Expression of Plet1 controls interstitial migration of murine small intestinal dendritic cells

Abstract Under homeostatic conditions, dendritic cells (DCs) continuously patrol the intestinal lamina propria. Upon antigen encounter, DCs initiate C‐C motif chemokine receptor 7 (CCR7) expression and migrate into lymph nodes to direct T cell activation and differentiation. The mechanistic underpinnings of DC migration from the tissues to lymph nodes have been largely elucidated, contributing greatly to our understanding of DC functionality and intestinal immunity. In contrast, the molecular mechanisms allowing DCs to efficiently migrate through the complex extracellular matrix of the intestinal lamina propria prior to antigen encounter are still incompletely understood. Here we show that small intestinal murine CD11b+CD103+ DCs express Placenta‐expressed transcript 1 (Plet1), a glycophoshatidylinositol (GPI)‐anchored surface protein involved in migration of keratinocytes during wound healing. In the absence of Plet1, CD11b+CD103+ DCs display aberrant migratory behavior, and accumulate in the small intestine, independent of CCR7 responsiveness. RNA‐sequencing indicated involvement of Plet1 in extracellular matrix‐interactiveness, and subsequent in‐vitro migration assays revealed that Plet1 augments the ability of DCs to migrate through extracellular matrix containing environments. In conclusion, our findings reveal that expression of Plet1 facilitates homeostatic interstitial migration of small intestinal DCs.

If the revision of the paper is expected to take more than three months, please inform the editorial office.
Revisions taking longer than six months may be assessed by new referees to ensure the relevance and timeliness of the data.
Once again, thank you for submitting your manuscript to European Journal of Immunology. We look forward to receiving your revision.  (ECM). Findings presented in this manuscript is novel, and beneficial for immunologists to understand molecular mechanisms of DC migration, but there are concerns to be addressed.

1.
Plet1-deficient DCs do not migrate from the small intestine to mesenteric lymph node. However, it is unclear whether Plet1-deficient DCs do not move at all, or they migrate and accumulate at particular areas where is potentially rich for ECM. DC localization or motility should be tested by histology or intravital microscopy to understand the migratory characteristics of Plet1-deficient DCs.

2.
Page 3, line 37. that microbial signals control Since there is no description about the association between microbiota and Plet1 prior to this sentence, â€˜microbial signalsâ€™ only in conclusion seems to be abrupt.
3. Figure 4D seems to be missing, and current Figure 4D is supposed be Figure 4E. The proper results should be added, and the figure should be corrected. Figure 5C are too small, and hard to read.

5.
Magnifications of diagrams in Supplementary Figure 4 are different among proteins, and hard to compare. Also, amino acid sequence besides the 3D diagram could be helpful to compare their similarity.

6.
Since 1) The authors to not describe any experiments designed to test the impact of the altered DC migration on immune responses. As dendritic cells are key players in the initiation of immune responses, such an experiment would strengthen the paper considerably.
2) It is unclear where, exactly, the increased number of double positive DCs in the small intestinal lamina propria reside. Immunohistochemistry might provide helpful and interesting information in this regard, although it is probably not essential to the paper.

30-Oct-2018
We are grateful to the reviewers for their positive comments and constructive criticism on our manuscript, and the opportunity to revise and improve our study. In response to the reviewers

comments we have included new experiments and altered text and figures of the manuscript. Please find a detailed point-by-point reply below.
Reviewer 1 1. Plet1-deficient DCs do not migrate from the small intestine to mesenteric lymph node. However, it is unclear whether Plet1-deficient DCs do not move at all, or they migrate and accumulate at particular areas where is potentially rich for ECM. DC localization or motility should be tested by histology or intravital microscopy to understand the migratory characteristics of Plet1-deficient DCs.
The reviewer raises an interesting point, that was also mentioned by reviewer 2. To address the question of DC localization in the intestine we labeled small intestinal sections from Plet1+/-and Plet1-/-mice with CD103 and CD3. This allowed us to visualize CD103 expressing T cells and CD103 expressing (CD3 negative) DCs. In line with our flow cytometric analyses, we identified villi that appeared to contain a higher number of CD103+ DC in Plet1-/-mice compared to the heterozygous controls. Even though this was consistently seen in 5 vs 5 mice, the distribution throughout the small intestine was not uniform. While some villi appeared to have a higher number of DC, other villi in the same section did not show this difference. In terms of location, the increased number of CD103+ DCs seemed mostly located to the bottom half of the villus. At this moment it is unclear whether this might correlate with altered composition of the ECM at these locations. We have added representative histological images as new Figure 4A, showing villi with normal and increased numbers of DC. Because of this heterogeneity we believe that the quantification by flow cytometry as in Figures 4B-C is still the most truthful representation of the differences now visualized in Figure 4A.

