Fat-associated lymphoid clusters control local IgM secretion during pleural infection and lung inflammation

Fat-associated lymphoid clusters (FALC) are inducible structures that support rapid innate-like B-cell immune responses in the serous cavities. Little is known about the physiological cues that activate FALCs in the pleural cavity and more generally the mechanisms controlling B-cell activation in FALCs. Here we show, using separate models of pleural nematode infection with Litomosoides sigmodontis and Altenaria alternata induced acute lung inflammation, that inflammation of the pleural cavity rapidly activates mediastinal and pericardial FALCs. IL-33 produced by FALC stroma is crucial for pleural B1-cell activation and local IgM secretion. However, B1 cells are not the direct target of IL-33, which instead requires IL-5 for activation. Moreover, lung inflammation leads to increased IL-5 production by type 2 cytokine-producing innate lymphoid cells (ILC2) in the FALC. These findings reveal a link between inflammation, IL-33 release by FALC stromal cells, ILC2 activation and pleural B-cell activation in FALCs, resulting in local and antigen-specific IgM production.

infection and lung inflammation" Jackson-Jones and colleagues investigate the role of B cells in Fat-associated Lymphoid Clusters (FALCs) in the pleural cavity. The authors describe an unexpected role for the mediastinal FALCs in innate immune responses in pleural nematode infection and lung inflammation, where IL-33, produced by adipose tissue stromal cells, activates group 2 innate lymphoid cells (ILC2) that produce IL-5 needed for pleural B1 cell activation and IgM production.
The current study is a continuum of an earlier one from the same authors demonstrating inflammation-induced expansion and activation of FALCs and FALC-associated B cells. However the notion of a large population of IgM+ B cells in the peritoneal and pleural cavities is not novel, their function has remained elusive. Jackson-Jones et al. have used two models of pleural inflammation characterizing B1 cell activation in FALCs. The most significant finding of the study is that pleural FALC sense and respond to lung inflammation, suggesting their FALC and FALC B cell involvement in various pathological conditions of the lung, such as infections, allergies or asthma.
The following points are raised: • As the Ls parasite model is not widely used, would be great to have more details about infection, such as parasite burden on different days when samples were acquired. Also, references to relevant literature using the Ls model are missing (Babayan et al, Infection and Immunity 2003) or others.
• How relevant are B cells and IgM for Ls resistance? Some discussion of proof of this would be valuable.
• Figure 1 includes data acquired at several different timepoints, such as day 8, day 11 and day 18. What is the reasoning for that? What does it mean for the infection with Ls (parasite burden)? • The mesenteric and pericardial adipose compartments are populated by ILC2s. ILC2s react to IL-33 and other epithelial factors induced by parasite infection by proliferation and production of IL-13 and IL-5. Is there an expansion/activation of ILC2 during Ls infection? • Lung type 2 pneumocytes constitutively express IL-33 at high levels that is release rapidly upon infection, such as Nippostrongylus brasiliensis or lung inflammation (papain) model for example. The data in Figure 4 demonstrated that per g of tissue the mediastinum produces similar amount of IL-33 compared to the lung. However, in infection situation, there is so much more lung tissue that releases IL-33, making the relevance of IL-33 from FALC questionable. • The data in Figure 4 to claim that IL-33 derives from stromal cell is weak. Why not use mesenteric FALCs as a comparison? It is known that GAT has little FALCs and is a very distant non-relevant tissue. Were negative/positive controls performed for the immunostaining? The data should be complemented by analyzing tissues from Ls infected animals. • On Figure 5e it is evident that p Lavage provides folds more IgM than pl lavage even at steady state. This is not consistent with Fig.1, where total IgM levels at steady state in p lavage is close to 0. This should be discussed.
• What other cells produced IL-5 in the pericardium/lung? Do lung ILC2 produce IL-5? • It is surprising that that ILC2s which are central in the story linking IL-33 and B cell-produced IgM, received so little attention in the analysis. Some inclusion of this analysis would improve the manuscript.
