0.7 (Figure 8).Co-immunoprecipitation experiments were also performed; however, the Lrba antibody did not sufficiently immunoprecipitate this protein, and only a weak signal was observed in both Lrba and TGFbRII immunoprecipitates (Supplementary

Discussion
LRBA has emerged as an important protein in immune processes, murine models can currently be used to study cellular functions involving Lrba.Here, we focused on elucidating the possible mechanism underlying the high levels of IgA in Lrba-/-mice.
We determined the levels of IgA in the serum and feces of young (10 weeks old) and old (12 months old) mice.Notably, the mice were not immunized.The data presented here reproduces previously published information on increased IgA levels.In contrast to previously published data (6), Significantly higher levels of IgG were detected in the sera of young mice in this study.Burnett et al. reported an increase in the IgG2b subclass in mice with a homozygous small deletion in the Lrba gene generated using CRISPR/Cas9 (19); however, we also determined the levels of IgG2b, which were not reproduced (Supplementary Figure 1).Additionally, we explored the in vitro induction of IgA+ and IgG2b+ B cells and their possible inhibition by TGFbRI SB505124.Upon stimulation, IgA+ B cells were properly induced, and the use of a TGFbR inhibitor (SB505124) blocked IgA+ B cell formation.This effect was not observed in the induction of IgG2b+ B cells (Supplementary Figure 3), suggesting that the increased levels of Lrba-/-mice are partly due to TGFbR activation; however, this switch to IgA requires other signals.
To determine whether the high levels of IgA production were due to a predominant B cell population, analysis of different B cell differentiation in the spleen and B1 subpopulations in the peritoneal cavity was performed.No alterations were detected in the proportion of B cells in the spleen; however, B1 cells were detected at reduced levels, which is consistent with the data reported by Gamez-Diaz et al.Explanations for the reduced B1 levels could result in poor survival, which has been reported for human LRBA-deficient B cells (2); B1 increased migration towards the intestine might be another possibility.B1 cells are precursors to a substantial percentage of intestinal IgA+ plasma cells in the lamina propria (20)(21)(22).Alterations in B1 cell maintenance, differentiation, or migration should be investigated in future studies.
To determine the source of the increased IgA levels, different secondary lymphoid organs, such as the spleen, MLNs, PPs, and small intestine, were examined.In all these organs, B cells can undergo isotype switching, not only to IgA, but also to other isotypes (23).Notably, all lymphoid organs explored, including the small intestine of Lrba-/-mice, showed a significantly higher number of IgA+ plasma and/or B cells than in WT mice.Importantly, although normal levels of IgA+ plasma cells were observed, significantly higher numbers and sizes were detected.These data suggest that all the explored organs contributed to the increased IgA production observed in both the feces and serum of Lrba-/-mice.
The increased sera IgA levels occur in non-immunized mice and the increased counts of IgA+ B and plasma cells suggest that the isotype switching to IgA occurs spontaneously in the Lrba -/-mice.The crucial role of TGFb1 and TGFbR in T-dependent and independent class switch to IgA+ led us to propose TGFbR is an ideal candidate to be analyzed to explain the increased IgA production in Lrba deficiency.
To address this issue, TGFbR (I and II) expression was evaluated; the endocytosis of this receptor defines the intensity and duration of signaling upon contact with TGFb1 (24).No differences were detected in the B-cell surface expression of the receptor.Importantly, a significant reduction in TGFbR I and II expression were found in permeabilized B cells from Lrba -/-mice.The altered expression of TGFbRI and II led us to evaluate the activation of SMAD2, the first signaling molecule activated after stimulation with the recombinant TGFb1.High SMAD2 phosphorylation levels were observed in Lrba-/-B cells under non-stimulated conditions, confirming that Lrba participates in the TGFbR signaling pathway.This result was consistent with the hypothesis of spontaneous class switching to IgA.A previous report describes that high phosphorylation of SMAD2 leads to a preference for B cells to switch to IgA (25); however, that report described increased membrane expression of TGFbR in B cells that cannot form clathrin-coated vesicles.TGFbR signaling depends on endocytosis of the receptor upon interaction