Page 3, line 37. '… that microbial signals control…'
Since there is no description about the association between microbiota and Plet1 prior to this sentence, 'microbial signals' only in conclusion seems to be abrupt.
We agree with the reviewer that this sentence disrupted the flow of the abstract and it has therefore been removed. For this same reason the microbiota reference has also been removed from the manuscript title.
3. Figure 4D seems to be missing, and current Figure 4D is supposed be Figure 4E. The proper results should be added, and the figure should be corrected. Figure 5C are too small, and hard to read.

4.Letters in
We have re-made the panels in figure 5C to increase font size and general readability. Supplementary Figure 4 are different among proteins, and hard to compare. Also, amino acid sequence besides the 3D diagram could be helpful to compare their similarity.

Magnifications of diagrams in
We thank the reviewer for pointing out this issue. We have replaced the diagrams in Supplemental figure 4A and have added protein sequence alignments of mouse Plet1-Reelin and human Plet1-Reelinin supplemental figure 4B.
6. Since FLT3L-induced BMDCs are more sensitive to CCR7-ligand chemokines than GM-CSF-induced BMDCs (J Immunol 193: 4904), the former cells could be more useful to test the function of PLET1 in DC migration in future studies.
The reviewer raises an important point relevant to most studies using in vitro generated dendritic cells. multiple methods have been described to generate in vitro DCs, yet a definitive consensus on which culture method is appropriate in individual cases remains to be established. The reviewer is right in pointing out that Flt3l-induced DC could have different, or most likely even better CCR7 responses in our in vitro 3D culture system. We have added the following text to the discussion to highlight this fact:

"Our in-vitro experiments were performed with isolated lamina propria DC as well as with GM-CSF matured BMDC. It is important to realize that expression of CCR7 on BMDC is lower in GM-CSF-derived cultures compared to FLT3L-matured BMDC [32]. This might imply that the differences observed could be even more pronounced when using alternative methods of cell preparation."
Reviewer 2 1. The authors to not describe any experiments designed to test the impact of the altered DC migration on immune responses. As dendritic cells are key players in the initiation of immune responses, such an experiment would strengthen the paper considerably.
We agree with the reviewer that linking functional alterations in downstream immune responses with absence of Plet1 expression would have been of added value to our study. We show in our study that steady state, ex-vivo, interstitial migration is reduced when DCs lack Plet1 expression. To define whether this would lead to alterations in immune parameters at steady state we evaluated immune cell composition and IgA production as shown in the original Figure 3. Currently, using the models available in our lab, we have not been able to identify alterations in immune responses after challenge. Oral gavage of the TLR7/8 ligand R848 induces egress of DCs from the lamina propria into the gut-draining mesenteric LN (MLN), both in Plet1+/-controls and in Plet1-/-animals (shown below as reviewer figure 1a). This suggests that Plet1-deficiency does not impair migration of activated DC, and if anything, a trend towards a higher percentage of CD11b+CD103+ DC in the MLN is seen. On the other hand, the DC distribution within the few remaining intestinal DCs is not altered (reviewer figure  1b), suggesting that the trend towards more CD103+CD11b+ DC in the lymph nodes likely reflects the presence of more of these cells in the intestine prior to activation. Differential regulation of migration by Plet1 in activated vs non-activated DC could be of interest, but goes beyond the scope of the current study.
To test whether Plet1 deficiency would alter pathology after small intestinal damage, we exposed Plet1-/-mice and littermate controls to Methotrexate, a model we have previously used to study tissue damage and repair ( In sum we conclude that Plet1-mediated regulation of migration might be altered by (strong) activation of DCs, and that with the models presently available to us we are unable to define differences in pathology in the absence of Plet1. The limited scope of our model and the possible loss of Plet1 effects after activation make that we are not able to draw any definitive conclusions from these observations, that will require in-depth analyses in future studies. For these reasons we have not included these preliminary experiments in the manuscript but show them below for the reviewers discretion only.
2. It is unclear where, exactly, the increased number of double positive DCs in the small intestinal lamina propria reside. Immunohistochemistry might provide helpful and interesting information in this regard, although it is probably not essential to the paper.
The reviewer raises an interesting point, that was also mentioned by reviewer 1. To address the question of DC localization in the intestine we labeled small intestinal sections from Plet1+/-and Plet1-/-mice with CD103 and CD3. This allowed us to visualize CD103 expressing T cells and CD103 expressing (CD3 negative) DCs. In line with our flow cytometric analyses, we identified villi that appeared to contain a higher number of CD103+ DC in Plet1-/-mice compared to the heterozygous controls. Even though this was consistently seen in 5 vs 5 mice, the distribution throughout the small intestine was not uniform. While some villi appeared to have a higher number of DC, other villi in the same section did not show this difference. In terms of location, the increased number of CD103+ DCs seemed mostly located to the bottom half of the villus. At this moment it is unclear whether this might correlate with altered composition of the ECM at these locations. We have added representative histological images as new Figure 4A, showing villi with normal and increased numbers of DC. Because of this heterogeneity we believe that the quantification by flow cytometry as in Figures 4B-C is still the most truthful representation of the differences now visualized in Figure 4A.