Lrig1-expression confers suppressive function to CD4+ cells and is essential for averting autoimmunity via the Smad2/3/Foxp3 axis

Regulatory T cells (Treg) are CD4+ T cells with immune-suppressive function, which is defined by Foxp3 expression. However, the molecular determinants defining the suppressive population of T cells have yet to be discovered. Here we report that the cell surface protein Lrig1 is enriched in suppressive T cells and controls their suppressive behaviors. Within CD4+ T cells, Treg cells express the highest levels of Lrig1, and the expression level is further increasing with activation. The Lrig1+ subpopulation from T helper (Th) 17 cells showed higher suppressive activity than the Lrig1- subpopulation. Lrig1-deficiency impairs the suppressive function of Treg cells, while Lrig1-deficient naïve T cells normally differentiate into other T cell subsets. Adoptive transfer of CD4+Lrig1+ T cells alleviates autoimmune symptoms in colitis and lupus nephritis mouse models. A monoclonal anti-Lrig1 antibody significantly improves the symptoms of experimental autoimmune encephalomyelitis. In conclusion, Lrig1 is an important regulator of suppressive T cell function and an exploitable target for treating autoimmune conditions.

authors should use the same therapeutic strategy for 6F01 and anti-IL17A to analyze the therapeutic potential of modulating LRIG1.
-The authors propose that activation of LRIG1 induces activation of p-Smad2/3 and increases Foxp3 mRNA expression through an Akt/mTOR/Foxo mechanism. The authors didn't prove that and should analyze more carefully the signaling following LRIG1 activation and the outcome of this stimulation.
-The CD45RBHigh colitis model was performed using total CD4+ sufficient or deficient for LRIG1. This model should be performed using LIRG1+ or LIRG1-Tregs cells instead of total CD4+ cells. The conclusion of the current experiment is not clear and doesn't show that LIRG1+ Treg cells present a better suppressive capacity. This is an interesting but preliminary study. The authors present what looks like an interesting discovery; that is, they show evidence suggesting that the transmembrane protein Lrig1 both marks and regulates regulatory T cells (Tregs). Furthermore, they generated an anti-Lrig-1 antibody, AF01, that promoted Treg function in vitro and in vivo and supressed experimental autoimmune diseases. The strength of the paper is the novelty of the Treg-Lrig1 finding and the possible therapeutic potential of the developed Lrig1 antibody. However, on the downside, the paper contains overinterpretation of the data, it is mechanistically shallow, important controls are missing, and there are a series of questionmarks regarding the 'therapeutic' antibody.
Specific points regarding the biological discovery: 1. The authors analyze cytokine-stimulated T cells that they named Th0, Th1, Th2, and Th17 without confirming their respective characteristics. The authors should confirm their subtype identities through transcriptional profiling, cytokine production, or functionality or rename the cell populations according to their treatment rather than their believed phenotypes.
2. The authors claim in the Title and elsewhere that Lrig1 acts through Smad2/3 signaling, which is not firmly supported by the data presented. The only data pertaining to Smad2/3 that is presented is an experiment showing that treatment with the non-validated anti-Lrig1-antibody 6F01 for 2 or 3 days during iTreg differentiation results in enhanced pSmad2/3 levels (Fig. 5b). There could be numerous different explanations for this result. If

Response 1.2 =>
We appreciate the Reviewer's comment on the proposed molecular mechanism of Lrig1 signaling. It has been known that Lrig1 interacts with EGFR and induces the degradation of EGFR by recruiting E3 ligase c-Cbl to the cytoplasmic domain of EGFR in the cancer cells 1 . Also, EGFR has an immunological role in inflammatory settings mediated by its downstream signaling molecules such as AKT or mTOR 2 . Phosphorylated AKT induced by EGFR stimulation inhibits Smad3 phosphorylation, suppressing TGF-b-mediated Foxp3 expression and also inhibit the functions of FoxO1 protein essential for Foxp3 expression and activates mTORC1 pathway to suppress Foxp3 expression in Treg cells [3][4][5][6] . Therefore, we hypothesized that in Treg cells, Lrig1 stimulation induces EGFR degradation by c-Cbl recruitment and further inhibits AKT and mTOR phosphorylation, thereby increasing Foxp3 expression by First, we examined the level of EGFR on the surface of Treg cells using flow cytometry, and EGFR expression in Treg cells was analyzed by western blot. As shown in the revised Fig. 5e-h, EGFR expression in iTreg cells was reduced by stimulating Lrig1 with Lrig1-specific monoclonal antibody 6F01. In addition, the treatment of CD4 + Foxp3 + T cells with 6F01 antibody significantly inhibited the phosphorylation of AKT, mTOR, and FoxO1 in a concentrationdependent manner. We assume that the down-regulation of EGFR expression on the surface of Treg cells by Lrig1 stimulation mediates the induction of Foxp3 expression via suppressing the inhibitory signals for Foxp3 expression.

Response 1.3 =>
We thank the Reviewer for the valuable comment. Our report demonstrated that Lrig1-expressing CD4 + T cells have a strong suppressive activity by highly expressing anti-inflammatory molecules such as Foxp3, IL-10, or PD-1, as presented in Figures 1, 2, and 4. Therefore, we performed the CD45RB hi cell-mediated colitis model using Lrig1 + or Lrig1cells in total CD4 + T cells to see whether the Lrig1 + population in CD4 + T cells can inhibit autoimmune symptoms in vivo model, and the results were included in Fig. 4a We have performed the new experiment by transferring Lrig1-expressing or Lrig1-knockout (KO) iTreg cells into the CD45RB hi -mediated inflammatory bowel disease (IBD) model, as the Reviewer suggested. In the revised Supplementary Fig. 8, we showed that the adoptive transfer of iTreg cells from Lrig1-knockout mice didn't alleviate the IBD symptoms well and suppress the colonic inflammation induced by CD45RB hi T cell transfer when compared to that iTreg cells from Lrig1 +/mice. Further, we observed that Lrig1 +/-iTreg cell-recipient mice showed decreased histological scores of colon inflammation compared to the recipients of CD45RB hi cell only or Lrig1 KO iTreg cells. Consistent with CD4 + Lrig1 high or Lrig1 low cell transfer results, Lrig1-expressing iTreg cell-recipient had a higher percentage of Foxp3 + T cells in intraepithelial lymphocytes (IEL) in the colon and also mesenteric lymph node (mLN) than Lrig1-knockout iTreg cell recipient. Furthermore, the proportion of CD4 + IFNg + T cells was significantly decreased by Lrig1 +/-iTreg cell transfer in both CD45.1 + and CD45.1cells in IEL and mLN compared to Lrig1-deficient iTreg cell transfer. These results demonstrate that Lrig1-expressing iTreg cells have a higher suppressive activity to reduce inflammatory symptoms in the colitis model than Lrig1-deficient iTreg cells.