• The biological importance of the pathways described is not evident from the data presented. There is no connection to pathogen resistance or tissue protection in inflammation. Loss-offunction and gain-of-function experiments are missing for a high-impact journal! • The term 'FALC' is sometimes confusing in the text as it is hard to understand which compartment is under discussion 'mediastinal FALC' or 'pericardial FALC' 'pleural FALC'. Page 10 talks about FALCs but figure 4 says "mediastinum". On figure 5 mediastinum has turned into 'mediastina". The is confusing to the reader.

Reviewer #1
Expert in pulmonary immune responses (Remarks to the Author): 1. The differentiation of PCs in IL-33R-/-mice seems not as poor as the authors want us to believe. In figure 3C, there is a huge population of SSChiIgDlo cells in the IL-33R-/mice. The absolute numbers go down (probably because the total inflammatory infiltrate goes down), but their ability to differentiate seems fine.
We entirely agree with the analysis of the reviewer. This was already stated in the text: 'Even though B2 cells failed to accumulate in pericardial FALCs of Il1rl1-/-mice, the differentiation of B2 cells into SSC high IgDplasma cells was not completely abrogated ( Fig. 3c and d), suggesting that IL-33R was only partially involved in plasma cell differentiation'. We now have modified this sentence as: 'Even though B2 cells failed to accumulate in pericardial FALCs of Il1rl1 -/mice, the differentiation of B2 cells into SSC high IgDplasma cells was not completely abrogated ( Fig. 3c and d), suggesting that IL-33R was not or only partially involved in plasma cell differentiation'.
2. It is interesting that the FALC size seems to go down in IL-33-/-mice prior to infection (Fig 3a) and after infection. Is this observation supported by total cell numbers?
Indeed in naïve animals, FALCs of IL-33R -/mice seemed smaller and there were less cells in the mediastinal adipose tissue of the IL-33R -/mice compared to BALB/c mice but this did not reach significance. A plot showing total cell numbers in pericardial adipose tissue has been added to Figure 3 and the following sentence has been added to the text: 'Immunofluorescence staining indicated that mediastinal FALCs of naïve Il1rl1 -/mice were smaller than their BALB/c counterparts, flow cytometric analysis of digested pericardial FALCs showed a trend for fewer cells in the mediastinal adipose but this did not reach significance (Fig.   3b).' 3. The division of cells into ki67hi and KI67med seems artificial and I have not seen this division before. I would gate all the KI67+ cells together. In this type of analysis, the flow plots (i.e in Fig 5c) would suggest much different conclusions. In fact, a huge proportion of B cells are responding in IL-33R-/-mice. Again, I would conclude that B cell proliferation and plasma cell differentiation are probably OK in the IL-33R-/-mice, but that the total numbers of cells is less.
We realize that this important point needs clarification. Our group has assessed Ki67 co-staining with BrDu following a 3hr BrDu pulse while Phil Taylor's group has done   similar experiments with pHH3 staining, which provides a definitive marker  "As eosinophils are the other main target of IL-5 3, 4 , we wanted to assess the contribution of eosinophils to the induction of B cell proliferation and IgM secretion. First, we analysed the impact of the delivery of anti-IL-5 antibody in the pleural cavity. As expected, we found a reduction in the number of eosinophils within the pleural exudate and a trend for reduced eosinophilia within the pericardial FALCs. However, this did not reach significance (Fig. 6h). To completely rule out that the effect we were seeing on B cells was dependent on eosinophils we performed Alt experiments in ΔdblGATA mice that lack eosinophils. At 48h following delivery of Alt, pericardial FALC B1a and B1b cells of ΔdblGATA mice were proliferating significantly more than their BALB/c counterparts (Fig. 6i), there was enhanced proliferation within the mediastinum as assessed by immunofluorescence staining (Fig. 6j) and there was no defect in the secretion of IgM within the pleural lavage (Fig. 6k). These data indicated that the induction of B cell proliferation and IgM secretion was independent of eosinophils. However, since both B cells and eosinophils are dependent on IL-5, they may be in competition for its access. In the absence of eosinophils, B cells would have more IL-5 available, providing an explanation for the enhanced B cell proliferation we found in ΔdblGATA mice." 5. One could make the argument that IL-33 is necessary for a type 2 inflammatory response, either from Th2 cells or ILC2 cells and that in the absence of IL-33, the whole response is impaired and fewer cells are recruited to the pleural cavity and mediastinum (data showing this are already published). Fewer eosinophils could mean less APRIL secretion and less support for plasma cells. IL-5 depletion may also reduce eosinophils and APRIL and lead to fewer PCs.
