Macrophages and Epithelial Cells Mutually Interact through NLRP3 to Clear Infection and Enhance the Gastrointestinal Barrier

Round 1

Reviewer 1 Report

The authors have investigated the role of the inflammasome in murine intestinal epithelial cells during infection with C. rodentium. The authors have used an in vitro co-culture of J774A.1 and CMT-93 cells to mimic the interaction of macrophage and colonic epithelial cells. Although the results are different from the initial hypothesis, the authors have done a good job of properly interpreting the data and further investigating the mechanism. I recommend the publication of this manuscript given the following minor concerns are addressed:

1. Since inflammasome activation requires translation of NLRP3, the protein expression of NLRP3 should be measured after treatment with ATP or YVAD (figure 1).
2. How was secreted IL-1β measured from the apical membrane (figure 3C)

Author Response

The authors have investigated the role of the inflammasome in murine intestinal epithelial cells during infection with C. rodentium. The authors have used an in vitro co-culture of J774A.1 and CMT-93 cells to mimic the interaction of macrophage and colonic epithelial cells. Although the results are different from the initial hypothesis, the authors have done a good job of properly interpreting the data and further investigating the mechanism. I recommend the publication of this manuscript given the following minor concerns are addressed:

We thank the reviewer for this positive summary and account of our manuscript and hope that we have provided sufficient clarifications for their comments. Changes within the manuscript are noted in each response as well as red font in the text.

1. Since inflammasome activation requires translation of NLRP3, the protein expression of NLRP3 should be measured after treatment with ATP or YVAD (figure 1).

We thank the reviewer for highlighting the importance of measuring protein level changes, along with RNA level. While we agree that NLRP3 protein level could provide some insight to its bioavailability and activity levels within the cytosol and activity level, we would like to highlight the value of measuring NLRP3 activity via measurement of IL-1B and IL-18 secretion, which indicated the final outcome of this pathway. Furthermore, the presence of NLRP3 protein does not confer NLRP3 activity or pathway outcomes, which is the target of ATP (induces NLRP3 inflammasome activity) and YVAD (inhibits caspase-1 activity and thereby IL-1B/IL-18). Additionally, the presence of NLRP3 in epithelial cells is still quite controversial and therefore we were more interested in the output (IL-18 and IL-1 beta); we have attempted to measure this protein in the past with variable success and have found cytokine release to be much more consistent. We do, however, acknowledge this limitation and have added it to the discussion (lines 321-324, page 12 paragraph 1).

1. How was secreted IL-1β measured from the apical membrane (figure 3C)

Thank you for highlighting that we could be more clear in our description of this method. In order to clarify the text provided in the methods section (added Lines 114-116, page 3 paragraph 3, and 144-146, page 3 paragraph 5), we have also now labelled the basolateral and apical compartments in our model (Supplemental Figure 1) to visually describe the method used to examine secretion from the apical membrane. This method utilized Transwell cultures, in which the supernatants were collected from the apical compartment of the Transwell; a similar method has been used by other laboratory teams (Belic et al, Front. Microbiol., 08 January 2019 PMID: 30671036)

Reviewer 2 Report

Nlrp3 is shown to be critical for clearing attaching/effacing pathogens like Citrobacter rodentium by regulating Il1b/Il-18 secretion. But the role of Nlrp3 in tissue repair process has not been studied well. In this manuscript authors studied the epithelial barrier integrity after Citrobacter infection using a co-culture method using a colonic epithelial cell line and a macrophage cell line. Authors found that Citrobacter infection induced Nlrp3 mRNA levels slightly and it did not change Il1b and IL-18 secretion. Supernatants from infected epithelial cells augmented the clearance of Citrobacter Rodentium by macrophages in an inflammasome independent manner. On the other hand inflammasome activation in macrophages promoted pathogen clearance and improved barrier integrity of epithelial cells.

 I find this study to be interesting. Authors have designed experiments well to test their idea. It is a well written manuscript. However I find there are some minor issues that need to be taken care.