Response 1.4 =>
We appreciate the Reviewer's point and now provide the quantification of effector cell proliferation activity in vitro suppression assay. As shown in updated Fig. 2A (see below), Lrig1-expressing CD4 + and CD4 + Foxp3 + T cells have a significantly higher suppressive activity than CD4 + Lrig1 -T cells. This result suggests that the Lrig1 + population in CD4 + T cells represents a strong suppressive subpopulation of CD4 + T cells like conventional Treg cells. In addition to this quantification, we analyzed the expression of Treg cell markers such as CD25, PD-1, CTLA-4, or Foxp3 in CD4 + Lrig1 -, Lrig1 + , or Foxp3 + T cells to investigate the suppressive function of CD4 + Lrig1 + T cells. Interestingly, we found that CD4 + Lrig1 + T cells show a higher expression level of these Treg cell markers than CD4 + Lrig1 -T cells, and the expression level of these markers in CD4 + Lrig1 + T cells was similar to that in CD4 + Foxp3 + T cells except CD25 expression (Revised Supplementary Fig. 3). Based on these results (Fig. 1f, 2b, 4g and Supplementary Fig. 3), we found that Lrig1 + T cells had a suppressive potential comparable to Foxp3 + T cells by highly expressing Treg cellassociated regulatory markers.

Response 1.5 =>
We thank the Reviewer for this suggestion. We included the quantification graph for the Fig. 2c data. Compared to Lrig1 low cells, Lrig1 hi cells among CD4 + Foxp3 + T cells exhibit a higher suppressive activity even though all cells express a similar level of Foxp3. Based on this result, we demonstrate that Lrig1 is a representative marker of a strong suppressive subpopulation in Treg cells.

Response 1.6 =>
We appreciate the Reviewer's suggestion. We repeated the in vitro suppression assay and now included the data for the suppression level of WT Treg cells (Lrig1 +/+ ) in the updated Fig. 3d with the representative flow cytometry data and graph (See updated Fig. 3d). There was a significant difference between Lrig1 +/+ and Lrig1 +/or Lrig1 -/-iTreg cells in each effector: Treg ratio. WT Treg cells showed the highest suppressive function compared to Lrig1 +/and Lrig1-deficient Treg cells.

Response 1.7 =>
We appreciate the Reviewer's suggestion. We carefully analyzed the RNA sequencing data to compare the gene expression profile of Lrig1 + with that of Lrig1in mouse CD4 + T cells. Now the heatmap of the top 20 differentially

Ratio (Effector T cells [E] : iTreg cells [T]
Cell proliferation dye-eFlour670 We identified 468 upregulated DEGs in Lrig1 + CD4 + T cells compared to Lrig1 -CD4 + T cells and 309 upregulated DEGs in Lrig1 -CD4 + T cells compared to Lrig1 + CD4 + T cells. Interestingly, the top 20 DEGs contained various Treg cell markers such as Foxp3, Il2ra, and Ctla4. The expression of these genes was higher in Lrig1 + T cells than in Lrig1 -T cells in all three samples (Revised Supplementary Fig. 10a). Next, we used the Metascape 9 to do the enrichment analysis in each gene set. The Metascape carries out pathway and process enrichment analysis with the following ontology sources: GO Biological Processes, KEGG Pathway, Reactome Gene Sets, CORUM, WikiPathways and PANTHER Pathway. In addition to the pathway and process enrichment analysis, we also performed a more careful analysis to investigate which cell type is similar to Lrig1 + T cells or Lrig1 -T cells in terms of gene expression profile and which transcription factor regulates the expression of DEGs in Lrig1 + CD4 + T cells.
As shown in revised Supplementary Fig. 10b, we demonstrated that the DEGs expressed by the Lrig1 + T cells are involved in the upregulation of cytokine production and negative regulation of the immune system process (Upper panel). In contrast, the upregulated pathways in Lrig1 -T cells are related to immunoglobulin production or lymphocyte activation (Lower panel). Surprisingly, our results showed that the cell type with the gene set most similar to the upregulated genes in Lrig1 + T cell is FoxP3 + T cells (Upper panel). And the cell type with the highest relation to the upregulated genes in Lrig1 -T cell is B220 + B cells (Lower panel) as shown in the revised Supplementary Fig. 11a. Moreover, using TRRUST enrichment analysis, we demonstrated that the expression of the upregulated genes in Lrig1 + T cell are regulated by Nkfb1, Rela or Foxp3 which is associated with the expression of suppressive molecules in Treg cells (Revised Supplementary Fig. 11b).

Response 1.8 =>
We appreciate the Reviewer's suggestion, and we have shown the functional association between Lrig1 and the natural thymic Treg cells (tTreg) development using Lrig1 -/mice in revised Supplementary Fig. 7. It is known that the tTreg precursor (CD4 + CD25 + Foxp3 -) is developed into tTreg (CD4 + CD25 + Foxp3 + ) through the tTreg development process in Thymus 10,11 . Therefore, we analyzed the populations of tTreg precursor (early stage) and tTreg (late stage) in the thymus of Lrig1 +/+ and Lrig1 -/mice by flow cytometry.
As shown in the revised Supplementary Fig. 7a, similar levels of tTreg precursor and the tTreg populations were detected in Lrig1 +/+ and Lrig1 -/mice. In addition, the difference in the level of CD4 + Foxp3 + T cell population in the spleen, thymus, and inguinal LNs of Lrig1 +/+ and Lrig1 -/mice was not observed (Revised Supplementary Fig. 7b). These results suggest that the absence of Lrig1 did not affect the development of tTreg cells.

Enrichment Analysis in TRRUST of Upregulated Genes in Lrig1 + T cells
Fos Atf2

Enrichment Analysis in PaGenBase of Upregulated Genes in Lrig1 + T cells 12
Cell-specific: Mast cells Cell-specific: Microglia Cell-specific: CD4+ T cells Cell-specific: Neuro2a Tissue-specific: Lymph node Cell-specific: NK cells

Response 1.9 =>
We appreciate the Reviewer's point. We now added the statistical analysis in every figure.