We concur that aIL-5 does deplete eosinophils in the pleural cavity, however we do not see a significant reduction in the FALCs themselves. These data have been added to Figure 6 (h) and the text altered as above. In addition, we are aware that eosinophils are required for the maintenance of plasma cells in the bone marrow (Chu et al 2011) but we do not believe this will have a role at this early time point (48h) following pleural cavity activation 6. The authors make the case that the FALCs of the mediastinum and pericardium are the most important source of IgM production following Ls infection. This may be true, but any evidence is lacking. Is there a way to eliminate the FALCs or the B cells in the pleural cavity?
In this study we clearly showed that during Ls infection, the levels of IgM rose only in the pleural cavity and not the serum or the peritoneal cavity, which indicated that IgM was secreted only in the pleural cavity (Figure 1a-b). In this cavity, we demonstrated that only B cells in FALCs were competent to produce antibody as shown by our in vitro culture study ( Figure 1g). We believe that this evidence though not entirely direct is strong. It is not currently possible to eliminate FALCs in the pleural cavity. Surgical removal of the pericardium is possible but total removal of FALC containing adipose is not.
Furthermore, such surgery would disrupt the pleural membranes where the nematode resides. We could use intra-pleural injection of anti-CD19 antibody to deplete B cells, however this wouldn't be a specific depletion of the cells in the FALCs but would target fluid phase B-cells as well, and thus we don't feel such experiments would add any new information to our manuscript.
7. The authors focus on IgM, however, IgE is also an important Ig isotype in these infections. Is IgE produced in the FALCs?
We typically have been unable to detect IgE within the substantial (2ml) volume of pleural lavage fluid that we routinely assessed and therefore had not pursued this isotype. However, because of the reviewers comment, 2) It is intriguing why the authors did not examine the antigen-specific response in Alternaria model. In Fig. 5e and f, the authors should show Alternaria-specific IgM titers in addition to the concentration of total IgM. We do not find any antigen specific IgM at 48h following Alternaria instillation, we believe that this time point is too early for an antigen specific response to have been generated.
3) In Fig. 6f, there seem to be two populations in anti-IL-5 treated group; half of population was not really affected by the anti-IL-5 treatment. Why?
Yes, we noted this as well. The immunofluorescence analysis in Figure 6f includes the % area of Ki67 within whole 'FALCs' and not specifically B-cells within FALCs as assessed by flow-cytometry, thus non B-cell types within the clusters are the most likely explanation for this population that exhibits higher Ki67, and are unaffected by anti-IL-5 treatment.
4) ILC2 is able to produce IL-5 in response to PMA and ionomycin even in naïve mice. IL-33 produced by Alternaria treatment is expected to induce the proliferation of ILC2s Therefore, the authors should show the absolute number of IL-5+ ILC2s in Fig. 6h.