with TGFb1 (26), and it has already been reported that it is internalized in endosomes coated with clathrin and caveolin (27).It is possible that in the absence of clathrin, the caveolin-dependent TGFbR endocytosis led to increased intensity of the signaling and promoting TGFbR expression.Additionally, another explanation for the normal expression of TGFbRI and II may be that TGFbR signaling also activates other pathways, such as PI3K-Akt-mTOR or MAPK (28, 29).In particular, Akt activation has been demonstrated to promote TGFbR membrane expression (30).However, we did not find any reports regarding PI3K-Akt-mTOR activation by TGFbR in B cells.Future studies should explore this issue.
SMAD2 phosphorylation was also examined in B1 and B cells derived from MLNs (Supplementary Figure 4).However, similar levels of pSMAD were detected between Lrba-/-and Lrba+/+ B cells and the expression of CD38, an activation marker, was detected at similar levels in both murine strains, indicating that basal activation of the TGFbR pathway may be favored in splenic B cells.
Reduced intracellular TGFbRI/II and high SMAD2 phosphorylation in Lrba-/-splenic B cells confirmed alterations in this signaling pathway.Human and murine Lrba have high levels of similarity, therefore, these proteins may have similar functions.Additionally, CTLA4 membranal expression is decreased in Lrba-/mice (6,19).Human LRBA interacts with the Rab11 GTPase, a protein involved in recycling membrane receptors.TGFbR is recycled through the endocytic system, and also depends on Rab11-coated recycling endosomes for this (31).It is very likely that similar to CTLA4, Lrba is also required for effective TGFbR recycling, and its absence may favor TGFbRII degradation, which may explain the reduced intracellular expression observed in Lrba-/-B cells.LRBA role in CTLA-4 recycling was recently explored by Janman et al., who suggested, that the interaction between CTLA4 and Rab11 occurs only when LRBA is correctly expressed (32).
TGFbR also depends on Rab11 recycling endosomes, suggesting that Lrba can interact with TGFbR.Co-localization and coimmunoprecipitation experiments were performed to determine if Lrba can interact with TGFbRI and/or TGFbRII.The confocal microscopy experiments shown in Figure 8 show a Mander's correlation of Lrba colocalizing with both TGFbRI and II subunits with mean values of approximately 0.7, suggesting that Lrba may interact with both proteins and/or that Lrba is involved in the TGFbRI/II signaling pathway.As all commercially available antibodies are directed to human proteins, the negative control for Lrba staining is included in Supplementary Figure 5, indicating that the staining observed for Lrba is specific.
Finally, co-immunoprecipitation assays were performed, and the possible interaction between Lrba and TGFbR was determined, a faint Lrba band was detected in the immunoprecipitated TGFbRII and vice versa, suggesting the physical interaction between both proteins, Lrba-/-B cells were used as negative controls for this assay, observing a lack of Lrba bands in both Lrba and TGFbR immunoprecipitates (Supplementary Figure 2).It is possible that the high molecular weight of Lrba (320 KDa) may make it difficult to immunoprecipitate, in addition to the lack of antibodies directed to mice.The interaction between Lrba and TGFbR should be explored in future research to determine the exact Lrba domain (s) responsible for the interaction with TGFbRII and/or RI.Previously, Lo et al., demonstrated an interaction between LRBA and CTLA-4 via concanavalin A-like and PH-like domains (5).Given that TGFbR and CTLA-4 share no homology, the other domains of Lrba are likely responsible for such interactions.
Altogether, our results position Lrba as an essential molecule in controlling TGFbR signaling, influencing the differentiation mechanism of IgA+ B and plasma cells.Another important aspect is that while defects associated with TGFbR signaling have not been extensively described in humans, this work is an important starting point for investigating this signaling pathway in human diseases, such as cancer.

3. 3
The Lrba-deficient mice have more IgA + B and plasma cells in secondary lymphoid organs and the small intestine

5
FIGURE 5 IgA+ B and plasma cells in histological sections from small bowel.(A) Representative histological slices from small bowel stained with anti-IgA (green), anti-CD19 or anti-CD138 (red), and DAPI.(B) Statistical analysis of IgA+ B and plasma cells counts per field.n=3 for both Lrba+/+ and Lrba-/mice.The Student´s t-test compared the means between the Lrba+/+ and Lrba-/mice, **p<0.01.