Response 1.10 =>
We now updated and unified the font size on each figure.

Response 1.11 =>
We appreciate the Reviewer's comment. We now included the isotype control for each group. 89. 9 6.38

Response 1.12 =>
We thank the Reviewer for the suggestion. As the Reviewer suggested, we revised the format of the representative dot plot -IL-10-PE (y-axis) vs. CD4-APC (x-axis), including isotype control-PE antibody in Fig. 1f. IL-10 expression in CD4 + Lrig1 + T cells was significantly higher compared to CD4 + Lrig1 -T cells in the spleen.

Response 1.13 =>
We appreciate the Reviewer's suggestion. We repeated the IL-10 ELISA assay using the same group in Fig. 2a and the updated group in the updated Fig. 2b. The level of IL-10 secretion from CD4 + Lrig1 + T cells was significantly higher than that of CD4 + Lrig1 -T cells. This result supports that Lrig1-expressing CD4 + T cells have a higher suppressive function than CD4 + Lrig1 -T cells. According to the Reviewer's comment, we analyzed the Lrig1 level on the surface of CD4 + T cells after treating the cells with control or the different doses of Lrig1-targeting siRNA. The expression level of Lrig1 protein on the surface was significantly reduced by siLrig1 treatment in a concentration-dependent manner compared to scrambled siRNA (Revised Fig. 3a).

Response 1.15 =>
We appreciate the Reviewer's comment. As shown in Comment 1.1, we repeated the EAE animal model using a Lrig1-targeting antibody and analyzed the ratio of Lrig1 + Foxp3 + , total Foxp3 + , IFNg + or IL-17A + in CD4 + T cells in a lymph node, or spinal cord of the treated mice. In the representative dot plots (see below figures), the increase of Lrig1 + Foxp3 + or total Foxp3 + population in lymph node and spinal cord CD4 + T cells was observed by the 6F01 treatment or anti-IL-17A mAb treatment compared to isotype IgG2a injection. According to the Reviewer's comment, we added the representative images of the isotype control-PE for the Lrig1-PE antibody using the same gating strategy for all samples in the updated Fig. 6c or data now shown in the manuscript below.

Response 1.16 =>
We appreciate that isotype control for Lrig1-PE is needed. In the updated Supplementary Fig. 1b, we added the isotype control goat IgG-PE-stained samples. Compared to isotype control-PE, we examined Lrig1 expression on each T cell population using Lrig1-PE antibody. Fig. 1b

Response 1.17 =>
We appreciate the Reviewer's suggestion. We now provide the quantification of LRIG1 expression in western blot analysis using Image J software. As shown in the revised Supplementary Fig. 1c, the relative expression level of Anti-IL-17A mAb 1.59

Response 1.19 =>
We thank the Reviewer for this suggestion. We added the survival curve of lupus-prone mice in Supplementary Fig.  9b. It was observed that 20% of the untreated and CD4 + Lrig1 -T cell-transferred mice died at 28 weeks of age. Weeks of age Survival rate (%)

Reviewer #2 (Remarks to the Author):
This is an interesting but preliminary study. The authors present what looks like an interesting discovery; that is, they show evidence suggesting that the transmembrane protein Lrig1 both marks and regulates regulatory T cells (Tregs). Furthermore, they generated an anti-Lrig-1 antibody, AF01, that promoted Treg function in vitro and in vivo and supressed experimental autoimmune diseases. The strength of the paper is the novelty of the Treg-Lrig1 finding and the possible therapeutic potential of the developed Lrig1 antibody. However, on the downside, the paper contains overinterpretation of the data, it is mechanistically shallow, important controls are missing, and there are a series of questionmarks regarding the 'therapeutic' antibody.
Specific points regarding the biological discovery:

Response 2.1 =>
We appreciate the Reviewer's point. Whenever we induce the differentiation of naïve T cells into each T cell subset (Th0=activated T cells, Th1, Th2, Th17 or Treg cells) under each T cell subset-specific polarizing culture condition, which was published in numerous previous papers by other groups and our group, the characteristics of each T cell subset including transcription factor expression, cytokine production or functionality is always confirmed before performing the subsequent experiments.
In this report, whenever the level of Lrig1 is measured at the RNA or protein level in each T cell subset, we examined the specific markers, including transcription factors and cytokines of each T cell subset, after inducing the differentiation of each T cell subset; Th0 (activated T cells)= CD69 + CD25 + , Th1= T-bet + IFNg + , Th2= GATA3 + IL-4 + , Th17= RORgt + IL-17A + , iTreg=Foxp3 + IL-10 + (See the representative analysis below). After the differentiation induction of naïve T cells into each T cell subset, we gated the T cell subset-positive population and analyzed the level of Lrig1 protein on the surface of the highly differentiated T cell subset population to measure the protein levels correctly.

Th0
Isotype control Naïve T cell

Response 2.2 =>
The Reviewer asks an essential question about the mechanism of action of Lrig1 through Smad2/3 signaling in Treg cells. As the Reviewer commented, we demonstrated that Lrig1-targeting monoclonal antibody (6F01) treatment during iTreg cell differentiation induced the expression of Foxp3 (RNA and protein levels) via the phosphorylation of Smad2/3 in the revised Fig. 5a, 5b and 5d. In addition to iTreg cells, we presented that Lrig1 stimulation enhanced the Foxp3 + population through the increase of p-Smad2/3 in Th17 differentiation in Supplementary Fig. 13a and b.
We also agree with the Reviewer's point that more supporting data would be needed to explain this mechanism.
To address this point, we performed 1) an inhibitor assay using SIS3, which is a potent inhibitor of Smad3 phosphorylation, and 2) a phospho-flow assay to find if 6F01 treatment inhibits AKT phosphorylation which is an upstream suppressor of Smad3 phosphorylation with shorter kinetics.
First, naïve T cells were induced to differentiate into Treg cells under the suboptimal iTreg differentiation condition in the presence of isotype control IgG2a or 6F01 (5 µg ml -1 ) and were treated with different concentrations of Smad3 phosphorylation inhibitor, SIS3. Then the level of CD4 + Foxp3 + T cell population was examined using flow cytometry. As described in the revised Fig. 5c, the level of CD4 + Foxp3 + T cells with 6F01 treatment was similar to that with isotype control treatment when SIS3 1 or 5 µM was used. But, SIS3-untreated cells showed an increased Foxp3 + T cells with 6F01 treatment compared to those with isotype control treatment. This result suggests that the increase of Smad2/3 phosphorylation accomplishes the increment of the Foxp3 level by 6F01 treatment, and this effect was completely abrogated by the blockage of Smad2/3 phosphorylation.
Second, to investigate whether this stimulatory effect of 6F01 is mediated by the down-regulation of inhibitory signaling by phospho-AKT, we analyzed the level of phosphorylated AKT in iTreg cells with shorter kinetics using flow cytometry. Within 1 min to 30 min, the level of AKT phosphorylation in Ser473 was considerably diminished by 6F01 stimulation compared to isotype control or no stimulation, as shown in the revised Supplementary Fig. 15. Therefore, it is confirmed that Lrig1 stimulation using 6F01 inhibits EGFR-AKT signaling in a short time and could induce Smad2/3 phosphorylation to mediate the increase of Foxp3 expression (Fig. 5).