We have removed the ILC data from Figure 6 and the absolute numbers of ILC2s have now been added to the new Figure 7 that also includes additional Ls ILC data. This paragraph has been added to the text:

FALC ILC2s increase following induction of pleural inflammation
Finally, we determined the cellular origin of IL-5 in pericardial FALCs during Ls infection by analyzing the intra-cellular levels of IL-5 within digested pericardium from C57BL/6 mice. We found here that Lineage -CD90.2 + ILCs, that represent 0.5-2% of total pericardial CD45 + FALC cells, constitute the only reservoir of IL-5 producing cells within the pericardium ( Fig. 7a-c). All other pericardial FALC cell populations assessed (CD45 -Gp38 + stromal cells, CD19 + MHC-II + B cells, TCRβ + MHC-II -SSC lo T cells, CD11b + F4/80/Ly6c + myeloid cells, Ly6G/SigF + SSC hi MHC-II -Granulocytes) had no detectable intracellular IL-5 compared to the fluorescence minus one control (Fig. 7a). ILCs expressed significantly more IL-5 than all other cell populations assessed (Fig. 7b). However, there was no significant difference in the geometric mean fluorescence intensity (gMFI) of IL-5 expression when comparing naïve and Ls infection, nor an increase in the percentage IL-5 expression within ILCs following infection (Fig. 7c). There was however a significant increase in the total number of ST2 + ILC2s within the pericardium following Ls infection (Fig. 7D). IL-5 + ILCs were also present within digested pericardium from BALB/c mice (Fig. 7e) and a trend toward an increase in the number of ST2 + ILC2s was seen at 48h following Alt instillation, however this did not reach significance. Thus our data indicate that increased numbers of IL-5 producing ILC2s are the most likely source of IL-5 for FALC B cell activation following pleural inflammation induced by two distinct experimental models." 5) Antigen-specific IgM is produced by FALC cells but not PLEC cells (Fig. 1g) but cell numbers in PLEC increased by Alternaria treatment (Fig. 5c, d), indicating that cells were mobilized to pleural cavity. It is intriguing that whereas adoptive transfer of PLEC cells resulted in the migration of B cells to FALC, antigen-specific B cells do not come out from the FALC to the pleural cavity. Where do PLEC cells come from? Are they from blood stream?
We believe that antigen-specific B cells are competent to migrate from the FALCS to the pleural cavity but when they do they are no longer able to produce antibodies since this is dependent on a close interaction with FALCs. FALC stromal cells provide the homeostatic chemokine Cxcl13 (previously shown in 5 ) and IL-33 creating a niche bringing together B cells and ILC2 as a source of the IL-5 required to support antibody production. Other cells in the pleural influx are likely recruited from the blood stream or an expansion of resident cells. The dynamics of recruitment from the blood and proliferative expansion of PLEC cells under type 2 conditions has been described 1 . Fig. 7, please add explanation of T, H, and L, which seem to be thymus, heart and lung.

6) In
We apologize for this oversight; the explanations for T, H and L have now been added to the figure legend for Fig 8 (previously Fig 7) 7) On page 4 th line, "Type 2" Innate Lymphoid Cells should be "Group 2" Innate Lymphoid Cells.
We have exchanged the word Type for the word Group in the text. are grossly imparied in the BALB/c μMT strain 11 . As these mice only lack IgM (unlike the C57BL/6 μMT) it suggests that IgM is needed for proper worm development. We are excited to unravel this complicated system in future work but feel that it is currently outside the scope of this manuscript. We hope to acquire reagents and mouse lines in the future that will allow us to address the role of secreted IgM in resistance to this infection. We believe that the contribution of pleural FALCs to local IgM production remains an important novel finding.
The following paragraph has been added to the discussion: In addition, B1 cells are implicated in resistance to both Ls 22 and human filariasis 26 . However, it is not practical to remove FALC from the pleural space during Ls infection, so we cannot directly address their role in protection. C57BL/6 µMT mice are not more susceptible to Ls primary infection 38 but that data is difficult to interpret because B cells are a major source of IL-10 22 , which is essential for susceptibility to Ls 39 and the development of female Ls adults is grossly imparied in the absence of IgM 27, 40 . There is marked accumulation of M2 macrophages in the serous cavities of Ls mice 29 and thus it will be important to use more refined models to determine whether IgM recognition facilitates parasite killing by macrophages, as has been described in a related parasite model 41 .
• Figure 1 includes data acquired at several different timepoints, such as day 8, day 11 and day 18. What is the reasoning for that? What does it mean for the infection with Ls (parasite burden)?
To be honest, Day 8, 11 and 18 were chosen because our initial hypothesis was that FALCs would be a major site of macrophage proliferation and these time points straddle the peak of proliferation. However, we made the novel discovery that FALCs were instead important site for B cell proliferation and antibody production. Nonetheless, we believe these are relevant time points as they allow us to assess the response at a stage of infection, when there is no significant difference in parasite burden between susceptible and resistant strains.