Minor comments:

1. Use LPS+ATP control for the experiments in figure 1. I understand that Citrobacter rodentium cell wall has LPS but it’s availability during infection may be critical to induce Nlrp3 and Il1b secretion in epithelial cells.
2. Heat treat the epithelial/ macrophage culture supernatants to determine if the factors are cytokines/proteins.
3. In the discussion comment on on Il1b receptor signaling role on epithelial/macrophages to explain the results in figure 4 and 5

     4. In Figure 5 align + sign properly to the representing bar in the figure

Author Response

Nlrp3 is shown to be critical for clearing attaching/effacing pathogens like Citrobacter rodentium by regulating Il1b/Il-18 secretion. But the role of Nlrp3 in tissue repair process has not been studied well. In this manuscript authors studied the epithelial barrier integrity after Citrobacter infection using a co-culture method using a colonic epithelial cell line and a macrophage cell line. Authors found that Citrobacter infection induced Nlrp3 mRNA levels slightly and it did not change Il1b and IL-18 secretion. Supernatants from infected epithelial cells augmented the clearance of Citrobacter Rodentium by macrophages in an inflammasome independent manner. On the other hand inflammasome activation in macrophages promoted pathogen clearance and improved barrier integrity of epithelial cells.

 I find this study to be interesting. Authors have designed experiments well to test their idea. It is a well written manuscript. However I find there are some minor issues that need to be taken care.

We thank the reviewer for this positive summary and account of our manuscript and hope that we have provided sufficient clarifications for their comments. Changes within the manuscript are noted in each response as well as red font in the text.

Minor comments:

1. Use LPS+ATP control for the experiments in figure 1. I understand that Citrobacter rodentium cell wall has LPS but it’s availability during infection may be critical to induce Nlrp3 and Il1b secretion in epithelial cells.

We thank the reviewers for their suggestion about adding LPS + ATP as a control, as this is commonly used for exploring NLRP3 activation in immune cells such as macrophages. Although C. rodentium surface LPS may not be available for interaction with TLR4, if LPS was not readily available in Figure 1 then we should have seen a subsequent increase in IL-1 beta secretion in the Transwells where gentamicin was added during infection and left for 24 hours. In addition, although overnight C. rodentium is centrifuged and resuspended in DMEM prior to infection, there would still be sufficient free LPS within the solution. In any case, we have found that C. rodentium provided a consistent activation of the inflammasome in our model, justifying its use as a positive control.

1. Heat treat the epithelial/ macrophage culture supernatants to determine if the factors are cytokines/proteins.

This suggestion by the reviewers, to determine if the factors are cytokines or other proteins by heat treating the supernatants, is indeed valuable and we thank them for this. This would be an interesting experiment for a future study as there are likely to be many factors involved in macrophage-epithelial cell interaction during bacterial infection. However, heat inactivating the supernatants would only show whether a heat labile protein is involved, and most soluble proteins are heat labile. A much more intensive study is required to determine this mechanism, and we have highlighted the need for further characterization of the secreted factor in our Discussion (Lines 392-394, page 13 paragraph 2).

1. In the discussion comment on on Il1b receptor signaling role on epithelial/macrophages to explain the results in figure 4 and 5

We thank the reviewer for their suggestion on IL-1b receptor signalling pathways, in regard to macrophage and epithelial cell interactions. As IL-1b has been shown to be produced by immune cells (particularly macrophages) in the intestinal tract, the majority of studies only look at its role in these cells and less in the epithelium. In the discussion, we have alluded to the fact that we have previously shown extracellular ATP to induce a shift towards M2 class of macrophages and this may explain the improved barrier integrity (Lines 381-385, page 13 paragraph 1).

4. In Figure 5 align + sign properly to the representing bar in the figure

Thank you to the reviewer for bringing this to our attention. We have fixed the alignment issue highlighted in Figure 5.