Response 2.3 =>
We appreciate the Reviewer's comment on the proposed molecular mechanism of Lrig1 signaling. It has been known that Lrig1 interacts with EGFR and induces the degradation of EGFR by recruiting E3 ligase c-Cbl to the cytoplasmic domain of EGFR in the cancer cells 1 . Also, EGFR has an immunological role in inflammatory settings mediated by its downstream signaling molecules such as AKT or mTOR 2 . Phosphorylated AKT induced by EGFR stimulation inhibits Smad3 phosphorylation, suppressing TGF-b-mediated Foxp3 expression and also inhibit the *** # *** *** *** ## functions of FoxO1 protein essential for Foxp3 expression and activates mTORC1 pathway to suppress Foxp3 expression in Treg cells [3][4][5][6] . Therefore, we hypothesized that in Treg cells, Lrig1 stimulation induces EGFR degradation by c-Cbl recruitment and further inhibits AKT and mTOR phosphorylation, thereby increasing Foxp3 expression by lifting the inhibition of AKT-mediated suppression of Smad3 or FoxO1 signaling. We have carefully investigated this hypothesis and included the data in the revised Figure 5e-h. First, we examined the level of EGFR on the surface of Treg cells using flow cytometry, and EGFR expression in Treg cells was analyzed by western blot. As shown in the revised Fig. 5e-h, EGFR expression in iTreg cells was reduced by stimulating Lrig1 with Lrig1-specific monoclonal antibody 6F01. In addition, the treatment of CD4 + Foxp3 + T cells with 6F01 antibody significantly inhibited the phosphorylation of AKT, mTOR, and FoxO1 in a concentrationdependent manner. We assume that the down-regulation of EGFR expression on the surface of Treg cells by Lrig1 stimulation mediates the induction of Foxp3 expression via suppressing the inhibitory signals for Foxp3 expression.

Response 2.4 =>
We appreciate the Reviewer's point. As the Reviewer suggested, we confirmed the antibody specificity with iTreg cells isolated from Lrig1 wild-type and knockout mice. Consistent with the Lrig1 siRNA experiment, the anti-Lrig1 polyclonal antibody specifically detects Lrig1 expression on the surface of iTreg cells from Lrig1 WT mice. The Lrig1 antibody doesn't bind to Lrig1-deficient iTreg cells with a similar level of isotype control (See the Figure  below). Also, we now provide the results about Lrig1 expression level in Lrig1 +/and Lrig1 -/mice in updated Supplementary Fig. 5, including isotype control-PE or other Treg cell marker staining. Anti-Lrig1-PE polyclonal antibody only stained Lrig1 WT and Lrig1 +/-Treg cells.

Response 2.5 =>
As the Reviewer commented, we now add the detailed methods, including an antigen used for the immunization and the selection and the validation process to check if this antibody binds specifically to Lrig1 in the Method section. Also, the flow cytometry data is provided to evaluate the binding affinity of affinity-maturation clones, showing the 6F01 antibody has the highest binding affinity to mouse iTreg cells among affinity-matured clones.
The amendments were made in Methods section of the manuscript:

Methods:
"Anti-Lrig1 monoclonal antibodies were generated using the antibody phage display and the panning method using mouse Lrig1 gene transfected-L cell as an antigen. After validating the antibody with high binding affinity among the generated anti-Lrig1 monoclonal antibodies through flow cytometric analysis, affinity maturation was performed

Response 2.7 =>
We appreciate the Reviewer's point about the specificity of 6F01. As the Reviewer mentioned in the comment, we used Lrig1 -/mice to investigate whether 6F01 has a function through Lrig1 binding. To confirm whether 6F01 has the specificity for Lrig1, naïve T cells obtained from Lrig1 WT or KO mice were differentiated into iTreg cells. Then the differentiated iTreg cells were treated with 6F01, and the level of binding percent was analyzed by flow cytometry. As a result, 6F01 only bound to Lrig1 +/+ iTreg cells expressing Lrig1 (See below figure). In addition, we investigated the function of 6F01 in both Lrig1 +/+ and Lrig1 -/-iTreg cells. When Lrig1 +/+ iTreg cells were treated with 6F01, the Foxp3 + T cell population increased, and phosphorylation of AKT or mTOR was inhibited. However, 6F01 did not show the inhibitory functions when treating Lrig1 -/-iTreg cells (See below figure).

days differentiated iT reg cells
Anti-mouse IgG2a-PE

Response 2.8 =>
We agree with the Reviewer's comment that references and descriptions of previous findings related to some defects in Lrig1 knockout mice need to be mentioned in our manuscript. Earlier studies on Lrig1-deficient mice have mainly investigated the defects induced by non-immune cell populations like intestinal or epithelial stem cells or neuronal cells.
In previous studies, Lrig1 is known to be expressed in intestinal or epithelial stem cells and some cancer cells: Our study is the first case to investigate the functionality of Lrig1 in immune cells, especially in Treg cells, in vitro and in vivo, which other studies might neglect. Importantly, Lrig1 is highly specific to Treg cells, and the level of Lrig1 is higher in the activated Treg cells than in the resting Treg cells. These results suggest that 6F01 mAb showed significant therapeutic potential in the autoimmune setting.