We have modified the following sentence in the text with the inclusion of additional references: We chose to assess resistant C57BL/6 mice at days 8-18 post infection, a time prior to immune mediated parasite killing but at which point an active immune response is occurring in the pleural cavity 28,29 • The mesenteric and pericardial adipose compartments are populated by ILC2s. ILC2s react to IL-33 and other epithelial factors induced by parasite infection by proliferation and production of IL-13 and IL-5. Is there an expansion/activation of ILC2 during Ls infection?
We have added new data to Figure 7 showing that ILC2s are increased in the pericardial FALCs during Ls infection (Fig. 7a-d). The text has been modified as shown in response to reviewer 2, question 4.
Lung type 2 pneumocytes constitutively express IL-33 at high levels that is release rapidly upon infection, such as Nippostrongylus brasiliensis or lung inflammation (papain) model for example. The data in Figure 4 demonstrated that per g of tissue the mediastinum produces similar amount of IL-33 compared to the lung. However, in infection situation, there is so much more lung tissue that releases IL-33, making the relevance of IL-33 from FALC questionable.
We acknowledge that following major lung infection/inflammation total levels of IL-33 will be higher in this tissue, but we don't believe that this IL-33 will have a relevant role in the B cell processes we observe in the FALCs. Cytokines typically act locally in controlled cell-to-cell interactions. Even circulating cytokines are normally bound to carrier molecules and not necessarily functional. It would seem unlikely and potentially dangerous for IL-33 to act in a broadly systemic manner. It makes more biological sense for there to be local control with locally produced IL-33 activating local B cell responses.
Indeed, our data highlight this point as we don't observe a full proliferative response in Bcells of the pleural cavity during either Ls or alternaria.
• The data in Figure 4 to claim that IL-33 derives from stromal cell is weak. Why not use mesenteric FALCs as a comparison? It is known that GAT has little FALCs and is a very distant non-relevant tissue. Were negative/positive controls performed for the immunostaining? The data should be complemented by analyzing tissues from Ls infected animals.
The mesenteric adipose has comparatively few FALCs in comparison to the omental adipose (see Benezech et al 2015), and thus omentum was used as a positive control.
We used the GAT as a negative control site that lacks clusters, enabling us to extrapolate that IL-33 expression is linked to stromal cells within FALC containing tissues and not within adipose depots with a more dispersed immune network. In Figure   4 c-f we have now expanded our analysis of FALC stromal cell IL-33 production. This new data includes flow cytometric analysis of digested pericardium, including secondary only controls for IL-33 staining, whole mount immunofluorescence staining for IL-33 also including secondary only controls and ELISA analysis of mediastinum IL-33 release from both naïve and day 11 Ls infected C57BL/6 mice.
We have modified the text as follows: 'We next addressed whether FALC stromal cells contained IL-33 during Ls infection (Fig. 4c).
Flow cytometric analysis of digested pericardial FALCs confirmed that >95% of CD45 -GP38 + stromal cells from C57BL/6 mice expressed intracellular IL-33 (Fig. 4d) compared to only ~2% of CD45 + cells, when gated based on a secondary antibody only control (Fig. 4d). At day 11 following Ls infection, no difference in the levels of IL-33 within stromal or haematopoietic cells was detected by flow cytometry (Fig. 4d), ELISA analysis of spontaneous IL-33 release during 1h in vitro culture of the mediastinum (Fig. 4e) or whole mount immunoflouresence staining as compared to naive controls (Fig. 4f)' • On Figure 5e it is evident that p Lavage provides folds more IgM than pl lavage even at steady state. This is not consistent with Fig.1, where total IgM levels at steady state in p lavage is close to 0. This should be discussed.
The explanation for the difference is that the data in Figure 1 are from C57BL/6 mice, while the data in Figure 5e are from BALB/c mice. A key point is that the peritoneal lavage reflects the serum IgM and naïve C57BL/6 animals have markedly lower levels of serum IgM (~10 μg) than naïve BALB/c animals (3 mg). The important finding is that in contrast to the pleural lavage, there is no increase in the amounts of IgM in the peritoneal lavage following Ls infection or Alternaria instillation, with the peritoneal lavage mirroring the serum response (though folds lower in the C57BL/6 mice).
• What other cells produced IL-5 in the pericardium/lung? Do lung ILC2 produce IL-5?
The only cells that we can show to be making IL-5 within the pericardium are ILCs, this data along with representative intracellular IL-5 staining of pericardial stromal cells, Bcells, T-cells, myeloid cells and granulocytes has been added to Figure 7. And the text has been modified to state: 'We found here that Lineage -CD90.2 + ILCs, that represent 0.5-2% of total pericardial CD45 + FALC cells, constitute the only reservoir of IL-5 producing cells within the pericardium (Fig. 7ac). All other pericardial FALC cell populations assessed (CD45-Gp38+ stromal cells, CD19 + MHC-II + B-cells, TCRβ + MHC-II -SSC lo T-cells, CD11b + F4/80/Ly6c + myeloid cells, Ly6G/SigF + SSC hi MHC-II -Granulocytes) had no detectable intracellular IL-5 compared to the fluorescence minus one control (Fig. 7a). ILCs expressed significantly more IL-5 than all other cell populations assessed (Fig. 7b)' • It is surprising that that ILC2s which are central in the story linking IL-33 and B cellproduced IgM, received so little attention in the analysis. Some inclusion of this analysis would improve the manuscript.
We have added additional ILC data from both Ls and Alternaria to revised Figure 7.
Please see above for changes made to the text.
• The biological importance of the pathways described is not evident from the data presented. There is no connection to pathogen resistance or tissue protection in inflammation. Loss-of-function and gain-of-function experiments are missing for a highimpact journal! We have spent considerable time considering how to directly address the role of the FALCs in resistance vs susceptibility to this parasite. The most obvious step, to specifically ablate the FALCs of the pleural space, is simply not possible. The use of B cell KOs is compromised as discussed above. However, we are excited to unravel this complicated system in future work and our next step will be to acquire mouse lines that will allow us to address the role of secreted IgM in resistance to this infection. However, as discussed above, the results may not provide immediate or straightforward answers.
We believe our finding that pleural FALCs are a critical site for local IgM production remains an important novel finding with implications for a variety of conditions beyond this parasite model.
Minor concerns: • Fig 1c to supplementary figures. This is not part of relevant findings We believe that most readers will not be familiar with these tissues and that their visual identification in figure 1c, will help with understanding. We can move this to the supplementary section if required.
• Figure 2a has no Ki67-posiitve B220+ cells, however this is claimed in the text. These errors have been corrected in the text.
• The term 'FALC' is sometimes confusing in the text as it is hard to understand which compartment is under discussion 'mediastinal FALC' or 'pericardial FALC' 'pleural FALC'. Page 10 talks about FALCs but figure 4 says "mediastinum". On figure 5 mediastinum has turned into 'mediastina". The is confusing to the reader.
We have altered Figure 5 so that it now reads Mediastinum rather than Mediastina and have modified the text in an attempt to be more consistent For the most part the authors have appropriately addressed my previous comments. I still believe that proof of the the essential role of the mediastinal/pericardial FALCs is lacking (point 6), but as the authors point out, selectively eliminating the FALCs may be technically impossible. As a result, we have to rely on circumstantial evidence, which are reasonably compelling.
Reviewer #2 (Remarks to the Author): Jackson-Jones et al. revised their paper with new experimental data. The authors promptly addressed most of the points raised by reviewers and the paper is much improved. There are still several minor points that need to be addressed.
1) The authors labeled PLEC with CFSE and CTV in the experiments shown in Fig. 6 and page 13. The cell composition of PLEC is unclear. The authors should provide the composition of PLEC such as X% of B1a, Y% of B1b, Z% of B2 etc. In addition, the authors should describe how they identified B1 cells in these experiments when they analyzed transferred cells.
2) The 4th line on page 7, PLEC should be spelt out here instead of those on the 12th line.
5) The 4th line from the bottom of page 15, (Fig. 7f) should be cited after "Alt instillation" or "did not reach significance". 6) At the end of the first paragraph of Discussion, (Fig. 7) should be (Fig. 8).