And, as with other surface proteins expressed in both immune cells and non-immune cells, the function of Lrig1 in T cells might be unique and different from that in other non-immune cells.
In addition to the fact that Lrig1 -/mice did not exhibit any significant immune-related abnormality, our results showing that 6F01 mAb did not show any single dose-and repeated dose-toxicity may hold the careful promise of minimal toxicity upon therapeutic application to autoimmunity. Currently, the standard toxicity study in rodents and monkeys is being undertaken with 6F01 mAb produced by the GMP facility.
Other general and specific points:

Response 2.9 =>
We appreciate the Reviewer's point. Some results in the previous manuscript version didn't have a statistical analysis by mistake. We now add the experimental statistics in every graph and the quantification data.

Response 2.10 =>
According to the Reviewer's comment, we included the reference to describe the Lrig1 knockout mouse strain. The strain used in this study was provided by Prof. Fiona Watt and is the same strain used in previous publications.

Methods:
"Lrig1 heterozygous and knockout (B6.129-Lrig1 tm1.1Hhed ) mice were generated and characterized as previously described. When used, the Lrig1-null mice did not have any abnormalities and were age-sex matched with Lrig1 +/mice."

Response 2.11 =>
We thank the Reviewer and agree that the sentence should be listed in the Authors contribution section. As described in Response 2.10, we added the mouse information in the Methods section, and the sentence was moved to the Authors contribution section.

Response 2.12 =>
We now described the pSmad2/3 assay in more detail in the Method section.

Response 2.13 =>
We agree with the Reviewer's comment that we need to include the reference and descriptions of previous findings related to the association of LRIG1 and BMP signaling in lipid metabolism.
The report from Herdenberg et al. 17 is cited and discussed in our manuscript.
Herdenberg et al. describe the function of LRIG proteins, including LRIG1 as regulators of lipid metabolism in adipocyte differentiation of fibroblast. This study generated Lrig-null mouse embryonic fibroblasts (MEF) and found that the role of LRIG proteins promotes adipogenesis when treated with an adipogenic cocktail. Further, they showed that Lrig-deficient MEFs had a lower sensitivity of BMP4 and BMP6 compared to wild type. The BMP sensitivity was recovered by ectopic LRIG1 and LRIG3 expression by analyzing pSmad1/5 phosphorylation with immunocytochemistry. On the other hand, the phosphorylation level of Smad3 in response to TGF-b1 was not changed in Lrig-null MEFs, suggesting that LRIG proteins, especially LRIG1, are associated with adipogenesis via BMP signaling, not required for TGF-b or receptor tyrosine kinase signaling.
This study is highly relevant to our paper, and the amendments were made in the Discussion section: "This study revealed that the expression and function of Lrig1 are dependent on TGF-b1. And we propose that there is a possibility that bone morphogenetic protein (BMP) signaling, one of the TGF-b superfamily members, may also be involved in Lrig1 signaling in Treg cells. The previous report has shown that LRIG proteins, including LRIG1, regulate lipid metabolism via BMP signaling through the phosphorylation of Smad1/5 in mouse embryonic fibroblasts, suggesting LRIG proteins may play an important role as BMP sensitizers. Furthermore, several studies have found that BMP signaling through BMPR1a controls Treg lineage and stability via Smad1/5/8 phosphorylation, and BMPR1a is highly expressed in activated T cells and Treg cells compared to naïve CD4 + T cells. In addition, TGF-b and BMP signals have a synergistic effect on the induction of iTreg cells even though the TGF-b signaling pathway is a key factor for the development of Foxp3 + Treg cells. Thus, we hypothesize that Lrig1 may play a role in the expression of suppressive markers through both TGF-b and BMP signaling to induce the phosphorylation of Smad1/5/8 from BMP and Smad2/3 from TGF-b. The molecular basis for the association of Lrig1 and BMP signaling remains to be elucidated."

Response 2.14 =>
We appreciate the Reviewer's point.
For the first part, we showed in the graph below that the expression of the reference gene, Hprt, was not changed, and there was no significant difference for each sample. Therefore, we could confirm that the expression of Lrig1 was different for each cell type, not by the change of the reference gene expression (Hprt) in Fig. 1a.
For the second question, we agree with the reviewer's concerns. To resolve this issue, we re-verified the relative expression of Foxp3 using Hprt as a first reference gene used for the comparative analysis of Lrig1 expression and GAPDH as a second reference gene. As shown in the updated Fig. 5d, it was confirmed that the relative expression level of Foxp3 was increased in response to the Lrig1-targeting antibody (6F01) compared to isotype IgG treatment regardless of the reference genes.