7) The 8th line of page 17, (Fig. 3A-D) should be ( Fig. 3a-d). In general, the authors have significantly improved the manuscript by including novel data and modifying the text to facilitate an uninformed reader. I understand that gain-and loss-of-function experiments to address functional relevance of FALCs in physiology and infection are technically complicated due to unavailability of tools. However, I have some minor comments on modifications: 1) It is surprising that there is no change in IL-5-producing ILC2s, whereas, there is an increase in ILC2 numbers in the tissue upon Ls infection or alternaria. Is there a change when analyzing ILC2 proportions to CD45+ cells? The data on Fig. 7 is presented as change of FMO control staining that is very misleading at first glance. Raw MFI would be more informative. You are demonstrating fold change but apparently there is none when comparing naïve vs. infected condition. The rate of IL-5+ ILC2s is very high, higher than published for ILC2s in any other condition. Therefore, this data warrants for comparison with some tissue earlier published to include IL-5+ ILC2s, such as the colon.
2) On figure 4 you demonstrate the difference in IL-33 release from different tissues. Does IL-5 follow the same pattern? The ILC2s in FALC seem to be highly activated according to your data.

Reviewer #1 (Remarks to the Author):
For the most part the authors have appropriately addressed my previous comments. I still believe that proof of the the essential role of the mediastinal/pericardial FALCs is lacking (point 6), but as the authors point out, selectively eliminating the FALCs may be technically impossible. As a result, we have to rely on circumstantial evidence, which are reasonably compelling.

Reviewer #2 (Remarks to the Author):
Jackson-Jones et al. revised their paper with new experimental data. The authors promptly addressed most of the points raised by reviewers and the paper is much improved. There are still several minor points that need to be addressed.
1) The authors labeled PLEC with CFSE and CTV in the experiments shown in Fig. 6 and page 13. The cell composition of PLEC is unclear. The authors should provide the composition of PLEC such as X% of B1a, Y% of B1b, Z% of B2 etc. In addition, the authors should describe how they identified B1 cells in these experiments when they analyzed transferred cells.
The text and figure legend has been amended as shown below: In the results section p13: "To test this, we labeled total PLEC from BALB/c and Il1rl1 -/donor mice with CFSE and Cell Trace Violet (CTV) respectively, co-injected labeled PLEC into the pleural space of a recipient BALB/c animal, instilled Alt 18h later and compared the CFSE (WT) and CTV (Il1rl1 -/-) labeled B cell populations after 48h (Fig. 6a). The injected PLEC were composed on average of 60% B cells of which 60% were B1 cells and 40% B2 cells. After transfer, between 2 and 4% of the PLEC were of donor origin. We could detect both CFSE and CTV positive CD45 + CD19 + CD11b + B1 and CD45 + CD19 + CD11b -B2 cells within the PLEC and in the FALCs, indicating that IL-33R signaling was not necessary for the recruitment of pleural B cells into FALCs (Fig. 6b and not shown)." In the figure legend p28: "Flow cytometric analysis of digested pericardial and mediastinal FALCs showing gating of CFSE and CTV labelled B1 cells (gated as CD45 + CD19 + CD11b + )." 2) The 4th line on page 7, PLEC should be spelt out here instead of those on the 12th line.
We have now corrected this in the text. tissues. Does IL-5 follow the same pattern? The ILC2s in FALC seem to be highly activated according to your data.
As discussed in 1), Molofsky et al. demonstrated with reporter mice that gonadal adipose tissue ILC2s were prone to produce IL-5. We found that the levels of IL-33 produced per gram of tissue by the gonadal adipose was low compared to the omentum or mediastinum, indicating that the high level of IL-5 production in naïve mice we observed in pericardial and mediastinal FALCs was not directly correlated to the amount of IL-33 produce per g of tissue.
However, we think that inside FALCs, IL-33 producing stromal cells provide a IL-33 rich environment enabling ILC2 to produce IL-5 and allowing maintenance of IL-5 dependent B1 cells. The existence and the nature of a niche for an IL-33 dependent activation of ICL2 in gonadal adipose tissue remain to be investigated.