REVIEWER COMMENTS
Reviewer #1 (Remarks to the Author): The authors have satisfactorily addressed the issues I have raised in my review.
Reviewer #2 (Remarks to the Author): This remains an interesting study. Many of the concerns previously raised by me have been addressed adequately; however, some major and minor issues remain to be adressed.
MAJOR AND MINOR COMMENTS: 1. Fig. 1b, c, and elsewhere, the authors discuss Lrig1 levels, although the data rather show the fractions of Lrig1-positive cells. Percentage of Lrig1-positive cells should not be confused with Lrig levels.
2. The authors claim that Lrig1 "defines", is "specific" for, or "essential" for Treg cells. These claims are confusing given that Lrig1 is expressed by many other cell types in addition to the Tregs, and not all Tregs express Lrig1. Hence, it is suggested that less confusing expressions are used, indicating that enrichment and levels are the defining features, rather than suggesting absolute specificity or dependency.
3. Fig. 2c, e, f. What does "Effector only" at the different effector:regulatory ratios mean? Most of these "Effector onlys" show 80-90% proliferation. 80-90% in relation to what -what does 100% proliferation represent? 4. Fig. 3a, b. The Lrig1 siRNA reduced the fraction of Lrig1-positive CD4+ T cells and the proliferation of the effector cells. However, siRNAs are known to have prominent off-target effects. To account for the possible off-target effects, at least two different siRNAs against the target mRNA should be used. I apologize for not having noticed this shortcoming in my original review.
5. It is claimed that "the body weight… [was] …unchanged (Fig. 3f, Supplementary Fig. 6a)" (lines 132-133). However, I do not think any body weight data are presented in these figures or elsewhere.
6. It is claimed that "…the level of various T cell subsets in the spleen were unchanged (Fig.  3f, Supplementary Fig. 6a)" (lines 132-133). Here, the authors analyzed IFNg, IL-4, IL-17A, and Foxp3 in the CD4+ splenocytes in three Lrig1+/+ mice and three Lrig1-/-mice. According to Fig. 3f, the Lrig1-/-mice show a downregulation of IL-4+ cells by approximately 70% and of IL-17A+ cells by approximately 25%. Nevertheless, based on a "two-way ANOVA with Tukey's multiple comparisons test", the authors conclude that there was no difference between the genotypes. However, in my opinion, this is not an appropriate statistical test for comparing means of two different populations (Lrig1+/-vs. Lrig1-/-). Hence, with appropriate statistical testing, e.g., t-tests, it would not be surprising if the Lrig1-/-mice actually show significant changes in the T cell subset populations.
7. The mechanistic claim that Lrig1 should induce Tregs via the downregulation of EGFR, resulting in reduced pAKT signaling, which in turn would enhance TGFb-pSmad2/3 signaling that drives the differentiation of the Tregs remains speculative and is not firmly supported by the data.
(i) First, the whole model seems to be mostly based on data obtained by cultivating the T cells on a surface coated with the Lrig1 antibody 6F01. There is no evidence suggesting that this stimulus corresponds to a physiological signal. This follows from the fact that there is no known physiological ligand for Lrig1 and in fact very little is known about the actual molecular function of Lrig1. Hence, plastic surface-coated antibodies might indeed stimulate the physiological function of Lrig1, but it may equally well inhibit its function or do something completely different. Hence, it cannot be concluded that this artificial stimulus "stimulates the function of Lrig1". It also follows that one cannot assume that 6F01 is agonistic, because it might equally well be antagonistic or something else.
(ii) Second, regarding the downregulation of EGFR after 2-3 days of cultivation, this represents a correlation rather than implying causation. The fact that EGF did not affect the differentiation of the Tregs may rather suggest (but not prove) that regulation of EGFR is not involved. The same argument holds for the pAKT and pmTOR results. These are also only correlations seen after 2-3 days in culture, which could possibly result from the differentiation of the cells or because of other secondary causes.
(iii) Furthermore, the idea that EGFR signaling should suppress Treg differentiation or functionality seems to be at odds with previous prominent publications showing that EGFR signaling promotes rather than suppresses Treg function [e.g., Zaiss  . This apparent discrepancy with published work needs to be addressed, in the Discussion, at least. (iv) The acute effect of 6F01 on pAKT levels shown in Supplementary Fig. 15 is difficult to understand. First, what happened at time 0 which made the pAKT levels go down by more than 50% during the subsequent 60 minutes, irrespective of the treatment given? Second, as I understood it, the treatment here was the transfer of the differentiated cells into wells coated with 6F01. If so, one would imagine that the response to 6F01 would require that the cells first contacted the bottom or sides of the wells. How then can a strong response be recorded already after 1 minute? For most cell types, it takes minutes or hours for the cells to sediment to the bottoms of the wells. Furthermore, except for the rapid dip of 6F01treated cells at 1 minute, the response curves appear more or less identical, or superimposable on each other. How should that be understood? (v) It is stated that 6F01 induces the phosphorylation of Smad2/3 followed by the enhanced Foxp3 transcription (lines 306-307). However, no time-resolved data showing this sequence of events is provided.
(vi) As pointed out by the authors, LRIG1 has been shown to enhance BMP signaling, but to not affect TGFb signaling. This is now discussed by the authors but it also warrants further experimental clarification, not the least because the sera used for T cell cultivation presumably contain physiologically relevant BMP levels. For example, is the antibody used to evaluate pSmad2/pSmad3 levels specific for pSmad2/3 or does it crossreact with BMPregulated pSmad1/5? Such crossreactivity between pSmad2/3 and pSmad1/5 is common among "specific" pSmad antibodies and this reviewer was not able to find any information for the used antibody in this regard.
8. The Lrig1 mouse strain used remains ambigously defined. In M&M (lines 374-375) it is stated that B6.129-Lrig1tm1.1Hhed mice were generated as described by Mao et al. (Mao et al., 2018). However, in the Results section (line 125), it is stated that the Lrig1 mice used were the mice generated by Suzuki et al. (Suzuki et al., 2002). Hence, it remains unclear which Lrig1 knockout mouse strain was used.
9. Furthermore, in addition to the identity of the Lrig1 knockout allele used it would be important to know the genetic background of the mice. That is, was the Lrig1 knockout allele back-crossed onto a genetically clean mouse strain and, if so, for how many generations? Or were the mice of a mixed genetic background, and if so, what was the mix?
10. Also in vivo, the mechanism of action for 6F01 remains obscure. For example, it is unclear to what degree 6F01 could (i) function by affecting other cells than the suspected Tregs, or (ii) function by binding to other proteins than Lrig1 (e.g., Lrig2, Lrig3, or other). These issues deserves a discussion or could be resolved by treating mice after adoptive transfers of Lrig1+/+ or Lrig1-/-cells into Lrig1+/+ or Lrig1-/-mice. 11. The apparent abscence of observed 6F01 side effects is promising, indeed. Nevertheless, possible side effects based on previous observations in knockout mice should be discussed to give a more complete picture. Furthermore, the statement that "Lrig1-null mice did not have any abnormalities" (lines 375-376) has to be specified -what were the features that were analyzed and found not to be abnormal?
12. It is stated that the cells were cultured in 100 ug/ml penicillin/streptomycin (lines 383 and 387). This information is ambigous because penicillin and streptomycin are two different compounds. Were they both at 100 ug/ml or was it 100 ug/ml in total, and if so, how were the proportions between the two compounds?
13. I was unable to find information about the BDsprint cDNA syntesis kit online and did not understand why SYBR green was included during cDNA synthesis (line 403)?

Reviewer #2 (Remarks to the Author):
This remains an interesting study. Many of the concerns previously raised by me have been addressed adequately; however, some major and minor issues remain to be addressed.

Response 2.1 =>
We appreciate the Reviewer's comment on the confusion of the unit definition in the figure and manuscript. As the Reviewer suggested, we changed "Lrig1 protein levels on the cell surface" into "the percentage of Lrig1 + or Lrig1population" in the figure and manuscript.

Response 2.2 =>
We thank the Reviewer's valuable comment on Lrig1 specificity and expression in the manuscript. As you understand, we found that Lrig1-expressing Treg cells had a higher immunosuppressive function than Lrig1-nonexpressing Treg cells. This evidence was also confirmed through mRNA sequencing, and Lrig1-expressing CD4 + T cells expressed higher levels of immunosuppressive markers too. Therefore, we suggested that Lrig1 is a surface marker that defines the suppressive population of CD4 + cells. However, we also agree with the Reviewer's comment that the expression 'defines' or 'specific' could be confused because Lrig1 is also expressed in the other cell types.
We now use less confusing representations of the specificity of Lrig1 + CD4 + T cells and change the title of this manuscript to "Lrig1 enriched in the suppressive population of CD4 + cells is essential for alleviating autoimmunity via the Smad2/3/Foxp3 axis".  Fig. 2c, e,

Response 2.3 =>
We appreciate the Reviewer's comment. As you understand from the Methods section protocol, i) "Effector only" means that only effector T cells were added in the tissue culture well, providing a positive control for proliferating T cells. The well with different Effector T cell : Regulatory T cell ratio contains the same number of effector T cells plus the different numbers of regulatory T cells. ii) Before co-culture of effector T cells with the different numbers of Treg cells, we first labeled the effector cells using eBioscience Cell Proliferation Dye eFluor 670 (Cat. 65-0840-85). Because the dye is evenly diluted into both daughter cells during the cell division, each division round can be monitored individually over time as an independent peak. Therefore, as the effector cell proliferates, the proliferation dye dilutes, resulting in several low-intensity picks during FACS analysis.
The first peak with the highest intensity is considered to be the population without proliferation, and the subsequent lower intensity peaks are judged as proliferating cell population with cell divisions. In this way, we quantified the level of cell proliferation, adding up the total intensity peaks except for the first highest intensity peak (nonproliferating population) by flow cytometry analysis. Therefore, "100% proliferation" means "the sum of all intensity peaks including the first highest peak, and "the proliferation % of effector only (80-90%)" is "100% proliferation minus the first highest peak % (non-proliferating population)". Fig. 3a, b

Response 2.4 =>
We appreciate the Reviewer's comment. We also agree with the comment to include at least two different siRNAs against the target mRNA to avoid possible off-target effects. In the revised Fig. 3a and 3b, the results using the different and second siRNA targeting Lrig1 or a scrambled siRNA (Cat No: AM4611, AM4613 for scrambled siRNA and AM16708, 4390771 for Lrig1 siRNA from Thermo Fisher) were added. Using two different siRNA, we confirmed that there is no off-target effect by the siRNAs targeting Lrig1, and these siRNAs show similar knockdown effects for Lrig1 protein expression. Also, the decreased level of Lrig1 expression on the CD4 + T cell surface by two different siRNAs causes lower suppressive activity.
Revised Fig. 3a, b Fig. 3f, Supplementary Fig. 6a)" (lines 132-133). However, I do not think any body weight data are presented in these figures or elsewhere.

Response 2.5 =>
We appreciate the Reviewer's point. The "spleen weight" was misused by mistake by "the body weight" in the figures. We changed "the body weight" to "the spleen weight" in the manuscript. (Fig. 3f,  Supplementary Fig. 6a)" (lines 132-133)

Response 2.6 =>
We appreciate the Reviewer's comment. As the Reviewer suggested, we have performed the two-tailed unpaired Student t-test to compare the results from two different samples in Fig. 3f. We concluded that there is no significant difference among various CD4 + T cell populations in Lrig1 +/and Lrig1 -/mice. In the Figure 3 legend, we provided the statistical information for Fig. 3f. The exact p-values are; IFNr + -p= 0.7209, IL-4 + -p=0.132, IL-17A + -p= 0.1007, Foxp3 + -p= 0.1329. In addition to Fig. 3f result, we have shown that the differentiation potential of naïve CD4 + T cells from Lrig1 WT or KO mouse to various T cell subsets is similar in Fig. 3e. Taking these results together, Lrig1 deficiency doesn't affect T cell subset polarization and balance.  Fig. 6.

(Comment 2.7) The mechanistic claim that Lrig1 should induce Tregs via the downregulation of EGFR, resulting in reduced pAKT signaling, which in turn would enhance TGFb-pSmad2/3 signaling that drives the differentiation of the Tregs remains speculative and is not firmly supported by the data. (i) First, the whole model seems to be mostly based on data obtained by cultivating the T cells on a surface coated with the Lrig1 antibody 6F01. There is no evidence suggesting that this stimulus corresponds to a physiological signal. This follows from the fact that there is no known physiological ligand for Lrig1 and in fact very little is known about the actual molecular function of Lrig1. Hence, plastic surface-coated antibodies might indeed
However, as the Reviewer mentioned, there is no known physiological ligand, and very little is known about the actual molecular function of Lrig1 in Treg cells, and it cannot be concluded that 6F01-mediated stimulation might not be agonistic and might be artificial.
Among the mAb specific to Lrig1 we generated, there is another agonist-like mAb (1C07), which may recognize the similar epitope in Lrig1, and two antagonist-like mAbs (K12 and K46) which may recognize the different epitope in Lrig1. In our new and unpublished data below, two agonist-like mAbs (6F01 and 1C07) induced EGFR degradation as compared to isotype IgG control in iTreg cells differentiated from naïve T cells. At the same time, two other antagonist-like mAbs (K12 and K46) inhibited the EGFR degradation, increasing EGFR level in iTreg cells. In addition to the level of EGFR change, we confirmed the reverse correlation of EGFR level and Foxp3/p-Smad2/3 levels in Treg cells between these mAbs (See below data). The stimulation of Treg cells by agonist-like mAbs (6F01/1C07) showed a lower level of EGFR and an increased level of Foxp3 and phosphorylated Smad2/3 as compared to isotype IgG control. On the other hand, stimulation of Treg cells with antagonist-like K12 or K46 mAb increased EGFR protein and showed a similar or decreased levels of Foxp3 and p-Smad2/3 compared to isotype IgG treatment. The antagonistic function of K46 mAb was also tested in a mouse tumor model (unpublished data) in that the tumor size of K46-treated mice was reduced, and the population of CD4 + Foxp3 + Treg cells decreased among tumor-infiltrated lymphocytes. Taking these results together, we assume that 6F01 is specific to Lrig1 functions as an agonistic antibody toward the increase of Treg-ness by induction of EGFR degradation.

Response 2.7 (ii) =>
We appreciate the Reviewer's comment. In our experimental design, the induced level of Foxp3 and Lrig1 is highest at 2-3 days after differentiation induction from naive T cells to iTreg cells. This is why we measured EGFR level and AKT/mTOR phosphorylation on day 2 or 3 after differentiation into Treg cells. Also, it has been well-known from previous studies that EGFR transmits a signal through the AKT/mTOR pathway by phosphorylation 4

Response 2.7 (iii) =>
We appreciate the Reviewer's comment that the discrepancy between their works and this issue would be addressed in the Discussion of the revised manuscript.
The reports from Zaiss et al. 6 and Wang et al. 7 are cited and discussed in this manuscript. Our study demonstrated that Lrig1-targeting monoclonal antibody directly stimulates EGFR degradation on Treg cells in normal conditions. And Lrig1-mediated EGFR degradation induces Foxp3 mRNA level via p-AKT/p-mTOR/p-Foxo1.
In summary, while previous studies provide an advanced understanding of AREG-EGFR signaling in Treg cells in inflammatory or tumor settings, our study provides key functional insights of Lrig1 in Treg cells during Treg celldifferentiation and activation processes through Lrig1/EGFR/Smad2/3/Foxp3 axis. As mentioned in Response 2.7 (i) and (ii), we confirmed that each of the agonist and antagonist mAbs specific to Lrig1 could upregulate or downregulate the differentiation of Treg cells through the opposite functions of EGFR degradation.
The amendments were made in the Discussion section: "In the immunological environment, CD4 + T cells express EGFR, one of the ErbB family, and the expression level of EGFR is higher on Treg cells than that on effector CD4 + or CD8 + T cells in tumor or inflammatory microenvironment 6,7

Response 2.7 (iv) =>
We appreciate the Reviewer's comment. i) As you are already familiar with T cell physiology, phosphorylation induction of intracellular signal mediators proximal to the T cell receptor complex is very rapid 11 . Upon T cell receptor stimulation by ligand or mAb, phosphorylation of these signal mediators peaks within 3-10 minutes after stimulation returning to the basal level rapidly. These responses are much faster in the primary T cells than in the established T cell line or transformed T cell. Akt can also be rapidly dephosphorylated if no activation signal stimulates T cell receptors or EGFR signaling 12,13 .
As we demonstrated in the manuscript, the differentiated iTreg cells at day 3 were transferred into another well plate coated with isotype IgG or 6F01, and the transferred iTreg cells cannot receive the activation signals through TcR stimulated with anti-CD3/28 antibodies or cytokines, and Akt starts to dephosphorylate very quickly. ii) Technically, we agree with the Reviewer's comment that the cells need to be sedimented to the bottom of the wells to receive a signal from 6F01. To facilitate this process, we usually spin down the cells into the bottom of the coated wells with weak centrifugation at 300 g for 3 min and start to count the incubation time. This step was added to the protocol in the Methods section.

Response 2.7 (v) =>
We appreciate the Reviewer's comment. It is well known that Foxp3 transcription is induced by the phosphorylation of Smad2/3 14,15 . In addition, we showed that Foxp3 expression did not increase when Smad2/3 phosphorylation was inhibited by Smad2/3 inhibitor (SIS3) when Treg cells were treated with 6F01 (Fig. 5c). Therefore, together with the previous studies, our results reveal the sequence of events of pSmad2/3 and Foxp3.

Response 2.7 (vi) =>
We appreciate the Reviewer's point. We have investigated whether 6F01-mediated Lrig1 stimulation could induce the Smad1/5 phosphorylation known to be triggered by Lrig1-BMP stimulation 16 . As shown in the figure below, 6F01 treatment didn't induce the phosphorylation of Smad1/5. However, the level of pSmad2/3 was increased by Lrig1 stimulation through 6F01 mAb treatment (Fig. 6b). Also, we confirmed that TGF- signaling doesn't mediate the phosphorylation of Smad1/5 in Treg cells. In addition to this result, we confirmed with the manufacturer that the pSmad2/3 antibody used in our experiments does not have a cross-reactivity with pSmad1/5.

Response 2.8 =>
Thank the Reviewer for checking this confusing information. The mouse strain we used in our study is the same mouse strain described in Suzuki et al. 2002 17 . This information is included In the Results and Methods section.

Response 2.9 =>
Thank the Reviewer for the comment that we have to provide the genetic background of the Lrig1 knockout mouse.
The Lrig1 KO mouse we used in our study was back-crossed onto a genetically clean C57BL/6 mouse strain. It does not have any mixed genetic background. Also, we crossed Lrig1 heterozygous mice and used the mice in our study which do not express the Lrig1 gene through gene screening. Whenever a new generation of Lrig1 KO mice is born, the Lrig1 KO status is confirmed through gene screening and FACS analysis, but the accurate information about the number of generations still needs to be correctly monitored. (e.g., Lrig2, Lrig3, or other). These issues deserves a discussion or could be resolved by treating mice after adoptive transfers of Lrig1+/+ or Lrig1-/-cells into Lrig1+/+ or Lrig1-/-mice.

Response 2.10 =>
We appreciate the Reviewer's comment. The sentences about these issues were newly added in the discussion section. We repeatedly confirmed Lrig1 mAb 6F01 doesn't bind to Lrig1-null iTreg cells with a similar level of isotype control (See below figure). Also, other Lrig family proteins, such as Lrig2 and Lrig3, are rarely expressed in CD4 + T cells, including Treg cells (See below). Therefore, we think 6F01 specifically binds to the Lrig1 protein and functions as Treg-ness agonistic antibody to regulate autoimmune responses.

Amendments in the manuscript: In Discussion section, Page 19, line 348
Data not shown in the manuscript (Comment 2.11) The apparent abscence of observed 6F01 side effects is promising, indeed. Nevertheless, possible side effects based on previous observations in knockout mice should be discussed to give a more complete picture. Furthermore, the statement that "Lrig1-null mice did not have any abnormalities" (lines 375-376) has to be specified -what were the features that were analyzed and found not to be abnormal?

Response 2.11 =>
We appreciate the Reviewer's comment and agree with the Reviewer that we need to demonstrate a possible side effect of Lrig1-targeting monoclonal antibody (mAb) and what abnormalities we observed in Lrig1 KO mice. The critical physiological changes and immunological abnormalities have been examined in Lrig1 knockout mice. There were no physiological abnormalities in Lrig1 knockout mice except minor bone mineral density and coat texture abnormality. Any abnormal immunological features such as splenomegaly or inflammatory CD4 + T cell population in lymph nodes and lungs were not detected. In addition to the fact that Lrig1 -/mice did not exhibit any significant immune-related abnormality, our results demonstrated that 6F01 mAb did not show any single dose-and repeated dose-toxicity, which may hold the careful promise of minimal toxicity upon therapeutic application to autoimmunity. Currently, the standard toxicity study in rodents and monkeys is being undertaken with 6F01 mAb produced by the GMP facility.